THE BIOLOGICAL POLYOMA
ACTIVITY OF DIFFERENT
FORMS OF
VIRUS DNA AND VIRAL DNA FRAGMENTS
M. GRAESSMANN, A. GRAESSMANN, E. HOFFMANN, J. NIEBEL, AND K. PILASKI
Institut fiir Molekularbiologie und Biochemie der Freien Universitdit Berlin, Berlin-W., F.R.G.
(Received 19 November, 1973) ABSTRACT. Mouse tissue culture cells were infected with different forms of Polyoma Virus (PV) DNA or virus DNA fragments by means of a microinjection technique and stained for PV-tumor (T) antigen and virus capsid (V) antigen 48 hr after injection. The efficiency of PV-DNA 1 (20 S) to induce T- and V-antigen was within the same range as the efficiency of the full virus particle. DNA II (16S) showed a reduced capacity for both T- and V-antigen induction. Single stranded DNA molecules (16 or 18S) and double stranded DNA fragments (12S) led to T-antigen but not V-antigen synthesis. Simultaneous transfer of 16S and 18S DNA revealed T- and V-antigen formation. I. INTRODUCTION The infectious agent of Polyoma Virus (PV) is the viral DNA [1]. Several groups observed cytopathic effects and plaque formation [2, 4] following exposure of tissue culture cells to PV-DNA. The infectivity of the virus DNA is very small. Substances like DEAE-D increase the infectivity of the viral DNA [5] but compared with the full virus particle the efficiency of infection is still lower than 1~o. It is likely that this phenomenon is caused by the cell surface membrane [5, 6]. Some years ago we developed a microinjection technique which enables us to inject nucleic acids directly into mammalian tissue culture cells [7]. This technique is very sensitive and reproducible for the assay of infectious viral DNA. In this study we describe the different capacities of PV-DNA I, DNA II, single stranded DNA molecules and double stranded DNA fragments to induce tumor (T)- and capsid (V)-antigen synthesis in mouse tissue culture cells, using the microinjection technique. II. MATERIAL AND METHODS PV-DNA was obtained from infected primary mouse kidney cells by the selective extraction procedure of Hirt [8], modified by Westphal [9]. (PV strain IL 11 was a gift of M. Fogel, Weizmann Institute.) The DNA preparation contained both cellular and viral molecules (DNA I and II). From these we subsequently separated cellular and viral DNA by velocity sedimentation (5-20~ sucrose, 1 M NaC1, 0.001 M Tris, 0.001 M EDTA, pH 7.4). Centrifugation was carried out in a Spinco SW 65 rotor, 54000 rpm, 2 hr at 20~ Fractions were collected in an Isco fractionator. DNA I was separated from DNA II by two different methods. First by CsCl-ethidium bromide [10] equilibrium centrifugation (1.56 g CsCI m1-1, 10-1 mg ethidium bromide ml -a in 0.02 M 233 Molecular Biology Reports 1 (1973) 233-241. All Rights Reserved Copyright 9 1973 by D. Reidel Publishing Company, Dordrecht-Holland
Tris, p H 8.0) in a Sprinco angle head Ti 50 rotor at 40000 rpm, 40 hr, 15 ~ The second method involved centrifugation over by alkaline sucrose gradients (5-20% sucrose, 0.5 M NaC1, 1 m M EDTA, p H 12.0) in a Spinco SW 65 rotor, 54000 rpm, 2 hr, 20~ Single stranded D N A molecules (16S and 18 S) were obtained by centrifugation of D N A II in alkaline sucrose (5-20%, 0.5 M NaCI, 1 m M EDTA, p H 12.5) for 5.5 hr, 54000 r p m at 20~ (Spinco SW 65 rotor). To obtain double stranded fragments D N A II was treated with pancreatic DNase I (Worthington). The incubation mixture (1 ml) contained 6 x 1 0 - 3 mg D N A , 2.5 m M MgC12 solution. After digestion for 15 min at 22 ~ the reaction was stopped by addition of SDS. Subsequent to phenol extraction, dialysis and concentration the fragments were centrifugated on a
I
I
.
t
!
I
I
II
3
1,5
2
1,0
x t3
E~ ol
1
0.5
|
5
10
I
!
15 20 Fractions
I
25
Fig. 1. PV-DNA, prepurified on neutral sucrose gradient (Spinco rotor SW 65, 20~ 2 hr, 54000 rpm), was centrifuged in CsC1-EtBr (spinco angle head rotor Ti 50, 40 hr, 40000 rpm). Tritium labeled PV-marker DNA was added. The gradient was fractionated from the bottom of the tube. We extracted the dye by isopropanol treatment and determined optical density (O O) and radioactivity (9 (3). DNA I banded at a density of 1.589 g cm-8 and DNA II at 1.556 gcm -3. 234
neutral sucrose gradient (conditions mentioned above). Fractions with D N A smaller than 12S were pooled and recentrifugated in neutral sucrose gradient 5-15%. Radioactivity was measured according to the m e t h o d o f Bollum [11 ]. All preparation steps were done under sterile conditions. Microinjection by means of glass capillaries was performed as described [7]. Secondary mouse culture cells grown on small glass slides (divided into numbered squares o f 1 m m 2) were used as recipient cells (Fig. 3). T-antigen was detected by indirect, and V-antigen by direct immunfluorescence technique [12] (antisera were kindly supplied by Dr. M. Fogel).
I
I
I
16-18s
I
5 3s
3
%
2,0
Z
t
E 2
0
,cz Ix) 0"1
- 1,0
5
10
15
20
25
Fractions Fig. 2. Mouse kidney ceils were Lqfected with PV (20 PFU cell-l) and labeled with 25 laCi 3H Thymidine per dish (90 man diameter) between 29-31 hr post infection (O O infected culture, V V mock-infected culture). DNA was obtained by the selective extraction method and centrifuged on alkaline sucrose gradient (Spinco SW 65 rotor, 2 hr, 54000 rpm, 20~ pH 12.0). The gradient was eluted from the top of the tube and simultaneously measured for optical density from infected culture at 254 nm 9 9 235
o~
Fig. 3. The microinjection technique. - (a) demonstrates the injection needle near the recipient cell - (b) the tip of the injection needle inside of the cell (c) the injection needle inside of the cell immediately after injection- (d) the recipient cell after transfer of virus DNA, the needle is again outside of the cell.
III. RESULTS A N D DISCUSSION
1. The Infectivity of DNA I The native form of PV-DNA (DNA I) is a double stranded, supercoiled molecule with a sedimentation coefficient of 20S [13]. The viral DNA can be obtained either by isolation from the full virus particle [2] or by the selective extraction method from virus infected cells [8]. DNA I was purified by two different methods. First by equilibrium centrifugation in a neutral CsC1 gradient containing ethidium bromide. In this condition DNA I bands at a higher density than D N A II [14] (Fig. 1). Alternatively we used centrifugation in alkaline sucrose gradient. At pH 12.0 D N A I is converted into a cyclic coiled D N A (53S) while DNA II is denatured into a linear (16S) and a circular (18S) single stranded molecule (Fig. 2). Following neutralization (pH 7.4) the cyclic coiled D N A cosediments with a marker supercoiled D N A I in neutral sucrose gradient [9]. To test the biological activity of D N A I, fractions 17-20 from the alkaline sucrose gradient (Fig. 2) were pooled, neutralized and dialyzed against diluted phosphate buffer. The DNA solution was then concentrated under vacuum to a final concentration of 0.3 mg ml- 1. The microinjection was carried out under a microscope at 400 fold magnification. The recipient cells were grown on glass slides. The injection needle was placed in a micromanipulator. About 5 • 10-12-1 • 10-1 ~ ml of D N A solution was injected into each cell. Fig. 3 illustrates the injection procedure. As indicated in Table I 100)/o of the injected cells stained for T-antigen and 32~ for V-antigen 48 hr after injection (Fig. 4). We did not observe any difference in biological activity of D N A I dependent on the purification method.
TABLE I Percentage of T- and V-antigen positive cells 48 hr after injection of PV and different PV-DNA forms into mouse tissue culture cells Injected substances
T-antigen in ~o
V-antigen in
Full virus particle (1011 P F U m1-1) D N A I (20S) (0.3 mg ml-1) D N A II (16S) (0.3 mg ml- 1) Single stranded D N A (0.3 mg ml- ~) linear molecule (16 S) Cyclic molecule (18 S) Linear and cyclic molecules (16S and 18S) D N A fragments (12 S) (0.1 mg m l - 1)
100
42
100
35
34
12
32
0
32 32
0 6
8
0
Each measurement is based on a count of 3 • l0 s injected ceils. The cells were stained for T- and V-antigen 48 hr after injection. About 5 • 10-1e-10-11 ml of solution was injected into each cell. 237
The results obtained demonstrate that the infectivity of injected PV-DNA I is within the same range as that of the full virus particle using either microinjection or conventional absorption method. These data show clearly that the low efficiency of virus D N A I using the absorption method is due to the cell surface membrane,
Fig. 4. Secondary mouse culture cells injected with PV-DNA I and stained for T-antigen 48 hr after DNA injection. 1:400. 238
I
I
I
I
I
0.5 16s 18s -,4 LD
1.1
E3
0.3
0.1
I
I
I
5
10
I
15 Froction
I
I
20
25
Fig. 5. DNA II, obtained from CsC1-EtBr gradient, was centrifuged in alkaline sucrose gradient (5-20%, pH 12.5, 5.5 hr, 54000 rpm, 20~ We eluted the gradient from the top of the tube and measured optical density simultaneously at 254 nm.
2. The Infectivity of DNA II The DNA II is a double stranded cyclic molecule (16S) containing at least one single-strand break. This DNA was obtained by a mild pancreatic DNase digestion [13, 15] or after spontaneous conversion of the supercoiled molecule. DNA II was separated from DNA I by equilibrium centrifugation in neutral CsCI-EtBr (Fig. 1). Fractions 19-21 were pooled, dialysed and concentrated by vacuum evaporation. Previous experiments have shown that single chain breaks do not abolish the infectivity of viral D N A [13, 16]. 48 hr after injection of DNA II we found 34% of the cells staining for T- and 12% for V-antigen (Table I).
3. The Infectivity of Single Stranded DNA We separated the linear (16S) from the cyclic (18S) molecule by centrifugation of DNA II in an alkaline sucrose gradient (pH 12.5) (Fig. 5). 239
The 16 S and 18 S DNA were injected separately into mouse cells. We found that 16 S and 18 S DNA had the same capacity to induce T-antigen but did not cause V-antigen synthesis. These results prove that both forms of DNA contain early strand. Following injection of 16S and 18 S DNA simultaneously, we observed both T- and V-antigen formation. The reason for this difference in biological response cannot yet be explained. It may be possible that the amount of V-antigen, synthesized within 48 hr after microinjection of linear of cyclic single stranded molecules, is too small for detection by the immunfluorescence technique. Another explanation could be that there is no viral DNA replication, an avent thought to be necessary for V-antigen synthesis [17, 18]. V-antigen detection after simultaneous transfer of 16 S and 18 S DNA consequently should be due to formation of replicative DNA molecules. For example in SV 40, early and late gene transcription has been observed in cells without viral DNA replication [18] although in these V-antigen was not made. Further experiments are necessary to elucidate this question. 4. The Infectivity of Double Stranded DNA Fragments To prove whether PV-DNA fragments can also induce virus specific reaction following microinjection, DNA II was treated with pancreatic DNase (2.5 x 10-3 mg ml-1) for 15 min at 22~ The fragments obtained were centrifuged on a sucrose gradient 5-20~. Fractions containing multiply nicked virus DNA smaller than 12S were pooled and recentrifuged. DNA fragments obtained from the second gradient were injected into mouse cells. Analysis of the DNA agarose gel electrophoresis [19] revealed no molecules larger than 12S (data not shown). As shown in Table I the double stranded DNA fragments had the capacity to induce T-antigen but failed to induce V-antigen formation. Since pancreatic DNase does not cleave DNA at a specific site but randomly, it is not possible to assign T-antigen synthesis to a specific DNA fragment. Restriction enzymes from Hemophilus influenzae and H. parainfluenzae introduce double strand cleavage at specific sites into virus DNA [19, 20]. The physical order of these fragments is known for SV 40 DNA but not for Polyoma DNA. Using these SV 40 fragments for microinjection we may be able to get further information about the biological potency of these molecules. This type of experiments is in progress. ACKNOWLEDGEMENTS We are grateful to H. Koch and K. Fischer for technical assistance and Dr. S. Modak for careful reading. This work was supported by the Deutsche Forschungsgemeinschaft. REFERENCES 1. Di Mayorca, G., Eddy, B., Stewart, S., Hunter, W., Friend, C., and Bendich, A., Proc. Nat. Acad. Sci. U.S. 45, 1805 (1959). 2. Weil, R., Virology 14, 46 (1961). 3. Vogt, M. and Dulbecco, R., Proc. Nat. Acad. Sci. U.S. 46, 365 (1960). 4. Crawford, L., Dulbecco, R., Fried, M., Montagnier, L., and Stoker, M., Proc. Nat. Acad. Sci. U.S. 52 148 (1964). 5. Pagano, J. S. : in K. Habel and N. P. Salzmann, Fundamental Techniques in Virology, Academic Press, New York, London, 1969, p. 184. 6. Gruen, R., Graessmann, M., Graessmann, A., and Fogel, M., Virology, in press (1973). 7. Graessmann, A. and Graessmann, M., Hoppe-Seyler's Z. Physiol. Chem. 352, 527 (1971). 8. Hirt, B., J. Mol. Biol. 26, 365 (1967). 240
9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
Westphal, H., J. Mol. Biol. 50, 407 (1970). Vinograd, J., Lebowitz, J., and Watson, R., J. Mol. Biol. 33, 173 (1968). Bollum, F. J., J. Biol. Che. 234, 2733 (1959). Fogel, M., Gilden, R. and Defendi, V., Proc. Soc. Exp. Biol. Med. 124, 1047 (1967). Vinograd, J., Lebowitz, J., Radloff, R., Watson, R. and Laipis, P., Proc. Nat. Acad. Sci. U.S. 53, 1104 (1965). Crawford, L. V. and Waring, M. J., J. Mol. Biol. 25, 23 (1967). Dingman, C. W., Fisher, M. P., and Kakefuda, T., Biochemistry 11, 1242 (1972). Dulbecco, R. and Vogt, M., Proc. Nat. Acad. Sci. U.S. 50, 236 (1963). P6tursson, G. and Weil, R., Arch. gesamte Virusforsch. 24, 1 (1968). Khoury, G., Byrne, J. C., and Martin, M. A., Proc. Nat. Acad. Sci. U.S. 69, 1925 (1972). Sharp, A. P., Sugden, B., and Sambrook, J., Biochemistry 12, 3055 (1973). Danna, K. and Nathans, D., Proc. Nat. Acad. Sci. U.S. 69, 3097 (1972).
241
ACTIVITY OF DIFFERENT
FORMS OF
VIRUS DNA AND VIRAL DNA FRAGMENTS
M. GRAESSMANN, A. GRAESSMANN, E. HOFFMANN, J. NIEBEL, AND K. PILASKI
Institut fiir Molekularbiologie und Biochemie der Freien Universitdit Berlin, Berlin-W., F.R.G.
(Received 19 November, 1973) ABSTRACT. Mouse tissue culture cells were infected with different forms of Polyoma Virus (PV) DNA or virus DNA fragments by means of a microinjection technique and stained for PV-tumor (T) antigen and virus capsid (V) antigen 48 hr after injection. The efficiency of PV-DNA 1 (20 S) to induce T- and V-antigen was within the same range as the efficiency of the full virus particle. DNA II (16S) showed a reduced capacity for both T- and V-antigen induction. Single stranded DNA molecules (16 or 18S) and double stranded DNA fragments (12S) led to T-antigen but not V-antigen synthesis. Simultaneous transfer of 16S and 18S DNA revealed T- and V-antigen formation. I. INTRODUCTION The infectious agent of Polyoma Virus (PV) is the viral DNA [1]. Several groups observed cytopathic effects and plaque formation [2, 4] following exposure of tissue culture cells to PV-DNA. The infectivity of the virus DNA is very small. Substances like DEAE-D increase the infectivity of the viral DNA [5] but compared with the full virus particle the efficiency of infection is still lower than 1~o. It is likely that this phenomenon is caused by the cell surface membrane [5, 6]. Some years ago we developed a microinjection technique which enables us to inject nucleic acids directly into mammalian tissue culture cells [7]. This technique is very sensitive and reproducible for the assay of infectious viral DNA. In this study we describe the different capacities of PV-DNA I, DNA II, single stranded DNA molecules and double stranded DNA fragments to induce tumor (T)- and capsid (V)-antigen synthesis in mouse tissue culture cells, using the microinjection technique. II. MATERIAL AND METHODS PV-DNA was obtained from infected primary mouse kidney cells by the selective extraction procedure of Hirt [8], modified by Westphal [9]. (PV strain IL 11 was a gift of M. Fogel, Weizmann Institute.) The DNA preparation contained both cellular and viral molecules (DNA I and II). From these we subsequently separated cellular and viral DNA by velocity sedimentation (5-20~ sucrose, 1 M NaC1, 0.001 M Tris, 0.001 M EDTA, pH 7.4). Centrifugation was carried out in a Spinco SW 65 rotor, 54000 rpm, 2 hr at 20~ Fractions were collected in an Isco fractionator. DNA I was separated from DNA II by two different methods. First by CsCl-ethidium bromide [10] equilibrium centrifugation (1.56 g CsCI m1-1, 10-1 mg ethidium bromide ml -a in 0.02 M 233 Molecular Biology Reports 1 (1973) 233-241. All Rights Reserved Copyright 9 1973 by D. Reidel Publishing Company, Dordrecht-Holland
Tris, p H 8.0) in a Sprinco angle head Ti 50 rotor at 40000 rpm, 40 hr, 15 ~ The second method involved centrifugation over by alkaline sucrose gradients (5-20% sucrose, 0.5 M NaC1, 1 m M EDTA, p H 12.0) in a Spinco SW 65 rotor, 54000 rpm, 2 hr, 20~ Single stranded D N A molecules (16S and 18 S) were obtained by centrifugation of D N A II in alkaline sucrose (5-20%, 0.5 M NaCI, 1 m M EDTA, p H 12.5) for 5.5 hr, 54000 r p m at 20~ (Spinco SW 65 rotor). To obtain double stranded fragments D N A II was treated with pancreatic DNase I (Worthington). The incubation mixture (1 ml) contained 6 x 1 0 - 3 mg D N A , 2.5 m M MgC12 solution. After digestion for 15 min at 22 ~ the reaction was stopped by addition of SDS. Subsequent to phenol extraction, dialysis and concentration the fragments were centrifugated on a
I
I
.
t
!
I
I
II
3
1,5
2
1,0
x t3
E~ ol
1
0.5
|
5
10
I
!
15 20 Fractions
I
25
Fig. 1. PV-DNA, prepurified on neutral sucrose gradient (Spinco rotor SW 65, 20~ 2 hr, 54000 rpm), was centrifuged in CsC1-EtBr (spinco angle head rotor Ti 50, 40 hr, 40000 rpm). Tritium labeled PV-marker DNA was added. The gradient was fractionated from the bottom of the tube. We extracted the dye by isopropanol treatment and determined optical density (O O) and radioactivity (9 (3). DNA I banded at a density of 1.589 g cm-8 and DNA II at 1.556 gcm -3. 234
neutral sucrose gradient (conditions mentioned above). Fractions with D N A smaller than 12S were pooled and recentrifugated in neutral sucrose gradient 5-15%. Radioactivity was measured according to the m e t h o d o f Bollum [11 ]. All preparation steps were done under sterile conditions. Microinjection by means of glass capillaries was performed as described [7]. Secondary mouse culture cells grown on small glass slides (divided into numbered squares o f 1 m m 2) were used as recipient cells (Fig. 3). T-antigen was detected by indirect, and V-antigen by direct immunfluorescence technique [12] (antisera were kindly supplied by Dr. M. Fogel).
I
I
I
16-18s
I
5 3s
3
%
2,0
Z
t
E 2
0
,cz Ix) 0"1
- 1,0
5
10
15
20
25
Fractions Fig. 2. Mouse kidney ceils were Lqfected with PV (20 PFU cell-l) and labeled with 25 laCi 3H Thymidine per dish (90 man diameter) between 29-31 hr post infection (O O infected culture, V V mock-infected culture). DNA was obtained by the selective extraction method and centrifuged on alkaline sucrose gradient (Spinco SW 65 rotor, 2 hr, 54000 rpm, 20~ pH 12.0). The gradient was eluted from the top of the tube and simultaneously measured for optical density from infected culture at 254 nm 9 9 235
o~
Fig. 3. The microinjection technique. - (a) demonstrates the injection needle near the recipient cell - (b) the tip of the injection needle inside of the cell (c) the injection needle inside of the cell immediately after injection- (d) the recipient cell after transfer of virus DNA, the needle is again outside of the cell.
III. RESULTS A N D DISCUSSION
1. The Infectivity of DNA I The native form of PV-DNA (DNA I) is a double stranded, supercoiled molecule with a sedimentation coefficient of 20S [13]. The viral DNA can be obtained either by isolation from the full virus particle [2] or by the selective extraction method from virus infected cells [8]. DNA I was purified by two different methods. First by equilibrium centrifugation in a neutral CsC1 gradient containing ethidium bromide. In this condition DNA I bands at a higher density than D N A II [14] (Fig. 1). Alternatively we used centrifugation in alkaline sucrose gradient. At pH 12.0 D N A I is converted into a cyclic coiled D N A (53S) while DNA II is denatured into a linear (16S) and a circular (18S) single stranded molecule (Fig. 2). Following neutralization (pH 7.4) the cyclic coiled D N A cosediments with a marker supercoiled D N A I in neutral sucrose gradient [9]. To test the biological activity of D N A I, fractions 17-20 from the alkaline sucrose gradient (Fig. 2) were pooled, neutralized and dialyzed against diluted phosphate buffer. The DNA solution was then concentrated under vacuum to a final concentration of 0.3 mg ml- 1. The microinjection was carried out under a microscope at 400 fold magnification. The recipient cells were grown on glass slides. The injection needle was placed in a micromanipulator. About 5 • 10-12-1 • 10-1 ~ ml of D N A solution was injected into each cell. Fig. 3 illustrates the injection procedure. As indicated in Table I 100)/o of the injected cells stained for T-antigen and 32~ for V-antigen 48 hr after injection (Fig. 4). We did not observe any difference in biological activity of D N A I dependent on the purification method.
TABLE I Percentage of T- and V-antigen positive cells 48 hr after injection of PV and different PV-DNA forms into mouse tissue culture cells Injected substances
T-antigen in ~o
V-antigen in
Full virus particle (1011 P F U m1-1) D N A I (20S) (0.3 mg ml-1) D N A II (16S) (0.3 mg ml- 1) Single stranded D N A (0.3 mg ml- ~) linear molecule (16 S) Cyclic molecule (18 S) Linear and cyclic molecules (16S and 18S) D N A fragments (12 S) (0.1 mg m l - 1)
100
42
100
35
34
12
32
0
32 32
0 6
8
0
Each measurement is based on a count of 3 • l0 s injected ceils. The cells were stained for T- and V-antigen 48 hr after injection. About 5 • 10-1e-10-11 ml of solution was injected into each cell. 237
The results obtained demonstrate that the infectivity of injected PV-DNA I is within the same range as that of the full virus particle using either microinjection or conventional absorption method. These data show clearly that the low efficiency of virus D N A I using the absorption method is due to the cell surface membrane,
Fig. 4. Secondary mouse culture cells injected with PV-DNA I and stained for T-antigen 48 hr after DNA injection. 1:400. 238
I
I
I
I
I
0.5 16s 18s -,4 LD
1.1
E3
0.3
0.1
I
I
I
5
10
I
15 Froction
I
I
20
25
Fig. 5. DNA II, obtained from CsC1-EtBr gradient, was centrifuged in alkaline sucrose gradient (5-20%, pH 12.5, 5.5 hr, 54000 rpm, 20~ We eluted the gradient from the top of the tube and measured optical density simultaneously at 254 nm.
2. The Infectivity of DNA II The DNA II is a double stranded cyclic molecule (16S) containing at least one single-strand break. This DNA was obtained by a mild pancreatic DNase digestion [13, 15] or after spontaneous conversion of the supercoiled molecule. DNA II was separated from DNA I by equilibrium centrifugation in neutral CsCI-EtBr (Fig. 1). Fractions 19-21 were pooled, dialysed and concentrated by vacuum evaporation. Previous experiments have shown that single chain breaks do not abolish the infectivity of viral D N A [13, 16]. 48 hr after injection of DNA II we found 34% of the cells staining for T- and 12% for V-antigen (Table I).
3. The Infectivity of Single Stranded DNA We separated the linear (16S) from the cyclic (18S) molecule by centrifugation of DNA II in an alkaline sucrose gradient (pH 12.5) (Fig. 5). 239
The 16 S and 18 S DNA were injected separately into mouse cells. We found that 16 S and 18 S DNA had the same capacity to induce T-antigen but did not cause V-antigen synthesis. These results prove that both forms of DNA contain early strand. Following injection of 16S and 18 S DNA simultaneously, we observed both T- and V-antigen formation. The reason for this difference in biological response cannot yet be explained. It may be possible that the amount of V-antigen, synthesized within 48 hr after microinjection of linear of cyclic single stranded molecules, is too small for detection by the immunfluorescence technique. Another explanation could be that there is no viral DNA replication, an avent thought to be necessary for V-antigen synthesis [17, 18]. V-antigen detection after simultaneous transfer of 16 S and 18 S DNA consequently should be due to formation of replicative DNA molecules. For example in SV 40, early and late gene transcription has been observed in cells without viral DNA replication [18] although in these V-antigen was not made. Further experiments are necessary to elucidate this question. 4. The Infectivity of Double Stranded DNA Fragments To prove whether PV-DNA fragments can also induce virus specific reaction following microinjection, DNA II was treated with pancreatic DNase (2.5 x 10-3 mg ml-1) for 15 min at 22~ The fragments obtained were centrifuged on a sucrose gradient 5-20~. Fractions containing multiply nicked virus DNA smaller than 12S were pooled and recentrifuged. DNA fragments obtained from the second gradient were injected into mouse cells. Analysis of the DNA agarose gel electrophoresis [19] revealed no molecules larger than 12S (data not shown). As shown in Table I the double stranded DNA fragments had the capacity to induce T-antigen but failed to induce V-antigen formation. Since pancreatic DNase does not cleave DNA at a specific site but randomly, it is not possible to assign T-antigen synthesis to a specific DNA fragment. Restriction enzymes from Hemophilus influenzae and H. parainfluenzae introduce double strand cleavage at specific sites into virus DNA [19, 20]. The physical order of these fragments is known for SV 40 DNA but not for Polyoma DNA. Using these SV 40 fragments for microinjection we may be able to get further information about the biological potency of these molecules. This type of experiments is in progress. ACKNOWLEDGEMENTS We are grateful to H. Koch and K. Fischer for technical assistance and Dr. S. Modak for careful reading. This work was supported by the Deutsche Forschungsgemeinschaft. REFERENCES 1. Di Mayorca, G., Eddy, B., Stewart, S., Hunter, W., Friend, C., and Bendich, A., Proc. Nat. Acad. Sci. U.S. 45, 1805 (1959). 2. Weil, R., Virology 14, 46 (1961). 3. Vogt, M. and Dulbecco, R., Proc. Nat. Acad. Sci. U.S. 46, 365 (1960). 4. Crawford, L., Dulbecco, R., Fried, M., Montagnier, L., and Stoker, M., Proc. Nat. Acad. Sci. U.S. 52 148 (1964). 5. Pagano, J. S. : in K. Habel and N. P. Salzmann, Fundamental Techniques in Virology, Academic Press, New York, London, 1969, p. 184. 6. Gruen, R., Graessmann, M., Graessmann, A., and Fogel, M., Virology, in press (1973). 7. Graessmann, A. and Graessmann, M., Hoppe-Seyler's Z. Physiol. Chem. 352, 527 (1971). 8. Hirt, B., J. Mol. Biol. 26, 365 (1967). 240
9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
Westphal, H., J. Mol. Biol. 50, 407 (1970). Vinograd, J., Lebowitz, J., and Watson, R., J. Mol. Biol. 33, 173 (1968). Bollum, F. J., J. Biol. Che. 234, 2733 (1959). Fogel, M., Gilden, R. and Defendi, V., Proc. Soc. Exp. Biol. Med. 124, 1047 (1967). Vinograd, J., Lebowitz, J., Radloff, R., Watson, R. and Laipis, P., Proc. Nat. Acad. Sci. U.S. 53, 1104 (1965). Crawford, L. V. and Waring, M. J., J. Mol. Biol. 25, 23 (1967). Dingman, C. W., Fisher, M. P., and Kakefuda, T., Biochemistry 11, 1242 (1972). Dulbecco, R. and Vogt, M., Proc. Nat. Acad. Sci. U.S. 50, 236 (1963). P6tursson, G. and Weil, R., Arch. gesamte Virusforsch. 24, 1 (1968). Khoury, G., Byrne, J. C., and Martin, M. A., Proc. Nat. Acad. Sci. U.S. 69, 1925 (1972). Sharp, A. P., Sugden, B., and Sambrook, J., Biochemistry 12, 3055 (1973). Danna, K. and Nathans, D., Proc. Nat. Acad. Sci. U.S. 69, 3097 (1972).
241
