Cell-free coupling of influenza virus RNA transcription and translation.

(1)

Cell-Free Coupling

of

Influenza Virus RNA Transcription

and

Translation

JEAN CONTENT,* LUCAS DE WIT, AND MICHEL HORISBERGER

Department of Virology, Institut Pasteur du Brabant,B-i040 Brussels, Belgium,* andResearch Department, PharmaceuticalDivision,CIBA-GEIGYLtd.,4002Basel,Switzerland

Received forpublication 13December1976

Acell-freecoupledsystem forthetranscription and translation offowl plague

virus RNA isdescribed.The system utilizes a newnuclease-preincubatedrabbit

reticulocyte lysate that has a high sensitivity toexogenous mRNA and a very low level of nuclease activity. Translation of the viral proteins inthe coupled system isstrictlydependentupon theviraltranscriptase activity. Inthe coupled system the optimal concentration ofmagnesium is intermediate betweenthe optimum for transcription and that for translation. Translation of the viral proteins seemsfaithful.The productsrepresentthe major viral peptides M and NP and two peptides with the same electrophoretic mobility as HA and P2. Virion NA is not resolved in the kind ofpolyacrylamidegelsdescribed. Proteins Mand NPwereimmunoprecipitablewithmonospecific antisera. It isconcluded that the virion-associated RNA polymerase transcribes the negative-stranded

segments of the viral genome coding for these major structural proteins into

fullyfunctional mRNA's.

Influenza viruses are enveloped animal vi-ruses whose genome is composed of a

seg-mented single-stranded RNA, probably

en-tirely "negativestranded" (15, 19). This finding isbasedonseveral lines ofevidence,asfollows:

(i) the virion RNA is not a template for viral protein synthesis in the wheat germ cell-free

system (20), whereas insimilarconditions the cytoplasmic polyadenylic acid [poly(A)]-rich RNA from influenza virus-infected cells pro-motes the synthesis ofthe major virus

struc-tural proteins (4, 6, 16); (ii)poly(A)-containing mRNA isolatedfrom infected-cell polysomesis

entirely complementarytothe virion RNA (5, 14); and (iii) the virion contains an RNA-de-pendent RNA polymerase (19).

Recentlyoneofusdescribed the possibility of stimulating the influenza virion-associated

po-lymerase activity with mammalian cell

ribo-somesinthe absenceofmanganeseions, butin

the presence of magnesium ions, and at low

ionic strength (7). These new conditions were

compatible with thoserequiredforprotein syn-thesis;therefore, itwas of interest to determine inacoupled system fortranscription and trans-lation whether the products of this activated RNA polymerase could represent fully func-tional viral mRNA's. In this paper we show

that, despite the low activity of the influenza virion-associated polymerase, it is possible to

establishacoupledsystemyieldingmostof the

virus structural proteins, provided that a suita-ble cell-free system ischosen.

MATERIALS AND METHODS

Cell and virus. Fowlplague virus(FPV), Rostock

strain, was grown in the allantoic cavity of

embryo-nated eggs and purified as previously described (7) by polyethylene glycol-NaCl precipitation followed by isopycnic centrifugation on a 5 to 40% (wt/vol) potassium tartrate gradient. The viral band was collected, diluted with 5 volumes of 10 mM Tris-hydrochloride (pH 7.4)-100 mM NaCl-1 mM EDTA (TNE), pelleted for 80 min at 109,000 x g,

resus-pendedinTNE, andsedimented in the same condi-tions.The pellet was resuspendedin asmallvolume of TNE and stored in smallportions inliquid nitro-gen. The virus protein concentration was 10 to 40 mg/mlinTNE.

[35S]methionine virus, cytoplasmic extracts, and cytoplasmic RNA from controlorinfected cultures of chicken embryo fibroblasts were prepared by the methodology described recently (4).

Cell-free translation in the reticulocyte lysate

translation system. Lysates were prepared from

rabbits (2 to 3kg) that had been made anemic by phenylhydrazine injection as previously described

(8). Theblood, containing 50 to 80% reticulocytes, was collectedandprocessedasdescribedby Pelham

andJackson (13). Preincubation of thethawed

ly-sate with the calcium-dependent micrococcal

nu-clease was as described previously (13), with the following modifications. Hemin (1 mM) was

dis-solved before use in concentrated KOH (21). The

concentration of the 19 unlabeled amino acids

(L-247

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

248 CONTENT, DE WIT, AND HORISBERGER

methionine wasomitted)addedtothe"mastermix"

wascalculated to be 110 gM in the final reaction mixture. The concentration of L-methionine in the final reaction wasoptimal at10 gM. Weroutinely used6 x 108cpm of[35S]methionine per ml. Never-theless, tokeep the specificradioactivityof the

me-thionine as high aspossible,wediluted itto160,000 cpm/pmol, with a final concentration of L-methio-nine at4 ,uM.

To assay protein synthesis, 3- or 5-gl samples from the reaction mixture werespottedonto What-man 3MM paper, and the hot trichloroacetic acid-insolubleradioactivitywasdetermined (17).

Disruption of the virion and RNA polymerase assay.Thepurifiedvirions weredisruptedin 8mM HEPES (N-2-hydroxyethyl piperazine-N'-2-ethane-sulfonic acid; pH 8.0)-0.3% Triton N-101-1% sodium deoxycholate-200 mMurea-10mM

2-mercaptoetha-nol-10mM potassiumacetatefor2minat20'C (7). The reaction mixture concentrations in the

tran-scriptase assay were: HEPES (pH 8.0), 50 mM; Mg2+,5mM (or asindicated); potassium acetate,110

mM; dithiothreitol, 1 mM; creatine kinase, 80 ug/ ml; creatinephosphate, 10 mM;ATP, 1 mM;UTP, GTP, and CTP,0.5mM; [3H]UTP(42Ci/mmol), 100

,uCi/ml; anddisrupted FPV,0.4 to 1.6mg/ml.

The reaction mixture (usually 25 ,l) was incu-bated at 31°C for 60 min and terminated by the addition of2volumes ofsaturated sodiuminorganic orthophosphate and sodium inorganic

pyrophos-phate (1:1)followed by20volumes of10% trichloroa-ceticacid. After15 min at0°C,theprecipitateswere

collectedby filtrationonWhatmanGF/C glassfiber disks or cellulosenitratefilters(Sartorius;0.45,um). Thefilters were washed atleast eight times with5%

cold trichloroacetic acid, dried, and counted in a

toluene-based scintillating solution, with an effi-ciencyof55%for 3H.

Experimental conditions for thecoupled system. The reaction mixture concentrationswere: HEPES (pH8.0), 25mM;L-methionine, 3,tM; Krebs ascites cell tRNA, 60 ,ug/ml; ATP, 1 mM; UTP, CTP, and GTP, 0.5 mM;anddisrupted FPV, 1mg/ml. A total of80 to90% of thefinal reaction volume was

com-prised of the nuclease-preincubated reticulocyte

ly-sate.When RNAsynthesiswasmonitored, [3H]UTP (42Ci/mmol) was added at100 ,tCi/ml. When pro-tein synthesis was assayed, 6 x 108 cpm of [35S]methionine per mlwasadded.Incubation,

usu-allyin 25,ul, was at31°C.

Productidentification. Samples (2 ,l) from the in vitro translation reaction, [35S]methionine-labeled

virion, or cytoplasmic extracts were analyzed by electrophoresison 7 to 20%polyacrylamide slab gel gradients (80 by 125 mm) (4). After fluorography (10), the X-ray films were scanned at 540 nm in a Gilfordspectrophotometer equipped for microdensi-tometrywitha model2520 linear transport system. Immunoprecipitation. Samples (50 to 100 ,ul) from the translation reactionweretreatedwith 0.1

mgofpancreatic RNase permland0.1 MEDTA for

15min at 37°C. Afterclarification for 5 min at 9,000 x g, the supernatant was incubated for 1 h at 20°C

with 0.1volume ofnormal rabbit serum and for 30

min with 0.05 volume of goat anti-rabbit

immuno-globulinG (IgG) (IgG fraction). Aftercentrifugation for 10 min at 3,000 x g, the supernatants were

divided into twofractions and treated with 0.1 vol-ume of either anti-M or anti-NP rabbit antibodies for1hat20°C. After the addition of 0.05 volume of goat anti-rabbit IgG (IgG fraction) for 30 min at

20°C, the samples were kept overnight at 0°C, washed three times by centrifugation for 2 minat

9,000 x g in phosphate-buffered saline containing

1%Triton X-100 and 1%sodium deoxycholate. The pellet was dissolved in 0.05 M Tris-hydrochloride (pH 6.8)-1% sodium dodecyl suflate-1% 2-mercapto-ethanol-10%glycerol, heated for3minat100°C, and analyzed by electrophoresis on a polyacrylamide slabgel as described above.

Materials. Micrococcal nuclease (EC 3.1.4.7) (29,000 U/mg) was from P.L. Biochemicals, Mil-waukee, Wis., and EGTA

[ethyleneglycol-bis(,8-aminoethyl ether)-NN-tetraacetic acid], hemin,

and Triton N-101 were from Sigma Chemical Co. (St. Louis, Mo.). The goat anti-rabbit IgG fraction wasobtained from Nordic Immunological Labora-tories (Tilburg, The Netherlands). Rabbit anti-bodies anti-X47(M) and anti-X42(NP)werekindly provided by G. C. Schild (World Health Organiza-tion International Laboratory for Biological Stan-dards).

[3H]UTP and [35S]methionine were obtained from the Radiochemical Centre (Amersham, United Kingdom).

RESULTS

Virion-associated polymerase. The virion-associated RNApolymerase ofFPVhasa rela-tively lowactivity in vitro. Figure1shows that, with atypical preparation of purifiedFPV, we

could obtain an incorporation of 40 pmol of UMPper mgofviral protein perh.

McGeoch and Kitron (11) and Plotch and Krug (S. J. Plotch and R. M. Krug, Abstr. Annu. Meet. Am. Soc. Microbiol. 1976, S114, p. 223) haveused some oligonucleotides to stimu-late this activity in vitro. We confirmed this observationby using ApGat aconcentration of 0.5 mM and found that this substance stimulated5 to 20times thetranscriptase activ-ity associated with FPV. The optimal

magne-sium concentration for this reaction has been reproducibly found to be 3 to 5 mM, and this requirement is notinfluenced by the addition of the dinucleotide (Fig. 1). Under these condi-tions the amount of [3H]UMP incorporated

amounts to 240pmol of UMP per mg of viral protein per h. This corresponds to 800 pmol of RNA per mgof viral protein per h and 0.2 to 0.4

pgg

of RNAper ml per h. In some early experi-ments (data not shown) we also observed a stimulation of the virion-associated polymerase activity by wheat germ ribosomes and to a much lower extent by the wheat germ extract. With the wheat germ lysate, several attempts weremade to couple theviral transcription and

J. VIROL.

http://jvi.asm.org/

(3)

INFLUENZA VIRUS: TRANSCRIPTION AND TRANSLATION 249

In

D 5.0_

U

z

z I-.

2

1 2 44 5

mM Mg

FIG. 1. Mg2+ requirement for the FPV-associated RNApolymerase activity and protein synthesis in the coupled cell-freesystem.The two dashed lines

repre-sent the incorporation of[3H]UMP in a standard transcriptase reaction mixture as a function of the magnesium concentration.Symbols: 0-*- (;O,in the absence ofApG; A----A, with 0.5 mMApG. The solid line represents the incorporationof [VS]methio-nine inthecoupled cell-free system. Disrupted FPV is added to the nuclease-preincubated reticulocyte

lysate,asindicated in the text. In this case, the con-centration of Mg2+ represents what isadded to the preincubated cell-free system; it is not the final con-centration sincethe preexistingMg2+inthe reticulo-cytelysate,is notknown.

translation by addingasuitable concentration

ofdisrupted FPV to the complete translation mixture in the presence of the four

ribonucleo-tide triphosphates. No significant stimulation of amino acid incorporation was detected. Whentheproducts of these reactions were

ex-aminedby polyacrylamide gelelectrophoresis, multiple bands wereobserved, with molecular

weights rangingfrom 15,000 to 100,000, andat

least30%of theproductswereofalow

molecu-larweight (below 15,000). None of the products

seemed to correspond to the virus structural proteins.

Use of rabbit reticulocyte lysate. Since this

failure could be the result of a high RNase content in the wheat germ extract, we decided to use a cell-free system that is known for its very low RNase activity. Recently, Pelham and

Jackson (13) described aprocedure that

elim-inates the endogenous mRNA activity of the rabbit reticulocyte lysate by means of a

cal-cium-dependent nuclease, thereby renderingit

extremelysensitive to alowlevelofexogenous mRNA. The efficiency of translation of this system becomes up to five times higher than that of the wheat germlysate.

A cell-free system was prepared by this

pro-cedure and found to be very efficient for the

translation ofseveral mRNA's. With purified rabbit globin mRNA or tobacco mosaic virus

RNA, theincorporation oflabeled aminoacids increased linearly between concentrations of

0.5and12,ugof RNAper ml.Thesaturation in

mRNAwasreachedat a concentrationof20,ugl ml.

Suchahighsensitivity to exogenousmRNA should allow the translation of the putative

mRNA synthetized in vitroby the polymerase

reaction, which under the best conditions

pro-duces0.5,ugofRNA permlper h.Thecoupling of the two reactions would be possible ifthe

ionicconditions arecompatible andifthe

prod-ucts of the polymerase reaction arefunctional mRNA's that are not too firmly bound to the RNAtemplate (19).

We have checked several parameters of the preincubated reticulocyte cell-free system. In

thissystem 80 to 90%of thereactionvolume is occupied by the undiluted reticulocyte lysate itself, whose final ionic concentration is un-known. Thus, the ionic parametersof the

trans-lation system cannot be measured, except for the 0.4 mM magnesium chloride and 95 mM

potassium chloride added

during

the

prepara-tionof thelysate (13). Wehave observed that

any further addition ofmagnesium is

inhibi-tory for the translating activity. With most

mRNA's,exceptglobin mRNA, the translation

isstimulated by theaddition of60 pg ofmouse

(Krebs ascitescells) tRNA per ml. Spermidine (0.05to 0.4mM) hadnoeffectontheactivityof thiscell-free system.

Itwas ofinterest to knowwhether this cell-free system translates accurately and effi-ciently the influenza virus mRNA obtained from FPV-infected cells. For comparison, the left panel of Fig. 2 shows the

electrophoretic

distributionof the labeledprotein components

A

, \

. 10

VOL. 22, 1977

I

II

II II II II

I

j.;

http://jvi.asm.org/

[image:3.504.46.239.63.414.2]

(4)

b

M

NP

I

HA

J\,

C f

5 4 2

cm cm

FIG. 2. Analysis of products synthesized in the mRNA-dependent reticulocyte lysate programmed

with total RNA from infected cells. Products were analyzed by slab gel electrophoresis on a 7to20% polyacrylamide gradient and scanned at 540 nm

afterfluorography. (a)[35S]methionine-purified vir-ion. The molecular weights of the virus structural proteinswereasfollows:P1,89,000; P2,80,000;HA,

74,000; NP, 59,000; andM, 28,000 (4). (b) Cyto-plasm fromchicken embryo fibroblasts that had been infected with1PFU ofFPVpercell andlabeled with

[35S]methionine from6to9 hafter infection.Slightly ahead ofHA,onecandistinguisharesidualamount ofa majorcomponentfrom the uninfectedcell

(mo-lecularweight,47,000) which is probably actin. The smallshoulder movingahead ofMprobably

repre-sents NS,. (c) [35S]methionine-labeled cytoplasm

from uninfected chicken embryo fibroblasts. (d)

[35S]methionine-labeled purified FPV. (e) In vitro

products in the reticulocyte cell-free system

pro-grammed with 200 pg of total RNA per ml from

FPV-infectedcells. RNA wasextracted2h after

in-fection. (f Same as(e), except thatRNA was

ex-tracted 4 h after infection. The full scale for the

microdensitometric scanning was setatan optical

density of1.2for(e) and 3.0 for all othertracings.

Endogenous products of the reticulocyte cell-free systemareshowninFig. 4and5.

from a purified FPV preparation (Fig. 2a), a

cytoplasmic extract obtained from

FPV-in-fected cells (Fig. 2b), and uninfected chicken

embryofibroblasts (Fig. 2c). Therightpanel of

Fig. 2 shows the labeled protein components

from a purified FPV preparation (Fig. 2d) as

compared with the products of the in vitro

translation in the reticulocyte cell-free system of totalcytoplasmicRNA extracted from chicken embryo fibroblastsat 2 (Fig. 2e) and 4 h (Fig.

20after injection with FPV. These cytoplasmic RNAs have notbeen enriched for poly(A)-con-tainiftg material by oligodeoxythymidylic acid [oligo(dT)]-cellulosechromatography.

RNA obtained as soon as 2 h afterinfection promotesthe synthesis of proteins M, NP, and HA(Fig. 2e). After 4 h, M and NPpredominate at the expense of putativeactin, a major host cell component (Fig. 2f), and a minor peak,

perhaps correspondingto P2, becomes

detecta-ble (Fig. 20.Except for their immunoprecipita-tion with an anti-FPV rabbit immune serum, these in vitro products have not been further characterized. We do not know whether the in vitro peptide at 2.5 cm (Fig. 2d, e, and f) is genuine HA (the precursor of

HA,

and HA2) andwhether it ispartiallyglycosylated.A simi-larobservation wasreported veryrecently(13). It isinteresting to note that in the wheat germ cell-free system the synthesis of such a compo-nent was neverobserved with poly(A)-contain-ing mRNA preparations obtained from infected cells (4, 6, 16). The difference between the two systems could reflect apeculiardeficiencyora higher nuclease content of the wheat germ ex-tract. It could also originate from losses

occur-ringduring the oligo(dT)-cellulose

chromatog-raphyutilizedinthe purification ofmRNAfor

the wheat germ system. This procedure could cause the loss of either some mRNA species with shorter or no poly(A) stretches or some essential tRNA species present in the infected chickenembryo fibroblasts.

Coupled system. (i) Kinetics and Mg2+ re-quirement. FPV wasdisrupted with urea,

so-dium deoxycholate, and Triton N-101 and

added to thenuclease-preincubatedreticulocyte lysate supplemented with the four

ribonucleo-tide triphosphates and either [3H]UTP or

[35S]methionine, as described above.

Figure3shows thatinthis reaction the incor-poration ofUMPcontinued for1to2h(Fig.3A

[a and c]), whereas the incorporation of

P15S]methioninereached its maximum between

2 and 5 h (Fig. 3B [c]). This incorporation is

weak butreproducibly higherthan that found

inthe endogenous reaction (Fig. 3B [a]). Except foratwofold stimulation after1 hof

J. VIROL.

http://jvi.asm.org/

[image:4.504.62.255.58.417.2]

(5)

incubation, the stimulatory effect of ApG (0.5 mM) of the RNA polymerase reaction was no longerobservedinthepresenceof reticulocyte

lysate (Fig. 3A [band c]);indeed, ApG exerted apronouncedinhibitoryeffectonthe

incorpora-tionof

[35S]methionine

(atleast50%oinhibition

ofthenet countsperminute) (Fig. 3B

[b

and

ci). This effect is alsoshown in Fig. 5.

As mentioned above, the actual Mg2+

con-centrationinthe lysate isnotknown, but any addition of Mg2+ to the reticulocyte lysate in-hibits protein synthesis programmedwith

ex-ogenousmRNA, whereas theoptimal Mg2+

con-centrationfor thepolymerase reaction is3 to 5

mM (Fig. 1). For this reason wemeasuredthe effect of Mg2+ on the rate of protein synthesis

in the reticulocyte lysate supplemented with disrupted FPV. For this reaction, Fig. 1shows a very sharp Mg2+ optimum at 2 mM. This requirement isthusintermediatebetweenthat

forproteinsynthesisand that for RNA

synthe-sis. This may explain whyboth transcription

and translation are suboptimal inthe coupled

reaction. This is in contrast to an analogous

system described recently with vesicular

sto-matitis virus (VSV) (1), in which both viral RNAtranslationand transcription are

signifi-cantlystimulated andprolongedwhencoupled.

Figure 3 shows that the situation is entirely

differenthere.The viral transcriptase alonein

theabsence of the lysate, but with0.5mMApG

and 5 mM Mg2+, was three tofour timesmore

active (datanotshown).

(ii) Product of the coupled reaction. Since

the

amount

of

[35S]methionine

incorporatedin

the coupled reactionwasrather low,we

exam-ined the protein products of this reaction by

polyacrylamide slab gel electrophoresis and fluorography. When compared with the

distri-bution ofthevirion proteins (Fig. 4a),the major

products of the coupled reaction coincide with proteins Mand NP (Fig. 4b).

An additional protein at 2.6 cm (Fig. 4b)

(which seems to correspond to actin) is also

detected inthe endogenous reaction (Fig. 4c).

For unexplained reasons, the synthesis of the

endogenous proteinwasalways increased when

thedisruptedviruswaspresent inthe reaction,

evenwhen the transcription ofviral RNAwas

inhibited (Fig. 4c and d; see section ivbelow).

Two distinct peptides run more slowly then NP. When this kind of experiment was

re-peated, these components, as compared with

theproducts of the infected cytoplasm (Fig. 40

and the endogenous reaction (Fig. 4h), were

betterdetectedintheproducts ofasimilar

reac-tion (Fig. 4g). It is clear that they have the

same mobility as HA and

P2.

P1 cannot be

ul1* _ b

- - a

I I

HOURS

FIG. 3. Kineticsofincorporationof[3H]UMP(A)

and [35S]methionine (B.) in the coupled cell-free system. (a)Controlreticulocyte Iysatewithout addi-tionof disruptedFPV; (b) reticulocyteIysatewithI

mgofdisruptedFPVpermland 0.5 mMApG;- (c)

reticulocyte Iysate with 1 mgof disruptedFPVper ml.

detected becauseof thepresenceofone

peptide

with thesame

mobility

inthe

endogenous,

reac-tion

(Fig.

4h). These

components

weredetected

aftera

long

exposureof the

gel.

(iii) Inhibition

by ApG.

It was mentioned

earlier that 0.5 mM

ApG

had an

inhibitory

effecton the total

protein

synthesis;

this was

also demonstrated

by

adecreased

synthesis

of

Mand

NP,

the

major

products

of in vitro

trans-lation.The

inhibitory

effect of

ApG

seemstobe relatedtoa

posttranscriptional

eventsincethe

transcription

of viral RNA was stimulated

by

ApG (Fig.

1and3A).On the other

hand,

Fig.

5

shows that the inhibition of NP and M

syn-theses in the

coupled

system was about 50% under the conditions that do not

modify

the translation of the

endogenous product.

Since

the same concentration of

ApG

also had no

detectable effectuponthe translation of

exoge-nousrabbit

globin

m1RNAin the

preincubated

reticulocyte lysate

orinthe wheatgermsystem

(data not shown), the most

likely

explanation

isthat

ApG

prevented

some

posttranscriptional

modification of the mRNA5'terminusor

intro-ducedanabnormal 5' terminus. Recent

obser-22, 251

http://jvi.asm.org/

[image:5.504.252.448.74.331.2]

(6)

252 CONTENT, DEWIT, AND HORISBERGER

6 5 4 3 2

cm

FIG. 4. Analysis of products synthesized in the coupled reaction under different conditions. (a)

[35S]methionine-purifledFPV; (b) in vitroproducts obtained in thecoupled cell-free systemwith1 mgof

disruptedFPVperml and incubated for2hat310C;(c)products ofthesamecoupled reaction supplemented with0.1 mMspermidine hydrochloride; (d) products of the coupled reaction without spermidineand thefour ribonucleotide triphosphates; (e) endogenous products of the reticulocyte cell-free system reaction without disrupted FPV; (f) 35S-labeled cytoplasmic proteins fromFPV-infected cells;(g) coupled reactionincubated

for3.5 hat310C; (h) endogenous reaction(without disrupted FPV). For (b), (c), (d),and(e),theexposure timewas96h.Thefull scale for microdensitometricscanningwasset atanopticaldensity of6.0for(a)and

3.0forthe other tracings.The productswereanalyzedasdescribed in thelegendtoFig.2.Theexposuretime was10days for (g)and(h)and 24 hfor (f). The full scale for themicrodensitometricscanningwassetatan

optical density of3.0.

vations showingthat ApG or GpG (11; Plotch

andKrug,Abstr. Annu. Meet. Am. Soc.

Micro-biol. 1976,S114,p.223),when usedtoprimethe viral transcriptase, are incorporated at the 5'

endof the newly synthesized RNA would sup-portthe secondhypothesis.

(iv) Effect ofsome inhibitors of

transcrip-tion. If the translation of the viralproteins in

the systemdescribedhere results from its

cou-plingwithviral mRNAsynthesisandnotfrom the presence ofsome mRNA contaminants in

the virion (4)orfrom the virionRNAitself(20),

it should be possible to prevent it by simply

inhibitingthe transcriptasereaction. Thiscan

be doneveryeasily by omittingthefour

ribonu-cleotide triphosphates from the reaction

mix-ture. Itmustbe noted thatno addition of GTP orATPisrequiredfor translation in the

prein-f

JXA4J401

VLYX

9 h

i & it*

J. VIROL.

1

http://jvi.asm.org/

[image:6.504.112.415.63.457.2]

(7)

INFLUENZA VIRUS: TRANSCRIPTION AND TRANSLATION 253

6 5 4 3 2 1

cm

FIG. 5. Effect of ApGonprotein synthesis in the coupledreaction. (a)Coupledreactionsupplemented

with0.5 mMApG; (b) coupled reaction;(c)

endoge-nous reaction.Incubation wasfor 8hat31`C. The productswereanalyzedasdescribedinthelegendto

Fig.2. Theexposuretimewas24 hfor(a), (b),and (c).Thefullscalewasset atanopticaldensity of3.0.

cubatedreticulocyte system (13). Whenthe

ri-bonucleotide triphosphates were omitted (Fig.

4d), thecell-freesystemcontinuedtosynthesize putative actin (anendogenous peptide), but the synthesisof viral proteinsMandPwasreduced by at least 95% (compared with Fig. 4b). The

observed residualamountofviralpeptidesmay

resultfrom the reticulocyte endogenous pool of nucleotidetriphosphate.

As mentionedearlier,0.05to0.4mM

spermi-dine hasnoeffecton thetranslating efficiency

of the reticulocyte lysate programmed with exogenous mRNA. Nevertheless, itcompletely inhibitsthe FPV RNApolymerasereaction(18)

(M. Horisberger, unpublished data). In the

coupledsystemthe addition of0.1 mM

spermi-dinecompletely abolished thesynthesisofviral

proteins (Fig. 4c).

(v) Further characterization of in vitro

products obtained in the coupled reaction.

Since theamountsof labeledproteinsobtained

from the coupled reaction were too low for a

determination of their [mS]methionine tryptic peptidemaps, onefurthercharacterizationwas

performed by immunoprecipitation with two

different monospecific rabbit antibody

prepara-tions. A comparison with thedistribution of the labeled proteins fromanFPV-infected cytoplas-micextract(Fig. 6a) shows that, after immuno-precipitation withananti-NPrabbit antiserum (Fig. 6b) followed by polyacrylamide gel elec-trophoresis, morethan 80% of theprecipitated

radioactive material comigrates with protein NP, whereas withan anti-M rabbit antiserum more than 95% of the precipitated material runs as a single component, with the same electrophoretic mobilityasproteinM (Fig. 6c).

DISCUSSION

We haveestablisheda cell-free coupled

sys-tem for the transcription and translation of FPV mRNA. Inthissystem,translation of the viral proteins is strictly dependent upon the functional integrity of the transcriptase

reac-tion. Both transcription and translation are

faithful andresult inthesynthesis of the main viral proteins M and NP and probablyP2 and HA.These proteins have been characterized by their electrophoretic mobility on

polyacryl-amidegels. OnlyMand NPhave been

charac-terizedby their antigenicity. Protein NA,even

from purified virion, hasnotbeen resolved on

the kind of gels used in these experiments. P1, which is the least abundant of all the

vir-ion proteins, cannotbedistinguishedfromone

ofthe endogenouscomponents of the

translat-ing cell-freesystem.

The coupled reaction hastobesupplemented

with Mg+; the final concentration ofMg2+ is practically halfway between the optimal

con-centrationof the viralpolymerase (3to5mM)

and that requiredfor the translationof

exoge-nousmRNA in thereticulocytecell-freesystem

(wherenoaddition ofMg2+isrequired). Due to this different requirement for the two

reac-tions,whichmaybe takenasoneexample,itis

not surprising that the coupled system used

here translatedwitharelativelylowefficiency

and thattherewasnocooperationor

reinforce-ment between the two reactions, as was

ob-served in thecaseof VSV(1).The virion-associ-ated RNA polymerase of VSV is much more

active (20 to 50 times), and the ion

require-M NP

a

b

~~~~

a i

VOL. 22, 1977

http://jvi.asm.org/

[image:7.504.51.243.63.460.2]

(8)

7 6 5 4 3 2 1

cm

FIG. 6. Product analysis ofthe coupled reaction after immunoprecipitation with anti-FPV IgG. (a) 35S-labeled cytoplasmic protein from FPV-infected cells; (b) immunoprecipitateobtainedfrom the stan-dard coupled reaction with anti-NP rabbit antise-rum;(c) immunoprecipitateofthesamereactionwith anti-Mrabbitantiserum.Theexposuretimewas4h.

Thefull scalewasset atanoptical density of3.0. ments are very similar for transcription and

translation; thus, both reactions areactivated

andprolongedas aresult of thecoupling.

In the present work we have taken

advan-tageofanewly described preincubated

reticulo-cyte cell-free system (13). Because of its

effi-ciency, high sensitivity to low amounts of

mRNA, and probably a very low RNase

con-tent, this systemcould be used here in

conjunc-tion withavirus-associatedRNApolymeraseof

verylowactivity.

The maximal concentration of viral RNA synthesized by FPV-associated transcriptase

wasabout 0.5jigpermlperh in thepresenceof

ApG. In the coupled reaction, where both

tran-scription and translation are suboptimal, for

reasonsstated above, theamountof RNA

syn-thesized (ascalculated from Fig. 3A) wouldnot

exceed0.4

tug

permlperh. This is atleast 20

times less than the amountof VSV RNA

syn-thesized in the coupled cell-free system de-scribed byBall andWhite (1). In addition, the

amount ofvirus usedis 20 timeshigher with FPVthan with VSV. Nevertheless, itmustbe kept in mind that the translation efficiency of themRNAgeneratedin suchacoupled cell-free

system might be much higherthan thatofan exogenousnmRNA (1).

Inhibition of translation of the viral products by ApG, whichwasobserved exclusivelyinthe coupled system, might be posttranscriptional since the dinucleotidedid not inhibitthe

trans-lation ofexogenousmRNA and stimulated the

virion-associated transcriptase. This could be explained if ApG serves as a primer for the transcriptase and is incorporatedatthe 5'

ter-minus of the native mRNA's as suggested

re-cently (11, 18; Plotch and Krug, Abstr. Annu. Meet. Am. Soc. Microbiol. 1976,S114,p.223). It

is conceivable that thepresenceof ApGatthe5' end of the mRNA's may introduce a wrong

signal for an eventual capping enzyme or for

thecorrectinitiation of translation (9). Totest

thispossibility,itwill be ofinteresttomeasure

the extentof mRNA capping and methylation

inthiscell-freesystem.

Onestriking observationthatcanbe derived

from the coupled cell-free system is that the proportion of the different proteins synthesized isnot toodifferent from that existing in vivo. It would therefore be possible to determine whether this proportion is regulated at the translation or transcription level, or even

whether itdependson someposttranscriptional

modificationof themRNA's (3, 9).

Finally, the fact that one can synthesize in

vitromostof the authentic viral proteins ina cell-free translating system that is only fed

with disrupted FPV particles and the four

ri-bonucleotidetriphosphatesprovesthat the

vir-ion-associated RNA polymerase transcribes faithfully and completely the corresponding negative-strandedsegmentsof the virion RNA. Untilnowthe products of the virion-associated

RNA polymerase had been examined by elec-trophoresis and annealing to purified virion RNA(2, 19).Fromthepresentinvitro observa-tionsitcanbe safely concluded that this RNA

polymerase really functions as a viral tran-scriptase.

NP

M

a

b

CJ

-1 I I _I~~_I I_X

7

1

J. VIROL.

http://jvi.asm.org/

[image:8.504.85.232.74.475.2]

(9)

ACKNOWLEDGMENTS

Wethank G. C.Schild for providing monospecific anti-sera, A.Ball foradvice,B.Mechlerfor suggesting the use of thenuclease-preincubatedreticulocytecell-free system, H. R. B.Pelham forcommunicating a preprint on the subject, and L.Thiryforcriticallyreadingthemanuscript.H.Van Haelen is gratefullyacknowledged for experttechnical as-sistance.

Two of us (J.C. and L.D.) were supported by grants from theBelgiumFonds dela RechercheScientifiqueM6dicale, Fonds National de la Recherche Scientifique, and the Fondation Rose et Jean Hoguet.

LITERATURE CITED

1. Ball, L. A., and C. N. White. 1976. Order of transcrip-tionof genesof vesicular stomatitis virus. Proc. Natl. Acad. Sci. U.S.A. 73:442-446.

2. Bishop, D. H. L., J. F.Obijeski,and R. W. Simpson. 1971.Transcriptionoftheinfluenza ribonucleic acid genomebyavirionpolymerase. II. Natureof thein vitro polymeraseproduct. J. Virol. 8:74-80. 3. Breindl, M., andJ. J.Holland.1976.Studiesonthein

vitrotranscription and translation of vesicular sto-matitis virusmRNA. Virology 73:106-118.

4. Content,J. 1976.Cell-free translationofinfluenzavirus mRNA. J. Virol.18:604-618.

5. Etkind, P. R.,andR. M. Krug. 1974.Influenza viral messenger RNA.Virology 62:38-45.

6. Etkind,P. R., and R. M. Krug. 1975. Purificationof influenza viral complementary RNA: its genetic con-tentand activity in wheat germ cell-free extracts. J. Virol.16:1464-1475.

7. Horisberger,M. 1976.The invitrostimulationof the influenza transcriptase by ribosomes. FEBS Lett. 63:134-136.

8. Housman, D., M. Jacobs-Lorena, U. L.RaJbhandary,

andH. F. Lodish. 1970. Initiationofhaemoglobin synthesis by methionyl tRNA. Nature (London) 227:913-918.

9. Krug, R.M., M. A. Morgan,and A. J. Shatkin. 1976. Influenza viral mRNA contains internal

N'-meth-yladenosine and 5'-terminal 7-methylguanosine in capstructures.J. Virol. 20:45-53.

10. Laskey,R.A., andA. D. Mills.1975. Quantitativefilm detection of 3H and 14C inpolyacrylamidegelby fluo-rography.Eur.J. Biochem. 56:335-341.

11. McGeoch, D., and N. Kitron. 1975. Influenza virion RNA-dependent RNA polymerase: stimulation by guanosine and related compounds. J. Virol. 15:686-695.

12. Morrison,T.G., and H. F. Lodish. 1975. Site of synthe-sis ofmembrane and nonmembraneproteins of vesic-ular stomatitis virus.J. Biol. Chem.250:6955-6962. 13. Pelham,H. R. B., and R. J. Jackson. 1976. An efficient

mRNA-dependent translation system from reticulo-cytelysates.Eur. J. Biochem. 67:247-256.

14. Pons, M. W. 1973. The inhibition of influenza virus RNA synthesisby actinomycin D andcycloheximide. Virology 51:120-128.

15. Pons, M. W. 1975. Influenza virus RNA(s), p. 145-170. In E. D. Kilbourne (ed.),The influenza virusesand influenza. Academic Press Inc., New York. 16. Ritchey,M. B.,and P. Palese. 1976. In vitro

transla-tion of influenza virus messenger RNA's. Virology 72:410-420.

17. Roberts, B. E., and B. M. Paterson. 1973. Efficient translation of tobacco mosaic virus RNA and rabbit globin 9S RNA in a cell-free system from commercial wheat germ. Proc. Natl. Acad. Sci. U.S.A. 70:2330-2334.

18. Rochovansky, 0. M. 1976. RNA synthesis by ribonu-cleoprotepolymerase complexes isolated from in-fluenza virus.Virology73:327-338.

19. Simpson, R. W., and W. J. Bean. 1975.Influenza tran-scriptase activity of cells and virions, p. 125-143. In E. D.Kilbourne (ed.), The influenza viruses and in-fluenza.Academic Press Inc., New York.

20. Tekamp, P. A., and E. E. Penhoet. 1976. Message activity of influenza viral RNA. J. Virol. 18:812-816. 21. Weber, L. A., E. R. Feman, and C.Baglioni. 1975. A cell free system fromHeLa cells active in initiation of protein synthesis. Biochemistry 14:5315-5321. 22, 1977

http://jvi.asm.org/