Memorandum submitted by Professor R D
Combes (Scientific Director, FRAME, Nottingham) (M1)
EXECUTIVE SUMMARY
The UK shellfish industry has recently been
affected by the statutory closure of several cockle beds, following
the detection of samples causing rapid and severe reactions in
the regulatory approved test for diarrhetic shellfish poisoning
(DSP) toxins, the mouse bioassay (MBA). Several studies have failed
to identify their cause. Following a detailed assessment of the
MBA[1]it
has been concluded that the test should never have been developed
for the routine screening of shellfish samples, as it has a substantially
severe endpoint and is not used in a tiered testing strategy with
non-animal methods. Moreover, it has been used without an optimised
and universal protocol, and apparently without due regard to the
principles of basic scientific methodology, during the UK monitoring
programme for DSP toxins. In view of this, the atypical results
obtained for cockle samples cannot be relied on as being evidence
of a human health hazard. It is recommended that the use of the
MBA should be discontinued as soon as possible, in favour of other
methods, especially those involving non-animal techniques. In
the short-term, these methods should be based on analytical chemical
detection systems and the essential availability of the relevant
pure toxin standards. The lack of any known toxins in samples
should be taken as evidence of lack of contamination. The suitability
of the existing non-animal methods needs to be assessed as a matter
of urgency. It is crucial that all new methods should be properly
validated and their acceptability for their stated purposes endorsed
by recognised criteria and validation centres, before being recommended
to or required by regulatory agencies. In this way, the possibility
that scientifically-unsuitable methods will once again used for
monitoring for the contamination of shellfish with toxins can
be avoided. This gross misuse of laboratory animals and ill-judged
application of science should never be allowed to happen again.
Acronyms
CEFAS—Centre for Environment, Fisheries
and Aquaculture Sciences; DARD—Department of Agriculture
and Rural Development (Belfast); FRS—Fisheries Research
Services (Aberdeen); FSA—Food Standards Agency.
The UK shellfish industry has suffered the statutory
closure of several important cockle beds because samples from
cockles caused so-called atypical results characterised by unusually
rapid, severe and extreme reactions in the EU-stipulated test
for diarrhetic shellfish poisoning (DSP).
1. The results of several investigations
have raised important questions about the use of the MBA for the
above purposes: (a) its suitability for the detection of DSP toxins
in cockles; (b) the scientific rigour of the protocols being used;
(c) interlaboratory variations in the conduct of the regulatory
assay and in the interpretation of the data; and (d) the justification
for the animal licences that were granted for the tests on mice,
in view of the un-anticipated severity associated with the atypical
nature of the effects observed later.
2. DSP is the mildest form of the four types
of shellfish poisoning in humans, the main symptoms being diarrhoea,
nausea, vomiting and abdominal pain with sufferers normally recovering
within a few days.
3. Although there have been genuine cases
of DSP contamination due to consumption of cockles, to the best
of my knowledge there have never been any medically-endorsed instances
of diarrehtic shellfish poisoning.
4. High molecular weight fat-soluble acidic
polyether chemicals, namely okadaic acid (OA) and its derivatives,
and dinophysis toxins (DTX-1 and DTX-2), are primarily responsible
for DSP. Other potentially important toxins include neutral polyether
lactone pectenotoxins (PTX) and yessotoxins (YTX; including homo-yessotoxin
(45-OH YTX)). Several other toxins have been associated with DSP,
including spiramino acid and the azaspiracids (AZAs).
5. OA and the DTX toxins are potent inhibitors
of the PPP family of serine/threonine protein phosphatases that
reversibly phosphorylate many proteins with the concomitant hyperphosphorylation
of proteins in ion channels of the intestinal epithelia, disrupting
mechanisms which maintain water balance thus leading to other
degenerative changes in the absorptive capacity of the small intestine.
PTX toxins also affect the cytoskeleton of cultured cells, and
OA can disrupt mitotic spindle formation in cells by binding to
phosphatases, and these effects could be related to the ability
of the toxin to induce diarrhoea.
6. A 24-hour mouse bioassay (MBA) is the
standard test for DSP toxins, and is stipulated in the current
EU Directive. However, the test fails to detect all PTX toxins
and is insensitive to AZAs.
7. The MBA involves a series of solvent
extraction steps of shellfish samples, before the final extract
is dissolved in Tween 80 and injected intraperitoneally (ip) into
mice. The death of 2 or 3 of the mice within 24 hours following
exposure is recorded as a positive result.
8. Exposed mice can undergo a variety of
abnormal responses, including prostration, hypothermia, and tachycardia,
with death occurring between 2-5 hours following dosing.
9. The use of solvents is crucial, as it
dictates the sensitivity and specificity of the test, by affecting
which toxins are extracted and detected. Also, the assay can be
affected by several problems, including the presence of zinc,
incomplete separation of the water phase during aqueous washing
of the diethyl ether and acetone, and carry-over of organic solvents,
resulting in lipophobic chemicals or solvent, respectively, being
in the extract for dosing.
10. Atypical MBA results were first detected
during routine monitoring programmes conducted by CEFAS in June
2001, but for 6-9 months these results were reported as normal
positives.
11. Atypical results are characterised by
the mice dying very rapidly following dosing, from heart failure,
preceded by shock and extensive trauma, accompanied by more violent
and rapid leg and body movements, and agonal breathing after collapse.
12. Atypical results in the MBA have not
been detected in any laboratory, other than CEFAS, in the UK,
and samples that gave atypical results at CEFAS have proved negative
when re-tested in other laboratories. Furthermore, there have
been no cases of adverse effects occurring in individuals that
have ingested cockles from beds yielding samples that have generated
atypical results.
13. There is further evidence for interlaboratory
variation in data obtain with the MBA, and also for results being
dependent on body weight, sex and strain of mice.
14. The FSA contends that the atypical positive
are attributable to a real effect caused by an unidentified substance
in the cockles that elicits symptoms in mice that are similar
to those produced by neurotoxins.
15. A recent independent audit of the three
laboratories (CEFAS, FRS and DARD) involved in the monitoring
of cockle beds for DSP in the UK found that: (a) each laboratory
used a different protocol; (b) there were differences between
the requirements for a positive and negative result; (c) there
were no provisions for internal and independent quality assurance;
(d) controls were lacking; and (e) there was a lack of precision
in the methods detailed in the Standard Operating Procedures being
used.
16. There is evidence in a report produced
by the FSA that carry-over of both organic solvents used, and
also of water, into the final extract did occur in some of the
tests undertaken. The extent of this carry-over varied between
replicate samples, laboratories and also between shellfish species.
17. The conclusion in the FSA report that
diethyl ether is not the cause of the atypical response might
be erroneous as the conditions of the confirmatory experiments
that were undertaken to investigate this possibility were not
the same as those that applied during the routine use of the MBA.
18. The MBA can result in severe welfare
costs to the individual animals involved. Thus, mice can rapidly
(within 30 mins following dosing) become subdued, unresponsive
with bluish extremities, and cold to the touch. This is followed
by prostration and extension of the rear legs. The animals can
also display clear signs of disorientation, paralysis of the hind
limbs, breathing difficulties, and a violent jumping reaction,
just prior to death. The MBA is conducted without anaesthesia
and the use of humane endpoints.
19. It is clear that experiments involving
the MBA, particularly when atypical results for DSP have been
obtained that could be due to a substance acting like a neurotoxin,
fall into the substantial category of experiments, under the UK
(Animal Procedures) 1986 Act, and this fact should be taken into
account when consideration is being given to granting a licence
for the test.
20. The MBA for DSP has been developed specifically
as a test for the presence of the relevant shellfish toxins, and
the endpoint detected in the assay (the death of mice) makes no
attempt to model the non-lethal clinical signs (diarrhoea) observed
in humans following their ingestion of such toxins via the consumption
of infected shellfish.
21. There are many deficiencies in the experimental
design of the MBA, as undertaken in laboratories in the UK. The
most important of these is the lack of negative and positive controls,
despite the fact that this was criticised in a report of a mission
conducted by the Health & Consumer Protection Directorate-General
of the EC (DG SANCO) made in July 2002. The inclusion of the controls
is absolutely crucial for correct data interpretation, to ensure
the absence of false-positives due to the generation of artefacts
during extraction and of any false negatives due to the lack of
extraction of toxic material that was actually present. False
positives can also be due to the carry-over of either a toxic
organic solvent, or possibly of lipophobic material in the aqueous
phase into the final test extract.
22. Some controls should be undertaken concurrently
with the test, and others should have been used when the test
was first developed, or when any significant part of the protocol
had been altered (eg when the extraction conditions (solvents
used, their order and the times and temperatures of extraction)
and/or the starting material (species of shellfish, or tissues
extracted) were changed).
23. These controls should verify the scientific
robustness of the test, and the fact that each laboratory undertaking
the test can obtain positive results under appropriate conditions.
Thus, each laboratory should establish historical positive and
negative control databases. On a routine basis, only a solvent-negative
control needs to be run concurrently with every test.
24. The MBA also involves ip exposure by
injection (ip) even though humans ingest shellfish. Such dosing
in the MBA has resulted in an over-estimate of toxicity.
25. The benefits to human health to be gained
by MBA testing for DSP are dubious since at worst the condition
is not life-threatening, and the test is likely to be scientifically
unjustified producing meaningless data. On the other hand, the
adverse welfare costs to the animals are substantial, and there
are other in vivo and in vitro tests that might be more suitable
for routine screening.
26. The MBA for DSP should never have been
developed as: (a) it involves death as an endpoint, without the
possibility of applying more-humane endpoints; (b) it often results
in animals suffering severe shock and trauma within a very short
time after dosing; (c) despite the extreme severity of the assay,
anaesthesia is not used; (d) it lacks scientific justification
concerning the relevance of the endpoint to human health; (e)
there are important shortcomings in the experimental design of
the assay; and (f) the test is not used on the basis of prior
data from in vitro screening.
27. The atypical results obtained when using
the MBA for detecting DSP toxins cannot be assumed to be indicative
of actual hazard. The continued use of the MBA, especially to
detect atypical results for DSP toxins, has been a gross misuse
of laboratory animals and a lack of application of sound principles
of scientific methodology. As a consequence, HO licences for the
use of the MBA for DSP toxins should be revoked and no more should
be issued, whether or not atypical results are expected.
28. Several in vitro methods can be used
to detect DSP toxins, including: (a) a cytotoxicity assay; (b)
immunological tests; (c) analytical methods, such as TLC, (HP)LC
and MS; and (d) phosphatase inhibition assays. However, none of
these has been formally validated, one problem being the lack
of a complete set of pure toxin standards.
29. Existing non-animal test alternatives
to the MBA should be assessed and improved urgently, taking account
of the availability of all relevant toxin standards. Mechanistically-based
assays, involving immunological and specific phosphatase binding
systems, should be considered first, and then (Q)SAR modelling
in the longer term.
30. In the immediate term and short term,
the monitoring of cockle beds for DSP toxins should continue by
using analytical chemistry methods. Where evidence for a DSP toxin
or any other toxin cannot be found, cockle beds should be re-opened.
The references supporting the statements made
in the above evidence can be found in this paper which is about
to be published.
December 2003
1 Combes, R.D. (2003). The Mouse Bioassay for diarrhetic
shellfish poisoning: a gross misuse of laboratory animals and
of scientific methodology. Alternatives to Laboratory Animals
31, (in press). Back
| 
