Studies on reverse transcriptase of RNA tumor viruses. I. Localization of thermolabile DNA polymerase and RNase H activities on one polypeptide.

Copyright@1975 AmericanSocietyfor Microbiology Printed inU.SA.

Studies

on

Reverse Transcriptase

of

RNA Tumor Viruses

I.

Localization

of

Thermolabile

DNA

Polymerase and RNase H

Activities on One Polypeptide

INDER M. VERMA

TumorVirologyLaboratory, TheSalk Institute,SanDiego,California 92112

Received forpublication6September1974

Purified reverse transcriptase from avian myeloblastosis virus or Rous sarcoma virus consists of two subunits of average mol wt of 100,000 and 60,000. The

lower-molecular-weight subunit, a, has been isolated from avianmyeloblastosis virus, Roussarcomavirus, andatemperature-sensitivemutantofRous sarcoma virus, LA337. Subunit a manifests both the DNA polymerase and RNase H activities associated with purified reverse transcriptase of avian RNA tumor viruses. Thethermal inactivation of these enzymatic activities ofasubunit from LA337was comparedwith the isolated a subunit from thewild-typevirus.The

resultsshow thatbothDNApolymeraseand RNase H activities associated with the a subunit of LA337 are five to seven times more thermolabile than the

correspondingasubunit fromthewild-type virus. It isconcluded that (i) both thepolymeraseandnucleaseactivities resideonthe samepolypeptidechain, and

(ii)atleast the lower-molecular-weight subunit a is coded forby the viral RNA.

Avian RNA tumor viruses contain a DNA

polymerase whichcanfaithfully transcribe the viral RNA into complementary DNA (11, 12). The DNA polymerase can besolubilized from thevirions and purifiedbycolumn chromatog-raphy (3, 6, 8, 13). Purified DNA polymerase from avianmyeloblastosis virus (AMV)orRous sarcoma virus (RSV), uponanalysisonsodium dodecyl sulfate-containing polyacrylamide gels, dissociates intoequimolaramountsoftwo poly-peptide chains (subunits) with average mol wt of100,000 and 60,000 (4, 8, 11, 15). The

lower-molecular-weight subunit (a) hasbeen isolated

from AMV (4) and has been shown to contain

the threeknownenzymaticactivitiesassociated

with the DNA polymerase from avian RNA

tumor viruses: RNA-directed DNA synthesis,

DNA-directed DNA synthesis, and RNase H (which selectively degradesthe RNAmoiety of RNA-DNAhybrids)(11, 12). DNApolymerases

isolated from two temperature-sensitive mu-tants ofRSVwith defectsin a veryearly event

ofthe virusgrowth cycle (10) have been shown

to be morethermolabile than the DNA

polym-erase fromthewild-type parent (14). Since the

purified DNA polymerase studied in those

ex-periments contained bothsubunits, it could not be conclusively established that the

tempera-ture-sensitive lesion residesinthesubunit

con-taining the polymerase and nuclease activities. To demonstrate that the polypeptide chain

manifesting the synthetic and the degradative activities isthermolabile, the lower-molecular-weight subunit a has been isolated from the temperature-sensitive mutant LA337. The heat-inactivation properties of the DNA polym-eraseand RNase H activities ofasubunit from the mutant virus were compared with the a

subunitisolated from thewild-type parent. The results indicate that both the DNApolymerase

and the RNase H activities of a from LA337 are five to seven times morethermolabile than those oftheasubunit of thewild-typeparent.

MATERIALS AND METHODS

Materials. Tritium-labeled and unlabeled

deoxy-ribonucleoside triphosphates were obtained from

Schwarz/Mann and New England Nuclear.

Polynu-cleotides wereobtainedfromeither MilesLaboratory

or P-L Biochemicals. Oligonucleotides were 8 to 18

in length andwere obtained fromCollaborative

Re-search. AMV inplasmawasprovidedbythe Office of

ProgramResources andLogisticsofthe Virus Cancer

Program of the National Cancer Institute.

[3HJpoly(A)

(2). Escherichia coliDNApolymerasewas agiftfrom

A.KornbergtoD. Baltimore.

Purificationof virion DNApolymerase.Purified

virions of LA337and wild-type RSV, Prague strain

(subgroup C), wereobtainedas describedpreviously

(10). AMV was purified from chicken plasma as

described(13). About 4.5 mg ofpurifiedvirionsfrom

LA337and 14.0 mg ofpurified virions fromwild-type

RSVwerelysed with Nonidet P-40(ShellChemicals), 121

on November 10, 2019 by guest

http://jvi.asm.org/

andDNA polymerase was purified by column chro-matography on DEAE-Sephadex A-25 (13). En-zymatic assays were performed as described previ-ously (13). The peak fractions with enzymatic activi-ties weredialyzed overnight against buffer B

contain-ing 50 mM Tris-hydrochloride (pH 7.5), 0.1 mM

EDTA, 0.1 M ft-mercaptoethanol, 20% glycerol, and

0.15%Nonidet P-40. Thedialyzed enzyme was further

purified on a phosphocellulose column (13). AMV

DNA polymerase was purified as previously described

(13).

RNA-directed DNA synthesis was assayed using

poly(C)-oligo(dG) as template-primer; DNA-directed DNA synthesis was assayed usingpoly(dC) -oligo(dG) astemplate-primer. RNase H activities were assayed as described previously (2, 14). All enzyme assays were carried out for times during which synthesis or

degradationwasalinear function of time. The details

ofvarious reactions are described in thelegends to the

figures.

RESULTS

Isolation and characterization of a from AMV, wild-type RSV, and LA337. The lower-molecular-weight subunit a was isolated by

sequential chromatographyonphosphocellulose columns. Figure 1 shows the chromatographic profiles of DNA polymerases isolated from

LA337 and wild-type RSV. The major peak of

enzymaticactivityobtainedfrom the first

phos-phocellulose chromatography was dialyzed overnight against buffer B and rechromato-graphed on asecond phosphocellulose column. When assayed with poly(C) -oligo(dG) as

tem-plate-primer, two peaks of enzymatic activity

could be distinguished. The minor peak of

enzyme activity, a, elutingat asalt

concentra-tion of 0.1 M, represents 3 to 10% ofthe total

enzyme activity. The major peak of enzyme

activity, a,, elutes at a salt concentration of 0.24 M KCl. Similar separationof two enzyme

activities can be obtained from AMV. In all

subsequent experiments, a and af3 obtained from the second phosphocellulose columnwere

used.

The molecularsize of a and

a#

col-umns. Figure 2represents a composite of three

sets ofexperiments and was drawn by aligning

the E. coli DNA polymerase I activity in all

three experiments. It can be seen that aC, from

AMV,which has anaverage mol wt of 150,000 to

170,000, elutes inthevoid volumeconsiderably

ahead of themarker,E. coli DNApolymeraseI,

which has a mol wt of 110,000 (7). This is

consistentwith the observation that the

holoen-zyme is composed of twosubunits, with atotal

molwtof170,000(6, 8,15).Subunitafromboth

AMV and wild-type RSV elutes from the

col-umnafterE. coli DNApolymeraseI.Analysis of

afrom AMV in sodiumdodecyl

sulfate-contain-ingagarose slab gels shows one major

polypep-tide band with an average mol wt of 60,000

(I. M. Verma and W. Gibson, unpublished

data). Thus, based upon Sephadex G-100

col-umn chromatography and polyacrylamide gel electrophoresis, afrom AMV orRSV wild-type

appearstoconsist ofonlyonepolypeptidechain

with a mol wt of 60,000 to 70,000. This is in

agreementwith the results ofGrandgenettetal.

(4), although their estimate of the molecular

weight of a of AMV is somewhat higher. The

molecular size ofa isolatedfromLA337was not

characterized because ofinsufficient material. It has, however, been shown previously that ad

obtained either from LA337or wild-type RSV

has the same molecular size and number of

subunits (14).

DNA polymerase and RNase H assay ofa

subunit of LA337 and wild-type RSV. The

three known enzymatic activities associated

with theDNA polymerase of avian RNAtumor

viruses were compared using poly(C) oligo(dG)

to measure RNA-directed DNA synthesis,

poly(dC) oligo(dG) toassayDNA-directed DNA

synthesisand [3H]poly(A) poly(dT) to assay for

RNase H activity (Fig. 3). The average time of

inactivationofone-half of the activity of a from

LA337andwild-type RSV for these three

activi-ties has been calculated (Table 1). The a

subunit isolated from the virions of LA337

showed a five- to sevenfold greater

thermolabil-ity than the a subunit isolated from wild-type

RSV for all three activities.

Comparison ofa and a(3 from LA337 and wild-type RSV. Both the RNA-directed DNA

synthesis and DNA-directed DNA synthesis

activities associated with a subunit of LA337

are slightly more resistant tothermal

inactiva-tion than the activities of the corresponding

holoenzyme, af (Fig. 4). Subunit a from the

wild-type parent is, however, slightly more

thermolabile than the af3 complex. The RNase

H activity of a from both LA337 and wild-type

parent arereproduciblymore thermolabile than

that of the two-subunit complex.

DISCUSSION

DNA polymerase isolated from two

condi-tional temperature-sensitive mutants of RSV

(LA335 and LA337) has been shown to be

three-to fivefold more thermolabile than the DNA

polymerase from thewild-type parent (14). The

thermal inactivationprofileofthe DNA

on November 10, 2019 by guest

http://jvi.asm.org/

Wild Type RSV

123

LA337

x

xI

Fraction No.

FIG. 1. Sequential column chromatography ofDNApolymerase from wild-type RSVandLA337. Themajor peak of enzymatic activity from the first phosphocellulosecolumnwaspooled, dialyzed overnight against buffer

B, andrechromatographedon asecondphosphocellulosecolumn. Twopeaks of enzymatic activities,aanda#,

could be distinguished. Peak awasstored in 50%glycerolat -20C,whereasa# wasstoredat -70C. The

recoveryof the enzymatic activity from the second phosphocellulosecolumnwasover50%.Enzymeassayswere

performed utilizing poly(C) -oligo(dG) as template-primer and 3[H]dGTP (100 counts per min/pmol) as

substrate. Peaksaand a#from LA337 virionswere isolated inan identicalfashion, exceptthedialysiswas

carried out for 5 to 8 h instead of overnight to minimize thermal inactivation ofthe enzyme during the

purification procedure. Therecoveryof totalenzymeactivity after first phosphocellulose columnwas over20%.

On the secondphosphocellulose column over 75% ofthe input enzymatic activity was recovered. Peak a

constitutedabout 10% of the totalenzymeactivity. Bothaandaftobtainedfrom LA337 virionswerestoredat

-70 C.

eraseactivity from the wild-typerevertants(wt

LA335 andwtLA337) is indistinguishable from

that of wild-type virus (14). Mason et al.

(Virology, in press) have recently characterized

the biological and biochemical properties of

these mutants, their revertants, and several

recombinants with leukosis viruses and have

shown that the lesion in these mutants is the

increasedthermolability of theirDNA

polymer-aseactivity. Varmusetal. (Cold SpringHarbor

Symp. Quant. Biol., in press) have reported

that less viral DNA is produced in LA335 and

LA337at the nonpermissive temperature than

isproduced by thewild-type virus.

Sinceall three knownenzymeactivities

asso-ciated with the RNAtumorvirusDNA

polym-erase werethermolabile, it was suggested that

all three activities were located on the same

polypeptide chain (14). To establish that the

polypeptide chainmanifesting all threeenzyme

CY

N

0

x

10 0 0

0. CL

0

'IO

E

IL

on November 10, 2019 by guest

http://jvi.asm.org/

VoLw volume UNAP'olymerose1

a) I ~~~~~~~~~~0

o

10

40-20 25 Q.

p h coln3lofAMVa aofrA M

Peakafw-RaRSV

20tie 0m ri-yrclrd (p 7.) 02M

polmeas 0aIde sasanadmre.Dx

-~0/

E

xo -~~~

O-x~~ ~ ~~~~-~~>

10 15 20 25 30

Effluent m2

FIG. 2. Size determinationofaandaffromsecond

phosphocellulosecolumn. Peaksaanda#fromAMV

were obtained as described in the legend to Fig. 1.

Peakafrom wild-typeRSV andaand az fromAMV were separately chromatographed on a Sephadex G-100 column (1 by 70 cm). The elution buffer

contained50mMTris-hydrochloride (pH 7.9),0.2M

KCI,0.1 mMEDTA, 10 mM mercaptoethanol, and

0.1% Nonidet P-40. In each case, E. coli DNA

polymeraseIwas added as astandard marker.

Dex-tran blue was used to determine the void volume. Peaks a and a# were assayed for enzyme activity

usingpoly(C) oligo(dG)astemplate-primer(15). The

Poly(c)-oligo(dG)

activities isthermolabile and isencoded by the viralRNA, I isolatedasubunit from LA337and

compared its thermal inactivation properties with isolated a subunit from wild-type RSV. The a subunitwasisolated from purified DNA polymerase which had bothsubunitsin equimo-lar ratio. This approach assures that the iso-lated lower-molecular-weight subunit a is part

oftheviral DNA polymerase. The mechanism

ofseparation of a from

a#

com-plex) is not clear. A comparison ofthe tryptic peptides of isolatedaand,8(large subunit)from

either AMV or wild-type RSV reveals thatthe two subunits share at least a portion oftheir amino acid sequences (Gibson and Verma,

un-published data). Thus, based upon the

struc-tural relatednessofthetwo subunits and poly-acrylamide gel electrophoretic profile of aged

reverse transcriptase from AMV (K. Molling,

Cold Spring Harbor Symp. Quant. Biol., in

E. coli DNA polymerase was assayed utilizing poly-(dA)-oligo(dT) as template-primer and 8[H]TTP as substrate (15). The figurewasdrawn by aligning the E. coli DNA polymerase activities of the three

sepa-rateexperiments. The recovery of enzyme activity of afifrom the column was over90%1, whereas a and E. coli DNA polymerase showed a recovery of over 30% of

theinputenzymeactivity.

PolytdC) oligo (dG) Ribonuclease H

Min. Incubotedat450 Min. Incubotedat460

5

Min.Incubatedot470

FIG. 3. Thermalinactivationprofiles ofDNApolymeraseand RNase H activitiesofisolatedasubunitfrom

LA337andwild-typeRSV.Assaysofheatlability wereperformed asdescribed (15).Briefly, the enzymewas

preincubated in reaction mixture 1, which contained 100 mM Tris-hydrochloride (pH 8.3), 20 mM

dithiothreitol, 12 mM Mg acetate, and 120 mMNaCl. The reaction mixture was incubated at the high

temperature showninthefigure,andatvariousintervalsportions (0.05ml)werewithdrawn and addedto0.05 mlof reaction mixture 2, which containedtemplate-primerand thesubstrates. This mixturewaskeptonicefor amaximum of10to 15 min, thenfurther incubatedat35Cfor30min.Acidprecipitable radioactivitywas

determinedasdescribedpreviously (1). (a)Approximately0.005to0.05unitsofpurifiedasubunitwasused per assay. (A unit of enzyme activity is defined as theamount ofenzyme required to incorporate 100pmol of

s[H]dGMPat37C in15min.) Onemicrogramofpoly(C)and0.5jigofoligo(dG)wereusedastemplate-primer,

and10nmolofs[H]dGTP (1,000countspermin/pmol)wasusedassubstrate.The enzymewaspreincubatedat 45C. (b) The samereaction conditions as described above were employed, except 1.0 gg of poly(dC) was substitutedforpoly(C),and thepreincubationwascarriedout at46C.(c)RNaseHassayswereperformedas

previously described [Vermaetal.,Nature(London),inpress].About1 to2unitsofenzymeactivitywasused,

and thepreincubationswerecarriedout at47C.

on November 10, 2019 by guest

http://jvi.asm.org/

TAMZ 1. Average one-half time for inactivation of DNA polymerase and RNase H activity of isolated a

from LA337 and wild-type RSV

t%,(min)"

RNA-

DNA-Source directed directed RNase H

Source DNA DNA R

e[H]

synthesis synthesis poly(A)-poly(C)* poly(dC) poly(dT) oligo(dG) oligo(dG)

LA337 1.8 2.1 1.5

Wild-type RSV 8.8 9.5 8.0

a

t%

activity at the temperatures indicated in Fig. 3.

LA 337

0

If)

re)

0 >-.

Z

.1-C c

press), it appears that a is derived from

f.

Experiments performed on the mechanism of

RNase H (5) and heat stability of isolated a subunit in the presence or absence of template (A. Panet, I. M. Verma, and D. Baltimore,

Cold Spring Harbor Symp. Quant. Biol., in press) suggest that , subunit enhances the affinity ofaforbindingtothetemplate.Weare

now investigating whether 6 is enzymatically activeand how a isgenerated from,.

ACKNOWLEDGMENT1S

I am verygratefulto W.S. Masonforgenerously providing me with mutant andwild-typevirus. I thank D.Baltimorefor hissustainedinterest inthisworkand helpinthepreparation

Wild Type RSV

Min. Incubated at470 Min. Incuboted at470

FIG. 4. Comparison of thermal inactivation profiles of DNA polymerase and RNase H activities ofaanda#, fromLA337 andwild-typeRSV. Assays for heatlabilitywereperformedasdescribed. About0.5 to 1.0unitsof

a,@wasused per assay. Preincubationswerecarried outattemperaturesshown in thefigure.

125

VOL.

on November 10, 2019 by guest

http://jvi.asm.org/

of this manuscript. I am grateful to Amos Panet for many valuable discussions.

Thiswork wascarried out in thelaboratoryofD. Baltimore at the Department of Biology and Center for Cancer Re-search, Massachusetts Institute ofTechnology, Cambridge, Mass. It was supportedbyacontract fromthe VirusCancer Program, Public HealthService grant lP01-CA-14051 from the National Cancer Institute to D.Baltimore, PublicHealth Service grant CA-16561-01 fromthe National Cancer Insti-tuteandaresearchgrant, 320, fromthe Jane Coffin Childs Memorial Fund to I.M.V.

LITERATURE CITED

1. Baltimore, D., A. S. Huang, and M. Stampfer. 1970. Ribonucleic acidsynthesisofvesicular stomatitis virus. II. AnRNApolymerase in the virion. Proc. Nat. Acad. Sci. U.S.A.66:572-576.

2. Baltimore, D., and D. Smoler. 1972. Association of an endoribonuclease with the avian myeloblastosisvirus DNApolymerase. J. Biol.Chem.247:7282-7287. 3. Faras, A. J., J. M. Taylor, J. P. McDonald, W. E.

Levinson, and J. M. Bishop. 1972. Purification and characterization of thedeoxyribonucleicacid polymer-aseassociatedwith Roussarcomavirus.Biochemistry

11:2334-2342.

4. Grandgenett,D.P.,G. F.Gerard,and M.Green.1973. A single subunit from avian myeloblastosis virus with both RNA-directed DNApolymerase activityand ribo-nuclease H activity. Proc. Nat. Acad. Sci. U.S.A. 70:230-234.

5. Grandgenett,D.P., and M. Green.1974.Different mode of action of ribonuclease H in purified a and a,B ribonucleicacid-directeddeoxyribonucleicacid polym-erasefrom avian myeloblastosisvirus. J. Biol. Chem. 249:5148-5152.

6. Hurwitz, J., and J. P. Leis. 1972.RNA-dependentDNA polymerase activity of RNA tumor viruses. I.Directing

influence of DNA in the reaction. J. Virol.9:116-129. 7. Jovin, T. M., P. T. Englund, and L. L.Bertsch. 1969. Enzymatic synthesis ofdeoxyribonucleic acid. XXVI. Physical andchemical studies of a homogeneous deoxy-ribonucleic acid DNA polymerase. J. Biol. Chem. 244:2996-3008.

8. Kacian,D. L., K. F. Watson, A. Burny, and S. Spiegel-man. 1971. Purification of DNApolymerase of avian myeloblastosis virus. Biochim. Biophys. Acta 246:365-383.

9. Leis,J.,I.Berkower,and J.Hurwitz. 1973. RNA-depend-entDNApolymerase activity in RNA tumor viruses, p. 287-308.In R. D. Wells and R. B. Inman(ed.), DNA synthesis in vitro. University Park Press, Baltimore. 10. Linial, M., and W.S.Mason. 1973.Characterizationof

twoconditional earlymutantsofRoussarcoma virus. Virology 53:258-273.

11. Temin, H.M., and D. Baltimore. 1972. RNA directed DNAsynthesis of RNA tumor viruses, p. 129-186. In K. M.Smith and M. A. Lauffer (ed.), Advances in virus research, vol. 17. Academic Press Inc., New York. 12. Tooze, J. (ed.). 1973. The molecular biology of tumor

viruses. Chapter 11. ColdSpringHarborLaboratory, New York.

13. Verma, I. M., and D. Baltimore. 1973. Purification of the RNA-directed DNApolymerase from avian myeloblas-tosis virusand its assay withpolynucleotidetemplates, p. 125-131. In L. Grossman and K. Moldave (ed.), Methods in enzymology, vol. 29. Academic Press Inc., NewYork.

14. Verma, I. M., W. S. Mason, S. D. Drost, and D. Baltimore. 1974. DNA polymerase activity from two temperature-sensitive mutants ofRous sarcoma virus isthermolabile. Nature (London) 251:27-31. 15. Verma, I. M., N. L. Meuth, H. Fan, and D. Baltimore.

1974. Hamster leukemia virus: lack of endogenous DNA synthesis and unique structure of its DNA polymerase. J. Virol. 13:1075-1082.

on November 10, 2019 by guest

http://jvi.asm.org/