- Email: [email protected]
A monoclonal antibody against a hepatitis B e antigen epitope borne by six amino acids encoded by the precore region
ELSEVIER
Journal of Virological Methods Journal of Virological Methods 68 (1997) 207 215
A monoclonal antibody against a hepatitis B e antigen epitope borne by six amino acids encoded by the precore region Sadakazu Usuda a Hiroaki Okamoto a Hitoshi Ohnuma b Takeshi Tanaka Atsuhiko Machida c Yuzo Miyakawa d, Makoto Mayumi a,,
b
Immunology Division, Jichi Medical School, Tochigi-Ken 329-04, Japan u Japanese Red Cross Saitama Blood Center, Saitama-Ken 338, Japan c Yamanashi Institute fi)r Public Health, Yarnanashi-Ken 400, Japan d M~vakawa Memorial Research Foundation, Tokyo 107, Japan
Received 4 April 1997; received in revised form 21 July 1997: accepted 27 July 1997
Abstract
Hepatitis B e antigen (HBeAg) polypeptide in the circulation (p17 e) is composed of ten amino acids (aa) coded for by the precore region and 149 aa by the core gene of hepatitis B virus. A monoclonal antibody (Y0583A) was raised against the N-terminal ten amino acids (SKLCLGWLWG) encoded by the precore region. The binding of Y0583A with a panel of 203 decapeptides on multipins, which covered the precursor of HBeAg polypeptide made of 212 aa shifting by one aa, recognized an epitope sequenced L G W L W G representing the C-terminal six aa coded for by the precore region. This HBeAg epitope was not readily accessible on HBeAg in serum, but it became exposed and bound with Y0583A by treatment with 0.2 N NaOH. Using Y0583A, an enzyme-linked immunosorbent assay was developed for specific determination of HBeAg. The test sample was incubated with the monoclonal antibody to an HBeAg determinant encoded by the core gene (904) that had been immobilized on a solid support. Captured HBeAg was treated with 0.2 N NaOH, neutralized and released into the fluid phase. The reactant was then tested for a sandwich between monoclonal antibody (C33) to the C-terminus of the HBeAg polypeptide immobilized on a solid support and Y0583A labeled with horseradish peroxidase. © 1997 Elsevier Science B.V. Keywords: Hepatitis B virus: Hepatitis B e antigen: Monoclonal antibody
* Corresponding author. Tel.: + 81 285 4421 ll; fax: + 81 285 441557. 0166-0934/97/$17.00 ~© 1997 Elsevier Science B.V. All rights reserved. PII S0166-0934(97)00125-0
208
S. Usuda et al./Journal of Virological Methods' 68 (1997) 207-215
1. Introduction
2. Materials and methods
Two kinds of genomic m R N A of approximately 3.5 kb are known for the hepatitis B virus (HBV) which differ slightly in size (Will et al., 1987; Yaginuma et al., 19871). The shorter m R N A acts as the pregenome and codes for core protein and reverse-transcriptase/DNA polymerase. Longer transcripts called precore mRNA, have diverse transcript initiation sites, and can translate the precore region and read through the core gene to encode the hepatitis B e antigen (HBeAg). The precursor of HBeAg is made of 29 amino acids (aa) coded for by the precore region and 183 aa encoded by the core gene. It loses 19 aa at the N-terminus as well as 34 aa at the C-terminus, and makes the HBeAg polypeptide of 159 aa (p17 e) (Ou et al., 1986; Roossinck et al., 1986). Therefore, p17 ° contains 10 aa coded for by the precore region at the N-terminus followed by 149 aa encoded by the core gene. pl 7e circulates in two molecular entities, and both are associated with plasma proteins such as albumin and IgG (Takahashi et al., 1978). Persons infected with the wild-type HBV of an HBeAg-plus phenotype have HBeAg in serum initially, and seroconvert to antibody to HBeAg (anti-HBe) sooner or later. HBeAg or anti-HBe influences the replication of HBV and the severity of hepatitis in the host (Miyakawa and Mayumi, 1985), and therefore, many immunological methods have been developed to detect these markers. All the presently available methods, however, involve antigenic determinants borne by C-terminal 149 aa in the HBeAg polypeptide (p17 ~) that are encoded by the core gene. These HBeAg determinants are carried also by the core polypeptide of 183 aa (p21 ~) (Takahashi et al., 1983), which may be released from damaged hepatocytes. Hence, current HBeAg assay methods can create false positive results, especially when extensive liver damage occurs such as in patients with fulminant hepatitis. Thus, antibody reagents for detecting HBeAg determinant(s) borne by the N-terminal ten aa on p17 e are necessary for a specific detection of HBeAg. Monoclonal antibody to the ten aa in pl 7~ coded for by the precore region was raised and an enzyme-linked immunosorbent assay (ELISA) for HBeAg was developed.
2.1. Synthetic oligopeptides Amino acids in p17 e were numbered from - 10 to - 1 for those encoded by the precore region followed by 1-149 for those coded for by the core gene which were deduced from the nucleotide sequence of an HBV D N A clone (pNDR260) of subtype adr (Okamoto et al., 1988). A decapeptide representing aa from - 10 to - 1 with a sequence of S K L C L G W L W G (designated preC-10), a 17mer oligopeptide representing aa 1-17 with a sequence of M D I D P Y K E F G A S V E L L S (C1) and an icosapeptide spanning aa 126-145 in the C-terminus of p l 7 e with a sequence of IRTPPAYRPPN A P I L S T L P E (C126) were synthesized by the solid-phase method of Merrifield (1969). The peptides were used as antigens to raise monoclonal antibodies (mAbs). A total of 203 decapeptides, which covered the precursor of HBeAg (p25 e) consisting of aa from - 29 to - 1 coded for by the precore region and aa 1-183 by the core gene with a shift of one aa, were synthesized as multipin peptides by commercial kits (Chiron Mimotopes Pty, Clayton, Australia). The peptides were tested for binding with the mAb, for the purpose of mapping the epitope recognized by it.
2.2. mAb An mAb was raised against preC-10 conjugated with keyhole limpet hemocyanin by the method described elsewhere (Nomura et al., 1982). The mAb was screened by the binding with preC-10 immobilized on wells of an immunoplate (Greiner, Frickenhausen, Germany). Bound antibodies were detected by goat IgG antibody to mouse IgG/Fc labeled with horseradish peroxidase (Cappel Laboratories, West Chester, USA). Two other mAb (19C1-8 and C33) were raised against p21 c obtained by denaturing core particles recovered from circulating HBV, and screened by C1, Cl26 and p21 c fixed on a solid support. Two mAbs raised against HBeAg purified from plasma (904 and 905) have been described previously (Imai et al., 1982). mAb T2212 (Machida et
S. Usuda et al./Journal of Virological Methods 68 (1997) 207-215
al., 1991) has been raised against p21 ° derived from human hepatoma cells propagated in nude mice (PLC/342) (Matsui et al., 1986). It served as a negative control because it recognizes an extreme C-terminal sequence of p21 ° and does not bind with p l 7e. A guinea pig polyclonal antibody raised against a synthetic undecapeptide (ASKLCLGWLWG) spanning aa positions from - 11 to 1 (Takahashi et al., 1991) was used as a positive control, -
2.3. Western blotting To prepare pl7 e and p21 c, respectively, HBeAg purified from plasma (0.3 mg/ml) and core particles isolated from PLC/342 cells (0.01 mg/ml) were treated with 1% (wt/vol) sodium dodecyl sulfate (SDS) and 2.5% (vol/vol) 2-mercaptoethanol at 95°C for 5 rain. They were then run on a 14% polyacrylamide gel electrophoresis (PAGE) in the presence of SDS, transferred to nitrocellulose membrane, blocked with bovine serum albumin and tested for the binding with mAb. Detection was by goat IgG antibody to mouse IgG/Fc labeled with horseradish peroxidase.
2.4. Specificity of mAb The specificity of the mAb was evaluated by two distinct methods, mAb was labeled with horseradish peroxidase, and tested for binding with 203 overlapping decapeptides covering the HBeAg precursor by the multipin method. To avoid non-specific reactions, multipins with decapeptides were blocked with 25% calf serum in 0.15 M NaC1 before the binding with the labeled mAb. In the other method, three mAbs were tested for the binding with each of three synthetic oligopeptides spanning various regions of p17 ~ (preC-10, C1 and C126) at a concentration of 1 /lg/ml, as well as with p17 e itself (0.3 itg/ml), diluted serially (up to 1:3000) in sodium carbonate buffer (20 mM, pH 9.5) and coated onto wells in an immunoplate. Bound mAb was detected by colorimetry with goat IgG antibody to mouse IgG/Fc labeled with horseradish peroxidase. Contents of the wells with an absorbance at 492 nm
209
(A492) greater than 2.5 were diluted 5-fold with 1.3 N H : S O 4 and tested again, so as to induce A492 values less than 2.5 and the reading was multiplied by five. 2.5. Chemical modification of HBeAg For an alkali treatment, HBeAg purified from plasma and dissolved in 40 /~1 of 150 mM NaC1 (0.6 mg/ml) was added with 4/ll of 0.3% (vol/vol) Tween 20 and 200/~1 of 0.5 N NaOH and left at room temperature for 30 min. Then, it was neutralized with 200/ll of 0.6 M NaH2PO4. Likewise, an acid treatment of HBeAg was performed with 0.5 N HCI in the presence of Tween 20, followed by neutralization with 0.5 N NaOH in phosphate buffer (0.1 M, pH 7.2). HBeAg (0.6 mg/ml, 40 ~tl) was reduced with dithiothreitol (1 M, 1 /tl) at room temperature for 1 h, or alkylated with iodoacetamide (1 M, 4.5/ll) at 4°C for 1 h. The reaction was terminated by adding 400/tl of 150 mM NaC1 and 4/tl of 0.3% Tween 20. Reduction and alkylation of HBeAg was carried out with dithiothreitol followed by iodoacetamide, and the reaction was terminated as described above. Chemically treated HBeAg samples were diluted to 20 ~g/ml in Tris-HC1 buffer (20 raM, pH 7.5) supplemented with 15% (vol/vol) fetal calf serum, 0.1% (wt/vol) NaN3 and 150 mM NaC1. The mixture of untreated HBeAg (0.6 mg/ml, 40 /~1), NaC1 (150 raM, 400/tl) and Tween 20 (0.3%, 4 /tl) was diluted to 20/~g/ml in the same buffer and tested in parallel as a control.
2.6. An ELISA for HBeAg with YO583A Plasma samples obtained from blood donors in Japan were tested by an ELISA using an mAb raised against the precore sequence (Y0583A). They included 30 donors with hepatitis B surface antigen (HBsAg) and HBeAg, 15 with HBsAg and anti-HBe and 50 without serological markers of HBV infection. HBsAg was determined by reversed passive hemagglutination (MyCell, Institute of Immunology, Tokyo, Japan) and HBeAg and anti-HBe by ELISA (Immunis HBeAg/Ab EIA, Institute of Immunology, Tokyo, Japan).
210
S. Usuda et al. / Journal of Virological Methods 68 (1997) 207 215
Wells of an immunoplate received 100 yl of the m A b against HBeAg (904) at a concentration of 5 ¢tg/ml in phosphate buffer (10 raM, p H 7.5). The test plasma (50 /tl) was delivered to a well and agitated on a shaker ( I K A Labortechnik, Janke and Kunkel G m b H , Staufen, Germany) for 3 h at room temperature. The plate was washed with 150 m M NaC1 containing 0.05% Tween 20 three times and twice with 150 m M NaC1. Thereafter, each well received 50/~1 of 0.2 N N a O H and 10 Itl of 0.03% Tween 20, and the plate was shaken at 37°C for 30 min. Then the well was neutralized by adding 5 0 / l l of 0.24 M N a H 2 P Q . The reactant was transferred to a well in an immunoplate coated with m A b C33 (5 /tg/ml in phosphate buffer [10 mM, p H 7.5], 100 /tl per well). The plate was shaken for 90 rain at r o o m temperature and washed. Finally, each well received 50 itl of Tris-HC1 buffer (20 raM, p H 7.5) containing 2.5 Itg/ml of m A b Y0583A labeled with horseradish peroxidase and supplemented with 25% fetal calf serum, 0.025% Tween 20, 0.02% (wt/vol) thimerosal and 150 m M NaC1. The plate was agitated at room temperature for 90 min, then washed. The antibody bound to the wells was determined by colorimetry at A492.
3. Results
3. I. Mapping epitopes recognized by the mAb with multipin peptides covering the H B e A g precursor (p25 e) Fig. 1 illustrates the binding of an m A b (Y0583A), which was raised against the ten aa positioned from - 1 0 to - 1 in the HBeAg polypeptide (p17 e) and coded for by the precore region, to 203 decapeptides synthesized on multipins. They shifted by one aa and covered the HBeAg precursor (p25 e) made of 29 aa encoded by the precore region (positions from - 2 9 to - 1) and 183 aa by the core gene (from 1 to 183). Y0583A showed the strongest binding with a decapeptide spanning aa positions from - 6 to 4 with a sequence of L G W L W G M D I D (aa coded for by the precore region underlined). The sequence contained four aa encoded by the core
gene at the C-terminus. Since the four aa were not included in the decapeptide used as an immunogen (preC-10), they would not be involved in the epitope recognized by Y0583A. Therefore, the hydrophobic sequence of six aa coded for by the precore region ( L G W L W G ) would be the HBeAg epitope recognized by Y0583A. The epitopes recognized by two m A b s raised against the HBV core polypeptide (p21 c) are m a p p e d in Fig. 1. m A b 19C1-8 bound with a high avidity to seven decapeptides having start positions from - 4 to 3. Hence, the sequence of four aa (IDPY) at positions 3 - 6 shared by them would be the epitope recognized by 19C1 8. Another m A b raised against p21 c (C33) bound strongly to three decapeptides with start positions from 131 to 133. Therefore, C33 would be directed to the sequence of eight aa (RPPNAP1L) at
A) Y0583A 8-
E
t(N
O3
~J
c) z< m no o9 en <
6420 10. 86420
B) 19C1-8
10. C) C33 8. 6, 4. 2. 0 i -29
a'l
e'l
91
lal
lgl 174
START POSITIONS OF MULTIPIN DECAPEPTIDES
Fig. 1. Mapping of epitopes recognized by mAb with multipin peptides. Three monoclonal antibodies were tested for the binding with 203 decapeptides covering the precursor of HBeAg (p25e) made of 29 amino acids encoded by the precore region and 183 aa by the core gene. Numbers in the abscissa represent the start positions of 203 decapeptides. Amino acids encoded by the precore region were numbered from - 2 9 to - 1, and those by the core gene from 1 to 174.
S. Usuda et al./Journal (~f Virological Methods 68 (1997) 207-215
Y0583A
19C 1-8
211
C33
C
- p2_ e -pl a
b
a
b
a
b
Fig. 2. Western blotting of HBeAg polypeptide (p17 e) and HBV core polypeptide (p21 ~) by mAb. Polypeptides were run on SDS-PAGE (p17 e in lanes labeled a and p21 ~ in lanes labeled b), and tested for the binding with three mAbs.
positions 133 140 which overlapped among them. Since the HBeAg polypeptide (p17 ~) contains aa 1-149 of p21 ~, it would be recognized by both 19C1 8 and C33. As shown in Fig. 1, the two mAbs (19C1-8 and C33) raised against the core polypeptide (p21 c) bound more strongly to epitopes on multipins than the mAb against a precore decapeptide (Y0583A). Therefore, the epitope borne by the hydrophobic sequence ( L G W L W G [aa from - 6 to - 1 ] ) in the precore sequence would be less immunogenic than the two epitopes in p21 ° (IDPY [aa 3-6] and R P P N A P I L [aa 133-140]).
3.2. The specificity of mAb evaluated by Western blotting and binding with synthetic oligopeptides Y0583A bound with p17 e in Western blotting (Fig. 2). It did not, however, bind with p21 c that is devoid of the precore sequence. By contrast, both 19C1-8 and C33 bound with p17 e as well as with p2 lC. Fig. 3 shows the binding of mAb with three synthetic oligopeptides mimicking sequences of p17 e, as well as p17 e itself, fixed on a solid support. Y0583A, 19C1-8 and C33 bound, respectively, with preC-10 (made of aa from - l 0 to - 1), C1 (1-17) and C126 (126-145) in a manner dependent on the concentration of oligopeptides
used for coating wells of an immunoplate. However, they did not react with the two other peptides which did not contain the corresponding epitope. Three mAbs bound with p17 e to various degrees; the binding was strongest for 19C1-8 and weakest for Y0583A.
3.3. Effects of chemical modifications of HBeAg on the binding with mAb HBeAg was isolated from plasma to a homogeneity, and treated with alkali or acid. It was also reduced or alkylated, or reduced and then alkylated. Chemically modified HBeAg was captured by C33 immobilized on a solid support and tested for the capacity to bind with Y0583A or 19C1-8 labeled with horseradish peroxidase (Table 1). The binding with labeled Y0583A increased remarkably after the treatment of HBeAg with 0.4 N NaOH; it was not affected by acid treatment, reduction or alkylation, or reduction followed by alkylation. Likewise, the binding of captured HBeAg with labeled 19C1 8 increased by pretreatment with alkali, albeit not to the extent of labeled Y0583A. Labeled T2212 with a specificity for an extreme C-terminal sequence of p21 c did not bind to captured HBeAg, whether it had been untreated or modified chemically.
S. Usuda eta/./Journal o/" Virological Methods' 68 (1997) 207-215
212
3.4. A sandwich enzyme immunoassay Jor HBeAg with immobilized C33 and labeled YO583A An ELISA was developed for specific determination of HBeAg by sandwiching it between C33 immobilized on a solid support and Y0583A labeled with horseradish peroxidase. The specificity of assay, for detecting HBeAg purified from plasma and treated with NaOH, was confirmed by blocking tests in which the binding with labeled Y0583A was competed by increasing concentrations of unlabeled Y0583A (Fig. 4). Unlabeled Y0583A inhibited the binding of labeled Y0583A to HBeAg almost completely. Unlabeled T2212, recognizing a C-terminal sequence of p2V not included in p17 ~, did not inhibit the binding at all. B) C1
A) PreC- 10 • Y0583A 5 4 E3 c~2 O~
z
i
i
i
r
i
i
D) pl 7 e J
//¢'!
<5 4.
i
i
c) c126
,"
i
i
3~
2'
0
1
101
10 z
103 0
1
101
102
103
iMMOBILIZED PEPTIDES (ng/rnl)
Fig. 3. Binding of three mAbs with three synthetic oligopeptides representing sequences of HBeAg polypeptide (p17 e) as well as with p17 e. Oligopeptides (A-C) as well as p17 ~ (D) in increasing concentrations were coated onto wells of an immunoplate, mAb (10/ig/ml) was delivered to wells and bound antibodies were detected with goat anti-mouse lgG/Fc labeled with horseradish peroxidase.
The binding of labeled Y0583A to HBeAg was also blocked by guinea pig polyclonal antibodies, raised against undecapeptide representing aa positioned from - 1 1 to - 1 , in a dose-dependent fashion. Pre-immune serum from the same guinea pig did not inhibit binding.
3.5. Determination of HBeAg #1 plasma samples from blood donors by ELISA with YO583A When plasma samples from blood donors containing HBeAg or anti-HBe, or neither, were tested by ELISA with immobilized C33 and labeled Y0583A, they all showed high A492 readings ( > 0.5) even when they had been diluted 20-fold. Such a high non-specific reaction, however, was abolished when plasma samples had been pretreated with 0.1-0.5 N NaOH and neutralized. An ELISA for HBeAg was developed with mAb C33 fixed on a solid support and labeled Y0583A (see Methods). Prior to the assay, HBeAg in the test plasma (50/tl) was captured by mAb 904 raised against purified HBeAg (Imai et al., 1982) immobilized onto a well of an immunoplate and washed. Then, wells received 0.2 N NaOH and were neutralized. The concentration of 0.2 N was adopted because it induced the highest m e a n A492 values for ten randomly selected plasma samples containing HBeAg. Thereafter, the reactant was evaluated as a sandwich between immobilized C33 and labeled Y0583A. Since 50 plasma samples without HBV markers showed a mean+S.D. A492 value at 0.028 + 0.009, a cut-off limit was set at 0.055 representing the mean + 3 S.D. of negative controls. A total of 30 plasma samples, positive for HBsAg and HBeAg by a commercial ELISA kit, had a mean-6 S.D. A492 value of 1.747 __ 1.238 (range: 0.145-4.888). By contrast, 15 samples positive for HBsAg and anti-HBe gave that of 0.025 + 0.005 (range: 0.021-0.039); they were all negative for HBeAg. The A492 reading of ten randomly selected samples with HBeAg was inhibited invariably by unlabeled antibodies directed to the precore sequence. Thus, mAb Y0583A (30 ILg/ml) and guinea pig polyclonal antibody raised against aa from - 11 to - 1 (1:30 dilution), when present in the reaction with labeled Y0583A, decreased
S, Usuda et al./Journal of Virological Methods' 68 (1997) 207 215
213
Table 1 Effect of chemical modifications on the detection of HBeAg by ELISA with immobilized C33 and Y0583A labeled with horseradish peroxidasea Labeled mAb
Blank
Y0583A 19C1-8d T2212e
HBeAg pretreated with
0.021 0.008 0.015
None
NaOH (0.4 N)
HC1 (0.4 N)
DTTb (25 mM)
IAAc (100 mM)
DTT/IAA
0.144 2.012 0.029
1.272 3.765 0.013
0.215 4.305 0.013
0.155 0.813 0.017
0.106 1.714 0.027
0.203 1.978 0.031
a HBeAg purified from plasma (0.6 mg/ml) was subjected to various chemical modifications with conditions specified in Materials and Methods. It was diluted to 20 #g/ml and tested for the capacity to be sandwiched between immobilized C33 (directed to aa 133-140) and Y0583A (aa from - 6 to - 1 ) or other mAb labeled with horseradish peroxidase. Values indicate absorbance at 492 rim. b Dithiothreitol.
c Iodoacetamide. d mAb directed to aa 3 6 in p21 c. e mAb directed to the C-terminus of p21 c. A492 by 9 7 . 6 + 1.7% (range: 9 3 . 4 - 9 9 . 8 % ) an d 83.8 + 9.4% (61.6-95.7%), respectively. Fig. 5 illustrates the results o f H B e A g tests on 30 p l a s m a samples with H B s A g a n d H B e A g . T h e r e was a close c o r r e l a t i o n (r = 0,868) b e t w e e n
A492 readings 100"
o b t a i n e d by E L I S A with C33 an d
mAb
Y 0 5 8 3 A an d t h o se by c o m m e r c i a l E L I S A with m A b 904 a n d 905 ( I m m u n i s H B e A g / A b E I A , Institute o f I m m u n o l o g y ) . Test p l a s m a were diluted 1:50 an d a 50-ltl p o r t i o n was used for the assay with c o m m e r c i a l E L I S A , a n d an alkali-pret r e a t m e n t was n o t p e r f o r m e d before the test. Since the E L I S A with C33 a n d Y 0 5 8 3 A r e q u i r e d 5 0 / , 1 o f test plasma, it w o u l d be a b o u t 50 times less sensitive t h a n the c o m m e r c i a l E L I S A .
• Y0583A
z
80-
O_
o T2212 Guinea Pig Serum •
5-
Post
i
•
/
-&-
60-
CO
z
m 40O
~3
~I
""/'"
20-
0
•
•
I-- 2
,< 'J 1
_51
~(/o °
. . . . . . . . .
i
l laooo
u
1;aoo
i
l:ao
0
DILUTION C O M M E R C I A L ELISA (A492) Fig. 4. Competition of unlabeled mAb and guinea pig polyclonal antibodies against labeled Y0583A for the binding with HBeAg. mAb (1 mg/ml) as well as preimmune and postimmune guinea pig sera were diluted and tested for the binding with HBeAg pretreated with NaOH and captured by C33 on a solid support.
Fig. 5. Correlation of A492readings obtained with ELISA with Y0583A and those with commercial ELISA (Immunis HBeAg/ Ab EIA, Institute of Immunology, Tokyo, Japan) with 904/ 905 in the detection of HBeAg in 30 plasma samples from blood donors carrying HBsAg.
214
S. Usuda et al. /Journal of ~qrological Methods 68 (1997) 207 215
4. Discussion
A monoclonal antibody (Y0583A) was raised against the N-terminal l0 aa in the HBeAg polypeptide (p17 e) which are encoded by the precore region. By epitope mapping with 203 decapeptides on multipins, Y0583A was deduced to be directed to a sequence of six aa (LGWLWG) representing a C-terminus of the precore region product. Y0583A bound with p17 e, but not with p21 c representing the product of the HBV core gene. Furthermore, it bound with a synthetic oligopeptide including the six aa but not with the mimicking sequence of the core-gene product in pl7L In addition, the binding of Y0583A with HBeAg, purified from plasma and pretreated with alkali, was blocked by unlabeled Y0583A as well as guinea pig polyclonal antibody raised against a synthetic oligopeptide containing the L G W L W G sequence, but not by mAb directed to a C-terminal sequences of p21 ° (T2212). Taken together, Y0583A would specifically recognize the epitope borne by the hexapeptide in the precore region. The binding of Y0583A with HBeAg purified from plasma was not strong. The association of HBeAg with plasma proteins such as albumin and IgG (Takahashi et al., 1978) may bury the epitope recognized by Y0583A in HBeAg. The pretreatment of HBeAg with alkali would be able to dissociate some plasma proteins for an easier access of Y0583A to the corresponding epitope on HBeAg. Many methods have been developed for the detection of HBeAg, such as two-dimensional immunodiffusion, passive hemagglutination, radioimmunoassay and ELISA (Magnius and Espmark, 1972; Takahashi et al., 1977; Miyakawa et al., 1979; Imai et al., 1982). However, all antiHBe reagents used are directed to the sequences in HBeAg that are encoded by the core gene. Commercial HBeAg tests are found to have unsatisfactory specificity (Westmoreland et al., 1994), which would be attributable to antibody reagents not directed to the precore sequence in HBeAg. To obtain specific antibodies to HBeAg, guinea pig and rabbit have been immunized with synthetic oligopeptides covering the ten aa in HBeAg encoded by the precore region (Takahashi et al., 1991; Dienes et al., 1995).
There are a number of clinical applications for HBeAg in the circulation. It is used as a marker for the response to interferon in patients with chronic hepatitis B (Tine et al., 1993; Wong et al., 1993). The lack of HBeAg in the patients infected with HBeAg-minus mutants is implicated in severe disease in the patients with HBeAg-negative or anti-HBe-positive chronic hepatitis (Bonino et al., 1986; Akahane et al., 1990) and those with fuhninant hepatitis B (Kosaka et al., 199l; Sato et al., 1995). A specific detection of HBeAg would be prerequisite to the molecular biological analysis of HBeAg-negative infections. An ELISA for HBeAg was developed by placing it between mAb (C33) to the core-gene sequence fixed on a solid support and Y0583A labeled with horseradish peroxidase. To increase the sensitivity and to avoid nonspecific reactions, HBeAg in the test plasma was captured by 904 on the solid support and treated with 0.2 N NaOH prior to ELISA with C33 and Y0583A. The ELISA detected HBeAg in 50/~1 of plasma samples from blood donors with A492 readings in a close correlation with those by commercial ELISA for HBeAg (lmmunis HBeAg/Ab EIA, Institute of Immunology, Tokyo, Japan). Although the sensitivity of Y0583A EL1SA was less than that of the commercial kits, it would be increased by using more volume of test samples. It is to be hoped that specific detection of HBeAg by ELISA involving Y0583A would contribute to better application and understanding of HBeAg in clinical and epidemiological settings. Furthermore, it will be useful in defining HBV variants with an HBeAg-minus phenotype and those with decreased expression of HBeAg, which are induced by mutations in the precore region and core promoter, respectively (Miyakawa et al., 1997).
References
Akahane, Y., Yamanaka, T., Suzuki, H., Sugai, Y., Tsuda, F., Yotsumoto, S., Omi, S., Okamoto, H.0 Miyakawa, Y., Mayumi, M., 1990. Chronic active hepatitis with hepatitis B virus DNA and antibody against e antigen in the serum. Disturbed synthesis and secretion of e antigen from hepatocytes due to a point mutation in the precore region. Gastroenterology99, 1113-1119.
S. Usuda et al. /Journal of Virological Methods 68 (1997) 207 215
Bonino, F., Rosina, F., Rizzetto, M., Rizzi, R., Chiaberge, E., Tardanico, R., Callea, F., Verme, G., 1986. Chronic hepatitis in HBsAg carriers with serum HBV-DNA and antiHBe. Gastroenterology 90, 1268-1273. Dienes, H.P., Gerken, G., Goergen, B., Heermann, K., Gerlich, W., Meyer zum Buschenfelde, K.H., 1995. Analysis of the precore DNA sequence and detection of precore antigen in liver specimens from patients with anti-hepatitis B e-positive chronic hepatitis. Hepatology 21, 1-7. Imai, M., Nomura, M., Gotanda, T., Sano, T., Tachibana, K., Miyamoto, H., Takahashi, K., Toyama, S., Miyakawa, Y., Mayumi, M., 1982. Demonstration of two distinct antigenic determinants on hepatitis B e antigen by monoclonal antibodies. J. Immunol. 128, 69-72. Kosaka, Y., Takase, K., Kojima, M., Shimizu, M., lnoue, K., Yoshiba, M., Tanaka, S., Akahane, Y., Okamoto, H., Tsuda, F., Miyakawa, Y, Mayumi, M., 1991. Fulminant hepatitis B: induction by hepatitis B virus mutants defective in the precore region and incapable of encoding e antigen. Gastroenterology 100, 1087-1094. Machida, A., Ohnuma, H., Tsuda, F, Yoshikawa, A., Hoshi, Y., Tanaka, T., Kishimoto, S., Akahane, Y., Miyakawa, Y., Mayumi, M., 1991. Phosphorylation in the carboxylterminal domain of the capsid protein of hepatitis B virus: evaluation with a monoclonal antibody. J. Virol. 65, 60246030. Magnius, L.O., Espmark, J.A., 1972. New specificities in Australia antigen positive sera distinct from the Le Bouvier determinants. J. Immunol. 109, 1017 1021. Matsui, T., Takano, M., Miyamoto, K., Itoh, Y., Yoshizawa, H., Koike, M., Mochizuki, T., Tanaka, E., Okamoto, H., lmai, M., Mishiro, S., Miyakawa, Y., Mayumi, M., 1986. Nude mice bearing human primary hepatocellular carcinoma that produces hepatitis B surface, core, and e antigens, as well as deoxyribonucleic acid polymerase. Gastroenterology 90, 135-142. Merrifield, R.B., 1969. Solid-phase peptide synthesis.. Adv. Enzymol. 32, 221-296. Miyakawa, Y., Mayumi, M., 1985. Hepatitis B e antigen and antibody (HBeAg/Anti-HBe). In: R.J. Gerety (Ed.), Hepatitis B, Academic Press, Orlando, pp. 47-76. Miyakawa, Y., Okamoto, H., Mayumi, M., 1997. The molecular basis of hepatitis B e antigen (HBeAg)- negative infections. J. Viral Hepatitis 4, 1 8. Miyakawa, Y., Tsuda, F., Akahane, Y., Mayumi, M., 1979. Application of microtiter solid-phase radioimmunoassay to the determination of hepatitis B e antigen. Vox Sang. 36, 307-311. Nomura, M., Imai, M., Nakamura, T., Miyakawa, Y., Mayumi, M., 1982. Demonstration of two distinct antigenic determinants on human e-foetoprotein by monoclonal antibodies. Mol. Immunol. 19, 1691-1697. Okamoto, H., Tsuda, F., Sakugawa, H., Sastrosoewignjo, R.I., Imai, M., Miyakawa, Y., Mayumi, M., 1988. Typing
215
hepatitis B virus by homology in nucleotide sequence: comparison of surface antigen subtypes. J. Gen. Virol. 69, 2575-2583. Ou, J.H., Laub, O., Rutter, W.J., 1986. Hepatitis B virus gene function: the precore region targets the core antigen to cellular membranes and causes the secretion of the e antigen. Proc. Natl. Acad. Sci. U.S.A. 83, 1578 1582. Roossinck, M.J., Jameel, S., Loukin, S.H, Siddiqui, A., 1986. Expression of hepatitis B viral core region in mammalian cells. Mol. Cell. Biol. 6, 1393 1400. Sato, S., Suzuki, K., Akahane, Y., Akamatsu, K., Akiyama, K., Yunomura, K., Tsuda, F., Tanaka, T., Okamoto, H., Miyakawa, Y., Mayumi, Y., 1995. Hepatitis B virus strains with mutations in the core promoter in patients with fulminant hepatitis. Ann. Intern. Med. 122, 241-248. Takahashi, K., Fukuda, M., Baba, K., Imai, M., Miyakawa, Y., Mayumi, M., 1977. Determination of e antigen and antibody to e by means of passive hemagglutination method. J. Immunol. 119, 1556-1559. Takahashi, K., Imai, M., Miyakawa, Y., Iwakiri, S., 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. Natl. Acad. Sci. U.S.A. 75, 1952-1956. Takahashi, K., Kishimoto, S., Ohori, K., Yoshizawa, H., Machida, A., Ohnuma, H., Tsuda, F., Munekata, E., Miyakawa, Y., Mayumi, M., 1991. Molecular heterogeneity of e antigen polypeptides in sera from carriers of hepatitis B virus. J. Immunol. 147, 3156-3160. Takahashi, K., Machida, A., Funatsu, G., Nomura, M., Usuda, S., Aoyagi, S., Tachibana, K., Miyamoto, H., Imai, M., Nakamura, T., Miyakawa, Y., Mayumi, M., 1983. Immunochemical structure of hepatitis B e antigen in the serum. J. Immunol. 130, 2903-2907. Tine, F., Liberati, A., Craxi, A., Almasio, P., Pagliaro, L., 1993. Interferon treatment in patients with chronic hepatitis B: a meta-analysis of the published literature. J. Hepatol. 18, 154-162. Westmoreland, D., Parr. A., Thomas, A.J., 1994. Assessment of hepatitis B 'e' antigen and 'e' antibody status-tests of status or status of tests?. J. Med. Microbiol. 41, 374-377. Will, H., Reiser, W., Weimer, T., Pfaff, E., Buscher, M., Sprengel, R., Cattaneo, R., Schaller, H., 1987. Replication strategy of human hepatitis B virus. J. Virol. 61,904 911. Wong, D.K., Cheung, A.M., O'Rourke, K., Naylor, C.D., Detsky, A.S., Heathcote, J., 1993. Effect of alpha-interferon treatment in patients with hepatitis B e antigen-positive chronic hepatitis B: a meta-analysis. Ann. Intern. Med. 119, 312-323. Yaginuma, K., Shirakata, Y., Kobayashi, M., Koike, K., 1987. Hepatitis B virus (HBV) particles are produced in a cell culture system by transient expression of transfected HBV DNA. Proc. Natl. Acad. Sci. U.S.A. 84, 2678-2682.
Journal of Virological Methods Journal of Virological Methods 68 (1997) 207 215
A monoclonal antibody against a hepatitis B e antigen epitope borne by six amino acids encoded by the precore region Sadakazu Usuda a Hiroaki Okamoto a Hitoshi Ohnuma b Takeshi Tanaka Atsuhiko Machida c Yuzo Miyakawa d, Makoto Mayumi a,,
b
Immunology Division, Jichi Medical School, Tochigi-Ken 329-04, Japan u Japanese Red Cross Saitama Blood Center, Saitama-Ken 338, Japan c Yamanashi Institute fi)r Public Health, Yarnanashi-Ken 400, Japan d M~vakawa Memorial Research Foundation, Tokyo 107, Japan
Received 4 April 1997; received in revised form 21 July 1997: accepted 27 July 1997
Abstract
Hepatitis B e antigen (HBeAg) polypeptide in the circulation (p17 e) is composed of ten amino acids (aa) coded for by the precore region and 149 aa by the core gene of hepatitis B virus. A monoclonal antibody (Y0583A) was raised against the N-terminal ten amino acids (SKLCLGWLWG) encoded by the precore region. The binding of Y0583A with a panel of 203 decapeptides on multipins, which covered the precursor of HBeAg polypeptide made of 212 aa shifting by one aa, recognized an epitope sequenced L G W L W G representing the C-terminal six aa coded for by the precore region. This HBeAg epitope was not readily accessible on HBeAg in serum, but it became exposed and bound with Y0583A by treatment with 0.2 N NaOH. Using Y0583A, an enzyme-linked immunosorbent assay was developed for specific determination of HBeAg. The test sample was incubated with the monoclonal antibody to an HBeAg determinant encoded by the core gene (904) that had been immobilized on a solid support. Captured HBeAg was treated with 0.2 N NaOH, neutralized and released into the fluid phase. The reactant was then tested for a sandwich between monoclonal antibody (C33) to the C-terminus of the HBeAg polypeptide immobilized on a solid support and Y0583A labeled with horseradish peroxidase. © 1997 Elsevier Science B.V. Keywords: Hepatitis B virus: Hepatitis B e antigen: Monoclonal antibody
* Corresponding author. Tel.: + 81 285 4421 ll; fax: + 81 285 441557. 0166-0934/97/$17.00 ~© 1997 Elsevier Science B.V. All rights reserved. PII S0166-0934(97)00125-0
208
S. Usuda et al./Journal of Virological Methods' 68 (1997) 207-215
1. Introduction
2. Materials and methods
Two kinds of genomic m R N A of approximately 3.5 kb are known for the hepatitis B virus (HBV) which differ slightly in size (Will et al., 1987; Yaginuma et al., 19871). The shorter m R N A acts as the pregenome and codes for core protein and reverse-transcriptase/DNA polymerase. Longer transcripts called precore mRNA, have diverse transcript initiation sites, and can translate the precore region and read through the core gene to encode the hepatitis B e antigen (HBeAg). The precursor of HBeAg is made of 29 amino acids (aa) coded for by the precore region and 183 aa encoded by the core gene. It loses 19 aa at the N-terminus as well as 34 aa at the C-terminus, and makes the HBeAg polypeptide of 159 aa (p17 e) (Ou et al., 1986; Roossinck et al., 1986). Therefore, p17 ° contains 10 aa coded for by the precore region at the N-terminus followed by 149 aa encoded by the core gene. pl 7e circulates in two molecular entities, and both are associated with plasma proteins such as albumin and IgG (Takahashi et al., 1978). Persons infected with the wild-type HBV of an HBeAg-plus phenotype have HBeAg in serum initially, and seroconvert to antibody to HBeAg (anti-HBe) sooner or later. HBeAg or anti-HBe influences the replication of HBV and the severity of hepatitis in the host (Miyakawa and Mayumi, 1985), and therefore, many immunological methods have been developed to detect these markers. All the presently available methods, however, involve antigenic determinants borne by C-terminal 149 aa in the HBeAg polypeptide (p17 ~) that are encoded by the core gene. These HBeAg determinants are carried also by the core polypeptide of 183 aa (p21 ~) (Takahashi et al., 1983), which may be released from damaged hepatocytes. Hence, current HBeAg assay methods can create false positive results, especially when extensive liver damage occurs such as in patients with fulminant hepatitis. Thus, antibody reagents for detecting HBeAg determinant(s) borne by the N-terminal ten aa on p17 e are necessary for a specific detection of HBeAg. Monoclonal antibody to the ten aa in pl 7~ coded for by the precore region was raised and an enzyme-linked immunosorbent assay (ELISA) for HBeAg was developed.
2.1. Synthetic oligopeptides Amino acids in p17 e were numbered from - 10 to - 1 for those encoded by the precore region followed by 1-149 for those coded for by the core gene which were deduced from the nucleotide sequence of an HBV D N A clone (pNDR260) of subtype adr (Okamoto et al., 1988). A decapeptide representing aa from - 10 to - 1 with a sequence of S K L C L G W L W G (designated preC-10), a 17mer oligopeptide representing aa 1-17 with a sequence of M D I D P Y K E F G A S V E L L S (C1) and an icosapeptide spanning aa 126-145 in the C-terminus of p l 7 e with a sequence of IRTPPAYRPPN A P I L S T L P E (C126) were synthesized by the solid-phase method of Merrifield (1969). The peptides were used as antigens to raise monoclonal antibodies (mAbs). A total of 203 decapeptides, which covered the precursor of HBeAg (p25 e) consisting of aa from - 29 to - 1 coded for by the precore region and aa 1-183 by the core gene with a shift of one aa, were synthesized as multipin peptides by commercial kits (Chiron Mimotopes Pty, Clayton, Australia). The peptides were tested for binding with the mAb, for the purpose of mapping the epitope recognized by it.
2.2. mAb An mAb was raised against preC-10 conjugated with keyhole limpet hemocyanin by the method described elsewhere (Nomura et al., 1982). The mAb was screened by the binding with preC-10 immobilized on wells of an immunoplate (Greiner, Frickenhausen, Germany). Bound antibodies were detected by goat IgG antibody to mouse IgG/Fc labeled with horseradish peroxidase (Cappel Laboratories, West Chester, USA). Two other mAb (19C1-8 and C33) were raised against p21 c obtained by denaturing core particles recovered from circulating HBV, and screened by C1, Cl26 and p21 c fixed on a solid support. Two mAbs raised against HBeAg purified from plasma (904 and 905) have been described previously (Imai et al., 1982). mAb T2212 (Machida et
S. Usuda et al./Journal of Virological Methods 68 (1997) 207-215
al., 1991) has been raised against p21 ° derived from human hepatoma cells propagated in nude mice (PLC/342) (Matsui et al., 1986). It served as a negative control because it recognizes an extreme C-terminal sequence of p21 ° and does not bind with p l 7e. A guinea pig polyclonal antibody raised against a synthetic undecapeptide (ASKLCLGWLWG) spanning aa positions from - 11 to 1 (Takahashi et al., 1991) was used as a positive control, -
2.3. Western blotting To prepare pl7 e and p21 c, respectively, HBeAg purified from plasma (0.3 mg/ml) and core particles isolated from PLC/342 cells (0.01 mg/ml) were treated with 1% (wt/vol) sodium dodecyl sulfate (SDS) and 2.5% (vol/vol) 2-mercaptoethanol at 95°C for 5 rain. They were then run on a 14% polyacrylamide gel electrophoresis (PAGE) in the presence of SDS, transferred to nitrocellulose membrane, blocked with bovine serum albumin and tested for the binding with mAb. Detection was by goat IgG antibody to mouse IgG/Fc labeled with horseradish peroxidase.
2.4. Specificity of mAb The specificity of the mAb was evaluated by two distinct methods, mAb was labeled with horseradish peroxidase, and tested for binding with 203 overlapping decapeptides covering the HBeAg precursor by the multipin method. To avoid non-specific reactions, multipins with decapeptides were blocked with 25% calf serum in 0.15 M NaC1 before the binding with the labeled mAb. In the other method, three mAbs were tested for the binding with each of three synthetic oligopeptides spanning various regions of p17 ~ (preC-10, C1 and C126) at a concentration of 1 /lg/ml, as well as with p17 e itself (0.3 itg/ml), diluted serially (up to 1:3000) in sodium carbonate buffer (20 mM, pH 9.5) and coated onto wells in an immunoplate. Bound mAb was detected by colorimetry with goat IgG antibody to mouse IgG/Fc labeled with horseradish peroxidase. Contents of the wells with an absorbance at 492 nm
209
(A492) greater than 2.5 were diluted 5-fold with 1.3 N H : S O 4 and tested again, so as to induce A492 values less than 2.5 and the reading was multiplied by five. 2.5. Chemical modification of HBeAg For an alkali treatment, HBeAg purified from plasma and dissolved in 40 /~1 of 150 mM NaC1 (0.6 mg/ml) was added with 4/ll of 0.3% (vol/vol) Tween 20 and 200/~1 of 0.5 N NaOH and left at room temperature for 30 min. Then, it was neutralized with 200/ll of 0.6 M NaH2PO4. Likewise, an acid treatment of HBeAg was performed with 0.5 N HCI in the presence of Tween 20, followed by neutralization with 0.5 N NaOH in phosphate buffer (0.1 M, pH 7.2). HBeAg (0.6 mg/ml, 40 ~tl) was reduced with dithiothreitol (1 M, 1 /tl) at room temperature for 1 h, or alkylated with iodoacetamide (1 M, 4.5/ll) at 4°C for 1 h. The reaction was terminated by adding 400/tl of 150 mM NaC1 and 4/tl of 0.3% Tween 20. Reduction and alkylation of HBeAg was carried out with dithiothreitol followed by iodoacetamide, and the reaction was terminated as described above. Chemically treated HBeAg samples were diluted to 20 ~g/ml in Tris-HC1 buffer (20 raM, pH 7.5) supplemented with 15% (vol/vol) fetal calf serum, 0.1% (wt/vol) NaN3 and 150 mM NaC1. The mixture of untreated HBeAg (0.6 mg/ml, 40 /~1), NaC1 (150 raM, 400/tl) and Tween 20 (0.3%, 4 /tl) was diluted to 20/~g/ml in the same buffer and tested in parallel as a control.
2.6. An ELISA for HBeAg with YO583A Plasma samples obtained from blood donors in Japan were tested by an ELISA using an mAb raised against the precore sequence (Y0583A). They included 30 donors with hepatitis B surface antigen (HBsAg) and HBeAg, 15 with HBsAg and anti-HBe and 50 without serological markers of HBV infection. HBsAg was determined by reversed passive hemagglutination (MyCell, Institute of Immunology, Tokyo, Japan) and HBeAg and anti-HBe by ELISA (Immunis HBeAg/Ab EIA, Institute of Immunology, Tokyo, Japan).
210
S. Usuda et al. / Journal of Virological Methods 68 (1997) 207 215
Wells of an immunoplate received 100 yl of the m A b against HBeAg (904) at a concentration of 5 ¢tg/ml in phosphate buffer (10 raM, p H 7.5). The test plasma (50 /tl) was delivered to a well and agitated on a shaker ( I K A Labortechnik, Janke and Kunkel G m b H , Staufen, Germany) for 3 h at room temperature. The plate was washed with 150 m M NaC1 containing 0.05% Tween 20 three times and twice with 150 m M NaC1. Thereafter, each well received 50/~1 of 0.2 N N a O H and 10 Itl of 0.03% Tween 20, and the plate was shaken at 37°C for 30 min. Then the well was neutralized by adding 5 0 / l l of 0.24 M N a H 2 P Q . The reactant was transferred to a well in an immunoplate coated with m A b C33 (5 /tg/ml in phosphate buffer [10 mM, p H 7.5], 100 /tl per well). The plate was shaken for 90 rain at r o o m temperature and washed. Finally, each well received 50 itl of Tris-HC1 buffer (20 raM, p H 7.5) containing 2.5 Itg/ml of m A b Y0583A labeled with horseradish peroxidase and supplemented with 25% fetal calf serum, 0.025% Tween 20, 0.02% (wt/vol) thimerosal and 150 m M NaC1. The plate was agitated at room temperature for 90 min, then washed. The antibody bound to the wells was determined by colorimetry at A492.
3. Results
3. I. Mapping epitopes recognized by the mAb with multipin peptides covering the H B e A g precursor (p25 e) Fig. 1 illustrates the binding of an m A b (Y0583A), which was raised against the ten aa positioned from - 1 0 to - 1 in the HBeAg polypeptide (p17 e) and coded for by the precore region, to 203 decapeptides synthesized on multipins. They shifted by one aa and covered the HBeAg precursor (p25 e) made of 29 aa encoded by the precore region (positions from - 2 9 to - 1) and 183 aa by the core gene (from 1 to 183). Y0583A showed the strongest binding with a decapeptide spanning aa positions from - 6 to 4 with a sequence of L G W L W G M D I D (aa coded for by the precore region underlined). The sequence contained four aa encoded by the core
gene at the C-terminus. Since the four aa were not included in the decapeptide used as an immunogen (preC-10), they would not be involved in the epitope recognized by Y0583A. Therefore, the hydrophobic sequence of six aa coded for by the precore region ( L G W L W G ) would be the HBeAg epitope recognized by Y0583A. The epitopes recognized by two m A b s raised against the HBV core polypeptide (p21 c) are m a p p e d in Fig. 1. m A b 19C1-8 bound with a high avidity to seven decapeptides having start positions from - 4 to 3. Hence, the sequence of four aa (IDPY) at positions 3 - 6 shared by them would be the epitope recognized by 19C1 8. Another m A b raised against p21 c (C33) bound strongly to three decapeptides with start positions from 131 to 133. Therefore, C33 would be directed to the sequence of eight aa (RPPNAP1L) at
A) Y0583A 8-
E
t(N
O3
~J
c) z< m no o9 en <
6420 10. 86420
B) 19C1-8
10. C) C33 8. 6, 4. 2. 0 i -29
a'l
e'l
91
lal
lgl 174
START POSITIONS OF MULTIPIN DECAPEPTIDES
Fig. 1. Mapping of epitopes recognized by mAb with multipin peptides. Three monoclonal antibodies were tested for the binding with 203 decapeptides covering the precursor of HBeAg (p25e) made of 29 amino acids encoded by the precore region and 183 aa by the core gene. Numbers in the abscissa represent the start positions of 203 decapeptides. Amino acids encoded by the precore region were numbered from - 2 9 to - 1, and those by the core gene from 1 to 174.
S. Usuda et al./Journal (~f Virological Methods 68 (1997) 207-215
Y0583A
19C 1-8
211
C33
C
- p2_ e -pl a
b
a
b
a
b
Fig. 2. Western blotting of HBeAg polypeptide (p17 e) and HBV core polypeptide (p21 ~) by mAb. Polypeptides were run on SDS-PAGE (p17 e in lanes labeled a and p21 ~ in lanes labeled b), and tested for the binding with three mAbs.
positions 133 140 which overlapped among them. Since the HBeAg polypeptide (p17 ~) contains aa 1-149 of p21 ~, it would be recognized by both 19C1 8 and C33. As shown in Fig. 1, the two mAbs (19C1-8 and C33) raised against the core polypeptide (p21 c) bound more strongly to epitopes on multipins than the mAb against a precore decapeptide (Y0583A). Therefore, the epitope borne by the hydrophobic sequence ( L G W L W G [aa from - 6 to - 1 ] ) in the precore sequence would be less immunogenic than the two epitopes in p21 ° (IDPY [aa 3-6] and R P P N A P I L [aa 133-140]).
3.2. The specificity of mAb evaluated by Western blotting and binding with synthetic oligopeptides Y0583A bound with p17 e in Western blotting (Fig. 2). It did not, however, bind with p21 c that is devoid of the precore sequence. By contrast, both 19C1-8 and C33 bound with p17 e as well as with p2 lC. Fig. 3 shows the binding of mAb with three synthetic oligopeptides mimicking sequences of p17 e, as well as p17 e itself, fixed on a solid support. Y0583A, 19C1-8 and C33 bound, respectively, with preC-10 (made of aa from - l 0 to - 1), C1 (1-17) and C126 (126-145) in a manner dependent on the concentration of oligopeptides
used for coating wells of an immunoplate. However, they did not react with the two other peptides which did not contain the corresponding epitope. Three mAbs bound with p17 e to various degrees; the binding was strongest for 19C1-8 and weakest for Y0583A.
3.3. Effects of chemical modifications of HBeAg on the binding with mAb HBeAg was isolated from plasma to a homogeneity, and treated with alkali or acid. It was also reduced or alkylated, or reduced and then alkylated. Chemically modified HBeAg was captured by C33 immobilized on a solid support and tested for the capacity to bind with Y0583A or 19C1-8 labeled with horseradish peroxidase (Table 1). The binding with labeled Y0583A increased remarkably after the treatment of HBeAg with 0.4 N NaOH; it was not affected by acid treatment, reduction or alkylation, or reduction followed by alkylation. Likewise, the binding of captured HBeAg with labeled 19C1 8 increased by pretreatment with alkali, albeit not to the extent of labeled Y0583A. Labeled T2212 with a specificity for an extreme C-terminal sequence of p21 c did not bind to captured HBeAg, whether it had been untreated or modified chemically.
S. Usuda eta/./Journal o/" Virological Methods' 68 (1997) 207-215
212
3.4. A sandwich enzyme immunoassay Jor HBeAg with immobilized C33 and labeled YO583A An ELISA was developed for specific determination of HBeAg by sandwiching it between C33 immobilized on a solid support and Y0583A labeled with horseradish peroxidase. The specificity of assay, for detecting HBeAg purified from plasma and treated with NaOH, was confirmed by blocking tests in which the binding with labeled Y0583A was competed by increasing concentrations of unlabeled Y0583A (Fig. 4). Unlabeled Y0583A inhibited the binding of labeled Y0583A to HBeAg almost completely. Unlabeled T2212, recognizing a C-terminal sequence of p2V not included in p17 ~, did not inhibit the binding at all. B) C1
A) PreC- 10 • Y0583A 5 4 E3 c~2 O~
z
i
i
i
r
i
i
D) pl 7 e J
//¢'!
<5 4.
i
i
c) c126
,"
i
i
3~
2'
0
1
101
10 z
103 0
1
101
102
103
iMMOBILIZED PEPTIDES (ng/rnl)
Fig. 3. Binding of three mAbs with three synthetic oligopeptides representing sequences of HBeAg polypeptide (p17 e) as well as with p17 e. Oligopeptides (A-C) as well as p17 ~ (D) in increasing concentrations were coated onto wells of an immunoplate, mAb (10/ig/ml) was delivered to wells and bound antibodies were detected with goat anti-mouse lgG/Fc labeled with horseradish peroxidase.
The binding of labeled Y0583A to HBeAg was also blocked by guinea pig polyclonal antibodies, raised against undecapeptide representing aa positioned from - 1 1 to - 1 , in a dose-dependent fashion. Pre-immune serum from the same guinea pig did not inhibit binding.
3.5. Determination of HBeAg #1 plasma samples from blood donors by ELISA with YO583A When plasma samples from blood donors containing HBeAg or anti-HBe, or neither, were tested by ELISA with immobilized C33 and labeled Y0583A, they all showed high A492 readings ( > 0.5) even when they had been diluted 20-fold. Such a high non-specific reaction, however, was abolished when plasma samples had been pretreated with 0.1-0.5 N NaOH and neutralized. An ELISA for HBeAg was developed with mAb C33 fixed on a solid support and labeled Y0583A (see Methods). Prior to the assay, HBeAg in the test plasma (50/tl) was captured by mAb 904 raised against purified HBeAg (Imai et al., 1982) immobilized onto a well of an immunoplate and washed. Then, wells received 0.2 N NaOH and were neutralized. The concentration of 0.2 N was adopted because it induced the highest m e a n A492 values for ten randomly selected plasma samples containing HBeAg. Thereafter, the reactant was evaluated as a sandwich between immobilized C33 and labeled Y0583A. Since 50 plasma samples without HBV markers showed a mean+S.D. A492 value at 0.028 + 0.009, a cut-off limit was set at 0.055 representing the mean + 3 S.D. of negative controls. A total of 30 plasma samples, positive for HBsAg and HBeAg by a commercial ELISA kit, had a mean-6 S.D. A492 value of 1.747 __ 1.238 (range: 0.145-4.888). By contrast, 15 samples positive for HBsAg and anti-HBe gave that of 0.025 + 0.005 (range: 0.021-0.039); they were all negative for HBeAg. The A492 reading of ten randomly selected samples with HBeAg was inhibited invariably by unlabeled antibodies directed to the precore sequence. Thus, mAb Y0583A (30 ILg/ml) and guinea pig polyclonal antibody raised against aa from - 11 to - 1 (1:30 dilution), when present in the reaction with labeled Y0583A, decreased
S, Usuda et al./Journal of Virological Methods' 68 (1997) 207 215
213
Table 1 Effect of chemical modifications on the detection of HBeAg by ELISA with immobilized C33 and Y0583A labeled with horseradish peroxidasea Labeled mAb
Blank
Y0583A 19C1-8d T2212e
HBeAg pretreated with
0.021 0.008 0.015
None
NaOH (0.4 N)
HC1 (0.4 N)
DTTb (25 mM)
IAAc (100 mM)
DTT/IAA
0.144 2.012 0.029
1.272 3.765 0.013
0.215 4.305 0.013
0.155 0.813 0.017
0.106 1.714 0.027
0.203 1.978 0.031
a HBeAg purified from plasma (0.6 mg/ml) was subjected to various chemical modifications with conditions specified in Materials and Methods. It was diluted to 20 #g/ml and tested for the capacity to be sandwiched between immobilized C33 (directed to aa 133-140) and Y0583A (aa from - 6 to - 1 ) or other mAb labeled with horseradish peroxidase. Values indicate absorbance at 492 rim. b Dithiothreitol.
c Iodoacetamide. d mAb directed to aa 3 6 in p21 c. e mAb directed to the C-terminus of p21 c. A492 by 9 7 . 6 + 1.7% (range: 9 3 . 4 - 9 9 . 8 % ) an d 83.8 + 9.4% (61.6-95.7%), respectively. Fig. 5 illustrates the results o f H B e A g tests on 30 p l a s m a samples with H B s A g a n d H B e A g . T h e r e was a close c o r r e l a t i o n (r = 0,868) b e t w e e n
A492 readings 100"
o b t a i n e d by E L I S A with C33 an d
mAb
Y 0 5 8 3 A an d t h o se by c o m m e r c i a l E L I S A with m A b 904 a n d 905 ( I m m u n i s H B e A g / A b E I A , Institute o f I m m u n o l o g y ) . Test p l a s m a were diluted 1:50 an d a 50-ltl p o r t i o n was used for the assay with c o m m e r c i a l E L I S A , a n d an alkali-pret r e a t m e n t was n o t p e r f o r m e d before the test. Since the E L I S A with C33 a n d Y 0 5 8 3 A r e q u i r e d 5 0 / , 1 o f test plasma, it w o u l d be a b o u t 50 times less sensitive t h a n the c o m m e r c i a l E L I S A .
• Y0583A
z
80-
O_
o T2212 Guinea Pig Serum •
5-
Post
i
•
/
-&-
60-
CO
z
m 40O
~3
~I
""/'"
20-
0
•
•
I-- 2
,< 'J 1
_51
~(/o °
. . . . . . . . .
i
l laooo
u
1;aoo
i
l:ao
0
DILUTION C O M M E R C I A L ELISA (A492) Fig. 4. Competition of unlabeled mAb and guinea pig polyclonal antibodies against labeled Y0583A for the binding with HBeAg. mAb (1 mg/ml) as well as preimmune and postimmune guinea pig sera were diluted and tested for the binding with HBeAg pretreated with NaOH and captured by C33 on a solid support.
Fig. 5. Correlation of A492readings obtained with ELISA with Y0583A and those with commercial ELISA (Immunis HBeAg/ Ab EIA, Institute of Immunology, Tokyo, Japan) with 904/ 905 in the detection of HBeAg in 30 plasma samples from blood donors carrying HBsAg.
214
S. Usuda et al. /Journal of ~qrological Methods 68 (1997) 207 215
4. Discussion
A monoclonal antibody (Y0583A) was raised against the N-terminal l0 aa in the HBeAg polypeptide (p17 e) which are encoded by the precore region. By epitope mapping with 203 decapeptides on multipins, Y0583A was deduced to be directed to a sequence of six aa (LGWLWG) representing a C-terminus of the precore region product. Y0583A bound with p17 e, but not with p21 c representing the product of the HBV core gene. Furthermore, it bound with a synthetic oligopeptide including the six aa but not with the mimicking sequence of the core-gene product in pl7L In addition, the binding of Y0583A with HBeAg, purified from plasma and pretreated with alkali, was blocked by unlabeled Y0583A as well as guinea pig polyclonal antibody raised against a synthetic oligopeptide containing the L G W L W G sequence, but not by mAb directed to a C-terminal sequences of p21 ° (T2212). Taken together, Y0583A would specifically recognize the epitope borne by the hexapeptide in the precore region. The binding of Y0583A with HBeAg purified from plasma was not strong. The association of HBeAg with plasma proteins such as albumin and IgG (Takahashi et al., 1978) may bury the epitope recognized by Y0583A in HBeAg. The pretreatment of HBeAg with alkali would be able to dissociate some plasma proteins for an easier access of Y0583A to the corresponding epitope on HBeAg. Many methods have been developed for the detection of HBeAg, such as two-dimensional immunodiffusion, passive hemagglutination, radioimmunoassay and ELISA (Magnius and Espmark, 1972; Takahashi et al., 1977; Miyakawa et al., 1979; Imai et al., 1982). However, all antiHBe reagents used are directed to the sequences in HBeAg that are encoded by the core gene. Commercial HBeAg tests are found to have unsatisfactory specificity (Westmoreland et al., 1994), which would be attributable to antibody reagents not directed to the precore sequence in HBeAg. To obtain specific antibodies to HBeAg, guinea pig and rabbit have been immunized with synthetic oligopeptides covering the ten aa in HBeAg encoded by the precore region (Takahashi et al., 1991; Dienes et al., 1995).
There are a number of clinical applications for HBeAg in the circulation. It is used as a marker for the response to interferon in patients with chronic hepatitis B (Tine et al., 1993; Wong et al., 1993). The lack of HBeAg in the patients infected with HBeAg-minus mutants is implicated in severe disease in the patients with HBeAg-negative or anti-HBe-positive chronic hepatitis (Bonino et al., 1986; Akahane et al., 1990) and those with fuhninant hepatitis B (Kosaka et al., 199l; Sato et al., 1995). A specific detection of HBeAg would be prerequisite to the molecular biological analysis of HBeAg-negative infections. An ELISA for HBeAg was developed by placing it between mAb (C33) to the core-gene sequence fixed on a solid support and Y0583A labeled with horseradish peroxidase. To increase the sensitivity and to avoid nonspecific reactions, HBeAg in the test plasma was captured by 904 on the solid support and treated with 0.2 N NaOH prior to ELISA with C33 and Y0583A. The ELISA detected HBeAg in 50/~1 of plasma samples from blood donors with A492 readings in a close correlation with those by commercial ELISA for HBeAg (lmmunis HBeAg/Ab EIA, Institute of Immunology, Tokyo, Japan). Although the sensitivity of Y0583A EL1SA was less than that of the commercial kits, it would be increased by using more volume of test samples. It is to be hoped that specific detection of HBeAg by ELISA involving Y0583A would contribute to better application and understanding of HBeAg in clinical and epidemiological settings. Furthermore, it will be useful in defining HBV variants with an HBeAg-minus phenotype and those with decreased expression of HBeAg, which are induced by mutations in the precore region and core promoter, respectively (Miyakawa et al., 1997).
References
Akahane, Y., Yamanaka, T., Suzuki, H., Sugai, Y., Tsuda, F., Yotsumoto, S., Omi, S., Okamoto, H.0 Miyakawa, Y., Mayumi, M., 1990. Chronic active hepatitis with hepatitis B virus DNA and antibody against e antigen in the serum. Disturbed synthesis and secretion of e antigen from hepatocytes due to a point mutation in the precore region. Gastroenterology99, 1113-1119.
S. Usuda et al. /Journal of Virological Methods 68 (1997) 207 215
Bonino, F., Rosina, F., Rizzetto, M., Rizzi, R., Chiaberge, E., Tardanico, R., Callea, F., Verme, G., 1986. Chronic hepatitis in HBsAg carriers with serum HBV-DNA and antiHBe. Gastroenterology 90, 1268-1273. Dienes, H.P., Gerken, G., Goergen, B., Heermann, K., Gerlich, W., Meyer zum Buschenfelde, K.H., 1995. Analysis of the precore DNA sequence and detection of precore antigen in liver specimens from patients with anti-hepatitis B e-positive chronic hepatitis. Hepatology 21, 1-7. Imai, M., Nomura, M., Gotanda, T., Sano, T., Tachibana, K., Miyamoto, H., Takahashi, K., Toyama, S., Miyakawa, Y., Mayumi, M., 1982. Demonstration of two distinct antigenic determinants on hepatitis B e antigen by monoclonal antibodies. J. Immunol. 128, 69-72. Kosaka, Y., Takase, K., Kojima, M., Shimizu, M., lnoue, K., Yoshiba, M., Tanaka, S., Akahane, Y., Okamoto, H., Tsuda, F., Miyakawa, Y, Mayumi, M., 1991. Fulminant hepatitis B: induction by hepatitis B virus mutants defective in the precore region and incapable of encoding e antigen. Gastroenterology 100, 1087-1094. Machida, A., Ohnuma, H., Tsuda, F, Yoshikawa, A., Hoshi, Y., Tanaka, T., Kishimoto, S., Akahane, Y., Miyakawa, Y., Mayumi, M., 1991. Phosphorylation in the carboxylterminal domain of the capsid protein of hepatitis B virus: evaluation with a monoclonal antibody. J. Virol. 65, 60246030. Magnius, L.O., Espmark, J.A., 1972. New specificities in Australia antigen positive sera distinct from the Le Bouvier determinants. J. Immunol. 109, 1017 1021. Matsui, T., Takano, M., Miyamoto, K., Itoh, Y., Yoshizawa, H., Koike, M., Mochizuki, T., Tanaka, E., Okamoto, H., lmai, M., Mishiro, S., Miyakawa, Y., Mayumi, M., 1986. Nude mice bearing human primary hepatocellular carcinoma that produces hepatitis B surface, core, and e antigens, as well as deoxyribonucleic acid polymerase. Gastroenterology 90, 135-142. Merrifield, R.B., 1969. Solid-phase peptide synthesis.. Adv. Enzymol. 32, 221-296. Miyakawa, Y., Mayumi, M., 1985. Hepatitis B e antigen and antibody (HBeAg/Anti-HBe). In: R.J. Gerety (Ed.), Hepatitis B, Academic Press, Orlando, pp. 47-76. Miyakawa, Y., Okamoto, H., Mayumi, M., 1997. The molecular basis of hepatitis B e antigen (HBeAg)- negative infections. J. Viral Hepatitis 4, 1 8. Miyakawa, Y., Tsuda, F., Akahane, Y., Mayumi, M., 1979. Application of microtiter solid-phase radioimmunoassay to the determination of hepatitis B e antigen. Vox Sang. 36, 307-311. Nomura, M., Imai, M., Nakamura, T., Miyakawa, Y., Mayumi, M., 1982. Demonstration of two distinct antigenic determinants on human e-foetoprotein by monoclonal antibodies. Mol. Immunol. 19, 1691-1697. Okamoto, H., Tsuda, F., Sakugawa, H., Sastrosoewignjo, R.I., Imai, M., Miyakawa, Y., Mayumi, M., 1988. Typing
215
hepatitis B virus by homology in nucleotide sequence: comparison of surface antigen subtypes. J. Gen. Virol. 69, 2575-2583. Ou, J.H., Laub, O., Rutter, W.J., 1986. Hepatitis B virus gene function: the precore region targets the core antigen to cellular membranes and causes the secretion of the e antigen. Proc. Natl. Acad. Sci. U.S.A. 83, 1578 1582. Roossinck, M.J., Jameel, S., Loukin, S.H, Siddiqui, A., 1986. Expression of hepatitis B viral core region in mammalian cells. Mol. Cell. Biol. 6, 1393 1400. Sato, S., Suzuki, K., Akahane, Y., Akamatsu, K., Akiyama, K., Yunomura, K., Tsuda, F., Tanaka, T., Okamoto, H., Miyakawa, Y., Mayumi, Y., 1995. Hepatitis B virus strains with mutations in the core promoter in patients with fulminant hepatitis. Ann. Intern. Med. 122, 241-248. Takahashi, K., Fukuda, M., Baba, K., Imai, M., Miyakawa, Y., Mayumi, M., 1977. Determination of e antigen and antibody to e by means of passive hemagglutination method. J. Immunol. 119, 1556-1559. Takahashi, K., Imai, M., Miyakawa, Y., Iwakiri, S., 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. Natl. Acad. Sci. U.S.A. 75, 1952-1956. Takahashi, K., Kishimoto, S., Ohori, K., Yoshizawa, H., Machida, A., Ohnuma, H., Tsuda, F., Munekata, E., Miyakawa, Y., Mayumi, M., 1991. Molecular heterogeneity of e antigen polypeptides in sera from carriers of hepatitis B virus. J. Immunol. 147, 3156-3160. Takahashi, K., Machida, A., Funatsu, G., Nomura, M., Usuda, S., Aoyagi, S., Tachibana, K., Miyamoto, H., Imai, M., Nakamura, T., Miyakawa, Y., Mayumi, M., 1983. Immunochemical structure of hepatitis B e antigen in the serum. J. Immunol. 130, 2903-2907. Tine, F., Liberati, A., Craxi, A., Almasio, P., Pagliaro, L., 1993. Interferon treatment in patients with chronic hepatitis B: a meta-analysis of the published literature. J. Hepatol. 18, 154-162. Westmoreland, D., Parr. A., Thomas, A.J., 1994. Assessment of hepatitis B 'e' antigen and 'e' antibody status-tests of status or status of tests?. J. Med. Microbiol. 41, 374-377. Will, H., Reiser, W., Weimer, T., Pfaff, E., Buscher, M., Sprengel, R., Cattaneo, R., Schaller, H., 1987. Replication strategy of human hepatitis B virus. J. Virol. 61,904 911. Wong, D.K., Cheung, A.M., O'Rourke, K., Naylor, C.D., Detsky, A.S., Heathcote, J., 1993. Effect of alpha-interferon treatment in patients with hepatitis B e antigen-positive chronic hepatitis B: a meta-analysis. Ann. Intern. Med. 119, 312-323. Yaginuma, K., Shirakata, Y., Kobayashi, M., Koike, K., 1987. Hepatitis B virus (HBV) particles are produced in a cell culture system by transient expression of transfected HBV DNA. Proc. Natl. Acad. Sci. U.S.A. 84, 2678-2682.