Introduction

The increasing demand for new therapeutic drugs from natural products led to greater interest towards marine microorganisms which are prolific producers of bioactive secondary metabolites in response to ecological parameters such as competition for space and maintenance of un-fouled surfaces (Konig et al. 1994). Colonization of sessile eukaryotic host surfaces by bacteria is common in the marine environment (Thomas et al. 2008), and macroalgae have long been known to support abundant populations of bacteria (Armstrong et al. 2000; Rao et al. 2005). Seaweeds are highly productive components of the coastal ecosystem releasing dissolved organic carbon into surrounding waters thus harboring suitable living substrata for microbial colonization (Hengst et al. 2010). Chemically driven interactions are important in the establishment of cross relationships between marine surface-associated microorganisms and their eukaryotic host (Imhoff et al. 2011). Epiphytic bacteria produce antimicrobial compounds that may be protective for algae from colonization by other microbiota and macrobiota (Holmstrom et al. 2002; Rao et al. 2006, 2007; Hengst et al. 2010). Other benefits can be attributed to the symbiotic relation; Croft et al. (2005) suggested that seaweeds, as microalgae, may acquire vitamin B12 from closely associated bacteria.

Few data is available on epiphytic bacteria of red algae. Recently, Kanagasabhapathy et al. (2008) identified epibiotic bacteria of nine non-calcareous species of red algae from Japanese waters with antimicrobial activities. Other studies concerned the bacterial biota on crustose coralline algae (Lewis et al. 1985; Johnson et al. 1991; Barott et al. 2011). It has been suggested that the characteristic bacteria associated with such species influence invertebrate larvae recruitments. Sakami et al. (1999) studied epiphytic bacteria of the branched coralline alga Jania sp. and their effects on the growth of toxic dinoflagellate. Among the isolated bacteria, one Flavobacterium strain inhibited positively Gambierdiscus toxicus, a microalga known to be causative of ciguatera fish food poisoning. In addition, Boyd et al. (1999) reported antifouling properties of epiphytic bacteria associated with the calcareous red alga Corallina officinalis.

Despite the limitations of cultivation-based studies, cultivation remains essential as it provides opportunities to study and understand microbial ecology and physiology and design antibiotic screening assays (Mearns-Spragg et al. 1998; Armstrong et al. 2000).

The branched coralline alga Jania rubens is widespread along Tunisian coasts, especially during spring season. It has been studied for its cytotoxic activity (Ktari et al. 2000). Still, no data are available concerning its epibionts and their biological potential. Thus, the present study provides information on the diversity of culturable bacterial communities intimately associated with J. rubens surfaces as well as their antimicrobial potential against several human and fish pathogens.

Materials and methods

Samples of J. rubens (Linnaeus) J.V. Lamouroux were collected within the period of February to August 2007 from Cap Zebib zone (37° 16.2′ N, 10° 3.6′ E), northern coast of Tunisia. Algae samples were transferred in the dark in sterile plastic bags filled with water from same location.

Isolation of epiphytic bacteria

Seaweed samples were washed three times with autoclaved seawater to remove free living and associated bacteria (Gil-Turnes et al. 1989; Cheng et al. 1999; Jiang et al. 1999; Burgess et al. 2003). Subsequently, firmly attached epiphytic bacteria were extracted by vortexing 10 g of alga in 90 mL autoclaved seawater for 6 min. Bacteria were isolated by serial dilution to 10−3 using sterile seawater. From each dilution 100 μL was spread-plated in triplicate on marine agar (MA: Pronadisa Laboratories, CONDA). The plates were incubated at 20°C until colonies appeared or at least 7 days (Lemos et al. 1985). Visually distinct bacterial colonies were selected and further plated on MA until clonal cultures were obtained. The pure cultures were stored at −80°C in marine broth (Pronadisa Laboratories, CONDA) supplemented with 20% glycerol.

Extraction of DNA and PCR

Single colonies from plates were suspended in sterile MilliQ water and used as template in the PCR reactions using the universal 16S ribosomal RNA (rRNA) gene primers B8F and U1492R (Table 1). PCR reactions were performed using a Gene Biometra T1 DNA thermal cycler (Perkin-Elmer Co., USA) in 25 μL (final volume) reaction mixtures containing 0.1 μL of Hot start DNA polymerase (Sigma), each primer at a final concentration of 10 pmol μL−1, each deoxynucleoside triphosphate at a concentration of 200 μM, 1.25 μL of 100% DMSO (dimethylsulfoxide), 2.5 μL of BSA (bovine serum albumin) at 0.2 mg mL−1 final concentration, 2.5 μL of PCR buffer (including MgCl2) and 1 μL of DNA template. The PCR protocol consisted of 40 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 30 s and extension at 72°C for 1 min 50 s. The cycles were preceded by 15 min of denaturation at 94°C and ended with a final extension for 7 min at 72°C. Negative controls contained all of the components of the PCR mixture except the DNA template.

Table 1 Primers used for PCR and DNA sequencing
Full size table

Analysis of PCR products

PCR products (2 μL) were analyzed by agarose gel electrophoresis (1% w/v agarose, 1× TAE running buffer containing 40 mM Tris acetate and 1 mM EDTA, pH 8). Electrophoresis was performed at 100 V for 45 min. The gels were stained for 45 min with SYBR Gold nucleic acid gel stain (Invitrogen Corp.) and photographed under UV illumination.

DNA sequencing

Approximately 100 ng template DNA was sequenced using the BigDye Terminator chemistry (Big Dye Terminator v3.1 Cycle Sequencing Kit, Applied Biosystem) according to the manufacturer’s instructions. The sequence products were analyzed on an ABI 377 DNA sequencing system (Applied Biosystem) using the primers C5, C26, C72 and C112 (Table 1).

Phylogenetic analysis

The closest phylogenetic relatives of each isolate were identified by comparison of the 16S rRNA gene sequence to the National Center for Biotechnology Information (NCBI) GenBank database using the Basic Local Alignment Search Tool (BLAST) analysis tools (http://www.ncbi.nlm.nih.gov/BLAST). Phylogenetic analysis of J. rubens isolates sequences and that of other red alga isolates sequences was performed using the Neighbor Joining method available in MEGA4.0 (Tamura et al. 2004).

Nucleotide sequence accession numbers

Nucleotide sequences of the isolates sequenced in this study have been added to the DNA GenBank with the following accession numbers: J1 (JN391160), J2 (JN391161), J3 (JN391162), J4 (JN391163), J5 (JN391164), J6 (JN391165), J7 (JN391166), J8 (JN391167), J9 (JN391168), J10 (JN391169), J11 (JN391170), J12 (JN391171), J13 (JN391172), J14 (JN391173), J15 (JN391174), J16 (JN391175), J17 (JN391176), J18 (JN391177) and J19 (JN391178).

Preparing algal extracts

About 20 g of dried and powdered alga was extracted consecutively with two organic solvents with increasing polarity: dichloromethane (D) and dichloromethane/methanol (D/M) (1:1 v/v). Each extraction was carried out three times by maceration for 24 h at room temperature. These extracts were pooled, filtered and concentrated under reduced pressure in a rotary evaporator apparatus. Extracts were stored at −20°C until use.

Antimicrobial activity

Antimicrobial activity of both algal isolates and algal extracts was evaluated against the following pathogen strains: Escherichia coli O126B16, E. coli ATCC 25922, E. coli ATCC 8739, Vibrio tapetis CECT4600, V. anguillarum, V. alginoliticus ATCC 17749 T, Pseudomonas cepacia, P. fluorescens AH2, P. aeruginosa ATCC 27853, Aeromonas salmonicida, A. hydrophila, Salmonella typhimurium, Streptococcus sp., Staphylococcus aureus, S. aureus ATCC 25923, S. aureus ATCC 6538, Enterococcus feacalis ATCC 29212, Micrococcus and the yeast Candida albicans. Algal extracts were also tested against Jania isolates and four strains from alga surrounding water.

Bacterial isolates were screened for their antimicrobial activity against pathogens using drop test assay on T soy agar (TSA, BIO RAD) plates containing 20 g L−1 NaCl as described by James et al. (1996) and Rao et al. (2005) with slight modifications. Briefly, drops of the full cell suspension of an overnight culture were spotted onto the agar plates containing a confluent lawn of the target strain (dried at 30°C for 30 min) and subsequently incubated at 30°C. Growth inhibition was evaluated after incubation (24 h at 30°C) by measuring the zone of inhibition around the spots.

The antibacterial assay of J. rubens extracts against its proper isolates, indicators bacteria and four Jania surrounding water isolates was evaluated by using standard paper disc method. Briefly, 500 μg of the crude extract dissolved in appropriate solvent (10 μL) was applied to sterile filter paper discs (6 mm). After solvent evaporation, the discs were placed on TSA plates, inoculated with an 18-h cultured associated strain (106 bacteria mL−1) in tryptic soy broth (TSB, BIO RAD) containing 20 g L−1 NaCl. As control, a disc loaded with solvent was simultaneously prepared. Plates were incubated overnight at 30°C. The diameter (in millimeters) of growth inhibition halo was measured after 24-h incubation. Assays were carried out in triplicate.

Results

In this study we first examined the composition on the culturable surface-associated bacterial community on J. rubens using 16S rRNA gene sequence analysis.

Nineteen epiphytic bacteria (J1–J19) were isolated as pure culture on marine agar. The 16S rRNA gene sequences revealed that the main bacterial groups present in the surface-associated community were Alphaproteobacteria and Gammaproteobacteria, Bacteroidetes and Firmicutes (Table 2). Results demonstrated that Proteobacteria (about 73%) constitute the majority of bacterial cells on the surface of J. rubens and were assigned especially to seven families, with the predominant one being Rhodobacteraceae containing five strains (J1, J3, J4, J7 and J16) followed by Pseudomonadaceae (J11, J13 and J14), Pseudoalteromonadaceae (J17 and J18), Vibrionaceae (J19), Oceanospirillaceae (J10) and Rhizobiaceae (J8). The second most represented phylogenetic group was Bacteroidetes representing 21% of the epiphytic totality and belonging to three genera: Flavobacteria, Aquimarina and Cytophaga. However, only one strain was found assigned to the Firmicutes and was identified as closely related to Bacillus. These data demonstrate that the J. rubens-associated bacterial community was diverse with Simpson’s index D of 0.035. Phylogenetic analysis of the 16S rRNA gene sequences of the J. rubens bacterial isolates was performed using the Neighbor Joining method as available in MEGA4.0 (Tamura et al. 2004) with 1000 replications. Most of the members of the Alphaproteobacteria group formed one clade with a bootstrap value of 48%. Members of group Bacteroidetes, i.e. J9, J6, J12, J15 and J19, cluster in same clade with strong bootstrap value (74%) as shown in Fig. 1.

Table 2 Identification of 19 bacterial isolates associated with J. rubens based on partial 16S rRNA gene sequence identity obtained by BLAST analysis
Full size table
Fig. 1
figure 1

Neighbor-joining tree based on partial 16S rRNA gene sequences derived from J. rubens bacterial isolates and those derived from other red alga isolates (identified via a BLAST search). The scale bar indicates 20% of sequence variation

Full size image

J. rubens-identified isolates were subsequently tested for their antimicrobial potential against 19 pathogens. The antimicrobial assay showed that amongst isolates, seven strains corresponding to about 36% were active against pathogens with different antimicrobial spectrum (Table 3). These active isolates coded J2, J9, J11, J13, J16, J17 and J18 were assigned to five genera: Bacillus, Aquimarina, Pseudomonas, Pseudoalteromonas and Paracoccus. Three of them, J2, J9 and J13, identified respectively as Bacillus pumilus, Aquimarina sp. and Pseudomonas sp. showed the strongest inhibition especially against S. aureus, Micrococcus and C. albicans, with inhibition diameters of 25 to 35 mm. The largest antimicrobial spectrum was obtained for J2 which inhibited about 68% of the pathogens, followed by J13 and J11 strains which inhibited 52% and 42% of the tested pathogens respectively.

Table 3 Antibacterial spectra of Jania rubens-associated bacteria
Full size table

The three strains of S. aureus were the most sensitive, since they were inhibited by J2, J9, J11, J13 and J17 isolates, while E. faecalis and Streptococcus sp. were only inhibited by J2. However, E. coli (three strains) and P. fluorescens AH2 were resistant towards all J. rubens isolates. Figure 2a, b shows antibacterial activity of J2, J9, J11, J13 and J17 against S. aureus by drop method and antibacterial activity of J2 isolates against Streptococcus sp. respectively.

Fig. 2
figure 2

a Antibacterial activity of J2, J9, J11, J13 and J17 against Staphylococcus aureus by drop method. b Antibacterial activity of J2 isolates against Streptococcus sp

Full size image

Strains J16, J17 and J18 were weakly active, and each one inhibited only four of the 19 tested pathogens. Despite its weak antimicrobial spectrum, J16 isolate identified as Paracoccus sp. displayed strong activity against P. cepacia. The growth of S. typhimurium was inhibited by the two isolates coded J2 and J18.

In order to verify the close relationship between host and isolated bacteria, toxicity of J. rubens organic extracts (polar and non-polar) was tested against 18 human and fish pathogens, the yeast C. albicans, four bacteria isolated from J. rubens surrounding water and against the 19 surface epibionts (J1–J19). Antimicrobial tests revealed activity of J. rubens extracts against the pathogens S. aureus, Micrococcus sp., Streptococcus sp. and P. cepacia, as well as one of the four surrounding water isolates which was identified by 16S rRNA gene sequence as closely related to Shewanella sp. Because the extracts of Jania did not inhibit the growth of any of the strains isolated from the surface of this organism, we conceived that the Jania isolates were intimately associated with their host.

Discussion

According to the present study, J. rubens harbors a diverse community of bacteria associated with its surface of which its members are able to grow on marine agar. 16S rRNA gene sequence analyses were used for bacteria identification.

In bacterial taxonomy it is commonly accepted that two bacteria do belong to the same species unless the 16S rRNA gene sequence similarity is < 97% (Stackebrandt and Goebel 1994; Hagström et al. 2000; Cabaj et al. 2006). 16S rRNA gene sequence of strains J10, J15 and J19 displayed a similarity with other published sequences of 95% for J10 and 96% for J15 and J19, which suggests that these are novel species, hitherto unknown species. This has still to be confirmed by further studies; universally conserved protein-coding genes such as gyrB, rpoB, recA and lepA may be used as an alternative to 16S rRNA gene sequencing.

The J. rubens bacterial community was considered as diverse according to Simpson’s index value. We notice that a number of bacteria associated with other algae and other marine macroorganisms were amongst the closest relatives of phylotypes associated with J. rubens (Sakami et al. 1999; Staufenberger et al. 2008).

Flavobacterium spp. were also isolated from Jania spp., from Tahiti (Sakami et al. 1999).

In previous investigations (Ismail-Ben Ali et al. unpublished data), we isolated bacteria of the genera Pseudomonas, Pseudoalteromonas, Paracoccus and Bacillus from brown alga Padina pavonica sampled from the same locality as J. rubens (Cap Zebib). Moreover, we reported isolates from Ulva rigida surfaces belonging to the genera Rhodobacteraceae, Roseovarius, Bacillus, Paracoccus, Loktanella, Shewanella and Pseudomonas. Bacteria of the genus Loktanella was also isolated from Ulva intestinalis surface (Ismail-Ben Ali et al. 2010). Further studies dealt with the interaction of cultured bacteria with the brown alga Laminaria spp. Members of the genera Flavobacterium and Pseudomonas, with the ability to utilize algal compounds such as mannitol, alginate and laminaran as substrates were isolated (Staufenberger et al. 2008). Bacteria that belong to Cytophaga-Flavobacterium-Bacteriodes (CFB) are often found in natural microbial communities in marine environments (Nedashkovskaya et al. 2003). These bacteria play an important role in the normal development of green algae in the marine coastal environment (Matsuo et al. 2003). In addition, members of the CFB group were abundant as epiphytic bacteria on two common freshwater macrophytes, the macroalga Chara aspera and the angiosperm Myriophyllum spicatum (Hempel et al. 2008).

In our study, identification of Cytophaga strain indicates that J. rubens harbors bacteria that are important for thallus structural maintenance as reported by Tujula (2006). Matsuo et al. (2003) identified Cytophaga species capable of inducing morphogenesis on the green marine alga Monostroma oxyspermum, demonstrating the importance of these associated strains in algal morphology.

Strains of the Cytophaga and Pseudomonas group were previously reported associated with coralline algal surface (Lewis et al. 1985). In addition, algicidal bacteria belonging to the genera Alteromonas, Pseudoalteromonas and Cytophaga were isolated from Ulva sp. and Gelidium sp. from the coast of Osaka Bay, Japan and have been also reported in coastal red tide areas (Imai et al. 2006).

The marine bacteria Pseudomonas sp. and Pseudoalteromonas sp. were isolated from the seaweed Diginea sp. and the sponge Halisarca ectofibrosa (Rungprom et al. 2008; Dahiya and Gautam 2011). Similarly, Bernan et al. (1997) isolated Pseudomonas sp. from the surface of red algae.

Members of the Roseobacter clade are abundant and widespread in marine habitats (Gonzalez and Moran 1997; Rappé et al. 2000; Zubkov et al. 2001; Selje et al. 2004; Martens et al. 2007). Organisms of this group are also associated with cephalopods (Barbieri et al. 2001) or algae (Shiba 1991; Lafay et al. 1995; Ashen and Goff 2000; Grossart et al. 2005). Shiba (1991) isolated Roseobacter litoralis and R. denitrificans from surfaces of green seaweeds. Also, as reported by Wang et al. (2009), bacteria of Roseovarius genus have been isolated from sediment from South China Sea. Ruiz-Ponte et al. (1999) and Wagner-Dobler et al. (2004) reported antibiotic activity from some species of Roseobacter, did not agree with our observations since strain J1 identified as closely related to Roseovarius was found inactive against tested pathogens. It may have another role as reported by others who proposed that members of the Roseobacter lineage play a key role in dimethylsulfoniopropionate (DMSP) cleavage and demethylation/demethiolation and correlated their presence and activity on algal surfaces with DMSP-producing algae (Gonzalez and Moran 1997). In this study, we isolated members of genus Agrobacterium from J. rubens surface; in previous investigation, five species of genus Agrobacterium were isolated from northeastern Atlantic Ocean bottom sediments by Rüger and Höfle (1992). Misawa and Shimada (1998) isolated the crt gene clusters responsible for the biosynthesis of carotenoids, from the marine bacterium Agrobacterium aurantiacum.

Shewanella species represent one of the most numerically abundant microorganisms among readily cultivated marine proteobacteria (Ivanova et al. 2003). Members of this genus have been studied extensively because of their important role in co-metabolic bioremediation of halogenated organic pollutants (Petrovskis et al. 1994), destructive souring of crude petroleum (Semple and Westlake 1987), the dissimilatory reduction of manganese and iron oxides (Myers and Nealson 1988), and their ability to produce high amounts of polyunsaturated fatty acids (Russell and Nichols 1999). Ivanova et al. (2003) reported and characterized new bacteria of the genus Shewanella (Shewanella waksmanii) isolated from the marine sipuncula. In other investigation, Ivanova et al. (2004) characterized Shewanella pacifica, a polyunsaturated fatty acid-producing bacterium isolated from seawater.

Bacteria of genera Pseudomonas, Pseudoalteromonas, Bacillus, Roseovarius and Cytophaga have been consistently isolated from marine algae, while bacteria of the genera Paracoccus, Shewanella and Cytophaga were occasionally isolated from algae surfaces, whereas compared to previous investigations, bacteria belonging to Aquimarina genus are rarely identified from algae and seem to be specific to J. rubens epiphytic communities. Hengst et al. (2010) reported epiphytic bacterial communities living on intertidal seaweeds at the northern coast of Chile and identified bacteria belonging to Aquimarina, using clonal approach. Nedashkovskaya et al. (2005) described a novel Aquimarina muelleri isolated from seawater. Similarly, J10 isolate was identified as closely related to Bermanella, was for the first time isolated from J. rubens surface. The novel gammaproteobacterium Bermanella marisrubri, designed strain RED65T, was isolated from the Red Sea at a depth of 1 m and its genome was sequenced (Pinhassi et al. 2009).

Results of J. rubens isolates screened for antimicrobial potentialities highlighted that bacteria of the genera Bacillus, Aquimarina, Pseudomonas, Pseudoalteromonas and Paracoccus were active against pathogens with a varying and wide antimicrobial spectrum. These findings suggest a beneficial relationship between algae and these epiphytic bacteria which may be involved in preventing fouling. Previous studies have found novel compounds possessing antibiotic activities have been identified from several seaweed-associated bacteria. Our results agree with previous research in which Pseudomonas, Pseudoalteromonas and B. pumilus species isolated from Laminaria saccharina were found to produce antibacterial substances (Wiese et al. 2009). Similarly, Pseudoalteromonas tunicata is known as a successful competitor on marine surfaces due to its ability to produce a number of inhibitory substances. As such P. tunicata has become the model organism for studies of surface colonization and eukaryotic host–bacteria interactions (Thomas et al. 2008).

A natural cyclotetrapeptide cyclo-(isoleucyl-prolyl-leucyl-alanyl) has been isolated from the marine bacteria Pseudomonas sp. and Pseudoalteromonas sp., associated with the seaweed Diginea sp. and the sponge H. ectofibrosa (Rungprom et al. 2008; Dahiya and Gautam 2011) and displayed antibacterial activity against P. aeruginosa and K. pneumoniae, and antifungal activity against pathogenic C. albicans (Dahiya and Gautam 2011). Algicidal bacteria belonging to the genera Alteromonas, Pseudoalteromonas and Cytophaga were also isolated from Ulva sp. and Gelidium sp. Kamei and Isnansetyo (2003) reported lysis of methicillin-resistant S. aureus by 2,4-diacetylphloroglucinol produced by marine algal associated species Pseudomonas sp.

Similarly, Bernan et al. (1997) mentioned that surugatoxin, tetrodotoxin and anhydrotetrodotoxin have been isolated from a microbial source and some of these microbial sources include a Pseudomonas sp. isolated from the surface of red algae. Yasumoto et al. (1986) mentioned that tetrodotoxin (TTX) is produced as a fermentation product of Pseudomonas sp. that was isolated from Jania sp.

Furthermore, marine Bacillus isolates have been found to be able to produce peptide compounds with antimicrobial activity (Jaruchoktaweechai et al. 2000; Barsby et al. 2001; Hentschel et al. 2001). Okarni et al. (1980) discovered a new enzyme that degrades the glucan of Streptococcus mutans, which is the cause of dental caries. The enzyme was isolated from a marine Bacillus and showed optimum activity at 37°C, which makes it favorable for use in the oral cavity. Burgess et al. (2003) isolated several bacteria with high antifouling activity and found that most of these bacteria belonged to Bacillus, such as B. pumilus, B. licheniformis and B. subtilus. Kanagasabhapathy et al. (2006) have isolated B. pumilus and other Bacillus species from different brown alga. Kanagasabhapathy et al. (2008) investigated antimicrobial activity of epiphytic bacteria from several red algae and found that the highest activity was produced by certain Bacillus species especially B. cereus and B. pumilus. Our results are in agreement with these studies since B. pumilus was found active against S. aureus, Vibrio spp. and S. typhimurium.

While it is known that several epiphytic bacteria lose their ability to produce antimicrobial compounds after many subcultures on artificial growth media, in the present report all active isolates, except J9, preserved their abilities to produce inhibitory substances against sensitive pathogens after successive subcultures on marine agar giving the same inhibition of growth of pathogens. However, as mentioned by Penesyan et al. (2010), marine surface-associated microorganisms may require conditions that resemble their native environment in order to produce the maximum amount of bioactives, which is the case for J9, which showed a decrease of activity when grown after several transfers on marine agar. This species may require different growth conditions for optimal production of desired metabolites.

Jania rubens harbors a high diversity of epiphytic bacteria on its surface. In the aim to determine the degree of affinity of J. rubens epibionts to their host, we tested inhibition of Jania organic crude extracts against all isolated bacterial species as well as against human and fish pathogens and bacteria isolated from the surrounding water. Results revealed activity of J. rubens extracts against the pathogens S. aureus, Micrococcus sp., Streptococcus sp. and P. cepacia as well as one of the four surrounding water isolates identified as Shewanella sp., while both dichloromethane and dichloromethane/methanol crude extracts were inactive towards all Jania isolates. Other investigations agree with our finding, and antimicrobial activity of J. rubens has been previously reported; the aqueous extract of J. rubens showed high antibacterial activity against B. subtilis and a low activity against S. aureus (Soliman et al. 1994). Similarly, Karabay-Yavasoglu et al. (2007) found that J. rubens extracts, especially methanol and chloroform extracts, possess antimicrobial activity. Moreover, volatile constituents, fatty alcohols and hydrocarbon fractions isolated from J. rubens displayed varying antibacterial activity against B. subtilis, S. aureus and E. coli (Awad 2002). In view of these results, J. rubens epibionts should be intimately associated with their host.

We conclude that the bacteria isolated from the surface of J. rubens are closely associated with their host and may represent a new source of antimicrobial secondary metabolites highly active against several Gram-negative and Gram-positive pathogens as well as the yeast C. albicans. These epibionts might be beneficial to the algae by limiting or preventing the development of competing or fouling bacteria. Moreover, we reported here an isolate of Aquimarina highly active against the pathogens Staphylococcus, Micrococcus and C. albicans pathogens and which is rarely isolated from red algae and newly isolated from J. rubens surface.