Summary

The human polyomavirus JCV shows significant neurotropism to cells from the central nervous system (CNS) and productively infects human oligodendrocytes, the myelin-producing cells of the brain. Like other polyomaviruses, this virus can cause tumors when intracerebrally inoculated at high titer into developing rodents. JCV is the only polyomavirus to cause brain tumors in nonhuman primates. Tumori-genecity is most likely induced by the viral early gene product, T-antigen, as is demonstrated by the occurrence of tumors in T-antigen transgenic animals. T-antigen may well interact with p53 and the pRb family of proteins, and resulting in the stimulation of S-phase specific genes via E2F-1. The possibility that JCV may be involved in human brain tumors has been raised both by the occurrences of these tumors coincident with productive JCV infection of the brain leading to the demyelinating disease Progressive Multifocal Leukoencephalopathy. The recent findings of both JCV DNA and T-antigen in human brain tumor samples points to a strong association of JCV with some forms of human brain tumors, namely medulloblastoma.

Similar to other polyomaviruses, the JCV genome consists of a closed, circular, supercoiled DNA molecule that is 5130 nucleotides in size [1]. The DNA of the virus can be divided into three functional domains: a viral early region encoding the tumor antigens, large and small T-antigens, which are expressed throughout viral infection [2]; a viral late region which encodes the three major capsid proteins, VP1, VP2 and VP3 during the late phase of the lytic cycle; and a noncod-

ing regulatory region located between the early and late regions which encompasses several transcription regulatory modules and the viral origin of DNA replication (Fig. 1). Viral proteins are encoded on both DNA strands and are transcribed divergently from the central promoter region. Large and small T-antigens are produced by alternative splicing of a single transcript from the early region which share a common amino terminus. An additional open reading frame located in the leader of the early RNA with the capacity to produce a small peptide termed JELP (JCV early leader protein) is thought to be expressed from the early region. VP1 is encoded by the 3'end of the late region, while VP2 and VP3 are translated from a common mRNA at the 5'end of the late region. Transcripts from the late promoter encode a polypeptide of 71 amino acids contained within the 5'region of the late transcripts termed agnoprotein. The viral regulatory region of the prototype Mad-1 strain of JCV consists of two exact 98-base pair (bp) repeats which contain cis elements required for expression of viral genes [3], This region also contains an origin of viral DNA replication located to the early side of the 98-bp repeats. A TATA box has been identified within each 98-bp repeat. Subsequent viral isolates have shown that the regulatory region is heterogeneous in different strains, and several isolates do not contain exact 98-bp repeats

The lytic cycle of JCV in primary human fetal glial cells begins with the synthesis of RNAs initiated from the early side of the viral genome during early post-infection and prior to DNA replication. DNA replication occurs by day 5 post-infection and continues for more than 15 days. At day 5, the synthesis of

Figure 1. Genomic structure of the human polyomavirus, JCV. Schematic representation of the JC virus genome depicting coding regions for the early genes, large and small T-antigens (gray) and the late genes, capsid proteins VP1, VP2, VP3 and the agnoprotein (black). The coding regions are separated by a bi-directional promoter containing the 98 base pair sequence elements and the origin of viral DNA replication.

Figure 1. Genomic structure of the human polyomavirus, JCV. Schematic representation of the JC virus genome depicting coding regions for the early genes, large and small T-antigens (gray) and the late genes, capsid proteins VP1, VP2, VP3 and the agnoprotein (black). The coding regions are separated by a bi-directional promoter containing the 98 base pair sequence elements and the origin of viral DNA replication.

late RNAs initiates and continues for 10-15 days [5], The lytic cycle of JCV, like that of other papovaviruses such as SV40, depends on the presence of a functional T-antigen, which through binding to the multiple DNA binding sites near the origin of DNA replication: (i) mediates down regulation of the level of early gene expression through a repressor-like function; (ii) stimulates the initiation of viral DNA replication; and (iii) both directly and indirectly activates the late transcriptional processes. The role of JCV T-antigen in stimulating viral DNA replication has been directly investigated; however, its function in autoregulation of the early promoter remains to be established. Recently, we have shown that the interplay between the viral early protein, T-antigen, and the cellular proteins, PurĀ« and YB-1, may determine the level of viral early and late promoter activity [6,7]. Furthermore, we have demonstrated that overexpression of Puree in glial cells down-regulates replication of JCV DNA [8],

In vivo studies have indicated that human oligodendrocytes are the only cells that are productively infected by JCV [9], Although human cells from em bryonic lung, intestine, liver and testis do not support JCV replication, previous studies have demonstrated that JCV DNA can replicate in any primate cell type, provided that JCV or SV40 large T-antigen is en-dogenously produced in those cells [10], However, replication of JCV DNA in primate cells requires prior expression of JCV large T-antigen which appears to be regulated at the transcriptional level [11, 12]. The host-range restriction of JCV at the early stage of infection can be determined on at least two levels: (1) tissue-specific transcriptional regulation of viral RNA synthesis; and (2) species-specific replication of the viral DNA. Whether or not there are additional restrictions in the viral life-cycle that contribute to the cell-type specificity of the virus is unclear. Although the involvement of the initial stages, i.e. adsorption, penetration and uncoating, do not exhibit tissue specificity, restriction in later stages such as viral late gene transcription and capsid protein synthesis have yet to be investigated. Although the precise mechanism responsible for the restricted transcriptional activity of JCV to glial cells has not been well identified, studies from several laboratories including our laboratory, have indicated that the JCV control region has an enhancer function that is glial cell-specific [3, 11-13]. This property of host-cell specificity of an enhancer apparently plays a key role in determining the hostrange of JCV and may also be a critical factor in determining the disease potential of this virus [14].

The initial observations that polyomaviruses had tumorigenic potential came from experiments in which the viruses were inoculated into animal species unrelated to the natural hosts. In the case of JCV, hamsters inoculated intracerebrally, intraocularly, in-traperitoneally, or subcutaneously with the Mad-1 strain of JCV developed neuroectodermal tumors (Fig. 2). Many of these appear to derive from neuronal elements (medulloblastomas, pineocytomas and neuroblastomas) while other (glioblastomas) are probably glial in origin [15-17]. Similarly, nude mice injected intracerebrally with JCV develop primitive neuroectodermal tumors [Gordon et al., unpublished observations]. JCV represents the only polyomavirus that induces tumors in nonhuman primates [18]. As is the case in rodents, intracerebral inoculation of JCV in owl and squirrel monkeys results in malignant tumors with both neuronal and glial components [19, 20]. Control animals inoculated with SV40 or human pa-povavirus BKV were seropositive for viral antigen, but did not develop tumors. Like Mad-1, other strains of JCV exhibit a high incidence of tumor development. Evidently, Mad-4 virus predominantly caused tumors of pineal gland origin [21], whereas a strain isolated in Tokyo produced cerebellar medullablastomas in hamster and undifferentiated neuroectodermal tumors in cerebra of rats [22].

There are certain obvious limitations to these rodent inoculation models in understanding the pathogenic mechanism through which this virus might act in humans. Since JCV does not replicate efficiently in rodents, tissue tropism, the pattern of JCV dissemination in the animal, and the molecular pathways surrounding replication are different from humans. Although the injection of high titers of virus, often directly into the CNS, circumvents many barriers, it clearly does not provide a true parallel to the viral life-cycle in humans. Some of these limitations have been circumvented by investigating the ability of JCV or its gene products to transform cells in culture, as both human and hamster glial cells can be transformed with JCV T-antigen [21, 23], As shown in Fig. 3, JCV T-antigen transformed hamster cells can generate tumors when injected sub-cutaneously in nude mice. Since polyomaviruses induce tumors most readily in nonpermissive cells, it has been assumed that the transformation is dependent upon expression of polyoma early gene product accumulation of tumor antigen past a certain threshold level sufficient to alter the normal cell cycle. Thus, it is not surprising that normal cells transformed by JCV in culture express T-antigen. Similarly, cells taken from JCV owl monkey gliomas also express T-antigen in vitro [24,25], and following explantation into the nude mouse flank [Gordon, unpublished observations].

Other limitations of the direct viral inoculation model have been overcome by creating transgenic mice that constitutively produce T-antigen. When Tantigen is produced under the control of the Mad-1 early promoter/enhancer, mice develop adrenal neuroblastomas and neuroectodermal origin tumors [26, 27], Mice that express T-antigen under the control of the archetype (98 bp without repeats) promoter/enhancer [4] developed purely CNS neuronal tumors resembling medulloblastomas, one of the most common malignancies of childhood [28], These results suggest that JCV T-antigen is directly associated with tumorigenesis in these animal models. They also suggest that promoter/enhancer elements may exert tissue-specific transcriptional control on viral RNA

synthesis thereby determining the cell types that develop pathology in vivo.

There are numerous reports linking polyomaviruses (JCV, BKV, SV40) with human tumors. It is of interest that all these viruses have been associated with brain tumors, emphasizing their neurotropism. In order to establish this association, one must reliably detect the virus in the tumor under study and show that its presence is not incidental to the pathologic process. In the setting of lytic JCV infection of human oligodendrocytes, primary CNS tumors have been documented on several occasions (for a review see [29]). There have been several documented cases of patients with central nervous system neoplasms and concomitant progressive multifocal leukoen-cephalopathy (PML). The first of these, reported in 1961 [30] was a 58 year-old man with chronic lymphocytic leukemia who at autopsy was found to have PML and an incidental oligodendroglioma. In another neoplasm, discovered at autopsy in the brain of a patient with PML, electron microscopy revealed polyomavirus in one tumor cell [31], In a case of PML in a patient with concomitant gliomas, JCV was detected using immunofluorescence, hemagglutination and electron microscopy [32]. A case of atypical PML and primary cerebral malignant lymphoma was reported by GiaRusso and Koeppen [33]. In that patient, no intranuclear inclusions were present, nor was virus detectable by immunofluorescence or electron microscopy.

Since PML generally occurs in immunosuppressed individuals, its has seemed logical to document the presence of JCV DNA in the immunosuppressed population which is PML-free. Thus, using PCR, JCV DNA has been found in more than 30% of CNS tissue and peripheral blood lymphocytes of HIV-1 positive individuals with PML [34, 35]. In normal individuals, similar techniques have detected JCV DNA in CNS samples, urine and peripheral blood lymphocytes [36-39], Two recent studies have also detected JCV in primary human CNS tumors. In one, nested PCR revealed genome sequences of the LTR and VP1 regions of JCV in a pleomorphic xanthroastrocytoma from a 9-year-old that were not found in the patient's peripheral blood [40], Sequence analysis showed a mutated region most consistent with the Mad-4 variant which has previously demonstrated oncogenic potential in animal models. Another report [41] describes an immunocompetent individual with an oligoastrocy-

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook


Post a comment