Host’s genetic background determines the outcome of reciprocal faecal transplantation on life-history traits and microbiome composition

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/ Host’s genetic background determines the outcome of reciprocal faecal transplantation on life-history traits and microbiome composition © The Author(s) 2022 Published version Juottonen, Heli; Moghadam, Neda N.; Murphy, Liam; Mappes, Johanna; Galarza, Juan A. Juottonen, H., Moghadam, N. N., Murphy, L., Mappes, J., & Galarza, J. A. (2022). Host’s genetic background determines the outcome of reciprocal faecal transplantation on life-history traits and microbiome composition. Animal microbiome, 4, Article 67. https://doi.org/10.1186/s42523-022-00210-y 2022 Juottonenetal. Animal Microbiome (2022) 4:67 https://doi.org/10.1186/s42523-022-00210-y RESEARCH © The Author(s) 2022. 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/. Open Access Animal Microbiome Host’s genetic background determines theoutcome ofreciprocal faecal transplantation onlife-history traits andmicrobiome composition Heli Juottonen1†, Neda N. Moghadam1†, Liam Murphy1, Johanna Mappes1,2 and Juan A. Galarza1,2* Abstract Background: Microbes play a role in their host’s fundamental ecological, chemical, and physiological processes. Host life-history traits from defence to growth are therefore determined not only by the abiotic environment and genotype but also by microbiota composition. However, the relative importance and interactive effects of these factors may vary between organisms. Such connections remain particularly elusive in Lepidoptera, which have been argued to lack a permanent microbiome and have microbiota primarily determined by their diet and environment. We tested the microbiome specificity and its influence on life-history traits of two colour genotypes of the wood tiger moth (Arctia plantaginis) that differ in several traits, including growth. All individuals were grown in the laboratory for several generations with standardized conditions. We analyzed the bacterial community of the genotypes before and after a reciprocal frass (i.e., larval faeces) transplantation and followed growth rate, pupal mass, and the production of defensive secretion. Results: After transplantation, the fast-growing genotype grew significantly slower compared to the controls, but the slow-growing genotype did not change its growth rate. The frass transplant also increased the volume of defensive secretions in the fast-growing genotype but did not affect pupal mass. Overall, the fast-growing genotype appeared more susceptible to the transplantation than the slow-growing genotype. Microbiome differences between the genotypes strongly suggest genotype-based selective filtering of bacteria from the diet and environment. A novel cluster of insect-associated Erysipelotrichaceae was exclusive to the fast-growing genotype, and specific Enterococcaceae were characteristic to the slow-growing genotype. These Enterococcaceae became more prevalent in the fast-growing genotype after the transplant, which suggests that a slower growth rate is potentially related to their presence. Conclusions: We show that reciprocal frass transplantation can reverse some genotype-specific life-history traits in a lepidopteran host. The results indicate that genotype-specific selective filtering can fine-tune the bacterial community at specific life stages and tissues like the larval frass, even against a background of a highly variable community with stochastic assembly. Altogether, our findings suggest that the host’s genotype can influence its susceptibility to being colonized by microbiota, impacting key life-history traits. †Heli Juottonen and Neda N. Moghadam contributed equally to this work *Correspondence: [email protected] 1 Department of Biological and Environmental Sciences, University of Jyväskylä, P.O. Box 35, 40014 Jyväskylä, Finland Full list of author information is available at the end of the article Page 2 of 16 Juottonenetal. Animal Microbiome (2022) 4:67 Keywords: Lepidoptera, Arctia plantaginis, Wood tiger moth, Bacterial community, Growth, Genotype, 16S rRNA, Gut, Long amplicon Background Variation in traits within a population can be partly determined by genetic polymorphisms. Uncovering genotype–phenotype associations allows the analysis of the evolution and adaptive advantages of the traits. It is increasingly recognized that phenotype may also be influenced by the microbiome, which all animals, including insects, possess [20]. In general, the microbiome can potentially influence the host’s life history and fitness [31, 58, 99]. In insects, the microbiome has been related to behavioural, nutritional, and life-history traits [11, 114, 119]. Moreover, microbiome composition can vary according to the host’s genetic background [58, 119]. For instance, the gut microbiome can mediate genotype effects on the phenotype: In Drosophila, the host genotype influences the microbiome composition, leading to differences in nutrition between phenotypes [11]. A stable, symbiotic microbiome can confer benefits on the insect host, such as aiding its growth [18, 41, 48]. On the other hand, even mutualistic symbionts incur costs [75], and opportunistic pathogenic bacteria can severely disadvantage the host [39, 97]. The outcome of these associations can depend on genotype-genotype interactions between the microbe and its host [79], as well as among microbes [55, 98]. For example, in the pea aphid (Acyrthosiphon pisum), the host’s genotype influences the protection given by bacterial symbionts against pathogens [79, 121]. In turn, host-to-microbe effects can play an important role in microbiome assembly in the host [23]. Selective mechanisms that impact the establishment of microbes in insects include specialized organs [53, 73] and mechanisms that vary with host genetic background, such as innate immunity [54, 72]. Such filtering due to host traits and genetic background could influence the host’s fitness and life-histories [52]. The gut bacteria of lepidopteran larvae show metabolic potential to benefit the host by digesting and detoxifying food plants [117, 124] and by producing antimicrobial compounds against invaders [92]. However, disruptions of the gut during moulting and metamorphosis, a highly alkaline pH (up to 11–12), lack of specialized gut structures, and fast passage of food can constrain the development of a consistent symbiotic microbiome [20]. Accordingly, several studies have concluded that there is no stable microbiome in Lepidoptera [35, 61, 103]. Despite reports on the effects of diet, habitat, and developmental stage on gut bacteria [7, 29, 44, 92, 93, 103], no clear consensus exists on the ecological roles of bacteria in Lepidoptera [77]. Lepidopteran larval growth has been found to be correlated [90] and not correlated [12] with microbiome composition. Antibiotic treatment of lepidopteran larvae has similarly led to increased growth [27, 116], decreased growth [124], or no effect on growth [35]. Moreover, increased growth has been observed in axenic larvae [62]. Thus, the causal connections between microbiota and Lepidoptera growth traits remain elusive. One way to identify such connections is through microbiota transplants. Transplants (or bacteriotherapy) have been extensively applied in the biomedical field to study the potential of microbes to impact health and disease [1, 118, 126]. The principle is to transfer microbes from a healthy individual to an unhealthy individual aiming to enrich beneficial microbes and restore a balanced microbiome. In insects, gut microbiota transplants are starting to reveal the importance of microbes in host development, immune response, and survival in dung beetles, cockroaches, bumblebees and parasitoid wasps [42, 68, 78, 80, 113]. However, such studies are lacking in Lepidoptera, one of the most species-rich and ecologically important groups of insects, in which less than 0.1% of species have been screened for microbes [77]. The uncertainties of the functional role and specificity of microbes in Lepidoptera make this group a particularly important target for transplant experiments. Here, we investigate the effect of microbiome transplantation on wood tiger moths (Arctia plantaginis) with distinct genetic backgrounds and life-history traits. A reciprocal faecal transplantation was carried out between wood tiger moths of two colour genotypes that differ in the duration of their larval stage by adding frass (i.e., larval faeces) to their diet. The experimental insects have been reared in the laboratory for several generations, kept in similar conditions and fed the same diet. We followed the microbiome’s compositional changes during larval development and in the resulting adults. Thus, any consistent microbiome differences between host genotypes could reflect selective filtering of bacteria. We ask (i) if each genotype has its own associated microbiome, (ii) if it is stable across life-stages, and (iii) if microbiome transplant can reverse the growth rate between the genotypes. We also examine if the transplantation impacts other important fitness traits, such as pupal mass and the volume of defensive secretions. Page 3 of 16 Juottonenetal. Animal Microbiome (2022) 4:67 Methods Study species andgenotype lines The wood tiger moth is an aposematic species distributed throughout the Holarctic [37]. In Europe, males display colour polymorphism, having yellow or white hindwings and co-occur at variable frequencies within populations [25, 37]. The two colour morphs differ in key fitness traits such as mating success [30, 69], immune responses [70], protection against predators [56, 87], and flight activity [88]. The yellow-white hindwing polymorphism is determined by a single Mendelian locus with two alleles in which the yellow allele (y) is recessive to the white (W) allele [71]. Hence, the white colouration is produced by WW and Wy allelic combinations, whereas yy produces yellow. Analyses of selection lines show that the homozygous genotypes differ in the length of their larval stage (i.e., from egg hatching to pupation). Individuals of the WW genotype have a significantly shorter larval stage than those of the yy genotype (Additional file1: Fig. S1). Adults of both colour morphs release defensive secretions from their anal cavity, which are effective at deterring invertebrate predators, with yellows having stronger chemical defence than whites [86]. Thus, this species offers a good opportunity to study the impact of microbes on life-history and fitness traits in relation to the host’s genetic background. Larval sampling andrearing beforefrass transplant Genotype selection lines of wood tiger moths have been maintained for over 12 generations at the University of Jyväskylä, Central Finland. For this study, we selected four families of WW and four of yy genotypes to characterize their bacterial communities as they develop with or without faecal transplantation (see below). We included four families to cover for possible variation among families in the analysis of genotype effects, and we did not analyze family effects. The general rearing protocol and the pedigree are described in detail in Nokelainen etal. [71] and De Pasqual etal. [17]. In the rearing protocol, larvae are fed with dandelion (Taraxacum spp.) collected from the wild without disinfection or antibiotic supplementation. Here we modified the rearing protocol as follows. Immediately after hatching and before being given any food, we collected larvae to assess the bacteria in newly hatched larvae. Newly hatched larvae are too small for dissection (~ 2mm), and hence, the whole larva was used. The larvae were surface sterilized to exclude microbial contamination from the environment. We cut the filter from a 1-ml filter tip and placed it inside a 1.5-ml tube. We pooled four larvae into a sample on the filter and added 450μl of autoclaved doubledistilled water (AddH2O), creating a whirlpool with a 1-ml filtered pipette tip for 2min. The water was collected and the procedure was repeated three times. The collected washing water (Table1) was stored at −20°C until DNA extraction to represent bacteria on the outside of the larva, including environmental contamination. The washed larvae were then transferred to a new 1-ml filter tip, rinsed with 450μl of 5% sodium hypochlorite (NaOCl) solution, and centrifuged at 2000rpm for 30s. The rinsing process was repeated three times, after which the surface sterilized larvae were stored at −20°C until DNA extraction (hatched larvae in Table1). The remaining larvae (374/genotype) were split into groups of 20–25 larvae and reared inside sterile petri dishes. The dishes were kept in climate chambers in a Table 1 Sample types included in bacterial 16S rRNA gene sequencing *Uneven n caused by difficulties obtaining PCR products or sample collection (i.e., abdominal fluids, guts) Sample type Sample type definition n per genotype* Total no. of samples Hatched larvae Surface sterilized newly hatched larvae analyzed as whole 4 8 Washing water Water from washing newly hatched larvae 4 8 Control diet Artificial diet na 2 Transplant diet Mix of control diet and frass 2 4 Frass before Frass from larvae fed with control diet before transplantation 2 4 Frass after Frass from larvae fed with transplant diet of the other genotype WW: 3, yy: 2 5 Frass after control Frass from larvae fed with transplant diet of the same genotype 2 4 Gut Gut of adult from larvae fed with transplant diet of the other genotype WW: 2, yy: 1 3 Gut control Gut of adult from larvae fed with transplant diet of the same genotype WW: 3, yy: 2 5 Abdominal fluid Abdominal fluid of adult from larvae fed with transplant diet of the other genotype WW: 10, yy: 4 14 Abdominal fluid control Abdominal fluid of adult from larvae fed with transplant diet of the same genotype WW: 8, yy: 4 12 Water control Water used in artificial diet and sample storage na 3 Page 4 of 16 Juottonenetal. Animal Microbiome (2022) 4:67 18:6h light:dark cycle at 21°C during the light and 14°C during the dark. An artificial diet was prepared consisting of 3g agar, 32.1g semolina, 8.58g yeast, 8.3g wheat germ, 1.76g Vanderzant vitamin mix, 1.8ml nipagin and 180μl acetic acid in 200ml freshly boiled AddH2O to minimize diet-derived bacteria. Two samples of this diet were stored at −20°C until DNA extraction (control diet in Table1). Roughly 5g of this diet was presented to the larvae on top of a sterilized microscope slide inside the petri dish. After 48h, approximately 0.5g of larval faeces, hereafter frass, was collected from the bottom of the petri dish using sterilized tweezers and stored at −20°C until DNA extraction (frass before in Table1). Frass transplantation andrearing To prepare diets for frass transplantation between genotypes, approximately 10 frass pellets from each petri dish were collected after 48h of giving the control diet and pooled to obtain ~ 1 g/genotype and mixed with 50 g of the control diet. Four samples of this transplant diet (two of each genotype) were mixed with 2ml of boiling AddH2O and stored at −20°C until DNA extraction (transplant diet of genotypes WW and yy in Table1). The larvae, all in their 3rd or 4th instar, were then divided into treatment and control groups in sterile petri dishes with 10–15 larvae of the same genotype per petri dish. The treatment group was fed the opposite genotype’s transplant diet: each petri dish of WW larvae received ~ 5g of yy transplant diet, and each petri dish of yy larvae received ~ 5g of WW transplant diet. The control group larvae received ~ 5g of transplant diet of their own genotype. The petri dishes were kept in the climate cabinets in the same conditions as above. Twenty-four hours after the food was given, approximately 1g of frass was collected from each genotype as above and stored at −20°C until DNA extraction (frass after, frass after control in Table1). The rearing continued until all larvae pupated or died. The pupae were placed individually in 150-ml plastic containers, kept in the climate chambers in the same conditions and weighed to the nearest milligram. From a subset of the emerging adults, we dissected the gut following Moghadam etal. [67] with minor modifications. Briefly, each adult moth was placed on a sterile petri dish and its head was removed using a sterilized scalpel. A drop of AddH2O was placed next to the abdomen and the gastrointestinal tract (i.e., gut) including the crop, foregut, midgut, and hind gut was pulled out using sterilized forceps under a light stereoscope with a Bunsen burner next to it to reduce the risk of contamination. The dissected guts were placed individually in 30μl of AddH2O and stored at −20°C until DNA extraction (gut in Table1). Likewise, we collected the abdominal defensive secretions from the adults. We gently pressed the abdomen of live adults with sterilized tweezers until the secretion was released from the anal cavity. The secretion was collected using UV-sterilized 10-μl glass capillaries under a laminar flow, measured with a digital caliper, and placed individually in 30μl of AddH2O and stored at −20°C until DNA extraction (abdominal fluid in Table1). Finally, we took 30μl of the AddH2O batch used to prepare all the samples above and stored it at −20°C until DNA extraction (water control in Table1). Life histories We followed several life-history traits of individual larvae, pupae, and adults from the different genotypes and treatments. The overall developmental rate was determined by counting the number of days elapsed from egg hatching until adult eclosion. This included the larval and pupal stages. We further analyzed the development rate within the larval stage (i.e., from egg hatching until pupation), as well as within the pupal stage (i.e., from pupation to adult eclosion). In addition, we recorded the weight of all individual pupa, and at the adult stage, we measured the volume of abdominal defensive secretions as described above. DNA extraction, PCR andPacBio amplicon sequencing DNA was extracted by homogenizing the sample (larvae, frass, gut, abdominal fluid, diet) in 30μl of water with a metal bead (∅ 2.3mm) in a Bead Ruptor (OMNI) at speed 3.93m/s for 2 × 30s. After homogenization, the samples were boiled at 100°C for 10min and stored at -20°C until further use. DNA quantification was performed with the Qubit BR DNA kit (ThermoFisher). To assess bacterial diversity in the larvae, their frass, and adult moths, we amplified ~ 1550bp of the 16S ribosomal RNA (rRNA) gene using custom primers (forward 5′-AGA GTT TGATCMTGG CTC AG-3′, reverse 5′-CCT TGT TAC GAC TTC ACC CCAG-3′). The primers were designed using Primer3 [112] from Lepidoptera-associated 16S rRNA gene sequences downloaded from National Center for Biotechnology Information (NCBI) and aligned using ClustalW [95]. Polymerase chain reactions (PCR) were performed in a C1000 thermal cycler (Bio-Rad) using 5μl of DNA template, 1 × Platinum SuperFI Mastermix (ThermoFisher), 1 × enhancer buffer, 0.5μM of each primer, and 0.5mM MgCl2 in reaction volume of 20μl. The cycling conditions were as follows: 98°C for 30s, 40 cycles of 98°C for 10s, 49°C for 10s, 72°C for 1min, and a final elongation of 5min at 72°C. The PCR products were run in 3% agarose gels, and the bands were excised using gel cutting tips (Axygen) and purified by centrifuging through 1-ml filter tips at 6000rpm for 15min. DNA concentration of the purified PCR products was Page 5 of 16 Juottonenetal. Animal Microbiome (2022) 4:67 measured using PicoGreen dsDNA Assay Kit (ThermoFisher), and a sequencing library was prepared according to the PacBio multiplexed amplicon library preparation protocol. The samples were barcoded and sequenced in a PacBio Sequel at the Novogene sequencing laboratories. In addition, a mock community (ZymoBIOMICS Microbial Community DNA Standard, Zymo Research) andthe water control (i.e., negative control) were amplified and sequenced with each library. Sequence processing andquality control Sequence reads were processed with PacBio tools distributed in Bioconda. PacBio subreads were combined into consensus sequences with ccs in package pbccs (v.6.0.0) with the default settings. The consensus sequences (797,984 reads) were demultiplexed based on barcodes with lima (v. 2.0.0) with the settings -peakguess, -different- -ccs, -min-length 1440, -max-inputlength 1580, -min-end-score 26, and -split-bam-named. The resulting bam files were converted into fastq with bam2fastq in the package bam2fastx (v. 1.3.1). Sequences were submitted to the National Center for Biotechnology Information Sequence Read Archive under accession code PRJNA804133. The reads were processed further and amplicon sequence variants (ASVs) inferred in DADA2 (v. 1.16.0, [8] following guidelines for PacBio data [9], https:// benjj neb. github. io/ LR A SM anusc ript/ L R A Sms_ fec al. html) in R (v. 4.0.4, [84] and RStudio (v. 1.4.1106). Primers were removed with the command removePrimers. Reads were filtered with the command filterAndTrim and the settings minQ = 3, minLen = 1300, maxLen = 1600, maxN = 0, and maxEE = 2. The reads were dereplicated and denoised using PacBio-specific error estimation function. Chimeras were removed from the denoised reads with the command removeBimeraDenovo and setting minFoldParentOverAbundance = 3.5. Taxonomy was assigned against the Silva database (v. 138.1, [83]. The ASV data was imported into phyloseq (v. 1.32.0, [64], and ASVs assigned to chloroplasts or mitochondria or not assigned to Bacteria were removed. Potential contaminants were examined based on ASVs in water control and washing water samples (Table 1) with the package decontam based on frequency (threshold 0.2), prevalence (threshold 0.5), and inspection of frequency vs. DNA concentration plots [16], and none were detected. This resulted in 226 ASVs (when excluding ASVs in the mock community controls) and on average 6898 reads per sample (in total 565,705 reads). Statistical analyses andphylogenetic diversity measures All analyses were carried out in R (v. 4.0.4) through RStudio. The package ggplot2 [122] was used for generating plots. To examine the effect of the frass transplantation on the life histories of the genotypes, we implemented a Kruskal–Wallis one-way analysis of variance (ANOVA) followed by pairwise Dunn’s tests to the traits measured. All tests’ significance values were corrected for multiple comparisons. This non-parametric approach was chosen because the samples violate parametric assumptions of normality and/or equality of variance (Additional file1: Fig. S2). For comparing bacterial diversity and community composition, the ASV table was rarefied to the median number of reads (6915 reads) with the function rrarefy in vegan (v. 2.5.7, [74]. If a sample had fewer reads than the median, all its reads were included. Then, the ASV table was standardized to relative abundances. Sequences of the ASVs were aligned and a phylogenetic tree was constructed using RAxML (model GRT + gamma, [101]) on the SILVA ACT server [82]. Faith’s phylogenetic diversity (PD) and ASV richness were determined in picante (v. 1.8.2, [49]. The phylogenetic tree was converted into a phylogenetic distance matrix with the function cophenetic. Measures of phylogenetic relatedness among communities (phylogenetic beta diversity) were calculated in picante as mean pairwise distance (MPD, [120], function comdist, abundance weighted) and mean nearest taxon distance (MNTD, function comdistnt, abundance weighted). MPD emphasizes the clustering of basal clades in the phylogenetic tree, whereas MNTD emphasizes patterns closer to the tips of the tree. These values were used as distance measures in non-metric multidimensional scaling (NMDS) with function metaMDS in vegan and permutational multivariate analysis of variance (PERMANOVA) with function adonis2 in vegan [2]. Separate phylogenetic trees of Erysipelotrichaceae and Enterococcus ASVs were constructed by aligning the sequences and selected reference sequences (described strains and similar environmental sequences identified by Blast searches) with SINA v. 1.2.11 on the SILVA ACT server and inferring a maximum likelihood tree with RAxML (model GTR + gamma, [101]) in QIIME2 (v. 2021.8.0, [4]. We used the same phylogenetic relatedness measures as above to assess phylogenetic clustering of bacterial communities against null models across sample types. This analysis aims to determine whether bacterial community assembly and turnover across life stages are driven by niche-based (i.e., environmental filtering) or stochastic processes [104]. For example, if gut conditions favour the proliferation of specific bacterial taxa, it would be shown as phylogenetic clustering. Net relatedness index (NRI) Page 6 of 16 Juottonenetal. Animal Microbiome (2022) 4:67 was calculated as the standardized effect size of MPD (function ses.mpd in picante, abundance weighted) multiplied by -1. Nearest taxon index (NTI) was calculated as the standardized effect size of MNTD (function ses.mntd, abundance weighted) multiplied by -1. Positive values of NRI and NTI indicate phylogenetic clustering (taxa are more closely related than by chance), whereas negative values indicate phylogenetic overdispersion (taxa are less closely related than by chance). Values of NRI and NTI differing from 0 and thus showing more clustering by chance were identified by Welch’st-test (function t.test). To compare mechanisms of phylogenetic turnover along life history and experimental stages, we calculated βNTI values for pairs of communities according to Stegen etal. [104] and https:// github. com/ stegen/ Stegen_ etal_ ISME_ 2013. βNTI > 2 indicates significantly higher community turnover than by chance driven by deterministic selection [104, 105]. βNTI between -2 and 2 indicates community assembly driven by stochastic processes. βNTI < -2 indicates less community turnover than by chance. Results Life‑history effects offrass transplantation Overall developmental time (i.e., from egg hatching to adult eclosion) differed between the controls of genotypes WW and yy (Additional file1: Fig. S3), as in the stock selection lines (Additional file1: Fig. S1). This difference was mainly due to the faster growth of WW individuals at the larval stage (Fig.1). The length of the pupal stage did not differ between the genotypes (Kruskal–Wallis one-way ANOVA statistic = 7.74, P > 0.05) (data not shown). When WW larvae received the frass transplant from genotype yy, they grew slower than the control WW larvae that received WW frass (Fig.1, Additional file1: Fig. S3). The opposite was not observed: WW frass transplant did not affect the growth of yy larvae. Pupal weight differed with genotype in the controls, with WW pupae being lighter than yy pupae (Additional file1: Fig. S4). However, we observed no differences in pupal weight between the controls and the pupae that received the frass transplant of the other genotype. This suggests that the frass transplant did not affect pupal weight. Transplantation also did not affect pupal mortality (controls WW 10%, yy 12%; treatments yy to WW 7%, WW to yy 15%). The adults of genotype WW secreted a smaller volume of defensive abdominal fluids than the adults of yy in the controls. Transplantation with yy frass significantly increased abdominal fluid secretion in WW adults (Fig.2). As in the case of growth rate, this could point to greater susceptibility or adaptability of the WW genotype to the transplantation and the presence of foreign bacteria. Larger volumes in defensive secretions were also observed in the yy genotype transplanted with WW frass. However, this difference was not significant relative to the yy controls, likely because of the unbalanced number of samples. Overall, it can be suggested that frass transplantation generally increased the volume of defensive secretions. Bacterial community composition anddiversity withgenotype, life stage andfrass transplant Bacteria detected in newly hatched larvae showed no difference in phylogenetic diversity or ASV richness between the genotypes (Additional file1: Fig. S5). Compared to the newly hatched larvae, larval frass before transplantation had lower phylogenetic diversity (Additional file1: Fig. S5), which indicated that frass contained a reduced set of bacteria. However, the frass bacteria were not solely a subset of the bacteria detected in newly hatched larvae. Frass before transplantation shared only one out of five ASVs (genotype WW) or three out of 11 ASVs (genotype yy) with the newly hatched larvae (Fig.3). Unlike in newly hatched larvae, the bacterial community in frass differed between the genotypes (Fig.4, Additional file1: Fig. S6; PERMANOVA hatched larvae R2 = 0.27, P = 0.24; frass R2 = 0.38, P = 0.001). Frass of genotype WW collected before transplantation had lower ASV richness and phylogenetic diversity than yy frass (Additional file1: Fig. S5). Bacteria in WW frass were dominated by Erysipelotrichaceae (Firmicutes; Fig. 5), which were only detected in genotype WW: in addition to larval frass in WW adult gut and abdominal fluids (Fig.3). These Erysipelotrichaceae belong to a novel cluster with no described species that includes uncultured bacteria detected in termite gut (Additional file1: Fig. S7; [40, 65]. Frass of genotype yy was instead dominated by Enterococcaceae (Firmicutes,Fig.5). These ASVs included ASV1 and ASV5 (Fig.3) that occurred across the genotypes in frass, adult gut, and abdominal fluids and had 100% sequence similarity to Enterococcus gallinarum and E. casseliflavus strains from Lepidoptera (Additional file1: Fig. S8; [14, 15]. These Enterococcaceae ASVs were among the only four common core ASVs detected in one or more sample types with 50% prevalence (Fig.3). When WW larvae received the yy frass transplant, their frass retained the WW-specific Erysipelotrichaceae. Notably, WW frass also gained or showed an increased relative abundance of the two common Enterococcaceae ASVs detected in yy frass (ASV1, ASV5) (Fig.5). The prominence of these ASVs in WW frass after transplantation suggests they may have originated from the yy frass transplant. Enterococcus ASVs became more prevalent also in one of the WW controls that received WW frass, but these ASVs represented a Page 7 of 16 Juottonenetal. Animal Microbiome (2022) 4:67 different cluster within Enterococcus not detected in yy frass (ASV3, ASV12, 99–100% sequence similarity to E. mundtii strains, Additional file1: Fig. S9). Bacteria in yy frass, on the other hand, changed almost completely both with WW frass transplant and in the controls that received genotype yy’s own frass (Figs.4B, 5). After the transplant, yy frass bacteria consisted of sporadic ASVs representing bacterial groups detected in the control diet and transplant diet fed to the larvae (Fig.5, Additional file1: Fig. S9). Therefore, we found no evidence pointing to the transfer of WW frass bacteria to genotype yy. Variation of bacterial community in adult gut and abdominal fluid could not be linked to the transplant treatment or to genotype, partly due to Fig. 1 Days as larvae of A. plantaginis in reciprocal frass transplantation between genotypes WW and yy and in control transplantations within the genotypes Page 8 of 16 Juottonenetal. Animal Microbiome (2022) 4:67 large variation among individuals (Fig.4C, Additional file1: Fig. S10). The ASVs affiliated with E. casseliflavus/gallinarum and E. mundtii occurred in adult gut with no consistent pattern. In adult abdominal fluids, both Enterococcus types were present in both genotypes, but E. casseliflavus/gallinarum ASVs dominated in genotype WW and E. mundtii type in genotype yy (Additional file1: Fig. S10). Phylogenetic clustering andturnover ofbacteria withlife stage andgenotype We used phylogenetic clustering and turnover measures to identify potential ecological mechanisms structuring Fig. 2 Secretion of defensive abdominal fluids of A. plantaginis adults in reciprocal frass transplantation between genotypes WW and yy and in control transplantations within the genotypes Page 15 of 16 Juottonenetal. Animal Microbiome (2022) 4:67 71. Nokelainen O, Galarza JA, Kirvesoja J, Suisto K, Mappes J. Genetic colour variation visible for predators and conspecifics is concealed from humans in a polymorphic moth. J Evol Biol. 2022;35:467–78. 72. Nyholm SV, Graf J. Knowing your friends: invertebrate innate immunity fosters beneficial bacterial symbioses. Nat Rev Microbiol. 2012;10:815–27. 73. Ohbayashi T, Takeshita K, Kitagawa W, Nikoh N, Koga R, Meng X-Y, Tago K, Hori T, Hayatsu M, Asano K, Kamagata Y, Lee BL, Fukatsu T, Kikuchi Y. Insect’s intestinal organ for symbiont sorting. Proc Natl Acad Sci. 2015;112:E5179–88. 74. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, Minchin PR, O’Hara RB, Simpson GL, Solymos P, Stevens MHH, Szoecs E, Wagner H. vegan: Community Ecology Package. R package version 2.5–7. https:// CRAN.Rproje ct. org/ packa ge= vegan (2020). 75. Oliver KM, Moran NA, Hunter MS. Costs and benefits of a superinfection of facultative symbionts in aphids. Proc R Soc B: Biol Sci. 2006;273:1273–80. 76. Osborn F, Berlioz L, Vitelli-Flores J, Monsalve W, Dorta B, Rodrıguez Lemoine V. Pathogenic effects of bacteria isolated from larvae of Hylesia metabus Crammer (Lepidoptera: Saturniidae). J Invertebrate Pathol. 2002;80:7–12. 77. Paniagua Voirol LR, Frago E, Kaltenpoth M, Hilker M, Fatouros NE. Bacterial symbionts in Lepidoptera: their diversity, transmission, and impact on the host. Front Microbiol. 2018;9:556. 78. Parker ES, Moczek AP, Macagno ALM. Reciprocal microbiome transplants differentially rescue fitness in two syntopic dung beetle sister species (Scarabaeidae: Onthophagus). Ecol Entomol. 2021;46:946–54. 79. Parker BJ, McLean AHC, Hrček J, Gerardo NM, Godfray HCJ. Establishment and maintenance of aphid endosymbionts after horizontal transfer is dependent on host genotype. Biol Let. 2017;13:20170016. 80. Pietri JE, Tiffany C, Liang D. Disruption of the microbiota affects physiological and evolutionary aspects of insecticide resistance in the German cockroach, an important urban pest. PLoS ONE. 2018;13: e0207985. 81. Pilon FM, Visôtto LE, Guedes RNC, Oliveira MGA. Proteolytic activity of gut bacteria isolated from the velvet bean caterpillar Anticarsia gemmatalis. J Comp Physiol B. 2013;183:735–47. 82. Pruesse E, Peplies J, Glöckner FO. SINA: accurate high-throughput multiple sequence alignment of ribosomal RNA genes. Bioinformatics. 2012;28:1823–9. 83. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Peplies J, Glöckner FO. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 2013;41(D1):D590–6. 84. R Core Team. 2021. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https:// www.Rproje ct. org/. 85. Risely A. Applying the core microbiome to understand host–microbe systems. J Anim Ecol. 2020;89:1549–58. 86. Rojas B, Burfield-Steel E, Pakkanen H, Suisto K, Maczka M, Schulz S, Mappes J. How to fight multiple enemies: target-specific chemical defences in an aposematic moth. Proc R Soc B: Biol Sci. 2017;284:20171424. 87. Rojas B, Mappes J, Burdfield-Steel E. Multiple modalities in insect warning displays have additive effects against wild avian predators. Behav Ecol Sociobiol. 2019;73:37. 88. Rojas B, Gordon S, Mappes J. Frequency-dependent activity in the aposematic wood tiger moth, Parasemia plantaginis. Curr Zool. 2015;61:765–72. 89. Rudman SM, Greenblum S, Hughes RC, Rajpurohit S, Kiratli O, Lowder DB, Lemmon SG, Petrov DA, Chaston JM, Schmidt P. Microbiome composition shapes rapid genomic adaptation of Drosophila melanogaster. Proc Natl Acad Sci. 2019;116:20025–32. 90. Ruokolainen L, Ikonen S, Makkonen H, Hanski I. Larval growth rate is associated with the composition of the gut microbiota in the Glanville fritillary butterfly. Oecologia. 2016;181:895–903. 91. Schimming O, Challinor VL, Tobias NJ, Adihou H, Grün P, Pöschel L, Richter C, Schwalbe H, Bode HB. Structure, biosynthesis, and occurrence of bacterial pyrrolizidine alkaloids. Angew Chem Int Ed. 2015;54:12702–5. 92. Shao Y, Chen B, Sun C, Ishida K, Hertweck C, Boland W. Symbiontderived antimicrobials contribute to the control of the Lepidopteran gut microbiota. Cell Chem Biol. 2017;24:66–75. 93. Shao Y, Arias-Cordero E, Guo H, Bartram S, Boland W. In vivo pyro-SIP assessing active gut microbiota of the cotton leafworm, Spodoptera littoralis. PLoS ONE. 2014;9: e85948. 94. Shapira M. Gut microbiotas and host evolution: scaling up symbiosis. Trends Ecol Evol. 2016;31:539–49. 95. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Söding J, Thompson JD, Higgins DG. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol. 2011;7:539. 96. Siva-Jothy MT, Moret Y, Rolff J. Insect immunity: an evolutionary ecology perspective. In: Simpson SJ, editor. Advances in Insect Physiology. Academic Press; 2005. p. 1–48. 97. Smee MR, Baltrus DA, Hendry TA. Entomopathogenicity to two hemipteran insects is common but variable across epiphytic Pseudomonas syringae strains. Front Plant Sci. 2017;8:2149. 98. Smee MR, Raines SA, Ferrari J. Genetic identity and genotype × genotype interactions between symbionts outweigh species level effects in an insect microbiome. ISME J. 2021;15:2537–46. 99. Spor A, Koren O, Ley R. Unravelling the effects of the environment and host genotype on the gut microbiome. Nat Rev Microbiol. 2011;9:279–90. 100. Stagaman K, Burns AR, Guillemin K, Bohannan BJ. The role of adaptive immunity as an ecological filter on the gut microbiota in zebrafish. ISME J. 2017;11:1630–9. 101. Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics. 2014;30:1312–3. 102. Stanley D, Hughes RJ, Geier MS, Moore RJ. Bacteria within the gastrointestinal tract microbiota correlated with improved growth and feed conversion: challenges presented for the identification of performance enhancing probiotic bacteria. Front Microbiol. 2016;7:187. 103. Staudacher H, Kaltenpoth M, Breeuwer JAJ, Menken SBJ, Heckel DG, Groot AT. Variability of bacterial communities in the moth Heliothis virescens indicates transient association with the host. PLoS ONE. 2016;11: e0154514. 104. Stegen J, Lin X, Konopka A, Fredrickson JK. Stochastic and deterministic assembly processes in subsurface microbial communities. ISME J. 2012;6:1653–64. 105. Stegen JC, Lin X, Fredrickson JK, Chen X, Kennedy DW, Murray CJ, Rockhold ML, Konopka A. Quantifying community assembly processes and identifying features that impose them. ISME J. 2013;7:2069–79. 106. Tang X, Freitak D, Vogel H, Ping L, Shao Y, Cordero EA, Andersen G, Westermann M, Heckel DG, Boland W. Complexity and variability of gut commensal microbiota in polyphagous Lepidopteran larvae. PLoS ONE. 2012;7: e36978. 107. Tamariz J, Burgueño-Tapia E, Vázquez MA, Delgado F. Chapter One: Pyrrolizidine Alkaloids. In: Knölker H-J, editor. The Alkaloids: Chemistry and Biology. Berlin: Academic Press; 2018. p. 1–314. 108. Tegtmeier D, Riese C, Geissinger O, Radek R, Brune A. Breznakia blatticola gen. nov. sp. nov. and Breznakia pachnodae sp. nov., two fermenting bacteria isolated from insect guts, and emended description of the family Erysipelotrichaceae. Syst Appl Microbiol. 2016;39:319–29. 109. Teh B-S, Apel J, Shao Y, Boland W. Colonization of the intestinal tract of the polyphagous pest Spodoptera littoralis with the GFP-tagged indigenous gut bacterium Enterococcus mundtii. Front Microbiol. 2016;7:928. 110. Thakur A, Dhammi P, Saini HS, Kaur S. Pathogenicity of bacteria isolated from gut of Spodoptera litura (Lepidoptera: Noctuidae) and fitness costs of insect associated with consumption of bacteria. J Invertebr Pathol. 2015;127:38–46. 111. Turnbaugh PJ, Bäckhed F, Fulton L, Gordon JI. Diet-induced obesity is linked to marked but reversible alterations in the mouse distal gut microbiome. Cell Host Microbe. 2008;3:213–23. 112. Untergasser A, Cutcutache I, Koressaar T, Ye J, Faircloth BC, Remm M, Rozen SG. Primer3-new capabilities and interfaces. Nucleic Acids Res. 2012;40:e115–e115. 113. van Opstal EJ, Bordenstein SR. Phylosymbiosis impacts adaptive traits in Nasonia wasps. MBio. 2019;10:e00887-e919. 114. Vernier CL, Chin IM, Adu-Oppong B, Krupp JJ, Levine J, Dantas G, BenShahar Y. The gut microbiome defines social group membership in honey bee colonies. Sci Adv. 2020;6:eabd3431. Page 16 of 16 Juottonenetal. Animal Microbiome (2022) 4:67 • 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: 115. Vilanova C, Baixeras J, Latorre A, Porcar M. The generalist inside the specialist: gut bacterial communities of two insect species feeding on toxic plants are dominated by Enterococcus sp. Front Microbiol. 2016;7:1005. 116. Visôtto LE, Oliveira MGA, Guedes RNC, Ribon AOB, Good-God PIV. Contribution of gut bacteria to digestion and development of the velvetbean caterpillar, Anticarsia gemmatalis. J Insect Physiol. 2009;55:185–91. 117. Visôtto LE, Oliveira MGA, Ribon AOB, Mares-Guia TR, Guedes RNC. Characterization and identification of proteolytic bacteria from the gut of the velvetbean caterpillar (Lepidoptera: Noctuidae). Environ Entomol. 2009;38:1078–85. 118. Walter J, Maldonado-Gómez MX, Martínez I. To engraft or not to engraft: an ecological framework for gut microbiome modulation with live microbes. Curr Opin Biotechnol. 2018;49:129–39. 119. Walters AW, Hughes RC, Call TB, Walker CJ, Wilcox H, Petersen SC, Rudman SM, Newell PD, Douglas AE, Schmidt PS, Chaston JM. The microbiota influences the Drosophila melanogaster life history strategy. Mol Ecol. 2020;29:639–53. 120. Webb CO, Ackerly DD, McPeek MA, Donoghue MJ. Phylogenies and community ecology. Annu Rev Ecol Syst. 2002;33:475–505. 121. Weldon SR, Russell JA, Oliver KM. More is not always better: coinfections with defensive symbionts generate highly variable outcomes. Appl Environ Microbiol. 2020;86:e02537-e2619. 122. Wickham H. ggplot2: elegant graphics for data analysis. New York: Springer; 2016. 123. Winters AE, Lommi J, Kirvesoja J, Nokelainen O, Mappes J. Multimodal aposematic defenses through the predation sequence. Front Ecol Evol. 2021;9:512. 124. Xia X, Lan B, Tao X, Lin J, You M. Characterization of Spodoptera litura gut bacteria and their role in feeding and growth of the host. Front Microbiol. 2020;11:1492. 125. Zhang C, Zhang M, Wang S, Han R, Cao Y, Hua W, Mao Y, Zhang X, Pang X, Wei C, Zhao G, Chen Y, Zhao L. Interactions between gut microbiota, host genetics and diet relevant to development of metabolic syndromes in mice. ISME J. 2010;4:232–41. 126. Zhang F, Cui B, He X, Nie Y, Wu K, Fan D, Feng B, Chen D, Ren J, Deng M, Li N, Zheng P, Cao Q, Yang S, He X, Liu Y, Nie Y, Zhou Y, Fan D, Wu K, Nie Y, Ji G, Li P, Cui B, Zhang F, FMT-standardization Study Group. Microbiota transplantation: concept, methodology and strategy for its modernization. Protein Cell. 2018;9:462–73. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.