The S2 gene nucleotide sequences of prototype strains of the three reovirus serotypes: characterization of reovirus core protein sigma 2.

(1)

The

S2

Gene Nucleotide Sequences

of

Prototype

Strains of

the

Three

Reovirus Serotypes:

Characterization

of

Reovirus

Core

Protein

o-2

TERENCES. DERMODY,

12t*

LESLIE A.

SCHIFF,lt

MAXL.

NIBERT,13

KEVIN M. COOMBS,1§

AND BERNARD N. FIELDS" 2'4

Departmentof Microbiology and Molecular Genetics' and ShipleyInstituteof Medicine,4 Harvard Medical School, and

Departments

of Medicine2

and

Pathology,3 Brigham

& Women's

Hospital,

Boston,

Massachusetts 02115

Received 11 April 1991/Accepted 22 July 1991

The S2genenucleotidesequencesofprototypestrainsof the three reovirusserotypesweredeterminedtogain

insight into thestructureandfunction of the S2 translation product, virioncoreprotein (72. The S2sequences

of thetype1Lang,type2 Jones, andtype3Dearing strainsare1,331 nucleotidesin length and containasingle

largeopenreading framethatcould encodeaprotein of 418 amino acids, correspondingto r2. The deduced

(2aminoacidsequencesofthesestrainsareveryconserved, being identicalat94% ofthesequencepositions.

Predictions of i2secondarystructureandhydrophobicitysuggestthatthe proteinhasatwo-domainstructure.

Alarger domain is suggestedto beformed from the amino-terminal three-fourths ofcr2sequence, which is

separated fromasmallercarboxy-terminaldomain byaturn-rich hinge region. The carboxy-terminal domain

includessequencesthataremorehydrophilicthan those in therestof the proteinandcontainssequenceswhich

arepredicted toformana-helix. Aregionof striking similarity wasfound between amino acids 354 and 374

of r2andamino acids 1008and 1031ofthe,1subunit of theEscherichia coli DNA-dependent RNApolymerase.

We suggestthat theregions withsimilarsequencein (r2 and the

0

subunit form amphipathic a-helices which may play a related role in the function of each protein. We have also performed experiments to further

characterize the double-stranded RNA-binding activity of f2 and found that the capacity to bind

double-strandedRNAisaproperty of the ff2proteinofprototype strains andoftheS2 mutanttsC447.

Virionsof the mammalian reoviruses consist ofa double

protein shell formed by an innercore and an outer capsid.

The virion core contains the 10 double-stranded RNA

(dsRNA) genome segments and consists of three major

proteins (Xl, X2, and a2) andtwominorproteins (A3 and ,u2) (63). Previous studies ofcorestructure suggest that Al and a2arecomplexed (33) andthatcr2 islocated primarilyonthe

innersurface of thecore(68). Pentamers of the othermajor

coreproteinX2 formthecorespikeslocatedatthevertices of

the virion icosahedron (54). Crystals of the reovirustype 3

Dearing (T3D) core have been isolated and characterized

(16); however, structural resolution at the atomic level

awaits furtherstudies.

The intact reovirus core contains all of the enzymatic

activitiesnecessaryfor viral RNAtranscription and

replica-tion(13); however, solubilization of the core results in the

loss of these activities (68). Insight into the function of individualcoreproteinshas beenpossible through biochem-ical and genetic studies and through an analysis of their

deduced amino acid sequences. The X1 protein has been

shown to bind reovirus dsRNA and zinc in blotting assays

(60), and Al has a nucleotide-binding sequence and a TF

IIIA-likezincfingermotif(6).The X2proteinisthe reovirus

guanylyltransferase (15),and X2 hasaGTP-bindingsequence

*Correspondingauthor.

tPresentaddress: Lamb Center for PediatricResearch,Division

of Pediatric Infectious Diseases, Vanderbilt University Medical Center,Nashville,TN37232.

tPresent address: Department of Microbiology, University of

MinnesotaMedicalSchool, Minneapolis,MN 55455.

§Present address: Departmentof MedicalMicrobiology,

Univer-sityofManitoba, Winnipeg, Manitoba,Canada R3EOW3.

(62). The pH dependence of transcription is genetically

associated with the X3-encoding Li gene segment (20),

suggestingthatthe X3protein is involved inreovirus

RNA-dependent RNA polymerase activity. Furthermore, X3 has

sequence similarity to the RNA-dependent RNA

polymer-ases of several viruses, further suggesting that it is a

com-ponent of the reovirus RNA polymerase (43, 44). The a2

protein is encoded by the S2 gene segment (42, 46), and

similarto theother reovirus coreproteins, thefunctions of

r2 are not precisely understood. However, a

temperature-sensitive(ts) mutantof T3D whose mutationmapstothe S2 gene(tsC447) synthesizesapproximately5%of thewild-type

level ofsingle-strandedRNA(ssRNA)and assembles virions which contain only 0.1% of the normal amount of dsRNA when grown at the restrictive temperature (57). Further-more, Schiffetal. (60) have shown that c2, like Xl, binds

reovirus dsRNAinablottingassay.Thus,these datasuggest that cr2 may play arole in viral RNA synthesis orin virion

assembly.

In ordertolearnmoreaboutthestructureand function of

a2, we determined the S2 gene nucleotide sequences of prototypestrains of the three reovirus serotypes. Wefound

thatthe deduced cr2 amino acidsequencesofthetype1Lang

(T1L), type 2 Jones (T2J), and T3D strains are of equal

length and are highly conserved. Predictions of protein

secondary structure, hydrophobicity, and surface probabil-itysuggest atwo-domain modelfor the structure of cr2.The

amino-terminal three-fourthsof the cr2sequenceissuggested

to form alarge domain; the carboxy-terminal one-fourth of

cr2 is suggested to form a small domain which contains a

regionofsequence with similarityto aregion in the

Esche-richiacoliDNA-dependentRNApolymerase ,Bsubunit. We

have also extendedstudies of thedsRNA-bindingactivityof 5721

0022-538X/91/115721-11$02.00/0

CopyrightC) 1991, American Society forMicrobiology

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

u2 in a Northwestern (RNAblotting) assay and have found that thecharacteristics ofu2 binding todsRNA are clearly

distinct from those of other reovirus dsRNA-binding

pro-teins. This studyrepresents aninitialstep in the understand-ing of structure-function relationships of the reovirus core protein, u2.

MATERIALSANDMETHODS

Cells and viruses. Spinner-adapted mouse L929 (L) cells

were grown in either suspension or monolayer cultures in

Joklik's modified Eagle's minimal essential medium (Irvine Scientific, Santa Ana, Calif.) that was supplemented to

contain5% fetal calfserum (HycloneLaboratories, Logan,

Utah),2 mMglutamine, 1Uofpenicillinperml,and 1 ,ugof streptomycin per ml (Irving Scientific). Reovirus strains

T1L, T2J, and T3D are laboratory stocks. Second- and

third-passage L-cell lysate stocks of twice-plaque-purified

reovirus were used to makepurified virionpreparations(25).

Sequencing dsRNA with primers. Genomic dsRNA was

purified from virions according to the method describedby Nibert et al. (49). The dsRNA sequences were determined

by using dideoxy sequencing reactions witholigonucleotide

primers, reverse transcriptase (Boehringer Mannheim

Bio-chemicals, Indianapolis, Ind.), and

[35S]dATP

(1,000 Ci/ mmol;Amersham, Arlington Heights, Ill.),accordingtothe

methoddescribed byBassel-Dubyetal.(8). Oligonucleotide primers weremade with anApplied Biosystems oligonucle-otide synthesizer (Applied Biosystems, FosterCity, Calif.) and werepurified by C-18 Sep-Pak column chromatography (WatersAssociates, Milford, Mass.) (3). Initial

primers

were

synthesized corresponding to the T3D S2 gene nucleotide sequence (12), and additionalprimers corresponding to the TlLandT2JS2genenucleotidesequenceswere

synthesized

as necessary. Primers were synthesized at approximately

200-nucleotide intervalsinordertodeterminethefull-length

sequence of each complementary strand.

Analysis of the S2 nucleotide and deduced u2 amino acid

sequences. Sequence alignmentwasfacilitated bytheglobal alignmentprogramGAP (19) fromtheUniversity of Wiscon-sin GeneticsComputer Group (UWGCG;

Madison, Wisc.).

Open reading frames in the S2 nucleotide sequences were

determined by using the David Mount sequence

analysis

package (Genetics Software Center, Tucson, Ariz.).

Deter-minations ofdeduced

ur2

protein amino acid composition,

molecular weight, and isoelectric point were facilitated by

the protein sequence analysis program PEPfrom the

Intel-ligenetics Suite (IntelIntel-ligenetics, Mountain View, Calif.).

Predictions ofprotein structure. Secondary-structure

pro-pensity values for a2 were assigned to each amino acid residue according to the method of Chou and Fasman (14)

(a-helix,

,-strand, and ,B-turn) and Leszczynski and Rose

(40)

(fQ-loop),

and plots of secondary-structure

predictions

foru2 weregenerated by the

technique

of Nibertetal. (49).

Secondary-structure

predictions were also derived by the

technique of Ralph et al. (55) as applied in the program

PRSTRUC, provided by the Molecular Biology Computer

Research Resource (MBCRR; Boston, Mass.), using the

secondary-structure propensity valuesofChou and Fasman

(14) and Leszczynski and Rose (40). For each

secondary-structure propensitytype, the seedthreshold andextension limit values defined by Ralph et al. (55) were used. This

technique

generated morethan a

single

secondary-structure prediction for some positions in the

u2

amino acid

se-quences. Hydropathicity index valuesfor r2 were

assigned

toeachamino acid residue

according

tothemethod ofKyte

and Doolittle(36),and

hydropathicity plots

were

generated

by

a modification ofthe

technique

of Nibert et al.

(49),

in which the

hydropathicity

valuesof individual residues were

averaged

for each seven consecutive

positions

in the u2 sequence. Surface

probability

valuesforcr2were

assigned

to

eachamino acidresidue

according

tothe method of Eminiet

al.

(22),

and surface

probability plots

were

generated by

using

the program PEPTIDESTRUCTURE from the

UWGCG.

Data basesearches. Searchesofthe National Biomedical

Research Foundation

(NBRF)

data base were

performed

with the BLASTP program

(1)

provided by

the MBCRR.

Searches of the NBRF and Swiss

protein

data bases were

also

performed

withtheFASTAprogram

(52)

through

BIO-NET

(Mountain

View, Calif.).

Searchesofa

protein

sequence

pattern data base were conducted with the PLSEARCH

program(64),

provided by

theMBCRR.Searches for

protein

sequences of

potential biologic

significance

werefacilitated

by

the PROSITEprogram from PCGENE

(Intelligenetics).

SDS-PAGE ofviral

proteins.

Discontinuous sodium

dode-cyl

sulfate-polyacrylamide gel

electrophoresis (SDS-PAGE)

was performed as described by Laemmli (37). Gels were

stained with Coomassie brilliant blue R-250

(Sigma

Chemical

Co.,

St.

Louis, Mo.)

and dried between

cellophane.

Alter-natively, gels containing 35S-labeled

proteins

were treated with

Enlightning

(New

England

Nuclear

Corp., Boston,

Mass.),

dried under a vacuum, and

exposed

to XAR film

(Eastman .Kodak

Co., Rochester, N.Y.)

at

-70°C.

Northwestern

blotting.

Proteins were

electrophoretically

transferred from

SDS-polyacrylamide gels

tonitrocellulose

(Trans-Blot;

Bio-Rad,

Richmond, Calif.) by

the Western

immunoblotting technique (66)

with

Tris-glycine

transfer

buffer that contained 0.01% SDS and 20%

(vol/vol)

metha-nol. Northwestern blots

(10)

were

performed

asdescribed

by

Schiffet al.

(60).

After the

electrophoretic

transfer of

pro-teins to

nitrocellulose,

unreacted sites on the membrane

were blocked with standard

binding

buffer

(10

mM

Tris-hydrochloride [pH 7.0],

1 mM

EDTA,

50 mM

NaCl,

0.04%

bovine serum

albumin,

0.04% Ficoll

400,

0.04%

polyvi-nylpyrrolidone-40).

The blotted

proteins

were

probed

with

32P-labeled

reovirus

genomic

dsRNA

(1

x

104

to 5 x 104

cpm/ml)

in

binding

buffer modifiedtocontain25 mM NaCl

and 10 mM

Tris-hydrochloride

(pH

6.5).

UnboundRNAwas

removed

by briefly

washing

the nitrocellulose filterin mod-ified

binding

buffer. Nitrocellulosefilterswereair-dried and

exposed

toXARfilm

(Kodak)

at

-70°C

withan

intensifying

screen

(Cronex

Lightning-Plus;

E. I.du Pont de Nemours &

Co.,

Wilmington, Del.).

Thereovirus

genomic

dsRNA used

as a

probe

was

prepared

from virions ofT3D

(60).

Chemicalcleavageof r2 with formic acid.The

protocol

for chemical

cleavage

withformic acid described

by

Andersand

Consigli (2)

was used to generate

peptide

fragments

ofu2.

Formic acid

efficiently

and

specifically

cleaves

aspartate-proline linkages (39).

Virion

proteins

were

subjected

to

electrophoresis

ina 5to 10%

polyacrylamide-SDS

gel.

The

gel

wasfixed in30%

isopropanol-10%

acetic

acid,

stainedin

a

fixing

solution

containing

0.5% Coomassie brilliant blue

R-250,

and then destained in 16.5% methanol-5% acetic acid. Visualized

protein

bandswereexcised from the

gel

and

dried

by

lyophilization.

Forformic acidtreatment,

gel

slices

containing u2

wereswollenin 0.1 mlof 75% formic acid and

then incubated in screw-cap vials at

37°C

for 48 h. The

reactionwas

stopped by

removing

the formicacid

by

lyoph-ilization. Gel slices were

equilibrated

with2x Laemmli

gel

sample

buffer

(37)

until the

bromphenol

blue

(Sigma) pH

indicator remained blue.

Cleavage products

were

analyzed

on November 10, 2019 by guest

http://jvi.asm.org/

(3)

by electrophoresis of equilibrated gel slices in a S to 20% polyacrylamide-SDS gel.

RESULTS

Comparison ofthe S2 gene nucleotide sequences of

proto-type strains of the three reovirus serotypes. We determined

thenucleotide sequencesofthe S2 genes ofreovirus strains

T1L, T2J, and T3D by using S2 genomic dsRNA that was

purified from virions. The full-length S2 coding-strand se-quences are showninFig. 1. The S2sequenceofeach strain wasfound to be 1,331 nucleotidesin length and to contain a

single large open readingframe (ORF) that could encode a

protein of418 amino acids. When the S2 sequences of the threestrainswerecompared,theywere found to beidentical

at 958 of the 1,331 nucleotide positions (72% sequence

identity). In pairwise comparisons, the TlL and T3D S2

nucleotide sequences were found to have greater similarity

with each other (86% sequence identity) than either was

found to have with the T2J S2 sequence (78% identity for eachtwo-sequence comparison).

These results include the first S2 nucleotide sequence

reported forstrainT2J. TheS2sequences of strainsTlL and

T3D were reported previously (12, 27, 73) but differ in

several instances from the sequences determined in this study. In the case of the TlL S2 sequence, our results

include anadditional cytosine between nucleotides986and

987ofthepreviously reported sequence (27). In the caseof

the T3D S2 sequence, our results confirm the sequence

reported by Wiener et al. (73); both sequences include an

additional cytosine between nucleotides 945 and 946ofthe

previously reported sequence

(12).

The corrections of pre-vious TlL and T3D sequences aresignificantin thatthey add a single nucleotide to the cr2-encoding region of each S2 sequence andtherebypermittheORFencodingcr2 toextend

to nearer the 3' endofeach S2gene. As aconsequence, the

T1L, T2J, andT3D S2 genes havecoding-strand sequences

that contain small, identically sized 5' and 3' noncoding regions: 18and 56nucleotides, respectively (Table 1).

The determination offull-length S2nucleotide sequences for prototype strains of the three reovirus serotypes has

allowed us to identify several additional ORFs in the S2 genes ofthese strains.ThelargestORF that isconservedin all three S2 sequences (otherthan that

proposed

to encode

cr2) is 102 nucleotides in length (nucleotides 1214 to 1315)

and potentially encodes a

protein

of 34 amino acids. The

resulting T1L, T2J, and T3D

proteins

would have 44% sequence identity, but there is no evidence to suggest that

this shortORF istranslated into

protein

in

reovirus-infected

cells. Infact, the S2 sequences ofthe three strains contain

cytosines

at nucleotides 1211 and 1217

(the

-3 and +4

positions,respectively, withrespectto theinitiator AUG of

theshortORF),sothatthis ORFis

unlikely

tobetranslated accordingtotherulesof Kozak(35).

Comparison of the deduced

af2

amino acid sequences of prototype strainsof the three reovirus serotypes. We

exam-ined thededuced cr2 amino acid sequencesofthe

T1L,

T2J,

and T3D strainsin

order

to

identify

conservedand variable featuresofcr2.Protein lengthwasfoundtobeconservedin

30 60

TlL GCUAUUCGC UGG UCAGUUAUG GCU CGC GCU GCGUUC CUA UUC AAG ACU GUU CGAUUUGGU

T2J --- --- --- --- --- --- --A --- --C --- --- --- --- --- --C--- ---

--T3D --- --- --- --- --- --- --- --- --- --- --- --- --- --- -G

90 120

T1L GGU CUG CAA AAUGUGCCAAUU AAUGAU GAGUUA UCG UCA CAU CUA CUUCGAGCC GGUAAU

T2J --G--- --- --- --A--U --- --C --- ---C-UG-U --- --- --GU-A --U--A

---T3D --- --- --- --- --- --- --- --C--C--AC-- --U --- --- ---C- --U ---

---150 180

TlL UCGCCAUGGCAG CUGACU CAGUUCUUAGAU UGG AUA AGUCUU GGAAGAGGAUUAGCU ACA

T2J --U --- --- --- --- --G--A---C-U --C --- --U --- --C --G--- --- --G--- --C T3D --A--- --- ---U-A --A --- --U --- --C--- --- --C--- --G --G --U ---

---210 240

TlL UCA GCU CUUGUUCCA ACCGCUGGUUCAAGAUAU UAU CAGAUG AGUUGUUUA UUGAGU GGU T2J --U--AU-A--G --U --G --G --A--GC-U--C---A --- --- --C C---

---T3D --G --- --C--- --G --- C --A --- --- --CC-UC-A --- --C

270 300

TlL ACC CUC CAG AUUCCAUUUCGUCCUAAU CAU CGAUGG GCA GAU AUUAGG Wuu GUG CGUCUA

T2J --G--G --A --- --U--- --A--A--- ---A-- --- --- --CG-AC-A ---A --- --G

T3D --U --- --- --- --CG--C --- --- --C--C --- --- --- --C--- --- --CU-A--C

U--330 360

TlL CUG UGG UCA GCU CCU ACG CUU GACGCC UUGCUUCUUGCCCCACCC CAGGUC UUAGCU CAG

T2J --A --- --U--- --C--C --- --U--A---A-- --A --- --- -CC A-UC-U --G--A

T3D --- --- --- --- ---- -C--U --A --A --C--A --U --- --A--A--U--C --

--390 420

TlL CCA GCG UUA CAG GCCCAG GCAGAUCGAGUGUAU GAUUGUGAU GAU UAC CCAUUCUUA GCA

T2J --U---A-- --- --- --A--G--- A---CU--C--C --- --- --C--U --U---C-U

---T3D --C --U--G--A--A--- --- --- --- --C--C--C --- --- --U---U C----G

450 480

TlL CGU GACCCCAGA UUUAAGCAU CGA CUGUAU CAACAAUUGAGCGCCGUAACU CUGCUUAAC

T2J ---U --C C-C --C --A---A-- --A--- --G---C-U --U --G A-U --C --- --U T3D -U---U--A --- --C--A --- --G -C---- --G--- --- --U--U --- ---A---

---510 540

TlL UUG ACG GGG UUUGGU CCA AUU UCCUAUCUU CGAGUAGACGAG GAU AUGUGCAGU GGG GAU T2J C----U--C --C----G--A--A --C--G--G--U --U --- --C --- --- --- --U--C

T3D --- --A--U --- --C --G--- --- --C----C---G--U--A--- --- --- --- --A

---570 600

TlL GUGAAC CAGCUU CUCAUGAACUACUUCGGGCAU ACGUWWGCGGAAAUU GCGUAUACAUUA

T2J --- -G- U-A --- --- --U -U--U --U --C --U--C --C --- --C--- --C---

---T3D --- --- --- --- --- -- - U---- ---C-- ---A--C ---G

630 660

TlL UGUCAAGCU UCA GCCAAUAGA CCU UGGGAG CAU GAU GGU ACGUAU GCGAGGAUGACU CAA

T2J --- --- --- --CG--U--CC-- --A --- --A --- --- --A--A--- --- C-U --- ---

---T3D --- --- --C--C--U--- --G --AU----C --- --A--- --U--- --- --- --G

690 720

TlL AUU AUA CUGUCCUUAUUUUCG UUAUCAUAUGUCGGUGUA AUUCAU CAG CAGAACACG UAC T2J --AC-CU ---C - A--- -- --C -- --U

T3D ---G-GU-A ---G --C---C-- --G --- --- --- ---C--- --- --- --U--- --U

750 780

TlL CCGACGUUCUAUUUC CAAUGCAAUCGG CCC CGUGAUGCU GCU GMGUAUGCAUU CUUUCC

T2J --U --- --- --- --- --- --U---A-A--U --- --- --C--CG - G--- --A--G --G

T3D --- --A --- --- --U--C --U --- --A--- --C--C ----C---G--- --- --- --U

810 840

TlL UGCUCAUUAAACCACUCCGCC CAGAUUAGACCCGGUAAU CCC ACU CUAUUC GUCAUGCCA T2J --- --- --CG-CU--- --G--A --- --A--G G-U --A --- --U--C --U --- --- --C T3D --U -G --- --AU---A-- --- --- --- --- --- --U--CU-- --- --U---

---870 900

TlL ACAAGU CCA GAU UGGAAUAUG GACGUCAAUUWGAUUUUAAGUUCG ACG UUG ACA GGG UGC T2J --U--C--U --- --- --C--- --U --G--- --- --- C-C--- --AC-U --U--A

---T3D --U --C --- --- --- --C--- --- --- --- --- --CC-G- --A--- --- --G--- --U

930 960

TlL UWGUGU UCG GGC UCU CAG UUA CCGCUU AUUGACAAUAAC UCAGUGCCU GCGGUUUCG CGU T2J C-- --- --U--G--C --- --C--U U-G --- --U--C --- --U--C--AAAC--- --UA-G T3D --- --- --- --U--A--- C-G--A--G--- --- --- --U--- --A --- -- --GC ---

---990 1020

TlL AACAUCCAUGGUUCG ACU GGCAGAGCU GGCAAC CAG CUGCAUGCUUUC CAAGUGCGACGA

T2J --- --U--- --A --- --G---C-- -G- --U--U--AU-- --- --G--- --- --U--- --C

T3D --- --- ---C--- --- --U--- --- --AU---AU--- --G--- --G---A--

---1050 1080

TlL AUG GUGACUGAAUUCUGUGACAGAUUAAGACGCGAUGGA GCU AUG ACUCAAGCU CAGCAA

T2J --- A-U--- --- -AU --C --U--GC-G --G--U --- --C- -C -CC--G--- --G

T3D --- --- --- --U--- --- --G--G--- --- --- --U--C --C --- --- --- --G

1110 1140

TIL AAU CMAUU GAAGCG UUGGCAGAU CAAACU CAACAGUUUAAGAGG GAUMA CWCGAGGCG

T2J -UG--- --A --- --- --A-GG--- --G --G --G- ---C-U--C--G--- --- --C

T3D --- G-- --- --- --- --- --- --C--- --- --- --- --- --C--G--C --A

A--1170 1200

TIL UGGGCUAGG CMGAUGAU CAGUAUMUCAGGCUMU CCCMUUCU ACGAUGWUCCGUACG

T2J --- --- UU- --G--C---A--- --- -GC --GC----A C --G--A --- --UA-G--U T3D --- --G -- --- -ACC --A--- --- --- --- C-- --C--C--C--A--- --- ---

---1230 1260

TlL MG CCAUUUACG MU CGCAMUGG GGACGAGGAMUACCGGAGCG ACUAGUGCCGCAAUW

T2J --- --- --- --C--- --U --- --- --C--G--- --- -A--C- --- --- --C --A--G

---T3D --A --- --- --- --- --- --- --- --- --- --U----C---G--- --- --- ---

--G---1290 1320

TlL GCA GCCCUUAUC UMUCG UCU UGGAGUGAGCGG AUCCCCCCA CACCCCUCA CGACUGACC T2J --G --U --C --- ---C-- --- G-A--C -U----G-- --- --UC-U--G ---

---T3D --- --- --- --- -G- --- --- --- --- G-- --- --- --- --- --- --- ---

---1331

TlL ACACAU UCA UC T2J --CU-- ---

-T3D --- -

--FIG. 1. Coding-strandnucleotidesequencesofthereovirusT1L,

T2J,and T3DS2genesegments.Nucleotidepositionsarenumbered abovethe sequences,and residuesin T2JandT3D identicaltothose in TlLareindicatedby dashes.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.612.325.566.80.691.2]

(4)

[image:4.612.53.297.89.343.2]

TABLE 1. Lengthsof noncoding regions of reovirus

gene segments

Length of

Gene segment noncoding region

(references)

Virus strain(s)"

(no. ofbases)b

5' 3'

Li (72) T1L,T2J, T3D 18 32

L2 (62) T3D 13 33

L3 (6) T3D 13 181

M1 (69) T3D 13 80

M2(32,65,71) TiL, T2J, T3D 29 47

M3(69) T3D 18 57

Si (7, 11,21, TiL 13 34

45, 47, 49) T2J 13 38

T3D 12 36

S2(12, 27, 73; TiL, T2J, T3D 18 56

this study)

S3(26, 58, 70) TiL,T2J, T3D 27 70

S4 (4, 28) TiL,T3D 32 66

a Reovirus strains whose indicated gene segment nucleotide sequence has

beendeterimined.

bDetermined from referenced gene segment nucleotide sequences; the

termination codon ineach gene segment is not considered part of the 3'

noncoding region.

the strains (418 amino acids), and no gaps were needed to

align the sequences (Fig. 2). The deduced cr2 amino acid

sequences of the three strains are very conserved, being

identical at 393 of the 418 amino acid positions (94%

se-quence identity). As was the case for the S2 nucleotide

sequences, the deduced TlL and T3D cr2 sequences have

greater similarity with each other (99% sequence identity)

thaneithersequencehas withT2Jcr2

(94%

sequenceidentity for each two-sequence comparison). The most variable region inthecr2sequencesexhibits 88%sequenceidentity in

the three strains and is located in the carboxy-terminal

one-fourth of the protein (amino acids 311 to 418). The cr2

protein ofeachstrain containseight cysteine residues, which

areentirely conserved in thea2 sequences (Fig. 2). Proline residues account for 20 to 21 amino acids and aromatic residues account for 45 to 46 amino acids in each cr2

sequence. There are alarge number of charged residues in

each

o2

sequence, a cluster of which are located in the

carboxy-terminal one-fourth oftheprotein (amino acids 333 to380) (Fig. 2). Ofthecharged residues, basic amino acids predominate inthe cr2 sequences, so thateach

cr2

protein has

a calculated isoelectric point of 8.33. This value

approxi-matestheisoelectric point of7.9reportedpreviouslyforthe type 3 Abney cr2protein (23). Themolecularmasses ofthe

threea2proteins calculated fromthe deduced cr2 sequences

are as follows: T1L, 47,130 Da; T2J, 47,065 Da;and T3D,

47,181Da. Thesemasses arewithinthe rangeofMrs(45,500

to 53,500) previously determined for

(r2

by SDS-PAGE (9, 56, 59).

Computer-based predictions of r2 structure. We used a

number ofcomputer-based techniques to gain insight into thestructure ofa2. Because the cr2 sequencesof T1L, T2J,

andT3Dare so

similar, analyses

carriedoutwitheachofthe

individual sequences exhibited

only

minor variations from

those oftheothers (datanot

shown).

A

hydropathicity plot

derived forT3Dcr2

by

using

the

hydropathic

indexvaluesof KyteandDoolittle

(36)

is shownin

Fig.

3.This

analysis

was

notable fora

region

nearthe

carboxy

terminus of

cr2

(amino

acids 340to

410)

that is more

consistently

hydrophilic

than

theremainderofthe

protein.

Incontrast, theamino-terminal three-fourths ofcr2 is characterized

by

nonregular

fluctua-tions between

regions

of moderate to

high

hydropathicity

and

hydrophilicity.

The

unique

natureofthe

carboxy-termi-nal one-fourth of

cr2

was also demonstrated in a surface probability plot derived forT3D cr2

by

using

the

technique

of Emini et al. (22)

(Fig. 3).

This

technique

is useful for

determining

more

specifically

whethera

given

amino acid in

a protein sequence is

likely

to be

exposed

on the

protein

surface.Amino

acids

340to410incr2havemore

consistently

high

probabilities

of

occurring

on the

protein

surface than

residues intheremainder of the

protein.

Theamino-terminal three-fourths of

cr2

is characterized

by

small

regions

of moderate to

high

surface

probability

that are

separated by

regions

oflow surface

probability.

We next made

predictions

of a2

secondary

structure

by

using

the

technique

of Nibert et al.

(49).

Plots of

a-helix,

,B-strand,

and

1-turn

scoresforT3Da2are shown in

Fig.

4.

In these

analyses,

a

carboxy-terminal region

of cr2 also

appears to be distinct from the remainder ofthe protein.

Amino acids 330to 380 in cr2 are characterized

by

consis-tently

high

a-helix scores. This

region

is bounded oneither side

by

a

region

with

high

1-turn

scores

(amino

acids 300to

330and 380to

410).

In

addition, ,B-strand

scores

throughout

this

region (amino

acids 300 to

418)

are low except for a

region

between aminoacids 330and 340. A small

region

at

the extreme

carboxy

terminus of

cr2

also has

high

a-helix

scores. In contrast to this

carboxy-terminal region,

the

amino-terminal three-fourths of

c2

are characterized

by

alternations between

regions

of

high

1-strand

and

high

,B-turn scores. Several small

regions

of

high

a-helix scores

are also found in this

region, including

one at the extreme

aminoterminusofthe

protein.

Inanalternative

approach

to

investigating

cr2

secondary

structure,weusedthe

technique

described

by Ralph

et al.

(55).

The results for T3Dcr2 are

shown in

Fig.

2 and are in

general

agreement with those

obtained fromthe

plots

ofcr2

secondary

structure

(Fig.

4).

The

cr2 protein

appearstoincludea

carboxy-terminal

region

that is formed from a-helices and

1-turns

and a

larger

amino-terminal

region

that is formed

predominantly

from

1-strands

and ,B-turns.

Sequences

in thereovirus r2 proteinand the

0i

subunitof

bacterial RNA

polymerase

aresimilar. To learn more about

c2

structure and

function,

we used several

techniques

to

searchfor

proteins

with sequence

similarity

tothededuced T1L, T2J, and T3D

cr2

amino acid sequences.

Using

the

BLASTP

algorithm (1),

we identifieda

region

in each ofthe

cr2 sequences that has

similarity

with a

region

in the v

subunit of the E. coli

DNA-dependent

RNA

polymerase

(50).

The cr2

region

with sequence

similarity

to the RNA

polymerase

13 subunit

corresponds

toamino acids354to374

and is shownfor

TiL

a2 in

Fig.

5A. The

regions

of similar

sequence in cr2 and the 13 subunitincludea

large

numberof

theacid and amide amino acidresidues andare

predicted

to

assume an a-helical conformation

according

to

secondary-structure

algorithms (Fig.

2 and

4) (29).

When we

arranged

these sequences on a helical wheel

projection

(61),

it was

evident that these a-helices in the two

proteins

would be

amphipathic,

with

apolar

residues

occupying

anarrow

strip

on November 10, 2019 by guest

http://jvi.asm.org/

(5)

10 20 30 40 50 60 70

*

+

00

0 O

+

+0

MARAAFLFKTVGFGGLQNVPINDELSSHLLRAGNSPWQLTQFLDWISLGRGLATSALVPTAGSRYYQMSC

---A---aaaaawacxc

a(

TTTT

2a aaaaa aaaaaa

TTTT

TTTTTTT

aaaaaa

TTTT TTTTTTT

80 90 100 110 120 130 140

*

O

* *

O

0*

0.00 #0 * #

LLSGTLQIPFRPNHRWGDIRFLRLVWSAPTLDGLVVAPPQVLAQPALQAQADRVYDCDDYPFLARDPRFK

---

-V---I---

Pi---I----A---

.---__---Po

ppppp

TTTT TTTTTTT

awxaaaa

pppppppp

aaaaa

aaaaaaacaaaaaa

aaaaa

TTTTTTTT TTTT

150 160 170 180 190 200 210

*

+~~~~~

000

0

0 *

*

O O

HRVYQQLSAVTLLNLTGFGPISYVRVDEDMWSGDVNQLLMNYFGHTFAEIAYTLCQASANRPWEHDGTYA

---I---

S---_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _v_ _ _ _ _

wacraaaa

Poppop

PoPPoppoP

_PPPPP

aaaaa aaaaaa

aaaaaaaaa

-

~~~PPPPPP

1TTT71'T TTTTTT

acr

TTTTT

220 230 240 250 260 270 280

*

+

*

0 O

O

0

*

* ,

RMTQIILSLFWLSYVGVIHQQNTYRTFYFQCNRRGDAAEVWILSCSLNHSAQIRPGNRSLFVMPTSPDWN

---V---H---T---A---

V.---__________________________

aaa

aaaaa

aaaaaaaaac

13PPPpppppppppppp13A3

ppppp

TTTTTT TTTTTT TTTT

aTTTT

TTTTTTT TTTTTTTT

290 300 310 320 330 340 350

O

O0

4 *

0

c**c

TiL

MDVNLILSSTLTGCLCSGSQLPLIDNNSVPAVSRNIHGWTGRAGNQLHGFQVRRMVTEFCDRLRRDGVMT

T2J

---N---G---I--Y---T3D

---TTTT TTTTT TTTT TTTT TTTT

aaaaaaa aa

TTTTTTTT

360 370 380 390 400 410 418

O O

+***

0

000

* *

QAQQNQIEALADQTQQFKRDKLEAWAREDDQYNQANPNSTMFRTKPFTNAQWGRGNTGATSAAIAALI

P---M---G---L---R-H---A---

---V---T---H---TTTT TTTTTT TTTT

caaaacraaa

TTTTTTT

FIG. 2. Alignmentof the deduced cr2amino acidsequencesofT1L,T2J,andT3D.Thesingle-letteraminoacidcodeisused.Aminoacid

positionsarenumbered abovethesequences, andresidues in T2JandT3Didentical to those inTlLareindicatedbydashes. Symbols: 0,

positionsofconservedcysteineresidues;O,positionsof conservedacidicresidues(D,E); *,positionsof conserved basicresidues(K,R).

Secondary-structurepredictions (14)forT3Du2 weredeterminedbythetechniqueofRalphet al.(55)andareindicatedfortheappropriate

amino acid positionin thea2sequence (a,a-helix; P, p-strand;T,,-tum).

TlL T2J

T3D

TlL

T2J T3D

TlL T2J

T3D

TlL

T2J T3D

TTTPP00

TTT

TlL

T2J T3D

aac

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.612.68.540.72.678.2]

(6)

x)

C2--o 1

-- :

0-I -3

--o

2-:t,

0 -0 0

Dl

(L u

0

n3

0 50 100 150 200 250

AminoAcidPosition

[image:6.612.79.292.80.220.2]

300 350 400

FIG. 3. Hydrophobicity and surface probability plots for T3D cr2. Hydrophobicity values (36) for T3D cr2 were plotted by the technique of Nibert et al. (49). The mean hydrophobicity value

(-0.4) is marked withahorizontal line. Surface probabilityvalues (22) for T3D cr2wereplotted by using thePEPTIDESTRUCTURE

program from the UWGCG. The mean surface probability value (1.0) is marked withahorizontal line.

A

TlL O2

aa 354-374

RNA pOl a

aa 1008-1031

Q N Q I E A L A D QTQ Q F K R D - - -K L EA ***:* Q *L*:AE*Q * Q N Q L E Q L AE Q Y D EL K H E F E K K LE A

B

TlL c2

0a 354-370

KA

Q

DE

L 1 1 E

~~H

y i

18

L 31 17

Q Q

EQ

RNA pol

a0 1008-1024

on one side of each a-helix and polar residues occupying

mostofthe positions onthe other side (Fig. 5B). The large

polar side of each amphipathic a-helix would be formed

almostentirely from the numerous acidand amide residues

found within each sequence. These findings suggest that

these a-helices in cr2 and the a subunit of the E. coli

DNA-dependent RNA polymerase may have related

func-tions.

Characterization of the dsRNA-binding activity of a2.

Although the functions of c2 are largely unknown, a ts

mutant whose mutation mapsto the S2gene (tsC447)

syn-thesizes5% ofthe normalamountof ssRNA and assembles

particles with 0.1% of the normal dsRNA content when

grown at the restrictive temperature (57). In addition, cr2 bindsreovirusdsRNA inablottingassay(60). These

obser-vationssuggestthat native c2mayalso be capable of binding

dsRNA and may play a role in viral RNA synthesis or in

packaging the viralgenome segments during virion

assem-bly.

In order to gain further insight into the dsRNA-binding activity of cr2,we used the Northwestern blotting technique

1.5

I1.3-I0

ZS

0.7-I0

-0

Cl)

1.5

-1.3 0-;

1.1 c

,0.9

-0.7

.-0.5

0 50 100 150 200 250 300 350 400

Amino Acid Position

FIG. 4. Secondary-structure plotsfor T3D cr2.

Secondary-struc-ture propensity scores (a-helix, P-strand, or ,-turn) (14) were plotted by the technique of Nibertetal. (49). The neutral preference

[image:6.612.324.557.85.292.2]

valueforsecondarystructures(1.0)ismarked withahorizontal line.

FIG. 5. (A) Region of similarity between the reovirus TlL cr2 protein and theE.coliDNA-dependentRNApolymerase ,Bsubunit (RNA pol Ia). Symbols: *, identical amino acid (aa) residues; :, closely relatedresidues (17). (B)Helical wheelprojection (61)ofa

portion ofthe cr2-RNApolymerase ,B subunitsequence similarity. Acid (D, E) and amide(N, Q) residues areshown inlargerletters thanthe other residues.

(10, 60)toanalyzecr2 fromvirionsoftheTiL, T2J,and T3D strains. In this assay, viral proteins were separated by

SDS-PAGE, electrophoretically transferred to

nitrocellu-lose, andprobedwith [5'-32P]pCp-labeled reovirusgenomic

RNA.Theresults oftheNorthwestern blot shown inFig.6A

indicatethat dsRNAbindingisapropertyof the a2proteins

of all three prototype strains. In orderto define the condi-tions foroptimal detection of the binding of cr2 todsRNA,

we determined the effects of varying the pH and NaCl

concentrationonthedsRNA-binding activityof T3D cr2. The cr2dsRNA-binding activitycould be detectedatpH 5.5 and

6.0, butwas never detected at a pH greater than 6.5 (Fig.

6B). Furthermore, the dsRNA-binding activity of cr2 was

detectedonlyinbindingbufferswithNaClconcentrationsof

c50mM(datanotshown). Thesecharacteristics areclearly

distinct from those of the other reovirus dsRNA-binding proteinscr3 and X1(Fig. 6B anddata notshown).

We recognized that the pH and NaCl dependence of

dsRNA bindingby cr2might enhance ourabilityto identify

differences in this activity between reovirus strains; such

differences mightallowusto identifyaspecificregion ofcr2

involved in dsRNA binding. The deduced cr2 amino acid

sequenceoftsC447, carryingatsmutationmappedtotheS2 gene,wasfoundtocontainonly three amino aciddifferences (valine 188, valine 323, and aspartate 383) from that of the

T3Dstrain from whichit wasderived (73). We investigated

whether the cr2 dsRNA-binding activity of tsC447 differed

from thatofT3D c2. Inexperiments in which either the pH (Fig. 6B) orthe NaClconcentration(data notshown) of the

binding buffer was varied, we found that there was no

difference inthe dsRNA-bindingactivity ofcr2between the

mutant and wild-type strains. In subsequent experiments

withTlLandT3D,strainswithdifferencesatfiveamino acid positionsina2,we werealso unabletodetectanydifference

inthe cr2dsRNA binding by varying either the pHorNaCl

5-

4-

3-

2-1 .--v ~

0

A I A

on November 10, 2019 by guest

http://jvi.asm.org/

[image:6.612.71.306.549.693.2]

(7)

zz Q_)

Q%

Q3

B

I.C\JN Ic.

pH

5.5

6.0

6.5

7.0

7.5

8.0

8.5

.:. C4t C4t

{ -

1*

~

*.~~~~~~~~~~~~~~~~~V

II.

.._ .,

lo*

i-X{--{

hI

>Llc

I.

o-3

FIG. 6. (A)Binding of reovirusproteins to reovirus genomic RNA. Viral proteins from 5x 1011purified reovirusT1L, T2J, andT3Dwere

separatedon a5to10% polyacrylamide-SDSgradient gel and electrophoretically transferred to nitrocellulose filters. The filtershown on the

leftwasstained with amidoblack immediately after electrophoretic transfer. The filter shown on the right was probed with 5 x 104 cpm of

[5'-32P]pCp-labeled reovirus genomic RNA per ml in binding buffer that had been adjusted to pH 6.0 and 25 mM NaCl after initial

protein-binding siteswereblocked. The filter probed with dsRNA was washed briefly in binding buffer and exposed to XAR film (Kodak) for 24 h at-70°C withanintensifying screen. Reovirus proteins are indicated between the filters. (B) Analysis of the dsRNA binding of T3D and

tsC447. Viral proteins from 5 x 1011purified reovirusT3D and tsC447 virions were separated on a 5 to 20% polyacrylamide-SDS gradient

gel and electrophoretically transferredto anitrocellulose filter. Protein-binding sites were blocked in standard bindingbuffer, and thefilter was cut sothat pairs of lanes could be probed with5 x 104 cpmof[5'-32P]pCp-labeledreovirus genomic RNA per ml in binding buffer that

had beenadjusted to the indicated pH. The filters were washed briefly in binding buffer and exposed to XAR film (Kodak) for 24 h at -70°C

withanintensifyingscreen. Reovirus proteins are indicated on the left.

concentration (data not shown). Thus, using this approach, we could not implicate any of the residues in cr2 that differ between these reovirus strains in the dsRNA-binding activ-ity of the cr2 protein.

We also performed Northwestern blotting studies of a2

peptide fragmentsgeneratedbychemicalcleavagewith75%

formic acid. Treatment with formic acid has been used to cleave proteins specifically between aspartate and proline residues (39). In the case of cr2, formic acid treatment

resulted in the generation of two fragments with Mrs of

17,000 and 29,000 (data not shown), consistent with the

presence ofa single aspartate-proline pair (amino acids 136

and 137) in the deduced amino acid sequence of a2. In

experiments thatusedamounts ofcr2 thatpermitthe

detec-tion of the dsRNA-binding activity of intact cr2, we were

unabletodetect dsRNAbinding byeitherfragment (datanot

shown).

DISCUSSION

Reovirus gene segments have short 5' and 3' noncoding

sequences. In this study, wedetermined the S2gene

nucle-otide sequences ofprototype strains of the three reovirus

serotypes.TheS2 sequences ofreovirus strains TlL and T2J inthis report and of strain T3D in this andapreviousreport

(73) have short 5' and 3' noncoding sequences (18 and 56

nucleotides, respectively). These S2 gene sequences differ frompreviouslypublished S2 sequences (12, 27)that

identi-fiedlonger3'noncoding

regions.

In theS2 sequences

deter-mined for each of the prototype strains in this study, the long

ORF in S2proposedto encode the cr2protein is 418 amino

acids in length. The molecular mass of cr2 calculated from

these deducedamino acid sequences ofTlL, T2J, and T3D

(-47 kDa) is consistent with the Mrs ofcr2 determined in

previous studies (9, 56, 59).

Havingshort 5' and 3' noncoding sequences is aquality

that appears to be common to all gene segments of the

mammalian reoviruses, including S2. Reovirus gene

seg-mentsforwhich nucleotide sequences have beenreportedto

date have 5'noncoding regionsof 32nucleotides(S4)orless and 3' noncoding regions of 181 nucleotides (L3) or less

(Table 1). It has been observed that the addition ofa single

basenearthe3'-terminal end of thepublishedL3nucleotide

sequence(6) would result ina Xl ORF with a 3'

noncoding

region ofonly 35 nucleotides (48); therefore, it is possible

that the publishedL3 sequence containsan errornearits 3' end and that thelongest 3'noncoding

region

in the reovirus

genome is 80 nucleotides

(Ml).

The economy exhibited in

the size of the noncoding sequences of the mammalian

reovirusesalso appears tobecommonamongother viruses

containing segmented genomes,

including

simian rotavirus (43), bluetongue virus (24), and influenza virus

(38).

Thus,

thereplication strategiesof members of the Reoviridae and the

Orthomyxoviridae

must

permit

noncoding

sequences to

constitute aminimal part of their genomes.

Comparison of the S2 nucleotide and i2 amino acid

se-quences of prototype strains ofthe three reovirusserotypes.

J-(T

I

aiwo."M10-

..0-2..

on November 10, 2019 by guest

http://jvi.asm.org/

[image:7.612.71.542.76.308.2]

(8)

TheS2genenucleotide sequences of T1L, T2J, and T3Dare

very similar, and the sequences of TlL and T3D are more

similarto each otherthan either sequence is tothat ofT2J.

Therelationship observedbetween the S2 sequencesofTIL,

T2J, andT3D isidenticaltothatofthe sequencesofallgene

segments that have been reported for these strains with the

exception of the Si gene(72). This observationhas ledtothe

suggestion thatgene segmentsoftype 1 andtype 3 reovirus strains, with the exception ofthe Si gene, diverged from a

commonancestormore recently thanthose oftype2 strains

(72). While we agree that the TlL and T3D S2 genes have

diverged from a common ancestor more recently than T2J

S2,it islikely thattheevolution of reovirusgenes otherthan Si is independent of serotype (18). Therefore, it will be necessarytocomparetheS2 nucleotide sequencesofseveral additional strains representing all three serotypes to more

accurately understand the evolution of the S2 gene.

The deduced c2 amino acid sequences of the T1L, T2J,

and T3D strains are even more conserved than their S2 sequences. The extent of cr2sequence conservation among these strains is similar to that ofthe deduced amino acid

sequences of other reovirus proteins that have been

re-ported, with theexception of those encodedby the Si gene

(72). This extent of amino acid sequence conservation may

be due to evolutionary constraints placed on either the

structure orfunction of reovirusproteins other than theSi

gene products, a1 and crls. Alternatively, there may be

unique selection pressures (e.g., immune) operating on the

Si geneproducts, resultingin theobserved sequence

heter-ogeneity of thecrl anduls proteins.

Two-domain modelforthestructure ofcr2. Ananalysis of computer-based predictions of cr2 structure (Fig. 2 through 4) suggeststhat the cr2proteincontains twodistinct regions

which form independent domains. In particular, the

car-boxy-terminalone-fourth of the cr2sequenceischaracterized by large hydrophilic regions, whichcorrespondtosequences

predicted to form a-helix and P-turn structures.

Further-more, a large number of residues in this region ofcr2 are

likelytobeexposedontheprotein surface. Incontrast, the

amino-terminal three-fourths ofthe cr2 sequence is

charac-terized by large hydrophobic regions, and predictions of protein secondary structure suggestthat this region ofa2 is

formed predominantly from 1-strands which alternate with

1-turns. Theappearance ofregions of lowsurface

probabil-ity separated by regions of high surface probability in this region of cr2 also suggests that it is formed predominantly

fromalternating 13-strands and ,3-turns. Theamino-terminal

and carboxy-terminal domains appear to be separated by

sequenceswithstrong 13-turn(Fig.2and4)andQt-loop(data

not shown) predictions, and this region may form a hinge

betweenthetwo domains.

Themultidomainstructural model thatwe proposeforcr2,

a major core protein of reovirus, is reminiscent of the

structures of the capsid proteins of several other small

RNA-containing viruses that have beendeterminedby X-ray

crystallography (31). Recently, thesamemultidomain

struc-tural motifhas been observed in a small isometric

DNA-containing virus (67). The multiple domains of several of

thesesmallisometricviruscapsid proteinsperform different

functions; a small (usually amino-terminal) basic domain

interacts with the viral genome, and other domains are

involvedintheprotein-protein interactionsthatgeneratethe

virus capsid shell. The proposed structural similarity be-tween cr2 and the capsid proteins ofthese other viruses suggests that, like these other virus capsid proteins, the differentdomains in cr2 might also have different functions.

The large amino-terminal domain in (r2 might form part of

the icosahedral shell of the virioncore. The small carboxy-terminal domain could mediate the interaction ofcr2 with the

genomic dsRNA, and/or it could mediate protein-protein

interactions (either with other cr2 subunits or with another reovirus core protein, e.g.,X1).

Searches of the NBRF data base revealed a region in the carboxy-terminal domain of the (r2 protein which has strik-ing similarity to a region in the If subunit of the E. coli

DNA-dependent RNA polymerase (Fig. 5A). The

corre-sponding sequences in the RNA polymerase

P

subunit are immediately amino-terminal to a region involved in binding theinitiating substrate (30) and in the transition from initia-tion to elongation (34). Furthermore, Glass etal. (29) have

found that a deletion that spans the sequences in the

P

subunit which are similar to sequences in (X2 results in

altered promoter selectivity of themutantpolymerase. Thus, these data suggest that the regions of similar sequence in the Xsubunit and in

cr2

areeither

directly

involved in the

binding

of nucleic acid or interact with another protein region to regulate nucleic acidbinding.

The regions of sequence similarity in a2 and the RNA

polymerase f subunit are predicted to form amphipathic

ac-helices which include a number of acidic residues and would beexpected to have a net negative charge (Fig.SB). Thea2 sequencewithsimilaritytothe

P

subunit is within the

region predicted to form the largest a-helix in cr2 (amino

acids 348to381). This large a-helicalregionhasanetcharge of-3 in TlL andT3Da2 and -4 in T2J c2; therefore, it is unlikely that it isdirectlyinvolved in the binding of nucleic acid (5, 51). Negativelycharged amphipathicc-heliceshave beenimplicated as playingkey roles in proteins capable of

transcriptional activation through the interaction of the

a.-helical regions with other proteins (53). Therefore, the acidic amphipathic a.-helix in the carboxy-terminal domain

ofc2maybe involved inprotein-protein interactions involv-ing another reovirus protein(s).

The r2 protein bindsdsRNA. In this study, weextended

the findings of Schiff et al. (60) and showed that the cr2 proteins ofprototype strains of the three reovirus serotypes

canbindreovirusgenomicRNAinaNorthwestern blotting

assay. Thea2proteins of strainsT1L, T2J, and T3D appear

tobind dsRNA with similaraffinity, and theseproteins bind

genomic RNA weakly in comparison to the other reovirus

dsRNA-binding proteins,a3 andAl. Whiletheprobe used to

study the dsRNA-binding activity of cr2 (reovirus genomic

RNA) would have allowed us to detect sequence-specific

nucleic acid-binding activities, our experiments do not di-rectly address the specificity of

c2-dsRNA

binding. In fact,

our experiments suggest that cr2 binding to dsRNA is not likely to be sequence specific. The conditions which allow detection ofcr2 binding to dsRNA in this assay (pH 5.5 to 6.0; 0to50mMNaCl)areverysimilar to those described for

the RNAbinding of rotavirusVP2(10)and the DNAbinding

of bovinepapillomavirusE2protein(41),both of which have

been characterized as sequence-independent interactions.

Wefound that the

c2-dsRNA

binding of tsC447 was

indis-tinguishablefrom that of T3D. Although the phenotype of tsC447 suggests that the mutant cr2 protein might have an

altered capacity to bind genomic dsRNA, we could not

demonstrateadifference between the activities of the

wild-type and mutant cr2 proteins by using the Northwestern

blotting assay. This suggests that the lesion(s) in tsC447

responsible for its phenotype does not directly affect the

dsRNA-binding activity of the mutant protein and may

on November 10, 2019 by guest

http://jvi.asm.org/

(9)

reflect altered interactions between mutant o2 and other proteins within the reovirus core.

We were unable to localize the dsRNA-binding activity of

(r2

to a particular region of the protein in experiments done

with

c2

peptide fragments generated by chemical cleavage

with formic acid. This may have been due to the relatively

weak dsRNA-binding activity ofu2observed in

Northwest-ern blots, the modification of specific residues in u2 by treatment with formic acid, or the possibility that the binding

of

a2

to dsRNA may require sequences in each of the a2

fragments generated by cleavage with formic acid. If the dsRNA-binding activity ofa2 requires sequences in each of

these fragments, this would suggest thatu2-dsRNA binding

resides in the large amino-terminal domain ofur2 proposed to form part of the shell of the reovirus core.

Understanding the reovirus core. This study represents an initial step in the characterization of the u2 protein and is partof a continuing analysis of the reovirus core. The intact core plays an important role in packaging the reovirus genome and also mediates several distinct enzymatic activ-ities during reovirus replication. The deduced amino acid

sequences of u2 reported here will be essential for the

interpretation of X-ray crystallographic data obtained from

studies of the reovirus core currently in progress in our

laboratory (16). However, given the complexity of the core particle, it will be necessary to pursue othertypes ofstudies, such as cryoelectron microscopy with image reconstruction,

in order to gain insight into core structure. In addition,

biochemical studies of core proteins and analyses oftheir amino acid sequences will allow us to learn more about the structure and function of individual core proteins. Our study will serve as aframework for future investigations

concern-ing the role thata2plays in formation of theinner shellofthe

core, the binding ofa2 to reovirus dsRNA, and possible

regulatory functions ofu2 with respect to the functions of

the intact core in reovirus transcription and replication.

ACKNOWLEDGMENTS

We express our appreciation to Elaine Freimontand Richard van den Broek for expert technical assistance and toMarcia Kazmier-czak for help with preparing the manuscript. We are grateful to Susan Russo and Randy Smith of the MBCRR forassistance with algorithms to predict protein secondary structure and searches of protein sequencedatabases andtoRoger Chalkley fordiscussions

of nucleicacid-bindingproteins.We alsothankLyndaMorrison and Kenneth Tyler for reviews of the manuscript.

This work was supported byresearch grant5R37 1AI13178from the National InstituteofAllergyandInfectiousDiseases. Support to B.N.F. was also provided by the Shipley Institute of Medicine. K.M.C. was supported by training grant 5T32-AI07061 from the National Institute of Allergy andInfectiousDiseases; M.L.N. was supported by PublicHealthServiceAward5T32GM07753 from the National Institute ofGeneral Medical Sciences; L.A.S. was sup-ported by Public Health Service Award 1F32 GM12478 from the National Institute ofGeneralMedicine; and T.S.D. was the recipi-ent ofPhysician-Scientist Award 5K11 A100865 from the National

Institute ofAllergy and Infectious Diseases. REFERENCES

1. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403-410.

2. Anders, D. G.,and R. A. Consigli. 1983. Chemical cleavage of polyomavirus major structural protein VP1: identification of cleavageproducts andevidence that the receptor moiety resides inthe carboxy-terminal region. J. Virol. 48:197-205.

3. AppliedBiosystems. 1984. User bulletin. 13:1-28.

4. Atwater, J. A., S. M. Munemitsu, and C. E. Samuel. 1986.

Biosynthesis of reovirus-specified polypeptides: molecular cDNAcloning and nucleotide sequence of the reovirusserotype

1Langstrain S4mRNAwhich encodesthemajorcapsidsurface

polypeptide cr3. Biochem. Biophys. Res. Commun. 136:183-192.

5. Bandziulis, R. J., M. S. Swanson, and G. Dreyfuss. 1989.

RNA-binding proteinsasdevelopmentalregulators.Genes Dev. 3:431-437.

6. Bartlett, J. A., and W. K. Joklik. 1988. The sequence ofthe reovirus serotype 3 L3 genome segment which encodes the major coreprotein Al. Virology 167:31-37.

7. Bassel-Duby, R., A. Jayasuriya, D. Chatterjee, N. Sonenberg,

J. V.Maizel, Jr., and B. N. Fields. 1985. Sequenceofreovirus

hemagglutininpredictsacoiled-coilstructure.Nature(London)

315:421-423.

8. Bassel-Duby, R., D. R. Spriggs,K. L. Tyler, andB. N. Fields. 1986. Identification of attenuating mutations on the reovirus

type 3S1 double-strandedRNAsegment witharapid sequenc-ingtechnique.J.Virol. 60:64-67.

9. Both, G. W., S. Lavi, and A.J. Shatkin. 1975.Synthesisof all the geneproducts ofthereovirusgenome in vivoandin vitro. Cell 4:173-180.

10. Boyle, J. F., and K. V. Holmes. 1986.RNA-bindingproteinsof bovine rotavirus. J. Virol.58:561-568.

11. Cashdollar, L. W., R. A. Chmelo, J. R. Wiener, and W. K. Joklik. 1985. The sequence of the S1 genes of the three serotypesof reovirus. Proc.Natl. Acad. Sci. USA 82:24-28.

12. Cashdollar, L. W., J. Esperza, G. R. Hudson, R. Chmelo,

P. W. K.Lee, and W. K. Joklik. 1982. Cloning ofthe double-stranded RNA genes of reovirus: sequence of the cloned S2 gene. Proc. Natl. Acad. Sci. USA79:7644-7648.

13. Chang, C.-T., and H. J. Zweerink. 1971. Fate of parental

reovirusininfectedcell. Virology46:544-555.

14. Chou, P. Y., and G. D. Fasman. 1978. Empiricalpredictions of

proteinconformation. Annu. Rev. Biochem.47:251-276. 15. Cleveland, D. R., H. Zarbl, and S. Millward. 1986. Reovirus

guanylyltransferase is L2 gene product lambda 2. J. Virol.

60:307-311.

16. Coombs, K. M., B. N. Fields, and S. C. Harrison. 1990.

Crystallization of the reovirus type 3 Dearing core. Crystal packing is determinedbythe X2protein.J.Mol. Biol. 215:1-5.

17. Dayhoff,M.O., W. C.Barker,and L. T.Hunt. 1983. Establish-ing homologies in protein sequences. Methods Enzymol. 91: 524-545.

18. Dermody,T. S.,M. L.Nibert,R. Bassel-Duby,and B. N. Fields. 1990. Sequence diversity in S1 genes andS1 translation

prod-uctsof eleventype 3 reovirus strains. J. Virol. 64:4842-4850.

19. Devereux, J., P.Haeberli,and0.Smithies. 1984. A

comprehen-sive set ofsequence analysis programs for the VAX. Nucleic Acids Res. 12:387-395.

20. Drayna, D., andB. N. Fields. 1982. Activation and

characteri-zation ofthe reovirus transcriptase: genetic analysis. J. Virol. 41:110-118.

21. Duncan, R., D. Horne, L. W. Cashdollar, W. K. Joklik, and P. W. K. Lee. 1990. Identification of conserved domains in the cell attachment proteins of the three serotypes of reovirus. Virology 174:339-409.

22. Emini, E. A., J. V. Hughs, D. S. Perlow, and J. Boger. 1985. Induction ofhepatitis Avirus-neutralizing antibodyby a

virus-specific synthetic peptide. J. Virol. 55:836-839.

23. Ewing, D. D., M. D. Sargent, and J. Borsa. 1985. Switch-on

transcriptase function in reovirus: analysis of

polypeptide

changes using 2-D gels.Virology 144:442-456.

24. Fukusho, Y., S. Yu, and P. Roy. 1989. Completion of the sequence ofbluetongue virus serotype 10by the characteriza-tion ofa structural protein, VP6, and a nonstructural

protein,

NS2. J. Gen. Virol. 70:1677-1689.

25. Furlong, D. B., M. L. Nibert, and B. N. Fields. 1988. Sigma 1

protein of mammalian reoviruses extends from the surfaces of viralparticles. J. Virol. 62:246-256.

26. George, C. X.,J. A.Atwater, andC.E. Samuel. 1986.

Biosyn-thesisofreovirus-specifiedpolypeptides: molecular cDNA

clon-ing and nucleotide sequence ofthe reovirus serotype 1 Lang

on November 10, 2019 by guest

http://jvi.asm.org/

(10)

strainS3mRNAwhich encodesthenonstructural RNA-binding proteinC NS. Biochem. Biophys. Res. Commun. 139:845-851.

27. George,C. X., A. Crowe,S. M. Munemitsu, J. A. Atwater, and

C.E.Samuel.1987. Biosynthesisofreovirus-specified

polypep-tides: molecular cDNAcloning andnucleotide sequenceofthe reovirus serotype 1 Lang strain S2 mRNA which encodes the virioncorepolypeptide a2. Biochem. Biophys. Res. Commun. 147:1153-1161.

28. Giantini,M., L.S.Seliger, Y. Furuichi, and A. J.Shatkin. 1984.

Reovirustype3genomesegmentS4: nucleotidesequenceofthe

geneencodingamajor virionsurface protein.J. Virol. 52:984-987.

29. Glass,R. E., S. T. Jones, V. Nene,T.Nomura,N.Fujita, and A.

Ishihama. 1986.Genetic studiesonthePsubunit of Escherichia

coliRNA polymerase. VIII. Localisationofaregion involved in

promoterselectivity. Mol. Gen. Genet. 203:487-491.

30. Grachev, M. A., E. A. Lukhtanov, A. A. Mustaev, E. F.

Zaychikov,M. N. Abdukayumov,I.V. Rabinov,V.I.Richter,

Y. S. Skoblov, and P. G. Chistyakov. 1989. Studies of the

functional topography of Escherichiacoli RNA polymerase: a methodfor localisation of the sites ofaffinity labelling. Eur. J. Biochem. 180:577-585.

31. Harrison, S.C. 1984. Multiple modesofsubunit associationin

thestructureof simple spherical viruses.Trends Biochem. Sci. 9:345-351.

32. Jayasuriya, A. K., M. L. Nibert, and B. N. Fields. 1988.

Complete nucleotide sequence of the M2 gene segment of

reovirustype3 Dearing andanalysisof its protein product,u1.

Virology 163:591-602.

33. Joklik, W. K. 1983. The reovirus particle, p. 9-78. In W. K.

Joklik (ed.), The reoviridae. Plenum Publishing Corp., New York.

34. Kashlev, M., J. Lee, K. Zalenskaya, V. Nikiforov, and A.

Goldfarb. 1990. Blocking of the initiation-to-elongation transi-tion by a transdominant RNA polymerase mutation. Science 248:1006-1009.

35. Kozak, M. 1981. Possibleroleof flanking nucleotidesin

recog-nition of the AUG initiator codon by eukaryotic ribosomes. NucleicAcidsRes.9:5233-5252.

36. Kyte, J., and R. F. Doolittle. 1982. A simple method for

displayingthehydropathic characterofaprotein.J.Mol. Biol. 157:105-132.

37. Laemmli,U. K.1970.Cleavageof structural proteinsduring the

assembly of the head of bacteriophage T4. Nature (London) 227:680-685.

38. Lamb, R.A.1983. Theinfluenza virus RNAsegmentsand their

encodedproteins, p.21-69. InP. Palese and D. W. Kingsbury

(ed.), Genetics of influenza viruses. Springer-Verlag, New

York.

39. Landon, M. 1977. Cleavage of aspartyl-prolyl bonds. Methods

Enzymol.47:145-149.

40. Leszczynski, J. F., and G. D. Rose. 1986. Loops in globular

proteins: a novel category of secondary structure. Science

234:849-855.

41. Mallon,R. G.,D. Wojciechowicz, and V.Defendi. 1987.

DNA-bindingactivity ofpapillomavirus proteins. J. Virol.

61:1655-1670.

42. McCrae, M. A., and W. K. Joklik. 1978. The nature ofthe

polypeptide encoded by eachof the tendouble-stranded RNA

segmentsof reovirus type 3. Virology89:578-593.

43. Mitchell, D. B., and G. W. Both. 1990. Completion of the

genomic sequence of the simian rotavirus SAil: nucleotide sequencesofsegments 1, 2, and 3. Virology 177:324-331.

44. Morozov, S. Y. 1989. A possible relationshipof reovirus

puta-tive RNA polymeraseto polymerases of positive-strand RNA

viruses. Nucleic AcidsRes. 17:5394.

45. Munemitsu, S. M., J. A. Atwater, and C. E. Samuel. 1986.

Biosynthesis of reovirus-specified polypeptides: molecular

cDNA cloningand nucleotidesequenceof the reovirusserotype 1 Lang strain bicistronic S1 mRNA whichencodes the minor

capsid polypeptide oala and the nonstructural polypeptide

albns. Biochem. Biophys. Res.Commun. 140:508-514.

46. Mustoe,T. A., R. F. Ramig, A. H. Sharpe, and B. N.Fields.

1978. Genetics of reovirus: identification of the dsRNA seg-ments encoding the polypeptides of the

,.

anda size classes. Virology 89:594-604.

47. Nagata, L., S. A. Masri, and D. C. Mah. 1984. Molecularcloning and sequencing of the reovirus (serotype 3) Si gene which

encodes theviral cellattachmentproteinsigma1.Nucleic Acids

Res. 12:8699-8710.

48. Nibert, M. L. Unpublished observation.

49. Nibert, M. L., T. S. Dermody, and B. N. Fields. 1990. Structure of thereovirus cell-attachment protein: amodelforthedomain

organization ofal. J. Virol. 64:2976-2989.

50. Ovchinikov, Y. A., G. S. Monastyrskaya, V. V. Gubanov, S.0. Guryev,0. Y. Chertov, N. N. Modyanov, U. A.Grinkevich,I.A. Makarova, T. V. Marchenko,I.N.Prolovnikova, V. M. Lipkin, and E. D. Sverdlov. 1981. Theprimary structureof Escherichia

coli RNA polymerase. Nucleotide sequence of the rpoB gene and amino acid sequence of thep subunit. Eur. J. Biochem.

116:621-629.

51. Pabo, C.O.,and R. T. Sauer. 1984. Protein-DNA recognition. Annu. Rev. Biochem. 53:293-321.

52. Pearson, W. R., and D. L. Lipman. 1988. Improved tools for biological sequence comparison. Proc. Natl. Acad. Sci. USA

85:2444-2448.

53. Ptashne, M. 1988. How eukaryotic transcriptional activators work. Nature (London) 335:683-689.

54. Ralph, S. J., J. D. Harvey, and A. R. Bellamy. 1980. Subunit structure of the reovirus spike. J. Virol. 36:894-896.

55. Ralph, W. W., T. Webster, and T. F. Smith. 1987. A modified Chou and Fasman protein structure algorithm. Comput.

Appl.

Biosci. 3:211-216.

56. Ramig, R. F., R. K. Cross, and B. N. Fields. 1977. Genome RNAs and polypeptides of reovirus serotypes 1, 2, and 3. J. Virol. 22:726-733.

57. Ramig, R. F., T. A. Mustoe, A. H. Sharpe, and B. N. Fields. 1978. A genetic map of reovirus.II. Assignment of the double-stranded RNA-negative mutant groups C, D, and E genome segments. Virology 85:531-534.

58. Richardson, M. A., and Y. Furuichi. 1983. Nucleotide sequence of reovirus genome segment S3 encoding nonstructural protein sigma NS. Nucleic Acids Res. 11:6399-6408.

59. Samuel, C. E. 1983. Biosynthesis of reovirus-specified peptides, p. 219-230. In R. W. Compans and D. H. L. Bishop (ed.), Double-stranded RNA viruses. Elsevier Biomedical, New York.

60. Schiff,L. A., M. L. Nibert, M. S. Co, E. G. Brown, and B. N. Fields. 1988. Distinct binding sites for zinc and double-stranded RNA in the reovirus outer capsid proteincr3. Mol. Cell. Biol. 8:273-283.

61. Schiffer, M., and A. B. Edmundson. 1967. Use.of helical wheels to represent the structures of proteins and to identify segments with helical potential. Biophys. J. 7:121-135.

62. Seliger, L. S., K. Zheng, and A. J. Shatkin. 1987. Complete nucleotide sequence of reovirus L2 gene and deduced amino acid sequence of viral mRNA guanylyltransferase. J. Biol. Chem. 262:16289-16293.

63. Smith, R. E., H. J. Zweerink, and W. K.Joklik. 1969. Polypep-tide components of virions, top component and cores of reovi-rus type 3. Virology39:791-810.

64. Smith, R. F., and T. F. Smith. 1990. Automatic generation of primary sequence patterns from sets of related protein se-quences. Proc. Natl. Acad. Sci. USA 87:118-122.

65. Tarlow, O., J. G. McCorquodale, and M. A. McCrae. 1988. Molecular cloning and sequencing of the gene M2 encoding the major virion structural protein

Rul-,ulC

of serotypes 1 and 3 of mammalian reoviruses. Virology 164:141-146.

66. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA79:4350-4354.

67. Tsao, J., M. S. Chapman, M. Agbandje, W. Keller, K. Smith, H. Wu, M. Luo, T. J. Smith, M. G. Rossman, R. W. Compans, and C. R. Parrish. 1991. The three-dimensional structure of canine