Proc. Nati. Acad. Sci. USA Vol. 73, No. 10, pp. 3705-3709, October 1976

Nucleotide sequences in mouse DNA and RNA specific for Moloney (murine sarcoma virus/RNA tumor virus/nucleic acid hybridization/gene evolution/sarcoma-specific nucleotide sequence) ARTHUR E. FRANKEL AND PETER J. FISCHINGER Laboratory of Viral , National Institute, Bethesda, Maryland 20014 Communicated by Howard M. Temin, July 12,1976

ABSTRACT Complementary DNA (cDNA) synthesized (3). Laboratory strains of oncoviruses do contain information by Moloney murine sarcoma virus (M-MSV) was separated into that is different from the spontaneously released oncovirus of two parts, the first, termed MSV-specific cDNA, composed of the species Of nucleotide sequences found only in M-MSV viral RNA, and the (6, 7). these, the sarcomagenic oncoviruses gen- second, termed MSV-MuLV common cDNA, composed of nu- erally contain both a set of nucleotide sequences shared with cleotide sequences that were found in both M-MSV and murine leukosis- , and another set of sequences that are leukemia virus (MuLV) viral . RNA complementary to the dissimilar from those of other oncoviruses of the species (6-10). MSV-specific cDNA was not found in several other MSV iso- Complementary DNA (cDNA) can be transcribed from sar- lates, nor in ecotropic MuLV, mouse mammary tumor virus, or coma virus RNA by the viral endogenous DNA polymerase. several murine xenotropic oncoviruses. Cellular DNA of several The cDNA represents both the "shared" and the species was examined for the presence of nucleotide sequences "specific" complementary to MSV-specific cDNA. Cells transformed by moieties of the sarcoma virus . We were able to isolate M-MSV did contain MSV-specific cDNA in their DNA. Normal these distinct moieties from the genome of the Moloney isolate mouse cell DNA apparently contained the majority of MSV- of murine sarcoma virus (M-MSV) (9). By hybridizing the total specific nucleotide sequences. Cellular DNA of related species M-MSV-MuLV cDNA with Moloney NMuLV (M-MuLV) viral contained proportionally less MSV-specific cDNA. Hybrids of high-molecular-weight RNA and separating the hybridized MSV-specific cDNA and cellular DNA of related species melted from the unhybridized cDNA by at lower temperatures than hybrids of MSV-specific cDNA and hydroxylapatite chroma- mouse cellular DNA. tography and repeating this procedure using viral RNA from RNA from normal mouse adult or embryonic cells did not the feline leukemia virus pseudotype of M-MSV [MSV(FeLV)], contain detectable nucleotide sequences complementary to we were able to isolate cDNA fractions that represented dif- MSV-specific cDNA. Transformation of cells with M-MSV re- ferent portions of the M-MSV genome, "MSV-specific cDNA" sulted in transcription of RNA hybridizing with MSV-specific and "MSV-MuLV common" cDNA (9). The isolation of cDNA. Methylcholanthrene-induced mouse and cell lines derived from them did not contain RNA complementary MSV-specific cDNA allowed us to approach and answer two to MSV-specific cDNA. Mouse cell lines transformed with avian critical questions: first, whether MSV-specific cDNA was sarcoma virus or Kirsten MSV also did not contain RNA com- present in normal cellular DNA; and, if it was, whether there plementary to the MSV-specific cDNA. RNA homologous to was RNA that was complementary to MSV-specific DNA in MSV-specific nucleotide sequences is measurably present only cells transformed by homologous virus, by different sarcoma in cells transformed by M-MSV and not in cells transformed by viruses, or by chemical or physical agents; other biological or chemical agents that also cause sarcomas. Extensive evidence demonstrates that vertebrate species contain MATERIALS AND METHODS genes related to endogenous RNA tumor virus in the form of The Murine Sarcoma Virus-Specific Complementary DNA that behaves genetically as cellular DNA. The usual RNA DNA. The nature and the specificity of the MSV-specific cDNA form of these viruses is released from cells of many species by derived from the sarcoma + leukemia - (S+L-) M-MSV ge- physical or chemical agents (1-5). Functionally, the oncoviruses nome has been described (9). Two sets of distinct cDNAs were that are spontaneously released from normal cells of a species isolated from MSV: that which was found only in MSV, oper- are nontransforming and can be categorized as leukosis-leu- ationally designated as "MSV-specific cDNA" (26% of the ge- kemia, helper-type viruses (2-5). Although most of these agents nome) and that which was shared by MSV and M-MuLV, la- are not known to be oncogenic, others such as the AKR type of beled "MSV-MuLV common cDNA" (74% of the genome) (9). (MuLV) are capable of causing neoplasia The genetic content of "MSV-specific cDNA" is unknown. The sources and the preparation of Kirsten (Ki) MSV and Harvey Abbreviations: MSV, murine sarcoma virus; M-MSV, Moloney isolate (Ha) MSV RNA and cDNA were previously described (9). of MSV; MuLV, murine leukemia virus; M-MuLV, Moloney isolate DNA-cDNA Hybridizations. We used the single-stranded of MuLV; M-MSV(FeLV), the feline leukemia virus pseudotype of MSV-specific cDNA as a to M-MSV; cDNA, complementary DNA made by the endogenous probe detect MSV-specific nucle- RNA-dependent DNA polymerase (reverse transcriptase) reaction; otide sequences in cell DNA. Unlabeled DNA was extracted MSV-specific cDNA, operational term describing cDNA representing from livers and spleens of BALB/c, NIH Swiss, C3H mice, that specific region of the M-MSV genome not shared with MuLV; Fischer , Sprague-Dawley rats, hamsters, and chickens (13). MSV-MuLV common cDNA, cDNA representing those sequences of DNA was also extracted from monolayers of mouse 3T3FL M-MSV which are also found in M-MuLV; MMTV, mouse mammary cells, M-MSV-transformed 3T3FL S+L- cells, and cat, dog, tumor virus; MuX, murine xenotropic oncovirus; SV40, simian virus and human normal cell lines as well as re-extracted from 40; ASV, avian sarcoma virus; Ki-MSV, Kirsten isolate of MSV; Ha- MSV, Harvey isolate of MSV; S+L-, sarcoma+ leukemia-; MCA, commercial salmon sperm DNA (Worthington). All cellular 3-methylcholanthrene; tm, melting temperature; Cot and Crt, product DNA preparations were sheared by sonication to a size similar of DNA and RNA concentration, respectively, and hybridization to the DNA polymerase products, 4-6 S. Alkaline sucrose gra- time. dients (10-30% sucrose) were centrifuged at 50,000 rpm in a 3705 Downloaded by guest on September 27, 2021 3706 Microbiology: Frankel and Fischinger Proc. Natl. Acad. Sci. USA 73 (1976) Beckman SW56 rotor at 40 for 18 hr. An internal standard of Table 1. Hybridization of oncovirus RNAs with 6S [32P]DNA from simian virus 40 (SV40) was used. Fractions complementary DNA specific for two parts of the (0.25 ml) were precipitated with trichloroacetic acid, and 32p M-MSV genome cpm were measured. Cellular DNA was mixed with 2000 cpm of MSV-specific cDNA (0.1 ng), heat-denatured, adjusted to Percent hybridization of 0.75 M Na+ with NaCl, and incubated at 70°. The cellular DNA MSV-MuLV concentration was 8-9 mg/ml in the hybridization mixture, and MSV-specific common there was a nucleotide molar excess of 1.5 X 107-fold of cellular Viral RNA cDNA cDNA* DNA to MSV-specific cDNA. At various times aliquots were removed, and the single-stranded DNA was separated from M-MSV(M-MuLV) 84 95 double-stranded DNA by hydroxylapatite chromatography at M-MSV(FeLV) 82 90 500 with 0.14 M and 0.3 M phosphate buffers (14). The acid- M-MuLV 0-1 90 precipitable radioactive material and the absorbances at 260 Rauscher MuLV 0 77 nm were determined. Calculations of the product of nucleotide MuX-BALB-2* 0 49 concentration and time (Cot) were carried out as described by MuX-NZB* 0 24 Britten and Kohne and corrected for salt concentration (15). Ki-MSV(Ki-MuLV) 1 48 Thermal Denaturation Profiles of DNA-cDNA Hybrids. Ha-MSV(FeLV) 5 59 Hybrids of MSV-specific cDNA and cellular at a Cot FeLV 1 13 value of 8000-12,000 mole-sec/liter were suspended in 0.12 MMTV 1 7 Avian sarcoma virus 0 2 M sodium phosphate buffer along with 1000 cpm of 32P-labeled MSV-MuLV cDNA hybridized to high-molecular-weight 500-1000 cpm of each cDNA fraction was hybridized in 0.22 M MSV-MuLV RNA serving as internal standard. The 32P-labeled phosphate buffer at 630 to a Crt of at least 10 mole-sec/liter with MSV-MuLV cDNA was prepared identically to the other total virion RNAs from each of the indicated viruses. The extent of starting cDNA except [32P]dTTP was used. It hybridized 97% hybridization was analyzed by hydroxylapatite chromatography. with high-molecular-weight MSV-MuLV RNA with a melting The background was 6% for MSV-specific cDNA and 2% for MSV- temperatures, of 83 ± 10. Hybrids were eluted from hy- MuLV common cDNA and was subtracted from each value. The tm, RNAs from the 10 viruses were tested with homologous comple- droxylapatite with the 0.12 M phosphate buffer at 50 incre- mentary cDNA prepared in the presence of 100 Ag/ml of actino- ments. A second internal standard consisted of mouse globin mycin D and were found to hybridize to about 70% with a 0-10% [32P]cDNA hybridized to mouse cell DNA. This was added to background. NZB liver RNA hybridized about 50% with a Ki- incubation mixtures containing MSV-specific [3H]cDNA, and MuLV cDNA; it was used because no homologous cDNA for murine the thermal elution profile of globin-mouse cell [32P]DNA xenotropic oncovirus (MuX) ",B" type virus could be obtained (19). * Cell RNA from A673 human cells chronically infected with hybrid served as standard for mouse cell DNA hybridizations. BALB-2 virus (MuX "a") from C3H/Hin induced mouse cells was As a third internal standard the absorbances of the 0.12 M used instead of viral RNA (19). Hybridization was taken to a phosphate buffer washes were measured for each 50 increment, final Crt of 2,500 mole-sec/liter. and the thermal denaturation profile of unique sequences was plotted for each cell DNA. MuLV's, both families of murine xenotropic viruses, and the RNA-cDNA Hybridizations. RNA was extracted from cell mouse mammary tumor virus (MMTV) did not hybridize with monolayers or liver homogenates by two techniques. In the first MSV-specific cDNA. The MSV-MuLV common cDNA nu- procedure the cells were treated with 1% sodium dodecyl sul- cleotide sequences did hybridize in some measure with all fate and 0.75 mg/ml Pronase for 30 min in 0.15 M.NaCl/15 strains of murine oncoviruses tested, except MMTV. Compat- mM Tris/1 mM EDTA, pH 7.4, at 370. The suspensions were ible with previous observations Ki-MSV and Ha-MSV isolates extracted three times with 680 phenol, once with room-tem- did not contain MSV-specific cDNA sequences of M-MSV. perature phenol and twice with phenol/chloroform/isoamyl Based on hybridization with MSV-MuLV common cDNA, both alcohol and were reprecipitated with ethanol. This RNA pellet Kirsten (Ki)-MSV and Harvey (Ha)-MSV shared sequences that was suspended in water, dialyzed against water, and lyophi- are also a part of M-MuLV and M-MSV (10, 11). lized. The second technique was Scherrer extraction (16). RNA Presence of M-MSV-Specific Sequences in Normal Cell was prepared by both techniques for each cell line. Two DNA. Several individually prepared MSV-specific cDNA methods of hybridization were used as well. In the first case probes were used to assess the presence of homologous nucle- RNA was dissolved in hybridization buffer (0.9 M NaCl/20 mM otide sequences in normal DNA. Fig. 1 and Table 2 depict the Tris-HCl/1 mM EDTA, pH 7.4) at 10-14 mg/ml, and MSV- reactivity of two different MSV-specific cDNA preparations specific cDNA was added. The hybridization mixtures were with cellular DNAs. The MSV-specific cDNA preparations incubated at 680 for 24-120 hr. The hybridization mixtures failed to hybridize significantly to salmon or chicken DNA. were then fractionated on hydroxylapatite as for the DNA. They hybridized to some degree to hamster DNA [41% relative Hybridizations were also performed by dissolving RNA in 0.5 to 3Bll-IC (MSV-MuLV-infected cell) DNA] and to DNA M NaCl at 0.1-15 mg/ml with MSV-specific cDNA. These (56% relative to 3B11-IC DNA) with low but reproducible Tm mixtures were incubated at 630 for 48 hr and fractionated on values (720 and 770, respectively). However, when the MSV- hydroxylapatite as in the other hybridizations. The Crt curves specific cDNA probes were hybridized with normal mouse liver were calculated based on the method of Leong et al. (17). DNA or normal murine cell line DNA, they hybridized 79% relative to cell DNA from M-MSV, MuLV-producing mouse RESULTS cells with a CotI/2 of approximately 2000 (Fig. 1). DNA from Spectrum of Reactivity of cDNA fractions from M-MSV. normal NIH Swiss mouse liver and 3T3FL cells hybridized 71% The specificity of the two subsets of cDNA from M-MSV was relative to producer cell DNA. Normal mouse cells apparently examined by hybridization with various viral RNAs (Table 1). contained about one copy of the majority of MSV-specific se- Sequences complemenitary to MSV-specific cDNA were present quences per cell genome. The MSV-transformed S+L- murine only in the two pseudotypes of M-MSV. Several ecotropic cell line also demonstrated a high percent hybridization (79%) Downloaded by guest on September 27, 2021 Microbiology: Frankel and Fischinger Proc. Natl. Acad. Sci. USA 73 (1976) 3707 Table 2. Presence of M-MSV-specific nucleotide sequences in different cell DNAs tl% Percent* tmit of hybridi- MSV- zation of specific MSV- C0tl/ cDNA 0 -~~~~~~~~0specific (mol-sec/ hybrid 0 Cell, description cDNA liter) (0C) 3B11-IC, 3T3FL mouse line producing M-MSV and MuLV 100 500 81 P521, CCC cat line yielding MSV(FeLV) and FeLV 100 300 81 C ;t ( he) FG10, S+L-mouse liver 79 1600 81 BALB/c, mouse liver 79 2600 81 FIG. 1. Hybridization of a single Moloney sarcoma virus-specific C3H, mouse liver 79 80 3H probe (XV) to DNAs extracted from cell lines or livers. Absorbance mouse liver 79 81 measurements were used to follow cell DNA seif-reannealing, and C57BL, acid-precipitable cpm were used to follow cDNA hybridization. NIH Swiss, mouse liver 71 2900 81 Background for the cDNA hybridization was 0%. The hybridization 3T3FL, Swiss mouse line 71 81 values to 3B11-IC DNA obtained with the probe used (here MSV- Sprague-Dawley rat liver 56 2000 77 specific cDNA XV) was 54%. Annealing of M-MSV-specific [3HIDNA Fischer rat, liver 56 77 to cell DNA of: *, 3B11-IC mouse cell line derived from Swiss mice Hamster, liver 41 72 infected with and preducing M-MSV (MuLV) and M-MuLV;@*, F49-I, human amnion line 27 500 70 BALB/c liver; A, Fischer rat liver; v, F-49-I human cells; and *, CCC CCC, cat cell line 26 70 normal cat cells is plotted with a least squares computer-determined DK, dog kidney line 21 68 second-order fit. Reannealing of BALB/c liver DNA O is plotted. The Salmon, testes DNA 9 54 unique sequence cellular DNA in the above hybridizations had Cot112 values of 1,000 mole-sec/liter for F49-I cell DNA, 1500 mole-sec/liter Chicken, liver 6 55 for CCC cat cell DNA, 1400 mole-sec/liter for BALB/c mouse DNA, 0, No DNA 0 56 and 2000 mole-sec/liter for 3B11-IC cellular DNA. * Hybridization was performed with 2 x 107-fold excess cellular of cDNA. The actual with a slightly lower Got112 value. The mouse and cat M-MSV DNA and a single preparation MSV-specifi, amount of this probe that hybridized to 3B11-IC cell DNA was producer cell lines hybridized essentially completely with a still 70%. The numbers represent normalized maximal hybridization lower Cot112 value, indicating that multiple MSV-specific percentages obtained at Cot > 12,000. The hybridization reactions cDNA copies were present in their cell DNA. Thermal dena- were carried out as described in the text. The values are nor- turation profiles were done on hydroxylapatite. As seen in Table malized to P521 cell DNA, with which MSV-specific cDNA hybridized to a maximum of 70%. 2 and Fig. 2 the thermal denaturation profiles of the hybrid t Thermal denaturations were performed as described in the text. composed of MSV-specific cDNA and normal cellular DNA of The internal standard consisted of 32P-labeled MSV-MuLV several strains of mice were generally similar. The hybrid of cDNA hybridized to 65S MSV-MuLV RNA. This hybrid had a MSV-specific cDNA with normal cell DNA of five strains of tm of 83 + 10. Additional internal standards included globin mice had a somewhat different slope and was biphasic. The Tm [32P]cDNA hybridized to mouse cell DNA to a Cot of 6,000- values of 80 81° indicated that there were nucleotide sequences 10,000 mole-sec/liter with a tm of 77 i 1° and reannealed cellular DNA which has a tm of 84 i 2° at a Cot of 104 mole-sec/liter. The in normal mouse cells closely related to the MSV-specific nu- tm is the temperature at which half of the cDNA bound 600 is cleotide sequences. The DNA that melted at 50°-60° in 0.12 eluted in 0.12 M phosphate buffer. M phosphate buffer consisted of nonspecific, unstructured hybrids which were also 99% digestible with 51 nuclease. Such hybrids represented material below the criterion (0.75 M Na~l, mals. In contrast, calf, chicken, and salmon DNAs displayed 70° = 0.12 M phosphate buffer, 600). We used only the struc- hybridization values of always less than 10%. Thermal dena- tured portion of the melting profile to establish the tm, i~e., the turation profiles for hybrids of mammalian cell DNAs and midpoint of cDNA elution above 60°. As can be seen from Fig. MSV-specific cDNA generally contained a distinct complement 2, the hamster, cat, dog, and human hybrids did appear to be melting above 600. The reactions of mammalian DNAs suggest biphasic, a portion of the curves being hybrid DNA with that at least some MSV-specific sequences were conserved in structure above background. The broader denaturation profile cellular DNA. The reduced proportion of DNA which hybri- of endogenous rat or hamster sequenc related to MSV-specific dized with the MSV-specific cDNA and the lower tM values sequences suggested further sequence heterogeneity of suggest the sequences have diverged phylogenetically (22). MSV-specific sequences and the endogenous nucleotide se- Presence of MSV-Specific Nucleotide Sequences in Mouse quences in normal cellular DNA. Cell RNA. Because MSV-specific cDNA was found in the DNA Five strains of mice have been tested: C3H, BALB/c, G57, of normal mice, it was of interest to determine whether those NIH Swiss, and outbred Swiss; these yielded similar reactions, sequences were transcribed into RNA of normal embryonic or as shown in Table 2. Rat DNA hybridized 56% and hamster adult mouse cells. Secondly, a number of murine cell lines DNA 41% With the MSV-SPeCifiC CDNA. A limited portion of transformed by different agents were examined to see if they the melting curve of MSV-specific CDNA and dog, cat, or contained RNA that was complementary to MSV-specific human DNAS appeared significant as well. The percent hy- cDNA. Table 3 shows that S+L- mouse cell or mouse cells bridization based on 0.14 M phosphate buffer wash at 500 producing both M-MSV and M-MuLV did contain M-MSV- ranges from 20 to 27% with a 35% error for the above mam- specific nucleotide sequences in their RNA. The Crt values Downloaded by guest on September 27, 2021 3708 Microbiology: Frankel and Fischinger Proc. Natl.-Acad. Sci. USA 73 (1976) Table 3. Presence of M-MSV-specific sequences in 100 different mouse cell RNAs Percent 90 hybridi- Cell, description zation Crt½/, 80s 3B11-IC, 3T3FL yielding M-MSV and M-MuLV 100 102 FG10, M-MSV-transformed 3T3FL line 100 103 70 KB23, Ki-MSV-transformed BALB line 16 > 104-5 B77, ASV-transformed BALB line (21) 12 > 104-5 I- liA81, SV40-transformed BALB line (23) 12 >104-5 60 MC55, BALB/c MCA tumor line* 10 > 104-5 Lai MCA1, MCA tumor in BALB/c mice t 0 > 104-5 I- 0 > 50 F MCA2, MCA tumor in BALB/c micet 104-5 z BALB/3T12, high density line 0 >104-5 wi 3T3FL, Swiss mouse line 0 > 104-5 40 1C-3T3, 3T3FL yielding M-MuLV (10) 0 >104-5 BALB/c embryo, whole embryo 0 > 104-5 NIH embryo, whole embryo 0 > 104-5 30 mlSR-10, revertant of FG10 (12) 5 > 104-5 mlSR-12, revertant of FG10 (12) 0 > 104-5 BALB/c, mouse liver 0 >104-5 20 The actual percent hybridization for 3B11-IC and FG10 cell RNA is 80% for each with the MSV-specific cDNA. MSV(FeLV) viral 10 RNA hybridizes with 80% of the cDNA. A single MSV-specific cDNA preparation with a background of 7% was used. Hybridiza- tion values were corrected for background and are made relative n to hybridization with the MSV(FeLV) RNA. ASV, avian sarcoma 50 60 70 80 90 100 virus. °C * A 3-methylcholanthrene (MCA)-induced fibrosarcoma cell line FIG. 2. Thermal denaturation profiles of hybrids formed between from a BALB/c mouse. M-MSV-specific [3H]cDNA (XVIII) and DNA extracted from cell t MCA1 and MCA2 are 3-methylcholanthrene-induced fibrosar- lines and livers. Thermal denaturation profiles, performed as de- comas in BALB/c mice developed by S. Rosenberg and C. Boone. scribed in the text, are shown for MSV-specific DNA hybrid with: V, BALB/c mouse liver DNA; *, Fischer rat liver DNA; 0, Syrian hamster liver DNA: 0, P521, MSV-producer cat cell line DNA; *, otide sequences were not closely related to those of MSV-spe- CCC, normal cat cell line DNA; *, DK, dog kidney cell line DNA: A, cific cDNA; the thermal denaturation profiles showed tm values F49-I, human amnion cell line DNA. Internal controls included 0, of 57°-64°. MSV-MuLV [32P]cDNA hybridized with 65S MSV-MuLV RNA with To determine that the present experimental conditions could a Crt of 10 mole-sec/liter. Another internal control used was A, mouse detect virus-specific RNA in virus-transformed cells, mouse cells globin [32P]cDNA incubated with BALB/c DNA to a Cot of 4 X 103 transformed by Ki-MSV, B77 avian sarcoma virus, or SV40 were mole-sec/liter. Finally, o, BALB/c cellular DNA denaturation was monitored with percent absorbance eluted as the ordinate scale. examined with appropriate cDNAs (Table 4). Ki-MSV-trans- formed nonproducer mouse cells are known to express Kir- indicated that a larger number of RNA copies was present in sten-sarcoma-virus-specific and Ki-MuLV-specific RNA (10). cells actively producing M-MSV. Although not shown, het- Accordingly, cDNA probe derived from Ki-MuLV should de- erologous cat cells transformed by M-MSV also contained tect virus-specific RNA in the same Ki-MSV-transformed MSV-specific cDNA-specific RNA. In contrast, normal mouse nonproducer mouse cell subjected to the identical conditions cell RNA from animals or from cells in tissue cultures did not under which no hybridization was detected with M-MSV- contain RNA complementary to MSV-specific cDNA. BALB/c derived MSV-specific cDNA. In such control experiments with or NIH Swiss mouse embryo RNA also had no detectable RNA Kirsten MuLV cDNA (made from virus grown in mouse cells), complementary to MSV-specific DNA. of normal Ki-MSV-transformed mouse cells contained approximately 100 mouse cells with M-MuLV did not lead to transcription of copies of Ki-MuLV related RNA per cell (Table 4). Similarly, M-MSV-specific RNA, although maximal hybridization of the SV40 minus strand sequences were found in SV40-transformed BALB 3T3 and ASV sequences were found in ASV- cell RNA with MuLV cDNA with a Crt1/2 of 30 was obtained cells, (Table 4). Revertants of M-MSV-transformed mouse cells transformed BALB 3T3 cells (Table 4). Accordingly, the data similarly did not contain M-MSV-specific RNA. Two different obtained from mouse cells transformed by different means do or methylcholanthrene (MCA)-induced sarcomas of BALB/c mice not indicate that chemical transformation transformation tran- as well as one cell line obtained from a methylcholanthrene- by heterologous sarcoma viruses is mediated via the induced sarcoma in a BALB/c mouse also did not contain RNA scription of Moloney-sarcoma-virus-specific RNA. complementary to MSV-specific DNA. Mouse cells transformed by an avian sarcoma virus genome or by Ki-MSV, where neither DISCUSSION virus contains M-MSV-specific cDNA, did not contain RNA Nucleotide sequences homologous to MSV-specific cDNA were complementary to MSV-specific cDNA at the highest Cot found both in DNA of several mammalian cells transformed (about 104.5) values used. Of interest were the 10-16% values by M-MSV and also in normal mouse cell DNA. Normal mouse obtained with RNAs of mouse cells transformed with Ki-MSV, cells had approximately one set of MSV-specific sequences per ASV, SV40, and one of the MCA tumors. However, the nucle- haploid genome. The tm of the MSV-specific cDNA hybrid with Downloaded by guest on September 27, 2021 Microbiology: Frankel and Fischinger Proc. Natl. Acad. Sci. USA 73 (1976) 3709 Table 4. Detection of viral RNA in mouse cells quences during DNA to RNA to DNA information transfer (24). transformed with Ki-MSV, ASV, or SV40, or infected The above data are compatible with the interpretation that with M-MuLV sarcoma viruses are recombinants with leukemia type viruses that transduce some cellular information, which in some as yet Maximal undefined way leads to cell transformation (10, 20). Transforming cDNA hybrid- We thank Cy Cabradilla for the RNA extraction technique, John Mouse cell virus as probe ized Crty2 Martin for providing the methylcholanthrene tumors, George Todaro for the MC55 cell line, J. Schlom for MMTV RNA, H. Varmus and N. BALB-3T3 Ki-MSV Ki-MuLV 40 200 Quintrell for the avian sarcoma virus cDNA, and P. Vogt for the B77 BALB-3T3 ASV ASV 80 4000 cell line. We thank P. Leder for the 32p globin cDNA, and G. Khoury BALB-3T3 SV40 SV40 55 400 for the minus strand 32p SV40 cDNA. 3T3FL M-MuLV* M-MuLV 90 50 1. Temin, H. M. (1971) Annu. Rev. Microbiol. 25,609-648. Cell RNA was prepared as in Table 3. Hybridization was carried 2. Lowy, D. R., Rowe, W. P., Teich, N. & Hartley, J. W. (1971) out as described in the text. Science 174, 155-156. * Nontransforming MuLV. 3. Chattopadhyay, S. K., Lowy, D. R., Teich, N. M., Levine, A. S. & Rowe, W. P. (1974) Proc. Natl. Acad. Sci. USA 71, 167- all mouse DNAs was 800 to 81°, indicating that MSV-specific 171. sequences in the cell DNA matched closely the MSV-specific 4. Fischinger, P. J., Peebles, P. T., Nomura, S. & Haapala, D. K. viral sequences. Less of MSV-specific cDNA was found in rat (1973) J. Virol. 11, 978-985. DNA or hamster DNA, and the nucleotide sequences were 5. Benveniste, R. E., Lieber, M. M., Livingston, D. M., Shen, C. J., much less related, as seen by reduced tm values. Cat, human, Todaro, G. J. & Kalter, S. S. (1974) Nature 248, 17-20. 6. Neiman, P. E., Wright, J. E., McMillin, C. & MacDonnel, D. and dog cell DNA had little homology to MSV-specific cDNA, (1974) J. Virol. 13, 837-846. whereas salmon and chicken DNA had essentially no homology 7. Stehelin, D., Guntaka, R. V., Varmus, H. E. & Bishop, J. M. (1976) to MSV-specific cDNA. Evolutionary relationships can be es- J. Mol. Biol. 101, 349-365. tablished among species with MSV-specific cDNA, as has been 8. Scolnick, E. M., Howk, K. S., Anisowicz, A., Peebles, P. T., Scher, done with unfractionated cDNAs from endogenous viruses of C. D. & Parks, W. P. (1975) Proc. Natl. Acad. Sci. USA 72, several species (22). Accordingly, these findings generally agree 4650-4654. with the presence of cDNAsrc sequences of ASV in the DNA of 9. Frankel, A. E., Neubauer, R. L. & Fischinger, P. J. (1976) J. Virol. normal chicken cells and other avian species (18). Total cell 18,481-490. RNA from M-MSV-transformed cells and normal mouse cells 10. Scolnick, E. M., Rands, E., Williams, D. & Parks, W. P. (1973) was reacted with MSV-specific cDNA to pres- J. Virol. 12,458-463. determine the 11. Scolnick, E. M. & Parks, W. P. (1974) J. Virol. 13, 1211-1219. ence of MSV-specific RNA in normal adult, embryonic, or 12. Nomura, S., Fischinger, P. J., Mattern, C. F. T., Peebles, P. T., transformed mouse cells. The final Crt values were about 104-5, Bassin, R. H. & Friedman, G. P. (1972) 50,51-64. which were the highest we could attain experimentally. Al- 13. Thomas, C. A., Bens, V. I. & Kelly, T. J., Sr. (1966) in Procedures though RNA, homologous to MSV-specific cDNA, was found in Nucleic Acid Research, eds. Antoni, G. C. & Davies, D. R. in M-MSV-transformed cells of several species examined, none (Harper and Row, New York), p. 535. was detected in normal adult mouse cells, indicating less than 14. Britten, R. J. & Kohne, D. E. (1967) Carnegie Inst. Washington about 3 to 10 copies per cell. In mouse cells transformed by Yearb. 65, 78-106. Ki-MSV, ASV, or SV40 virus essentially no RNA complemen- 15. Britten, R. J. & Kohne, D. E. (1968) Science 161, 529. tary to MSV-specific cDNA was detected. 16. Scherrer, K. (1969) Fundamental Techniques in Virology, ed. Habel, K. & Salzman, H. P. (Academic Press, New York), p. The presence of the majority of MSV-specific sequences in 413. the DNA of normal mouse cells and the presence of these se- 17. Leong, J. A., Garapin, A. C., Jackson, N., Fanshier, L., Levinson, quences in the RNA in M-MSV-transformed mouse cells raised W. E. & Bishop, J. M. (1972) J. Virol. 9,891. the possibility that there was one set of DNA sequences in all 18. Stehelin, D., Varmus, H. E., Bishop, J. & Vogt, P. K. (1976) Nature mouse cells which might be transcribed into RNA during cell 260, 170-173. transformation. The present data are not compatible with the 19. Callahan, R., Lieber, M. M. & Todaro, G. J. (1975) J. Virol. 15, hypothesis that the M-MSV-specific nucleotide sequences are 1378-1384. present in RNA in all transformed mouse cells. It may be pos- 20. Fischinger, P. J. & Haapala, D. K. (1974) Immunol. Cancer Prog. sible that several of these apparently unrelated sarcoma viruses Exp. Tumor Res. 19,1-22. induce the transcription of unknown common cellular genes 21. Varmus, H. E., Bishop, J. M. & Vogt, P. K. (1973) J. Mol. Biol. in 74,613-626. involved pleiotropic responses. An explanation for the origin 22. Benveniste, R. & Todaro, G. (1974) Proc. Natl. Acad. Sci. USA of oncoviruses with different nucleotide sequences involved in 71,4513-4518. cell transformation has been proposed by Temin (24). The genes 23. Khoury, G., Byrne, J. C., Takemoto, K. K. & Martin, M. A. (1973) responsible for both virus production and cell transformation J. Virol. 11, 54-60. would have misevolved from normal cellular nucleotide se- 24. Temin, H. M. (1974) Cancer Res. 34, 2835-2841. Downloaded by guest on September 27, 2021