The titration of vaccinia virus by the pock counting technique

Abstract

Since the discovery that certain viruses produce discrete macroscopic lesions on the chorionic aspect of the chick allantoic membrane, this phenomenon has been widely used in virus titration. This method, together with bacteriophage plaque counting and plant virus titration by necrotic lesion counts on leaves, has been regarded as analogous to bacterial colony counting and involves the same two basic assumptions: (a) that the growth medium is optimal for the microbe under consideration; and (b) that each lesion, plaque or colony, is derived under optimal conditions from one infective particle. The information provided by such an assay is regarded as absolute, in that it gives a direct estimate of the number of infective units per unit volume, as opposed to the relatively imprecise indirect estimate of the number derived from a quantal titration. Counting techniques are reckoned on theoretical grounds to be superior to quantal assays not only for this reason but also because they are statistically subject to a lower error.There is, however, little published work on the error of virus lesion counting techniques, although they have been extensively used, especially in experimental work with bacteriophages and now with animal viruses titrated in tissue cell monolayers. Miles & Misra (1938) and Davis (1940) showed that bacterial colonies from single drop inocula followed a Poissonian distribution, and it has been assumed that plaque and pock counts are distributed the same way.No particular merit attaches to the fact that plaque, pock and colony counts conform to any one type of statistical distribution. The close approximation of such counts to a Poisson distribution would, however, carry the utilitarian advantage that the standard deviation of the mean count is then easily predictable, being equal to its square root. A large series of vaccinia pock counts was carried out by one of us (C.K.) under routine conditions and, together with a similar series obtained from another laboratory, examined with the specific purpose of ascertaining whether (1) vaccinia pock counts follow a Poissonian distribution; (2) ifthe distribution is not Poissonian, the class of distribution can be identified; and (3) any estimate can be obtained from these data of the standard deviation of vaccinia pock-count tritrations. The results of this analysis are presented here.

Attenuation of Vaccinia Tian Tan Strain by Removal of Viral TC7L-TK2L and TA35R Genes

“…Titers of the mutants in different cell lines were determined by plaque assay and expressed as plaque-forming units (pfu)/ml virus suspension [24].…”

“…The five cell lines infected with VTT and the mutants, and the viruses were harvested after 96 h and titrated in BHK-21 cells. Titers of the mutants in different cell lines were determined by plaque assay and expressed as plaque-forming units (pfu)/ml virus suspension [24] .…”

Single-Dose Assay Technique for Variola Virus12

“…However, a literature survey indicated that, regardless of the pox virus that was being investigated, the variance obtained in pock counting was far greater than expected for a Poisson distribution. Kaplan and Belyavin (1957), who made no adjustments in their assays of vaccinia virus, found that the variance of their pock counts was so wide that no estimate of the coefficient of variation was possible. In other cases, adjustments were made in the counts to reduce the coefficient of variation to about 25 %.…”