Diversity and antimicrobial activities of microbes from two Irish marine sponges, Suberites carnosus and Leucosolenia sp.

Authors

  • B. Flemer,

    1.  Marine Biotechnology Centre, Environmental Research Institute, University College Cork, Cork, Ireland
    Search for more papers by this author
  • J. Kennedy,

    1.  Marine Biotechnology Centre, Environmental Research Institute, University College Cork, Cork, Ireland
    Search for more papers by this author
  • L.M. Margassery,

    1.  Marine Biotechnology Centre, Environmental Research Institute, University College Cork, Cork, Ireland
    Search for more papers by this author
  • J.P. Morrissey,

    1.  Marine Biotechnology Centre, Environmental Research Institute, University College Cork, Cork, Ireland
    2.  Department of Microbiology, University College Cork, Cork, Ireland
    Search for more papers by this author
  • F. O’Gara,

    1.  Marine Biotechnology Centre, Environmental Research Institute, University College Cork, Cork, Ireland
    2.  Department of Microbiology, University College Cork, Cork, Ireland
    3.  BIOMERIT Research Centre, University College Cork, Cork, Ireland
    Search for more papers by this author
  • A.D.W. Dobson

    1.  Marine Biotechnology Centre, Environmental Research Institute, University College Cork, Cork, Ireland
    2.  Department of Microbiology, University College Cork, Cork, Ireland
    Search for more papers by this author

Alan D.W. Dobson, Environmental Research Institute, University College Cork, Lee Road, Cork, Ireland. E-Mail: a.dobson@ucc.ie

Abstract

Aims:  To evaluate the diversity and antimicrobial activity of bacteria from the marine sponges Suberites carnosus and Leucosolenia sp.

Methods and Results:  Two hundred and thirty-seven bacteria were isolated from the sponges S. carnosus (Demospongiae) and Leucosolenia sp. (Calcarea). Isolates from the phyla Actinobacteria, Bacteroidetes, Firmicutes and Proteobacteria were obtained. Isolates of the genus Pseudovibrio were dominant among the bacteria from S. carnosus, whereas Pseudoalteromonas and Vibrio were the dominant genera isolated from Leucosolenia sp. Approximately 50% of the isolates from S. carnosus displayed antibacterial activity, and c. 15% of the isolates from Leucosolenia sp. demonstrated activity against the test fungal strains. The antibacterial activity observed was mostly from Pseudovibrio and Spongiobacter isolates, while the majority of the antifungal activity was observed from the Pseudoalteromonas, Bacillus and Vibrio isolates.

Conclusions:  Both sponges possess a diverse range of bioactive and potentially novel bacteria. Differences observed from the sponge-derived groups of isolates in terms of bioactivity suggest that S. carnosus isolates may be a better source of antibacterial compounds, while Leucosolenia sp. isolates appear to be a better source of antifungal compounds.

Significance and Impact of the Study:  This is the first study in which cultured bacterial isolates from the marine sponges S. carnosus and a Leucosolenia sp. have been evaluated for their antibacterial activity. The high percentage of antibacterial isolates from S. carnosus and of antifungal isolates from Leucosolenia sp. suggests that these two sponges may be good sources for potentially novel marine natural products.

Introduction

Marine sponges are known to harbour a wide variety of bacteria with 26 bacterial phyla being detected in a recent study (Lee et al., 2011). One of the roles that has been proposed for these sponge-associated microbes is protection of the sponge from predation, potential pathogens, competitors and fouling organisms, by producing biologically active secondary metabolites (Taylor et al. 2007). Marine sponges themselves have proven to be a very rich source of biologically active and pharmaceutically important natural products, many with biotechnological relevant properties, including anticancer, antiviral and anti-inflammatory activities (Brady et al. 2009; Cragg et al. 2009; Piel 2009). Indeed, sponges are the most prolific marine producers of novel compounds, with more than 3500 new metabolites having been reported from sponges between 1985 and 2008 (Hu et al. 2011). These sponge-derived compounds include a wide variety of different chemical classes such as alkaloids, polyketides and terpenoids among others. The occurrence of structural similarities between some of these compounds from sponges and those from the sponge microbiota has led to the hypothesis that at least some of these bioactive compounds may in fact be of microbial origin (Wang 2006). In addition, a number of studies have shown that sponge-associated bacterial isolates produced the same compounds that had been isolated from the sponges themselves (Stierle et al. 1988; Bewley et al. 1996; König et al. 2006). Thus, it is clear that the sponge–microbe association makes sponges an ideal source for biologically active micro-organisms producing potentially novel chemicals and enzymes (Kennedy et al. 2008a).

In recent years, sponge-associated marine micro-organisms have received renewed attention with respect to their production of secondary metabolites, with numerous studies focusing on culturing these micro-organisms and screening them for the production of bioactive compounds (Hentschel et al. 2001; Muscholl-Silberhorn et al. 2008; Kennedy et al. 2009). This has involved culturing bacteria predominantly belonging to the actinobacteria, α-Proteobacteria and γ-Proteobacteria phyla, with limited reports from other bacterial phyla such as δ-Proteobacteria, Planctomycetes and Verrucomicrobia (Radjasa et al. 2011). Among bacterial isolates from marine sources, actinobacteria are the most significant producers of secondary metabolites, with c. 50% of the novel compounds derived from bacteria in 1985–2008 being from these genera (Hu et al. 2011). However, α- and γ-Proteobacteria together with Bacillus species have also been shown to produce compounds with antimicrobial activity (Hentschel et al. 2001; Pabel et al. 2003; Thakur et al. 2005; O’Halloran et al. 2011). The marine genus Pseudovibrio has only recently been described (Shieh et al. 2004) and to date comprises of three described species (P. denitrificans, P. ascidiaceicola and P. japonicus). Recent studies describe biologically active Pseudovibrio species isolated from different organisms including marine sponges (Hentschel et al. 2001; Thiel and Imhoff 2003; Kennedy et al. 2009). As little is currently known about the nature of these biologically active compounds derived from Pseudovibrio spp., they may prove to be a particularly promising source of potentially novel metabolites.

With the growing importance of drug-resistant infections over the last number of years, increases in infections by multidrug-resistant Escherichia coli, methicillin-resistant Staphylococcus aureus and other drug-resistant bacteria such as Salmonella spp. have become a more and more prevalent problem in the food industry and in clinical settings. Thus, there is an ever-increasing need to discover new antibiotics, with the most recent fatal outbreak of the Shiga toxin–producing Escherichia coli O104:H4 in Germany, which displayed resistance to multiple antibiotics providing a stark reminder of this fact (Bielaszewska et al. 2011).

Thus, this study focused on the isolation and identification of bacteria from two marine sponges namely Suberites carnosus and Leucosolenia sp. and the subsequent screening of these bacteria for antibacterial activities against three clinically relevant bacteria such as E. coli (NCIMB 12210), Bacillus subtilis (IA40) and S. aureus (NCIMB 9518), and against five fungal test strains. While some work has been performed on Suberites domuncula with respect to biosilica production (Wang et al. 2011), and Leucosolenia sp. regarding the production of novel bioactive aminoimidazole alkaloids (Ralifo et al. 2007), these sponges have not previously been targeted for their culturable biodiversity nor for their potential as sources for novel antibiotics. In addition, they represent two different classes of sponges, namely the Demospongiae (S. carnosus) and the Calcarea (Leucosolenia sp.), and in general, the calcareous sponges to date remain a largely overlooked class with respect to biodiscovery.

Materials and methods

Sponge sampling

Samples of S. carnosus and Leucosolenia sp. were collected at a depth of 15 m by SCUBA in Lough Hyne, Co. Cork, Ireland (51°50·556′ N, 09°30·389′ W) in November 2008. Whole sponge samples were either transported to the onsite laboratory in sea water, rinsed with sterile artificial sea water and directly processed for the cultivation of bacteria or frozen immediately following collection.

Isolation of bacteria

After collection, 1 g of each sponge sample was finely chopped with a sterile razor blade, mixed with sterile glass beads and 1 ml sea water and vortexed for 30 s. The homogenate was then diluted, and 100 μl of the dilutions 10−1–10−5 was spread on agar plates using sterile glass beads. For the cultivation, three different media were employed: starch-yeast extract-peptone-sea water (SYP-SW) agar (1 l distilled water, 10 g starch, 4 g yeast extract, 2 g peptone, 33·3 g artificial sea salts and 15 g agar), MMA agar (1 l natural, filtered autoclaved sea water, 50 μg yeast extract, 0·5 mg tryptone (enzymatic digest from casein), 0·1 mg Na-glycerol-PO4 and 15 g agar) and chitin medium (1 l distilled water, 4 g colloidal chitin, 33·3 g artificial sea water and 15 g agar). All agar plates were then incubated at 18°C for c. 2 months and checked regularly for the formation of distinguishable colonies. Colonies were picked from these isolation plates and re-streaked on SYP-SW agar plates. The colonies were not kept on the isolation medium in the case of MMA and chitin agar because they grew faster on SYP-SW agar. Colonies were selected based on their morphology and colour with the aim of picking as many different isolates as possible. Single colonies were repeatedly re-streaked to obtain pure colonies. For long-term storage, isolates were grown in liquid SYP-SW for 3–5 days at 18°C, sterile glycerol was added to 15%, and stocks were stored at −80°C.

Deferred antagonism assays

These assays were performed as previously described (Kennedy et al. 2009); 5 μl of the stock culture was spotted onto an SYP-SW agar plate and incubated at 18°C until the culture was c. 0·5–1 cm in diameter (for most isolates, 3–5 days of incubation was sufficient). The culture was then overlayed with 10 ml of soft agar seeded with a fresh culture of bacterial or fungal test strains. For bacterial test strains (E. coli NCIMB 12210, B. subtilis IA 40 and S. aureus NCIMB 9518), LB soft agar (1 l distilled water, 20 g LB broth, 5 g agar) was used. For fungal test strains (Candida albicans Sc5314, Candida glabrata CBS138, Saccharomyces cerevisiae BY4741, Kluyveromyces marxianus CBS86556 and Aspergillus fumigatus Af293), YPD soft agar (1 l distilled water, 10 g yeast extract, 20 g peptone, 20 g d-glucose, 7 g agar) was used. The overlayed plates were incubated for c. 12–24 h at 28°C for the fungal test strains and 37°C for the bacterial test strains. A zone of clearance in the overlay agar indicated the production of an antimicrobial compound by an isolate.

Well diffusion assays

Isolates that showed clear activity in the deferred antagonism assay were also assayed in the well diffusion assay. The isolates were grown up in 50 ml liquid SYP-SW in 250-ml Erlenmeyer flasks at 28°C and 180 rev min−1 for 14 days. Twice a week, 1 ml of the culture was taken and centrifuged at 20 238 g for 10 min to obtain a cell-free supernatant. Parallel, LB and YPD agar plates were prepared for testing of bacterial and fungal test strains, respectively. Plates were first seeded with 100 μl of a 1 : 50 dilution of an overnight culture of each test strain that was spread out on the plates using sterile glass beads. Holes were then punched into the agar with a heat-sterilized cork borer (diameter 5 mm). Then, 100 μl of the cell-free supernatant or SYP-SW broth as a negative control was applied to the wells. The plates were incubated for 12–24 h at 28°C for fungal test strains and 37°C for bacterial test strains. Zones of inhibition around the punched wells indicating antimicrobial activity of the according supernatant were recorded.

PCR amplification of 16S rRNA genes

A single colony of each isolate was added to 100 μl of autoclaved TE buffer, pH 8·0, and incubated for 15 min at 98°C. The mixture was centrifuged at 4500 g for 10 min, and 3 μl of the crude extracts was used in the PCR. To amplify the 16S rRNA gene, the universal primers 27f (5′-AGAGTTTGATCMTGGCTCAG-3′) and 1492r (5′-TACGGYTACCTTGTTACGACTT-3′) were used (Lane 1991), generating a PCR product of c. 1500 bp. The following PCR conditions were used: initial denaturation at 95°C for 5 min, followed by 35 cycles of 94°C for 30 s, 50°C for 30 s and 72°C for 2 min, and a final elongation step of 72°C for 10 min. The reaction mixture (30 μl) contained 3 μl 10× buffer (Fermentas, St Leon-Rot, Germany), 3 μl dNTPs (2 mmol l−1 each; Fermentas), 1·5 μl 27f forward primer (10 μmol l−1; Sigma-Aldrich, Arklow, Ireland), 1·5 μl 1492 r reverse primer (10 μmol l−1; Sigma-Aldrich), 0·15 μl TAQ-polymerase (DreamTaq™ DNA polymerase; Fermentas, 5 U μl−1), 17·85 μl molecular biology grade water and 3 μl DNA. The PCR products were analysed by agarose gel electrophoresis.

Sequencing and phylogenetic analysis of 16S rRNA gene products

The 16S rRNA gene PCR products were purified and partially sequenced using primer 27f (carried out by Macrogen, Seoul, Korea). The partial 16S rRNA gene sequences were manually checked for quality using Finch TV (http://www.geospiza.com/Products/finchtv.shtml) and then grouped into operational taxonomic units (OTUs) based on 98·5% sequence similarity with FastGroupII (Yu et al. 2006). Sequences representing isolates with antibacterial activity, antifungal activity and both antibacterial and antifungal activity (all from both sponges) and isolates from S. carnosus and Leucosolenia sp. without activity were uploaded and grouped separately. Representatives of each OTU obtained were aligned with neighbouring sequences obtained from RDP using the seqmatch tool (Cole et al. 2009). Only type strains with a sequence length of ≥1200 bp and of good quality were initially analysed. For isolates that formed deep branches in the phylogenetic tree, the closest blast hit sequence was included in the analysis. The sequences were aligned, and phylogenetic trees were calculated using mega5, using the neighbour-joining algorithm (Tamura et al. 2011). The evolutionary history was inferred using the neighbour-joining method (Saitou and Nei 1987). The optimal trees are shown, and the sum of branch lengths is given in the caption for each tree. The percentage of replicate trees (values below 50 deleted) in which the associated taxa clustered together in the bootstrap test (1000 replicates) are shown next to the branches (Felsenstein 1985). The trees are drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic trees. The evolutionary distances were computed using the maximum composite likelihood method (Tamura et al. 2004) and are in the units of the number of base substitutions per site. All positions containing gaps and missing data were eliminated.

Accession numbers

DNA sequences have been deposited at GenBank and have been given the accession numbers JQ042918JQ043154.

Results

Isolation of bacteria

One hundred and three and 134 bacteria were isolated from the S. carnosus (W13) and Leucosolenia sp. (W15) samples, respectively. The bacteria were picked from the original plates based on colony appearance to obtain a diverse collection of isolates. For all 237 bacteria, a 16S rRNA sequence was obtained. The phyla distribution of isolated bacteria from S. carnosus and Leucosolenia sp. is shown in Fig. 1 and in supplementary Figs S1–5. In both sponges, the most dominant isolates were γ-Proteobacteria (36 and 55% for S. carnosus and Leucosolenia sp., respectively), α-Proteobacteria (34 and 7%), Actinobacteria (11 and 5%), Firmicutes (10 and 11%), Bacteroidetes (7 and 22%) and β-Proteobacteria (3% and 0%). Thus, on the phylum level, the isolates from both sponges are dominated by bacteria belonging to the Proteobacteria with 72% and 63% of isolates from S. carnosus and Leucosolenia sp. grouping into this phylum, respectively. The remaining c. 30% are shared more or less equally for both sponges between the phyla Actinobacteria, Bacteroidetes and Firmicutes, with the exception that isolates from the phylum Bacteroidetes were slightly more dominant and diverse among the bacteria from Leucosolenia sp. Among the Bacteroidetes, a total of 19 different OTUs were isolated from Leucosolenia sp. and only 5 OTUs from S. carnosus (Supplementary Fig. S2).

Figure 1.

 Phyla distribution in percentage of isolates from Scarnosus (black bar, 100% = 103 isolates) and Leucosolenia sp. (grey bar, 100% = 134 isolates).

Pseudovibrio was the most common genus isolated from S. carnosus (30 isolates), followed by Spongiobacter (10), Microbulbifer (8), Pseudoalteromonas (8), Vibrio (6), Bacillus (5) and Staphylococcus (4). For Leucosolenia sp., the most common isolates were of the genera Pseudoalteromonas (33) and Vibrio (27). Other frequently found genera were Bacillus (9), Shewanella (8), Staphylococcus (8), Maribacter (7), Polaribacter (5) and Formosa (5).

The four bacterial phyla obtained from the two sponges have previously been detected with culture-dependent and culture-independent methods (Taylor et al. 2007). This is also true on the genus level with representatives of, for example, Pseudovibrio (Kennedy et al. 2009; O’Halloran et al. 2011), Pseudoalteromonas (Dieckmann et al. 2004; Kennedy et al. 2009), Vibrio (Dieckmann et al. 2004), Spongiobacter (Thiel et al. 2007; Kennedy et al. 2008a,b; Mohamed et al. 2008), Pseudomonas (Thakur et al. 2005; Kennedy et al. 2008a,b) and Bacillus (Webster and Hill 2001; Kennedy et al. 2009) being commonly isolated from marine sponges or detected with culture-independent methods. But some isolates, especially from Leucosolenia sp., have a sequence similarity <98% to the closest hit in a BLAST search and therefore probably represent novel species or genera (data not shown). The majority of these potentially novel species are Bacteroidetes, with all but one isolated from the Leucosolenia sp. sponge sample. The bacteria were isolated on different growth media (chitin agar, minimal agar and SYP-SW agar), with the distribution of bacterial groups shown in Table 1.

Table 1.   Phyla distribution of 16S rRNA gene sequences obtained from isolates from Leucosolenia sp. (L) and S. carnosus (Sc) related to the isolation medium. Values are given as % of the total isolates of any given phylum, that is, 33% of all α-Proteobacteria isolates from Leucosolenia sp. were isolated on chitin agar plates
 ProteobacteriaActinobacteriaBacteroidetesFirmicutes
αβγ
LScLScLScLScLScLSc
Chitin335733161601819292920
MMA4411331227437352712410
SYP22313372575793004770

Antimicrobial activity (deferred antagonism assay)

All isolates obtained from the two sponge samples were tested in an overlay assay against three clinically relevant bacterial and five fungal test strains: E. coli (NCIMB 12210), B. subtilis (IA40), S. aureus (NCIMB 9518), C. albicans (Sc5314), C. glabrata (CBS138), S. cerevisiae (BY4741), K. marxianus (CB86556) and A. fumigatus (Af293). In total, 69 isolates showed clear activity against at least one of the test strains (Table 2) with 45 isolates from S. carnosus being active against bacterial test strains, two isolates active against fungal test strains and two isolates active against both fungal and bacterial test strains. Thus, in total, 49 isolates (i.e. 48%) of the 103 isolates from S. carnosus inhibited microbial growth in the initial activity assay. The main contributors to antimicrobial activity belonged to the phylum α-Proteobacteria (55% of the 49 active isolates from S. carnosus were α-Proteobacteria) followed by γ-Proteobacteria isolates (31%). Isolates from Leucosolenia sp. did not show antibacterial activity, but 20 isolates (15%) showed activity against the fungal test strains. The main contributors to antifungal activity belonged to the phylum γ-Proteobacteria (55% of the 20 active isolates from Leucosolenia sp. were γ-Proteobacteria) followed by Firmicutes isolates (40%). Only one isolate (5%) from the phylum Bacteroidetes showed activity. Thus, bacterial isolates from the phyla Proteobacteria and Firmicutes proved to be most active in this study.

Table 2.   Active Isolates. All 69 isolates that showed clear activity in the overlay assay against at least one test strain. Also given are the closest blast hits to the 16S rRNA sequence including the sequence similarity
IsolateClosest blast hitSimilarity (%)ECBSSACACGSCKMAF
  1. W13 = isolate from S. carnosus; W15 = isolate from Leucosolenia sp.; C, M and S = isolate obtained from chitin, MMA or SYP agar, respectively; the test strains used were as follows: EC = E. coli; BS = B. subtilis; SA = S. aureus; CA = C. albicans (Sc5314); CG = C. glabrata (CBS138); SC = S. cerevisiae (BY4741); KM = K. marxianus (CB86556) and AF = A. fumigatus (Af293); + = clear zone of inhibition visible in overlay agar; (+) = faint zone of inhibition; − = no zone of inhibition; nt = not tested.

W13M60Aquimarina muelleri strain KMM 602198+
W13M61AAquimarina muelleri strain KMM 602198+
W13M62BAquimarina muelleri strain KMM 602198+
W13C11Arthrobacter sp. 7A9S3100+++
W13S8Bacillus pumilus strain Ba9100++
W13S32Bacillus sp. RS114(2010)99(+)++
W13M2Pseudoalteromonas sp. STAB 201100+++
W13S34Pseudoalteromonas sp. STAB 201100++
W13S28Pseudoalteromonas tetraodonis100+
W13M4Pseudomonas sp. C12799++
W13C16Pseudovibrio sp. Ad299(+)+
W13C12Pseudovibrio sp. Ad3599(+)++
W13C15Pseudovibrio sp. Ad42100(+)++
W13C24Pseudovibrio sp. Ad42100(+)+(+)
W13C25Pseudovibrio sp. Ad42100(+)++
W13C28Pseudovibrio sp. Ad42100++
W13C32Pseudovibrio sp. Ad42100(+)+(+)
W13S49Pseudovibrio sp. Ad5399+++
W13C22Pseudovibrio sp. Ad54100(+)+
W13C30Pseudovibrio sp. Ad5499++
W13S4Pseudovibrio sp. Ad57100+++
W13S21Pseudovibrio sp. Ad5899+++
W13C10Pseudovibrio sp. Hs3100(+)+
W13C14Pseudovibrio sp. Hs3100(+)+
W13C17Pseudovibrio sp. Hs3100(+)++
W13C18Pseudovibrio sp. Hs3100(+)+
W13C21Pseudovibrio sp. Hs3100(+)+
W13C23Pseudovibrio sp. Hs3100(+)+
W13C27Pseudovibrio sp. Hs3100(+)+
W13C34Pseudovibrio sp. Hs3100++
W13S13Pseudovibrio sp. Hs3100++
W13S16Pseudovibrio sp. Hs3100+(+)
W13S22Pseudovibrio sp. Hs3100++nt
W13S23Pseudovibrio sp. Hs399+++
W13S26Pseudovibrio sp. Hs3100+++
W13S31Pseudovibrio sp. Hs399+++
W13C33Pseudovibrio sp. Pb2100++
W13S54Psychrobacter faecalis strain UCL-NF 159099+
W13M7Ralstonia sp. 1F2 16S100+
W13S29Shewanella sp. E505-799+
W13C2Spongiobacter sp. S229398(+)+
W13C5Spongiobacter sp. S229398(+)+
W13C7Spongiobacter sp. S229399(+)+
W13S12Spongiobacter sp. S229398++
W13S18Spongiobacter sp. S229398+
W13S2Spongiobacter sp. S229398+
W13S30Spongiobacter sp. S229398+
W13S46Spongiobacter sp. S229398+
W13S51Spongiobacter sp. S229398+
W15C18aBacillus amyloliquefaciens strain Z100+
W15C2Bacillus amyloliquefaciens strain Z99+
W15M1ABacillus amyloliquefaciens strain Z99++++
W15S58aBacillus simplex99+
W15S84aBacillus sp. A-05100+
W15S84bBacillus thuringiensis strain W8B-8099+
W15S32Formosa algae strain F8999+
W15S83Pseudoalteromonas issachenkonii99+
W15S11Pseudoalteromonas sp. A2B1099+
W15M16Pseudoalteromonas sp. K2B-299+
W15S10Pseudoalteromonas sp. LJ1100+
W15S18Pseudoalteromonas sp. LJ1100+
W15S14Pseudoalteromonas sp. S317899+
W15M34Staphylococcus saprophyticus100+
W15S87aStaphylococcus sp. HJB00399+
W15S67Vibrio litoralis strain MANO22P100+
W15S24bVibrio sp. SC-C1-599+++
W15C15Vibrio sp. BSw21697100++++
W15C16Vibrio splendidus LGP3299++++
W15C28Vibrio splendidus LGP3299+++

All of the growth media used recovered bioactive isolates with chitin agar and SYP agar, resulting in the isolation of greater numbers and proportions of active isolates (Fig. 2) than MMA. In terms of obtaining bioactive isolates, chitin agar proved to be most successful for both sponges.

Figure 2.

 Effect of culture medium on recovery of bioactive isolates, that is, 67% of the Suberites isolates (black bar) derived from chitin isolation medium were active against at least one test strain in the initial deferred antagonism assay. Only 20% of the Leucosolenia sp. isolates (grey bar) derived from the same medium were active in this assay.

Antibacterial isolates (Table 3, Figs 3 and 4) were most commonly found in the genera Pseudovibrio (27), Spongiobacter (9), Aquimarina (3) and Bacillus (2). Antifungal isolates were most common from Pseudoalteromonas (8), Bacillus (7), Vibrio (5) and Staphylococcus (2). Representatives of many of those genera have been frequently shown to produce antimicrobial compounds (Hentschel et al. 2001; Muscholl-Silberhorn et al. 2008; Gram et al. 2009; Kennedy et al. 2009; O’Halloran et al. 2011), but only one report could be found for antimicrobial activity from a Spongiobacter isolate (Gram et al. 2009).

Table 3.   Summary of antibacterial activities detected in isolates from S. carnosus and Leucosolenia sp.
PhylumGenusS. carnosusLeucosolenia sp.
NABB&FF%NABB&FF%
  1. The different isolates are grouped into genera and are distinguished by their activity profile in the deferred antagonism assay: NA, not active; B, antibacterial activity; B&F, antibacterial and antifungal activity; F, antifungal activity; %, percentage of bioactive isolates relative to the total number of isolates in this genus.

ActinobacteriaArthrobacter110030
BacteroidetesAquimarina310010
BacteroidetesFormosa4120
FirmicutesBacillus311403667
FirmicutesStaphylococcus406225
α-ProteobacteriaPseudovibrio32790
β-ProteobacteriaRalstonia2133
γ-ProteobacteriaSpongiobacter18190
γ-ProteobacteriaMicrobulbifer711310
γ-ProteobacteriaPseudoalteromonas5123827618
γ-ProteobacteriaPsychrobacter110010
γ-ProteobacteriaShewanella115080
γ-ProteobacteriaVibrio6022519
Figure 3.

 Total number of isolates (black bar) and number of bioactive isolates (grey bar) from Scarnosus (a) and Leucosolenia sp. (b). Shown are genera for which at least one bioactive isolate in either of the two sponges was detected in the deferred antagonism assay.

Figure 4.

 Phylogenetic analysis of 16S rRNA gene sequences from bioactive sponge isolates. Sequences obtained in this study are marked (inline image). Where a strain represents more than one isolate, the corresponding number is given in brackets. Bioactivities of isolates are marked with B (antibacterial), F (antifungal) or BF (both antibacterial and antifungal). W13 = isolate from S. carnosus; W15 = isolate from Leucosolenia sp.; C, M and S = isolate derived from chitin, MMA or SYP medium, respectively. The optimal tree with the sum of branch length = 1·59214718 is shown. The analysis involved 50 nucleotide sequences. There were a total of 492 positions in the final data set.

Well diffusion assay

All 69 bioactive isolates were tested in a well diffusion assay, with nine isolates showing activity in this assay (three from S. carnosus and six from Leucosolenia sp.). None of the supernatants tested showed activity against more than one indicator strain, and no activity was found from any S. carnosus isolate against bacterial test strains. The three isolates from S. carnosus were active against A. fumigatus, and the isolates from Leucosolenia sp. were active against C. glabrata (two isolates) and C. albicans (four isolates) (Table 4).

Table 4.   Well diffusion assays. In total, 69 isolates were tested, and nine showed activity against one of the fungal test strains for which the results are shown
IsolateClosest blast hitCACGSCKMAF
  1. None of the isolates tested showed activity against any of the bacterial test strains. CA, Calbicans (Sc5314); CG, Cglabrata (CBS138); SC, Scerevisiae (BY4741); KM, Kmarxianus (CB86556) and AF, Afumigates (Af293); +, clear zone of inhibition; −, no zone of inhibition.

W13M2Pseudoalteromonas sp. STAB 201+
W13S28Pseudoalteromonas tetraodonis+
W13S8Bacillus pumilus strain Ba9+
W15C15Vibrio sp. BSw21697+
W15C18ABacillus amyloliquefaciens strain Z+
W15C28Vibrio splendidus LGP32+
W15M16Pseudoalteromonas sp. K2B-2+
W15S14Pseudoalteromonas sp. S3178+
W15S87AStaphylococcus sp. HJB003+

Discussion

The bacteria isolated from both sponges were diverse, with microbes from four phyla and a combined total of 85 OTUs being isolated from both sponges. Whereas for S. carnosus this is perhaps not surprising because demosponges are known to host a diverse range of bacteria, calcareous sponges for their part have not to date been thoroughly analysed for their microbial biodiversity. Detailed phylogenetic analysis of these isolates showed that only 11 of the 85 OTUs were obtained from both sponges. Both Pseudovibrio and Spongiobacter, the dominant isolates from the demosponge S. carnosus, have been commonly isolated from demosponges and other marine invertebrates and are likely to have a symbiotic relationship with the sponge. The common isolates from Leucosolenia sp., Pseudoalteromonas and Vibrio, while also commonly isolated from sponges, are also highly abundant in sea water, perhaps implying a less close, more opportunistic relationship. In a parallel study, metagenomic analysis of sea water collected at the same site and time revealed the most dominant phyla to be Vibrio and Pseudoalteromonas (S. Jackson, personal communication).

Most of the isolates from both sponges have >98% sequence identity to 16S rRNA genes from cultured bacteria in the GenBank database. However, several isolates, especially those derived from Leucosolenia sp., have a 16S rRNA gene sequence similarity equal to or <97% identity, indicating potentially novel species or genera (data not shown). These isolates did not inhibit the growth of any of the test strains in the initial assays, but nonetheless, the potential of the investigated sponge as a source for novel micro-organisms is evident. As sponges of the class Calcarea have so far been almost largely overlooked, it is perhaps not surprising that the calcareous sponge investigated in this study reveals more novelty in its culturable microbiota. This finding also highlights the potential of this genus of calcareous sponges as a potential source for novel metabolites, enzymes or other activities produced by their associated microbes. Some of the most promising isolates are currently being analysed for description as new species or potentially novel genera.

A wide range of bacteria with antimicrobial activity were isolated from both sponges. Representatives of 13 bacterial genera (21 OTUs) and four phyla exhibited activity against at least one of the test strains used (Table 3). The isolates from S. carnosus, especially of the genus Pseudovibrio, were more likely to exhibit antimicrobial activity (48% of all isolates showed activity) than isolates from Leucosolenia sp. (15%). Especially interesting in this respect is the observation that 46% of all Suberites isolates were biologically active against bacterial test strains and only 4% of isolates were antifungal (2% of Suberites isolates were both antibacterial and antifungal). Conversely, none of the isolates from Leucosolenia sp. inhibited bacterial growth in the deferred antagonism assay, while 15% displayed antifungal activity. Differences in the activity profiles of the sponge isolates are also reflected in the genera isolated from the sponges. For Leucosolenia sp., the clearly dominant genera were Pseudoalteromonas (27 isolates) and Vibrio (23), while the dominant genera isolated from S. carnosus were Pseudovibrio (30 isolates) and Spongiobacter (10). The Pseudovibrio and Spongiobacter isolates comprised a large fraction of the antibacterial isolates from S. carnosus, and neither of these genera were obtained from Leucosolenia sp., which could explain the lesser extent of antibacterial activity. It is also apparent that both sponges share isolates of the same genera which demonstrated antibacterial activity if isolated from S. carnosus and no, or antifungal, activity if isolated from Leucosolenia sp. This is true for isolates of the genus Arthrobacter, Aquimarina, Bacillus, Microbulbifer, Psychrobacter, Shewanella, Staphylococcus and Vibrio (Table 3). The only exception to this are isolates of the genus Pseudoalteromonas which showed antifungal activity regardless of their origin. While this study was designed to access culturable microbial diversity (rather than measure relative abundances) and is subject to the biases of all culture-based approaches, nonetheless, the apparent sponge-specific activity profile is striking and could be due to subtle genetic differences between members of the same bacterial species residing in the different sponges (e.g. plasmid-encoded activities). In summary, the different activity profiles of isolates from the two sponges can partly be explained by the different genera and species isolated, but intraspecies variation also appears to play a role.

It is especially interesting that isolates belonging to genera not often or not all reported to exhibit antimicrobial activity were found to inhibit the growth of the test microbes. For good reason, the focus of biodiscovery researchers has been on bacteria belonging to actinobacterial genera, particularly Streptomyces and Micromonospora. These bacteria have been routinely shown to be prolific producers of secondary metabolites and to be abundant members of the sponge-associated microbiota, but it has also been reported that their abundance varies greatly from sponge to sponge. From the sponges targeted in the present study, no isolate grouped with the above-mentioned genera. However, representatives from different genera such as Spongiobacter and Aquimarina were shown to exhibit antimicrobial activity. To our knowledge, this is one of the first reports for representatives of these genera to inhibit the growth of microbes. The other main genera found (Bacillus, Pseudovibrio, Pseudoalteromonas and Vibrio) have been previously reported to exhibit antimicrobial activity, while the genus Pseudovibrio has gathered the interest of researchers recently because of the broad range of activities detected in isolates of this genus (Hentschel et al. 2001; Thiel and Imhoff 2003; Kennedy et al. 2009; O’Halloran et al. 2011). The different bioactivity profiles of Pseudovibrio isolates that are phylogenetically similar indicate the production of different compounds from closely related bacteria. As only two antimicrobial compounds from a Pseudovibrio isolate (Sertan-de Guzman et al. 2007; Penesyan et al. 2011) have to date been characterized, this genus represents a very promising source of potentially novel bioactive metabolites.

In summary, a wide variety of bacteria have been isolated with a high percentage exhibiting antimicrobial activity. The isolates of Leucosolenia sp., a calcareous sponge, only showed activity against different fungal test strains, whereas almost 50% of the isolates from S. carnosus showed activity against bacterial test strains. Also, biological activity was detected for genera that have not previously been shown to inhibit microbial growth. This not only highlights the potential of the genus Suberites, but it also shows that the so far almost overlooked calcareous sponges could also play an important role in the biodiscovery of novel compounds. This is especially true because they appear to harbour hitherto unknown bacterial species. The differences in the antibacterial and antifungal profiles of the isolates from the two sponges are intriguing. While from this study it is clear that the isolates from the Calcareous sponge Leucosolenia sp. appear to be a better source for antifungal agents and those from S. carnosus a better source of antibacterial activity, the biological reason for this is unknown, and whether this is because of the different competing microbes or pathogens in the two niches or biases in culturing may be worthy of further studies with additional samples. Finally, while in this study many antimicrobial isolates were found, antimicrobial activity was subsequently difficult to establish in shake flask cultures. Other approaches such as extraction from agar plates, changing the culture medium and concentrating the broth before testing it in the bioassay have also been tried with selected isolates but have proven unsuccessful. Little is known regarding antimicrobials produced by the genera Spongiobacter, Aquimarina and Pseudovibrio; however, new approaches to induce the production of bioactive compounds such as co-cultivation techniques (Nützmann et al. 2011; Pérez et al. 2011) and biofilm formation (Yan et al. 2003; Wilson et al. 2011) may help in the further characterization of the bioactive component of these strains.

Acknowledgements

This research was supported by a grant awarded by the Irish Marine Institute under the strategic Marine Biodiscovery RTDI Programme and by the Marine Biodiscovery Research award funded by the Irish Government under the National Development Plan (2007–2013). We would like to thank Bernard Picton (Ulster Museum) and Dr R. McAllen (UCC) for their help in the collection and identification of sponge samples.

Ancillary