Sponges, one of the most ancient metazoan phyla, are an excellent model system to
investigate the origin of animal-bacteria interactions. Their associated microbiome is very
complex (up to 40 different phyla of bacteria and archaea in a single sponge specimen), yet it
is sponge-specific. The main challenge, studying this system, is that symbionts cannot be
cultured, and sponges hardly survive aquaria conditions. The advent of -omics and
bioinformatics has provided new approaches to gain insights on the molecular mechanisms
underlying sponge-bacteria interactions. We use these approaches to investigate mechanisms
of recognition between these ancient hosts and their microbial symbionts. I will discuss how
metagenomics and genome-binning of yet-unculturable sponge symbiont genomes coupled
with comparative genomics enabled us to reveal functions related to a free-living versus a
symbiotic lifestyle. From the host-perspective, we are analyzing host gene expression
(transcriptomics) in response to symbiont acquisition/loss, using in situ manipulation of the
sponge-cyanobacteria interaction.