Detection of Antimicrobial Resistance Genes in Escherichia coli Isolated from Black Howler Monkeys (Alouatta pigra) and Domestic Animals in Fragmented Rain-Forest Areas in Tabasco, Mexico

Antimicrobial resistance (AMR) is a major global health challenge and is an important topic at the interface of human, animal, and ecosystem health (Butaye et al. 2014). The occurrence of AMR in bacteria in wild animals has been increasingly reported in many species (Ramey and Ahlstrom 2020), but its prevalence may be underestimated. Anthropogenic factors such as habitat overlap can increase the risk of transmission of bacteria carrying AMR genes among humans, livestock, and wild animals, creating new host reservoirs (Rwego et al. 2008). Prevalence of AMR bacteria among wildlife appears to be multifactorial, depending on habitat use and foraging strategy of the species, particularly as they relate to anthropogenic inputs into the environment (Ramey and Ahlstrom 2020).

The black howler monkey (Alouatta pigra) is a neotropical primate inhabiting tropical rain-forest areas from southern Mexico and is considered a critically endangered species on the Red List (International Union for Conservation of Nature 2019). Being arboreal, these primates are not expected to directly interact often with either humans or domestic animals (Ramey and Ahlstrom 2020). However, in Balancán, Tabasco, Mexico, the high degree of habitat fragmentation has led the monkeys to live in small forest patches that they share with domestic fauna and humans, often descending from the trees to move between forest patches through livestock roaming areas (Supplementary Material [video]; Estrada and Coates-Estrada 1996; Pozo-Montuy and Serio-Silva 2007). We believe that these behaviors expose howler monkeys to the transmission of microbes from livestock as reported in several studies conducted with other primate species (Rolland et al. 1985; Rwego et al. 2008; Tegner et al. 2019).

Escherichia coli is a common gastrointestinal bacterium that has been used as a model for microbial transmission among wildlife, domestic animals, and humans; moreover, these studies have provided valuable information in formulating conservation recommendations to reduce pathogen transmission (Rwego et al. 2008; Allen et al. 2011). The objective of this study was to detect AMR genes in E. coli isolates from black howler monkeys and domestic animals in rain-forest fragments with areas of agricultural and livestock activities.

The study was performed in the region of the Usumacinta River basin in Balancán, Tabasco, Mexico (17°46′48″N, 91°30′22″W; Fig. 1). This area is part of a highly fragmented landscape where original forests have lost around 80% of their original cover (Pozo-Montuy and Serio-Silva 2007; Hernández-Rodríguez et al. 2019). During January 2016, samples were collected from two troops of black howler monkeys inhabiting two forest fragments and from domestic animals such as cattle (Bos taurus), sheep (Ovis aries), and horses (Equus caballus) from the surrounding open grazing areas. Recently excreted fecal samples were collected using fecal swab^™ (Copan Diagnostics Inc., Murrieta, California, USA), avoiding soil contamination, placed in phosphate-buffered saline (pH 7.4; Thermo Fisher, Waltham, Massachusetts, USA; Cristóbal-Azkarate et al. 2014), and stored in a cooler box with ice packs (5±3 C) followed by electrical refrigeration once available (Tegner et al. 2019). Only one sample per individual was collected. Fecal swabs were cultured for E. coli on eosin–methylene blue agar overnight at 37 C using standard protocols (Renoux and Terdjman 1951). From each animal, a pool of three representative colonies compatible with E. coli morphologic characteristics was selected for molecular identification. We extracted DNA from bacteria using the DNeasy blood and tissue kit (Qiagen, Hercules, California, USA). Molecular identification of E. coli by PCR was performed using a previously published protocol (Chen and Griffiths 1998). If the pool of E. coli from a sample was confirmed as positive, the animal was considered a carrier of E. coli. For the detection of AMR genes in E. coli DNA, we performed four different multiplex PCRs using the multiplex PCR kit (Qiagen) and 10 previously published pairs of specific primers for AMR genes (Table 1): sulfonamides (sul1, sul2, and sul3), tetracycline (tet[A], tet[B], and tet[C]), β-lactamase (blaTEM, blaSHV, and blaCMY-2), and chloramphenicol (cmlA).

We collected 51 fecal samples, one per individual: 16 from howler monkeys, 20 from cattle, 10 from sheep, and five from horses. We detected E. coli in 94% (48/51) of fecal samples, which were further analyzed for AMR genes; E. coli was not detected in three samples from cattle. Table 2 summarizes the number of samples analyzed per animal species and the prevalence of AMR genes.

Each AMR gene tested was detected in at least one sample. Resistance genes were detected in 33% (16/48) of samples carrying E. coli. The most prevalent resistance gene was tet(B) at 18% (9/48) and was the only gene detected in at least one sample from each animal species. The less prevalent resistance gene was tet(C) at 4% (2/48). We detected all AMR genes in samples from domestic animals, with sheep carrying all of those except for blaSHV. Horses and sheep showed high prevalence of AMR genes, with 60% (3/5) and 50% (5/10) respectively, whereas black howler monkeys showed the lowest prevalence, at 12% (2/16) with sul1 and tet(B) detected in one independent sample each. Most of the analyzed genes were detected in more than one animal species except for cmlA, which was found only in sheep.

Antimicrobial agents are indispensable for the control of bacterial infections; however, their indiscriminate use and management favor the selection of AMR bacteria (Butaye et al. 2014). Previous studies have detected AMR genes in bacteria from many wildlife species (Allen et al. 2011; Chen et al. 2018; Ramey and Ahlstrom 2020), including nonhuman primates (Rolland et al. 1985; Rwego et al. 2008; Tegner et al. 2019). However, few studies address this issue in endangered New World primates (Cristóbal-Azkarate et al. 2014).

Nonhuman primates can be asymptomatic carriers of AMR bacteria such as E. coli, Campylobacter spp., or Salmonella spp.; however, these bacteria can cause fatal diseases and are a concern regarding the conservation of free-ranging primate populations (Tegner et al. 2019). Results from this study showed a total AMR gene prevalence of 12% (2/16) for samples from black howler monkeys, lower compared with the domestic animals analyzed. The prevalence reported here for howler monkeys is also lower compared with a previous report for a Mexican primate species (Alouatta palliata) that had a prevalence of 43–47% for tetracycline and 12–19% for sulfonamide (Cristóbal-Azkarate et al. 2014).

The acquisition of bacteria in wild primates can occur through contact with human food waste, contact with other species, soil, feces, or contaminated water (Tegner et al. 2019). The arboreal habits of black howler monkeys reduce the contact with bacteria that may carry AMR genes (Estrada and Coates-Estrada 1996), unlike domestic animals that, in addition to receiving prophylactic treatments, are exposed to waste from other animals and humans (Allen et al. 2011; Ramey and Ahlstrom 2020). Howler monkeys from our study area have been reported to descend from the trees to move through livestock areas between forest fragments, and for direct water intake from tree holes and from a lagoon near the study site (Pozo-Montuy and Serio-Silva 2007; Supplementary Material [video]). This behavior in howler monkeys may increase their exposure to bacteria from other animals, also increasing the risk of acquisition of AMR bacteria as they are prevalent among the domestic animals that roam the area.

Our results showed a high prevalence of AMR genes in domestic animals compared with other studies. Rwego et al. (2008) reported a prevalence of 27% and Butaye et al. (2014) a range of 3.9% to 43%, whereas in our study the prevalence ranged between 35% and 60%. Increased AMR in domestic animals is associated with the use of antimicrobial agents and with direct exposure to bacteria carrying these genes (Allen et al. 2011). In Mexico, veterinary antibiotics are easily obtained without a prescription and can be used indiscriminately. In the Balancán area, antibiotics are commonly used for domestic animals and these practices could be the cause of the high prevalence of AMR genes in our study sample.

In conclusion, we report AMR genes in samples from black howler monkeys in Mexico and in domestic animals in Balancán. On the basis of our data and given the local habitat fragmentation, we infer that the transmission of ARM genes between domestic animal species and black howler monkeys is possible, but rare in the studied location. This study helps to emphasize the importance of the proper use of antimicrobials from the perspective of conservation medicine.

We thank the Estación de Investigación Primatológica y Vida Silvestre established in Balancán, Tabasco, Mexico by Instituto de Ecología A.C. for their support; Juan Carlos Serio Siva and Francisca Vidal for support with sample collection, and Dolores Tejero Geronimo for help in the field. We also thank Alejandro Azaola Espinosa and the Laboratorio de Biotecnología at the Universidad Autónoma Metropolitana-Xochimilco for their support in bacterial isolation and Hilda Montero Ladrón de Guevara from the Instituto de Salud Pública de la Universidad Veracruzana for support in the molecular biology laboratory.

Supplementary material for this article is online at http://dx.doi.org/10.7589/2020-10-243.