(~) |NSTITUTPASTEUR/ELsEVIER Paris 1992

Res. Viroi.

1992, 143, 179-191

Human and murine monoclonal antibodies directed against a conserved sequence from gp41 (aa583-599) of human immunodeficiency virus type 1 A. Ebersold o)('), V. Boyer (I), p . j . Klasse (2), M. Holnigue (3), C. Fraisier o), J.M. Cocchi o), R. Pipkorn (4), j. Blomberg (2) and C. Desgranges O) fo Unit# de Recherche sur les H#patites, le SIDA et les Rdtrovirus humains, I N S E R M U271, 151 Cours Albert Thomas, 69424 Lyon Cedex 03 (France), c2) Section o f Virology, Lurid University, L u n d (Sweden), ¢3) Centre Rdgional de Transfusion sanguine, 59000 Lille (France), and r~J Novabiochem A G, Sandhausen, Heidelberg (Germany)

SUMMARY

Human spleen cells from an HIV-seropositive donor were immunized in vitro with the aa583-599 peptide conjugated to an heptalysyl core. This sequence was derived from the putatively HIV-immunosuppressive region of HIV1 gp41. The same conjugated peptide was used to immunize mice. One human and one mouse IgM monoclonal antibody (mAb) directed against the aa583-599 peptide were obtained. The t w o mAb had distinct patterns of reactivity against a panel of 42 peptides with modified sequences. Neither of the mAb inhibited the immunosuppressive effect of aa583-599 octopus-lysconjugated peptide on anti-CD3 Ab-induced lymphoproliferation. In addition, both mAb did not neutralize cell-free virus transmission or enhance HIV infection. However, HmAb inhibited formation of syncytia between HIVl-infected (but not HIV2-infected cells) and non-infected target cells at concentrations above 20 i~g/ml, whereas MmAb did not have any effect. The degree of conservation of the aa583-599 region makes HmAb a candidate for use as a group-specific reagent in future HIV1 passive immunotherapy protocols.

Key-words: Lymphocyte, Peptide, HIV, AIDS, mAb, IgM, Immunotherapy; gp41, Immunosuppression, Syncytia, pHIVIS, Heptalysine.

INTRODUCTION Antibodies (Ab) specific to the human immunodeficiency virus type 1 (HIV1) are produced in all infected individuals. Several groups have reported Ab that neutralize HIV1 infectivity in vitro (Ho et aL, 1987; RobertGuroff et al., 1985 ; Weiss et al., 1985). The pre-

Submitted February 5, 1992, accepted April 24, 1992. (*) Correspondingauthor.

cise location and structure of epitopes on HIV1 proteins that elicit neutralizing Ab, or have other effects on HIV1 infection, such as cell cytotoxicity or immunomodulation, are now being defined. Although a variety of epitopes which elicit neutralizing Ab are located on gpl20 (CD4-receptor-binding site), a highly conserved region of the extracellular domain of gp41

A. EBERSOLD ET AL.

180

(Modrow et al., 1987), called CS3 (Cianciolo et al., 1988) or pHIVIS (for putatively HIVimmunossuppressive sequence) (Klasse et ai., 1988) is a potentially important epitope. This region has 6 exact amino acid (aa) matches with a 17-aa murine peptide described to be immunosuppressive in vitro (Cianciolo et al., 1988). Ruegg et al. (1989) have found that the aa583-599 peptide specifically inhibits human and murine lymphoproliferation induced by antiCD3 Ab in vitro. It has also been shown that this peptide binds specifically to CD4-positive cell lines and peripheral blood mononuclear cells (PBMC) (Qureshi et al., 1990). This binding inhibits HIVl-mediated cytopathology and infection, suggesting a possible role for a second receptor in the infectious process of HIV1. A recent report showed that the aa584-595 peptide is recognized by cytotoxic T lymphocytes (CTL) induced by immunization of humans with recombinant HIV1 gpl60 ( H a m m o n d et ai., 1991), suggesting a role in cellular immunity for this region. An association between Ab to pHIVIS and absence of AIDS among HIVl-infected persons has been described (Klasse et al., 1988) but could not be found in any o t h e r c o h o r t (Lange et al., 1989; Cheingsong-Popov et al., 1990). More recently, it has been shown that vaccination of rhesus macaques with a aa582-602 [3gal-conjugated peptide from simian immunodeficiency virus (SIV) envelope homologous with aa583-599 from HIV, induced peptide-specific virus-neutralizing Ab (Shafferman et aL, 1991). We report here the generation of a h u m a n monoclonal antibody (HmAb) after in vitro immunization of spleen cells from a seropositive donor with the aa583-599 coupled to an hep-

Ab

=

aa

=

antibody. amino acid.

CTL FCS FITC

= = =

cytotoxic T lymphocyte. foetal calf serum. fluorescein isothiocyanate.

gp

=

HIV HmAb lg

= -=

talysyl core bearing 8 reactive NH 2 termini (aa583-599 octopus-lysine-conjugated peptide), and of mouse monoclonal antibody (MmAb) generated after immunization with the same conjugated peptide. The m A b did not have any inhibitory effect against the immunosuppressive activity of the aa583-599 peptide. We also have compared their capacity to neutralize or enhance HIV1 infection. M m A b directed against aa583-599 peptide did not show any neutralizing or enhancing activity, whereas H m A b directed against the same region inhibited fusion between infected and non-infected cells at a concentration of 20 ~tg/ml without having any enhancing activity.

MATERIALS

Peptide synthesis

The pHIVIS peptide analogous to aa sequence 583 to 599 from the HIV-IIIB (clone BH10) isolate of HIVI was synthesized by means of FMOC solidphase technology (Klasse et al., 1988), and 99 070purified by high-performance chromatography. The 17 aa covered by the aa583-599 peptide are LQARILAVERYLKDQQL. The peptide was directly synthesized on a lysine "octopus" core bearing 8 reactive NH 2 termini for immunization. To screen clones secreting mAb directed against HIVI, we used unconjugated aa583-599 peptide and, as a negative control, aa252-272 peptide (kindly provided by the "Agence Nationale de Recherche sur le Sida") from the aa sequence of the BRU isolate (Neosystem, Strasbourg, France). Generation of mAb

To obtain HmAb directed against aa583-599, 6 x 107 lymphocytes isolated from the spleen of an HIV-seropositive asymptomatic 17-year old hae-

IU

=

lys

=

MmAb

=

PBMC

=

PBS

=

glycoprotein.

pHIVIS

=

human immunodeficiency virus.

RT

=

human monoclonal antibody.

SIV

=

immunoglobulin.

AND METHODS

infectious unit. lysine. mouse monoclonal antibody. peripheral blood mononuclear cell. phosphate-bu ffered saline. putatively HlV-immunosuppressive sequence. reverse transcriptase. simian immunodeficiency virus.

H U M A N A N D M U R I N E m A b A G A I N S T H I V I gp41 P E P T I D E

mophilia patient were immunized in vitro with 50 ~g of the aa583-599 octopus-lys-conjugated peptide in 60 ml of RPMI-1640 supplemented with l0 % FCS and 20 gg/ml phytohaemagglutinin (Wellcome Diagnostics, Dartford, England). Immunized lymphocytes and P3bX63 mouse myeloma cells were fused on day 7 at a ratio of 3:1, using 45 °70 polyethylene glycol and 7 % dimethyl sulphoxid. Cells were suspended in RPMI-1640 supplemented with l0 % FCS, 100 IzM hypoxanthine, 0.4 ~.M aminopterin, 16 ~.M thymidine, and seeded in 96-well plates. After 2-3 weeks, supernatant fluids from 451 wells with hybrid growth were screened by ELISA. To obtain MmAb directed against aa583-599, 3 to 5-weekold BALB/c mice were immunized with 100 ~.g of aa583-599 octopus-lys-conjugated peptide in Freund's complete adjuvant. Two weeks later, mice were boosted in Freund's incomplete adjuvant and two weeks later in saline buffer. Three days prior to fusion, the mice received 50 ~g of conjugated peptide. Spleen cells were fused with SP2/0 Agl4 mouse myeloma cells following classical procedures. Immunized mice sera and supernatants from wells with hybrid growth were screened by ELISA.

181

Determination of Ab concentration

The determination of HmAb concentration in culture supernatants was performed by ELISA, as already described (Trabaud et al., 1989). For MmAb, Nunc plates were coated with rat anti-mouse IgM (Biosoft, Paris, France) and incubated with culture supernatants or ascites fluid dilutions. Bound Ig were detected with alkaline-phosphatase-conjugated goat anti-mouse IgGAM (Zymed). Mouse Ig reference serum (ICN ImmunoBiologicals, IL, USA) was used to produce standard curves. Purification of human monoclonal IgM

The harvested culture supernatants were loaded onto a 30-mi cation exchange "Mono S" column (Pharmacia, Uppsala, Sweden). The active fraction from the Mono S flow was then gel-filtered on a "Superdex 20" column equilibrated with PBS pH 7.2 (300 ml bed volume). The purity was monitored by gel electrophoresis and anti-HIV activity tested by ELISA.

Anti-HIV1 Ab specificity determination

Immunosuppressive effect on the anti-CD3 Ab proliferation assay

A commercial HIV Ab ELISA (ELAVIA-1, Marnes La Coquette, France) was used according to the manufacturer's instructions. To determine peptide specificity, peptide ELISA was performed as previously described (Boyer et al., 1991 ; Klasse et al., 1988). The mAb were assessed for specificity to HIVI antigens with a commercial Western immunobiot kit (Diagnostics Pasteur) according to the manufacturer's instructions. To determine whether Ab reactivity was directed against cell surface HIV-expressed antigenic determinants, cytofluorimetric analysis (FACScan, Becton Dickinson, Mountain View, CA) was carried out with HIV-infected H9 cells and uninfected H9 cells as previously described (Boyer et al., 1991).

The proliferation assay was performed as previously described (Cianciolo et al., 1988). PBMC 005 in 50 i~l) were stimulated by anti-CD3 mAb (5 ng/ml, lmmunotech, Marseille, France) in the presence or absence of the aa583-599 octopus-lys-conjugated peptide or aal05-117 as negative control (from 50 ng/ml to 50~.g/ml), for 7 2 h at 37°C with 5 °7o CO 2. 3H-thymidine incorporation (l v.Ci/well) was measured during the final 6 h of incubation. AntiCD3-stimulated PBMC in the presence of medium alone usually incorporated 30,000-50,000 cpm/well and in the presence of the highest concentration of aa583-599-conjugated peptide, 1,000-10,000 cpm/well. Neutralization assays

Isotype determination

Heavy and light chains of the HmAb were determined by intracytoplasmic and membrane immunofluorescence on ethanol-fixed and viable cells, reacting with FITC-conjugated goat F(ab') 2 to human )., V-,~, × and "I' (Kallestad, TX, USA), and sheep anti-human IgG1, 2, 3 or 4 (Organon Teknika, Serifontalne, France) revealed by a FITC-conjugated rabbit anti-sheep IgG (Biosys, Compi/~gne, France). Heavy and light chains of the MmAb were determined by "Mono-Id EIA" kit (Zymed, CA, USA) according to the manufacturer's instructions.

Purified HmAb and ascites fluid were tested in 4 neutralization assays. Cell to cell virus transmission

As already described (Boyer et al., 1991), the syncytia formation assay was performed by incubating 50 g.I (104 cells) of H9 cells infected with 1 of 4 different HIVI strains (HIV-IIIB, HIV-MN, HIVRF, HIV-CH88) or Molt4 cells infected with HIV2, with 50 izl of the tested Ab for 1 h at 37°C. A volume

A. E B E R S O L D E T A L .

182

Cell-free H I V 1 neutralization assays

of 50 ~I o f SUPTl-uninfected target cells ( 2 x 104 cells) were added to each well o f a 96-well flatbottomed microtitre plate and incubated at 37°C with 5 % CO 2. After 5 to 18 h o f incubation, the number of giant cells per well was determined by microscopic examination.

A.

?

The supernatants from de novo HIV-llIB-infected H9 cells were collected, separated from cells, filtered on a 0.45-1zm filter membrane, aliquoted and stored at - 80°C until use. The infectious titre of the viral

B.

C.

•.• , ~ • . • .

! !

!

.

~

"

g~

".

•,,~ ¢ 4

-r !

• "~"3

-:"

.

• "-'..~"~. '.I ~ "

FSC--HxFSC-Helgh t - - >

FSC-t~FSC-He lOh t - - )

4:HgI619L~4 Ouad Stats F l l t : 4~HgI619L~4 Sa~plt~ Date: 1/11~9"d Gate GI,, RJ Parameters: FSC-H(LIN), FL1Total5tee Gated= Ouad Events ~ Gated 1 2 3 4

UL UR LL UR

e 2L>81 e 939

e.ee 78.84 e.eo 29, 16

4, H91619~--'eI x F L I ~ I - H e

.

i ght

t

:

"0

FSC-HxFSC-HelOh t ~

4zHgI6192L~ 0uad Stats

>

4:H91619~31

F i l e ~ 4~H9161~26 Sample= Dater 1/1~-

O

0

JO

o

U C

"~

~-

0 C

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

I

I

585



I

584

L

E

R Y

L

K

585

I

586

I

587

I

shifted single residues

A V

.i . i . , . i. . , . ~ . i. . i . i .

l) 1 2 - m e r s

I I I : L Q A R

/

A

|

i

&

588

I

D Q L

i

&

b ) E f f e c l s of s i n g l e - r e s i d u e deletions

~d

R

r,n

oo 4~

185

H U M A N A N D M U R I N E mAb A G A I N S T HIV1 gp41 PEPTIDE

the same conjugated peptide. Sera from two immunized mice gave a strong positive signal in E L I S A and spleen cells from one o f these mice were fused with mouse m y e l o m a S P 2 / O Ag 14 cells. One clone secreting about 5 ixg/ml o f antiaa583-599 Ig was selected.

Table I. Absorbance differences relative to that of

Serum f r o m the HIV-seropositive d o n o r reacted with HIV antigens in ELISA, but did not react with aa583-599 peptide, whereas serum from the immunized mouse gave a positive signal in both assays (data not shown).

For each sample, a reactivityvaluewas calculatedfrom the differencebetweenthe absorbance(OD 492 nm) of the peptidecoatedwelland the mock-coatedwell(0.01-0.05).The ahsorhance differencefor eachpeptidehas beendividedby that for aa583-599. The absorbance differencesobtained are referredto as "reactivities".

Both M m A b and H m A b were IgM),. H m A b F432B2 cell culture supernatant bound to H9 infected with HIV-IIIB strain (fig. lc), whereas the non-specific IgM negative control did not (fig. lb). No staining was observed with the H9 cells infected with other viral strains tested (MN, RII and CH88), and uninfected H9 cells (data not shown). The m A b were further characterized by Western blotting. Only H m A b F432B2 recognized a band o f 41 k D a molecular weight (fig. 2), whereas M m A b 12G10F5 failed to detect any specific HIV viral proteins (data not shown).

Epitope characterization

Both h u m a n and mouse m A b were tested in E L I S A with 42 sequence variants o f the aa583-599 peptide (their coating efficiencies being comparable) (table I and fig. 3): T589 (LKARILTVERYLKDQQL), aa581-599, aa586-606 (table I), 11 peptides with systematic pairwise threonine substitutions (fig. 3a), 16 with single-residue deletions (fig. 3b), 6 single-residue-

aa583-599.

F432B2 12GIOF5

T589

581-599

586-606

1.4 1.0

I. l 0.4

0.7 1.4

shifted 17-mers (fig. 3c), 6 single-residue-shifted 12-mers (fig. 3d). H m A b F432B2 presented unmodified reactivity with T589 and aa581-599, but low reactivity with aa586-606 (table I). All residue pairs except A R and ER seem to be important in the structure recognized by H m A b F432B2 (fig. 3a). The smallest modification (the single T for L at 599) had the strongest effect on antigenicity (fig. 3a). Deletions mostly increased reactivity to different degrees. L - A R I L A - E - Y - D Q L all seemed to be important to the F432B2 epitope (fig. 3b). This m a y be due to a better conformational simulation o f the epitope by this peptide. The aa583-599 seemed to stimulate the H m A b F432B2 epitope best a m o n g the 17-mers o f the region (fig. 3c). Twelve residues appeared to be too few to simulate the H m A b F432B2 epitope, suggesting a high c o n f o r m a t i o n a l dependence (fig. 3d). We found that M m A b 12G10F5 supernatant gave a strong positive signal with aa586-606 and

Fig. 3. The reactivities in peptide-ELISA are expressed relative to that of the parental peptide in a), b), to 583-599 in c) and d). The median of 7 tests with F432B2 diluted 1/2 is represented by open triangles. The median OD in 5 tests with 12GIOF5 is represented by filled circles, a) F432B2 and 12GIOF5 tested against peptides with threonine-pair substitutions in the sequence aa581-599; a single threonine at position 599. b) Deletion of simple residues in the sequence aa583-599, c) The 17-reefs shifted single residues in the sequence, starting at position 580 to 586. d) The 12-reefs shifted single residues in the sequence, starting at position 583 to 588.

186

A. E B E R S O L D E T A L .

T589 (which has the sequence derived from the neutralization escape m u t a n t previously described (Reitz et al., 1988; Wilson et ak, 1990), but presented a low reactivity with aa581-599 (table I). Threonine substitutions did not abrogate the reactivity of MmAb 12G10F5 (fig. 3a). R586 seems to be important in the 12G10F5 epitope (fig. 3b), but this does not correlate with the effect of substituting TT for AR (fig. 3a). Figure 3c shows that peptide "585", i.e. aa585-601, lacking the amino-terminal LQ of the immunogen and having an additional carboxyterminal LG, bound MmAb 12G10F5 notably stronger than aa583-599. MmAb 12G10F5 bound significantly to all 12-mers (fig. 3d). The sequence they all have in common is aa588-594 LAVERYL, which may thus constitute the epitope.

Inhibition of immunosuppression

In view of the immunosuppressive effect of the aa583-599 region on anti-CD3-Ab-induced PBMC proliferation, we investigated the neu-

tralizing effect of both HmAb and MmAb in this assay. Anti-CD3-stimulated PBMC in the presence of medium alone usually incorporated 30,000-50,000 cpm/well. The lowest concentration at which the aa583-599 octopus-lysconjugated peptide was found to reduce the antiCD3-induced proliferation to 1,000-10,000 cpm/well was 20 vg/ml. Neither H m A b nor MmAb (even at concentrations of up to 100 ~g/ml) were able to inhibit immunosuppression after pre-incubation with the conjugated peptide.

Neutralization transmission

assay

of

cell-to-cell

viral

The mAb were assessed for their effect on syncytia formation between uninfected SUPT1 target cells and cells infected with 5 different HIV strains. When HmAb F432B2 was incubated with infected H9 cells, the size and number of syncytia was reduced by 90 070 at 100 ~tg/ml and by 50 070 at 10 i~g/ml, after 18 h of coculture (fig. 4). The effect of H m A b on syncytia formation was dose-dependent (fig. 4). At

100 80

8o

!

!:

60

40

20

0

20

40

60

80

100

120

IIIB

t

pg/ml of IgM Fig. 4. Dose-dependent effect of the HmAb F432B2 ( t ) on syncytia formation between HIV-IIIB infected H9 cells and SUPT1 target cells. Negative control (F-l) is a human monoclonal IgM directed against tetanus antigen. Experiments were performed in triplicate.

MN

RF

CH88

HIV-1

I HIV-2

Different HIV Isolates

Fig. 5. Effect of the HmAb F432B2 on syncytia between H9 cells infected with different viral isolates and SUPTI target cells. (l) HIV-IIIB, (2) HIVMN, (3) HIVRF, (4) HIVCH88, (5) HIV2. A 20-1~g/ml concentration of F432B2 mAb was used in this assay.

H U M A N A N D M U R I N E m A b A G A I N S T H I V I gp41 P E P T I D E

187

20 ~tg/ml, more than 50 07o inhibition was obtained with the 4 HIVI isolates tested. In contrast, syncytia between HIV2-infected Molt4 cells and SUPT1 cells were not inhibited by HmAb F432B2 (fig. 5). These syncytia were unaffected by the HIV-positive serum of the spleen donor (1:100) and even with the highest concentration of MmAb 12G10F5 tested (50 I.tg/ml).

Cell-free HIV1 neutralization assays of mAb

To assess the specificity of inhibition, 50 ~.1 of aa583-599 octopus-lys-conjugated peptide (or aal05-117 as negative control) at 100 I.tg/ml was incubated with 50 I.d of HmAb F432B2 at 100 ~.g/ml for 1 h prior to the addition of the infected H9 cells. The mixture was then added to the HIV-IIIB-infected cells for 1 h at 37°C with 5 °70 CO 2, and incubated overnight with SUPT1 target cells, as described in "Materials and Methods". Preincubation of HmAb F432B2 with peptide abrogated the neutralizing effect of the mAb (fig. 6). This result suggests that aa583-599 octopus-lys-conjugated peptide competed with the antigenic determinant recognized by F432B2 on gp41.

Enhancing activity of HIVI infection

IO0

!! 0

20

40

60

80

100

120

pg/ml d peptlde

Fig. 6. Dose-dependent effect of pre-incubation of the aa583-599-poly-L-lys peptide ([3) or aal05-117 as negative control (0) with 100 ~tg/ml of HmAb F432B2 on syncytia formation between HIV-IIIB-infected H9 cells and SUPTI target cells.

In the MT4 colorimetric assay, H9 and CEMSS assays, no inhibition of cell-free virus infectivity was observed, even at 100 ~.g/ml of HmAb or 50 i.tg/ml of MmAb.

No enhancing activity was detected in the MT2 assay for the mAb concentrations tested ( H m A b , 1 n g / m l to 10g.g/ml; M m A b , 50 ng/ml to 5 g.g/ml).

DISCUSSION

To determine the biological significance of mAb directed against the conserved sequence localized between aa583 and aa599 of HIVI gp41, and to examine the possible differences in antigen recognition by human and mouse Ab, we produced a murine and a human hybridoma secreting mAb against aa583-599 peptide, and studied their biological activities. HmAb F432B2 was produced by in vitro immunization of spleen cells from an HIV 1-seropositive individual with the aa583-599 octopus-lys-conjugated peptide. PBMC had been previously immunized in culture against a recombinant fragment of gpl20 from HIVI (Borrebaeck et al., 1988). In vitro immunization would overcome a severe limitation in human hybridoma technology and allow the production of H m A b against therapeutically valuable antigens of HIV 1. We observed that the serum of the donor did not show any reactivity in ELISA with the aa583-599 peptide, suggesting that in vitro immunization corresponds to a primary activation of an aa583-599-antigenspecific B cell in culture. Amino-terminal residues in aa583-599 and perhaps also aa581-582 are important in the antigenic structure recognized by H m A b F432B2, which is probably directed to an epitope other than aa586-606 (D6pel et al., 1989). The low reactivity of MmAb 12G10F5 with aa581-599 may indicate that the amino-terminal leucine must be free, as in the immunogen, but this is

188

A. EBERSOLD E T AL.

Table II. Naturally occurring HIV sequence variability in the pHIVIS epitope.

Isolate BRU, BHI0, HXB2 MN, BRVA, OYI RF, SF2, HAN HIV2 (ROD) SIV

Sequence 5s3LQARILAVERYLKDQQL599 583LQARVLAVERYLKDQQL599 5s3LQARVLAVERYLRDQQL599 583LQARVTAIEKYLQDQAR599 583EKYLEDQAQs9t

Syncytia inhibition + + + ND

Sequence data are taken from Bisance, Comp. Appl. in Biosciences (1990) 6, 355-356. Inhibition of syncytia formation by HmAb F432B2 between target cells and H9 cells infected by isolate in italics ( + ). No inhibition observed ( - ).

r e f u t e d by the high reactivity with TTLQARILAVERYLKDQQL. In summary, the human and mouse mAb differed in their reactivities with the variant peptides: these results may be of value in illustrating two distinct modes of Ab reaction to a functionally important region of gp41. HmAb F432B2 directed against aa583-599 inhibited syncytia formation between HIVl-infected H9 cells and uninfected SUPT1 target cells, whereas it failed to neutralize the cell-free transmission of the virus in 3 different assays. Up to now, 3 MmAb (Matsushita et al., 1988 ; Skinner et al., 1988b ; Durda et al., 1990) and one HmAb (Scott et al., 1990) with fusion inhibitory activity at a concentration of 0.5 g.g/ml to more than 200 i.tg/ml have been described, but they have always been mapped to the highly variable V3 loop of HIV1, resulting in an HIV-type-specific neutralization. Moreover, recent work described cell-free neutralization of different HIV strains by 8 to 64 g.g/ml of rat mAb directed against gpl20 (Cordell et al., 1991). In the present work, we have obtained an HmAb directed against the aa583-599 region of gp41 which neutralizes syncytia formation in 4 out of the 5 isolates tested. This is consistent with the fact that this region is a conserved portion of the envelope among different HIV1 isolates (Modrow et al., 1987; H a m m o n d et al., 1991) but differs from the HIV2 sequence (table II). Shafferman et al. (1991) have induced protec-

tion of macaques by vaccination with aa582-602 [3gal-conjugated peptide from a region of the SIV envelope homologous to aa583-599 from HIV (table II). In this study, virus suppression correlated with prechallenge neutralizing Ab titres as measured in a cell-free SIV syncitia formation assay. This suggests that homologous HIV peptides, and also mAb directed against this region, may be important components of future immunization strategies. The separation of a neutralization phenotype in which Ab is capable of blocking syncytia formation without blocking cell-free virus neutralization is unusual but not unpreecedented. Ab that recognize viral envelope proteins have been shown to sometimes inhibit viral infection at a post-binding step (Dalgleish et al., 1988; Skinner et al., 1988a; Celada et al., 1990). A possible dissociation between syncytia formation and cell-free virus infection is suggested. Since the aa581-597 peptide suppresses the in vitro proliferative response of murine and human T lymphocytes to mitogenic stimulation (Cianciolo et al., 1988; Ruegg et aL, 1989), we investigated if mAb against aa583-599 region were able to abrogate this immunosuppressive effect. We did not find that the mAb had any capacity for inhibiting the immunosuppressive effect of the aa583-599 peptide. Other functions have been described for the aa583-599 region. Indeed, Robinson et al. (1990) showed that H m A b directed against aa586-620 enhanced HIV1 infection in a complement-mediated

HUMAN AND MURINE mAb AGAINST

a n t i b o d y - d e p e n d e n t enhancement fashion and did not neutralize HIV1, even at a high concentration. We did not find any enhancing activity of human and mouse m A b against the aa583-599 peptide. These conflicting results could stem from the fact that an immunodominant sequence that elicited enhancing antibodies could be Iocated between aa600 and aa620 o f gp41 (D6pel et al., 1989). A n o t h e r possibility may be the different isotype of the H m A b produced (IgG 1K for R o b i n s o n et al., IgM), for F432B2). Thus, their different capacities to bind and activate c o m p l e m e n t could play a role in the enhancing properties. Moreover, Qureshi et al. (1990) have suggested that an additional receptor on HIV1 target cells exists which is able to bind to region aa583-599 o f gp41. Thus, a possible mechanism for cell fusion inhibition by the H m A b F432B2 here described, could be the prevention o f the binding o f gp41 to this additional cell surface receptor on target cells. Until now, preliminary experiments undertaken in our l a b o r a t o r y did not allow us to conclude on this point. Fusion o f H I V l - i n f e c t e d cells and target cells cannot be blocked by the M m A b 12GIOFS. Mice and humans m a y recognize antigens differently, and Ab response to epitopes o f the same region m a y be different. The interpretation o f the protective potential o f such epitopes in future vaccines cannot be based solely on information from m A b from immunized mice. Thus, it is important to develop a large panel o f H m A b in order to identify neutralizing epitopes on the HIV1 envelope glycoproteins and to better understand the mechanism by which neutralizing Ab act on the infection process. Use of octopuslysine-conjugated peptides as antigen in in vitro immunization procedures is a rational approach to the production o f H m A b with biological activities against such HIV1 epitopes.

Acknowledgements This work was supported by the Institut National de la Sant~ et de la Recherche M~dicale (INSERM) and by the Agence nationale de recherches sur le SIDA (ANRS), France. A. Ebersold is a recipient of a CIFRE contract with the Centre R~giona[ de Transfusion Sanguine de Lille.

H I V I gp41 P E P T I D E

189

Spleen was kindly provided by Dr. Mouton from CHU Lille. We thank Drs. J.M. Huart and H. Broly for financial assistance and helpful discussions.

Anticorps monoclonaux humain et murin dirig~s contre une s~quence conserv~e de la gp41 (aa583-599) du virus de I'immunod~ficience humaine Les lymphocytes de ia rate d'un donneur s~ropositif pour le VIH ont ~t~ immunis6s in vitro par un peptide synth~tique constitu~ des acides amines 583 A 599 de la r~gion immunosuppressive de la gp41 du V1H. Pour l'immunisation, le peptide a ~t~ coupl~ /~ un

Human and murine monoclonal antibodies directed against a conserved sequence from gp41 (aa583-599) of human immunodeficiency virus type 1.

Human spleen cells from an HIV-seropositive donor were immunized in vitro with the aa583-599 peptide conjugated to an heptalysyl core. This sequence w...
1MB Sizes 0 Downloads 0 Views