Poliovirus protease 3C (P3-7c) does not cleave P220 of the eucaryotic mRNA cap-binding protein complex.

(1)

0022-538X/85/080489-05$02.00/0

Microbiology

Poliovirus Protease 3C (P3-7c) Does Not Cleave P220 of the

Eucaryotic

mRNA

Cap-Binding

Protein

Complex

KEVIN A. W. LEE,' ISAAC EDERY,' RONNIE HANECAK,2t ECKARD WIMMER,2 ANDNAHUM SONENBERGl 3*

DepartmentofBiochemistry' and McGill Cancer Center,3 McGill University, Montreal, Quebec, CanadaH3GIY6, and Department of Microbiology, School of Medicine, State University ofNew YorkatStonyBrook, Stony Brook,

New York J17942

Received24 January1985/Accepted 19April 1985

Infection of HeLa cells by poliovirus results in proteolysis of the large subunit(P220) of the cap-binding protein complex. This is believed to cause the rapid shut-off of host protein synthesis during poliovirus infection.Inthiscommunicationweexamined thepossible involvement ofpoliovirus proteins3C(aproteinase)

and 2C in cleavage of P220. Using antisera against these two viral polypeptides, we wereunable to inhibit

proteolysisof P220 inanin vitroassay. Theseresults indicate that viralproteins3Cand 2C arenotdirectly

involved incleaving P220 and hence do notcauseshut-off of cellular protein synthesis.

The mechanism by which poliovirus inhibits HeLa cell protein synthesis,asubject of intense study for severalyears

(4), has recently been clarified in some respects. In vivo, polioviruscauses arapid and extensive inhibition of cellular (capped) mRNA translation, whereas translation of the naturally uncapped poliovirus RNA proceeds with high efficiency (1). Many lines of evidence have demonstrated thatthe failureofcappedmRNAsto enterpolysomes is due to avirally induced defect in the translation initiation ma-chineryof the host cell (forarecentreviewseereference4). Thefact thatcellextractsprepared frompoliovirus-infected

cellsarealsospecifically deficient inanactivity requiredfor capped mRNA translation (3, 11, 12, 14), andthusfaithfully

mimicthe in vivosituation, providedanassayfor the factor

which isinactivated. Consequently, it has been shown that thecap-binding protein (CBP) complex (also termed eIF-4F or CBP II) can restore translation of capped mRNAs in extracts from poliovirus-infected cells (3, 14) or in a re-constituted translationsystemfrompoliovirus-infected cells (5), and thus it isthought, that poliovirus achieves inhibition ofcellular protein synthesis by somehow inactivating the CBPcomplex.

The CBP complex consists of three polypeptides, the

24,000-molecular-weight CBP (also termed CBP I or eIF-4E), eIF-4A, and an -220-kilodalton (kDa) polypeptide (2, 7). Etchisonet al. (6) havepresented evidence which indi-cates that the 220-kDa polypeptide is cleaved by a virus-dependent protease, yielding cleavage fragments of -130 kDa.Subsequently,weisolatedamodified CBP complex (by using m7GDP affinity chromatography) from poliovirus-infected HeLa cells, which contains proteolytic fragments of P220 withapparentmolecularmassesof 110to130kDa (10). Although it remains to be demonstrated directly that proteolysis of P220 results in loss of activity of the CBP complex, itappearsmostreasonable that proteolysis of P220 is thecause of inhibition of cellular protein synthesis.

It is currently not known whether the protease activity

*Corresponding author.

tPresent address: Department of Molecular Biology and Bio-chemistry, University of California, Irvine, CA 92717.

which cleaves P220is encoded by the poliovirusgenomeor whether it is an induced cellular activity. The poliovirus

genome encodes at least one protease activity (protein 3C [8]) and mayhaveprotease activities mapping elsewhere in

the genome. Protein 3C (formerly known as P3-7c [13]) is

knownto process the viral polyprotein toproduce mostof the viral polypeptides by cleavage between the Gln-Gly amino acid pairs (8). There are, however, other cleavage sites(one Asn-Ser andtwoTyr-Gly) thatarenotcleaved by protein 3C. The protease(s) responsible for these other cleavage events is unidentified. It has been shown (8), however, that the activity does not appear to reside in

protein 2C (formerly known as P2-X [13]), as had been previously claimed (9).

Several studies have shown that the virus-dependent activity which is responsible for inactivating the CBP com-plex and consequently inhibiting cellular translation can be assayed in vitro. Originally, Rose et al. (12) showed that translational restoring activity (i.e., the activity which can restore capped mRNA function in extractsfrom poliovirus-infected HeLacells)can beslowly inactivatedupon incuba-tion with a cell extract from poliovirus-infected cells. We have confirmed these results (11) and have also shown that crude initiation factorpreparations from infected cells have anactivity whichcan slowlyimpair thecap-binding activity of polypeptides present in crude initiation factors from uninfectedcells(11). These observationsareconsistent with thecontention that the mechanismby whichcellularprotein synthesis is inhibited in poliovirus infected cells is catalytic innatureand isnotaresultof steric hindrancebysomeviral

protein. Finally, Etchison et al. (6) have shown that the protease activity which cleaves P220 of the CBP complex can be detected in crude initiation factors from poliovirus-infectedcells.

Using polyclonal antisera against P220 of sheep erythro-cytes,weprobedextractsfromeitheruninfected(U-S10)or

poliovirus-infected (I-S10) cells and obtained the same re-sults as Etchison et al. (6). The assay and immunoblotting procedures were as follows. HeLa S3 cellswere grown in mediumsupplementedwith5% calfserum.Poliovirustype1

(Mahoney strain) infection of HeLacells was performedas

489

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

490 NOTES

12 3 4 5 6 7 8

P220--FIG. 1. Invitroassayfortheproteasewhich cleaves P220 of the CBPcomplex. Lanes: 1, 10,ulofS10extractfrom uninfected cells

(U-S10) and 5 of S10 extract from poliovirus-infected cells (I-S10); 2, 10 ,ul of U-Si0 and 2.5 ,u ofI-S10;3, 10 pu1 of U-Sl0 and 1 pIofI-S10; 4, 10 ,ul of U-Sl0 only; 5,10 1.IofU-Sl0and 5 pu1 of I-S10 which were not preincubated; 6, a sample of the reaction

mixture usedinlane5, after 30 min ofincubationat37°C;7 and 8, U-Sl0andI-S10, respectively, whichwerenotincubated.

previously described with 10 to 20 PFU per cell, and preparation of cellextractswas aspreviously described (11, 12). Extractsweremixedas indicated below and incubated for30 min at 37°C. Reactions were stopped by addition of electrophoresis sample buffer and resolved on 10% poly-acrylamide gels containing sodium dodecyl sulfate. After electrophoresis, polypeptides were transferred to nitro-cellulose paper according to the method of Towbin et al., (15). Nitrocellulose blotswere presaturated with 1% bovine serumalbumin inTBS(10 mM Tris [pH 7.5], 150 mM NaCl) for 30minatroom temperature. Blots were incubatedwith anti-P220 antiserum in 1% bovineserumalbumin in TBSfor 3 hatroomtemperature. The antiserawereraised in rabbits against sheet CBP complex injected intradermally (16) as describedelsewhere (10) and werediluted 2,000-fold in 1% bovineserumalbumin in TBSbeforeuse. Blotswere subse-quently washed with six changes of TBSover aperiod of 30

min, followed by incubation with peroxidase-conjugated

goat anti-rabbit immunoglobulin G (Boehringer Mannhein Biochemicals) diluted 1,000-fold in 1% bovine serum albu-min in TBS for 1 h. Immunoreactive species were then visualized by staining with diaminobenzidene as described elsewhere (15).

Figure 1 is an immunoblot showing that the anti-P220

serumreactedmainly witha220-kDapolypeptidepresentin

the U-Sl0 extract (lane 7), whereas this polypeptide was absent in the I-S10 extract (lane 8). Instead, the antisera recognized in I-SlO polypeptides that were presumably cleavage products of P220 with molecular weights between 110,000 and 130,000 (lane 8). In an attempt to assay the protease activity in vitro, we mixed U-Sl0 and I-S10 and monitoredproteolysis of P220 by probing with anti-P220. As a control we incubated U-Sl0 alone for 30 min and found thatP220 is stable under these conditions (lane 4). We have repeated this experiment with many different cell extracts

and have never detected degradation of P220, even after

longer incubation times (data not shown), an observation suggesting that P220 is not intrinsically unstable. Lanes 1

through 3 show mixtures ofU-Sl0withdecreasingamounts

ofI-S10. The results show thataratio ofU-Sl0toI-S10of

2(10 ofU-Sl0 and 5 ,ul ofI-S10, in which bothextracts

contained

equal protein

concentrations as determined by

A280/A260

readings) was sufficient to completely proteolyze P220 after30minofincubation

(lane 1).

The samebands of 110 to 130 kDa in lane 8arealsoseeninlanes 1

through

3. It

is reasonableto

assumne

that the

disappearance

ofP220from

U-S1O upon

mixing

with

I-S10

occurred because the P220

was cleaved to yield the smaller products. In accord with

thisassumptionwehaveperformedmorequantitativeassays

using 125I-labeled protein

Aand have detectedappearance of

the

putative cleavage products (data

not

shown).

With a U-S1O-to-I-S1O ratio of

4,

there was almost

complete

proteolysis

ofP220aftera30-minincubation

(lane

2). Witha U-S1O-to-I-S1O ratio of 10, there was

clearly

some P220 remaining after 30min(lane 3). Thus,aU-S1O-to-I-S1O ratio of 4was

approximately

the

endpoint

for titration of theI-SlO extract

against

theprotease

activity

underourassay

condi-tions. Lanes 5 and 6 show that the loss of P220 is time

dependent.

Lane 5 shows P220-related antigens after a

simple mixing

of U-S1O and I-S10 without

incubation,

and lane 6 shows a

portion

ofthe same

sample

after30 minof incubation. The

intensity

ofthe P220

antigen

in lane 5 is

weaker than thatofthe

P220

cleavage

products, although

the

opposite

might

be

expected

based ontheamountsof

U-Sl0

and I-S10 added. Thisapparentcontradiction most

probably

resulted from the fact that P220 was less

efficiently

trans-ferred to nitrocellulose paper than the

cleavage products

under our transfer conditions. It is also

possible

that the

cleavage

products

reacted better with the

antibody

on nitro-cellulose paper than P220. It should be noted that the

incubations shown in lanes 5 and 6 are from a different

experiment

than those in lanes 1

through

4.

Thus,

the

absolute

amountofP220detectedwasless inlane 5 thanin

lane4dueto variationin

staining intensity

between

experi-ments. Insummary, the results

presented

in

Fig.

1 confirm

activity

in extracts from

poliovirus-infected

HeLa cells which candegrade P220.

Wenextwantedto determinewhether

poliovirus proteins

3C or2C are involved in P220

proteolysis.

Protein 3C is a

likely

candidate for such a protease, since it is the viral

protein

involved in mostof the

cleavages

of viral precursor

polypeptides

to

yield

both structural andnonstructural

pro-teins (8). Ithas been

reported

that

protein

2Chas protease

activity

involved in

processing

of

poliovirus protein

pre-cursors

(9),

butthiswasnotverified ina more recent

study

(8).

Initially

we tested the

activity

of our

preparations

of antibodies

against

proteins

3C and 2C. Anti-3Cis knownto

inhibit

cleavages

at

Gln-Gly

pairs

which occurwhen

proc-essing

of viralprecursors is

assayed during

invitro transla-tion of

endogenous

poliovirus

RNA in extracts from

poliovirus-infected

cells

(8).

Consequently

we used this

assayto testthe

activity

of anti-3C. Translation incubations with extractsfrom

poliovirus-infected

HeLacells were car-riedout

according

tothe methodofLeeand

Sonenberg (11),

except that theextracts werenotnuclease treated. Thecell extract was

preincubated

for 60 minat

4°C

with

antibody

buffer or the

desired

antibody.

The translation incubations werethen

performed.

Reaction mixtures contained

(in

atotal volume of 25

,ul)

9

RI

of cell extract, 130 mM

potassium

acetate, 0.4 mM

magnesium

acetate, 20 mM HEPES

[N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic

acid,

pH

7.5],

1 mM

ATP,

200

,M

GTP,9 mMcreatine

phosphate,

22mgof creatine

phosphokinase

per

ml,

2.5 mM

dithiothreitol,

0.2

mM

spermidine,

19amino acids

(10 ,uM

each,

minus

methio-nine),

20

,uCi

of

[35S]methionine

(>1,000

Ci/mmol;

New J.VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:2.612.43.285.57.209.2]

(3)

1 2 3

I

4

5

6

7

8 -v

1ABCD-

3ABCD-3CD---- _

} a

FIG. 2. Effects of anti-3C and anti-2Conprocessingofpoliovirusprecursorproteins. Lanes 1through3 and 4through6 showdifferent

exposuresof thesameautoradiograph (exposuretimeswere10 h and 5min, respectively).Lanes: 1 and4,translationproductsintheabsence

ofantibody;2 and5,translationproductsin thepresenceof 60,ugofanti-3C;3 and6,translationproductsin thepresenceof 60pLgofanti-2C; 7 and 8, animmunoblot ofextractsfrom uninfected (lane 7)and infected (lane 8)cells probedwithanti-2C. Blotting conditions were as

describedinthetext.

England Nuclear Corp.), and 6 ,ul of antibody buffer (10mM

Tris[pH 8.0], 10 mMKCI)orimmunoglobulin G fractions of the antibodies indicated below. Antisera topoliovirus pro-teins 3C and 2C and purification of immunoglobulin G

fractionwas asdescribed elsewhere (8). After incubation for 60 min at 37°C, samples were resolved on 10% sodium dodecyl sulfate-polyacrylamide gels, followed by autoradi-ography.

Figure 2 (lanes 1 through 6) shows the [35S]methionine-labeled proteins produced in an extract from poliovirus-infectedHeLa cells. Lanes 1through 3 and4through 6 show different exposures of the samegel. Lanes 1 through 3 are overexposed to show the lower-molecular-weight polypep-tides. Lanes 4through 6 areunderexposed to showclearly thehigher-molecular-weightprecursors.Lanes 1 and 4 show endogenous translation products in the absence of antibody. Major bands which correspond tothe three precursor pro-teins,asdetermined from theirmigration relativeto molecu-lar weight markers (1ABCD, 3ABCD, and 3CD; formerly P1-la, P3-ib, and P3-2, respectively), are indicated to the left oflane 4. Inaddition thereare several lower-molecular-weight bands which correspondtothe various viral proteins derivedfrom thehigher-molecular-weightprecursors. Addi-tion of anti-3C to the translation incubation resulted in inhibition ofprocessing,asindicatedby the disappearance of thelower-molecular-weight bands and the build-up of high-er-molecular-weight polypeptides (lanes 2 and 5). The high-molecular-weight polypeptide in lane 5 (indicated by an arrowhead) migratedas an -150-kDa polypeptide and con-sisted of the combined amino acid sequences of 2ABC and 3ABCD as previously shown (8). Inhibition of 3C activity was specific. This is indicated by the fact that the Tyr-Gly cleavage that produces protein 3D' (formerly known as P3-6b) and which is notcarried outby protein 3C was also

not inhibitedby anti-3C, as previously shown (8; compare lanes1 and2). Densitometry of the lower-molecular-weight bands(indicated by arrowheadstothe leftof lane 1)showed

that under the conditions ofourassay,morethan90% of 3C activitywas blocked by anti-3C. Additionofanti-3Ctothe

translation incubation had no effect onthe total incorpora-tion of[35S]methionine into trichloroacetic acid-precipitable material(datanotshown). Lanes 3 and 6 showtheeffectsof

anti-2Conpoliovirus proteinprocessing. Itcanbeseenthat

anti-2C had no effect on processing of poliovirus polypeptides,aspreviously shown (8).To ascertain that the anti-2C antibodywasactive, weprobedHeLa cellextracts from uninfected and poliovirus-infected HeLa cells. The immunoblot is shown in lanes 7 and 8, representing the uninfected and infected cell extracts, respectively. The

antibody reacted with protein 2C in extractsfrom infected cells(lane 8) andgavenoreaction witha similar-molecular-weightpolypeptide inextractsfromuninfected cells (lane 7). There was also a weak reaction with a higher-molecular-weight band in extracts from infected cells which most probably correspondedtotheprecursorpolypeptide 2ABC. Thus, weconclude that the antibodies weusedwere active in inhibiting the activity or recognizing their cognate anti-gens. However,itshouldbeemphasized that the onlyassay

wehavefor the anti-2Cantibodywasimmunoreactivityona nitrocellulose blot, and it is possible that anti-2C cannot

inhibit theenzymatic activity of protein 2C.

Wenextasked whetheranti-3Coranti-2C could inhibit the proteolysis of P220. The protease assay was performed

under the conditions used for Fig. 1, lane 1, toensure that efficient proteolysis was achieved but that the protease activity was not in vast excess. Figure 3 (lanes 1 and 2) shows the U-Sl0 extract and a mixture of the U-Sl0 and I-S10extracts, respectively, incubated for 30 min.In lanes3

30---2ABC

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.612.142.474.70.315.2]

(4)

492 NOTES

12 3 4 5 6 7 8 Ehrenfeld (Proc.

Natl.

Acad. Sci. U.S.A., in

press),

which

demonstrated that 3C activity can be separated from the P220 proteolyzing activity and that antibodies against

pro-tein 3C donotinhibit P220proteolytic cleavage.

If proteins 3C and 2C are not directly responsible for cleavage of P220, then the question of the identity of this protease remains unanswered. It is possible that a hitherto-uncharacterizedpoliovirus-encoded protease is involved, or alternatively that poliovirus infection induces a cellular activity that cleaves P220. If the latter is true, it would be

interestingto know whether such anactivity plays a role in

[image:4.612.72.285.73.219.2]

regulationof protein synthesis in situations other than during poliovirus infection of HeLa cells.

FIG. 3. Effects of anti-3Cananti-2Conproteolysis of P220. The

protease assay was carried out and samples were processed for

immunoblottingasdescribed in thetext.Lanes: 1, 4,ul of U-S10; 2

through8, 4,ul of U-Sl0 and2,ul ofI-S10;3through 5, 2, 10, and 20

,ug of anti-3C, respectively;6through8, 2, 10, and 20,ug of anti-2C, respectively. Theextractswerepreincubated for60minat4°C with the desired antiseraasdescribed fortheexperimentshownin Fig.2.

through 5, increasing amountsof anti-3C were addedto the incubation under the same conditions as for the in vitro translation experiments. That is, the infected extract was preincubated with anti-3C for 60 min at 4°C. The highest amount of antibody added (expressed as micrograms of antibodypermicroliter of I-S10)wasinexcessof theamount which resulted inmorethan90% inhibition of the 3C activity (lane 5). It is clear fromourdata that anti-3C hadnoeffecton the protease activity whichcleaved P220, as evidenced by the absenceof P220 in lanes 3 through 5. Lanes 6 through 8 show that anti-2C (added in the same amounts as anti-3C) also hadnoeffecton proteolysis of P220. We conclude that the activity which cleaves P220 is not the same as that (protein 3C) which cleaves poliovirus precursor polypep-tides. The data alsosuggests thatprotein 2C is not directly involved in proteolysis of P220.

The resultspresented here and in previousreports estab-lishthat P220of the CBP complex is proteolytically cleaved in poliovirus-infected cells. However, it remains to be proven that the cleavage is indeed the cause for loss of activity of the CBP complex. The identification of a viral

proteaseresponsible for the degradation of P220 would lend support tothe proposedmechanism of inhibition of host cell protein synthesis. The results shown here indicate that the poliovirus proteinase 3C is notinvolved in the cleavage of P220, because anti-3C antibody didnotinhibit P220cleavage under the same conditions as itinhibited poliovirus protein cleavage. A similar conclusion canbe made forpolypeptide 2C, but with the reservationpointedoutabove.

It remainspossible that 3C amino acid sequences harbor different proteaseactivities dependingon whetherthey are partofhigher-molecular-weight forms (3CDor3C')orof the

matureform of protein 3C. If thiswerethecase,then anti-3C might not block both putative activities, thus raising the

possibility that protein 3C might still be involved in

proteolysis of P220. However, the fact that the polyclonal antisera used in our experiments recognized all forms of

protein 3C (8) argues against this possibility. In anyevent,

the main conclusion from our data is that the activity required for processing ofpoliovirusprecursors at the Gln-Gly amino acid pairs and the activity which proteolyses P220

are not the same. Our conclusion is in accord with recent

results obtained by R. E. Lloyd, D. Etchison, and E.

We thank Ellie Ehrenfeld forcommunicating her results to us

beforepublicationand Alice Newman forvaluablehelpin someof

theexperiments.

This research was supported by agrant from the Medical Re-search Council of CanadatoN.S. and PublicHealth Servicegrants AI-15122 and CA-28146 from the NationalInstitutes of Healthto E.W. N.S. isarecipientofaTerryFox CancerResearch Award

fromtheNational CancerInstitute ofCanada. K.A.W.L.andI.E.

arerecipientsofpredoctoralresearchfellowships fromtheCancer

ResearchSociety (Montreal).

LITERATURECITED

1. Baltimore, D. 1969. The replication of picornaviruses, p.

101-176. In H. B. Levy (ed.), The biochemistry of viruses.

Marcel Dekker, Inc.NewYork.

2. Edery, I., M.Humbelin,A.Darveau, K. A. W. Lee,S. Milburn, J. W. B. Hershey, H. Trachsel, and N. Sonenberg. 1983. In-volvementof eukaryotic initiation factor4Ain thecap

recogni-tionprocess.J. Biol. Chem.258:11398-11403.

3. Edery, I., K. A. W. Lee, and N. Sonenberg. 1984. Functional characterizationofeukaryoticmRNAcapbinding protein

com-plex: effects on translation ofcapped andnaturally uncapped

RNAs.Biochemistry23:2456-2462.

4. Ehrenfeld, E. 1982. Poliovirus-induced inhibition of host cell

protein synthesis.Cell 28:435-436.

5. Etchison, D., J. Hansen, E.Ehrenfeld,I.Edery, N.Sonenberg,S.

Milburn, and J. W. B. Hershey. 1984. Demonstration in vitro thateucaryoticinitiation factor 3 is activebut thatacap-binding protein complexis inactive inpoliovirus-infected HeLa cells. J.

Virol. 51:832-837.

6. Etchison, D., S. C. Milburn, I. Edery, N. Sonenberg, and

J. W. B.Hershey. 1982. Inhibition ofHeLacellprotein

synthe-sisfollowing poliovirusinfection correlateswith theproteolysis

of a 220,000-dalton polypeptide associated with eucaryotic

initiation factor 3 andacapbinding protein complex.J. Biol. Chem. 257:14806-14810.

7. Grifo,J. A.,S. M.Tahara,M. A. Morgan, A.J.Shatkin,and

W. C. Merrick. 1983. New initiator factoractivity requiredfor

globinmRNA translation. J. Biol. Chem.258:5804-5810. 8. Hanecak, R.,B. L.Semler, C. W.Anderson,and E. Wimmer.

1982. Proteolytic processing of poliovirus polypeptides: anti-bodies to polypeptide P3-7c inhibit cleavage at glutamine-glycine pairs. Proc.Natl. Acad. Sci. U.S.A. 79:3973-3977. 9. Korant, B., N.Chow, M. Lively,and J. Powers. 1979.

Virus-specifiedprotease inpoliovirus-infectedHeLa cells. Proc.Natl.

Acad. Sci. U.S.A. 76:2992-2995.

10. Lee, K. A.W.,I.Edery,and N.Sonenberg.1985. Isolation and

structural characterization of cap-binding proteins from

poliovirus-infectedHeLa cells. J. Virol.54:515-524.

11. Lee, K. A. W., and N. Sonenberg. 1982. Inactivation of

cap-binding proteins accompanies theshut-offof hostprotein syn-thesis by poliovirus. Proc. Natl. Acad. Sci. U.S.A. 79: 3447-3451.

12. Rose, J. K., H. Trachsel, K. Leong, and D. Baltimore. 1978. Inhibition of translationby poliovirus:inactivation ofaspecific

initiation factor. Proc. Natl.Acad. Sci. U.S.A. 75:2732-2736. 13. Rueckert, R.R.,and E. Wimmer. 1984. Systematic

nomencla-

P220--J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

(5)

tureofpicornaviral proteins. J. Virol. 50:957-959.

14. Tahara, S., M.A. Morgan, and A. J. Shatkin. 1981. Two forms ofpurified m7G-cap binding protein with different effects on capped mRNA translation in extracts of uninfected and poliovirus-infected HeLa cells. J. Biol. Chem. 256:7691-7694. 15. Towbin, H.,T.Staehelin,andJ. Gordon. 1979. Electrophoretic

transfer ofproteins from polyacrylamide gels to nitrocellulose sheets: procedures and some applications. Proc. Natl. Acad. Sci. U.S.A. 76:4350-4354.

16. Vaitukaitis, J.L. 1981. Production of antisera with small doses of immunogen: multiple intradermal injections. Methods Enzymol. 73:46-52.