Impact of host intraspecies genetic variation, diet, and age on bacterial and fungal intestinal microbiota in tigers

Abstract The bacterial microbiota in the gut varies among species, as well as with habitat, diet, age, and other factors. Intestinal microbiota homeostasis allows a host to adjust metabolic and immune performances in response to environmental changes. Therefore, potential implications of the gut microbiota in sustaining the health of the host have gained increasing attention in the field of endangered animal conservation. However, the effect of host intraspecies genetic variation on the gut microbiota is unknown. Moreover, little is known about the complexity of the gut mycobiota. Tigers are listed as endangered species, raising worldwide concern. Potential influences of subspecies, diet, and age on the gut microbiota in tigers were investigated in this study to provide a better understanding of the response of the tiger gut microbiota to external changes. The results revealed that the impacts of the factors listed above on gut bacterial and fungal communities are versatile. Host intraspecies genetic variation significantly impacted only fungal alpha diversity of the gut microbiota. Differences in diet, on the other hand, had a significant impact on alpha diversity of the gut microbiota, but exerted different effects on beta diversity of gut bacterial and fungal communities. Host age had no significant impact on the diversity of the gut fungal communities, but significantly impacted beta diversity of gut bacterial communities. This comprehensive study of tiger gut microbiota is an essential reference for tiger conservation when considering feeding and management strategies, and will contribute to a better understanding of the mycobiota in wildlife.

Upon habitat changes, wild animals were found to display differences in microbial community composition and bacterial diversity (Huang, Chang, Huang, Gao, & Liao, 2018;Watson et al., 2019).
The gut microbiota is known to vary because of host genetic variation above the species level in mammals (Ley et al., 2008;Nishida & Ochman, 2017;Song et al., 2020), but the effect of intraspecies differences on the gut microbiota is still unknown. Microbiome fluctuations allow the host to adjust metabolic and immune performances in response to environmental changes (Candela, Biagi, Maccaferri, Turroni, & Brigidi, 2012). Fecal microbiota transplantation has been used in humans to treat diseases, such as Clostridium difficile infection, irritable bowel syndrome, and inflammatory bowel diseases (Bajaj et al., 2017;Kelly et al., 2015).
So far, investigators have mostly focused on the bacteria present in the gut. Other microorganisms that are less common are often overlooked, such as archaea, viruses, and microbial eukaryotes (Cotter, Huseyin, Scanlan, & O'Toole, 2017). In recent years, fungi have increasingly been recognized as defining constituents of the gut microbiota, cofactors in disease, and interaction partners of the immune system (Huffnagle & Noverr, 2013;Sam, Chang, & Chai, 2017;Underhill & Iliev, 2014;Wheeler et al., 2016). Moreover, the gut mycobiome may act as a reservoir for opportunistic pathogens in immunocompromised hosts (Polvi, Li, O'Meara, Leach, & Cowen, 2015). An earlier report showed that fungi in the gut were stable over time (Scanlan & Marchesi, 2008), whereas other reports indicated that the gut mycobiota was transient and lacked core species (Hallen-Adams, Kachman, Kim, Legge, & Martínez, 2015). Studies in humans and mice have implied that the dynamics of mycobiota in the gut are influenced by the underlying pathophysiology and environment Iliev et al., 2012). However, studies on the fungal microbiota are still rare and focus mainly on humans. Little is known about how the gut mycobiota is influenced by various factors, and whether these factors exert similar effects on the fungal and bacterial intestinal microbiota.
The potential implications of the gut microbiome in endangered animal conservation have increasingly been drawn attention to in recent years, and a new branch of conservation biology, that is, conservation metagenomics, has been proposed (Stumpf et al., 2016;Wei et al., 2019;West et al., 2019). A total of 13,482 animal species worldwide are currently threatened with extinction, including vulnerable, endangered, and critically endangered species, according to the IUCN Red List of Threatened Species version 2018-2. More and more threatened species have been brought under intensive management or into captivity to save these animals. However, captive programs are often impeded by disease epidemics and poor health, as well as lower reproductive rates and higher offspring mortality compared to their wild-born counterparts (Jiang et al., 2017;Lowenstine, McManamon, & Terio, 2015;Seimon et al., 2013;Wasimuddin et al., 2017;Yang et al., 2017;Zhao et al., 2017). During intensive management or captivity, animals experience a range of changes regarding diet, drug treatment, and habitat, which may significantly alter the gut microbiome of these threatened animals. Numerous studies have already recorded the changes in the gut microbiome between captive and wild populations of endangered animals, such as those observed in dugongs (Dugong dugon), black howler monkeys (Alouatta pigra), nonhuman primates, Australian sea lions (Neophoca cinerea), and crocodile lizards (Amato et al., 2013;Clayton et al., 2016;Delport, Power, Harcourt, Webster, & Tetu, 2016;Eigeland et al., 2012;Jiang et al., 2017). A detailed understanding of the gut microbiome will have important implications for the management of endangered animal health in captivity.
Tigers are classified as endangered, with a decreasing population trend (Goodrich et al., 2015), and are further listed in the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) Appendix I. The total number of mature tigers worldwide was only 2,154-3,159 in 2014 (Goodrich et al., 2015). Moreover, SouthCN tigers have not been directly observed in the wild since the 1970s and are possibly extinct in the wild. Only 178 SouthCN tigers are conserved in zoos. Tiger cubs depend on artificial feeding, because cubs nursed by their mothers show low survival rates. In addition, the mother tigers are often unwilling to take care of their cubs when kept in zoos (Li, Lei, & Chen, 2013). Tiger cubs often die of indigestion and gastrointestinal diseases during artificial feeding . Therefore, a detailed understanding of the gut microbiota of tigers will prove important for raising tiger cubs and to improve survival rates.
This study was designed to investigate the intestinal microbiota in healthy tigers, and the impact of host genetic variation within species (subspecies and inheritance), diet, and age on the gut microbiota, to provide a better understanding of the gut microbiome, reduce the knowledge gap concerning fungal intestinal communities, and provide support for healthy tiger management. Bacterial and fungal microbial communities were studied to determine whether the effects of the above factors on the gut microbiota were the same for both communities.

| Sample collection
All tigers were housed at Guangzhou Zoo (China). When fresh feces of tigers were observed, sterile swabs were used to collect the feces from the center of the dunghill. These samples were placed in sterile tubes, immediately frozen in liquid nitrogen, and then transported to the laboratory for DNA extraction.
A total of 24 tigers were sampled, including 19 adults or subadults (3-16 years old), and 5 cubs (no more than 6 months old). Adult and subadult animals were fed meat, more specifically, a mixture of lean meat, chicken, and beef at a fixed ratio. Cubs were fed milk (KMR ® kitten milk replacer powder, Pet-Ag, Inc.) before 30-40 days of age; 40-to 60-day-old cubs were fed a mixture of milk and meat  Similarly, 32 samples were sequenced for fungal community analysis. When the effects of subspecies were studied, samples from adult meat-fed tigers were selected and grouped into three groups, namely Bengal (N = 6), Amur (N = 6), and SouthCN tigers (N = 7). The effects of different diets were studied by separating the samples into two groups: meat-fed (N = 5) and mix-fed tigers (N = 8). For the milk-fed group, a total of eight samples from milk-fed tigers were collected, but seven of them failed to generate an ITS1 amplicon band.
When the effect of age was studied, the samples were separated into two groups: adult (N = 7) and young tigers (N = 5). To study the effect of inheritance, the same four families were studied that had been used for bacterial microbiota analysis.

| Data analysis
Sequenced reads were submitted to analysis using the microbial diversity platform of BMKCloud (http://www.biocl oud.net/) for quality control, operational taxonomic unit (OTU) assignment, OTU annotation, alpha and beta diversity estimations, principal coordinate analysis (PCoA), unweighted pair group method with arithmetic means (UPGMA) clustering, evaluation of linear discriminatory analysis (LDA) effect size (LEfSe), and function prediction. In brief, sequences with 97% similarity were set as the same OTU using QIIME (Caporaso et al., 2010). OTU abundances were normalized to the sample with the low- and FUNGuild (Nguyen et al., 2016), respectively. For comparisons of alpha diversity, a t test was performed to analyze the significance of the differences between groups using the R package. For comparisons of beta diversity, PERMANOVA test was conducted to test the significance of the differences among groups using the R package.

| Data assessment in 16S rDNA analysis
A total of 34 samples were sequenced for analysis of bacterial communities in tiger guts (Table 1). Each sample contained more than 58,697 effective sequences. Rarefaction curves showed that these sequence depths were sufficient for capturing the major bacterial groups in each sample ( Figure A1). More than 96% of OTUs could be annotated at the genus level in each sample (Table A1).

| General pattern of tiger intestinal bacterial microbiota
According to their relative abundances, bacterial communities in the tiger gut microbiota were similar among subspecies, but differed with diet and age ( Figure 1). The composition of the gut bacterial communities of meat-fed and mix-fed tigers was similar, but differed from that of milk-fed tigers. Generally, the bacterial microbiota of the tiger gut showed a predominance of Firmicutes, Fusobacteria, Bacteroidetes,

| Comparison of bacterial diversity indices among groups
In the gut bacterial microbiota, the comparison of Shannon indices showed that milk-fed tigers had significantly lower community diversity than meat-fed and mix-fed tigers, whereas mix-fed tigers showed no significant differences in community diversity compared In terms of beta diversity, according to PCoA analysis, the samples were separated by diet and age, but no clusters by subspecies were observed ( Figure 3). These separation patterns were also supported by UPGMA cluster analyses ( Figure 4) and PERMANOVA tests (p < .05, Figure 3). These results indicated that the beta diversity of gut bacterial microbiota was significantly different depending on diet and age, whereas the differences in beta diversity among subspecies were not significant. In addition, PCoA analysis of four tiger families also showed that these samples were not clustered by family according to the genetic relationship, but by age ( Figure A2).

| Comparison of bacterial taxa among groups
In the LEfSe analyses, only genus and species levels were selected for use in screening differential bacteria among groups, as more than 96% of OTUs from 16S rDNA sequencing were well annotated at the F I G U R E 1 Relative abundances of intestinal bacteria at phylum (a-c) and genus (d-f) levels genus level (Table A1).

| Function prediction of bacterial communities and comparison between groups
Distributions of functional classes were similar among the samples. The main functions of gut bacteria in tigers were carbohydrate metabolism, amino acid metabolism, energy metabolism, and metabolism of cofactors and vitamins, among others, according to the annotation in the KEGG database ( Figure A3).
The comparison of meat-fed and milk-fed tigers revealed significant differences in energy metabolism and infectious diseases (parasitic) ( Figure 6).
When the effects of age on intestinal microbial functions were analyzed, significant differences were noted in global and overview maps, as well as in the biosynthesis of other secondary metabolites in adult and young tigers ( Figure 6).

| Data assessment in ITS1 analysis
A total of 32 samples were sequenced for analyzing intestinal fungal composition in tigers (Table 1). Each sample contained at least 57,336 effective sequences. Rarefaction curves showed that these sequence depths were sufficient for capturing the major fungal groups in each sample ( Figure A4). However, OTU annotation of ITS1 rDNA was not as detailed as that of 16S rDNA (Table A2).

| General pattern of tiger intestinal mycobiota
The composition of the fungal community in the gut of tigers was similar among subspecies, as well as between different diet, and age groups F I G U R E 2 Comparison of Chao1 (a-c) and Shannon (d-f) indices for intestinal bacterial communities categorized by subspecies, diet, and age. ** represents p < .01 in statistical tests ( Figure 7). In sum, the tiger intestinal fungi belonged to 11 phyla, 31 classes, 69 orders, 171 families, 304 genera, and 344 species.

| Comparison of fungal diversity indices among groups
For alpha diversity, no significant differences existed in community richness (Chao1 indices) among subspecies, diet, and age groups (Figure 8a-c). Comparison of Shannon indices revealed that community diversity of the gut mycobiota was significantly For beta diversity, the tigers' intestinal fungal samples could not be separated by subspecies, diet, or age according to PCoA (Figure 9) and UPGMA cluster analyses. PERMANOVA tests also showed no significant differences in beta diversity among subspecies, diet, or age groups (p > .05, Figure 9).

| Comparison of fungal taxa among groups
Among the three subspecies studied, the SouthCN tigers showed significantly higher abundances of Alternaria and Pleosporaceae (Figure 10a).
Regarding dietary effects on the tiger intestinal fungal community, abundances of Microascales, Microascaceae, and Microascus were significantly increased in the guts of meat-fed tigers compared with that of mix-fed tigers (Figure 10b). Compared to adult tigers, young tigers were characterized by significantly higher abundances of Basidiomycota and Cladosporium in the gut mycobiota (Figure 10c).

F I G U R E 6
Difference in the function of the intestinal bacterial microbiome regarding diet (a) and age (b)

| Function prediction of fungal communities and comparison between groups
The function of tiger intestinal fungi was categorized into eight trophic modes. Among these trophic modes, saprotrophs were the most abundant fungal guild. Notably, 9.83% of fungal sequences, on average, were predicted to be derived from pathogenic fungi ( Figure 11).
Among the predicted pathogens, 13 genera were predicted to be animal pathogens or fungal parasites: Cutaneotrichosporon,

| D ISCUSS I ON
Understanding the factors that cause changes in the composition and function of the gut microbiota could provide insights into possible measures to promote the health of endangered animals by targeting these microbial communities in clinical treatment and health management settings. Using high-throughput sequencing technology, the characteristics of the tiger intestinal microbiota and the factors underlying the observed differences were comprehensively investigated in this study.
The gut bacterial microbiota in tigers is generally dominated by Firmicutes, Fusobacteria, Bacteroidetes, Proteobacteria, and Actinobacteria. This finding is consistent with previous studies on tiger intestinal bacterial composition (Peng, 2017;Tian, 2013). The pattern of the bacterial microbiota in the tiger gut is similar to that in other mammals, except for the significantly higher abundance of Fusobacteria (21.71% in adult tigers) (Ley et al., 2008). Fusobacteria were all contributed by Fusobacterium. A higher abundance of Fusobacterium has been associated with a series of diseases in humans, such as colorectal carcinoma and acute appendicitis (Brennan & Garrett, 2016;Zhong, Browersinning, Firek, & Morowitz, 2014).
In addition to Fusobacterium, some species of Clostridium, another dominant genus, have been reported to be associated with colorectal carcinoma in humans. Clostridium perfringens type A was also F I G U R E 8 Comparison of Chao1 (a-c) and Shannon (d-f) indices for the intestinal fungal communities categorized by subspecies, diet, and age. * represents p < .05 in statistical tests reported to be associated with severe hemorrhagic enterocolitis in tigers (Zhang, Hou, & Ma, 2012). However, the rate of colorectal carcinoma in tigers is far below that observed in humans (Tian, 2013 The gut microbiota of endangered animal offspring has rarely been characterized. In this study, the effects of different diets on the gut microbiota of young tigers under 6 months of age were investigated. A comparison of alpha and beta diversity indices among groups showed that the diversity of bacterial microbiota in the gut of milk-fed tigers was significantly different from that in the mix-fed and meat-fed tigers' gut. This was consistent with F I G U R E 1 0 Linear discriminatory analysis (LDA) effect size (LEfSe) of the intestinal mycobiota regarding subspecies (a), diet (b), and age (c) categories

F I G U R E 11
Distribution of functional categories of the gut mycobiota. The mean percentages of each sector are presented previous studies in other vertebrates, implying that the diet could shape the gut microbiota (Ley et al., 2008;Nishida & Ochman, 2017). The intestinal bacterial microbiota in milk-fed cubs was characterized by low diversity and was dominated by Escherichia/ Shigella, Bifidobacterium, and Lactobacillus. This finding was similar to observations made in formula-fed human infants (Azad et al., 2013;Penders et al., 2006) and corresponded to the milk-fed tiger diet, which included some probiotics. Milk-fed tigers showed significantly higher abundances of Bifidobacterium and Lactobacillus, and a lower abundance of Fusobacterium. Bifidobacterium and Lactobacillus species are well known for their role in milk fermentation and are hence widely used as probiotics (Widyastuti & Febrisiantosa, 2014;Xiao et al., 2003). In addition, bifidobacterial diversity is believed to play a role in the development of the host immune system in early life (Sjogren et al., 2009). Although the periods of diet change were short, young tigers with a change in diet from milk to meat displayed increased alpha diversity of the intestinal bacterial microbiota, changes in dominant groups of bacteria, and decreased levels of bacteria that could improve milk digestion, such as Bifidobacterium and Lactobacillus (Jiang, Mustapha, & Savaiano, 1996;Widyastuti & Febrisiantosa, 2014;Xiao et al., 2003). In contrast, bacteria that are associated with meat digestion gradually increased, such as Fusobacterium and Bacteroides (Bachrach et al., 2004;Davila et al., 2013;Macfarlane, Allison, Gibson, & Cummings, 1988;Riepe, Goldstein, & Alpers, 1980). Differences in intestinal bacterial function showed that milk-fed tigers had significantly lower abundances of bacteria associated with energy metabolism and higher abundances of pathogenic bacteria. This result implied that the diet not only changed the metabolism, but might also be important for pathogen invasion. For example, guts of milk-fed tigers were enriched for Escherichia/Shigella, which were significantly associated with diarrheal disease in infants (Pop et al., 2014). In addition to milk-fed cubs, meat-fed cubs also showed a significant difference in beta diversity of their gut bacterial microbiota compared with adult tigers. This was consistent with previous results showing that the composition of the gut microbiota varies with age (Rothschild et al.,

2018). Age-related shifts in bacterial abundance in tigers included
Bacteroides, Megamonas, Anaerotruncus, Phascolarctobacterium, Peptostreptococcus, and Clostridium groups. Among these genera, changes with age in Bacteroides and Clostridium groups have also been observed in humans (Clark & Walker, 2018). The different intestinal bacterial microbiota between adult and young tigers is explained by the metabolic function of global and overview maps and biosynthesis of other secondary metabolites (Figure 6), which are consistent with active metabolism in young tigers.
Previous studies of the mammalian gut microbiota showed that host taxonomy structured the intestinal bacterial microbiota at the order, family, and genus levels; the fecal microbial communities in the same species were more similar to each other than to communities of different species (Ley et al., 2008;Nishida & Ochman, 2017). However, at the intraspecies level, the diversity of the bacterial microbiota in the tiger gut was not significantly changed by the tiger subspecies. Moreover, when the effect of heredity is considered, the gut microbiota clustered by age instead of family. These results implied that the bacterial microbiota was less influenced by host intraspecies genetic variation.
Although the gut mycobiome is receiving increased research attention for its potential role in the etiology of gut-associated diseases, research on the gut mycobiome is still in its infancy (Cotter et al., 2017). Current knowledge on fungal diversity in the gut of vertebrates is still limited to a small number of species, such as humans, mice, dogs, bats, and several fishes (Cotter et al., 2017;Foster, Dowd, Stephenson, Steiner, & Suchodolski, 2013;Li et al., 2018;Siriyappagouder et al., 2018). At the phylum level, tiger intestinal fungal composition was similar to that observed in human, mouse, dog, bat, and fish guts, which are dominated by Ascomycota and Basidiomycota (Cotter et al., 2017;Foster et al., 2013;Li et al., 2018;Siriyappagouder et al., 2018). Like bacterial communities, the gut mycobiota differed among vertebrate species. Dominant genera of the tiger mycobiota were Aspergillus, Cladosporium, Nigrospora, Gibberella, Sphaceloma, and Alternaria. Aspergillus and Cladosporium were also found dominant in the human gastrointestinal tract (Lai, Tan, & Pavelka, 2019), but the other dominant genera were different between tigers and other vertebrates. For example, Candida was the most abundant genus found in dogs, humans, and phytophagous bats (Foster et al., 2013;Lai et al., 2019), but it was not found in the tiger gut. This difference may be associated with the high protein content in the tiger diet, because Candida was negatively correlated with diets high in amino acids, proteins, and fatty acids, but was positively correlated with diets high in carbohydrates .
Similar to bacterial communities, beta diversity of the fungal microbiota was not significantly influenced at the subspecies level.
However, alpha diversity of the fungal microbiota was significantly different among the subspecies analyzed. Previous studies have shown that changes in diet can regulate mycobiota diversity in humans and bats (Hallen-Adams & Suhr, 2016;Li et al., 2018). In contrast, no significant differences in intestinal fungal beta diversity were found between meat-fed and mix-fed tigers. Remarkable similarities in intestinal fungal diversity between diets were observed in laboratory-reared zebrafish (Siriyappagouder et al., 2018). However, the similarity of intestinal fungal beta diversity in tigers might have resulted from the similarity of diets between meat and mix groups. Significant differences in intestinal fungal alpha diversity were found. In addition, for seven out of eight samples from milk-fed tigers, we failed to amplify ITS1 sequences, which implied that milk-fed tigers had fewer intestinal fungi than meat-fed or mix-fed tigers. Regarding the effect of age, which is inconsistent with research on the human gut mycobiota, which has shown significant differences with respect to age and gender (Strati et al., 2016), the tiger gut mycobiota was not significantly influenced by age. In contrast to the intestinal bacterial community of universal character, changes in the intestinal fungal community may vary with species. Future research needs to further explore the variation in gut mycobiota in other animals.
More than one-third of the fungal species present in the tiger gut were saprotrophic ( Figure 11). Saprotrophic fungi decompose and redistribute nutrients (Hättenschwiler, Tiunov, & Scheu, 2005) and could therefore provide benefits to a growing host. In addition, commensal fungi in the gut may have potential health benefits or probiotic effects, such as those provided by yeasts (Hatoum, Labrie, & Fliss, 2012). Although all tigers included were apparently healthy, approximately 9.83% of intestinal fungi were categorized as pathogenic guilds. These fungi may serve as opportunistic pathogens, especially the genera that were cataloged into animal pathogens and detected in only a few samples. For example, Purpureocillium lilacinum has been reported to cause infections in both immunocompromised and immunocompetent humans (Saghrouni et al., 2013). Engyodontium album causes diseases in bovines and humans (Balasingham et al., 2011;Thamke, Mendiratta, Dhabarde, & Shukla, 2015). Moreover, a high frequency of azole resistance has been found in fungal isolates from human guts (Strati et al., 2016). The intestine appears to serve as a reservoir of opportunistic pathogens in plants and animals.
The role and pathogenic potential of intestinal fungi need further investigation.

| CON CLUS ION
Overall, this study provides comprehensive information on the gut microbiota of tigers and further describes the different impacts of host intraspecies genetic variation, diet, and age on gut bacterial and fungal communities. showed no significant differences in beta diversity among subspecies and diet or age groups. However, significant differences were observed in alpha diversity of the gut mycobiota among different subspecies and diets.
Our findings expand the current understanding of the gut microbiota in tigers and even in vertebrates, and these findings provide a reference for the healthy management of tigers by modulating the gut microbiota. Moreover, this baseline information will be crucial for future studies on wildlife mycobiome.

ACK N OWLED G M ENTS
The authors thank Prof.

CO N FLI C T O F I NTE R E S T S
None declared.

E TH I C S S TATEM ENT
This study was carried out following the recommendations of the animal