Antimicrobial Activities of Bacteria Associated with the Brown Alga Padina pavonica

Macroalgae belonging to the genus Padina are known to produce antibacterial compounds that may inhibit growth of human- and animal pathogens. Hitherto, it was unclear whether this antibacterial activity is produced by the macroalga itself or by secondary metabolite producing epiphytic bacteria. Here we report antibacterial activities of epiphytic bacteria isolated from Padina pavonica (Peacocks tail) located on northern coast of Tunisia. Eighteen isolates were obtained in pure culture and tested for antimicrobial activities. Based on the 16S rRNA gene sequences the isolates were closely related to Proteobacteria (12 isolates; 2 Alpha- and 10 Gammaproteobacteria), Firmicutes (4 isolates) and Actinobacteria (2 isolates). The antimicrobial activity was assessed as inhibition of growth of 12 species of pathogenic bacteria (Aeromonas salmonicida, A. hydrophila, Enterobacter xiangfangensis, Enterococcus faecium, Escherichia coli, Micrococcus sp., Salmonella typhimurium, Staphylococcus aureus, Streptococcus sp., Vibrio alginoliticus, V. proteolyticus, V. vulnificus) and one pathogenic yeast (Candida albicans). Among the Firmicutes, isolate P8, which is closely related to Bacillus pumilus, displayed the largest spectrum of growth inhibition of the pathogenic bacteria tested. The results emphasize the potential use of P. pavonica associated antagonistic bacteria as producers of novel antibacterial compounds.


INTRODUCTION
In aquatic environments, planktonic microbial communities live as free-living forms or attached to biotic or abiotic surfaces (Martin et al., 2014). In the marine environment, competition for space and nutrients is intense; the surfaces of marine eukaryotes such as seaweeds or invertebrates represent a suitable nutrient-rich habitat for microbial colonization and biofilm formation (Egan et al., 2008;Goecke et al., 2010). As the surfaces of theses eukaryotes are highly complex and differentiated, marine microbial biofilms should constitute a huge source of diversity, and the microbial communities forming them are expected to differ considerably in composition from pelagic microbial communities (Martin et al., 2014). Microbial community living on seaweed surfaces are highly complex, dynamic and consist of bacteria, fungi, diatoms, protozoa, spores and larvae of marine invertebrates (Lachnit et al., 2009(Lachnit et al., , 2011Goecke et al., 2010;Burke et al., 2011). Because especially bacteria interact with their host in multiple and complex ways, they constitute an interesting source of novel bioactive compounds with biotechnological potential (Martin et al., 2014). These bacteria may protect their host from harmful entities present in pelagic areas (Singh and Reddy, 2014) but can also produce compounds important for human and animal health (Kanagasabhapathy et al., 2006;Penesyan et al., 2010).
During summer, the brown alga Padina pavonica is abundantly present along the Tunisian coast. Despite its accessibility, nothing is known about the diversity and biological potential of its epibionts. Seaweeds of the genus Padina were previously reported to harbor epiphytic bacteria with variety of biological activities such as antibacterial activities of epiphytic bacteria isolated from P. arborescens, which was collected from Awaji Island, Japan (Kanagasabhapathy et al., 2006). Bacterial isolates from a close relative of P. pavonica, P. tetrastromatica collected from southeastern India were previously studied for their antibacterial activities against a panel of human pathogens (Chellaram et al., 2013;Ravisankar et al., 2013). However, since the geographic location is amongst the factors that impact on a bacterial community's diversity, composition and secondary metabolites production, the study of the epiphytic bacteria of P. pavonica, which is abundant at the Tunisian coast, was considered to be promising for finding novel antibacterial compounds.
The present study aimed at assessing the culturable bacterial diversity of the community attached to P. pavonica, by isolating and identifying the associated strains, and by testing their potential antimicrobial activity using a variety of pathogens as target organisms.

Sampling
The seaweed P. pavonica (Class: Phaeophyceae, Order: Dictyotales, Family: Dictyotaceae) was collected during two seasons: winter and summer of 2007 from Cap Zebib (37 • 16.2 ′ N, 10 • 3.6 ′ E) at the northern coast of Tunisia. The algae were transferred in dark sterile plastic bags filled with water from the same location.

Isolation of Epiphytic Bacteria
Seaweed samples were washed three times with autoclaved seawater in order to remove free-living and loosely attached bacteria (Burgess et al., 2003). Subsequently, the firmly attached epiphytic bacteria were extracted by vortexing 10 g of algal biomass in 90 ml autoclaved seawater for 6 min. Bacteria were isolated by serial dilution to 10 −3 using autoclaved seawater. From each dilution 100 µl was spread-plated in triplicate on marine agar plates (MA: Pronadisa Laboratories, CONDA). The plates were incubated at 20 • C until colonies appeared or at least for 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 (v/v).

Extraction of DNA and PCR Conditions
Single colonies from the agar plates were suspended in sterile Milli-Q water and used as template in the PCR reactions using the bacterial 16S rRNA gene primer B8F and the universal primer U1492R (Lane, 1991; Table 1). PCR reactions were performed using a Gene Biometra T1 DNA Thermal Cycler (Perkin-Elmer Co. Norwalk-CT, USA) in 25 µl (final volume) reaction mixtures containing 0.1 µl of Hot Star Taq DNA polymerase (Qiagen), 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 (incl. MgCl 2 ) 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 components of the PCR mixture except the DNA template.

Analysis of PCR Products
PCR products (2 µl) were analyzed by agarose gel electrophoresis (1% agarose, 1 × TAE 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.

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 the sequences of the isolates and those of their selected top five best hit blast close relatives was performed using the Neighbor Joining method as available in MEGA version 4.1 software (Tamura et al., 2007).

Nucleotide Sequence Accession Numbers
Partial sequences of the 16S rRNA genes were submitted to GenBank and received the following accession numbers: P1

Microorganisms Used for Antimicrobial Tests
Bacterial type strains (18 strains) used for the evaluation of antimicrobial activity of all isolated epiphytes (and also for alga extracts) were: Aeromonas salmonicida LMG3780, A. hydrophila B3 (

Antimicrobial Assay
Bacterial isolates were screened for their antimicrobial activity against human and fish pathogenic bacteria. The following two methods were used. The drop test assay on Trypto-Casein-Soy agar (TSA, BIO RAD) plates containing 20 g l −1 NaCl was carried out as described by James et al. (1996) and Rao et al. (2005) with slight modifications. Briefly, drops of cell suspension of an overnight culture were spotted onto the agar plates containing a confluent lawn of the target strain (dried for 30 min at 30 • C) and then incubated at 30 • C. The overlay assay was carried out as described by Wiese et al. (2009) with some modifications. Briefly, drops (10 µl) of an overnight culture of the isolates were spotted onto TSA plates supplemented with 20 g l −1 NaCl and incubated for 24 h at 30 • C before they were covered with an overlay containing the target strains in soft agar [3 g l −1 Tryptone Soy Broth-(TSB-BIO RAD, 9 g l −1 agar (PRS, Panreac), 20 g l −1 NaCl, pH 7.2]. The suspension with the target strains contained ∼10 7 cells ml −1 . Growth inhibition was evaluated after incubation (24 h at 30 • C) by measuring the zone of inhibition around the spots with the test organism.
The antibacterial assay of algal crude extracts against its isolates was evaluated by disk diffusion method. Briefly, 500 µg of the total alga crude extract dissolved in dichloromethane and dichloromethane/methanol solvents (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 (10 6 bacteria. ml −1 ) in TSB 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 millimeter) of growth inhibition halo was measured after 24 h incubation considering the disc diameter. Assays were carried out in triplicate. These extracts were also tested in our previous work (Ismail-Ben Ali et al., 2010a) against the 18 indicator bacterial pathogens used in this study and cited above in material and methods.

Preparing Algal Extracts
Fresh algae were dried during 15 days at room temperature in the dark and powdered. About 20 g of dried algae were 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. The extracts were pooled, filtered, and concentrated under reduced pressure in a rotary evaporator. Algal extracts were stored at −20 • C until use.

Preparation of the Cell-Free Supernatant
Ten ml of culture of the isolates P9, P11, P12, and P20 grown for 48 h at 30 • C in TSB was centrifuged at 5000 rpm for 15 min and the supernatants were transferred to new tubes. Activities of the cell-free supernatants were evaluated by the drop method on TSA plates previously inoculated with Micrococcus sp., A. salmonicida LMG3780T or S. aureus. The same protocol was used for isolate P8 using cultures taken after 24, 48, and 72 h of growth.

Protease effect
The enzymatic effect on the antibacterial activity was evaluated on cell-free supernatants of the active isolates. Samples of 1 ml were treated at 37 • C for 2 h with 1 mg ml −1 final concentration of trypsin and pepsin. The residual activities of the samples were assayed by the drop method.

Heat effect
To analyze thermal stability, supernatants were heated to 60 • C for 30 min, 80 • C for 10 min, and 100 • C for 5 min and then cooled to room temperature before being tested for antibacterial activity.

Alga Epiphytic Bacteria Identification
Epiphytic bacteria were extracted from P. pavonica thallus and plated on MA. A total number of 26 isolates was obtained after the first round of cultivation of algal samples. Colony morphology dereplication resulted in 18 bacterial isolates that were obtained in pure culture. Analysis of the 16S rRNA gene sequences showed that all isolates were affiliated to three major groups within the bacterial domain, Alpha-and Gammaproteobacteria, Firmicutes and Actinobacteria ( Table 2). Most of the isolates belongs to the proteobacterial group (twelve isolates: P1-P5, P9, P11, P12, P14, P17, P20, and P26). Two isolates belong to Alphaproteobacteria: Paracoccus (P1) and Devosia (P2) and ten isolates affiliated

Host and Epibionts Affinity
To verify the close relationship between host and isolated bacteria, antibacterial activity of P. pavonica organic crude extracts (polar and non-polar) were tested against the 18 isolates. The results revealed that extracts of the alga harvested in summer exhibited a weak activity (1 mm diameter) against three of the 18 isolates, P6, P7, and P8. While the same extracts had a pronounced inhibitory activity against human and animal pathogens indicator bacteria (9 amongst 18 were inhibited) (Ismail-Ben Ali et al., 2010a; Figure 4).

Epiphytic Bacteria Potential Activity
The 18 isolates were tested for their antimicrobial activities using two different methods and a number of selected pathogens. Five isolates: P8, P9, P11, P12, and P20 showed antibacterial activities against a broad range of pathogens (Table 3). Phylogenetic analysis of the 16S rRNA gene sequences of these active isolates was performed using the Neighbor Joining method (Figure 1). Table 3, isolates P9, P11, P12, and P20 had the same antibacterial spectrum and were effective against nine   indicator strains tested. The antibiotic properties were most effective against the pathogens S. aureus, S. typhymurium, and Micrococcus sp. Growth of the yeast C. albicans was not inhibited by these isolates. Isolate P8 (related to Bacillus pumilus) displayed the strongest antimicrobial activity and was effective against 15 tested bacteria and the yeast C. albicans. A weak inhibition of P8 was observed against V. tapetis and no inhibition of V. anguillarum, P. cepacia, and P. aeruginosa was observed. Inhibition zones depended on the method applied. Generally larger inhibition zones were observed when using the overlay assays compared to the drop assay ( Table 3). The strongest effect was recorded for isolate P8 against S. aureus (inhibition zones varied from 20 to 45 mm; Figure 2). On the basis of its strong activity, isolate P8 was selected to test its antagonistic activity against nine Vibrio species, which are well known marine (fish, shellfish) pathogens. The overlay method was chosen for this test, showing a higher effectiveness when compared to the drop method. As shown in Table 4, two strains of Vibrio species tested were inhibited (V. proteolyticus and V. vulnificus) and furthermore E. xiangfangensis and E. faecium were inhibited as well.

Test of Antibiotic Sensitivity of Isolated Strains
The five isolates with clear antimicrobial activity were further characterized. Their own sensitivity against various antibiotics was investigated. Isolate P8 showed resistance to cefoxitin, oxacillin, and ceftriaxon. P9, P11, and P20 were resistant to amoxicillin, cefoxitin, streptomycin, tetracycline, penicillin G, oxacillin, furans, and ceftriaxone. P11 was also resistant to chloramphenicol. P12 was resistant to amoxicillin, cefoxitin, streptomycin, tobramycin, penicillin G and oxacillin.

Effects of Enzymatic and Heat Treatments on Antibacterial Activity
The cell-free supernatants of cultures of the five active isolates were tested for presence of antimicrobial activity against Micrococcus sp. and S. aureus (the most sensitive species retained). Supernatants of cultures of the 5 isolates inhibited growth of both test strains ( Table 5). The effect of incubation time on the production of active compounds was tested in isolate P8 after 24, 48, and 72 h of incubation. Antibacterial activity was detected for all three incubation periods, with the strongest activity after 48 h incubation, compared by the activity detected after 24 and 72 h.
Results of the sensitivity tests to proteases and heat treatment ( Table 5) showed that bioactive substances produced by isolates P8, P9, P11, P12, and P20 were insensitive to trypsin and pepsin treatments. However, the bioactive substances were thermolabile. Antibacterial activity for isolates P9, P11, P12, and P20, was totally lost after being heated at 60 • C for 30 min.
The active substance produced by isolate P8 was more resistant to increased temperatures since activity was not completely lost at 60 • C but decreased with increasing temperature. About 50% of the diffusion zone remained FIGURE 2 | Plates testing antimicrobial activity of strain P8 isolated from P. pavonica by using the drop method and the overlay assay against A. salmonicida and Staphylococcus aureus. after incubation at 60 • C and 20% at 80 • C, while the activity was totally lost after a treatment for 5 min at 100 • C (Figure 3).

DISCUSSION
The brown alga P. pavonica is widespread along the Tunisian coastline, especially in the northern part (Cap Zebib). This alga can be found up to 10 m depth with a total estimated biomass of 112.5 tons of fresh matter (25 tons dry matter; Ksouri et al., 2008). A previous study by Ktari and Guyot (1999) highlighted the cytotoxic activity of a dichloromethane extract of P. pavonica against KERATIN-forming tumor cell line HeLa (KB cells). Besides, Omezzine et al. (2009), reported antifungal potential of this alga collected from Tunisian coast. Several studies reported interactions between brown algae and their epibionts (Lee et al., 2006). Some brown algaeassociated bacteria may degrade algal polysaccharides, such as fucoidan (Nedashkovskaia et al., 2002), or alginates (Preston et al., 1986;Brown and Preston, 1991). Brown algae may produce biologically active compounds that are capable of killing bacteria (bactericidal) or inhibiting bacterial growth (bacteriostatic; Nagayama et al., 2002).
In this study, a number of 26 isolates were firstly extracted from P. pavonica thallus at a first round cultivation on MA, then restricted to 18 after subculturing to obtain pure morphological different colonies since algal samples were subjected to a harsh pre-treatment in order to be sure that only attached bacteria will be isolated. Also in other previous study by Kanagasabhapathy et al. (2006), numbers of culturable isolates from some brown algae collected from Awaji Island Japan; were 10 for Padina arborescens and 21, 19, 17, 15, 12, 9, 8, and 5 for Ecklonia cava, Colpemenia sinusa, Sargassum serratifolium, S. fusiforme, Petalonia fascia, Scytosiphon lomentaria, S. filicinum, and Undaria pinnatifida respectively.
In other previous study by Albakosh et al. (2016) on the brown alga Splachnidium rugosum, a number of 41 isolates were obtained using 4 different medium; MA, nutrient sea water agar, nutrient agar and thiosulfate-citrate-bile-salts-sucrose agar. In this same study by Albakosh et al. (2016), initially 29 isolates were obtained on MA medium and after identification, the number was narrowed down to 22 unique isolates. From our results and these previous work and observations it seems that MA is a suitable medium to isolates algal epiphytic bacteria as compared to other medium. However, the number of culturable epiphytic bacteria will increase when using different culturing media.
After characterization, Gram-negative Alpha-and Gammaproteobacteria and Gram-positive Firmicutes and Actinobacteria were identified from P. pavonica. Most of the isolates belonged to Gram-negative bacteria, as was the case for epibionts of seaweeds (Lakshmanaperumalsamy and Purushothaman, 1982;Albakosh et al., 2016). Among the different P. pavonica isolates, isolates related to the genera Paracoccus, Acinetobacter, and Staphylococcus were reported in previous studies isolated from marine environments and   seaweeds (Gallardo et al., 2004;Schulze et al., 2006;Li, 2009;Zeng et al., 2010). Bacteria of the genera Planomicrobium were previously isolated from seawater and sediment and have algicidal effect, affecting harmful algal blooms (Giudice et al., 2007;Zeng et al., 2010). The actinobacterial genus Brevibacterium (P6, P19) is found in marine and terrestrial environments (Jones and Keddie, 1953;Collins, 1992) and were previously also isolated from degraded thallus of the brown alga Fucus evanescens (Ivanova et al., 2004). Two of the P. pavonica associated isolates reported in this study, P2 and P26, closely related to the Proteobacteria: Devosia and Erwinia, respectively, were previously not known from the surfaces of marine seaweeds. Species of the genus Devosia have been isolated from beach sediment (Lee, 2007), from freshwater legume Neptunia natans (Rivas et al., 2003), as endosymbionts of the marine ciliate Euplotes magnicirratus (Vannini et al., 2004), from a nitrifying inoculum (Vanparys et al., 2005) and from soil and greenhouse soil (Yoo et al., 2006;Yoon et al., 2007). Bacteria of the genus Erwinia are known as plant pathogens (Paulert et al., 2007) and common inhabitants of the rhizosphere of cultivated crops and weeds (Burr and Schroth, 1977). Besides, in this study we notice that P9, P11, P14, P17, and P27 isolates may be considered as potentially novel species since they were related with less than 95% identity to Pseudomonas sp. (P9, P11), Pseudoalteromonas sp. (P14), and Acinetobacter sp. (P17). Isolate P27 may be a potentially novel genus, since it is less than 90% identical with its closest relative Staphylococcus. These isolates may be of biotechnological interest and need further characterization.
P. pavonica harbors several culturable epiphytic bacteria on its surface and our harsh isolation procedure indicates that these epibionts are intimately associated with their host. We tested the resistance of the isolates to crude extracts of the P. pavonica. Only extracts from P. pavonica harvested in summer revealed a low inhibitory effect toward three of the isolates. Organic crude extracts of P. pavonica were toxic against human and fish pathogens such as V. tapetis, V. anguillarum, V. alginoliticus, Ps. cepacia, A. salmonicida, A. hydrophila, E. faecalis ATCC 29212, and S. aureus (Ismail-Ben Ali et al., 2010a). In Figure 4 a diagram is depicted showing the number of pathogenic bacteria inhibited by P. pavonica extracts, which represent 50% of the total number of pathogens tested (Ismail-Ben Ali et al., 2010a), compared to P. pavonica isolates that are inhibited by their host extracts, which represent only 16.6% of the total isolates tested. Chbani et al. (2011) and Zerrin et al. (2011) also reported toxicity of P. pavonica against pathogenic indicator bacteria.
In this study 5 of the 18 isolates possessed antimicrobial activities. In our previous work we isolated 10 bacteria from the green alga U. intestinalis with two strains showing antimicrobial potential and 19 from J. rubens with 7 strains displaying antibacterial activity against pathogenic indicator bacteria (Ismail-Ben Ali et al., 2010bAli et al., , 2012. Many Pseudoalteromonas species are associated with eukaryotic hosts in the marine environment (Holmström and Kjelleberg, 1999). Such species have been isolated from a variety of animals, such as mussels, tunicates and sponges, as well as from a range of marine plants (Holmström and Kjelleberg, 1999) and from the red alga Jania rubens collected from Tunisian coast (Ismail-Ben Ali et al., 2012). These Pseudoalteromonas species displayed different extracellular biological activities ranging from antibacterial, agarolytic, antiviral, antifouling, to algicidal activities (Sawabe et al., 1998;Holmström and Kjelleberg, 1999;Yuexin et al., 2010). Pseudoalteromonads may play an important role in the control of harmful algal blooms within the marine environment by the production of an extracellular inhibitor that causes rapid lyses of algal species within the genera Chattonella, Gymnodinium, and Heterosigma (Lovejoy et al., 1998). Pseudomonas, Pseudoalteromonas and Vibrio species have also been isolated from the brown alga Laminaria saccharina and possess antibacterial substances (Wiese et al., 2009). Previously, we isolated Pseudoalteromonas, Pseudomonas and B. pumilus species from the surface of the red alga J. rubens collected from the Tunisian coast. These isolates showed a varying spectrum of antimicrobial activity (Ismail-Ben Ali et al., 2012). Vibrio sp., Pseudomonas sp. and B. pumilus are known as probiotic bacteria used in aquaculture (Hill et al., 2009). Epiphytic Vibrio strains produced various antibacterial compounds (Oclarit et al., 1994). However, in our study 13 strains of bacteria that were isolated from the surface of P. pavonica did not show any antibacterial activity. These isolates may have another ecological function. For instance, they may have a protective role by releasing compounds into the surrounding seawater that would help preventing extensive fouling of the surface (Burgess et al., 1999;Kanagasabhapathy et al., 2008).
Bacteria of the genus Bacillus, to which our isolate strain P8 belongs, have been reported to be associated with various marine organisms including brown algae (Thakur and Anil, 2000). Lee et al. (2006) described B. psychrodurans as a representative strain associated with the brown alga Undaria pinnatifida. B. pumilus was isolated by Wiese et al. (2009) from the surfaces of the brown alga L. saccharina as well as from the surface of other brown algae such as Sargassum fusiforme, Colpomenia sinuosa, and Ecklonia cava (Kanagasabhapathy et al., 2006). Similarly, Parvathi et al. (2009) isolated B. pumilus strains from sediment, crab, fish, starfish and oyster. In the Table 6 the main studies on B. pumilus isolated from algae and other sources are summarized and the biological activities and compounds responsible for these inhibitory activities are reported. Bacillus is well known for producing antimicrobial compounds (Wiese et al., 2009). Since 1955, B. pumilus was described as antibiotic producer; Pumilin is an antibiotic isolated from this bacterial strain by Bhate (1955) and it is mainly active against gram positive microorganisms. Various metabolites with antibiotic activity, including the important group of antibiotic peptides like polymixin B, gramicidin and bacitracin are produced by Bacillus species (Crisley, 1964;Vandamme and Demain, 1976;Ishihara et al., 2002). Results obtained in this study are in agreement with these reports. P8 bacteria showed a broad inhibition spectrum against human and fish pathogens and the strongest inhibitory activity was observed against S. aureus and A. salmonicida. Bacillus sp. isolated from seaweeds and sediment by Suvega and Arunkumar (2014) had strong activities against bacteria plant pathogens (Xanthomonas axonopodis pv. citri, X. oryzae pv. oryzae) and fungi (Ustilaginoidea virens) causing diseases in rice paddy fields. In our previous study (Ismail-Ben Ali et al., 2012) a B. pumilus related strain was isolated from the red alga J. rubens and had a different antibacterial spectrum than the isolate from P. pavonica, confirming that the algal species not only determines epiphytic diversity but also affects the type of secondary metabolite that is produced, even among the same species of epiphyte. Kumar et al. (2014) reported antibacterial activities cell free supernatant of a culture of a Bacillus strain isolated from entomopathogenic nematode against E. coli. Only a few epiphytic bacteria isolated from algae are known to have inhibitory potential against the human pathogen E. coli. Inhibition of E. coli by isolates, P9, P11, P12, and P20 and especially by P8 suggests that these isolates may produce novel antimicrobial substances. P8 inhibits both Gram+ and Gram-pathogens. This finding is supported by other studies such as those of Chatterjee et al. (2008) and Prieto et al. (2012) who reported antibacterial activities of Bacillus strains isolated from reef sediments and marine alga against P. aeruginosa, S. aureus, Salmonella enterica, Bacillus sp., Proteus vulgaris, E. coli, and S. typhimurium. Gram negative isolates such as Pseudomonas and Pseudoalteromonas strains (P9, P10, P11, and P20) have variable antibacterial activities toward both Gram+ and Gram− tested pathogens. Qi et al. (2009) andHorta et al. (2014) reported Gram− marine isolates (from Bifurcaria bifurcata and deep see sediment) such as P. rhizosphaerae, Vibrio sp., Alteromonas sp., Shewanella sp., and Serratia sp. active against Gram+ pathogens.
Antimicrobial tests were carried out by two different methods, the drop technique and the overlay assay. The overlay assay was better suitable for the detection of antibacterial and antifungal activity of bacterial isolates. Spots of isolates that are grown overnight may already have secreted bioactive compounds in the agar before they were overlaid by the target strain. Therefore, it was concluded that the five active isolates closely related to Bacillus, Pseudomonas and Pseudoalteromonas, produced and excreted bioactive(s) inhibitory substance(s) during 24 h growth. When these organisms were dropped simultaneously with the target organism the zone of inhibition was much less apparent. Perhaps the target organism prevented growth of the test organism or its production or excretion of the bioactive compound.
Inhibitory compounds produced by the five active isolates were present in the cell-free spent medium. The inhibitory compounds produced by the Pseudomonas and Pseudoalteromonas isolates were resistant to proteolytic enzymes (trypsin and pepsin) and sensitive to heat treatment. Inhibitory compounds produced by isolate P8 were also resistant to proteolytic enzymes and the activity was only lost after heat treatment for 5 min at 100 • C.
In summary, the five isolates P8, P9, P11, P12, and P20, isolated from the surface of Padina pavonica and identified as closely related to B. pumilus, Pseudomonas sp., and Pseudoalteromonas produce bioactive compounds that act against human pathogens such as S. aureus; the most pathogenic specie of the genus Staphylococcus and of which multi-resistant strains represent a major problem in hospital (Le Loir et al., 2003). They also act against V. alginolyticus, which is one of the major causes of bacterial infections on shrimp farms (Hill et al., 2009), and against the virulent Gram-bacterium V. vulnificus causing gastroenteritis, necrotic skin infections, meningitis and fatality (Mayer et al., 2014), as well as against a number of fish pathogens. Given the broad range of pathogens that are inhibited, these isolates may produce compounds that might prove promising candidates for potentially novel antibiotic and antifungal drugs.

AUTHOR CONTRIBUTIONS
AI performed all the experiments, analyzed and interpreted data, wrote and revised the manuscript. LK contributed in designation of the research plan, organization of the study, the development of work and corrected and evaluated the manuscript. MH assisted the molecular manipulations and sequencing, contributed to the blast, phylogentic analysis. HB was involved in design of the experiments, evaluation of outcome and involved in manuscript writing. MC did the Vibrio strains identification, contributed to data analyzes, interpretation and manuscript final finalization. AB and LS revised critically the manuscript and approved final version to be published. ME designed the research plan, organized and supervised the development of work, corrected and evaluated the manuscript, submitted the full sequences for assignment in GENBANK and supervised the whole work.

FUNDING
This work was co-funded and supported by the two projects "MOTOX" and "BIOMER" from the National Institute of Marine Sciences and Technologies (INSTM).