Skip to main content
NCBI home page
Journal of Virology logoLink to Journal of Virology
. 2000 Apr;74(7):3177–3187. doi: 10.1128/jvi.74.7.3177-3187.2000

DNA Methylation of Helper Virus Increases Genetic Instability of Retroviral Vector Producer Cells

Won-Bin Young 1,2,†, Gary L Lindberg 3, Charles J Link Jr 1,2,*
PMCID: PMC111818  PMID: 10708434

Abstract

Retroviral vector producer cells (VPC) have been considered genetically stable. A clonal cell population exhibiting a uniform vector integration pattern is used for sustained vector production. Here, we observed that the vector copy number is increased and varied in a population of established LTKOSN.2 VPC. Among five subclones of LTKOSN.2 VPC, the vector copy number ranged from 1 to approximately 29 copies per cell. A vector superinfection experiment and Northern blot analysis demonstrated that suppression of helper virus gene expression decreased Env-receptor interference and allowed increased superinfection. The titer production was tightly associated with helper virus gene expression and varied between 0 and 2.2 × 105 CFU/ml in these subclones. In one analyzed subclone, the number of integrated vectors increased from one copy per cell to nine copies per cell during a 31-day period. Vector titer was reduced from 1.5 × 105 CFU to an undetectable level. To understand the mechanism involved, helper virus and vectors were examined for DNA methylation status by methylation-sensitive restriction enzyme digestion. We demonstrated that DNA methylation of helper virus 5′ long terminal repeat occurred in approximately 2% of the VPC population per day and correlated closely with inactivation of helper virus gene expression. In contrast, retroviral vectors did not exhibit significant methylation and maintained consistent transcription activity. Treatment with 5-azacytidine, a methylation inhibitor, partially reversed the helper virus DNA methylation and restored a portion of vector production. The preference for methylation of helper virus sequences over vector sequences may have important implications for host-virus interaction. Designing a helper virus to overcome cellular DNA methylation may therefore improve vector production. The maintenance of increased viral envelope-receptor interference might also prevent replication-competent retrovirus formation.

Retroviral vectors are the most commonly used gene delivery vehicles in human gene transfer trials (50). Permanent modifications of a targeted cell's genotype by the integration of a retroviral vector and the consistent production of retroviral vectors from permanent packaging cells (5, 32) offer significant advantages over other gene delivery vectors. After infection of target cells, retroviral single-stranded RNA genome is converted into double-stranded DNA (proviral DNA) by retroviral reverse transcriptase (RT). The proviral DNA is permanently integrated into the cellular chromosomes as a provirus and then replicates as part of the host genome. The proviral DNA is then transmitted from one generation to the next if the virus is integrated into a germ line cell. The deletion of a provirus from its integration site is observed in only approximately 1 of 106 cells (53, 59).

A major biosafety concern with retroviral vectors is replication-competent retrovirus (RCR) outbreak (1). RCR has been correlated with the occurrence of T-cell lymphomas in monkeys that had received RCR-contaminated bone marrow cell transplants (7, 58). Vector and helper virus rearrangement caused by abnormal template switches during the reverse transcription process has been considered the major cause of RCR formation (46, 58). Without Env-receptor interference, cocultivation of vector producer cells (VPC) with different host ranges, such as ecotropic and amphotropic VPC, can cause RCR formation in 3 to 10 days depending on the number of recombination events necessary. The driving force for these recombination events is RT enzyme activity imported by the vector infection from other VPC (38).

During the vector assembly process, viral Env protein produced by the helper virus is transported to the cell membrane, where it binds the cellular receptor used for virus entry (31, 54, 60). This Env-receptor interference has been considered an efficient barrier to the superinfection of vectors. The transduction efficiency of amphotropic vector on amphotropic PA317 packaging cells is reduced by 6 orders of magnitude, because of the amphotropic Env-receptor interference mechanism, from the transduction efficiency of the same vector on NIH 3T3 cells (33). However, Env-receptor interference is not a complete block to superinfection. Most retroviral packaging cells currently used in gene transfer studies are based on either murine cells (i.e., NIH 3T3 cells in GP+E86 [29] and PA317 [32]) or human cells (i.e., 293T cells in BOSC [47]) and express virus receptors. Thus, these packaging cells can still be infected by the vector they produce, albeit at low frequency (33, 36). These vector reentry events may lead to RCR formation.

Suicide gene therapy against cancer cells by the combination of HSVtk and ganciclovir is a well-developed approach in human gene therapy trials (37, 43). We previously established murine LTKOSN.2 VPC to transfer the HSVtk gene by transducing an LTKOSN vector from GP+E86 to PA317 cells and demonstrated effective suicide gene therapy against breast cancer cells in vitro (27), colon carcinoma in vitro and in vivo (41), and human ovarian adenocarcinoma xenografts in vivo (28). We have previously found an HSVtk deletion mutant vector in addition to the original LTKOSN vector in the same VPC (Fig. 1) (62). In the current study of LTKOSN.2 VPC and derivative subclones, we observed significant genetic instability of several types, including increased copies of both vectors, a deleted provirus integration site, and a complete loss of vector production. To understand the mechanisms involved in these observations, we analyzed the DNA methylation status and the gene expression of vectors and helper virus. We demonstrated that transcriptional inactivation occurred in helper virus but not vectors. Significant DNA methylation of the helper virus 5′ long terminal repeat (LTR) region was observed in VPC subclones that lacked helper virus gene expression. These same VPC subclones also had a loss of vector production and high vector copy number. A superinfection experiment demonstrated that these subclones are very susceptible to the reentry of vector, compared to the other subclones. These findings suggest that DNA methylation may lead to VPC genome instability by reducing Env-receptor interference and permitting high levels of vector reentry.

FIG. 1.

FIG. 1

Open in a new tab

Increased copy number of LTKOSN and ΔLTKOSN in LTKOSN.2 subclones. (A) Schematic diagram of LTKOSN and ΔLTKOSN vectors. LTKOSN contains an HSVtk gene, which was cloned into the EcoRI site of LXSN. Ψ, extended packaging signal region; Neo, neomycin phosphate transferase gene; SV, promoter sequence of SV40 early gene. Bp, BpmI; E, EcoRI; H, HindIII; K, KpnI. (B) Genomic DNA was extracted from GP+E86 transfected with LTKOSN (lane 1), parental LTKOSN.2 (lane 3), and its derived subclones 1 to 5 and 10 (lanes 4 to 9, respectively). PA317-derived DNA (lane 2) was used as a control. KpnI releases a full-length proviral LTKOSN without the 5′ LTR (4.0 kb) and a ΔLTKOSN without its 5′ LTR (2.5 kb). Integrated LTKOSN and ΔLTKOSN were detected from parental LTKOSN.2 and all subclones, but not PA317 cells. Only LTKOSN vector was detected in LTKOSN-transfected GP+E86 cells. Note the relative variance in ΔLTKOSN copy number of different subclones compared to LTKOSN. (C) The same Southern blot was hybridized with a BstXI fragment probe (0.68 kb, located at gag coding region) that detects a 2.5-kb fragment of KpnI-digested endogenous retroviral sequences showed similar loading of DNA samples.

MATERIALS AND METHODS

Generation of LTKOSN.2 VPC and derived subclones.

The construction of pLTKOSN and LTKOSN.2 VPC has been previously described (28). Briefly, 1,201 bp of PCR-amplified HSVtk gene was cloned into the EcoRI site of pLXSN (35) (kindly provided by A. Dusty Miller, Fred Hutchinson Cancer Research Center, Seattle, Wash.) to produce pLTKOSN retroviral vector (Fig. 1A). Ecotropic packaging cell line GP+E86 (29), a generous gift from Arthur Bank, Columbia University, New York, N.Y., was transiently transfected with pLTKOSN. Supernatants from these GP+E86 cells were used to transduce the amphotropic retroviral packaging line PA317 (32), kindly provided by A. Dusty Miller. Forty different VPC clones were isolated, and the LTKOSN.2 VPC produced the highest titer, 1.6 × 106 CFU/ml (28). The subclones of LTKOSN.2 in this study were obtained by limiting dilution of LTKOSN.2 VPC onto two 96-well plates and examined by light microscopy to ensure that only single-cell subclones were further analyzed. Six single-cell clones (1 to 5 and 10) were identified and expanded for further study.

Cell culture and drug treatment.

Cell cultures were maintained in Dulbecco's modified Eagle medium (DMEM; GIBCO BRL Life Technologies, Gaithersburg, Md.)–10% fetal calf serum with 5% CO2, at 37°C. To reverse the DNA methylation, cells were grown in DMEM containing 5 μM 5-azacytidine (5-aza-C) (Sigma Chemical Co., St. Louis, Mo.) for 72 h (26) and then subjected to titer assay, vector RNA slot blotting, and DNA methylation analysis.

Retroviral infection.

Vector titers were determined by the transduction of NIH 3T3 tk(−) cells (American Type Culture Collection [ATCC] CRL1658), A375 cells (ATCC CRL1619; human melanoma), and IGROV cells (human ovarian carcinoma [57]) with 10-fold serial dilutions of vector stocks. Transduction was performed in 1 ml of DMEM by incubating 105 cells/well in six-well plates with vector in the presence of 10 μg of protamine sulfate (Fujisawa USA Inc., Deerfield, Ill.) per ml. After transduction for 24 h, the cells were selected in medium containing G418 (1 mg/ml) for 10 to 14 days. Titers were obtained by multiplying the number of resistant colonies by the dilution factor.

To perform the superinfection assays on each subclone of LTKOSN.2 VPC, a LEIN retroviral vector carrying an enhanced green fluorescent protein (EGFP) (6, 10) reporter gene was used to transduce LTKOSN.2 VPC subclones. The LEIN vector is an LXSN vector with an internal ribosomal entry site (IRES) (19) replacing the simian virus 40 (SV40) promoter. The EGFP gene was inserted in front of the IRES to demonstrate transduction efficiency in target cells. Cells from NIH 3T3, PA317, parental LTKOSN.2 VPC, and LTKOSN.2 VPC subclones were seeded at 5 × 105 cells per well on 12-well plates and exposed to 0.5 ml of supernatant of LEIN vector (3.4 × 105 CFU/ml) in the presence of 10 μg of protamine sulfate per ml 24 h after seeding. Two days after a single exposure to LEIN vector, the transduction efficiency of these cell lines and VPC subclones was determined by a fluorescence-activated cell sorter (FACS) analysis of EGFP expression (39) on an EPICS Profile II Analyzer (Coulter Co., Miami, Fla.).

Southern blot analysis of genomic DNA.

Total cellular DNA was extracted from cell cultures using phenol-chloroform extraction and then dissolved in Tris-EDTA buffer (pH 8.0) overnight at 55°C (51). Genomic DNA was digested by KpnI to release nearly full-length LTKOSN (4.0-kb) and ΔLTKOSN (2.5-kb) fragments (Fig. 1) and then Southern blotted for Neor probe detection of the copy number of integrated vectors. A 0.68-kb BstXI fragment of pPAM3 was used as a probe to detect a 2.5-kb DNA fragment of endogenous retroviral elements to demonstrate similar loading of DNA samples. To analyze for multiple integration sites in each subclone, BamHI was used to generate DNA fragments containing both proviral vector sequences and flanking chromosomal sequences. Southern blot analysis was performed, and membrane duplicates were hybridized with Neor and HSVtk probes (Fig. 2).

FIG. 2.

FIG. 2

Open in a new tab

Genetic instability of integrated LTKOSN in VPC genome. (A) Genomic DNA extracted from PA317 (lane 1), LTKOSN.2 VPC (lane 2), and its subclones (lanes 3 to 8) was subjected to BamHI digestion and detected by Neor and/or HSVtk probes in Southern blot analysis. Only two integration sites were detected on parental LTKOSN.2 and subclones 1, 4, and 10, but subclones 2, 3, and 5 showed more integration sites. The arrows indicate the positions of integrated LTKOSN vector. Note the absence of these bands from subclone 5 on both Neor and HSVtk probe hybridizations. (B) Schema of LTKOSN and ΔLTKOSN integration sites shows the locations of BamHI sites on the flanking chromosomal sequences. The BamHI site on ΔLTKOSN is deleted; therefore, digestion with BamHI excises the entire ΔLTKOSN from flanking chromosomal DNA sequences (4.5 kb) at the integration site. BamHI digestion excises the LTKOSN backbone 5′ to the SV40 promoter and results in two fragments of 3.7 and 2.2 kb including 5′ and 3′ junctional chromosomal DNA sequences, respectively.

RNA analysis of LTKOSN VPC and vector supernatants.

Total cellular RNA was isolated from individual subclones using the RNAeasy kit (Qiagen, Inc., Valencia, Calif.). Viral supernatants were subjected to 20% sucrose gradient ultracentrifugation (125,000 × g) for 2 h at 4°C, and virion pellets were then extracted for RNA with RNAzol (Biotecx, Houston, Tex.). Cellular and virion RNAs from each subclone were subjected to Northern blot analysis on a 1% agarose–0.4 M formaldehyde gel. Vector transcripts were detected by a Neor probe, and helper viral transcripts were detected by a 1.4-kb env probe, which was digested from pPAM3 by XhoI. RNA slot blotting for titer determination (40) was performed by applying 200 μl of supernatant, directly collected from VPC medium and centrifuged at 2,800 × g to eliminate the cellular debris, onto a slot blot apparatus (SlotBlot; Hoefer Scientific Instruments, San Francisco, Calif.). Vector RNA was UV cross-linked onto this membrane and detected by a Neor probe.

Methylation analysis.

The methylation status of provirus and vectors at the SmaI site in the 5′ LTR was determined by digestion of genomic DNA with DraI and EcoRV to reduce the DNA fragment size. The DNA was then precipitated with ethanol, redissolved in sterile water, and divided into two equal portions, one of which was subjected to methylation-sensitive SmaI restriction endonuclease digestion for Southern blotting analysis. The Southern blot membrane was hybridized with a 428-bp fragment of gag sequence (PvuII/DraI) from pPAM3 to detect helper virus, a 273-bp EcoRI/EcoRV fragment of HSVtk sequence for LTKOSN vector, and a Neor probe to detect both LTKOSN and ΔLTKOSN. This Southern blot was also probed with an env probe to determine the methylation status of the SmaI site in the helper virus RT region. DNAs derived from NIH 3T3 and PG13 (ATCC CRL 10686), a packaging cell line for gibbon ape leukemia virus (GaLV) pseudotyped vector (34), were used as negative controls. A 0.68-kb BstXI fragment of pPAM3 was used as a probe to detect a 1.2-kb DNA fragment of endogenous retroviral sequences to demonstrate similar loading of DNA samples. Densitometric analyses were performed with a Hoefer Densitometer GS300 (Hoefer Scientific Instruments) to measure the densities of SmaI-sensitive bands relative to those of DraI/EcoRV bands. Due to the interference from endogenous retroviral elements, the fraction of SmaI methylation in 5′ LTR was calculated as 1 minus the intensity ratio of the SmaI-sensitive band (1.5 kb) divided by that of the DraI/EcoRV band (1.8 kb).

RESULTS

Dynamics of proviral DNA integration.

We have previously detected an HSVtk deletion mutant vector (ΔLTKOSN) coexisting with full-length LTKOSN vector (Fig. 1A) in our LTKOSN.2 VPC population (62). To investigate whether these ΔLTKOSN and LTKOSN vectors coexisted within the same individual cellular genome or existed separately in portions of the LTKOSN.2 VPC population, limiting dilution cloning on LTKOSN.2 VPC was performed to isolate six single-cell subclones (1 to 5 and 10). After expansion of the subclones, each clone was then analyzed for integrated vectors by KpnI digestion on both LTR regions to release the integrated proviral vectors of ΔLTKOSN (2.5 kb [Fig. 1A]) and LTKOSN (4.0 kb [Fig. 1A]). Southern blot analysis demonstrated that ΔLTKOSN and LTKOSN coexisted in the same individual cellular genome, since both vectors were detected in all six subclones (Fig. 1B). Varying copy numbers of ΔLTKOSN and LTKOSN were observed in subclones 2, 3, and 5 by comparing the intensities of detected signals of both vectors. In contrast, only one single copy each of ΔLTKOSN and LTKOSN vector was detected in parental LTKOSN.2 VPC and subclones 1, 4, and 10. To estimate the copy numbers of integrated vectors, genomic DNA from each subclone was subjected to BamHI digestion and Southern blot analysis for the integration sites of both ΔLTKOSN and LTKOSN (Fig. 2). The BamHI site is unique in the LTKOSN sequence, and restriction digestion yields two fragments of the LTKOSN vector with adjacent chromosomal sequences. Since the BamHI site was deleted with the HSVtk gene in ΔLTKOSN vector, only one genomic DNA fragment of integrated ΔLTKOSN vector with flanking chromosomal sequences is observed after BamHI digestion (Fig. 2B). The copy number of ΔLTKOSN was approximately six copies per cell in subclone 3 and eight copies per cell in subclone 2, while only one copy of LTKOSN exists per cell in these two subclones. An increased copy number of LTKOSN was observed only in subclone 5, which showed approximately 9 copies of LTKOSN and approximately 20 copies of ΔLTKOSN per cell. Therefore, significant changes in the vector copy numbers have occurred between the parental VPC and its subclones.

Deletion of proviral integration site.

In subclone 5, a possible deletion of an integration site of LTKOSN vector was observed (Fig. 2A). A 2.2-kb fragment (arrow) containing the 3′ portion of proviral LTKOSN fragment with adjacent chromosomal sequences was detected by the Neor DNA probe on parental LTKOSN.2 VPC and all of the subclones except subclone 5 (Fig. 2A, lane 7). A 4.5-kb fragment containing the entire ΔLTKOSN vector with adjacent cellular sequences on both ends was detected by the same Neor DNA probe on all of the subclones and parental LTKOSN.2 VPC. In addition to the 4.5- and 2.2-kb bands detected in subclones 2 and 3, other bands that are larger than 2.2 kb should be ΔLTKOSN vector, since only one integration site of LTKOSN (3.7 kb) was detected by the HSVtk probe on the same membrane in subclones 2 and 3. This 3.7-kb fragment (arrow) containing the 5′ portion of LTKOSN with adjacent chromosomal sequences was detected in the LTKOSN.2 VPC and all subclones except subclone 5 (Fig. 2A, lane 7). Taken together, the original LTKOSN vector integration site, present in the parental LTKOSN.2 VPC and the other subclones, appears absent in subclone 5 (Fig. 2A, lane 7). Since subclone 5 still contains the same integration site (4.5-kb band) of ΔLTKOSN as seen in parental LTKOSN.2 VPC and the other subclones, subclone 5 is a true subclone derived from parental LTKOSN.2 VPC. We also suspect that subclone 5 may be derived from subclones 2 and 3, since subclone 5 has an integration pattern of ΔLTKOSN very similar to those of subclones 2 and 3.

Gene expression and packaging of ΔLTKOSN and LTKOSN.

After Northern blot analysis of RNA transcripts from subclones 1, 3, 4, 5, and 10 (subclone 2 was lost to contamination), the results revealed that the RNA transcript ratio of ΔLTKOSN to LTKOSN in all of the subclones was approximately 2:1 or greater (Fig. 3A). This ratio was observed in subclones 1, 4, and 10, which contain only one copy each of ΔLTKOSN and LTKOSN vector. This could result from differences in the integration sites or the smaller size of ΔLTKOSN. Previous reports demonstrated that the internal SV40 promoter, present in LTKOSN but deleted from ΔLTKOSN, can interfere with transcription from the 5′ LTR (48). Subclone 5, which contains the highest copy number of both ΔLTKOSN and LTKOSN vectors, showed more abundant vector transcripts than the other subclones. Unexpectedly, these abundant vector transcripts in subclones 1, 3, and 5 were not packaged into virions (Fig. 3B).

FIG. 3.

FIG. 3

Open in a new tab

Differential gene expression of vectors and helper virus in LTKOSN.2 VPC subclones. (A) Total RNA was extracted from LTKOSN.2 VPC subclones (lanes 3 to 7) and detected by Neor, env, and human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA probes. RNA transcripts of LTKOSN and ΔLTKOSN are 4.0 and 2.5 kb in size, respectively. RNA extracted from NIH 3T3 (lane 8) was used as a negative control. Hybridization with env probe detected helper virus gene expression. Compared to the spliced env transcript observed in all subclones, full-length MoMLV transcript (gag-pol-env) was detected in only LTKOSN.2 VPC (lane 2) and subclones 4 and 10 (lanes 5 and 7, respectively) but was barely visible in other subclones and PA317 cells. (B) Viral RNA extracted from virions was detected by Neor probe and showed that LTKOSN and ΔLTKOSN transcripts could be detected only in the supernatants collected from parental LTKOSN.2 VPC and subclones 4 and 10. Thus, despite abundant LTKOSN and ΔLTKOSN transcripts in the VPC, little or no vector was packaged and released from subclones 1, 3, and 5.

Transcriptional inactivation of helper virus.

To determine the mechanism for the observed reduction of vector production in subclones 1, 3, and 5, Northern blots of subclones were probed with amphotropic env DNA to evaluate the level of helper virus gene expression (Fig. 3). Spliced env transcripts were detected in all subclones, but significant levels of full-length helper virus transcripts were observed only in subclones 4 and 10 (Fig. 3A), which were previously shown to produce virions containing ΔLTKOSN and LTKOSN vectors (Fig. 3B). It has been reported that spliced RNA (e.g., env transcript) is more stable than precursor RNA (e.g., full-length helper virus) as a result of increased polyadenylation efficiency on RNA 3′ processing (13). Therefore, the overall low-level gene expression of helper virus and faster degradation of full-length helper virus transcripts may explain why only env transcripts were observed in subclones 1, 3, and 5, not full-length helper virus (Fig. 3A). These results demonstrated that the reduction of packaged ΔLTKOSN and LTKOSN vectors in subclones 1, 3, and 5 correlated closely with low levels of helper virus gene expression.

Superinfection of LTKOSN.2 VPC subclones.

To test our hypothesis that the suppression of helper virus gene expression reduces the Env-receptor interference and then results in vector reentry (superinfection), LTKOSN.2 VPC subclones were transduced with an EGFP expression vector, LEIN (see Materials and Methods). The transduction efficiency of LEIN vector on these subclones was determined by FACS analysis of EGFP expression compared to that in NIH 3T3, PA317, and parental LTKOSN.2 VPC (Fig. 4). NIH 3T3 cells that do not express Env protein on the cell surface were 21.9% EGFP positive 2 days after single exposure to LEIN vector. LTKOSN.2 VPC subclones with significant DNA methylation in the 5′ LTR region were transduced at rates comparable to those for NIH 3T3 cells (15.7 to 19.3% in subclones 1, 3, and 5). Subclones 4 and 10, which exhibited more detectable helper virus transcripts, were transduced at lower rates of only 0.6 and 1.4%, which were comparable to those of PA317 (0.2%) and parental LTKOSN.2 VPC (0.4%). These results demonstrate that gene expression of helper virus in these subclones correlated closely with the superinfection of LEIN vector. These results also suggested that increased copy numbers of LTKOSN and ΔLTKOSN in subclones 3 and 5 were due to superinfection. Subclone 1 showed significant inactivation of helper virus gene expression and superinfection of LEIN vector as well as subclones 3 and 5 but did not exhibit increased copy number of vectors. We suspect that the inactivation of helper virus gene expression by DNA methylation occurred during or after the subcloning procedure, since we did observe that the titer of subclone 4 was reduced to completely nondetectable levels by DNA methylation during a 31-day period of cell culture (see below).

FIG. 4.

FIG. 4

Open in a new tab

Superinfection susceptibility of LTKOSN.2 VPC subclones. Cells from NIH 3T3, PA317, parental LTKOSN.2, and LTKOSN.2 VPC subclones were seeded at 5 × 105 cells per well on 12-well plates and exposed to 0.5 ml of supernatant of EGFP-expressing vector, LEIN (3.5 × 105 CFU/ml; see Materials and Methods). FACS analysis of EGFP expression was performed 2 days after the single exposure of LEIN vector. The average rate of three transductions of each target cell line was as follows: NIH 3T3, 21.9%; PA317, 0.2%; LTKOSN.2, 0.4%; subclone 1, 18.4%; subclone 3, 15.7%; subclone 4, 0.6%; subclone 5, 19.3%; and subclone 10, 1.4%. Error bars were calculated from triplicate data with each target cell line.

Methylation status of helper virus and retroviral vectors.

DNA methylation has been demonstrated to be associated with the repression of retroviral vector gene expression in different cells and tissues in vitro (12, 22, 24) and in vivo (3). The methylation of CpG sequence(s) within an SmaI site by mammalian methyltransferase will prevent SmaI digestion (49). The methylation status of the SmaI site in the 5′ LTR has been employed to demonstrate an inverse correlation with the gene expression levels of retroviral vector (3). Therefore, the same analysis was performed to detect helper virus DNA methylation (Fig. 5) compared to helper virus gene expression in VPC subclones (Fig. 3A).

FIG. 5.

FIG. 5

Open in a new tab

DNA methylation status of helper virus and vectors. Genomic DNA digested with DraI and EcoRV was divided into two equal portions for SmaI digestion (lanes 1 to 9) or without SmaI digestion (lanes 10 to 18). (A) Hybridization of gag DNA probe to detect helper virus 5′ LTR. SmaI digestion reduced the 1.8-kb band (lanes 11 to 18) to 1.5 kb (lanes 2 to 9). DNA from NIH 3T3 cells was used to demonstrate the presence of endogenous retroviral elements (lanes 1 and 10). Since a 1.8-kb band was generated from endogenous retroviral element after SmaI digestion (lane 1), the values for SmaI resistance were measured by densitometry and calculated as 1 minus the relative intensities of the 1.5-kb bands (SmaI digestion; lanes 1 to 9) and the 1.8-kb bands (without SmaI digestion; lanes 10 to 18). Note the absence of helper virus methylation in parental LTKOSN.2 (lane 3) and completely methylated helper virus 5′ LTR in subclone 5 (lane 7). (B) Rehybridization of env DNA probe for the SmaI methylation status in helper virus RT region. SmaI resistance was measured as the relative intensities of the 4.6-kb and 3.3- plus 4.6-kb bands generated after SmaI digestion. Note that multiple copies of helper virus env sequences were detected in PA317 and derived cells. (C) HSVtk DNA probe was used to detect the 5′ LTR of LTKOSN vector. SmaI digestion reduced the 1.5-kb fragment to 1.2 kb. The varied signal intensities are secondary to the copy number of LTKOSN vector in each subclone. (D) Neor probe was used to determine the SmaI methylation status in both the HSVtk gene of LTKOSN and the 5′ LTR region of ΔLTKOSN. SmaI reduced the 2.4-kb fragment of ΔLTKOSN to 2.1 kb and the 2.1-kb band of LTKOSN to 1.5 kb. (E) A 0.68-kb DNA fragment from the gag gene digested by BstXI was used as a probe to detect a 1.2-kb band of endogenous retroviral sequence to demonstrate equivalent DNA loading in paired samples of SmaI digestion. (F) Schema of helper virus and vectors showing the locations of restriction enzyme sites and the probes used for the methylation analysis. D, DraI; E, EcoRV; S, SmaI; AAA, SV40 poly(A) signal. Drawings are not to scale.

Genomic DNA was first digested with EcoRV and DraI to generate a DNA fragment convenient for Southern blot analysis. An equal portion of this EcoRV/DraI-digested DNA was subjected to SmaI digestion. Since the gag region is not present in the retroviral vector, a 428-bp gag probe hybridizes only to the helper virus DNA (Fig. 5A and E). All of the subclones and parental VPC showed a 1.8-kb EcoRV/DraI band (lanes 11 to 18), which was reduced to a 1.5-kb band upon SmaI digestion if no methylation occurred at the SmaI site. This 1.5-kb band was not detected in either NIH 3T3 cells or subclone 5 (Fig. 5A, lanes 1 and 7) but was observed in all other samples. NIH 3T3 cells were used to show the presence of endogenous retroviral elements that hybridized with the gag probe (Fig. 5A, lanes 1 and 10). After SmaI digestion of NIH 3T3-derived DNA, a band of approximately 1.8 kb was generated from endogenous retroviral elements, complicating our analysis (Fig. 5A, lane 1). Therefore, the methylation status of the SmaI site was evaluated by the relative intensities of the observed 1.5-kb bands from SmaI/EcoRV/DraI digestions (lanes 2 to 9) compared to 1.8-kb bands from EcoRV/DraI digestions (lanes 11 to 18), which do not overlap in size with endogenous sequences (Fig. 5A, lane 10). Although 40% of helper virus 5′ LTR in PA317 is methylated (Fig. 5A, lane 2), complete digestion by SmaI shows the apparent absence of methylation of the helper virus 5′ LTR in parental LTKOSN.2 VPC (Fig. 5A, lane 3). This suggests that the prior selection of LTKOSN.2 VPC, a clone that was chosen for its high titer compared to those of the other VPC clones, is likely the result of both high-level expression of vector and low methylation of helper virus. A high degree (60%) of helper virus methylation is also found in the 5′ LTR of PG13, a GaLV pseudotyped VPC (Fig. 5A, lanes 9 and 18). This indicates that the methylation of helper virus 5′ LTR in PA317 is not a single packaging cell line phenomenon. In contrast to the low level of helper virus methylation in parental LTKOSN.2 VPC, higher degrees of helper virus DNA methylation were observed in the subclones (Fig. 5A). This high level of methylation was inversely proportional to the helper virus gene expression (Fig. 3A) and subsequently to the packaged vector titer (Fig. 3A and Table 1). Subclone 5 has an absence of the 1.5-kb band, which indicates that the methylation of the SmaI site in the helper virus 5′ LTR (Fig. 5A, lane 7) occurred in the entire subclone 5 population.

TABLE 1.

Titer of LTKOSN.2 subclones treated with 5-aza-C

Subclone 5-aza-C Target cell line
NIH 3T3 A375 IGROV
1 + 4 × 102 5 × 102 4 × 102
− 1 × 101 2 × 100 0
3 + 4 × 102 6 × 102 5 × 101
− 0 0 0
4 + 5 × 105 4 × 105 1 × 105
− 2.2 × 105 4 × 105 8 × 104
5 + 2 × 102 3 × 102 1.8 × 101
− 0 0 0
10 + 3 × 104 1 × 105 1.2 × 104
− 4 × 104 2 × 105 1.6 × 104
Open in a new tab

The methylation status of the SmaI site in the RT region of the helper virus was evaluated by the rehybridization of this Southern blot with a 1.4-kb env probe (Fig. 5B). In subclones 1, 3, and 5, which had highly methylated helper viral 5′ LTR, the methylation level of SmaI in the RT region ranged from 46 to 100%. In contrast, only 3 to 10% methylation was observed in parental LTKOSN.2 VPC and subclones 4 and 10. Thus, a strong correlation exists between the observed methylation of the SmaI site in 5′ LTR and within the RT gene (Fig. 5A and B). In addition to the 4.6-kb band in the DNA samples without SmaI digestion (lanes 10 to 18), three additional bands were observed in parental LTKOSN.2 VPC and all VPC subclones (lanes 2 to 8 and 10 to 17) but not in NIH 3T3 and PG13 (lanes 1 and 10 and lanes 9 and 18). These results demonstrate that the env probe is specific for the env sequence of helper virus pPAM3 and does not detect GaLV env or Moloney murine leukemia virus (MoMLV) gag-pol from PG13 cells. Furthermore, at least four copies of env sequence are present in the PA317 genome. These multiple copies of helper virus env sequence were confirmed by Southern blot analysis of PA317 genomic DNA with different methylation-insensitive restriction endonuclease digestions (data not shown). In contrast, hybridization with probes for either the HSVtk (Fig. 5C) or Neor (Fig. 5D) gene reveals no significant methylation within either LTKOSN or ΔLTKOSN vectors. Extremely low methylation of the 5′ LTR of LTKOSN and ΔLTKOSN vectors was observed only in subclone 5 (Fig. 5D, lane 7), which showed 100% methylation of the SmaI sites in both the helper virus 5′ LTR and RT region (Fig. 5A and B, lane 7).

Parental LTKOSN.2 VPC do not show significant methylation of either helper virus or vectors. However, the subclones derived from LTKOSN.2 VPC show various methylation levels of the helper virus but not of their retroviral vectors. These results suggest a sequential time line of methylation that developed within the helper virus DNA in these five subclones. These data also support our previously theorized cell lineage relationship, based upon the similarity of vector integration patterns observed in Fig. 2A.

Reversal of DNA methylation.

To reactivate silenced helper viruses, subclones were treated with 5-aza-C, a cytidine analog that can inhibit DNA methylase and reverse methylation to restore gene expression (8, 21, 23, 26). After coculture with 5-aza-C for 72 h, the rescue of vector RNA from these VPC subclones was demonstrated by RNA slot blot analysis of supernatants (Fig. 6). Partial recovery of vector production was also demonstrated by titer assay (Table 1) along with the partial reversal of DNA methylation by 5-aza-C treatment (Fig. 7). The titers of subclones 1, 3, and 5 were increased by 2 orders of magnitude from ≤10 CFU/ml to up to 6 × 102 CFU/ml. No significant change was observed in subclones 4 and 10, which exhibited high vector production and less helper virus methylation before 5-aza-C treatment. The most significant reduction of SmaI resistance was observed in subclone 5, in which the methylation of the 5′ LTR and RT regions was reduced from 100 to about 45% on a Southern blot (Fig. 7). In this model system, the methylation of helper virus is responsible for the reduction of vector production.

FIG. 6.

FIG. 6

Open in a new tab

Treatment with 5-aza-C partially restores the vector production ability in subclones. Cell cultures were coincubated with 5 μM 5-aza-C for 72 h. Supernatant (200 μl) collected from each of the cell cultures with or without 5-aza-C treatment was loaded onto the slot blot apparatus and UV cross-linked onto a nylon membrane. Neor probe was used to detect the LTKOSN and ΔLTKOSN vector transcripts to quantify the vector production ability of each subclone. Supernatant collected from NIH 3T3 was used as a negative control.

FIG. 7.

FIG. 7

Open in a new tab

Treatment with 5-aza-C reverses the DNA methylation of helper virus. (A and B) Genomic DNA digestion and hybridization with the gag (A) and env (B) DNA probes were performed the same way as described in the legend to Fig. 5 to evaluate the SmaI methylation status in the helper virus 5′ LTR and RT regions. Note the new appearance of a 1.5-kb band (A) and a 3.3-kb band (B) in subclone 5 (lane 4) that were not seen in Fig. 5, lane 7, before 5-aza-C treatment. (C) A 0.68-kb DNA fragment from the gag gene digested by BstXI was used as a probe to detect a 1.2-kb band of endogenous retroviral sequence to demonstrate equivalent DNA loading in paired samples of SmaI digestion.

DNA methylation rate of helper virus 5′ LTR.

To study the DNA methylation rate of helper virus 5′ LTR, we examined the DNA methylation of subclone 4, which showed the least DNA methylation with the most highly activated gene expression among these subclones. During 31 days of continuous cell culture, DNA methylation of helper virus 5′ LTR in subclone 4 increased rapidly from 36 to 100% (Fig. 8A). The methylation rate of the SmaI site in helper virus 5′ LTR was calculated to be as high as 2% of the cell population per day. The copy numbers of vector were increased from one copy (parental LTKOSN.2 VPC and subclone 4 on day 0) to five copies (day 7) and subsequently to 9 copies (days 23 and 31) per cell (Fig. 8B and C). The titer of subclone 4 was reduced from 1.5 × 105 CFU/ml (day 0) to 7 × 104 CFU/ml (day 7) and then to undetectable levels by days 23 and 31. The reduction in vector titer correlated with increasing inactivation of helper virus gene expression (Fig. 9A). However, vector gene expression still remained at high levels (Fig. 9B). Only minimal DNA methylation (10%) of 5′ LTR was observed in LTKOSN vector (Fig. 8B), and no detectable DNA methylation was noted in the ΔLTKOSN vector (Fig. 8C).

FIG. 8.

FIG. 8

Open in a new tab

Increased DNA methylation in helper virus of subclone 4 over time. Genomic DNA was first digested with DraI and EcoRV and divided into two equal portions for SmaI digestion as described in the legend to Fig. 5. (A) Hybridization of gag DNA probe to detect helper virus 5′ LTR. SmaI digestion reduced the 1.8-kb band (even-numbered lanes 4 to 14) to 1.5 kb (odd-numbered lanes 3 to 13). SmaI resistance was calculated as described for Fig. 5. (B) HSVtk DNA probe was used to detect the 5′ LTR of LTKOSN vector. SmaI digestion reduced the 1.5-kb fragment to 1.2 kb. The values for SmaI resistance were calculated as described for Fig. 5. The varied signal intensities are secondary to increasing copy numbers of LTKOSN vector over time. Copy number was estimated by comparing the relative intensity of HSVtk signals and intensities detected by BstXI fragment probe from endogenous retroviral elements in panel D to standardize loading. Increased copy numbers were observed on days 7 (five copies, lane 10) and 23 and 31 (nine copies, lanes 12 and 14), while only one copy in parental LTKOSN.2 and subclone 4 was detected on day 0 (lanes 6 and 8). (C) Neor probe was used to determine the SmaI methylation status in both the HSVtk gene of LTKOSN and the 5′ LTR region of ΔLTKOSN. (D) A 0.68-kb DNA fragment from the gag gene digested by BstXI was used as probe to detect a 1.2-kb band of endogenous retroviral sequence to demonstrate equivalent DNA loading in paired samples of SmaI digestion.

FIG. 9.

FIG. 9

Open in a new tab

Transcription activity of helper virus and retroviral vectors in LTKOSN.2 VPC subclone 4. (A) Northern blot analysis of cellular RNA extracted from LTKOSN.2 VPC subclone 4 at different time points and hybridized with env probe. Unspliced MoMLV transcripts (gag-pol-env) were detected at different sizes as a result of the presence of at least four different integrated pPAM3 plasmids detected in PA317 with different adjacent sequences. (B) Gene expression of LTKOSN and ΔLTKOSN vectors. Rehybridization of this Northern blot membrane with Neor probe to detect LTKOSN vector (4.0 kb), ΔLTKOSN vector (2.5 kb), and a Neor RNA transcript (1.2 kb) derived from internal SV40 promoter activity was carried out. (C) Hybridization with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA probe demonstrated similar levels of RNA loading.

DISCUSSION

The genetic stability of retroviral vectors and a VPC genome was examined in LTKOSN.2 VPC and derived subclones. We also compared the methylation status of helper virus and vectors in these individual subclones. Our results demonstrate that DNA methylation occurred in helper viral sequences rather than vector in subclones derived from LTKOSN.2 VPC (Fig. 5). The increase in methylation of the helper virus (Fig. 5 and 8) correlated directly with the inactivation of helper virus gene expression (Fig. 3 and 9), which resulted in a loss of vector production. In contrast, vector gene expression was constant, and no significant methylation of vector sequences was detected. We conclude that the key limitation for vector production in LTKOSN.2 VPC is the inactivation of helper virus gene expression by DNA methylation. This conclusion was supported by the partial restoration of vector production (Table 1) and the reversal of DNA methylation (Fig. 7) from VPC subclones treated with 5-aza-C. In previous studies, de novo DNA methylation has been suggested to cause transcriptional inactivation of retroviral vectors in infected embryonic cells (3, 12, 22). DNA methylation had not been a focus of investigation with regard to the genetic instability of retroviral VPC that are designed for continuous vector production. A cascade of complex events resulting in the genetic instability of VPC caused by the methylation of helper virus 5′ LTR is summarized in Fig. 10. We propose that the cascade occurs in the following sequence: (i) repression of helper virus 5′ LTR transcriptional activity by methylation reduces the production of vector, (ii) decreased Env protein synthesis reduces Env-receptor interference and allows vector superinfection to cause increased vector copy number in VPC, and (iii) random integration of vectors caused by superinfection results in a more fluid VPC cellular genome.

FIG. 10.

FIG. 10

Open in a new tab

Cascade reactions initiated by increased DNA methylation of helper virus 5′ LTR. DNA methylation inactivates helper virus gene expression, and then vector production is reduced. Without sufficient Env-receptor interference, susceptibility to superinfection of vector from its own or other VPC in the same population is increased and is seen in the increased vector copy number.

The preference for methylation of the helper virus rather than the retroviral vectors is not clear and requires further study. However, chromosomal location has been suggested as a key factor for facilitating methylation (12, 20). In PA317, four copies of env sequences of pPAM3 helper virus were observed (Fig. 5B). Restriction analysis of these helper virus sequences demonstrated that the sequences are located at different chromosomal sites (data not shown). The probability that all four copies of helper virus sequences are located within a hypermethylation region of PA317 seems remote. No significant methylation of vectors was observed in subclones 3 and 5, even though these subclones contain multiple copies of vectors in different integration sites. About 100 copies of integrated MoMLV provirus have been observed in infected murine embryonic carcinoma cells, and these proviruses were hypermethylated with no significant gene expression (55). Presumably, these 100 copies of provirus were not all located in hypermethylation regions. Therefore, the preference of methylation for helper virus is probably related to both helper virus sequence and chromosomal location. The idea that sequence has an important role is also suggested by the findings that the presence of retrotransposon and provirus sequences enhances the de novo DNA methylation of adjacent sequences (11, 16). Most recently, evidence that DNA methylation acts as a defense system against retrotransposons invading the mammalian genome has been shown elsewhere (44). Infection of retrovirus, such as human immunodeficiency virus, increased the overall DNA methyltransferase gene expression and activity, which resulted in suppression of cytokine gene expression for evading immune surveillance (30). The spreading of de novo methylation from a provirus into adjacent cellular sequences was observed in either the 5′ or 3′ region of the integration site in Mov-derived mice carrying a MoMLV provirus in a distinct chromosomal location (16). In the pPAM3 LTR, the methylation observed could be the result of spreading of methylation from adjacent gag sequences, since the LTR in either LTKOSN or ΔLTKOSN vector did not demonstrate any significant methylation. However, we have not excluded a role of chromosomal position and preferred DNA methylation (12). These subclones of VPC may provide a useful model to study DNA methylation preferences that are key in host-retrovirus interactions (11, 17, 18, 30, 44).

Alternatively, G418 selection of VPC could have eliminated VPC cells with methylated vectors. After the establishment of LTKOSN.2 VPC, VPC were not maintained on G418 selection. Other LTKOSN parental lines require only a single copy of the Neor gene for G418 resistance. Subclones 3 and 5 contained multiple copies of LTKOSN and ΔLTKOSN. Even if partial methylation of the vector, as noted for the helper virus, had occurred, adequate Neor gene expression would nevertheless provide G418 resistance. However, almost no methylation was observed in either LTKOSN or ΔLTKOSN. An extremely low degree of methylation was observed only in subclone 5, which contained approximately 29 copies of integrated vectors. This could be attributed to the random integration of some vectors into hypermethylation regions of the cellular genome. In contrast to the moderate degree of methylation (40%) of the helper virus 5′ LTR in PA317, the helper virus in parental LTKOSN.2 VPC did not show significant methylation of SmaI sites in the 5′ LTR. This demonstrates that a diverse range of methylation statuses exists within PA317 cells and that the selection of the highest-titer LTKOSN VPC (LTKOSN.2) from among the other 39 clones (see Materials and Methods) may have selected for a clone with hypomethylated helper viruses.

It has been well documented that Env-receptor interference is important to reducing the risk of superinfection (31, 33, 54, 60). Recent data demonstrated that at least four copies of retrovirus per cell are required for sufficient Env protein synthesis to interfere with the cellular receptor (42). Superinfection experiments on these LTKOSN.2 subclones (Fig. 4) demonstrated that DNA methylation (Fig. 5) and helper virus mRNA level (Fig. 3) are highly correlated with the reduction of Env-receptor interference. The reduction of env gene expression by DNA methylation would logically result in the reentry of multiple vectors into subclones 3 and 5. Subclone 1 may represent a transitional state between the parental LTKOSN.2 and subclone 3, since only one copy of each LTKOSN and ΔLTKOSN vector is present while helper virus gene expression is low.

DNA methylation in subclones 4 and 10 (38 and 50%, respectively) and PA317 (40%) is significantly higher than that in parental LTKOSN.2 cells (0%). However, the resistance to superinfection in subclones 4, 10, and PA317 is still comparable to that observed in parental LTKOSN.2 cells (Fig. 4). A significant increase in superinfection was observed in subclones only when at least 60% methylation of the helper virus 5′ LTR occurred. This suggests a possible threshold effect. It has been demonstrated that the amount of Env product per cell, the interference, and virion release did not increase linearly but increased abruptly once an infected cell reached the threshold of proviral copy number which correlated with mRNA level (42). This is due to the fact that oligomerization of Env protein is essential for functional virus assembly and interference (2, 14, 42). Since the concentration of Env protein determines the kinetics of oligomerization, reduction of helper virus mRNA level by DNA methylation resulted in low Env protein concentration and oligomerization and, therefore, significantly decreased the interference and virion release. Compared to the previously reported high level of resistance to the superinfection of PA317, which was transduced with the vector collected from another PA317 VPC (33), the apparent reentry frequency of vectors in LTKOSN.2 VPC is relatively high (50% of population, three out of six subclones [Fig. 1]). This superinfection was enhanced by both the DNA methylation of helper virus and the presence of extensive vector exposure in cell culture. It has also been observed that human immunodeficiency virus provirus was accumulated in infected cells to multiple copies by superinfection of 70% of cells in the culture over time (45).

Amphotropic murine leukemia virus vectors are used widely for human gene transfer. However, most amphotropic murine leukemia virus VPC are derived from murine cells and are susceptible to vector reentry. The consequence of vector reentry is the import of active RT enzyme which may lead to RCR formation and the random integration of additional vectors that can interrupt functional genes or activate proto-oncogenes (9, 15, 25, 52). Our results imply that DNA methylation decreases the Env-receptor interference, which results in vector reintegration and subsequent further instability of the VPC genome. Recently, progress in this direction has been made: PG13 VPC were established using NIH 3T3 cells for packaging xenotropic GaLV-pseudotyped vector. GaLV packaged vector will not reinfect PG13 cells that lack the receptor for GaLV envelope (34, 61). PG13 VPC may diminish the vector reentry problem; however, helper virus gene expression in PG13 is still not monitored and selected by a selection marker. Our data with PG13 show significant DNA methylation (60%) of helper virus 5′ LTR (Fig. 5A, lanes 9 and 18). The observations in this study strongly suggest that improvement of the helper virus gene expression in overcoming DNA methylation in VPC is important for sustained, stable vector production. Treatment with 5-aza-C of LTKOSN.2 subclones did not completely restore the titer (Table 1). The effect of 5-aza-C treatment is transient (8), and the treatment cannot be applied to long-term cell culture since 5-aza-C is toxic to the treated cells (21). Several alternative strategies that ensure sustained vector gene expression in target cells might improve the gene expression of helper virus in VPC. Insertion of a demethylation fragment of the murine Thy-1 gene in front of 5′ LTR can inhibit methylation (4, 56) and may be useful for helper virus. Another strategy is to ligate an IRES (19) with a selection marker gene downstream of the helper virus so that drug selection would ensure active enhancer-promoter function of helper virus for high vector packaging and preventing superinfection by enhancing Env-receptor interference (W.-B. Young and C. J. Link, Jr., submitted for publication).

ACKNOWLEDGMENTS

We thank John Levy, A. Dusty Miller, and Tatiana Seregina for helpful discussions. We also thank Ginger Dreifurst and Julie Seiwert for technical assistance with flow cytometry.

This work was supported in part by a grant from the Iowa Health System, Des Moines, and Research Project Grant RPG-98-091-01-MBC from the American Cancer Society. W.-B. Young is a recipient of a competitive research fellowship from the Molecular, Cellular and Developmental Biology Program, Iowa State University, Ames.

REFERENCES

  • 1.Anderson W F, McGarrity G J, Moen R C. Report to the NIH Recombinant DNA Advisory Committee on murine replication-competent retrovirus (RCR) assays. Hum Gene Ther. 1993;4:311–321. doi: 10.1089/hum.1993.4.3-311. [DOI] [PubMed] [Google Scholar]
  • 2.Bernstein H B, Tucker S P, Kar S R, McPherson S A, McPherson D T, Dubay J W, Lebowitz J, Compans R W, Hunter E. Oligomerization of the hydrophobic heptad repeat of gp41. J Virol. 1995;69:2745–2750. doi: 10.1128/jvi.69.5.2745-2750.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Challita P M, Kohn D B. Lack of expression from a retroviral vector after transduction of murine hematopoietic stem cells is associated with methylation in vivo. Proc Natl Acad Sci USA. 1994;91:2567–2571. doi: 10.1073/pnas.91.7.2567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Challita P M, Skelton D, el-Khoueiry A, Yu X J, Weinberg K, Kohn D B. Multiple modifications in cis elements of the long terminal repeat of retroviral vectors lead to increased expression and decreased DNA methylation in embryonic carcinoma cells. J Virol. 1995;69:748–755. doi: 10.1128/jvi.69.2.748-755.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Cone R D, Mulligan R C. High-efficiency gene transfer into mammalian cells: generation of helper-free recombinant retrovirus with broad mammalian host range. Proc Natl Acad Sci USA. 1984;81:6349–6353. doi: 10.1073/pnas.81.20.6349. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Cormack B P, Valdivia R H, Falkow S. FACS-optimized mutants of the green fluorescent protein (GFP) Gene. 1996;173:33–38. doi: 10.1016/0378-1119(95)00685-0. [DOI] [PubMed] [Google Scholar]
  • 7.Donahue R E, Kessler S W, Bodine D, McDonagh K, Dunber C, Goodman S, Agricola B, Byrne E, Raffeld M, Moen R, Bacher J, Zsebo K M, Nienhuis A W. Helper virus induced T cell lymphoma in nonhuman primates after retroviral mediated gene transfer. J Exp Med. 1992;176:1125–1135. doi: 10.1084/jem.176.4.1125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Gram G J, Nielsen S D, Hansen J E. Spontaneous silencing of humanized green fluorescent protein (hGFP) gene expression from a retroviral vector by DNA methylation. J Hematother. 1998;7:333–341. doi: 10.1089/scd.1.1998.7.333. [DOI] [PubMed] [Google Scholar]
  • 9.Gray D A. Insertional mutagenesis: neoplasia arising from retroviral integration. Cancer Investig. 1991;9:295–304. doi: 10.3109/07357909109021327. [DOI] [PubMed] [Google Scholar]
  • 10.Haas J, Park E C, Seed B. Codon usage limitation in the expression of HIV-1 envelope glycoprotein. Curr Biol. 1996;6:315–324. doi: 10.1016/s0960-9822(02)00482-7. [DOI] [PubMed] [Google Scholar]
  • 11.Hasse A, Schulz W A. Enhancement of reporter gene de novo methylation by DNA fragments from the alpha-fetoprotein control region. J Biol Chem. 1994;269:1821–1826. [PubMed] [Google Scholar]
  • 12.Hoeben R C, Migchielsen A A, van der Jagt R C, van Ormondt H, van der Eb A J. Inactivation of the Moloney murine leukemia virus long terminal repeat in murine fibroblast cell lines is associated with methylation and dependent on its chromosomal position. J Virol. 1991;65:904–912. doi: 10.1128/jvi.65.2.904-912.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Huang M T, Gorman C M. Intervening sequences increase efficiency of RNA 3′ processing and accumulation of cytoplasmic RNA. Nucleic Acids Res. 1990;18:937–947. doi: 10.1093/nar/18.4.937. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Hunter E. Macromolecular interactions in the assembly of HIV and other retroviruses. Semin Virol. 1994;5:71–83. [Google Scholar]
  • 15.Ihle J N, Smith-White B, Sisson B, Parker D, Blair D G, Schultz A, Kozak C, Lunsford R D, Askew D, Weinstein Y, Isfort R J. Activation of the c-H-ras proto-oncogene by retrovirus insertion and chromosomal rearrangement in a Moloney leukemia virus-induced T-cell leukemia. J Virol. 1989;63:2959–2966. doi: 10.1128/jvi.63.7.2959-2966.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Jahner D, Jaenisch R. Retrovirus-induced de novo methylation of flanking host sequences correlates with gene inactivity. Nature. 1985;315:594–597. doi: 10.1038/315594a0. [DOI] [PubMed] [Google Scholar]
  • 17.Jähner D, Jäenisch R. Chromosomal position and specific demethylation in enhancer sequences of germ line-transmitted retroviral genomes during mouse development. Mol Cell Biol. 1985;5:2212–2220. doi: 10.1128/mcb.5.9.2212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Jähner D, Stuhlmann H, Stewart C L, Harbers K, Lohler J, Simon I, Jaenisch R. De novo methylation and expression of retroviral genomes during mouse embryogenesis. Nature. 1982;298:623–628. doi: 10.1038/298623a0. [DOI] [PubMed] [Google Scholar]
  • 19.Jang S K, Pestova T V, Hellen C U, Witherell G W, Wimmer E. Cap-independent translation of picornavirus RNAs: structure and function of the internal ribosomal entry site. Enzyme. 1990;44:292–309. doi: 10.1159/000468766. [DOI] [PubMed] [Google Scholar]
  • 20.Jolly D J, Willis R C, Friedmann T. Variable stability of a selectable provirus after retroviral vector gene transfer into human cells. Mol Cell Biol. 1986;6:1141–1147. doi: 10.1128/mcb.6.4.1141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Juttermann R, Li E, Jaenisch R. Toxicity of 5-aza-2′-deoxycytidine to mammalian cells is mediated primarily by covalent trapping of DNA methyltransferase rather than DNA demethylation. Proc Natl Acad Sci USA. 1994;91:11797–11801. doi: 10.1073/pnas.91.25.11797. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Kempler G, Freitag B, Berwin B, Nanassy O, Barklis E. Characterization of the Moloney murine leukemia virus stem cell-specific repressor binding site. Virology. 1993;193:690–699. doi: 10.1006/viro.1993.1177. [DOI] [PubMed] [Google Scholar]
  • 23.Kuriyama S, Sakamoto T, Kikukawa M, Nakatani T, Toyokawa Y, Tsujinoue H, Ikenaka K, Fukui H, Tsujii T. Expression of a retrovirally transduced gene under control of an internal housekeeping gene promoter does not persist due to methylation and is restored partially by 5-azacytidine treatment. Gene Ther. 1998;5:1299–1305. doi: 10.1038/sj.gt.3300738. [DOI] [PubMed] [Google Scholar]
  • 24.Laker C, Meyer J, Schopen A, Friel J, Heberlein C, Ostertag W, Stocking C. Host cis-mediated extinction of a retrovirus permissive for expression in embryonal stem cells during differentiation. J Virol. 1998;72:339–348. doi: 10.1128/jvi.72.1.339-348.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Lazo P A, Tsichlis P N. Recombination between two integrated proviruses, one of which was inserted near c-myc in a retrovirus-induced rat thymoma: implications for tumor progression. J Virol. 1988;62:788–794. doi: 10.1128/jvi.62.3.788-794.1988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Lengauer C, Kinzler K W, Vogelstein B. DNA methylation and genetic instability in colorectal cancer cells. Proc Natl Acad Sci USA. 1997;94:2545–2550. doi: 10.1073/pnas.94.6.2545. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Link C J, Jr, Kolb E M, Muldoon R R. Preliminary in vitro efficacy and toxicity studies of the herpes simplex thymidine kinase gene system for the treatment of breast cancer. Hybridoma. 1995;14:143–147. doi: 10.1089/hyb.1995.14.143. [DOI] [PubMed] [Google Scholar]
  • 28.Link C J, Jr, Moorman D, Seregina T, Levy J P, Schabold K J. A phase I trial of in vivo gene therapy with the herpes simplex thymidine kinase/ganciclovir system for the treatment of refractory or recurrent ovarian cancer. Hum Gene Ther. 1996;7:1161–1179. doi: 10.1089/hum.1996.7.9-1161. [DOI] [PubMed] [Google Scholar]
  • 29.Markowitz D, Goff S, Bank A. A safe packaging line for gene transfer: separating viral genes on two different plasmids. J Virol. 1988;62:1120–1124. doi: 10.1128/jvi.62.4.1120-1124.1988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Mikovits J A, Young H A, Vertino P, Issa J P, Pitha P M, Turcoski-Corrales S, Taub D D, Petrow C L, Baylin S B, Ruscetti F W. Infection with human immunodeficiency virus type 1 upregulates DNA methyltransferase, resulting in de novo methylation of the gamma interferon (IFN-γ) promoter and subsequent downregulation of IFN-γ production. Mol Cell Biol. 1998;18:5166–5177. doi: 10.1128/mcb.18.9.5166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Miller A D. Cell-surface receptors for retroviruses and implications for gene transfer. Proc Natl Acad Sci USA. 1996;93:11407–11413. doi: 10.1073/pnas.93.21.11407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Miller A D, Buttimore C. Redesign of retrovirus packaging cell lines to avoid recombination leading to helper virus production. Mol Cell Biol. 1986;6:2895–2902. doi: 10.1128/mcb.6.8.2895. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Miller A D, Chen F. Retrovirus packaging cells based on 10A1 murine leukemia virus for production of vectors that use multiple receptors for cell entry. J Virol. 1996;70:5564–5571. doi: 10.1128/jvi.70.8.5564-5571.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Miller A D, Garcia J V, Von Shur N, Lynch C M, Wilson C, Eiden M V. Construction and properties of retrovirus packaging cells based on gibbon ape leukemia virus. J Virol. 1991;65:2220–2224. doi: 10.1128/jvi.65.5.2220-2224.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Miller A D, Rosman G J. Improved retroviral vectors for gene transfer and expression. BioTechniques. 1989;7:980–982. , 984–986, 989–990. [PMC free article] [PubMed] [Google Scholar]
  • 36.Miller A D, Trauber D R, Buttimore C. Factors involved in production of helper virus-free retrovirus vectors. Somat Cell Mol Genet. 1986;12:175–183. doi: 10.1007/BF01560664. [DOI] [PubMed] [Google Scholar]
  • 37.Moolten F L. Tumor chemosensitivity conferred by inserted herpes thymidine kinase genes: paradigm for a prospective cancer control strategy. Cancer Res. 1986;46:5276–5281. [PubMed] [Google Scholar]
  • 38.Muenchau D D, Freeman S M, Cornetta K, Zwiebel J A, Anderson W F. Analysis of retroviral packaging lines for generation of replication-competent virus. Virology. 1990;176:262–265. doi: 10.1016/0042-6822(90)90251-l. [DOI] [PubMed] [Google Scholar]
  • 39.Muldoon R R, Levy J P, Kain S R, Kitts P A, Link C J., Jr Tracking and quantitation of retroviral-mediated transfer using a completely humanized, red-shifted green fluorescent protein gene. BioTechniques. 1997;22:162–167. doi: 10.2144/97221rr03. [DOI] [PubMed] [Google Scholar]
  • 40.Nelson D M, Wahlfors J J, Chen L, Onodera M, Morgan R A. Characterization of diverse viral vector preparations, using a simple and rapid whole-virion dot-blot method. Hum Gene Ther. 1998;9:2401–2405. doi: 10.1089/hum.1998.9.16-2401. [DOI] [PubMed] [Google Scholar]
  • 41.Nielsen C S, Moorman D W, Levy J P, Link C J., Jr Herpes simplex thymidine kinase gene transfer is required for complete regression of murine colon adenocarcinoma. Am Surg. 1997;63:617–620. [PubMed] [Google Scholar]
  • 42.Odawara T, Oshima M, Doi K, Iwamoto A, Yoshikura H. Threshold number of provirus copies required per cell for efficient virus production and interference in Moloney murine leukemia virus-infected NIH 3T3 cells. J Virol. 1998;72:5414–5424. doi: 10.1128/jvi.72.7.5414-5424.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Oldfield E H, Ram Z, Culver K W, Blaese R M, DeVroom H L, Anderson W F. Gene therapy for the treatment of brain tumors using intra-tumoral transduction with the thymidine kinase gene and intravenous ganciclovir. Hum Gene Ther. 1993;4:39–69. doi: 10.1089/hum.1993.4.1-39. [DOI] [PubMed] [Google Scholar]
  • 44.O'Neill R, O'Neill M, Graves J A. Undermethylation associated with retroelement activation and chromosome remodelling in an interspecific mammalian hybrid. Nature. 1998;393:68–72. doi: 10.1038/29985. [DOI] [PubMed] [Google Scholar]
  • 45.Ott D E, Nigida S M, Jr, Henderson L E, Arthur L O. The majority of cells are superinfected in a cloned cell line that produces high levels of human immunodeficiency virus type 1 strain MN. J Virol. 1995;69:2443–2450. doi: 10.1128/jvi.69.4.2443-2450.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Otto E, Jones-Trower A, Vanin E F, Stambaugh K, Mueller S N, Anderson W F, McGarrity G J. Characterization of a replication-competent retrovirus resulting from recombination of packaging and vector sequences. Hum Gene Ther. 1994;5:567–575. doi: 10.1089/hum.1994.5.5-567. [DOI] [PubMed] [Google Scholar]
  • 47.Pear W S, Nolan G P, Scott M L, Baltimore D. Production of high-titer helper-free retroviruses by transient transfection. Proc Natl Acad Sci USA. 1993;90:8392–8396. doi: 10.1073/pnas.90.18.8392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Robbins P B, Yu X J, Skelton D M, Pepper K A, Wasserman R M, Zhu L, Kohn D B. Increased probability of expression from modified retroviral vectors in embryonal stem cells and embryonal carcinoma cells. J Virol. 1997;71:9466–9474. doi: 10.1128/jvi.71.12.9466-9474.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Roberts R J. Restriction enzymes and their isoschizomers. Nucleic Acids Res. 1990;18(Suppl.):2331–2365. doi: 10.1093/nar/18.suppl.2331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Roth J A, Cristiano R J. Gene therapy for cancer: what have we done and where are we going? J Natl Cancer Inst. 1997;89:21–39. doi: 10.1093/jnci/89.1.21. [DOI] [PubMed] [Google Scholar]
  • 51.Sambrook J, Fritsch E F, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory Press; 1989. pp. 9.14–9.23. [Google Scholar]
  • 52.Schwartz J R, Duesberg S, Duesberg P. DNA recombination is sufficient for retroviral transduction. Proc Natl Acad Sci USA. 1995;92:2460–2464. doi: 10.1073/pnas.92.7.2460. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Seperack P K, Strobel M C, Corrow D J, Jenkins N A, Copeland N G. Somatic and germ-line mutation rates of the retrovirus-induced dilute coat-color mutation of DBA mice. Proc Natl Acad Sci USA. 1988;85:189–192. doi: 10.1073/pnas.85.1.189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Sommerfelt M A, Weiss R A. Receptor interference groups of 20 retroviruses plating on human cells. Virology. 1990;176:58–69. doi: 10.1016/0042-6822(90)90230-o. [DOI] [PubMed] [Google Scholar]
  • 55.Stewart C L, Stuhlmann H, Jahner D, Jaenisch R. De novo methylation, expression, and infectivity of retroviral genomes introduced into embryonal carcinoma cells. Proc Natl Acad Sci USA. 1982;79:4098–4102. doi: 10.1073/pnas.79.13.4098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Szyf M, Tanigawa G, McCarthy P L., Jr A DNA signal from the Thy-1 gene defines de novo methylation patterns in embryonic stem cells. Mol Cell Biol. 1990;10:4396–4400. doi: 10.1128/mcb.10.8.4396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Teyssier J R, Benard J, Ferre D, Da Silva J, Renaud L. Drug-related chromosomal changes in chemoresistant human ovarian carcinoma cells. Cancer Genet Cytogenet. 1989;39:35–43. doi: 10.1016/0165-4608(89)90227-6. [DOI] [PubMed] [Google Scholar]
  • 58.Vanin E F, Kaloss M, Broscius C, Nienhuis A W. Characterization of replication-competent retroviruses from nonhuman primates with virus-induced T-cell lymphomas and observations regarding the mechanism of oncogenesis. J Virol. 1994;68:4241–4250. doi: 10.1128/jvi.68.7.4241-4250.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Varmus H E, Quintrell N, Ortiz S. Retroviruses as mutagens: insertion and excision of a nontransforming provirus alter expression of a resident transforming provirus. Cell. 1981;25:23–36. doi: 10.1016/0092-8674(81)90228-2. [DOI] [PubMed] [Google Scholar]
  • 60.Weiss R A, Tailor C S. Retrovirus receptors. Cell. 1995;82:531–533. doi: 10.1016/0092-8674(95)90024-1. [DOI] [PubMed] [Google Scholar]
  • 61.Wilson C, Reitz M S, Okayama H, Eiden M V. Formation of infectious hybrid virions with gibbon ape leukemia virus and human T-cell leukemia virus retroviral envelope glycoproteins and the gag and pol proteins of Moloney murine leukemia virus. J Virol. 1989;63:2374–2378. doi: 10.1128/jvi.63.5.2374-2378.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Young, W.-B., E. J. Beecham, G. L. Lindberg, and C. J. Link, Jr. Restriction mapping of retroviral vector episomal DNA. BioTechniques, in press. [DOI] [PubMed]

Articles from Journal of Virology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES