(ICAM-1) Is Controlled by Cell Spreading and the Activation
of Integrin LFA-1 ^
Mairi P. S te w a rt/ Carlos C a b a n a s /^ and Nancy Hogg"*
Many leukocyte integrins require activation before they can adhere to their ligands. For exam ple, stimulation o f T cells enables the integrin LFA-1 to bind to ligand. This study com pares tw o w ell known protocols for inducing T cell LFA-1 -m ediated adhesion to intercellular adhesion m olecule-1 (ICAM)-I. W e show that treatm ent with high concentrations o f the divalent cation Mg^^ induces a high affinity state o f LFA-1, w hich is reflected in the binding o f soluble ICAM-1 and correlates with the expression of the epitope recognized by mAb 24. The second stimulation protocol with the phorbol ester phorbol-12 ,1 3-dibutyrate (PDBu) does not induce a high affinity state o f LFA-1, and in this situation, adhesion is dependent on cell spreading and intracellular events involving protein kinase C, [Ca^’’’];, and actin polym erization. These low affinity LFA-1 receptors are responsible for the initial contact with im mobilized ligand because, unlike the Mg^^-stimulated receptors, adhesion is not blocked by soluble ICAM-1. Finally, w e used a third m ethod o f inducing LFA-1 -m ediated adhesion by stimulation o f T cells through the TCR/CD3 com plex. This procedure, which is considered to be a m ore physiologic trigger for LFA-1 activation, resem bles the phorbol ester protocol in that high affinity LFA-1 receptors are not induced and cell adhesion depends on involvem ent o f the cytoskeleton and cell spreading. The Journal o f Im m unology, 1 9 9 6 ,1 5 6 : 1 8 1 0 - 1 8 1 7 .
I nteraction of the integrin LFA-1 (C Dlla/CD 18; aLj32) with its hgands intercellular adhesion molecules (ICAM^)-l, -2, and -3 (CD54, CD 102, CD50) is central to many leukocyte adhesive events (reviewed in Refs. 1 and 2). LFA-1 is maintained in an inactive form on resting leukocytes and becomes activated following signalling through other cell surface receptors such as the TCR/CD3 complex. Phorbol esters such as phorbol-12,13- dibutyrate (PDBu) cause formation o f LFA-1/ICAM-1-dependent cell aggregates and have been used as another means of inducing integrin-mediated adhesion (3, 4).
The mechanism by which phorbol esters contribute to cell ad hesion has not yet been fully elucidated. Their best studied effect is to activate protein kinase C (PKC), which in turn is responsible for key phosphorylation events in signal transduction (5, 6). Al-
L eukocyte A d h esio n Laboratory, Im perial C a n c e r R esearch Fund, London, U n ited Kingdom
R eceived fo r p u b licatio n O c to b e r 1 9 ,1 9 9 5 . A ccep ted for p u b licatio n D ecem b er 2 2, 1 9 95.
The costs o f p u b licatio n o f this article w e re d efray e d in p a rt b y th e p ay m en t o f p ag e ch arg es. This article m u st th e refo re b e h ereb y m a rk ed a d vertisem en t in a c c o rd a n c e w ith 18 U .S.C. S ection 1 7 3 4 solely to in d icate th is fact.
’ This w ork w as su p p o rted by th e Im perial C a n c e r R esearch F und (ICRF). C.C. is a re c ip ie n t o f a P ostd o cto ral Fellow ship from th e M inisterio Espanol d e Educa- c id n y C iencia and is a ICRF Visiting R esearch F ellow from D ep artm en to d e B ioqufm ica y Biologia M o lecu lar, Facultad d e M ed icin a, U niversidad C om - p lu ten se, 2 8 0 4 0 M adrid, S pain. W e also gratefully a c k n o w le d g e an A ccio n es Integradas g ran t (1 9 9 3 -1 9 9 5 ) from th e British C o u n cil a n d S p an ish M inistry o f Education a n d S cien ce.
^ Both au th o rs c o n trib u te d eq u a lly to this w ork.
^ P resen t ad d ress: D ep arta m en to d e B ioqufm ica y B iologfa M o lec u lar III, Eac- ultad d e M ed ic in a, U n iv ersid ad C o m p lu ten se, 2 8 0 4 0 M adrid, Spain.
^ A ddress c o rre s p o n d e n c e an d rep rin t req u ests to Dr. N an cy H ogg, Leukocyte A d h esio n L aboratory, Im perial C an c er R esearch F und, 4 4 L in co ln 's Inn Fields, London, W C 2A 3PX U.K.
® A b b rev iatio n s u sed in this p ap er: ICAM, in tercellu la r a d h e sio n m o lecu le;
though phorbol ester treatment causes direct phosphorylation of both a and j8 subunits of the j82 integrins (7-9), mutational anal ysis has shown that this phosphorylation is not necessary for PMA- stimulated adhesion to ICAM-1 (10). More generally, phosphory lation could alter interactions of the j82 integrins with cytoskeletal proteins (8 ,1 1 ) and/or promote adhesion by increasing the avidity of cell contacts through integrin clustering (12), which has been reported for a second j82 integrin Mac-1 (CD 1 lb/CD 18) (13). Phorbol esters are also reported to cause an affinity increase in murine LFA-1 (14). Apart from effects on integrins, phorbol ester causes morphologic changes such as cell spreading. This could benefit the adhesive state by several means; it could reduce cell surface tension generated by shear stress of vascular flow (15,16), be required for integrin clustering and potentially make more in tegrin available for interaction with ligand through a process of diffusion of receptors and hgand trapping (17). Cell spreading oc curs following physiologic stimulation o f T cells in that it is ob served upon engagement of the TCR/CD3 complex (8, 18) and during T cell interactions with antigen-pulsed presenting cells (19). It is also possible to directly convert LFA-1 to an active form using the divalent cations Mg^"^ or Mn^"^ (20, 21). Exposure to divalent cations is speculated to circumvent physiologic triggering procedures and directly bring about changes in the integrin, per mitting ligand binding to occur. For example, addition of Mn^"^ to isolated Mac-1 induces it to bind to fibrinogen (22), and Mg^"^ can induce a4]31 lacking an a subunit cytoplasmic domain to bind ligand (23). Divalent cations such as Mg^“^ probably regulate li gand interactions through selective binding to several sites on in tegrins (reviewed in Refs. 2,2 4 , and 25) and are thought to directly associate with the hgand binding site and/or control access to a cryptic binding site(s) through altering the conformation of the integrin.
In this study we initially use two different models to analyze T cell adhesion to lCAM-1 and investigate characteristics of the
The Journal of immunology 1811
aflSnity receptors and adhesion is facilitated though cell spreading, which is dependent on actin polymerization, PKC, and intracellu lar Ca^^ ([Ca^'*“]j). In contrast, a high concentration of the divalent cation Mg^"*" induces a similar level o f cell adhesion by direct induction of high affinity LFA-1 receptors as measured by the abil ity to bind soluble (s)ICAM-l. This high affinity receptor state correlates with expression of an epitope recognized by the activa tion reporter mAb 24. Mg^^-stimulated adhesion has less depen dence on the above-mentioned intracellular events.
Materials and Methods
Cells
T lym phoblast cells were expanded from unstim ulated peripheral blood m ononuclear cells by culture for 1-2 w k in R P M I1640 medium containing recombinant IL-2 (20 ng/ml; Eurocetus, U.K. Limited, Harefield, U.K.) with details as previously described (21). C ells w ere used betw een days 10 and 14.
M onoclonal antibodies and other reagents
Purification and FITC-conjugation o f mAb 24 has been previously de scribed (20). Other mAbs used in this study were the LFA-1 a subunit mAb 38 (20), which blocks LFA-1 function, and the anti-CD3 mAh U C H Tl which was obtained from Dr. Peter Beverley (University College, London, U.K.). PDBu, phalloidin-TRTTC and cytochalasin D were purchased from Sigma Chemical Co. (Dorset, U.K.). The intracellular Ca^"^ chelator bis- (o-aminophenoxy)-ethane-A,A,A^'A'-tetraacetic acid, tetraacetoxymethyl ester (BAPTA-AM) was purchased from Life Technologies (Paisley, U.K.). The PKC inhibitor Ro 31-8220 was obtained from Dr. Trevor H d - 1am, Roche Research Centre. Stock concentrations o f inhibitors were pre pared in DMSO, and in every relevant experiment an equivolume of DM SO was added in the control sample.
T Cell adhesion to im m obilized ICAM-1 Fc
A dimeric form of an IC A M -lFc chimeric protein consisting o f the five domains of lCAM -1 fused to the Fc fragm ent o f human Ig G l was prepared as previously described (26). The m ethod for quantitation of T cell adhe sion to ICAJM-lFc protein has also been previously described (27, 28). Flat-bottom 96-well Immulon 1 plates (Dynatech, Chantilly, VA) were precoated with 0.24 jug/well IC A M -lFc in PBS-A overnight at 4°C, blocked w ith 2.5% BSA in PBS-A for 1 h at room temperature, and then w ashed four times in PBS-A and twice in 20 mM H E P K , 140 mM NaCl, 2 m g/m l glucose, pH 7.4 (HEPES/NaCl buffer). 5 X 10’ cultured T cells were labeled with 200 juCi ^ 'C r for 1 h at 37°C or 25 jxCi of thymidine overnight at 37°C and then washed three times in HEPES/NaCl buffer. Fifty m icroliters of the cell suspension (5 X 10® cells/ml) was combined on IC A M -lFc-coated wells with 50 jllI o f HEPES/NaCl buffer containing ti trations o f M g’ "*", various combinations o f 50 nM PDBu, 1 mM EGTA, 1 m M Ca^"^, and mAbs. For TCR/CD3 triggering, 10 ju,g/ml U C H Tl was added to radiolabeled cells, which were maintained in RPM I 1640. After incubation on ice for 20 min, plates were centrifuged at 40 X g for 1 min, and the incubation was continued for 30 min at 37°C. Plates were washed gently four tim es with prewarmed RPMI, any bound cells were lysed in 1% Triton X-100, and the incorporated radioactivity was m easured using a Betaplate counter (LKB Instruments Inc., Bromma, Sweden).
Immunofluorescence microscopy
For immunofluorescence analysis, 13 mm round glass coverslips were pre coated with 0.72 jug o f IC A M -lFc in PBS-A overnight at 4°C. Coverslips were blocked with 2.5% BSA in PBS-A for 1 h at room temperature and then washed three times with PBS-A and once with HEPES/NaCl buffer. T cells were washed three times in HEPES/NaCl buffer before addition onto coverslips (5 X 10^ cells/coverslip) in the presence o f divalent cat ions, 1 m M EGTA or 50 nM PDBu and mAbs at 10 /ig/m l. For staining of intracellular polymerized actin, prefixed adherent cells were permeabilized with 0.2% Triton X-100 in HEPES/NaCl buffer for 10 min, incubated with 0.25 jug/ml TRITC-conjugated phalloidin for 20 min, washed twice, and mounted for fluorescence microscopy.
Flow cytom etry and assessment o f soluble ICAM-lFc binding
and incubated at 2 X 10® cells per well with the indicated concentrations o f either sIC A M -lFc or the control protein CD14Fc (a gift from Dr. David Simmons, Oxford, U.K.) (29). After a 30-min incubation at 37°C, cells w ere washed twice in ice-cold PBS-A with 0.2% BSA and then incubated with 10 jug/ml FITC-conjugated goat anti-human IgG Fc specific Ah (Jack son Immunoresearch Labs, W est Grove, PA) for 20 m in on ice. Unbound secondary Ab was removed by washing twice in ice-cold PBS-A with 0.2% BSA, and then fluorescence of live cells was detected using a FACScan (Becton Dickinson, San Jose, CA) flow cytometer. For TCR/CD3 trigger ing experiments, T cells were maintained in RPM I 1640 and stimulated with 10 jug/ml U C H T l for the 30-min incubation at 37°C and analyzed as described above. Flow cytometric analysis o f the m Ah 24 epitope was conducted as previously described (20). Briefly, purified FITC-conjugated mAb 24 (10 /xg/ml) was added to T cells suspended in H EPES/NaCl buffer in the presence o f divalent cations, EGTA and PDBu where appropriate. Following a 30-rain incubation at 37°C, the cells were washed three times in ice cold PBS-A with 0.2% BSA before analysis as above.
Results
A comparison o f two LFA-1-activating protocols: a) T cell adhesion to im m obilized lCAM-1
To define the parameters that are important in the promotion of T cell adhesion, we determined the percentage o f activated T cells adherent to immobilized lCAM-1 by means of the integrin LFA-1. The divalent cation Mg^'*', when used at concentrations above 1 mM, was able to induce LFA-l-dependent adhesion when EGTA was used to chelate Ca^"^ (Fig. lA; EGTA). There was little en hancing effect o f additional treatment with PDBu (EGTA + PDBu). In confirmation of previous findings (21), the presence of 1 mM extracellular Ca^"^ inhibited adhesion induced by Mg^“^ (Fig. lA; Ca"+).
When PDBu was used to induce adhesion in the presence of Ca^'^, there was an enhancing effect on adhesion at suboptimal levels ofMg^^ (—0.3 mM Mg^""") (Ca^“^ + PDBu). This adhesion- potentiating effect o f Ca^"^ in the presence of phorbol ester is dose dependent, requiring —0.5 mM Ca^"^ for induction of half maxi mal adhesion (data not shown). The anti-LFA-1 (CD 11 a) mAh 38 or the anti-lCAM-1 (CD54) mAh 15.2 inhibited the adhesion o f T cells in all the conditions, indicating the specificity o f binding for LFA-1 to lCAM-1 (data not shown). Thus there are two distinct methods for inducing comparable amounts o f LFA-l-mediated ad hesion of T cells to lCAM-1; the Mg^"^ protocol (using 5 mM Mg^"^ in the presence of 1 mM EGTA) and the phorbol ester protocol (using 50 nM PDBu in the presence of 1 mM Ca^'^ and 0.3 mM Mg^"^). fii these two activating procedures, Ca^^ appears to have opposing roles.
A comparison o f two LFA-1-activating protocols: b) expression o f activation reporter epitope recognized b y m Ab 24
The active state of LFA-1 is reflected in the expression of an epitope detected with mAh 24, which serves as an LFA-1 activa tion reporter (28,30). Therefore, the two LFA-1 adhesion-inducing protocols were assessed for expression o f this epitope. Figure IB illustrates that Mg^“^, at concentrations over 1 mM and in the ab sence of Ca^^ (EGTA), causes high expression o f the 24 epitope on LFA-1 and PDBu does not alter this expression further (EGTA + PDBu). As expected from previous studies, the addition o f Ca^^ had a substantial inhibitory effect on the expression of the Mg^"^- induced 24 epitope (Ca^“^) (21), which paralleled the adhesion results (Fig. lA ).
In the second activating procedure, PDBu in the presence of Ca^'*’ did not increase the expression o f the mAh 24 epitope
1812 ACTIVATED LFA-1 A N D CELL SPREADING CO NTRO L T CELL ADHESIO N 100-1 — 0 - - EGTA EGTA +PdBu
I
5 80- Ca-^^+PdBu 60-I
0 4—/' 0 .1 1 10 Mg * concentration (mM) B — EGTA 500- EGTA +PdBu +PdBu 400-I -
GI '
I 200- 100 4— /■ 0 .1 1 10 Mg concentration (mM) 350 1 •s g 3001 250- « 2001 — EGTA — • — Ca +PdBu IM/- 150-T/- 0 Mg concentration (mM)(compare Fig. 1, A and B\ Ca^"^ + PDBu) and suggest that the Mg^'^ and PDBu protocols do not promote adhesion by the same means, and one explanation for this difference could be in the nature of the LFA-1 receptors present under both conditions.
A comparison o f the two LFA-1 activating protocols: c) binding to soluble ICAM-1
To further define characteristics of the LFA-1 receptors activated following the two different activating protocols, T cells were ex amined for their ability to bind recombinant soluble (s)ICAM-l. A sICAM-lFc chimera in dimer form bound to LFA-1 following maximal Mg^'*“/EGTA stimulation with saturating concentration reached at ~ 4 0 0 jag/ml or ~ 1.8 pM (M. P. Stewart and N. Hogg, manuscript in preparation). This concentration of sICAM-1 can bind to T cells when stimulated with the Mg^^ protocol, but the same saturating amount o f sICAM-1 showed no binding to T cells stimulated according to the phorbol ester protocol (Fig. 1C). A control protein comprising the Ig superfamily member CD14 fused to the same Fc portion o f IgG did not bind to the T cells, and mAh 38 blocked the binding o f sICAM-1 to these cells indicating the specificity of the reaction for LFA-l/ICAM-1 (data not shown). As the binding o f soluble hgand to receptor is considered a measure of receptor affinity, we conclude that high affinity LFA-1 receptors are induced by the Mg^^ but not by the phorbol ester protocol. When Figure 1, B and C are compared, it is evident that 24 epitope is expressed under conditions that support sICAM-1 binding, in dicating that 24 epitope expression correlates with high affinity LFA-1 receptors.
Soluble ICAM-1 can block Mg^~^ -stimulated adhesion to im m obilized ICAM-1
Although the phorbol ester treatment protocol appeared not to in duce detectable numbers of high affinity LFA-1 receptors, it was possible that a small number o f such receptors had a critical role in initiating adhesion. Therefore, we next investigated the ability of sic AM -1 to interfere with the adhesion of T cells to immobilized ICAM-1. Incubation o f T cells with saturating concentrations of sic AM -1 (400 /xg/ml) could substantially inhibit Mg^'^-stimulated adhesion to immobilized ICAM-1, illustrating a dependence on high affinity LFA-1 for the adhesion process (Fig. 2). In contrast, the adhesion of phorbol ester-stimulated T cells was not inhibited by incubation with sICAM-1. This provides further evidence that phorbol ester-stimulated LFA-1 receptors are not in a high affinity state and suggests that adhesion occurs by other means.
tio n s w e r e te s t e d in t h e p r e s e n c e o f 5 0 n M o f t h e p h o r b o l e s t e r P D B u (Ca^"^ + P D B u ; E G T A -f P D B u ). A, T h e a d h e s i o n a s s a y w a s p e r f o r m e d a s d e s c r i b e d in Materials and Methods a n d r e s u lts e x p r e s s e d a s t h e p e r c e n t a g e o f a d h e r e n t c e l ls c a l c u l a t e d fr o m t h e to ta l s a m p l e . D a ta