Dynamics, diversity, and roles of bacterial transmission modes during the first asexual life stages of the freshwater sponge Spongilla lacustris

Dynamics, diversity, and roles of bacterial transmission modes during the first asexual life stages of the freshwater sponge Spongilla lacustris | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Dynamics, diversity, and roles of bacterial transmission modes during the first asexual life stages of the freshwater sponge Spongilla lacustris Benoit Paix, Elodie van der Valk, Nicole J. de Voogd This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3973150/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 08 Jun, 2024 Read the published version in Environmental Microbiome → Version 1 posted 9 You are reading this latest preprint version Abstract Background: Sponges-associated bacteria play important roles in the physiology of their host, whose recruitment processes are crucial to maintain symbiotic associations. However, the acquisition of bacterial communities within freshwater sponges is still unexplored. Spongilla lacustris is a model sponge widely distributed in European rivers and lakes, producing dormant cysts (named gemmules) for their asexual reproduction, before winter. Through an in vitro experiment, this study aims to describe the dynamics of bacterial communities and their transmission modes following the hatching of these gemmules. Results: An overall change of bacterial β -diversity was observed through the ontology of the new juvenile sponges. These temporal differences were potentially linked first to the osculum acquisition and the development of a canal system, and then, the increasing colonization of the Chlorella -like photosymbionts. Gemmules hatching with a sterilized surface were found to have a more dispersed and less diverse microbiome, revealing the importance of gemmule epibacteria for the whole holobiont stability. These epibacteria were suggested to be vertically transmitted from the maternal tissues to the gemmule surface. Vertical transmission through the incorporation of bacterial communities inside of the gemmule, was also found as a dominant transmission mode, especially with the nitrogen fixers Terasakiellaceae . Finally, we showed that almost no ASVs were shared between the free-living community and the juveniles, suggesting that horizontal recruitment is unlikely to happen during the first stages of development. However, the free-living bacteria filtered are probably used as a source of nutrients, allowing an enrichment of copiotrophic bacteria already present within its microbiome. Conclusions: This study brings new insight for a better understanding of the microbiome acquisition during the first stages of freshwater sponge development. We showed the importance of epibacterial communities on gemmules for the whole holobiont stability, and demonstrated the near absence of recruitment of free-living bacteria during the first stages. Freshwater sponge Holobiont Microbiome Vertical transmission Horizontal acquisition Gemmule Ontogeny Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Sponges (phylum Porifera) are filter-feeding and sessiles animals known for their tight interactions with their microbiome. Described as a functional entity, the association of a sponge and its microbiome called “sponge-holobiont” represents an important engineering system of benthic environments (e.g. through the cycling of the organic matter, [ 1 ]). Within these holobionts, the host provides a habitat for microbial symbionts, who in return displays key functions for the host (e.g. by producing chemical defenses, or supplying additional nutrients; [ 2 ]). Sponge microbiomes are often described to be diverse, stable and host-specific [ 3 ], playing an essential role for the resilience of their host observed during environmental changes (e.g. under heat-stress or acidification, [ 4 , 5 ]). The mechanisms associated with the selection of microbial symbionts and their stable maintenance are increasingly studied for marine sponge models [ 6 , 7 ]. Among others, stochasticity processes seem to be involved through the acquisition of microbial communities [ 8 , 9 ]. The current body of knowledge suggests that marine sponges can acquire their microbial communities through both vertical and horizontal transmission (“VT” and “HT”, respectively; [ 7 , 9 – 11 ]). This intermediate transmission mode was initially named “leaky vertical transmission” [ 12 ], or “mixed-mode transmission” [ 13 ]. While VT was confirmed for a large diversity of sponge models [ 14 – 18 ], it appears insufficient to explain the origin of the overall bacterial community [ 8 , 19 , 20 ]. For example, larvae of eight Mediterranean sponge species shared only 17% of the bacterial ASVs with their parents, on average [ 8 ]. Additionally, the importance of horizontal transmission was discussed for 19 sponge species from Vietnam, as their bacterial core community was found to be highly shared (> 50%) with the core bacterial community of the ambient seawater [ 19 ]. In this latter study, species-specific recognition mechanisms were hypothesized for the sponges, allowing an enrichment of specific planktonic bacteria. However, the diversity and relative importance of these transmission modes are still unknown within freshwater sponges, as their bacterial diversity remains largely underexplored [ 21 ]. Freshwater sponges (Order Spongillida) comprise approximately 240 species [ 22 ], and display a wide range of environmental adaptation within diverse ecosystems such as lakes, rivers, streams, ponds, or urban canals [ 23 ]. They can survive drastic changes of temperatures and light, but also desiccation and anoxic conditions, and can also tolerate high levels of pollutants [ 23 ]. Under light-exposed conditions, the establishment of stable interaction with photosymbionts (e.g. microalgae such as Chlorella spp.) constitutes an important factor for the physiology of freshwater sponges harboring a green coloration [ 24 ]. Up to 20% of the photosynthates produced by the microalgal partners can be translocated directly to the host. [ 25 , 26 ]. However, symbiotic association within freshwater sponges can be disturbed, leading for example to a dysbiosis state for the baikal sponge Lubomirskia baikalensis harboring the brown rot syndrome associated with bleached tissues [ 27 ]. Genetic responses to the presence of the Chlorella -like green symbionts revealed their importance for the host, as immunity mechanisms are associated with the algal symbiont recognition [ 28 ]. These results were demonstrated through the infection of aposymbiotic sponge Ephydatia muelleri with its algae, suggesting that such symbiosis can also be horizontally acquired [ 28 ], in addition to the vertical transmission [ 29 ]. However, the mechanisms associated with the recognition and the selection of a host-specific bacterial community from their environment are yet to be explored for freshwater sponges [ 30 ]. Indeed, the bacterial composition has been rarely compared between freshwater sponge species, but also with those from the ambient environment (e.g. from freshwater or sediment, [ 21 ]). Sugden et al. showed that bacterial communities of E. muelleri are mainly distinct from those from the surrounding freshwater and biofilms [ 31 ]. Geographical differences were also observed between rivers, suggesting that the environment could also shape the host-microbe specificity. In this latter study, horizontal acquisition from ambient bacteria was suggested as an important factor that could explain the rivers-specific microbiome, while the relative importance of the vertical transmission was still uncertain. Spongilla lacustris (Linnaeus, 1759) is one of the most widespread freshwater sponge species in temperate areas of the Northern Hemisphere [ 32 ]. As for E. muelleri , the asexual reproduction of the S. lacustris model has been extensively described through the production (in autumn) of diapausing cysts, named gemmules [ 33 ]. These gemmules contain undifferentiated dormant stem cells (named thesocytes) and provide protection against adverse conditions during winter or exposure to air. These gemmules can be hatched and cultivated under diverse in vitro conditions (allowing a detailed description of their first developmental stage). Hence, gemmule-producing sponges constitute models of interest for recent imaging and molecular techniques [ 28 , 34 – 36 ]. However, the microbiome composition of S. lacustris has been underexplored to date, in regards to their wide distribution area. Two studies suggested the dominance of Alphaproteobacteria, Actinobacteria, Bacteroidota, Gammaproteobacteria, and Betaproteobacteria [ 37 , 38 ], but the mechanisms underlying the early assembly of the bacterial consortium within these freshwater sponges are yet to be explored. Our study aims to decipher the transmission modes of bacterial communities during the first steps of S. lacustris asexual cycle. We hypothesize that the mixed mode transmission occurs, with both VT and HT. First, the VT would be achieved through the transmission of maternal bacteria within the gemmule, but also on its surface, as suggested by Sugden et al. [ 31 ]. Once the gemmule has hatched and the filtration system is formed, the HT of planktonic bacteria could then constitute an additional factor shaping the S. lacustris microbiome. To test these hypotheses, an experimental design is developed to determine the respective importance over time of (i) the microbiome transmitted on the gemmule’s surface, and (ii) the ambient free-living bacteria filtered by the juveniles. Material and methods The overall study design was based on the cultivation of the S. lacustris juveniles under two crossed factors (resulting in 4 treatments, Fig. 1 ). The first factor was related to the absence/presence of epibiotic microbiome (± “EM”) on the gemmule, depending on the surface sterilization. In parallel, the second factor was related to the absence/presence of free-living bacteria (± “FB”) in the cultivation medium, depending on the pore size of the filters used for the freshwater filtration. Sampling of the in situ adult sponges Four replicates of adult S. lacustris specimens were collected in situ from a wooden jetty located at an inlet of the river Rhine at Oegstgeest (the Netherlands, 52°10'24.5"N, 4°27'01.7"E, ~ 0.5m depth), at three different sampling times a month apart in 2021: M1 (October 15th ), M2 (November 10th ) and M3 (December 15th ). Sponge samples were rinsed with surrounding freshwater filtered upon 0.2µm pore size filters (FF0.2), and cut with sterilized tweezers and scalpels. The identification of S. lacustris was based on skeletal and gemmoscleres examination. A voucher specimen is deposited at the sponge collection of Naturalis Biodiversity Center (RMNH.POR.12472). Samples for DNA metabarcoding were preserved in sterilized plastic vials filled with 96% ethanol and conserved at -20°C until DNA extraction. After M3, only encrusting parts of dead sponges were left: the sponge skeletons were observed without tissues, but full of gemmules. One of these “encrusting gemmules patches” (resulting from specimens collected at M3) was collected at t 0 (February 17th) in sterilized plastic vials filled with surrounding FF0.2. These samples collected for the experiment were kept at 4°C until the washing steps of the gemmules. Sampling and washing of the gemmules One day after the sampling, gemmules from the encrusting patch were separated from the maternal sponge skeleton according to [ 34 ], using sterilized tweezers and teasing needles. Approximately 450 gemmules were evenly distributed in two distinct 15mL falcon tubes. Gemmules from the first tube were washed according to [ 34 ], while gemmules from the second tube were washed with the same protocol but with Strekal’s medium instead of the 1% hydrogen peroxide (H 2 O 2 ) solution (Fig. 1 ). For this treatment without H 2 O 2 , particular cautions were taken (using a stereomicroscope) to sort and remove the dead and damaged gemmules, as well as important debris of the sponge maternal skeleton attached to the gemmule surface. As the H 2 O 2 solution induces the sterilization of the gemmule surfaces [ 34 ], these two washing conditions allowed to separate the gemmules in two groups: (i) the -EM for the sterilized gemmules without any potential epibiotic microbiome, and (ii) the + EM for the unsterilized gemmules with their potential epibiotic microbiome still attached the surface (Fig. 1 ). Gemmules were stored at 4°C in Strekal’s medium for one hour before the plating. The efficiency of the sterilization through the use of H 2 O 2 was investigated as described in Supplementary Information (SI). Plating and cultivation of the juveniles For the t 0 samples (gemmules collected before the plating), 5 replicates for both + EM and -EM treatments were collected from the 15mL falcon tubes and stored in 1.5mL eppendorf tubes filled with 96% EtOH at -20°C until DNA extraction. A total of 120 juvenile samples were expected for the experiment, based on the calculation of 6 sampling times (t 1 to t 6 ), 2 conditions of gemmule surface (± EM), 2 conditions of filtered freshwater (± FB), and 5 replicates. Under sterile conditions, the + EM and -EM gemmules were distributed (on ice) in 120 wells of 24-well plates (approximately 3 to 4 gemmules per well), as described in Fig. 1 . The Strekal’s medium was then removed and replaced by 1 mL of filtered freshwater at room temperature. Plates were filled with freshwater filtered upon 0.45µm size filters (FF0.45) or FF0.2, for + FB and -FB respectively (Fig. 1 ). The filtered freshwater used was collected 1 hour before the filtration, on the same sampling site as the in situ sponges. The plates were then transferred in an incubation chamber (Fitotron®, Weiss Technik, GmbH, Germany) at 20°C, with 70% of humidity and a day/night cycle of 14:10h (960 lux for the day light, SD: ±65). Details of the sampling timeline of the experiment are described in note S1 (Supplementary Information, SI). For each treatment of each sampling time (t 1 to t 6 ,), the hatching rates obtained (80.2% in average, SD: ± 21.2%) allowed us to collect at least one hatched juvenile for each replicates. The unhatched gemmules were discarded, resulting in approximately 1 to 3 juveniles gathered per replicate. Briefly, samples from t 1 to t 6 were collected at 3, 7, 11, 17, 24, and 31 days after the plating (t 0 ), respectively. The experiment was designed up to 31 days to follow potential changes occurring after the formation of the osculum and the development of a mature canal system. For each sampling time, the replicates of juvenile sponges were collected in 1.5mL eppendorf tubes filled with 96% EtOH at -20°C until DNA extraction. The FF0.45 and FF0.2 were refreshed at room temperature every 3 or 4 days with freshly collected freshwater from the same sampling site described above (see details in SI, note S1 ). This duration between two refreshments was slightly longer compared to those recommended in [ 34 ] (every 2 days). This choice was considered to have a longer exposure time enabling potential recognition mechanisms allowing the horizontal acquisition of planktonic bacteria. To investigate the structure of the free-living bacterial community amended at each refreshment for the + FB condition, the FF0.45 was also successively filtered upon 0.2µm pore size PES filters used for DNA extraction. These samples named “FF0.45 samples” were collected for refreshment occurring before each time (except t 1 ), and preserved in the CTAB extraction buffer at -20°C until extraction. Before each sampling time from t 1 to t 6 , the juveniles sponges from each treatment (cross-conditions of ± EM with ± FB) were photographed using a stereomicroscope (SteREO Discovery.V20, ZEISS, Germany) mounted with a camera (AxioCam MRc5, ZEISS, Germany). DNA extractions, library preparation and high throughput sequencing of 16S rRNA gene amplicons DNA from in situ adults (M1, M2, and M3), gemmules (t 0 ), and in vitro juveniles (t 1 to t 6 ) was extracted using the FastDNA™ SPIN Kit for Soil (MP Biomedicals, Inc.) following the manufacturer’s instructions. In situ adult sponge samples were cut into small pieces of approximately 3*1*0.5mm using sterilized tweezers and scalpel blades. Special care was taken to collect only sponge tissues, avoiding gemmules. For t 0 samples, the gemmule coating was broken (by crushing them using the top of 10µL filter pipette tips), to facilitate the extraction of the DNA from the inside of the gemmule (see note S2 in SI). The CTAB DNA extraction of the FF0.45 samples was performed as described in [ 39 ]. The library preparation was conducted through a two-step PCR protocol for all samples together with the extraction blank and two negative controls (mQ water instead of template DNA). For the first PCR, the V3-V4 regions of the 16S rRNA gene were targeted and amplified with the PCR primers 341F 5’-CCTACGGGNGGCWGCAG-3’ and 785R 5’-GACTACHVGGGTATCTAATCC-3’ [ 40 ] and the KAPA HiFi HotStart Ready Mix PCR Kit (Roche Molecular Systems, Inc.). Reactions were performed in a T100 Thermal Cycler (Bio-Rad, Hercules, CA, United States). The following thermal cycling scheme was conducted: initial denaturation at 95°C for 3 min, 30 cycles of denaturation at 98°C for 20 s, annealing at 57°C for 30 s, followed by extension at 72°C for 30 s. The final extension was carried out at 72°C for 1 min. PCR products from the samples were checked using an E-Gel™ (agarose gels at 2%), and the absence of amplification was validated for the negative controls and two extraction blanks. In several gemmule and juvenile samples, PCR amplifications were unsuccessful due to low DNA concentrations. At least 4 replicates was successful for every treatment at each time, except for + EM + FB at t 5 , with 3 successful replicates ( Table S1 ). PCR products were then cleaned with NucleoMag NGS-Beads (bead volume at 0.9 times the total volume of the sample, Macherey Nagel, Düren, Germany) using the VP 407AM-N 96 Pin Magnetic Bead Extractor stamp (V&P Scientific, San Diego, CA, United States). Through a second PCR, 3 µL of the cleaned PCR products were then amplified and labeled using the MiSeq Nextera XTDNA library preparation kit (Illumina, San Diego, CA, United States), with the same thermal cycling scheme limited to 8 cycles. PCR products were then analyzed with the Fragment Analyzer Agilent 5300 using the DNF-910-33 dsDNA Reagent Kit (35–1,500 bp) protocol (Agilent Technologies, Santa Clara, CA, United States) to confirm the successful labeling of the DNA fragments. Negative controls and extraction blanks remained negative after this step. The pooling at the equimolar concentration was performed with QIAgility (Qiagen, Hilden, Germany). The final pool was then cleaned with NucleoMag NGSBeads, eluted in Milli-Q, and the DNA concentration was verified using Tapestation 4150 (Kit HSD 5000, Agilent Technologies, Santa Clara, CA, United States). The sequencing was performed on an Illumina MiSeq V3 PE300 platform at BaseClear B.V. (Leiden, Netherlands). 16S rRNA gene metabarcoding data processing The raw reads were first treated by BaseClear B.V. for demultiplexing (using bcl2fastq version 2.20, Illumina), and filtering based on two quality controls (using Illumina Chastity filtering, and a PhiX control signal filtering). The following reads were then processed with the DADA2 workflow allowing an inference to Amplicon Sequence Variant (ASV) [ 41 , 42 ], using the “dada2” R package following the workflow described in [ 43 ] and guidelines described in the online tutorial ( https://benjjneb.github.io/dada2/tutorial.html ). The parameters used for filtering and trimming reads were as follows: truncation length of 270 and 240 base pairs for forward and reverse reads, respectively, maxN = 0, maxEE = 2, and truncQ = 2. After the construction of the ASV table, chimeric sequences were filtered and the taxonomic assignment was performed using the Silva v138 reference database [ 44 ]. The ASV and taxonomy tables produced by the pipeline were then combined into a phyloseq object, together with the sample metadata table, using the “phyloseq” R package [ 45 ]. The dataset was then filtered by removing all sequences from Eukaryota, chloroplast and mitochondria (representing on average > 0.01% [SD: +/-0.002], 36.9% [SD: +/-12.8] and 0.9% [SD: +/-1.9] of all reads per sample, respectively). Data was then decontaminated with the negative controls and the two extraction blanks used as control samples, using the “decontam” R package [ 46 ]. Metabarcoding and statistical analyses The α -diversity measures were estimated with Chao1 (estimated richness), Pielou (evenness), and Shannon (both richness and evenness) indices using the “phyloseq” and “vegan” R packages [ 45 , 47 ] and the rarefied datasets (rarefaction performed to the minimum library size, i.e. 5332 reads). According to Shapiro tests, the diversity metrics were significantly different from the normal distribution. Consequently, differences between sample groups within these metrics were investigated through non-parametric tests (Kruskal-Wallis followed by pairwise Wilcoxon tests) using the “agricolae” R package [ 48 ]. Following recommendations for compositional approaches from [ 49 , 50 ], all other analyses were conducted without rarefaction, using the “phyloseq” R package and the datasets normalized to the total number of sequences per sample. The β -diversity was analyzed with non-metric multidimensional scaling (NMDS) using Bray-Curtis dissimilarity. Differences of β -diversity between groups were statistically checked with one-way PERMANOVA tests followed by pairwise Adonis tests, using the “vegan” R package. Differences between the dispersion of the β -diversity within each treatment were calculated through the PERMDISP2 analysis using the betadisper () function from the “vegan’ R package, and tested using an ANOVA followed by a Tukey’s HSD pairwise test. Focusing on t 0 and t 1 samples, phylogenetical heat trees were performed to identify the significant pioneer taxa differentially abundant between the different treatment conditions (± EM and ± FB), using the metacoder package [ 51 ]. The differential analyses were performed with a subset of the compositional dataset excluding rare ASV (relative abundance < 0.02%). Analyses of the ASVs shared between conditions were performed with the core () function using the “microbiome” R package [ 52 ]. An ASV from juveniles from one specific treatment at a specific sampling time was considered shared with the in situ adult sponges, when the ASV is present at least in 3 replicates of each of the two groups. Similarly, an ASV from juveniles from one specific treatment at a specific sampling time was considered shared with the filtered freshwater (FF0.45), when the ASV is present at least in 3 samples of each of the two groups. Using the “eulerr” R package [ 53 ], the Venn diagrams were plotted to represent the number of shared ASVs between the juveniles, the in situ adult sponges and the prefiltered freshwater with the venn () function. Results Development of the juveniles sponges The hatching of the gemmules was observed 2 days after their plating (t 0 ), whatever the treatment. No significant differences of hatching rates were observed between the gemmules from each treatment (ANOVA test: p = 0.07). These hatching rates reached in average 76.3% (SD: ±20.7) for + EM + FB, 81.9% (SD: ±22.6) for + EM-FB, 75.3% (SD: ±21.4) for -EM + FB and 87.2% (SD: ±20.0) for -EM-FB. The same developmental stages were observed between treatments for each sampling time ( Figure S1 ). The first sampling time (t 1 ) was characterized with the development of the sponge tissues on the gemmule coat (i.e. the gemmule husk) and colonizing the substrate. The early formation of a canal system and an osculum was observed for t 2 samples. At t 3 , the development of a mature canal system was noticed, together with a light green coloration. The coloration was more saturated at t 4 and thereafter ( Figure S1 ). Alpha-diversity analyses Significant differences between the different sample types were observed only for the Pielou index ( Figure S2 , Table S2 ), with lower values within in situ adult sponges compared to gemmules and juvenile sponges ( Figure S2 , Wilcoxon test). For gemmules and juveniles, significant differences between treatments were observed according to the sampling time together with the ± EM (with epibiotic microbiome) factor, for the Shannon (Fig. 2 , Table S3) , Chao1 and Pielou indices ( Figures S3A and S3B , respectively, Table S3 ). No significant differences were observed according to the ± FB factor (Kruskal-Wallis test: p > 0.05 for the three indices). For the Shannon index, a decrease from t 0 to t 2 was observed for the -EM (without epibiotic microbiome) samples (Fig. 2 ). At t 1 , t 2 , and t 3 , Shannon values in -EM samples were significantly lower compared to those in + EM (Fig. 2 , Wilcoxon test). The same results were observed for the two other indices (Pielou and Chao1), except at t 2 for the Chao1 ( Figure S3 , Wilcoxon test). For all indices, a general tendency of increasing α -diversity was observed after t 3 up to t 6 , for the juveniles in -EM samples, while the values for the + EM samples stayed stable all along the experiment. Beta-diversity analyses The NMDS plot and PERMANOVA test, conducted with all samples, showed significant differences between the β -diversity of each sample type (freshwater, in situ adult tissues, gemmules and in vitro juveniles, Fig. 3 A, Table S4 ). Additionally, significant temporal changes were observed on the NMDS along (i) the first axis from M3 to t 0 ( in situ adult sponges and gemmules), and (ii) the second axes from t 1 up to t 6 ( in vitro juveniles) (Fig. 3 A, Table 5A ). Within the in vitro juveniles, the main temporal shifts appear between (i) t 1 and t 2 , and (ii) t 4 and t 5 (Fig. 3 A). Pairwise comparison tests confirmed the significant differences between all sampling times, except between (i) t 2 and t 3 , (ii) t 3 and t 4 , and (iii) t 5 and t 6 ( Table S5B ). For in vitro juvenile samples, the three-way PERMANOVA ( Table S5A ) also indicated significant differences between sampling times, in combination with the two treatment factors (± EM and ± FB). Pairwise comparison tests conducted with all juvenile samples, showed an overall significant difference between the β -diversity the 4 treatments, except between + EM + FB and + EM-FB ( Table S5C ). The NMDS of all samples plotted with a color code related to the treatments ( Figure S4 ) revealed that + EM samples from t 1 to t 4 are grouped, while differences between + FB and -FB can be distinguished for the -EM samples. Additionally, differences between + EM and -EM can also be observed with t 5 and t 6 samples ( Figure S4 ). The NMDS (Fig. 3 B ) and two-way PERMANOVA ( Table S6A ) analyses, performed at each sampling time separately, revealed significant differences of β -diversity between treatments. At t 0 (for the gemmules), significant differences between + EM and -EM samples were observed on the first NMDS axis (Fig. 3 B, Tables S6A and S6B ). For the NMDS plots from t 1 to t 6 , significant differences can also be observed between juveniles + EM and -EM (Fig. 3 B, Tables S6A and S6B ). From t 1 to t 3 , a similar pattern can be observed with three clusters differentiating (i) + EM samples (including both + and -FB), (ii) -EM + FB samples, (iii) and -EM-FB samples (Fig. 3 B). More precisely, the differences between + FB and -FB appear more important within the -EM group than the + EM group. From t 4 to t 6 , a larger separation between the + FB and -FB clusters was observed for the + EM samples. Finally at t 6 , 4 distinct clusters can be observed for each of the 4 treatments from the two cross-conditions: the first NMDS axis is involved in the ± EM differences, while the second axis seems to explain the ± FB differences. However, according to the pairwise comparisons, these ± FB differences at t 6 were not significant, in both + EM and -EM conditions ( Table S6B ). The β -diversity dispersion within each treatment (± EM and ± FB) was determined at each sampling time of the in vitro juveniles ( Figure S5 ). A global tendency of increasing dispersion is observed with a higher dispersion over time for the + EM samples, confirming the previous observations on the NMDS plots (Fig. 3 B). For -EM juvenile sponges, the highest dispersion values were found at t 1 , followed by a decrease at t 2 ( Figure S5 ). At t 1 , a significantly higher dispersion was observed in -EM compared to + EM, whatever the presence or not of FB ( Figure S5 , Table S7 ). At t 2 , a similar observation was made, with significantly higher dispersion in -EM compared to + EM, only in the + FB condition ( Figure S5 , Table S7 ). For t 3 , t 4 , and t 6 , no significant differences of β -diversity dispersion were observed between treatments ( Figure S5 ). At t 5 , significant differences were observed with higher dispersion in + EM + FB juveniles compared to all -FB juveniles ( Figure S5 ). Compositional and differential analyses The composition of the bacterial community at the family level showed overall differences between each sample type (filtered freshwater, in situ adult sponge, the gemmules, and the in vitro juveniles, Figure S6 ). For the filtered freshwater, the main free-living bacterial communities were dominated by Sporichthyaceae and Microbacteriaceae (Actinobacteria), and Burkholderiaceae and Methylophilaceae (Gammaproteobacteria) ( Figure S6A ). For the in situ adult sponges, the composition was dominated by Sporichthyaceae (Actinobacteria), Chitinophagaceae and Flavobacteriaceae (Bacteroidia), Elsteraceae and Terasakiellaceae (Alphaproteobacteria) and Comamonadaceae (Gammaproteobacteria) ( Figure S6A ). Differences in composition for these adult sponge tissues were observed over time from M1 (October) to M3 (December), with an increase of Flavobacteriaceae and Elsteraceae and a decrease of Sporichtyaceae . Additionally, higher relative abundances of Terasakiellaceae can be observed specifically at M2 compared to M1 and M3. For the gemmules (t 0 ), the bacterial composition of the -EM was dominated by Terasakiellaceae , while the + EM was dominated by the Rhodobacteraceae and Comamonadaceae ( Figure S6A ). For the in vitro juveniles ( Figure S6B ), important changes can be observed, with a dominance of Flavobacteriaceae with exclusively ASVs from Flavobacterium (Bacteroidia;) at t 1 . From t 2 to t 4 , Comamonadaceae , Alteromonadaceae , and Pseudomonadaceae (Gammaproteobacteria) are the dominant families. Alphaproteobacteria was also found as one of the most abundant classes ( Rhodobacteraceae , Sphingomonadaceae , Rhizobiaceae ), without clear changes observed over time. Additionally, an increase of Pirellulaceae can be noticed all along the experiment up to t 6 . Differences in relative abundances of families between treatments were observed, for example at t 1 , with higher percentages of Comamonadaceae in + EM compared to -EM. For + EM at t 1 , higher abundances of Oxalobacteraceae were observed in + FB compared to -FB samples, while the opposite is noticed for Alteromonadaceae . At t 2 higher relative abundances of Pseudomonadaceae were found in -EM samples compared to + EM samples, while the opposite was observed for the Comamonadaceae . The differential analysis with t 0 samples (gemmules before hatching) was performed to identify significant taxa differentially abundant in + EM gemmules compared to -EM gemmules, and conversely ( Figure S7 ). The analysis confirmed observations from the barplots for the most abundant families ( Figure S6A ) with a higher abundance of Terasakiellaceae in the -EM, and a higher abundance of Rhodobacteraceae and Comamonadaceae + EM. These two last families were found mainly represented by the genus Pseudorhodobacter and Hydrogenophaga , respectively, being also found differentially more abundant in + EM samples. Other taxa with lower percentages were also found to be differentially abundant, such as Burkholderiaceae (genus Ralstonia ) and Hyphomonadaceae (genus Hirschia ) in + EM and Chitinophagaceae (genus Ferruginibacter ) and Pseudomonadaceae (genus Pseudomonas ) in -EM. In addition to the t 0 samples, another focus was made at t 1 for differential analyses performed between treatment of the in vitro juveniles. The differential analysis between + EM and -EM was conducted for both + FB and -FB samples, separately. (i) For + FB samples, (Fig. 4 A) significantly higher abundances of Caulobacterales and Sphingomonadales (Alphaproteobacteria), Cytophagales (Bacteroidia), Burkholderiales, Xanthomonadales and Enterobacterales (Gammaproteobacteria) were observed in + EM compared to -EM. Conversely, a significantly higher abundance of Terasakiellaceae (Alphaproteobacteria) and Moraxellaceae (Gammaproteobacteria) was observed in -EM compared to + EM. (ii) For -FB samples (Fig. 4 B), significantly higher abundance of all Bacteroidia taxa (including Cytophagales), Rhodobacteraceae (Alphaproteobacteria), Enterobacterales and Comamonadaceae (Gammaproteobacteria) were observed in + EM compared to -EM, while higher abundance of Oxalobacteraceae , Burkholderiaceae and Caulobacteraceae and Rhizobiales were observed in -EM. The differential analysis between + FB and -FB was also conducted for both + EM and -EM samples, separately. (i) For + EM samples (Fig. 4 C), significant higher abundance of Oxalobacteraceae (genera Undibacterium and Janthinobacterium ) and Shewanellaceae were observed for + FB samples compared to -FB samples, while higher Moraxellaceae , Crocinitomicaceae , Alteromonadaceae ( Rheinheimera ), Rhodocyclaceae where significantly more abundant in -FB samples compared to the + FB ones. (ii) For -EM samples (Fig. 4 D), the differential analysis revealed significant higher abundances of Flavobacteriaceae and Pseudomonadales (including Moraxellaceae ) in + FB, while -FB samples were enriched with Caulobacterales, Rhizobiales, and Sphingomonadaceae . ASVs from gemmules or juveniles shared with adult sponges or freshwater samples Venn diagrams ( Figure S8 ) were obtained to estimate the number of ASVs shared between three groups of samples: (i) the juveniles (from a specific treatment at a specific time), (ii) adult sponges and (iii) freshwater samples. No ASVs were found to be shared between the freshwater and the gemmules, or between the freshwater and the juveniles ( Figure S8 ), except (i) ASV73 ( Polynucleobacter ) shared at t 1 and t 3 with the + EM + FB juveniles, and at t 4 with the -EM-FB juveniles, (ii) ASV80 (Candidatus Limnoluna ) shared at t 5 with the + EM + FB, -EM + FB and -EM-FB juveniles, and (iii) ASV734 ( Bradyrhizobium ) shared at t 6 with the -EM-FB juveniles. These three ASVs (73, 80, and 734) were found with an average relative abundance below 0.04%. Additionally, a total of 17 ASVs were found to be shared between the freshwater and the in situ adult sponges ( Figure S8 ). Based on the Venn diagram results ( Figure S8 ), the numbers ASVs shared between in vitro juveniles and in situ adult sponges were summarized in Fig. 5 . These numbers were higher for + EM juveniles, compared to -EM juveniles. For example, at t 1 , 34 ASVs from the adult sponges were shared in + EM + FB, while only 9 were shared between the -EM + FB and the in situ adult sponges (Fig. 5 and S10). Similarly, 21 ASVs from in situ adult sponges were shared with the + EM-FB samples, while 12 ASVs were shared with the -EM-FB samples (Fig. 5 ). These numbers were also found higher in + EM compared to -EM for t 0 , t 2 , t 3 , and t 4 . Additionally, a decreasing number of shared ASVs in + EM can be noticed over time (from 34 shared ASVs at t 1 , reaching 11 ASVs at t 6 , Fig. 5 ). In addition to numbers of ASVs shared between the juveniles and the in situ adult sponges, the relative abundance of sequences from these ASVs and their taxonomy was investigated ( Figure S9 ). At t 0 , the dominant shared ASVs within -EM gemmules belong to the Terasakiellaceae family (ASV6: unaffiliated genus, average relative abundance > 20%), while their relative abundance was lower in + EM samples ( Figure S9 ). At t 1 and t 2 , major shared ASVs (average relative abundance > 3%) identified belong to (i) the Flavobacteriaceae (ASV9: Flavobacterium ), (ii) the Rhodobacteraceae (ASV18 and ASV21: Defluviimonas and Tabrizicola , respectively), and (iii) the Terasakiellaceae families (ASV6: unaffiliated genus). From t 2 to t 4 , a higher relative percentage of shared ASVs from Gammaproteobacteria was observed, with the Alteromonadaceae (ASV14: Rheinheimera ) and the Comamonadaceae (ASV65: Paucibacter ). Finally, the Rhodobacteraceae (ASV6: Flavobacterium ) were found mainly dominant in the t 5 and t 6 samples ( Figure S9) . When considering all shared ASVs together, important variations of their relative abundances were observed between and within the treatments, especially at t 1 , t 3 , and t 4 within -EM + FB samples. However, despite these variations, a significant temporal decrease of percentages of shared ASVs was observed from t 0 to t 6 (ANOVA test: p < 0.001; Figure S9 ). Discussion Our experiment was designed (i) to investigate the dynamics of bacterial communities during the first steps of the asexual cycle of Spongilla lacustris , and (ii) to decipher the transmission modes involved in the microbiome assembly. Only few studies experimentally investigated bacterial HT and VT hypotheses during the first steps of the sponge ontogeny [ 16 , 54 – 56 ], and to date, our study brings the first insights for freshwater sponges. Through their sampling accessibility, but also the easiness of hatching and culturing, freshwater sponges such as S. lacustris or Ephydatia muelleri , are promising ressources to better understand such mechanisms [ 28 , 35 ]. Additionally, the gemmule surfaces sterilization protocol [ 34 ], provides a good experimental condition to test the VT scenario on the gemmule surface. Links between the ontogeny of S. lacustris juveniles and the temporal dynamics of their bacterial communities The development of S. lacustris juveniles after gemmule hatching observed during this experiment was similar to E. muelleri [ 34 , 35 ] and Ephydatia fluviatilis [ 57 , 58 ]. In these previous studies, five stages of juveniles development were described: (i) stage 1 (1–2 days after plating, pre- or just-hatching stage): the first stem cells are migrating out of the gemmules through the micropyle; (ii) stage 2 (1–3 days after hatching): the first tissues are growing around the gemmule husk, or on the substrate; (iii) stage 3 (2–4 days after hatching): canal system and choanocytes are formed; (iv) stage 4 (3–5 days after hatching): an osculum is starting to forms while the aquiferous system is still being organized; and (v) stage 5 (4–7 days after hatching): complex branched canals are formed and the osculum is developed. Based on this description and our observation, the first sampling time of our experiment (t 1 : 3 days after plating and 1 day after hatching) corresponds to stage 2, the second (t 2 : 5 days after hatching) corresponds to stage 4, while following ones (from t 3 to t 6 : 9 days after hatching, and beyond) correspond to stage 5. Additionally, another important change in the development of the S. lacustris juveniles can be observed mainly after t 3 with the colonization of Chlorella -like symbionts explaining the slight green coloration observed around the choanocyte chambers. The Chlorella -like colonization was found to be more important from t 4 to t 6 , with a large proportion of tissues showing a green coloration more saturated. These results suggest that the symbiosis with the Chlorella -like symbionts is acquired before t 3 but fully established in the whole juvenile body only after t 4 , under the light condition of our experiment (960 lux for day-light). When looking at the dynamics of the bacterial diversity after t 0 and regardless of the treatment conditions, the main temporal changes can be observed in particular with a continuous shift from the β -diversity from t 1 to t 6 . The specific diversity associated with t 1 samples could be linked to the early development stage of these juveniles since the osculum is not formed yet, while the first cells are emerging around the gemmule husk, and colonizing the substrate (stage 2). This β -diversity difference seems to be explained in terms of composition by a higher relative abundance of Bacteroidota, and more specifically the Flavobacteriaceae (dominated by Flavobacterium ) observed at t 1 compared to the other following sampling times, but also compared to t 0 . These results suggest that a specific development of these pioneer bacterial taxa could be involved in these first steps following the hatching. Interestingly, diverse Bacteroidota strains (including one Flavobacterium sp.) were found to promote the settlement of the larvae of the marine sponge Tedania sp. [ 55 , 59 ]. These strains were either found to form biofilm enhancing the settlement of the larvae, or to excrete chemical cues inducing the larvae settlement through direct secretion or through the production of extracellular vesicles. A similar scenario (Fig. 6 ) could be considered for the settlement of the gemmules of S. lacustris , with an enrichment of Flavobacterium symbionts allowing the first sponge cells to better colonize their substrate (in our case: the gemmule husk and the flat bottom of the 24-well plate). From t 2 to t 4 (stage 4 to 5), the formation of the osculum and the development of the aquiferous system are observed. The β -diversity differences with t 1 can be linked to the ability of the sponge to actively filter the environment. Similar observations were demonstrated for the marine sponge Crambe crambe [ 56 ], as its prokaryotic community was found to change with regard to the osculum formation. In terms of composition, t 2 to t 4 samples are highlighted by a high abundance of Gammaproteobacteria, with three dominant families: Comamonadaceae , Alteromonadaceae and Pseudomonadaceae . These families are commonly occurring in freshwater sponges [ 21 , 31 , 60 ] and might be related to chemical defenses [ 61 ]. For example, Comamonadaceae were associated with defense mechanisms such as CRISPR and intracellular trafficking within E. muelleri [ 31 ], while diverse PKS genes were found within Pseudomonadaceae and Alteromonadaceae (dominant genus: Rheinheimera ) associated with the freshwater sponges Rezinkovia echinata (Lake Baikal) and E. fluviatilis (Vinkeveense Plassen, the Netherlands), respectively [ 62 , 63 ]. This specific bacterial community acquired during the formation of the aquiferous system and the first filter-feeding activity steps could be linked to the early development of bacterial symbionts acting for the chemical defense of the sponge. From t 4 to t 6 (stage 5), the development of an important colonization of the Chlorella -like symbionts in the juvenile tissues, can be considered as a major factor explaining the differences with the previous sampling times (Fig. 6 ). Even if the bacterial communities associated with the Chlorella -like symbionts are similar to those of the freshwater sponge host [ 29 ], a part of this temporal change might be associated with the microalgal enrichment. Additionally, these microalgal symbionts provide a source of nutrients for the juveniles through the production of photosynthates which might also affect the bacterial composition in return. The bacterial community transmitted from the gemmule surface plays an important role in the microbiome stability of the juveniles Our study tends to confirm the efficiency of the sterilization protocol of the gemmule surface developed by Leys et al . [ 34 ]. More precisely, no 16S rRNA sequences were successfully amplified when the gemmules were washed with hydrogen peroxide (-EM gemmules) and their DNA extracted without breaking the gemmule coating. Interestingly, the only way to amplify the 16S rRNA gene from the -EM gemmules was to break the gemmules coatings (by crushing them) before the DNA extraction. This result proves that gemmules host epibacteria within their core, and confirms that bacteria can be transmitted through the gemmule surface but also within the gemmule, as previously suggested [ 29 , 31 ]. However, this result comes in opposition with culture-based observations from Rozenfeld and Curtis, [ 64 ], suggesting a “spontaneous bacterial sterility” inside of E. fluviatilis gemmules. Here, we suggest that the methodology employed didn’t allow to correctly assess such sterility, considering the challenges associated with the culture of sponge-associated bacterial endosymbionts [ 65 ]. qPCR or microscopy analyses (e.g. FISH techniques or environmental SEM) could be considered in future studies to confirm our observation. The sterilization of the gemmule surfaces, resulting in the absence of the EM, was found to be a major factor impacting the bacterial diversity of the gemmules and juveniles. More precisely, the removal of the EM reduces the α -diversity of the juveniles during the first steps of their growth, but also significantly changes the β -diversity all along the experiment (including before the hatching, at t 0 ). Additionally, the β -diversity dispersion was significantly higher without the EM for the early stages. This result suggests that such an epibacterial community is not only distinct from the bacterial diversity transmitted inside of the gemmule, but also participates in the stability of the whole microbiome during early juvenile development (Fig. 6 ). The Anna Karenina principle adapted to sponge holobionts provides a good understanding on the importance of stable microbiomes [ 66 ]. This principle coined that high β -diversity dispersion could result from various dysbiosis scenarios induced by environmental stresses at the holobiont scale. The ability to regulate a stable and less dispersed microbiome over time is then an indicator of healthy microbiomes. In marine environments, the importance of such stability was demonstrated with the Haplosclerida Petrosia ficiformis , during the acquisition of host-specific cyanobacterial symbionts which might provide antioxidants protections for the host [ 67 ]. Differential analysis performed at t 0 and t 1 allowed to identify enriched taxa in + EM compared to -EM, to target bacterial taxa specifically associated with the gemmule surface. The Comamonadaceae and its main genus Hydrogenophaga , appear as major potential taxa attached to the gemmule surface (enriched in + EM condition). This genus is often described as autotrophic hydrogen oxidizers and denitrifiers [ 68 ], living in biofilms such as biofilm reactors [ 69 – 71 ] or epilithic biofilms in lake Baikal [ 72 ]. In this latter environment, this genus has also been observed within the freshwater sponge Baikalospongia fungiformis [ 73 ]. These observations suggest that such taxa might be adapted to colonize biofilms within freshwater sponges, such as the external surface of their gemmules. In this study, the Comamonadaceae family was already hypothesized to play an important role during the development of the aquiferous system. Consequently, the attachment of bacteria to the gemmule surface could constitute an important transmission mode shaping the development of juvenile holobionts. Two hypotheses can be considered about the origin of taxa transmitted on the gemmule surface before the hatching: (i) a VT hypothesis where these taxa are vertically transmitted from the maternal tissue, or (ii) an “early HT” hypothesis where planktonic bacteria directly colonize the gemmule biofilm before or during the planktonic phase of the gemmule (Fig. 6 ). No stable association with the maternal cells would be involved in this case. This early HT hypothesis came then in contrast with both VT hypothesis and the traditional HT hypothesis, where in the latter the acquisition of planktonic colonizers occurs only once the juvenile has hatched, and filters the freshwater. A similar scenario to the early HT was suggested for E. muelleri [ 31 ], since epilithic biofilms were found to have similar bacterial communities to those from gemmules described in [ 35 ]. In our study, only a limited number of ASVs were found to be shared between the gemmules/juveniles and the in situ adult sponges, representing between 2 and 44.7% of the sequences. Surprisingly, almost no ASVs were shared between the gemmules/juveniles and the planktonic community (FF0.45 samples), suggesting that the remaining part of the community is acquired from (i) ultra-rare free-living taxa being undetected, or (ii) other sources than maternal tissues (VT) and free-living planktonic bacteria. In line with this hypothesis, we suggest that the colonization of the gemmule surface could also be achieved through an early HT involving planktonic particle-attached bacteria, even if their contribution could not be directly assessed through this study. As described in marine environments, the contribution of planktonic colonizers in the formation of biofilm can be challenging to assess, but particle-attached bacteria and the ultra-rare taxa are important to consider [ 74 ]. The Terasakiellaceae as a dominant family vertically transmitted inside of the gemmules The β- diversity analysis conducted at t 0 indicated a significant effect of the sterilization of the surface, confirming the difference of bacterial composition transmitted inside of the gemmules compared to their surface. The Terasakiellaceae family are good candidates for these types of taxa transmitted specifically from the maternal tissues to the inside of a gemmule. More precisely, the differential analysis revealed their specificity for the -EM samples at t 0 and t 1 . For the same sampling times, this family also gathers the dominant ASVs shared between the adult sponge and the -EM gemmules. Additionally, within the in situ adult sponge samples, this family was found mainly dominant in the M2 samples but less abundant in the M3 samples. This observation can be explained since the formation of the gemmules and their thesocytes mainly occurred in late November (after M2), while in December (M3) most of the tissue from the adult sponge was found absent and full of gemmules within its skeleton. As the in situ adult sponge samples were extracted mostly with unbroken gemmules, the formation of the gemmules could lead to a lower relative abundance of Terasakiellaceae in the M3 sponge tissues, which are mostly transmitted and located within the gemmules. Consequently, these specific taxa might be important endosymbionts, transmitted to the thesocytes within the gemmule (Fig. 6 ). Further studies are needed to confirm their location and their role as potential endosymbionts. Terasakiellaceae were found in the primmorphs of the diseased freshwater sponge Lubomirskia baikalensis [ 75 ], but also within marine holobionts such as the sponge Suberites massa [ 76 ], as well as an important diversity of corals [ 77 – 80 ]. Taxa within this family are also known as nitrogen fixers [ 81 ], and may play an important role for the nitrogen regulation within the gemmule. Others taxa enriched in the -EM samples such as the Rhizobiaceae family and its dominant genus Rhizobium (covering also Allorhizobium , Neorhizobium and Parararhizobium ) known for its denitrification role [ 82 , 83 ], might also play similar functions within the gemmule. To date, little is known about the role of bacterial symbionts within sponge cells undergoing diapause states such as the thesocytes cells within the gemmules. The expression of glutamine metabolism, apoptotic process, and oxidation-reduction system was found to be specific to stage 0 [ 35 ]. Further studies are needed to better investigate the potential link between these specific metabolisms and these putative endosymbionts stored in the gemmules. Effect of the presence of exogenous free-living bacteria during the juveniles growth: horizontal acquisition of symbionts, or additional food source (FS)? Another major result of this study was the effect of the absence/presence of ambient free-living bacteria on the bacterial β -diversity of the juveniles. These differences can be explained through two hypotheses: (i) an HT hypothesis, related to the acquisition of bacterial symbionts from the medium, and (ii) a food source (FS) hypothesis, where the bacterioplankton is filtered by the sponge for its nutrition. In this scenario, this specific food source would be linked to a different physiology of the juveniles impacting indirectly their microbiome. As mainly discussed for marine sponges, the delineation between these two processes can be hard to distinguish [ 7 ]. In this study, only three low abundant ASVs ( Polynucleobacter , Candidatus Limnoluna , and Bradyrhizobium ; each < 0.04%) were occasionally shared between the prefiltered water (FF0.45 samples) and the juvenile sponges, suggesting that the horizontal acquisition of FL bacteria by the juveniles was nearly nonexistent during this experiment. The effect of the presence of the FL bacteria in the medium on the β -diversity is then more likely explained by the FS hypothesis (Fig. 6 ). Additionally, the α -diversity analysis also goes in favor of the rejection of the HT hypothesis, since the richness (estimated with Chao1 index) of the + FB juveniles was not significantly higher compared to the -FB. From t 1 to t 3 , the difference of β -diversity between + FB and -FB samples was mainly observed under the -EM condition, while being less important in + EM. Consequently, the absence of epibiotic microbiome (EM) increases the effect of the absence of ambient free-living bacteria (FB). Their absence in the medium might then represent cumulative stress to the juveniles growing with an unstable microbiome due to the EM absence. In line with the FS hypothesis, the absence of FB as a food source might explain this as an additional source of stress in + EM, leading to a distinct bacterial community. As the differences between + FB and -FB were linked to the food source availability, these differences can be investigated in terms of trophic strategy within the bacterial community. Indeed, the juveniles growing with more food could provide more nutrients for their associated microbiome. Interestingly, this assumption appears to be particularly consistent with the specific taxa found enriched in + FB or -FB samples, based on differential analyses at t 1 . For example, several + FB enriched taxa such as Oxalobacteraceae (within + EM samples), but also Pseudomonas and Flavobacterium (within -EM samples) are classically known as copiotroph [ 84 , 85 ]. Conversely, -FB enriched taxa such as Rheinemera (in + EM) or Brevundimonas and Caulobacter (in both -EM and + EM) are typically oligotrophic bacteria [ 85 , 86 ]. This observation strongly supports the FS hypothesis, indicating that in the absence of FB, less nutrients can be provided to the microbiome, resulting in a higher relative abundance of oligotrophic taxa. Conversely, the presence of FB turned into a source of organic matter by the juveniles, could favor copiotrophs that could take advantage of this condition and quickly grow within the sponge (i.e. r-strategists bacteria). Finally, even if almost no ASVs from the prefiltered water were found to be horizontally acquired within the juveniles, a slightly higher number of these free-living ASVs were found to be shared with the in situ sponges (17 in total, with dominant ones belonging to Sporichthyaceae ). This observation suggests that horizontal acquisition within S. lacustris could still occasionally happen after a longer term of development. Further studies on these transmissions in natural conditions could provide relevant insight to better consider the complete dynamics of the holobiont through its full life cycle. Conclusion The bacterial diversity within S. lacustris juveniles was found to be shaped by three factors: the life cycle stage, the presence of epibacteria on the gemmule, and finally the presence of ambient free-living bacteria. The osculum formation together with the development of a canal system, could lead to an active filtration of the environment which might induce a change in the microbiome β -diversity. Thereafter, the colonization of the Chlorella -like symbionts could also provide an additional niche for the development of new bacterial symbionts within the holobiont. Our study revealed a complex diversity of microbial acquisition modes within the S. lacustris holobiont model. For instance, multiple vertical acquisition scenarios can be considered with both transmission within the gemmule or on its surface. Importantly, the transmission of the microbiome on the gemmule surface was found to be essential for the whole holobiont stability during the first days. In line with the Anna Karenina principle [ 66 ], the absence of these epibacterial communities might represent a stress condition. While a recruitment of free-living bacteria by filtration of the juveniles was found to be nearly impossible, an alternative horizontal transmission scenario can however be considered with a colonization of planktonic bacteria on the gemmule biofilm, before the hatching. The community of the gemmule biofilm would then be composed of both vertically and horizontally transmitted bacteria, in line with the mixed-mode transmission hypothesis [ 9 ]. Such diversity and complexity of transmission modes need to be better considered in future studies, and our results associated with the importance of the microbiome on the gemmule surface provide new perspectives that could also be investigated for sponge larvae. Abbreviations ASV: Amplicon Sequence Variant EM: Epibiotic Microbiome FF0.2: Freshwater filtered upon 0.2μm pore size filters FF0.45: Freshwater filtered upon 0.45μm pore size filters FB: Free-living Bacteria FS: Food Source HT: Horizontal Transmission SD: Standard Deviation VT: Vertical Transmission Declarations Availability of data and material 16S rRNA gene sequences were deposited and are publicly available in the NCBI Sequences Read Archive (SRA) under the BioProject ID PRJNA1077127, accession number. The R scripts used for all the 16S rRNA gene metabarcoding analysis can be found at https://github.com/BenoitPAIX/Gemmules_microbiome Competing interests The authors declare that they have no competing interests. Funding This work was funded by the NWO-VIDI with project number 16.161.301. Authors' contributions BP, EvdV and NdV designed the experiment and performed the fieldwork, BP and EvdV performed the lab work, BP processed the data, analyzed the results and wrote the first draft of the manuscript. NdV and BP reviewed the manuscript and finalized the last draft. Acknowledgements We thank Mike Hynes, Anouk Langerak and Jerry Weidema for their help during the sampling in October. References de Goeij JM, Lesser MP, Pawlik JR. Nutrient Fluxes and Ecological Functions of Coral Reef Sponges in a Changing Ocean. In: Carballo JL, Bell JJ, editors. Clim Change Ocean Acidif Sponges Impacts Mult Levels Organ. Cham: Springer International Publishing; 2017 p. 373–410. Slaby BM, Franke A, Rix L, Pita L, Bayer K, Jahn MT, et al. Marine sponge holobionts in health and disease. In: Li Z, editor. Symbiotic Microbiomes Coral Reefs Sponges Corals . Dordrecht: Springer Netherlands; 2019 p. 81–104. Thomas T, Moitinho-Silva L, Lurgi M, Björk JR, Easson C, Astudillo-García C, et al. Diversity, structure and convergent evolution of the global sponge microbiome. Nat Commun. 2016;7:11870. Posadas N, Baquiran JIP, Nada MAL, Kelly M, Conaco C. Microbiome diversity and host immune functions influence survivorship of sponge holobionts under future ocean conditions. ISME J. 2022;16:58–67. De Castro-Fernández P, Ballesté E, Angulo-Preckler C, Biggs J, Avila C, García-Aljaro C. How does heat stress affect sponge microbiomes? Structure and resilience of microbial communities of marine sponges from different habitats. Front Mar Sci. 2023;9. Pita L, Rix L, Slaby BM, Franke A, Hentschel U. The sponge holobiont in a changing ocean: from microbes to ecosystems. Microbiome. 2018;6:46. Díez-Vives C, Koutsouveli V, Conejero M, Riesgo A. Global patterns in symbiont selection and transmission strategies in sponges. Front Ecol Evol . 2022;10. Björk JR, Díez-Vives C, Astudillo-García C, Archie EA, Montoya JM. Vertical transmission of sponge microbiota is inconsistent and unfaithful. Nat Ecol Evol. 2019;3:1172–83. de Oliveira BFR, Freitas-Silva J, Sánchez-Robinet C, Laport MS. Transmission of the sponge microbiome: moving towards a unified model. Environ Microbiol Rep. 2020;12:619–38. Sipkema D, de Caralt S, Morillo JA, Al-Soud WA, Sørensen SJ, Smidt H, et al. Similar sponge-associated bacteria can be acquired via both vertical and horizontal transmission. Environ Microbiol. 2015;17:3807–21. Carrier TJ, Maldonado M, Schmittmann L, Pita L, Bosch TCG, Hentschel U. Symbiont transmission in marine sponges: reproduction, development, and metamorphosis. BMC Biol. 2022;20:100. Vrijenhoek RC. Genetics and evolution of Deep-Sea chemosynthetic bacteria and their invertebrate hosts. In: Kiel S, editor. Vent Seep Biota Asp Microbes Ecosyst . Dordrecht: Springer Netherlands; 2010. p. 15–49. Ebert D. The epidemiology and evolution of symbionts with mixed-mode transmission. Annu Rev Ecol Evol Syst. 2013;44:623–43. Ereskovsky AV, Gonobobleva E, Vishnyakov A. Morphological evidence for vertical transmission of symbiotic bacteria in the viviparous sponge Halisarca dujardini Johnston (Porifera, Demospongiae, Halisarcida). Mar Biol. 2005;146:869–75. Sharp KH, Eam B, Faulkner DJ, Haygood MG. Vertical transmission of diverse microbes in the tropical sponge Corticium sp. Appl Environ Microbiol. 2007;73:622–9. Engelberts JP, Abdul Wahab MA, Maldonado M, Rix L, Marangon E, Robbins SJ, et al. Microbes from Mum: symbiont transmission in the tropical reef sponge Lanthella basta . ISME Commun. 2022;2:1–12. Carrier TJ, Schmittmann L, Jung S, Pita L, Hentschel U. Maternal provisioning of an obligate symbiont in a sponge. Ecol Evol. 2023;13:e10012. Strano F, Micaroni V, Thomas T, Woods L, Davy SK, Bell JJ. Marine heatwave conditions drive carryover effects in a temperate sponge microbiome and developmental performance. Proc R Soc B Biol Sci. 2023;290:20222539. Turon M, Cáliz J, Garate L, Casamayor EO, Uriz MJ. Showcasing the role of seawater in bacteria recruitment and microbiome stability in sponges. Sci Rep. 2018;8:15201. de Oliveira BFR, Lopes IR, Canellas ALB, Muricy G, Dobson ADW, Laport MS. Not that close to mommy: horizontal transmission seeds the microbiome associated with the marine sponge Plakina cyanorosea . Microorganisms. 2020;8:1978. Lo Giudice A, Rizzo C. Freshwater Sponges as a neglected reservoir of bacterial biodiversity. Microorganisms. 2024;12:25. Pronzato R, Pisera A, Manconi R. Fossil freshwater sponges: taxonomy, geographic distribution, and critical review. Acta Palaeontol Pol. 2017;62:468–95. Manconi R, Pronzato R. Global diversity of sponges (Porifera: Spongillina) in freshwater. In: Balian EV, Lévêque C, Segers H, Martens K, editors. Freshw Anim Divers Assess . Dordrecht: Springer Netherlands; 2008 Gilbert JJ, Allen HL. Studies on the physiology of the green freshwater sponge Spongilla lacustris : primary productivity, organic matter, and chlorophyll content. SIL Proc 1922-2010. 1973;18:1413–20. Wilkinson CR. Nutrient translocation from green algal symbionts to the freshwater sponge Ephydatia fluviatilis . Hydrobiologia. 1980;75:241–50. Cook CB. Metabolic interchange in algae-invertebrate symbiosis. Intern Rev Cytol Suppl. 1983;177–210. Kulakova NV, Sakirko MV, Adelshin RV, Khanaev IV, Nebesnykh IA, Pérez T. Brown rot syndrome and changes in the bacterial сommunity of the Baikal sponge Lubomirskia baikalensis . Microb Ecol. 2018;75:1024–34. Hall C, Camilli S, Dwaah H, Kornegay B, Lacy C, Hill MS, et al. Freshwater sponge hosts and their green algae symbionts: a tractable model to understand intracellular symbiosis. PeerJ. 2021;9:e10654. Hustus K, Díez-Vives C, Mitsi K, Nutakki J, Kering V, Nguyen IT, et al. Algal symbionts of the freshwater sponge Ephydatia muelleri . Symbiosis. 2023 Costa R, Keller-Costa T, Gomes NCM, da Rocha UN, van Overbeek L, van Elsas JD. Evidence for selective bacterial community structuring in the freshwater sponge Ephydatia fluviatilis . Microb Ecol. 2013;65:232–44. Sugden S, Holert J, Cardenas E, Mohn WW, Stein LY. Microbiome of the freshwater sponge Ephydatia muelleri shares compositional and functional similarities with those of marine sponges. ISME J. 2022;16:2503–12. Manconi R, Pronzato R. Rediscovery of the type material of Spongilla lacustris (L., 1759) in the Linnean herbarium. Ital J Zool. 2000;67:89–92. Manconi R, Pronzato R. Life cycle of Spongilla lacustris (Porifera, Spongillidae): a cue for environment-dependent phenotype. Hydrobiologia. 1991;220:155–60. Leys S, Grombacher L, Hill A. Hatching and freezing gemmules from the freshwater sponge Ephydatia muelleri . 2019 Kenny NJ, Francis WR, Rivera-Vicéns RE, Juravel K, de Mendoza A, Díez-Vives C, et al. Tracing animal genomic evolution with the chromosomal-level assembly of the freshwater sponge Ephydatia muelleri . Nat Commun. 2020;11:3676. Ruperti F, Becher I, Stokkermans A, Wang L, Marschlich N, Potel C, et al. Molecular profiling of sponge deflation reveals an ancient relaxant-inflammatory response. Curr Biol. 2024 Gernert C, Glöckner FO, Krohne G, Hentschel U. Microbial diversity of the freshwater sponge Spongilla lacustris . Microb Ecol. 2005;50:206–12. Graffius S, Garzón JFG, Zehl M, Pjevac P, Kirkegaard R, Flieder M, et al. Secondary metabolite production potential in a microbiome of the freshwater sponge Spongilla lacustris . Microbiol Spectr. 2023;11:e04353-22. Turner CR, Miller DJ, Coyne KJ, Corush J. Improved methods for capture, extraction, and quantitative assay of environmental DNA from asian bigheaded carp ( Hypophthalmichthys spp.). PLOS ONE. 2014;9:e114329. Klindworth A, Pruesse E, Schweer T, Peplies J, Quast C, Horn M, et al. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 2013;41:e1. Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: High-resolution sample inference from Illumina amplicon data. Nat Methods. 2016;13:581–3. Callahan BJ, McMurdie PJ, Holmes SP. Exact sequence variants should replace operational taxonomic units in marker-gene data analysis. ISME J. 2017;11:2639–43. Callahan BJ, Sankaran K, Fukuyama JA, McMurdie PJ, Holmes SP. Bioconductor Workflow for Microbiome Data Analysis: from raw reads to community analyses . F1000Research; 2016 Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 2013;41:D590–6. McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLOS ONE. 2013;8:e61217. 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;6:226. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, et al. vegan: Community ecology package. 2019. Mendiburu F de. Agricolae: Statistical procedures for agricultural research. 2019. McMurdie PJ, Holmes S. Waste not, want not: why rarefying microbiome data is inadmissible. PLOS Comput Biol. 2014;10:e1003531. Gloor GB, Macklaim JM, Pawlowsky-Glahn V, Egozcue JJ. Microbiome datasets are compositional: and this is not optional. Front Microbiol. 2017;8:2224. Foster ZSL, Sharpton TJ, Grünwald NJ. Metacoder: An R package for visualization and manipulation of community taxonomic diversity data. PLOS Comput Biol. 2017;13:e1005404. Lahti L, Shetty S, Blake T, Salojarvi J. Tools for microbiome analysis in R. Version. 2017;1:28. Larsson J, Godfrey AJR, Gustafsson P, algorithms) DHE (geometric, code) EH (root solver, Privé F. eulerr: Area-Proportional Euler and Venn Diagrams with Ellipses . 2022. Fieth RA, Gauthier M-EA, Bayes J, Green KM, Degnan SM. Ontogenetic changes in the bacterial symbiont community of the tropical demosponge Amphimedon queenslandica : metamorphosis is a new beginning. Front Mar Sci . 2016;3. Wu S, Ou H, Liu T, Wang D, Zhao J. Structure and dynamics of microbiomes associated with the marine sponge Tedania sp. during its life cycle. FEMS Microbiol Ecol. 2018;94:fiy055. Turon M, Ford M, Maldonado M, Sitjà C, Riesgo A, Díez-Vives C. Microbiome transmission through the ontogeny of the marine sponge Crambe crambe . 2023; Funayama N, Nakatsukasa M, Hayashi T, Agata K. Isolation of the choanocyte in the fresh water sponge, Ephydatia fluviatilis and its lineage marker, Ef annexin. Dev Growth Differ. 2005;47:243–53. Bart MC, Vet SJ de, Bakker DM de, Alexander BE, Oevelen D van, Loon EE van, et al. Spiculous skeleton formation in the freshwater sponge Ephydatia fluviatilis under hypergravity conditions. PeerJ. 2019;6:e6055. Li M, Wang K, Jia C, Liu T, Yang S, Ou H, et al. Bacteroidetes bacteria, important players in the marine sponge larval development process. iScience. 2021;24:102662. Laport MS, Pinheiro U, Rachid CTC da C. Freshwater sponge Tubella variabilis presents richer microbiota than marine sponge species. Front Microbiol . 2019;10. Gaikwad S, Shouche YS, Gade WN. Microbial community structure of two freshwater sponges using Illumina MiSeq sequencing revealed high microbial diversity. AMB Express. 2016;6:40. Keller-Costa T, Jousset A, van Overbeek L, van Elsas JD, Costa R. The freshwater sponge Ephydatia fluviatilis harbours diverse Pseudomonas species (Gammaproteobacteria, Pseudomonadales) with broad-spectrum antimicrobial activity. PloS One. 2014;9:e88429. Kaluzhnaya OV, Itskovich VB. Diversity of potential producers of bioactive metabolites having polyketide nature in the Baikal sponge community of Rezinkovia echinata . Limnol Freshw Biol. 2020;423–8. Rozenfeld F, Curtis ASG. A technique for sterile culture of fresh water sponges. Experientia. 1980;36:371–3. Dat TTH, Steinert G, Cuc NTK, Smidt H, Sipkema D. Bacteria cultivated from sponges and bacteria not yet cultivated from sponges—a review. Front Microbiol . 2021;12. Zaneveld JR, McMinds R, Vega Thurber R. Stress and stability: applying the Anna Karenina principle to animal microbiomes. Nat Microbiol. 2017;2:1–8. Britstein M, Cerrano C, Burgsdorf I, Zoccarato L, Kenny NJ, Riesgo A, et al. Sponge microbiome stability during environmental acquisition of highly specific photosymbionts. Environ Microbiol. 2020;22:3593–607. Willems A, Gillis M. Hydrogenophaga . In: Whitman WB, editor. Bergeys Man Syst Archaea Bact . 1st ed. Wiley; 201. p. 1–15. Zhao H-P, Van Ginkel S, Tang Y, Kang D-W, Rittmann B, Krajmalnik-Brown R. Interactions between perchlorate and nitrate reductions in the biofilm of a hydrogen-based membrane biofilm reactor. Environ Sci Technol. 2011;45:10155–62. Zhou L, Xu X, Xia S. Effects of sulfate on simultaneous nitrate and selenate removal in a hydrogen-based membrane biofilm reactor for groundwater treatment: Performance and biofilm microbial ecology. Chemosphere. 2018;211:254–60. Rezaei R, Aghapour AA, Chavshin AR, Bargeshadi R. Biodegradation of the atenolol and nitrogen removal using the sequencing batch biofilm reactor. Bioresour Technol Rep. 2022;18:101109. Sukhanova EV, Shtykova YuR, Suslova MYu, Pestunova OS, Kostornova TYa, Khanaev IV, et al. Diversity and physiological and biochemical properties of heterotrophic bacteria isolated from lake Baikal epilithic biofilms. Microbiology. 2019;88:324–34. Kaluzhnaya OV, Itskovich VB. Features of diversity of polyketide synthase genes in the community of freshwater sponge Baikalospongia fungiformis . Russ J Genet. 2022;58:336–46. Catão E, Pollet T, Garnier C, Barry-Martinet R, Rehel K, Linossier I, et al. Temperate and tropical coastal waters share relatively similar microbial biofilm communities while free-living or particle-attached communities are distinct. Mol Ecol. 2021;30:2891–904. Chernogor L, Klimenko E, Khanaev I, Belikov S. Microbiome analysis of healthy and diseased sponges Lubomirskia baicalensis by using cell cultures of primmorphs. PeerJ. 2020;8:e9080. Lamb CE, Watts JEM. Microbiome species diversity and seasonal stability of two temperate marine sponges Hymeniacidon perlevis and Suberites massa . Environ Microbiome. 2023;18:52. Weiler BA, Verhoeven JTP, Dufour SC. Bacterial communities in tissues and surficial mucus of the cold-water coral Paragorgia arborea . Front Mar Sci . 2018;5. Rosales SM, Clark AS, Huebner LK, Ruzicka RR, Muller EM. Rhodobacterales and Rhizobiales are associated with stony coral tissue loss disease and its suspected sources of transmission. Front Microbiol . 2020;11. Li J, Long L, Zou Y, Zhang S. Microbial community and transcriptional responses to increased temperatures in coral Pocillopora damicornis holobiont. Environ Microbiol. 2021;23:826–43. Quintanilla E, Rodrigues CF, Henriques I, Hilário A. Microbial associations of abyssal Gorgonians and Anemones (>4,000 m depth) at the Clarion-Clipperton fracture zone. Front Microbiol . 2022;13 Tiedje J. Ecology of denitrification and dissimilatory nitrate reduction to ammonium. Methods Soil Anal Part 2 Chem Microbiol Prop. 1988. p. 179–244. Lv Y, Chen X, Zhang X, Zhu C, Pan Y, Sun T, et al. Denitrification for acidic wastewater treatment: Long-term performance, microbial communities, and nitrous oxide emissions. J Biosci Bioeng. 2022;134:513–20. Long Y, Ma Y, Wan J, Wang Y, Tang M, Fu H, et al. Denitrification efficiency, microbial communities and metabolic mechanisms of corn cob hydrolysate as denitrifying carbon source. Environ Res. 2023;221:115315. Schlatter DC, Hansen JC, Schillinger WF, Sullivan TS, Paulitz TC. Common and unique rhizosphere microbial communities of wheat and canola in a semiarid Mediterranean environment. Appl Soil Ecol. 2019;144:170–81. Hellauer K, Michel P, Holland SI, Hübner U, Drewes JE, Lauro FM, et al. Inferring trophic conditions in managed aquifer recharge systems from metagenomic data. Sci Total Environ. 2021;772:145512. Wilhelm RC. Following the terrestrial tracks of Caulobacter - redefining the ecology of a reputed aquatic oligotroph. ISME J. 2018;12:3025–37. Additional Declarations No competing interests reported. Supplementary Files SIgemmulesgrowthmicrobiomesubmissionEM.docx Cite Share Download PDF Status: Published Journal Publication published 08 Jun, 2024 Read the published version in Environmental Microbiome → Version 1 posted Editorial decision: Revision requested 21 May, 2024 Reviews received at journal 29 Apr, 2024 Reviews received at journal 23 Apr, 2024 Reviewers agreed at journal 18 Apr, 2024 Reviewers agreed at journal 17 Apr, 2024 Reviewers invited by journal 17 Apr, 2024 Editor assigned by journal 03 Mar, 2024 Submission checks completed at journal 21 Feb, 2024 First submitted to journal 20 Feb, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3973150","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":274152446,"identity":"58c5fd65-706c-4596-8d69-185a24390780","order_by":0,"name":"Benoit Paix","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA50lEQVRIiWNgGAWjYBACAxCRACKYGRgfACkePuK0JIC1MIM4PGxEaYFYw8AmASYJaTFn7z344OEPhjz+dh6zyq85djJsDMwPH93Ao8Wy51yyAdBhxRKH2dJuy25LBjqMzdg4B5/DbuSYSQC1JDYcZj52W3IbM1ALD5s0Xi3330C0zD/M2FYsua2eCC03eCBaNgBtYfy47TBhLZY9OcYGCWkSxYaH2ZKlGbcd52FjJuAXc/Yzhg9/2NjkyZ0/Y/jx57Zqe3725oeP8WmBAokEEMnMAyYJKwcDsBbGH0SqHgWjYBSMgpEFAJsuQQZzR/ntAAAAAElFTkSuQmCC","orcid":"","institution":"Naturalis Biodiversity Center","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Benoit","middleName":"","lastName":"Paix","suffix":""},{"id":274152447,"identity":"3afe911d-febb-4aed-b9a2-3c3771c3af7a","order_by":1,"name":"Elodie van der Valk","email":"","orcid":"","institution":"Naturalis Biodiversity Center","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Elodie","middleName":"van der","lastName":"Valk","suffix":""},{"id":274152448,"identity":"5ad006a8-b763-4f46-b2bb-32c8cdd81c98","order_by":2,"name":"Nicole J. de Voogd","email":"","orcid":"","institution":"Naturalis Biodiversity Center","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Nicole","middleName":"J.","lastName":"de Voogd","suffix":""}],"badges":[],"createdAt":"2024-02-20 15:16:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3973150/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3973150/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s40793-024-00580-7","type":"published","date":"2024-06-08T14:48:47+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":51526316,"identity":"4b5d4df6-5d0e-46c2-8f41-d095115b47e6","added_by":"auto","created_at":"2024-02-23 05:48:18","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":473430,"visible":true,"origin":"","legend":"\u003cp\u003eWorkflow summarizing the experimental design of the study. Abbreviations: EM for epibiotic microbiome, and FB for free-living bacteria\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-3973150/v1/821bfb05f94f4861446cb0f4.png"},{"id":51525999,"identity":"8b679d2a-88f6-465b-8159-81d3667a822b","added_by":"auto","created_at":"2024-02-23 05:40:18","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":130993,"visible":true,"origin":"","legend":"\u003cp\u003eDynamics of the \u003cem\u003eα\u003c/em\u003e-diversity (Shannon index) of the bacterial communities associated with the gemmules (t\u003csub\u003e0\u003c/sub\u003e) and the \u003cem\u003ein vitro\u003c/em\u003e juvenile sponges (t\u003csub\u003e1\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e). Abbreviations: EM for epibiotic microbiome, and FB for free-living bacteria.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-3973150/v1/61b61b658536f818db14131a.png"},{"id":51526004,"identity":"a2c0a7c0-fe40-4305-965f-e4d1ac93d4e5","added_by":"auto","created_at":"2024-02-23 05:40:18","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":245898,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eβ\u003c/em\u003e-diversity of the bacterial communities (Bray-Curtis distances). \u003cstrong\u003eA\u003c/strong\u003e NMDS plot performed with all samples represented according to their sample type and sampling time. \u003cstrong\u003eB\u003c/strong\u003eNMDS plots performed for each sampling time of the \u003cem\u003ein vitro\u003c/em\u003e juveniles, separately, with samples represented according to their treatments. Abbreviations: EM for epibiotic microbiome, and FB for free-living bacteria.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-3973150/v1/80480111abefe48495c35fbc.png"},{"id":51526002,"identity":"9de366cc-28f2-418c-a2b3-2d5db00f2532","added_by":"auto","created_at":"2024-02-23 05:40:18","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1206974,"visible":true,"origin":"","legend":"\u003cp\u003ePhylogenetical heat trees performed with juvenile sponges samples collected at t\u003csub\u003e1\u003c/sub\u003e and representing the taxa significantly and differentially abundant between the different treatments. \u003cstrong\u003eA\u003c/strong\u003e and \u003cstrong\u003eB\u003c/strong\u003e Differentially abundant taxa between +EM and -EM samples, for +FB and -FB samples, respectively. \u003cstrong\u003eC\u003c/strong\u003e and \u003cstrong\u003eD\u003c/strong\u003e Differentially abundant taxa between +FB and -FB samples, for -EM and +EM samples, respectively. For each taxon, (i) the colors of their associated nodes corresponds to the log2 fold change between the ratio of the mean relative abundance within each treatment, (ii) the size of the nodes corresponds to the relative abundance of each taxon. Abbreviations: EM for epibiotic microbiome, and FB for free-living bacteria.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-3973150/v1/a192cd6cb56f5b2151ebaebd.png"},{"id":51526005,"identity":"3e699e0d-2576-4e9d-984b-000bae6ef5c1","added_by":"auto","created_at":"2024-02-23 05:40:19","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":103311,"visible":true,"origin":"","legend":"\u003cp\u003eDynamics of number of ASVs shared between gemmules/\u003cem\u003ein vitro\u003c/em\u003e juveniles and \u003cem\u003ein situ \u003c/em\u003eadult sponges. Abbreviations: EM for epibiotic microbiome, and FB for free-living bacteria.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-3973150/v1/ede1940321525aa543fd409e.png"},{"id":51526001,"identity":"c4e71e3f-d172-4089-9564-a281ac02e166","added_by":"auto","created_at":"2024-02-23 05:40:18","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":518118,"visible":true,"origin":"","legend":"\u003cp\u003eIllustration summarizing the potential transmission modes and the dynamics of the epibacterial community during the first life stages of the asexual cycle of \u003cem\u003eS. lacustris\u003c/em\u003e. The gemmule formed within the maternal tissue, encapsulate undifferentiated dormant cells: the thesocytes. Three transmission modes can be considered during the gemmule formation: (\u003cstrong\u003e1A\u003c/strong\u003e) a vertical transmission of bacteria inside of the gemmule (e.g. \u003cem\u003eTerasakiellaceae\u003c/em\u003e that might be associated with the thesocytes); (\u003cstrong\u003e1B\u003c/strong\u003e) a vertical transmission of bacteria on the gemmule surface within its biofilm (e.g. \u003cem\u003eHydrogenophaga\u003c/em\u003e), increasing the overall \u003cem\u003eα\u003c/em\u003e-diversity; (\u003cstrong\u003e1C\u003c/strong\u003e) a horizontal acquisition of ambient bacteria which are not directly associated with the maternal tissues (e.g. potential particle-attached planktonic colonizers). Stage 0: following its planktonic phase, the gemmule is in contact with a substrate, and (\u003cstrong\u003e2\u003c/strong\u003e) the epibacteria from the gemmule biofilm can colonize this new surface. These epibacteria (e.g. Bacteroidota such as \u003cem\u003eFlavobacterium\u003c/em\u003e) might enhance the settlement of the gemmule and promote its hatching. Stage 1 to 2: the gemmule is hatching, with the thesocytes emerging out of the micropyle, together with the bacteria initially inside (\u003cstrong\u003e3\u003c/strong\u003e). The bacteria from the gemmule surface play an important role for the whole stability of the juvenile holobiont (\u003cstrong\u003e4\u003c/strong\u003e). Stage 3 to 4: the juvenile is developing its canal system and forming an osculum, while (\u003cstrong\u003e5\u003c/strong\u003e) the filtered free-living bacteria (FB) are used as a source of nutrients allowing an enrichment of copiotroph bacteria (\u003cstrong\u003e6\u003c/strong\u003e). Stage 5 and beyond: the increasing colonization of \u003cem\u003eChlorella\u003c/em\u003e-like symbionts, is associated with a shift of bacterial composition (\u003cstrong\u003e7\u003c/strong\u003e). Adult stage of \u003cem\u003eSpongilla lacustris\u003c/em\u003e: the tissues are forming a new generation of gemmules before winter. The different elements of the illustration are not to scale.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-3973150/v1/3e176f1ef4470395f1196982.png"},{"id":58822089,"identity":"1b060439-ed17-40dc-995c-c7bcada97a3a","added_by":"auto","created_at":"2024-06-21 16:30:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3291671,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3973150/v1/a7e9d81a-14b7-4841-bedc-cf2e5894b13b.pdf"},{"id":51526003,"identity":"2e8e9cf5-a9de-47c7-b187-01118d9d23ed","added_by":"auto","created_at":"2024-02-23 05:40:18","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3005455,"visible":true,"origin":"","legend":"","description":"","filename":"SIgemmulesgrowthmicrobiomesubmissionEM.docx","url":"https://assets-eu.researchsquare.com/files/rs-3973150/v1/b54a5b77aadfa74336dc61b7.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Dynamics, diversity, and roles of bacterial transmission modes during the first asexual life stages of the freshwater sponge Spongilla lacustris","fulltext":[{"header":"Introduction","content":"\u003cp\u003eSponges (phylum Porifera) are filter-feeding and sessiles animals known for their tight interactions with their microbiome. Described as a functional entity, the association of a sponge and its microbiome called \u0026ldquo;sponge-holobiont\u0026rdquo; represents an important engineering system of benthic environments (e.g. through the cycling of the organic matter, [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]). Within these holobionts, the host provides a habitat for microbial symbionts, who in return displays key functions for the host (e.g. by producing chemical defenses, or supplying additional nutrients; [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]). Sponge microbiomes are often described to be diverse, stable and host-specific [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e], playing an essential role for the resilience of their host observed during environmental changes (e.g. under heat-stress or acidification, [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]). The mechanisms associated with the selection of microbial symbionts and their stable maintenance are increasingly studied for marine sponge models [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Among others, stochasticity processes seem to be involved through the acquisition of microbial communities [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. The current body of knowledge suggests that marine sponges can acquire their microbial communities through both vertical and horizontal transmission (\u0026ldquo;VT\u0026rdquo; and \u0026ldquo;HT\u0026rdquo;, respectively; [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]). This intermediate transmission mode was initially named \u0026ldquo;leaky vertical transmission\u0026rdquo; [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], or \u0026ldquo;mixed-mode transmission\u0026rdquo; [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. While VT was confirmed for a large diversity of sponge models [\u003cspan additionalcitationids=\"CR15 CR16 CR17\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e], it appears insufficient to explain the origin of the overall bacterial community [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. For example, larvae of eight Mediterranean sponge species shared only 17% of the bacterial ASVs with their parents, on average [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Additionally, the importance of horizontal transmission was discussed for 19 sponge species from Vietnam, as their bacterial core community was found to be highly shared (\u0026gt;\u0026thinsp;50%) with the core bacterial community of the ambient seawater [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. In this latter study, species-specific recognition mechanisms were hypothesized for the sponges, allowing an enrichment of specific planktonic bacteria. However, the diversity and relative importance of these transmission modes are still unknown within freshwater sponges, as their bacterial diversity remains largely underexplored [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFreshwater sponges (Order Spongillida) comprise approximately 240 species [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e], and display a wide range of environmental adaptation within diverse ecosystems such as lakes, rivers, streams, ponds, or urban canals [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. They can survive drastic changes of temperatures and light, but also desiccation and anoxic conditions, and can also tolerate high levels of pollutants [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Under light-exposed conditions, the establishment of stable interaction with photosymbionts (e.g. microalgae such as \u003cem\u003eChlorella\u003c/em\u003e spp.) constitutes an important factor for the physiology of freshwater sponges harboring a green coloration [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. Up to 20% of the photosynthates produced by the microalgal partners can be translocated directly to the host. [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. However, symbiotic association within freshwater sponges can be disturbed, leading for example to a dysbiosis state for the baikal sponge \u003cem\u003eLubomirskia baikalensis\u003c/em\u003e harboring the brown rot syndrome associated with bleached tissues [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. Genetic responses to the presence of the \u003cem\u003eChlorella\u003c/em\u003e-like green symbionts revealed their importance for the host, as immunity mechanisms are associated with the algal symbiont recognition [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. These results were demonstrated through the infection of aposymbiotic sponge \u003cem\u003eEphydatia muelleri\u003c/em\u003e with its algae, suggesting that such symbiosis can also be horizontally acquired [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e], in addition to the vertical transmission [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. However, the mechanisms associated with the recognition and the selection of a host-specific bacterial community from their environment are yet to be explored for freshwater sponges [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Indeed, the bacterial composition has been rarely compared between freshwater sponge species, but also with those from the ambient environment (e.g. from freshwater or sediment, [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]). Sugden \u003cem\u003eet al.\u003c/em\u003e showed that bacterial communities of \u003cem\u003eE. muelleri\u003c/em\u003e are mainly distinct from those from the surrounding freshwater and biofilms [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Geographical differences were also observed between rivers, suggesting that the environment could also shape the host-microbe specificity. In this latter study, horizontal acquisition from ambient bacteria was suggested as an important factor that could explain the rivers-specific microbiome, while the relative importance of the vertical transmission was still uncertain.\u003c/p\u003e \u003cp\u003e \u003cem\u003eSpongilla lacustris\u003c/em\u003e (Linnaeus, 1759) is one of the most widespread freshwater sponge species in temperate areas of the Northern Hemisphere [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. As for \u003cem\u003eE. muelleri\u003c/em\u003e, the asexual reproduction of the \u003cem\u003eS. lacustris\u003c/em\u003e model has been extensively described through the production (in autumn) of diapausing cysts, named gemmules [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. These gemmules contain undifferentiated dormant stem cells (named thesocytes) and provide protection against adverse conditions during winter or exposure to air. These gemmules can be hatched and cultivated under diverse \u003cem\u003ein vitro\u003c/em\u003e conditions (allowing a detailed description of their first developmental stage). Hence, gemmule-producing sponges constitute models of interest for recent imaging and molecular techniques [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan additionalcitationids=\"CR35\" citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. However, the microbiome composition of \u003cem\u003eS. lacustris\u003c/em\u003e has been underexplored to date, in regards to their wide distribution area. Two studies suggested the dominance of Alphaproteobacteria, Actinobacteria, Bacteroidota, Gammaproteobacteria, and Betaproteobacteria [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e], but the mechanisms underlying the early assembly of the bacterial consortium within these freshwater sponges are yet to be explored.\u003c/p\u003e \u003cp\u003eOur study aims to decipher the transmission modes of bacterial communities during the first steps of \u003cem\u003eS. lacustris\u003c/em\u003e asexual cycle. We hypothesize that the mixed mode transmission occurs, with both VT and HT. First, the VT would be achieved through the transmission of maternal bacteria within the gemmule, but also on its surface, as suggested by Sugden \u003cem\u003eet al.\u003c/em\u003e [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Once the gemmule has hatched and the filtration system is formed, the HT of planktonic bacteria could then constitute an additional factor shaping the \u003cem\u003eS. lacustris\u003c/em\u003e microbiome. To test these hypotheses, an experimental design is developed to determine the respective importance over time of (i) the microbiome transmitted on the gemmule\u0026rsquo;s surface, and (ii) the ambient free-living bacteria filtered by the juveniles.\u003c/p\u003e"},{"header":"Material and methods","content":"\u003cp\u003eThe overall study design was based on the cultivation of the \u003cem\u003eS. lacustris\u003c/em\u003e juveniles under two crossed factors (resulting in 4 treatments, Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The first factor was related to the absence/presence of epibiotic microbiome (\u0026plusmn; \u0026ldquo;EM\u0026rdquo;) on the gemmule, depending on the surface sterilization. In parallel, the second factor was related to the absence/presence of free-living bacteria (\u0026plusmn; \u0026ldquo;FB\u0026rdquo;) in the cultivation medium, depending on the pore size of the filters used for the freshwater filtration.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eSampling of the\u003c/b\u003e \u003cb\u003ein situ\u003c/b\u003e \u003cb\u003eadult sponges\u003c/b\u003e\u003c/p\u003e \u003cp\u003eFour replicates of adult \u003cem\u003eS. lacustris\u003c/em\u003e specimens were collected \u003cem\u003ein situ\u003c/em\u003e from a wooden jetty located at an inlet of the river Rhine at Oegstgeest (the Netherlands, 52\u0026deg;10'24.5\"N, 4\u0026deg;27'01.7\"E, ~\u0026thinsp;0.5m depth), at three different sampling times a month apart in 2021: M1 (October 15th ), M2 (November 10th ) and M3 (December 15th ). Sponge samples were rinsed with surrounding freshwater filtered upon 0.2\u0026micro;m pore size filters (FF0.2), and cut with sterilized tweezers and scalpels. The identification of \u003cem\u003eS. lacustris\u003c/em\u003e was based on skeletal and gemmoscleres examination. A voucher specimen is deposited at the sponge collection of Naturalis Biodiversity Center (RMNH.POR.12472). Samples for DNA metabarcoding were preserved in sterilized plastic vials filled with 96% ethanol and conserved at -20\u0026deg;C until DNA extraction. After M3, only encrusting parts of dead sponges were left: the sponge skeletons were observed without tissues, but full of gemmules. One of these \u0026ldquo;encrusting gemmules patches\u0026rdquo; (resulting from specimens collected at M3) was collected at t\u003csub\u003e0\u003c/sub\u003e (February 17th) in sterilized plastic vials filled with surrounding FF0.2. These samples collected for the experiment were kept at 4\u0026deg;C until the washing steps of the gemmules.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eSampling and washing of the gemmules\u003c/h2\u003e \u003cp\u003eOne day after the sampling, gemmules from the encrusting patch were separated from the maternal sponge skeleton according to [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e], using sterilized tweezers and teasing needles. Approximately 450 gemmules were evenly distributed in two distinct 15mL falcon tubes. Gemmules from the first tube were washed according to [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e], while gemmules from the second tube were washed with the same protocol but with Strekal\u0026rsquo;s medium instead of the 1% hydrogen peroxide (H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e) solution (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). For this treatment without H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e, particular cautions were taken (using a stereomicroscope) to sort and remove the dead and damaged gemmules, as well as important debris of the sponge maternal skeleton attached to the gemmule surface. As the H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e solution induces the sterilization of the gemmule surfaces [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e], these two washing conditions allowed to separate the gemmules in two groups: (i) the -EM for the sterilized gemmules without any potential epibiotic microbiome, and (ii) the +\u0026thinsp;EM for the unsterilized gemmules with their potential epibiotic microbiome still attached the surface (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Gemmules were stored at 4\u0026deg;C in Strekal\u0026rsquo;s medium for one hour before the plating. The efficiency of the sterilization through the use of H\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e2\u003c/sub\u003e was investigated as described in Supplementary Information (SI).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003ePlating and cultivation of the juveniles\u003c/h2\u003e \u003cp\u003eFor the t\u003csub\u003e0\u003c/sub\u003e samples (gemmules collected before the plating), 5 replicates for both +\u0026thinsp;EM and -EM treatments were collected from the 15mL falcon tubes and stored in 1.5mL eppendorf tubes filled with 96% EtOH at -20\u0026deg;C until DNA extraction. A total of 120 juvenile samples were expected for the experiment, based on the calculation of 6 sampling times (t\u003csub\u003e1\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e), 2 conditions of gemmule surface (\u0026plusmn;\u0026thinsp;EM), 2 conditions of filtered freshwater (\u0026plusmn;\u0026thinsp;FB), and 5 replicates.\u003c/p\u003e \u003cp\u003eUnder sterile conditions, the +\u0026thinsp;EM and -EM gemmules were distributed (on ice) in 120 wells of 24-well plates (approximately 3 to 4 gemmules per well), as described in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. The Strekal\u0026rsquo;s medium was then removed and replaced by 1 mL of filtered freshwater at room temperature. Plates were filled with freshwater filtered upon 0.45\u0026micro;m size filters (FF0.45) or FF0.2, for +\u0026thinsp;FB and -FB respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The filtered freshwater used was collected 1 hour before the filtration, on the same sampling site as the \u003cem\u003ein situ\u003c/em\u003e sponges. The plates were then transferred in an incubation chamber (Fitotron\u0026reg;, Weiss Technik, GmbH, Germany) at 20\u0026deg;C, with 70% of humidity and a day/night cycle of 14:10h (960 lux for the day light, SD: \u0026plusmn;65).\u003c/p\u003e \u003cp\u003eDetails of the sampling timeline of the experiment are described in \u003cb\u003enote S1\u003c/b\u003e (Supplementary Information, SI). For each treatment of each sampling time (t\u003csub\u003e1\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e,), the hatching rates obtained (80.2% in average, SD: \u0026plusmn; 21.2%) allowed us to collect at least one hatched juvenile for each replicates. The unhatched gemmules were discarded, resulting in approximately 1 to 3 juveniles gathered per replicate. Briefly, samples from t\u003csub\u003e1\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e were collected at 3, 7, 11, 17, 24, and 31 days after the plating (t\u003csub\u003e0\u003c/sub\u003e), respectively. The experiment was designed up to 31 days to follow potential changes occurring after the formation of the osculum and the development of a mature canal system. For each sampling time, the replicates of juvenile sponges were collected in 1.5mL eppendorf tubes filled with 96% EtOH at -20\u0026deg;C until DNA extraction.\u003c/p\u003e \u003cp\u003eThe FF0.45 and FF0.2 were refreshed at room temperature every 3 or 4 days with freshly collected freshwater from the same sampling site described above (see details in SI, \u003cb\u003enote S1\u003c/b\u003e). This duration between two refreshments was slightly longer compared to those recommended in [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e] (every 2 days). This choice was considered to have a longer exposure time enabling potential recognition mechanisms allowing the horizontal acquisition of planktonic bacteria.\u003c/p\u003e \u003cp\u003eTo investigate the structure of the free-living bacterial community amended at each refreshment for the +\u0026thinsp;FB condition, the FF0.45 was also successively filtered upon 0.2\u0026micro;m pore size PES filters used for DNA extraction. These samples named \u0026ldquo;FF0.45 samples\u0026rdquo; were collected for refreshment occurring before each time (except t\u003csub\u003e1\u003c/sub\u003e), and preserved in the CTAB extraction buffer at -20\u0026deg;C until extraction.\u003c/p\u003e \u003cp\u003eBefore each sampling time from t\u003csub\u003e1\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e, the juveniles sponges from each treatment (cross-conditions of \u0026plusmn;\u0026thinsp;EM with \u0026plusmn;\u0026thinsp;FB) were photographed using a stereomicroscope (SteREO Discovery.V20, ZEISS, Germany) mounted with a camera (AxioCam MRc5, ZEISS, Germany).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eDNA extractions, library preparation and high throughput sequencing of 16S rRNA gene amplicons\u003c/h2\u003e \u003cp\u003eDNA from \u003cem\u003ein situ\u003c/em\u003e adults (M1, M2, and M3), gemmules (t\u003csub\u003e0\u003c/sub\u003e), and \u003cem\u003ein vitro\u003c/em\u003e juveniles (t\u003csub\u003e1\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e) was extracted using the FastDNA\u0026trade; SPIN Kit for Soil (MP Biomedicals, Inc.) following the manufacturer\u0026rsquo;s instructions. \u003cem\u003eIn situ\u003c/em\u003e adult sponge samples were cut into small pieces of approximately 3*1*0.5mm using sterilized tweezers and scalpel blades. Special care was taken to collect only sponge tissues, avoiding gemmules. For t\u003csub\u003e0\u003c/sub\u003e samples, the gemmule coating was broken (by crushing them using the top of 10\u0026micro;L filter pipette tips), to facilitate the extraction of the DNA from the inside of the gemmule (see \u003cb\u003enote S2\u003c/b\u003e in SI). The CTAB DNA extraction of the FF0.45 samples was performed as described in [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe library preparation was conducted through a two-step PCR protocol for all samples together with the extraction blank and two negative controls (mQ water instead of template DNA). For the first PCR, the V3-V4 regions of the 16S rRNA gene were targeted and amplified with the PCR primers 341F 5\u0026rsquo;-CCTACGGGNGGCWGCAG-3\u0026rsquo; and 785R 5\u0026rsquo;-GACTACHVGGGTATCTAATCC-3\u0026rsquo; [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e] and the KAPA HiFi HotStart Ready Mix PCR Kit (Roche Molecular Systems, Inc.). Reactions were performed in a T100 Thermal Cycler (Bio-Rad, Hercules, CA, United States). The following thermal cycling scheme was conducted: initial denaturation at 95\u0026deg;C for 3 min, 30 cycles of denaturation at 98\u0026deg;C for 20 s, annealing at 57\u0026deg;C for 30 s, followed by extension at 72\u0026deg;C for 30 s. The final extension was carried out at 72\u0026deg;C for 1 min.\u003c/p\u003e \u003cp\u003ePCR products from the samples were checked using an E-Gel\u0026trade; (agarose gels at 2%), and the absence of amplification was validated for the negative controls and two extraction blanks. In several gemmule and juvenile samples, PCR amplifications were unsuccessful due to low DNA concentrations. At least 4 replicates was successful for every treatment at each time, except for\u0026thinsp;+\u0026thinsp;EM\u0026thinsp;+\u0026thinsp;FB at t\u003csub\u003e5\u003c/sub\u003e, with 3 successful replicates (\u003cb\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). PCR products were then cleaned with NucleoMag NGS-Beads (bead volume at 0.9 times the total volume of the sample, Macherey Nagel, D\u0026uuml;ren, Germany) using the VP 407AM-N 96 Pin Magnetic Bead Extractor stamp (V\u0026amp;P Scientific, San Diego, CA, United States). Through a second PCR, 3 \u0026micro;L of the cleaned PCR products were then amplified and labeled using the MiSeq Nextera XTDNA library preparation kit (Illumina, San Diego, CA, United States), with the same thermal cycling scheme limited to 8 cycles. PCR products were then analyzed with the Fragment Analyzer Agilent 5300 using the DNF-910-33 dsDNA Reagent Kit (35\u0026ndash;1,500 bp) protocol (Agilent Technologies, Santa Clara, CA, United States) to confirm the successful labeling of the DNA fragments. Negative controls and extraction blanks remained negative after this step. The pooling at the equimolar concentration was performed with QIAgility (Qiagen, Hilden, Germany). The final pool was then cleaned with NucleoMag NGSBeads, eluted in Milli-Q, and the DNA concentration was verified using Tapestation 4150 (Kit HSD 5000, Agilent Technologies, Santa Clara, CA, United States). The sequencing was performed on an Illumina MiSeq V3 PE300 platform at BaseClear B.V. (Leiden, Netherlands).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e16S rRNA gene metabarcoding data processing\u003c/h2\u003e \u003cp\u003eThe raw reads were first treated by BaseClear B.V. for demultiplexing (using bcl2fastq version 2.20, Illumina), and filtering based on two quality controls (using Illumina Chastity filtering, and a PhiX control signal filtering). The following reads were then processed with the DADA2 workflow allowing an inference to Amplicon Sequence Variant (ASV) [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e], using the \u0026ldquo;dada2\u0026rdquo; R package following the workflow described in [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e] and guidelines described in the online tutorial (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://benjjneb.github.io/dada2/tutorial.html\u003c/span\u003e\u003cspan address=\"https://benjjneb.github.io/dada2/tutorial.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The parameters used for filtering and trimming reads were as follows: truncation length of 270 and 240 base pairs for forward and reverse reads, respectively, maxN\u0026thinsp;=\u0026thinsp;0, maxEE\u0026thinsp;=\u0026thinsp;2, and truncQ\u0026thinsp;=\u0026thinsp;2. After the construction of the ASV table, chimeric sequences were filtered and the taxonomic assignment was performed using the Silva v138 reference database [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe ASV and taxonomy tables produced by the pipeline were then combined into a phyloseq object, together with the sample metadata table, using the \u0026ldquo;phyloseq\u0026rdquo; R package [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. The dataset was then filtered by removing all sequences from Eukaryota, chloroplast and mitochondria (representing on average\u0026thinsp;\u0026gt;\u0026thinsp;0.01% [SD: +/-0.002], 36.9% [SD: +/-12.8] and 0.9% [SD: +/-1.9] of all reads per sample, respectively). Data was then decontaminated with the negative controls and the two extraction blanks used as control samples, using the \u0026ldquo;decontam\u0026rdquo; R package [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eMetabarcoding and statistical analyses\u003c/h2\u003e \u003cp\u003eThe \u003cem\u003eα\u003c/em\u003e-diversity measures were estimated with Chao1 (estimated richness), Pielou (evenness), and Shannon (both richness and evenness) indices using the \u0026ldquo;phyloseq\u0026rdquo; and \u0026ldquo;vegan\u0026rdquo; R packages [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e] and the rarefied datasets (rarefaction performed to the minimum library size, i.e. 5332 reads). According to Shapiro tests, the diversity metrics were significantly different from the normal distribution. Consequently, differences between sample groups within these metrics were investigated through non-parametric tests (Kruskal-Wallis followed by pairwise Wilcoxon tests) using the \u0026ldquo;agricolae\u0026rdquo; R package [\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. Following recommendations for compositional approaches from [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e], all other analyses were conducted without rarefaction, using the \u0026ldquo;phyloseq\u0026rdquo; R package and the datasets normalized to the total number of sequences per sample. The \u003cem\u003eβ\u003c/em\u003e-diversity was analyzed with non-metric multidimensional scaling (NMDS) using Bray-Curtis dissimilarity. Differences of \u003cem\u003eβ\u003c/em\u003e-diversity between groups were statistically checked with one-way PERMANOVA tests followed by pairwise Adonis tests, using the \u0026ldquo;vegan\u0026rdquo; R package. Differences between the dispersion of the \u003cem\u003eβ\u003c/em\u003e-diversity within each treatment were calculated through the PERMDISP2 analysis using the \u003cem\u003ebetadisper\u003c/em\u003e() function from the \u0026ldquo;vegan\u0026rsquo; R package, and tested using an ANOVA followed by a Tukey\u0026rsquo;s HSD pairwise test.\u003c/p\u003e \u003cp\u003eFocusing on t\u003csub\u003e0\u003c/sub\u003e and t\u003csub\u003e1\u003c/sub\u003e samples, phylogenetical heat trees were performed to identify the significant pioneer taxa differentially abundant between the different treatment conditions (\u0026plusmn;\u0026thinsp;EM and \u0026plusmn;\u0026thinsp;FB), using the \u003cem\u003emetacoder\u003c/em\u003e package [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. The differential analyses were performed with a subset of the compositional dataset excluding rare ASV (relative abundance\u0026thinsp;\u0026lt;\u0026thinsp;0.02%).\u003c/p\u003e \u003cp\u003eAnalyses of the ASVs shared between conditions were performed with the \u003cem\u003ecore\u003c/em\u003e() function using the \u0026ldquo;microbiome\u0026rdquo; R package [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e]. An ASV from juveniles from one specific treatment at a specific sampling time was considered shared with the \u003cem\u003ein situ\u003c/em\u003e adult sponges, when the ASV is present at least in 3 replicates of each of the two groups. Similarly, an ASV from juveniles from one specific treatment at a specific sampling time was considered shared with the filtered freshwater (FF0.45), when the ASV is present at least in 3 samples of each of the two groups. Using the \u0026ldquo;eulerr\u0026rdquo; R package [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e], the Venn diagrams were plotted to represent the number of shared ASVs between the juveniles, the \u003cem\u003ein situ\u003c/em\u003e adult sponges and the prefiltered freshwater with the \u003cem\u003evenn\u003c/em\u003e() function.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eDevelopment of the juveniles sponges\u003c/h2\u003e \u003cp\u003eThe hatching of the gemmules was observed 2 days after their plating (t\u003csub\u003e0\u003c/sub\u003e), whatever the treatment. No significant differences of hatching rates were observed between the gemmules from each treatment (ANOVA test: \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.07). These hatching rates reached in average 76.3% (SD: \u0026plusmn;20.7) for\u0026thinsp;+\u0026thinsp;EM\u0026thinsp;+\u0026thinsp;FB, 81.9% (SD: \u0026plusmn;22.6) for +\u0026thinsp;EM-FB, 75.3% (SD: \u0026plusmn;21.4) for -EM\u0026thinsp;+\u0026thinsp;FB and 87.2% (SD: \u0026plusmn;20.0) for -EM-FB. The same developmental stages were observed between treatments for each sampling time (\u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). The first sampling time (t\u003csub\u003e1\u003c/sub\u003e) was characterized with the development of the sponge tissues on the gemmule coat (i.e. the gemmule husk) and colonizing the substrate. The early formation of a canal system and an osculum was observed for t\u003csub\u003e2\u003c/sub\u003e samples. At t\u003csub\u003e3\u003c/sub\u003e, the development of a mature canal system was noticed, together with a light green coloration. The coloration was more saturated at t\u003csub\u003e4\u003c/sub\u003e and thereafter (\u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/b\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eAlpha-diversity analyses\u003c/h2\u003e \u003cp\u003eSignificant differences between the different sample types were observed only for the Pielou index (\u003cb\u003eFigure S2\u003c/b\u003e, \u003cb\u003eTable S2\u003c/b\u003e), with lower values within \u003cem\u003ein situ\u003c/em\u003e adult sponges compared to gemmules and juvenile sponges (\u003cb\u003eFigure S2\u003c/b\u003e, Wilcoxon test). For gemmules and juveniles, significant differences between treatments were observed according to the sampling time together with the \u0026plusmn;\u0026thinsp;EM (with epibiotic microbiome) factor, for the Shannon (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, \u003cb\u003eTable S3)\u003c/b\u003e, Chao1 and Pielou indices (\u003cb\u003eFigures S3A\u003c/b\u003e and \u003cb\u003eS3B\u003c/b\u003e, respectively, \u003cb\u003eTable S3\u003c/b\u003e). No significant differences were observed according to the \u0026plusmn;\u0026thinsp;FB factor (Kruskal-Wallis test: \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05 for the three indices). For the Shannon index, a decrease from t\u003csub\u003e0\u003c/sub\u003e to t\u003csub\u003e2\u003c/sub\u003e was observed for the -EM (without epibiotic microbiome) samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). At t\u003csub\u003e1\u003c/sub\u003e, t\u003csub\u003e2\u003c/sub\u003e, and t\u003csub\u003e3\u003c/sub\u003e, Shannon values in -EM samples were significantly lower compared to those in +\u0026thinsp;EM (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Wilcoxon test). The same results were observed for the two other indices (Pielou and Chao1), except at t\u003csub\u003e2\u003c/sub\u003e for the Chao1 (\u003cb\u003eFigure S3\u003c/b\u003e, Wilcoxon test). For all indices, a general tendency of increasing \u003cem\u003eα\u003c/em\u003e-diversity was observed after t\u003csub\u003e3\u003c/sub\u003e up to t\u003csub\u003e6\u003c/sub\u003e, for the juveniles in -EM samples, while the values for the +\u0026thinsp;EM samples stayed stable all along the experiment.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eBeta-diversity analyses\u003c/h2\u003e \u003cp\u003eThe NMDS plot and PERMANOVA test, conducted with all samples, showed significant differences between the \u003cem\u003eβ\u003c/em\u003e-diversity of each sample type (freshwater, \u003cem\u003ein situ\u003c/em\u003e adult tissues, gemmules and \u003cem\u003ein vitro\u003c/em\u003e juveniles, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, \u003cb\u003eTable S4\u003c/b\u003e). Additionally, significant temporal changes were observed on the NMDS along (i) the first axis from M3 to t\u003csub\u003e0\u003c/sub\u003e (\u003cem\u003ein situ\u003c/em\u003e adult sponges and gemmules), and (ii) the second axes from t\u003csub\u003e1\u003c/sub\u003e up to t\u003csub\u003e6\u003c/sub\u003e (\u003cem\u003ein vitro\u003c/em\u003e juveniles) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, \u003cb\u003eTable\u0026nbsp;5A\u003c/b\u003e). Within the \u003cem\u003ein vitro\u003c/em\u003e juveniles, the main temporal shifts appear between (i) t\u003csub\u003e1\u003c/sub\u003e and t\u003csub\u003e2\u003c/sub\u003e, and (ii) t\u003csub\u003e4\u003c/sub\u003e and t\u003csub\u003e5\u003c/sub\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). Pairwise comparison tests confirmed the significant differences between all sampling times, except between (i) t\u003csub\u003e2\u003c/sub\u003e and t\u003csub\u003e3\u003c/sub\u003e, (ii) t\u003csub\u003e3\u003c/sub\u003e and t\u003csub\u003e4\u003c/sub\u003e, and (iii) t\u003csub\u003e5\u003c/sub\u003e and t\u003csub\u003e6\u003c/sub\u003e (\u003cb\u003eTable S5B\u003c/b\u003e). For \u003cem\u003ein vitro\u003c/em\u003e juvenile samples, the three-way PERMANOVA (\u003cb\u003eTable S5A\u003c/b\u003e) also indicated significant differences between sampling times, in combination with the two treatment factors (\u0026plusmn;\u0026thinsp;EM and \u0026plusmn;\u0026thinsp;FB). Pairwise comparison tests conducted with all juvenile samples, showed an overall significant difference between the \u003cem\u003eβ\u003c/em\u003e-diversity the 4 treatments, except between\u0026thinsp;+\u0026thinsp;EM\u0026thinsp;+\u0026thinsp;FB and +\u0026thinsp;EM-FB (\u003cb\u003eTable S5C\u003c/b\u003e). The NMDS of all samples plotted with a color code related to the treatments (\u003cb\u003eFigure S4\u003c/b\u003e) revealed that +\u0026thinsp;EM samples from t\u003csub\u003e1\u003c/sub\u003e to t\u003csub\u003e4\u003c/sub\u003e are grouped, while differences between +\u0026thinsp;FB and -FB can be distinguished for the -EM samples. Additionally, differences between +\u0026thinsp;EM and -EM can also be observed with t\u003csub\u003e5\u003c/sub\u003e and t\u003csub\u003e6\u003c/sub\u003e samples (\u003cb\u003eFigure S4\u003c/b\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe NMDS (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB\u003cb\u003e)\u003c/b\u003e and two-way PERMANOVA (\u003cb\u003eTable S6A\u003c/b\u003e) analyses, performed at each sampling time separately, revealed significant differences of \u003cem\u003eβ\u003c/em\u003e-diversity between treatments. At t\u003csub\u003e0\u003c/sub\u003e (for the gemmules), significant differences between +\u0026thinsp;EM and -EM samples were observed on the first NMDS axis (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB, \u003cb\u003eTables S6A and S6B\u003c/b\u003e). For the NMDS plots from t\u003csub\u003e1\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e, significant differences can also be observed between juveniles\u0026thinsp;+\u0026thinsp;EM and -EM (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB, \u003cb\u003eTables S6A and S6B\u003c/b\u003e). From t\u003csub\u003e1\u003c/sub\u003e to t\u003csub\u003e3\u003c/sub\u003e, a similar pattern can be observed with three clusters differentiating (i)\u0026thinsp;+\u0026thinsp;EM samples (including both +\u0026thinsp;and -FB), (ii) -EM\u0026thinsp;+\u0026thinsp;FB samples, (iii) and -EM-FB samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). More precisely, the differences between +\u0026thinsp;FB and -FB appear more important within the -EM group than the +\u0026thinsp;EM group. From t\u003csub\u003e4\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e, a larger separation between the +\u0026thinsp;FB and -FB clusters was observed for the +\u0026thinsp;EM samples. Finally at t\u003csub\u003e6\u003c/sub\u003e, 4 distinct clusters can be observed for each of the 4 treatments from the two cross-conditions: the first NMDS axis is involved in the \u0026plusmn;\u0026thinsp;EM differences, while the second axis seems to explain the \u0026plusmn;\u0026thinsp;FB differences. However, according to the pairwise comparisons, these \u0026plusmn;\u0026thinsp;FB differences at t\u003csub\u003e6\u003c/sub\u003e were not significant, in both +\u0026thinsp;EM and -EM conditions (\u003cb\u003eTable S6B\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eThe \u003cem\u003eβ\u003c/em\u003e-diversity dispersion within each treatment (\u0026plusmn;\u0026thinsp;EM and \u0026plusmn;\u0026thinsp;FB) was determined at each sampling time of the \u003cem\u003ein vitro\u003c/em\u003e juveniles (\u003cb\u003eFigure S5\u003c/b\u003e). A global tendency of increasing dispersion is observed with a higher dispersion over time for the +\u0026thinsp;EM samples, confirming the previous observations on the NMDS plots (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). For -EM juvenile sponges, the highest dispersion values were found at t\u003csub\u003e1\u003c/sub\u003e, followed by a decrease at t\u003csub\u003e2\u003c/sub\u003e (\u003cb\u003eFigure S5\u003c/b\u003e). At t\u003csub\u003e1\u003c/sub\u003e, a significantly higher dispersion was observed in -EM compared to +\u0026thinsp;EM, whatever the presence or not of FB (\u003cb\u003eFigure S5\u003c/b\u003e, \u003cb\u003eTable S7\u003c/b\u003e). At t\u003csub\u003e2\u003c/sub\u003e, a similar observation was made, with significantly higher dispersion in -EM compared to +\u0026thinsp;EM, only in the +\u0026thinsp;FB condition (\u003cb\u003eFigure S5\u003c/b\u003e, \u003cb\u003eTable S7\u003c/b\u003e). For t\u003csub\u003e3\u003c/sub\u003e, t\u003csub\u003e4\u003c/sub\u003e, and t\u003csub\u003e6\u003c/sub\u003e, no significant differences of \u003cem\u003eβ\u003c/em\u003e-diversity dispersion were observed between treatments (\u003cb\u003eFigure S5\u003c/b\u003e). At t\u003csub\u003e5\u003c/sub\u003e, significant differences were observed with higher dispersion in\u0026thinsp;+\u0026thinsp;EM\u0026thinsp;+\u0026thinsp;FB juveniles compared to all -FB juveniles (\u003cb\u003eFigure S5\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eCompositional and differential analyses\u003c/h2\u003e \u003cp\u003eThe composition of the bacterial community at the family level showed overall differences between each sample type (filtered freshwater, \u003cem\u003ein situ\u003c/em\u003e adult sponge, the gemmules, and the \u003cem\u003ein vitro\u003c/em\u003e juveniles, \u003cb\u003eFigure S6\u003c/b\u003e). For the filtered freshwater, the main free-living bacterial communities were dominated by \u003cem\u003eSporichthyaceae\u003c/em\u003e and \u003cem\u003eMicrobacteriaceae\u003c/em\u003e (Actinobacteria), and \u003cem\u003eBurkholderiaceae\u003c/em\u003e and \u003cem\u003eMethylophilaceae\u003c/em\u003e (Gammaproteobacteria) (\u003cb\u003eFigure S6A\u003c/b\u003e). For the \u003cem\u003ein situ\u003c/em\u003e adult sponges, the composition was dominated by \u003cem\u003eSporichthyaceae\u003c/em\u003e (Actinobacteria), \u003cem\u003eChitinophagaceae\u003c/em\u003e and \u003cem\u003eFlavobacteriaceae\u003c/em\u003e (Bacteroidia), \u003cem\u003eElsteraceae\u003c/em\u003e and \u003cem\u003eTerasakiellaceae\u003c/em\u003e (Alphaproteobacteria) and \u003cem\u003eComamonadaceae\u003c/em\u003e (Gammaproteobacteria) (\u003cb\u003eFigure S6A\u003c/b\u003e). Differences in composition for these adult sponge tissues were observed over time from M1 (October) to M3 (December), with an increase of \u003cem\u003eFlavobacteriaceae\u003c/em\u003e and \u003cem\u003eElsteraceae\u003c/em\u003e and a decrease of \u003cem\u003eSporichtyaceae\u003c/em\u003e. Additionally, higher relative abundances of \u003cem\u003eTerasakiellaceae\u003c/em\u003e can be observed specifically at M2 compared to M1 and M3.\u003c/p\u003e \u003cp\u003eFor the gemmules (t\u003csub\u003e0\u003c/sub\u003e), the bacterial composition of the -EM was dominated by \u003cem\u003eTerasakiellaceae\u003c/em\u003e, while the +\u0026thinsp;EM was dominated by the \u003cem\u003eRhodobacteraceae\u003c/em\u003e and \u003cem\u003eComamonadaceae\u003c/em\u003e (\u003cb\u003eFigure S6A\u003c/b\u003e). For the \u003cem\u003ein vitro\u003c/em\u003e juveniles (\u003cb\u003eFigure S6B\u003c/b\u003e), important changes can be observed, with a dominance of \u003cem\u003eFlavobacteriaceae\u003c/em\u003e with exclusively ASVs from \u003cem\u003eFlavobacterium\u003c/em\u003e (Bacteroidia;) at t\u003csub\u003e1\u003c/sub\u003e. From t\u003csub\u003e2\u003c/sub\u003e to t\u003csub\u003e4\u003c/sub\u003e, \u003cem\u003eComamonadaceae\u003c/em\u003e, \u003cem\u003eAlteromonadaceae\u003c/em\u003e, and \u003cem\u003ePseudomonadaceae\u003c/em\u003e (Gammaproteobacteria) are the dominant families. Alphaproteobacteria was also found as one of the most abundant classes (\u003cem\u003eRhodobacteraceae\u003c/em\u003e, \u003cem\u003eSphingomonadaceae\u003c/em\u003e, \u003cem\u003eRhizobiaceae\u003c/em\u003e), without clear changes observed over time. Additionally, an increase of \u003cem\u003ePirellulaceae\u003c/em\u003e can be noticed all along the experiment up to t\u003csub\u003e6\u003c/sub\u003e.\u003c/p\u003e \u003cp\u003eDifferences in relative abundances of families between treatments were observed, for example at t\u003csub\u003e1\u003c/sub\u003e, with higher percentages of \u003cem\u003eComamonadaceae\u003c/em\u003e in +\u0026thinsp;EM compared to -EM. For +\u0026thinsp;EM at t\u003csub\u003e1\u003c/sub\u003e, higher abundances of \u003cem\u003eOxalobacteraceae\u003c/em\u003e were observed in +\u0026thinsp;FB compared to -FB samples, while the opposite is noticed for \u003cem\u003eAlteromonadaceae\u003c/em\u003e. At t\u003csub\u003e2\u003c/sub\u003e higher relative abundances of \u003cem\u003ePseudomonadaceae\u003c/em\u003e were found in -EM samples compared to +\u0026thinsp;EM samples, while the opposite was observed for the \u003cem\u003eComamonadaceae\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eThe differential analysis with t\u003csub\u003e0\u003c/sub\u003e samples (gemmules before hatching) was performed to identify significant taxa differentially abundant in +\u0026thinsp;EM gemmules compared to -EM gemmules, and conversely (\u003cb\u003eFigure S7\u003c/b\u003e). The analysis confirmed observations from the barplots for the most abundant families (\u003cb\u003eFigure S6A\u003c/b\u003e) with a higher abundance of \u003cem\u003eTerasakiellaceae\u003c/em\u003e in the -EM, and a higher abundance of \u003cem\u003eRhodobacteraceae\u003c/em\u003e and \u003cem\u003eComamonadaceae\u003c/em\u003e\u0026thinsp;+\u0026thinsp;EM. These two last families were found mainly represented by the genus \u003cem\u003ePseudorhodobacter\u003c/em\u003e and \u003cem\u003eHydrogenophaga\u003c/em\u003e, respectively, being also found differentially more abundant in +\u0026thinsp;EM samples. Other taxa with lower percentages were also found to be differentially abundant, such as \u003cem\u003eBurkholderiaceae\u003c/em\u003e (genus \u003cem\u003eRalstonia\u003c/em\u003e) and \u003cem\u003eHyphomonadaceae\u003c/em\u003e (genus \u003cem\u003eHirschia\u003c/em\u003e) in +\u0026thinsp;EM and \u003cem\u003eChitinophagaceae\u003c/em\u003e (genus \u003cem\u003eFerruginibacter\u003c/em\u003e) and \u003cem\u003ePseudomonadaceae\u003c/em\u003e (genus \u003cem\u003ePseudomonas\u003c/em\u003e) in -EM.\u003c/p\u003e \u003cp\u003eIn addition to the t\u003csub\u003e0\u003c/sub\u003e samples, another focus was made at t\u003csub\u003e1\u003c/sub\u003e for differential analyses performed between treatment of the \u003cem\u003ein vitro\u003c/em\u003e juveniles. The differential analysis between +\u0026thinsp;EM and -EM was conducted for both +\u0026thinsp;FB and -FB samples, separately. (i) For +\u0026thinsp;FB samples, (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA) significantly higher abundances of Caulobacterales and Sphingomonadales (Alphaproteobacteria), Cytophagales (Bacteroidia), Burkholderiales, Xanthomonadales and Enterobacterales (Gammaproteobacteria) were observed in +\u0026thinsp;EM compared to -EM. Conversely, a significantly higher abundance of \u003cem\u003eTerasakiellaceae\u003c/em\u003e (Alphaproteobacteria) and \u003cem\u003eMoraxellaceae\u003c/em\u003e (Gammaproteobacteria) was observed in -EM compared to +\u0026thinsp;EM. (ii) For -FB samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB), significantly higher abundance of all Bacteroidia taxa (including Cytophagales), \u003cem\u003eRhodobacteraceae\u003c/em\u003e (Alphaproteobacteria), Enterobacterales and \u003cem\u003eComamonadaceae\u003c/em\u003e (Gammaproteobacteria) were observed in +\u0026thinsp;EM compared to -EM, while higher abundance of \u003cem\u003eOxalobacteraceae\u003c/em\u003e, \u003cem\u003eBurkholderiaceae\u003c/em\u003e and \u003cem\u003eCaulobacteraceae\u003c/em\u003e and Rhizobiales were observed in -EM. The differential analysis between +\u0026thinsp;FB and -FB was also conducted for both +\u0026thinsp;EM and -EM samples, separately. (i) For +\u0026thinsp;EM samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC), significant higher abundance of \u003cem\u003eOxalobacteraceae\u003c/em\u003e (genera \u003cem\u003eUndibacterium\u003c/em\u003e and \u003cem\u003eJanthinobacterium\u003c/em\u003e) and \u003cem\u003eShewanellaceae\u003c/em\u003e were observed for +\u0026thinsp;FB samples compared to -FB samples, while higher \u003cem\u003eMoraxellaceae\u003c/em\u003e, \u003cem\u003eCrocinitomicaceae\u003c/em\u003e, \u003cem\u003eAlteromonadaceae\u003c/em\u003e (\u003cem\u003eRheinheimera\u003c/em\u003e), \u003cem\u003eRhodocyclaceae\u003c/em\u003e where significantly more abundant in -FB samples compared to the +\u0026thinsp;FB ones. (ii) For -EM samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD), the differential analysis revealed significant higher abundances of \u003cem\u003eFlavobacteriaceae\u003c/em\u003e and Pseudomonadales (including \u003cem\u003eMoraxellaceae\u003c/em\u003e) in +\u0026thinsp;FB, while -FB samples were enriched with Caulobacterales, Rhizobiales, and \u003cem\u003eSphingomonadaceae\u003c/em\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eASVs from gemmules or juveniles shared with adult sponges or freshwater samples\u003c/h2\u003e \u003cp\u003eVenn diagrams (\u003cb\u003eFigure S8\u003c/b\u003e) were obtained to estimate the number of ASVs shared between three groups of samples: (i) the juveniles (from a specific treatment at a specific time), (ii) adult sponges and (iii) freshwater samples. No ASVs were found to be shared between the freshwater and the gemmules, or between the freshwater and the juveniles (\u003cb\u003eFigure S8\u003c/b\u003e), except (i) ASV73 (\u003cem\u003ePolynucleobacter\u003c/em\u003e) shared at t\u003csub\u003e1\u003c/sub\u003e and t\u003csub\u003e3\u003c/sub\u003e with the\u0026thinsp;+\u0026thinsp;EM\u0026thinsp;+\u0026thinsp;FB juveniles, and at t\u003csub\u003e4\u003c/sub\u003e with the -EM-FB juveniles, (ii) ASV80 (Candidatus \u003cem\u003eLimnoluna\u003c/em\u003e) shared at t\u003csub\u003e5\u003c/sub\u003e with the\u0026thinsp;+\u0026thinsp;EM\u0026thinsp;+\u0026thinsp;FB, -EM\u0026thinsp;+\u0026thinsp;FB and -EM-FB juveniles, and (iii) ASV734 (\u003cem\u003eBradyrhizobium\u003c/em\u003e) shared at t\u003csub\u003e6\u003c/sub\u003e with the -EM-FB juveniles. These three ASVs (73, 80, and 734) were found with an average relative abundance below 0.04%. Additionally, a total of 17 ASVs were found to be shared between the freshwater and the \u003cem\u003ein situ\u003c/em\u003e adult sponges (\u003cb\u003eFigure S8\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eBased on the Venn diagram results (\u003cb\u003eFigure S8\u003c/b\u003e), the numbers ASVs shared between \u003cem\u003ein vitro\u003c/em\u003e juveniles and \u003cem\u003ein situ\u003c/em\u003e adult sponges were summarized in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e. These numbers were higher for +\u0026thinsp;EM juveniles, compared to -EM juveniles. For example, at t\u003csub\u003e1\u003c/sub\u003e, 34 ASVs from the adult sponges were shared in\u0026thinsp;+\u0026thinsp;EM\u0026thinsp;+\u0026thinsp;FB, while only 9 were shared between the -EM\u0026thinsp;+\u0026thinsp;FB and the \u003cem\u003ein situ\u003c/em\u003e adult sponges (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e and S10). Similarly, 21 ASVs from \u003cem\u003ein situ\u003c/em\u003e adult sponges were shared with the +\u0026thinsp;EM-FB samples, while 12 ASVs were shared with the -EM-FB samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). These numbers were also found higher in +\u0026thinsp;EM compared to -EM for t\u003csub\u003e0\u003c/sub\u003e, t\u003csub\u003e2\u003c/sub\u003e, t\u003csub\u003e3\u003c/sub\u003e, and t\u003csub\u003e4\u003c/sub\u003e. Additionally, a decreasing number of shared ASVs in +\u0026thinsp;EM can be noticed over time (from 34 shared ASVs at t\u003csub\u003e1\u003c/sub\u003e, reaching 11 ASVs at t\u003csub\u003e6\u003c/sub\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn addition to numbers of ASVs shared between the juveniles and the \u003cem\u003ein situ\u003c/em\u003e adult sponges, the relative abundance of sequences from these ASVs and their taxonomy was investigated (\u003cb\u003eFigure S9\u003c/b\u003e). At t\u003csub\u003e0\u003c/sub\u003e, the dominant shared ASVs within -EM gemmules belong to the \u003cem\u003eTerasakiellaceae\u003c/em\u003e family (ASV6: unaffiliated genus, average relative abundance\u0026thinsp;\u0026gt;\u0026thinsp;20%), while their relative abundance was lower in +\u0026thinsp;EM samples (\u003cb\u003eFigure S9\u003c/b\u003e). At t\u003csub\u003e1\u003c/sub\u003e and t\u003csub\u003e2\u003c/sub\u003e, major shared ASVs (average relative abundance\u0026thinsp;\u0026gt;\u0026thinsp;3%) identified belong to (i) the \u003cem\u003eFlavobacteriaceae\u003c/em\u003e (ASV9: \u003cem\u003eFlavobacterium\u003c/em\u003e), (ii) the \u003cem\u003eRhodobacteraceae\u003c/em\u003e (ASV18 and ASV21: \u003cem\u003eDefluviimonas\u003c/em\u003e and \u003cem\u003eTabrizicola\u003c/em\u003e, respectively), and (iii) the \u003cem\u003eTerasakiellaceae\u003c/em\u003e families (ASV6: unaffiliated genus). From t\u003csub\u003e2\u003c/sub\u003e to t\u003csub\u003e4\u003c/sub\u003e, a higher relative percentage of shared ASVs from Gammaproteobacteria was observed, with the \u003cem\u003eAlteromonadaceae\u003c/em\u003e (ASV14: \u003cem\u003eRheinheimera\u003c/em\u003e) and the \u003cem\u003eComamonadaceae\u003c/em\u003e (ASV65: \u003cem\u003ePaucibacter\u003c/em\u003e). Finally, the \u003cem\u003eRhodobacteraceae\u003c/em\u003e (ASV6: \u003cem\u003eFlavobacterium\u003c/em\u003e) were found mainly dominant in the t\u003csub\u003e5\u003c/sub\u003e and t\u003csub\u003e6\u003c/sub\u003e samples (\u003cb\u003eFigure S9)\u003c/b\u003e. When considering all shared ASVs together, important variations of their relative abundances were observed between and within the treatments, especially at t\u003csub\u003e1\u003c/sub\u003e, t\u003csub\u003e3\u003c/sub\u003e, and t\u003csub\u003e4\u003c/sub\u003e within -EM\u0026thinsp;+\u0026thinsp;FB samples. However, despite these variations, a significant temporal decrease of percentages of shared ASVs was observed from t\u003csub\u003e0\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e (ANOVA test: \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001; \u003cb\u003eFigure S9\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur experiment was designed (i) to investigate the dynamics of bacterial communities during the first steps of the asexual cycle of \u003cem\u003eSpongilla lacustris\u003c/em\u003e, and (ii) to decipher the transmission modes involved in the microbiome assembly. Only few studies experimentally investigated bacterial HT and VT hypotheses during the first steps of the sponge ontogeny [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan additionalcitationids=\"CR55\" citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e], and to date, our study brings the first insights for freshwater sponges. Through their sampling accessibility, but also the easiness of hatching and culturing, freshwater sponges such as \u003cem\u003eS. lacustris\u003c/em\u003e or \u003cem\u003eEphydatia muelleri\u003c/em\u003e, are promising ressources to better understand such mechanisms [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Additionally, the gemmule surfaces sterilization protocol [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e], provides a good experimental condition to test the VT scenario on the gemmule surface.\u003c/p\u003e \u003cp\u003e \u003cb\u003eLinks between the ontogeny of\u003c/b\u003e \u003cb\u003eS. lacustris\u003c/b\u003e \u003cb\u003ejuveniles and the temporal dynamics of their bacterial communities\u003c/b\u003e\u003c/p\u003e \u003cp\u003eThe development of \u003cem\u003eS. lacustris\u003c/em\u003e juveniles after gemmule hatching observed during this experiment was similar to \u003cem\u003eE. muelleri\u003c/em\u003e [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e] and \u003cem\u003eEphydatia fluviatilis\u003c/em\u003e [\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e]. In these previous studies, five stages of juveniles development were described: (i) stage 1 (1\u0026ndash;2 days after plating, pre- or just-hatching stage): the first stem cells are migrating out of the gemmules through the micropyle; (ii) stage 2 (1\u0026ndash;3 days after hatching): the first tissues are growing around the gemmule husk, or on the substrate; (iii) stage 3 (2\u0026ndash;4 days after hatching): canal system and choanocytes are formed; (iv) stage 4 (3\u0026ndash;5 days after hatching): an osculum is starting to forms while the aquiferous system is still being organized; and (v) stage 5 (4\u0026ndash;7 days after hatching): complex branched canals are formed and the osculum is developed. Based on this description and our observation, the first sampling time of our experiment (t\u003csub\u003e1\u003c/sub\u003e: 3 days after plating and 1 day after hatching) corresponds to stage 2, the second (t\u003csub\u003e2\u003c/sub\u003e: 5 days after hatching) corresponds to stage 4, while following ones (from t\u003csub\u003e3\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e: 9 days after hatching, and beyond) correspond to stage 5. Additionally, another important change in the development of the \u003cem\u003eS. lacustris\u003c/em\u003e juveniles can be observed mainly after t\u003csub\u003e3\u003c/sub\u003e with the colonization of \u003cem\u003eChlorella\u003c/em\u003e-like symbionts explaining the slight green coloration observed around the choanocyte chambers. The \u003cem\u003eChlorella\u003c/em\u003e-like colonization was found to be more important from t\u003csub\u003e4\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e, with a large proportion of tissues showing a green coloration more saturated. These results suggest that the symbiosis with the \u003cem\u003eChlorella\u003c/em\u003e-like symbionts is acquired before t\u003csub\u003e3\u003c/sub\u003e but fully established in the whole juvenile body only after t\u003csub\u003e4\u003c/sub\u003e, under the light condition of our experiment (960 lux for day-light).\u003c/p\u003e \u003cp\u003eWhen looking at the dynamics of the bacterial diversity after t\u003csub\u003e0\u003c/sub\u003e and regardless of the treatment conditions, the main temporal changes can be observed in particular with a continuous shift from the \u003cem\u003eβ\u003c/em\u003e-diversity from t\u003csub\u003e1\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e. The specific diversity associated with t\u003csub\u003e1\u003c/sub\u003e samples could be linked to the early development stage of these juveniles since the osculum is not formed yet, while the first cells are emerging around the gemmule husk, and colonizing the substrate (stage 2). This \u003cem\u003eβ\u003c/em\u003e-diversity difference seems to be explained in terms of composition by a higher relative abundance of Bacteroidota, and more specifically the \u003cem\u003eFlavobacteriaceae\u003c/em\u003e (dominated by \u003cem\u003eFlavobacterium\u003c/em\u003e) observed at t\u003csub\u003e1\u003c/sub\u003e compared to the other following sampling times, but also compared to t\u003csub\u003e0\u003c/sub\u003e. These results suggest that a specific development of these pioneer bacterial taxa could be involved in these first steps following the hatching. Interestingly, diverse Bacteroidota strains (including one \u003cem\u003eFlavobacterium\u003c/em\u003e sp.) were found to promote the settlement of the larvae of the marine sponge \u003cem\u003eTedania\u003c/em\u003e sp. [\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e]. These strains were either found to form biofilm enhancing the settlement of the larvae, or to excrete chemical cues inducing the larvae settlement through direct secretion or through the production of extracellular vesicles. A similar scenario (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) could be considered for the settlement of the gemmules of \u003cem\u003eS. lacustris\u003c/em\u003e, with an enrichment of \u003cem\u003eFlavobacterium\u003c/em\u003e symbionts allowing the first sponge cells to better colonize their substrate (in our case: the gemmule husk and the flat bottom of the 24-well plate).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFrom t\u003csub\u003e2\u003c/sub\u003e to t\u003csub\u003e4\u003c/sub\u003e (stage 4 to 5), the formation of the osculum and the development of the aquiferous system are observed. The \u003cem\u003eβ\u003c/em\u003e-diversity differences with t\u003csub\u003e1\u003c/sub\u003e can be linked to the ability of the sponge to actively filter the environment. Similar observations were demonstrated for the marine sponge \u003cem\u003eCrambe crambe\u003c/em\u003e [\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e], as its prokaryotic community was found to change with regard to the osculum formation. In terms of composition, t\u003csub\u003e2\u003c/sub\u003e to t\u003csub\u003e4\u003c/sub\u003e samples are highlighted by a high abundance of Gammaproteobacteria, with three dominant families: \u003cem\u003eComamonadaceae\u003c/em\u003e, \u003cem\u003eAlteromonadaceae\u003c/em\u003e and \u003cem\u003ePseudomonadaceae\u003c/em\u003e. These families are commonly occurring in freshwater sponges [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e] and might be related to chemical defenses [\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e]. For example, \u003cem\u003eComamonadaceae\u003c/em\u003e were associated with defense mechanisms such as CRISPR and intracellular trafficking within \u003cem\u003eE. muelleri\u003c/em\u003e [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], while diverse PKS genes were found within \u003cem\u003ePseudomonadaceae\u003c/em\u003e and \u003cem\u003eAlteromonadaceae\u003c/em\u003e (dominant genus: \u003cem\u003eRheinheimera\u003c/em\u003e) associated with the freshwater sponges \u003cem\u003eRezinkovia echinata\u003c/em\u003e (Lake Baikal) and \u003cem\u003eE. fluviatilis\u003c/em\u003e (Vinkeveense Plassen, the Netherlands), respectively [\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e, \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e]. This specific bacterial community acquired during the formation of the aquiferous system and the first filter-feeding activity steps could be linked to the early development of bacterial symbionts acting for the chemical defense of the sponge.\u003c/p\u003e \u003cp\u003eFrom t\u003csub\u003e4\u003c/sub\u003e to t\u003csub\u003e6\u003c/sub\u003e (stage 5), the development of an important colonization of the \u003cem\u003eChlorella\u003c/em\u003e-like symbionts in the juvenile tissues, can be considered as a major factor explaining the differences with the previous sampling times (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). Even if the bacterial communities associated with the \u003cem\u003eChlorella\u003c/em\u003e-like symbionts are similar to those of the freshwater sponge host [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e], a part of this temporal change might be associated with the microalgal enrichment. Additionally, these microalgal symbionts provide a source of nutrients for the juveniles through the production of photosynthates which might also affect the bacterial composition in return.\u003c/p\u003e \u003cp\u003e \u003cb\u003eThe bacterial community transmitted from the gemmule surface plays an important role in the microbiome stability of the juveniles\u003c/b\u003e \u003c/p\u003e \u003cp\u003eOur study tends to confirm the efficiency of the sterilization protocol of the gemmule surface developed by Leys \u003cem\u003eet al\u003c/em\u003e. [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. More precisely, no 16S rRNA sequences were successfully amplified when the gemmules were washed with hydrogen peroxide (-EM gemmules) and their DNA extracted without breaking the gemmule coating. Interestingly, the only way to amplify the 16S rRNA gene from the -EM gemmules was to break the gemmules coatings (by crushing them) before the DNA extraction. This result proves that gemmules host epibacteria within their core, and confirms that bacteria can be transmitted through the gemmule surface but also within the gemmule, as previously suggested [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. However, this result comes in opposition with culture-based observations from Rozenfeld and Curtis, [\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e], suggesting a \u0026ldquo;spontaneous bacterial sterility\u0026rdquo; inside of \u003cem\u003eE. fluviatilis\u003c/em\u003e gemmules. Here, we suggest that the methodology employed didn\u0026rsquo;t allow to correctly assess such sterility, considering the challenges associated with the culture of sponge-associated bacterial endosymbionts [\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e]. qPCR or microscopy analyses (e.g. FISH techniques or environmental SEM) could be considered in future studies to confirm our observation.\u003c/p\u003e \u003cp\u003eThe sterilization of the gemmule surfaces, resulting in the absence of the EM, was found to be a major factor impacting the bacterial diversity of the gemmules and juveniles. More precisely, the removal of the EM reduces the \u003cem\u003eα\u003c/em\u003e-diversity of the juveniles during the first steps of their growth, but also significantly changes the \u003cem\u003eβ\u003c/em\u003e-diversity all along the experiment (including before the hatching, at t\u003csub\u003e0\u003c/sub\u003e). Additionally, the \u003cem\u003eβ\u003c/em\u003e-diversity dispersion was significantly higher without the EM for the early stages. This result suggests that such an epibacterial community is not only distinct from the bacterial diversity transmitted inside of the gemmule, but also participates in the stability of the whole microbiome during early juvenile development (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). The Anna Karenina principle adapted to sponge holobionts provides a good understanding on the importance of stable microbiomes [\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e]. This principle coined that high \u003cem\u003eβ\u003c/em\u003e-diversity dispersion could result from various dysbiosis scenarios induced by environmental stresses at the holobiont scale. The ability to regulate a stable and less dispersed microbiome over time is then an indicator of healthy microbiomes. In marine environments, the importance of such stability was demonstrated with the Haplosclerida \u003cem\u003ePetrosia ficiformis\u003c/em\u003e, during the acquisition of host-specific cyanobacterial symbionts which might provide antioxidants protections for the host [\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eDifferential analysis performed at t\u003csub\u003e0\u003c/sub\u003e and t\u003csub\u003e1\u003c/sub\u003e allowed to identify enriched taxa in +\u0026thinsp;EM compared to -EM, to target bacterial taxa specifically associated with the gemmule surface. The \u003cem\u003eComamonadaceae\u003c/em\u003e and its main genus \u003cem\u003eHydrogenophaga\u003c/em\u003e, appear as major potential taxa attached to the gemmule surface (enriched in +\u0026thinsp;EM condition). This genus is often described as autotrophic hydrogen oxidizers and denitrifiers [\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e], living in biofilms such as biofilm reactors [\u003cspan additionalcitationids=\"CR70\" citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e] or epilithic biofilms in lake Baikal [\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e]. In this latter environment, this genus has also been observed within the freshwater sponge \u003cem\u003eBaikalospongia fungiformis\u003c/em\u003e [\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e]. These observations suggest that such taxa might be adapted to colonize biofilms within freshwater sponges, such as the external surface of their gemmules. In this study, the \u003cem\u003eComamonadaceae\u003c/em\u003e family was already hypothesized to play an important role during the development of the aquiferous system. Consequently, the attachment of bacteria to the gemmule surface could constitute an important transmission mode shaping the development of juvenile holobionts.\u003c/p\u003e \u003cp\u003eTwo hypotheses can be considered about the origin of taxa transmitted on the gemmule surface before the hatching: (i) a VT hypothesis where these taxa are vertically transmitted from the maternal tissue, or (ii) an \u0026ldquo;early HT\u0026rdquo; hypothesis where planktonic bacteria directly colonize the gemmule biofilm before or during the planktonic phase of the gemmule (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). No stable association with the maternal cells would be involved in this case. This early HT hypothesis came then in contrast with both VT hypothesis and the traditional HT hypothesis, where in the latter the acquisition of planktonic colonizers occurs only once the juvenile has hatched, and filters the freshwater. A similar scenario to the early HT was suggested for \u003cem\u003eE. muelleri\u003c/em\u003e [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], since epilithic biofilms were found to have similar bacterial communities to those from gemmules described in [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn our study, only a limited number of ASVs were found to be shared between the gemmules/juveniles and the \u003cem\u003ein situ\u003c/em\u003e adult sponges, representing between 2 and 44.7% of the sequences. Surprisingly, almost no ASVs were shared between the gemmules/juveniles and the planktonic community (FF0.45 samples), suggesting that the remaining part of the community is acquired from (i) ultra-rare free-living taxa being undetected, or (ii) other sources than maternal tissues (VT) and free-living planktonic bacteria. In line with this hypothesis, we suggest that the colonization of the gemmule surface could also be achieved through an early HT involving planktonic particle-attached bacteria, even if their contribution could not be directly assessed through this study. As described in marine environments, the contribution of planktonic colonizers in the formation of biofilm can be challenging to assess, but particle-attached bacteria and the ultra-rare taxa are important to consider [\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003cb\u003eThe\u003c/b\u003e \u003cb\u003eTerasakiellaceae\u003c/b\u003e \u003cb\u003eas a dominant family vertically transmitted inside of the gemmules\u003c/b\u003e\u003c/p\u003e \u003cp\u003eThe \u003cem\u003eβ-\u003c/em\u003ediversity analysis conducted at t\u003csub\u003e0\u003c/sub\u003e indicated a significant effect of the sterilization of the surface, confirming the difference of bacterial composition transmitted inside of the gemmules compared to their surface. The \u003cem\u003eTerasakiellaceae\u003c/em\u003e family are good candidates for these types of taxa transmitted specifically from the maternal tissues to the inside of a gemmule. More precisely, the differential analysis revealed their specificity for the -EM samples at t\u003csub\u003e0\u003c/sub\u003e and t\u003csub\u003e1\u003c/sub\u003e. For the same sampling times, this family also gathers the dominant ASVs shared between the adult sponge and the -EM gemmules. Additionally, within the \u003cem\u003ein situ\u003c/em\u003e adult sponge samples, this family was found mainly dominant in the M2 samples but less abundant in the M3 samples. This observation can be explained since the formation of the gemmules and their thesocytes mainly occurred in late November (after M2), while in December (M3) most of the tissue from the adult sponge was found absent and full of gemmules within its skeleton. As the \u003cem\u003ein situ\u003c/em\u003e adult sponge samples were extracted mostly with unbroken gemmules, the formation of the gemmules could lead to a lower relative abundance of \u003cem\u003eTerasakiellaceae\u003c/em\u003e in the M3 sponge tissues, which are mostly transmitted and located within the gemmules. Consequently, these specific taxa might be important endosymbionts, transmitted to the thesocytes within the gemmule (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). Further studies are needed to confirm their location and their role as potential endosymbionts.\u003c/p\u003e \u003cp\u003e \u003cem\u003eTerasakiellaceae\u003c/em\u003e were found in the primmorphs of the diseased freshwater sponge \u003cem\u003eLubomirskia baikalensis\u003c/em\u003e [\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e], but also within marine holobionts such as the sponge \u003cem\u003eSuberites massa\u003c/em\u003e [\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e], as well as an important diversity of corals [\u003cspan additionalcitationids=\"CR78 CR79\" citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e80\u003c/span\u003e]. Taxa within this family are also known as nitrogen fixers [\u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e81\u003c/span\u003e], and may play an important role for the nitrogen regulation within the gemmule. Others taxa enriched in the -EM samples such as the \u003cem\u003eRhizobiaceae\u003c/em\u003e family and its dominant genus \u003cem\u003eRhizobium\u003c/em\u003e (covering also \u003cem\u003eAllorhizobium\u003c/em\u003e, \u003cem\u003eNeorhizobium\u003c/em\u003e and \u003cem\u003eParararhizobium\u003c/em\u003e) known for its denitrification role [\u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e82\u003c/span\u003e, \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e83\u003c/span\u003e], might also play similar functions within the gemmule. To date, little is known about the role of bacterial symbionts within sponge cells undergoing diapause states such as the thesocytes cells within the gemmules. The expression of glutamine metabolism, apoptotic process, and oxidation-reduction system was found to be specific to stage 0 [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Further studies are needed to better investigate the potential link between these specific metabolisms and these putative endosymbionts stored in the gemmules.\u003c/p\u003e \u003cp\u003e \u003cb\u003eEffect of the presence of exogenous free-living bacteria during the juveniles growth: horizontal acquisition of symbionts, or additional food source (FS)?\u003c/b\u003e \u003c/p\u003e \u003cp\u003eAnother major result of this study was the effect of the absence/presence of ambient free-living bacteria on the bacterial \u003cem\u003eβ\u003c/em\u003e-diversity of the juveniles. These differences can be explained through two hypotheses: (i) an HT hypothesis, related to the acquisition of bacterial symbionts from the medium, and (ii) a food source (FS) hypothesis, where the bacterioplankton is filtered by the sponge for its nutrition. In this scenario, this specific food source would be linked to a different physiology of the juveniles impacting indirectly their microbiome. As mainly discussed for marine sponges, the delineation between these two processes can be hard to distinguish [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn this study, only three low abundant ASVs (\u003cem\u003ePolynucleobacter\u003c/em\u003e, Candidatus \u003cem\u003eLimnoluna\u003c/em\u003e, and \u003cem\u003eBradyrhizobium\u003c/em\u003e; each \u0026lt;\u0026thinsp;0.04%) were occasionally shared between the prefiltered water (FF0.45 samples) and the juvenile sponges, suggesting that the horizontal acquisition of FL bacteria by the juveniles was nearly nonexistent during this experiment. The effect of the presence of the FL bacteria in the medium on the \u003cem\u003eβ\u003c/em\u003e-diversity is then more likely explained by the FS hypothesis (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). Additionally, the \u003cem\u003eα\u003c/em\u003e-diversity analysis also goes in favor of the rejection of the HT hypothesis, since the richness (estimated with Chao1 index) of the +\u0026thinsp;FB juveniles was not significantly higher compared to the -FB.\u003c/p\u003e \u003cp\u003eFrom t\u003csub\u003e1\u003c/sub\u003e to t\u003csub\u003e3\u003c/sub\u003e, the difference of \u003cem\u003eβ\u003c/em\u003e-diversity between +\u0026thinsp;FB and -FB samples was mainly observed under the -EM condition, while being less important in +\u0026thinsp;EM. Consequently, the absence of epibiotic microbiome (EM) increases the effect of the absence of ambient free-living bacteria (FB). Their absence in the medium might then represent cumulative stress to the juveniles growing with an unstable microbiome due to the EM absence. In line with the FS hypothesis, the absence of FB as a food source might explain this as an additional source of stress in +\u0026thinsp;EM, leading to a distinct bacterial community.\u003c/p\u003e \u003cp\u003eAs the differences between +\u0026thinsp;FB and -FB were linked to the food source availability, these differences can be investigated in terms of trophic strategy within the bacterial community. Indeed, the juveniles growing with more food could provide more nutrients for their associated microbiome. Interestingly, this assumption appears to be particularly consistent with the specific taxa found enriched in +\u0026thinsp;FB or -FB samples, based on differential analyses at t\u003csub\u003e1\u003c/sub\u003e. For example, several\u0026thinsp;+\u0026thinsp;FB enriched taxa such as \u003cem\u003eOxalobacteraceae\u003c/em\u003e (within +\u0026thinsp;EM samples), but also \u003cem\u003ePseudomonas\u003c/em\u003e and \u003cem\u003eFlavobacterium\u003c/em\u003e (within -EM samples) are classically known as copiotroph [\u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e84\u003c/span\u003e, \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e85\u003c/span\u003e]. Conversely, -FB enriched taxa such as \u003cem\u003eRheinemera\u003c/em\u003e (in +\u0026thinsp;EM) or \u003cem\u003eBrevundimonas\u003c/em\u003e and \u003cem\u003eCaulobacter\u003c/em\u003e (in both -EM and +\u0026thinsp;EM) are typically oligotrophic bacteria [\u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e85\u003c/span\u003e, \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e86\u003c/span\u003e]. This observation strongly supports the FS hypothesis, indicating that in the absence of FB, less nutrients can be provided to the microbiome, resulting in a higher relative abundance of oligotrophic taxa. Conversely, the presence of FB turned into a source of organic matter by the juveniles, could favor copiotrophs that could take advantage of this condition and quickly grow within the sponge (i.e. r-strategists bacteria).\u003c/p\u003e \u003cp\u003eFinally, even if almost no ASVs from the prefiltered water were found to be horizontally acquired within the juveniles, a slightly higher number of these free-living ASVs were found to be shared with the \u003cem\u003ein situ\u003c/em\u003e sponges (17 in total, with dominant ones belonging to \u003cem\u003eSporichthyaceae\u003c/em\u003e). This observation suggests that horizontal acquisition within \u003cem\u003eS. lacustris\u003c/em\u003e could still occasionally happen after a longer term of development. Further studies on these transmissions in natural conditions could provide relevant insight to better consider the complete dynamics of the holobiont through its full life cycle.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThe bacterial diversity within \u003cem\u003eS. lacustris\u003c/em\u003e juveniles was found to be shaped by three factors: the life cycle stage, the presence of epibacteria on the gemmule, and finally the presence of ambient free-living bacteria. The osculum formation together with the development of a canal system, could lead to an active filtration of the environment which might induce a change in the microbiome \u003cem\u003eβ\u003c/em\u003e-diversity. Thereafter, the colonization of the \u003cem\u003eChlorella\u003c/em\u003e-like symbionts could also provide an additional niche for the development of new bacterial symbionts within the holobiont.\u003c/p\u003e \u003cp\u003eOur study revealed a complex diversity of microbial acquisition modes within the \u003cem\u003eS. lacustris\u003c/em\u003e holobiont model. For instance, multiple vertical acquisition scenarios can be considered with both transmission within the gemmule or on its surface. Importantly, the transmission of the microbiome on the gemmule surface was found to be essential for the whole holobiont stability during the first days. In line with the Anna Karenina principle [\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e], the absence of these epibacterial communities might represent a stress condition. While a recruitment of free-living bacteria by filtration of the juveniles was found to be nearly impossible, an alternative horizontal transmission scenario can however be considered with a colonization of planktonic bacteria on the gemmule biofilm, before the hatching. The community of the gemmule biofilm would then be composed of both vertically and horizontally transmitted bacteria, in line with the mixed-mode transmission hypothesis [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Such diversity and complexity of transmission modes need to be better considered in future studies, and our results associated with the importance of the microbiome on the gemmule surface provide new perspectives that could also be investigated for sponge larvae.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eASV: Amplicon Sequence Variant\u003c/p\u003e\n\u003cp\u003eEM: Epibiotic Microbiome\u003c/p\u003e\n\u003cp\u003eFF0.2: Freshwater filtered upon 0.2\u0026mu;m pore size filters\u003c/p\u003e\n\u003cp\u003eFF0.45: Freshwater filtered upon 0.45\u0026mu;m pore size filters\u003c/p\u003e\n\u003cp\u003eFB: Free-living Bacteria\u003c/p\u003e\n\u003cp\u003eFS: Food Source\u003c/p\u003e\n\u003cp\u003eHT: Horizontal Transmission\u003c/p\u003e\n\u003cp\u003eSD: Standard Deviation\u003c/p\u003e\n\u003cp\u003eVT: Vertical Transmission\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAvailability of data and material\u003c/h2\u003e\n\u003cp\u003e16S rRNA gene sequences were deposited and are publicly available in the NCBI Sequences Read Archive (SRA) under the BioProject ID PRJNA1077127, accession number. The R scripts used for all the 16S rRNA gene metabarcoding analysis can be found at \u003cu\u003ehttps://github.com/BenoitPAIX/Gemmules_microbiome\u003c/u\u003e\u003c/p\u003e\n\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThis work was funded by the NWO-VIDI with project number 16.161.301.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eAuthors\u0026apos; contributions\u003c/h2\u003e\n\u003cp\u003eBP, EvdV and NdV designed the experiment and performed the fieldwork, BP and EvdV performed the lab work, BP processed the data, analyzed the results and wrote the first draft of the manuscript. NdV and BP reviewed the manuscript and finalized the last draft.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eAcknowledgements\u003c/h2\u003e\n\u003cp\u003eWe thank Mike Hynes, Anouk Langerak and Jerry Weidema for their help during the sampling in October.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003ede Goeij JM, Lesser MP, Pawlik JR. Nutrient Fluxes and Ecological Functions of Coral Reef Sponges in a Changing Ocean. In: Carballo JL, Bell JJ, editors. Clim Change Ocean Acidif Sponges Impacts Mult Levels Organ. Cham: Springer International Publishing; 2017 p. 373\u0026ndash;410.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSlaby BM, Franke A, Rix L, Pita L, Bayer K, Jahn MT, et al. Marine sponge holobionts in health and disease. In: Li Z, editor. Symbiotic Microbiomes Coral Reefs Sponges Corals . Dordrecht: Springer Netherlands; 2019 p. 81\u0026ndash;104.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eThomas T, Moitinho-Silva L, Lurgi M, Bj\u0026ouml;rk JR, Easson C, Astudillo-Garc\u0026iacute;a C, et al. Diversity, structure and convergent evolution of the global sponge microbiome. Nat Commun. 2016;7:11870.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003ePosadas N, Baquiran JIP, Nada MAL, Kelly M, Conaco C. Microbiome diversity and host immune functions influence survivorship of sponge holobionts under future ocean conditions. ISME J. 2022;16:58\u0026ndash;67.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eDe Castro-Fern\u0026aacute;ndez P, Ballest\u0026eacute; E, Angulo-Preckler C, Biggs J, Avila C, Garc\u0026iacute;a-Aljaro C. How does heat stress affect sponge microbiomes? Structure and resilience of microbial communities of marine sponges from different habitats. Front Mar Sci. 2023;9.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003ePita L, Rix L, Slaby BM, Franke A, Hentschel U. The sponge holobiont in a changing ocean: from microbes to ecosystems. Microbiome. 2018;6:46.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eD\u0026iacute;ez-Vives C, Koutsouveli V, Conejero M, Riesgo A. Global patterns in symbiont selection and transmission strategies in sponges. Front Ecol Evol . 2022;10.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eBj\u0026ouml;rk JR, D\u0026iacute;ez-Vives C, Astudillo-Garc\u0026iacute;a C, Archie EA, Montoya JM. Vertical transmission of sponge microbiota is inconsistent and unfaithful.\u0026nbsp;Nat Ecol Evol. 2019;3:1172\u0026ndash;83.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003ede Oliveira BFR, Freitas-Silva J, S\u0026aacute;nchez-Robinet C, Laport MS.\u0026nbsp;Transmission of the sponge microbiome: moving towards a unified model.\u0026nbsp;Environ Microbiol Rep. 2020;12:619\u0026ndash;38.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSipkema D, de Caralt S, Morillo JA, Al-Soud WA, S\u0026oslash;rensen SJ, Smidt H, et al.\u0026nbsp;Similar sponge-associated bacteria can be acquired via both vertical and horizontal transmission. Environ Microbiol. 2015;17:3807\u0026ndash;21.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eCarrier TJ, Maldonado M, Schmittmann L, Pita L, Bosch TCG, Hentschel U. Symbiont transmission in marine sponges: reproduction, development, and metamorphosis. BMC Biol. 2022;20:100.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eVrijenhoek RC. Genetics and evolution of Deep-Sea chemosynthetic bacteria and their invertebrate hosts.\u0026nbsp;In: Kiel S, editor. Vent Seep Biota Asp Microbes Ecosyst . Dordrecht: Springer Netherlands; 2010. p. 15\u0026ndash;49.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eEbert D. The epidemiology and evolution of symbionts with mixed-mode transmission. Annu Rev Ecol Evol Syst. 2013;44:623\u0026ndash;43.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eEreskovsky AV, Gonobobleva E, Vishnyakov A. Morphological evidence for vertical transmission of symbiotic bacteria in the viviparous sponge \u003cem\u003eHalisarca dujardini\u003c/em\u003e Johnston (Porifera, Demospongiae, Halisarcida). Mar Biol. 2005;146:869\u0026ndash;75.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSharp KH, Eam B, Faulkner DJ, Haygood MG. Vertical transmission of diverse microbes in the tropical sponge \u003cem\u003eCorticium\u003c/em\u003e sp.\u0026nbsp;Appl Environ Microbiol. 2007;73:622\u0026ndash;9.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eEngelberts JP, Abdul Wahab MA, Maldonado M, Rix L, Marangon E, Robbins SJ, et al.\u0026nbsp;Microbes from Mum: symbiont transmission in the tropical reef sponge \u003cem\u003eLanthella basta\u003c/em\u003e. ISME Commun. 2022;2:1\u0026ndash;12.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eCarrier TJ, Schmittmann L, Jung S, Pita L, Hentschel U. Maternal provisioning of an obligate symbiont in a sponge. Ecol Evol. 2023;13:e10012.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eStrano F, Micaroni V, Thomas T, Woods L, Davy SK, Bell JJ. Marine heatwave conditions drive carryover effects in a temperate sponge microbiome and developmental performance. Proc R Soc B Biol Sci. 2023;290:20222539.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eTuron M, C\u0026aacute;liz J, Garate L, Casamayor EO, Uriz MJ. Showcasing the role of seawater in bacteria recruitment and microbiome stability in sponges. Sci Rep. 2018;8:15201.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003ede Oliveira BFR, Lopes IR, Canellas ALB, Muricy G, Dobson ADW, Laport MS. Not that close to mommy: horizontal transmission seeds the microbiome associated with the marine sponge \u003cem\u003ePlakina cyanorosea\u003c/em\u003e. Microorganisms. 2020;8:1978.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLo Giudice A, Rizzo C. Freshwater Sponges as a neglected reservoir of bacterial biodiversity. Microorganisms. 2024;12:25.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003ePronzato R, Pisera A, Manconi R. Fossil freshwater sponges: taxonomy, geographic distribution, and critical review. Acta Palaeontol Pol. 2017;62:468\u0026ndash;95.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eManconi R, Pronzato R. Global diversity of sponges (Porifera: Spongillina) in freshwater. In: Balian EV, L\u0026eacute;v\u0026ecirc;que C, Segers H, Martens K, editors. Freshw Anim Divers Assess . Dordrecht: Springer Netherlands; 2008\u003c/li\u003e\n \u003cli\u003eGilbert JJ, Allen HL. Studies on the physiology of the green freshwater sponge \u003cem\u003eSpongilla lacustris\u003c/em\u003e : primary productivity, organic matter, and chlorophyll content. SIL Proc 1922-2010. 1973;18:1413\u0026ndash;20.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eWilkinson CR. Nutrient translocation from green algal symbionts to the freshwater sponge \u003cem\u003eEphydatia fluviatilis\u003c/em\u003e. Hydrobiologia. 1980;75:241\u0026ndash;50.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eCook CB. Metabolic interchange in algae-invertebrate symbiosis. Intern Rev Cytol Suppl. 1983;177\u0026ndash;210.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eKulakova NV, Sakirko MV, Adelshin RV, Khanaev IV, Nebesnykh IA, P\u0026eacute;rez T. Brown rot syndrome and changes in the bacterial\u0026nbsp;сommunity of the Baikal sponge \u003cem\u003eLubomirskia baikalensis\u003c/em\u003e. Microb Ecol. 2018;75:1024\u0026ndash;34.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eHall C, Camilli S, Dwaah H, Kornegay B, Lacy C, Hill MS, et al. Freshwater sponge hosts and their green algae symbionts: a tractable model to understand intracellular symbiosis. PeerJ. 2021;9:e10654.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eHustus K, D\u0026iacute;ez-Vives C, Mitsi K, Nutakki J, Kering V, Nguyen IT, et al. Algal symbionts of the freshwater sponge \u003cem\u003eEphydatia muelleri\u003c/em\u003e. Symbiosis. 2023\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eCosta R, Keller-Costa T, Gomes NCM, da Rocha UN, van Overbeek L, van Elsas JD. Evidence for selective bacterial community structuring in the freshwater sponge \u003cem\u003eEphydatia fluviatilis\u003c/em\u003e.\u0026nbsp;Microb Ecol. 2013;65:232\u0026ndash;44.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSugden S, Holert J, Cardenas E, Mohn WW, Stein LY.\u0026nbsp;Microbiome of the freshwater sponge \u003cem\u003eEphydatia muelleri\u003c/em\u003e shares compositional and functional similarities with those of marine sponges. ISME J. 2022;16:2503\u0026ndash;12.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eManconi R, Pronzato R. Rediscovery of the type material of \u003cem\u003eSpongilla lacustris\u003c/em\u003e (L., 1759) in the Linnean herbarium. Ital J Zool. 2000;67:89\u0026ndash;92.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eManconi R, Pronzato R. Life cycle of \u003cem\u003eSpongilla lacustris\u003c/em\u003e (Porifera, Spongillidae): a cue for environment-dependent phenotype. Hydrobiologia. 1991;220:155\u0026ndash;60.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLeys S, Grombacher L, Hill A. Hatching and freezing gemmules from the freshwater sponge \u003cem\u003eEphydatia muelleri\u003c/em\u003e. 2019\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eKenny NJ, Francis WR, Rivera-Vic\u0026eacute;ns RE, Juravel K, de Mendoza A, D\u0026iacute;ez-Vives C, et al.\u0026nbsp;Tracing animal genomic evolution with the chromosomal-level assembly of the freshwater sponge \u003cem\u003eEphydatia muelleri\u003c/em\u003e. Nat Commun. 2020;11:3676.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eRuperti F, Becher I, Stokkermans A, Wang L, Marschlich N, Potel C, et al. Molecular profiling of sponge deflation reveals an ancient relaxant-inflammatory response. Curr Biol. 2024\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eGernert C, Gl\u0026ouml;ckner FO, Krohne G, Hentschel U. Microbial diversity of the freshwater sponge \u003cem\u003eSpongilla lacustris\u003c/em\u003e.\u0026nbsp;Microb Ecol. 2005;50:206\u0026ndash;12.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eGraffius S, Garz\u0026oacute;n JFG, Zehl M, Pjevac P, Kirkegaard R, Flieder M, et al.\u0026nbsp;Secondary metabolite production potential in a microbiome of the freshwater sponge \u003cem\u003eSpongilla lacustris\u003c/em\u003e. Microbiol Spectr. 2023;11:e04353-22.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eTurner CR, Miller DJ, Coyne KJ, Corush J. Improved methods for capture, extraction, and quantitative assay of environmental DNA from asian bigheaded carp (\u003cem\u003eHypophthalmichthys\u003c/em\u003e spp.). PLOS ONE. 2014;9:e114329.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eKlindworth A, Pruesse E, Schweer T, Peplies J, Quast C, Horn M, et al. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 2013;41:e1.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eCallahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: High-resolution sample inference from Illumina amplicon data. Nat Methods. 2016;13:581\u0026ndash;3.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eCallahan BJ, McMurdie PJ, Holmes SP. Exact sequence variants should replace operational taxonomic units in marker-gene data analysis. ISME J. 2017;11:2639\u0026ndash;43.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eCallahan BJ, Sankaran K, Fukuyama JA, McMurdie PJ, Holmes SP. Bioconductor Workflow for Microbiome Data Analysis: from raw reads to community analyses . F1000Research; 2016\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eQuast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 2013;41:D590\u0026ndash;6.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eMcMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLOS ONE. 2013;8:e61217.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eDavis 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;6:226.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eOksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, et al. vegan: Community ecology package. 2019.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eMendiburu F de. Agricolae: Statistical procedures for agricultural research. 2019.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eMcMurdie PJ, Holmes S. Waste not, want not: why rarefying microbiome data is inadmissible. PLOS Comput Biol. 2014;10:e1003531.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eGloor GB, Macklaim JM, Pawlowsky-Glahn V, Egozcue JJ. Microbiome datasets are compositional: and this is not optional. Front Microbiol. 2017;8:2224.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eFoster ZSL, Sharpton TJ, Gr\u0026uuml;nwald NJ. Metacoder: An R package for visualization and manipulation of community taxonomic diversity data. PLOS Comput Biol. 2017;13:e1005404.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLahti L, Shetty S, Blake T, Salojarvi J. Tools for microbiome analysis in R. Version. 2017;1:28.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLarsson J, Godfrey AJR, Gustafsson P, algorithms) DHE (geometric, code) EH (root solver, Priv\u0026eacute; F. eulerr: Area-Proportional Euler and Venn Diagrams with Ellipses . 2022.\u003c/li\u003e\n \u003cli\u003eFieth RA, Gauthier M-EA, Bayes J, Green KM, Degnan SM. Ontogenetic changes in the bacterial symbiont community of the tropical demosponge \u003cem\u003eAmphimedon queenslandica\u003c/em\u003e: metamorphosis is a new beginning. Front Mar Sci . 2016;3.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eWu S, Ou H, Liu T, Wang D, Zhao J. Structure and dynamics of microbiomes associated with the marine sponge \u003cem\u003eTedania\u003c/em\u003e sp. during its life cycle. FEMS Microbiol Ecol. 2018;94:fiy055.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eTuron M, Ford M, Maldonado M, Sitj\u0026agrave; C, Riesgo A, D\u0026iacute;ez-Vives C. Microbiome transmission through the ontogeny of the marine sponge \u003cem\u003eCrambe crambe\u003c/em\u003e. 2023;\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eFunayama N, Nakatsukasa M, Hayashi T, Agata K. Isolation of the choanocyte in the fresh water sponge, \u003cem\u003eEphydatia fluviatilis\u003c/em\u003e and its lineage marker, Ef annexin.\u0026nbsp;Dev Growth Differ. 2005;47:243\u0026ndash;53.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eBart MC, Vet SJ de, Bakker DM de, Alexander BE, Oevelen D van, Loon EE van, et al.\u0026nbsp;Spiculous skeleton formation in the freshwater sponge \u003cem\u003eEphydatia fluviatilis\u003c/em\u003e under hypergravity conditions. PeerJ. 2019;6:e6055.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLi M, Wang K, Jia C, Liu T, Yang S, Ou H, et al. Bacteroidetes bacteria, important players in the marine sponge larval development process. iScience. 2021;24:102662.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLaport MS, Pinheiro U, Rachid CTC da C. Freshwater sponge \u003cem\u003eTubella variabilis\u003c/em\u003e presents richer microbiota than marine sponge species. Front Microbiol . 2019;10.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eGaikwad S, Shouche YS, Gade WN. Microbial community structure of two freshwater sponges using Illumina MiSeq sequencing revealed high microbial diversity. AMB Express. 2016;6:40.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eKeller-Costa T, Jousset A, van Overbeek L, van Elsas JD, Costa R. The freshwater sponge \u003cem\u003eEphydatia fluviatilis\u003c/em\u003e harbours diverse \u003cem\u003ePseudomonas\u003c/em\u003e species (Gammaproteobacteria, Pseudomonadales) with broad-spectrum antimicrobial activity. PloS One. 2014;9:e88429.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eKaluzhnaya OV, Itskovich VB. Diversity of potential producers of bioactive metabolites having polyketide nature in the Baikal sponge community of \u003cem\u003eRezinkovia echinata\u003c/em\u003e. Limnol Freshw Biol. 2020;423\u0026ndash;8.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eRozenfeld F, Curtis ASG. A technique for sterile culture of fresh water sponges. Experientia. 1980;36:371\u0026ndash;3.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eDat TTH, Steinert G, Cuc NTK, Smidt H, Sipkema D. Bacteria cultivated from sponges and bacteria not yet cultivated from sponges\u0026mdash;a review. Front Microbiol . 2021;12.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eZaneveld JR, McMinds R, Vega Thurber R. Stress and stability: applying the Anna Karenina principle to animal microbiomes. Nat Microbiol. 2017;2:1\u0026ndash;8.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eBritstein M, Cerrano C, Burgsdorf I, Zoccarato L, Kenny NJ, Riesgo A, et al. Sponge microbiome stability during environmental acquisition of highly specific photosymbionts. Environ Microbiol. 2020;22:3593\u0026ndash;607.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eWillems A, Gillis M. \u003cem\u003eHydrogenophaga\u003c/em\u003e. In: Whitman WB, editor. Bergeys Man Syst Archaea Bact . 1st ed. Wiley; 201. p. 1\u0026ndash;15.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eZhao H-P, Van Ginkel S, Tang Y, Kang D-W, Rittmann B, Krajmalnik-Brown R. Interactions between perchlorate and nitrate reductions in the biofilm of a hydrogen-based membrane biofilm reactor. Environ Sci Technol. 2011;45:10155\u0026ndash;62.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eZhou L, Xu X, Xia S. Effects of sulfate on simultaneous nitrate and selenate removal in a hydrogen-based membrane biofilm reactor for groundwater treatment: Performance and biofilm microbial ecology. Chemosphere. 2018;211:254\u0026ndash;60.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eRezaei R, Aghapour AA, Chavshin AR, Bargeshadi R. Biodegradation of the atenolol and nitrogen removal using the sequencing batch biofilm reactor. Bioresour Technol Rep. 2022;18:101109.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSukhanova EV, Shtykova YuR, Suslova MYu, Pestunova OS, Kostornova TYa, Khanaev IV, et al. Diversity and physiological and biochemical properties of heterotrophic bacteria isolated from lake Baikal epilithic biofilms. Microbiology. 2019;88:324\u0026ndash;34.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eKaluzhnaya OV, Itskovich VB. Features of diversity of polyketide synthase genes in the community of freshwater sponge \u003cem\u003eBaikalospongia fungiformis\u003c/em\u003e. Russ J Genet. 2022;58:336\u0026ndash;46.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eCat\u0026atilde;o E, Pollet T, Garnier C, Barry-Martinet R, Rehel K, Linossier I, et al.\u0026nbsp;Temperate and tropical coastal waters share relatively similar microbial biofilm communities while free-living or particle-attached communities are distinct. Mol Ecol. 2021;30:2891\u0026ndash;904.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eChernogor L, Klimenko E, Khanaev I, Belikov S. Microbiome analysis of healthy and diseased sponges \u003cem\u003eLubomirskia baicalensis\u003c/em\u003e by using cell cultures of primmorphs. PeerJ. 2020;8:e9080.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLamb CE, Watts JEM. Microbiome species diversity and seasonal stability of two temperate marine sponges \u003cem\u003eHymeniacidon perlevis\u003c/em\u003e and \u003cem\u003eSuberites massa\u003c/em\u003e. Environ Microbiome. 2023;18:52.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eWeiler BA, Verhoeven JTP, Dufour SC.\u0026nbsp;Bacterial communities in tissues and surficial mucus of the cold-water coral \u003cem\u003eParagorgia arborea\u003c/em\u003e. Front Mar Sci . 2018;5.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eRosales SM, Clark AS, Huebner LK, Ruzicka RR, Muller EM. Rhodobacterales and Rhizobiales are associated with stony coral tissue loss disease and its suspected sources of transmission. Front Microbiol . 2020;11.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLi J, Long L, Zou Y, Zhang S. Microbial community and transcriptional responses to increased temperatures in coral \u003cem\u003ePocillopora damicornis\u003c/em\u003e holobiont. Environ Microbiol. 2021;23:826\u0026ndash;43.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eQuintanilla E, Rodrigues CF, Henriques I, Hil\u0026aacute;rio A. Microbial associations of abyssal Gorgonians and Anemones (\u0026gt;4,000 m depth) at the Clarion-Clipperton fracture zone. Front Microbiol . 2022;13\u003c/li\u003e\n \u003cli\u003eTiedje J. Ecology of denitrification and dissimilatory nitrate reduction to ammonium.\u0026nbsp;Methods Soil Anal Part 2 Chem Microbiol Prop. 1988. p. 179\u0026ndash;244.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLv Y, Chen X, Zhang X, Zhu C, Pan Y, Sun T, et al.\u0026nbsp;Denitrification for acidic wastewater treatment: Long-term performance, microbial communities, and nitrous oxide emissions.\u0026nbsp;J Biosci Bioeng. 2022;134:513\u0026ndash;20.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLong Y, Ma Y, Wan J, Wang Y, Tang M, Fu H, et al.\u0026nbsp;Denitrification efficiency, microbial communities and metabolic mechanisms of corn cob hydrolysate as denitrifying carbon source. Environ Res. 2023;221:115315.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSchlatter DC, Hansen JC, Schillinger WF, Sullivan TS, Paulitz TC.\u0026nbsp;Common and unique rhizosphere microbial communities of wheat and canola in a semiarid Mediterranean environment.\u0026nbsp;Appl Soil Ecol. 2019;144:170\u0026ndash;81.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eHellauer K, Michel P, Holland SI, H\u0026uuml;bner U, Drewes JE, Lauro FM, et al.\u0026nbsp;Inferring trophic conditions in managed aquifer recharge systems from metagenomic data. Sci Total Environ. 2021;772:145512.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eWilhelm RC. Following the terrestrial tracks of \u003cem\u003eCaulobacter\u003c/em\u003e - redefining the ecology of a reputed aquatic oligotroph. ISME J. 2018;12:3025\u0026ndash;37.\u0026nbsp;\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"environmental-microbiome","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sigs","sideBox":"Learn more about [Environmental Microbiome](https://environmentalmicrobiome.biomedcentral.com)","snPcode":"40793","submissionUrl":"https://submission.nature.com/new-submission/40793/3","title":"Environmental Microbiome","twitterHandle":"@bmc","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Freshwater sponge, Holobiont, Microbiome, Vertical transmission, Horizontal acquisition, Gemmule, Ontogeny","lastPublishedDoi":"10.21203/rs.3.rs-3973150/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3973150/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003e Sponges-associated bacteria play important roles in the physiology of their host, whose recruitment processes are crucial to maintain symbiotic associations. However, the acquisition of bacterial communities within freshwater sponges is still unexplored. \u003cem\u003eSpongilla lacustris\u003c/em\u003e is a model sponge widely distributed in European rivers and lakes, producing dormant cysts (named gemmules) for their asexual reproduction, before winter. Through an \u003cem\u003ein vitro\u003c/em\u003eexperiment, this study aims to describe the dynamics of bacterial communities and their transmission modes following the hatching of these gemmules.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e An overall change of bacterial \u003cem\u003eβ\u003c/em\u003e-diversity was observed through the ontology of the new juvenile sponges. These temporal differences were potentially linked first to the osculum acquisition and the development of a canal system, and then, the increasing colonization of the \u003cem\u003eChlorella\u003c/em\u003e-like photosymbionts. Gemmules hatching with a sterilized surface were found to have a more dispersed and less diverse microbiome, revealing the importance of gemmule epibacteria for the whole holobiont stability. These epibacteria were suggested to be vertically transmitted from the maternal tissues to the gemmule surface. Vertical transmission through the incorporation of bacterial communities inside of the gemmule, was also found as a dominant transmission mode, especially with the nitrogen fixers \u003cem\u003eTerasakiellaceae\u003c/em\u003e.\u003cem\u003e \u003c/em\u003eFinally, we showed that almost no ASVs were shared between the free-living community and the juveniles, suggesting that horizontal recruitment is unlikely to happen during the first stages of development. However, the free-living bacteria filtered are probably used as a source of nutrients, allowing an enrichment of copiotrophic bacteria already present within its microbiome.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions:\u003c/strong\u003e This study brings new insight for a better understanding of the microbiome acquisition during the first stages of freshwater sponge development. We showed the importance of epibacterial communities on gemmules for the whole holobiont stability, and demonstrated the near absence of recruitment of free-living bacteria during the first stages.\u003c/p\u003e","manuscriptTitle":"Dynamics, diversity, and roles of bacterial transmission modes during the first asexual life stages of the freshwater sponge Spongilla lacustris","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-02-23 05:40:13","doi":"10.21203/rs.3.rs-3973150/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-05-21T10:13:00+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-04-29T14:36:52+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-04-23T09:45:13+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"5ed39a41-f602-450f-92e2-4c20eedeaeb3","date":"2024-04-18T06:23:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"840df42d-1a49-4ead-9034-7beeac9c7aff","date":"2024-04-17T18:10:20+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-04-17T12:46:32+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-03-03T12:42:37+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-02-21T17:54:51+00:00","index":"","fulltext":""},{"type":"submitted","content":"Environmental Microbiome","date":"2024-02-20T15:11:42+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"environmental-microbiome","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sigs","sideBox":"Learn more about [Environmental Microbiome](https://environmentalmicrobiome.biomedcentral.com)","snPcode":"40793","submissionUrl":"https://submission.nature.com/new-submission/40793/3","title":"Environmental Microbiome","twitterHandle":"@bmc","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"82878eba-8d47-4c18-ba2f-fc881d17cf50","owner":[],"postedDate":"February 23rd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-06-21T14:48:48+00:00","versionOfRecord":{"articleIdentity":"rs-3973150","link":"https://doi.org/10.1186/s40793-024-00580-7","journal":{"identity":"environmental-microbiome","isVorOnly":false,"title":"Environmental Microbiome"},"publishedOn":"2024-06-08 14:48:47","publishedOnDateReadable":"June 8th, 2024"},"versionCreatedAt":"2024-02-23 05:40:13","video":"","vorDoi":"10.1186/s40793-024-00580-7","vorDoiUrl":"https://doi.org/10.1186/s40793-024-00580-7","workflowStages":[]},"version":"v1","identity":"rs-3973150","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3973150","identity":"rs-3973150","version":["v1"]},"buildId":"cTy_lsJlmDsVRNrSptgXS","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}