Bats Are Carriers of Antimicrobial-Resistant Staphylococcaceae in Their Skin

Bats have emerged as potential carriers of zoonotic viruses and bacteria, including antimicrobial-resistant bacteria. Staphylococcaceae has been isolated from their gut and nasopharynx, but there is little information about Staphylococcaceae on bat skin. Therefore, this study aimed to decipher the Staphylococci species in bat skin and their antimicrobial susceptibility profile. One hundred and forty-seven skin swabs were collected from bats during the spring and summer of 2021 and 2022. Bats were captured in different areas of the Metropolitan Region of São Paulo, Brazil, according to the degree of anthropization: Area 1 (Forested), Area 2 (Rural), Area 3 (Residential-A), Area 4 (Slum-– up to two floors), Area 5 (Residential-B—condo buildings), and Area 6 (Industrial). Swabs were kept in peptone water broth at 37 °C for 12 h when bacterial growth was streaked in Mannitol salt agar and incubated at 37 °C for 24 h. The disc-diffusion test evaluated antimicrobial susceptibility. Staphylococcaceae were isolated from 42.8% of bats, mostly from young, from the rural area, and during summer. M. sciuri was the most frequent species; S. aureus was also isolated. About 95% of isolates were resistant to at least one drug, and most strains were penicillin resistant. Eight isolates were methicillin resistant, and the mecA gene was detected in one isolate (S. haemolyticus). Antimicrobial resistance is a One Health issue that is not evaluated enough in bats. The results indicate that bats are carriers of clinically meaningful S. aureus and antimicrobial-resistant bacteria. Finally, the results suggest that we should intensify action plans to control the spread of resistant bacteria.

The Staphylococcaceae is a family of Gram-positive bacteria, including the significant Staphylococcus genus. Staphylococcus spp. is a Gram-positive cocci genus whose species may be members of the normal gut and skin microbiota. Some Staphylococcus (e.g., S. epidermidis) play an essential role as the first defense against pathogenic bacteria in a mutualistic relationship with the host [12]. However, few species are either opportunistic or pathogenic/zoonotic. For instance, S. aureus colonizes humans and other animals, and most of them show no symptoms. However, S. aureus can become pathogenic after breaking physical and immunological barriers, leading to mild to severe infections [13,14].  (Table 1). The 63 isolates belonged to 10 species of Staphylococcaceae. M. sciuri was the most frequent species, isolated in 66.7% (06/09) of bat species (Table 1), and the Staphylococcus isolated were S. aureus, S. saprophyticus, S. warneri, S. xylosus, S. kloosii, S. epidermidis, S. haemolyticus, and S. nepalensis (Figure 1). The S. aureus belonged to spa type t645 (N = 05) or t1451 (N = 02).

Number of Bats Captured and Sampled
Finally, 33 isolates were Gram-negative bacteria whose results will not be described here. No growth was observed in 51 samples.

Diversity Analysis
The diversity was numerically higher in Artibeus lituratus, followed by Glossophaga soricina, Artibeus fimbriatus, Platyrrhinus lineatus, and Sturnira lilium (Figure 2A), but the Shannon index analysis revealed no differences between bat species and Staphylococcaceae. (P = 0.43). The same analysis was conducted for Staphylococcaceae and zone and showed no differences (P = 0.41) ( Figure 2B).

Diversity Analysis
The diversity was numerically higher in Artibeus lituratus, followed by Glossophaga soricina, Artibeus fimbriatus, Platyrrhinus lineatus, and Sturnira lilium (Figure 2A), but the Shannon index analysis revealed no differences between bat species and Staphylococcaceae. (P = 0.43). The same analysis was conducted for Staphylococcaceae and zone and showed no differences (P = 0.41) ( Figure 2B).

Discussion
Wildlife animals are relevant carriers of antimicrobial-resistant bacteria worldwide [33]. Bats belong to a particular group of species that may be potential carriers of resistant bacteria due to their synanthropization and contact with contaminated environments and other species. Their potential to fly to different regions makes them potential disseminators of these bacteria.
In this present research, antimicrobial-resistant Staphylococcaceae were isolated from the bats' skin, which makes them important from the One Health perspective. These bacteria can easily be transferred among individuals in the same colony, to other species or even through occupational transmission. Professionals working with bats are at risk since vampire bat culling is recommended to control cattle rabies. Although bats are handled with proper gloves, this equipment may act as fomites and transfer bacteria to professionals. Feeding and roost habits may favor the spread of resistant microorganisms. For example, the vampire bat (Desmodus rotundus) feeds exclusively on blood, particularly from domestic animals, although they can feed on wildlife and humans if needed [34]. An antimicrobialresistant bacteria on the skin may be transferred to the food source by contact, which can be spread to other animals, humans, or the environment through their daily activities. Bat species, including Platyrrhinus spp., Artibeus spp., Pygoderma spp., and Sturnira spp., prefer roosting in trees where several other individuals are. This particular habit favors the microorganism exchange through close contact. The same situation involves the transfer of these bacteria to the trees, which may play a role as carriers and transferors to humans and other animals, mainly when roosts are found in the cities [35].
Wild small mammals (mice, voles, and shrews) were recently characterized as sentinels for the environmental transmission of antimicrobial resistance [36] by detecting more resistant E. coli strains in coastal animals than inland ones, in addition to human-associated pathogenic strains. The authors also highlighted concerns about the overall presence of antimicrobial resistance regarding the interaction between animals, the environment, and humans. In this regard, most bat species are long-lived and gregarious, and migration is present in some species [37]. Facing these habits and the consequent close contact with other animals and the environment, bats may also play a role as sentinels for antimicrobial resistance.
Bats were mostly captured in rural and forested areas, which may explain why most Staphylococcaceae isolates were obtained there. However, both sites present low anthropization, which increases the sources of Staphylococcaceae. The genus Staphylococcus is the most clinically relevant and comprises ubiquitous species that can be found in animals [38], soil [39], water [40], and air [41].
The seasonal activity of bats is usually higher in summer because it is the period with the increased food supply; therefore, it is the right time for pregnant and lactating females to forage [42,43]. Thus, higher activity implies increased contact with other animals and plants, which may explain why most Staphylococcaceae isolates were obtained in summer.
The Staphylococcus genus comprises 53 recognized pathogenic and commensal species [44] that can be isolated from animals and the environment. M. sciuri was the most frequent species isolated in this research, which agrees with the fact that M. sciuri is mainly found in animals [45]. However, its importance in human medicine is increasing due to its association with wound infections [46], peritonitis [47], and urinary tract infections [48]. S. aureus is a significant pathogen that causes several infections in animals and humans, and it is worth mentioning that these clinically significant species were also isolated from bats.
It is worth mentioning that Staphylococcaceae was mainly isolated from young bats, which may be related to the colonization in females. Similar behavior is observed in bat flies as females usually have a high degree of infestation and transfer the parasites to their offspring [49].
Regarding the genus Staphylococcus, few studies have evaluated their colonization and diversity in bat skin. Swab samples were taken from the wing membranes of healthy Livingstone's fruit bats (Pteropus livingstonii) from the Jersey Zoo, United Kingdom [32], and the species were similar to those isolated in the present study. Still, in terms of frequency, P. livingstonii mainly were colonized by S. xylosus, followed by S. saprophyticus, S. nepalensis, and S. aureus, differing from the present study [32]. S. aureus was also isolated from the skin wounds of bats in Berlin [31].
Antimicrobial resistance is a global concern that may cause about 10 million deaths by 2050 [21]. Therefore, studies on the frequency of antimicrobial-resistant bacteria are encouraged not only in humans but in the environment and animals. In addition to deciphering the Staphylococcaceae species in bat skin, antimicrobial susceptibility studies are essential for determining their role as carriers of resistant bacteria.
In the present study, most isolates were resistant to penicillin, a beta-lactam agent extensively used in methicillin-susceptible staphylococcal infections [50]. Since penicillin was massively used after its discovery, Staphylococcaceae, particularly the genus Staphylococcus and the M. sciuri, can become resistant by two central mechanisms: the primary system is the production of a penicillinase encoded by the blaZ gene [51], and the second is the production of a penicillin-binding protein (PBP2), which has a low affinity for beta-lactams, encoded by mecA and mecC genes [51].
In research conducted in Chile, Sacristán et al. (2020) [55] collected fecal samples of wild guignas (Leopardus guigna) and observed an association between anthropization degree and antimicrobial resistance genes' prevalence in wildlife. In the present study, however, the bacteria isolated from bat skin was not a good indicator of wildlife and/or environmental exposure to human activity ("zone"; P > 0.05).
The skin Staphylocociccaceae of the bats sampled here did not exhibit a multidrugresistance profile, which differs from feces studies [24]. Fecal samples usually show more diverse microbiota than the skin, mainly due to the ingestion of microbes from food and water. This ingestion reflects in the gut colonization with resistant bacteria that can transfer mobile genetic elements containing resistance genes to non-resistant bacteria. In addition, food and water may have antibiotic residues, increasing the selective pressure in the gut. Therefore, we hypothesize that sample type may impact antimicrobial resistance studies. A cross-sectional study of dairy cattle metagenomes of heavy metal-contaminated and non-contaminated areas was conducted by Gaeta et al. (2020) [56]. The resistome of rumen liquid, deep nasopharyngeal swabs, and fecal samples were compared, and the results showed that fecal samples had the highest frequency of antimicrobial resistance genes. Finally, the possibly less selective pressure on the skin microbiota (compared with other anatomical sites, such as the gut) may also explain the absence of multidrug-resistant bacteria in the captured bats.
Antimicrobial resistance is a One Health issue [22] studied in humans, pets, and farm animals but not enough in wild animals, such as bats. Brazil was the third largest consumer of antimicrobials in livestock in 2010, and it is also projected to be the second country to increase antimicrobial consumption between 2010 and 2030 [57]. These antimicrobials may be dumped in water courses, soil, and plants and reach bats, and this behavior may favor the spread of antimicrobial-resistant bacteria. In addition, bacteria from bats could acquire resistance genes from isolates of humans and other animals through their intense microbial exchange [58].
Although no multidrug-resistant bacteria were detected, the presence of isolates resistant to clinically important drugs warns the need to conduct epidemiologic studies to understand the role of bats as carriers of these bacteria.

Bat Capture and Sample Collection
In previous more extensive research, captures were performed from June 2021-April 2022, every three months (one capture in each season of the year), in the optimal conditions for capture, such as new moon, less light, and no rain (when possible). For this study, samples were obtained during summer and autumn ("moment of collection" variable).
Bats were captured in six different areas, according to an increasing order of anthropization ("zone" variable): Area 1 (Forested): a private reserve of environmental preservation of 10,000 acres composed of a dense mountain rainforest and altitude of around 900 m; Area 2 (Rural): a 100 hectares portion of secondary ombrophilous forest with the presence of exotic trees and ornamental species and an average altitude of 800 m; Area 3 (Residential-A): space with several fruit trees, including papaya, banana, an avocado, and a fig; Area 4 (Slum-up to two floors): a small fragment of secondary Atlantic forest, with several Piper sp. and eucalyptus trees near to the area; Area 5 (Residential-B-condo buildings): a 1800 m 2 condominium building and garage with a vegetable garden and orchard located next to a highway, with several fruit trees including avocados, bananas and mulberries; Area 6 (Industrial): area from the University of São Paulo close to a nearby industrial area, with some non-fruit trees.
Bats were captured using mist nets, which were verified every 30 min. The networks were opened at dusk and closed after 4-5 h [59]. Animals were kept in cloth bags and handled calmly, one by one, by the trained participants.
Morphometric measurements were estimated for each bat, such as forearm length, sex, age estimate, and weighing using cloth bags and precision spring scales. Then, sterile swabs were rubbed on 147 bats' wings and backs and immediately stored in ice. Samples were transported to the laboratory, placed in tubes with Peptone Water Buffered (Difco, BD) and incubated at 35 • C ± 2 • C for 12 h.

Microbiologic Tests and Antimicrobial Susceptibility
Ten microliters of each broth with bacterial growth were streaked in Salt Mannitol Agar plates (Difco, BD) following incubation at 35 ± 2 • C for 24 h. Then, bacterial strains were selected and identified using matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS).
The antimicrobial susceptibility profile was determined by the disc-diffusion test (Kirby-Bauer method) according to the Clinical Laboratory Standard Institute [60]. Each isolate was streaked in Mueller Hinton agar plates (Difco, BD) and incubated at 35 ± 2 • C for 18 h. Then three to five colonies from the pure cultures were added to a sterile saline buffer, and the suspension was adjusted to the 0.5 McFarland turbidity scale. In three directions, the suspension was streaked onto a Muller Hinton agar plate (Difco, BD). After distributing all antibiotic discs, the plates were incubated at 36 ± 1 • C for 18-24 h. The diameter of inhibition zones around the discs was measured, and the strains were classified as sensitive, resistant or intermediate against each antimicrobial tested according to the CLSI M100 guidelines [60]. The following drugs with respective concentrations (DME, Brazil) were tested: sulfamethoxazole-trimethoprim (SUT 25), gentamycin (GEN 10), chloramphenicol (CLO 30), tetracycline (TET 30), cefoxitin (CFO 30), penicillin (PEN 10), erythromycin (ERI 15), and ciprofloxacin (CIP 05). The Staphylococcus aureus ATCC 25923 was used as a negative (sensitive) control. A laboratory strain (Staphylococcus epidermidis strain USP-LZB-G06) [61] was used as a positive control for resistance to CFO, PEN, CIP, SUT, and ERI.

DNA Extraction
DNA was obtained using the protocol described by Fan et al. (1995) [62]. Pure colonies were added to 50 microliters of phosphate buffer saline 1X (2.6 mM NaH2PO4, 7.4 mM NaHPO 4 , 10 mM NaCl, pH 7.2), boiled at 100 • C for 10 min, and put on ice for five minutes. At the end of the previous step, microtubes were centrifuged at 14,000× g rpm for 10 min, and the supernatant was stored in new sterile microtubes at −20 • C until processing.

Spa Typing
Spa Typing analysis was performed for all S. aureus isolates. The protein A region was amplified by PCR, according to Harmsen et al. (2003) [63]. Briefly, the polymerase chain reaction (PCR) was performed in a total volume of 25 µL consisting of: 12.5 µL of 2× GoTaq Green Master Mix (Promega), 10 µM of each primer, and 9 µL of nuclease-free water. The amplification cycle was the first denaturation at 95 • C for five minutes, 35 cycles of denaturation at 92 • C for 45 s, annealing at 60 • C for 45 s, extension at 72 • C for 90 s, and a final extension step at 72 • C for 10 min. Ten microliters of each amplicon were purified using four microliters of ExoProStar (Sigma-Aldrich, St. Louis, MO, USA) followed by incubation (37 • C for 15 min and 80 • C for 15 min, respectively). Then, a mixture containing one microliter of BigDye (Thermo Fisher™), 0.5 µL of each primer, and 1.5 µL of buffer 5× (included in the BigDye Terminator v3.1 cycle sequencing kit). Three microliters of this mixture were distributed in a 96-deep well plate, followed by seven microliters of the purified DNA. After centrifugation at 1000 rcf for 60 s, the plate was incubated in a thermocycler: one cycle at 96 • C for 60 s, 60 cycles of 96 • C for 10 s, 50 • C for five seconds and 60 • C for four minutes. Then, 2.5 µL of EDTA 125 mM and 30 µL of 100% ethanol were added to each well for DNA precipitation. After drying at room temperature, the plate was centrifuged at 2250 rcf for 30 min at 15 • C, and 30 µL of 70% ethanol was added to each well. The plate was centrifuged at 1650 rcf for 15 min at 15 • C. Then, the plate was incubated at 95 • C for 10 min with the lid up. Finally, 10 µL of Hi-Di™ was added to each well, and the plate was placed in the Eppendorf sequencer for sequencing by the Sanger method. The obtained files were uploaded to Bionumerics (bioMérieux, France) for complete characterization of the Spa Type.

Amplification of Resistance Genes
A polymerase chain reaction was performed to detect the blaZ, mecA and mecC genes. The blaZ gene was amplified according to Martineau et al. (2000) [64]. Briefly, the PCR was performed in a total volume of 25 µL consisting of: 12.5 µL of 2× GoTaq Green Master Mix (Promega), 75 µM of each primer, and 9.5 µL of nuclease-free water. The amplification cycle was the first denaturation at 94 • C for five minutes, 30 cycles of denaturation at 94 • C for 20 s, annealing at 60 • C for 30 s, and extension at 72 • C for 90 s, and a final extension step at 72 • C for five minutes. The mecA gene amplification was performed according to Mehrotra et al. (2000) [65]. The PCR was performed in a total volume of 15 µL consisting of 7 µL of 2× GoTaq Green Master Mix (Promega), 10 µM of each primer, and 5.4 µL of nuclease-free water. The amplification cycle was the first denaturation at 94 • C for five minutes, 34 cycles of denaturation at 94 • C for two minutes, annealing at 52 • C for two minutes, extension at 72 • C for 60 s, and a final extension step at 72 • C for seven minutes. Finally, the mecC gene was amplified according to Paterson et al. (2012) [66]. The PCR was performed in a total volume of 15 µL consisting of 7 µL of 2× GoTaq Green Master Mix (Promega), 10 µM of each primer, and 5.4 µL of nuclease-free water. The amplification cycle was the first denaturation at 94 • C for five minutes, 34 cycles of denaturation at 94 • C for 30 s, annealing at 52 • C for 30 s, extension at 72 • C for 60 s, and a final extension step at 72 • C for 10 min. The amplicons were observed by ultraviolet light after 40-min gel electrophoresis for 35 min. Laboratory isolates were used as positive controls.

Data Analysis
Results were shown as absolute and relative frequencies. Comparisons between the number of isolates and variables were performed using Pearson's Chi-Square test with continuity correction, except for "collection area", when the Kruskal-Wallis test was used. The association between "resistance", "collection areas", and "sampling moment" was evaluated using the Fisher exact test. Diversity analysis was performed using the "vegan" package from RStudio [67], and diversity indexes were compared using the Kruskal-Wallis test. Figures were elaborated using the "ggplot2" package [68] from RStudio [67] and R v.4.0.5. Variables were considered significant when P < 0.05.  Institutional Review Board Statement: The study was conducted according to the guidelines of the Ethics Committee on the Use of Animals in Experimentation, FMVZ-USP (protocol code #2332030320; date of approval 3/3/2022), and Biodiversity Authorization and Information System (protocol #71136-2).

Informed Consent Statement: Not applicable.
Data Availability Statement: All data are available upon request.