Diversity and Composition of Gastrointestinal Helminths and Gram-Negative Bacteria of the West African Rainbow Lizard ( Agama picticauda) in a Human-Modified Landscape: Implications for Conservation and Zoonosis

Emerging infectious diseases are negatively impacting herpetofaunal populations globally, posing a major conservation threat. Anthropogenic perturbations of natural habitats may infuence wildlife disease dynamics and spillover events. Although viral and fungal diseases of vertebrates have received considerable studies, there is scant information on the enteric bacteria and helminths of herpetofauna. Here, we assessed the gastrointestinal helminths and Gram-negative bacteria of the West African rainbow lizard ( Agama picticauda ) in a human-dominated landscape in Accra, Ghana. We used the fotation method and standard cultural, morphological, and biochemical methods, for helminths and bacteria, respectively. Four helminth species, Ascaris spp., Enterobius spp., Pharyngodon spp., and Oxyurid spp., were identifed, with overall high (71.9%) prevalence. Ascaris spp. ( χ 2 =33.33, p < 0 . 001) and Pharyngodon spp. ( χ 2 =14.5, p < 0 . 001) were more prevalent in suburban lizards during the wet season than dry season. Te prevalence of Ascaris spp. and Enterobius spp. was signifcantly higher ( χ 2 =12.34, p < 0 . 001) in urban than suburban lizards, while Oxyurid spp. was more prevalent ( χ 2 =4.75, p � 0 . 03) in the suburban lizards during the wet season. Te gut bacteria identifed (prevalence) were Escherichia spp. (86.5%), Enterobacter spp. (75.7%), Proteus spp. (67.6%), Shigella spp. (40.5%), Klebsiella spp. (21.6%), and Salmonella spp. (16.2%). Te prevalence of Escherichia spp. and Proteus spp. was signifcantly higher ( χ 2 =5.77–7.34, p � 0 . 01–0 . 02) in suburban lizards than urban lizards during the wet season. Some Ascaris spp., Enterobius spp., Escherichia spp., Enterobacter spp., Shigella spp., and Salmonella spp. are found in humans. Our data can serve as baseline information for long-term monitoring studies of reptile-parasites-environment interactions in the Accra plains of Ghana. Further studies using molecular techniques are needed to ascertain the zoonotic helminths and bacteria of A. picticauda .


. Introduction
Emerging infectious diseases are on the rise and impacting negatively wildlife, particularly, herpetofaunal (reptiles and amphibians) populations globally. Tis poses a major conservation challenge, particularly in developing countries where natural habitats are being destroyed at alarming rates [1,2]. Parasites infuence host behaviour, resource utilization, body condition, population dynamics, and ftness [3,4]. Infectious diseases result from the interactions of host and pathogens, and the interactions involve physiological and ecological mechanisms that have evolved over several years under a given set of environmental conditions [5]. Terefore, land use and other environmental changes that alter the structural and ecological conditions of the environment will modulate host-pathogen interactions, leading to altered disease risks [6]. the biotic and abiotic components of the environment [7], changing species richness, distribution, and community composition [8]. Rapid urbanization also causes dramatic changes in local climate, food and habitat availability, water fow, and waste accumulation, which directly or indirectly afect the composition of soil microbial and pathogen communities and their interaction with host species. Tis infuences the distribution, intensity, and dynamics of infectious diseases [9]. Furthermore, the urban environment may pose serious and novel stressors to wildlife that may compromise their immunity and make them more susceptible to infectious diseases.
It is widely acknowledged that many emerging infectious diseases, such as those related to coronaviruses, are associated with anthropogenic modifcations of the environment [10,11]. Wildlife in their natural habitats may decrease the transmission of infectious diseases via the dilution or amplifcation efect. According to the dilution hypothesis, an increase in species diversity leads to a decrease in pathogen prevalence [12]. An increase in the number of nonhost species decreases the number of intraspecies encounters between infected and susceptible hosts, leading to a decrease in the pathogen transmission rate and prevalence [5]. Te amplifcation efect, on the other hand, argues that there is a positive correlation among species diversity, disease risk, and infection prevalence [12].
Clearing of forests and other natural habitats for urbanization destroys wildlife habitats and reduces species richness as habitat specialist species become extirpated [13]. However, habitat degradation and fragmentation could also increase species richness by adding new habitats, such as forest edges that attract invasive or generalist species [8]. Te increase or decrease in species richness as a result of anthropogenic habitat modifcations may lead to the loss or gain of particular host or vector species, which can afect disease exposure and transmission [12]. Terefore, in principle, loss of biodiversity due to rapid urbanization could either increase or decrease disease transmission. However, mounting evidence indicates that biodiversity loss frequently increases disease transmission [5].
Te complex social-ecological system and several characteristics of the urban environment, including high human population densities, loss of natural habitats, and humans' close proximity to wildlife, increase human interactions with wild animals and their pathogens [9]. Urban areas are therefore key venues for potential spillover events because of the presence of zoonotic pathogens transmitted by hosts and vectors living in close proximity to dense human settlements [5,9]. Already, it is estimated that zoonotic diseases account for about 2.5 billion cases of infections and 2.7 million deaths annually, with substantial socioeconomic consequences [14]. Te ongoing COVID-19 pandemic is also a stark reminder of the devastating consequences of pathogen spillover from wildlife to human hosts in urban areas [9].
Te rapidly expanding cities in developing tropical countries and the ever growing human populations in urban areas enhance human-wildlife interactions, which increase the risks of infectious zoonotic diseases of wildlife origin. Te prevention of future zoonotic disease is contingent on a better understanding of host-parasite interactions as well as informed surveillance for known and novel threats across diverse human-wildlife interfaces, particularly in humandominated landscape [9].
Te African rainbow lizard (Agama agama complex) is the most abundant and widespread lizard species in rural, suburban, and urban areas of sub-Saharan Africa [15,16]. Agama agama plays a fundamental role in ecosystem structure and functioning. Tey form an integral part of animal biomass and important actors in the energy fow and materials cycle of natural and human-modifed landscapes. Tey feed on insects and other arthropods, thereby controlling insect population to prevent them from becoming a nuisance in the ecosystem. Tey also serve as food for several birds of prey, snakes, and other reptilian, mammalian, and even arthropod predators, especially when the lizards are young. Agama agama is a rich source of protein for some local communities [17,18] and a model organism for scientifc research.
Agama lizards have been the subject of many ecological, physiological, and epidemiological studies [19][20][21][22][23][24][25][26][27][28]. Oyekunle et al. [29] assessed the suitability of using Agama lizards as a biomarker for environmental pollution of some heavy metals and concluded that the liver and kidney of Agama lizards are suitable for the study of cadmium, zinc, arsenic, barium, manganese, and lead contamination in soils. Despite the extensive distribution and abundance of A. picticauda (a member of the Agama agama complex) in West Africa, data on its population ecology and life history are scant. Anthropogenic induced land-use change such as urbanization may infuence the species' interactions with pathogens and humans. Tis necessitates studies of the parasite community in the species across its geographic range in order to understand how it adapts to humanmodifed environment to facilitate its conservation. Also, data on the zoonotic risk potential associated with the lizards are vital for efective mitigation.
To the best of our knowledge, no study has simultaneously assessed the enteric bacteria and helminths of Agama picticauda in human-dominated landscape in Ghana and West Africa. Terefore, the present study assessed the diversity, prevalence, and composition of enteric bacteria and helminths of A. picticauda in urban and suburban areas of the Accra plains of Ghana. We hypothesized that the prevalence of enteric helminths and bacteria will be higher in male lizards than females, in the wet season than dry season in the suburban area. We also expected the prevalence of helminths and bacteria to be higher in the urban area than suburban area during the wet season. We also expected some genera of the bacteria and helminths found in the lizards to be present in humans and therefore potentially zoonotic. Our fndings can serve as baseline information for long-term monitoring studies of reptile-parasite-environment interactions in the rapidly urbanizing Accra plains of Ghana as these parasites are good bio-indicators of environmental change.

2
International Journal of Zoology

Study Area.
Te study was carried out in developed urban area and suburban area in the Greater Accra Region of Ghana ( Figure 1). Urbanization can be defned by population size, population density, or impervious surface area or built structures [30]. In this study, urbanization was defned a priori using the proportion of impervious surface, i.e., roads and buildings within the surrounding landscape [31]. Urbanization increases the coverage of impervious surface with a corresponding decrease of the original vegetation cover. Te suburban area was defned to cover areas with about 40 to 70% of impervious surface, while the urban area was characterized by more than 70% of impervious surface. In the urban area, samples were collected within the developed areas of the University of Ghana, Legon main campus (N 05.650°, W 00.186°), which is located about 8 km north of the Central Business District (CBD) of Accra. For the suburban area, samples were collected at Oyibi (N 5.808°, W −0.118°), which is about 27 km northeast of the CBD of Accra.
Te climate of the study area is typical of that of the Accra plains. Te mean annual rainfall ranges from 733 to 1118 mm, and the mean annual temperature is 25 to 28°C. Te vegetation is mainly coastal savanna comprising shrubs interspersed with thickets, woodland, and forest [32].

Study Species.
Agama lizards are diurnal insectivores that range in snout-vent length from approximately 47 mm to 133 mm [33]. Te distribution of the species extends from Mauritania and Sierra Leone in the west to Ethiopia and Kenya in the east [34]. Te species is highly territorial with dominant males, particularly conspicuous, having blueblack bodies, orange-yellow heads, and tricolored tails [16,35]. Females, immatures, and subdominant males are uniformly greenish brown [16,24]. Agama lizards reproduce all year round, but reproduction peaks in the wet season [36]. Te female lizards become sexually matured in 14 to 18 months, while it takes about 24 months for males to become sexually matured [36]. Tey are strictly diurnal [33], with bimodal peak activity occurring in the morning and late afternoon [16]. Tey are sit-and-wait predators, with ants, beetles, and grasshoppers being the most common consumed prey item [37]. Agama lizards are highly adapted to human-dominated landscape, with their continental distribution coinciding strongly with human settlements [37]. Overall, 57 adult Agama lizard individuals were captured and analyzed from the developed urban area and suburban area. In the developed urban area, 16 lizards (10 males and 6 females) were captured during the wet season only. In the suburban area, a total of 41 individual lizards were captured. Of these, 21 adult lizards (12 males and 9 females) were captured during the wet season and 20 (7 males, 13 females) were captured during the dry season. We searched for the lizards at random in both the urban and suburban areas, and when they were found, we captured at most three individuals (one adult male and two adult females) where possible, from a family group. Family groups were easily identifable because Agama lizards are territorial and usually live in a family group of up to 11 individuals (personal observation). We measured the snout-vent length (SVL) of all the captured lizard individuals using a meter rule. We analyzed all the 57 individuals for enteric helminths, but only the 37 individuals captured during the wet season (16 in the urban area and 21 in the suburban area) were analyzed for enteric bacteria.

Sample Collection and Gastrointestinal Helminth
Identifcation. Te guts of the Agama lizards were sectioned, and its content was emptied into dry cleaned labeled Petri dishes containing physiological saline. Te mixture was mixed with a stirrer and then observed under a dissecting microscope for the presence of adult helminths. After, the mixture was transferred into labeled 15 ml tubes and centrifuged at 1500 rpm for two minutes. Te supernatant was discarded, and 5 ml of zinc sulfate (ZnSO 4 , 33% v/v, specifc gravity, 1.180) solution was added and capped. Te mixture was then reconstituted and centrifuged for two minutes at 1500 rpm. Following this, a Pasteur pipette was used to collect part of the topmost supernatant onto a microscope glass slide, covered with a cover slip, and observed under a Leica 2000 compound microscope with a magnifcation of ×10 and ×40 for parasite ova. Parasites detected were confrmed by a chief microscopist.

Bacterial Sampling and Isolation.
Buccal, cloacal, and faecal swabs were collected from lizards with sterile cotton swab and stored in Eppendorf tubes containing 200 µl phosphate bufered saline. Each swab was streaked on MacConkey agar and incubated at 37°C for 18 to 24 hours aerobically. For enrichment and purifcation, morphologically similar colonies were transferred onto soft nutrient agar plate and incubated at 37°C for 18-24 hrs. Te pure bacteria isolates were identifed by morphological characteristics, Gram staining, motility test, and biochemical tests (indole, catalase, oxidase, urease, Triple Sugar Iron (TSI), Methyl-Red-Voges-Proskauer, and citrate). Table 1 shows a summary of how the Gram-negative bacteria were identifed (biochemical profle).
Te chi-square test was used to determine the statistically signifcant diference of the prevalence of parasites between sites (suburban vs. urban, for the wet season data only), season (wet vs. dry, for the suburban data only), and sexes (male vs. female, for both wet and dry season data for suburban and urban areas) of the lizards. Pearson's correlation coefcient (r) was used to assess the correlation between the snout-vent length (SVL) of lizards and diversity (richness) of enteric helminths infestations in the lizards after log-transforming the data. Te analysis was conducted using the R software (Version 4.0.5), and signifcance was set at an alpha level of 0.05.

Prevalence and Zoonotic Risk Potential of Enteric Helminths in Agama
Lizards. Fifty-seven Agama lizard individuals, 41 from the suburban area and 16 from the urban area, were analyzed for enteric helminths. A total of at least four helminth parasite genera, Ascaris spp., Pharyngodon spp., Enterobius spp., and Oxyurid spp., were detected. Te overall prevalence of enteric helminth infections in the lizards at the study area was high (71.9%), with Oxyurid spp. being the most (59.6%) prevalent and Enterobius spp. the least (8.8%) prevalent (Table 2). In the suburban area, the overall prevalence of enteric helminth was 68.3%, while in the urban area, it was 81.3%. Te prevalence of enteric helminths in the lizards was signifcantly higher (χ 2 = 5.25-33.3, p ≤ 0.02) in the wet season than dry season. Although the prevalence of enteric helminths was higher in lizards from the urban area than the suburban area during the wet season, the diference was statistically signifcant (χ 2 = 12.34, p < 0.001) for Ascaris spp. only. Oxyurid spp. was signifcantly more prevalent (χ 2 = 4.75, p � 0.03) in lizards from the suburban area than the urban area during the wet   (Table 3). Interestingly, Ascaris spp. was not observed in the lizards captured during the dry season at the suburban area. Also, the prevalence of enteric helminths was signifcantly higher in male than female lizards from the suburban area for both wet and dry seasons. For the lizards from the urban area, enteric helminths were generally more prevalent in female than male lizards (Table 3). We found a positive but statistically insignifcant correlation between SVL and diversity of enteric helminths for male lizards captured in the suburban and urban areas. However, for female lizards, we found a negative correlation between SVL and diversity of enteric helminths for the individuals captured at the suburban area and positive correlation for the individuals captured at the urban area ( Figure 2). Ascaris spp. and Enterobius spp. may pose potential zoonotic risks to humans.  2%). Generally, the prevalence of the identifed Gram-negative bacteria was slightly higher in males than females, but the diference was not statistically signifcant (χ 2 � 0.02-1.89, p � 0.17-0.90; Figure 3). Escherichia spp. was the most prevalent bacteria in male (90.9%) and female (80%) lizards, while Salmonella spp. was the least prevalent species, with 18.2% and 13.3% in male and female lizards, respectively. Also, the prevalence of enteric bacteria was higher in lizards from the suburban area than urban area, but the diference was not statistically signifcant (χ 2 � 0.59-3.34, p � 0.07-0.44). However, the prevalence of Escherichia spp. and Proteus spp. was signifcantly higher (χ 2 � 5.77-7.34, p � 0.01-0.02; Figure 3) in lizards from the suburban area than the urban area. Again, Escherichia sp. was the most prevalent bacteria in lizards from both suburban and urban areas, while Salmonella sp. was the least prevalent species (Figure 3).

Prevalence and Zoonotic Risk Potential of Enteric
Escherichia spp., Enterobacter spp., Shigella spp., and Salmonella spp. could be of a potential zoonotic risk to humans.

Diversity, Prevalence, and Zoonotic Risk Potential of
Enteric Helminths in A. picticauda. Parasitic infections are ubiquitous in wildlife, and healthy ecosystems are often parasite rich. Parasites play an important role in natural communities, infuencing host behaviour, resource utilization, predation risks, body condition, population dynamics, and ftness [2][3][4]38]. Infected reptiles may also pose a serious threat for public health because it may be a source of zoonotic pathogens for humans and livestock. We identifed four genera of enteric helminths in A. picticauda from suburban and urban areas in Accra, Ghana. Te number of genera of enteric helminths that was recovered from A. picticauda was similar to what was found (4 genera) in Agama agama from the Nsugbe, Anambra State, Nigeria [39], but it was higher than that observed (2 genera) in Agama agama from Ibadan, Nigeria [40], and lower than that observed (7 genera) in Agama agama from Legos, Nigeria [41].
All the helminths recovered from A. picticauda were nematodes, and this fnding corroborates the fndings of other studies that suggest that nematodes form the bulk of the enteric helminths of Agama species [39][40][41][42]. Likewise, Rataj et al. [43] found that most of the 18 parasite taxa in specimens of lizards imported into Slovenia were nematodes. Similarly, the endoparasites of reptiles in Namibia were mostly nematodes [44].  Te prevalence of helminth infections in A. picticauda was high and supports information in the literature, which suggests that helminth infection among lizards can be high [45]. For instance, high prevalence of enteric helminths has been reported in Agama agama from Legos, Ibadan, Ile-Ife, Anambra State, Okrika Rivers State, and Southwest Nigeria [39][40][41][42][43][44][45][46]. Furthermore, studies on house lizards (Gekkonidae): Cosymbothus platyurus and Hemidactylus frenatus from Tailand revealed high levels of infection [47].
We found higher prevalence of enteric helminths in male lizards than their female counterparts, and this corroborates the fndings of Robert et al. [46] who reported higher prevalence of enteric helminths in male Agama agama than females from the Okrika Rivers State in Nigeria. However, the prevalence of enteric helminths in Agama agama from Lagos and Anambra State, Nigeria, was similar for male and female lizards [39,41]. Adeoye and Ogunbanwo [41] and Sowemimo and Oluwafemi [42] reported signifcantly higher mean intensity of helminth infection in male than female Agama agama, but the present study did not consider helminth intensity in A. picticauda. Tis indicates that both male and female lizards are susceptible to helminth infection, but the larger and highly active males are more prone to infections because of increased contact with parasites during feeding.  International Journal of Zoology Also, we found higher prevalence of enteric helminths in A. picticauda during the wet season than dry season and in suburban area than urban area. Tis was expected given that most parasitic helminths are transmitted through soil and become abundant during the wet season when the environment is warm and humid [48]. Te urban area had a greater percentage of impervious surfaces that were not favourable for helminth growth and survival. Terefore, the Agama lizards in this area encountered less helminths compared to those in the suburban area. Tis fnding supports the results of some studies in the literature; for instance, Dugarov et al. [49] found that the prevalence, intensity, and diversity of helminths in the Mongolian racerunner (Eremias argus; Lacertidae, Reptilia) decreased with the increasing degree of urbanization. Also, Sitko and Zaleśny [50] found that urbanization signifcant reduced the helminth fauna of the Eurasian blackbird (Turdus merula).
We found a positive correlation between SVL and diversity of enteric helminths for male lizards captured in the suburban and urban areas, but for females, we found a negative correlation between SVL and diversity of enteric helminths for the individuals captured at the suburban area and positive correlation for the individuals captured at the urban area. Sowemimo and Oluwafemi [41] found a positive correlation between the SVL and helminth prevalence and intensity in Agama agama from Nigeria. Mature female Agama lizards in the suburban area where food resources are abundant perhaps use a small area, usually around the center of the dominant male's territory and therefore encounter less helminths during feeding. In the urban area where food resources are less, females extend their feeding ground in order to meet their energy requirement and thus encounter more helminths while feeding.
We found higher prevalence of bacteria in male than female lizards, but both sexes had similar genera richness and composition. Tis contrasts with the results of Martin et al. [57], who found sex asymmetry in microbial community structure and composition of free-ranging striped plateau lizards (Sceloporus virgatus), with males having signifcantly higher microbial diversity and richness than females. Sex bias in susceptibility to parasitic infection may be linked to sex diferences in the production of sexual hormones that afect directly or indirectly the host's immunological response [58,59]. Te males, which are the larger sex, are more active and invest more energy into vigilance and territory defense. Tis may reduce the efcacy of their immune function, and so, the males tend to be more exposed to parasitic infections [60]. Te larger home range of the males might also enhance their exposure to parasites [61]. Also, we found higher prevalence of bacteria in lizards from the suburban area than urban area, but the genera richness and composition did not change. Changes in community composition and behavior, such as a switch in diet to include more anthropogenic food associated with urbanization, might have led to the lower parasite prevalence in the suburban lizards. Modupe al. [62] isolated 10 bacteria genera including Shigella spp., Yersinia spp., Salmonella spp., Citrobacter spp., Enterobacter spp., Klebsiella spp., Proteus spp., Pseudomonas spp., Serratia spp., and Escherichia spp. from birds in a protected forest in Ghana. Te fact that A. picticauda shared all the six bacteria genera with wild-caught birds from Ghana suggests that these bacteria genera are common in the environment and are easily picked up by individual animals during foraging.

Conclusion
Our study highlights the parasite-host and environment relationship in A. picticauda and is the frst to assess simultaneously the enteric bacteria and helminths of A. picticauda from Ghana. Our results showed that enteric helminths and bacteria were more prevalent in males than females, in the wet season than dry season and in the suburban area than urban area. We found a positive correlation between SVL and diversity of helminths in male lizards. Te helminths Ascaris and Enterobius and bacteria Escherichia, Enterobacter, Shigella, Klebsiella, Proteus, and Salmonella are all potentially pathogenic to humans. Tis raises public health concerns given the close association of the rainbow lizards with human dwellings. Low sample sizes and lack of replication were the main limitations of the present study. Te methods we used to identify helminths and bacterial could only identify to the genus level and not species. Terefore, future studies should increase the International Journal of Zoology number of lizard individuals captured from each site, and also, several urban and suburban sites should be considered to increase the power of the statistical analysis. Molecular techniques should be used in future studies to determine the identity the actual species of enteric helminths and bacteria of A. picticauda in order to establish their zoonotic status.

Data Availability
Te data that support the fndings of this study are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare no conficts of interest.