Antimicrobial susceptibility profile of oral and rectal microbiota of non‐human primate species in Ghana: A threat to human health

Abstract Background The potential for the transfer of zoonotic diseases, including bacteria between human and non‐human primates (NHPs), is expected to rise. It is posited that NHPs that live in close contact with humans serve as sentinels and reservoirs for antibiotic‐resistant bacteria. Objectives The objective was to characterize the oral and rectal bacteria in Ghanaian NHPs and profile the antimicrobial susceptibility of the isolated bacteria. Methods Oral and rectal swabs were obtained from 40 immobilized wild and captive NHPs from 7 locations in Ghana. Standard bacteriological procedures were used in the isolation, preliminary identification, automated characterization and antimicrobial susceptibility test (AST) of bacteria using the Vitek 2 Compact system. Results Gram‐negative bacteria dominated isolates from the rectal swabs (n = 76, 85.4%), whereas Gram‐positive bacteria were more common in the oral swabs (n = 41, 82%). Staphylococcus haemolyticus (n = 7, 14%) was the most occurring bacterial species isolated from the oral swabs, whereas Escherichia coli (n = 32, 36%) dominated bacteria isolates from rectal swabs. Enterobacter spp. had the highest (39%) average phenotypic resistance to antimicrobials that were used for AST, whereas a trend of high resistance was recorded against norfloxacin, Ampicillin and Tetracycline in Gram‐negative bacteria. Similarly, among Gram‐positive bacteria, Staphylococcus spp. had the highest (25%) average phenotypic resistance to antimicrobials used for AST, and a trend of high resistance was recorded against penicillin G and oxacillin. Conclusions This study has established that apparently healthy NHPs that live in anthropized environments in Ghana harbour zoonotic and antimicrobial resistant bacteria.


INTRODUCTION
Although wildlife is generally not the driver for pandemics, it is known to serve as a constant source for microorganisms that have the potential to cause pandemics and panzootics (Can et al., 2019).In particular, pathogens that are transmitted among wildlife, livestock and humans pose a challenge to wildlife conservation and threat to human and animal health (Rwego et al., 2008).Over the years, the contact between wild non-human primates (NHPs) and human in Africa has intensified.Some of the reasons for this are human encroachment and resulting habitat destruction and (eco-)tourism (Rwego et al., 2008).
The close relationship between NHPs and human further increases the potential for the exchange of zoonotic pathogens (Vore et al., 2001).
Antimicrobial resistance (AMR) is a pandemic that is of global concern.It goes largely unrecognized even though 700,000 people die from infection with AMR annually (Pokharel et al., 2020).Although the One Health approach is more and more accepted, there is still inadequate cross-sectorial collaboration with regard to AMR (O'Neill, 2016;Pokharel et al., 2020).
Isolation of potentially zoonotic pathogenic bacteria in apparently healthy NHPs has been reported in West Africa (Okwori et al., 2014(Okwori et al., , 2013)).The number of transmission routes, such as faecal-oral or through bites (Devaux et al., 2019) highlights the risk of the exchange of pathogenic bacteria between human and NHPs.
There are indications that bacteria recovered from bite wounds are generally reflective of the oral bacteria of the biting animal.This is particularly significant in human-NHP interactions as it has been reported that NHPs that have close contact with humans usually inflict bite wounds (Abrahamian & Goldstein, 2011).A number of pathogenic bacteria, such as Enterobacteria, Staphylococcus spp., Streptococcus spp., Neisseria spp.and Haemophilus spp.have been isolated from the oral cavity of NHPs (Devaux et al., 2019;Sobreira et al., 2019).With regards to the faecal-oral route of transmission, apparently healthy NHPs have been reported to harbour pathogenic enteric bacteria, such as Escherichia coli, Klebsiella spp., Salmonella spp., Shigella spp.and Campylobacter spp.(Devaux et al., 2019).NHPs and humans are generally susceptible to infections caused by pathogenic enteric bacteria through zoonotic and anthroponotic transmissions even though they can be asymptomatic carriers (Medkour et al., 2021;Okwori et al., 2014).
Proximity of NHPs to human or anthropized environments can facilitate the transfer of antibiotic-resistant bacteria (ARB) between human and NHPs.This is significant as wild animals, such as NHPs with the exception of captive animals, usually do not come into contact with antimicrobials (Parsons et al., 2021).The exchange of ARB between human and NHPs in close proximity threatens not only human health but also animal health (Albuquerque et al., 2020).Given the reports of apparently healthy NHPs harbouring zoonotic bacteria pathogens (Egbetade et al., 2020;Okwori et al., 2014Okwori et al., , 2013)), it becomes important to conduct surveillance on the oral and rectal bacteria and their antimicrobial susceptibility profile of NHPs to provide information relevant to public health and NHP conservation.In this study, we therefore characterized the oral and rectal bacteria of NHPs in Ghana and determined their antimicrobial susceptibility profile.

Study design
This is a cross-sectional study which involved NHPs in Ghana that have high interaction with humans.Permits were obtained from the Wildlife Division of the Forestry Commission of Ghana with number 0095189 for the sampling of NHPs according to their approved guidelines.Good veterinary practice rules were applied to all sampling procedures, and all samples were taken by a licensed veterinarian.

Study areas
We sought information from the Ghana Wildlife Division on possi-
Immobilized NHPs were first weighed with a scale (KERNCH) before  being placed on the dorsal recumbency for sampling to commence.A physical examination of the NHPs was done, and vital parameters, such as body temperature, respiratory rate, heart rate and blood oxygen saturation, were taken every 5 min during sampling.The pulse, heart rate and blood oxygen saturation were monitored with a pulse oximeter (OxiMax N65, Nellcor).One oral (FITZCO CEP swabs) and one rectal (Polyester-tipped swabs) swab samples were obtained aseptically from the oral cavity and rectum of each of the animals, respectively.
The samples were transported on ice to the laboratories of the Kumasi Centre for Collaborative Research in Tropical Medicine (KCCR) for subsequent analysis.

Culture of swab samples
The samples were subjected to aerobic culture at 37

Automated bacteria identification
Polystyrene tubes (12 mm × 75 mm) (Sarstedt) were filled with 3 mL of 0.45% sterile saline (CareFusion).Homogenous suspensions of bacteria were prepared in the 3 mL 0.45% sterile saline.Suspensions were examined using the DensiCheck Plus (BioMérieux) device to ensure that the suspensions were within the turbidity range of 0.5-0.63 (McFarland standard) for both Gram-positive and Gramnegative bacteria.The appropriate Vitek cards (Vitek GP and Vitek GN) (BioMérieux), which contained 64 wells with miniaturized selective media and biochemical reagents, were used for bacterial identification.
The Vitek system identified the bacteria by filling the Vitek wells with the standard turbidity bacteria suspension in 3 mL of 0.45% sterile saline.The colour codes from the miniaturized selective media as well as the results of the miniaturized biochemical tests were analysed by the Vitek 2 Compact system software (Version 08.01, BioMérieux) to identify the bacteria species.

Data analysis
Data

Diversity of bacteria isolates
A total of 175 bacterial isolates were isolated from the oral and rectal swabs.Of this, 36 (20.6%) could not be identified by the Vitek 2 system.
Of the remaining 139 isolates, 85 (61.2%) and 54 (38.8%) were identified as Gram-negative and Gram-positive bacteria, respectively.Out of these 139 isolates, automated identification and AST was achieved for 119 (85.6%) bacterial isolates.The 119 isolates were composed of 85 (71.4%)Gram-negative bacteria and 34 (28.6%)Gram-positive bacteria.All these isolates had their antimicrobial susceptibility profile performed.A summary of the distribution of bacteria isolated from the oral and rectal swabs is shown in Figure 2.

Proportion of bacteria species harboured by non-human primates
The 40 NHPs which were included into the study harboured bacteria in the oral and rectal region in various proportions; in E. patas, E. coli (n = 15, 11.4%) was the most prevalent bacterium isolated.Staphylococcus spp.were the dominant isolates in both P. anubis and C. mona.Klebsiella spp.(n = 10, 7.8%) also dominated in C. sabaeus (Figure 3a).
In the distribution of oral and rectal bacterial isolates, E. patas had the highest proportion (n = 28, 20.1%) of rectal bacterial isolates followed by C. sabaeus (n = 26, 18.7%).Both P. anubis and E. patas had the highest proportion (n = 17, 12.2%) of oral bacterial isolates.Across all the NHP species, there was a higher proportion of rectal bacterial isolates compared to oral bacterial isolates (Figure 3b).More bacterial species were isolated from the oral and rectal regions of captive NHPs compared to wild NHPs.Isolates of Enterobacter spp., Klebsiella spp.and Rothia spp.were absent in the cultured samples of wild NHPs (Figure 3c).

F I G U R E 2
Distribution of bacteria isolated from the oral and rectal swabs.Out of 175 bacteria isolates, 85 isolates were Gram-negative and 90 were Gram-positive.Out of the 90 Gram-positive bacteria, 36 were unidentified, whereas 20 of the remaining 54 Gram-positive isolates were identified by the Vitek 2 system without an automated antimicrobial susceptibility test (AST).There were 85 Gram-negative and 34 Gram-positive isolates that had their automated identification and AST performed by the Vitek 2 system.

Antimicrobial susceptibility profile of isolated bacteria
Automated AST was performed by Vitek 2 employing the AST-N and AST-P cards.The results were recorded according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines.Staphylococcus spp.recorded the highest average phenotypic resistance (25%) against the antibiotics used for Gram-positive bacteria, whereas Enterococcus spp.recorded an average of 82% sensitivity to the antibiotics.Enterobacter spp.recorded the highest average phenotypic resistance (39%) against antibiotics used for Gram-negative bacteria, whereas E. coli recorded an average of 84% sensitivity to the same antibiotics (Figure 4).Among the 85 Gram-negative bacterial isolates, the highest frequency of phenotypic resistance was found against Norfloxacin (98%) followed by Ampicillin (56%) and Tetracycline (36%), and a high frequency of sensitivity to Meropenem (100%), Ciprofloxacin (99%) and Gentamicin (99%) (Table 3).Among the 34 Gram-positive bacterial isolates, the highest frequency of phenotypic resistance was against penicillin G (56%) followed by oxacillin (35%), and a high frequency of sensitivity to Ciprofloxacin (91%), Gentamicin (94%), Vancomycin (97%), and Erythromycin (91%) (Table 4).

Association between antimicrobial susceptibility profile and body part bacteria were isolated from as well as sex and origin of non-human primates
A chi-square test was performed to determine the relationship between the antimicrobial susceptibility profile and the body part (oral cavity and rectal region) bacteria were isolated from the origin and sex of NHPs.There was a significant association (p = 0.000119) between the antimicrobial susceptibility profile and body part, but not with the TA B L E 2 Distribution of identified bacteria species.

DISCUSSION
In this study, we aimed to characterize the oral and rectal microbiota of NHPs in Ghana and their antimicrobial susceptibility profile.
Opportunistic pathogens, such as E. coli, Klebsiella spp., Proteus spp., isolated from the mucosa of NHPs in many studies (Albuquerque et al., 2020;Carvalho et al., 2014;Egbetade et al., 2020) and are important bacterial pathogens that are known to cause pneumonia, nephritis and other debilitating conditions in NHPs (Carvalho et al., 2014).All NHPs in this study had frequent contact with humans especially those that were kept as pets.Bites inflicted by NHPs usually serve as a means of dissemination of oral pathogenic bacteria (Abrahamian & Goldstein, 2011).
In our study, we isolated pathogenic enteric bacteria, such as E.
coli, Klebsiella spp., Proteus mirabilis, and Enterobacter spp.NHPs have been reported to harbour or serve as reservoirs of enteric bacterial pathogens (Okwori et al., 2014;Parmar et al., 2013).E. coli was the most prevalent enteric pathogen followed by K. pneumoniae.This was similar to a study conducted in the United States of America where E. coli was the most common enteric Gram-negative bacilli isolated from clinically healthy NHPs in captivity (Carrier et al., 2009).This may be because E. coli, as part of the normal flora, is one of the most common bacteria in the gastrointestinal tract of mammals.However, there are pathogenic strains of E. coli that are known to cause pathologic conditions, such as diarrhoea, septicaemia and other life-threatening illnesses (Carvalho et al., 2014) AMR is a global health issue with predicted 10 million deaths annually by 2050 (Band et al., 2016;O'Neill, 2016).In Africa, the easy accessibility (over the counter) and low cost of antimicrobial agents have led to its abuse in humans and livestock (Weiss et al., 2018).
Although there is a minimal use of antimicrobial agents in captive NHPs, it has been reported that wild NHPs acquire ARB through close contact with humans (Parsons et al., 2021), livestock and environmental contaminants.In this study, ARB was isolated from NHPs that live in anthropized environments and have high contact with humans.Bacterial isolates from captive NHPs that are more exposed to antibiotics had higher resistance in their antimicrobial susceptibility profile compared to wild NHPs that are less exposed (Table 5).There was an observed trend of phenotypic resistance against Norfloxacin, Ampicillin and Tetracycline among Gram-negative bacteria (mostly Enterobacteria).Ampicillin and Tetracycline are broad-spectrum antibiotics commonly used to manage a wide range of infections.Their widespread indiscriminate use has resulted in acquired resistance among many bacterial pathogens (Chopra & Roberts, 2001;Kaushik et al., 2014).Ampicillin and Tetracycline are classified as 'critically important' and 'highly important' antimicrobials by the World Health Organization (WHO) (WHO Advisory Group on Integrated Surveillance of Antimicrobial Resistance, 2018).This study also recorded that Enterobacter spp.had the highest average phenotypic resistance among Gram-negative bacterial isolates.This is quite alarming because there is currently an emergence of ARB among Enterobacter spp.and other members of the family Enterobacteriaceae in humans (Band et al., 2016).Enterobacter spp. is one of the most common Carbapenemresistant Enterobacteriaceae with multidrug resistance to last resort Carbapenems in humans (Annavajhala et al., 2019).Enterobacter spp.
are clinically important pathogens associated with multidrug-resistant nosocomial infections in humans (Band et al., 2016).
Among Gram-positive bacterial isolates, there was a trend of resistance against penicillin G and oxacillin.This is in tandem with a study in Nigeria that reported resistance against Penicillin antibiotics among bacterial isolates from oral and rectal swabs of captive NHPs (Egbetade et al., 2020).The resistance to Penicillin antibiotics may be due to the inappropriate use of these broad-spectrum antibiotics in humans and in animal husbandry leading to the selection for resistant bacteria (Yevutsey et al., 2017).Penicillin G is classified as 'highly important' haemolyticus was the most occurring of the resistant staphylococcal species.S. haemolyticus is an opportunistic pathogen known to cause nosocomial infections as well as infections in several body systems in humans (Eltwisy et al., 2020).Therefore, the isolation of resistant staphylococcal species in the NHPs is important for public health.
We found a significant association between the body part (oral and rectal) from which the bacteria were isolated from and the antimicrobial susceptibility profile.This was expected and in tandem with studies in humans where AMR was associated with enteric bacterial composition (Covington & Parmer, 2017;Isaac et al., 2017).Exposure to antimicrobials, even at low levels, in the environment leads to the selection for AMR bacteria in microflora-rich environments such as the oral cavity and the gastrointestinal tract (Kim et al., 2020).The oral and rectal microbiota of NHPs have been associated with the transmission of pathogenic AMR bacteria through bites and the faecal-oral route (Abrahamian & Goldstein, 2011;Egbetade et al., 2020;Nguema et al., 2021;Sobreira et al., 2019).
We note here that the culture-based method of identifying bacteria limited the isolation of bacteria to only isolates that could grow aerobically on artificial culture media which underestimated the total diversity and richness of the bacterial community.Future studies should, therefore, consider a combination of culture-based methods and molecular methods such as metagenomic sequencing to circumvent the limitations imposed by culture-dependent methods on the diversity of the bacterial community of NHPs.

CONCLUSION
Despite some limitations, the results of our study revealed that apparently healthy NHPs in Ghana harbour zoonotic pathogenic bacteria.
It was also discovered that NHP-associated bacteria were resistant to commonly used antibiotics in humans and livestock.This study demonstrates the necessity for surveillance systems for zoonotic pathogens and AMR among NHPs within the One Health concept.This is crucial in the prevention of the dissemination of zoonotic and antimicrobial-resistant bacteria among humans and NHPs.

F
I G U R E 1 A map showing the sample locations generated with ArcGIS version 3.8.2.Papio anubis were the only non-human primate (NHP) species sampled in Mole National Park in the Savannah Region; Cercopithecus mona and Erythrocebus patas were the NHP species sampled in the Ashanti Region in Kumasi Zoo.All four NHP species were found and sampled in Tema and its environs in the Greater Accra Region.
Polystyrene tubes (12 mm × 75 mm) were filled with 3 mL 0.45% sterile saline.A volume of 145 and 280 µL of McFarland turbidity standard of homogenized bacteria suspension in 3 mL 0.45% sterile saline was pipetted into the polystyrene tubes for Gram-positive and Gram-negative ASTs, respectively.AST results were obtained after the Vitek system had analysed the minimum inhibitory concentration of the bacterial species against the prescribed antibiotics of the Vitek AST cards (AST-N and AST-P) (BioMérieux).The antibiotics included in the AST-N cards were Cefuroxime, Ampicillin, Ciprofloxacin, Tetracycline, Trimethoprim-Sulfamethoxazole, Amoxicillin/clavulanic acid, Gentamicin, Meropenem, Ceftazidime, Norfloxacin and Nitrofurantoin.The antibiotics included in the AST-P cards were Vancomycin, Clindamycin, Ciprofloxacin, Tetracycline, Trimethoprim-Sulfamethoxazole, Erythromycin, Gentamicin, penicillin G and oxacillin.
Staphylococcus spp.were the most prevalent bacteria isolated from the oral swabs (Staphylococcus haemolyticus: n = 7, 14%; Staphylococcus gallinarum: n = 6, 12%; Staphylococcus sciuri: n = 6, 12%).E. coli (n = 32, 36%) and Klebsiella pneumoniae ssp.pneumoniae (n = 16, 18%) were the most occurring bacteria isolated from rectal swabs (Table Enterobacter spp.and Staphylococcus spp., were isolated from both oral and rectal swabs.Enterobacter spp.had the highest average phenotypic resistance to antibiotics used for AST among Gram-negatives, whereas Staphylococcus spp.had the highest average phenotypic resistance to antibiotics used for AST among Gram-positives.The bacterial isolates from the oral and rectal swabs were resistant to antibiotics, such as Norfloxacin, Ampicillin, Tetracycline, penicillin G and oxacillin.In this study, Staphylococcus spp.were the dominant isolates from oral swabs, specifically, S. haemolyticus.Staphylococcus spp.have been F I G U R E 3 Proportion of bacterial species (a) based on bacterial proportion per non-human primate (NHP): Escherichia coli was the most abundant species in Erythrocebus patas, whereas Staphylococcus spp.were the dominant isolates in both Papio anubis and Cercopithecus mona.Klebsiella spp.were the dominant species in Chlorocebus sabaeus; (b) based on anatomical origin: E. patas harboured the highest proportion of rectal bacterial isolates, whereas both E. patas and P. anubis had the highest oral bacterial isolates; (c) based on origin: Enterobacter spp., Klebsiella spp.and Rothia spp.were absent in isolates from wild NHPs.
under the WHO antimicrobial classification (WHO Advisory Group on Integrated Surveillance of Antimicrobial Resistance, 2018).Resistance to penicillin G is therefore significant for public health.Staphylococcus spp.recorded the highest average phenotypic resistance to antibiotics used for AST among Gram-positive bacterial isolates.In this study, S.

TA B L E 1 Detailed information about sampled non-human primates. NHP species No. (%) No. (%) male No. (%) female No. (%) adult No. (%) sub-adult No. (%) juvenile
Chi-square test was used to determine whether there was statistical difference between antimicrobial susceptibility profile and the body part (oral cavity and rectal region) from which bacteria were isolated, as well as the sex and origin (wild and captive) of NHPs.Any result with a p-value ≤0.05 was considered statistically significant.
on the captured NHPs and characterized bacteria with AST were entered into Microsoft Office Excel 2019.Descriptive analyses were performed to generate the proportion of bacteria populations, and the frequency and distribution of Gram-positive, Gram-negative, resistant or sensitive bacteria were analysed with Microsoft Office Excel 2019 and Graph Pad Prism (Version 8).Figures were generated with Adobe Illustrator 2022.
Chi-square test assessing association between antimicrobial susceptibility profile and body part bacteria were isolated from, origin of non-human primate and sex of non-human primate.Observed results of sensitivity of isolated bacteria to antibiotics tested.b Observed results of resistance of isolated bacteria to antibiotics tested.c Expected results of sensitivity of isolated bacteria to antibiotics tested.d Expected results of resistance of isolated bacteria to antibiotics tested.
. Pathogenic strains of E. coli such as enteropathogenic E. coli (EPEC) have been reported in both human TA B L E 5 a