Identification of DNA Fragments Carrying Ecotropic Proviruses of AKR Mice

Please download to get full document.

View again

of 5
12 views
PDF
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
Identification of DNA Fragments Carrying Ecotropic Proviruses of AKR Mice
Document Share
Document Tags
Document Transcript
  Proc. Natl. Acad. Sci. USA Vol.76, No. 9, pp. 4554-4558, September 1979 Genetics Identification of DNA fragments carrying ecotropic proviruses of AKR mice (Southern blotting/endogenousviruses/murine leukemia virus/congenicmice) DAVID STEFFEN*, STEPHANIE BIRD*, WALLACE P. ROWEt, AND ROBERT A. WEINBERG *Center for Cancer Research and Department of Biology,Massachusetts Institute of Technology, Cambridge, Massachusetts02139; and tLaboratory of ViralDiseases, National Institute of Allergy and Infectious Diseases, Bethesda, Maryland 20205 Contributed by Wallace P. Rowe, June 7, 1979 ABSTRACT The provirusesof the N-tropic, ecotropic virus (AKV) of AKR mice (Akv-1, Akv-2) have been studied by the Southern gel-filter transfer technique. These proviruses can be detected by cleavage of cell DNA by BamHI endonuclease,which- yieldscharacteristic subgenomic DNA fragments upon cleavage of this typeof provirus. Provirusesintegrated into different sites in the mouse genome can be resolved with EcoRI endonuclease, which does not cleave the AKV proviruses. Use ofcongenic and backcrossed mice and a radioactive DNA probe enriched for AKV sequences has allowed identificationof the EcoRI fragments carrying the proviruses of the genetically defined Akv-1 and Akv-2 loci. Novel proviruses introduced by superinfection of cultured AKR cells with AKV and present in leukemic cells from AKR mice have also been identified. Comparison of substrainsof AKR mice indicates some hetero- geneity in theirspectra of proviruses. The AKR strain of mouse is characterized by a life-longex-pression of an ecotropic N-tropic murine leukemia virus, AKV, and a 70-95 incidence of leukemia  1, 2 . Both ofthese characteristicsare the result of dominant alleles present at two loci, Akv-1 and Akv-2  3 . Congenicmice havebeen constructed that possess either the Akv-1 or Akv-2 allele in the genetic background of the NIH/Swiss strain of mouse  4 , a strain which otherwise lacks such alleles. Hybridization of DNA from such congenic mice to an AKVcDNA probe demonstrated that the Akv-1 allele segregates with and presumably represents a provirus for AKV. The allele at Akv-2 is associated with a sec- ond,apparently identical  5 , provirus. The provirusespresent at Akv-1 and Akv-2 arenot necessarily expressed. Embryonic AKR mice and many cell lines derived from AKR embryos do not produce virus. In vitro, expression of virus occurs at a low frequency spontaneously and at a much higher frequency upon induction with BrdUrd  6 . Because AKR cells are sensitiveto exogenous infection with AKV, such transient induction of virus expressionleads to horizontal in- fection of the entire culture and subsequent continuous virus production. In vivo, expression of virus occurs fromsome cells in all postnatal AKR mice. The in vivo expression of virus is probably also the result of transient induction of virus by a few cells followed by horizontal infection of other cells in the an- imal. The number of AKV provirusespresent in virus-negative, virus-producing, and leukemic AKR cells has been estimated by liquid hybridization  7,8 . This technique allowsrelatively accuratequantitationof the amount of AKV sequences present in a given DNA sample. It provides no information, however, about the organizationof these sequences. Recently, the Southern gel-filtertransfer technique  9 has been used suc- cessfully to analyze single integrated murine leukemia virus (MuLV) proviruses in rodent cells  10 . We apply this technique here to identify the Akv-1 and Akv-2 genetic loci with specific DNA fragments produced by restriction endonucleases, to identify specific restriction endonuclease fragments as markers for the presence of the AKV genome in different genetic backgrounds, and to identify newly introduced AKV and AKV-related proviruses in the DNAs of AKV-infected and leukemic AKR cells. MATERIALS AND METHODS Mice. AKR/N and NFS/N mice were obtained from the mouse colony at the National Institute of Health. Thehomo- zygous congenic mice containing Akv-1 and Akv-2 in an NIH/Swissbackground were constructed in one of our labo- ratories (W.P.R.) as has been described  4 . The NFS/N mice heterozygous for Akv-1 used for thesegregationanalysis were derived by six generations of backcrossingAkv-1 from an NIH/Swiss Akv-1 congenic into NFS/N mice. AKR/J mice used for the studiesof tumor DNAs were obtained from Jackson Laboratory.Virus and Cell Lines. Isolation of the virus coded for by the Akv-1 locus has been described  5 . Akv-1 was grown in an NIH 3T3 cell line selected for its ability to grow MuLVs. These cells were infected and cloned, and a clone producing a high titer of virus was used as a sourceof virus. Production and purifi- cation of virus has been described  10 . Isolation of the AKR2B cell line has been described  11 . The AKV-infected clones were obtained as follows. Supernatant fluids from a culture of AKR2B cells that had spontaneously begun virus production were passed ontoa non-virus-producing cultureof AKR2B. These infected cells were carried for several months and then cloned at 0.16 cells perwell in microtiter wells. Clones were screened for virus production. Isolation of DNA. Extraction of DNA from cells and animal tissues (10) has been described. A simplified procedurewas used for purification of much of the DNA used here. The DNA was extracted once or twice with 1 volof water-saturated phenol (pH 8) and once with 1 volof 96 CHC13/4 isoamyl alcohol (vol/vol). Isopropyl alcohol (2.5vol) was added at 250C, the DNA was mixed until a white clump resulted, and the clump was removed and redissolved in 10 mM TrisIHCl, pH 8/1 mM EDTA. This DNA was cleaved with restriction endonucleases as described (10) by using in each case the buffer recommended by the manufacturer of the enzyme. All restriction enzymes were obtained from New England Biolabs,except for EcoRI,which was obtained from Boehringer Mannheim. Completeness of digestion was monitored as described(10) or by adding X Abbreviations: AKV, virus specified by the Akv-1 or Akv-2 allele of AKR mice; MuLV, murine leukemia virus; NaDodSO4, sodium do- decyl sufate; NaCI/Cit, 0.15 M NaCl/0.015 sodium citrate at pH 7.4. 4554 The publication costs of this article were defrayed in part by page charge payment. This article must therefore by hereby marked  ad- vertisement in accordance with 18 U. S. C. §1734 solely to indicate this fact.  Proc. Natl. Acad. Sci. USA 76 (1979) 4555 DNA to an aliquot of the reaction and examining the A DNA for completeness of digestion by gel electrophoregis.. Gel Electrophoresis of DNAs and Hybridization. Isolation of RNA from virions has been described  12 . cDNA was syn- thesized from purified RNA by useof avianmyeloblastosis virus reversetranscriptase (unpublished data). The cDNA had a specific activity of 2 X 108 to 109 dpm/,gg. This cDNA was used for Southern hybridization directlyor after prehybridization to Moloney MuLV RNA as follows: 70S Moloney MuLVRNA at 1000 Ag/ml and cDNA at 1 Atg/ml in 6X NaCl/Cit, pH 7.4 (1X is 0.15 M NaCl/0.015 M sodium citrate) and 0.1 sodium dodecyl sulfate (NaDodSO4) were hybridized for 60 min at 670C and thenused for Southern hybridization. Gels were run and the DNA was transferred to the filters as described  13 . Hybridization to the DNA on the filters was modified as follows: the filter was soaked before hybridization for 2hr at 670C in 6X NaCI/Cit, lOX Denhardt s reagent (1X is 0.02 each of bovine serum albumin, Ficoll, and polyvinyl- pyrrolidone), 1 00 Mg of boiled salmon DNA per ml, and 0.1 NaDodSO4.The filter was rinsed with 2X NaCI/Cit and dried at room temperature for severalhours. Hybridizations were performed at 670C for 16-24 hr. The hybridization reaction contained 6X NaCI/Cit, lOX Denhardt s reagent, 100 Mg of boiled salmon DNA perml, 0. 1 NaDodSO4, 10 Mg of poly(A) per ml, and 2.5 X 106 dpm of AKV cDNA per ml. Prehybrid- ized cDNA was used at twice this concentration. Afterhybridization, the filter was rinsedseveral times briefly with 2X NaCl/Cit at room temperature and then washed threetimes for 60 min and twice for 30 min at 670C with 100-200 ml of 3X NaCI/Cit, lOX Denhardt s reagent, 100 Mg of boiledherring DNA per ml, and 0 1 NaDodSO4. If the filter had been hybridized with cDNA prehybridized to Moloney MuLV RNA, the filter was then rinsed several times at room temperature with 2X NaCI/Cit, incubated 30-45 min in 2X NaCI/Cit, 0.5 Mig of boiled RNase A per ml, andwashed twice for 60 min and once for 30 min at 670C with 100 ml of 0.3X NaCl/Cit and 0.1 NaDodSO4. RESULTS Use of Endonucleases BamHI and EcoRI in Analysis of AKV Proviruses. As the first step in our analysisof the AKV proviruses of AKR mice, we determined the sites within the AKV genome at which BamHI cleaves. The sites of cleavage by this enzyme are shown in Fig. 1. Data supporting this map will be reported elsewhere. Because BamHI cleavesthe AKV genome at three sites, cleavage of an AKV provirus integrated into the genome of a mouse cell will result in two viral fragments linked to thead- jacent cellular sequences and two internal fragments containing only viral sequences. The sizes ofthese two internal fragments will be invariant and will not depend on the sequences into which the provirus happens to be integrated. These two internal fragments can be used to screen for the presence or absence of an AKV provirus in a given sample of DNA. DNA samples were screened for the presence of AKV pro- viruses by cleaving the DNA with BamHI and detecting the 5 1.9 1.9 3.0 1.9 A B C D FIG. 1. Sites on the AKV genome cleaved by BamHI. Arrows indicate sites on the genome cleaved by BamHI. Numbers above the line indicate the sizes  in kilobases)of the fragments. The 3 and 5 orientationof the map is defined with respect to the viral RNA. The fragments are labeled A-D as indicated below the line. Fragments A, B, and D are coincidentallythe same size and thus migrate the same distance during gel electrophoresis (Fig. 2 . resulting fragments by the Southern procedure. When DNA -fi-fim an AKR/N mouse was analyzed, the complex pattern of Fig. 2, lane b was seen. The large number of fragments hybri-dizing to the AKV probe represent diverse, endogenous, ge-netically acquired sequences with homology to the AKV probe used fordetection. These endogenous sequences are observed in many strains of mice, some of which lack AKV-type provi- ruses  14 . Among the fragments seen in Fig. 2, lane b, two of them (termed B and C) were the same size as BamHI fragments resulting from digestionof theunintegrated AKV genome (Fig. 2, lane a . We tentatively conclude that these result from the provirus at Akv-1 and Akv-2 in the AKR/N mouse DNA. The NIH-Swiss strain of mouse does not contain an AKV provirus (14). Analysis of BamHI-cleavedNIH/Swiss DNA again reveals many fragments with homology to the AKV probe (Fig. 2, lane d) but none of the size of the AKV internal frag- ments B and C. If DNA from a clone of AKV-infected NIH/ Swiss mouse cells was analyzed (lane c , the B and C fragments a b C d 23in 9 8 9.8 -_ 4.5- rnbm -.~ C 2.5i . 2 2__ B FIG. 2. Identification of AKV specific BamHI fragments in mouse DNAs. BamHI-digested DNAs were analyzed by the Southern procedure. The gel was run for 16 hr at 1.2 V/cm. The molecular sizes indicated to the left of the figure in kilobases were determined with HindIII-digested X DNA. The BamHI B and C fragments (seeFig. 1 are shown to the right of thefigure. Lane a, unintegrated AKV linear DNA; lane b, AKR2B DNA; lane c, AKV-infected NIH 3T3 cell DNA; lane d, NIH/Swiss mouse DNA. All lanes were hybridized with unselected AKV cDNA. Genetics: Steffen et al.  Proc. Natl. Acad. Sci. USA 76 (1979) were observed. Thus, the B and C fragments represent reliable markers for thepresence of the AKV genome amid the com- plex, endogenousbackground of'NIH/Swiss mouse DNA. Unlike BamHI, EcoRI does not cleave AKV proviruses to produce fragments of characteristic size. Rather, because EcoRI does not cleave the AKV genome (datanot shown), cleavage of DNAs containing AKV proviruses with EcoRI willresultin intact AKV proviruses flanked on eitherside by cellular se- quences. The size of the DNA fragment containing an AKV provirus will depend on the  distance between the viral se- quences and the nearest EcoRI site in the cellular DNA. Con- sequently, proviruses integrated into different sites in the mouse cell genome will in general reside in different sized EcoRI fragments and therefore canbe resolved by gelelectrophoresis. All EcoRI fragments containinga provirus must be larger than 8800 base pairs (8.8 kilobases), the size of the AKV provirus. Development of a Probe Selective for AKV. Analysis of EcoRI-cleaved AKR/N mouse DNA by the Southern proceduredescribed above revealed many hybridizing DNA fragments (Fig. 3, lane a . Most of these fragments do notrepresent AKV proviruses, but rather other endogenous sequenceswith some homology to the AKV probe. We were able to use a minor variationofa technique developed byLee Bachelor of theSalk Institute (personal communication) to remove most of the se- quences from the AKV probe that hybridize to these non-AKV endogenous sequences. We annealed the AKVcDNA probe to a 1000-fold excess of Moloney MuLVRNA before incubating the probe with theSouthern filter containing mouse DNA. The sequences in the probe with homology to the MuLVRNA become unavailable for hybridization to DNA on the filter. Hybridization of EcoRI-cleaved AKR/N mouse DNA with this preselected probe (Fig. 3, lane c) revealed a much simpler pattern of DNA f rag- ments than was observedwith the unselected probe (lane a . Some of the fragments detected must still represent sequences other than AKV proviruses, as they are smaller insize than the intact AKV genome. There are, in fact, only three fragments large enough (>8.8 kilobases) to contain a complete AKV ge- nome. Two of these must represent the AKV proviruses at the Akv-1 and Akv-2 loci. Use of Congenic Mice to Identify Akv 1 and Akv-2. We used congenic NIH/Swissmice carrying Akv 1 and Akv-2 to identify the EcoRI fragments on which these alleles reside. The congenic mice contain either the Akv-l or the Akv-2 allele amid thegenetic background of NIH/Swiss mice. The respective proviruses were identified as EcoRI fragments absent fromNIH/Swissmouse DNA, but present both in the congenic mouse DNA and the AKR/N mouse DNA. Comparison of the DNAs of these mice is shown in Fig. 3. DNA from the NIH/ Swiss Akv-2 congenic mouse is displayed in lane b. This DNA exhibited two fragments (one at 23 and one at 9.8 kilobases)that were absent from the NIH/Swiss DNA (lane e andwere large enough to contain a complete AKV provirus. Only one of these was present in the DNA of the AKR/N strain of mouse. This fragment must contain the sequence of the Akv-2 allele. The fragment absent from the AKR/N and NIH/Swiss DNAs but present in the congenic could represent eithervariability in the NIH/Swiss background or integration ofa new provirus during construction of the congenic line. DNA from the NIH/Swiss Akv-1 congenic (laned) contained three fragments that were absent from the NIH/Swiss DNA and that were large enough to contain a complete AKV provi- rus. Only one of these was present in DNA from the AKR mouse and, therefore, we identify this fragment as containingAkv-1. Because this fragmentcomigrated with a group of very high molecularweight fragments, its size was impossible to deter- a b c rj e   kv- 1 -Akv --2 9 8- mi- .i- 8. W 88 qp 40 .W;* RMD*. 6.6 FIG. 3. Identificationof DNA fragments associated with Akv-1 and Akv-2 in EcoRI-digested mouse DNAs. EcoRI-digested DNAs were analyzed by the Southern procedure. The gel was run for 40 hr at 1.2 V/cm. The molecular sizes indicated to the left  in kilobases) were determined with HindIII-cleaved X DNA. The mobilities ofthe fragments associated with Akv-1 and Akv-2 and the mobility esti- mated for an intact, unintegrated AKV genome  8.8 kilobases) are shown to the right. Lane a was hybridized with unselected AKV cDNA; the remaining lanes were hybridized with AKV cDNA prehybridized to Moloney MuLV RNA. DNAs were extracted frommouse livers. Lane a, AKR/N DNA: lane b, NIH/Swiss Akv-2 con- genic DNA; lane c, AKR/N DNA; lane d, NIH/Swiss Akv-1 congenic DNA; lane e, NIH/Swiss DNA  a mouse obtained from the Massa- chusetts Institute of Technology Center for Cancer Research Animal Facility). mine here and its intensity was frequently low due to technical factors in electrophoresis and transfer. To provide further evidence for our identification of .the Akv-1 -containing DNA fragment, we followed thesegregationof the Akv-1 allele in a genetic cross. A mouse with thegenetic background of the NFS/N strain (an inbred strain derived from the NIH/Swiss strain and heterozygous for Akv-1 (Akv-l /-) was crossed with an NFS/N mouse lacking any Akv-1 alleles (-/-). About half the progeny of this cross should inherit the Akv-1 allele (Akv-1/-) while the remaining mice will lack Akv-1 alleles (-/-). Eight suckling mice resulting from this cross were tested for IdUrd-inducible virus. Three micewere virus positive. DNA from all eight micewas cleaved with either BamHI or EcoRI. As shown in Fig. 4, the same three mice  nos. 1,5, and 7 that were positive for virus expression also possessed an AKV provirus, as scored by generation of the B and C AKV marker fragments upon cleavage of their DNA with BamHI. 4556 Genetics: Stef fen et al. 27.5   23-  Proc. Natl. Acad. Sci. USA 76 (1979) 4557   2 3 4 5 6 7 8 a b c de f g h   27.5- ___A v-1 50 A 23 _AAkv 2 23- 4 16 n   9 8 8  .   8 2.5- 2.2- 28- 23- 9.8-   _- B _- Akv- 1 FIG. 4. Segregation of the Akv-1 allele in a backcross. Eight progeny resulting from a cross between an NFS/N mouse congenic and heterozygous for the Akv-1 allele (Akv-1/-) and another lacking any Akv alleles (-/-) were analyzed for infectiousvirus in their tails. A  + below the number of each mouse indicates virus was detected, a  - indicates it was not. Resultsofanalysis of DNA from each mouse are shown in the lanes directly below its number. DNA was extracted from a pool of liver, spleen, andthymus, cleaved with BamHI (Upper) or EcoRI (Lower), and analyzed by the Southern procedure. The gel was run for 16 hr (Upper) or 40 hr (Lower) at 1.2 V/cm. The molecular sizes indicated in kilobases were determined with HindIlI-cleaved A DNA. The mobilities of the BamHIB and C fragments (Fig. 1) or the EcoRI fragment associated with Akv-1 (Fig. 3) are shown to the right.All lanes were hybridized with unselected  KV cDNA. Finally, we note that nos. 1, 5, and 7 were the only mice con- taining the EcoRI-generated DNA fragment identified in Fig. 3 as containing Akv-1, thus confirming the identification. Although the provirus-containing fragments associated with the Akv-1 and Akv-2 alleles were identifiable in the DNA of AKR/N mice, subsequent examination of DNA from two other lines of AKR mice, AKR/Cu (Fig. 5, laned) and AKR/J (lane e) revealed the Akv-1 -associated fragment but failed to show theAkv-2-associated fragment. Noting yet other variabilities in the proviruses of thesethree lines of the AKR strain, we 6.6- 6.6- FIG. 5. Identification of novelproviruses in AKV-infected and leukemic AKR cell DNAs. EcoRI-digested DNAs were analyzed by the Southern procedure. The gels wererun for 40hr at 1.2 V/cm. The molecular sizes indicated in kilobases to the left of eachpanel were determined withuncleaved,HindIll-cleaved, or BamHI/EcoRI- cleaved A DNA. The mobilitiesof the fragments associated with Akv-1 andAkv-2and themobility estimated for an intact, unintegrated AKV genome (8.8 kilobases) are shown.Novel provirus-containing fragments are located between the 9.8- and 23-kilobase markers in lanes b and c and are indicated by small arrows in lanes f-j. Lane a, AKR2B DNA; lane b, clone 1 of AKV-infected AKR2B DNA; lane c, clone 2 of AKV-infected AKR2B DNA; lane d, AKR/Cu liver DNA; lane e, AKR/J embryo DNA; lane f, AKR/J lymphoma 1 DNA; lane g, AKR/J lymphoma 2 DNA; lane h, AKR/J lymphoma 3 DNA; lane i AKR/J lymphoma cell line clone 1 DNA; and lane j, AKR/J lym- phoma cell line clone 2 DNA. All lanes were hybridizedwith AKV cDNA prehybridized to Moloney MuLV RNA. conclude that there is substantial variation between thesethree lines. As a consequence, our identificationofthe Akv-2 associ- ated fragment is applicableonly to the DN of AKR/1N mice. Identification of Novel Proviruses in AKV-Infectedand Leukemic Cells. The Akv loci in cell lines derived from AKR embryos are generally not expressed. Such cells can be infected in vitro with AKV and will thereafter express virus at high levels. Does this infection result in theintroduction of new ac- tive proviruses, or alternatively, in theinduction of expression ofthe preexisting proviruses? It has been reported  8 that, after infectionof AKR embryo fibroblasts and a cell line derived from AKR embryo fibroblasts, no increase in the number of AKV proviruses was detectable by liquid hybridization. The relatively small number of EcoRI fragments detected in AKR DNA with our selected AKV probe encouraged us to re-ex- amine this question by theanalysisdescribedabove. Results of this analysis are shown in Fig. 5. Lane a contains DNA from AKR2B, a cloned line of AKR cells derived from AKR embryo fibroblasts  11 . This line does not express virus. The pattern of EcoRI fragments observed is very similar to that seen when AKR/N mouse DNA was analyzed. AKR2B cells were infected with AKV, and DNA from two subsequently derived clones is displayed in lanes band c ofFig. 5. DNA fragments are present in the infected cell lines but absent from the uninfected parent line. These fragments contain AKV proviruses integrated into new sites in the cellular genome after in vitro infection. Each clone contains several new proviruses and in each clone the proviruses are integrated in different sites. Thus, infectionof AKR fibroblasts in vitro results in integration of AKV proviruses in a number of new sites in the mouse genome. -_-Akv-1 -8.8 Genetics: Steffen et al. *E  |  Proc.Natl. Acad. Sci. USA 76 (1979) DNA from tumor cells of spontaneous leukemias in AKR mice has more AKV sequences than DNA from normal cells, as measured by liquid hybridization  8 . Analysis of such DNAs with restriction endonucleases and procedures similar to those described here, except with an unselected AKV probe, has re- vealednovel DNA fragments in tumor DNAs (15). Using our selected AKV probe, we compared DNAs from leukemic cell DNA with embryo DNA of AKR/J mice (Fig. 5 . As expected, a number of fragments were detected in the tumor DNAs that were absent from the embryo DN (lane e . DNA from each tumor displays a unique pattern of novelfragments. These two observations taken together indicate that the tumors consist largelyof the clonal descendants of one transformed cell  8 . Consistent with this is theobservation that two clonal cell lines derived from the same tumor have the same pattern of novel proviruses (lanes i and j . Not expected is theobservation of fragments in the embryo DNA absent from the tumor DNAs. We do not understand this observation. A number of non-AKVMuLVs, some of which are deriva- tives of AKV, can be isolated from leukemic AKR mice (16-20). Thus, some of the novel fragments observed in tumor DNAs may result from non-AKV proviruses. We point out that some of the novel fragments are smaller than the complete AKV genome, indicating that these fragments cannot result from the integration of an unaltered AKV genome. DISCUSSIONWe have identified DNA fragments containing the genetically acquired ecotropic proviruses corresponding to the Akv-1 and Akv-2 alleles. These alleles, and thusthe associated DNA fragments, are present in AKR mice but absent from NIH/Swiss mice. By introducing the Akv-1 and Akv-2 alleles into the NIH/Swiss genetic background, we were able to correlate physically defined DNA fragments with these genetically de- fined loci. Ihle and Joseph (21) have presentedevidence supporting the existence of a third AKV provirus in AKR mice. Our identifi- eation of two AKV proviruses in AKR mice (Akv-1 and Akv-2) does not bear on existence of additional AKV proviruses in these mice. We note the presence in the DNA of AKR/N mice of a third EcoRIfragment (s 12-15 kilobases) that hybridizesstrongly to our selective probe and that is large enough to contain a third AKV provirus (Fig. 3 . The Akv-1 -associated fragment was identified by two in- dependent procedures, analysis of an NIH/Swiss Akv-1 con- genic mouse (Fig. 3) and segregation of Akv-J in a cross (Fig. 4 . This fragment is present in all three lines of AKR mice ex- amined: AKR/N, AKR/J, and AKR/Cu. The Akv-2-associated fragment, identified by analysis of an NIH/Swiss Akv-2 con- genic mouse (Fig. 3 , is present in the AKR/N line  as well as the cell line AKR2B) but is absent from the AKR/J and AKR/ Cu lines. Thisdifference, and other differences between the bands observed in these lines, demonstrates that there is sub-stantial genetic variability between lines of the AKR strain of mouse. AKR cells can be infected by the AKV specified by their endogenous proviruses. Here we have shown that this infection results in the presence of additional proviruses beyond the ge- netically acquired AKV proviruses. These additional proviruses are difficult to resolve from the background of endogenous sequences in the AKR mouse DNA when an unselected AKV probe is used for detection(data not shown),but are clearly resolved by usingour preselected AKV probe. We have iden- tified such additionalproviruses in AKR fibroblasts infected in vitro, as well as in AKR leukemic cells. Due to the diversityof virus types isolated from leukemic AKR mice, the identityof theproviruses in the leukemic cells is not clear. In fact, the subgenomic size of some of the novel fragments identified in the leukemic cells is inconsistent with their containing unaltered AKR proviruses (Fig. 5 . Itis likely thatvirus production in infected AKR cells results from transcription ofthese newly introduced proviruses and that cells producing virus in adult AKR mice also contain newly introduced proviruses. An un- resolvedquestion concerns the differences between the poorlyexpressed endogenous AKV sequences and the apparently ef- ficiently expressed proviruses introduced via infection. We especially thank Lee Bachelor for describing her technique for producing aspecific hybridization probe and for heradvice in devel- oping it for our system. We also thank Eli Canaani and Stuart Aaronson for communicating to ustheir results before publication. We thank Cliff Tabin, Julie Sexton, and Ann Dannenberg for help in theprep- aration of DNA and with cell culture. We acknowledge the American Cancer Society Grant VC 140 and acoregrant to theMassachusetts Institute of Technology Center for Cancer Research from the National Cancer Institute (CA14051). D.S. was supported by fellowships from the AmericanCancer Society and subsequently theMassachusetts Institute of Technology Weizmann Fellowship Fund. R.A.W. is a Rita Allen Scholar. 1. Rowe, W. P.   Pincus, T. (1972) J. Exp. Med. 135,429-436. 2. Lynch, C. J. (1954) J. Natl. Cancer Inst. 15, 161-176. 3. Rowe, W. P. (1972) J. Exp. Med. 136, 1272-1285. 4. Chattopadhyay, S. K., Rowe, W. P., Teich, N. M.   Lowy, D. R. (1975) Proc. Natl. Acad. Sci. USA 72,906-910. 5. Rommelaere, J., Faller, D. V.   Hopkins, N. (1977) J. Virol. 24, 690-694. 6. Hartley, J. W., Rowe, W. P., Capps, W. I.   Huebner, R. J. (1969) J. Virol. 3,126-132. 7. Cattopadhyay, S. K., Rowe, W. P.   Levine, A. S. (1976) Proc. Natl. Acad. Sci. USA 73,4095-4099. 8. Berns, A.   Jaenisch, R. (1976) Proc. Natl. Acad. Sci. USA 73, 2448-2452. 9. Southern, E. M. (1975) J. Mol. Biol. 98,503-517. 10. Steffen, D.   Weinberg, R.A. (1978) Cell 15, 1003 1010 11. Rowe, W. P., Hartley, J. W., Lander, M. R., Pugh, W. E.   Teich, N. (1971) Virology 46,864-874. 12. Faller, D. V.   Hopkins, N. (1977) J. Virol. 23,188-195. 13. Yoshimura, F. K.   Weinberg, R.A. (1979) Cell 16,323-332. 14. Lowy, D. R., Chattopadhyay, S. K., Teich, N. M., Rowe, W. P.   Levine, A. S. (1974) Proc. Natl. Acad. Sci. USA 71, 3555- 3559. 15. Canaani, E.   Aaronson, S. A. (1979) Proc. Natl. Acad. Sci. USA 76, 1677-1681. 16. Kawashima, K., Ikeda, H., Hartley, J. W., Stockert, E., Rowe, W. P.   Old, L. J. (1976) Proc. Natl. Acad. Sci. USA 73, 4680- 4684. 17. Hartley, J. W., Wolford, N. K., Old, L. J.  Rowe, W. P. (1977) Proc. Natl. Acad. Sci. USA 74,789-792. 18. Hays, E. F.   Vredevoe, D. L. (1977) Cancer Res. 37, 726- 730. 19. Staal, S. P., Hartley, J. W.  Rowe, W. P. (1977) Proc. Natl. Acad. Sci. USA 74,3065-3067. 20. Nowinski, R. C., Hays, E. F., Doyle, T., Linkhart, S., Medeiros, E.   Pickering, R. (1977) Virology 81,363-370. 21. Ihle, J. N.   Joseph, D. R. (1978) Virology 87,287-297. 4558 Genetics: Steffen et al.
Similar documents
View more...
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks