frequent integration of precore/core mutants of hepatitis b virus in human hepatocellular carcinoma...

Frequent integration of precore/core mutants of hepatitis Bvirus in human hepatocellular carcinoma tissuesS. Zhong,1 J. Y. H. Chan,1 W. Yeo,1 J. S. Tam2 and P. J. Johnson1

Departments of 1Clinical Oncology and2Microbiology, Sir Y. K. Pao Centre for Cancer, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong

Received 30 July 1999; accepted for publication 8 October 19991

INTRODUCTION

Hepatocellular carcinoma (HCC) is a common malignancy in

many parts of the world, particularly in the Far East and

sub-Saharan Africa. On the basis of extensive epidemiologi-

cal, experimental and molecular studies [1±5], persistent

HBV infection has been shown to be strongly associated with

HCC. Proposed mechanisms of hepatitis B virus (HBV)-

associated carcinogenesis include insertional mutagenesis

[6] and the transactivating role of the hepatitis B X protein

[7±9] or truncated PreS2/S gene [10±12]. However, the

precise mechanism by which HBV leads to the development

of HCC remains unclear.

Development of HBV-associated HCC frequently follows a

prolonged period of virus replication and disease progression

from acute to chronic hepatitis. During this time, subge-

nomic fragments of HBV DNA may integrate randomly into

the hepatocyte chromosomes. The detection of integrated

HBV DNA in liver biopsies from persistently infected indi-

viduals without tumours [13,14] implies clonal expansion

following the initial integration event. Such expanded clones

may be subjected to selection (e.g. by the immune response

to viral gene products that may be expressed) and further

mutagenic events.

The HBV precore/core open reading frame (ORF) encodes

two closely related proteins: the secreted hepatitis B e anti-

gen (HBeAg) and the nucleocapsid core protein ± hepatitis B

c antigen (HBcAg). Frequent mutations of the pre-C/C region

of HBV DNA have been identi®ed in patients with chronic

hepatitis [15±17].

Abbreviations: HBcAg, hepatitis B core antigen; HBeAg, hepatitis B e

antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; nt,

nucleotide; PCR, polymerase chain reaction; pre-C/C, precore/core.

Correspondence: Professor Philip J. Johnson, Department of Clinical

Oncology, Sir Y. K. Pao Centre for Cancer, The Chinese University of

Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong.

Journal of Viral Hepatitis 2000, 7, 115±123

Ó 2000 Blackwell Science Ltd

SUMMARY. The development of hepatitis B virus (HBV)-

associated hepatocellular carcinoma (HCC) frequently fol-

lows persistent HBV infection and may arise in individuals

who are hepatitis B e antigen (HBeAg) negative, indicating

the possible presence of precore/core mutants. It is unclear

whether precore/core mutants are associated with tumour

development or are selected for after chromosomal integra-

tion of the wild-type viral DNA. We studied the status and

sequence variation of the precore/core region of HBV in 56

patients with HBV-associated HCC and in various corres-

ponding non-tumour tissues by Southern blot analysis,

polymerase chain reaction and direct sequencing. Southern

blot showed that integrated HBV DNA existed in 43 of 56

HCC tissues. Sequence analysis revealed mutations in 65% of

the HCC (26/40) and 45% (14/31) of the corresponding

non-tumour tissues. The mutation at nucleotide (nt) 1896,

known to prevent HBeAg synthesis, was detected in 40%

(16/40) of the tumours and in 35.4% (11/31) of the non-

tumour tissues. Other mutations were found at nt 1899

(eight of 40 in HCC; three of 31 in non-tumour tissues), nt

1898 (seven of 40 in HCC; two of 31 in non-tumour tissues),

nt 1912 (seven of 40 in HCC; none of 31 in non-tumour

tissues) and nt 1886 (three of 40 in HCC; none of 31 in non-

tumour tissues). To determine whether this ®nding merely

re¯ected the prevalence of such mutants in this geographical

region, HBV DNA from the sera of patients (also in this

region) with acute and chronic hepatitis were sequenced.

The nt 1896 mutant was found in 5.6% (one of 18) of

patients with acute hepatitis B and in 22.8% (nine of 35) of

patients with chronic hepatitis B. However, the nt 1898

mutation was not found in any of these sera. The precore/

core mutant was observed with increasing frequency from

acute hepatitis to chronic hepatitis, non-tumour and HCC,

and this difference in frequency was signi®cant between HCC

and acute hepatitis B groups (P < 0.01), suggesting that the

precore/core mutant or hepatocytes harbouring this mutant

may be under immune selection and that such mutations

may facilitate integration and subsequent tumour develop-

ment.

Keywords: HBeAg, hepatitis B virus, hepatocellular carcinoma,

precore/core mutant.

Patients with HBV-associated HCC are often negative for

HBeAg, a marker of replication of the wild-type virus. Such

patients may remain viraemic with the HBeAg-negative

variant described above, or may have cleared replicating

virus since the initial integration event(s). In the former

situation, evolution and immune selection of the HBeAg-

negative variants may have occurred following integration

into preneoplastic cells. Alternatively, replication and inte-

gration of precore/core may be a risk factor for the devel-

opment of HCC [18]. Thus, investigation of the precore/

core region of HBV DNA sequences from paired tumour

and non-tumour samples may provide information on the

possible role of the HBeAg-negative mutant on tumour

development.

To assess the frequency of integration of precore/core

mutants into HCC tissues, we investigated the status of HBV

genomes by Southern blotting, and compared sequences of

the HBV precore/core region in HCC tissues and adjacent

non-tumorous tissues in HBV-associated HCC patients. In

addition, HBV DNA from the sera of patients with acute and

chronic HBV infection was also compared. Our results indi-

cate that there is an increasing frequency of integrated

precore/core mutants from acute hepatitis B infection to

chronic hepatitis B infection to HCC tissues and non-tumour

tissues, suggesting that the precore/core mutant viruses may

be under immune selection and have preferential integra-

tion, which may further contribute to tumorigenesis.

PATIENTS AND METHODS

Patients

Surgically resected HCC and non-tumorous liver tissues from

56 patients with HCC were collected between 1990 and

1995. Fresh tumour tissue and adjacent liver were each

divided into two parts. One part was immediately frozen in

liquid nitrogen and stored at ±80 °C until used for DNA

isolation. The other part was processed for histological

examination. Tissue samples of 0.5 g were used for DNA

extraction and for subsequent analysis of HBV DNA by

Southern blot hybridization and polymerase chain reaction

(PCR) ampli®cation. Serum was collected from each patient

for analysis of HBV markers using commercial kits (Abbott

Laboratories, North Chicago, IL).

All DNA samples were tested for HBV DNA by Southern

hybridization or PCR using multiple primers based on dif-

ferent regions of the HBV genome. Eleven samples were

negative and excluded from the study group for Southern

analysis and sequence analysis. The details of the remain-

ing 45 patients (25 from Hong Kong and 20 from

Shanghai) are given in Table 1. Seven were hepatitis B

surface antigen (HBsAg) seronegative and the other 38

were HBsAg positive. Patients were born in China and

Hong Kong and were presumed to have acquired HBV

early in life.

Non-HCC controls

Serum samples were obtained from 66 HBsAg-positive Chi-

nese patients (in Hong Kong) with acute or chronic hepatitis

B, none of whom had any evidence of HCC. Fifty-three were

viraemic by nested PCR (18 acute and 35 chronic hepatitis).

Human placenta tissues and sera from healthy individuals

without serological markers for HBV were used as negative

controls.

Southern blot analysis for HBV sequences

DNA was isolated from tissues or sera using proteinase K

digestion followed by phenol±chloroform extraction.

Extracted DNA was dissolved in 50 ll TE buffer

(10 mmol l±1 of Tris-HCl, 0.5 mmol l±1 of disodium EDTA,

pH 8.0). Southern blotting was performed as described pre-

viously [6]. Brie¯y, puri®ed cellular DNA (10 lg) was com-

pletely digested with the restriction enzymes HindIII or EcoRI

and then subjected to electrophoresis in a 0.7% agarose gel.

DNA fragments in the gel were transferred onto Hybond-N

nylon ®lters (Amersham, Little Chalfont, Bucks, UK). The

®lters were hybridized with a 32P-labelled HBV DNA (adr)

probe (2 ´ 108 counts per minute [c.p.m.] lg±1 of DNA)

labelled with [a-32P]dCTP using a labelling kit (Amersham)

and then analysed by X-ray autoradiography.

PCR for the pre-C/C gene of HBV

The nucleotide sequences and positions of the primers are

listed in Table 2. Viral DNA from tissues and sera was

extracted, as described previously [6]. For the sera, 200 ll of

samples were incubated at 60 °C for 4 h in 20 mM Tris-HCl,

pH 8.0, 10 mM EDTA, 0.1% sodium dodecyl sulphate (SDS)

and 800 lg ml±1 proteinase K. The samples were then

extracted with an equal volume of phenol/chloroform and

precipitated with two volumes of ethanol using 20 lg of

yeast tRNA as carrier. The extracted DNA was dissolved in

20 ll TE buffer. Five microlitres (100 ng) of the DNA

extracted from tissues was used for PCR ampli®cation in a

50-ll reaction mixture containing: 10 mM Tris-HCl, pH 8.8,

50 mM KCl, 1.5 mM MgCl2, 0.2 lM of each of the two oli-

gonucleotide primers, 200 lM of each of the four dNTPs

(dATP, dGTP, dCTP and dTTP) and 1 U of Taq polymerase

(Stratagene, La Jolla, CA). The ampli®cations were carried

out for 35 cycles as follows: 95 °C for 1 min, 56 °C for

1 min and 72 °C for 1.5 min, followed by a ®nal extension of

5 min. Nested PCR for serum HBV DNA was performed

under similar conditions, except for an elongated extension

time (72 °C for 2.5 min2 ). In all of the PCR reactions, the

cloned HBV DNA (GenBank accession no: M38454) was

used as a positive control. DNA from placenta tissues of an

HBsAg-negative subject and reaction mixtures without any

DNA template were used as negative controls. False positives

were avoided by strict application of published control

Ó 2000 Blackwell Science Ltd, Journal of Viral Hepatitis, 7, 115±123

116 S. Zhong et al.

Table 1 Status of hepatitis B virus (HBV) DNA in hepatocellular carcinoma (HCC) and HBV serological markers

HBV DNA HBV serological markers

Samples Gender/age Southern blot PCR HBsAg/HBsAb HBeAg/HBeAg

H1 M/37 Int/rep + +/) NA/NA

H2 M/60 Int/rep + +/) +/)H3 M/39 Int + +/) +/)H4 M/71 NA + +/) )/+

H5 F/54 NA + +/) )/+

H6 M/12 Int + NA/NA NA/NA

H7 M/53 Int + +/) )/+

H8 M/58 Int + +/) )/+

H9 F/40 Int + +/) +/)H10 M/41 Int + +/) +/)H11 M/68 Int/rep + +/) )/+

H12 F/59 Int + +/) )/+

H13 M/45 Int ± +/) )/+

H14 F46 Int + +/) +/)H15 M/55 Int/rep + +/) )/)H16 F/66 Int + +/) +/)H17 M/66 Int + )/+ )/+

H18 M/64 Int/rep + +/) +/)H19 M/64 Int ) +/) )/+

H20 F/69 NA ) +/) )/)H21 M/74 Int + )/) )/)H22 M/68 Int + )/) )/+

H23 M/41 Int + +/) )/+

H24 M/48 Int/rep + +/) +/)H25 M/67 Int + +/) )/+

S1 M/37 Int ) +/) )/)S2 F/53 Int + +/) )/)S3 M/36 Int + )/) )/)S4 M/63 Int + +/) )/)S5 M/41 Int + +/) )/)S6 M/53 Int + ) )/)S7 F/32 Int + +/) )/)S8 M/22 Int + +/) )/)S9 M/45 Int + +/) )/)S10 M/61 Int + +/) )/)S11 M/59 Int + +/) )/+

S12 M/56 Int + )/+ )/)S13 M/38 Int + +/) )/)S14 M/34 Int + +/) )/)S15 M/45 Int + ) )/)S16 M/47 Int ) ) )/)S17 M/53 Int + +/) )/+

S18 M/47 Int + +/) )/)S19 M/41 Int + +/) )/)S20 F/32 Int + +/) )/)

HBeAb, antibody to hepatitis B e antigen; HBeAg, hepatitis B e antigen; HBsAb, antibody to hepatitis B surface antigen;

HBsAg, hepatitis B surface antigen; Int, integrated form of HBV DNA; Int/rep, integrated and replicated forms of HBV

DNA; NA, not available; +, positive; ), negative.


Integration of HBV from precore/core mutants in HCC tissues 117

measures. After PCR, an aliquot of the reaction mixture was

electrophoresed on a 1.5% agarose gel and then blotted onto

a nylon membrane. The membrane was then hybridized

with the labelled HBV DNA as described above.

Direct sequencing of PCR products

HBV DNA mutations were detected by sequencing. Brie¯y,

PCR-ampli®ed DNA bands were cut out from the agarose gel

and puri®ed using a SephaglasTM BandPrep Kit (Pharmacia

Biotech, Uppsala, Sweden). The DNA sequence of the reac-

tion product was determined with the two PCR primers

using a Sequenase Kit (Amersham Life Science, Cleveland,

OH), according to the manufacturer's instructions.

RESULTS

Analysis of integration of HBV subgenome in HCCand non-tumour tissues by Southern blotting

Southern blotting of DNA from HCC tissues using the HBV

genomic probe revealed that all the 45 HBV DNA-positive

HCC samples contained the integrated form of HBV DNA. Six

of these samples contained both the integrated and replica-

tive forms of HBV DNA. (Table 2). The replicative form of

HBV DNA was also found in the corresponding non-tumour

regions of the six tumour samples that contained the repli-

cative HBV DNA, while the integrated form of HBV DNA was

found in all of the corresponding non-tumorous tissues.

Mutations of the precore/core region of HBVin HCC and non-tumour tissues

The sequence of the precore/core positive samples from 40

HCC and 31 non-tumour tissues were examined by direct

sequencing of the PCR products. The precore/core nucleotide

sequences of the DNA from tumour and non-tumour tissues

and the wild-type HBV genotype D (adr subtype) (GenBank

accession number: M38454) are illustrated in Fig. 1.

Twenty-six out of the 40 HCC (65%) contained mutations at

the precore/core region of HBV DNA, while 24 out of the 40

(60%) had at least a G to A mutation at nucleotides (nts)

1896, 1898 and 1899. Of these sequences, 16 of 24 con-

tained a G to A conversion at nt 1896, creating a stop codon

(TAG) mutation which prevented the synthesis of HBeAg.

Seven of 24 contained the conversion at nt 1898, changing

the corresponding amino acid Gly to Ser. An amino acid

substitution caused by a G to A mutation at position 1899

was also found in eight samples. This mutation occurred

together with the 1896 nonsense mutation in six of the

samples.

In addition, missense mutations were found in one sample

at nt 1882 from T to C, changing Leu to Pro, and in three

samples at nt 1884 (G to A), changing the corresponding

amino acid from Gly to Arg. Other missense mutations were

found in seven samples at nt 1912 (C to A in ®ve and C to G

in two), changing the corresponding amino acid Pro to Thr

and Ala, respectively, and at nt 1913 from C to A in a single

sample, changing the corresponding amino acid from Pro to

His. A silent mutation of C to A was also found at nt position

1914.

Deletions of the core sequence of HBV were found in two

samples of HCC examined. Sample HT14 contained an

additional fragment with lower molecular mass as compared

to the other PCR products. Sequencing data indicated that

this sample contained a deletion of 168 bp from codon 56 to

codon 111 in the core gene. For the other sample (HT8), a

12-bp deletion was found from codon 73±76 in the middle of

the core gene. However, the deletion was in multiples of

three, and was in-frame with the rest of the sequence of the

core gene ORF.

Among the 31 non-tumour tissues analysed, 14 samples

(45.6%) contained G to A mutated precore/core sequence at

nts 1896, 1898 or 1899 (Fig. 1). Mutations were found in

11 samples at nt position 1896, two at nt 1898 and three at

nt 1899. Interestingly, ®ve non-tumour tissues (HN-2, -4,

-22, -24 and SN-8) contained the same mutations as the

corresponding tumour tissues. However, the other 10 non-

tumour samples showed an absence of mutations as com-

pared to their corresponding tumour tissues (Fig. 2 HN-5,

-8, -16, -21, -23, and SN-2, -4, -5, -7 and -17, respectively).

In these 10 tumour tissues, nine out of the 10 had at least

one G to A mutation at nt 1896, 1898 or 1899 (seven of 10

at nt 1896, six of 10 at nt 1899 and two of 10 at nt 1898).

Mutations of the precore/core region of HBVin acute and chronic hepatitis

The frequent presence of the precore/core mutations des-

cribed above might be associated with tumour development

Table 2 Oligonucleotide primers used in this study

Primer Sequence Position Size

Outer primers 9 5¢-TGCCAAGTGTTTGCTGACGC-3¢ 1176±1195 1110 bp

14 5¢-AGTGCGAATCCACACTC-3¢ 2286±2270

Inner primers 11 5¢-CATGGAGACCACCGTGAAC-3¢ 1627±1609 370 bp

12 5¢-AAGGAAAGAAGTCAGAAGG-3¢ 1978±60


118 S. Zhong et al.

or might merely re¯ect the predominant geographical strain

of HBV. Therefore, the precore/core region of HBV DNA from

sera of 18 acute and 35 chronic hepatitis B patients was

analysed. As shown in Fig. 1, only one of 18 (5.6%) serum

samples from patients with acute hepatitis contained the nt

1896 and nt 1899 mutations, while 10 of 35 (28.5%) serum

samples from patients with chronic hepatitis had mutations:

nine (25.6%) with a G to A mutation at nt 1896 and one

(2.8%) with a G to A mutation at nt 1899. However, none of

the samples from patients with acute and chronic hepatitis B

had the G to A mutation at nt 1898.

Our data showed that precore/core mutations, especially

the nt 1896 stop codon, are frequently present in HCC

tissues. Comparison of the frequency of the mutant HBV in

HCC tissues with other non-malignant conditions, such as

acute hepatitis B, chronic hepatitis B and corresponding

non-tumour tissues, showed non-signi®cant differences

between HCC and corresponding non-tumour tissues

(P � 0.18) and chronic hepatitis B (P � 0.17); but a signi-

®cant difference was found between HCC and acute hepatitis

B (P < 0.01) (Table 3).

DISCUSSION

We have analysed the frequency of precore/core mutant

integration in HCC tissues and corresponding non-tumour

1880 1890 1900 1910 1920 1930Mutation * * * * * *pattern Precore Start-C Coreadr(M38454) TGCCTTGGGTGGCTTTGGGGCATGGACATTGACCCGTATAAAGAATTTGGA No.

CysLeuGlyTrpLeuTrpGlyMetAspLleAspProTyrLysGluPheGlyHCC tissues1 --------------------------------------------------- 142 ----------------A---------------------------------- 53 ------------------A-------------------------------- 54 ----------------A--A------------------------------- 35 ----------------A---------------A------------------ 36 ----C-A------------A------------------------------- 17 ------------------A-------------A-----------G------ 18 ----------------A--A------------A------------------ 19 ----------------A--A------------G------------------ 110 ----------------A-A-------------G------------------ 111 ------A---------A--A------------------------------- 112 ---------------------------------A----------------- 113 ----------------A-----------------A---------------- 114 ------A-------------------------------------------- 115 -------------------A------------------------------- 1Non-tumourtissues1 --------------------------------------------------- 172 ----------------A---------------------------------- 83 ----------------A--A--------------- ---------------- 24 ------------------A-------------------------------- 25 -------------------A------------------------------- 2Acutehepatitis1 --------------------------------------------------- 172 ----------------A--A------------------------------- 1Chronichepatitis1 --------------------------------------------------- 252 ----------------A---------------------------------- 73 ----------------A--A------------------------------- 14 ----------------A---------------A------------------ 15 ---------------------------------A----------------- 1

Fig. 1 Hepatitis B virus (HBV) DNA mutation pattern in the precore/core region (1880±1930) of tumour and non-tumour

tissues and sera of acute and chronic hepatitis B patients. The wild-type sequence of genotype D (adr subtype) is shown at the

top. The number of each mutation pattern in this region is shown on the left5 .



tissues. The frequency of such mutants in the sera of

patients, from the same region, with acute and chronic

hepatitis B, were also compared. Our results show that the

precore/core mutants were present in 60% of HCC tissues in

the integrated form. The mutation at nt 1896, which abol-

ishes the synthesis of HBeAg, was the predominant mutant

found in our HCC tissues. An increased incidence of this

mutant was found in HCC tissues compared with non-

malignant lesions, as found in patients with acute and

chronic hepatitis B from the same geographical region,

suggesting that such mutants may be under immune se-

lection and have oncogenic potential.

There have been several reports of nt 1896 mutation of

HBV in HCC [19±25], but only a limited number of cases

have been investigated and the results have been contra-

dictory. Studies from Italy and southern African Blacks

have reported that the pre-C/C 1896 mutants were absent

[25] and present [21] in the HCC tissue, but amino acid

substitution at precore codon 23, changing Cys to other

amino acids, was present exclusively in tumour tissues [25].

However, whether or not the mutant HBV genome was in-

tegrated was not determined in these studies. In accordance

with other investigators of Chinese patients [19,22±24], we

found that this mutant was present in both tumour and non-

tumour tissues in an integrated form. However, we found no

mutation at precore codon 23, as reported previously in

Italian HCC tissues [25].

These discrepancies may re¯ect geographical variation in

HBV genotype. The 1896 mutation occurs frequently in

Chinese populations, where genotype D is prevalent [1]. HBV

genotypes, other than genotype A, are known to contain T at

nts 1858 and 1855, which pairs with mutated A at nts 1896

and 1899, respectively, in the stem±loop structure of the

packaging signal e [25±27]. The nt 1896 and nt 1899

mutations stabilize the stem±loop structure by generating

1880 1890 1900 1910 1920 1930Samples * * * * * *

Precore Start-C coreadr(M38454) TGCCTTGGGTGGCTTTGGGGCATGGACATTGACCCGTATAAAGAATTTGGA

CysLeuGlyTrpLeuTrpGlyMetAspLleAspProTyrLysGluPheGlyHT1 ----------------A---------------------------------- MutantHN1 --------------------------------------------------- Wild-typeHT5 ------A-------------------------------------------- MutantHN5 --------------------------------------------------- Wild-typeHT16 ----------------A--A------------------------------- MutantHN16 --------------------------------------------------- Wild-typeHT21 ----------------A--A------------G------------------ MutantHN21 --------------------------------------------------- Wild-typeHT23 ----------------A-----------------A---------------- MutantHN23 --------------------------------------------------- Wild-typeST2 ----------------A-A-------------G------------------ MutantSN2 --------------------------------------------------- Wild-typeST4 ------A---------A--A------------------------------- MutantSN4 --------------------------------------------------- Wild-typeST5 -------------------A------------------------------- MutantSN5 --------------------------------------------------- Wild-typeST7 ----------------A--A------------------------------- MutantSN7 --------------------------------------------------- Wild-typeST17 ----------------A-----------------A--------------- MutantSN17 --------------------------------------------------- Wild-type

Fig. 2 Nucleotide sequence and deduced amino acid sequence of the precore/core region (1880±1930) of hepatitis B virus

(HBV) DNA. The wild type sequence of genotype D (adr subtype) is shown at the top. The mutant HBV precore/core sequences

were found in 10 tumour (T6 ) tissues, while wild-type sequences of HBV DNA were found in their corresponding non-tumour

tissues.

Table 3 The frequency of precore/core mutations in hepa-

tocellular carcinoma (HCC) and in acute and chronic

hepatitis B infection

Mutant Wild

Groups (nt 1896) type Total

HCC 16 (40)* 24 40

Non-tumour 10 (32.3) 21 31

Acute hepatitis 1 (5.6)* 17 18

Chronic hepatitis 9 (29.7) 26 35

* P < 0.01(v2-test).

nt, nucleotide.


120 S. Zhong et al.

additional matched base pairs of A±T [16±18]. A G to A

mutation at position 1899 was found in our HCC and non-

tumorous tissues and sera of hepatitis B patients. This mu-

tation usually occurred together with the 1896 nonsense

mutation, as seen in our six samples of HCC tissues, four

samples of non-tumour tissues and two samples of sera from

hepatitis B patients, suggesting that such mutant viruses may

have an enhanced replication ability. In accordance with this

view, G to A mutations at either nt 1896 or nt 1899, or both,

were found in six samples of HCC with replicative virus. Such

mutant viruses may therefore, through enhanced replication

frequency, contribute to persistent infection and lead to

an increased incidence of viral integration.

In addition, a G to A mutation at position 1898 was found

in seven of the 40 HCC tissues and in two of the 31 non-

tumour tissues sequenced. However, this mutation was not

found in our control hepatitis B sera. This mutation occurred

on the bulge of e and may destabilize e in all of the HBV

genotypes [25±27]. This mutant may accumulate unen-

capsidated HBV DNA replicative intermediates in cells at an

earlier stage of infection. The accumulation of such forms of

HBV DNA in liver cells might stimulate3 HBV DNA to inte-

grate into the host genome.

Interestingly, of the 10 tumour samples without a pre-

core/core mutation in their corresponding non-tumour tis-

sues, nine out of the 10 HCC had at least one G to A

mutation at nt 1896, 1898 or 1899 (seven at nt 1896, six

at nt 1899 and two at nt 1898). Seven of the 10 tumour

tissues had mutations at nt 1898 or 1899, which does not

lead to HBeAg negativity. We speculate that these mutations

may promote integration. Integrated HBV DNA can result in

genomic instability which, in turn, could result in the

expression of oncogenes or deletion of genes that regulate

hepatocyte growth (suppressor genes) [28].

Like HBcAg, HBeAg has been considered as a target for

both cytotoxic T lymphocyte (CTL) and antibody-dependent

cellular cytotoxicity (ADCC) [29±33]. Abrogation of the

synthesis of HBeAg might thus diminish the immune

response to HBV and thereby account for its `escape' during

the course of chronic infection or the rapid selection of

mutant virus in fulminant hepatitis. However, the patho-

genic role of such a mutant virus is still unclear, as this

mutant is not only associated with severe and progressive

forms of chronic hepatitis and fulminant hepatitis [15,16],

but is also present in asymptomatic carriers [34±36]. There

have also been reports indicating that such virus leads to the

same outcome as wild-type virus [37]. In fact, a recent study

showed that there was no association of the stop codon

mutant with severe liver damage [38]. On the contrary,

HBeAg-negative patients with precore mutations of HBV

that prevent synthesis of HBeAg had less in¯ammation and

®brosis than those with a wild-type precore region [39]. It is

unlikely that the frequent nt 1896 mutant HBV that was

found in our HCC tissues can be explained by predisposing

the patient to more severe liver disease, which in turn leads

to HCC. It is more probable that these mutations evolved

and/or were selected for under immunological pressure. In

this manner, such mutants that escape immunosurveillance

would ensure persistence of the virus during chronic

infection. Alternatively, given that HBeAg is an important

immunological target, these mutations may favour clonal

expansion of the transformed hepatocytes that contain the

precore/core mutants, thereby contributing to oncogenesis.

Consistent with this view, mutations at the precore

promoter, which can down-regulate HBeAg expression, have

also been observed frequently in HCC tissues [24,40±42].

Missense mutations at nts 1912 and 1913 in the core

gene were found in HCC tissues and sera from chronic

hepatitis B patients. Moreover, deletions were also found in

the core region of HBV in two of our samples, located at

codons 56±1114 and 73±76 of the HBcAg, respectively,

con®rming previous reports that a hot-spot deletion occurred

within the core gene in chronic carriers [43,44], which

included the critical region for antibody binding [29]. Given

that HBcAg is an important immunological target during

chronic infection, it is probable that these mutations were

also selected for under immunological pressure and may

favour the clonal expansion of the transformed hepatocytes

containing the core gene mutants.

In conclusion, the integration of precore/core mutant

viruses was found in HCC tissues and was signi®cantly more

frequent than in acute hepatitis B, suggesting that such

mutants may be under immune selection and undergo

preferential integration. The fact that the mutation at nt

1898 was found exclusively in HCC and non-tumour tissues

suggests that it may also facilitate integration of the HBV

genome into the cellular genome.

ACKNOWLEDGEMENTS

This research was supported by The Chinese University of

Hong Kong (CUHK Direct Grant 2040520) and a postdoc-

toral fellowship of the CUHK to S. Zhong. We are indebted to

Dr Yuan Shao (Zhongshan Hospital) and Tsai-Ping Li

(Shanghai Institute of Biochemistry, Shanghai, China) for

the donation of HCC samples and the HBV probe used in this

study. Thanks are also extended to Dr Stephen Ho for his

assistance in obtaining clinical data. We are indebted to Dr

William F. Carman (Institute of Virology, University of

Glasgow, Glasgow, UK) for his advice and criticism during

the preparation of this manuscript.

REFERENCES

1 Kremsdof D, Brechot C. Genetic variability of hepatitis B

virus: potential clinical implications. In: Brechot C, eds. Pri-

mary Liver Cancer: Etiological and Progression Factors. London:

CRC Press, 1994: 149±123.

2 Johnson PJ. Hepatitis B virus and primary liver cancer. In:

Brechot C, eds. Primary Liver Cancer: Etiological and Progres-

sion Factors. London: CRC Press, 1994: 31±38.



3 Fourel G, Tiollais PG. Molecular biology of hepatitis B virus.

In: Brechot C, eds. Primary Liver Cancer: Etiological and Pro-

gression Factors. London: CRC Press, 1994: 89±124.

4 Robinson WS. The role of hepatitis B virus in the develop-

ment of primary hepatocellular carcinoma. J Gastroenterol

Hepatol 1992; 7: 622±638.

5 Kidd-Ljunggren K. Variability in hepatitis B virus DNA:

phylogenetic, epidemiological and clinical implications.

Scand J Infect Dis 1996; 28: 111±116.

6 Brechot C, Pourcel C, Louise A, Rain B, Tiollais P. Presence of

integrated hepatitis B virus DNA sequences in cellular DNA

of human hepatocellular carcinoma. Nature 1980; 286:

533±535.

7 Twu JS, Schloemer RH. Transcriptional trans-activating

function of hepatitis B virus. J Virol 1987; 61: 3448±3453.

8 Wollersheim M, Debelka U, Hofschneider PH. A transacti-

vating function encoded in the hepatitis B virus X gene is

conserved in the integrated state. Oncogene 1988; 3: 545±

552.

9 Zahm P, Hofschneider PH, Koshy R. The HBV X-ORF encodes

a trans-activator: a potential factor in viral hepatocarcino-

genesis. Oncogene 1988; 3: 169±177.

10 Caselmann WH, Meyer M, Kekule AS, Lauer U, Hofschneider

PH, Koshy R. A trans-activator function is generated by

integration of hepatitis B virus preS/S sequences in human

hepatocellular carcinoma DNA. Proc Natl Acad Sci USA

1990; 87: 2970±2972.

11 Kekule AS, Lauer U, Meyer M, Caselmann WH. The preS2/S

region of integrated hepatitis B virus DNA encodes a tran-

scriptional transactivator. Nature 1990; 343: 457±461.

12 Lauer U, Weiss L, Hofschneider PH, Kekule AS. The hepatitis

B virus preS/St transactivator is generated by 3¢ truncations

within a de®ned region of the S gene. J Virol 1992; 66:

5284±5289.

13 Shafritz DA, Shouval D, Sherman HI, Hadziyannis SJ, Kew

MC. Integration of hepatitis B virus DNA into the genome of

liver cells in chronic liver disease and hepatocellular carci-

noma: studies in percutaneous liver biopsies and post-

mortem tissue specimens. N Engl J Med 1981; 305:

1067±1073.

14 Harrison TJ, Anderson MG, Murray-Lyon IM, Zuckerman AJ.

Hepatitis B virus DNA in the hepatocyte: a series of 160

biopsies. J Hepatol 1986; 2: 1±10.

15 Carman WF, Fagan EA, Hadziyannis S et al. Association of a

pre-core variant of hepatitis B virus with fulminant hepatitis.

Hepatology 1991; 14: 219±222.

16 Kosaka Y, Takase K, Kojima M. Fulminant hepatitis B:

induction by hepatitis B virus mutants defective in the

precore region and incapable of encoding e antigen. Gast-

roenterology 1991; 324: 1087±1094.

17 Lee YI, Hur GM, Suh DJ, Kim SH. Novel pre-C/C gene

mutants of hepatitis B virus in chronic active hepatitis:

naturally occurring escape mutants. J Gen Virol 1996; 77:

1129±1138.

18 Hadziyannis S, Bramou T, Alexopoulou A, Makris A.

Immuno-pathogenesis and natural course of anti-HBe-posi-

tive chronic hepatitis with replicating B virus. In: Hollinger

SF, Lemon BM, Margolis HS, eds. Viral Hepatitis and Liver

Disease. Baltimore: Williams & Wilkins, 1991: 673±676.

19 Loncarevic IF, Schranz P, Zentgraf H et al. Replication of

hepatitis B virus in a hepatocellular carcinoma. Virology

1990; 174: 158±168.

20 Manzin A, Menzo S, Bagnarelli P et al. Sequence analysis of

the hepatitis B virus preC region in hepatocellular carcinoma

(HCC) and non-tumor liver tissues from HCC patients.

Virology 1992; 188: 890±895.

21 Clementi M, Manzin A, Paolucci S et al. Hepatitis B virus

preC mutants in human hepatocellular carcinoma tissues.

Res Virol 1993; 144: 297±301.

22 Hosono S, Tai PC, Wong W et al. Core antigen mutations

of hepatitis B virus in hepatomas accumulate in MHC

class II-restricted T cell epitope. Virology 1995; 212: 151±

162.

23 Minami M, Poussin K, Kew M, Okanoue T, Brechot C, Pat-

erlini P. Precore/core mutations of hepatitis B virus in

hepatocellular carcinomas developed on noncirrhotic liver.

Gastroenterology 1996; 111: 691±700.

24 Fang ZL, Ling R, Wang SS, Nong J, Huang CS, Harrison TJ.

HBV core promoter mutations prevail in patients with

hepatocellular carcinoma from Guangxi, China. J Med

Virol 1998; 56: 18±24.

25 Tong SP, Li JS, Vitvitski L, Trepo C. Replication capacities of

natural and arti®cial precore stop codon mutants of hepatitis

B virus: relevance of pregenome encapsidation signal.

Virology 1992; 191: 237±245.

26 Tong SP, Lu JS, Vitvitski L, Kay A, Trepo C. Evidence for a

base-paired region of hepatitis B virus pregenome encapsi-

dation signal that in¯uences the patterns of precore muta-

tions abolishing HBe protein expression. J Virol 1993; 67:

5651±5655.

27 Lok ASF, Akarca U, Greene S. Mutations in the pre-core

region of the hepatitis B virus serve to enhance the stability

of the secondary structure of the pre-genome encapsidation

signal. Proc Natl Acad Sci USA 1994; 91: 4077±4081.

28 Talor E. Tumor suppressor genes, growth factor, and onco-

genes in hepatitis B virus-associated hepatocellular carcino-

ma. J Med Virol 1994; 42: 357±365.

29 Milich DR, Penna A, Mclachlan A, Moriarty A, Thornton GB.

Immune response to hepatitis B virus core antigen (HBcAg):

localization of T cell recognition sites within HBcAg/HBeAg.

J Immunol 1987; 139: 1223±1231.

30 Missale G, Redeker A, Person J et al. HLA-A31 and HLA-

Aw68 restricted cytotoxic T cell responses to a single hepa-

titis B virus nucleocapsid epitope during acute viral hepatitis.

J Exp Med 1993; 177: 751±762.

31 Moriyama K, Nakajima E, Hohioh H, Asayama R, Okochi K.

Immuno-selected hepatitis B virus mutant. Lancet 1991;

337: 125±128.

32 Penna A, Chisari FV, Bertoletti A. Cytotoxic T lymphocytes

recognise an HLA-A2 restricted epitope within the hepatitis

B virus nucleocapsid antigen. J Exp Med 1991; 174: 1565±

1572.

33 Pircher H, Moskophidis D, Rohrer U, Burki K, Hengartner

H, Zinkernagle RM. Viral escape by selection of cytotoxic T

cell-resistant virus variants in vivo. Nature 1990; 346: 629±

633.

34 Scaglioni PP, Melegari M, Wands JR. Biologic properties of

hepatitis B viral genomes with mutations in the precore


122 S. Zhong et al.

promoter and precore open reading frame. Virology 1997;

233: 374±381.

35 Tur-Kaspa R, Klein A, Aharanson S. HBV precore mutants

are identical in carriers from various ethnic origins and

associated with a range liver disease severity. Hepatology

1992; 16: 1338±1342.

36 Lai ME, Solinas A, Mazzoleni AP et al. The role of hepatitis B

virus mutants on the long term outcome of chronic hepatitis

B virus hepatitis: a longitudinal study. J Hepatol 1994; 20:

773±781.

37 Akarca US, Greene S, Lok AS. Detection of precore hepatitis B

virus mutants in asymptomatic HBsAg-positive family

members. Hepatology 1994; 19: 1366±1370.

38 Mphahlele MJ, Shattock AG, Boner W, Quinn J, McCormick

PA, Carman WF. Transmission of homogeneous hepatitis B

virus population of A1896-containing strains leading to mild

resolving acute hepatitis and seroconversion to hepatitis B e

antigen antibodies in a adult. Hepatology 1997; 26: 743±746.

39 Lindh M, Horal P, Dhillon AP, Furuta Y, Norkrans G. Hep-

atitis B virus carriers without precore mutations in hepatitis

B e antigen-negative stage show more severe liver damage.

Hepatology 1996; 24: 494±501.

40 Okamoto H, Tsuda F, Akahane Y et al. Hepatitis B virus with

mutations in the core promoter for an e antigen-negative

phenotype in carriers with antibody to e antigen. J Virol

1994; 71: 1058±1071.

41 Hsia CC, Yuwen H, Tabor E. Hot-spot mutations in hepatitis

B virus X gene in hepatocellular carcinoma. Lancet 1996;

348: 625±626.

42 Gunther S, Pivon N, Will H. Wild-type levels of pregenomic

RNA and replication but reduced pre-C RNA and e-antigen

synthesis of hepatitis B virus with C (1653) to T, A (1762) to

T and G (1764) to A mutations in the core promoter. J Gen

Virol 1998; 79: 375±380.

43 Okamoto H, Tsuda F, Mayumi M. Defective mutants of

hepatitis B virus in the circulation of symptom-free carriers.

Jpn J Exp Med 1987; 57: 217±221.

44 Wakita T, Kakumu S, Shibata M et al. Detection of pre-C and

core region mutants of hepatitis B virus in chronic hepatitis

B virus carriers. J Clin Invest 1991; 88: 1793±1801.



frequent integration of precore/core mutants of hepatitis b virus in human hepatocellular carcinoma...

Documents