Review Article

Mutagenic Effects of DNA-Containing Oncogenic Viruses and Malignant Transformation of Mammalian Cells

Nicolai I. Shapiro, Marina I. Marshak, and Nina B. Varshaver

ABSTRACT: It was discovered in the 1970s that oncogenic viruses could induce gene mutations in mammalian cells. The phenomenon seems to be widespread: it was observed with all groups of DNA-containing viruses and some retroviruses. The mutagenic effects of the tested viruses at gene level are not locus specific. The viruses induce point mutations, including base sub- stitutions, as well as deletions and insertions. The mutagenic effect of SV40 is controlled by the activity of the early A gene, which encodes the T antigen. Presumably, the process of integration creates the possibility for occurrence of mutations early after infection. Mutagen- esis seems to be induced by an integrated virus, though to a much smaller extent. Virus- induced mutagenesis may be connected with an activation of the cell error-prone repair sys- tems. The sum total of the experimental data shows that virus-induced mutagenesis and transformation are interrelated: (A) viruses, like other carcinogenes, display mutagenic activ- ity; (B) viruses that are far removed from each other systematically, whose only similarity lay in being oncogenic and capable of integration, simultaneously showed the ability to induce gene mutations; (C) agents changing the rate of transformation also changed the rate of gene mutations: (D) The function of mutagenicity was mapped in the oncagene of SV40 (gene A); and the DNA of (E) mouse mammary carcinoma virus (MMTV) and avian leukosis virus (ALLV) induced tumors has been found to contain nucleotide sequences that transform 3T3NIH cells but do not carry any viral genetic information. Mutagenesis induced by onco- genic viruses may play a part in the multistage process of malignant transformation, though its contribution may be different in various specific cases and for different groups of viruses. Further studies of the uncommon mutagens, which viruses seem to be, may greatly increase our knowledge of the virus-cell relationship. An understanding of the extent of genetic dan- ger inherent in viruses and live viral vaccines is necessary for practical medicine.

INTRODUCTION

By the early 1970s, many findings confirmed the mutat ion theory of cancer [1, 2]. One of the most conclusive proofs consisted in the mutagenici ty of"practically all tested chemical and physical carcinogens [3]. These results indicated that mutagen- esis was involved in the process of malignant transformation and that there might be genes (protooncogenes) whose mutat ions caused malignancy. At present, the ex- istence of protooncogenes has been confirmed at the molecular level [4].

The situation is more complex in viral carcinogenesis, for the virus introduces its own genetic information into the cell. Unti l recently, mal ignant transformation

Address request for reprints to Dr. Nicolai I. Shapiro, Insti tute of Molecular Genetics, USSR A c a d e m y of Sciences, Moscow, USSR.

Received July 20, 1983; accepted December 14, 1983.

1 6 7

© 1984 by Elsevier Science PublishinR Co., Inc. Cancer Genetics and Cvtoeenetics 13. 167-179 f19841

168 N. 1. Shapiro et al.

of cells was bel ieved to depend entirely on the constant act ivi ty of the viral onco- gene integrated in the cell genome. However, facts were gradual ly accumula ted that contradic ted that concept. This par t icular ly concerns the DNA-containing viruses and some retroviruses. It was shown, thus, that in spite of the integrat ion of viruses in a vast majori ty of cells and the t ranslat ion of their oncogenes, mal ignant trans- formation is a rare event [5-9].

On the other hand, v i rus- t ransformed cells occas ional ly stop synthesiz ing viral antigens, lose viral DNA, but retain their mal ignancy [9, 10]. Hence, the constant presence of the viral genome in the cell is not always necessary for main ta in ing the t ransformed phenotype .

After a whi le it became quite clear that mal ignant t ransformat ion depends on some rare events in the cell 's own genome, apart from the integrat ion of the viral oncogene. Virus- induced muta t ions were supposed to be the events in question. Yet, the hypothet ica l involvement of mutat ions in viral carcinogenesis was devoid of exper imenta l evidence. It was not known whether or not oncogenic viruses in- duced gene muta t ions in host cells; in fact, no informat ion was available about v i rus- induced gene muta t ions in vertebrates.

Considering the par t ic ipa t ion of gene muta t ions as a working hypothes is on the nature of viral t ransformat ion of cells, it was necessary, first of all, to discover if oncogenic viruses caused gene muta t ions in mammal ian cells. That was the task facing our group at the Inst i tute of Molecular Genetics in Moscow and the Insti tute of Molecular Biology and Genetics in Kiev. This article is a review of the investi- gations carried out in our laboratories. The data obtained by other researchers on the mutagenic effect of oncogenic viruses also are cited and discussed. If oncogenic viruses, indeed, were shown to induce muta t ions in some genes of the host cell, this would mean that they have certain features in common wi th other carcinogenic factors. Therefore, new exper imenta l approaches might be used to s tudy the process of malignizat ion.

In order to clarify the role of viruses as possible inducers of gene mutat ions, it was essential to not only s tudy mal ignant growth, but also to reveal new aspects of the v i rus -ce l l re la t ionship, which is of utmost impor tance for unders tand ing the genetic danger of these ubiqui tous envi ronmenta l agents, especia l ly in the context of the wide use of live vaccines in medicine.

IS MUTAGENIC ACTIVITY A UNIVERSAL PROPERTY OF ONCOGEN~C VIRUSES?

We chose SV40 as the puta t ive mutagen. SV40 is one of the best s tudied oncogenic viruses. Its genome contains only a few genes. SV40 DNA has been ful ly sequenced, and the specific features of its t ranscr ip t ion and t ranslat ion are wel l known [11].

For the s tudy of mutagenesis , we chose cell cul tures where the virus had li t t le or no cytopathic effect. Because nothing was yet known as to whether or not the virus could induce mutat ions, mutagenesis was invest igated in two kinds of cell systems: human cell l ines suppor t ing weak mul t ip l ica t ion of SV40 and prac t ica l ly nonpermiss ive Chinese hamster cells.

It seemed plausible that oncogenic viruses might exercise their mutagenic effect at the t ime of integrat ion in the host cell genome, when the recombina t ion of viral and cell DNA occurs and the latter 's integri ty is dis turbed. Mutat ions are known to occur in the modera te phage-bacter ium system dur ing phage integrat ion [12]. The integrat ion of SV40 into ~he cell genome is comple ted 30-48 hrs after infect ion [13]. Therefore, to begin with, mutagenesis was invest igated early after infection.

The first results were obtained in 1973 [14], and in 1975, more deta i led data were publ i shed [15]. Table 1 shows that SV40 induces muta t ions of resistance to pur ine

DNA-Containing Oncogenic Viruses 169

Table 1 SV40-Induced mutat ions of resistance to purine analogues in human and Chinese hamster cells

Frequency of mutants ( × 10-5)b

Exp. Purine Expression Cells no. analogue ° time (days} Control S V 4 0 Difference pC

Man 1 8AG 4 0.8 7.4 + 6.6 <0.05 2 8AG 4 11.2 20.6 +9.4 <0.01 3 8AG 1 0.9 3.6 +2.7 <0.01

4 0.3 1.0 +0.7 ~0.05

Chinese 1 6MP 1 4.0 144.0 + 140.0 <0.001 hamster 4 3.6 19.7 + 16.1 <0.01

2 6MP 1 18.0 43.0 + 25.0 <0.01 4 18.0 20,1 +2.1 >0.05

3 6MP 1 1.2 7.4 + 6.2 <0.01

°8AG, 8-azaguanine; 6MP, 6-mercaptopurine. bin all the tables, the frequency of mutants is shown with correction for survival.

CThe significance of the difference was estimated according to the Fisher's exact test [14].

analogues 8-azaguanine (8AG) and 6-mercaptopurine (6MP}, both in semipermis- sive human cells and in nonpermissive Chinese hamster cells. The mutat ions were induced on the first day after infection, and their number usual ly dropped by the fourth day. No SV40-induced mutants were found on the eighth day.

Thus, the very first oncogenic virus studied proved to induce gene mutations; these results were later confirmed [16]. It remained unclear whether or not SV40 was the only virus to cause mutagenesis. To answer that question, the highly on- cogenic bovine adenovirus 3 (BAV3), which is systematically remote from SV40, was investigated [17, 18].

The experiments have shown that BAV3 is also mutagenic, with the level of induct ion increasing with the mult ipl ic i ty of infection (Table 2). Mutations to 6MP resistance are induced at a very low titer, about 1 plaque-forming uni t (PFU) per cell; hence, one or very few viral particles are sufficient to cause mutat ion in the cell (the BAV3 strain was not likely to contain defective particles, as it had been purified by repeated cloning). At 80 PFU/cell, the process tends to be saturated, probably because either only a l imited number of viral particles can penetrate into the cell or cells are killed by an overload of the virus.

Similar results were obtained for human adenovirus 5 (HAV5) [9]. An increased rate of SAG and aminopter ine (AP) resistant colonies in the first days after the in- fection of mur ine cells was reported. Later, the mutagenic effect of one more aden° ovirus (Ad2) was detected [19, 20]. The virus induced back and forward mutat ions in the hypoxanthine phosphoribosyltransferase (HPRT) locus. In 1982, it was shown that Herpes simplex virus type 1 (HSV) also displayed mutagenic activity [21]. Thus, studies of representatives of all three groups of DNA-containing onco- genic viruses (papova, adeno, and herpes) have revealed the induct ion of mutat ions after infection. Recently, the mutagenic effect of Moloney mur ine leukemia virus was detected in mouse somatic and germ cells [22-25]. It can be concluded, there- fore, that the ability to induce gene mutat ions is a property shared by all groups of DNA-containing oncogenic viruses and by retroviruses, as well.

Reliable experimental evidence of virus- induced gene mutat ions was slow to ac- cumulate. This was probably due to the methodologic difficulties involved. Above

170 N.I . Shapiro et al.

Table 2 Dependence of the rate of induced muta t ions to 6MP resistance on the mul t ip l ic i ty of infection of BAV3"

Frequency of BAV3 induced mutants (minus control)

Exp. no. BAV3 PFU/cell × 10 ~ p

1 0.8 25.89 <0.001 8.0 76.65 <0.001

86.0 80.42 "<0.001 2 0.8 4.43 <0.01

8.0 7.30 <0.001 3 0.8 1.62 >0.05

8.0 2.72 <0.05 80.0 4.85 <o.ool

°The expression time is 2 days 1171.

all, the successful detect ion of mutagenesis largely depends on the correct choice of a v i rus -ce l l system where the infect ion is effective but the virus exercises l i t t le or no cytopathic effects. Thus, the Ad2 virus, which was not pathogenic for the cul tured cells concerned, was recent ly shown to induce 8AG-resistant mutat ions, whereas the highly active, cell~killing virus, Ad12, induced none [20]. One way of making the s tudy of mutagenesis poss ible in such cases consists of part ia l inacti- vat ion of the virus to reduce its pathogenici ty. This method was used wi th some success in a s tudy of the effects p roduced by the herpes virus on human cells [21]. However, this procedure seems to be barred in the case of Ad12. The invest igators bel ieve that the apparent lack of effect is due to the high frequency of lethal chro- mosome damage wherever viral part icles have penetra ted the cell and induced po lyp lo idy . Besides, if an HPRT muta t ion occurs when the po lyp lo id cell is a l ready formed, the muta t ion may go undetec ted in the surviving cells [20].

Of course, in a s tudy of viral mutagenesis , the op t imum condi t ions of mutan t detect ion must be strictly observed [26]; specifically, the r isk of metabol ic cooper- at ion be tween mutant and nonmutan t cells must be taken into account. It is no less impor tant to find the op t imum express ion t ime and to remember that, after a long enough post infect ion period, the pic ture may be distorted by popu la t ion processes that i ndependen t ly change the frequency of mutants in control and exper imenta l s i tuat ions, as we observed weeks and months after the infect ion of h u m a n cells wi th SV40.

CHARACTERIZATION OF VIRUSES AS MUTAGENS

Viral Mutagenesis Is Not Locus Specific

The fact that mutagenesis was detected in the very first r andomly chosen HPRT locus suggested that the virus effects were not locus specific in semipermiss ive and nonpermiss ive cells. Fur ther research confirmed this assumption. SV40 was shown to induce muta t ions at another r andomly chosen locus responsible for revers ions from glutamine t s -auxothrophy to pro to t rophy [27, 28]. These data are presented in Table 3.

Mutagenesis also occurs at other loci, e.g., those control l ing AP and colchic ine resistance [29]. Thus, SV40-induced mutagenesis was detected in all the genes stud- ied and in different cell systems with respect to both forward and back mutat ions.

DNA-Containing Oncogenic Viruses 171

Table 3 SV40 induct ion of g l u - t s -~ glu+tr in Chinese hamster cells °

Frequency of mutants ( x 10 -s) Expression time (days) Control SV40 Difference P

2 0.19 8.40 + 8.21 <0.001 1.37 3.42 + 2.05 <0.001

3 0.52 4.55 +4.03 <0.001 9.4 13.5 +4.1 <0.01

°The table shows the results of typical experiments [27].

Al l this indicates that viral mutagenesis at the gene level is nonspecific, though the poss ibi l i ty that the cell genome has as yet undiscovered "ho t" spots cannot be ex- cluded.

Viruses Induce Different Types of Genetic Changes, Including Point Mutations

It was natural to a t tempt to determine the molecular types of muta t ions induced by the virus. Al though only the resistance to pur ine analogues result ing from a de- crease or complete loss of HPRT activi ty was s tudied [30, 31], it could be surmized that this character, in pr inciple , might be control led by different types of mutat ions, inc luding poin t mutat ions as well as insert ions and delet ions in the HPRT gene. It became possible to unders tand if the v i rus- induced heredi tary changes inc luded point muta t ions after we invest igated back mutat ions usual ly arising at the level of ind iv idua l nucleot ides.

SV4O was capable of causing back muta t ions to glutamine prototrophy. The char- acter proved to be stable. The frequency of mutat ion was about 1.5 orders of mag- n i tude lower than the rate of forward mutat ions at the HPRT locus (Tables 1 and 3) [27, 28]. A relat ively low rate is characterist ic of back mutations. The tested gluta- mine-auxotrophic strain had a temperature-sensi t ive phenotype. This p rov ided ad- di t ional informat ion about the molecular nature of the v i rus- induced mutat ions. The temperature sensi t ivi ty of the ini t ial strain indica ted that the gene in quest ion has only a slight defect, as an active enzyme is synthesized at 33°C. The muta t ion is most probably a DNA base substi tution, for muta t ions of other molecular types do not normal ly lead to a ts phenotype. In that case, a reversion to the wi ld type may also result from a base substi tution. Of course, one cannot exclude the possi- bi l i ty that part of the reversions are due ta suppressor muta t ions in other genes, but, according to the data for microorganisms, suppressor muta t ions are usua l ly also base substi tut ions. Thus, SV40 proved to be capable of causing reversions from a temperature-sensi t ive to a temperature-resis tant phenotype by dint of point mu- tations.

SV40 and HAV2 were also shown to induce reversions at the HPRT locus, where forward muta t ions had been observed [19, 29]. The Ad2 revertants induced dis- p layed a b iochemica l ly al tered enzyme [19].

The s tudy of reversions revealed point mutat ions, though it seemed l ikely that the virus could also induce other types of mutations. An at tempt to find more dras- tic changes of the genetic material was made in the s tudy of a col lect ion of pur ine analogue-resis tant mutants. The resistance may arise as a result of point mutat ions or after a part ial or complete loss of the HPRT gene [30]. Biochemical analysis of four 6MP-resistant clones induced by SV40 and 14 BAV3-induced clones carried out at our laboratories (unpubl ished data) has shown the cells of most clones to contain reduced but detectable HPRT activity. A number of clones d i sp layed a

172 N. 1. Shapiro et al.

changed affinity of HPRT to the substrate [31[. Only a few of the BAV3-induced mutant clones showed no HPRT activi ty at all.

The invest igat ion of four 8AG-resistant mutant clones induced by SV40 also re- vealed res idual enzyme activity; in two more cases, HPRT was inactive, but its presence could be detected immunolog ica l ly [32, 33]. Thus, even though the selec- tive system a l lowed different k inds of mutat ions, all 10 tested mutants induced by SV40 and most of the BAV3-induced mutants (10 of 14) kept the genetic informa- tion for HPRT synthesis . Most probably, in all these cases, point muta t ions had occurred at the HPRT locus. The four BAV3-induced clones that did not show any active enzyme might contain an inact ive HPRT protein whose detect ion requires further research. Alternat ively, the lack of enzyme may be due to dele t ions in the HPRT gene.

Thus, the muta t ions induced by viruses are often point mutat ions, probably base subst i tut ions. This is rather unexpected , as moderate phages (a well s tudied analo- gous prokaryot ic system) most ly induce insert ions and delet ions [12]. The occur- rence of that type of muta t ions was detected after infect ion by the Moloney mur ine leukemia virus (M-MuLV) [22, 23]. The base subst i tut ions in infected cells are most readi ly expla ined by errors in t roduced by the repair systems, which "hea l" the vi- rus- induced damage in DNA. The increased rate of v i rus- induced forward and back muta t ions in the presence of agents causing DNA breaks and unschedu led DNA synthesis may be regarded as indirect evidence of the par t ic ipat ion of cell repair and/or recombina t ion systems in viral mutagenesis (for details see above) [18, 29, 34, 35].

The fact that viruses can induce different types of muta t ions was i ndependen t ly confirmed by the results of molecular research. Various DNA rearrangements and dele t ions were detected near the integrat ion sites of SV40 and the po lyoma virus [36-39]. In addi t ion, during the evolut ion of all l ines t ransformed by these viruses, the viral sequences were t ransposed to new sites [38, 39]. This could induce new insert ion mutat ions.

Along with Gene Mutations There Are Chromosome Aberrations

A large series of s tudies carried out in the 1960s showed near ly all the known viruses to cause chromosome damage [40, 41]. Sometimes, it was conjectured that the chromosome aberrat ions might indicate a paral le l muta t ion process at gene level, but no exper imenta l evidence of that para l le l ism was furnished [40, 41]. To discover if v i rus - induced mutagenesis occurs both at gene and chromosome levels, the effect of SV40 [15] and BAV3 [18] were s tudied in Chinese hamster cells that are nonpermiss ive for both viruses. In all cases, there was a paral le l y ie ld of gene muta t ions and chromosome aberrations. The t ime course of chromosome aberra- tions is highly s imilar for both viruses: the ma x imum of aberrant metaphases occurs on days 1-2 after infection; the control level is regained on days 3-4. It is quite l ikely that the appearance of chromosome aberrat ions induced by oncogenic viruses does indicate the occurrence of gene muta t ions in many cases. Yet, even then, the chromosome aberrat ions can only be regarded as indicat ions that gene muta t ions are induced and can reveal nothing of their quanti tat ive pattern.

When Do Mutations Take Place?

In most cases, v i rus- induced muta t ions were observed in the first few generat ions after infection, after which, the number of induced mutants decreased ]9, 15, 17]. The surge of mutagenesis co inc ided in t ime with the integration process of viral genomes in the cell. Nevertheless, it seems that a weak mutagenic effect may be

DNA-Containing Oncogenic Viruses 17 3

caused by an integrated virus. An increased mutation rate in transformed cells was demonstrated by using the fluctuation test and by introducing additional copies of viral genomes into cell DNA [42]. The first experiments gave indications of a pos- sible mutagenic activity of the polyoma virus. Significant results were obtained, however, only after integration of 3-5 additional SV40 genome equivalents per cell. The maximum effect in these double transformants amounted to 7 x 1 0 - 7 induced mutations per cell per generation.

One can try to compare the mutation rates at the HPRT locus for early postinfec- tion stages [15] and for integrated viruses [42]. In the experiments with SV40 on the first day after infection, the mean frequency of induced mutations exceeded 10 -4 (see Table 1). Hence, an integrated SV40 has a much weaker mutagenic effect (by 2-3 orders of magnitude) than the virus in the process of integration. Presumably, an integrated virus increases the mutation rate by continuously synthesizing the T antigen, which is a mutagenic factor (see last section). In that case, the A gene of SV40 (the g6ne encoding the T antigen) may be regarded as a new mutator gene. However, in spite of the constant activity of the A gene in transformed cells, its product only makes for a very weak mutagenic effect. It seems that a more effective mutagenesis requires some other processes that occur during integration.

VIRUS-INDUCED MUTAGENESIS AND MALIGNANT TRANSFORMATION

The study of the mutagenic action of oncogenic viruses has shown them to be no exception, in that they are mutagens like all other carcinogens. This seemed to support the idea that mutations made a contribution to the viral transformation of cells. Experiments were carried out to test the hypothesis.

First of all, it was necessary to compare the mutagenic effects of viruses with the mutagenesis induced by other carcinogens under similar experimental conditions. We made this comparison for reversions to glutamine prototrophy in hamster cells [28]. SV40 was found to induce a significant but smaller number of mutations than the supermutagen N-methyl-N'-nitro-N-nitrosoguanidine (MNNG): on an average, 6 x 10 -6 compared to 4.5 × 10 -5 (with a spontaneous rate of 1.9 x 10 -~ mutations per cell per generation). A similar comparison for the induction of reversions at the HPRT locus by the herpes simplex virus and 4-nitroquinoline-l-oxide has revealed roughly equal mutation rates [21]. It should be noted that quantitative comparisons are not quite reliable, as it is very difficult to equalize the dose of the virus, which has nearly no effect on survival, and that of the chemical mutagen, which produces a considerable lethal effect.

The study of such typical transformation characters as anchorage independence [9] and growth at low serum concentrations [43] by means of the fluctuation test showed both characters to arise spontaneously at typical rates of gene mutations. Treatment witil chemical mutagens (MNNG, EMS) and viruses (Ad5 and SV40) in- creased the yield of transformed cells, with the induction values and kinetics being similar. A study of Ad5-induced anchorage-independent clones showed that they contained no viral information but were capable of inducing tumors [9]. These re- suits support the hypothesis that these properties arise through mutations.

Another approach is based on the assumption that if mutations play a part in viral transformation, there should be a parallelism of virus-induced mutagenesis and transformation, not only under normal infection conditions, but also in various modified set-ups where the rate of either is increased or reduced. Of course, the parallelism will not prove the identity of the two processes, but it will evidence their interrelation.

The transformation rate may be increased by treatment with agents causing DNA breaks [44, 45]. We asked whether such an agent, i.e., 5-bromodeoxyuridine (BrdU),

1 7 4 N.I . Shapiro et al.

would also increase the rate of SV40-induced mntagenesis [34, 35, 46]. The incor- porat ion of BrdU in both DNA strands caused the rate of SV40-induced glu = mu- tations to rise several t imes over the expected value for an addi t ive effect of the two agents. A s tudy of the combined action of BrdU and BAV3 y ie lded the same result (see Figs. 1 and 2) [18]. A synergist ic effect was found for both viruses wi th respect to chromosome aberrat ions [18, 35]. The pat tern observed in a s tudy of mal ignant t ransformation pract ica l ly repeated itself at gene and chromosome levels. The rate of SV40-induced muta t ions was later shown to be increased by other agents that enhance transformation, inc luding ul t raviolet {UV) light [29]. Thus, when the trans- forming activi ty of a virus is increased, its mutagenic act ivi ty changes in the same direction.

We have also tr ied to map the SV40 gene responsible for the mutagenic activi ty [46-48] and to compare its locat ion with that of the gene control l ing mal ignant transformation. If the two locat ions coincided, it would mean that mutagenesis and t ransformation depend on the product of the same gene and are therefore closely interrelated. Because mutagenesis was observed in SV40-nonpermiss ive hamster cells, where the early genes are t ranscr ibed but the late ones are inactive, only the early A region, encoding the T and t antigens, could claim the role of the "muta tor" gene. Of course, one could not exclude that mutagenesis might be merely due to the in t rus ion of foreign DNA into the cell.

Because the t antigen does not seem to be necessary for t ransformat ion (mutants wi th a dele t ion of the unique t antigen region of DNA were able to induce mal ignant t ransformation [11]), it was natural to concentrate on the A gene of SV40, which encodes the T antigen. From a col lect ion of SV40 mutants, we chose the ts A239 mutant , which synthesizes a temperature-sensi t ive T antigen. All tsA SV40 mutants t ransform mammal ian cells at 33°C, but fail to do so at 40°C [111. This made it poss ible to switch on (at 33°C) and off (at 40°C} the " t ransforming" A gene of SV40 whi le observing its role in the induct ion of mutagenesis.

Table 4 shows that, at the permiss ive temperature (33°C), the mutant virus in-

Figure 1

7

. - 6

5 o

X 4

~ 3 E

r =

i . _

Mutagenesis induced by SV40 and BUdR (g lu t s --+ glu+tr).

T r//.~///.I

Control SV40 BUdR

^ . . ^ J r

5V40+BU, dR

DNA-Containing Oncoge:,lic Viruses 175

80

70

6o

0

~. 50

~ 40

E

- 30

g 20 .-i

" 1 0

Figure 2

Control. BRV3 BUdR

XX Y~ 'h , v vvv \

< × " / ' - / ~z

~XXXY KXXXY KX>O<b< x . / ' ~ .X .X X ~XX×× X 'Y YX ' x ×XXX :X

>~XXX>

w x x I . . ,

X"~" XX Y K~XX)~ <XXX)~ <XXM~ <XXXY <XXX~ 4 "X~OX)< <XXvX . ,X

N BFIV3+BUdR

M u t a g e n e s i s i n d u c e d b y B A V 3 a n d B U d R (HPRT ÷ --~ H P R T - ) .

creased the yield of 6MP-resistant colonies after 2 days in each of 6 experiments, with the excess over control being significant. The induction was larger with a larger mutant background. In the same experiments, the situation was quite differ- ent if the cells were incubated at 40°C. Insignificant deviations both ways from the control mutant frequency were observed. Similar results were obtained for chro- mosome aberrations. It should be noted that we used a temperature-resistant clone of Chinese hamster cells whose viability was not affected at 40°C. Thus, SV40 mu- tagenicity is controlled by the same A gene as the initiation and maintenance of transformation: the ts A239 mutation simultaneously inactivates these functions at nonpermissive temperature.

The results of all experiments patently demonstrate that the mutagenic and transforming effects of DNA-containing oncogenic viruses are closely interrelated.

Table 4 Effect of incubation temperature on mutagenesis induced by tsA239 mutant of SV40

F r e q u e n c y of 6MP res i s tant m u t a n t s in t s A 2 3 9 in fec ted ce l l s ( m i n u s control ) × 10 -5

Exp. no. 33°C p 40°C ° p

1 + 1 2 . 4 < 0 . 0 1 - 2 . 1 > 0 . 0 5

2 + 9.6 < 0 . 0 1 + 3.6 > 0 . 0 5

3 + 3.7 < 0 . 0 5 + 2.5 > 0 . 0 5 4 + 13.9 < 0 . 0 1 - 1.4 > 0 . 0 5 5 + 3.7 < 0 . 0 1 + 0 . 3 > 0 . 0 5 6 + 24.9 < 0 . 0 1 - 0.6 > 0 . 0 5

°The expression time at both temperatures is 2 days. The mean number of cell generations during expres- sion time is 1.7 at 33°C and 1.9 at 40°C [47].

176 N.I . Shapiro et al.

Moreover, both phenomena proved to depend on one and the same A gene of SV40. It seems l ikely that the processes of t ransformation and mutagenesis have some links in common. These links may be connected with viral genome integration.

The exper iments involving the tsA mutant of SV40 somewhat clarify the mech- anisms of viral mutagenesis . At 40°C, there is no mutagenesis in ts A239-infected cells, even though viral DNA is t ranslated (the T antigen is detected). Hence, at 40°C, the cell accumulates foreign macromolecules that are not themselves involved in mutagenesis . Obviously, what matters is the information carried by the A gene of SV40--hence , the funct ioning of its product , the T antigen, which is inhibi ted at 40°C. However, in what way the T antigen can par t ic ipate in inducing muta t ions is as yet obscure. An independen t conclus ion about the role of viral genetic informa- t ion in induc ing mutagenesis was reached by other researchers [49] who used a different method: the frequency of gene muta t ions was compared in cells t reated with different DNAs (SV40, calf, phage). Only SV40 DNA proved to be mutagenic. According to our data, BAV3 DNA also d i sp layed mutagenic activi ty [50], whereas normal hamster cell DNA was ineffective as a mutagen (unpubl i shed data). How- ever, the viral gene responsible for the induct ion of mutat ions could not be ident i- fied.

The fact the viral act ion is not locus-specific in somatic and germ cells indicates that viruses can induce muta t ions in any genes, inc luding the cells ' protoonco- genes. In this connect ion, we should consider the studies where M-MuLV-induced mutagenesis was invest igated in RSV-transformed rat cells. This system may be regarded as a model wi th the single integrated RSV genome playing the part of the cell 's t ransforming gene [22, 23]. The src oncogene of RSV is known to be homolo- gous to the cel lular src oncogene [51]. A super infect ion of this cell l ine wi th M- MuLV was shown to induce morphologica l reversions that were control led by dif- ferent types of mutat ions: a complete or part ial loss of the RSV genome, muta t ions in the SRC gene, and insert ions of M-MuLV DNA sequences into the RSV genome. Thus, direct exper iments have demonst ra ted that the oncogene may be the target for the mutagenic act ion of a virus. More evidence of the involvement of muta t ions in viral t ransformat ion has been furnished by another exper imenta l procedure. Ma- l ignant phenotyp ic characters were transferred via DNA from cells t ransformed by the avian leukosis virus (ALLV) or the mouse mammary carcinoma virus (MMTV) to rec ip ient cells [52-54]. The t ransforming DNA did not contain any viral se- quence. This suggests that ALLV- and MMTV-induced carcinogenesis is based on the act ivat ion of cel lular t ransforming genes through mutat ions or rearrangements causing their anomalous expression.

Of course, the mechanisms of viral carcinogenesis cannot be reduced to muta- genesis alone. The contr ibut ion of mutagenesis to t ransformation may be different for different viruses. However, the available data demonstra te that muta t ions are involved in the mult is tage process of v i rus- induced transformation.

REFERENCES

1. Shapiro NI (1967): Mutagenesis and carcinogenesis. Usp Sovr Biol USSR 63:163-168. 2. Sandberg AA (1980): The Chromosomes in Human Cancer and Leukemia. Elsevier North

Holland, New York. 3. Miller EC, Miller JA (1971): The mutagenicity of chemical carcinogens: Correlations, prob-

lems and interpretations. In: Chemical Mutagens. Plenum Press, New York-London, pp 83-119.

4. Cooper GM (1982}: Cellular transforming genes. Science 217:801-806. 5. Risser R, Pollack R (1974): A nonselective analysis of SV40 transformation of mouse 3T3

cells. Virology 59:477-489.

DNA-Conta in ing O n c o g e n i c Viruses 17 7

6. Stiles CD, Desmond W, Sato G, Saier MH (1975): Failure of human cells transformed by Simian virus 40 to form tumors in athymic nude mice. Proc Natl Acad Sci USA 72:4971-4975.

7. Gotoh S, Gelb L, Schlessinger D (1979): SV40 transformed human diploid cells that remain transformed throughout their limited lifespan. J Gen Virol 42:409-414.

8. Teich N, Wyke J, Mak T, Bernstein A, Hardy W (1982): Pathogenesis of retrovirus induced disease. In: Molecular Biology of Tumor Viruses. RNA Tumor Viruses, R Weiss, N Teich, H Varmus, J Coffin, eds. Cold Spring Harbor Lab., New York, pp 792-794.

9. Bellett AJD, Younghusband HB (1979): Spontaneous, mutagen-induced and adenovirus- induced anchorage independent tumorigenic variants of mouse cells. J Cell Physiol 101:33-48.

10. Paraskeva C, Brown KW, Dunn AR, Gallimore PH (1982): Adenovirus type 12 transformed rat embryo brain and rat liver epithelial cell lines: Adenovirus type 12 genome content and viral protein expression. J Virol 44:759-764.

11. Tooze J, ed (1980): Molecular Biology of Tumor Viruses. Part 2. DNA Tumor Viruses. Cold Spring Harbor Lab., New York.

12. Boram W, Abelson J (1971): Bacteriophage Mu integration: On the mechanism of Mu- induced mutations. J Mol Biol 62:171-178.

13. Hirai K, Lehman J, Defendi V (1971): Integration of Simian virus 40 DNA into the DNA of primary infected Chinese hamster cells. J Virol 8:708-715.

14. Marshak MI, Varshaver NB, Shapiro NI (1973): Induction of gene mutations in cultured mammalian cells by the Simian virus 40. Genetika USSR 9:138-141.

15. Marshak MI, Varshaver NB, Shapiro NI (1975): Induction of gene mutations and chromo- somal aberrations by Simian virus 40 in cultured mammalian ceils. Mutat Res 30:383- 396.

16. Theile M, Scherneck S, Geissler E (1976): Mutagenesis by Simian virus 40. I. Detection of mutations in Chinese hamster cell lines using different resistance markers. Mutat Res 37:111-124.

17. Lukash LL, Buzhievskaya TI, Varshaver NB, Shapiro NI (1979): Induction of gene muta- tions by adenovirus in mammalian cells. Dokl Acad Nauk USSR 245:970-973.

18. Lukash LL, Buzhievskaya TI, Varshaver NB, Shapiro NI (1981): Oncogenic adenovirus as mutagen for Chinese hamster cells in vitro. Somat Cell Genet 7:133-146.

19. Marengo C, Mbikay M, Weber J, Thirion JP (1981): Adenovirus-induced mutations at the hypoxanthine phosphoribosyltransferase locus of Chinese hamster ceils. J Virol 38:184- 190.

20. Paraskeva C, Roberts C, Biggs P, Gallimore PH (1983): Human adenovirus type 2 but not adenovirus type 12 is mutagenic at the hypoxanthine phosphoribosyltransferase locus of cloned rat liver epithelial cells. J Virol 46:131-136.

21. Schlehofer JR, zur Hausen H (1982): Induction of mutations within the host cell genome by partially inactivated Herpes simplex virus type 1. Virology 122:471-475.

22. Opperman H, Levinson AD, Varmus HE (1981): The structure and protein kinase activity of proteins encoded by nonconditional mutants and back mutants in the src gene of avian sarcoma virus. Virology 108:47-70.

23. Varmus HE, Quintrell N, Ortiz S (1981): Retroviruses as mutagens: Insertion and excision of a nontransforming provirus after expression of a resident transforming provirus. Cell 25:23-36.

24. Jenkins N, Copeland N, Taylor B, Lee B (1981): Dilute (d) coat colour mutation of DBA/2J mice is associated with the site of integration of an ecotropic MuLV genome. Nature 293:370-374.

25. Jaenisch R, Harbers K, Schnieke A, L6hler J, Chumakov Y, J~hner D, Grottkopp D, Hoff- man E (1983): Germline integration of Moloney murine leukemia virus at the Mov 13 locus leads to recessive lethal mutation and early embryonic death. Cell 32:210-216.

26. Shapiro NI, Varshaver NB (1975): Mutagenesis in cultured mammalian cells. Meth Cell Biol 10:209-234.

27. Varshaver NB, Marshak MI, Shapiro NI (1976): Reversions to glutamine independence induced by the oncogenic virus SV40 in mammalian cells. Dokl Acad Nauk USSR 230:716-718.

28. Varshaver NB, Marshak MI, Luss EV, Gorbunova LV, Shapiro NI (1977): Spontaneous,

178 N. 1. Shap i ro et al.

chemical and viral mutagenesis in temperature-sensitive glutamine-requiring Chinese hamster cells. Murat Res 43:263-278.

29. Theile M, Strauss M, Luebbe L, Scherneck S, Krause H, Geissler E (1980): SV40-induced somatic mutations: Possible relevance to viral transformation. Cold Spring Harbor Syrup Quant Biol 44:377-381.

30. Caskey CT, Kruh G (1979): The hypoxanthine guanine phosphoribosyltransferase locus. Cell 16:1-10.

31. Volkova LV, Luss EV, Moiseenko EV, Pankova NV, Petrova ON, Shapiro NI (1981): Study of Chinese hamster cells mutant for the hypoxanthine-guanine phosphoribosyl transferase locus. 1. The collection of mutants and experiments on intragenic complementation. Genetika USSR 17:297-307.

32. Theile M, Strauss M (1977): Mutagenesis by Simian virus 40 II. Changes in substrate affin- ities in mutant hypoxanthine-guanine phosphoribosyl transferase enzymes at different pH values. Murat Res 45:111-123.

33. Strauss M, Theile M, Eckert R, Geissler E (1978): Detection of antigenically active mutant HGPRT after mutagenesis with Simian virus 40. Mutat Res 51:297-300,

34. Varshaver NB, Gorbunova LV, Lukash LL, Shapiro N1 (1978): 5-bromodeoxyuridine and mutagenic effect of the oncogenic virus SV40, Dokl Acad Nauk USSR 242:707-710.

35. Varshaver NB, Marshak MI, Gorbunova LV, Lukash LL, Shapiro NI (1980): The synergistic effects of SV40 and BUdR on induction of gene mutations and chromosomal aberrations in Chinese hamster cells. Mutat Res 70:351-364.

36. Botchan M, Stringer J, Mitchison T, Sambrook I (1980): Integration and excision of SV40 DNA from the chromosome of a transformed cell. Cell 20:143-152.

37. Stringer JR (1982): DNA sequence homology and chromosomal deletion at a site of SV40 DNA integration. Nature 296:363-366.

38. Hiscott JB, Murphy D, Defendi V (1981): instability of integrated viral DNA in mouse cells transformed by SV40. Proc Natl Acad Sci USA 78:1736-1740.

39. Della Valle G, Fenton RG, Basilico C (1981): Polyoma large T antigen regulates the inte- gration of viral DNA sequences into the genome of transformed cells. Cell 23:347-355.

40. Nichols WW (1963): Relationship of viruses, chromosomes and carcinogenesis. Hereditas 50:53-80.

41. Stich HF, Yohn DS (1970): Viruses and chromosomes. Prog Med Virol 12:78-127.

42. Goldberg S, Defendi V (1979): Increased mutation rates in doubly viral transformed Chinese hamster cells. Somat Cell Genet 5:887-895.

43. Varshaver NB, Marshak MI, Shapiro NI (1983): The mutational origin of serum indepen- dence in Chinese hamster cells in vitro. Int J Cancer 31:471-475.

44. Casto BC, Pieczynski WJ, Janosko N, DiPaolo JA (1976): Significance of treatment interval and DNA repair in the enhancement of viral transformation by chemical carcinogens and mutagens. Chem Biol Interact 13:105-125.

45. Hirai K, Defendi V, Diamond L (1974): Enhancement of Simian virus 40 transformation and integration by 4-nitroquinoline-1 oxide. Cancer Res 34:3497-3500,

46. Shapiro NI (1978): Viral mutagenesis in cultured mammalian cells in vitro. XIV Interna- tional Congress on Genetics, Plenary Session Symposia, Moscow, pp 78-79 (abstr)

47. Gorbunova LV, Varshaver NB, Shapiro NI (1980): Induction of gene mutations in mam- malian cells by a temperature-sensitive mutant of SV40. Dokl Acad Nauk USSR 255:209- 211.

48. Gorbunova LV, Varshaver NB, Marshak MI, Shapiro NI (1982): The role of the transform- ing A gene of SV40 in the mutagenic activity of the virus, Mol Gen Genet 187:473-476.

49. Theile M, Scherneck S, Geissler E (1980): DNA of Simian virus 40 mutates Chinese ham- ster cells. Arch Virol 55:293-309.

50. Charikova EV, Lukash LL, Zalmanson ES (1983): The mutagenic effect of a highly onco- genic bovine adenovirus type 3 DNA. Biol Nauk USSR 4:86-88.

51. Parker RC, Varmus HE, Bishop JM (1981): Cellular homologue {c-src) of the transforming gene of Rous sarcoma virus: Isolation, mapping and transcriptional analysis of c-src and flanking regions. Proc Natl Acad Sci USA 78:5842-5845.

DNA-Containing Oncogenic Viruses 179

52. Cooper GM, Neiman PE (1981): Two distinct candidate transforming genes of lymphoid leukosis virus-induced neoplasms. Nature 292:857-858.

53. Cooper GM, Neiman PE (1980): Transforming genes of neoplasms induced by avian lym- phoid leukosis viruses. Nature 287:656-659.

54. Lane MA, Sainten A, Cooper GM (1981): Activation of related transforming genes in mouse and human mammary carcinomas. Proc Natl Acad Sci USA 78:5185-5189.