Abstract

Black Band Disease (BBD) is a well-described disease plaguing corals worldwide. It has been established that ecological and environmental stress factors contribute to the appearance and progression of the disease, believed to be caused by a diverse microbial consortium. We have identified and characterized Vibrio sp. associated with BBD in Eilat reef corals using both culture-dependent and -independent methods. Direct sampling using 16S rRNA gene clone libraries showed seasonal dynamics in the diversity of BBD-associated Vibrios. In the two sampling periods, BBD-associated Vibrio clones showed similarities to different groups: October samples were similar to known pathogens, while December samples were similar to general aquatic Vibrio sp. Cultured bacterial isolates of Vibrio sp. were highly homologous (≥99%) to previously documented BBD-associated bacteria from the Caribbean, Bahamas and Red Seas, and were similar to several known coral pathogens, such as Vibrio coralliilyticus. The proteolytic activity of Vibrio sp., as measured using casein- and azocasein-based assays, directly correlated with temperature elevation and peaked at 26–28 °C, with the microorganisms producing more proteases per bacterial cell or increasing the rate of proteolytic activity of the same proteases (potentially metalloproteases). This activity may promote coral tissue necrosis and aid in ensuing progression of the coral BBD.

Introduction

Black Band Disease (BBD) is a widespread and well-described disease plaguing corals worldwide (Antonius, 1985; Bruckner & Bruckner, 1997; Dinsdale, 2002). Several ecological and environmental factors can contribute to the appearance and progression of BBD (Kuta & Richardson, 2002). When active, BBD affects <1% of the total coral reef (Kuta & Richardson, 1996; Dinsdale, 2002); nevertheless, the disease is considered to be a major factor for coral reef decline, due to the lack of recovery and regeneration of the affected corals (Kuta & Richardson, 1997; Edmunds, 2000). Indeed, Bruckner & Bruckner (1997) reported that once BBD penetrates a reef, it will persist and spread throughout the population, slowly causing the demise of susceptible coral.

Even though BBD has been documented in coral reefs for over 30 years, there is still very little unequivocal information regarding its etiology, and many aspects of this disease still need clarification. BBD symptoms include a dark-red or black band mat that moves across a coral colony, destroying healthy tissues and exposing the calcium carbonate skeleton. Environmental parameters, such as water depth and clarity, as well as temperature, are known to be correlated with disease prevalence and dispersal (Rützler et al., 1983; Antonius, 1985; Kuta & Richardson, 2002). BBD is most active during the summer when water temperatures are elevated, reaching temperatures above 28 °C, and yet disappears almost completely during winter months, when the water temperature declines below 20 °C (Rützler et al., 1983; Kuta & Richardson, 1996).

The BBD mat is made up of a diverse community that includes cyanobacteria, sulfur reducers and oxidizers, vibrios and numerous other microbial groups (Garrett & Ducklow, 1975; Rützler & Santavy, 1983; Bythell et al., 2002; Cooney et al., 2002; Frias-Lopez et al., 2002, Frias-Lopez et al., 2004; Viehman & Richardson, 2002; Richardson, 2004; Sekar et al., 2006; Sussman et al., 2006; Barneah et al., 2007). This community is distinct from that of healthy corals or of healthy tissues from diseased corals (Pantos et al., 2003; Barneah et al., 2007). The morphology and chemical characteristic of the black band were found to be similar to typical microbial mats (Carlton & Richardson, 1995; Richardson, 1997). The fact that BBD communities all contained populations of the same (physiologically) functional groups of microorganisms led Carlton & Richardson (1995) to propose that BBD is caused by a pathogenic microbial consortium and that there is no primary pathogen.

BBD in the Red Sea corals exhibits an aggregated distribution, which provides evidence for local transmission of the disease, which does not necessarily require direct contact (Zvuloni et al., 2009). These authors also suggested that the mechanism of BBD transmission is likely to be a water-borne infection. In addition, although the BBD phenotype disappears during the winter, the infected corals contribute to the reintroduction of the reef in the following summer. In Red Sea corals affected by BBD, the mat is made up mainly of cyanobacteria, sulfate-reducing bacteria (SRB), and several strains of Vibrios (Barneah et al., 2007). Because several well-characterized coral diseases are known to be caused by pathogenic strains of Vibrio (Kushmaro et al., 1996, Kushmaro et al., 1997; Ben-Haim & Rosenberg, 2002; Ben-Haim et al., 2003a, Ben-Haim et al., 2003b; Cervino et al., 2004), it is likely that pathogenic strains of these bacteria may also play a role in BBD. This study, therefore, focused on the characterization of the Vibrio sp. in the BBD consortium and evaluation of their possible role in the disease, using microbiological and molecular tools.

Materials and methods

Sample collection

Samples from BBD-affected corals were collected from the Eilat coral reef [Inter-University Institute (IUI) for Marine Science, Northern Red Sea, 29°51′N, 34°94′E] during October and December 2007. The seven sampled Favia sp. corals were numbered and marked to enable repeated samplings. October samples were collected while the disease was very active and the black band was morphologically evident. The ensuing December samples were collected from the interface area (i.e. between the healthy tissue and the exposed skeleton) of the same marked corals, even when the band was not evident. Mucus and tissue samples were collected using 10-mL syringes and immediately transported to the laboratory. Water temperatures were measured in situ adjacent to the colonies in October, 26 °C, and in December, 24 °C (IUI monitoring).

Bacterial isolation and selection media

In the laboratory, the collected volume was spun down by centrifuge and the resuspended pellet was used for cultures and DNA extraction. For isolation, several different selective isolation media were tested: thiosulfate citrate bile sucrose agar (TCBS) with a pH indicator for Vibrio selection (Merck), Marine Broth and Marine Agar 2216 (HiMedia Laboratories, Mumbai, India) plates (10% of the recommended final concentration) were used for isolation of bacteria of marine origin (DSMZ GmbH) and Desulfovibrio medium (M63) (http://www.dsmz.de/index.html) for isolation and enrichment of anaerobic SRB. Marine agar (MA) and M63 media were modified using sea salt water [NaCl 4% (w/v)] and used as liquid and for 1.5% agar or 0.8% gellan gum (Fluka, Sigma-Aldrich, St. Louis, MO) plates. M63 cultures were incubated in an anaerobic chamber (BD Gas Pak EZ). All media were incubated at room temperature (22–24 °C).

DNA extraction

Total genomic DNA was extracted from BBD environmental samples using a MoBio Power Soil DNA isolation kit (MoBio Laboratories Inc., Solana Beach, CA) or using the physical bead-shearing technique and precipitation. Silica beads (0.5 mm diameter) in 300 μL double distilled water were used for the shearing, achieved by a 10-min vortex step at maximum speed. The precipitation lasted 10 min at maximum centrifugation speed (13 200 g).

Primer design and PCR amplification

Total DNA was amplified using a Biometra TGradient thermocycler (Biometra, Göttingen, Germany) by PCR, using specific modified 16S rRNA gene primers for bacteria: 8F (GGATCCAGACTTTGATYMTGGCTCAG, modified by shortening from the 5′-end; Falske et al., 1997) and 907R (CCGTCAATTCCTTTRAGTTT; Lane et al., 1985). The reaction mixture included 12.5 μL ReddyMix (ABgene, Surrey, UK), 1 μL of 10 mM of each primer (forward and reverse), 1 μL of 25 mM bovine serum albumin, 1–2 μL of the sample genomic DNA (environmental, 5–90 ng μL−1; culture, 20–100 ng μL−1) and water to bring the total volume to 25 μL. An initial denaturation hot start of 4 min at 95 °C was followed by 30 cycles of the following incubation pattern: 94 °C for 30 s, 54 °C for 30 s and 72 °C for 1 min. In the case of clone library building, a final extension at 72 °C for 30 min concluded the reaction.

Clone library construction and sequencing

The heterologous 16S rRNA gene PCR products were purified from the gel using the Wizard PCR Prep kit (Promega, Madison, WI), cloned into the pCRII-TOPO-TA cloning vector as specified by Invitrogen (Carlsbad, CA) and transformed into BioSuper CaCl2-competent HD5αEscherichia coli cells (Bio-Lab, Israel), according to the manufacturer's instructions. Clones were checked for inserts by PCR amplification using M13-Forward and Reverse primers (Invitrogen) and sent to McLab (San Francisco, CA) for sequencing.

Sequence analysis

The sequences were aligned using clustalw found in the molecular evolutionary genetics analysis package (mega, version 4.0; Tamura et al., 2007). Positions not sequenced in all isolates or with alignment uncertainties were removed. Phylogenetic trees were constructed using the neighbor-joining method by the mega package. The blast database and Ribosomal Database Project II (RDPII) (Cole et al., 2005) were used to classify all 16S rRNA gene sequences.

Metabolic tests

Culture strains were characterized biochemically using the API20 NE and API ZYM systems, according to the manufacturer's instructions (bioMerieux SA, Marcy l'Etoile, France). Oxidase activities of the strains were examined using an oxidase test kit (Fluka, Sigma-Aldrich), according to the manufacturers' instructions. Catalase activity was tested using 3% analytical H2O2.

Sensitivity to common commercial antibiotics was tested according to the disk diffusion assay (Awerbuch et al., 1989), using Oxoid antimicrobial susceptibility test disks (Oxoid, Hampshire, UK) containing the following antibiotics: tetracycline, sulfamethoxazole trimethoprim, nalidixic acid, erythromycin, chloramphenicol, methicillin and ampicillin.

Proteinase assay

Casein test

Modified calcium caseinate agar (Sigma-Aldrich) was used for a protease test. Calcium caseinate agar (2 × with sea water) was autoclaved and added to 1% skim milk (2 ×), and heated at 70 °C for 1 h to prepare casein plates. Bacterial isolates were spread evenly and incubated overnight at different temperatures to observe lysed areas produced by extracellular proteinases (Chantawannakul et al., 2002).

Azocasein test and EDTA inhibition

The main principle of the azocasein test is the separation of the azo-molecule from the casein protein during azocasein degradation by bacterial extracellular proteases. The azo-molecules released in this reaction have a unique absorption of ∼420 nm. Chosen isolates were grown overnight in marine broth medium at different temperatures (i.e. 20, 22, 24, 26, 28 and 30 °C). The cultures (1.5 mL each) were centrifuged to precipitate the bacteria and 1 mL of each supernatant was added to 1% azocasein in 0.1 M Tris buffer (pH 8) for 1 h of incubation at different temperatures. The reaction was stopped using 2 mL of 5% trichloroacetic acid and left for 20 min at 4 °C. The solution was centrifuged and an identical volume of 1 M NaOH was added to the supernatant. Absorption of the azo-molecules was measured at 420 nm (Vazquez et al., 1995). Relative protease activity units (RAU) were calculated using the following formula (Denkin & Nelson, 1999):  

formula

Bacterial isolate supernatants were tested for inhibition of protease activities at different concentrations of EDTA (from 0.1 to 50 mM). Samples were incubated for 1 h at 30 °C and then tested for proteolytic activity by the azocasein assay described above. A control sample contained isolate supernatant with no EDTA.

Nucleotide sequence accession number

The 16S rRNA gene sequences from this study have been deposited in the NCBI GenBank database under accession numbers GQ215061GQ215096 for the isolate sequences and GQ215097GQ215227 for the clone library sequences.

Results

BBD-infected Favia sp. corals were found on the reef near the IUI shore of the Gulf of Eilat. Diseased Favia sp. corals were sampled in October and December 2007 from the black band. Subsequently, 148 sequences were generated from appropriate 16S rRNA gene libraries. Diversity assessment on the class level in black band samples collected in October and December revealed an abundance of Cyanobacteria (32% at both time points), 31% and 22%Alphaproteobacteria, respectively, followed by 11% and 8%Gammaproteobacteria, a class that includes the Vibrio sp., respectively (Fig. 1). Only four classes were common to both time periods, namely Flavobacteria, Cyanobacteria, Alpha- and Gammaproteobacteria. Bacteroidetes and Clostridia classes that appeared in October libraries disappeared in December, when several new classes appeared (e.g. Sphingobacteria, Lentisphaerae and Deltaproteobacteria). The proportion of unclassified bacteria had grown from 5% to 17% in the December samples.

1

Bacterial diversity (b class) obtained from clone libraries constructed with universal 16S rRNA gene primers from diseased corals of samples collected in (a) October (n=85) and (b) December (n=64) 2007. Classification was performed using RDPII and blast.

1

Bacterial diversity (b class) obtained from clone libraries constructed with universal 16S rRNA gene primers from diseased corals of samples collected in (a) October (n=85) and (b) December (n=64) 2007. Classification was performed using RDPII and blast.

Five ribotypes from October and one ribotype from December of the Gammaproteobacteria class were found to be highly similar (≥99%) to Vibrio species (Fig. 2). October clones Oct07-4BB-60, Oct07-5BB-67, 70 and 72, and Oct07-6BB-81 were highly homologous to several well-known pathogenic Vibrios (e.g. EU372929 – white syndrome pathogen Vibrio sp. PM6A and AJ440004 –Vibrio coralliilyticus), and clones that were previously documented in BBD (EF123487 – BBD-associated Gammaproteobacteria clone). December clone Dec07-1BB-21 was highly homologous (≥99) to previously documented BBD clone BB2INT16SI-5 (EF089440; Barneah et al., 2007) found in the interface between the healthy apparent tissue adjacent to the black band. Other December clones were highly similar (100%) to different aquatic Vibrio sp. (AM159567, EF587962, AJ630202, etc.; data not shown).

2

Phylogenetic trees of the Gammaproteobacteria class representatives of the BBD clone library constructed using the 16S rRNA gene and environmental genomic DNA of samples obtained from the marked corals in October and December 2007. The numbers in parentheses indicate the total number of similar clones from the same coral colony on the basis of ≥99% identity for each representative sequence. ●– October 2007, coral BB area; ▪– December 2007, coral BB area; unmarked – clones with high homology to GenBank entries. WS, white syndrome.

2

Phylogenetic trees of the Gammaproteobacteria class representatives of the BBD clone library constructed using the 16S rRNA gene and environmental genomic DNA of samples obtained from the marked corals in October and December 2007. The numbers in parentheses indicate the total number of similar clones from the same coral colony on the basis of ≥99% identity for each representative sequence. ●– October 2007, coral BB area; ▪– December 2007, coral BB area; unmarked – clones with high homology to GenBank entries. WS, white syndrome.

A total of 110 bacterial isolates from the October 2007 samples and 45 isolates from the December 2007 samples were obtained by culturable methods using TCBS, MA and Desulfovibrio media. All isolated bacteria were obtained from the seven marked coral colonies taken from diseased tissue. No Vibrio isolates were obtained using TCBS plates from the apparently healthy area of the coral. Alignment analysis of the 16S rRNA genes identified 88% and 93% of isolates as Gammaproteobacteria from the October and December samples, respectively. At the genus level, most Gammaproteobacteria belonged to the genus Vibrio (65% and 87%) and Beneckea (15% and 5%) in the October and December samples, respectively.

Phylogenetic tree analysis clustered the sequences from isolates retrieved from different time periods and different corals into five separate Vibrio subclusters (Fig. 3). All sequences showed high homology (≥99%) to diverse Vibrio sp., most of which were previously associated with different coral diseases. The sequences of cluster 1 (isolated from marked corals 1, 3, 6 and 7 at different time periods) displayed high similarity to V. coralliilyticus (AJ440004), a well-known coral pathogen that causes bleaching and tissue lysis (Ben-Haim et al., 2003a), and to a BBD clone (EF123487) from the Caribbean. The sequences of clusters 2 and 4 (Fig. 3) showed high resemblance to apparent Vibrio clones (EF089442 and EF089451) previously retrieved from BBD by Barneah et al. (2007) showing high homology to Vibrio sp. (AY876051) associated with the brown band syndrome from the Great Barrier Reef. Cluster 3 showed isolates highly similar to Vibrio shiloi (AF007115), a known pathogen of Mediterranean coral Oculina patagonica that causes coral bleaching (Kushmaro et al., 1996, Kushmaro et al., 1997). Cluster 5 included an isolate similar to a pathogen Photobacterium eurosenbergii (AJ842345) associated with coral bleaching (Thompson et al., 2005). The last Alphaproteobacteria cluster includes isolates from different corals and time periods that were similar to previously studied BBD clones (i.e. DQ446157 and EF123485; Sekar et al., 2006).

3

Phylogenetic tree of bacterial isolates representatives from Favia sp. corals infected with the BBD, together with best-matched sequences from GenBank. Subtrees are marked by genus or class (according to RDPII, blast). ●– October 2007, coral BB area; ▪– December 2007, coral BB area; unmarked — clones with high homology to GenBank entries.

3

Phylogenetic tree of bacterial isolates representatives from Favia sp. corals infected with the BBD, together with best-matched sequences from GenBank. Subtrees are marked by genus or class (according to RDPII, blast). ●– October 2007, coral BB area; ▪– December 2007, coral BB area; unmarked — clones with high homology to GenBank entries.

Out of the 110 isolates obtained in the October 2007 sampling, 20 isolates were chosen for further research. The choice was made according to isolate abundance and similarity to a known coral pathogen (metabolic and antibiotic resistance test results are summarized in Supporting Information, Tables S1 and S2).

Bacterial proteolytic activity is known to be a key virulent characteristic of pathogenic Vibrio sp. (Filkelstein & Hanne, 1982). In this study, we used the qualitative casein test and the semi-quantitative azocasein test to assess the proteolytic activities of our strains. Following incubation at 20 °C, no lysed area was observed on the casein agar plates around the bacterial colonies (Fig. 4), although several isolates showed slight activity beneath the colony itself. As the temperature increased, so did the activity area around all tested isolates (Fig. 4). Isolates MA-3BB-3 and M63-7BB-2 display relatively high activity levels, with the latter being the most active, taking over the entire Petri dish upon overnight incubation at 30 °C (Fig. 4).

4

Protease activity of Vibrio sp. isolates as detected on casein agar plates. The figure summarizes the diameter of the lysed area under and around the bacterial colony, as a function of temperature. Fifty microliters of bacterial culture, grown overnight in marine broth (∼109 CFU) at 22°C, was placed on casein agar plates and incubated at different temperatures overnight.

4

Protease activity of Vibrio sp. isolates as detected on casein agar plates. The figure summarizes the diameter of the lysed area under and around the bacterial colony, as a function of temperature. Fifty microliters of bacterial culture, grown overnight in marine broth (∼109 CFU) at 22°C, was placed on casein agar plates and incubated at different temperatures overnight.

The azocasein test allows us to quantify protease activity as a function of CFU. Seventy-five percent of isolates showed relatively moderate or high proteolytic activity. Several examples of this activity are exemplified in Fig. 5. Isolate M63-7BB-2 showed a relatively high activity at all temperatures and peaked around 26–28 °C. Isolate MA-3BB-3 was linearly dependent on temperature between 20 and 24 °C, showing equivalent maximal magnitudes at 25 °C (Fig. 5). Similar activity was observed with isolate MA-2BB-3, with the activity peaking at 28 °C. Isolate MA-5BB-3 was almost inactive at all the temperatures tested. Isolates TCBS-6BB-11, TCBS-6BB-12 and TCBS-7BB-3 showed moderate protease activity at all temperatures, peaking around 26 °C (data not shown). Isolate TCBS-6BB-7 had the most distinct profile of all, showing no activity at low temperatures, linear dependence with temperature between 20 and 24 °C, a clear peak of activity at 24 °C and a sharp decrease above this value (Fig. 5).

5

Proteinase (azocasein) assay of representative Vibrio sp. isolates. RAU vs. temperature are plotted. The values were calculated using the azocasein formula and mean CFU at a specific temperature.

5

Proteinase (azocasein) assay of representative Vibrio sp. isolates. RAU vs. temperature are plotted. The values were calculated using the azocasein formula and mean CFU at a specific temperature.

In addition, the most active isolate M63-7BB-2 was grown overnight at three different temperatures (22, 26 and 30 °C) and extracellular proteases secreted by the strain were tested as a function of temperature (measured between 20 and 30 °C at intervals of 2 °C). Extracellular proteases secreted at 22oC showed a high shift in the relative activity, in comparison with the proteases secreted during high-temperature growth (26 and 30 °C) (data not shown) that display low temperature dependence.

Extracellular proteolytic activity was inhibited using EDTA and bacterial isolate supernatants, as revealed by the azocasein assay. Eighty-seven percent of active (75% moderate and highly active) isolates were totally inhibited by 50 mM EDTA. Others showed partial, but significant inhibition. For example, isolate M63-7BB-2 showed 34% inhibition with 0.1 mM EDTA, 51% inhibition with 5 mM EDTA, up to 68% inhibition with 10 mM and 81% inhibition with 50 mM EDTA.

Protease activity can be characterized as a function of time and temperature. Three isolates were grown overnight at two different temperatures, namely 20 and 30 °C. The growth media, including extracellular proteases, were incubated with azocasein at the same temperatures. The reaction was interrupted every 10 min and the sample absorption was measured. The slope in the figure of the relative activity as a function of time represents the relative activity rate of proteases in 1 mL culture as a function of temperature, CFU and the amount of protein. Isolate M63-7BB-2 showed high activity at all the temperatures tested. Even at 20 °C, it was more active (1.0471 RAU min−1) than isolate MA-3BB-3 at 30 °C (0.9608 RAU min−1). At 30 °C, the former multiplied its activity by 1.53-fold (a=1.6128 RAU min−1). Isolates MA-2BB-2 and MA-3BB-3 showed somehow low rates of 0.3654 and 0.3016 RAU min−1 at 20 °C, and 1.2324 and 0.9608 RAU min−1 at 30 °C, respectively.

Discussion

In this study, we focused on the specific microbial group of Vibrio sp. appearing in BBD-affected corals in the Red Sea. Our BBD samples were collected from the same corals in October 2007 (when the disease was very active and the black band was evident, at a water temperature of 26 °C) and again during December 2007 (when the water temperature declined to 24 °C and the black band disappeared). Our findings on the general microbial diversity from the 16S rRNA gene libraries obtained from direct sampling during October and December 2007 were also compared with the findings of Barneah et al. (2007). These earlier results showed that diseased corals contained mainly Cyanobacteria (25%) and Alphaproteobacteria (30%) sequences, which were comparable to values of 32% and 27%, respectively, in the present study. In addition, 10% of the sequences from the clone libraries (Figs 1 and 2) demonstrated high similarity (≥99%) to Gammaproteobacteria, in general, and to Vibrio genus sequences reported in Barneah et al. (2007). These findings strengthen our hypothesis that Vibrio sp. plays a role in BBD appearance and progression. October and December 2007 Vibrio sequences showed similarities to different Vibrio sp. groups. While the October samples included clones similar to known coral pathogens (Fig. 2), the December clones were similar to general aquatic Vibrio sp. sequences.

In addition to clone library sequences, Vibrio sp. originating from the BBD coral mucus and tissue samples from October and December 2007 were collected and cultured (Fig. 3). No Vibrio sp. were isolated from the apparently healthy area of the coral, probably due to their low concentration in the coral tissue and mucus. In addition, Vibrio sp. found in the BB area are distinct from those found in the apparently healthy area (Barneah et al., 2007). The possibility exists, therefore, that the TCBS selective medium was not suitable for isolation of these strains. Some of the Vibrio strains isolated in this study were highly similar (≥99%) to BBD Vibrio clones identified previously by Barneah et al. (2007) using molecular techniques. Moreover, on examining the sequences, we noticed high similarity between isolates from different coral colonies sampled in the same time period. Similarities were also found between isolates from samples collected from the same coral, but at different times (see Fig. 3). These findings suggested that some Vibrio sp. may be ubiquitously associated with the diseased coral, both in the symptomatic and in the asymptomatic state. Furthermore, we isolated several strains (9%) that were highly homologous (>99%) to V. coralliilyticus, a coral pathogen known to cause bleaching and tissue lysis at elevated temperatures in the branching coral Pocillopora damicornis (Ben-Haim et al., 2003a). We also succeeded in isolating several bacterial strains that were similar to other Vibrio sp. associated with known coral diseases, such as the brown band syndrome (Great Barrier Reef, AY876051) and coral bleaching (Mediterranean Sea, AF007115; Kushmaro et al., 1996) (Fig. 3). Additional Gammaproteobacteria ribotypes were obtained in this study. These P. eurosenbergii-like isolates were found to belong to a group known to be associated with coral bleaching (Thompson et al., 2005). Interestingly, we also isolated a group of Alphaproteobacteria, a class of bacteria that appear in all BBD studies to date (Cooney et al., 2002; Frias-Lopez et al., 2003; Richardson, 2004; Barneah et al., 2007). This class also includes several known pathogens and particularly a specific coral pathogen, Aurantimonas coralicida, which is known to cause white plague type II disease in several coral hosts (Denner et al., 2003) and that is often a natural inhabitant of corals (Rohwer et al., 2001, Rohwer et al., 2002). This pathogen is a constant feature of BBD mats in several locations in the world and, therefore, should be taken into consideration in assessing BBD etiology.

Vibrios are known pathogens with characteristic pathogenic and virulence properties (Thompson et al., 2004). One characteristic of pathogenic Vibrios is the activity of their extracellular proteases. This activity could be a mechanism by which the coral tissue is penetrated and degraded (Ben-Haim & Rosenberg, 2002; Ben-Haim et al., 2003b; Halpern et al., 2006; Sussman et al., 2008). Two protease activity tests were conducted in this study, namely the casein and azocasein tests. All isolates tested by the casein test displayed elevated temperature-dependent activity (Fig. 4). We found significant differences (Kruskal–Wallis nonparametric test, followed by multiple comparisons analysis, P<0.05) between activity at 20 °C, and at the other tested temperatures (i.e. 26, 28 and 30 °C), as well as between 30 °C and the other tested temperatures (i.e. 22 and 24 °C). Interestingly, isolate M63-7BB-2 showed the highest levels of proteolytic activity on casein agar plates (Fig. 4). This enhanced activity was reinforced by the additional azocasein proteolytic test (Fig. 5). In the azocasein test, most of the isolates' proteolytic activity peaked at around 26–28 °C. This finding is consistent with BBD appearance, which is associated with elevated water temperatures (Rützler et al., 1983; Kuta & Richardson, 1996). Accordingly, we can speculate that when water temperatures increase, the general proteolytic activity of the black band microbial mat also increases, probably due, at least in part, to the increased activity of Vibrio sp. (in addition to the increase in the biomass). This phenomenon could explain the increased progression of the disease during periods of high temperatures, and may shed light on the mechanism of the disease. It is also possible to attribute the shift in bacterial virulence properties as a function of temperature to the different natures/types of the extracellular proteases secreted at different incubation times.

EDTA inhibition indicates that the extracellular proteases are a family of metalloproteases. This protein family was previously identified with several Vibrio sp. pathogens, such as the human pathogen Vibrio cholerae, V. coralliilyticus that causes bleaching in P. damicornis coral and Vibrio sp. associated with Indo-Pacific coral white syndrome (Ben-Haim et al., 2003b; Halpern et al., 2006; Sussman et al., 2009).

The proteolytic activity of the M63-7BB-2 isolate, which may be an important component of BBD, was quantified as a function of time and temperature using ‘relative activity units.’ Our results demonstrated that this strain was highly active, showing 1.0471 RAU min−1 at 20 °C and 1.6128 RAU min−1 at 30 °C. The strain thus either produces more extracellular protease or enhances protease activity at the higher temperature. This isolate may be a major contributor to the tissue necrosis observed in BBD; however, its activity on coral tissue in situ still needs to be assessed. Additional isolates, for example MA-2BB-2 and MA-3BB-3, also increased their protease activity between 20 and 30 °C (by more than threefold) and may also play a role in tissue necrosis. Our results indicate a shift in bacterial virulence as a function of temperature.

We showed that proteolytic activity increases with temperature between 20 and 30 °C (Fig. 4), either due to the production of more proteases per bacterial cell (Fig. 5) or by an increase in the rate of proteolytic activity of the same proteases. The shift in the proteolytic activity usually occurred between 20 and 24 °C, peaking around 24–28 °C. These features correlate highly with the appearance and progression of the BBD in the Red Sea corals (Zvuloni et al., 2009).

Carlton & Richardson (1995) suggested that the microbial consortium in the BBD mat operates synergistically to exploit a source of nutrients, as the BBD migrates across the coral. They suggested that tissue degradation releases a rich variety of organic nutrients for use by the mat population. Our findings regarding the possible increase in the proteolytic activity of BBD-associated Vibrio sp. as a function of elevated temperatures support this hypothesis. Such proteolytic activity could, therefore, play a major role in the observed black band progression as a function of elevated seawater temperature, as described previously (Rützler et al., 1983; Kuta & Richardson, 1996).

Acknowledgements

This research was supported by ISF grant no. 1167/07 and Levi-Eshkol scholarship to N.S. from the Israeli Ministry of Science, Culture and Sports. We would like to thank D. Ben-Yossef, O. Barneah and O. Dotan for their help in sample collection and IUI in Eilat for use of their facilities.

References

Antonius
A
(
1985
)
Coral diseases in the Indo-Pacific: a first record
.
PSZNI Mar Ecol
 
6
:
197
218
.
Awerbuch
TE
Lustman
L
Piret
JM
(
1989
)
A numerical method to determine minimal inhibitory concentrations (MICs) of antibiotics directly from disc-diffusion susceptibility tests
.
J Microbiol Meth
 
9
:
1
7
.
Barneah
O
Ben-Dov
E
Kramarsky-Winter
E
Kushmaro
A
(
2007
)
Characterization of the black band disease in the Red sea stony corals
.
Environ Microbiol
 
9
:
1995
2006
.
Ben-Haim
Y
Rosenberg
E
(
2002
)
A novel Vibrio sp. pathogen of the coral Pocillopora damicornis
.
Mar Biol
 
141
:
47
55
.
Ben-Haim
Y
Thompson
FL
Thompson
CC
Cnockaert
MC
Hoste
B
Swings
J
Rosenberg
E
(
2003a
)
Vibrio coralliilyticus sp. nov., a temperature-dependent pathogen of the coral Pocillopora damicornis
.
Int J Syst Evol Micr
 
53
:
309
315
.
Ben-Haim
Y
Zicherman-Keren
M
Rosenberg
E
(
2003b
)
Temperature-regulated bleaching and lysis of coral Pocillopora damicornis by the novel pathogen Vibrio coralliilyticus
.
Appl Environ Microb
 
69
:
4236
4242
.
Bruckner
AW
Bruckner
RJ
(
1997
)
The persistence of black-band disease in Jamaica: impact on community structure
.
Proc 8th. Int Coral Reef Symp
 
1
:
601
606
.
Bythell
JC
Barer
MR
Cooney
RP
Guest
JR
O'Donnell
AG
Pantos
O
(
2002
)
Histopathological methods for the investigation of microbial communities associated with disease lesions in reef corals
.
Lett Appl Microbiol
 
34
:
359
364
.
Carlton
R
Richardson
LL
(
1995
)
Oxygen and sulfide dynamics in a horizontal migration cyanobacterial mat: black band disease of corals
.
FEMS Microbiol Ecol
 
18
:
155
162
.
Cervino
JM
Hayes
RL
Polson
SW
Polson
SC
Goreau
TJ
Martinez
RJ
Smith
GW
(
2004
)
Relationship of Vibrio species infection and elevated temperatures to yellow blotch/band disease in Caribbean corals
.
Appl Environ Microb
 
70
:
6855
6864
.
Chantawannakul
P
Oncharoen
A
Klanbut
K
Chukeatirote
E
Lumyong
S
(
2002
)
Characterization of proteases of Bacillus subtilis strain 38 isolated from traditionally fermented soybean in Northern Thailand
.
Science Asia
 
28
:
241
245
.
Cole
JR
Chai
B
Farris
RJ
Wang
Q
Kulam
SA
McGarrell
DM
Garrity
GM
Tiedje
JM
(
2005
)
The Ribosomal Database Project (RDP-II): sequences and tools for high-throughput rRNA analysis
.
Nucleic Acids Res
 
33
:
D294
D296
.
Cooney
RP
Pantos
O
Le Tissier
MDA
Barer
MR
O'Donnell
AG
Bythell
JC
(
2002
)
Characterization of the bacterial consortium associated with black band disease in coral using molecular microbiological techniques
.
Environ Microbiol
 
4
:
401
413
.
Denkin
SM
Nelson
DR
(
1999
)
Induction of protease activity in Vibrio anguillarum by gastrointestinal mucus
.
Appl Environ Microb
 
65
:
3555
3560
.
Denner
EBM
Smith
GW
Busse
HJ
Schumann
P
Narzt
T
Polson
SW
Lubitz
W
Richardson
LL
(
2003
)
Aurantimonas coralicida gen. nov., sp. nov., the causative agent of white plague type II on Caribbean scleractinian corals
.
Int J Syst Evol Micr
 
53
:
1115
1122
.
Dinsdale
EA
(
2002
)
Abundance of black-band disease on corals from one location on the Great Barrier Reef: a comparison with abundance in the Caribbean region
.
Proc 9th Int Coral Reef Symp
 
2
:
1239
1243
.
Edmunds
P
(
2000
)
Recruitment of scleractinians onto the skeletons of corals killed by black band disease
.
Coral Reefs
 
19
:
69
74
.
Falske
A
Rheims
H
Wolterink
A
Stackebrandt
E
Akkermans
AD
(
1997
)
Ribosome analysis reveals prominent activity of an uncultured member of the class Actinobacteria in grassland soil
.
Microbiology
 
143
:
2983
2989
.
Filkelstein
RA
Hanne
LF
(
1982
)
Purification and characterization of the soluble haemagglutinin (cholera lectine) produced by Vibrio cholerae
.
Infect Immun
 
36
:
1199
1208
.
Frias-Lopez
J
Zerkle
AL
Boneheyo
GT
Fouke
BW
(
2002
)
Partitioning of bacterial communities between sea water and healthy black band disease and dead coral surfaces
.
Appl Envion Microb
 
68
:
2214
2228
.
Frias-Lopez
J
Bonheyo
GT
Jin
Q
Fouke
BW
(
2003
)
Cyanobacteria associated with coral black band disease in Caribbean and Indo-Pacific reefs
.
Appl Environ Microb
 
69
:
2409
2413
.
Frias-Lopez
J
Klaus
JS
Bonheyo
GT
Fouke
BW
(
2004
)
Bacterial community associated with black band disease in corals
.
Appl Environ Microb
 
70
:
5955
5962
.
Garrett
P
Ducklow
H
(
1975
)
Coral diseases in Bermuda
.
Nature
 
253
:
349
350
.
Halpern
M
Raats
D
Lavion
R
Mittler
S
(
2006
)
Dependent population dynamics between chironomids (nonbiting midges) and Vibrio cholerae
.
FEMS Microbiol Ecol
 
55
:
98
104
.
Kushmaro
A
Loya
Y
Fine
M
Rosenberg
E
(
1996
)
Bacterial infection and coral bleaching
.
Nature
 
380
:
396
.
Kushmaro
A
Rosenberg
E
Fine
M
Loya
Y
(
1997
)
Bleaching of the coral Oculina Patagonica by Vibrio AK-1
.
Mar Ecol Prog Ser
 
147
:
159
165
.
Kuta
KG
Richardson
LL
(
1996
)
Abundance and distribution of Black Band Disease of corals in the northern Florida Keys
.
Coral Reefs
 
15
:
219
223
.
Kuta
KG
Richardson
LL
(
1997
)
Black Band Disease and the fate of diseased coral colonies in the Florida Keys
.
Proc 8th Int Coral Reef Symp
 
1
:
575
578
.
Kuta
KG
Richardson
LL
(
2002
)
Ecological aspects of Black Band Disease of corals: relationships between disease incident and environmental factors
.
Coral Reefs
 
21
:
393
398
.
Lane
DJ
Pace
B
Olsen
GJ
Stahlt
DA
Sogint
ML
Pace
NR
(
1985
)
Rapid determination of 16S ribosomal RNA sequences for phylogenetic analyses
.
P Natl Acad Sci USA
 
82
:
6955
6959
.
Pantos
O
Cooney
RP
Le Tissier
MDA
Barer
MR
O'Donnell
AG
Bythell
JC
(
2003
)
The bacterial ecology of a plague-like disease affecting the Caribbean coral Montastrea annularis
.
Environ Microbiol
 
5
:
370
382
.
Richardson
LL
(
1997
)
Occurrence of the black band disease cyanobacterium on healthy corals of the Florida Keys
.
B Mar Sci
 
61
:
485
490
.
Richardson
LL
(
2004
)
Black band disease
.
Coral Health and Disease
  (
Rosenberg
E
Loya
Y
eds), pp.
325
336
.
Springer-Verlag
,
Heidleberg, Germany
.
Rohwer
F
Breitbart
M
Jara
J
Azam
F
Knowlton
N
(
2001
)
Diversity of bacteria associated with the Caribbean coral Montastraea Franksi
.
Coral Reefs
 
20
:
85
91
.
Rohwer
F
Seguritan
V
Azam
F
Knowlton
N
(
2002
)
Diversity and distribution of coral-associated bacteria
.
Mar Ecol Prog Ser
 
243
:
1
10
.
Rützler
K
Santavy
D
(
1983
)
The black band disease of Atlantic reef corals, I. Description of the cyanophyte pathogen
.
PSZNI Mar Ecol
 
4
:
301
319
.
Rützler
K
Santavy
D
Antonius
A
(
1983
)
The black band disease of Atlantic reef corals. III. Distribution, ecology and development
.
PSZNI Mar Ecol
 
4
:
329
358
.
Sekar
R
Mills
DK
Remily
ER
Voss
JD
Richardson
LL
(
2006
)
Microbial communities in the surface mucopolysaccharide layer and the black band microbial mat of black band-diseased Siderastrea sidereal
.
Appl Environ Microb
 
72
:
5963
5973
.
Sussman
M
Bourne
DG
Willis
BL
(
2006
)
A single cyanobacterial ribotype is associated with both red and black bands on diseased corals from Palau
.
Dis Aquat Organ
 
69
:
111
118
.
Sussman
M
Willis
BL
Victor
S
Bourne
DG
(
2008
)
Coral pathogens identified for White Syndrome (WS) Epizootics in the Indo-Pacific
.
PloS ONE
 
3
:
e2393
.
Sussman
M
Mieog
JC
Doyle
J
Victor
S
Willis
BL
Bourne
DG
(
2009
)
Vibrio zinc-metalloprotease causes photoinactivation of coral endosymbionts and coral tissue lesions
.
PloS ONE
 
4
:
e4511
.
Tamura
K
Dudley
J
Nei
M
Kumar
S
(
2007
)
MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0
.
Mol Biol Evol
 
24
:
1596
1599
.
Thompson
FL
Iida
T
Swings
J
(
2004
)
Biodiversity of Vibrios
.
Microbiol Mol Biol R
 
68
:
403
431
.
Thompson
FL
Thompson
CC
Naser
S
Hoste
B
Vandemeulebroecke
K
Munn
C
Bourne
D
Swings
J
(
2005
)
Photobacterium rosenbergii sp. nov. and Enterovibrio coralii sp. nov., vibrios associated with coral bleaching
.
Int J Syst Evol Micr
 
55
:
913
917
.
Vazquez
SC
Rios Merino
LN
MacCormack
WP
Fraile
ER
(
1995
)
Protease-producing psychrotrophic bacteria isolated from Antarctica
.
Polar Biol
 
15
:
131
135
.
Viehman
S
Richardson
LL
(
2002
)
Motility patterns of Beggiatoa and Phormidium corallyticum in black band disease
.
Proc 9th Int Coral Reef Symp
 
2
:
1251
1255
.
Zvuloni
A
Artzy-Randrup
Y
Stone
L
Kramarsky-Winter
E
Barkan
R
Loya
Y
(
2009
)
Spatio-temporal transmission patterns of black-band disease in a coral community
.
PLoS ONE
 
4
:
e4993
.

Supporting Information

Supporting Information

Author notes

Editor: Riks Laanbroek