OUP user menu

Effects of salinity and light on organic carbon and nitrogen uptake in a hypersaline microbial mat

Anthony C. Yannarell, Hans W. Paerl
DOI: http://dx.doi.org/10.1111/j.1574-6941.2007.00384.x 345-353 First published online: 1 December 2007


Utilization of dissolved organic matter (DOM) is thought to be the purview of heterotrophic microorganisms, but photoautotrophs can take up dissolved organic nitrogen (DON) and dissolved organic carbon (DOC). This study investigated DOC and DON uptake in a laminated cyanobacterial mat community from hypersaline Salt Pond (San Salvador, Bahamas). The total community uptake of 3H-labeled substrates was measured in the light and in the dark and under conditions of high and low salinity. Salinity was the primary control of DOM uptake, with increased uptake occurring under low-salinity, ‘freshened’ conditions. DOC uptake was also enhanced in the light as compared with the dark and in samples incubated with the photosystem II inhibitor 3(3,4-dichlorophenyl)-1, 1-dimethylurea, suggesting a positive association between photosynthetic activity and DOC uptake. Microautoradiography revealed that some DOM uptake was attributed to cyanobacteria. Cyanobacteria DOM uptake was negatively correlated with that of smaller filamentous microorganisms, and DOM uptake by individual coccoid cells was negatively correlated with uptake by colonial coccoids. These patterns of activity suggest that Salt Pond microorganisms are engaged in resource partitioning, and DOM utilization may provide a metabolic boost to both heterotrophs and photoautrophs during periods of lowered salinity.

  • microbial mat
  • hypersalinity
  • organic matter uptake
  • microautoradiography


The production and consumption of dissolved organic matter (DOM) represents an important component of the energy and material budget of microbial communities. Most ecological research is conducted from the perspective that autotrophs produce DOM and heterotrophs consume and recycle DOM. However, there is evidence from laboratory-based studies and from a variety of natural systems that microorganisms generally considered to be autotrophic (e.g. microalgae) can at times take up both dissolved organic nitrogen (DON) – especially urea and free amino acids (Antia et al., 1991; Paerl, 1991; Paerl et al., 1993; Nedoma et al., 1994; Nilsson & Sundback, 1996; Hietanen et al., 2002; Tyler et al., 2003, 2005; Zotina et al., 2003; Zubkov et al., 2003; Glibert et al., 2004; Linares & Sundback, 2006) – and dissolved organic carbon (DOC) (Smith, 1982; Paerl, 1991; Paerl et al., 1993; Kirkwood et al., 2003; Zotina et al., 2003; Gobler et al., 2004; Tuchman et al., 2006). Understanding the conditions under which DOC and DON are taken up by autotrophs and heterotrophs is crucial for accurately assessing the flow of matter and energy through the microbial components of ecosystems and to characterize the interactions of microbes as communities and consortia (Paerl & Pinckney, 1996; Paerl et al., 2000).

The ability of autotrophic organisms to utilize DON may be particularly important under conditions of inorganic nitrogen limitation (Antia et al., 1991; Paerl, 1991; Tyler et al., 2005). In some microorganism-dominated systems, like cyanobacterial mats, DON can account for a substantial fraction of the total nitrogen uptake (Davey, 1993; Rondell et al., 2000). DOC and DON may serve as a supplemental energy source or an alternative to photosynthesis under low light (Smith, 1982; Antia et al., 1991; Grotzschel et al., 2002; Zotina et al., 2003; Tuchman et al., 2006) or in systems subjected to frequent disturbances such as burial by sediment (Nilsson & Sundback, 1996). The usage of DOC and DON may be particularly important survival strategies in extreme environments that periodically challenge autotrophs with conditions that are unfavorable to crucial growth-supporting processes like photosynthesis and nitrogen fixation.

The islands of the Bahamas contain numerous hypersaline lakes where evaporation commonly exceeds freshwater input, and organisms in these environments are subjected to periods of extreme water stress, osmotic shock, and desiccation (Paerl et al., 2003). One such lake, Salt Pond on the island of San Salvador, contains a laminated, benthic microbial mat dominated by nonheterocystous filamentous (order Oscillatoriales) and colonial coccoid (orders Chroococcales and Pleurocapsales) cyanobacteria, anoxygenic photosynthetic bacteria, and a highly diverse bacterial community (Paerl et al., 2000, 2003; Yannarell et al., 2006). Previous work on this system has shown that rates of photosynthesis and nitrogen fixation are greatly reduced at salinities above 160 PSU, but experimental ‘freshening’ can lead to increases in the rates of photosynthesis and nitrogen fixation within a day (Pinckney et al., 1995; Paerl et al., 2003; Yannarell et al., 2007). The arid climate in this region of the Bahamas ensures that the Salt Pond community persists in the hypersaline state for much of the year. However, freshening rain events occur with increased frequency in the wet season between September and January. Some of these rain events are in the form of intense tropical storms or hurricanes capable of burying the mat community beneath wind-transported carbonate sand sediment (Paerl et al., 2003; Yannarell et al., 2007). Thus, the Salt Pond microbial mat community has evolved in an extreme environment subjected to pulses of increased water availability and periodic burial.

The pulsed nature of water availability in Salt Pond means that flexibility in microbial activities and interactions during freshening events are important for mat growth, community and ecosystem dynamics. Here, observations are presented regarding DOC and DON uptake by the total community and by individual groups of autotrophic and heterotrophic organisms in the upper oxygenated (i.e. ‘photosynthetic’) layer of the Salt Pond mat. The purpose is to investigate the environmental impact of salinity and light availability on DOM uptake and to characterize the organisms that utilize DOM.


Experimental conditions

All experiments were conducted on a benthic microbial mat community located in hypersaline Salt Pond, San Salvador Island, the Bahamas (24°01′22″N, 74°27′02″W). Two experiments were conducted on 11 June 2005 to determine the impacts of salinity and light energy availability on the uptake of low-molecular-weight DOC (glucose) and DON (leucine). Two more were carried out on 26 February 2006 using a more diverse mixture of substrates. These experimental periods were chosen to investigate DOM uptake at the beginning (February) and height (June) of the dry season in this system. All substrates utilized in this study were labeled with tritium (3H), which has previously been shown to be an appropriate tracer for substrate utilization even though it is not directly associated with the molecular ‘backbone’ of organic molecules (Paerl, 1991; Paerl et al., 1993). The specific details of each experiment are given below, but the following conditions and procedures were common to all experiments. Sections of a microbial mat were collected from Salt Pond and transported to the Gerace Research Station (24°07′09″N, 74°27′49″W), where they were maintained at the appropriate salinity and temperature before experimentation. Replicate cores were obtained using a cylindrical cork borer (internal diameter=11 mm) and were trimmed to include only the upper 4–5 mm of biomass above a sand layer deposited in September 2004 by Hurricane Frances (Yannarell et al., 2007). Cores were placed in glass scintillation vials with 20 mL of water containing trace concentrations of the appropriate substrate. Ambient Salt Pond water was used for high-salinity treatments, while seawater was used for low-salinity treatments. Although these two sources of water often differ in dissolved N concentration (Paerl et al., 2003), parallel incubations with NH4+ and NO3 amendments did not differ from the results reported here (data not shown). Vials for dark treatments were completely wrapped in two layers of heavy-duty aluminum foil, and all incubation vials were placed outdoors in a small wading pool through which running seawater circulated to maintain ambient temperature. Vials were shaded from full sunlight with a single layer of fiberglass window screening (2 mm mesh). At the conclusion of the experiments, each vial was drained and rinsed for 10 min with seawater to remove unincorporated substrates. Cores for scintillation counting were air-dried for 24 h in preparation for transport. Cores for microautoradiography were preserved with 2% w/v formaldehyde.

In June 2005, experiments followed a 2-by-2 fully factorial design investigating the effects of low salinity (46 PSU) vs. high salinity (340 PSU) and incubations conducted in the light vs. in the dark. The DOC uptake experiment utilized 3H-labeled glucose (2.07 μCi nmol−1; MP Biomedicals Inc.) added to a final concentration of 7.69 nM. The DON experiment utilized 3H-labeled l-leucine (13 μCi nmol−1; MP Biomedicals Inc.) added to a final concentration of 48.36 nM. Glucose and leucine incubations were run from 11 : 10 to 14 : 30 and 10 : 30 to 14 : 30, respectively. Three replicate cores per treatment were collected for scintillation counting, and one core per treatment was preserved for microautoradiography.

In February 2006, experiments followed a 2-by-3 fully factorial design investigating the effects of low salinity (36 PSU) vs. high salinity (183 PSU) and three treatment levels reflecting different light conditions. In addition to the light and dark treatments of the June experiments, a third set of samples was incubated in the light but with the addition of 3(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU) to a final concentration of 20 μM in order to inhibit oxygenic photosynthesis (Paerl, 1991). The DOC uptake experiment utilized a mixture consisting of final concentrations of 10.20 nM 3H-glucose (25 μCi mmol−1), 3.47 nM 3H-mannitol (20 μCi mmol−1), and 8.54 nM 3H-acetate (10 μCi mmol−1). The DON experiment utilized a 3H-labeled l-amino acid ‘algal protein hydrolysate’ mixture (MP Biomedicals Inc.) with an overall specific activity of 35.7 μCi nmol−1. The final incubation concentrations of the compounds in this mixture were as follows: 99 pM l-alanine, 77 pM l-arginine, 471 pM l-aspartic acid, 357 pM l-glutamic acid, 129 pM l-glycine, 35 pM l-histidine, 104 pM l-isoleucine, 123 pM l-lecuine, 128 pM l-lysine, 195 pM l-phenylalanine, 78 pM l-proline, 174 pM l-serine, 333 pM l-threonine, 167 pM l-tyrosine, and 333 pM l-valine. All incubations began at 09:45 hours and ended at 15:45 hours. Five replicate cores per treatment were collected for scintillation counting, and a single core per treatment was preserved for microautoradiography.

Scintillation counting

Samples were placed in 5 mL of CytoScint liquid scintillation cocktail (MP Biomedicals Inc.) and extracted for 48 h. Radioactivity per core was determined with a Beckman-Coulter LS6500 liquid scintillation counter. Counts were corrected for quenching using a quench curve developed for microbial mats (Paerl et al., 1993) with calibrated 3H-hexadecane standards (ICN Inc.). Because experimental incubations were not conducted with uptake-kinetics saturating substrate concentrations, and because the impacts of isotope dilution were unknown, no attempt was made to convert radioactive measurements into uptake rate estimates; data were recorded as disintegrations per minute (DPM) for each sample core.


Thin-layer grain-density microautoradiography was performed on all low-salinity samples following the procedure of Paerl (1974) with the following modifications. Formaldehyde-preserved cores were subsampled with a cylindrical cork borer (internal diameter=5.5 mm), and subsamples were manually homogenized with a sterile plastic pestle into 1.5 mL of 2% formaldehyde. Fifty microliter of this mixture was then suspended in reverse osmosis/deionized (RO/DI) water and filtered onto 0.45 μm Millipore HA nitrocellulose filters. Filters were rinsed three times with 60 mL of RO/DI water to remove traces of salt and formaldehyde from samples. Filters were stained with 2% erythrosin in 5% phenol, destained with RO/DI water, air-dried, placed onto clean microscope slides, and optically cleared under acetone fumes. Slides were then dipped in Kodak-type NTB2 autoradiography emulsion (diluted 1 : 1 with water) and placed in the dark for radiographic exposure. Slides from June 2005 and February experiments were exposed for 56 and 21 days, respectively. Slides were then developed, fixed, rinsed, and dried (Paerl, 1974; Paerl et al., 1993), after which the uptake of radiolabeled substrates was microscopically determined by the presence of silver grains precipitated in the emulsion (Fig. 1).


Dissolved organic matter uptake as determined by liquid scintillation counting of 3H-labeled mat cores. Error bars show the 95% confidence interval of the mean. (a) Results from June 2005 experiments (n=3); (b) Results from February 2006 experiments (n=5).

Enumeration of radiolabeled organisms was performed with phase contrast at × 200 using a Leica DMIRB inverted microscope. Three counting transects (1 cm long) were scribed with an ultra-fine point marker onto glass coverslips before mounting with Cargille Laboratories Inc. ‘Type B’ high-viscosity immersion oil. For each counting transect, a total of 20 fields of view (as defined by a 490 μm × 490 μm square counting reticle) were examined and all organisms taking up substrates (as indicated by silver grains) were enumerated and pooled for a transect total. No attempt was made to distinguish between different densities of silver grains (i.e. ‘quantitative’ microautoradiography), and neither were differences in organism biomass or cell density for colonial or filamentous organisms taken into account.

Statistical analyses

Liquid scintillation count data (DPM) were analyzed by two-way anova to account for salinity effects, treatment effects (i.e. light, dark, DCMU), and the salinity–treatment interaction. For February experiments, significant overall treatment effects were further explored with post hoc pairwise comparisons using the procedure of Tukey's Honest Significant Differences. Because DPM data were not converted to standardized uptake rates, and because of experiment-to-experiment differences in the specific activity of substrates, no statistical comparisons were attempted between treatments from different months or different substrate experiments. Calculations were performed using the R statistical environment (R Development Core Team, 2005).

Microautoradiographic enumerations of substrate-utilizing organisms were analyzed using a nonmetric, multivariate framework. Counts for each organism type were standardized to proportions through division by the total number of labeled organisms enumerated for the transect. A transect-by-transect distance matrix was assembled using the Bray–Curtis (dis)similarity coefficient [coefficient D14 of Legendre & Legendre (1998)]. This distance matrix was used to produce a nonmetric multidimensional scaling (NMDS) plot to summarize patterns of change in the substrate-utilizing community across the various low-salinity samples. NMDS plots graphically represent the underlying structure of a distance matrix while preserving the ranked dissimilarity between all pairs of observations (Kruskal, 1964). Thus, in the present case, the two counting transects that were most similar in the makeup of their substrate-utilizing communities plotted closest together, the two least similar plotted farthest apart, and so on for all pairs of counting transects. NMDS was conducted from 20 random starting configurations, with the optimal two-dimensional solution [lowest Kruskal stress formula 1 (Kruskal & Wish, 1978) value=0.12] arising eight times. A type of biplot was produced by correlating (Pearson's R) the loadings of the sample scores on each NMDS axis to the proportional representation of labeled organisms using the Pearson correlation coefficient.

In order to quantify the degree to which different groups of organisms modified their substrate uptake under different conditions, the similarity percentages (SIMPER) method (Clarke, 1993) was used. Briefly, this procedure calculates the average contribution of each species to the dissimilarity of samples in different treatment groups (i.e. light vs. dark), as well as the standard deviation (SD) of the contribution of each species to between-group dissimilarity. Species with a high average contribution and a low SD can be deemed to be consistently important in discriminating between samples in those two groups (Clarke, 1993).

Distance matrix calculation, NMDS, and SIMPER were all performed using primer 5 for Windows (Clarke & Gorley, 2001).


In June 2005, the uptake of glucose and leucine was higher under conditions of lower salinity as compared with high salinity, and in the light as compared with the dark (Table 1, Fig. 1a). The salinity–light interaction was significant for glucose but not for leucine, indicating that freshening enhanced glucose uptake in the light much more than in the dark (Table 1). In February 2006, the uptake of sugars and amino acids was higher at low salinity as compared with high salinity (Table 1, Fig. 1b). Light treatment effects and the water–light interaction were significant for sugars, but not for amino acids (Table 1). Post hoc comparisons revealed that the uptake of sugars was higher in the light than in the dark (adjusted P=0.004) and with DCMU (adjusted P=0.045), but there was no detectable difference between uptake in the dark and uptake in the DCMU-treated samples (adjusted P=0.281).

View this table:

Analysis of variance of total substrate uptake

June 2005, glucose
Low : high salinity148.401<0.001***
Light : dark148.661<0.001***
June 2005, leucine
Low : high salinity132.700<0.001***
Light : dark110.4590.012*
Interaction11.8680.21 NS
February 2006, sugars
Low : high salinity179.758<0.001***
Light : dark : DCMU28.6000.002**
February 2006, amino acids
Low:high salinity175.387<0.001***
Light : dark : DCMU22.2880.12 NS
Interaction21.7910.19 NS
  • *α=0.05; **α=0.01; ***α=0.001. NS, not significant; DCMU, 3(3,4-dichlorophenyl)-1,1-dimethylurea.

Microautoradiography revealed that over 99% of all substrate-utilizing organisms could be classified into seven categories based on morphology (Fig. 2). These categories were: (1) large filamentous cyanobacteria, predominantly Microcoleus spp. (these were enumerated as single bundles/trichomes and not on a per cell basis); (2) irregular, plaque-like colonies of polygonal or sarcinoid cyanobacteria, e.g. Chroococcidium spp. (enumerated as single colonies); (3) round colonies of polygonal or sarcinoid cyanobacteria, e.g. Cyanosarcina spp. (enumerated as single colonies); (4) colonies of small coccoid organisms (enumerated as single colonies); (5) small filamentous organisms (enumerated as single filaments); (6) individual coccoid organisms whose morphology was partially obscured by silver grains; and (7) mixed-species assemblages where localization of silver grains to individual cells or colonies was not possible (enumerated as single assemblages).


Microautoradiographs showing the seven groups of organisms enumerated in the present study. The arrowheads in each panel indicate examples of silver grain labeling, which may appear as light or dark spots depending on the relation of grains to the focal plane in each panel. The 10-μm bar in panel A applies to all panels, except F. (a) Microcoleus; (b) small filaments; (c) Chroococcidium; (d) small coccoid colonials; (e) mixed-species assemblage; (f) small coccoid organisms; (g) Cyanosarcina.

Regardless of light treatment level or substrate type, there were consistent differences in the substrate-utilizing communities of June and February, which were separated along axis 1 of Fig. 3. Overall, there was a high degree of overlap in labeled communities between different light treatment levels and also between DOC and DON substrates (Fig. 3).


Nonmetric multidimensional scaling biplot showing the standardized composition of dissolved organic matter-utilizing communities from microautoradiographs of low-salinity experiments. Each point represents the labeled community from a single counting transect. Vectors correspond to the seven organisms depicted in Fig. 2. Final stress for this biplot=0.12.

Across all microautoradiographs and all substrates, the labeling of large cyanobacteria (Microcoleus, Cyanosarcina, Chroococcidium) was negatively correlated with the labeling of small filamentous organisms (Fig. 3). Individual coccoid organisms were negatively correlated with mixed-species assemblages and colonial coccoid organisms (Fig. 3). The labeling of the cyanobacteria/filaments was nearly independent of that of the coccoids/mixed species, with the former associated with axis 1 and the latter associated with axis 2 (Fig. 3).

Differences between June and February substrate-utilizing communities were primarily due to small filamentous organisms, which constituted a larger fraction of the labeled communities in June, while Microcoleus, Cyanosarcina, and small round organisms constituted a larger fraction of the labeled communities in February (Fig. 3). Changes in the labeling patterns of these four groups accounted for over 80% of the average dissimilarity of June and February samples.


Bulk community DOM uptake

As has been shown previously for rates of photosynthesis and nitrogen fixation in the Salt Pond mat (Pinckney et al., 1995; Paerl et al., 2003; Yannarell et al., 2007), the rates of DOM uptake were primarily controlled by salinity (Table 1; Fig. 1), with increased uptake rates rapidly following experimental freshening. The ability of the mat community to rapidly respond to pulses of fresh water is a crucial adaptation to the fluctuating environment of this system. Halotolerant organisms must take advantage of periodic low-salinity windows of opportunity to acquire the energy and nutrient sources needed for cellular and population growth. Increased sequestration of new carbon and nitrogen (Pinckney et al., 1995; Paerl et al., 2003; Yannarell et al., 2007) and increased utilization of DOM (present study) may represent an adaptation of organisms or microbial consortia (Paerl et al., 2000) to utilize all available carbon and nitrogen sources for growth in Salt Pond's extreme environment. The short-term nature of the present study does not allow determination of whether this type of DOM utilization is a persistent feature of the Salt Pond community throughout the September to January wet season, or whether increased DOM uptake occurs as a pulse following freshening events. The nature of this activity will substantially impact the seasonal carbon and nitrogen budgets in this system.

Light was a further control on DOM uptake rates, with light stimulating the uptake of both DOC and DON, but only under freshened conditions when the community was already ‘activated’ by low salinity (Table 1; Fig. 1). For DON, this light stimulation was only significant in June. Light-stimulated DON uptake has been shown before for marine phyto- and bacterioplankton, including axenic cultures of the cyanobacterium Synechococcus (Paerl, 1991). DON uptake by plankton at station ALOHA followed a diel pattern that mirrored photosynthetic activity (Church et al., 2004), and surface water planktonic communities from the Mediterranean and Atlantic showed a positive relationship between PAR and DON uptake (Moran et al., 2001). These observations may be indicative of increased nitrogen demand for photosynthesis. However, previous work from a variety of microbial mat communities showed that DON uptake was not consistently enhanced by light (Paerl et al., 1993), and other workers found no short-term influence of light on DON uptake by estuarine phytobenthic communities (Linares & Sundback, 2006). The discrepancies between these various studies may reflect differences between planktonic and benthic habitats where organisms face a different set of conditions and constraints in relation to the light field.

Autotrophs may utilize DOC to compensate for inadequate photosynthesis (Rondell et al., 2000; Zotina et al., 2003); however, the present study revealed light stimulation of DOC uptake (Fig. 1). Even cyanobacterial uptake of DOC was enhanced by the light, as exemplified by the light response of the Cyanosarcina group to the carbon sugars mixture (Fig. 3). If DOC were serving to compensate for inadequate photosynthesis, then the addition of the photosystem II inhibitor DCMU should have enhanced DOC uptake as was seen previously in microbial mats (Paerl et al., 1993). In the present study, DOC uptake by the whole community (Fig. 1b) and by the three groups of large cyanobacteria (Fig. 3) was inhibited by DCMU under illumination as compared with the illuminated conditions alone. This suggests that photosynthetic activity may stimulate DOC uptake in Salt Pond. This may be an indirect effect, for example, stimulation of heterotrophy by oxygen production, or it may be the product of a tight coupling of microbial activities through consortial relationships (Paerl & Pinckney, 1996; Paerl et al., 2000).

DOM-utilizing community at low salinity

The large cyanobacteria Microcoleus, Cyanosarcina, and Chroococcidium were seen to take up both DOC and DON under freshened conditions, showing that autotrophic microbes utilize organic matter. When all microradiographs were taken into consideration, the proportions of these three groups in the labeled communities were highly correlated with each other (Fig. 3), suggesting that these cyanobacteria were taking up DOM in response to similar sets of conditions. Furthermore, because the labeling of Microcoleus, Cyanosarcina, and Chroococcidium was negatively correlated with that of the small filaments group (Fig. 3), it is possible that the large cyanobacteria and the small filaments constitute two separate functional guilds in terms of DOM utilization. By the same reasoning, the individual coccoid organisms and the colonial coccoid organisms/mixed-species assemblages may constitute separate DOM-utilizing guilds (Fig. 3).

The most definable differences in the composition of DOM-utilizing communities were observed between the June experiments and the February experiments (Fig. 3), with June communities dominated by the small filaments and February communities showing a greater diversity of DOM-utilizing organisms. Several potential explanations may account for this pattern. These organisms may be acting as distinct metabolic guilds whose activities vary seasonally. Alternatively, it is possible that the higher diversity of labeled organisms in February samples reflects the higher diversity of substrates used in both the DOC and DON experiments in February. This is particularly likely if competition for DOM has selected for a high degree of resource partitioning in the Salt Pond mat. It is also possible that the differences in DOM-utilizing communities reflected underlying temporal differences in the total community. Previous work has shown that there are seasonal differences in community composition in the Salt Pond mat (Yannarell et al., 2006), and these seasonal differences could be reflected in the communities present during the different experimental periods. There may additionally have been lingering differences as a result of disturbance wrought by Hurricane Frances in September of 2004 (Yannarell et al., 2007). This storm resulted in a dominance shift among the cyanobacterial community, with small filamentous cyanobacteria of the Lyngbya/Phormidium/Plectonema (LPP) group replacing previously dominant Oscillatoriales (Yannarell et al., 2007). The posthurricane LPP organisms were morphologically similar to some of the small filaments of the present study (but see below) and the dominant prehurricane Oscillatoriales was most likely Microcoleus. The shift from small filaments in June 2005 to Microcoleus and others in February may thus have been part of a longer-term pattern of community recovery from disturbance.

While DOM uptake increased in the light at low salinities, there was no detectable difference in the composition of substrate-utilizing communities in the light vs. in the dark (or with DCMU; Fig. 3). This may mean that only uptake rates are influenced by the light, with the same sets of organisms taking up DOM in the dark at reduced rates. However, the morphological similarity of some microorganisms, particularly heterotrophic bacteria, may have masked shifts in substrate utilization. For instance, the small filamentous organisms are morphologically similar to both the photoautotrophic LPP seen to dominate in Salt Pond after Hurricane Frances (Yannarell et al., 2007) and to the photoorganotrophic Chloroflexus shown to dominate in a similar hypersaline microbial mat (Nübel et al., 2001; Ley et al., 2006). Further molecular characterization of the Salt Pond microbial community will be useful in the development of molecular probes with high taxonomic resolution, so that techniques such as combined microautoradiography and fluorescent in situ hybridization (Nielsen et al., 1999) may help clarify the role of specific organisms involved in DOM utilization in this system.

Concluding remarks

This study demonstrates that hypersalinity limits microbially mediated uptake of DOC and DON in Salt Pond, but that uptake rapidly increases with lowered salinity. This pattern mirrors what has previously been seen for primary productivity and diazotrophy (Pinckney et al., 1995; Paerl et al., 2003; Yannarell et al., 2007). Increased organic matter uptake following freshening allows organisms to recover C and N from the environment and may serve as a boost for increased production after periods of inactivity. Under low-salinity conditions, light enhances the uptake of DOC – and to a lesser extent that of DON – over uptake in the dark, and this observation, coupled with the observation that DCMU inhibits DOM uptake, suggests that photosynthetic activity is positively related to DOM utilization. At least some of the DOC and DON uptake in the Salt Pond mat was attributable to cyanobacteria, providing further evidence of DOM uptake of microorganisms generally considered to be autotrophic. Organism-specific patterns of DOM utilization suggest that bulk community uptake of DOM is influenced by the metabolic diversity and niche specialization within the Salt Pond community. These observations also suggest that microbial uptake of DOM may have play an important role in extreme environments, particularly during periods of lowered environmental stress.


For assistance with field work, sample analysis, and constructive advice on this manuscript, the authors gratefully acknowledge V. Voegeli, A. Kent, K. Rossignol, M. Leonard, J. Braddy, B. Peierls, and the staff at Gerace Research Center, San Salvador, Bahamas. This work was funded by the US National Science Foundation's Microbial Observatories Program (MCB-0132528).


  • Editor: Riks Laanbroek


View Abstract