|Home | About | Journals | Submit | Contact Us | Français|
JC virus (JCV) is a human neurotropic polyomavirus whose replication in the central nervous system induces the fatal demyelinating disease, progressive multifocal leukoencephalopathy (PML). JCV particles have been detected primarily in oligodendrocytes and astrocytes of the brains of patients with PML and in the laboratory its propagation is limited to primary cultures of human fetal glial cells. In this short communication, the development of a new cell culture system is described through the fusion of primary human fetal astrocytes with the human glioblastoma cell line, U-87MG. The new hybrid cell line obtained from this fusion has the capacity to support efficiently expression of JCV and replication of viral DNA in vitro up to 16 passages. This cell line can serve as a reliable culture system to study the biology of JCV host cell interaction, determine the mechanisms involved in cell type specific replication of JCV, and provide a convenient cell culture system for high throughput screening of anti-viral agents.
The lack of a convenient and reliable cell culture system has significantly hampered the ability to study the mechanisms involved in the life cycle of the human polyomavirus, JC virus (JCV), the causative agent of the demyelinating disease, progressive multifocal leukoencephalopathy (PML) (Khalili et al., 2003; Major et al., 1995). Replication of JCV in vitro takes place on an average of three or more weeks in culture (Radhakrishnan et al., 2003). As such, propagation of wild type and mutant viral stocks is a labor intensive and time consuming undertaking. In comparison, the life cycle of the highly related simian virus 40 (SV40) is significantly shorter with death of host cells evident in as little as 24 h post infection. The availability of several kidney epithelial cell lines including CV-1, for cultivation and study of SV40 have helped to decipher the mechanisms involved in gene regulation, viral replication, and virus-host interaction. Unlike SV40, knowledge about the pathways involved in the life cycle of JCV remains more limited, mainly due to the absence of a continuous well characterized cell line that permits efficient replication of JCV. In the laboratory, replication of JCV has been limited to primary cultures of human fetal glial cells and thus far no small animal model that supports viral replication has been identified (for review, see Khalili et al, 2003). The difficulties associated with obtaining human fetal brain tissues for the preparation of primary cultures, which are costly, labor intensive, and can vary in purity from preparation to preparation, have prompted several investigators to develop cell lines including SVG and POJ that support the JCV infection cycle (Mandl et al., 1987; Major et al., 1985; Frye et al., 1997). However, the utility of these lines for studying initial events that stimulate viral gene expression and replication is limited as the culture systems are transformed with either SV40 (SVG-A) or JCV (POJ) genomes and constitutively express the T-antigens of these viruses. Thus, the constitutive presence of T-antigen in these cells bypasses the immediate early events in the JCV infection cycle including activation of the early promoter and expression of T-antigen. To overcome this issue, an alternative strategy of cell fusion was employed between permissive primary human fetal astrocytes and the non-permissive human glioblastoma cell line, U-87MG and subsequently several hybrid cell lines were developed to study JCV life cycle.
To create hybrid cell lines, a PEG-mediated cell fusion method with negative selection by HPRT between primary human fetal astrocytes (PHFA) and HPRT deficient U-87MG cells was employed. The efficiency of fusion was evaluated by the appearance of bi-nucleated cells that had a distinct flattened morphology two hours after cell fusion (Figure 1). At twenty-four hours, the morphology of the cells was more elongated and spindle-shaped, and the hybrid clones retained this morphology thereafter (Figure 1). The clones were plated in selective media (HAT-medium) to confirm that they were deficient in HPRT activity (data not shown). Two clones, named HC-7 and HC-15, were selected for further study from a total of thirty-two.
Examination by Western blot analysis of whole cell extracts of the astrocyte marker protein, glial fibrillary acidic protein (GFAP) revealed that, unlike PHFA, which expresses GFAP, neither of these proteins were detected in extracts from U-87MG cells, or in the hybrid cells, HC-7 and HC-15 (Figure 2, Panel A). To compare the growth properties and DNA content of the hybrid cells with the parental cells flow cytometry assay was performed. By determining the cell cycle profile of the clones, the distribution of the cells at G0/G1, S, and G2/M phases and determined the percentage of apoptotic cells was determined. As shown in Figure 2 (Panel B), a modest decline in the cell distribution at G0/G1 of the hybrid clones compared to their parental cells was noted. At the G2/M stage, the profile of the hybrid clones appear to be closer to PHFA than U-87MG (HPRT-) cells. Furthermore, a larger number of apoptotic cells was noted in clonal cells than those in the parental cells as was observed, apoptosis was detected in HC-15 (19.1%) versus PHFA (10.4%) and U-87MG (9.7%). These observations showed a subtle variation in the cell cycle distribution of the U-87MG and PHFA cells upon their fusion. However, no significant differences in the doubling times of the parental and the hybrid cells were observed, which ranged between 27 and 30 hours (data not shown).
To examine the infectivity of these cells with JCV, hybrid cells at passage six after fusion, along with PHFA and U-87MG parental cells were infected with JCV Mad-1 at a multiplicity of infection (MOI) of one. At 7 and 14 days post-infection, cells were harvested and examined for replication of viral DNA and expression of viral proteins by Southern and Western blot analyses, respectively. As expected viral DNA and the early and late viral proteins were detected in infected PHFA but not U-87MG cells (Figure 3, compare lanes 1-3 and 4-6). Viral DNA and proteins were detected in both HC-7 and HC-15 cells at days 7 and 14 after infection (lanes 8 and 9, and 11 and 12). While the levels of viral gene expression and production of T-antigen and VP1 in HC-7 and HC-15 were comparable to those from PHFA, the level of viral DNA replication in HC-15 was noticeably higher than in the PHFA and HC-7 cells. Expression of the unrelated cellular protein, Grb2, served as a control for equal protein loading of the Western blot. Further studies are required to understand the molecular events involved in high efficiency of JCV DNA replication in HC-15 cells.
Immunocytochemical staining of viral proteins in JCV infected cells was determined at day 8 post infection. As seen in Figure 4, consistent with previous observations (Radhakrishnan et al, 2003), T-antigen and VP1 were detected predominantly in the nucleus of PHFA and HC-7 and HC-15. Further, the late auxiliary protein, agnoprotein, which is critical for viral replication (Akan et al., 2006; Khalili et al., 2005; Sariyer et al., 2006) displayed a mostly perinuclear appearance in both PHFA and the hybrid cells. These observations indicate that similar to primary human fetal astrocytic cultures, the hybrid cells have the capacity to efficiently support viral gene expression and replication.
To test the longevity of the hybrid cells in supporting JCV replication, HC-15 cells at various passages ranging from 6 to 26 were infected with JCV and replication of viral DNA and expression of the viral late protein were determined at day 8 post-infection. As shown in Figure 5, replication of JCV DNA and expression of VP1 was detected in passage 6, 11 and 16, but not 26, suggesting that the hybrid cells lose their ability to support JCV replication after repeated passaging in culture, i.e. >16 passages. Similar results were obtained with HC-7 cells (data not shown). To rule out the involvement of receptors in the loss of permissiveness of high passage hybrid clones, in the next series of experiments, cells were transfected with JCV DNA and replication of the DNA and expression of viral late proteins was determined. As shown in Figure 5C, consistent with the findings from Figure 5A, the replication of the viral genome in the cells at passage 16 is barely detectable and completely abrogated by passage 26, supporting the notion that the loss of the replication of the viral genome in high passage hybrid lines is not due to the loss of receptors or any other barriers imposed during the viral entry process. In addition, results from Western blotting showed complete disappearance of viral late protein expression by passage 26. Altogether, these results show that fusion of PHFA and U-87MG produces cell lines that are permissive for JCV for several passages, at least up to 16.
After attachment of JC virus to the cell membrane and endocytosis, viral DNA enters the nucleus of the infected cells where it can be stimulated by specific transcription factors to produce the early gene product, T-antigen. T-antigen then recruits cellular proteins to stimulate viral DNA replication and late gene expression that results in the production of progeny capsids. In addition, several studies have ascribed a critical role for Agnoprotein in successful virus production (Johannessen et al., 2008; Matoba et al., 2008; Sariyer et al., 2006, 2008). The major distinction between the more ubiquitous SV40 and JCV rests on the tissue specificity of JCV gene expression, which is more restricted. In addition, SV40 and JCV exhibit a high degree of species specificity with viral replication restricted to the natural host, i.e. simian and human.
One of the main drawbacks of hybridoma cell technology is the instability of the genome of hybrid clones (Duelli et al., 2005, 2007; Ganem et al., 2007; Hernandez et al., 1996). It is possible that the gradual loss of JCV replication at the higher passages of the hybrid lines reflects genomic instability of the hybrid clones. It should be noted that JCV proteins, T-antigen and Agnoprotein have been shown to interfere with DNA repair machinery and to induce chromosomal instability (White and Khalili, 2005; Khalili et al., 2003, 2005). Thus, the sensitivity of the hybrid clones to chromosomal damage may be enhanced upon JCV infection, leading to the loss of genes upon repeated passaging that are responsible for expression of specific factors that potentiate the JCV genome. Another possibility for restricted JCV replication at higher passages, though not mutually exclusive, may come from the activation of cell-type specific inhibitors of viral gene expression. This possibility is consistent with earlier fusion studies in which mouse fibroblasts were fused with transformed hamster glial cells and the expression of JCV large T-antigen was lost in the hybrid cells (Beggs et al., 1988; 1990). These observations sugget that expression of negative regulatory protein(s) in the hybrid cells can suppress JCV gene expression and replication. Further studies are required to understand the mechanism for extinction of JCV in the hybrid cells.
In addition to glial origin cells including primary human fetal astrocytes and the T-antigen transformed glial origin cell lines POJ and SVG described above, some strains of JCV have been able to grow in non-glial cell culture systems (Khalili et al., 2003). These include primary cultures of schwann cells and neuroblastoma cell lines, as well as kidney epithelial cells (Akatani et al., 1994; Assouline and Major 1991; Hara et al., 1998). In addition, JCV has been shown to replicate in hemotopoietic progenitor cells, B lymphocytes, and tonsillar stromal cells with varying efficiency (Monaco et al., 1996). Of note, JCV infection does not appear to be restricted at the level of binding and cell entry (Ashok and Atwood, 2006). The cell lines capable of supporting JCV infection have all been of human origin, with a few notable exceptions being of simian origin (Hara et al., 1998). These findings are consistent with the high degree of species specificity seen with the polyomavirus family and also with the natural history of JCV infection in humans.
As an alternative to primary cultures of human fetal astrocytes, hybrid human cell lines were produced by fusing primary human fetal astrocytes with the human malignant glioblastoma cell line, U-87MG by negative selection using HPRT. Several JCV-permissive hybrid cell lines were identified and two of these clones were examined in detail with respect to hybrid properties and viral propagation. Both hybrid cell lines and PHFA cells were infected and compared with respect to viral transcription and replication. It was demonstrated that the newly-generated hybrid cell lines support JCV replication and viral protein production at multiple passages. The hybrid lines supported JCV replication and propagation at early passages but lost that ability at late passages, suggesting that these cells may provide a system for identifying positive and negative regulators of viral transcription and replication. In addition, these newly-generated hybrid cell lines may serve as good tissue culture model system for JCV replication and propagation and may help us develop more effective strategies to prevent JCV infection in glial cells.
Thank you to past and present members of the Department of Neuroscience and Center for Neurovirology for sharing of ideas and reagents and Cynthia Schriver for preparation of the manuscript. This work was supported by grants awarded by NIH to KK.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.