Bacteria and parasites in Podarcis sicula and P. sicula klemmerii

Background New epidemiological data on bacterial and parasitic infections in 24 Italian wall lizards, namely Podarcis sicula (mainland population) and P. sicula klemmerii (insular population) in southern Italy were provided. To achieve this goal, samples were collected from individuals belonging to the two populations and analysed by microbiological and parasitological methods. Results A wide range of bacteria (e.g. Pantoea spp., Citrobacter spp., Morganella spp., Pseudomonas, Enterobacter spp., Staphylococcus spp. and Escherichia coli) and parasites (e.g. Ophionyssus natricis, coccidia, Dicrocoelidae) were detected in both P. sicula and P. sicula klemmerii individuals. Insular population presented similar bacterial and parasitic diversity to its mainland counterpart. Ampicillin was the antimicrobial with the highest resistance rate. Conclusion This study highlighted various bacteria and parasites, some of them potentially zoonotic. Further studies are needed to better understand the epidemiology and transmission routes of these pathogens along with their impact on the welfare and behaviour of Italian wall lizards.


Background
The Italian wall lizard, Podarcis sicula, is one of the most common lizards in Italy [1]. The snout-vent length (SVL) is, on average, 15 to 25 cm long. The colour pattern is characterized by a green or brown back and whitish belly, although melanic variants, with either shorter, or longer SVL, are known to occur in several islands in the Mediterranrean sea. As a common, and easily managed study model, the Italian wall lizard was subjected to several studies, regarding aspects as disparate as phenotypic response to predation [2], feeding behaviour [3], ontogeny [4], adaptation to novel environments [5], biogeography [6,7], and its role as a biological indicator [8].
Podarcis sicula klemmeri [9] is one of the subspecies belonging to P. sicula. It is confined to a small, 1 km 2 large islet, Licosa, off the western coast of Italy. As with many other insular populations of the Italian wall lizard, P.s. klemmeri is melanic, meaning the back appears tinged with blue, and the pale underside is bluish as well, rather than the usual white of continental populations.
Melanic variants have been investigated for the so-called Island syndrome [10], which is a suite of phenotypic character shifts in insular populations including, besides melanism, changed body size, feeding behaviour and ecology, patterns of aggressiveness, and life history traits [11][12][13][14][15]. The link between melanism and characters shifts on islands was found to be in the activity levels of melano-cortin receptors, MCRs [13]. Melanocortins form a suite of five receptors activated pleyotropically by a single DNA locus, the proopiomelanocortin POMC gene, that happen to regulate feeding and sexual activity, immunocompetence and body colour [16][17][18]. Interestingly, Monti et al. [19] tested whether P. s. klemmeri individuals have different ectoparasite loads (i.e. the density of ticks and mites on the skin) as compared to the continental individuals. They found reduced load in insular individuals, consistently with their comparatively higher α-melanocyte-stimulating hormone (MSH) levels.
Herein, we deepen microbiological and parasitological investigations on P. s. klemmeri. Besides a few studies [20,21], these aspects have not been scrutinized so far, yet remain very important in the case of melanic insular lizards, whose immune system is expected to be depressed by great investment in reproduction, and life at high density [13,19,22]. This study was undertaken with the aim to evaluate the presence of potentially zoonotic bacteria and parasites in wild-caught insular individuals of P. sicula klemmeri, along with mainland individuals of P. sicula.

Results
A wide range of bacteria and parasites were detected in both P. sicula and P. sicula klemmerii individuals. Among the 24 analysed animals, 23 (95.9%) were positive to at least one bacterium and 19 (79.1%) were positive to at least one parasite.
With respect to antimicrobial susceptibility testing, Citrobacter spp. showed the highest resistance profile. Specifically, two out of ten strains of Citrobacter spp. were resistant to three or more drugs ("multidrug-resistant") of which, one strain was resistant to ampicillin, doxycycline and streptomycin and the other one was resistant to ampicillin, amoxicillin-clavulanic acid, doxycycline and nitrofurantoin. The remaining strains were resistant to ampicillin, one was also resistant to doxycycline and one other was also resistant to amoxicillin-clavulanic acid. Escherichia coli O145 and all the strains of Enterobacter spp. were resistant to ampicillin. In contrast, Pseudomonas aeruginosa were susceptible to all antimicrobials tested.

Discussion
Our results provide new epidemiological data on bacterial and parasitic infections in P. sicula and P. sicula klemmerii. These species were poorly investigated from a sanitary perspective so far. Bacterial and parasitic infections in reptiles have recently gained scientific relevance. However, the majority of studies were carried out on captive-bred individuals of Cryptodira (tortoises, [23]), Serpentes and Sauria (snakes and other 'lizards' , [24]). Conversely, data on infections in wild lizards are scarce.
The results of our study showed the presence of endoparasites (coccidia, pinworms and liver flukes) and ectoparasites (O. natricis) in P. sicula and P. sicula klemmeri. With respect to the presence of pinworms, our findings are in line with those by Casanova et al. [20] who detected the presence of a nematode belonging to the Oxyuroidea superfamily in P. sicula. Nevertheless, our results showed the first epidemiological information on parasites infecting P. sicula klemmeri. Regarding the presence of O. natricis, there are no studies in literature that highlight its presence in P. sicula and P. sicula klemmeri, although the infestation by this mite has been described in various genus of Sauria as Lacerta, Podarcis and Darevskia. In addition, the carrier role of O. natricis in the transmission of a blood parasite belonging to the genus Karyolysus has been reported in lizards [25]. An interesting finding of our study was the detection at necropsy of liver flukes (Dicrocoelidae) in the continental lizards, although species identification was not performed.
Bacteriological results also added new epidemiological data in P. sicula due to detection of some potential zoonotic species as P. aeruginosa and E. coli O145. The bacterial isolation performed on oral swabs gave as result the presence of bacteria belonging to the genera Pantoea, Pseudomonas, Morganella, Staphylococcus, Citrobacter, Shewanella never detected before in P. sicula and P. sicula klemmeri. However, these bacteria are frequent in reptiles, along with other bacteria such as Enterobacter spp., E. coli and Providencia spp. [26][27][28]. Bacteria isolated from cloacal swabs were Enterobacter spp., Citrobacter spp., Morganella spp., E. coli, Providencia spp. Shewanella spp., and Pseudomonas spp., gram-negative bacteria frequently isolated in other reptiles [29] as well as Staphylococcus spp. The isolation of E. coli O145 in one P. sicula klemmeri is noteworthy due to the potential zoonotic role of this serogroup which is considered a shigatoxin-producing E. coli.
It is interesting to notice that the insular population presents similar bacterial diversity of its mainland counterpart, although the differences in sample size urge caution in interpreting these results. Noteworthy, 9 different bacterial genera were identified in P.s. klemmeri (up to six in a single individual), against 4 in mainland individuals (up to three in a single individual). A weakness of this study was the failure to isolate bacteria in the oral cavity of some lizards due to the difficult to keep viable some strains during the isolation procedures. To our knowledge this is the first study to assess the antimicrobial resistance profiles of potentially zoonotic bacteria carried by Podarcis spp. It is difficult to speculate regarding the results of the antimicrobial resistance recovered in the present study. However, one possible mechanism by which lizards acquire antimicrobial resistant bacteria in their environment range may be directly The figure for parasites is even harder to interpret given the smaller sample size. Yet, we found coccidia, liver flukes and pinworms in the insular populations, and pinworms only (in just one individual) within continental lizards. While difference in collecting season, sample size, and the instance of necropsy in two individuals only suggest great caution, data seem to indicate an overall higher parasite/bacterial load in insular lizards. Assuming this is true, it remains to be elucidated whether this depends on population density on Licosa, or on less competent immune system in insular melanic individuals [12,13].

Conclusion
In conclusion, the results of this study highlighted various bacteria and parasites, some of them pathogenic, able to infect the species P. sicula and the subspecies P. sicula klemmeri. Further studies are needed to better understand the epidemiology and transmission routes of these pathogens along with their impact on the welfare and behaviour of Italian wall lizards.

Study animals
From November 2015 to May 2016, we collected and examined, within 24 h from capture, a total of 24 individuals, 20 (14 males and 10 females) belonging to the subspecies P. sicula klemmeri, and 4 individuals belonging to the species P. sicula. The P. sicula klemmeri specimens were from the small islet of Licosa, whereas the P. sicula specimens from Punta Licosa mainland (Campania region of southern Italy). The study site, collection and housing methods were described in Raia et al. [12,13]. Lizards captured on Licosa islet were identified with the code "LIC" and a progressive number (from LIC 1001 to 1020); the lizards from Licosa mainland were identified with the code "PLIC" and a progressive number (from PLIC 1001 to 1004).
In order to perform the microbiological analysis, oral and cloacal samples were individually collected by sterile cotton-tipped swabs and then inoculated in Phosphate Buffered Saline (PBS). For the parasitological analyses, different methods were used: faecal samples were collected from each lizard, scotch tape test was performed on the skin of each animal and, eventually, a necropsy with stereo microscope was carried out on two lizards found dead on the insular study site. After sampling, animals were released in the same area where they were captured. All experiments described were performed in accordance with the local and national guidelines governing animal experiments (86/ 609/CEE and its modifications).

Microbiological and parasitological analyses
Oral and cloacal swabs stored in PBS were inoculated in buffered peptone water (BPW), Campylobacter-selective enrichment broth (CSEB), cooked meat medium (CMM), modified tryptone soy broth (MTSB). The samples inoculated in BPW were incubated at both 25°C and 37°C for 24 h and then inoculated/streaked into Rappaport-Vassiliadis broth (RV), Columbia blood agar base (CBA; Oxoid, Milan, Italy), Pseudomonas cetrimide agar (PCA; Oxoid), MacConkey agar (MCA; Oxoid) and Baird-Parker agar (BPA; Oxoid). The samples inoculated in MTSB were incubated at both 25°C and 37°C for 24 h and then spread on sorbitol MacConkey agar (SMCA; Oxoid). The samples inoculated in CSEB were incubated in microaerophilic atmosphere at both 25°C and 42°C for 48 h and then streaked on Campylobacter blood-free selective agar (CBFA; Oxoid). Instead, the samples inoculated in CMM were incubated in anaerobic atmosphere at both 25°C and 37°C for 24 h, and then streaked on anaerobe basal agar (ABA; Oxoid). The remaining PBS were incubated at 4°C for 14 days, spread on Yersinia selective agar base (cefsulodin-irgasan-novobiocin, CIN Agar; Oxoid) and incubated at 30°C for additional 24-48 h. CBA, PCA, MCA, SMCA, CEOA and BPA were incubated at both 25°C and 37°C for 24-48 h, whereas RV was incubated at both 25°C and 42°C for 24-48 h and then streaked on xylose lysine desoxycholate agar (XLD) and brilliant green agar (BGA); CBFA were incubated in microaerophilic atmosphere at both 25°C and 42°C for 24-48 h, ABA plates were anaerobically incubated both at 25°C and 37°C for 48 h and checked daily for a further week before discarding. All the isolates were previously identified according to their morphological features, their growing need, Gram colouring, motility and pigments production test and through conventional biochemical and phenotypic test. Progressively, the isolates were identified through the biochemical systems API 20 E, API 20 NE (bioMerieux, Mercy-l'Etoile, France) and RapID ANA II, RapID NF PLUS, RapID STAPH PLUS (Oxoid). Escherichia coli strains were serogrouped with antisera poly-and monospecific (Sifin) whereas, suspected Campylobacter strains were identified by PCR as reported by Gargiulo et al. [30].
In order to detect and count parasitic elements (eggs, larvae, cysts, oocysts), each faecal sample collected from the 24 lizards was analysed with the FLOTAC pellet technique [24,31]. The analytic sensitivity of the FLO-TAC pellet technique varied on the basis of the weight of each pellet μμ [24] and expressed as eggs/oocysts per gram (EPG/OPG) of faeces. Two flotation solutions were used: sodium chloride (NaCl, specific gravity = 1200) and zinc sulfate (ZnSO4, specific gravity = 1350) to detect protozoa/nematoda and trematoda, respectively. The scotch tape test was performed on the skin of the lizards, specifically, at the level of skin folds, around the eye, the cloaca and the eardrum recesses. The necropsy of the two lizard carcasses was performed under a stereomicroscope using the reptile necropsy protocol reported in Jacobson [32].

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Availability of data and materials
The datasets used and analyzed are available from the corresponding author upon reasonable request.