A study was undertaken on the effect in vivo, in the guinea pig, of equine encephalomyelitis virus antiserum upon the antigenic response to active, as compared with that to formolized, inactive virus. It was found that when animals were given subcutaneously a proper amount of hyperimmune serum 1 hour before inoculation, in the subcutis, of either active or of inactive virus, no immunity was induced against an intracerebral test of more than 1,000 and less than 10,000 M.L.D. of virus. This preventive power of the serum was lost by its dilution, the loss being proportional to the dilution, and, on the other hand, more serum was needed to obtain the blocking effect as the quantity of virus was increased. When an insufficient amount of serum was introduced into the animals along with the same quantities of active virus or formolized vaccine, a certain number of those receiving the untreated virus succumbed to virus infection in the course of the inoculations, but the survivors were rendered resistant to the intracerebral test; all the guinea pigs treated with higher dilutions of serum and with formolized material were brought safely to an immune state.
The point to be stressed then is that antigenic stimuli present in untreated active virus and in formolized virus tissue suspensions in which no active virus is demonstrable by drastic tests (1) and which are wholly noninfective in animals (1), are completely inhibited from acting by the use of proper amounts of immune serum.
The mechanism underlying this preventive power of adequate amounts of serum may be explained on the basis of facts deduced in preceding papers of this series (1, 3) and in the present article. We have shown that 3 x 107 m.i.u. of active virus contains a sufficient amount of antigen to induce immunity without the necessity of its multiplication in the animal body. This has been fully established by the similar degree of resistance brought about by 3 x 107 m.i.u. of virus formolized to a degree in which no active virus could be revealed (1). The assumption that the blocking effect of serum in the quantity employed prevents multiplication of the virus which is reflected in the production of inadequate amounts of antigen, is therefore untenable, since this effect was obtained when a sufficient amount of antigen was present in "living" as well as in "killed" virus. On the other hand, with insufficient amounts of immune serum (to be noted in higher dilutions shown in Table II), only the active virus could multiply—the formolized vaccine was not affected in respect to its antigenicity by these quantities of serum—and so produce more antigenic substance. This substance, in turn, brought about greater resistance in the host.
The precise action of proper amounts of serum in preventing development of immunity by both active and inactive virus is not definitely known. However, two hypotheses are offered for consideration: the first implies that the action of the serum is direct, that is, by entering into combination with the antigens to bar antigenic capacity; the second ascribes to the serum an indirect action, on the cells of the body, in such a way as to make them unable to react to the antigenic stimuli present in the inoculated materials.
The identity of these antigenic stimuli in virus suspensions containing the active, infective agent or this agent inactivated by formalin is at the present time undetermined. If virus were obtainable in pure state, free from extraneous material, the answer to this question might be readily given, but it is quite a different matter when the substance called virus is a mixture of the infective agent, of inflammatory tissue products, of tissue, etc. We have, however, shown that induced immunity is not due to the presence of "living" virus, but whether the antigenic action originates from "killed" virus or from another constituent of the suspension is not clear. On the other hand, Sabin (7) suggests the possibility that the virus may not be the direct antigenic stimulus but that some substance on which it acts and which becomes liberated from infected cells may be the factor responsible. While this subject awaits the results of further study, we believe that formalin inactivates the infective agent in virus suspensions and preserves the antigenic component therein, whatever its nature may be.
It would be of interest if this phenomenon of prevention of antigenic capacity by proper amounts of immune serum might apply to such materials which by their very nature do not multiply in the body of the host, e.g., toxin and antitoxin. Theobald Smith (8) and later Park (9) demonstrated that in mixtures of diphtheria toxin-antitoxin, when smaller amounts of immune serum (antitoxin) are used, the toxicity of the mixture is retained and immunity results; if the serum is increased, toxicity is reduced and immunity occurs irregularly, and if more serum is added, no toxicity nor immunity results. This is supported by the experiments of Hartley (10) on washed precipitates from underneutralized, neutral, and overneutralized mixtures of antitoxin and toxin: those derived from underneutralized material are toxic and powerfully antigenic; those from neutralized, atoxic and of good antigenic action, and from overneutralized, atoxic and of low antigenicity. Hartley states, moreover, that the precipitate reactions of toxicity and antigenicity bear a close relationship to the nature of the mixture from which they are produced.
There is, therefore, a connection between the preventive reactions of the serum on the two forms of virus and of antitoxin on toxin in respect to toxicity and antigenicity. Furthermore, the toxin is rendered atoxic with retention of immunizing capacity by formalin: the production of toxoid or anatoxin (Glenny and Hopkins (11), Ramon (12))—again a condition related to the effects following formolization of the virus. It has, however, been stated that "in an immunizing mixture prepared with modified [formolized, but partially detoxified] toxin the antitoxin present does not within wide limits affect the antigenic power" (Glenny, Hopkins, and Pope (13)). It is not known whether a preliminary injection of antitoxic serum could have prevented the antigenic power of fully detoxified toxin, that is, after the passive immunity induced by the serum disappears. If a preventive action of antitoxic serum could be shown under these circumstances, a remarkable correlation of the reactions of proper amounts of antitoxin to toxoid and of proper amounts of immune serum on the virus would be evident.
Finally, the inhibition of antigenic power of both active and inactive virus by immune serum has been demonstrated to apply to the virus of equine encephalomyelitis in guinea pigs and no generalizations of the application of the phenomenon to other viruses are intended.