Introduction

An expanding human population has increased the risk of zoonosis (Devaux et al. 2019), particularly between humans and wild primates, given our close evolutionary histories and shared ecologies (Fuentes and Hockings 2010). Among the major classes of zoonotic agents, some enteropathogenic bacteria (family Enterobacteriaceae) can cause lethal diarrheal disease and pose a serious threat to all primates (McLennan et al. 2018). Enterobacteriaceae may enter primate populations from livestock, soil, human foods, and water (McLennan et al. 2018). Epidemiological studies in wild primates, both endangered species (gorillas, Gorilla gorilla; chimpanzees, Pan troglodytes) and widely distributed species that thrive in (peri)urban environments (rhesus macaques, Macaca mulatta; olive baboons, Papio anubis) speculate that primates, aside from being at risk and showing disease symptoms themselves, also act as reservoirs or carriers of Enterobacteriaceae and pose risks of bacterial spillover into overlapping humans and other wildlife (reviewed in McLennan et al. 2018).

Among the most widely distributed, socioecologically flexible primates, long-tailed macaques (Macaca fascicularis) are found throughout Southeast Asia, where frequent occurrence of human–macaque interactions have led to calls for macaque population control (Malaivijitnond et al. 2011). Such interactions may also increase macaques’ exposure to zoonotic Enterobacteriaceae, which nevertheless remain underassessed among wildlife populations in the Asian tropics. Here we report the prevalence of three Enterobacteriaceae {Salmonella spp., Shigella spp., and Shiga toxin producing Escherichia coli (STEC)] in long-tailed macaques living in human-impacted environments in Thailand.

Methods

Between July 22, 2019 and July 26, 2019, we collected and processed 168 samples from 143 macaques from three sites in Thailand—Wat Haad Moon (WHM: 16.3°N,100.1°E), Wat Ta Sung Tai (WTST: 15.5°N,99.5°E), and Wat Tam Mi Ka Wat (WTMK: 14.5°N,100°E)—across which samples were fairly evenly distributed (Electronic Supplementary Material [ESM] Table SI). All three sites are temples or towns connected to fragmented forests that have wild boars (Sus scrofa) and feral dogs (Malaivijitnond pers. obs.), which may shed Enterobacteriaceae. The monkeys are protected by the Wildlife Act of Thailand (B.E.2562). Macaques at all three sites forage in fragmented forests and rely heavily on human-provisioned foods.

Eighty-three percent of the samples (140/168) were rectal swabs from trapped, anesthetized individuals (Malaivijitnond et al. 2008; details in the ESM). For trapped animals, we recorded the identity tag, sex, and body- mass (kg) and estimated their age (yr) from their state of dental eruption (details in Malaivijitnond et al. 2008). The remaining samples (28/168, 17%) were fresh feces collected opportunistically from unidentified individuals when they were seen defecating. We collected these samples from the ground without soil contamination. We stored all samples on ice and transported them to the Department of Veterinary Public Health, Chulalongkorn University, on the day of collection.

We followed the procedures of the International Organization for Standardization to test for Salmonella spp. (ISO-6579) and Shigella spp. (ISO-21567). We used a procedure for Escherichia coli and coliforms to test for STEC, in accordance with the U.S. Food and Drug Administration (U.S. FDA) Bacteriological Analytical Manual (BAM) (ESM). We calculated the prevalence of each bacteria for each population as the number of samples in which we confirmed bacterial presence over the total number of samples processed.

To explore the impact of site, and macaque age, sex, and body mass, on bacterial infection, we ran a multivariate logistic general linear model (GLM) on an “effective” sample size of 140 macaques (details in the ESM). The outcome variable was the presence or absence of at least one type of Enterobacteriaceae in a macaque. Given the close evolutionary histories, similar ecologies, and similar epidemiology of the three enterobacterial taxa we processed, we assumed that environmental and host-specific variables will affect the likelihoods of infection with these bacteria in similar ways.

For more information on macaque trapping-and-release procedures, sample processing, the effective sample size, and data analysis, see the ESM.

Ethical Note

All protocols were approved by the Institutional Animal Care and Use Committee of the National Primate Research Centre of Thailand, Chulalongkorn University (NPRCT-CU) (No.1975007) and adhered to the principles for the ethical treatment of primates set by the American Society of Primatologists.

Data Availability

The dataset used in this study is also available in the ESM document provided.

Results

The prevalence of each type of Enterobacteriaceae was low in all three populations. Overall, Salmonella spp. showed the highest prevalence, followed by STEC and Shigella spp. (Table I). Across sites, Salmonella spp. was markedly more prevalent at WTMK than at WHM and WTST, while Shigella spp. and STEC showed similar prevalence (Table I).

Table I Summary of bacterial prevalence among long-tailed macaques at three sites in Thailand, in July 2019
Full size table

The GLM met standard diagnostic criteria (ESM) and was a significantly better fit than a null model (likelihood ratio test: χ2 (5) = 14.27, P = 0.01). The likelihood of infection was significantly negatively associated with body mass (B = −0.55, P = 0.02), positively but not significantly associated with age (B = 0.05, P = 0.09), and not significantly affected by sex (males vs. females: B = −0.62, P = 0.24) or study site (WTMK vs. WHM: B = 0.47, P = 0.53; WTST vs. WHM: B = 0.35, P = 0.64; WTST vs. WTMK: B = −0.12, P = 0.80).

Discussion

Long-tailed macaques at all three sites were exposed to human food and waste, contaminated water, and the feces of other animals with which they overlap spatially (wild boar, feral dogs, other macaques) or interact with socially (other macaques). All these are potential environmental sources of enterobacterial infection (Devaux et al. 2019; Malaivijitnond pers. obs.).

The low enterobacterial prevalence is consistent with studies of other wild primates (reviewed in McLennan et al. 2018). We mostly processed one sample from each macaque, which can decrease pathogen detectability and result in low prevalence. More frequent sampling may also establish location- or pathogen-specific differences in prevalence. However, we detected little evidence for differences in prevalence across pathogen types, except for a relatively high prevalence of Salmonella spp. among macaques at WTMK. This finding may relate to the denser macaque population and therefore potentially greater rates of contact/overlap with humans, other macaques, and other animals at WTMK compared to the other locations (Malaivijitnond pers. obs.).

The negative relationship between macaque body mass and bacterial infection may have implications for animal health. This pattern is unlikely to be because young, immunologically naive macaques weigh less and are more prone to infection, since we found a positive (rather than a negative) association between age and infection. Instead, lower body mass may be a consequence (rather than a predictor) of bacterial infection, since weight loss is a commonly observed symptom of gastroenteritis and diarrheal disease in humans and other mammals.

Our findings add to existing evidence of Enterobacteriaceae infection in wild primates living in human-impacted environments (McLennan et al. 2018). Throughout Southeast Asia, long-tailed macaques live in dense populations that share ecological space with both humans and other wildlife. These primates, in addition to being vulnerable to diarrheal disease themselves, may act as “superspreaders” of pathogens to overlapping humans, livestock, and other wildlife (Devaux et al. 2019). From both conservation and public health perspectives, it is therefore imperative to conduct more in-depth epidemiological and socioecological assessments of enterobacterial transmission and pathogenesis at human–nonhuman primate interfaces. Comparison with macaques living in less anthropogenic environments would be valuable to further our understanding of the epidemiology of infection.