Purification and characterization of the reovirus cell attachment protein σ1

Purification and characterization of the reovirus cell attachment protein σ1

VIROLOGY 156, 377-385 (1987) Purification and Characterization of the Reovirus Cell Attachment Protein al MICHAEL C. YEUNG, M. JOHN GILL, SULEIMAN ...

2MB Sizes 0 Downloads 52 Views

VIROLOGY

156, 377-385

(1987)

Purification and Characterization of the Reovirus Cell Attachment Protein al MICHAEL C. YEUNG, M. JOHN GILL, SULEIMAN S. ALIBHAI, MAHMOUD S. SHAHRABADI, AND PATRICK W. K. LEE’ Department

of Microbiology

and Infectious

Diseases, Received

University August

of Calgary

4, 1986;

accepted

Health October

Sciences

Centre,

Calgary,

Alberta,

Canada

TZN 4N 1

16, 1986

It has previously been shown that of all the soluble reovirus-specified proteins present in the infected cell lysate, protein 01 alone possesses the capacity to bind to host cells (P. W. K. Lee, E. C. Hayes, and W. K. Joklik, 1981, Virology 108, 156-163). We found that 01 from urea-disrupted reovirus particles was also capable of such specific binding. Reovirions were therefore used as a source of functional al. Accordingly, a simple procedure has been developed to purify ~1 by subjecting urea-disrupted reovirions to DEAE ion-exchange chromatography. Protein 01 thus isolated was electrophoretically homogeneous and the recovery was estimated to be 50 to 60% of the theoretical yield. The purified protein presumably maintained its native conformation since it was recognized by a panel of monoclonal anti~1 antibodies previously isolated, and was capable of specifically binding to host cell receptors, agglutinating human erythrocytes and inducing neutralization and hemagglutination-inhibition antibodies. Subsequent chemical crosslinking studies revealed the presence of oligomeric (mostly dimeric) al forms in the preparation. The amino acid composition of the purified ~1 was found to closely match that inferred from the Sl gene sequence. However, attempts to determine its amino-terminal sequence have not been successful. The p/of the purified protein was determined to be 6.8. Circular dichroic measurements of the purified 01 indicated that 54 and 19% of its residues were arranged in a-helical and o 1987 Academic PWSS, Inc. &sheet secondary structures, respectively.

in obtaining this protein in the purified form. This is mainly due to the scantiness of al in either infected cells or intact virions (Smith et al., 1969; Zweerink and Joklik, 1970; Levin and Samuel, 1980). Certain salient features of ~1 have nevertheless been predicted through molecular cloning and sequence analysis of the gene encoding this protein (gene Sl) (Nagata et al., 1984; Cashdollar et a/., 1985; Bassel-Duby et a/., 1985). More recently the cloned gene has been expressed in Escherichia co/i to yield a al fusion protein capable of binding to host cells and to erythrocytes (Masri et a/., 1986). Despite advances of this nature, it is clear that a detailed analysis of authentic (viral) ~1 in its mature form is crucial for the ultimate understanding of the interaction of this protein with cellular receptors, with the host immune system, and with other reovirus proteins. This report documents a fast, one-step purification scheme whereby electrophoretically homogeneous ~1 can be obtained in bulk from intact virions by DEAESephacel ion-exchange chromatography. Protein ~1 thus purified was found to be functionally active and was characterized by biochemical and biophysical means.

INTRODUCTION The reovirus cell attachment protein, al, plays important roles in reovirus infection in that it specifies how reovirus particles interact with host cells and with the host. This protein possesses the remarkable capability of attaching to host cells by itself (Lee et a/., 1981 b). It is also the reovirus hemagglutinin since it interacts with erythrocyte receptors as well (Weiner et al., 1978; Masri et al., 1986). Epitope mapping studies using a library of monoclonal anti-al antibodies have revealed that the host cell binding function and the hemagglutinating function are likely controlled by distinct regions of the al protein (Burstin et al., 1982; Spriggs et a/., 1983). Perhaps because of its cell receptor recognizing property, al also defines reovirus tissue tropism and virulence in the host. This protein has further been shown to be responsible for the triggering of host immune (both humoral and cellular) responses (for reviews, see Joklik, 1985; Sharpe and Fields, 1985). In view of the multifaceted roles al plays in reovirus infection, it is evident that an understanding of the functional uniqueness of al necessitates an in-depth characterization of this protein at the molecular level. Thus far, efforts aimed at defining structure-function relationships of 01 have been frustrated by the difficulty

’ To whom

requests

for reprints

should

MATERIALS

AND

METHODS

Cells and virus The virus used was reovirus serotype 3 (strain Dearing). It was propagated in suspension cultures of Earle’s

be addressed 377

0042-6822187

$3.00

Copyright 0 1987 by Academic Press, Inc. All rights of reproduction in any form reserved.

378

YEUNG ET AL.

L-929 strain of mouse fibroblasts and was purified according to Smith et a/. (1969). Reovirus labeled with [35S]methionine was grown and purified as described by McCrae and Joklik (1978). Purification

of ~1

Twenty milligrams of purified reovirus (7 mg/ml) in PBS were disrupted by adding solid ultrapure grade urea (United States Biochemicals, Cleveland) to a final concentration of 10 n/l, followed by incubation at 22” for 30 min. The clear, disrupted reovirus solution was then diluted fivefold with 10 mM sodium phosphate (pH 8.0). Protein precipitates formed as a result were pelleted at 13,500 g for 5 min at 4’. The supernatant was then applied at a speed of 40 ml/hr onto a l-ml (bed volume) DEAE-Sephacel (Pharmacia Chemicals) column previously equilibrated with 10 mM sodium phosphate (pH 8.0). The column was then washed with 20 bed vol of 10 mM sodium phosphate (pH 8.0). Protein al was eluted with a 14-ml linear NaCl gradient of O-O.25 M in 10 mM sodium phosphate (pH 8.0) at a speed of 40 ml/hr. Fourteen l-ml fractions were collected and aliquots were subjected to SDS-polyacrylamide gel electrophoresis (PAGE) followed by silver staining (or fluorography when [35S]methionine-labeled virus was used) to locate ol-containing fractions and to assess the purity of al isolated. Purified al was used directly off the column, or was concentrated and dialyzed against PBS before use. SDS-PAGE SDS-PAGE of proteins was performed as previously described (Lee et al., 1981 a).

Preparation

of rabbit antisera

Female New Zealand white rabbits (5 to 6 lb) were inoculated subcutaneously and intramuscularly with either purified reovirus (500 pg) or purified ~1 (5 pg) in Freund’s complete adjuvent for the initial dose. Two subsequent boostings (same amounts of antigens but prepared in Freund’s incomplete adjuvant) were made at 14 and 28 days. Sera were collected 7 days after the third inoculation. Protein al cell attachment assays and hemagglutination studies These were carried out essentially as described previously (Lee et al., 1981 b; Masri et al., 1986). Crosslinking

studies of purified al

Just before use, 10 mg of dimethyl suberimidate (DMS) (Pierce Chemical Co., IL) was dissolved in 272 ~1of dimethyl sulfoxide (DMSO) and was subsequently diluted with PBS to a final concentration of 0.184 mg/ ml. Equal volumes (50 ~1)of the freshly prepared DMS solution and the purified [35S]methionine-labeled ~1 were mixed and incubated at 4” for 30 min. The reaction mixture was then added to 100 ~1of the Laemmli solubilizing buffer, boiled for 5 min, and subjected to SDS-PAGE. Crosslinked al products were visualized by fluorography. Radioimmunoprecipitation

lmmunoprecipitation of radiolabeled al with various antibodies was carried out as described previously (Lee et a/., 1981 a). Viral neutralization

Silver staining Following electrophoresis, SDS-polyacrylamide gels were soaked in 50% methanol overnight before they were silver stained according to Wray et al. (1981). Coomassie of al yield

blue staining and quantitation

To quantitate the yield of purified ~1, gels were stained for 1 hr in methanol:water:glacial acetic acid (5:5:1) containing 0.1% Coomassie blue. They were then destained and the ~1 bands from gradient fractions were compared to those on marker lanes containing known amounts of reovirus using a laser scanning densitometer (LKB202) and a reporting integrator (Hewlett-Packard 3390A). In all cases, quantitation was carried out on ~1 bands that registered responses on the linear portion of the dose-response curve.

of ~1

tests

These procedures (Hayes et a/., 1981). Radioiodination

were as described

previously

of al

Purified ~1 (100 ~1, 25 pg/ml in PBS) was mixed with 1 mCi of carrier-free iodine-125 (Amersham IMS300) in a borosilicate tube that had previously been coated at the bottom with 1 pg of IODO-GEN (Pierce Chemical). After 30 set at 22” with occasional mixing, the sample was taken out of the tube and allowed to pass through a desalting PD-10 column (Pharmacia) previously equilibrated with PBS. Peak fractions of i251-ul were then pooled; specific activity of radiolabeled al was 1 to 2 X 10’ dpmlpg. lsoelectrofocusing lsoelectrofocusing studies of purified al was performed in a BioRad mini-slab gel unit (Model 360) under

PURIFICATION

OF

nondenaturing and nonreducing conditions. The isoelectrofocusing slab gel is made of 5.5% acrylamide, 0.5% N-AI’-methylene-bis-acrylamide, 10% glycerol, and 2.2% ampholyte (range 3.5-10, LKB). The gel was prerun at 100 V for 30 min at 4” with 0.02 M acetic acid at the anode and 0.02 M NaOH at the cathode. lodinated 01 samples were then mixed with equal volumes of sample buffer containing 15% glycerol and 2.4% ampholyte (range 3.5-l 0). lsoelectrofocusing was then carried out at 100 V for 1 hr, 300 V for 4 hr, and 500 V for 30 min. Protein bands were located by autoradiography; pl values were determined by slicing 0.5-cm gel portions on both sides of sample lanes. The gel slices were each soaked in 0.5 ml boiled distilled water overnight and pH was determined on the supernatant with a Fisher Accumet pH meter. Amino acid analysis Amino acid analysis was performed on purified al on a Beckman 6300 amino acid analyzer following HCI hydrolysis (6 N HCI, 1 OO”, 24 hr in vacua) of the protein (ca. 1 nmol). N-terminal

amino acid sequence

analysis

Purified ~1 was subjected to N-terminal protein sequencing on an Applied Biosystems 470A gas phase protein sequencer using methanolic HCI conversion and reagents from Applied Biosystems as described by McKay et al. (1985). Circular dichroic measurements Circular dichroic (CD) spectra of al were recorded at 25” under constant nitrogen flush with a Jasco J5OOC spectropolarimeter coupled to a Jasco DP-500N data processor, using 1 .O-mm cells. The instrument was calibrated using a standard solution of ~-1 O-camphorsulfonic acid. Peak fractions containing al from the DEAE-Sephacel column were used directly for CD measurements. Circular dichroic data were recorded in 1 -nm increments from 190 to 240 nm and reported as mean residue elliptic@ (in deg cm* dmol-‘). An amino acid mean residue weight of 107 for ~1 was calculated from our previously published Sl gene sequence (Nagata et al., 1984). Estimations of the a-helical and b-strand contents of al were derived from the CD spectra based on the algorithm reported by Siegel et al. (1980). l

l

RESULTS Intact virion as a source of functional

~1

Although al present in the reovirus-infected cell lysate has been demonstrated to be “biologically active”

REOVIRUS

PROTEIN

379

01

in that it is capable of binding to cellular receptors (Lee era/., 1981 b), intact reovirion was chosen as a source of 01 to avoid possible contamination problems due to the myriad of host proteins present in the infected cell lysate. An additional advantage of using the intact virion is that “virion al” would represent the most mature form of al available. It was important to first determine whether disruption of the reovirion by chaotropic agents significantly alters the “biological activity” of al. To this end, CsCl density gradient purified r5S]methioninelabeled reovirus (1.6 mg/ml) was disrupted by adding solid ultrapure urea to a final concentration of 10 M. The opaque reoviral suspension instantly clarified upon disintegration of virus particles by urea. To ensure complete virus disruption, the urea-treated virus preparation was further incubated at 22” for 30 min and was then dialyzed against PBS at 4” overnight to remove urea. The dialysate was centrifuged at 13,500 g for 5 min to pellet any protein precipitates. The supernatant containing soluble reoviral proteins (and RNAs) was then directly applied to L cell monolayers for al cell binding assay as described under Materials and Methods. Figure 1 (lane 3) shows that as is the case with the reovirus-infected cell lysate (Lee eta/., 1981 b), only ul, of all virion proteins, was capable of binding to L cells by itself. Further, this binding of al to L cells was specific since the presence of excess reovirus effectively blocked such attachment (Fig. 1, lane 2). These results indicate that al, after exposure to a high concentration of urea, still retains a functional cell attachment domain. Thus, intact reovirions can serve as an excellent source of biologically active al. One-step purification

of virion ~1

A simple procedure was then developed to isolate al in bulk for studies requiring highly purified 01. Routinely, solid urea was added to 20 mg reovirus to a final concentration of 10 M. Disruption of the virion was allowed to continue for 30 min at 22”, after which the clear lysate was diluted with 10 mM sodium phosphate (pH 8.0) centrifuged to pellet any protein precipitates, and the supernatant was loaded onto a 1-ml (bed volume) DEAE-Sephacel column as described under Materials and Methods. Protein al was eluted with a linear salt gradient of O-O.25 M NaCl in 10 mM sodium phosphate (pH 8.0). Figure 2A shows an elution profile of al when purified [35S]methionine-labeled reovirus was used as the starting material. A sharp radioactivity peak appeared at the second half of the elution gradient. Upon subsequent SDS-PAGE analysis, this peak was found to correspond to an electrophoretically homogeneous preparation of al (Fig. 28, lanes 7 to 11). To eliminate the possibility that reoviral RNAs might have

380

YEUNG

1

2

3

ET AL.

(i.e., O-O.25 l\/I NaCI) is critical in ensuring a pure preparation of al ; employment of a higher salt gradient (e.g., O-O.5 M NaCI) would cause coelution of al with other viral proteins such as a3 and plC, as well as reoviral RNAs (data not shown). Third, minor contamination problems sometimes arise if the reovirus preparations used are not fresh. These contaminating proteins probably represent degraded virion proteins since they do not electrophoretically correspond to any known reoviral proteins (data not shown). The yield of ~1 using this method was quantitated by subjecting gradient fractions to SDS-PAGE, followed by Coomassie blue staining and scanning laser densitometry (see Materials and Methods). The recovery

IJ ,, [

-01 a A

FIG. 1. Binding of ul (from urea-disrupted reovirions) to mouse L fibroblasts. [36S]Methionine-labeled virions (1.6 mg/ml) were treated with 10 M urea, followed by dialysis against PBS at 4” overnight. The dialysate was then centrifuged at 13,500 g for 5 min and the supernatant was then mixed with type 3 reovirus or BSA (final concentration, 3 mg/ml). The mixtures were then applied to monolayers of mouse L fibroblasts growing in 35 X 10 mm petri dishes After incubation for 1 hr at 4” with intermittent rocking, the monolayers were washed extensively with PBS and the cells were lysed with 1% Triton X-l 00 in PBS. The nuclei were removed by centrifugation and the lysates were analyzed by SDS-PAGE. Lane 1, [35S]methioninelabeled reovirus; lanes 2 and 3, reoviral proteins (from urea-disrupted virions) bound to L cells in the presence of reovirus and BSA, re-

-16-

I

-0.25

N-129

I A z - 0.125 -. v

3. 8v

z”

4 1

J-0

1

-

FRACTION

coeluted with al but failed to show up in the fluorogram, (as the RNAs were not radioactively labeled), the ~1 peak was subjected to SDS-PAGE followed by silver staining according to the procedure of Wray et al. (1981). In this technique, both proteins and nucleic acids are stained. Since no bands other than those of al could be visualized on the silver stained gel (data not shown), it was concluded that al isolated from our purification scheme was devoid of any contamination due to other reoviral proteins or RNA species. Several observations pertaining to the above al isolation procedure require comment. First, the dilution step, following urea disruption of virions and prior to DEAE chromatography, was found to be important since it results in the precipitation of considerable amounts of non-al proteins, allowing for their subsequent removal by centrifugation (Fig. 2B, lane 2). This step also serves to minimize the precipitation of such proteins in the DEAE column which would otherwise decrease the efficiency of the column. Second, the choice of the salt gradient as described in our protocol

~

NUMBER

FIG. 2. Purification of virion al. Twenty milligrams of [%]methioninelabeled reovirus (7 mg/ml) were disrupted with urea added to a final concentration of 10 M. After incubation at 22” for 30 min, the lysate was diluted fivefold with 10 mM sodium phosphate (pH 8.0) and centrifuged at 13.500 g for 5 min. The supernatant was then applied to a 1 -ml DEAE-Sephacel column equilibrated with 10 mM sodium phosphate (pH 8.0). The column was developed with a 14-ml linear NaCl gradient (O-O.25 M). (A) Radioactivity profile of column fractions. (B) SDS-PAGE analysis of peak fractions from (A). Lane 1, [%]methionine-labeled reovirus; lane 2, reoviral protein precipitates obtained after dilution of urea in lysate; lane 3, reoviral proteins applied to column; lane 4, column flowthrough; lanes 5-l 2. peak fractions (7-l 4) from column; lane 13, postgradient fraction (eluted with 0.5 M NaCI).

PURIFICATION

OF

of 01 was routinely found to be in the range of 50 to 60% of the theoretical yield (data not shown). Cell binding and hemagglutinating activities of purifed ~1 To determine whether the functional domains on al remain active after the above isolation procedure, the purified protein was subjected to competitive cell attachment assays carried out in the presence of reovirus or bovine serum albumin (BSA) (control). The results, illustrated in Fig. 3, show clearly that the purified ~1 was capable of recognizing and binding to L cell surface receptors (Fig. 3, lane 3) and that this binding was specific since it was interfered with in the presence of reovirus (Fig. 3, lane 2). Another functional domain, believed to be distinct from that responsible for host cell attachment, is the hemagglutination domain (Burstin eta/., 1982; Spriggs et al., 1983). Direct evidence that 01 is the reovirus hemagglutinin came from the recent demonstration that f. co/i-expressed al aggregates agglutinate human red blood cells (Masri et a/., 1986). It would therefore be of interest to see whether the hemagglutination domain of the purified protein remains functionally intact. Accordingly, purified 01 was assayed for its ability to agglutinate human type 0 erythrocytes. The results, illustrated in Fig. 4, show that the purified protein had hemagglutinating activity. Peak al fractions from the column routinely exhibited an HA titer of 1 :16 to 1:32. The above experiments thus demonstrate that the two assayable biological activities (i.e., host cell binding and hemagglutination) of al are preserved in the purified protein.

FIG. 3. Competitive binding of purified ~1 to mouse L fibroblasts. Purified PQmethionine-labeled al was subjected to competitive cell attachment assay as described in the legend to Fig. 1. Lane 1, j?Slmethionine-labeled reovirus marker; lanes 2 and 3, binding of purified 01 to L cells in the presence of 3 mg/ml of reovirus and BSA, respectively.

REOVIRUS

PROTEIN

ul

2

381

4

8

16

32

64

128

256

PBS CONTROL’

FIG. 4. Hemagglutinating activity of purified ~1. Serial twofold dilutions of reovirus (10 pg/ml) or purified ~1 (20 rglml) were made in PBS in a microtiter plate. An equal volume of a 2% (v/v) suspension of human type 0 erythrocytes was then added to each well and the agglutination reaction was carried out at 22” for 4 hr. The numbers on top designate dilution factors of reovirus or purified ~1 used.

Oligomeric

forms of purified al

It has been shown previously that two molecules of ~1 are located at the tip of each vertex of the viral icosahedron (Lee et al., 1981 b). Based on the Sl gene sequence, it was also suggested that 01 might occur as dimers on intact virions (Bassel-Duby et al., 1985). Our present observation that the purified al is capable of agglutinating red blood cells further supports the notion that this protein is oligomeric in nature. Thus, it appears that al dimeric structures on the virion might be nonsusceptible to urea disruption, and were subsequently isolated as such. Alternatively, it is possible that monomeric ~1 forms were indeed produced after urea treatment, but spontaneously aggregated to yield oligomeric structures upon subsequent removal of the chaotropic agent. To demonstrate the presence of al oligomers in the preparation, purified al was crosslinked with DMS and subjected to SDS-PAGE. As shown in Fig. 5A, two crosslinked forms of al were resolved on the gel. The molecular weights of these two forms, determined on the basis of their electrophoretic mobilities (in 10% SDS-polyacrylamide gels) relative to those of marker proteins of known molecular weights, were 88K and 176K, corresponding precisely to those expected of dimers and tetramers, respectively, of al (Fig. 5B). The absence of trimers suggests that the dimeric form was the prevalent, if not the sole, oligomeric species in the purified al preparation, and that the tetramers were most likely produced by crosslinking between such dimers (Davies and Stark, 1970). Assuming there is only one HA domain on a al monomer, the crosslinking results may explain why purified ~1 is able to agglutinate human red blood cells as documented in the last section. Immunological

activities of purified al

To determine if antigenic determinants on the purified al are also preserved, the preparation was subjected

382

YEUNG

ET AL.

MYOSIN t

200K

97K

t-

0 v-

PHOSPHORYLASE

: IO

66K

3 . =

9 8 7I

ALBUMIN,

BOVINE

6 .43K 5 A/

0 DISTANCE FIG. 5. described with DMS. Molecular weight vs

MIGRATED

(cm)

Crosslinking studies of purified ~1. Purified [35S]methionine-labeled ~1 was treated with DMS and subjected to SDS-PAGE as under Materials and Methods. (A) Lane 1, [36S]methionine-labeled reovirus marker; lane 2, purified al ; lane 3, purified 01 crosslinked The bands marked a and b represent ~1 crosslinked products migrating at positions corresponding to 88K and 176K, respectively. weight standards: 200K, myosin; 97K, phospholylase b; 68K, bovine serum albumin; 43K, ovalbumin. (B) Semilog plot of molecular distance migrated (measured from the top of the gel) for 01 crosslinked products (from A).

to radioimmunoprecipitation/SDS-PAGE using a panel of monoclonal anti-al antibodies previously characterized [clones 10, 56, 86 and 102 (Lee et al., 1981 a)]. It was found that the purified al was recognized specifically by all four monoclonal anti-al antibodies, but not by irrelevant antibodies such as an anti-c2 antibody [clone 24 (Lee et al., 1981a)] (data not shown). Furthermore, if the neutralization domain(s) on the purified al is intact, prior incubation of the purified protein with these monoclonal antibodies (all previously shown to inhibit reovirus binding to host cells and to possess reovirus neutralizing activity) should result in a blocking of the binding of the purified al to L cells. That this was in fact the case is illustrated in Fig. 6: whereas the anti-a2 antibody showed no blocking activity (lane 3) all four monoclonal anti-al antibodies exhibited total inhibition of the binding of purified al to L cells (lanes 4-7). Purified al was also used to raise antibodies in rabbits as described under Materials and Methods. Serum was collected after three bimonthly injections of the purified protein (5 pg per injection). The serum was found to be capable of precipitating native al from a reovirus-infected cell lysate and of recognizing al on a Western blot (data not shown). It also blocked the binding of purified al to L cells (Fig. 6, lane 8) possessed neutralizing activity (a 1: 1000 dilution resulted in a plaque reduction of 1 log; data not shown), and

inhibited reovirus- and purified al -induced hemagglutination (Fig. 7). Collectively, these data have provided evidence that the four antigenic determinants assayed, which also

12345678

FIG. 6. Binding of purified 01 to mouse L fibroblasts in the presence of various antibodies, Purified [35S]methionine-labeled 01 was incubated with various antibody preparations at 22” for 30 min, and the mixtures were then applied to monolayers of L cells for al binding assay as described in the legend to Fig. 1. Lane 1, purified [36S]methionine-labeled reovirus; lane 2, al preincubated with a 1:50 dilution of rabbit preimmune serum (control); lane 3, al preincubated with 25 fig/ml of monoclonal anti-a2 IgG [from clone 24 (Lee et a/., 1981 a)]; lanes 4-7, al preincubated with 25 pglml of various monoclonal anti-al IgGs [from clones 10, 56. 86, and 102 (Lee et a/., 1981 a), respectively]; lane 8, al preincubated with a 150 dilution of rabbit anti-al serum.

PURIFICATION ANTIBODY

OF

DILUTION

REOVIRUS

PROTEIN

observations strongly suggest that al thus purified is in its native state. Biochemical

ANTI-01 ANTIBODY* i;

ANTI-l% ANTIBODY’

FIG. 7. Hemagglutination inhibition of purified al. Serial twofold dilutions of rabbit anti-al serum (originally diluted 1 :lO) or monoclonal anti-u2 IgG [from clone 24 (Lee et a/., 1981 a)] were made in PBS in a microtiter plate. Eight HA units of reovirus or purified al were added to each well. After 1 hr incubation at 22”, human type 0 erythrocytes were added and incubation was continued for 4 hr before the plate was examined.

happen to be neutralization domains, are preserved in the purified ~1, and that the antibodies elicited by the purified protein manifest activities that mirror those previously demonstrated with monoclonal anti-al antibodies (Lee et a/., 1981a, b). Coupled with the demonstration that the purified ~1 is functional in terms of its cell binding and hemagglutinating activities, these

TABLE AMINO ACID COMPOSITION

Amino acid

1 OF PROTEIN al

Determined by hydrolysis of 01 a (residues/molecule)

Ala Aw Asn Asp CYS Glu Gln GIY His Ile Leu LYS Met Phe Pro Ser Thr Trp Tyr Val B Determination based Trp not included. b Nagata et a/. (1984). ’ Not determined.

Predicted from genesequenceb (residues/molecule)

27.5 26.3

29 29

52.7

57

ND”

1

41.2

36

42.4 6.9 26.3 52.7 12.6 8.0 12.6 17.2 47.0 32.1 ND 8.0 35.5

38 5 31 52 7 8 12 14 52 37 5 8 34

on 449 amino

acids

383

01

for protein

Sl

01; Cys and

characterizations

of purified al

Purified al was also subjected to amino acid analysis. Table 1 shows that the amino acid composition of the purified protein was in close agreement with that inferred from the Sl gene sequence (Nagata et al., 1984). The pl of the purified protein was determined to be 6.8 (Fig. 8). Attempts were also made to determine the N-terminal amino acid sequence of the purified al using the Edman reaction (McKay et al., 1985). However, no phenylthiohydantoic (PTH)-amino acids could be detected after three degradation cycles (data not shown). It was concluded that like all other reovirus proteins (with the exception of plC), (Pett et a/., 1973) virion al has a blocked amino terminus. Secondary

structure

of purified al

To probe the secondary structure of purified al, the protein was subjected to circular dichroic (CD) determinations as outlined under Materials and Methods. Figure 9 shows the CD spectrum for ~1. The mean residue ellipticity at 220 nm had a value of -15100 deg. cm’. dmol-‘, indicating a relatively high helical content for al. The CD spectrum was further analyzed from 210 to 240 nm according to the procedure of Siegel et al. (1980); values of 54% a-helix and 19% P-sheet were predicted for the purified al.

IO9-

a57 654-

diiiiic, DISTANCE

(cm)

FIG. 8. lsoelectrofocusing of purified ~1. Purified 01 was labeled with iodine-l 25 and subjected to isoelectrofocusing in a 0.75-mmthick slab gel containing 5.696 acrylamide, 0.5% NW-methylene-bisacrylamide. and 2.2% ampholyte (range 3.5-10, LKB) as described under Materials and Methods. The 01 band was located by autoradiography and the pl of 01 was estimated from the pH gradients established in the blank lanes on both sides of the sample lane.

384

YEUNG

-7-T

l--l

190

200

210 WAVELENGTH

220 (nm)

230

240

FIG. 9. Circular dichroic spectrum of purified 01. Conditions are described under Materials and Methods. Protein al concentration was 24 pg/ml. Experimental data were corrected for a control sample lacking protein and are expressed as the mean residue ellipticity, [S], based on an amino acid mean residue weight of 107 for al calculated from the St gene sequence (Nagata et a/., 1984). The CD spectrum was repeated for three different samples.

DISCUSSION Despite the immediate relevance of al to reoviralinduced pathogenesis, the inability to isolate al has so far prevented an in-depth probing into the molecular biology of this protein. We have now developed a simple, rapid procedure whereby the most mature form of ~1 can be purified from intact virions to an apparent electrophoretic homogeneity. Our protocol calls for first the disruption of reovirus particles by the “mild” chaotropic agent, urea, followed by an ion-exchange chromatography step on DEAE-Sephacel. Protein al thus purified was found to be devoid of any detectable contamination by other viral proteins or RNA and is believed to be in its native state as judged by its cell binding and HA activities, its recognizability by a panel of monoclonal anti-al antibodies, and its ability to induce neutralizing and HA-inhibiting antibodies. This is important since many studies requiring the most mature form of authentic al can now be undertaken, the most obvious of which being X-ray diffraction studies of al crystals to yield an atomic resolution structure of this protein. Controlled fragmentization of the purified protein by various biochemical means, followed by studies similar to those described in this report, should further lead to the identification of domains that interact with cellular receptors and those that trigger host immune responses. The finding that the purified al exists as oligomers (mostly dimers) is consistent with the current notion that ~1 occurs in the virion as dimers, as deduced from stoichiometric considerations (Lee et al., 1981 b), as well as from feature analysis of the predicted amino acid sequence of ~1 (Bassel-Duby et a/., 1985). It is

ET

AL.

not known at this time whether these dimers exist as such after urea disruption of reovirions or are products of spontaneous aggregation of monomeric al molecules upon urea removal. We think that the latter possibility is more likely since al contains only one cysteine residue (Nagata et al., 1984) and virion ~1 monomers do not appear to be linked to each other via disulfide bridges (unpublished observation). Also, the relatively low HA titer of the purified protein suggests that a significant portion of it may still be in its monomeric form as a result of the urea treatment. It is not known whether binding of al to host cell receptors requires the protein to be in its dimeric form. The amino acid composition of the purified al closely matches the composition inferred from the Sl gene sequence (Table 1). Hence the possibility of any substantial post-translational processing of al is unlikely. It is possible that some minor processing such as the removal of a presumptive signal peptide as suggested previously (Nagata et al., 1984) does occur. However, evidence pertaining to this possibility is not available since we were unable to determine the amino-terminal sequence of the purified protein. Circular dichroic studies yielded values of 54% a-helix and 19% P-sheet structures for the purified ~1. Although confirmation of these figures awaits X-ray crystallographic data, values for a-helical content of proteins calculated from their CD spectra are generally quite accurate (Chen et a/., 1974; Siegel et a/., 1980; Foster et al., 1985). In this regard, it is of interest to note that previously, based on the Sl sequence, it was deduced that most a-helical regions on al are confined to the amino-terminal third of the molecule (Nagata et a/,, 1984; Bassel-Duby et a/., 1985) and that the (Yhelices of two al molecules are wound around each other to yield a coiled-coil structure (Bassel-Duby et a/., 1985). Experiments are now in progress to determine the actual topographical distribution of a-helical and other secondary structures on the purified al, as well as the correlation between these structures and the biological activities of this protein.

ACKNOWLEDGMENTS We thank Don McKay for analyzing the amino acid composition of our purified protein. We also acknowledge the help of Navin Khanna and David Waisman who assisted us with the circular dichroism studies, This work was supported by the Medical Research Council of Canada. M.C.Y. is a recipient of an Alberta Heritage Foundation for Medical Research (AHFMR) Studentship. M.J.G. is an AHFMR Clinical Investigator, and P.W.K.L. is an AHFMR Scholar.

REFERENCES BASSELL-DUBY, R., JAYASURIYA, A., CHATTERJEE, D., SONENBERG, N., MAIZEL, J. V., JR., and FIELDS, B. N. (1985). Sequence of reovirus

PURIFICATION

OF

haemagglutinin predicts a coiled-coil structure. Nature (London) 315,421-423. BURSTIN, S. J., SPRIGGS, D. R., and FIELDS, B. N. (1982). Evidence for functional domains on the reovirus type 3 hemagglutinin. Virology 117, 146-155. CASHDOLLAR, L. W., CHMELO, R. A., WEINER, J. R., and JOKLIK, W. K. (1985). Sequences of the Sl genes of the three serotypes of reovirus. Proc. Nat/. Acad. Sci. USA 82, 24-28. CHEN, Y.-H., YANG, J. T., and CHAU, K. H. (1974). Determination of the helix and /3 form of proteins in aqueous solution by circular dichroism. Biochemistry 13, 3350-3359. DAVIES, G. E., and STARK, G. R. (1970). Use of dimethyl suberimidate, a cross-linking reagent, in studying the subunit structure of oligomerit proteins. Proc. Nat/. Acad. Sci. USA 66, 651-656. FOSTER, D. L., MOWSRAY, S. L., JAP, B. K., and KOSHLAND, D. E., JR.. (1985). Purification and characterization of the aspartate receptor. /. Ho/. Chem. 260, 11706-l 17 10. HAYES, E. C., LEE, P. W. K., MILLER, S. A., and JOKLIK, W. K. (1981). Interaction of a series of hybridoma IgGs with reovirus particles: Demonstration that the core protein X2 is exposed on the particle surface. Virology 108, 147-155. JOKLIK, W. K. (1985). Recent Genet. 19, 537-575.

progress

in reovirus

research.

Ann.

Rev.

LEE, P. W. K., HAYES, E. C., and JOKLIK, W. K. (1981 a). Characterization of anti-reovirus immunoglobulins secreted by cloned hybridoma cell lines. virology 108, 134-l 46. LEE, P. W. K., HAYES, E. C., and JOKLIK, W. K. (1981 b). Protein 0l the reovirus cell attachment protein. virology 108, 156-l 63.

is

LEVIN, K. H., and SAMUEL, C. E. (1980). Biosynthesis of reovirus-specified polypeptides. Purification and characterization of the smallsized class mRNAs of reovirus type 3: Coding assignments and translational efficiencies. Virology 106, l-l 3. MASRI, S. A., NAGATA, Functional expression

L., MAH, D. C. W., and LEE, P. W. K. (1986). in fscherichia co/i of cloned reovirus Sl

REOVIRUS

PROTEIN

al

385

gene encoding the viral cell attachment protein ul . Virology 149, 83-90. MCCRAE, M. A., and JOKLIK, W. K. (1978). The nature of the polypeptide encoded by each of the double-stranded RNA segments of reovirus type 3. Virology 89, 578-593. MCKAY, D. J.. RENAUX, B. S., and DIXON, G. H. (1985). The amino acid sequence of human sperm protamine Pl. Biosci. Rep. 5, 383391. NAGATA, L., MASRI, S. A., MAH, D. C. W., and LEE, P. W. K. (1984). Molecular cloning and sequencing of the reovirus (serotype 3) Sl gene which encodes the viral cell attachment protein 01. Nucleic Acids Res. 12, 8699-87 10. PETE, D. M., VANAMAN, T. C., and JOKLIK, W. K. (1973). Studies on the amino and carboxyl terminal amino acid sequences of reovirus capsid polypeptides. Virology 52, 174-l 86. SHARPE, A. H., and FIELDS, B. N. (1985). Pathogenesis of viral infections. Basic concepts derived from the reovirus model. N. Engl. /. Med. 312,486-497. SIEGEL, J. B., STEINMETZ, W. E., and LONG, G. L. (1980). A computerassisted model for estimating protein secondary structure from circular dichroic spectra: Comparison of animal lactate dehydrogenases. Anal. Biochem. 104, 160-l 67. SMITH, R. E., ZWEERINK, H. J.. and JOKLIK, W. K. (1969). Polypeptide components of virions, top component and cores of reovirus type 3. Virology 39, 791-810. SPRIGGS, D. R.. KAYE, K., and FIELDS, B. N. (1983). Topological analysis of the reovirus type 3 hemagglutinin. Virology 127, 220-224. WEINER, H. L., RAMIG, R. F., MUSTOE, T. A., and FIELDS, B. N. (1978). Identification of the gene coding for the hemagglutinin of reovirus. Virology 86, 581-584. WRAY, W., BOULIKAS, T., WRAY, V. P., and HANCOCK, R. (1981). Silver staining of proteins in polyactylamide gels. Anal. Biochem. 118, 197-203. ZWEERINK, H. J., and JOKLIK, W. K. (1970). Studies on the intracellular synthesis of reovirus-specific proteins. Virology 41, 501-5 18.