Microbiology of marine sponges: from community structure to symbiont function

Abstract

Marine sponges are filter-feeding metazoans that can host complex microbial communities which comprise as much as 35% of total sponge biomass. In this thesis I have employed high-throughput, next-generation sequencing technologies to study the sponge microbiota at two different scales. Firstly, I studied complex communities associated with different sponge assemblages, then subsequently conducted an in-depth investigation of an enigmatic sponge symbiont which has largely escaped attention until now. Analysis of the marine sponge microbiota poses unique conceptual and analytical challenges, as microbial species may number in the thousands. One way to overcome this issue is to consider only the persistent and/or abundant species, i.e. the „core‟ community. While this approach has been widely used to analyse diverse biological systems, including sponge microbiota, to date its robustness has not been rigorously evaluated. Thus, in this thesis I systematically evaluated the applicability of the core microbiota approach for the complex microbial communities of three Xestospongia species from southeast Sulawesi (Indonesia), using 16S rRNA gene-based amplicon sequencing (Illumina MiSeq). Different factors for OTU selection were then considered to generate a set of different core communities, including percentage occurrence, minimum abundance threshold and sample set selection. Alpha- and beta- diversity analyses conducted on the core communities were largely insensitive to major changes in core microbiota definition, thus revealing the robustness of this approach when considering closely related sponge species. Furthermore, none of the applied core definitions altered ecological network structure summarising interactions among bacteria within the sponges. Sponge reefs often comprise an array of different and sometimes phylogenetically distant sponge species, with most of them hosting distinct microbial communities. Thus, to further assess the strength and sensitivity of the core microbiota approach in complex sponge assemblages, I analysed the associated bacterial communities of 20 co-occurring sponge species from the south coast of Wellington (New Zealand), using the same 16S rRNA genebased amplicon sequencing approach described above. The application of different core definitions resulted in a marked (and uneven at sponge species level) decrease in bacterial OTU and phylum richness. As a consequence of this decrease in richness, alpha- and betadiversity patterns changed significantly. Therefore, although the application of a core microbiota approach may seem appropriate in closely related systems (e.g. congeneric sponges), I showed that this approach can have a profound influence on the results obtained when studying complex host species assemblages. While sponge microbiota surveys have tended to focus on the study of a few dominant symbionts, other, less prominent members of these diverse communities remain poorly understood. To shed light on one abundant but under-studied community member, I investigated the distribution and phylogenetic status of the sponge symbiont SAUL (spongeassociated unclassified lineage). A meta-analysis of the available literature revealed the ubiquitous distribution of this clade and its association with taxonomically different sponge species. Additionally, the phylogeny of SAUL was revisited using both a 16S rRNA genebased phylogeny and a concatenated set of single-copy marker genes. Phylogenetic analyses revealed the monophyletic nature of this clade and, consequently, I suggest its status as a novel putative candidate phylum. To provide the first information on the putative function of SAUL clade members, I conducted a comprehensive analysis of two draft genomes assembled from sponge metagenome data, revealing novel insights into the physiology of this ubiquitous symbiont. This included the identification of genes encoding several symbiosis factors such as eukaryotic-like repeats (involved in symbiont recognition) and the presence of a CRISPR-Cas defense system, as well as the genomic capability of secondary metabolite production. This thesis represents the first systematic evaluation of the widely applied core microbiota approach, and highlights the importance of testing data sensitivity before its implementation. Moreover, the phylogenetic and genomic analyses of the SAUL lineage conducted here have contributed to expand the knowledge of less prominent and poorly understood spongeassociated microorganisms.