Adaptation of cyanobacteria to the sulfide-rich microenvironment of black band disease of coral


  • Present address: Jamie L. Myers, Miami Institute for Human Genomics, University of Miami Miller School of Medicine, Miami, FL 33136, USA

  • Editor: Gary King

Correspondence: Jamie L. Myers, Miami Institute for Human Genomics, 12500 SW 152nd St, UM South Campus Building C, Miami, FL 33177, USA. Tel.: +1305 243 1044; fax: +1305 969 3563; e-mail:


Black band disease (BBD) is a cyanobacteria-dominated microbial mat that migrates across living coral colonies lysing coral tissue and leaving behind exposed coral skeleton. The mat is sulfide-rich due to the presence of sulfate-reducing bacteria, integral members of the BBD microbial community, and the sulfide they produce is lethal to corals. The effect of sulfide, normally toxic to cyanobacteria, on the photosynthetic capabilities of five BBD cyanobacterial isolates of the genera Geitlerinema (3), Leptolyngbya (1), and Oscillatoria (1) and six non-BBD cyanobacteria of the genera Leptolyngbya (3), Pseudanabaena (2), and Phormidium (1) was examined. Photosynthetic experiments were performed by measuring the photoincorporation of [14C] NaHCO3 under the following conditions: (1) aerobic (no sulfide), (2) anaerobic with 0.5 mM sulfide, and (3) anaerobic with 0.5 mM sulfide and 10 μM 3-(3′,4′-dichlorophenyl)-1,1-dimethylurea (DCMU). All five BBD cyanobacterial isolates tolerated sulfide by conducting sulfide-resistant oxygenic photosynthesis. Five of the non-BBD cyanobacterial isolates did not tolerate sulfide, although one Pseudanabaena isolate continued to photosynthesize in the presence of sulfide at a considerably reduced rate. None of the isolates conducted anoxygenic photosynthesis with sulfide as an electron donor. This is the first report on the physiology of a culture of Oscillatoria sp. found globally in BBD.


Black band disease (BBD) is a pathogenic microbial community that infects corals worldwide (Richardson, 2004). It is characterized as a dark-colored mat, several millimeters to centimeters wide, that typically migrates across coral colonies at an average rate of 3 mm day−1 killing live coral tissue and exposing coral skeleton (Antonius, 1982; Rützler et al., 1983). The dominant BBD mat-building organisms in terms of biomass are filamentous nonheterocystous cyanobacteria (Rützler & Santavy, 1983; Rützler et al., 1983; Taylor, 1983). The community also includes sulfate-reducing bacteria (Garrett & Ducklow, 1975; Viehman et al., 2006), sulfide-oxidizing bacteria (Garrett & Ducklow, 1975; Mitchell & Chet, 1975; Ducklow & Mitchell, 1979; Richardson, 1996), >60 species of heterotrophic bacteria (Cooney et al., 2002; Frias-Lopez et al., 2002, 2004; Sekar et al., 2006, 2008), and marine fungi (Ramos-Flores, 1983).

Both microscopic and molecular studies, have shown that there are multiple genera and species of cyanobacteria in BBD. Rützler et al. (1983) identified three cyanobacteria, Schizothrix calciola Schizothrix mexicana, and Spirulina subsalsa, in BBD samples using light and scanning electron microscopy (SEM). Ducklow & Mitchell (1979) reported detection of Oscillatoria and Spirulina sp. in BBD based on analysis with SEM. A molecular study by Cooney et al. (2002) reported the presence of Oscillatoria in BBD from the Caribbean, and Frias-Lopez et al. (2002) reported both Trichodesmium and Oscillatoria sequences in BBD samples from the Caribbean and Papua New-Guinea. Previous studies by our group revealed members of the genera Geitlerinema, Leptolyngbya, Oscillatoria, and Phormidium (Myers et al., 2007) as well as Lyngbya hieronymusii (Sekar et al., 2006) in BBD from the wider Caribbean and the Philippines using molecular techniques. Of particular interest is the finding that most of the molecular studies of BBD, which include samples from five regions of the wider Caribbean [Barbados, the Bahamas, Curaçao, the Florida Keys, and St Croix, US Virgin Islands (USVI)], the Philippines, the Northern Red Sea, and the Indo-Pacific, have detected one 16S rRNA gene sequence that corresponds to one Oscillatoria sp. (Cooney et al., 2002; Frias-Lopez et al., 2002, 2003, 2004; Sekar et al., 2006, 2008; Sussman et al., 2006; Barneah et al., 2007; Myers et al., 2007). Although one of these reports (Sussman et al., 2006) stated that an isolate of this BBD Oscillatoria was successfully cultured from BBD in the Indo-Pacific, no physiological studies have been reported from this group and the culture was ultimately lost (D.G. Bourne, pers. commun.).

BBD microorganisms, including cyanobacteria, are exposed to sulfide-rich conditions within the band, the result of sulfidogenesis by BBD sulfate-reducing bacteria that grow in the anaerobic microniche within the BBD microbial mat using sulfate from seawater. Sulfide has been shown experimentally to cause coral tissue lysis at concentrations measured within the band (Richardson et al., 1997) and is thus considered to be involved in BBD pathogenicity. A study that used microelectrodes showed that the oxygen and sulfide dynamics in BBD are virtually identical to those found in other cyanobacterial mats in sulfide-rich environments (Carlton & Richardson, 1995). During the day, the oxygen/sulfide interface is typically 200–300 μm below the band surface (Carlton & Richardson, 1995), with the microenvironment below the interface being completely anoxic and sulfide-rich. Because the BBD microbial mat is typically only 1 mm deep, much of the microenvironment of the band during the day would be expected to include both sulfide and light (Kühl & Jørgensen, 1992; Kühl et al., 1994; Oren et al., 1995).

Sulfide is toxic to most oxygenic photosynthetic organisms, including cyanobacteria, because it irreversibly blocks photosystem II (PSII) (Cohen et al., 1986; Garcia-Pichel & Castenholz, 1990). Previous studies have determined that there are four photosynthetic responses of cyanobacteria to sulfide (Cohen et al., 1986). The most common response, exhibited by most cyanobacteria, is sulfide-sensitive oxygenic photosynthesis, in which sulfide exposure results in the total and permanent inhibition of photosynthesis by blocking electron transport associated with PSII (Cohen et al., 1986). This inhibition may be a result of sulfide targeting the donor side of PSII (Oren et al., 1979; Miller & Bebout, 2004) or cytochromes on the acceptor side of PSII (Cohen et al., 1986). Less common are three strategies by which cyanobacteria can tolerate sulfide. A few cyanobacteria conduct sulfide-resistant oxygenic photosynthesis, which allows the continuation of oxygenic photosynthesis during exposure to sulfide. Cyanobacteria that can utilize this strategy include Synechococcus lividus, Synechococcus elegans, and Oscillatoria terebriformis (reclassified as a member of the genus Geitlerinema; Castenholz, 2001), which were all isolated from sulfide-rich hot spring outflows (Castenholz, 1977; Richardson & Castenholz, 1987). The second strategy is the partial inhibition of oxygenic photosynthesis combined with stimulation of anoxygenic photosynthesis (Cohen et al., 1986), in which sulfide is an electron donor to photosystem I (PSI). The ability to conduct simultaneous oxygenic and anoxygenic photosynthesis allows for efficient CO2 assimilation by cyanobacteria in environments that exhibit diel transitions between oxic and anoxic conditions. Cyanobacteria that utilize this strategy include Microcoleus chthonoplastes and Oscillatoria amphigranulata (Castenholz & Utkilen, 1984; Cohen et al., 1986; Garcia-Pichel & Castenholz, 1990). The third strategy is the total inhibition of PSII (oxygenic photosynthesis) and stimulation of PSI to allow anoxygenic photosynthesis using sulfide as an electron donor (Cohen, 1984; Miller & Bebout, 2004). An example of a cyanobacterium that exhibits this strategy is Oscillatoria limnetica (Cohen, 1984), which has been reclassified as Geitlerinema PCC9228 (Castenholz, 2001). Based on the illuminated, sulfide-rich nature of the BBD environment, it seems likely that BBD cyanobacteria must possess one of the relatively rare abilities to tolerate and/or utilize sulfide.

In previous studies by our group (Richardson & Kuta, 2003; Myers et al., 2007), we demonstrated that three BBD cyanobacteria isolated into culture were capable of conducting sulfide-resistant oxygenic photosynthesis in the presence of 0.5 mM sulfide. These consisted of two strains of Geitlerinema, one isolated from the Bahamas (accession no. EF110974) and the other from the Florida Keys (accession no. DQ151461), and one Leptolyngbya strain (accession no. EF110975) from the Florida Keys. To further understand the ecology and etiology of BBD, it would be worthwhile to determine the extent of sulfide tolerance of additional strains of BBD cyanobacteria. Sulfide tolerance may influence the ability of coral reef cyanobacteria to colonize the nutrient-rich band. Thus, the sulfide tolerance of BBD cyanobacteria should be compared with that of cyanobacteria isolated from non-BBD sources on the reef to determine whether BBD cyanobacterial species are uniquely adapted to the sulfide-rich BBD environment. Such studies require isolation of different species and strains of coral reef cyanobacteria into culture to allow physiological studies.

In the present study, we examined the effect of sulfide on the photosynthetic capabilities of five new BBD cyanobacterial isolates. Thus, in addition to the three strains that were previously investigated, as discussed above (Richardson & Kuta, 2003; Myers et al., 2007), our studies now include physiological results for eight cultured BBD cyanobacteria. The five new isolates include three additional isolates of Geitlerinema, one additional Leptolyngbya isolate, and an isolate of the apparently ubiquitous BBD Oscillatoria. The three additional Geitlerinema BBD isolates and the additional Leptolyngbya BBD isolate were characterized molecularly in Myers et al. (2007) and the physiological data for these isolates are presented here. The photosynthetic capabilities of six mat-forming non-BBD cyanobacteria, isolated from the surfaces of apparently healthy coral colonies and sandy areas on reefs of the Florida Keys, were also assessed.

Materials and methods

Sample collection

BBD samples were collected from reefs of the Northern Florida Keys, Lee Stocking Island in the Bahamas, St Croix in the USVI, and the Philippines (Table 1). Non-BBD samples were collected from four reefs in the Northern Florida Keys along 17 km of the reef tract (Table 1). The six nonpathogenic cyanobacterial mats were obtained from sediment patches on coral colonies (n=5) and a mat overlying the benthic sediment (n=1). These non-BBD samples were representative of typical nonpathogenic cyanobacterial patches and mats as assessed in a previous (microscopy based) study of cyanobacteria on Florida reefs (Richardson, 1997). All of the samples were collected using sterile, needleless 10-mL syringes while SCUBA diving. Freshly collected samples were stored in a low light environment at ambient temperature until arrival at the laboratory, where the samples were transferred to test tubes containing sterile seawater or ASN III medium, a mineral medium that supports photoautotrophic marine algal growth (Rippka et al., 1979).

Table 1.   Cyanobacteria isolated into culture from BBD samples and non-BBD samples from the Northern Florida Keys, Lee Stocking Island in the Bahamas' Exuma Chain, St Croix in the USVI, and Negros Island in the Philippines
Isolate ID
(accession no.)
Source (host coral, reef, region, and sample date)Sequence
length (bp)
Closest relative in the GenBank database% similarity
  • *

    Partial sequence provided by M. Gantar and R. Sekar not submitted to the GenBank database.

  • Lyngbya–Phormidium–Plectonema group.

  • FLK, Northern Florida Keys; LSI, Lee Stocking Island; NI, Negros Island.

BBD cyanobacterial isolates
 HS223 (DQ680351)BBD on Siderastrea siderea, Horseshoe Reef, LSI, 8/041439Geitlerinema sp. BBD (DQ151461)99
 W-1 (EF154084)BBD on Siderastrea siderea, Watson's Reef, FLK, 5/051396Geitlerinema sp. BBD (DQ151461)99
 P2b-1 (EF372580)BBD on Porites lutea, Agan-an, NI, 8/051461Geitlerinema sp. BBD (DQ151461)99
 P2b-2 (EF372581)BBD on Porites lutea, Agan-an, NI, 8/051454Leptolyngbya sp. PCC7375 (AB039011)98
 101-1BBD on Diploria strigosa, Frederiksted Reef, St Croix, USVI, 6/07356*Uncultured BBD Oscillatoria (AF473936)99
Non-BBD cyanobacterial isolates
 AR-1 (EF372583)Cyanobacterial mat on sediment, Algae Reef, FLK, 6/041460Leptolyngbya sp. 0BB30S02 (AJ639892)92
 9 (EU196364)Cyanobacterial mat on Colpophyllia natans, Horseshoe Reef, FLK, 6/041464Uncultured Leptolyngbya sp. (DQ917878)94
 62-2 (EU223007)Cyanobacterial mat on Dendrogyra cylindrus, Conch Shallow, FLK, 6/041461Uncultured Leptolyngbya sp. (DQ917878)93
 73-2 (EU196366)Cyanobacterial mat on Montastraea cavernosa, North North Dry Rocks, FLK, 6/041426Phormidium sp. MBIC10003 (AB058201)LPP-group MBIC10012 (AB058209)9797
 63-1 (EF110976)Cyanobacterial mat on Dendrogyra cylindrus, Conch Shallow, FLK, 6/041460Pseudanabaena sp. PCC7367 (AB039018)98
 72-1 (EU196365)Cyanobacterial mat on Montastraea cavernosa, North North Dry Rocks, FLK, 6/041435Pseudanabaena sp. PCC7367 (AB039018)98

Isolation of cyanobacteria

The gliding method described by Castenholz (1988) was used to isolate cyanobacteria from each sample. Single cyanobacterial filaments were cut out of agar plates using sterile watchmaker's forceps and introduced into test tubes containing sterile ASN III medium. Inoculated tubes were covered with paper towels for 2–3 weeks in order to reduce the incident light intensity and prevent photoinhibition (Richardson & Kuta, 2003). Once enough biomass had accumulated to allow for cyanobacterial clumping, the paper towels were removed. Cyanobacterial cultures were maintained in a temperature-controlled incubator at 30 °C with a light source of 30 μmol photons m−2 s−1 on a 12 : 12 h light/dark cycle (Richardson & Kuta, 2003). Cultures were transferred every 3 weeks to fresh culture media.


Photographs of cyanobacterial isolates were taken using a Leitz DMR microscope (Rockleigh, NJ) equipped with a Leica DC500 digital camera, which was connected to a computer with digital imagining software. All images were taken at × 1000 magnification.

Extraction of DNA and 16S rRNA gene amplification and sequencing

DNA from the BBD cyanobacterial isolates (with the exception of the BBD Oscillatoria) was extracted and sequenced in a previous study that focused on molecular identification of BBD cyanobacteria (Myers et al., 2007). For the present study, DNA was extracted from the BBD Oscillatoria and non-BBD cyanobacterial isolates using the FastDNA® SPIN kit for soil (Q-Biogene, Vista, CA) as described previously (Mills et al., 2003; Sekar et al., 2006).

Cyanobacterial 16S rRNA genes were amplified from the DNA extracts with the bacterial primers 27F (5′-AGA GTT TGA TCM TGG GTC AG-3′) and 1492R (5′-TAC GGY TAC CTT GTT ACG ACT T-3′) (Muyzer et al., 1995) and then used in cloning and sequencing reactions as described in Myers et al. (2007). The sequence data for the non-BBD cyanobacterial isolates have been submitted to the GenBank database under accession nos EF110976, EF372583, EU196364, EU196365, EU196366, and EU223007, which are listed in Table 1. Sequences were analyzed using the blastn queuing system ( to identify their closest relatives in the GenBank database.

Sulfide-tolerant photosynthesis experiments

Experiments were performed to compare the abilities of the 11 cyanobacterial isolates to conduct photosynthesis under five different experimental conditions: (1) light, aerobic, no sulfide (baseline oxygenic photosynthesis); (2) light, anaerobic with the addition of 0.5 mM sulfide (sulfide-tolerant photosynthesis); (3) light, anaerobic with the addition of 0.5 mM sulfide and DCMU [10 μM 3-(3′,4′-dichlorophenyl)-1,1-dimethylurea] (anoxygenic photosynthesis); (4) light, aerobic with the addition of 10 μM DCMU (control); and (5) dark, aerobic (second control). The experimental condition of sulfide plus DCMU assesses the ability to conduct ‘DCMU-forced’ anoxygenic photosynthesis, in which DCMU blocks electron flow in PSII (from H2O as electron donor) while allowing donation of electrons to PSI. One potential electron donor to PSI is sulfide.

Photosynthesis was measured by photoincorporation of [14C] NaHCO3. Triplicate 20-mL scintillation vials were used for each experimental condition, and experiments were repeated for a total of three times for each of the 11 isolates. Aerobic vials were prepared by adding 5 mL of sterile ASN III medium to sterile 20-mL scintillation vials. Anaerobic vials were prepared as described in Myers et al. (2007). For experiments involving the addition of sulfide, a sterile needle was used to introduce sulfide (from a stock solution of 0.1 M Na2S·9H2O) through the septum of each vial to produce a final concentration of 0.5 mM (Myers et al., 2007). DCMU was added to a final concentration of 10 μM.

Cyanobacterial cultures that had been incubated overnight in fresh media were used to inoculate all photosynthesis experiments. For those experiments that included sulfide, cyanobacterial cultures were exposed to 0.5 mM sulfide for 6 h before experimental inoculation to allow for sulfide adaptation. Samples were inoculated into experimental vials as described in Myers et al. (2007). After inoculation, [14C] NaHCO3 was added (through the septum for the anaerobic vials) to a final specific activity of 0.05 μCi mL−1. Experimental vials were incubated for 2 h at 30 °C at a light intensity of 150 μmol photons m−2 s−1 in a temperature-controlled incubator (Myers et al., 2007).

After 2 h, photoincorporation was stopped by adding formalin to each vial to a final concentration of 1.5%. Each sample was then filtered onto a 934-AH glass fiber filter, followed by rinsing with sterile ASN III medium and a 2.0% HCl solution, which was used to remove unincorporated inorganic 14C. Each filter was inserted into a 7-mL scintillation vial containing an Ecolume scintillation cocktail (MP Biomedicals, Solon, OH) and counted using a Beckman LS-6500 liquid scintillation counter (Beckman Coulter, Fullerton, CA). The amount of photoincorporated 14C was expressed as kilo counts per minute (kCPM) per biomass by use of cyanobacterial dry weights. The dry weights were obtained from triplicate vials containing ASN III medium (without [14C] NaHCO3) that were inoculated at the same time as the experimental vials. The contents of these three vials were immediately filtered onto preweighed filters, dried for 48 h at 55 °C, weighed (with preweight values subtracted), and averaged. The photoincorporation values of the dark controls were subtracted from the light-dependent values.

Photosynthesis vs. irradiance

The cyanobacterial isolates that were able to tolerate sulfide, as determined in the experiments described above, were further investigated to assess photosynthetic capability in terms of light intensity. Experiments were conducted in a photosynthetron (Lewis & Smith, 1983), which allows measurement of photosynthesis vs. irradiance at a range of light intensities in a single experiment, as described previously in Myers et al. (2007). Photosynthesis vs. irradiance (P vs. I) experiments were performed for the two conditions in which photosynthesis was detected in the first set of experiments, i.e. under aerobic (no sulfide) conditions and under anaerobic conditions with 0.5 mM sulfide. Experimental vials were inoculated (as described in Myers et al., 2007), placed in the photosynthetron, and incubated for a 2-h time period at 30 °C, which was maintained by keeping the photosynthetron in a flow-through water bath.

After 2 h, photoincorporation was stopped by adding formalin to each vial to a final concentration of 1.5%. Each sample was filtered, rinsed, placed in 7 mL of Ecolume scintillation cocktail (MP Biomedicals), and counted as described above. All of the experiments were performed in triplicate for each isolate (N=45 for 15 light levels) plus dark and DCMU controls.

The light level at which the maximum rate of photosynthesis (Pmax) was achieved was determined as the point where photoincorporation of 14C leveled off as the light intensity increased when plotted as a P vs. I curve (Lewis & Smith, 1983). The average maximum rate of photosynthesis (average Pmax) was determined by calculating the average of the photoincorporation values across the range of light intensities where Pmax was attained.


16S rRNA gene identification of BBD and non-BBD cyanobacteria

The molecular identifications of the cyanobacteria isolated into culture and used in this study are presented in Table 1, together with their source (host coral species, reef, and region). BBD isolates were from reefs of the Florida Keys, the Bahamas, St Croix in the USVI, and the Philippines and consisted of three Geitlerinema, one Leptolyngbya, and one Oscillatoria species. The latter proved to be most closely related (99% similarity) to the Oscillatoria detected using molecular techniques in BBD samples collected worldwide (discussed previously), deposited as ‘uncultured’ in the GenBank database as accession no. AF473936 by Cooney et al. (2002). The six non-BBD isolates, all collected from reefs of the Florida Keys from nonpathogenic patches on the surfaces of three species of corals (n=5 isolates) and a benthic cyanobacterial mat overlying the sediment (n=1), were most closely related to three genera: Leptolyngbya (n=3), Pseudanabaena (n=2), and Phormidium (n=1).

Morphology of BBD and non-BBD coral reef cyanobacteria

The cyanobacteria isolated from BBD were all filamentous, nonheterocystous, and contained phycoerythrin as the dominant light-harvesting pigment. The three Geitlerinema isolates all have trichomes with isodiametric cells that are 4–5 μm in length and width. One example, isolate HS223, is shown in Fig. 1a. The BBD Leptolyngbya isolate P2b-2 has apparent sheaths and cells that are 2 μm in width and length (Fig. 1b). The Oscillatoria isolate 101-1 has cells that are 5–6 μm wide and 2–3 μm long (Fig. 1c).

Figure 1.

 Photomicrographs of representative cyanobacterial isolates investigated in this study. Isolates are as follows: (a) BBD Geitlerinema HS223, (b) BBD Leptolyngbya P2b-2, (c) BBD Oscillatoria (photo taken by M. Gantar), (d) Leptolyngbya AR-1, (e) Leptolyngbya 9, (f) Pseudanabaena 63-1, (g) Pseudanabaena 72-1, and (h) Phormidium 73-2. All images were taken at a magnification of × 1000. Scale bars=10 μm.

All of the coral reef cyanobacterial isolates from non-BBD mats were also filamentous and nonheterocystous. Five of the six non-BBD cyanobacterial isolates exhibit red pigmentation due to predominance of phycoerythrin. One Leptolyngbya isolate (AR-1, Fig. 1d) has cells that are longer (3–4 μm) than wide (2 μm) while a second Leptolyngbya (isolate 9) has apparent constrictions between the cells, which are 2–3 μm in width and length (Fig. 1e). A third Leptolyngbya (62-2) exhibits green pigmentation and has isodiametric cells, which are 2 μm in width and length (not shown). One Pseudanabaena isolate (63-1, Fig. 1f) has slightly rounded cells that are 2–3 μm in width and length with apparent constrictions between the cells, while a second (72-1) has short filaments consisting of only 5–10 cells, which are 3–4 μm wide and 3–8 μm long (Fig. 1g). The Phormidium isolate (73-2) has very thin filaments with cells that are 1–2 μm in width and length (Fig. 1h).

Assessment of sulfide-tolerant photosynthesis

As shown in Fig. 2a, all five BBD cyanobacterial isolates continued to conduct photosynthesis in the presence of sulfide, with only the Leptolyngbya isolate exhibiting photosynthetic rates lower than when sulfide was absent. None of the BBD isolates were able to conduct photosynthesis in the presence of DCMU and sulfide, and thus were not able to use sulfide as an electron donor for anoxygenic photosynthesis. Under this condition, all incubations exhibited the same results as both the dark control and the DCMU control in which no electron donor was provided.

Figure 2.

 Photoincorporation rates for BBD (a) and non-BBD (b) cyanobacterial isolates under light, aerobic; light, anaerobic with 0.5 mM H2S; light, anaerobic with 0.5 mM H2S and 10 μM (DCMU); light, anaerobic with 10 μM DCMU (control); and dark, aerobic (second control) conditions. BBD samples (a) are as follows: Geitlerinema strains [HS223 (□), W-1 (○), and P2b-1 (▵)]; Leptolyngbya P2b-2 (◊); and BBD Oscillatoria (•). Non-BBD samples (b): Pseudanabaena strains [63-1 (▪) and 72-1 (▴)]; Leptolyngbya strains [AR-1 (□), 9 (▵), and 62-2 (○)]; and Phormidium 73-2 (◆). Each data point represents the average of nine experimental values (three experiments, each with triplicate samples). Samples were incubated at a light intensity of 150 μmol photons m−2 s−1. Error bars indicate the SE of the mean.

Figure 2b shows the results of the same set of experiments for the six non-BBD cyanobacterial isolates. While five of the six isolates were unable to conduct photosynthesis in the presence of sulfide, one of the Pseudanabaena isolates (63-1) continued to photosynthesize, although at a much lower rate when compared with photosynthesis when sulfide was absent. As with the BBD isolates, none of the non-BBD isolates were capable of photosynthesizing when exposed to DCMU and sulfide, indicating that the non-BBD cyanobacterial isolates could not conduct anoxygenic photosynthesis with sulfide as an electron donor.

Photosynthesis vs. irradiance of sulfide-tolerant BBD and non-BBD cyanobacteria

P vs. I curves were generated for aerobic conditions (no sulfide) (Fig. 3a) and anaerobic conditions with sulfide (Fig. 3b) for the isolates that could conduct photosynthesis in the presence of sulfide, with the exception of the BBD Oscillatoria isolate. This set of experiments was not performed for BBD Oscillatoria due to insufficient biomass, as this isolate is extremely slow growing in culture. We did not conduct P vs. I experiments for the five non-BBD isolates that did not exhibit photosynthetic activity in the presence of sulfide (Fig. 2b).

Figure 3.

P vs. I curves for sulfide-tolerant BBD and non-BBD cyanobacterial isolates in the absence (a) and presence (b) of 0.5 mM sulfide. Samples are as follows: BBD Geitlerinema strains [HS223 (□), W-1 (○), and P2b-1 (▵)] and Leptolyngbya P2b-2 (◊); non-BBD Pseudanabaena 63-1 (▪). Solid lines correspond to BBD isolates and the dashed line corresponds to the non-BBD isolate. Photosynthetic rates were corrected for background radioactivity levels and dark controls were subtracted. Each data point represents the average of nine measurements (three experiments with three replicates for each experiment). Error bars indicate the SE of the mean.

The results for the aerobic experiments indicated that the two Caribbean BBD Geitlerinema isolates (HS223 and W-1) and the Philippines BBD Leptolyngbya isolate (P2b-2) were able to attain Pmax at a minimum light intensity of 82 μmol photons m−2 s−1 (Fig. 3a). As can be seen in Fig. 3a, photosaturation was maintained from 82 to 150 μmol photons m−2 s−1 for these isolates. Pmax for the Philippines Geitlerinema isolate (P2b-1) occurred at a minimum light intensity of 73 μmol photons m−2 s−1. For the non-BBD Pseudanabaena isolate 63-1 Pmax was maintained from 94 to 150 μmol photons m−2 s−1.

The P vs. I curves generated under anaerobic conditions with the addition of 0.5 mM sulfide again demonstrated that all of the BBD cyanobacterial isolates continued performing photosynthesis in the presence of sulfide (Fig. 3b). These experiments, however, revealed that photosynthetic rates at light levels below those that supported Pmax were suppressed when compared with comparable light levels in the absence of sulfide. This can be seen in the differences between the shapes and slopes of the curves below 82 μmol photons m−2 s−1 in Fig. 3a vs. 3b. Once Pmax was attained, the photoincorporation rates for all three BBD Geitlerinema isolates in the presence of sulfide were comparable to the photosynthetic rates in the absence of sulfide (Fig. 3; Table 2), ranging from 94% to 104%. The BBD Leptolyngbya isolate had a lower (79%) photoincorporation rate in the presence of sulfide when compared with the rate in the absence of sulfide (Table 2).

Table 2.   Average photosynthetic maximum (Pmax) obtained by sulfide-tolerant coral reef cyanobacterial isolates under aerobic conditions and anaerobic conditions with 0.5 mM sulfide
IsolateAverage Pmax (kCPM mg−1 dry wt h−1)
AerobicAnaerobic, sulfide
(% of aerobic value)
BBD cyanobacterial isolates
 Geitlerinema HS2238.198.16 (99.6%)
 Geitlerinema W-17.157.44 (104%)
 Geitlerinema P2b-17.316.87 (94%)
 Leptolyngbya P2b-25.774.56 (79%)
Non-BBD cyanobacterial isolates
 Pseudanabaena 63-16.362.49 (39%)

In contrast to the BBD cyanobacterial isolates, the single non-BBD cyanobacterium (Pseudanabaena 63-1) that was able to conduct photosynthesis when exposed to sulfide did so at a much lower rate that was only 39% of the rate in the absence of sulfide (Fig. 3, Table 2).


All of the BBD cyanobacteria investigated in this study, which included representative members of the genera Geitlerinema, Leptolyngbya, and Oscillatoria, could perform oxygenic photosynthesis in the presence of sulfide; none were capable of carrying out anoxygenic photosynthesis with sulfide as an electron donor. These results are in agreement with those of our two previous studies of a 1991 BBD Geitlerinema isolate (referred to as Phormidium corallyticum in Richardson & Kuta, 2003) and two additional Geitlerinema and Leptolyngbya BBD isolates (Myers et al., 2007). Thus, of eight BBD cyanobacteria investigated in laboratory physiological studies to date, all exhibit the same strategy of sulfide tolerance – sulfide–resistant oxygenic photosynthesis. In the case of the 1991 isolate, this strain was maintained under aerobic (sulfide-free) conditions for 5 years before the experiments were conducted (Richardson & Kuta, 2003). Additionally, all of the isolates have been confirmed (Myers et al., 2007) as being present in field samples of BBD using denaturing gradient gel electrophoresis.

Our findings are significant because relatively few cyanobacteria are capable of tolerating sulfide. In fact, Cohen et al. (1986) showed that sulfide levels as low as 10–50 μM (0.2–1% of the level used in our study) completely and irreversibly blocked CO2 photoassimilation by 10 strains of sulfide-sensitive cyanobacteria. Furthermore, these authors suggested that the majority of cyanobacteria have a similar sensitivity to sulfide (Cohen et al., 1986). In relation to BBD, our previous studies (Carlton & Richardson, 1995; Richardson et al., 2001) using cored corals with intact BBD revealed that sulfide concentrations ranged from 0 μM in the top few hundred micrometers of the band (during the day) to >800 μM near the base of the band. The sulfide concentration varied with the light intensity, the depth within the band, and the position of the oxygen/sulfide interface (Carlton & Richardson, 1995; Richardson et al., 2001). Because cyanobacteria are found throughout the band matrix, in order to survive in the BBD environment they must tolerate sulfide at all concentrations that occur within the band to prevent irreversible cessation of oxygenic photosynthesis.

The current study is the first report on the physiology of our recently cultured BBD cyanobacterium identified as most closely related to the BBD Oscillatoria (accession no. AF473936, Cooney et al., 2002) that is found worldwide (Cooney et al., 2002; Frias-Lopez et al., 2002, 2003, 2004; Sekar et al., 2006, 2008; Sussman et al., 2006; Barneah et al., 2007; Myers et al., 2007). Here we report that this Oscillatoria isolate exhibits the same strategy of sulfide-resistant oxygenic photosynthesis as the other BBD cyanobacterial isolates. This culture is now axenic and is currently the subject of an in-depth study to fully characterize this potentially critical member of the BBD microbial community.

This study determined that members of three genera of BBD cyanobacteria, Geitlerinema, Leptolyngbya, and Oscillatoria, isolated from widely varying regions (the Caribbean and the Philippines), are able to tolerate the sulfide-rich BBD environment. Although the three Geitlerinema isolates (HS223, W-1, and P2b-1) were collected from three different geographic locations, they have identical 16S rRNA sequences and morphology, and thus are likely strains of the same species. Additional BBD cyanobacteria reported in previous studies, such as Schizothrix and Lyngbya, have not yet been isolated into culture, but should also be investigated for sulfide tolerance. (Reports of Trichodesmium by Frias-Lopez et al., 2002 as an important member of BBD should be discounted as this genus is planktonic and extremely common in tropical and subtropical waters, and thus was most likely an artifact in BBD sampling.)

Based on microscopic observations and molecular studies, it appears that different BBD cyanobacteria occupy the same ecological niche within BBD. This niche, in addition to being sulfide-rich, is nutrient and organic carbon-rich due to active coral tissue lysis. The 1991 BBD Geitlerinema isolate is capable of using exogenous organic carbon to sustain long-term survival in darkness (Richardson & Ragoonath, 2008), suggesting that BBD cyanobacteria can take advantage of the abundance of organic carbon.

In addition to sharing the unique BBD microenvironment, each BBD cyanobacterium may contribute differently to BBD pathogenicity. While we still do not know the full extent and nature of the individual contributions of BBD microorganisms to coral tissue death, we do know that in addition to BBD sulfidogens there are toxin-associated heterotrophic bacteria abundant in BBD (Sekar et al., 2006, 2008) and that BBD cyanobacteria produce the cyanotoxin microcystin (Richardson et al., 2007). Furthermore, we know that different BBD cyanobacterial isolates produce different variants of microcystin and that different variants are found in BBD field samples (Richardson et al., 2007). As there are differences between the toxicity of microcystin variants, the roles of different cyanobacteria in the same BBD niche may be governed by differences in contributions to disease etiology as well as a uniform sulfide tolerance.

In conclusion, the present study demonstrates that all BBD cyanobacterial isolates that have been tested for sulfide tolerance are specifically adapted for survival in the sulfide-rich BBD environment through their ability to conduct sulfide-insensitive oxygenic photosynthesis. In contrast, the non-BBD cyanobacteria isolated in this study were either not capable of tolerating sulfide (five of six) or exhibited very low photosynthetic rates (a 61% decrease) when exposed to sulfide. This inability to tolerate sulfide may result in the exclusion of some coral reef cyanobacteria from the BBD community and may confer a competitive advantage to the sulfide-tolerant BBD cyanobacteria, which can colonize the nutrient- and organic carbon-rich band. Our results suggest that sulfide tolerance may play a key role in determining which cyanobacterial species inhabit the BBD microbial community.


We thank L. Kaczmarsky, J. Pinzón, E. Remily, and J. Voss for sample collection, L. Ochoa for assistance with photosynthetic experiments, and M. Gantar for providing cultures of both non-BBD Psuedanabaena strains, one of the non-BBD Leptolyngbya strains (EU223007), and the BBD Oscillatoria. We also thank the Florida Keys National Marine Sanctuary for boat support. Sample collection in the Florida Keys National Marine Sanctuary was conducted under permit numbers FKNMS-2003-011 and FKNMS-2005-010. This research was supported by NIH (NIH/NIGMS SO6GM8205), NOAA's National Undersea Research Center (FKRP-2004-11A), and NOAA's Caribbean Marine Research Center (CMRC-04-PRJV-01-04C). This is contribution 154 of the Tropical Biology Program at Florida International University.