[CANCER RESEARCH 54, 3971-3973,
August 1, 1994]
Advances in Brief
Determination of a Putative Recombinogenic Human Hepatitis B Virus Sequence
and Its Binding Cellular Protein'
Kazunori
Kajino,
Yasuo
Hotta,
and Okio Hino2
Department ofExperimental Pathology, Cancer Jastitute@1-37-1, Kami4kebukuro, Toshima-ku@Tokyo 170 (K K, 0. H.], and Department ofBiology, School ofScience, Nagoya
University, Nagoya 464 /1'. H.], Japan
a 60-base pair fragment of HBV DNA (nt.1855-1915)
Abstract
Previously, we reported that C4Bg111196,a 196-base pair subgenomlc
fragment
@
ofhepatitis
B virus (HBV) covering its precore region, enhances
in vitm recombination in the presence of extracts from actively dividing
cells (Hino, 0., etal. Proc. NatLAcad. Sd, USA, 88: 9248-9252, 1991). The
results indicated that HBV may play some role in causing genomic insta
blilty during chronic hepatitis. In the present study, we showed that 15AB,
a 60-base
subgenomlc fragment of HBV DNA (nudeotides 1855-
1914) wIthinC4Bgllll% Is indispensablefor enhancementof in vitm
recombinatlon, using the mouse leukemia cell 70713, as the cellular extract
source. 15AB, thought to be the encapsidatlon signal of HBV pregenomic
RNA and US-like retrovirus long terminal repeat, was found to bind
specifically to an approximately 100 kDa protein of7OZI3 by southwestern
blotting.
Production
of a mutation
in the 1MB region decreased
both its
binding activity to 100 kDa protein and the in vitro recombinatlon activity.
Our present results thus suggest that 15AB might be a recombinogenic
is indispensa
ble for in vitro recombination and that it specifically binds to a 100
kDa cellular protein.
Materials
and Methods
In a previousstudy, recombinationactivity was detectedonly in the pres
ence ofcellular proteins extracted from proliferating cells, such as human HCC
tissue, melanoma cells, mouse spleen, or bone marrow cells (2); therefore for
the present study we used a marine B-cell leukemia line, 70Z/3 (3), as the
source of cellular extract because of its easy handling. DNA-binding proteins
were extracted from 70713 by the method reported previously (2, 4).
C4Bg1II196, a subgenomic fragment of HBV DNA covering the precore and
core region, was previously shown to enhance in vitro recombination (2). We
divided C4Bg111196into 5 smallerparts,consisting of 4 differentsequences,
designated15AB,17AB,18AB,and 2OAB(Fig. 1), preparedby synthesisof
oligomers and annealing.In a mutatedfragmentnamed 19AB, GCI'Gcorre
sequenceand the 100-kDaprotein may be a putativerecombinogenic spendingto nt. 1876—1879
in 15ABwerechangedto AAAA.Thosefragments
protein in eukaryotes, triggering genomicinstability and facilitatingcar
were used for protein-bindingor in vitro recombinationassays(2).
cinogenesis.
Gel Shift Many. The reaction mixture contained the following in a total
volume of 12.5 @.tl:
0.5—1.0
ng of 5'-end-IabeledDNA fragments;3—5@g
of
701/3 extract proteins, 80 p@g/mldouble-strand poly(deoxyinosinate-deoxy
Introduction
Epidemiologically it is clear that HCC,3 regardless of the etiology,
develops
at high incidence
in patients
suffering
from chronic
active
liver diseases, where death and regeneration of hepatocytes occur
cytidylate); 20 mM Hepes (pH 7.6); 0.1 mM EDTA; 1 maa MgC12; 40 mM Kcl;
and 0.5 mM Dli'. Incubation was at room temperature for 30 min, and the
reaction products were run on 6% nondenaturing polyacrylamide gels, which
were driedand autoradiographed.
repeatedly (1). This suggests that hepatitis-related proliferative change
Southwestern Blot Analysis. Ten @g
of 70Z/3 extract were separated on
may itself have some role in hepatocarcinogenesis. The accumulation
of DNA damage, which is likely to occur during continuous cycles of
cell division, may eventually transform some hepatocytes through a
multistage process involving mutations of oncogenes and/or tumor
8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred
to nitrocellulose (Schleicher & Schuell; BA85) by electroblotting. Immobi
lized proteins were incubated at room temperature for 3 h, with 1.3 or 2.3 @M
suppressor
genes.
Previously we showed that a subgenomic fragment of HBV DNA
increases in vitro recombination events in the presence of extracts
from actively dividing cells (2). This result supports the notion that, in
some instances, integrated HBV DNAS can cause genomic instability,
possibly involving specific cellular protein(s). The fact that extracts
from nondividing, normal liver did not increase recombination events
indicates that genomic instability may depend upon active cell pro
liferation, a prominent feature of chronic active liver diseases. Deter
mination of recombinogenic
HBV DNA sequence(s)
and isolation of
HBV DNA-binding
cellular protein(s)
involved
in recombination
therefore offers a potential new approach to analyzing genomic insta
concentrationsof 5'-end labeled,double-strandDNAprobesin bindingbuffer
[10 m@Hepes (pH 8.0)-SOmMNaCI-lO [email protected]]2-0.1sass EDTA-1 mM
DTf.0.25% skim milk (Difco)] and 100 gig/mi salmon sperm DNA of which
one-halfhad been denaturedby boiling.Then the filters were washedwith
binding buffer supplemented with 0.2 M KCI, air-dried, and autoradiographed.
In Vitro Recombination Assay. This was carried out as reported previ
ously (2). Briefly, two tetracycline-sensitive(1@cS)
pBR322 derivatives,mut 1
and mut 2, were preparedby introducinga mutationinto EcoRV (nt. 185) or
XmaIII (nt. 939) sites of pBR322, respectively. DNA fragments, 15AB or
19AB, were inserted into the EcoRI site of mut 1 and mut 2 in the same
orientation.Thereactionmixture,containing70713cellextract,andmut 1 and
mut 2, having the insert, were incubated in 35 mM Hepes (pH 7.8)-b mM
MgCI2-1 fliM D1T-2 mM spermidine-5 mM ATP-10% polyvinyl alcohol at
biity of cancer cells. In the present study, using the mouse leukemia
37°Cfor 60 min. DNA was purified by phenol/chloroformextraction and
ethanol precipitationand then used to transformEscherichia coli (recA@,
DH1). The ratio of tetracycline-resistantcolonies over ampicillin-resistant
cell 70713 (3) as a source of proliferating cell extract, we showed that
colonies(Tcr/Ampr)was assessedas the recombinationfrequency.
Results
Received 4/8/94; accepted 6/15/94.
The costs of publication of this article were defrayed in part by the payment of page
charges.Thisarticlemustthereforebe herebymarkedadvertisementin accordancewith
18U.S.C.Section1734solelyto indicatethisfact.
1 This
work
was
supported
in
part
by
a
Grant-in
Aid
for
Cancer
Research
from
the
Ministry of Education, Science and Culture of Japan.
2 To
whom
requests
for
reprints
should
be
addressed.
3The abbreviationsusedare:HCC, hepatocellularcarcinoma;HBV, hepatitisB virus;
nt, nucleotide;Hepes, 4-(2-hydroxyethyl)-1-piperuzineethanesulfonic
acid;Dli.', dithio
threitol; DR1, direct repeat 1.
Gel shift assay showed that two main bands were shifted by the
binding of 1SAB to 701/3, whereas the other DNA fragments did not
cause any shifts. Excess amounts of nonlabeled 17AB, 18AB, or
2OAB, added as competitors, did not change the intensity of these two
bands, but nonlabeled 15AB efficiently diminished them (Fig. 2A).
Southwestern
blot analysis
also showed
that the 15AB bound to an
3971
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1994 American Association for Cancer Research.
@
@; ;
P1T@ATIVERECOMBINOGENICSEQUENCE AND PROTEIN
Discussion
C4BgAhl96
>1
pro-core
I
@
I
I@
@
@
r'@―i
17AB
2OAB
(nt.1820.1854)
@
core
%7A@@' Theexactreason
forthestrong
epidemiological
linkbetween
persistent HBV infection and HCC is still open to debate, although
I
_•-.•——35
!
bp—@--@--—61bp
@—44bp—
15AB
-@-28 bp-@
U
several mechanisms have been put forward. For example, “trans―
18AB @.
2OAB
(nt.1855-1915)
(nt.1916-1959) (nt.1960-1987)
Fig. 1. The locations of 15AB (61 base pairs (bp)], 17AB (35 base pairs), 18AB (44
base pairs), and 2OAB(28 base pairs) in C4Bg(11196.C4Bg111196has an inverted repeat
structure at both ends. Large arrows, coding regions for precore and core protein of HBV;
Hatched area, the 28-base pair inverted repeat fragment (2OAB) of C4BgI11196;at,
nucleotide numbers of HBV DNA reported previously (8); DRI, an 11-base pair direct
repeatsequence1 of HBVDNA(nt. 1824—1834).
activation of cellular growth-controlling genes by viral proteins, such
as the X gene protein, has been suggested to have some role in
hepatocarcinogenesis (5). There is, however, no evidence to prove a
direct relation between X gene expression and the occurrence of
human HCC. It is improbable that tumors arise from activation )f
known cellular oncogenes, unlike in the woodchuck HCC case (6).
Previously we proposed that HBV can cause an increase in the
incidence
of HCC
by a combination
of two mechanisms:
(a)
cell
killing and stimulation of mitosis that facilitates accumulation of the
approximately 100-kDa protein in the 70713 extract, while no signals
were evident with the other fragments (Fig. 18). These results sug
gested that only 15AB, covering from nt. 1855 to fit. 1915 of the HBV
genome, is able to bind specifically to 70713 protein.
To confirm whether 15AB enhances recombination activity through
the specific binding to 70Z/3 protein, we made the mutant fragment
19AB, identical to 15AB except for the 4-base pair replacement in nt.
1876—1879. When both 15AB and 19AB were used for in vitro
recombination
recombination
assay (Table 1), 15AB showed significantly
higher
activity (a 10-fold increase) as compared
to control,
multiplicity
of events necessary
for transformation;
and (b) increase
of chromosomal instability by insertion of HBV DNAS, apparently
mediated by proteins that can stimulate recombination during chronic
hepatitis (2, 7, 8). This twin mechanism would explain why among
HBV carriers, HCCS usually develop only in patients with chronic
hepatitis and/or cirrhosis.
C4Bg1II196 covers the precore and core region of HBV and has an
inverted repeat structure at both ends (Fig. 1). It includes the DR1,
packaging signal (9) of HBV, and the US-like sequence of retroviral
long terminal repeat (10). The US-like region nucleotide sequence is
whereas 19AB gave only slight increase. Although absolute values of
these increases were rather small, they were reproducible and signif
Table 1 In vitro recombinationactivity
andMethods―).1st2ndAv.1.3
The assay was repeatedandresultswerereproducible.(see“Materials
icant statistically (P < 0.01: t test). We preliminarily tried a similar
assay using extract from mouse spleen cells and obtained essentially
the same results (data not shown). Subsequently we checked the
@
binding activity of 19AB to 701/3 extracts. Both gel shift assay and
southwestern blot analysis showed 19AB to have much weaker bind
ing activity to 70Z/3 extracts than 15AB, although the binding was not
abolished completely (Fig. 3). Thus, these binding activities are cor
15AB
X iO@
3.0 X i0@
1.2 X [email protected]
19AB
Negative control°
a mut
1 and mut 2 without
any insert
I
@
I5AB
@--——ext(+)—--—@
II
I@ competitor
zE::@: z@
a,
an
an
J
used for the recombination
B
18AB 2OAB
II
site were
bp<001
A
II
in the EcoRI
Jb) b
assay.
related to the recombinogenic activities.
17AB
X iO@
3.2 x iO@
1.3 X i0@1
x i0@
3.3 x i0@
1.4 x [email protected]
15AB
—@
@:
17AB
18AB
2OAB
200 kD
a
92.5 kD
69 kD
@
S
46 kD
F
30 kD
Fig. 2. Specific binding of 15AB to 70713 extract. A, gel shift assay, in which O3-ng aliquots of5'-end-labeled 15AB, 17AB, 18AB, and 2OAB were used as probes. For the 15AB
probe reaction, 5 ng (0) or 50 ng (S) of nonlabeled DNAS were also added as competitors. ext (+) or ext (-), probe DNA incubated with or without 70713 extract (5 sag), respectively;
Arrowhead, shifted bands due to the specific binding of 15AB to 70713 extract; F, free DNA probe. B, southwestern blot analysis. Immobilized 70713 extract on nitrocellulose was
incubated with 2.3 @M
5'-end-labeled 15AB (4.5 X 10'@cpm), 17AB (3.5 X 1O@
cpm), 18AB (6.0 X 10@cpm), or 2OAB(2.8 X 106 cpm) in a volume of 2 ml. Al), lcDa.
3972
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1994 American Association for Cancer Research.
PUTATIVERECOMBINOGENIC
SEQUENCEAND PROTEIN
A
B
be effective. It should be also noted that the sequence at the top of the
stem and loop structure, 5'-GCTGTGCC-3', shares 6 of 8 nucleotides
with the Clii sequence, 5'-GCTGGTGG-3', which is a hotspot for
4“a,4
I5AB
recombination in E. coli (13).
The drawback in our in vitro recombination assay system is that we
cannot exclude the possibility of a bacterial, nonspecific contribution
to the recombination. However, compatible results supporting conclu
sion of a significance of 15AB were gained from two independent
assays: i.e., 15AB showed specific binding activity and high recom
I9AB
200 kD
bination activity, and both activities were diminished by introducing
the mutation. Several sequences in the hepadnavirus genome are
known to be binding sites for cellular regulatory proteins (14), but
15AB has not been included in those thus far reported.
We conclude that 15AB, covering from nt. 1855 to nt. 1915 of the
HBV genome, binds specifically to a 100-kDa protein of 70713, and
92.5kD
@, ,
.@..
69kD
this is essential for the induction of recombination activity. Thus
15AB might be a recombinogenic
sequence
and the 100-kDa
protein
may be a putative recombinogenic protein in eukaryotes, although at
present it is not clear whether the binding protein is in fact directly
responsible for recombination. Isolation of HBV DNA binding cellu
lar protein(s), which are abundant in dividing cells but not in resting
46 kD
cells, provides an exciting approach to understanding the genomic
instability of cancer cells. Currently we are trying to isolate the
Fig.3. Decreasedbindingactivityof 19ABto 70713.A, gel shift assay.One-ng
aliquots of 5'-end-labeled 15AB (1.5 X UP cpm) and I9AB (1.6 X 1O@cpm) were
recombinogenic protein from a complementary DNA library of 70Z/3
by southwestern screening using 15AB as a probe.
incubated with 3 pg of 70713 extract. B, southwestern blot analysis. Aliquots (1.3 pM)of
5' end-labeled 15AB (5.8 x i0@cpm) or 19AB (8.1 X 106cpm) were used as probes in
a reaction volume of 2 nil.
Acknowledgments
We thankDrs. Alfred G. Knudsonand BaruchS. Blumbergfor interestin
this work. We also thank Des. H. Sugano
most highly conserved in the HBV genome among hepadnaviruses
(10), and although its function is not yet known, it partially overlaps
the HBV packaging signal. The fragment corresponding to the U5-like
sequence was designated as 15AB, and the 5'-end and 3'-end from
for their encour
making available 707.13 cells, Dr. Y. Kikuchi for statistical consultation,
and
Y. Hirayamaand T. Takaharafor their assistance.
References
15AB, excluding the inverted repeat as 17AB or 18AB, respectively
(Fig. 1), for the present
and T. Kitagawa
agement throughthis work. We thank Dr. J. Inoue for his suggestions and
1. Beasley, R. P., Un, C. C., Hwang, L. Y., and Chien, C. S. Hepatocellular carcinoma
and hepatitis B virus. A perspective study of 22,707 men in Taiwan. Lancet, 2:
study.
DR1 is indispensable for replication of HBV (11) but is apparently
not required for in vitro recombination in our system, because 4AB,
the fragment of 15 base pairs covering DR1, did not induce in vitro
recombination (data not shown), and 17AB which includes DR1 did
not bind to 70713. Gordenin et al. (12) earlier reported that the long
1129—1133,
1981.
2. Hino, 0., Tabata, S., and Hotta, Y. Evidence for increased in vitro recombination with
insertion
of human hepatitis
B virus DNA. Proc. Nail. Acad. Sd.
USA, 88:
9248—9252,1991.
3. Paige, C. J., Kincade, P. W., and Ralph, P. Murine B cell leukemia line with inducible
surface immunoglobulin expression. J. Immunol., 121: 641—647,1978.
4. Andrews,N. C., andFaller,D. V. A rapidmicropreparationtechniquefor extraction
inverted repeats can enhance genomic recombination, but the inverted
of DNA-binding proteins from limiting numbers of mammalian cells. Nucleic Acids
repeat structure in C4Bg1II196 proved to be not essential for in vitro
5. Kim, C.-M., Koike, K., Saito, I., Miyamura, T., and Jay, G. HBx gene of hepatitis B
recombination, because, as mentioned in the previous report (2),
another subgenomic fragment of HBV, which contains the whole area
of C4Bg1II196 but without the inverted repeat, also induces recom
bination.
15AB overlaps the packaging signal of HBV pregenome, which is
predicted to take a secondary stem and loop form (9). Although 15AB
is double stranded DNA, it is possible
that it makes cruciform
Rca., 19: 2499, 1991.
virus induces liver cancer in transgenic mice. Nature (Lond.), 351: 317—320,1991.
6. Fourel,0., Trepo, C., Bougueleret,L, Henglein, B., Ponzetto,A., Tiollais, P., and
Buendia, M-A. Frequent activation of N-myc genes by hepadnavirus insertion in
woodchuckliver tumours.Nature(Lond.),347: 294-298, 1990.
7. Hino, 0., Shows, T. B., and Rogler, C. E. Hepatitis B virus integration site in
hepatocellular carcinoma at chromosome
17;18 translocation. Proc. NatI. Aced. Sd.
USA, 83: 8338—8342,1986.
8. Hino, 0., Ohtake, K., and Rogler, C. E. Features oftwo hepatitis B virus(HBV) DNA
integrations suggest mechanisms ofHBV integration. J. Virol., 63: 2638—2643, 1989.
or stem
9. Junker-Niepmann,M., Bartenschlager,R., andSchilar,H. A shortcia-actingsequence
and loop structures in reaction mixtures or in vivo and that this may
work as the binding site for cellular proteins. The fact that replace
10.
ment of 4 nucleotides in 15AB corresponding to the top of the stem
and loop structure from 5'-GCFGTGCC-3' to 5'-AAAATGTC-3' to
11.
make 19AB, resulted in decreased activity in terms of both specific
12.
binding and recombination, suggests that this sequence is indeed
involved in the two processes and that they are linked. However, the
12-base pair fragment, 5'-AAGCTGTGCCTF-3',
covering the “top―13.
was itself not sufficient for specific binding (data not shown), imply
14.
ing that a relatively longer sequence within 15AB is required for it to
is required for hepatitis B virus pregenome encapsidation and sufficient for packaging
of foreign RNA. EMBO, 9: 3389-3396, 1990.
Miller, R. H., Kaneko, S., Clung, C. T., Girones, R., and Purcell, R. H. Compact
organization of the hepatitis B virus genome. Hepatology, 9: 322—327,1989.
Ganem, D., and Varmus, H. E. The molecularbiology of the hepatitis B viruses.
Annu. Rev. Biochem., 56: 651—694,1987.
Gordenin, D. A., Lobachev, K. S., Degtyareva, N. P., Malkova, A. L., Perkins, E., and
Resnick, M. A. Inverted DNA repeats: a source of eukaryotic genomic instability.
Mol. Cell. Biol., 13: 5315—5322,
1993.
Smith, G. R., Kunes,S. M., Schultz, D. W., Taylor,A., andTrinan,K. L Structure
of Chi hotspotsof generalizedrecombination.Cell, 24: 429—436,1981.
Schaller,H., andFischer,M. Transcriptionalcontrolof hepadnavirusgene expression.
Curr. Top. Microbiol. Immunol., 168: 21—39,1991.
3973
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1994 American Association for Cancer Research.
Determination of a Putative Recombinogenic Human Hepatitis B
Virus Sequence and Its Binding Cellular Protein
Kazunori Kajino, Yasuo Hotta and Okio Hino
Cancer Res 1994;54:3971-3973.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/54/15/3971
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1994 American Association for Cancer Research.