Avian leukosis virus-induced tumors have common proviral integration sites and synthesize discrete new RNAs: oncogenesis by promoter insertion

Unlike other RNA tumor viruses, avian leukosis viruses (which cause lymphomas and occasionally other neoplasms) lack discrete "transforming genes". We have analyzed the virus-related DNA and RNA of avian leukosis virus (ALV)-induced tumors in an attempt to gain insight into the mechanism of ALV oncogenesis. Our results show that viral gene products are not required for maintenance of neoplastic transformation. Primary and metastatic tumors are clonal and thus presumably derived from a single infected cell. Most importantly, tumors from different birds have integration sites in common. Tumor cells synthesize discrete new poly(A) RNAs consisting of viral sequences covalently linked to cellular sequences. These RNA species are expressed at high levels in tumor cells. Our results suggest that in lymphoid tumors, an ALV provirus is integrated adjacent to a specific cellular gene, and the insertion of the viral promoter adjacent to this gene results in its enhanced expression, leading to neoplasia. These results have potentially important implications for the mechanism of non-viral carcinogenesis.     

Integration of bovine leukemia virus DNA in the genomes of bovine lymphosarcoma cells

Integration of bovine leukemia virus (BLV) in the genomes of infected cells was investigated in cattle with enzootic bovine leukosis (EBL) and sporadic bovine leukosis (SBL). Southern blot hybridization of BLV cDNA to Eco RI and Xba I restriction fragments of EBL tumor DNAs revealed that: 1) one to four or more copies of proviral DNA were integrated per genome; 2) the restriction pattern of the integrated proviral DNA was the same in two or three different tumors from the same animals; and 3) different patterns were observed among tumors from four different animals. These findings suggest the monoclonal origin of different tumors in an individual animal and the existence of multiple chromosomal integration sites of BLV provirus. DNAs from several SBL tumors were also analyzed with the same restriction enzymes, but with both representative and cDNA3'-enriched's of BLV RNA. No hybridization bands reactive with representative BLV cDNA could be detected, while several bands appeared to hybridize with cDNA3'-enriched.     

Chicken leukosis virus genome sequences in DNA from normal chick cells and virus-induced bursal lymphomas.

Genome sequences of two recent field isolates of avian leukosis viruses in the DNA of normal and neoplastic chicken cells were studied by DNA-RNA hybridization under conditions of DNA excess. Comparisons were made between 60-70S RNA from these viruses and that of a chicken endogenous type C virus (RAV-0), and of a series of "laboratory" leukosis and sarcoma viruses, by competitive hybridization analysis. A minimum of 18% of the genome sequences of both ALV isolates detected in DNA from lymphomas they induced were not detected in normal chicken DNA. The vast majority of the fraction of RNA sequences from ALV which do form hybrids with normal chick DNA appear to be reacting with the endogenous provirus of RAV-0. The genomic representation of a variety of avian leukosis and sarcoma viruses in normal chicken cells could not be distinguished by these methods (except that 13% of the RAV-0 genome was not shared with any of the other viruses). In contrast, the portion of the ALV genome exogenous to the normal chicken geome showed significant divergence from that of two sarcoma viruses (Pr RSV-C and B-77). The increased hybridization of ALV RNA with lymphoma DNA was used to detect the appearance of ALV specific sequences in the bursa of Fabricius following infection.increased hybridization was correlated with both the time after infection and the extent of replacement of the bursa by lymphoma. About one half of the increase in hybridization preceded histologic evidence of transformation.     

Persistence of Marek''s disease virus in a subpopulation of B cells that is transformed by avian leukosis virus, but not in normal bursal B cells.

Previous studies have described an augmentation of avian leukosis virus (ALV)-induced lymphoid leukosis in chickens that were coinfected with a serotype 2 Marek's disease virus (MDV) strain, SB-1. As a first step toward understanding the mechanism of this augmentation, we have analyzed the tropism of the MDV for the ALV-transformed B cell. After hatching, chickens were coinfected with ALV and a nonpathogenic strain of MDV, SB-1. Seventy primary and metastatic ALV-induced lymphomas that developed in chickens between 14 and 20 weeks of age were found, with only one exception, to carry SB-1 DNA. The MDV genome was maintained in cell lines derived from the tumors. However, MDV DNA could not be detected in nontransformed bursal B cells from chickens carrying ALV lymphomas. Moreover, during and after the lytic phase of MDV infection, SB-1 DNA was near or below the level of detection in bursal cells, suggesting that MDV most likely infects only a small subpopulation of bursal cells. By contrast, ALV-transformed B cells from MDV-free chickens could be persistently infected with MDV in vitro. These findings indicate that ALV lymphoma cells, unlike nontransformed bursal B cells, are susceptible to persistent MDV infection and can serve as a reservoir of MDV that can potentially influence the physiology of the transformed cell.     

Differential Selection of Cells with Proviral c-myc and c-erbB Integrations after Avian Leukosis Virus Infection

Avian leukosis virus (ALV) infection induces bursal lymphomas in chickens after proviral integration within the c-myc proto-oncogene and induces erythroblastosis after integration within the c-erbB proto-oncogene. A nested PCR assay was used to analyze the appearance of these integrations at an early stage of tumor induction after infection of embryos. Five to eight distinct proviral c-myc integration events were amplified from bursas of infected 35-day-old birds, in good agreement with the number of transformed bursal follicles arising with these integrations. Cells containing these integrations are remarkably common, with an estimated 1 in 350 bursal cells having proviral c-myc integrations. These integrations were clustered within the 3′ half of c-myc intron 1, in a pattern similar to that observed in bursal lymphomas. Bone marrow and spleen showed a similar number and pattern of integrations clustered within 3′ c-myc intron 1, indicating that this region is a common integration target whether or not that tissue undergoes tumor induction. While all tissues showed equivalent levels of viral infection, cells with c-myc integrations were much more abundant in the bursa than in other tissues, indicating that cells with proviral c-myc integrations are preferentially expanded within the bursal environment. Proviral integration within the c-erbB gene was also analyzed, to detect clustered c-erbB intron 14 integrations associated with erythroblastosis. Proviral c-erbB integrations were equally abundant in the bone marrow, spleen, and bursa. These integrations were randomly situated upstream of c-erbB exon 15, indicating that cells carrying 3′ intron 14 integrations must be selected during induction of erythroblastosis.     

Sint1, a common integration site in SL3-3-induced T-cell lymphomas, harbors a putative proto-oncogene with homology to the septin gene family

The murine retrovirus SL3-3 is a potent inducer of T-cell lymphomas when inoculated into susceptible newborn mice. Previously, DNAs from twenty SL3-3-induced tumors were screened by PCR for provirus integration sites. Two out of 20 tumors demonstrated clonal provirus insertion into a common region. This region has now been isolated and characterized. The region, named SL3-3 integration site 1 (Sint1), maps to the distal end of mouse chromosome 11, corresponding to human chromosome 17q25, and may be identical to a mouse mammary tumor virus integration site in a T-cell lymphoma, Pad3. Two overlapping genomic lambda clones spanning about 35 kb were isolated and used as a starting point for a search for genes in the neighborhood of the virus integration sites. A genomic fragment was used as a hybridization probe to isolate a 3-kb cDNA clone, the expression of which was upregulated in one of two tumors harboring a provirus in Sint1. The cDNA clone is predicted to encode a protein which shows 97.0% identity to a human septin-like protein encoded by a gene which has been found as a fusion partner gene of MLL in an acute myeloid leukemia with a t(11;17)(q23;q25). Together these findings raise the possibility that a proto-oncogene belonging to the septin family, and located about 15 kb upstream of the provirus integration sites, is involved in murine leukemia virus-induced T-cell lymphomagenesis.     

The integration sites of endogenous and exogenous Moloney murine leukemia virus

Specific cDNA probes of Moloney and AKR murine leukemia viruses have been prepared to characterize the proviral integration sites of these viruses in the genomes of Balb/Mo and Balb/c mice. The genetically transmitted Moloney provirus of Balb/Mo mice was detected in a characteristic Eco RI DNA fragment of 16 x 10(6) daltons. No fragment of this size was detected in tissue DNAs from Balb/c mice infected as newborns with Moloney virus. We conclude that a viral integration site, occupied in preimplantation mouse embryos, is not necessarily occupied when virus infects cells in post-natal animals. Balb/Mo and Balb/c mice do carry the AkR structural gene in an Eco RI DNA fragment of 12 x 10(6) daltons. Further restriction analysis of this fragment indicated that both mouse lines carry one AKR-type provirus. Leukemogenesis in Balb/Mo and newborn infected Balb/c mice is accompanied by reintegration of Moloney viral sequences in new chromosomal sites of tumor tissues. Part of the reintegrated Moloney viral sequences are of subgenomic size. The AKR viral sequences, however, are not found in new sites. Further restriction analysis revealed that the development of Moloney virus-induced leukemia in Balb/Mo mice does not lead to detectable structural alteration of the genetically transmitted Moloney and AKR structural genes. Possible mechanisms of the reintegration process are also discussed.     

Methylation and amplification of mouse mammary tumor virus DNA in normal, premalignant, and malignant cells of GR/A mice.

The methylation and amplification of mouse mammary tumor virus (MuMTV) proviral DNA was investigated in normal, premalignant, and malignant tissues of GR/A mice. The proviral methylation pattern was examined with the restriction enzyme HhaI, which fails to cleave methylated DNA. MuMTV proviral DNA from liver, kidney, and heart was highly methylated. Proviral DNA was somewhat undermethylated in mammary gland cells from virgin and lactating mice and extensively undermethylated in cells from premalignant outgrowths, pregnancy-dependent tumors, and pregnancy-independent tumors. The restriction enzyme SacI was used to detect additional proviruses in the same cells. No additional proviral copies of MuMTV were detected in liver, kidney, or heart cells or in mammary gland cells from virgin mice. Some mammary gland cells from lactating mice appeared to contain additional copies of the endogenous, highly oncogenic GT-MTV-2 provirus. Premalignant outgrowth, pregnancy-dependent tumor, and pregnancy-independent tumor cells contained an average of two to three additional copies per cell of the GT-MTV-2 provirus. Thus, neoplasia in GR/A mice was directly associated with quantized increases in MuMTV proviral DNA undermethylation and GR-MTV-2 proviral DNA amplification. Restriction enzyme analysis suggested that premalignant outgrowths and pregnancy-dependent tumors both consisted largely of heterogenous cell populations, although some evidence of clonal dominance was detected.     

Chromosome integration domain for bovine leukemia provirus in tumors.

The 3'-end host-virus junction fragments from two bovine leukemia virus (BLV)-induced lymphoid tumors (tumors 15-4 and 1351), each containing a single provirus, were used as probes to detect large restriction fragments flanking these proviruses. The DNAs from 28 other independent BLV-induced tumors were checked by Southern analysis of their restriction fragments for possible rearrangement due to the insertion of a BLV provirus in the cellular sequences corresponding to those flanking the proviruses in tumors 15-4 and 1351. In no case did proviral integration occur in cellular sequences corresponding to those implicated in the tumors of origin. According to the statistical analysis performed, if a preferential domain for BLV integration exists, it has a size of 1,304 kilobases when the probability of not observing an integration event in the cellular fragments considered in tumors 15-4 and 1351 is 0.50.     

Identification of the avian myeloblastosis virus genome. II. Restriction endonuclease analysis of DNA from lambda proviral recombinants and leukemic myeoblast clones.

Two lambda proviral DNA recombinants were characterized with a number of restriction endonucleases. One recombinant contained a complete presumptive avian myeloblastosis virus (AMV) provirus flanked by cellular sequences on either side, and the second recombinant contained 85% of a myeloblastosis-associated virus type 1 (MAV-1)-like provirus with cellular sequences adjacent to the 5' end of the provirus. Comparing the restriction maps for the proviral DNAs contained in each lambda hybrid showed that the putative AMV and MAV-1-like genomes shared identical enzyme sites for 3.6 megadaltons beginning at the 5' termini of the proviruses with respect to viral RNA. Two enzyme sites near the 3'-end of the MAV-1-like provirus were not present in the putative AMV genome. We also examined a number of leukemic myeloblast clones for proviral content and cell-provirus integration sites. The presumptive AMV provirus was present in all the leukemic myeloblast clones regardless of the endogenous proviral content of the target cells or the AMV pseudotype used for conversion. Multiple cellular sites were suitable for integration of the putative AMV genome and the helper genomes. The proviral genomes were all integrated colinearly with respect to linear viral DNA.     

Restriction endonuclease mapping of the proviral DNA of the exogenous RIII murine mammary tumor virus

Cellular DNA containing integrated murine mammary tumor virus (MuMTV) was isolated from FeI/C6 feline kidney cells and CCL64 mink lung cells infected with milkborne RIII MuMTV. By using restriction enzyme HpaI, intact RIII MuMTV provirus (length, 8.7 kilobases [kb]) was excised from the cellular DNA. Subsequent restriction endonuclease analysis of this HpaI fragment with KpnI, HindIII, EcoRI, BamHI, BglII, PstI, SstI, SalI, and XhoI enabled us to construct a map of the RIII virus genome. A comparison of this map with the maps of the GR and C3H MuMTV's revealed that there are greater sequence differences between the RIII virus and the GR and C3H MuMTV proviruses than there are between the GR and C3H proviruses. The following are features of the restriction map unique to the RIII provirus: the presence of three BamHI and two EcoRI cleavage sites, a HpaI cleavage site in the terminal 3'-5' repeat unit of the provirus, and the absence of an XhoI cleavage site. Another distinguishing feature of the RIII provirus is that the sizes of some of the restriction fragments produced by cleavage of the RIII provirus with PstI are different from the sizes of the fragments obtained by PstI cleavage of the GR and C3H proviruses. Like the GR proviral DNA, the RIII proviral DNA has three SstI (SacI) cleavage sites, whereas the C3H provirus has only two SstI sites. HpaI digestion of MuMTV-infected mink lung cell DNA revealed only one class of provirus (an 8.7-kb fragment); however, we observed several minor classes of RIII proviral DNA in addition to the major class of provirus DNA in infected cat kidney cells. PstI digestion of the HpaI 8.7-kb fragments from both feline and mink cells generated a 3.7-kb DNA fragment identical in size to a PstI-generated fragment that has been found in GR and C3H milkborne virus-infected cells. Although a fragment similar in size to the milkborne 3.7-kb PstI fragment has been found as an endogenous component in many C3H and GR mouse tissues, we did not observe such an endogenous fragment in the RIII mouse strain. Therefore, the 3.7-kb fragment may be useful as a marker for the milkborne RIII MuMTV provirus in RIII mice.     

AKR thymic lymphomas involving mink cell focus-inducing murine leukemia viruses have a common region of provirus integration

Newly acquired proviruses related to a mink cell focus-inducing murine leukemia virus were detected in low copy number in restriction endonuclease-digested DNAs from thymic lymphomas of AKR/J mice. These extra proviruses were not present in DNAs of either normal thymus or leukemic brain tissues. Extra tumor-specific DNA fragments generated by restriction endonucleases either were identical in size or fell into similar size classes, suggesting a common site(s) of provirus integration. Characterization of extra EcoRI DNA fragments for mink cell focus-inducing viral sequences revealed that all of them contained large terminal repeat sequences and that a significant number represented proviruses with deletions.