|Home | About | Journals | Submit | Contact Us | Français|
Marine bacterial isolates cultured from the digestive tracts of blue mussels (Mytilus edulis) contaminated with paralytic shellfish toxins (PSTs) were screened for the ability to reduce the toxicity of a PST mixture. Seven isolates reduced the overall toxicity of the algal extract by ≥90% within 3 days. These isolates shared at least 99% 16S rRNA gene sequence similarity with five Pseudoalteromonas spp. Phenotypic tests suggested that all are novel strains of Pseudoalteromonas haloplanktis.
Among the marine algal biotoxins identified to date; paralytic shellfish toxins (PSTs) constitute the most serious threat to the safety of the food supply, mainly due to their high acute toxicities and the absence of antidotes or effective medical treatments (8). Paralytic shellfish poisoning is caused by ingestion of one or more of the chemically related PSTs (see Fig. S1 in the supplemental material). PSTs are mainly produced by marine dinoflagellates, including Alexandrium spp., Gymnodinium catenatum, and Pyrodinium bahamense var. compresssum (16). Since bivalve molluscs filter-feed on marine algae, they tend to concentrate PSTs largely, but not exclusively, in their digestive organs (7, 9, 10, 29). Not affected by commercial sterilization (14, 18) or cooking, PSTs present significant risks to the food supply, particularly during periods of toxic algal blooms. Practical methods for PST detoxification of living shellfish do not exist (5).
Transformations of PSTs by bacteria have been reported in the literature (23-25, 31, 35, 36, 38); early studies focused on the conversion of hydroxysulfate carbamate derivatives (gonyautoxins 1 and 4) to the more highly toxic saxitoxin (STX) (23-25). In addition, several reports have noted the high capacity of the digestive gland for PST transformation (12, 28, 32, 39), suggesting the presence of toxin-transforming enzymes and/or microorganisms in bivalve molluscs. The partial degradation of gonyautoxins 1 and 4 and C1/C2 by marine bacteria has also been reported (38). In addition, Stewart et al. (37) discovered the bacterial degradation of domoic acid (another marine toxin that causes amnesic shellfish poisoning), collectively suggesting that bacteria might play a role in the elimination of marine toxins from toxic bivalve molluscs. The capacity to catabolize domoic acid is greater in cultures isolated from blue mussels that rapidly eliminate domoic acid than in bacterial isolates from bivalves known to retain the toxin for longer time periods (e.g., scallops), suggesting these bacteria play a role in the elimination of marine toxins.
Recently, we reported the kinetics of PST destruction for a group of marine bacteria isolated from toxic blue mussels (11). Here we report the phenotypic and taxonomic characterization of these unique marine bacteria.
Toxic blue mussels were recovered from around Atlantic Canada by the Canadian Food Inspection Agency (CFIA; Dartmouth, Nova Scotia, Canada) as part of its routine shellfish inspection program. The digestive gland microflora from affected mussels was sampled and streaked on marine agar 2216 (Difco Laboratories, Detroit, MI), from which 69 bacterial isolates were recovered based on distinct colony morphologies and purity. Isolates were identified numerically and further grouped into the “clear” (C) or “opaque” (O) groups according to differences in colony appearance detected upon subculture. The rationale for processing digestive glands was that PSTs tend to concentrate in this organ (7, 9, 10, 29), likely creating an enriched environment for PST-degrading bacteria.
All 69 isolates were tested for their capacity to break down PSTs in 1 ml of sterile screening medium consisting of marine broth 2216 (MB) (Difco Laboratories, Detroit, MI) (600 μl), a toxic algal extract (100 μl) prepared from Alexandrium tamarense (strain Pr18b) as described by Donovan et al. (11), and a mussel extract (300 μl) prepared from fresh blue mussels (11). Nontoxic controls were prepared by replacing the algal extract with 100 μl of sterile water. The bacterial inoculum was prepared by growing selected isolates in MB, harvesting by centrifugation (2,000 x g for 10 min) and resuspending the cells in fresh MB to yield a suspension with an A650 of 1 unit. The screening medium was then inoculated with 100 μl of the bacterial suspension, and controls were prepared with 100 μl of fresh sterile MB. Cultures were incubated at 25°C for 5 days in a shaking incubator set at 130 rpm. Samples were taken on days 0 and 5 and analyzed for PSTs by high-performance liquid chromatography (HPLC) using methods described elsewhere (11).
A confirmatory screening was performed only on those isolates demonstrating potential toxin breakdown from the initial (HPLC) screening. Samples were prepared as in the initial screening, except volumes were increased to accommodate requirements for the mouse bioassay (MBA) on days 0 and 5 and daily sampling for HPLC analyses. The MBAs were performed at the CFIA (Dartmouth, Nova Scotia, Canada) following AOAC official method 959.08 (2).
Only seven isolates were considered “fast detoxifiers,” completely eliminating at least one PST (see Fig. S2 in the supplemental material) and reducing the overall toxin level of the screening medium by no less than 90% in ≤3 days (Table (Table1).1). In all cases, the PST levels in sterile control samples remained constant over the 5-day period (see Fig. S3 in the supplemental material). It should be noted that all seven isolates were unable to grow on purified PSTs as a sole source of carbon or nitrogen, suggesting that the toxins were cometabolized. The degradation kinetics of each PST found in the algal extract has been previously reported for all seven isolates (11). The reduced toxicity of screening medium treated with the seven isolates was confirmed by the MBA (Table (Table22 and Fig. Fig.1),1), implying that true biodegradation rather than simple PST biotransformations had taken place. It is perhaps interesting to note in Fig. Fig.11 that for all seven cultures, each tested with three mice, percentage survival rates were identical, and therefore all 21 data points are shown as single determinations. The isolates reduced the toxicity of the screening medium by a factor of 3. At this point, we can only speculate on the mode(s) of PST degradation. Microbial degradation of N-heterocyclic compounds similar in structure to the PSTs has been reviewed by Xu et al. (40). The most efficient biodegradation of such compounds generally occurs under aerobic conditions, where oxidations catalyzed by oxidases and peroxidases constitute key biodegradation reactions (13). It is likely that these reactions played an important role in the conversion of PSTs into nontoxic metabolites, simply because our isolates are aerobic. Once oxidized, it seems reasonable to surmise that PSTs could be subsequently metabolized in central pathways such as those for purine and/or arginine catabolism (30).
Procedures for preparation, amplification, cloning, sequencing, and analyzing the 16S rRNA genes were performed as previously described (6). rRNA (16S) gene sequence analyses indicated that all of the fast PST-degrading bacteria aligned within a single clade in the genus Pseudoalteromonas (Fig. (Fig.2),2), having >99% sequence similarity to previously reported species. Phylogenetic relationships among all 69 isolates are shown in Fig. S4 in the supplemental material. Pseudoalteromonas spp. are readily cultivated from marine environments and are often found in association with marine eukaryotes, including toxic dinoflagellates (1, 15, 20). They have been reported to produce a variety of biologically active metabolites that include polyketides, a multitude of extracellular enzymes and polysaccharides, as well as antibiotics and antimycotics (17, 19, 21, 41). Certain Pseudoalteromonas spp. have also demonstrated algicidal activity against dinoflagellates capable of producing PSTs (27, 33). However, no previous reports have alluded to the remarkable, broad PST-degrading abilities of the Pseudoalteromonas genus.
The PST-degrading Pseudoalteromonas spp. were further characterized according to Bergey (4) and as shown in Table Table3.3. Transmission electron microscopy was carried out on negatively stained specimens (3), revealing that all seven isolates were rod shaped, occurring as single cells and short chains (see Fig. S5 in the supplemental material). Isolate C20-C was unusual in that the majority of cells were arranged as filaments. All isolates produced pili and flagella as well as outer membrane vesicles (see Fig. S5 in the supplemental material). Biochemical and physiological tests were performed as described by Smibert and Krieg (34). Glucose oxidation and fermentation were performed using the modified oxidation/fermentation medium of Leifson (26) for marine bacteria. Alginate hydrolysis was tested as described by Kitamikado et al. (22). Substrate utilization profiles were generated using 96-well Biolog-GN MicroPlates (Biolog, Inc., Hayword, CA) as described by Smith et al. (35). All other biochemical and physiological tests were carried out as described by Smibert and Krieg (34). Results from the phenotypic characterization indicated that the PST-degrading isolates were more similar to Pseudoalteromonas haloplanktis, the type species of the genus.
One-third of the 69 isolates from the digestive gland of blue mussels had some ability to degrade one or more of the PSTs. However, nearly complete destruction was accomplished by only seven of these isolates. It is possible that these and similar bacteria aid marine bivalves in the natural metabolism and elimination of PSTs and other marine biotoxins. Work is currently in progress to develop a biological process to expedite the elimination of PSTs from bivalves in vivo.
This work was supported by a research grant from AquaNet (Canadian Network of Centres of Excellence in Aquaculture; T.A.G., M.A.Q., and R.A.G.) and a postgraduate scholarship from the Natural Sciences and Engineering Research Council of Canada (C.J.D.).
We thank the CFIA (Dartmouth, Nova Scotia, Canada) for providing toxic mussels and assisting with the MBAs, the National Research Council of Canada Certified Reference Materials Program (Halifax, Nova Scotia, Canada) for providing purified PST calibration standards, and Health Canada (Ottawa, Ontario, Canada) for sequencing the 16S rRNA genes of the PST-degrading isolates.
Published ahead of print on 28 August 2009.
†Supplemental material for this article may be found at http://aem.asm.org/.