No evidence for associations between brood size, gut microbiome diversity and survival in great tit (Parus major) nestlings

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ No evidence for associations between brood size, gut microbiome diversity and survival in great tit (Parus major) nestlings © The Author(s) 2023. Published version Liukkonen, Martta; Hukkanen, Mikaela; Cossin-Sevrin, Nina; Stier, Antoine; Vesterinen, Eero; Grond, Kirsten; Ruuskanen, Suvi Liukkonen, M., Hukkanen, M., Cossin-Sevrin, N., Stier, A., Vesterinen, E., Grond, K., & Ruuskanen, S. (2023). No evidence for associations between brood size, gut microbiome diversity and survival in great tit (Parus major) nestlings. Animal Microbiome, 5, Article 19. https://doi.org/10.1186/s42523-023-00241-z 2023 Liukkonenetal. Animal Microbiome (2023) 5:19 https://doi.org/10.1186/s42523-023-00241-z RESEARCH Open Access © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. Animal Microbiome No evidence forassociations betweenbrood size, gut microbiome diversity andsurvival ingreat tit (Parus major) nestlings Martta Liukkonen1*, Mikaela Hukkanen2,3, Nina Cossin‑Sevrin2, Antoine Stier2,4,5, Eero Vesterinen2, Kirsten Grond6 and Suvi Ruuskanen1,2 Abstract Background The gut microbiome forms at an early stage, yet data on the environmental factors influencing the development of wild avian microbiomes is limited. As the gut microbiome is a vital part of organismal health, it is important to understand how it may connect to host performance. The early studies with wild gut microbiome have shown that the rearing environment may be of importance in gut microbiome formation, yet the results vary across taxa, and the effects of specific environmental factors have not been characterized. Here, wild great tit (Parus major) broods were manipulated to either reduce or enlarge the original brood soon after hatching. We investigated if brood size was associated with nestling bacterial gut microbiome, and whether gut microbiome diversity predicted survival. Fecal samples were collected at mid‑nestling stage and sequenced with the 16S rRNA gene amplicon sequencing, and nestling growth and survival were measured. Results Gut microbiome diversity showed high variation between individuals, but this variation was not significantly explained by brood size or body mass. Additionally, we did not find a significant effect of brood size on body mass or gut microbiome composition. We also demonstrated that early handling had no impact on nestling performance or gut microbiome. Furthermore, we found no significant association between gut microbiome diversity and short‑term (survival to fledging) or mid‑term (apparent juvenile) survival. Conclusions We found no clear association between early‑life environment, offspring condition and gut microbi‑ ome. This suggests that brood size is not a significantly contributing factor to great tit nestling condition, and that other environmental and genetic factors may be more strongly linked to offspring condition and gut microbiome. Future studies should expand into other early‑life environmental factors e.g., diet composition and quality, and paren‑ tal influences. Keywords Avian microbiome, Brood size, Gut microbiome, Parus major, 16S rRNA gene Introduction The digestive tract hosts a large community of different microorganisms (i.e., gut microbiome) and is known to be a fundamental part of organismal health and a powerful proximate mechanism affecting host performance [1, 2]. The gut microbiome has been studied across a wide range of animal taxa e.g., humans [3–5], fish [6], and economically important species such as poultry [7], and data from wild populations is slowly increasing [8]. Generally, a more diverse gut microbiome is considered beneficial for individual health [9], but there are also community structure effects that define the functionality [10]. For example, laboratory-bred mice *Correspondence: Martta Liukkonen martta.a.liukk[email protected] Full list of author information is available at the end of the article Page 2 of 16 Liukkonenetal. Animal Microbiome (2023) 5:19 with a less diverse gut microbiome have a substantially lower chance of surviving an influenza infection compared to their wild counterparts unless receiving a gut microbiota transplant from their wild counterparts [11, 12]. Moreover, gut microbiome had been linked to host fitness and survival in the Seychelles warbler (Acrocephalus sechellensis). Individuals that harbored opportunistic pathogens, i.e., microbes that usually do not cause disease in healthy individuals,but may become harmful in individuals that are immunocompromised, in their gut microbiome showed higher mortality [13, 14]. Therefore, understanding how gut microbiome affects fitness within and between individuals is necessary for not only understanding species survival but also evolution [15–17]. Gut microbiome forms at a young age and remains somewhat stable in adulthood as found for example in laboratory bred mice [18–20]. Disruption in the gut microbiome that leads to a microbiome imbalance at a young age could result in both short-term and longterm changes in the gut microbiome [21, 22]. Of the environmental effects, diet [23], including e.g., macronutrient balance (carbohydrates, fats, amino acids) [3, 24] has been concluded to be major determinants of rat and mouse gut microbiome, and this effect has recently been seen in avian models as well [25–28]. Moreover, macronutrient balance has been linked to intestinal microbiome composition [3, 24] and the functioning of individual immune response [29, 30]. However, as a large part of the prior research has focused strictly on humans or species living in controlled environments in which environmental effects on both the microbiome and host are sidelined [31, 32], many species, including most birds [8], have only started to attract attention [33]. The mechanisms of bacterial colonization of the bird gut are somewhat unique as avian life-histories differ significantly from those of e.g., mammals [34]. In mammals, the offspring are exposed to bacterial colonization during vaginal birth [35] and lactation [36, 37], whereas bird hatchlings are exposed to bacteria first upon hatching [20, 38]. Few studies have investigated the possibility of bacterial colonization in ovo, but results are still lacking [39]. Genetics [40–42] as well as the post-hatch environment [20, 43–46] have a significant effect on the formation of the avian gut microbiome. Once hatched, most altricial birds feed their young, which exposes the hatchlings to various bacteria that originate from the parents i.e., via vertical transmission [47]. It has also been shown that environmental factors are major contributors in the formation of gut microbiome [48–51], one of these being the rearing environment in the nest [44]. As early-life environment is connected to the establishment of gut microbiome, brood size may affect gut microbiome [52]. Brood size is often associated with parents’ performance and ability to feed their young [53], and the trade-off between offspring quality and quantity has been studied widely [54, 55]. Food quantity per nestling can decrease in enlarged broods, as parents may not be able to fully compensate for the additional amount of food an enlarged brood requires [56, 57]. For example, in great tits (Parus major) it has been shown that nestlings from reduced broods may have a higher body mass [58] and tend to survive better [59]. Importantly, great tit nestling body mass has been connected to gut microbiome diversity and composition: body mass positively correlates with gut microbiome richness [52]. This could imply that good physiological condition and high food availability would allow the host to have a diverse gut microbiome that promotes a healthy gut. Alterations in early-life gut microbiome could have long-term consequences on individual performance [60], yet such effects have rarely been studied in wild organisms. In wild birds, some bacterial taxa have been linked to better survival. For example, a high abundance of bacteria in the order Lactobacillales of the phylum Firmicutes is related to higher individual fitness in Seychelles warblers [14] and great tits [61]. These bacteria are also known for the benefits for bird health in economically important species such as poultry, in which Lactobacilli are used as probiotics to boost immune functioning [62]. Besides Lactobacillales, gut bacteria belonging to other genera such as Clostridium and Streptococcus are important for the degradation of non-starch polysaccharides and for the synthesis of essential molecules such as the short-chain fatty acids [63, 64]. Short-chain fatty acids are important in host energy metabolism [65] and therefore crucial for performance. Changes in nestling’s earlylife gut microbiome could affect such key physiological processes that could influence for example nestling body mass, which is tightly linked to survival to fledging [58, 59]. Because the gut microbiome establishes at a young age and is less plastic later in life [18–20], gut microbiome and changes to its richness can have long-term effects on juvenile and adult survival [21, 22]. For example, antibiotic treatment at infancy can affect the expression of genes involved in immune system functioning and lead to long-term effects on host metabolism [20]. Moreover, changes in the rearing environment can affect individual physiology and these effects can carry over to later stages of an individual’s life such as survival to fledging and lifetime reproductive success [66]. Here, we use an experimental approach to investigate whether brood size manipulation influenced wild great tit nestlings’ bacterial gut microbiome diversity on day 7 post-hatch. We also investigated whether brood size influenced nestling body mass on day 7 or on day 14 post-hatch, and if the gut microbiome predicts Page 3 of 16 Liukkonenetal. Animal Microbiome (2023) 5:19 short-term (i.e., survival to fledging) and mid-term (i.e., apparent juvenile) survival. The great tit is a well-studied species in the fields of ecology and evolution, and it is easy to monitor in the wild due to its habit of breeding in nest boxes. Great tit nestlings’ gut microbiome undergoes profound shifts during early life [52], and it has been linked to nestling natal body mass and body size [52, 61], yet studies focusing on gut microbiome associations with survival are still scarce. Here, we manipulated wild great tit broods by reducing or enlarging the original brood size in order to analyze if this affected the gut microbiome. In large broods, nestlings need to compete for their food more [67, 68], and the lower food availability could result in a lower gut microbiome diversity. This might impair nestling body mass and fitness prospects [13, 52]. We used a partial cross-fostering design that enabled us to disentangle the relative contributions of genetic background, early maternal effects, and rearing environment such as parents, nest and nestmates on gut microbiome. Furthermore, we used an unmanipulated control group in which no nestling was cross-fostered to control for the possible effects of moving the nestlings between nests. For example, early human handling such as marking and weighing at day 2 post-hatch could influence gut microbiome later on. We hypothesized that (1) in reduced broods nestlings would have a higher body mass, (2) in reduced broods nestling gut microbiome would be more diverse than in enlarged broods, and (3) higher gut microbiome diversity on day 7 post-hatch would increase survival to fledging and potentially reflect apparent juvenile survival. Such knowledge could provide new information about gut microbiome in wild passerine bird population and how the early-life environment may associate with nestling gut microbiome, body mass, and short-term and mid-term survival. Methods Study area andspecies The great tit is a small passerine bird, which breeds in secondary holes and artificial nest-boxes, making it a suitable model species. Great tits breed throughout Europe and inhabit parts of Northern Africa and Asia as well, and the breeding areas differ in environment and diet [69]. In Finland the great tit is a common species with an estimate of 1.5 to 2 million breeding pairs. They lay 6 to 12 eggs between April and May and the female incubates the eggs for 12–15days. The nestlings fledge approximately 16 to 21days after hatching. The study was conducted during the breeding season (May–July 2020) on Ruissalo island (60°25′59.99″N 22°09′60.00″E). Ruissalo island habitat is a mostly temperate deciduous forest and meadows, and some areas have small patches of coniferous trees. Brood size manipulation experiment Nest boxes were first monitored weekly and later daily when clutches were close to the estimated hatching date. Brood size manipulation took place on day 2 after hatching. Increases in great tit brood size can lead to lowered weight in both the nestlings and adults [70–75], and our decision on the number (i.e., + 2 or − 2) of manipulated nestlings (i.e, + 2 or − 2) followed the cited studies. We had four treatment groups (see Fig.1): in the ‘enlarged group (henceforward called E)’, we increased the brood size by two individuals that were taken from a ‘reduced brood’. Correspondingly, in the ‘reduced group (henceforward called R)’, we decreased the brood size by two individuals, that were added to the enlarged broods. In the ‘control group (henceforward called C)’, we swapped nestlings between nests but did not change the brood size. And lastly, in the ‘unmanipulated control group (henceforward called COU)’, we only weighed and collected fecal samples on day 7 but did not move the nestlings between nests. We also moved nestlings between the reduced nests to ensure that all nests except for COU had both original and fostered nestlings. Control nests were used to control for potential cross-fostering effects unrelated to brood size. Additionally, in the unmanipulated control group nestlings were not moved or weighed on day 2 in order to control for any handling effects per se. This study design enabled us to test the potential impacts of handling nestlings and swapping the nest early after hatching. We aimed to move approximately half of the chicks in the manipulated nests, so that the number of original and the fostered nestlings would be the same in each nest after manipulation. Before they were moved, nestlings were weighed using a digital scale with a precision of 0.1g and identified by clipping selected toenails. We aimed to add/remove nestlings that were of similar weight to avoid changing the sibling hierarchy in the brood. The moving procedure was performed as quickly as possible to minimize the risk of stress and the nestlings were kept in a warmed box during transportation. For each pair of nests in the brood size manipulation experiment, we selected nests that had a similar hatching date. In case of uneven number of nests hatching within a day, one or three nest(s) was/were allocated to the COU group. To avoid potential bias from hatching date, we allocated nests in any given day evenly to each treatment. We also checked that the treatments had an equal brood size on average i.e., we did not want to only reduce the larger clutches and enlarge the smaller clutches. These is also a significant bias towards COU nests being later in the season on average (Table1). Page 4 of 16 Liukkonenetal. Animal Microbiome (2023) 5:19 Fecal sample collection To study the effects that brood size may have on the nestling gut microbiome and its links to individual nestling body mass, survival to fledging and apparent juvenile survival, we used a subset of data from a larger experiment (Cossin-Sevrin et al., unpublished data). In this subset, we use individuals from which fecal samples were collected on day 7 after hatching and analyzed for microbiome diversity and composition (C = 23 nestlings/15 nests, COU = 22/13, E = 23/15, R = 24/16) We aimed to collect two samples (one from original and one from foster nestlings) per nest. Fecal samples from the nestlings were collected gently by stimulating the cloaca with the collection tube. Samples were collected straight into a sterile 1.5ml Eppendorf tube to avoid possible contamination of the sample. At time of sampling, each nestling was weighed (0.1g), and the nestlings were ringed for individual identification using aluminum bands. The samples were stored in cool bags onsite and afterwards moved to a -80°C freezer for storage until DNA extraction. Apparent juvenile survival We monitored all study nests until fledging to measure short-term survival. On day 14 post-hatch, the sampled nestlings were weighed, and wing-length was measured to detect if the manipulation had any effects on nestling growth. Nests were subsequently monitored for fledging success. Additionally, we monitored our study population for apparent juvenile survival (i.e., mid-term survival) after the breeding season (i.e., approximately 3months after fledging) to assess the association between gut microbiome and post-fledging survival. We captured juvenile great tits by mist netting during the autumn– winter 2020 at six different feeding stations that had a continuous supply of sunflower seeds and suet blocks. Feeding stations were located within the previously mentioned nest box population areas. For each site mist netting with playback was conducted on three separate days during October–November 2020 for three hours at a time, leading to a total of 69h of mist netting. A total of 88 individuals from the brood size manipulation experiment were caught, and the caught juvenile great tits were weighed, and wing length was measured. Our catching method provides an estimate of post-fledging survival yet, it could be slightly biased based by dispersal. In a previous study in our population [76], none of the birds ringed as nestlings were recaptured outside the study area, suggesting that dispersal is likely limited. Before manipulation After manipulation Control Unmanipulated control Moved across broods but original and manipulated brood sizes were identical. Not moved at all, but only sampled for a fecal sample at day 7 post-hatch. Enlarged nest gained four nestlings from reduced nest. Additionally, two nestlings from the original enlarged nest were moved to the reduced nest. Reduced nest lost two nestlings when four of the nestlings were moved to the enlarged nests and only two nestlings were moved back to the reduced nest. E R C COU Fig. 1 Brood size manipulation experiment schematic diagram. 2‑day‑old nestlings were moved between nestboxes to enlarge or to reduce original brood size (an example with brood size of seven is given). Some nests were kept as control nests (nestlings were moved but brood size remained the same) and some were kept as unmanipulated control nests (nestlings were not moved at all to test whether early‑life handling affects gut microbiome). The original brood size varied between nests Page 5 of 16 Liukkonenetal. Animal Microbiome (2023) 5:19 DNA extraction andsequencing We chose two samples per nest for DNA extraction, yet in such a way that both fledged and not-fledged nestlings would be included in the dataset. DNA was extracted from nestling fecal samples using the Qiagen QIAamp PowerFecal Pro DNA Kit (Qiagen; Germany) following the manufacturer’s protocols. Additionally, we included negative (RNAse and DNAse free ddH2O) controls to control for contamination during DNA extraction and additional controls to confirm successful amplification during PCR. A short fragment of hypervariable V4 region in the 16S rRNA gene was amplified using the purified DNA samples as template with the following primers: 515F_Parada (5’-GTG YCA GCMGCC GCG GTAA-3’) and 806R_Apprill (5’-GGA CTA CNVGGG TWT CTAAT-3’) [77, 78]. PCRs were performed in a total volume of 12µL using MyTaq RedMix DNA polymerase (Meridian Bioscience; Cincinnati, OH, USA). The PCR cycling conditions were as follows: first, an initial denaturation at 95°C for 3min followed by 30 cycles of 95°C for 45s, 55°C for 60s, and 72°C for 90s, and finished with a 10-min extension at 72°C. After the first round of PCR, a second round was conducted to apply barcodes for sample identification [79]. For this, PCR cycling conditions were as follows: first, an initial denaturation at 95°C for 4min followed by 18 cycles of 98°C for 20s, 60°C for 15s, and 72°C for 30s, and finished with a 3-min extension at 72°C. We performed replicate PCR reactions to control for errors during the amplification. Further on, the PCR products were measured for DNA concentration with Quant-IT PicoGreen dsDNA Assay Kit (ThermoFischer Scientific; Waltham, MA, USA) and for quality with TapeStation 4200 (Agilent; Santa Clara, CA, USA). The samples from each of the PCR replicates were pooled equimolarly creating two separate pools and purified using NucleoMag NGS Clean-up and Size Select beads (Macherey–Nagel; Düren, Germany). Finally, pooled samples were sequenced (2 × 300 bp) on the Illumina MiSeq platform (San Diego, CA, USA) at the Finnish Functional Genomic Center at the University of Turku (Turku, Finland). Sequence processing All statistical analyses were performed with R (v. 4.11.0; R Development Core Team 2021) unless otherwise stated. The demultiplexed Illumina sequence data was Table 1 (A) Brood size before and after manipulation, (B) hatching date across treatments (E) enlarged brood size, (R) reduced brood size, (C) control brood size, (COU) unmanipulated control brood size. Brood size was successfully either reduced or enlarged by two chicks (A) Brood size Before manipulation (mean ± SD) After manipulation (mean ± SD) Enlarged broods (E) 7.700 ± 1.61 9.650 ± 1.309 Reduced broods (R) 8.375 ± 1.637 6.375 ± 1.637 Control broods (C) 7.565 ± 1.805 7.565 ± 1.805 Unmanipulated broods 7.810 ± 2.112 na ANOVA F3 = 0.987, p = 0.403 (B) Hatching date Mean ± SD Enlarged broods (E) 58.60 ± 5.77 Reduced broods (R) 59.83 ± 6.41 Control broods (C) 58.74 ± 5.34 Unmanipulated broods 63.81 ± 4.79 (B) Tukey’s post-hoc for between-group comparisons Average hatching date ANOVA F3 = 3.964, p = 0.011* Contrasts Estimate SE t.ratio p COU‑C 5.070 1.70 2.983 0.019* COU‑E 5.210 1.76 2.961 0.020* COU‑R 3.976 1.68 2.363 0.092 C‑E 0.139 1.72 0.081 0.100 C‑R − 1.094 1.64 − 0.666 0.910 E‑R − 1.233 1.70 − 0.723 0.888 Page 6 of 16 Liukkonenetal. Animal Microbiome (2023) 5:19 first processed with Cutadapt version 2.7 [80] to remove locus-specific primers from both R1 and R2 reads. Then, the DADA2 pipeline (v. 1.24.0; [81]) was used to filter the reads based on quality, merge the paired-end (R1 and R2) reads, to define the single DNA sequences i.e., Amplicon Sequence Variants (henceforward ASV), and to construct a ‘seqtab’. Seqtab is a matrix also known as otutable or readtable: ASVs in columns, samples in rows, number of reads in each cell, using default parameter settings. In total, our seqtab consisted of 6,929,537 high-quality reads. Reads were assigned to taxa against the SILVA v132 reference database [82] resulting in 8658 ASVs. To control for contamination, negative DNA extraction and PCR controls were used to identify contaminants (60 ASVs) using the decontam package (v. 1.12; [83]) and all were removed from the dataset. Sequencing runs (replicate PCR’s) were merged using the phyloseq package (v. 1.32.0) and non-bacterial sequences (mainly Chlorophyta) were removed from the data as they were not of interest in this study resulting in a total of 6566 ASVs (a total of 4,107,028 high-quality reads in all samples; mean per sample: 15,155.085; mean range per sample: 0–97,264). Singleton reads were removed from the dataset by the DADA2 pipeline. Data was further analyzed with the phyloseq package (v. 1.32.0; [84]), and the microbiome package (v. 1.18.0; [85]) and visualized with the ggplot2 package (v. 3.3.6; [86]). The final dataset contained 92 samples from great tit nestlings resulting in a total of 3,161,696 reads (mean per sample: 34,366.261; mean range per sample 108 – 189,300 reads), which belonged to 6,505 ASVs. The dataset was then rarefied for alpha diversity analyses at a depth of 5000, as this was where the rarefaction curves plateaued (see Additional file 2). The rarefied dataset contained 4,791 ASVs in 88 samples. For beta diversity, the unrarefied dataset was used after confirming that the beta diversity statistics were quantitatively similar for the rarefied and unrarefied datasets. Bacterial relative abundances were summarized at the phylum and genus level and plotted based on relative abundance for all phyla and genera. A Newick format phylogenetic tree with the UPGMA algorithm to cluster treatment groups together was used to visualize sample relatedness (see Additional file 3) and was constructed using the DECIPHER (v. 2.24.0; [87]), phangorn (v. 2.8.1; [88]), and visualized with ape (v. 5.6-2; [89]), and ggtree (v. 3.4.0; [90]) packages. Statistical analyses Nestling body mass First, to analyze whether brood size manipulation affected nestling body mass in the C, E, and R treatment groups, we ran two linear mixed-effects models with the lme4 package (v. 1.1-29; [91]). In these models we used either body mass on day 7 or 14 as the dependent variable and brood size manipulation treatment, hatching date, body mass on day 2 post-hatch and original brood size as predicting variables. Hatching date is used as a predicting variable because it is known to affect nestling body mass during the breeding season [92] and there were differences in hatching date between the COU and other treatment groups (see Table1). We included the interaction between original brood size and brood size manipulation treatment in both models as the effect of manipulation may depend on the original brood size. For example, there could be stronger effect of enlargement in already large broods. Nest of origin and nest of rearing were used as random intercepts to control for the nonindependence of nestlings sharing the same original or foster nests. Here, we did not include the COU group in the analysis because we wanted to measure the effects of treatment on nestling body mass, and only enlarged, reduced or control broods’ nestlings were moved between nests. Second, to analyze whether the actual brood size affected nestling body mass, we ran two models where we used it as a continuous dependent variable to explain body mass either on day 7 or on day 14 post-hatch. Hatching date and body mass on day 2 post-hatch were used as predicting variables and nest of origin and nest of rearing as random intercepts to control for the nonindependency of samples. We included the interaction between manipulated brood size and hatching date in the models because the effect of brood size may depend on the hatching date. For example, hatching date can reflect environmental conditions and large broods may perform poorly late in the season due to poorer food availability. The COU group was initially excluded from this model to see which of the two random effects, nest of origin or nest of rearing, explained a larger portion of variation in the treatment groups. In the COU group, nest of origin and nest or rearing were the same, which meant we could not include both random effects in models where all treatment groups were present due to the model failing to converge. Nest of origin explained more of the variation in the first model (see Additional file4) and therefore, we used it in the full models with all treatment groups: C, COU, E and R. In these models, nestling body mass either on day 7 and or on day 14 post-hatch was used as a dependent variable and manipulated brood size as the explanatory variable. Hatching date and body mass on day 2 post-hatch were set as predicting variables. Nest of rearing was used as a random intercept to control for the non-independence of nestlings sharing the same foster nests. The significance of factors included in the models were tested using the F-test ratios in analysis of variance (Type III ANOVA). Page 7 of 16 Liukkonenetal. Animal Microbiome (2023) 5:19 Alpha diversity For alpha diversity analyses, which measures withinsample species diversity, we ran two linear mixed-effects models with the lme4 package (v. 1.1-29; [91]) to measure if either brood size manipulation or manipulated brood size as a continuous variable were associated with gut microbiome diversity. We used two alpha diversity metrics: the Shannon Diversity Index, which measures the number of bacterial ASVs and their abundance evenness within a sample, and Chao1 Richness, which is an estimation of the number of different bacterial ASVs in a sample. Both metrics were used to check if alpha diversity results were consistent across different metrics. Each diversity index was used as the dependent variable at a time and either brood size manipulation treatment or manipulated brood size as a predicting variable. In both models we included original brood size, weight on day 7 post-hatch and hatching date as covariates. We included interaction between brood size manipulation treatment and original brood size as there could be a stronger effect of enlargement in initially large broods. We also included interaction between manipulated brood size and weight on day 7 post-hatch because effect of brood size on microbiome may depend on nestling weight. We also tested whether alpha diversity predicted weight on day 7 post-hatch, as weight and gut microbiome diversity have been connected in previous studies. In this analysis we used weight on day 7 post-hatch as the dependent variable and alpha diversity (Shannon Diversity Index and Chao1 Richness), treatment and hatching date as predicting variables and nest of rearing as the random effect. In these sets of models, we first excluded the COU group to see which of the two random effects, nest of origin or nest of rearing, explained a larger proportion of variation in the treatment groups. Nest of rearing explained more of the variation in this model (see Additional file4) and therefore, we used it in the full model with all treatment groups: C, COU, E and R. The significance of factors included in the models were tested using the F-test ratios in analysis of variance (ANOVA). Short-term survival To explore whether alpha diversity associated with survival to fledging (i.e., short-term survival) and with apparent juvenile survival in Autumn 2020 (i.e., midterm survival), we used generalized linear models with binomial model (v. 1.1-29; lme4 package, [91]), and then tested the significance of factors with type 2 ANOVA from the car package (v. 3.0-13; [93]). Type 2 ANOVA was used because the model did not contain interaction between predicting and there was no order between covariates, as they could not be ranked. Survival to fledging and recapture in Autumn 2020 were used as the binomial response variable (yes–no) in each model. Alpha diversity (Shannon Diversity Index and Chao1 Richness) was the main predicting variable, and weight on day 7 post-hatch (same time as sampling the fecal gut microbiome), hatching date and manipulated brood size were included as covariates in the model. We did not include brood size manipulation treatment in the survival models as not enough birds from each treatment group were recorded for fledging and juvenile survival. Moreover, we excluded random effects from this model as the model failed to converge. 65 nestlings fledged successfully, while 8 nestlings were found dead in nest boxes. For 15 nestlings we had no fledging record, so these were excluded from the survival to fledging analysis. In apparent juvenile survival, 19 birds out of 92 (with data on microbiome diversity) were recaptured as juveniles. For all analyses, the R package car (v. 3.0-13; [93]) was used to test Variance Inflation Factors (VIFs) and the package DHARMa (v. 0.4.5; [94]) to test model diagnostics for linear mixed-effects and generalized linear models. Beta diversity For visualizing beta diversity, i.e., the similarity or dissimilarity between the treatment group gut microbiomes, non-metric multidimensional scaling (NMDS) was used with three distance matrices: Bray–Curtis [95], weighted UniFrac, and unweighted UniFrac [96]. Permutational multivariate analysis of variance (PERMANOVA) using the Euclidean distance matrix and 9999 permutations was tested with the R package vegan (adonis2 function; v. 2.6-2; [97]) to investigate if any variables affected to the variation in gut microbiome composition. Nest of rearing was set as a blocking factor in the PERMANOVA to control for the non-desirable effects of the repeated sampling of foster siblings. The test for homogeneity of multivariate dispersions was used to measure the homogeneity of group dispersion values. We used the phyloseq package (v. 1.32.0; [84]) to run a differential abundance analysis with a significance cut-off p < 0.01 to test the differential abundance of ASVs between the treatment groups. Results The effects ofbrood size manipulation onnestling body mass Brood size manipulation did not significantly affect nestling body mass on day 7 post-hatch (ANOVA: F2, 25.832 = 0.441, p = 0.648; see Additional file 5). Moreover, there was no significant interaction between brood size manipulation and original brood size (ANOVA: F2, 24.610 = 0.678, p = 0.517; see Additional file5). On day 14 post-hatch, brood size manipulation did not significantly affect nestling body mass (ANOVA: F2, 24.335 = 0.831, Page 8 of 16 Liukkonenetal. Animal Microbiome (2023) 5:19 p = 0.448; see Additional file 5). However, body mass increased with increasing hatching date (ANOVA: F1 24.070 = 13.367, p = 0.001; see Additional file5). Next, we did not find any significant associations between manipulated brood size and nestling body mass (ANOVA for weight on day 7: F1, 35.149 = 1.777, p = 0.191; ANOVA for weight on day 14: F1, 29.491 = 2.156, p = 0.153; see Additional file6). Nest of origin explained a larger proportion of the variation in weight than the nest of rearing on both day 7 (nest of origin 41.1% and nest of rearing 24.4%) and day 14 (nest of origin 65.5% and nest of rearing 21.9%) post-hatch, but this result was not statistically significant (Pr > χ2 = 1) (see Additional file4). Alpha diversity As 7-day-old nestlings, most bacterial taxa belonged to the phyla Proteobacteria, Firmicutes, and Actinobacteria (Fig.2). Brood size manipulation did not significantly influence alpha diversity (Shannon Diversity Index) (ANOVA: F3, 47.488 = 1.026, p = 0.390, Fig. 3; see Additional file7). Moreover, original brood size (ANOVA: F1, 50.269 = 0.388, p = 0.536; see Additional file 7), weight on day 7 post-hatch (ANOVA: F1, 80.551 = 0.003, p = 0.959; see Additional file 7), and hatching date (ANOVA: F1, 50.276 = 1.073, p = 0.305; see Additional file7) did not significantly associate with alpha diversity. There was no significant interaction between brood size manipulation and original brood size (ANOVA: F3, 48.053 = 0.126, p = 0.944; see Additional file7). Results for Chao1 Richness were quantitatively similar: brood size manipulation did not affect alpha diversity (ANOVA: F3, 45.936 = 0.358 p = 0.784, Fig. 3; see Additional file7). Nest of rearing explained a larger proportion of the observed variance in alpha diversity (27.7%) than nest of origin (10.8%), but the result was not statistically significant (Pr > χ2 = 1) (see Additional file4). Next, we tested whether the manipulated brood size as a continuous variable was associated with alpha diversity (Shannon Diversity Index), but found no significant association (ANOVA: F1, 63.001 < 0.001, p = 0.984; see Additional file8) in this analysis either. Weight on day 7 post-hatch (ANOVA: F1, 82.840 = 0.015, p = 0.903; see Additional file 8) and hatching date (ANOVA: F1, 59.734 = 0.137, p = 0.713; see Additional file 8) did not correlate with alpha diversity in this model either. There was no significant interaction between manipulated Fig. 2 Bacterial relative abundances on Phylum level across the four treatment groups. Each bar represents an individual sample. Treatment groups are control (C), unmanipulated control (COU), enlarged (E), and reduced (R). N = 88 samples divided into treatment groups as follows: C = 23, COU = 21, E = 20, R = 24. Phyla with less that 10% in relative abundance is collapsed into the category “< 10% abundance” Page 15 of 16 Liukkonenetal. Animal Microbiome (2023) 5:19 70. Smith HG, Källander H, Nilsson JÅ. Effect of experimentally altered brood size on frequency and timing of second clutches in the great tit. Auk. 1987;104(4):700–6. 71. Hõrak P, Tegelmann L, Ots I, Møller AP. Immune function and survival of great tit nestlings in relation to growth conditions. Oecologia. 1999;121(3):316–22. 72. Sanz JJ, Tinbergen JM. Energy expenditure, nestling age, and brood size: an experimental study of parental behavior in the great tit Parus major. Behav Ecol. 1999;10(5):598–606. 73. Tinbergen JM, Verhulst S. A fixed energetic ceiling to parental effort in the great tit? J Anim Ecol. 2000;69(2):323–34. 74. Neuenschwander S, Brinkhof MW, Kölliker M, Richner H. Brood size, sibling competition, and the cost of begging in great tits (Parus major). Behav Ecol. 2003;14(4):457–62. 75. Tinbergen JM, Sanz JJ. Strong evidence for selection for larger brood size in a great tit population. Behav Ecol. 2004;15(4):525–33. 76. Cossin‑Sevrin N, Hsu BY, Marciau C, Viblanc VA, Ruuskanen S, Stier A. Effect of prenatal glucocorticoids and thyroid hormones on develop‑ mental plasticity of mitochondrial aerobic metabolism, growth and survival: an experimental test in wild great tits. J Exp Biol. 2022;225(9): jeb243414. 77. Caporaso JG, Lauber CL, Walters WA, Berg‑Lyons D, Huntley J, Fierer N, Knight R, et al. Ultra‑high‑throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. ISME J. 2012;6(8):1621–4. 78. Parada AE, Needham DM, Fuhrman JA. Every base matters: assess‑ ing small subunit rRNA primers for marine microbiomes with mock communities, time series and global field samples. Environ Microbiol. 2016;18(5):1403–14. 79. de Muinck EJ, Trosvik P, Gilfillan GD, Hov JR, Sundaram AY. A novel ultra high‑throughput 16S rRNA gene amplicon sequencing library preparation method for the Illumina HiSeq platform. Microbiome. 2017;5(1):1–15. 80. Martin M. Cutadapt removes adapter sequences from high‑throughput sequencing reads. EMBnet Journal. 2011;17(1):10–2. 81. Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: high‑resolution sample inference from Illumina amplicon data. Nat Methods. 2016;13(7):581–3. 82. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Glöck‑ ner FO, et al. The SILVA ribosomal RNA gene database project: improved data processing and web‑based tools. Nucleic Acids Res. 2012;41(D1):D590–6. 83. Davis NM, Proctor DM, Holmes SP, Relman DA, Callahan BJ. Simple statistical identification and removal of contaminant sequences in marker‑gene and metagenomics data. Microbiome. 2018;2018(6):1–14. 84. McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE. 2013;8(4): e61217. 85. Lahti L & Sudarshan Shetty et al. Tools for microbiome analysis in R. 2017; http:// micro biome. github. com/ micro biome 86. Wickham H. Data analysis. In ggplot2 (pp. 189–201). 2016; Springer, Cham. 87. Wright ES. Using DECIPHER v2. 0 to analyze big biological sequence data in R. R Journal. 2016; 8:1. 88. Schliep KP. phangorn: phylogenetic analysis in R. Bioinformatics. 2011;27(4):592–3. 89. Paradis E, Blomberg S, Bolker B, Brown J, Claude J, Cuong HS, Didier G et al. Package ‘ape’. Analyses of phylogenetics and evolution, version. 2019; 2(4). 90. Yu G, Smith DK, Zhu H, Guan Y, Lam TTY. ggtree: an R package for visualization and annotation of phylogenetic trees with their covariates and other associated data. Methods Ecol Evol. 2017;8(1):28–36. 91. Bates D, Maechler M, Bolker B, Walker S. Fitting Linear Mixed‑Effects Models Using lme4. J Stat Softw. 2015;67(1):1–48. https:// doi. org/ 10. 18637/ jss. v067. i01. 92. Monrós JS, Belda EJ, Barba E. Post‑fledging survival of individual great tits: the effect of hatching date and fledging mass. Oikos. 2002;99(3):481–8. 93. Fox J, Weisberg S, Adler D, Bates D, Baud‑Bovy G, Ellison S, Monette G, et al. Package ‘car.’ Vienna: R Foundation for Statistical Computing; 2012. p. 16. 94. Hartig F & Hartig MF. Package ‘DHARMa’. R package. 2017. 95. Bray JR, Curtis JT. An ordination of the upland forest communities of southern Wisconsin. Ecol Monogr. 1957;27(4):326–49. 96. Lozupone C, Knight R. UniFrac: a new phylogenetic method for comparing microbial communities. Appl Environ Microbiol. 2005;71(12):8228–35. 97. Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’hara RB, Oksanen MJ etl al. Package ‘vegan’. Community ecology package, ver‑ sion. 2013; 2(9):1–295. 98. Moreno J, Cowie RJ, Sanz JJ, Williams RS. Differential response by males and females to brood manipulations in the pied flycatcher: energy expenditure and nestling diet. J Anim Ecol. 1995;64:721–32. 99. Askenmo C. Effects of addition and removal of nestlings on nestling weight, nestling survival, and female weight loss in the Pied Flycatcher Ficedula hypoleuca (Pallas). Ornis Scand. 1977;8:1–8. 100. Nur N. The consequences of brood size for breeding blue tits II. Nestling weight, offspring survival and optimal brood size. J Anim Ecol. 1984;53:497–517. 101. Sanz JJ. Clutch size manipulation in the pied flycatcher: effects on nestling growth, parental care and moult. J Avian Biol. 1998;28:157–62. 102. Burness GP, McClelland GB, Wardrop SL, Hochachka PW. Effect of brood size manipulation on offspring physiology: an experiment with pas‑ serine birds. J Exp Biol. 2000;203(22):3513–20. 103. Naef‑Daenzer L, Naef‑Daenzer B, Nager RG. Prey selection and foraging performance of breeding Great Tits Parus major in relation to food avail‑ ability. J Avian Biol. 2000;31(2):206–14. 104. Rytkönen S, Orell M. Great tits, Parus major, lay too many eggs:experimental evidence in mid‑boreal habitats. Oikos. 2001;93:439–50. 105. van der Most PJ, de Jong B, Parmentier HK, Verhulst S. Trade‑off between growth and immune function: a meta‑analysis of selection experiments. Funct Ecol. 2011;25(1):74–80. 106. Rauw WM. Immune response from a resource allocation perspective. Front Genet. 2012;3:267. 107. Naef‑Daenzer B, Keller LF. The foraging performance of great and blue tits (Parus major and P. caeruleus) in relation to caterpillar development, and its consequences for nestling growth and fledging weight. J Anim Ecol. 1999;6(84):708–18. 108. Barba E, Gil‑Delgado JA. Seasonal variation in nestling diet of the great tit Parus major in orange groves in eastern Spain. Ornis Scand. 1990;21:296–8. 109. Tinbergen JM, Dietz MW. Parental energy expenditure during brood rearing in the great tit (Parus major) in relation to body mass, tempera‑ ture, food availability and clutch size. Funct Ecol. 1994;8:563–72. 110. Ilmonen P, Hasselquist D, Langefors Å, Wiehn J. Stress, immunocompe‑ tence and leukocyte profiles of pied flycatchers in relation to brood size manipulation. Oecologia. 2003;136(1):148–54. 111. Kohl KD, Brun A, Caviedes‑Vidal E, Karasov WH. Age‑related changes in the gut microbiota of wild House Sparrow nestlings. Ibis. 2019;161(1):184–91. 112. Waite DW, Taylor MW. Characterizing the avian gut microbiota: mem‑ bership, driving influences, and potential function. Front Microbiol. 2014;5:223. 113. Davidson GL, Wiley N, Cooke AC, Johnson CN, Fouhy F, Reichert MS, Quinn JL, et al. Diet induces parallel changes to the gut microbiota and problemsolving performance in a wild bird. Sci Rep. 2020;10(1):1–13. 114. Murray MH, Lankau EW, Kidd AD, Welch CN, Ellison T, Adams HC, Hernandez SM, et al. Gut microbiome shifts with urbanization and potentially facilitates a zoonotic pathogen in a wading bird. PLoS ONE. 2020;15(3): e0220926. 115. Bodawatta KH, Koane B, Maiah G, Sam K, Poulsen M, Jønsson KA. Species‑specific but not phylosymbiotic gut microbiomes of New Guinean passerine birds are shaped by diet and flight‑associated gut modifications. Proc R Soc B. 2021;288(1949):20210446. 116. Bodawatta KH, Klečková I, Klečka J, Pužejová K, Koane B, Poulsen M, Sam K, et al. Specific gut bacterial responses to natural diets of tropical birds. Sci Rep. 2022;12(1):1–15. 117. Alberdi A, Aizpurua O, Bohmann K, Zepeda‑Mendoza ML, Gilbert MTP. Do vertebrate gut metagenomes confer rapid ecological adaptation? Trends Ecol Evol. 2016;31(9):689–99. Page 16 of 16 Liukkonenetal. Animal Microbiome (2023) 5:19 • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: 118. Wilkin TA, King LE, Sheldon BC. Habitat quality, nestling diet, and provisioning behaviour in great tits Parus major. J Avian Biol. 2009;40(2):135–45. 119. Rytkönen S, Vesterinen EJ, Westerduin C, Leviäkangas T, Vatka E, Mutanen M, Orell M, et al. From feces to data: a metabarcoding method for analyzing consumed and available prey in a bird‑insect food web. Ecol Evol. 2019;9(1):631–9. 120. Drobniak SM, Cichoń M, Janas K, Barczyk J, Gustafsson L, Zagalska‑ Neubauer M. Habitat shapes diversity of gut microbiomes in a wild population of blue tits Cyanistes caeruleus. J Avian Biol. 2022;2: e02829. 121. Fleischer R, Risely A, Hoeck PE, Keller LF, Sommer S. Mechanisms gov‑ erning avian phylosymbiosis: genetic dissimilarity based on neutral and MHC regions exhibits little relationship with gut microbiome distribu‑ tions of Galápagos mockingbirds. Ecol Evol. 2020;10(23):13345–54. 122. Kloos WE. Natural populations of the genus Staphylococcus. Annu Rev Microbiol. 1980;34(1):559–92. 123. Gharibzahedi SMT, Razavi SH, Mousavi SM. Characterization of bacteria of the genus Dietzia: an updated review. Ann Microbiol. 2014;64(1):1–11. 124. Hanning I, Diaz‑Sanchez S. The functionality of the gastrointestinal microbiome in non‑human animals. Microbiome. 2015;3:5. https:// doi. org/ 10. 1186/ s40168‑ 015‑ 0113‑6. 125. Doak DF, Bigger D, Harding EK, Marvier MA, O’malley RE, Thomson D. The statistical inevitability of stability‑diversity relationships in com‑ munity ecology. Am Nat. 1998;151(3):264–76. 126. Naef‑Daenzer B, Widmer F, Nuber M. Differential post‑fledging survival of great and coal tits in relation to their condition and fledging date. J Anim Ecol. 2001;70(5):730–8. 127. Nilsson JA, Svensson M. Sibling competition affects nestling growth strategies in marsh tits. J Anim Ecol. 1996;65:825–36. 128. Magrath RD. Nestling weight and juvenile survival in the blackbird, Turdus merula. J Anim Ecol. 1991;60:335–51. 129. Ringsby TH, Sæther BE, Solberg EJ. Factors affecting juvenile survival in house sparrow Passer domesticus. J Avian Biol. 1998;29:241–7. 130. Both C, Visser ME, Verboven N. Density–dependent recruitment rates in great tits: the importance of being heavier. Proc R Soc Lond Seri B Biol Sci. 1999;266(1418):465–9. 131. Cleasby IR, Nakagawa S, Gillespie DO, Burke T. The influence of sex and body size on nestling survival and recruitment in the house sparrow. Biol J Lin Soc. 2010;101(3):680–8. 132. Tinbergen JM, Boerlijst MC. Nestling weight and survival in individual great tits (Parus major). J Anim Ecol. 1990;59:1113–27. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations.