Antigenic sites on the arginine-rich carboxyl-terminal domain of the capsid protein of hepatitis B virus distinct from hepatitis B core or e antigen

Molecular immunology,

Vol. 26,

0161-5890/89 $3.00+ 0.00

No. 4, pp. 413-421, 1989

Printedin Great Britain.

Pergamon Press plc

ANTIGENIC SITES ON THE ARGININE-RICH ~AR~~XYL-TERMINAL DOMAIN OF THE CAPSID PROTEIN OF HEPATITIS B VIRUS DISTINCT FROM HEPATITIS B CORE OR e ANTIGEN ATSUHIKQMACHIDA,*WITCHHIOHNUMA,*EMIKOTAKAI,* FUMIOTSUDA,* TAKESHITANAKA,? MASANORINAITO,$ EISUKEMUNEKATA,~ Yuzo MIYAKAWA§ and MAKOTO MAYUMI~~ *Section of Immunology, Kitasato Institute, Minato-Ku, Tokyo 108, Japan TJapanese Red Cross Blood Center, Saitama-Ken 362. Japan IInstitute of Applied Biochemistry, University of Tsukuba, Tsukuba City 305, Japan §Institute of Immunology, Tokyo 112, Japan ~~IrnmunoI~g~Division, Jichi Medical School, Mjnamikawachi-Macho, To&g&Ken 329-04, Japan

(Erst received 20 October 1988; accepted in rmsed

form9

December

1988)

Abstract-The capsid protein of hepatitis B virus (Pl9) is made of 183 amino acids and carries the antigenic sites of hepatitis B core antigen (HBcAg) and hepatitis B e antigen (HBeAg) on the aminoterminal domain. The ~rboxyl-behind domain of PI9 (amino acids 15@-183)is arginine-rich (47%) and faces the interior of the nucleocapsid for the binding with DNA. Mono&ma1 antibody was raised against an antigenic site on this protamine-like region of P19, which was distinct from HBcAg or HBeAg sites, and the novel antigenic site(s) was provisionally designated as hepatitis B inner core antigen (HBicAg). When PI9 in a low concn (150 ng/ml) was immobilized on the solid surface, HBicAg sites were preserved, while HBcAg or HBeAg sites were no longer available on it. This allowed the detection of antibodies against HBicAg (anti-~Bic), by sandwiching them between immobilized P19 and anti-IgG labeled with horseradish peroxidase. Anti-HBic was detected in sera from HBsAg carriers, typically those seropositive for antibody to HBeAg. A synthetic arginine-rich decapeptide, with a sequence of Arg-Arg-Arg-GlyArg-Ser-Pr+Arg-Arg-Arg, representing amino acids 1X&159 of PI9 and conserved in the majority of reported hepatitis B virus, absorbed the activity to bind with PI9 in seven (44%) out of 16 sera containing anti-HBic. These results indicate that the decapeptide carries an HBicAg epitope and the remaining amino acid sequence of the arginine-rich carboxyf terminal domain (160-183) may be responsible for the other HBicAg epitopes.

INTRODUCTION

A number of immune systems are recognized in association with infection with hepatitis B virus (HBV). Hepatitis B surface antigen (HBsAg) is carried by the viral coat and coded for by the envelope gene, consisting of the pre-Si and pre-S2 regions and the S gene friollais et al., 1985). Antibody against HBsAg denotes the te~~nation of infection and is useful in preventing primary infection

~“~-

-_~--

BAuthor to whom correspondence should be addressed. Abbreviations used: HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen; HBcAg, hepatitis B core antigen; anti-HBc. antibody to HBcAg; anti-HBcol, monoclonal antibody against the CI site of HBcAg; anti-HBc& monoclonal antibody against the fi site of HBcAg; HBeAg, hepatitis B e antigen; anti-HBe, antibody to HBeAg; anti-HBea, monoclonal antibody against the a site of HBeAg; anti-HBeb, monoclonal antibody against the b site of HBeAg; HBicAg, hepatitis B inner core antigen, representing the arginine-rich carboxylterminal domain of P19 (amino acids 150-183) and distinct from HElcAg or HBeAg; anti-HBic, antibody to HBicAg.

(Szmuness et al., 1980). Both hepatitis B core antigen (HBcAg) and e antigen (NBeAg) are present in the capsid protein of HBV (P19), coded for by the C gene (Takahashi et al., 1981). Antibody against HBcAg (anti-HBc) in the serum, when detected in high titers, indicates an ongoing infection (Hoofnagle ef al., 1973), while antibody against HBeAg (anti-HBe) signals low levels of HBV replication and circulating virions (Okada et al., 1976). In addition, antibody responses to the product of the X gene have been described, typically in patients with hepatoceIlular carcinoma ~Mo~arty ef al., 1985). By means of monoclonal antibodies, two distinct antigenic sites are defined for each HBcAg and HBeAg (Imai et al., 1982; Takahashi er al., 1983). P19 is made of 183 amino acids, and carries HBeAga and HBeAgf, sites, as well as HBcAga and HBcAgfi sites. P15.5 represents the amino-terminal domain of P19 (amino acids 1-149) and possesses both HBeAga and HBeAgb sites, as well as the HBcAga site; HBcAgP site is not detectable on it (Takahashi et af., 1983). P15.5 is secreted by mamm~ian &Is transfected with HBV DNA (Ou et al., 1986;Bruss and Gerlich, 1988; Garcia et al., 1988). 413

414

ATWHIKO

MACHIDA et ai.

HBcAg sites are exposed on the surface of a hepatitis B core particle, representing the nucleocapsid of HBV; core particles aggregate in the presence of anti-HBc in immune electron microscopy (Almeida et ni., 1971). HBeAg occurring in the serum of infected individuals is composed of P15.5 and serum protein (albumin or IgG) and is associated with two different physicochemical forms (Takahashi et al., 1978; Yamada et al., 1981). We have characterized a category of antigenic sites that is carried by P19, but is distinct from either HBcAg or HBeAg sites. The novel epitopes on Pi9 are borne by the arginine-rich carboxyl-terminal domain of P19 (amino acids 150-183) that faces the interior of the nucleocapsid and has a strong affinity for HBV DNA (Petit and Pillot, 1985). Antibodies against these epitopes were produced by hybridomas and detected in sera from individuals persistently infected with HBV. MATERIALS

Dane particie~ and HBV

AND METHODS

cupsid proteins

Core particles were isolated from Dane particles, which were derived from the plasma of asymptomatic carriers of HBsAg who were seropositive for HBeAg, by the method previously described (Takahashi et al., 1976). PI9 was solubilized from core particles by treatment with sodium dodecylsulfate (SDS) and 2-mercaptoethanol (2-ME) and purified by electrophoresis in SDS-polyacrylamide gel by the method of Takahashi et al. (1979), with a slight modification, i.e. by using a higher concn of 2-ME (5%) and at a higher temp (100°C). P15.5, representing amino acids l-149 of Pl9, was obtained from HBeAg in the serum of HBsAg carriers (Takahashi et al., 1980). Dane particles were phosphorylated with “P ([y -j*P]ATP: 3000 Ci/mmol; Amersham, Buckinghamshire, England) by means of endogenous protein kinase, after the method of Albin and Robinson (1980). They were then broken into their constituent poly~ptides and radiolabeled PI9 was obtained. Monoclonal

antibodies

Monoclonal antibodies were raised against P19 by a modification (Oi and Herzenberg, 1980) of the method originally described by Kohler and Milstein (1975). Each of two female BALBic mice, 6 weeks of age, received i.p. 100~1 of a solution of P19 (100 pg/ml), emulsified in Freund’s complete adjuvant (Difco Laboratories, Detroit, MI, U.S.A.). They were boosted by the same inoculum 3 weeks later and by a further 10 pg of P19 in IO0 pi of saline i.v., after an additional 3 weeks. Spleen cells were harvested 3 days after the final inoculation and fused with NS-1 cells by means of polyethyleneglycol. Hybridomas producing antibodies directed to the arginine-rich carboxyl-terminal domain of P19 were selected by the following procedure. Antibodies in the culture medium were screened by their reactivity with

P19, but not with P15.5, fixed on the solid surface. The wells of a microtiter plate (Immunoplate, Nune, Roskilde, Denmark) received a solution of P19 (150 ngiml) in phosphate buffer (0.01 M. pH 8.) and were then overcoated with buffer, supplemented with 40% (v/v) fetal calf serum. On the wells coated with P19 in concns less than 200 ng/ml, no HBeAg or HBcAg sites were available for binding with anti-HBc or anti-HBe (see Results). The wells of another plate were coated with PI55 (150 ngiml) and post-coated with fetal calf serum. Culture medium (50~1) was delivered to a well and captured antibodies were detected by rabbit anti-mouse immunoglobulin, labeled with horseradish peroxidase. Hybridomas that produced antibodies capable of binding with immobilized P19, but not with immobilized P15.5, were cloned. A clone producing such antibodies (No. 19-27) was propagated in the peritoneal cavity of mice, which had been pretreated with Pristane (Wako Pure Chemicals, Osaka, Japan). Ascites fluid was harvested -7-10 days after inoculation and the antibodies were precipitated with 2 M ammonium sulfate. Monoclonal anti-HBc antibodies against two distinct HBcAg sites, anti-HBcr (No. 3 105) and antiHBcp (No. 3120) as well as monoclonal anti-HBe antibodies directed to two different HBeAg sites, anti-HBea (No. 904) and anti-HBeb (No. 905). were characterized previously (Imai c’t fr2., 1982; Takahashi et al., 1983). Human sera Sera were obtained from individuals who carried HBsAg persistently. They were tested for HBsAg by reversed passive hemagglutination with a commercial assay kit (Mycell; Institute of Immunology Co. Ltd, Tokyo, Japan), for HBeAg and anti-HBe by enzyme immunoassay kits (IMMUNIS HBeAg/Ab EIA; Institute of Immunology), and for anti-HBc by hemagglutination inhibition (Shimizu et al., 1983). The titer of anti-HBe was expressed by the reciprocal of the dilution of serum that gave 50% inhibition in the enzyme immunoassay. The results of hemagglutination tests were expressed by the highest two-fold dilution (2”) of serum that induced positive reaction. S?Jnthetic P 19 peptides Eight synthetic peptides, representing parts of the arginine-rich carboxyl-terminal domain (amino acids 150-183) of Pi9 (deduced from an HBV DNA clone of subtype adr; Ono ef al., 1983), were prepared by the solid-phase method of Merrifield (1969). In addition, a tridecapeptide, representing amino acids 1488160 of Pl9, was synthesized. Solid-phase

enzyme immunoassays

The wells of an immunoplate were coated with phosphate buffer containing P19 (150 ng/ml) and overcoated with the buffer supplemented with 40% fetal calf serum. The test serum was diluted five-fold

Antigenic sites of HBV capsid protein with buffer and a 50 ~1 portion was delivered to a well. The plate was incubated at room temp for 1 hr and washed. Antibodies bound to a well were detected by monoclonal anti-human IgGy (No. 19G; Takai et al., 1986) labeled with horseradish peroxidase. An immunoplate was coated with tridecapeptide (IO~gjml), representing amino acids 148-160 of P19, and antibodies capable of binding with it were detected in five-fold diluted sera by enzyme-labeled monoclonal anti-human IgGy. Sera from 30 healthy individuals, without evidence of HBV infection, were processed similarly and the mean absorbance at 492nm was calculated. The result of test serum was expressed by sample/normal (S/N) ratio. An S/N ratio >2.5 was considered to be positive. Inhibition

tests

Human sera, containing antibodies able to bind with immobilized P19 or a tridecapeptide (148-160) in the enzyme immunoassay, were diluted appropriately. A 25 ~1 portion of diluted sera was mixed with an equal volume of phosphate buffer, supplemented with 30% fetal calf serum and containing one or other of synthetic peptides (2mg/ml), P15.5 (400 ng/ml) or P19 (500ng/ml), then incubated at 4°C overnight. The mixture was then delivered to a well of the immunoplate coated with P19 or the tridecapeptide (148-160) and the plate was incubated at room temp for 1 hr. Another 25 ~1 portion of diluted sera was mixed with 25 ~1 of buffer, supplemented with 30% fetal calf serum and processed similarly. Thereafter, bound antibodies were detected by enzyme-labeled monoclonal anti-human IgGy. The absorbance at 492nm was determined and the per cent inhibition was calculated as follows: (absorbance of buffer) - (absorbance of inhibitor) (absorbance Statistical

of buffer)

x 100.

analysis

Differences in the mean values between two groups were evaluated by the Wilcoxon rank sum test. RESULTS

Antigenic determinants exhibited before and after solubilization 2-mercaptoethanol

by core particles with SDS and

A monoclonal antibody (No. 19-27) with a specificity for the arginine-rich carboxyl-terminal domain of P19, was tested for the binding with core particles, before and after incubation at 100°C for 5 min, in the presence of 1% (w/v) SDS and -0.03-3% (v/v) 2-ME. HBcAg and HBeAg sites on the core particles were determined also. For this purpose, core particles, before and after treatment, were sandwiched between a monoclonal antibody

415

fixed on a solid support and another monoclonal antibody, with a different specificity, labeled with horseradish peroxidase (Fig. 1A and B). Untreated core particles displayed HBcAgcc and HBcAgP sites, but few (if any) HBeAga or HBeAgb sites; they did not bind with No. 19-27. After the treatment of core particles with SDS and 2-ME, the determinant detectable by No. 19-27 appeared along with HBeAga and HBeAgb. HBcAgcl was retained while HBcAgp disappeared. These results indicated that the antigenic site detectable by the monoclonal antibody (No. 19-27) was buried inside the core particles. D@erences in the expression of antigenic determinants on PI9 immobilized directly on the solid surface and on P19 captured by antibody P19 was fixed onto the wells of an immunoplate, in concns ranging from 25 to 12,800 ng/ml, and then tested for binding with the monoclonal antibody against the arginine-rich carboxyl-terminal domain of P19 (No. 19-27). Two monoclonal antibodies with a specificity for HBcAg, anti-HBco! and anti-HBc/I, and two with a specificity for HBeAg, anti-HBea and anti-HBeb, were tested in parallel (Fig. 2). Only No. 19-27 bound with P19 that had been immobilized on the solid surface in concns of less than 200 ng/ml; no monoclonal anti-HBc and antiHBe antibodies bound, to any appreciable extent, with P19 immobilized in these low concns. HBeAga and HBeAgb determinants were detetable on PI9 immobilized in higher concns, but HBcAg determinants were not. The monoclonal antibody, No. 19-27, bound with the synthetic tridecapeptide (148-160) representing the C-terminus sequence of the protamine-like region of P19, which had been immobilized on the solid surfacae, while none of the monoclonal anti-HBc and anti-HBe antibodies did. P19, labeled with ‘*P by the endogenous protein kinase, was immobilized onto a solid surface in concns of 100 and 400 ng/ml and various determinants were detected by monoclonal antibodies in the enzyme immunoassay. In another experiment, [32P]P19 was captured by anti-HBea, that had been fixed on a solid support, and then tested for various determinants. The results are shown in Table 1. Based on the radioactivity of 32P, the amount of [32P]P19 directly fixed on the solid support in a concn of lOOng/ml was similar to that immobilized via anti-HBea in a concn of 400 ng/ml. This was reflected in the number of sites detectable by No. 19-27, represented by the absorbance at 492nm in the enzyme immunoassay, which was comparable between [32P]P19 directly fixed in 100 ng/ml and [32P]P19 fixed via anti-HBea in 400ng/ml. No HBcAg or HBeAg sites were detectable on [32P]P19 which had been directly immobilized and bound only with No. 19-27. In contrast, HBcAg and HBeAg sites were available on [32P]P19 captured by anti-HBea, in addition to the site detectable by No. 19-27.

ATSUHIKO MACHIDA et ai.

416

P-----O

(A)

I

Antibodies Fixed Labeled 0 0 . P

SDS (%)O Z-ME(%)0

1 0.03

c:a c/u c/p e/a

1 0.3

Antibodies Fixed Labeled

C/P eib elb elb

0 0

1 3

Treatment

1 0 03

1 0.3

1 3

of core parttcles

Fig. 1. Antigenic sites detectable on core particles treated with sodium dodecylsulfate (SDS) and 2-mer~aptoethanol (2-ME). Core particles, purified from Dane particles in the sera of asymptomatic carriers, were treated with 1% SDS and -0.03-3% 2-ME at 100°C for 5 min, then subjected to the solid-phase enzyme immunoassasy. Core particles and PI9 soiubilized from them were sandwiched between a monoclonal antibody of a certain specificity, fixed on a solid support, and another monoclonal antibody of a different specificity labeled with horseradish peroxidase. Seven combinations of two monoclonal antibodies were tested, four in (A) and four in (B). Specificities of monoclonal antibodies against HBcAg and HBeAg sites are indicated in abbreviated forms. The monoclonal antibody No. 19-27 was directed to the arginine-rich carboxyl-terminal domain of P19.

l Antl-P19 A Antl-HBeia a Antl-HBeib l Ant!-H&/a CI Anti-HBci@

(No.1 9-271 (No 904) (No.9051 (No.3105) (No.31 20)

I-/I

25

50

100

200

400

Concentratton

800

7,600

3,200

6.400

12,800

of P19. ngiml

Fig. 2. Binding of monoclonal antibodies with immobilized specificities were tested for the binding with P19 immobilized The bound antibodies were detected by rabbit anti-mouse peroxidase in the enzyme immunoassay. Anti-PI9 (No. carboxyl-terminal domain

P19. Monoclonal antibodies of various onto the solid surface in increasing concns. immunoglobulin labeled with horseradish 17-29) was directed to the arginine-rich of P19.

Antigenic sites of HBV capsid protein Table

417

1. Determinants detectable by mo”oclonal antibodies on PI9 immobilized directly onto a solid support or on PI9 captured by immobilized antibody”

PI9 (“n/ml) 400 100 0

PI9 fixed directly onto a solid support Anti-ea Anti-eb Anti-cm 19-21 0.012 0.01 I 0.010

0.012 0.012 0.010

0.035 0.033 0.033

2.185 0.917 0.015

PI9 fixed via anti-HBea Anti-eb Anti-cm 19-27 2.902 I .982 0.044

1.429 0.163 0.027

1.223 0.198 0.032

“PI9 labeled with 32P was immobilized onto wells of an immunoplate, in a cone” of 400 or 100 “g/ml, and overcoated with fetal calf serum. Antigenic determinants, detectable by monoclonal antibodies, were tested for by enzyme immunoassay and expressed by absorbance at 492 “m. In the other experiment, [‘2P]P19 in two different concns was captured by anti-HBea, that had been fixed onto wells, and then tested for determinants available by monoclonal antibodies. The amount of [32P]P19 directly immobilized using a cone” of 100 “g/ml was comparable with that captured by anti-HBea at a concn of 400 “g/ml, as estimated by the radioactivity of 32P bound to the wells of immunoplates.

Antibodies in human sera capable of binding with P 19 immobilized in a low concn (150 ng/ml) Forty sera from HBsAg carriers positive for antiHBe (A-l - A-40) and 40 sera from carriers positive for HBeAg (B- 1 - B-40) were tested for binding with PI9 immobilized in 150 ng/ml. All of them contained high hemagglutination titers for anti-HBc (> 2”). Antibodies capable of binding with PI9 were detected in 23 (57.5%) out of 40 sera containing anti-HBe, while they were found in only one (2.5%) of 40 sera containing HBeAg. Among 40 anti-HBe positive sera, the mean titer of anti-HBe in 23 sera with such antibodies was significantly higher than that in 17 sera without such antibodies, in the enzyme immunoassay (antibody titer: 15.1 f 9.2 us 10.2 _t 6.7 x IO*, p < 0.05). Among 80 HBsAg positive sera tested, the titer of anti-HBc in 24 sera with such antibodies was not different from that in 56 without (the geometric mean of the hemagglutination titer: 13.5 f 1.6 us 13.6 k 1.8, p > 0.1). Table 2 lists the 24 sera containing antibodies that bound with immobilized P19. The specificity of the antibodies was evaluated in 16 of them with an S/N ratio higher than 20 (-A-l-A-l5 and B-l). The binding was invariably inhibited by P19, but not by P15.5. The antibodies reactive with PI9 immobilized in low concns, distinct from anti-HBc or anti-HBe, were directed to the arginine-rich carboxyl-terminal domain of P19 (amino acids 15&183), since they were not absorbed by P15.5, representing the aminoterminal domain of P19. Antigenic sites recognized by these antibodies were collectively designated as hepatitis B inner core antigen and will be referred to as HBicAg hereafter. The antibody against HBicAg, detectable by the binding with PI9 immobilized in a low concn (150 ng/ml), will be called anti-HBic. Absorption oj’human anti-HBic with syntheticpeptides copying the amino acid sequences of the arginine-rich carboxyl- terminal domain of P 19 Eight peptides, that copied parts rich carboxyl-terminal domain of deduced from an HBV DNA clone (Ono et al., 1983)], were synthesized. peptides, along with a synthetic

of the arginineP19 [(15CLl83) of subtype adr These synthetic tridecapeptide

(148-160), were evaluated for their inhibition on the activity of antibodies to bind with immobilized P19, in the 16 sera with an S/N ratio >20. Table 3 gives the results with three peptides: 148-160, representing the N-terminal region of the arginine-rich protamine-like domain of P19; 161-183, representing the remaining C-terminal region; and 157-l 66, covering the junction in between. Significant inhibition was observed only by the N-terminal tridecapeptide (148-160). It inhibited the binding of antibodies with immobilized PI9 by > 50% in seven (44%) out of the 16 sera tested. No appreciable inhibition was observed by the other two peptides on the binding of any of the antibodies to immobilized P19.

Table 2. Antibodies in sera from HBsAg carriers capable of binding with PI9 immobilized onto the solid surface in a low cone” Cl50 neiml~ Serum No.

A-1 A-2 A-3 A-4 A-5 A-6 A-7 A-8 A-9 A-IO A-II A-12 A-13 A-14 A-15 A-16 A-11 A-18 A-19 A-20 A-21 A-22 A-23 B-I

Anti-HBc titer (2”) IS 14 12 15 II I6 I5 15 I3 I3 I3 14 IO 14 16 I4 I2 I3 I4 I3 II I2 I3 16

Anti-HBe titer ( x IO*)* 8 I4 5 20 9 24 I6 50 4 I7 16 12 20 I6 8 I4 I6 IO 18 I8 9 17 _*

Binding with PI9 (SIN)’ 121.2 100.0 94.8 85.7 75.1 13.5 71.0 70.9 68.1 65.0 62. I 42.6 40.9 40.7 26.3 17.6 12.0 7.6 1.5 6.6 3.6 3.5 3.4 60.6

Per cent inhibition on the binding with PI9 by PI9 P15.5 99 99 97 99 99 94 96 99 97 96 98 95 95 95 95 NT“ NT NT NT NT NT NT NT 96

5 4 3 5 4 0 I 6 3 2

I 2 4 2 4 NT NT NT NT NT NT NT NT 4

The highest dilution of serum that induced a positive reaction in hemagglutination inhibition. ‘Dilution of serum that gave 50% inhibition in the determination of anti-HBe by enzyme immunoassay. ‘Ratio between the absorbance of the sample and the mea” absorbance of negative controls in enzyme immunoassay. Values greater than 2.5 were considered positive. “Not tested. ‘HBeAg was detected.

418

ATSUHIKO MACHVJA Table 3. Per cent inhibmon by synthetic peptides, representing parts of the argininerich carboxyl-terminal domain of Pl9. on the binding of human antibodres with P19 immobilized in a iox. concn (1 SOng;ml)” Serlim NO. A-l A-2 A-3 A-4 A-i A-6 A-7 A-X .A-9 A-IO A-11 A-12 A-13 A-14 A-15

B-I

Amino acid positions of synthetic peptides 14X-160 157-166 161-183 II 39 3N 91 YI 27 91 07 7 83 20 25 15 Yl 70

2

0 2 3 5 3

5 3 2 7 0

I

18

0 9 6 0 IO 5 0 3

er al

tide sequence, was immobilized on a solid support. Captured human antibodies were then detected by enzyme-labeled monoclonal anti-I@. The specificity of detected antibodies for the decapeptide (150-159) was established, when it inhibited the binding with immobilized tridecapeptide (148-160) by > 50%. By this assay, anti-HBic with a specificity for the decapeptide was detected in seven out 01 16 sera that showed high binding activities with immobilized P19.

I?

I0

DISCUSSION

0 2 8 5 16 9

1

s

16

4

“Synthetic peptides were tested for the capacity to inhibil the binding of human antibodies with Immobilized P19. Sera labeled A were from carriers of HBsAg seropositive for anti-HBe and the serum labeled B WL~S from a carrier seropositive for HBeAg.

The epitope to which anti-HBic was directed was further specified by the inhibition with six synthetic peptides (representing parts or the whole of amino acids 150-160) on six of the seven sera {in Table 3) containing antibodies whose binding to PI9 was inhibited by the tridecapeptide (148--160). The two peptides listed at the top of Table 4 inhibited the binding of six sera extensively (>80%). The inhibition by the other four peptides was mostly low and variable, on the pane1 of sera containing antiHBic. The least common denominator required for the full inhibition was deduced to be the decapeptide with a sequence of Arg--Arg-Arg--Gly--Arg-SerPro-Arg-Arg-Arg, seven amino acid residues of which were arginines. This decapeptide inhibited the binding with PI9 of the remaining one serum in Table 3 (A-15). These results indicated that the decapeptide carried one of the HBicAg epitopes.

A solid-phase enzyme immunoassay.fiv anti-HBic with th4 spec$city for an HBicAg decapcptide (150-I 59) Synthetic HBicAg peptides were applied to develop a sandwich enzyme immunoassay for anti-HBic. A tridecapeptide (148-I 60), containing the decapep-

The antigenic configuration of HBV is classified into two entities, i.e. envelope and capsid antigens. The envelope antigen of HBV is collectively referred to as HBsAg and borne by pol~eptides, coded for by a part or the whole of the envelope gene, comprising the pre-Sl and pre-S2 regions and the S gene (Tiollais rt ul., 1985). The products of these regions and gene, which constitute the viral envelope and occur as small non-infectious particles also, elicit respective antibody responses in the host and their corresponding antibodies have various clinical, epidemiologic and immunoprophylactic bearings (Szmuness er al., 1980; Neurath et al., 1985; Itoh e’t al., 1986; Ise ft al., 1988; Tsuda et al., 1988). The capsid antigen of HBV is carried by the poiypeptidc made of 183 amino acid residues (P19f, which, along with HBV DNA and DNA polymerase, assembles the core particle. Two groups of antigenic sites are presently recognized on P19, i.e. HBcAg and HBeAg (Almeida et al., 1971; Magnius and Espmark, 1972; Takahashi et al., 1979). They are distinguished by HBcAg sites, which are available on the exterior of the core particle. white HBeAg sites are not. Carried by the identical polypeptide (P19), HBcAg and HBeAg elicit distinct antibody responses in the host and the determination of anti-HBc and anti-HBe provide useful information in various clinical and epidemiologic settings (Hoofnagte et uf., 1973; Okada et al., 1976). Here we report a third group of antigenic sites on P19, that is present on the arginine-rich carboxyl-terminal domain (amino acids 150-183), and is distinct from either HBcAg or HBeAg sites. We would like to propose that it be called the hepatitis B inner core antigen, or HBicAg in an abbreviated manner, until a more sophisticated and systematic nomenclature for the envelope and capsid antigens of HBV has

Table 4. Inhibition by synthetic HBicAg peptides on the binding of human anti-HBx wth PI9 Immoblhaed in a low concn (I 50 neiml) Synthetic HBicAg peptldes” (position of amino acid residues) RRRGRSPRRRT RRRGRSPRRR RRRGRSPRR RRRGRSPR RRGRSPRRRT RGRSPRRRT

(1X&160) (1.5&159) (150~1.58) (156157) (151.160) (152-160)

Per cent inhibition on human anti-HBic A-IO A-14 A-S A-7 A-8 A-4 91 01 20 0 79 28

85 84 0 0 13 9

91 x9 41 9 76 2

98 -98 52 n 19 8

88 90 17 0 47 37

91 92 82 49 74 0

“Smgle letter abbreviations for the amino acid residues are: G. glycine: P. proline: R. arginine. S. wins: T. thrconine.

Antigenic

sites of HBV capsid

been invented. The antibody to HBicAg will hence be referred to as anti-HBic. Human hosts recognize HBicAg determinants and raise humoral antibodies. There remains little doubt that the arginine-rich carboxyl-terminal domain of P19 can stimulate a B cell response in human beings, although they are not recognized by T cells in congenic mice (Milich et al., 1987). The specificity of human anti-HBic for the carboxyl-terminal domain of PI9 was ascertained by the failure of P15.5, carrying HBeAga and HBeAgb sites as well as the HBcAgcr site, in absorbing its capacity to bind with immobilized P19. HBcAgp was no longer available on the P19 obtained under the conditions we employed. It was possible to preserve HBcAgP on Pl9, however, under more mild conditions with a lower concn of 2-ME and at low temp, for solubilization of core particles (Takahashi et al., 1983). These observations would indicate that, unlike the HBcAg? site, the HBcAgB site is dependent on the conformation of P19. Proteins, poly- and oligopeptides adhere to the surface of polyvinyl or polystylene; this phenomenon has been widely applied to the sensitive determination of antigens and antibodies in solid-phase radio- and enzyme immunoassays. The mechanism whereby they are immobilized remains unclear, but may not be as haphazard as it seems. There would be active sites on a molecule by which it can attach to the solid surface. Polypeptides may change their conformation after they are immobilized on the solid surface, so that some of the epitopes exhibited by them, when they are in solution, are no longer available. In converse, the immobilization of a molecule onto the surface may reveal epitopes that are not accessible on it in solution. For example, a determinant of a-fetoprotein, detectable by the monoclonal antibody, doubles itself after the molecule is fixed on a solid support (Nomura et al., 1982); a cryptic epitope is exposed by a conformational change brought about by the immobilization. Different expression of antigenic determinants, between molecules immobilized on a solid support and molecules in solution, was typically observed for PI9 and this might be due to the specific manner in which PI9 binds onto the solid surface. PI9 would adhere to the surface involving its sites, which are essential for the expression of HBcAg and HBeAg sites, until its population exceeds a certain level. When P19 molecules crowd on the surface, however, some would no longer be able to stretch themselves liberally. The result would be that PI9 molecules would anchor to the surface in such a manner that would allow HBcAg and HBeAg sites to be accessible for binding with corresponding antibodies. HBicAg determinants were carried by amino acid sequences in the arginine-rich carboxyl-terminal domain of P19, a protamine-like region implicated in the binding with HBV DNA (Petit and Pillot, 1985). A synthetic decapeptide, containing seven arginine

protein

419

residues (150-159) with a sequence of Arg-ArgArg-Gly-Arg-Ser-PrcArg-Arg-Arg, inhibited appreciably (> 70%) the activity to bind with immobilized PI9 in seven (44%) out of 16 sera containing anti-HBic. These results indicate that this argininerich decapeptide constitutes an HBicAg epitope and that there could be HBicAg epitopes other than the one borne by the decapeptide. The decapeptide is conserved in six HBV genomes of subtype U&J (Retno et al., 1985; Okamoto et al., 1987, 1988; Vaudin et al., 1988) four adr genomes (Fujiyama et al., 1983; Ono et al., 1983; Kobayashi and Koike, 1984; Gan et al., 1987) two uyw genomes (Galibert et al., 1979; Bichko et al., 1985) and one ayr genome (Okamoto et al., 1986). The ubiquity of the HBicAg epitope borne by the decapeptide, expressed by the majority (13 out of 16) of reported HBV DNA clones, would be useful for determining anti-HBic in individuals infected with HBV of various subtypes. The three adw HBV genomes, two originated in the United States (Valenzuela et al., 1980; Ono et al., 1983) and one in the Philippines (Estacio et al., 1988) appear exceptional, because they have an insertion of -Asp-Argbetween the third Arg and Gly in the decapeptide. This difference might account for, at least in part, the failure in detecting anti-HBic by immobilized decapeptide in some human sera. The validity of this assumption would have to be evaluated by testing these sera for binding with synthetic dodecapeptide, which has an insertion of -AspArg-. Core particles, purified from infected liver or circulating virions, display an activity for protein kinase (Albin and Robinson, 1980) which phosphorylates serine residues of the arginine-rich carboxyl-terminal domain of PI 9, both in vivo and in vitro (Feitelson et al., 1982; Gerlich et al., 1982; Roossinck and Siddiqui, 1987). Based on the conformation of capsid proteins deduced from nucleotide sequences of hepadnaviruses, serine residues at positions 155, 162, 168 and 170 are proposed as candidates for phosphorylation (Argos and Fuller, 1988). The state of phosphorylation of serine residue(s) in the argininerich carboxyl-terminal domain of P19, which might affect the expression of HBicAg sites, would deserve a further analysis.

REFERENCES Albin C. and Robinson W. S. (1980) Protein kinase activity in hepatitis B virus. J. Viral. 34, 2977302. Almeida J. D., Rubenstein D. and Stott E. J. (1971) New antigen-antibody system in Australia-antigen-positive hepatitis. Lancet ii, 122551227. Argos P. and Fuller S. D. (1988) A model for the hepatitis B virus core protein: prediction of antigenic sites and relationship to RNA virus capsid proteins. EMBO J. 7, 819-824. Bichko V., Pushko P., Dreilina D., Pumpen P. and Gren E. (1985) Subtype uq’w variant of hepatitis B virus. DNA primary structure analysis. FEBS Lcttr 185, 2088212. Bruss V. and Gerlich W. H. (1988) Formation of transmembraneous hepatitis B e-antigen by cotranslational in

420

ATSUHIKO MACHIVA et ai

vitro processing of the viral precore protein. Viro/og?, 163, 268-275. Estacio R. C., Chavez C. C., Okamoto H., Lingao A. L., Reyes M. T.. Domingo E. and Mayumi M. (1988) Nucleotide sequence of a hepatitis B virus genome of subtype a& isolated from a Philippino: comparison with the reported three genomes of the same subtype. J. Gastroenternl. Hepatol. 3, 215-222. Feitelson M. A.. Marion P. L. and Robinson W. S. (1982) Core particle of hepatitis B virus and ground squirrel hepatitis virus. II. Characterization of the protein kinase reaction associated with ground squirrel hepatitis virus and hepatitis B virus. J. Virof. 43, 741 748. Fujiyama A., Miyanohara A.. Nozaki C., Yoneyama T., Ohtomo N. and Matsubara K. (1983) Cloning and structural analyses of hepatitis B virus DNAs, subtype adr. Nucl. Acid.s Res. 11, 4601-4610. Galibert F., Mandart E., Fitoussi F., Tiollais P. and Charnay P. (1979) Nucleotide sequence of the hepatitis B virus genome (subtype aq’n*) cloned in E. et&. Narure 281, 646650. Gan R., Chu M., Shen L., Qian S. and Li Z. (1987) The complete nucleotide sequence of the cloned DNA of hepatitis B virus subtype adr in pADR- 1. Scientia sirzicn 30, 507752 1. Garcia P. D.. Ou J.-H., Rutter W. J. and Walter P. (1988) Targeting of the hepatitis B virus precore protein to the endoplasmic reticulum membrane: after signal peptide cleavage translo~ation can be aborted and the product reieased into the cytoplasm. J. Cell Biol. 106, 1093-l 104. Gerlich W. H., Goldmann U., Miiller R., Stibbe W. and Wolff W. (1982) Specificity and localization of the hepatitis B virus-associated protein kinase. J. Viral. 42, 761&766. Hoofnagle J. H., Gerety R. J. and Barker L. F. (1973) Antibody to hepatitis-B-virus core in man. Lancer ii, 869-873. Imai M., Nomura M., Gotanda T.. Sano T., Tachibana K.. Miyamoto H., Takahashi K., Toyama S., Miyakawa Y. and Mayumi M. (1982) Demonstration of two distinct antigenic determinants on hepatitis B e antigen by monoclonal antibodies. 1. Immun. 128, 6972. ise I., Tsuda F., Aihara S.. Machida A.. Takai E.. Miyamoto H., Akahane H., Miyakawa Y. and Mayumi M. (1988) Antibodies to translation products of the pre-Sl and preS2 regions of the envelope gene of hepatitis B virus in fulminant hepatitis B. Hepafology 8, 1089-1093. Itoh Y., Takai E., Ohnuma H., Kitajima K., Tsuda I?., Machida A., Mishiro S., Nakamura T., Miyakawa Y. and Mayumi M. (1986) A synthetic peptide vaccine involving the product of the pre-S(2) region of hepatitis B virus DNA: protective efficacy in chimpanzees. Proc. nutn. Acad. Sci. U.S.A. 83, 9174-9178. Kobayashi M. and Koike K. (1984) Complete nucleotide sequence of hepatitis B virus DNA of subtype adr and its conserved gene organization. Gene 30, 2277232. Kijhler G. and Milstein C. (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256, 495497. Magnius L. 0. and Espmark J. A. (lY72) New specificities in Australia antigen positive sera distinct from the Le Bouvier determinants. J. Immun. 109. 1017~~1021. Merrifield R. B. (1969) Solid-phase ‘peptide synthesis. Adz>. Enzymol. 32, 221-296. Mihch D. R., McLachlan A., Moriarty A. M. and Thornton G. B. (1987) Immune response to hepatitis B virus core antigen (HBcAg): localization of T cell recognition sites within HBcAg/HBeAg J. Immun. 139, 122331231. Moriarty A. M., Alexander H., Lerner R. A. and Thornton G. 8. (1985) Antibodies to peptides detect new hepa-

titis B antigen: serological correlation with hepatocellular carcinoma. Science 227, 4299433. Neurath A. R.. Kent S. B. H., Strick N., Taylor P. and Stevens C. E. (1985) Hepatitis B virus contains pre-S gene-encoded domains. Nature 315, 154156. Nomura M., Imai M., Nakamura T., Miyakawa Y. and Mayumi M. (1982) Demonstration of two distinct antigenic determinants on human z-foetoprotein by monoclonal antibodies. M&c. Immun. 19, 1691- 1697. Oi V. T. and Herzenberg L. A. (1980) lmmunoglobulin producing hybrid cell-lines. In Selecred Methods in Cellukrr Immunofog~ (Edited by Mishell B. B. and Shiigi S. M.), pp. 351 572. W. H. Freeman, San Francisco. Okada K...Kamiyama I., Inomata M., Imai M., Miyakawa Y. and Mayumi M. (1976) e Antigen and anti-e in the serum of asymptomatic carrier mothers as indicators of positive and negative transmission of hepatitis B virus to their infants. Nen Engl. J. Med. 294, 746 749. Okamoto H., imai M., Kametani M., Nakamura T. and Mayumi M. (1987) Genomic heterogeneity of hepatitis B virus in a 54-year-old woman who contracted the infection through materno-fetal transmission. Jpn. J. exp. Med. 57, 231-236. Okamoto H., Imai M., Shimoraki M., Hoshi Y., Iizuka H.. Gotdnda ‘T., Tsuda F., Miyakawa Y. and Mayumi M. (1986) Nucleotide sequence of a cloned hepatitis B virus genome, subtype ayr: comparison with genomes of the other three subtypes. J. gen. Viral. 67, 23052314. Okamoto H.. Tsuda F., Sakugawa H., Sastrosoewignjo R. I., Imai M., Miyakawa Y. and Mayumi M. (1988) Typing hepatitis B virus by homology in nucleotide sequence: comparison of surface antigen subtypes. J. gen. Viral. 69, 25752583. Ono Y., Onda H., Sasada R., Jgarashi K., Sugino Y. and Nishioka K. (1983)The complete nucleotide sequences of the cloned hepatitis B virus DNA: subtype uclr and a&. Nuci. Acids drs. 11, 1747-1757. -. Ou J.-H.. Laub 0. and Rutter W. J. (1986) Henatitis B virus gene function: the precore region targets the core antigen to cellular membranes and causes the secretion of the e antigen. Proc. natn. Acad. Sci. U.S.A. 83, 1578 1582. Petit M.-A. and Pillot J. (1985) HBc and HBe antigenicity and DNA-binding activity of major core protein P22 in hepatitis B virus core particles isolated from the cytoplasm of human liver cells. J. Vi&. 53, 543 -55 i Retno I. S., Okamoto H., Mayumi M., Warsa U. C. and Sujudi (1985) The complete nucleotide sequence of an HBV DNA clone of subtype adw (pRTB299) from Indonesia. ICMR A. 5, 39-50. Roossinck M. J. and Siddiqui A. (1987) In kw phosphorylation and protein analysis of hepatitis B virus core antigen. J. piroi. 61, 955-961. Shimizu M., Ohyama M., Takahashi Y., Udo K,, Kojima M., Kametani M.. Tsuda F., Takai E.. Miyakawa Y. and Mayumi M. (1983) lmmunoglobulin M antibody against hepatitis B core antigen for the diagnosis of fulminant type B hepatitis. Gastroenterology 84, 604610. Szmuness W., Stevens C. E., Harley E. J., Zang E. A., Oleszko W. R.. William D. C.. Sadovsky R., Morrison J. M. and Keilner A. (1980) Hepatitis B vaccine. Demonstration of efficacy in a controlled clinical trial in a high-risk population in the United States. Nebe Engl. J. Med. 303, 833 841. Takahashi K.. Akahane Y.. Gotanda T.. Mishiro S., lmai M., Miyakawa Y. and Mayumi M. (1979) Demonstration of hepatitis B e antigen in the core of Dane particles. f. Inzmru2. 122, 275-279. Takahashi K., Imai M.. Gotanda T.. Sano T., Oinuma A., Mishiro S., Miyakawa Y. and Mayumi M. (1980) Hepatitis B e antigen polypeptides isolated from sera of individuals infected with hepatitis B virus: comparison with HBeAg polypeptide derived from Dane particles. J. pw. Viral. 50, 49 -57.

Antigenic

sites of HBV capsid

Takahashi K., Imai M., Miyakawa Y., Iwakiri S. and Mayumi M. (1978) Duality of hepatitis B e antigen in serum of persons infected with hepatitis B virus: evidence for the nonidentity of e antigen with immunoglobulins. Proc. natn. Acad. Sci. U.S.A. 15, 1952-1956. Takahashi K., Imai M., Nomura M., Oinuma A., Machida A., Funatsu G., Miyakawa Y. and Mayumi M. (1981) Demonstration of the immunogenicity of hepatitis B core antigen in a hepatitis B e antigen polypeptide (P19). J. gen. Virol. 57, 325-330. Takahashi K., Machida A., Funatsu G., Nomura M., Usda S., Aoyagi S., Tachibana K., Miyamoto H., Imai M., Nakamura T., Miyakawa Y. and Mayumi M. (1983) Immunochemical structure of hepatitis B e antigen in the serum. J. Immun. 130, 2903-2907. Takahashi T., Nakagawa S., Hashimoto T., Takahashi K., Imai M., Miyakawa Y. and Mayumi M. (1976) Largescale isolation of Dane particles from plasma containing hepatitis B antigen and demonstration of a circular double-stranded DNA molecule extruding directly from their cores. J. Zmmun. 117, 1392-1397. Takai E., Machida A., Ohnuma H., Miyamoto H., Tanaka T.. Baba K., Tsuda F., Usuda S., Nakamura T., Miyakawa Y. and Mayumi M. (1986) A solid-phase enzyme immunoassay for the determination of IgM and

protein

421

IgG antibodies against translation product of pre-Sl and pre-S2 regions of hepatitis B virus. J. Immun. Meth. 95, 23-30. Tiollais P., Pourcel C. and Dejean A. (1985) The hepatitis B virus. Nafure 317, 489-495. Tsuda F., Masuko K., Mitsui T., Akahane H., Machida A., Miyakawa Y. and Mayumi M. (1988) Antibody against the translation product of the pre-S2 region of hepatitis B virus and the resolution of infection after accidental needlestick exposure. Am. J. Gastroenferol. 83, 943-947. Valenzuela P., Quiroga M., Zaldivar J., Gray P. and Rutter W. J. (1980) The nucleotide sequence of the hepatitis B viral genome and the identification of the major viral genes. In Animal Virus Genetics (Edited by Fields B. N., Jaenisch R. and Fox C. F.), pp. 57770. Academic Press, New York. Vaudin M., Wolstenholme A. J., Tsiquaye K. N., Zuckerman A. J. and Harrison T. J. (1988) The complete nucleotide sequence of the genome of a hepatitis B virus isolated from a naturally infected chimpanzee. J. gen. Virol. 69, 1383-1389. Yamada E., Ohori H. and Ishida N. (1981) Molecular heterogeneity of hepatitis B virus e antigen in liver and serum. J. gen. Viral. 55, 75-86.