The crocodile lizard (Shinisaurus crocodilurus Ahl, 1930) is the only species in the monotypic genus Shinisaurus and the monotypic family Shinisauridae. It is a relict reptile that now survives only in separated Pleistocene refugia. This species is distributed in southern China (Guangdong and Guangxi Provinces) and northern Vietnam (Qung Ninh and Bac Giang Provinces) with severely fragmented populations (van Schingen et al., 2014, 2016). This species is essential not only for taxonomical systematics but also for understanding the origin, adaptation, and evolution of reptiles. However, it faces extinction due to the pressure of being hunted; environmental changes; and habitat destruction (Huang et al., 2008; Nguyen and Ziegler, 2015). It has been listed as an endangered species on the IUCN Red List of Threatened Species (Nguyen et al., 2014), a class I protected species in China, and an appendix I species by the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES I). According to a recent survey, the total number of crocodile lizards in the wild has decreased from 6000 in 1978 to approximately 1200 in China (Huang et al., 2008; unpublished survey conducted by Wu et al., 2012). What’s worse, the population continues to decline sharply. Similarly, the wild population in Vietnam has decreased to fewer than 150 individuals in recent years (van Schingen et al., 2016). Some nature reserves, such as Guangdong Luokeng S. crocodilurus National Nature Reserve and Guangxi Daguishan Crocodile Lizard National Nature Reserve, are conducting captive breeding and release programs with the hope of restoring the wild populations. However, the captive individuals can become infected with serious diseases that cause many deaths each year (Figure 1). Crocodile lizards have two major types of diseases that can be distinguished by their symptoms. Disease A is characterized by one or more nodules in the underjaw or limbs covered by lesions (Figure 1A), while disease B is characterized by varying degrees of rot in the four limbs (Figure 1B). In the wild, individuals with disease B have been observed, while individuals with disease A have not been found.

Recent studies have revealed how variations and changes in the composition of gut microbial communities influence normal physiology and contribute to diseases (Clemente et al., 2012; Martin et al., 2014; Boursier et al., 2016). In addition, gut microbes affect host immunity, behavior, reproductive isolation, and metabolism (Cryan and Dinan, 2012; Brucker and Bordenstein, 2013; Ramakrishna, 2013; Thaiss et al., 2016; Sylvia et al., 2017). Conversely, many factors such as diet or host genetics can shape the microbial community (David et al., 2014; Goodrich et al., 2014).

Study on the gut microbiota has been conducted in a host of vertebrates, including mammals (Thaiss et al., 2016), birds (Hird et al., 2015; Waite and Taylor, 2015), fishes (Gajardo et al., 2016), amphibians (Bletz et al., 2016), and reptiles (Costello et al., 2010; Yuan et al., 2015; Kohl et al., 2016, 2017; Ren et al., 2016). However, the majority of these studies have been conducted in mammalian hosts. Surveys of the gut microbiota in reptiles, an ancient group with more than 10,000 extant species (Uetz et al., 2016), remain rare except in the case of some economically important species, such as snakes (Costello et al., 2010; Colston et al., 2015) and turtles (Huang and Zhang, 2013; Yuan et al., 2015). A few studies have been conducted on the gut microbiota of lizards, which represent about 60% of reptiles (Martin et al., 2010; Hong et al., 2011; Kohl et al., 2016, 2017; Ren et al., 2016). Nevertheless, the interaction between the gut microbiome and disease in reptiles remains unclear.

According to previous studies, the reptilian immune system differs from those of other vertebrates in several aspects (Zimmerman et al., 2010). Here, two questions are raised. Is the gut microbiota associated with disease susceptibility in crocodile lizards? How does cultivation shape the gut microbiome of crocodile lizards? To explore these questions and to facilitate the protection of this endangered species, we analyzed the amplicon-based microbiome of cloacal swab samples from crocodile lizards using 16S rRNA gene sequencing. We compared the gut microbiota of sick, healthy, captive, and wild lizards to identify the interactions among specific diseases, diets, and the gut microbiota in crocodile lizards.

We analyzed the bacterial composition of 30 crocodile lizard cloacal swab samples (Table 1). Each sample contained at least 30,000 effective sequences (Supplementary Figure 1). The rarefaction curves showed that these sequence depths were sufficient for capturing the major microbiota in each sample (Supplementary Figure 2). More than 99% of the OTUs could be well annotated at the family level in each sample (Supplementary Figure 3).

Determining the role of the intestinal bacterial community in digestion and pathogenesis depends critically upon defining the “wild” and “normal” states. With high-throughput sequencing, it is now possible to comprehensively identify the bacteria in a given community, including fastidious and unculturable taxa. This study aimed to describe, for the first time, the wild state of the crocodile lizard gut microbiota as well as potential alterations in the composition of the gut microbiota of crocodile lizards with different diseases and diets compared with that in wild and healthy control subjects.

The composition of the gut microbiota in crocodile lizards is unique. Previous reports have indicated that the gut microbiota of lizards is dominated by the phyla Firmicutes (33.2–73%), Bacteroidetes (6.2–45.7%), and Proteobacteria (5.7–62.3%) (Martin et al., 2010; Nelson et al., 2010; Hong et al., 2011; Ren et al., 2016; Kohl et al., 2017). In other reptiles, the gut microbiota also appeared to be consistently dominated by Firmicutes, followed by Bacteroidetes, while Proteobacteria ranged from dominant to minor components (Costello et al., 2010; Colston et al., 2015; McLaughlin et al., 2015; Yuan et al., 2015). An exception is the alligator gut microbiota, which was dominated by Fusobacteria (Keenan et al., 2013). In other vertebrates, Firmicutes and Bacteroidetes also represent overwhelming majority of the gut microbiome (summarized by Keenan et al., 2013). Conversely, in the crocodile lizard gut microbiota, the proportion of Firmicutes was low, in all samples (0.13–14.56%), while Proteobacteria and Bacteroidetes were dominant (Figure 2). Moreover, in the wild samples from Luokeng Nature Reserve, the high prevalence of phylum Deinococcus–Thermus in the gut microbiome has not been found in other vertebrates. Deinococcus–Thermus spp. are usually found in extreme environments (Theodorakopoulos et al., 2013). They have also been detected in the feces of some animals with only a few clones (Lagier et al., 2012; McLaughlin et al., 2012). The role of Deinococcus–Thermus in the intestinal tract is not clear. However, Deinococcus–Thermus bacteria have been reported to show remarkable resistance to a range of stresses such as ionizing radiation, UV radiation, oxidizing agents, and desiccation (Theodorakopoulos et al., 2013). In addition, the relative abundance of Deinococcus–Thermus was significantly higher in the wild group than in the captive groups from the same place (Figures 2, 7). Therefore, the high proportion of Deinococcus–Thermus may help crocodile lizards adapt to the wild environment in Luokeng Nature Reserve. The wild samples from the Daguishan Nature Reserve had high abundances of Mycoplasma and Helicobacter. Particularly, the total abundances of Mycoplasma and Helicobacter reached up to 71.2–91.4% in samples WDG02, WDG03, WDG05, and WDG10. Because Mycoplasma and Helicobacter were detected in all samples, they are likely commensal inhabitants of crocodile lizards. Mycoplasma is a genus of bacteria that lack a cell wall and is primarily obligate commensals or parasites. Some species of Mycoplasma are significant pathogens of birds, mammals, fish, and reptiles, although many of them are harmless commensals to the hosts (Bano et al., 2007; Ossiboff et al., 2015). The overwhelming dominance of Mycoplasma have been reported in the fish gut or oyster stomach (Bano et al., 2007; King et al., 2012). However, the dominance of Mycoplasma in reptile guts was first reported. In the terms of Helicobacter, the most widely known member is H. pylori, which is strongly associated with peptic ulcers, chronic gastritis, duodenitis, and stomach cancer in humans (Khalifa et al., 2010). However, it is unclear whether the dominant Mycoplasma and Helicobacter would be harmful for the crocodile lizards.

Many studies showed that the gut microbiota had a relationship with a variety of diseases. In crocodile lizards, there was no significant correlation between the two diseases (A and B) and the gut microbiota. This may be because both disease A and B are located on the skin surface and do not have direct contact with the gut system.

Previous studies have shown that captivity can change the diversity of the gut microbiota (Nelson et al., 2013; Kohl and Dearing, 2014). In a study of Anolis sagrei insectivorous lizards also demonstrated that captivity led to a shift in microbial diversity (Ren et al., 2016). However, an investigation of omnivorous and herbivorous lizards showed that captivity had no significant effect on gut microbial diversity in terms of alpha diversity (Kohl et al., 2017). In this study, a comparison of alpha diversity indices, PCoA analysis, and UPGMA cluster suggested that the effect of captivity in shaping of gut microbiota was depend on diet. The diet of loach significantly changed the community richness of the gut microbiota of crocodile lizards but the diet of earthworm did not. However, captivity had no effect on the community diversity of gut microbiota of crocodile lizards according to the Shannon index.

The effect of captivity in shaping of gut microbiota was also reflected in the enrichment of specific microbes. The earthworm-fed group exhibited a gut microbiota enriched in Petrimonas, Bacteroides, Fusobacterium, Microbacter, and Proteiniclasticum compared with the wild and loach-fed groups (Figures 6, 7). Moreover, the captive lizards were consistently had a high abundance of Bacteroides compared with the wild samples (Figure 7). Petrimonas and Microbacter are fermenters isolated from environment (Grabowski et al., 2005; Sánchez-Andrea et al., 2014). The Proteiniclasticum, a genus of Clostridiaceae, is known as polysaccharide degrader in the gut (Zhang et al., 2010; Wust et al., 2011). A high relative abundance of Clostridiaceae was also observed in Anolis lizards (Hong et al., 2011). Proteiniclasticum may facilitate energy consumption in crocodile lizards. Notably, the diet of earthworm resulted in enrichment of Fusobacterium, which has been reported as human and animal pathogens (Signat et al., 2011). In addition, Fusobacterium has been commonly isolated from infected reptiles (Stewart, 1990). In the gut microbiota of other lizards, the abundance of Fusobacterium, if any, was very low, similar to that of the wild group of crocodile lizards (Martin et al., 2010; Hong et al., 2011; Kohl et al., 2016, 2017; Ren et al., 2016).

The loach-based diet resulted in the significant predominance of Elizabethkingia, Halomonas, Morganella, and Salmonella in the crocodile lizard gut microbiota. These genera are pathogens or opportunistic pathogens in animals and/or humans and have also been found in disease cases in reptiles (O’Hara et al., 2000; Bernardet et al., 2006; Stevens et al., 2009; Kim et al., 2013; Yeo et al., 2016). Particularly, attention should be paid to Salmonella, which causes enteritis and typhoid fever in mammalian and avian species. The Salmonella was detected in all samples, and its relative abundance in loach-fed crocodile lizards reached up to an average of 11.42% (Figure 8). Moreover, the zoonotic potential of Salmonella in reptiles has been widely documented (Murphy and Oshin, 2015). In addition, the Salmonella isolated from the focus of crocodile lizard with disease A caused the death of the Chinese skink in our lab (data have not been published). Moreover, Elizabethkingia was isolated from the focus of crocodile lizard with disease B (data have not been published). Therefore, hand washing should be recommended after contact with crocodile lizards, especially for those who contact these animals frequently. Although these genera have been detected as normal flora in many reptiles, there are many reports that associate these bacteria with reptile diseases such as bacterial pneumonia, osteomyelitis, septicemia, and hepatitis (Huchzermeyer, 1991; Ramsay et al., 2002; Grupka et al., 2006; Chinnadurai and Devoe, 2009). The high content of potential pathogenic bacteria in loach-fed group suggests that the diet of loach not only altered the structure of gut microbiota but also increased the risk of infection for crocodile lizards.

The composition of the crocodile lizard gut microbiota is unique compared with other animals. Diets altered the bacterial community richness and the relative abundance of certain bacteria in the intestine. The gut microbiota of loach-fed crocodile lizard was significantly different from the gut microbiota of the wild and the earthworm-fed crocodile lizards. The earthworm-fed crocodile lizards had a higher abundance of Fusobacterium in the gut compared with the wild lizards. The intestine of loach-fed crocodile lizard was enriched in Elizabethkingia, Halomonas, Morganella, and Salmonella. These bacteria were reported to be pathogens or opportunistic pathogens in human or other animals. This may be a consequence of unbalanced nutrition, as the crocodile lizards in this study were routinely fed with only earthworms or loaches at the Nature Reserves. It seems that the diet of loach was not suitable for crocodile lizards. However, there is no sufficient evidence that the gut microbiota contributes to either disease A or disease B.

This study provides an overview of the gut microbiota of the crocodile lizard, an extremely endangered lizard, at different states as well as the first examination of the relationship between disease and the gut microbiota in lizards. These findings have numerous implications for the practice of crocodile lizard conservation, from the perspectives of both captivity and disease prevention. For instance, the results emphasize that a more diverse diet could improve the care of crocodile lizards in wildlife rescue centers and nature reserves. In addition, the diet of earthworm is better for crocodile lizard than the diet of loach.

This study was carried out in accordance with the recommendations of Guidelines of Animal Experiments, the Committee on the Ethics of Animal Experiments of the Guangdong Institute of Applied Biological Resources. The protocol was approved by the Committee on the Ethics of Animal Experiments of the Guangdong Institute of Applied Biological Resources.

JPC and RCH designed the research. HYJ, NH, HYL, SYL, and ZJW collected the samples. HYJ, LML, and XJZ conducted the research. HYJ, JEM, and JL analyzed the data. HYJ and JPC wrote the manuscript. All authors approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewers AM and KK, and handling Editor declared their shared affiliation, and the handling Editor states that the process nevertheless met the standards of a fair and objective review.