Eur. J. Irnrnunol. 1990. 20: 999-1005
Christine Couturier, Nicole Haeffner-Cavaillon, Laurence Weiss, Elisabeth Fischer and Michel D. Kazatchkine
INSERM U 28, H6pital Broussais, Paris
Induction o f IL 1 by stimulation of LFA-1 and CR.7
99Y
Induction of cell-associated interleukin 1 through stimulation of the adhesion-promoting proteins LFA-1 (CDlldCD18) and CR3 (CDllbKD18) of human monocytes* Serum-free culture of human monocytes in the presence of monoclonal antibodies to the LFA-1 a chain (CD lla) , CR3 u chain (CD llh) or p chain (CD18) bound to Sepharose induced the dose-dependent production of cellassociated interleukin (IL) I activity and of IL la and IL 1p antigens, but no release of extracellular IL 1 activity or antigen in the culture medium. Triggering of IL 1 production was also observed with insolubiiized anti-CDlI/CD18 F(ab’)? antibodies. Two cross-linked antibodies recognizing distinct epitopes on the C D l l b molecule induced cell-associated IL 1. Soluble antibodies did not induce IL 1 production. The kinetics of induction of IL 1 by stimulation of adhesionpromoting proteins differed from those of IL 1 induction by adhesion to plastic. The lack of induction of IL 1 release by stimulation of the CDlI/CDl8 molecules resembled the intracellular accumulation of IL 1 induced by lipid A. Induction of IL 1 by adhesive processes may be a mechanism by which Tcells trigger 1L I production by monocytes during antigen presentation.
1 Introduction The CDIl/CD18 complex comprises three molecules of the integrin supergene family which mediate cell adhesion [l-31. The three members of the CD1 K D 1 8 family are heterodimers with distinct u chains, C D l l a , C D l l b and CDllc, and a non-covalently bound identical 95-kDa f3 chain, CD18 [4]. Human peripheral blood monocytes express CDlla/CD18 (LFA-l), CDllWCD18 (CR3) and, to a lower extent CDllc/CD18 (p150,95) (2.51. Adhesion processes mediated by t h e CDlUCD 18 complex are cationand temperature dependent and may involve an ArgGly-Asp (RDG) sequence in the ligands. CD 1la/CD18 plays an important role in homotypic adhesion of IFN:/-stimulated monocytes [6] and in monocyte binding to endothelial cells 171. A major ligand for CDllalCDl8 is the intercellular adhesion molecule, ICAM-1 [8].CDll b/CD IS mediates adhesion reactions of monocytes to leukocytes, endothelial cells, target particles of phagocytosis and targets of antibody-dependent cellular cytotoxicity [7. 9. lo]. C D l l b also functions as a receptor for the iC3b cleavage fragment of C3 when presented as a multivalent ligand covalently bound to opsonized surfaces 111, 121. Evidence from blocking experiments using mAb indicates that C D l l b exhibits at least two functional binding sites, one for iC3b and other RDG-containing proteins, and another for carbohydrates and LPS [13. 141. A third functional site involved in the enhancement of granulocyte adhesion-dependent functions by CR3 has also been pos-
[I 76901
* This study was supported by Institut National de la Sante et de la Recherche MCdicale (INSERM) and by Association pour la Recherche contre le Cancer (ARC 67-66).
Correspondence: Nicolc Haeffner-Cavaillon. Unite d‘lrnrnunopathologie. INSERM U 28, HBpital Broussais. 96. rue Didot. F-75014 Paris, France 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
tulated [15]. CR3 mediates phagocytosis of iC3b-coated particles by activated human monocytes [16, 171. The molecule has been implicated in triggering enhanced expression of class 11antigens and macrophage activation in mice [ 181. The present study indicates that triggering of specific epitopes o n C D l l a , C D l l b and CD18 induces the production of cell-associated IL 1 by serum-free cultured human monocytes. Induction of IL 1by adhesive processes may he of relevance for Tcell activation during antigen pre5entation.
2 Materials and methods 2.1 Materials Polymyxin B and indomethacin were purchased from Sigma (St. Louis, MO). LPS from Neisseria meningitidis (NmLPS), extracted according to Westphal’s procedure [19], was a gift from Dr. D. Schulz, Institut MCrieux, Lyon, France. 2.2 Antibodies The following mouse mAb were obtained as indicated in parentheses: anti-cr chain of LFA-1 (CD lla) IgGl antibody 25.3.1 ([30];a gift from Dr. D. Olive, Marseille, France), anti-a chain of CR3 ( CD llb) IgGl antibody MN41 ([21]; a gift from Dr. G. D. Ross, Chapel Hill. NC) and IgGl antibody 60.1 ( [ 2 ] :a gift from Dr. P. G. Beatty, Seattle, WA). anti-common (3 chain (CD18) IgGl antibody BL-5 ([22]; a gift from Dr. J. Brochier, Montpellier, France) and IgG2, antibody 60.3 ((2, 231; Dr. l? G . Beatty), anti-CR2 (CD21) IgGI antibody BL-13 ([24]; Dr. J. Brochier).W6/32, a mAb against a human non polymorphic determinant of HLA class I antigens.W6/32 ( [ 2 5 ] ;Dr. D. J. Charron, Paris. France). IgG was purified from ascites fluid by contrapre-
lo00
Eur. J. Immunol. 1990. 20: 999-1005
C. Couturier, N. Haeffner-Cavaillon,L. Weiss et al.
cipitation with octanoic acid [26]. F(ab')2 fragments were prepared by pepsin (Millipore, Bedford, MA) digestion (2.5% ,w/w for 2 to 6 h a t pH 3.5 and at 37 "C; [27]) followed by chromatography on Sephadex G-150 (Pharmacia Fine Chemicals, Upssala, Sweden) and chromatography on protein A-Sepharose (Pharmacia). The purity of F(ab')z fragments was assessed using SDS-PAGE. Purified IgG or F(ab'h fragments were coupled to CNBr-activated Sepharose (Phamacia) according to the manufacturer's recommendations.The amount of antibodies bound was 30 pg/mg of dry gel. For coating with F(ab')z fragments, 24-well culture plates (Falcon, Lincoln Park, NJ) were incubated with various concentrations of F(ab')2 in NaHC03 buffer, pH 8.8, for 1h at 37 "C. For binding experiments, 25.3.1, BL-5 and MN41 IgG were radiolabeled with 1251(Amersham, Les Ulis, France) to a sp. act. of 6.5 x 105 c p d p g using iodogen (Pierce, Rockford, IL). Anti-CDllb mAb MN41 and 60.1, and anti-CD18 mAb BL-5 and 60.3, recognize distinct epitopes on the a and p chains of the CDllb/CD18 heterodimer, respectively, as assessed by competitive binding studies using cytofluorometry and inhibition of formation of SRBC C3bi-rosettes. PE-conjugated anti-monocyte/macrophage antibody anti-Leu-M3 (Becton Dickinson, Mountain View, CA), FITC-conjugated rabbit Fab anti-mouse IgG antibody (Biosys, Compikgne, France) were purchased from the indicated sources.
activity was quantitated by means of a conventional Con A co-mitogenic assay using C3H/HeJ thymocytes [34]. Units of IL 1 activity/ml were calculated as follows:
2.3 Monocytes and monocyte cultures
2.5 Binding of antibodies to cell membrane antigens
PBMC were prepared by centrifugation of citrated venous blood from healthy adult donors on MSL (Eurobio, Paris, France). PBMC were enriched in monocytes by allowing the cells to adhere to plastic culture dishes (Falcon) for 45 min at 37 "C in the absence of serum. Under these conditions, adherent cells contained > 85% monocytes as assessed by morphological analysis by phase contrast microscopy and histochemical staining for nonspecific esterase activity (NSE) using a-naphthyl acetate as a substrate [28]. Mononuclear adherent cells (5 x 105 NSE+ cells/well) were cultured in RPMI 1640 medium (Whittaker MA Bioproducts, Walkersville, MD) containing 100 IU/ml penicillin, 100 p g / d streptomycin, 4 pg/ml polymyxin B and 1 pg/ml indomethacin in the presence or absence of IL 1 inducers and in the total absence of serum. Indomethacin was used to inhibit the neosynthesis of P G in activated monocytes. PG down-regulate IL 1transcription in monocytes [29] and inhibit the thymocyte proliferation used to quantitate IL 1 activity [30]. RPMI contained no detectable LPS as assessed in the limulus amebocyte lysate assay.
The ability of F(ab')2 fragments prepared from anti-CDlla and CD18 antibodies to bind to their respective antigens on monocytes was assessed by dual-color immunofluorescence and FCM using PE conjugated anti-Leu-M3 and FITCrabbit Fab anti-mouse IgG.
U/ml
cpm of test sample - background cprn background cpm
X
dilution of test sample,
where cpm of test sample was the mean [3H]dThd incorporation in triplicate cultures of thymocytes performed in the presence of SN from monocyte cultures, and background cpm [3H]dThd incorporation in thymocytes cultured in RPMI alone. The freeze-thawing procedure yielded reproducible results since the mean cell-associated IL 1 activity measured in eight lysates of LPS-stimulated monocytes from normal individuals (10 pg of Nm LPS/5 X 105 cells) was 796 U/ml k 61 (mean k SD). Each assay of IL 1 activity included cultures performed in the presence of a constant source of monocyte-associated IL 1 obtained by stimulation of monocytes of one individual with 10 pg/5 x 105cells of LPS as internal standard. The internal standard contained 1000 U/ml of thymocyte-stimulating activity and 2.2 ng/ml of IL 1p antigen. IL la and IL 1p antigen concentrations were determined using a commercial radioimmunoassay for IL la (Amersham) and an ELISA for I L l p (Cistron Biotechnology, Pine Brook, NJ).
The binding characteristics of antibodies 25.3.1, BL-5 and MN41 to NSE+ mononuclear adherent cells were studied by incubating adherent cells in culture wells (1 x 106celldwell) with increasing concentrations of '"I-labeled antibodies ranging between 12.5 and 400 pg in 200 pl of RPMI 1640 in the presence or absence of a 100-fold molar excess of cold antibody for 1 h at room temperature. The cells were extensively washed with RPMI 1640 and lyzed in 2 mM EDTA 1% SDS overnight [35]. Radioactivity in cell lysates was assessed using an LKB minigamma counter (LKB, Orsay, France). The binding data were analyzed according to Scatchard.
3 Results
2.4 Induction of IL 1 and assays for IL 1
3.1 Expression of CDlla, CDllb and CD18 by adherent mononuclear cells
Mononuclear adherent cells were cultured in the presence of inducers as described above. At various time points, SN from culture wells were harvested, centrifuged and were assessed for extracellularly released IL 1 activity. In order to determine cell-associated IL 1 activity, mononuclear adherent cells were lyzed by three freeze-thaw cycles at - 80 "C in RF'MI containing antibiotics [31-331. Monocyte lysates were Centrifuged at 2000 x g for 10 min and IL 1 activity o r concentration were measured in the SN. IL 1
Incubation of adherent mononuclear cells with 1251-labeled purified 25.3.1, MN41 and BL-5 IgG resulted in dosedependent saturable binding of the antibodies (Fig. 1). Scatchard plots of the binding data indicated that adherent monocytes express an average of 120OOO, 40 000 and 58 OOO anti-CDlla, anti-CDllb and antLCD18 accessible binding sites, respectively. The estimated affinity of antibodies 25.3.1, MN41 and BL-5 for their respective antigens was 1.1 X 108 M-I, 1.3 X 108 M-l and 5.2 X lo8 M-l. The
Eur. J. Immunol. 1990. 20: 999-1005
Induction of IL 1 by stimulation of LFA-1 and CR3
1001
10
/
5
rn0I.C"I..
0
I.10.')
Ir""ndlC.ll
J
0:1
[ '=I].antibody
0.2
0:4
0
input (pg/1rlO"cells)
Figure 1. Binding of '=I-labeled anti-CDllaantibody25.3.1(A), 1251-anti-CDllbantibody h4N41 (B) and 1251-anti-CD18 antibody BL-5 (C) to normal adherent monocytes. Shown are specificbinding curves obtained in the presence of a 100-fold molar excess of cold antibody. Each point represents the mean of duplicate measurements. Inserts: Scatchard plots of the binding data.
calculated affinity of MN41 for CDllb on adherent monocytes was similar to that found for CDllb on polymorphonuclear leukocytes in suspension (data not shown).
3.2 Induction of cell-associated IL 1 by anti-CDlla, anti-CDllb and anti-CD18 antibodies
shows that cell-associated IL 1 activity in cells stimulated with anti-CDllb was approximately 1.5-fold lower than in cells triggered with LPS for similar amounts of generated IL la protein.
In order to ascertain that induction of cell-associated IL 1 by Sepharose-boundantibodies was not mediated by the Fc portion of antibodies and FcyRII on monocytes, F(ab');! Serum-free culture of mononuclear adherent cells in the fragments were prepared from anti-adhesive protein antipresence of soluble purified antibodies 25.3, MN41 and bodies and control antibody W6/32 and either coupled to BL-5 for 24 h resulted in no significant induction of Sepharose or insolubilized on culture dishes. Sepharosecell-associated IL 1 activity and no release of extracellular bound F(ab');! from anti-CDllb antibody MN41 induced IL 1 (data not shown). When bound to Sepharose, the the production of cell-associated IL 1in a dose-dependent antibodies induced a dose-dependent cell-associated IL 1 manner (Fig. 4). Cell-associated IL 1 activity was generactivity (Fig. 2). There was some heterogeneity in the ated by anti-CDllb and anti-CD18 F(ab');! antibodies response to each antibody between the donors tested which recognize distinct epitopes on the a and fi chains of although all antibodies bound in a similar fashion to the the CDllb/CD18 heterodimer. The relative ability of cells of the various donors. The maximal amount of antibodies to trigger IL 1 production differed between cell-associated IL 1 generated in the presence of the donors (Fig. 5). The generation of IL 1 activity by insoluantibodies represented 35% to 100% of the cell-associated bilized F(ab')2 antibodies correlated with that of IL la and IL 1 response induced by 10 pg of LPS in cells from the IL 1fi antigens (data not shown). Insolubilized F(ab')2 same donors. As shown in the Figure, no release of fragments from anti-HLA class I antibody W6/32 did not extracellular IL 1 activity was observed despite the high induce IL 1 activity in monocytes from three donors that amounts of cell-associated IL 1that had been generated. In were tested . contrast, LPS induced both cell-associated and extracellularly released IL 1in cultures of cells from the same donors. Uncoated Sepharose beads did not induce any significant 3.3 Induction of cell-associated IL 1 by adhesion of cells-associated IL 1 activity. Triggering of IL 1production monocytes to plastic by anti-CDll/CDlS antibodies was not mediated by contaminating LPS since experiments were performed in the Cultures for 24 h of mononuclear adherent cells in FWMI presence of polymyxin B and since no extracellularIL 1was 1640containing polymyxin B in the absence of IL 1inducer resulted in generation of cell-associated IL 1activity but in generated in the presence of insolubilized antibodies. no release of extracellular IL 1 activity. Mean cellInduction of cell-associated IL 1activity in cells stimulated associated IL 1activity was 135 & 30 U/ml (mean f SEM) with Sepharose-boundantibodies correlated with the gen- in cells from 35 individuals. No significant cell-associated eration of cell-associated IL la and IL 1/3antigens. Fig. 3 IL 1activity (11.6 k 3.2 Wml) was found in cells that had shows the kinetics of induction of IL la and IL l p by adhered for 45 min during the process of monocyte Sepharose-bound anti-CDllb antibody. At 24 h, the isolation before the culture was started. Fig. 6 depicts the amount of IL 1proteins generated was 280 pg 106cells and kinetics of production of cell-associated IL 1 activity and 340 pg/106 cells for IL 1fiand IL l a , respectively. Fig. 3 also IL la and IL lfiantigens in monocytes from one individual
Eur. J. Immunol. 1990. 20: 999-1005
C. Couturier, N. Haeffner-Cavaillon, L. Weiss et al.
1002
20
-
0510
25
50
0510
25
50
0510
25
50
.->>, .c, 4-
oa?
r
O
--za go --
4 g 10-
Q)
0
E
c,
X
Q)
0-
---
yg 5-25.3/well
yg S,MN41/well
ug S-BLS/welI
LPSNm/well
Figure 2. Induction of cell-associated IL 1 (upper panel) and extracellular IL 1 activity (lower panel) by Sepharose-bound (S-) anti-CDlla, anti-CDllb, anti-CD18 mAb and LPS. Mononuclear adherent cells from four donors (I, 0 ,A and A) were cultured ( 5 x 105 NSE+ celldwell) for 24 h in the absence of serum, in the presence of indomethacin (0.5 &well), polymyxin B (2 &well) and in the presence of increasing amounts of Sepharose-bound mAb or LPS. No polymyxin B was present in the cultures performed in the presence of LPS. IL 1activity was measured by using a C3H/HeJ thymocyte co-stimulatory assay in the presence of Con A (0.075 pg/0.75 x 106 cells).The results are expressed as mean [3H]dThd cpm k SD incorporated in triplicate thymocyte cultures. [3H]dThd cpm incorporated in control experiments (uncoated Sepharose beads) were subtracted.These values were 2000,2960,1750 and 400 cpm for the highest input of beads for donors I,0 , A and A, respectively.
during a 24-h culture in serum-free medium. In kinetic studies, a constant feature in normal individuals was the finding of maximal IL 1 generation at 6 h. In most individuals, cell-associated IL 1activity decreased thereafter and often reached base line values at 24 h.
4 Discussion Monocyte/macrophages do not constitutively produce IL 1. Stimulation of monocytes with IL 1 inducers results in transcription of two distinct genes coding for IL l a and IL l p , intracellular accumulation of high molecular weight IL 1precursors and processing of pro-IL 1into the mature 17.5-kDa forms of IL l a and IL 16 [36, 371. The cellassociated IL 1generated comprises intracytosolic biologically active pro-IL l a , inactive pro-IL l p [38] and a biolog-
ically active membrane-associated form of IL la [39,40]. In addition t o the production of cell-associated IL 1, specific signals may trigger the processing of pro-IL 1fJ and the extracellular release of biologically active IL 1p [41].Thus, intact LPS induces both intracellular and extracellular IL 1 whereas the lipid A moiety of LPS or LPS encapsulated in liposomes induce accumulation of cell-associated IL 1[31, 421. The present study demonstrates that triggering of the adhesion-promoting proteins CDlldCD18 and CD1lb/CD18 of human monocytes induces the production of both intracellular IL la and IL 1p but no release of IL 1p. Thus, IL 1 may be produced upon physiological or pathological conditions involving the adhesion of monocytes to surfaces. mAb to the a and p chains of the CDlla/CD18 and CDllbKD18 complexes induced the production of cell-associated IL 1in serum-free cultures of monocytes in a dose-dependent manner when coupled to
Eur. J. Immunol. 1990.20: 999-1005
Induction of IL 1 by stimulation of LFA-1 and CR3
0
6
1003
12 18 24
'0
(D
x
- 0 ) -
2 - 'P 5
6
-01 z 'p - 2-1 \uc
4
Q
u)
D 0
6
12
18 24
0 S
time (hours)
Sepharose beads. Monomeric antibodies in solution had no effect indicating a requirement for a high density of ligand or a cross-linking of receptors in order to induce IL 1 production. As for other IL 1 inducers including LPS [43] and C3a [44],different amounts of IL 1were found to be generated by monocytes from various donors and in a few donors no intracellular IL 1 was detectable. These differences may pertain to the different ability of monocytes from different individuals to be activated upon adhesion to plastic. Several lines of evidence indicated that the generation of IL 1 by mAb bound to Sepharose beads was not mediated by contaminating LPS: (a) cultures were performed in the presence of polymyxin B at concentrations
i 2 1'8 24
time ( hours)
Figure 3. Kinetics of induction of cell-associated IL 1 (upper panel) and extracellular IL 1 (lower panel) activity (01, IL la antigen (A) and IL l p antigen (0) by Sepharosebound anti-CDllb antibodies (left panel) and LPS (right panel). Culture conditions and the assay for IL 1 activity were those depicted in Fig. 2. IL la and IL 1fi concentrations were determined by RIA and ELISA, respectively.
which inhibit the capacity of all LPS species to generate IL 1 [45],(b) anti-CDlla, anti-CDllband anti-CD18mAb did not induce IL 1production when present in a soluble form and (c) polymeric antibodies did not induce IL 1 release although they resulted in the generation of intracellular IL 1 in amounts that could be associated with extracellular release in monocytes stimulated with LPS. The IL 1-inducingcapacity of anti-CDlla, anti-CDllb and anti-CD18mAb did not involve the Fc part of the antibody molecule since F(ab')z fragments bound to Sepharose or insolubilized on culture plates also induced cell-associated IL 1.F(ab'h fragments of an antibody directed against the cell-associated IL1 activity (U/ml) 0 20 40 60
*O1 F(W2
'
cell-associated I L1 activity (U.W/rnl) 0 20 40 60
i:lP k 80.3 BL5
pg S.F(ab'), / w e l l
Figure 4 . Generation of cell-associatedIL 1activity by Sepharosebound F(ab')2 fragments from anti-CDllb antibody. Mononuclear adherent cells were cultured in the presence of Sepharose-bound F(ab');! anti-CDllb mAb MN41 ( O ) , Sepharose-bound F(ab')z from control anti-CD21 mAb BL-13 (0)and uncoated Sepharose beads (A).Culture conditions and assay for IL 1activity were those described in Fig. 2.
Figure 5. Induction by insolubilized F(ab')2 anti-CDllb mAb (MN41 and 60.1) and antLCD18 mAb (60.3 and BL.5) of cell-associated IL 1 activity in serum-free cultured mononuclear adherent cells from two donors (left and right panels). Culture wells were coated with F(ab')z fragments from anti-CDllb, anti-CD18 mAb or control antLCD21 mAb BL-13. Monocytes were allowed to adhere to precoated culture wells for 45 min at 37 "C and cultured in RPMI 1640 containing polymyxin B and indomethacin for 24 h. Cell-associatedIL 1 activity was measured as described in Fig. 2. Results are expressed in U/ml (see Sect. 2.4).
1004
Eur. J. Immunol. 1990.20: 999-1005
C. Couturier, N. Haeffner-Cavaillon, L. Weiss et al.
The induction of intracellular IL 1has been associated with that of a membrane form of IL la which, at least in mice, plays a role inTcell activation during the process of antigen presentation [55]. Triggering of IL 1 production through stimulation of adhesion-promoting proteins. may thus be relevant to physiological and pathological circumstances involving cell-cell interactions such as macrophage-T cell cooperation during antigen presentation or adherence of activated monocytes to endothelial cells.
time (hours)
We wish to thank Drs. C.D. Ross, l? Beatq, J. Brochier, D. Olive and D. J. Charronfor providing us with monoclonal antibodies to leukocyte antigens. The secretarial assistance of A . Vioux is gratefully acknowledged.
Figure 6. Kinetics of adhesion-dependent induction of cell-associated IL 1 activity (O), IL la antigen (A) and U, l p antigen (0). Monocytes were allowed to adhere to plastic culture dishes for Received June 1, 1989; in final revised form January 29, 1990. 45 min in the absence of serum of 37 "C and cultured in RPMI 1640 containing polymyxin B and indomethacin.
5 References 1 Wright, S. D. and Detmers, P. A., J. Cell. Sci. Suppl. 1988. 9: 99. 2 Patarroyo, M., Prieto, J., Beatty, P. G., Clark, E. A. and Gahmberg, C. G., Cell. Immunol. 1988. 113: 278. 3 Hynes, R. O., Cell 1987. 48: 549. 4 SBnchez-Madrid,F., Nagy, J. A., Robbins, E., Simon, F! and Springer,T. A., J. Exp. Med. 1983. 158: 1785. 5 Schwarting, R., Stein, H. and Wang, C.Y., Blood 1985. 65: 974. The kinetics of generation of cell-associated IL 1activity by 6 Mentzer, S. J., Faller, D. V. and Burakoff, S. J., J. Immunol. 1986. 137: 108. triggering C D l l and CD18 molecules were similar to those of LPS, lipid A and other inducers of intracellular IL 1[44]. 7 Amaout, M. A., Lanier, L. L. and Faller, D.V., J. Cell. Physiol. 1988. 137: 305. Induction of IL 1activity was associated with that of IL la 8 Dustin, M. L., Rothlein, R., Bhan, A. K., Dinarello, C. A. and and I L l p antigens. However, the amount of activity Springer,T. A., J. Immunol. 1986. 137: 245. relative to that of antigens induced by anti-CDllb antibody 9 Brown, E. J., Bohnsack, J. F. and Gresham, H. D., J. Clin. was significantly higher than that induced by LPS. These Invest. 1988. 81: 365. observations suggest either differential post-translational 10 Capron, M., Kazatchkine, M. D., Fischer, E., Joseph, M., modification and/or differential kinetics induced by CR3 as Butterworth, A. E., Kusnierz, J. P., Prin, L., Papin, J. P. and compared to LPS, or that triggering of CR3 induces Capron, A., J. lmmunol. 1987. 139: 2059. 11 Wright, S. D., Rao, F! E.,VanVoorhis,W. C., Craigmyle, L. S., co-secretion of IL 1 and of an IL 1 functional inhibitor. Iida, K., Talle, M. A., Westberg, E. E , Goldstein, G. and Silverstein, S. C., Proc. Natl. Acad. Sci. USA 1983. 80: Adhesion of monocytes to plastic in the absence of serum at 5699. 37 "C results in the production of intracellular IL 1 in L. and Kazatchkine, M. D., in Kazatchkine, M. D. amounts that greatly differ between donors but not in the 12 Weiss, (Ed.), Clinical Immunology and Allergy, Bailliere Tindall, release of extracellular IL 1. C D l l a , C D l l b and CD18 are Eastbome 1988, vol. 2, 387. involved in the process of adhesion of monocytes to plastic 13 Ross, G. D., Cain, J. A., Myones, B. L., Newman, S. L. and as indicated by the defective adherence of cells from Lachmann, P. J., Complement 1987. 4: 61. patients with CD11/CD18 deficiency [20, 48, 491. A 14 Wright, S. D., Levin, S. M., Jong, M. T. C., Chad, Z. and Kabbash, L. G., J. Exp. Med. 1989. 169: 175. frequent feature of adhesion-dependent intracellular IL 1 production is the transient character of the generation of 15 Dana, N., Styrt, B., Griffin, J. D.,Todd, R. F. m.,Klempner, M. S. and Amaout, M. A., J. Immunol. 1986. 137: 3259. IL 1message [50] and IL 1activity and antigens (Fig. 6). In this respect, induction of IL 1by adhesion to plastic differs 16 Wright, S. D. and Silverstein, S. C., J. Exp. Med. 1982. 156: 1149. from that observed by triggering of the CD11/CD18 17 Wright, S. D., Craigmyle, L. S. and Silverstein, S. C., J. Exp. complex with antibodies. Med. 1983. 158: 1338. 18 Ding, A.,Wright, S. D. andNathan, C., J. Exp. Med. 1987.165: There is evidence that the CDllb/CD18 molecule 733. expresses distinct binding sites for RGD-containing pep- 19 Westphal, O., Liideritz, 0.and Bister, F., Z . Naturforsch. 1952. 7: 148. tides [51, 521 and for carbohydrates on yeasts and unopsonized bacteria [13,53,54]. Experiments using inhibition 20 Fischer, A., Seger, R., Durandy, A., Lisowska-Grospierre,B., Virelizier, J. L., Le Deist, F., Griscelli, C., Fischer, E., of rosette formation suggested that CDllWCDlS binds Kazatchkine, M., Bohler, M. C., Descamps-Latscha,B.,Trung, intact LPS [14]. It is unclear at the present time whether the I? H., Springer,T. A., Olive, D. and Mawas, C., J. Clin. Invest. carbohydrate binding site on C D l l b binds lipid A or the 1985. 76: 2385. polysaccharidic moiety of LPS. There is, however, an 21 Eddy, A., Newman, S. L., Cosio, F., Lebien, T. and Michael, analogy between the IL 1-inducing capacity of lipid A A., Clin. Immunol. Immunopathol. 1984. 31: 371. which is restricted to intracellular IL 1 [31] and that of 22 Schmitt, D., Souteyrand, F,! Brochier, J., Czemielewski,J. and CDllb. Thivollet, J., Acta Derm. Venerol. 1982. 62: 193.
CR2 complement receptor which is not expressed on human monocytes and F(ab')z fragments of an antibody directed against a nonpolymorphic determinant of class I molecules had no effect. In addition, human FcyRII are known to poorly bind mouse antibodies of the IgGl subclass [46, 471.
Eur. J. Immunol. 1990. 20: 999-1005 23 Beatty, P. G., Ledbetter, J. A., Martin, P. S., Price,T. H. and Hansen, J. A., J. lmmunol. 1983. 131: 2913. 24 Cohen, J. H. M., Fischer, E., Kazatchkine, M. D., Brochier, J. and Revillard, J. P., Scand. J. Immunol. 1986. 23: 279. 25 Barnstable, C. J., Bodmer, W. E, Brown, G., Galfre, G., Milstein, C., Williams, A. F. and Ziegler, A., Cell 1978. 14: 9. 26 Steinbuch, M. and Audran, R., Arch. Biochem. Biophys. 1969. 134: 279. 27 Parham, P., J. lmmunol. 1983. 131: 2895. 28 Tucker, S. B., Pierre, R.Y. and Jordon, R. E., J. lmmunol. Methoah 1977. 14: 267. 29 Konket, S. L., Chensue, S. N. and Phan, S. H., J. Immunol. 1986. 136: 186. 30 Hayori,Y., Kurulansky,T. and Globuron, A., Eur. J. lmmunol. 1985. 15: 43. 31 Caroff, M., Cavaillon, J. M., Fitting, C. and HaeffnerCavaillon, N., Infect. lmmunol. 1986. 54: 465. 32 Lepe-Zunica, J. L. and Guery, I., Clin. Immunol. Immunopathol. 1984. 31: 222. 33 Arend,W. P. and Massoni, R. J., Clin. Exp. lmmunol. 1986.64: 656. 34 Gery, I., Gershon, R. K. and Waksman, B. H., J. Enp. Med. 1972. 136: 128. 35 Haeffner-Cavaillon, N., Klein, M. and Domngton, K. J., J. lmmunol, 1979. 123: 1905. 36 Durum, S. K., Schmidt, J. A. and Oppenheim, J. J., Annu. Rev. lmmunol. 1985. 3: 263. 37 March, C. J., Mosley, B., Larsen, A., Cerretti, D. F!, Braedt, G., Price, V., Gillis, S., Henney, C. S., Kronheim, S. R., Grabstein, K., Conlon, F! J., Hopp, T. F! and Cosrnan, D., Nature 1985. 315: 641. 38 Mosley, B., Urdal, D. L., Prickett, K. S., Larsen, A., Cosman, D., Conlon, P. J., Gillis, S. and Dower, S. K., J. Biol. Chem. 1987. 262: 2941. 39 Conlon, I! J., Grabstein, K. H., Alpert, A., Prickett, K. S., Hopp, T. P. and Gillis, S., J. lmmunol. 1987. 139: 98.
Induction of IL 1 by stimulation of LFA-1 and CR3
1005
40 Kurt-Jones, E. A., Beller, D. I., Mizel, S. B. and Unanue, E. R., Proc. Natl. Acad. Sci. USA 1985. 82: 1204. 41 Auron, P. E., Warner, S. J. C., Webb, A. C., Cannon, J. G., Bernheim, H. A., McAdam, K. J. P.W., Rosenwasser, L. J., Lo Preste, G., Mucci, S. E and Dinarello, C A., J. lmmunol. 1987. 138: 1447. 42 Bakouche, O., Brown, D. C. and Lachman, L. B., J. lmmunol. 1987. 138: 4256. 43 Haeffner-Cavaillon, N., Caroff, M. and Cavaillon, J. M., Mol. lmmunol. 1989. 26: 485. 44 Haeffner-Cavaillon, N., Cavaillon, J. M.,Laude, M. and Kazatchkine, M. D., J. Immunol. 1987. 139: 794. 45 Cavaillon, J. M and Haeffner-Cavaillon, N., Mol. Immunol. 1986. 23: 965. 46 Anderson, C. L. and Looney, R. J., lmmunol. Today 1986. 7: 264. 47 Tax,W. J. M., Herrnes, F. F. M.,Willems, R. W., Capel, P. J. A. and Koene, R. A. I!, J. lmmunol. 1984. 133: 1185. 48 Dana, N. and Amaout, M. A., in Kazatchkine, M. D. (Ed.), Clinical Immunology and Allergy, Baillikre Tindall, Eastborne 1988. vol. 2, p. 453. 49 Anderson, D. C., Miller, L. J., Schmalstieg, F. C., Rothlein, R. and Springer,T. A., J. Immunol. 1986. 137: 15. 50 Fuhlbrigge, R. C., Chaplin, D. D., Kiely, J. M. and Unanue, E. R., J. lmmunol. 1987. I38: 3799. 51 Wright, S. D., Reddy, I? A., Jong, M. T. C. and Erickson, B.W., Proc. Natl. Acad. Sci. USA 1987. 84: 1965. 52 Russel, D. G. and Wright, S. D., J. Exp. Med. 1988. 168: 279. 53 Wright, S. D. and Jong, M. T. C., J. Exp. Med. 1986. 164: 1876. 54 Bullock, W. E. and Wright, S. D., J. Exp. Med. 1987, 165: 195. 55 Wasik, M. A., Donnelly, R. €! and Beller, D. I., J. Immunol. 1988. 141: 3456.
Christine Couturier, Nicole Haeffner-Cavaillon, Laurence Weiss, Elisabeth Fischer and Michel D. Kazatchkine
INSERM U 28, H6pital Broussais, Paris
Induction o f IL 1 by stimulation of LFA-1 and CR.7
99Y
Induction of cell-associated interleukin 1 through stimulation of the adhesion-promoting proteins LFA-1 (CDlldCD18) and CR3 (CDllbKD18) of human monocytes* Serum-free culture of human monocytes in the presence of monoclonal antibodies to the LFA-1 a chain (CD lla) , CR3 u chain (CD llh) or p chain (CD18) bound to Sepharose induced the dose-dependent production of cellassociated interleukin (IL) I activity and of IL la and IL 1p antigens, but no release of extracellular IL 1 activity or antigen in the culture medium. Triggering of IL 1 production was also observed with insolubiiized anti-CDlI/CD18 F(ab’)? antibodies. Two cross-linked antibodies recognizing distinct epitopes on the C D l l b molecule induced cell-associated IL 1. Soluble antibodies did not induce IL 1 production. The kinetics of induction of IL 1 by stimulation of adhesionpromoting proteins differed from those of IL 1 induction by adhesion to plastic. The lack of induction of IL 1 release by stimulation of the CDlI/CDl8 molecules resembled the intracellular accumulation of IL 1 induced by lipid A. Induction of IL 1 by adhesive processes may be a mechanism by which Tcells trigger 1L I production by monocytes during antigen presentation.
1 Introduction The CDIl/CD18 complex comprises three molecules of the integrin supergene family which mediate cell adhesion [l-31. The three members of the CD1 K D 1 8 family are heterodimers with distinct u chains, C D l l a , C D l l b and CDllc, and a non-covalently bound identical 95-kDa f3 chain, CD18 [4]. Human peripheral blood monocytes express CDlla/CD18 (LFA-l), CDllWCD18 (CR3) and, to a lower extent CDllc/CD18 (p150,95) (2.51. Adhesion processes mediated by t h e CDlUCD 18 complex are cationand temperature dependent and may involve an ArgGly-Asp (RDG) sequence in the ligands. CD 1la/CD18 plays an important role in homotypic adhesion of IFN:/-stimulated monocytes [6] and in monocyte binding to endothelial cells 171. A major ligand for CDllalCDl8 is the intercellular adhesion molecule, ICAM-1 [8].CDll b/CD IS mediates adhesion reactions of monocytes to leukocytes, endothelial cells, target particles of phagocytosis and targets of antibody-dependent cellular cytotoxicity [7. 9. lo]. C D l l b also functions as a receptor for the iC3b cleavage fragment of C3 when presented as a multivalent ligand covalently bound to opsonized surfaces 111, 121. Evidence from blocking experiments using mAb indicates that C D l l b exhibits at least two functional binding sites, one for iC3b and other RDG-containing proteins, and another for carbohydrates and LPS [13. 141. A third functional site involved in the enhancement of granulocyte adhesion-dependent functions by CR3 has also been pos-
[I 76901
* This study was supported by Institut National de la Sante et de la Recherche MCdicale (INSERM) and by Association pour la Recherche contre le Cancer (ARC 67-66).
Correspondence: Nicolc Haeffner-Cavaillon. Unite d‘lrnrnunopathologie. INSERM U 28, HBpital Broussais. 96. rue Didot. F-75014 Paris, France 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
tulated [15]. CR3 mediates phagocytosis of iC3b-coated particles by activated human monocytes [16, 171. The molecule has been implicated in triggering enhanced expression of class 11antigens and macrophage activation in mice [ 181. The present study indicates that triggering of specific epitopes o n C D l l a , C D l l b and CD18 induces the production of cell-associated IL 1 by serum-free cultured human monocytes. Induction of IL 1by adhesive processes may he of relevance for Tcell activation during antigen pre5entation.
2 Materials and methods 2.1 Materials Polymyxin B and indomethacin were purchased from Sigma (St. Louis, MO). LPS from Neisseria meningitidis (NmLPS), extracted according to Westphal’s procedure [19], was a gift from Dr. D. Schulz, Institut MCrieux, Lyon, France. 2.2 Antibodies The following mouse mAb were obtained as indicated in parentheses: anti-cr chain of LFA-1 (CD lla) IgGl antibody 25.3.1 ([30];a gift from Dr. D. Olive, Marseille, France), anti-a chain of CR3 ( CD llb) IgGl antibody MN41 ([21]; a gift from Dr. G. D. Ross, Chapel Hill. NC) and IgGl antibody 60.1 ( [ 2 ] :a gift from Dr. P. G. Beatty, Seattle, WA). anti-common (3 chain (CD18) IgGl antibody BL-5 ([22]; a gift from Dr. J. Brochier, Montpellier, France) and IgG2, antibody 60.3 ((2, 231; Dr. l? G . Beatty), anti-CR2 (CD21) IgGI antibody BL-13 ([24]; Dr. J. Brochier).W6/32, a mAb against a human non polymorphic determinant of HLA class I antigens.W6/32 ( [ 2 5 ] ;Dr. D. J. Charron, Paris. France). IgG was purified from ascites fluid by contrapre-
lo00
Eur. J. Immunol. 1990. 20: 999-1005
C. Couturier, N. Haeffner-Cavaillon,L. Weiss et al.
cipitation with octanoic acid [26]. F(ab')2 fragments were prepared by pepsin (Millipore, Bedford, MA) digestion (2.5% ,w/w for 2 to 6 h a t pH 3.5 and at 37 "C; [27]) followed by chromatography on Sephadex G-150 (Pharmacia Fine Chemicals, Upssala, Sweden) and chromatography on protein A-Sepharose (Pharmacia). The purity of F(ab')z fragments was assessed using SDS-PAGE. Purified IgG or F(ab'h fragments were coupled to CNBr-activated Sepharose (Phamacia) according to the manufacturer's recommendations.The amount of antibodies bound was 30 pg/mg of dry gel. For coating with F(ab')z fragments, 24-well culture plates (Falcon, Lincoln Park, NJ) were incubated with various concentrations of F(ab')2 in NaHC03 buffer, pH 8.8, for 1h at 37 "C. For binding experiments, 25.3.1, BL-5 and MN41 IgG were radiolabeled with 1251(Amersham, Les Ulis, France) to a sp. act. of 6.5 x 105 c p d p g using iodogen (Pierce, Rockford, IL). Anti-CDllb mAb MN41 and 60.1, and anti-CD18 mAb BL-5 and 60.3, recognize distinct epitopes on the a and p chains of the CDllb/CD18 heterodimer, respectively, as assessed by competitive binding studies using cytofluorometry and inhibition of formation of SRBC C3bi-rosettes. PE-conjugated anti-monocyte/macrophage antibody anti-Leu-M3 (Becton Dickinson, Mountain View, CA), FITC-conjugated rabbit Fab anti-mouse IgG antibody (Biosys, Compikgne, France) were purchased from the indicated sources.
activity was quantitated by means of a conventional Con A co-mitogenic assay using C3H/HeJ thymocytes [34]. Units of IL 1 activity/ml were calculated as follows:
2.3 Monocytes and monocyte cultures
2.5 Binding of antibodies to cell membrane antigens
PBMC were prepared by centrifugation of citrated venous blood from healthy adult donors on MSL (Eurobio, Paris, France). PBMC were enriched in monocytes by allowing the cells to adhere to plastic culture dishes (Falcon) for 45 min at 37 "C in the absence of serum. Under these conditions, adherent cells contained > 85% monocytes as assessed by morphological analysis by phase contrast microscopy and histochemical staining for nonspecific esterase activity (NSE) using a-naphthyl acetate as a substrate [28]. Mononuclear adherent cells (5 x 105 NSE+ cells/well) were cultured in RPMI 1640 medium (Whittaker MA Bioproducts, Walkersville, MD) containing 100 IU/ml penicillin, 100 p g / d streptomycin, 4 pg/ml polymyxin B and 1 pg/ml indomethacin in the presence or absence of IL 1 inducers and in the total absence of serum. Indomethacin was used to inhibit the neosynthesis of P G in activated monocytes. PG down-regulate IL 1transcription in monocytes [29] and inhibit the thymocyte proliferation used to quantitate IL 1 activity [30]. RPMI contained no detectable LPS as assessed in the limulus amebocyte lysate assay.
The ability of F(ab')2 fragments prepared from anti-CDlla and CD18 antibodies to bind to their respective antigens on monocytes was assessed by dual-color immunofluorescence and FCM using PE conjugated anti-Leu-M3 and FITCrabbit Fab anti-mouse IgG.
U/ml
cpm of test sample - background cprn background cpm
X
dilution of test sample,
where cpm of test sample was the mean [3H]dThd incorporation in triplicate cultures of thymocytes performed in the presence of SN from monocyte cultures, and background cpm [3H]dThd incorporation in thymocytes cultured in RPMI alone. The freeze-thawing procedure yielded reproducible results since the mean cell-associated IL 1 activity measured in eight lysates of LPS-stimulated monocytes from normal individuals (10 pg of Nm LPS/5 X 105 cells) was 796 U/ml k 61 (mean k SD). Each assay of IL 1 activity included cultures performed in the presence of a constant source of monocyte-associated IL 1 obtained by stimulation of monocytes of one individual with 10 pg/5 x 105cells of LPS as internal standard. The internal standard contained 1000 U/ml of thymocyte-stimulating activity and 2.2 ng/ml of IL 1p antigen. IL la and IL 1p antigen concentrations were determined using a commercial radioimmunoassay for IL la (Amersham) and an ELISA for I L l p (Cistron Biotechnology, Pine Brook, NJ).
The binding characteristics of antibodies 25.3.1, BL-5 and MN41 to NSE+ mononuclear adherent cells were studied by incubating adherent cells in culture wells (1 x 106celldwell) with increasing concentrations of '"I-labeled antibodies ranging between 12.5 and 400 pg in 200 pl of RPMI 1640 in the presence or absence of a 100-fold molar excess of cold antibody for 1 h at room temperature. The cells were extensively washed with RPMI 1640 and lyzed in 2 mM EDTA 1% SDS overnight [35]. Radioactivity in cell lysates was assessed using an LKB minigamma counter (LKB, Orsay, France). The binding data were analyzed according to Scatchard.
3 Results
2.4 Induction of IL 1 and assays for IL 1
3.1 Expression of CDlla, CDllb and CD18 by adherent mononuclear cells
Mononuclear adherent cells were cultured in the presence of inducers as described above. At various time points, SN from culture wells were harvested, centrifuged and were assessed for extracellularly released IL 1 activity. In order to determine cell-associated IL 1 activity, mononuclear adherent cells were lyzed by three freeze-thaw cycles at - 80 "C in RF'MI containing antibiotics [31-331. Monocyte lysates were Centrifuged at 2000 x g for 10 min and IL 1 activity o r concentration were measured in the SN. IL 1
Incubation of adherent mononuclear cells with 1251-labeled purified 25.3.1, MN41 and BL-5 IgG resulted in dosedependent saturable binding of the antibodies (Fig. 1). Scatchard plots of the binding data indicated that adherent monocytes express an average of 120OOO, 40 000 and 58 OOO anti-CDlla, anti-CDllb and antLCD18 accessible binding sites, respectively. The estimated affinity of antibodies 25.3.1, MN41 and BL-5 for their respective antigens was 1.1 X 108 M-I, 1.3 X 108 M-l and 5.2 X lo8 M-l. The
Eur. J. Immunol. 1990. 20: 999-1005
Induction of IL 1 by stimulation of LFA-1 and CR3
1001
10
/
5
rn0I.C"I..
0
I.10.')
Ir""ndlC.ll
J
0:1
[ '=I].antibody
0.2
0:4
0
input (pg/1rlO"cells)
Figure 1. Binding of '=I-labeled anti-CDllaantibody25.3.1(A), 1251-anti-CDllbantibody h4N41 (B) and 1251-anti-CD18 antibody BL-5 (C) to normal adherent monocytes. Shown are specificbinding curves obtained in the presence of a 100-fold molar excess of cold antibody. Each point represents the mean of duplicate measurements. Inserts: Scatchard plots of the binding data.
calculated affinity of MN41 for CDllb on adherent monocytes was similar to that found for CDllb on polymorphonuclear leukocytes in suspension (data not shown).
3.2 Induction of cell-associated IL 1 by anti-CDlla, anti-CDllb and anti-CD18 antibodies
shows that cell-associated IL 1 activity in cells stimulated with anti-CDllb was approximately 1.5-fold lower than in cells triggered with LPS for similar amounts of generated IL la protein.
In order to ascertain that induction of cell-associated IL 1 by Sepharose-boundantibodies was not mediated by the Fc portion of antibodies and FcyRII on monocytes, F(ab');! Serum-free culture of mononuclear adherent cells in the fragments were prepared from anti-adhesive protein antipresence of soluble purified antibodies 25.3, MN41 and bodies and control antibody W6/32 and either coupled to BL-5 for 24 h resulted in no significant induction of Sepharose or insolubilized on culture dishes. Sepharosecell-associated IL 1 activity and no release of extracellular bound F(ab');! from anti-CDllb antibody MN41 induced IL 1 (data not shown). When bound to Sepharose, the the production of cell-associated IL 1in a dose-dependent antibodies induced a dose-dependent cell-associated IL 1 manner (Fig. 4). Cell-associated IL 1 activity was generactivity (Fig. 2). There was some heterogeneity in the ated by anti-CDllb and anti-CD18 F(ab');! antibodies response to each antibody between the donors tested which recognize distinct epitopes on the a and fi chains of although all antibodies bound in a similar fashion to the the CDllb/CD18 heterodimer. The relative ability of cells of the various donors. The maximal amount of antibodies to trigger IL 1 production differed between cell-associated IL 1 generated in the presence of the donors (Fig. 5). The generation of IL 1 activity by insoluantibodies represented 35% to 100% of the cell-associated bilized F(ab')2 antibodies correlated with that of IL la and IL 1 response induced by 10 pg of LPS in cells from the IL 1fi antigens (data not shown). Insolubilized F(ab')2 same donors. As shown in the Figure, no release of fragments from anti-HLA class I antibody W6/32 did not extracellular IL 1 activity was observed despite the high induce IL 1 activity in monocytes from three donors that amounts of cell-associated IL 1that had been generated. In were tested . contrast, LPS induced both cell-associated and extracellularly released IL 1in cultures of cells from the same donors. Uncoated Sepharose beads did not induce any significant 3.3 Induction of cell-associated IL 1 by adhesion of cells-associated IL 1 activity. Triggering of IL 1production monocytes to plastic by anti-CDll/CDlS antibodies was not mediated by contaminating LPS since experiments were performed in the Cultures for 24 h of mononuclear adherent cells in FWMI presence of polymyxin B and since no extracellularIL 1was 1640containing polymyxin B in the absence of IL 1inducer resulted in generation of cell-associated IL 1activity but in generated in the presence of insolubilized antibodies. no release of extracellular IL 1 activity. Mean cellInduction of cell-associated IL 1activity in cells stimulated associated IL 1activity was 135 & 30 U/ml (mean f SEM) with Sepharose-boundantibodies correlated with the gen- in cells from 35 individuals. No significant cell-associated eration of cell-associated IL la and IL 1/3antigens. Fig. 3 IL 1activity (11.6 k 3.2 Wml) was found in cells that had shows the kinetics of induction of IL la and IL l p by adhered for 45 min during the process of monocyte Sepharose-bound anti-CDllb antibody. At 24 h, the isolation before the culture was started. Fig. 6 depicts the amount of IL 1proteins generated was 280 pg 106cells and kinetics of production of cell-associated IL 1 activity and 340 pg/106 cells for IL 1fiand IL l a , respectively. Fig. 3 also IL la and IL lfiantigens in monocytes from one individual
Eur. J. Immunol. 1990. 20: 999-1005
C. Couturier, N. Haeffner-Cavaillon, L. Weiss et al.
1002
20
-
0510
25
50
0510
25
50
0510
25
50
.->>, .c, 4-
oa?
r
O
--za go --
4 g 10-
Q)
0
E
c,
X
Q)
0-
---
yg 5-25.3/well
yg S,MN41/well
ug S-BLS/welI
LPSNm/well
Figure 2. Induction of cell-associated IL 1 (upper panel) and extracellular IL 1 activity (lower panel) by Sepharose-bound (S-) anti-CDlla, anti-CDllb, anti-CD18 mAb and LPS. Mononuclear adherent cells from four donors (I, 0 ,A and A) were cultured ( 5 x 105 NSE+ celldwell) for 24 h in the absence of serum, in the presence of indomethacin (0.5 &well), polymyxin B (2 &well) and in the presence of increasing amounts of Sepharose-bound mAb or LPS. No polymyxin B was present in the cultures performed in the presence of LPS. IL 1activity was measured by using a C3H/HeJ thymocyte co-stimulatory assay in the presence of Con A (0.075 pg/0.75 x 106 cells).The results are expressed as mean [3H]dThd cpm k SD incorporated in triplicate thymocyte cultures. [3H]dThd cpm incorporated in control experiments (uncoated Sepharose beads) were subtracted.These values were 2000,2960,1750 and 400 cpm for the highest input of beads for donors I,0 , A and A, respectively.
during a 24-h culture in serum-free medium. In kinetic studies, a constant feature in normal individuals was the finding of maximal IL 1 generation at 6 h. In most individuals, cell-associated IL 1activity decreased thereafter and often reached base line values at 24 h.
4 Discussion Monocyte/macrophages do not constitutively produce IL 1. Stimulation of monocytes with IL 1 inducers results in transcription of two distinct genes coding for IL l a and IL l p , intracellular accumulation of high molecular weight IL 1precursors and processing of pro-IL 1into the mature 17.5-kDa forms of IL l a and IL 16 [36, 371. The cellassociated IL 1generated comprises intracytosolic biologically active pro-IL l a , inactive pro-IL l p [38] and a biolog-
ically active membrane-associated form of IL la [39,40]. In addition t o the production of cell-associated IL 1, specific signals may trigger the processing of pro-IL 1fJ and the extracellular release of biologically active IL 1p [41].Thus, intact LPS induces both intracellular and extracellular IL 1 whereas the lipid A moiety of LPS or LPS encapsulated in liposomes induce accumulation of cell-associated IL 1[31, 421. The present study demonstrates that triggering of the adhesion-promoting proteins CDlldCD18 and CD1lb/CD18 of human monocytes induces the production of both intracellular IL la and IL 1p but no release of IL 1p. Thus, IL 1 may be produced upon physiological or pathological conditions involving the adhesion of monocytes to surfaces. mAb to the a and p chains of the CDlla/CD18 and CDllbKD18 complexes induced the production of cell-associated IL 1in serum-free cultures of monocytes in a dose-dependent manner when coupled to
Eur. J. Immunol. 1990.20: 999-1005
Induction of IL 1 by stimulation of LFA-1 and CR3
0
6
1003
12 18 24
'0
(D
x
- 0 ) -
2 - 'P 5
6
-01 z 'p - 2-1 \uc
4
Q
u)
D 0
6
12
18 24
0 S
time (hours)
Sepharose beads. Monomeric antibodies in solution had no effect indicating a requirement for a high density of ligand or a cross-linking of receptors in order to induce IL 1 production. As for other IL 1 inducers including LPS [43] and C3a [44],different amounts of IL 1were found to be generated by monocytes from various donors and in a few donors no intracellular IL 1 was detectable. These differences may pertain to the different ability of monocytes from different individuals to be activated upon adhesion to plastic. Several lines of evidence indicated that the generation of IL 1 by mAb bound to Sepharose beads was not mediated by contaminating LPS: (a) cultures were performed in the presence of polymyxin B at concentrations
i 2 1'8 24
time ( hours)
Figure 3. Kinetics of induction of cell-associated IL 1 (upper panel) and extracellular IL 1 (lower panel) activity (01, IL la antigen (A) and IL l p antigen (0) by Sepharosebound anti-CDllb antibodies (left panel) and LPS (right panel). Culture conditions and the assay for IL 1 activity were those depicted in Fig. 2. IL la and IL 1fi concentrations were determined by RIA and ELISA, respectively.
which inhibit the capacity of all LPS species to generate IL 1 [45],(b) anti-CDlla, anti-CDllband anti-CD18mAb did not induce IL 1production when present in a soluble form and (c) polymeric antibodies did not induce IL 1 release although they resulted in the generation of intracellular IL 1 in amounts that could be associated with extracellular release in monocytes stimulated with LPS. The IL 1-inducingcapacity of anti-CDlla, anti-CDllb and anti-CD18mAb did not involve the Fc part of the antibody molecule since F(ab')z fragments bound to Sepharose or insolubilized on culture plates also induced cell-associated IL 1.F(ab'h fragments of an antibody directed against the cell-associated IL1 activity (U/ml) 0 20 40 60
*O1 F(W2
'
cell-associated I L1 activity (U.W/rnl) 0 20 40 60
i:lP k 80.3 BL5
pg S.F(ab'), / w e l l
Figure 4 . Generation of cell-associatedIL 1activity by Sepharosebound F(ab')2 fragments from anti-CDllb antibody. Mononuclear adherent cells were cultured in the presence of Sepharose-bound F(ab');! anti-CDllb mAb MN41 ( O ) , Sepharose-bound F(ab')z from control anti-CD21 mAb BL-13 (0)and uncoated Sepharose beads (A).Culture conditions and assay for IL 1activity were those described in Fig. 2.
Figure 5. Induction by insolubilized F(ab')2 anti-CDllb mAb (MN41 and 60.1) and antLCD18 mAb (60.3 and BL.5) of cell-associated IL 1 activity in serum-free cultured mononuclear adherent cells from two donors (left and right panels). Culture wells were coated with F(ab')z fragments from anti-CDllb, anti-CD18 mAb or control antLCD21 mAb BL-13. Monocytes were allowed to adhere to precoated culture wells for 45 min at 37 "C and cultured in RPMI 1640 containing polymyxin B and indomethacin for 24 h. Cell-associatedIL 1 activity was measured as described in Fig. 2. Results are expressed in U/ml (see Sect. 2.4).
1004
Eur. J. Immunol. 1990.20: 999-1005
C. Couturier, N. Haeffner-Cavaillon, L. Weiss et al.
The induction of intracellular IL 1has been associated with that of a membrane form of IL la which, at least in mice, plays a role inTcell activation during the process of antigen presentation [55]. Triggering of IL 1 production through stimulation of adhesion-promoting proteins. may thus be relevant to physiological and pathological circumstances involving cell-cell interactions such as macrophage-T cell cooperation during antigen presentation or adherence of activated monocytes to endothelial cells.
time (hours)
We wish to thank Drs. C.D. Ross, l? Beatq, J. Brochier, D. Olive and D. J. Charronfor providing us with monoclonal antibodies to leukocyte antigens. The secretarial assistance of A . Vioux is gratefully acknowledged.
Figure 6. Kinetics of adhesion-dependent induction of cell-associated IL 1 activity (O), IL la antigen (A) and U, l p antigen (0). Monocytes were allowed to adhere to plastic culture dishes for Received June 1, 1989; in final revised form January 29, 1990. 45 min in the absence of serum of 37 "C and cultured in RPMI 1640 containing polymyxin B and indomethacin.
5 References 1 Wright, S. D. and Detmers, P. A., J. Cell. Sci. Suppl. 1988. 9: 99. 2 Patarroyo, M., Prieto, J., Beatty, P. G., Clark, E. A. and Gahmberg, C. G., Cell. Immunol. 1988. 113: 278. 3 Hynes, R. O., Cell 1987. 48: 549. 4 SBnchez-Madrid,F., Nagy, J. A., Robbins, E., Simon, F! and Springer,T. A., J. Exp. Med. 1983. 158: 1785. 5 Schwarting, R., Stein, H. and Wang, C.Y., Blood 1985. 65: 974. The kinetics of generation of cell-associated IL 1activity by 6 Mentzer, S. J., Faller, D. V. and Burakoff, S. J., J. Immunol. 1986. 137: 108. triggering C D l l and CD18 molecules were similar to those of LPS, lipid A and other inducers of intracellular IL 1[44]. 7 Amaout, M. A., Lanier, L. L. and Faller, D.V., J. Cell. Physiol. 1988. 137: 305. Induction of IL 1activity was associated with that of IL la 8 Dustin, M. L., Rothlein, R., Bhan, A. K., Dinarello, C. A. and and I L l p antigens. However, the amount of activity Springer,T. A., J. Immunol. 1986. 137: 245. relative to that of antigens induced by anti-CDllb antibody 9 Brown, E. J., Bohnsack, J. F. and Gresham, H. D., J. Clin. was significantly higher than that induced by LPS. These Invest. 1988. 81: 365. observations suggest either differential post-translational 10 Capron, M., Kazatchkine, M. D., Fischer, E., Joseph, M., modification and/or differential kinetics induced by CR3 as Butterworth, A. E., Kusnierz, J. P., Prin, L., Papin, J. P. and compared to LPS, or that triggering of CR3 induces Capron, A., J. lmmunol. 1987. 139: 2059. 11 Wright, S. D., Rao, F! E.,VanVoorhis,W. C., Craigmyle, L. S., co-secretion of IL 1 and of an IL 1 functional inhibitor. Iida, K., Talle, M. A., Westberg, E. E , Goldstein, G. and Silverstein, S. C., Proc. Natl. Acad. Sci. USA 1983. 80: Adhesion of monocytes to plastic in the absence of serum at 5699. 37 "C results in the production of intracellular IL 1 in L. and Kazatchkine, M. D., in Kazatchkine, M. D. amounts that greatly differ between donors but not in the 12 Weiss, (Ed.), Clinical Immunology and Allergy, Bailliere Tindall, release of extracellular IL 1. C D l l a , C D l l b and CD18 are Eastbome 1988, vol. 2, 387. involved in the process of adhesion of monocytes to plastic 13 Ross, G. D., Cain, J. A., Myones, B. L., Newman, S. L. and as indicated by the defective adherence of cells from Lachmann, P. J., Complement 1987. 4: 61. patients with CD11/CD18 deficiency [20, 48, 491. A 14 Wright, S. D., Levin, S. M., Jong, M. T. C., Chad, Z. and Kabbash, L. G., J. Exp. Med. 1989. 169: 175. frequent feature of adhesion-dependent intracellular IL 1 production is the transient character of the generation of 15 Dana, N., Styrt, B., Griffin, J. D.,Todd, R. F. m.,Klempner, M. S. and Amaout, M. A., J. Immunol. 1986. 137: 3259. IL 1message [50] and IL 1activity and antigens (Fig. 6). In this respect, induction of IL 1by adhesion to plastic differs 16 Wright, S. D. and Silverstein, S. C., J. Exp. Med. 1982. 156: 1149. from that observed by triggering of the CD11/CD18 17 Wright, S. D., Craigmyle, L. S. and Silverstein, S. C., J. Exp. complex with antibodies. Med. 1983. 158: 1338. 18 Ding, A.,Wright, S. D. andNathan, C., J. Exp. Med. 1987.165: There is evidence that the CDllb/CD18 molecule 733. expresses distinct binding sites for RGD-containing pep- 19 Westphal, O., Liideritz, 0.and Bister, F., Z . Naturforsch. 1952. 7: 148. tides [51, 521 and for carbohydrates on yeasts and unopsonized bacteria [13,53,54]. Experiments using inhibition 20 Fischer, A., Seger, R., Durandy, A., Lisowska-Grospierre,B., Virelizier, J. L., Le Deist, F., Griscelli, C., Fischer, E., of rosette formation suggested that CDllWCDlS binds Kazatchkine, M., Bohler, M. C., Descamps-Latscha,B.,Trung, intact LPS [14]. It is unclear at the present time whether the I? H., Springer,T. A., Olive, D. and Mawas, C., J. Clin. Invest. carbohydrate binding site on C D l l b binds lipid A or the 1985. 76: 2385. polysaccharidic moiety of LPS. There is, however, an 21 Eddy, A., Newman, S. L., Cosio, F., Lebien, T. and Michael, analogy between the IL 1-inducing capacity of lipid A A., Clin. Immunol. Immunopathol. 1984. 31: 371. which is restricted to intracellular IL 1 [31] and that of 22 Schmitt, D., Souteyrand, F,! Brochier, J., Czemielewski,J. and CDllb. Thivollet, J., Acta Derm. Venerol. 1982. 62: 193.
CR2 complement receptor which is not expressed on human monocytes and F(ab')z fragments of an antibody directed against a nonpolymorphic determinant of class I molecules had no effect. In addition, human FcyRII are known to poorly bind mouse antibodies of the IgGl subclass [46, 471.
Eur. J. Immunol. 1990. 20: 999-1005 23 Beatty, P. G., Ledbetter, J. A., Martin, P. S., Price,T. H. and Hansen, J. A., J. lmmunol. 1983. 131: 2913. 24 Cohen, J. H. M., Fischer, E., Kazatchkine, M. D., Brochier, J. and Revillard, J. P., Scand. J. Immunol. 1986. 23: 279. 25 Barnstable, C. J., Bodmer, W. E, Brown, G., Galfre, G., Milstein, C., Williams, A. F. and Ziegler, A., Cell 1978. 14: 9. 26 Steinbuch, M. and Audran, R., Arch. Biochem. Biophys. 1969. 134: 279. 27 Parham, P., J. lmmunol. 1983. 131: 2895. 28 Tucker, S. B., Pierre, R.Y. and Jordon, R. E., J. lmmunol. Methoah 1977. 14: 267. 29 Konket, S. L., Chensue, S. N. and Phan, S. H., J. Immunol. 1986. 136: 186. 30 Hayori,Y., Kurulansky,T. and Globuron, A., Eur. J. lmmunol. 1985. 15: 43. 31 Caroff, M., Cavaillon, J. M., Fitting, C. and HaeffnerCavaillon, N., Infect. lmmunol. 1986. 54: 465. 32 Lepe-Zunica, J. L. and Guery, I., Clin. Immunol. Immunopathol. 1984. 31: 222. 33 Arend,W. P. and Massoni, R. J., Clin. Exp. lmmunol. 1986.64: 656. 34 Gery, I., Gershon, R. K. and Waksman, B. H., J. Enp. Med. 1972. 136: 128. 35 Haeffner-Cavaillon, N., Klein, M. and Domngton, K. J., J. lmmunol, 1979. 123: 1905. 36 Durum, S. K., Schmidt, J. A. and Oppenheim, J. J., Annu. Rev. lmmunol. 1985. 3: 263. 37 March, C. J., Mosley, B., Larsen, A., Cerretti, D. F!, Braedt, G., Price, V., Gillis, S., Henney, C. S., Kronheim, S. R., Grabstein, K., Conlon, F! J., Hopp, T. F! and Cosrnan, D., Nature 1985. 315: 641. 38 Mosley, B., Urdal, D. L., Prickett, K. S., Larsen, A., Cosman, D., Conlon, P. J., Gillis, S. and Dower, S. K., J. Biol. Chem. 1987. 262: 2941. 39 Conlon, I! J., Grabstein, K. H., Alpert, A., Prickett, K. S., Hopp, T. P. and Gillis, S., J. lmmunol. 1987. 139: 98.
Induction of IL 1 by stimulation of LFA-1 and CR3
1005
40 Kurt-Jones, E. A., Beller, D. I., Mizel, S. B. and Unanue, E. R., Proc. Natl. Acad. Sci. USA 1985. 82: 1204. 41 Auron, P. E., Warner, S. J. C., Webb, A. C., Cannon, J. G., Bernheim, H. A., McAdam, K. J. P.W., Rosenwasser, L. J., Lo Preste, G., Mucci, S. E and Dinarello, C A., J. lmmunol. 1987. 138: 1447. 42 Bakouche, O., Brown, D. C. and Lachman, L. B., J. lmmunol. 1987. 138: 4256. 43 Haeffner-Cavaillon, N., Caroff, M. and Cavaillon, J. M., Mol. lmmunol. 1989. 26: 485. 44 Haeffner-Cavaillon, N., Cavaillon, J. M.,Laude, M. and Kazatchkine, M. D., J. Immunol. 1987. 139: 794. 45 Cavaillon, J. M and Haeffner-Cavaillon, N., Mol. Immunol. 1986. 23: 965. 46 Anderson, C. L. and Looney, R. J., lmmunol. Today 1986. 7: 264. 47 Tax,W. J. M., Herrnes, F. F. M.,Willems, R. W., Capel, P. J. A. and Koene, R. A. I!, J. lmmunol. 1984. 133: 1185. 48 Dana, N. and Amaout, M. A., in Kazatchkine, M. D. (Ed.), Clinical Immunology and Allergy, Baillikre Tindall, Eastborne 1988. vol. 2, p. 453. 49 Anderson, D. C., Miller, L. J., Schmalstieg, F. C., Rothlein, R. and Springer,T. A., J. Immunol. 1986. 137: 15. 50 Fuhlbrigge, R. C., Chaplin, D. D., Kiely, J. M. and Unanue, E. R., J. lmmunol. 1987. I38: 3799. 51 Wright, S. D., Reddy, I? A., Jong, M. T. C. and Erickson, B.W., Proc. Natl. Acad. Sci. USA 1987. 84: 1965. 52 Russel, D. G. and Wright, S. D., J. Exp. Med. 1988. 168: 279. 53 Wright, S. D. and Jong, M. T. C., J. Exp. Med. 1986. 164: 1876. 54 Bullock, W. E. and Wright, S. D., J. Exp. Med. 1987, 165: 195. 55 Wasik, M. A., Donnelly, R. €! and Beller, D. I., J. Immunol. 1988. 141: 3456.
