Selective decrease in the rate of cleavage of an intracellular precursor to Rauscher leukemia virus p30 by treatment of infected cells with actinomycin D.

(1)

Selective

Decrease in

the Rate

of

Cleavage of

an

Intracellular

Precursor

to

Rauscher Leukemia Virus

p30

by Treatment of

Infected

Cells with Actinomycin

D

G. A. JAMJOOM, R. B. NASO, AND R. B. ARLINGHAUS*

BiologyDepartment, The University of Texas System CancerCenter, M. D.Anderson Hospital and Tumor

Institute, Houston, Texas 77030

Received for publication26January1976

The cleavage ofan intracellular 67,000- to 70,000-dalton precursor, termed

Pr4, toRauscher leukemia virus (RLV) p30 protein proceededata slowerrate

when virus-producingcellsweretreated with actinomycin D (AMD). Treatment

with AMD also caused a slight accumulation of Pr4 in purified early virus

particles produced by a cell line which usually produces virions that contain

little Pr4. The cleavage of other intracellular viral precursorpolypeptides was notaffected bytreatmentwith AMD. Treatment ofinfected cells with cyclohexi-mide,onthe other hand, allowed the cleavage of Pr4toproceedatthe usualrate

for a short period of time before further cleavage was drastically slowed or

prevented. The cleavage of several other viral precursor polypeptides wasalso

inhibited bytreatmentwithcycloheximide. Different lines of evidence suggest

that themechanism of action of AMD isnotduetoapossible indirect effecton

protein synthesis. Thus, the rate ofcleavage of Pr4 was not affected by the

length ofpretreatment with AMDbetween 1 to8h. Inaddition, the combined effectof AMD andcycloheximide, attheir maximal inhibitory concentrations,

wasgreaterthan the effect of either drug alone, indicating the involvement of

two atleast partially different mechanisms in the action of AMD and

cyclohexi-mide. Furthermore, AMD did not affect the pulse labeling ofviral precursor

polypeptides. These results suggest that the interaction with viral RNA, whose production is inhibited by AMD, accelerates the cleavage ofPr4 top30 during virus assembly. A hypothetical model is presented to illustrate the possible advantages ofhavingastepin virusassembly in which genomicRNA interacts

witha precursortocapsid proteins before the cleavage of that precursor.

The cleavage of precursor proteins during the maturation of viral particles is a phenomenon

ofwidespreadoccurrence. Ithas been observed

in many types of viruses, including

bacterio-phages T4 andX, picornaviruses, Sindbisvirus, vaccinia virus, influenza virus, Sendai virus,

and adenoviruses, among others (for reviews see8, 21, 27, 46, 48). Although the purposesof

suchproteolytic cleavage are not definitely de-termined, they evidently include the formation ofthe finalstructural proteins and the facilita-tionand stabilizationof the protein-proteinand

protein-nucleic acid interactions that occur dur-ing assembly.

Differenttypes of proteolytic cleavage can be

distinguished in certain instances. For

exam-ple, inpicornaviruses, nascent chain cleavage, which mayfacilitate the release of segments of thepolyprotein chain from the ribosome (7), is

distinguishedfrom eitherintermediate or

mor-phogenetic-type cleavages. Morphogenetic

cleavage is strongly linked with virus

assem-bly. The mostprominent morphogenetic cleav-age is thatwhichaccompanies thepackagingof

theviralgenome. This type of cleavage in most cases seems torequire an interaction with the

viral nucleicacidsince itusually doesnot occur in emptyparticles.

Proteolytic cleavage also plays a prominent role in the formation of oncornavirus proteins (2, 37, 51; R. B. Arlinghaus, R. B. Naso, G. A.

Jamjoom, L. J. Arcement, andW. L. Karshin,

in D. Baltimore, A. S. Huang, and C. F. Fox,

ed.,Animal Virology, ICN-UCLA Symp. Mol. Cell. Biol., in press). In Rauscher leukemia

virus (RLV) the group-specific core protein,

p30, is made by way of cleavage of several

precursors of higher molecular weight. The most prominent of these are two precursors withmolecularweightsof80,000 and67,000to

70,000, designated Pr3 and Pr4,

respectively.

Less prominent, based on their occurrence in

1054

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

CLEAVAGE OF RLV p30 PRECURSOR 1055

small quantities in infected cell extracts, are

larger precursors with a molecular weight of about 200,000,designated Prla+b (2).

The effect of actinomycin D (AMD) on the inhibition of oncornaviral RNA synthesis is well established (4, 10, 33, 50). Treatment of

virus-producing cells with AMD, however,

al-lows for several hours the synthesis of viral proteins and the maturation ofRNA-deficient viral particles that have normal morphology (33). Thus, treatment with AMDmakesit

pos-sibletoexamine the role ofgenomicRNA inthe maturationandprocessing of theviralproteins

duringvirusassembly.

In this paper, we describe the observation

thattreatmentofchronically infected cellswith AMD noticeablyslowed the rateofcleavage of Pr4 and caused its accumulation in infected cells and,to asmallextent, inearlyvirus

parti-cles. Pr4, previously designated p70,is an

inter-mediate intracellular precursor polypeptide, but is present in mature virions produced by severalinfected cell lines (24).

MATERIALS AND METHODS

Cells and virus. RLV-infected NIH Swiss mouse

embryo (JLS-V16) (37) and RLV-infected BALB/c

mouse spleen and thymuscells (JLS-V5) (49) were

usedinthis study. Theculturemedium contained a

modified Eagle amino acid formula and 10% fetal

calf serum, asdescribedpreviously (49). Cellswere

grown in 2-ounce (about 60-ml) prescription glass

bottles or, for virus production, in l-quart (about

0.95-liter) glass bottles. The cultures wereused 3 to

6 daysafter passage and werenearly confluent.

For viruspurification, culturefluid was collected

and clarified at 10,000xg for 10 min.The virus was

pelleted at 78,000 x g for 2 h, suspended in TNE

buffer (0.01 M Tris, pH 7.5, 0.1 M NaCl, 0.001 M

EDTA), and banded by isopycnic centrifugation on a

15 to60% (wt/vol) sucrose gradient in TNE for 16 h

at13,000 rpmin aBeckmanSW27rotor.Torecover

cell-associated viral particles, the cell sheet was

rinsed with rinsing buffer (0.14 M NaCl, 5 mMKCl,

0.3 mM

Na2,HPO47H20,

0.4 mMKH2PO4, 5.5 mM

glucose, 4 mM NaHCO3, 100 mg of neomycin per liter) and treated for 5 min with 0.02% trypsin in

rinsing buffer. The resultant cell suspension was

pooled with the culture fluid, and the cells were

pelleted. The supernatant fluids wereprocessed for

virus purification asdescribed above. The sucrose

gradients were fractionated into 1-ml portions, and trichloroacetic acid-insoluble radioactivity and

den-sity were measured. Theradioactivity peak at

dens-ities of 1.13 to 1.16g/cm3waspooled, dilutedinTNE,

andpelletedat78,000 xg for2h.

Labeling of cells and virus.Labeling of cells for

pulse-chase studies was done as follows: the cells

wererinsed with warm Hanks solution and labeled

in Hanks solution containing 100

ACi

of

[35S]methionine

per ml (288 Ci/mmol; Amersham/

Searle) or in complete growth medium (49) minus

methionine containing 5%dialyzedfetal calfserum

and no tryptose phosphate. The pattern of

radioac-tive proteins was identical in either case. For a

chase, the cells were rinsed with Hanks solution and

incubated in complete growth medium. For virus

labeling, the cells were rinsed with Hankssolution

and incubated for 4 h in growth medium which

contained 45 ,MCi of [35S]methionine per ml, 1/10

Eagle's concentration of unlabeled methionine, 5%

dialyzed fetal calf serum, and no tryptose phos-phate.

Immunoprecipitation and gel electrophoresis. Cell lysis, the preparation of anti-RLV antiserum, and the immunoprecipitation of virus-specific

pro-teins in infected cell lysates were described

previ-ously (37). Sodium dodecyl sulfate (SDS)-polyacryl-amide gel electrophoresis of proteins was done by using the buffer system described by Laemmli (28);

RNA urea-acrylamide-agarose gel electrophoresis

wasdone as described by Floyd et al. (17) except that

SDS was used instead of lithium dodecyl sulfate,

andelectrophoresis was at room temperature. The

gels were processed for fluorography as described by

Bonner andLaskey (6). To obtain a linear response

to radioactivity, the X-ray films were preflashed

(32). To quantitate the amount of radioactivity in

different bands, the films were scanned at 590 nm in

aGilfordspectrophotometer, and the relative areas

ofdifferentpeaksintheresultingcurves were

mea-suredinaDuPont310 curveresolver.

RESULTS

Effect of AMD on viral RNA and protein

synthesis inchronicallyinfected cells. Inhibi-tion ofoncornaviral RNA synthesis by treat-ment with AMD occursrapidly (e.g., reference 4). On the other hand, the synthesis of viral

proteins and the maturation of RNA-deficient particles in murine leukemia virus-infected

cellscontinuesforseveral hours in the presence of AMD (33). These two results are shown in

Fig. 1 for the RLV-infected JLS-V16 cell line used in these studies. In this experiment, the cells were treated with 5,g of AMD per ml for 75 min before they were pulse-labeled for 75

min with mediumcontaining both [H]uridine and

[P5S]methionine.

After a chase for 3 h in

complete medium, the virus waspurified ona

sucrose gradient andthe radioactivity was

de-termined. It can be seen that whereas the

[PH]uridine

radioactivity was drastically

re-duced ascompared with the

control,

no

signifi-cantdifference had occurred in the

[35Jmethio-nine radioactivity during this period. This re-sult confirms thefinding that viralprotein

syn-thesis andmaturationof viral particles can oc-curintheabsenceof viral RNAsynthesis(33),

indicatingthe stability of the viral mRNA for severalhours (34).

We have estimated the rate ofsynthesis of intracellular virus-specific polypeptides by

VOL. 19, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

(3)

1,400

1,200

1,000

2

800

u 600 0

cr-400

200

15

14

13

12

II z

E

0-z

5 10 15 20 5 10 15 20

FRACTION NUMBER

FIG. 1. Effect ofAMDon viral RNAandproteinsynthesis.RLV-infectedJLS-16 cells were pretreated for

75 min with 5

lAg

ofAMDper ml in complete medium, pulsed for 75 min with methionine-free medium

containing

/P5SImethionine, /PIHluridine,

and 5 Ag of AMD per ml, and chased for 3 h in complete medium

containing 5pgof AMD per ml. Virus was collected and purifiedona 15to60% sucrose gradient. Fractions

wereassayed for their density and trichloroacetic acid-insoluble radioactivity. For control virus, the cells were

similarly treated, exceptthat AMDwasomitted.

measuring the amount ofradioactivity after a 15-minpulselabel withradioactiveaminoacids that is immunoprecipitable in cell extracts by antiserapreparedagainst mature virions.Such

measurements have indicated that 5 h of pre-treatment with 5 to 10

gg

of AMD per mlonly

reduced by 15 to 20% the incorporation of I:5S]methionineintointracellularvirus-specific polypeptides, whereas a 7-h pretreatment was

required to reduce incorporationby 50%. These observations suggest that the viral mRNA has

a functional half-life ofapproximately 7 h. A

similar value has been reported by Levin and Rosenak (34).

Effect of AMI) on the rate ofcleavage of Pr4 to p30. Figure 2 shows the effect of

pre-treatment of RLV-infected cells with AMD on

the rate ofcleavage of Pr4 and appearance of p30. After a 1-h treatment of infected cells either with AMD in growth medium or with

fresh medium without AMD, cells were pulse labeled for 15 min with

[t5S]methionine

in

Hankssolution in the presence or absence of5

,ug of AMD per ml. The label was chased in

complete growth medium for different times

while maintaining the presence or absence of

AMD. Cells were then lysed, and the

cytoplas-micvirus-specificpolypeptides wereisolatedby

immunoprecipitation with anti-RLVserum. In

control cultures (Fig. 2B, D, F, and H) Pr4

disappeared during a chase period of3 h at a

ratesimilar tothatatwhich p30appeared dur-ing thisperiod. The fact thatPr4 is aprecursor

top30 has been verified by pulse chase studies

and trypticdigestion (2). In AMD-treated

cul-tures (Fig. 2A, C, E, and G), cleavage of Pr4to

p30 also occurred, but at a much slower rate.

Quantitation of the results of Fig. 2 isshownin

Fig. 4A as the ratio of p30 to Pr4 obtained at

different times ofchase in the presence or

ab-sence ofAMD.

The disappearance of the p30 precursors

Prla +b and Pr3 during the chase was also evident in Fig. 2. However, in contrast tothe disappearance of Pr4, the rates at which Prla+b and Pr3 disappeared were similar in

AMD-treated and control cultures. This shows

that AMD does not affect the cleavage of all

viral precursor polypeptides indiscriminately.

Dose-response analyses of the effect of AMD on the cleavage of Pr4. The effect of pretreatment with different doses of AMD on thecleavageof Pr4 top30 was examined.Table 1 shows the effect of a 1-h pretreatment with differentconcentrations ofAMDonthe p30/Pr4 ratio obtained aftera subsequent 15-min pulse and 1-hchase. It can beseen that the effect of AMDbecame noticeable under the conditions of the experiment at 0.1

gg/ml.

This

concentra-tion of AMD is known to inhibit viral RNA

synthesis (10). Theeffect ofAMD washigherat 1

gg/ml,

but nofurther increase was observed

when theconcentration was raisedto 10 ,ug/ml. This result suggests that the effect ofAMD is

specificand not due togeneral cytotoxic effects, which would beexpected toincrease at higher

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.505.125.408.68.261.2]

(4)

!I

tJ

"I

W. ~~~~~~

- -...

FIG. 2. EffectofAMD on the cleavage ofPr4 top30.RLV-infected JLS-V16 cellswerepretreatedwith 5pg

of AMD per ml for1 h incomplete growthmedium.-The culturefluidwasremoved, and the cultureswere

pulselabeledfor 15minwith[35S]methionineinHanksbuffercontaining5pgofAMD per ml.The cell sheet

wasrinsedand incubated incompletegrowthmedium containing 5pgofAMD permlfor15 min(A),30min

(C),1h(E),and 3 h(G).Control cultures: Parallel cultureswereprocessedasabovewithout AMD andchased

for15 min(B),30 min(D),1 h(F),and 3 h(H).Cytoplasmicextracts weretreated with antiserum toRLV

proteins, and the immunoprecipitates wereanalyzed bySDS-polyacrylamide gel electrophoresis (6 to 12%

[image:4.505.58.447.72.388.2]

lineargradientgel). Similaramountsofradioactivitywereapplied.

TABLE 1. Dose-response analysis of the effect of

AMD on thecleavageof Pr4a

Concn ofAMD p3O/Pr4

(,ug/ml)

0 1.67

0.1 1.20

1.0 0.54

10.0 0.53

a RLV-infected JLS-V16 cultures were treated

with freshcomplete growth mediumcontaining the

different AMD concentrations for 1 h. They were

then pulsed with [35S]methionine in Hanks buffer

for 15min and chasedin complete growthmedium

for1h, while maintaining thesamedrug

concentra-tions during the pulse and chase. Virus-specific polypeptides were immunoprecipitated from

cyto-plasmic extracts and analyzed by

SDS-polyacryl-amidegel electrophoresis andautoradiography.The

autoradiogram was scanned, and thep30 and Pr4

peakswerequantitated.

concentrations of the drug. The duration of

AMD treatment inthisexperiment (2 hand 15 min)didnotresultinnoticeable changes in cell

morphology. Cells treated with 10 ,ug ofAMD per ml could still make viral proteins and ex-port virusfor several morehours.

Effect of cycloheximide on the rate of

cleavage ofPr4 to

p30.

The clarification of the mechanism by which AMD slows the cleavage

ofPr4requires an examination of the effect of inhibition of protein synthesis on the cleavage of Pr4 (seebelow). We therefore tested the ef-fect ofcycloheximide onthe rate ofcleavage of Pr4 (Fig. 3). In this experiment, the cells were

pulsedfor 15 minin Hankssolution containing

[35S]methionine

and then chased in the

pres-enceorabsenceof100 ,g ofcycloheximideper ml for different periods of time. Virus-specific

polypeptides in the cytoplasmic extracts were

VOL. 19, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

[image:4.505.54.247.498.562.2]

(5)

1058 JAMJOOM, NASO, AND ARLINGHAUS

A B C D E F G H

Prla+b- .'Z---

-Pr2a+b

----Pr3 --- _

Pr4 - 4---- awm...2-.

a's0

dMMM

.~

-.

p30 -- - m -;w"n_E_I__

_

p30/Pr4 0 0.14 0.16 0.49 0.49 1.4

mm~~~~--P1B

p----81

SE

p12

[image:5.505.65.453.76.344.2]

0,68 7' 0.82

FIG. 3. Effect ofcycloheximideonthecleavageofPr4top30.RLV-infectedJLS-V16 cells werepulsedfor

15minwith[35S]methionineinHanks buffer(A) andchased for thefollowingtimes:(B,C)15min;(D, E)30

min;(F,G)60 min;(H, I) 3 h. B,D, F,and Hwerechased incomplete medium;C, E,G,and Iwerechased in

complete medium containing 100

lAg

ofcycloheximideperml.Cytoplasmicextracts wereimmunoprecipitated

withanti-RLV, and the immunoprecipitatewasanalyzedongelsasinFig.2.

then examined after immunoprecipitation by anti-RLV antiserum.

>The

cycloheximide

con-centration used was inexcess of that

required

for fastand maximal inhibition of protein

syn-thesis. Figure3B, D, F, and H shows the

pat-tern of viral polypeptides in control cultures after chases for 15 min, 30 min, 1 h, and 3 h, respectively. This is similar to the control chases shown in Fig. 2. When

cycloheximide

wasadded during the chase (Fig. 3C, E, G, and

I), the cleavage ofPr4 continued atthe same rate as the control for about 30 min before it

wasdrastically reduced.Quantitative

measure-mentsof the ratio ofp30 to Pr4 from Fig. 3 is showninFig.4B. Thisresultsuggeststhat the cleavage ofPr4 requiresthe function ofa pro-tein(s) that is initially present in excess

amounts, ifcompared on a functional basisto

the amount ofPr4 present,butthat isquickly

depleted. Candidates for such a protein would

include a cleavage enzyme or a protein that directsastepinviralassemblythatisrequired

for the cleavage ofPr4. However, other more complexmechanisms forthe effectof inhibition ofproteinsynthesis onthecleavage ofPr4 are

alsopossible.

In contrast to AMD,

cycloheximide

also

re-10

0

rK)_-C

c-8 6

4

2 A

3 0

TimeofChase(Hrs.)

FIG. 4. Effectof AMD and cycloheximide onthe

cleavage ofPr4 top30inthechase. Graphic

represen-tation of

p30/Pr4

of Fig.2andFig. 3 isshown in

(A) and (B),respectively.

duced the cleavage of the precursors

polypep-tides Prla +bandPr2a+b (Fig. 3).

Puromycin,

at 2 x 10-3 M,hadaneffectonthe

cleavage

of viral precursor

polypeptides,

similarto that of

cycloheximide, indicating

thatthe effect is due

to the inhibition ofprotein

synthesis

and not

particularto

cycloheximide

(not

shown).

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.505.273.458.403.550.2]

(6)

Quantitation of the combined effect of

AMD and cycloheximide on the cleavage of

Pr4. The effect ofAMD onthe cleavage of Pr4 could be due either to its direct inhibition of viral RNA synthesis or to the inhibition of a species ofmRNA, with a shorthalf-life, which codes for a protein whose function is required for the cleavage ofPr4. Todistinguish between

these alternatives we have used two ap-proaches, the first of which is to test whether

treatment with AMD can increase the

inhibi-tionof cleavage of Pr4causedbycycloheximide used at levels that are high enough to insure maximal inhibition of proteinsynthesis. Figure 5 shows the amount ofPr4 present in a pulse

(A), its reduction after a subsequent chase in

duplicate (B, B'), the effect of200

Ag

of cyclo-heximidepermladded

during

thechase (C,C'), theeffect of10 ,ugofAMD permladded15 min

before and during the pulse and chase (D, D'), and the effect of the combined AMDand

cyclo-heximide treatment (E, E') on the extent of

cleavage of Pr4 in the chase. Measurements of

theratioofp30 to Pr4 afterthe different treat-ments are indicated beneath the different col-umns.The experimentwasdoneinduplicateto give an idea about the statistical variation in

such quantitative measurements. It can be

seenthat theratioof p30toPr4 waslowerafter the combined treatment with AMD and

cyclo-heximide than aftertreatmentwith eitherdrug alone. This indicates that at least part of the effect ofAMD onthecleavage ofPr4 is notdue

tothe inhibitionof proteinsynthesis.

Although the increase in the inhibition of cleavage ofPr4 after the combined treatment with the two drugs over the separate treat-ments with either drug alone has been

repro-ducible, it seems tobeslight. However, it must

be noted that simple mathematical additivity should notbe

expected

in suchinteracting sys-tems. Moreover, examination of the kinetics of cleavage ofPr4 inthepresenceof thetwodrugs

(Fig. 4A, B) points to the difficulty in

demon-FIG. 5. Combinedeffect ofAMD andcycloheximideontheextentof cleavage ofPr4 top3O.RLV-infected

cellswerepulse-labeled for15minwith1:5S]methionine(A)andparallel cultures,induplicate,werechased

for60 min incomplete growth medium (B, B')orchasedfor60 min incomplete growthmediumcontaining

200mgofcycloheximideper ml(C, C').Inthesameexperimentparallelcultureswerepretreatedwith10pgof

AMD permlfor15min incomplete growthmedium and thenpulselabeledfor15minandchasedfor60 min inthe presenceof10pgofAMDper ml(D, D'),orchasedfor60 mininthe presenceof10pgofAMDplus200

pgof cycloheximideperml(E, E'). Cytoplasmicextracts wereprepared andprocessedasinFig.2.

VOL. 19, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.505.59.447.309.599.2]

(7)

strating large additivity in the inhibition of that cleavage after the combined treatment withthe drugs.

We have also tested the combined effect of AMDand cycloheximide when AMD was added only during the chase, in order to exclude any

possible effectof AMD onthe synthesisof

virus-specific proteins. In this case, the combined effect of the two drugs was similarly greater than the effect of either drug alone (data not

shown).

It must be noted that therateofcleavage of

Pr4 top30in control cultures varies somewhat in different experiments. We have found the

rate ofthe cleavage to be affected by the age and confluency of the cell culture. The varia-tion, however, is not large enough to interfere

withthestatistical significance of the resultsif comparisons are only made among culturesin one experiment and ifthe compared cultures

are similarly prepared and used within a few

hours of each other.

It is to be emphasized, however, that

al-though these results indicate that the

mecha-nisms of action ofAMD andcycloheximideare atleastpartially different, they donotruleout

thepossibility that part ofthe effects ofthese twodrugsonthe cleavage ofPr4maybe dueto acommonmechanism. Thus, itispossible that

part of the effect ofcycloheximideisdue tothe inhibition of the synthesis or migration of viral

RNA. Cycloheximide was found to interfere with total RNA synthesis (14), the synthesis and processing of rRNA in particular (e.g., ref-erences 14, 53) and the nucleocytoplasmic translocation ofRNA (12) in the different sys-tems examined. Similarly, inhibition of RNA synthesis by AMD was found to have an indi-rect inhibitory effect on the initiation of protein synthesis, which is not a result of thedecay of available mRNA (19, 47).

Effect of the length of pretreatment with AMD on the cleavage of Pr4. A better way of testingwhethertheeffectof AMD on the cleav-ageof Pr4 is due to indirect inhibitionofprotein synthesis is to examine the effect of thelength of pretreatment with AMD on the cleavage of Pr4 in a subsequent pulse-chase experiment.

Thus, if AMD causes a depletion of a protein whose function is required for the cleavage of Pr4, then it is expected that the longer the pretreatment with AMD, the slower the rate obtained for thecleavageof Pr4.Figure 6 shows that the extent ofcleavage of Pr4 was

propor-tionaltothelengthof pretreatment with AMD

within30 minof pretreatmentbefore chase. No additional decrease in the rate of cleavage of Pr4, however, was obtained by lengthening the time ofpretreatment up to 8 h. The simplest

interpretation of these resultsisthat the effect

of AMD on thecleavageof Pr4 is not mediated by the depletion of a labile protein with a

short-r

[image:7.505.121.403.411.601.2]

0

FIG. 6. Effect ofthelength ofpretreatment with AMD onthecleavage ofPr4 top30.Insert:RLV-infected

cells werepulse-labeled for15 min inHanks solutioncontaining[35S]methionineandchasedfor60 min in

complete growth medium (A). Parallel culturesweretreatedasfollows: (B)AMD(5,pg/ml)addedduringthe

chase; (C)AMDaddedduringthepulse andthechase;(D)AMDadded15min, (E)1h, (F)4h,and(G)8h

before the pulse and chase.Cytoplasmic extracts wereprocessedasinFig.2. The ratiosof p30 toPr4are

plotted versusthelength ofAMDpretreatment.

on November 10, 2019 by guest

http://jvi.asm.org/

(8)

lived mRNA, but is more likely due to the

direct inhibition byAMDofgenomic viral RNA

synthesis.

The results also indicate that the mRNA(s) for thecleavage enzyme(s)involved in the mat-uration of p30 has a relatively long half-life that is comparable to the half-life of the viral mRNA.

Ifourinterpretation of the effect of AMD on the cleavage of Pr4, as being a result of the inhibition by the drug of the synthesis of gen-omic RNA, is correct, then the above results indicate that there is a pool ofgenomic viral RNA thatisavailable for catalyzingthe cleav-age of newly made Pr4 and that this pool is

depleted in about 30 min. This period is not

likelytoberequired for AMDto exert its inhi-bition of RNAsynthesis, sinceAMD isknown

toact within afewminutes(e.g., reference 44).

Note, however,that sucharesultmeasuresthe pool ofgenomicviral RNA only interms ofits

availability for accelerating the cleavage of newly made Pr4 and not in terms ofits total

amount inthecell.

Theabove results would indicate, moreover,

thattheeffect ofgenomic RNA onthe cleavage ofPr4iscatalyticrather than obligatory, since

cleavage stilloccurred after8hofpretreatment

withAMD, at a rate that was constant between 1 to 8 h, after the drug had attained maximal inhibition of viral RNA synthesis.

Effect ofAMD on the pattern ofsynthesis of viral precursor

polypeptides.

To confirm that AMDdid not haveany effect onthe

syn-thesis of the virus-specific

polypeptides,

pulse-labelingand pulse-chaseexperiments were

per-formed in the presence and absence of AMD. Thedrug (5

,g/ml)

wasaddedto

RLV-infected

cells1hbeforeaddition of

[35S]methionine.

Fig-ure 7showsacontrol15-minpulse labeling (A) and a control 15-min pulse-60-min chase

(C).

AMD-treated pulse and chase cultures are

showninFig.7Band D,respectively. No

signif-icant differences were

observed

in the pulse labeling ofAMD-treated (Fig. 7B) and control (Fig. 7A) cultures. In the chase, however, the usual increase inthe amountofuncleaved Pr4 wasseen in AMD-treated cells. AMD thus did

not affect the pattern of synthesis of the viral

proteins, but only the rate of cleavage of Pr4. Occurrenceof uncleaved Pr4 inempty par-ticles. We have previously reported the occur-renceofuncleavedPr4 inparticlesproduced by JLS-V5 and JLS-V9, but to a much lesser ex-tent inparticlesproducedby JLS-V16 cells (24).

The fact that Pr4 occurs in virusparticles

pro-duced naturally by some cell lines indicates thatacertain amount ofuncleaved Pr4 canbe

incorporated into mature virions. Since AMD reduced therate of cleavage ofPr4 ininfected cells producing RLV, we

tested

whether it is possible to cause a buildup of Pr4 in RNA-deficient particles produced by RLV-infected JLS-V16 cellsinthepresenceof AMD. Figure 8

shows the [35S]methionine-labeled proteins of RLVproduced inJLS-V16 cells inthepresence

of 5 ,g of AMD per ml added 4.5 h before labeling and subsequent virus isolation (Fig.

8A). It is evident that AMD caused a slight

accumulation ofPr4 in virusparticles relative

tocontrolvirus (Fig. 8B). Such a buildup of Pr4 inAMDvirusparticleswasreduced if a period

of chasewasallowedbefore virus isolation (not

shown). Inspection of the virus-specificproteins in the AMD-treated cells at the time ofvirus

isolation (Fig. 8C) showsalargerproportionof Pr4 than was present in virus produced by these cells. This agrees with the idea,

men-tioned above, that the relative amount of

un-cleaved Pr4,inthepresenceof AMD,is

propor-tionaltothe period of chase. Thus, because the

maturation and budding of virus requires a certain time,therelativeamount ofuncleaved

Pr4 should always be higher in cells than in virus.This alsoexplains why it maysometimes be difficult to observe any increase in Pr4 in

AMDvirusparticles (34), especiallyifaperiod ofchase is allowed between labeling

and

virus

isolation. Recent observations have

indicated

that incubation of Nonidet

P-40-disrupted

viri-ons at

37°C

resultsina

preferential

cleavageof

Pr4. This cleavage ofPr4 did not occur if the incubationwascarriedout inthe absenceof the

detergent

(datanotshown). These observations

canbeinterpretedtomeanthat the enzymatic activity

required

for the cleavage ofPr4is pres-entinvirionpreparations, butthey alsomean

thataresidual level ofPr4 may notbe accessi-bletothecleavageenzyme(s) intheintactviral

particles.

As mentioned above, treatment with AMD greatly reduces or inhibits the synthesis

and

incorporationofnewRNAinsubsequently

pro-duced viral

particles.

Concerning the preexist-ingpool of viral RNAinthe cell, Levin et al. (33)found that if the cellswere

prelabeled

with

[3H]uridine

and then treated with AMD for 2 h,

then the viralparticlesproduced lacked the60 to

708

viralRNA.Similarly, Paskindetal. (40) found that 4 h of pretreatment with AMD

greatly depletes the pool of preexisting viral RNA in maturing virions. Thus, the particles

whoseproteins areshowninFig. 8A should be mostly, if notcompletely,devoid of viral RNA, sincetheseparticleswerenewly made after4.5

h of AMD pretreatment. Because these

parti-VOL. 19, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

(9)

4

-.Av#*j

.'

4g_

*- _

4-... .

-U., s_,7S

p30/Pr4

0 0

2.23

1.33

FIG. 7. Effect ofAMD on thepulse labeling ofRLV-specificprecursor polypeptides. Control cultures:

Infected cellswereincubated for 1 h incompletegrowthmedium andthenpulse labeled with[P5S]methionine

for15min(A), andaparallelculture wasthenchasedfor60 min(C). Cytoplasmicextracts werepreparedand

analyzedas in Fig. 2 exceptthat the gels were10%polyacrylamide.AMD treatment:RLV-infectedcellswere

incubated for1 h incomplete growth medium containing5pgofAMD per ml and thenpulselabeledfor15

minwithl'Simethioninein5jAgof AMDperml(B), andaparallelculturewasthenchasedfor60 min in

completegrowth medium containing5 mgofAMDperml(D).

cles contain a level ofp30that issimilarto the

levelof

p30

inuntreated particles,thecleavage of Pr4 to p30 must occur in the formation of empty particles. This confirms the conclusion mentionedabove, that the viral RNAcatalyzes

the cleavage of Pr4 but is not absolutely

re-quiredfor thatcleavagetooccur.

We have considered the possibility that in the absence of genomic viral RNA the viral

capsids are filled with cellular RNA. In this

on November 10, 2019 by guest

http://jvi.asm.org/

[image:9.505.68.459.76.538.2]

(10)

.0

Om

es

-~~_<

*-_

[image:10.505.55.449.76.563.2]

p30/Pr4

1

2

39

0.82

2.57

FIG. 8. Effect ofAMDon theamountofPr4 in virusparticles. RLV-infectedcells wereincubated in the presenceandabsenceofAMD(5

pgIml)

for4.5h. The cultureswererinsedtwicewith Hanks balancedsalt

solution and incubated with[35S]methionineinthepresenceand absenceofAMD(5

p.gIml)

for4hingrowth

mediumcontaining1120 Eagle methionineconcentrationand5%dialyzedfetal calfserum. Cell-associated

viruswithculturefluidviruswaspurified,and anti-RLVimmunoprecipitates fromthecytoplasmicextracts wereprepared. (A)SDS-polyacrylamidegel electrophoresis ofvirusfromAMD-treatedcells; (B) virusfrom

control cells; (C) anti-RLVimmunoprecipitate from AMD-treatedcells; (D) anti-RLV immunoprecipitate

fr-om

control cells. Similar amountsofradioactiveproteinfrompurified virus(A andB)orfrom immune precipitatesofcytoplasmicextraction(C and D)wereappliedtoa6to12%gradientslabpolyacrylamide gel.

VOL. 19, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

(11)

case, it is possible that such cellular RNA would substitute for viral RNA in catalyzing thecleavageof Pr4.Toexaminesucha

possibil-ity, we have labeled cells for3h withmedium containing both [3H]uridine and

[15S]methio-nine, added AMD for 15 min to stop newviral RNA synthesis, and then collected virusat2-h intervals in the presence ofAMD. The same content of [3S]methionine was maintained throughout the experiment, whereas

13H]uri-dine was removed and replaced with excess cold uridine before virus collection. The col-lected virus was purified on a sucrose

gradi-ent, and the trichloroacetic

acid-insoluble

ra-dioactivity of

[35S]methionine

and [:H]uridine

wasmeasured. The

[35S]methionine

radioactiv-ity would be a measure ofthe viral proteins,

according

towhich the

[:'Hiuridine

RNAcanbe normalized. Figure 9 shows that the [3H]uri-dine incorporation in viral particles was

re-duced after AMD treatment. Note that the sH/:5S in the virus was not a reflection of the

ratio inthecelldebris orthemedia (Fig. 9and legend). This result indicates that it is not

likely that cellular RNA quantitatively

substi-tutes forviral RNA inviralparticles after the inhibition ofgenomic viral RNA synthesis

by

AMD.

The RNA present in viral particles after

AMD treatment wasexamined under

denatur-X

E 150

Q0

U

> 100

.)

0 50

-o

0

cr

A) ~ 3H-Ur I.2

N 200_35S-met

B)

3H-Ur

= 0 64 35S-met

ing

conditions

by electrophoresis

in

urea-acryl-amide-agarose gels (Fig.

10). The results

indi-cate that whereas virus collected

during

the

first 2 h after AMD treatment contained 35S

subunit viral

RNA,

virus collected at

subse-quent

periods

wasdevoidofthis RNA. A simi-lar

finding

hasbeenpreviously reported (33).It isinteresting,however, that particles collected

after several hoursofpretreatmentwith

AMD.

containedabroad

peak

of RNA in theregion 15

to 28S. An increase in the 4S RNAcontent of AMD virus has alsobeenreported (33). Part of thiscellular RNAcouldbe present dueto

con-tamination ofthe viral preparation

by

cellular

debris,

whereas another part could be due to

theentrapmentofcellularmaterial (e.g.,

ribo-somes)inside the virusenvelope. However,the

possibility exists that some cellular RNA is

encapsidated in place of viral RNA. Further

work isrequiredtoinvestigate this

possibility.

Effect of AMD on the cleavage of Pr4 in

JLS-V5 cells. The effect ofAMDonthe rate of

cleavage

of Pr4top30wasevenmorenoticeable inJLS-V5 than in JLS-V16cells. JLS-V5 cells

produced virus particles that contained a

higheramountofuncleavedPr4 than ispresent

in virus produced by JLS-V16 cells (24).

Also,

the rate ofcleavageof Pr4 top30wasnoticeably

slower in JLS-V5 than in JLS-V16 cells (not

shown).

Treatment ofJLS-V5 cells with AMD

c) D)

3H-Ur 51 3HHUr =

35S-met 35S-met 9

5 10 15 5 10 151 5 10 15 5 10 15

Fraction Number

FIG. 9. Quantitative measurement of RNA content in AMD viral particles. RLV-infectedJLS-V16cells

were labeledfor3 h with [3H]uridine and [15Slmethionine in medium containing 1/10 Eagle methionine

contentand 5%dializedcalf serum. AMD (5/IgIml)and 6

M.M

unlabeled uridine were then added for 15min. Themedium wasthen replaced with a portion of the same medium lacking

[:'H1uridine.

Medium changes of the same volume weremadeat2 h (A),4h(B),6h(C), and 8 h (D). Virus was collected from the different mediumchangesandpurifiedon a15 to 60%sucrose gradient. Fractions of the gradient were assayed for their

densitiesandtrichloroacetic acid-insolubleradioactivity.Thebarsindicate the viral peakat adensity of1.13

to1.15glcm3. The ratio of:PHIuridine(3H-Ur) and[35Slmethionine(35S-met) ofthe viruspeakis indicated

withinthebars. The 3H-UrI35S-"'e'ratioof the celldebrismaterial thatpelletedthroughthe sucrose gradient was(A)2.1, (B)1.9,(C)2.6, and(D)2.1, whereas in the supernatantgrowthmedia, aftervirusisolation,it

was(A) 0.81, (B)0.51, (C)0.48, and (D) 0.43.

J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:11.505.70.457.389.561.2]

(12)

(Fig. liB) slowed the cleavage rate of Pr4 in

these cellsto agreaterextentthan insimilarly treated JLS-V16 cells (Fig. liD). These results

suggestthat theincreased accumulation ofPr4 in JLS-V5 virus is due to the slower rate of

cleavageofthisprotein intracellularly.A

num-ber of factorscouldcausesuchaneffect,among

them a limitingamount of genomic RNA syn-thesis relativetoviralproteinsynthesis.

However, examination of virusparticles pro-duced by JLS-V5 aftertreatmentwith AMDdid notrevealadrastic accumulation of Pr4,as was seenin the cells(notshown). Thissuggests that

although it ispossibletoincorporate some

un-cleaved Pr4 intomature virions,cleavage ofPr4 top30to alarge extentisprobably an essential

step inassembly.

Effectofcordycepin andcytochalasin B on

therateofcleavage of Pr4. Itwasalso possible

to cause a slowdown of the cleavage ofPr4by

treatingcells with 10 ,ugofcytochalasin B per

ml (Fig. 12D) or200 ,ug ofcordycepin (3'-deoxy-adenosine) perml (Fig. 12C). The reason that wetested thesetwodrugswasthat, duetosome

oftheir known effects, theymayinterfere with the availability of viral RNAinthecytoplasm. Thus, cytochalasin B causes cell enucleation

(42), whichwould interfere with the migration

of nuclear RNA to the cytoplasm. Cordycepin inhibits poly(A) additiontoheterogeneous

nu-clear RNAand,as aresult, intereferes with the

processing and migration of that RNA to the cytoplasm (1). This could apply to viral RNA, whichisknownto contain poly(A) (20, 31, 43).

Cordycepin has been showntoinhibit induction ofmurineleukemia virus production by 5-iodo-2'-deoxyuridine (54), to depress virus produc-tion invirus-producing cell (54), andtoinhibit transformationby murinesarcomaviruses(35, 54).

However, we have not determined that the effects of cytochalasin B

and

cordycepin

onthe

cleavage

of Pr4 are

actually

duetotheir effects

onviral RNAsynthesis, anditispossible that the effects ofthese drugsmay be dueto other cytotoxic effects that these

drugs

may have. Thus,

cordycepin

wasfoundtohaveinhibitory

FIG. 10. RNA in viral particles isolated after

treatment with AMD. Viralparticles werepelleted

from the peakfractions of the gradients describedin

Fig. 9, suspended in TNE, made 1% inSDS, and

extracted with one-fourth the volume ofphenol and

one-fourth the volume ofchloroform. The RNAinthe

aqueouslayerwasprecipitatedwith twice thevolume

of ethanol, boiled for1 min, andelectrophoresedon

2%polyacrylamide-0.5%agarose ureagels(17)for6

hat75V.(A)Marker rRNA; (B) marker RLV RNA;

(C-F)RNAextractedfromthe viralpreparations

A-DofFig. 9,respectively.Arrow indicatespositionof

35S mengovirusRAVA.

VOL. 19, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

[image:12.505.75.221.54.669.2]

(13)

1066 JAMJOOM, NASO, AND ARLINGHAUS

C

a.

Pr2

a +

b <

Pr4

_

..

.4o A4

a.-4

Ja

o- .a*

p3-FIG. 11. Cleavage ofPr4top30 inRLV-infectedJLS-V5cells and JLS-V16 cellstreated withAMD.

RLV-infected JLS-V5cells (B) and JLS-V16 cells (D) were incubated for90min incomplete growthmedium,

containing 5 ,ug ofAMDperml, thenlabeledfor75mininAMDwith[35S]methionineinmediumcontaining

1/10 Eagle's methionine concentration, and finally chasedfor 3 h inAMDcompletemedium.Controlcultures

ofJLS-V5 (A) andJLS-V16(C) wereprocessedinthesamefashionexceptthatAMDtreatmentwasomitted.

Anti-RLVimmunoprecipitatesfromRLV-infectedJLS-V5cells (Aand B)wereanalyzedon6to12%gradient gels;immunoprecipitatesfromRLV-infectedJLS-V16 cells (Cand D) wereanalyzedon10%gels.

effectsontotal RNAsynthesis (35)and protein

synthesis (52). Further studies are needed to

determine the mechanismby which these two drugsinterfere with the cleavage of Pr4.

In our studies, treatment with cordycepin

also resulted in the appearance of a distinct

high-molecular-weight

polypeptide that is

im-munoprecipitable

withanti-RLV antisera(Fig.

A

B

D

flw..

*,

""I

on November 10, 2019 by guest

http://jvi.asm.org/

[image:13.505.67.454.81.555.2]

(14)

A

B

C

D

91]I a ji) <

_~ _*w

_

1. _

pr'I- t <

t)

tia-4

oi3G0

-p30/Pr4

0 -_3-_~~~~~~~l

-_

0'w

4

0.82

__v_s|p_ -5

F-0.75

FIG. 12. Effect of cordycepin and cytochalasin B on the cleavage ofPr4 top30. RLV-infected cells were

pulse labeled for 15 min with[35S]methionine(A)and chased for 60 min (B) .Cordycepin treatment: A culture

waspretreated for1 h with 200

Mg

ofcordycepin per ml and pulse labeled and chased in the presence of

cordycepin (C). CytochalsinB treatment:Aculturewaspretreatedfor1 h with 10 pg ofcytochalasinB per

ml; the cells werecentrifuged for2 min at2,000 xg, resuspended,and thenpulselabeledfor15minand

chased for60min (D). The cytoplasmic extracts were processed as in Fig. 2.

12C, arrow),which is not usually observed. The

identity ofthispolypeptide is under investiga-tion.

DISCUSSION

We have shown thatAMD significantly

re-duced the rate ofcleavage ofintracellular Pr4 to p30 while not affecting the processing of

other p30 precursors, such as Pr3 or Prla +b.

The results are consistent with the conclusion

that AMD exerts its effect on the cleavage of Pr4by preventing the synthesis of viral geno-mic RNA, thereby preventing interaction of viral RNA with viral precursor proteins during virus assembly. This interpretation makes it possible toclassify thecleavage(s)of Pr4 top30 as amorphogenetic cleavage (21) and onethat

is acceleratedbyinteraction with viralRNA.

Pr4, previously called p70, has been detected inmature RLVparticles in variable amounts,

VOL. 19, 1976

on November 10, 2019 by guest

http://jvi.asm.org/

[image:14.505.56.449.78.500.2]

(15)

dependingon thecell type producing thevirus

and the conditions of cell growth (24). It has beenpossibletoslightlyincrease the amountof

Pr4 in virus preparations produced by RLV-infected JLS-V16 cells by pretreatment of

cul-tureswith AMD. However, evenlong pretreat-ments (up to 4h)with AMDdidnotpreventthe almostcomplete cleavage ofPr4 to p30insuch

viruspreparations. Theseresultsindicate that theinteraction with viral genomicRNA is not

absolutely required for the cleavage of Pr4 to

p30 and supportsthe idea that the viral RNA only accelerates the rate ofcleavage ofPr4 to

p30.

Three findings indicate that the effect of AMD on the cleavageofPr4 is not dueto the indirect inhibition ofprotein synthesis. First, thecombined effect of AMD and

cycloheximide

was greaterthanthe effect of eitherdrug alone. Second, the length of the

period

ofpretreatment

with AMD between1 and8hdidnotaffect the subsequentrateofcleavage ofPr4. Third, AMD did not affect the pattern of viral protein

syn-thesis during pulse-labelingexperiments.

Our results agree with the finding reported by other investigators (34, 40) on the presence

oftwo non-equilibrating pools ofvirus-specific

RNAininfectedcells: (i)mRNA thatfunctions

in translation; and (ii) genomic RNA that is

packagedintovirions. BecauseAMDexerts its maximal effect on Pr4 cleavage after treatment

for1h,wededuce that the pool ofviral mRNA,

which has a functional half-life of 7 h, is not

structurally involved in the interaction which accelerates Pr4 cleavage. Thismeansthatit is

only

the genomic RNA which functionsinthe enhancement ofPr4 cleavage.

Anotherpoint inthisstudyconcernsthepool sizeof genomic viral RNA thatisavailable for

accelerating the cleavage of newly made Pr4.

Our results indicate that suchapoolisrapidly depleted. Thus, whereas less inhibition of

cleavage of Pr4 to

p30

was observed in cells pretreated withAMD for 15 to 30 minthanfor 60min, longerpretreatment with AMD (up to 8

h) hadnofurther inhibitory effect thandida 1-h treatment on the rate ofcleavage of Pr4 to

p30.

However, these results do not measure the rate of depletionof total genomic viral RNA in the cell, which is known to require several

hours(about 4) to bedepleted (40). We suggest

that this latter pool of genomic viral RNA is

already associatedwithviralproteinsandthus cannotaffect thecleavage of newly made Pr4.

Our results suggest a long half-life of more

than 8 h either for the enzyme(s) involved in theprocessing of Pr4 or for the mRNA(s) coding

for this enzyme(s). Thedecreaseinthe rate of

cleavageof Pr4 after treatment with inhibitors

ofprotein synthesis may indicate that the en-zymes themselves donot havea long half-life.

It is possible, however, that inhibitors of pro-teinsynthesis interferewiththecleavage ofPr4

by a more complex mechanism that oversha-dows the direct inhibition of the synthesis of such processing enzymes.

Thedependence of the cleavageofprecursors to capsid proteins on the interaction with the viral genome is a commonpattern inviral

as-sembly (e.g., reference 22, 25, 46). Such a de-pendencemay stemfrom the role thatcleavage playsin the packaging of the genome. For in-stance, in T4the

packaging

of theviral DNAis

accompanied

by the cleavage of the core pro-teins p22and IPIII (30). Moreover, thecleavage ofthe major head proteinp23leads to a

struc-turaltransformation that resultsin an expan-sion of the empty

head,

which is

probably

re-quired

for the

packaging

ofthe DNA(29).

How-ever, a

relationship

between

cleavage

and

packaging

of thegenomeisnotclearinseveral other systems. For example, in

poliovirus

the cleavage of theprecursorproteinvp0tovp2and

vp4 seems to require the interaction with the viral RNA, since it does not occur in empty

capsids. However, this cleavageis not

required

to bring about the initial association between the viral RNA and the

capsid

proteins, since

RNA-containing provirion particles have been foundtocontainvp0(15). InRLV,asmentioned

above,

the cleavage of Pr4 to p30,

although

catalyzed by

the genomic

RNA,

still occursin

theformation ofempty cores, thereby suggest-ingthatthe role of this cleavage isnot in pack-aging of thegenome. However, examination of the cleavageproducts of Pr4, together with the

nature of the influence ofgenomicRNAonthe cleavage ofPr4 that was discussed above, has leadus to aconsideration ofamodel thatpoints to several advantages of having adependence of therateofcleavage ofPr4 onthe interaction withthe genomicRNA. Wediscuss this model

inthe sectionbelow.

Theresults presented above have suggested

tousthat the interaction between thegenomic

RNA and Pr4 is a prominent step in RLV

as-sembly. Suchinteraction islikelytoresultfrom a direct affinity between Pr4 and the viral RNA, althoughanindirectinteraction is

possi-ble. Although directinteraction ofpurifiedPr4 topurified viral RNA hasnotyetbeen demon-strated, several recentfindings havepointedto the likelihoodthat such interaction does occur.

Thus, Pr4 has been shown to contain two pro-teinsthat have affinityfor the RNA:

plO,

which

isfoundinassociationwith theviral RNA (16)

andwhich is richinarginine andlysine, anda

protein that migrates onguanidine

on November 10, 2019 by guest

http://jvi.asm.org/

(16)

VOL. 19, 1976

ride agarosecolumns in the p12 region (L. Kar-shin, inpreparation). (The p12 region in

SDS-polyacrylamide

gels contains another compo-nent, designated pl2E, which migrates on

GuHCl agarose columns in the void region. This latter protein contains a

methionine-con-taining tryptic peptide present in pl5E and

Pr2a+b, the precursor of the viralgp69/71

[38].)

The othercomponents ofPr4arep30 andp15 (2;

Arlinghaus et al., in press). Sen et al. have demonstratedastrain-specific binding ofp12 to

genomic RNA (45). Davisetal. have

described

abasic proteininRLVwithamolecularweight of 9,800 that binds to viral RNA and to other heterologous single-stranded RNA and DNA

(9).Thisprotein isthoughtobethesame asthe

plO thatis a component ofPr4. Inaddition, a

70,000-dalton

precursorproteinthatshares

an-tigenic specificities with RLV p30 has been foundinmouse L929cells, whichproducealow level ofa C-type virus. This protein exhibits DNA-binding properties and hasbeenisolated by affinitychromatography on

single-stranded

DNA columns (39). RLV p30 does not bind to

single-stranded

DNA under the same condi-tions (39). Recently, this precursor protein has also been found to share antigenic

determi-nantswith RLVp15, p12, andplO(C.W. Long,

personal

communication)and is thusvery simi-lar to our Pr4. Theseresults suggestthat Pr4 probably interactsdirectly with the RNA. Our results provide evidence for the occurrence of

aninteraction betweenthe viral RNAandPr4 incells. Whatarethepossible advantages ofan

arrangement in viral assembly in which the cleavage ofa precursor

polypeptide

is

affected

by the interaction of that precursor with the genomic RNA? To answer this question, we

propose the model

depicted

in Fig. 13 for the

CLEAVAGE OF RLV

p30

PRECURSOR 1069

assembly

of the RLV core. In this

model,

a

"procapsid" structure, consisting of subunits

composed

of

uncleaved

Pr4, is

formed before

interactionwith the viral RNA. The formation ofthisstructureisdetermined only by the

spec-ificity of the interactions of the protein sub-units. ThePr4components thatprobably inter-act with the RNA, e.g., plO and p12, are de-pictedontheinside,and the other components,

e.g.,

p30,

are on the outside. Entrance of the RNA into this structure results in

neutraliza-tion of the positive charges of the basic region plO of the molecules in the center and allows the proteinsubunitstocome in acloser

proxim-ity and in a more favorable orientation. This facilitates the cleavage ofPr4. Cleavage ofPr4 removes the electrostatic (and

possibly

the

steric)hindranceand allows the

p30

subunitsto

forma morestable association. Cleavage ofPr4 occurs both inthe presence ofviral

RNA,

and,

at a slower rate, in its absence. Specificity is

determined by theinteractionof theRNA

bind-ing components (e.g.,

p12

and

plO)

with the viral RNA. In the absence of the viral RNA, charge neutralization is

achieved

by

counter-ions orbyspeciesof host cell RNA.

Incorpora-tion of hostcell RNA is not quantitative (i.e.,

doesnotresultinanRNA/proteinratiosimilar

totheratioof the normal viralparticle), but the

extentofincorporation and the species of RNA

incorporated

are not yet clear. Alarge portion of Pr4 must be cleaved in order to allow the

majorcapsid protein subunitstoassume aclose proximity to permit the formation ofa stable

particle,

although a number of

uncleaved

Pr4

moleculescanbe incorporated.

Inadditiontotheevidenceontheinfluence of RNA on the cleavage of Pr4, further evidence forthismodelcanbe derived from the present

Pr4 (Uncleaved)

"Procopsid"Structure

Mature Proteins p30, p15, plO, p/2

Assembled Core

[image:16.505.116.396.480.596.2]

VRNA

FIG. 13. Hypotheticalmodelof RLVcoreassembly.Thismodelisbasedonthepresumptive evidencethat

the viralgenomicRNA interactswiththeuncleaved precursorPr4ratherthan with thecleavedmatureviral

proteins. Protein subunits ofPr4forma loose andopen arrangement which ispenetrated by thegenomic

RNA. The RNA binding proteins (e.g.,p12 and plO)aredepictedontheinside. Thepositivecharge(ofp1O)

prevents theuncleavedPr4 subunitsfrom formingacompact structure.Charge neutralization (e.g., by the

RNA)facilitatesthecleavageofPr4. Cleavageisrequiredtoallowadjacent subunitstoget incloseproximity

so as toformastablecompactstructure.

on November 10, 2019 by guest

http://jvi.asm.org/

(17)

study. Thus, the proposal that the protein-pro-tein interaction is the only major factor that

determines the core structure is based on the very close similarity in composition between

particles containing or lacking the viral RNA

(Fig. 8; 34). Capsid assembly in several viral

systems has been shown to be determined

mainly by protein-protein interaction (22, 25). The importance of cleavage for the formationof

amatureparticle is evidenced bythedifficulty

ofbuildingahigh ratio ofPr4 top30in mature

virus particles, despite several hours of

pre-treatment with AMD, even in virus produced

by cells(e.g.,JLS-V5) that exhibitahigh

inter-cellular Pr4/p30 ratio. No evidence exists yet regarding the proposed procapsid structure,

which may ormay notbe stable enoughto be

isolated.

This. model offers several advantages for

viral assembly. First, the covalent linkage of the majorcoreprotein, p30,toproteins havinga

high affinity for the RNA, e.g., the basic

pro-tein plO, would facilitate encapsidation. This

would beparticularly essential if p30 hadalow

affinityfortheviral RNA. Thearrangementof a basic region in the capsid or core protein

seemstohave been conservedduringevolution.

Thus, inRNA bacteriophages there isa

cluster-ing ofbasic and acidic amino acids (26). Second, the arrangement of the positively charged proteins on the inside results in the

presenceofacenterwithastrongattraction for

theRNA.This arrangement, atthesametime,

ensuresthat theprotein subunits donotforma

compact structure. The existence of an open

structure is ofimportance for thepackaging of

RNA viruses because of the secondary struc-ture of theRNAgenome (26). The assemblyof

DNA viruses,ontheother hand, takes place by the formation of an empty capsid which

ex-pands in size as packaging of the genome

pro-ceeds (23, 29). A similar arrangement of the basic amino acids on the inside of the capsid

structure has beenproposed by Matthewsand

Cole (36) for the capsid off2 bacteriophage. Third, since RNA catalyzes the cleavage of Pr4 and since cleavage is, to a large extent, visualized as required for assembly, the model

predicts that the formation of RNA-containing cores proceeds faster than the formation of

empty cores. This would be of obvious

advan-tage for infectious core formation. Since our

results (Fig. 1) and earlier results by Levinet al. (33)donotindicateasignificant reductionin

the rate of maturation of virions after

treat-mentwithAMD, itmaybepostulated thatcore

formation is not the limitingstep in virus

as-sembly. However, recent resultsby Levin and

Rosenak show areduction, after treatment of

cells with AMD, in the amount of produced

virus, as more accurately determined by radio-immune assay of p30 and the level of activity of the reverse transcriptase (34).

The fourthadvantage of the model is that, if there is low affinity between the major core protein and the ribonucleoprotein complex

in-side it, the genome can then be free to move

inside thecapsid. Such movement may be im-portant for assuming a structurally favorable arrangement. Structural rearrangement of ge-nomic RNA has been described to occur in a murine sarcoma-leukemia virus after matura-tion (11). Moreover, the lowaffinityofthe

ma-joI core protein to the RNA complex would

facilitate the uncoating and release of the ge-nome in the next cycle of infection.

Insummary, the advantages in this model of

havinggenomic RNA intereact with a precur-sor to coreproteins, rather thantothe mature

cleaved proteins themselves, are to ensure the existence ofan open procapsid structure that

can be easily penetratedby the genomic RNA, to make full core formation faster than the

formation ofempty cores, and to allow

encapsi-dation of the RNAin acoat structure for which itmighthave low affinity, thereby facilitating its structural rearrangement and subsequent

release. Further work is required to test the

differentaspects of the model.

Finally, our present results suggest that it may be useful to consider a possible role of genomic RNA in the cleavage ofviral precur-sors in other situations. For example, Eisen-man et al. (13) reported the presence of a stable

precursor-like polypeptidein Rous sarcoma

vi-rus-transformed hamster cells and suggested that the lack of cleavage ofsuch a precursor

mightbe due to theabsence of the appropriate

proteolyticenzymes in thehamster cells.

Simi-larly, the cleavage of precursor polypeptides proceeds slowlyin some cell lines, e.g., mouse L929cells, whichproduce low levels of C-type

particles (39), or, as we noticed, in C243 cells

(5), which produce noninfectious C-type parti-clesthat are devoid of 60to70S viral RNA (41)

(unpublished observation). If genomic viral RNA plays a role in the cleavage of oncorna-viral proteins, then it would be useful to con-sider the absence of viral genomic RNA, as

opposed to viral mRNA, as a possible mecha-nism that plays a role in the cleavage of the

precursor polypeptides.

ACKNOWLEDGMENTS

Thiswork wassupportedinpartby PublicHealth Serv-ice contractCP-61017 and grantCA-15495, bothfrom the National Cancer Institute, andagrant from The Robert A. WelchFoundation (G-429). One of us(G.A.J.) isa

on November 10, 2019 by guest

http://jvi.asm.org/

(18)

VOL. 19, 1976

toral fellow supported by the Riyadh University, Saudi Arabia.

Wethank J. Levin and C. Long forcommunicating re-sultsbeforepublication, J. Davis for useful discussions, and JamesSyrewicz for excellent technical assistance.

LITERATURE CITED

1. Adesnik, M., W.Salditt, W. Thomas, and J. E. Darnell. 1972. Evidence that all messenger RNA molecules (except histone messenger RNA) contain poly(A) se-quencesandthat the poly(A) has a nuclear function. J.Mol. Biol. 71:21-30.

2. Arcement, L. J., W. L. Karshin, R. B. Naso, G. A. Jamjoom, and R. B. Arlinghaus. 1976. Biosynthesis ofRauscher leukemia viral proteins: presence ofp30 and envelope p15 sequences in precursor polypro-teins.Virology 69:763-774.

3. Bader, J. P. 1970. Synthesis of the RNA of RNA-con-taining tumor viruses. I. The interval between syn-thesisandenvelopment. Virology 40:494-504. 4. Bases, R. E., and A. S. King. 1967. Inhibition of

Rauscher murineleukemia virus growth in vitro by Actinomycin D.Virology 32:175-183.

5. Bassin, R. H., L. A. Phillips, M. J. Kramer, D. K. Haapala, P. T. Peebles, S. Nomura, and P. J. Fis-chinger. 1971.Transformation of mouse 3T3 cells by murine sarcoma virus: release ofvirus-like particles intheabsence of replicating murine leukemia helper virus. Proc. Natl.Acad. Sci. U.S.A. 68:1520-1524. 6. Bonner, W.M., and R. A. Laskey. 1974. A film

detec-tionmethod fortritium-labelledproteins and nucleic acids inpolyacrylamidegels. Eur. J. Biochem. 46:83-88.

7. Butterworth, B.E., L.Hall, C. M. Stoltzfus,and R. R. Rueckert. 1971. Virus-specific proteinssynthesizedin

encephalomyocarditisvirusinfected HeLa cells. Proc. Natl. Acad. Sci. U.S.A. 68:3083-3087.

8. Casjens, S.,and J.King. 1975. Virusassembly. Annu. Rev.Biochem.44:555-611.

9. Davis, J., M. Scherer, W. P. Tsai, and C. Long. 1976. Low-molecular-weight Rauscher leukemia virus pro-tein with preferential binding for single-stranded

RNA and DNA. J. Virol. 18:709-718.

10. Duesberg, P. H., and W. S. Robinson. 1967. Inhibition of mouseleukemia virus (MLV) replicationby Acti-nomycin D.Virology 31:742-746.

11. East, J. L., P. T. Allen, J. E. Knesek, J. C.Chan, J. M. Bowen, and L. Dmochowski. 1973. Structural rear-rangement and subunit composition of RNA from released Soehner-Dmochowskimurinesarcoma viri-ons.J. Virol. 11:709-720.

12. Eckert, W. A., W. W. Franke, and U. Scheer. 1975. Nucleocytoplasmic translocation of RNA in Tetrahy-menapyriformisand its inhibition by ActinomycinD andcyclohemixide. Exp.Cell Res. 94:31-46. 13. Eisenman, R., V. M. Vogt, and H. Diggelman. 1974.

Synthesis of avian RNA tumor virusstructural pro-teins. Cold Spring Harbor Symp. Quant. Biol. 39:1067-1075.

14. Fakan, S. 1971. Inhibition of nucleolar RNP synthesis by cycloheximide asstudiedby high resolution ra-dioautography. J. Ultrastruct. Res. 34:586-596. 15. Fernandez-Thomas, C. B., and D. Baltimore. 1973.

Morphogenesis of poliovirus. II. Demonstration of a new intermediate, theprovirion. J. Virol. 12:1122-1130.

16. Fleissner, E., and E. Tress.1973. Isolation of a ribonu-cleoprotein structure fromoncornaviruses. J. Virol. 12:1612-1615.

17. Floyd, R. W., M. P. Stone, and W. K. Joklik. 1974.

Separation ofsingle-stranded ribonucleic acids by

acrylamide-agarose-urea gel electrophoresis. Anal. Biochem.59:599-609.

18. Gillespie, D., S. Marshall, and R. C. Gallo. 1972. RNA of RNA tumor viruses contains poly(A). Nature

(Lon-don) NewBiol. 236:227-231.

19. Goldstein, E. S., and S. Penman. 1973. Regulation of protein synthesis in mammalian cells. V. Further studies on the effect of Actinomycin D on translation control inHeLa cells. J. Mol. Biol.80:213-254. 20. Green, M., and M. Cartas. 1972. The genome of RNA

tumorviruses containspolyadenylic acidsequences. Proc. Natl. Acad. Sci. U.S.A. 69:.791-794.

21. Hershko, A., and M. Fry. 1975. Post-translational

cleavage of polypeptide chains: role in assembly. Annu.Rev. Biochem. 44:775-797.

22. Hohn, T., and B. Hohn. 1970. Structure and assembly of simple RNA bacteriophages. Adv. Virus Res. 16:43-98.

23. Hohn, B., M. Wurtz, B. Klein, A. Lustig, and T. Hohn. 1974. PhagelambdaDNA packaging in vitro. J. Su-pramol. Struct. 2:302-317.

24. Jamjoom, G. A., W. L. Karshin, R. B. Naso, L. J. Arcement, and R. B. Arlinghaus. 1975. Proteinsof Rauscher murine leukemia virus: resolution of a 70,000dalton, non-glycosylatedpolypeptide contain-ingp30 peptidesequences. Virology68:134-145. 25. Klug, A. 1972. The polymorphism of tobacco mosaic

virusproteinanditssignificance for theassemblyof the virus, p. 207-215. InCiba FoundationSymp. 7: Polymerization inbiological systems, p. 207-215. El-sevier,Amsterdam.

26. Knolle,P., and T. Hohn. 1975. Morphogenesis ofRNA

phages, p. 147-201.In X. Zinder(ed.),RNAphages.

ColdSpringHarborMonogr.Ser.ColdSpringHarbor Laboratory, ColdSpringHarbor,N.Y.

27. Korant, B. D. 1975.Regulationofanimalvirus

replica-tionby protein cleavage, p. 621-644. In E. Reich(ed.),

Cold Spring Harbor symposium on control of cell proliferation, vol.2:Proteasesandbiologicalcontrol. ColdSpringHarborLaboratory, ColdSpringHarbor,

N.Y.

28. Laemmli, U. K. 1970. Cleavage ofstructural proteins during the assembly of the head ofbacteriophageT4. Nature(London)227:680-685.

29. Laemmli, U. K., L. A. Amos, and A. Klug. 1976. Corre-lation betweenstructural transformation and cleav-age of the major head protein of T4bacteriophage.

Cell 7:191-203.

30. Laemmli, U.K., and M. Favre. 1973. Maturation of the head of bacteriophage T4. The DNA packaging

events.J. Mol. Biol. 80:575-599.

31. Lai, M. M. C.,and P. H. Duesberg. 1972.Adenylic

acid-rich sequences inRNAsof Rous sarcomavirusand

Rauscher mouse leukemia virus. Nature (London) 235:383-386.

32. Laskey, R.A.,andA.D.Mills.1975.Quantitative film detection of3H and 14C in polyacrylamide gels by

fluorography. Eur. J. Biochem. 56:335-341. 33. Levin, J.G.,P. M.Grimley,J. M.Ramseur,and I. K.

Berezesky.1974. Deficiency of60 to70SRNAin mu-rine leukemia virus particles assembled in cells

treatedwithactinomycinD.J. Virol. 14:152-161. 34. Levin, J. G., and M. J. Rosenak. 1976. Synthesis of

murineleukemia virus proteinsassociatedwith viri-ons assembled in Actinomycin D-treated cells: evi-dence forthe persistence ofviral messenger RNA. Proc.Natl. Acad. Sci. U.S.A. 73:1154-1158. 35. Lovinger, G. G., R. A. Klein, R. V. Gilden, and M.

Hatanaka. 1973. The effect of cordycepin on cell transformation by RNA tumor viruses. Virology

55:524-526.

36. Matthews, K. S., andR. D.Cole. 1972.Shellformation bycapsidprotein off2bacteriophage. J. Mol. Biol. 65:1-15.

37. Naso, R. B., L. J.Arcement, and R. B. Arlinghaus.

on November 10, 2019 by guest

http://jvi.asm.org/

(19)

1975. Biosynthesis of Rauscher leukemia viral

pro-teins.Cell4:31-36.

38. Naso, R. B., L.J. Arcement, W. L. Karshin, G. A. Jamjoom,and R. B.Arlinghaus. 1976. A fucose-defi-cient glycoprotein precursor to Rauscher leu' mia

virusgp69/71.Proc.Natl. Acad. Sci. U.S.A. 73:2326-2330.

39. Oroszlan, S., C. W. Long, and R. V. Gilden. 1976. Isolationofamurine type C virusp30precursor

pro-tein by DNA-cellulose chromatography. Virology

72:523-526.

40. Paskind, M. P., R. A. Weinberg, and D. Baltimore. 1975. DependenceofMoloneymurineleukemia virus productiononcell growth. Virology67:242-248.

41. Phillips,L. A., V.W. Hollis, Jr., R. H. Bassin, and P. J. Fischinger. 1973. Characterization ofRNAfrom non-infectiousvirionsproduced bysarcoma positive-leukemianegativetransformed 3T3 cells.Proc. Natl. Acad. Sci. U.S.A. 70:3002-3006.

42. Prescott, D. M., D. Myerson, and J. Wallace. 1972. Enucleationofmammalian cellswith cytochalasin B. Exp. CellRes. 71:480-485.

43. Ross, J., S. R. Tronick, and E. M. Scolnick. 1972. Polyadenylate-rich RNAinthe 70S RNA of murine leukemia-sarcomavirus.Virology49:230-235. 44. Scherrer, K., H. Latham, and J.E.Darnell.1963.

Dem-onstrationofanunstable RNA andofa precursorto

ribosomal RNAinHeLacells. Proc.Natl. Acad. Sci. U.S.A.49:240-248.

45. Sen, A., C. J. Sherr, and G. J. Todaro. 1976. Specific binding of the type C viral core protein p12 with purified viral RNA.Cell 7:21-32.

46. Showe, M. H.,and E. Kellenberg.1975.Control mecha-nisms in virusassembly,p.407-438.InControl

proc-esses in virusmultiplication, 25th Symp. Soc. Gen. Microbiol. p. 407. Cambridge University Press, Cam-bridge, England.

47. Singer, R.,and S. Penman. 1972.Biological stabilityof HeLa cell mRNA inActinomycin. Nature(London) 240:100-102.

48. Smith, A. E. 1975. Control of translation of animal virusmessenger RNA, p. 183-223. In Control proc-essesinvirusmultiplication, 25th Symp. Soc. Gen. Microbiol. Cambridge University Press, Cambridge, England.

49. Syrewicz, J. J., R. B. Naso, C. S. Wang, and R. B. Arlinghaus. 1972. Purification oflarge amounts of murineribonucleicacid tumor viruses in roller bottle cultures. Appl. Microbiol. 24:488-498.

50. Temin, H. M. 1963. The effects of Actinomycin D on growth of Rous sarcoma virus in vitro. Virology 20:577-582.

51. Vogt, V. M., and R. Eisenman. 1973. Identificationofa large polypeptide precursor of avian oncornavirus proteins. Proc. Natl.Acad. Sci. U.S.A.70:1934-1938. 52. Weiss, S. R., andM.A.Bratt. 1975.Effectofcordycepin (3'-deoxyadenosine) on virus-specific RNA species

synthesized in New Castle disease virus-infected cells.J.Virol. 16:1575-1583.

53. Willems, M., M. Penman, andS. Penman. 1969. The regulation ofRNA synthesis and processing in the nucleolus during inhibition of protein synthesis. J. Cell Biol. 41:177.

54. Wu, A. M., R.C. Ting, M. Paran, andR.C.Gallo.1972. Cordycepin inhibits induction of murine leukovirus production by 5-iodo-2'-deoxyuridine. Proc. Natl. Acad. Sci. U.S.A. 69:3820-3824.