BACTERIAL DEGRADATION OF PARALYTIC SHELLFISH TOXINS

CLAIM OF PRIORITY fOOOl] This application claims the benefit of U.S. Provisional application No. 60/976,557 filed October 1 , 2007 which is herein incorporated in its entirety by reference.

TECHNICAL FIELD OF THE INVENTION

[0002] The present invention provides for bacterial methods of isolating, screening, and using bacterial cells and their cellular components in order to detoxify paralytic shellfish toxins. The present invention provides new technologies for bioremediation and decontamination of toxins as described.

BACKGROUND OF THE INVENTION

[0003] Paralytic shellfish toxins (PSTs) are a group of potent neurotoxins possessing a tricyclic perhydropurine skeleton. Considered purine alkaloids, PSTs resemble those compounds produced by higher plants, such as caffeine. They are among the most toxic substances known. PSTs are traditionally considered to be synthesized by marine dinoflagellates. However, in recent years, marine bacteria have also been investigated as a primary source. Most researchers consider PSTs and related plant alkaloids to be secondary metabolites, implying that they are not essential to the life and growth of the producing organism. The role PSTs play includes intrinsic roles in DNA metabolism, nitrogen storage, and/or an extrinsic role as a defensive compound against predation. [0004] The structures of PSTs are based on a tricylic perhydropurine skeleton with two guanidinium groups. The parent PST molecule, saxitoxin (STX), gives the molecular formula CK)HI ₇ N ₇ O ₄ and a molecular weight of 299.3g/mole. STX is a white hygroscopic solid which is very soluble in water, but insoluble in most non-polar solvents and only partly soluble in methanol and ethanol. It is an unstable alkaloid that easily oxidizes at elevated pH levels and, with the exception of the N-sulfocarbamoyl derivatives, is heat stable in acidic conditions.

[0005] Although PSTs vary in structure and potency, they all act through a reversible blockage of voltage-gated sodium channels in nerve and muscle plasma membranes, interrupting the formation of an action potential.

[0006] The group of shellfish most susceptible to accumulating PSTs is filter- or suspension- feeding bivalves. The most frequently implicated bivalves are mussels and clams, but

oysters, scallops and cockles are also commonly contaminated. PSTs accumulate primarily in these shellfish because they feed directly upon toxic dinoflagellates. Other than their role as primary consumers of algae, a number of factors facilitate toxin accumulation in bivalves. These include their limited mobility, the relative insensitivity of some species to PSTs, and their ability to filter large volumes of water per unit time, concentrating the algae in their systems.

[0007] The standard AOAC mouse bioassay (MBA) is currently the official method by which the net toxicity of PST-contaminated samples is measured, forming the basis of shellfish safety monitoring programs in most countries. The MBA was developed in 1937 and later refined and standardized by the Association of Official Analytical Chemists (AOAC). The test is an in vivo assay that measures the total PST content of toxic shellfish samples. It involves injecting 1 ml of an acid extraction of the shellfish sample into a triplicate of mice weighing 19-22 g. The injection is given intraperitoneally and the time taken for the animal to die is recorded. Highly toxic samples are diluted accordingly to cause death within 5-15 min of injection. After the death of the mice, the lethality of the samples is calculated and expressed in mouse units (MU). One mouse unit (MU) is equivalent to 0.16- 0.23 μg STX eq. and is defined as the minimum amount of toxin that would kill a 20 g white mouse in 15 min when injected intraperitoneally with 1 ml of the acid extract of a shellfish sample. The detection limit of the assay is 32-58 μg STX eq./100 g. Most countries have adopted a regulatory limit of 80 μg STX eq./100 g. Shellfish containing PST levels under this value are considered safe for human consumption. Although the MBA has proven useful in providing data on trends in net toxin concentration, the assay may not always give an accurate picture of shellfish sample toxicity.

[0008] Several methods of detoxifying PST-contaminated bivalves have been utilized over the last several years. These have included self-depuration in uncontaminated water, various methods of thermal processing, chemical and pH treatments, and removal of contaminated tissues. Thus far, none of these processes have proven effective in reducing toxicity in contaminated bivalves as they may be unsafe, costly, too slow, and/or have resulted in products the consumer may reject due to unacceptable properties. It is clear that new technologies are needed to develop an effective method of bivalve decontamination that is commercially viable. To address these limitations in current technologies, we present below the isolation of unique bacteria capable of effective decontamination of paralytic shellfish toxins in selected environments and food supplies.

BRIEF SUMMARY OF THE INVENTION

[0009] The embodiments disclosed herein include; (1) isolated bacteria from the genus Pseudoalteromonas capable of degrading paralytic shell fish toxins, (2) isolated bacteria capable of degrading PSTs wherein the bacteria can be identified by their 16S rDNA gene sequences described herein, (3) isolated bacteria capable of degrading PSTs wherein the bacteria can be identified by their phenotypic and morphological characterizations described herein, (4) isolated bacteria capable of degrading PSTs wherein the bacteria can be identified by their unique plasmid profile described herein, and (5) isolated bacteria capable of degrading PSTs wherein the bacteria can be identified by their unique protein profile described herein.

[0010] Also included in the embodiments are; (1) compositions according to the isolated bacteria described above consisting of whole cells, cellular fractions, cellular components, and lysed cells, of said bacterium, wherein said composition degrades PSTs, (2) compositions that include agents and additives that enhance degradation properties, (3) compositions according to the isolated bacteria described above wherein said composition degrades PSTs in an aqueous environment, and (4) compositions with the isolated bacterium wherein said bacterium is selected from the group of species within the genera Pseudoalteromonas. [0011] Also included in the embodiments are; (1) formulations comprising of bacteria described herein which were isolated for their capability of PST degradation, (2) processes of reducing PSTs in selected environment, (3) methods for screening bacteria for their ability to degrade PSTs, (4) isolated plasmids that encode proteins assisting in the degradation PSTs, (5) isolated proteins that are capable of assisting in the degradation of PSTs, and lastly (6) shellfish that have internal concentrations of bacteria decribed herein that are greater than concentrations found naturally occurring in shellfish.

ABBREVIATIONS

ADI Arginine deiminase pathway

AE Toxic algal extract

AOAC Association of Official Analytical Chemists

APC Aerobic plate count

APS Ammonium persulfate

AST Arginine succinyltransferase pathway

ASW Artificial seawater

BAM Bacteriological Analytical Manual

BHI Brain heart infusion bp Base pairs

C Crosslinker

C1/C2 C1/C2 toxins (N-sulfocarbamoyl-gonyautoxin 2/3)

CFIA Canadian Food Inspection Agency

CFSAN Center for Food Safety and Applied Nutrition

CFU Colony forming unit

CIFT Canadian Institute of Fisheries Technology

DA Domoic acid dcGTX Decarbamoyl gonyautoxin dcNEO Decarbamoyl neosaxitoxin dcSTX Decarbamoyl saxitoxin

DDW Distilled deionized water

DNA Deoxyribonucleic acid

EDTA Ethylenediamine tetraacetic acid

EPS Extracellular polymeric substances

F Fermentative

FAO Food and Agriculture Organization of the United Nations

FDA Food and Drug Administration

FAME Fatty acid methyl esters

GTX Gonyautoxin

HAB Harmful algal bloom

HPLC High performance liquid chromatography

LC-MS Liquid chromatography coupled with mass spectroscopy

LC-pcr-FLD Liquid chromatography coupled with post-column oxidation and fluorescence detection

LD50 Lethal dose, 50%

LPS Lipopolysaccharide

MA Marine agar 2216

MB Marine broth 2216

MBA Mouse bioassay

ME Mussel extract

MNB Mouse neuroblastoma

MU Mouse units

NEO Neosaxitoxin

N-ST N-sulfotransferase

O Oxidative

OD Optical density

OMV Outer membrane vesicle

O-ST O-sulfotransferase

PAGE Polyacrylamide gel electrophoresis

PAPS 3 ' -phosphoadenosine 5 ' -phosphosulfate

PBS Phosphate buffered saline

PCR Polymerase chain reaction

PST Paralytic shellfish toxin rDNA Ribosomal deoxyribonucleic acid

RISA Ribosomal intergenic spacer analysis

RNA Ribonucleic acid

SCB Sodium channel blocking

SD Standard deviation

SDS Sodium dodecyl sulfate

STX Saxitoxin

T Total acrylamide

TAE Tris-acetate-EDTA

TE Tris-EDTA

TEM Transmission electron microscopy

TEMED N, N, N', N'-tetramethylethylenediamine

TSA Tryptic soy agar

TSB Tryptic soy broth

TTC 2,3,5-triphenyltetrazolium chloride

BRIEF DESCRIPTION OF THE FIGURES

[0012] Fig. 1 Changes in total PST concentration determined by LC on AE incubated with isolate C20C at 25°C over 5 d (mean ± SD, n = 2). Fitted exponential decay regression: y = 79.96 e ^~ ' ^28t ; r ² = 1.0 (p<0.01), where t = incubation time (d).

[0013] Fig. 2 Linearized plot of changes in total PST concentration determined by LC on AE incubated with isolate C20C at 25°C over 5days (mean ± SD, n = 2). Graphical determination of D value is illustrated.

[0014] Fig. 3 Molar composition of individual PSTs in AE at day 0.

[0015] Fig. 4 Changes in NEO concentration determined by LC on AE incubated with isolate C20C at 25°C over 5 d (mean ± SD, n = 2). Fitted exponential decay regression: y = 38.68 e ^{~ 2 1Ot} ; r ² = 1.00 (p<0.01), where t = incubation time (d). Insert shows initial molar composition. [0016] Fig. 5 Changes in STX concentration determined by LC on AE incubated with isolate C20O at 25°C over 5 d (mean ± SD, n = 2). Fitted exponential decay regression: y = 43.45 e ^~ ' ^02t ; r ² = 0.99 (p<0.01), where t = incubation time (d). Insert shows initial molar composition.

[0017] Fig. 6 Changes in GTX 2 concentration determined by LC on AE incubated with isolate Cl IO at 25°C over 5 d (mean ± SD, n = 2). Fitted exponential decay regression: y = 3.09 e ^" ° ^38t ; r ² = 0.98 (p<0.01), where t = incubation time (d). Insert shows initial molar composition.

[0018] Fig. 7 Changes in Cl toxin concentration determined by LC on AE incubated with isolate C3O at 25 ⁰ C over 5 d (mean ± SD, n = 2). Fitted exponential decay regression: y = 0.03 e ^~ ° ^76t ; r ² = 0.97 (p<0.01), where t = incubation time (d). Insert shows initial molar composition.

[0019] Fig. 8 Changes in C2 toxin concentration determined by LC on AE incubated with isolate C3O at 25 ⁰ C over 5 d (mean ± SD, n = 2). Fitted exponential decay regression: y = 0.17 e ^{~ O 55t} ; r ² = 0.97 (p<0.01), where t = incubation time (d). Insert shows initial molar composition.

[0020] Fig. 9 Changes in GTX 3 concentration determined by LC on AE incubated with isolate C3C at 25°C over 5 d (mean ± SD, n = 2). Fitted cubic regression: y = 1.83 + 0.18t - 0.28t ² + 0.04t ³ ; r ² = 0.98 (p<0.01), where t = incubation time (d). Insert shows initial molar composition.

[0021] Fig. 10 Changes in dcGTX 3 concentration determined by LC on AE incubated with isolate C3C at 25°C over 5 d (mean ± SD, n = 2). Fitted cubic regression: y = 1.09 + 0.2Ot -

0.24t ² + 0.03t3; r ² = 1.00 (pO.Ol), where t = incubation time (d). Insert shows initial molar composition.

[0022] Fig. 11 Changes in dcGTX 2 concentration determined by LC on AE incubated with isolate C20C at 25°C over 5 d (mean ± SD, n = 2). Fitted cubic regression: y = 1.12 - O.Olt -

0.19t ² + 0.03t ³ ; r ² = 0.98 (p<0.01), where t = incubation time (d). Insert shows initial molar composition.

[0023] Fig. 12 Changes in total PST concentration determined by LC on the AE incubated with isolate C20O at 25 ⁰ C over 5 d (mean ± SD, n = 2) graphed with logio CFU/ml for the same isolate under the same growth conditions (mean ± SD, n = 2). ^a T = toxin; isolate grown with AE, ¹⁷ NT = no toxin; isolate grown without AE.

[0024] Fig. 13 Unrooted phylogenetic tree of the eight PST-degrading isolates (indicated in bold text) based on near full length 16S rDNA sequences. The bar represents a 2 % sequence divergence. Bootstrap values are shown at the branch points.

[0025] Fig. 14 Unrooted phylogenetic sub-tree of the eight PST-degrading isolates (indicated in bold text) based on near full length 16S rDNA sequences. The bar represents a 0.2 % sequence divergence. Bootstrap values are shown at the branch points.

[0026] Fig. 15 DNA ladder and plasmid profiles of the eight PST-degrading isolates run on

1.2 % agarose gel.

[0027] Fig. 16 RISA PAGE profiles of the eight PST-degrading isolates. Resolving gel: 6 %

T, 3.3 % C; Stacking gel: 3.5 % T, 3.3 % C.

[0028] Figure 17 Chart based on principal component analysis of the cellular fatty acid compositions of the eight PST-degrading isolates. Isolates in different quadrants have significantly different fatty acid profiles (p < 0.05).

[0029] Fig. 18 SDS-PAGE whole-cell protein profile patterns of the eight PST-degrading isolates run with Broad Range SDS-PAGE Molecular Weight Standards. Resolving gel: 12

% T, 2.6 % C; Stacking gel: 6 % T, 2.6 % C.

[0030] Fig. 19 SDS-PAGE LPS profile patterns of the eight PST-degrading isolates run with

Broad Range SDS-PAGE Molecular Weight Standards. Resolving gel: 10 % T, 2 % C;

Stacking gel: 4.5 % T, 2 % C.

[0031] The examples described in the figures above are not intended to limit the scope of the present invention.

BRIEF DESCRIPTION OF SEQUENCE LISTINGS SEQ.ID No.l is the sequence of the 16S rDNA from bacterial isolate C3C SEQ.ID No.2 is the sequence of the 16S rDNA from bacterial isolate C3O SEQ.ID No.3 is the sequence of the 16S rDNA from bacterial isolate ClOC SEQ.ID No.4 is the sequence of the 16S rDNA from bacterial isolate ClOO SEQ.ID No.5 is the sequence of the 16S rDNA from bacterial isolate CI lC SEQ.ID No.6 is the sequence of the 16S rDNA from bacterial isolate CI lO SEQ.ID No.7 is the sequence of the 16S rDNA from bacterial isolate C20C SEQ.ID No.8 is the sequence of the 16S rDNA from bacterial isolate C20O SEQ.ID No.9 is the 16S ribosomal DNA (rDNA) primer S-D-Bact-0008-a-S-20 SEQ.ID No.10 is the 16S ribosomal DNA (rDNA) primer S-*-Univ-1492-a-A-19 SEQ.ID No.l 1 is the 16S ribosomal intergenic spacer analysis primer B 1055 SEQ.ID No.12 is the 16S ribosomal intergenic spacer analysis primer 23 SOR SEQ.ID No.13 is the universal primer lόSInternal for sequencing the 16S rRNA internal region of the gene

DETAILED DESCRIPTION OF THE INVENTION [0032] Screening bacterial isolates for PST degradation

[0033] Seventy-three isolates were recovered from toxic shellfish, nineteen of which demonstrated the potential ability to individually degrade PSTs during a presumptive screening test for changes in the LC PST profile. Those nineteen isolates were then subjected to the confirmatory screening protocol for PST degradation by MBA. Throughout preliminary MBAs, four isolates demonstrated the greatest PST-degrading potential. However, during this preliminary confirmatory screening, each of the four 'pure' aforementioned cultures appeared to develop two colony phenotypes: opaque (the original and dominant colony type) and translucent (the apparent new colony type). At the time of the observation, the two colony phenotypes of each isolate were separated and treated as different cultures. The new identification scheme for the eight isolates was as follows: C3O ('O' = opaque), C3C ('C = clear/translucent), ClOO, ClOC, Cl 10, Cl 1C, C20O, and C20C. The final confirmatory screening by MBA was conducted on these eight isolates. The fact that only four isolates from several hundred live toxic shellfish samples supplied by the Canadian Food Inspection Agency were capable of PST degradation, suggests that the presence of PST- degrading bacteria are relatively rare in nature.

[0034J The experimental evidence indicated the successful degradation of PSTs by the eight selected bacterial isolates. The bacteria isolated in this study reduced the toxicity of the total PST concentrations in the AE by at least 90 % in 3 d. The initial total PST concentrations used in the experiments translated to 718 μg STX eq./100 g edible tissue. Assuming a minimum total toxicity reduction of 90 %, the samples would have contained 71.8 μg STX eq./100 g after 3 days of incubation, which is under the regulatory limit of 80 μg STX eq./100 g for safe consumption in North America. This is the first evidence demonstrating bacterial degradation of STX, NEO, dcGTX 2, and dcGTX 3. As STX and NEO are the two most potent PSTs, these findings are very significant. In addition, we are reporting significant degradation of GTX 2, GTX 3, Cl, and C2, which often comprise a large proportion of shellfish toxicity in temperate waters.

[0035] Current evidence suggests that the biotransformation and/or degradation of PSTs by bacteria are largely enzyme-mediated. PST-specific enzymes include decarbamoylases and sufotransferases. In addition, reductive elimination by natural reductants can result in transformations and de novo appearance of specific PST analogues. The AE contained carbamate (STX, NEO, GTX 2/3), decarbamoyl (dcGTX 2/3), and N-sulfocarbamoyl (C1/C2) toxins. Degradation of the PSTs was monitored every 24 h by LC. Daily changes in the toxin concentrations (μM STX eq.) were used to propose potential mechanisms of degradation.

[0036] The daily changes in the individual PST profiles of the AE show more evidence of degradation rather than toxin interconversions. There was no significant evidence of reductive elimination of C-11 hydroxysulfate or N-I hydroxyl moieties. Reductive eliminations may not have been favored in the aerobic conditions of the present study as all of the competent bacteria studied to date have been shown to be obligate aerobes. Acid hydrolysis of N-sulfocarbamates was also unlikely due to the relatively mild conditions employed during preparation of the bacterial extracts for LC. Epimerization may have occurred under the neutral conditions of incubation, but, like other spontaneous toxin transformations, it was difficult to monitor and distinguish them from other interconversions in a mixture of PSTs.

[0037] The rate at which a bacterium is capable of degrading PSTs is a characteristic of paramount importance. Although enzymatic digestion is often very complex and involves multiple mechanisms, knowledge of the kinetics can provide insight into the pathway(s) responsible for degradation. In this particular analysis, the D values obtained from the degradation curves are also valuable tools in determining which cultures possess more

effective degradative metabolisms.

[0038] For all eight isolates, the total PST concentrations were found to degrade exponentially with time and were appropriately fitted to general exponential decay equations. The overall kinetics of degradation was first order and the relationship between the toxicity and time on plots of logio total PST concentration versus time was highly linear. This tells us that, in all cases, incubation time is an excellent predictor of overall toxicity of the AE for any of the competent bacteria tested. The time required to reduce total toxin concentrations by 90 % ranged from -1.5-3 d. Isolates C20O and C20C appeared to perform the best, with respect to total PST degradation rates, having the lowest D values of 1.41 days and 1.61 d, respectively.

[0039] The overall kinetics of degradation were largely determined by STX and NEO, as they were responsible for the AE toxicity and experienced almost complete degradation in most cases. Like STX and NEO, the kinetics of GTX 2, Cl and C2 were also first order with respect to time and also closely resembled the overall PST degradation kinetics. Although the kinetics were first order, the pathway(s) involved may be so complex that the resultant behavior is indistinguishable from that of first order reactions. However, it is likely that the overall degradation would have looked very different and perhaps not first order had different PSTs been present in different concentrations in the AE. Altering the temperature, pH, and reaction medium would also have significantly affected the degradation kinetics. [0040] Preliminary experiments showed that the bacteria would not grow on PSTs alone when tested with purified STX, NEO, GTX 1, GTX 4, GTX 2, GTX 3, Cl, and C2 at levels of 3.34, 3.21, 3.18, 0.68, 2.64, 0.67, 3.34 and 30.1 μM STX eq., indicating the requirement of an additional carbon source for growth. The lack of growth in the absence of a supplementary carbon source suggests that co-metabolism may occur. Co-metabolism in the metabolism of an organic compound in the presence of a growth substrate which is used as the primary carbon and energy source. Under these circumstances, the isolates would grow on the preferred carbon source in the media (main substrate) and catabolize the co-substrate (PSTs) gratuitously. Additionally, the filter-sterilized mussel extract (ME) included in the growth medium was hypothesized to contain cofactors and/or coenzymes required by the bacteria in order to degrade and/or transform the toxins. Although the PSTs were shown to be unaffected by ME alone, the ME provided additional nutrients and carbon to the isolates and the PSTs were largely degraded during the 5 d incubation. Since PSTs were not used as sole carbon sources in preliminary experiments and other studies, it is improbable they were used preferentially for growth and more likely that they were degraded as a result of co-

metabolism, possibly in the presence cofactors and/or coenzymes in the ME. [0041] According to nearly full length 16S rDNA sequences, basic phenotypic characteristics, and fatty acid profiles, all of the isolates were characterized as Pseudoalteromonas species. Members of Pseudoalteromonas are readily cultivated from marine environments and have been isolated from sea water, marine invertebrates, sediments, marine algae, and sea ice as well as salted foods. The genus Pseudoalteromonas are Gram- negative, straight rods that are motile by means of sheathed or unsheathed polar, bipolar, or lateral flagella. Colonies may be pigmented or nonpigmented. Nonpigmented species tend to share a very high 16S rDNA sequence similarity; indeed the sequences of several species have been known to diverge by less than 2 %, which is within the range of sequence divergence for a single species. Members of the genus are chemoheterotrophic and possess a strictly respiratory metabolism with no growth occurring anaerobically. An arginine dihydrolase system is usually absent. All species decompose gelatin and Tween 80 and require sodium ions for growth. Many species are capable of metabolizing a wide range of carbon substrates. They can grow in temperatures ranging from 0-40°C and most nonpigmented strains can tolerate 9-12 % NaCl.

[0042] Several morophological and phenotypical characteristics of the PST-degrading Pseudoalteromonas sp. were determined. The cells were generally curved rods surrounded by outer membrane vesicle (OMV) networks of varying sizes and configurations. OMVs have been found extensively in Gram-negative bacteria and have been mainly studied in relation to their roles in protein trafficking with a focus on the transport of virulence factors by pathogens. Bacteria are unlikely to produce OMVs without function and they might play a protective role in non-pathogenic bacteria by reducing intracellular levels of toxic compounds.

[0043] The most important characteristics with respect to biodegradation may be the presence of a strict aerobic metabolism and oxidase and catalase production. In general, the most rapid and complete degradation of many organic compounds is an oxidative process catalyzed by oxidases and peroxidases. Furthermore, the induction of anaerobic conditions involving bacterial transformation of PSTs seems to enhance toxicities due to increases in reductive reactions.

[0044] Several of the PST-degrading isolates were found to contain plasmids. Plasmids are important genetic components of pseudomonads and often confer resistance to antibiotics, chemical and physical agents, bacteriophage propagation, and bacteriocins. In addition, degradative plasmids have been widely reported within genus Pseudomonas, allowing some

species to degrade unusual carbon sources including environmental pollutants and hydrocarbons. Plasmids were isolated from six of the bacteria that may encode enzymes capable of degrading or assisting in the degradation of PSTs and related compounds. Isolates CI lO and CI lC did not contain any plasmids and still degraded 90 % of the PSTs in the AE in 3 d. However, the other isolates degraded the PSTs to a slightly higher degree (-95 % in 3 d), largely due to the complete degradation of STX, a small percentage (-13-15 %) of which still remained in the AE after incubation with isolates CI lO and CI lC. The evidence suggests that plasmid-encoded enzymes are involved in, but not entirely responsible for, the degradation of PSTs in vivo, particularly STX.

[0045] The occurrence of opaque (O) and translucent (C) phenotypes was a major point of differentiation among the isolates. The development of two colony phenotypes was not noticed until part way through the project, when they were separated and treated as individual isolates. In all four original cultures, the dominant phenotype was O. The phenomenon was initially observed in isolates ClO and CI l, where there was a much higher occurrence of the C phenotype than in the other two isolates. It was more difficult to isolate C colonies from the C3 and C20 cultures, as they rarely appeared on the plates when streaked from glycerol stocks. After separating the two phenotypes, there was no spontaneous 'switching' between them during sub-culturing, even when the isolates were streaked onto marine agar (MA) from liquid cultures. This 'switching' between O and C phenotypes, a type of phase variation, has been known to occur in several pathogenic Vibrio sp. of marine origin. In phase variation, the expression of a certain factor is periodically altered so that it is either 'on' or 'off. Factors commonly controlled by phase variation include structures like flagella, fimbriae, outer membrane proteins, and exo- and lipopolysacchardies that are associated with the cell surface. Phase variation can occur at different frequencies depending on the type of media and growth conditions employed. The transition from O to C phenotypes occurs more readily than the reverse which is less frequent and sometimes not observed at all. This particular type of phase variation has been attributed to cell organization within a colony, production of extracellular polysaccharides, and variations in capsular polysaccharide expression. [0046] The morphology of the O and C phenotypes for isolates ClO and CI l were roughly the same with the exception of a slightly higher occurrence of filamentous cells in the C cultures. The morphology of the two phenotypes for C3 and C20 showed a much greater discrepancy; the O cultures contained a combination of straight and slightly curved rods while the translucent contained shorter, straight rods and filamentous cells, respectively. The variation in the O and C colonies could have been caused by cellular organization; the

morphologically different cells in the C colonies might have been 'packed' more closely together resulting in a differential transmission of light. In addition, the opaque colonies might have produced more abundant levels of an EPS, which is not uncommon in Pseudoalteromonas sp. and other marine bacteria. However, images captured did not show any evidence of this. Neither phenotype contained capsules, so variation in capsular polysaccharide expression is unlikely. In any case, the frequency of phenotype switching was so low that it was not observed during routine sub-culturing, and the phenotypes could be treated as separate isolates. Although the occurrence of the two different phenotypes is of interest from a microbiological standpoint, it does not have a significant effect on the PST- degrading capabilities of the bacteria.

[0047] There were no major differences observed among the phenotypic tests performed on the eight PST-degrading isolates. There were some discrepancies, however, in maximum growth temperatures, casein hydrolysis, carbon substrate utilization (Biolog GN microplates), and plasmid profiles. All isolates grew at a maximum temperature 34°C, except for C3C and C20C, which grew to 33 ⁰ C. With respect to plasmid profiles (see Figure 15), the profiles between the O and C phenotypes for each original culture were the same. Plasmid profiling showed unique profiles for Cl 1 and C20 and the C3 and ClO cultures shared almost identical profiles. Isolate C3C did not produce a caseinase, whereas as the others did so, albeit to varying degrees. There were some differences in carbon utilization profiles among the isolates. Caution must be exercised, however, when interpreting Biolog results as the microplates were read visually and the designation of a positive result was somewhat subjective.

[0048] The definitive method of differentiating among the isolates was the 16S rDNA sequencing. The distance matrix calculated a mere 0.2-1 % sequence divergence among the isolates. A value of 3 % is often used as the maximum sequence divergence for a single species. Based on this cut-off, as well as the phenotypic analyses, all of the isolates appear to be strains of the same species. Yet, looking at the distance matrix, there is also less then 1% sequence divergence among the PST-degrading isolates and sequences in ARB database for P. haloplanktis, P. espejiana, P. carrageenovora, P. distincta, P. translucida, and three unspeciated members of the genus. However, as mentioned previously, the nonpigmented Pseudoalteromonas share a high sequence similarity, as six Pseudoalteromonas sp., possess sequences that diverge less than 2 %.

[0049] The six aforementioned isolates have been designated their own species classification based on specific phenotypic characteristics differentiating them from the rest of the genus.

Major differential phenotypic characteristics of these species are compared with those of the PST-degrading isolates in Table 1. Based mainly on morphology, the hydrolysis of specific compounds, growth temperatures, and NaCl tolerance, the PST-degrading isolates are phenotypically identified as Pseudoalteromonas haloplanktis.

[0050] Table 1 Phenotypic characteristics distinguishing the PST-degrading strains from closely related and previously described Pseudoalteromonas sp. All strains listed are aerobic, Gram-negative rods, require Na ⁺ ions for growth, grow in 3-6 % NaCl, are positive for oxidase and catalase, negative for arginine dihydrolase, nitrogen reduction/denitrification, indole production, and agarase, and positive for gelatinase, lipase, and DNAse. +, positive; -, negative; +/-, variable reaction; nd, no data available

C

Characteristics

Pigments - - - - - - - - - - - -

Melanin-like - - - - - - - - - + - - - +

Flagellation

P P P P P P P P P P P P B P,L (unsheathed)

Production of

Urease + + + + + + + + + - + nd nd nd

Caseinase + - + + + + + + + + + + + +

Amylase - - - - - - - - +/- H- + + + -

Alginase - - - - - - - - - + + + + + κ-carrageenanase - - - - - - - - - + - - nd -

Growth at/in

4°C + + + + + + + + - + +/- - + - e ^y ao

34°C + - + + + + + - + + + + nd nd

35°C - + + + nd nd

37°C - - - - - - - - - + + + - -

8% ^" NaCl + + + + + + + + + + + + + +/-

9% NaCl + + + + + + + + + + + + - -

12% NaCl + + + + + + + + + + + -

Utilization of

D-glucose + + + + + + + + + + + + - +/-

D-mannose - - - - - - - - - - + +

D-galactose + - + + + + + - +/- - + + + -

D-fructose + + + + + + + + +/- + +/- + nd +

Sucrose - - +/- + + + nd +

Maltose + + + + + + + + + + + + - -

Mehbiose - - - - - - - - - + + +/- - +/-

Lactose - - - - - - - - + + + - +/-

D-mannitoI + + + + + + + + +/- + +/- + + +/-

D-gluconate + + + + - + + +/- - - +/- nd -

Citrate - - - + + +/- + + +/- - +/-

Glycerol - - - - - - - - +/- + +/- +/- nd +

Trehalose - - - +/- - + - nd -

D-glucosamine - - - - - - - - +/- +/- - - nd -

[0051] PST-contaminated blue mussels and soft-shell clams seem to provide an enriched environment for PST-degrading bacteria. The eight strains isolated from toxic bivalves reduced the overall toxicity of an AE containing STX, NEO, GTX 2, GTX 3, dcGTX 2, dcGTX 3, Cl, and C2 by 90.0-98.5 % within 3 d, depending on the isolate. The majority of the toxicity reduction was due to the nearly complete destruction of NEO and STX, although the other toxins also experienced significant degradation throughout the incubation period. There was little evidence of PST-specific biotransformations described in similar previous studies. The exception was possible enzymatic decarbamoylation indicated by a decrease in C2 and concurrent increase in dcGTX 3 on days 4 and 5 of incubation. Nonetheless, it is likely that the degradation was enzymatic and, under the aerobic culture conditions employed, largely catalyzed by oxygen-demanding enzymes. Metabolites of toxin breakdown may enter established bacterial degradative pathways such as purine catabolism. As PSTs are largely biosynthesized from arginine molecules, the AST pathway of arginine catabolism may also be involved in their destruction.

[0052] For all isolates, the degradation of total PST concentrations was appropriately fitted to an exponential decay model and the overall kinetics appeared to be first order, as was the case for NEO, STX, GTX 2, Cl, and C2. The degradation of toxins GTX 3, dcGTX 2, and dcGTX was not first order and was best described by cubic functions. Changing the incubation conditions and composition of the AE would likely affect the overall kinetics of degradation. All isolates experienced growth, albeit to various degrees, in the presence of PSTs. As the isolates were not found to grow on PSTs as a sole carbon source in preliminary studies, co-metabolism in the presence of a primary carbon source in the media is possible. In addition, the ME included in the growth medium might have provided cofactors and/or coenzymes.

[0053] Although all isolates were easily cultured and drastically reduced the toxicity of the PSTs in a reasonable amount of time, some appeared to perform better than others. Overall, isolates C20O and C20C achieved the most complete PST destruction in the shortest amount of time, reducing the overall toxicity of the AE by 96.8 and 98.5 % in 3 d, respectively, with D values of 1.41 and 1.61 d, respectively. Although we are reporting the successful degradation of PSTs by single isolates, it is more likely that a mixed microbial community would have the greatest biodegradative potential. With the increased genetic diversity offered by a consortium of microorganisms, there is a greater chance that they collectively possess the enzymatic capabilities to degrade more toxins than one isolate alone. Although almost complete degradation of the toxicity was observed within 3 d, the use of a bacterial

community of 'PST degraders' might result in more rapid PST degradation, at least in vitro. [0054] Having these bacteria that were isolated from naturally occurring populations of shellfish allows exploitation of the mutualistic relationships established between these organisms. Shellfish with higher than normal concentrations of these isolated bacteria capable of degradation of PSTs would provide proactive (probiotic) protection against paralytic shellfish poisoning (PSP).

[0055] The eight PST-degrading isolates are likely strains of Pseudoalteromonas haloplanktis, based on similar 16S rDNA sequences and phenotypic characteristics. Further testing might warrant the separation of the isolates into different strains. The O and C variants of the four original isolates may be differentiated initially by colony morphology and subsequently by plasmid profiles, cell morphology, or carbon substrate utilization profiles.

[0056] The LC analyses indicated that all bacterial isolates significantly reduced the total PST concentration of the AE by greater than 92 % over the 5 days incubation period. The degradation of the toxins was concurrently confirmed by MBA. These theoretical and actual total PST concentrations (μM STX eq.) obtained from the MBA and LC data, respectively, for each bacterial isolate during the confirmatory screening are given in Tables 6.1 and 6.2. For all isolates, the theoretical changes in total PST concentration could be adequately described by a first order general exponential decay equation:

[0057] A = A _o e ^{" kt} ; (1)

[0058] where 'A' is the final total PST concentration (μM STX eq.) in the growth medium after 5 days incubation; 'A ₀ ' is the initial total PST concentration (μM STX eq.) in the growth medium; 'k' is the reaction rate constant (days ^"1 ); and 't' is the number of incubation days

(d). The exponential decay regression equations for total PST degradation by each isolate can be seen in Table 6.3. Figure 1 illustrates a representative regression curve for total PST degradation by isolate C20C.

[0059] A plot of log io [A] versus time (excluding baseline values) for each isolate gives a straight line described by the following equation:

[0060] A = mt + b; (2)

[0061] where 'A' is the logio of the total PST concentration (μM STX eq.) in the growth medium; 'm' is the slope (μM STX eq.d ^'1 ); 't' is incubation time in days (d); and 'b' is the y intercept (μM STX eq.).

[0062] The slopes of the linearized degradation plots were determined in order to find the decimal reduction times (D value) at 25 ⁰ C for each isolate as follows:

[0063] D = -l/m ; (3)

[0064] In the context of these experiments, the D value is the time (d) at a given temperature (25°C) that is required to reduce total or individual PST concentrations by 90 % or one log cycle. Alternatively, D values can be obtained graphically from the plots of logio PST concentrations (μM STX eq.) versus time (d) as shown in Figure 2 using the daily LC data for isolate C20C.

[0065] Although both the LC analyses and MBAs demonstrated a clear reduction in overall toxicity, there did not seem to be any correlation between these two methods of toxicity quantification. The day 5 values for both methods appeared to be similar for each isolate; the average day 5 values from the LC analysis spanned averages of 1.19-5.28 μM STX eq., whereas the average day 5 values for the MBA ranged from 6.06-8.42 μM STX eq. As expected, the day 5 MBA values were slightly higher than those obtained by LC, likely due to hydrolysis of N-sulfocarbamoyl toxins to carbamates during MBA sample preparation. On the other hand, there were large, inexplicable discrepancies in the initial (day 0) PST concentrations between the LC and MBA data.

[0066] Table 6.1 Changes in total PST concentrations determined by LC (μM STX eq.) on AE incubated with isolates at 25 ⁰ C over 5 days (mean ± standard deviation (SD), n = 2)

Isolate Total PST concentration (μM STX eq.) on days 0- 5 Total

0 1 2 3 4 5 degradation (%)

C3O 83.64 ± 27.70 ± 9.48 ± 2.93 ± 2.74 ± 3.10 ± 96.25 ± 1 1.90 15.82 3.67 0.38 0.50 0.13 0.69 C3C 74.64 ± 58.40 ± 8.14 ± 4.88 ± 2.25 ± 2.46 ± 96.72 ± 7.43 37.60 5.64 3.15 0.20 0.58 0.45

ClOO 57.05 ^a 10.24 3.08 1.64 1.81 1.80 96.84

ClOC

CI lO 90.03 ± 48.98 ± 12.60 ± 9.00 ± 8.03 ± 6.94 ± 92.28 ±

5.76 9.30 1.05 1.47 8.26 0.28 0.19

CI lC 89.66 ± 40.15 ± 9.90 ± 8.85 ± 3.50 ± 5.28 ± 94.08 ±

7.05 2.80 5.52 1.43 3.22 0.26 0.76

C20O 80.10 ± 16.87 ± 3.11 ± 2.57 ± 2.53 ± 2.42 ± 96.96 ±

4.14 1.58 0.07 0.13 0.47 0.35 0.60

C20C 79.84 ± 23.04 ± 4.60 ± 1.18 ± 1.14 ± 1.19 ± 98.51 ±

5.68 0.61 1.89 0.21 0.06 0.01 0.09 ^a Data lost due to error (one replicate) bData lost due to error (both replicates)

[0067] Table 6.2 Changes in total theoretical PST concentrations determined by MBA (μM STX eq.) on AE incubated with isolates at 25°C over 5 d. Day 0 toxicity values for all extracts are based on T (toxic) control. MBAs were conducted only on days 0 and 5 (mean ± SD, n = 2)

Isolate Total PST concentration (μM STX eq.) on days 0 & 5 0 5

C3O 20.47 ± 1.85 7.38 ± 1.17

C3C 20.47 ± 1.85 8.23 ± 0.58

ClOO 20.47 ± 1.85 6.06 ± 1.07

ClOC 20.47 ± 1.85 6.72 ± 0.46

CI lO 20.47 ± 1.85 6.94 ± 1.17

CI lC 20.47 ± 1.85 6.54 ± 0.31

C20O 20.47 ± 1.85 8.42 ± 0.98

C20C 20.47 ± 1.85 7.30 ± 1.26

[0068] Table 6.3 Exponential decay regression equations for the degradation of total PST concentrations in AE incubated with the PST-degrading isolates at 25 ⁰ C over 5 days (mean ± SD, n = 2), where t = incubation time (d)

Isolate Total PST concentrations

C3O y = 83.54 e ^" ' ^09t ; r ² = 1.00 (p<0.01)

C3C y = 79.76 e ^" ° ^68t ; r ² = 0.90 (p<0.01)

ClOO y = 56.99 e ^" ' ^66t ; r ² = 1.00 (p<0.01)

CI lO y = 91.06 e ^{~ 0 74t} ; r ² = 0.98 (p<0.01)

CI lC y = 90.02 e ^{" 088t} ; r ² = 0.99 (p<0.01)

C20O y = 80.07 e ^" ' ^55t ; r ² = 1.00 (p<0.01 )

C20C y - 79.96 e ^" ' ^28t ; r ² = 1.00 (p<0.01)

[0069] Table 6.4 gives the slopes, D values, and coefficients of determination (r ) for the linearized exponential decay regression curves of total and individual theoretical PST concentration versus time. Like the daily LC data for total PST concentration, that of NEO, STX, GTX 2, Cl and C2 appear to be adequately described by the general exponential decay equation, as indicated by significant r ² values (p < 0.05). Representative regression curves for PST destruction by isolate C20C are given in Figures 4-8. Incorporated into each figure is a pie chart insert showing the percent molar contribution of the individual PST to the AE at day 0, as indicated by an exploded slice. Figure 3 shows a labeled pie chart of the percent molar composition of each PST in the AE at day 0. The r ² values of the linear regression curves for the degradation of all PSTs by all isolates were either highly significant (p < 0.01) or significant (p < 0.05), with the exception of the degradation of STX by isolate ClOO (p > 0.05).

[0070] NEO and STX are the most potent of the PSTs and, as illustrated in Figure 3, together accounted for approximately 75 % of the total toxicity of the AE. All of the isolates reduced NEO concentrations by 90 % in less than 12 hours (no linear regression could be conducted on NEO data for isolates ClOO and C20O, as they reduced NEO to undetectable levels in less than 24 h). Similarly, all isolates except CI lO and CI lC reduced STX concentrations by 90 % in less than 24 h. GTX 2, Cl, and C2 comprised significantly less of the total PST concentration in the AE; approximately 11, 0.32, and 5 %, respectively. D values for GTX 2 were 6-7 d, whereas Cl and C2 toxins required 2.5-4 days of incubation with the isolates in order to achieve 90 % toxin degradation.

[0071] The degradation data for toxins GTX 3, dcGTX 3 and dcGTX 2 were not suitably described by the general exponential decay equation and were more appropriately fitted to cubic functions, with the exception of the degradation of dcGTX 2 by isolate ClOO, which was fitted to a quadratic function. The r ² values of the regression curves were all highly significant (p < 0.01). The degradation of these PSTs was generally characterized by a decrease in PST concentration until approximately day 4, after which there was typically a slight increase in toxicity. GTX 3 accounted for approximately 7 % of the total PST concentration in the AE, whereas dcGTX 2 and 3 together comprised less than 1.5 %. The regression equations and representative regression curves for the degradation of GTX 3, dcGTX 3, and dcGTX 2 can be seen in Table 6.5 and Figures 9-11, respectively.

[0072] Table 6.4 Slopes (m), D values, and coefficients of determination (r) for the log _I0 individual and total PST concentrations (μM STX eq.) versus time (d) plots for the degradation of PST concentrations determined by LC on AE incubated with PST-degrading isolates at 25°C after 5 d (n = 2 ± SD , except isolate ClOO where n = 1). Data for isolate C 1 OC were lost due to error.

Extracts NEO STX GTX 2 Cl C2 Total PSTs ^a

C3O m" = -2.26 m = -1.44 m = -0.14 m = -0.38 M = -0.25 m = -0.48

D ^L = 0.44 D = 0.69 D = ^" 7.14 D = 2.63 D = 4.00 D = 2.07

R ² = 0.88 (p r = 0.77(p< r ² = 0.99 (p < ^• r = 1.00 (p < ^• R ² = 0.98 (p r ² = 1.00 (p

^• '0.01) 0.01) 0.01) 0.01) < 0.01) 0.01)

C3C m = -2.26 m = -1.03 m = -0.14 ND ^d ND m = -0.41

D = 0.44 D = 0.97 D = 7.I4 D = 2.43

R ² = 0.78 (p r ² = 0.76 (p < r ² = 0.92 (p < r ² = 0.94 (p

< 0.05) 0.01) 0.01) 0.01)

ClOO NR ^e m=-2.26 m = -0.17 m = -0.34 M = -0.30 m = -0.52

D = 0.44 D = 5.88 D = 2.94 D = 3.33 D = 1.94 r = 0.86(p> r ² = 0.99 (p < r ² = 0.97 (p < R ² = 0.96 (p r ² = 0.96 (p

0.05) 0.01) 0.05) <0.05) 0.05)

CIlO m = -2.30 m = -0.21 m = -0.16 m = -0.31 M = -0.25 m = -0.36

D = 0.43 D = 4.76 D = 6.25 D = 3.22 D = 4.00 D = 2.79

R ² - 0.82 (p r ² = 0.90 (p < r ² = 0.99 (p < r = 0.96 (p < R ² = 0.98 (p r = 0.95 (p

<0.05) 0.01) 0.01) 0.05) <0.01) 0.01) cue m - -2.29 m = -0.33 m = -0.16 m = -0.31 M = -0.29 m = -0.35

D = 0.44 D = 3.03 D = 6.25 D = 3.22 D = 3.44 D = 2.88

R ² = 0.84 (p r ² = 0.94 (p < r ² = 0.99 (p < r = 0.99 (p < R ² = 0.97 (p r ² = 0.95 (p

<0.01) 0.01) 0.01) 0.01) <0.05) 0.01)

C20O NR ^e m = -2.30 m = -0.14 ND ND m = -0.70

D = 0.43 D = 7.14 D= 1.41 r ² = 0.83 (p < r = 0.88(p< r = 1.00 (p

0.05) 0.01) 0.01)

C20C m - -2.30 m = -1.46 m = -0.16 m = -0.37 M = -0.40 m = -0.62

D - 0.43 D = 0.68 D = 6.25 D = 2.70 D = 2.5 D= 1.61

R ² = 0.89 (p r ² = 0.84 (p < r= 0.90 (p R ² = 0.94 (p R ² = 0.90 (p r ² = 1.00 (p

< 0.01) 0.01) < 0.01) <0.05) <0.05) 0.01) ^d λlso incϊudes ^~ dcGTX 3 aπcfdcGTX 2, ^b Units for m are μM STX ^~ eq d ^'1 , ^c Units for D are d, ^d ND, no data. ⁰ NR, no linear regression performed (not enough data points).

|0073] Table 6.5 Regression equations for the degradation of GTX 3, dcGTX 3 and dcGTX 2 concentrations determined by LC on AE incubated with PST-degrading isolates at 25 ^C C after 5 days (n =2 ± SD . except isolate ClOO where n = 1). Data for isolate ClOC were lost due to error.

^~ Exlraet GTX 3 deGTX ^~ 3 dcGTX2

C3O y = 1.76- 0.211 -0.08t ² + y = 1.07 - 0.13t - 0.05t ² + Y = 1.23 - 0.16t - 0.04t ² + 0.02t ³ O.Olt ³ O.Olt ³ r = 0.98(p<0.01) r ² = 0.92 (p<0.01) r ² = 0.79 (p<0.01)

C3C y = 1.83 + 0.18t-0.28t ² + y = 1.09 + 0.2Ot - 0.24t ² Y = 1.27 + 0.14t - 0.23t ²

0.04t ³ + 0.03t ³ + 0.03t ³ r ² = 0.98(p<0.01) r ² = 1.00 (p<0.01) r ² = 0.97 (pO.Ol)

ClOO y = 1.16-0.36t + 0.02t ² + y = 0.74 + 0.03t - 0.13t ² Y = 0.87 + 0.21t - 0.12t ² 0.005t ³ +0.02t ³ + O.Olt ³ r ² = 0.97(p<0.01) r ² = 0.68 (pO.Ol) r ² = 0.76 (p<0.01)

CIlO y = 1.85-0.24t-0.12t ² + y = 1.04 - O.Olt - 0.13t ² + Y = 1.39 - 0.27t - 0.03t ² + 0.02t ³ 0.02t ³ O.Olt ³ r = 0.99(p<0.01) r ² = 0.98(p<0.01) r ² = 0.88 (p<0.01)

CIlC y = 1.93 -0.23t-0.12r+ y = 0.51 + 0.76t - 0.39t ² Y= 1.46-0.33t-0.03t ²

0.02t ^J + 0.04t ³ r ² = 0.93(p<0.01) r ² = 1.00 (p<0.01) r ² = 0.95 (p<0.01)

C20O y= 1.70-0.57t + 0.10r- y = 1.03 - 0.56t - 0.2 It ² Y= 1.21 -0.67t-0.20t ² +

0.004f 0.03f O.Olt ³ r ² = 1.00(p<0.01) r = 0.97(p<0.01) r = 0.92(p<0.01)

C20C y= 1.55 -0.4It- O.Olt ² - y = 0.81 - 0.1 It - 0.2Ot ² + Y = 1.12 - O.Olt - 0.19t ² + O.Olt ¹ 0.03t ³ 0.03t ³ r = 0.99(p<0.01) r ² = 0.95(p<0.01) r ² = 0.98 (pO.Ol)

[0074] DIFFERENTIAL BACTERIAL DEGRADATION OF INDIVIDUAL AND TOTAL PST CONCENTRATIONS

[0075 J The PST degradation plots showing changes in the total PST concentrations over 5 days (see Figure 1 ) indicated that the majority of the total toxicity (> 90 %) disappeared within 3 d of incubation with the PST-degrading cultures. Most of this reduction in toxicity was due to the nearly complete destruction of NEO and STX within 2-3 d. As mentioned earlier. NEO and STX together comprised approximately 75 % of the total toxicity, and their

rapid degradation was therefore largely responsible for the total decrease in toxicity. After the first 3 d of incubation, the individual PSTs subjected to exponential decay decreased very little while some of those described by cubic functions increased slightly. Therefore, the theoretical individual and total PST concentrations were expressed in terms of percent degradation after 3 and 5 d of incubation with the PST-degrading cultures. These data are shown in Table 6.6.

[0076] Table 6.6 Average percent reduction of individual and total PST concentrations determined by LC on AE incubated with PST-degrading isolates at 25°C after 3 and 5 d (n = 2 ± SD . except isolate ClOO and toxins Cl and C2 where n = 1 ). Data for isolate ClOC were lost due to error.

Isolate N FO STX G I X 3 <%) G I X 2 (%) dcGTλ 3 dcGTX 2 Cl C2 Total PSTs

<%) <%) (%) <%) (%) ( %) (%) 3 d 5 d 3 d 5 d 3 d 5 d 3 d 5 d 3 d S d 3 d 5 d 3 d S d 3 d 5 d 3 d 5 d

C lO 100 100 100 100 53.4 58.5 64.8 73.4 57.2 39.5 49.4 25.2 87.1 89.2 81.5 84.6 96.4 96.2

± ± ± ± ± i- ± ± ±

0 00 0 00 0 00 0 00 7 15 S O l 7 19 1 98 13 9 13 9 20 7 19 0 0 96 0 69

C 1C 100 100 95.0 100 49.9 63.4 66.9 76.1 57.5 59.8 51 9 45.7 ND ¹ ND ND ND 93.6 96.7

± ± ± ± ± ± ± -t ± ± ± ± 0 00 0 00 7 1 1 0 00 3 35 5 24 2 50 1 91 5 33 17 5 5 89 13 5 3 58 0 45

C. KX) 100 100 100 100 67.4 57.6 68.5 77.4 100 55.8 20.4 34.1 89.9 91.3 86.8 89.7 97.1 96.8

C I K) 100 100 84.8 90.2 70.3 80.5 68.2 76.8 56.1 35.3 63.5 27.3 ND ND ND ND 90.0 92.3

± ± ± ± ± ± ± ± i ± ± ± ± 0 (K) 0 00 1 99 0 45 2 12 1 07 0 40 0 40 8 36 2 86 5 45 0 24 1 00 0 19

C 1 I C 100 100 87.2 91.2 64.5 81.4 67.1 84.8 42.6 100 43.0 67.2 86.2 95.1 81.8 94.1 90.0 94.1

± i ± ± ± ± ± ± ± ± ± ±

0 00 0 00 3 88 0 98 16 7 1 1 8 I 94 0 27 4 30 0 00 6 53 4 92 2 38 0 76

C 200 100 100 100 100 55.1 57.2 69.9 73.7 61.6 100 53.5 22.0 ND ND ND ND 96.8 97.0

± ± ± ± ± ± ± ± ± ± ± 0 00 0 00 0 00 0 00 4 47 6 51 2 30 4 82 5 02 0 00 7 57 20 1 0 32 0 60

C20C 100 100 100 100 80.2 100 79.5 81.0 100 100 73.8 40.8 92.2 99.5 88.4 99.3 98.5 98.5

± ± ± ± ± ± ± ± ± ± ± ± ± 0 00 0 00 0 00 0 00 3 45 0 00 1 59 2 22 0 00 0 00 4 98 0 14 0 16 0 09

¹ ND no daU

[0077] All isolates reduced 100 % of NEO concentrations within 3 d and all but three completely degraded STX within the same time frame. Only isolates CI l O and C I l C did not completely destroy STX concentrations by the end of the incubation period. Even so, they still degraded the majority of the STX (> 84 %) within 3 d. The isolates differed somewhat in their capacity to degrade GTX 3; isolate C20C seemed to be the most competent, reducing concentrations by approximately 80 % by day 3. whereas the remaining isolates degraded 50- 70 % by day 3. There were few differences in the percent degradation of GTX 2 among the isolates. In most cases, the extra 2 d of incubation resulted in only an additional 10 % degradation. Again, C20C seemed to be the most efficient in this respect, reducing the

concentration by almost 80 % in 3 d. In most cases, shortening the incubation time with dcGTX 3 and dcGTX 2 to 3 d resulted in more complete degradation of these toxins due to the slight increases in toxicity observed on days 4 and 5, likely due to enzymatic decarbamoylation of GTX 3 and GTX 2. The Cl and C2 toxins together only comprised about 1.5 % of the total toxicity of the AE. Nonetheless, over 80 % of these toxins were degraded in 3 d by all isolates for which data was available. In general, all isolates successfully degraded at least 90 % of the total toxicity within 3 d, and most achieved over 95 % toxin degradation over the entire 5 d incubation.

[0078] Each isolate was grown with and without the AE over the 5 d incubation period to examine the effects of PSTs on bacterial growth. To facilitate this comparison as well as relate bacterial growth to total PST degradation, the two sets of enumeration data were graphed on the same axis along with degradation data on a separate axis. The bacterial enumeration data for all eight PST-degrading isolates were similar and are represented by isolate C20O in Figure 12.

(0079] There did not appear to be any trends in the relationship between the enumeration data of the isolates grown with and without AE. While the growth of some isolates appeared slightly enhanced in the presence of PSTs. others seemed to grow slightly better in their absence with no apparent effects on overall toxin degradation. Most isolates achieved maximum cell numbers (CFU/ml) on day 2 or 3, followed by a subsequent decrease on days 4 and/or 5, regardless of the presence or absence of AE. The vast majority of the PST degradation took place during early stages of bacterial growth.

|0080] CHARACTERIZATION OF PST-DEGRADING BACTERIA [0081] During routine plating under standard culture conditions on MA, the four "O ^" isolates produced opaque, nonpigmented, circular, convex, smooth colonies with entire margins and mucoid textures. The colony diameters measured 1-1.5 mm. Under the same growth conditions, the four 'C" isolates produced translucent, non-pigmented, circular, convex, smooth colonies with entire margins and waxy textures. The diameter of the colonies was measured as 0.5-1.5 mm. Both colony phenotypes formed a white 'haze" around colonies after several days of incubation.

[0082] Phenotypic characteristics tested are summarized in Table 6.7 along with cell morphology, cultural characteristics and optimal growth conditions as described above. All eight PST-degrading isolates were Gram-negative and produced OMVs. They were positive for cytochrome oxidase, catalase, urease, gelatinase, lipase (Tween 80), DNAse, and

phosphotase. All isolates but C3C were positive for caseinase. However, isolates C l OC, C I l C. C20O. and C20C produced only weak reactions after 7 days of incubation. All isolates also produced H2S from cysteine. None of the isolates hydrolyzed esculin, starch, alginate, agar, or κ-carrageenan. None of the isolates possessed an arginine dihydrolase system nor were they capable of denitrification, glucose fermentation, or indole production.

[0083] Table 6.7 Cell morphology, cultural characteristics, phenotypic characteristics, and optimal growth conditions of the eight PST-degrading isolates. Culture conditions used for separate tests were as described in Section 5.1.2.

Characteristics C3O C3C ClOO ClOC CI lO CIlC C20O C20C

Gram stain GN ^a GN GN GN GN GN GN GN

Colony diameter after 1 - 0.5- 1 -1 .5 0.5-1 1 -1.5 0.5-1 1 -1.5 0.5-1

48 h (mm) 1.5 1

Cell shape St ^b St St St St St St Fl ^c

Average cell length

2 1 .5 5 4 3.5 15 ( uin)

Average cell diameter ] 1 1 1 1 1 1 ]

(μm)

Pigments d -

Unsheathed polar + + + + + + + + flagellum

Outer membrane + + + + + + + + vesicles

Motility + + + + + + + +

Obligate aerobe + + + + + + + +

Capsule - - - - - - - -

Oxidase test + + + + + + + +

Catalase test + + + + + + + +

Arginine dihydrolase - - - - - - - -

Nitrate reduction - - - - - - - -

Oxidation/fermentation of glucose

Production of:

Urease + + + + + + + +

B-glucosidase (esculin) - - - - - - - - ϋelatinase + + + + + + + +

Characteristics C3O C3C ClOO ClOC CIlO CIlC C20O C20C

Caseinase + - + + +**

Lipase (Tween 80) + + + + + + + +

Amylase - - - - - - - -

Agarase - - - - - - - -

Alginase - - - - - - - -

K-carrageenanase - - - - - - - -

DNAse + + + + + + +

Phosphotase + + + + + + + +

Indole - - - - - -

HTS production from + + + + + + + cysteine

Growth at/in:

4°C + + + + + + + +

33 ⁰ C + + + + + + + +

34°C + _ + + + + +

35°C - - _ _ _ pH 5 - - - - - - - - pH 5.5 + + + + + + + + pH 9.5 + + + + + + + + pH 10 - - - - - - - -

0 % NaCl - - - - - - - -

0.5 % NaCl + + + + + + + +

12 % NaCl + + + + + + + +

Optimal:

Temperature range ( ⁰ C) 25-30 25-30 25-30 25-30 25-30 25-30 25-30 25-30 pH range 7-8.5 7-8.5 7-8.5 7-8.5 7-8.5 7-8.5 7-8.5 7-8.5

NaCl range (%) 2-4 2-4 2-4 2-4 2-4 2-4 2-4 2-4

⁸ GN, Gram-negative

''straight or slightly curved rods

^filamentous d+, positive; -. negative

* growth in oxidative tubes was difficult to determine

** produced very weak reaction

[0084| Utilization of 95 carbon substrates was tested using Biolog GN microplates. Results of these tests are given in Table 6.2. All isolates utilized α-cyclodextrin, dextrin, glycogen, Tween 40. Tween 80, erythritol, D-fructose, α-D-glucose. maltose, D-mannitol. pyruvic acid methyl ester, succinic acid mono-methyl ester, acetic acid, β-hydroxybutyric acid, α-keto butyric acid, propionic acid, succinic acid, succinamic acid. L-alanine. L-alanyl-glycine, L- asparagine. L-aspartic acid. L-glutamic acid, glycyl-L-aspartic acid, glycyl-L-glutamic acid. L-proline. L-threonine, inosine, and uridine. No isolates utilized N-acetyl-D-galactosamine, N-acetyl-D-glucosamine, adonitol, L-arabinose, D-arabitol, D-cellobiose. L-fucose, gentiobiose, m-inositol, α-D-lactose, lactulose. D-mannose, D-melibiose. β-methyl-D- glucoside. D-raffinose, L-rhamnose, D-sorbitol, sucrose, D-trehalose, turanose. xylitol, formic acid, D-galactonic acid lactone, D-galacturonic acid. D-glucosaminic acid, D- gluciironic acid, α-hydroxybutyric acid, γ-hydroxybutyric acid, p-hydroxy phenylacetic acid, itaconic acid, α-keto glutaric acid, α-keto valeric acid. D.L-lactic acid, malonic acid, quinic acid. D-saccharic acid sebacic acid, glucuronamide. D-alanine, L-histidine, hydroxyl-L- proline. L-ornithine. L-phenylalanine, L-pyroglutamic acid. D-serine, D.L-carnitine, γ-amino

butyric acid, urocanic acid, phenyethylamine, putresine, 2-aminoethanol. 2.3-butanediol, glycerol. D.L-α-glycerol phosphate. The differential results are given in Table 6.8.

[0085] Table 6.8 Differential carbon substrate utilization profiles of the PST-degrading isolates based on Biolog-GN microplates. Plates were read visually after 48 hours incubation at 25 ⁰ C.

Carbon Source C3O C3C ClOO ClOC CIlO CIlC C20O C20C

D-galactose + - + + + + + -

D-psicose + + + + - - - -

Cis-aconitic acid - - - - - - + +

Citric acid - - - - - - + +

D-gluconic acid + + + + - - + +

Bromosuccinic acid + + + + - + + -

L-alaninamide + + + + + + + -

L-leucine + + + + + + - +

L-serine - + - - - - - +

Thymidine + + + + + + - +

A-D-glucose-1 - phosphate

D-glucose-6-phosphate + + + + - - - -

[0086| PHYLOGENETIC ANALYSIS

|0087] Nearly full length 16S rDNA sequences of the eight isolates were used to construct the phylogenetic trees in Figures 13 and 14. Figure 13 presents the sequences in the context of several genera, whereas Figure 14 focuses on their position among more closely related species. The gene sequence analyses revealed that all of the PST-degrading bacteria are members of the y-Proteobacteria and. more specifically, a clade formed within the genus Psendoalteromonas. ClOO clustered with a subclade distinct from the rest of the PST- degrading cultures which included P. distincta, P. carrageenovora, and P. espejiana. The rest of the isolates were located in a subclade with P. haloplanktis and P. translucida. Isolates C3O and Cl OO clustered robustly with Pseudoalteromonas sp. AY383041 and SUR560. respectively, neither of which have been given a species designation. According to the distance matrix calculated, none of 16S rDNA sequences of the PST degrading bacteria diverged more than 1 % from each other. Additionally, P. haloplanktis, P. distincta, P. carrageenovora, P. espejiana, and P. translucida all diverged less than 1% from the isolates in the current study.

[0088] PLASMID PROFILES AND RISA [0089] Bacterial plasmids are extrachromosomal DNA molecules capable of autonomous

replication. They are often circular and double-stranded and may confer a selective advantage to the harboring bacterium, such as antibiotic resistance or the ability to degrade specific compounds. In addition to confirming the presence or absence of plasmids, plasmid profiling can be helpful in differentiating among bacterial strains. An image of the agarose gel containing the plasmid profiles of the eight isolates is shown in Figure 15. The profiles for the corresponding "O" and 'C ^" cultures for C3, C 10. C 1 1. and C20 appear to be almost identical. The "C" cultures, however, produced stronger bands of genomic DNA in and near in the sample wells than the 'O ^" cultures. The plasmid profiles for the 'O' and 'C cultures of C3 and ClO are very similar. They all possess a plasmid band just under 2000 base pairs (bp) and another just above 1000 bp. Additionally, isolates C3C and ClOC have a faint plasmid band at around 1000 bp. which does not seem to be present in C3O and ClOO. Isolates CI lO and Cl 1C do not appear to have any plasmid bands, where C20O and C20C have three and two. respectively, that do not match the banding patterns seen in isolates C3O, C3C, Cl OO. and C 1 OC. C20O has one plasmid band above 3000 bp. another at 2000 bp, and a third at 1000 bp. C20C shares the latter two plasmid bands at 2000 bp and 1000 bp. but lacks the former at 3000 bp.

[0090J Ribosomal intergenic spacer analysis (RISA) is a method of microbial community analysis. It involves polymerase chain reaction (PCR) amplification of a region of the rRNA gene operon between the small (16S) and large (23S) subunits called the intergenic spacer region. The taxonomic value of the intergenic spacer region lies in the significant heterogeneity in both length and nucleotide sequence. The resulting PCR product will be a mixture of fragments contributed by several dominant community members resulting is a complex banding pattern on a polyacrylamide gel electrophoresis (PAGE) gel that provides a community-specific profile, with each DNA band corresponding to a bacterial population in the original group. In these experiments. RISA was used to differentiate among isolates rather than entire microbial communities. Figure 16 shows an image of the PAGE gel illustrating the RISA profiles of the eight isolates. All of the profiles are identical with the exception of C3O. which contains an additional band below the second one common to all isolates.

[0091 ] Long chain fatty acids (9 to 20 carbons) found in bacterial membranes can be used to characterize and identify bacteria. The fatty acid profile is generally unique to a species. Fatl) acid profiles for the eight PST-degrading isolates are shown in Table 6.9. The major fatty acids and their percent composition ranges among the isolates were 15:0 (1.77-4.97 %). 16:0 (27.47-31.71 %), 16:l(n-9) (5.92-23.56 %). 16: l (n-7) (24.99-42.39 %), 17: l(n-8) (2.24-

4.44 %), and 18: l(n-7) (3.00-4.42 %). A principal component analysis showing the relationships among the eight isolates with respect to their fatty acid profiles is given in Figure 17. The analysis indicates that, with respect to fatty acid profiles, the isolates fall into four different strains: 1 ) C3O and C3C: 2) Cl 10. Cl 1C, and C2OO; 3) C20C and ClOO; and 4) ClOC. However, according to this analysis, none of the isolates appear to cluster too closely with any other.

|0092| Table 6.9 Cellular fatty acids of the eight PST-degrading isolates. Isolates were cultured at 25°C in MB to late log stage of growth (approximately 48 hours). Major fatty acids are indicated in bold type

14:0 2.23 1.91 2 .31 2.29 ~> .12 3.35 2.72 1 .79 2.0 (0.4-8 •4)

14: 1 (/7- 1.41 1.52 1 .39 1.40 1 .77 1.81 1.37 1 .18

7)

15:0 3.06 1.77 2 .79 2.46 4 .97 4.66 2.21 1 .91 3.3 (0.2-8 .5) iso- 0.47 0.32 0.47 0.35 0.66 0.69 0.64 0.34

16:0

16:0 29.86 31.03 30.52 30.05 27.96 27.47 30.46 31.71 23.7 (14.0-33.8)

16: 1 (/7- 9.62 14.87 5.92 12.92 8.60 17.10 8.27 23.56

9)

16: 1 (n- 38.75 33.20 43.15 36.83 38.37 30.97 42.39 24.99 39.8 (29.6-47.8)

7)

ISO- 0.33 0.19 0.32 0.26 0.30 0.35 0.35 0.26 0.3 (0.0-0. 9)

17:0

17:0 1.33 1.48 1.13 1.04 1.61 1.60 0.73 1.33 3.6 (0.5-8. 7)

16:3(π- 0.34 0.65 0.13 0.47 0.42 1.20 0.19 1.43

4)

17: l (w- 4.03 2.73 3.46 3.00 4.44 3.78 2.34 2.24 7.6 (0.2-19 .0)

8)

18:0 1.56 2.19 1.57 1.59 1.16 1.05 1.37 1.71

1 8 : 1 (>7- 0.22 0.21 0.24 0.25 0.22 0.21 0.21 0.16

13)

1 8 : 1 (77- 0.60 0.50 0.55 0.43 0.47 0.32 0.47 0.45

9)

1 8 : 1 (77- 4.16 4.42 4.02 4.40 3.32 3.00 4.19 4.24 6.1 (1.3-18 .5)

7)

1 8:2(/7- 0.18 0.14 0.21 0.16 0.17 0.14 0.22 0.16

6)

Total 98.97 98.56 99.02 98.73 98.56 98.44 99.03 98.75 87.6 aMean value b Range (min and max values)

*Source: Ivanova et al. (2000)

(0093] PAGE patterns of whole-cell proteins can be used to characterize and identify bacterial strains. Strains of a given species will often have similar patterns, especially of the major proteins. However, some strains may yield patterns that are not easily distinguishable from those of other species. Figure 18 shows the SDS-PAGE whole cell protein profiles of the PST-degrading isolates. The protein profiles of the eight isolates are virtually identical. The only discernable difference seems to be that the bands of the "C cultures appear slightly more faint than those of the "O ^" cultures.

(0094) Lipopolysaccharide (LPS) is a major surface component of Gram-negative bacteria which contributes to the organism ^' s structural integrity and varies immensely among different strains and species. The EPS is made up of three main parts: polysaccharide (O) side chains (O-antigens). a core oligosaccaride. and Lipid A. Some strains of bacteria produce both a smooth (S) and rough (R) form of LPS which correspond to smooth and rough colony phenotypes. The main difference between these two forms of LPS is that the S form contains O antigens, while the R form does not.

[00951 The SDS-PAGE LPS profiles of the eight isolates are shown in Figure 19. The LPSs of the isolates are very similar, with the exception of the C20C sample which may have experienced an error during the loading and/or running of the gel. All isolates possess polysaccharide cores of approximately the same size at the 14.4 kDa protein markers. All of the isolates appear to have the same O-antigen banding pattern between the 21.5-97.4 kDa protein markers, with the exception of the three bands between 66.2-97.4 kDa which do not appear to be present in the CI lC and C20O samples. Given the presence of O-antigens and the colony phenotypes, it safe to say the isolates all possess the S form of LPS.

METHODS

|0096] SCREENING BACTERIAL ISOLATES FOR PST DEGRADATION

|0097] Toxic algal extract (AE)

|0098] Tv\o (2) grams (wet weight) of toxic algal biomass AJexandrium tamarense (strain

PrI 8b) were weighed into a 20 ml tube and diluted 1 :1 with distilled deionized water (DDW).

The tube was placed in a boiling hot water bath at 100 ⁰ C for 10 min. After heating and cooling, 12 niL of 0.1 M HCl was added the tube, which was sonicated (Vibra Cell High

Intensity Ultra Sonic Processor, model VC375 Watts. Sonics, Newtown. CT) on ice for 10 min. The extract was frozen at -2O ⁰ C until use. Prior to use, the AE was thawed and filtered

twice with 0.45 μm syringe filters (Sarstedt Ag. & Co., Nϋmbrecht. Germany) and then filter- sterilized with a 0.20 μm syringe filter (Sarstedt Ag. & Co.). The composition of the AE is given in Table 5.1.

|0099] Table 5.1 PST composition of AE from Alexandrium tamarense (strain Prl8b).

PST Concentration Specific Toxicity Concentration

(μM) (MU/μmole) (μM STX eq) dcGTX 3 3.27 935 1.23 dcGTX 2 10.36 382 1.59 GTX 3 22.33 1584 14.24 GTX 2 57.54 892 20.67 Cl 103.77 15 0.63 C2 102.43 239 9.86 NEO 77.78 2295 71 .89 STX 73.25 2483 73.25

Total μM STX eq: 193.36

[0100| Sterile mussel extract (ME)

[0101] Fresh blue mussels (Mytilus edulis) (2.3 kg) were purchased from Clearwater (Halifax, NS). The mussels were divided among four 1 L beakers. The beakers were placed in a microwave oven (Kenmore Micro/convection, Sears, Hoffman Estates, IL) and heated for 6 min at a power setting of 6. The mussels were pulled apart by hand and the tissue was removed from the shell. Mussels that did not open with the initial heat treatment were reheated at the above microwave oven settings for an additional 3-4 min. The collected tissues and their liquids were divided into five equal parts in 250 ml centrifuge bottles (plus one bottle of equal weight containing only water) in preparation for centrifugation in the IEC B-20A Centrifuge (Thermo Fisher Scientific, Waltham, MA) using the 6 x 250 ml rotor. The tissues were centrifuged for 30 min at 8.000 ^χ g at 5°C. At the end of the run. the liquid was separated from the tissue, resulting in the recovery of about 450 ml. The liquid was further clarified by centrifugation for 1 hour at 100,000 x g (Beckman model L2 centrifuge, Beckman Coulter, Inc.. Fullerton. CA) using the SW-27 swing bucket 6 ^χ 28 ml rotor at 5°C. The liquid was pooled and filtered using a 1 L glass Millipore filtration flask (Millipore, Billerica. MA) with a 47 mm x 0.45 μm nylon filter (Whatman, Middlesex. UK). The tap water aspirator was used as the vacuum source. Many changes of the filter were required to clarify all of the mussel extract (ME). The filtration flask was placed in an ice bath, maintaining the temperature of the extract at around 5 ⁰ C. The filtration step was repeated a second time. The third filtration step required the same filtration flask, only this time a 0.47 mm x 0.2 μm filter was used (Pall, East Hills. NY). Several changes of the filter were

required. The final filtration step involved sterilization filtration using a sterile 150 ml bottle with an attached 0.2 μm filter funnel unit (Nalgene, Rochester, NY). The vacuum was supplied by the portable vacuum pump (Brinkman vacuum aspirator model B- 169, Mississauga, ON) and the filtration was carried out in a laminar flow hood. Three bottles of ME were collected, totalling 400-450 ml. This material was stored at 4°C until pickup, after which it was frozen at -20 ⁰ C. Prior to use, the ME was centrifuged at 4.000 rpm for 10 min (Universal 32 R centrifuge. Hettich Zentrifugen. Tuttlingen. Germany). |0102| Bacterial isolates and standard culture conditions

[0103] Bacteria were isolated from toxic blue mussels (Mytilus edulis) and soft-shell clams (Mya arenaria) harvested from throughout Atlantic Canada by the Canadian Food Inspection Agency (CFIA; Dartmouth, NS) in the summer of 2003. The digestive glands of the mussels were dissected, pooled, and weighed. Then, a 1/10 dilution of the digestive glands was made in sterile peptone water (Oxoid Ltd.. Basingstoke. Hampshire, England). This dilution was stomached (Lab-Blender 400 model BA6021 , Bury St. Edmunds, England) and spread plated on sterile marine agar (MA) 2216 (Difco Laboratories. Detroit, MI) plates and incubated at 25°C until isolated colonies appeared (about 48 h). Morphologically different colonies were picked and streaked onto MA plates to obtain isolated colonies. Individual colonies were picked from the streak plates and restreaked to ensure culture purity. Glycerol stocks were made for each of the pure cultures as follows. A 5 ml tube of marine broth (MB) 2216 (Difco Laboratories) was inoculated with a single colony from a MA plate of pure culture and grown overnight at 25°C. After incubation, 800 μl of the culture and 200 μl of sterile glycerol were added to a sterile 2.0 ml cryogenic tube (Nalgene) and frozen at -80 ⁰ C. The isolates were routinely cultivated on MA at 25°C for 48 hours and in MB at 25 ⁰ C for 24 h. unless otherwise indicated. All media used in the experiments were autoclaved at 121 ⁰ C for 15 min. [0104] Instrumentation for LC coupled with post-column oxidation and fluorescence detection (LC-pcr-FLD) analyses

[0105| The LC analyses were conducted with an Agilent 1 100 LC equipped with quaternary solvent delivery system and a variable injector (0 to 25 μl) (Agilent. Palo Alto, CA). The post-column reaction system consisted of a Waters Post-Column Reaction Module (Waters, Milford, MA) operated at 80 ⁰ C. A knitted Teflon tubing reaction coil was used with 1000 μl volume (Waters PN 030805). The system used two Agilent Gl 31 OA isocratic pumps, which could be controlled by the Agilent ChemStation. The fluorescence detector was a Shimadzu RF551 detector fitted with either a large (12 μl) or micro (2 μl) flow cell (Shimadzu, Kyoto, Japan). PSTs were detected by excitation and emission wavelengths set to 330 and 390 nm.

respectively.

(0106| For the combined analysis of the GTX and STX groups, the LC method involved a mobile phase with a step gradient between three solvents using a Zorbax Bonus-RP column from Agilent (5 μm, 4.6 x 250 mm). Solvent A was deionized water containing 20 mM HSA and 10 mM ammonium phosphate, adjusted to pH 7.1 with ammonium hydroxide. Solvent B was acetonitrile/water (30:70 v/v) containing 30 mM ammonium phosphate adjusted to pH 7.1 with ammonium hydroxide. Solvent C was deionized water. The mobile phase used for chromatography was 4 0% A and 60 % C for the first 17 min and then stepped up to 45 % A, 35 % B and 20 % C at 17.1 min for the rest of the run. The flow rate of the mobile phase was set at 0.8 mL/min. Post-column reagent #1 ("oxidant") was either tert-butyl hydrogen peroxide (tBHPO) or sodium periodate (both adjusted to pH 7.8 using 5M NaOH) and post- column reagent #2 ("acid") was 0.75 M HNO3. Both reagents were pumped at 0.4 mL/min. [0107] For the C toxins, analyses were performed using an isocratic run with 0.8 mL/min of 2 mM tetrabutylammonium dihydrogenphosphate (TBAP) in deionized water, adjusted to pH 5.8 with ammonium hydroxide. The column used was a Beta Basic C8 (5 μm, 4.6 x 250 mm; Thermo Keystone, Waltham, MA). The post-column reaction system was the same as above, but only periodate was used as the oxidant.

[0108] For the analysis of the GTX and STX groups on the narrow bore column (5 μm, 2.1 x 150 mm: Agilent Zorbax Bonus-RP) system, the same three solvents were used. The mobile phase was 55 % A and 45 % C until 22 min, when it was stepped up to 55 % A, 35 % B and 10 %C at 22.1 min for the rest of the run. The flow rate of the mobile phase was set at 0.2 mL/min. The post-column reaction system was the same as above, but only periodate was used as the oxidant. The flow for oxidant and acid pumps were set at 0.1 mL/min. [0109] For C toxins, analyses were performed using an isocratic run with 0.2 mL/min of 2 mM TBAP in deionized water, adjusted to pH 5.8 with ammonium hydroxide. The column used was a Beta Basic C8 from Thermo Keystone (5 μm, 2.1 x 250 mm). The post-column reaction system was the same as for the GTX + STX group. [0110] Modifed LC-pcr-FLD method for analysis of bacterial extracts [011 1 ] The LC method used to screen for bacteria with PST degradation ability was slightly different from the method being developed. Due to a fewer number of PST analogues presented in the sample and the large number of samples that had to be processed, a faster LC method was used. A BDS-Hypersil-C8 (5 μm, 10 x 4 mm: Thermo Electron, Waltham. MA) guard column was put in place before the actual column. [0112] For the combined analysis of the GTX and STX groups, the LC method involved a

mobile phase with a step gradient between three solvents using a Zorbax Bonus-RP column

(5 μm. 4.6 x 250 mm) from Agilent. Solvent A was deionized water containing 20 mM HSA and 10 mM ammonium phosphate, adjusted to pH 7.1 with ammonium hydroxide. Solvent B was acetonitrile/water (30:70 v/v) containing 30 mM ammonium phosphate adjusted to pH

7.1 with ammonium hydroxide. Solvent C was deionized water. The mobile phase used for chromatography was 40 % A and 60 % C for the first 17 min and then stepped up to 45 % A,

35 % B and 20 % C at 17.1 min for the rest of the run. The flow rate of the mobile phase was set at 0.8 mL/min. Post-column reagent #1 ("oxidant ^" ) was sodium periodate (both adjusted to pH 7.8 using 5M NaOH) and post-column reagent #2 ('acid') was 0.75 M HNO3. Both reagents were pumped at 0.4 mL/min.

[0113] For the C toxins, analyses were performed using an isocratic run with 0.8 mL/min of

2 mM TBAP in deionized water, adjusted to pH 5.8 with ammonium hydroxide. The column used was a Beta Basic C8 (5 μm, 4.6 x 250 mm; Thermo Keystone. Waltham. MA). The post-column reaction system was the same as described above.

[0114] Mouse colony

[01 15| MBAs were carried out with 19-22 g white mice from the stock colony used for rouline MBAs at CFIA (Dartmouth. NS) as specified by the AOAC Official Method 959.08

(2000).

[0116] CHARACTERIZATION OF PST-DEGRADING BACTERIA

[0117| Bacterial isolates and standard culture conditions

10118] The marine bacteria were routinely cultivated on MA at 25°C for 48 hours and in MB at 25 ⁰ C for 24 hours. Control bacteria were routinely cultivated on tryptic soy agar (TSA)

(Oxoid Ltd.) or tryptic soy broth (TSB) (Oxoid Ltd.) at 37°C for 24 hours. Unless otherwise indicated, cultures for phenotypic tests were incubated at 25°C and 37°C for marine and control bacteria, respectively, for 48 hours and performed in duplicate. Bacteria were cultivated under aerobic conditions unless otherwise indicated. All media used in the experiments were autoclaved at 121 ⁰ C for 15 min.

[0119| Presumptive screening of individual bacterial isolates

[0120] Bacterial cultures isolated from toxic marine bivalves were streaked from glycerol stocks onto MA plates and incubated. Sterile 50 ml centrifuge tubes containing 10 ml of MB were inoculated with single colonies from plates and grown overnight at 25°C. After incubation, the A ₆ so of each culture was measured (Ultrospec 1 100 pro UV/Visible spectrophotometer, Amersham Biosciences. Buckinghamshire, England) and adjusted to approximately 0.1 OD/ml by diluting the culture with MB or centrifuging the culture at 2,000

rprn for 15 min (Universal 32 R centrifuge) and resuspending the pellet in less MB until the target A ₆₅₀ was obtained. The cultures were centrifuged at 2,000 rpm for 15 min and the supernatant was carefully poured off and discarded. The cell pellets were resuspended in a medium consisting of MB. filter-sterilized ME, and filtered-sterilized AE, as given in Table 5.2, to a final volume of 1 ml per sample in sterile 1.5 ml microtubes for day 0 samples and sterile 5 ml plastic culture tubes for day 5 samples. This was done for each isolate a total of four times - in duplicate for both day 0 and day 5 tubes. In addition to these samples, controls were prepared: the toxic (T) control contained AE, MB, and ME. but no bacteria; and the non-toxic (NT) control contained MB, and ME. but no AE. The controls were also produced in duplicate for days 0 and 5. Immediately after the day 0 sampling for LC analysis, day 0 samples and controls were frozen at -80 ⁰ C and day 5 tubes were incubated at 25°C for 5 da\ s in a shaking incubator at 130 rpm. All toxin work was conducted in the laminar flow hood to minimize contamination.

[0121] Table 5.2 Control and sample tubes for days 0 and 5 of the presumptive screenings.

Tubt-s Bacteria AE (μl) MJT(μϊ) MϊT(μϊ)

T control - 100 300 600

NT control - - ^c 300 600 aPrepa7e^ffor ^~ each of the isolates. ^b P, tubes contained bacterial pellet, ^C NT control contained l OOμl of sterile DDW in the place of AE

(0122] Bacterial enumeration

[0123] During the incubation period, bacterial growth was monitored using the standard spot-plate technique. Spot-plating was done at 0, 24. 48 hours, and 5 days in order to monitor cell numbers. The aerobic plate count (APC) was conducted according to the conventional plate count method described by the US FDA/CFSAN"s Bacteriological Analytical Manual

(BAM).

[0124J Sample preparation for LC-pcr-FLD analysis

[0125| After spot-plating on day 5, samples were transferred to sterile 1.5 ml microtubes and frozen at -80 ⁰ C overnight or longer. When ready for analysis, all samples (both day 0 and 5) from -80 ⁰ C were thawed and tubes covered in Parafilm M (SPI Supplies, Toronto, ON).

Each tube was sonicated with a 3 mm probe (Branson, Danbury, CT) inserted through the

Parafilm just below the surface of the liquid. Each sample (1 ml) was sonicated at 20 % power for three rounds of 10 sec with 1 min on ice between each round (Branson Sonifier model S-150D. Branson). The probe was wiped with DDW and then 95 % ethanol between

runs. Complete cell lysis was verified by phase contrast microscopy (Nikon Optiphot Biological Microscope, Nikon, Japan). Sonication was done in the laminar flow hood with ear and eye protection. After sonication, samples were centrifuged for 2 min at 13,000 rpm (Mikro 20 centrifuge, Hettich Zentrifugen) and then filtered with 0.20 μm filters into new sterile 1.5 ml microtubes and stored at -8O ⁰ C until LC analysis. [0126] LC-pcr-FLD analysis

[0127| In preparation for LC-pcr-FLD analysis, the 0.5 ml bacterial extract aliquots were thawed at room temperature. Two hundred (200) μl of extract was pipetted into the 10,000 NMWL Ultrafree-MC Centrifugal Filter Unit (Millipore). The sample unit was centrifuged at 16,000 ^χ g for 25 min (Eppendorf Centrifuge 5415. Eppendorf. Hamburg. Germany). The filtrate was pipetted into LC sample vials with inserts. These samples were stored at 4°C until analysis. The methods used in the LC analyses were followed as described above. |0128| Confirmatory screening of individual bacterial isolates for PST degradation [0129] Only those isolates showing partial or complete disappearance of toxin peaks in the LC PST profile during the presumptive screening advanced to the confirmatory screening. Nonetheless, glycerol stocks of those cultures not demonstrating PST utilization from the first screening were maintained for prospective future studies. The methods used in the confirmatory screenings were the same as those employed in the presumptive screening except the sample volumes were increased (shown in Table 5.3) to provide enough extract for both LC and MBA analyses. In addition, spot-plating and sampling for LC analysis were conducted every 24 hours throughout the 5 days incubation.

[0130| Table 5.3 Control and sample tubes for LC and MBAs.

Tubes Bacteria AE (μl) ME (μl) MB (μl)

T control 0.7 2.1 4.2 NT control C 2.1 4.2 Samples ^a p b 0.7 2.1 4.2

''Prepared for each of the PST-utilizing isolates. P. tubes contained bacterial pellet CNT control contained 0.7 ml of sterile DDW in the place of AE

[0131 J Sonication and preparation of the samples was performed as described above, except three rounds of 1 min at 80 % power were used. Complete cell lysis of samples was verified by phase contrast microscopy, as done previously. After sonication, samples were centrifuged for 10 min at 4000 rpm (Mikro 20 centrifuge) and filter- sterilized with 0.20 μm filters into new sterile 15 ml centrifuge tubes and stored at -2O ⁰ C until ready to perform LC

analyses or to prepare the acid extractions for the MBAs. [0132] Acid extraction

(01331 The acid extraction methods were modified from the AOAC Official Method 959.08 (2000). The acid extraction was performed on sonicated extracts showing possible PST degradation from the presumptive screenings. Samples were removed from the freezer and thawed. Two (2) ml of each sample was removed and pipetted into new sterile 50 ml centrifuge tubes. Two (2) ml of 0.1 N HCl was pipetted into each tube with the 2 ml of sample ( 1 : 1 ratio of sample:acid) and the pH was adjusted to approximately 3 by adding 0.1 N NaOH dropwise with constant stirring to prevent local alkalinization. Samples were boiled in a hot water bath at 100 ⁰ C for 5 min. After cooling, enough of the acid extraction for each sample was pipetted into new tubes containing enough DDW to achieve appropriate dilution factors for the samples and controls. The pH of the samples was adjusted to 2-4 by adding 0.1 N HCl or 0.1 N NaOH dropwise with constant stirring. After dilution and pH adjustment, the diluted extracts were filter-sterilized with 0.20 μm syringe filters and frozen at -20 ⁰ C until MBAs were performed.

[0134] Appropriate dilution factors were determined by estimating the PST concentration (μM STX eq) in the AE and then determining the initial PST concentration in the T control (and therefore the samples). After the initial calculation of the T control toxicity, "test" dilutions of the T control and one isolate were each injected into three individual mice. This was done to preserve mice as well as save time. Test dilutions were made until median death times of 5-7 min for the samples and the T control were achieved. When this was accomplished, dilutions were made for each isolate and then adjusted according to individual MBA results.

|0135| Mouse bioassays (MBAs)

I0136J The MBAs were performed at CFIA (Dartmouth, NS) as described by the AOAC Official Method 959.08 (2000) as modified from Mcfarren (1959). [0137) Graphical and statistical analyses

[0138] Graphs and regression analyses were produced in SigmaPlot 8.0 (Systat Software Inc.. San .lose. CA). Only LC data were included in these analyses. Non-linear regressions were performed on the total and individual toxin concentration data for each isolate over the 5 days incubation period. The significance of the coefficients of determination (Pearson r ) for each regression analysis was determined (Rohlf and Sokal 1981). In addition, those toxicity data exhibiting first order kinetics (excluding baseline values) were plotted on a semi-logarithmic scale for analyses of initial degradation rates. A linear regression was

performed on each plot and decimal reduction times (D values; days) were determined by taking the negative reciprocal of the slopes (μM STX eq d ^"1 ). Again, the significance of the coefficients of determination for each linear regression was assessed.

[0139] MORPHOLOGY AND CULTURAL CHARACTERIZATION ¹ |0140| Gram reaction

[0141 ] This test was performed following the standard methods conducted by those skilled in the art. The bacterial controls for this test were: Gram-positive. Bacillus cereus; Gram- negative. Escherichia coli. [0142] Transmission election microscopy (TEM)

[0143] The marine bacteria were negatively stained following the standard methods conducted by those skilled in the art. The prepared samples were examined in a JEOL JEM- 1230 electron microscope (JEOL, Toyko. Japan) using an accelerating voltage of 80 kV. The images were captured digitally by a Hamamatsu digital camera ORCA-HR (model C4742- 51 -12HR, Hamamatsu Photonics KK, Hamamatsu City. Japan). [0144] Colony morphology

[0145] Marine bacteria were streaked onto MA from glycerol stocks. Pigmentation and colony morphology were observed every 24 h for several days under standard culture conditions.

[0146] Oxygen requirements

[0147] Oxygen relationships were investigated following the standard methods conducted by those skilled in the art using MB prepared with the addition of 0.5 % agar. After autoclaving, five (5) ml aliquots were dispensed into sterile glass test tubes and allowed to set. After cooling, the tubes were stabbed (to about ³ λ depth) using an inoculating needle with single colonies of marine bacteria from MA plates and incubated under standard culture conditions. After incubation, the location of growth in the tubes was examined. To test for growth in anaerobic conditions, an anaerobic hood (Coy Hoods, Ltd) was used. Twenty-four (24) h prior to inoculation, sterile MA plates were placed in the anaerobic hood containing an atmosphere of H2/CO2 (20:80 % v/v) to facilitate oxygen diffusion from the plates. The plates were then inoculated and incubated in the hood at room temperature for 72 hours after which they were observed for growth. [0148] Motility

Examples of standard protocols of microbiological tests which are commonly performed and known by those skilled in the art can be found in the Protocol References on page 82.

|0149] This test was performed as modifying the standard methods conducted by those skil led in the art using soft agar plates made with MB and TSB prepared with 0.5 % agar and 0.01 % 2.3.5-triphenyltetrazolium chloride (TTC; Allied Chemical Corporation. New York, NY). Standard culture conditions were used. The bacterial controls for this test were: positive. Proteus vulgaris; negative, Staphylococcus aureus. An uninoculated plate was included as an additional control. In addition, cultures were examined by light microscopy (Nikon Optiphot Biological Microscope, Nikon. Tokyo, Japan) for swimming motility at a magnification of 100 ^χ . [015Oj Optimum growth temperature

[01511 O ^ne 0 ) ^m ' aliquots of MB in sterile 1.5 ml microtubes were inoculated with single colonies from MA plates and vortexed to make homogenous suspensions. Sterile 50 ml centrifuge tubes containing 10 ml of MB were inoculated with 20 μl of the 1 ml bacterial suspensions. The tubes were incubated at 10. 15, 20. 25, 30, and 35 ⁰ C in a shaking water bath at 84 rpm for 24 h. Growth was measured with A^o measurements after 24 h. To test growth at 4°C. the inoculated tubes were incubated at 4°C with no shaking. The cultures were tested for growth at 4°C on a daily basis. [0152] pH optimum

|0153] Prior to autoclaving, the pH of MB was adjusted to 4.5. 5, 5.5. 6. 6.5. 7, 7.5, 8. 8.5. 9. 9.5., and 10 by dropwise addition of either 1 M HCl or 1 M NaOH. Bacterial suspensions were prepared as described in Section 5.2.2.1.6 using the pH-adjusted media. Sterile 50 ml centrifuge tubes containing 10 ml of the pH adjusted MB were inoculated with 20 μl of the bacterial suspensions and incubated at 25°C in a shaking incubator at 130 rpm. Growth was measured with A _6? o measurements after 24 h. [0154] Sodium requirement and optimum

[0155] Artificial seawater (ASW) was prepared as described by Smibert and Krieg (1994) but without FeSO ₄ . It contained: MgCl ₂ . 5 g/L; MgSO ₄ , 2 g/L; CaCl ₂ , 0.5 g/L: KCl 1 g/L; and tryptone (Difco Laboratories), 5 g/L. Prior to autoclaving, these ingredients were dissolved in DDW and enough NaCl was added to give NaCl concentrations of 0, 0.5. 1. 2, 3. 4, 5, 6. 7. 8. 9, 10. 1 1. 12, 13, 14 and 15 %. Bacterial suspensions were prepared as described in Section 5.2.2.1 .6 using the NaCl-adjusted ASW. Sterile 50 ml centrifuge tubes containing 10 ml of NaCl-adjusted ASW were inoculated with 20 μl of the bacterial suspensions and incubated at 25°C in a shaking incubator at 130 rpm. Growth was measured with A ₆ So measurements after 24 h. [0156] PHENOTYPIC CHARACTERIZATION

10157] Oxidase and catalase activity

[0158] The test for catalase activity was performed following the Standard methods conducted by those skilled in the art using 3 % hydrogen peroxide. The bacterial controls for this test were: positive. Pseudomonas aeruginosa; negative. Bifidobacterium bifidum. The test for oxidase activity was performed using oxidase sticks (Oxoid Ltd.) according to the manufacturer's directions. The bacterial controls for this test were: positive, Pseudomonas aeruginosa: negative. Escherichia coli.

[0159| Biolog-GN microplates (carbon substrate utilization)

[0160| The substrate utilization profiles were generated using 96 well Biolog-GN

MicroPlates (Biolog Inc., Hayword. CA) as described by Smith and others (2001 ). The microplates were incubated under standard culture conditions after which they were read visually.

[0161 ] Nitrate reduction and denitrification

[0162] This test was performed following the standard methods conducted by those skilled in the art using MB and TSB prepared with the addition of 0.1 % KNO3 and 0.17 % agar. MB and TSB containing 0.17 % agar were also prepared without the 0.1 % KNO3. Standard culture conditions were used. The bacterial controls for this test were: denitrification.

Pseudomonas aeruginosa; nitrate to nitrite. Escherichia coli: negative. Bacillus cereus.

Uninoculated tubes of MB and TSB with and without the 0.1 % KNO3 were included as additional controls.

[0163] Indole production

[0164] This test was performed following the standard methods conducted by those skilled in the art using ASW prepared with the addition of 1 % tryptone. Standard culture conditions were used. The bacterial controls for this test were: positive, Escherichia coli; negative.

Enlerobacter aerogenes. An uninoculated tube was included as an additional control.

(0165] Glucose oxidation/fermentation catabolism

[0166] This test was performed following the standard methods conducted by those skilled in the art using the modified oxidation/fermentation medium of Leifson (1963) for marine bacteria. Standard culture conditions were used. The bacterial controls for this test were: fermentative (F) reaction with glucose. Staphylococcus epidermitis: oxidative (O) reaction with glucose. Micrococcus luteus; negative reaction with glucose, Pseudomonas areuginosa.

[0167] Arginine dihydrolase activity

[0168] This test was performed following the standard methods conducted by those skilled in the art using MB and TSB. Standard culture conditions were used. The bacterial controls for

this test were: positive. Enterobacter cloacae; negative. Proteus vulgaris.

|0169] Urease activity

[0170] This test was modified from the standard methods conducted by those skilled in the art. Christensen Urea Agar was prepared as described with the addition of 20.0 g of NaCl/L to meet the needs of marine bacteria. Standard culture conditions were used. The bacterial controls for this test were: positive. Proteus vulgaris; negative, Escherichia coli. An uni tioculated tube was included as an additional control.

[0171 J Esculin hydrolysis

[0172] This test was performed following the standard methods conducted by those skilled in the art using MB and TSB prepared with the addition of 0.01 % esculin (Sigma-Aldrich) and

0.05 % ferric citrate (Sigma-Aldrich). Standard culture conditions were used. The bacterial controls for this test were: positive. Enter ococcus faecalis; negative. Streptococcus epidermidis. An uninoculated tube was included as an additional control.

[0173] Gelatin hydrolysis

[0174] This test was performed following the standard methods conducted by those skilled in the art using MA and TSA prepared with the addition of 0.4 % Bacto Gelatin (Difico

Laboratories). Standard cultures conditions were used. The bacterial controls for this test were: positive, Pseudomonas aeruginosa; negative, Escherichia coli. An uninoculated plate was included as an additional control.

[0175] Agar hydrolysis

[0176] This test was conducted following the standard methods conducted by those skilled in the art using MA. The plates were incubated for several days under otherwise standard culture conditions. An uninoculated plate was included as a control.

[0177] κ-Carrageenan hydrolysis

[0178] MB was prepared according to the manufacturer's directions, with the addition of 1.5

% κ-carrageenan (CP Kelco US Inc., Wilmington. DE). After autoclaving and cooling, the medium was poured into petri dishes. The plates were streaked with single colonies of marine bacteria from agar plates and incubated for several days under otherwise standard culture conditions. The plates were periodically examined for pitting or corrosion around areas of growth. An uninoculated plate was included as a control.

[0179] Alginate hydrolysis

[0180] This test was conducted following the standard methods conducted by those skilled in the art. Standard culture conditions were used. The bacterial controls for this test were: positive. Vibrio sp.; negative. Escherichia coli. An uninoculated tube was included as an

additional control.

[0181 ] Starch hydrolysis

(0182] This test was performed following the standard methods conducted by those skilled in the art using MA and TSA prepared with the addition of 0.2 % soluble starch (BDA

Chemicals. Toronto, ON). Standard culture conditions were used. The bacterial controls for this test were: positive, Aeromonas hydrophila; negative, Escherichia coli. An uninoculated plate was included as an additional control.

[0183] Casein hydrolysis

|0184] This test was performed following the standard methods conducted by those skilled in the art using ASW with 3 % agar and nutrient agar, both supplemented with Bacto Skim Milk

(dehydrated; Difco Laboratories). Plates were incubated up to 14 days under otherwise standard culture conditions. The bacterial controls for this test were: positive. Serratia marcesens; negative. Escherichia coli. An uninoculated plate was included as an additional control.

[0185] Cellulase activity

[0186] This test was performed following the standard methods conducted by those skilled in the art using ASW. Tubes were incubated up to 14 d under otherwise standard culture conditions. An uninoculated tube was included as a control.

[0187] Lipase activity

[0188] This test was performed following the standard methods conducted by those skilled in the art using MA and TSA prepared with the addition of 1 % Tween 80 (Sigma-Aldrich) and

0.01 % CaClτ • H2O. Standard culture conditions were used. The bacterial controls for this test were: positive, Psendomonas aeruginosa; negative, Escherichia coli. An uninoculated plate was included as an additional control.

[0189] Phosphatase activity

[0190] This test was performed following the standard methods conducted by those skilled in the art using MA and TSA prepared with the addition of 0.01 % phenolphthalein bisphosphate tetrasodium salt (Sigma-Aldrich). Plates were incubated for 2 to 5 d under otherwise standard culture conditions. The bacterial controls for this test were: positive.

Staphlococcus aureus: negative. Micrococcus luteus. An uninoculated plate was included as an additional control.

[0191 ] Hydrogen sulfide production from cysteine

[0192] This test was performed following the standard methods conducted by those skilled in the art using MB and TSB prepared with the addition of 0.05 % L-cysteine (Sigma-Aldrich).

Standard culture conditions were used. The bacterial controls for this test were: positive, Aeromonas hydrophila; negative. Escherichia coli. An uninoculated tube was included as an additional control.

[0193] Deoxyribonucleic acid (DNA) hydrolysis

[0194] This test was performed following the standard methods conducted by those skilled in the art using MA and TSA prepared with the addition of 0.2 % DNA sodium salt (Type XlV from herring testes; Sigma-Aldrich). Standard culture conditions were used. The bacterial conirols for this test were: positive, Pseudomonas aeruginosa; negative, Escherichia coli. An uninoculated plate was included as an additional control. |0195] GENETIC ANALYSES [0196] DNA amplification, cloning, and sequencing

|0197| The marine bacteria were lysed as follows. Single colonies of each marine bacterium were selected from MA plates and added to a sterile 1.5 ml microtubes containing 100 μl of a lysing solution of the following composition: Triton X-100 (Sigma-Aldrich), 1 ml/L; and lysozyme (Sigma-Aldrich), 344 μl/L (of a 1 mg/ml solution). The tubes were boiled for 10 min in a water bath at 100 ⁰ C and cooled to room temperature.

[0198] The 16S ribosomal DNA (rDNA) was amplified as described by Leser and others (2002) using primers S-D-Bact-0008-a-S-20 (5'-AGAGTTTGATCMTGGCTCAG-S '. see SEQ. ID No.9) and S-*-Univ-1492-a-A-19 (5'-GGTTACCTTGTTACGACTT-S '. see SEQ.ID No.10). A primer solution of the following composition was prepared: S-D-Bact-0008-a-S- 20, 1 μl/reaction; S-*-Univ-1492-a-A-19. 1 μl/reaction; and sterile DDW. 18 μl/reaction. Twenty (20) μl of the primer solution and 5 μl of the supernatant from each isolate containing the template DNA were added to PuReTaq Ready-To-Go PCR Beads (Amersham Biosciences). Polymerase chain reaction (PCR) was performed in a Biometra T Gradient Thermoblock (Biometra, Gottingen, Germany). DNA amplicons were verified by electrophoresis by running 8 μl aliquots of the PCR products plus 2 μl of 6 ^χ loading buffer (bromophenol blue. 2.5 g/L; sucrose, 400 g/L) in 1.2 % agarose gels (UltraClean Agarose. MoBio Laboratories. Inc.. Carlesbad. CA) made with 1 * Tris-acetate-EDTA (TAE) (Sigma- Aldrich). The electrophoresis unit was the HE 99X Max Submarine Unit (Amersham Biosciences). The gel was run in 1 x TAE buffer at 120 mV for approximately 1 h. A DNA ladder (100-3000 bp; Bio-Rad) was run with the samples. A picture of the gel was captured using the BIS303PC Bio-Imaging system (DNR Ltd., Montreal Biotechnologies, Inc.. Montreal, QC). The PCR products were recovered by excising them from the gels under a portable UV light (Blak-Ray Long Wave UV Lamp model B 100 AP, UVP, Upland. CA)

using a sterile scalpel blade. The gel slices were placed in filter tips (VWR International, Mississauga. ON) inserted in microtubes and centrifuged at 13.000 rpm for 20 sec (Mikro 20 centrifuge).

[01991 PCR products were cloned using a TOPO TA Cloning Kit with the pCR 2.1 -TOPO vector (Invitrogen. Carlesbad, CA) according to the manufacturer's directions. Two different volumes from each reaction (25 μl and 50 μl) were plated on pre-warmed selective Brain heart infusion (BHl) agar (Oxoid Ltd.) plates (warmed at 37 ⁰ C for 30 min) containing 100 μl/nil ampicillan (Sigma-Aldrich). The plates were incubated overnight at 37°C. After incubation, single colonies from each selective BHI plate were chosen and grown in BHI broth (Oxoid Ltd.) overnight at 37°C. Several colonies were chosen from each plate to ensure at least one colony contained plasmids with the insert. One (1) ml of each culture was pipetted into a provided collection tube and centrifuged at 13.000 rpm for 1 min. The plasmids from the transformed bacteria were then extracted using the GenElute Plasmid Miniprep Kit (70 reactions; Sigma-Aldrich) according to the manufacturer ^' s directions. The eluent containing the plasmid was frozen at -20 ⁰ C until ready for sequencing. |0200] The transformations were verified by a restriction endonuclease digestion of each cloning reaction using EcoRI (New England Biolabs, Ipswich, MA). Each reaction was set up as follows in 1.5 ml microtubes: EcoRI. 1 μl; EcoRI buffer (New England Biolabs. Ipswich. MA). 1 μl; sterile DDW, 4 μl; and plasmid. 4 μl. The tubes were placed in a 37°C water bath for 1 hours to digest. After digestion. 2 μl of 6 ^χ loading buffer (see above) was added to each reaction and they were run under the same electrophoresis conditions given above using the HE 33 Mini Submarine Unit (Amersham Biosciences). A DNA ladder (see above) was run with the samples. A picture of the gel was captured using the BIS303PC Bio- Imaging system. After presence of the inserts was verified, the plasmids were sent to Health Canada (Ottawa, ON) to be sequenced. DNA sequencing reactions were carried out using the BigDye Terminator v3.1 cycle sequencing kit and run on an ABI 3130 genetic analyzer (Applied Biosystems, Foster City, CA) using a 36 cm capillary column containing POP7 polymer. Cloned 16S rRNA genes were sequenced from each end using the Ml 3 forward and reverse primers. The internal region of the gene was sequenced using the following internal universal primer: l όSInternal 5'-TCACRRCACCJAGCTGACG A-3'(see SEQTD No.13). Fragments were aligned using Sequencer v4.5 (Gene Codes Corp. Ann Arbor. MI). [0201 J Phylogenetic analysis

[0202] The 16S rDNA sequences of the eight isolates were added to the database of the ARB program (Ludwig and others 2004). They were automatically and then manually aligned with

nearest relatives in the database. An unrooted phylogenetic tree was constructed with the aligned sequences in the PHYLO WIN program (Galtier and others 1996). using the Nearest Neighbor Algorithm with a Jukes and Cantor one-parameter correction and pairwise gap removal. In order to statistically evaluate the tree, a bootstrap analysis of 1000 replications was conducted. A matrix distance was calculated (without correction) to compare the aligned sequences.

[02031 Plasmid profiles

[0204] The marine bacteria plasmid extractions were performed on 20 ml MB cultures according to the small-scale cold alkaline pH method described by Crosa and others (1994). A plasmid DNA clean-up was performed using a phenol/chloroform extraction and ethanol precipitation.

[0205| Electrophoresis was performed by running 20 μl of the plasmid DNA solutions plus 4 μl of 6 x loading buffer under the same electrophoresis conditions given above using the HE Max 99X Submarine Unit (Amersham Biosciences). A DNA ladder (100-3000 bp; Bio-Rad) was run with the samples. A picture of the gel was captured using the BIS303PC Bio- Imaging system.

[0206] Ribosomal intergenic spacer analysis (RISA)

[0207] The marine bacteria were lysed as described above. The ribosomal internal transcribed spacers plus a fragment stretch of the 16S rRNA for RISA were amplified as described by Jan-Roblero and others (2004) using primers B1055 (5'- AATGGCTGTCGTCAGCTCGT-3'. see SEQ.ID No.l 1 ) and 23SOR (5 ^f - TGCC AAGGC ATCC ACCGT-3', see SEQ.ID No.12). A primer solution of the following composition was prepared: B 1055. 1 μl/reaction: 23SOR, 1 μl/reaction; and sterile DDW, 18 μl/reaction. Twenty (20) μl of the primer solution and 5 μl of the supernatant from each isolate containing the template DNA were added to PuReTaq Ready-To-Go PCR Beads. [0208] PAGE was performed as described by Ausubel and others (2002). The electrophoresis unit used was the LKB 2001 Vertical Electrophoresis System (LKB Produkter AB, Bromma, Sweden). For staining, the gels were placed in glass dishes filled with DDW and a few drops of ethidium bromide (Bio-Rad). They were left to stain for approximately 30 min after which pictures of the gels were captured using the BIS303PC Bio-Imaging system. PAGE details are given in Table 5.4.

[0209] Table 5.4 Percent (%) T and C of resolving and stacking gels used for RISA PAGE. Resolving gels measured 14 cm x 12 cm x 1.5 mm.

a% l\ percent total acrylamide. % C. relative percentage of crosslinker (bis-acrylamide)

[0210] Fatty acid methyl esters (FAME)

[0211 J Single colonies of each marine bacterium were selected from MA plates and grown in 10 ml MB under standard culture conditions. These 'starter ^' cultures were used to inoculate flasks containing 400 ml MB which were incubated under standard culture conditions in a shaking incubator at 130 rpm. Each isolate was grown in triplicate. After incubation, the cultures were centrifuged in sterile centrifuge bottles (Nalgene) at 5,000 rpm (Sorvall Superspeed RC-2 automatic refrigerated centrifuge. Ivan Sorvall, Inc.. Norwalk, CT) for 15 min. The supernatant was poured off and pellets for each sample were pooled in a 50 ml centrifuge tube. The centrifuge bottles were periodically rinsed with small volumes of a sterile 2.5 % NaCl solution to assist in pellet removal. After the samples were pooled in their respective tubes, excess media and rinsing solution were removed by centrifugation at 5,000 rpm ( Universal 32 R centrifuge) for 10 min. The supernatants were carefully poured off and the pellets transferred to glass test tubes for immediate lipid extraction. The lipid extraction and FAME analysis were performed as described by Budge and others (2006). [0212] Glycine-SDS-PAGE of proteins

[0213] Sample preparation for glycine-sodium dodecyl sulfate (SDS)-PAGE was modified from the method described by Smibert and Kreig (1994). Single colonies of each marine bacterium were selected from MA plates and grown in 10 ml of MB under standard culture conditions. The cells were centrifuged at 8,000 rpm for 10 min (Universal 32 R centrifuge) and the supernatents were poured off. The cells in each tube were then disrupted by sonication (Branson sonifier model S-150D) for 1 min at 100 % power. A protein standard was also prepared according to the manufacturer's directions to run with the samples (SDS- PAGE Molecular Weight Standards, Broad Range. Bio-Rad). Glycine-SDS-PAGE was performed according to the methods described by Laemmli (1970). The electrophoresis unit used was the same as described above. The gels were stained with Coomassie Brilliant Blue G-250 (Bio-Rad). Glycine-SDS-PAGE details are given in Table 5.5.

[0214] Table 5.5 Percent (%) T and C of resolving and stacking gels used for glycine - SDS- PAGE. Resolving gels measured 14 cm x 12 cm x 1.5 mm. Resolving gel Stacking gel

|02I5| Tricine-SDS-PAGE of lipopolysaccharide (LPS)

[0216] The sample preparation, analysis, and silver staining of the LPS were performed as described by Hitchcock and Brown (1983). The electrophoresis unit used was the same as described above. Tricine-SDS-PAGE details are given in Table 5.6.

(0217] Table 5.6 Percent (%) T and C of resolving and stacking gels used for tricine- SDS- PAGE. Resolving gels measured 14 cm ^x 12 cm x 1.5 mm.

Resolving gel Stacking gel

% T ^a % C ^b % T % C

I U I 44..53 λ a% f, percent total acrylamide. % C. relative percentage of crosslinker (bis-acrylamide)

[0218] Statistical analyses

[0219] A principal component analysis based on the cellular fatty acid profiles was performed using The Unscrambler 9.7 (CAMO Software Inc., Woodbridge, NJ).

PROTOCOL REFERENCES

AOAC, 2000. AOAC Official Method 959.08 Paralytic shellfish poison. In: Official Methods of Analysis, 17 ^th ed. Arlington, VA: AOAC International.

Ausubel, FM, Brent, R, Kingstone, RE, More, DD, Seidman, JJ, Smith, JA, and Struhl, K. 2002. Short protocols in molecular biology. 5 ^th ed. John Wiley and Sons, Inc., NY.

Ball, RJ, Sellers, W. 1966. Improved motility medium. Appl Microbiol 14(4):670-673.

Beveridge, TJ, Popkin, TJ, Cole, RM. 1994. Electron Microscopy. In: Methods for general and molecular bacteriology. Gerhart P, Murray RGE, Wood WA, Kreig NR, editors. 1 ^st ed. Washington, DC: American Society for Microbiology, pp 42-71.

Budge, S, Iverson, S, Koopman, HN. 2006. Studying trophic ecology in marine ecosystems using fatty acids: a primer on analysis and interpretation. Mar Mammal Sci 22(4):759-801.

Crosa, JH, Tolmasky, ME, Actic, LA, Falkow, S. 1994. Plamids. In: Methods for general and molecular bacteriology. Gerhart P, Murray RGE, Wood WA, Kreig NR, editors. 1 ^st ed. Washington, DC: American Society for Microbiology, pp 365-386.

Galtier, N, Gouy, M, Gautier, C. 1996. SEAVIEW and PHYLO_WIN: two graphic tools for sequence alignment and molecular phylogeny. Comput Appl Biosci 12:543-548.

Hitchcock, PJ, Brown, TM. 1983. Morphological heterogeneity among Salmonella lipopolysaccharide chemotypes in silver-stained polyacrylamide gels. J Bacteriol 154(1):269-

277.

Ivanova, EP, Chun, J, Romanenko, LA, Matte, ME, Mikhailov, VV, Frolova, GM, Huq, A, Colwell, RR. 2000. Reclassification of Alteromonas distincta Romanenko et al. 1995 as Pseudoalteromonas distincta comb. nov. Int J Syst Evol Micr 50:141-144.

Jan-Roblero, J, Magos, X, Fernandez, L, Hernandez-Rodriguez, C, Le Borgne, S. 2004. Phylogenetic analysis of bacterial populations in waters of the former Texcoco Lake, Mexico. Can J Microbiol 50(12):1049-1059.

Kitamikado, M, Yamaguchi, K, Tseng, C-H, Okabe, B.1990. Method designed to detect alginate-degrading bacteria. Appl Environ Microbiol 56(9):2939-2940.

Laemmli, UK. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.

Leifson, E. 1963. Determination of carbohydrate metabolism in marine bacteria. L Bacteriol 85:1183-1184.

Leser, TD, Amenuvor, JZ, Jensen, TK, Lindecrona, RH, Boye, M, Møller, K. 2002. Culture- independent analysis of gut bacteria: the pig gastrointestinal tract microbiota revisited. Appl Environ Microbiol 68(2):673-690.

McFarren, EF. 1959. Report on collaborative studies of the bioassay for paralytic shellfish poison. J AOAC 42:263-271.

Rohlf, FJ, Sokal, RR. 1981. Table 25: Critical values for correlation coefficients. In: Statistical Tables. 2 ^nd ed. New York: W.H. Freeman and Company, pp 166-168. (Table was reproduced from: Snedecor W. 1956. Statistical Methods. 5 ^th ed. Iowa State University Press.)

Smibert, KM, Kreig, NR. 1994. Phenotypic characterization. In: Methods for general and molecular bacteriology. Gerhart, P, Murray, RGE, Wood, WA, Kreig, NR, editors. 1 ^st ed. Washington, DC: American Society for Microbiology, pp 607-654.

Smith, EA, Grant, F, Ferguson, CMJ, Gallacher, S. 2001. Biotransformation of paralytic shellfish toxins by bacteria isolated from bivalve molluscs. Appl Environ Microb 67(5):2345- 2353.