75, 481-483 (1976)
of Herpes Simplex MARY
of Microbiology The Pennsylvania
Virus DNA from the “Hirt
and Specialized Cancer Research Center, The Milton S, Hershey Medical State University College of Medicine, Hershey, Pennsylvania 17033 Accepted August
Herpes simplex virus DNA was isolated from infected cells by its preferential release into the “Hirt supernatant” and its subsequent sedimentation in neutral glycerol gradients. The DNA isolated by this procedure was free from contamination with cellular DNA and was identical in its kinetic complexity to the DNA isolated from intact virions.
Polyoma virus DNA and simian virus 40 DNA can be isolated from infected cells by the preferential release of virus DNA into the “Hirt supernatant” (2, 6). This method has proven to be highly effective for the separation of virus DNA from the vast majority of cellular DNA because the high molecular weight cell DNA is precipitated under the conditions of extraction (2). In this communication, we demonstrate that herpes simplex virus (HSV) DNA, together with some small DNA fragments of cellular origin, is also preferentially released into the Hirt supernatant. Pure virus DNA is obtained by subsequent band velocity centrifugation of the supernatant DNA in neutral glycerol gradients. Vero cell monolayers (approximately lo7 cells per 16-02 bottle) grown in medium 199 supplemented with 10% fetal calf serum were infected with HSV-2 (strain 324) at a multiplicity of infection of one plaqueforming unit per cell. After 20 hr at 37”, the cells were washed with phosphatebuffered saline and lysed gently for 15 min at room temperature with a lysing solution (2 ml per bottle) containing 0.01 M Tris (pH 8.0), 0.6% sodium dodecyl sulfate, and 0.01 M EDTA (2). NaCl was added to a final concentration of 1.0 M, and the lysate was gently poured into a small centrifuge tube and maintained at 4” for 10 hr. The supernatant was separated by centrif1 Author dressed.
to whom reprint
ugation at 17,000 g and centrifuged isopycnically in ethidium bromide-CsCl gradients (6). As illustrated by tube B in Fig. 1, most of the DNA in the supernatant is in band 2, which is of higher buoyant density. The density of band 2 is characteristic
FIG. 1. Buoyant density of HSV DNA in ethidium bromide-CsCl gradients. Infected Vero cells were lysed and kept cold after the addition of 1.0 M NaCl. The pellet was separated from the supernatant and resuspended in Tris-EDTA (pH 8.0) and digested with Pronase (1 mgiml) at 37” for 30 min. Part of the solution containing DNA from the pellet was mixed with part of the supernatant (gradient A), and the remainder of the supernatant was centrifuged separately (gradient B) in ethidium bromideCsCl gradients in a Ti50 rotor at 42,000 rpm for 50 hr at 4”. The starting density of CsCl was 1.566 g/cm” and the concentration of ethidium bromide was 100 pgiml (6).
should be ad481
Copyright All rights
0 1976 by Academic Press, Inc. of reproduction in any form reserved.
of the increased guanine plus cytosine content of herpesvirus DNA over that of the host DNA (3). To examine the size of the DNA released into the Hirt supernatant, infected cells were labeled with [14Clthymidine for 15 hr beginning at 5 hr postinfection. The cells were lysed and the supernatant was collected. The DNA in the supernatant was then sedimented in a neutral glycerol gradient (3) with 3H-T4 DNA as marker. As shown in Fig. 2, the majority of label in the supernatant sediments as expected for intact HSV DNA (4). Genome-length DNA molecules were also observed in the electron microscope. Cellular DNA released into the supernatant (band 1, tube B in Fig. 1) sedimented slowly in glycerol gradients (unpublished observation). Thus, to remove the contaminating host DNA, the total DNA released into the Hirt supernatant was preparatively centrifuged in a glycerol gradient. The rapidly sedimenting HSV DNA was collected and its purity was examined by
FIG. 2. Glycerol gradient sedimentation of HSV DNA. Supernatant from HSV-infected cells containing ‘YLlabeled DNA was centrifuged with 3H-T4 DNA in a lo-30% glycerol gradient containing 0.02 M Tris (pH 8.01, 1.0 t&J EDTA, and 1.0 M NaCl in an SW41 rotor at 40,000 rpm for 3 hr at 20” (3). Twenty-drop fractions were dripped from the bottom and the DNA in each fraction was collected onto nitrocellulose filters after precipitation with cold trichloroacetic acid. The amount of radioactivity in each fraction was then determined. Direction of sedimentation is from right to left. The arrow indicates the position of T4 DNA.
3. CsCl buoyant density centrifugation of purified HSV DNA. Rapidly sedimenting 3H-Iabeled DNA from a glycerol gradient was collected and mixed with 14C-labeled viral and cellular DNA and centrifuged in a CsCl gradient containing 0.1 M Tris (pH 7.51, 1.0 mM EDTA, and 1% sarkosyl in a Ti50 rotor at 40,000 rpm for 65 hr at 20” (5). Ten-drop fractions were collected and the amount of radioactivity in each fraction was determined as in Fig. 2. Direction of sedimentation is from right to left. 3H. A-A a 92. , O-O, FIG.
isopycnic centrifugation in CsCl gradients (5). As shown in Fig. 3, the DNA purified by this method is free from contamination by cellular DNA. The kinetic complexity of the 14C-labeled HSV DNA purified by the above method was compared to the kinetic complexity of 3H-labeled HSV DNA obtained from intact virions. The two DNAs were mixed, and the rate of reassociation of the two was monitored by digestion of single-stranded DNA by S,-nuclease (5). As shown in Fig. 4, the Cotliz values of the two DNAs are identical and are as expected for the kinetic complexity of HSV DNA (1). Recovery of virus DNA by this procedure is approximately 72%, and the yield is about 10 pg of HSV DNA per lo7 cells. In our experience, this amount is 50 times greater than that obtained by isolating DNA from intact virions grown in the same number of cells. Thus, this method
DNA; if size alone is the predominant factor, HSV DNA should have been precipitated in the same manner as the high molecular weight DNA of the host. Since the HSV DNA appeared in the supernatant, it appears that the genome size of the DNA viruses is not the predominant factor in the preferential release of DNA into the Hirt supernatant. This phenomenon is clearly open to further investigation. ACKNOWLEDGMENTS
Cot imol * sec/lltl
4. Reassociation kinetics of HSV DNA. Rapidly sedimenting ‘*C-labeled DNA from a glycerol gradient was mixed with L3HlTdR-labeled virions. The DNA was isolated by Pronase treatment followed by phenol extraction. The rate of reassociation of sonicated DNA was measured by the digestion of single-stranded DNA by S,-nuclease in zincacetate buffer (pH 4.5) for 30 min (5). Specific activity of 14C-DNA was 2 x lo4 cpm/mg and that of 3HDNA was 4.8 x loj cpmimg. Resistance of heatdenatured DNA to S,-nuclease was 2.8% for 14C!DNA and 2.6% for 3H-DNA. A-A, 3p1[;O-O, “C. FIG.
should enable the isolation of the large quantity of HSV DNA which is required for the study of the molecular biology of herpesviruses. The size of the HSV genome (lo8 daltons) is approximately 30 times greater than that of polyoma and simian virus 40
Encouraging discussions by A. Pater are acknowledged. This work was supported by Contract NO1 CP 53516 within the Virus Cancer Program of the National Cancer Institute, NIH, PHS, and by Grant Number 1 P30 CA18450, awarded by the National Cancer Institute, DHEW. Note added in proof. Pseudorabies virus DNA is also released into the Hirt supernatant (Ben-Porat et al., Virology 69, 547-560, 1976). REFERENCES N., and ROIZMAN, B., J. Viral. 10, 565-572 (1972). 2. HIRT, B., J. Mol. Biol. 26: 365-369 (1967). 3. LINDAHL, T., ADAMS, A., BJURSELL, G., BORNKAMM, G. W., KASCHKA-DIERICH, C., and JEHN, U., J. Mol. Biol. 102, 511-530 (1976). 4. KIEFF, E. D., BACHENHEIMER, S. L., and ROIZMAN, B., J. Viral. 8, 125-132 (1971). 5. PATER, M. M., and MAK, S., Nature (London) 258, 636-639 (1975). 6. SEBRING, E. D., KELLY, T. J., JR., THOREN, M. M., AND SALZMAN, N. P., J. Virol. 8, 478-490 (1971). 1. FRENKEL,