Exploration of host-gut microbiome dynamics of wild hosts is important to better understand the flexibility, resilience and long-term associations of symbiotic interactions under dietary changes. By examining the flexibility and resilience of wild omnivorous passerine bird gut microbiomes through diet manipulation, our findings document a rapid and significant impact of diet, aligning with other studies on passerine birds [51, 52]. We observed a significant deviation in microbial community structure from the initial gut microbiomes after diet manipulation, but we found no significant differences in gut communities between the three diet groups (Fig. 3). Despite this, there were noteworthy diet-associated trends in the bacterial community variation and differentially abundant bacterial genera. Furthermore, gut microbiomes partially recovered after the diet reversal, implying that the association between P. major and their gut communities is somewhat resilient to diet-induced changes.
P. major gut microbiomes respond to diet changes, but not as flexibly as predicted
The deviation of gut microbial communities from the initial diet to three different diet groups exemplifies the ability of P. major to respond to dietary changes. However, the magnitude of the microbial response varied among diets, suggesting that responses occur according to both the physical (ratios of different diet components) and the macronutrient compositions. Although the macronutrient content of standard, mixed and seed diets did not differ markedly, differences in gut microbial communities indicate that the physical content of diets can impact the gut microbiomes (Figs. 3 and 4). A diet that is both compositionally and macronutritionally different, such as the mealworm diet (Fig. 1a and 3c), has the strongest and most consistent impact on microbiome structure (Figs. 3a and S4).
The differences in individual variation in gut microbiomes between seed and mealworm diets provided some intriguing insights into gut microbiome dynamics (Figure S4). Microbiomes of birds fed with mealworms experienced significantly lower individual variation than birds fed with a standard or a seed diet (Figure S4). The reduced individual variation in insectivores has been demonstrated in wild birds [14] and in another diet manipulation study of P. major [52], suggesting that a strictly insectivorous diet (mealworms) might be associated with a specific and more narrow set of gut microbes [58] compared to the seed diet. This is potentially associated with more homogeneous nutrient availability in the mealworm diet [58] compared to the more heterogeneous nutrient availability from multiple types of seeds (e.g., sunflower seeds [59], millet [60], and wheat [61]). Additionally, the taxonomic diversity of initial gut microbiomes of P. major in this study differed markedly from the other diet manipulation study [52], where Proteobacteria dominated the gut microbiomes, indicating regional differences. However, the observed similarity in the overall community-level response of gut microbiomes for individuals on seed or insect diets in the two studies, suggest that despite the regional differences of P. major gut microbiomes, microbial communities respond in similar ways to similar dietary changes. The microbes responding to these differences may depend on the taxonomic composition of the starting microbiome. This species-level regional variation in gut microbiomes and the flexible nature of passerine microbiomes to dietary changes further support the lack of [16, 20] or weak [62] association between bird gut microbiome structure and host phylogeny.
We also documented significant diet-specific responses of a few bacterial genera in different diet groups (Figs. 1, 3 and Table S6). The seed diet led to a significant increase in relative abundance of the genera Fusobacterium (Fusobacteria), Blautia (Firmicutes), and Ruminococcus (Firmicutes). Some Fusobacterium members are animal pathogens [63], but their consistent presence in wild bird guts [2, 14, 64, 65], especially in herbivorous species [64–66], suggests a possible symbiotic role. Bacteria from the genus Blautia may facilitate the metabolism of plant secondary metabolites [67, 68], consistent with an increase in the relative abundance of this genus in birds feeding on seeds. Ruminococcus bacteria (e.g., the R. gauvrauii and R. gnavus groups) that increased significantly with seed diet are within the family Lachnospiraceae and their possible functions are poorly resolved [69]. However, the Ruminococcus gnavus group has been proposed to degrade mucins [69], and the mucin levels in digestive tracts tend to increase on plant-based diets with high fibre contents [70]. These bacteria may thus be opportunistic and utilize the increased amount of mucins.
Birds on the mealworm diet experienced a significant increase in the relative abundance of e.g., Rombutisia and Akkermansia that are generally presumed to be associated with protein metabolism [71]. Relative increases in the Firmicutes orders Lactobacillales, Bacillales and Clostridiales (Table S6.5) in birds on the mealworm diet are consistent with a previous study on insectivorous passerines [14]. Most of these lineages are believed to play roles related to protein fermentation and degradation of toxic by-products from protein metabolism [68, 72–74], thus enabling hosts to sustain an insect diet. Furthermore, a study has demonstrated the prebiotic effects of mealworms on mice gut microbiomes, where mealworm exuviae led to an increase in lactic acid bacteria (e.g., family Lactobacillaceae) [75], similar to what we observe in P. major fed mealworms (Figs. 1a, 3c and 4). However, understanding how macronutrient content affects wild bird gut microbiome lineages and their associated functional roles, requires further studies with nutritionally markedly different diets and data to decipher their association with gut microbial processes.
Although we cannot rule out that foodborne microbes could impact gut microbial community structures, as has been shown for a small fraction of lactic-acid bacteria from fermented foods in humans [26], we find this unlikely to be a main driver of our results. We did observe new bacterial ASVs in the three diet groups and reversed microbiomes compared to initial communities (Figures S1 and S2), suggesting that new lineages colonise bird gut microbiomes along with dietary changes. However, the most abundant ASVs observed in manipulated gut microbiomes were shared with the initial communities (Table S1 and Figure S2), supporting that major changes in gut microbiomes are accounted for by pre-existing community members. Previous studies on wild birds have demonstrated low bacterial diversity and community composition differences in the midgut region (stomach and small intestine) compared to the crop microbiota. This suggests that the highly-acidic conditions in the midgut region acts as a barrier for environmental and foodborne bacteria [14, 66, 76, 77]. This is further evident when we look at previously published mealworm gut microbiomes [78], where only a small fraction of the microbiome consists of bacterial genera that we identified in the gut microbiomes of mealworm-fed P. major. Furthermore, cloacal swabs appear to adequately capture the microbial diversity of entire digestive tracts [79], suggesting that the observed differences in the cloacal microbial communities are unlikely to be driven by foodborne microbes.
Wild-bird gut microbiomes are partially resilient to diet-induced changes
For the first time, we illustrate the resilience of wild bird gut microbiomes to diet changes. However, the recovered bacterial communities did not completely mirror the initial gut communities, despite the recuperated bacterial diversity. There were differences in relative abundances of dominant bacterial genera compared to initial community compositions, and ASV richness was significantly lower in reversed than initial gut communities (Figs. 2, 4 and S1). Intriguingly, we found a consistent pattern that the majority of the taxa that significantly increased in abundance during specific diet treatments decreased again when birds were returned to the standard diet, while taxa that decreased significantly during the diet manipulation rarely recovered following diet reversal (Table S6, Fig. 4).
Partial recovery suggests that ecologically relevant diet changes render available gut niches, that are subsequently filled by either the original microbes filling these niches or that niches are replaced by novel symbiont lineages. Lineages filling particular niches may have the opportunity to flourish by outcompeting bacteria playing similar roles due to functional redundancies inherent in complex gut communities [68, 80, 81]. To understand whether this indeed plays a role, exploration of microbial gene expression between initial and reversed gut bacterial communities will be needed. Our results also demonstrate that P. major microbiomes are resilient to natural dietary changes that individuals might experience due to seasonality (e.g., seed-dominated diet during winter and insect-dominated diet in the spring) [36], but these recovery trajectories can be impacted by the competition among bacterial lineages that fill the same functional niches. A partial restoration of gut bacterial communities after dietary changes imply that natural diet changes could lead to microbial symbiont losses (Figure S1), further suggesting that the nature of digestive tract microbiotas in wild birds is plastic to some extent. Hosts may utilize other mechanisms, such as acquiring bacteria from the environment [40] or hosting bacteria with similar metabolic capacities [68, 80] to maintain stability after disruptions of their symbiotic associations. The flexible yet resilient nature of avian gut microbiomes may thus provide an additional level of plasticity for bird hosts to cope with natural (i.e., seasonality, migration, and interspecific competition) [32–39] and anthropogenic (i.e., habitat degradation and invasive species) [82, 83] dietary fluctuations, as predicted by many avian gut microbiome studies [28, 30, 40–49].