Abstract
The consumption of wildlife meat is traditionally accepted in the Peruvian Amazon; however, little is known about the pathogens present in this type of food. One of the most frequently consumed species is a rodent, the paca (Cuniculus paca) or “majaz” in the Peruvian language. The objective of this study was to determine the presence of Salmonella enterica and its antimicrobial resistance profile in paca carcasses sold in the Belen Market of Iquitos-Peru. An observational, descriptive, cross-sectional study was carried out. Fresh and smoked paca carcasses (72 samples) were evaluated during the low-rain period (July 2019) in the traditional market of Iquitos, in the Amazonian Region. Meat samples were swabbed, and International Standards Organization (ISO) 6579-1: 2017 protocol was followed to Salmonella isolation. Antimicrobial susceptibility analysis was performed by the disk diffusion method. In addition, serotyping was performed by using the Kauffmann-White scheme. A total of 25 strains of S. enterica were isolated in the paca carcasses, mainly in fresh carcasses (48.6%). The serovars isolated were Agona (45.8%), Infantis (41.7%), Wangata (8.3%), and Javiana (4.2%). A considerable number of the isolated strains were multidrug resistant (40%). The highest prevalence of resistance corresponded to trimethoprim-sulfamethoxazole (64%) followed by nitrofurantoin (44%), chloramphenicol (40%), cefotaxime (40%), and nalidixic acid (40%). Ten strains isolated (40%) were identified as producers of extended spectrum beta lactamases, all in S. enterica serovar Infantis. This study describes the presence of Salmonella Infantis with multidrug resistance profiles in wildlife meat carcasses, making the consumption of this type of products a risk factor for the development of foodborne diseases in the Amazon region. Institutional Review Board: Approval Resolution of Thesis Project: N° 024-DACMVZ-DAFCVB-U.CIENTÍFICA-2019.
Introduction
Wildlife meat consumption is widely practiced in both rural and urban areas of the Amazon, being an important source of protein for the local population (Bodmer et al., 2004; Coomes et al., 2020). The hunting of wild animals in the Amazonian rainforest varies according to the season and the availability of natural resources, which have an impact on the volume of wildlife hunted and the type of presentation of the final product as fresh, smoked, or salt-cured meat (Bodmer et al., 2004; Cáceres, 2015). Traditional preservation techniques are usually performed in the Amazon region to preserve meat during the transfer from the forest to cities (Cáceres, 2015).
One of the major cities where bushmeat is commercialized in Peru is Iquitos (Bodmer et al., 2004). Indeed, 82.8% of the urban population consume bushmeat and mainly acquire these products in the Belen Market, which is the largest center for the sale of wildlife meat in Peru (García, 1995) and is also one of the most important in the Amazon region (van Vliet et al., 2014). However, most meat merchants sell their products in street stalls with low food safety conditions (Cáceres, 2015).
One of the wild mammal frequently consumed in the Neotropics is the paca (Cuniculus paca) (Bodmer and Lozano, 2001; Koster et al., 2010; Santos-Fita et al., 2012), a large rodent (weight around 6 to 12 kilos) of the same suborder as the guinea pig and capibara (Ortega and Arita, 2014). This animal has a thick skin and is usually sold as a carcass or cut portions of a carcass. The meat of the paca is appreciated for its tenderness and flavor in several countries (Koster et al., 2010; Santos-Fita et al., 2012; Cáceres, 2015). Although the authors are aware that bushmeat trade can reach the capital of Peru (Lima), this occurs sporadically. Paca consumption is mainly limited to the high-land and low-land Amazonian forests (Bodmer et al., 2004; Coomes et al., 2020). Despite the usual consumption of this type of meat in some regions, there is no sanitary control at any stage of the trade chain, and therefore, potential foodborne pathogens may be present.
Salmonella enterica is an important foodborne pathogen in low- and middle-income countries (Graham, 2002; Majowicz et al., 2010) and is mainly transmitted by the consumption of contaminated food, causing moderate to severe diarrhea, and in some cases, severe complications (WHO, 2018). Approximately 150,000 people die annually from salmonellosis (Majowicz et al., 2010). In the past decade, there has been an increase in the number of cases of salmonellosis in Peru often caused by multidrug-resistant (MDR) and extended spectrum beta lactamases (ESBL)-producing strains (Riveros and Ochoa, 2015; Silva et al., 2017). However, more information on the epidemiology of this disease is lacking in the South American region, especially in tropical cities where 151,000 salmonellosis cases are estimated annually (Majowicz et al., 2010).
Therefore, the aim of this study was to determine the presence of S. enterica in paca carcasses sold in the Belen Market, as well as identify the serotypes and the antimicrobial resistance of the strains isolated.
Materials and Methods
Study site
The paca carcasses analyzed were from the informal area of the Belen Market in the city of Iquitos, in the Loreto Region, located in the northeast of the Peruvian amazon. The market has a 1500 m2 formal area (occupied by 260 vendors) and a bigger informal commercialization area located in the adjacent streets (occupied by 5379 vendors) (Cáceres, 2015; PRODUCE, 2015), involving retail areas with rustic wooden tables, with or without paper covers, a lack of potable water, no refrigerators, and a general lack of care for avoiding cross-contamination with other types of meat. The sampling occurred in the low-rain season during the month of July 2019, when climatic conditions were characterized by a maximum temperature of 31.2°C, a minimum temperature of 21.3°C, and an average rainfall of 157.75 mm per month (SENAHMI, 2020).
Sample collection
The minimum sample size, with a limit prevalence of 8.3%, was n = 35 samples, based on the Disease Detection formula and the limit prevalence of Salmonella spp. found in ronsoco carcasses (Hydrochoerus hydrochaeris) (Álvarez and Barragán, 2014). Convenience sampling was performed, choosing only market stalls that sold pacas carcasses exclusively and agreed to participate in the study. The vendor bought pacas from different Amazonian hunters. A total of 72 samples of paca carcasses (37 fresh and 35 smoked) were randomly selected. An area of 25 cm2 of the medial plane of hind leg skin was swabbed (Avagnina et al., 2012). Cary-Blair medium (Oxoid) was used for transportation at room temperature to the Food Microbiology Laboratory of the Universidad Nacional de la Amazonía Peruana, in Iquitos, where the samples were processed.
Isolation and identification of Salmonella
Isolation was performed according to the International Standards Organization (ISO) 6579-1: 2017 protocol (ISO, 2017). Briefly, the swabs from the transport medium were transferred to the nonselective Buffered Peptonated Water (Merck) enrichment broth for 18 h at 37°C. Then, 0.1 mL was transferred to Rappaport-Vassiliadis Soy Broth (Merck) for a 24 h incubation at 41°C, after which it was cultured by streaking in Xylose Lysine Deoxycholate agar (Merck) and incubated for 24 h at 37°C. Colonies suspected of Salmonella spp. were identified by the classical biochemistry tests: Triple Sugar Iron Agar (Oxoid), Lysine Iron Agar (Oxoid), Urea Broth (Merck), Tryptone Broth (prepared manually) for Indol Test, and Simmons Citrate Agar (Oxoid). In addition, late enrichment was performed, leaving the previously incubated tubes of Rappaport-Vassiliadis Broth at room temperature to incubate for 5 d and then proceed in the same way by culture in Xylose Lysine Deoxycholate agar (Davies et al., 2000).
Serotyping and antimicrobial susceptibility testing
Positive samples were sent to the Enteropathogen Reference Laboratory of the Instituto Nacional de Salud in Lima for serotyping and antimicrobial susceptibility testing. The Kauffmann-White scheme based on surface antigens was performed for serotyping the strains isolated. To determine antimicrobial susceptibility, the disk diffusion method was used following the Clinical & Laboratory Standards Institute (CLSI, 2016) guideline. Eleven clinically important antibiotics were used (WHO, 2017) and distributed in cephalosporins: ceftriaxone (30 μg), cefotaxime (30 μg), ceftazidime (30 μg); aminopenicillins: ampicillin (10 μg) and amoxicillin plus clavulanic acid (20/10 μg); quinolones: nalidixic acid (30 μg) and ciprofloxacin (5 μg); tetracyclines: tetracycline (30 μg); nitrofurans: nitrofurantoin (300 μg); phenicols: chloramphenicol (30 μg); and also folate inhibitors: trimethoprim-sulfamethoxazole (23.75/1.25 μg).
Data analysis
The statistical analysis included the 72 samples analyzed. Statistical associations between the variables “Presence of Salmonella” and “Type of carcass (smoked, fresh)” were evaluated by the Fisher test and were performed by using the GraphPad Prism 8 software. A p-value was considered significant when p was <0.05.
Ethical conditions
This study was approved by the Ethics Committee of the Universidad Científica del Sur (Certificate No. 92-CIEI-AB-CIENTIFICA-2019).
Results
A total of 25 strains of Salmonella spp. were isolated in 29.2% (21/72) of the paca carcasses evaluated (Table 1). The presence of Salmonella was 48.6% (18/37) in fresh paca carcasses and 8.6% (3/35) in smoked paca carcasses; a statistically significant association exists between the presence of Salmonella spp. and the type of paca carcass, Fisher's exact test (p < 0.001).
Isolation Frequency and Association of Salmonella enterica Serotypes in Two Types of Paca (Cuniculus paca) Carcasses (n = 72)
Fisher's exact test.
Most strains were isolated from normal enrichment, and four strains were isolated by late enrichment (Table 1). The serovars isolated were Agona (45.8%), Infantis (41.7%), Wangata (8.3%), and Javiana (4.2%). In three samples, there were both Salmonella Agona and Infantis, and in one sample both Salmonella Agona and Wangata were present. In addition, one isolate was unable to be serotyped, being then classified as “indetermined.”
Of the 25 strains analyzed, 76% were resistant to at least one antibiotic, 40% were resistant to at least three classes of antibiotics and were considered MDR (Table 2) (Parry and Threlfall, 2008). The highest resistance rate for the Agona serotype corresponded to trimethoprim-sulfamethoxazole (63.6%). The Wangata and Javiana serovars were susceptible to all antibiotics, except for a strain of Wangata that was resistant to ampicillin. For the Infantis serovar, resistance was evidenced for the antibiotics cefotaxime (100%), nalidixic acid (100%), tetracycline (100%), and nitrofurantoin (100%), followed by trimethoprim-sulfamethoxazole (90%), ampicillin (90%), and ceftriaxone (90%), in addition to showing decreased resistance against ciprofloxacin (100%). Moreover, all strains isolated from this serotype were ESBL producers.
Antimicrobial Resistance Profiles of Salmonella enterica Serotypes Isolated in Fresh and Smoked Carcasses of Paca (Cuniculus paca) in the Belen Market
POS: positive for MDR pattern or BLEE production.
NEG: negative for MDR pattern or BLEE production.
AMC, amoxicillin plus clavulanic acid; AMP, ampicillin; AM-PE, aminopenicillins; C, chloramphenicol; CAZ, ceftazidime; CEPH, cephalosporins; CIP, ciprofloxacin; CRO, ceftriaxone; CTX, cefotaxime; F, nitrofurantoin; FO-IN, folate inhibitors; MDR, multidrug-resistant; NA, nalidixic acid; NITR, nitrofurans; PHEN, phenicols; QUIN, quinolones; R/I/S, resistant/intermediate resistant/sensible; SXT, trimethoprim-sulfamethoxazole; TE, tetracycline; TETR, tetracyclines.
Discussion
This study demonstrates for the first time the presence of Salmonella in paca carcasses in the Amazon region of Peru, with greater contamination being found in fresh carcasses (48.6%) compared with other studies evaluating carcasses of fauna in the South American and African tropics (Bachand et al., 2012; Álvarez y Barragán, 2014), and others in which the presence of this pathogen was not found (Lozano et al., 2012). Smoked carcasses showed 8.6% of contamination by Salmonella, with levels being similar to those reported in smoked domestic animal meat (Yang et al., 2019). Although traditional preservation techniques are used for wildlife meat, Salmonella is a genus that can survive in foods with low water activity (Chitrakar et al., 2019).
All the serotypes found in this study have been reported as pathogens for humans (Silva et al., 2017; ECDC/EFSA, 2018; Simpson et al., 2019) and some as emerging serovars (Collins et al., 2019; Ferrari et al., 2019). The Agona serovar was the most frequent in this study (45.8%) and has mainly been reported in pigs and chickens from different continents (Haley et al., 2012; Torii et al., 2019; Zhou et al., 2019). However, it is also found in water buffaloes and turkeys in Brazil (Oliveira et al., 2014; Silva et al., 2018) and in other rodents consumed in Asia (Saengthongpinit et al., 2019). Nevertheless, it should be mentioned that some of the first reports of Salmonella Agona were from Peruvian hydrobiological products (fishmeal) (Clark et al., 1973), expanding the distribution of this serotype in the country.
The second most frequent serovar was Salmonella Infantis (41.7%), and it is usually reported in chickens and pigs (Valderrama et al., 2014; Vallejos-Sánchez et al., 2019; Tîrziu et al., 2020). This serovar has important relevance in Peru, since it has been isolated in different regions of the country (Quino et al., 2019) and is associated with epidemic outbreaks in the population (Silva et al., 2017; Garcia et al., 2019). The increase of this serotype has been reported during diarrheal symptoms in humans, displacing other serovars such as Enteritidis and Typhimurium (Garcia et al., 2019). This serovar has also been isolated in the Loreto region, which would indicate that this serovar has been circulating in this region for years (Valderrama et al., 2014).
Two other serovars reported in this study are Wangata (8.3%) and Javiana (4.2%), both of which have been isolated in birds (Obi et al., 2016; Khan et al., 2018; Collins et al., 2019), reptiles (Collins et al., 2019; Mukherjee et al., 2019), and the Javiana serotype has also been described in rodents (Saengthongpinit et al., 2019). In addition, there is evidence that the presence of these serotypes is associated with high temperature weather conditions (Simpson et al., 2019) and humidity (Huang et al., 2017; Simpson et al., 2019), similar to the climatic conditions described in the present study. Finally, it should be noted that the Javiana serovar could only be isolated through the late enrichment described in this study, corroborating the utility of this method (Davies et al., 2000).
The highest frequency of antimicrobial resistance of the strains isolated corresponded to trimethoprim-sulfamethoxazole (64%), similar to cephalosporins, cefotaxime (40%), and ceftriaxone (36%), with these results being higher than other reports on domestic animals (Ríos et al., 2019), fresh meat (Yang et al., 2019), and hydrobiological products (Yen et al., 2020). However, there are also studies describing higher antimicrobial resistance in trimethoprim-sulfamethoxazole (Nayarit-Ballesteros et al., 2016) and in cephalosporins for isolates in clinical samples of diarrhea due to the Infantis serovar (Granda et al., 2019).
Regarding quinolones, resistance to nalidixic acid (40%) was shown in this study, with this result being similar to values found in domestic animals (Centeno et al., 2018; Yang et al., 2019). In contrast, ciprofloxacin had 40% intermediate resistance, unlike other studies in Peru in which no resistance was reported (Ríos et al., 2019) or is minimal (Centeno et al., 2018). Therefore, it is very important to carry out studies to evaluate antimicrobial resistance in the main Salmonella serotypes in Peru (Quino et al., 2019) because of the presence of resistance to first-line drugs used in the treatment of salmonellosis (Medalla et al., 2016; Godínez-Oviedo et al., 2020; Nadi et al., 2020).
All the Salmonella Infantis isolates were MDR and were ESBL producers. These results are consistent with other studies conducted at a regional level (Vinueza-Burgos et al., 2016, 2019; Quino et al., 2019), and they highlight the importance of evaluating antimicrobial resistance based on Salmonella serovars (Table 3) (Michael and Schwarz, 2016; Saengthongpinit et al., 2019). The presence of MDR- and ESBL-producing Salmonella Infantis complicates the prognosis of cases of diarrhea that are usually treated with fluoroquinolones (Lübbert, 2016), in addition to complications in extraintestinal infections caused by Salmonella (Crump et al., 2015).
Antimicrobial Resistance Profiles in Salmonella enterica Isolates in Paca Carcasses (Cuniculus paca) from the Belen Market
AMC, amoxicillin plus clavulanic acid; AMP, ampicillin; C, chloramphenicol; CAZ, ceftazidime; CIP, ciprofloxacin; CRO, ceftriaxone; CTX, cefotaxime; F, nitrofurantoin; NA, nalidixic acid; SXT, trimethoprim-sulfamethoxazole; TE, tetracycline.
Although the origin of Salmonella contamination was unknown, different serotypes and antimicrobial resistance patterns in the samples analyzed could indicate that different sources of carcass contamination were involved. It was of note that the contamination values in fresh carcasses coincided with those found in rectal swabs of rodents (Ribas et al., 2016; Saengthongpinit et al., 2019), animals that are usually hosts of bacteria such as Salmonella, among which the paca (Adesiyun et al., 1998) and other rodents are found (Farikoski et al., 2019). Therefore, it is possible that, in part, contamination is due to poor handling techniques during evisceration, especially with Agona serovar, which demonstrate low-resistance profiles. On the contrary, it has been hypothesized that the Infantis serovar could correspond to cross-contamination during distribution or commercialization in the Belen Market, since it is normal for potential buyers to handle the carcasses before acquisition, possibly explaining the high-resistance profile. Further, cross-contamination could occur during transportation to the sale point, since the carcasses are usually shipped with other types of meat, including poultry, river fish, reptile meat, and other wild mammalian meat. It is also important to highlight that although this market is large, it is largely unregulated (Cáceres, 2015), and thus, no hygiene rules are applied. In this regard, the researchers observed that when sellers finished their workday, animals such as dogs, cats, and vultures were looking for food leftovers at the market stalls, and were even standing at tables where meat would be sold the next day. Nonetheless, the significant difference on the number of Salmonella among fresh and smoked paca carcasses suggests an effect of the meat processing, and subsequently that the Salmonella are in the meat before heating processes.
It was not possible to include dried salted paca carcassess among the samples. However, the data found are still relevant since the consumption of fresh carcasses is higher. During sample collection, the minimum number of samples were collected to ensure investigator well-being due to the unregulated and even illegal conditions of the Belen Market. Nonetheless, the number of samples collected was adequate according to the Detection of Disease formula and allowed achieving significant results.
This study describes the presence of Salmonella in paca carcasses, which is usually consumed by children and the elderly because of the health benefits attributed to this type of meat (Cáceres, 2015). Those who regularly consume these types of products should ensure that appropriate preparation methods are used to eliminate pathogenic bacteria, and special care should be taken in meat handling to avoid cross-contamination. The results of this study provide information that can help to develop public policies since there is currently no agency in Peru monitoring the food safety of bushmeat, despite its frequent consumption in the Amazon region. Further studies are needed to determine the presence of different bacteria in other important commercialization species, as well as analyze the epidemiology of Salmonella in paca meat. In addition, the results of this study can serve as a baseline for the analysis of antimicrobial resistance trends in the city of Iquitos and in products in the main markets of Loreto, and also highlight the importance of monitoring zoonotic bacteria and other pathogens in bushmeat in South America.
Conclusion
This study describes the presence of S. enterica in paca (Cuniculus paca) carcasses in the Belen Market in Iquitos, showing that potential pathogens are present in this type of wild meat. It highlights that the high proportion presented an MDR strain's profile with ESBL producers and its implication in public health. Continuous surveillance studies of antimicrobial resistance are required to establish possible links between reservoirs and humans.
Footnotes
Acknowledgments
The authors acknowledge the Instituto Nacional de Salud for serotyping the samples. They thank the Bromatologist Marilia Vales and Biologist Jessy Vasquez from the Food Microbiology Laboratory in the Universidad Nacional de la Amazonia Peruana.
Disclosure Statement
No competing financial interests exist.
Funding Information
This project was funded by the Universidad Científica del Sur through the “Concurso de Fondos para Proyectos de Tesis 2018-2.” (RD No. 12-DGIDI-CIENTIFICA-2018).
