Marine sponges host bacterial communities with important ecological and economic roles in nature and society, yet these benefits depend largely on the stability of host–symbiont interactions and their susceptibility to changing environmental conditions. Here, we investigated the temporal stability of complex host–microbe symbioses in a temperate, seasonal environment over three years, targeting sponges across a range of symbiont density (high and low microbial abundance, HMA and LMA) and host taxonomy (six orders). Symbiont profiling by terminal restriction fragment length polymorphism analysis of 16S rRNA gene sequences revealed that bacterial communities in all sponges exhibited a high degree of host specificity, low seasonal dynamics and low interannual variability: results that represent an emerging trend in the field of sponge microbiology and contrast sharply with the seasonal dynamics of free-living bacterioplankton. Further, HMA sponges hosted more diverse, even and similar symbiont communities than LMA sponges and these differences in community structure extended to core members of the microbiome. Together, these findings show clear distinctions in symbiont structure between HMA and LMA sponges while resolving notable similarities in their stability over seasonal and inter-annual scales, thus providing insight into the ecological consequences of the HMA-LMA dichotomy and the temporal stability of complex host–microbe symbioses.

INTRODUCTION

Symbiotic microorganisms inhabit nearly all forms of multicellular life and play critical roles in our understanding of animal ecology and evolution (Fraune and Bosch 2010; McFall-Ngai et al.2013). In the marine environment, sessile invertebrates harbor diverse microbial symbiont taxa that confer considerable advantages to their immobile hosts, including enhanced growth rates via supplemental autotrophic nutrition (Erwin and Thacker 2008; Freeman et al.2013) and reduced disease susceptibility when residential symbionts remain intact (Cebrian et al.2011; Fan et al.2013). Further, microbial symbionts have been implicated in the production of secondary metabolites from marine invertebrates (Flowers et al.1998; Kwan et al.2012): natural products that play important ecological (e.g. defense against predation; Pawlik 2011) and economic (e.g. pharmaceutical value; Erwin, López-Legentil and Schuhmann 2010) roles in nature and society. Such potential and realized benefits depend largely on the stability of these symbiotic partnerships and their impact on host resilience under changing environmental conditions (Webster 2007). Understanding the dynamics of host–symbiont interactions is a critical first step in assessing symbiont stability, host resilience and how the interaction of these two factors may determine larger shifts in natural marine communities in the face of growing anthropogenic disturbances.

Marine sponges represent an ideal system to study the dynamics and impact of host–symbiont interactions across broad ecological scales: they are distributed worldwide in marine environments (Van Soest et al.2012), form abundant components of past and present benthic communities (Bell et al.2013) and contribute to benthic–pelagic coupling by actively filtering large volumes of seawater (McMurray, Pawlik and Finelli 2014). Sponge also host complex consortia of prokaryotic and eukaryotic microorganisms that are generally dominated by bacterial symbionts (Simister et al.2012) and contributed greatly to overall genetic diversity within the holobiont (Li et al.2014). Within the sponge microbiome, microbial-mediated nutrient cycling pathways add to the metabolic repertoire of the host sponge, including key transformation in the carbon (e.g. carbon fixation) and nitrogen (e.g. nitrification) cycles (Hoffmann et al.2009; Freeman and Thacker 2011; Ribes et al.2012). Coupling symbiont physiological capabilities with the abundance and high filtration rates of host sponges, these holobionts may transform the chemical composition of seawater in these habitats and play important roles in marine nutrient cycles of coastal ecosystems (Jiménez and Ribes 2007).

A major conceptual development in the field of sponge microbiology has been the recognition of high microbial abundance (HMA) and low microbial abundance (HMA) host sponge species and the ecological implications of the HMA-LMA dichotomy (Gloeckner et al.2014). Early observations of sponge-associated bacteria revealed two distinct categories of host sponges: those that host a remarkably high biomass of bacteria in their body (up to 35% of the total sponge volume; Vacelet 1975; Wilkinson 1978) and those with few or no bacterial cells in their tissues (Reiswig 1974, 1981; Vacelet and Donadey 1977). These two categories were originally termed bacteriosponges and non-bacteriosponges and later studies changed the nomenclature to HMA and LMA sponges, more accurately reflecting the diversity of archaeal and eukaryotic microbes also present within sponge tissues (Hentschel et al.2003). In addition to the namesake differences in microbial symbiont abundance, distinctions between HMA and LMA sponges extend to symbiont composition (Blanquer, Uriz and Galand 2013), diversity and specificity (Weisz et al.2007; Kamke, Taylor and Schmitt 2010; Moitinho-Silva et al.2014), as well as host sponge morphology and physiology (Vacelet and Donadey 1977; Weisz, Lindquist and Martens 2008; Schläppy et al.2010; Ribes et al.2012; Poppell et al.2014).

A growing body of research suggests that the sponge microbiome, for both LMA and HMA species, exhibits a high degree of host specificity and temporal stability. The sponge microbiome is comprised of unique bacterial and archaeal lineages, including novel candidate phyla (e.g. Poribacteria; Fieseler et al.2004), that are absent (‘sponge-specific’ sensu Hentschel et al.2002) or rare (‘sponge-enriched’ sensu Moitinho-Silva et al.2014) in free-living communities (Simister et al.2012; Taylor et al.2013). At the community level, these symbionts exhibit a high degree of host specificity (e.g. Easson and Thacker 2014), even among congeneric and sympatric host sponges (Erwin et al.2012a; Hardoim et al.2012). These host-specificity patterns are generally maintained over time and across seasons, as investigations of temporal variability in sponge-associated microorganisms report high symbiont stability (Taylor et al.2004; Thiel et al.2007; Anderson, Northcote and Page 2010; Erwin et al.2012b; Björk et al.2013; Simister et al.2013; Hardoim and Costa 2014), although some exceptions exist (Wichels et al.2006; Cao et al.2012; White et al.2012). As a diverse metazoan phylum consisting of over 8500 species (Van Soest et al.2012), a greater breadth of host sponge coverage is required to assess general trends in the stability of sponge-associated microbial communities. In this study, a replicated, hierarchical sampling design was applied to six taxonomically and phylogenetically diverse host sponge species, representing three HMA and three LMA sponges, sampled over a three-year time period. The targeted species represent all three major phylogenetic lineages of Demospongiae (Morrow and Cárdenas 2015): Keratosa (Dysidea avara; Redmond et al.2013), Verongimorpha (Chondrosia reniformis; Redmond et al.2013) and Heteroscleromorpha (Agelas oroides, Axinella damicornis, Petrosia ficiformis, Spriastrella cunctatrix; Riesgo et al.2014). To accommodate this broad, replicated study design, we employed the high-throughput DNA fingerprinting technique terminal restriction fragment length polymorphism (T-RFLP) analysis (Liu et al.1997), a high-resolution tool for profiling complex microbial communities in nature (Van Dorst et al.2014), including the sponge microbiome (Erwin, Olson and Thacker 2011; Lee et al.2011).

The Mediterranean Sea represents an ideal system to study the temporal dynamics of sponge–microbe associations. Sponges comprise an important ecological component of infralittoral benthic communities (Ballesteros 2006) and often dominate low irradiance habitats in coastal Mediterranean ecosystems (e.g. caves, vertical walls). Further, the Mediterranean Sea is subject to seasonal fluctuations in abiotic factors (e.g. temperature, irradiance) that directly and indirectly dictate the structure of free-living microorganisms (Pinhassi et al.2006; Alonso-Sáez et al.2007), providing a comparative framework for assessing the stability of host-associated microbiomes. Previous work has shown that Mediterranean sponges harbor diverse microbial symbiont communities (Erwin et al.2012a; Blanquer, Uriz and Galand 2013), that these symbionts can contribute to local nutrient cycling (Jiménez and Ribes 2007) and that sponge–microbe interactions are largely stable across seasons (Erwin et al.2012b; Björk et al.2013). Here, we extend our knowledge of temporal stability in Mediterranean sponges by expanding host sponge coverage and sampling design to examine the structure, dynamics and core microbial communities of six phylogenetically diverse HMA and LMA sponge species across both seasonal and annual scales.

METHODS

Sample collection

The HMA sponge species Agelas oroides (Schmidt, 1864), Chondrosia reniformisNardo, 1847 and Petrosia (Petrosia) ficiformis (Poiret, 1789) and the LMA sponge species Axinella damicornis (Esper, 1794), Dysidea avara (Schmidt, 1862) and Spirastrella cunctatrixSchmidt, 1868 were collected close to the Medes Islands marine reserve in the NW Mediterranean Sea (42º30 N, 3º130 E) by SCUBA at depths between 5 and 10 m. Representing common Mediterranean sponges with distinct morphologies, these species were identified based on their characteristic gross morphological features. The exception was S. cunctatrix, an encrusting red-orange sponge with superficial similarity to the sympatric species Crambe crambe, where micromorphological features (spicule preparations) were used to confirm identification by gross morphology (Rützler 2002). The HMA or LMA status of each species was taken from the literature (Vacelet and Donadey 1977; Uriz, Martin and Rosell 1992; Ribes et al.2012). Separate (i.e. different sponges each sampling time) replicate individuals (n = 4) from each species and ambient seawater samples (n = 2) were collected quarterly (February, May, August and November) for three consecutive years beginning in May 2009. For S. cunctatrix, four sampling months yielded insufficient replicates and individuals collected during the previous month (October 2011, January 2012) or following month (March 2010, March 2011) were used. Specimens were sublethally sampled and excised fragments collected in separate plastic bottles, brought to the surface and carefully transferred to separate 2 L jars containing filtered seawater (0.22-μm filter) to remove food microbes or loosely associated microbes from the sponges. Samples of the ambient water were taken at 5 m depth in two separate 5 L jars. All jars were transported in an insulated cooler to the laboratory (<2 h) where small pieces (ca. 2 mm2) of endosome tissue were dissected with sterilized scalpels and frozen in liquid nitrogen, and then stored at −80ºC until DNA extraction. Aliquots of seawater (300–500 mL each, 1 aliquot per sample jar) were concentrated on 0.2-μm polycarbonate filters, submerged in lysis buffer and stored at −80ºC until DNA extraction.

DNA extraction and PCR amplification

DNA extracts were prepared from sponge samples (n = 288) and seawater filters (n = 24) using the DNeasy Blood & Tissue kit (Qiagen), following the manufacturers Animal Tissue protocol. Full strength and 1:10 dilutions of DNA extracts were used as templates for PCR amplification of 16S rRNA gene sequences (ca. 1500 bp) with the universal bacterial forward primer 8F (Turner et al.1999), with a 5-end 6-carboxyfluorescein label attached to the 5 end, and reverse primer 1509R (Martínez-Murcia, Acinas and Rodriguez-Valera 1995), as described previously (Erwin et al.2012b).

T-RFLP analysis

Triplicate PCR products were gel purified and cleaned using the QIAquick Gel Extraction kit (Qiagen). Purified PCR products (ca. 50 ng) were digested separately with the restriction endonucleases HaeIII and MspI (Promega) for 8 h at 37ºC. Following incubation, samples were ethanol precipitated to remove excess salts from enzyme buffers, eluted in 10 μl highly deionized formamide with 0.5 μl GeneScan 600-LIZ size standard (Applied Biosystems). Samples were heated for 2 min at 94ºC, cooled on ice and analyzed by capillary electrophoresis on an automated sequencer (ABI 3730 Genetic Analyzer; Applied Biosystems) at the Scientific and Technical Services of the University of Barcelona. The lengths of individual terminal restriction fragments (T-RFs or peaks) were determined by comparison with the internal size standards using the program PeakScanner (Applied Biosystems) and T-RFs beyond the resolution of the size standard (50–600 bp) were removed to avoid inaccurate sizing calculations. To discriminate between ‘true’ T-RFs and background signal, T-RFs with peak areas less than 50 fluorescence units were discarded and the objective filtering algorithm of Abdo et al. (2006) was applied to the remaining datasets using TREX (cutoff value = 2 standard deviations; Culman et al.2009). The remaining T-RFs were aligned across samples using a 1-bp clustering threshold and standardized to relative abundance (percentage of total fluorescence) within each sample.

Statistical analyses of T-RFLP data

Bacterial community similarity was compared among samples using Bray–Curtis similarity matrices constructed from square-root transformations of relative T-RF abundance data and visualized in non-metric multidimensional scaling (nMDS) and cluster plots. Permutational multivariate analyses of variance (PERMANOVA) were used to determine significant differences in bacterial community structure across three factors: source (sponge species or seawater), season within source (nested analysis) and year within source (nested analysis). Estimated components of variation (CV) were calculated to determine the variability among bacterial communities attributable to each factor. Significant PERMANOVA outcomes were followed with multiple pairwise comparisons among levels of each factor, with P-values corrected based on the Benjamini–Yekutieli (B-Y) false discovery rate control (Benjamini and Yekutieli 2001), and visualized in heat maps. nMDS plots and PERMANOVA calculations were performed using Primer v6 and PERMANOVA+ (Plymouth Marine Laboratory, UK).

Bacterial community structure was compared among sponge and seawater sources using common ecological metrics for richness (number of T-RFs) and evenness (relative abundance of T-RFs) calculated for each sample. Species richness (S) calculates the total number of unique T-RFs, while Simpson's evenness (E1/D) provides a measure independent of richness and ranging from 0 to 1. Simpson's inverse index (1/D) is a reciprocal version of Simpson's heterogeneity index (D = ∑pi2, where pi is the proportion of individuals in species i) that incorporates both dimensions of diversity (richness and evenness) into the calculation. The Berger Parker index (d) is a simple dominance metric that calculates the relative abundance of the most dominant taxon (i.e. max pi). One-way analyses of variance (ANOVA) were performed on ranked values (Kruskal–Wallis), due to failed normality of the raw data (Shapiro–Wilk, P < 0.05). Pairwise multiple comparisons were conducted using Dunn's method, due to uneven sample sizes among groups (sponges vs. seawater). Statistical analyses were performed using the software Sigmaplot v11.0.

Determination of core bacterial communities

Core bacterial communities were strictly defined as bacterial T-RFs that were present in all samples (n = 48) within a species across all seasons and years. Rare members of core communities were identified as those exhibiting <1% average relative abundance across samples within a species (Pita et al.2013).

RESULTS

Host-related variation in bacterial communities

T-RFLP analyses provided high-resolution profiles of bacterial communities inhabiting six sponge species and ambient seawater, recovering 213 distinct T-RFs (105–130 per source) with the HaeIII enzyme and 223 distinct T-RFs (108–143 per source) with the MspI enzyme. nMDS plots and dendrograms constructed from T-RFLP profiles consistently clustered bacterial communities by their source for both enzymes across the three-year sampling period (Figs 1 and 2). Accordingly, statistical analyses revealed a significant effect of source on bacterial community similarity (PERMANOVA, P < 0.001, Table 1), including significant pairwise comparisons among all sponge species and seawater (P < 0.002, Table S1, Supporting Information). In fact, half of the observed variation among bacterial communities was explained by the source of each sample (sponge species or seawater, Table 1).

Figure 1.

nMDS plots of bacterial community similarity from three HMA sponges (black symbols), three LMA sponges (gray symbols) and ambient seawater (white symbols) over the three-year sampling period. nMDS ordination based on Bray–Curtis similarity of T-RFLP profiles for HaeIII (A, B) and MspI (C, D) data sets, with stress values for 2D ordination shown in parentheses. Data points are coded by source with dashed circles encompassing all samples from each source (sponge species or seawater) and solid circles encompassing all HMA sponges.

Figure 1.

nMDS plots of bacterial community similarity from three HMA sponges (black symbols), three LMA sponges (gray symbols) and ambient seawater (white symbols) over the three-year sampling period. nMDS ordination based on Bray–Curtis similarity of T-RFLP profiles for HaeIII (A, B) and MspI (C, D) data sets, with stress values for 2D ordination shown in parentheses. Data points are coded by source with dashed circles encompassing all samples from each source (sponge species or seawater) and solid circles encompassing all HMA sponges.

Figure 2.

Seasonal changes in the structure of bacterioplankton communities over the three-year monitoring period, showing consistent clustering of seawater samples by season: summer (gray circles), spring (gray half-circles), fall (black half-circles) and winter (black circles). Dendrograms based on Bray–Curtis similarity values from T-RFLP profiles with HaeIII (left) and MspI (right) data sets. Sponge samples exhibited no seasonal structure and were collapsed at host-specific nodes for clarity. See full (uncollapsed) dendrogram in Fig. S1 (Supporting Information).

Figure 2.

Seasonal changes in the structure of bacterioplankton communities over the three-year monitoring period, showing consistent clustering of seawater samples by season: summer (gray circles), spring (gray half-circles), fall (black half-circles) and winter (black circles). Dendrograms based on Bray–Curtis similarity values from T-RFLP profiles with HaeIII (left) and MspI (right) data sets. Sponge samples exhibited no seasonal structure and were collapsed at host-specific nodes for clarity. See full (uncollapsed) dendrogram in Fig. S1 (Supporting Information).

Table 1.

Permutational statistical analyses of variation in T-RFLP profiles by source (sponge species or seawater), season (month) and yeara.

  HaeIII MspI 
Factor df Pseudo-F P (perm) CV (%) Pseudo-F P (perm) CV (%) 
Source 28.127 0.001 49.94 24.13 0.001 48.43 
Month (source) 21 2.2534 0.001 9.98 1.7473 0.001 8.50 
Year (source) 14 2.3684 0.001 9.03 2.4486 0.001 10.25 
Month (source) × year (source) 42 2.3845 0.001 11.77 3.0113 0.001 13.91 
Residual 228 – – 19.28 – – 18.91 
  HaeIII MspI 
Factor df Pseudo-F P (perm) CV (%) Pseudo-F P (perm) CV (%) 
Source 28.127 0.001 49.94 24.13 0.001 48.43 
Month (source) 21 2.2534 0.001 9.98 1.7473 0.001 8.50 
Year (source) 14 2.3684 0.001 9.03 2.4486 0.001 10.25 
Month (source) × year (source) 42 2.3845 0.001 11.77 3.0113 0.001 13.91 
Residual 228 – – 19.28 – – 18.91 

adf, degrees of freedom; Pseudo-F, multivariate analog of Fisher's F statistic (i.e. ratio of variance); P (perm), P-values based on 999 permutations; CV, component of variation (i.e. percentage of variation among samples explained by each factor level).

Table 1.

Permutational statistical analyses of variation in T-RFLP profiles by source (sponge species or seawater), season (month) and yeara.

  HaeIII MspI 
Factor df Pseudo-F P (perm) CV (%) Pseudo-F P (perm) CV (%) 
Source 28.127 0.001 49.94 24.13 0.001 48.43 
Month (source) 21 2.2534 0.001 9.98 1.7473 0.001 8.50 
Year (source) 14 2.3684 0.001 9.03 2.4486 0.001 10.25 
Month (source) × year (source) 42 2.3845 0.001 11.77 3.0113 0.001 13.91 
Residual 228 – – 19.28 – – 18.91 
  HaeIII MspI 
Factor df Pseudo-F P (perm) CV (%) Pseudo-F P (perm) CV (%) 
Source 28.127 0.001 49.94 24.13 0.001 48.43 
Month (source) 21 2.2534 0.001 9.98 1.7473 0.001 8.50 
Year (source) 14 2.3684 0.001 9.03 2.4486 0.001 10.25 
Month (source) × year (source) 42 2.3845 0.001 11.77 3.0113 0.001 13.91 
Residual 228 – – 19.28 – – 18.91 

adf, degrees of freedom; Pseudo-F, multivariate analog of Fisher's F statistic (i.e. ratio of variance); P (perm), P-values based on 999 permutations; CV, component of variation (i.e. percentage of variation among samples explained by each factor level).

While each sponge species hosted a unique bacterial symbiont community, relationships among sponge hosts revealed different specificity patterns between HMA and LMA sponge species. HMA species formed a distinct group based on bacterial community similarity in nMDS (Fig. 1) and cluster plots (Fig. 2), clearly separated from LMA sponge and seawater communities. Within the HMA species cluster, C. reniformis and P. ficiformis consistently hosted more similar bacterial communities compared to A. oroides (Figs 1 and 2). In contrast, the bacterial communities in LMA hosts did not cluster together based on similarity. Two LMA species (D. avara and S. cunctatrix) hosted bacterial communities more similar to seawater than to the third LMA host (A. damicornis) or any of the HMA sponges (Figs 1 and 2), with the microbiome of D. avara exhibiting the greatest similarity to seawater bacterioplankton.

Bacterial community structure

The bacterial symbiont communities present in HMA and LMA sponge hosts were also differentiated based on common ecological diversity metrics. HMA sponges hosted significantly more species rich, diverse and even symbiont communities compared to LMA sponges (ANOVA, P < 0.05). Notably, these symbiont community trends were consistent across enzyme datasets and among all pairwise comparisons by host species (Dunn's multiple comparisons tests, P < 0.05; Fig. 3), indicating that a few, dominant symbionts occur in LMA hosts and many similarly abundant symbionts occur in HMA hosts. For example, the three most abundant bacterial taxa in LMA sponges accounted for over 50% of the total symbiont community, while the three most abundant bacterial taxa in HMA sponges accounted for only 22–27% of the total symbiont community.

Figure 3.

Diversity metrics comparing the richness and evenness of bacterial communities in three HMA sponges (black bars), three LMA sponges (gray bars) and ambient seawater (white bars). Different letters above bars indicate significantly different means among sources (P < 0.05) and error bars represent ±1 standard deviation.

Figure 3.

Diversity metrics comparing the richness and evenness of bacterial communities in three HMA sponges (black bars), three LMA sponges (gray bars) and ambient seawater (white bars). Different letters above bars indicate significantly different means among sources (P < 0.05) and error bars represent ±1 standard deviation.

Seasonal and annual variation in bacterial communities

Some variability was evident across temporal scales, as statistical analyses revealed a significant effect of season, year and an interaction term (season × year) on bacterial community similarity (PERMANOVA, P < 0.001, Table 1). Together, these two factors and their interaction term accounted nearly one third of the observed variation among bacterial communities in sponges and seawater (Table 1). Pairwise comparisons within each sponge host were used to further investigate these statistical outcomes and revealed some significant differences in bacterial community similarity across sampling times (Fig. 4). However, these differences were not consistent across years and were often detected with a single restriction enzyme dataset. Further, no consistent clustering of bacterial communities by season or year was observed in similarity dendrograms for any sponge host (Fig. S1, Supporting Information). In contrast, seawater bacteria exhibited clear seasonal patterns in community similarity, with samples from spring and summer months and from fall and winter months forming distinct clusters (Figs 2 and S1, Supporting Information) and significant differences observed among months for nearly all pairwise comparisons (Fig. 4). Even within the spring/summer and fall/winter seawater bacteria groups, samples clustered consistently by month across all three years, with the exception of spring and summer samples during the third year (Fig. 2).

Figure 4.

Heatmap summary of temporal variation in bacterial communities of six sponge species and ambient seawater, highlighting the stability of sponge-associated communities and high variability in ambient bacterioplankton. Results of pairwise comparisons of community structure (PERMANOVA) are shown as black (significant for both enzyme datasets), gray (significant for one dataset) and white (non-significant) boxes following correction based on the B-Y false discovery rate control. Exact P-values are provided in Table S2 (Supporting Information).

Figure 4.

Heatmap summary of temporal variation in bacterial communities of six sponge species and ambient seawater, highlighting the stability of sponge-associated communities and high variability in ambient bacterioplankton. Results of pairwise comparisons of community structure (PERMANOVA) are shown as black (significant for both enzyme datasets), gray (significant for one dataset) and white (non-significant) boxes following correction based on the B-Y false discovery rate control. Exact P-values are provided in Table S2 (Supporting Information).

Compared to the seasonal structure observed in ambient seawater bacteria, sponge-associated bacterial communities exhibited a high degree of seasonal stability. In particular, HMA sponge hosts showed high bacterial community similarity (73.6–83.1%) across all seasons and years, with no significant differences detected across months for A. oroides and P. ficiformis and no consistent differences detected for C. reniformis (i.e. both enzyme datasets; Fig. 4, Table S2, Supporting Information). Lower bacterial community similarity was observed in LMA sponges (48.4–64.7%) across the sampling period and all three LMA hosts showed significant differences in some pairwise comparisons by month, although these differences were not consistently detected each year, indicating the presence/absence of transient bacterial symbionts unexplained by temporal factors included in the multivariate analyses.

Core bacterial communities and symbiont overlap

The repeated sampling of sponge hosts and ambient seawater over time also allows for the analysis of core members of bacterial symbiont communities, thereby distinguishing stable from transient bacterial groups. Consistent with diversity metrics of the entire symbiont communities, a higher number of core bacterial groups were present in HMA sponges (10–35 T-RFs) compared to LMA sponges (1–6 T-RFs). These core bacterial groups accounted for the majority of symbiont communities in each host sponge species, based on one or both enzyme datasets (Table 2). Surprisingly, several members of the core microbiota in HMA hosts were rare symbionts taxa, particularly in the hosts A. oroides and P. ficiformis. These rare symbionts were detected in all samples of each sponge host yet averaged <1% relative abundance (Table 2). None of the rare members of the core microbiota reached a relative abundance above 2.89% in any sample from any time point.

Table 2.

Core bacterial communities in six Mediterranean sponge species, showing the total number of core T-RFs (present in all samples within a species) and rare core T-RFs (average relative abundance <1%) for both restriction enzyme datasets (HaeIII and MspI). Percentages of total symbiont communities accounted for by core T-RFs (% RA) and minimum and maximum relative abundances (RA Range) for rare T-RFs are shown.

  HaeIII MspI 
  Core T-RFs Rare core T-RFs Core T-RFs Rare core T-RFs 
Host sponge Type No. % RA No. RA Range No. % RA No. RA range 
A. oroides HMA 17 66.90 0.12–2.37% 21 66.71 0.17–2.34% 
C. reniformis HMA 10 49.51 − 19 58.81 0.17–2.89% 
P. ficiformis HMA 27 78.83 0.15–1.47% 35 69.19 0.13–2.37% 
A. damicornis LMA 51.08 − 31.07 − 
D. avara LMA 47.66 − 62.25 − 
S. cunctatrix LMA 63.05 − 40.78 − 
Seawater − 13 73.83 − 15 72.32 0.15–2.88% 
  HaeIII MspI 
  Core T-RFs Rare core T-RFs Core T-RFs Rare core T-RFs 
Host sponge Type No. % RA No. RA Range No. % RA No. RA range 
A. oroides HMA 17 66.90 0.12–2.37% 21 66.71 0.17–2.34% 
C. reniformis HMA 10 49.51 − 19 58.81 0.17–2.89% 
P. ficiformis HMA 27 78.83 0.15–1.47% 35 69.19 0.13–2.37% 
A. damicornis LMA 51.08 − 31.07 − 
D. avara LMA 47.66 − 62.25 − 
S. cunctatrix LMA 63.05 − 40.78 − 
Seawater − 13 73.83 − 15 72.32 0.15–2.88% 
Table 2.

Core bacterial communities in six Mediterranean sponge species, showing the total number of core T-RFs (present in all samples within a species) and rare core T-RFs (average relative abundance <1%) for both restriction enzyme datasets (HaeIII and MspI). Percentages of total symbiont communities accounted for by core T-RFs (% RA) and minimum and maximum relative abundances (RA Range) for rare T-RFs are shown.

  HaeIII MspI 
  Core T-RFs Rare core T-RFs Core T-RFs Rare core T-RFs 
Host sponge Type No. % RA No. RA Range No. % RA No. RA range 
A. oroides HMA 17 66.90 0.12–2.37% 21 66.71 0.17–2.34% 
C. reniformis HMA 10 49.51 − 19 58.81 0.17–2.89% 
P. ficiformis HMA 27 78.83 0.15–1.47% 35 69.19 0.13–2.37% 
A. damicornis LMA 51.08 − 31.07 − 
D. avara LMA 47.66 − 62.25 − 
S. cunctatrix LMA 63.05 − 40.78 − 
Seawater − 13 73.83 − 15 72.32 0.15–2.88% 
  HaeIII MspI 
  Core T-RFs Rare core T-RFs Core T-RFs Rare core T-RFs 
Host sponge Type No. % RA No. RA Range No. % RA No. RA range 
A. oroides HMA 17 66.90 0.12–2.37% 21 66.71 0.17–2.34% 
C. reniformis HMA 10 49.51 − 19 58.81 0.17–2.89% 
P. ficiformis HMA 27 78.83 0.15–1.47% 35 69.19 0.13–2.37% 
A. damicornis LMA 51.08 − 31.07 − 
D. avara LMA 47.66 − 62.25 − 
S. cunctatrix LMA 63.05 − 40.78 − 
Seawater − 13 73.83 − 15 72.32 0.15–2.88% 

While considerable overlap was observed among sponge and seawater communities, restricting host-specificity analysis to core bacteria revealed further differences between LMA and HMA sponges (Fig. 5). Among HMA sponge hosts, several core members were shared across all three (5–6 T-RFs) or two of three (7–11 T-RFs) host species, indicating some overlap in the membership of each hosts’ symbiont community. In stark contrast, no core bacteria were shared among any of the LMA host species, indicating a higher degree of host specificity of these dominant symbiont groups (Fig. 5). It is also noteworthy that HMA hosts shared more core bacterial members with seawater (2–5 T-RFs) compared to LMA hosts (1–2 T-RFs) and that several core members were shared between LMA and HMA hosts (4–6 T-RFs).

Figure 5.

Venn diagrams depicting the host specificity of total (top row) and core (bottom row) bacterial communities in three HMA sponges, three LMA sponges and ambient seawater. Circles represent bacterial communities by group (LMA or HMA) or host species, with overlapping portions indicating the number of shared bacterial T-RFs and non-overlapping portions the number of host-specific T-RFs (top values = HaeIII, bottom values = MspI).

Figure 5.

Venn diagrams depicting the host specificity of total (top row) and core (bottom row) bacterial communities in three HMA sponges, three LMA sponges and ambient seawater. Circles represent bacterial communities by group (LMA or HMA) or host species, with overlapping portions indicating the number of shared bacterial T-RFs and non-overlapping portions the number of host-specific T-RFs (top values = HaeIII, bottom values = MspI).

DISCUSSION

In this study, we assessed the temporal stability of complex host–microbe associations in six taxonomically diverse Mediterranean sponges, revealing low seasonal and interannual dynamics in bacterial communities of both HMA and LMA hosts. This remarkable temporal stability contrasted sharply with the clear seasonal fluctuations in free-living bacterioplankton present in ambient seawater surrounding the investigated sponges. The microbiomes of HMA sponges were also clearly differentiated from LMA sponges based on community structure and diversity metrics, with HMA sponges hosting more diverse, even and similar symbiont communities than LMA sponges. These differences extend to core members of the microbiome (i.e. present in all samples of a sponge species), which were more diverse and exhibited greater overlap between HMA sponge species than LMA sponges species. In fact, no core symbiont taxa were shared between LMA sponge species. Together, these findings highlight clear distinctions in the structure of symbiont communities between HMA and LMA sponges while resolving notable similarities in their temporal stability over seasonal and interannual scales.

Source was the main driver of symbiont structure across all investigated sponge species and seawater, accounting for over half of the variation observed among microbial communities. The role of the host in dictating symbiont structure is emerging as a paradigm in sponge microbiology, reported in numerous LMA and HMA host species collected from diverse geographic regions (Blanquer, Uriz and Galand 2013; Webster et al.2013; Easson and Thacker 2014). More surprising, we observed high similarity (>40%) between microbial communities of HMA hosts. Several studies have shown that HMA sponges are separate from LMA sponges based on symbiont community similarity (Gerçe et al.2011; Blanquer, Uriz and Galand 2013; Schöttner et al.2013; Poppell et al.2014), though generally clustering at low similarity values (<10%). Clustering of HMA sponges is a pattern that likely reflects a higher percentage of shared, generalist symbiont taxa in HMA hosts compared to LMA hosts (Kamke, Taylor and Schmitt 2010; Erwin, Olson and Thacker 2011; Schmitt et al.2011; Giles et al.2013; Simister et al.2013). The comparison of core symbiont communities herein provides a clear illustration of this pattern, with several shared core taxa among HMA hosts and no shared core taxa among LMA hosts.

Our study also yielded consistent differences in symbiont richness and evenness that underlie these overall patterns of community structure. HMA sponges hosted symbiont communities with greater richness and evenness compared to LMA hosts, a result observed for both total and core symbiont communities. Indeed, the microbiomes of LMA sponges are typically dominated by a few specialist (i.e. host-specific) symbiont taxa (Kamke, Taylor and Schmitt 2010; Erwin, Olson and Thacker 2011; Giles et al.2013). However, exceptions to these general trends have been reported (Schöttner et al.2013; Easson and Thacker 2014) and potentially ascribed to specific habitats (cold water) or host characteristics (phototrophic sponges). Given the broad taxonomic and ecological diversity of species in the phylum Porifera (Van Soest et al.2012), such exceptions are not unexpected and the HMA-LMA dichotomy remains a useful framework for comparative analyses in sponge microbiology (Gloeckner et al.2014).

The host specificity and diversity patterns of symbiont communities in HMA and LMA sponge hosts were maintained across seasons and years, despite the large temporal fluctuations in abiotic and biotic factors that characterize the Mediterranean Sea. The temporal stability of the sponge microbiome, even in seasonal environments, is also emerging as a common theme in the field. To date, at least 22 sponge species have been investigated for in situ symbiont stability over time, representing 11 HMA and 11 LMA hosts from 15 families, 11 orders and all three major subclasses (Morrow and Cárdenas 2015) of Demospongiae (Table 3). High temporal stability was reported for most (82%) sponge hosts, including the six species herein, indicating high stability of symbiont communities in both HMA and LMA sponges from diverse taxonomic groups. Further, these studies were conducted in different geographic locations (South Pacific Ocean, Western Atlantic Ocean, Mediterranean Sea) using different sampling schemes (monthly, quarterly, yearly) and symbiont characterization methodologies (multiple DNA fingerprinting techniques, first- and second-generation DNA sequencing). The consistency of these diverse studies suggests an overall high degree of host–symbiont fidelity in the complex marine sponge–bacteria symbioses.

Table 3.

Taxonomic diversitya and microbial abundance category (LMA vs. HMA) of sponge species used in this studyb and previous investigations of temporal variation in sponge-associated microbial communitiesc.

Subclass Order Family Species Category Citation 
Keratosa Dictyoceratida Dysideidae Dysidea avara LMA This study, (Björk et al.2013
  Irciniidae Ircinia fasciculata HMA (Erwin et al.2012b
   Ircinia oros HMA (Erwin et al.2012b
   Ircinia variabilis HMA (Erwin et al.2012b
   Sarcotragus spinosulus HMA (Hardoim and Costa 2014
  Spongiidae Spongia officinalis HMA (Bauvais et al.2015
Verongimorpha Chondrosida Chondrillidae Chondrilla nucula HMA (Thiel et al.2007
  Chondrosidae Chondrosia reniformis HMA This study, (Björk et al.2013
 Verongida Aplysinidae Aplysina cauliformis HMA (Olson, Thacker and Gochfeld 2013
Heteroscleromorpha Agelasida Agelasidae Agelas oroides HMA This study, (Björk et al.2013
 Axinellida Axinellidae Axinella corrugata LMA (White et al.2012
   Axinella damicornis LMA This study 
   Cymbastela concentrica LMA (Taylor et al.2004
 Clionaida Spirastrellidae Spirastrella cunctatrix LMA This study 
 Haplosclerida Callyspongidae Callyspongia sp. LMA (Taylor et al.2004
  Petrosiidae Petrosia ficiformis HMA This study 
 Poecilosclerida Mycalidae Mycale hentscheli LMA (Anderson et al.2010
 Subertida Halichondridae Halichondria panicea LMA (Wichels et al.2006
   Hymeniacidon sinapium LMA (Cao et al.2012
   Stylinos sp. LMA (Taylor et al.2004
 Tethyida Tethyidae Tethya stolonifera LMA (Simister et al.2013
 Tetractinellida Ancorinidae Ancorina alata HMA (Simister et al.2013
Subclass Order Family Species Category Citation 
Keratosa Dictyoceratida Dysideidae Dysidea avara LMA This study, (Björk et al.2013
  Irciniidae Ircinia fasciculata HMA (Erwin et al.2012b
   Ircinia oros HMA (Erwin et al.2012b
   Ircinia variabilis HMA (Erwin et al.2012b
   Sarcotragus spinosulus HMA (Hardoim and Costa 2014
  Spongiidae Spongia officinalis HMA (Bauvais et al.2015
Verongimorpha Chondrosida Chondrillidae Chondrilla nucula HMA (Thiel et al.2007
  Chondrosidae Chondrosia reniformis HMA This study, (Björk et al.2013
 Verongida Aplysinidae Aplysina cauliformis HMA (Olson, Thacker and Gochfeld 2013
Heteroscleromorpha Agelasida Agelasidae Agelas oroides HMA This study, (Björk et al.2013
 Axinellida Axinellidae Axinella corrugata LMA (White et al.2012
   Axinella damicornis LMA This study 
   Cymbastela concentrica LMA (Taylor et al.2004
 Clionaida Spirastrellidae Spirastrella cunctatrix LMA This study 
 Haplosclerida Callyspongidae Callyspongia sp. LMA (Taylor et al.2004
  Petrosiidae Petrosia ficiformis HMA This study 
 Poecilosclerida Mycalidae Mycale hentscheli LMA (Anderson et al.2010
 Subertida Halichondridae Halichondria panicea LMA (Wichels et al.2006
   Hymeniacidon sinapium LMA (Cao et al.2012
   Stylinos sp. LMA (Taylor et al.2004
 Tethyida Tethyidae Tethya stolonifera LMA (Simister et al.2013
 Tetractinellida Ancorinidae Ancorina alata HMA (Simister et al.2013

aClassifications based on the recent, revised taxonomic framework for Demospongiae (Morrow and Cárdenas 2015).

bSpecies targeted in the current study are shown in bold.

cReferences cited refer to studies examining symbiont community variability in situ using molecular (culture-independent) techniques. Additional temporal studies have focused on sponges maintained ex situ (Friedrich et al.2001; Mohamed et al.2008a,b; Webster et al.2011) and used culture-dependent techniques (Webster and Hill 2001).

Table 3.

Taxonomic diversitya and microbial abundance category (LMA vs. HMA) of sponge species used in this studyb and previous investigations of temporal variation in sponge-associated microbial communitiesc.

Subclass Order Family Species Category Citation 
Keratosa Dictyoceratida Dysideidae Dysidea avara LMA This study, (Björk et al.2013
  Irciniidae Ircinia fasciculata HMA (Erwin et al.2012b
   Ircinia oros HMA (Erwin et al.2012b
   Ircinia variabilis HMA (Erwin et al.2012b
   Sarcotragus spinosulus HMA (Hardoim and Costa 2014
  Spongiidae Spongia officinalis HMA (Bauvais et al.2015
Verongimorpha Chondrosida Chondrillidae Chondrilla nucula HMA (Thiel et al.2007
  Chondrosidae Chondrosia reniformis HMA This study, (Björk et al.2013
 Verongida Aplysinidae Aplysina cauliformis HMA (Olson, Thacker and Gochfeld 2013
Heteroscleromorpha Agelasida Agelasidae Agelas oroides HMA This study, (Björk et al.2013
 Axinellida Axinellidae Axinella corrugata LMA (White et al.2012
   Axinella damicornis LMA This study 
   Cymbastela concentrica LMA (Taylor et al.2004
 Clionaida Spirastrellidae Spirastrella cunctatrix LMA This study 
 Haplosclerida Callyspongidae Callyspongia sp. LMA (Taylor et al.2004
  Petrosiidae Petrosia ficiformis HMA This study 
 Poecilosclerida Mycalidae Mycale hentscheli LMA (Anderson et al.2010
 Subertida Halichondridae Halichondria panicea LMA (Wichels et al.2006
   Hymeniacidon sinapium LMA (Cao et al.2012
   Stylinos sp. LMA (Taylor et al.2004
 Tethyida Tethyidae Tethya stolonifera LMA (Simister et al.2013
 Tetractinellida Ancorinidae Ancorina alata HMA (Simister et al.2013
Subclass Order Family Species Category Citation 
Keratosa Dictyoceratida Dysideidae Dysidea avara LMA This study, (Björk et al.2013
  Irciniidae Ircinia fasciculata HMA (Erwin et al.2012b
   Ircinia oros HMA (Erwin et al.2012b
   Ircinia variabilis HMA (Erwin et al.2012b
   Sarcotragus spinosulus HMA (Hardoim and Costa 2014
  Spongiidae Spongia officinalis HMA (Bauvais et al.2015
Verongimorpha Chondrosida Chondrillidae Chondrilla nucula HMA (Thiel et al.2007
  Chondrosidae Chondrosia reniformis HMA This study, (Björk et al.2013
 Verongida Aplysinidae Aplysina cauliformis HMA (Olson, Thacker and Gochfeld 2013
Heteroscleromorpha Agelasida Agelasidae Agelas oroides HMA This study, (Björk et al.2013
 Axinellida Axinellidae Axinella corrugata LMA (White et al.2012
   Axinella damicornis LMA This study 
   Cymbastela concentrica LMA (Taylor et al.2004
 Clionaida Spirastrellidae Spirastrella cunctatrix LMA This study 
 Haplosclerida Callyspongidae Callyspongia sp. LMA (Taylor et al.2004
  Petrosiidae Petrosia ficiformis HMA This study 
 Poecilosclerida Mycalidae Mycale hentscheli LMA (Anderson et al.2010
 Subertida Halichondridae Halichondria panicea LMA (Wichels et al.2006
   Hymeniacidon sinapium LMA (Cao et al.2012
   Stylinos sp. LMA (Taylor et al.2004
 Tethyida Tethyidae Tethya stolonifera LMA (Simister et al.2013
 Tetractinellida Ancorinidae Ancorina alata HMA (Simister et al.2013

aClassifications based on the recent, revised taxonomic framework for Demospongiae (Morrow and Cárdenas 2015).

bSpecies targeted in the current study are shown in bold.

cReferences cited refer to studies examining symbiont community variability in situ using molecular (culture-independent) techniques. Additional temporal studies have focused on sponges maintained ex situ (Friedrich et al.2001; Mohamed et al.2008a,b; Webster et al.2011) and used culture-dependent techniques (Webster and Hill 2001).

Exceptions to the general trend of high stability of the sponge microbiome include reports of temporal variation in sponge-associated microbial communities in four sponge species: Halichondria panicea (Wichels et al.2006), Hymeniacidon sinapium (Cao et al.2012), Ax. corrugata (White et al.2012) and Aplysina cauliformis (Olson, Thacker and Gochfeld 2013). The sole HMA host in this species list, A. cauliformis, also exhibited significant variability in symbiont communities across locations and disease status, yet maintained a common core of stable bacterial taxa (Olson, Thacker and Gochfeld 2013). The remaining host species represent LMA sponges and two are closely related (H. panicea, Hy. sinapium, family Halichondridae), leading to the preliminary hypothesis that some sponge lineages within the LMA category may exhibit greater flexibility in host–symbiont interactions over time. While our study reported greater symbiont variability across replicates of LMA sponges, these differences did not follow any seasonal or interannual pattern and the factors collection month and year had low explanatory power compared to host sponge species. Similarly, three of the aforementioned studies reported temporal variability, rather than clear seasonal shifts (Wichels et al.2006; White et al.2012; Olson, Thacker and Gochfeld 2013). Indeed, it is difficult to assess whether these temporal changes reflect responses of the symbiont community to periodic fluctuations in environmental factors (i.e. seasonal dynamics), due to the limited ability of the sampling designs (two time points; White et al.2012) and statistical methods (no quantitative comparisons of symbiont fingerprint data; Wichels et al.2006) to detect seasonal trends. In contrast, the Japanese sponge Hy. sinapium was shown to exhibit seasonal shifts in symbiont communities over 1.5 years in Yellow Sea (Cao et al.2012), an area characterized by drastic changes in environmental conditions throughout the year, including sedimentation rate and large temperature fluctuations (<0º to 30ºC). Thus, exceptions to the trend of high temporal stability of sponge microbiomes may occur in particular host species (e.g. A. cauliformis) or environments where extreme seasonal fluctuations may exceed the physiological tolerance levels of specific symbionts, resulting in the observed seasonal community shifts.

The investigation of microbial communities over temporal scales also allows for the detection of core microbial taxa, those that occur consistently in particular habitats. Such core taxa are hypothesized to have ecological relevance in the functioning of the overall microbial communities (Shade and Handelsman 2012) and are a particularly relevant aspect of sponge microbiomes, given that bacteria detected in the sponge body may represent permanent symbionts or transient microbes (Olson, Thacker and Gochfeld 2013), such parasites (Webster et al.2002) or food source bacterioplankton (Reiswig 1975). Accordingly, a large proportion (61–66%) of non-core (transient) bacteria were shared between sponges and seawater while most (74–78%) of the core bacteria were present only in sponges. The analysis of core communities in the six host sponges investigated herein revealed parallel patterns with analyses of total symbiont communities, including a higher diversity and greater overlap of core taxa in HMA hosts compared to LMA hosts. The comparatively low complexity of the LMA microbiome (one to six core taxa, herein) offers a tractable model for in-depth characterization of symbiont community structure and function. For example, a recent study utilized next-generation sequencing and fluorescence in situ hybridization techniques to identify and localize a single bacterial species that dominates the microbiome of the LMA sponge Crambe crambe, suggesting a putative role of this symbiont in the production of bioactive secondary metabolites isolated from the host sponge (Croué et al.2013). In contrast to the LMA hosts, the core microbiomes of HMA hosts were not only more diverse (10–35 core taxa) but included rare symbiont taxa (averaging <1% relative abundance), indicating that not all rare members of the sponge microbiome are transient symbionts. Notably these core taxa remained rare throughout the three-year monitoring period, rather than exploiting temporal windows of favorable conditions and increasing in abundance (‘conditionally’ rare) as reported in marine bacterioplankton (Alonso-Sáez, Díaz-Pérez and Morán 2015). Additional studies focusing on the rare components of the sponge microbiome are required to assess their role in overall symbiont community function, which may be disproportionate to their abundance as observed in other rare environmental taxa (Musat et al.2008; Pester et al.2010).

Monitoring the stability of the sponge microbiome over seasonal and annual scales provides baseline levels of natural variation and aids in the development of sponge–microbe symbioses as biomonitoring tools. Given the complexity of the sponge microbiome, it is necessary to define baseline shifts in symbiont communities prior to detecting abnormal shifts and disease incidence in response to changing environment conditions, such as acute anthropogenic disturbances or chronic climate change (Webster 2007). In addition to studies of symbiont structure, future studies targeting specific functional guilds of symbionts (Fan et al.2012) and their temporal fluctuations (Zhang, Vicente and Hill 2014) will provide insight into the consequences of structural shifts, the potential for functional redundancy and cascading effects of host sponge health and resilience. Currently, sponges are being developed as biomonitors of toxic metal pollutants (Cebrian, Uriz and Turon 2007; Davis et al.2014) and coupling microbiome data with toxicological analyses may prove an even more sensitive measure of sublethal stress in marine organisms (Fan et al.2013). In the face of growing anthropogenic disturbances, further development of sponges and their microbiome as models for basic ecological research and applied biomonitoring initiatives will enhance our understanding of host–symbiont dynamics and clarify the role of symbiont stability and host resilience in determining larger shifts in natural marine communities.

SUPPLEMENTARY DATA

Supplementary Data.

We are grateful to the ‘Parc Natural del Montgrí, les Illes Medes i el Baix Ter’, ‘Parc Natural del Cap de Creus’ and ‘Reservas Marinas de España, Dirección General de Recursos Pesqueros y Acuicultura, Ministerio de Agricultura, Alimentación y Medio Ambiente’ for their support to our research and sampling permissions.

FUNDING

This work was supported by the Spanish Government projects MARSYMBIOMICS [grant number CTM2013-43287-P] and CSI-Coral [grant number CGL2013-43106-R to RC and MR]. This is a contribution from the Marine Biogeochemistry and Global Change research group funded by the Catalan Government [grant number 2014SGR1029].

Conflict of interest. None declared.

REFERENCES

Abdo
Z
Schüette
UME
Bent
SJ
et al. 
Statistical methods for characterizing diversity of microbial communities by analysis of terminal restriction fragment length polymorphisms of 16S rRNA genes
Environ Microbiol
 
2006
8
929
38
Alonso-Sáez
L
Balagué
V
EL
et al. 
Seasonality in bacterial diversity in north-west Mediterranean coastal waters: assessment through clone libraries, fingerprinting and FISH
FEMS Microbiol Ecol
 
2007
60
98
112
Alonso-Sáez
L
Díaz-Pérez
L
Morán
XAG
The hidden seasonality of the rare biosphere in coastal marine bacterioplankton
Environ Microbiol
 
2015
DOI: 10.1111/1462-2920.12801
Anderson
SA
Northcote
PT
Page
MJ
Spatial and temporal variability of the bacterial community in different chemotypes of the New Zealand marine sponge Mycale hentscheli
FEMS Microbiol Ecol
 
2010
72
328
42
Ballesteros
E
Mediterranean coralligenous assemblages: a synthesis of present knowledge
Oceanogr Mar Biol
 
2006
44
123
95
Bauvais
C
Zirah
S
Piette
L
et al. 
Sponging up metals: bacteria associated with the marine sponge Spongia officinalis
Mar Environ Res
 
2015
104
20
30
Bell
JJ
Davy
SK
Jones
T
et al. 
Could some coral reefs become sponge reefs as our climate changes?
Glob Change Biol
 
2013
19
2613
24
Benjamini
Y
Yekutieli
D
The control of false discovery rate in multiple testing under dependency
Ann Stat
 
2001
29
1165
88
Björk
JR
Díez-Vives
C
Coma
R
et al. 
Specificity and temporal dynamics of complex bacteria-sponge symbiotic interactions
Ecology
 
2013
94
2781
91
Blanquer
A
Uriz
MJ
Galand
PE
Removing environmental sources of variation to gain insight on symbionts vs. transient microbes in high and low microbial abundance sponges
Environ Microbiol
 
2013
15
3008
19
Cao
H
Cao
X
Guan
X
et al. 
High temporal variability in bacterial community, silicatein and hsp70 expression during the annual life cycle of Hymeniacidon sinapium (Demospongiae) in China's Yellow Sea
Aquaculture
 
2012
358-359
262
73
Cebrian
E
Uriz
MJ
Garrabou
J
et al. 
Sponge mass mortalities in a warming Mediterranean Sea: are cyanobacteria-harboring species worse off? Thrush S (ed.)
PLoS One
 
2011
6
e20211
Cebrian
E
Uriz
M-J
Turon
X
Sponges as biomonitors of heavy metals in spatial and temporal surveys in northwestern Mediterranean: multispecies comparison
Environ Toxicol Chem
 
2007
26
2430
9
Croué
J
West
NJ
Escande
M-L
et al. 
A single betaproteobacterium dominates the microbial community of the crambescidine-containing sponge Crambe crambe
Sci Rep
 
2013
3
2583
Culman
SW
Bukowski
R
Gauch
HG
et al. 
T-REX: software for the processing and analysis of T-RFLP data
BMC Bioinformatics
 
2009
10
171
Davis
AR
de Mestre
C
Maher
W
et al. 
Sponges as sentinels: metal accumulation using transplanted sponges across a metal gradient
Environ Toxicol Chem
 
2014
33
2818
25
Easson
CG
Thacker
RW
Phylogenetic signal in the community structure of host-specific microbiomes of tropical marine sponges
Front Microbiol
 
2014
5
532
Erwin
P
Thacker
R
Phototrophic nutrition and symbiont diversity of two Caribbean sponge–cyanobacteria symbioses
Mar Ecol Prog Ser
 
2008
362
139
47
Erwin
PM
López-Legentil
S
Schuhmann
PW
The pharmaceutical value of marine biodiversity for anti-cancer drug discovery
Ecol Econ
 
2010
70
445
51
Erwin
PM
Olson
JB
Thacker
RW
Phylogenetic diversity, host-specificity and community profiling of sponge-associated bacteria in the northern Gulf of Mexico. López-García P (ed.)
PLoS One
 
2011
6
e26806
Erwin
PM
López-Legentil
S
González-Pech
R
et al. 
A specific mix of generalists: bacterial symbionts in Mediterranean Ircinia spp
FEMS Microbiol Ecol
 
2012a
79
619
37
Erwin
PM
Pita
L
Lopez-Legentil
S
et al. 
Stability of sponge-associated bacteria over large seasonal shifts in temperature and irradiance
Appl Environ Microb
 
2012b
78
7358
68
Fan
L
Liu
M
Simister
R
et al. 
Marine microbial symbiosis heats up: the phylogenetic and functional response of a sponge holobiont to thermal stress
ISME J
 
2013
7
991
1002
Fan
L
Reynolds
D
Liu
M
et al. 
Functional equivalence and evolutionary convergence in complex communities of microbial sponge symbionts
P Natl Acad Sci USA
 
2012
109
E1878
87
Fieseler
L
Horn
M
Wagner
M
et al. 
Discovery of the novel candidate phylum ‘Poribacteria’ in marine sponges
Appl Environ Microb
 
2004
70
3724
32
Flowers
AE
Garson
MJ
Webb
RI
et al. 
Cellular origin of chlorinated diketopiperazines in the dictyoceratid sponge Dysidea herbacea (Keller)
Cell Tissue Res
 
1998
292
597
607
Freeman
CJ
Thacker
RW
Complex interactions between marine sponges and their symbiotic microbial communities
Limnol Oceanogr
 
2011
56
1577
86
Freeman
CJ
Thacker
RW
Baker
DM
et al. 
Quality or quantity: is nutrient transfer driven more by symbiont identity and productivity than by symbiont abundance?
ISME J
 
2013
7
1116
25
Fraune
S
Bosch
TCG
Why bacteria matter in animal development and evolution
BioEssays
 
2010
32
571
80
Friedrich
AB
Fischer
I
Proksch
P
et al. 
Temporal variation of the microbial community associated with the Mediterranean sponge Aplysina aerophoba
FEMS Microbiol Ecol
 
2001
38
105
13
Gerçe
B
Schwartz
T
Syldatk
C
et al. 
Differences between bacterial communities associated with the surface or tissue of Mediterranean sponge species
Microb Ecol
 
2011
61
769
82
Giles
EC
Kamke
J
Moitinho-Silva
L
et al. 
Bacterial community profiles in low microbial abundance sponges
FEMS Microbiol Ecol
 
2013
83
232
41
Gloeckner
V
Wehrl
M
Moitinho-Silva
L
et al. 
The HMA-LMA dichotomy revisited: an electron microscopical survey of 56 sponge species
Biol Bull
 
2014
227
78
88
Hardoim
CCP
Costa
R
Temporal dynamics of prokaryotic communities in the marine sponge Sarcotragus spinosulus
Mol Ecol
 
2014
23
3097
112
Hardoim
CCP
Esteves
AIS
Pires
FR
et al. 
Phylogenetically and spatially close marine sponges harbour divergent bacterial communities. Harder T (ed.)
PLoS One
 
2012
7
e53029
Hentschel
U
Fieseler
L
Wehrl
M
et al. 
Microbial diversity of marine sponges
Prog Mol Subcell Biol
 
2003
37
59
88
Hentschel
U
Hopke
J
Horn
M
et al. 
Molecular evidence for a uniform microbial community in sponges from different oceans
Appl Environ Microb
 
2002
68
4431
40
Hoffmann
F
Radax
R
Woebken
D
et al. 
Complex nitrogen cycling in the sponge Geodia barretti
Environ Microbiol
 
2009
11
2228
43
Jiménez
E
Ribes
M
Sponges as a source of dissolved inorganic nitrogen: nitrification mediated by temperate sponges
Limnol Oceanogr
 
2007
52
948
58
Kamke
J
Taylor
MW
Schmitt
S
Activity profiles for marine sponge-associated bacteria obtained by 16S rRNA vs 16S rRNA gene comparisons
ISME J
 
2010
4
498
508
Kwan
JC
Donia
MS
Han
AW
et al. 
Genome streamlining and chemical defense in a coral reef symbiosis
P Natl Acad Sci USA
 
2012
109
20655
60
Lee
OO
Wang
Y
Yang
J
et al. 
Pyrosequencing reveals highly diverse and species-specific microbial communities in sponges from the Red Sea
ISME J
 
2011
5
650
64
Li
Z-Y
Wang
Y-Z
He
L-M
et al. 
Metabolic profiles of prokaryotic and eukaryotic communities in deep-sea sponge Lamellomorpha sp. indicated by metagenomics
Sci Rep
 
2014
4
3895
Liu
W-T
Marsh
TL
Cheng
H
et al. 
Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16S rRNA
Appl Environ Microb
 
1997
63
4516
22
McFall-Ngai
M
Hadfield
MG
Bosch
TC
et al. 
Animals in a bacterial world, a new imperative for the life sciences
P Natl Acad Sci USA
 
2013
110
3229
36
McMurray
S
Pawlik
J
Finelli
C
Trait-mediated ecosystem impacts: how morphology and size affect pumping rates of the Caribbean giant barrel sponge
Aquat Biol
 
2014
23
1
13
Martínez-Murcia
AJ
Acinas
SG
Rodriguez-Valera
F
Evaluation of prokaryotic diversity by restrictase digestion of 16S rDNA directly amplified from hypersaline environments
FEMS Microbiol Ecol
 
1995
17
247
55
Mohamed
NM
Enticknap
JJ
Lohr
JE
et al. 
Changes in bacterial communities of the marine sponge Mycale laxissima on transfer into aquaculture
Appl Environ Microb
 
2008a
74
1209
22
Mohamed
NM
Rao
V
Hamann
MT
et al. 
Monitoring bacterial diversity of the marine sponge Ircinia strobilina upon transfer into aquaculture
Appl Environ Microb
 
2008b
74
4133
43
Moitinho-Silva
L
Bayer
K
Cannistraci
CV
et al. 
Specificity and transcriptional activity of microbiota associated with low and high microbial abundance sponges from the Red Sea
Mol Ecol
 
2014
23
1348
63
Morrow
C
Cárdenas
P
Proposal for a revised classification of the Demospongiae (Porifera)
Front Zool
 
2015
12
7
Musat
N
Halm
H
Winterholler
B
et al. 
A single-cell view on the ecophysiology of anaerobic phototrophic bacteria
P Natl Acad Sci USA
 
2008
105
17861
6
Olson
JB
Thacker
RW
Gochfeld
DJ
Molecular community profiling reveals impact of time, space, and disease status on the bacterial community associated with the Caribbean sponge Aplysina cauliformis
FEMS Microb Ecol
 
2013
87
268
79
Pawlik
JR
The chemical ecology of sponges on Caribbean reefs: natural products shape natural systems
BioScience
 
2011
61
888
98
Pester
M
Bittner
N
Deevong
P
et al. 
A ‘rare biosphere’ microorganism contributes to sulfate reduction in a peatland
ISME J
 
2010
4
1591
602
Pinhassi
J
Gómez-Consarnau
L
Alonso-Sáez
L
et al. 
Seasonal changes in bacterioplankton nutrient limitation and their effects on bacterial community composition in the NW Mediterranean Sea
Aquat Microb Ecol
 
2006
45
241
52
Pita
L
Turon
X
López-Legentil
S
et al. 
Host rules: spatial stability of bacterial communities associated with marine sponges (Ircinia spp.) in the Western Mediterranean Sea
FEMS Microbiol Ecol
 
2013
86
268
76
Poppell
E
Weisz
J
Spicer
L
et al. 
Sponge heterotrophic capacity and bacterial community structure in high- and low-microbial abundance sponges
Mar Ecol
 
2014
35
414
24
Redmond
NE
Morrow
CC
Thacker
RW
et al. 
Phylogeny and systematics of Demospongiae in light of new small-subunit ribosomal DNA (18S) sequences
Integr Comp Biol
 
2013
53
388
415
Reiswig
HM
Water transport, respiration and energetics of three tropical marine sponges
J Exp Mar Biol Ecol
 
1974
14
231
49
Reiswig
HM
Bacteria as food for temperate-water marine sponges
Can J Zool
 
1975
53
582
9
Reiswig
HM
Partial carbon and energy budgets of the bacteriosponge Verongia fistularis (Porifera: Demospongiae) in Barbados
Mar Ecol
 
1981
2
273
93
Ribes
M
Jiménez
E
Yahel
G
et al. 
Functional convergence of microbes associated with temperate marine sponges
Environ Microbiol
 
2012
14
1224
39
Riesgo
A
Novo
M
Sharma
PP
et al. 
Inferring the ancestral sexuality and reproductive condition in sponges (Porifera)
Zool Scripta
 
2014
43
101
17
Rützler
K
Hooper
JNA
Van Soest
RWM
Family Spirastrellidae Ridley & Dendy, 1886
Systema Porifera: A Guide to the Classification of Sponges
 
2002
New York
Kluwer Academic/Plenum Publishers
220
3
Schläppy
M-L
Schöttner
SI
Lavik
G
et al. 
Evidence of nitrification and denitrification in high and low microbial abundance sponges
Mar Biol
 
2010
157
593
602
Schöttner
S
Hoffmann
F
Cárdenas
P
et al. 
Relationships between host phylogeny, host type and bacterial community diversity in cold-water coral reef sponges. Gilbert JA (ed.)
PLoS One
 
2013
8
e55505
Schmitt
S
Deines
P
Behnam
F
et al. 
Chloroflexi bacteria are more diverse, abundant, and similar in high than in low microbial abundance sponges
FEMS Microbiol Ecol
 
2011
78
497
510
Shade
A
Handelsman
J
Beyond the Venn diagram: the hunt for a core microbiome
Environ Microbiol
 
2012
14
4
12
Simister
RL
Deines
P
Botté
ES
et al. 
Sponge-specific clusters revisited: a comprehensive phylogeny of sponge-associated microorganisms
Environ Microbiol
 
2012
14
517
24
Simister
R
Taylor
MW
Rogers
KM
et al. 
Temporal molecular and isotopic analysis of active bacterial communities in two New Zealand sponges
FEMS Microbiol Ecol
 
2013
85
195
205
Taylor
MW
Tsai
P
Simister
RL
et al. 
‘Sponge-specific’ bacteria are widespread (but rare) in diverse marine environments
ISME J
 
2013
7
438
43
Taylor
MW
Schupp
PJ
Dahllöf
I
et al. 
Host specificity in marine sponge-associated bacteria, and potential implications for marine microbial diversity
Environ Microbiol
 
2004
6
121
30
Thiel
V
Leininger
S
Schmaljohann
R
et al. 
Sponge-specific bacterial associations of the Mediterranean sponge Chondrilla nucula (Demospongiae, Tetractinomorpha)
Microb Ecol
 
2007
54
101
11
Turner
S
Pryer
KM
Miao
VP
et al. 
Investigating deep phylogenetic relationships among cyanobacteria and plastids by small subunit rRNA sequence analysis
J Eukaryot Microbiol
 
1999
46
327
38
Uriz
MJ
Martin
D
Rosell
D
Relationships of biological and taxonomic characteristics to chemically mediated bioactivity in Mediterranean littoral sponges
Mar Biol
 
1992
113
287
97
Vacelet
J
Etude en microscopie electronique de l'assocation entre bacteries et spongiaires du genre Verongia (Dictyoceratida)
J Microsc Biol Cell
 
1975
23
271
88
Vacelet
J
Donadey
C
Electron microscope study of the association between some sponges and bacteria
J Exp Mar Biol Ecol
 
1977
30
301
14
Van Dorst
J
Bissett
A
Palmer
AS
et al. 
Community fingerprinting in a sequencing world
FEMS Microbiol Ecol
 
2014
89
316
30
Van Soest
RWM
Boury-Esnault
N
Vacelet
J
et al. 
Global diversity of sponges (Porifera). Roberts JM (ed.)
PLoS One
 
2012
7
e35105
Webster
NS
Sponge disease: a global threat?
Environ Microbiol
 
2007
9
1363
75
Webster
NS
Cobb
RE
Soo
R
et al. 
Bacterial community dynamics in the marine sponge Rhopaloeides odorabile under in situ and ex situ cultivation
Mar Biotechnol
 
2011
13
296
304
Webster
NS
Hill
RT
The culturable microbial community of the Great Barrier Reef sponge Rhopaloeides odorabile is dominated by an α-Proteobacterium
Mar Biol
 
2001
138
843
51
Webster
NS
Luter
HM
Soo
RM
et al. 
Same, same but different: symbiotic bacterial associations in GBR sponges
Front Microbiol
 
2013
3
444
Webster
NS
Negri
AP
Webb
RI
et al. 
A spongin-boring α-proteobacterium is the etiological agent of disease in the Great Barrier Reef sponge Rhopaloeides odorabile
Mar Ecol Prog Ser
 
2002
232
305
9
Weisz
JB
Hentschel
U
Lindquist
N
et al. 
Linking abundance and diversity of sponge-associated microbial communities to metabolic differences in host sponges
Mar Biol
 
2007
152
475
83
Weisz
JB
Lindquist
N
Martens
CS
Do associated microbial abundances impact marine demosponge pumping rates and tissue densities?
Oecologia
 
2008
155
367
76
White
JR
Patel
J
Ottesen
A
et al. 
Pyrosequencing of bacterial symbionts within Axinella corrugata sponges: diversity and seasonal variability
PLoS One
 
2012
7
e38204
Wichels
A
Würtz
S
Döpke
H
et al. 
Bacterial diversity in the breadcrumb sponge Halichondria panicea (Pallas)
FEMS Microbiol Ecol
 
2006
56
102
18
Wilkinson
CR
Microbial associations in sponges. III. Ultrastructure of the in situ associations in coral reef sponges
Mar Biol
 
1978
49
177
85
Zhang
F
Vicente
J
Hill
RT
Temporal changes in the diazotrophic bacterial communities associated with Caribbean sponges Ircinia stroblina and Mycale laxissima
Front Microbiol
 
2014
5
561

Supplementary data