Precise Identification of Endogenous Proviruses of
http://www.100md.com
病菌学杂志 2005年第8期
0Laboratory of Persistent Viral Diseases, Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, Hamilton, Montana 59840
ABSTRACT
Polytropic murine leukemia viruses (MuLVs) are generated by recombination of ecotropic MuLVs with env genes of a family of endogenous proviruses in mice, resulting in viruses with an expanded host range and greater virulence. Inbred mouse strains contain numerous endogenous proviruses that are potential donors of the env gene sequences of polytropic MuLVs; however, the precise identification of those proviruses that participate in recombination has been elusive. Three different structural groups of proviruses in NFS/N mice have been described and different ecotropic MuLVs preferentially recombine with different groups of proviruses. In contrast to other ecotropic MuLVs such as Friend MuLV or Akv that recombine predominantly with a single group of proviruses, Moloney MuLV (M-MuLV) recombines with at least two distinct groups. In this study, we determined that only three endogenous proviruses, two of one group and one of another group, are major participants in recombination with M-MuLV. Furthermore, the distinction between the polytropic MuLVs generated by M-MuLV and other ecotropic MuLVs is the result of recombination with a single endogenous provirus. This provirus exhibits a frameshift mutation in the 3' region of the surface glycoprotein-encoding sequences that is excluded in recombinants with M-MuLV. The sites of recombination between the env genes of M-MuLV and endogenous proviruses were confined to a short region exhibiting maximum homology between the ecotropic and polytropic env sequences and maximum stability of predicted RNA secondary structure. These observations suggest a possible mechanism for the specificity of recombination observed for different ecotropic MuLVs.
INTRODUCTION
Vertebrate species harbor a very large number of endogenous retroviral elements in their genomes that have originated from the infrequent insertion of exogenous viruses into the germ line. It is estimated that 8% of the human genome is composed of retrovirus sequences and that the number of distinct retroviral elements approaches the total number of structural genes identified in the genome (35). Endogenous retroviruses or their gene products are differentially expressed in tissues and in certain physiological or pathological conditions (3, 13, 46, 51, 57, 69); however, the control and consequences of such expression are poorly understood. It has been suggested that endogenous retrovirus expression may be involved in human pathogenesis, including neoplasms (40, 41) and autoimmune disorders (44), as well as normal physiological processes such as placental development (43).
One of the more extensively investigated groups of endogenous retroviruses are those participating in recombination with exogenous ecotropic retroviruses in mice giving rise to host range variants. Ecotropic murine leukemia viruses (MuLVs) are infectious for murine cells but not cells of other species. Upon infection of mice, these MuLVs frequently recombine with members of a group of endogenous proviruses to generate polytropic MuLVs (18, 19, 28, 30, 54, 55). Recombination involves substitution of the env gene sequences encoding the receptor-binding region of the ecotropic MuLV with analogous env gene sequences of the endogenous proviruses (1, 4, 7, 9, 16, 18, 19, 34, 52, 62). The recombinant MuLVs utilize a different cell-surface receptor for infectious entry and are capable of infecting murine cells as well as cells of other species, including humans.
The generation of polytropic viruses is intimately involved in a number of proliferative diseases in mice including lymphocytic leukemia induced by Moloney MuLV (M-MuLV) (5, 12, 64), erythroleukemia induced by the Friend MuLV (F-MuLV) (55), and spontaneous leukemia observed in highly leukemic mouse strains (6, 28, 59, 61, 62, 65). In addition, polytropic MuLVs may contribute to neurological disorders (48).
There are about 20 to 40 endogenous proviruses in different inbred strains of mice that bear very close homology to the sequences found in polytropic recombinant viruses; however, the participation of the proviruses in recombination with ecotropic viruses is not random. Polytropic MuLVs belong to one of two antigenic subclasses based on their reactivity to one of two monoclonal antibodies (MAbs), termed Hy 7 and MAb 516 (39). Reactivities to the antibodies are mutually exclusive and signal recombination with distinct structural groups of endogenous proviruses. The ecotropic viruses Akv and F-MuLV generate predominantly Hy 7-reactive polytropic MuLVs with few MAb 516-reactive viruses, whereas infection with M-MuLV results in approximately equal numbers of Hy 7- and MAb 516-reactive polytropic viruses (37, 38). Thus, M-MuLV recombines with a distinct structural group of endogenous proviruses reactive with MAb 516.
There are a number of possible mechanisms for the specificity of recombination with different ecotropic MuLVs, ranging from tissue-specific expression of the proviruses to specific interactions of the ecotropic MuLVs with the endogenous proviruses or their RNA transcripts. To gain an understanding of this specificity, it is first necessary to obtain a more complete description of the phenomenon, namely, the precise identification of those proviruses that actually participate in recombination. Efforts to precisely identify the endogenous origin of polytropic MuLVs have been hampered by the very close homology of the env gene sequences of the endogenous proviruses and the rate of divergence of these sequences by random mutation once they are incorporated into a replicating MuLV (31, 32, 33, 58). In our previous study, we found that nearly all of the endogenous proviruses of NFS/N mice were distinguishable by sequence microheterogeneity (22). Thus, if the divergence of polytropic MuLV sequences could be minimized, comparisons to env gene sequences in recombinant polytropic MuLVs should identify which of the proviruses participate in recombination. In the present study, we utilized procedures to minimize the number of replication cycles prior to sequence analyses of the recombinant viruses and were able to unambiguously identify the endogenous proviral origin of numerous polytropic MuLVs derived after inoculation of M-MuLV. We found that a surprisingly small number of endogenous proviruses are major participants in recombination with M-MuLV and that recombination with a single provirus accounts for the difference in the polytropic MuLVs generated after inoculation of M-MuLV compared to other ecotropic viruses. Furthermore, examination of the recombination patterns of the polytropic MuLVs suggested a mechanism that may account for this specificity. Additional findings suggest that the polytropic MuLVs, once generated, may continue to recombine with one another and evolve to more virulent forms in the host.
MATERIALS AND METHODS
Mice. NFS/N mice were maintained as an inbred strain at Rocky Mountain Laboratories, Veterinary Branch. All animals were treated in accordance with National Institutes of Health regulations and the guidelines of the Animal Care and Use Committee of Rocky Mountain Laboratories.
Isolation of recombinant polytropic MuLVs. Polytropic MuLVs were isolated from NFS/N mice that had been inoculated within 24 h of birth with the ecotropic virus M-MuLV1387 (20). The infected mice were sacrificed 3 to 5 weeks after inoculation, the spleens and thymuses were dissociated, and serial dilutions of cells from the tissues were distributed as infectious centers on NIH 3T3, Mus dunni (36), or mink (ATTC CCL64) cell cultures as previously described (18). At confluence, the monolayers were examined for infection by a focal immunofluorescence assay (56). The assay was used in conjunction with MAb 516 (8) or Hy 7 (10) to detect polytropic MuLVs corresponding to the two major antigenic subclasses of polytropic isolates (39) and MAb 538 (37) to detect M-MuLV. Infected cells from individual foci developed with Hy 7 or MAb 516 were subcultured and expanded to obtain virus stocks. Care was taken to exclude uninfected cells from the subcultures, to minimize virus spread during expansion. The viruses were biologically cloned by endpoint dilution to obtain cultures infected with a single polytropic MuLV and lacking ecotropic M-MuLV. In some instances, genomic DNA lambda libraries were constructed from infected M. dunni cells after cocultivation with tissues of M-MuLV-infected mice, and the libraries were screened for polytropic proviruses as previously described (22). M. dunni cells lack endogenous polytropic proviruses (31, 36); thus, positive clones obtained from the libraries correspond to newly integrated polytropic proviruses.
PCR and reverse transcription-PCR (RT-PCR) analyses and DNA sequencing. DNA from lambda clones containing integrated proviruses from infected M. dunni genomic libraries was prepared as previously reported (22) and used as the template for PCRs. An approximately 900-bp product that encompassed sequences encoding the receptor-binding region of the surface glycoprotein (SU) protein was generated with a forward primer termed FORPOLY (GCAGTACAACGAGAGGTCTGG) and a reverse primer termed REVPOLY (GGGTCAAAGAGAACCGGGTCAC). In some instances where the ecotropic-polytropic recombination boundary was 3' to the polytropic forward primer, an ecotropic forward primer termed FORECO (CTGGTCCAGCACGAAGTCTGG) was employed. PCRs were conducted using Pfu Turbo DNA polymerase (catalog no. 600153; Stratagene), Platinum Pfx DNA polymerase (catalog no. 11708-021; Invitrogen), or the puReTaq Ready-To-Go PCR system (catalog no. 27-9558-01; Amersham) according to the manufacturer's instructions.
RNA from virions released from infected cultures was used for RT-PCR analysis. The tissue culture medium from infected cultures was collected at 24-h intervals and sedimented in an SW 28.1 Beckman ultracentrifuge rotor at 25,000 rpm for 90 min. The pelleted virions from 200 to 500 ml of medium were resuspended in 0.1 M NaCl, 0.01 M Tris HCl, 0.001 M EDTA, and the RNA purified with the QIAamp Viral RNA Mini Kit (catalog no. 52906; QIAGEN) with the following modifications of the manufacturer's instructions. Lysis of the sample with buffer AVL was done in the absence of carrier RNA, and elution of the RNA from the spin column was accomplished with two successive elutions, each with 60 μl of distilled water DNase and RNase free (catalog no. 10977-015; Invitrogen). The second elution typically increased the total yield of RNA from 50 to 80%. Synthesis of cDNA was performed using the SuperScript First-Stand Synthesis system for RT-PCR (catalog no. 11904-018; Invitrogen) according to the manufacturer's instructions with the primer REVPOLY at 70 pmol and 1 μg of virion RNA per reaction mixture. PCR was performed on the cDNAs with the primers FORPOLY and REVPOLY with Platinum Pfx DNA polymerase to generate the 900-bp product. Primers used for sequencing the 900-bp PCR and RT-PCR products included FORPOLY, REVPOLY, MIDFORPOLY (AGCTAATGCTACCTCCCTCCTG), MIDREVPOLY (ACATTGAAGACCTGATGAGGG), and 329MX33RC (CAGTCTCATCCCAGTCGTCC). Primers used to for sequence analysis of additional env gene regions included 510MX33 (CCCTTAAGCGAGGAAACACCCCTCGG), 897MX33RC (GCTTGGTAGGCTCCATCTACCAGGT), 820MX33 (CATGCTCCCCAGGCCTCCTC), 1203MX33RC (AGCCCGGTGTTGCAAGCCC), and 1645MX33RC (CTACGCCAGTGTGGTCCGCG). Sequence analyses were performed using an ABI model 377 DNA sequencer by the DNA Sequencing and Synthesis Facility at Iowa State University or by an ABI model 3700 DNA sequencer at Rocky Mountain Laboratories.
Alignments, homology, and free-energy plots. Sequence alignments were performed with the CLUSTAL_X sequence alignment program (63). Homology at each nucleotide was calculated as the percent homology of a 31-base sequence alignment that included 15 bases to the 3' and 5' sides of the nucleotide. Free-energy computations of predicted RNA conformations of the overlapping 251-bp env gene sequences were done with the Mfold program (version 3.1) (71). The free energy of conformation at a particular position was plotted as the free energy of the RNA sequence, which included 125 bases to the 3' and 5' sides of the nucleotide.
Nucleotide sequence accession numbers. The sequences in the report have been deposited in GenBank under accession no. AY714500 through AY714525.
RESULTS
Minimizing sequence heterogeneity of polytropic MuLV nucleotide sequences due to random mutation. We recently characterized the endogenous polytropic proviruses in the NFS/N mouse genome (22). It was found that nearly all of the proviruses were distinguishable by base differences within an 893-base region, encompassing sequences encoding the receptor-binding domain that is acquired during recombination by all polytropic MuLV isolates. An average divergence of 2.2% was observed among the sequences in this region. In the present study, we obtained numerous polytropic MuLVs from M-MuLV-infected NFS/N mice, utilizing isolation procedures designed to limit the number of virus replication cycles and thereby minimize random mutations. Comparisons of the nucleotide sequence of the 893-base regions of the polytropic virus isolates to those of the endogenous proviruses identified the endogenous parent of each polytropic MuLV (Table 1). This included polytropic MuLVs of the Hy 7-reactive and MAb 516-reactive antigenic subclasses, both of which are generated as major populations in M-MuLV-infected mice (37, 38). In most cases (14 of 24 isolates), the polytropic sequence determined in the recombinant viruses exactly matched the sequence of their respective endogenous parents. Several other recombinant viruses differed from their endogenous parent by a single base change within the 893-base sequence, and three polytropic MuLVs differed from their endogenous parents by two base changes in this region. In each instance, the base differences were unique in that none of the specific changes were observed in any other recombinant virus or endogenous polytropic proviral sequence and were almost certainly the result of random mutation during active replication of the recombinant viruses. The rate of mutation of an MuLV has been determined to be approximately 1.4 x 10–5 (42); thus, a sequence of 900 bases would, on the average, acquire a point mutation every 70 or 80 replication cycles. The frequency of mutation we observed (11 in 21,000 bases examined) agreed closely to what would be expected from the mutation rate of MuLVs and the estimated number of replication cycles (30 to 50) undergone by the viruses before sequence analyses.
Two additional polytropic MuLVs arose from multiple recombination events and therefore exhibited multiple base differences compared to any single endogenous provirus (M-RV 7B and M-RV 15B). These viruses are considered in a separate section of the results.
Three endogenous polytropic proviruses give rise to nearly all recombinant polytropic MuLVs in M-MuLV-infected NFS/N mice. In contrast to other ecotropic MuLVs such as Akv and F-MuLV, which generate predominantly Hy 7-reactive polytropic recombinant viruses, M-MuLV generates both Hy 7-reactive and MAb 516-reactive polytropic MuLVs in approximately equal amounts (17, 23, 37, 38). Three distinct structural subgroups of endogenous polytropic proviruses have been described previously (22, 58) and have been termed the polytropic (PT), modified polytropic (mPT), and intermediate polytropic (iPT) subgroups. Hy 7-reactive polytropic MuLVs correspond to recombinants with the PT subgroup of endogenous polytropic proviruses, while MAb 516-reactive polytropic MuLVs correspond to recombinants with either the mPT or iPT subgroups (22). Surprisingly, we found that nearly all of the polytropic MuLVs generated by M-MuLV were derived by recombination with only three endogenous polytropic proviruses. Nine of the 11 Hy 7-reactive polytropic MuLVs were derived from one of two PT subgroup proviruses, NC3 (4 MuLVs) or NP1 (5 MuLVs). Even more striking was the observation that 12 of the 13 MAb 516-reactive polytropic MuLVs were derived from a single iPT subgroup provirus, NB1. Thus, the prominent population of MAb 516-reactive recombinant viruses characteristic of M-MuLV infection can be ascribed to the recombination of M-MuLV with this particular provirus. It is noteworthy that NB1, while it possesses an intact receptor-binding region, does not encode a functional SU or transmembrane protein as a result of a frameshift mutation in the 3' half of the SU-encoding sequences (22).
Other proviruses also participate in recombination with M-MuLV to yield polytropic MuLVs; however, they were identified at a much lower frequency. These include two PT subgroup proviruses, NI1 and NQ1, and one mPT subgroup provirus, NA1. It seems likely that additional proviruses may also contribute sequences to the polytropic MuLVs at a low frequency.
M-MuLV and endogenous polytropic proviruses recombine in a short, highly homologous region of the env gene. The sequence analyses above were extended to include the sites of recombination between the proviruses and M-MuLV within the env gene of the recombinant viruses. Surprisingly, we found that almost all of the junctions between ecotropic and polytropic sequences in the env genes of the recombinants were confined to a short region in the SU-encoding sequences. The region lies immediately to the 3' side of the sequences encoding the proline-rich hinge region of the SU protein and corresponds to the region of highest homology between the ecotropic and polytropic env gene sequences and very near a region of high predicted secondary structural stability in the M-MuLV env gene (Fig. 1). Within this region, the recombination sites of several different viruses were indistinguishable from one another. This was particularly striking for recombinants with the NB1 provirus (Fig. 1A). As noted earlier, NB1, the most frequent endogenous participant in recombination, contains a frameshift mutation disrupting the translation of the carboxyl end of the Env polyprotein. This suggests that recombination of NB1 with M-MuLV must occur on the 5' side of the mutation to yield a replication-competent MuLV. We found this to be true in all recombinants of M-MuLV with this provirus (Fig. 1A).
A polytropic MuLV can be derived by recombination with more than one endogenous polytropic-like provirus. Two polytropic MuLV isolates (M-RV 7B and M-RV 15B) exhibited multiple base differences when compared to any of the endogenous proviruses (Table 1). Comparison to the closest related provirus revealed that the differences were not randomly distributed but rather were localized to particular regions of the 893-base sequence. Alignments of subsections of the viral sequences with those of the proviruses revealed that each of the viruses was derived by recombination of M-MuLV with two different endogenous proviruses. M-RV 7B was found to be a recombinant with a provirus of the PT subgroup and a provirus of the iPT subgroup (Fig. 2). The iPT subgroup parent (NB1) of M-RV 7B could be identified unambiguously; however, the region contributed by the PT subgroup parent was indistinguishable from three distinct PT proviruses (NC1, NC2, and NC3). A second polytropic virus isolate, M-RV 15B, exhibited polytropic sequences identical to the PT provirus, NQ1, except for 2 bases flanking a short 14-base sequence near the middle of the polytropic env sequences. It is conceivable that these base differences were due to point mutations. However, in contrast to the point mutations we have identified in other MuLVs, which are not found in any other virus isolates or proviruses, the bases in M-RV 15B differing from NQ1 are common to numerous endogenous proviruses. This strongly suggests that they were the result of a second recombination event. Alignments with the remaining polytropic proviruses identified 14 different proviruses, any one of which could have recombined to yield M-RV 15B.
Our analyses of polytropic virus isolates suggest that approximately 10% of the viruses examined were derived by multiple recombination events with more than one endogenous provirus. It is quite possible that this is an underestimation of the actual frequency of such events, considering the close homology of the endogenous proviruses, particularly those of the PT and mPT subclasses. These subgroups exhibit long regions of sequence identity to one another, and recombination within an identical region would be undetectable.
Multiple polytropic MuLVs are generated in an individual mouse. It is possible that the double recombinants were formed by sequential recombination in which the polytropic MuLV generated after the initial recombination event subsequently underwent recombination with a second endogenous provirus or its transcript. Alternatively, two different polytropic viruses may have been generated and thereafter infected the same cell, where they recombined with one another to generate the double recombinant. The latter mechanism predicts the simultaneous replication of multiple recombinant viruses in a single animal. Our earlier studies suggested that at least two different polytropic recombinant viruses were detectable in nearly all M-MuLV-infected mice. This was based on the presence of both Hy 7- and MAb 516-reactive subclasses of viruses in individual animals (37, 38) and our subsequent finding that the antigenic subclasses correspond to different subgroups of endogenous polytropic-like proviruses (22). However, the extent of polytropic MuLV heterogeneity in a single animal was not known. To address this question, we isolated and characterized multiple polytropic MuLVs derived from a single mouse. Of the five polytropic MuLVs derived from this mouse, only two (M-RV 10D and M-RV 10E) were derived from the same endogenous provirus (NB1) (Fig. 3). These two viruses differed in their sites of recombination at both the 5' and 3' ecotropic and/or polytropic sequence boundaries. Thus, each of the five polytropic MuLVs derived from this animal were derived by different recombination events.
DISCUSSION
The unambiguous identification of endogenous polytropic-like proviruses gives rise to polytropic recombinant MuLVs. The identity of the endogenous proviruses participating in recombination to generate polytropic MuLVs has been considered in numerous studies (31, 32, 33, 58); however, the precise identification of those proviruses has been difficult. There are numerous very closely related proviruses in the genomes of inbred mice that are potential donors of the envelope sequences found in polytropic viruses. Proviral sequences diverge very slowly when replicating as components of the mouse genome; however, upon incorporation into a replicating retrovirus their rate of divergence increases by several orders of magnitude (26). In a recent study, we were able to distinguish nearly all of the endogenous polytropic proviruses of NFS/N mice by microheterogeneity in a 900-base sequence of their genome that was incorporated into the recombinant viruses (22). In the present study, utilizing procedures to minimize the number of replication cycles undergone by the recombinant viruses, we were able to unambiguously identify the proviral parents for all of the polytropic MuLV isolates examined.
Selectivity of endogenous proviruses contributing to polytropic MuLVs. Our analyses indicate that the overwhelming majority of polytropic MuLVs detected after infection with M-MuLV were derived by recombination with only three endogenous proviruses. At present, it is unclear what factors influence this selectivity. Among the possible factors is replicative selection of the recombinant viruses. Replicative selection could be the result of inefficient spread of certain recombinant viruses or less subtle effects, such as the generation of recombinant viruses that are replication defective. Although replicative selection likely influences the types of polytropic MuLVs that are detected in infected mice, its effect may be dampened by the extensive pseudotyping of the polytropic MuLV genomes within ecotropic virions (37).
A second possible factor that influences the selectivity of the proviruses participating in recombination is the availability of the proviral transcripts for recombination with the ecotropic MuLV. The finding that only three proviruses were major contributors to the polytropic MuLVs found in M-MuLV-infected mice could reflect the relative abundance of their RNA transcripts. Endogenous proviruses appear to be expressed in a tightly controlled manner in mice, as well as in humans (3, 13, 46, 51, 57, 69), but the identity of the transcripts and the mechanism of control are poorly understood. It is possible that high levels of transcription of the endogenous polytropic proviruses are limited to the three proviruses identified as major participants in recombination. In contrast to other ecotropic MuLVs, infection by M-MuLV results in a major population of polytropic MuLVs derived from a single provirus reactive with MAb 516. Ecotropic MuLVs exhibit distinct in vivo tropisms for replication in different tissues (14, 24, 53); thus, the different types of recombinants observed after infection with different ecotropic MuLVs could simply reflect differences in the endogenous proviruses expressed in those tissues in which the ecotropic MuLV is propagating. In this regard, Akv, the ecotropic MuLV of AKR/J mice, induces thymic lymphomas similar to M-MuLV and exhibits a similar temporal pattern of replication in the spleen and thymus (17, 24). With both M-MuLV-infected mice and AKR/J mice, polytropic recombinant viruses are initially detected in the spleen. However, in contrast to mice infected with M-MuLV, predominantly Hy 7-reactive recombinants are generated in AKR/J mice (17, 23). Although it cannot be excluded that the two viruses exhibit subtle differences in the cellular subpopulations in which they are propagating, the observations suggest that the different types of recombinants generated by M-MuLV and Akv are not the result of tissue tropism differences. Furthermore, studies with chimeric viruses between M-MuLV and F-MuLV indicate that sequences encoding the nucleocapsid protein (NC) and a portion of the protease (NCPR sequence) strongly influence recombination of the MuLVs to generate MAb 516-reactive polytropic MuLVs (38). The NCPR sequences, however, have no obvious effect on tissue tropism or the pathology induced by the viruses. Thus, it seems unlikely that the different polytropic MuLVs generated by M-MuLV compared to other MuLVs are entirely the result of replication in different tissues.
It is possible that the selectivity of proviruses involved in recombination is the result of events that influence the recombination process between the ecotropic MuLV and polytropic proviruses. The prevailing mechanism proposed for retroviral recombination involves the formation of a virion RNA heterodimer consisting of one transcript of each parental genome (29, 60). Recombination is thought to occur by a copy choice mechanism, involving template switching during RT of the heterodimer (11). The selectivity of proviruses involved in recombination could be manifested at various processes occurring prior to or during recombination. These include possible changes in transcription levels of the proviruses, the ability to form or package heterodimeric RNAs, or the ability to efficiently transcribe the heterodimers upon infection.
A "hot spot" for recombination between M-MuLV and endogenous polytropic proviruses. We found that the points of recombination in the env genes of M-MuLV and the polytropic proviruses are highly constrained to a small region of the SU-encoding sequences, coinciding with the highest degree of homology between the two genes, as well as the highest predicted secondary structural stability in the M-MuLV env gene (Fig. 1). This finding may have important implications with regard to the mechanism of recombination and possibly to the specificity of recombination of the endogenous proviruses with different ecotropic MuLVs. Accurate template switching during M-MuLV RT has been reported to be severely compromised under conditions of limited homology (47). Thus, recombination at the peak of highest homology may reflect the requirement of the newly synthesized negative-strand DNA to correctly anneal to the second RNA strand. It is unlikely that strand switching during RT of a heterodimer could proceed accurately in a region of low homology. In this regard, no recombination points were observed within the receptor-binding region of the SU-encoding sequences which exhibited low homology (Fig. 1A). It is also possible that the hot spot for recombination reflects, to some extent, the replicative selection of viruses. The recombinants encode chimeric SU proteins, some of which may function better than others. In this regard, any recombination that did occur in the nonhomologous receptor-binding region of the SU protein may result in the disruption of receptor-binding activity and the elimination of such recombinants.
Template switching during recombination is frequently envisioned as a forced copy choice mechanism involving a break in the RNA encountered during RT, requiring switching of templates to continue synthesis (11). However, breaks would be expected to occur randomly along the RNA, and it is not obvious that compensation by template switching would be restricted to regions of near-sequence identity. Moreover, comparisons of the sequences of M-MuLV and the endogenous polytropic-like proviruses reveal other regions of their SU genes that are highly homologous (Fig. 1A), including substantial regions (ca. >20 bases) that are identical in sequence.
Template switching by the polymerase can occur not only when random breakage is encountered, but also at defined points along the genome that exhibit complex secondary and possibly tertiary structures (15, 68, 70). The observation that recombination between M-MuLV and the polytropic proviruses occurs most frequently near the region of maximal homology and secondary structural stability may reflect a region of complex structure, resulting in a higher occurrence of template switching. The NC protein facilitates transcription through regions of complex RNA structures during RT, thereby decreasing the frequency of strand switching (and recombination) in these regions (67, 70). Subtle differences in the NC protein markedly alter the interactions of the protein with nucleic acids as well as the frequency of strand switching (27, 67, 70). Thus, different NC proteins (e.g., M-MuLV NC and F-MuLV NC proteins) may differ in their ability to facilitate transcription through complex RNA structures. As noted earlier, the provirus NB1, which gives rise to the major MAb 516-reactive polytropic MuLV recombinants with M-MuLV, contains a frameshift mutation in the 3' region of the SU protein-encoding sequences. Thus, if the NC protein of F-MuLV was unable to promote transcription through a troublesome region on the 3' side of the mutation and required template switching to proceed, the resulting recombinants would be replication defective. Conversely, if the M-MuLV NC protein could facilitate the successful negotiation of this region, transcription would proceed past the mutation until another region of difficulty was reached. Recombination in that region would yield a replication-competent MuLV. This hypothesis is consistent with the different types of polytropic MuLVs observed after infection with M-MuLV and F-MuLV and the finding that this difference is strongly influenced by the NCPR sequence.
Potential for in vivo evolution of polytropic MuLVs. Several examples exist of highly virulent viruses that are the result of multiple recombination and/or deletion events. The highly oncogenic mink cell focus-forming viruses of AKR/J and HRS/J mice are polytropic MuLVs that have undergone additional recombination to acquire a xenotropic MuLV long terminal repeat (49, 59) and are proximal agents in the transcriptional activation of proto-oncogenes (45, 50). The spleen focus-forming viruses which cause an acute erytholeukemia in adult mice appear to be the result of several deletion and/or recombination events involving the deletion of a polytropic MuLV and the acquisition of a short xenotropic env sequence (2, 21, 66). Similarly, FMCF98, a neurovirulent polytropic MuLV (48) (GenBank accession number AF133256) appears to have sequences derived from a polytropic provirus and a xenotropic MuLV. Recombination between viruses in an infected individual giving rise to a more virulent form appears to have been recently documented in a human HIV infection in which the emergence of the recombinant was associated with a relapse in the disease (25).
The finding that some of the polytropic viruses are recombinants with more than one endogenous sequence indicates that the viruses continue to undergo recombination subsequent to their generation. Furthermore, there is a heterogeneous array of recombinant viruses generated in an individual animal that could recombine with one another to yield new variants. Pseudotyping within ecotropic virions circumvents polytropic viral interference and increases the likelihood of coinfection of the same cell, a prerequisite for recombination. Pseudotyping may also serve to maintain defective viruses in the population, possibly including viruses that carry defects inherent to the endogenous polytropic provirus parent, thereby increasing the available pool of sequences that might contribute to new variants. These observations suggest a high potential for the rapid evolution of polytropic MuLVs in M-MuLV-infected mice. Evolution of polytropic MuLVs might be reflected in polytropic MuLVs that are the proximal agents of transformation in M-MuLV-induced tumors. In light of these considerations, a more detailed structural analysis of polytropic MuLVs integrated near proto-oncogenes in M-MuLV-induced leukemia is warranted.
Present address: Bio-Inova Life Sciences International, 78372 Plaisir Cedex, France.
REFERENCES
Adachi, A., K. Sakai, N. Kitamura, S. Nakanishi, O. Niwa, M. Matsuyama, and A. Ishimoto. 1984. Characterization of the env gene and long terminal repeat of molecularly cloned Friend mink cell focus-inducing virus DNA. J. Virol. 50:813-821.
Amanuma, H., A. Katori, M. Obata, N. Sagata, and Y. Ikawa. 1983. Complete nucleotide sequence of the gene for the specific glycoprotein (gp55) of Friend spleen focus-forming virus. Proc. Natl. Acad. Sci. USA 80:3913-3917.
Blaise, S., N. de Parseval, L. Benit, and T. Heidmann. 2003. Genomewide screening for fusogenic human endogenous retrovirus envelopes identifies syncytin 2, a gene conserved on primate evolution. Proc. Natl. Acad. Sci. USA 100:13013-13018.
Bosselman, R. A., F. van Straaten, C. Van Beveren, I. M. Verma, and M. Vogt. 1982. Analysis of the env gene of a molecularly cloned and biologically active Moloney mink cell focus-forming proviral DNA. J. Virol. 44:19-31.
Brightman, B. K., A. Rein, D. J. Trepp, and H. Fan. 1991. An enhancer variant of Moloney murine leukemia virus defective in leukemogenesis does not generate detectable mink cell focus-inducing virus in vivo. Proc. Natl. Acad. Sci. USA 88:2264-2268.
Chattopadhyay, S. K., M. W. Cloyd, D. L. Linemeyer, M. R. Lander, E. Rands, and D. R. Lowy. 1982. Cellular origin and role of mink cell focus-forming viruses in murine thymic lymphomas. Nature 295:25-31.
Chattopadhyay, S. K., M. R. Lander, S. Gupta, E. Rands, and D. R. Lowy. 1981. Origin of mink cytopathic focus-forming (MCF) viruses: comparison with ecotropic and xenotropic murine leukemia virus genomes. Virology 113:465-483.
Chesebro, B., W. Britt, L. Evans, K. Wehrly, J. Nishio, and M. Cloyd. 1983. Characterization of monoclonal antibodies reactive with murine leukemia viruses: use in analysis of strains of friend MCF and Friend ecotropic murine leukemia virus. Virology 127:134-148.
Chien, Y. H., I. M. Verma, T. Y. Shih, E. M. Scolnick, and N. Davidson. 1978. Heteroduplex analysis of the sequence relations between the RNAs of mink cell focus-inducing and murine leukemia viruses. J. Virol. 28:352-360.
Cloyd, M. W., B. Chesebro, J. L. Portis, and M. Weir. 1982. MCF-specific murine monoclonal antibodies made against AKR-247 MCF virus recognize a unique determinant associated with the gp70-p15(E) complex. J. Virol. 41:1112-1117.
Coffin, J. M. 1979. Structure, replication, and recombination of retrovirus genomes: some unifying hypotheses. J. Gen. Virol. 42:1-26.
Cuypers, H. T., G. Selten, W. Quint, M. Zijlstra, E. R. Maandag, W. Boelens, P. van Wezenbeek, C. Melief, and A. Berns. 1984. Murine leukemia virus-induced T-cell lymphomagenesis: integration of proviruses in a distinct chromosomal region. Cell 37:141-150.
de Parseval, N., V. Lazar, J. F. Casella, L. Benit, and T. Heidmann. 2003. Survey of human genes of retroviral origin: identification and transcriptome of the genes with coding capacity for complete envelope proteins. J. Virol. 77:10414-10422.
DesGroseillers, L., E. Rassart, and P. Jolicoeur. 1983. Thymotropism of murine leukemia virus is conferred by its long terminal repeat. Proc. Natl. Acad. Sci. USA 80:4203-4207.
DeStefano, J. J., R. A. Bambara, and P. J. Fay. 1994. The mechanism of human immunodeficiency virus reverse transcriptase-catalyzed strand transfer from internal regions of heteropolymeric RNA templates. J. Biol. Chem. 269:161-168.
Donoghue, D. J., E. Rothenberg, N. Hopkins, D. Baltimore, and P. A. Sharp. 1978. Heteroduplex analysis of the nonhomology region between Moloney MuLV and the dual host range derivative HIX virus. Cell 14:959-970.
Evans, L. H. 1986. Characterization of polytropic MuLVs from three-week-old AKR/J mice. Virology 153:122-136.
Evans, L. H., and M. W. Cloyd. 1984. Generation of mink cell focus-forming viruses by Friend murine leukemia virus: recombination with specific endogenous proviral sequences. J. Virol. 49:772-781.
Evans, L. H., and M. W. Cloyd. 1985. Friend and Moloney murine leukemia viruses specifically recombine with different endogenous retroviral sequences to generate mink cell focus-forming viruses. Proc. Natl. Acad. Sci. USA 82:459-463.
Evans, L. H., and P. H. Duesberg. 1982. Isolation of a transformation-defective deletion mutant of Moloney murine sarcoma virus. J. Virol. 41:735-743.
Evans, L. H., P. H. Duesberg, D. L. Linemeyer, S. K. Ruscetti, and E. M. Scolnick. 1981. Structural and functional studies of the Friend spleen focus-forming virus: structural relationship of SFFV to dualtropic viruses and molecular cloning of a biologically active subgenomic fragment of SFFV DNA. Hamatol. Bluttransfus. 26:472-478.
Evans, L. H., M. Lavignon, M. Taylor, and A. S. Alamgir. 2003. Antigenic subclasses of polytropic murine leukemia virus (MLV) isolates reflect three distinct groups of endogenous polytropic MLV-related sequences in NFS/N mice. J. Virol. 77:10327-10338.
Evans, L. H., and F. G. Malik. 1987. Class II polytropic murine leukemia viruses (MuLVs) of AKR/J. mice: possible role in the generation of class I oncogenic polytropic MuLVs. J. Virol. 61:1882-1892.
Evans, L. H., and J. D. Morrey. 1987. Tissue-specific replication of Friend and Moloney murine leukemia viruses in infected mice. J. Virol. 61:1350-1357.
Fang, G., B. Weiser, C. Kuiken, S. M. Philpott, S. Rowland-Jones, F. Plummer, J. Kimani, B. Shi, R. Kaul, J. Bwayo, O. Anzala, and H. Burger. 2004. Recombination following superinfection by HIV-1. AIDS 18:153-159.
Gojobori, T., and S. Yokoyama. 1985. Rates of evolution of the retroviral oncogene of Moloney murine sarcoma virus and of its cellular homologues. Proc. Natl. Acad. Sci. USA 82:4198-4201.
Guo, J., T. Wu, B. F. Kane, D. G. Johnson, L. E. Henderson, R. J. Gorelick, and J. G. Levin. 2002. Subtle alterations of the native zinc finger structures have dramatic effects on the nucleic acid chaperone activity of human immunodeficiency virus type 1 nucleocapsid protein. J. Virol. 76:4370-4378.
Hartley, J. W., N. K. Wolford, L. J. Old, and W. P. Rowe. 1977. A new class of murine leukemia virus associated with development of spontaneous lymphomas. Proc. Natl. Acad. Sci. USA 74:789-792.
Hu, W. S., and H. M. Temin. 1990. Genetic consequences of packaging two RNA genomes in one retroviral particle: pseudodiploidy and high rate of genetic recombination. Proc. Natl. Acad. Sci. USA 87:1556-1560.
Ishimoto, A., A. Adachi, K. Sakai, T. Yorifuji, and S. Tsuruta. 1981. Rapid emergence of mink cell focus-forming (MCF) virus in various mice infected with NB-tropic friend virus. Virology 113:644-655.
Jung, Y. T., T. Wu, and C. A. Kozak. 2003. Characterization of recombinant nonecotropic murine leukemia viruses from the wild mouse species Mus spretus. J. Virol. 77:12773-12781.
Khan, A. S. 1984. Nucleotide sequence analysis establishes the role of endogenous murine leukemia virus DNA segments in formation of recombinant mink cell focus-forming murine leukemia viruses. J. Virol. 50:864-871.
Khan, A. S., W. P. Rowe, and M. A. Martin. 1982. Cloning of endogenous murine leukemia virus-related sequences from chromosomal DNA of BALB/c and AKR/J. mice: identification of an env progenitor of AKR-247 mink cell focus-forming proviral DNA. J. Virol. 44:625-636.
Koch, W., W. Zimmermann, A. Oliff, and R. Friedrich. 1984. Molecular analysis of the envelope gene and long terminal repeat of Friend mink cell focus-inducing virus: implications for the functions of these sequences. J. Virol. 49:828-840.
Lander, E. S., et. al. 2001. Initial sequencing and analysis of the human genome. Nature 409:860-921.
Lander, M. R., and S. K. Chattopadhyay. 1984. A Mus dunni cell line that lacks sequences closely related to endogenous murine leukemia viruses and can be infected by ectropic, amphotropic, xenotropic, and mink cell focus-forming viruses. J. Virol. 52:695-698.
Lavignon, M., and L. Evans. 1996. A multistep process of leukemogenesis in Moloney murine leukemia virus-infected mice that is modulated by retroviral pseudotyping and interference. J. Virol. 70:3852-3862.
Lavignon, M., J. Richardson, and L. H. Evans. 1997. A small region of the ecotropic murine leukemia virus (MuLV) gag gene profoundly influences the types of polytropic MuLVs generated in mice. J. Virol. 71:8923-8927.
Lavignon, M., J. L. Walker, S. M. Perryman, F. G. Malik, A. S. Khan, T. S. Theodore, and L. H. Evans. 1994. Characterization of epitopes defining two major subclasses of polytropic murine leukemia viruses (MuLVs) which are differentially expressed in mice infected with different ecotropic MuLVs. J. Virol. 68:5194-5203.
Lower, R. 1999. The pathogenic potential of endogenous retroviruses: facts and fantasies. Trends Microbiol. 7:350-356.
Lower, R., J. Lower, and R. Kurth. 1996. The viruses in all of us: characteristics and biological significance of human endogenous retrovirus sequences. Proc. Natl. Acad. Sci. USA 93:5177-5184.
Monk, R. J., F. G. Malik, D. Stokesberry, and L. H. Evans. 1992. Direct determination of the point mutation rate of a murine retrovirus. J. Virol. 66:3683-3689.
Muir, A., A. Lever, and A. Moffett. 2004. Expression and functions of human endogenous retroviruses in the placenta: an update. Placenta 25(Suppl. A):S16-S25.
Nakagawa, K., and L. C. Harrison. 1996. The potential roles of endogenous retroviruses in autoimmunity. Immunol. Rev. 152:193-236.
O'Donnell, P. V., E. Fleissner, H. Lonial, C. F. Koehne, and A. Reicin. 1985. Early clonality and high-frequency proviral integration into the c-myc locus in AKR leukemias. J. Virol. 55:500-503.
Oliver, P. L., and J. P. Stoye. 1999. Genetic analysis of Gv1, a gene controlling transcription of endogenous murine polytropic proviruses. J. Virol. 73:8227-8234.
Pfeiffer, J. K., and A. Telesnitsky. 2001. Effects of limiting homology at the site of intermolecular recombinogenic template switching during Moloney murine leukemia virus replication. J. Virol. 75:11263-11274.
Portis, J. L., S. Czub, S. Robertson, F. McAtee, and B. Chesebro. 1995. Characterization of a neurologic disease induced by a polytropic murine retrovirus: evidence for differential targeting of ecotropic and polytropic viruses in the brain. J. Virol. 69:8070-8075.
Quint, W., W. Boelens, P. van Wezenbeek, T. Cuypers, E. R. Maandag, G. Selten, and A. Berns. 1984. Generation of AKR mink cell focus-forming viruses: a conserved single-copy xenotrope-like provirus provides recombinant long terminal repeat sequences. J. Virol. 50:432-438.
Reicin, A., J.-Q. Yang, K. B. Marcu, E. Fleissner, C. F. Koehne, and P. V. O'Donnell. 1986. Deregulation of the c-myc oncogene in virus-induced thymic lymphomas of AKR/J mice. Mol. Cell. Biol. 6:4088-4092.
Rigby, R. J., S. J. Rozzo, H. Gill, T. Fernandez-Hart, B. J. Morley, S. Izui, B. L. Kotzin, and T. J. Vyse. 2004. A novel locus regulates both retroviral glycoprotein 70 and anti-glycoprotein 70 antibody production in New Zealand mice when crossed with BALB/c. J. Immunol. 172:5078-5085.
Rommelaere, J., D. V. Faller, and N. Hopkins. 1978. Characterization and mapping of RNase T1-resistant oligonucleotides derived from the genomes of Akv and MCF murine leukemia viruses. Proc. Natl. Acad. Sci. USA 75:495-499.
Rosen, C. A., W. A. Haseltine, J. Lenz, R. Ruprecht, and M. W. Cloyd. 1985. Tissue selectivity of murine leukemia virus infection is determined by long terminal repeat sequences. J. Virol. 55:862-866.
Rulli, K., P. A. Lobelle-Rich, A. Trubetskoy, J. Lenz, and L. S. Levy. 2001. Tissue distribution and timing of appearance of polytropic envelope recombinants during infection with SL3-3 murine leukemia virus or its weakly pathogenic SL3Myb5 mutant. J. Virol. 75:522-526.
Ruscetti, S., L. Davis, J. Feild, and A. Oliff. 1981. Friend murine leukemia virus-induced leukemia is associated with the formation of mink cell focus-inducing viruses and is blocked in mice expressing endogenous mink cell focus-inducing xenotropic viral envelope genes. J. Exp. Med. 154:907-920.
Sitbon, M., J. Nishio, K. Wehrly, D. Lodmell, and B. Chesebro. 1985. Use of a focal immunofluorescence assay on live cells for quantitation of retroviruses: distinction of host range classes in virus mixtures and biological cloning of dual-tropic murine leukemia viruses. Virology 141:110-118.
Stauffer, Y., G. Theiler, P. Sperisen, Y. Lebedev, and C. V. Jongeneel. 2004. Digital expression profiles of human endogenous retroviral families in normal and cancerous tissues. Cancer Immun. 4:2.
Stoye, J. P., and J. M. Coffin. 1987. The four classes of endogenous murine leukemia virus: structural relationships and potential for recombination. J. Virol. 61:2659-2669.
Stoye, J. P., C. Moroni, and J. M. Coffin. 1991. Virological events leading to spontaneous AKR thymomas. J. Virol. 65:1273-1285.
Stuhlmann, H., and P. Berg. 1992. Homologous recombination of copackaged retrovirus RNAs during reverse transcription. J. Virol. 66:2378-2388.
Thomas, C. Y., and J. M. Coffin. 1982. Genetic alterations of RNA leukemia viruses associated with the development of spontaneous thymic leukemia in AKR/J mice. J. Virol. 43:416-426.
Thomas, C. Y., R. Khiroya, R. S. Schwartz, and J. M. Coffin. 1984. Role of recombinant ecotropic and polytropic viruses in the development of spontaneous thymic lymphomas in HRS/J. mice. J. Virol. 50:397-407.
Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876-4882.
van der Putten, H., W. Quint, J. van Raaij, E. R. Maandag, I. M. Verma, and A. Berns. 1981. M-MuLV-induced leukemogenesis: integration and structure of recombinant proviruses in tumors. Cell 24:729-739.
Wirschubsky, Z., P. Tsichlis, G. Klein, and J. Sumegi. 1986. Rearrangement of c-myc, pim-1 and Mlvi-1 and trisomy of chromosome 15 in MCF- and Moloney-MuLV-induced murine T-cell leukemias. Int. J. Cancer 38:739-745.
Wolff, L., E. Scolnick, and S. Ruscetti. 1983. Envelope gene of the Friend spleen focus-forming virus: deletion and insertions in 3' gp70/p15E-encoding region have resulted in unique features in the primary structure of its protein product. Proc. Natl. Acad. Sci. USA 80:4718-4722.
Wu, W., B. M. Blumberg, P. J. Fay, and R. A. Bambara. 1995. Strand transfer mediated by human immunodeficiency virus reverse transcriptase in vitro is promoted by pausing and results in misincorporation. J. Biol. Chem. 270:325-332.
Xu, H., and J. D. Boeke. 1987. High-frequency deletion between homologous sequences during retrotransposition of Ty elements in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 84:8553-8557.
Yi, J. M., H. M. Kim, and H. S. Kim. 2004. Expression of the human endogenous retrovirus HERV-W family in various human tissues and cancer cells. J. Gen. Virol. 85:1203-1210.
Zhang, W. H., C. K. Hwang, W. S. Hu, R. J. Gorelick, and V. K. Pathak. 2002. Zinc finger domain of murine leukemia virus nucleocapsid protein enhances the rate of viral DNA synthesis in vivo. J. Virol. 76:7473-7484.
Zuker, M. 2003. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res. 31:3406-3415.(A. S. M. Alamgir, Nick Ow)
ABSTRACT
Polytropic murine leukemia viruses (MuLVs) are generated by recombination of ecotropic MuLVs with env genes of a family of endogenous proviruses in mice, resulting in viruses with an expanded host range and greater virulence. Inbred mouse strains contain numerous endogenous proviruses that are potential donors of the env gene sequences of polytropic MuLVs; however, the precise identification of those proviruses that participate in recombination has been elusive. Three different structural groups of proviruses in NFS/N mice have been described and different ecotropic MuLVs preferentially recombine with different groups of proviruses. In contrast to other ecotropic MuLVs such as Friend MuLV or Akv that recombine predominantly with a single group of proviruses, Moloney MuLV (M-MuLV) recombines with at least two distinct groups. In this study, we determined that only three endogenous proviruses, two of one group and one of another group, are major participants in recombination with M-MuLV. Furthermore, the distinction between the polytropic MuLVs generated by M-MuLV and other ecotropic MuLVs is the result of recombination with a single endogenous provirus. This provirus exhibits a frameshift mutation in the 3' region of the surface glycoprotein-encoding sequences that is excluded in recombinants with M-MuLV. The sites of recombination between the env genes of M-MuLV and endogenous proviruses were confined to a short region exhibiting maximum homology between the ecotropic and polytropic env sequences and maximum stability of predicted RNA secondary structure. These observations suggest a possible mechanism for the specificity of recombination observed for different ecotropic MuLVs.
INTRODUCTION
Vertebrate species harbor a very large number of endogenous retroviral elements in their genomes that have originated from the infrequent insertion of exogenous viruses into the germ line. It is estimated that 8% of the human genome is composed of retrovirus sequences and that the number of distinct retroviral elements approaches the total number of structural genes identified in the genome (35). Endogenous retroviruses or their gene products are differentially expressed in tissues and in certain physiological or pathological conditions (3, 13, 46, 51, 57, 69); however, the control and consequences of such expression are poorly understood. It has been suggested that endogenous retrovirus expression may be involved in human pathogenesis, including neoplasms (40, 41) and autoimmune disorders (44), as well as normal physiological processes such as placental development (43).
One of the more extensively investigated groups of endogenous retroviruses are those participating in recombination with exogenous ecotropic retroviruses in mice giving rise to host range variants. Ecotropic murine leukemia viruses (MuLVs) are infectious for murine cells but not cells of other species. Upon infection of mice, these MuLVs frequently recombine with members of a group of endogenous proviruses to generate polytropic MuLVs (18, 19, 28, 30, 54, 55). Recombination involves substitution of the env gene sequences encoding the receptor-binding region of the ecotropic MuLV with analogous env gene sequences of the endogenous proviruses (1, 4, 7, 9, 16, 18, 19, 34, 52, 62). The recombinant MuLVs utilize a different cell-surface receptor for infectious entry and are capable of infecting murine cells as well as cells of other species, including humans.
The generation of polytropic viruses is intimately involved in a number of proliferative diseases in mice including lymphocytic leukemia induced by Moloney MuLV (M-MuLV) (5, 12, 64), erythroleukemia induced by the Friend MuLV (F-MuLV) (55), and spontaneous leukemia observed in highly leukemic mouse strains (6, 28, 59, 61, 62, 65). In addition, polytropic MuLVs may contribute to neurological disorders (48).
There are about 20 to 40 endogenous proviruses in different inbred strains of mice that bear very close homology to the sequences found in polytropic recombinant viruses; however, the participation of the proviruses in recombination with ecotropic viruses is not random. Polytropic MuLVs belong to one of two antigenic subclasses based on their reactivity to one of two monoclonal antibodies (MAbs), termed Hy 7 and MAb 516 (39). Reactivities to the antibodies are mutually exclusive and signal recombination with distinct structural groups of endogenous proviruses. The ecotropic viruses Akv and F-MuLV generate predominantly Hy 7-reactive polytropic MuLVs with few MAb 516-reactive viruses, whereas infection with M-MuLV results in approximately equal numbers of Hy 7- and MAb 516-reactive polytropic viruses (37, 38). Thus, M-MuLV recombines with a distinct structural group of endogenous proviruses reactive with MAb 516.
There are a number of possible mechanisms for the specificity of recombination with different ecotropic MuLVs, ranging from tissue-specific expression of the proviruses to specific interactions of the ecotropic MuLVs with the endogenous proviruses or their RNA transcripts. To gain an understanding of this specificity, it is first necessary to obtain a more complete description of the phenomenon, namely, the precise identification of those proviruses that actually participate in recombination. Efforts to precisely identify the endogenous origin of polytropic MuLVs have been hampered by the very close homology of the env gene sequences of the endogenous proviruses and the rate of divergence of these sequences by random mutation once they are incorporated into a replicating MuLV (31, 32, 33, 58). In our previous study, we found that nearly all of the endogenous proviruses of NFS/N mice were distinguishable by sequence microheterogeneity (22). Thus, if the divergence of polytropic MuLV sequences could be minimized, comparisons to env gene sequences in recombinant polytropic MuLVs should identify which of the proviruses participate in recombination. In the present study, we utilized procedures to minimize the number of replication cycles prior to sequence analyses of the recombinant viruses and were able to unambiguously identify the endogenous proviral origin of numerous polytropic MuLVs derived after inoculation of M-MuLV. We found that a surprisingly small number of endogenous proviruses are major participants in recombination with M-MuLV and that recombination with a single provirus accounts for the difference in the polytropic MuLVs generated after inoculation of M-MuLV compared to other ecotropic viruses. Furthermore, examination of the recombination patterns of the polytropic MuLVs suggested a mechanism that may account for this specificity. Additional findings suggest that the polytropic MuLVs, once generated, may continue to recombine with one another and evolve to more virulent forms in the host.
MATERIALS AND METHODS
Mice. NFS/N mice were maintained as an inbred strain at Rocky Mountain Laboratories, Veterinary Branch. All animals were treated in accordance with National Institutes of Health regulations and the guidelines of the Animal Care and Use Committee of Rocky Mountain Laboratories.
Isolation of recombinant polytropic MuLVs. Polytropic MuLVs were isolated from NFS/N mice that had been inoculated within 24 h of birth with the ecotropic virus M-MuLV1387 (20). The infected mice were sacrificed 3 to 5 weeks after inoculation, the spleens and thymuses were dissociated, and serial dilutions of cells from the tissues were distributed as infectious centers on NIH 3T3, Mus dunni (36), or mink (ATTC CCL64) cell cultures as previously described (18). At confluence, the monolayers were examined for infection by a focal immunofluorescence assay (56). The assay was used in conjunction with MAb 516 (8) or Hy 7 (10) to detect polytropic MuLVs corresponding to the two major antigenic subclasses of polytropic isolates (39) and MAb 538 (37) to detect M-MuLV. Infected cells from individual foci developed with Hy 7 or MAb 516 were subcultured and expanded to obtain virus stocks. Care was taken to exclude uninfected cells from the subcultures, to minimize virus spread during expansion. The viruses were biologically cloned by endpoint dilution to obtain cultures infected with a single polytropic MuLV and lacking ecotropic M-MuLV. In some instances, genomic DNA lambda libraries were constructed from infected M. dunni cells after cocultivation with tissues of M-MuLV-infected mice, and the libraries were screened for polytropic proviruses as previously described (22). M. dunni cells lack endogenous polytropic proviruses (31, 36); thus, positive clones obtained from the libraries correspond to newly integrated polytropic proviruses.
PCR and reverse transcription-PCR (RT-PCR) analyses and DNA sequencing. DNA from lambda clones containing integrated proviruses from infected M. dunni genomic libraries was prepared as previously reported (22) and used as the template for PCRs. An approximately 900-bp product that encompassed sequences encoding the receptor-binding region of the surface glycoprotein (SU) protein was generated with a forward primer termed FORPOLY (GCAGTACAACGAGAGGTCTGG) and a reverse primer termed REVPOLY (GGGTCAAAGAGAACCGGGTCAC). In some instances where the ecotropic-polytropic recombination boundary was 3' to the polytropic forward primer, an ecotropic forward primer termed FORECO (CTGGTCCAGCACGAAGTCTGG) was employed. PCRs were conducted using Pfu Turbo DNA polymerase (catalog no. 600153; Stratagene), Platinum Pfx DNA polymerase (catalog no. 11708-021; Invitrogen), or the puReTaq Ready-To-Go PCR system (catalog no. 27-9558-01; Amersham) according to the manufacturer's instructions.
RNA from virions released from infected cultures was used for RT-PCR analysis. The tissue culture medium from infected cultures was collected at 24-h intervals and sedimented in an SW 28.1 Beckman ultracentrifuge rotor at 25,000 rpm for 90 min. The pelleted virions from 200 to 500 ml of medium were resuspended in 0.1 M NaCl, 0.01 M Tris HCl, 0.001 M EDTA, and the RNA purified with the QIAamp Viral RNA Mini Kit (catalog no. 52906; QIAGEN) with the following modifications of the manufacturer's instructions. Lysis of the sample with buffer AVL was done in the absence of carrier RNA, and elution of the RNA from the spin column was accomplished with two successive elutions, each with 60 μl of distilled water DNase and RNase free (catalog no. 10977-015; Invitrogen). The second elution typically increased the total yield of RNA from 50 to 80%. Synthesis of cDNA was performed using the SuperScript First-Stand Synthesis system for RT-PCR (catalog no. 11904-018; Invitrogen) according to the manufacturer's instructions with the primer REVPOLY at 70 pmol and 1 μg of virion RNA per reaction mixture. PCR was performed on the cDNAs with the primers FORPOLY and REVPOLY with Platinum Pfx DNA polymerase to generate the 900-bp product. Primers used for sequencing the 900-bp PCR and RT-PCR products included FORPOLY, REVPOLY, MIDFORPOLY (AGCTAATGCTACCTCCCTCCTG), MIDREVPOLY (ACATTGAAGACCTGATGAGGG), and 329MX33RC (CAGTCTCATCCCAGTCGTCC). Primers used to for sequence analysis of additional env gene regions included 510MX33 (CCCTTAAGCGAGGAAACACCCCTCGG), 897MX33RC (GCTTGGTAGGCTCCATCTACCAGGT), 820MX33 (CATGCTCCCCAGGCCTCCTC), 1203MX33RC (AGCCCGGTGTTGCAAGCCC), and 1645MX33RC (CTACGCCAGTGTGGTCCGCG). Sequence analyses were performed using an ABI model 377 DNA sequencer by the DNA Sequencing and Synthesis Facility at Iowa State University or by an ABI model 3700 DNA sequencer at Rocky Mountain Laboratories.
Alignments, homology, and free-energy plots. Sequence alignments were performed with the CLUSTAL_X sequence alignment program (63). Homology at each nucleotide was calculated as the percent homology of a 31-base sequence alignment that included 15 bases to the 3' and 5' sides of the nucleotide. Free-energy computations of predicted RNA conformations of the overlapping 251-bp env gene sequences were done with the Mfold program (version 3.1) (71). The free energy of conformation at a particular position was plotted as the free energy of the RNA sequence, which included 125 bases to the 3' and 5' sides of the nucleotide.
Nucleotide sequence accession numbers. The sequences in the report have been deposited in GenBank under accession no. AY714500 through AY714525.
RESULTS
Minimizing sequence heterogeneity of polytropic MuLV nucleotide sequences due to random mutation. We recently characterized the endogenous polytropic proviruses in the NFS/N mouse genome (22). It was found that nearly all of the proviruses were distinguishable by base differences within an 893-base region, encompassing sequences encoding the receptor-binding domain that is acquired during recombination by all polytropic MuLV isolates. An average divergence of 2.2% was observed among the sequences in this region. In the present study, we obtained numerous polytropic MuLVs from M-MuLV-infected NFS/N mice, utilizing isolation procedures designed to limit the number of virus replication cycles and thereby minimize random mutations. Comparisons of the nucleotide sequence of the 893-base regions of the polytropic virus isolates to those of the endogenous proviruses identified the endogenous parent of each polytropic MuLV (Table 1). This included polytropic MuLVs of the Hy 7-reactive and MAb 516-reactive antigenic subclasses, both of which are generated as major populations in M-MuLV-infected mice (37, 38). In most cases (14 of 24 isolates), the polytropic sequence determined in the recombinant viruses exactly matched the sequence of their respective endogenous parents. Several other recombinant viruses differed from their endogenous parent by a single base change within the 893-base sequence, and three polytropic MuLVs differed from their endogenous parents by two base changes in this region. In each instance, the base differences were unique in that none of the specific changes were observed in any other recombinant virus or endogenous polytropic proviral sequence and were almost certainly the result of random mutation during active replication of the recombinant viruses. The rate of mutation of an MuLV has been determined to be approximately 1.4 x 10–5 (42); thus, a sequence of 900 bases would, on the average, acquire a point mutation every 70 or 80 replication cycles. The frequency of mutation we observed (11 in 21,000 bases examined) agreed closely to what would be expected from the mutation rate of MuLVs and the estimated number of replication cycles (30 to 50) undergone by the viruses before sequence analyses.
Two additional polytropic MuLVs arose from multiple recombination events and therefore exhibited multiple base differences compared to any single endogenous provirus (M-RV 7B and M-RV 15B). These viruses are considered in a separate section of the results.
Three endogenous polytropic proviruses give rise to nearly all recombinant polytropic MuLVs in M-MuLV-infected NFS/N mice. In contrast to other ecotropic MuLVs such as Akv and F-MuLV, which generate predominantly Hy 7-reactive polytropic recombinant viruses, M-MuLV generates both Hy 7-reactive and MAb 516-reactive polytropic MuLVs in approximately equal amounts (17, 23, 37, 38). Three distinct structural subgroups of endogenous polytropic proviruses have been described previously (22, 58) and have been termed the polytropic (PT), modified polytropic (mPT), and intermediate polytropic (iPT) subgroups. Hy 7-reactive polytropic MuLVs correspond to recombinants with the PT subgroup of endogenous polytropic proviruses, while MAb 516-reactive polytropic MuLVs correspond to recombinants with either the mPT or iPT subgroups (22). Surprisingly, we found that nearly all of the polytropic MuLVs generated by M-MuLV were derived by recombination with only three endogenous polytropic proviruses. Nine of the 11 Hy 7-reactive polytropic MuLVs were derived from one of two PT subgroup proviruses, NC3 (4 MuLVs) or NP1 (5 MuLVs). Even more striking was the observation that 12 of the 13 MAb 516-reactive polytropic MuLVs were derived from a single iPT subgroup provirus, NB1. Thus, the prominent population of MAb 516-reactive recombinant viruses characteristic of M-MuLV infection can be ascribed to the recombination of M-MuLV with this particular provirus. It is noteworthy that NB1, while it possesses an intact receptor-binding region, does not encode a functional SU or transmembrane protein as a result of a frameshift mutation in the 3' half of the SU-encoding sequences (22).
Other proviruses also participate in recombination with M-MuLV to yield polytropic MuLVs; however, they were identified at a much lower frequency. These include two PT subgroup proviruses, NI1 and NQ1, and one mPT subgroup provirus, NA1. It seems likely that additional proviruses may also contribute sequences to the polytropic MuLVs at a low frequency.
M-MuLV and endogenous polytropic proviruses recombine in a short, highly homologous region of the env gene. The sequence analyses above were extended to include the sites of recombination between the proviruses and M-MuLV within the env gene of the recombinant viruses. Surprisingly, we found that almost all of the junctions between ecotropic and polytropic sequences in the env genes of the recombinants were confined to a short region in the SU-encoding sequences. The region lies immediately to the 3' side of the sequences encoding the proline-rich hinge region of the SU protein and corresponds to the region of highest homology between the ecotropic and polytropic env gene sequences and very near a region of high predicted secondary structural stability in the M-MuLV env gene (Fig. 1). Within this region, the recombination sites of several different viruses were indistinguishable from one another. This was particularly striking for recombinants with the NB1 provirus (Fig. 1A). As noted earlier, NB1, the most frequent endogenous participant in recombination, contains a frameshift mutation disrupting the translation of the carboxyl end of the Env polyprotein. This suggests that recombination of NB1 with M-MuLV must occur on the 5' side of the mutation to yield a replication-competent MuLV. We found this to be true in all recombinants of M-MuLV with this provirus (Fig. 1A).
A polytropic MuLV can be derived by recombination with more than one endogenous polytropic-like provirus. Two polytropic MuLV isolates (M-RV 7B and M-RV 15B) exhibited multiple base differences when compared to any of the endogenous proviruses (Table 1). Comparison to the closest related provirus revealed that the differences were not randomly distributed but rather were localized to particular regions of the 893-base sequence. Alignments of subsections of the viral sequences with those of the proviruses revealed that each of the viruses was derived by recombination of M-MuLV with two different endogenous proviruses. M-RV 7B was found to be a recombinant with a provirus of the PT subgroup and a provirus of the iPT subgroup (Fig. 2). The iPT subgroup parent (NB1) of M-RV 7B could be identified unambiguously; however, the region contributed by the PT subgroup parent was indistinguishable from three distinct PT proviruses (NC1, NC2, and NC3). A second polytropic virus isolate, M-RV 15B, exhibited polytropic sequences identical to the PT provirus, NQ1, except for 2 bases flanking a short 14-base sequence near the middle of the polytropic env sequences. It is conceivable that these base differences were due to point mutations. However, in contrast to the point mutations we have identified in other MuLVs, which are not found in any other virus isolates or proviruses, the bases in M-RV 15B differing from NQ1 are common to numerous endogenous proviruses. This strongly suggests that they were the result of a second recombination event. Alignments with the remaining polytropic proviruses identified 14 different proviruses, any one of which could have recombined to yield M-RV 15B.
Our analyses of polytropic virus isolates suggest that approximately 10% of the viruses examined were derived by multiple recombination events with more than one endogenous provirus. It is quite possible that this is an underestimation of the actual frequency of such events, considering the close homology of the endogenous proviruses, particularly those of the PT and mPT subclasses. These subgroups exhibit long regions of sequence identity to one another, and recombination within an identical region would be undetectable.
Multiple polytropic MuLVs are generated in an individual mouse. It is possible that the double recombinants were formed by sequential recombination in which the polytropic MuLV generated after the initial recombination event subsequently underwent recombination with a second endogenous provirus or its transcript. Alternatively, two different polytropic viruses may have been generated and thereafter infected the same cell, where they recombined with one another to generate the double recombinant. The latter mechanism predicts the simultaneous replication of multiple recombinant viruses in a single animal. Our earlier studies suggested that at least two different polytropic recombinant viruses were detectable in nearly all M-MuLV-infected mice. This was based on the presence of both Hy 7- and MAb 516-reactive subclasses of viruses in individual animals (37, 38) and our subsequent finding that the antigenic subclasses correspond to different subgroups of endogenous polytropic-like proviruses (22). However, the extent of polytropic MuLV heterogeneity in a single animal was not known. To address this question, we isolated and characterized multiple polytropic MuLVs derived from a single mouse. Of the five polytropic MuLVs derived from this mouse, only two (M-RV 10D and M-RV 10E) were derived from the same endogenous provirus (NB1) (Fig. 3). These two viruses differed in their sites of recombination at both the 5' and 3' ecotropic and/or polytropic sequence boundaries. Thus, each of the five polytropic MuLVs derived from this animal were derived by different recombination events.
DISCUSSION
The unambiguous identification of endogenous polytropic-like proviruses gives rise to polytropic recombinant MuLVs. The identity of the endogenous proviruses participating in recombination to generate polytropic MuLVs has been considered in numerous studies (31, 32, 33, 58); however, the precise identification of those proviruses has been difficult. There are numerous very closely related proviruses in the genomes of inbred mice that are potential donors of the envelope sequences found in polytropic viruses. Proviral sequences diverge very slowly when replicating as components of the mouse genome; however, upon incorporation into a replicating retrovirus their rate of divergence increases by several orders of magnitude (26). In a recent study, we were able to distinguish nearly all of the endogenous polytropic proviruses of NFS/N mice by microheterogeneity in a 900-base sequence of their genome that was incorporated into the recombinant viruses (22). In the present study, utilizing procedures to minimize the number of replication cycles undergone by the recombinant viruses, we were able to unambiguously identify the proviral parents for all of the polytropic MuLV isolates examined.
Selectivity of endogenous proviruses contributing to polytropic MuLVs. Our analyses indicate that the overwhelming majority of polytropic MuLVs detected after infection with M-MuLV were derived by recombination with only three endogenous proviruses. At present, it is unclear what factors influence this selectivity. Among the possible factors is replicative selection of the recombinant viruses. Replicative selection could be the result of inefficient spread of certain recombinant viruses or less subtle effects, such as the generation of recombinant viruses that are replication defective. Although replicative selection likely influences the types of polytropic MuLVs that are detected in infected mice, its effect may be dampened by the extensive pseudotyping of the polytropic MuLV genomes within ecotropic virions (37).
A second possible factor that influences the selectivity of the proviruses participating in recombination is the availability of the proviral transcripts for recombination with the ecotropic MuLV. The finding that only three proviruses were major contributors to the polytropic MuLVs found in M-MuLV-infected mice could reflect the relative abundance of their RNA transcripts. Endogenous proviruses appear to be expressed in a tightly controlled manner in mice, as well as in humans (3, 13, 46, 51, 57, 69), but the identity of the transcripts and the mechanism of control are poorly understood. It is possible that high levels of transcription of the endogenous polytropic proviruses are limited to the three proviruses identified as major participants in recombination. In contrast to other ecotropic MuLVs, infection by M-MuLV results in a major population of polytropic MuLVs derived from a single provirus reactive with MAb 516. Ecotropic MuLVs exhibit distinct in vivo tropisms for replication in different tissues (14, 24, 53); thus, the different types of recombinants observed after infection with different ecotropic MuLVs could simply reflect differences in the endogenous proviruses expressed in those tissues in which the ecotropic MuLV is propagating. In this regard, Akv, the ecotropic MuLV of AKR/J mice, induces thymic lymphomas similar to M-MuLV and exhibits a similar temporal pattern of replication in the spleen and thymus (17, 24). With both M-MuLV-infected mice and AKR/J mice, polytropic recombinant viruses are initially detected in the spleen. However, in contrast to mice infected with M-MuLV, predominantly Hy 7-reactive recombinants are generated in AKR/J mice (17, 23). Although it cannot be excluded that the two viruses exhibit subtle differences in the cellular subpopulations in which they are propagating, the observations suggest that the different types of recombinants generated by M-MuLV and Akv are not the result of tissue tropism differences. Furthermore, studies with chimeric viruses between M-MuLV and F-MuLV indicate that sequences encoding the nucleocapsid protein (NC) and a portion of the protease (NCPR sequence) strongly influence recombination of the MuLVs to generate MAb 516-reactive polytropic MuLVs (38). The NCPR sequences, however, have no obvious effect on tissue tropism or the pathology induced by the viruses. Thus, it seems unlikely that the different polytropic MuLVs generated by M-MuLV compared to other MuLVs are entirely the result of replication in different tissues.
It is possible that the selectivity of proviruses involved in recombination is the result of events that influence the recombination process between the ecotropic MuLV and polytropic proviruses. The prevailing mechanism proposed for retroviral recombination involves the formation of a virion RNA heterodimer consisting of one transcript of each parental genome (29, 60). Recombination is thought to occur by a copy choice mechanism, involving template switching during RT of the heterodimer (11). The selectivity of proviruses involved in recombination could be manifested at various processes occurring prior to or during recombination. These include possible changes in transcription levels of the proviruses, the ability to form or package heterodimeric RNAs, or the ability to efficiently transcribe the heterodimers upon infection.
A "hot spot" for recombination between M-MuLV and endogenous polytropic proviruses. We found that the points of recombination in the env genes of M-MuLV and the polytropic proviruses are highly constrained to a small region of the SU-encoding sequences, coinciding with the highest degree of homology between the two genes, as well as the highest predicted secondary structural stability in the M-MuLV env gene (Fig. 1). This finding may have important implications with regard to the mechanism of recombination and possibly to the specificity of recombination of the endogenous proviruses with different ecotropic MuLVs. Accurate template switching during M-MuLV RT has been reported to be severely compromised under conditions of limited homology (47). Thus, recombination at the peak of highest homology may reflect the requirement of the newly synthesized negative-strand DNA to correctly anneal to the second RNA strand. It is unlikely that strand switching during RT of a heterodimer could proceed accurately in a region of low homology. In this regard, no recombination points were observed within the receptor-binding region of the SU-encoding sequences which exhibited low homology (Fig. 1A). It is also possible that the hot spot for recombination reflects, to some extent, the replicative selection of viruses. The recombinants encode chimeric SU proteins, some of which may function better than others. In this regard, any recombination that did occur in the nonhomologous receptor-binding region of the SU protein may result in the disruption of receptor-binding activity and the elimination of such recombinants.
Template switching during recombination is frequently envisioned as a forced copy choice mechanism involving a break in the RNA encountered during RT, requiring switching of templates to continue synthesis (11). However, breaks would be expected to occur randomly along the RNA, and it is not obvious that compensation by template switching would be restricted to regions of near-sequence identity. Moreover, comparisons of the sequences of M-MuLV and the endogenous polytropic-like proviruses reveal other regions of their SU genes that are highly homologous (Fig. 1A), including substantial regions (ca. >20 bases) that are identical in sequence.
Template switching by the polymerase can occur not only when random breakage is encountered, but also at defined points along the genome that exhibit complex secondary and possibly tertiary structures (15, 68, 70). The observation that recombination between M-MuLV and the polytropic proviruses occurs most frequently near the region of maximal homology and secondary structural stability may reflect a region of complex structure, resulting in a higher occurrence of template switching. The NC protein facilitates transcription through regions of complex RNA structures during RT, thereby decreasing the frequency of strand switching (and recombination) in these regions (67, 70). Subtle differences in the NC protein markedly alter the interactions of the protein with nucleic acids as well as the frequency of strand switching (27, 67, 70). Thus, different NC proteins (e.g., M-MuLV NC and F-MuLV NC proteins) may differ in their ability to facilitate transcription through complex RNA structures. As noted earlier, the provirus NB1, which gives rise to the major MAb 516-reactive polytropic MuLV recombinants with M-MuLV, contains a frameshift mutation in the 3' region of the SU protein-encoding sequences. Thus, if the NC protein of F-MuLV was unable to promote transcription through a troublesome region on the 3' side of the mutation and required template switching to proceed, the resulting recombinants would be replication defective. Conversely, if the M-MuLV NC protein could facilitate the successful negotiation of this region, transcription would proceed past the mutation until another region of difficulty was reached. Recombination in that region would yield a replication-competent MuLV. This hypothesis is consistent with the different types of polytropic MuLVs observed after infection with M-MuLV and F-MuLV and the finding that this difference is strongly influenced by the NCPR sequence.
Potential for in vivo evolution of polytropic MuLVs. Several examples exist of highly virulent viruses that are the result of multiple recombination and/or deletion events. The highly oncogenic mink cell focus-forming viruses of AKR/J and HRS/J mice are polytropic MuLVs that have undergone additional recombination to acquire a xenotropic MuLV long terminal repeat (49, 59) and are proximal agents in the transcriptional activation of proto-oncogenes (45, 50). The spleen focus-forming viruses which cause an acute erytholeukemia in adult mice appear to be the result of several deletion and/or recombination events involving the deletion of a polytropic MuLV and the acquisition of a short xenotropic env sequence (2, 21, 66). Similarly, FMCF98, a neurovirulent polytropic MuLV (48) (GenBank accession number AF133256) appears to have sequences derived from a polytropic provirus and a xenotropic MuLV. Recombination between viruses in an infected individual giving rise to a more virulent form appears to have been recently documented in a human HIV infection in which the emergence of the recombinant was associated with a relapse in the disease (25).
The finding that some of the polytropic viruses are recombinants with more than one endogenous sequence indicates that the viruses continue to undergo recombination subsequent to their generation. Furthermore, there is a heterogeneous array of recombinant viruses generated in an individual animal that could recombine with one another to yield new variants. Pseudotyping within ecotropic virions circumvents polytropic viral interference and increases the likelihood of coinfection of the same cell, a prerequisite for recombination. Pseudotyping may also serve to maintain defective viruses in the population, possibly including viruses that carry defects inherent to the endogenous polytropic provirus parent, thereby increasing the available pool of sequences that might contribute to new variants. These observations suggest a high potential for the rapid evolution of polytropic MuLVs in M-MuLV-infected mice. Evolution of polytropic MuLVs might be reflected in polytropic MuLVs that are the proximal agents of transformation in M-MuLV-induced tumors. In light of these considerations, a more detailed structural analysis of polytropic MuLVs integrated near proto-oncogenes in M-MuLV-induced leukemia is warranted.
Present address: Bio-Inova Life Sciences International, 78372 Plaisir Cedex, France.
REFERENCES
Adachi, A., K. Sakai, N. Kitamura, S. Nakanishi, O. Niwa, M. Matsuyama, and A. Ishimoto. 1984. Characterization of the env gene and long terminal repeat of molecularly cloned Friend mink cell focus-inducing virus DNA. J. Virol. 50:813-821.
Amanuma, H., A. Katori, M. Obata, N. Sagata, and Y. Ikawa. 1983. Complete nucleotide sequence of the gene for the specific glycoprotein (gp55) of Friend spleen focus-forming virus. Proc. Natl. Acad. Sci. USA 80:3913-3917.
Blaise, S., N. de Parseval, L. Benit, and T. Heidmann. 2003. Genomewide screening for fusogenic human endogenous retrovirus envelopes identifies syncytin 2, a gene conserved on primate evolution. Proc. Natl. Acad. Sci. USA 100:13013-13018.
Bosselman, R. A., F. van Straaten, C. Van Beveren, I. M. Verma, and M. Vogt. 1982. Analysis of the env gene of a molecularly cloned and biologically active Moloney mink cell focus-forming proviral DNA. J. Virol. 44:19-31.
Brightman, B. K., A. Rein, D. J. Trepp, and H. Fan. 1991. An enhancer variant of Moloney murine leukemia virus defective in leukemogenesis does not generate detectable mink cell focus-inducing virus in vivo. Proc. Natl. Acad. Sci. USA 88:2264-2268.
Chattopadhyay, S. K., M. W. Cloyd, D. L. Linemeyer, M. R. Lander, E. Rands, and D. R. Lowy. 1982. Cellular origin and role of mink cell focus-forming viruses in murine thymic lymphomas. Nature 295:25-31.
Chattopadhyay, S. K., M. R. Lander, S. Gupta, E. Rands, and D. R. Lowy. 1981. Origin of mink cytopathic focus-forming (MCF) viruses: comparison with ecotropic and xenotropic murine leukemia virus genomes. Virology 113:465-483.
Chesebro, B., W. Britt, L. Evans, K. Wehrly, J. Nishio, and M. Cloyd. 1983. Characterization of monoclonal antibodies reactive with murine leukemia viruses: use in analysis of strains of friend MCF and Friend ecotropic murine leukemia virus. Virology 127:134-148.
Chien, Y. H., I. M. Verma, T. Y. Shih, E. M. Scolnick, and N. Davidson. 1978. Heteroduplex analysis of the sequence relations between the RNAs of mink cell focus-inducing and murine leukemia viruses. J. Virol. 28:352-360.
Cloyd, M. W., B. Chesebro, J. L. Portis, and M. Weir. 1982. MCF-specific murine monoclonal antibodies made against AKR-247 MCF virus recognize a unique determinant associated with the gp70-p15(E) complex. J. Virol. 41:1112-1117.
Coffin, J. M. 1979. Structure, replication, and recombination of retrovirus genomes: some unifying hypotheses. J. Gen. Virol. 42:1-26.
Cuypers, H. T., G. Selten, W. Quint, M. Zijlstra, E. R. Maandag, W. Boelens, P. van Wezenbeek, C. Melief, and A. Berns. 1984. Murine leukemia virus-induced T-cell lymphomagenesis: integration of proviruses in a distinct chromosomal region. Cell 37:141-150.
de Parseval, N., V. Lazar, J. F. Casella, L. Benit, and T. Heidmann. 2003. Survey of human genes of retroviral origin: identification and transcriptome of the genes with coding capacity for complete envelope proteins. J. Virol. 77:10414-10422.
DesGroseillers, L., E. Rassart, and P. Jolicoeur. 1983. Thymotropism of murine leukemia virus is conferred by its long terminal repeat. Proc. Natl. Acad. Sci. USA 80:4203-4207.
DeStefano, J. J., R. A. Bambara, and P. J. Fay. 1994. The mechanism of human immunodeficiency virus reverse transcriptase-catalyzed strand transfer from internal regions of heteropolymeric RNA templates. J. Biol. Chem. 269:161-168.
Donoghue, D. J., E. Rothenberg, N. Hopkins, D. Baltimore, and P. A. Sharp. 1978. Heteroduplex analysis of the nonhomology region between Moloney MuLV and the dual host range derivative HIX virus. Cell 14:959-970.
Evans, L. H. 1986. Characterization of polytropic MuLVs from three-week-old AKR/J mice. Virology 153:122-136.
Evans, L. H., and M. W. Cloyd. 1984. Generation of mink cell focus-forming viruses by Friend murine leukemia virus: recombination with specific endogenous proviral sequences. J. Virol. 49:772-781.
Evans, L. H., and M. W. Cloyd. 1985. Friend and Moloney murine leukemia viruses specifically recombine with different endogenous retroviral sequences to generate mink cell focus-forming viruses. Proc. Natl. Acad. Sci. USA 82:459-463.
Evans, L. H., and P. H. Duesberg. 1982. Isolation of a transformation-defective deletion mutant of Moloney murine sarcoma virus. J. Virol. 41:735-743.
Evans, L. H., P. H. Duesberg, D. L. Linemeyer, S. K. Ruscetti, and E. M. Scolnick. 1981. Structural and functional studies of the Friend spleen focus-forming virus: structural relationship of SFFV to dualtropic viruses and molecular cloning of a biologically active subgenomic fragment of SFFV DNA. Hamatol. Bluttransfus. 26:472-478.
Evans, L. H., M. Lavignon, M. Taylor, and A. S. Alamgir. 2003. Antigenic subclasses of polytropic murine leukemia virus (MLV) isolates reflect three distinct groups of endogenous polytropic MLV-related sequences in NFS/N mice. J. Virol. 77:10327-10338.
Evans, L. H., and F. G. Malik. 1987. Class II polytropic murine leukemia viruses (MuLVs) of AKR/J. mice: possible role in the generation of class I oncogenic polytropic MuLVs. J. Virol. 61:1882-1892.
Evans, L. H., and J. D. Morrey. 1987. Tissue-specific replication of Friend and Moloney murine leukemia viruses in infected mice. J. Virol. 61:1350-1357.
Fang, G., B. Weiser, C. Kuiken, S. M. Philpott, S. Rowland-Jones, F. Plummer, J. Kimani, B. Shi, R. Kaul, J. Bwayo, O. Anzala, and H. Burger. 2004. Recombination following superinfection by HIV-1. AIDS 18:153-159.
Gojobori, T., and S. Yokoyama. 1985. Rates of evolution of the retroviral oncogene of Moloney murine sarcoma virus and of its cellular homologues. Proc. Natl. Acad. Sci. USA 82:4198-4201.
Guo, J., T. Wu, B. F. Kane, D. G. Johnson, L. E. Henderson, R. J. Gorelick, and J. G. Levin. 2002. Subtle alterations of the native zinc finger structures have dramatic effects on the nucleic acid chaperone activity of human immunodeficiency virus type 1 nucleocapsid protein. J. Virol. 76:4370-4378.
Hartley, J. W., N. K. Wolford, L. J. Old, and W. P. Rowe. 1977. A new class of murine leukemia virus associated with development of spontaneous lymphomas. Proc. Natl. Acad. Sci. USA 74:789-792.
Hu, W. S., and H. M. Temin. 1990. Genetic consequences of packaging two RNA genomes in one retroviral particle: pseudodiploidy and high rate of genetic recombination. Proc. Natl. Acad. Sci. USA 87:1556-1560.
Ishimoto, A., A. Adachi, K. Sakai, T. Yorifuji, and S. Tsuruta. 1981. Rapid emergence of mink cell focus-forming (MCF) virus in various mice infected with NB-tropic friend virus. Virology 113:644-655.
Jung, Y. T., T. Wu, and C. A. Kozak. 2003. Characterization of recombinant nonecotropic murine leukemia viruses from the wild mouse species Mus spretus. J. Virol. 77:12773-12781.
Khan, A. S. 1984. Nucleotide sequence analysis establishes the role of endogenous murine leukemia virus DNA segments in formation of recombinant mink cell focus-forming murine leukemia viruses. J. Virol. 50:864-871.
Khan, A. S., W. P. Rowe, and M. A. Martin. 1982. Cloning of endogenous murine leukemia virus-related sequences from chromosomal DNA of BALB/c and AKR/J. mice: identification of an env progenitor of AKR-247 mink cell focus-forming proviral DNA. J. Virol. 44:625-636.
Koch, W., W. Zimmermann, A. Oliff, and R. Friedrich. 1984. Molecular analysis of the envelope gene and long terminal repeat of Friend mink cell focus-inducing virus: implications for the functions of these sequences. J. Virol. 49:828-840.
Lander, E. S., et. al. 2001. Initial sequencing and analysis of the human genome. Nature 409:860-921.
Lander, M. R., and S. K. Chattopadhyay. 1984. A Mus dunni cell line that lacks sequences closely related to endogenous murine leukemia viruses and can be infected by ectropic, amphotropic, xenotropic, and mink cell focus-forming viruses. J. Virol. 52:695-698.
Lavignon, M., and L. Evans. 1996. A multistep process of leukemogenesis in Moloney murine leukemia virus-infected mice that is modulated by retroviral pseudotyping and interference. J. Virol. 70:3852-3862.
Lavignon, M., J. Richardson, and L. H. Evans. 1997. A small region of the ecotropic murine leukemia virus (MuLV) gag gene profoundly influences the types of polytropic MuLVs generated in mice. J. Virol. 71:8923-8927.
Lavignon, M., J. L. Walker, S. M. Perryman, F. G. Malik, A. S. Khan, T. S. Theodore, and L. H. Evans. 1994. Characterization of epitopes defining two major subclasses of polytropic murine leukemia viruses (MuLVs) which are differentially expressed in mice infected with different ecotropic MuLVs. J. Virol. 68:5194-5203.
Lower, R. 1999. The pathogenic potential of endogenous retroviruses: facts and fantasies. Trends Microbiol. 7:350-356.
Lower, R., J. Lower, and R. Kurth. 1996. The viruses in all of us: characteristics and biological significance of human endogenous retrovirus sequences. Proc. Natl. Acad. Sci. USA 93:5177-5184.
Monk, R. J., F. G. Malik, D. Stokesberry, and L. H. Evans. 1992. Direct determination of the point mutation rate of a murine retrovirus. J. Virol. 66:3683-3689.
Muir, A., A. Lever, and A. Moffett. 2004. Expression and functions of human endogenous retroviruses in the placenta: an update. Placenta 25(Suppl. A):S16-S25.
Nakagawa, K., and L. C. Harrison. 1996. The potential roles of endogenous retroviruses in autoimmunity. Immunol. Rev. 152:193-236.
O'Donnell, P. V., E. Fleissner, H. Lonial, C. F. Koehne, and A. Reicin. 1985. Early clonality and high-frequency proviral integration into the c-myc locus in AKR leukemias. J. Virol. 55:500-503.
Oliver, P. L., and J. P. Stoye. 1999. Genetic analysis of Gv1, a gene controlling transcription of endogenous murine polytropic proviruses. J. Virol. 73:8227-8234.
Pfeiffer, J. K., and A. Telesnitsky. 2001. Effects of limiting homology at the site of intermolecular recombinogenic template switching during Moloney murine leukemia virus replication. J. Virol. 75:11263-11274.
Portis, J. L., S. Czub, S. Robertson, F. McAtee, and B. Chesebro. 1995. Characterization of a neurologic disease induced by a polytropic murine retrovirus: evidence for differential targeting of ecotropic and polytropic viruses in the brain. J. Virol. 69:8070-8075.
Quint, W., W. Boelens, P. van Wezenbeek, T. Cuypers, E. R. Maandag, G. Selten, and A. Berns. 1984. Generation of AKR mink cell focus-forming viruses: a conserved single-copy xenotrope-like provirus provides recombinant long terminal repeat sequences. J. Virol. 50:432-438.
Reicin, A., J.-Q. Yang, K. B. Marcu, E. Fleissner, C. F. Koehne, and P. V. O'Donnell. 1986. Deregulation of the c-myc oncogene in virus-induced thymic lymphomas of AKR/J mice. Mol. Cell. Biol. 6:4088-4092.
Rigby, R. J., S. J. Rozzo, H. Gill, T. Fernandez-Hart, B. J. Morley, S. Izui, B. L. Kotzin, and T. J. Vyse. 2004. A novel locus regulates both retroviral glycoprotein 70 and anti-glycoprotein 70 antibody production in New Zealand mice when crossed with BALB/c. J. Immunol. 172:5078-5085.
Rommelaere, J., D. V. Faller, and N. Hopkins. 1978. Characterization and mapping of RNase T1-resistant oligonucleotides derived from the genomes of Akv and MCF murine leukemia viruses. Proc. Natl. Acad. Sci. USA 75:495-499.
Rosen, C. A., W. A. Haseltine, J. Lenz, R. Ruprecht, and M. W. Cloyd. 1985. Tissue selectivity of murine leukemia virus infection is determined by long terminal repeat sequences. J. Virol. 55:862-866.
Rulli, K., P. A. Lobelle-Rich, A. Trubetskoy, J. Lenz, and L. S. Levy. 2001. Tissue distribution and timing of appearance of polytropic envelope recombinants during infection with SL3-3 murine leukemia virus or its weakly pathogenic SL3Myb5 mutant. J. Virol. 75:522-526.
Ruscetti, S., L. Davis, J. Feild, and A. Oliff. 1981. Friend murine leukemia virus-induced leukemia is associated with the formation of mink cell focus-inducing viruses and is blocked in mice expressing endogenous mink cell focus-inducing xenotropic viral envelope genes. J. Exp. Med. 154:907-920.
Sitbon, M., J. Nishio, K. Wehrly, D. Lodmell, and B. Chesebro. 1985. Use of a focal immunofluorescence assay on live cells for quantitation of retroviruses: distinction of host range classes in virus mixtures and biological cloning of dual-tropic murine leukemia viruses. Virology 141:110-118.
Stauffer, Y., G. Theiler, P. Sperisen, Y. Lebedev, and C. V. Jongeneel. 2004. Digital expression profiles of human endogenous retroviral families in normal and cancerous tissues. Cancer Immun. 4:2.
Stoye, J. P., and J. M. Coffin. 1987. The four classes of endogenous murine leukemia virus: structural relationships and potential for recombination. J. Virol. 61:2659-2669.
Stoye, J. P., C. Moroni, and J. M. Coffin. 1991. Virological events leading to spontaneous AKR thymomas. J. Virol. 65:1273-1285.
Stuhlmann, H., and P. Berg. 1992. Homologous recombination of copackaged retrovirus RNAs during reverse transcription. J. Virol. 66:2378-2388.
Thomas, C. Y., and J. M. Coffin. 1982. Genetic alterations of RNA leukemia viruses associated with the development of spontaneous thymic leukemia in AKR/J mice. J. Virol. 43:416-426.
Thomas, C. Y., R. Khiroya, R. S. Schwartz, and J. M. Coffin. 1984. Role of recombinant ecotropic and polytropic viruses in the development of spontaneous thymic lymphomas in HRS/J. mice. J. Virol. 50:397-407.
Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins. 1997. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876-4882.
van der Putten, H., W. Quint, J. van Raaij, E. R. Maandag, I. M. Verma, and A. Berns. 1981. M-MuLV-induced leukemogenesis: integration and structure of recombinant proviruses in tumors. Cell 24:729-739.
Wirschubsky, Z., P. Tsichlis, G. Klein, and J. Sumegi. 1986. Rearrangement of c-myc, pim-1 and Mlvi-1 and trisomy of chromosome 15 in MCF- and Moloney-MuLV-induced murine T-cell leukemias. Int. J. Cancer 38:739-745.
Wolff, L., E. Scolnick, and S. Ruscetti. 1983. Envelope gene of the Friend spleen focus-forming virus: deletion and insertions in 3' gp70/p15E-encoding region have resulted in unique features in the primary structure of its protein product. Proc. Natl. Acad. Sci. USA 80:4718-4722.
Wu, W., B. M. Blumberg, P. J. Fay, and R. A. Bambara. 1995. Strand transfer mediated by human immunodeficiency virus reverse transcriptase in vitro is promoted by pausing and results in misincorporation. J. Biol. Chem. 270:325-332.
Xu, H., and J. D. Boeke. 1987. High-frequency deletion between homologous sequences during retrotransposition of Ty elements in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 84:8553-8557.
Yi, J. M., H. M. Kim, and H. S. Kim. 2004. Expression of the human endogenous retrovirus HERV-W family in various human tissues and cancer cells. J. Gen. Virol. 85:1203-1210.
Zhang, W. H., C. K. Hwang, W. S. Hu, R. J. Gorelick, and V. K. Pathak. 2002. Zinc finger domain of murine leukemia virus nucleocapsid protein enhances the rate of viral DNA synthesis in vivo. J. Virol. 76:7473-7484.
Zuker, M. 2003. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res. 31:3406-3415.(A. S. M. Alamgir, Nick Ow)