The isolation of defective tobacco mosaic virus strains.

This paper describes a method for the isolation of strains of tobacco mosaic virus (TMV) which consist of infective particles that are labile under conditions where ordinary TMV is extremely stable. Two such strains have been isolated, PML and PM2, which are distinctly different in their properties. Evidence will be presented to demonstrate that tobacco plants infected with strain PM1 fail to synthesize viral protein and that plants infected with the PM2 strain synthesize a virus-like protein which fails to aggregate in vivo with the viral nucleic acid to form complete virus particles. The possibility that defective strains of TMV could be recovered was suggested by the results of studies dealing with the induction of mutations with nitrous acid." 2 It was noted that a class of survivors of nitrous acid treatment existed that would induce necrotic lesions on a local lesion host but from which infectivity could not be recovered. We postulated that perhaps some of these nontransferable lesions contained a labile infectious entity which was destroyed during transfer manipulation. Such defective strains have now been isolated by the direct inoculation of systemic-type host seedlings with nitrous acid-treated TMV preparations at limit dilution. Isolation of the Defective Strains.-A preparation of the U1 strain' of TMV was treated with nitrous acid according to the technique of Mundry and Giererl to three levels of survival: 4.2%, 0.28%, and 0.015%. The treated preparations were inoculated to tobacco (Nicotiana tabacum L. var. Samsun) seedlings at a concentration estimated to infect half or fewer of the seedlings with the expectation that many of the infections would result from a single infective particle. The infection present in those seedlings appearing diseased was tested for ease of transfer by macerating a small piece of leaf tissue and applying the macerate to a leaf of the local lesion host Xanthi (Nicotiana tabacum L. var. Xanthi-nc). Those seedlings yielding few or no lesions were set aside and allowed to grow for about a month in 6 in. pots before being retested. They were then tested for the presence of labile infectious material in the following manner: A mature leaf exhibiting symptoms symmetrically on each side of the midrib was divided along the midrib. One of the half-leaves was ground in a mortar with the addition of an approximately equal weight of 11/15, pH 7, phosphate buffer, and the grindate was allowed to incubate for two hr at room temperature. The other half-leaf was extracted with phenol according to the method of Schlegel.4 The two leaf extracts were then applied to opposite halves of 6 Xanthi leaves. In a typical experiment, many (at least 40) lesions were induced on each of the Xanthi half-leaves rubbed with the phenol extract and an average of only 2.5 on the half-leaves rubbed with the buffer extract. An attempt was made to demonstrate infectivity in the above lesions by cutting them out and macerating them between two ground glass spatulas with a drop of 11/15, pH 7, phosphate buffer containing celite. The macerate was then rubbed