CELLULAR
IMMUNOLOGY
145,287-298
(1992)
Down-Regulation of Cell Surface CD4 Molecule Expression by Anti-CD4 Antibodies in Human T Lymphocytes P. MOREL,*
Induced
C. VINCENT,* J. WIJDENES,~AND J. P. REVILLARD*
*Immunology Laboratory, INSERM lJ80, CNRS URA 1177, Lyon, France; and tlnnotherapie Laboratoires, Besancon. France Received June 16. 1992; acceptedAugust 19, I992 Antigenic modulation was defined as the down-regulation of a cell surface antigen expression induced by exposureto specificantibody. We investigatedthe modulation of CD4 surfaceexpression in human peripheral blood lymphocytes incubated in vitro with anti-CD4 monoclonal antibodies (mAbs). Modulation of surface CD4 was achieved at 37°C but not at 4°C. with five different murine anti-CD4 mAbs of IgGl and IgC2a subclasses,with different epitope specificities. Modulation wasdosedependentwith a maximum at nonsaturating mAb concentration. It was reversible upon culture in mAb-free medium. It was acceleratedand amplified in the presenceof monocytes or after cross-linking of anti-CD4 mAbs. It could be induced with solid phase anti-CD4 mAbs. but not with soluble F(ab’)r fragments. Its magnitude was identical on all CD4+ lymphocytes. It was associatedwith a moderate down-regulation of CD2 and CD3 but not of CD8 and HLA class I surfaceexpression.Modulation was slightly augmentedby addition of inhibitors of the endosome/ lysosome pathway but not by protein synthesis inhibitors. The anti-CD4 mAb initially bound to cell surfacewas no longer detectableafter 24 hr of culture. Most of surfaceCD4 proteins complexed with antibody were rapidly internalized and transiently replaced by CD4 from an intmcytoplasmic pool and then no longer were expressed.CD4 mRNA was moderately decreasedin cells incubated with anti-CD4 mAb while @-actinand µgJobulin mRNAs remained at stable levels. It was concluded that down-regulation of CD4 surfaceexpression induced by anti-CD4 mAb concerned only a part of CD4 molecules and was associatedwith a decreasedsynthesis.The delay required to achieve maximal modulation is likely to reflect exhaustion of the intracytoplasmic recycling 0 1992 Academic press, Inc. pool of CD4 IIIOkCUkS.
INTRODUCTION The CD4 antigen is a 55-kDa glycoprotein expressedon the surface of a functionally distinct subset of human T lymphocytes and on some cells of the reticuloendothelial system (1, 2). This molecule interacts with class II antigens expressed on antigenpresenting cells and functions as a coreceptor in the activation of T cells by the TcR/ CD3 pathway (3,4). The CD4 molecule has also been identified as a receptor for HIV (5,6). On human and murine T cells, CD4 is physically associatedwith the &-related tyrosine protein kinase ~56“‘~ (7, 8). CD4 can function as a signal transducer and tyrosine kinase phosphorylation events may be important in CDCmediated signaling. Cross-linking of the CD4 receptor induces a rapid increase in the tyrosine-specific protein kinase activity of ~56”~ and is associated with the rapid phosphorylation of one of the subunits ({) of the TcR complex on tyrosine residues(9, 10). T cell activation by the TcR/CD3 or by the CD2 pathways, as well as direct stimulation by phorbol 287 0008~8749192 $5.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
288
MOREL
ET AL.
esters,induces receptor cointernalization and a transient defective expression of both CD4 and TcR/CD3 on T lymphocyte surface ( 11- 13). Modulation of CD4 molecule expression can therefore be induced by T cell activation, e.g., with anti-CD2 and antiCD3 antibodies. With one known exception (14) anti-CD4 antibodies alone cannot trigger T cell activation. However, during the course of in vivo administration of antiCD4 mAb in patients with rheumatoid arthritis or psoriasis, we observed a markedly decreasedexpression of CD4 molecules at the cell surface ( 15). Therefore, the present study was addressedto the mechanisms whereby anti-CD4 antibodies could regulate cell surface expression of CD4 antigen in vitro. MATERIALS AND METHODS Monoclonal Antibodies (mAbs) Five anti-CD4 mAbs were used in this study (Table 1). The equilibrium constant, K, was calculated according to Scatchard analysis of saturation curves established with ‘251-labeledmAbs. The saturating concentration was defined as the lowest amount of antibody needed to obtain a plateau in the conditions of cell labeling for immunofluorescence. The following anti-CD4 mAbs were used: BL4, initially produced by J. Brochier in our laboratory (15). BL4 is a mouse IgG2a which recognizes an epitope of group 2B. BB 14 and BF8 are mouse IgG 1 purified from ascites by chromatography, produced by J. Wijdenes at the Centre Regional de Transfusion Sanguine. The 13B8.2, IgGl, was obtained from Dr. Mawas (INSERM U 119, Marseille, France) and FlO l-69, IgG 1, from Dr. Poncelet (Sanofi Research Center, Montpellier, France). F(ab’)* fragments from BB 14 were prepared by bromelain digestion and purified by exclusion on protein A-Sepharose chromatography (Pharmacia Uppsala Sweden). A sample of BB14 was biotinylated with Act-biotin (IBF, France) according to the instruction sheet. Its fixation was revealed with streptavidin phycoerythrin (SAPE) (Caltag, San Francisco, CA). Leu 3a-phycoerythrin (PE) (IgG 1, Becton-Dickinson, Mountain View, CA) and OKTCfluorescein isothiocyanate (FITC) (IgG2b, Ortho Pharmaceutical Corp., Raritan, NJ), which do not cross-reactwith the anti-CD4 mAb described above, were used to assessCD4 expression. MAbs of other specificity were used as control: anti-CD2, LeuSb-FITC; anti-CD3, Leu 4-FITC; anti-CD8, Leu2a-PE (Becton-Dickinson), and a mouse anti-p2-microglobulin mAb (BE104, IgG2a), developed in our laboratory (16). Chemicals Chloroquine, monensin, bacitracin, cycloheximide, L-leucine methyl ester, and phorbol myristate acetate (PMA) were obtained from Sigma (St. Louis, MO). IodoTABLE
1
Characteristics of the Anti-CD4
Isotype Saturating dose (wg/ml)
K, L X hi-'
mAbs
BB14
BF8
BL4
13B8.2
FlOl-69
IgG I 5-10 3.7 x lo*
IgG 1 5-10
IgG2a 10
5.6 X IO'
5.5 x IO'
IgG 1 10 1.3 x 10s
IgGl 10 1.9 x lo*
CD4 DOWN-REGULATION
BY ANTI-CD4
mAbs
289
gen was obtained from Pierce (Rockford, IL) and phthalate oils were obtained from Fluka (Buchs CH). Ionomycin was obtained from Calbiochem (La Jolla, CA). lmmunojluorescence Assay PBMC from healthy donors were separated by centrifugation on Ficoll-Hypaque (Lymphoprep, Nycone A.S., Norway). To assessthe degreeof CD4 saturation by the mAb, 20 ~1 of serial dilutions of anti-CD4 mAb were added on PBMC pellets (5 X 105),30 min at 4°C. Cells were then washed with PBS, stained with polyclonal goatanti-mouse Ig (H+L) antibodies conjugated with FITC (GAMIg-FITC, Tago Immunologicals, Burlingame, CA), and were fixed with phosphate-buffered saline (PBS) containing bovine serum albumin (BSA), azide ( 1%) and formaldehyde (1%). CD4 antigen expression on PBMC was measured by flow cytometry using a FACScan analyzer (Becton-Dickinson) and was analyzed with the Lysis software. Analysis was performed with a scatter gated on lymphocyte fraction, excluding monocytes, polymorphs, and dead cells. Antigen density on the cell surface was evaluated by specific mean fluorescence intensity (MFI) measured in arbitrary units (log scale). Results are expressedas the percentage of the initial MFI obtained in saturating conditions, after substraction of MFI measured on the same cells in the absenceof mAb. To assessthe effect of anti-CD4 mAb on CD4 antigen expression, PBMC were incubated with an anti-CD4 mAb (10 pg/106 cells/ml) at 37°C 7.5% COz, in RPM1 1640culture medium supplemented with glutamine (2 mM), penicillin, streptomycin, and 10% fetal calf serum (FCS). At different time intervals, cells were sampled and washed twice in PBS and then CD4 expression was assessedby staining with a fluoresceinated anti-CD4 mAb. As control, the expressions of CD2, CD3, CD8, and & microglobulin were analyzed in the same experiment. Control of the cycloheximide activity was assessedby measuring [3H]thymidine incorporation into cell nuclei during the last 16 hr of 3-day cultures after stimulation with PMA plus ionomycin. In some experiments, monocytes were eliminated from PBMC with L-leucine methyl ester (Sigma, 5 mM, 5 X IO6 cells/ml, 40 min at room temperature) according to Thiele et al. ( 17). L-Leucine methyl ester is a lysosomotropic agent rapidly metabolized into leucine inside the lysosomes, causing osmotic swelling of these organelles and subsequent cell death. Monocyte depletion was assessedby flow cytometric analysis after staining with LeuM3 (CD14) mAb (Becton-Dickinson). After L-leucine methyl ester treatment less than 1% of the monocytes were detected in the cell suspensions. These cell suspensions are referred to as peripheral blood lymphocytes (PBL). pzMicroglobulin concentrations were measured with an ELISA method described elsewhere ( 16). Assays with Radiolabeled Antibodies MAbs were isolated either by filtration on DEAE-Sephadex A50 in 0.1 M phosphate buffer at pH 7.0 or by chromatography on protein A-Sepharose with a 20 mM phosphate buffer, pH 7.80, with 1 A4 NaCl, followed by an elution with 0.1 M glycine/ HCl, pH 2.7, buffer. Antibody molecules were labeled by the Iodo-gene method with a specific activity of 300-350 MBq/mg. Labeled molecules were diluted in RPM1 1640 medium supplemented with 10% FCS and antibiotics were then used within 1 hr.
290
MOREL
ET AL.
For binding assays,PBL (l-4 X 106/ml) were incubated at 4°C or at 37°C with 10 &ml of ‘251-labeledCD4 mAb in RPM1 1640 medium containing 10% FCS. After various periods of incubation, unbound 1251-labeled mAb was removed by layering of 50 ~1 samples on a cushion of 300 ~1 of phthalate oil (1 vol of phthalate-dibutylester and 1.1 vol of bis(2-ethylhexyl) phthalate) in a 0.7-ml Eppendorf microtube. After 4 min of centrifugation at 12,OOOg at 4°C the bottom of the tube which contained the cell pellet was cut and counted in a gamma counter. The remaining part of the tube was also counted in order to calculate total activity. Acid resistant activity was measured by addition of 50 ~1of 0.1 A4glycine/HCl, pH 2.7, buffer on the cushion of phthalate oil together with the cells. Centrifugation was performed after 5 min of incubation in acid buffer. Isolation and Identification of mRNA We used the following DNA probes specific for CD4 RNA (clone pT4B, a kind gift from Dr. Piatier, CNRS Nogent-sur-Maine), µglobulin RNA (18) (gift from F. Lemonnier, Villejuif), and P-actin RNA (pACP, gift from J. P. Rouault, Lyon). These probes were labeled by random priming (random priming kit, Amersham) with [32P]dCTP. For Northern blot analysis, total RNA from 5 X lo6 PBL was prepared by a modified phenol/chloroform/isoamyl alcohol extraction method. Five-microgram samples of RNA were electrophoresed through a 1% agarosegel containing 6% formaldehyde in the running buffer. RNAs were blotted onto a nylon membrane (Hybond N, Amersham) by the standard capillary blotting technique. The blots were fixed by heating 2 hr at 80°C and prehybridized 2 hr at 42°C in 50% formamide, 5X SSC,5X Denhardt’s solution, 0.5% SDS, 5% dextran, and 0.25 mg/ml of denatured salmon sperm DNA. Blots were incubated overnight at 42°C with 20-30 ng of radiolabeled DNA probes. Blots were rinsed in 2X SSC,0.1% SDS at room temperature and were washed twice in the samesolution 20 min at 65°C. A last wash step was done in 2X SSC,0.2% SDS at 65°C and they were finally exposed to Kodak XOMAT film for 12-48 hr at 70°C. Size of RNA were determined by comparison with 0.24- to 9.5-kb RNA Ladder (GIBCO BRL). RESULTS DecreasedMembrane Density of CD4 Molecules on T Cells Incubated with Anti-CD4 mAbs Kinetics of CD4 density changes on human PBL in the presence of anti-CD4 mAb. PBMC were incubated at 37°C with BB 14 mAb at saturating concentration ( 1O6cells/ 10 pg/ml). At various time intervals, cells were stained with a phycoerythrin-conjugated anti-CD4 mAb (Leu3a-PE) which binds to another epitope than BB14. CD4 downregulation occurred within the first 24 hr. It was maximal between 24 and 48 hr with a loss of 40 to 50% of the initial MFI (Fig. IA). Despite a minor spontaneous increase after 48 hr of culture, the MFI remained 26% lower than the initial value after 5 days of culture (Fig. 1A). As control, CD4 expression remained stable on PBMC incubated in antibody-free medium. During all the culture period, the percentage of CD4+stained cells remained stableand CD4 expressionon the cells was homogenous, showing that the antigenic modulation occurred in all the CD4’ lymphocytes with the same
CD4 DOWN-REGULATION A
100
3 E80 8 6 a960 s !i 40
291
BY ANTI-CD4 mAbs
100
L I
0
24
80 60
I
I
I
48
72
96
Hours
I
120
0
48
96
Hours
144
0
24
48
72
Hours
FIG. 1. In vitro down-regulation of CD4 surface antigen induced by anti-CD4 mAbs on lymphocytes. (A) PBMC were cultured at 37°C with(m) or without BB14 (0) and at 4°C with BB14 (e). SurfaceCD4 proteins were labeled at various time intervals with Leu3a-PE. Resultsare expressedas percentageof the MFI obtained with saturating doses,as described under Material and Methods. (B) Effect of BB14 on CD2, CD3, CD8, and µglobulin expression. PBMC were cultured with BB14 at 37°C and were stained with antibodies against CD2 (0) CD3 (0) CD8 (0) &-microglobulin (A), or CD4 (m). (C) PBMC were cultured with three different IgGl anti-CD4 mAbs (0, 0, A) or one IgG2a anti-CD4 mAb, BL4 (Cl) at 37°C. CD4 expression was assessedwith OKT4-FITC or Leu3a-PE. intensity. CD4 expression was not decreasedwhen culture was realized at 4 instead of 37°C (Fig. IA). In the same cultures with anti-CD4 mAbs, CDS, and µglobulin antigen expression remained unchanged, whereas a slight decreaseof surface CD3 (lo- 15%) and CD2 (15-20%) MFIs occurred (Fig. 1B). As shown in Fig. lC, CD4 modulation was of similar magnitude with the three other IgGl anti-CD4 mAbs which recognize different epitopes. Modulation was also observed during incubation with BL4 (IgG2a) but with a slightly lower magnitude than with the other anti-CD4 mAbs (Fig. 1C). Dose-dependent down-regulation of CD4. PBL were incubated at 37°C with BB14 mAb at saturating (10 pg/ml) or at lower concentrations (0.02 pg/ml). At different time intervals, cells were sampled,washed,and CD4 expressionwasassessedby labeling with Leu 3a-PE. CD4 modulation was already important at low mAb concentration (0.02 and 0.2 pg/ml) and reached a maximum at nonsaturating (2 pg/ml) mAb concentration (Fig. 2A). Two other experiments reproduced these data, showing that the CD4 down-regulation induced by anti-CD4 mAb was dose dependent and occurred even at low concentrations of anti-CD4 mAb. Reexpression of CD4 after removal of anti-CD4 mAb from medium. In order to evaluate the kinetics of CD4 antigen reexpression in a medium free of BB14 mAb, PBL were incubated at 37°C with BB14 mAb as above, and after a 48-hr period, the medium containing BB14 was removed and cells were washed and incubated in antibody-free medium. Modulation of CD4 persisted for up to 120 hr in presence of mAb, whereas a progressive recovery of cell surface CD4 was observed in absenceof antibody (Fig. 2B). CD4 expression raised from 70 to 87% of the initial density during the 24 hr following antibody removal, but was still slightly below the initial value after 4 days. Role of the Fc part of antibody and cross-linking in CD4 antigenic modulation. Cross-linking of CD4 mAbs with a Gamig (H+L) at 4°C with subsequent incubations at 37°C acceleratedthe kinetics of CD4 modulation but did not increase its magnitude at 24 hr (Fig. 3A). Partial removal of monocytes by L-leucine methyl ester treatment decreasedthe magnitude of CD4 down-regulation, but the phenomenon was still de-
292
MOREL
24
0
40
72
96
ET AL.
120
46
0
Hours
144
96
Hours
FIG. 2. Dose-dependent down-regulation of CD4 and reexpression of CD4 antigen after removal of antiCD4 mAb from medium. (A) PBMC were cultured with increasing concentrations (pg/ml) of BB14 mAb: 0 (a), 0.02 (Cl), 0.2 (0). and 2 (0). CD4 expression was assessed as described in the legend for Fig. 1A. The MFIs measured with 10 pg/ml of BB14 was superimposed with those observed at 2 rg/ml. Results are expressed as in the legend for Fig. 1. (B) PBMC were cultured at 37°C with BB14 at 10 @g/ml. After 48 hr of culture a sample of these cells was put in fresh medium with (0) or without (m) BB14 (10 pg/mI).
monstrable with PBL containing no more than 1% CD14+ monocytes (Fig. 3B). No CD4 modulation could be demonstrated in PBMC cultures incubated with F(ab’)z antibody fragments of BB 14 used at saturating concentration (not shown). However, modulation was partially restored by cross-linking of the F(ab’)z by addition of antiIg (HSL) antibodies (Fig. 3C). Finally to determine whether internalization of CD4anti-CD4 complexes was required for CD4 modulation to occur, PBMC were cultured in plates coated with BB14 mAb. As shown in Fig. 3D, modulation was still demonstrable when using solid phase anti-CD4 mAbs, but it was less pronounced than with soluble mAb. Detection of initially bound anti-CD4 mAbs on lymphocyte surface. To determine whether binding of anti-CD4 mAbs to cell surface CD4 antigens was followed by internalization of all the antigen-antibody complexes, cells were incubated with bi-
201-IO
15
30
60
Minutes
120
,440
0
2448nss,20 HO”M
0
30
60
180 ,200 2400
Minutes
0
30
90
180
1440
Minutes
FIG. 3. Role of the Fc part of antibody and of cross-linking in CD4 antigenic down-regulation. (A) PBMC were treated at 4°C with BBl4 (10 rg/ml) and then with Gamig, followed by washes and a subsequent incubation at 37°C with BB14 (0). In parallel, PBMC were treated in the same way, but without addition of Gamig (0). CD4 expression was assessed with Leu3a-PE. (B) CD4 molecule expression was assessed after culture of PBMC (m) or PBL (0) with BB14 (10 &ml). (C) CD4 molecule expression was assessed after treatment of PBMC at 4°C with BB14 (10 pg/ml) (m) or with F(ab’)z fragments of BB14 at saturating concentration, followed by washes and a subsequent incubation at 37°C in the medium (0) or with Gamig (H+L) (@). (D) PBMC were cultured in plates coated with BB14 mAb (10 &ml, Cl) or with soluble BB14 (0) at 37°C. CD4 expression was assessed by staining with Leu3a-PE.
CD4 DOWN-REGULATION
0.01
0.1
293
BY ANTI-CD4 mAbs
Timb (hr)
10
100
FIG. 4. Decreaseof anti-CD4 mAb bound on the surface of lymphocytes. PBL were submitted to a 30min exposure to biotinylated BB14 and then incubated at 37°C with unlabeled mAb (10 pg/ml). Cells were stained with SAPE (0) which revealed the residual biotinylated BB14 or with Gamig-FITC (0) which stained all the BB 14 bound to the cell surface. Results are expressedas in the legend for Fig. 1.
otinylated BB14 (10 pug/ml) 30 min at 4°C and then washed and incubated at 37°C in culture medium containing unlabeled BB14 (10 pg/ml). At various time intervals, cells were harvested and labeled with SAPE or FITC-Gamig and were processedfor immunofluorescence analysis.The data show that most of the initially bound antibodies were no longer detectable on the cell surface after 24 hr (Fig. 4). Kinetics of Surface CD4 Molecule Expression in Presenceof Anti-CD4 mAb In order to quantify the expression of CD4 molecules on the cell surface, cells were saturated with unlabeled BB14 (10 pg/ml) at 4°C as in previous experiments and then were washed and incubated in the presenceof an identical concentration of ‘251-BB14. At different time intervals, cells were collected and surface-bound radioactivity was determined as the difference between total radioactivity before and after exposure to acid pH. The results show that newly expressedCD4 antigens could bind ‘251-mAbs (= 13,500 antibody molecules/cell) during the first hour of incubation at 37°C and then the number of surface-bound ‘251-mAbmolecules progressively decreased.Intracellular (acid-resistant) radioactivity increased from 1 to 3 hr and then plateaued and
0.1
1 Time (hr)
10
100
0.01
0.1
10 Tim:
100
(hr)
FIG. 5. Expression of new CD4 molecules on lymphocyte surface,as measured by binding of radiolabeled mAbs. PBL were saturated with unlabeled BB14, washed, and then were cultured in the presence of iz51BB14 mAb at 37°C (A) or 4°C (B). Total radioactivity (O), acid-resistant radioactivity (V), membranebound radioactivity (0).
294
MOREL
ET AL.
progressively decreasedafter 20 hr (Fig. 5A). At 4°C a small number (x5000) of free CD4 molecules were expressedon the cell surface at 20 hr but not at 1 hr (Fig. 5B). Nonprotein-bound radioactivity in culture supernatants at 37°C remained within 512%of total activity and did not increase during the 4-day culture period. Therefore, it was not possible to calculate a rate of degradation of ‘251-mAbs. In order to evaluate the relative contribution, in the reexpression of surface CD4 molecules, of de nova synthesis versus recycling of CD4 molecules cleared of their ligand by passagein the endosome/lysosome pathway, we performed the same experiments as above, but with addition of chloroquine, monensin, or cycloheximide. Addition of chloroquine (50 pm, monensin (25 &‘), or a mixture of both agents did not alter the kinetics of binding of ‘25I anti-CD4 mAb (Fig. 6A) during the initial period but slightly decreasedit at 24 and 48 hr. As a control of its activity, the mixture of inhibitors prevented the modulation of the CD3/TcR complexes induced by activation with PMA (10 rig/ml) plus ionomycin (1 pg/ml) (Fig. 6B). It was concluded that blockade of the endosome/lysosome pathway slightly amplified antibody-induced 25000 20000 i
5400 I
OC 0.01
0.1
1
10
0
100
15
24
Time (hr)
Time (hr) B2-m assay
25000 1
3H-Thymidine incorporation
c 50
40
40 ‘:
z 3o F 20
x E,
:
10 0.01
0.1
Tim: (hr)
10
100
0 IhLb. R
20
= 10 0 300 d A
R
A
FIG. 6. Effects of chloroquine, monensin, or cycloheximide on CD4 molecule expression. (A) Cells were preincubated 1 hr with culture medium (0) 50 @chloroquine (O), 25 &I monensin (@), or both inhibitors (6) prior to saturation with unlabeled mAb and culture at 37°C with radiolabeled mAb (10 pg/ml). Data are expressed as number of cell associated mAb molecules. (B) Controls of inhibitor activity, cells treated with (dark bars) or without (clear bars) chloroquine plus monensin were activated with PMA plus ionomycin in order to induce CD3 molecule modulation. The CD3 density on cells was expressed as MFI. (C) Cells were preincubated I hr with (e) or without (0) 10 fig/ml of cycloheximide prior to saturation with unlabeled mAb and culture at 37°C with radiolabeled mAb. Data are expressed as in the legend for Fig. 4A. (D) Controls of cycloheximide activity, cells treated with (dark bars), or without cycloheximide (clear bars) were activated with PMA plus ionomycin (R, resting cells; A, activated cells). At 48 hr, &-microglobulin was assayed in the culture supernatant and proliferation was measured by [‘Hlthymidine incorporation.
CD4 DOWN-REGULATION
295
BY ANTI-CD4 mAbs
down-regulation of surface CD4 expression. Next, PBL were cultured in presence of cycloheximide (10 pg/ml) under which condition [3H]thymidine incorporation and µglobulin release into supernatants by PMA plus ionomycin-activated PBL were completely abrogated (Fig. 6D). The kinetics of 125Ianti-CD4 mAb fixation to cell surfacewas not modified by cycloheximide (Fig. 6C), indicating that CD4 molecules expressedon cell surface during culture with anti-CD4 mAb were preformed but not synthesized during the culture period. Attempts to measure directly the intracellular pool of CD4 molecules by fixation of “‘1-mAb after membrane permeabilization by saponin and saturation of surface CD4 by unlabeled antibodies failed to provide consistent results. Expression of mRNA PBL were cultured in the presence of the anti-CD4 BB14 mAb (10 pg/ml) in the same experimental conditions as described above and at various time intervals, RNA was isolated and processed for Northern blot analysis. f12-Microglobulin and actin mRNAs remained at unchanged levels throughout the culture but CD4 mRNA levels decreasedfrom 20 to 50% at 48 or 72 hr according to experiments. No change in CD4 mRNA levels was observed at 2 and 24 hr. Results of a typical experiment are shown in Fig. 7. DISCUSSION We demonstrate here that anti-human CD4 mAbs of the IgG 1 and IgG2a subclasses which recognize different epitopes can induce the decreaseof cell surface CD4 antigen expression by PBL. Maximal modulation was achieved after 24 to 48 hr of incubation
02-m .5 .5
CD4
-
100
100
/ .4
90
80
70
0.24
1 -
1234
1
2
60 50 ~ 3
4
FIG. 7. Northern blot analysis of µglobulin and CD4 mRNAs in lymphocytes. RNA prepared from lymphocytes were probed with &-microglobulin or CD4 cDNA. They were also hybridized with pactin probe to control the mRNA concentration. Lanes 1 and 3, anti-CD4 treated lymphocytes; lanes 2 and 4, untreated lymphocytes.
296
MOREL
ET AL.
with antibody. It was demonstrated both by measuring the distribution of fluorescence intensities with an anti-CD4 mAb that did not cross-reactwith the modulating antibody, and by counting cell surface radioactivity of lymphocytes incubated with ‘25I antiCD4 mAb. Modulation was preceded by a rapid disappearance of surface-bound antibodies by internalization of CD4-anti-CD4 complexes as already reported with antiCD4 antibodies applied on T cell lines (19) or on other cell lines transfected with the CD4 gene (20, 21). After 24 hr of culture the antibodies initially bound to blood lymphocytes can no longer be demonstrated on the cell surface (Fig. 4). Internalization occurs at 37°C but not at 4°C and the kinetics of disappearance of antibodies do not reveal an heterogeneity among CD4+ lymphocytes. Internalization of CD4 was shown to occur in the absenceof CD4 ligand (22) and it is likely to be enhanced by divalent antibodies. However, internalization of CD4-anti-CD4 complexes does not seem to be required becauseCD4 modulation could be induced by solid phaseanti-CD4 mAbs. Antibody-induced down-regulation of CD4 expression in the present experiments fulfilled the criteria of antigenic modulation as initially described by Boyse et al. (23). It was achieved at 37°C but not at 4°C reversible upon removal of antibody from the culture medium, and induced by anti-CD4 but not by irrelevant mouse IgGl and IgG2. In agreement with Cole and co-workers (24), we observed that CD4 modulation was increased by cross-linking anti-CD4 antibodies with second antibodies or with monocytes, as it has been demonstrated for the modulation of CD3 (25) and CD5 (26). The nearly complete failure of F(ab’)2 fragments of anti-CD4 to induce CD4 modulation is a further evidence for a role of the cross-linking of CD4 molecules. Surface CD4 protein was equally down-regulated on cell CD4+ lymphocytes. Modulation of CD4 was associatedwith a moderate decreaseof surface expression of CD3, asalready reported by Cole et al. (24). The comodulating CD3 molecules may represent the subset of CD3/TcR complexes physically associated with CD4 in resting T cells (11, 27) or those which become associated after addition of anti-CD4 or anti-TcR/ CD3 antibodies (28-30) or after stimulation by antigen (31). The comodulation of CD2 and CD4 has been already reported in cells stimulated with anti-CD2 mAbs (32). It is noteworthy that at variance from some reported anti-CD4 mAbs (33) none of the anti-CD4 mAbs used in our experiments could activate T cells, as evidenced by the lack of proliferative response and the absence of increase of intracellular Ca*+ (data not shown). Therefore, the modulation of surface CD4 reported here must be distinguished from that induced by phorbol esters( 13, 33), which requires phosphorylation of (a) serine residue(s) in the intracytoplasmic tail of CD4 (34, 35) and mobilization of intracellular Ca*+ (33). Internalized CD4 molecules were replaced on the cytoplasmic membrane by preformed CD4 molecules from an intracytoplasmic pool. This pool itself derives primarily from recycled surface CD4 (20,22) and we show here that blockade of the endosome/ lysozome pathway slightly enhances CD4 down-regulation. Conversely, the rate of de lzovoCD4 synthesisshould be very low in resting peripheral blood T cells since cultures with protein synthesis inhibitors in the absence of anti-CD4 mAb induced only a minor decreaseof surfaceCD4 density, not demonstrable before 72 hr (data not shown). Furthermore, cycloheximide did not significantly enhance CD4 modulation in the presenceof anti-CD4 mAb (Fig. 5), suggestingthat de IZOVO CD4 synthesis was already minimal in modulated cells. However, a slightly decreasedexpression of CD4 mRNA was observedin cells incubated with anti-CD4 mAb, asalready reported after activation with anti-CD3 plus PMA (36, 37). The down-regulation of surface CD4 expression
CD4 DOWN-REGULATION
BY ANTI-CD4 mAbs
297
was reported in cells transfected with the HIV- 1 nefgene (38) and in T cells incubated with dextran sulfate (39) or gangliosides (40,4 1). Similarly, in cells infected with HIV, a down-regulation of CD4 expression has been extensively documented, but different mechanisms could account for this alteration including decreasedsteady-state levels of CD4 transcripts, reduced levels of CD4 proteins and formation of CD4-gp120 complexes (42-44). The present data with anti-CD4 mAbs should also be compared with the alterations of surface CD4 protein expression induced by HIV envelope glycoproteins in noninfected cells. Weinhold et al. showed that gp 120 exhibited a dosedependent suppressive effect of T lymphocyte responsesto specific antigens or OKT3 antibodies but failed to demonstrate a down-regulation of surface CD4 expression in cells incubated for 12 hr with gp120 (45). Conversely, Juszczak and co-workers (46) recently reported that exposure of the T cell line CEM-CM3 to gp 120 (5 fig/ml) resulted in a down-regulation of surface CD4 protein expression comparable to that reported here with anti-CD4 mAbs. In their experiments the authors also described rapid dissociation of the CD4-~56”’ complex which was also observed in the presence of the Leu3a anti-CD4 mAb but not with another anti-CD4 mAb. One may postulate that common signals generated by CD4 ligands could induce the down-regulation of CD4 protein and the inhibition of T cell responses,but the precise relationship between these effects remains to be investigated. ACKNOWLEDGMENTS This work was supported by Institut National de la Sante et de la recherche medicale and Centre National de la Recherche Scientitique and by a grant from the Department of Human Biology, University Lyon I. Patricia Morel was recipient of a fellowship from the Fondation Marcel Merieux. We thank G. Panaye and G. Lizard (Service commun de cytofluorometrie) for expert assistancein flow cytometry and M. J. Gariazzo for her skillful1 assistance.
REFERENCES 1. 2. 3. 4. 5.
Reinherz, E. L., and Schlossman,F., Cell 19, 821, 1980. Terhost, C., Van Agthoven, A., Reinherz, E., and Schlossman,S., Science 209, 520, 1980. Biddinson, W., Rao, P., Talle, M., Goldstein, G., and Shaw, S., J. Immunol. 131, 152, 1983. Doyle, C., and Strominger, J. L., Nature 330,256, 1987. Dalgleish, A. G., Beverley, P. C. L., Clapham, P. R., Crawford, D. H., Greaves, M. F., and Weiss, R. A., Nature 312, 763, 1984. 6. Klatzmann, D., Champagne, E., Chamaret, S., Gruest, J., Guetard, D., Hercend, T., Gluckman, J. C., and Montagnier, L., Nature 312, 767, 1984. 7. Rudd, C., Trevillyan, J., Dasgupta, J., Wong, L., and Schlossman, S., Proc. Natl. Acad. Sci. USA 85, 5190, 1988. 8. Veillette, A., Bookman, M., Horak, E., and Bolen, J., Cell 55, 301, 1988. 9. Veillette, A., Bookman, M. A., Horak, E. M., Samelson, L. E., and Bolen, J. B., Nature 338, 257, 1989. 10. Veillette, A., Bolen, J. B., and Bookman, M. A., Mol. Cell Biol. 9, 4441, 1989. 11. Anderson, P., Blue, M. L., and Schlossman,S. F., J. Immunol. 140, 1732, 1988. 12. Weyand, C. M., Goronzy, J., and Fathman, C. G., J. Zmmunol. 138, 1351, 1987. 13. Hoxie, J. A., Matthews, D. M., Callahan, K. J., Cassel, D. L., and Cooper, R. A., J. Immunol. 137, 1194, 1986. 14. Carrel, S., Moretta, A., Pantaleo, G., Tambussi, G., Isler, P., Perussia, B., and Cerottini, J. C., Eur. J. Immunol. 18, 333, 1988.
15. Morel, P., Nicolas, J. F., Wijdenes, J., and Revillard, J. P., C/in. Immunol. Immunopathol. 64, 248, 1992. 16. Vincent, C., and Revillard, J. P., In “Methods of Enzymatic Analysis” (H. U. Bergmeyer, Ed), Vol. 9, pp. 248-265. VCH VerlagsgesellschafimbH, D-6940 Weinheim, Germany. 17. Thiele, D. L., Kurosaka, M., and Lipsky, P. E., J. Immunol. 131, 2282, 1983.
298
MOREL
ET AL.
18. Suggs,S. V., Wallace, R. B., Hirose, T., Kawashima, E. H., and Itakura, K., Proc. N&l. Acud. USA 78, 6613, 1981. 19. Caniere, D., Artier, J. M., Derocq, J. M., Fontaine, C., and Richer, G., Exp. Cell. Rex 182, 114, 1989. 20. Pelchen-Matthews, A., Armes, J. A., and Marsh, M., EMBO J. 8, 3641, 1989. 21. Marsh, M., Armes, J. E., and Pelchen-Matthews, A., Biochem. Sot. Trans. 18, 139, 1989. 22. Pelchen-Matthews, A., Armes, J. E., Griffiths, G., and Marsh, M., J. Exp. Med. 173, 575, 1991. 23. Boyse, E. A., Stocker& E., and Old, L. J., Proc. Natl. Acad. Sri. USA 58, 954, 1967. 24. Cole, J. A., McCarthy, A., Rees, M. A., Sharrow, S., and Singer, A., J. Immunol. 143, 397, 1989. 25. Rinnooy Kan, E. A., Platzer, E., Welte, K., and Wang, C. Y., J. Immunol. 133, 2979, 1984. 26. Schroff, R. W., Klein, R. A., Farrel, M. M., and Stevenson, H. C., J. Immunol. 133,2270, 1984. 27. Gallagher, P. F., De St. Growth, B. F., and Miller, J. F., Proc. Nat/. Acud. Sci. USA 86, 10044, 1989. 28. Chuck, R. S., Cantor, C. R., and Tse, D. B., Proc. Natl. Acad. USA 87, 5021, 1990. 29. Dianzani, U., Shaw, A., al-Ramadi, B. K., Kubo, R. T., and Janeway, C. A. J., J. Immunol. 148,678, 1992. 30. Rojo, J. M., Saizawa, K., and Janeway, C. A. J., Proc. Natl. Acud. Sci. USA 86, 3311, 1989. 31. Mittler, R. S., Goldman, S. J., Spitalny, G. L., and Burakoff, S. J., Proc. Nat!. Acad. Sci. USA 86, 853 1, 1989. 32. Bueso-Ramos,C. E., Donahoe, R. M., Nicholson, J. K. A., Madden, J. J., and Falek, A., J. Immunol. 140, 1414, 1988.
33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46.
Bigby, M., Wang, P., Fierro, J. F., and Sy, M. S., J. Immunoi. 144, 3 111, 1990. Shin, J., Doyle, C., Yang, Z., Kappes, D., and Strominger, J. L., EMBO J. 9, 425, 1990. Shin, J., Dunbrack, R. L., Lee, S., and Strominger, J. L., J. Biol. Chem. 266, 10658, 1991. Paillard, F., Sterkers, G., and Vaquero, C., EM30 J. 9, 1867, 1990. Neudorf, S., Jones, M., Parker, S., and Lattier, D., J. Immunol. 146, 2836, 1991. Garcia, J. V., and Miller, A. D., Nature 350, 508, 1991. Thiele, B., Braig, H. R., Ehm, I., Kunze, R., and Ruf, B., Eur. J. Immunol. 19, 1161, 1989. Offner, H., Thieme, T., and Vandenbark, A. A., J. Immunol. 139, 3295, 1987. Suthanthiran, M., J. Exp. Med. 171, 1965, 1990. Hoxie, J. A., Alpers, J. D., Rackowski, J. L., Huebner, K., Haggarty, B. S., Cedarbaum, A. J., and Reed, J. C., Science 234, 1123, 1986. Yuille, M. A., Hugunin, M., John, P., Sacks,L. V., Poiesz, B. J., Tomar, R. H., and Silverstone, A. E., J. Acquired Immune Defic. Syndrome 1, 13 1, 1988. Salmon, P., Olivier, R., Riviere, Y., Brisson, E., Gluckman, J. C., Kieny, M. P., Montagnier, L., and Klatzmann, D., J. Exp. Med. 168, 1953, 1988. Weinhold, K. J., Lyerly, H. K., Stanley, S. D., Austin, A. A., Matthews, T. J., and Bolognesi, D. P., J. Immunol. 142,3091, 1989. Juszczak, R. J., Turchin, H., Truneh, A., Culp, J., and Kassis, S., J. Biol. Chem. 266, 11176, 1991.
IMMUNOLOGY
145,287-298
(1992)
Down-Regulation of Cell Surface CD4 Molecule Expression by Anti-CD4 Antibodies in Human T Lymphocytes P. MOREL,*
Induced
C. VINCENT,* J. WIJDENES,~AND J. P. REVILLARD*
*Immunology Laboratory, INSERM lJ80, CNRS URA 1177, Lyon, France; and tlnnotherapie Laboratoires, Besancon. France Received June 16. 1992; acceptedAugust 19, I992 Antigenic modulation was defined as the down-regulation of a cell surface antigen expression induced by exposureto specificantibody. We investigatedthe modulation of CD4 surfaceexpression in human peripheral blood lymphocytes incubated in vitro with anti-CD4 monoclonal antibodies (mAbs). Modulation of surface CD4 was achieved at 37°C but not at 4°C. with five different murine anti-CD4 mAbs of IgGl and IgC2a subclasses,with different epitope specificities. Modulation wasdosedependentwith a maximum at nonsaturating mAb concentration. It was reversible upon culture in mAb-free medium. It was acceleratedand amplified in the presenceof monocytes or after cross-linking of anti-CD4 mAbs. It could be induced with solid phase anti-CD4 mAbs. but not with soluble F(ab’)r fragments. Its magnitude was identical on all CD4+ lymphocytes. It was associatedwith a moderate down-regulation of CD2 and CD3 but not of CD8 and HLA class I surfaceexpression.Modulation was slightly augmentedby addition of inhibitors of the endosome/ lysosome pathway but not by protein synthesis inhibitors. The anti-CD4 mAb initially bound to cell surfacewas no longer detectableafter 24 hr of culture. Most of surfaceCD4 proteins complexed with antibody were rapidly internalized and transiently replaced by CD4 from an intmcytoplasmic pool and then no longer were expressed.CD4 mRNA was moderately decreasedin cells incubated with anti-CD4 mAb while @-actinand µgJobulin mRNAs remained at stable levels. It was concluded that down-regulation of CD4 surfaceexpression induced by anti-CD4 mAb concerned only a part of CD4 molecules and was associatedwith a decreasedsynthesis.The delay required to achieve maximal modulation is likely to reflect exhaustion of the intracytoplasmic recycling 0 1992 Academic press, Inc. pool of CD4 IIIOkCUkS.
INTRODUCTION The CD4 antigen is a 55-kDa glycoprotein expressedon the surface of a functionally distinct subset of human T lymphocytes and on some cells of the reticuloendothelial system (1, 2). This molecule interacts with class II antigens expressed on antigenpresenting cells and functions as a coreceptor in the activation of T cells by the TcR/ CD3 pathway (3,4). The CD4 molecule has also been identified as a receptor for HIV (5,6). On human and murine T cells, CD4 is physically associatedwith the &-related tyrosine protein kinase ~56“‘~ (7, 8). CD4 can function as a signal transducer and tyrosine kinase phosphorylation events may be important in CDCmediated signaling. Cross-linking of the CD4 receptor induces a rapid increase in the tyrosine-specific protein kinase activity of ~56”~ and is associated with the rapid phosphorylation of one of the subunits ({) of the TcR complex on tyrosine residues(9, 10). T cell activation by the TcR/CD3 or by the CD2 pathways, as well as direct stimulation by phorbol 287 0008~8749192 $5.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
288
MOREL
ET AL.
esters,induces receptor cointernalization and a transient defective expression of both CD4 and TcR/CD3 on T lymphocyte surface ( 11- 13). Modulation of CD4 molecule expression can therefore be induced by T cell activation, e.g., with anti-CD2 and antiCD3 antibodies. With one known exception (14) anti-CD4 antibodies alone cannot trigger T cell activation. However, during the course of in vivo administration of antiCD4 mAb in patients with rheumatoid arthritis or psoriasis, we observed a markedly decreasedexpression of CD4 molecules at the cell surface ( 15). Therefore, the present study was addressedto the mechanisms whereby anti-CD4 antibodies could regulate cell surface expression of CD4 antigen in vitro. MATERIALS AND METHODS Monoclonal Antibodies (mAbs) Five anti-CD4 mAbs were used in this study (Table 1). The equilibrium constant, K, was calculated according to Scatchard analysis of saturation curves established with ‘251-labeledmAbs. The saturating concentration was defined as the lowest amount of antibody needed to obtain a plateau in the conditions of cell labeling for immunofluorescence. The following anti-CD4 mAbs were used: BL4, initially produced by J. Brochier in our laboratory (15). BL4 is a mouse IgG2a which recognizes an epitope of group 2B. BB 14 and BF8 are mouse IgG 1 purified from ascites by chromatography, produced by J. Wijdenes at the Centre Regional de Transfusion Sanguine. The 13B8.2, IgGl, was obtained from Dr. Mawas (INSERM U 119, Marseille, France) and FlO l-69, IgG 1, from Dr. Poncelet (Sanofi Research Center, Montpellier, France). F(ab’)* fragments from BB 14 were prepared by bromelain digestion and purified by exclusion on protein A-Sepharose chromatography (Pharmacia Uppsala Sweden). A sample of BB14 was biotinylated with Act-biotin (IBF, France) according to the instruction sheet. Its fixation was revealed with streptavidin phycoerythrin (SAPE) (Caltag, San Francisco, CA). Leu 3a-phycoerythrin (PE) (IgG 1, Becton-Dickinson, Mountain View, CA) and OKTCfluorescein isothiocyanate (FITC) (IgG2b, Ortho Pharmaceutical Corp., Raritan, NJ), which do not cross-reactwith the anti-CD4 mAb described above, were used to assessCD4 expression. MAbs of other specificity were used as control: anti-CD2, LeuSb-FITC; anti-CD3, Leu 4-FITC; anti-CD8, Leu2a-PE (Becton-Dickinson), and a mouse anti-p2-microglobulin mAb (BE104, IgG2a), developed in our laboratory (16). Chemicals Chloroquine, monensin, bacitracin, cycloheximide, L-leucine methyl ester, and phorbol myristate acetate (PMA) were obtained from Sigma (St. Louis, MO). IodoTABLE
1
Characteristics of the Anti-CD4
Isotype Saturating dose (wg/ml)
K, L X hi-'
mAbs
BB14
BF8
BL4
13B8.2
FlOl-69
IgG I 5-10 3.7 x lo*
IgG 1 5-10
IgG2a 10
5.6 X IO'
5.5 x IO'
IgG 1 10 1.3 x 10s
IgGl 10 1.9 x lo*
CD4 DOWN-REGULATION
BY ANTI-CD4
mAbs
289
gen was obtained from Pierce (Rockford, IL) and phthalate oils were obtained from Fluka (Buchs CH). Ionomycin was obtained from Calbiochem (La Jolla, CA). lmmunojluorescence Assay PBMC from healthy donors were separated by centrifugation on Ficoll-Hypaque (Lymphoprep, Nycone A.S., Norway). To assessthe degreeof CD4 saturation by the mAb, 20 ~1 of serial dilutions of anti-CD4 mAb were added on PBMC pellets (5 X 105),30 min at 4°C. Cells were then washed with PBS, stained with polyclonal goatanti-mouse Ig (H+L) antibodies conjugated with FITC (GAMIg-FITC, Tago Immunologicals, Burlingame, CA), and were fixed with phosphate-buffered saline (PBS) containing bovine serum albumin (BSA), azide ( 1%) and formaldehyde (1%). CD4 antigen expression on PBMC was measured by flow cytometry using a FACScan analyzer (Becton-Dickinson) and was analyzed with the Lysis software. Analysis was performed with a scatter gated on lymphocyte fraction, excluding monocytes, polymorphs, and dead cells. Antigen density on the cell surface was evaluated by specific mean fluorescence intensity (MFI) measured in arbitrary units (log scale). Results are expressedas the percentage of the initial MFI obtained in saturating conditions, after substraction of MFI measured on the same cells in the absenceof mAb. To assessthe effect of anti-CD4 mAb on CD4 antigen expression, PBMC were incubated with an anti-CD4 mAb (10 pg/106 cells/ml) at 37°C 7.5% COz, in RPM1 1640culture medium supplemented with glutamine (2 mM), penicillin, streptomycin, and 10% fetal calf serum (FCS). At different time intervals, cells were sampled and washed twice in PBS and then CD4 expression was assessedby staining with a fluoresceinated anti-CD4 mAb. As control, the expressions of CD2, CD3, CD8, and & microglobulin were analyzed in the same experiment. Control of the cycloheximide activity was assessedby measuring [3H]thymidine incorporation into cell nuclei during the last 16 hr of 3-day cultures after stimulation with PMA plus ionomycin. In some experiments, monocytes were eliminated from PBMC with L-leucine methyl ester (Sigma, 5 mM, 5 X IO6 cells/ml, 40 min at room temperature) according to Thiele et al. ( 17). L-Leucine methyl ester is a lysosomotropic agent rapidly metabolized into leucine inside the lysosomes, causing osmotic swelling of these organelles and subsequent cell death. Monocyte depletion was assessedby flow cytometric analysis after staining with LeuM3 (CD14) mAb (Becton-Dickinson). After L-leucine methyl ester treatment less than 1% of the monocytes were detected in the cell suspensions. These cell suspensions are referred to as peripheral blood lymphocytes (PBL). pzMicroglobulin concentrations were measured with an ELISA method described elsewhere ( 16). Assays with Radiolabeled Antibodies MAbs were isolated either by filtration on DEAE-Sephadex A50 in 0.1 M phosphate buffer at pH 7.0 or by chromatography on protein A-Sepharose with a 20 mM phosphate buffer, pH 7.80, with 1 A4 NaCl, followed by an elution with 0.1 M glycine/ HCl, pH 2.7, buffer. Antibody molecules were labeled by the Iodo-gene method with a specific activity of 300-350 MBq/mg. Labeled molecules were diluted in RPM1 1640 medium supplemented with 10% FCS and antibiotics were then used within 1 hr.
290
MOREL
ET AL.
For binding assays,PBL (l-4 X 106/ml) were incubated at 4°C or at 37°C with 10 &ml of ‘251-labeledCD4 mAb in RPM1 1640 medium containing 10% FCS. After various periods of incubation, unbound 1251-labeled mAb was removed by layering of 50 ~1 samples on a cushion of 300 ~1 of phthalate oil (1 vol of phthalate-dibutylester and 1.1 vol of bis(2-ethylhexyl) phthalate) in a 0.7-ml Eppendorf microtube. After 4 min of centrifugation at 12,OOOg at 4°C the bottom of the tube which contained the cell pellet was cut and counted in a gamma counter. The remaining part of the tube was also counted in order to calculate total activity. Acid resistant activity was measured by addition of 50 ~1of 0.1 A4glycine/HCl, pH 2.7, buffer on the cushion of phthalate oil together with the cells. Centrifugation was performed after 5 min of incubation in acid buffer. Isolation and Identification of mRNA We used the following DNA probes specific for CD4 RNA (clone pT4B, a kind gift from Dr. Piatier, CNRS Nogent-sur-Maine), µglobulin RNA (18) (gift from F. Lemonnier, Villejuif), and P-actin RNA (pACP, gift from J. P. Rouault, Lyon). These probes were labeled by random priming (random priming kit, Amersham) with [32P]dCTP. For Northern blot analysis, total RNA from 5 X lo6 PBL was prepared by a modified phenol/chloroform/isoamyl alcohol extraction method. Five-microgram samples of RNA were electrophoresed through a 1% agarosegel containing 6% formaldehyde in the running buffer. RNAs were blotted onto a nylon membrane (Hybond N, Amersham) by the standard capillary blotting technique. The blots were fixed by heating 2 hr at 80°C and prehybridized 2 hr at 42°C in 50% formamide, 5X SSC,5X Denhardt’s solution, 0.5% SDS, 5% dextran, and 0.25 mg/ml of denatured salmon sperm DNA. Blots were incubated overnight at 42°C with 20-30 ng of radiolabeled DNA probes. Blots were rinsed in 2X SSC,0.1% SDS at room temperature and were washed twice in the samesolution 20 min at 65°C. A last wash step was done in 2X SSC,0.2% SDS at 65°C and they were finally exposed to Kodak XOMAT film for 12-48 hr at 70°C. Size of RNA were determined by comparison with 0.24- to 9.5-kb RNA Ladder (GIBCO BRL). RESULTS DecreasedMembrane Density of CD4 Molecules on T Cells Incubated with Anti-CD4 mAbs Kinetics of CD4 density changes on human PBL in the presence of anti-CD4 mAb. PBMC were incubated at 37°C with BB 14 mAb at saturating concentration ( 1O6cells/ 10 pg/ml). At various time intervals, cells were stained with a phycoerythrin-conjugated anti-CD4 mAb (Leu3a-PE) which binds to another epitope than BB14. CD4 downregulation occurred within the first 24 hr. It was maximal between 24 and 48 hr with a loss of 40 to 50% of the initial MFI (Fig. IA). Despite a minor spontaneous increase after 48 hr of culture, the MFI remained 26% lower than the initial value after 5 days of culture (Fig. 1A). As control, CD4 expression remained stable on PBMC incubated in antibody-free medium. During all the culture period, the percentage of CD4+stained cells remained stableand CD4 expressionon the cells was homogenous, showing that the antigenic modulation occurred in all the CD4’ lymphocytes with the same
CD4 DOWN-REGULATION A
100
3 E80 8 6 a960 s !i 40
291
BY ANTI-CD4 mAbs
100
L I
0
24
80 60
I
I
I
48
72
96
Hours
I
120
0
48
96
Hours
144
0
24
48
72
Hours
FIG. 1. In vitro down-regulation of CD4 surface antigen induced by anti-CD4 mAbs on lymphocytes. (A) PBMC were cultured at 37°C with(m) or without BB14 (0) and at 4°C with BB14 (e). SurfaceCD4 proteins were labeled at various time intervals with Leu3a-PE. Resultsare expressedas percentageof the MFI obtained with saturating doses,as described under Material and Methods. (B) Effect of BB14 on CD2, CD3, CD8, and µglobulin expression. PBMC were cultured with BB14 at 37°C and were stained with antibodies against CD2 (0) CD3 (0) CD8 (0) &-microglobulin (A), or CD4 (m). (C) PBMC were cultured with three different IgGl anti-CD4 mAbs (0, 0, A) or one IgG2a anti-CD4 mAb, BL4 (Cl) at 37°C. CD4 expression was assessedwith OKT4-FITC or Leu3a-PE. intensity. CD4 expression was not decreasedwhen culture was realized at 4 instead of 37°C (Fig. IA). In the same cultures with anti-CD4 mAbs, CDS, and µglobulin antigen expression remained unchanged, whereas a slight decreaseof surface CD3 (lo- 15%) and CD2 (15-20%) MFIs occurred (Fig. 1B). As shown in Fig. lC, CD4 modulation was of similar magnitude with the three other IgGl anti-CD4 mAbs which recognize different epitopes. Modulation was also observed during incubation with BL4 (IgG2a) but with a slightly lower magnitude than with the other anti-CD4 mAbs (Fig. 1C). Dose-dependent down-regulation of CD4. PBL were incubated at 37°C with BB14 mAb at saturating (10 pg/ml) or at lower concentrations (0.02 pg/ml). At different time intervals, cells were sampled,washed,and CD4 expressionwasassessedby labeling with Leu 3a-PE. CD4 modulation was already important at low mAb concentration (0.02 and 0.2 pg/ml) and reached a maximum at nonsaturating (2 pg/ml) mAb concentration (Fig. 2A). Two other experiments reproduced these data, showing that the CD4 down-regulation induced by anti-CD4 mAb was dose dependent and occurred even at low concentrations of anti-CD4 mAb. Reexpression of CD4 after removal of anti-CD4 mAb from medium. In order to evaluate the kinetics of CD4 antigen reexpression in a medium free of BB14 mAb, PBL were incubated at 37°C with BB14 mAb as above, and after a 48-hr period, the medium containing BB14 was removed and cells were washed and incubated in antibody-free medium. Modulation of CD4 persisted for up to 120 hr in presence of mAb, whereas a progressive recovery of cell surface CD4 was observed in absenceof antibody (Fig. 2B). CD4 expression raised from 70 to 87% of the initial density during the 24 hr following antibody removal, but was still slightly below the initial value after 4 days. Role of the Fc part of antibody and cross-linking in CD4 antigenic modulation. Cross-linking of CD4 mAbs with a Gamig (H+L) at 4°C with subsequent incubations at 37°C acceleratedthe kinetics of CD4 modulation but did not increase its magnitude at 24 hr (Fig. 3A). Partial removal of monocytes by L-leucine methyl ester treatment decreasedthe magnitude of CD4 down-regulation, but the phenomenon was still de-
292
MOREL
24
0
40
72
96
ET AL.
120
46
0
Hours
144
96
Hours
FIG. 2. Dose-dependent down-regulation of CD4 and reexpression of CD4 antigen after removal of antiCD4 mAb from medium. (A) PBMC were cultured with increasing concentrations (pg/ml) of BB14 mAb: 0 (a), 0.02 (Cl), 0.2 (0). and 2 (0). CD4 expression was assessed as described in the legend for Fig. 1A. The MFIs measured with 10 pg/ml of BB14 was superimposed with those observed at 2 rg/ml. Results are expressed as in the legend for Fig. 1. (B) PBMC were cultured at 37°C with BB14 at 10 @g/ml. After 48 hr of culture a sample of these cells was put in fresh medium with (0) or without (m) BB14 (10 pg/mI).
monstrable with PBL containing no more than 1% CD14+ monocytes (Fig. 3B). No CD4 modulation could be demonstrated in PBMC cultures incubated with F(ab’)z antibody fragments of BB 14 used at saturating concentration (not shown). However, modulation was partially restored by cross-linking of the F(ab’)z by addition of antiIg (HSL) antibodies (Fig. 3C). Finally to determine whether internalization of CD4anti-CD4 complexes was required for CD4 modulation to occur, PBMC were cultured in plates coated with BB14 mAb. As shown in Fig. 3D, modulation was still demonstrable when using solid phase anti-CD4 mAbs, but it was less pronounced than with soluble mAb. Detection of initially bound anti-CD4 mAbs on lymphocyte surface. To determine whether binding of anti-CD4 mAbs to cell surface CD4 antigens was followed by internalization of all the antigen-antibody complexes, cells were incubated with bi-
201-IO
15
30
60
Minutes
120
,440
0
2448nss,20 HO”M
0
30
60
180 ,200 2400
Minutes
0
30
90
180
1440
Minutes
FIG. 3. Role of the Fc part of antibody and of cross-linking in CD4 antigenic down-regulation. (A) PBMC were treated at 4°C with BBl4 (10 rg/ml) and then with Gamig, followed by washes and a subsequent incubation at 37°C with BB14 (0). In parallel, PBMC were treated in the same way, but without addition of Gamig (0). CD4 expression was assessed with Leu3a-PE. (B) CD4 molecule expression was assessed after culture of PBMC (m) or PBL (0) with BB14 (10 &ml). (C) CD4 molecule expression was assessed after treatment of PBMC at 4°C with BB14 (10 pg/ml) (m) or with F(ab’)z fragments of BB14 at saturating concentration, followed by washes and a subsequent incubation at 37°C in the medium (0) or with Gamig (H+L) (@). (D) PBMC were cultured in plates coated with BB14 mAb (10 &ml, Cl) or with soluble BB14 (0) at 37°C. CD4 expression was assessed by staining with Leu3a-PE.
CD4 DOWN-REGULATION
0.01
0.1
293
BY ANTI-CD4 mAbs
Timb (hr)
10
100
FIG. 4. Decreaseof anti-CD4 mAb bound on the surface of lymphocytes. PBL were submitted to a 30min exposure to biotinylated BB14 and then incubated at 37°C with unlabeled mAb (10 pg/ml). Cells were stained with SAPE (0) which revealed the residual biotinylated BB14 or with Gamig-FITC (0) which stained all the BB 14 bound to the cell surface. Results are expressedas in the legend for Fig. 1.
otinylated BB14 (10 pug/ml) 30 min at 4°C and then washed and incubated at 37°C in culture medium containing unlabeled BB14 (10 pg/ml). At various time intervals, cells were harvested and labeled with SAPE or FITC-Gamig and were processedfor immunofluorescence analysis.The data show that most of the initially bound antibodies were no longer detectable on the cell surface after 24 hr (Fig. 4). Kinetics of Surface CD4 Molecule Expression in Presenceof Anti-CD4 mAb In order to quantify the expression of CD4 molecules on the cell surface, cells were saturated with unlabeled BB14 (10 pg/ml) at 4°C as in previous experiments and then were washed and incubated in the presenceof an identical concentration of ‘251-BB14. At different time intervals, cells were collected and surface-bound radioactivity was determined as the difference between total radioactivity before and after exposure to acid pH. The results show that newly expressedCD4 antigens could bind ‘251-mAbs (= 13,500 antibody molecules/cell) during the first hour of incubation at 37°C and then the number of surface-bound ‘251-mAbmolecules progressively decreased.Intracellular (acid-resistant) radioactivity increased from 1 to 3 hr and then plateaued and
0.1
1 Time (hr)
10
100
0.01
0.1
10 Tim:
100
(hr)
FIG. 5. Expression of new CD4 molecules on lymphocyte surface,as measured by binding of radiolabeled mAbs. PBL were saturated with unlabeled BB14, washed, and then were cultured in the presence of iz51BB14 mAb at 37°C (A) or 4°C (B). Total radioactivity (O), acid-resistant radioactivity (V), membranebound radioactivity (0).
294
MOREL
ET AL.
progressively decreasedafter 20 hr (Fig. 5A). At 4°C a small number (x5000) of free CD4 molecules were expressedon the cell surface at 20 hr but not at 1 hr (Fig. 5B). Nonprotein-bound radioactivity in culture supernatants at 37°C remained within 512%of total activity and did not increase during the 4-day culture period. Therefore, it was not possible to calculate a rate of degradation of ‘251-mAbs. In order to evaluate the relative contribution, in the reexpression of surface CD4 molecules, of de nova synthesis versus recycling of CD4 molecules cleared of their ligand by passagein the endosome/lysosome pathway, we performed the same experiments as above, but with addition of chloroquine, monensin, or cycloheximide. Addition of chloroquine (50 pm, monensin (25 &‘), or a mixture of both agents did not alter the kinetics of binding of ‘25I anti-CD4 mAb (Fig. 6A) during the initial period but slightly decreasedit at 24 and 48 hr. As a control of its activity, the mixture of inhibitors prevented the modulation of the CD3/TcR complexes induced by activation with PMA (10 rig/ml) plus ionomycin (1 pg/ml) (Fig. 6B). It was concluded that blockade of the endosome/lysosome pathway slightly amplified antibody-induced 25000 20000 i
5400 I
OC 0.01
0.1
1
10
0
100
15
24
Time (hr)
Time (hr) B2-m assay
25000 1
3H-Thymidine incorporation
c 50
40
40 ‘:
z 3o F 20
x E,
:
10 0.01
0.1
Tim: (hr)
10
100
0 IhLb. R
20
= 10 0 300 d A
R
A
FIG. 6. Effects of chloroquine, monensin, or cycloheximide on CD4 molecule expression. (A) Cells were preincubated 1 hr with culture medium (0) 50 @chloroquine (O), 25 &I monensin (@), or both inhibitors (6) prior to saturation with unlabeled mAb and culture at 37°C with radiolabeled mAb (10 pg/ml). Data are expressed as number of cell associated mAb molecules. (B) Controls of inhibitor activity, cells treated with (dark bars) or without (clear bars) chloroquine plus monensin were activated with PMA plus ionomycin in order to induce CD3 molecule modulation. The CD3 density on cells was expressed as MFI. (C) Cells were preincubated I hr with (e) or without (0) 10 fig/ml of cycloheximide prior to saturation with unlabeled mAb and culture at 37°C with radiolabeled mAb. Data are expressed as in the legend for Fig. 4A. (D) Controls of cycloheximide activity, cells treated with (dark bars), or without cycloheximide (clear bars) were activated with PMA plus ionomycin (R, resting cells; A, activated cells). At 48 hr, &-microglobulin was assayed in the culture supernatant and proliferation was measured by [‘Hlthymidine incorporation.
CD4 DOWN-REGULATION
295
BY ANTI-CD4 mAbs
down-regulation of surface CD4 expression. Next, PBL were cultured in presence of cycloheximide (10 pg/ml) under which condition [3H]thymidine incorporation and µglobulin release into supernatants by PMA plus ionomycin-activated PBL were completely abrogated (Fig. 6D). The kinetics of 125Ianti-CD4 mAb fixation to cell surfacewas not modified by cycloheximide (Fig. 6C), indicating that CD4 molecules expressedon cell surface during culture with anti-CD4 mAb were preformed but not synthesized during the culture period. Attempts to measure directly the intracellular pool of CD4 molecules by fixation of “‘1-mAb after membrane permeabilization by saponin and saturation of surface CD4 by unlabeled antibodies failed to provide consistent results. Expression of mRNA PBL were cultured in the presence of the anti-CD4 BB14 mAb (10 pg/ml) in the same experimental conditions as described above and at various time intervals, RNA was isolated and processed for Northern blot analysis. f12-Microglobulin and actin mRNAs remained at unchanged levels throughout the culture but CD4 mRNA levels decreasedfrom 20 to 50% at 48 or 72 hr according to experiments. No change in CD4 mRNA levels was observed at 2 and 24 hr. Results of a typical experiment are shown in Fig. 7. DISCUSSION We demonstrate here that anti-human CD4 mAbs of the IgG 1 and IgG2a subclasses which recognize different epitopes can induce the decreaseof cell surface CD4 antigen expression by PBL. Maximal modulation was achieved after 24 to 48 hr of incubation
02-m .5 .5
CD4
-
100
100
/ .4
90
80
70
0.24
1 -
1234
1
2
60 50 ~ 3
4
FIG. 7. Northern blot analysis of µglobulin and CD4 mRNAs in lymphocytes. RNA prepared from lymphocytes were probed with &-microglobulin or CD4 cDNA. They were also hybridized with pactin probe to control the mRNA concentration. Lanes 1 and 3, anti-CD4 treated lymphocytes; lanes 2 and 4, untreated lymphocytes.
296
MOREL
ET AL.
with antibody. It was demonstrated both by measuring the distribution of fluorescence intensities with an anti-CD4 mAb that did not cross-reactwith the modulating antibody, and by counting cell surface radioactivity of lymphocytes incubated with ‘25I antiCD4 mAb. Modulation was preceded by a rapid disappearance of surface-bound antibodies by internalization of CD4-anti-CD4 complexes as already reported with antiCD4 antibodies applied on T cell lines (19) or on other cell lines transfected with the CD4 gene (20, 21). After 24 hr of culture the antibodies initially bound to blood lymphocytes can no longer be demonstrated on the cell surface (Fig. 4). Internalization occurs at 37°C but not at 4°C and the kinetics of disappearance of antibodies do not reveal an heterogeneity among CD4+ lymphocytes. Internalization of CD4 was shown to occur in the absenceof CD4 ligand (22) and it is likely to be enhanced by divalent antibodies. However, internalization of CD4-anti-CD4 complexes does not seem to be required becauseCD4 modulation could be induced by solid phaseanti-CD4 mAbs. Antibody-induced down-regulation of CD4 expression in the present experiments fulfilled the criteria of antigenic modulation as initially described by Boyse et al. (23). It was achieved at 37°C but not at 4°C reversible upon removal of antibody from the culture medium, and induced by anti-CD4 but not by irrelevant mouse IgGl and IgG2. In agreement with Cole and co-workers (24), we observed that CD4 modulation was increased by cross-linking anti-CD4 antibodies with second antibodies or with monocytes, as it has been demonstrated for the modulation of CD3 (25) and CD5 (26). The nearly complete failure of F(ab’)2 fragments of anti-CD4 to induce CD4 modulation is a further evidence for a role of the cross-linking of CD4 molecules. Surface CD4 protein was equally down-regulated on cell CD4+ lymphocytes. Modulation of CD4 was associatedwith a moderate decreaseof surface expression of CD3, asalready reported by Cole et al. (24). The comodulating CD3 molecules may represent the subset of CD3/TcR complexes physically associated with CD4 in resting T cells (11, 27) or those which become associated after addition of anti-CD4 or anti-TcR/ CD3 antibodies (28-30) or after stimulation by antigen (31). The comodulation of CD2 and CD4 has been already reported in cells stimulated with anti-CD2 mAbs (32). It is noteworthy that at variance from some reported anti-CD4 mAbs (33) none of the anti-CD4 mAbs used in our experiments could activate T cells, as evidenced by the lack of proliferative response and the absence of increase of intracellular Ca*+ (data not shown). Therefore, the modulation of surface CD4 reported here must be distinguished from that induced by phorbol esters( 13, 33), which requires phosphorylation of (a) serine residue(s) in the intracytoplasmic tail of CD4 (34, 35) and mobilization of intracellular Ca*+ (33). Internalized CD4 molecules were replaced on the cytoplasmic membrane by preformed CD4 molecules from an intracytoplasmic pool. This pool itself derives primarily from recycled surface CD4 (20,22) and we show here that blockade of the endosome/ lysozome pathway slightly enhances CD4 down-regulation. Conversely, the rate of de lzovoCD4 synthesisshould be very low in resting peripheral blood T cells since cultures with protein synthesis inhibitors in the absence of anti-CD4 mAb induced only a minor decreaseof surfaceCD4 density, not demonstrable before 72 hr (data not shown). Furthermore, cycloheximide did not significantly enhance CD4 modulation in the presenceof anti-CD4 mAb (Fig. 5), suggestingthat de IZOVO CD4 synthesis was already minimal in modulated cells. However, a slightly decreasedexpression of CD4 mRNA was observedin cells incubated with anti-CD4 mAb, asalready reported after activation with anti-CD3 plus PMA (36, 37). The down-regulation of surface CD4 expression
CD4 DOWN-REGULATION
BY ANTI-CD4 mAbs
297
was reported in cells transfected with the HIV- 1 nefgene (38) and in T cells incubated with dextran sulfate (39) or gangliosides (40,4 1). Similarly, in cells infected with HIV, a down-regulation of CD4 expression has been extensively documented, but different mechanisms could account for this alteration including decreasedsteady-state levels of CD4 transcripts, reduced levels of CD4 proteins and formation of CD4-gp120 complexes (42-44). The present data with anti-CD4 mAbs should also be compared with the alterations of surface CD4 protein expression induced by HIV envelope glycoproteins in noninfected cells. Weinhold et al. showed that gp 120 exhibited a dosedependent suppressive effect of T lymphocyte responsesto specific antigens or OKT3 antibodies but failed to demonstrate a down-regulation of surface CD4 expression in cells incubated for 12 hr with gp120 (45). Conversely, Juszczak and co-workers (46) recently reported that exposure of the T cell line CEM-CM3 to gp 120 (5 fig/ml) resulted in a down-regulation of surface CD4 protein expression comparable to that reported here with anti-CD4 mAbs. In their experiments the authors also described rapid dissociation of the CD4-~56”’ complex which was also observed in the presence of the Leu3a anti-CD4 mAb but not with another anti-CD4 mAb. One may postulate that common signals generated by CD4 ligands could induce the down-regulation of CD4 protein and the inhibition of T cell responses,but the precise relationship between these effects remains to be investigated. ACKNOWLEDGMENTS This work was supported by Institut National de la Sante et de la recherche medicale and Centre National de la Recherche Scientitique and by a grant from the Department of Human Biology, University Lyon I. Patricia Morel was recipient of a fellowship from the Fondation Marcel Merieux. We thank G. Panaye and G. Lizard (Service commun de cytofluorometrie) for expert assistancein flow cytometry and M. J. Gariazzo for her skillful1 assistance.
REFERENCES 1. 2. 3. 4. 5.
Reinherz, E. L., and Schlossman,F., Cell 19, 821, 1980. Terhost, C., Van Agthoven, A., Reinherz, E., and Schlossman,S., Science 209, 520, 1980. Biddinson, W., Rao, P., Talle, M., Goldstein, G., and Shaw, S., J. Immunol. 131, 152, 1983. Doyle, C., and Strominger, J. L., Nature 330,256, 1987. Dalgleish, A. G., Beverley, P. C. L., Clapham, P. R., Crawford, D. H., Greaves, M. F., and Weiss, R. A., Nature 312, 763, 1984. 6. Klatzmann, D., Champagne, E., Chamaret, S., Gruest, J., Guetard, D., Hercend, T., Gluckman, J. C., and Montagnier, L., Nature 312, 767, 1984. 7. Rudd, C., Trevillyan, J., Dasgupta, J., Wong, L., and Schlossman, S., Proc. Natl. Acad. Sci. USA 85, 5190, 1988. 8. Veillette, A., Bookman, M., Horak, E., and Bolen, J., Cell 55, 301, 1988. 9. Veillette, A., Bookman, M. A., Horak, E. M., Samelson, L. E., and Bolen, J. B., Nature 338, 257, 1989. 10. Veillette, A., Bolen, J. B., and Bookman, M. A., Mol. Cell Biol. 9, 4441, 1989. 11. Anderson, P., Blue, M. L., and Schlossman,S. F., J. Immunol. 140, 1732, 1988. 12. Weyand, C. M., Goronzy, J., and Fathman, C. G., J. Zmmunol. 138, 1351, 1987. 13. Hoxie, J. A., Matthews, D. M., Callahan, K. J., Cassel, D. L., and Cooper, R. A., J. Immunol. 137, 1194, 1986. 14. Carrel, S., Moretta, A., Pantaleo, G., Tambussi, G., Isler, P., Perussia, B., and Cerottini, J. C., Eur. J. Immunol. 18, 333, 1988.
15. Morel, P., Nicolas, J. F., Wijdenes, J., and Revillard, J. P., C/in. Immunol. Immunopathol. 64, 248, 1992. 16. Vincent, C., and Revillard, J. P., In “Methods of Enzymatic Analysis” (H. U. Bergmeyer, Ed), Vol. 9, pp. 248-265. VCH VerlagsgesellschafimbH, D-6940 Weinheim, Germany. 17. Thiele, D. L., Kurosaka, M., and Lipsky, P. E., J. Immunol. 131, 2282, 1983.
298
MOREL
ET AL.
18. Suggs,S. V., Wallace, R. B., Hirose, T., Kawashima, E. H., and Itakura, K., Proc. N&l. Acud. USA 78, 6613, 1981. 19. Caniere, D., Artier, J. M., Derocq, J. M., Fontaine, C., and Richer, G., Exp. Cell. Rex 182, 114, 1989. 20. Pelchen-Matthews, A., Armes, J. A., and Marsh, M., EMBO J. 8, 3641, 1989. 21. Marsh, M., Armes, J. E., and Pelchen-Matthews, A., Biochem. Sot. Trans. 18, 139, 1989. 22. Pelchen-Matthews, A., Armes, J. E., Griffiths, G., and Marsh, M., J. Exp. Med. 173, 575, 1991. 23. Boyse, E. A., Stocker& E., and Old, L. J., Proc. Natl. Acad. Sri. USA 58, 954, 1967. 24. Cole, J. A., McCarthy, A., Rees, M. A., Sharrow, S., and Singer, A., J. Immunol. 143, 397, 1989. 25. Rinnooy Kan, E. A., Platzer, E., Welte, K., and Wang, C. Y., J. Immunol. 133, 2979, 1984. 26. Schroff, R. W., Klein, R. A., Farrel, M. M., and Stevenson, H. C., J. Immunol. 133,2270, 1984. 27. Gallagher, P. F., De St. Growth, B. F., and Miller, J. F., Proc. Nat/. Acud. Sci. USA 86, 10044, 1989. 28. Chuck, R. S., Cantor, C. R., and Tse, D. B., Proc. Natl. Acad. USA 87, 5021, 1990. 29. Dianzani, U., Shaw, A., al-Ramadi, B. K., Kubo, R. T., and Janeway, C. A. J., J. Immunol. 148,678, 1992. 30. Rojo, J. M., Saizawa, K., and Janeway, C. A. J., Proc. Natl. Acud. Sci. USA 86, 3311, 1989. 31. Mittler, R. S., Goldman, S. J., Spitalny, G. L., and Burakoff, S. J., Proc. Nat!. Acad. Sci. USA 86, 853 1, 1989. 32. Bueso-Ramos,C. E., Donahoe, R. M., Nicholson, J. K. A., Madden, J. J., and Falek, A., J. Immunol. 140, 1414, 1988.
33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46.
Bigby, M., Wang, P., Fierro, J. F., and Sy, M. S., J. Immunoi. 144, 3 111, 1990. Shin, J., Doyle, C., Yang, Z., Kappes, D., and Strominger, J. L., EMBO J. 9, 425, 1990. Shin, J., Dunbrack, R. L., Lee, S., and Strominger, J. L., J. Biol. Chem. 266, 10658, 1991. Paillard, F., Sterkers, G., and Vaquero, C., EM30 J. 9, 1867, 1990. Neudorf, S., Jones, M., Parker, S., and Lattier, D., J. Immunol. 146, 2836, 1991. Garcia, J. V., and Miller, A. D., Nature 350, 508, 1991. Thiele, B., Braig, H. R., Ehm, I., Kunze, R., and Ruf, B., Eur. J. Immunol. 19, 1161, 1989. Offner, H., Thieme, T., and Vandenbark, A. A., J. Immunol. 139, 3295, 1987. Suthanthiran, M., J. Exp. Med. 171, 1965, 1990. Hoxie, J. A., Alpers, J. D., Rackowski, J. L., Huebner, K., Haggarty, B. S., Cedarbaum, A. J., and Reed, J. C., Science 234, 1123, 1986. Yuille, M. A., Hugunin, M., John, P., Sacks,L. V., Poiesz, B. J., Tomar, R. H., and Silverstone, A. E., J. Acquired Immune Defic. Syndrome 1, 13 1, 1988. Salmon, P., Olivier, R., Riviere, Y., Brisson, E., Gluckman, J. C., Kieny, M. P., Montagnier, L., and Klatzmann, D., J. Exp. Med. 168, 1953, 1988. Weinhold, K. J., Lyerly, H. K., Stanley, S. D., Austin, A. A., Matthews, T. J., and Bolognesi, D. P., J. Immunol. 142,3091, 1989. Juszczak, R. J., Turchin, H., Truneh, A., Culp, J., and Kassis, S., J. Biol. Chem. 266, 11176, 1991.
