return to

 Email received from Dr Ruth Watkins  Sept 22 2008

" I have gone into some detail but in the climate of rumour and partial information around BTV8 the dangerous practice of not using the the vaccine because of misinformation is gaining ground and going to be difficult to shift."

Very interesting postings from the research lab and the vaccine manufacturer, I am so glad you are able to get these excellent replies and post them on the net.  It is uniquely useful.  May I add my own as a diagnostic virologist and clinician?
a) It is not disclosed what test was used in Romania to detect the response of the imported sheep to their vaccination.  Was this an in house assay or a commercial kit?  The quality of the test is very important. The test used to detect antibody does not necessarily always detect antibody even when it is there in response to the vaccine for example.  The production of virus antigen for use in a test, it may not be BTV8 but another serotype in this case, and its dilution and preparation for coated plastic ELISA test plates may remove or alter important antigens.  In particular in human virology there is an attempt to correlate antibody tests for vaccine response with protection so the result has a meaning.  That is why the challenge work done by the manufacturer of vaccine is so important in BTV8 when there are no in vitro established tests of vaccine response correlated with protection.
b) It is curious that no vaccination date seems to be known for the imported sheep.  This is critical for interpretation of any test.  Were they a uniform bunch from a single flock, or a dealers mix?  Why was there no vaccination certificate dated and signed by a vet for identified sheep?  Clearly the blood could have been taken too soon after the test for an immune response to have occurred in all the vaccinees.  Some may never have been vaccinated.  Occasional animals may not respond or the vaccine may have been incorrectly stored or administered.  I would have thought the antibody testing should have been done by the UK before export of sheep, also the rules state they should be vaccinated 60 days before export to a BTV8 free zone from a BTV8 protection zone if testing is not done if I recall the rules right.
c)  It is clear that protection remains in humans long after the detected antibody response has faded to levels that are no longer detectable.  Presumably this is because memory immune cells remain, and antibody may still circulate at levels below detection but still afford protection against exposure to small amounts of virus (the challenge experiments probably involve amounts of virus rarely reached in the natural encounter).  But how long? Certainly not for ever. (viz studies on native Alaskan populations with high HBV infection rates and vaccination against HBV; eventually protection against infection and carriage of HBV is lost after successful vaccination if there are no boosters given after completion of the original course of vaccination)  Protection does not result from vaccinations with no detectable antibody response in the first instance when an appropriate antibody test is done.  This has been clearly documented for HBV and rubella for instance in humans.
d) As far as I can gather from the small amount of info I have on saponins (the adjuvant in bovilis BTV8 vaccine) they induce a strong memory cell and antibody response.  They are not cited as producing the specific cytotoxic T-cells, CD8 cells, as would occur if live vaccine is used (and of course in infection by the wild virus).  But intranasal vaccination with saponin and respiratory virus antigens in the form of ISCOMS see below, for example can induce good mucosal antibody level of IgA in addition to a good serum antibody level IgG to the respiratory virus, the combination providing a better protection against respiratory infection than serum IgG alone.  Protective vaccination using saponins should provide a positive test result for antibody when it is successful if the right test is used. 
e) I also note that in pregSure vaccine for BVD produced by Pfizer for use in cattle, safe for use in any stage of a cow's pregnancy, the adjuvant consists of quil A (derived from saponin)  together with 3 other adjuvants, cholesterol, amphigen base and liquid paraffin.  I believe this would produce cages for the vaccine antigen, the quil A and cholesterol assembling into structures called ISCOMS, to make the inactivated BVD more antigenic.  BVD is related to rubella virus of humans, very difficult to culture in the lab (I note BVD virus is titred in guinea pigs).  It may be very difficult to make good antibody tests for BVD that can be correlated with protective response to vaccination.  The protective neutralising antibody can be present but undetected by any test because the crucial antigen to measure it cannot be generated for use in a laboratory test; some antigens must also have the natural shape of the protein preserved, something that is easily lost in preparations used to test for antibody.
I have gone into some detail but in the climate of rumour and partial information around BTV8 the dangerous practice of not using the the vaccine because of misinformation is gaining ground and going to be difficult to shift. 
I recall that when vaccination against HBV first came out coupled with better testing for immunity when I was a junior virologist, my ancient professor believed he had been exposed to HBV, by breathing in blood aerosol so often in the post mortem room etc, that he must have gained immunity through this natural medical contact (not the normal routes of transmission, sex. blood and mother to neonate).  Surely he would not need the HBV vaccine.  We tested him and found him susceptible so he was vaccinated!  Miracles don't often happen in virology.   
Interesting further information is provided on Promed by Chris Oura on the standard commercial test being used to detect BTV8 antibody- a competitive ELISA.
This type of test is not the most sensitive but is relatively more specific than it is sensitive.  It is probable that the BTV antigen used in this test, though Chris Oura does not specify this, is not BTV8 but another BTV serotype.  It is not a test designed to detect response to vaccines but to screen for past infection. 
The basis of a competitive ELISA is that any antibody in the test serum (the vaccinee in this instance) is applied together with antibody from a donor known to have had infection in the past (this serum is porvided as part of the kit). Both sera are put in the same well on the test plate together and compete with each other for binding to the virus antigen 'fixed' to the solid surface of the plate.  Unless peptide or recombinant antigens are used the antigen in the preparation must either be all the virus structural proteins prepared from virus released into the cell culture supernantant, or a cell lysate containing at least some of the non-structural proteins involved in virus replication as well as the virus structural proteins.  Thus a broad range of virus antigens are used to provide the target for the spectrum of antibodies produced.
Initially when antibody is produced in response to an inactivated vaccine it will start at a low level and a low affinity for the antigen.  As time passes, in terms of months, up to 6 months in humans, the antibody matures so that clones of high affinity antibody remain at a relatively high concentration to all the virus proteins that all act as antigens.  Some proteins may be more dominant than others in stimulating a high concentration of antibody.  Thus it is a big ask that vaccinees in the early stage of their response produce sufficiently high concentrations of antibody to most virus proteins particularly the dominant immunogenic proteins, at affinities high enough to successfully compete with that produced by a past infection. That is why this type of antibody test is not the most suitable for assessing vaccine response, nor for detecting acute infection.
I do not know if the test used at Pribright, a commercial test, is specifically formulated for each different animal species- It would not have to be to work, so I suspect there could be differences in sensitivity when testing sheep for instance if the test uses cattle serum to compete with the test serum from sheep for instance.
In human diagnostic virology it is particularly important to get the testing for HIV infection correct, and before the advent of highly sensitive PCR testing for HIV virus the strategy for the type of antibody tests used was crucial.  In acute primary infection with HIV the last type of antibody test to become positive was a competitive ELISA and it took roughly a month to reach a strong positive after seroconversion by a more sensitive test (the antibody tests used were not quantitative and if they were they may have shown a rise in antibody concentration over months.  The affinity of IgG matured at about 6 months).  Tests for total antibody, IgM or IgA antibodies, microparticle agglutination, indirect ELISAs, Western blots, all preceeded the competitive ELISA in showing the presence of antibody in acute infection.  There is of course no vaccine against HIV.
I assume Pirbright are using the same test in acute infection as in screening for past infection and vaccine response screening. The competitive ELISA test used at Pirbright detects antibody fairly promptly in acute BTV8 infection- by the time an animal is severely ill and the DEFRA vet arrives to take blood, Pirbright have found both antibody and virus to be present in most instances of acute infection they have diagnosed though they could find the presence of virus alone if they have samples early enough (it probably takes at least a week after onset of clinical illness in the animal on average to collect serum).  The immune response is clearly greater in an infected ill animal than in a vaccinee exposed to relatively a tiny amount of antigen. 
It is difficult to make in house tests as quality control is a very important issue and difficult to maintain to an evenly high standard for many reasons.  Hence the importance of commercial manufacture.  Ideally a peptide or recombinant BTV8 surface protein or part of, needs to be generated for specific BTV8 vaccine response testing by microparticle agglutination or indirect ELISA testing to create a sensitive and specific test.  Would there be a sufficient market to justify it?  I think not.  So we will have to live with the limitations of what is provided.  I am confident that Pirbright's standard of testing is second to none in the veterinary world.
These are both quite long and detailed.  I think the apparent investigation of a cell mediated response might seem as though the antibody testing is ultimately not to be relied upon but I think the problem lies with the antibody testing used.  I think this is confusing for farmers who are paying for testing or reading about testing.  They are subject to untrue rumours about the side effects of vaccination, the test results are not clearly set out and no clinical group answers their questions thoroughly and honestly.  The vaccine manufacturers must protect themselves against being sued.
I had difficulty selling my vaccinated ewe lambs in Llandovery mart the other day: just like at Penrith, mine were the only vaccinated female sheep offered for sale in the 4026 to be auctioned that day.