Association of hepatitis B viral precore mutations with fulminant hepatitis B in Japan

VIROLOGY

l&460-463

(1991)

Association

of Hepatitis

B Viral Precore Mutations

with Fulminant

Hepatitis

B in Japan

KIYOSHIHASEGAWA,JIAKANGHUANG, JACKR. WANDS, *HIROSHI OBATA, AND T. JAKELIANG Department of Medicine, Harvard Medical School, Molecular Hepafology Laboratory, Massachusetts General Hospital Cancer Center, Charlestown, Massachuserrs 02129; Gastrointestinal Unit, Massachusetts General Hospital, Boston, Massachuserts 02114; and *The Institute of Gasrroenterology, Tokyo Women’s Medical College, Tokyo, Japan 162 Received July 1, 199 1; accepted August 9, 199 1 We studied the precore DNA sequences of hepatitis B viral genomes in five patients with fulminant hepatitis B and in five with acute self-limited hepatitis B from Japan. Using the polymerase chain reaction, three to four independent HBV DNA clones from each patient were obtained and analyzed. We demonstrated that patients with fulminant hepatitis B carried HBV genomes with a G to A mutation at nucleotide positions 1898 (five of five patients; 18 of 18 clones, 100%) and 1901 (five of five patients; 12 of 18 clones, 66%) in the precore region. The first mutation results in an in-phase stop codon (TAG) in the precore open reading frame and the absence of HBeAg production. In contrast, a G to A mutation was found in 6 of 16 clones (37%) in position 1898 and in 0 of 16 clones (0%) in position 1901 from patients with acute self-limited hepatitis. We concluded that both of the precore mutations are commonly associated with fulminant hepatitis B and may contribute to the pathogenesis of fulminant hepatitis. A hypothetical model for the biological significance of these two mutations is proposed. Q 1991Academic PW. IIIC.

Infection with hepatitis B virus (HBV) leads to a wide spectrum of liver injury ranging from acute self-limited infection, fulminant hepatitis, chronic hepatitis with progression to cirrhosis and liver failure as well as to an asymptomatic chronic carrier state. Fulminant hepatitis B is a rare disease but carries a high mortality (I, 2). Co-infection or superinfection with hepatitis D virus (HDV) often contributes to a poor prognosis (I). There is no conclusive evidence at this point that hepatitis C virus is associated with acute fulminant hepatitis (3). In this report, we analyzed the DNA sequences of HBV genomes isolated from Japanese patients with fulminant hepatitis B and established that precore mutations are commonly associated with fulminant hepatitis. Five Japanese patients with fulminant hepatitis B and five with acute self-limited hepatitis B from the Tokyo Women’s Medical College were studied. Diagnosis was based on their clinical course, laboratory data, serologic studies, and liver’biopsy which is commonly performed in patients with acute hepatitis in this Japanese institution. Patients who presented with encephalopathy within 2 weeks of the onset of jaundice were defined as having fulminant hepatitis( 7). All serum samples were obtained during the acute phase of illness. DNA was extracted from the serum of each patient and the HBV DNA fragment spanning the X, precore, and

core regions was amplified as described previously (4). The primers used are 5’ GAAAGCTKTGCGACGCGGCGAlTGAGA 3’ (adw 2429-2410) and 5’ TGGAATTCGCATGGAGACCACCGTGAAC 3’ (adw 1608-1627). The underlined sequences of the first primer represent a HindIll restriction site and those of the second primer an EcoRl site. The nucleotide positions were based on the published HBV sequence for the subtype adw from GenBank. All the proper precautions were observed to minimize PCR contaminations (5). The 822-bp fragments of HBV DNA generated by the above two primers spanning the X, precore, and core regions from all IO patients were cloned into pGEM7Zf(+) (Promega Biotech, Madison, WI) and sequenced using the T7 Sequencing kit from Pharmacia (Piscataway, NJ). The difference in the prevalences of HBV precore mutations between fulminant and acute self-limited cases was analyzed for statistical significance by the two-tailed Fisher’s exact test. HBsAg was measured by Ausria II (Abbott Laboratories, North Chicago, IL); anti-HBc, anti-HBc IgM, HBeAg, anti-HBe, anti-HBs, and anti-HDV were measured by Corab, Corzyme-M, HBe(rDNA), Ausab, and Anti-delta EIA (Abbott Laboratories), respectively. Enzyme-linked immunosorbent assay (ELISA) to detect antibody to hepatitis C virus (6, 7) was performed according to the manufacturer’s instructions (Abbott Laboratories). Samples were assayed in duplicate and only samples positive in duplicate were recorded as reactive. For detection of HCV genomes, RNA was extracted from serum, treated with reverse-transcriptase in the presence of random hexamer (Pharmacia) and

’ Address all correspondences and reprint requests to Dr. T. J. Liang, Molecular Hepatology Laboratory, Massachusetts General Hospital, Cancer Center, 149 13th Street; 7th Floor, Charlestown, MA 02129. 0042-6822191

$3.00

Copyright 0 1991 by Academic Press. Inc. All rights of reproduction in any form reserved.

460

SHORT COMMUNICATIONS

461

TABLE 1 THE CLINICAL COURSE, SEROL~C~IC PROFILE, AND PRECORE SEQUENCE ANALYSIS OF PATIENTS WITH FULMINANT Age

Case No. Fulminant 1

Sex

Days from onset*

ALTIASTb

HEsAg/ antiHBs

HBeAg/ antiHBe

IgM antiHBc

AND AWE

SELF-L&+1TEo

Outcome

No. of clones studied

Survival Death Death Death Death Survival Survival

HEPATITB

6

nt 1698

nt 1901

4 4 4 3 3

4A 4A 4A 3A 3A

4A 2A/2G lA/3G 2A/1G 3A

4 3 3 3 3

4A 1AI2G 3G 3G le/2G

4G 3G 3G 3G 3G

hepatitis

2 3 4 5 Acute self-limited hepatitis 6 7 8 9 10

22F

7

1010/1870

--/-I-

-l-

+

27F 25F 61F 35M

4 5 5 5

87912925 508014165

i-lfi-

-/-/-

+ +

1084/650

i-l-

-/+

+

1182/1712

-i-l-+

-I-/-

+

53M 38M 41M 28F 61M

8 15 23 15 20

242712550 71511287 3551542 3861639 312/831

+/-+IA-/--I+-l-l-

-/+/+/-/-+ +/-

+ + + + +

Survival

Survival Survival

a Onset was defined as the beginning of jaundice. b Peak levels of ALT and AST in IWliter.

deoxynucleotides, and subjected to nested PCR amplification using two sets of primers spanning the 5’ noncoding region of HCV genome (8). Serum HBV DNA was successfully amplified from all the patients. To eliminate a mutation introduced by the Taq polymerase during PCR amplification and to ensure that the HEW sequence obtained was representative of the predominant viral species, we sequenced three to four independent clones from separate PCR amplifications of the HBV DNA in each patient’s serum. The results are summarized in Table 1. All clones from the five patients with fulminant hepatitis contained a G to A mutation at nucleotide position 1898 (18 of 18 clones). Two patients (Nos. 1 and 5) also had a G to A mutation at nucleotide position 190 1 and three other patients (Nos. 2, 3 and 4) harbored a mixture of viral strains with either the A (mutant) or the G (wild type) nucleotide at the same position. In contrast, all the viral strains from patients with acute selflimited hepatitis contained the wild-type sequence of G at nucleotide position 190 1 (16 of 16 clones). Two of them (Nos. 8 and 9) had the wild-type sequence of G, two others (Nos. 7 and 10) harbored a mixture of A and G, and the fifth (No. 6) contained the mutation of A at nucleotide position 1898. The precore sequences containing the mutations are shown in Fig. 1. The difference in the numbers of HBV isolates containing the precore mutations between fulminant hepatitis and acute self-limited hepatitis was significant at a P value of
tive. Anti-HCV antibodies may not ba positive at an early stage of acute HCV infection (3) and the currently available anti-HCV test is not sensitive enough to detect all cases of HCV infection ( 70, 7 7). In order to elimi-

A.

B c

A.

B

c

AGCT

AGCT

AGCT

r Core r. . .GGGTGGCTTTGGGGCATGGACAT Precore

,.‘i”“I’*e+ A

G

1 . ..“ip A A StP

FIG. 1. Precore mutations in HBV DNA clones from patients with fulminant and acute &f-limited hepatitis 8. ~~~~~~ WA sequencing gels of three different precore sequgnceer 91& stwwn st tie top and the actual DNA sequences are at tha bottom. (A) The wildtype sequence. (B) The precore sequence with a G to A mutation at nucleotide position 1898. (C) The precore sequence with G to A mutations at both nucleotide positions 1898 and 190 1.

462

SHORT COMMUNICATIONS

nate any possibility of HCV co-infection, serum ribonucleic acids were extracted and subjected to reversetranscription PCR using primers specific for the 5’ noncoding region of HCV genome. No HCV-specific sequences were detected using this method (data not shown). Patients with fulminant hepatitis B have a high mortality rate. The prognosis is more favorable with recent advances in intensive medical care and orthotopic liver transplantation (7). Clinically, these patients have a short course of HBV antigenemia (HBsAg and HBeAg) and an early enhanced production of antibodies to HBV antigens (7, 72). The pathogenesis of liver injury due to acute and chronic HBV infection is not well understood. Hepatocellular injury is a likely sequela of the host immune response to HBV antigens and not due to a direct cytopathic effect of the virus (73). Taken together, these observations suggest that an enhanced host’s immune response to the infection may be more important than the level of HBV replication in the pathogenesis of fulminant hepatitis. Since studies on fulminant hepatitis B were usually performed late in the course of the disease, it is difficult to reconstruct the sequence of virologic and immunologic events leading to massive hepatocellular injury. Several groups have described the same HBV mutation of G to A at nt 1898 resulting in a stop codon at the end of the precore region, and this mutation has been associated with a clinically more aggressive form of chronic liver disease in a group of Mediterranean patients with anti-HBe and high levels of circulating HBV DNA (74 76). A second precore mutation of G to A at nt 1901 has also been reported in combination with the first mutation in this group of patients, Such patients also demonstrated a poor response to interferon therapy (7 7). These observations suggest that the mutation(s) may influence the host’s immune response and/ or alter the antigenic properties of HBV. Thus, a more severe and progressive liver disease may be found in individuals chronically infected with this strain. In addition, various mutations in the precore region leading to elimination of the open reading frame and absence of HBeAg production have been observed to arise de nova in patients chronically infected with wild-type HBV (78). Therefore, it is possible that these mutations or the absence of HBeAg confers a biologic advantage to the survival and persistence of viral genomes carrying these mutations. We have previously described the association of an HBV variant containing these two precore mutations with an epidemic of fulminant hepatitis B (79). This nosocomial outbreak of five cases was remarkable for an unusually severe course of acute HBV infection leading to 100% mortality within a few days of hepatic and

renal failure. Kosaka et a/. recently reported the presence of the G to A mutation at position 1898 in 9 of 10 patients with sporadic fulminant hepatitis B and the wild-type precore sequences in 10 patients with acute self-limited hepatitis B from Japan (20). None of the patients contained a second mutation of G to A at nucleotide position 1901. Omata et al. reported similar findings in other Japanese patients with fulminant and acute self-limited hepatitis (27). In contrast, our current study demonstrated the common occurrence of the second mutation in our Japanese patients with fulminant hepatitis B. Our study also demonstrated that the HBV strain with the G to A mutation at nt 1898 can also be detected in patients with acute self-limited hepatitis and is not exclusively associated with fulminant hepatitis, as suggested by Kosaka et al. This discrepancy could be due to a geographic variation of the predominant HBV strains; the cases described by Kosaka eta/. and Omata et al. occurred in several cities outside Tokyo and our cases were in Tokyo. One of the patients with fulminant hepatitis was seroreactive for HBeAg, although he contained HBV genomes which are incapable of directing HBeAg synthesis. One possible explanation for this finding is that intracellular core particles may be released as a result of massive hepatocellular necrosis, accounting for the HBeAg immunoreactivity; HBcAg in a nonparticular or partially degraded form is detectable by the currently available assay for HBeAg and will be recorded as a positive test (22). In this study, the predominance of females in the group of patients with fulminant hepatitis could be a mere coincidence. There has been no evidence in the literature suggesting that females are more predisposed to fulminant hepatitis B than males (7,2,2(I). It is of interest that one of our patients with acute selflimited hepatitis (No. 6) who contained exclusively the first mutation at nucleotide position 1898 presented with very high levels of transaminases and a short incubation period not too dissimilar from those of patients with fulminant hepatitis. However, he did not progress into encephalopathy as the other patients with fulminant hepatitis did. These observations suggest that both of the precore mutations may be important in the pathogenesis of fulminant hepatitis. It is possible that other mutations within the viral genome in addition to those described here (Fig. 1) may contribute to a more fulminant course of hepatitis B. At present, the biological significance of these two unique precore mutations in the precore region is not well understood. The first mutation at nucleotide position 1898 might alter the pathogenesis of acute HBV infection by virtue of the absence of HBeAg production associated with the HBV strain carrying this mutation. Since HBc/eAg has been proposed to be the target of the host immune

SHORT

system eliminate infected hepatocytes (23, 24, the HBeAg-negative HBV mutant may have an advantage in its replication and dissemination in the infected host over the wild-type virus at the initial phase of acute infection. In addition, these two nucleotides lie within the genomic RNA “packaging signal” proposed by Junker-Niepmann et al. (25). By conducting deletion anaiysis and studying the secondary structure of the precore nucleotide sequences, Junker-Niepmann et al. postulated a complicated stem-loop structure with extensive intrastrand base pairing as an important regulatory sequence in HBV replication. This structure is highly conserved among the hepadnaviruses (26). The G to A mutation at both nucleotide positions 1898 and 190 1 will provide perfect base pairing matches at the stem of this structure. Therefore it is conceivable that these two mutations may alter the regulation of HBV transcription and replication, leading to a more severe course of hepatitis. At present, we are conducting experiments to address these issues.

10. 11. 12. 13. 14. 15. 16. 17.

ACKNOWLEDGMENTS We are grateful for the superb secretarial assistance Kristin Cambria, excellent technical support from Mrs. Brown and Mr. Rolf Carlson, and helpful advices from Drs. Blum and Kurt J. issefbacher during the preparation of script. This work was supported by Grants CIA DKO1952 CA54524 from the National Institute of Health.

463

COMMUNICATIONS

18. from Ms. Nancy V. Hubert E. the manuand ROl

REFERENCES 1. BERNUAU, J., RUEFF, B., and BENHAMOU, J-P., Semin. Liver Dis. 6, 97-106 (1986). 2. Fulminant hepatitis. MMWR. 33, 76-77 (1984). 3. READ, A. E., DONEGAN, E., LAKE, J., FERRELL, L., GALBRAITH. C., KURAMOTO, I. K., ZELDIS, J. B., ASCHER, N. L., ROBERTS, J., and WRIGHT, T. L., Ann. Intern. Med. 114, 282-284 (1991). 4. LIANG, T. J., ISSELBACHER, K. J., and WANDS, J. R., J. C/in. invest 64, 1367-l 371 (1989). 5. KWOK, S., and HIGUCHI, R., Nature 339, 237-238 (1989).

19. 20.

CHOO, Q-L, Kuo, G., WEINER, A. J., QVER~Y, L. R., BRALXDI, D. W., and HOUGHTON, M., Science 244,359-362 (1989). Kuo, G., CHOO. G-L, ALTERS, H. J., and et&., Science 244,362364 (1989). KANAI, K., IWATA, K., NAKAO, K., KAKo, M.. and OKAMOTO, H., Lancer 336,245 (1990). ALTER, H.. PURCELL, R., SHIH, J. W., MELPOUMR, J. C., H-TON, M., CHOO, G-L, and Kuo, G., N. Engl J. Med. 321, 1494-l 500 (1989). WEINER, A., Kuo, G., BFZACILEY, D., BOMNO, F., SAARACCO. G., LEE, C., ROSENBLAT, J.. and et a/., Lamer 335, l-3 (1990). GARSON, 1. A., TEDDER, R. S., BRKXX, M., TUKE, P.. G~J&zE~~~xx, J. A., TRUTE, A., PARKER, D., BARWARA, J. A., CONIRERAS, M., and ALOVSIUS, S., Lancet335, 1419-1422 (1990). SARACCO, G., MAMAS, S., Ro~#A, F., and et al., Ann. Intern. Med. 108,380-383 (1988). MONDELU, M., and EDDLESTON, A. L., Semin. Liver Dis. 4,47-58 (1984). CARMAN, W. F., JACYNA, M. R., HADZIYAMIS. S., and etal., Lancer ii, 588-590 (1989). BRUNETTO, M. R., STEMLER. M., S~HODEC, F., and et sl., Ital. 1. Gestroenferol. 21, 15 1 - 154 (1989). BRUNEITO. M. R., STEMLER, M., BONINO, F., and et al., 1. Hepatol. 10,258-261 (1990). BRUNEITO, M. R., OLIVERI, F., R~(.xA. G., and eta/., Hepatology 10,198-202 (1989). OKAMOTO, H., YOTSUMOTO, S., AKAHANE, Y., and et al., J. Viral. 64, 1298-l 303 (1990). LIANG, T. J., HA~E~WA, K.. RIMON, N., WANDS, J. R., and BENPORATH, E., N. Engl. J. Med. 324, 1705-l 708 (199 1). KO~AKA, Y., TAKASE. K., KOJIMA, M., and er al., Gastroenterology

324,1087-1094 (1991). 21. OUATA, M., EHATA, T., YOKOSUKA, O., HOSODA, K., and OHTO. M., N. Engl. J. Med. 324, 1699-l 704 (1991). 22. SALFELD, J., PFAFF, E., NOAH, M., and SCHALLER, H., J. Viral. 63, 798-808 (1989). 23. MONDELLI, M., VERGANI, G. M., ALBERTI, A., VE@X@JI, D., PORTMANN, B., EDDLESTON, A. W. F., and WILLIAMS, R., 1. Immunol. 129,2773-2777 (1982). N. V., MODELLI. M., ALE~ANMR, G. J. M., TEOOER, R. S., EDDLESTON, A. L. W. F., and WILLIAMS. R., Hepato/ogy 4, 6368 (1984). JUNKER-NEPMANN. M., BARTENSCHLAMR, R., and SCHALLER. H., EMS0 J. 9,3389-3396 (1990). MILLER, R. H., KANEKO, S., CHUNG, C. T.. GwIoNEs, R., and PURCELL, R. H., Heparology 9, 322-327 (1989).

24. NAUMOV, 25. 26.