Sponges have a great ecological and economic value, in part due to their three-dimensional complex structure, which supports rich metazoans communities, by providing micro-habitats and ideal shelters (Gerovasileiou et al., 2016). A major component of these communities are polychaetes (phylum: Annelida), often the dominant group in terms of abundance, richness, or biomass (Pansini, 1970; Rützler, 1976; Amoureux et al., 1980; Pansini and Daglio, 1980; Alos et al., 1981; Peattie and Hoare, 1981; Westinga and Hoetjes, 1981; Dauer, 1984; Koukouras et al., 1985, 1992, 1996; Voultsiadou-Koukoura et al., 1987; Cinar and Ergen, 1998; Çinar et al., 2019). Sponge-associated fauna, and specifically sponge-associated polychaetes, have been studied extensively around the world, from the tropics to the poles (Klitgaard, 1995; Ribeiro et al., 2003; Abdo, 2007; Fiore and Jutte, 2010; Schejter et al., 2012; Kersken et al., 2014; Gerovasileiou et al., 2016). Among polychaetes, only a few species may be considered truly obligatory to sponges, presenting adaptations to life within a sponge, e.g., parasitic forms such as Dorvillea sociabilis, Branchiosyllis exilis, Branchiosyllis oculata, Haplosyllis spongicola, Exogone spp., or commensal forms such as Potamilla symbiotica, Hydroides spongicola, and Sphinther arcticus (Martin and Britayev, 1998). However, in most cases the nature of the relationships between the polychaetes and their sponge hosts remains unknown. Moreover, most studies agree that polychaetes represent an opportunistic part of the sponge-inhabiting assemblage, without any selection of species tightly associated with sponges (Long, 1968; Frith, 1976; Alos et al., 1981; Koukouras et al., 1985, 1996; Gherardi et al., 2001). The main reason for the latter statement is due to the fact that a large number of sponge-associate polychaete species are known to be common on other substrates, therefore, their composition is related to the polychaete fauna of the surrounding benthos, rather than to the sponge itself (Long, 1968; Alos et al., 1981; Peattie and Hoare, 1981; Koukouras et al., 1985, 1996). By contrast, Westinga and Hoetjes (1981) and Koukouras et al. (1985) have stated that sponge-associated fauna constitutes a real “ecological community” characterized by a constant composition.

Based on the above statement and on our own observations, we hypothesize that sponge-associated polychaetes communities are not a random assembly. The current study examined this apparent discrepancy and aimed to shed light on polychaetes affinity to (massive) sponges. Massive sponges are defined in the “Thesaurus of sponge morphology” (Boury-Esnault and Rützler, 1997) as large compact sponges without definable shape. Most of the studies investigating sponge-inhabitants deals with massive sponges, as they usually contain large oscula and inner spaces, that can support greater diversity of organisms. By using data mined from the literature and applying Clarke and Warwick’s randomization test (1998), we wished to determine the extent to which sponge-associated polychaete species lists represent random subset from the larger benthic polychaete species pool (Clarke and Warwick, 1998; Warwick et al., 2002; Somerfield et al., 2009). We further aimed to find whether polychaetes constitute a constant and stable community inside sponges, across several zoogeographic regions.

Literature study showed that in total 210 polychaete species from 30 families were recorded (Supplementary Table 1) in 11 species of host sponges (Supplementary Table 2). The sponge-associated polychaete species represent ∼19% of the benthic polychaete fauna of the Mediterranean and about 43% of the families. The number of species reported from sponges in the Mediterranean and the number of studies considered in this research are shown in Figure 3. No clear pattern of an increase or a decrease in richness was found according to geographic gradients. The sponge-associated polychaete fauna of the Aegean was found to be the richest, but it is also the best studied area (seven studies), while the Adriatic and the Central Basin are by far the most impoverished and least studied (with one study each area). There is a strong positive correlation between the number of species in a region and the number of studies conducted in it (Pearson correlation, df = 3, Rho = 0.91, p = 0.03). A similar positive correlation was found between the number of polychaete species and the number of host sponge species studied in each region (Pearson correlation, df = 3, Rho = 0.97, p = 0.004; Figure 4).

Estimation of resemblance among the polychaete species of the five Mediterranean regions by group-average-linkage cluster analysis, revealed a similarity of 40% between the Western Mediterranean and the Aegean and Levant Seas. The first two regions had a 50% similarity between them. These results are visualized in an nMDS ordination (Figure 5).

An estimation of the resemblance among the polychaete families (not shown) of the five Mediterranean regions yielded the same pattern but with 60% similarity between the Aegean, Levant and Western Mediterranean, and 70% similarity between the Aegean Sea and the Western Mediterranean.

Five feeding guilds were characterized in the list of the sponge-associated polychaetes in the Mediterranean Sea (Figure 6): carnivores (72 spp.), deposit feeders (54 spp.), omnivores (40 spp.), filter feeders (39 spp.), and herbivores (10 spp.). Carnivores were the richest group in the Western Mediterranean, the Aegean and the Levant Seas, followed by omnivores. In the Adriatic Sea, both carnivores and omnivores were the most speciose groups, and in the Central Mediterranean, omnivores were the richest group followed by carnivores. The trophic composition, did not differ between the regions. As the Adriatic Sea had only eight species, it was excluded from this analysis.

The relationship between sponges and their associated macrofaunal communities has been studied in the Mediterranean Sea more intensively than anywhere else in the world. However, this study is the first to consider sponge-associated polychaete communities on a large biogeographic scale. The sponge-associated polychaetes are usually considered to be facultative transient part of the community, whose composition is probably governed mostly by the zoogeographic region (Pavloudi et al., 2016) and nearby habitats (Voultsiadou-Koukoura et al., 1987; Çinar et al., 2002; Ávila and Ortega-Bastida, 2015). Our results, suggest that it is a specialized community, and that the species in it tend to be more closely related to each other than would be expected if species were assembled at random from the regional or Mediterranean species pool.

Only at the very local scale (study), some of the communities could be considered as a random assembly of the regional pool. These cases probably represent the effect of specific sponge species and individuals’ morphology as well as sponge internal architecture, on the diversity and community composition of the polychaetes. The sponge general morphology (Koukouras et al., 1996); diameter and volume of canals (Koukouras et al., 1996; Ávila and Ortega-Bastida, 2015); sponge volume (Cinar and Ergen, 1998; Ávila and Ortega-Bastida, 2015); the presence of an ectosome (Gherardi et al., 2001); oxygen saturation within sponge areas; and the dynamics of water flow within them (Leys et al., 2011). All of which create many potential sub-habitats within the sponge, that probably have a profound effect on the composition of their inhabitants’ communities.

Greater taxonomic similarity of the sponge-associated polychaetes shown by the low Δ+ and high Λ+, indicate that environmental conditions, rather than inter-specific interactions, are probably the dominant force in shaping their communities (Elton, 1946; Webb, 2000; Somerfield et al., 2009). This suggests that in general, habitat conditions (i.e., sponge conditions), play a greater role in shaping its inhabitants’ community than historical evolutionary processes. A decrease in Δ+ and an increase in Λ+ are considered to be a feature of degraded environments (Clarke and Warwick, 2001; Warwick et al., 2002), but since the richness of sponge-associated polychaete communities is rather high compared with polychaete communities from other habitats in the Mediterranean (Box et al., 2010; Chatzigeorgiou et al., 2017), it should not be considered as an indication of habitat degradation. Rather it indicates that the polychaete communities inhabiting sponges share similar ecological adaptations to survive in this specific habitat, i.e., they are a specialized community.

The trophic structure of the sponge-associated polychaete community, remains stable across the entire Mediterranean Sea. No meaningful differences between the ratios of feeding guilds in each region (except for the Adriatic Sea which suffered from lack of data) were found. The main guild was carnivores, as it is often the case in other areas of the world (Frith, 1976; Westinga and Hoetjes, 1981; Dauer, 1984). This finding further undermines the notion of random assembly of sponge-associated polychaete communities.

No relationship was found between the sponge-associated polychaete community composition with major zoogeographic patterns that affect the entire benthic fauna, such as the North-West to South-East gradient of decreasing diversity that was found in benthic polychaetes and other organisms in the Mediterranean, which relates to differences in salinity and temperatures (Arvanitidis et al., 2002; Voultsiadou, 2009). This finding may be linked to the stability of polychaete communities inside of sponges. However, this latter result can also be attributed to the uneven research effort across the Mediterranean Sea, as areas such as the Aegean Sea, with five studies considered in this work, had greater richness than the Adriatic or the Central Basin, with a single study in each. Both cluster analysis and nMDS support this notion, and show that regions with higher richness, were more similar to each other.

Taken together, the results of this study show the importance of sponges as ecosystem engineers in temperate marine environment as they provide shelter and habitat to diverse assemblages of invertebrates. The results further emphasize that polychaete assemblages inhabiting sponges are not a transient facultative assembly of species, but rather stable, diverse and specialized communities which are well adapted for life in this habitat. However, in order to better understand the engineering function of the hosting sponges and their relationship with their surroundings and associated fauna, further research on the role of sponge morphology and zoogeographical gradients on sponge-associated fauna is necessary, especially in the Mediterranean Sea.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

LG organized the database and wrote the first draft of the manuscript. LG and TI performed the statistical analysis. All authors contributed to the conception and design of the study, manuscript revision, and read and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.