Caracterizacion de Clostridium perfringens circulante en muestras fecales de humanos y animales en el altiplano Cundiboyacense Colombiano AnnyCamargoMancipe - Bacterias y bacteriosis de importancia médica veterinaria | Studenta (2024)

Vista previa del material en texto

1 Caracterización de Clostridium perfringens circulante en muestras fecales de humanos y animales en el altiplano Cundiboyacense Colombiano Anny Jineth Camargo Mancipe Documento de tesis presentado como requisito para optar al título de Doctora en Ciencias Biomédicas y Biológicas DOCTORADO EN CIENCIAS BIOMÉDICAS Y BIOLÓGICAS UNIVERSIDAD DEL ROSARIO BOGOTÁ, D.C. 2024 2 Caracterización de Clostridium perfringens circulante en muestras fecales de humanos y animales en el altiplano Cundiboyacense Colombiano Estudiante: Anny Jineth Camargo Mancipe Directora: Marina Muñoz Díaz, Ph.D. Profesora Asociada Instituto de Biotecnología-UN (IBUN) Universidad Nacional de Colombia Co-Director: Juan David Ramírez González, Ph.D. Profesor Titular Facultad de Ciencias Naturales Universidad del Rosario Documento de tesis presentado como requisito para optar al título de Doctora en Ciencias Biomédicas y Biológicas DOCTORADO EN CIENCIAS BIOMÉDICAS Y BIOLÓGICAS UNIVERSIDAD DEL ROSARIO BOGOTÁ, D.C. 2024 3 “Si he logrado ver más lejos, ha sido porque he subido a hombros de gigantes”. Isaac Newton. A mis maestros: Mis padres y profesores 4 AGRADECIMIENTOS A Dios, mi guía y mi fortaleza. A la Dra. Marina Muñoz, Mi directora de tesis, quien con su pasión por la ciencia, paciencia y sabiduría es mi brújula en este viaje llamado ciencia. Gracias por enseñar que detrás del éxito se esconde el fruto del esfuerzo y la dedicación. Al Dr. Juan David Ramírez, Mi maestro en ciencia y ejemplo de excelencia. Como dijo Séneca, "No hay viento favorable para el barco que no sabe a dónde va", y gracias a su guía experta, navegué con determinación hacia la meta de esta tesis doctoral. Al Dr. Manuel Elkin Patarroyo, mi maestro y quien como a muchos niños colombianos me inspiró en el sueño de ser científica, al Dr. Manuel Alfonso Patarroyo y su equipo, por cimentar las bases de mi inicio en la investigación. A mi familia, mi mayor inspiración: un padre que me enseñó que, “Caminante no hay camino, se hace camino al andar”, a una madre inquebrantable por su fe y ejemplo de valentía. A mis hermanos, que con amor y presencia son refugio en las tormentas y celebración en las victorias. Y a mi amor, por ese abrazo que alentó mis días, por su paciencia y amor incondicional. A mis amigos, por los años venideros juntos, sembrando sueños y cosechando triunfos. Agradezco, además, a los jurados que, con compromiso, excelencia académica y aportes, son el estímulo constante en búsqueda del conocimiento. A la Universidad de Boyacá, a la Dra. Rosita Cuervo, al Dr. Osmar Correal y al Dr. Andrés Correal, cuya misión de promover la investigación y la educación han sido el pilar en el cual se ha construido mi formación profesional. A la Universidad del Rosario, que, a través de la Dirección de Investigación e Innovación, el Comité de becas de la Dirección Académica y el Centro de Investigaciones en Microbiología y Biotecnología (CIMBIUR), junto con el apoyo de Minciencias me brindaron el respaldo necesario para hacer realidad los sueños académicos y de investigación que han trascendido fronteras. A todas las comunidades de Gámeza, Jenesano y Paipa, el más sincero agradecimiento por abrirme las puertas de sus hogares y permitirme hacer ciencia en los hermosos campos del Departamento de Boyacá. Este logro es el resultado de un esfuerzo colectivo, y lo celebro con profunda humildad, junto a todos aquellos que lo hicieron posible. Gracias. 5 TABLA DE CONTENIDO 1. LISTA DE PUBLICACIONES............................................................................................6 2. LISTA DE ANEXOS...........................................................................................................7 3. LISTA DE FIGURAS..........................................................................................................8 4. LISTA DE ABREVIATURAS............................................................................................9 5. RESUMEN.........................................................................................................................10 6. MARCO TEÓRICO...........................................................................................................12 7. OBJETIVOS.......................................................................................................................28 8. INTRODUCCIÓN A LOS CAPÍTULOS..........................................................................29 Capítulo 1...............................................................................................................................31 Capítulo 2........................................................................................................................... ....32 Capítulo 3...............................................................................................................................33 9. CONCLUSIONES...........................................................................................................139 10. PERSPECTIVAS....................................................................................................... ....140 11. PRODUCTOS DE LA TESIS.................................................................................. ......142 12. BIBLIOGRAFÍA............................................................................................................146 6 1. LISTA DE PUBLICACIONES Esta Tesis Doctoral corresponde a un compendio de artículos científicos publicados en revistas internacionales e indexadas en el Science Citation Reports. A continuación, se listan todos los artículos que fueron publicados, los cuales están adjuntos a este documento. Cualquier información suplementaria y/o tablas se incluirán en archivos comprimidos organizados según el número asignado a continuación: Artículo 1: Camargo A, Páez-Triana L, Camargo D, García-Corredor D, Pulido-Medellín M, Camargo M, Ramírez J.D. and Muñoz M*. Carriage of Clostridium perfringens in Domestic and Farm Animals across the Central Highlands of Colombia: Implications for Gut Health and Zoonotic Transmission. Vet Res Commun. 2024. Artículo 2: Camargo A., Bohórquez L., López D., Ferrebuz-Cardozo A., Castellanos-Rozo J., Díaz J., Rada M., Camargo M., Ramírez J. D. and Muñoz M. Clostridium perfringens in central Colombia: Frequency, Toxin Genes, and Risk Factors. (SOMETIDO en Gut Pathogens) Artículo 3: Camargo A., Guerrero-Araya E, Castañeda S, Vega L, Cardenas-Alvarez MX, Rodríguez C, Paredes-Sabja D, Ramírez JD, Muñoz M. Intra-species diversity of Clostridium perfringens: A diverse genetic repertoire reveals its pathogenic potential. Front Microbiol. 2022 Jul 22; 13:952081. doi: 10.3389/fmicb.2022.952081. PMID: 35935202; PMCID: PMC9354469. Artículo 4: Camargo A., Bohorquez L., Cáceres T., Ferrebuz-Cardozo A, Díaz J, Castellanos-Rozo J, Diaz J., Kiu R., Hall L. J., Rámirez J. D. and Muñoz M. Insights into Clostridium perfringens Dispersal Hotspots, Toxins, and Virulence Factors through Integrated Genomic and Phenotypic Profiling. (EN CONSTRUCCIÓN) Artículo 5: Camargo A, Rámirez J. D., Kiu R., Hall L.J., Muñoz M. Unveiling the pathogenic mechanisms of Clostridium perfringens toxins and virulence factors. Emerg Microbes Infect. 2024 Apr 9:2341968. doi: 10.1080/22221751.2024.2341968. Epub ahead of print. PMID: 38590276.Artículo 6: Herrera G, Vega L, Camargo A, Patarroyo MA, Ramírez JD, Muñoz M. Acquisition site-based remodelling of Clostridium perfringens- and Clostridioides difficile-related gut microbiota. Comp Immunol Microbiol Infect Dis. 2023 Nov; 102:102074. doi: 10.1016/j.cimid.2023.102074. Epub 2023 Oct 10. PMID: 37832162. 7 2. LISTA DE ANEXOS 2.1 Anexo de artículos Artículo 1 Artículo 2 Artículo 3 Artículo 4 Artículo 5 Artículo 6 2.2 Anexo de productos de la tesis 2.2.1 Lista de publicaciones 2.2.2 Presentación en eventos científicos 2.2.3 Pasantía internacional 2.2.4 Cursos 2.2.5 Becas 2.2.6 Orientación de trabajos de grado Anexos tesis Doctoral Anny Camargo https://uredu-my.sharepoint.com/:f:/g/personal/anny_camargo_urosario_edu_co/ErluOcOCrYNGhpU4wE7FPmkBclDb4JUBpXjtaPpVK8SFqQ?e=agjrB5 8 3. LISTA DE FIGURAS Figura 1. Representación esquemática de los principales hospedadores para cada toxinotipo de C. perfringens. Figura 2. Mecanismos de acción de las principales toxinas de C. perfringens utilizadas para la toxinotipificación. Figura 3. Microfotografías electrónicas de transmisión de esporas de cepas de C. perfringens que contienen genes cromosómicos cpe. Figura 4. Enfermedades humanas causadas por C. perfringens. Figura 5. Enfermedades animales causadas por C. perfringens. 9 4. LISTA DE ABREVIATURAS ADN: Desoxirribonucleico CDC: Center for Disease Control and Prevention CPA: Alfa-toxina de C. perfringens CPB: Beta-toxina de C. perfringens CPE: Enterotoxina de C. perfringens DAA: Diarrea asociada a antibióticos ECN: Enterocolitis necrotizante EN: Enteritis necrotizante ETA: Enfermedades transmitidas por alimentos ETX: Epsilon-toxina GTP: Guanosín trifosfato ITX: Iota-toxina LSR: Receptor de lipoproteínas estimulado por lipólisis MRA: Marcadores de resistencia a antibióticos NETB: Necrotic enteritis B-like toxin OMS: Organización Mundial de la Salud PCR: Reacción en cadena de polimerasa PFOA: Perfingolisina O TSC: Triptosa Sulfito Cicloserina UFC: Unidades Formadoras de Colonias 10 5. RESUMEN Clostridium perfringens es una bacteria anaerobia Gram positiva, formadora de esporas y productora de toxinas, que puede encontrarse en el tracto intestinal humano y animal, así como en alimentos, suelo y agua en forma esporulada. Esta bacteria es reconocida como causante de brotes de intoxicación alimentaria, necrosis intestinal y mionecrosis grave en humanos y animales, ocupando el tercer lugar en incidencia de enfermedades transmitidas por alimentos (ETA) en los Estados Unidos. Estudios recientes han revelado la presencia de aislamientos toxigénicos de C. perfringens, portadores de la toxina perfingolisina O (PFOA) en individuos sanos. Esta toxina se relacionó con toxicidad celular, hemólisis completa, respuestas proinflamatorias y una mayor capacidad de esporulación. La combinación de es estas características biológicas en aislamientos de C. perfringens se correlacionan con un avance desalentador de la patología intestinal, estando asociado con infección/enfermedad y una mayor propagación entre hospederos. Asimismo, se ha observado un aumento en los factores de virulencia y marcadores de resistencia a los antibióticos (MRA) que transporta esta especie bacteriana, lo que representa un desafío considerable tanto a nivel clínico como de salud pública a escala global. A pesar de la relevancia para la salud pública, es evidente que en Colombia y en Suramérica existe un conocimiento limitado sobre esta bacteria. Por lo tanto, es crucial avanzar en la descripción de su arquitectura genómica, así como de sus factores de virulencia y resistencia antibiótica. Estos conocimientos son fundamentales para mejorar las estrategias de prevención y manejo de las enfermedades causadas por C. perfringens. Por lo tanto, esta tesis doctoral tuvo como objetivo describir la frecuencia de detección, diversidad genética y principales factores de virulencia de C. perfringens en muestras fecales de humanos y animales en el altiplano Cundiboyacense Colombiano. La tesis se dividió en tres capítulos: El Capítulo I se enfocó en determinar la frecuencia de detección de C. perfringens en humanos y animales en el altiplano Cundiboyacense colombiano mediante pruebas moleculares. Las muestras fecales de humanos con y sin síntomas gastrointestinales, así como muestras fecales de diversas especies animales, fueron colectadas en los Departamentos de Cundinamarca y Boyacá. Se realizó la extracción de ADN y PCR dirigida al gen 16S-rRNA y al gen de la toxina alfa (cpa) para la detección de C. perfringens. Los resultados revelaron una alta frecuencia de detección de C. perfringens a nivel general en animales domésticos y cerdos. En humanos la toxina beta 2 de C. perfringens (CPB2) se asoció con la presencia de diabetes, lo que sugiere interacciones con el sistema inmunológico del huésped. 11 El Capítulo II tuvo como objetivo describir la estructura genética poblacional y los factores de virulencia de C. perfringens utilizando genomas públicos para evaluar la diversidad genética, linajes y toxinotipos circulantes en diferentes hospederos, así como los principales factores de virulencia a nivel global. Los análisis bioinformáticos revelaron que la mayoría de los genomas provienen de países desarrollados como EE. UU., Francia y China, cuyos aislamientos fueron establecidos principalmente de alimentos, aves y humanos. La clasificación filogenética mostró rutas de dispersión entre diferentes hospederos, además se encontró una alta frecuencia de toxinas como toxina alfa (CPA) y enterotoxina (CPE), así como el incremento en la detección de MRA asociados a tetraciclinas y macrólidos. Por otra parte, se detectó una escasa representación de aislamientos en países en desarrollo, destacando la necesidad de investigaciones locales para comprender mejor la diversidad genética y los mecanismos de transmisión y virulencia de este patógeno. El propósito del Capítulo III fue caracterizar la arquitectura genómica, factores de virulencia y MRA de aislamientos colombianos de C. perfringens, obtenidos de muestras positivas de humanos y animales recolectadas en el Capítulo I. Además, se llevaron a cabo pruebas fenotípicas de hemólisis, inhibición del crecimiento celular, esporulación y susceptibilidad a antibióticos en un grupo representativo de aislamientos. El criterio de selección de los aislamientos se basó en su capacidad para producir la toxina PFOA, una toxina formadora de poros que juega un papel crucial en la patogenia intestinal, con el objetivo de evaluar su impacto biológico. El análisis microgeográfico de aislamientos obtenidos de diversas fuentes en una región central de Colombia, reveló una amplia diversidad genética y posibles eventos de dispersión entre humanos, y animales domésticos como perros y gatos. La presencia de toxinas como PFOA en aislamientos de individuos asintomáticos plantea un potencial riesgo de infección en dicha población, debido a su asociación con hemolisis, inhibición del crecimiento celular y mayor capacidad de esporulación. Además, se observó una reducción de la susceptibilidad a varios antibióticos, incluyendo gentamicina, eritromicina, metronidazol y tetraciclina. En resumen, este estudio ha enriquecido el entendimiento de C. perfringens en Colombia al proporcionar datos de epidemiología molecular y genómica de C. perfringens circulante en la región de análisis. Además, ofreció el primer reporte de genomas obtenidos degatos a nivel mundial e incluyó el mayor número de genomas registrados en América Latina hasta la fecha. Nuestros hallazgos permitieron fortalecer las acciones de promoción de la salud y prevención de la transmisión de enfermedades infecciosas a través de campañas educativas con la comunidad y las entidades de salud, con implicaciones significativas para mejorar la calidad de vida de poblaciones vulnerables. 12 6. MARCO TEÓRICO 6.1 Generalidades sobre Clostridium perfringens C. perfringens (anteriormente conocido como Bacillus aerogenes capsulatus, Bacillus perfringens, Bacillus welchii o Clostridium welchii) es un bacilo Gram positivo, anaerobio que carece de flagelos pero exhibe motilidad de deslizamiento mediada por pili tipo IV [1], productor de toxinas y formador de esporas, lo que le permite sobrevivir en condiciones extremas o con pocos nutrientes, una característica que favorece su rápida dispersión [2]. Esta bacteria fue identificada y aislada por primera vez por William H. Welch en 1891 a partir de la autopsia de un hombre donde se observaron burbujas de gas dentro de sus vasos sanguíneos, signos posteriormente asociados con gangrena gaseosa presentada por soldados británicos durante la Primera Guerra Mundial [3]. C. perfringens es un enteropatógeno oportunista capaz de colonizar el intestino de humanos y animales sin causar síntomas (colonización). Sin embargo, cuando la presencia de esta bacteria se combina con otros factores de riesgo como enfermedades inflamatorias intestinales, cambios dietéticos a dietas ricas en proteínas, o la coexistencia de otros patógenos, puede desencadenar infecciones con síntomas como diarrea, dolor abdominal, vómitos y en los casos graves, necrosis intestinal y afectación sistémica [4, 5]. La producción de toxinas juega un papel crucial en el efecto patógeno de esta bacteria, predisponiendo a infecciones graves. Dentro de los factores de riesgo clave asociados con la producción de toxinas se destacan el uso de antiácidos y la edad avanzada pueden aumentar la susceptibilidad a estas infecciones, no obstante, los mecanismos exactos que influyen en el riesgo individual aún no están completamente claros [5, 6]. 6.2 Epidemiología Según el Centro de Control y Prevención de Enfermedades de los EE. UU., durante el periodo de 2009 a 2015 se presentaron más de 800 brotes de ETA que generaron 800 hospitalizaciones y más de 20 muertes [7]. Dentro de los organismos comúnmente asociados a ETA C. perfringens hace parte de las cinco causas principales junto a norovirus, Salmonella, Campylobacter y Staphylococcus aureus [8, 9]. En Inglaterra, se estima que entre el 8 y el 13% de los brotes gastrointestinales de origen alimentario están asociados a esta bacteria (90.000 casos de C. perfringens al año) [10]. Según el informe de la Autoridad Europea de Seguridad Alimentaria y el Centro Europeo para la Prevención y el Control de Enfermedades, donde se presentan los resultados de las actividades de seguimiento de las zoonosis llevadas a cabo en 2017 en 37 países europeos, se informaron 818 (16,1%) brotes de origen alimentario, asociados con toxinas bacterianas especialmente de C. perfringens, Staphylococcus y Bacillus cereus [11]. 13 6.3 Características del genoma de C. perfringens El tamaño del genoma de C. perfringens varía de 3.0 a 4.1 Mb, con un contenido de G-C relativamente bajo entre 27 y 28% y un potencial codificante de 2.500 a 3.600 genes aproximadamente [12]. Además, presenta un pangenoma compuesto por 11.667 genes (12,6% de genes centrales y 87,4% de genes accesorios), estas características representan un indicador de la alta plasticidad genómica de C. perfringens [13-15]. Los genomas de C. perfringens incluyen genes que codifican para homólogos del sistema feoAB de captación de hierro, un elemento esencial para la supervivencia bacteriana, así como genes asociados a actividades glucolíticas, lo que refleja su capacidad para adaptarse a diversas condiciones ambientales [16]. Además, los genomas de dichas poblaciones bacterianas contienen genes para factores putativos que facilitan la respuesta adaptativa al estrés, incluyendo la enzima superóxido dismutasa, reductasas de metionina sulfóxido y glutatión peroxidasas, contribuyendo a su resiliencia y supervivencia en ambientes hostiles [17]. 6.4 Factores de virulencia de C. perfringens 6.4.1 Toxinas principales de C. perfringens El impacto de las infecciones por C. perfringens se debe en parte al potencial de la bacteria para secretar múltiples toxinas extracelulares, incluyendo la alfa-toxina (CPA, gen cpa/plc), la beta-toxina (CPB, gen cpb), la epsilon-toxina (ETX, gen etx), la iota-toxina (ITX, genes binarios iap e ibp), la enterotoxina de C. perfringens (CPE, gen cpe) y Necrotic enteritis B-like toxin (NetB). La presencia diferencial de estas toxinas permite la clasificación en siete toxinotipos del A al G cada uno implicado con una enfermedad en un hospedero particular [18] (Figura 1). 14 Figura 1. Representación esquemática de los principales hospedadores para cada toxinotipo de C. perfringens. Cada recuadro representa un toxinotipo diferente. Tomado de Camargo y colaboradores. La toxina alfa (CPA), compuesta por 370 aminoácidos, es producida por todos los toxinotipos de C. perfringens y se asocia con mionecrosis clostridial en individuos inmunocomprometidos. Esta patología se caracteriza por edema subcutáneo, necrosis muscular, enfisema, shock y falla multiorgánica [19]. CPA actúa sobre la proteína de unión a Guanosín Trifosfato (GTP), desencadenando la apoptosis y necrosis celular [20]. La CPA también induce la activación de la vía del ácido araquidónico, generando prostaglandinas, tromboxanos y leucotrienos que contribuyen a la inflamación y a la disminución de la perfusión sanguínea [21, 22] (Figura 2A). 15 Figura 2. Mecanismos de acción de las principales toxinas de C. perfringens utilizadas para la toxinotipificación. Mecanismo de acción molecular de las principales toxinas de C. perfringens. (A). Toxina CPA: La toxina CPA interactúa con GM1a, hidrolizando la fosfatidilcolina (PC) y la esfingomielina (SM) dando lugar a la formación de diacilglicerol (DAG) y ceramida (CER) con la activación del receptor Tropomiosina quinasa A (TrKA) y desencadena la activación de una cascada de señalización intracelular con liberación de Interleucina - 8 (IL-8). Se produce la activación del fosfatidil inositol 3 (IP3) que promueve la entrada de calcio (Ca+) intracitoplasmático (B). Toxina CPB: La CPB se une a la molécula-1 de adhesión de las células endoteliales plaquetarias (PECAM-1) con la consiguiente liberación de trifosfato de adenosina (ATP) y formación de poros que permiten el intercambio de iones hacia y desde la célula (C). Toxina ETX: La toxina ETX interactúa con la proteína "mielina y linfocitos" (MAL) formando un poro activo que induce el transporte e intercambio de iones a través de la membrana celular (D). Toxina ITX: La unión de Ib al receptor de lipoproteínas estimulado por lipólisis (LSR) media su entrada en la célula huésped, promoviendo la formación de canales para la entrada de Ia por endocitosis con la consiguiente despolimerización de los filamentos de actina, generando cambios morfológicos y alteración de la permeabilidad celular (E). Toxina CPE: la toxina CPE se une a los receptores de claudina, contribuyendo a la formación de un poro en la 16 superficie celular con intercambio iónico y desequilibrio osmótico. (F). La toxina NetB reconoce regiones libres de colesterol en las membranas celulares formando poros hidrofílicos heptaméricos que permiten la entrada de iones como Na+, Cl- y Ca 2+. Tomado de Camargo y colaboradores. La toxina beta (CPB), compuesta por 336 aminoácidosproducida por C. perfringens tipo B y C, está vinculada con disentería hemorrágica en ovejas y enteritis necrótica en ganado y humanos [23, 24]. Su mecanismo de acción ocurre a nivel de la mucosa intestinal, generando poros en las membranas celulares y causando hemorragias y trombos fibrinosos en el intestino [31]. La patología de la enfermedad se caracteriza principalmente por hemorragias y trombos de fibrina que obstruyen la microvasculatura en la lámina propia del intestino, desencadenando necrosis del epitelio [25] (Figura 2B). La toxina Épsilon (ETX), producida por C. perfringens tipo B y D, se asocia con enterotoxemias en rumiantes y puede provocar complicaciones cerebrales graves [25]. Su mecanismo de acción implica la formación de poros en la membrana plasmática, conduciendo a la muerte celular y a la alteración de la transmisión eléctrica en el cerebro [26]. Además, se ha demostrado que induce inflamación, necrosis, daño mitocondrial y alteraciones en la membrana celular [27, 28] (Figura 2C). La Iota toxina (ITX), producida por el toxinotipo E, se vincula con enteritis hemorrágica. ITX es una toxina binaria implicada en daño al citoesqueleto celular y apoptosis [29, 30]. Su ingreso a las células se facilita a través del receptor de lipoproteínas estimulado por lipólisis (LSR), lo que conduce a la apoptosis celular y a un aumento de la permeabilidad de las monocapas de las células intestinales [29, 31, 32] (Figura 2D). El gen cpe, que codifica la toxina CPE, se halla en el cromosoma de la mayoría de los aislamientos de C. perfringens asociadas a intoxicaciones alimentarias o en plásmidos conjugativos grandes, acompañado de secuencias de inserción que facilitan su movilización y diseminación [33]. La toxina CPE es producida durante la esporulación de C. perfringens tipo F, es crucial en intoxicaciones alimentarias y diarrea [34]. Actúa formando poros en las membranas celulares, desencadenando la muerte celular por necrosis del epitelio ileal y colónico humano a través de la liberación de citocromo C y activación de caspasa-3 [33, 35] Figura 2E. Finalmente, la toxina de enteritis necrótica tipo B (NetB), es una toxina formadora de poros codificada por el gen netB ubicada en plásmidos conjugativos [36]. Es producida por C. perfringens tipo G [18]. Esta toxina está implicada en enteritis necrótica aviar, forma poros hidrófilos que causan lisis celular y destrucción de la lámina propia del intestino [37, 38] (Figura 2F). 6.4.2 Toxinas accesorias de C. perfringens Perfingolisina O La Perfingolisina O (PFOA), clásicamente conocida como toxina θ, es codificada por el gen pfoA, situado en el cromosoma de C. perfringens, y actúa como una toxina formadora de poros en las membranas que contienen colesterol. Aunque se ha sugerido que casi todos los aislamientos de C. 17 perfringens codifican el gen pfoA. Algunos estudios han revelado que la mayoría de los aislamientos productores de enterotoxinas asociadas a intoxicaciones alimentarias carecen de este gen [17, 39]. La toxina PFOA se asocia con citotoxicidad en células Caco-2 y hemólisis completa, lo que está vinculado con la patología de las células intestinales y posiblemente con los resultados de infección o enfermedad en neonatos con enterocolitis necrotizante. Por otro lado, los aislamientos de C. perfringens portadores de pfoA tanto en individuos sanos como en aquellos con enterocolitis necrotizante, muestran una mayor tolerancia al oxígeno y una capacidad de formación de esporas/germinación más pronunciada, lo que parece favorecer su propagación [40]. Además, la toxina PFOA se vincula con el desarrollo de la gangrena gaseosa en humanos, ya que parece actuar de forma sinérgica con la toxina CPA para afectar desde la leucostasis periférica hasta la mionecrosis y la coagulopatía intravascular [41, 42]. Asimismo, la PFOA puede aumentar la expresión de moléculas de adhesión, como CD11b / CD18 en leucocitos y la molécula de adhesión intracelular 1 (ICAM-1) y el factor activador de plaquetas en células endoteliales humanas, contribuyendo así a la trombosis y a la disminución del flujo sanguíneo en la gangrena gaseosa [43, 44]. Alveolisina La alveolisina es una toxina poco estudiada que es dependiente de colesterol y guarda similitudes con la PFOA [45]. Se activa a través de compuestos reductores, específicamente mediante la activación de tiol, y su receptor principal es el colesterol. Esta toxina exhibe actividad lítica contra los eritrocitos de varias especies animales. Se ha descrito que esta toxina está implicada en la inducción de la expresión de IL8 en polimorfonucleares humanos y en poblaciones de células linfocitos-monocitos-basófilos [46]. A medida que se acumula más información sobre el papel de esta familia de toxinas como factores de virulencia importantes, se hace cada vez más evidente que sus efectos pueden ir más allá de la simple destrucción de eritrocitos mediante la formación de poros. Los efectos aparentes sobre la función de las células inmunitarias y la inducción de vías inflamatorias parecen ser su principal mecanismo de acción [46]. Clostridium perfringens beta 2 Clostridium perfringens beta 2 (CPB2), es una toxina accesoria localizada en un plásmido que se ha asociado a enteritis porcina, equina y bovina [47]. La estructura y el receptor de CPB2 siguen sin estar claros. Sin embargo, se ha sugerido que sólo una fracción muy pequeña de segmentos de aminoácidos adopta una conformación α-hélice, lo que es insuficiente para atravesar una membrana. En contraste, se han identificado algunos segmentos de aminoácidos que podrían formar cadenas β transmembrana [48]. 18 Esta toxina formadora de poros está implicada en la formación de canales selectivos de cationes de aproximadamente 1,4 nm de diámetro en bicapas lipídicas, lo que provoca una alteración del flujo de iones y un aumento de la permeabilidad intestinal que desencadena la sintomatología durante la enteritis [49]. Sialidasas Las sialidasas, también conocidas como neuraminidasas y exosialidasas, son codificadas por los genes nanH, nanI y nanJ, y desempeñan un papel crucial en la patogénesis de bacterias patógenas. Estas enzimas promueven la colonización bacteriana, la adhesión, la internalización, la formación de biopelículas y la unión de toxinas a las células huésped al hidrolizar el enlace α-glucósido del ácido siálico terminal en glucoproteínas y glucolípidos [50, 51] Los aislamientos de C. perfringens, producen diversas sialidasas como: NanH, NanI y NanJ, codificadas por genes ubicados en diferentes regiones del cromosoma [17, 52]. La producción de estas enzimas varía entre aislamientos, siendo NanH un posible factor de virulencia accesorio que aún continua en estudio, mientras que para NanI se ha demostrado que juega un papel crucial en la adherencia a células y tejidos del huésped, así como en la modulación de la actividad de toxinas como CPA, CPB, ETX y NetF. Estudios sugieren que estas enzimas pueden regular positivamente la producción de toxinas, mejorar su citotoxicidad, aumentar la adhesión bacteriana a las células huésped y proporcionar sustratos para el crecimiento bacteriano y el metabolismo [53-55]. 6.4.3 Esporulación de C. perfringens La capacidad de C. perfringens para formar esporas juega un papel clave durante la transmisión de esta bacteria, su papel en las enfermedades y su supervivencia en muchos nichos ambientales como el suelo, aguas residuales, heces y alimentos. Las esporas de C. perfringens presentan una compleja estructura multicapa que desempeña distintas funciones esenciales para su supervivencia y reproducción. La capa más externa está formada por más de 50 proteínas específicas, las cuales ofrecen protección contra enzimas líticas [56] Figura 3. 19 Figura 3. Microfotografías electrónicas de transmisiónrepresentativas de esporas de cepas de C. perfringens que contienen genes cromosómicos cpe. Cada microfotografía está etiquetada con la designación de la cepa respectiva. Los componentes estructurales medidos por tamaño están etiquetados: ct, capas de recubrimiento proteínico de la espora; cx, región de la corteza de la espora; y c, el núcleo de la espora con ribosomas que dan un aspecto granular. Las barras más bajas de cada micrografía representan 1,0 AM. Tomado de Novak y colaboradores [57]. Subsiguientemente, se encuentra el pelaje de la espora, seguido por la membrana externa. Aunque esta membrana no proporciona protección a las esporas inactivas, es fundamental para el proceso de formación de esporas. Bajo estas capas, se halla la corteza, compuesta principalmente de peptidoglicano, similar a la pared celular en crecimiento. Esta corteza es crucial para mantener la hidratación del núcleo de la espora, contribuyendo significativamente a la resistencia de las esporas frente al estrés ambiental y a agentes químicos [56]. Procediendo hacia el interior, se encuentra la pared de la célula germinal, la cual, a diferencia de las capas exteriores, no juega un rol en la resistencia de las esporas. Durante el proceso de germinación, esta pared se transforma en la pared celular activa. La membrana interna, situada debajo de la pared de la célula germinal, actúa como una barrera protectora para el núcleo, que constituye la capa más interna de la espora. El núcleo alberga el ADN, el ARN y la mayoría de las enzimas críticas para las esporas, siendo esencial para su viabilidad y funcionamiento [58]. Las esporas mueren por daño a varios componentes diferentes, como el ADN, la membrana interna de la espora, las proteínas del núcleo de la espora, entre otros. Sin embargo, su resistencia está dada por el bajo contenido de agua del núcleo de la espora (20-50%), su capa protectora de peptidoglicano, sus altos niveles de ácido Ca-dipicolínico (Ca-DPA) (25%) y la saturación de ADN con pequeñas proteínas solubles en ácido (SASPs) [59, 60]. Durante la intoxicación alimentaria, causada por C. perfringens, las esporas germinan en alimentos con temperaturas de cocción muy bajas (<50 ºC) seguido de una rápida multiplicación de células vegetativas. Con un tiempo de duplicación muy corto de ~10 minutos, lo que le permite crecer 20 rápidamente en alimentos hasta alcanzar una carga bacteriana (>106 células vegetativas/gramo de alimento) necesaria para iniciar la enfermedad gastrointestinal [58]. Una vez ingeridos los alimentos contaminados, muchas de estas células mueren por exposición al pH ácido del estómago; sin embargo, algunas sobreviven y llegan al intestino delgado donde se inicia el proceso de esporulación en presencia de fosfato inorgánico en el medio ambiente. Durante la esporulación la toxina CPE involucrada en la patogenésis de la intoxicación alimentaria, se libera en la luz cuando la célula madre esporulante se lisa para liberar su espora madura [58, 61]. En los casos de mionecrosis, las células vegetativas o esporas de C. perfringens pueden ingresar al tejido muscular a través de una herida y germinar en condiciones de oxido – reducción (redox), las células vegetativas resultantes luego crecen rápidamente para reducir aún más las condiciones de Redox del tejido, promoviendo un crecimiento bacteriano adicional. Las células vegetativas producen CPA, CPE y PFOA que provocan necrosis local, daño del tejido, entre otros [62]. 6.5 Impacto de C. perfringens en humanos Se han descrito múltiples patologías humanas asociadas a C. perfringens (Figura 4). Figura 4. Enfermedades humanas causadas por C. perfringens [13, 23, 63]. RN: Recién nacido. Fuente: Autor 6.5.1 Intoxicación alimentaria 21 Cada año en todo el mundo se registran aproximadamente 600 millones de casos de enfermedades transmitidas por alimentos y 420.000 muertes. La intoxicación alimentaria por C. perfringens se encuentra entre las enfermedades gastrointestinales más prevalentes en países desarrollados [64]. La historia natural de la intoxicación alimentaria por C. perfringens ocurre cuando las esporas de las cepas tipo F que portan el gen cpe cromosómico sobreviven a la cocción inadecuada de alimentos (rango de temperatura de 10 a 50ºC), lo que permite la germinación de células vegetativas [65] [66]. Algunas células vegetativas ingeridas en alimentos contaminados sobreviven al pH ácido del estómago y pasan al intestino donde comienzan a esporular, la célula madre se lisa liberando la toxina CPE a la luz del intestino, donde actúa a través del mecanismo descrito anteriormente [33, 67-69]. Los síntomas clásicos de intoxicación alimentaria por C. perfringens ocurren dentro de las 8 a 14 h posteriores a la ingestión de alimentos contaminados con esporas y se caracterizan por la aparición de dolor abdominal, diarrea acuosa y deshidratación [13, 70]. El diagnóstico de intoxicación alimentaria por C. perfringens se basa en el aislamiento de colonias reductoras de sulfito que se vuelven negras en agar que contiene hierro y es suplementado con cicloserina; la presencia de C. perfringens es confirmada por microscopia [71]. 6.5.2 Diarrea esporádica y diarrea asociada a antibióticos (DAA) La DAA es una complicación grave que sigue siendo un problema sanitario importante, tanto en los pacientes hospitalizados como en la comunidad. Los aislamientos de C. perfringens tipo F que llevan el gen cpe en plásmidos de ~ 70-75 kb son asociadas con casos de enfermedad gastrointestinal no transmitida por alimentos como diarrea esporádica y DAA [34, 72, 73]. Se ha estimado que 5% de todos los pacientes con DAA están infectados por C. perfringens enterotoxigénica [74, 75]. La DAA se presenta como consecuencia del tratamiento antimicrobiano con penicilina, trimetroprim, cefalosporinas, o clotrimoxazol (Trimetroprim-sulfametoxazol), mientras que la diarrea esporádica se desarrolla independiente de cualquier tratamiento antibiótico. En contraste a las grandes cantidades de células bacterianas cpe+ cromosómicas que deben ser ingeridas en alimentos contaminados para causar intoxicación alimentaria, la DAA y la diarrea esporádica son causadas por un pequeño inóculo de células positivas para cpe transmitidas por plásmido seguido de la transferencia conjugativa del plásmido cpe+ a poblaciones intestinales de C. perfringens cpe- presentes en el entorno intestinal [70, 76, 77]. 6.5.3 Gangrena gaseosa Las infecciones de la piel y los tejidos blandos van desde leves, moderadas a graves con afectación sistémica, de progresión rápida, que requieren tratamiento médico de inmediato. Estas infecciones pueden ser mono o polimicrobianas causadas por microorganismos aeróbicos, anaeróbicos o una combinación de ambos, los cuales colonizan áreas expuestas de la piel debido principalmente a traumatismos, heridas penetrantes, intervenciones quirúrgicas o úlceras en pacientes diabéticos donde C. perfringens representa el 80-95% de las infecciones después de una lesión o cirugía [78, 79]. 22 En los Estados Unidos, la incidencia de mionecrosis es de 1.000 casos por año, sin embargo, en los países menos desarrollados con menor acceso a la atención médica, tratamiento antibiótico y diagnóstico oportuno, la incidencia es probablemente mayor, pero se desconoce el número exacto [79]. La gangrena gaseosa causada por la invasión exitosa de heridas traumáticas por células vegetativas o esporas de cepas de C. perfringens productoras de CPA y PFOA, se caracteriza por mionecrosis grave (necrosis muscular), acumulación de leucocitos intravascular, trombosis significativa y hemólisis [13, 80]. En los casos más severos, la infección puede llevar a sepsis grave, lo que incluye choque séptico, síndrome de dificultad respiratoria del adulto, coagulación intravascular diseminada y anemiahemolítica [79, 81]. 6.5.4 Enteritis necrotizante (EN) La EN causada por C. perfringens tipo C productor de la toxina beta, es una enfermedad infecciosa potencialmente mortal que se caracteriza por necrosis segmentaria del yeyuno proximal [13, 82-85]. Los factores de riesgo son diabetes mellitus, desnutrición, ingesta repentina de comidas proteicas, ingestión de carne de cerdo contaminada con C. perfringens tipo C y altas cantidades de inhibidores de tripsina en la nutrición. La patogenia de la EN tipo C se atribuye principalmente a CPB, una exotoxina, que causa daño celular a través de la formación de poros multiméricos en las membranas plasmáticas de las células sensibles [86-88]. Los síntomas de la enfermedad son el dolor abdominal severo principalmente en el abdomen superior, vómito, fiebre, distensión abdominal, diarrea sanguinolenta y en los casos más severos puede progresar a sepsis, shock y muerte [89, 90]. 6.5.5 Enterocolitis necrotizante del recién nacido La enterocolitis necrotizante (ECN) es una de las principales causas de mortalidad en los recién nacidos prematuros (mortalidad de hasta un 35% en lactantes de peso extremadamente bajo al nacer <1000 g) [91], caracterizada por inflamación intestinal aguda con o sin perforación intestinal [92]. Se cree que la ECN es una enfermedad multifactorial que requiere un intestino inmaduro, alimentación enteral y colonización bacteriana para su desarrollo, además las preocupaciones sobre la infección intrauterina oculta que precipita el parto prematuro, la rotura prematura de membranas y la corioamnionitis a menudo provocan el inicio de un tratamiento antibiótico empírico en la primera semana de vida del lactante, con posible disrupción de la microbiota intestinal, lo que se asocia con un mayor riesgo de ECN [93-95]. El aislamiento de C. perfringens tipo A y C en lactantes con ECN sugiere su implicación en la fisiopatología de la enfermedad [96], donde la isquemia del intestino dada por suministro vascular reducido con baja tensión de oxígeno, podrían favorecer la conversión de esporas clostridiales en bacilos invasores productores de toxinas [97]. La presentación clínica de ECN puede variar entre los recién nacidos. Los signos clínicos característicos son distensión abdominal, llanto frecuente, heces con sangre, vómitos, ruidos 23 intestinales ausentes, en los casos más severos se observa celulitis en la pared abdominal, peritonitis, letargo, shock y la muerte [98-100]. 6.6 Impacto de C. perfringens en animales C. perfringens no sólo afecta humanos, sino que también, causa diversas enfermedades en hospedadores animales (Figura 5). Figura 5. Enfermedades animales causadas por C. perfringens [3, 52, 109]. Fuente: Autor 6.6.1 Enteritis necrótica en cerdos La EN grave y letal en lechones recién nacidos es causada principalmente por C. perfringens tipo C que se define por llevar los 2 genes de toxina de tipificación cpa+ y cpb+, además algunas cepas pueden portar el gen de la toxina CPE [18]. Las cepas de tipo C pueden producir otras toxinas, que, no se utilizan para la tipificación, como CPB2, PFOA y la toxina clostridial Tpel [18, 101]. C. perfringens tipo C causa enteritis con mayor frecuencia en animales recién nacidos como terneros, ovejas, cabras y, en particular, cerdos [102]. Esta enfermedad en cerdos puede propagarse rápidamente, los lechones con un curso clínico de subagudo a crónico más prolongado tienen diarrea no hemorrágica y crecimiento reducido [103-105]. 24 6.6.2 Abomastitis y enteritis en ruminates Diversas enfermedades como enterocolitis hemorrágica, enterotoxemia, enfermedad del riñón pulposo, abomastitis, entre otras son causadas por deferentes tipos de C. perfringens. El tipo A es un aislamiento cada vez más común asociado con abomasitis por clostridios en rumiantes, el tipo E también se ha aislado en casos de abomasitis, pero parece ser raro [106-108]. La tasa de letalidad de la abomasitis por clostridios parece ser alta cuando no se instituye un tratamiento rápido (75% -100%) [109]. Los signos clínicos incluyen distensión abdominal, letargo, distensión líquida del abomaso (último compartimento del complejo aparato estomacal de los rumiantes), timpanismo abdominal, cólicos y bruxismo (movimiento de diente contra diente), en los casos graves los animales pueden presentar fiebre, sepsis sistémica o peritonitis [106, 110]. C. perfringens tipo A también puede causar enterotoxemia en ovejas (enfermedad del cordero amarillo), un trastorno aparentemente raro pero muy fatal que se manifiesta como una enfermedad hemolítica aguda [111, 112]. El tipo C se asocia con enterocolitis hemorrágica en terneros y corderos y se sospecha que ocurre en raras ocasiones en cabras [106]. C perfringens tipo D es responsable de enterotoxemia en pequeños rumiantes de todas las edades, en ovejas es típicamente una enfermedad hiperaguda, y muchos casos simplemente se encuentran muertos, mientras que en terneros es muy rara [106]. Aunque el tipo E es supuestamente raro y aparece como una causa poco común de enterotoxemia en corderos, terneros y conejos, algunos estudios sugieren que el tipo E puede desempeñar un papel importante en la enteritis de terneros recién nacidos que pueden cursar con diarrea y muerte súbita, abomasitis y enteritis hemorrágica [113]. 6.6.3 Síndrome de diarrea hemorrágica aguda canina Se ha descrito que C. perfringens toxinotipo F productor de CPE asociado con la liberación de toxinas NetF, posiblemente junto con otras toxinas como NetE y NetG, es el responsable de la enterocolitis necrótica en perros, aunque se necesita más trabajo para comprender su epidemiología y los factores que predisponen a los perros a la infección [114] . La enteritis necrótica canina se caracteriza por la aparición diarrea sanguinolenta severa, vómitos, y pérdida significativa de líquidos a través del tracto gastrointestinal [115, 116]. 6.6.4 Enteritis necrótica en aves de corral La enteritis necrótica en pollos es una enfermedad bacteriana común en la industria avícola, que impone una carga económica significativa debido a la reducción de la eficiencia de producción y el costo de las medidas de control. Es una enfermedad principalmente de pollos de engorde, pero también puede afectar a gallinas ponedoras y pavos [117]. En los estudios de lesiones de aves enfermas generalmente se han encontrado aislamientos de C. perfringens que portan el gen netB, mientras que el transporte del gen es menos frecuente en los 25 aislamientos de aves sanas. Las toxinas netB y TpeL se han asociado con la EN clínica en aves de corral [118, 119]. La enteritis necrótica clínica se caracteriza por un aumento de mortalidad repentino con tasas de letalidad que pueden llegar al 50%, los signos clínicos incluyen pérdida de peso, diarrea, intestino delgado distendido y daño o necrosis de la mucosa intestinal, lo que genera un aumento repentino de la mortalidad [117, 120, 121]. 6.7 Detección y aislamiento de C. perfringens Los métodos estándar para identificar C. perfringens tradicionalmente se basan en medios selectivos para el aislamiento (p. Ej., Triptosa-sulfito-cicloserina con agar yema de huevo [TSC-EYA][122], agar sangre con polimixina B [123] o placas de agar C. perfringens de membrana [mCP; Thermo Fisher Scientific, Reinach, Suiza])[124]. Los aislamientos de C. perfringens se caracterizan por colonias reductoras de sulfito, que aparecen negras en agar que contiene hierro complementado con cicloserina. En agar sangre, C. perfringens normalmente forma una zona doble característica de hemólisis. La zona interna de hemólisis completa es causada por la toxina PFOA; la zona exterior de hemólisis incompleta es causada por CPA [125]. Como alternativa práctica, rápida y sensible para la identificación de C. perfringens, se viene utilizandola detección molecular mediante la técnica de Reacción en Cadena de la Polimerasa (PCR). La detección molecular de C. perfringens dirigida a genes constitutivos de la especie resulta efectiva, debido a la elevada especificidad y sensibilidad analítica. Dentro de los genes más utilizados en este campo se incluyen el gen codificante para la toxina CPA (cpa), ubicado en una región estable del cromosoma y codificado universalmente por todos los aislamientos de C. perfringens [126], así como el gen de la subunidad ribosomal 16S (ARNr 16S). Este último es un gen conservado a nivel de especie, con múltiples copias en una misma célula, lo cual aumenta la sensibilidad analítica de la prueba [127]. Utilizar más de un marcador molecular para la detección de patógenos reduce sesgos asociados con la eficiencia de amplificación, uniones inespecíficas de los cebadores e incluso perdida de anillamiento por la alta plasticidad genética del agente. Estudios realizados en países europeos y asiáticos han empleado cultivo in vitro y/o pruebas moleculares para detectar C. perfringens, revelando frecuencias de detección que oscilan alrededor del 38,4% en China [128], 25,0% en Reino Unido [40], 42,0% en Canadá [129] y 49,0% en Egipto [130]. Sin embargo, en países en vías de desarrollo, especialmente en América Latina, la investigación sobre la frecuencia de detección de C. perfringens es limitada, un estudio pionero en Colombia reportó frecuencias de infección del 32,7% en individuos de la comunidad [131]. 6.8 Tratamiento de la infección por C. perfringens El tratamiento de la intoxicación alimentaria por C. perfringens toxinotipo F es sintomático, principalmente se basa en el manejo del dolor abdominal e hidratación. El mejor enfoque para prevenir este tipo de enfermedades es cocinar bien los alimentos y mantenerlos a temperaturas adecuadas (<4ºC - >65ºC). En el caso de la EN, el único tratamiento es una cirugía intestinal rápida, para resecar la parte del intestino afectado [132]. 26 El tratamiento de las enfermedades causadas por C. perfringens dependen del cuadro clínico que se presente. Sin embargo, algunas medidas son el manejo de la distensión abdominal, particularmente si la respiración está comprometida debido a la presión abdominal sobre el diafragma (en los casos de patología gastrointestinal), soporte sistémico y nutricional con líquidos intravenosos si está indicado, prevención de la proliferación bacteriana en curso mediante el uso de antibióticos como penicilina G, Clindamicina o Metronidazol, restaurar la microbiota intestinal normal y el manejo del dolor según sea necesario [106]. 6.9 Resistencia a antibióticos de C. perfringens Como patógeno emergente humano y animal que puede causar diversas enfermedades, C. perfringens ha ganado una atención creciente a nivel mundial en los últimos años debido especialmente a la gran cantidad de factores de virulencia y el aumento de resistencia a los antibióticos, lo que podría deberse a las presiones selectivas causadas por el uso inadecuado de antibióticos, tanto en animales como en humanos, especialmente durante los últimos 30 años [133]. La preocupación por la resistencia a los antibióticos en C. perfringens, se fundamenta en: i) el aumento de la resistencia a los antibióticos más comúnmente utilizados en el ámbito clínico para tratar infecciones en humanos, como la tetraciclina, clindamicina, gentamicina y eritromicina, que favorece su capacidad de proliferación y daño, siendo una de las características de este patógeno oportunista, y ii) el incremento cada vez mayor de la resistencia a antibióticos específicamente empleados en el tratamiento de infecciones por clostridiales, como la penicilina y el metronidazol. Estos fenómenos se atribuyen principalmente a la combinación del uso indiscriminado de antibióticos junto a la capacidad intrínseca de las bacterias para desarrollar mutaciones espontáneas en sus genomas, así como para llevar a cabo la transferencia horizontal de genes de resistencia a antibióticos [134]. Muchos MRA pueden moverse entre el cromosoma bacteriano y los plásmidos, dentro de la misma especie o entre diferentes especies o incluso entre géneros, a través de diferentes mecanismos de movilización como conjugación, transducción y transformación [135]. Aunque los mecanismos de transferencia horizontal de genes a menudo se consideran los principales mediadores de la resistencia a los antibióticos, la emergencia y fijación de mutaciones representa un fenómeno complejo, que puede afectar la unión del antibiótico al receptor, evitar el acceso del antibiótico a la célula diana o la proteger los blancos de los fármacos [136, 137], lo cual induce mecanismos de resistencia a antibióticos convirtiéndose en una amenaza continua a los tratamientos eficaces. La integración de datos genómicos que reflejen la emergencia y propagación de mutaciones/genes asociados a resistencia antibióticos, con procesos de caracterización fenotípica podrían aportar información clave para fortalecer los regímenes de tratamiento, la vigilancia y el diseño de fármacos para ayudar a combatir la diseminación de la resistencia a los antibióticos [138]. Los plásmidos conjugativos son moléculas autorreplicativas que se transfieren entre aislamientos, a través del sistema de secreción tipo IV. El plásmido de resistencia a la tetracicilina 47 kb pCW3 de C. perfringens es un plásmido que comprende 47.263 pb, con un contenido de G + C del 27,6%, 27 similar al del cromosoma de C. perfringens [139], que porta dos genes de resistencia a la tetraciclina: proteína de resistencia al flujo de tetraciclina (tetA (P)) y proteína de resistencia a la tetraciclina (tetB (P)) [140, 141]. tetA(P) codifica una supuesta proteína transmembrana de 46 kDa que media la salida activa de tetraciclina de la célula, mientras que tetB(P) codifica una supuesta proteína de 72,6 kDa que tiene una similitud significativa con las proteínas de resistencia a tetraciclina similares a TetM [142]. Diversos estudios han encontrado en aislamientos de C. perfringens marcadores de resistencia a cloranfenicol [cat(P), cat(Q)], macrólido-lincosamida-estreptogramina B (MLSB) [erm(B), erm(Q)], tetraciclina [tetA(P), tetB(P), tet(M) and tet(Q)] y un marcador recientemente descubierto de resistencia a los macrólidos [mef (A)], por lo tanto, el tratamiento con estos antibióticos se considera ineficaz [143-146]. 6.10 Estrategias de prevención de la transmisión de C. perfringens Según el Center for Disease Control and Prevention (CDC), C. perfringens puede encontrarse en el medio ambiente y en el tracto intestinal de animales y humanos, y participa frecuentemente en el desarrollo de enteritis. Las heces de animales infectados y la carne contaminada son las fuentes más comunes de infección. Los alimentos mantenidos a temperaturas inseguras entre 4ºC y 60ºC facilitan la rápida multiplicación bacteriana y la germinación de esporas, convirtiéndose en bacterias activas que se propagan en los alimentos. Los alimentos más comúnmente asociados con brotes de intoxicación alimentaria son el pavo, el pollo, la carne de res o cerdo, las salchichas y las salsas. Por lo tanto, es esencial que las personas cocinen estos alimentos a temperaturas superiores a 60ºC y los refrigeren a -5ºC dentro de las 2 horas posteriores a la cocción. Además, los alimentos refrigerados deben ser calentados nuevamente a temperaturas superiores a 60ºC antes de su consumo [147]. Así mismo, se recomienda mantener separadas las carnes de aves, pescados y mariscos crudos de otros alimentos en el refrigerador de casa. Colocar estos alimentos en recipientes sellados o bolsas de plástico en el estante inferior del refrigerador evita que los jugos crudos goteen sobre otros alimentos. Los alimentos refrigerados pueden durar un máximo de 2-3 días, y en el congelador a -18ºC,un máximo de 8-12 meses. Además, otras medidas de higiene, como lavarse las manos correctamente, disponer adecuadamente de los desechos, limpiar las superficies con toallas de papel y el manejo adecuado de excretas, pueden prevenir el riesgo de contaminación cruzada y la propagación de bacterias. Implementar estas prácticas de prevención en la comunidad contribuirá significativamente a la seguridad alimentaria, y prevención de la transmisión de C. perfringens [147]. 28 7. OBJETIVOS OBJETIVO GENERAL: Describir la frecuencia de detección, diversidad genética y principales factores de virulencia de Clostridium perfringens en muestras fecales de humanos y animales en el altiplano Cundiboyacense Colombiano. OBJETIVOS ESPECÍFICOS: 1. Determinar la frecuencia de detección de C. perfringens en humanos y animales en el altiplano Cundiboyacense Colombiano mediante pruebas moleculares. 2. Describir la estructura genética y factores de virulencia de C. perfringens a partir de genomas públicos. 3. Caracterizar la arquitectura genómica, marcadores de resistencia y factores de virulencia de aislamientos de C. perfringens obtenidos de las poblaciones de estudio. 4. Describir los perfiles de hemolisis, citotoxicidad, esporulación y susceptibilidad a antibióticos de C. perfringens en las poblaciones de estudio. 29 8. INTRODUCCIÓN A LOS CAPÍTULOS C. perfringens, un microorganismo que reside en los intestinos de humanos y animales tiene la notable capacidad de proliferar y liberar una variedad de toxinas, desempeñando un papel crucial en la aparición de diversas enfermedades de tejidos. Este patógeno se ha identificado como uno de los cinco principales causantes de ETA en países desarrollados [8, 9], representando hasta 90.000 casos anuales de brotes gastrointestinales en Inglaterra [10] y 1000 casos por año de gangrena gaseosa solo en los EE.UU. [148]. Sin embargo, en países en vías de desarrollo, como Colombia, los estudios son limitados desconociéndose datos epidemiológicos y trabajos experimentales publicados que involucren a este patógeno. En un estudio pionero reportado en Colombia se recolectaron un total de 220 muestras de heces de pacientes con diarrea, durante el período de septiembre de 2015 a abril de 2017, en dos centros de atención de salud ubicados en la ciudad de Bogotá, Colombia. Este estudio encontró que 32.7% de los participantes estaba infectado con C. perfringens. Es de particular interés que la frecuencia de infección fue considerablemente más alta en pacientes de la comunidad que a nivel intrahospitalario, lo que sugiere una prevalencia significativa de este patógeno en el entorno comunitario [131]. Este hallazgo subrayó la necesidad de investigar la frecuencia de detección de C. perfringens en humanos y animales a nivel microgeográfico e identificar factores de riesgo para la colonización o desarrollo de infección y características genómicas mediante estudios de secuenciación de genoma completo. Los estudios de genómica en bacterias permiten una caracterización detallada de los patógenos, revelando aspectos críticos como las relaciones filogenéticas, los factores de virulencia y marcadores de resistencia a antibióticos. El análisis de 206 genomas públicos de C. perfringens ha arrojado luz sobre la complejidad genética de este microorganismo, revelando la existencia de cinco filogrupos distintos. Este descubrimiento subraya la notable diversidad genómica y la capacidad de adaptación de C. perfringens, evidenciando su alta plasticidad genética. Estos hallazgos son cruciales para comprender mejor la dinámica de este patógeno y su relación con diversas enfermedades [149]. A pesar de estos avances, existe una marcada escasez de investigaciones a nivel microgeográfico en países en desarrollo, lo que pone de manifiesto la urgente necesidad de estudios focalizados en la diversidad genética de C. perfringens, que aporten en la descripción de una diversidad potencialmente desconocida de este patógeno. Especialmente, es de interés enfocarse en la investigación de factores de virulencia y los MRA circulantes en diferentes contextos poblacionales, lo que podría aportar conocimiento sobre las variaciones en la patogenicidad y la resistencia a tratamientos. Adicionalmente, la reciente detección, de aislamientos toxigénicos de C. perfringens que producen PFOA en bebés sanos y en neonatos con ECN en Inglaterra, y su correlación directa con toxicidad celular, hemólisis completa, respuestas proinflamatorias y mayor tolerancia al oxígeno, destacan la importancia de PFOA en la patología intestinal, en el establecimiento de la infección y en el desarrollo de enfermedad [40]. Estos rasgos plantean un desafío importante para las medidas de control de 30 infecciones y subrayan la necesidad de un examen detallado de C. perfringens en poblaciones aparentemente sanas que habitan en comunidad, para monitorear y prevenir la transmisión de aislamientos virulentos. Esto incluiría además de estudios genómicos, evaluaciones fenotípicas de la hemólisis, capacidad de esporulación y viabilidad de las esporas, así como pruebas de susceptibilidad a los antibióticos. Dicha investigación no solo ampliaría nuestra comprensión de C. perfringens, sino que también contribuye significativamente al desarrollo de medidas de prevención y manejo más efectivas. Este enfoque integral hacia la investigación de C. perfringens es esencial para enriquecer nuestro conocimiento sobre este patógeno, lo que, a su vez, fortalecerá las estrategias de salud pública dirigidas a mitigar las infecciones gastrointestinales asociadas a este microorganismo, tanto en Colombia como en otros países. Por estas razones, se identificó la necesidad de describir la frecuencia de detección, diversidad genética, principales factores de virulencia, MRA y características fenotípicas de aislamientos de C. perfringens obtenidos a partir de muestras fecales de humanos y animales en el altiplano Cundiboyacense Colombiano. Para alcanzar los objetivos planteados, esta tesis fue dividida en tres capítulos. A continuación, se describen los capítulos y los objetivos a los que impacta cada uno de ellos: CAPÍTULO I: Determinar la frecuencia de detección de C. perfringens en humanos y animales en el altiplano Cundiboyacense Colombiano mediante pruebas moleculares (Objetivo específico 1). CAPÍTULO II: Describir la estructura genética y factores de virulencia de C. perfringens a partir de genomas públicos (Objetivo específico 2). CAPÍTULO III: Caracterizar la arquitectura genómica y perfiles fenotípicos de hemolisis, citotoxicidad, esporulación y susceptibilidad a antibióticos de aislamientos colombianos de C. perfringens (Objetivos específicos 3 y 4). 31 CAPITULO I: Determinar la frecuencia de detección de C. perfringens en humanos y animales en el altiplano Cundiboyacense Colombiano mediante pruebas moleculares. El Capítulo I se centra en la determinación de la frecuencia de detección de C. perfringens en humanos y animales en el altiplano Cundiboyacense colombiano usando pruebas moleculares. Se recolectaron muestras fecales de 114 humanos, tanto con síntomas gastrointestinales como sin ellos, así como 347 muestras fecales de diversas especies animales sin síntomas, que incluyeron perros, cerdos, gatos, bovinos, cabras y ovejas en los Departamentos de Cundinamarca y Boyacá. Todas las muestras se transportaron inmediatamente en recipientes estériles al laboratorio de procesamiento, tras la disrupción mecánica, se procedió a la extracción de ADN y a la PCR dirigida al gen 16S-rRNA y al gen de la toxina alfa (cpa) para la detección de C. perfringens. Los resultadosrevelaron una frecuencia de detección de C. perfringens en todos los animales del 22,1% (n=77/347), siendo más frecuente en gatos (34,1%), perros (30,0%) y cerdos (22,0%). Por otro lado, los rumiantes presentaron una frecuencia de detección menor (<11,1%). La mayor frecuencia de detección en animales domésticos podría estar relacionada con factores predisponentes, como la alteración de la microbiota intestinal debido a dietas ricas en proteínas y carbohidratos, o la posible infección entérica por otros patógenos. En cuanto a los humanos, la frecuencia general fue del 19,3% (n=22/114) con un 21,2% en individuos asintomáticos y un 16,6% en sintomáticos. Se observó una asociación entre la presencia de la toxina CPB2 y la diabetes, lo que sugiere interacciones con el sistema inmunológico del huésped. Este estudio resalta una frecuencia de detección elevada de C. perfringens en comparación con otros países, además de una asociación entre la diabetes o enfermedad autoinmune y el transporte de la C. perfringens toxigénico. Estos resultados fueron consolidados en dos artículos científicos derivados de este capítulo. Artículo 1: Camargo A, Páez-Triana L, Camargo D, García-Corredor D, Pulido-Medellín M, Camargo M, Ramírez J.D. and Muñoz M*. Carriage of Clostridium perfringens in Domestic and Farm Animals across the Central Highlands of Colombia: Implications for Gut Health and Zoonotic Transmission. Vet Res Commun. 2024. Artículo 2: Camargo A., Bohórquez L., López D., Ferrebuz-Cardozo A., Castellanos-Rozo J., Díaz J., Rada M., Camargo M., Ramírez J. D. and Muñoz M. Clostridium perfringens in central Colombia: Frequency, Toxin Genes, and Risk Factors. (SOMETIDO en Gut Pathogens) 32 CAPÍTULO II: Describir la estructura genética y factores de virulencia de C. perfringens a partir de genomas públicos. El Capítulo II describe la estructura genética poblacional y los factores de virulencia de C. perfringens utilizando genomas públicos. Para esto, se descargaron 370 genomas de C. perfringens de bases de datos públicas, además se incluyeron 2 genomas provenientes de Chile secuenciados en este estudio. Mediante herramientas bioinformáticas, se realizó el ensamblaje de los datos y se llevó a cabo un análisis de MLST para tipificar los aislamientos, identificar complejos clonales e inferir posibles rutas de dispersión. Se elaboró un análisis de pangenoma y se construyó un árbol filogenético basado en el alineamiento completo del genoma central para definir grupos filogenéticos. Asimismo, se identificaron factores clave de virulencia, toxinotipos y MRA. La mayoría de los genomas reportados tuvieron su origen geográfico en países desarrollados como EE. UU., Francia y China, y en gran parte correspondieron a fuentes de alimentos, aves y humanos. El 53% de los genomas se asignaron al toxinotipo A, seguido del F (32%) y G (7%). Se identificaron toxinotipos circulantes en humanos poco comunes, como los es el tipo G asociado con enfermedades en aves, junto con la presencia de genomas del toxinotipo D y E vinculados a enfermedades en rumiantes. Mediante el análisis de MLST y el alineamiento del genoma central, se identificaron 5 filogrupos. Se encontraron factores de virulencia como fosfolipasa C (plc), sialidasa (nanH), alfa-clostripaína (ccp) y colagenasa (colA) implicados en el daño a la membrana celular endotelial, y un mayor número de MRA a tetraciclinas, macrólidos y aminoglucósidos. El hallazgo en humanos de los toxinotipos D, E y G, los cuales generalmente se asocian con enfermedades en animales, junto con los resultados del análisis filogenético basado en MLST y en el alineamiento del genoma central que evidenció una estrecha relación entre aislamientos de diferentes hospederos, señala la adaptabilidad de C. perfringens a diferentes hospederos y resalta su potencial zoonótico. Además, el alto contenido de factores de virulencia y de MRA en los genomas de C. perfringens destacan su capacidad para adquirir y transmitir estos elementos, planteando preocupaciones sobre el tratamiento antibiótico eficaz. Por ende, se enfatiza la necesidad de una vigilancia continua para implementar medidas preventivas efectivas. Los hallazgos de este capítulo resaltan además la importancia de fortalecer la vigilancia epidemiológica y mejorar la recopilación de datos genómicos en países en vías de desarrollo a través de un esfuerzo conjunto para ampliar la disponibilidad de información acerca de la arquitectura genómica de microorganismos de interés para la salud humana y animal. Como producto de este capítulo se adjunta el siguiente artículo científico: Artículo 3: Camargo A., Guerrero-Araya E, Castañeda S, Vega L, Cardenas-Alvarez MX, Rodríguez C, Paredes-Sabja D, Ramírez JD, Muñoz M. Intra-species diversity of Clostridium perfringens: A diverse genetic repertoire reveals its pathogenic potential. Front Microbiol. 2022 Jul 22; 13:952081. doi: 10.3389/fmicb.2022.952081. PMID: 35935202; PMCID: PMC9354469 33 CAPÍTULO III: Caracterizar la arquitectura genómica y perfiles fenotípicos de hemolisis, citotoxicidad, esporulación y susceptibilidad a antibióticos de aislamientos colombianos de C. perfringens. El objetivo del Capítulo III fue caracterizar la arquitectura genómica y los perfiles fenotípicos de aislamientos colombianos de C. perfringens, obtenidos a partir de muestras positivas de humanos y animales colectadas en el Capítulo I. Se cultivaron un total de 55 muestras positivas para C. perfringens en agar selectivo Triptosa Sulfito Cicloserina (TSC), obteniendo entre dos y cinco colonias por muestra, para un total de 185 aislamientos. El ADN extraído fue secuenciado mediante la plataforma Illumina. A través de análisis bioinformáticos, se realizaron análisis genómicos microgeográficos utilizando los 185 aislamientos colombianos. Además, se realizó un análisis comparativo a nivel global que incluyó 617 genomas desreplicados, es decir, se incorporaron genomas representativos y de mayor calidad, de los cuales 71 fueron colombianos y 546 de acceso público. Se identificaron rutas de dispersión, marcadores de virulencia y MRA a nivel genómico, y se llevaron a cabo pruebas fenotípicas de hemólisis, citotoxicidad, esporulación y susceptibilidad a antibióticos en un grupo representativo de aislamientos. El criterio de selección de los aislamientos se basó en su capacidad para producir la toxina PFOA, una toxina formadora de poros que juega un papel crucial en la patogenia intestinal, con el objetivo de evaluar su impacto biológico. El análisis de la estructura poblacional de C. perfringens a nivel microgeográfico reveló una alta diversidad genética y sugirió eventos de dispersión zoonótica entre caninos, felinos y humanos, destacando la necesidad de mejorar las prácticas de higiene efectivas para prevenir la transmisión comunitaria. Además, la comparación con genomas globales mostró una agrupación de aislamientos de fuentes humanas y alimenticias, resaltando el papel de C. perfringens en las ETA y la necesidad de garantizar una manipulación segura de alimentos para evitar la propagación de esta bacteria. Asimismo, en individuos asintomáticos se detectaron aislamientos de C. perfringens portadores de pfoA, que a nivel fenotípico indujeron hemolisis completa, inhibición del crecimiento celular y mayor eficiencia de esporulación. Estos hallazgos subrayan la importancia de comprender mejor el impacto de las toxinas accesorias y el potencial riesgo de transmisión de aislamientos toxigénicos que puedan desencadenar la enfermedad a nivel comunitario. Por último, la vigilancia genómica y las pruebas fenotípicas revelaron una alta prevalencia de MRA y una susceptibilidad reducida a gentamicina, eritromicina, metronidazol y tetraciclina. Estos resultados resaltan la importancia deajustar las estrategias terapéuticas según las características locales y enfatizan la necesidad de realizar más estudios epidemiológicos para abordar y comprender mejor la resistencia a los antimicrobianos en C. perfringens. Los resultados del análisis genómico y fenotípico de aislamientos colombianos de C. perfringens se encuentran depositados en el siguiente artículo: 34 Artículo 4: Camargo A., Bohorquez L., Cáceres T., Ferrebuz-Cardozo A, Díaz J, Castellanos-Rozo J, Diaz J., Kiu R., Hall L. J., Rámirez J. D. and Muñoz M. Insights into Clostridium perfringens Dispersal Hotspots, Toxins, and Virulence Factors through Integrated Genomic and Phenotypic Profiling (EN CONSTRUCCIÓN). 1 Carriage of Clostridium perfringens in Domestic and Farm Animals across the Central 1 Highlands of Colombia: Implications for Gut Health and Zoonotic Transmission 2 3 Anny Camargo1,2, Luisa Páez-Triana1, Diego Camargo1, Diego García-Corredor3, Martin 4 Pulido-Medellín3, Milena Camargo1,4, Juan David Ramírez1,5 and Marina Muñoz1,6*. 5 6 1 Centro de Investigaciones en Microbiología y Biotecnología-UR (CIMBIUR), Facultad de 7 Ciencias Naturales, Universidad del Rosario, Bogotá, Colombia. 8 2 Universidad de Boyacá, Tunja, Colombia. 9 3 Grupo de Investigación en Medicina Veterinaria y Zootecnia (GIDIMEVETZ), Universidad 10 Pedagógica y Tecnológica de Colombia (UPTC), Tunja, Colombia. 11 4 Centro de Tecnología en Salud (CETESA), Innovaseq SAS, Funza, Cundinamarca, 12 Colombia. 13 5 Molecular Microbiology Laboratory, Department of Pathology, Molecular and Cell-Based 14 Medicine, Icahn School of Medicine at Mount Sinai, New York city, NY 10029, USA 15 6 Molecular Epidemiology Laboratory, Biotechnology Institute, Universidad Nacional de 16 Colombia, Bogotá, Colombia 17 * Corresponding autor: claudia.munoz@urosario.edu.co 18 Carrera 26 # 63B - 48, Bogotá – Colombia 19 (+57-1) 297 0200 Ext. 3359 20 21 2 Abstract 22 Clostridium perfringens inhabits the guts of humans and animal species. C. perfringens can 23 proliferate and express an arsenal of toxins, promoting the development of multiple gut 24 illnesses. Healthy animals carrying C. perfringens represents a risk of transmission to other 25 animals or humans through close contact and an increased likelihood of acquisition of toxin 26 plasmids. The aim of this study was to evaluate the frequency of C. perfringens carriage in 27 domestic and farm animals in the central highlands of Colombia. 28 C. perfringens was detected in six animal species using PCR targeting alpha toxin (cpa) and 29 16S ribosomal RNA (16S-rRNA) genes from 347 fecal samples collected in two 30 Departments: 177 from farm animals of Boyacá and 170 from domestic animals of both 31 Cundinamarca and Boyacá. 32 The overall frequency of C. perfringens detection was 22.1% (n=77/347), with the highest 33 frequency observed in cats 34.2% (n=41/120), followed by dogs 30.0% (n=15/50). The 34 lowest frequency was detected in ruminants: goats 11.1% (n=3/27), sheep 8.0% (n=4/50) and 35 cattle 6.0% (n=6/50). 36 Domestic animals showed a higher frequency of C. perfringens carriage than farm animals. 37 This difference could be associated with dietary patterns, as domestic animals have diets rich 38 in proteins and carbohydrates, while ruminants have low-carbohydrate diets, resulting in high 39 production of endopeptidase-type enzymes and differences in pH due to the anatomy of 40 gastrointestinal tract, which can influence bacterial proliferation. These findings indicate a 41 potential risk of transmission of C. perfringens among animals and from animals to humans 42 through close contact. 43 44 3 Keywords 45 Clostridium perfringens; bacterial carriage; healthy animals; domestic animals; farm 46 animals. 47 Introduction 48 Clostridium perfringens is an anaerobic bacterium that can produce a diverse range of toxins 49 (Uzal et al., 2014). The presence of toxin-encoding genes is the basis of the intra-taxa 50 diversity classification scheme for this species, which is supported by seven toxinotypes (A-51 G) of clinical, epidemiological, and diagnostic importance (Kiu & Hall, 2018). 52 In humans and animals, C. perfringens toxinotype A can be a commensal of the intestinal 53 microbiota. However, the loss of homeostasis due to inflammatory diseases, a sudden change 54 to a protein-rich diet, or even enteric infection by other pathogens or systemic diseases are 55 predisposing factors for developing C. perfringens enterotoxemia (Silva & Lobato, 2015; 56 Turk et al., 1992). 57 C. perfringens is an important opportunistic pathogen in domestic and farm animals, with 58 prevalence rates in healthy animals of 76.0% (44/58) in dogs (Marks, Kather, Kass, & Melli, 59 2002), 63.0% in cats (34/54) (Queen, Marks, & Farver, 2012), 47.1% (33/70) in cattle, 58.0% 60 (29/50) in goats, and 65.4% (36/55) in sheep (Hamza, Dorgham, Elhariri, Elhelw, & Ismael, 61 2018) reported in the United States, India, and the United Kingdom. However, there is limited 62 information available on the frequency of carriage of this bacterial pathogen in developing 63 countries such as Colombia. This knowledge gap limits the design of control measures to 64 control the spread of infection among different hosts, representing a potential risk of 65 implications for human and animal health. 66 Therefore, this study aims to evaluate the frequency of C. perfringens in apparently healthy 67 domestic and farm animals by molecular testing. This work provides an accurate description 68 4 of the epidemiological overview of this bacterium in six animal species (cattle, goats, sheep, 69 pigs, canines, and felines) sampled in the departments of Boyacá and Cundinamarca in the 70 central highlands of Colombia. 71 Materials and methods 72 Study population 73 For this purpose, a total of 347 stool samples were collected from apparently healthy 74 domestic and farm animals in two departments of Colombia: i) Boyacá department, located 75 in the central-eastern part of the country, which has a large rural area with farming 76 communities focused on livestock and food production, and ii) Cundinamarca department, 77 specifically the city of Bogotá, the capital of Colombia, characterized for being a large, highly 78 populated urban area (see Supplementary fig. 1). Fecal sampling was distributed as follows: 79 120 cats (60 from Boyacá and 60 from Cundinamarca), 50 dogs (25 from Boyacá and 25 80 from Cundinamarca), and all other animal species were sampled exclusively from Boyacá 81 department, as follows: 50 cows, 50 pigs, 50 sheep and 27 goats. All fecal samples were 82 collected in sterile recipients with airtight seal (to avoid direct exposure to oxygen) and 83 without transport medium for preservation. Samples were stored frozen -30°C until 84 processing. This study was approved by the Research Ethics Committee of Universidad del 85 Rosario (CEI-UR). 86 DNA extraction from feces and molecular detection of C. perfringens 87 DNA extraction was conducted using the Stool DNA Isolation Kit (Norgen), following the 88 manufacturer's instructions. DNA concentration was measured using the 89 NanoDrop/2000/2000c spectrophotometer (Thermo Fisher Scientific, Massachusetts, USA). 90 Subsequently, C. perfringens carriage was detected through conventional PCR, utilizing two 91 primer sets targeted specific genes: i) 16S-rRNA gene, and ii) cpa gene. Both PCR reactions 92 5 were carried out under conditions that have been reported previously (Roussan, Shaheen, 93 Totanji, Khawaldeh, & Al Rifai, 2012). The positive control used was the C. perfringens 94 ATCC® strain 13124, while primers targeting the V4 hypervariable region of the bacterial- 95 and Archaea-specific 16S-rRNA marker (Caporaso et al., 2011) were employed as an internal 96amplification control to ensure that negative samples were not inhibited. Presence of an 97 amplified product for the two molecular markers, 481 base pairs (bp) for 16S-rRNA and 400 98 bp for cpa, was considered a positive sample for C. perfringens. 99 Statistical analysis 100 Quantitative variables were described in terms of percentages and were analyzed using a Chi 101 square test or Fisher's exact test depending to the observed frequencies. Binomial logistic 102 regression models were carried out to evaluate a possible association between the collected 103 variables (origin, animal use, and animal species) and the identification of C. perfringens. 104 All analyses were two-tailed with a statistical significance value of <0.05, using the 105 STATA17® program. 106 107 Results 108 The results showed an overall frequency of C. perfringens in all animals of 22.1% (n=77). 109 The frequency for domestic animals was higher (32.9%; 25.9-40.5 95%CI) compared to farm 110 animals (11.6%; 7.4-17.5 95%CI), indicating a statistically significant difference (p= 0.001, 111 Chi2 tests). 112 At the species level, the frequency of C. perfringens was higher in cats (34.2%; 25.7-43.3 113 95%CI) and dogs (30.0%; 17.8-44.6 95%CI), compared to cows, that presented the lowest 114 frequency of C. perfringens (6.0%; 1.2-16.5% 95%CI). This difference was statistically 115 significant (p=0.001, Fisher's exact test) (Fig. 1A). 116 6 Additionally, the frequency of C. perfringens for domestic animals according to the 117 geographical origin (Boyacá and Cundinamarca with 25 dogs and 60 cats for each 118 department), did not show significant differences for dogs (p= 0.75, Chi2 tests). However, 119 for cats, a significantly higher carriage was found in Cundinamarca (50.0%; 36.8-63.2 120 95%CI) compared to those from Boyacá (18.3%; 9.5-30.4 95%CI) (Fig. 1B) 121 A logistic regression analysis was carried out to establish the strength of association between 122 the characteristics of the population analyzed (origin and use of the animal) and the 123 identification of C. perfringens in the population. The results showed that domestic animals 124 had a positive association (aOR: 2.77; 1.36-5.61 95.0% CI) in C. perfringens identification 125 with respect to farm animals. In contrast, a lower probability of carriage was found in animals 126 from Boyacá (aOR: 0.38; 0.19-0.73 95.0% CI) (Table 1). 127 Finally, the strength of association between C. perfringens presence and animal species was 128 evaluated (Table 1). The results showed that pigs (aOR: 4.55; 1.18-17.50 95%CI), cats (aOR: 129 4.96; 1.37-17.87 95%CI), and dogs (aOR: 4.05; 1.02-16.01 95%CI) were the animal species 130 that showed a positive association with bacterial detection. 131 Discussion 132 These findings provide the first epidemiological overview of frequency of C. perfringens 133 carriage in several animal species in Colombia. The study detected a high frequency of this 134 pathogenic bacterium in domestic animals (Fig. 1A), consistent with research conducted by 135 other groups in Brazil where detection frequencies in animals without diarrhea of up to 55.0% 136 in dogs (22/40) (Silva et al., 2013) and 47.9% (112/234) in cats have been reported (Silva et 137 al., 2020). These findings may be related to predisposing factors such as altered intestinal 138 microbiota in these species due to their high protein and carbohydrate diets or potential 139 7 enteric infection by other pathogens (Sabshin et al., 2012; Silva & Lobato, 2015; Turk et al., 140 1992). 141 The higher frequencies of C. perfringens in cats from Cundinamarca compared to those from 142 Boyacá (Fig. 1B) reported here may be due to sociodemographic differences in the collection 143 sites. Samples in Boyacá were collected mostly in rural areas where the animals are free-144 living and acquire food through hunting and/or human food remains. In contrast, the vast 145 majority of samples in Cundinamarca were collected in urban areas where the animals 146 probably have access to processed food rich in carbohydrates and proteins that are broken 147 down by proteinases into peptides and amino acids before being absorbed by the intestinal 148 villi, which are important for bacterial growth (Bermingham, Maclean, Thomas, Cave, & 149 Young, 2017; Huang, Pan, Yang, Bi, & Xiong, 2020). 150 Regarding farm animals, pigs presented a higher carriage of C. perfringens than ruminants 151 (Fig. 1A). These results could be correlated with dietary factors in rural areas of the 152 Department of Boyacá. The vast majority of animals in these areas are fed with food remains 153 that could carry resistant spores from microbial species, fats, and high levels of crude protein. 154 These factors can affect the intestinal microbiota by increasing the availability of nitrogen as 155 well as amino acids, favoring the proliferation of pathogenic bacteria such as C. perfringens 156 (Duarte & Kim, 2022; Yu, Zhu, & Hang, 2019). 157 In contrast, ruminants (cows, sheep, and goats) that are polygastrophic animals showed low 158 C. perfringens carriage (Fig. 1A). The reduced circulation of C. perfringens in ruminants 159 could be due to factors such as diet based on a high content of fodder and pasture that induces 160 a higher transport of Gram-negative bacteria compared to Gram-positive bacteria in the 161 rumen (Plaizier et al., 2018), low consumption of carbohydrates and proteins that provide a 162 nutrient-poor environment for C. perfringens growth, and the consumption of poor quality 163 8 silage that generates excessive accumulation of lactic acid in the digestive system, causing 164 acidosis and alteration of the intestinal microbiota (Cholewińska, Górniak, & Wojnarowski, 165 2021). Additionally, mechanisms that facilitate nutrient digestion and absorption (microbial 166 fermentation versus enzymes in the mammalian intestinal tract), nutrient, as well as the 167 morphology and histology of the gastrointestinal tract may influence discrepancies in C. 168 perfringens carriage between domestic and farm animals such as ruminants 169 (Abdolmohammadi Khiav & Zahmatkesh, 2021; Constable, Hinchcliff, Done, & Grünberg, 170 2016; Nazki et al., 2017). 171 The results presented in this study provide an initial insight into the context of the frequency 172 of C. perfringens carriage in healthy farm and domestic animals in Colombia. The high 173 frequency of domestic animals and pigs carrying C. perfringens, as indicated in this study, 174 highlights the risk of transmission of this bacterium not only between animals, but also from 175 animals to humans through meat consumption or contact with feces of carrier animals that 176 share the same living environment as humans, as indicated above (Song et al., 2013). 177 According to the World Health Organization (WHO) (Narayan, Sinha, & Singh, 2023), 178 zoonoses represent a growing global threat as diseases naturally transmitted from animals to 179 humans. Approximately 60% of infectious agents are pathogenic to humans, 75% of 180 pathogens associated with emerging diseases are zoonotic, 25% of zoonoses originate in 181 domestic animals, and according to etiology bacteria account for the majority of zoonotic 182 diseases (Rahman et al., 2020). 183 This makes pathogen surveillance through molecular epidemiology, especially in domestic 184 animals such as dogs and cats, crucial because of their critical role in transmission and 185 increased interaction with humans (Meyer, Gastmeier, Kola, & Schwab, 2012; Tomori & 186 Oluwayelu, 2023). 187 9 Furthermore, it is relevant to note that low- and middle-income countries, particularly in rural 188 areas with closer contact with domestic and farm animals, face additional challenges. These 189 challenges include lack of access to safe drinking water, inadequatefood refrigeration and 190 environmental conditions conducive to the transmission of bacteria such as C. perfringens. 191 These regions not only face higher zoonosis burdens, but also have limited capacity to 192 effectively address disease risk (Worsley-Tonks et al., 2022). 193 This, highlighting the urgent need for awareness and focus on One Health to protect both 194 animal and human health, including improved surveillance of potential pathogen transport in 195 animal hosts (Qiu et al., 2023; Worsley-Tonks et al., 2022). 196 Finally, it is worth noting that most studies on the biology and genomics of pathogens such 197 as C. perfringens focus on isolating strains from diseased individuals. However, studying the 198 evolutionary adaptation of strains reveals the role of C. perfringens in the healthy gut 199 microbiota, emphasizing the need to develop studies that investigate genome-scale metabolic 200 patterns and the differential presence of toxin-bearing strains and other virulence factors in 201 the gut of healthy animals. 202 Acknowledgments 203 We would like to thank Minciencias, Universidad de Boyacá, Universidad del Rosario and 204 Universidad Pedagógica y Tecnológica in Colombia for their support. The authors are 205 thankful to the school principal Francisco Lara Rodríguez, teacher Orlando Camargo 206 Camargo and students of Institución Educativa Técnica Pecuaria de Saza – Gámeza, we thank 207 the Fundación Dejando Huella and Dr. Fabian Danilo Reyes Secretario de Agricultura y 208 Desarrollo Rural de Paipa – Boyacá for helping in collection of samples, and Emanuella and 209 Mariana, for helping in processing samples. Anny Camargo and Diego Camargo would like 210 to thank engineer María Magdalena Mancipe Núñez, their mother, for supporting their 211 10 training as physicians and for being an example of tenacity, resilience, and courage in the 212 complex path of life. 213 Funding 214 This research was funded by the Ministerio de Ciencia Tecnología e Innovación 215 (Minciencias), Universidad de Boyacá, Universidad del Rosario and Universidad Pedagógica 216 y Tecnológica de Colombia within the framework of the project “Identificación y 217 caracterización de Clostridium perfringens circulante en humanos, animales y alimentos en 218 el Departamento de Boyacá – Colombia”, Code: 722289684653 Contract No. 613-2021. This 219 work was supported by Dirección de Investigación e Innovación from Universidad del 220 Rosario. 221 Author contributions 222 A.C., J.D.R. and M.M. designed the study and drafted the manuscript. A.C. and L.P. carried 223 out the processing of the samples. A.C, M.C and M.M. performed the epidemiological 224 analyses. A.C, D.C, L.P, D.G.C and M.P.M, helped in collection of samples. J.D.R. and M.M. 225 revised the manuscript. All authors read and approved the final version of the manuscript. 226 227 Conflict of interest 228 The authors declare that they have no conflict of interest. 229 230 11 References 231 Abdolmohammadi Khiav, L., & Zahmatkesh, A. J. T. A. H. (2021). Vaccination against 232 pathogenic clostridia in animals: A review. 53(2), 284. 233 https://doi.org/10.3382/ps.2012-02368 234 Bermingham, E. N., Maclean, P., Thomas, D. G., Cave, N. J., & Young, W. J. P. (2017). Key 235 bacterial families (Clostridiaceae, Erysipelotrichaceae and Bacteroidaceae) are 236 related to the digestion of protein and energy in dogs. 5, e3019. 237 https://doi.org/10.7717/peerj.3019 238 Caporaso, J. G., Lauber, C. L., Walters, W. A., Berg-Lyons, D., Lozupone, C. A., Turnbaugh, 239 P. J.,Knight, R. J.(2011). Global patterns of 16S rRNA diversity at a depth of millions 240 of sequences per sample. 108(supplement_1), 4516-4522. 241 https://doi.org/10.1073/pnas.1000080107 242 Cholewińska, P., Górniak, W., & Wojnarowski, K. J. B. v. r. (2021). Impact of selected 243 environmental factors on microbiome of the digestive tract of ruminants. 17(1), 1-10. 244 doi: 10.1186/s12917-021-02742-y. https://doi: 10.1186/s12917-021-02742-y. 245 Constable, P. D., Hinchcliff, K. W., Done, S. H., & Grünberg, W. (2016). Veterinary 246 medicine: a textbook of the diseases of cattle, horses, sheep, pigs and goats: Elsevier 247 Health Sciences. 248 Duarte, M. E., & Kim, S. W. J. A. N. (2022). Intestinal microbiota and its interaction to 249 intestinal health in nursery pigs. 8, 169-184.https://doi:10.1016/j.aninu.2021.05.001. 250 Hamza, D., Dorgham, S. M., Elhariri, M., Elhelw, R., & Ismael, E. J. J. o. V. R. (2018). New 251 insight of apparently healthy animals as a potential reservoir for Clostridium 252 perfringens: A public health implication. 62(4), 457. https://doi:10.2478/jvetres-253 2018-0073. 254 Huang, Z., Pan, Z., Yang, R., Bi, Y., & Xiong, X. J. G. M. (2020). The canine gastrointestinal 255 microbiota: early studies and research frontiers. 11(4), 635-654. https://doi: 256 10.1080/19490976.2019.1704142 257 Kiu, R., & Hall, L. J. (2018). An update on the human and animal enteric pathogen 258 Clostridium perfringens. 7(1), 1-15. https://doi:10.1038/s41426-018-0144-8. 259 Marks, S. L., Kather, E. J., Kass, P. H., & Melli, A. C. J. J. o. v. i. m. (2002). Genotypic and 260 phenotypic characterization of Clostridium perfringens and Clostridium difficile in 261 diarrheic and healthy dogs. 16(5), 533-540. https://doi:10.1892/0891-262 6640(2002)016<0533:gapcop>2.3.co;2 263 Meyer, E., Gastmeier, P., Kola, A., & Schwab, F. J. I. (2012). Pet animals and foreign travel 264 are risk factors for colonisation with extended-spectrum β-lactamase-producing 265 Escherichia coli. 40, 685-687. https://doi:10.1007/s15010-012-0324-8 266 Narayan, K. G., Sinha, D. K., & Singh, D. K. (2023). Zoonoses. In Veterinary Public Health 267 & Epidemiology: Veterinary Public Health-Epidemiology-Zoonosis-One Health (pp. 268 21-33): Springer. 269 Nazki, S., Wani, S. A., Parveen, R., Ahangar, S. A., Kashoo, Z. A., Hamid, S., . . . Dar, P. A. 270 J. V. W. (2017). Isolation, molecular characterization and prevalence of Clostridium 271 perfringens in sheep and goats of Kashmir Himalayas, India. 10(12), 1501. 272 https://doi:10.14202/vetworld.2017.1501-1507 273 Plaizier, J., Mesgaran, M. D., Derakhshani, H., Golder, H., Khafipour, E., Kleen, J., . . . 274 Zebeli, Q. J. A. (2018). Enhancing gastrointestinal health in dairy cows. 12(s2), s399-275 s418. https://doi:10.1017/S1751731118001921 276 12 Qiu, Y., Guitian, J., Webster, J. P., Musallam, I., Haider, N., Drewe, J. A., & Song, J. J. P. 277 T. o. t. R. S. B. (2023). Global prioritization of endemic zoonotic diseases for 278 conducting surveillance in domestic animals to protect public health. 378(1887), 279 20220407. https://doi:10.1098/rstb.2022.0407 280 Queen, E., Marks, S. L., & Farver, T. B. J. J. o. V. I. M. (2012). Prevalence of selected 281 bacterial and parasitic agents in feces from diarrheic and healthy control cats from 282 Northern California. 26(1), 54-60. https://doi:10.1111/j.1939-1676.2011.00843.x 283 Rahman, M. T., Sobur, M. A., Islam, M. S., Ievy, S., Hossain, M. J., El Zowalaty, M. E., . . 284 . Ashour, H. M. J. M. (2020). Zoonotic diseases: etiology, impact, and control. 8(9), 285 1405. https://doi:10.3390/microorganisms8091405 286 Roussan, D., Shaheen, I. A., Totanji, W., Khawaldeh, G., & Al Rifai, R. J. P. s. (2012). 287 Simultaneous detection of Clostridium perfringens and Clostridium colinum by 288 duplex-polymerase chain reaction. 91(12), 3080-3085. https://doi:10.3382/ps.2012-289 02368 290 Sabshin, S. J., Levy, J. K., Tupler, T., Tucker, S. J., Greiner, E. C., & Leutenegger, C. M. J. 291 J. o. t. A. V. M. A. (2012). Enteropathogens identified in cats entering a Florida 292 animal shelter with normal feces or diarrhea. 241(3), 331-337. 293 https://doi:10.2460/javma.241.3.331 294 Silva, R. O. S., & Lobato, F. C. F. J. A. (2015). Clostridium perfringens: A review of enteric 295 diseasesin dogs, cats and wild animals. 33, 14-17. 296 https://doi:10.1016/j.anaerobe.2015.01.006 297 Silva, R. O. S., Ribeiro, M. G., de Paula, C. L., Pires, I. H., Junior, C. A. O., Diniz, A. N., . . 298 . Lobato, F. C. F. J. A. (2020). Isolation of Clostridium perfringens and Clostridioides 299 difficile in diarrheic and nondiarrheic cats. 62, 102164. 300 https://doi:10.1016/j.anaerobe.2020.102164. 301 Silva, R. O. S., Santos, R. L. R., Pires, P. S., Pereira, L. C., Pereira, S. T., Duarte, M. C., 302 Lobato, F. C. F. J. B. J. o. M. (2013). Detection of toxins A/B and isolation of 303 Clostridium difficile and Clostridium perfringens from dogs in Minas Gerais, Brazil. 304 44, 133-137. https://doi:10.1590/S1517-83822013005000008 305 Song, S. J., Lauber, C., Costello, E. K., Lozupone, C. A., Humphrey, G., Berg-Lyons, D., . . 306 . Nakielny, S. J. e. (2013). Cohabiting family members share microbiota with one 307 another and with their dogs. 2, e00458. https://doi: 10.7554/eLife.00458. 308 Tomori, O., & Oluwayelu, D. O. J. A. R. o. A. B. (2023). Domestic animals as potential 309 reservoirs of zoonotic viral diseases. 11, 33-55. https://doi:10.1146/annurev-animal-310 062922-060125 311 Turk, J., Fales, W., Miller, M., Pace, L., Fischer, J., Johnson, G., . . . Rottinghaus, A. J. J. o. 312 t. A. V. M. A. (1992). Enteric Clostridium perfringens infection associated with 313 parvoviral enteritis in dogs: 74 cases (1987-1990). 200(7), 991-994.PMID: 1315727. 314 Uzal, F. A., Freedman, J. C., Shrestha, A., Theoret, J. R., Garcia, J., Awad, M. M., McClane, 315 B. A. J. F. m. (2014). Towards an understanding of the role of Clostridium 316 perfringens toxins in human and animal disease. 9(3), 361-377. 317 https://doi:10.2217/fmb.13.168 318 Worsley-Tonks, K. E., Bender, J. B., Deem, S. L., Ferguson, A. W., Fèvre, E. M., Martins, 319 D. J.,Ogada, (2022). Strengthening global health security by improving disease 320 surveillance in remote rural areas of low-income and middle-income countries. 10(4), 321 e579-e584. https://doi:10.1016/S2214-109X(22)00031-6 322 13 Yu, D., Zhu, W., & Hang, S. J. A. (2019). Effects of long-term dietary protein restriction on 323 intestinal morphology, digestive enzymes, gut hormones, and colonic microbiota in 324 pigs. 9(4), 180. https://doi:10.3390/ani9040180. 325 326 14 Figure legends 327 Figure 1. Carriage of C. perfringens in animal species from central Colombia 328 (Cundinamarca and Boyacá). A. Overall carriage of C. perfringens in cats (n=120), dogs 329 (n=50), pigs (n=50), goats (n=27), sheep (n=50), cattle (n=50). B. Frequency of carriage in 330 domestic animals sampled in the departments of Boyacá (cats n=25 and dogs n=25) and 331 Cundinamarca (cats n=25 and dogs n=25). 332 333 Tables 334 Table 1. Logistic regression analysis of the association between the features of analyzed 335 population and C. perfringens carriage. 336 337 15 Supplementary Material Legends 338 Supplementary Figure 1. Schematic representation of the map of Colombia showing the 339 departments of Boyacá and Cundinamarca located in the center of the country, where 340 sampling was carried out. 341 1 Micro-Geographical Insights into Clostridium perfringens: Frequency, Toxin-1 Encoding Genes, and Risk Factors in a Colombian Population 2 Anny Camargo1,2, Laura Bohorquez1, Diana Paola López2, Atilio Ferrebuz-Cardozo2, José 3 Castellanos-Rozo2, Javier Díaz2, Mariana Rada2, Milena Camargo1,3, Juan David Ramírez1,4 4 and Marina Muñoz1,5*. 5 6 1 Centro de Investigaciones en Microbiología y Biotecnología-UR (CIMBIUR), Facultad de 7 Ciencias Naturales, Universidad del Rosario, Bogotá, Colombia. 8 2 Universidad de Boyacá, Tunja, Colombia. 9 3 Centro de Tecnología en Salud (CETESA), Innovaseq SAS, Funza, Cundinamarca, 10 Colombia. 11 4 Molecular Microbiology Laboratory, Department of Pathology, Molecular and Cell-Based 12 Medicine, Icahn School of Medicine at Mount Sinai, New York city, NY 10029, USA 13 5 Molecular Epidemiology Laboratory, Institute of Biotechnology, Universidad Nacional de 14 Colombia, Bogotá, Colombia 15 * Corresponding author: 16 Marina Muñoz 17 cmmunozd@unal.edu.co; claudia.munoz@urosario.edu.co 18 19 2 Abstract 20 Clostridium perfringens is an opportunistic bacterium that causes intestinal diseases in both 21 humans and animals. This study aimed to assess the frequency of C. perfringens and the 22 presence of toxin-encoding genes in fecal samples from individuals with or without 23 gastrointestinal symptoms in the Department of Boyacá, Colombia. Additionally, risk factors 24 associated with colonization and disease development were analyzed. 25 A total of 114 stool samples were analyzed using a molecular test based on specific 26 polymerase chain reaction (PCR) targeting 16S-rRNA and alpha toxin (cpa) genes. Bacterial 27 isolates obtained from positive samples were screened for toxin-encoding genes related to 28 gastrointestinal diseases. In addition, sociodemographic, and clinical data from 77 29 individuals were also analyzed. 30 The overall frequency of C. perfringens was 19.3%, with an infection frequency of 16.6% 31 among symptomatic individuals and a colonization frequency of 21.2% among asymptomatic 32 individuals. All 56 isolates obtained carried the cpa gene, while cpb2 was present in 10.7% 33 (n=6/56); cpe and cpb genes were not detected. Notably, diabetes/autoimmune disease 34 demonstrated a significant association with an increased risk of C. perfringens (adjusted OR 35 8.41: 95% CI 1.32-35.89). This study highlights an elevated frequency of C. perfringens and 36 the presence of cpb2 gene in asymptomatic individuals compared with their symptomatic 37 counterparts. 38 These findings offer insights into the distribution and virulence factors of C. perfringens at a 39 micro-geographical level. This information supports the need for developing tailored 40 prevention strategies based on local characteristics to promote active surveillance programs 41 based on molecular epidemiology. 42 3 Key words: Clostridium perfringens, colonization, infection, symptomatic, asymptomatic, 43 toxins. 44 Introduction 45 Clostridium perfringens, an opportunistic enteropathogenic bacterium affecting both humans 46 and animals, has been linked to multiple intestinal and systemic diseases, including food 47 poisoning, antibiotic-associated diarrhea (AAD), and intestinal necrosis (1, 2). Colonization 48 by opportunistic bacterial pathogens, such as C. perfringens, capable of releasing toxins, 49 often precedes infections characterized by disease development (3). In many instances, the 50 factors influencing or mitigating this risk in individual patients remain unclear. Some studies 51 have indicated that colonization by C. perfringens, along host-associated risk factors such as 52 the use of antacids (1) and older age (>50 years) (4), may predispose individuals to infection 53 development. 54 Understanding predisposing host factors in specific contexts is essential for designing 55 tailored preventive and therapeutic strategies, thereby enhancing public health efforts to 56 address the incidence and impact of these diseases more effectively. Similarly, risk factors 57 associated with the microorganism, such as a rapid growth rate, production of high-58 temperature resistant spores, the release of toxins, and antibiotic resistance mechanisms, 59 influence the development of severe infections that are difficult to control (5). 60 In this context, it is known that C. perfringens can codify a diverse array of toxins, including 61 pore-forming toxins such as perfringolysin O (pfoA), enterotoxin (cpe), and necrotic enteritis 62 B-like toxin (netB) (3). Based on toxin production, C. perfringens isolates are classified into 63 7 toxinotypes, labeled A to G, each associated with a specificdisease in a particular host (5, 64 6). The CPA, CPE, CBPB, and CPB2 toxins are linked to gastrointestinal disease (6), 65 highlighting the importance of their routine surveillance and the implementation of improved 66 4 control measures for circulating strains carrying these toxins. Therefore, molecular 67 epidemiology studies could contribute to the understanding of the clinical, 68 sociodemographic, and biological factors that influence the development of diseases caused 69 by C. perfringens and other pathogens. 70 Although molecular epidemiology has explored the detection of C. perfringens in 71 symptomatic individuals in Europe, Asia, and North America, reporting infection frequencies 72 ranging from approximately 5 to 20% in cases of antibiotic-associated diarrhea (AAD) and 73 sporadic non-foodborne diarrhea (1, 7-11), comparative studies between the colonization 74 frequency in asymptomatic individuals and infection among symptomatic patients, as well 75 as information about predisposing factors, especially in developing countries, are limited. 76 In Colombia, a reported 32.7% infection rate was observed in patients with diarrhea, but this 77 data lacks detailed sociodemographic and clinical information, hindering the establishment 78 of significant clinical associations (12). The absence of relevant data poses challenges to 79 effectively contributing to the implementation of public health measures aimed at reducing 80 the disease burden. Despite the significance of C. perfringens in public health due to its 81 extensive toxigenic arsenal and involvement in various diseases, there is a clear lack of 82 epidemiological data on detection frequency, with few studies characterizing this bacterium 83 in asymptomatic patients (approximately 5%) (13, 14). 84 Therefore, we aimed to identify the presence of C. perfringens in fecal samples from 85 individuals with gastrointestinal symptoms and asymptomatic individuals in a central region 86 of Colombia. Additionally, we identified potential circulating toxins and collected relevant 87 sociodemographic and clinical data, aiming to understand the characteristics of the 88 population colonized by C. perfringens. By exploring these dimensions, our goal was to 89 contribute to the understanding of the relationship between the presence of C. perfringens, 90 5 toxin circulation, and population characteristics. The results of this study contribute to the 91 local understanding of the molecular epidemiology of C. perfringens in Colombia. 92 Methods 93 94 Study Population 95 Between May and September 2022, a total of 114 fecal samples were collected from adults 96 aged between 26 and 84 years. These samples were obtained from individuals seeking 97 healthcare services at three hospitals located in the Department of Boyacá, Colombia (Figure 98 supplementary 1). Participant selection included individuals with gastrointestinal symptoms 99 such as diarrhea, abdominal pain, fever, or vomiting, who were referred for stool sample 100 collection due to their symptoms. Asymptomatic participants were also included in the study, 101 selected as part of routine examinations, primarily screening tests (fecal occult blood test) 102 due to risk factors such as advanced age (>60 years), chronic diseases like diabetes, 103 hypertension, hypothyroidism, gastritis, or autoimmune diseases (Supplementary table 1) 104 Comprehensive data were collected from 77 out of the 114 participants, with a few choosing 105 not to disclose information due to privacy concerns or time constraints in responding to the 106 survey. This included sociodemographic details (gender, education level, access to clean 107 water, food refrigeration, presence of animals at home, and age), as well as clinical 108 information such as antibiotic consumption in the last 3 months and the presence of chronic 109 diseases. 110 DNA extraction and molecular detection of C. perfringens 111 DNA extraction was performed from a 300 mg fecal sample aliquot using the Norgen fecal 112 DNA isolation kit following the manufacturer's instructions. DNA concentration was 113 measured with the NanoDrop/2000/2000c spectrophotometer (Thermo Fisher Scientific, 114 Massachusetts, USA). Subsequently, the presence of C. perfringens was detected through 115 6 conventional PCR using two primer sets targeting the genes: i) 16S rRNA gene, and ii) cpa 116 gene. Both PCR reactions were carried out under conditions described in previous studies 117 (15). C. perfringens ATCC® strain 13124 was included as a positive control. The presence 118 of an amplified product for both genes, 16S rRNA (product size 481 bp), and cpa (product 119 size 400 bp), was considered a positive result for C. perfringens. 120 121 Bacterial culture and identification of potential toxins 122 From samples that tested positive for C. perfringens during PCR, a second aliquot of the 123 sample was used for in vitro cultivation on Tryptose Sulfite Cycloserine agar (TSC), 124 incubated for 24 hours at 37°C under anaerobic conditions. Subsequently, 2-5 colonies per 125 sample were selected. Colonies were checked by Gram staining; the biomass was increased 126 on blood agar and bacterial genomic DNA was extracted using Promega's Wizard genomic 127 DNA purification kit. DNA concentration was measured with the NanoDrop/2000/2000c 128 spectrophotometer (Thermo Fisher Scientific, Massachusetts, USA). 129 The presence of genes encoding CPA, CPE, CPB, and CPB2 toxins on C. perfringens isolates 130 was evaluated by PCR using specific primer pairs detailed in Table 1. Amplification 131 reactions were carried out in a 12.5 μL volume, with a template DNA volume of 1.5 μL (100 132 ng/μL), following the previously described conditions (15, 16). PCR products were 133 visualized through electrophoresis on a 1.5% agarose gel with a 1000 bp molecular weight 134 marker. 135 136 Table 1. Primers used in the PCR for the detection of potential toxins from C. perfringens. Toxin Primer name Sequence (5′–3′) Gene Product size (bp) Ref 7 CPA CPAF CPAR TGCATGAGCTTCAATTAGGT TTAGTTTTGCAACCTGCTGT Alpha-toxin 400 (15) CPE CPEL CPER GGGGAACCCTCAGTAGTTTCA ACCAGCTGGATTTGAGTTTAATG Entero-toxin 506 (16) CPB CPBL CPBR TCCTTTCTTGAGGGAGGATAAA TGAACCTCCTATTTTGTATCCCA Beta-toxin 611 (16) CPB2 CPB2L CPB2R CAAGCAATTGGGGGAGTTTA GCAGAATCAGGATTTTGACCA Beta2 - toxin 200 (16) 137 Statistical analysis 138 Given the distribution of the quantitative variable of age, we chose to use the median, and 139 reported its measure of dispersion, the interquartile range (IQR). The qualitative variables 140 (detailed in Supplementary Table 1.), such as sex, presence of animals in the home, access 141 to drinking water, hypertension, gastritis, diabetes, and hypothyroidism, were presented in 142 terms of frequencies and percentages. 143 To evaluate associations between C. perfringens detection, identified toxin, and 144 sociodemographic and clinical variables, conditional logistic regression was employed. The 145 model provided estimates of odds ratios (OR) in crude and adjusted form, accompanied by 146 their 95% confidence intervals (CI). All two-tailed statistical tests were performed using the 147 STATA 14 program, with p values < 0.05 being considered statistically significant. 148 149 Results 150 This study involved the collection and analysis of 114 stool samples from individuals seeking 151 care at three hospital centers in the Department of Boyacá, Colombia. Sociodemographic and 152 clinical data from 77 participants were obtained through a detailed survey (see 153 Supplementary Table 1). 154 8 Globally, the detection frequency of C. perfringens was 19.3% (n=22/114). The metadata 155 for the 77 individuals revealed that 64.9% (n=50/77) were women, 20.7% (n=16/77)had 156 completed secondary education, and the average age was 57.2 years. The infection frequency 157 among symptomatic individuals was 16.6% (n=5/30), while the colonization detection 158 among asymptomatic individuals was 21.2% (n=10/47) (Figure 1A). 159 Among the infected individuals, the average age was 56.6 years, 80.0% (n=4/5) were women, 160 and all were over 50 years old. Additionally, 40.0% (n=2/5) had consumed antibiotics in the 161 last 3 months, lived with animals, and had hypertension as the most common comorbidity. 162 Colonized individuals had an average age of 62.1 years, with 40% being women (n=4/10). 163 Additionally, 20% (n=2/10) had taken antibiotics in the last 3 months, and 40% (n=4/10) 164 lived with animals. Associated comorbidities included hypertension (50%, n=5/10), diabetes 165 (30%, n=3/10), hypercholesterolemia (20%, n=2/10), and gastritis or hypothyroidism (10%, 166 n=1/10). 167 The analysis of associated comorbidities in the population revealed an increased odds ratio 168 for Diabetes/Autoimmune Disease (adjusted OR 8.41: 95% CI 1.32-35.89) (Figure 1B, 169 Supplementary Table 2). No statistically significant associations were found between the 170 variables and the group of patients infected versus colonized. 171 A total of 56 isolates of C. perfringens were obtained from 15 stool samples. The cpa gene 172 was detected in all isolates, while cpb2 was present in 10.7% (n=6/56); the cpe and cpb genes 173 were not detected. Individuals carrying the CPB2 toxin showed no gastrointestinal 174 symptoms. Importantly, a strong association was found between the detection of the CPB2 175 toxin and Diabetes/Autoimmune Disease (adjusted OR 27.52: 95% IC 1.68-50.67). 176 Discussion 177 9 The human intestine harbors a wide diversity of microorganisms and symbiotic bacteria that 178 play an essential role in promoting immune responses and preventing the colonization of 179 opportunistic pathogens (17). However, the alteration of the microbiota, caused by 180 environmental and genetic factors, creates a conducive environment for opportunistic 181 pathogens such as C. perfringens to develop survival strategies detrimental to the host's 182 health. This includes colonization by C. perfringens strains carrying cytotoxic toxins (18, 19) 183 acquired through zoonotic transmission, consumption of contaminated food, or person-to-184 person transmission. Additionally, gut-resident toxinotype A strains of C. perfringens can 185 acquire other virulence factors through mobile genetic elements and release toxins. This 186 process triggers inflammatory intestinal and tissue diseases, underscoring the critical 187 importance of preventive strategies. 188 In this context, understanding the local epidemiology of these pathogens through micro-189 geographical studies is crucial for anticipating and effectively addressing public health 190 threats. Therefore, comparing the detection frequencies of C. perfringens with studies 191 conducted in other countries highlights variability, emphasizing the importance of assessing 192 epidemiology locally. 193 While the infection frequencies reported in this study (16.6%) (See Figure 1A), are lower 194 than those found in patients with diarrhea in Bogotá, Colombia (32.7%) (12), they fall within 195 the range reported in other regions of the world. In Europe and Asia, the frequencies of C. 196 perfringens infection range from 5% to 20% in cases of antibiotic-associated diarrhea (AAD) 197 and sporadic non-foodborne diarrhea (1, 7, 8). In the United States, C. perfringens ranks as 198 the second leading cause of foodborne bacterial illnesses (11, 20). 199 Despite the infection percentage in Colombia being high compared to studies in countries 200 such as Germany (4.1%) (21) and India (8.6%) (22), it is similar to infection frequencies in 201 10 China (13.8%) (23) and Iran (22.4%) (8). These data significantly emphasize the relevant 202 presence of C. perfringens in the central region of Colombia, highlighting the pressing need 203 to consider the local context when interpreting detection frequencies in epidemiological 204 studies. The variability in these data among different regions could be attributed to factors 205 such as hygiene practices, environmental conditions, and demographic characteristics unique 206 to each region, emphasizing the complexity of the epidemiology of C. perfringens. 207 Regarding the frequency of colonization in asymptomatic individuals, our study revealed a 208 rate of 21.2% (Figure 1A), which is lower than that reported in healthy individuals in 209 Germany (40.0% - 20/50), North America (51.0% - 22/43) (24), and northern Mexico (63.0% 210 - 126/200) (25). The low colonization frequency reported here, compared to studies in other 211 countries, could be attributed to differences in innate and acquired immune responses and to 212 determinants of the host's genetic susceptibility to specific enteric infections (26). 213 The analysis of colonization frequency in asymptomatic individuals provides a unique 214 perspective that underscores the imminent risk this population faces when exposed to virulent 215 strains. Similarly, the higher frequency of colonization observed in this study compared to 216 infection emphasizes the significance of these findings. This is particularly crucial when 217 considering that colonized patients have a significantly greater propensity to experience 218 episodes of invasive infection compared to non-colonized individuals. 219 Furthermore, future research should take into account recent information from another study 220 suggesting that the presence of C. perfringens at the intestinal level in asymptomatic 221 individuals can trigger brain inflammation, oxidative stress, apoptosis, and cell damage (27). 222 This aspect becomes particularly relevant when exploring different age groups, especially in 223 the context of the presence of metabolic disorders. 224 11 The significant association between the detection of the CPB2 toxin of C. perfringens and 225 Diabetes/Autoimmune Disease (adjusted OR 27.52: 95% IC 1.68-50.67) reveals possible 226 pathways of interaction between the microorganism and the host's immune system. The 227 hypersensitivity of diabetic tissues to colonization factors due to low levels of UDP-glucose 228 (28) and the cytotoxic effects of the CPA and CPB2 toxins (29), raise questions about the 229 specific mechanisms involved, which should be studied. 230 Although we did not find other statistically significant associations, the most common 231 underlying comorbidities in patients where C. perfringens was detected were cardiovascular 232 diseases, diabetes, and/or hypercholesterolemia—findings similar to other studies (8, 23). 233 Probably, the dysfunction of the reticuloendothelial system in cardiovascular diseases and 234 high cholesterol levels, precursors to the production of bile acids important for spore 235 germination, may play a key role in colonization and infection by C. perfringens (30). These 236 data provide valuable information for guiding preventive measures in the colonized 237 population. 238 The identification of some limitations in our work, such as the sample size and specific 239 microgeographic representation, acknowledges the need for broader and more diversified 240 future investigations. Exploring additional factors that may predispose to symptomatic 241 disease and the genomic and phenotypic characterization of virulent strains carrying toxins 242 like pfoA and hypovirulent or "commensal-like" strains that significantly encoded fewer 243 virulence factors and plasmids, underscores research areas that can deepen our understanding 244 of microbial interactions and their clinical consequences. 245 In conclusion, this study explores the results within an internationalcontext, delving into 246 factors associated with patients experiencing intestinal symptoms. It underscores the 247 significance of colonization in individuals without intestinal symptoms. Moreover, it 248 12 indicates future directions, emphasizing the necessity to investigate complex microbial 249 interactions in the human intestine and to consider specific preventive measures for the at-250 risk population. 251 The presented findings contribute to the local understanding of C. perfringens epidemiology 252 in Colombia, offering valuable insights for public health management and prevention 253 strategies in the central region of the country. 254 255 Declarations 256 Ethical Considerations 257 This study was conducted with the approval of the Research Ethics Committee of the 258 Universidad del Rosario (Approval Law No. 339). The study was deemed low risk according 259 to Resolution 8430 of 1993 from the Ministry of Health of Colombia. Samples were coded 260 to protect the identity of the participants following national ethical guidelines and the 261 Declaration of Helsinki. Informed consent was obtained for the use of samples in research, 262 in accordance with the committee's authorization. Data and information about individuals 263 were collected through a data collection instrument endorsed by the ethics committee. 264 265 Conflict of interest 266 The authors declare that they have no conflict of interest. 267 268 Funding 269 This research was funded by the Ministerio de Ciencia Tecnología e Innovación 270 (Minciencias), Universidad de Boyacá, Universidad del Rosario and Universidad Pedagógica 271 y Tecnológica de Colombia within the framework of the project “Identificación y 272 13 caracterización de Clostridium perfringens circulante en humanos, animales y alimentos en 273 el Departamento de Boyacá – Colombia”, code: 722289684653 contract No. 613-2021. This 274 work was supported by Dirección de Investigación e Innovación from Universidad del 275 Rosario. 276 277 Authors' contributions 278 A.C., J.D.R. and M.M. designed the study and drafted the manuscript. A.C., L.B., and J.D. 279 carried out the processing of the samples. M.C and M.R. performed the epidemiological 280 analyses. A.C, D.P.L, A.F and J.C, helped in collection of samples. J.D.R. and M.M. revised 281 the manuscript. All authors read and approved the final version of the manuscript. 282 283 Acknowledgments 284 We express our gratitude to the staff at San Vicente de Paul Hospital, E.S.E Centro de Salud 285 de Gámeza, and E.S.E Centro de Salud de Jenesano for their crucial role in sample collection. 286 287 Legends: 288 Figure 1. A) Frequency of detection of C. perfringens in symptomatic (infected) vs 289 asymptomatic (colonized) individuals. B.) Schematic representation of the adjusted Odds 290 ratio (OR) together with its corresponding 95% confidence interval, which measures the 291 strength of association between the risk factors evaluated and the detection of C. perfringens. 292 293 Figure Supplementary 1. Sampling areas in the Department of Boyacá Colombia. The 294 hospitals included in the study are part of the municipalities highlighted in orange. 295 296 14 References 297 1. Asha N, Tompkins D, Wilcox MJJocm. Comparative analysis of prevalence, risk 298 factors, and molecular epidemiology of antibiotic-associated diarrhea due to Clostridium 299 difficile, Clostridium perfringens, and Staphylococcus aureus. 2006;44(8):2785-91. 300 2. Kyne L, Merry C, O'Connell B, Kelly A, Keane C, O'Neill DJA, et al. Factors 301 associated with prolonged symptoms and severe disease due to Clostridium difficile. 302 1999;28(2):107-13. 303 3. Casadevall A, Pirofski L-aJI, immunity. Host-pathogen interactions: basic concepts 304 of microbial commensalism, colonization, infection, and disease. 2000;68(12):6511-8. 305 4. Larcombe S, Hutton ML, Lyras DJTim. Involvement of bacteria other than 306 Clostridium difficile in antibiotic-associated diarrhoea. 2016;24(6):463-76. 307 5. Uzal FA, Freedman JC, Shrestha A, Theoret JR, Garcia J, Awad MM, et al. 308 Towards an understanding of the role of Clostridium perfringens toxins in human and 309 animal disease. 2014;9(3):361-77. 310 6. Kiu R, Hall LJJEm, infections. An update on the human and animal enteric 311 pathogen Clostridium perfringens. 2018;7(1):1-15. 312 7. Vaishnavi C, Kaur SJIJoP, Microbiology. Clostridium perfringens enterotoxin in 313 antibiotic-associated diarrhea. 2008;51(2):198. 314 8. Azimirad M, Gholami F, Yadegar A, Knight DR, Shamloei S, Aghdaei HA, et al. 315 Prevalence and characterization of Clostridium perfringens toxinotypes among patients 316 with antibiotic-associated diarrhea in Iran. 2019;9(1):7792. 317 9. Grass JE, Gould LH, Mahon BEJFp, disease. Epidemiology of foodborne disease 318 outbreaks caused by Clostridium perfringens, United States, 1998–2010. 2013;10(2):131-6. 319 10. Dewey-Mattia D, Manikonda K, Hall AJ, Wise ME, Crowe SJJMSS. Surveillance 320 for foodborne disease outbreaks—United States, 2009–2015. 2018;67(10):1. 321 11. Scallan E, Hoekstra RM, Angulo FJ, Tauxe RV, Widdowson M-A, Roy SL, et al. 322 Foodborne illness acquired in the United States—major pathogens. 2011;17(1):7-15. 323 12. Forero AJ, Muñoz M, Camargo M, Soto-De León SC, Ríos-Chaparro DI, Birchenall 324 C, et al. High frequency of toxigenic Clostridium difficile and Clostridium perfringens 325 coinfection among diarrheic patients at health care facility-onset (HCFO) and community-326 onset (CO) centers in Bogotá, Colombia. 2019;11:1-6. 327 13. Feng Y, Fan X, Zhu L, Yang X, Liu Y, Gao S, et al. Phylogenetic and genomic 328 analysis reveals high genomic openness and genetic diversity of Clostridium perfringens. 329 2020;6(10). 330 14. Kiu R, Shaw AG, Sim K, Acuna-Gonzalez A, Price CA, Bedwell H, et al. Particular 331 genomic and virulence traits associated with preterm infant-derived toxigenic Clostridium 332 perfringens strains. 2023:1-16. 333 15. Roussan D, Shaheen IA, Totanji W, Khawaldeh G, Al Rifai RJPs. Simultaneous 334 detection of Clostridium perfringens and Clostridium colinum by duplex-polymerase chain 335 reaction. 2012;91(12):3080-5. 336 16. Baums CG, Schotte U, Amtsberg G, Goethe RJVm. Diagnostic multiplex PCR for 337 toxin genotyping of Clostridium perfringens isolates. 2004;100(1-2):11-6. 338 17. Pickard JM, Zeng MY, Caruso R, Núñez GJIr. Gut microbiota: Role in pathogen 339 colonization, immune responses, and inflammatory disease. 2017;279(1):70-89. 340 15 18. Barbara AJ, Trinh HT, Glock RD, Songer JGJVm. Necrotic enteritis-producing 341 strains of Clostridium perfringens displace non-necrotic enteritis strains from the gut of 342 chicks. 2008;126(4):377-82. 343 19. Kobayashi S, Wada A, Shibasaki S, Annaka M, Higuchi H, Adachi K, et al. Spread 344 of a large plasmid carrying the cpe gene and the tcp locus amongst Clostridium perfringens 345 isolates from nosocomial outbreaks and sporadic cases of gastroenteritis in a geriatric 346 hospital. 2009;137(1):108-13. 347 20. Carey J, Cole J, Venkata SLG, Hoyt H, Mingle L, Nicholas D, et al. Determination 348 of genomic epidemiology of historical Clostridium perfringens outbreaks in New York 349 State by use of two web-based platforms: national center for biotechnology information 350 pathogen detection and FDA GalaxyTrakr. 2021;59(2):e02200-20. 351 21. Abrahao C, Carman R, Hahn H, Liesenfeld OJEJoCM, Diseases I. Similar 352 frequency of detection of Clostridium perfringens enterotoxin and Clostridium difficile 353 toxins in patients with antibiotic-associated diarrhea. 2001;20(9):676. 354 22. Joshy L, Chaudhry R, Dhawan B, Kumar L, Das BJJoHI. Incidence and 355 characterization of Clostridium perfringens isolated from antibiotic-associated diarrhoeal 356 patients: a prospective study in an Indian hospital. 2006;63(3):323-9. 357 23. Wang B, Dong W,Ma L, Dong Y, Wang S, Yuan Y, et al. Prevalence and genetic 358 diversity of Clostridium perfringens isolates in hospitalized diarrheal patients from Central 359 China. 2021:4783-93. 360 24. Carman RJ, Sayeed S, Li J, Genheimer CW, Hiltonsmith MF, Wilkins TD, et al. 361 Clostridium perfringens toxin genotypes in the feces of healthy North Americans. 362 2008;14(2):102-8. 363 25. Vela M, Heredia NL, Feng P, Garcı́a-Alvarado JSJDm, disease i. DNA probe 364 analysis for the carriage of enterotoxigenic Clostridium perfringens in feces of a Mexican 365 subpopulation. 1999;35(2):101-4. 366 26. Flores J, Okhuysen PCJCoiid. Genetics of susceptibility to infection with enteric 367 pathogens. 2009;22(5):471. 368 27. Lee H, Lee S, Kim S, Lee J, Ha J, Choi Y, et al. Asymptomatic Clostridium 369 perfringens inhabitation in intestine can cause inflammation, apoptosis, and disorders in 370 brain. 2020;17(1):52-65. 371 28. Flores-Dı́az M, Alape-Girón A, Titball RW, Moos M, Guillouard I, Cole S, et al. 372 UDP-glucose deficiency causes hypersensitivity to the cytotoxic effect of Clostridium 373 perfringens phospholipase C. 1998;273(38):24433-8. 374 29. Singh V, Rai R, Mathew BJ, Chourasia R, Singh AK, Kumar A, et al. 375 Phospholipase C: underrated players in microbial infections. 2023;13. 376 30. Francis MB, Allen CA, Shrestha R, Sorg JAJPp. Bile acid recognition by the 377 Clostridium difficile germinant receptor, CspC, is important for establishing infection. 378 2013;9(5):e1003356. 379 380 381 Frontiers in Microbiology 01 frontiersin.orgIntra-species diversity of Clostridium perfringens: A diverse genetic repertoire reveals its pathogenic potentialAnnyCamargo 1,2, EnzoGuerrero-Araya 3, SergioCastañeda 1, LauraVega 1, MaríaX.Cardenas-Alvarez 1,4, CésarRodríguez 5, DanielParedes-Sabja 3,6, JuanDavidRamírez 1,7 and MarinaMuñoz 1,3*1 Centro de Investigaciones en Microbiología y Biotecnología-UR (CIMBIUR), Facultad de Ciencias Naturales, Universidad del Rosario, Bogotá, Colombia, 2 Faculty of Health Sciences, Universidad de Boyacá, Tunja, Colombia, 3 ANID, Millennium Science Initiative Program, Millennium Nucleus in the Biology of the Intestinal Microbiota, Santiago, Chile, 4 Department of Pharmacology, University of North Carolina, Chapel Hill, NC, United States, 5 Laboratorio de Investigación en Bacteriología Anaerobia, Facultad de Microbiología, Centro de Investigación en Enfermedades Tropicales, Universidad de CostaRica, San José, Costa Rica, 6 Department of Biology, Texas A&M University, College Station, TX, United States, 7 Molecular Microbiology Laboratory, Department of Pathology, Molecular and Cell-Based Medicine, Icahn School of Medicine at Mount Sinai, New York, NY, United StatesClostridium perfringens is the causative agent of many enterotoxic diseases in humans and animals, and it is present in diverse environments (soil, food, sewage, and water). Multilocus Sequence Typing (MLST) and Whole Genome Sequencing (WGS) have provided a general approach about genetic diversity of C. perfringens; however, those studies are limited to specific locations and often include a reduced number of genomes. In this study, 372 C. perfringens genomes from multiple locations and sources were used to assess the genetic diversity and phylogenetic relatedness of this pathogen. In silico MLST was used for typing the isolates, and the resulting sequence types (ST) were assigned to clonal complexes (CC) based on allelic profiles that differ from its founder by up to double-locus variants. A pangenome analysis was conducted, and a core genome-based phylogenetic tree was created to define phylogenetic groups. Additionally, key virulence factors, toxinotypes, and antibiotic resistance genes were identified using ABRicate against Virulence Factor Database (VFDB), TOXiper, and Resfinder, respectively. The majority of the C. perfringens genomes found in publicly available databases were derived from food (n  = 85) and bird (n  = 85) isolates. A total of 195 STs, some of them shared between sources such as food and human, horses and dogs, and environment and birds, were grouped in 25 CC and distributed along five phylogenetic groups. Fifty-three percent of the genomes were allocated to toxinotype A, followed by F (32%) and G (7%). The most frequently found virulence factors based on > 70% coverage and 99.95% identity were plc (100%), nanH (99%), ccp (99%), and colA (98%), which encode an alpha-toxin, a sialidase, an alpha-clostripain, and a collagenase, respectively, while tetA (39.5%) and tetB (36.2%), which mediate tetracycline resistance determinants, were the most common antibiotic TYPE Original ResearchPUBLISHED 22 July 2022DOI 10.3389/fmicb.2022.952081OPEN ACCESSEDITED BYAndrew Spiers, Abertay University, UnitedKingdomREVIEWED BYJoseph C. S. Brown, Aparon, UnitedKingdomAnn-Katrin Llarena, Norwegian University of Life Sciences, Norway*CORRESPONDENCEMarina Muñoz claudia.munoz@urosario.edu.coSPECIALTY SECTIONThis article was submitted to Evolutionary and Genomic Microbiology, a section of the journal Frontiers in MicrobiologyRECEIVED 24 May 2022ACCEPTED 28 June 2022PUBLISHED 22 July 2022CITATIONCamargo A, Guerrero-Araya E, Castañeda S, Vega L, Cardenas-Alvarez MX, Rodríguez C, Paredes-Sabja D, Ramírez JD and Muñoz M (2022) Intra-species diversity of Clostridium perfringens: A diverse genetic repertoire reveals its pathogenic potential.Front. Microbiol. 13:952081.doi: 10.3389/fmicb.2022.952081COPYRIGHT© 2022 Camargo, Guerrero-Araya, Castañeda, Vega, Cardenas-Alvarez, Rodríguez, Paredes-Sabja, Ramírez and Muñoz. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orghttp://crossmark.crossref.org/dialog/?doi=10.3389/fmicb.2022.952081&domain=pdf&date_stamp=2022-07-22https://www.frontiersin.org/articles/10.3389/fmicb.2022.952081/fullhttps://www.frontiersin.org/articles/10.3389/fmicb.2022.952081/fullhttps://www.frontiersin.org/articles/10.3389/fmicb.2022.952081/fullhttps://www.frontiersin.org/articles/10.3389/fmicb.2022.952081/fullhttps://www.frontiersin.org/journals/microbiology#editorial-boardhttps://www.frontiersin.org/journals/microbiology#editorial-boardhttps://doi.org/10.3389/fmicb.2022.952081mailto:claudia.munoz@urosario.edu.cohttps://doi.org/10.3389/fmicb.2022.952081http://creativecommons.org/licenses/by/4.0/http://creativecommons.org/licenses/by/4.0/Camargo et al. 10.3389/fmicb.2022.952081Frontiers in Microbiology 02 frontiersin.orgresistance genes detected. The analyses conducted here showed a better view of the presence of this pathogen across several host species. They also confirm that the genetic diversity of C. perfringens is based on a large number of virulence factors that vary among phylogroups, and antibiotic resistance markers, especially to tetracyclines, aminoglycosides, and macrolides. Those characteristics highlight the importance of C. perfringens as a one of the most common causes of foodborne illness.KEYWORDSClostridium perfringens, intra-species diversity, multilocus sequence typing, genomic epidemiology, toxinotypesIntroductionClostridium perfringens is an anaerobic, Gram-positive, spore-forming bacillus capable of surviving extreme conditions such as high temperature (> 60°C) and low nutrient levels (Hassan etal., 2015). In 2002, the first C. perfringens genome wassequenced showing 2,660 protein-codifying regions, ten rRNA genes, and a low G + C content (28.6%; Shimizu etal., 2002). Since then, many studies have revealed the presence of genes that codify to multiple virulence factors involved in the pathogenicity of this species (Li and BA, 2014; Revitt-Mills etal., 2015; Kiu etal., 2017; Abdel-Glil etal., 2021), such as the alpha (CPA), beta (CPB), epsilon (ETX), iota (ITX), enterotoxin (CPE), and necrotic B-like (NetB) toxins (Awad etal., 1995; Sarker etal., 1999; Keyburn etal., 2008; Garcia et al., 2013; Rood et al., 2018), that contribute to neurologic, histotoxic, and intestinal infections in animals and humans and are used to classify strains in seven different toxin types (A to G; Kiu and Lindsay, 2018).Despite the usefulness of toxin typing in epidemiology and diagnosis, Multilocus Sequence Typing (MLST) has been implemented as an alternative approach for C. perfringens typing (Chalmers etal., 2008; Deguchi etal., 2009; Verma etal., 2020). This method is based on the presence and combination of internal fragments of eight housekeeping genes, which in C. perfringens includes colA, groEL, sodA, plc, gyrB, sigK, pgk, and nadA. The sequences are assigned as distinct alleles creating a unique allelic profile or sequence type (ST). Using the goeBURST algorithm, the STs are classified into groups of genetically related organisms called clonal complexes (CC), which provide insights about evolution and diversification processes and allow intra-species analyses (Larsen etal., 2012; Uzal etal., 2014; Page etal., 2017) such as epidemiological and phylogenetical associations (Maiden, 2006). In addition, the identification of shared STs allows the assessment of possible transmission routes between hosts, a third shared source, and genetic stability in the lineage (Xiao etal., 2012).Other methods, like Whole Genome Sequencing (WGS) have enabled a better understanding of bacterial pathogens through a high-resolution characterization of their genetic variation and evolution (Raskin etal., 2006). Recently, a genomic analysis allocated 206 publicly available C. perfringens genomes into five phylogroups (I–V) linked to different disease outcomes and hosts. Phylogroup Iincluded human food poisoning strains and phylogroup II mostly grouped isolates from enteric lesions in horses and dogs. Phylogroup III was the most abundant and heterogeneous group, containing a variety of strains from different hosts causing multiple diseases, while phylogroups IV and V were less abundant (Abdel-Glil etal., 2021). Additionally, the use of WGS has allowed the identification of virulence factors such as sialidases and hyaluronidases along with other toxins associated with clinical outcomes (Kiu et al., 2017; Abdel-Glil et al., 2021). Furthermore, over the last few years molecular markers linked to antibiotic resistance to tetracycline, rifamycin, and aminoglycoside among others, have been recognized as a potential risk for treatment of the infections caused by C. perfringens (Kiu and Lindsay, 2018).So far, C. perfringens genomic studies have been limited to a few geographic locations or to a small number of genomes. Hence, to better understand the global diversity of C. perfringens, weconducted a comparative genome analysis of 372 genomes from multiple locations and sources. Our goal was to determine the intra-species diversity and phylogenetic relationships of C. perfringens, as well as to identify and characterize key molecular markers associated with its pathogenicity, virulence, and antibiotic resistance from whole genome analyses.Materials and methodsStrain and genome collectionA total of 372 C. perfringens genomes were included in our analysis. Two strains sequenced specifically for this study were recovered from human feces and water, in 2013 and 2019, respectively. Briefly, samples were collected in sterile screw cap containers to avoid direct oxygen exposure (Siah etal., 2014) and were grown on tryptose sulfite cycloserine (TSC) agar (Merck) under anaerobic conditions by using anaerobic jars with anaerobic atmosphere generation pouches (AnaeroGen, Thermo Scientific, Oxoid) at 37°C for 24 h. Genomic DNA was extracted using the Wizard Genomic DNA Purification Kit (Promega). The quantity of the extracted DNA was assessed using a Qubit 2.0 Fluorometer. https://doi.org/10.3389/fmicb.2022.952081https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orgCamargo et al. 10.3389/fmicb.2022.952081Frontiers in Microbiology 03 frontiersin.orgDNA library preparation and paired-end whole genome sequencing (2 × 150 bp) were conducted on the Illumina NextSeq platform 2000 at the Microbial Genome Sequencing Center (MiGS). The assembled genomes were deposited at DDBJ/ENA/GenBank as part of a Bioproject under the accession number PRJNA836622.Publicly available sequence data of 370 C. perfringens genomes was also included in our study for comparative analyses. These data comprised 197 genome assemblies from the Pathosystems Resource Integration Center (PATRIC; Wattam et al., 2014; accessed on May 20th, 2021), 40 assemblies from the Public databases for molecular typing and microbial genome diversity (PubMLST; Jolley etal., 2018; accessed on May 14th, 2021), and 133 reads that were retrieved from the European Nucleotide Archive (ENA) database1 using the keywords “C. perfringens” (accessed on Jun 20th, 2021). Metadata and access numbers were verified to avoid duplicates (Supplementary Table S1).Taxonomic classification, genome assembly and annotationQuality stats of reads were collected using FastQC.2 The 133 raw read pairs from ENA were processed with Trimmomatic v. 0.38 to remove low-quality bases below Phred 30 and adapter sequences (Bolger etal., 2014). De novo assembly was conducted using SPAdes v. 3.14.1(Bankevich etal., 2012) with the default settings. The quality of the resulting assemblies was assessed with Quast v5.0.2 (Gurevich etal., 2013).The taxonomic classification of the assemblies was verified using Kraken v1.1.1 (Wood and Salzberg, 2014) with default parameters and using standard databases. Additionally, average nucleotide identity (ANI; Richter and Rossello-Mora, 2009) was calculated with pyANI.3 An average identity percentage above 95% denotes strains that belong to the same species. C. perfringens ATCC 13124 was used as a reference strain for this analysis. The assemblies generated by SPAdes and the ones obtained from public databases were annotated with PROKKA v. 1.11 (Seemann, 2014) with the default parameters and using the UniprotKB (SwissProt) database (Research UCJNa, 2017), considering kingdom specific databases for Bacteria. Annotated assemblies were taken to calculate the pangenome using Roary (Page etal., 2015). Genes present in 95% of the genomes with at least 95% of identity were designated as core genes.In silico MLST assignmentTo assign isolates to sequence type (ST), in silico MLST was performed using FastMLST v0.0.144 (Guerrero-Araya etal., 2020) 1 https://www.ebi.ac.uk/ena/browser/home2 http://www.bioinformatics.babraham.ac.uk/projects/fastqc/3 https://github.com/widdowquinn/pyani4 https://github.com/EnzoAndree/FastMLSTwith the default parameters. Concatenated sequences of the eight housekeeping genes of the scheme were extracted (colA, groEL, sodA, plc, gyrB, sigK, pgk, and nadA) and novel STs were submitted to the PubMLST database for identification (accessed on May 14th, 2021). Global optimal eBURST (goeBURST)5 was used to visualize CC, defined in this study as closely related STs that differ from a common founder at up to two of the eight loci used in the MLST scheme (DLV, double locus variant; Francisco etal., 2012).Phylogenetic reconstruction based on MLST and core genome gene sequencesConcatenated sequences of the eight housekeeping genes belonging to theMLST scheme were used as input for a multiple alignment. Likewise, core genes were concatenated and aligned using MAFFT v7.407. The best-fit model for base substitution in IQ-TREE v2.0.3 was selected to infer phylogenetic relationships by maximum likelihood (ML) and ultrafast bootstrap (1,000 repetitions; Nguyen et al., 2015). A clade was considered well supported when the bootstrap was ≥ 80%, as previously indicated (Wróbel, 2008; Minh et al., 2013). The resulting trees were visualized and edited using iTOL v.5 (Letunic and Bork, 2021). To confirm the phylogenetic clades generated by IQ-TREE, a NeighbourNet network was reconstructed in SplitsTree4 software v4.17.0 (Huson and Bryant, 2006) from a core genome alignment and producing a splits graph representing sequence distances (Huson, 1998).Reference genomes were included to classify isolates into phylogenetic groups. For Phylogroup I, C. perfringens Darmbrand NCTC8081 (ERR1656460) was selected, which is responsible for human enteritis Necroticans cases in Germany in the 1940s and is genetically related to strains carrying chromosomal cpe (Lindström etal., 2011; Ma etal., 2012). For Phylogroup II, strains involved in necrotizing enteritis in foals and hemorrhagic gastroenteritis in dogs, and denoted with the prefix JFP were chosen (GCA_001949795.1, GCA_001949805.1, GCA_001949775.1; Abdel-Glil etal., 2021). For Phylogroup III, C. perfringens ATCC3626 (GCA_000171155.1) and C. perfringens ATCC13124 (GCA_000013285.1) were used, and for Phylogroup IV, C. perfringens type D (GCA_006385425.1) isolated from a sheep. Phylogroup V was established based on the strain C. perfringens Tumat (GCA_003990265.1), Which was isolated from the mummified remains of an ancient puppy found in Siberian Permafrost (Abdel-Glil etal., 2021).In silico identification of virulence and antimicrobial resistance genesToxin detection (CPA, CPB, ETX, IAP, CPE, and netB) and typing of the 372 isolates were conducted with ABRicate v. 0.5 5 http://goeburst.phyloviz.nethttps://doi.org/10.3389/fmicb.2022.952081https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orghttps://www.ebi.ac.uk/ena/browser/homehttp://www.bioinformatics.babraham.ac.uk/projects/fastqc/https://github.com/widdowquinn/pyanihttps://github.com/EnzoAndree/FastMLSThttp://goeburst.phyloviz.netCamargo et al. 10.3389/fmicb.2022.952081Frontiers in Microbiology 04 frontiersin.orgusing the TOXIper database6 (available until October 20th, 2020). Additional virulence factors such as sialidases, collagenases, and secondary toxins like pfo, iam, and cpb2 were screened using the Virulence Factor Database (VFDB; March 17th, 2017; Chen etal., 2016). The Resfinder database (available until July 8th, 2017; Zankari etal., 2012) was used for the Antimicrobial Resistance (AMR) evaluation since it was the most complete database available with 2,228 AMR sequences at the moment of the analysis.ResultsLong-range dispersion of Clostridium perfringensA total of 372 genomes from different sources were collected for our study: two de novo sequenced genomes from Chile, 237 previously assembled sequences, and 133 raw reads (Supplementary Table S1). The number of contigs among the assemblies was variable, with an average of 82 contigs per genome. On overage, the genome size was 3.3 Mb, with a low GC content of around 28% (Supplementary Table S2). In addition, ANI values were invariably > 95% for the 372 genomes when compared to the reference strain C. perfringens ATCC 13124, confirming their classification as C. perfringens (Supplementary Figure S1). Most of the publicly available genomes were obtained from the US (123/372, 33.1%), France (48/372, 12.9%), and China (44/372, 11.8%; Figure1A), and were recovered mainly from food (85/372, 22.8%), birds (85/372, 22.8%), and humans (74/372, 19.8%; Figure1B).Wide toxinotype diversity among multiple hostsThe seven different toxinotypes defined until now were seen among our 372 isolates. Toxinotype A was the most frequently found (198/372, 53%), followed by toxinotype F (120/372, 32%), and G (24/372, 7%; Figure 1C). Interestingly, toxinotype distribution varied among hosts. In this regard, 60% (45/74) and 32% (24/74) of the human isolates were classified as toxinotype A and F, respectively. Two human isolates (2.7%) belonged to toxinotype E, and toxinotype C, D, and G had one (1.3%) isolate each. Food isolates were also distributed in a similar way, as 32% (27/85) were toxinotype A, 67% (57/85) F, and 1% (1/85) G.As for animals, 72% (61/85) of the bird isolates were classified as toxinotype A, 26% (22/85) as G, and 1% as C (1/85) and F (1/85). Fifty percent of the swine isolates (8/16) corresponded to toxinotype A, 43.7% (7/16) to C, and 6.3% (1/16) to F. Likewise, the ovine isolates were classified as toxinotype A (2/15, 14%), B (3/15, 20%), C (5/15, 33%), and D (5/15, 33%). Isolates from canines (16/22, 73%) and equines (15/16, 94%) belonged to 6 https://github.com/raymondkiu/TOXIpertoxinotype F and only a few to toxinotype A (6/22, 27% and 1/16, 6%, respectively). The high diversity of toxinotypes in the multiple evaluated hosts is an indicator of heterogeneous toxins that can support the differential pathogenic effect of C. perfringens populations circulating in each of these species.Clostridium perfringens clones of heterogeneous originsMultilocus sequence typing (MLST) was used to classify C. perfringens genomes. A total of 195 STs were identified among 322 genomes. One or more genes from the MLST scheme could not berecovered from 50 genomes, thus they were not assigned an ST. cpa, colA, and nadA were the genes with the most alleles (84, 84, and 73 respectively) compared to other MLST genes, which had less than 50 alleles (Supplementary Table S3).MLST loci from the 322 genomes, were concatenated and aligned to generate a phylogenetic tree (Figure2A), where five clusters were identified using the phylogenetic relationships and biological features as criteria, as reported before for C. perfringens (Abdel-Glil etal., 2021) Overall, Cluster III isolates were more frequent in our dataset (226/322, 70%). This cluster grouped isolates from diverse sources, including animals (120/226, 53%), humans (44/226, 20%), food (34/226, 15%), unknown sources (21/226, 9%), and environment (21/226, 3%). Cluster I(67/322, 20%) grouped mostly isolates from food (42/67, 63%) along with isolates from human (19/67, 28%), animal (3/67, 4.5%), and unknown sources (3/67, 4.5%). Interestingly, Cluster II (23/322, 7%) only grouped isolates from animals (19/23, 83%) and humans (4/23, 17%), whereas Cluster IV was formed by 3 isolates (3/322, 1%) from animal origin, and Cluster V by 3 isolates (3/322, 1%) from animal (2/3, 66%) and food (1/3, 34%) sources (Figure2A).Of the 195 STs detected, five were the most common: ST-147, -248, -80, -251, and ST-73 (Supplementary Figure S2). Isolates within these STs tended to have the same toxinotype, however, they were located across different clusters. For example, 14 toxinotype F isolates from food and human sources from the US and Italy isolated during 2017 and 2019 were classified as ST-147 and grouped in Cluster I. In the same way, 13 toxinotype F isolates from dogs and horses from Canada and the US, were classified as ST-80 located in Cluster II. Furthermore, ST-251 included 11 isolates from toxinotype F in cluster Irecovered in the US. In contrast, ST-73 grouped 11 isolates from toxinotypes G and A from birds, all of them grouped in Cluster III.The MLST-based minimum spanning tree built using the goeBURST algorithm led to evolutionary inferences. By identifying founder STs in the 25 CC and 95 singletons (Figure2B; Supplementary Figure S3). Overall, CC1 was the most commonly represented CC. It was found within Cluster III, grouping 23 STs from 50 genomes mostly from birds (ST-73, -106, -118),human stool and blood samples (ST-5, -262 y ST-271), and food (ST-5, -195, -225, -299 y ST-302). The founder ST for this group was ST-225, which is associated with food. https://doi.org/10.3389/fmicb.2022.952081https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orghttps://github.com/raymondkiu/TOXIperCamargo et al. 10.3389/fmicb.2022.952081Frontiers in Microbiology 05 frontiersin.orgUsually, the predicted founder corresponds to the most predominant ST in a CC (Feil et al., 2004), however this ST contained only one isolate whereas DLV descendant ST-73 contained 11 isolates (Supplementary Figure S3). This interesting finding might be caused by an origin bias due to either the relatively low number of samples used or by the natural selection pressure within the population leading to the emergence of strains with a strong adaptive advantage (Feil etal., 2004).Most of the isolates from human stools and food were grouped in Cluster Ias CC-2 along with ST-139 as founder ST. ST-39 and ST-253 from swine farms in China were classified as CC-18in Cluster III, and founder ST-80, -93, and -79 from canine and equine isolates were in cluster II as CC-9. Some of the STs varied in terms of hosts, for instance ST-80 and ST-78 included isolates from dogs and horses, ST-5, -33, -132, -139, -147, -248, and -251 included human and food isolates, and ST-39 grouped isolates from swine and birds, while ST-143 grouped birds and environment, and ST-215 birds and food (Figure2B).Five Clostridium perfringens phylogroups identified by core genome-based phylogenetic analysisA total of 35,876 genes are included in the pangenome of this dataset. Of them, 34,917 (97.3%) genes are accessory and only 959 (2.7%) are considered part of the core genome. In the resulting tree, five main branches were observed matching the ABCDFIGURE1Origin and toxinotype profile of 372 Clostridium perfringens genomes. (A) Distribution of isolates by country of origin. (B) Distribution of isolates by source. (C) Distribution of isolates by toxin type. (D) Toxinotype diversity among hosts. All figures were created with R software.https://doi.org/10.3389/fmicb.2022.952081https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orgCamargo et al. 10.3389/fmicb.2022.952081Frontiers in Microbiology 06 frontiersin.orgABFIGURE2Multilocus sequence typing (MLST)-based phylogeny and goeBURST full minimum spanning tree of 195 C. perfringens MLST profiles among 322 genomes. (A) MLST-based phylogeny tree obtained with fastMLST. The outer ring shows the origin of the isolates as indicated in the legend. Clusters are represented by different colors on the inside of the ring. (B) MLST-based minimum spanning tree obtained with goeBURST. Sequence types (STs) are displayed as circles. Founder STs were defined as the STs with the greatest number of single-locus variants. Circle size indicates the number of isolates in every particular ST, with each color representing a different source type. Lines represent closely related isolates and line length illustrates STs that vary by one, two, or more alleles in their MLST profile. Clonal complexes (CC), defined as closely related STs that differ up to two of the eight loci used in the MLST scheme from a common founder (DLV, double locus variant), are designated by dashing lines.https://doi.org/10.3389/fmicb.2022.952081https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orgCamargo et al. 10.3389/fmicb.2022.952081Frontiers in Microbiology 07 frontiersin.orgclusters that were established by using the reference genomes and a bootstrap ≥ 80% (Figure 3A). The core genome-based phylogenetic tree matches with the MLST-based topology, as well as with the phylogenetic network generated using neighbor-joining, confirming the classification of C. perfringens population in five distinct clusters as reported previously (Figure3B).Phylogroup I (n = 69) clustered most of the isolates from humans and food from toxinotype F (n = 57) that were cpa and cpe positive, along with some isolates from toxinotype A (n = 7) from humans and animals. Phylogroup II (n = 36) grouped canine and equine isolates from toxinotype F (n = 26), as well as one cattle and two human isolates from toxinotype E. Phylogroup III (n = 257) was the most diverse group, where the novel isolates recovered from water and a human in Chile were located. In this phylogroup isolates from birds, food, environment, and humans classified as toxinotype A (n = 177), toxinotype F (n = 31), toxinotype G (n = 24), and in less proportion toxinotype C (n = 14), D (n = 7), and B (n = 3) were also found. Phylogroup IV (n = 3) and V (n = 7) included a majority of toxinotype A isolates (Figure3A).Clostridium perfringens is a highly versatile pathogen with a large number of virulence factorsA total of 372 WGS were analyzed to evaluate the distribution of virulence genes. The alpha-toxin gene plc associated with gangrene in humans and several animals, and possibly involved in enterotoxemia and gastrointestinal disease in ruminants, horses, and swine was present in most of the isolates, along with genes encoding extracellular enzymes such as alpha-clostripain (ccp), hyaluronidases (nagH, nagI and nagJ), and alpha-clostripain collagenase (colA), as expected (Canard etal., 1994; Sakurai etal., 2004; Goossens etal., 2017; Geier etal., 2021). Other genes present in the majority of the isolates were pfoA (perfringolysin O), tpeL (toxin perfringens large), and cpb2 (beta2 toxin), which are protein coding genes involved in gastrointestinal outcomes and gangrene (Coursodon etal., 2012; Bueschel etal., 2013; Chen and McClane, 2015), and the sialidases nanH, nanI y nanJ, which play an important role in colonization and immunomodulation (Figure4A).The presence of virulence genes varied among the phylogroups. For example, all of the isolates in phylogroup Icarried nanH, however nanI, nanJ, nagH, pfoA, and cpb2 were absent in this group, in contrast to phylogroup II, where these genes were found in all of the isolates along with nagI and nagK. Additionally, 64% (23/36) of the isolates in this phylogroup also carried netE and 58% (21/36) netF. While the presence of these virulence genes in phylogroup III was variable, this was the only group where TpeL, a member of the large clostridial toxin (LCT) family involved in cell cytotoxicity was present, especially in isolates of toxinotype A, B, C, and G. However, this gene was not detected in type D or F isolates carrying the cpe and itx toxin genes. This difference in toxin profiles can be attributed to a potential incompatibility between plasmids carrying these genes (Chen and McClane, 2015).Isolates in phylogroup IV carried the coding genes for the hyaluronidases nagH, nagI, and nagJ, as well as nanI and pfoA. Phylogroup V, which cluster toxinotype A isolates carried nanI and pfoA, unlike the toxinotype A isolates in phylogroup I(Figure4A). These findings suggest that the differential presence of virulence factors in phylogroups may be due to selective advantage conferred by determinants in different niches (Sawires and Songer, 2006), routine treatment with clostridial toxoids especially in ruminants, or even environmental differences in the geographical regions (Simpson etal., 2018).Prevalence of antibiotic resistance genes in Clostridium perfringensAMR gene were found in 72.8% (271/372) of the genomes, with tetA (107/271, 39.5%) and tetB (98/271, 36.2%) involved in tetracycline resistance, being the most frequent and commonly found in phylogroup III isolates (Figure4B). Interestingly, the water isolate from Chile harbors genes for tetracycline resistance (tetA), while the one from human stool possesses genes that encoded tetracycline (tetA, tet44) and aminoglycoside resistance (ant(6)-Ib1). Likewise, this approachidentified genomes carrying ermQ, (28/372, 7.5%) an erythromycin resistance methylase gene mainly found in birds and environment isolates grouped in phylogroup III and classified as toxinotype A. ant(6)-Ib1 genes encoding aminoglycoside resistance were found in some toxinotype A and C isolates from swine and birds from phylogroup III (Figure4B).DiscussionClostridium perfringens is a clinically relevant pathogen due to its presence across several host species and its capacity to cause numerous medically important intestinal diseases in humans and animals (Mehdizadeh Gohari etal., 2021). To assess the genetic variation within the species, as well as to establish the phylogenetic relatedness, wecollected the publicly available genomes of 370 isolates collected between 2010 and 2020 from different countries, with a majority of the isolates obtained in developed countries such as the US, France, and China (Figure1A). There is a poor representation of isolates from developing countries in our dataset, possibly due to limited epidemiologic surveillance and genomic data collection, especially in South America, from where only one genome from Argentina was found and two more from Chile were added in this study.The dataset also has a high percentage of strains isolated from stool samples of animals used for human consumption as well as isolates from food origin (Figures1B, 2B), which along with the evidence of food products such as milk, meat, poultry, and pork https://doi.org/10.3389/fmicb.2022.952081https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orgCamargo et al. 10.3389/fmicb.2022.952081Frontiers in Microbiology 08 frontiersin.orgamong others as a source of infection, confirm the key role that C. perfringens plays as one of the most common causes of foodborne illness (Brynestad etal., 1997; Grass etal., 2013; Bintsis, 2017; Xu etal., 2021). However, despite the high percentage of isolates of human and food origin, it should be noted that environmental or commensal isolates are scarce (Figure1B). In addition, some metadata variables are unknown, such as the possible association of genomes with outbreaks, which could underestimate the diversity of C. perfringens and affect the results of AMR prevalence, emphasizing the need for further studies with a greater number of genomes.Clostridium perfringens toxinotypes are associated with heterogeneous diseases such as clostridial myonecrosis (gas gangrene) or gastrointestinal infections in humans and animals (Rood etal., 2018). Toxinotype G, associated with necrotic enteritis and gangrenous dermatitis in birds (Yang etal., 2019; Kiu etal., 2019a), and toxinotypes D and E that cause illness in sheep and cattle (Layana etal., 2006; Uzal and Songer, 2008; Nazki etal., 2017; Rood etal., 2018), have not been described in humans before, however, weidentified one isolate from a raw chicken patty and another from human blood as toxinotype G (Figure 1D). Likewise, we found toxinotype D and E strains isolated from human stools. These findings support previous studies that have reported C. perfringens type D and E strains harboring etx and iap in humans (Al-Shukri etal., 2021), which can explain potential routes of transmission in subjects that have been in contact with infected animals or have consumed contaminated food A BFIGURE3Phylogenetic grouping of C. perfringens using two different approaches. (A) Phylogenetic tree based on the core genome of 372 genomes. Five main phylogroups are highlighted in different colors. Source type and toxinotype are shown on the right panels. (B) Five phylogenetic networks based on the Neighbor-Joining (NJ) algorithm are shown with different colors.https://doi.org/10.3389/fmicb.2022.952081https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orgCamargo et al. 10.3389/fmicb.2022.952081Frontiers in Microbiology 09 frontiersin.org(Songer, 2010; Kiu etal., 2019b). Furthermore, these findings might berelated to C. perfringens strains considered as normal intestinal microbiota that are in contact with acquired strains carrying conjugative plasmids that are often associated with insertion sequences that can mobilize toxin genes between different strains. This could lead to the conversion into virulent toxin-producing strains and the emergence of specific toxinotypes in new hosts (Freedman et al., 2015). Future studies should include genomes assembled with a standard pipeline that includes an approach to recover extrachromosomal elements in order to describe the plasmidome of C. perfringens and thus contribute to the biological knowledge of this species.MLST and geoBURST analyses have not been widely used for C. perfringens: However, the identification of the founder STs, that differs from other STs at only one locus, provide a tool for epidemiological and evolutionary investigations of emergent pathogens (Feil and Enright, 2004). Furthermore, the identification of 25 CC in this work allowed us to compare the distribution of C. perfringens isolates from animals, humans, food and environment (Figure 2B; Supplementary Figure S2), evidencing a close relationship between isolates from different hosts. These findings support the strong association with foodborne diseases and suggesting their zoonotic potential and high diversification of this specie, as previously described (Verma etal., 2020; Hassani etal., 2022; Xiu etal., 2022).The CC including toxinotype G isolates from human and birds, and the evidence of the distribution of human, food, and animal isolates in the same ST or CC matches the zoonotic potential of C. perfringens as demonstrated by other published studies (Immerseel etal., 2004; Mwangi etal., 2019; Verma etal., 2020). Population structure analyses based on MLST (Figure2A) revealed five clusters in line with those generated here using the core genome (Figures 3A,B), and with results from previous studies (Abdel-Glil etal., 2021). Although MLST has a limited ability to establish phylogenetic relations since it only uses fragments from eight housekeeping genes, it is still a useful tool for interspecies C. perfringens typing due to its reproducibility, high discriminatory power, and easy accessibility (Pérez-Losada etal., 2013; Jolley etal., 2018; Guerrero-Araya etal., 2020). Despite these advantages, WGS has emerged as a more robust and complete tool contributing to the investigation of phylogenetic A BFIGURE4Virulence-associated and antimicrobial resistance (AMR) genes in 372 C. perfringens genomes. The core genome-based phylogenetic tree showed phylogroups highlighted in different colors. Matrix of absence and presence of genes shows (A) virulence genes (B) antimicrobial resistance (AMR) genes. The minimum coverage threshold needed for detection of these genes was 70% and the percentage of identity was 99.95%. The resulting trees were visualized and edited using iTOL v.5 (Letunic and Bork, 2021).https://doi.org/10.3389/fmicb.2022.952081https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orgCamargo et al. 10.3389/fmicb.2022.952081Frontiers in Microbiology 10 frontiersin.orgrelatedness among isolates and allowing a deeper understanding of transmission dynamics, emerging clones, key virulence loci, and the presence of AMR genes (Salipante etal., 2015; Quainoo etal., 2017; Pightling etal., 2018).The pangenome analysis of C. perfringens conducted in this study showed an accessory genome of 97.32%, an extremely high percentage in comparison with other closely related bacteria such as C. paraputrificum, species with an accessory genome of 67%, C. tertium with 37.6% of accessory genes (Muñoz etal., 2019), or C. baratii with an accessory genome of 24.43% (Silva-Andrade etal., 2022). This high level of genome plasticity is similar to the one found in Clostridioides difficile,which has an accessory genome of 87.2% (Knight etal., 2021). C. difficile and C. perfringens are part of the normal intestinal flora but can become gastrointestinal pathogens, which can beexplained by the ability to express different phenotypes in response to particular environmental conditions. The high genome plasticity of C. perfringens can give rise to the emergence of populations carrying new toxigenic profiles by acquired virulence factors due to horizontal gene transfer (HGT) leading to rapid genetic evolution. Thus, this genomic plasticity of C. perfringens is a determinant in the adaption to different hosts, as well as in the increase of its pathogenic potential and survival in different environments (Brüggemann, 2005).The importance of the acquisition of different virulence factors in C. perfringens is given by their adaptation within different disease niches. When exploring the virulence factors present in the C. perfringens genomes, wefound that plc, colA, and nanH are present in the majority of the genomes, hence they cannot be considered markers with high host discrimination capacity (Goossens et al., 2017; Mahamat Abdelrahim etal., 2019). The identification of isolates carrying netE and netF toxins in two animal species (dogs and foals) of phylogroup I, as well as previous reports of the prevalence of these two toxins in C. perfringens isolates from dogs with acute diarrhea hemorrhagic syndrome (Sindern etal., 2019), suggest their adaptability to these specific hosts. Likewise, the presence of three sialidases (nanH, nanI, and nanJ) in isolates from the same clinical outcomes highlights the role of these enzymes in increasing the adhesion of C. perfringens to host cells (Carman, 1997; Chiarezza etal., 2009; Li and McClane, 2021) and suggests an important role in the intestinal pathogenesis in these hosts, as was previously reported (Li et al., 2015; Li and McClane, 2021). However, a better understanding of the role and specificity of these molecules in canine acute hemorrhagic diarrhea and necrotizing enteritis of foals is required.Another virulence-related gene, pfoA, was also found in most of the genomes included here (Figure4A), however, it was absent in toxinotype F isolates clustered in the phylogroup I, which correlates with prior studies that have revealed that some strains that produce enterotoxins and therefor cause food poisoning lack pfoA (Myers etal., 2006; Deguchi etal., 2009). This suggests that this cytolytic toxin is not the main cause of most gastrointestinal outcomes in humans, however, it could beassociated with the enteritis necro-hemorrhagic or bovine enterotoxemia, as recent studies have shown (Verherstraeten et al., 2013a,b). Thus, the high number of virulence-related genes in the 372 genomes analyzed, especially in the novel isolates from Chile included in our study (Figure4A), reveals the need to continue with the epidemiological surveillance and the molecular study of virulence factors, mainly in unexplored countries, to provide more data for a deeper knowledge of the global diversity of C. perfringens.Another group of molecular markers of importance in health is the one associated with antibiotic resistance. Wefound that a high percentage of the genomes (72.8%, 270/372) harbored some type of AMR gene (Figure4B), where a large number of isolates carried genes putatively linked to resistance to tetracycline, macrolides, and aminoglycosides. The presence of a high number of AMR genes in isolates from pigs and birds could be a consequence of the use of antibiotics as growth promoters in these animals and may berelated to the appearance of resistance in zoonotic pathogens (Osman and Elhariri, 2013). Although WHO has questioned the use of additives due to the risk of antibiotic residues in meat products for human consumption, several countries continue with this practice that can berelated to the increase of antibiotic resistant strains (Salvarani et al., 2012). Many factors such as the production of high concentrations of antibiotics in the global industry (Nijsingh etal., 2019; Tell et al., 2019), the indiscriminate use of antibiotics in the community (Graham etal., 2019), the contamination of natural sources by hospital waste (Larsson and Flach, 2021), the impact of intrahospital infections caused by multidrug resistant strains (Avershina etal., 2021), the use of antibiotics for animal growth, the poor management of organic waste and the use of animal excrement in the agricultural sector (Tang etal., 2017), and the antimicrobial drugs overused during the first wave of COVID-19 (Manesh and Varghese, 2021), pose a high risk for possible pathways of antimicrobial resistance (Chokshi etal., 2019).The analysis of 372 genomes conducted here, is the largest effort to snapshot the global genomic diversity of C. perfringens to date. The genomic plasticity of this microorganism due to its low GC content (~ 28%; Uzal etal., 2014), its short doubling time (~ 7 min; Maiden, 2006), and a high percentage of horizontally transferred toxin encoding genes (Xiao etal., 2012; Uzal et al., 2014) contributes to the spread of existing toxinotypes to new hosts, as well as to the increase of food poisoning outbreaks and the growth of AMR. The use of MLST and WGS in a “One Health” framework that connects the health of humans and animals in a shared environment represents an optimal approach to advance knowledge of the global genetic diversity of C. perfringens. Our findings emphasize the need for further studies using a larger number of isolates from different ecological niches to elucidate the genetic characteristics, diversity, and zoonotic potential of C. perfringens and to improve strategies to reduce the growing threat to public health by this pathogen.https://doi.org/10.3389/fmicb.2022.952081https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orgCamargo et al. 10.3389/fmicb.2022.952081Frontiers in Microbiology 11 frontiersin.orgData availability statementThe datasets presented in this study can befound in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material.Author contributionsAC, EG-A, and MM designed the study and performed the data analyses. AC and MM performed methodology, formal analysis, data curation, and visualization and wrote the original draft preparation. JDR, DP-S, CR, SC, LV, and MC-A validated the results and revised and edited the manuscript. JDR and MM supervised the study and contributed to review and editing the manuscript. All authors contributed to the article and approved the submitted version.FundingThis study was financially supported by Dirección de Investigación e Innovación from Universidad del Rosario, Bogotá D.C. and the Universidad de Boyacá, Tunja, Colombia. In addition, this work mas supported by start-up funds from Texas A&M University and by ANID, Millennium Science Initiative Program–NCN17_093 to DP-S.AcknowledgmentsWe would like to thank the Universidad de Boyacá and the Universidad del Rosario in Colombia for their support.Conflict of interestThe authors declare that the research was conducted in the absence of any commercial or financial relationships that could beconstrued as a potential conflict of interest.Publisher’s noteAll claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.Supplementary materialThe Supplementary Material for this article can befound online at: https://www.frontiersin.org/articles/10.3389/fmicb.2022.952081/full#supplementary-materialReferencesAbdel-Glil, M. Y., Thomas, P., Linde, J., Busch, A., Wieler, L. H., Neubauer, H., et al. (2021). Comparative in silico genome analysis of Clostridium perfringens unravels stable phylogroups with different genome characteristics and pathogenic potential. Sci. Rep. 11:6756. doi: 10.1038/s41598-021-86148-8Al-Shukri, M. S., Hmood, A. M., and Al-Charrakh, A. H. (2021). Sequencing of Clostridium perfringens toxin genes (cpa, etx, iap) from Iraqi hospitals and detection by PCR of the genes encoding resistance to metronidazole, tetracycline, and clindamycin. Indian J. Med. Microbiol. 39, 289–294. doi: 10.1016/j.ijmmb.2021.03.017Avershina, E., Shapovalova, V., and Shipulin, G. J. (2021). Fighting antibiotic resistance in hospital-acquired infections: current state and emerging technologies in disease prevention, diagnostics and therapy. Front. Microbiol. 12:707330. doi: 10.3389/fmicb.2021.707330Awad, M. M., Bryant, A. E., Stevens, D. L., and Rood, J. I. (1995). Virulence studies on chromosomal α-toxin and Θ-toxin mutants constructed by allelic exchange provide genetic evidence for the essential role of α-toxin in Clostridium perfringens-mediated gas gangrene. Mol. Microbiol. 15, 191–202. doi: 10.1111/j.1365-2958.1995.tb02234.xBankevich, A., Nurk, S., Antipov, D., Gurevich, A. A., Dvorkin, M., Kulikov, A. S., et al. (2012). SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J. Comput. Biol. 19, 455–477. doi: 10.1089/cmb.2012.0021Bintsis, T. J. (2017). Foodborne pathogens. AIMS Microbiol. 3, 529–563. doi: 10.3934/microbiol.2017.3.529Bolger, A. M., Lohse, M., and Usadel, B. J. B. (2014). Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120. doi: 10.1093/bioinformatics/btu170Brüggemann, H. (2005). Genomics of clostridial pathogens: implication of extrachromosomal elements in pathogenicity. Curr. Opin. Microbio. 8, 601–605. doi: 10.1016/j.mib.2005.08.006Brynestad, S., Synstad, B., and Granum, P. E. J. M. (1997). The Clostridium perfringens enterotoxin gene is on a transposable element in type A human food poisoning strains. Microbiology 143, 2109–2115. doi: 10.1099/00221287- 143-7-2109Bueschel, D. M., Jost, B. H., Billington, S. J., Trinh, H. T., and Songer, J. G. (2013). Prevalence of cpb2, encoding beta2 toxin, in Clostridium perfringens field isolates: correlation of genotype with phenotype. Vet. Microbiol. 94, 121–129. doi: 10.1016/s0378-1135(03)00081-6Canard, B., Garnier, T., Saint-Joanis, B., and Cole, S. T. (1994). Molecular genetic analysis of the nagH gene encoding a hyaluronidase of Clostridium perfringens. Mol. Gen. Genet. 243, 215–224. doi: 10.1007/BF00280319Carman, R. J. (1997). Clostridium perfringens in spontaneous and antibiotic-associated diarrhoea of man and other animals. Rev. Med. Microbiol. 8:S46. doi: 10.1097/00013542-199712001-00024Chalmers, G., Bruce, H., Hunter, D., Parreira, V., Kulkarni, R., Jiang, Y.-F., et al. (2008). Multilocus sequence typing analysis of Clostridium perfringens isolates from necrotic enteritis outbreaks in broiler chicken populations. J. Clin. Microbiol. 46, 3957–3964. doi: 10.1128/JCM.01548-08Chen, J., and McClane, B. A. (2015). Characterization of Clostridium perfringens TpeL toxin gene carriage, production, cytotoxic contributions, and trypsin sensitivity. Infect. Inmun. 83, 2369–2381. doi: 10.1128/IAI.03136-14Chen, L., Zheng, D., Liu, B., Yang, J., and Jin, Q. (2016). VFDB 2016: hierarchical and refined dataset for big data analysis—10 years on. Nucleic Acids Res. 44, D694–D697. doi: 10.1093/nar/gkv1239Chiarezza, M., Lyras, D., Pidot, S. J., Flores-Díaz, M., Awad, M. M., Kennedy, C. L., et al. (2009). The NanI and NanJ sialidases of Clostridium perfringens are not essential for virulence. Infect. Immun. 77, 4421–4428. doi: 10.1128/IAI.00548-09Chokshi, A., Sifri, Z., Cennimo, D., and Horng, H. (2019). Global contributors to antibiotic resistance. J. Glob. Infect. Dis. 11, 36–42. doi: 10.4103/jgid.jgid_110_18Coursodon, C., Glock, R., Moore, K., Cooper, K., and Songer, J. J. A. (2012). TpeL-producing strains of Clostridium perfringens type A are highly virulent for broiler chicks. Anaerobe 18, 117–121. doi: 10.1016/j.anaerobe. 2011.10.001https://doi.org/10.3389/fmicb.2022.952081https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orghttps://www.frontiersin.org/articles/10.3389/fmicb.2022.952081/full#supplementary-materialhttps://www.frontiersin.org/articles/10.3389/fmicb.2022.952081/full#supplementary-materialhttps://doi.org/10.1038/s41598-021-86148-8https://doi.org/10.1016/j.ijmmb.2021.03.017https://doi.org/10.3389/fmicb.2021.707330https://doi.org/10.1111/j.1365-2958.1995.tb02234.xhttps://doi.org/10.1089/cmb.2012.0021https://doi.org/10.3934/microbiol.2017.3.529https://doi.org/10.1093/bioinformatics/btu170https://doi.org/10.1093/bioinformatics/btu170https://doi.org/10.1016/j.mib.2005.08.006https://doi.org/10.1099/00221287-143-7-2109https://doi.org/10.1099/00221287-143-7-2109https://doi.org/10.1016/s0378-1135(03)00081-6https://doi.org/10.1016/s0378-1135(03)00081-6https://doi.org/10.1007/BF00280319https://doi.org/10.1097/00013542-199712001-00024https://doi.org/10.1128/JCM.01548-08https://doi.org/10.1128/IAI.03136-14https://doi.org/10.1093/nar/gkv1239https://doi.org/10.1128/IAI.00548-09https://doi.org/10.4103/jgid.jgid_110_18https://doi.org/10.1016/j.anaerobe.2011.10.001https://doi.org/10.1016/j.anaerobe.2011.10.001Camargo et al. 10.3389/fmicb.2022.952081Frontiers in Microbiology 12 frontiersin.orgDeguchi, A., Miyamoto, K., Kuwahara, T., Miki, Y., Kaneko, I., Li, J., et al. (2009). Genetic characterization of type A enterotoxigenic Clostridium perfringens strains. PLoS One 4:5598. doi: 10.1371/journal.pone.0005598Feil, E. J., and Enright, M. C. (2004). Analyses of clonality and the evolution of bacterial pathogens. Curr. Opin. Microbiol. 7, 308–313. doi: 10.1016/j.mib.2004.04.002Feil, E. J., Li, B. C., Aanensen, D. M., Hanage, W. P., and Spratt, B. G. (2004). eBURST: inferring patterns of evolutionary descent among clusters of related bacterial genotypes from multilocus sequence typing data. J. Bacteriol. 186, 1518–1530. doi: 10.1128/JB.186.5.1518-1530.2004Francisco, A. P., Vaz, C., Monteiro, P. T., Melo-Cristino, J., Ramirez, M., and Carriço, J. A. (2012). PHYLOViZ: phylogenetic inference and data visualization for sequence-based typing methods. BMC Bioinform. 13, 1–10. doi: 10.1186/1471- 2105-13-87Freedman, J. C., Theoret, J. R., Wisniewski, J. A., Uzal, F. A., Rood, J. I., and McClane, B. (2015). Clostridium perfringens type A–E toxin plasmids. Res. Microbiol. 166, 264–279. doi: 10.1016/j.resmic.2014.09.004Garcia, J., Adams, V., Beingesser, J., Hughes, M. L., Poon, R., Lyras, D., et al. (2013). Epsilon toxin is essential for the virulence of Clostridium perfringens type D infection in sheep, goats, and mice. Infect. Inmun. 81, 2405–2414. doi: 10.1128/IAI.00238-13Geier, R. R., Rehberger, T. G., and Smith, A. H. (2021). Comparative genomics of Clostridium perfringens reveals patterns of host-associated phylogenetic clades and virulence factors. Front. Microbiol. 12:649953. doi: 10.3389/fmicb.2021.649953Goossens, E., Valgaeren, B. R., Pardon, B., Haesebrouck, F., Ducatelle, R., Deprez, P. R., et al. (2017). Rethinking the role of alpha toxin in Clostridium perfringens-associated enteric diseases: a review on bovine necro-haemorrhagic enteritis. Vet. Res. 48, 9–17. doi: 10.1186/s13567-017-0413-xGraham, D. W., Giesen, M. J., and Bunce, J. (2019). Strategic approach for prioritising local and regional sanitation interventions for reducing global antibiotic resistance. Water 11:27. doi: 10.3390/w11010027Grass, J. E., Gould, L. H., and Mahon, B. E. (2013). Epidemiology of foodborne disease outbreaks caused by Clostridium perfringens, UnitedStates, 1998–2010. Foodborne Pathog. Dis. 10, 131–136. doi: 10.1089/fpd.2012.1316Guerrero-Araya, E., Munoz, M., Rodríguez, C., and Paredes-Sabja, D. J. (2020). FastMLST: a multi-core tool for multilocus sequence typing of draft genome assemblies. Bioinform. Biol. Insights 15. doi: 10.1177/11779322211059238Gurevich, A., Saveliev, V., Vyahhi, N., and Tesler, G. J. B. (2013). QUAST: quality assessment tool for genome assemblies. Bioinformatics 29, 1072–1075. doi: 10.1093/bioinformatics/btt086Hassan, K. A., Elbourne, L. D., Tetu, S. G., Melville, S. B., Rood, J. I., and Paulsen, I. T. (2015). Genomic analyses of Clostridium perfringens isolates from five toxinotypes. Res. Mirobiol. 166, 255–263. doi: 10.1016/j.resmic.10.003Hassani, S., Pakbin, B., Brück, W. M., Mahmoudi, R., and Mousavi, S. (2022). Prevalence, antibiotic resistance, toxin-typing and genotyping of Clostridium perfringens in raw beef meats obtained from Qazvin City Iran. Antibiotics 11:340. doi: 10.3390/antibiotics11030340Huson, D. H. J. B. (1998). SplitsTree: analyzing and visualizing evolutionary data. Bioinformatics 14, 68–73. doi: 10.1093/bioinformatics/14.1.68Huson, D. H., and Bryant, D. J. (2006). Application of phylogenetic networks in evolutionary studies. Mol. Biol. Evol. 23, 254–267. doi: 10.1093/molbev/msj030Immerseel, F. V., Buck, J. D., Pasmans, F., Huyghebaert, G., Haesebrouck, F., and Ducatelle, R. (2004). Clostridium perfringens in poultry: an emerging threat for animal and public health. Climate Change Responses 33, 537–549. doi: 10.1186/s40665-016-0025-0Jolley, K. A., Bray, J. E., and Maiden, M. C. (2018). Open-access bacterial population genomics: BIGSdb software, the PubMLST. Org website and their applications. Wellcome Open Res. 3:124. doi: 10.12688/wellcomeopenres. 14826.1Keyburn, A. L., Boyce, J. D., Vaz, P., Bannam, T. L., Ford, M. E., Parker, D., et al. (2008). NetB, a new toxin that is associated with avian necrotic enteritis caused by Clostridium perfringens. PLoS Pathog. 4:e26. doi: 10.1371/journal.ppat. 0040026Kiu, R., Brown, J., Bedwell, H., Leclaire, C., Caim, S., Pickard, D., et al. (2019a). Genomic analysis on broiler-associated Clostridium perfringens strains and exploratory caecal microbiome investigation reveals key factors linked to poultry necrotic enteritis. Anim. Microbiome. 1:12. doi: 10.1186/s42523-019-0015-1Kiu, R., Caim, S., Alexander, S., Pachori, P., and Hall, L. J. (2017). Probing genomic aspects of the multi-host pathogen Clostridium perfringens reveals significant pangenome diversity, and a diverse array of virulence factors. Front. Microbiol. 8:2485. doi: 10.3389/fmicb.2017.02485Kiu, R., and Lindsay, J. H. (2018). An update on the human and animal enteric pathogen Clostridium perfringens. Emerg. Microbes Infect. 7, 1–15. doi: 10.1038/s41426-018-0144-8Kiu, R., Sim, K., Shaw, A., Cornwell, E., Pickard, D., Kroll, J. S., et al. (2019b). Genomic analysis of Clostridium perfringens BEC/CPILE-positive, Toxinotype D and E strains isolated from healthy children. Toxins 11:543. doi: 10.3390/toxins11090543Knight, D. R., Imwattana, K., Kullin, B., Guerrero-Araya, E., Paredes-Sabja, D., Didelot, X., et al. (2021). Major genetic discontinuity and novel toxigenic species in Clostridioides difficile taxonomy. Elife 10:10. doi: 10.7554/eLife.64325Larsen, M. V., Cosentino, S., Rasmussen, S., Friis, C., Hasman, H., Marvig, R. L., et al. (2012). Multilocus sequence typing of total-genome-sequenced bacteria. J. Clin. Microbiol. 50, 1355–1361. doi: 10.1128/JCM.06094-11Larsson, D., and Flach, C. (2021). Antibiotic resistance in the environment. Nat. Rev. Microbiol. 20, 257–269. doi: 10.1038/s41579-021-00649-xLayana, J. E., Miyakawa, M. E. F., and Uzal, F. (2006). Evaluation of different fluids for detection of Clostridium perfringens type D epsilon toxin in sheep with experimental enterotoxemia. Anaerobe 12, 204–206. doi: 10.1016/j.anaerobe.2006.05.001Letunic, I., and Bork, P. (2021). Interactive tree Of life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res. 49, W293–W296. doi: 10.1093/nar/gkab301Li, J., and BA, M. C. (2014). The sialidases of Clostridium perfringens type D strain CN3718 differ in their properties and sensitivities to inhibitors. Appl. Environ. Microbiol. 80, 1701–1709. doi: 10.1128/AEM.03440-13Li, J., Freedman, J. C., and McClane, B. A. (2015). NanI sialidase, CcpA, and CodY work together to regulate epsilon toxin production by Clostridium perfringens type D strain CN3718. J. Bacteriol. 197, 3339–3353. doi: 10.1128/JB.00349-15Li, J., and McClane, B. (2021). NanH is produced by sporulating cultures of Clostridium perfringens type F food poisoning strains and enhances the cytotoxicity of C. perfringens enterotoxin. mSphere 6. doi: 10.1128/mSphere.00176-21Lindström, M., Heikinheimo, A., Lahti, P., and Korkeala, H. J. (2011). Novel insights into the epidemiology of Clostridium perfringens type A food poisoning. Food Microbiol. 28, 192–198. doi: 10.1016/j.fm.2010.03.020Ma, M., Li, J., and McClane, B. A. (2012). Genotypic and phenotypic characterization of Clostridium perfringens isolates from Darmbrand cases in post-world war II Germany. Infect. Human. 80, 4354–4363. doi: 10.1128/IAI.00818-12Mahamat Abdelrahim, A., Radomski, N., Delannoy, S., Djellal, S., Le Négrate, M., Hadjab, K., et al. (2019). Large-scale genomic analyses and toxinotyping of Clostridium perfringens implicated in foodborne outbreaks in France. Front. Microbiol. 10:777. doi: 10.3389/fmicb.2019.00777Maiden, M. C. J. A. R. M. (2006). Multilocus sequence typing of bacteria. Annu. Rev. Microbiol. 60, 561–588. doi: 10.1146/annurev.micro.59.030804.121325Manesh, A., and Varghese, G. M. (2021). Rising antimicrobial resistance: an evolving epidemic in a pandemic. Lancet Microbe. 2, e419–e420. doi: 10.1016/S2666-5247(21)00173-7Mehdizadeh Gohari, I. A., Navarro, M., Li, J., Shrestha, A., Uzal, F., and McClane, B. A. (2021). Pathogenicity and virulence of Clostridium perfringens. Virulence 12, 723–753. doi: 10.1080/21505594.2021.1886777Minh, B. Q., Nguyen, M. A. T., and von Haeseler, A. J. (2013). Ultrafast approximation for phylogenetic bootstrap. Mol. Biol. Evol. 30, 1188–1195. doi: 10.1093/molbev/mst024Muñoz, M., Restrepo-Montoya, D., Kumar, N., Iraola, G., Herrera, G., Ríos-Chaparro, D. I., et al. (2019). Comparative genomics identifies potential virulence factors in Clostridium tertium and C. paraputrificum. Virulence 10, 657–676. doi: 10.1080/21505594.2019.1637699Mwangi, S., Timmons, J., Fitz-Coy, S., and Parveen, S. (2019). Characterization of Clostridium perfringens recovered from broiler chicken affected by necrotic enteritis. Poult. Sci. 98, 128–135. doi: 10.3382/ps/pey332Myers, G. S., Rasko, D. A., Cheung, J. K., Ravel, J., Seshadri, R., DeBoy, R. T., et al. (2006). Skewed genomic variability in strains of the toxigenic bacterial pathogen, Clostridium perfringens. Genome Res. 16, 1031–1040. doi: 10.1101/gr.5238106Nazki, S., Wani, S. A., Parveen, R., Ahangar, S. A., Kashoo, Z. A., Hamid, S., et al. (2017). Isolation, molecular characterization and prevalence of Clostridium perfringens in sheep and goats of Kashmir Himalayas, India. Vet. World. 10, 1501–1507. doi: 10.14202/vetworld.2017.1501-1507Nguyen, L.-T., Schmidt, H. A., Von Haeseler, A., and Minh, B. Q. J. (2015). IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol. Biol. Evol. 32, 268–274. doi: 10.1093/molbev/msu300Nijsingh, N., Munthe, C., and Larsson, D. (2019). Managing pollution from antibiotics manufacturing: charting actors, incentives and disincentives. Environ. Health 18, 1–17. doi: 10.1186/s12940-019-0531-1Osman, K., and Elhariri, M. (2013). Antibiotic resistance of Clostridium perfringens isolates from broiler chickens in Egypt. Rev. Sci. Tech. 32, 841–850. doi: 10.20506/rst.32.2.2212Page, A. J., Alikhan, N.-F., Carleton, H. A., Seemann, T., Keane, J. A., and Katz, L. S. (2017). Comparison of classical multi-locussequence typing software for https://doi.org/10.3389/fmicb.2022.952081https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orghttps://doi.org/10.1371/journal.pone.0005598https://doi.org/10.1016/j.mib.2004.04.002https://doi.org/10.1016/j.mib.2004.04.002https://doi.org/10.1128/JB.186.5.1518-1530.2004https://doi.org/10.1186/1471-2105-13-87https://doi.org/10.1186/1471-2105-13-87https://doi.org/10.1016/j.resmic.2014.09.004https://doi.org/10.1128/IAI.00238-13https://doi.org/10.1128/IAI.00238-13https://doi.org/10.3389/fmicb.2021.649953https://doi.org/10.1186/s13567-017-0413-xhttps://doi.org/10.3390/w11010027https://doi.org/10.1089/fpd.2012.1316https://doi.org/10.1177/11779322211059238https://doi.org/10.1093/bioinformatics/btt086https://doi.org/10.1093/bioinformatics/btt086https://doi.org/10.1016/j.resmic.10.003https://doi.org/10.3390/antibiotics11030340https://doi.org/10.1093/bioinformatics/14.1.68https://doi.org/10.1093/molbev/msj030https://doi.org/10.1186/s40665-016-0025-0https://doi.org/10.1186/s40665-016-0025-0https://doi.org/10.12688/wellcomeopenres.14826.1https://doi.org/10.12688/wellcomeopenres.14826.1https://doi.org/10.1371/journal.ppat.0040026https://doi.org/10.1371/journal.ppat.0040026https://doi.org/10.1186/s42523-019-0015-1https://doi.org/10.3389/fmicb.2017.02485https://doi.org/10.1038/s41426-018-0144-8https://doi.org/10.1038/s41426-018-0144-8https://doi.org/10.3390/toxins11090543https://doi.org/10.3390/toxins11090543https://doi.org/10.7554/eLife.64325https://doi.org/10.1128/JCM.06094-11https://doi.org/10.1038/s41579-021-00649-xhttps://doi.org/10.1016/j.anaerobe.2006.05.001https://doi.org/10.1093/nar/gkab301https://doi.org/10.1128/AEM.03440-13https://doi.org/10.1128/JB.00349-15https://doi.org/10.1128/mSphere.00176-21https://doi.org/10.1016/j.fm.2010.03.020https://doi.org/10.1128/IAI.00818-12https://doi.org/10.3389/fmicb.2019.00777https://doi.org/10.1146/annurev.micro.59.030804.121325https://doi.org/10.1016/S2666-5247(21)00173-7https://doi.org/10.1016/S2666-5247(21)00173-7https://doi.org/10.1080/21505594.2021.1886777https://doi.org/10.1093/molbev/mst024https://doi.org/10.1080/21505594.2019.1637699https://doi.org/10.3382/ps/pey332https://doi.org/10.1101/gr.5238106https://doi.org/10.14202/vetworld.2017.1501-1507https://doi.org/10.1093/molbev/msu300https://doi.org/10.1186/s12940-019-0531-1https://doi.org/10.20506/rst.32.2.2212Camargo et al. 10.3389/fmicb.2022.952081Frontiers in Microbiology 13 frontiersin.orgnext-generation sequencing data. Microb. Genom. 3:e000124. doi: 10.1099/mgen.0.000124Page, A. J., Cummins, C. A., Hunt, M., Wong, V. K., Reuter, S., Holden, M. T., et al. (2015). Roary: rapid large-scale prokaryote pan genome analysis. Bioinformatics 31, 3691–3693. doi: 10.1093/bioinformatics/btv421Pérez-Losada, M., Cabezas, P., Castro-Nallar, E., and Crandall, K. A. (2013). Pathogen typing in the genomics era: MLST and the future of molecular epidemiology. Genet. Evol. 16, 38–53. doi: 10.1016/j.meegid.2013.01.009Pightling, A. W., Pettengill, J. B., Luo, Y., Baugher, J. D., Rand, H., and Strain, E. (2018). Interpreting whole-genome sequence analyses of foodborne bacteria for regulatory applications and outbreak investigations. Front. Microbiol. 9:1482. doi: 10.3389/fmicb.2018.01482Quainoo, S., Coolen, J. P., van Hijum, S. A., Huynen, M. A., Melchers, W. J., van Schaik, W., et al. (2017). Whole-genome sequencing of bacterial pathogens: the future of nosocomial outbreak analysis. BMC Infect. Dis. 20, 1015–1063. doi: 10.1186/s12879-019-4743-3Raskin, D. M., Seshadri, R., Pukatzki, S. U., and Mekalanos, J. J. J. C. (2006). Bacterial genomics and pathogen evolution. Cell 124, 703–714. doi: 10.1016/j.cell.2006.02.002Research UCJNa (2017). The universal protein resource (UniProt). Nucleic Acids Res. 36, D190–D195. doi: 10.1093/nar/gkm895Revitt-Mills, S. A., Rood, J. I., and Adams, V. J. M. A. (2015). Clostridium perfringens extracellular toxins and enzymes: 20 and counting. Microbiol. Aust. 36, 114–117. doi: 10.1071/MA15039Richter, M., and Rossello-Mora, R. (2009). Shifting the genomic gold standard for the prokaryotic species definition. Proc. Natl. Acad. Sci. U. S. A. 106, 19126–19131. doi: 10.1073/pnas.0906412106Rood, J. I., Adams, V., Lacey, J., Lyras, D., McClane, B. A., Melville, S. B., et al. (2018). Expansion of the Clostridium perfringens toxin-based typing scheme. Anaerobe 53, 5–10. doi: 10.1016/j.anaerobe.2018.04.011Sakurai, J., Nagahama, M., and Oda, M. (2004). Clostridium perfringens alpha-toxin: characterization and mode of action. J. Biochem. 136, 569–574. doi: 10.1093/jb/mvh161Salipante, S. J., SenGupta, D. J., Cummings, L. A., Land, T. A., Hoogestraat, D. R., and Cookson, B. T. (2015). Application of whole-genome sequencing for bacterial strain typing in molecular epidemiology. J. Clin. Microbiol. 53, 1072–1079. doi: 10.1128/JCM.03385-14Salvarani, F. M., Silva, R. O. S., Pires, P. S., Cruz Júnior, E. C. C., Albefaro, I. S., Guedes, R. M. C., et al. (2012). Antimicrobial susceptibility of Clostridium perfringens isolated from piglets with or without diarrhea in Brazil. Braz. J. Microbiol. 43, 1030–1033. doi: 10.1590/S1517-838220120003000027Sarker, M. R., Carman, R. J., and BA, M. C. (1999). Inactivation of the gene (cpe) encoding Clostridium perfringens enterotoxin eliminates the ability of two cpe-positive C. perfringens type A human gastrointestinal disease isolates to affect rabbit ileal loops. Mol. Microbiol. 33, 946–958. doi: 10.1046/j.1365-2958.1999.01534.xSawires, Y. S., and Songer, J. G. J. A. (2006). Clostridium perfringens: insight into virulence evolution and population structure. Anaerobe 12, 23–43. doi: 10.1016/j.anaerobe.2005.10.002Seemann, T. J. B. (2014). Prokka: rapid prokaryotic genome annotation. Bioinformatics 30, 2068–2069. doi: 10.1093/bioinformatics/btu153Shimizu, T., Ohtani, K., Hirakawa, H., Ohshima, K., Yamash*ta, A., Shiba, T., et al. (2002). Complete genome sequence of Clostridium perfringens, an anaerobic flesh-eater. J. Proc. Natl. Acad. Sci. U. S. A. 99, 996–1001. doi: 10.1073/pnas.022493799Siah, S. P., Merif, J., Kaur, K., Nair, J., Huntington, P. G., Karagiannis, T., et al. (2014). Improved detection of gastrointestinal pathogens using generalised sample processing and amplification panels. Pathology 46, 53–59. doi: 10.1097/PAT.0000000000000022Silva-Andrade, C., Martin, A. J., and Garrido, D. J. M. (2022). Comparative genomics of Clostridium baratii reveals Strain-level diversity in toxin abundance. Microorganisms 10:213. doi: 10.3390/microorganisms10020213Simpson, K. M., Callan, R. J., and Metre, V. (2018). Clostridial abomasitis and enteritis in ruminants. Vet. Clin. North Am. Food Anim. Pract. 34, 155–184. doi: 10.1016/j.cvfa.2017.10.010Sindern, N., Suchodolski, J. S., Leutenegger, C. M., Mehdizadeh Gohari, I., Prescott, J. F., Proksch, A. L., et al. (2019). Prevalence of Clostridium perfringens netE and netF toxin genes in the feces of dogs with acute hemorrhagic diarrhea syndrome. J Vet Intern Med 33, 100–105. doi: 10.1111/jvim.15361Songer, J. G. (2010). Clostridia as agents of zoonotic disease. Vet. Microbiol. 140, 399–404. doi: 10.1016/j.vetmic.2009.07.003Tang, K. L., Caffrey, N. P., Nóbrega, D. B., Cork, S. C., Ronksley, P. E., Barkema, H. W., et al. (2017). Restricting the use of antibiotics in food-producing animals and its associations with antibiotic resistance in food-producing animals and human beings: a systematic review and meta-analysis. Lancet Planet Health. 1, e316–e327. doi: 10.1016/S2542-5196(17)30141-9Tell, J., Caldwell, D. J., Häner, A., Hellstern, J., Hoeger, B., Journel, R., et al. (2019). Science-based targets for antibiotics in receiving waters from pharmaceutical manufacturing operations. Integr. Environ. Assess. Manag. 15, 312–319. doi: 10.1002/ieam.4141Uzal, F. A., Freedman, J. C., Shrestha, A., Theoret, J. R., Garcia, J., Awad, M. M.,et al. (2014). Towards an understanding of the role ofClostridium perfringenstoxins in human and animal disease. Future Microbiol. 9, 361–377. doi: 10.2217/ fmb.13.168Uzal, F. A., and Songer, J. G. J. (2008). Diagnosis of Clostridium perfringens intestinal infections in sheep and goats. J. Vet. Diagn. Invest. 20, 253–265. doi: 10.1177/104063870802000301Verherstraeten, S., Goossens, E., Valgaeren, B., Pardon, B., Timbermont, L., Haesebrouck, F., et al. (2013a). Perfringolysin O: the underrated Clostridium perfringens toxin? Vet. Res. 44, 1702–1721. doi: 10.1186/1297-9716-44-45Verherstraeten, S., Goossens, E., Valgaeren, B., Pardon, B., Timbermont, L., Vermeulen, K., et al. (2013b). The synergistic necrohemorrhagic action of Clostridium perfringens perfringolysin and alpha toxin in the bovine intestine and against bovine endothelial cells. Vet. Res. 44, 1–8. doi: 10.1186/1297- 9716-44-45Verma, A. K., Abdel-Glil, M. Y., Madesh, A., Gupta, S., Karunakaran, A. C., Inbaraj, S., et al. (2020). Multilocus sequence typing of Clostridium perfringens strains from neonatal calves, dairy workers and associated environment in India. Anaerobe 63:102212. doi: 10.1016/j.anaerobe.2020.102212Wattam, A. R., Abraham, D., Dalay, O., Disz, T. L., Driscoll, T., Gabbard, J. L., et al. (2014). PATRIC, the bacterial bioinformatics database and analysis resource. Nucleic Acids Res. 42, D581–D591. doi: 10.1093/nar/gkt1099Wood, D. E., and Salzberg, S. L. (2014). Kraken: ultrafast metagenomic sequence classification using exact alignments. Genom. Biol. 15, R46–R12. doi: 10.1186/gb-2014-15-3-r46Wróbel, B. J. (2008). Statistical measures of uncertainty for branches in phylogenetic trees inferred from molecular sequences by using model-based methods. J. Appl. Genet. 49, 49–67. doi: 10.1007/BF03195249Xiao, Y., Wagendorp, A., Moezelaar, R., Abee, T., Wells-Bennik, M. H. J. A., and Microbiology, E. (2012). A wide variety of Clostridium perfringens type A food-borne isolates that carry a ChromosomalcpeGene belong to one multilocus sequence typing cluster. Appl. Environ. Microbiol. 78, 7060–7068. doi: 10.1128/AEM.01486-12Xiu, L., Zhu, C., Zhong, Z., Liu, L., Chen, S., Xu, W., et al. (2022). Prevalence and multilocus sequence typing of Clostridium perfringens isolated from different stages of a duck production chain. Food Microbiol. 102:103901. doi: 10.1016/j.fm.2021.103901Xu, W., Zhang, H., Hu, Z., Miao, Z., Zhang, Y., Wang, H. J. V. M., et al. (2021). Prevalence and multilocus sequence typing of Clostridium perfringens isolated from retail chicken products and diseased chickens in Tai'an region China. Vet. Med. Sci. 7, 2339–2347. doi: 10.1002/vms3.616Yang, W.-Y., Lee, Y.-J., Lu, H.-Y., Branton, S. L., Chou, C.-H., and Wang, C. (2019). The netB-positive Clostridium perfringens in the experimental induction of necrotic enteritis with or without predisposing factors. Poult. Sci. 98, 5297–5306. doi: 10.3382/ps/pez311Zankari, E., Hasman, H., Cosentino, S., Vestergaard, M., Rasmussen, S., Lund, O., et al. (2012). Identification of acquired antimicrobial resistance genes. J. Antimicrob. Chemother. 67, 2640–2644. doi: 10.1093/jac/dks261https://doi.org/10.3389/fmicb.2022.952081https://www.frontiersin.org/journals/microbiologyhttps://www.frontiersin.orghttps://doi.org/10.1099/mgen.0.000124https://doi.org/10.1099/mgen.0.000124https://doi.org/10.1093/bioinformatics/btv421https://doi.org/10.1016/j.meegid.2013.01.009https://doi.org/10.3389/fmicb.2018.01482https://doi.org/10.1186/s12879-019-4743-3https://doi.org/10.1016/j.cell.2006.02.002https://doi.org/10.1016/j.cell.2006.02.002https://doi.org/10.1093/nar/gkm895https://doi.org/10.1071/MA15039https://doi.org/10.1073/pnas.0906412106https://doi.org/10.1016/j.anaerobe.2018.04.011https://doi.org/10.1093/jb/mvh161https://doi.org/10.1093/jb/mvh161https://doi.org/10.1128/JCM.03385-14https://doi.org/10.1590/S1517-838220120003000027https://doi.org/10.1046/j.1365-2958.1999.01534.xhttps://doi.org/10.1016/j.anaerobe.2005.10.002https://doi.org/10.1016/j.anaerobe.2005.10.002https://doi.org/10.1093/bioinformatics/btu153https://doi.org/10.1073/pnas.022493799https://doi.org/10.1097/PAT.0000000000000022https://doi.org/10.3390/microorganisms10020213https://doi.org/10.1016/j.cvfa.2017.10.010https://doi.org/10.1111/jvim.15361https://doi.org/10.1016/j.vetmic.2009.07.003https://doi.org/10.1016/S2542-5196(17)30141-9https://doi.org/10.1002/ieam.4141https://doi.org/10.1002/ieam.4141https://doi.org/10.2217/fmb.13.168https://doi.org/10.2217/fmb.13.168https://doi.org/10.1177/104063870802000301https://doi.org/10.1186/1297-9716-44-45https://doi.org/10.1186/1297-9716-44-45https://doi.org/10.1186/1297-9716-44-45https://doi.org/10.1016/j.anaerobe.2020.102212https://doi.org/10.1093/nar/gkt1099https://doi.org/10.1186/gb-2014-15-3-r46https://doi.org/10.1186/gb-2014-15-3-r46https://doi.org/10.1007/BF03195249https://doi.org/10.1128/AEM.01486-12https://doi.org/10.1128/AEM.01486-12https://doi.org/10.1016/j.fm.2021.103901https://doi.org/10.1016/j.fm.2021.103901https://doi.org/10.1002/vms3.616https://doi.org/10.3382/ps/pez311https://doi.org/10.1093/jac/dks261 1 Insights into Clostridium perfringens Dispersal Hotspots, Toxins, and Virulence Factors 1 through Integrated Genomic and Phenotypic Profiling 2 3 Anny Camargo1,2, Laura Bohorquez1, Tatiana Cáceres1, Atilio Ferrebuz2, Diana Paola López2, José 4 Castellanos2, Javier Diaz Ovalle2, Raymond Kiu3, Lindsay J. Hall3.4,5, Juan David Ramírez1,6 and 5 Marina Muñoz1,7*. 6 7 1 Centro de Investigaciones en Microbiología y Biotecnología-UR (CIMBIUR), Facultad de Ciencias 8 Naturales, Universidad del Rosario, Bogotá, Colombia. 9 2 Universidad de Boyacá, Tunja, Colombia. 10 3 Gut Microbes and Health, Quadram Institute Bioscience, Norwich Research Park, Norwich, UK 11 4 Norwich Medical School, University of East Anglia, Bob Champion Research and Education 12 Building, James Watson Road, Norwich Research Park, Norwich, UK 13 5 Chair of Intestinal Microbiome, School of Life Sciences, ZIEL-Institute for Food and Health, 14 Technical University of Munich, Freising, Germany 15 6 Molecular Microbiology Laboratory, Department of Pathology, Molecular and Cell-Based 16 Medicine, Icahn School of Medicine at Mount Sinai, New York City, NY 10029, USA 17 7 Instituto de Biotecnología - UN (IBUN), Universidad Nacional de Colombia, Bogotá, Colombia 18 19 * Corresponding author: cmmunozd@unal.edu.co 20 2 Abstract 21 22 Clostridium perfringens, an anaerobic bacterium known for causing intestinal and tissue diseases in 23 humans and animals, requires local insights into its virulence factors, antibiotic resistance markers 24 (ARM), and dispersal dynamics as a critical step to develop effective prevention and management 25 strategies. This study conducted whole genome sequencing (WGS) of C. perfringens isolates obtained 26 from symptomatic and asymptomatic individuals and animals in the Colombian Cundiboyacense 27 highlands in 2022. A comparative genomic analysis was carried out at the microgeographic level with 28 185 Colombian genomes and another at the global level with 617 representative genomes, of which 29 546 were publicly available and 71 were Colombian. The main objective was to evaluate dispersal 30 pathways, AMR and virulence factors. In addition, a phenotypic characterization of 30 representative 31 isolates from all phylogroups was performed, taking into account the presence of perfingolysin O 32 (pfoA) coding. These tests included the evaluation of hemolysis, cell growth inhibition, sporulation 33 capacity and antibiotic sensitivity. 34 35 Comparative genomic analyses at the microgeographic level identified eight phylogroups exhibiting 36 wide genetic diversity in the region. Phylogroup I comprised 55 genomes from humans, canines, and 37 felines, while phylogroups II, IV, and VI had lower representation, including isolates from humans 38and multiple animal species. Phylogroup VII comprised genomes obtained from cats, which lacked 39 pfoA and a multi-peptide resistance factor (mprF). These deficiencies might correlate with heightened 40 susceptibility to aminoglycosides/defensins and reduced tolerance to acidic environments. 41 Additionally, pfoA, complete hemolysis, and increased cell growth inhibition rate were identified in 42 asymptomatic individuals, underscoring the potential risk of carrying toxigenic isolates in this 43 population. At the genomic level, we detected MRAs associated with aminoglycosides and 44 macrolides, which resulted in lower susceptibility to antibiotics such as gentamicin (87.0%), 45 erythromycin (20.0%) and metronidazole (17.0%). These results underscore the urgency of strict 46 regulation of antibiotic use in low- and middle-income countries to curb the proliferation of antibiotic 47 resistance. The integration of genomic, phenotypic, and epidemiological analyses allows us to better 48 understand the virulence factors and antibiotic resistance of locally circulating C. perfringens strains. 49 This knowledge is essential to improve public health interventions at the local level. 50 51 Keywords 52 Clostridium perfringens, comparative genomics, perfringolysin O, virulence factors, antibiotic 53 resistance markers, cytotoxicity, sporulation. 54 3 INTRODUCTION 55 56 Clostridium perfringens, an anaerobic and sporulating bacterium, has emerged as a global health 57 threat. It is recognized as one of the top five causes of foodborne illnesses (FBD) in the United States 58 (US) [1]. Despite extensive research on its impact in developed countries, the need for detailed 59 statistics in developing countries represents a crucial challenge. 60 61 C. perfringens is classified into seven toxinotypes (A-G) based on its ability to produce toxins, 62 including alpha (CPA), beta (CPB), enterotoxin (CPE), epsilon (Etx), iota (Itx), and necrotic enteritis 63 B-like (NetB) toxin, each with specific clinical implications. Toxinotype A, which carries the CPA 64 toxin, is ubiquitous and naturally inhabits the human and animal intestines. In vulnerable populations 65 with risk factors such as chronic diseases or immune deficiencies, this opportunistic pathogen can 66 quickly multiply, acquire, and release toxins capable of destroying epithelial tissue, thereby 67 contributing to the progression of intestinal and tissue diseases [2]. This toxinotype has been 68 implicated in pathologies such as clostridial myonecrosis [3, 4] and necrotizing enterocolitis in 69 premature infants [5, 6]. Toxinotype B isolates are associated with dysentery in sheep [7]. Toxinotype 70 C is associated with necrotizing enteritis and enterotoxemia in humans, sheep, foals, and piglets [8]. 71 Toxinotype D is related to symptoms of enterotoxemia and enterocolitis in sheep, goats, and, rarely, 72 cattle [9]. Toxinotype E is associated with enterotoxemia in calves and lambs [10]. Toxinotype F, 73 carriers of CPE toxin, are mainly related to human food poisoning, non-foodborne diarrhea, and 74 antibiotic-associated diarrhea (AAD) [11, 12]. Finally, toxinotype G is associated with avian necrotic 75 enteritis [13]. 76 77 While most studies have concentrated on detecting C. perfringens and characterizing major toxins 78 and virulence factors in symptomatic individuals, the specific role of this pathogen in asymptomatic 79 people, including the circulating accessory toxins and their contribution to disease development, still 80 needs to be completed. Some genotyping studies of C. perfringens carried out in healthy populations 81 have revealed the presence of toxigenic isolates in asymptomatic adults [14, 15] and premature 82 neonates [16], implying a potential risk as a reservoir of toxigenic isolates implicated in the 83 progression of natural history of diseases caused by this bacterium. 84 85 Approaches based on WGS enable the analysis of dissemination dynamics and the evaluation of 86 potential risks posed by circulating isolates to human and animal health [17]. Genomic studies of C. 87 perfringens have revealed the existence of five phylogroups, with clustering between isolates 88 4 involved in FBD from food and human sources and between isolates from intestinal lesions in foals 89 and canines, suggesting key relationships between host species [18]. In turn, the carriage of virulence 90 genes and ARM, as an evolutionary strategy to adapt to a specific habitat, may profoundly impact 91 both the phylogeny and the ability of this opportunistic pathogen to inhabit different ecological niches 92 [13]. 93 94 Complementary to WGS-based approaches, biological characterization of pathogenic bacteria is 95 crucial in understanding circulating strains' ecological and physiological attributes [19]. Phenotypic 96 analysis of isolates from the pediatric population revealed the role of toxins such as PFOA in diseases 97 such as necrotizing enterocolitis, impacting cellular toxicity, hemolysis, and increased 98 proinflammatory responses [20, 21]. This shows the importance of phenotypic testing for 99 characterizing C. perfringens toxins. 100 101 The increasing spread of virulence factors, the increase of genomic AMR over time [22] , and the 102 rising emergence of phenotypic resistance to antibiotics such as aminoglycosides, macrolides, and 103 nitroimidazoles [23] complicate the implementation of effective treatments aimed at eradicating 104 infectious diseases. This complexity is compounded by the limited biological knowledge of C. 105 perfringens, particularly in developing countries, and the population's need for more information on 106 antibiotic resistance markers. These factors converge to create a significant barrier that impedes the 107 implementation of effective preventive measures and treatments. 108 109 Integrating genomic data with experimental validation through phenotypic testing deepens the 110 understanding of C. perfringens biology. This study aimed to characterize 185 C. perfringens isolates 111 collected from symptomatic humans and asymptomatic animals in a central region of Colombia at 112 the genomic level, addressing the need for more data in developing nations. These findings were 113 contextualized globally through comparative genomics analysis with reported data worldwide. 114 Additionally, phenotypic tests were conducted to assess distinctive biological traits such as hemolysis 115 capacity, cytotoxicity, sporulation, and antibiotic susceptibility across representative phylogroups of 116 C. perfringens. This research expands the understanding of this bacterium's genomic and phenotypic 117 features to a microgeographic model, shedding light on hotspots and highlighting its threat to human 118 and animal health in the Colombian region. 119 120 MATERIALS AND METHODS 121 Sample Collection 122 5 A total of 59 C. perfringens-positive fecal samples were analyzed, previously identified by a specific 123 polymerase chain reaction (PCR) targeting the 16S-rRNA and alpha toxin (cpa) genes [24]. These 124 samples were collected from various sources in the Colombian Cundiboyacense highland region, 125 encompassing 18 humans (14 asymptomatic and 4 with gastrointestinal symptoms) and 41 126 asymptomatic animals (29 cats, nine dogs, one goat, one sheep, and one pig) (Supplementary Table 127 1). 128 129 Ethics Approval and Consent to Participate 130 The current study received approval from the Research Ethics Committee of the Universidad del 131 Rosario (CEI-UR approval number 449). It was classified as low risk by Resolution 8430/1993 of the 132 Colombian Ministry of Health. 133 134 Isolate Establishment, Whole Genome Sequencing (WGS), and Assembly 135 C. perfringens isolates were established by directly culturing fecal samples on selective medium 136 Tryptose Sulfite Cycloserine (TSC) agar through streaking technique [25], incubatingfor 24 hours at 137 37ºC under anaerobic conditions. Between 2 and 5 colony forming units (CFUs) per sample 138 exhibiting black appearance were selected, resulting in 185 isolates. The biomass of each CFU was 139 amplified on blood agar. Microscopic confirmation was achieved through Gram staining. Genomic 140 DNA was extracted from pure colonies matching the macroscopic and microscopic characteristics 141 using the Wizard Genomic DNA Purification Kit (Promega) following the manufacturer's 142 instructions. Quantification of the extracted genomic DNA was performed using a Qubit 2.0 143 Fluorometer. 144 145 The whole-genome paired-end library (2x150 bp) was sequenced on Illumina HiSeq500. The quality 146 of the reads (fastq files) and trimming was performed with Fastp v.0.20.0 [26]. De novo assembly 147 was performed using Unicycler v 0.4.9b [27]. Contigs <500 bp in each assembly were filtered out 148 before subsequent analyses, and the quality of the assemblies was assessed with sequence-stat v1.1 149 (https://github.com/raymondkiu/sequence-stats). The taxonomy of the assemblies was confirmed 150 with GTDB-TK v.1.5.1 [28]. Subsequently, the assemblies that presented contamination of less than 151 10% and completeness greater than 80% using CheckM V.1.1.3 [29] were selected to be included in 152 the subsequent analyses. 153 154 Public data set for comparative genomics analysis 155 6 The comparative genomic analyses included 1,083 public C. perfringens assemblies. These 156 assemblies, in fasta format, were compiled from the NCBI assembly database, accessible at 157 https://www.ncbi.nlm.nih.gov/assembly/?term=Clostridium+perfringens. Quality control was 158 implemented following the previously described methodology. 159 The complete dataset, comprising 1,268 genomes (1,083 public and 185 Colombian), underwent 160 analysis in dRep v.3.2.2 [30] to remove genomes with a 99.9% or higher similarity. This filtering 161 process yielded a final set of 617 genomes, consisting of 546 publicly accessible genomes and 71 of 162 Colombian origin. 163 164 Phylogenomic analyses based on the core genome and whole genome SNPs 165 The genomes were annotated with Prokka v.1.14.6 [31], and the resulting .gff files were employed as 166 input for pangenome analysis using Panaroo v.1.2.8. The core genome was identified with an identity 167 threshold of 98% and presence in at least 98% of the compared genomes [32]. Single Nucleotide 168 Polymorphisms (SNPs) were extracted from the core-genome alignment using snp-sites-2.3.3 [33]. 169 170 Alignments were conducted for two datasets: one at a microgeographic level, comprising all 171 Colombian C. perfringens isolates (n=185) to analyze potential local transmission hotspots, and the 172 other at a global level, containing the public genomes along with dereplicated Colombian genomes 173 (n=617) to assess the phylogenetic relationships between the Colombian genomes and those reported 174 globally, we utilized data representative of the diversity. Maximum likelihood phylogenetic trees 175 were constructed using IQtree v.2.0.5 [34] for each dataset, employing the ultrafast bootstrap 176 (UFBoot) function (-bb option) and the SH-aLRT test (-alrt 1000) to define a reliable clade with SH-177 aLRT >= 80% and UFboot >= 95%. The resulting phylogenetic trees were visualized using iTol [35], 178 and clusters were identified using an R package that implements the hierBAPS algorithm, a method 179 for hierarchically clustering DNA sequence alignments to reveal nested population structures [36]. 180 181 Antimicrobial resistance and in-silico virulence factors 182 The presence of AMR-associated genes and virulence factors, specifically toxins, was determined 183 using Abricate v.1.0.1 (Seemann T, Abricate, GitHub https://github.com/tseemann/abricate) with the 184 CARD database [37], the VFDB database [38], and the Toxiper database (Kiu R, TOXIper: rapid 185 toxinotyping of Clostridium perfringens genomes, GitHub https://github.com/raymondkiu/ 186 TOXIper), respectively. Sequence identification was conducted with a percentage of identity greater 187 than or equal to 90% and coverage exceeding 80% based on previous work [5]. 188 189 7 In vitro phenotypic tests 190 Selection of isolates and hemolysis assay 191 Thirty representative isolates from the eight phylogroups were selected, ensuring a proportional 192 representation of each phylogroup in the total number identified. This approach aimed to adequately 193 capture the evaluated isolate set's genetic diversity. To analyze phenotypic traits related to the PFOA 194 toxin found in asymptomatic individuals, associated with necrotizing enterocolitis and linked to 195 increased cellular toxicity in intestinal cell lines as well as complete hemolysis [39], isolates were 196 classified according to the presence of the pfoA gene. 197 The isolates were cultured on blood agar plates for 24 hours at 37ºC under anaerobic conditions. 198 Complete hemolysis was assessed manually by observing the presence of transparent halos formed 199 inside and around the colonies. 200 201 Sulforhodamine B cytotoxicity assay 202 A cellular cytotoxicity assay using sulforhodamine B (SRB) was performed. This assay measures 203 cellular biomass by utilizing SRB's capability to bind to cellular proteins, forming complexes under 204 acidic conditions. Upon release in basic environments, dye release correlates with cell biomass, which 205 is quantified at 565 nm. This assay is independent of metabolic activity, thereby minimizing 206 interference [40]. 207 The SRB assay was conducted following previously documented procedures [40]. Vero-CCL-81 cells 208 sourced from ATCC were cultured for 24 hours at 37°C and 5% CO2 in Roswell Park Memorial 209 Institute (RPMI) 1640 medium supplemented with 5% fetal bovine serum. After forming a confluent 210 monolayer within 24 hours, the cells were rinsed, and bacterial supernatants (diluted 1:5) that had 211 been filtered (0.22 μm filter) were added for 48 hours. Air-dried plates were then stained with 100 μl 212 of 0.04% SRB and incubated for 20 minutes at room temperature. Unbound dye was eliminated by 213 washing rapidly five times with 200 μl/well of 1% acetic acid. Subsequently, 100 μl of 10 mM Tris 214 base was added to the wells to solubilize the dye. Absorbance was measured at a wavelength of 525 215 nm using a SPECTROstar nano microplate reader from BMG Labtech [41]. 216 217 Spore count and viability 218 Assays were conducted following previously outlined protocols [42-44]. In brief, 0.1 ml of the 219 bacterial culture was retrieved and inoculated into 10 ml of liquid thioglycollate (FTG) medium, 220 followed by incubation at 37°C until reaching an optical density at 620 nm (OD600) of 0.5 for all 221 inoculum to normalize bacterial biomass. 222 223 8 The spores were obtained by centrifuging at 16,000 × g at four °C for 20 minutes, followed by three 224 washes and resuspension in Phosphate Buffer Saline (PBS). Subsequently, the spores were stored at 225 -20°C for at least 48 hours to induce the lysis of vegetative cells. Density gradient separation was 226 then conducted using a 50% iso-osmotic Accudenz® solution [45-47]. Confirmation of spore 227 extraction and removal of vegetative cells was achieved through microscopic observation using the 228 Schaeffer-Fulton staining method [48]. Mechanical disaggregation of spores was performed with 229 Tween 80, and the spores were diluted 20 times in Guava® ViaCount™ reagent and quantified by 230 flow cytometry. Spores were identified as particles with a diameter of ≤1 μm and granularity of ≤ 3 231 μm (potentially aggregating up to three spores) [49]. To assess spore viability, 100 μL of spores were 232 plated on Brain Heart Infusion (BHI) agar plates supplemented with 0.1% taurocholate (a potent 233 germinant for spores). These agar plates were anaerobically incubatedfor 48 hours before counting 234 the CFU to determine viability. 235 236 Antibiotic susceptibility testing 237 The minimum inhibitory concentration (MIC) tests were conducted for a panel of nine antibiotics: 238 ceftriaxone, clindamycin, chloramphenicol, erythromycin, gentamicin, imipenem, metronidazole, 239 penicillin, and tetracycline. These antibiotics were selected based on their usage in humans and 240 animals and previous literature on antibiotic resistance in C. perfringens [50]. Upon reaching an 241 optical density at 620nm (OD600) of 0.5, bacterial inoculum was plated on Brucella agar 242 supplemented with hemin (5 µg/mL), vitamin K, and lamb blood. Commercial e-test strips were then 243 placed on the plates, which were incubated for 48 hours under anaerobic conditions. Resistance was 244 determined following the Clinical and Laboratory Standards Institute (CLSI) 2022 guidelines. For 245 antibiotics without established breakpoints for C. perfringens. Staphylococcus aureus was used as a 246 reference, consistent with previous reports [51]. 247 248 Results 249 250 Microgeographic analysis of C. perfringens reveals eight globally maintained phylogroups and 251 critical dispersal hotspots 252 253 Genomic Analysis at a microgeographic level uncovers dispersal hotspots of 185 C. perfringens 254 isolates from humans and animals in the Colombian Cundiboyacense highlands, revealing an average 255 genome size of 3.4 mb, with 51.8 contigs and a GC Percentage of 27.9% (Supplementary Table 1). 256 257 9 SNP-based phylogenetic reconstruction reveals eight clusters corresponding to phylogroups, with a 258 loss of clustering by host. It highlights critical relationships between human hosts and domestic 259 animals at the local level. Phylogroup I was the most representative, including 55 isolates from 260 humans, canines, and felines, while Phylogroup VI contained only two genomes, from a canine and 261 a human. Phylogroup VII comprised 41 C. perfringens genomes isolated exclusively from cats 262 (Figure 1). 263 264 Figure 1. Maximum likelihood tree representing phylogenomics relationships of 185 C. perfringens 265 genomes from humans and animals at a microgeographic level. Eight phylogroups were identified by 266 hierBAPS. The first column indicates the host, with unique (dereplicated) genomes highlighted in black 267 in the following column. A presence-absence matrix of antibiotic resistance markers (AMR), toxins, 268 tet44mprFtetAtetBermQant(6)-IbPfoAccpcolAcpb2-1cpb2-2cpeAlveolisinplcnanHnanInanJnagHnagInagJnagKnagLAMR Toxins ColonisationHostdrepTree scale: 0.1HostFelineHumanCanisPigRuminant050100150200No. of isolatesToxinotypeA F PhylogroupI VIIIII VIIIV VIII I VI 10 virulence factors, and associated genes, as well as colonization, is presented in the final column. The 269 bottom of the figure shows the number of isolates carrying each inspected molecular marker. 270 271 A dereplicated dataset comprising 546 public genomes and 71 Colombian genomes as representative 272 of C. perfringens diversity was used to evaluate the phylogenomic relationships from Colombia 273 compared to Global Data. The pangenome of this global dataset (n=617) comprised 17,590 genes, 274 with 1,875 core genes (10.6%) shared by 99% of the genomes. The resulting phylogeny based on core 275 genome SNPs identified eight distinct phylogroups (Figure 2). 276 277 Insights into phylogroup distribution in C. perfringens were inferred from this phylogenomic 278 reconstruction. Phylogroup I was the most frequently detected, with 281 genomes primarily sourced 279 from humans, ruminants, birds, and some felines. Phylogroup II comprised genomes from humans 280 and food sources, whereas phylogroups III, IV, and VI included genomes isolated from multiple 281 species. Phylogroup V featured a significant presence of pig and human genomes, while a 282 predominant association among human, feline, and ruminant genomes characterized phylogroups VI, 283 VII, and VIII. The genomes recovered from Colombian samples were distributed across all 284 phylogroups, with a higher concentration of cat genomes in phylogroup VII. Conversely, genomes of 285 human origin remained dispersed alongside other animal sources in different phylogroups without 286 distinctive clustering by countries or continents. This loss of clustering by biological characteristics 287 supports the high frequency of circulation among hosts and the diversification of this species. 288 289 11 290 Figure 2. Maximum likelihood tree derived from SNP alignment of core genomes exploring 291 phylogenomic relationships of 617 dereplicated genomes. This comprehensive view of Colombian and 292 Global Representatives reveals eight phylogroups identified through hierBAPS. The tree includes a 293 column specifying the host, followed by the continent and country of isolation. The locations of the 71 294 dereplicated Colombian genomes are highlighted in black. Furthermore, the tree presents information on 295 antibiotic resistance markers (AMR), toxins, virulence factors, and genes associated with colonization and 296 toxinotype. 297 298 Exploring the distribution of virulence factors and other distinctive characteristics across 299 phylogroups through phylogenomic reconstruction 300 301 The impact of C. perfringens infections is attributed to its vast repertoire of virulence factors. These 302 include major toxins, which are crucial for classifying the bacteria into toxinotypes associated with 303 disease development, as well as accessory toxins, sialidases, and hyaluronidases, which facilitate 304 colonization and play an essential role in the pathogenesis of this bacterium. 305 306 The genomic identification of main toxin genes at a microgeographic level showed that the 307 phospholipase C (plc) toxin gene, associated with clostridial myonecrosis in immunocompromised 308 HostContinentCountryplcnanHnanInanJnagHnagInagJnagKnagLmprFtetAtetBtet44ermQermGermBant(6)-Ibaac(6')ant(6)-IaoptrAfexAinuCPfoAccpcolAcpb2-1cpb2-2cpeAlveolisinNetBetxcpbTpeLiapibpiam becAbecBAMR Toxins ColonisationNetFColombia dRepToxinotypeTree scale: 0.1Tree scale: 0.1HostFelineHumanCanisPigFoodRuminantEnvironmentBirdHorseOtherContinentAmericaAfricaAsiaEuropeOceaniaEnvironmentNDCountryColombiaAustraliaChinaFranceUnited KingdomUSAFinlandOtherToxinotypeTumatATCC13124ATCC3626ATCC8081Type Dstr13PhylogroupA B C D E F GI VIIIII VIIIV VIII VI 12 patients, was present in 100% (n=185) of the Colombian isolates. In contrast, the enterotoxin (cpe) 309 gene, linked to gastrointestinal disease in humans, was found in only 1.63% of the isolates and showed 310 no association with symptoms in carrier individuals. No other main toxins were identified, indicating 311 that 98.3% of the isolates were classified within toxinotype A, while the remaining 1.63% 312 corresponded to toxinotype F. 313 314 All analyzed isolates contained genes encoding for Alpha Clostripain (ccp) and Collagenase A 315 (ColA), two accessory toxins known for their role in tissue inflammation induction in clostridial 316 myonecrosis cases. Screening the distribution of other accessory toxins by each phylogroup revealed 317 distinctive patterns. In phylogroup I, the presence of the pfoA gene was identified, linked to the 318 pathogenesis of necrotizing enterocolitis and intravascular hemolysis in humans, along withthe 319 presence of sialidases (nanH, nanI, nanJ) and hyaluronidases (nagH, nagI, nagJ and nagK) associated 320 with increased cytotoxicity and toxin spread, respectively. Furthermore, 29.0% (n=16/55) of the 321 isolates located in phylogroup I harbored the C. perfringens beta 2 gene (cpb2), a pore-forming toxin 322 implicated in the pathogenesis of necrotic enteritis. Phylogroup II was composed of hypovirulent 323 isolates lacking several toxins and virulence factors, only with the presence of the sialidase genes 324 nanH and nanI involved in bacterial colonization [52]. Phylogroups III and V were characterized by 325 carrying several sialidase and hyaluronidase genes. All genomes included in phylogroup VII obtained 326 exclusively from cats lacked pfoA, like phylogroups II and III, and showed alveolsin (alv), a 327 chromosomal toxin with hemolytic activity considered crucial in the damage of the eukaryotic cell 328 membrane [53, 54]. 329 Phylogenomic analyses at a global scale revealed that Colombian isolates were distributed across all 330 phylogroups, exhibiting similar virulence characteristics to public genomes. For instance, in 331 Phylogroups I, IV, VI, VII, and VIII, the Colombian isolates were identified alongside others from 332 different countries carrying the pfoA gene. Conversely, Phylogroup VII contained isolates from 333 Colombian cats, humans, and other sources worldwide carrying alv but lacking pfoA. The single 334 isolate carrying cpe clustered with C. perfringens isolates from human and food sources. 335 336 C. perfringens isolates carry AMR associated with tetracycline, macrolides, and 337 aminoglycosides 338 339 Genomic surveillance of AMR from genomes obtained at a microgeographic level allowed the 340 identification that 93.5% of genomes (n=173/185) carry tetracycline resistance (tet) genes, being 341 distributed throughout all phylogroups. Genes that confer resistance to multiple peptides and human 342 13 defensins (mprF) were identified in 77.2% (n=143/185) of the genomes, except those from cats 343 located in phylogroup VII. Furthermore, 9.1% of the isolates (n=17), located mostly in phylogroups 344 II and V, exhibited the aminoglycoside O-nucleotidyltransferases gene (ant(6)-Ib), indicative of 345 resistance to aminoglycosides. In a smaller proportion, 5.9% of the genomes (n=11) carried the 346 erythromycin ribosome methylase Q (ermQ) gene, associated with resistance to macrolides, 347 lincosamides, and streptogramins B, grouped mostly in phylogroup I (Figure 1). 348 349 Globally, the mprF gene was present in 92.0% of the genomes analyzed (n=586/617). The tet44, tetA, 350 and tetB genes were found in 70.6% (n=436), followed by the ermQ gene, which was identified in 351 9.2% (n=57) of the genomes (Figure 2). Regarding geographical and species distribution, isolates 352 from pigs in China, mainly located in phylogroup V, showed the highest amount of AMR, with 353 between four and five AMR per genome. In contrast, most genomes from phylogroups I and II, which 354 include isolates from birds, ruminants, humans, and foods, had a lower AMR load, with between one 355 and two per genome. 356 357 PFOA toxin is associated with complete hemolysis and inhibition of cell growth 358 359 A recent pediatric study suggests that C. perfringens isolates producers of PFOA toxin may induce 360 hemolysis and heightened cellular toxicity [21]. Considering this, we opted to assess the hemolytic 361 capacity of blood agar and the impact of culture supernatants on Vero cells using isolates from 362 asymptomatic individuals, including both carriers and non-carriers of PFOA. 363 The hemolysis profiles of the 30 representative isolates for the phylogroups identified in this study 364 were examined, classifying them based on the presence or absence of the pfoA gene. These analyses 365 revealed that all isolates carrying pfoA+ demonstrated complete hemolysis on blood agar (n=16), 366 while isolates lacking pfoA- exhibited partial hemolysis. When assessing the biological impact of 367 toxin supernatants from each isolate (1:5 dilution) on Vero cells, diverse behaviors were observed 368 among them. Those situated above the IC50 line showed a growth inhibitory effect at 48 hours (Figure 369 3A). Regarding the presence of the pfoA gene, a significant increase (p<0.05) in the percentage of 370 growth inhibition was observed in isolates carrying pfoA+ (Figure 3B). 371 14 372 Figure 3. Phenotypic characteristics of cytotoxicity and sporulation were assessed in 30 C. 373 perfringens isolates obtained from various sources in a central region of Colombia. (A) activity 374 of culture supernatants (1:5 dilution) for 30 representative isolates of C. perfringens lineages to assess 375 their potential toxigenic effect on Vero cells. The comparison was made between pfoA+ (n=16) and 376 pfoA− (n=14) isolates. pfoA+ isolates encode the PFOA toxin, while pfoA− isolates do not. (B) 377 Cellular toxicity of C. perfringens isolates was evaluated using the Sulphorodamine B assay, 378 comparing between the pfoA+ and pfoA− groups. (C) Percentage of growth inhibition caused by toxin 379 supernatants of isolates from different phylogroups on cell growth. (D) Sporulation efficiency of C. 380 perfringens isolates. (E) The count of colony-forming units (CFU) was detected after spore 381 purification and subsequent growth induction on a medium supplemented with taurocholate. 382 The behavior of isolates from different phylogroups on cell growth was analyzed through an ANOVA 383 analysis with the Holm-Sidak multiple comparisons test. The isolates in phylogroups II, IV, VII, and 384 VIII are responsible for the biological effect observed in the tests. It should be noted that phylogroups 385 II and IV exhibit a more pronounced biological effect than the other phylogroups (Figure 3C). 386 387 The ability to sporulate and resist antibiotics confers propagation potential to C. perfringens 388 389 100IC500pfoA+ pfoA-nanJnanInanHalvcpb2colAccppfoAcpeCP172CP145CP136CP192CP171CP143CP55CP182CP113CP72CP148CP25CP26CP125CP94CP48CP22CP70CP70CP167CP138CP181CP109CP107CP233CP208CP88CP193CP202CP200pfoA+ pfoA-AMRtetB(P)+tetA(P)+tet44+ErmQ+ANT(6)-Ib+Sialidase nanJnanInanHToxinsalvcpb2cpecolAccppfoAplcID LABCP172CP145CP136CP192CP171CP143CP55CP182CP233CP208CP88CP113CP193CP202CP72CP148CP25CP26CP125CP94CP48CP22CP70CP170CP167CP138CP200CP181CP109CP107SourceSHEEPDOGDOGHUMA N PIGDOGCATHUMA NHUMA NHUMA NCATCATHUMA NHUMA NCATDOGCATCATCATCATCATCATCATPIGGOAT DOGHUMA NHUMA NCATCATID IntOV28.1DOG12.1CAN148.2PH25.5POR42.5DOG09.4MG20.3PH07.5PH03.1PH46.2MG44.3MG60.2PH29.1PH37.1MG37.2DOG17.4GB46.1GB46.2MG65.4MG48.4MG14.2GB30.3MG35.5POR42.3CAP09.1CAN148.4PH31.4PH07.4MG58.3MG56.3VIII I III I IV V VI III II V III VIIPhylogroup PhylogroupCP172CP145CP136CP192CP171CP143CP55CP182CP233CP208CP88CP113CP193CP202CP72CP148CP25CP26CP125CP94CP48CP22CP70CP170CP167CP138CP200CP181CP109CP1070100Clostridium perfringens strainPercentofinhibitionIC50pfoA+ pfoA-plc7500050000250000Colony Forming Units (n) Spore concentration (spores/ml) Toxin genes Percentage of inhibition020000400006000080000PhylogroupI VIIIII VIIIV VIII I VIConcentrationI II III IV V VI VII VIII0100Clostridium perfringens PhylogroupPercentofinhibitionIC50a: p<0.0001 vs III; p<0.001 vs V; p<0.01 vs I; p<0.05 vs VI.b: p<0.001 vs III; p<0.05 vs V.abbb1000Phylogroup!: p<0.0001 vs III;p<0.001 vs V; p<0.01 vs I; <0.05 vs VI β: p<0.001 vs III;p<0.05 vs V 0 20 40 60 80 100pfoA+pfoA-Percent of inhibition** Unpaired t test, p=0,0273Percentage of inhibition*Unpair t test, p=0,0273pfoA-pfoA+ 0 20 40 60 80 100 IC50C. perfringens isolatesCP172CP145CP136CP192CP171CP143CP55CP182CP113CP72CP148CP25CP26CP125CP94CP48CP22CP70CP170CP167CP138CP181CP109CP107CP233CP208CP88CP193CP202CP200109876543210Colony Forming Units (n) Percentage of inhibition Toxin genes105PhylogroupI II III V VI VII VIIII II III IV V VI VII VIIICFUs2.55.07.510.012.5A. B.D. C.E. F.CP170 15 Phenotypic evaluation of sporulation efficiency revealed that isolates from phylogroups I, carrying 390 pfoA+, showed higher levels of sporulation compared to other phylogroups (Figure 3D). Specifically, 391 isolates CP55 and CP143, obtained from a cat and a dog, respectively, that belong to phylogroup I 392 and carry both pfoA+ and cpb2, demonstrated the highest sporulation efficiency. 393 394 Spore viability was measured as the isolates' ability to generate new CFU after purification and in 395 vitro cultivation (Figure 3E). It was observed that 56.6% (n=17/30) of the isolates formed CFU after 396 48 hours of incubation. Isolates from phylogroup I, which exhibited a wide distribution among diverse 397 sources such as humans, dogs, and cats, showed higher spore viability (Figure 3E), a characteristic 398 that can contribute to their survival in many environmental niches facilitating the dispersal of C. 399 perfringens among different hosts. 400 401 Finally, the antibiotic susceptibility profiles of 30 C. perfringens isolates were evaluated against a 402 panel of nine antibiotics from multiple classes, including aminoglycosides, macrolides, lincosamides, 403 penicillins, third-generation cephalosporins, carbapenems, tetracyclines, amphenicols, and 404 nitroimidazoles (Figure 4). The detection of antibiotic susceptibility followed the criteria established 405 by CLSI in 2022 for C. perfringens. 406 407 408 Figura 4. Antibiotic susceptibility pattern of C. perfringens isolates. Percentage of antibiotic 409 susceptibility (resistant, intermediate or sensitive) of 30 isolates of C. perfringens collected from a 410 central region of Colombia. 411 2665430 0 0 012212090 0 03224 24272130 30 300%10%20%30%40%50%60%70%80%90%100%Gentamycin Erythromycin Metronidazole Clindamycin Penicillin Tetracycline Ceftriaxone Chloramphenicol ImipenemANTIBIOTICSResistente Intermedio SensiblePercentage of isolatesResistance Intermediate SensitiveAntibiotics 16 All isolates demonstrated sensitivity to Imipenem, Chloramphenicol and Ceftriaxone, and 90.0% and 412 80.0% showed sensitivity to Penicillin and Clindamycin respectly. 413 414 On the other hand, 87.0% of the isolates resisted Gentamicin, while 20.0% resisted Erythromycin and 415 17.0% to Metronidazole. Additionally, 10.0% resisted Penicillin, as detailed in Figure 4. Although 416 70.0% of the isolates showed sensitivity to tetracycline, 30% showed intermediate susceptibility. 417 418 Discussion 419 420 Genomic studies of C. perfringens have primarily focused on isolates from limited geographical 421 regions in developed countries, creating an information gap in developing nations [55]. To address 422 this gap, our study obtained and characterized 185 genomes of C. perfringens from humans and 423 animals in the Altiplano Cundiboyacense, a central region of Colombia with a significant agricultural 424 impact. 425 426 Studying the population genetic structure of potentially pathogenic bacteria, coupled with 427 understanding their ecology, epidemiology, and geographical distribution, is a crucial tool for 428 unraveling their dispersal pathways. This study revealed a significant genetic diversity and two types 429 of profiles: i) groupings by animal species, such as in the case of phylogroup VII, and ii) signals of 430 potential zoonotic dispersal events between canines, felines, and humans, as evidenced in 431 phylogroups I and III (Figure 1). The dispersal of C. perfringens among animals or from animals to 432 humans can occur through three main pathways. Firstly, direct transmission, where most bacteria 433 multiply in the intestine and are excreted in feces, while spores present can contaminate surfaces in 434 contact with other individuals. Secondly, the presence of C. perfringens spores in tissues of 435 decomposing dead animals represents focal points of environmental contamination and increases 436 exposure for humans. Thirdly, transmission through animal meat products for human consumption 437 [56]. These findings highlight the evolutionary capacity that clostridial have developed to obtain 438 nutrients from multiple hosts, allowing them to adapt to a heterogeneity of niches. The high adaptive 439 capacity of C. perfringens is a concern, particularly in low- and middle-income developing countries 440 like Colombia, where factors such as increased exposure to domestic animals within the household, 441 lack of access to clean water and basic sanitation, as well as limited food refrigeration due to lack of 442 access to electricity, contribute to increased vulnerability to the spread of emerging high-impact 443 zoonoses. 444 445 17 In comparative bacterial genomics, analyzing organism spread globally stands out for identifying 446 potential health risks and establishing a baseline for evaluating intervention effectiveness [57]. 447 Utilizing phylogenomic analysis at a microgeographic level allowed for identifying possible 448 dissemination routes between canines and felines, apparently asymptomatic domestic animals in a 449 region with significant agricultural activity, revealing the potential risk of transmission through direct 450 contact of humans with animals carrying C. perfringens, underscoring a possible threat to human 451 health. Comparing C. perfringens genomes studied locally with globally available information 452 revealed that phylogroup I encompassed the highest number of cosmopolitan isolates originating from 453 diverse sources. Meanwhile, phylogroup II featured isolates from human and food sources, and 454 phylogroups IV and V from pigs and humans (Figure 2), supporting the role of C. perfringens in 455 foodborne diseases (FBD) [13] and its spread between human and animal sources [12]. These findings 456 emphasize the importance of safe food handling practices and the need to promote hygiene measures 457 such as handwashing and surface disinfection to prevent transmission. Remarkable geographic 458 differences could be due to specific genomic traits,such as the clustering of pig isolates from Asia, 459 cats from South America, and food sources in Europe. However, limitations in sampling C. 460 perfringens isolates in other geographic regions restrict our global understanding. This highlights the 461 urgency of further sampling efforts to trace the origins and spread of infectious diseases, especially 462 in developing countries in South America, Oceania, and Africa. 463 464 C. perfringens can release toxins, such as CPA, CPE, CPB, ETX, ITX, and NETB, in the intestinal 465 tract of humans and animals. These main toxins classify C. perfringens into seven toxinotypes linked 466 to intestinal and tissue diseases, especially in symptomatic hosts [1, 10, 18]. Although the impact of 467 accessory toxins and toxigenic isolates in asymptomatic populations is not fully evidenced, recent 468 studies suggest that the PFOA toxin associated with gas gangrene and necrotizing enterocolitis in 469 newborns [5, 58, 59], this could increase the risk of intestinal damage in asymptomatic carriers [5]. 470 This study revealed the presence of isolates carrying the pfoA+ gene in asymptomatic humans and 471 animals, which demonstrated complete hemolysis induction and exhibited a significant increase 472 (p<0.05) in the percentage of in-vitro cell growth inhibition compared to pfoA- isolates (ver Figura 473 3A y 3B). These findings are supported by previously reported results of toxigenic C. perfringens 474 isolates producing PFOA in healthy babies and neonates with NEC in England, which directly 475 correlated with cellular toxicity, complete hemolysis, proinflammatory responses, and increased 476 oxygen tolerance. This highlights the importance of PFOA in intestinal pathology, infection 477 establishment, and disease development [21]. On the other hand, alveolysin, a poorly studied 478 chromosomal toxin activated by thiol, which binds to cholesterol and shares a high identity (~86% 479 18 nucleotide identity) with PFOA, was predominantly present in isolates from Colombian cats located 480 in phylogroup VII (Figure 1). Its unique group specificity in asymptomatic domestic cats at a 481 microgeographic level, the high percentage of cell growth inhibition found in this study (Figure 3A), 482 and its presence in isolates from dogs with hemorrhagic diarrhea and foals with necrotizing enteritis 483 reported in North America and Switzerland [53, 60], highlight the importance of expanding 484 knowledge about its potential virulence in domestic animals. 485 486 The risk of C. perfringens transmission among individuals is due to its unique ability to survive in 487 adverse environments, forming spores resistant to high temperatures and hostile conditions, which 488 becomes a crucial factor during the dissemination of this bacterium [49, 61]. The increased 489 sporulation efficiency in pfoA+ carrying isolates (Figure 3D) implies a higher risk of transmission of 490 toxigenic isolates capable of generating intestinal pathologies in previously healthy populations. On 491 the other hand, the differences in cellular compositions of the spores related to surface proteins, 492 mineralization, and central water, as well as nutritional requirements and the time required for 493 germination, may explain the variation in germination rates at 48 hours (Figure 3E and 3F) [62]. 494 495 Phylogenomic studies are also crucial for monitoring AMR's mobilization, persistence, and 496 abundance in microbial populations [63]. Genomic studies in C. perfringens have revealed a 497 conserved presence of the gene encoding the multiple peptide resistance factor (MprF) [64, 65], which 498 confers resistance to antimicrobial peptides [66]. Despite being described as conserved, this study 499 identified a notable absence of MprF in isolates from cats in phylogroup VII (Figure 2). This absence 500 could be attributed to differences in selective pressures among hosts and may also indicate a 501 competitive disadvantage against other bacteria or the host's immune response in these isolates [67], 502 warranting further investigation. 503 504 Penicillin and clindamycin remain relevant for treatments of clostridial infections, with penicillin 505 currently being the drug of choice. Our study highlights that rates of resistance to penicillin and 506 clindamycin are notably low, similar to Hungary, Slovenia, and northern Taiwan [68]. However, the 507 presence of circulating isolates with reduced susceptibility to these antibiotics may lead to therapeutic 508 failures in treating serious pathologies caused by C. perfringens. Nevertheless, the reduction in 509 susceptibility to metronidazole in Colombia (17.0%) (Figure 4) contrasts markedly with other 510 countries such as Pakistan [69], Australia [50], and Canada [70], where susceptibility to 511 metronidazole is lower, raising questions about differences in prescription practices or specific local 512 factors such as diversity in the intestinal microbiota of hosts, genetic characteristics, and 513 19 sociodemographic factors. Since metronidazole is crucial for treating anaerobic and protozoal 514 infections, its resistance threatens the effectiveness of conventional treatment, underscoring the need 515 for more epidemiological studies to adapt therapeutic strategies to local realities [50, 71]. 516 517 Our study has some limitations, including the loss of precise correlation between genotype and 518 phenotype for AMR, which may be due to variability in gene expression under culture conditions, 519 limitations of the databases used for identification, and even the existence of alternative mechanisms 520 of bacterial persistence [72, 73]. However, it is noteworthy that such studies transcend local 521 epidemiology, contributing to understanding the evolution and spread of C. perfringens and 522 identifying relevant genomic markers. Vero cells were used to evaluate the cytotoxic effect of C. 523 perfringens, which may not directly reflect the intestinal scenario. Hence, cells are proposed to model 524 the intestinal epithelial barrier or an enteroid model to compare our results [74]. Future research 525 should include a broader and more diverse variety of samples from different ecological niches, such 526 as humans, animals, soil, water, and food. These studies will improve the accuracy in detecting and 527 understanding this microorganism and discover more efficient and strategic methods to mitigate its 528 growing impact on health. 529 530 Conclusions 531 This study underscores the relevance of WGS for understanding the evolution and spread of C. 532 perfringens, identifying relevant genomic markers, and overcoming limitations in local 533 epidemiology. By integrating genomic and experimental analyses, results allow for more precise 534 implementation of treatments and prevention measures for infectious diseases. The findings reported 535 here indicate, for the first time, the genetic diversity and presence of virulence and antimicrobial 536 resistance factors in C. perfringens in Colombia, highlighting notable dissemination between humans 537 and domestic animals and differences in toxin distribution such as PFOA and alveolysin. The 538 correlation of phenotypic traits of the PFOA toxin and its presence in asymptomatic individuals 539 suggests a potential risk to intestinal health. The observation of reduced susceptibility to antibiotics 540 such as gentamicin, erythromycin, and metronidazole underscores the importance of genomic and 541 phenotypic analyses that allow tracking of clonal expansion and transmission of existing and 542 emerging pathogens, as well as AMR that capture and propagate antibiotic resistance. The 543 comprehensive approach of genomic and phenotypic data analysis within the One Health framework 544 is essential for understanding the ecology and epidemiology of pathogens of public health concern, 545 enabling the identification of markersassociated with virulence and antimicrobial resistance and 546 contributing to understanding interactions with the environment and hosts. This multidisciplinary and 547 20 collaborative approach acknowledges the interconnectedness between human, animal, and 548 environmental health, providing valuable information to develop more precise and sustainable 549 strategies to mitigate these challenges at the community level. 550 551 Disclosure statement 552 The authors declare that they have no potential conflicts of interest 553 554 Funding 555 This study was financially supported by the Ministerio de Ciencia Tecnología e Innovación 556 (Minciencias), within the framework of the project 722289684653 contract. 613-2021. A.C. was 557 supported by the Universidad de Boyacá, Tunja, Colombia and by the Scholarships for doctoral 558 students Universidad del Rosario. L.J.H. was supported by Wellcome Trust Investigator Award 559 220876/Z/20/Z; the Biotechnology and Biological Sciences Research Council (BBSRC) Institute 560 Strategic Programme, Gut Microbes and Health BB/R012490/1, and its constituent projects 561 BBS/E/F/000PR10353 and BBS/E/F/000PR10356; and the BBSRC Institute Strategic Programme 562 Food Microbiome and Health BB/X011054/1 and its constituent project BBS/E/F/000PR13631. 563 References 564 565 1. Scallan, E., et al., Foodborne illness acquired in the United States--major 566 pathogens. Emerg Infect Dis, 2011. 17(1): p. 7-15. 567 2. Lee, H., et al., Asymptomatic Clostridium perfringens inhabitation in intestine can 568 cause inflammation, apoptosis, and disorders in brain. 2020. 17(1): p. 52-65. 569 3. Sakurai, J., M. Nagahama, and M. Oda, Clostridium perfringens alpha-toxin: 570 characterization and mode of action. J Biochem, 2004. 136(5): p. 569-74. 571 4. Valeriani, R.G., et al., Gas gangrene-associated gliding motility is regulated by the 572 Clostridium perfringens CpAL/VirSR system. Anaerobe, 2020. 66: p. 102287. 573 5. Kiu, R., et al., Particular genomic and virulence traits associated with preterm 574 infant-derived toxigenic Clostridium perfringens strains. Nat Microbiol, 2023. 8(6): 575 p. 1160-1175. 576 6. Moschino, L., et al., The Metabolome and the Gut Microbiota for the Prediction of 577 Necrotizing Enterocolitis and Spontaneous Intestinal Perforation: A Systematic 578 Review. Nutrients, 2022. 14(18). 579 7. Uzal, F.A. and J.G. Songer, Diagnosis of Clostridium perfringens intestinal 580 infections in sheep and goats. J Vet Diagn Invest, 2008. 20(3): p. 253-65. 581 8. Uzal, F.A. and B.A. McClane, Recent progress in understanding the pathogenesis 582 of Clostridium perfringens type C infections. Vet Microbiol, 2011. 153(1-2): p. 37-583 43. 584 9. Uzal, F.A., et al., Comparative pathogenesis of enteric clostridial infections in 585 humans and animals. 2018. 53: p. 11-20. 586 21 10. Redondo, L.M., et al., Clostridium perfringens type E virulence traits involved in 587 gut colonization. PLoS One, 2015. 10(3): p. e0121305. 588 11. Kiu, R. and L.J. Hall, An update on the human and animal enteric pathogen 589 Clostridium perfringens. Emerg Microbes Infect, 2018. 7(1): p. 141. 590 12. Uzal, F.A., et al., Towards an understanding of the role of Clostridium perfringens 591 toxins in human and animal disease. Future Microbiol, 2014. 9(3): p. 361-77. 592 13. Grenda, T., et al., Clostridium perfringens-Opportunistic Foodborne Pathogen, Its 593 Diversity and Epidemiological Significance. Pathogens, 2023. 12(6). 594 14. Carman, R.J., et al., Clostridium perfringens toxin genotypes in the feces of healthy 595 North Americans. Anaerobe, 2008. 14(2): p. 102-8. 596 15. Heikinheimo, A., et al., Humans as reservoir for enterotoxin gene--carrying 597 Clostridium perfringens type A. Emerg Infect Dis, 2006. 12(11): p. 1724-9. 598 16. Shaw, A.G., et al., Dynamics of toxigenic Clostridium perfringens colonisation in a 599 cohort of prematurely born neonatal infants. BMC Pediatr, 2020. 20(1): p. 75. 600 17. Karlsen, S.T., et al., From genotype to phenotype: computational approaches for 601 inferring microbial traits relevant to the food industry. FEMS Microbiol Rev, 2023. 602 47(4). 603 18. Abdel-Glil, M.Y., et al., Comparative in silico genome analysis of Clostridium 604 perfringens unravels stable phylogroups with different genome characteristics and 605 pathogenic potential. Scientific Reports, 2021. 11(1): p. 6756. 606 19. Barberán, A., et al., Hiding in Plain Sight: Mining Bacterial Species Records for 607 Phenotypic Trait Information. mSphere, 2017. 2(4): p. 10.1128/msphere.00237-17. 608 20. Kuo, J., et al., Toxigenic Clostridium perfringens Isolated from At-Risk Paediatric 609 Inflammatory Bowel Disease Patients. Journal of Crohn's and Colitis, 2024. 610 21. Kiu, R., et al., Particular genomic and virulence traits associated with preterm 611 infant-derived toxigenic Clostridium perfringens strains. Nature Microbiology, 612 2023. 8(6): p. 1160-1175. 613 22. Feng, Y., et al., Phylogenetic and genomic analysis reveals high genomic openness 614 and genetic diversity of Clostridium perfringens. Microb Genom, 2020. 6(10). 615 23. Slavić, D., et al., Antimicrobial susceptibility of Clostridium perfringens isolates of 616 bovine, chicken, porcine, and turkey origin from Ontario. Can J Vet Res, 2011. 617 75(2): p. 89-97. 618 24. Roussan, D., et al., Simultaneous detection of Clostridium perfringens and 619 Clostridium colinum by duplex-polymerase chain reaction. 2012. 91(12): p. 3080-620 3085. 621 25. Harmon, S.M., D.A. Kautter, and J.T. Peeler, Improved medium for enumeration of 622 Clostridium perfringens. Appl Microbiol, 1971. 22(4): p. 688-92. 623 26. Chen, S., et al., fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics, 624 2018. 34(17): p. i884-i890. 625 27. Wick, R.R., et al., Unicycler: Resolving bacterial genome assemblies from short 626 and long sequencing reads. PLoS Comput Biol, 2017. 13(6): p. e1005595. 627 28. Chaumeil, P.-A., et al., GTDB-Tk: a toolkit to classify genomes with the Genome 628 Taxonomy Database. Bioinformatics, 2019. 36(6): p. 1925-1927. 629 29. Parks, D.H., et al., CheckM: assessing the quality of microbial genomes recovered 630 from isolates, single cells, and metagenomes. Genome Res, 2015. 25(7): p. 1043-55. 631 22 30. Olm, M.R., et al., dRep: a tool for fast and accurate genomic comparisons that 632 enables improved genome recovery from metagenomes through de-replication. Isme 633 j, 2017. 11(12): p. 2864-2868. 634 31. Seemann, T., Prokka: rapid prokaryotic genome annotation. Bioinformatics, 2014. 635 30(14): p. 2068-2069. 636 32. Tonkin-Hill, G., et al., Producing polished prokaryotic pangenomes with the 637 Panaroo pipeline. Genome Biol, 2020. 21(1): p. 180. 638 33. Page, A.J., et al., SNP-sites: rapid efficient extraction of SNPs from multi-FASTA 639 alignments. Microb Genom, 2016. 2(4): p. e000056. 640 34. Minh, B.Q., et al., IQ-TREE 2: New Models and Efficient Methods for Phylogenetic 641 Inference in the Genomic Era. Molecular Biology and Evolution, 2020. 37(5): p. 642 1530-1534. 643 35. Letunic, I. and P. Bork, Interactive Tree Of Life (iTOL): an online tool for 644 phylogenetic tree display and annotation. Bioinformatics, 2006. 23(1): p. 127-128. 645 36. Tonkin-Hill, G., et al., RhierBAPS: An R implementation of the population 646 clustering algorithm hierBAPS. Wellcome Open Res, 2018. 3: p. 93. 647 37. Jia, B., et al., CARD 2017: expansion and model-centric curation of the 648 comprehensive antibiotic resistance database. Nucleic Acids Res, 2017. 45(D1): p. 649 D566-d573. 650 38. Chen, L., et al., VFDB 2016: hierarchical and refined dataset for big data analysis-651 -10 years on. Nucleic Acids Res, 2016. 44(D1): p. D694-7. 652 39. Kiu, R., et al., Particular genomic and virulence traits associated with preterm 653 infant-derived toxigenic Clostridium perfringens strains. 2023: p. 1-16.654 40. Vichai, V. and K. Kirtikara, Sulforhodamine B colorimetric assay for cytotoxicity 655 screening. Nat Protoc, 2006. 1(3): p. 1112-6. 656 41. Skehan, P., et al., New colorimetric cytotoxicity assay for anticancer-drug 657 screening. J Natl Cancer Inst, 1990. 82(13): p. 1107-12. 658 42. Lin, T., et al., Inactivation of Clostridium perfringens C1 Spores by the 659 Combination of Mild Heat and Lactic Acid. Foods, 2022. 11(23). 660 43. Juneja, V., et al., Evaluation of methylxanthines and related compounds to enhance 661 Clostridium perfringens sporulation using a modified Duncan and Strong Medium 662 1. 1993. 2(3): p. 203-218. 663 44. Akhtar, S., et al., Strategy to inactivate Clostridium perfringens spores in meat 664 products. Food Microbiol, 2009. 26(3): p. 272-7. 665 45. Gellert-Mortimer, S.T., et al., Evaluation of Nycodenz and Percoll density gradients 666 for the selection of motile human spermatozoa. Fertil Steril, 1988. 49(2): p. 335-41. 667 46. Sbracia, M., et al., Sperm function and choice of preparation media: comparison of 668 Percoll and Accudenz discontinuous density gradients. J Androl, 1996. 17(1): p. 61-669 7. 670 47. Galarza, D.A., et al., Sephadex filtration as successful alternative to density-671 gradient centrifugation procedures for ram sperm selection with improved kinetics. 672 Anim Reprod Sci, 2018. 192: p. 261-270. 673 48. Hamouda, T., A.Y. Shih, and J.R. Baker, Jr., A rapid staining technique for the 674 detection of the initiation of germination of bacterial spores. Lett Appl Microbiol, 675 2002. 34(2): p. 86-90. 676 49. Li, J., et al., Clostridium perfringens Sporulation and Sporulation-Associated Toxin 677 Production. Microbiol Spectr, 2016. 4(3). 678 23 50. Santos, R., et al., Clostridium perfringens associated with dairy farm systems show 679 diverse genotypes. Int J Food Microbiol, 2022. 382: p. 109933. 680 51. Anju, K., et al., Toxinotyping and molecular characterization of antimicrobial 681 resistance in Clostridium perfringens isolated from different sources of livestock 682 and poultry. Anaerobe, 2021. 67: p. 102298. 683 52. Li, J. and B.A.J.M. McClane, NanH is produced by sporulating cultures of 684 Clostridium perfringens type F food poisoning strains and enhances the cytotoxicity 685 of C. perfringens enterotoxin. 2021. 6(2): p. 10.1128/msphere. 00176-21. 686 53. Kiu, R., et al., Genomic Analysis of Clostridium perfringens BEC/CPILE-Positive, 687 Toxinotype D and E Strains Isolated from Healthy Children. Toxins (Basel), 2019. 688 11(9). 689 54. Geoffroy, C., et al., Alveolysin, the thiol-activated toxin of Bacillus alvei, is 690 hom*ologous to listeriolysin O, perfringolysin O, pneumolysin, and streptolysin O 691 and contains a single cysteine. J Bacteriol, 1990. 172(12): p. 7301-5. 692 55. Camargo, A., et al., Intra-species diversity of Clostridium perfringens: A diverse 693 genetic repertoire reveals its pathogenic potential. Front Microbiol, 2022. 13: p. 694 952081. 695 56. Songer, J.G., Clostridia as agents of zoonotic disease. Vet Microbiol, 2010. 140(3-696 4): p. 399-404. 697 57. Hay, S.I., et al., Global mapping of infectious disease. Philos Trans R Soc Lond B 698 Biol Sci, 2013. 368(1614): p. 20120250. 699 58. Stevens, D.L. and A.E. Bryant, The role of clostridial toxins in the pathogenesis of 700 gas gangrene. Clin Infect Dis, 2002. 35(Suppl 1): p. S93-s100. 701 59. Li, Z., et al., Severe Intravascular Hemolysis from Clostridium perfringens 702 Septicemia in a Neonate with Necrotizing Enterocolitis in China: A Case Report. 703 Infect Drug Resist, 2022. 15: p. 1461-1465. 704 60. Mehdizadeh Gohari, I., et al., NetF-producing Clostridium perfringens: Clonality 705 and plasmid pathogenicity loci analysis. Infect Genet Evol, 2017. 49: p. 32-38. 706 61. Hu, W.S., et al., Biofilm and Spore Formation of Clostridium perfringens and Its 707 Resistance to Disinfectant and Oxidative Stress. Antibiotics (Basel), 2021. 10(4). 708 62. Liang, D., et al., Effects of sporulation conditions on the growth, germination, and 709 resistance of Clostridium perfringens spores. Int J Food Microbiol, 2023. 396: p. 710 110200. 711 63. Djordjevic, S.P., et al., Genomic surveillance for antimicrobial resistance - a One 712 Health perspective. Nat Rev Genet, 2024. 25(2): p. 142-157. 713 64. Zhong, J.X., et al., Molecular characteristics and phylogenetic analysis of 714 Clostridium perfringens from different regions in China, from 2013 to 2021. Front 715 Microbiol, 2023. 14: p. 1195083. 716 65. Kiu, R., et al., Probing Genomic Aspects of the Multi-Host Pathogen Clostridium 717 perfringens Reveals Significant Pangenome Diversity, and a Diverse Array of 718 Virulence Factors. Front Microbiol, 2017. 8: p. 2485. 719 66. Ernst, C.M., et al., The bacterial defensin resistance protein MprF consists of 720 separable domains for lipid lysinylation and antimicrobial peptide repulsion. PLoS 721 Pathog, 2009. 5(11): p. e1000660. 722 67. Nawrocki, K.L., E.K. Crispell, and S.M. McBride, Antimicrobial Peptide 723 Resistance Mechanisms of Gram-Positive Bacteria. Antibiotics (Basel), 2014. 3(4): 724 p. 461-92. 725 24 68. Sárvári, K.P. and D. Schoblocher, The antibiotic susceptibility pattern of gas 726 gangrene-forming Clostridium spp. clinical isolates from South-Eastern Hungary. 727 Infect Dis (Lond), 2020. 52(3): p. 196-201. 728 69. Haider, Z., et al., Isolation, toxinotyping and antimicrobial susceptibility testing of 729 Clostridium perfringens isolated from Pakistan poultry. 2022. 73: p. 102499. 730 70. Slavić, Đ., et al., Antimicrobial susceptibility of Clostridium perfringens isolates of 731 bovine, chicken, porcine, and turkey origin from Ontario. 2011. 75(2): p. 89-97. 732 71. Álvarez-Pérez, S., J.L. Blanco, and M.E. García, Clostridium perfringens Type A 733 Isolates of Animal Origin with Decreased Susceptibility to Metronidazole Show 734 Extensive Genetic Diversity. Microb Drug Resist, 2017. 23(8): p. 1053-1058. 735 72. Doyle, R.M., et al., Discordant bioinformatic predictions of antimicrobial 736 resistance from whole-genome sequencing data of bacterial isolates: an inter-737 laboratory study. Microb Genom, 2020. 6(2). 738 73. Rose, R., et al., Comparing antimicrobial resistant genes and phenotypes across 739 multiple sequencing platforms and assays for Enterobacterales clinical isolates. 740 BMC Microbiol, 2023. 23(1): p. 225. 741 74. Stewart, C.J., M.K. Estes, and S. Ramani, Establishing Human Intestinal 742 Enteroid/Organoid Lines from Preterm Infant and Adult Tissue. Methods Mol Biol, 743 2020. 2121: p. 185-198. 744 745 Full Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=temi20Emerging Microbes & InfectionsISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/temi20Unveiling the pathogenic mechanisms ofClostridium perfringens toxins and virulence factorsAnny Camargo, Juan David Rámirez, Raymond Kiu, Lindsay J. Hall & MarinaMuñozTo cite this article: Anny Camargo, Juan David Rámirez, Raymond Kiu, Lindsay J. Hall & MarinaMuñoz (09 Apr 2024): Unveiling the pathogenic mechanisms of Clostridium perfringens toxinsand virulence factors, Emerging Microbes & Infections, DOI: 10.1080/22221751.2024.2341968To link to this article: https://doi.org/10.1080/22221751.2024.2341968© 2024 The Author(s). Published by InformaUK Limited, trading as Taylor & FrancisGroup, on behalf of Shanghai ShangyixunCultural Communication Co., LtdAccepted author version posted online: 09Apr 2024.Submit your article to this journal Article views: 256View related articles View Crossmark datahttps://www.tandfonline.com/action/journalInformation?journalCode=temi20https://www.tandfonline.com/journals/temi20?src=pdfhttps://www.tandfonline.com/action/showCitFormats?doi=10.1080/22221751.2024.2341968https://doi.org/10.1080/22221751.2024.2341968https://www.tandfonline.com/action/authorSubmission?journalCode=temi20&show=instructions&src=pdfhttps://www.tandfonline.com/action/authorSubmission?journalCode=temi20&show=instructions&src=pdfhttps://www.tandfonline.com/doi/mlt/10.1080/22221751.2024.2341968?src=pdfhttps://www.tandfonline.com/doi/mlt/10.1080/22221751.2024.2341968?src=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1080/22221751.2024.2341968&domain=pdf&date_stamp=09 Apr 2024http://crossmark.crossref.org/dialog/?doi=10.1080/22221751.2024.2341968&domain=pdf&date_stamp=09 Apr 2024 1 Publisher: Taylor & Francis & The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group, on behalf of Shanghai Shangyixun Cultural Communication Co., Ltd Journal: Emerging Microbes & Infections DOI: 10.1080/22221751.2024.2341968 Unveiling the pathogenic mechanisms of Clostridium perfringens toxins and virulence factors Anny Camargo1,2, Juan David Rámirez1,3, Raymond Kiu4,5, Lindsay J. Hall4,5,6, and Marina Muñoz1,7*. 1 Centro de Investigaciones en Microbiología y Biotecnología-UR (CIMBIUR), Facultad de Ciencias Naturales, Universidad del Rosario, Bogotá, Colombia. 2 Health Sciences Faculty, Universidad de Boyacá, Tunja, Colombia. 3 Molecular Microbiology Laboratory, Department of Pathology, Molecular and Cell-Based Medicine, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA 4 Institute of Microbiology and Infection, College of Medical and Dental Sciences, University of Birmingham, Birmingham, UK 5 Gut Microbes and Health, Quadram Institute Bioscience, Norwich Research Park, Norwich, UK 6 Norwich Medical School, University of East Anglia, Bob Champion Research and Education Building, James Watson Road, Norwich Research Park, Norwich, UK 7 Instituto de Biotecnología-UN (IBUN), Universidad Nacional de Colombia, Bogotá, Colombia * Corresponding author: cmmunozd@unal.edu.co; claudia.munoz@urosario.edu.co Abstract Clostridium perfringens causes multiple diseases in humans and animals. Its pathogenic effect is supported by a broad and heterogeneous arsenal of toxins and other virulence factors associated with a specific host tropism. Molecular approaches have indicated that most C. perfringens toxins produce membrane pores, leading to osmotic cell disruption and mailto:cmmunozd@unal.edu.comailto:claudia.munoz@urosario.edu.cohttp://crossmark.crossref.org/dialog/?doi=10.1080/22221751.2024.2341968&domain=pdf 2 apoptosis. However, identifying mechanisms involved in cell tropism and selective toxicity effects should be studied more. The differential presence and polymorphisms of toxin-encoding genes and genes encoding other virulence factors suggest that molecular mechanisms might exist associated with host preference, receptor binding, and impact on the host; however, this information has not been reviewed in detail. Therefore, this review aims to clarify the current state of knowledge on the structural features and mechanisms of action of the major toxins and virulence factors of C. perfringens and discuss the impact of genetic diversity of toxinotypes in tropism for several hosts. Keywords: Clostridium perfringens, toxin type, genetic diversity, host, enteritis, toxinotypes. Introduction Clostridium perfringens is a Gram-positive anaerobic bacterium that can form spores that are crucial during transmission. C. perfringens spores are exceptionally resistant to stressful environments, such as high temperatures, the presence of oxygen, or low nutrient levels [1]. These features facilitate its survival in different environmental niches, including soil, faeces, sewage, food, and the intestinal tract of humans and animals [2]. This ‘survivability’ means that C. perfringens has been associated with many infections and diseases, including being the second leading cause of foodborne bacterial disease in the United States (causing one million illnesses yearly) and Europe's fourth leading cause [3, 4, 5]. In England, 8-13% of foodborne gastrointestinal outbreaks are estimated to be associated with this bacterium (90,000 cases of C. perfringens per year) [6]. The impact of C. perfringens infections is due in part to the bacterium’s potential to secrete multiple extracellular toxins, including alpha-toxin (CPA, cpa/plc gene), beta-toxin (CPB, cpb gene), epsilon-toxin (ETX, etx gene), iota-toxin (ITX, iap, and ibp binary genes), C. perfringens enterotoxin (CPE, cpe gene) and Necrotic enteritis B-like toxin (NetB). These toxins correspond to essential toxin genes used in the current toxinotyping scheme (toxinotypes A-G) (Figure supplementary 1) [7]. 3 Each toxinotype is associated with specific diseases. For example, toxinotype A, characterised by the presence of the cpa+ toxin genes, is linked to gas gangrene in humans and animals. Toxinotype B carries the cpa+, cpb+, and etx+ genes, and it is associated with haemorrhagic enteritis in calves, foals, and sheep, as well as dysentery in lambs. Toxinotype C, with the toxin genes, cpa+, cpb+, and cpe+/-, is linked to enterotoxaemia in sheep and necrotising enteritis in humans (pigbel), pigs, calves, goats, and foals. Toxinotype D, featuring cpa+, etx+, and cpe+/-, is associated with enterotoxaemia in lambs (pulpy kidney disease), goats, and cattle. Toxinotype E, containing cpa+, itx+, and cpe+/- genes, is linked to enterotoxaemia in calves and lambs. Toxinotype F, with cpa+ and cpe+, is associated with food poisoning and antibiotic-associated diarrhoea (AAD). Finally, toxinotype G, presenting cpa+ and netB+, is associated with avian necrotic enteritis [7]. Additionally, C. perfringens produces other clinically relevant accessory toxins, such as Perfringolysin O or theta toxin (PFO) and C. perfringens beta 2 toxin (CPB2). Although not used for toxinotyping, these toxins can act synergistically with extracellular toxins, impacting other toxins' expression, production levels, and virulence factors, thereby influencing overall disease progression [8]. The diversity of clostridial toxins and other virulence factors remains a critical study point. Therefore, in this review, we present a brief updated description of the structural and molecular characteristics, mechanisms of action, and genetic diversity of the main toxins and virulence factors associated with the tropism of C. perfringens toxinotypes across a diversity of hosts, including links to specific disease phenotypes. C. perfringens toxins: Biological properties, function, host tropism and diversity Localisation of toxins of C. perfringens and plasmids families The virulence of C. perfringens largely depends on its ability to produce toxins, which can be encoded on plasmids, chromosomally, or both. So far, seven groups of plasmids have been described for C. perfringens: transfer of clostridial plasmids (Tcp), pCP13 C. perfringens (Pcp), pIP404, phage-like, small plasmids or an unclassified group [9], and 4 recently the botulinum conjugation in C. perfringens (Bcp) group of plasmids encoding a new putative conjugation locus. The first significant group of plasmids carries the Tcp conjugation locus and genes encoding clinically relevant toxins such as cpb2 and cpe, iota binary toxin genes iap and ibp, as well as antibiotic resistance genes, including chloramphenicol, clindamycin, erythromycin, bacitracin, and lincomycin [10, 11]. Although a C. perfringens strain can carry up to five similar plasmids, it was recently demonstrated that the stability and inheritance of these plasmids are favoured by the ResP recombinase that catalyses the multimer resolution system and by differences in the type II partitioning systems (ParMRC) [13]. A second major group of C. perfringens conjugative plasmids is the pCP13-like plasmid family, which shares a highly conserved sequence at the Pcp locus [14, 15]. These plasmids carry a cpb2 consensus variant associated with disease in horses and piglets [16]. Theyalso have the novel C. perfringens binary enterotoxin (BEC/CPILE) [14], detected in the faeces of gastroenteritis patients, food [17], and healthy UK children [18]. The PiP404 plasmid family consists of small non-conjugative plasmids that encode a bacteriocin; however, they have not been identified to carry any toxin genes [19]. The newly identified Bcp plasmids contain a novel putative conjugation locus (Bcp) with a sequence like Clostridium botulinum plasmids and contigs that have the same plasmid mobilisation gene, mobC and the same zeta toxin-encoding gene as pJGS1984_5. Additionally, these plasmids encode VirB4 hom*ologous proteins and VirD4-like conjugation proteins distinct from the variants encoded by the Tcp and Pcp plasmids [20]. The presence of toxin-carrying plasmids and antibiotic-resistance genes in C. perfringens has been associated with its survival ability in multiple environments. Moreover, the high rate of plasmid transfer in this bacterium (2.9 × 10-1 to 3.8 × 10-2) also allows it to overcome fitness costs and segregation loss. This influences the transformation of toxinotypes in a given environment, especially within the gastrointestinal tract, implying impacts on disease outcomes [21]. 5 However, in some cases, the metabolic load and additional energy resources required by plasmid replication, disruption of essential host genes by plasmid gene integration, and plasmid-encoded molecules may negatively impact native host proteins, leading to decreased fitness in bacterial cells [21, 22]. In addition to plasmid-localized toxins, other toxins genes such as cpa (also plc) and pfo are located only on chromosomes [9]. Furthermore, cpe can be located on the chromosome, as shown in food poisoning isolates or on a large plasmid, as observed in non-foodborne gastrointestinal disease and veterinary isolates. The location of toxin-associated genes is expected to influence disease outcomes, as plasmid-encoded toxins can be transferred by conjugation to strains of C. perfringens toxinotype A that reside as part of the ‘normal’ intestinal microbiota, giving rise to new strains with enhanced colonisation and virulence traits. Conversely, chromosomal variants may be lost over evolutionary time due to mutation or deletion events and may cause shorter disease durations due to low colonisation levels and spread [21]. Toxin structure C. perfringens toxins are diverse and widely distributed. Structurally, clostridial toxins may have similar binding domains but different catalytic domains, suggesting possible recombination in toxin evolution. This recombination could occur through horizontal plasmid transfer or insertion-deletion processes, serving as adaptation, survival, and tropism mechanisms. This recombination could occur through horizontal plasmid transfer or insertion-deletion processes, serving as adaptation, survival, and tropism mechanisms [23]. As outlined below, some toxins are composed of different domains. In addition to allowing flexibility and dynamics in the toxin, these domains promote and facilitate the recognition of specific substrates in the target cell. Major toxins of C. perfringens 6 CPA is composed of 370 amino acids (aa) and is divided into two domains: the catalytic N-domain (CP1-249) and the membrane-binding C-terminal domain (CP247-370) [24] (Figure 1a). It involves cell membrane colonisation, haemolytic activity, and damaging action [24, 25, 26, 27]. CPB toxin, a 336 aa protein (Figure 1b) in the alpha-haemolysing family, is suggested to bind to its receptor or form the oligomer via its C-terminus (although the exact structure-function relationship is still under study) [28]. ETX toxin is composed of three domains: i) the amino-terminal domain, involved in the receptor binding process; ii) the central region domain, responsible for membrane insertion plus channel formation; and iii) the carboxy-terminal domain, involved in the activation of proteolysis [29] (Figure 1c). ITX is a binary toxin composed of two components: an enzymatic Ia (454 amino acids) and a binding Ib (875 amino acids), separated by 243 non-coding nucleotides. Both components, Ia and Ib, exhibit cytotoxic properties [30] (Figure 1d). CPE is a single-chain polypeptide consisting of 319 amino acids and three domains: I, II, and III. Domain I binds to the claudin-specific receptor, while domains II and III constitute the N-terminal region associated with pore-forming activity [31] (Figure 1e). NetB, a putative toxin gene encoding a 323 aa protein (Figure 1f), including a 30 aa secretion signal sequence, which has similarity to CPB toxin (38% identity), leading to its naming as necrotic enteritis β-like toxin, NetB [32]. Another clinically relevant accessory toxins PFO has 500 aa residues and a 27-residue signal peptide. It is composed of four domains, with domain four (D4) mediating the toxin's binding to the eukaryotic cell's plasma membrane [33]. The structure and receptor of CPB2 remain unclear. It is suggested that only a small fraction of amino acid segments adopt an α-helix conformation in both beta toxins (atypical and consensus), which is insufficient to cross a membrane. However, in the case of consensus 7 CPB2, some amino acid segments have been identified that could form transmembrane β-chains [34]. Mechanisms of action of C. perfringens toxin C. perfringens is a pathogen of significant clinical and veterinary importance, attributed to its capacity to induce disease in various hosts by producing different toxins (Table 1. Section A, B). Sphingomyelin and phosphatidylcholine degradation CPA, a toxin in all toxinotypes, is the primary determinant of virulence in clostridial myonecrosis caused by toxinotype A. The process is initiated by the toxin binding to the ganglioside receptor GM1a [35] (Figure 2a). This binding results in the degradation of sphingomyelin and phosphatidylcholine in the cell's plasma membrane, triggering proinflammatory responses during the early phases of gas gangrene [36]. This inflammatory process may advance to aggravated tissue infection, accompanied by secondary gas crepitation in the tissue and the development of necrotic in humans [37] (Figure 3). Formation of pores Pore-forming toxins (CPB, ETX, ITX, CPE, and NetB) allow Ca2+, Na+, Cl- entry and K+ loss with subsequent cellular electrochemical gradient depolarisation, ionic homeostasis alteration, and cell death [38]. CPB is the primary virulence determinant in necrotising human enteritis and enterotoxaemia caused by toxinotype C in sheep. It is also involved in haemorrhagic dysentery in sheep caused by toxinotype B (Figure 3). CPB binds with platelet endothelial cell adhesion molecule 1 (PECAM-1, also known as CD31) to trigger the formation of pores in the endothelial cell membrane. This process causes damage to the endothelial cells that make up the vasculature of the intestinal mucosa (Figure 2b), resulting in haemorrhagic and necrotising enteritis [39, 40]. The ETX toxin, found in toxinotypes B and D, is linked to enterotoxaemia in ruminants (Figure 3). It binds to receptors dependent on cholesterol and sphingomyelin present on the 8 outer membranes of endothelial cells lining the intestine and vascular cells in organs such as the brain, kidneys, and liver in ruminants. While the exact identity of the receptor is unknown, it is confirmed that ETX forms active pores, causing direct damage to intestinal endothelial cells. This process increases vascular and intestinal wall permeability, leading to intestinal epithelial cell detachment and necrosis, triggering enterotoxaemia, especially in goats (Figure 2c) [41, 42]. For ITX, present only in toxinotype E and associated with haemorrhagic enteritis in cattle (Figure 3), it is described that the lipolysis-stimulatedlipoprotein receptor (LSR) mediates the entry of toxin Ib into the host cell [43], allowing the formation of functional channels for ion movement and entry of Ia by endocytosis [44]. This interaction leads to degenerative changes in the mucosal epithelium of the small intestine, increased permeability of intestinal cell monolayers, and haemorrhagic lesions in the serosa and mucosa characteristic of enteritis (Figure 2d) [45]. The CPE toxin, found in toxinotype F, is recognised as the primary toxin causing food poisoning and gastrointestinal illnesses in humans, including AAD (Figure 3). Its mechanism of action involves binding to claudin receptors, crucial components of the tight junctions between epithelial or endothelial cells at the cell surface. This binding facilitates the formation of pores in the plasma membrane [46], removing claudins from the cell membrane and disrupting tight junctions between cells. Consequently, this disruption increases barrier permeability in the small intestine and colon, impairing absorption and diarrhoea (Figure 2e) [47, 48]. NetB, found in toxinotype G of C. perfringens and associated with avian necrotic enteritis (Figure 3), recognises cholesterol-free regions in the membrane of intestinal epithelial cells. It forms heptameric hydrophilic pores that allow the entry of ions, resulting in increased permeability. This process leads to focal, multifocal, or coalescent necrosis of enterocytes and, in severe cases, coagulative necrosis of the entire superficial mucosa separating the intestinal lamina of birds (Figure 2f) [49]. PFOA is encoded in most disease-associated C. perfringens strains, except for strains carrying cpe on the chromosome [50]. PFOA interacts with cell membrane cholesterol to 9 form pores [33]. It acts synergistically with CPA toxin to affect leukostasis [51] and promotes the expression of adhesion molecules and platelet-activating factors. These toxin-associated features contribute to developing myonecrosis (gas gangrene), haemorrhagic enteritis in calves, and septicaemia (intravascular haemolysis) in humans, including neonates [52, 53, 54]. Most recently, it was described that pfoA+ strains were associated with preterm infants, including those with necrotising enterocolitis, and were shown in vitro to cause significantly more intestinal cell damage than pfoA- strains [11]. CPB2 is an accessory toxin produced by C. perfrignens associated with porcine, equine, and bovine enteritis. While the structure of CPB2 and the identification of residues responsible for its antigenicity and association with the membrane of intestinal epithelial cells are under study [55], CPB2 has been reported as a pore-forming toxin. Its action mechanism involves forming cation-selective channels approximately 1.4 nm in diameter in lipid bilayers, leading to altered ion flux and increased intestinal permeability [56]. Genetic diversity and host specificity of C. perfringens toxins Understanding the molecular patterns influencing interspecies transmission and host adaptation (Figure 3 and Table 1) is crucial for disease prevention and control. The amino acid sequence diversity of the CPA toxin from 15 strains of C. perfringens suggests variation within the signal sequences (six positions), the N-domain (10 positions), C-domain (three positions), and the N- to C-domain linker peptide (one position), with most of these changes conserved potentially impacting the toxin’s resistance to degradation and biological activity [57]. The limited variability of the CPA toxin enhances its affinity for endothelial cells in the host, destroying the endothelial structure; this disruption in peripheral circulation results in myonecrosis. The study of CPB diversity in isolates of toxinotypes B and C has identified that some isolates of type C have four conserved amino acid changes, resulting in two natural variants of the toxin. The sequence modifications in CPB found in toxinotype C isolates lead to increased toxin sensitivity to trypsin, a higher affinity phenotype for host endothelial cells, and increased cytotoxicity. As a result, the molecular diversity of CPB in toxinotype C may 10 contribute to enterotoxaemia and necrotising enteritis in hosts deficient in trypsin, such as neonatal animals, individuals with diets rich in trypsin inhibitors, and those with pancreatic dysfunction [59]. The genetic diversity of the ETX toxin still needs to be fully elucidated. It has been described that the N-terminal amino acid sequence of ETX is identical in all strains examined to date [60]. Amino acid mutations within domain II of ETX affect pore characteristics and result in changes in cytotoxicity on renal distal tubule cells, leading to pulpy kidney disease in ruminants [61]. The diversity analysis of the ITX components, Ia and Ib, obtained from E-type strains from a calf and a human in France, reveals a very high identity, above 99%, at the amino acid level [62]. Although both toxins show conserved sequences, studies on iota toxin diversity are limited, and a better understanding of their contribution to enterotoxaemia in lambs, rabbits, and calves is required. The genetic diversity analysis of the CPE toxin gene located on the chromosome, associated with food poisoning, and on plasmids linked to AAD reveals identical nucleotide sequences with similar cytotoxic activity, suggesting a physiological basis for the genotype-disease relationship [63]. Strains harbouring the CPE toxin on the chromosome exhibit superior heat resistance compared to non-food human gastrointestinal disease strains carrying the plasmid CPE gene. This resistance favours their survival in food and contributes to the development of food poisoning [64]. Analysis of NetB sequence diversity in different avian isolates reveals a high degree of conservation despite the presence of an amino acid variant, A168T, in some isolates. This variant does not significantly affect the physical properties of the encoded protein and does not differ in cytotoxicity. These genetic characteristics of the NetB toxin provide additional evidence that NetB constitutes an essential virulence factor in the pathogenesis of necrotic enteritis in birds [65]. Although the genetic diversity of the PFOA toxin and its influence on disease development is not well understood, the nucleotide sequence of the pfoA gene has been reported to exhibit 11 approximately 86% identity with a membrane-damaging thiol-activated alveolysin detected in C. perfringens isolates IQ2 (type E) and IQ3 (BEC-positive) [18]. This toxin also shares a 60% identity with the streptolysin O gene and a 48% identity with the pneumolysin gene [66]. The expression of CPB2 toxin in C. perfringens strains varies according to host species, with two variants identified: consensus in porcine strains and atypical in non-porcine species. While the genotype-phenotype correlation is high in the consensus variant, the atypical variant shows only a 50% correlation. Furthermore, the atypical genes in types D and E are more similar to each other than isolates for non-porcine kinds A, B, and C, suggesting divergent evolution and a possible explanation for tropism by the host [67]. A detailed description of other accessory toxins is presented in Table 1—section B. Conclusions In recent years, efforts have been made to understand how chemical structures underpin the molecular and cellular mechanisms of action of bacterial toxins, including the regions involved in host cell recognition and the receptors that confer specificity to each toxin in different hosts, leading to the development of the pathological state. However, although C. perfringens is part of the 'resident' microbiota, studies need to probe the genetic diversity of each toxin in healthy animals and humans. This is crucial in understandingwhether changes in the toxins render them less toxic or lead to their deactivation, ultimately contributing to a more 'commensal' lifestyle. Future research could focus on additional sampling of isolates from healthy and diseased hosts and ecological niches to unravel broader evolutionary aspects, phylogenetic relationships, and genetic diversity of this bacterium, including a comprehensive examination of toxin-associated gene makeup. Further elucidation of the precise association between toxin-specific molecular and structural changes linked with host affinity will enhance knowledge about the diversity of this important pathogen. 12 In addition to the above, gut microbiota studies are needed to understand how they may influence resistance to C. perfringens colonisation, competition for space and specific nutrients, or toxin function. Understanding the interactions, genetic diversity, precise targets, and selective toxicity mechanism of C. perfringens toxins may open new avenues for preventative and therapy development. This could involve the rational design of potent pharmacological inhibitors/compounds and identifying new vaccine targets to counteract the effects of toxins, ultimately improving human and animal health. Funding This study was financially supported by the Ministerio de Ciencia Tecnología e Innovación (Minciencias) within the framework of the project 722289684653 contract. 613-2021. A.C. was supported by the Universidad de Boyacá, Tunja, Colombia. Wellcome Trust Investigator Award 220876/Z/20/Z; the Biotechnology and Biological Sciences Research Council (BBSRC) Institute Strategic Programme, Gut Microbes and Health BB/R012490/1, and its constituent projects BBS/E/F/000PR10353 and BBS/E/F/000PR10356; and the BBSRC Institute Strategic Programme Food Microbiome and Health BB/X011054/1 and its constituent project BBS/E/F/000PR13631. Conflict of interest The authors declare that they have no conflict of interest. References 1. Li J, Paredes-Sabja D, Sarker MR, et al. Clostridium perfringens sporulation and sporulation‐associated toxin production. Microbiol Spectr. 2016:331-347. doi: 10.1128/microbiolspec 2. Kiu R, Hall LJ. An update on the human and animal enteric pathogen Clostridium perfringens. Emerging microbes infections. Emerg Microbes Infect. 2018;7(1):1-15. doi: 10.1038/s41426-018-0144-8 3. Grass JE, Gould LH, Mahon BE. Epidemiology of foodborne disease outbreaks caused by Clostridium perfringens, United States. Foodborne Pathog Dis.1998–2010. 2013;10(2):131-136. doi: 10.1089/fpd.2012.1316. 13 4. Scallan E, Hoekstra RM, Angulo FJ, et al. Foodborne illness acquired in the United States major pathogens. Emerg Infect Dis. 2011;17(1):7. doi: 10.3201/eid1701.p11101. 5. Mahamat Abdelrahim A, Radomski N, Delannoy S, et al. Large-scale genomic analyses and toxinotyping of Clostridium perfringens implicated in foodborne outbreaks in France. Front Microbiol. 2019;10:777. doi: 10.3389/fmicb.2019.00777. 6. Bhattacharya A, Shantikumar S, Beaufoy D, et al. Outbreak of Clostridium perfringens food poisoning linked to leeks in cheese sauce: an unusual source. Epidemiol Infect. 2020;148. doi: 10.1017/S095026882000031X. 7. Rood JI, Adams V, Lacey J, et al. Expansion of the Clostridium perfringens toxin-based typing scheme. Anaerobe. 2018;53:5-10. doi: 10.1016/j.anaerobe.2018.04.011. 8. Gohari IM, A. Navarro M, Li J, et al. Pathogenicity and virulence of Clostridium perfringens. Virulence.2021;12(1):723-753. doi: 10.1080/21505594.2021. 9. Li J, Adams V, Bannam TL, et al. Toxin plasmids of Clostridium perfringens. Microbiol Mol Biol Rev. 2013;77(2):208-233. doi: 10.1128/MMBR.00062-12. 10. Adams V, Han X, Lyras D, et al. Antibiotic resistance plasmids and mobile genetic elements of Clostridium perfringens. Plasmid. 2018;99:32-39. doi: 10.1016/j.plasmid.2018.07.002. 11. Kiu R, Shaw AG, Sim K, et al. Particular genomic and virulence traits associated with preterm infant-derived toxigenic Clostridium perfringens strains. Nat Microbiol. 2023:1-16. doi: 10.1038/s41564-023-01385-z. 12. Revitt-Mills S, Lao C, Archambault M, et al. The Tcp plasmids of Clostridium perfringens require the resP gene to ensure stable inheritance. Plasmid. 2020;107:102461. doi: 10.1016/j.plasmid.2019.102461. 13. Watts TD, Traore DA, Atkinson SC, et al. The Specificity of ParR Binding Determines the Incompatibility of Conjugative Plasmids in Clostridium perfringens. mBio. 2022;13(4):e01356-22. doi: 10.1128/mbio.01356-22. 14. Ueda K, Kawahara K, Kimoto N, et al. Analysis of the complete genome sequences of Clostridium perfringens strains harbouring the binary enterotoxin BEC gene and comparative genomics of pCP13-like family plasmids. BMC Genomics. 2022;23(1):1-14. doi: 10.1186/s12864-022-08453-4. 15. Watts TD, Vidor CJ, Awad MM, et al. pCP13, a representative of a new family of conjugative toxin plasmids in Clostridium perfringens. Plasmid. 2019;102:37-45. doi: 10.1016/j.plasmid.2019.02.002. 16. Gibert M, Jolivet-Reynaud C, Popoff MR. Beta2 toxin, a novel toxin produced by Clostridium perfringens. Gene.1997;203(1):65-73. doi: 10.1016/s0378-1119(97)00493-9. 17. Yonogi S, Matsuda S, Kawai T, et al. BEC, a novel enterotoxin of Clostridium perfringens found in human clinical isolates from acute gastroenteritis outbreaks. Infect Immun.2014;82(6):2390-2399. doi: 10.1128/IAI.01759-14. 18. Kiu R, Sim K, Shaw A, et al. Genomic analysis of Clostridium perfringens BEC/CPILE-positive, toxinotype D and E strains isolated from healthy children. Toxins (Basel).2019;11(9):543. doi: 10.3390/toxins11090543. 19. Garnier T, Cole ST. Complete nucleotide sequence and genetic organization of the bacteriocinogenic plasmid, pIP404, from Clostridium perfringens. Plasmid. 1988;19(2):134-150. doi: 10.1016/0147-619x(88)90052-2. 20. Gulliver EL, Adams V, Marcelino VR, et al. Extensive genome analysis identifies novel plasmid families in Clostridium perfringens. Microb Genom. 2023;9(4). doi: 10.1099/mgen.0.000995. 14 21. Brynestad S, Sarker MR, McClane BA, et al. Enterotoxin plasmid from Clostridium perfringens is conjugative. Infect Immun. 2001;69(5):3483-3487. doi: 10.1128/IAI.69.5.3483-3487.2001. 22. Carroll AC, Wong A. Plasmid persistence: costs, benefits, and the plasmid paradox. Can J Microbiol. 2018;64(5):293-304. doi: 10.1139/cjm-2017-0609. 23. Kumar R, Feltrup TM, Kukreja RV, et al. Evolutionary features in the structure and function of bacterial toxins. Toxins (Basel). 2019;11(1):15. doi: 10.3390/toxins11010015. 24. Oda M, Terao Y, Sakurai J, et al. Membrane-binding mechanism of Clostridium perfringens alpha-toxin. Toxins (Basel). 2015;7(12):5268-5275. doi: 10.3390/toxins7124880. 25. Nagahama M, Mukai M, Morimitsu S, et al. Role of the C‐domain in the biological activities of Clostridium perfringens alpha‐toxin. Microbiol Immunol. 2002;46(10):647-655. doi: 10.1111/j.1348-0421.2002.tb02748.x. 26. Naylor CE, Eaton JT, Howells A, et al. Structure of the key toxin in gas gangrene. Nat Struct Biol. 1998;5(8):738-746. doi: 10.1038/1447. 27. Xu C, She Y, Lin Y, et al. Molecular structure and function of the carboxy‐terminus of the alpha‐toxin from Clostridium perfringens type A. J Anim Physiol Anim Nutr (Berl). 2020;104(2):725-734. doi: 10.1111/jpn.13274. 28. Sakurai J, Duncan CL. Some properties of beta-toxin produced by Clostridium perfringens type C. Infect Immun. 1978;21(2):678-680. doi: 10.1128/iai.21.2.678-680.1978. 29. Stiles BG, Barth G, Barth H, et al. Clostridium perfringens epsilon toxin: a malevolent molecule for animals and man? Toxins. 2013 Nov 12;5(11):2138-60. doi: 10.3390/toxins5112138. 30. Perelle S, Gibert M, Boquet P, et al. Characterization of Clostridium perfringens iota-toxin genes and expression in Escherichia coli. Infect Immun. 1993;61(12):5147-5156. doi: 10.1128/iai.61.12.5147-5156.1993. 31. Kitadokoro K, NishimuraK, Kamitani S, et al. Crystal structure of Clostridium perfringens enterotoxin displays features of β-pore-forming toxins. J Biol Chem. 2011;286(22):19549-19555. doi: 10.1074/jbc.M111.228478. 32. Rood JI, Keyburn AL, Moore RJ. NetB and necrotic enteritis: the hole movable story. Avian Pathol. 2016;45(3):295-301. doi: 10.1080/03079457.2016.1158781. 33. Verherstraeten S, Goossens E, Valgaeren B, et al. Perfringolysin O: the underrated Clostridium perfringens toxin? Toxins (Basel). 2015;7(5):1702-1721. doi: 10.3390/toxins7051702. 34. Benz R, Piselli C, Hoxha C, et al. Clostridium perfringens Beta2 toxin forms highly cation-selective channels in lipid bilayers. Eur Biophys J. 2022;51(1):15-27. doi: 10.1007/s00249-021-01577-7. 35. Oda M, Kabura M, Takagishi T, et al. Clostridium perfringens alpha-toxin recognizes the GM1a-TrkA complex. J Biol Chem. 2012;287(39):33070-33079. doi: 10.1074/jbc.M112.393801. 36. Takagishi T, Oda M, Kabura M, et al. Clostridium perfringens alpha-toxin induces Gm1a clustering and Trka phosphorylation in the host cell membrane. PLoS One. 2015;10(4):e0120497. doi: 10.1371/journal.pone.0120497. 37. Garcia NM, Cai J. Aggressive soft tissue infections. Surg Clin North Am. 2018;98(5):1097-1108. doi: 10.1016/j.suc.2018.05.001. 38. Gohari IM, Parreira VR, Nowell VJ, et al. A novel pore-forming toxin in type A Clostridium perfringens is associated with both fatal canine hemorrhagic gastroenteritis and 15 fatal foal necrotizing enterocolitis. PLoS One. 2015;10(4):e0122684. doi: 10.1371/journal.pone.0122684. 39. Lencer WI. Everything Illuminated-Clostridium perfringens β-toxin. Cell Host Microbe. 2020;28(1):5-6. doi: 10.1016/j.chom.2020.06.006. 40. Bruggisser J, Tarek B, Wyder M, et al. CD31 (PECAM-1) serves as the endothelial cell-specific receptor of clostridium perfringens β-toxin. Cell Host Microbe. 2020;28(1):69-78. e6. doi: 10.1016/j.chom.2020.05.003. 41. Goldstein J, Morris WE, Loidl CF, et al. Clostridium perfringens epsilon toxin increases the small intestinal permeability in mice and rats. PLoS One. 2009;4(9):e7065. doi: 10.1371/journal.pone.0007065. 42. Alves GG, de Ávila RAM, Chávez-Olórtegui CD, et al. Clostridium perfringens epsilon toxin: the third most potent bacterial toxin known. Anaerobe. 2014; 30:102-107. doi: 10.1016/j.anaerobe.2014.08.016. 43. Schmidt G, Papatheodorou P, Aktories K. Novel receptors for bacterial protein toxins. Curr Opin Microbiol. 2015; 23:55-61. doi: 10.1016/j.mib.2014.11.003. 44. Richard JF, Mainguy G, Gibert M, et al. Transcytosis of iota‐toxin across polarized CaCo‐2 cells. Mol Microbiol. 2002; 43(4):907-917. doi: 10.1046/j.1365-2958.2002.02806.x. 45. Redondo LM, Redondo EA, Dailoff GC, et al. Effects of Clostridium perfringens iota toxin in the small intestine of mice. Anaerobe. 2017;48:83-88. doi: 10.1016/j.anaerobe.2017.07.007. 46. Freedman JC, Shrestha A, McClane BA. Clostridium perfringens enterotoxin: action, genetics, and translational applications. Toxins (Basel). 2016;8(3):73. doi: 10.3390/toxins8030073. 47. Eichner M, Augustin C, Fromm A, et al. In colon epithelia, Clostridium perfringens enterotoxin causes focal leaks by targeting claudins which are apically accessible due to tight junction derangement. J Infect Dis. 2018;217(1):147-157. doi: 10.1093/infdis/jix485. 48. Sonoda N, Furuse M, Sasaki H, et al. Clostridium perfringens enterotoxin fragment removes specific claudins from tight junction strands: evidence for direct involvement of claudins in tight junction barrier. J Cell Biol. 1999;147(1):195-204. doi: 10.1083/jcb.147.1.195. 49. Keyburn AL, Bannam TL, Moore RJ, et al. NetB, a pore-forming toxin from necrotic enteritis strains of Clostridium perfringens. Toxins (Basel). 2010;2(7):1913-1927. doi: 10.3390/toxins2071913. 50. Deguchi A, Miyamoto K, Kuwahara T, et al. Genetic characterization of type A enterotoxigenic Clostridium perfringens strains. PLoS One. 2009;4(5):e5598. doi: 10.1371/journal.pone.0005598. 51. Ellemor DM, Baird RN, Awad MM, et al. Use of genetically manipulated strains of Clostridium perfringens reveals that both alpha-toxin and theta-toxin are required for vascular leukostasis to occur in experimental gas gangrene. Infect Immun. 1999;67(9):4902-4907. doi: 10.1128/IAI.67.9.4902-4907.1999. 52. Bryant AE, Chen RY, Nagata Y, et al. Clostridial Gas Gangrene. II. Phospholipase C-Induced Activation of Platelet gpIIbIIIa Mediates Vascular Occlusion and Myonecrosis in Clostridium perfringens Gas Gangrene. J Infect Dis. 2000;182(3):808-815. doi: 10.1086/315757. 53. Bryant AE, Stevens DL. Clostridial myonecrosis: new insights in pathogenesis and management. Curr Infect Dis Rep. 2010;12(5):383-391. doi: 10.1007/s11908-010-0127-y. 16 54. Bryant A, Bayer C, Aldape M, et al. Clostridium perfringens phospholipase C-induced platelet/leukocyte interactions impede neutrophil diapedesis. J Med Microbiol. 2006;55(5):495-504. doi: 10.1099/jmm.0.46390-0. 55. Gurjar AA, Yennawar NH, Yennawar HP, et al. Expression, crystallization and preliminary X-ray diffraction studies of recombinant Clostridium perfringens β2-toxin. Acta Crystallogr Sect F Struct Biol Cryst Commun. 2007;63(6):484-487. doi: 10.1107/S1744309107020313. 56. Benz R, Piselli C, Hoxha C, et al. Clostridium perfringens Beta2 toxin forms highly cation-selective channels in lipid bilayers. Eur Biophys J. 2021:1-13. doi: 10.1007/s00249-021-01577-7. 57. Titball RW, Naylor CE, Basak AK. The Clostridium perfringens α-toxin. Anaerobe.1999;5(2):51-64. doi: 10.1006/anae.1999.0191. 58. Takehara M, Bandou H, Kobayashi K, et al. Clostridium perfringens α-toxin specifically induces endothelial cell death by promoting ceramide-mediated apoptosis. Anaerobe.2020;65:102262. doi: 10.1016/j.anaerobe.2020. 59. Theoret JR, Uzal FA, McClane BA, et al. Identification and characterization of Clostridium perfringens beta toxin variants with differing trypsin sensitivity and in vitro cytotoxicity activity. Infect Immun. 2015;83(4):1477-1486. doi: 10.1128/IAI.02864-14. 60. Freedman JC, Li J, Uzal FA, et al. Proteolytic processing and activation of Clostridium perfringens epsilon toxin by caprine small intestinal contents. mBio. 2014;5(5):e01994-14. doi: 10.1128/mBio.01994-14. 61. Dorca-Arévalo J, Gómez de Aranda I, Blasi J. New mutants of epsilon toxin from Clostridium perfringens with an altered receptor-binding site and cell-type specificity. Toxins (Basel). 2022;14(4):288. doi: 10.3390/toxins14040288. 62. Diancourt L, Sautereau J, Criscuolo A, et al. Two Clostridium perfringens type E isolates in France. Toxins (Basel). 2019;11(3):138. doi: 10.3390/toxins11030138. 63. Collie RE, Kokai-Kun JF, McClane BA. Phenotypic Characterization of Enterotoxigenic Clostridium perfringens Isolates from Non-foodborne Human Gastrointestinal Diseases. Anaerobe. 1998;4(2):69-79. doi: 10.1006/anae.1998.0152. 64. Sparks SG, Carman RJ, Sarker MR, et al. Genotyping of enterotoxigenic Clostridium perfringens fecal isolates associated with antibiotic-associated diarrhea and food poisoning in North America. J Clin Microbiol. 2001;39(3):883-888. doi: 10.1128/JCM.39.3.883-888.2001. 65. Keyburn AL, Yan XX, Bannam TL, et al. Association between avian necrotic enteritis and Clostridium perfringens strains expressing NetB toxin. Vet Res. 2010;41(2). doi: 10.1051/vetres/2009069. 66. Tweten RK. Nucleotide sequence of the gene for perfringolysin O (theta-toxin) from Clostridium perfringens: significant hom*ology with the genes for streptolysin O and pneumolysin. Infect Immun. 1988;56(12):3235-3240. doi: 10.1128/iai.56.12.3235-3240.1988. 67. Jost BH, Billington SJ, Trinh HT, et al. Atypical cpb2 genes, encoding beta2-toxin in Clostridium perfringens isolates of nonporcine origin. Infect Immun. 2005;73(1):652-656. doi: 10.1128/IAI.73.1.652-656.2005. 68. Uzal FA, Freedman JC, Shrestha A, et al. Towards an understanding of the role of Clostridium perfringenstoxins in human and animal disease. Future Microbiol. 2014;9(3):361-377. doi: 10.2217/fmb.13.168. 17 69. Chakravorty A, Awad MM, Hiscox TJ, et al. The cysteine protease α-Clostripain is not essential for the pathogenesis of Clostridium perfringens-mediated myonecrosis. PLoS One.2011;6(7):e22762. doi: 10.1371/journal.pone.0022762. 70. Roos S, Wyder M, Candi A, et al. Binding studies on isolated porcine small intestinal mucosa and in vitro toxicity studies reveal lack of effect of C. perfringens beta-toxin on the porcine intestinal epithelium. Toxins (Basel). 2015;7(4):1235-1252. doi: 10.3390/toxins7041235. 71. Songer JG, Miskimins DW. Clostridial abomasitis in calves: case report and review of the literature. Anaerobe. 2005;11(5):290-294. doi: 10.1016/j.anaerobe.2004.12.004. 72. Uzal FA, Vidal JE, McClane BA, et al. Clostridium perfringens toxins involved in mammalian veterinary diseases. Open Toxinology J. 2010;2:24. doi: 10.2174/1875414701003020024 73. Revitt-Mills SA, Rood JI, Adams V. Clostridium perfringens extracellular toxins and enzymes: 20 and counting. Microb. Australia 2015;36(3):114-117. doi:10.1071/MA15039 74. Coursodon C, Glock R, Moore K, et al. TpeL-producing strains of Clostridium perfringens type A are highly virulent for broiler chicks. Anaerobe. 2012;18(1):117-121. doi: 10.1016/j.anaerobe.2011.10.001. 75. Van Damme L, Cox N, Callens C, et al. Protein truncating variants of colA in Clostridium perfringens type G strains. Front Cell Infect Microbiol. 2021;11. doi: 10.3389/fcimb.2021.645248. 76. Wang YH. Sialidases from Clostridium perfringens and their inhibitors. Front Cell Infect Microbiol. 2020;9:462. doi: 10.3389/fcimb.2019.00462. 77. Li J, McClane BA. NanH Is Produced by Sporulating Cultures of Clostridium perfringens Type F Food Poisoning Strains and Enhances the Cytotoxicity of C. perfringens Enterotoxin. mSphere. 2021;6(2): e00176-21. doi: 10.1128/mSphere.00176-21. 78. Hynes WL, Walton SL. Hyaluronidases of Gram-positive bacteria. FEMS Microbiol Lett. 2000;183(2):201-207. doi: 10.1111/j.1574-6968.2000.tb08958.x. 79. Canard B, Garnier T, Saint-Joanis B, et al. Molecular genetic analysis of the nagH gene encoding a hyaluronidase of Clostridium perfringens. Mol Gen Genet. 1994;243(2):215-224. doi: 10.1007/BF00280319. 18 Table 1. Classification of toxins and other virulence factors of C. perfringens Virulence factor or toxin Gene Biological Properties and Function Associated disease Reference Section A. Toxins used for C. perfringens typing alpha-toxin plc, cpa - Modifies cell membranes by enzymatic activity. - Zinc-dependent phospholipase C which degrades phosphatidylcholine and sphingomyelin, in the eukaryotic cell membranes. - Gas gangrene. - Necrotic enteritis in chickens. - Enteritis in calves and in piglets. - Sudden infant death syndrome (SIDS). - Inflammatory diseases. [37] Clostridium perfringens enterotoxin cpe - Intestinal damage, severe villus shortening, along with epithelial necrosis and desquamation. - Human foodborne illness - Antibiotic-associated diarrhoea - Human enteritis necroticans [68] 19 - Sporadic diarrhoea - Enteric diseases in swine, cattle, horses, sheep and goats and wild animals such as deer and bears. Alpha- clostripain ccp - Processing of secreted proteins - Potential to affect the levels of active extracellular toxins. - Possible Clostridial myonecrosis-associated [69] Epsilon toxin etx - Aerolysin-like b-pore-forming toxin family. - Intestine: Epithelial cell detachment and cell necrosis, lamina propria haemorrhaging and polymorphonucleocyte infiltration. - Affect renal system, brain, cardiorespiratory system, and pleura. - Hemolysin. - Enterotoxaemia in domestic ruminants (sheep and goats) - Illness in humans related to the epsilon toxin? [42] 20 - Blood pressure elevation, increased contractility of smooth muscle, vascular permeability increase. Beta toxin cpb - Beta-barrel pore-forming toxin family and forms oligomeric pores in several susceptible immune cell lines. - Vascular necrosis and marked inflammatory reactions result from a direct interaction of CPB with vascular endothelial cells. - Destruction of jejunal and ileal villous tip epithelium. - Necrotizing enteritis (NE) in pigs, sheep, goats, calves, and humans. [70] Iota toxin itx - Necrosis of the superficial epithelium with relative sparing of - Lethal necrotizing - Enteritis and sudden death [71] 21 the crypt epithelium, and submucosal haemorrhage and transmural haemorrhage. in beef calves. Section B. Accessory toxins Clostridium Perfringens beta 2 toxin cpb2 - Haemorrhage and necrosis of the small and large intestines. - Degenerated and necrotic desquamated epithelial cells, cell debris, inflammatory cells. - Enteric diseases in swine, cattle, horses, sheep and goats and wild animals such as deer and bears. [72] Perfringolysin O pfoA - Tissue destruction and an anti-inflammatory response. - Vascular accumulation of leukocytes within blood vessels and the extracellular matrix of host tissues. - Disruption of endothelial, local - Myonecrosis (gas gangrene), haemorrhagic enteritis in calves and, septicaemia (intravascular haemolysis) in humans. - Associated with [11] 22 edema and systemic shock and multiorgan failure. Necrotising Enterocolitis in human neonates. Necrotic enteritis E-like toxin netE - Pore-forming toxin within the Leukocidin/Hemolysin superfamily. [73] Necrotic enteritis F-like toxin netF - Putative beta-pore-forming toxin. - Pore-forming toxin within the Leukocidin/Hemolysin superfamily. - Canine haemorrhagic gastroenteritis. - Foal necrotizing enteritis. [38] Necrotic enteritis G-like toxin netG - Putative beta-pore-forming toxin. - Cytotonic effects, such as proinflammatory effects. - No confirmed association with disease. [73] Necrotic enteritis B-like toxin netB - Pore-forming toxin. - Necrotic enteritis of chickens. [49] Toxin C. perfringens large cytotoxin TpeL - Ras-specific glucosyltransferase activity inactivating the Ras - Associated with avian necrotic enteritis. [74] 23 signalling pathway leading to apoptosis. - Cytotoxic effects (morphological changes such as enlargement and the rounding of Vero cells). Collagenase ColA - Degradation of the extracellular matrix due to their ability to digest native collagen. - Associated with avian necrotic enteritis. [75] Binary enterotoxin of C. perfringens becA,becB - Enterotoxic activity. - Fluid accumulation in mice. - Acute gastroenteritis in humans. [17] Neuramidases Secreted neuramidase nanJ - Increase CPE-induced cytotoxicity and CH-1 pore formation in Caco-2 cells. - Contribute to food poisoning caused by F-type c-cpe strains carrying the nanH and nanJ genes. [76] 24 Secreted major neuramidase nanI - Synergy with CPA toxin, ETX toxin, CPB-toxin and CPE toxin - Sialidase activity. - Promotes the colonization of C. perfringens in the intestinal tract and enhances the cytotoxic activity. - Gas gangrene? [76] Non-secreted neuramidase nanH - Enhance CPE cytotoxicity. - Reducing host cell surface charge repulsion during CPE binding or removing sialic acid residues that sterically interfere with CPE binding. - Intestinal pathology. [77] Hyaluronidases Hyaluronidases nagI,nagJ, nagH, nagK nagL - Facilitates the spread of the major tissue-damaging α-toxin, thereby potentiating its cytolytic activity. - Gas gangrene [78, 79] 25 - Degrade hyaluronate cell surface coatings. - Act on connective tissue during gas gangrene. - Increased permeability of the connective tissues. Figure legends Figure 1. Schematic representation showing domain structure of Clostridium pefringens main toxins (A). CPA toxin (B). CPB toxin (C). ETX toxin (D). ITX toxin (E). CPE toxin and (F). NetB toxin are shown. Numbers indicate amino acids that mark domain boundaries. Figure 2. Action mechanisms of the primary C. perfringens toxins used for toxin typing. Molecular mechanism of action of the major toxins of C. perfringens. (A). CPA toxin: CPA toxin interacts with GM1a, hydrolysing phosphatidylcholine (PC) and sphingomyelin (SM), resulting in the formation of diacylglycerol (DAG) and ceramide (CER) with Tropomyosin kinase A receptor (TrKA) activation and triggers the activation of an intracellular signalling 26 cascade with Interleukin - 8 (IL-8) release. The activation of phosphatidyl inositol 3 (IP3) promotes intracytoplasmic calcium (Ca+) entry (B). CPB toxin: CPB binds to platelet endothelial cell adhesion molecule-1 (PECAM-1) with subsequent release of adenosine triphosphate (ATP) and formation of pores that allow ion exchange to and from the cell (C). ETX toxin: ETX toxin interacts with protein "myelin and lymphocytes" (MAL), forming an active pore that induces ion transport and exchange across the cell membrane (D). ITX toxin: The binding of Ib to the lipolysis-stimulated lipoprotein receptor (LSR) receptor mediates its entry into the host cell, promoting the formation of channels for the entry of Ia by endocytosis with subsequent depolymerization of actin filaments, generating morphological changes and alteration of cell permeability (E). CPE toxin: the CPE toxin binds to claudin receptors, contributing to the formation of a pore on the cell surface with ion exchange and osmotic imbalance. (F). NetB toxin recognizes cholesterol-free regions in cell membranes by forming heptameric hydrophilic pores that allow the entry of ions such as Na+, Cl- y, and Ca 2+. Figure 3. Schematic representation of principal hosts for each C. perfringens toxinotypes—each box representing a different toxinotype. Supplementary material Figure supplementary 1. Toxinotyping scheme of C. perfringens. The main toxins are used as the primary typing scheme in seven toxin types (A-G). 27 28 29 30 Comparative Immunology, Microbiology and Infectious Diseases 102 (2023) 102074Available online 10 October 20230147-9571/© 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).Acquisition site-based remodelling of Clostridium perfringens- and Clostridioides difficile-related gut microbiota Giovanny Herrera a, Laura Vega a, Anny Camargo a,b, Manuel Alfonso Patarroyo c,d,e, Juan David Ramírez a,f, Marina Muñoz a,* a Centro de Investigaciones en Microbiología y Biotecnología -UR (CIMBIUR), Facultad de Ciencias Naturales, Universidad del Rosario, Bogotá 111221, Colombia b Health Sciences Faculty, Universidad de Boyacá, Tunja, Colombia c Molecular Biology and Immunology Department, Fundación Instituto de Inmunología de Colombia (FIDIC), Bogotá D.C. 111321, Colombia d Microbiology Department, Faculty of Medicine, Universidad Nacional de Colombia, Bogotá D.C. 111321, Colombia e Health Sciences Division, Main Campus, Universidad Santo Tomás, Bogotá D.C. 110231, Colombia f Molecular Microbiology Laboratory, Department of Pathology, Molecular and Cell-Based Medicine, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA A R T I C L E I N F O Keywords: C. perfringens C. difficile Gut microbiota A B S T R A C T Introduction: Clostridium perfringens is a gram-positive, anaerobic sporulating bacillus which can infect several hosts, thereby being considered the causative agent of many gut illnesses. Some studies have suggested that C. perfringens’s virulence factors may negatively affect gut microbiota homeostasis by decreasing beneficial bacteria; however, studies have failed to evaluate the simultaneous presence of other pathogenic bacteria, such as C. difficile (another sporulating bacillus known to play a role in gut microbiota imbalance). Conscious of the lack of compelling data, this work has ascertained how such microorganisms’ coexistence can be associated with a variation in gut microbiota composition, compared to that of C. perfringens colonisation. Methods: PCR was thus used for identifying C. perfringens and C. difficile in 98 samples. Amplicon-based sequencing of 16S- and 18S-rRNA genes’ V4 hypervariable region from such samples was used for deter-mining the microbiota’s taxonomical composition and diversity. Results: Small differences were observed in bacterial communities’ taxonomic composition and diversity; such imbalance was mainly associated with groups having hospital-acquired diarrhoea. Conclusion: The alterations reported herein may have been influenced by C. difficile and diarrhoea acquisition site, despite C. perfringens’ ability to cause alterations in microbiota due to its virulence factors. Our findings highlight the need for a holistic view of gut microbiota. 1. Introduction Multiple pathogens can affect gut microbiota homeostasis due to the production of virulence factors (such as toxins) by detrimentally modi-fying the gut environment [1]. Clostridium perfringens (CPF) is a gram-positive, anaerobic, spore-forming bacteria which is found in varied environments, e.g., soil, food and human and animal gut micro-biota [2]. Clinical manifestations are usually associated with this bac-terium, i.e. gas gangrene, necrotising enteritis, food poisoning, colitis and other non-specific gastrointestinal alterations [2]. The CPF genome can encode more than 20 toxins, six of which are clinically relevant and useful for toxinotyping: alpha (CPA), beta (CPB), epsilon (ETX), iota (ITX), enterotoxin (CPE) and necrotic enteritis-causing B-like (NetB) toxins [3]. Some CPF strains can carry accessory enzymes, thereby increasing their virulence (considering this species’ strong genomic plasticity). CPF is considered to be the second cause of food poisoning in the USA and Canada; it causes around 5% of outbreaks and 4% of hospitalisations [4]. Few CPF infection studies have been carried out in Colombia; they have revealed 18.3–41.3% infection frequency in patients who have contracted community- or intrahospital-acquired diarrhoea [5]. Such high frequency has been explained by CPF’s occurrence as a gut Abbreviations: CPF, Clostridium perfringens; CDI, Clostridioides difficile Infection; CO, community; HCFO, healthcare facility onset; ASV, amplicon sequence variants; SCFA, short-chain fatty acids. * Corresponding author. E-mail address: claudia.munoz@urosario.edu.co (M. Muñoz). Contents lists available at ScienceDirect Comparative Immunology, Microbiology and Infectious Diseases journal homepage: www.elsevier.com/locate/cimid https://doi.org/10.1016/j.cimid.2023.102074 Received 7 August 2023; Received in revised form 21 September 2023; Accepted 29 September 2023 mailto:claudia.munoz@urosario.edu.cowww.sciencedirect.com/science/journal/01479571https://www.elsevier.com/locate/cimidhttps://doi.org/10.1016/j.cimid.2023.102074https://doi.org/10.1016/j.cimid.2023.102074https://doi.org/10.1016/j.cimid.2023.102074http://crossmark.crossref.org/dialog/?doi=10.1016/j.cimid.2023.102074&domain=pdfhttp://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/Comparative Immunology, Microbiology and InfectiousDiseases 102 (2023) 1020742microbiota member in healthy individuals [2,6]. Further research is needed to ascertain the effect of CPF colonisation on human gut microbiota, considering that some of this bacterium’s strains can acquire genes from plasmids and cause intestinal disease [6]. A few studies have evaluated CPF-related gut microbiota changes [7–9], finding a decrease in beneficial species such as the Bacteroides fragilis group, Bifidobacterium spp. and Lactobacillus spp. However, such studies have not considered the presence of other pathogens, such as Clostridioides difficile whose coexistence with CPF can directly and/or indirectly affect gut microbiota composition [10–12], thereby biasing microbiota profiling estimation. Clostridioides difficile Infection (CDI)’s prominent role is related to its ability to produce toxins (A, B and binary toxins) and other virulence factors (genetic diversity, sporulation and antibiotic resistance) [13]. Such virulence factors can negatively affect patients’ health, causing various clinical manifestations and disrupting intestinal ecosystem homeostasis [11]. Future studies should determine whether bacterial pathogens’ coexistence is associated with changes in gut microbiota regarding patients suffering diarrhoea. This study was thus aimed at describing the gut microbiota compo-sition (bacteria, archaea, and eukaryotes) of patients suffering CPF- and/ or CDI-related community- and intrahospital-acquired diarrhoea (considering the lack of Latin American studies in this field). Prokaryote and eukaryote composition was described through amplicon-based sequencing of either the 16S-rRNA or 18S-rRNA V4 hypervariable marker regions, respectively. Bacterial communities had differences regarding taxonomic composition and alpha and beta diversity, mainly influenced by diarrhoea acquisition site, whilst such indexes remained constant for eukaryotic communities. Interestingly, significant differences were not observed due to the lack of samples having simultaneous CPF and CDI detection. The greatest imbalance occurred within groups having intrahospital- acquired diarrhoea when stratifying the groups by place of diarrhoea acquisition, suggesting this factor’s influence on microbiota modification. 2. Materials and methods 2.1. Sample selection and group consolidation Ninety-eight DNA samples were randomly selected from the Uni-versidad de Rosario’s (UR) Microbiology and Biotechnology Research Centre (CIMBIUR) biobank, based on the following quantity standards: > 20 ng/µL, 260/280 > 1.8. A Norgen Biotek Corp (Ontario, Canada) kit was used for extracting DNA from faecal samples, following the manu-facturer’s instructions; the samples had been PCR screened for CPF targeting the CPA toxin gene [5] as it has been detected in all CPF toxinotypes [3]. The samples were concurrently screened for CDI, as described elsewhere [14]. Briefly, PCR was used for CDI molecular detection, using primers targeting 16S-rRNA and glutamate dehydrogenase (gdh) genes. A posi-tive PCR result targeting CPF and CDI was denoted as coexistence. Sample groups were established according to diarrhoea acquisition site, following the Society for Healthcare Epidemiology of America (SHEA) and the Infectious Diseases Society of America (IDSA) criteria [15]. This gave four groups: a) community (CO)-associated diarrhoea positive for CPF ‘+ ’ (CO/+, n = 27), b) CO-associated diarrhoea negative for CPF ‘-’ (CO/- n = 23), c) healthcare facility onset (HCFO)- associated diarrhoea positive for CPF (HCFO/+, n = 7) and d) HCFO- associated diarrhoea negative for CPF (HCFO/-, n = 41). Groups were constructed based on bacterial coexistence: + /+ : CDI and CPF coex-istence, + /-: the presence of just CPF, -/+ : the presence of just CDI, -/-: a lack of both types of bacteria. 2.2. Quality assessment and sequencing A NanoDrop/2000/2000c spectrophotometer (Thermo Fisher Scientific, Massachusetts, USA) was used for verifying DNA quality by agarose gel electrophoresis, along with concentration measurement; a 260:280 ratio at 1.8 and 2.0 and 20 ng/µL minimum concentration were confirmed for each sample. The NovaSeq PE-250 platform (Illumina Inc) was used for sequencing all samples that met the quality criteria, using a minimum of 100,000 raw reads per sample. The Novogene Corporation Inc. (Sacra-mento, CA, USA) used 515-F (5′-GTGCCAGCMGCCGCGGTAA-3′) and 806-R primers (5′-GGACTACHVGGGTWTCTAAT-3′) [16] targeting the 16S-rRNA gene’s V4 hypervariable region for bacteria and archaea sequencing; by contrast, 528 F (5′-GCGGTAATTCCAGCTCCAA-3′) and 706 R primers (5′-AATCCRAGAATTTCACCTCT-3′) [17] were used for amplifying and sequencing the 18S-rRNA gene’s V4 hypervariable re-gion for eukaryotes. 2.3. Taxonomic assignment FastQC [18] and MultiQC [19] summary tools were used for assessing sequence quality, considering parameters such as Phred score (minimum Q20) and the presence of adapters. The sequences were then merged, and chimeras eliminated; amplicon sequence variants (ASV) were obtained (defined as sequences varying by at least one nucleotide). The DADA2 pipeline package [20] was used for inferring exact ASVs from high-throughput amplicon sequencing data in R studio [21], using the recommended parameters. The DADA2 formatted SILVA database (version 132) [22] was used for ASV taxonomic assignment of bacteria and Archaea and the Protist Ribosomal Reference database (PR2) [23] for eukaryotes. 2.4. Alpha and beta diversity analysis Phyloseq, Vegan, DESeq2, RCy3, FSA, ggplot2 and reshape2 R packages were used for analysing diversity. Alpha diversity was evalu-ated for determining differences regarding ASV richness and abundance amongst groups by calculating Shannon and inverted Simpson indexes. Beta diversity was analysed by Bray Curtis similarity matrix-based principal coordinates analysis (PCoA) for establishing potential differ-ences concerning sample clustering according to group. 2.5. Statistical analysis Kruskal-Wallis tests with Benjamini-Hochberg correction for multi-ple comparisons and post hoc analysis using Dunn’s test were used for assessing differences amongst groups regarding alpha diversity and genus abundance. A Man-Whitney test was used for comparing CPF infection status. Permutational multivariate analysis of variance (PER-MANOVA) using analysis and partitioning sums of squares using dis-similarities (adonis) was used for evaluating differences in centroids in PCoA plots; < 0.05 p-values were assumed to be significant for the sta-tistical analysis. 3. Results 3.1. C. perfringens infection and microbiota composition Thirty-four of the 98 samples tested (34.7%) were CPF- and CPA- gene positive, 60 (61.2%) were CDI gene-positive and coexistence was seen in 24 (24.5%) samples. No samples from patients having hospital- acquired diarrhoea contained CPF whilst lacking CDI (HCFO +/-) (Supplementary Table 1). Microbiota composition did not vary signifi-cantly amongst groups as similar relative abundance of the phyla Bacillota, Bacteroidota and Pseudomonadota was observed, along with low relative abundance of Actynomicetota, Verrucomicrobia and Fuso-bacteria phyla (Fig. 1A and B). Increased relative abundance of the phylum Bacteroidota was only observed in CPF-positive samples (p = 0.0349) (Fig. 1A). Increased G. Herrera et al. Comparative Immunology, Microbiology and Infectious Diseases 102 (2023) 1020743Fig. 1. Taxonomic composition of bacterial and eukaryotic microbiota by C. perfringens/C. difficile coexistence status and by group. A) Distribution of bacterial phyla involving C. perfringens; B) 16S-rRNA taxonomical composition by group (mixing diarrhoeaacquisition site and C. perfringens and C. difficile detection); C) Bacterial phyla distribution by C. perfringens detection status in groups lacking simultaneous C. perfringens and C. difficile detection; D) Bacterial phyla distribution by group (mixing diarrhoea acquisition site and C. perfringens and C. difficile detection) in groups lacking simultaneous C. perfringens and C. difficile detection; E) Eukaryotic class distribution by C. perfringens detection status; F) Eukaryotic class distribution by group; G) Eukaryotic class distribution by C. perfringens detection status in groups lacking simultaneous C. perfringens and C. difficile detection; H) Eukaryotic class distribution by group re groups lacking simultaneous C. perfringens and C. difficile detection. Statistically significant differences are represented by an asterisk. Fig. 2. Alpha and beta diversity measurement differences: statistically significant differences between groups are represented by a star; A) Alpha diversity indexes by bacterial sequence-based C. perfringens detection status; B) Alpha diversity indexes by bacterial sequence-based group (mixing diarrhoea acquisition site and C. perfringens and C. difficile detection); C) Alpha diversity indexes by eukaryotic sequence-based C. perfringens detection status; D) Alpha diversity indexes by eukaryotic sequence-based group in groups lacking simultaneous C. perfringens and C. difficile detection; E) PCoA by bacterial sequence-based C. perfringens detection status; F) PCoA by bacterial sequence-based group (mixing diarrhoea acquisition site and C. perfringens and C. difficile detection); G) PCoA by eukaryotic sequence- based C. perfringens detection status; H) PCoA by eukaryotic sequence-based group; I) Alpha diversity indexes by C. perfringens detection status-based bacterial sequences in samples lacking simultaneous C. perfringens and C. difficile detection; J) Alpha diversity indexes by group (mixing diarrhoea acquisition site and C. perfringens and C. difficile detection) based on bacterial sequences in samples lacking simultaneous C. perfringens and C. difficile detection; K) PCoA by eukaryotic sequence-based C. perfringens detection status; L) PCoA by eukaryotic sequence-based group. G. Herrera et al. Comparative Immunology, Microbiology and Infectious Diseases 102 (2023) 1020744relative abundance of this phylum was evident in the HCFO + /+ group, although this difference was not observed when stratifying groups by coinfection status (Fig. 1B). No statistically significant differences were observed between phyla when re-analysing samples; this involved excluding groups in which CDI and CPF had been detected (Figs. 1C and 1D). The groups’ eukaryote composition had differential profiles charac-terised by high relative Saccharomycetes abundance, followed by Blas-tocystis-group and Exobasidiomycetes classes (Figs. 1E and 1F). Decreased relative Saccharomycetes class abundance was observed in positive samples (p = 0.002474) when considering CPF presence (Fig. 1E). Statistical analysis revealed a decrease in the CO/+ group versus the HCFO/- group (p = 8.601011e-07) in this class (Fig. 1F); conversely, no changes were observed regarding eukaryotic community abundance amongst groups when CDI-positive samples were eliminated from analysis (Figs. 1G and 1H). 3.2. Differences in alpha and beta diversity Moderate bacterial diversity values were observed regarding CPF and CDI infection samples (Shannon p = 0.006, Simpson p = 0.008); higher bacterial diversity was observed in the CO -/- group compared to the HCFO -/- group (Shannon p = 0.04) (Fig. 2A and B). A lack of dif-ference regarding diversity was observed when groups were analysed without CDI- and CPF-positive samples (Figs. 2C and 2D). Bacterial PCoA highlighted spatial sample clustering by CPA result; the adonis test gave differences regarding centroid positions regarding infection status (PERMANOVA F = 3.2526, p = 0.001) along with great dispersion, especially of HCFO groups (Figs. 2E and 2F). Spatial clus-tering remained when analysing data without CPF- and CDI-positive samples (PERMANOVA F = 1.9555, p = 0.031) (Figs. 2G and 2H). Eukaryote diversity was moderate and did not vary regarding infection status or group (Figs. 2I and 2J); spatial clustering was not observed (Figs. 2K and 2L). 4. Discussion Forero et al., reported up to 33.3% frequency for simultaneous CPF and CDI infection in Colombia [5]; such coexistence’s effect on micro-biota had not been explored previously. Although much research has shown CDI’s direct and indirect effect on gut microbiota, i.e. common pathogens increase whereas beneficial bacteria become depleted [10,11, 24], other factors must be taken into account when analysing microbiota from patients suffering CPF- and/or CDI-related community- and/or intrahospital-acquired diarrhoea. The pathogen’s possible acquisition site must be analysed in depth since its presence alone does not neces-sarily trigger drastic changes in host microbiota [11]. Bacteroidota’s increased relative abundance observed in CPF-posi-tive samples (Fig. 1A) conflicted with previous research reporting a decrease in specific genera belonging to this phylum in Bacteroidota- positive samples [7,12]. However, such increase was largely due to the hospital-acquired diarrhoea group, along with simultaneous CDI and CPF (HCFO +/+) (Fig. 1B), where a considerable increase in Bacteroides’ relative abundance was observed (despite few samples: n = 7) (data not shown). The beneficial role played by many members of this genus [25,26] could suggest that patients in this group preserve a balance within their microbiota (even when two bacteria are associated with diarrheal symptoms). Such balance is promoted by their carbohydrate use ma-chinery which could influence butyrate production in other microor-ganisms [27] (butyrate being a significant metabolite for gut microbiota homeostasis maintenance). This poses a challenge for studying microbiota since the probable pathogen acquisition site adds to the list of factors that could influence results. This became evident when eliminating CO + /+ and HCFO + /+ groups from analysis, as statistically significant differences could no longer be observed (Figs. 1C and 1D). It must be mentioned that information regarding patients’ health status and sociodemographic characteristics was lacking as this could have directly affected the results [11,28,29]. CDI and CPF interaction is another variable contributing to modifi-cations within microbiota due to damage to the epithelium from toxins and/or other virulence factors [3,10]. Both bacterial species’ potential for producing toxins was not evaluated as it is difficult to determine toxigenic profiles from DNA extracted directly from samples due to limitations regarding molecular test sensitivity (i.e. few copies of these genes during initial infection phases) or the potential presence of more than one toxinotype, thereby giving unreliable results. This highlights the need for reliable in vitro culture procedures for obtaining isolates from these species for genomic characterisation for determining viru-lence factors’ true effect within gut microbiota’s complex relationships. The infection status and group differences observed for the class Saccharomycetes (Fig. 1E - 1F) contrasted with previous reports con-cerning inflammatory disease (i.e. irritable bowel syndrome (IBD); an increase in such fungi by as-yet-unknown mechanisms was shown [30]. Such differences were not observed when eliminating CDI- and CPF-positive samples, suggesting a complex interaction between bacte-ria and eukaryotes meriting further research. Moderate alphadiversity values (mainly associated with positive samples - Fig. 2A), suggested complex relationships between microbiota members and CPF and CDI, despite the diarrhoea status of all patients included in this study. Eliminating samples having both CPF and CDI caused the lack of difference amongst groups, thereby highlighting the fact that microbiota modifications did not result from a single group, but rather the product of many components’ interaction [10,24,28]. The groups of patients having diarrhoea associated with an in- hospital setting could have influenced the differences, since stratifying the results showed that the HCFO -/-, HCFO -/+ and HCFO + /+ groups had the greatest data dispersion diversity (i.e. a downward trend - Fig. 2B). It was evident that decreased diversity could have resulted from the bacillus’ interactions with other intestinal ecosystem members in CPF-negative groups having CDI; this could have been promoted by factors such as nutrient competition, toxins and/or a decrease in short- chain fatty acids (SCFA) [31]. This has been the first report regarding bacterial and eukaryotic microbiota associated with CPF and CDI coexistence. Composition and diversity index variations suggested that such microorganisms’ coexis-tence and pathogen acquisition sites could have affected gut microbiota, and that CDI could have further affected it by modulating ecosystem balance and contributing to an increase in its diversity. Our study has limitations concerning a lack of clinical and de-mographic factors associated with the target patients from whom the samples were taken and a lack of information concerning CPF–positive samples’ toxinotyping which could have affected the results. Further research is required for evaluating CPF’s effect and that of the toxinotypes and their load regarding gut microbial ecology. Col-lecting and evaluating clinical and sociodemographic data is also rec-ommended for determining the effect of other factors on CPF and CDI role regarding gut microbiota. This study has highlighted the impor-tance of analysing gut microbiota in a broad context for determining the effect of the relationships between microorganisms and intestinal ho-meostasis, rather than concentrating on their influence on a single element concerning this ecosystem. (Table 1). Funding source This research was funded by the Colombian Ministry of Science, Technology, and Innovation (Minciencias) within the framework of a project entitled, “Determining the intestinal microbiome in patients suffering intensive care unit- and community-acquired Clostridioides difficile infection-associated diarrhoea,” code 212477758147, contract number 606–2018, call for research projects 777/2017. We would like G. Herrera et al. Comparative Immunology, Microbiology and Infectious Diseases 102 (2023) 1020745to thank the Universidad del Rosario’s Academic Affairs Office and Natural Sciences Faculty for granting Giovanny Herrera a graduate as-sistant scholarship. Ethical approval statement The study was considered low risk according to Colombian Ministry of Health resolution 8430/1993. Samples were coded according to Colombian ethical guidelines and the Declaration of Helsinki to avoid patient identification. This project was approved by the Universidad del Rosario’s (UR) Research Ethics Committee (approval 339). Written informed consent was obtained for using the samples in this research, as authorised by the UR ethics committee. CRediT authorship contribution statement GH, JDR, and MM: conceptualisation and methodology; GH and AC: investigation and data curation; GH and MM: software, validation, and formal analysis; MAP, JDR and MM: funding acquisition, review and editing the final manuscript. All authors have read and approved the final manuscript. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements We would like to thank the Universidad del Rosario’s advanced computing laboratory for data analysis (CENTAURO - laboratory for research and teaching needs requiring high computing capacity and/or processing large volumes of information). We would like to thank Jason Garry for translating the manuscript. Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.cimid.2023.102074. References [1] S.-H. Han, J. Yi, J.-H. Kim, S. Lee, H.-W. Moon, Composition of gut microbiota in patients with toxigenic Clostridioides (Clostridium) difficile: Comparison between subgroups according to clinical criteria and toxin gene load, PLOS ONE vol. 14 (2) (2019), e0212626, https://doi.org/10.1371/journal.pone.0212626. [2] R. Kiu, L.J. Hall, An update on the human and animal enteric pathogen Clostridium perfringens, Emerg. Microbes Infect. vol. 7 (1) (2018) 1–15. [3] M.A. Navarro, B.A. McClane, F.A. Uzal, Mechanisms of action and cell death associated with clostridium perfringens toxins, Toxins (Basel) vol. 10 (5) (2018) 212, https://doi.org/10.3390/toxins10050212. [4] D. Dewey-Mattia, K. Manikonda, A.J. Hall, M.E. Wise, S.J. Crowe, Surveillance for foodborne disease outbreaks — United States, 2009–2015, MMWR Surveill. Summ. vol. 67 (10) (2018) 1–11, https://doi.org/10.15585/mmwr.ss6710a1. [5] A.J. Forero, et al., High frequency of toxigenic Clostridium difficile and Clostridium perfringens coinfection among diarrheic patients at health care facility-onset (HCFO) and community-onset (CO) centers in Bogotá, Colombia, Gut Pathog. vol. 11 (2019) 27, https://doi.org/10.1186/s13099-019-0308-7. [6] I. Mehdizadeh Gohari, M.A. Navarro, J. Li, A. Shrestha, F. Uzal, B.A. McClane, Pathogenicity and virulence of Clostridium perfringens, Virulence vol. 12 (1) (2021) 723–753. [7] B. Lakshminarayanan, et al., Prevalence and characterization of Clostridium perfringens from the faecal microbiota of elderly Irish subjects, J. Med. Microbiol. vol. 62 (3) (2013) 457–466. [8] Z. Li, W. Wang, D. Liu, Y. Guo, Effects of Lactobacillus acidophilus on gut microbiota composition in broilers challenged with Clostridium perfringens, PLoS One vol. 12 (11) (2017), e0188634. [9] S.M. Finegold, P.H. Summanen, J. Downes, K. Corbett, T. Komoriya, Detection of Clostridium perfringens toxin genes in the gut microbiota of autistic children, Anaerobe vol. 45 (2017) 133–137, https://doi.org/10.1016/j. anaerobe.2017.02.008. [10] A. Abbas, J.P. Zackular, Microbe–microbe interactions during Clostridioides difficile infection, Curr. Opin. Microbiol. vol. 53 (2020) 19–25. [11] G. Herrera, D. Paredes-Sabja, M.A. Patarroyo, J.D. Ramírez, M. Muñoz, Updating changes in human gut microbial communities associated with Clostridioides difficile infection, Gut Microbes vol. 13 (1) (2021), 1966277, https://doi.org/ 10.1080/19490976.2021.1966277. [12] R. Nagpal, et al., Gut dysbiosis following C-section instigates higher colonisation of toxigenic Clostridium perfringens in infants, Benef. Microbes vol. 8 (3) (2017) 353–365. [13] S. Horvat, A. Mahnic, M. Breskvar, S. Dzeroski, M. Rupnik, Evaluating the effect of Clostridium difficile conditioned medium on fecal microbiota community structure, Sci. Rep. vol. 7 (1) (2017), https://doi.org/10.1038/s41598-017-15434- 1. [14] M. Muñoz, et al., New Insights into Clostridium difficile (CD) infection in latin america: novel description of toxigenic profiles of diarrhea-associated to CD in Bogotá, Colombia, Front. Microbiol. vol. 9 (2018) 74, https://doi.org/10.3389/ fmicb.2018.00074. [15] S.H. Cohen, et al., Clinical practice guidelinesfor Clostridium difficile infection in adults: 2010 update by the society for healthcare epidemiology of America (SHEA) and the infectious diseases society of America (IDSA), Infect. Control Hosp. Epidemiol. vol. 31 (5) (2010) 431–455, https://doi.org/10.1086/651706. [16] J.G. Caporaso, et al., Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample, Proc. Natl. Acad. Sci. vol. 108 (Supplement 1) (2011) 4516–4522. [17] M.K. Cheung, C.H. Au, K.H. Chu, H.S. Kwan, C.K. Wong, Composition and genetic diversity of picoeukaryotes in subtropical coastal waters as revealed by 454 pyrosequencing, ISME J. vol. 4 (8) (2010) 1053–1059. [18] S. Andrews, FastQC: a quality control tool for high throughput sequence data. Babraham Bioinformatics, Babraham Institute, Cambridge, United Kingdom, 2010. [19] P. Ewels, M. Magnusson, S. Lundin, M. Käller, MultiQC: summarize analysis results for multiple tools and samples in a single report, Bioinformatics vol. 32 (19) (2016) 3047–3048, https://doi.org/10.1093/bioinformatics/btw354. [20] B.J. Callahan, P.J. McMurdie, M.J. Rosen, A.W. Han, A.J.A. Johnson, S.P. Holmes, DADA2: high resolution sample inference from Illumina amplicon data, Nat. Methods vol. 13 (7) (2016) 581–583, https://doi.org/10.1038/nmeth.3869. [21] R Core Team, R: A language and environment for statistical computing. Vienna, Austria, 2013. [22] C. Quast, et al., The SILVA ribosomal RNA gene database project: improved data processing and web-based tools, Nucleic Acids Res vol. 41 (2013) D590–D596, https://doi.org/10.1093/nar/gks1219. [23] L. Guillou, et al., The Protist Ribosomal Reference database (PR2): a catalog of unicellular eukaryote Small Sub-Unit rRNA sequences with curated taxonomy, Nucleic Acids Res vol. 41 (2013) D597–D604, https://doi.org/10.1093/nar/ gks1160. [24] G. Herrera, L. Vega, M.A. Patarroyo, J.D. Ramírez, M. Muñoz, Gut microbiota composition in health-care facility-and community-onset diarrheic patients with Clostridioides difficile infection, Sci. Rep. vol. 11 (1) (2021), https://doi.org/ 10.1038/s41598-021-90380-7. [25] H.M. Wexler, Bacteroides: the good, the bad, and the nitty-gritty, Clin. Microbiol Rev. vol. 20 (4) (2007) 593–621, https://doi.org/10.1128/CMR.00008-07. [26] H. Zafar, M.H. Saier Jr, Gut Bacteroides species in health and disease, Gut Microbes vol. 13 (1) (2021), 1848158. [27] G.P. Rodriguez-Castaño, M.R. Dorris, X. Liu, B.W. Bolling, A. Acosta-Gonzalez, F. E. Rey, Bacteroides thetaiotaomicron starch utilization promotes quercetin degradation and butyrate production by Eubacterium ramulus, Front. Microbiol. vol. 10 (2019) 1145. [28] M.C. Abt, P.T. McKenney, E.G. Pamer, Clostridium difficile colitis: pathogenesis and host defence, Nat. Rev. Microbiol vol. 14 (10) (2016) 609–620, https://doi. org/10.1038/nrmicro.2016.108. Table 1 C. perfringens and C. difficile found in the samples. Group CPA positive CPA negative Total CO 27 23 50 HCFO 7 41 48 Total 34 64 98 Coinfection Status CPA positive CPA negative Total Cdiff positive 24 36 60 Cdiff Negative 10 28 38 Total 34 64 98 Coinfection by group Status CPA positive CPA negative Total Cdiff/+ CO 17 13 30 HCFO 7 23 30 Cdiff/- CO 10 10 20 HCFO 0 18 18 Total 34 64 98 G. Herrera et al. https://doi.org/10.1016/j.cimid.2023.102074https://doi.org/10.1371/journal.pone.0212626http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref2http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref2https://doi.org/10.3390/toxins10050212https://doi.org/10.15585/mmwr.ss6710a1https://doi.org/10.1186/s13099-019-0308-7http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref6http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref6http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref6http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref7http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref7http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref7http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref8http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref8http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref8https://doi.org/10.1016/j.anaerobe.2017.02.008https://doi.org/10.1016/j.anaerobe.2017.02.008http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref10http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref10https://doi.org/10.1080/19490976.2021.1966277https://doi.org/10.1080/19490976.2021.1966277http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref12http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref12http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref12https://doi.org/10.1038/s41598-017-15434-1https://doi.org/10.1038/s41598-017-15434-1https://doi.org/10.3389/fmicb.2018.00074https://doi.org/10.3389/fmicb.2018.00074https://doi.org/10.1086/651706http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref16http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref16http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref16http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref17http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref17http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref17https://doi.org/10.1093/bioinformatics/btw354https://doi.org/10.1038/nmeth.3869https://doi.org/10.1093/nar/gks1219https://doi.org/10.1093/nar/gks1160https://doi.org/10.1093/nar/gks1160https://doi.org/10.1038/s41598-021-90380-7https://doi.org/10.1038/s41598-021-90380-7https://doi.org/10.1128/CMR.00008-07http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref24http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref24http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref25http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref25http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref25http://refhub.elsevier.com/S0147-9571(23)00132-7/sbref25https://doi.org/10.1038/nrmicro.2016.108https://doi.org/10.1038/nrmicro.2016.108Comparative Immunology, Microbiology and Infectious Diseases 102 (2023) 1020746[29] W. Sangster, et al., Bacterial and fungal microbiota changes distinguish C. difficile Infection from other forms of diarrhea: results of a prospective inpatient study, Front Microbiol vol. 7 (2016), https://doi.org/10.3389/fmicb.2016.00789. [30] C.E. Huseyin, P.W. O’Toole, P.D. Cotter, P.D. Scanlan, Forgotten fungi—the gut mycobiome in human health and disease, FEMS Microbiol. Rev. vol. 41 (4) (2017) 479–511, https://doi.org/10.1093/femsre/fuw047. [31] C. Martin-Gallausiaux, L. Marinelli, H.M. Blottière, P. Larraufie, N. Lapaque, SCFA: mechanisms and functional importance in the gut, Proc. Nutr. Soc. vol. 80 (1) (2021) 37–49, https://doi.org/10.1017/S0029665120006916. G. Herrera et al. https://doi.org/10.3389/fmicb.2016.00789https://doi.org/10.1093/femsre/fuw047https://doi.org/10.1017/S0029665120006916 139 9. CONCLUSIONES CAPÍTULO I: ¨ El primer panorama epidemiológico sobre la frecuencia de detección de C. perfringens en varias especies animales en Colombia revela una elevada presencia de esta bacteria en animales domésticos y cerdos. ¨ La relación de la toxina CPB2 de C. perfringens con la diabetes, sugiere interacciones con el sistema inmunitario del huésped. Factores de riesgo como la inmunidad comprometida, la hiperglucemia, resaltan la importancia de comprender esta dinámica en pacientes diabéticos. ¨ Comprender la epidemiología local de C. perfringens mediante estudios microgeográficos es crucial para anticiparse a las amenazas para la salud pública y hacerles frente con eficacia. CAPÍTULO II: ¨ La representación de genomaspúblicos de C. perfringens en países en desarrollo es escasa, probablemente debido a la falta de vigilancia epidemiológica y a la limitada recopilación de datos genómicos. ¨ Los análisis basados en MLST y el genoma central, utilizando genomas públicos de C. perfringens, revelan una estrecha relación entre aislamientos de diferentes hospederos. Estos hallazgos respaldan la marcada asociación de C. perfringens con las ETA y sugieren su potencial zoonótico. ¨ La identificación de toxinotipos de C. perfringens poco comunes en humanos, como el tipo G asociado con enfermedades en aves, junto con la presencia de aislamientos de tipo D y E vinculados con enfermedades en rumiantes, refleja la adaptabilidad de C. perfringens a diferentes hospedadores y resalta su papel zoonótico. ¨ Se requieren llevar a cabo estudios de caracterización genómica de microorganismos relevantes para la salud pública, particularmente en países en desarrollo, con el fin de comprender la diversidad genética, los patrones de dispersión y los marcadores de virulencia circulantes, que permitan instaurar medidas sanitarias y tratamientos adaptados a la realidad local. 140 CAPÍTULO III: ¨ El análisis microgeográfico de la estructura poblacional a partir de genomas colombianos de C. perfringens, reveló una amplia diversidad genética, sugiriendo posibles eventos de dispersión entre caninos, felinos y humanos. Estos hallazgos resaltan la necesidad de incrementar prácticas de higiene eficaces, especialmente en los cuidadores de animales domésticos con el objetivo de prevenir la transmisión a nivel comunitario. ¨ La presencia de alveolisina en aislamientos de C. perfringens de gatos asintomáticos en la región colombiana, junto con su alta capacidad de inhibición del crecimiento celular in vitro, sugiere un potencial significativo de virulencia en animales domésticos. Se requieren investigaciones adicionales para comprender mejor el papel de la alveolisina en la patogénesis de C. perfringens. ¨ Los aislamientos de C. perfringens portadores de la toxina PFOA, relacionada con la enterocolitis necrotizante, están presentes en individuos asintomáticos. Las pruebas fenotípicas, revelan su capacidad para generar hemólisis completa, inhibición del crecimiento celular in vitro y una mayor capacidad de esporulación. Esto sugiere un potencial riesgo de transmisión e infección a nivel comunitario y destaca la necesidad de comprender mejor el impacto de las toxinas accesorias a través de la caracterización fenotípica. ¨ Los hallazgos de MRA a aminoglucósidos, macrólidos y tetraciclinas a nivel genómico, junto con la reducida susceptibilidad fenotípica de aislamientos de C. perfringens a gentamicina, eritromicina, tetraciclina y metronidazol, representan una amenaza para el tratamiento adecuado de las infecciones causadas por esta bacteria. Esta capacidad para desarrollar un fenotipo resistente a múltiples fármacos de uso convencional supone un riesgo en el abordaje clínico y subraya la necesidad de realizar estudios microgeográficos para adaptar medidas terapéuticas eficientes a nivel local. 10. PERSPECTIVAS ü La mayor detección de C. perfringens en animales domésticos en comparación con los rumiantes resalta la importancia de factores como la dieta, la coinfección por otros patógenos y el papel de la microbiota intestinal. Por ende, los estudios futuros deberán incorporar datos acerca de hábitos alimenticios, factores demográficos y perfiles de dichas comundiades microbianas para profundizar en el entendimiento de estas complejas interacciones. ü La inclusión de un mayor número de participantes, tanto individuos de la comunidad asintomáticos como pacientes hospitalizados con síntomas gastrointestinales, de diferentes regiones de Colombia, es esencial para obtener una visión más completa y representativa de epidemiología de C. perfringens. 141 ü La ampliación de investigaciones futuras incluyendo una variedad más amplia y diversa de muestras de diferentes nichos ecológicos como humanos, animales, muestras de suelo, agua y alimentos permitirá avanzar en la comprensión del potencial de transmisión de C. perfringens. Estos estudios, aportarán a mejorar la precisión en la detección y entendimiento de este microorganismo y descubrir métodos más eficientes y estratégicos para mitigar su creciente impacto en la salud. ü La asociación entre la presencia de aislamientos de C. perfringens portadores del gen de la toxina CPB2 y diabetes mellitus se convierte en un interesante campo de investigación. Profundizar en la frecuencia de detección de C. perfringens entre la población diabética y examinar muestras de tejido necrótico de úlceras diabéticas podría transformar el diagnóstico y tratamiento de estas afecciones, marcando un avance significativo en el manejo clínico de los pacientes afectados. ü El análisis de los datos de secuenciación de genomas completos de C. perfringens deberá incluir el análisis funcional de genes centrales y accesorios, para comprender mejor la plasticidad genómica de esta bacteria. Este proceso permitirá identificar genes relacionados con aspectos cruciales como el metabolismo, los mecanismos de defensa, la recombinación y los procesos de reparación, lo que a su vez enriquecerá el entendimiento sobre su elevada plasticidad genómica. ü Investigar el rol patológico de los factores de virulencia de C. perfringens, posiblemente a través de modelos murinos o modelos de colon humano, puede ofrecer información valiosa sobre cómo este patógeno afecta al organismo y cómo podría ser contrarrestado de manera más efectiva. ü Evaluar los niveles de expresión mediante ensayos de PCR cuantitativa en tiempo real de toxinas como CPA, CPE, CPB2 y PFOA y su asociación con el daño sobre células Caco2, podría aportar un mayor conocimiento acerca de la patogénesis de aislamientos toxigénicos de C. perfringens ü Las condiciones de esporulación afectan significativamente el crecimiento, la germinación y la resistencia de las esporas de C. perfringens. Por lo tanto, se deberá investigar los efectos de la temperatura, el pH y la actividad de desinfectantes sobre el crecimiento y la germinación de las esporas de C. perfringens. ü Es fundamental llevar a cabo investigaciones genómicas enfocadas en la población pediátrica, prestando especial atención a los recién nacidos prematuros con bajo peso al nacer que padecen de enterocolitis necrotizante por C. perfringens en las Unidades de Cuidados Intensivos. El objetivo principal podría estar enfocado a caracterizar los aislamientos microbianos implicados en la patogénesis de esta condición, con la intención de contribuir significativamente al desarrollo de tratamientos más efectivos. 142 11. PRODUCTOS DE LA TESIS Los productos generados durante el desarrollo de la presente tesis doctoral se listan a continuación: LISTA DE PUBLICACIONES: 1. Artículo 1: Camargo A, Páez-Triana L, Camargo D, García-Corredor D, Pulido-Medellín M, Camargo M, Ramírez J.D. and Muñoz M*. Carriage of Clostridium perfringens in Domestic and Farm Animals across the Central Highlands of Colombia: Implications for Gut Health and Zoonotic Transmission. Vet Res Commun. 2024. 2. Artículo 2: Camargo A., Bohórquez L., López D., Ferrebuz-Cardozo A., Castellanos-Rozo J., Díaz J., Rada M., Camargo M., Ramírez J. D. and Muñoz M. Clostridium perfringens in central Colombia: Frequency, Toxin Genes, and Risk Factors. (SOMETIDO en Gut Pathogens). 3. Artículo 3: Camargo A., Guerrero-Araya E, Castañeda S, Vega L, Cardenas-Alvarez MX, Rodríguez C, Paredes-Sabja D, Ramírez JD, Muñoz M. Intra-species diversity of Clostridium perfringens: A diverse genetic repertoire reveals its pathogenic potential. Front Microbiol. 2022 Jul 22; 13:952081. doi: 10.3389/fmicb.2022.952081. PMID: 35935202; PMCID: PMC9354469. 4. Artículo 4: Camargo A., Bohorquez L.,Cáceres T., Ferrebuz-Cardozo A, Díaz J, Castellanos-Rozo J, Diaz J., Kiu R., Hall L. J., Rámirez J. D. and Muñoz M. Insights into Clostridium perfringens Dispersal Hotspots, Toxins, and Virulence Factors through Integrated Genomic and Phenotypic Profiling. (EN CONSTRUCCIÓN) 5. Artículo 5: Camargo A, Rámirez J. D., Kiu R., Hall L.J., Muñoz M. Unveiling the pathogenic mechanisms of Clostridium perfringens toxins and virulence factors. Emerg Microbes Infect. 2024 Apr 9:2341968. doi: 10.1080/22221751.2024.2341968. Epub ahead of print. PMID: 38590276. 6. Artículo 6: Herrera G, Vega L, Camargo A, Patarroyo MA, Ramírez JD, Muñoz M. Acquisition site-based remodelling of Clostridium perfringens- and Clostridioides difficile-related gut microbiota. Comp Immunol Microbiol Infect Dis. 2023 Nov; 102:102074. doi: 10.1016/j.cimid.2023.102074. Epub 2023 Oct 10. PMID: 37832162. 143 PRESENTACIÓN EN EVENTOS CIENTÍFICOS: Julio/2022 Seattle, EE. UU. THE 16TH BIENNIAL CONGRESS OF THE ANAEROBE SOCIETY OF THE AMERICAS Tipo de evento: Congreso Ámbito: Internacional Tipo de producto: Póster Nombre del producto: Intra-species diversity of Clostridium perfringens: A diverse genetic repertoire reveals its pathogenic potential Diciembre/2022 Bogotá, Colombia XVIII CONGRESO COLOMBIANO DE PARASITOLOGÍA Y MEDICINA TROPICAL Tipo de evento: Congreso Ámbito: Nacional Tipo de producto: Ponencia oral Nombre del producto: Detección de Clostridium perfringens en animales domésticos y de granja en el altiplano central de Colombia: Implicaciones para la salud intestinal y la transmisión zoonótica PASANTÍA INTERNACIONAL Norwich, Inglaterra Enero – Junio de 2023 Tipo de producto: Pasantía doctoral Ámbito: Internacional Lugar: Early life microbiota-host interactions laboratory, Quadram Institute Biosciences. Líder: Lindsay Hall CURSOS Noviembre 2020 Curso virtual Curso Introducción a la filogenómica de bacterias 9 horas BioSciences App Noviembre 2020 Curso virtual Curso Metagenómica 9 horas BioSciences App Septiembre 2021 Curso virtual Curso Ciencia de datos en R 16 horas Biofreelancer 144 Marzo 2024 Curso presencial Bogotá, Colombia. Curso Clinical Research During Outbreaks – CREDO 24 horas Universidad de Oxford – ISARIC- UCTS USabana BECAS Enero 2023 Bogotá/Colombia Beca para el fortalecimiento académico - Pasantía Doctoral Estudiantes doctorales Universidad del Rosario Tunja/Colombia Beca para cursar estudios doctorales Facultad de Ciencias de la Salud Universidad de Boyacá ORIENTACIÓN DE TRABAJOS DE GRADO Tesis de pregrado 2023 Universidad del Rosario Tipo de producto: Tesis de pregrado en Biología. Estudiante: Emanuella de la Cruz Facultad: Facultad de Ciencias Naturales Nombre de la tesis: Caracterización de Clostridium paraputrificum y Clostridium tertium en muestras de heces de animales de Boyacá y Cundinamarca - Colombia Estado: Aprobado y finalizado Tesis de pregrado 2023 Universidad de Boyacá Tipo de producto: Tesis de pregrado en Bacteriología y Laboratorio Clínico. Estudiante: María Alejandra Díaz Facultad: Facultad de Ciencias de la Salud Nombre de la tesis: Calidad microbiológica de productos cárnicos crudos frescos en el municipio de Gámeza, departamento de Boyacá, Colombia Estado: Aprobado y finalizado Trabajo de grado 2023 Universidad Pedagógica y Tecnológica de Colombia - UPTC Tipo de producto: Trabajo de grado en Medicina Veterinaria y Zootecnia. Estudiante: Valentina Rodríguez Montaña Facultad: Nombre del trabajo: Detección de C. perfringens circulante en animales domésticos del Departamento de Boyacá Estado: Aprobado y finalizado Tesis de pregrado 2024 Universidad de Boyacá Tipo de producto: Tesis de pregrado en Bacteriología y Laboratorio Clínico. Estudiante: Ingrid Dayana Cano Tipazoca Facultad: Facultad de Ciencias de la Salud 145 Nombre de la tesis: Características fenotípicas de resistencia antimicrobiana de Clostridium perfringens obtenidas de muestras de heces Estado: En curso Tesis de Maestría 2024 Universidad Colegio Mayor de Cundinamarca Tipo de producto: Tesis de Maestría en Microbiología Estudiante: Dayana Sofía Torres Facultad: Facultad de Ciencias de la Salud, Universidad Colegio Mayor de Cundinamarca Nombre de la tesis: Resistencia antimicrobiana de Clostridium perfringens en humanos y animales del Departamento de Boyacá Estado: En curso 146 9. BIBLIOGRAFÍA 1. Melville, S. and L. Craig, Type IV pili in Gram-positive bacteria. Microbiol Mol Biol Rev, 2013. 77(3): p. 323-41. 2. Li, J., et al., Clostridium perfringens Sporulation and Sporulation-Associated Toxin Production. Microbiol Spectr, 2016. 4(3). 3. Welsh, W. and G.J.J.H.H.B. Nuttall, A Gas Producing Bacillus (Bacillus aerogenes capsulatus, nov, spec.) Capable of Rapid Development in the Blood Vessels after Death. 1892. 3: p. 81-91. 4. Silva, R.O. and F.C. Lobato, Clostridium perfringens: A review of enteric diseases in dogs, cats and wild animals. Anaerobe, 2015. 33: p. 14-7. 5. Asha, N.J., D. Tompkins, and M.H. Wilcox, Comparative analysis of prevalence, risk factors, and molecular epidemiology of antibiotic-associated diarrhea due to Clostridium difficile, Clostridium perfringens, and Staphylococcus aureus. J Clin Microbiol, 2006. 44(8): p. 2785-91. 6. Kiu, R. and L.J. Hall, An update on the human and animal enteric pathogen Clostridium perfringens. Emerg Microbes Infect, 2018. 7(1): p. 141. 7. Gould, L.H., et al., Surveillance for foodborne disease outbreaks - United States, 1998-2008. MMWR Surveill Summ, 2013. 62(2): p. 1-34. 8. Gould, L.H., et al., Surveillance for foodborne disease outbreaks—United States, 2009–2010. 2013. 62(3): p. 41. 9. Hall, B.G., Predicting the evolution of antibiotic resistance genes. Nat Rev Microbiol, 2004. 2(5): p. 430-5. 10. Bhattacharya, A., et al., Outbreak of Clostridium perfringens food poisoning linked to leeks in cheese sauce: an unusual source. Epidemiol Infect, 2020. 148: p. e43. 11. Authority, E.F.S.J.E.j., The European Union summary report on trends and sources of zoonoses, zoonotic agents and food-borne outbreaks in 2016. 2017. 15(12). 12. Kiu, R., et al., Probing genomic aspects of the multi-host pathogen Clostridium perfringens reveals significant pangenome diversity, and a diverse array of virulence factors. 2017. 8: p. 2485. 13. Kiu, R., L.J.J.E.m. Hall, and infections, An update on the human and animal enteric pathogen Clostridium perfringens. 2018. 7(1): p. 1-15. 14. Möllby, R., et al., Production of phospholipase C (alpha-toxin), haemolysins and lethal toxins by Clostridium perfringens types A to D. 1976. 96(1): p. 137-144. 15. Hassan, K.A., et al., Genomic analyses of Clostridium perfringens isolates from five toxinotypes. 2015. 166(4): p. 255-263. 16. Andrews, S.C., A.K. Robinson, and F.J.F.m.r. Rodríguez-Quiñones, Bacterial iron homeostasis. 2003. 27(2-3): p. 215-237. 17. Myers, G.S., et al., Skewed genomic variability in strains of the toxigenic bacterial pathogen, Clostridium perfringens. 2006. 16(8): p. 1031-1040. 18. Rood, J.I., et al., Expansion of the Clostridium perfringens toxin-based typing scheme. 2018. 53: p. 5-10. 147 19. Oda, M., et al., Membrane-binding mechanism of Clostridium perfringens alpha-toxin. 2015. 7(12): p. 5268-5275. 20. Awad, M.M., et al., Virulence studies on chromosomal alpha-toxin and theta-toxin mutants constructed by allelic exchange provide genetic evidence for the essential role of alpha-toxin in Clostridium perfringens-mediated gas gangrene. Mol Microbiol, 1995. 15(2): p. 191-202. 21. Titball, R.W., C.E. Naylor, and A.K.J.A. Basak, The Clostridium perfringensα-toxin. 1999. 5(2): p. 51-64. 22. Takehara, M., et al., Clostridium perfringensα-toxin impairs innate immunity via inhibition of neutrophil differentiation. 2016. 6(1): p. 1-12. 23. Uzal, F.A., et al., Towards an understanding of the role of Clostridium perfringens toxins in human and animal disease. 2014. 9(3): p. 361-377. 24. Garcia, J.P., et al., The pathology of enterotoxemia by Clostridium perfringens type C in calves. 2013. 25(3): p. 438-442. 25. Uzal, F., et al., Clostridium perfringens toxins involved in mammalian veterinary diseases. 2010. 2: p. 24. 26. Chassin, C., et al., Pore-forming epsilon toxin causes membrane permeabilization and rapid ATP depletion-mediated cell death in renal collecting duct cells. 2007. 293(3): p. F927-F937. 27. Takagishi, T., et al., Oligomer formation of Clostridium perfringens epsilon-toxin is induced by activation of neutral sphingomyelinase. 2016. 1858(11): p. 2681-2688. 28. Petit, L., et al., Clostridium perfringens epsilon toxin induces a rapid change of cell membrane permeability to ions and forms channels in artificial lipid bilayers. 2001. 276(19): p. 15736-15740. 29. Hilger, H., et al., The long-lived nature of Clostridium perfringens iota toxin in mammalian cells induces delayed apoptosis. 2009. 77(12): p. 5593-5601. 30. Songer, J.G.J.C.D.o.A. and F. Uzal, Songer, JG, Prescott, JF, Popoff, MR, Eds, Infections by Clostridium perfringens type E. 2016: p. 173-176. 31. Richard, J.F., et al., Transcytosis of iota-toxin across polarized CaCo-2 cells. 2002. 43(4): p. 907-917. 32. Tsuge, H., et al., Crystal structure and site-directed mutagenesis of enzymatic components from Clostridium perfringens iota-toxin. 2003. 325(3): p. 471-483. 33. Freedman, J.C., A. Shrestha, and B.A.J.T. McClane, Clostridium perfringens enterotoxin: action, genetics, and translational applications. 2016. 8(3): p. 73. 34. Carman, R.J.J.R.i.M.M., Clostridium perfringens in spontaneous and antibiotic-associated diarrhoea of man and other animals. 1997. 8: p. S46. 35. Chakrabarti, G. and B.A.J.C.m. McClane, The importance of calcium influx, calpain and calmodulin for the activation of CaCo-2 cell death pathways by Clostridium perfringens enterotoxin. 2005. 7(1): p. 129-146. 36. Bannam, T.L., et al., Necrotic enteritis-derived Clostridium perfringens strain with three closely related independently conjugative toxin and antibiotic resistance plasmids. 2011. 2(5). 148 37. Rood, J.I., A.L. Keyburn, and R.J.J.A.P. Moore, NetB and necrotic enteritis: the hole movable story. 2016. 45(3): p. 295-301. 38. Savva, C.G., et al., Molecular architecture and functional analysis of NetB, a pore-forming toxin from Clostridium perfringens. 2013. 288(5): p. 3512-3522. 39. Deguchi, A., et al., Genetic characterization of type A enterotoxigenic Clostridium perfringens strains. 2009. 4(5): p. e5598. 40. Kiu, R., et al., Particular genomic and virulence traits associated with preterm infant-derived toxigenic Clostridium perfringens strains. Nature Microbiology, 2023. 8(6): p. 1160-1175. 41. Awad, M.M., et al., Synergistic effects of alpha-toxin and perfringolysin O in Clostridium perfringens-mediated gas gangrene. 2001. 69(12): p. 7904-7910. 42. Ellemor, D.M., et al., Use of genetically manipulated strains of Clostridium perfringens reveals that both alpha-toxin and theta-toxin are required for vascular leukostasis to occur in experimental gas gangrene. 1999. 67(9): p. 4902-4907. 43. Bryant, A.E., et al., Clostridial Gas Gangrene. II. Phospholipase C—Induced Activation of Platelet gpIIbIIIa Mediates Vascular Occlusion and Myonecrosis in Clostvidium perfringens Gas Gangrene. 2000. 182(3): p. 808-815. 44. Bryant, A., et al., Clostridium perfringens phospholipase C-induced platelet/leukocyte interactions impede neutrophil diapedesis. 2006. 55(5): p. 495-504. 45. Geier, R.R., T.G. Rehberger, and A.H. Smith, Comparative Genomics of Clostridium perfringens Reveals Patterns of Host-Associated Phylogenetic Clades and Virulence Factors. Front Microbiol, 2021. 12: p. 649953. 46. Billington, S.J., B.H. Jost, and J.G. Songer, Thiol-activated cytolysins: structure, function and role in pathogenesis. FEMS Microbiology Letters, 2000. 182(2): p. 197-205. 47. Gurjar, A.A., et al., Expression, crystallization and preliminary X-ray diffraction studies of recombinant Clostridium perfringens β2-toxin. 2007. 63(6): p. 484-487. 48. Benz, R., et al., Clostridium perfringens Beta2 toxin forms highly cation-selective channels in lipid bilayers. 2022. 51(1): p. 15-27. 49. Benz, R., et al., Clostridium perfringens Beta2 toxin forms highly cation-selective channels in lipid bilayers. 2021: p. 1-13. 50. Traving, C., R.J.C. Schauer, and M.L.S. CMLS, Structure, function and metabolism of sialic acids. 1998. 54(12): p. 1330-1349. 51. Li, J., et al., Sialidases affect the host cell adherence and epsilon toxin-induced cytotoxicity of Clostridium perfringens type D strain CN3718. 2011. 7(12): p. e1002429. 52. Li, J., B.A.J.A. McClane, and e. microbiology, The sialidases of Clostridium perfringens type D strain CN3718 differ in their properties and sensitivities to inhibitors. 2014. 80(5): p. 1701-1709. 53. Cioffi, D.L., et al., Terminal sialic acids are an important determinant of pulmonary endothelial barrier integrity. 2012. 302(10): p. L1067-L1077. 149 54. Theoret, J.R., et al., Native or proteolytically activated NanI sialidase enhances the binding and cytotoxic activity of Clostridium perfringens enterotoxin and beta toxin. 2018. 86(1): p. e00730-17. 55. Wang, Y.-h.J.F.i.c. and i. microbiology, Sialidases from Clostridium perfringens and their inhibitors. 2020. 9: p. 462. 56. Popham, D., et al., The bacterial spore: from molecules to systems. 2015. 57. Novak, J.S., V.K. Juneja, and B.A. McClane, An ultrastructural comparison of spores from various strains of Clostridium perfringens and correlations with heat resistance parameters. Int J Food Microbiol, 2003. 86(3): p. 239-47. 58. Li, J., et al., Clostridium perfringens sporulation and sporulation-associated toxin production. 2016. 4(3): p. 4.3. 08. 59. Setlow, P.J.T.i.m., I will survive: DNA protection in bacterial spores. 2007. 15(4): p. 172-180. 60. Setlow, P.J.M.s., Spore resistance properties. 2014. 2(5): p. 2.5. 11. 61. McClane, B., S. Robertson, and J.J.C.p. Li, Food microbiology: fundamentals and frontiers. 2013: p. 465-489. 62. O'Brien, D.K. and S.B. Melville, Effects of Clostridium perfringens alpha-toxin (PLC) and perfringolysin O (PFO) on cytotoxicity to macrophages, on escape from the phagosomes of macrophages, and on persistence of C. perfringens in host tissues. Infect Immun, 2004. 72(9): p. 5204-15. 63. Mehdizadeh Gohari, I., et al., Pathogenicity and virulence of Clostridium perfringens. 2021. 12(1): p. 723-753. 64. Lee, H. and Y.J.F.S.o.A.R. Yoon, Etiological agents implicated in foodborne illness world wide. 2021. 41(1): p. 1. 65. Gould, L., et al., Centers for Disease C, Prevention: Surveillance for foodborne disease outbreaks-United States, 1998-2008. 2013. 62(2): p. 1-34. 66. Sarker, M.R., et al., Comparative experiments to examine the effects of heating on vegetative cells and spores of Clostridium perfringens isolates carrying plasmid genes versus chromosomal enterotoxin genes. 2000. 66(8): p. 3234-3240. 67. Caserta, J.A., et al., Development and application of a mouse intestinal loop model to study the in vivo action of Clostridium perfringens enterotoxin. 2011. 79(8): p. 3020-3027. 68. McDonel, J., In F. Dorner and J. Drews (ed.), Pharmacology of Bacterial Toxins. 1986, Pergamon Press, Oxford. 69. Bos, J., et al., Fatal necrotizing colitis following a foodborne outbreak of enterotoxigenic Clostridium perfringens type A infection. 2005. 40(10): p. e78-e83. 70. Lindström, M., et al., Novel insights into the epidemiology of Clostridium perfringens type A food poisoning. 2011. 28(2): p. 192-198. 71. Rhodehamel, E. and S.J.B.A.M. Harmon, 8th Edition, RevisionA, Clostridium perfringens. 1998. 150 72. Collie, R.E. and B.A.J.J.o.c.m. McClane, Evidence that the enterotoxin gene can be episomal in Clostridium perfringens isolates associated with non-food-borne human gastrointestinal diseases. 1998. 36(1): p. 30-36. 73. Miyamoto, K., et al., Complete sequencing and diversity analysis of the enterotoxin-encoding plasmids in Clostridium perfringens type A non-food-borne human gastrointestinal disease isolates. 2006. 188(4): p. 1585-1598. 74. Asha, N., D. Tompkins, and M.J.J.o.C.M. Wilcox, Comparative analysis of prevalence, risk factors, and molecular epidemiology of antibiotic-associated diarrhea due to Clostridium difficile, Clostridium perfringens, and Staphylococcus aureus. 2006. 44(8): p. 2785-2791. 75. Azimirad, M., et al., Prevalence and characterization of Clostridium perfringens toxinotypes among patients with antibiotic-associated diarrhea in Iran. 2019. 9(1): p. 1-9. 76. Heikinheimo, A., M. Lindström, and H.J.J.o.c.m. Korkeala, Enumeration and isolation of cpe-positive Clostridium perfringens spores from feces. 2004. 42(9): p. 3992-3997. 77. Sparks, S.G., et al., Genotyping of enterotoxigenic Clostridium perfringens fecal isolates associated with antibiotic-associated diarrhea and food poisoning in North America. 2001. 39(3): p. 883-888. 78. Lu, J., et al., Gas gangrene without wound: both lower extremities affected simultaneously. 2008. 26(8): p. 970. e3-970. e4. 79. Buboltz, J.B. and H.M.J.S. Murphy-Lavoie, Gas gangrene. 2020. 80. Titball, R.W.J.M., Gas gangrene: an open and closed case. 2005. 151(9): p. 2821-2828. 81. Sarvari, K.P., et al., Fatal Clostridium perfringens sepsis due to emphysematous gastritis and literature review. 2016. 40: p. 31-34. 82. Gui, L., et al., Fatal enteritis necroticans (pigbel) in a diabetic adult. 2002. 15(1): p. 66-70. 83. Fick, K. and A.J.T.L. Wolken, Necrotic jejunitis. 1949. 253(6552): p. 519-521. 84. Murrell, T., et al., Pig-bel: enteritis necroticans. A study in diagnosis and management. 1966: p. 217-22. 85. Egerton, J., P.J.J.o.P. Walker, and Bacteriology, The Isolation of Clostridium perfringens Type C from Necrotic Enteritis of Man in Papua-New Guinea. 1964. 88(1): p. 275-8. 86. Sayeed, S., et al., Beta toxin is essential for the intestinal virulence of Clostridium perfringens type C disease isolate CN3685 in a rabbit ileal loop model. 2008. 67(1): p. 15-30. 87. Matsuda, T., et al., Enteritis necroticans ‘pigbel’in a Japanese diabetic adult. 2007. 57(9): p. 622-626. 88. Johnson, S. and D.N. Gerding, Enterotoxemic infections, in The Clostridia. 1997, Elsevier. p. 117-140. 89. Wormald, J.C.R., et al., Adult necrotising enterocolitis—pig-bel disease: a Pacific disease in London. 2016. 2016: p. bcr2016217903. 151 90. Miclard, J., et al., Clostridium perfringensβ-toxin binding to vascular endothelial cells in a human case of enteritis necroticans. 2009. 58(6): p. 826-828. 91. Fitzgibbons, S.C., et al., Mortality of necrotizing enterocolitis expressed by birth weight categories. 2009. 44(6): p. 1072-1076. 92. Hsueh, W., et al., Necrotizing enterocolitis of the newborn: pathogenetic concepts in perspective. 1998. 1(1): p. 2-16. 93. Goldenberg, R.L., J.C. Hauth, and W.W.J.N.E.j.o.m. Andrews, Intrauterine infection and preterm delivery. 2000. 342(20): p. 1500-1507. 94. Kuppala, V.S., et al., Prolonged initial empirical antibiotic treatment is associated with adverse outcomes in premature infants. 2011. 159(5): p. 720-725. 95. Sim, K., et al., Dysbiosis anticipating necrotizing enterocolitis in very premature infants. 2015. 60(3): p. 389-397. 96. Pedersen, P.V., et al., Necrotising enterocolitis of the newborn—is it gas-gangrene of the bowel? 1976. 308(7988): p. 715-716. 97. Dittmar, E., et al., Necrotizing enterocolitis of the neonate with Clostridium perfringens: diagnosis, clinical course, and role of alpha toxin. 2008. 167(8): p. 891-895. 98. Berdon, W.E., et al., Necrotizing enterocolitis in the premature infant. 1964. 83(5): p. 879-887. 99. Thompson, A.M. and M.J.J.D. Bizzarro, Necrotizing enterocolitis in newborns. 2008. 68(9): p. 1227-1238. 100. Petty, J.K. and M.M. Ziegler. Operative strategies for necrotizing enterocolitis: the prevention and treatment of short-bowel syndrome. in Seminars in pediatric surgery. 2005. Elsevier. 101. Amimoto, K., et al., A novel toxin hom*ologous to large clostridial cytotoxins found in culture supernatant of Clostridium perfringens type C. 2007. 153(4): p. 1198-1206. 102. Songer, J.G.J.C.m.r., Clostridial enteric diseases of domestic animals. 1996. 9(2): p. 216-234. 103. Jäggi, et al., Retrospective study on necrotizing enteritis in piglets in Switzerland. 2009. 151(8): p. 369-375. 104. Songer, J.G. and F.A.J.J.o.v.d.i. Uzal, Clostridial enteric infections in pigs. 2005. 17(6): p. 528-536. 105. Posthaus, H., et al., Clostridium perfringens type C necrotic enteritis in pigs: diagnosis, pathogenesis, and prevention. 2020. 32(2): p. 203-212. 106. Simpson, K.M., R.J. Callan, and D.C.J.V.C.F.A.P. Van Metre, Clostridial abomasitis and enteritis in ruminants. 2018. 34(1): p. 155-184. 107. Songer, J.G. and D.W.J.A. Miskimins, Clostridial abomasitis in calves: case report and review of the literature. 2005. 11(5): p. 290-294. 108. Van Kruiningen, H.J., et al., Clostridial abomasal disease in Connecticut dairy calves. 2009. 50(8): p. 857. 152 109. Roeder, B., et al., Experimental induction of abdominal tympany, abomasitis, and abomasal ulceration by intraruminal inoculation of Clostridium perfringens type A in neonatal calves. 1988. 49(2): p. 201-207. 110. Roeder, B., et al., Isolation of Clostridium perfringens from neonatal calves with ruminal and abomasal tympany, abomasitis, and abomasal ulceration. 1987. 190(12): p. 1550-1555. 111. Fernández Miyakawa, M.E. and F.A.J.A.j.o.v.r. Uzal, Morphologic and physiologic changes induced by Clostridium perfringens type A α toxin in the intestine of sheep. 2005. 66(2): p. 251-255. 112. Goossens, E., et al., Rethinking the role of alpha toxin in Clostridium perfringens-associated enteric diseases: a review on bovine necro-haemorrhagic enteritis. 2017. 48(1): p. 1-17. 113. Songer, J.G. and D.W.J.A. Miskimmins, Clostridium perfringens type E enteritis in calves: two cases and a brief review of the literature. 2004. 10(4): p. 239-242. 114. Sindern, N., et al., Prevalence of Clostridium perfringens netE and netF toxin genes in the feces of dogs with acute hemorrhagic diarrhea syndrome. 2019. 33(1): p. 100-105. 115. Burrows, C.J.J.-A.A.H.A., Canine hemorrhagic gastroenteritis [Diseases of unknown etiology, dogs]. 1977. 116. Gohari, I.M., et al., NetF-producing Clostridium perfringens: clonality and plasmid pathogenicity loci analysis. 2017. 49: p. 32-38. 117. Skinner, J.T., et al., An economic analysis of the impact of subclinical (mild) necrotic enteritis in broiler chickens. 2010. 54(4): p. 1237-1240. 118. Keyburn, A.L., et al., Association between avian necrotic enteritis and Clostridium perfringens strains expressing NetB toxin. 2010. 41(2): p. 1-8. 119. Allaart, J.G., et al., NetB-producing and beta2-producing Clostridium perfringens associated with subclinical necrotic enteritis in laying hens in the Netherlands. 2012. 41(6): p. 541-546. 120. Mwangi, S., et al., Characterization of Clostridium perfringens recovered from broiler chicken affected by necrotic enteritis. 2019. 98(1): p. 128-135. 121. Immerseel, F.V., et al., Clostridium perfringens in poultry: an emerging threat for animal and public health. 2004. 33(6): p. 537-549. 122. Kotsanas, D., et al., Novel use of tryptose sulfite cycloserine egg yolk agar for isolation of Clostridium perfringens during an outbreak of necrotizing enterocolitis in a neonatal unit. 2010. 48(11): p. 4263-4265. 123. Møller, K. and P.J.A. Ahrens, Comparison of Toxicity Neutralization-, ELISA-andPCR Tests for Typing ofClostridium perfringensand Detection of the Enterotoxin Gene by PCR. 1996. 2(2): p. 103-110. 124. Gobeli, S., et al., Antimicrobial susceptibility of canine Clostridium perfringens strains from Switzerland. 2012. 154(6): p. 247. 153 125. Xiao, Y., et al., A wide variety of Clostridium perfringens type A food-borne isolates that carry a chromosomal cpe gene belong to one multilocus sequence typing cluster. 2012. 78(19): p. 7060-7068. 126. Meer, R.R. and J.G. Songer, Multiplex polymerase chain reaction assay for genotyping Clostridium perfringens. Am J Vet Res, 1997. 58(7): p. 702-5. 127. Wang, R.F., et al., A 16S rDNA-based PCR method for rapid and specific detection of Clostridium perfringens in food. Mol Cell Probes, 1994. 8(2): p. 131-7. 128. Xu, W., et al., Prevalence and characterization of Clostridium perfringens isolated from different chicken farms in China. Anaerobe, 2021. 72: p. 102467. 129. Leal, J., et al., Epidemiology of Clostridium species bacteremia in Calgary, Canada, 2000-2006. J Infect, 2008. 57(3): p. 198-203. 130. Hamza, D., et al., New Insight of Apparently Healthy Animals as a Potential Reservoir for Clostridium Perfringens: A Public Health Implication. J Vet Res, 2018. 62(4): p. 457-462. 131. Forero, A.J., et al., High frequency of toxigenic Clostridium difficile and Clostridium perfringens coinfection among diarrheic patients at health care facility-onset (HCFO) and community-onset (CO) centers in Bogotá, Colombia. 2019. 11(1): p. 1-6. 132. Shrestha, A., F.A. Uzal, and B.A.J.M.s. McClane, Enterotoxic clostridia: Clostridium perfringens enteric diseases. 2018. 6(5): p. 6.5. 07. 133. Feng, Y., et al., Phylogenetic and genomic analysis reveals high genomic openness and genetic diversity of Clostridium perfringens. 2020. 6(10). 134. Hall, B.G.J.N.R.M., Predicting the evolution of antibiotic resistance genes. 2004. 2(5): p. 430-435. 135. Vrancianu, C.O., et al., Targeting plasmids to limit acquisition and transmission of antimicrobial resistance. 2020. 11. 136. Davies, T.A., et al., Effects of the 7-valent pneumococcal conjugate vaccine on US levofloxacin-resistant Streptococcus pneumoniae. 2008. 14(3): p. 187-196. 137. Martinez, J.L. and F. Baquero, Mutation frequencies and antibiotic resistance. Antimicrob Agents Chemother, 2000. 44(7): p. 1771-7. 138. Knopp, M. and D.I. Andersson, Predictable Phenotypes of Antibiotic Resistance Mutations. mBio, 2018. 9(3). 139. Shimizu, T., et al., Complete genome sequence of Clostridium perfringens, an anaerobic flesh-eater. 2002. 99(2): p. 996-1001. 140. Rood, J.I., V.N. Scott, and C.L.J.P. Duncan, Identification of a transferable tetracycline resistance plasmid (pCW3) from Clostridium perfringens. 1978. 1(4): p. 563-570. 141. Bannam, T.L., et al., Functional identification of conjugation and replication regions of the tetracycline resistance plasmid pCW3 from Clostridium perfringens. 2006. 188(13): p. 4942-4951. 142. Lyras, D., J.I.J.A.A. Rood, and Chemotherapy, Genetic organization and distribution of tetracycline resistance determinants in Clostridium perfringens. 1996. 40(11): p. 2500-2504. 154 143. Slavić, Đ., et al., Antimicrobial susceptibility of Clostridium perfringens isolates of bovine, chicken, porcine, and turkey origin from Ontario. 2011. 75(2): p. 89-97. 144. Soge, O., et al., A conjugative macrolide resistance gene, mef (A), in environmental Clostridium perfringens carrying multiple macrolide and/or tetracycline resistance genes. 2009. 106(1): p. 34-40. 145. Ngamwongsatit, B., et al., Multidrug resistance in Clostridium perfringens isolated from diarrheal neonatal piglets in Thailand. 2016. 38: p. 88-93. 146. Han, X., et al., Functional analysis of a bacitracin resistance determinant located on ICE Cp1, a novel Tn 916-like element from a conjugative plasmid in Clostridium perfringens. 2015. 59(11): p. 6855-6865. 147. Yao, P.Y. and P. Annamaraju, Clostridium perfringens Infection, in StatPearls. 2024, StatPearls Publishing Copyright © 2024, StatPearls Publishing LLC.: Treasure Island (FL) companies. Disclosure: Pavan Annamaraju declares no relevant financial relationships with ineligible companies. 148. Leiblein, M., et al., Clostridial Gas Gangrene - A Rare but Deadly Infection: Case series and Comparison to Other Necrotizing Soft Tissue Infections. Orthop Surg, 2020. 12(6): p. 1733-1747. 149. Abdel-Glil, M.Y., et al., Comparative in silico genome analysis of Clostridium perfringens unravels stable phylogroups with different genome characteristics and pathogenic potential. 2021. 11(1): p. 1-15. 139 9. CONCLUSIONES CAPÍTULO I: ¨ El primer panorama epidemiológico sobre la frecuencia de detección de C. perfringens en varias especies animales en Colombia revela una elevada presencia de esta bacteria en animales domésticos y cerdos. ¨ La relación de la toxina CPB2 de C. perfringens con la diabetes, sugiere interacciones con el sistema inmunitario del huésped. Factores de riesgo como la inmunidad comprometida, la hiperglucemia, resaltan la importancia de comprender esta dinámica en pacientes diabéticos. ¨ Comprender la epidemiología local de C. perfringens mediante estudios microgeográficos es crucial para anticiparse a las amenazas para la salud pública y hacerles frente con eficacia. CAPÍTULO II: ¨ La representación de genomas públicos de C. perfringens en países en desarrollo es escasa, probablemente debido a la falta de vigilancia epidemiológica y a la limitada recopilación de datos genómicos. ¨ Los análisis basados en MLST y el genoma central, utilizando genomas públicos de C. perfringens, revelan una estrecha relación entre aislamientos de diferentes hospederos. Estos hallazgos respaldan la marcada asociación de C. perfringens con las ETA y sugieren su potencial zoonótico. ¨ La identificación de toxinotipos de C. perfringens poco comunes en humanos, como el tipo G asociado con enfermedades en aves, junto con la presencia de aislamientos de tipo D y E vinculados con enfermedades en rumiantes, refleja la adaptabilidad de C. perfringens a diferentes hospedadores y resalta su papel zoonótico. ¨ Se requieren llevar a cabo estudios de caracterización genómica de microorganismos relevantes para la salud pública, particularmente en países en desarrollo, con el fin de comprender la diversidad genética, los patrones de dispersión y los marcadores de virulencia circulantes, que permitan instaurar medidas sanitarias y tratamientos adaptados a la realidad local. 140 CAPÍTULO III: ¨ El análisis microgeográfico de la estructura poblacional a partir de genomas colombianos de C. perfringens, reveló una amplia diversidad genética, sugiriendo posibles eventos de dispersión entre caninos, felinos y humanos. Estos hallazgos resaltan la necesidad de incrementar prácticas de higiene eficaces, especialmente en los cuidadores de animales domésticos con el objetivo de prevenir la transmisión a nivel comunitario. ¨ La presencia de alveolisina en aislamientos de C. perfringens de gatos asintomáticos en la región colombiana, junto con su alta capacidad de inhibición del crecimiento celular in vitro, sugiere un potencial significativo de virulencia en animales domésticos. Se requieren investigaciones adicionales para comprender mejor el papel de la alveolisina en la patogénesis de C. perfringens. ¨ Los aislamientos de C. perfringens portadores de la toxina PFOA, relacionada con la enterocolitis necrotizante, están presentes en individuos asintomáticos. Las pruebas fenotípicas, revelan su capacidad para generar hemólisis completa, inhibición del crecimiento celular in vitro y una mayor capacidad de esporulación. Esto sugiere un potencial riesgo de transmisión e infección a nivel comunitario y destaca la necesidad de comprender mejor el impacto de las toxinas accesorias a través de la caracterización