Surveys of eleven species of wild and zoo birds and feeding experiments in white-tailed eagles reveal differences in the composition of the avian gut microbiome based on dietary habits between and within species (2025)

Keywords: avian, dietary habitat, microbiome, next-generation sequencing, white-tailed eagle

Animals and fecal sample collection

Experiment 1: Comparison between bird species: Feces were collected fromIndian peafowls (Pavo cristatus), domestic geese (Ansercygnoides), ducks (Anas platyrhynchos domesticus), ostriches(Struthio camelus), night herons (Nycticoraxnycticorax), Humboldt penguins (Spheniscus humboldti), andgreat horned owls (Bubo virginianus) kept at the Maruyama Zoo inHokkaido, Japan (Table1). Feces were collected using autoclaved spatulas and enclosed in 15 mL or 30mL sample tubes.

Feces were collected from three WTEs and one falcon (Falco peregrinus)housed at the Kushiro Shitsugen Wildlife Center, Japan, which were fed frozen meat (Table 1). The falcon and WTEs feces were collectedusing autoclaved disposable chopsticks and placed in 30 mL sample tubes. Fecal samplescollected from WTEs on their final feeding day were used for this experiment.

Feces were collected from black-tailed gulls (Larus crassirostris) andslaty-backed gulls (Larus schistisagus) on Rishiri Island, Japan (45°18’N, 141°24’ E) (Table 1). The high-fluidityfeces were collected using a pipette with a 1 mL pipette tip and placed in a 30 mL sampletube. The edge of the tip was cut to facilitate collection of small fragments in the fecalsample.

The ground of the animal house was basically concrete and was carefully cleaned beforethe study began. After feces samples were collected, they were cleaned in the same mannerso that fresh feces could always be collected. For outdoor feces, fresh feces wereselected as much as possible and collected so as not to be contaminated by environmentalsoil. All sample tubes were stored in liquid nitrogen for transfer to the laboratory. Uponarrival, samples were stored in a −80°C freezer until bacterial genomes wereextracted.

Experiment 2: WTEs feeding experiment: Animal experiments were performedat the Kushiro Shitsugen Wildlife Center under supervision and with the endorsement of theInstitutional Animal Care and Use Committee of Hokkaido University, Japan. Individual WTEswere housed separately in outdoor individual breeding huts, and were fed a specific dietfor approximately one month (Fig.1). Control samples were obtained from WTEs (n=7) before dietarytreatment, and WTEs were fed a diet of both meat and fish. Excluding one individual, WTEs(n=6) were assigned into either the meat feeding group (meat-diet group, n=3) or the fishfeeding group (fish-diet group, n=3), and fed thawed meat or fish (approximately 10types). Feeding and fecal collection took place at around 9 a.m. daily, for about a month.The reason for month-long duration of the experiment was to verify any compositionalfluctuations in the gut microbiome. According to David et al. [7], the diversity of bacterial flora in humanstransitioning from an ordinary diet to a plant- or animal-based diet changed in only twodays. However, as no published research has examined such compositional changes in birdsover time, it was difficult to predict how long it would take for the gut microbiome toadapt to diets limited to either meat or fish. Therefore, the WTEs were treated with alimited diet for about one month in this experiment.

In addition, the WTEs were not force-fed, so the amount of food that each eagle consumedvaried daily. On some days, the eagles did not spontaneously ingest the food, while onothers, they were not fed to ensure their health. In addition, we observed a fastingperiod, which lasted several days after the experiment commenced, in each individual. Toidentify the influence more clearly, we swapped the diet of two individuals (IDs: meat2,fish2) on the 33rd and 34th days after commencing the experiment (i.e., meat2 was givenfish, and fish2 was given meat). The ground of the animal house was basically concrete andwas carefully cleaned before the study began. After feces samples were collected, theywere cleaned in the same manner so that fresh feces could always be collected. Feces werecollected daily, using autoclaved disposable chopsticks, placed in 30 mL sample tubes, andstored in liquid nitrogen until they were transferred to the laboratory. After transfer tothe laboratory, samples were kept in a −80°C freezer.

Although feces were collected daily, the number of fecal samples with sufficient qualityfor sequencing was limited, and only 119 out of the 185 samples collected were used forthe bacterial flora analysis. Furthermore, because sampling dates were inconsistent acrossindividuals, the data were sorted into five-day averages for comparison. Term 0 includesthe days until meat-only or fish-only treatment, and one of the Term 0 samples was used asthe control group for each individual. Term 7 consists of the 33rd and 34th days afterswapping the diets of individuals fish2 and meat2.

Next-generation sequencing

Next-generation sequencing (NGS) was performed as described below. Briefly, a QIAamp FASTDNA Stool Mini Kit (QIAGEN, Nordrhein-Westfalen, Germany) was used to extract bacterialDNA from 150–200 mg fecal samples, according to the manufacturer’s instruction. A Qubit ™4 Fluorometer (Invitrogen™, Carlsbad, CA, USA) was used to measure the DNA concentration.PrimeSTAR^® Max DNA Polymerase (TaKaRa Inc., Kusatsu, Japan) and followingprimers (final concentration 3 µM) were used to amplify the V4 region of 16S ribosomal RNA(rRNA): F515, 5′-GTGCCAGCMGCCGCGGTAA-3′ and R806, 5′-GGACTACVSGGGTATCTAAT-3′ [5]. PCR was performed using a thermal cycler (Life ECO,BIOER, Zhejiang, China), with the following settings: 98°C for 10 sec, 54°C for 5 sec, and72°C for 5 sec, repeated for 30 cycles. We confirmed that the band of the first PCRproduct was observed at ⁓350 bp. Purification of PCR products was performed using AMPureXP (BECKMAN COULTER, Brea, CA, USA). A second PCR was also performed withPrimeSTAR^® Max DNA Polymerase and barcode primers (forward primer,IonA-barcode[i]-F515; reverse primer, ionP1-F806; final concentration 3 µM) on the thermalcycler, with the following settings 98°C for 10 sec, 54°C for 5 sec, and 72°C for 5 sec,repeated for five cycles. We confirmed that the products of the second PCR contained a⁓400 bp band. Purification and concentration measurements were performed for the productsof the second PCR, using the same procedure described for the first PCR. In order toadjust the concentration of each sample more accurately, the concentration of severalsamples was measured using the BioAnalyzer Agilent 2100 (Agilent Technologies, SantaClara, CA, USA), with LabChip from the DNA 1000 kit, according to the protocol from themanufacturer’s website(https://www.chem-agilent.com/pdf/BioA_SII_DNA_v04_05_20160830.pdf).

Library preparation was performed using an Ion Chef™ Instrument and Ion PGM™ Hi-Q™ ViewChef 400 Kit (Thermo Fisher Scientific, Waltham, MA, USA). To compare sampleconcentrations against those that were measured in Qubit, a calibration curve wasgenerated using samples whose concentrations were measured using the Bioanalyzer. AllQubit-measured sample concentrations were corrected using the calibration curve. Equalmolar amounts from each sample were used to prepare the sample mix, which was diluted to50 pmol/L.

An Ion PGM™ System (Thermo Fisher Scientific Inc.) was used for DNA sequencing. Theoutput data were sent to Torrent Browser (supplied by Thermo Fisher Scientific Inc.),verified, and downloaded as compressed files, which were primarily used for analysis withQiime (1.9.1, built in Bio-Linux-8.0.7 [6]).

Data analysis was performed primarily using Qiime, R (386 3.4.1 using RStudio), and JMP(Pro 14, SAS Institute Inc., Cary, NC, USA). For Qiime, terminal commands were executed tocreate an operational taxonomic unit (OTU) table, and to perform α-rarefaction andβ-diversity analyses (weighted and unweighted UniFrac analyses). Details of the analysisprocedures for experiment 1 and 2 are described below.

Experiment 1: Comparison between bird species: Steel-Dwass tests wereconducted in R, using the NSM3 package, to identify significant differences in the gutmicrobiome ratios among the groups of each diet habit category. A P-valuebelow 0.05 was considered to indicate a significant difference. Principal coordinateanalysis (PCoA) was also performed on the data from the β-diversity analysis using R.

The Shannon diversity index was calculated from the OTU table, using the “MASS” and“vegan” packages in R. The samples were categorized by diet, and Steel-Dwass tests wereperformed to detect differences.

Experiment 2: WTE feeding experiments: In order to investigatefluctuations in the gut microbiome, samples were categorized by each individual,normalized using the “DESeq 2” package, and compared using the OTU count number in R. Whena significant difference was observed for the OTU, the composition ratio for eachindividual was calculated for each term, to confirm fluctuations. Steel-Dwass tests werealso conducted to identify significant differences in the ratio of gut microbiomecomposition between the control, the meat-diet group, and the fish-diet group in R, usingthe “NSM3” package; P<0.05 was considered significant. Graphs of thegut microbiome composition for each individual were created in JMP.

To identify changes before and after completion of the diet treatment, we first used datafrom all seven individuals before treatment as the control, including one individual thatwas not assigned to either the meat-diet group or the fish-diet group. Next, the samplesthat were collected during the final phase of the feeding period that had the highest OTUcount in each individual were identified. The average of these samples from the two dietgroups was calculated. Term 7 samples were excluded. Composition graphs were created atthe phylum, class, order, and family levels in JMP.

Comparison between avian species (Experiment 1)

A) Diversity: α-Rarefaction analysis:Figure 2 shows the α-rarefaction analysis after grouping by diet. The blue, green, and brownlines reflect the fish, grain, and meat-diet groups, respectively. The fish- and meat-dietgroups had similar trend of represents of OTUs. Whereas, the grain-diet group showedgreater represents of OTUs.

Shannon index: Because rarefaction analysis showed that the gutmicrobiome in herbivorous birds was most diverse, we calculated the Shannon index usingthe OTU table (Fig. 3) and found a significant difference (P=0.0017) between thegrain-diet and the fish-diet groups. However, there was no significant difference betweenthe grain-diet and the meat-diet group (P=0.0853).

Unweighted UniFrac analysis: An unweighted UniFrac analysis wasconducted and plotted using PCoA, as shown in Fig.4. The blue, green, and brown points denote the fish-, grain-, and the meat-dietgroups, respectively. The PCoA 1 values differed between the grain-diet group and thefish- and meat-diet groups; the grain-diet group had negative PCoA 1 values, whereas thefish- and meat-diet groups showed near-positive values. No differences between PCoA 2values were observed between groups, although the plot position of each sample showedrough trends based on diet. Furthermore, individuals of the same species were plotted atfairly homologous positions, depending on the host species and diet contents.

B) Gut microbiome composition: Steel-Dwass tests were conducted at eachbiological classification level, and the bacteria species that showed significantdifferences were extracted (Table 2). Many bacteria within the Bacteroidetes,Tenericutes, Proteobacteria, andFirmicutes phyla were significantly different between the grain-dietgroup and the fish- and meat-diet groups, and comprised a higher ratio in the OTU of thegrain-diet group.

On the other hand, there was also a significant difference in some bacteria between thefish-diet and meat-diet groups. Veillonellaceae (phylum:Firmicutes) was significantly higher in the meat-diet than thegrain-diet group (P=0.0152). In the Clostridiaceaefamily (P=0.0207), the proportion of Eubacteriaceae(P=0.0113) and Lacnospiraceae(P=0.0029) was significantly higher in the meat-diet group than in thefish-diet group.

Furthermore, Flavobacteriia were significantly higher in the fish-dietgroup than the grain-diet group (P=0.012).

WTEs feeding experiment (Experiment 2)

WTEs, which eat both meat and fish, were used in experiment 2. The WTEs were divided intotwo groups and given only deer meat or fish for approximately one month. Then, we examinedwhether there were differences in the composition of their bacterial flora.

Compositional ratios of bacterial flora per term in each WTEs: Thevariation in composition of intestinal bacterial flora for each individual at the classlevel is shown in Fig. 5. No distinctive fluctuations in overall bacterial flora composition were observedfor either diet groups at the class level, or the phylum, order, and family levels.

Compositional fluctuation: Some bacteria groups showed trends towardsrough fluctuations (Fig. 6). At the Phylum level, a large increase in the ratio ofActinobacteria in the fish-diet group was observed once, around term 3,and was followed by a decrease. Nevertheless, the ratio of Actinobacteriastill remained a high level compared with pre-treatment. On the other hand, the ratio ofActinobacteria in the meat-diet group remained the same, or decreased,when compared with pre-treatment ratios. Additionally, in term 7, the WTE individual inthe fish-diet group with the swapped diet (fish2) had a decreasedActinobacteria ratio after being fed meat. Conversely, theActinobacteria ratio for the individual in the meat-diet group given aswapped diet (meat2) increased in term 7.

At the order level, the ratio of Lactobacillales in the fish-diet groupconsistently remained at a higher level than in the meat-diet group. Conversely, in themeat-diet group, the Lactobacillales ratio tended to decrease whencompared with pre-treatment. However, the ratios of the two individuals with swapped dietsthat changed diet after Term 7 did not change much.

At the class level, the ratio of Bacilli was decreased in the meat-dietgroup. The ratio of Bacilli in the fish-diet group remained high duringthe one-month experimental period, but nevertheless decreased when compared topre-treatment ratios, except in one individual. In addition to Bacilli,Lactobacillales and Actinobacteria also showed a trendwhere their ratios increased significantly at one point and then were somewhatreduced.

Coriobacteriales was found to gradually increase in proportion in thefish-diet group during the experiment period. There was no significant change in the ratioof Coriobacteriales in the meat-diet group.

The average composition of the gut microbiome in each group: Figure 7 shows the composition of bacterial flora in the control, meat-diet, and fish-dietgroups. Different patterns of changes were observed in the fish-diet and meat-diet groupsrelative to the control group. At the family level, the ratio ofLactobacillaceae increased in the fish-diet group (17.50%) comparedwith the control group (8.78%) and the meat-diet group (0.74%). In addition, the ratio ofClostridiaceae decreased in the fish-diet group (10.40%) compared withthe control group (28.73%).

Pasteurellaceae was higher in the meat-diet group (4.66%) than in thefish-diet group (0.37%). Whereas, Veillonellaceae levels were higher inthe fish-diet group (10.52%) than in the meat-diet group (3.18%) in this study.

Comparison between avian species

As mentioned in Introduction, the gut microbiome plays essential roles in many ways, andis affected by various factors such as food habitat. In experiment 1, we tried to comparethe diversity between species. α-Rarefaction analysis revealed that the grain-diet grouphad a wider variety of gut microbiome diversity than the other two groups (Fig. 2), which is in accordance with previousstudies showing that herbivorous mammals have a high diversity of bacterial flora [15]. Herbivorous animals are thought to have a highdiversity of bacterial flora because their diet comprises large amounts of non-digestibledietary fiber. In addition, they are also exposed to plant secondary compounds, whichinclude toxic substances [9].

The calculation of Shannon index showed no significant difference between the grain-dietand the meat-diet groups (P=0.0853), whereas there was a significantdifference (P=0.0017) between the grain-diet and the fish-diet groups(Fig. 3). This is likely caused by the highShannon index for the great horned owl (3.4222), which was included in the meat-dietgroup. Generally, the cecum is long and developed in herbivorous birds, but is short ordegenerated in carnivorous birds. Therefore, it is interesting to note that owls have adeveloped caecum, despite being carnivorous [18].This may account for the high Shannon index observed in the great horned owl. In fact,when the great horned owl is excluded, the Steel-Dwass test shows a significant differencebetween the grain-diet and the meat-diet groups (P=0.0231).

From an unweighted UniFrac analysis, we clarified the rough trends based on diet (Fig. 4). Moreover, we also found the different plotpositions depending on the host species and diet contents, even belonged at same dietgroups. For example, the Indian peafowls, domestic geese, and ducks have the same diet,such as clover and oyster shells. These three species had similar bacterial flora, exceptfor one individual. Ostriches also ate similar foods as the above species, such as oystershells, but their microbiome composition differed. The ostrich may have a differentmicrobiome composition due to its large cecum and relatively long body and intestinaltract. These findings reveal the strength of the influence of the diet and host strainfactors on microbiome composition.

Gut microbiome composition differed between diet groups

As shown in Table 2, there were significantdifferences of bacteria species between the grain-diet group and the fish- and meat-dietgroups. A significant difference was observed for Bacteroidetes, whichdegrade complex biopolymers and polysaccharides, such as carbohydrates and plant cell wallcomponents, in the gastro-intestinal (GI) tract. The number ofBacteroidetes has been shown to decrease as a result of ingesting ahigh-fat diet [22]. Furthermore,Bacteroidetes primarily produce acetic and propionic acid as finalmetabolites, which could contribute to short-chain fatty acid (SCFA) production inherbivores. Firmicutes are known to be involved in SCFA production, andare especially involved in the production of butyric acid [8]. SCFA production from non-digestible dietary fiber may play an important rolein the nutrition of herbivorous birds, because the proportion of many bacterial specieswas larger in the grain-diet group. For example, Lactobacillaceae areknown to be involved in acetic acid production [11].

When focused on the difference between fish-diet and meat-diet groups, meat- diet groupsshowed higher composition than fish-diet groups, especiallyVeillonellaceae (phylum: Firmicutes),Eubacteriaceae and Lacnospiraceae (both belonged tothe Clostridiaceae family). As mentioned in the Introduction, SCFAs maybe produced by fermenting cartilage and collagen, as opposed to non-digestible dietaryfiber. The significant presence of these bacteria in the meat-diet group suggests thatthey may contribute to SCFA production in a carnivorous diet. Indeed, previous studiesindicate that Lachnospiraceae and Eubacteriaceae areknown as SCFA-producing and butyric-acid–producing bacteria,respectively [20, 21].Veillonellaceae utilizes the partially degraded products of bacterialpolysaccharides to produce acetic and propionic acid [10], and its ratio in kittens that ate moderate-protein, moderate-carbohydratefood has been shown to increase in comparison to kittens that ate high-protein,low-carbohydrate food [13]. It was concluded thatthe balance between the protein and the carbohydrate content of the diet affects the ratioof Veillonellaceae; therefore, the intestinal environment of themeat-diet group may be suitable for Veillonellaceae, compared with theintestinal environment of the grain-diet group. Additionally,Veillonellaceae likely contributes to SCFA production when theequilibrium of the intestinal flora that is characteristic of the meat-diet group ispreserved.

In the fish-diet group, Flavobacteriia was the only bacteria class thatwas present in a significantly greater proportion than in the other two diet groups.Flavobacteriia is widely present in the GI tract, soil, and aqueousenvironments, among other areas [3, 22]. Flavobacteriia also comprisesspecies that can become pathogens for birds and mammals, but it is still unknown how theirincrease influences the onset of disease.

Impact of fish diet on WTEs

The fish-diet group had a higher ratio of Lactobacillales, the order inwhich Lactobacillaceae (mentioned in experiment 1) belongs, than themeat-diet group. Since Lactobacillaceae are involved in the production ofacetic acid, this observation suggests that acetic acid production may be increased byeating fish. Indeed, the ratio of Lactobacillaceae rose sharply whenchanging the meat diet of the meat2 individual to fish in term 7. Therefore, it can besaid that Lactobacillales fluctuated due to the influence of the fishdiet. Moreover, the same is true for Actinobacteria at the Phylum level.Actinobacteria include Bifidobacteriaceae, whichproduces a large amount of acetic and lactic acid. Lactic acid is very important becauseit is converted to butyric acid [4]. At the familylevel, the average ratio of Bifidobacteriaceae in the fish-diet group(1.36%) was higher than in the control (0.27%) or meat-diet groups (0.36%; Fig. 7d).

The increase in the ratio of Actinobacteria in the swapped-dietindividual, meat2, in term 7 was likely caused by the change to the fish diet.Furthermore, in term 7, there was no change in the ratio of Bacilli inthe fish2 individual, but the composition ratio in the swapped-diet individual in themeat-diet group (meat2) rose. This also indicates that compounds in the fish diet possiblypromote Bacilli growth. There was no change in the ratios observed in themeat2 individual after the diet swap in term 7. However, theCoriobacteriales ratio was greatly reduced in the fish2 individual whenit was given meat in term 7; therefore, it can be concluded thatCoriobacteriales increased as a result of consuming fish. Much of theinfluence of Coriobacteriales is unknown, so it is difficult to speculateas to what such an increase means. Thus, we can only surmise that the fish-based dietprovides suitable conditions for the growth of this bacterium.

Convergence of fluctuations in gut microbiome constituents

According to Fig. 6, we found some fluctuationtrends in the ratio of Bacilli, and Lactobacillales andActinobacteria as well. Considering these results, the intestinalmicrobial flora of WTEs may be more suited for a meat diet. If this hypothesis is correct,the meat diet may not cause sensitive fluctuations in bacterial flora, whereas thecontinuous feeding of fish, to which the intestinal flora do not adapt, could cause clearfluctuations. In this study, a continuous fish-only diet may have resulted in the gutmicrobiome of the fish-diet group adapting to their diet in around 10–20 days.Furthermore, as the fish-processing capacity of the bacterial flora gradually increased,the ratio of the bacteria which showed an increase would gradually fall.

As the fluctuation in each intestinal bacterium species should converge, the trend of anincreasing Coriobacteriales ratio may be shown in the latter half of theone-month experimental period.

Integrated consideration of the insights from experiments 1 and 2

Through experiment 1 and 2, we could consider the diet dependent effects to WTEs. Asmentioned in Result, there were different patterns of changes in the fish-diet andmeat-diet groups relative to the control group (Fig.7).

The tendency of increase of Lactobacillaceae in the fish-diet groupcompared with the control group and the meat-diet group was observed both in experiment 1and 2 (Table 2, Fig. 7d). Furthermore, we also got the tendency of decrease ofClostridiaceae in the fish-diet group compared with the control group(Table 2, Fig. 7d). Similarly, the ratio of Eubacteriaceae was alsosignificantly lower in the fish-diet group compared to the meat-diet group in experiment 1(Table 2). Therefore, the observed specificdiets had similar effects on gut microbiome composition in multiple avian species, werefurther confirmed among an allogeneic species.

Some bacteria species in the meat-diet group in experiment 2 showed characteristicchanges that were not found in experiment 1. For example, in experiment 2,Pasteurellaceae was higher in the meat-diet group than in the fish-dietgroup (Fig. 7d). Dong et al.[10] examined the relationship between fecalmicrobiota and SCFA concentration in children with or without cow milk protein allergies.They found that Pasteurellaceae was correlated with the concentration ofpropionic acid, suggesting that the levels of propionic acid production may be high in themeat-diet group. Indeed, in experiment 2, Pseudomonadaceae was found inhigher levels in the meat-diet group compared with the fish-diet group. ThePseudomonadaceae family has been found to cause infectious diseases insome birds [14]. Likewise, Pseudomonadaceaeis related to inflammatory enteritis in humans, and its toxins cause damage toepithelial cells. It remains unclear how the composition of the Pseudomonadaceae familyincreases, and whether such an increase has any meaningful effects.

In experiment 1, the average Veillonellaceae ratios in the meat and thefish-diet groups were almost the same, but this ratio was higher in the fish-diet groupthan in the grain-diet group (Table 2).Whereas, in experiment 2, the levels of Veillonellaceae differed betweenin the fish-diet group and the meat-diet group (Fig.7d). As described in experiment 1, it is likely that the ratio of protein andcarbohydrate content of the dietary components influences Veillonellaceaeproportion. Accordingly, it follows that the intestinal environment that is generated by afish diet appears to be most suitable for Veillonellaceae.Megasphaera, which belongs to the Veillonellaceaefamily, is a major butyric acid-producing bacterium; therefore, the prosperity of theVeillonellaceae family is an important factor in identifying influenceson the intestinal barrier [2].

Limitations of experiment 2

There were some differences in the gut microbiome composition between the control groupand the two diet groups, but these were not significant. Two factors may have contributedto these observations. First, the number of individuals in each group was small, andfluctuation strength varied depending on differences in individual bacterial flora. Thus,identifying significant differences could be difficult. Second, the WTEs were notforce-fed, so the amount of food that they consumed varied daily. This could account forthe fact that similar trends in fluctuation were not observed for each period. Therefore,it is necessary to repeat the research to increase the number of samples collected andverify these observations. In addition, it is important to optimize the experimentalmethod, for example by ensuring uniformity in the amount of food consumed.

Conclusion

We investigated influences of diet on the composition of the gut microbiome using thefeces of bird species. The gut microbiome compositions of the meat and the fish-dietgroups were similar when compared with the grain-diet group. The grain-diet group hadhigher gut microbiome diversity than the meat- and fish-diet group. The composition of theintestinal microflora differed between the fish and meat diet groups, even for the sameanimal species, and an increase in bacterial composition specific to the meat diet groupwas observed, indicating that the composition of the intestinal microflora differeddepending on the diet.

• 1.Adak A, Khan MR. 2019. An insight into gut microbiota and itsfunctionalities. Cell Mol LifeSci 76: 473–493. doi:10.1007/s00018-018-2943-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 2.Bermingham EN, Young W, Kittelmann S, Kerr KR, Swanson KS, Roy NC, Thomas DG. 2013. Dietary format alters fecal bacterialpopulations in the domestic cat (Felis catus).MicrobiologyOpen 2:173–181. doi: 10.1002/mbo3.60 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 3.Bernardet JF, Bowman JP. 2013. International committee on systematicsof prokaryotes subcommittee on the taxonomy of Flavobacterium and Cytophaga-likebacteria. Int J Syst EvolMicrobiol 63: 2752–2754.doi: 10.1099/ijs.0.053926-0 [DOI] [PubMed] [Google Scholar]
• 4.Binda C, Lopetuso LR, Rizzatti G, Gibiino G, Cennamo V, Gasbarrini A. 2018. Actinobacteria: a relevant minority forthe maintenance of gut homeostasis. Dig LiverDis 50: 421–428. doi:10.1016/j.dld.2018.02.012 [DOI] [PubMed] [Google Scholar]
• 5.Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Lozupone CA, Turnbaugh PJ, Fierer N, Knight R. 2011. Global patterns of 16S rRNA diversityat a depth of millions of sequences per sample. ProcNatl Acad Sci USA 108Suppl1: 4516–4522. doi: 10.1073/pnas.1000080107 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 6.Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Peña AG, Goodrich JK, Gordon JI, Huttley GA, Kelley ST, Knights D, Koenig JE, Ley RE, Lozupone CA, McDonald D, Muegge BD, Pirrung M, Reeder J, Sevinsky JR, Turnbaugh PJ, Walters WA, Widmann J, Yatsunenko T, Zaneveld J, Knight R. 2010. QIIME allows analysis ofhigh-throughput community sequencing data. NatMethods 7: 335–336.doi: 10.1038/nmeth.f.303 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 7.David LA, Maurice CF, Carmody RN, Gootenberg DB, Button JE, Wolfe BE, Ling AV, Devlin AS, Varma Y, Fischbach MA, Biddinger SB, Dutton RJ, Turnbaugh PJ. 2014. Diet rapidly and reproducibly altersthe human gut microbiome.Nature 505:559–563. doi: 10.1038/nature12820 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 8.den Besten G, van Eunen K, Groen AK, Venema K, Reijngoud DJ, Bakker BM. 2013. The role of short-chain fatty acids inthe interplay between diet, gut microbiota, and host energy metabolism.J Lipid Res 54:2325–2340. doi: 10.1194/jlr.R036012 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 9.Dearing MD, Foley WJ, McLean S. 2005. The influence of plant secondarymetabolites on the nutritional ecology of herbivorous terrestrialvertebrates. Annu Rev Ecol EvolSyst 36: 169–185. doi:10.1146/annurev.ecolsys.36.102003.152617 [DOI] [Google Scholar]
• 10.Dong Y, Fei P, Han Y, Guo L. 2018. Characterization of fecal microbiota,short-chain fatty acids and lactic acid concentrations in 5–8-year-old children with cowprotein allergy. Iran JPediatr 28: e64638. doi: 10.5812/ijp.64638 [DOI] [Google Scholar]
• 11.Gamage HKAH, Tetu SG, Chong RWW, Ashton J, Packer NH, Paulsen IT. 2017. Cereal products derived from wheat,sorghum, rice and oats alter the infant gut microbiota in vitro.Sci Rep 7: 14312. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 12.Grond K, Sandercock BK, Jumpponen AJ, Zeglin LH. 2018. The avian gut microbiota: community,physiology and function in wild birds. J AvianBiol 56:218–227. [Google Scholar]
• 13.Hooda S, Vester Boler BM, Kerr KR, Dowd SE, Swanson KS. 2013. The gut microbiome of kittens isaffected by dietary protein:carbohydrate ratio and associated with blood metabolite andhormone concentrations. Br JNutr 109: 1637–1646.doi: 10.1017/S0007114512003479 [DOI] [PubMed] [Google Scholar]
• 14.Kojima A. 2010. Encyclopesia of Companion Bird Diseases. SeibundoShinkosha Inc., Tokyo, Japan (in Japanese). [Google Scholar]
• 15.Ley RE, Hamady M, Lozupone C, Turnbaugh PJ, Ramey RR, Bircher JS, Schlegel ML, Tucker TA, Schrenzel MD, Knight R, Gordon JI. 2008. Evolution of mammals and their gutmicrobes. Science 320:1647–1651. doi: 10.1126/science.1155725 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 16.Levey DJ, Cipollini ML. 1996. Is most D-glucose absorbed passively inNorthern Bobwhite? Comp BiochemPhysiol 113A: 225–231.doi: 10.1016/0300-9629(95)02046-2 [DOI] [Google Scholar]
• 17.Li X, Bi R, Xiao K, Roy A, Zhang Z, Chen X, Peng J, Wang R, Yang R, Shen X, Irwin DM, Shen Y. 2022. Hen raising helps chicks establish gutmicrobiota in their early life and improve microbiota stability after H9N2challenge.Microbiome 10: 14. doi:10.1186/s40168-021-01200-z [DOI] [PMC free article] [PubMed] [Google Scholar]
• 18.Meyer W, Hellmann AN, Kummerfeld N. 2009. Demonstration of calcium transportmarkers in the ceca of owls (Aves: Strigiformes), with remarks on basic cecastructure. Eur J WildlRes 55: 91–96. doi:10.1007/s10344-008-0221-8 [DOI] [Google Scholar]
• 19.Nishijima S, Suda W, Oshima K, Kim SW, Hirose Y, Morita H, Hattori M. 2016. The gut microbiome of healthy Japaneseand its microbial and functional uniqueness. DNARes 23: 125–133. doi:10.1093/dnares/dsw002 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 20.Romick-Rosendale LE, Haslam DB, Lane A, Denson L, Lake K, Wilkey A, Watanabe M, Bauer S, Litts B, Luebbering N, Dandoy CE, Davies SM. 2018. Antibiotic exposure and reduced shortchain fatty acid production after hematopoietic stem cell transplant.Biol Blood Marrow Transplant 24:2418–2424. doi: 10.1016/j.bbmt.2018.07.030 [DOI] [PubMed] [Google Scholar]
• 21.San-Juan-Vergara H, Zurek E, Ajami NJ, Mogollon C, Peña M, Portnoy I, Vélez JI, Cadena-Cruz C, Diaz-Olmos Y, Hurtado-Gómez L, Sanchez-Sit S, Hernández D, Urruchurtu I, Di-Ruggiero P, Guardo-García E, Torres N, Vidal-Orjuela O, Viasus D, Petrosino JF, Cervantes-Acosta G. 2018. A Lachnospiraceae-dominated bacterialsignature in the fecal microbiota of HIV-infected individuals from Colombia, SouthAmerica. Sci Rep 8:4479. doi: 10.1038/s41598-018-22629-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 22.Thomas F, Hehemann JH, Rebuffet E, Czjzek M, Michel G. 2011. Environmental and gut bacteroidetes:the food connection. FrontMicrobiol 2: 93. doi: 10.3389/fmicb.2011.00093 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 23.Wienemann T, Schmitt-Wagner D, Meuser K, Segelbacher G, Schink B, Brune A, Berthold P. 2011. The bacterial microbiota in the ceca ofCapercaillie (Tetrao urogallus) differs between wild and captive birds.Syst Appl Microbiol 34:542–551. doi: 10.1016/j.syapm.2011.06.003 [DOI] [PubMed] [Google Scholar]

Supplementary Materials