233

Journal of Virological Methods, 33 (1991) 233-251 0 1991 Elsevier Science Publishers B.V. / 0168-8510/91/$03.50 ADONIS 0168851091002575

VIRMET 01192

Review Article

Antineoplasic activity of parvoviruses Jean Rommelaere1,2 and Jan J. Cornelis2 1Department of Molecular Biology, Universite’ Libre de Bruxelles, Rhode-Saint-Gent%e. Belgium and 2Unith d’Oncologie MolPculaire, INSERM U186/CNRS Institut Pasteur de Lille, France

(Accepted

URA 1160.

18 March 1991)

Summary

The family of Parvoviridae is composed of small, nuclear-replicating viruses that are without envelope and contain an essentially single-stranded, linear DNA genome. Certain parvoviruses proved to have the remarkable capacity to prevent the formation of spontaneous as well as virally- and chemicallyinduced tumors in laboratory animals. Established tumor cells serve as targets for the antineoplasic activity of parvoviruses, since the growth of preformed cancer cells transplanted in recipient animals can also be inhibited by these viruses. Furthermore, epidemiological studies in humans have revealed a correlation between serological evidence of parvoviral infection and a lower incidence of certain cancers. The parvoviral life-cycle appears to depend on cellular factors that are expressed as a function of proliferation and differentiation. This subordination may account for the oncotropism of parvoviruses in vivo and for the specificity of their interactions with (pre-)neoplasic cells under appropriate culture conditions. Thus, certain parvoviruses were found to preferentially lyse initiated or stably transformed cells in vitro, as a possible result of the stimulation of the production and/or activity of cytotoxic viral proteins. Parvoviruses can also have a cytostatic effect and cause the reversion of transformation traits, parallel to the down-modulation of the expression of defined genes, in particular oncogenes. Such direct disturbance of neoplasic cells or their precursors may participate in the oncosuppressive activity of parvoviruses, although indirect viral effects mediated by host defense mechanisms also deserve to be considered.

Correspondence to:

Jean Rommelaere, Laboratory of Biophysics and Radiobiology, Dept. of Molecular Biology, Universitk Libre de Bruxelles, 67 Rue des chevaux, B 1640 Rhode Saint Genbse, Belgium.

234

Altogether, these properties suggest the possible use of parvoviruses as probes to investigate the process of malignant transformation. Further research is needed to determine whether infections with parvoviruses may find a place in the prevention and treatment of human cancers, Parvovirus/Tumor

suppression; Neoplasic transformation

Since their discovery in the 1960s parvoviruses have been associated with cancer. Initially, these agents were envisaged as oncogenic viruses because most parvoviral strains isolated at that time were obtained from tumors or tumor transplants passaged in laboratory animals (Siegl, 1984). Subsequent work did or experimental infections with not support this view. Thus, natural parvoviruses failed to correlate with an increased risk of cancer (Toolan, 1967) and were rather found to be associated, in a number of instances, with a suppression of oncogenesis (see below). Arguments to be discussed in this review support the view that tumor-derived or in vitro transformed cells often provide a favorable milieu for parvovirus replication, suggesting that the preferential localization of these viruses in neoplasic lesions (oncotropism) may reflect an opportunistic relationship instead of a causal one. Parvoviridae are not the only virus family that comprises members capable of exerting an oncosuppressive activity. Thus, a slowing down or inhibition of the formation of primary tumors and metastases can also be achieved by certain myxoviruses (influenza A virus, Newcastle disease virus, Sendai virus) and vaccinia virus (Schirrmacher et al., 1986; Ito et al., 1990). Multiple mechanisms may account for such a virus surveillance against cancer, including production of cytotoxic or immunogenic viral polypeptides in infected tumor cells, induction of cytokines capable of interfering with the growth of neoplasic cells (e.g. interferons, tumor necrosis factors) and/or stimulation of host immune responses (Schirrmacher et al., 1986). The efficiency and range of the oncosuppressive activity of different viruses will depend on their respective abilities to trigger one and/or the other of these processes, and hence on the biology of virus-host interactions. Parvoviruses are especially interesting in this respect. Indeed, the lytic lifecycle of parvoviruses proved to be controlled by cellular factors that are expressed as a function of proliferation and differentiation (Cotmore and Tattersall, 1987). Therefore, it is not surprising that neoplasic transformation of various mammalian, including human cells, is often accompanied by a change in their sensitivity to parvoviral attack. Such privileged interactions of parvoviruses with transformed cells are the main topic of this paper, since they may be exploited not only to suppress neoplasic growth but also to reveal cellular changes associated with malignant transformation. Reference is also

235

made to recent comprehensive reviews dealing with the oncosuppressive activity of parvoviruses (Cukor et al., 1984; Rommelaere and Tattersall, 1990) and related aspects of the replication, gene regulation and pathogenicity of these viruses (Berns and Bohenzky, 1987; Berns and Labow, 1987; Cotmore and Tattersall, 1987; Siegl and Tratschin, 1987; Berns, 1990).

Parvovirus organization and biology The Parvoviridae comprise some thirty viruses that are parasites of certain insects, birds and many species of mammals, including man. The members of this family are closely related by their overall structure, and consist of spheroid non enveloped particles (approximately 20 nm in diameter) that contain a linear single-stranded DNA molecule (about 5000 nucleotides long) flanked by terminal hairpin sequences (Cotmore and Tattersall, 1987; Berns and Bohenzky, 1987; Bando et al., 1990). The low degree of genetic complexity of parvoviruses exacerbates the subordination of their replication to exogenous helper functions. This assistance can be provided by helper viruses and/or host cells. Two groups of parvoviruses were distinguished by the fact that productive infection by the former (adeno-associated viruses, AAV) usually requires cell co-infection with a helper virus (in particular adeno- and herpesviruses), while the latter (autonomous parvoviruses) replicate by themselves in appropriate host cells (Siegl and Tratschin, 1987). This classification is not absolute, however, since cells can be induced to produce AAV in the absence of helper viruses (Yakobson et al., 1987, 1989; Yalkinoglu et al., 1988); conversely, the inability of certain cells to replicate autonomous parvoviruses can be remedied, at least in part, by helper viruses (Ledinko et al., 1969; Faisst et al., 1989; Fox et al., 1990). As stated above, a major interest aroused by parvovirus replication is due to its requirement for cellular factors, the availability of which depends on proliferation and differentiation (Cotmore and Tattersall, 1987). Parvoviruses can thus be used as probes to study how these processes are controlled. The parvoviral genorne comprises only a limited number of overlapping genes, the organization of which is illustrated in Fig. 1 for prototype autonomous (minute virus of mice, MVM) and helper-dependent (AAV-2, a human virus) parvoviruses (Cotmore and Tattersall, 1987; Berns and Bohenzky, 1987; Berns, 1990). The genetic maps are basically similar and can be subdivided in left- and right-hand parts that encode regulatory and capsid proteins, respectively. Emphasis will be laid in this ‘paper on the regulatory proteins since they appear to be involved both in the control of parvoviral DNA replication and expression (Berns, 1990) and in virus cytopathogenicity (see below). The various regulatory functions are accomplished by several proteins that are denoted NS (MVM) or Rep (AAV) and are translated from differentially spliced mRNAs programmed by a single (P4 in MVM) or two (P4 and P19 in AAV-2) promoters. The uttermost right-hand

236 0

t

I

20

I

I

10

I

60 T

,

80 1

,

100 map units ,

,

A&V-2 P5p

P&0,0) Proteins lkd)

mRNAs (kb) 4.2 3.9

Pt+

Rep 178)

3

Rep (68)

3.6

Rep

(521

3.3

Rep

(40)

2.3

MVM

L.8

-

NSI

(83)

3.3

1

NS2

(25)

3.0

Fig. 1. Coding strategies of two representative parvovirus genomes. Transcripts are aligned beneath a tinear diagram of the viral genome. Introns are indicated by interruptions in the RNAs. Open reading frames thought to be translated are shown by boxes below each transcript. RNA sizes (in kilobases [kb]) and protein molecular masses (in kilodaitons [kd]) are given to the left and right, respectively. Symbol: flag, sites of transcription initiation from indicated promoters. Abbreviations: P, promoter; Rep/NS, nonstructural proteins; VP, virion structural proteins.

promoter (P38 in MVM and P40 in AAV-2) governs the expression of proteins that are designated as VP and are the major constituents of the capsid, although they may also play a regulating role (Antonietti et al., 1988; Gardiner and Tattersall, 1988). The successive stages of the parvoviral life-cycle generally lead to cell lysis, sometimes even when the virus cycle is not completed (Cornelis et al., 1988a; Guetta et al., 1990). However, the fact that parvoviral replication requires very specific assistance from the cell limits the potential target-cell population. Many cells resist parvovirus infection either for lack of surface receptors capable of adsorbing the parent particles, or - most frequently - because the intracellular milieu is inadequate for sustaining the viral cycle (Tattersall and Gardiner, 1990). These restrictions make parvoviral pathogenesis quite specific (Siegl, 1984). For instance, human cells can be infected by the autonomous parvovirus B19 which selectively replicates in, and destroys specific erythroid

237

precursor cells and causes aplastic crises in certain circumstances (Anderson, 1990). Other parvoviruses which frequently (AAV) or occasionally (H-l, an autonomous parvovirus) infect human hosts, are associated with no known pathology. This article mostly concerns these latter viruses.

In~bi~on of ~~cinog~n~is by p~o~rus

i~~tion

The parvovirus-related anticancer effect was first discovered ~by Helen Toolan, who reported as early as 1967 that hamsters persistently infected with H viruses developed 5 to 25 fewer spontaneous tumors than control animals. Later studies showed that this parvoviral surveillance can also affect various types of tumors induced experimentally in rodents and chicken by means of oncogenic viruses or a chemical carcinogen (Table 1). This oncosuppression could result from parvoviral interference with the induction of malignant transfo~ation and/or with the survival or proliferation of tumor cells. It will be considered below that parvoviruses have in vitro effects compatible with both types of mechanisms. The second possibility, suppression of preestablished transformed cells, has been demonstrated in vivo by a series of experiments showing that parvoviruses are able to prevent, in recipient animals, the development of tumors from transplanted transformed cells (Table 2). The degree of protection conferred by parvoviruses depends both on the inoculation program and on the relative quantities of virus and neoplasic cells. It is worth noting that cells transformed by helper viruses (i.e. adeno- and herpesviruses) appear to be preferential, although not exclusive targets for the an~neoplasic activity of AAV, whereas the autonomous pa~ovi~ses seem to TABLE 1 Inhibition

of tumorigenesis

Tumorigenic

by parvoviruse$’

agent

Host

Antineoplasic

effectb

Autonomous parvovirus

AAV

+

ReferencesC

Tumor viruses Adenoviruses 12 and 31 Simian virus 40 Moloney leukemia virus Marek’s disease virus

Hamster Hamster Rat Chicken

+ + + NR

RR +

14 3s 6 7

Chemical carcinogen 7,12-Dimethylbenzanthracene

Hamster

+

NR

8

a Parvovirus inoculated into newborn animals at time of, or prior to carcinogenic treatment. b + , ‘At least 50% inhibition of tumorigenicity; -, no significant effect on tumor rate; NR, not reported. ’ 1, Kirschstein et al., 1968; 2, Toolan and Ledinko, 1968; 3, Mayor et al., 1973; 4, de la Maza and Carter, 1981; 5, Siegl, 1984; 6, Bergs, 1969; 7, Paul and Monreal, 1989, Proc. EM30 Workshop Molecular Biology of Parvoviruses, Ha’ate Hach~isha, Israel, p. 45; 8, Toolan et al., 1982.

238 TABLE 2 Suppression Transplanted

of transplanted

tumors by parvoviruseSa

cells

Host

Origin

Transforming

Hamster Hamster Hamster Hamster

Adenoviruses 5112 Herpes simplex virus 2 Simian virus 40 9, I O-Dimethyl~n~nthracene Dimethylnitrosamine Ascites cells Activated c-Ha rus Transmissible venereal sarcoma Simian virus 40

Hamster Mouse Mouse Dog Human

agent

Antineoplasic

effectb

Referencesc

Autonomous parvoviruses

AAY

Hamster Hamster Hamster Hamster

NR NR NR NR

+ + -

Hamster Mouse Nude mouse Dog

NR NR +

;R

:, 7

Nude mouse

+

NR

8

-

192 ?f

I’

334

NR

a Parvovirus inoculated into tumor cells prior to transplantation or into animals at time of or after transplantation. b + , At least 50% inhibition of tumorigenicity; -, no significant effect on tumor rate; NR, not reported. ’ 1, Ostrove et al., 1981; 2, de la Maza and Carter, 1981; 3, Cukor et al., 1975; 4, Blacklow et al., 1978; 5, Guetta et al., 1986; 6, Katz and Carter, 1986; 7, Yang, 1987; 8, Dupressoir et al., 1989.

have a broader spectrum of oncosuppressive action. Interestingly, recent work carried out in our laboratory shows that parvovirus H-l exerts preventive, and to a lesser extent, curative antitumor action against transplants of human mammary epithelial cells in nude mice, in the absence of any detectable side effects (Dupressoir et al., 1989). At present, parvoviral oncosuppression is still a ‘black box’ phenomenon at the first stage of analysis. A possible, albeit extreme view, would be that parvoviruses do not come into contact with (pre-)neoplasic cells themselves but disturb them indirectly via physiological effecters (the immune system, interferons, tumor necrosis factors, interleukines, transforming growth factors) whose production or activity they modulate. These possibilities have not yet been fully explored, but it seems that parvoviruses are inefficient both as inducers and as targets of interferons (Wiedbrauk et al., 1986a,b). In this context, it is interesting to note that Newcastle disease virus, a myxovirus endowed with oncosuppressive potential, is capable of inducing tumor necrosis factors and augmenting their cytotoxicity (Lorence et al., 1988; Rood et al,, 1990). On the other hand, certain experimental results support the view that parvoviruses interact directly with at least a fraction of the transformed cells. For instance, parvoviral imprints as well as viral proteins have been detected in regressing tumors (Dupressoir et al., 1989 and unpublished data). Furthermore, it appeared in one of the systems studied that the use of a parvoviral strain capable of replication in the tumor cells is required (Guetta et al., 1986). It thus seems likely that the ability of parvoviruses to interact directly with neoplasic

239

cells cont~b~tes to the in~bition of oncogenesis, even though it remains dificult to ascertain to what extent this component contributes to the global level of oncosuppression.

The interference of parvovirus infwtion with cell metabolism In the absence of information on other possible anticancer actions of parvoviruses, we will emphasize the disturbances that these viruses can make in infected transformed cells. It should be stated, however, that these changes were demonstrated so far in cell cultures and that their occurrence in tumor cells in viva, hence their possible cont~bution to oncosuppression, remains unconfirmed. Direct effects on cell metabolism are attributed to the terminal sequences of the viral DNA as well as to the regulatory NS/Rep proteins. Analysis of defective particles suggests that the non-coding hairpin ends of the parvoviral genome, which contain replication and transcription control sequences (Cotmore and Tattersall, 1987; Berns, 1990), cont~bute to the inhibition of a response frequently associated with malignant transformation: DNA amplification (Bantel-Schaal and zur Hausen, 1988). It may be that these parvoviral genomic termini interact with cellular DNA-amfilification factors and thus prevent induction of this process by tumor-initiating agents. On the other hand, the parvoviral regulatory proteins exert pleiotropic functions which might contribute to the suppression of oncogenic transfo~ation and destruction of neoplasic cells or both. The NS/Rep proteins were indeed found to inhibit gene expression driven by heterologous promoters of which some control viral oncogenes (Labow et al., 1487; Rhode and Richard, 1987; Brandenburger et al., 1990; Antoni et al., 1990). Furthermore, the intracellular accumulation of parvoviral regulatory proteins may be toxic, as indicated by the disturbance and eventual death of neoplasic human cells induced to express NS proteins from an integrated hormone-dependent clone of corresponding genes (CailletFauquet et al., 1990). Altogether, these interfering effects may account for the reported involvement of NS/Rep proteins in the inhibition of various cellular responses such as DNA amplification (Heilbronn et al., 1990), stable DNA transfo~atio~ with selectable marker genes (Labow et al,, 1987; Rhode, 1987; Ozawa et al., 1988; Brandenburger et al., 1990) and in vitro oncogenic transformation (Hermonat, 1989). Lastly, parvoviruses may have a cytostatic action due to as yet unidentified constituents of the virions (Winocour et al., 1988). In addition to these direct effects, parvoviruses might influence the fate of infected tumor cells in other, more devious ways. Certain enveloped viruses, produced by budding at the plasma membrane, increase the immunogenic character of infected tumor cells and thus favor their rejection. At first glance, parvoviruses are not very effective in this respect because they are assembled in the nucleus and are thought to be liberated when the host cell lyses (Cotmore

240

and Tattersall, 1987). That parvoviral antigenic determinants might be displayed on the surface of infected cells cannot be excluded, however, and would account for certain observations suggesting that immunological phenomena are one component of the oncosuppressive response. This possibility is supported by the fact that parvoviral infection stimulates potential immunological indicators. For instance, metastases are arrested (Blacklow et al., 1978) and vaccination against implanted tumor cells can be achieved by combined pretreatment of animals with these cells and parvoviruses (Guetta et al., 1986). It thus may be that certain aspects of parvoviral oncosuppression, particularly late events, may involve immunological mediation caused by the presence of viral antigens on a fraction of the tumor cells.

Inhibitory effects of parvoviruseson in vitro cell transformation The hypothesis that transformed cells may be direct targets of parvoviral action is supported by a series of results obtained with cell cultures. These experiments indicate that parvoviruses inhibit induction and/or maintenan~ of a transformed phenotype in vitro, under conditions where only direct interactions between viruses and transformed cells or their precursors are possible. As for oncosuppression, parvoviral inhibition of in vitro transformaTABLE 3 Parvovirus-induced Gel1 origin

inhibition Parvovirus”

of in vitro cell transformation Transformation

Referenced

Agentb

Criterion

Inhibitionc + + + _ + + + + + +

Hamster embryo Hamster embryo

AAV-I AAV-2

Ad12, SiAd Ad5

Mouse embryo

AAV-2

Ha-rus

Mouse breast Mouse embryo Mouse embryo Newborn human kidney Human breast epithelium

AAV-2 MVMp MVMp H-l

BPV-1 sv40 Ha-rus, SV40 sv40

Foci formation Anchorage-independence High saturation Foci formation High saturation & cloning Foci formation Anchorage-independence Anchorage-independence Anchorage-independence

H-l

sv40

Anchorage-inde~ndence

1 i: 3 : :, 7 8

’ Inoculation of parvoviral particles or DNA before, at time of, or after ceil treatment with transforming agent. b Ad5/12, adenovirus type S/12; SiAd, simian adenovirus; Ha-ras, activated human ras gene from the EJ bladder carcinoma; BPV-1, bovine papilloma virus type 1; SV40, simian virus 40. ’ f , At least 50% inhibition of transformation; -, no significant effect on transformation. d 1, Casto and Goodheart, 1972; 2, Ostrove et al., 1981; 3, Katz and Carter, 1986; 4, Hermonat,1989; 5, Mousset and Rommelaere, 1982; 6, Mousset et al., 1986; 7, Su et al., 1988; 8, Dupressoir et al., 1989.

241

tion concerns a variety of human or murine cells and viral or cellular oncogenes (Table 3). Although most of these studies were performed with permanent cell lines, it was recently reported that the proliferation of non-established cultures of freshly isolated human tumor cells is severely impaired in the presence of AAV-2 or AAV-5 viruses (Bantel-Schaal, 1990). Among the above-mentioned viral effecters of cell disturbances, the Rep protein family is involved in the ability of virus AA%‘-2 to prevent in vitro transfo~ation of mouse cells by bovine papillomavirus (Hermonat, 1989). These observations lead to the question: how selective is parvoviral attack with respect to transformed cells? It seems likely that the reduced transformation frequency is due, at least in part, to non-specific effects. It has been demonstrated that AAV particles have a cytostatic action which may affect normal cells at least as strongly as some transformed cells ~inocour et al., 1988; Bantel-Schaal, 1990). This interference appears after inoculation with large quantities of virus and apparently does not require viral replication. On the other hand, parvoviruses can also act selectively against (pre-)transformed cells. For instance, virus MVMp efficiently inhibits the oncogenic transformation of mouse cells which, in their normal state, exhibit little or no sensiti~ty to the inocula (Mousset and Rommelaere, 1982). This specific component of parvoviral anti-transforming action is particularly interesting: it spares the normal untransformed cells and opens the perspective that these viruses might be used to detect cellular changes associated with neoplasic transformation. Two types of targeted alterations of transformed cells have been identified in the presence of parvoviruses and will be discussed in the following paragraphs. Parvovirus-caused suppression of the transformed phenotype A first type of specific interaction between transformed cells and parvoviruses, initially described for AAV, leads to reversion of transfo~ation characteristics. AAV infection does not give rise to a viral progeny unless assisted by helper viruses or unless the cell cycle is perturbed by special treatments (Yalkinoglu et al., 1988; Yakobson et al., 1987, 1989). However, various transformed cells can to low level express AAV genes in the absence of any appreciable replication of viral DNA. This parvoviral expression is apparently not cytotoxic but is associated with the extinction of various traits of the transformed phenotype. Thus, when adenovirus-transformed hamster cells are inoculated with AAV virus, there is a spectacular drop in the intracellular concentration of the 58-kDa tumor antigen encoded by region Elb of the transforming virus (Ostrove et al., 1981). The molecular mechanism of this reduction has not been elucidated but could involve one of the parvoviral Rep proteins, given its (their) above-mentioned role in suppressing in vitro cell transformation and in regulating gene expression. It is also worth mentioning that AAV viruses might have a similar inhibitory effect on the expression of cellular oncogenes (on transcription in particular). This possibility stems from

242

results for c-myc and c-ras, reported recently by Schlehofer and Faisst, and Hermonat, respectively [Proc. EMBO Workshop (1989) Molecular Biology of Parvoviruses, pp. 43 and 461. In conclusion, suppression of cellular or viral genes responsible for maintaining the transformed phenotype is worthy of consideration among the possible mechanisms of parvoviruses’ anti-transfo~ing action. Exacerbation of cellular stresses by AAV viruses

A correlation between cell transformation and the magnitude of parvoviruses’ cytopathic effect has been revealed in two distinct situations. The first concerns the interaction of AAV viruses with cells subjected to various types of stress, induced notably by genotoxic agents that initiate tumorigenesis. AAV viruses have long been regarded as defective agents unable to replicate without the assistance of a helper virus (Berns and Bohenzky, 1987). It has appeared recently that cells having undergone the above-mentioned stresses become transiently able to support at least a partial AAV virus life-cycle in the absence of a helper virus (Schlehofer et al., 1983,1986; Yakobson et al, 1987,1989; Bantel-Schaal and zur Hausen, 1988; Yalkinoglu et al., 1988). This autonomous replication of AAV viruses apparently interferes with the metabolism of initiated cells, probably via the parvoviral DNA’s terminal structures and/or the Rep proteins (see previous sections). As a result, infection with AAV-5 inhibits the mutagenicity of chemical and viral initiators (Schlehofer and Heilbronn, 1990). In the presence of AAV, initiated cells also exhibit a reduced amplification rate for chromosomal DNA sequences related (Schlehofer et al., 1983, 1986; Bantel-Schaal and zur Hausen, 1988; Schmitt et al., 1989) or not (Yalkinoglu et al., 1990) to integrated tumor viruses - and may even die (Heilbronn et al., 1984). In other words, AAV viruses synergically reinforce the intrinsic toxicity of initiating agents and are thus liable to decrease the risk that cells committed to transformation will appear or subsist. Furthermore, since expression of a transformed phenotype seems to predispose cells to induction of permissivity towards AAV viruses (Nahreini and Srivastava, 1989; Yakobson and Winocour, Proc. EMBO Workshop (1989) Molecular Biology of Parvoviruses, p. 5), infection by these agents could also compromise preferentially the survival of pre-established neoplasic cells exposed to some form of physiological stress. Interestingly, treatments having a cytocidal action in synergy with AAV viruses include heat shock and Tumor Necrosis Factor (Walz and Schlehofer, ibid., p. 37).

243

Preferential lysis of transformed cells by the autonomous parvoviruses H-l and MVM A second case of interrelation between cell transformation and parvovirus cytocidal capacity is exemplified by a series of recent studies concerning the autonomously replicating MVMp and H-l viruses. An analysis of the interaction of these parvoviruses with cells in culture has revealed that many human and murine lines of in vitro transformed cells from fibroblastic and epithelial origin are much more sensitive to viral infection than the normal cells from which they derive (Table 4). The increased severity of the cytopathic effect of parvoviruses correlates with in vitro transformation by various viral and cellular oncogenes [such as the genes coding for large (simian virus 40) and middle (polyomavirus) T antigens, v-myc, v-src, Ha-ras] and by physical and chemical carcinogens. Consequently, the H-l virus could be used as a selective agent to isolate rare ‘normalized’ variants present in a culture of transformed human kidney cells (Su et al., 1988). It seems that transformation causes the sensitization to parvoviral infection, because these two parameters vary concomitantly in cells conditionally transformed by a temperature-sensitive oncoprotein (Salomt et al., 1989). Changes in the overall duration of the cell cycle and immortalization do not appear to be major determinants of cell sensitization to parvoviral attack. Thus, pre- and post-crisis transformed human tibroblasts both proved to be more susceptible to H-l virus infection than normal parental cells proliferating at a similar rate (Chen et al., 1986; Cornelis et al., 1988a; Rhode and Paradiso, 1989). It should also be noted that the reinforcement of the cytopathic action of parvoviruses H-l and MVMp by transformation depends on the nature of both the oncogene and the cell. For instance, established ‘normal’ rat libroblasts (FR3T3) are sensitized to the killing effect of MVMp as a result of their transformation by a series of oncogenes among which Ha-ras, but not by the bovine papillomavirus type 1 (Salome et al., 1990). On the other hand, transformation of permanent lines of ‘normal’ human keratinocytes (HaCaT) and rat kidney cells (NRK) by Ha-rus is not associated with an increase in their susceptibility to parvovirus-induced killing (Chen et al., 1989; van Hille et al., 1989). Thus, although hyper-sensitivity to parvoviral infection is often acquired as a result of transformation, it is not required for expression of the transformed phenotype. Interestingly, abortive transformation induced by infection with simian virus 40 or treatment with the tumor promoter 12-0tetradecanoyl phorbol-13-acetate appears to be insufficient to sensitize hostcells to MVMp or H-l virus (Mousset and Rommelaere, 1988; Faisst et al., 1989). Similar to in vitro transformed cells, many human cell lines established from various tumors (tibrosarcoma, epidermoid and mammary carcinomas) are more liable to be killed by MVMp or H-l virus than the corresponding normal cells (Table 4). In addition, H-l virus also appears to be cytotoxic for in vitro established or tumor-derived B and T lymphoid cell lines of which some are

TABLE 4

+ + +

f + f

&X -b

+

i&t

Parvoviral gene expressiona .^_-_ NR +

-

12,13 9,t2 9,12-14

7-1 I 9 811

l-3 1,336

--

References’

a SV40, simian virus 40; PyV, polyoma virus; pyMT, polyoma middle T gene; AEV, avian erythroblastosis virus; 4NQO,4-nitroquinoline-l-oxide; Ad5-Ela, adenovirus 5 Ela gene; HPV-16, human papilloma virus type 16; EBV, Epstein Barr virus; NR, not reported. b Increase (+) or insignificant change (-) of mentioned parameters in transfarmed versus established parental (murine) or normal diploid (human) cells. ’ 1, Mousset et al., 1986; 2, Salome et al., 1989; 3, Salome et al., 1990; 4, Guetta et al., 1990; 5, van Hille et al., 1989; 6, Cornelis et al., 198813;7, Chen et al., 1986, 8, Cornelis et al., 1988a; 9, Faisst et al., 1989; 10, Rhode and Paradiso, 1989; 11, Cornelis et al., 1990; 12, Chen et al., 1989; 13, Dupressoir et al., 1989; 14, Toolan and Ledinko, 1965.

+

H-f H-f H-f

sv40 HPV- 16, Ha-m Ep~de~oid and breast carcinomas

-I-

f or + or -

-I-

Human epithehal celia

propertiesb Parvoviral DNA amplification

I:

C

Parvovirus-induced cell killing -^___1_. i i

Transformation-enhanced

H-l or MVMp

H-l, MVMp H-l H-l, MVMp

SV40,4NQO, AdS-Ela tibrosareoma

Human tibroblasts

y-rays

MVMp MVMp

v-rrlyc, V-SK, sv40 PyV, pyMT, Ha-m,

Murine fibroblasts

AEV

Parvoviruses

In vitro transforming agent or tumar of origina

semi tization of cells to parvoviruses

Cell origin

Transfo~ation-~sociated

245

able to sustain a productive infection (Bass and Hetrick, 1978; Faisst et al., 1989). Thus, certain parvoviruses can induce an oncolytic process in cell cultures. Currently, it is not known how much the establishment and possible in vitro drifting of cell lines contribute to their vulnerability to parvoviruses, as the analysis of freshly isolated tumor-derived cell cultures is only at its initial stages (Wanmin et al., 1989). Yet, the afore-mentioned observations justify taking oncolysis into consideration as a potential component of parvoviral oncosuppression. The mechanism by which transformation exacerbates parvoviral lytic action has not been fully elucidated. This phenomenon does not result, in any of the studied cases, from a stimulation of the adsorption and internalization of the inoculated vii-ions. Nor does it require production of infectious particles (Cornelis et al., 1988a; Guetta et al., 1990). In contrast, the parvovirus-sensitive transformed cell derivatives can be distinguished from resistant parental cells by their increased ability to express (and in particular to transcribe) or replicate (or both) the parvoviral genome (Table 4 and unpublished data). The effect of this stimulation of the viral life-cycle is to enrich the infected cells in nonstructural viral proteins whose toxicity (see previous sections) may contribute to the oncolytic process. Consistently, a correlation is observed, in certain systems, between the respective capacities of related cells for producing NS proteins and their susceptibilities to virus-induced killing (Chen et al., 1989; Dupressoir et al., 1989; van Hille et al., 1989; Cornelis et al., 1990). The possibility that cell transformation might also potentiate the cytotoxic action of parvoviral factors cannot, however, be excluded and is actually supported by some preliminary evidence (Salome et al., 1989). Parvovirus-induced oncolysis is interesting from a practical standpoint, as a means of eliminating transformed cells. It is also fascinating from a fundamental standpoint because it reflects how virus replication is controlled by cellular factors which are modulated by oncogenic transformation. These factors could either stimulate the parvovirus life-cycle and be activated in transformed cells or inversely, suppress virus replication and be inactivated in transformed cells. Such elements should be sought among the transcription factors whose activities are regulated as a function of neoplasic transformation and among other nuclear proteins that directly or indirectly interact with the parvoviral genome. Candidates for both types of cellular factors have recently been identified (Walton et al., 1989; Ahn et al., 1989; Blundell and Astell, 1989; Chang et al., 1989; Avalosse et al., 1989). These studies are at their very beginning but they herald the use of parvoviruses as means of investigating cellular markers of malignant transformation. Parvoviruses and human cancer

Man can be infected by various genuine human parvoviruses: the autonomous B19 virus and three or four serotypes of AAV viruses which are

246

associated with active adenovirus infections (Anderson, 1990). Moreover, the autonomous H-l virus, a more common contaminant of rodents, also proved to be infectious for humans (Toolan et al., 1965). Infected individuals undergo seroconversion, the maintenance of which suggests that a persistent viruscarrier state is established. In principle, anticancer surveillance could occur under these conditions, following parvovirus reactivation and expression of the viral antineoplasic potential when clones of transformed cells appear. Certain epidemiological data are compatible with this possibility: an inverse correlation has been established between the risk of cancer, namely cervical carcinoma, and the seropositivity index reflecting prior AAV virus infection (SprecherGoldberger et al., 1971; Mayor et al., 1976; Georg-Fries et al., 1984). These studies do not exclude, however, the possibility that other parameters of a socioeconomic nature might be responsible for the observed differences. At any rate, there is no doubt that tumors can appear in humans despite the presence of parvoviruses. For instance, chronic B19 infections have been described in leukemic patients (Kurtzman et al., 1988; Malarme et al., 1989). A possible cause of this escape from parvoviral surveillance could be an incompatibility between the tumor and the tropism of the virus. On the other hand, even if the tumor were a favorable terrain for viral replication, the magnitude of the antitumor response could be limited by the initial quantity of virus present at the site, by inefficient local production of new particles or by the host’s immune response (Rommelaere and Tattersall, 1990). These restrictions explain the fact that various parvoviruses have been isolated from primary or transplanted human tumors, or from human tumor cell lines (Siegl, 1984; Georg-Fries et al., 1984). These limitations might also explain the failure of a previous attempt to use virus H-l to treat two patients with advanced, disseminated osteosarcoma that proved resistant to conventional therapies (Toolan et al., 1965). Infection was followed by viremia and seroconversion but remained asymptomatic. Although the level of alkaline phosphatase in the serum was temporarily stabilized in one of these cases, tumor development was not stopped by the doses of H-l used in the treatment. It thus appears that parvoviral oncosuppression probably reflects, at least in part, the regulation of virus replication in relation to cell proliferation and differentiation. Consequently, the potential therapeutic use of parvoviruses as anticancer agents first requires that cellular factors controlling viral multiplication are defined and the mode of cytotoxicity of these viruses is unraveled. We cannot expect the modulation of parvovirus-host cell interactions to produce a universal, all-or-nothing-type response. Nevertheless, the apparent harmlessness of certain parvoviruses towards humans, combined with their oncotropism, seems to justify the hope of using these agents in conjunction with other treatments, to express parvoviral or heterologous cytotoxic factors in human tumors.

241

Acknowkdghents

Work performed in our laboratory on this subject was supported by the Commission of European Communities, by the Centre National de la Recherche Scientifique, Institut National de la Sante et de la Recherche Medicale, Fondation de France, Association pour la Recherche sur le Cancer and Ligue contre le Cancer (France) and by the Fonds National de la Recherche Scientifique, Services de Programmation de la Politique Scientifique, Commissariat General aux Relations Internationales, Oeuvre Belge du Cancer, Caisse Generale d’Epargne et de Retraite, Banque Nationale and Solvay et Cie (Belgium).

References Ahn, J.K., Gavin, B.J., Kumar, G. and Ward, DC. (1989) Transcriptional analysis of minute virus of mice P4 promoter mutants. J. Virol. 63, 5425-5439. Anderson, L.J. (1990) Human parvoviruses. J. Infect. Dis. 161, 603-608. Antoni, B.A., Rabson, A.B., Miller, I.L., Trempe, J.P., Chejanovski, N. and Carter, B.J. (1990) Adeno-associated virus rep protein inhibits human immunodefEiency virus type 1 production in human cells. J. Virol. 65, 396404. Antonietti, J-P., Sali, R., Beard, P. and Hirt, B. (1988) Characterization of the cell type-specific determinant in the genome of minute virus of mice. J. Virol. 62, 552-557. Avalosse, B.A., Berrijal, S., Chen, Y.Q., Cassiman, J.J. and Rommelaere, J. (1989) Identification of a transformation-sensitive nuclear protein from normal human Iibroblasts, that specitically interacts with minute virus of mice DNA and correlates with cell resistance to the parvovirus. Mol. Carcinogenesis 2, 245-252. Bando, H., Choi, H., Ito, Y. and Kawase, S. (1990) Terminal structure of a densovirus implies a hairpin transfer replication which is similar to the model for AAV. Virology 179, 5763. Bantel-Schaal, U. and zur Hausen, H. (1988) Adeno-associated viruses inhibit SV40 DNA amplification and replication of herpes simplex virus in SV40-transformed cells. Virology 164, 64-74. Bantel-Schaal, U. (1990) Adeno-associated parvoviruses inhibit growth of cells derived from malignant human tumors. Int. J. Cancer 45, 190-194. Bass, L.R. and Hetrick, F.M. (1978) Human lymphoblastoid cells as hosts for parvoviruses H-l and rat virus. J. Virol. 25, 486490. Bergs, V.V. (1969) Rat virus-mediated suppression of leukemia induction by Moloney virus in rats. Cancer Res. 29, 1669-1673. Berns, K.I. and Bohenzky, R.A. (1987) Adeno-associated viruses: an update. Adv. Virus Res. 32, 243-307. Berns, K.I. and Labow, M.A. (1987) Parvovirus gene regulation. J. Gen. Virol. 68, 601614. Bems, K.I. (1990) Parvovirus replication. Microbial. Rev. 54, 316-329. Blacklow, N.R., Cukor, G., Kibrick, S. and Quinnan, G. (1978) Interaction of adeno-associated viruses with cells transformed by herpes simplex virus. In: DC. Ward and P. Tattersall (Eds), Replication of Mammalian Parvoviruses. Cold Spring Harbor Laboratory, Cold Spring Harbor, pp. 87-98. Blundell, MC. and Astell, C.R. (1989) A GC-box motif upstream of the Bl9 parvovirus unique promoter is important for in vitro transcription. J. Virol. 63, 4814-4823. Bodnar, J.W. (1988) Sequence organization in regulatory regions of DNA of minute virus of mice. Virus Genes 2, 167-182. Brandenburger, A., Legendre, D., Avalosse, B. and Rommelaere, J. (1990) NS-1 and NS-2 proteins

248 may act synergisti~aiiy in the cytopathogenicity of parvovirus MVMp. Virology 174, 576-584. Caillet-Fauquet, P., Perros, M., Brandenburger, A., Spegelaere, P. and Rommelaere, J. (1990) Programmed killing of human cells by means of an inducible clone of parvoviral genes encoding non-structural proteins. EMBO J. 9, 2989-2995. Casto, B.C. and Goodheart, CR. (1972) Inhibition of adenovirus transformation in vitro by AAV1. Proc. Sot. Exptl. Biol. Med. 140, 72-78. Chang, L-S, Shi, Y. and Shenk, T. (1989) Adeno-associated virus P5 promoter contains an adenovirus ElA-inducible element and a binding site for the major late transcription factor. J. Virol. 63, 347993488. Chen, Y.Q., de Foresta, F., Hertoghs, J., Avalosse, B., Cornelis, J. and Rommelaere, J. (1986) SeIective killing of simian virus 40-transformed human libroblasts by parvovirus H-l. Cancer Res. 46, 35743579. Chen, Y.Q., Tuynder, M.C., Cornelis, J.J., Boukamp, P., Fusenig, NE. and Rommelaere, J. (1989) Sensitization of human keratinocytes to killing by parvovirus H-l takes place during their malignant transformation but does not require them to be tumorigenic. Carcinogenesis IO, 163167. Cornelis, J.J., Becquart, P., Duponchel, N., Salome, N., Avalosse, B.L., Namba, M. and Rommelaere, J. (1988a) Transformation of human libroblasts by ionizing radiation, a chemical carcinogen, or simian virus 40 correlates with an increase in susceptibility to the autonomous parvoviruses H-l virus and minute virus of mice. J. Virol. 62, 167991686. Cornelis, J.J., Spruyt, N., Spegelaere, P., Guetta, E., Darawshi, T., Cotmore, SF., Tat, J. and Rommelaerc, J. (1988b) Sensitization of transformed rat fibroblasts to kilhng by parvovirus minute virus of mice correlates with an increase in viral gene expression. J. Virol. 62, 3438-3444. Cornelis, J.J., Chen, Y.Q., Spruyt, N., Duponchel, N., Cotmore, SF., Tattersall, P. and Rommelaere, J. (1990) Susceptibility of human cells to killing by the parvoviruses H-l and minute virus of mice correlates with viraf transcription. J. Virol. 64, 2537-2544. Cotmore, SF. and Tattersall, P. (1987) The autonomously replicating parvoviruses of vertebrates. Adv. Virus Res. 33, 91-169. Cukor, G., Blacklow, N.R., Kibrick, S. and Swan, I.C. (1975) Effect of adeno-associated virus on cancer expression by herpesvirus-transformed hamster cells. J. Natl. Cancer Inst. 55, 957-959. Cukor, G., Blacklow, N.R., Hoggan, M.D. and Berns, K.I. (1984) Biology of adeno-associated virus. In: K.I. Bems (Ed), The Parvoviruses. Plenum Press, New York, pp. 33-66. de la Maza, L. and Carter, B.J. (1981) Inhibition of adenovirus oncogenicity in hamsters by adenoassociated virus DNA. J. Nat]. Cancer Inst. 67, 132331326. Dupressoir, T., Vanacker, J.M., Comelis, J., Duponchel, N. and Rommelaere, J. (1989) Inhibition by parvovirus H-l of the formation of tumors in nude mice and colonies in vitro by transformed human mammary epithelial cells. Cancer Res. 49, 3203-3208. Faisst, S., Schlehofer, J.R. and zur Hausen, H. (1989) Transformation of human cells by oncogenic viruses supports permissiveness for parvovirus H-l propagation. J. Virol. 63, 2152-2158. Fox, E., Moen, P.T. and Bodnar, J.W. (1990) Replication of minute virus of mice DNA in adenovirus-infected or adeno~rus-transformed cells. Virology 176, 403-412. Cardiner, E.M. and Tattersall, P. (1988) Evidence that developmentally regulated control of gene expression by a parvoviral allotropic determinant is particle-mediated. J. Virol. 62, 1713-1722. Georg-Fries, B., Biederlack, S., Wolf, J. and zur Hausen, H. (1984) Analysis of proteins, helper dependence and seroepidemiology of a new human parvovirus. Virology 134, 64-71. Guetta, E., Graziani, Y. and Tal, J. (1986) Suppression of Ehrlich ascites tumors in mice by minute virus of mice. J. Nat]. Cancer Inst. 76, 1177-l 180. Guetta, E., Mincberg, M., Mousset, S., ~~~nchamps, C., Rommelaere, J. and Tal, J. (1990) The selective killing of transformed rat cells by minute virus of mice correlates with elevated viral gene expression but does not require infectious virus production. J. Viral. 64, 458-462. Heilbronn, R., Schlehofer, J.R. and zur Hausen, H. (1984) Selective killing of carcinogen-treated SV40-transformed Chinese hamster ceils by a defective parvovirus. Virology 136, 439-441. Heilbronn, R., Burkle, A., Stephan, S. and zur Hausen, H. (1990) The adeno-associated rep gene suppresses herpes simplex virus-induced DNA ampli~~ation. J. Viral. 64, 3012-3018.

Hermonat, P.L. (1989) The ad~no-aviate virus Rep 78 gene inhibits cellular tmnsfo~ation induced by bovine papillomavirus. Virology 172, 2.53-261. Ito, T., Wang, D-Q, Maru, M., Nakajima, K., Kato, S., Kurimura, T. and Wakamiya, N. (1990) Antitumor efficacy of vaccinia virus-modified tumor cell vaccine. Cancer Res. 50, 69156918. Katz, E. and Carter, B.J. (1986) Effect of adeno-associated virus on transformation of NIH3T3 cells by ras gene and on tumorigenicity of an NIH3T3 transformed cell line. Cancer Res. 46, 30233026. Kirchstein, R.L., Smith, K.O. and Peters, E.A. (1968) Inhibition of adenovirus 12 oncogenicity by adeno-associated virus. Proc. Sot. Exp. Biol. Med. 128, 67&673. Kurtzman, G., Cohen, B., Meyers, P., Amunullah, A. and Young, N.S. (1988) Persistent B19 parvovirus infection as a cause of severe chronic anaemia in children with acute lymphocytic leukaemia. The Lancet 2, 1159-l 162. Labow, M.A., Graf, L.H. and Bems, K.I. (1987) Adeno-associated virus gene expression inhibits cellular tmnsfo~ation by heterologous genes. Mol. Ceil Biol. 7, 132@-1325. Ledinko, N., Hopkins, S. and Toolan, H. (1969) Relationship between potentiation of H-l growth by human adenovirus 12 and inhibition of the ‘helper’ adenovirus by H-l. J. Gen. Virol. 5, 19Lorence, R.M., Road, P.A. and Kelley, K.W. (1988) Newcastle disease virus as an antineoplastic agent: induction of tumor necrosis factor-cc and augmentation of its cytotoxicity. J. Natl. Cancer Inst. 80, 1305-1311. Malarme, M., Vandervelde, D. and Brasseur, M. (1989) Parvovirus infection, leukaemia, and immunode~ciency. The Lancet 1, 1457. Mayor, H.D., Houlditch, G.S. and Mumford, D.M. (1973) Influence of adeno-associated satellite virus on adenovirus-induced tumors in hamster. Nature New Biol. 241, 44-46. Mayor, H.D., Drake, S., Stahmann, J. and Mumford, D.M. (1976) Antibodies to adeno-associated satellite virus and herpes simplex in sera from cancer patients and normal adults. Am. J. Obstet. Gynecol. 126, 100-104. Mousset, S. and Rommelaere, J. (1982) Minute virus of mice inhibits cell transfo~ation by simian virus 40. Nature 300, 537-539. Mousset, S., Cornelis, J., Spruyt, N. and Rommelaere, J. (1986) Transformation of established murine fibroblasts with an activated cellular Harvey-ras oncogene or the polyoma virus middle T gene increases cell permissiveness to patvovirus minute virus of mice. Biochimie 68, 951-955. Mousset, S. and Rommelaere, J. (1988) Susceptibility to parvovirus minute virus of mice as a function of the degree of host cell transformation: little effect of simian virus 40 infection and phorbol ester treatment. Virus Res. 9, 107-I 17. Nahreini, P. and Srivastava, A. (1989) Rescue and replication of the adeno-associated virus 2 genome in mortal and immortal human cells. Intervirology 30,7485. Ostrove, J.M., Duckworth, D.H. and Berns, K.I. (1981) Inhibition of adenovirus-transformed cell oncogenicity by adeno-associated virus. Virology 113, 521-533. Ozawa, K., Ayub, J., Kajigaya, S., Shimada, T. and Young, N. (1988) The gene encoding the nonstructural protein of Bl9 (human) parvovirus may be lethal in transfected cells. J. Virol. 62, 28842889. Rhode, S.L. (1987) Construction of a genetic switch for inducible tram-activation of gene expression in eucaryotic cells. J. Virol. 61, 1448-1456. Rhode, S.L. and Richard, S.M. (1987) Characterization of the trans-activation-responsive element of the parvovirus H-l P38 promoter. J. Virol. 61, 2807-2815. Rhode, S.L. and Paradiso, P.R. (1989) Parvovirus replication in normal and transformed human cells correlates with the nuclear translocation of the early protein NSl. J. Virol. 63, 349-355. Rommelaere, J. and Tattersall, P. (1990) On~supp~ssion by parvoviruses. In: P. Tijssen (Ed), Handbook of Parvoviruses. Vol. 2. CRC Press, Boca Raton, pp. 41-57. Rood, P.A., Lorence, R.M. and Kelley, K.W. (1990) Serum protease inhibitor abrogation of Newcastle disease virus enhancement of cytolysis by recombinant tumor necrosis factors alpha and beta. J. Natl. Cancer Inst., 82, 213-217. Salome, N., van Hille, B., Geuskens, M. and Rommelaere, J. (1989) Partial reversion of conditional

transfo~ation correlates with a decrease in the sensitivity of rat cells to killing by the parvovirus minute virus of mice but not in their capacity for virus production: effect of a temperaturesensitive v-src oncogene. J. Viral. 63, 47974807. Salome, N., van Hille, B., Duponchel, N., Meneguzzi, G., Cuzin, F., Rommelaere, J. and Cornelis, J. (1990) Sensitization of transformed rat cells to parvovirus MVMp is restricted to specific oncogenes. Oncogene 5, 123- 130. Schirrmacher, V., Ahlert, T., Heicappel, R., Appelhans, B. and Von Hoegen. (1986) Successful application of non-oncogenic viruses for antimetastatic cancer immunotherapy. Cancer Rev. 5, 1949. Schlehofer, J.R., Heilbronn, R., Georg-Fries, B. and zur Hausen, H. (1983) Inhibition of initiatorinduced SV40 gene ampli~~ation in SV40-transformed Chinese hamster cells by infection with a defective parvovirus. Int. J. Cancer 32, 591-595. Schlehofer, J.R., Ehrbar, M. and zur Hausen, H. (1986) Vaccinia virus, herpes simplex virus, and carcinogens induce DNA amplification in a human cell line and support replication of a helper virus-dependent parvovirus. Virology 152, 110-I 17. Schlehofer, J.R. and Heilbronn, R. (1990) Infection with adeno-associated virus type 5 inhibits mutagenicity of herpes simplex virus type 1 or 4-nitroquinoline-l-oxide. Mutat. Res. 244, 317320. Schmitt, J., Schlehofer, J.R., Mergener, K., Gissman, L. and zur Hausen, H. (1989) Amplification of bovine papillomavirus DNA by N-methyl-N’-Nitro-N-Nitrosoguanidine, ultraviolet irradiation, or infection with herpes simplex virus. Virology 172, 73-81. Siegl, G. (1984) Biology and pathogenicity of autonomous parvoviruses. In: K.I. Berns (Ed), The Parvoviruses. Plenum Press, New York, pp. 297-362. Siegl, G. and Tratschin, J.-D. (1987) Parvoviruses: agents of distinct pathologic and molecular potential. FEMS Microbial. Rev. 46, 433450. Sprecher-Goldberger, S., Thiry, L., Lefebvre, N., Dekegel, D. and de Halleux, F. (1971) Complement-fixation antibodies to adenovirus-associated viruses, adenoviruses, cytomegaloviruses and herpes simplex viruses in patients with tumors and in control individuals. Am. J. Epidemiol 94, 351-358. Su, Z.Z., Luo, Z.Y., Guo, L.P., DuPont, F., Avalosse, B. and Rommelaere, J. (1988) Positive selection of human cells lacking several transformation parameters from an SV40-transformed culture by means of parvovirus H-l. Carcinogenesis 9, 13951400. Tattersall, P. and Gardiner, E.M. (1990) Autonomous parvovirus-host-cell interactions. In: P. Tijssen (Ed), Handbook of Parvoviruses. Vol. 1, CRC Press, Boca Raton, pp. 11 I-121. Toolan, H.W. and Ledinko, N. (1965) Growth and cytopathogenicity of H-viruses in human and simian cell cultures. Nature 208, 8 12-8 13. Toolan, H.W., Saunders, E.L., Southam, C.M., Moore, A.E. and Levin, A.G. (1965) H-l virus viremia in the human. Proc. Sot. Exp. Biol. Med. 119, 711-715. Toolan, H.W. (1967) Lack of oncogenic effect of the H-viruses. Nature 214, 1036. Toolan, H.W. and Ledinko, N. (1968) Inhibition by H-l virus of the incidence of tumors produced by adenovirus 12 in hamsters. Virology 35, 475478. Toolan, H.W., Rhode, S.L. and Gierthy, J.F. (1982) Inhibition of 7, 12-dimethylbenz(a)anthraceneinduced tumors in Syrian hamsters by prior infection with H-l parvovirus. Cancer Res. 42,255225.55. van HilIe, B., Duponchel, N., SalomC, N., Spruyt, N., Cotmore, S.F., Tattersall, P., Cornelis, J. and Rommelaere, J. (1989) Limitations to the expression of parvoviral nonstructural proteins may determine the extent of sensitization of ~-rub-transfo~ed rat cells to minute virus of mice. Virology 171, 89-97. Walton, T.H., Moen, P.T., Fox, E. and Bodnar, J.W. (1989) Interactions of minute virus of mice and adenovirus with host nucleoli. J. Virol. 63, 3651-3660. Wanmin, L., Yanzhen, C., Bogen, Y., Zuyu, L. and Zhiying, L. (1989) Preliminary comparison of sensitivity toward parvovirus H-l of human hepatoma cells and parahepatoma tissue. Chinese J. Cancer Res. 1, 15-20. Winocour, E., Callaham, M.F. and Huberman, E. (1988) Perturbation of the cell cycle by adeno-

251 associated virus. Virology 167, 393-399. Wiedbrauk, D.L., Bloom, M.E. and Lodmell, D.L. (1986a) Mink parvoviruses and interferons: in vitro studies. J. Viral. 60, 1179-I 182. Wiedbrauk, D.L., Hadlow, W.J., Ewalt, L.C. and Lodmell, D.L. (1986b) Interferon response in normal and aleutian disease virus-infected mink. J. Virol. 59, 514517. Yakobson, B., Koch, T. and Winocour, E. (1987) Replication of adeno-associated virus in synchronized cells without the addition of a helper virus. J. Viral. 61, 972-981. Yakobson, B., Hrynko, T.A., Peak, M.J. and Winocour, E. (1989) Replication of adeno-associated virus in cells irradiated with UV light at 254 nm. J. Virol. 63, 1023-1030. Yalkinoglu, A.O., Heilbronn, R., Burkle, A., Schlehofer, J.R. and zur Hausen, H. (1988) DNA amplification of adeno-associated virus as a response to cellular genotoxic stress. Cancer Res. 48, 3123-3129. Yalkinoglu, A.O., Schlehofer, J.R. and zur Hausen, H. (1990) Inhibition of N-methyl-N-nitro-Nnitroso~anidine-indu~ methotrexate and adriamycin resistance in CHO cells by adenoassociated virus type 2. Int. J. Cancer 45, 1195-1203. Yang, D.V.M. (1987). Parvovirus-induced regression of canine transmissible venereal sarcoma. Am. J. Vet. Res. 48, 799-800.