human RNA

CellBiology

Interferon messenger RNA content of human fibroblasts during induction, shutoff, and superinduction of interferon production

[humandiploidfibroblastcells/poly(I)poly(C)/McAuslan-Tomkinstranslational repressorhypothesis/5,6-dichloro-l-l-D- ribofuranosylbenzimidazole/Xenopuslaevisoocytes]

PRAVINKUMAR B. SEHGAL, BERNHARDDOBBERSTEIN*, ANDIGORTAMM

TheRockefellerUniversity, NewYork,NewYork10021

ContributedbyIgor Tamm, May 16, 1977

ABSTRACT Translation ofinjectedmRNA in oocytes of Xenopus laevishas been usedasahighlysensitiveand quanti- tative assay for interferon mRNA. Injection into oocytes of polyadenylylatedRNAextracted frompoly(I)poly(C)induced human diploid fibroblasts(FS-4)leadstothesynthesisofbio- logically activehumanfibroblast interferon over aperiodof 24-32hr.There is alinearrelationship betweentheamountof mRNAinjectedandthe interferonyieldobtainedoverarange of1-20ngofinjectedRNA.Injectionof40-80ngofmRNA into eachof15oocytes,homogenizedin0.3mlof incubation medi- um, gave a titerof128-256 interferonreference units/ml of homogenate.

FS-4 cells at the peak of interferon production-i.e., ap- proximately 2.5 hr after the beginning of induction with poly(I)poly(C-gavemRNAthatyielded24-48interferon ref- erenceunits/mlinthe oocyte assay (30 ng ofRNAinjectedper oocyte). Anequivalent amount of mRNA from FS-4cellsinthe shutoffphase, approximately6hrafterinduction,gave<4 in- terferonreferenceunits/ml. Incontrast, mRNA extractedfrom FS-4cells that hadbeen inducedandmaintainedinthepresence of40uM

5,6-dichloro-1-j-o-ribofuranosylbenzimidazole

for6 hr produced 64-128 interferon reference units/ml. Polyade- nylylatedRNAobtainedfrom uninducedFS-4cellsdidnotlead to detectable interferon synthesis (<4 interferon reference units/ml). These data provideadirect verification ofthehy- pothesisthattheshutoff of interferon productionin FS-4cells involves aregulatoryeventleadingtotheposttranscriptional inactivation ordegradationofinterferonmRNA. Becausethe inactivating mechanism is sensitive to inhibition by 5,6-di- chloro-1--D-ribofuranosylbenzimidazole, a selective inhibitor of nuclearheterogeneousRNAandmRNAsynthesis,itislikely thatsynthesisofanRNAmoleculeisnecessaryfortheshutoff ofinterferonproduction.

Humanfibroblastinterferonis aglycoproteinwithanapparent molecular weightof21,000(1,2)andischaracterized bythe F antigenic determinant (3). Culturesof human diploidfi- broblasts(FS-4strain)canbeinducedtosynthesizeandsecrete interferonintotheextracellularmediumbyexposure topoly- (inosinicacid)-poly(cyticylic acid)

[poly(I)'poly(C)]

(4-8).Ex- tensivestudies withinhibitorsofRNAorproteinsynthesisin- dicate that the regulation of interferon production in poly(I)-poly(C)-induced FS-4cellsiscomplex(7, 9-11). Fig.

1 presentsasummaryofthe keyeventsthought tounderlie poly(I)-poly(C)-induced interferon production in FS-4cells, asrevealedthroughinhibitor studies (10).

InterferonproductioninFS-4cells reachesdetectable levels approximately1hrafter thebeginning of induction,rises to a peakby2.5-3hr,andisrapidlyshut off by 6-8 hr (5, 10, 12).

Inhibitors of RNA or protein synthesis, applied before the shutoff,preventthe shutoff and leadtoacontinuousproduction ofinterferon forup to4days (7,9, 10,12,13).As aresult,there occurs a paradoxical increase ("superinduction") ofup to a 100-foldinthe cumulative interferon yield (7,9,13-15). De- tailed analyses of the kinetics of interferon production by

poly(I)-poly(C)-induced

FS-4cellshave ledtothesuggestion that the rateofinterferonproductioninresponse topoly(I).

3409

poly(C) isdeterminedbytwoprocesses (7, 16). Theinducer

brings

about therapid synthesisofinterferon mRNA, which islargely complete by3hr after thebeginningofinduction

(5,

13, 16,17). Indirect evidencesuggeststhat interferon mRNA hasahalf-life of12-15hrat370 inthepresenceofinhibitors ofRNAsynthesis(12,

13).

Concurrently with the induction of the synthesis of interferon mRNA,poly(I)-poly(C)isthoughtalsotobring about thesyn- thesis ofanRNAspecies,andpossiblyalso ofaproteinspecies, thatisresponsiblefor posttranscriptionalrepressionofinterferon mRNA.Thisrepressormechanismisthoughttoberesponsible for therapidshutoff of interferon productionby6-8hr after inductiondespitetheintrinsicstabilityofinterferonmRNA

(7,

16,18). Ithas been proposed that inhibitors ofmacromolecular synthesis, givenatappropriate times, interfere with thesyn- thesis ofcomponentsofthe shutoffmechanism and henceare abletoenhance interferonyield (5,7, 9, 10, 12,14,15,18,

19).

Themolecules mediating the shutoffappear tobe unstableand

tohaveanoveralllifetime of only 3-4 hr (7, 16).Furthermore, the shutoffmechanismappears to inactivateinterferon mRNA in an irreversible manner (7). It has been suggested thata translationalrepressorof the kindhypothesized by McAuslan (20)andbyTomkins(21-23)isresponsiblefor the shutoff of poly(I)-poly(C)-inducedinterferonproductioninhumandip- loid fibroblasts (7,9, 14, 15).

Ithas been discoveredrecently thatpolyadenylylated RNA extractedfrominterferon-producinghumanormousecellscan be translatedinwheat-germ, mouseEhrlichascites,Krebs-II ascites,and rabbitreticulocyte cell-free translation systems and inthe Xenopuslaets* oocyte, with the production ofbiologi- callyactiveandspecific interferon(24-26).

Wehave used theX. Iaevtsoocyteinjectionassayforinter- feron mRNA(26, 27)tocarry out adirect biochemicaltestof the hypothesis that the shutoff ofpoly(I)-poly(C)-inducedin- terferon productioninFS-4cells involves the irreversiblein- activationof interferonmRNA. The resultsdescribedinthis communicationprovidestrong supportfor thishypothesis and show that suchinactivationcanbepreventedby5,6-dichloro- 1-/3-D-ribofuranosylbenzimidazole(DRB), aninhibitor ofRNA synthesis (10,12,28,29).

MATERIALS AND METHODS

Cell Cultures.Ahumandiploidfibroblast strain (FS-4) (5, 30)has been usedinall experiments(7, 19, 31).Interferonti- trations werecarriedoutwith ahuman cell straintrisomic for chromosome21(GM258) (32) obtained from the HumanGe- neticMutantCellRepository,Camden,NJ.

Abbreviations:poly(I)-poly(C), poly(inosinicacid)-poly(cytidylic acid);

DRB, 5,6-dichloro-1-f,-D-ribofuranosylbenzimidazole; Pi/NaCl, phosphate-buffered saline.

*Presentaddress:EuropeanMolecularBiologyLaboratory,Postfach 10.2209,69Heidelberg, Germany.

Proc.Natl.Acad.Sci. USA 74(1977)

O2_

4Q Ec o 6.

0 ac r', c in

2 4 6 8 10 12

H-ours Interferon mRNA

Repressormechanism (RNA, ? Protein)

&-.

FIG. 1. Scheme of transcription ofinterferon mRNA and re- pressorRNA-determinants of interferon synthesis in a culture of FS-4 cellsinducedwithpoly(I)-poly(C).

Interferon Induction and Titration. Interferon wasinduced

by

exposingconfluentFS-4cultures, washedoncewithwarm phosphate-bufferedsaline

[Pi/NaCl

(ref.

33)],

to asolutionof poly(I)-poly(C)(12.5 or 20

lg/ml,

P-LBiochemicals,Inc.) in Eagle'sminimumessentialmedium for 1 hr at 37°. Interferon wasassayed bymeansofthe semimicromethodusingGM258 cells (30, 34). Interferontitersareexpressed in termsofthe 69/19referencestandard. Onereference unit/mltitratedat 2units/mlinthe assay we used.

Cell Fractionation and RNA Extraction. Cultures were rinsed four timeswith ice-cold

Pi/NaCl,

and the cellswere scrapedintocold

Pi/NaCl

andpelletedin arefrigeratedcen- trifuge (1000rpm, 3min). Thecellpelletwaseitherextracted withphenol directlyorthe nuclearandcytoplasmic fractions wereisolatedfirstandthenextractedwithphenol. Forfrac- tionation,the cellswereresuspendedin

one-third-strength

re- ticulocyte standardbuffer(3.3mMNaCl/0.5mMMgCI2/3.3 mMTris-HCI,pH7.4)for 10 min andthecellswerebrokenin aDouncehomogenizer.Nuclearandcytoplasmic fractionswere separated by centrifugation (2000rpm,3min)and the nuclear pellet was further washed withdetergent (0.5%

deoxycho-

late/1.0% Tween 40) beforeRNAextraction (35, 36). RNA fromthecytoplasmicfractionwas

phenol

extractedatneutral pH and room temperature whereas that from the nuclear fraction and from

whole

cellswasextractedatlowpH

(pH

5.1 inacetatebuffer)at650by

procedures

described earlier(29, 37-39).

Polyadenylylated

RNA was isolated

by poly(U)- Sepharose chromatography (37-39).

Translation ofmRNA inOocytes ofX.laevis.Oocytes were obtainedbydissection ofmature X.laevts females

purchased

from Nasco(FortAtkinson, WI).Oocyteswereused within2 daysofdissection. Clusters of

approximately

five

fully

devel- opedoocytesweremicroinjectedwith100-150nl per oocyte oftheappropriateRNAdissolvedindistilled wateror in in- jectionbuffer(88mMNaCI/15mM

Tris-HCI,

pH 7.6;ref.

40).

Ten to15oocytes wereusuallyinjectedpervariable. Five in- jectedoocyteswereincubatedper0.1 ml of sterilemodified Barth'smedium(40)at roomtemperature

(21-230).

Atthe end of the

incubation,

theoocyteswerebroken

by

meansofacap- illarypipetteinthesamealiquotofmedium thatwasused for incubation. Thehomogenatewasclarified

by centrifugation

for30minattopspeedinaBeckman

Microfuge

at40 andtwo 50-Alaliquots oftheclarifiedsupernatewereusedtomakeup a 2-fold dilutionseries totestforinterferon in

duplicate

as- says.

Cell-FreeTranslationofmRNA.Thiswascarriedout inthe wheat germandHeLa cell systems

by

using

previously

de- scribed

procedures

(24,41,

42).

256 E 128

C 64

1 32 16 8

C

4

<4

6 12 24 48 96

Hours

FIG. 2. Time-course of synthesis ofinterferon inoocytesinjected with interferon mRNA.Cytoplasmic polyadenylylated RNA(@)(see Table1) wasinjected into 50 oocytes (80 ng of RNA per oocyte). An equivalentamount ofnuclear polyadenylylated RNA(U)(Table 1) wasalsoinjectedinto 15oocytes (40 ng of RNA per oocyte). Groups of 10injectedoocytes wereharvested atdifferenttimes after injection andthe amount of interferon synthesized was determined.

Interferon Neutralization Test. Specific antisera directed against human fibroblast interferon (anti-F antiserum) and human leukocyte interferon (anti-Le antiserum) were a gift from E. A. Havell, New York University School of Medicine, New York, NY. Interferon neutralization by antiserum was tested as described (3).

RESULTS

Cell-Free Translation of Interferon mRNA. Our initial effortsweredirectedatestablishingareproducible cell-free translationassayfor interferon mRNA based on a bioassay (30, 34, 43) todetectinterferon synthesis. Polyadenylylated total cellularRNAaswell ascytoplasmicand nuclear poly(A)-con- tainingRNAwereobtained fromgram quantities ofFS-4 cells that hadbeen treated withpoly(I)-poly(C) inthe presence of cycloheximidefor 4 hr(25,26).The mRNA preparations were translated, togetherwith control preparations of rabbit globin and dog pancreas mRNAs, in the wheat-germ and HeLa translationsystemsbyusingpreviouslydescribedprocedures (24-26, 41, 42). It was our uniform experience that mRNA preparations frominduced FS-4 cells stimulated incorporation of radiocative aminoacidsinboth systems to an extent com-

parable

tothe stimulationproduced by rabbit globin and dog pancreas mRNAs(approximately 25-foldinthe wheat-germ systemand 7-foldinthe HeLasystem). Nevertheless, no in- terferonactivitywasdetectableinthe translationproducts (<10 referenceunits/ml).Varyingthe concentration of spermine in thetranslation systemandaddingamembrane fraction from dogpancreas(44,45)didnotalter this result. Inasmuch as the samecytoplasmic and nuclearRNA preparations wereeffi- ciently translated into biologically active human fibroblast interferonbyX.laevis oocytes 6months later(Figs.2and3;

Table1),itappearsthatsynthesisof the interferonpolypeptide isnotbyitself sufficienttogeneratebiologicallyactive inter- feron.

Translation of InterferonmRNA inOocytesofX.

laevis.

The same

sample

of

polyadenylylated cytoplasmic

mRNA obtained from inducedFS-4cells thathad been usedinsome ofthe cell-free translationexperimentswasusedtoobtaindata

presented

inTable1and Figs.2and3.Oocyteswere

injected

withmRNAandincubatedfor48hr. The clarified oocyte ho- mogenatesweretested for interferon activity.Ascanbeseen in Table 1,X. laevis oocytessynthesized biologically active interferoninhigh

yield

afterinjectionofmRNAobtainedfrom

poly(I)-poly(C)-induced

FS-4 cells. Interferon activity was neutralizedbyantiserumagainsthumanfibroblastinterferon 3410 Cell

Biology: Sehgal

etal.

28

(A

C

C

0.3 0.6 1.2 2.5 5 10 20 40 80 RNA injected, ng/oocyte

FIG. 3. Relationship between the amount of interferon mRNA injected per oocyte and the interferon yield. A serial 2-fold dilution of the interferon mRNA stock (Table 1) was prepared in injection buffer and 15 oocytes were injected with about 150 nl each of the variousRNA dilutions. Interferon synthesized by the end of 48 hr was determined.

but not by that against human leukocyte interferon. The oocyte supernate was inactive in the interferon assayin mouse L cells.

In separate experiments, we have observed that control unin- jected or mockinjected oocytes as well as oocytes injected with mRNA from uninduced FS-4 cells do not yield detectable in- terferon activity (see Fig. 4).

The time-course of synthesis of interferon in oocytes after injection of cytoplasmic polyadenylylated mRNA is described in Fig. 2. It can be seen that interferon was detectable within 6 hr of mRNA injection. Maximal interferon yields were ob- tainedafter 24-48 hr of incubation. In additional experiments, maximal yields wereobtained around 32 hr after injection (data notshown). Total cellular polyadenylylated RNA from induced cells gave similar results (data not shown). An incubationperiod of 48 hr was therefore chosen for subsequentexperiments.

It is noteworthythat nuclear poly(A)-containing RNA, when injected into oocytes, resulted inthe synthesis of interferon to a level approximately 15-20% ofthat obtained with cytoplasmic RNA (Fig. 2). Appropriate control experiments (not shown) indicated that these molecules did not represent cytoplasmic contamination of the nuclearRNA preparation.

The dependence of interferon synthesis in oocytes on the amount of mRNA injected is illustrated in Fig. 3. Detectable interferon yields were obtained by injection of approximately 1 ng of RNA per oocyte. There was a linear increase in thein- terferon yield withincreasing amounts of injected RNA in the range of 1.2-20 ng of RNA. A maximum titer of 192 reference units/mi wasobtained by injecting 80 ng of RNA per oocyte.

The leveling off of the dose-response curve above 20 ng may be the result of injecting sufficient mRNA to saturate the translational capacity of the oocytes. Alternatively, the leveling off maybe due to saturation of the posttranslational modifi- cationmachinery that may operate in the synthesis of biologi- cally activeinterferon in oocytes (46). The reproducibility of this assay is demonstrated by the observation that the same mRNA preparation gave identical interferon yields in three separate experiments (80 ng of RNA injected per oocyte, fol- lowed by incubationfor 48 hr, gave 192 reference units/ml in experiments summarized in Table 1 and Figs. 2 and 3). Our overall experience indicates that variation in the oocyte assay is no greater than variation in the interferon titrations (a2-fold error). We have used the oocyte assay to measure the concen- tration of biologically activeinterferon mRNA during different stages ofinterferon production in FS-4 cells.

Synthesis and Inactivation of Interferon mRNA in Human

Cells. The extraction of polyadenylylated mRNA and its

translation in X. Iaevis oocytes, described in thepreceding

Table 1. Synthesis of human fibroblast interferon inoocytes Interferontiter, Condition of interferonassay ref.units/ml

Human GM258 cells 192

Human GM258 cells+anti-F antiserum <8 Human GM258 cells+anti-Le antiserum 192

Mouse L cells <4*

Confluent FS-4 cells in 17 roller bottles (490 cm2 each;Corning) were exposed topoly(I)*poly(C) (12.5mg/mlin Eagle's minimumes- sential medium) in the presence of cycloheximide (50jg/ml)for4 hr.

Atthe end of this period, the cultures were washed four times with ice-coldPi/NaCland the cells were harvested by scraping. The cell pellet (about 1 ml packed volume) was resuspended in one-third- strengthreticulocyte standard buffer and the cytoplasmic and de- tergent-washed nuclear fractions were isolated. RNA in the two fractions was obtainedby phenol extraction, andpoly(A)-containing RNA was selected by poly(U)-Sepharose chromatography. Poly- adenylylated RNA was dissolved in 100 gl of distilled water. Each of 30oocytes was injected with approximately 80 ng ofcytoplasmic poly(A)-containing RNA (about 150 nl per oocyte) and the oocytes wereincubated for 48 hr in modified Barth's medium in a total volume of 0.6 ml. The oocytes were thenhomogenized in the same medium andaliquots of theclarified homogenate were assayed for interferon.

Homogenates frommock-infected oocytes did not contain detectable amountsof interferon (also see Fig. 4). Thesensitivity of the L cells to mouseinterferonand thespecificityof the antiserawereverified inseparate control assays.

*A dilution of 1:4 was tested.

section, was scaled down for smallerquantitiesofcells. Ten 150-mmpetri-dish cultures provided enough cells (about0.3 mlpacked volume)toallowthe isolation of sufficientquantities oftotalcellular polyadenylylated RNA. Answers to twokey questions were sought. First, did the shutoff of interferon production involve a loss of translatable interferon mRNA?

Second, what effectdidsuperinducing regimens such asthe presence of DRBthroughout the production periodhaveon interferonmRNAcontent?

Fig.4 andTable2presentasummary of the results obtained.

As reported earlier (see the introduction), poly(I)-poly(C)- induced FS-4cells showed aburst ofinterferon synthesis that peakedatapproximately3hr after induction andwasshut off by6-8 hr(Fig. 4). Inthe presenceof 40 ,uM DRB,the rate of interferonproduction showedadistinct lag and then continued atahighlevel for anextendedperiodof time(upto 4days) (12, 13).Wedetermined the relative content ofinterferonmRNA inuninduced cultures, in inducedcultures near the peak of production (2.5 hr),ininducedcultures in the shutoff phase (6 hr), andincultures induced and maintained in

40MgM

DRBfor 6hr. It is clearfromFig.4and Table2thatthe shutoff ofin- terferonproduction wasaccompaniedby a loss or inactivation oftranslatable interferon mRNA ininduced cells. Asimilar result was obtained when polyadenylylated RNA from the cytoplasmic fractionwastested forinterferonmRNAactivity.

Thepresence ofDRBinthecultures prevented this loss orin- activation(Fig.4and Table2). The datainTable2showthat the lowyield of interferon obtainedinoocytesinjected with mRNApreparationsfrom culturesintheshutoff phase wasnot anexperimentalartifact becausemRNAsamplesfromunin- duced culturesorfrom culturesinthe shutoffphasedid not interferewiththetranslation of known amounts of interferon mRNA.

Table2also showsthatRNAfrominterferon-producing cells thatdid notbindtopoly(U)Sepharose hadnointerferonmRNA activity. However,theseflow-throughRNA preparationsdid interfere with the translation of added interferon mRNA,

presumably

because of theirhighcontentofribosomal RNA andcontaminatingDNA. This result suggeststhatitmay be

Proc. Nati.Acad. Sci. USA 74(1977)

0

L4c

lL-

._ a)

Oc

.0

4m

0 Q-

OL ':

VAE

c

00

0

0

1 2 3 4 5 6 -

Hours

FIG. 4. Correlation between interferonproduction and interferon mRNAcontentof FS-4 cellsduring induction, shutoff,andsuperin- duction. Ten 150-mm Falconpetri-dishcultures of confluentFS-4 cells^wereused for eachanalysis. Thecultures wereinduced with poly(I).poly(C) (20Ag/ml)in 10mlEagle'sminimumessentialme-

dium for 1 hrstartingat"0hour." One batch ofcultures didnotre-

ceivepoly(I).poly(C)whereas another batch receivedpoly(I)-poly(C) togetherwith40MgMDRB. The DRB-treated batch^wasmaintained in DRBthroughoutthe experiment. Two of theinducedcultureswere

usedtomonitor therateof interferonproduction by hourlymedium (10 ml) change;oneculturewasfrom the DRB-freebatch(o)and the otherwastreatedwith DRB(o).DRB-containing samplesweredi- alyzed priortothe^assayfor interferon. At 2.5hr,abatch of induced (*)andabatch of uninduced(A)cultureswereharvested,andtotal cellular RNA was fractionated into polyadenylylated RNA and flow-throughRNAby usingapoly(U)-Sepharosecolumn. Another batch of induced cultures(U)and the DRB-treated cultures(0)were

similarlyharvestedat6 hr after induction. ThepolyadenylylatedRNA and theflow-throughRNAwereethanolprecipitatedand dissolved in 20and100,l,respectively, of injection buffer. Approximately2,l

of eachpolyadenylylatedRNApreparationwasinjectedinto 13-15 oocytes (30 ng of RNA peroocyte) and interferon synthesis was

monitored.

prematuretoconclude that the poly(A)-lacking fractiondoes notcontainanyinterferonmRNA(24-26).

DISCUSSION

In previousreports,wehavesystematically evaluated several different hypotheses that might explain superinduction of human interferonproduction(forreviewsseerefs. 10and 11).

Ithas becomeapparent,onthe basis ofindirectexperiments, thatthe enhancement of interferonproductioninFS-4cellsby inhibitorsofRNAorproteinsynthesis is notrelatedtoanin- hibitionofintralysosomal protein degradation (19) ortoan

inhibition of mRNAcompetitionduringthe shutoffphase (7).

However, evidence has beenobtained, byusinginhibitors such

asDRB,toimplicateaposttranscriptional regulatoryeventin the shutoff of human interferon production, mediatedby a

rapidlyturningoverRNAandpossiblybyprotein(9, 11-13, 17, 18,47). These dataaregenerally consistentwiththehy- pothesisthatthe shutoff ofpoly(I)-poly(C)-induced interferon synthesisinFS-4cellsismediatedbyatranslationalrepressor

of theMcAuslan-Tomkinstype(14,15,20-23).McAuslan(20) andTomkinsand hiscolleagues (21-23) suggestedthe possi- bilitythattranslation of mrRNA ineukaryoticcellsmightbe regulatedinaspecific mannerby rapidlyturningoverRNA andproteinmolecules thatservetorepress,inactivate,orde- grademRNA. Inthepresentstudy,wehave carriedoutadirect biochemicaltestof thishypothesisintheinterferonsystem.

Wehave demonstrated that interferon mRNA is rapidly inactivatedordegraded duringthe shutoffphasebutispro-

tectedinthepresenceofDRB,aselective inhibitor ofnuclear heterogeneousRNAandmRNAsynthesis(28,29,48).Thefact that interferon mRNAisinactivatedduringtheshutoffphase inFS-4 cells has also been observedbyR. L. Cavalieri,E. A.

Table 2. Interferon mRNA activity of RNA preparations Interferon titer,ref.

units/ml Flow- Poly(A)+ through

Cell cultures used RNA RNA

Uninduced,2.5hr <4 <4

Induced, 2.5 hr 48 <4

Induced, 6 hr 4 <4

Induced, DRB-treated, 6 hr <4

Alone 128

+Uninduced, 2.5hr,poly(A)+ RNA 48 (64*) +Induced, 6hr,poly(A)+ RNA 192 (64*) Induced, cycloheximide-treated, 4 hr

Alone 384

+Uninduced, 2.5hr,flow-through <4(192) +Induced, 6hr,flow-through <4 (192)

RNA preparationsdescribed in this tablecamefromtheexperi- mentspresented in Fig.4and Table1.Approximately 30 ng of poly- adenylylatedRNA[poly(A)+RNA]or500 ngofflow-throughRNA wasinjected per oocyte. In carryingoutthe mixingcontrols, equal volumes (2MO)ofthe appropriatesamples weremixedand then in- jectedinto oocytes.Hence,interferonyieldsfrominjectingmixtures would be half of those obtainedby injecting interferon-mRNA-con- taining samples alone.Expected titers are indicated in parentheses.

Interferon mRNA from induced,cycloheximide-treated cells was cytoplasmicpolyadenylylatedRNA as described in Table1.

*Deviationsindicate experimental variationandare notsignificant, asdemonstratedbyotherexperiments.

Havell,S. Pestka,andJ. Vil6ek (personalcommunication)using the oocyte assay. These results differ from those reported re- cently by Raj andPitha (49). The observation that interferon mRNAisinactivated or degraded during the shutoff phase is entirelyconsistentwiththeMcAuslan-Tomkins repressor hy- pothesis.

Aproper interpretationofthe data presented in this com- municationrequiresthat thebioassay for interferon correctly measurethequantityofinterferon proteinsynthesized. Atthe present time,thevalidityof thebioassay cannot be rigorously proven. Nevertheless, Fig. 3shows thatalinear relationship existsbetweentheamountofmRNAinjectedintooocytesand the interferonyieldobtainedoverthe rangeof 1-20 ng ofRNA injected per oocyte. Thisrealtionship (slope about 45°) suggests that thesynthesis ofbiologicallyactiveinterferoninthe oocyte translation assay correctly reflectstheamount ofinterferon synthesized,atleastoverthis rangeofinjectedmRNA.

Wehaveconfirmed that oocytes produce active interferon that possesses thesame speciesspecificity and the same im- munologicalcharacteristicsasinterferonsynthesized bycells from wihch themRNA wasextracted(26,27). Furthermore, ithas been shownbyother investigatorsthat,wheninjectedinto X.laevisoocytes,interferon mRNA obtained fromFS-4cells bringsabout thesynthesisof interferon which hasanapparent molecularweightof 21,000(26,27).

It is known that oocytes can

correctly cleave, acetylate,

phosphorylate, and'hydroxylate

polypeptides

synthesizedin responsetoinjectedmRNA(40,50).In

preliminary

experiments withratgrowthhormonemRNA(P.M.Sussman,P. B.Sehgal, andC.Bancroft,

unpublished data),

wehave observed thatX.

laevis oocytes accumulate the processed

growth

hormone polypeptideandnotpregrowthhormone(51).Itappearsthat X. laevis oocytesefficiently accomplishtheposttranslational modifications of the interferon molecule necessarytogenerate biologically^active

interferon,

atleastoverthe linear portion of thecurve'presentedinFig.3. Ourfailuretoobserveinter- feron synthesis in cell-free translation systems, such as the 3412 Cell

Biology: Sehgal

etal.

wheat-germ

system, mayreflect the

inability

ofour

ce11-free

systemsto

perform

these modifications. i

The loss oftranslatableinterferonmRNA

(Fig.

4and Table

2)

fromcellsinthe shutoff

phase

indicates thatthe shutoff of interferon

production

is notdueto

simple displacement

ofin- terferonmRNAfrom

polysomes by

competingcellularmRNA

(52).

^Itisapparent thatinterferonmRNAdoesnot survive in FS-4cellsinanuntranslatedbut

phenol-extractable

state ashas beenobserved

during

the shutoff of

vitellogenin synthesis

in avianliverafteraprimarystimulation withestrogen

(53).

The data obtainedintheinterferon systemarealsoquitedifferent fromthose

reported by

Killewichetal.

(54).

Theseinvestigators showed that theconcentrationofthemRNAfor

superinducible tryptophan 2,3-dioxygenase,

measuredinaKrebsascitescell- free translation system

during superinduction

of theenzyme activity

by

actinomycinD,wasidenticaltothatin

actinomycin

D-freecontrols. The interferonmRNAcontentof cellssuper- induced withDRBis

markedly higher

than that of

comparable

cultures6 hr after induction. It appears very

likely

that the

regulation

of human interferon

production

involvesaMcAus- lan-Tomkinstype

translational

repressorwhichinactivates or

degrades

interferonmRNA.

NoteAddedinProof.Wehaveobserved that therateofinterferon productionandtheconcentrationoftranslatableinterferonmRNA inFS-4cellsinduced and maintainedinthe presence ofDRBdecay with virtuallyidentical half-lives. Thus, continued interferon pro- ductioninDRB-treatedcellscanbeentirely explainedintermsofthe level of functional interferonmRNA.

WethankDsJ.Vil6ek andE.A.Havell for theircontinuedinterest inthisinvestigation.WealsothankthemaswellasDrs.S.Pestka and R.L.Cavalieriforsharingtheirunpublishedobservationswithus.We aregratefultoDrs.J.E.DarnellG.Blobel,andN.-H.Chuaforhelpful discussion.WethankDr.ArthurF.Wagner,MerckSharpand Dohme ResearchLaboratories,foragiftofDRB.Thisinvestigationwassup-

portedinpartbyResearchGrant CA-18608andbyProgram Project Grant CA-18213awardedbythe NationalCancer Institute. P.B.S. is apostdoctoralfellow(IF32CA05900-01)of theNationalCancer In- stituteand B.D. helda fellowshipfromthe DeutscheForschungs- gemeinschaft.

1. Knight,E., Jr.(1976)Proc.Natl.Acad.Sci. USA73,520-523.

2. Bose,S., Gurari-Rotman,D., Ruegg,U.Th., Corley,L. &An-

finsen,^{C. B.}(1976)J. Bio!.Chem. 251,1659-1662.

3. Havell,E.A., Berman, B.,Ogburn,^C.A., Berg, K.,Paucker,K.

&Vilek,J.(1975)Proc.Natl.Acad.Sci.USA72,2185-2187.

4. Field,A. K.,Tytell,A. A.,Lampson,G. P. &Hilleman,M. R.

(1967)Proc.Natl.Acad.Sci. USA58,1004-1010.

5.

Vikcek,

J. &Havell,E.A.(1973)Proc.Natl.Acad.Sci.USA70, 3909-3913.

6. Ho, M. & Armstrong,J. A. (1975) Annu.Rev. Microbiol. 29, 131-161.

7. Sehgal,P. B. &Tamm,I.(1976)^Proc.Natl.Acad.Sci.USA73, 1621-1625.

8. Falcoff,E.,Havell,E. A., Lewis, J.A.,Lande,M.A.,Falcoff,R.,

Sabatini,

D.D. &

Vikcek,

J.(1976)Virology 75,384-393.

9.

Vikcek,

J.,Havell,E. A.&Kohase,M.(1976)J.Infect.Dis. 133, A22-A29.

10. Tamm,I.&Sehgal,P. B.(1977)Adv.Virus Res.22,inpress.

11. Havell,E. A.(1977)inInterferon,ed. Stewart,W.(CRCPress,

Cleveland,

OH),inpress.

12. Sehgal, P.B., Tamm, I. &Vilcek,J. (1975)Science 190,282- 284.

13.

Sehgal,

P. B., Tamm,I. &Vilcek,J.(1976) Virology 70,532- 541.

14. Vilcek,J., Rossman,T.G. & Varacalli,F. (1969)Nature222, 682-683.

15. Tan, Y. H., Armstrong, J.A., Ke,Y.H. & Ho,M.(1970)Proc.

Nat!.Acd.Sd.. USA 67,464-471.

16. Kohase,M.&Vilkek,J.

(1977) Virology

76,47-54.

17.

Sehgal,

P.B.,Tamm, 1.&

Vil6ek,

J.

(1976) Virology

70,542-

544.

la Tamm, I. & Sehgal, P. B. (1977) J. Exp. Med. 145, 344-356.

19. Sehgal,P. B., Tamm, I. &Vilkek,J.(1975)J. Exp.Med. 142,

1283-1300.

20. McAuslan,B. R. (1963) Virology21,383-389.

21. Garren, L. D., Howell,R.R., Tomkins,G. M. &Crocco, R. M.

(1964)Proc.Natl.Acad. Sci. USA 52, 1121-1129.

22. Tomkins, G. M., Gelehrter, T., Granner, D., Martin, D. W., Samuels, H. & Thompson, E. B. (1969) Science 166, 1474- 1480.

23. Tomkins,G.M., Levinson,B.B., Baxter, J.D. & Dethlefsen, L.

(1972)Nature New Biol. 239, 9-14.

24. Thang,M.N., Thang, D. C., De Maeyer, E. & Montagnier, L.

(1975)Proc.Natl. Acad. Sci. USA 72,3975-3977.

25. Pestka, S.,McInnes, J., Havell,E. A. &Vil6ek, J. (1975)Proc.Natl.

Acad. Sci. USA 72,3898-3901.

26. Reynolds,F.H., Premkumar,E. &Pitha,P. M.(1975)Proc.

Natd.

Acad.Sci. USA 72,4881-4885.

27. Cavalieri,R.L., Havell,E.A.,Vilkek, J.& Pestka,S.(1977) Proc.

Nat!.Acad.Sci.USA, 74,3287-3291.

28. Tamm,I.,Hand,R. &Caliguiri,L. A. (1976)j. CellBiol. 69, 229-240.

29. Sehgal, P. B., Darnell, J. E. & Tamm, I. (1976) Cell 9, 473- 480.

30. Havell,E.A. & Vil6ek,J.(1972)Antimicrob.AgentsChemother.

2,476-484.

31. Eagle,H.(1959)Science130, 432-437.

32. Tan,Y. H.(1976)Nature260,141-143.

33. Dulbecco,R.&Vogt, M.(1954)J.Exp.Med. 99, 167-182.

34. Armstrong,J.A.(1971)Appl.Microbiol.21,723-725.

35. Penman, S.,Vesco,C. & Penman,M.(1968)J.Mol. Biol. 34, 49-69.

36. Penman,S.(1969)inFundamentalTechniquesinVirology,eds.

Habel,K. & Salzman,N. P.(AcademicPress, NewYork),pp.

35-48.

37. Adesnik,M.,Salditt,M.,Thomas,W. &Darnell,J. E.(1972)J.

Mol. Biol. 71, 21-30.

38. Molloy,G.R.,Jelinek,W.,Salditt,M.&Darnell,J.E.(1974) Cell 1,43-53.

39. Derman,E.&Darnell,J.E.(1974)Cell3,255-264.

40. Gurdon,J.B.(1974)inThe ControlofGeneExpressioninAn- imal Development (Harvard University Press, Cambridge, MA).

41. Villa-Komaroff,L., Guttman,N.,Baltimore,D.&Lodish,H. F.

(1975)Proc.Natl.Acad.Sci. USA 72,4157-4161.

42. Dobberstein, B. & Blobel,G. (1977) Biochem. Biophys. Res.

Commun.74,1675-1682.

43. Finter,N. B.,ed. (1973)InterferonsandInterferon Inducers (North-HollandPublishingCo.,Amsterdam;AmericanElsevier PublishingCo., Inc.,NewYork).

44. Blobel,G. &Dobberstein,B.(1975)J. CellBiol. 67,835-851.

45. Blobel,G. &Dobberstein,B. (1975)J.CellBiol. 67, 852-862.

46. Havell,E. A.,Vilcek,J.,Falcoff,E.&Berman, B.(1975)Virology 63,475-483.

47. Sehgal,P. 1.,Tamm, I. &Vilcek,J.(1976) Virology 70,256- 259.

48. Sehgal,P.B., Derman, E., Molloy,G. R., Tamm, I. &Darnell, J.E.(1976)Science194,431-433.

49. Raj,N.B. K.&Pitha,P. M.(1977)Proc.Natl.Acad.Sci.USA74, 1483-1487.

50. Lane,C. D. &Knowland,J. S.(1975)inBiochemistryofDevel- opment, ed. Weber, R. (William Heineman Medical Books, London; J. B. Lippincott Co., Philadelphia, PA), Vol. 3, pp.

145-181.

51. Sussman,P.M.,Tushinski,R.J. &Bancroft,F.C.(1976)Proc.

Nat!. Acad.Sc. USA73,29-33.

52. Palmiter, R. D. & Schimke, R. T. (1973)J. Biol. Chem. 248, 1502-1512.

53. Mullinix, K. P., Wetekam, W., Deeley, R. G., Gordon, J. I., Meyers, M., Kent,K.A. &Goldberger,R.F.(1976) Proc.Natl.

Acad.Sci. USA 73,1442-1446.

54. Killewich,L.,Schutz,G.&Feigelson,P.(1975)Proc.Nat!.Acad.

Sci.USA 72,4285-4287.