Detection of pre-S1 proteins in peripheral blood mononuclear cells from patients with HBV infection

Detection of pre-S, proteins in peripheral blood mononuclear patients with HBV infection

Fabicn

Zoulim’.

Ludmila

Vitvitski’,

Pascal

Bouffard’,

Christian

Pichood’,

Philippe

cells from

Rougierz,

Jean-Pierre Lamelin’ and Christian T&PO’,” Hr,.o”lir .nd ,L,DS R~rFurc,luna, INSERM 0271’. ond”cporo~~r,~~~,~mlogy swvicc. H&.,-DiU’, QO”. Frvnce (Received 5 December ,989)

The presence 01 pre-S, proteins in periphen! blood mononuclear cell (PBMC) samples from 115 patients with different forms of hepatitis B virus (HBV) infection was investigated by Western blot. Among 61 chronic HBsAg carriers, HBV antigens were detected in the PBMC in 80% for HBsAg. 21% for HBc/e Ag and 34% for prr-S, proteins. The detection of pre-S, proteins in PBMC was significantly associated with the presence of serum markers of HBV replication (HBV DNA and/or DNA polymerasc). In the group of 48 consecutive patients negative for serttm HBsAg, but positive for anti-HBc wth or without anti-HBs. HBsAg and prc-S, proteins could be detected in PBMC. Tbi? finding was more frequent ,+mong anti-HIV-positive patients (77 and 23% of the cases, respectively) than in the negative ones (23 and 4% of the .ws. respectively). The detection of HBV DNA and polyadenylatcd RNA in ~ottte of the PBMC samples positive for HBV proteins sugges:, that these proteins may be expressed in PBMC. especially during intense HBV replication. In patients negativc for strum HBsAg, PBMC may constitute a reservoir of HBV.

The hepatitis

B virus (HBV) envelope

contains

three

dtstmct proteins encoded by the pre-S and S regions of the viral genome: the major protein (S protein) encoded by the S region (PXiGP27), the middle protein (pre+ protein) encoded by the pre+ and S regions (GP33iGP36) and the large protein (pre-S, protein) encoded by the preS,. prc-S, and S regions (P39IGP42) (1,2). The pre-S, proteins are mainly expressed on Dane particles (l-4) and their detection in the strum of HBsAg chronic carriers is generally believed to correlate with viral replication (3-6). Some stodics. however, did not confirm this correlation since pre-S, proteins had been detected in the serum of non-viremic carriers (7,X). The few studies dealing with the expression of pre-S, proteins in the liver of chronic hepatitis B patieots sugges: that it may be associated with HBV replication ad/or secretion of Dane

particles into the blood stream (5,6,9,10). In previous studies, mostly dealingwith the detection of HBV DNA, peripheral blood mononuclear cells (PBMC) have been suspected to be target cells for the complete HBV cycle which suggests that PBMC may reflect the situation in infected hepatocytes (11-X). In a preliminary report, we have shown that preJ proteins may also be detected in PBMC of chronic hepatitis B patients (17). To further investigate the significance of the detection of preS, proteins in PBMC, we studied it using Western blot analysis on PBMC lysates in a series of 115 patients sufferinc. from HBV infection with different seroloeical orofiles. Sixteen healthy subjects were used as a cat&l group. The results of pre-S, protein detection in PBMC were correlated with HdV serological markers and HIV StatUs.

iMaterials and Methods

horn a panel of HBV-infected

(i) Parienrs and conwnh Sixty-two patients having histologically proven chronic hepatitis and HBs ant,generma for more rbao 6 months were studied: seven auffered from chronic pers~stcnt hep-

assocnted

paben!s wh I cleated tiBV replicaboo. smce it has heen previously dcscrtbcd that these anttbodlrs may he involved in HBV clearance (19). As shown m Fig. I, we selected ao antibody against prc-S protems from the cerom of a patient wffcring polyartcritis

This rrr~m was anti-HBc

Were anti-HIV positive. Five healthy chronic HBrAg carriers. positive for antiHBe with normal liver function tests. were also studied. Another group of 48 consecutive patients negative for serum HBsAg. but positive for anti-HBc with anti-HBs (39 patients) or without it (nine patients) was included to the study. This latter group consisted of 22 chronic hepatitis cases negative for anti-HIV (ten CPH, seven CAH. five cirrhosis). four subjects with resolved acow HBV infection and 22 patients seropoositive for anti-HIV with

rpeciflc

AIDS Related Complex (ARC) and normal serum traosaminases. Sixteen healthy subjects (ten negative for all HBV serological markers and six good responders to the Hevac B vaccine of Pasteur Vaccins) were iocluded in the control

nor anwHB5

from HBV-

“odosa who had cleared HBcAg.

atitis (CPH). 38 from chronic active hepatttis (CAH) nod 17 from cirrhosis. Forty-six were males and 16 wew tcmales. All were negative for anti-HD ;trmbady. but w

positive, but neither anti-HBe pmitive with commercmlly available tcst~.

This polyclonal

antibody

pcpnde

32-53

was duectrd

against tbc prc-S;

and the pp:-S: specific peptide

120-150 as determined by an ELISA uses!; :-podes cotresponding to qments of the pre-S sequence on toe sclid phase. Brietly. plastic plates were coatei: with synthettc

1

2

3

4

Kd

67,

group.

_GP42

_P39 (ii) Defection

ofHBV markers

-GP35

in serum

.,“, -GP33

antiHBe and anti-HBc were tested by radioimmunoassay using commercially available kits (Abbott Laboratories. Chicago). The serum !iBV DNA polymerase (DNAp) nctiwty HBsAe.

HBcAz

I

and HBeAe.

I

was determined pbosphonoformic

directly

BSwell 8s anti-HBs.

on 50 ~1 of serum

acid (PFA) inhibttable

where (12.16). nitrocellulose N&OH,

Briefly,

the specific

spot test as described else-

50~11 of serum were applied to a

filter. The DNA

was denatured

with 0.2 M

I M NaCl for 20 min. followed by a treatment

with

0.5 MTris,

I M NaCI (pH 7.4) and then witb2x SSC (lx SSC is 0.15 M NaCl10.015 M sodium citrate). The prehy-

bridization

and hybridization

those described for PBMC

procedures wete tdentical to (see below).

The sensitivity

was 0.1 pg for the limit of detection of HBV

DNA

per 50

pl of serum.

(iii) Antibody Anti-pr+SI

againstpre-Spmeins antibodies were selected

_GP 27 -P24

(‘H]dTTP incor-

poration was obtained by calculating the difference (A cpm) between the DNAp activity measured in the precence or absence of PFA (18). The DNAp activity was considered as positive when the d cpm exceeded 50 cpm. The detection of HBV DNA in serum was performed by the molecular hybridization

..bl(

by Western

blat

14-

*-I

!

I52

prc-S okgopcptidcs nturtnc

and incubated with tbc test serum. A

monwlon;d

;a&human

immuno8lobulin

body labelled with Puruxidax body. Furthermore.

anti-

was used as second anti-

in a compctitivc

inhibition

radi<.im-

munonsray. Ihi\ antibody was found to inhibit 7U% of the binding

of :m :mb-pre-S,

tnurine

monockmal

antibody

(F3S 2.5). spccifn for the pre-S, requcnce 32-53. tthody (F3S 251 w’~s kindly Frovidcd (4).

Anti-pre-S

Na,SO,

IgG

was purified

precipitation

followed

This tin-

by Dr. M.A. from

Petit

added for 2 h under agitation.

After

three washings.

IgG was added for I h. The presence of antibody WAS visualirsd

by enzymatic

dine/O. 1% H,O,

staining

in H,O.

with 0.01%

diaminobeszi-

In each experiment.

ant PBMC

sotnple from the control group was analyzed.

In order to

evaluate

35 random

PBMC

the reproducibility

of the method.

samples were studied in duplicate.

the serum by

by chromatograpty

on

DEAF-ccllulosc

(vii) Dermion Total

oftiBV

DNA

DNA in PBMC

was prepared

bation with proteinare

from 5-10.106

K (lOO&ml)

cells by incu-

and SDS (0.5%)

2 h at 31 “C. followed by two phenol-chloroform PRMC Briefly.

WSIC isolated

as described

previously

(13j.

blond collcctcd in hcparin was diluted (l:l.

vJvj

chloroform

extraction.

The DNA

was precipitated

Followmg

luted to ICQnl in 10x SSC. denatured bv he&teat

PBMC

were collected

25 min

at 400

at the interface,

x

g, the

washed

three

m&i EDTA

(pH 8.0). DNA

trocellulose filter (0.45rm)

solution

paratus.

was adjusted mostly

and counted.

to 2.IU’

lymphucytes

-80 “C in TEN

The

cell concentratton

cells per ml. The cells obtained. (90%).

were then

110 mM Tris.

IO mM

kept

frozen

EDTA.

at

0.15 mM

NaCI. pH 7.4) until testing. of the PBMC

preparation

by circ:*laiing HBV

from serwtt and the bind-

ing of HBV

to the surface of the cells. en aliquor of the fi-

“al wash was tested for the presence of HBV radmimmunoarsdy

and HBV

DNA

ern blot analysis of the DNA sampler RNA

and

extracted from ten PBMC

the detection

in three PBMC

nntigens by

by spot test. A South-

of rpecifzc

polyadenylote~i

samples were also performed

for

this purpose (see below).

80 “C. The filter-bound labelled

tion in a Bransonic

12 appamus

(60 W power)

twice for

10 min. the PBMC

lysates were tested for HBV

antigens

(Abbott

Laboratories,

Chicago).

100 “C

containing

was probed with HBV

[‘2P]dCTP

nick translation

(800 Cilmmol,

We used a purified

insert uf pBR

322, k:ndly provided

The hybridization solution

manifold

HBV

Western

ofpre-Sprordm

scribed (1). Briefly. by 1% SDS. min at IW

in PBMC

blot analysis was performed SOpl of PBMC

10% glycerol,

by Werrern

blor

as previously

dt:.

lyvate was denatured

1% ,&mermptocthanol

for 5

“C. Proteins were scparatcd in 12% polyac~4.

amidelh’.N’-methvlene

bisacrvt*.xide

eelt. After

elec:m-

phorests. the sep&ted

proteins were ttti sferred

to a ni-

!:ocellulo~e venal

membrane

electrophorests.

(Sch!cicher Non-specific

and Schuli) by trattsprotein

binding

wits

by Dr.

probe (an

F. Galibert).

was carried ottt, overnight at 42 ‘C, in a

containing

50%

formamide,

hardt’s solution,

lOO~glm1 denatured

and HBV

radiolabelled

DNA

5x

SSC, 3x

gent conditions

as follows: three times in 2x SSC. 0.1%

RP5 film

Cronex Plus intensifying screen (13).

A Southern blot analysis of the DNA PBMC

SSC, 0.1%

samples was performed

extracted from 10

as described briefly. DNA

was analyzed using three restriction

endonucleases:

RI. Hind111 ant! BarnHI.

was ei>xtroohoresed

The DNA

Eco-

for 3 hat

42 “C and then hybridized

for 18 h at 42 “C in the sohttion

described

were washed three times and

above.

Filters

then autoradiographed

at -80 ‘C. For each patient,

ditional

with

hybridization

pBR

with a new filter bearing EcoRI (viii) Derecrion Detection PBMC

ofpolyodenylafed of polyadenylated

probe

an ad-

was performed

digested DNA

(16).

RNA in PBMC RNA

was performed

in

samples from two chronic hepatitis B patients and

one control. reaction

a suitable amount of our against pre-S proteins was

as de-

SDS. The filters were

and exposed at -80 “C against Kodak

Triton X-100 in PBS. Thereafter, IgG dire&d

Den-

salmon s~attt DNA

by nick translation

scribed above. Filters were washed at 68 ‘C under stritt-

saturated with 10% fetal calf serum, 10% dry milk, 0.2%

human polyclonal

DNA kit in a

Amersham

DNA

on a 1% zgarose gel. Filters were prehybridizcd fvil Drrecrim

ap-

to a specific activity of 0.6-

International).

using a Dupont cycles, and ultrasonica-

(pH

was di-

and baked for 3 h at

using Amershsm’s

SDS; and twice in 0.5~

three freezit@thawitg

by radtoimmunoassay

DNA

by nick translation

1.2.108cpm/~g

air-dried After

using a hybridot

The filters were air-dried

reaction

In order to exclude the contaminatton

(lS&le)

for 10 mitt followed by quench cooling. and applied to a ni-

times with large volumes (50 ml) of Hanks’ balanced salt (HBSS).

with 2

vols. of ethanol and resuspended in 10 mM Trir-HCI 7.4)/l

for

for

and one

wnh PBS and loaded on top of 1 vol. of Ficoll Isopaque. centrtfugation

pcr-

oxidase-labelled second entibody directed agaost human

(20)

RNA with

was detected using polymerase a reverse transcriptase

scribed elsewhere (21). Briefly,

chain

step as de-

samples were treated be-

153 fore the revelse tranrcriptaee step with DNAre and RNAse H. The reverse transcriptase step was initlaied with olieodT in order to distineuish between RNA acd mRNA. Controls without reverse transcriptase. aligodr and RNAse H or with RNAse A were included m the reaction. Primers (MD03/06) in the SlPol gene of HBV. kindly provided iw Dr. J. Sninsky. were usedforehe polvmerasechain reaction over 30 cycles. (ix) De~ecdon of HBV anrrgens in live, Detection of HBsAg and HBcieAg in liver sectionswas performed by direct immunofluorescenceas reported earlier (22). Out of the 62 patients with chronic hepatitis B. 59 had simultaneousliver biopsy and PBMC testing.

The &test (with Yates’ correction when required) was used for statistical comparisonof different prevalences.

0) Specificity controls

All of the 16 negative

controls wexfound negative for

HBV markers in PBMC lysates as well as the last wash of the PBMC preparation (data no! shown). Fig. 2 shows the control experiments with preabsarptian of anti-pm8 IgG by HBsAg particles and with or+sion of the first antibody (anti-pre-S I@). These experiments gave further evidence for the specificity of pre-S detection. (ii) Patienrs with chronic heparirir and HBs anrigenemiu Prevalence of HBV markers in PBMC. As shown in Table 1, PBMC from the 62 patients were positive for HBsAg in 53 cases(85%). A.nong these 53 HBsAg-posi-

1

*I3 /

62 55 _i

’

__I~ -GP42 -P?9

30)

154 live casts. HBcieAg was detected m 18 cases and pre-S prowin* in 23 C”CCI. These 23 patients 811expressed pre-S, protcim. while only n frnction expressed pre+ proteins (4!23) Fig 3 +ws tbc result of one Western blotting

immunoassay were tested for the presence of polyadenylated RNA by polymerase chain reaction: both samples were positive, while the control from HBV free subject was negative (see Ref. 20).

analysts of PBMC from chronic hepatitis B patients. Among 26 PBMC sampler tested for FB” DNA by spot tc?~. only two were positive for HBV DNA and both reacted &r HBsAg. HBcieAp and pre-S, proteins. Ten other PBMC preparations were also tested for the presence of HBV DNA bv Southern blot: five were oositive for both pre-S, proteins and extrachromasomal forms of HBV DNA as high molecular weight free concatemers (for more details see Ref 16). Two PBMC samples positive for HBV DNA by Southern blot and HBsAg by radao-

Correlation berwen rho detection of HBV Ags in PBMC and the pmmcr ofserum markers of replicudon. Patients were classified according to serological evidence 01 HBV replication into four groups: (i) chronic hepatitis associated with high HBV replication (HBcAg, HBV DNA and HBV DNAp positive); (ii) chronic bcpatitis asso. ciated with weak HBV replication (HBeAg positive, HBV DNA and HBV DNAp negative); (iii) anti-HBepositive chronic hepatitis positive for HBV DNA and HBV DNAp; and (iv) anti-HBe-positive chronic hepatitis negative for HBV DNA and HBV DNAp. The prevalence of HBV markers in PBMC according to the viral replication status is shown in Table 1. The mean

12

3

4

5-q

serum HBV DNAp value was higher in patients positive far pre-St proteins in PBMC (658 cpm) than in those negative for it (197 cpm). As shown in Table 2. the detection of pre-St proteins and that of HBcIeAg in PBMC were associated with a serum HBV DNAp value higher than 400

KD

cpm (2.tes1.p < 0.05). However, the detection of preS, proteins il? PBMC did not correlate with the HBe/anti-

:8g,” -GPJG -0P33

20_

?4_

HBe statw (2.test.p > 0.05). Thirty-one chronic hepatitis cases associated with antiHBe negative for anti-HD were studied. Among these cases, only seven (23%) had detectahic ~crum HBV DNA and HBV DNAp. HBc/eAg was detected in the liver in 21 cases (68%) and in three cases in PBMC, while pre-S, proteins were positive in eleven cases (35%) in PBMC (Table 1). (iii) Healrhy HBs Ag chronic carrimposirivefor anti-HL?e In this group of five patients, serum HBV DNA and/or DNAp were not detectable. In PBMC, HBsAg wan detected in one case. while pre-S, proteins. HBcieAg and HBV DNA were negative.

membrane Moreover. the additional finding that HBsAg-positwe PBMC not oa,y cantam HB” DNA. but also polyadeoylated RNA. strongly wggesrs proteins are Indeed expressed I,, PBMC (2,).

anti-H& wirh OI wirhouranri-HBs In serum. HBsAe and HBV DNA and/or DNAo were negative in all of the 48 cases. Anti-HBs were positive in 39 patients. As shown in Table 3, the prevalence of HBV markers in PBMC in this group was lower than that of patients with HEsAg-positive chronic hepatitis. No HBV markers could be detected in PBMC of patients with resolved acute HBV infection and normal liver function tests. In the group of patients suffering from chronic hepatitis asswiated with anti-HBc/s, HBV markers were detected in PBMC in 6122 cases. Ii is noteworthy

that in case

of HIV infection, the persistence of HBV in PBMC despite the presence of anti-HBds could be documented in a significantly higher proportion (&test, p < 0.05) than in the anti-HIV-negative patients (Table 3).

,,,a, HB”

Several reports have documented thar HBcJeAg expression in the liver (22-24) an&or in PBMC (13.15.17) ES assoc!ated with HBV replication. In our study. the detection of HBcieAg in PBMC was almost always associated with the presence of serum markers of HBV replication. In agreement with previous studies indicating that the detection of pre-S, proteins in the serum (3-6) and/or in the liver (5.10) IS a marker of viral replicnrion, the pre-S, protein exprewon in PBMC was strongly associated with high serum HBV DNAp activity. However, m thus series, we did not find nny correlation between thr prear~c of pre-S, proteina in PBMC and the HBeianti-HBe bratus A rimilar absence of correlation between the HBeianti-HBe status and the pre-S, protein detection in wrum (7) or in liver (25) hat. bren reported. Despite the relative lack of sensitivity of Western blot. there were more cases positive for pre-S, proteins than for HBcleAe in PBMC. This mav be due either to the overproducti& of HBV ewelo& proteins in comparison to

‘theproductton

of the vtral core or to masking of core pro. wins by anti-HBc within PBMC. Chrome hepatitis cases positive for antt-HBe are usually associared wnh the presence o! liver HBcAg, especially m cytoplasm, and Iow levels of circulatmg HBV DNA (24,26,27). In these anti-HBe chronic hepatitis cases. the detection of pre-S, proteins in PBMC in addition to that of HBcAg in the bver suggested ongoing HBV infection m 26131 cases despite the low prevaience DNA and HBV DNAp. In contras,,

of serum HBV o”t of the five

As expected from ow previous obserations. this study further confirms in a large series the frequent detection of HBsAg and i?Bc/eAg in PBMC of chronic hepatitis patients with HBs antigenemia (13,15517). We also report here, that pre-S, proteins are often detectable in PBMC (37% of cases) by Western blot. The pre-S, proreins were detected in HBsAg-positive sampler only. The absence of HBV markers in the last wash of PBMC preparations and

healthy chronic HBsAg carriers positive for anti-HBe. none had detectable serum HBV DNA and HBV DNAp in agreement with the lack of pre-d, protein detection in PBMC, thus confirming the absence of viral replication. The detection of HBV markers (in&ding pre-S, proteins) in the PBMC of patiects negative for serum HBsAg but positwe for anti-HBc with or without anti-HBs confirxs that HBV infection may persist or resume in PBMC (14.28) and/or in the liver (29) despite anri-HBs reraconversion. Moreover, HBV was detected by PBMC analysis of HBV markers in xvera, cases of chronic hepatitis positive for anti-HBds suggesting that some of the HBsAgnegative chronic hepatitis may be related to HBV (11.13.30). In the case of HIV infection the persistence of

the dctcction of HBV DYA as free high molecular weight forms IX PBMC samples positive for pre-S, proteins indi-

the reactivation

cate that the presence oi pre-S, proteins in PBMC, like that of HBsAg, is more likely the result of the viral genome translation than of HBV binding to the leucocyte

viduals previously immune to HBV (31.32). In conclusion. these data suggest that: (i) pre-S, proteins may be expressed in PBMC. especially m case of m-

HBV in PBMC was a frequent event which could explain of HBV infection cbserved

in AlDS indi-

,e”se HHV voir of HHV

repltcatioo: (d) PBMC in ;mti-HBc-

may constitute B reser-

and anti-HBs-posttive

and (ail) tbc dcwctinn of pre-S, proteins in PRMC a ,elev”“,

“on-i”wsivc

marker

Acknowledgmena

patients;

of ongoing ti3V

may be infec-

tinn

We a,e grareful to Dr. acterization B&ski

of our

MA.

nnti-p-S

and 1. Chrmin

Petit for help in the charantibody,

for performing

the

and to Dr.

1.

PCK.

wilb hepatitis B virus (HBV) i”lecti&J Hepaml 19R7; 4.224. 20 Baginski I, Ferw A, Watson K. Ma&D. Delectl”” “f hcpadds B vi,w I,: PCR Pmtoeolr: a Guide t” Methods and Appliertir)nn. &“i Ynrk: AcademicPress. ,990: 348-55 21 Baginsti 1, Chemi” I. Bauffard P. Hantz 0. R&pa C. PalyIncraw chain reactiondC,CCfi””ofpalyadc”ylated RNA in hep nitis B vi,113 inftxted priphcral blood ma”a”ucka, cell5 (PBMCs). In: Proceedings0’ iiw ,990 ln,e,“a”o”al Symporivm on Viral Hcpatilis andLive, Disease. p. 56. 22 T,Qo CG. Magniur LO, SchaeferRA. Prince A,.,. Detection of e a”tilc,i and antibody: eorrelatianswith hcpatidsB surfaceand hepatbisB coreantigens. live, disease,and “u,E”~c in hepatitis B infection,. Gassoe”le,ology LW(1; 7,. 804-8. 23 Mondelli M, Tedder RS Ferns B. Pondsso P, Rcatdi G. ~lberti A. Differential distribudo” of hepatitis B coreand e antigens in heoaracv,er: ~nalwis b” mo”oelow~l antibodnes. Heoatolow _.

1986:6:199-x4

.?A Ramalho F. Brunelto MR. Rowa G. el PI. Serum markers of

““I hlm”“opathol19R7:4~ 1-8 14 Hadchoue,M. Pasqune,,, C. Faurnie, JG. ct al Delec”“” of