Journal of General Virology (2010), 91, 483–492
DOI 10.1099/vir.0.012740-0
Biological significance of amino acid substitutions
in hepatitis B surface antigen (HBsAg) for
glycosylation, secretion, antigenicity and
immunogenicity of HBsAg and hepatitis B virus
replication
Chunchen Wu,1 Xiaoyong Zhang,2 Yongjun Tian,3 Jianhua Song,1
Dongliang Yang,3 Michael Roggendorf,2 Mengji Lu1 and Xinwen Chen2,4
1
Correspondence
State Key Lab of Virology, Wuhan Institute of Virology, Chinese Academy of Sciences, Wuhan,
PR China
Xinwen Chen
[email protected]
Mengji Lu
2
Institute of Virology, University Hospital of Essen, Essen, Germany
3
Division of Clinical Immunology, Tongji Hospital, Huazhong University of Science and Technology,
Wuhan, PR China
[email protected]
4
Department of Microbiology, Tongji Medical College, Huazhong University of Science and
Technology, Wuhan, PR China
Received 20 April 2009
Accepted 7 October 2009
Amino acid substitutions of hepatitis B surface antigen (HBsAg) may affect the antigenicity and
immunogenicity of HBsAg, leading to immune escape and diagnostic failure. The amino acid
positions 122 and 160 are known as determinants for HBsAg subtypes d/y and w/r, respectively.
The substitution K122I has been shown to strongly affect HBsAg antigenicity. In this study, we
investigated the significance of naturally occurring amino acid substitutions K122I, T123N,
A159G and K160N. Both T123N and K160N substitutions resulted in additional N-glycosylated
forms of HBsAg, while the other mutations produced more glycosylated HBsAg compared with
the wild type (wt). Detection of HBsAg by ELISA and immunofluorescence staining indicated that
variant HBsAg (vtHBsAg) with K122I was not recognized by HBsAg immunoassays, while
vtHBsAg with T123N, A159G, K160N and A159G/K160N had reduced antigenicity. DNA
immunization in BALB/c mice revealed that wtHBsAg and vtHBsAg with T123N and K160N are
able to induce antibodies to HBsAg (anti-HBs), whereas K122I and A159G greatly impair the
ability of HBsAg to trigger anti-HBs responses. The cellular immune response to the HBsAg
aa 29–38 epitope was enhanced by the K160N substitution. Using replication competent clones
of hepatitis B virus (HBV), T123N and A159G substitutions were shown to strongly reduce virion
assembly. The amino acid substitution K160N appeared to compensate for the negative effect
of A159G on virion production. These results reveal complex effects of amino acid substitutions
on biochemical properties of HBsAg, on antigenicity and immunogenicity, and on the replication
of HBV.
INTRODUCTION
The three surface antigens of the hepatitis B virus (HBV)
(L-, M- and S-HBsAg) share 226 carboxyl-terminal amino
acid residues. The aa 99–169 central region of S-HBsAg has
been designated as a major hydrophilic region (MHR) and
harbours many B-cell epitopes. In particular, the immunodominant aa 124–147 ‘a’ determinant in HBsAg represents a cluster of epitopes targeted by neutralizing
antibodies. A number of variant HBsAgs with amino acid
Supplementary tables are available with the online version of this paper.
012740 G 2010 SGM
substitutions within or around the ‘a’ determinant have
been found in patients. Some have shown reduced affinity
for monoclonal antibodies (mAbs) to HBsAg (anti-HBs) or
they may escape detection by commercial HBsAg ELISA
assays (Carman, 1997; Carman et al., 1990, 1999;
Cooreman et al., 2001; Gunther et al., 1999; Lu &
Lorentz, 2003; Seddigh-Tonekaboni et al., 2000).
The amino acid residues at positions 122 and 160 of HBsAg
are known as determinants for HBsAg serotypes d/y and w/r,
respectively. HBsAg serotypes d/y and w/r are defined by
amino acid residues Arg/Lys at position 122 and Arg/Lys at
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:51:29
Printed in Great Britain
483
C. Wu and others
position 160 (Kay & Zoulim, 2007; Okamoto et al., 1987).
Substitutions at positions 122 and 160 have been identified
in different cases and were found to be associated with
immune escape or diagnostic failure. Substitutions at the
position 122 occurred in chronically HBV-infected patients
that were tested negative for HBsAg (Alexopoulou et al.,
2004; Grethe et al., 1998; Hou et al., 2001; Weinberger et al.,
2000). Tian et al. (2007) further confirmed that the K122I
substitution had a major influence on HBsAg antigenicity.
HBV isolates with a Lys to Asn substitution at amino acid
position 160 (K160N) were identified in a renal transplant
recipient who developed de novo HBV infection despite preexisting protective anti-HBs (Lu & Lorentz, 2003). The
K160N substitution was accompanied by an Ala to Gly
substitution at position 159 (A159G). The variant HBsAg
isolated from this patient had reduced antigenicity and
immunogenicity. However, the roles of amino acid
substitutions K160N and A159G in antigenicity and
immunogenicity are not clearly defined. Substitutions to
Asn were frequently found in HBsAg, for example, at the
position 123 in a liver transplant patient during the
treatment with anti-HBs immunoglobulin (Carman et al.,
1999) or in patients with atypical serological profiles for
HBV infection (Hou et al., 2001). The recognition of such
variants by anti-HBs appeared to be reduced (Ireland et al.,
2000; Tian et al., 2007). These published data indicate that
such amino acid substitutions may occur in HBV isolates of
different genotypes.
In this study, we investigated the significance of amino acid
substitutions at positions 122, 123, 159, and 160 of HBsAg
for the antigenicity and immunogenicity of HBsAg. We
generated a series of expression constructs of variant
HBsAg (vtHBsAg) with amino acid substitutions A159G,
K160N, and A159G/K160N. The expression of vtHBsAg
was examined by transient transfection in hepatoma cells
and was detected by specific HBsAg immunoassays and
immunofluorescence (IF) staining with anti-HBs antibodies. The ability of vtHBsAg to induce anti-HBs was
determined by in vivo DNA immunization to assess the
influence of single or combined mutations on the
immunogenicity of HBsAg. Finally, replication competent
HBV clones with corresponding amino acid substitutions
were constructed to study the impact of the amino acid
substitutions on HBV replication and viral assembly.
Replication competent clones were transfected into hepatoma cells and intracellular HBV replication intermediates
and virion DNA released in supernatants were detected.
RESULTS
Expression of wild type (wt)- and vtHBsAg in
transfected mammalian cells
Western blot analysis was performed to detect wt- and
vtHBsAg in culture supernatants and cell lysates of
transiently transfected Huh7 cells using a specific antibody
targeting the haemagglutinin (HA)-tag (Fig. 2a, b). The
484
wtHBsAg and vtHBsAgs with K122I, A159G and A159G/
K160N substitutions showed two bands with sizes 29 and
26 kDa that most likely represented HBsAg with and
without glycosylation (Peterson, 1981; Tian et al., 2007).
These proteins had slightly higher molecular masses than
natural HBsAg due to the HA-tag. However, the 29 kDa
band was stronger for vtHBsAgs with amino acid
substitutions K122I, A159G and A159G/K160N than for
wtHBsAg. An additional band with a size of about 32 kDa
was detected for vtHBsAgs with amino acid substitutions
T123N and K160N (Fig. 2a, b). Comparable results were
obtained by using HeLa cells (data not shown).
The amino acid substitutions T123N and K160N may
therefore have created new sites for N-glycosylation that
led to the generation of the 32 kDa band. It has long been
established that the classical N-glycosylation motif is AsnX-Ser/Thr (with X not Pro, Gavel & von Heijne, 1990;
Kaplan et al., 1987; Marshall, 1974; Mellquist et al., 1998).
While the T123N substitution created the sequence AsnCys-Thr corresponding to the N-glycosylation motif, the
K160N substitution resulted in the sequence Asn-Tyr-Leu
that is atypical for a N-glycosylation site. To further
investigate the glycosylation of vtHBsAg, Huh7 cells were
treated with the glycosylation inhibitor tunicamycin after
transfection with the expression vectors. As shown in Fig.
2(c), only the 26 kDa band was detected for wtHBsAg and
all five vtHBsAgs. This indicates that the 29 and 32 kDa
bands arose from glycosylation of 26 kDa HBsAg, as
reported previously (Peterson, 1981; Tian et al., 2007). To
identify the form of glycosylation, proteins obtained from
transfected Huh7 cells were treated with N-glycosidase F
(PNGase F) to remove N-linked glycans or with endoglycosidase H (EndoH) to specifically cleave high mannose and
certain types of hybrid sugars, respectively. Fig. 2(d, e) show
that only the 26 kDa band was detected for wtHBsAg and all
five vtHBsAgs in either PNGase F- or EndoH-treated
samples. These results clearly demonstrate that the amino
acid substitutions T123N and K160N create new sites for Nglycosylation. Consistently, glycosylated forms of HBsAg
were predominant in culture supernatants (Fig. 2a).
Reactivity of wt- and vtHBsAgs using commercial
ELISA kit and ELISAs based on anti-HBs mAbs
It was shown previously and in Fig. 3(a) that wtHBsAg
with an HA-tag was recognized by anti-HBs antibodies and
in HBsAg ELISAs (Tian et al., 2007). Here, wt- and
vtHBsAg were expressed, vtHBsAg with amino acid
substitution K122I was not detectable in culture supernatants and cell lysates. Although vtHBsAgs were expressed
at a comparable level in cell lysates, vtHBsAg K122I was not
reactive in several commercial ELISA kits used so far. A low
amount of vtHBsAg K122I was detected in the cell
supernatant by Western blot analysis (Fig. 2a).
The amino acid substitution K122I strongly impaired the
reactivity of their respective vtHBsAgs in commercial
HBsAg detection assays (Fig. 3a). Similar results were
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:51:29
Journal of General Virology 91
Amino acid substitutions in HBsAg
obtained with culture supernatants and lysates of transfected cells. The wtHBsAg and vtHBsAgs with substitutions
T123N, A159G, K160N and A159G/K160N were clearly
recognized by the commercial ELISA kit, indicating that
the protein was assembled properly and secreted into the
culture medium (Fig. 3a, b).
The vtHBsAgs with substitutions of A159G, K160N and
A159G/K160N had different reactivities to anti-HBs mAbs.
Although reduced reactivity of these vtHBsAgs was
observed in ELISAs based on mAbs S1, 2C8 and D12,
vtHBsAgs with K160N and A159G/K160N were recognized
well by an ELISA based on the 1C10 anti-HBs mAb and
produced a higher S/N ratio when compared with
wtHBsAg (Fig. 3b). Consistent with previous findings, no
reactivity was found for vtHBsAgs with K122I and T123N
substitutions with any mAbs when culture supernatants of
transfected cells were tested (Fig. 3b).
Different subcellular distributions of wt- and
vtHBsAg
The vtHBsAg with the T123N substitution was shown to
accumulate in the perinuclear region by using IF staining
with an anti-HBs mAb (Tian et al., 2007). We refined the IF
staining with the anti-HA mAb to detect all HBsAgs
expressed in transfected cells and examined the subcellular
distribution of wt- and vtHBsAgs (Fig. 4). The wtHBsAg and
all vtHBsAgs including T123N, A159G, K160N and A159G/
K160N were positively stained by anti-HA mAb with
comparable intensities (Fig. 4b, d, f, h, j). This result was
consistent with the previous finding that the amounts of wtand vtHBsAgs produced in transfected cells were comparable (Fig. 2). The staining showed an even distribution
through the cytoplasm of transfected cells for wt- and
vtHBsAgs. The S1 anti-HBs mAb was able to recognize
wtHBsAg and vtHBsAgs with amino acid substitutions
T123N, A159G, K160N and A159G/K160N (Fig. 4a, c, e, g, i).
IF staining of wtHBsAg with the anti-HBs mAb showed a
diffuse intracellular distribution. It is notable that IF
staining with the anti-HBs mAb showed a perinuclear
distribution for vtHBsAg with T123N, consistent with the
previous observation (Tian et al., 2007; Fig. 4c). The
positive IF staining of vtHBsAg T123N with anti-HBs mAb
was probably due to the high antibody concentration used
in the assay, allowing sufficient binding of mAb to
vtHBsAg. Although vtHBsAg with K160N was hyperglycosylated similar to T123N, it showed a diffuse distribution in
cells similar to other vtHBsAgs (Fig. 4g). The vtHBsAg
K122I was detected by using the anti-HA mAb but not the
anti-HBs mAb (data not shown), which is consistent with a
previous report (Tian et al., 2007). Taken together, IF
staining with the anti-HA mAb showed that all HBsAg
variants were expressed in the cytoplasmic compartment.
However, HBsAg variants differed in their detectability by
the anti-HBs mAb recognizing HBsAg ‘a’ determinant. The
fraction of HBsAg detected by the anti-HBs mAb appeared
to be located differently from that detected by the anti-HA,
http://vir.sgmjournals.org
as visible in the case of T123N. This observation may be
explained by the association of folded HBsAg with the
cellular ER/Golgi system. However, an optic effect of
different staining intensities is not completely ruled out.
Induction of anti-HBs antibodies by wt- and
vtHBsAgs
We next asked whether vtHBsAgs with amino acid
substitutions could induce production of antibodies
against HBsAg. Genetic immunizations in mice were
performed to test the ability of wt- and vtHBsAg to induce
anti-HBs antibody responses. Six groups of six mice were
each immunized with plasmids pHBsAg-WT, pHBsAgK122I, pHBsAg-T123N, pHBsAg-A159G, pHBsAg-K160N,
and pHBsAg-A159G/K160N. All six plasmids were able to
induce anti-HBs responses in mice but with different
efficiencies as measured by using an anti-HBs commercial
kit (Fig. 5). Notably, the commercial anti-HBs assay is
primarily designed to detect antibodies specifically against
wtHBsAg and may not recognize such antibodies only
reactive with vtHBsAg. Mice immunized with plasmids
expressing wtHBsAg and vtHBsAgs with the substitutions
T123N and K160N showed an anti-HBs response with an
average titre over 500 IU l21, comparable to previous
studies (Lu & Lorentz, 2003). The ability of vtHBsAgs with
the substitutions K122I, A159G and A159G/K160N to
induce anti-HBs antibody responses against wtHBsAg was
greatly impaired. Although five of six mice developed antiHBs titres after immunization with pHBsAg-K122I, these
mice showed very low antibody titres compared with
pHBsAg-WT (Fig. 5). Only one mouse from each group
showed measurable anti-HBs titres after immunization
with pHBsAg-A159G and pHBsAg-A159G/K160N (Fig. 5).
These results indicate that changes in HBsAg glycosylation
by T123N and K160N substitutions did not impair the
immunogenicity of the corresponding vtHBsAgs. However,
induction of anti-HBs antibodies was not efficient when
K122I and A159G substitutions were present in HBsAg.
Characterization of cellular responses to HBsAg
induced by wt- and vtHBsAgs
The question was raised as to whether the amino acid
substitutions in HBsAg influence the induction of cellmediated immune responses to HBsAg since vtHBsAgs are
biochemically different to wtHBsAg. The cellular immunity
elicited by the different variants was measured by using an
ELISPOT assay 3 weeks after the last immunization. The
major H-2Ld CTL epitope HBsAg aa 29–38 was used to
stimulate antigen-specific gamma interferon (IFN-c) secretion from splenocytes (Zheng et al., 2004). As shown in Fig.
6, all six different plasmids expressing wt- and vtHBsAg
induced specific cellular responses to the H-2Ld CTL epitope
of HBsAg. Notably, the most vigorous responses were
observed in BALB/c mice from the A159G/K160N- and
K160N-injected groups (in A159G/K160N, P,0.001; in
K160N, P,0.01) as compared with wt- and other vtHBsAgs.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:51:29
485
C. Wu and others
Detection of HBV DNA in culture supernatants
and transfected cells
The substitutions K122I, T123N and K160N in S-HBsAg in
the present study led to the substitutions rtQ476H,
rtN477K and rtI515L in HBV polymerase, K285I, T286N
and K323N in L-HBsAg, and K178I, T179N and K215N in
M-HBsAg, respectively. Thus, the impact of amino acid
substitutions on viral replication and assembly was
examined. Encapsidated HBV replicative intermediates
were extracted from Huh7 cells transfected with replication
competent clones and subjected to Southern blot analysis
(Fig. 7a). No apparent differences were noted for HBV
produced with wt- or vtHBsAg. We further quantified the
amount of virion-associated HBV DNA in culture supernatants to analyse virion assembly and secretion. The
amounts of virion-associated HBV DNA from cells
transfected with pHBV1.3-T123N, -A159G, -K160N and A159G/K160N were about 35, 17, 52 and 47 % of that from
cells transfected with pHBV1.3-WT, respectively (Fig. 7b).
Transfection with pHBV1.3-K122I produced the same
amount of virion-associated HBV DNA as transfection
with pHBV1.3-WT, despite the low antigenicity and
immunogenicity of K122I vtHBsAg. These observations
indicate that the HBsAg T123N, A159G and K160N
substitutions examined in this study may influence the
assembly and secretion of the HBV virion, but not the
formation of HBV replication intermediates.
It has been reported that transfection of replication
competent clones into hepatoma cells may lead to the
production and release of naked cores (Parekh et al., 2003).
Therefore, virions and naked cores in culture supernatants
of transfected Huh7 cells were separated by CsCl gradient
ultracentrifugation and their relative proportions were
determined by real-time PCR. The results indicated that
the fractions of naked cores ranged between 1 and 13.1 %
of the corresponding total HBV DNA amounts in supernatants (Supplementary Table S2, available in JGV Online).
DISCUSSION
In the present study, the significance of amino acid
substitutions at positions 122, 123, 159 and 160 of
HBsAg for glycosylation, secretion, antigenicity and
immunogenicity of HBsAg and HBV replication was
investigated. The results are summarized in Table 1.
Fig. 1. Schematic representation of the position and the type of
amino acid substitutions in HBsAg. The small 226 aa HBsAg is
indicated with depiction of aa 122 and 160 as determinants for d/y
and w/r subtype, aa 123 and 159 co-existing with 122 and 160,
respectively, and the positions and the nature of amino acid
substitutions introduced into the wtHBsAg sequence in the
present study. A series of expression vectors of HBsAg variants
with amino acid substitutions (pHBsAg-A159G, pHBsAg-K160N
and pHBsAg-A159G/K160N) was constructed based on
pHBsAg-WT. Mutations were introduced into the wtHBsAg
sequence (subtype adw2) by PCR-based mutagenesis. Two
previously constructed expression vectors pHBsAg-K122I and
pHBsAg-T123N expressing vtHBsAgs with the substitutions
K122I and T123N were also used.
We found that amino acid substitutions K122I, T123N,
A159G and K160N led to altered glycosylation patterns for
HBsAg. Both T123N and K160N substitutions led to the
production of extra 32 kDa products that represent new Nglycosylated forms of HBsAg (Fig. 1). vtHBsAgs with
amino acid substitutions K122I, A159G and A159G/K160N
were detected in their unglycosylated 26 kDa form and also
in an N-glycosylated 29 kDa form; these forms have been
shown to natively exist for wtHBsAg (Peterson, 1981; Tian
et al., 2007). However, the glycosylated 29 kDa form was
present in greater proportions for vtHBsAgs than for
wtHBsAg, indicating that the amino acid substitutions
altered the biochemical properties of HBsAg and facilitated
glycosylation. Interestingly, the A159G substitution prevented the production of the 32 kDa form of HBsAg that
was caused by the K160N substitution. Thus, the coexistence of A159G and K160N may be the result of
Table 1. Summary of characteristics of wtHBsAg and vtHBsAgs in the present study
HBsAg
wt
K122I
T123N
A159G
K160N
A159G/K160N
486
Glycosylation
Antibody recognition
Anti-HBs induction
Replication
Virion formation
Normal
Normal
Additional
Normal
Additional
Normal
Normal
Impaired
Impaired
Reduced
Reduced
Reduced
Normal
Impaired
Normal
Impaired
Normal
Impaired
Normal
Normal
Normal
Normal
Normal
Normal
Normal
Normal
Reduced
Reduced
Slightly reduced
Slightly reduced
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:51:29
Journal of General Virology 91
Amino acid substitutions in HBsAg
Fig. 3. Reactivity of expressed wt- and vtHBsAg to a commercial
kit or to mAbs. The wt- and vtHBsAg were expressed by transient
transfection. The culture supernatants and cell lysates were
collected at 48 h after transfection and used for ELISA. The
culture supernatants and cell lysates were detected with a
commercial kit (Shiye Kehua) (a) and the culture supernatants
were detected for ELISA with four mAbs S1, 2C8, D12 and 1C10
(b). The reactivity was expressed as the ratio of samples to the
negative control (S/N ratio) with S/N¢2.1 as the cut-off value. The
average value of four replicates was calculated and given as the
final reactivity of wt or vtHBsAg. ck, Cells transfected with empty
plasmid as a negative control.
selective pressure for prevention of HBsAg glycosylation at
locations other than aa 146.
Fig. 2. Expression of HBsAg in Huh7 or HeLa cells transfected
with different plasmids. Huh7 cell culture medium (a) and Huh7
cells without (b) or with 4 mg tunicamycin ml”1 (c) were collected
at 48 h after transfection with pHBsAg-WT (WT), pHBsAg-K122I
(K122I), pHBsAg-T123N (T123N), pHBsAg-A159G (A159G),
pHBsAg-K160N (K160N), pHBsAg-A159G/K160N (A159G/
K160N). Cell lysates were prepared and subjected to SDSPAGE. A mouse mAb to the HA-tag was used for detection of
expressed HBsAg. b-Actin was detected as a loading control. ck,
Cells transfected with empty plasmid as a negative control.
Another two aliquots of Huh7 cell lysates without tunicamycin
collected at 48 h after transfection were incubated with 2 ml
PNGase F (d) or 2 ml Endo H (e) at 37 6C overnight.
http://vir.sgmjournals.org
The exact sites of additional glycosylation for T123N and
K160N vtHBsAgs remain unknown, but the amino acid
substitutions at the positions 123 and 160 definitely created
two potential glycosylation sites. The consensus sequence
Asn-X-Ser/Thr is generally required for N-linked oligosaccharides (Gavel & von Heijne, 1990; Kaplan et al., 1987;
Marshall, 1974; Mellquist et al., 1998). The T123N
substitution created the sequence Asn-Cys-Thr, which
should allow glycosylation. However, the sequence AsnTyr-Leu that resulted from the K160N substitution does
not represent a typical N-glycosylation motif, as no such
site was described in the literature. HBV surface proteins
may also be subjected to O-glycosylation based on the
hydroxyl group of Ser/Thr (Schmitt et al., 1999). Thus, the
possibility exists that the extra 32 kDa product, resulting
from the T123N substitution, may represent new Oglycosylated forms of HBsAg. Nevertheless, results of
endoglycosidase digestion supported that both vtHBsAg
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:51:29
487
C. Wu and others
Fig. 4. IF staining of wt- and vtHBsAgs with anti-HBs and anti-HA mAbs. HepG2 cells were transfected with pHBsAg-WT,
pHBsAg-T123N, pHBsAg-A159G, pHBsAg-K160N and pHBsAg-A159G/K160N. At 48 h after transfection, cells were fixed
and stained with the anti-HBs mAb S1 (a, c, e, g, i, k) or anti-HA mAb (b, d, f, h, j, l). The nuclei were stained with Hoechst. The
images (a), (c), (e), (g), (i) and (k) were overlapped by the phase-contrast images to show the intracellular distribution of HBsAg.
Magnification ¾800. ck, Cells transfected with empty plasmid as a negative control.
possess additional N-glycosylation but not O-glycosylation,
consistent with the creation of the corresponding sites.
Further biochemical analysis is necessary to demonstrate
that the new glycosylated 32 kDa form of vtHBsAg is
indeed glycosylated at amino acid positions 123 and 160.
The S gene overlaps the gene for the viral DNA polymerase.
In this study, the substitutions sK122I, sT123N and
sK160N in HBsAg led to the substitutions rtQ476H,
rtN477K and rtI515L in the polymerase. However, further
studies demonstrated that the substitutions T123N and
K160N significantly reduced HBV DNA levels in culture
supernatants of transfected cells but did not change the
intracellular HBV DNA levels, indicating that both
substitutions mainly affect the virion assembly and
secretion of HBV virions, but not the formation of
replication intermediates. Apparently, the mutations on
the amino acid sequence of the polymerase due to the
substitutions in HBsAg had no phenotypic effect. Thus, the
additional glycosylation due to substitutions T123N and
K160N may be responsible for altered assembly.
Unexpectedly, however, the A159G substitution also
impaired virion secretion of HBV, although it did not
influence the secretion of HBsAg. Thus, the A159G
substitution likely altered the large or middle surface
488
protein such that virions could not be properly assembled.
The co-variation of K160N appeared to compensate for the
effect of A159G. This may explain the necessity of the covariation at aa 159 and 160.
The amino acid substitutions K122I and A159G appear to
have major impact on the antigenicity and immunogenicity
of HBsAg. The residue at aa 122 may be especially critical
for the HBsAg conformation and the induction of
antibodies reactive to wtHBsAg with the natural conformation. Certainly, vtHBsAgs may induce antibodies
reactive to themselves but not to wtHBsAg, as shown
previously (Zheng et al., 2004). However, the anti-HBsAg
assay used in this study would not allow us to detect
antibodies that are reactive exclusively to vtHBsAgs but not
wtHBsAg. Thus, the low anti-HBs responses seen for some
variants may be due to a lack of expression/secretion of
vtHBsAg, and/or a lack of recognition of specific antibodies
to vtHBsAg in the anti-HBsAg assay. To solve this issue, a
system allowing proper expression of the variant HBsAgs
would be required. In contrast, vtHBsAgs with substitutions T123N and K160N retained the ability to induce
antibody and cellular responses to wtHBsAg. This is
particularly interesting since the T123N substitution
reduced the antigenicity of HBsAg. Furthermore, the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:51:29
Journal of General Virology 91
Amino acid substitutions in HBsAg
Fig. 5. Anti-HBs responses in mice after DNA immunization with
plasmids expressing wt- and vtHBsAg. BALB/c (H-2Ld) mice were
immunized with pHBsAg-WT (WT), pHBsAg-K122I (K122I),
pHBsAg-T123N (T123N), pHBsAg-A159G (A159G), pHBsAgK160N (K160N) and pHBsAg-A159G/K160N (A159G/K160N),
and sacrificed 3 weeks after the last immunization. ck, Mice
immunized with empty plasmid as a negative control. Sera were
collected to test anti-HBs antibodies. Results are presented as
anti-HBs IgG (IU l”1) level for each mouse in each group. Empty
circles (#) represent IgG measurement of individual mice. Filled
diamonds (X) represent geometric mean values.
A159G substitution also slightly impaired antigenicity and
resulted in an anti-HBs antibody response with altered
specificity. On the other hand, the double substitution of
A159G/K160N more strongly impaired antigenicity and
induced an antibody response similar to that observed with
the A159G substitution. The results indicate that multiple
substitutions may have cumulative effects on the conformation of HBsAg, and that the A159G substitution may
have an important impact on the fine specificity of the
antibody response.
The K160N substitution enhanced the cellular immune
response. Previous studies reported that N-linked glycans
of viral proteins play important roles in modulating the
immune response (Tai et al., 2002; Wei et al., 2003; Zhang
et al., 2004). Thus, alteration of wtHBsAg conformation
due to the K160N substitution and its subsequent effect on
glycosylation may contribute to cell-mediated immune
responses.
The influence of amino acid substitutions on the properties
of HBsAg is not explained at the basis of biochemistry of
amino acid residues yet. However, no sufficient data about
different amino acid substitutions at these particular sites
are available for a reasonable analysis. We introduced a
number of different amino acid substitutions at the
position 145 of HBsAg (unpublished data). The preliminary results indicate that each amino acid substitution
http://vir.sgmjournals.org
Fig. 6. Induction of HBsAg-specific T-cell immune response in
mice by DNA immunization. BALB/c (H-2Ld) mice were immunized
with pHBsAg-WT (WT, n56), pHBsAg-K122I (K122I, n56),
pHBsAg-T123N (T123N, n55), pHBsAg-A159G (A159G, n56),
pHBsAg-K160N (K160N, n55) and pHBsAg-A159G/K160N
(A159G/K160N, n56), respectively; mice were sacrificed
3 weeks after the last immunization. ck, Mice immunized with
empty plasmid as a negative control (n53). Splenocytes from mice
were prepared and stimulated with the major H-2Ld CTL epitope
HBsAg aa 29–38 (10 mg ml”1) for 18 h. IFN-c secreting T cells
were determined by ELISPOT assay. The results are presented as
number of spots per 1¾105 cells.
may have its specific effect on the conformation of HBsAg.
Thus, more experimental data need to be accumulated for
the understanding of the influence of each single amino
acid substitution.
The double mutation was first identified from a renal
transplant recipient who developed de novo HBV infection
despite the presence of a pre-existing protective antibody
(Lu & Lorentz, 2003). Thus, the concurrence of A159G and
K160N in HBsAg may have some biological significance.
Our experiments found that the A159G substitution
reduced the antigenicity and immunogenicity of HBsAg.
However, this substitution also prevented efficient virion
production. The concurrence of K160N compensated for
the negative effect of A159G on virion production and
further reduced the antigenicity of HBsAg. Our results
demonstrate the importance of investigating amino acid
substitutions in HBsAg in the context of HBV replication.
In addition, more attention must be paid to the mutual
influences of concurrent amino acid substitutions in
HBsAg.
METHODS
Construction of plasmids encoding HBsAg with amino acid
substitutions by site-directed mutagenesis. A series of expression
vectors of HBsAg variants with amino acid substitutions at positions
159 or 160 was constructed based on plasmid pHBsAg-WT, which
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:51:29
489
C. Wu and others
K160N and pHBsAg-A159G/K160N, respectively, using primer pairs
XS1/XS2 (Supplementary Table S1) and inserted into the corresponding site in pHBV1.3. These procedures resulted in replication
competent clones pHBV1.3-WT, -K122I, -T123N, -A159G, -K160N
and -A159G/K160N. Correct construction of all plasmids was verified
by sequence analysis.
Transient transfection and detection of HBsAg by Western blot
analysis and ELISA. Expression of HBsAg in Huh7 and HeLa cells
by transient transfection and detection of HBsAg by Western blotting
and ELISA were performed according to Tian et al. (2007). For
Western blotting, a mouse anti-HA mAb (Cell Signaling Technology)
was used as the primary antibody. Detection of the reactivity of wtand vtHBsAg by a commercial HBsAg ELISA kit (Shiye Kehua) was
performed according to the manufacturer’s instructions. In addition,
four mAbs S1, 2C8, D12 and 1C10 generated to HBV particles were
tested against vtHBsAgs. These mAbs were not reactive with
denatured HBsAg and are therefore directed to conformational
epitopes. Their fine specificities to the epitope on HBsAg are not
known.
Tunicamycin treatment. The effects of amino acid substitutions on
Fig. 7. Analysis of HBV DNA level in culture supernatants and
cells. Huh7 cells were transiently transfected with HBV replication
competent clones expressing wt- and vtHBsAg: pHBV1.3-WT
(WT), pHBV1.3-K122I (K122I), pHBV1.3-T123N (T123N),
pHBV1.3-A159G (A159G), pHBV1.3-K160N (K160N) and
pHBV1.3-A159G/K160N (A159G/K160N), respectively. ck,
Cells transfected with empty plasmid as negative control.
Encapsidated HBV replicative intermediates were isolated from
transfected cells at 72 h after transfection and probed with a 32Plabelled full-length HBV genome (a). Virion-associated HBV DNA
was isolated from cell culture supernatants at 72 h after
transfection for real-time PCR analysis to assess secreted HBV
DNA in supernatants. Results are depicted as the per cent of wt
control pHBV1.3-WT (b).
glycosylation were studied via transient transfection using plasmids
expressing wt- and vtHBsAg. Transfected cells were treated with
tunicamycin (Sigma), an inhibitor of N-linked glycosylation (Shi &
Elliott, 2004). Tunicamycin was dissolved in DMSO at a concentration of 1 mg ml21 and then added to cells at 6 h after transfection at
a final concentration of 4 mg ml21. An equivalent concentration of
DMSO was included in control cultures. The cultures were
maintained for a further 48 h until cells were assayed.
Endoglycosidase digestion. PNGase F and Endo H were purchased
from New England Biolabs. These enzymes were used to remove Nlinked glycans from glycoproteins according to the manufacturer’s
protocols. Lysates of transfected cells were prepared by freeze–thaw
cycles. After adding Glycoprotein Denaturing Buffer provided with
the enzymes, cell lysates were heated to 100 uC for 10 min. Then, G7
Reaction Buffer, NP-40 and PNGase F were added and the reaction
mix was incubated overnight at 37 uC followed by Western blot
analysis using the anti-HA antibodies. Alternatively, G5 Reaction
Buffer and Endo H were added and processed accordingly.
IF staining of transfected cells. IF staining of transfected cells was
was modified from pXF3H (a kind gift from Feng Xinghua, Baylor
College of Medicine, Houston, TX, USA) with a wtHBsAg gene
[nt 160–841 according to the HBV genome sequence (GenBank
accession no. AF282918, adw)] fused with an amino-terminal HA-tag
and driven by a cytomegalovirus immediate-early promoter as
described previously (Tian et al., 2007). Mutations were introduced
into the wtHBsAg sequence by PCR-based mutagenesis using the
primer pairs SP1/A159GR or K160NR and SP2/A159GL or K160NL
(listed in Supplementary Table S1, available in JGV Online) as
described previously (Tian et al., 2007). Plasmids containing mutated
sequences were identified by sequence analysis. Three new plasmids
(pHBsAg-A159G, pHBsAg-K160N and pHBsAg-A159G/K160N) were
produced (Fig. 1a). Two expression vectors pHBsAg-K122I and
pHBsAg-T123N for vtHBsAg with K122I and T123N substitutions
(Tian et al., 2007) were also used in this study. Sequence analysis
excluded the presence of unintended mutations in the plasmids used
in this study.
For construction of the replication-competent plasmid with amino
acid substitutions, the wt HBV 1.3-fold over length plasmid pHBV1.3
(Lei et al., 2006) was used as a backbone. The region nt 252–689
containing amino acid substitutions was amplified from pHBsAgWT, pHBsAg-K122I, pHBsAg-T123N, pHBsAg-A159G, pHBsAg490
performed as described previously (Wang et al., 2008). Transfected
cells were fixed with 3.7 % paraformaldehyde and permeabilized in
0.2 % Triton X-100. Then the cells were incubated with a mouse antiHA mAb (Cell Signaling Technology) or anti-HBs mAb S1 as primary
antibodies and fluorescein isothiocyanate-labelled secondary antibodies (Novagen) for laser confocal scanning microscopy after cell
nuclei were stained with Hochest.
DNA immunization. Adult female BALB/c mice (6–8-weeks-old)
were kept under standard pathogen-free conditions in the Central
Animal Laboratory of Wuhan Institute of Virology, Chinese Academy
of Sciences and treated following the guidance of animal ethical
standard. In vivo electroporation was performed according to a
previously described protocol (Chen et al., 2005; Widera et al., 2000).
Mice were immunized with 30 mg plasmid DNA dissolved in 30 ml
PBS. Control mice were immunized with 30 ml PBS containing 30 mg
vector DNA without inserts, or they received 30 ml PBS alone. A pair
of electrode needles was inserted into the muscle 5 mm apart to cover
the DNA injection site, and electric pulses were delivered using an
electric pulse generator (Electro Square Porator T830M; BTX)
following injection into the right quadricep muscle. Three pulses of
100 V each, followed by three pulses of the opposite polarity, were
delivered into each injection site at a rate of one pulse per second.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:51:29
Journal of General Virology 91
Amino acid substitutions in HBsAg
Each pulse lasted for 50 ms. DNA vaccination was repeated after 3
and 6 weeks using the same procedure.
proportions of HBV DNA in separated fractions were determined
by real-time PCR.
Measurement of anti-HBs antibody responses. Three weeks after
Detection of encapsidated HBV replicative intermediates.
the last immunization, blood samples were collected from anaesthetized animals and sera were recovered by centrifugation. Anti-HBs
antibodies in sera were detected using a commercial enzyme
immunoassay AUSAB kit (Abbott Laboratory). Tests were performed
by a full automatic Architect i2000 (Abbott Laboratory) following the
manufacturer’s instructions. The amount of anti-HBs antibody was
expressed as IU l21.
Replication competent HBV clones were transfected into Huh7 cells
by using lipofectamine 2000. Encapsidated HBV replicative intermediates were isolated from transfected cells at 72 h after transfection
and probed with a 32P-labelled full-length HBV genome. Analysis of
HBV replicative intermediates was performed as described by Meng
et al. (2008). Briefly, cells were lysed in lysis buffer (50 mM Tris/HCl,
pH 7.4, 1 mM EDTA, 1 % NP-40). Nuclei were pelleted by
centrifugation. The supernatant was adjusted to 10 mM MgCl2 and
incubated for 2 h at 37 uC with 100 mg DNase I ml21 and EDTA at a
final concentration of 25 mM was added to stop the reaction. The
supernatant after centrifugation was treated with 0.5 mg proteinase K
ml21 and 1 % SDS for 2 h at 55 uC. Finally, the sample was extracted
with phenol/chloroform followed by 2-propanol precipitation. The
purified DNA was subjected to Southern blot analysis. DNA samples
were probed with a 32P-labelled full-length HBV genome.
ELISPOT assay. The ELISPOT assay was carried out using a Mouse
IFN-c Secreting Cell kit (U-Cytech) according to the manufacturer’s
instructions. In brief, 16105 mice splenocytes plus 5 mg ConA
(Sigma) ml21 and 0 or 10 mg HBsAg peptide (H-2Ld CTL epitope aa
29–38) ml21 were added to each well of 96-well plates coated with
anti-IFN-c mAb and incubated for 18 h at 37 uC in 5 % CO2. Then
biotinylated anti-IFN-c antibody, GABA solution and activator I/II
solution were added one by one. Finally, spots were counted using an
automated ELISPOT plate reader (BioReader 4000) and expressed as
the number of spots per 16105 cells.
Detection of virion-associated HBV DNA in culture supernatants. Replication competent HBV clones were transfected into
Huh7 cells by using lipofectamine 2000. Virion-associated HBV DNA
was isolated from cell culture supernatants at 72 h after transfection
for real-time PCR analysis to assess secreted HBV DNA in
supernatants. Virion-associated HBV DNA was extracted from cellculture supernatants as described previously (Karayiannis et al.,
1995). In brief, the clarified supernatants obtained after low-speed
centrifugation were treated with 100 mg DNase I (Promega) ml21 to
remove free DNA, then incubated for 10 min at 65 uC in the presence
of stop solution (Promega) to heat-inactivate the enzyme. Then, an
equal volume of 20 % PEG 8000 was added to supernatants to
precipitate the virions, followed by digestion with 20 mg proteinase K
ml21. Finally, HBV DNA was extracted with phenol/chloroform
followed by 2-propanol precipitation and resuspended in TE buffer at
pH 8.0 (10 mM Tris/HCl pH 7.5, 1 mM EDTA).
A SYBR Green Mix kit (Toyobo) was used for quantitative PCR
according to the manufacturer’s instructions with primer pairs F1/R1
(Supplementary Table S1). PCR parameters were as follows: 10 s at
95 uC, and then 38 cycles of 5 s at 95 uC, 15 s at 61 uC and 15 s at
72 uC. PCR specificity was tested by adding a 2 s data-acquisition step
at 82 uC. Additionally, PCR products were loaded onto 1.5 % agarose
gels for electrophoresis in order to further confirm the presence of
specific bands of the correct size.
The relative proportions of HBV virions and naked cores in cell
culture supernatants were determined. In brief, the clarified supernatants obtained after low-speed centrifugation were subjected to
CsCl gradient ultracentrifugation to further separate Dane particles
from naked core particles as previously described (Parekh et al.,
2003). The culture supernatants were concentrated approximately 30fold by using Centricon centrifugal filter devices with a 100 kDa cutoff (Amicon; Millipore) and centrifugation at 4 uC and 5000 r.p.m
(Eppendorf F-34-6-38 rotor). The concentrated samples were loaded
on to a CsCl gradient and spun at 15 uC and 35 000 r.p.m. for 24 h or
longer in a Beckman SW41 rotor. Fractions of 200 ml were taken from
the top, weighed to determine densities and used for DNA extraction.
The fractions with a density less and higher than 1.25 g ml21 were
separated and collected for analysis of virion-associated and naked
core-associated DNA. After treatment with 100 mg DNase I
(Promega) ml21 to remove free DNA, the solution was subjected to
extraction of HBV genome DNA by proteinase K digestion and
phenol/chloroform extraction, as described above. The relative
http://vir.sgmjournals.org
ACKNOWLEDGEMENTS
We are grateful to Mr Bin Yan for helping us to prepare high quality
antibodies. We are indebted to Ms. Xuefang An and Ms Jiaxin Liu for
technical support in animal experiments and confocal microscopy,
respectively. We also thank Ms Yuan Zhou for her excellent work in
cell culture. This work was supported by the National Basic Research
Priorities Program of China (2005CB522901, 2006CB933100 and
2007CB512901).
REFERENCES
Alexopoulou, A., Baltayiannis, G., Jammeh, S., Waters, J., Dourakis,
S. P. & Karayiannis, P. (2004). Hepatitis B surface antigen variant
with multiple mutations in the a determinant in an agammaglobulinemic patient. J Clin Microbiol 42, 2861–2865.
Carman, W. F. (1997). The clinical significance of surface antigen
variants of hepatitis B virus. J Viral Hepat 4 (Suppl. 1), 11–20.
Carman, W. F., Zanetti, A. R., Karayiannis, P., Waters, J., Manzillo, G.,
Tanzi, E., Zuckerman, A. J. & Thomas, H. C. (1990). Vaccine-induced
escape mutant of hepatitis B virus. Lancet 336, 325–329.
Carman, W. F., Owsianka, A., Wallace, L. A., Dow, B. C. & Mutimer, D. J.
(1999). Antigenic characterization of pre- and post-liver transplant
hepatitis B surface antigen sequences from patients treated with
hepatitis B immune globulin. J Hepatol 31, 195–201.
Chen, J., Fang, F., Li, X., Chang, H. & Chen, Z. (2005). Protection
against influenza virus infection in BALB/c mice immunized with a
single dose of neuraminidase-expressing DNAs by electroporation.
Vaccine 23, 4322–4328.
Cooreman, M. P., Leroux-Roels, G. & Paulij, W. P. (2001). Vaccine-
and hepatitis B immune globulin-induced escape mutations of
hepatitis B virus surface antigen. J Biomed Sci 8, 237–247.
Gavel, Y. & von Heijne, G. (1990). Sequence differences between
glycosylated and non-glycosylated Asn-X-Thr/Ser acceptor sites:
implications for protein engineering. Protein Eng 3, 433–442.
Grethe, S., Monazahian, M., Böhme, I. & Thomssen, R. (1998).
Characterization of unusual escape variants of hepatitis B virus
isolated from hepatitis B surface antigen-negative subject. J Virol 72,
7692–7696.
Gunther, S., Fischer, L., Pult, I., Sterneck, M. & Will, H. (1999).
Naturally occurring variants of hepatitis B virus. Adv Virus Res 52,
25–137.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:51:29
491
C. Wu and others
Hou, J., Wang, Z., Cheng, J., Lin, Y., Lau, G. K., Sun, J., Zhou, F.,
Waters, J., Karayiannis, P. & Luo, K. (2001). Prevalence of naturally
Schmitt, S., Glebe, D., Alving, K., Tolle, T. K., Linder, M., Geyer, H.,
Linder, D., Peter-Katalinic, J., Gerlich, W. H. & Geyer, R. (1999).
occurring surface gene variants of hepatitis B virus in nonimmunized
surface antigen-negative Chinese carriers. Hepatology 34, 1027–1034.
Analysis of the pre-S2 N- and O-linked glycans of the M surface
protein from human hepatitis B virus. J Biol Chem 274, 11945–
11957.
Ireland, J. H., O’Donnell, B., Basuni, A. A., Kean, J. D., Wallace, L. A.,
Lau, G. K. K. & Carman, W. F. (2000). Reactivity of 13 in vitro
expressed hepatitis B surface antigen variants in 7 commercial
diagnostic assays. Hepatology 31, 1176–1182.
Seddigh-Tonekaboni, S., Waters, J. A., Jeffers, S., Gehrke, R.,
Ofenloch, B., Horsch, A., Hess, G., Thomas, H. C. & Karayiannis, P.
(2000). Effect of variation in the common ‘‘a’’ determinant on the
Kaplan, H. A., Welply, J. K. & Lennarz, W. J. (1987). Oligosaccharyl
antigenicity of hepatitis B surface antigen. J Med Virol 60, 113–121.
transferase: the central enzyme in the pathway of glycoprotein
assembly. Biochim Biophys Acta 906, 161–173.
Shi, X. & Elliott, R. M. (2004). Analysis of N-linked glycosylation of
Karayiannis, P., Alexopoulou, A., Hadziyannis, S., Thursz, M., Watts,
R., Seito, S. & Thomas, H. C. (1995). Fulminant hepatitis associated
with hepatitis B virus e antigen-negative infection: importance of host
factors. Hepatology 22, 1628–1634.
Kay, A. & Zoulim, F. (2007). Hepatitis B virus genetic variability and
evolution. Virus Res 127, 164–176.
Lei, Y. C., Hao, Y. H., Zhang, Z. M., Tian, Y. J., Wang, B. J., Yang, Y.,
Zhao, X. P., Lu, M. J., Gong, F. L. & Yang, D. L. (2006). Inhibition of
hepatitis B virus replication by APOBEC3G in vitro and in vivo. World
J Gastroenterol 12, 4492–4497.
hantaan virus glycoproteins and the role of oligosaccharide side
chains in protein folding and intracellular trafficking. J Virol 78,
5414–5422.
Tai, P. C., Suk, F. M., Gerlich, W. H., Neurath, A. R. & Shih, C. (2002).
Hypermodification and immune escape of an internally deleted
middle-envelope (M) protein of frequent and predominant hepatitis
B virus variants. Virology 292, 44–58.
Tian, Y., Xu, Y., Zhang, Z., Meng, Z., Qin, L., Lu, M. & Yang, D. (2007).
The amino acid residues at positions 120 to 123 are crucial for the
antigenicity of hepatitis B surface antigen. J Clin Microbiol 45, 2971–
2978.
Lu, M. & Lorentz, T. (2003). De novo infection in a renal transplant
recipient caused by novel mutants of hepatitis B virus despite the
presence of protective anti-hepatitis B surface antibody. J Infect Dis
187, 1323–1326.
Wang, Q., Wang, Y., Liang, C., Song, J. & Chen, X. (2008).
Marshall, R. D. (1974). The nature and metabolism of the
carbohydrate-peptide linkages of glycoproteins. Biochem Soc Symp
40, 17–26.
Wei, X., Decker, J. M., Wang, S., Hui, H., Kappes, J. C., Wu, X.,
Salazar-Gonzalez, J. F., Salazar, M. G., Kilby, J. M. & other authors
(2003). Antibody neutralization and escape by HIV-1. Nature 422,
Mellquist, J. L., Kasturi, L., Spitalnik, S. L. & Shakin-Eshleman, S. H.
(1998). The amino acid following an asn-X-Ser/Thr sequon is an
Weinberger, K. M., Bauer, T., Bohm, S. & Jilg, W. (2000). High genetic
important determinant of N-linked core glycosylation efficiency.
Biochemistry 37, 6833–6837.
Meng, Z., Xu, Y., Wu, J., Tian, Y., Kemper, T., Bleekmann, B.,
Roggendorf, M., Yang, D. & Lu, M. (2008). Inhibition of hepatitis B
virus gene expression and replication by endoribonuclease-prepared
siRNA. J Virol Methods 150, 27–33.
Okamoto, H., Imai, M., Tsuda, F., Tanaka, T., Miyakawa, Y. & Mayumi, M.
(1987). Point mutation in the S gene of hepatitis B virus for a d/y or w/r
Identification of a hydrophobic domain of HA2 essential to
morphogenesis of Helicoverpa armigera nucleopolyhedrovirus. J Virol
82, 4072–4081.
307–312.
variability of the group-specific a-determinant of hepatitis B virus
surface antigen (HBsAg) and the corresponding fragment of the viral
polymerase in chronic virus carriers lacking detectable HBsAg in
serum. J Gen Virol 81, 1165–1174.
Widera, G., Austin, M., Rabussay, D., Goldbeck, C., Barnett, S. W.,
Chen, M., Leung, L., Otten, G. R., Thudium, K. & other authors
(2000). Increased DNA vaccine delivery and immunogenicity by
electroporation in vivo. J Immunol 164, 4635–4640.
subtypic change in two blood donors carrying a surface antigen of
compound subtype adyr or adwr. J Virol 61, 3030–3034.
Zhang, M., Gaschen, B., Blay, W., Foley, B., Haigwood, N., Kuiken, C.
& Korber, B. (2004). Tracking global patterns of N-linked glycosyla-
Parekh, S., Zoulim, F., Ahn, S. H., Tsai, A., Li, J., Kawai, S., Khan, N.,
Trepo, C., Wands, J. & Tong, S. (2003). Genome replication, virion
tion site variation in highly variable viral glycoproteins: HIV, SIV, and
HCV envelopes and influenza hemagglutinin. Glycobiology 14, 1229–
1246.
secretion, and e antigen expression of naturally occurring hepatitis B
virus core promoter mutants. J Virol 77, 6601–6612.
Peterson, D. L. (1981). Isolation and characterization of the major
Zheng, X., Weinberger, K. M., Gehrke, R., Isogawa, M., Hilken, G.,
Kemper, T., Xu, Y., Yang, D., Jilg, W. & other authors (2004). Mutant
protein and glycoprotein of hepatitis B surface antigen. J Biol Chem
256, 6975–6983.
hepatitis B virus surface antigens (HBsAg) are immunogenic but may
have a changed specificity. Virology 329, 454–464.
492
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 23:51:29
Journal of General Virology 91