CELLULAR

IMMUNOLOGY

Interleukin-4

129,329-340

Receptors

(1990)

on Human Blood Mononuclear

Cells

CAROLINE E. ZUBER*,’ JEAN-PIERREGALIZZI,* NOBUY~KI HARADA,~ ISABELLEDURAND,* AND JACQUES BANCHEREAU* *Schering-Plough (UNICET), Laboratory for Immunological Research, 69572 Dardilly, France; and TDNAX Institute of Molecular and Cellular Biology, Department oflmmunology, Palo Alto, California Received December 12, 1989; accepted April 8, 1990 We have studied regulation of the expression of the interleukin-4 receptor (IL-4R) on human blood mononuclear cells (PBMC) using both ‘251-IG4 binding assayand flow cytometric analysis of biotinylated IL-4 (B-B-4) binding. PBMC express = 300 high-affinity IL4R per cell (& = 25- 100 PM). Activation of PBMC for 60-80 hr by phytohemagglutinin (PHA) or concanavalin A (Con A) results in a 2- to 45fold increase of IL-4R number without alteration of IL4R affinity for IL-4. Binding of B-IL-4 showed that IL4R expression is upregulated on virtually all PHA-stimulated PBMC, whereas it mostly concerns larger cells among Con A-activated PBMC. Reculture of PHA-blasts with 1 nMIL-4 further upregulates IG4R expression to a level approximately IO-fold higher than observed on freshly isolated PBMC. Interestingly, IL-4 is able to reinduce high IL4R levels on cells that have been deprived of IL-4 for 20 hr and IL-2 is almost as efficient. Finally, SDS-PAGE analysis of IL-Cbinding molecules on unstimulated, PHA- and PHA/ILCactivated PBMC revealed the same three peptides of MW 140-l 30,80-75, and 7065 kDa. as shown on human cell lines. o 1990Academic press, Inc.

INTRODUCTION Human

interleukin-4 (IL-4)2 is a 18- to 19-kDa glycoprotein secreted by activated ( 1) which exhibits diverse biological activities depending both on the target cell lineage and on its stage of activation (2,3). In accordance with IL-4 pleiotropic activities, the specific receptor for IL-4 (IL4R) has been detected on human cell lines representative of various cell types (4,5). This receptor is characterized by a high affinity for its ligand (& = 20-100 PM) and a low level of expression on the various cell lines tested (~3000 IL4R per cell). Since the low spontaneous proliferative response of human peripheral blood lymphocytes to IL-4 is considerably increased upon lectin preactivation (6, 7), we have studied the regulation of IL-4R membrane expression on resting and in vitro activated peripheral blood mononuclear cells (PBMC). This investigation was performed using T lymphocytes

’ Caroline E. Zuber is supported by the Fondation Merieux (Lyon, France). The DNAX Research Institute of Molecular and Cellular Biology is supported by Schering-Plough Corporation. * Abbreviations used: B-IL4, biotinylated IL4, BS, bis(sulfosuccinimidy1) suberate; BSA, bovine serum albumin; Con A, concanavalin A; FctR2/CD23, low-affinity receptor for IgE on lymphocytes; IL-4, human interleukin-4; IL-4R, interleukin-4 receptor; MFI, mean fluorescence intensity; PBMC, peripheral blood mononuclear cells; PBS, phosphate-buffered saline; PHA, phytohemagglutinin; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis. 329 0008-8749/90$3.00 Copyright 8 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.

330

ZUBER ET AL.

both radiolabeled ‘251-IL-4 and biotinylated IL4 (B-K-4). It is shown that lectin activation increases the low number of high-affinity IG4R expressed on unstimulated PBMC by two to fourfold without affecting IG4R affinity and gross biochemical structure. The modulation of IG4R expression was further studied following reculture of PHA-blasts with and without IL-4 or IL-2. MATERIALS

AND METHODS

Cytokines and Reagents Highly purified human recombinant IL4 from transfected CHO cell supernatants was a kind gift of Dr. P. Trotta and Dr. T. Nagabhushan (Schering Plough Research, Bloomfield, NJ). Purified recombinant IL-2 (3 X lo6 U/mg) was from Amgen (Thousand Oaks, CA). Purified phytohemagglutinin (PHA) was from Wellcome, Concanavalin A (Con A) from Pharmacia. NHS-LC biotin and Bis (sulfosuccinimidyl) suberate (BS) were from Pierce Chemical Co. (Rockford, IL). Phycoerythrin-labeled streptavidin was from Becton-Dickinson (Mountain View, CA).

Preparation and Culture of PBMC PBMC were isolated from healthy donors by centrifugation on standard Ficoll/ Hypaque density gradient. Cells were cultured at 1 X lo6 cells/ml in complete culture medium which consisted of RPM1 1640 supplemented with 10% heat-inactivated fetal calf serum, 2 mM glutamine, 100 U/ml penicillin, 100 pg/ml streptomycin (all from Flow Laboratories, Irvine, Scotland), 50 pg/ml gentamicin (Laboratoires UNICET, Levallois-Perret, France), and 20 mM Hepes. At the end of the culture, cells were washed twice (200 g, 10 min) before being tested in binding experiments.

12jI-IL-4 Equilibrium Binding Experiments Equilibrium binding of ‘25I-IG4 was carried out as described in detail elsewhere (4). Highly purified IL-4 was iodinated according to a modified chloramine T method (8) to a specific radioactivity of 2000 Ci/mmol without significant loss of biological activity. Cell aliquots (2-3 X 106) were incubated for 4 hr at 4°C in the presence of various ‘251-IL-4 concentrations in 0.5 ml of RPM1 1640 medium supplemented with 2% BSA and 20 mMHepes, pH 7.2 (binding medium). Then, free ‘251-IL-4 was separated from surface-bound ‘251-IL-4 by centrifugation (200 g, 10 min at 4°C). Nonspecific binding was determined in the presence of 10 nM unlabeled IL-4. In the case of single point-binding studies, the 1251-IL4 concentration used was 150 PM. PBMC cultured without any activator were found to express a stable number of IL-4R during the first O-80 hr of culture. Thus we used binding values on noncultured cells (t = 0 hr) as a control for expressing binding data.

Preparation and Characterization of Biotinylated IL-4 IL-4 (final concentration: 12 nM) was incubated for 2 hr at room temperature with NHS-LC biotin ( 1: 100 molar ratio). Reaction was stopped by addition of Tris-HCl, pH 8.0 (final concentration: 83 mM). The biotinylated material was separated from free biotin by dialysis at 4°C against PBS. The final protein content (12 &f) was determined using a Bradford assay with BSA as a standard. SDS-PAGE analysis of

IL4R

EXPRESSION

ON HUMAN

LYMPHOID

CELLS

331

B-IL-4

FIG. 1. Specific B-IL-4 binding as a function of B-IL-4 concentration. MLA 144 cells were incubated for 3 hr at 4°C in the presence of various B-IL-4 concentrations, washed, and exposed for 20 min to streptavidinphycoerythrin conjugate. Alter washes, cell-bound fluorescence was analyzed on a FACScan using a linear scale. The specific binding is expressed as AMFI value, as described under Materials and Methods. This experiment is representative of three.

this material on a 15% polyacrylamide followed by Coomassie blue staining demonstrated the presence of a single band of MW 20 kDa slightly higher than unlabeled IL4 (MW 18 kDa). Moreover, standard streptavidin-peroxidase staining of this labeled material after transfer on nitrocellulose revealed one single band, thus demonstrating homogeneous biotinylation of IL4 (data not shown). Pilot experiments demonstrated that B-IL-4 inhibits ‘25I-IG4 binding to cells and is biologically active as demonstrated by induction of FceRdCD23 on Jijoye cells (9). Thus, IL-4 was efficiently biotinylated without significant degradation.

B-IL-4 Binding: Equilibrium Conditions and Flow Cytometric Analysis Equilibrium binding of B-IL4 was performed at 4°C as followed: cell aliquots ( 106) were incubated for 3 hr in the presence of a saturating concentration of B-IL4 (5 nM) in 0.1 ml of PBS supplemented with 0.1% BSA and 0.01% sodium azide. Cells were then incubated with a streptavidin-phycoerythrin conjugate and fluorescence was analyzed with a FACScan (Becton-Dickinson). Fluorescence parameters were collected using a linear scale after gating on the combination of forward light scatter and perpendicular light scatter, which was used to discriminate viable from nonviable cells. B-IL-4 binding was expressed as fluorescence profiles (10,000 cells analyzed per histogram) and quantified as mean fluorescence intensity (MFI). Specific B-IL-4 binding values were obtained after subtraction of the nonspecific signal obtained in the presence of a 1OO-fold excess of unlabeled IL-4 (AMFI). Pilot experiments were performed to test B-IL-4 ability to bind in a specific and saturable manner to cells expressing IL4R. Therefore, various B-IL-4 concentrations were allowed to bind in the presence or absence of a large excess of unlabeled IL-4 to cells of the gibbon T cell line MLA 144 (which express 4000 ‘251-IL-4 binding sites per cell). Figure 1 illustrates the results of such an experiment: the AMFI value which represents B-IL-4 specific binding was a saturable function of B-IL-4 concentration.

332

ZUBER ET AL.

0

2

4

6

125 l-IL-4 specifically

6

bound

10

12

@M)

FIG. 2. Scatchard plots of ‘251-IL4 binding to PHA- or Con A-stimulated PBMC. Unstimulated PBMC (O), or PBMC activated for 85 hr with either 0.5 &ml PHA (m) or 2 &ml Con A (A) were allowed to bind to various “‘I-IL-4 concentrations, as described under Materials and Methods.

B-IL-4 (5 r&f) was chosen for further studies because this concentration saturation with a low nonspecific signal.

allows 80%

Removal of &&ace-Bound IL-4 from Cells Cultured with IL-4 As described earlier for the Burkitt lymphoma cell line, Jijoye (lo), two methods were performed in parallel to wash out membrane-bound IL-4: (i) cells were recultured for 2 hr in fresh prewarmed culture medium; (ii) cell aliquots (1 ml) were exposed to 700 ~1 of glycine-HCl 0.1 M, NaCl 140 mM, pH 2.9, for 7 min. This short glycine-HCl treatment removes 80% of IL-4 bound to PHA-blasts and does not affect further binding of B-IL-4 to these cells (data not shown).

Covalent Crosslinking and Gel Electrophoresis These experiments

were performed exactly as described earlier ( 10). Briefly, 3 at 4°C with 250 pM i2’I-IL-4, washed and 50 H of crosslinking agent BS was added to the incubation medium. After 30 min, reaction was stopped with Tris-HCl. Cells were lysed in 1% Triton X- 100 in the presence of a cocktail of protease inhibitors (2 mMphenylmethylsulfony1 fluoride, 1 mMiodoacetamide, 1 mM 1, IO-phenanthroline, 1 $W pepstatin A). Lysates were prepared for sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) according to standard protocol and analyzed on a 5- 10% polyacrylamide gradient gel. Finally, the dried gel was exposed to XAR-5 film (Kodak, Paris, France). X lo7 cells in 3 ml were incubated

RESULTS

Lectin Activation of Human PBMC Increases the Number of IL-4R Normal PBMC express an average of 250 12’I-IL-4 binding sites per cell with Kd = 25- 100 pM (Fig. 2). PBMC ( lo6 cells/ml) were cultured for different periods of

IL-4R EXPRESSION

ON HUMAN

LYMPHOID

333

CELLS

TABLE 1 ‘?-IL-4

Binding on Lectin-Activated PBMC Bound “‘I-IL-4

Expt I

PHA Con A

(%)

Expt 2

36 hr

60 hr

115hr

36 hr

60 hr

N.D. 96 + 18

388227 230+ 3

420a13 226k 15

226+22 254 f 31

394+ 13 355 + 16

Expt 3 80 hr

60 hr

358+27 N.D.

438+21 287k22

80 hr 434 + 32 258k 18

Note. Freshly isolated PMBC were cultured with either PHA (0.5 pg/ml) or Con A (2 &ml). At the different times indicated, cells were washed with binding medium and incubated for 4 hr at 4°C with 150 pM ‘251-IL-4 as described under Materials and Methods. Results are expressed as mean percentages + SD of the average binding value obtained at time 0 of three determinations. N.D., not determined.

time (up to 115 hr) in the presence of 0.5 pg/ml PHA or 2 pg/ml Con A. The binding of 150 pA4 1251-IL-4 to these cells was examined. Results illustrated in Table 1 show that cells cultured for 60-80 hr with PI-IA or Con A bound, respectively, 2.6-4.4 fold and 2 . 3-3 . 5-fold more 12’I-IL-4 than unstimulated PBMC. Similar results were obtained for PHA concentrations within a range of 0.1-1.0 &ml and for Con A concentrations within a range of 2-6 pg/ml (data not shown). Figure 2 shows that the affinity for ‘25I-IG4 is the same on freshly isolated PBMC and PBMC cultured for 85 hr with PHA or Con A (& = 60 PM, 80 PM, and 65 PM, respectively) whereas the average number of IL-4R increases from 260 sites per unstimulated cell to 1150 and 630 sites per cell after PHA and Con A stimulation, respectively. Experiments performed with PBMC of three donors provided averages of 305 -t 50, 650 f 35, 1100 + 40 high-affinity ILCR per unstimulated, Con Aactivated and PHA-stimulated PBMC, respectively. Moreover, similar IL4R levels were measured after 2 hr reculture of these activated cells in medium alone, ruling out the hypothesis that some IL-4 produced by the cells during the culture was partially inhibiting 1251-IL-4 binding (not shown). Thus, in vitro lectin activation of normal PBMC results in a 2-4.5-fold increase of IL-4R number without alteration of IL4R affinity for its ligand.

Flow Cytometric Analysis of IL-4R Expression on Activated PBMC To investigate the distribution of IL-4R expression on lectin-activated PBMC, we used flow cytometry analysis of B-IL-4 which enables us to examine IG4R distribution on a cell per cell basis. Unstimulated, PHA- and Con A-activated (85 hr) PBMC were incubated with 5 nM B-IL-4, with and without an excess of unlabeled IL-4. After washes and exposure to a streptavidin-phycoerythrin conjugate, cell-bound fluorescence was analyzed with a FACScan. Figure 3 illustrates one typical B-IL-4 binding experiment out of three. It shows that the specific B-IL-4 binding (AMFI) increased from 1.57 to 17.1 and 9.6 afier 85 hr activation with PHA and Con A, respectively. The increased IL4R expression on PHA-activated PBMC affected virtually all the cells. In contrast, only a proportion (approximately 50%) of Con Astimulated PBMC expressed increased IL4R levels. Light scattering analysis (forward versus perpendicular light scatter) of Con A-activated PBMC showed the pres-

334

ZUBER ET AL. UNSTIMULATED AMFI

= 1.57

fluorescence

intensity

b 2 2 = 8 al .s m 2

5

PHA, 85 h AMFI

Con A, 85 h AMFI

= 17.1

= 9.6

$ 2 2 = 83 al .g s e!

fluorescence

intensity

SMALLER

CELLS

LARGER AMFI

fluorescence

CELLS = 19.7

intensity

FIG. 3. Flow cytometric analysis of B-IL-4 binding on unstimulated, PHA- and Con A (85 hr)- activated PBMC. Cells ( 106) were incubated for 3 hr at 4’C in the presence of 5 nM B-IL-4, washed, and exposed to streptavidin-phycoerythrin conjugate. Cell-bound fluorescence was then analyzed on a FACScan using a linear scale. The histograms show relative cell number in function of fluorescence intensity (both in arbitrary units). The histograms of B-IL-4 binding obtained in the presence of unlabeled IL-4 (nonspecific binding) and in its absence (total binding) are overlapped to yield specific binding. AMFI values correspond to the difference between the MFI values measured in these two conditions. Using light scattering analysis, Con A-activated cells have been divided into two subpopulations, corresponding to either the smaller cells (60%) or the larger cells. B-IL-4 binding to each of these subpopulations is shown. This experiment is representative of three.

IL4R

EXPRESSION

ON HUMAN

LYMPHOID

335

CELLS

TABLE 2 IL-4 Upregulates IL-4R Expression on PHA-blasts B-IL-4 binding (AMIT

‘251-IL-4 binding sites (number/cell) IL-4 Concentration (PM) 0 1 10 100 1000

20 hr

40 hr

60 hr

20 hr

40 hr

60 hr

1050 1370 2510 4050 3575

585 590 590 2600 3620

N.D. N.D. N.D. 1100 2990

16.5 18.5 37.4 62.0 54.0

7.9 9.5 10.9 37.0 55.0

N.D. N.D. N.D. 14.0 43.0

Note. PBMC were cultured for 3 days with 0.5 &ml PHA, washed, and recultured for the indicated times with various IL-4 concentrations as shown above. At the end of the culture, cells were washed extensively and incubated for 2 hr at 37°C in fresh culture medium without IL-4 to remove membrane-bound IL-4. Then, cells were incubated either for 4 hr with various 12%IL-4 concentrations or for 3 hr with 5 nM B-IL-4 as described under Materials and Methods. ‘251-IL-4 binding data are expressed as number of 12>1IL-4 binding sites per cell as determined by Scatchard analysis. B-IL-4 specific binding is expressed as AMPI value (arbitrary unit), which corresponds to the difference in the MFI obtained in the presence versus in the absence of unlabeled IL-4 (nonspecific and total binding, respectively). This experiment is representative of two.

ence of a smaller cell subpopulation (60% of total cells) and a larger cell subpopulation. The smaller cells bound little B-IL-4 (AMFI = 6.5), whereas the larger cells bound as much B-IL-4 (AMFI = 19.7) as PHA-activated cells.

IL-4 Regulates IL-4R Expression on PHA Blasts Our previous studies with Jijoye cells showed that IL-4 induces a decrease of IL4R expression after 2 hr but the IL-4R were reexpressed to control level after 20 hr of culture in spite of the presence of a large excess of IL-4 (10). Consequently, we investigated whether IL-4 would modulate IL-4R expression on PHA-blasts. PHAblasts were washed and recultured for 20,40, or 60 hr with various IL4 concentrations (1 to 1000 PM). In the first assay, membrane-bound IL-4 was dissociated by reculture for 2 hr at 37°C in prewarmed fresh medium. IL-4R expression was deterbinding and flow cytometric analysis mined using both Scatchard analysis of ?-IL-4 of B-IL-4 binding. As illustrated in Table 2, IL-4 induces a dose-dependent increase in both 1251-IL-4 binding sites and AMFI values of B-IGCspecific binding. Maximal IL-4R upregulation was obtained after 20-40 hr culture with lOO- 1000 pM of K-4. PHA-blasts cultured with 10 pM IL-4 expressed increased IL-4R levels at 20 hr but not later, whereas blasts cultured with 1 pM IL-4 did not express increased IG4R levels when compared to cells cultured without IL-4. Both ‘251-IL-4 binding sites and AMFI values simultaneously decreased after more than 40 hr culture with 100 pM of IL-4. These data show that AMFI values vary in parallel to the number of ‘251-IL4 binding sites, indicating that AMFI is a valid parameter to measure modulations of IL-4R number. Under these conditions the affinity of IG4R for 125I-IL-4 remained unaltered (data not shown). The present 2-hr reculture method to remove cold bound IL-4 has the advantage of being both gentle and highly efficient, but it does not allow direct measurement of

ZUBER ET AL.

336

TABLE 3 The IGCMediated

IL4R Upregulation Is Transient

B-IL4 binding (AMPI) PHA 65 hr/IL-4 40 hr PHA 65 hr Experiment 1 Experiment 2

15.0 23.6

Glvcine-HCl 31.6 38

2-hr reculture

PHA 65 hr/IG4 40 hr/MED 20 hr

55.5 53

13 21.3

Note. PBMC were first cultured for 65 hr with 0.5 &ml PI-IA, then for 40 hr with 1 nMIL-4 and finally for 20 hr with medium alone. At the end of each culture, an aliquot of cells was used to test cell ability to bind 5 n&f IL-4. In the case of cells pulsed with IL-4, B-IL-4 binding was preceded by a glycine-HCl treatment or by a 2-hr reculture in medium to remove bound IL-4. B-IL+specific binding is expressed in AMPI value (arbitrary unit). These experiments are representative of four.

the levels of IL4R at the end of IL4 culture. Thus, we investigated IL4R expression on PHA-blasts cultured for 40 hr with 1 riMof IL-4 directly after removing unlabeled IL-4 by a 7-min exposure to glycine-HCl. Results were compared to those obtained after a 2-hr reculture in IL-Cfree medium as described earlier and are shown in Table 3. PHA-blasts cultured with IL-4 and then treated with glycine-HCl express 1.6-fold more IG4R than unstimulated cells, whereas the same IL-4-pulsed cells recultured for 2 hr in fresh medium express a 2.3-fold increased IG4R level. This further enhancement of IL4R level after reculture in IGCfree medium can be observed for reculture periods of 2 to 7 hr (data not shown) but cannot be detected after a 20hr reculture in medium alone (Table 3). Thus, these data indicate that IL-4 slightly increases IG4R membrane expression on PHA-blasts when it is present during the whole culture period. Reculture without IL-4 for short periods of time results in a transient amplification of this IL-4-induced IG4R upregulation. IL-4 and IL-2 Can Reinduce IL-4R on PHA-Blasts Deprived of IL-4 Since removing of IL-4 from cultures of PHA-blasts finally results in a strong decrease of IG4R expression, we investigated whether readdition of IL-4 or IL2 could reinduce high levels of IG4R expression on these cells. Sixty-five-hour PHA-blasts cultured for 40 hr with 1 nM IL-4 were harvested, washed extensively, and recultured for 20 hr without any cytokine. Then cells were recultured for 40 hr with medium alone, 20 U/ml IL-2, 1 nM IL4 or a combination of both cytokines. Results illustrated in Table 4 show that addition of IL-4 leads to the reexpression of IL4R at levels comparable to those obtained following the first IL-4 culture (Table 3), both after glycine-HCl treatment and after 2 hr reculture without IL-4. Interestingly, the level of IL-4R observed at the end of IL-2 culture (20 U/ml) is similar to the level measured at the end of IL-4 culture, after glycine treatment. Similar results were obtained with 40 U/ml IL2 and the IG4R response was dose-dependent for IL-2 concentrations below 20 U/ml (data not shown). Nevertheless, cells cultured with IL4 expressed twofold higher level of IG4R after 2 hr reculture in medium alone than at the end of IL-2 or IL-4 (glycine-HCl treatment) cultures. Also, combinations of

IL-4R EXPRESSION

ON HUMAN

LYMPHOID

337

CELLS

TABLE 4 Reinduction by IL-4 or IL-2 of High IL-4R Level on PHA-blasts Deprived of IL4 B-IL-4 binding (AMFI) IL-4

Experiment 1 Experiment 2

Medium

IL-2

Glycine-HCl

7.5 12.0

34.4 36.0

45.6 37.0

IL-4 + IL-2 2-hr reculture 82.3 61.9

Glycine-HCl 46.2 42.0

2-hr reculture 78.8 58.8

Note. 65hr PHA-blasts were cultured for 40 hr with 1 nM IL4, washed, and cultured for 20 hr with medium as described in Table 3. Then, cells were washed and recultured for 40 hr with either medium, IL2 (20 U/ml), IL-4 (1 nM), or a combination of IL-2 and IL-4 as described above. After harvest and washes, cells were allowed to bind to 5 nkf B-K-4. In the case of IL-4 activation, cells were either treated with glycine-HCl or recultured for 2 hr without IL-4 prior to binding. B-IL-4 binding is quantified as AMFI value (arbitrary unit). Representative of three.

IL-2 and IL-4 were not more efficient in upregulating IL4R expression than IL-4 alone, both after glycine-HCl treatment and after 2 hr reculture.

SDS-PAGE Analysis of the IL-4R-12’I-IL-4

Crosslinked Complex

The gross biochemical structure of IL4R on PBMC was studied in comparison to that of MLA 144 cells by the crosslinking of 12’I-IL-4 and analysis of cell lysates by SDS-PAGE and autoradiography. The IG4R levels on unstimulated cells were too low for the detection of specifically labeled species (Fig. 4, lane 3). PHA-blasts (Fig.

Mr x

1o-3

123456

FIG. 4. SDS-PAGE analysis of IL-4R-‘251-IL4 complexes on unstimulated and activated PBMC. ‘*‘IIL-4 bindingjcrosslinking to MLA 144 cells (lane 2), unstimulated PBMC (lane 3), PHA-blasts (lane 4), PHA-blasts recultured with IL-4 (lane 5) were carried out, as described under Materials and Methods. 1% Triton X- 100 cell lysates were analyzed by SDS-PAGE on a 5- 10% polyacrylamide gel gradient followed by autoradiography. Nonspecific ‘*‘I-IL-4 binding/crosslinking was determined with a sample ofMLA 144 cells (lane I) and a sample of PHA-blasts recultured with IL-4 (lane 6) in the presence of 10 nkfunlabeled IL-4. This experiment is representative of two.

338

ZUBER ET AL.

4, lane 4) and PHA-blasts recultured for 40 hr with 1 nM IL-4 (Fig. 4, lane 5) displayed three labeled species of 140, 80-75, 70-65 kDa after subtraction of the IL-4 MW (18 kDa). Formation of these labeled complexes was prevented by inclusion of a loo-fold molar excess of unlabeled IL-4 (Fig. 4, lane 6). The high MW species of PHA-blasts is slightly larger than the 130~kDa high MW component of MLA 144 cells (Fig. 4, lane 2). The doublet at 80-70 and 70-65 kDa of PBMC comigrates with that observed in MLA 144 cells. Thus, the SDS-PAGE pattern of the IL-4R-‘*%IL4 complex on PHA-blasts and PHA-blasts recultured with IL-4 is almost identical. Moreover, it is similar to the pattern obtained with MLA 144 cells. DISCUSSION Once activated with PHA or Con A, PBMC strongly proliferate in response to IL4 whereas unstimulated cells are almost unreactive (6, 7). In the present study, we have investigated whether this difference might be attributable to differences of IL 4R expression and/or characteristics. Unstimulated PBMC express low levels of highaffinity IG4R (250 binding sites, Kd = 25-100 PM). Upon 85 hr activation with PHA or Con A, cells express 2- to 4.5-fold higher numbers of IL4R, whereas IL-4R affinity for its ligand is unmodified. In the majority of the experiments, PHA appears more potent than Con A when the whole activated populations are considered (Scatchard analysis of ‘251-IL-4 binding). This may be explained by the fact that PHA appears to stimulate virtually all cells whereas Con A activates only a fraction (40%) of the cells as illustrated by light-scattering properties of activated populations. Flow cytometry analysis of B-IL-4 binding demonstrates that PHA induces an increase of IL4R expression on virtually all cells, the largest cells staining the brightest. In contrast, Con A upregulates IL-4R expression on approximately 40% of the cells with light-scattering properties corresponding to a subpopulation of large size. Reculture of PHA-blasts with optimal IL-4 concentrations ( 1OO- 1000 PM) induces a 1.6-fold increase in IL 4R level. Such an IL-4 induced increase of IG4R on T cells was observed with murine BALB/c lymph nodes T cells (11) as well as long-term T cell lines (12). It was also observed with normal human tonsillar B cells (13). Furthermore, removal of IL-4 from these cultures for 2 hr results in a further enhancement of IL4R expression. As observed earlier with the CD23 inducible B cell line Jijoye (lo), and normal B cells ( 13) this further increase is transient, since after 20 h without IL-4, cells express even less IL-4R than PHA-blasts, a phenomenon coincidental with a decrease in viability. This transient upregulation of IL4R is likely to be due to the arrest of IL-4-IL-4R complex internalization after removal of IL-4 while IG4R proteins are still inserted at a high rate into the membrane, probably as a consequence of an IGCmediated increase of IL4R turnover (10). The low expression of IL4R on cells which have been weaned of IL-4 for 20 hr is reversible as readdition of IL-4 induces a new increase of IL-4R expression on cells. Interestingly, IL-2 is also able to reinduce high IL4R levels on these cells. Nevertheless, IL-4 seemed to be a more potent activator of IL4R turnover since the IL4R level was twofold higher 2 hr after removal of IL-4. The upregulating effects of IL-4 and IL-2 on IL-4R expression are not additive. The IGZinduced upregulation of IL4R on human T cells contrasts with results obtained with mouse long-term T cell lines where IL-2 does not upregulate IL4R expression ( 12). This discrepancy might be due to species differences or to differences in the experimental protocols and

IL4R

EXPRESSION

ON HUMAN

LYMPHOID

CELLS

339

clearly deserves further investigations (short-term cultured PHA blasts versus longterm T cell lines). When a cDNA coding for human IG4R protein is available, it will be of interest to compare IL2 and IL-4 ability to stimulate transcription of the IL4R gene as it has been demonstrated earlier with IL-2 and ~55 IL-2R (14, 15). Crosslinking studies performed on PHA-blasts demonstrate that IL-4 specifically binds to three species of MW 140, 80-75, and 70-65 kDa, as shown recently by Foxwell et al. ( 16). The same three species were detected on normal human B cells, various T and B cell lines as well as on the gibbon T cell line MLA 144 from which we purified a high molecular weight component that by itself binds IL-4 with high affinity (17). On PHA-blasts the high molecular weight species is slightly higher than that observed on the other cells tested (140 kDa versus 130 kDa). This difference might be due to heterogeneity in glycosylation but this remains to be determined. The 140- to 130-kDa component is likely to be the human counterpart of the murine 130-kDa IL-4 binding protein for which a cDNA was recently isolated ( 18, 19). Although the nature of the other two components remains to be definitely established, our preliminary results suggest that the 70- to 65-kDa component may represent a degradation product of the 140- to 130-kDa component, as removal of protease inhibitors results in a decrease of the 140- to 130-kDa component and a concomitant increase of the 70- to 65-kDa band (not shown). Alternatively, the 70- to 65-kDa component may be the product of an alternative splicing of the 140- to 130~kDa polypeptide cDNA, as shorter mRNA species have been found in murine cells (18) and as the 70-65 component is always detected in spite of the presence of a complex cocktail of protease inhibitors. The 80- to 75-kDa component may represent another polypeptide of the IL4R which is preferentially crosslinked to 1251-IL-4 with the crosslinker DST (Cabrillat, H., Galizzi, J. P., Banchereau, J., unpublished data). Thus, like the IL-2 R (20), the IL-6R (2 1,22) and the IFN-7 R (23), the IL-4R would be composed of at least two polypeptides. The molecular characterization of these different polypeptides is presently underway. Taken together, the present results indicate that peripheral blood mononuclear cells express low numbers of high-affinity IL-4 receptors which increase atter polyclonal activation induced by PHA or Con A. IL-4 is able to further increase this IL4R number by approximately twofold. Finally, PHA-blasts display three IL-Cbinding species of 140,80-75, and 70-65 kDa as described earlier with cell lines. ACKNOWLEDGMENTS We are grateful to J. P. Aubry for his help in flow cytometry and to Mrs. N. Courbiere and M. Vatan for excellent editorial assistance. We also thank Dr. A. Waitz for careful review of the manuscript and Dr. D. Blanchard for helpful discussions.

REFERENCES 1. Yokota, T., Otsuka, T., Mosmann, T., Banchereau, J., Defiance, T., Blanchard, D., de Vries, J. E., Lee F., and Arai K., Proc. Natl. Acad. Sci. USA 83,5894, 1986. 2. Paul, W. E., and Ohara, J., Annu. Rev. Immunol. 5,429,1987. 3. Yokota, T., Arai, N., de Vries, J. E., Spits, H., Banchereau, J., Zlotnick, A., Rennick, D., Howard, M., Takebe, Y., Miyatake, S., Lee, F., and Arai, K., Immunol. Rev. 102, 137, 1988. 4. Cabrillat, H., Galizzi, J. P., Djossou, O., Arai, N., Yokota, T., Arai, K., and Banchereau, J., Biochem. Biophys. Rex Commun. 149,995, 1987. 5. Park, L. S., Friend, D., Sassenfeld, H. M., and Urdal, D. L., J. Exp. Med. 166,476, 1987.

340

ZUBER ET AL.

6. Spits, H., Yssel, Y., Arai, N., Yokota, T., Lee, F., Arai, K., Banchereau, J., and de Vries, J. E., J. Immunol. 139,1142,1987. 7. Mitchell, L. C., Davis, L. S., and Lipsky, P. E., J. Zmmunol. 142, 1548, 1989. 8. Tejedor, F., and Ballesta, J. P. G., Anal. Biochem. 127, 143, 1982. 9. Rousset, F., de Waal Malefijt, R., Slierendregt, B., Aubry, J. P., Bonnefoy, J. Y., Defiance, T., Arai, K., Banchereau, J., and de Vries, J. E., J. Immunol. 140,2625, 1988. 10. Galizzi, J. P., Zuber, C. E., Cabrillat, H., Djossou, O., and Banchereau, J., J. Biol. Chem. 264,6984, 1989. 11. Ohara, J., and Paul, W. E., Proc. Natl. Acad. Sci. USA 85,8221, 1988. 12. Jankovic, D. L., Gibert, M., Baran, D., Ohara, J., Paul, W. E., and Theze, J., J. Zmmunol. 142,3 I 13, 1989. 13. Zuber C. E., Galizzi, J. P., Valle, A., Harada, N., Howard, M., and Banchereau, J., Eur. J. Zmmunol. 20,551, 1990. 14. Smith, K., and Cantrell, D., Proc. Natl. Acad. Sci. USA 82,864, 1985. 15. Malek, T., and Ashwell, J. D., J. Exp. Med. 161, 1126, 1985. 16. Foxwell, B. M. J., Woerly, G., and Ryffel, B., Eur. J. Immunol. 19, 1637, 1989. 17. Galizzi, J. P., Castle, B., Djossou, O., Harada, N., Cabrillat, H., Ait Yahia, S., Barrett, R., Howard, H., and Banchereau, J., J. Biol. Chem. 265,439, 1990. 18. Mosley, B., Beckmann, M. P., March, C. J., Idzerda, R. L., Gimpel, S. D., VandenBos, T., Friend, D., Alpert, A., Anderson, D., Jackson, J., Wignall, J. M., Smith, C., Gallis, B., Sims, J. E., U&l, D., Widmer, M. B., Cosman, D., and Park, L. S., CelZ59,335, 1989. 19. Harada, N., Castle, B. E., Gorman, D. M., Itoh, N., Schreurs J., Barrett, R. L., Howard, M., and Miyajima, A,, Proc. Natl. Acad. Sci. USA 87,857, 1990. 20. Smith, K. A., Immunol. Today9,36. 21. Yamasaki, K., Taga, T., Hirata, Y., Yawata, H., Kawanishi, Y., Seed, B., Taniguchi, T., Hirano, T., and Kishimoto, T. Science 241,825, 1988. 22. Taga, T., Hibi, M., Hirata, Y., Yamasaki, K., Yasukawa, K., Matsuda, T., Hirano, T., and Kishimoto, T., CeN58,573, 1989. 23. Aguet, M., Dembie, Z., and Merlin, G., Cell 55,273, 1988.