SHELLFISH UPDATED PROTOCOLS

SHELLFISH UPDATED PROTOCOL

ACCELERATED DETOXIFICATION SYSTEM FOR LIVE MARINE SHELLFISH CONTAMINATED BY PSP TOXINS CRAFT CONTRACT N° QLK1-CT-2002-72076 A DELIVERABLE D17 a 1 SCOPE The Shellfish updated protocol draw lessons from the experiments to update and adapt the protocol they used in order to apply it in the pre-industrial pilot. 2

GENERAL ENVIRONMENT 2.1

Biological material

Oysters (C. gigas) or clams (R. philippinarum) should be in sexual resting phase and with a mean total weight of 51.3 ± 5.1 g (oysters) / 38.4 ± 3.4 g (clams). Algal cultures to be used for detoxification purposes should be either live cultures of Isochrysis galbana or concentrated pastes (dead cells) of Skeletonema costatum according to existing facilities. 2.2

Toxin quantification in shellfish tissues

Methods recommended for PSP toxins analysis and described by the LCR Vigo should preferably be used. The mouse test with the AOAC extraction protocol is accepted by the EU regulation bodies and remain the reference bioassay in case of dispute. However, confirmation analysis by HPLC methods should also be carried out, at least at the early and last steps of the detoxification process. Such analysis should be performed by reverse-phase ion-pairing high-performance liquid chromatography (IP-HPLC) according to the method of Oshima (1989). Diluted solutions of PSP-NRC certified reference material (Marine Analytical Chemistry Standards Programme, NRC-Halifax, Canada) for oysters, BCR-CRM 542 (Community Bureau of Reference Brussels) for clams, should be used as external standard for quantitative detection. 3

REFERENCE MODEL 3.1

Background :

From previous experiments using PSP contaminated Pacific oyster Crassostrea gigas to compare the efficiency of different non toxic algal diets in reducing initial toxic level down to the safe limit (80 µg eq STX 100g-1 ) it appeared that : 1. no significant difference could be established between either Skeletonema costatum, Thalassiosira weissflogii, Tetraselmis suesica or Isochrysis galbana detoxification rates providing initial toxic levels were in a 200-300 µg eq STX 100g-1 range and providing experimental temperature was 16° 1°C and Total Particulate Matter in the flume = 0.5 mg L-1 (Lassus et al, 1999; 2000) SHELLFISH 200-01a/11-04

1/5

QLK1-CT-2002-72076

SHELLFISH UPDATED PROTOCOLS

2. as a result, all detoxification data could be melted together and used to build a kind of reference PSP detoxification trend, only suitable for further experiments performed in similar conditions. Considering the wide range of toxicities particularly observed during the first 3 days of detoxification an attempt to discard aberrant values (i.e. keeping only values in the range 50% around the arithmetic mean) was overtaken (Fig 1). The general first order exponential function obtained is : C = 189 e-0.21t, which fits with the general detoxification model proposed by Bricelj and Shumway (1998). 300

µg eq STX 100 g-1

250

y = 189.53 e -0,2141x R 2 = 0.884

200

150

100

50

0 0

2

4

6

8

10

12

14

16

detoxification time (d)

Figure 1

Detoxification curve

From this last definition the time (in days) required to detoxify oyster from 200 to 80 µg eq STX 100g-1 , i.e. what is claimed by shellfishermen, could be calculated from Ln (80 / 200) = -kt, or t = -Ln (80 / 200) / k, ie : t(d) = 0.91 / k, where k is the slope of the exponential detoxification trend : C = C0 e-kt. When applied to the reference detoxification trend it gives : t= 4.33 days for an initial toxic level of 200 instead of 189 µg eq STX 100g-1 . This result is in agreement with what is expected by the Industry. However, it was worth exploring i) whether a detoxification step using sea water and no algal food added was giving the same result and ii) whether subsequent treatments (temperature change, algal food increase) were capable of improving detoxification scores. In the same manner clams reference model needs further input to determine PSP detoxification trend. The model is used in the Man Machine Interface too in order to give an estimation of the remaining time up to the end of the process and display it to the user.

SHELLFISH 200-01a/11-04

2/5

QLK1-CT-2002-72076

SHELLFISH UPDATED PROTOCOLS

3.2

Sea water versus algal diet :

According to previous experiments, PSP contaminated oysters displaying more than 200 µg eq STX 100g-1 initial toxicities are assumed to depict a slow detoxification pattern when only supplied with sea water (almost no exponential trend) and a faster detoxification rate (exponential trend) when fed Isochrysis galbana at 16°C with a TPM content in sea water = 0.5 mg L-1 corresponding to T.ISO concentration : 12,000 cells ml-1. 3.3

Algae concentration effect on detoxification pattern

The results (da / sda = 0.24) show no significant difference in detoxifications trends, whatever Isochrysis concentration levels (32,000 or 36,000 or 72,000 cells ml-1). 3.4

Temperature effect on detoxification pattern

Adjustment of detoxification kinetics for each temperature was realised in a similar way. A significant difference between 20°C on one hand, and 12 or 16°C, on the other hand was revealed by the comparison of linear regression slopes, with 20°C giving the shortest detoxification time. 3.5

Factors likely to significantly modify PSP toxin bioaccumulation patterns

Differences in initial toxic levels (day = 0 detoxification) for a group of oysters can only be explained by changes of individual feeding activities in the population used for one experiment. This oyster population is made of calibrated animals (same size / weight) and is supposed to display an homogenous physiological behaviour. Conversely, the oyster population used for another experiment at a different time of the year can display some different behaviour, especially regarding feeding activity, even if the same size / weight ratio is kept. Besides some local environmental conditions may have changed in the farmer breeding systems thus modifying the way oyster feed on Alexandrium cells. Such effect can be easily detected by a daily monitoring of shell valve activity, cumulated biodeposition rates, filtration rates and ammonia concentration . Ammonia Ammonia is supposed to have some effects on bivalves filter feeding activity as soon as its concentration in sea water exceeds highest values encountered in coastal areas during max chlorophyll a concentration and bloom conditions, i.e. : NH4 = 20 atg.l-1 that is approximately 0.8 ppm. The ammonium level, kept under the threshold by a biological filter, will be checked by discrete measurements made manually (colorimetric method of Koroleff, 1969, is recommended), as there is no ammonia sensor able to provide measures on this threshold. Feeding cells discrete counting Feeding cells concentration should reach the amount of TPM generated by the toxic algae responsible of the bloom : a typical value will be 0.5 mg/l (corresponding for instance to 200 cells/ml A.minutum if it is the toxic strain, 12000 cells/ml I. galbana, 20000 cells/ml S.costatum). In a general manner, 0.5 to 0.7 mg.l-1 TPM is considered as enough to reach a correct detoxification rate and TPM values greater than 2 mg.l-1 are to be avoided (pseudo faeces production). SHELLFISH 200-01a/11-04

3/5

QLK1-CT-2002-72076

SHELLFISH UPDATED PROTOCOLS

Discrete seston analysis should give a useful complementary information during the setting phase of the prototype. It must be noted that fresh seawater supply will be filtered in order to eliminate organic and mineral particles and so avoid inducing significant changes in total seston TPM values. Shell valve activities and Bio-deposition Diurnal shell valve activity recording during all experiments showed a 80-100 % activity as soon as non toxic diets are supplied to contaminated oysters. As a results, it could be recommended to check daily the good physiological state of shellfish processed in industrial scale units. Considering bio-deposition rates they should be as close as possible to those observed during experimental modules attempts It is important to remind that pseudo faeces production, linked to the TPM feeding ratio, should be avoided in the scope of a correct detoxification process to be established. Bio-deposits are to be regularly discarded from the storage tanks. 4

RECOMMENDATIONS

Considering the two sets of experiments performed in this study it is clear that the main objective, i.e. to detoxify PSP contaminated oysters down to levels equal to or lower than the safe level in less than 6 days, was established in order to comply with the expectations of the shellfish industry. This objective should have been at least reached in a way similar to the reference model (see background section 3.1) or with significant improvement directly linked to temperature and non-toxic algae concentration increases. To reach a better understanding of the results the mathematical transformation of the detoxification exponential trend to express the number of days required to have a drop in toxicity from 200 down to 80 µg eq STX 100g-1 was used again (table 1). Experimental conditions 12°C 16°C 20°C 12,000 cells ml-1 36,000 cells ml-1 72,000 cells ml-1

Exponential trends C = 298 e-0.17t C = 228 e-0.14t C = 162 e-0.16t C = 194 e-0.48t C = 99 e-0.38t C = 104 e-0.41t

Time (d) to reach safe level (80) 5.4 6.5 5.7 1.9 2.4 2.2

Table 1 - Calculation of theoretical time needed to reach the safe level (80 µg eq STX 100g-1) from the k value of each exponential trend and for an initial theoretical toxic level of 200 µg eq STX 100g-1.

It appears that detoxification trends observed at either 12, 16 or 20°C display slopes (k values) slightly lower than the reference trend (k = -0.21), even if the initial toxic level is in the same scale of sizes as the expected one. As a consequence, expression of t = 0.91 / k gives detoxification times higher i.e. 5.4 to 6.5 days. At the opposite, k values at T.ISO concentrations of 12,000 to 72,000 cells ml-1 are higher than the k value of the reference model (factor : x 2) and therefore display a shorter detoxification time (1.9 to 2.4 days). From a practical point of view, that is to say to meet the requirements of shellfish industry, especially the economical feasibility of a detoxification pilot operating at industrial scale, it seems the following items can be outlined and particularly taken into consideration for the pre-industrial pilot : When comparing identical experimental conditions (i.e. : 12,000 cells ml-1 T.ISO, 16°C and initial toxic levels roughly identical (194 and 228 µg eq. STX SHELLFISH 200-01a/11-04

4/5

QLK1-CT-2002-72076

SHELLFISH UPDATED PROTOCOLS

-

-

-

100g-1) detoxification times may be as different as 1.9 and 6.5 days, respectively, with corresponding k : 0.48 and 0.14. Taking into account the expected reference detoxification trend with a k : 0.21 this observation would mean that some unidentified parameters, different from the experimental parameters followed in the study, might have a significant effect on the slope of the detoxification trends. An increase of either algae concentration or TPM value does not seem to drastically change the efficiency of detoxification process. As a result, a non toxic algal supply of 0.5 mg l-1 TPM could be appropriate in term of efficiency as well as in term of low cost investment. Similarly, when comparing all detoxification trends at the initial toxic level of 200 µg, it seems either 12, 16 or 20°C experimental temperatures do not significantly modify the k values. As a result, a detoxification temperature close to or not very different from environmental conditions at the time of the treatment could be apply for. Every time the toxic content in oyster will be close to 200 µg eq STX 100g-1 the detoxification process should follow a simple reference model, i.e an exponential trend with a k value as close as possible from 0.21 to ensure a rapid detoxification time, i.e. : 4 to 5 days. Detoxification patterns during post harvest industrial process should ensure ammonia levels lower than 20 µatg l-1 and shell valve activities between 80 and 100 %.

SHELLFISH 200-01a/11-04

5/5

QLK1-CT-2002-72076