* * * * * * * *

first submitted to Nature in July 1990; also rejected from Lancet, Intl.Immunol. and N.Y.Acad. of Sci. (of which I was a member);
no response from Science and AIDS Res.&Hum. Retrovir.


Summary: One of the dark secrets of modern medicine is that we still do not know what makes the smallpox vaccine (discovered 200 years ago) to function as a vaccine. It's an ancient tabooed theme: the fact is that none other than Louis Pasteur was ostracized and even challenged to a duel after he attempted to debate EXACTLY this same problem back in 1880 in French Academy. 
It is also still not known what makes any other live "attenuated" vaccine to work. This gap in our knowledge seems to be the most obvious cause of the failure to find an AIDS vaccine and, more generally, for failure of the whole 'new generation' of high-tech vaccines. So to say, new vaccines do not work because it is not known why old ones work.

This speculative paper shortly reviews the entangled political and historical background of this problem and presents a hypothetical solution involving a simple explanation of the transmutation of virulent viruses into vaccines (called 'attenuation' by L.Pasteur). It is based on two principal assumptions:
1. Contrary to the popular view that the immune system cannot cope with more than 25 antigens simultaneously, I think that this number is much larger, meaning that it may be possible to achieve protection from a rapidly mutating microbe by immunising against all its antigenic forms simultaneously.
2. This very technique is used at least in vaccines for rabies and smallpox, as these are actually cocktails containing complete repertoires of antigenic forms of the corresponding viruses, and not wild viruses undergone some miraculous 'attenuating' mutation.

If correct, these ideas promise nothing less than cures for AIDS, malaria and perhaps even for cancer. A great advantage is that experiments may be conveniently conducted on such disease as the common cold, for which a protective vaccine could be developed and tested after a few weeks of work.


It's a sad fact that failure in quest for AIDS vaccine was not a big surprise for immunology. There already exists a wide variety of diseases -- ranging from the common cold to cancer -- for which vaccines are also still unavailable despite a much longer history of research. Indeed, effective vaccines are very difficult to make. But why it is so difficult?

The simplest explanation is that it's difficult just because vaccinologists don't know how to make effective vaccines. Of course, this trivial idea is not popular among "leaders of science", yet sometimes it may be found in official sci. literature (though with some amount of equivoking added -- see citations below).

It means that there is nothing in vaccinology that may be called as "scientific basis", "guiding theory" or even "basic understanding" for the mysterious art of developing vaccines -- it is purely empirical technique relying exclusively on hit-and-miss methods.

Vaccinologists have at their disposal a broad (and rapidly expanding) variety of technical methods for constructing not a vaccine but a so-called "candidate vaccine" - i.e. a preparation with high immunogenicity and which is, perhaps, perfectly safe, but whose capacity for preventing disease is absolutely unpredictable. Hence the quest for vaccines is confined to an endless sifting in the hope that some day a "candidate" may turn out to be a true vaccine.

Not only is this procedure tedious, but the assessment of "candidates" proceeds very slowly and erratically -- to some extent resembling Brownian motion. This strategy was invented more than century ago and has succeeded in finding cures for a substantial number of infections. But even if a vaccine was occasionally discovered, there is absolutely no knowledge of the basic reasons for this success, such as to facilitate making vaccines for other microbes. For example, origins of smallpox vaccine are not known at all (2), so the great Jenner's discovery is today represented by a stock of vials labelled "vaccine" with no more benefit for scientific knowledge than if it was a souvenir from extraterrestrial visitors.

Thus, the real problem seems to be that we need an understanding how the existing effective vaccines were created.

All today's really effective vaccines are the so-called live vaccines which were derived from wild viruses by growing them in unnatural conditions which somehow transform them into vaccines. An understanding of what happens to microbes during this metamorphosis or, in other words, what is the relationship between the virulent microbe and the vaccine was confined to statements that 'the vaccine is a modified microbe' (before Pasteur) or that 'the vaccine is an attenuated microbe' (after him and until today). The trouble is that the meaning of attributives 'modified' and 'attenuated' never were defined.

It's really difficult to define official views of today's orthodox immunology on these problems. As an example I should mention special issue of Science journal devoted to a funny event: 10 years ago Institute of Medicine (IOM) announced an ambitious plan to develope a half-dozen of new miraculous vaccines during next decade. Decade has suddenly ended but NONE of the promised vaccines appeared. Apparently, this issue was meant by editors to discuss the causes why the actual result is pricisely nil. Yet, certainly, this commendable undertaking (real surprise from this tabloid) was not supported by contributors who showed absolutely no desire to talk about any troubles in vaccinology. So, the discussion turn out to be limited by two editorials by Jon Cohen.

The leading article (3) entitled "Bumps on the vacine road" starts very well:
"And when it came to describing the obstacles that hinder vaccine developement the respondents be they from Russia ... or Brazil - had remarkably similar views. The scientific unknowns are the highest hurdles..." Apparently, the talk is about something very similat to the content of my article. After such start, one may guess, the article should proceed to definitions of these unknowns and discussing ways of their resolving.
Alas, the article diverts instead to dull protracted complaints about funding problems, insufficient number of governing centers, etc. Nevertheless, at the very end, in two last paragraphs the author returns again to these highest hurdles quoting J.Salk saying: "Salk believes that one of the main scientific obstacles is that many scientists r&d vaccines "don't have a clue" what's required to make an effictive vaccine"
In my view, the trouble is that NOBODY has a clue what's required to make an effective vaccine. Salk's sentence is slightly different. But what does it mean exactly? How numerous are "many (stupid?) scientists"? Whether the rest of scientists have this clue? Perhaps, J.Salk has one?

The quotation goes further: "It's chaos... There is going to be a need for more awareness not of the pathogen, but of the host - says Salk". This seems to be just a useless conjecture where the clue, perhaps, may be found. Therefore, J.Salk has no clues. And at this point the discussion of highest hurdle on vaccine's road suddenly ends. All other questions remain unanswered.

I think it is quite typical way of talking about the problem of vaccines' understanding: vague statements about some tremendous difficulties so that it's almost impossible to understand what the talk is about. In contrast, the clear statement that absence of knowledge how to make vaccines may be the most important cause of failures in quest for new vaccines is almost unique for modern vaccinology.

The more frequent approach is that vague sentences about existence of some problems with vaccines' understanding are accompanied by statements implying that solving this problem is not important, not urgent, not possible, or not necessary.The following misperceptions (or, actually, little lies) underlying such logic are the commonest:

1. Our understanding of vaccines is usually presented as problematical only in connection with live vaccines -- which were declared old and obsolete in the 1970s. But none of their high-tech successors -- after over 20 years of ballyhoo about this 'new generation' of vaccines -- can be compared with Jenner's smallpox vaccine discovered almost 200 years ago. Though it is not mathematically rigorous, but it seems obvious that the questions 'what is needed to make an effective new vaccine' and 'what makes the old vaccines effective' have a common solution.

2. AIDS propagandists very frequently present the rapid antigenic variation of HIV virus as the main scientific obstacle for developing vaccine. To be true, such opinions quite frequently are followed by honest remarking that a number of effective vaccines are already available against other viruses who also display high rates of antigenic variation (poliovirus is usually mentioned). Therefore, the actual problem is to understand how existing vaccines overcome antigenic variation. Assuming that overcoming antigenic variation is the clue to making effective vaccine, this is obviously just a less general formulation of problem of vaccines' understanding.

3. Numerous genetic works on sequencing and comparing wild and vaccine strains of viruses cling to the view that, to understand how vaccines work, we just need to find out what 'precise genetic changes lead to attenuation'. This gives us the illusion that 'approximate' knowledge is already available, but this is not true.

4. It is often proudly stated that vaccination is 'as much a science as an art' or that it is based more on superficial intuition than on precise guiding theories (e.g. ref. 10). This probably reflects the cause of the desire to conceal the problem in question. Modern vaccinology is like a nervous child about to lose faith in Santa Claus who avoids discussing this theme with his friends and screams at every adult who discusses it with a fatuous smile. The threat of shifting 'from magic to science' is not more pleasant for the modern vaccinological establishment.

And finally, to illustrate once more the peculiar kind of logic prevailing in vaccinology, we see such a quotation as: 'the AIDS epidemic is too urgent for scientists to be worrying about how vaccines work' (4). Is not it like saying: 'Fish is too urgent for fishermen to be worrying about bait'?


Apparently, the major reason why this article might be considered iconoclastic is that it conflicts with the popular opinion that the ultimate scientific basis of vaccination was created by 'the great Pasteur'.

We may first mention another popular mainstream view that new generation of high-tech vaccines is really scientific while traditional vaccines are only 'empirical' (e.g. 10) which also suggests the existence of fatal troubles with this scientific basis. Yet, apparently, I have an even more surprising view that Pasteur actually did not pretend to be a founder of any scientific basis at all!

The first vaccine (against smallpox) appeared in 1798, a quarter of century before Pasteur was born, when the British physician Edward Jenner succeeded to publishing a pamphlet (7) reporting the discovery that a cowpox inoculation protects against subsequent exposure to smallpox.

It is generally agreed by historians that this was an extremely controversial work comprising a galaxy of mistakes, empty claims and, actually, fraudulent data (for more details see refs.3,5,8, and 9); yet, worst of all in the context of this article is that there already was an alternative quite widespread method of preventing smallpox by smallpox inoculation. So, Jenner's goal was substantially more difficult than just to show that his smallpox prophylactic was effective; It was necessary to prove that vaccination ensured better results than its competitor.

Alas, the first independent trials gave controversial results. And, apparently, one of the cleverest tricks under these unfortunate circumstances (together with treating the opposition as 'serpent's venom' and ascribing every case of vaccination failure to the use of 'spurious' vaccine) was Jenner's tactical attempt to present vaccination as 'scientific' in contrast to the barbarous smallpox inoculation.

Obviously, choices for the imitation of a scientific theoretical explanation of Jenner's vaccine was very limited at the end of the 18-th century. It was possible either to state that cowpox and smallpox are absolutely different diseases or to say that they are essentially the same. Jenner (quite correctly, as it turned out later) selected the second option and began (with his supporters) to fiercely protect this view.

Cowpox was announced by Jenner to be 'modified smallpox'; later his followers went further and called cowpox 'smallpox which was not smallpox but yet prevented smallpox' (5) . These statements stress two major ideas: first, that smallpox and cowpox are essentially the same disease; and, second, that they differ one from another. Of these ideas the first remained unproven for almost a century (a work showing that inoculating heifers with smallpox yields cowpox was mentioned by Creighton in 1889 but was not known to Pasteur in 1880). And the neeed to delineate the essence of the difference between smallpox and cowpox (what this present article is about) was simply 'disregarded'.

Yet, somehow Jenner's vaccination became a world-wide established practice within several years after its inception, and his self-contradictory claim that cowpox is 'modified' smallpox became the so-called 'Jenner doctrine'. As early as 1802 (just 3 years after publication of his paper) he was awarded a grant of 10,000 pounds by the British parliament, and in 1803 a Royal Vaccine Institute was founded, with himself as President.

But that could not have been the consequence of a real assessment of the merits of vaccination, since this was obviously impossible in such a short period of time. An accurate historical analysis of Jenner's miraculous success is still to be written. One view is that it reflected the usual opportunistic crowd psychology -- for whom the smallpox inoculation seemed more expensive and dangerous; the other view is that it reflected the unprecedented unabating enthusiasm of British parliament in promoting Jenner's discovery. I can't help to mention here that its vaccine board was put under investigation.

Naturally, despite the danger of political persecution, a number of people immediately attacked Jenner's doctrine at its controversial points. These people were dubbed as 'antivaccinists', and the struggle against them was conducted so successfully that even today the word 'antivaccinism' is usually associated in the public mind with the darkest obscurantism and with Torquemada's Inquisition rather than with serious science.

Antivaccinatism is most commonly interpreted as reflecting some absurd fear that humans will be transformed into cows after inoculation of cow's matter, as is seen, for instance, in the Encyclopedia Americana. In fact, this was just a rash joke by one of the first antivaccinationists, B.Moseley, later used skilfully by Jennerians as a serious example of the nonsensical obscurantism which a great discovery encounters. Moseley's real goal, as stated in his book, was to 'arrest the hurry of public credulity until the subject had undergone deep, calm, and dispassionate scrutiny'(6). Obviously, it is difficult to say anything against it. To be sure, the first wave of antivaccinationism at the beginning of 19-th century did not try to dig into the theoretical background of Jenner's doctrine, being more absorbed by practical questions of vaccine efficacy and adverse effects.

Antivaccinism almost abated at the middle of century, but after the recurring smallpox epidemic in the 1870s, the second wave of antivaccinists appeared and left several diligently researched and fundamental works (see 5,8,9). In my view, Charles Creighton was the brightest of these.

It is terrible to state such heresy, but Pasteur was one of these 'second generation' antivaccinationists. Soon after finding that a culture of fowl cholera vibrios exposed to air turns into a vaccine, he attempted in 1880 to attack Jenner's doctrine with purely antivaccinationist accusations -- trying to argue in the French Academie that every existing explanation of the relationship between cowpox and smallpox can be reduced to the statement that 'vaccine is vaccine', and that his opponents deliberately muddled this problem by meaningless speculations (e.g. ref 11). He argued that his new vaccine was more scientific than cowpox: indeed, he had at least proven that his vaccine was actually derived from a virulent microbe.

Pasteur also proposed a speculative explanation for what happens to a microbe when it turns into a vaccine: being a chemist, he suggested that atmospheric oxygen somehow 'attenuated' the microbe, apparently in the same way as it oxidised chemical substances. Obviously, it was just a hopeless hunt for a theory, as using the word 'attenuation' instead of 'modification' (which is now presented as Pasteur's biggest service) actually did not state anything at all.

The result of these debates was that Pasteur was challenged to a duel and, apparently, preferred to abandon activity in this field under an absolutely clear threat of ostracism from medical profession. He immediately cancelled further attacks on Jenner's follies, and in 1881 at a big congress in London he delivered a eulogy to Jenner, thus posing as a faithful follower rather than an opponent or critic (e.g. 11).

At the same time, it may be said with absolute confidence that the problem of understanding the relationship between vaccine and virulent microbe still disturbed Pasteur's mind. He continued to propose other hopeless theories attributing 'attenuation' to the ability to form spores (for anthrax) and even to the geometrical shape of microbes. It may be taken as certain that his invention of the term 'attenuation' was absolutely not related to vaccines for rabies and smallpox. Yet, apparently, any further focusing of attention on this problem soon conflicted with the prestige of his own vaccines (which were no bed of roses). So, it seems natural that Pasteur abandoned this field.

Regular antivaccinationism was also successfully buried and the problem has been suspended since that time.

It is difficult to say what Pasteur thought of his own speculation about the relationship between vaccines and virulent microbes. Perhaps he agreed with C.Creighton (5): 'If it is ever to be cleared up, it will need more something more of exact diligence than is implied in the invention pf phrases like the 'methodological attenuation of virus', or the construction of bold figures of speech like 'vaccines charboneux', or 'vaccines rabiques'. This may sound excessively desperate, but it is exactly what I think as well.


It may be just a question of vocabulary, but Pasteur's term 'attenuation' is not well suited to describe many classical vaccines. For instance, it has been known for a hundred years that the latent period before the patient's temperature rises is shorter when smallpox vaccine is inoculated than when smallpox itself is inoculated and that naturally acquired smallpox has the longest latent period of all (12). The same shortening of the latent period is seen with rabies vaccines as well. (13). In other words these vaccine strains show significantly stronger 'virulence' in early stages of parasitemia. Consequently, their vaccinating properties cannot be attributed to a low rate of self-reproduction, weaker pathogenicity, or restricted viability as it is intuitively implied by the word 'attenuation'.

In many other cases, a general weakness of the vaccine microbe seems a more reasonable hypothesis, but adduced examples suggest that apparent loss of 'virulence' has little or nothing to do with the protective properties of vaccines, and that the actual explanation of vaccines is to be found in some more complicated shift in host-parasite relations.

As a starting point for a logical approach to a new understanding of vaccines, we may take the fatal contradiction between vaccines and Darwin's law of evolution. For instant, if the virus of smallpox is cultivated on a cow's skin, then accordingly to Darwin, the smallpox virus ought to acquire the ability to adapt to the new host, most likely by acquiring some tolerance to the cow's immune system. But what happens instead is that the smallpox becomes catastrophically susceptible to the human immune system -- almost directly opposed to what Darwin's theory would predict. Similar effects are found with all other methods of 'attenuation', so it appears that this process has nothing to do with Darwin's 'survivival of the fittest'.

Accordingly to elementary genetics, Darwin's law embodies spontaneous mutations leading to variation. Thus, if it is possible to explain 'attenuation' by some sort of programmed phenotypic alterations, the task is nearly solved.

There is no big selection to choose from. Pasteur's methods of 'attenuation' are the same for different kinds of parasites, so the mechanisms involved must be common to viruses, bacteria, and other microorganisms. In fact the only universal phenomenon satisfying such conditions is so-called antigenic variation.

Actually any microscopic parasite must use tricks to escape the host immune response and survive. Antigenic variation is the simplest, most effective, and hence the most widespread mechanism of that kind. Microbes using it can exist in a huge number (usually hundreds) of antigenically distinct variants. In the course of parasitemia, as the specific immune response neutralises the primary invasion, the parasite population has time to generate a new antigenic form for which the immune response has to develop again from the start, triggering a new wave of parasitemia and establishing a chronic infection.

It is worth mentioning that the clinical manifestations of such diseases are not necessarily periodic fevers: the simultaneous coexistence of several switching forms means that the overall course of the disease may be absolutely arbitrary.

Molecular mechanisms of switching are different between species. For bacteria it has been found that switching is defined by complicated rearrangements of nucleotide sequences (14). New variants appear through replacing fragments of translated nucleotide sequences to other variants selected from a large number of silent genes. For viruses it is thought that a high rate of mutation is caused by the absence of enzymes ensuring faaithful reproduction of genetic information (15). But there are observations indicating no accumulation of changes during rapid viral mutation(16), so it is likely that viruses also switch among a restricted number of antigenic variants, thus allowing violation of Darwin's law.

There are actually only two basic methods of 'attenuation' - the long cultivation of parasite strains in vitro under influence of various artificial factors and repeatedly passing the strain through experimental animals. Assuming that the microbe subjected to 'attenuation' is capable of antigenic switching, it is rather obvious that the most remarkable and stable state which can be achieved with these methods, is the strain containing all or almost all antigenic variants of the microbe.

It is also absolutely clear why this strain will make a good vaccine: administering a complete repertoire of antigenic variants confers immunity simultaneously against all variants thus preventing the microbe varieties from persisting. Of course, instead of an increase in antigenic heterogeneity during 'attenuation', some antigenic variants might overpower others; in any case, success in achieving a complete repertoire will depend on the choice of experimental conditions.

The difficulty involved here approximately corresponds (by subjective comprehension) to the present severe difficulty of empirically finding the right method of 'attenuation'. For instance, one could naturally assume that sequential passages of microbe through different species of experimental animals can cause both an increase and a decrease of antigenic heterogeneity in function of the duration of the disease or even the health conditions of the animal used. Understanding vaccines as multivalent strains not only substantiates the fact of the very existence of vaccines, but also explains a number of well-known phenomena which otherwise are treated as pure paradoxes:

1. Antigenic switching was always considered a nearly fatal obstacle to creating vaccines. It is therefore unclear how immunologists of the past, unaware of the existence of this problem, nevertheless succeeded in designing effective vaccines for antigenically unstable microbes.

2. There is huge quantitative incompatibility between the high rates of genetic mutation and the amazing stability of vaccine strains: vaccines for rabies or smallpox were isolated from their 'wild' relatives more than century ago and have since evolved in absolutely independent directions. This obviously led to increasing differences between 'wild' viruses and vaccines and consequently to loss of antigenic similarity. But, in fact, no worsening of vaccine quality over time has been reported. What seems to be happening is that the apparent mutation rate obviously involves switching processes, while any loss of vaccine efficacy must be due to hyperevolution of the whole antigenic repertoire -- whose rate must be several orders of magnitude slower (17).

3. There already exists a clumsy explanation for the poor quality of inactivated vaccines in comparison with live vaccines involving speculation about a worsening of antigen 'presentation'. Our new analysis yields an alternative variant:

Any real vaccine includes different antigenic variants present in concentrations differing by several orders of magnitude. For live vaccines this is not significant. as any small population will yield its own wave of parasitemia providing sufficient specific immunity. But for inactivated vaccines the quality of the immunity elicited is obviously proportional to the concentration of antigen, so the protective effect of such vaccine will not be uniform for different antigenic variants.

4. And, in my opinion, the most striking argument in favor of our proposed hypothesis is its ability to explain the transmutation over time of virulent microbes into vaccines. This transmutation is usually a long continuous process. There is no clear opposition between 'deadly virus' and 'perfect vaccine'; there must be a set of intermediate forms through which the microbe strain passes during the process of attenuation...

This fact is quite incompatible with what we would expect from viruses. First, viruses are very tiny organisms, and even a single mutation usually means an abrupt change in virus properties, so coexistence of various sets of viral forms with continuous change in virulence seems a miracle. Second, even if such a sequence of forms exists, it is another miracle that a virus can be forced to pass through them by such simple measures as growing it in cows or in green monkey kidneys. Another possible opinion (of Pasteur, if I am not mistaken) is that attenuation is just the overgrowing of one 'virulent virus' by 'vaccine'. But this also does not fit the dynamics of attenuation.. Quite reliable models of population overgrowth (18) indicate that this process is described by exponential function, or, in other words, that it should be a very abrupt process.

A classical quantitative illustration of this transmutation phenomenon is the so-called fixation of rabies virus described by Pasteur (13). After transmission from a dog to sequential passages in rabbits, the duration of rabies abruptly decrease from about 20 days to about 10 days after first two or three passages and then for a very long time this value approaches a stable value of about 7 days.

This exactly coincides with our proposed understanding of vaccines. Assuming, for example, that the total number of antigenic forms of rabies virus is 100, and that wild strain contains 5 forms, such a time period of fixation is quite natural. The rapid initial phase is obviously due to the 10-fold growth in the number of antigenic forms -- from 5 to 50-60 during the first several passages (such a mixture may be interpreted as 'bad vaccine'). Then there must be a very slow phase of further slight shortening of the length of the disease -- corresponding to a gradual approaching of this number to 100 with very small 2-fold relative change in it.


The idea of antigenically heterogeneous vaccines does not pretend to be new -- polyvalent and stock vaccines have been in use since at least the beginning of this century. Antigenic heterogeneity of a single monovalent vaccine against rapidly mutating virus is also thought to be a positive factor for vaccine efficacy (19). But the maximum number of distinct antigens assumed possible in a single mixture is by 'grotesque superstition' restricted to about twenty. This is not an official number but rather a general opinion with no distinct underlying explanation. It is just that investigators in very diverse fields don't dare to mention larger numbers. So the maximal number of serovariants in a single polyvalent vaccine (20), the maximal declared number of revealed genetic variants of virus persisting simultaneously in AIDS patient (21), and even an asserted maximal number of antigens which can be expressed on the surface of vaccinia virus (22) -- all fall in the very narrow range between 15 and 25.

There are two major reasons why larger numbers are thought to be unrealistic. The first, of course, is the phenomenon of antigenic competition which holds that simultaneous administration of only two antigens may mutually suppress responses evoked by them. On an emotional level this phenomenon is perceived as a sign of weakness of the immune system, but that is absolutely wrong. In other reports, experimental effects of this kind are explained not by exhaustion of the immune system, but by normal regulatory reactions (23), which most mean optimising the work of the immune system.

This formulation does not prohibit multivalent vaccines: antigenic competition is actually a rather esoteric phenomenon of questionable practical significance; frequent reports of successful polyvalent vaccination simply mention that in this case 'antigenic competition is not a major factor'(20).

The second reason for mistrust of multivalent vaccines is the natural concern that, for instance, one hundred antigenic variants will provide a hundred-fold greater parasite concentration than that attained during the normal course of disease. Therefore, administration of live multivalent vaccine should evoke clinical symptoms which are a hundred-fold more severe -- leading to death. This argument can be refuted by using inactivated multivalent vaccines, and , as for live vaccines, mathematical modelling easily offers a number of theoretical ways of softening this effect. For instance, it is quite natural to assume the existence of some diffusion-like processes between the rise of parasitemia and its clinical manifestation; if so, the short-term peak of microbial concentration after multivalent vaccination may arrive with no clinical symptoms at all,. since they simply have not had the time to display themselves.

One can also assume that simple interference limits the total concentration of microbe to some more acceptable level; and there are other arguments as well, so this is not a fatal hurdle to multivalent vaccines.

Counting the number of antigens in a single isolate is actually a hopeless task. Every antigenically distinct microbe has on its surface a substantial number of different epitopes which (or groups of which) can also be treated as separate antigens. In addition to that, every real vaccine strain contains a huge number of contaminating antigens also interacting with the immune system. Adding up these factors, the maximal number of antigens which the immune system can cope with at once must be 'very large'.

Common sense alone tells us that multivalent vaccine do not seem very dangerous. For instance. it is impossible to imagine that a vaccine constructed from 100+ serovariants of the common cold virus could evoke severe immunodeficit or death; more likely, this would evoke mothing more than a common cold, so as vaccine for this disease would seem to be the most convenient target for applying the presented theory.

There are also interesting observations concerning AIDS patients with 1000+ sexual partners (24). Elementary numerical estimation shows that they must have contracted at least several hundred antigenic variants of HIV, but no unusual disease pattern has been reported. The authors, apparently embarrassed by this finding, have asserted the existence of some mystical mechanisms forbidding superinfection, but this does not seem well-founded.


I gratefuly thank Harris Coulter and Karin Schumacher (Vaccine Information and Awareness) for help in editing this manuscript.


1. Karpas A. - Nature, 368, 387, (1994).
2. Baxby D. Jenner's smallpox vaccine. (Heinemann, London, 1981).
3. Cohen J. - Science, 265, 1371 (1994).
4. Brown P. - New scientist, 132(1793), 21, (1991).
5. Creighton C. Jenner and vaccination. A strange chapter of medical history. - (London, Sonnenschein & Co, 1889).
6. Moseley B. - Medical tracts. (T.Cadell & M.Davis, London,1800).
7. Jenner E. An inquiry into the causes and effect of the variolae vaccinae. (Sampson Low, London, 1798).
8. Crookshank E.M. The history and pathology of vaccination (London, 1889).
9. White W. The story of great delusion. (Allen, London, 1885).
10. Beardsley T. - Sci.American. 272(1), 88-95, (1995).
11. Vallery-Radot R. - La vie de Pasteur. (Flammarion, Paris, 1946).
12. Hime T.W. - Brit.Med.J., 2,117-120,(1892).
13. Pasteur L. - Semaine medicale, 4, 318-320, (1884).
14. Borst P. & Greaves D.R. - Science, 235, 658-667, (1987).
15. Holland J. et.al. - Science,215,1577-1585,(1982).
16. Salinovich O. et.al. - J.Virol., 57, 71-80, (1986).
17. Pays E. - Trends in Genetics, 2, 21-26, (1986).
18. Eigen M. - Naturwissenshaften, 58, 465-523, (1976).
19. Bolwell C. et.al. - in: New approaches to immunization (Eds. Brown F. et al.), 51-56, (Cold Spring Harbor, 1986).
20. Ambrosino D.M. & Siber G.R. - J.Infect.Dis., 154,893- 896,(1986).
21. Saag M.S. et.al. - Nature, 334, 440-444, (1988).
22. Perkus M.E. et.al. - Science, 229(4717), 981-984, (1985).
23. Pross H.F. & Eidinger D. - Adv.immunol., 18, 133-168, (1974).
24. Hahn B.H. et.al. - Science, 232, 1548-1553, (1986).