The Neutralization of Vaccine Virus by Serum of Vaccine-Immune Animals

Abstract

Summary Previous investigations by ourselves and others have shown that infection of the rabbit with vaccinia appears to follow the suitable introduction into the skin of a single particle of virus. Based on this observation a method has been described (16, 17) (which is not original with us) by the use of which the number of infective particles in a suspension can be estimated with considerable accuracy. The present investigations were undertaken in order to learn whether a similar method could be applied in studies of the neutralization of vaccine virus by its immune serum. Our first experiments were carried out to learn the effect of adding a constant amount of serum to various dilutions of virus. It was found that the infectivity of the virus was reduced in a regular manner, and that when the percentage of takes resulting from inoculation of the serum-virus mixtures was plotted against the concentration of virus, a reasonably smooth curve was obtained. This curve, however, differed significantly from the one obtained by plotting concentration of virus suspended in normal serum against percentage of takes in being definitely flattened. This appeared to indicate that while the amount of virus neutralized was a function of the amount added, the concentration of serum remaining constant, it was not a fixed proportion of the total. On the contrary the proportion neutralized by a certain quantity of serum increased somewhat as larger amounts of virus were added to the mixtures. Although the titration curve of virus in the presence of immune serum differed significantly from that of virus in the presence of normal serum, its regularity indicated that by defining its midpoint a consistent measure of the ability of serum to neutralize virus should be obtained. Thus the neutralizing ability of the serum would be expressed as the difference between the titer of virus in the presence of normal serum and its titer in the presence of the serum being tested, these both expressed as the dilution of virus which would be expected on inoculation to give an equal number of positive and negative results. It was realized that the variance of this difference would be greater than the variance of either of the titrations of virus and would, therefore, be of appreciable magnitude. A number of neutralization tests were performed in order to determine whether this would be such as to impair seriously the value of the results obtained. The outcome indicated that while a greater range was covered than when virus alone was titrated, the range was not excessive. The results of the neutralization tests described revealed also that a regular relation existed between the concentration of serum present and the amount of virus neutralized. When the logarithm of the difference in titer between virus in the presence of normal and of immune serum was plotted against the logarithm of the dilution of immune serum, the experimental points were seen to distribute themselves in the form of a rectangular hyperbola. This indicated definitely that while the absolute amount of virus neutralized by serum decreased as the serum was diluted, the relative efficiency of the serum as a virus-neutralizing agent increased. A satisfactory explanation for this cannot be offered at the present time, although two comparisons immediately present themselves. A similar effect on the concentration of salt is exhibited in the dissociation of the salt of a weak acid and weak base on dilution. On the other hand, the disparity of size between the microscopic virus-bearing particles and antibody molecules suggests that physical factors may be important, and a curve similar to the one observed could be obtained if the reaction were one of adsorption.