VIROLOGY

185,72-79

(1991)

Rat Monoclonal Antibodies to Nonoverlapping Epitopes of Human lmmunodeficiency Virus Type 1 gp120 Block CD4 Binding in vitro JACKIE CORDELL,* JOHN P. MOORE,t CHRISTOPHER J. DEAN,* PER JOHAN KLASSE,t ROBIN A. WEISS,t AND JANE A. McKEATING?’ *The Institute of Cancer Research, Sutton, Surrey SM2 5NG; and tChester Beetty Laboratories, The Institute of Cancer Research, Fulham Road, London SW3 6/B, United Kingdom Received March 2 1, 199 1; accepted July 11, 199 1 Monoclonal antibodies (MAbs) to a recombinant form of the envelope glycoprotein gpl20 of human immunodeficiency virus type 1 (HIV-1 IIIB) were raised in rats and screened for their ability to block recombinant gpl20 binding to recombinant, soluble CD4 (sCD4) in vitro. Four such MAbs were identified and characterised. Each MAb bound strongly to gpl20 from eight widely divergent HIV-1 strains from the United States and Africa. Two MAbs were mapped to the fourth conserved (C4) region of gpl20, whereas the other two recognised an as yet undefined, conformationally sensitive epitope. MAbs to the latter epitope were the more potent in blocking the gp120-sCD4 interaction. None of the MAbs, however, had potent neutralising activity. o IWI Academic PUSS, Inc.

INTRODUCTION

(Sattentau et a/., 1989). However, the corresponding contact amino acids of gpl20 are poorly defined. In the absence of crystallographic data’ and to complement genetic analyses (Lasky et al., 1987; Kowalski et al., 1987; Olshevsky eta/., 1990), one approach is the generation of monoclonal antibodies (MAbs) capable of blocking the gp120-CD4 interaction. This was successfully applied to outline the gpl20 binding site on CD4 (Sattentau eta/., 1989). We have, therefore, made a panel of rat MAbs to recombinant gp120 and screened them for blocking of gpl20-sCD4 binding. Four MAbs were identified, which fell into two nonoverlapping groups. One of these mapped to the fourth conserved (C4) region of gpl20 which has previously been demonstrated as relevant to CD4 binding (Lasky et a/., 1987; Dowbenko et al., 1988; Cordonnier et a/., 1989a, 1989b). The other group recognises a nonlinear, conformational epitope that we have been unable to define, but which appears to be similar to an epitope recognised by a cross-neutralising human MAb (Ho et al., 1991).

For infection of most, but not all, cells human and simian immunodeficiency viruses (HIV-l, HIV-2, SIV) utilise the CD4 antigen as their primary receptor (Dalgleish et a/., 1984; Klatzmann et a/., 1984; Sattentau et al., 1988). The initial stage of binding is mediated by a high affinity interaction between CD4 and the viral external envelope glycoprotein gp120 (Lasky et al., 1987). Subsequently, a series of incompletely characterised steps involving the transmembrane protein gp41, CD4 (Camerini and Seed, 1990; Healey et al., 1990), and probably other undefined cell surface proteins (Ashorn eta/., 1990; Clapham eta/., 1991) results ultimately in the insertion of an N-terminal fusion domain of gp41 into the host-cell membrane (Kowalski et al., 1987; Gallaher, 1987). Fusion of the virus and cell membranes then occurs by a pH-independent mechanism (Stein et a/., 1987; McClure et al., 1988, 1990). In principle, HIV infection can be blocked at any of the above stages. A challenging prospect in vaccine design is the prevention of the initial virus-cell interaction by the induction of appropriate anti-gpl20 antibodies that would block CD4 binding. Knowledge of the regions of gpl20 and CD4 involved in the binding reaction would be useful toward this goal. The relevant CD4 amino acids have been delineated by crystallographic (Wang et a/., 1990; Ryu et a/., 1990), genetic (Petersen and Seed, 1988; Clayton et a/., 1988; Arthos et a/., 1989), and antibody epitope mapping analyses ’ To whom correspondence dressed. 0042-6822191

MATERIALS Source of reagents

Recombinant gp120 (HIV-1 1118; BHlO clone) was expressed in and purified from CHO cells by Celltech Ltd. for the UK MRC AIDS Directed Programme (ADP) resources programme. Some of its properties have been described elsewhere (Moore et al., 1990a). gpl20 and fragments thereof expressed in Escherichia Co/i(Morikawa eta/., 1990a) were gifts from I. M. Jones

and reprint requests should be ad-

$3.00

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

AND METHODS

72

CD4-BLOCKING

(NERC Institute of Virology, Oxford). Recombinant gpl20 (HIV-1 IltB) and gpll0 (HIV-2 ROD)from baculovirus expression systems (Morikawa et a/., 1990b, 1991) were obtained from American Biotechnology Inc. (Cambridge, MA) through the ADP resources programme. Recombinant soluble CD4 (sCD4) expressed in and purified from CHO cells (Deen eta/., 1988) was a gift from R. W. Sweet (Smith Kline Beecham, King of Prussia, PA). HIV-1 strains IIIB, RF, and MN (Popovic et al., 1984) were obtained from R. C. Gallo, SF-2 was obtained (Levy et al., 1984) from J. Levy, 234 was obtained from M. Martin, and GL3 and GL5 were obtained from G. Farrar. CBL4 was isolated in our own laboratory (Weiss ef a/., 1985). Reduced and alkylated CHO gpl20 was prepared by incubating gpl20 (1 Fg/ml) with 50 mM dithiothreitol for 15 min at 75”, then adding iodoacetamide to a concentration of 100 mM for 30 min at 4”. Dithiothreitol and iodoacetamide were removed by gel filtration on Sephacryl S-l 000. MAb production

and purification

CBH/Cbi rats (National Institute for Medical Research, United Kingdom, Mill Hill) were immunised twice, at 14-day intervals, via Peyer’s patches. The immunogen was recombinant CHO gpl20 emulsified in complete Freund’s adjuvant, with a subsequent boost in incomplete Freund’s. Three days after the second immunisation, mesenteric lymph node cells were fused with the rat myeloma YB-Agl.2.3 (Dean et al., 1986). Supernatants were screened for antibodies to HIV-l gp120 by an antigen-capture immunoassay (modified from Moore et al., 1989). The detection antigen was baculovirus-expressed recombinant gpl20 (HIV-1 IIIB), captured via its C-terminus by sheep antibody D7324 (Aalto BioReagents, Dublin). Bound rat antibodies were detected with ‘251-labelled sheep anti-rat immunoglobulin (Amersham International). Hybridomas from positive wells were cloned twice by limiting dilution and the resulting antibody was purified by precipitation with ammonium sulphate (45% saturated) and then ion-exchange chromatography on Whatman DE52. MAbs 536, 537 (Sun eta/., 1989) and 15e (Ho eta/., 1991) were gifts from D. D. Ho (Aaron Diamond Center, New York), and MAb 5C2E5 (Lasky et a/., 1987) was from T. Gregory and R. Ward (Genentech, San Francisco). MAb 110.5 to the gp120 V3 loop (KinneyThomas et al., 1988) was obtained from Genetic Systems Inc., Seattle. Binding of MAbs to gpl20:

Competition

with sCD4

Virus containing culture supernatants were inactivated with 1% Empigen-BB detergent as previously de-

ANTI-gpl20

73

MAbs

scribed (Moore er a/., 1989). The concentrations of gpl20 present in the preparations were estimated by ELISA using 07324 as a capture antibody and pooled HIV-positive serum to detect bound gpl20, with recombinant CHO gp120 used as a reference standard, as described previously (Moore et al., 1990b). The binding of MAbs and sCD4 to gp120 was assessed by ELISA using methods previously described (Moore et al., 1989, 1990a; Moore, 1990). For competition experiments, MAbs were added to 07324-immobilised gpl20 for 30 min prior to sCD4 addition. Biotinylation

of MAbs and cross-competition

assay

MAbs were dialysed against 100 mM sodium hydrogen carbonate buffer, pH 8.4, then incubated at 500 pg/ml for 16 hr at 4” with 100 Mg/ml of biotin-l\l-hydroxy succinimide ester (Pierce Chemical Co., Rockford, IL). Excess biotin was removed by dialysis against PBS. Biotinylated MAbs were titrated for gpl20 binding activity in ELISA essentially as described previously (Moore et a/., 1989) but using streptavidin-alkaline phosphatase (Novo Nordisk, Cambridge, UK) in the detection system. Biotin-MAb concentrations giving half-maximal binding were used for competition assays, in which unlabelled MAbs were titrated prior to addition to solid-phase immobilised gpl20 essentially as described above. Western

blotting

SDS-PAGE was performed essentially as described by Laemmli (1970) and the separated proteins were transferred electrophoretically to nitrocetfulose (Towbin et al., 1979). Blots were blocked by incubation for 1 hr at 37” in 50 mMTris buffer, pH 8.0, containing 2% (w/v) BSA, 5% (w/v) skimmed milk powder, and 80 mM NaCI. After washing once in Pl3SA/O.O5% Tween 20, the blots were rinsed with PBS and probed for 1 hr at room temperature with MAb-containing culture supernatants. After three washes with PBSIO.O5% Tween 20, the blots were incubated for a further 1 hr with 1251 sheep anti-rat F(ab), (2.1 O6 cpm/ml) (Amersham International) in PBS containing 3% normal sheep serum and then washed in PBS/0.05% Tween 20, d&d, and autoradiographed at -70” on prefogged Fuji RX film (Fuji Ltd., Swindon, UK) using an intensifying screen (Bonner and Lasky, 1974). Neutralisation

assays

Cell-free HIV (1 O4TCID,, units of infectious virus per 100 ~1)was incubated with a range of concantrations of MAb for 1 hr at 37” and then assewd for residual infectivity (TCID,,) on C8166 cells. The concentration

74

CORDELL ET AL.

of MAb able to reduce infectivity by 90% was defined as the neutralisation titre (McKeating et a/., 1989). 0.9 -

RESULTS Generation MAbs

and characterisation

E c

of gpl20-specific

s

Hybridoma culture supernatants were screened initially for antibodies that bound to recombinant gpl20 in antigen capture ELISA in the absence of detergent. Under these conditions, gpl20 binds sCD4 with high affinity (Moore, 1990). Forty-three wells were selected on the basis of positive reactivity and were further screened for the ability to inhibit gp120 binding to sCD4 in vitro. Four MAbs (ICR38.8f, ICR38.1 a, ICR39.13g, and ICR39.3b) were positive in this assay and were selected for further analysis. gpl20-binding

and sCDCblocking

0.6 -

Q 0.4 0.2 0.0 : 1

10

100 khb Concentmtlon

1000

I 10000

nglml

activity

The binding of the four ICR MAbs to gpl20 and their capacities to compete with sCD4 binding are shown in Fig. 1. MAbs ICR38.8f and ICR38.1 a had the highest affinity for gpl20, with half-maximal binding in ELISA at concentrations of 17 and 23 rig/ml, respectively (Fig. 1A). The other two MAbs bound less avidly, the corresponding half-maximal binding concentrations being 100 rig/ml for both (Fig. 1A). An independently derived human MAb, 15e, with gpl20-sCD4 blocking ability (Ho et a/., 1991) bound half-maximally to gpl20 at 48 rig/ml in a similar assay (see Table 1). Binding data for MAbs 536 and 537 (Sun et al., 1989) are also presented in Table 1 for comparison. Under the same conditions, sCD4 binds half-maximally to gpl20 at about 50 rig/ml (Moore, 1990; and data not shown). When binding concentrations are converted into molarities (Table l), it is apparent that MAbs ICR38.8f and ICR38.la have an affinity for gpl20 that is approximately 10 times greater than that of sCD4, and ICR39.139 and ICR39.3b have an affinity approximately twice that of sCD4. All four ICR MAbs prebound onto D7324-immobilised gp120 inhibited the binding of sCD4 when that reagent was added into the reaction mix (Fig. 1 B). The MAb concentrations giving half-maximal inhibition of sCD4 binding under these conditions were ICR38.1 a, 3.5 pg/ml; ICR38.8f, 3.5 pg/ml; ICR39.139, 3.8 pg/ml; and ICR39.3b, 2.1 pg/ml. MAb 15e blocked half-maximally at 1 .O pg/ml (data not shown). Thus despite their 1O-fold higher affinity than sCD4 for gpl20, a 1 O-fold higher concentration of MAbs ICR38.1 a and ICR38.8f was required to block sCD4 binding. The lower affinity MAbs ICR39.3b and 39.139 needed to be in 6- to 1O-

OY 10

1000

100 Yab Conwntmtion

10000

nglml

FIG. 1. Binding of MAbs to gpl20 and their inhibition of the gpl20sCD4 interaction. (A) MAbs ICR38.la (o), ICR38.8f (0) ICR39.3b (Cl), and ICR39.13g @) were added at the concentrations indicated to D7324-immobilised CHO gp120. Bound antibodies were detected with alkaline-phosphatase-conjugated goat anti-rat IgG. (B) MAbs ICR38.la (0) ICR38.8f (0) ICR39.3b (Cl), and ICR39.139 (N) were reacted for 30 min with D7324-immobilised CHO gpl20 at the concentrations indicated, and then sCD4 (0.1 pg/ml, 2.2 nAJ was added to the reaction mix. Bound sCD4 was detected with rabbit anti-sCD4 antiserum and alkaline-phosphatase-conjugated sheep anti-rabbit lg.

fold molar excess for sCD4 binding inhibition. These data are summarised in Table 1. The reactivity of the MAbs to gpl20 from other HIV-1 strains was assessed using detergent extracts of virus-infected cells as an antigen source in a gp120 capture ELISA (Moore et a/., 1989). Each MAb bound to gp120 from the North American HIV-l isolates RF, MN, and SF-2, and also from the Zairian isolates CBL4, 234, GL-3, and GL-5, suggesting that the epitopes for these MAbs are well conserved (Table 2). However, none of the MAbs cross-reacted with recombinant gpll0 from HIV-2 ROD (Table 2). Neutralisation

of HIV-1 by MAbs

MAbs that inhibit the interaction of gpl20 and CD4 should theoretically be capable of neutralising HIV-1 infection. We therefore investigated the ability of the

CDCBLOCKING

ANTI-gpl20

75

MAbs TABLE 3

TABLE 1

NEUTRALISATION OF HIV-1 BY MAbs

gpl 20-BINDING, sCD4-BLOCKING, AND NEUTRALISATION CAPACITYOF ICR MAbs

MAb concentration MAb concentration

for neutralisation

(&ml)

(W) HIV-1 strain

MAb

Epitope

ICR38.8f ICR38.la ICR39.3b ICR39.139 15e 536 537 (sCD4)

c4 c4 Conf Conf Conf c4 c4

gp120 binding

sCD4 blocking

Neutralisation

MAb

IllB

RF

MN

0.11 0.15 0.66 0.66 0.32 0.08 0.69 1 .l

23 23 14 25 7 23 235

2 1O-420 210-420 50-l 00 50-100 n.d. n.d. n.d.

ICR38.la ICR38.8f ICR39.139 ICR39.3b

64 32 16 8

32 64 8 8

64 32 8 16

Note. MAbs are classed by their recognition of the C4 region of gpl20 or an undefined, conformationally sensitive epitope (see text). Data shown are MAb (or sCD4) concentration (nM) required for halfmaximal binding to recombinant gpl20 in ELISA; MAb concentration (nM) that, when preincubated with recombinant gpl20 then left in its presence, blocks half-maximally the subsequent binding of sCD4 (2.2 n&f); MAbconcentration (r&f) blocking HIV-1 (1118)infectiviv of ~8166 cells by 90%. n.d., not determined in this study but see Ho et al. (1991) for MAb 15e and Sun et al. (1989) for MAb 536 and 537 neutralisation titres.

Note. Neutralisation of the HIV-1 strains indicated was assessed by the MAb concentration required to reduce infectivity (ACID,,) by 90%. Note that 1 pg/ml MAb is approximately 6.7 nM.

concentrations (32-64 pg/ml), whersas lCR39.13g and ICR39.3b neutralised at lower concentrations (816 &ml). At none of the concentrations ‘te@@d were any of the MAbs cytotoxic as judged by rH]thymidine incorporation assays (data not shown). Identification

ICR MAbs to block infection of the CD4-positive cell line C8166 by several HIV-1 isolates. All four MAbs reduced the infectivity (TCID,,) of IIIB, RF, and MN for C8166 cells by 90% (Table 3). However, ICR38.1 a and ICR38.8f neutralised HIV-1 (IIIB) only at very high MAb

TABLE 2 BINDINGOF MAbs TO gpl20 FROMDIFFERENTHIV STRAINS MAb bound (OD,,,) HIV strain

ICR38.8f

ICR38.1 a

ICR39.3b

ICR39.13g

IIIB MN RF SF-2 CBL4 234 GL3 GL5 HIV-2 ROD No gpl20

0.98 1.04 1.17 0.84 0.74 0.49 0.82 0.41 0.09 0.07

0.84 0.96 0.89 0.65 0.78 0.51 0.62 0.44 0.05 0.04

0.67 0.85 0.47 0.67 1.43 0.63 0.61 0.20 0.08 0.08

0.71 0.90 0.52 0.52 0.57 0.61 0.65 0.19 0.07 0.09

Note. Empigen extracts of H9 cells infected with the HIV strains indicated and containing l-2 rig/ml gp120 were reacted with the MAbs (10 &ml) for 1 hr before capture of the complexes onto D7324 and determination of gpl20-bound MAb. Data are means of triplicate determinations (with standard deviations ~15% of the mean value).

of nonoverlapping

gpl20

To test whether any of the MAbs bound to overlapping epitopes on recombinant CHO gpl20, w% carried out cross-competition studies using biotinylat%d MAbs. MAbs ICR38.3b and ICR39.13g cross-competed with each other, but not with IC@@.8f and ICR38.1 a (Table 4). Similarly, there was croals-competition between MAbs lCR38.8f and iCR38.1~3, but neither of these blocked the binding of ICR b or ICR39.139 to gp120. Thus, the MAbs s into two groups recognising nonoverlapping ep&op%s. Three additional MAbs previously reported to btock gpl20-CD4 binding were evaluated in the competition assay. MAbs 536 (Sun et al., 1989) and 5C2E5 (Lanky et a/., 1987), mapping within amino acids 430-447 of the C4 region, cross-compet%d with tCR38.8f and ICR38.la for gpl20 binding, suggesting that these MAbs recognised the same or a closely loEat%d epitope. In contrast, MAb 15e (Ho et al., 1991), specific for a conformational epitope, did not compete with ICR38.8f, ICR38.la, 536, or 5C2E5, but did compete with ICR39.138 and lCR39.3b (Tabte 4). 3x1 CR MAbs thus fall into two classes, mapping to distinct regions of gpl20 involved in CD4 binding. Reactivity

of MAbs to various reoomt&wnt

antigens

The abiiity of the MAbs to bind to den&& recombinant gpl20 was investigated by SDS-PAGE and West-

76

CORDELL ET AL. TABLE 4 CROSS-COMPETITIONBETWEENMAb BINDINGSITESON gpl20 MAb2 MAbl

ICR38.8f

ICR38.la

536

5C2E5

ICR38.8f ICR38.la 536 5C2E5 ICR39.3b ICR39.139 15e

+ + + +

+ + + + -

+ + + + -

+ + + + -

-

ICR39.3b + + +

ICR39.13g

15e

+ + +

+ + +

Note. MAbs were tested for their ability to cross-compete for binding to recombinant gpl20 in ELISA. (+) Strong inhibition (~80%) of MAb 2 binding by MAb 1; (-) no significant inhibition (~20%) of MAb 2 binding by MAb 1.

ern blotting. MAbs ICR38.8f and ICR38.la bound to denatured gpl20 under these conditions, suggesting that they recognise a linear epitope (data not shown). They also bound to the 50-kDa C-terminal fragment of gpl20 that had been proteolytically cleaved by thrombin within the V3 loop (Clements eta/., 1991). Because MAbs ICR38.8f and ICR38.la cross-competed with MAbs known to bind to the C4 region of gp120, we tested their binding to a solid-phase adsorbed synthetic peptide, env4 (aa 430-447 of LAV-1 BRU) from this region (McKeating et a/., 199 1). Both ICR38.8f and ICR38.la bound potently to the env4 peptide, as did MAbs 536 (Sun et al., 1989) and 5C2E5 (Lasky et al., 1987). This confirms the cross-competition data and defines approximately the epitope for MAbs ICR38.8f and ICR38.1 a. Fine-mapping of their epitopes within the C4 region of gp120 is described elsewhere (McKeating et al., 1991). In contrast, MAbs ICR39.3b and ICR39.139 failed to bind to denatured gpl20, recombinant E. Colienvfragments, and peptide env4 (data not shown). These findings suggest that their epitope(s) are likely to be conformation-dependent. We therefore investigated the ability of ICR39.3b and ICR39.139 to bind to reduced gp120. Neither ICR MAb bound to dithiothreitoltreated gpl20 (Fig. 2A), and the binding of MAb 15e was also destroyed by reduction of gp120 (Fig. 2C). However, MAbs ICR38.8f, ICR38.1 a, and 110.5 (antiV3 loop; Kinney-Thomas et a/., (1988)) bound strongly to reduced and alkylated gpl20 (Figs. 2B and 2C). The data for MAbs ICR39.3b and ICR39.13 g are similar to those repot-ted previously for MAb 15e (Ho et al., 1991). Taken together with the cross-competition experiments, they indicate strongly that MAbs ICR39.3b, ICR39.139, and 15e bind to the same, conformationdependent epitope.

DISCUSSION Initial progress toward defining the CD4 binding domains on gp120 was made by mutagenesis studies (Lasky et a/., 1987; Kowalski et a/., 1987; Dowbenko et al., 1988). Lasky eta/. (1987) demonstrated that a deletion of 12 amino acids (426-437) from the fourth conserved (C4) region of gp120 abrogated CD4 binding. Several point mutations in this region also affected the ability of gpl20 to bind CD4 (Lasky et al., 1987; Cordonnier et a/., 1989a, 1989b; Olshevsky et al., 1990). The binding to recombinant gp120 of antibodies directed against the C4 region blocked CD4 binding in vitro to variable extents (Lasky et al., 1987; Sun et a/., 1989; McKeating et al., 1991); conversely, anti-C4 antibody binding to gp120 was inhibited by prior CD4 binding (Ardman eta/., 1990; McKeating eta/., 1991). It has, however, generally been reported that anti-C4 antibodies are poorly or nonneutralising (Lasky et a/., 1987; Sun eta/., 1989; Ardman eta/., 1990; McKeating et al,, 1991). Mutations in regions of gp120 other than C4 also influence CD4 binding (Kowalski et al., 1987; Olshevsky et a/., 1990), and a human MAb to a conserved conformational epitope on gpl20 that does not cross-compete with anti-C4 antibodies has been reported to cross-neutralise divergent HIV-1 strains (Ho et a/., 199 1). Taken together with crystallographic estimates of the surface area of gpl20 likely to be covered by the relevant amino acids of CD4 (Wang et al., 1990; Ryu eta/., 1990), this supports the concept that several regions of gpl20 widely separated in the primary sequence fold together to form a topologically complex CDcbinding face or groove (Kieber-Emmons et al., 1990; Olshevsky et al., 1990; Ho et a/., 199 1; McKeating et a/., 1991). Our present data support this; two of

CDCBLOCKINGANTI-gpl20 MAbs

has not yet been defined, MAbs ICR39.13g and

A

ICR39.3b cross-compete withthe humanMAb1%

0.8 0.6-

---t --O-

39.3b+ 39.3b

-f-I)-

='%+ 39.139

Mab Concentmtion

rig/ml

B 1.2-

-

l.O-

---I)-

36.la+

" E 8 %

0.6 -

0.6 8

::;: ;

lb

100 Mab Concantmtlon

lob0

10000

rig/ml

C 1.2 l.O-

E 0.6~

* -f--D-

15% 15.3 110.5+ 110.5

c ffj

0.6 -

n 0

0.4 0.2-

&,

0.0 1

77

10

1000 100 MabConosnlrationng/ml

10000

FIG. 2. Effect of gp120 reduction on MAb recognition, (A) MAbs ICR39.3b (0, 0) and ICR39.13g @, Cl); (B) ICR38.la (A, A) and ICR38.8f (Cl, S); and (C) 15e (a, 0) and 110.5 (m, Cl) were added at the concentrations indicated to D7324-immobilised CHO gpl20 previously treated with (open symbols) or without (closed symbols) dithiothreitol and iodoacetamide. Bound MAbs were detected with alkaline-phosphatase-conjugated goat anti-rat IQG.

the MAbs (ICR38.8f and ICR38.1a) map to the C4 region and, although able to block recombinant gp120 binding to sCD4 in vitro, are only weakly neutralising. They therefore have similar properties to other MAbs to the same region (Lasky eta/., 1987; Sun eta/., 1989; McKeating et al., 199 1). In contrast, MAbs ICR39.139 and ICR39.3b do not bind to the C4 region but also weakly neutralise HIV infectivity. While their epitope

(Ho et al., 1991) and in general have similar properties to that antibody. It is noteworthy that this group of MAbs more potently neutralise HfV infectivity than MAbs to the C4 region despite their lower affinity for recombinant gpl20. They are, furthermore, more potent inhibitors of gpl20-sCD4 binding in vitro relative to their affinity for gp120 under the same conditions (Table 1). Thus antibodies to the conformational epitape(s) may more closely overlap the CD4-binding sites on gpl20 than do antibodies to the C4 region. We are attempting to define more precisely the conformational epitope recognised by this group of MAbs. The envelope glycoproteins of HIV are the major targets for the host humoral immune response (Weiss et al., 1986; Robey et a/., 1986; Ho et a/., 1987) and antibodies against gpl20 and gp41 are responsible for almost all the neutralising activity found in the sera of HIV-infected primates (Matthews et al., 1986; Steimer et a/., 1987; Berman et a/., 1988). Most HIV-1 positive human sera can block the binding of HIV-1 or recombinant gpl20 to sCD4 (McDougal et al., 1986; Skinner et a/., 1988; Schnittman et al., 1988; Moore 1990), suggesting that such activity may account for the crossneutralising activity observed in most human sera. However, we (unpublished data) and others (Sun et al., 1989) have been unable to detect human serum antibodies to the C4 region, suggesting that this region of gpl20 is immunosilent in humans. It remains possible that antibodies to the undefined conformational epitopes recognised by MAbs 15e (Ho et al., 1991) ICR39.13g, and ICR39.3b are responsible for the gpl20-CD4 blocking and cross-neutralising activities seen in HIV-positive human sera. Supporting this is the observation that a large proportion of HIV-positive human sera contains antibodies that compete for ICR39.3b binding to gp120 (P. J. Klasse and J. A. McKeating, unpublished data). If the conformationally sensitive site on gpl20 is crucial for sCD4 binding and virus neutralisation, it is relevant for vaccine design that we have been able to raise MAbs with these properties by immunising with recombinant gpl20. Although we used Freund’s as the adjuvant in our studies, this may be less than optimal for the preservation of gpl20 conformational epitopes which may be important for cross-neutralisation (Haigwood et al., 1990; Ho er a/., 1991). Thus the two groups of MAbs described in this study complement previously reported CD4-gpl20 blocking MAbs (Lasky et al., 1987; Sun et al., 1989; Ho et al., 1991; McKeating et al., 199 1) and may assist in the delineation of the CD4-binding regions of gpl20.

78

CORDELL ET AL. ACKNOWLEDGMENTS

VIROLOGY

185,80-89

tinn nf InA-nnoretinn

rat X rat hvhrirfnman

In “Mnthnrk

in Fwvmnl-

(199 1)

Identification of a Novel Transcription Factor, ACF, in Cultured Avian Fibroblast Cells That Interacts with a Marek’s Disease Virus Late Gene Promoter PAUL M. COUSSENS,’ VIRGINIA L. TIEBER, CHRISTINE S. MEHIGH, Molecular

AND

MICHELLE MARCUS

Virology Laboratory, Department of Animal Science, Michigan State University, East Lansing, Michigan 48824 Received March 26, 7997; accepted July 2, 199 1

Interactions between factors in duck and chick embryo fibroblast (DEF and CEF, respectively) nuclear extracts and the Marek’s disease virus (MDV) gp57-65 gene promoter were investigated. Results of in vitro transcription and gel mobility-shift assays indicated that multiple cellular factors interact with 5’-flanking sequences of the MDV gp57-85 gene. One sequence-specific DNA binding activity (termed ACF for avian cell factor(s)) was identified by interaction of DEF and CEF nuclear extract proteins with a particular site (nucleotides -193 to -177) in the MDV gp57-85 gene promoter. Binding of ACF to its apparent recognition sequence, contained within the 17-bp oligonucleotide 5’-CTAGlTTACTTGTITGT-3’ (ACF-12), was highly sequence-specific. Radiolabeled ACF-12 oligonucleotide bound significant ACF protein in the presence of a 400-fold molar excess of unlabeled nonspecific competitor DNA. A similar amount of specific competitor completely abolished ACF binding to probe DNA. Deletion of the ACF binding site from MDV gp57-65 gene promoters linked to a chloramphenicol acetyltransferase (CAT) reporter gene reduced expression of CAT activity by twofold relative to that seen with a gp57-65 promoter-CAT construct containing an intact ACF binding site. Transfection inhibition assays using double-stranded ACF binding site competitors reduced steady-state levels of gp57-65 mRNA in MDV infected cells by over twofold relative to those in control infected cells. Introduction of a similar amount of nonspecific double-stranded oligonucleotide had no adverse effect on gp57-65 mRNA levels. These data suggest that ACF is important for efficient expression of gp57-65. Q wgi Academic PWSS, I~C.

INTRODUCTION Marek’s disease virus (MDV), a member of the herpesviridae family, produces an acute lymphoproliferative disorder in poultry (Marek, 1907; Pappenheimer et a/., 1929a,b). Classified as a gammaherpesvirus because of its propensity for infecting lymphocytes and acute oncogenicity (Payne, 1982; Ross, 1982) MDV more closely resembles the alphaherpesviruses in terms of overall genome structure (Fukuchi er al., 1985) and apparent similarity of various genes (Buckmaster et a/., 1988; Coussens et a/., 1989; Brunovskis and Velicer, 199 1). As with most herpesviruses, genes in MDV appear to be temporally regulated in a cascade fashion (Maray et al., 1988). Although specific MDV transcriptional units have yet to be examined in detail, certain genes are expressed early in infection (immediate-early and early genes) while others are not expressed until after viral DNA replication is initiated (late genes). Expression of late genes is prevented by specific inhibitors of viral DNA replication, such as phosphonoacetic acid (PAA) (Lee et al., 1976,1978). Thus far, few studies have investigated the complex regulatory machinery required to elicit this pattern of gene expression in MDV infected cells. By analogy to herpes simplex virus ’ To whom

requests

for reprints

should

be addressed.

(HSV) and other herpesviruses, however, gene expression in MDV is likely to be regulated by specific interactions of both viral and cellular proteins with MDV gene promoters through formation of protein-DNA and protein-protein complexes (review, Wagner, 1982). An MDV strain GA “late” structural gene encoding gp57-65 and its associated promoter was previously localized, molecularly cloned, and sequenced (Coussens and Velicer, 1988; lsfort et al., 1987). Predicted amino acid sequences of gp57-65 show homology to herpes simplex virus type-l (HSV-1) glycoprotein C and gC-like proteins of other herpesviruses (Binns and Ross, 1989; Coussens et a/., 1989). More recently, nine additional MDV genes have been mapped within the unique short (U,) segment of the MDV genome (Brunovskis and Velicer, 1991; Cantello et a/., 1991). Nucleotide sequence analysis of these genes has revealed striking similarities to genes in the U, segment of HSV-1. Factors responsible for regulating expression of MDV genes have not been examined in detail. Regulation of gp57-65 is of particular interest in the MDV system for several reasons. First, as MDV strains are passaged in culture, expression of gp57-65 is lost or significantly reduced (Churchill et al., 1969; Ross, 1982; Schat eta/., 1985). In many cases, loss of gp5765 expression correlates with a loss of oncogenicity (Churchill et al., 1969; Ross, 1982; Schat eta/., 1985). Second, by analogy to HSV-1, efficient expression and

CDCBLOCKING (1990). The pH independence of mammalian retrovirus infection. 1. Gen. Viral. 71, 767-773. MCDOUGAL, J. S., NICHOLSON, J. K. A., CROSS, G. D., KENNEDY, M. S., and MAWLE, A. C. (1986). Binding of the human retrovirus HTLV-III/ LAVIARVIHIV to the CD4 (T4) molecule: Conformational dependence, epitope mapping, antibody inhibition and potential for idiotypic mimicry. /. Immunol. 137, 2937-2944. MCKEATING, 1. A., MCKNIGHT, A., MCINTOSH, K., CLAPHAM, P. R., MULDER,C. A., and WEISS, R. A. (1989). Evaluation of HIV and SIV plaque and neutialization assays. J. Gen. Viral. 70, 3327-3333. MCKEATING.J. A., MOORE, J. P., FERGUSON,M., GRAHAM, S., ALMOND, 1. W., EVANS,D. J., and WE!%, R. A. (1991). Monoclonal antibodies to the C4 region of human immunodeficiency virus type 1 gpl20: Use in topological analysis of a CD4 binding site. Submitted for publication. MATHEWS, T. J., LANGLOIS, A. J., ROBEY. W. G., CHANG, N. T., GALLO, R. C., FISCHINGER,P. J., and BOLOGNESI, D. P. (1986). Restricted neutralisation of divergent human T-iymphotropic virus type ill isolates by antibodies to the major envelope glycoprotein. froc. Nat/. Acad. Sci. USA 63,9709-97 14. MOORE,J. P. (1990). Simple methods for monitoring HIV-l and HIV-2 gpl20 binding to sCD4 by ELISA: HIV-2 has a 25fold lower affinity than HIV-1 for sCD4. A/OS 3, 297-305. MOORE, J. P., WALLACE, L. A., FOLLETT,E. A. C., and MCKEATING,J. A. (1989). An enzyme-linked immunosorbent assay for antibodies to the envelope glycoproteins of divergent strains of HIV-1 A/OS 3, 155-163. MOORE, J. P., MCKEATING,J. A., JONES,I. M., STEPHENS,P. E., CLEMENTS,G., THOMSON, S., and WEISS, R. A. (1990a). Characterisation of recombinant gp120 and gp160 from HIV-l: Binding to monoclonal antibodies and sCD4. AIDS 4, 307-3 15. MOORE, J. P., MCKEATING,J. A., WEISS, R. A., and SATENTAU, Q. 1. (1990b). Dissociation of gpl20 from HIV-l virions induced by soluble CD4. Science 250, 1139-l 142. MORIKAWA,Y., MOORE,1. P., and JONES,I. M. (1990a). HIV-1 envelope protein gpl20 expression by secretion in E. Co/i: Assessment of CD4 binding and use in epitope mapping. J. Wrol. Methods 29, 105-l 14. MORIKAWA,Y., OVERTON,H. A., MOORE, 1. P., WILKINSON,A. J., BRADY, R. L., LEWIS, S. J., and JONES,I. M. (1990b). Expression of HIV-1 gpl20 and human soluble CD4 by recombinant baculoviruses and their interaction in vitro. AIDS Res. Hum. Retroviruses 6,765-773. MORIKAWA,Y., MOORE,1. P., WILKINSON,A. J., and JONES,I. M. (1991). Reduction in CD4 binding affinity associated with removal of a single glycosylation site in the external glycoprotein of HIV-2. Virology 180, 853-856. OLSHEVSKY,U., HELSETH,E., FURMAN, C., LI, J., HASELTINE,W., and SODROSKI,1. (1990). Identification of individual human immunodeficiency virus type I gpl20 amino acids important for CD4 binding. J. Viral. 64, 5701-5707. PETERSEN,A., and SEED, B. (1988). Genetic analysis of monoclonal antibody and HIV binding sites on the human lymphocyte antigen CD4. Cell 54, 65-72. POPOVIC, M., SARNGADHARAN,M. G., READ, E., and GALLO, R. C. (1984). A method for detection, isolation, and continuous production of cytopathic human T-lymphotropic retroviruses of the HTLV family (HTLV-IIIB) from patients with AIDS and pre-AIDS. Science 224,497-500.

ANTI-gpl20

MAbs

79

ROBEY, W. G., ARTHUR, L. O., MAI-~HNVS, T. J., LAN~LoIs, A., COPELAND, T. D., LERCHE,N. W., OROSZLAN, S., BOLOQNESI, D. P., GILDEN, R. V., and FISCHINGER,P. J. (1986). Prospect for prevention of human immunodeficiency virus infection: Purified 120-kDa envelope glycoprotein induces neutralizing antibody. Proc. Nat/. Acad. Sci. USA 63,7023-7027. RYU, S. E., KWONG, P. D., TRUNEH,A., PORTER,T. G., ARTHUR. J.. ROSENBERG,M., DAI, M., XUONG,N.-H., AXEL, R., SWEET,R. W., and HENDRICKSON,W. A. (1990). Crystal structure of an HIV-binding recombinant fragment of CD4. Nature 346, 419-426. SAI-~ENTAU,Q. J., CLAPHAM, P. R., WEISS, R. A., BEVERLEY.P. C. L., MONTAGNIER,L., ALHALABI, M. F., GLUCK~AAN,J. C., and KLAT~MAN. D. (1988). The human and simian immunodeficiency viruses HIV1, HIV-2 and SIV interact with similar epitopes on their cellular receptor, the CD4 molecule. AIDS 2, 101-l 05. SATTENTAU,Q. J., ARTHOS,J., DEEN, K., HANNA, N., HEALEY, D., BEVERLEY,P. C. L., SWEET,R. W., and TRUNEH,A. (1989). Structural analysis of the human immunodeficiency virus-binding domain of CD4. J. Exp. Med. 170, 1319-1334. SCHNIITMAN, S. M., LANE, H. C., ROTH, J., BURROWS,A. M., FOLKS, T. M., KEHRL,J. H., KOENIG,S., BERMAN,P., and Faua, A. S. (1988). Characterisation of gpl20 binding to CD4 and an assay that measures ability of sera to inhibit this binding. J. Immunol. 141, 41814186. SKINNER,M. A., LANGLOIS,A. J., MCDANAL, C. B.. MCDOUGAL, J. S., BOLOGNESI,D. P., and MATTHEWS,T. J. (1988). Neutralizing antibodies to an immunodominant envelope sequence do not prevent gpl20 binding to CD4. J. Viral. 62, 4195-4200. STEIMER,K. S.. VAN NEXT, G., DINA, D., BARR,P. J., Luciw, P. A., and MILLER, E. T. (1987). In “Modern Approaches to New Vaccines” (R. M. Chanock, R. A. Lerner, F. Brown, and H. Ginsberg, Eds.), pp. 236-241. Cold Spring Harbor Laboratories, Cold Spring Harbor, NY. STEIN, B. S., GOWDA, S., LIFSON,J., PENHALLOW,R., BENSCH,K., and ENGLEMAN.E. (1987). pH-independent HIV entry into CD4-positive cells via virus envelope fusion to the plasma membrane. Cell 49, 659-668. SUN, N. C., Ho, D. D., SUN, C. R. Y., LIOU, R. S., GORDON,W., FUNG, M. S. C., LI, X. L., TING, R. C., LEE, T. H., CHANG.N. T., and CHANG, T. W. (1989). Generation and characterisation of monoclonal antibodies to the putative CD4-binding domain of HIV-1 gp120. I Viral. 63, 3579-3585. TOWBIN, H., STAEHELIN,T., and GORDON,1. (1979). Electrophoretic transfer of protein from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Nat/. Acad SC;. USA 76,4350-4354. WANG, J., YAN, Y., GARRETT,T. P. J.. LIU, J., RODQERS,D. W., GARLICK, R. L., TARR, G. E., HUSAIN,Y., REINHERZ,E. L., and HARRISON,S. C. (1990). Atomic structure of a fragment of human CD4 containing two immunoglobulin-like domains. Nature 346, 41 l-41 9. WEISS, R. A., CLAPHAM, P. R., CHEINGSONG-POPOV,R., DALGLEISH, A. G.. CARNE, C. A., WELLER, I. V. D.. and TEDDER,R. S. (1985). Neutralization of human T-lymphotropic virus type Ill by sera of AIDS and AIDS-risk patients. Nature 316, 69-72. WEISS, R. A., CLAPHAM, P. R., WEBER,1. N., DAL(~LEISH,A. G., &KY, L. A., and BERMAN,P. W. (1986). Variable and conserved neutralization antigens of HIV. Nature 324, 572- 575.