Mutation of the core or adjacent LVb elements of the Moloney murine leukemia virus enhancer alters disease specificity

Long terminal repeat enhancer core sequences in proviruses adjacent to c-myc in T-cell lymphomas induced by a murine retrovirus

The transcriptional enhancer in the long terminal repeat (LTR) of the T-lymphomagenic retrovirus SL3-3 differs from that of the nonleukemogenic virus Akv at several sites, including a single base pair difference in an element termed the enhancer core. Mutation of this T-A base pair to the C-G C-G sequence found in Akv significantly attenuated the leukemogenicity of SL3-3. Thus, this difference is important for viral leukemogenicity. Since Akv is an endogenous virus, this suggests that the C-G in its core is an adaptation to being minimally pathogenic. Most tumors that occurred in mice inoculated with the mutant virus, called SAA, contained proviruses with reversion or potential suppressor mutations in the enhancer core. We also found that the 72-bp tandem repeats constituting the viral enhancer could vary in number. Most tumors contained mixtures of proviruses with various numbers of 72-bp units, usually between one and four. Variation in repeat number was most likely due to recombination events involving template misalignment during viral replication. Thus, two processes during viral replication, misincorporation and recombination, combined to alter LTR enhancer structure and generate more pathogenic variants from the mutant virus. In SAA-induced tumors, enhancers of proviruses adjacent to c-myc had the largest number of core reversion or suppressor mutations of all of the viral enhancers in those tumors. This observation was consistent with the hypothesis that one function of the LTR enhancers in leukemogenesis is to activate proto-oncogenes such as c-myc.

AML1 is expressed in skeletal muscle and is regulated by innervation

Although most skeletal muscle genes are expressed at similar levels in electrically active, innervated muscle and in electrically inactive, denervated muscle, a small number of genes, including those encoding the acetylcholine receptor, N-CAM, and myogenin, are expressed at significantly higher levels in denervated than in innervated muscle. The mechanisms that mediate electrical activity-dependent gene regulation are not understood, but these mechanisms are likely to be responsible, at least in part, for the changes in muscle structure and function that accompany a decrease in myofiber electrical activity. To understand how muscle activity regulates muscle structure and function, we used a subtractive-hybridization and cloning strategy to identify and isolate genes that are expressed preferentially in innervated or denervated muscle. One of the genes which we found to be regulated by electrical activity is the recently discovered acute myeloid leukemia 1 (AML1) gene. Disruption and translocation of the human AML1 gene are responsible for a form of acute myeloid leukemia. AML1 is a DNA-binding protein, but its normal function is not known and its expression and regulation in skeletal muscle were not previously appreciated. Because of its potential role as a transcriptional mediator of electrical activity, we characterized expression of the AML1 gene in innervated, denervated, and developing skeletal muscle. We show that AML1 is expressed at low levels in innervated skeletal muscle and at 50- to 100-fold-higher levels in denervated muscle. Four AML1 transcripts are expressed in denervated muscle, and the abundance of each transcript increases after denervation. We transfected C2 muscle cells with an expression vector encoding AML1, tagged with an epitope from hemagglutinin, and we show that AML1 is a nuclear protein in muscle. AML1 dimerizes with core-binding factor beta (CBF beta), and we show that CGF beta is expressed at high levels in both innervated and denervated skeletal muscle. PEBP2 alpha, which is structurally related to AML1 and which also dimerizes with CBF beta, is expressed at low levels in skeletal muscle and is up-regulated only weakly by denervation. These results are consistent with the idea that AML1 may have a role in regulating gene expression in skeletal muscle.

AML1 is expressed in skeletal muscle and is regulated by innervation

Although most skeletal muscle genes are expressed at similar levels in electrically active, innervated muscle and in electrically inactive, denervated muscle, a small number of genes, including those encoding the acetylcholine receptor, N-CAM, and myogenin, are expressed at significantly higher levels in denervated than in innervated muscle. The mechanisms that mediate electrical activity-dependent gene regulation are not understood, but these mechanisms are likely to be responsible, at least in part, for the changes in muscle structure and function that accompany a decrease in myofiber electrical activity. To understand how muscle activity regulates muscle structure and function, we used a subtractive-hybridization and cloning strategy to identify and isolate genes that are expressed preferentially in innervated or denervated muscle. One of the genes which we found to be regulated by electrical activity is the recently discovered acute myeloid leukemia 1 (AML1) gene. Disruption and translocation of the human AML1 gene are responsible for a form of acute myeloid leukemia. AML1 is a DNA-binding protein, but its normal function is not known and its expression and regulation in skeletal muscle were not previously appreciated. Because of its potential role as a transcriptional mediator of electrical activity, we characterized expression of the AML1 gene in innervated, denervated, and developing skeletal muscle. We show that AML1 is expressed at low levels in innervated skeletal muscle and at 50- to 100-fold-higher levels in denervated muscle. Four AML1 transcripts are expressed in denervated muscle, and the abundance of each transcript increases after denervation. We transfected C2 muscle cells with an expression vector encoding AML1, tagged with an epitope from hemagglutinin, and we show that AML1 is a nuclear protein in muscle. AML1 dimerizes with core-binding factor beta (CBF beta), and we show that CGF beta is expressed at high levels in both innervated and denervated skeletal muscle. PEBP2 alpha, which is structurally related to AML1 and which also dimerizes with CBF beta, is expressed at low levels in skeletal muscle and is up-regulated only weakly by denervation. These results are consistent with the idea that AML1 may have a role in regulating gene expression in skeletal muscle.

AML1 fusion transcripts in t(3;21) positive leukemia: evidence of molecular heterogeneity and usage of splicing sites frequently involved in the generation of normal AML1 transcripts

The t(3;21)(q26;q22) is associated with chronic myelogenous leukemia in blast crisis (CML-BC), leukemia evolving from (therapy-related) myelodysplasia, and with leukemia following other hematopoietic proliferative diseases. Molecular cytogenetic analysis and cloning of a few t(3;21) cases indicate that the breakpoints are quite heterogeneous even within a specific clinical phenotype. Interestingly some of the (3;21) breakpoints involve the AML1 gene previously found rearranged in the t(8;21) associated with acute myelogenous leukemia. AML1 is related to the Drosophila gene runt and is the human counterpart of the gene for the alpha subunit of the nuclear polyoma enhancer binding protein (PEBP2) also known as the core binding factor (CBF). In the t(3;21) AML1 was found rearranged with EAP, a gene on chromosome 3 encoding a small ribosomal protein, as well as with EV11, another gene on chromosome 3. Here we report our study of six cases of t(3;21). By using fluorescence in situ hybridization (FISH) analysis and AML1 probes we could conclude that at least in two CML-BC cases the breakpoint occurred in the AML1 intron that is disrupted by the t(8;21). An AML1/EAP fusion transcript, different from the one described in a therapy-related myelodysplasia, was detected in both CML-BC cases. This transcript is expected to result in a predicted protein containing the AML1 nuclear binding domain with an attached stretch of 17 amino acids unrelated to the EAP small ribosomal protein. In the other t(3;21) patients we could not detect an AML1/EAP transcript or an AML1/EV11 transcript. This result suggests heterogeneity of the t(3;21) at the molecular level. The AML1 chimeric transcripts identified so far, both in the t(3;21) and in the t(8;21), diverge from the normal transcripts either after exon 5 or exon 6. Here we show that in normal AML1 transcripts different splicing events are seen to occur after AML1 exon 5 as well as exon 6.

Cloning of the MCF1233 murine leukemia virus and identification of sequences involved in viral tropism, oncogenicity and T cell epitope formation

MCF1233 is an oncogenic C57BL-derived retrovirus of the Murine Leukemia Virus (MuLV) family, that causes T and B lymphomas in an MHC-associated fashion. In this study, we cloned MCF1233, determined its nucleotide sequence and, by comparison with its MuLV relatives, identified the sequences that relate to the leukemogenic character of this virus. MCF1233 was found to have an ecotropic backbone, and carried acquired polytropic sequences in the 3' pol and 5' env region. The gag-region contained six specific nucleotides, determining the viral B-tropism. Short sequences within the U3 LTR shared specific homology with the xenotropic Bxv-1 MuLV, which is the U3 donor for leukemogenic MCF MuLV of AKR origin. These sequences, in combination with specific ecotropic sequences present in env p15E, most likely determine the viral oncogenicity. Currently, the deduced MCF1233 amino sequence is being exploited for T cell epitope analysis, which in this paper is discussed with respect to antigenically distinct Friend/Moloney/Rauscher types of MuLV. Identification of these T cell epitopes will contribute to our understanding of the fundamental aspects of immune control on MCF1233-induced lymphomagenesis. It will help to elucidate the mechanisms that underlie immune escape of T lymphomas, rarely arising in immunoresistant mice, and allow the development of vaccination protocols for tumor therapy.

Identification of ETS domain proteins in murine T lymphocytes that interact with the Moloney murine leukemia virus enhancer

The enhancer of Moloney murine leukemia virus (Mo-MuLV) contains an array of transcriptional control elements that direct viral gene expression in diverse cell types. The murine transcription factor Ets-1 was shown to bind to the LVb and LVc elements of the enhancer by DNase I protection and methylation interference assays. Enhancers containing disrupted Ets-1 binding sites were tested in transient expression assays in the murine T-cell line EL4.E1; alterations in the LVb element affected constitutive enhancer activity, while mutation of either the LVb or LVc element disrupted phorbol ester-induced enhancer activity. Members of the ets gene family of proteins display similar DNA-binding properties; therefore, we speculated that ets proteins other than Ets-1 also might bind these elements. Crude nuclear extracts of EL4.E1 cells were assayed to identify the protein(s) that potentially functions at the LVb element. The predominant binding activity was not Ets-1 but rather two independent DNA-protein complexes that comigrated in mobility shift assays. UV cross-linking and denaturing gel electrophoresis sized the two DNA-binding species, which we denoted p55 and p100. Immunoprecipitation combined with UV cross-linking identified p55 as the alpha subunit of GA-binding protein. The DNA-binding properties of p100 and several ets proteins were compared. Similarities suggested that p100 is also an ETS domain protein, possibly Elf-1. This strategy could be used to identify other ETS domain proteins in crude nuclear extracts. These findings suggest multiple ETS domain proteins could regulate gene expression of Mo-MuLV.

Identification of ETS domain proteins in murine T lymphocytes that interact with the Moloney murine leukemia virus enhancer

The enhancer of Moloney murine leukemia virus (Mo-MuLV) contains an array of transcriptional control elements that direct viral gene expression in diverse cell types. The murine transcription factor Ets-1 was shown to bind to the LVb and LVc elements of the enhancer by DNase I protection and methylation interference assays. Enhancers containing disrupted Ets-1 binding sites were tested in transient expression assays in the murine T-cell line EL4.E1; alterations in the LVb element affected constitutive enhancer activity, while mutation of either the LVb or LVc element disrupted phorbol ester-induced enhancer activity. Members of the ets gene family of proteins display similar DNA-binding properties; therefore, we speculated that ets proteins other than Ets-1 also might bind these elements. Crude nuclear extracts of EL4.E1 cells were assayed to identify the protein(s) that potentially functions at the LVb element. The predominant binding activity was not Ets-1 but rather two independent DNA-protein complexes that comigrated in mobility shift assays. UV cross-linking and denaturing gel electrophoresis sized the two DNA-binding species, which we denoted p55 and p100. Immunoprecipitation combined with UV cross-linking identified p55 as the alpha subunit of GA-binding protein. The DNA-binding properties of p100 and several ets proteins were compared. Similarities suggested that p100 is also an ETS domain protein, possibly Elf-1. This strategy could be used to identify other ETS domain proteins in crude nuclear extracts. These findings suggest multiple ETS domain proteins could regulate gene expression of Mo-MuLV.

Basic helix-loop-helix proteins in murine type C retrovirus transcriptional regulation

E boxes, recognition sequences for basic helix-loop-helix (bHLH) transcription factors, are detected in the enhancer and promoter regions of several murine type C retroviruses. Here we show that ALF1, a member of bHLH protein family of transcription factors, in vitro binds with differing affinities to distinct E-box sequences found in the U3 regulatory regions of Friend, Moloney, SL3-3, and Akv murine leukemia viruses (MLVs) as well as Friend spleen focus-forming virus (SFFV). In NIH 3T3 fibroblasts, ALF1 overexpression elevated transcription from the U3 region of Moloney MLV and the complete long terminal repeat regions of Friend SFFV, Akv MLV, and SL3-3 MLV but neither from the U3 region nor from the complete long terminal repeat of Friend MLV. Introduction of mutations in the Akv MLV E boxes showed the E-box cis elements to be required for the function of ALF1 as a transcription factor. ALF1 and the glucocorticoid receptor, with overlapping DNA binding sequences, did not act synergistically with respect to transcriptional trans activation of expression from the Akv MLV promoter-enhancer region. We conclude that ALF1 in vivo may be an important transcription regulator for Akv, SL3-3, and Moloney MLVs as well as for Friend SFFV.

Pathogenic determinants in the U3 region of recombinant murine leukemia viruses isolated from CWD and HRS/J mice

Recombinant murine leukemia viruses (MuLVs) from high-leukemia-incidence mouse strains typically acquire pathogenic U3 region sequences from the genome of the endogenous xenotropic virus, Bxv-1. However, a recombinant virus isolated from a leukemic HRS/J mouse and another from a CWD mouse contained U3 regions that lacked genetic markers of Bxv-1. The U3 regions of both recombinants were derived from the endogenous ecotropic virus Env-1 and had retained a single enhancer element. However, compared with that of Emv-1, the U3 region of each of the recombinant viruses contained five nucleotide substitutions, one of which was shared. To determine the biological significance of these substitutions, chimeric ecotropic viruses that contained the U3 region from one of the two recombinant viruses or from Emv-1 were injected into NIH Swiss mice. All three of the chimeric ecotropic viruses were leukemogenic following a long latency. Despite the presence of an enhancer core motif that is known to contribute to the leukemogenicity of the AKR MuLV SL3-3, the HRS/J virus U3 region induced lymphomas only slightly more rapidly than the allelic Emv-1 sequences. The chimeric virus with the U3 region of the CWD recombinant caused lymphomas more frequently and more rapidly than either of the other two viruses. The results support the hypothesis that one or more of the five nucleotide substitutions in the U3 regions of the recombinants contribute to viral pathogenicity. Comparison of DNA sequences suggests that the pathogenicity of the CWD virus U3 region was related to a sequence motif that is shared with Bxv-1 and is recognized by the basic helix-loop-helix class of transcription factors.

Origins of enhancer sequences of recombinant murine leukemia viruses from spontaneous B- and T-cell lymphomas of CWD mice

Recombinant murine leukemia viruses from the highly leukemic mouse strains AKR, HRS, and C58 usually acquire pathogenic U3 region sequences fro the endogenous xenotropic virus, Bxv-1. However, the majority of tumors from another highly leukemic strain, CWD, contained recombinant viruses that lacked Bxv-1-specific sequences. The nucleotide sequence of the U3 regions of two such CWD recombinants was nearly identical to that of the endogenous ecotropic virus parent Emv-1, but they shared three nucleotide substitutions immediately 3' of the enhancer core. These substitutions were found in recombinant proviruses from about one-third of spontaneous CWD lymphomas as determined by an oligonucleotide hybridization assay of proviral fragments that had been nucleotide substitutions in the CWD viruses were inherited from an endogenous polytropic provirus that is absent in the other highly leukemic strains. On the basis of the results of these and previous studies, we propose that CWD recombinants acquire pathogenic U3 region sequences through recombination with an endogenous polytropic virus or Bxv-1 and that the pathogenicity of these sequences may be related to a sequence motif that is known to bind members of the basic helix-loop-helix class of transcription factors.

Retroviral vectors containing chimeric promoter/enhancer elements exhibit cell-type-specific gene expression

Retroviral vectors were constructed in which the U3 promoter/enhancer of Moloney murine leukemia (Mo-MLV) was replaced by the corresponding region from five related murine retroviruses--AKR murine leukemia virus (AKV), Harvey murine sarcoma virus (HaMSV), myeloproliferative sarcoma virus (MPSV), SL3-3, and the NZB-xenotropic virus (Xeno). In these vectors the chimeric long terminal repeat (chLTR) drives the expression of the chloramphenicol acetyl transferase (CAT) reporter gene that is followed by an internal SV40 virus early region promoter linked to the neomycin phosphotransferase II (NEO) gene. As an initial measure of the relative promoter/enhancer strength of the chLTR vectors, the murine NIH-3T3 cell line and the human JURKAT cell lines were transfected and assayed for CAT reporter activity. Relative to the MoMLV vector, the HaMSV construct was the most active in NIH-3T3 cells whereas the SL3-3 vector displayed the greatest activity in JURKAT cells. Retroviral vector producer cell populations and cell clones were established for each chLTR vector, and all were capable of yielding high vector titers (> 10(5) G418R cfu/ml on NIH-3T3). Supernatant from these cells was used to transduce both mouse and human cell lines and primary cells. In NIH-3T3 cells and two murine fibrosarcoma cell lines, the HaMSV chLTR vector was slightly more active than the MoMLV chLTR vector. In the human HepG2 and HeLa cell lines, the MPSV chLTR vector was the most active. Data from the human JURKAT T-cell line and a T cell line derived from an ADA-deficient severe combined immunodeficiency (SCID) patient demonstrate that the SL3-3 chLTR is the most active in these lymphoid cell lines. The greatest difference in the comparison of the different chLTR vectors was observed in primary human umbilical vein endothelial cells, where the MoMLV vector produced up to 100 times more CAT activity than the SL3-3 vector. These data suggest that the use of specific promoter/enhancer elements may lead to higher levels of gene expression following retroviral-mediated gene transfer into specific cell types and these observations may be useful in the design of human gene therapy experiments.

Identification of a region of a murine leukemia virus long terminal repeat with novel transcriptional regulatory activities

The 93-bp region downstream of the enhancer (DEN) in the long terminal repeat (LTR) of the mink cell focus-forming virus (MCF13) has been shown to be important for transcriptional activation and viral lymphomagenicity (J. C. Tupper, H. Chen, E. F. Hays, G. C. Bristol, and F. K. Yoshimura, J. Virol. 66:7080-7088, 1992). In this report, we have further explored the role of the DEN region in transcriptional activation. We observed that it has enhancer-like abilities as well as some unique LTR properties. Transcriptional activation by the DEN region involved interactions with enhancer sequences that were either synergistic or additive, depending on the cell type. The most intriguing property of the DEN region is its ability to induce transcription in activated T cells. This activity is unique for the LTR in that no other LTR region can do this. We also examined the role of the DEN region in retroviral lymphomagenesis. We cloned and sequenced proviral LTRs integrated upstream of the cellular c-myc gene from DNA obtained from thymic tumors induced by DEN region deletion mutant viruses in AKR mice. We determined that for transcriptional activation of the c-myc proto-oncogene, enhancer sequences can substitute for the DEN region. This study identifies the significance of non-enhancer sequences in the LTR for the oncogenesis of the MCF13 retrovirus.

Transcriptional regulation of T-cell genes during T-cell development

The expression of many T cell specific genes has been shown to be regulated at the transcriptional level. Recent studies of T-cell specific promoters and enhancers have allowed the identification of a number of transcription factors that appear to play distinct but complementary roles in regulating gene expression during T-cell development and activation.

Inducible and cell type-specific expression of VL30 U3 subgroups correlate with their enhancer design

The murine VL30 elements constitute one family of retrotransposons represented in 100 to 200 copies that are dispersed among the mouse chromosomes. On the basis of sequence homology, we have subdivided mouse VL30 members into four distinct U3 subgroups. The use of subgroup-specific probes in Northern (RNA) blot analyses shows that individual VL30 U3 subgroups are expressed in a tissue-specific manner. We show by in situ hybridization of mouse skin treated with 12-O-tetradecanoylphorbol-13-acetate (TPA) that VL30 expression is induced in epidermal keratinocytes but not in dermal fibroblasts. Transient transfections of reporter gene plasmids together with in vitro binding analysis indicate that TPA-induced VL30 transcription specific for keratinocytes is mediated by two cooperating sequence motifs in juxtaposed position. One sequence motif is shown to constitutively bind CREB- and Jun-related proteins in both keratinocytes and fibroblasts, whereas the other is a target for TPA-induced c-Rel/p65(NF-kappa B)-binding activity specifically in keratinocytes. These binding sites are found to be conserved within U3 subgroups and individual U3 regions showing induced expression in TPA-treated mouse epidermis. These results together with a sequence comparison between different U3 subgroups indicate that cell type-specific activity of transcription factors known to regulate VL30 transcription and the presence or absence of their cognate binding sites within individual U3 regions determine inducible and cell type-specific VL30 expression. The variable VL30 U3 regions might thus be useful tools to study inducible and cell type-specific transcription in many different cell systems.

Protein interactions with DNA elements in variant equine infectious anemia virus enhancers and their impact on transcriptional activity

The long terminal repeats (LTRs) from various cloned equine infectious anemia virus (EIAV) proviruses differ significantly, but all contain cis-acting DNA elements identical to MDBP-, PEA2-, AP-1-, and PU.1 (ets)-binding sites. A prototype EIAV LTR would contain one of each of these conserved elements. The LTR variations originate from the insertion of novel sequences between the PEA2 and AP-1 elements in the transcriptional enhancer unit. Viewed in this way, the LTR from provirus clone lambda 12 has an 11-bp insertion containing a PEA2 site and the LTR of the lambda 6 provirus has a 31-bp insertion/duplication containing PEA2, AP-1, and PU.1 sites. Two other LTRs were cloned by amplification of cDNAs from the persistently infected cell line, EIAV-FEA. A third LTR was generated by site-directed mutagenesis of one of the LTRs from EIAV-FEA cells. The latter three had a single base change in the element next to the TATA box that abolished PU.1 binding; however, the variable regions of these LTRs were shown by gel mobility shift assays to contain one or two PU.1 sites. One variable region was shown to have an octamer site overlapping its tandem PU.1 elements. Basal, PMA-activated, and Tat trans-activated transcriptional activities of the LTRs were compared in several different cell lines by transient transfection. The various promoters displayed different relative levels of activity depending on the cell line used and the condition of activation. This natural set of variant promoters may help define how changes in the components of the transcription complex influence transactivation by Tat. The diverse LTRs could endow their respective proviruses with a unique pattern of expression and activation in vivo.

Identification of AML-1 and the (8;21) translocation protein (AML-1/ETO) as sequence-specific DNA-binding proteins: the runt homology domain is required for DNA binding and protein-protein interactions

The AML1 gene on chromosome 21 is disrupted in the (8;21)(q22;q22) translocation associated with acute myelogenous leukemia and encodes a protein with a central 118-amino-acid domain with 69% homology to the Drosophila pair-rule gene, runt. We demonstrate that AML-1 is a DNA-binding protein which specifically interacts with a sequence belonging to the group of enhancer core motifs, TGT/cGGT. Electrophoretic mobility shift analysis of cell extracts identified two AML-1-containing protein-DNA complexes whose electrophoretic mobilities were slower than those of complexes formed with AML-1 produced in vitro. Mixing of in vitro-produced AML-1 with cell extracts prior to gel mobility shift analysis resulted in the formation of higher-order complexes. Deletion mutagenesis of AML-1 revealed that the runt homology domain mediates both sequence-specific DNA binding and protein-protein interactions. The hybrid product, AML-1/ETO, which results from the (8;21) translocation and retains the runt homology domain, both recognizes the AML-1 consensus sequence and interacts with other cellular proteins.

Identification of AML-1 and the (8;21) translocation protein (AML-1/ETO) as sequence-specific DNA-binding proteins: the runt homology domain is required for DNA binding and protein-protein interactions

The AML1 gene on chromosome 21 is disrupted in the (8;21)(q22;q22) translocation associated with acute myelogenous leukemia and encodes a protein with a central 118-amino-acid domain with 69% homology to the Drosophila pair-rule gene, runt. We demonstrate that AML-1 is a DNA-binding protein which specifically interacts with a sequence belonging to the group of enhancer core motifs, TGT/cGGT. Electrophoretic mobility shift analysis of cell extracts identified two AML-1-containing protein-DNA complexes whose electrophoretic mobilities were slower than those of complexes formed with AML-1 produced in vitro. Mixing of in vitro-produced AML-1 with cell extracts prior to gel mobility shift analysis resulted in the formation of higher-order complexes. Deletion mutagenesis of AML-1 revealed that the runt homology domain mediates both sequence-specific DNA binding and protein-protein interactions. The hybrid product, AML-1/ETO, which results from the (8;21) translocation and retains the runt homology domain, both recognizes the AML-1 consensus sequence and interacts with other cellular proteins.

Sequences responsible for the distinctive hemolytic potentials of Friend and Moloney murine leukemia viruses are dispersed but confined to the psi-gag-PR region

Friend and Moloney murine leukemia viruses (F- and M-MuLV) induce distinct diseases in hematopoietic tissues following inoculation of newborn mice of susceptible strains. F-MuLV induces erythroleukemia preceded by severe early hemolytic anemia; M-MuLV induces thymomas and only very mild hemolysis. The major viral determinant of severe early hemolytic anemia residues in the env gene, but sequences located outside this gene can modulate this effect. By means of genetic chimeras of F- and M-MuLV, we have found that although they are confined to the 5' portion of the env gene intron, sequences that determine the distinctive hemolytic potentials of F- and M-MuLV are widely distributed over a region spanning the RNA encapsidation domain, the gag gene, and the portion of the pol gene encoding the viral protease. Within this large region, two fragments of M-MuLV, a 1.3-kb region encoding the matrix, pp12, and capsid proteins and a 0.8-kb region encoding the nucleocapsid and the viral protease, were capable, individually, of partially attenuating the capacity of F-MuLV for induction of severe early hemolytic anemia. In association, these two fragments conferred complete attenuation. Moreover, a second pair of adjacent fragments within this large region appeared to behave cooperatively to confer complete attenuation; a 0.36-kb region roughly corresponding to the encapsidation domain, although not detectably altering hemolytic potential on its own, deepened the attenuation conferred by the adjacent 1.3-kb region. Whether capable of inducing severe early hemolytic anemia or not and despite different efficiencies of induction of recombinant polytropic viruses, all chimeric viruses retained the erythroleukemogenicity of the F-MuLV parent.

Protection against retroviral diseases after vaccination is conferred by interference to superinfection with attenuated murine leukemia viruses

Cell cultures expressing a retroviral envelope are relatively resistant to superinfection by retroviruses which bear envelopes using the same receptor. We tested whether this phenomenon, known as interference to superinfection, might confer protection against retroviral diseases. Newborn mice first inoculated with the attenuated strain B3 of Friend murine leukemia virus (F-MuLV) were protected against severe early hemolytic anemia and nonacute anemiant erythroleukemia induced by the virulent strain 57 of F-MuLV. Vaccinated animals were also protected as adults against acute polycythemic erythroleukemia induced upon inoculation with the viral complex containing the defective spleen focus-forming virus and F-MuLV 57 as helper virus. Animals were inoculated as newborns, which is known to induce immune tolerance in mice, and the rapid kinetics of protection, incompatible with the delay necessary for the immune response to develop, indicated that protection was not due to an immune mechanism but rather was due to the rapid and long-lasting phenomenon of interference. This result was confirmed by combining parental and envelope chimeric MuLV from different interference groups as vaccinal and challenge viruses. Although efficient protection could be provided by vaccination by interference, we observed that attenuated replication-competent retroviruses from heterologous interference groups might exert deleterious synergistic effects.

Fusion between transcription factor CBF beta/PEBP2 beta and a myosin heavy chain in acute myeloid leukemia

The pericentric inversion of chromosome 16 [inv(16)(p13q22)] is a characteristic karyotypic abnormality associated with acute myeloid leukemia, most commonly of the M4Eo subtype. The 16p and 16q breakpoints were pinpointed by yeast artificial chromosome and cosmid cloning, and the two genes involved in this inversion were identified. On 16q the inversion occurred near the end of the coding region for CBF beta, also known as PEBP2 beta, a subunit of a heterodimeric transcription factor regulating genes expressed in T cells; on 16p a smooth muscle myosin heavy chain (SMMHC) gene (MYH11) was interrupted. In six of six inv(16) patient samples tested, an in-frame fusion messenger RNA was demonstrated that connected the first 165 amino acids of CBF beta with the tail region of SMMHC. The repeated coiled coil of SMMHC may result in dimerization of the CBF beta fusion protein, which in turn would lead to alterations in transcriptional regulation and contribute to leukemic transformation.

Cloning and characterization of subunits of the T-cell receptor and murine leukemia virus enhancer core-binding factor

Moloney murine leukemia virus causes thymic leukemias when injected into newborn mice. A major determinant of the thymic disease specificity of Moloney virus genetically maps to the conserved viral core motif in the Moloney virus enhancer. Point mutations introduced into the core site significantly shifted the disease specificity of the Moloney virus from thymic leukemia to erythroid leukemia (N.A. Speck, B. Renjifo, E. Golemis, T.N. Fredrickson, J.W. Hartley, and N. Hopkins, Genes Dev. 4:233-242, 1990). We previously reported the purification of core-binding factors (CBF) from calf thymus nuclei (S. Wang and N.A. Speck, Mol. Cell. Biol. 12:89-102, 1992). CBF binds to core sites in murine leukemia virus and T-cell receptor enhancers. Affinity-purified CBF contains multiple polypeptides. In this study, we sequenced five tryptic peptides from two of the bovine CBF proteins and isolated three cDNA clones from a mouse thymus cDNA library encoding three of the tryptic peptides from the bovine proteins. The cDNA clones, which we call CBF beta p22.0, CBF beta p21.5, and CBF beta p17.6, encode three highly related but distinct proteins with deduced molecular sizes of 22.0, 21.5, and 17.6 kDa that appear to be translated from multiply spliced mRNAs transcribed from the same gene. CBF beta p22.0, CBF beta p21.5, and CBF beta p17.6 do not by themselves bind the core site. However, CBF beta p22.0 and CBF beta p21.5 form a complex with DNA-binding CBF alpha subunits and as a result decrease the rate of dissociation of the CBF protein-DNA complex. Association of the CBF beta subunits does not extend the phosphate contacts in the binding site. We propose that CBF beta is a non-DNA-binding subunit of CBF and does not contact DNA directly.

Cloning and characterization of subunits of the T-cell receptor and murine leukemia virus enhancer core-binding factor

Moloney murine leukemia virus causes thymic leukemias when injected into newborn mice. A major determinant of the thymic disease specificity of Moloney virus genetically maps to the conserved viral core motif in the Moloney virus enhancer. Point mutations introduced into the core site significantly shifted the disease specificity of the Moloney virus from thymic leukemia to erythroid leukemia (N.A. Speck, B. Renjifo, E. Golemis, T.N. Fredrickson, J.W. Hartley, and N. Hopkins, Genes Dev. 4:233-242, 1990). We previously reported the purification of core-binding factors (CBF) from calf thymus nuclei (S. Wang and N.A. Speck, Mol. Cell. Biol. 12:89-102, 1992). CBF binds to core sites in murine leukemia virus and T-cell receptor enhancers. Affinity-purified CBF contains multiple polypeptides. In this study, we sequenced five tryptic peptides from two of the bovine CBF proteins and isolated three cDNA clones from a mouse thymus cDNA library encoding three of the tryptic peptides from the bovine proteins. The cDNA clones, which we call CBF beta p22.0, CBF beta p21.5, and CBF beta p17.6, encode three highly related but distinct proteins with deduced molecular sizes of 22.0, 21.5, and 17.6 kDa that appear to be translated from multiply spliced mRNAs transcribed from the same gene. CBF beta p22.0, CBF beta p21.5, and CBF beta p17.6 do not by themselves bind the core site. However, CBF beta p22.0 and CBF beta p21.5 form a complex with DNA-binding CBF alpha subunits and as a result decrease the rate of dissociation of the CBF protein-DNA complex. Association of the CBF beta subunits does not extend the phosphate contacts in the binding site. We propose that CBF beta is a non-DNA-binding subunit of CBF and does not contact DNA directly.

Characterization of a protein that binds multiple sequences in mammalian type C retrovirus enhancers

Mammalian type C retrovirus enhancer factor 1 (MCREF-1) is a nuclear protein that binds several directly repeated sequences (CNGGN6CNGG) in the Moloney and Friend murine leukemia virus (MLV) enhancers (N. R. Manley, M. O'Connell, W. Sun, N. A. Speck, and N. Hopkins, J. Virol. 67:1967-1975, 1993). In this paper, we describe the partial purification of MCREF-1 from calf thymus nuclei and further characterize the binding properties of MCREF-1. MCREF-1 binds four sites in the Moloney MLV enhancer and three sites in the Friend MLV enhancer. Ethylation interference analysis suggests that the MCREF-1 binding site spans two adjacent minor grooves of DNA.

A protein-binding site with dyad symmetry in the long terminal repeat of the MCF13 murine leukemia virus that contributes to transcriptional activity in T lymphocytes

We have previously identified regions in the long terminal repeat (LTR) of the MCF13 murine leukemia virus (MLV) that contribute to transcriptional activity in different cell types. We have observed that enhancer sequences and a region that resides 3' of the enhancer make significant contributions to transcriptional activity in T lymphocytes (T. Hollon and F. K. Yoshimura, J. Virol. 63:3353-3361, 1989). In this report, we have focused on the region of the MCF13 LTR that is 3' of the enhancer to identify binding sites for proteins that may play a role in the regulation of transcription in T cells. By gel shift and DNA footprint analyses, we have identified a single protein-binding site (MLPal) that includes a nucleotide sequence with dyad symmetry. A synthetic double-stranded oligonucleotide corresponding to this protein-binding site formed a specific protein-DNA complex. Deletion of this protein-binding site from the wild-type LTR decreased transcriptional activity in T lymphocytes but not in fibroblasts as determined by a transient expression assay. The MLPal sequence by itself cannot augment transcription in T cells but is able to do so in conjunction with enhancer sequences.

Two factors that bind to highly conserved sequences in mammalian type C retroviral enhancers

The transcriptional enhancers of the Moloney and Friend murine leukemia viruses (MLV) are important determinants of viral pathogenicity. We used electrophoretic mobility shift and methylation interference assays to study nuclear factors which bind to a region of these enhancers whose sequence is identical between Moloney and Friend viruses and particularly highly conserved among 35 mammalian type C retroviruses whose enhancer sequences have been aligned (E. Golemis, N. A. Speck, and N. Hopkins, J. Virol. 64:534-542, 1990). Previous studies identified sites for the leukemia virus factor b (LVb) and core proteins in this region (N. A. Speck and D. Baltimore, Mol. Cell. Biol. 7:1101-1110, 1987) as well as a site, overlapping those for LVb and core, for a third factor (N. R. Manley, M. A. O'Connell, P. A. Sharp, and N. Hopkins, J. Virol. 63:4210-4223, 1989). Surprisingly, the latter factor appeared to also bind two sites identified in the Friend MLV enhancer, Friend virus factor a and b1 (FVa and FVb1) sites, although the sequence basis for the ability of the protein to bind these diverse sites was not apparent. Here we describe the further characterization of this binding activity, termed MCREF-1 (for mammalian type C retrovirus enhancer factor 1), and the identification of a consensus sequence for its binding, GGN8GG. We also identify a factor, abundant in mouse T-cell lines and designated LVt, which binds to two sites in the Moloney MLV enhancer, overlapping the previously identified LVb and LVc binding sites. These sites contain the consensus binding site for the Ets family of proteins. We speculate on how distinct arrays of these factors may influence the disease-inducing phenotype.

Sequence specificity of the core-binding factor

The core-binding factor (CBF) binds the conserved core motif in mammalian type C retrovirus enhancers. We analyzed the phosphate contacts made by CBF on the Moloney murine leukemia virus enhancer by ethylation interference assay. The phosphate contacts span 9 bp centered around the consensus core site. To examine the sequence preferences for CBF binding, we employed the technique of selected and amplified binding sequence footprinting (T. K. Blackwell and H. Weintraub, Science 250:1104-1110, 1990). The consensus binding site for CBF defined by selected and amplified binding sequence footprinting is PyGPyG GTPy.

Mouse mammary tumor virus with rearranged long terminal repeats causes murine lymphomas

Mouse mammary tumor virus (MMTV) is a slowly transforming retrovirus associated primarily with the induction of mammary tumors. It is widely accepted that T-cell lymphomas of various mouse strains are associated with extra proviruses of MMTV. These extra proviruses showed site-specific rearrangements in the U3 region of long terminal repeats (LTRs), consisting of about 400 nucleotide deletions and occasional substitution resulting in unique tandem repeats. However, the question of whether these mutant MMTVs cause lymphomas has not been experimentally resolved. Here we present distinct evidence that they do. We constructed chimeric MMTVs by replacing the LTR of the recently constructed pathogenic MMTV provirus clone with rearranged LTRs of MMTV proviruses obtained from two DBA/2 mouse lymphoma cell lines, MLA and DL-8, and inoculated them into BALB/c mice. These mice developed lymphomas, but no mammary tumors, 4 to 11 months postinoculation, whereas the original pathogenic MMTV clone alone induced mammary tumors. These results showed that the tissue specificity of MMTV tumorigenesis is determined by the LTR structures.

A common integration locus in type B retrovirus-induced thymic lymphomas

Type-B leukemogenic retrovirus (TBLV) is a replication-competent type-B thymotropic retrovirus which lacks a transforming gene and whose genome is > 98% homologous to that of type-B mouse mammary tumor virus (MMTV). In contrast to MMTV, which induces mammary adenocarcinomas, TBLV induces a high incidence of T-cell thymic lymphomas in mice after a very short latent period. To investigate the molecular mechanisms by which TBLV induces T-cell lymphomas, we screened TBLV-induced tumor DNA for the frequent disruption of a particular cellular locus by TBLV proviral copies. In approximately 20% of the 55 primary tumors screened, the presence of proviruses in a common integration site was detected. This locus spans at least 53 kb of genomic DNA and maps to the mouse X chromosome. The presence of a functional gene at this locus is suggested by the conservation of nucleotide sequences from this locus among diverse animal species and by the expression of these sequences as mRNA in normal mouse tissues and tumors. The majority (17/18) of TBLV-induced primary tumors examined have elevated levels of this expressed mRNA.

Contributions to transcriptional activity and to viral leukemogenicity made by sequences within and downstream of the MCF13 murine leukemia virus enhancer

We have identified nucleotide sequences that regulate transcription in both a cell-type-specific and general manner in the long terminal repeat of the MCF13 murine leukemia virus. Besides the enhancer element, we have observed that the region between the enhancer and promoter (DEN) has a profound effect on transcription in different cell types. This effect, however, was dependent on the copy number of enhancer repeats and was detectable in the presence of a single repeat. When two enhancer repeats were present, the effect of DEN on transcription was abrogated except in T cells. DEN also makes a significant contribution to the leukemogenic property of the MCF13 retrovirus. Its deletion from the MCF13 virus dramatically reduced the incidence of thymic lymphoma and increased the latency of disease in comparison with the wild-type virus. This effect was most marked when one rather than two enhancer repeats was present in the mutant viruses. We also observed that the removal of one repeat alone remarkably reduced leukemogenicity by the MCF13 virus. A newly identified protein-binding site (MLPal) located within DEN affects transcription only in T cells, and its deletion attenuates the ability of an MCF13 virus with a single enhancer repeat to induce thymic lymphoma. This observation suggests that the MLPal protein-binding site contributes to the effect of the DEN region on T-cell-specific transcription and viral leukemogenicity. This study identifies the importance of nonenhancer sequences in the long terminal repeat for the oncogenesis of the MCF13 retrovirus.

Indistinguishable nuclear factor binding to functional core sites of the T-cell receptor delta and murine leukemia virus enhancers

We have previously shown that the delta E3 site is an essential element for transcriptional activation by the human T-cell receptor (TCR) delta enhancer and identified two factors, NF-delta E3A and NF-delta E3C, that bound to overlapping core (TGTGGTTT) and E-box motifs within delta E3. In this study, we show that protein binding to the core motif is necessary but not sufficient for transcriptional activation by the delta E3 element. In contrast, protein binding to the E-box motif does not contribute significantly to enhancer activity. A similar core motif present within the enhancers of T-cell-tropic murine retroviruses has been shown to contribute to transcriptional activity of the viral long terminal repeat in T lymphocytes and to viral T-cell tropism. We therefore determined the relationship between the nuclear factors that bind to the TCR delta and Moloney murine leukemia virus core motifs. On the basis of electrophoretic mobility shift binding and competition studies, biochemical analysis of affinity-labeled DNA-binding proteins, and the binding of a purified core binding factor, the proteins that bound to the TCR delta core site were indistinguishable from those that bound to the murine leukemia virus core site. These data argue that DNA-binding proteins that interact with the core site of murine leukemia virus long terminal repeats and contribute to viral T-cell tropism also play an essential role in the T-cell-specific expression of cellular genes.

Indistinguishable nuclear factor binding to functional core sites of the T-cell receptor delta and murine leukemia virus enhancers

We have previously shown that the delta E3 site is an essential element for transcriptional activation by the human T-cell receptor (TCR) delta enhancer and identified two factors, NF-delta E3A and NF-delta E3C, that bound to overlapping core (TGTGGTTT) and E-box motifs within delta E3. In this study, we show that protein binding to the core motif is necessary but not sufficient for transcriptional activation by the delta E3 element. In contrast, protein binding to the E-box motif does not contribute significantly to enhancer activity. A similar core motif present within the enhancers of T-cell-tropic murine retroviruses has been shown to contribute to transcriptional activity of the viral long terminal repeat in T lymphocytes and to viral T-cell tropism. We therefore determined the relationship between the nuclear factors that bind to the TCR delta and Moloney murine leukemia virus core motifs. On the basis of electrophoretic mobility shift binding and competition studies, biochemical analysis of affinity-labeled DNA-binding proteins, and the binding of a purified core binding factor, the proteins that bound to the TCR delta core site were indistinguishable from those that bound to the murine leukemia virus core site. These data argue that DNA-binding proteins that interact with the core site of murine leukemia virus long terminal repeats and contribute to viral T-cell tropism also play an essential role in the T-cell-specific expression of cellular genes.

a1/EBP: a leucine zipper protein that binds CCAAT/enhancer elements in the avian leukosis virus long terminal repeat enhancer

Avian leukosis virus (ALV) induces bursal lymphoma in chickens after integration of proviral long terminal repeat (LTR) enhancer sequences next to the c-myc proto-oncogene. Labile LTR-binding proteins appear to be essential for c-myc hyperexpression, since both LTR-enhanced transcription and the activities of LTR-binding proteins are specifically decreased after inhibition of protein synthesis (A. Ruddell, M. Linial, W. Schubach, and M. Groudine, J. Virol. 62:2728-2735, 1988). This lability is restricted to hematopoietic cells from ALV-susceptible chicken strains, suggesting that the labile proteins play an important role in lymphomagenesis. The major labile activity binding to the a1 LTR region (A. Ruddell, M. Linial, and M. Groudine, Mol. Cell. Biol. 12:5660-5668, 1989) was purified from bursal lymphoma cells by conventional and oligonucleotide affinity chromatography, yielding three proteins of 35, 40, and 42 kDa. More than one of these species binds the a1 LTR region, as judged by gel shift analysis. A gene encoding an a1-binding protein (designated a1/EBP) was cloned by screening a bursal lymphoma cDNA library for fusion proteins binding the a1 LTR site. DNase I footprinting and gel shift assays indicate that the a1/EBP fusion protein binds multiple LTR CCAAT/enhancer elements in a pattern similar to that of the purified B-cell protein. DNA sequence analysis shows that this 2.2-kb cDNA encodes a 209-amino-acid open reading frame containing carboxy-terminal basic and leucine zipper motifs, indicating that a1/EBP encodes a novel member of the leucine zipper family of transcription factors.

A phorbol ester response element within the human T-cell receptor beta-chain enhancer

The activity of the T-cell receptor beta-chain gene enhancer is increased by activators of the protein kinase C pathway during T-cell activation. Analysis of mutant enhancer constructs identified two elements, beta E2 and beta E3, conferring phorbol ester inducibility. Multimerized beta E2 acted in isolation as a phorbol ester-responsive element. Both beta E2 and beta E3, which contain a consensus Ets-binding site, were shown to bind directly to the product of the c-ets-1 protooncogene. Both regions also bound a second factor, core-binding factor. Mutation of the beta E2 Ets site abolished the inducibility of the beta E2 multimer. beta E2 and beta E3 Ets site mutations also profoundly affected activity and inducibility of the enhancer. In contrast, enhancer activity but not its inducibility was affected by mutation of the beta E2 core-binding factor site. Cotransfection studies showed that Ets-1 specifically repressed activity of the multimerized beta E2 element and the complete T-cell receptor beta-chain enhancer. These data show that the T-cell receptor beta-chain enhancer responds to protein kinase C-mediated activation signals via a functional domain, composed of two elements, which contains binding sites for Ets transcription factors and which is negatively regulated by Ets-1.

Phenotypes of murine leukemia virus-induced tumors: influence of 3' viral coding sequences

Murine leukemia viruses (MuLVs) induce leukemias and lymphomas in mice. We have used fluorescence-activated cell sorter analysis to determine the hematopoietic phenotypes of tumor cells induced by a number of MuLVs. Tumor cells induced by ecotropic Moloney, amphotropic 4070A, and 10A1 MuLVs and by two chimeric MuLVs, Mo(4070A) and Mo(10A1), were examined with antibodies to 13 lineage-specific cell surface markers found on myeloid cell, T-cell, and B-cell lineages. The chimeric Mo(4070A) and Mo(10A1) MuLVs, consisting of Moloney MuLV with the carboxy half of the Pol region and nearly all of the Env region of 4070A and 10A1, respectively, were constructed to examine the possible influence of these sequences on Moloney MuLV-induced tumor cell phenotypes. In some instances, these phenotypic analyses were supplemented by Southern blot analysis for lymphoid cell-specific genomic DNA rearrangements at the immunoglobulin heavy-chain, the T-cell receptor gamma, and the T-cell receptor beta loci. The results of our analysis showed that Moloney MuLV, 4070A, Mo(4070A), and Mo(10A1) induced mostly T-cell tumors. Moloney MuLV and Mo(4070A) induced a wide variety of T-cell phenotypes, ranging from immature to mature phenotypes, while 4070A induced mostly prothymocyte and double-negative (CD4- CD8-) T-cell tumors. The tumor phenotypes obtained with 10A1 and Mo(10A1) were each less variable than those obtained with the other MuLVs tested. 10A1 uniformly induced a tumor consisting of lineage marker-negative cells that lack lymphoid cell-specific DNA rearrangements and histologically appear to be early undifferentiated erythroid cell-like precursors. The Mo(10A1) chimera consistently induced an intermediate T-cell tumor. The chimeric constructions demonstrated that while 4070A 3' pol and env sequences apparently did not influence the observed tumor cell phenotypes, the 10A1 half of pol and env had a strong effect on the phenotypes induced by Mo(10A1) that resulted in a phenotypic consistency not seen with other viruses. This result implicates 10A1 env in an active role in the tumorigenic process.

Molecular cloning of osteoma-inducing replication-competent murine leukemia viruses from the RFB osteoma virus stock

We report the molecular cloning of two replication-competent osteoma-inducing murine leukemia viruses from the RFB osteoma virus stock (M. P. Finkel, C. A. Reilly, Jr., B. O. Biskis, and I. L. Greco, p. 353-366, in C. H. G. Price and F. G. M. Ross, ed., Bone--Certain Aspects of Neoplasia, 1973). Like the original RFB osteoma virus stock, viruses derived from the molecular RFB clones induced multiple osteomas in mice of the CBA/Ca strain. The cloned RFB viruses were indistinguishable by restriction enzyme analysis and by nucleotide sequence analysis of their long-terminal-repeat regions and showed close relatedness to the Akv murine leukemia virus.

Differential expression of subspecies of polyomavirus and murine leukemia virus enhancer core binding protein, PEBP2, in various hematopoietic cells

The core sequence of the enhancer of murine leukemia virus (MuLV) long terminal repeat is highly conserved in a large number of MuLV strains and appears to play an essential role when SL3-3 or Moloney strains induce T cell lymphoma in mice. We found by using the electrophoretic mobility shift assay that a polyomavirus enhancer core-binding protein, PEBP2, bound to this core motif of MuLV. We also noted that PEBP2 in several hematopoietic cell lines derived from B lymphocyte, macrophage and myelocyte lineages migrated significantly faster than the authentic PEBP2 detected in NIH3T3 fibroblasts. Interestingly, PEBP2 detected in the cell lines of T lymphocyte lineage appeared to contain both types, which were indistinguishable in electrophoretic mobility from those of NIH3T3 and of B lymphocyte, macrophage and myelocyte lineages. The treatment of the nuclear extract containing PEBP2 with phosphatase generated PEBP3, which is a subcomponent of PEBP2 and retained the same DNA-binding specificity as PEBP2. The altered mobility of hematopoietic cell-derived or T lymphocyte-derived PEBP2 was found to be due to the alteration of the mobility of PEBP3. Based on the distinct mobility of PEBP2/3 of T lymphocytes from those of other hematopoietic cells, we discuss the implication of PEBP2 in MuLV-induced T cell leukemia and T cell-specific gene expression.

Insertional mutagenesis of flvi-2 in tumors induced by infection with LC-FeLV, a myc-containing strain of feline leukemia virus

LC-FeLV is a myc-containing strain of feline leukemia virus (FeLV) which exhibits only partial transforming activity in vitro and in vivo. LC-FeLV infection in kittens may induce, but does not necessarily induce, thymic lymphosarcoma in viremic animals after a short latency. These observations suggest that infection with LC-FeLV is not sufficient to induce complete transformation and that another genetic event(s) is required. One possibility for such an event is that the integrating provirus acts as an insertional mutagen and thereby disrupts the structure or function of another proto-oncogene. Using a strategy of transposon tagging, this possibility was examined in eight feline T-cell lymphosarcomas, including four induced by experimental infection with LC-FeLV, three induced by natural infection with FeLV, and one FeLV-negative tumor. The analysis demonstrated one locus, termed flvi-2, to be structurally altered in six of the tumors examined, including three induced by LC-FeLV and three in which no activated myc oncogene is apparent. Inverse polymerase chain reaction was used to demonstrate the presence and transcriptional orientation of proviruses integrated at flvi-2 in five of these tumors. The flvi-2 locus does not hybridize to cloned probes representing 21 previously identified proto-oncogenes or common domains of retroviral integration. Thus, the data suggest that interruption of the flvi-2 locus cooperates with the myc oncogene in the induction of T-cell lymphomas by LC-FeLV; indeed, the observations indicate that the insertional mutagenesis of flvi-2 plays a role in T-cell lymphomagenesis even in the absence of feline v-myc.

Purification of core-binding factor, a protein that binds the conserved core site in murine leukemia virus enhancers

The Moloney murine leukemia virus causes thymic leukemias when injected into newborn mice. A major genetic determinant of the thymic disease specificity of the Moloney virus genetically maps to two protein binding sites in the Moloney virus enhancer, the leukemia virus factor b site and the adjacent core site. Point mutations introduced into either of these sites significantly shifts the disease specificity of the Moloney virus from thymic leukemia to erythroleukemia (N. A. Speck, B. Renjifo, E. Golemis, T. Frederickson, J. Hartley, and N. Hopkins, Genes Dev. 4:233-242, 1990). We have purified several polypeptides that bind to the core site in the Moloney virus enhancer. These proteins were purified from calf thymus nuclear extracts by selective pH denaturation, followed by chromatography on heparin-Sepharose, nonspecific double-stranded DNA-cellulose, and core oligonucleotide-coupled affinity columns. We have achieved greater than 13,000-fold purification of the core-binding factors (CBFs), with an overall yield of approximately 19%. Analysis of purified protein fractions by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis reveals more than 10 polypeptides. Each of the polypeptides was recovered from an SDS-polyacrylamide gel, and those in the molecular size range of 19 to 35 kDa were demonstrated to have core-binding activity. The purified CBFs were shown by DNase I footprint analyses to bind the core site in the Moloney virus enhancer specifically, and also to core motifs in the enhancers from a simian immunodeficiency virus, the immunoglobulin mu chain, and T-cell receptor gamma-chain genes.

Genetic differences accounting for evolution and pathogenicity of simian immunodeficiency virus from a sooty mangabey monkey after cross-species transmission to a pig-tailed macaque

We determined the nucleotide sequences of two related isolates of simian immunodeficiency virus from the sooty mangabey monkey (SIVsmm) that exhibit dramatic differences in virulence. These isolates are separated by one experimental cross-species transmission, from sooty mangabey to pig-tailed macaque. The parental virus (SIVsmm9), nonpathogenic in the original host (sooty mangabeys), causes a chronic AIDS-like disease in macaques. In contrast, the variant virus (SIVsmmPBj14) induces an acute lethal disease in various macaque species and is also pathogenic for sooty mangabeys. The combination of necessary and sufficient mutations that determined the acutely lethal phenotype on the SIVsmm9 genetic background is included within a maximal set of 57 point mutations, plus two insertions located in the long terminal repeat (22 bp spanning an NF-kappa B-like enhancer element) and in the surface envelope glycoprotein (5 amino acids). Comparisons of synonymous and nonsynonymous nucleotide substitutions in the genome of SIVsmm indicated that selective pressures, probably due to the host immune response, favored amino acid changes in the envelope. This immunoevolutionary mechanism could explain the increase in diversity and the apparition of new virulent phenotypes after cross-species transmission.

Interaction of SEWA sarcoma cell proteins with the intracisternal A-type particle long terminal repeat DNA sequence

Intracisternal A-type particle (IAP) transcripts are endogenous retrovirus-like sequences expressed during specific stages of normal development and in a variety of murine tumors. In this study, we have analyzed two cell lines derived originally from the SEWA murine osteosarcoma and grown either as ascites or as solid tumors, for proteins that might regulate IAP expression. We found that subline AA7-NA, originally derived from the ascites tumor, expressed about five times more IAP RNA than the AS12-AD subline, which was derived from a solid tumor. In view of this finding, we examined the binding of cellular proteins from the two cell lines to the 5' end of an IAP long terminal repeat sequence. Gel retardation assays of DNA-protein complexes and DNase I footprinting assays identified several DNA sequences within the long terminal repeat fragment that were protected by protein extracts from both SEWA sublines. Gel retardation assays using specific synthetic oligonucleotide sequences that correspond to two of these protected regions revealed different patterns of DNA-protein complexes with extracts from the two SEWA sublines. These data suggest that expression of IAP sequences is regulated by complex mechanisms involving several proteins that appear to differ between the two sublines.

Purification of core-binding factor, a protein that binds the conserved core site in murine leukemia virus enhancers

The Moloney murine leukemia virus causes thymic leukemias when injected into newborn mice. A major genetic determinant of the thymic disease specificity of the Moloney virus genetically maps to two protein binding sites in the Moloney virus enhancer, the leukemia virus factor b site and the adjacent core site. Point mutations introduced into either of these sites significantly shifts the disease specificity of the Moloney virus from thymic leukemia to erythroleukemia (N. A. Speck, B. Renjifo, E. Golemis, T. Frederickson, J. Hartley, and N. Hopkins, Genes Dev. 4:233-242, 1990). We have purified several polypeptides that bind to the core site in the Moloney virus enhancer. These proteins were purified from calf thymus nuclear extracts by selective pH denaturation, followed by chromatography on heparin-Sepharose, nonspecific double-stranded DNA-cellulose, and core oligonucleotide-coupled affinity columns. We have achieved greater than 13,000-fold purification of the core-binding factors (CBFs), with an overall yield of approximately 19%. Analysis of purified protein fractions by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis reveals more than 10 polypeptides. Each of the polypeptides was recovered from an SDS-polyacrylamide gel, and those in the molecular size range of 19 to 35 kDa were demonstrated to have core-binding activity. The purified CBFs were shown by DNase I footprint analyses to bind the core site in the Moloney virus enhancer specifically, and also to core motifs in the enhancers from a simian immunodeficiency virus, the immunoglobulin mu chain, and T-cell receptor gamma-chain genes.

Helix-loop-helix transcriptional activators bind to a sequence in glucocorticoid response elements of retrovirus enhancers

A family of nuclear proteins, designated SL3-3 enhancer factors 2 (SEF2), were found to interact with an Ephrussi box-like motif within the glucocorticoid response element in the enhancer of the murine leukemia virus SL3-3. Mutation of the DNA sequence decreased the basal enhancer activity in various cell lines. The important nucleotides for binding of SEF2 are conserved in most type C retroviruses. Various cell types displayed differences both in the sets of SEF2-DNA complexes formed and in their amounts. A cDNA which encoded a protein that interacted specifically with the SEF2-binding sequence was isolated from human thymocytes. The nucleotide sequence specificity of the recombinant protein, expressed in Escherichia coli, corresponded to that of at least one of the nuclear SEF2 proteins. Sequence analysis of the cDNA revealed that it belongs to the basic helix-loop-helix class of DNA-binding proteins. Several mRNA transcripts of different sizes were identified. Molecular analysis of cDNA clones revealed multiple related mRNA species containing alternative coding regions, which are most probably a result of differential splicing.

Deletion of a GC-rich region flanking the enhancer element within the long terminal repeat sequences alters the disease specificity of Moloney murine leukemia virus

Moloney murine leukemia virus (M-MuLV) is a replication-competent retrovirus which induces T-lymphoblastic lymphoma 2 to 4 months after inoculation. Enhancer sequences in the U3 region of the M-MuLV long terminal repeat, primarily the 75-bp tandem repeats, strongly influence the disease specificity and latency of M-MuLV. We investigated the role of GC-rich sequences downstream of the tandem repeats in the disease specificity of M-MuLV. A recombinant M-MuLV lacking 23 bases of a GC-rich sequence (-174 to -151), Delta 27A M-MuLV, was tested for pathogenesis in neonatal NIH Swiss mice. Delta 27A M-MuLV induced disease with a longer latency than did M-MuLV (7 versus 3 months) in greater than 85% of inoculated mice. More interestingly, this virus showed an expanded repertoire of hematopoietic diseases. Molecular analyses and histopathologic examinations indicated that while 39% of mice inoculated with Delta 27A M-MuLV developed T-cell lymphoblastic lymphoma typical of wild-type M-MuLV, the majority developed acute myeloid leukemia, erythroleukemia, or B-cell lymphoma. Viral DNA corresponding to Delta 27A M-MuLV was detectable in most of the tumors analyzed. These findings indicate that the GC-rich region significantly influences the disease specificity and latency of M-MuLV.

A single point mutation activates the Moloney murine leukemia virus long terminal repeat in embryonal stem cells

The expression of Moloney murine leukemia virus (Mo-MuLV) and Mo-MuLV-derived vectors is restricted in undifferentiated mouse embryonal carcinoma and embryonal stem (ES) cells. We have previously described the isolation of retroviral mutants with host range properties expanded to embryonal cell lines. One of these mutants, the murine embryonic stem cell virus (MESV), is expressed in ES cell lines. Expression of MESV in these cells relies on DNA sequence motifs within the enhancer region of the viral long terminal repeat (LTR). Here we show that replacement of the Mo-MuLV enhancer region by sequences derived from the MESV LTR results in the activation of the Mo-MuLV LTR in ES cells. The enhancer regions of MESV and Mo-MuLV differ by seven point mutations. Of these, a single point mutation at position -166 is sufficient to activate the Mo-MuLV LTR and to confer enhancer-dependent expression to Mo-MuLV-derived retroviral vectors in ES cells. This point mutation creates a recognition site for a sequence-specific DNA-binding factor present in nuclear extracts of ES cells. This factor was found by functional assays to be the murine equivalent to human Sp1.

SL3-3 enhancer factor 1 transcriptional activators are required for tumor formation by SL3-3 murine leukemia virus

The transcriptional enhancers of retroviruses that lack an oncogene are important determinants of their oncogenicity. However, no specific cellular transcriptional activator has yet been found to determine the oncogenicity for any of these viruses. The SL3-3 enhancer factor 1 (SEF1) cellular transcriptional activators are expressed preferentially in T lymphocytes. In the SL3-3 murine leukemia virus enhancer, two different sequences can bind SEF1 activators. We show that mutation of the SEF1 binding sites disrupts the disease potential of SL3-3 murine leukemia virus, implying that SEF1 transcriptional activators are required for tumor induction by SL3-3. The SEF1 site mutations did not appear to affect the pathogenicity of SL3-3 by impairment of virus multiplication, but rather by a specific defect in the ability of neoplastic transformation.