- Email: [email protected]
Expression of the E6 and E7 Genes of Human Papillomavirus (HPV16) Extends the Life Span of Human Myoblasts
Experimental Cell Research 248, 186 –193 (1999) Article ID excr.1999.4407, available online at http://www.idealibrary.com on
Expression of the E6 and E7 Genes of Human Papillomavirus (HPV16) Extends the Life Span of Human Myoblasts Hanns Lochmu¨ller,* Timothy Johns,* and Eric A. Shoubridge* ,† ,1 *Montreal Neurological Institute and †Department of Human Genetics, McGill University, Montreal, Quebec, Canada H3A 2B4
Primary human myoblasts (satellite cells), like other human cells, have a limited life span in vitro. Here we show that expression of the E6E7 early region from human papillomavirus type 16 can greatly extend the life span of both fetal and satellite cell-derived myoblasts and release them from dependence on the growth factors normally necessary for their proliferation. Expression of either the E6 or the E7 gene alone was not sufficient to confer this phenotype, although expression of E7 did delay cellular senescence. The steady-state level of E6E7 transcripts in clonal cultures correlated with proliferative capacity and inversely with the capacity to differentiate into multinuclear myotubes. The expression of E7 alone markedly inhibited cell fusion in both adult and fetal cultures. These effects on myoblast differentiation could be related in part to the level of retinoblastoma protein (pRb), the major cellular target of E7. Terminal differentiation of skeletal myoblasts is associated with permanent withdrawal from the cell cycle; however, continued expression of E6E7 in differentiated myotubes permits reentry of myotube nuclei into S phase in response to growth factor stimulation. These results support a key role for pRb in the acquisition and maintenance of the differentiated state in human skeletal muscle and, in cooperation with p53, in the control of proliferative capacity and response to external growth factors. © 1999 Academic Press Key Words: myoblasts; papillomavirus; immortalization; cell cycle; retinoblastoma protein.
INTRODUCTION
Satellite cells are dormant myoblasts, located between the plasma membrane and basal lamina in skeletal muscle, that can be induced to reenter the cell cycle in response to signals associated with growth and repair of skeletal muscle. They can be cloned [1] and propagated in vitro; however, like other mammalian cells they have a limited life span [2]. Although several immortal rodent myoblast cell lines are in common use, 1
To whom correspondence and reprint requests should be addressed. Fax: (514) 398-1509. E-mail: [email protected]. 0014-4827/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
immortal human myoblast lines have not been isolated. The molecular basis for this species difference is unclear but it may reflect the presence of more stringent cell cycle regulation in human cells. Rodent myoblasts can be readily immortalized with genes from DNA tumour viruses, such as the SV40 large T antigen [3], which targets p53 and retinoblastoma protein (pRB), key components of the cell cycle regulatory machinery at the G1/S checkpoint. Human myoblasts stably transfected with a plasmid driving inducible expression of the SV40 large T antigen are reported to have extended lifetimes, but are not immortalized [4, 5]. Infection of human myoblasts with polyoma large T antigen or a retroviral vector expressing a temperature-sensitive SV40 large T antigen [6] also failed to immortalize at the permissive temperature (E. Shoubridge, unpublished results). Origin-defective SV40 virus has been used to immortalize human myoblasts, but the cells retain little capacity to differentiate [7]. Human papillomaviruses (HPVs) are DNA viruses that normally infect squamous epithelial cells [8]. Over 60 different types are known and the so called “high risk” species such as HPV16 and HPV18 are often associated with cervical carcinomas. Previous studies have shown that expression of the E6E7 early region from HPV16 can efficiently immortalize human primary fibroblasts [9], epithelial cells [10, 11], keratinocytes [12], and smooth muscle cells [13] and that expression of the E7 gene alone is sufficient to immortalize some human epithelial cell types [11] and keratinocytes [12], albeit at a lower frequency. The principal cellular target of the E7 gene is the hypophosphorylated form of pRb which binds E7 in the conserved pocket region, a property shared with other DNA tumor virus genes such as SV40 large T antigen and adenovirus E1A [8]. This relieves the pRb-mediated block at the G1/S boundary, permitting the cell to enter S phase. E6 targets the p53 tumor suppressor gene, abrogating its transcriptional transactivation activity, by targeting p53 destruction in the ubiquitindependent proteolytic pathway [8]. In this study we tested whether expression of the HPV16-E6E7 genes could immortalize primary human myoblasts. Using recombinant retroviral vectors we
186
EXPRESSION OF HPV16-E6E7 IN HUMAN MYOBLASTS
introduced the E6 and E7 genes alone and in combination into human primary adult and fetal myoblasts and show that expression of the HPV16 E6 and E7 genes together significantly extends the life span, but does not result in true immortalization of either fetal or adult primary myoblasts. We also show that the differentiation potential of E6E7 transformed myoblasts is negatively correlated with the steady-state level of E6E7 transcripts in clonal cultures and that myonuclei in terminally differentiated myotubes expressing E6E7 can be induced to reenter the cell cycle in response to exogenous growth factors. MATERIAL AND METHODS Cell culture. Adult and fetal human myoblasts were obtained from biopsy material. Satellite cells were dissociated from the biopsy specimen by trypsinization and grown in supplemented Ham’s F10 medium containing 15% fetal bovine serum (FBS), hereafter referred to as growth medium, as in [14]. Myoblasts were labeled with an antibody to NCAM and purified by fluorescence-activated cell sorting (FACS) as previously described [14]. This routine procedure typically results in myoblast purities .99.5% as determined by postsorting FACS analysis. Fusion and differentiation of myoblasts were induced by replacing the growth medium with Dulbecco’s modified Eagle’s medium (DMEM) containing 2% horse serum (fusion medium) for 48 h. Myoblasts were grown in DMEM containing 10% FBS to test for dependence on exogenous growth factors. To monitor entry of nuclei into S phase, bromodeoxyuridine (BrdU) at a concentration of 0.1 mM was added to the medium. To visualize myoblast clones, cells were fixed in methanol at room temperature for 15 min on the culture plate and stained with 0.1% crystal violet for 15 min. Retroviral constructs and infection. Retroviral constructs, carrying the E6 region, the E7 region, or the E6E7 region of human papillomavirus type 16, were a kind gift of Dr. Denise Galloway (Fred Hutchinson Cancer Center, Seattle, WA) [10]. Packaging cells (PA317) were grown to near confluence in DMEM containing 10% FBS; 24 h later the medium containing retroviral particles was harvested and filtered through a 0.4-mm filter. Retroviral titers were determined by infecting NIH/3T3 cells. The 24-h supernatant (1 or 2 ml) was mixed with 2 or 3 ml of DMEM containing 20% FBS and polybrene (4 mg/ml). Human myoblasts (;10 5) in 100-mm tissue culture dishes were incubated with this medium for 2 h after which an additional 5 ml of medium was added to the dish and the incubation was continued from 3 h to overnight. The cells were washed once in fresh medium to remove the polybrene, and 48 h later selection for retroviral integration was performed in growth medium containing 400 mg/ml geneticin (G418) (GIBCO, Burlington, Ontario, Canada) and continued for at least 14 days. Individual G418-resistant myoblast clones were picked, expanded, and analyzed immediately. RNA extraction and Northern blots. Myoblasts were grown to confluence in 100-mm tissue culture dishes. Total RNA was extracted with an RNAeasy kit according to the manufacturer’s recommendations (Qiagen, Mississauga, Ontario, Canada). Electrophoresis was performed in 1% agarose gels containing formaldehyde. Gels were transferred to nylon membranes (Zeta-probe, Bio-Rad, Mississauga, Ontario, Canada) by vacuum blotting and UV-crosslinked following standard procedures. DNA fragments were labeled by random prime labeling (Stratagene, La Jolla, CA) with [a- 32P]dCTP. Hybridization was performed overnight. The following probes were used: as a specific probe for the E6E7 region of human papillomavirus type 16 (E6E7 probe) a 850-bp EcoRI/BamHI fragment from pLXSN16E6E7; for the E6 region (E6 probe) a 310-bp NsiI fragment from pLXSN16E6E7; and for the E7 region (E7 probe) a 310-bp
187
NsiI/BamHI fragment from pLXSN16E6E7. An 18S rRNA probe was used as an internal reference. After hybridization the membranes were washed for 1 h in three steps with buffers of increasing stringency. The final wash was done at 50°C (0.1% SDS, 0.53 SSC). Membranes were exposed overnight and analyzed on a PhosphorImager using the ImageQuant software (Molecular Dynamics, Sunnyvale, CA). Immunoblotting. Myoblasts and myotubes were harvested from 100-mm tissue culture dishes in cold PBS. Cells were taken up in Laemmli sample buffer and protein from approximately equal numbers of cells (determined by degree of confluency) was loaded on 7% polyacrylamide gels. Electrophoresis and Western blotting were done on the Bio-Rad miniblot apparatus. Membranes were incubated overnight at 4°C with the following primary antibodies: a monoclonal antibody raised against human pRb (1:500 dilution, clone G3-245; PharMingen, Mississauga, Ontario) and a monoclonal antibody against quail embryonic myosin heavy chain (MHC) (1:100 dilution, 47A; a kind gift by Dr. P. Merrifield, University of Western Ontario) [15]. Specific bands were detected using a 1:1000 dilution of a secondary antibody (anti-mouse IgG) coupled to HRP using enhanced chemiluminescence (ECL kit, Amersham, Oakville, Ontario, Canada) according to the manufacturer’s recommendations. Immunocytochemistry. Myoblasts, seeded on glass coverslips, were induced to differentiate in medium containing 2% horse serum for 48 h when nearly confluent. To visualize nuclei in S phase, the medium was changed to myoblast growth medium containing BrdU (0.1 mM) for 24 h. Coverslips were washed with PBS and fixed in 70% ethanol for 30 min at 220°C and then incubated in 2 N hydrochloric acid for 10 min, followed by 0.1 M sodium borate, pH 9, for 10 min. Coverslips were incubated in a humidified chamber with blocking buffer (10% horse serum in PBS) for 30 min, followed by incubation with the primary antibody in blocking buffer for 2 h. The following primary antibodies were used: a 1:20 dilution of a monoclonal antibody against BrdU (B44, Becton Dickinson, Franklin Lakes, NJ) or a 1:20 dilution of a monoclonal antibody against embryonic myosin heavy chain (Novocastra, Newcastle upon Tyne, UK). Nuclei were counterstained with bisbenzimide (Hoechst stain H33342, Calbiochem, La Jolla, CA) at a concentration of 1 mg/ml. The secondary antibody, a 1:100 dilution of biotinylated antimouse IgG raised in horse (Vector, Burlingame, CA), was applied for 45 min. Specific binding was detected with Cy3-conjugated streptavidin (1:1000; Jackson ImmunoResearch, West Grove, PA) for 20 min. Telomerase assays. Telomerase was assayed using the TRAP assay as in [16]. Glioblastoma cell lines were used as positive controls.
RESULTS
Expression of HPV16-E6E7 Extends the Life Span of Human Myoblasts FACS-sorted adult human myoblasts were infected with recombinant retroviral vectors expressing a neomycin resistance gene and the E6 or E7 genes from HPV16, alone or in combination. All retroviral infections were done with an infectious titer of ;10 5/ml. After selection in G418, resistant clones were pooled, passaged twice (1:100) at near confluence, and then plated at 1:1000 (;1000 cells/100 mm plate) to test for clonal proliferative capacity and growth factor dependence (Fig. 1). A large number of colonies was observed from cells that had been transduced with constructs expressing either E6E7 (;800 colonies) or E7 (;500 colonies) and plated in growth medium, while control uninfected myoblasts and cells transduced with E6
188
¨ LLER, JOHNS, AND SHOUBRIDGE LOCHMU
FIG. 1. Analysis of clonal proliferative potential and growth factor dependence in adult human primary myoblasts expressing the E6 and E7 regions of HPV16 alone or in combination. The plates in the top row were cultured in full supplemented growth medium; those in the bottom row in DMEM plus 10% FCS. Cells were transduced with retroviral constructs expressing the E6 or E7 genes as indicated.
alone produced few colonies (Fig. 1). Clones were tested for growth factor dependence by plating in medium that does not support proliferation of primary adult human myoblasts (DMEM plus 10% FCS). Only those clones expressing E6E7 proliferated under these conditions (Fig. 1). Clones expressing E6E7 were selected and passaged in growth medium for approximately 40 more population doublings (60 total). Of the 12 clones tested, 2 grew vigorously and did not exhibit any loss of viability or growth potential, 2 ceased growing and became senescent, and the remaining 8, although still proliferating, showed reduced population doubling times. This behavior correlated with the expression of E6E7 which showed a fivefold variation in the steady-state level of E6E7 by Northern blot analysis (Fig. 2). Clones with the greatest proliferative potential (no decrease in population doubling time) expressed the highest levels of E6E7 (Fig. 2, lanes 6 and 7); the lowest levels were expressed in senescent clones (Fig. 2, lanes 8 and 11). Parallel cultures of uninfected myoblast clones or clones infected with constructs transferring the E7region alone showed signs of reduced proliferative potential after 25–30 population doublings and all senesced shortly thereafter. The two vigorously growing clones (Fig. 2, lanes 6 and 7) were passaged further and both abruptly senesced after 4 more passages (12 population doublings). Fusion competence was assessed in the same 12 clones. Clones with the highest expression of E6E7 showed poor fusion with only a few scattered, small myotubes. However, clones with intermediate levels of expression showed variable fusion competence; some (Fig. 2, lane 9) fused very well and others
with similar E6E7 expression (Fig. 2, lane 5) showed modest fusion. Proliferative potential was also assessed in FACSsorted fetal human myoblasts expressing HPV16E6E7. G418-resistant clones were recloned twice and passaged in growth medium for several months. The cells proliferated without major changes in growth or doubling time until passage 38 (total population doublings 130 –140), when they abruptly senesced. Uncloned adult and fetal E6E7 cells were maintained for several weeks in growth medium after they had ceased growing (crisis); however, no foci of immortal cells emerged. Telomerase activity, measured by the TRAP assay, was not detectable in any of the adult or fetal E6E7-transformed cells (data not shown). Differentiation of Fetal Human Myoblast Clones and pRb Expression pRb is thought to play a key role in the differentiation of skeletal muscle through its interaction with the MyoD family members [17]. To test whether the expression of HPV16 E6 or E7 affected the capacity for differentiation of uncloned primary human myoblasts we first examined cell fusion and MHC expression in pooled fetal myoblasts expressing E6, E7, or E6E7 (Fig. 3). Large multinuclear myotubes expressing MHC, indistinguishable from control uninfected, uncloned cultures, were observed in cells expressing E6 or E6E7 60 h after transfer to fusion medium. Fusion was, however, markedly inhibited in cells expressing E7 alone where only a few scattered myotubes were observed, most with only two to three nuclei (Figs. 3C and
EXPRESSION OF HPV16-E6E7 IN HUMAN MYOBLASTS
189
FIG. 2. Northern blot analysis of E6E7 expression in adult myoblast clones. The 18S rRNA was used as a control for RNA loading. Proliferative capacity and fusion were scored semiquantitatively as follows: for proliferative capacity; 0, senescence; 1, slow growth, long population doubling time; 11, good growth, average population doubling time; 111, rapid growth, short population doubling time. For fusion: 0, poor fusion; 1, some fusion; 11, good fusion; 111, very good fusion. Fusion could not be assessed in the two clones that senesced (2, lanes 8 and 11). The steady-state ratio of E6E7 mRNA:18S rRNA levels is shown below the fusion index.
3G). Northern blot analysis with E6- and E7-specific probes showed that the difference between cells expressing E7 alone or E6E7 could not be attributed to a difference in the steady-state level of transcripts containing the E7-region (data not shown). To further examine the basis for this difference in the capacity for differentiation, pRb expression, fusion competence, and MHC expression were measured in fetal myoblast clones expressing E6E7. There was a wide range in differentiation potential in different clones; some fused readily and formed large multinucleate myofibers and others exhibited little fusion. Immunoblotting on these clones revealed that MHC expression was positively correlated with immunodetectable pRb protein (Fig. 4A). Clones with detectable pRb all expressed MHC, while those with no detectable pRb did not. As a control, uninfected fetal myoblast clones were analysed for fusion/differentiation capacity and pRb expression. As with E6E7-expressing clones, some fused rapidly and formed large multinucleate myofibers and others fused more slowly and less extensively. Immunoblotting on these clones revealed that differentiation capacity as measured by MHC expression also correlated to some degree with pRb expression (Fig. 4B). Compared to uninfected fetal myoblast clones, E6E7-expressing clones showed reduced fusion and lower levels of detectable pRb protein. Although we did not carry out careful quantitation, the increased pRb expression in control cells is clear from the pRb blots in Figs. 4A and 4B which were exposed for 5 s and 3 min, respectively. pRb levels were lower in differentiated myotubes from all clones, regardless of their fusion capacity.
E6E7-Expressing Myoblasts Can Reenter the Cell Cycle after Differentiation The continued presence of pRb is thought to be important for keeping the myonuclei in postmitotic myotubes in the G 0 phase of the cell cycle. Myotubes derived from pRb null mice can reenter the cell cycle in response to exogenous growth factors [18] and expression of the large T antigen from SV40 (which targets pRb and p53) in rodent myotubes can trigger new DNA synthesis [3]. To test whether nuclei in myotubes derived from clones expressing E6E7 could be induced to reenter the cell cycle, differentiated cultures containing large multinucleate myotubes were exposed to growth medium containing BrdU to label nuclei in S phase. After 24 h in this medium a large number of nuclei in myotubes could be detected with the antiBrdU antibody (data not shown). The intensity and pattern of anti-BrdU immunofluorescence varied considerably among nuclei in myotubes. Some showed a punctate staining pattern and others showed a uniform intense fluorescence. In some myotubes, nearly all nuclei were labeled; in others some nuclei were labeled more intensely than others. No BrdU-positive nuclei were seen in parallel control experiments using uninfected human fetal myoblast clones. DISCUSSION
This study demonstrates that the cooperative expression of the HPV16 E6 and E7 genes from a retroviral vector can greatly extend the life span of primary human myoblasts. Clonal analysis showed that this increased proliferative potential correlated with the
190
¨ LLER, JOHNS, AND SHOUBRIDGE LOCHMU
EXPRESSION OF HPV16-E6E7 IN HUMAN MYOBLASTS
FIG. 4. Immunoblot analysis of pRb and myosin heavy chain expression in (A) fetal myoblast clones expressing HPV16-E6E7 and (B) control uninfected fetal myoblast clones. Five different clones were analyzed in both panels. Each clone is represented in two lanes; unfused myoblasts in growth medium in the left lane and differentiated myotubes 72 h after withdrawal of growth factors in the right lane.
level of expression of E6E7; individual myoblast clones with the highest expression underwent a larger number of population doublings compared to control myoblast clones while those with the lowest expression senesced early. Fusion competence, which was not markedly affected in bulk cultures of E6E7-transduced myoblasts, was impaired in clones with very high steady-state levels of E6E7 mRNA. However, because even untransduced clones show marked differences in these parameters, the expression of E6E7 alone is not itself a reliable predictor of either proliferative potential or fusion competence in any individual clone. No truly immortal myoblast clones emerged from either adult or fetal primary myoblasts transduced with HPV16-E6E7. Immortalization of human cells by DNA tumor virus genes has been modeled as a two-step process in which introduction of an immortalizing agent initially extends the lifetime of the primary cell, usually by abrogating the function of the cell cycle regulatory proteins pRb and p53 [19, 20]. At some point the cells enter a crisis phase in which the telomeres have shortened beyond a critical length necessary for cell growth and division [21]. The emergence of truly immortal clones from cells in crisis is usually a relatively rare event. Although the molecular basis for this second stage of
191
the immortalization process remains obscure, it is thought to involve some genomic modification and is associated with reactivation of telomerase activity [22]. Consistent with this model we observed the abrupt onset of senescence (crisis) in both fetal and adult myoblast cultures expressing E6E7 and could not detect telomerase activity in either early or late passage cultures. The efficiency of immortalization of human cells with viral genes such as SV40 large T antigen or HPV16E6E7 varies greatly depending upon cell type. For instance, immortalization with SV40 large T antigen occurs at much higher frequencies in epithelial cells than fibroblasts [23]. The results we present here suggest that the frequency of immortalization of human myoblasts is also extremely low. Different populations of human mammary epithelial cells have been reported to be immortalized by either E6 or E7 alone, whereas others require the expression of both genes [11]. Human keratinocytes can be fully immortalized by expression of E7 alone, but not E6 [10]. The expression of E6 has been shown to reactivate telomerase activity in human keratinocytes and mammary epithelial cells in a fashion which is independent from its function in targeting p53 for degradation [24]. The fact that the expression of this gene is insufficient for immortalization of these cells shows that telomerase activation, while an intrinsic part of the transformation induced by HPV in these cells, is not sufficient in itself to confer the immortalized phenotype [24]. Recently the expression of the catalytic component of telomerase has been shown to extend the lifetime of human fibroblasts and retinal pigment epithelial cells in culture [25]. These results raise the possibility of increasing the efficiency of isolating immortal human myoblast lines by expressing the telomerase gene in E6E7-expressing myoblast cultures. Myoblasts can be induced to exit the cell cycle and differentiate in vitro by withdrawal of the growth factors from the medium. This program, which is orchestrated by the MyoD family of transcriptional activators, involves fusion of mononuclear myoblasts to form multinucleate myotubes and the transcriptional activation of p21 (a cyclin-dependent kinase inhibitor) and genes coding for muscle-specific proteins [17, 26]. The expression of p21 inhibits phosphorylation of pRb [26] and hypophosphorylated pRb plays an essential role in the differentiation process by interacting with MyoD. Since pRb, and the Rb family members p107 and p130, are the main cellular targets of E7, one would predict
FIG. 3. Fusion and myosin heavy chain expression in fetal myoblasts expressing E6 and E7 genes of HPV16. (A–D) Phase contrast photographs taken 60 h after withdrawal of growth factors to induce myoblast fusion. (E–H) The detection of the embryonic form of MHC by indirect immunofluorescence in parallel cultures. (A, E) Control, uninfected cells; (B, F) E6-infected cells; (C, G) E7-infected cells; (D, H) E6E7-infected cells.
192
¨ LLER, JOHNS, AND SHOUBRIDGE LOCHMU
that fusion/differentiation would be impaired in cells expressing this gene. This effect was apparent in clonal cultures from fetal and adult cells expressing E6E7 where expression of pRb was inversely related to fusion competence and MHC expression: clones in which pRb was not detectable did not express MHC. This suggests that the level of pRb in E6E7-infected cells is barely adequate to induce terminal differentiation. The inhibition of myoblast fusion was most pronounced in cultures from both fetal and adult muscle expressing E7 alone where only scattered, small myotubes were observed. Northern blot analysis suggested that the different effects on fusion in cells expressing E6E7 or E7 alone did not result from differences in transcription. It is possible that the level of E7 protein is lower in cells expressing E6E7 due to less efficient translation from a bicistronic transcript. Terminal differentiation of skeletal myoblasts is associated with permanent withdrawal from the cell cycle and an inability of myonuclei to respond to growth factor stimulation by initiating new DNA synthesis. However, forced expression of SV40 large T antigen in rodent myotubes can reverse this process, inducing nuclei in G 0 to reenter the cell cycle and cross the G 1/S checkpoint [17]. This process is mediated by pRb as evidenced by the reappearance of the hypophosphorylated form after induction of large T antigen in C2C12 myotube nuclei [17]. Likewise, nuclei in myotubes derived from pRb null mice remain responsive to growth factor stimulation and can enter S phase [18]. Myotubes expressing E6E7 can also be induced to reenter the cell cycle, although it was observed in a minority of myotube nuclei in the fetal myoblast clones we tested. As in the experiments with large T antigen in the C2C12 mouse skeletal muscle cell line [27], foci of DNA replication were visible as speckles of anti-BrdU immunofluorescence, characteristic of early S-phase nuclei and there was often asynchrony between myotubes and among nuclei even in the same cytoplasm. Because fusion/differentiation capacity in E6E7-expressing clones depends on the expression of pRb, it seems likely that insufficient pRb is available, even in those E6E7 clones which fuse and differentiate well, to keep cells in the differentiated state. This supports the concept that the terminally differentiated state is one that requires constant maintenance [28]. Retroviral transduction of human myoblasts with the HPV16-E6E7 region is a simple and effective way to extend the life span of adult satellite cells by 30 – 40 population doublings, and we have now used this system to establish cells lines from many patients with mitochondrial myopathies. The extent of muscle differentiation that can be induced in uncloned cultures is not remarkably different from control cultures. Differentiation in some clonal cultures is, however, significantly inhibited, and this is an undesireable property of the system. This difficulty could be overcome by
putting an inducible promoter, such as the tetracycline operon [29], on the expression cassette for E6E7. Alternatively one could potentially rescue the differentiation phenotype by overexpressing an unphosphorylatable form of pRb at the time growth factors are withdrawn. This work was supported by grants to E.A.S. from the Medical Research Council of Canada and the Muscular Dystrophy Association of Canada. E.A.S. is an MNI Killam Scholar. H.L. was supported by a grant of the Savoy Foundation, Canada, and the Deutshe Forshungsgemeinshaft, Germany.
REFERENCES 1.
Edom, F., et al. (1994). Clones of human satellite cells can express in vitro both fast and slow myosin heavy chains. Dev. Biol. 164, 219 –229.
2.
Karpati, G., et al. (1993). Myoblast transfer in Duchenne muscular dystrophy [see comments]. Ann. Neurol. 34, 8 –17.
3.
Iujvidin, S., et al. (1990). SV40 immortalizes myogenic cells: DNA synthesis and mitosis in differentiating myotubes. Differentiation 43, 192–203.
4.
Hurko, O., McKee, L., and Zuurveld, J. G. (1986). Transfection of human skeletal muscle cells with SV40 large T antigen gene coupled to a metallothionein promoter. Ann. Neurol. 20, 573– 582.
5.
Deschenes, I., et al. (1997). Increase in the proliferative capacity of human myoblasts by using the T antigen under the vimentin promoter control. Muscle Nerve 20, 437– 445.
6.
Frederiksen, K., et al. (1988). Immortalization of precursor cells from the mammalian CNS. Neuron 1, 439 – 448.
7.
Nakamigawa, T., et al. (1988). Generation of human myogenic cell lines by the transformation of primary culture with origindefective SV40 DNA. J. Neurol. Sci. 83, 305–319.
8.
Huibregtse, J. M., and Scheffner, M. (1994). Mechanisms of tumor suppressor protein inactivation by the human papillomavirus E6 and E7 oncoproteins. Semin. Virol. 67, 357–367.
9.
Compton, T. (1993). An immortalized human fibroblast cell line is permissive for human cytomegalovirus infection. J. Virol. 67, 3644 –3648.
10.
Halbert, C. L., Demers, G. W., and Galloway, D. A. (1991). The E7 gene of human papillomavirus type 16 is sufficient for immortalization of human epithelial cells. J. Virol. 65, 473– 478.
11.
Wazer, D. E., et al. (1995). Immortalization of distinct human mammary epithelial cell types by human papilloma virus 16 E6 or E7. Proc. Natl. Acad. Sci. USA 92, 3687–3691.
12.
Hudson, J. B., et al. (1990). Immortalization and altered differentiation of human keratinocytes in vitro by the E6 and E7 open reading frames of human papillomavirus type 18. J. Virol. 64, 519 –526.
13.
Perez-Reyes, N., et al. (1992). Immortalization of primary human smooth muscle cells. Proc. Natl. Acad. Sci. USA 89, 1224 – 1228.
14.
Shoubridge, E. A., Johns, T., and Boulet, L. (1995). Use of myoblast cultures to study mitochondrial myopathies. Methods Enzymol. 264, 465– 475.
15.
Pin, C. L., and Merrifield, P. A. (1993). Embryonic and fetal rat myoblasts express different phenotypes following differentiation in vitro. Dev. Genet. 14, 356 –368.
EXPRESSION OF HPV16-E6E7 IN HUMAN MYOBLASTS 16.
17.
18.
19.
20.
21.
22.
Shay, J. W., et al. (1995). Spontaneous in vitro immortalization of breast epithelial cells from a patient with Li-Fraumeni syndrome. Mol. Cell. Biol. 15, 425– 432. Gu, W., et al. (1993). Interaction of myogenic factors and the retinoblastoma protein mediates muscle cell commitment and differentiation. Cell 72, 309 –324. Schneider, J. W., et al. (1994). Reversal of terminal differentiation mediated by p107 in Rb2/2 muscle cells. Science 264, 1467–1471. Shay, J. W., Wright, W. E., and Werbin, H. (1991). Defining the molecular mechanisms of human cell immortalization. Biochim. Biophys. Acta 1072, 1–7. Shay, J. W., et al. (1993). E6 of human papillomavirus type 16 can overcome the M1 stage of immortalization in human mammary epithelial cells but not in human fibroblasts. Oncogene 8, 1407–1413. Counter, C. M., et al. (1992). Telomere shortening associated with chromosome instability is arrested in immortal cells which express telomerase activity. EMBO J. 11, 1921–1929. Kim, N. W., et al. (1994). Specific association of human telomerase activity with immortal cells and cancer [see comments]. Science 266, 2011–2015.
Received March 15, 1998 Revised version received December 15, 1998
23.
24.
25.
26.
27.
28. 29.
193
Shay, J. W., et al. (1993). The frequency of immortalization of human fibroblasts and mammary epithelial cells transfected with SV40 large T-antigen. Exp. Cell Res. 209, 45–52. Klingelhutz, A. J., Foster, S. A., and McDougall, J. K. (1996). Telomerase activation by the E6 gene product of human papillomavirus type 16. Nature 380, 79 – 82. Bodnar, A. G., et al. (1998). Extension of life-span by introduction of telomerase into normal human cells [see comments]. Science 279, 349 –352. Halevy, O., et al. (1995). Correlation of terminal cell cycle arrest of skeletal muscle with induction of p21 by MyoD [see comments]. Science 267, 1018 –1021. Cardoso, M. C., Leonhardt, H., and Nadal-Ginard, B. (1993). Reversal of terminal differentiation and control of DNA replication: Cyclin A and Cdk2 specifically localize at subnuclear sites of DNA replication. Cell 74, 979 –992. Blau, H. M. (1992). Differentiation requires continuous active control. Annu. Rev. Biochem. 61, 1213–1230. Baron, U., Gossen, M., and Bujard, H. (1997). Tetracyclinecontrolled transcription in eukaryotes: Novel transactivators with graded transactivation potential. Nucleic Acids Res. 25, 2723–2729.
Expression of the E6 and E7 Genes of Human Papillomavirus (HPV16) Extends the Life Span of Human Myoblasts Hanns Lochmu¨ller,* Timothy Johns,* and Eric A. Shoubridge* ,† ,1 *Montreal Neurological Institute and †Department of Human Genetics, McGill University, Montreal, Quebec, Canada H3A 2B4
Primary human myoblasts (satellite cells), like other human cells, have a limited life span in vitro. Here we show that expression of the E6E7 early region from human papillomavirus type 16 can greatly extend the life span of both fetal and satellite cell-derived myoblasts and release them from dependence on the growth factors normally necessary for their proliferation. Expression of either the E6 or the E7 gene alone was not sufficient to confer this phenotype, although expression of E7 did delay cellular senescence. The steady-state level of E6E7 transcripts in clonal cultures correlated with proliferative capacity and inversely with the capacity to differentiate into multinuclear myotubes. The expression of E7 alone markedly inhibited cell fusion in both adult and fetal cultures. These effects on myoblast differentiation could be related in part to the level of retinoblastoma protein (pRb), the major cellular target of E7. Terminal differentiation of skeletal myoblasts is associated with permanent withdrawal from the cell cycle; however, continued expression of E6E7 in differentiated myotubes permits reentry of myotube nuclei into S phase in response to growth factor stimulation. These results support a key role for pRb in the acquisition and maintenance of the differentiated state in human skeletal muscle and, in cooperation with p53, in the control of proliferative capacity and response to external growth factors. © 1999 Academic Press Key Words: myoblasts; papillomavirus; immortalization; cell cycle; retinoblastoma protein.
INTRODUCTION
Satellite cells are dormant myoblasts, located between the plasma membrane and basal lamina in skeletal muscle, that can be induced to reenter the cell cycle in response to signals associated with growth and repair of skeletal muscle. They can be cloned [1] and propagated in vitro; however, like other mammalian cells they have a limited life span [2]. Although several immortal rodent myoblast cell lines are in common use, 1
To whom correspondence and reprint requests should be addressed. Fax: (514) 398-1509. E-mail: [email protected]. 0014-4827/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
immortal human myoblast lines have not been isolated. The molecular basis for this species difference is unclear but it may reflect the presence of more stringent cell cycle regulation in human cells. Rodent myoblasts can be readily immortalized with genes from DNA tumour viruses, such as the SV40 large T antigen [3], which targets p53 and retinoblastoma protein (pRB), key components of the cell cycle regulatory machinery at the G1/S checkpoint. Human myoblasts stably transfected with a plasmid driving inducible expression of the SV40 large T antigen are reported to have extended lifetimes, but are not immortalized [4, 5]. Infection of human myoblasts with polyoma large T antigen or a retroviral vector expressing a temperature-sensitive SV40 large T antigen [6] also failed to immortalize at the permissive temperature (E. Shoubridge, unpublished results). Origin-defective SV40 virus has been used to immortalize human myoblasts, but the cells retain little capacity to differentiate [7]. Human papillomaviruses (HPVs) are DNA viruses that normally infect squamous epithelial cells [8]. Over 60 different types are known and the so called “high risk” species such as HPV16 and HPV18 are often associated with cervical carcinomas. Previous studies have shown that expression of the E6E7 early region from HPV16 can efficiently immortalize human primary fibroblasts [9], epithelial cells [10, 11], keratinocytes [12], and smooth muscle cells [13] and that expression of the E7 gene alone is sufficient to immortalize some human epithelial cell types [11] and keratinocytes [12], albeit at a lower frequency. The principal cellular target of the E7 gene is the hypophosphorylated form of pRb which binds E7 in the conserved pocket region, a property shared with other DNA tumor virus genes such as SV40 large T antigen and adenovirus E1A [8]. This relieves the pRb-mediated block at the G1/S boundary, permitting the cell to enter S phase. E6 targets the p53 tumor suppressor gene, abrogating its transcriptional transactivation activity, by targeting p53 destruction in the ubiquitindependent proteolytic pathway [8]. In this study we tested whether expression of the HPV16-E6E7 genes could immortalize primary human myoblasts. Using recombinant retroviral vectors we
186
EXPRESSION OF HPV16-E6E7 IN HUMAN MYOBLASTS
introduced the E6 and E7 genes alone and in combination into human primary adult and fetal myoblasts and show that expression of the HPV16 E6 and E7 genes together significantly extends the life span, but does not result in true immortalization of either fetal or adult primary myoblasts. We also show that the differentiation potential of E6E7 transformed myoblasts is negatively correlated with the steady-state level of E6E7 transcripts in clonal cultures and that myonuclei in terminally differentiated myotubes expressing E6E7 can be induced to reenter the cell cycle in response to exogenous growth factors. MATERIAL AND METHODS Cell culture. Adult and fetal human myoblasts were obtained from biopsy material. Satellite cells were dissociated from the biopsy specimen by trypsinization and grown in supplemented Ham’s F10 medium containing 15% fetal bovine serum (FBS), hereafter referred to as growth medium, as in [14]. Myoblasts were labeled with an antibody to NCAM and purified by fluorescence-activated cell sorting (FACS) as previously described [14]. This routine procedure typically results in myoblast purities .99.5% as determined by postsorting FACS analysis. Fusion and differentiation of myoblasts were induced by replacing the growth medium with Dulbecco’s modified Eagle’s medium (DMEM) containing 2% horse serum (fusion medium) for 48 h. Myoblasts were grown in DMEM containing 10% FBS to test for dependence on exogenous growth factors. To monitor entry of nuclei into S phase, bromodeoxyuridine (BrdU) at a concentration of 0.1 mM was added to the medium. To visualize myoblast clones, cells were fixed in methanol at room temperature for 15 min on the culture plate and stained with 0.1% crystal violet for 15 min. Retroviral constructs and infection. Retroviral constructs, carrying the E6 region, the E7 region, or the E6E7 region of human papillomavirus type 16, were a kind gift of Dr. Denise Galloway (Fred Hutchinson Cancer Center, Seattle, WA) [10]. Packaging cells (PA317) were grown to near confluence in DMEM containing 10% FBS; 24 h later the medium containing retroviral particles was harvested and filtered through a 0.4-mm filter. Retroviral titers were determined by infecting NIH/3T3 cells. The 24-h supernatant (1 or 2 ml) was mixed with 2 or 3 ml of DMEM containing 20% FBS and polybrene (4 mg/ml). Human myoblasts (;10 5) in 100-mm tissue culture dishes were incubated with this medium for 2 h after which an additional 5 ml of medium was added to the dish and the incubation was continued from 3 h to overnight. The cells were washed once in fresh medium to remove the polybrene, and 48 h later selection for retroviral integration was performed in growth medium containing 400 mg/ml geneticin (G418) (GIBCO, Burlington, Ontario, Canada) and continued for at least 14 days. Individual G418-resistant myoblast clones were picked, expanded, and analyzed immediately. RNA extraction and Northern blots. Myoblasts were grown to confluence in 100-mm tissue culture dishes. Total RNA was extracted with an RNAeasy kit according to the manufacturer’s recommendations (Qiagen, Mississauga, Ontario, Canada). Electrophoresis was performed in 1% agarose gels containing formaldehyde. Gels were transferred to nylon membranes (Zeta-probe, Bio-Rad, Mississauga, Ontario, Canada) by vacuum blotting and UV-crosslinked following standard procedures. DNA fragments were labeled by random prime labeling (Stratagene, La Jolla, CA) with [a- 32P]dCTP. Hybridization was performed overnight. The following probes were used: as a specific probe for the E6E7 region of human papillomavirus type 16 (E6E7 probe) a 850-bp EcoRI/BamHI fragment from pLXSN16E6E7; for the E6 region (E6 probe) a 310-bp NsiI fragment from pLXSN16E6E7; and for the E7 region (E7 probe) a 310-bp
187
NsiI/BamHI fragment from pLXSN16E6E7. An 18S rRNA probe was used as an internal reference. After hybridization the membranes were washed for 1 h in three steps with buffers of increasing stringency. The final wash was done at 50°C (0.1% SDS, 0.53 SSC). Membranes were exposed overnight and analyzed on a PhosphorImager using the ImageQuant software (Molecular Dynamics, Sunnyvale, CA). Immunoblotting. Myoblasts and myotubes were harvested from 100-mm tissue culture dishes in cold PBS. Cells were taken up in Laemmli sample buffer and protein from approximately equal numbers of cells (determined by degree of confluency) was loaded on 7% polyacrylamide gels. Electrophoresis and Western blotting were done on the Bio-Rad miniblot apparatus. Membranes were incubated overnight at 4°C with the following primary antibodies: a monoclonal antibody raised against human pRb (1:500 dilution, clone G3-245; PharMingen, Mississauga, Ontario) and a monoclonal antibody against quail embryonic myosin heavy chain (MHC) (1:100 dilution, 47A; a kind gift by Dr. P. Merrifield, University of Western Ontario) [15]. Specific bands were detected using a 1:1000 dilution of a secondary antibody (anti-mouse IgG) coupled to HRP using enhanced chemiluminescence (ECL kit, Amersham, Oakville, Ontario, Canada) according to the manufacturer’s recommendations. Immunocytochemistry. Myoblasts, seeded on glass coverslips, were induced to differentiate in medium containing 2% horse serum for 48 h when nearly confluent. To visualize nuclei in S phase, the medium was changed to myoblast growth medium containing BrdU (0.1 mM) for 24 h. Coverslips were washed with PBS and fixed in 70% ethanol for 30 min at 220°C and then incubated in 2 N hydrochloric acid for 10 min, followed by 0.1 M sodium borate, pH 9, for 10 min. Coverslips were incubated in a humidified chamber with blocking buffer (10% horse serum in PBS) for 30 min, followed by incubation with the primary antibody in blocking buffer for 2 h. The following primary antibodies were used: a 1:20 dilution of a monoclonal antibody against BrdU (B44, Becton Dickinson, Franklin Lakes, NJ) or a 1:20 dilution of a monoclonal antibody against embryonic myosin heavy chain (Novocastra, Newcastle upon Tyne, UK). Nuclei were counterstained with bisbenzimide (Hoechst stain H33342, Calbiochem, La Jolla, CA) at a concentration of 1 mg/ml. The secondary antibody, a 1:100 dilution of biotinylated antimouse IgG raised in horse (Vector, Burlingame, CA), was applied for 45 min. Specific binding was detected with Cy3-conjugated streptavidin (1:1000; Jackson ImmunoResearch, West Grove, PA) for 20 min. Telomerase assays. Telomerase was assayed using the TRAP assay as in [16]. Glioblastoma cell lines were used as positive controls.
RESULTS
Expression of HPV16-E6E7 Extends the Life Span of Human Myoblasts FACS-sorted adult human myoblasts were infected with recombinant retroviral vectors expressing a neomycin resistance gene and the E6 or E7 genes from HPV16, alone or in combination. All retroviral infections were done with an infectious titer of ;10 5/ml. After selection in G418, resistant clones were pooled, passaged twice (1:100) at near confluence, and then plated at 1:1000 (;1000 cells/100 mm plate) to test for clonal proliferative capacity and growth factor dependence (Fig. 1). A large number of colonies was observed from cells that had been transduced with constructs expressing either E6E7 (;800 colonies) or E7 (;500 colonies) and plated in growth medium, while control uninfected myoblasts and cells transduced with E6
188
¨ LLER, JOHNS, AND SHOUBRIDGE LOCHMU
FIG. 1. Analysis of clonal proliferative potential and growth factor dependence in adult human primary myoblasts expressing the E6 and E7 regions of HPV16 alone or in combination. The plates in the top row were cultured in full supplemented growth medium; those in the bottom row in DMEM plus 10% FCS. Cells were transduced with retroviral constructs expressing the E6 or E7 genes as indicated.
alone produced few colonies (Fig. 1). Clones were tested for growth factor dependence by plating in medium that does not support proliferation of primary adult human myoblasts (DMEM plus 10% FCS). Only those clones expressing E6E7 proliferated under these conditions (Fig. 1). Clones expressing E6E7 were selected and passaged in growth medium for approximately 40 more population doublings (60 total). Of the 12 clones tested, 2 grew vigorously and did not exhibit any loss of viability or growth potential, 2 ceased growing and became senescent, and the remaining 8, although still proliferating, showed reduced population doubling times. This behavior correlated with the expression of E6E7 which showed a fivefold variation in the steady-state level of E6E7 by Northern blot analysis (Fig. 2). Clones with the greatest proliferative potential (no decrease in population doubling time) expressed the highest levels of E6E7 (Fig. 2, lanes 6 and 7); the lowest levels were expressed in senescent clones (Fig. 2, lanes 8 and 11). Parallel cultures of uninfected myoblast clones or clones infected with constructs transferring the E7region alone showed signs of reduced proliferative potential after 25–30 population doublings and all senesced shortly thereafter. The two vigorously growing clones (Fig. 2, lanes 6 and 7) were passaged further and both abruptly senesced after 4 more passages (12 population doublings). Fusion competence was assessed in the same 12 clones. Clones with the highest expression of E6E7 showed poor fusion with only a few scattered, small myotubes. However, clones with intermediate levels of expression showed variable fusion competence; some (Fig. 2, lane 9) fused very well and others
with similar E6E7 expression (Fig. 2, lane 5) showed modest fusion. Proliferative potential was also assessed in FACSsorted fetal human myoblasts expressing HPV16E6E7. G418-resistant clones were recloned twice and passaged in growth medium for several months. The cells proliferated without major changes in growth or doubling time until passage 38 (total population doublings 130 –140), when they abruptly senesced. Uncloned adult and fetal E6E7 cells were maintained for several weeks in growth medium after they had ceased growing (crisis); however, no foci of immortal cells emerged. Telomerase activity, measured by the TRAP assay, was not detectable in any of the adult or fetal E6E7-transformed cells (data not shown). Differentiation of Fetal Human Myoblast Clones and pRb Expression pRb is thought to play a key role in the differentiation of skeletal muscle through its interaction with the MyoD family members [17]. To test whether the expression of HPV16 E6 or E7 affected the capacity for differentiation of uncloned primary human myoblasts we first examined cell fusion and MHC expression in pooled fetal myoblasts expressing E6, E7, or E6E7 (Fig. 3). Large multinuclear myotubes expressing MHC, indistinguishable from control uninfected, uncloned cultures, were observed in cells expressing E6 or E6E7 60 h after transfer to fusion medium. Fusion was, however, markedly inhibited in cells expressing E7 alone where only a few scattered myotubes were observed, most with only two to three nuclei (Figs. 3C and
EXPRESSION OF HPV16-E6E7 IN HUMAN MYOBLASTS
189
FIG. 2. Northern blot analysis of E6E7 expression in adult myoblast clones. The 18S rRNA was used as a control for RNA loading. Proliferative capacity and fusion were scored semiquantitatively as follows: for proliferative capacity; 0, senescence; 1, slow growth, long population doubling time; 11, good growth, average population doubling time; 111, rapid growth, short population doubling time. For fusion: 0, poor fusion; 1, some fusion; 11, good fusion; 111, very good fusion. Fusion could not be assessed in the two clones that senesced (2, lanes 8 and 11). The steady-state ratio of E6E7 mRNA:18S rRNA levels is shown below the fusion index.
3G). Northern blot analysis with E6- and E7-specific probes showed that the difference between cells expressing E7 alone or E6E7 could not be attributed to a difference in the steady-state level of transcripts containing the E7-region (data not shown). To further examine the basis for this difference in the capacity for differentiation, pRb expression, fusion competence, and MHC expression were measured in fetal myoblast clones expressing E6E7. There was a wide range in differentiation potential in different clones; some fused readily and formed large multinucleate myofibers and others exhibited little fusion. Immunoblotting on these clones revealed that MHC expression was positively correlated with immunodetectable pRb protein (Fig. 4A). Clones with detectable pRb all expressed MHC, while those with no detectable pRb did not. As a control, uninfected fetal myoblast clones were analysed for fusion/differentiation capacity and pRb expression. As with E6E7-expressing clones, some fused rapidly and formed large multinucleate myofibers and others fused more slowly and less extensively. Immunoblotting on these clones revealed that differentiation capacity as measured by MHC expression also correlated to some degree with pRb expression (Fig. 4B). Compared to uninfected fetal myoblast clones, E6E7-expressing clones showed reduced fusion and lower levels of detectable pRb protein. Although we did not carry out careful quantitation, the increased pRb expression in control cells is clear from the pRb blots in Figs. 4A and 4B which were exposed for 5 s and 3 min, respectively. pRb levels were lower in differentiated myotubes from all clones, regardless of their fusion capacity.
E6E7-Expressing Myoblasts Can Reenter the Cell Cycle after Differentiation The continued presence of pRb is thought to be important for keeping the myonuclei in postmitotic myotubes in the G 0 phase of the cell cycle. Myotubes derived from pRb null mice can reenter the cell cycle in response to exogenous growth factors [18] and expression of the large T antigen from SV40 (which targets pRb and p53) in rodent myotubes can trigger new DNA synthesis [3]. To test whether nuclei in myotubes derived from clones expressing E6E7 could be induced to reenter the cell cycle, differentiated cultures containing large multinucleate myotubes were exposed to growth medium containing BrdU to label nuclei in S phase. After 24 h in this medium a large number of nuclei in myotubes could be detected with the antiBrdU antibody (data not shown). The intensity and pattern of anti-BrdU immunofluorescence varied considerably among nuclei in myotubes. Some showed a punctate staining pattern and others showed a uniform intense fluorescence. In some myotubes, nearly all nuclei were labeled; in others some nuclei were labeled more intensely than others. No BrdU-positive nuclei were seen in parallel control experiments using uninfected human fetal myoblast clones. DISCUSSION
This study demonstrates that the cooperative expression of the HPV16 E6 and E7 genes from a retroviral vector can greatly extend the life span of primary human myoblasts. Clonal analysis showed that this increased proliferative potential correlated with the
190
¨ LLER, JOHNS, AND SHOUBRIDGE LOCHMU
EXPRESSION OF HPV16-E6E7 IN HUMAN MYOBLASTS
FIG. 4. Immunoblot analysis of pRb and myosin heavy chain expression in (A) fetal myoblast clones expressing HPV16-E6E7 and (B) control uninfected fetal myoblast clones. Five different clones were analyzed in both panels. Each clone is represented in two lanes; unfused myoblasts in growth medium in the left lane and differentiated myotubes 72 h after withdrawal of growth factors in the right lane.
level of expression of E6E7; individual myoblast clones with the highest expression underwent a larger number of population doublings compared to control myoblast clones while those with the lowest expression senesced early. Fusion competence, which was not markedly affected in bulk cultures of E6E7-transduced myoblasts, was impaired in clones with very high steady-state levels of E6E7 mRNA. However, because even untransduced clones show marked differences in these parameters, the expression of E6E7 alone is not itself a reliable predictor of either proliferative potential or fusion competence in any individual clone. No truly immortal myoblast clones emerged from either adult or fetal primary myoblasts transduced with HPV16-E6E7. Immortalization of human cells by DNA tumor virus genes has been modeled as a two-step process in which introduction of an immortalizing agent initially extends the lifetime of the primary cell, usually by abrogating the function of the cell cycle regulatory proteins pRb and p53 [19, 20]. At some point the cells enter a crisis phase in which the telomeres have shortened beyond a critical length necessary for cell growth and division [21]. The emergence of truly immortal clones from cells in crisis is usually a relatively rare event. Although the molecular basis for this second stage of
191
the immortalization process remains obscure, it is thought to involve some genomic modification and is associated with reactivation of telomerase activity [22]. Consistent with this model we observed the abrupt onset of senescence (crisis) in both fetal and adult myoblast cultures expressing E6E7 and could not detect telomerase activity in either early or late passage cultures. The efficiency of immortalization of human cells with viral genes such as SV40 large T antigen or HPV16E6E7 varies greatly depending upon cell type. For instance, immortalization with SV40 large T antigen occurs at much higher frequencies in epithelial cells than fibroblasts [23]. The results we present here suggest that the frequency of immortalization of human myoblasts is also extremely low. Different populations of human mammary epithelial cells have been reported to be immortalized by either E6 or E7 alone, whereas others require the expression of both genes [11]. Human keratinocytes can be fully immortalized by expression of E7 alone, but not E6 [10]. The expression of E6 has been shown to reactivate telomerase activity in human keratinocytes and mammary epithelial cells in a fashion which is independent from its function in targeting p53 for degradation [24]. The fact that the expression of this gene is insufficient for immortalization of these cells shows that telomerase activation, while an intrinsic part of the transformation induced by HPV in these cells, is not sufficient in itself to confer the immortalized phenotype [24]. Recently the expression of the catalytic component of telomerase has been shown to extend the lifetime of human fibroblasts and retinal pigment epithelial cells in culture [25]. These results raise the possibility of increasing the efficiency of isolating immortal human myoblast lines by expressing the telomerase gene in E6E7-expressing myoblast cultures. Myoblasts can be induced to exit the cell cycle and differentiate in vitro by withdrawal of the growth factors from the medium. This program, which is orchestrated by the MyoD family of transcriptional activators, involves fusion of mononuclear myoblasts to form multinucleate myotubes and the transcriptional activation of p21 (a cyclin-dependent kinase inhibitor) and genes coding for muscle-specific proteins [17, 26]. The expression of p21 inhibits phosphorylation of pRb [26] and hypophosphorylated pRb plays an essential role in the differentiation process by interacting with MyoD. Since pRb, and the Rb family members p107 and p130, are the main cellular targets of E7, one would predict
FIG. 3. Fusion and myosin heavy chain expression in fetal myoblasts expressing E6 and E7 genes of HPV16. (A–D) Phase contrast photographs taken 60 h after withdrawal of growth factors to induce myoblast fusion. (E–H) The detection of the embryonic form of MHC by indirect immunofluorescence in parallel cultures. (A, E) Control, uninfected cells; (B, F) E6-infected cells; (C, G) E7-infected cells; (D, H) E6E7-infected cells.
192
¨ LLER, JOHNS, AND SHOUBRIDGE LOCHMU
that fusion/differentiation would be impaired in cells expressing this gene. This effect was apparent in clonal cultures from fetal and adult cells expressing E6E7 where expression of pRb was inversely related to fusion competence and MHC expression: clones in which pRb was not detectable did not express MHC. This suggests that the level of pRb in E6E7-infected cells is barely adequate to induce terminal differentiation. The inhibition of myoblast fusion was most pronounced in cultures from both fetal and adult muscle expressing E7 alone where only scattered, small myotubes were observed. Northern blot analysis suggested that the different effects on fusion in cells expressing E6E7 or E7 alone did not result from differences in transcription. It is possible that the level of E7 protein is lower in cells expressing E6E7 due to less efficient translation from a bicistronic transcript. Terminal differentiation of skeletal myoblasts is associated with permanent withdrawal from the cell cycle and an inability of myonuclei to respond to growth factor stimulation by initiating new DNA synthesis. However, forced expression of SV40 large T antigen in rodent myotubes can reverse this process, inducing nuclei in G 0 to reenter the cell cycle and cross the G 1/S checkpoint [17]. This process is mediated by pRb as evidenced by the reappearance of the hypophosphorylated form after induction of large T antigen in C2C12 myotube nuclei [17]. Likewise, nuclei in myotubes derived from pRb null mice remain responsive to growth factor stimulation and can enter S phase [18]. Myotubes expressing E6E7 can also be induced to reenter the cell cycle, although it was observed in a minority of myotube nuclei in the fetal myoblast clones we tested. As in the experiments with large T antigen in the C2C12 mouse skeletal muscle cell line [27], foci of DNA replication were visible as speckles of anti-BrdU immunofluorescence, characteristic of early S-phase nuclei and there was often asynchrony between myotubes and among nuclei even in the same cytoplasm. Because fusion/differentiation capacity in E6E7-expressing clones depends on the expression of pRb, it seems likely that insufficient pRb is available, even in those E6E7 clones which fuse and differentiate well, to keep cells in the differentiated state. This supports the concept that the terminally differentiated state is one that requires constant maintenance [28]. Retroviral transduction of human myoblasts with the HPV16-E6E7 region is a simple and effective way to extend the life span of adult satellite cells by 30 – 40 population doublings, and we have now used this system to establish cells lines from many patients with mitochondrial myopathies. The extent of muscle differentiation that can be induced in uncloned cultures is not remarkably different from control cultures. Differentiation in some clonal cultures is, however, significantly inhibited, and this is an undesireable property of the system. This difficulty could be overcome by
putting an inducible promoter, such as the tetracycline operon [29], on the expression cassette for E6E7. Alternatively one could potentially rescue the differentiation phenotype by overexpressing an unphosphorylatable form of pRb at the time growth factors are withdrawn. This work was supported by grants to E.A.S. from the Medical Research Council of Canada and the Muscular Dystrophy Association of Canada. E.A.S. is an MNI Killam Scholar. H.L. was supported by a grant of the Savoy Foundation, Canada, and the Deutshe Forshungsgemeinshaft, Germany.
REFERENCES 1.
Edom, F., et al. (1994). Clones of human satellite cells can express in vitro both fast and slow myosin heavy chains. Dev. Biol. 164, 219 –229.
2.
Karpati, G., et al. (1993). Myoblast transfer in Duchenne muscular dystrophy [see comments]. Ann. Neurol. 34, 8 –17.
3.
Iujvidin, S., et al. (1990). SV40 immortalizes myogenic cells: DNA synthesis and mitosis in differentiating myotubes. Differentiation 43, 192–203.
4.
Hurko, O., McKee, L., and Zuurveld, J. G. (1986). Transfection of human skeletal muscle cells with SV40 large T antigen gene coupled to a metallothionein promoter. Ann. Neurol. 20, 573– 582.
5.
Deschenes, I., et al. (1997). Increase in the proliferative capacity of human myoblasts by using the T antigen under the vimentin promoter control. Muscle Nerve 20, 437– 445.
6.
Frederiksen, K., et al. (1988). Immortalization of precursor cells from the mammalian CNS. Neuron 1, 439 – 448.
7.
Nakamigawa, T., et al. (1988). Generation of human myogenic cell lines by the transformation of primary culture with origindefective SV40 DNA. J. Neurol. Sci. 83, 305–319.
8.
Huibregtse, J. M., and Scheffner, M. (1994). Mechanisms of tumor suppressor protein inactivation by the human papillomavirus E6 and E7 oncoproteins. Semin. Virol. 67, 357–367.
9.
Compton, T. (1993). An immortalized human fibroblast cell line is permissive for human cytomegalovirus infection. J. Virol. 67, 3644 –3648.
10.
Halbert, C. L., Demers, G. W., and Galloway, D. A. (1991). The E7 gene of human papillomavirus type 16 is sufficient for immortalization of human epithelial cells. J. Virol. 65, 473– 478.
11.
Wazer, D. E., et al. (1995). Immortalization of distinct human mammary epithelial cell types by human papilloma virus 16 E6 or E7. Proc. Natl. Acad. Sci. USA 92, 3687–3691.
12.
Hudson, J. B., et al. (1990). Immortalization and altered differentiation of human keratinocytes in vitro by the E6 and E7 open reading frames of human papillomavirus type 18. J. Virol. 64, 519 –526.
13.
Perez-Reyes, N., et al. (1992). Immortalization of primary human smooth muscle cells. Proc. Natl. Acad. Sci. USA 89, 1224 – 1228.
14.
Shoubridge, E. A., Johns, T., and Boulet, L. (1995). Use of myoblast cultures to study mitochondrial myopathies. Methods Enzymol. 264, 465– 475.
15.
Pin, C. L., and Merrifield, P. A. (1993). Embryonic and fetal rat myoblasts express different phenotypes following differentiation in vitro. Dev. Genet. 14, 356 –368.
EXPRESSION OF HPV16-E6E7 IN HUMAN MYOBLASTS 16.
17.
18.
19.
20.
21.
22.
Shay, J. W., et al. (1995). Spontaneous in vitro immortalization of breast epithelial cells from a patient with Li-Fraumeni syndrome. Mol. Cell. Biol. 15, 425– 432. Gu, W., et al. (1993). Interaction of myogenic factors and the retinoblastoma protein mediates muscle cell commitment and differentiation. Cell 72, 309 –324. Schneider, J. W., et al. (1994). Reversal of terminal differentiation mediated by p107 in Rb2/2 muscle cells. Science 264, 1467–1471. Shay, J. W., Wright, W. E., and Werbin, H. (1991). Defining the molecular mechanisms of human cell immortalization. Biochim. Biophys. Acta 1072, 1–7. Shay, J. W., et al. (1993). E6 of human papillomavirus type 16 can overcome the M1 stage of immortalization in human mammary epithelial cells but not in human fibroblasts. Oncogene 8, 1407–1413. Counter, C. M., et al. (1992). Telomere shortening associated with chromosome instability is arrested in immortal cells which express telomerase activity. EMBO J. 11, 1921–1929. Kim, N. W., et al. (1994). Specific association of human telomerase activity with immortal cells and cancer [see comments]. Science 266, 2011–2015.
Received March 15, 1998 Revised version received December 15, 1998
23.
24.
25.
26.
27.
28. 29.
193
Shay, J. W., et al. (1993). The frequency of immortalization of human fibroblasts and mammary epithelial cells transfected with SV40 large T-antigen. Exp. Cell Res. 209, 45–52. Klingelhutz, A. J., Foster, S. A., and McDougall, J. K. (1996). Telomerase activation by the E6 gene product of human papillomavirus type 16. Nature 380, 79 – 82. Bodnar, A. G., et al. (1998). Extension of life-span by introduction of telomerase into normal human cells [see comments]. Science 279, 349 –352. Halevy, O., et al. (1995). Correlation of terminal cell cycle arrest of skeletal muscle with induction of p21 by MyoD [see comments]. Science 267, 1018 –1021. Cardoso, M. C., Leonhardt, H., and Nadal-Ginard, B. (1993). Reversal of terminal differentiation and control of DNA replication: Cyclin A and Cdk2 specifically localize at subnuclear sites of DNA replication. Cell 74, 979 –992. Blau, H. M. (1992). Differentiation requires continuous active control. Annu. Rev. Biochem. 61, 1213–1230. Baron, U., Gossen, M., and Bujard, H. (1997). Tetracyclinecontrolled transcription in eukaryotes: Novel transactivators with graded transactivation potential. Nucleic Acids Res. 25, 2723–2729.