Human ubiquitous JCV(CY) T-antigen gene induces brain tumors in experimental animals

Article metrics


JCV is a papovavirus which is widespread in the human population. The prototype Mad-1 variant of JCV induces a fatal demyelinating disease of the central nervous system (CNS) called Progressive Multifocal Leukoencephalopathy (PML) in immunosuppressed individuals. The unique tropism of JCV (Mad-1) to the CNS is attributed to the tissue-specific regulation of the viral early promoter which is responsible for the production of the viral regulatory protein, T-antigen. The archetype form of this virus, JCV(CY), which has been repeatedly isolated from the urine of PML and non-PML individuals, is distinct from JCV(Mad-1) in the structural organization of the regulatory sequence. To characterize the tissue specific expression of JCV(CY) and to investigate its potential in inducing disease, transgenic mice containing the early region of JCV(CY) were generated. Some of these mice between 9 – 13 months of age exhibited signs of illness as manifested by paralysis of rear limbs, hunched posture, and poor grooming. Neuropathological examination indicated no sign of hypomyelination of the brain, but surprisingly, revealed the presence of primitive tumors originating from the cerebellum and the surrounding brain stem. The tumor masses also infiltrated the surrounding tissue. Results from RNA and protein studies revealed a high level of T-antigen mRNA expression in hindbrains of clinically normal and affected transgenic mice. However, higher levels of T-antigen RNA and protein were detected in brains of the animals exhibiting severe illness. The close resemblance of JCV(CY) induced tumor in transgenic mice to the human medulloblastoma/primitive neuroectodermal tumor (PNETs) in location, histologic appearance, and expression of marker proteins strongly suggests the utility of this novel animal model for the study of human brain tumors.


JC virus (JCV) is a human neurotropic polyomavirus infecting more than 70% of the human population (Frisque and White, 1992). This virus was originally isolated from the brains of patients with the chronic demyelinating disease Progressive Multifocal Leukoencephalopathy (PML) (Padgett et al., 1971; ZuRhein and Chou, 1965). PML is an opportunistic disease that affects patients with an impaired immune system due to various illnesses, immunosuppressive therapeutic treatments, and genetic disorders (Astrom et al., 1958; Berger and Concha, 1995). In addition to its causative role in PML, several studies have reported detection of JCV PML variants such as Mad-1, Mad-4, etc., from a variety of human brain tumors (Castaigne et al., 1974; Egan et al., 1980; Ho et al., 1980; Liberski et al., 1982; Rencic et al., 1996; Sima et al., 1983). Although there is no direct evidence for tumorigenesis of JCV in the human CNS, the oncogenic potential of JCV(Mad) in several experimental models has been well established. More than 80% of neonatal hamsters inoculated intracerebrally, intraperitoneally, or subcutaneously with JCV(Mad) developed glioblastomas, medulloblastomas, neuroblastomas, pineocytomas, and other tumors of central nervous system (CNS) origin (Walker et al., 1973; ZuRhein, 1983, 1987; ZuRhein and Varakis, 1975). Intraocular inoculation of prototype JCV(Mad-1) in hamsters has also induced abdominal neuroblastomas which subsequently metastasized to other organs such as lymph nodes, liver and bone marrow (Ohashi et al., 1978; Varakis et al., 1976).

The structural organization of the JCV genome isolated from the brains of PML patients exhibits striking similarity to that of the well-characterized simian virus 40 (SV40). The viral early gene encoding the regulatory protein, T-antigen, is separated from the late gene encoding the structural capsid protein by a non-coding sequence encompassing the regulatory motifs for viral gene transcription and the origin of viral DNA replication (Frisque et al., 1984). High level expression of the viral early protein, T-antigen, has been observed in hamster cells derived from JCV-induced glioblastomas (Frisque et al., 1980; Raj et al., 1995). Evidence for the direct involvement of JCV(Mad-1) T-antigen expression and tumor formation stems from earlier studies on transgenic mice that constitutively produced T-antigen from JCV(Mad-1) under the control of the viral control region. Four out of five transgenic mice developed adrenal neuroblastomas which metastasized to several other tissues. JCV T-antigen RNA was detected at high levels in the tumor tissues and at very low levels in normal tissues of the affected mice (Small et al., 1986a,b).

The control region of the PML isolates of JCV, such as the Mad-1 variant, is composed of two characteristic tandem 98 bp repeats that confer specificity for expression of JCV early and late genes in CNS cells (Frisque et al., 1984). Classic molecular techniques have been utilized in order to delineate the functional importance of various regulatory motifs of the viral promoter and to identify transcription factors, which upon interaction with the viral DNA sequence, modulate CNS-specific expression of the viral genome (Raj and Khalili, 1995). In addition to its productive replication in the brain, JCV particles have been isolated from the urine of PML patients as well as non-PML individuals who are experiencing mild to severe immunosuppression as well as in a large number of pregnant women (Arthur and Shah, 1989; Coleman et al., 1980; Gardner et al., 1984; Myers et al., 1989; Weiner et al., 1972). Viral sequences have also been demonstrated in kidneys and lymphocytes of healthy individuals leading to the speculation that these tissues may provide sites for JCV latency (Arthur et al., 1989; Dorries et al., 1994; Ishaq and Stoner, 1994; Matsushima et al., 1997; Tornatore et al., 1992; Vago et al., 1996; White et al., 1992). Of particular interest, in this respect, is the notion that the regulatory region of JCV from many of these healthy individuals is similar to that found in a Japanese urine isolate JCV(CY) (Yogo et al., 1990) and distinct from those originally obtained from PML brains. The control region of JCV(CY) contains only one copy of the 98 bp enhancer/promoter found in JCV(Mad) with insertions of 23 and 64 – 66 nucleotides (Figure 1). It appears that deletions of the 23 and 64 – 66 nucleotide sequences from JCV(CY) and duplication of the remaining 98 bp creates a structure which is found in JCV(Mad). These observations have led to the speculation that JCV(CY) can undergo, albeit under appropriate conditions, a DNA rearrangement process and gains a structure which is adapted to brain, i.e. JCV(Mad-1), and through cytolytic infection of glial cells induces demyelination in immunosuppressed individuals. Thus, JCV(CY) has been designated as the archetype strain. In order to gain insight into the in vivo expression of JCV(CY) early promoter and to evaluate the pathogenic potential of the viral early protein, T-antigen, in a whole animal system, we developed transgenic mice containing the JCV(CY) early region. Results from these studies reveal that the animals with the JCV(CY) early sequence develop brain tumors which resemble human medulloblastoma/primitive neuroectodermal tumors. A high level of JCV(CY) early protein was detected in the tumor cells.

Figure 1

Control region of JCV(CY). Top. Schematic diagram illustrating the regulatory region of JCV(CY). The direction of the viral early gene (T-antigen) and the late genes (capsid) are shown by the arrows on top. ORI represents the origin of viral DNA replication and the 64 and 23 boxes (solid) illustrate the positions of the nucleotide insertion which divide the 98 bp enhancer sequences found in JCV(Mad) into three segments (open boxes). For detection of the JCV(CY) regulatory sequence in transgenic mice and the evaluation of the nucleotide composition, primers derived from nucleotides 230 – 248 and 4987 – 5006, as depicted by the arrows at the bottom, were utilized in PCR. The DNA sequences of the amplified DNA are shown. The underlined nucleotides point to the positions of the primers, and the nucleotides in boldface highlight the 23 and 64 nucleotide insertions

Results and discussion

Recent reports on world-wide distribution of JCV(CY) in the non-PML human population (Loeber and Dorries, 1988; White et al., 1992) prompted us to investigate the cell type-specific expression of the JCV(CY) promoter and to evaluate the effect of T-antigen expression on cell function in a transgenic animal model. Toward this end, the JCV(CY) early region (see Figure 1), encoding the early genome (T-antigen) and the viral regulatory sequences, was injected into one cell FVB/N mouse embryos. Of 20 litters spanning the first two generations descended from the founders, birth weights were normal, litter size averaged seven pups and the average male: female ration was 1 : 1. Transgenic animals showed no sign of illness and were behaviorally normal from birth to as much as 9 months of age. The founder mice containing the JCV sequences and nearly 50% of their transgenic offspring (35 of 79 counted) showed signs of illness at 9 – 13 months or more of age manifested by poor grooming, paralysis of rear limbs, and hunched posture. None of the offspring exhibited shaking with motion or shivering, a phenotype which was observed in a JCV(Mad) transgenic line and was caused by dysmyelination in the CNS (Small et al., 1986a). Analysis of DNA from the affected animals by PCR and direct sequencing revealed the integrity of the transgene with no evidence for the rearrangement of the JCV(CY) control sequence in the transgenic animals (Figure 1). Three positive animals with JCV(CY) (BB4, FF6, II4) were selected for biochemical and four (BB4, FF6, HH6 and II2) for histopathological evaluation. FF6 became ill at 9 months of age when tissue samples were harvested. BB4 and II4, who were both positive mice with no sign of illness, were sacrificed at 13 and 10 months of age, respectively and their organs were collected for simultaneous studies with those from FF6. Figure 2 (top panels) illustrates results from RT – PCR analysis of RNA from various tissues of BB4, FF6, and II4, and the control age-matched negative littermates, KK3 (panels a – d, respectively). T-antigen RNA was detected in the hindbrains of all positive mice (panels a – c, lane 5). The brain of FF6 mouse contained detectable levels of T-antigen RNA in contrast to BB4 and II4 in which the T-antigen RNA was not detected. The levels of actin mRNA were comparable in various tissues of transgenic and control animals (Figure 2, bottom panels).

Figure 2

Detection of JCV(CY) early RNA in transgenic animals. Approximately 1 μg of total RNA obtained from various tissues of BB4 (a), FF6 (b), II4 (c), and KK3 (d) was used to amplify a 242 nucleotide sequence of the JCV early gene or the 400 nucleotide of the control β-actin sequence. The position of RT – PCR products representing T-antigen and β-actin are shown by the arrowheads

To assess the level of T-antigen protein in the brains of these transgenic animals, we performed immunoprecipitation and Western blot analysis of protein extracts from brains of BB4, FF6, and II4, and control age-matched negative, KK3. As shown in Figure 3, high level of T-antigen was produced in FF6 brain (lane 2). The absence of a band corresponding to T-antigen protein in BB4 (lane 1) and II4 (lane 3) indicate that the low levels of the viral early RNA, as detected by RT – PCR in the brains of those animals (shown in Figure 2a,c lane 1), may not be sufficient to produce detectable levels of T-antigen. Protein extract from CG4 cells that constitutively express T-antigen was used as a control for T-antigen detection (Figure 3a, lane 5). As expected, no signal corresponding to T-antigen was observed in KK3 brain extract (Figure 3a, lane 4).

Figure 3

Production of JCV(CY) T-antigen and p53 in brains of transgenic animals. (a) Protein extracts (60 μg) from brains of BB4 (lane 1), FF6 (lane 2), II4 (lane 3), and KK3 (lane 4), and the CG-4 T-antigen-producing cell line (lane 5) were prepared and after immunoprecipitation with 5 μl of anti-T-antigen antibody (clone 416, Oncogene Science), were analysed by Western blot utilizing anti-T-antigen antibody, according to the previously described method (Franks et al., 1996). The position of T-antigen is shown by an arrowhead. Asterisks point to the IgG heavy and light chains from the immunoprecipitation and serve as a control for loading. (b) Equal amounts (50 μg) of protein extract from brains of KK3 (lane 1), II4 (lane 2), FF6 (lane 3) were fractionated by SDS – PAGE and after transfer to nitrocellulose, the immunoblot was reacted with anti-p53 antibody (Ab-1, Oncogene Science). The position of p53 is shown by an arrowhead. (c) 100 μg of protein extract from brains of KK3 (lane 1) and FF6 (lane 3) were immunoprecipitated with anti-p53 antibody and the immunocomplexes were fractionated by SDS – PAGE. In lane 2, the immunocomplex obtained from CG4-T-antigen upon treatment with anti-T-antigen antibody was loaded on the gel. The blot was transferred to nitrocellulose, and subjected to Western blot analysis with antibody to T-antigen. The arrowhead indicates the presence of T-antigen and the asterisks indicate IgG heavy and light chains from immunoprecipitation

In the next study, we examined the level of expression of the cellular regulatory protein, p53, and its association with JCV T-antigen in brain extract from the FF6 animal. First, the level of p53 was evaluated in protein extracts from brains of KK3, II4, and FF6. As shown in Figure 3b, while no signal corresponding to p53 was detected in the extract from KK3, low and noticeably higher levels of p53 were detected in brain extracts from II4 and FF6, respectively. Next, protein extract from brains of KK3 and FF6 was immunoprecipitated with anti-p53 antibody (Oncogene Science) and the immunocomplex was analysed by Western blot utilizing anti-T-antigen antibody. Protein extract from T-antigen producing CG4 cells was analysed by Western blot and served as a control for T-antigen detection (Figure 3c, lane 2). The results in Figure 3c show the presence of a band corresponding to T-antigen in FF6 brain extracts (lane 3) suggesting that T-antigen is associated with p53 in this transgenic animal. Thus, it is likely that the association of T-antigen with p53 stabilizes p53 and leads to its detection by Western blot as shown in Figure 3b.

Gross examination of the various organs including brains of transgenic animals expressing either low or high levels of T-antigen showed no distinct signs of abnormality which could cause the illness in the affected animals. Microscopic examination of the brain from the ill or healthy JCV(CY) positive mice showed no evidence of abnormalities in myelin, abnormal oligodendrocytes, or abnormal astrocytes. Of particular interest, however, was the detection of primitive neuroectodermal tumors (PNETs) in the hindbrains of these animals resembling human medulloblastomas. The tumors consisted of sheets of cell with high nuclear/cytoplasmic ratios, occasionally Homer Wright rosettes, frequent pyknotic nuclei, and approximately one mitotic figure/40× high powered field. Figure 4a illustrates an island of tumor cells in a cerebellar hemisphere of FF6. The tumor cells abutting the cerebellar folia showed microscopic interdigitations in the molecular layer and early subarachnoid spread (Figure 4b). In the other three animals, tumors were larger and it was not possible to determine the anatomic region of their origin. However, they were located, at least partially, around the brain stem and cerebellum, and spread diffusely into the subarachnoid space and infiltrated CNS tissues. Figure 4c illustrates extensive tumor infiltration of the spinal cord by subarachnoid mestastases leaving residual dorsal white matter at the top of the photograph. The anaplastic nuclei, mitotic figures, and apoptotic cells of the tumor are demonstrated in Figure 4d. Homer-Wright rosettes are shown in Figure 5d. Electron microscopy of the tumor revealed apoptotic nuclei, granular cytoplasm, and extensions with numerous processes containing few, if any, differentiating organelles (Figure 4e,f) further supporting the primitive nature of these tumors.

Figure 4

Light and electron microscopy. (a – d) Light microscopy analysis of brains from test animals. At low power (a) the cerebellar location of the tumor in FF6 is demonstrated by an arrow and at higher power (b) its invasion of the molecular layer is indicated by an arrow. Even more extensive tumor infiltration of native tissue is seen in the spinal cord cross section of HH6 (c). Higher power views of the tumor (d) demonstrate anaplastic nuclei (black arrow), apoptotic cells (arrowhead) and mitotic figures (red arrow). (e – f) Electron microscopic analysis of tumors. In panel a, an arrow depicts an apoptotic nucleus within a group of tumor nuclei infiltrating white matter. In panel f, the processes of the tumor cells reveal granular cytoplasm with few if any definable organelles

Figure 5

Immunohistochemistry. Immunostaining of tumor cells for detection of JCV early gene product, p53 and cell markers. Immunopositivity (indicated by arrows) shows that a large percentage of tumor cells have nuclear staining for T-antigen (a), occasional cells are positive for synaptophysin (b), and neurofilament protein (c), and that some nuclei are strongly immunopositive for p53 (d). Homer-Wright rosettes are also prominent (indicated by arrowhead in panel d)

The non-CNS organs including kidneys and adrenal glands from JCV(CY) transgenic mice either with or without brain tumor showed no pathologic changes (data not shown).

Immunohistochemical staining of tumors for the viral protein showed nuclear staining for T-antigen in all tumor tissues with the positive percentage of positive nuclei ranging from approximately 25 – 75% (Figure 5a). The adjacent CNS tissue and the brain studied from the control transgenic mice failed to show any nuclear staining for T-antigen (data not shown). The tumor cells showed no evidence for expression of glial fibrillary acidic protein (GFAP), a marker for astrocytic differentiation. Synaptophysin and neurofilament protein, markers for neuronal differentiation, showed scattered immunoreactivity in the tumor cells (Figure 5b and c, respectively).

In accord with the results from Western blot analysis shown in Figure 3b, immunohistochemical staining of the tumor tissue for detection of p53 showed heavy staining of approximately 10% of tumor cells (Figure 5d).

Here, we describe the development of a transgenic line using the early region of the human ubiquitous JCV(CY) genome that drives the oncogenic protein, T-antigen. Earlier attempts utilizing JCV(Mad) which is commonly detected in PML brain, resulted in creation of transgenic animals which exhibit either dysmyelination of brain or development of undifferentiated mesenteric tumors in the abdominal region of the animal (Franks et al., 1996; Small et al., 1986b). While the coding region of JCV, which is responsible for expression of the viral early protein, T-antigen, is highly conserved between the Mad and CY, the regulatory sequences of these two variants exhibit distinct characteristics. Earlier molecular studies revealed several regulatory modules within JCV(Mad-1), which play an important role in the regulation of viral promoter activity in CNS cells. How structural variations in the regulatory sequence within the JCV(CY) may dictate expression of the viral T-antigen in the CNS, and perhaps other organs, remains to be investigated.

The development of primitive neuroectodermal CNS tumors in transgenic mice harboring the JCV(CY) early gene, a phenotype which has not previously been observed in transgenic animals containing JCV(Mad) T-antigen, suggests that the unique structural organization of the JCV(CY) promoter may provide specificity for T-antigen expression in CNS cells leading to this experimental tumor type. Although CNS PNETs have been observed in mice transgenic for the SV40 T-antigen, they have usually required non-viral promoter sequences specific for target CNS regions such as tyrosine hydroxylase (Suri et al., 1993), phenyethanolamine N-methyltransferase, luteinizing hormone or interphotoreceptor retinoid-binding protein (Fung and Trojanowski, 1995). Since previous work has demonstrated that the JCV(Mad-1) promoter/enhancer sequence has a significant role in determining tissue specific expression (Feigenbaum et al., 1992), it is not unreasonable that the JCV(CY) promoter/enhancer would specify a different tissue expression resulting in the tumor type observed in this transgenic model.

The data presented here argue for a novel animal model for brain tumorigenesis which closely parallels the human medulloblastomas/PNETs in location, histologic appearance, and expression of differentiation markers. This transgenic line, which presents no signs of myelin deficiency in brain or any abnormality in other tissues could be useful for studying the mechanism involved in the formation and progression of brain tumors.

Materials and methods

Transgenic mice generation and analysis

Transgenic animals were created by a technique detailed previously (Gordon and Ruddle, 1983) utilizing a 3.2 kilobase DNA fragment containing the early region of JCV(CY). For identification of the transgenic mice, genomic DNA was isolated (Franks et al., 1996) and analysed by polymerase chain reaction (PCR) utilizing specific primers as described in Figure 1.

Histological and ultrastructural analysis

For light microscopy, brains were fixed in 4% paraformaldehyde, 0.1 M phosphate buffer (pH 7.4) and embedded in paraffin. Following serial sectioning at 4 – 6 microns, H&E stained sections were obtained at 100 micron intervals. For ultrastructural analysis, the tumor sample was immersion-fixed in buffered 4% paraformaldehyde, 1% glutaraldehyde, embedded in Epon, and examined in thin sections stained with uranyl/lead.

For immunohistochemistry, sections were rehydrated with DH2O, quenced for endogenous peroxidase for 15 min at room temperature with 3% H2O2 in methanol, after washing in DH2O and incubation for 30 min at 95°C in 0.01 M citrate buffer (pH 6.0) to achieve maximum antigen retrieval, and washed with PBS. Primary monoclonal antibodies were applied for 16 h at room temperature. The antibodies used in this study were pAb416 (Oncogene Science) at 1 : 100 dilution for T-antigen, SF2 (Dako) at 1 : 100 dilution for glial filament protein, 2F11 (Dako) at 1 : 50 dilution for neurofilament, SY38 (Boehringer-Mannheim) at 1 : 50 dilution for synaptophysin, and p53 Ab-3 (Oncogene Science) at 1 : 20 dilution. Primary antigens were revealed with an Avidin-biotin complex technique using diaminobenzidine as the chromagen (Vectastain E-Lite Kit, Novocastra Laboratories, Ltd.).

Reverse transcriptase – polymerase chain reaction (RT – PCR)

One microgram of total RNA from brain, kidney, lung, and liver of transgenic mice was used in RT – PCR assay (Franks et al., 1996) to amplify a 242 base pair sequence of JCV T-antigen or the control 400 nucleotide β-actin. For synthesis cDNA, the JCV primer (5′-CCCCATACCAACATTAGCTTTC-3′) or the actin primer (5′-CTGGTTGCCAATAGTGATGA-3′) were mixed with 100 units of Moloney murine leukemia virus reverse transcriptase (RT) at 37°C for 30 min as described previously (Franks et al., 1996). The primers for the second strand synthesis for actin (5′-CCAGATCATGTTTGAGACCT-3′) and for JCV (5′-CCAGATTTGTAAGGCAGATAG-3′) were added to the RT reaction and PCR amplification was performed (Franks et al., 1996). The RT – PCR product was detected by Southern blot utilizing 80 mer oligonucleotide probes representing the internal sequences of the PCR product which have no overlapping sequences to the primers.

Immunoprecipitation/Western blot

Direct Western blot analysis of crude protein extracts from tumor and various other tissues were performed using 50 – 60 μg of protein. Protein extracts were prepared according to the procedure described previously (Franks et al., 1996). For detection of T-antigen, anti-T-antigen antibody (clone 416, Oncogene Science) was used, whereas for detection of p53 we used Ab-1 antibody (Oncogene Science) which detected both wild-type and mutant p53. For immunoprecipitation 100 μg of protein were mixed with the appropriate antibody and the immunocomplex was subsequently analysed by Western blot as previously described. (Franks et al., 1996; Krynska et al., 1997).


  1. Arthur RR and Shah KV. . 1989 Prog. Med. Virol. 36: 42–61.

  2. Arthur RR, Dagostin S and Shah KV. . 1989 J. Clin. Microbiol. 27: 1174–1179.

  3. Astrom K-E, Mancall EL and Richardson EP, Jr. . 1958 Brain 81: 93–127.

  4. Berger JR and Concha M. . 1995 J. NeuroVirology 1: 5–18.

  5. Castaigne P, Rondot P, Escourolle R, Ribadeau-Dumas J-L, Cathala F and Hauw J-J. . 1974 Rev. Neurol. 130: 379–392.

  6. Coleman DV, Wolfendale MR, Daniel RA, Dhanjal NK, Gardner SD, Gibson PE and Field AM. . 1980 J. Inf. Dis. 142: 1–8.

  7. Dorries K, Vogel E, Gunther S and Czub S. . 1994 Virology 198: 59–70.

  8. Egan JD, Ring BL, Reding MJ, Wells IC and Shuman RM. . 1980 Transplantation 29: 84–86.

  9. Feigenbaum L, Hinrichs SH and Jay G. . 1992 J. Virology 66: 1176–1182.

  10. Franks RR, Rencic A, Gordon J, Zoltick PW, Curtis M, Knobler RL and Khalili K. . 1996 Oncogene 12: 2573–2578.

  11. Frisque RJ, Rifkin DB and Walker DL. . 1980 J. Virol. 35: 265–269.

  12. Frisque RJ, Bream GL and Cannella MT. . 1984 J. Virol. 51: 458–469.

  13. Frisque RJ and White FA. . 1992 The molecular biology of JC virus causative agent of progressive multifocal leukencephalopathy. In: Roos R. (ed.) Molecular Neurovirology. Humana Press: Totowa, New Jersey. pp 25–158.

  14. Fung KM and Trojanowski JQ. . 1995 J. Neuropathol. Exp. Neurol. 54: 285–296.

  15. Gardner SD, MacKenzie EFD, Smith C and Porter AA. . 1984 J. Clin. Path. 37: 578–586.

  16. Gordon JW and Ruddle FH. . 1983 Methods Enzymol 101: 411–433.

  17. Ho K-C, Garancis JC, Paegle RD, Gerber MA and Borkowski WJ. . 1980 Acta Neuropathol. 52: 81–83.

  18. Ishaq M and Stoner GL. . 1994 Proc. Natl. Acad. Sci. USA 91: 8283–8287.

  19. Krynska B, Gordon J, Otte J, Franks R, Knobler R, Giordano A, De Luca A and Khalili K. . 1997 J. Cell. Biochem. 67: 223–230.

  20. Liberski PP, Alwasiak J and Wegrzyn Z. . 1982 Neuropat. Pol. 20: 3–4.

  21. Loeber G and Dorries K. . 1988 J. Virol. 62: 1730–1735.

  22. Matsushima T, Nakamura K, Oka T, Tachikawa N, Sata T, Murayama S, Nukina N and Kanazawa I. . 1997 Neurology 48: 279–282.

  23. Myers C, Frisque RJ and Arthur RR. . (1989), J. Virol. 63: 4445–4449.

  24. Ohashi T, ZuRhein GM, Varakis JN, Padgett BL and Walker DL. . 1978 J. Neuropathol. Exp. Neurol. 37: 667.

  25. Padgett BL, Walker DL, ZuRhein GM, Eckroade RJ and Dessel BH. . 1971 Lancet i: 1257–1260.

  26. Raj G and Khalili K. . 1995 Virology 213: 283–291.

  27. Raj GV, Gordon J, Logan TJ, Hall DJ, De Luca A, Giordano A and Khalili K. . 1995 Int. J. Oncol. 7: 801–808.

  28. Rencic A, Gordon J, Otte J, Curtis M, Kovatich A, Zoltick P, Khalili K and Andrews D. . 1996 Proc. Natl. Acad. Sci. USA 93: 7352–7357.

  29. Sima AF, Finkelstein SD and McLachlan DR. . 1983 Ann. Neurol. 14: 183–188.

  30. Small JA, Scangos G, Cork L, Jay G and Khoury G. . 1986a Cell 46: 13–18.

  31. Small JA, Khoury G, Jay G, Howley PM and Scangos GA. . 1986b Proc. Natl. Acad. Sci. USA 83: 8288–8292.

  32. Suri C, Fung BP, Tischler AS and Chikaraishi DM. . 1993 J. Neurosci. 13: 1280–1291.

  33. Tornatore C, Berger JR, Houff SA, Curfmann B, Meyers K, Winfield D and Major EO. . 1992 Ann. Neurol. 31: 454–462.

  34. Vago L, Cinque P, Sala E, Nebuloni M, Caldarelli R, Racca S, Ferrante P, Trabattoni GR and Costanzi G. . 1996 12: 139–146.

  35. Varakis J, ZuRhein GM, Padgett BL and Walker DL. . 1976 Cancer Res. 38: 1718–1722.

  36. Walker DL, Padgett BL, ZuRhein GM, Albert A and Marsh RF. . 1973 Science 181: 674–676.

  37. Weiner LP, Herndon RM, Narayan O, Johnson RT, Shah K, Rubenstein LJ, Preziosi TJ and Conley FK. . 1972 N. Engl. J. Med. 286: 385–390.

  38. White FA, III, Ishaq M, Stoner GL and Frisque RJ. . 1992 J. Virol. 66: 5726–5734.

  39. Yogo Y, Kitamura T, Sugimoto C, Ueki T, Aso Y, Hara K and Taguchi F. . 1990 J. Virol. 65: 2422–2428.

  40. ZuRhein GM and Chou SM. . 1965 Science 148: 1477–1479.

  41. ZuRhein GM and Varakis J. . 1975 Morphology of brain tumors induced by Syrian hamsters after inoculation with JC virus, a new human papovavirus. In: Kornyey S, Tariska S and Gosztonyi G . (eds.). Proceedings, VIIth International Congress of Neuropathology, vol.1. Akademia Kiado: Budapest. pp 479–481.

  42. ZuRhein GM. . 1983 Studies of JC virus-induced nervous system tumors in the Syrian hamster. A review in: Sever JL and Madden DM. (eds.). Polyomaviruses and Human Neurological Disease. Alan R Liss: New York. pp205–221.

  43. ZuRhein GM. . 1987 Human virus in experimental neurooncogenesis. In: Grundmann E. (ed.). Cancer Campaign: Experimental NeuroOncology, Brain Tumor, and Pain Therapy, vol 10. Gustav, Fischer: Stuttgart. pp19–46.

Download references


The authors wish to thank Drs Frisque and Yogo for kindly providing the JCV(CY) construct. We would like to thank past and present members of the Center for NeuroVirology and NeuroOncology for sharing of reagents, ideas, and their insightful discussion, Carlos Lorenzana for technical support in histopathological processing, and Cynthia Schriver for editorial assistance. This work was made possible by a grant awarded by NIH to KK.

Author information

Correspondence to Kamel Khalili.

Rights and permissions

Reprints and Permissions

About this article


  • human virus
  • brain tumors
  • transgenic mice

Further reading