of December 12, 2015.This information is current as

T Cell Adhesion to FibronectinNeuropeptides, Via Specific Receptors, Regulate

and Ofer LiderMia Levite, Liora Cahalon, Rami Hershkoviz, Lawrence Steinman

http://www.jimmunol.org/content/160/2/9931998; 160:993-1000; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/160/2/993.full#ref-list-1

, 16 of which you can access for free at: cites 42 articlesThis article

Subscriptionshttp://jimmunol.org/subscriptions

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/ji/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/cgi/alerts/etocReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 1998 by The American Association of9650 Rockville Pike, Bethesda, MD 20814-3994.The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on Decem

ber 12, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on D

ecember 12, 2015

http://ww

w.jim

munol.org/

Dow

nloaded from

Neuropeptides, Via Specific Receptors, Regulate T CellAdhesion to Fibronectin1

Mia Levite,* Liora Cahalon,* Rami Hershkoviz,* Lawrence Steinman,2*† and Ofer Lider3†

The ability of T cells to adhere to and interact with components of the blood vessel walls and the extracellular matrix is essentialfor their extravasation and migration into inflamed sites. We have found that the b1 integrin-mediated adhesion of restinghuman T cells to fibronectin, a major glycoprotein component of the extracellular matrix, is induced by physiologic concen-trations of three neuropeptides: calcitonin gene-related protein (CGRP), neuropeptide Y, and somatostatin; each acts via its ownspecific receptor on the T cell membrane. In contrast, substance P (SP), which coexists with CGRP in the majority of peripheralendings of sensory nerves, including those innervating the lymphoid organs, blocks T cell adhesion to fibronectin when inducedby CGRP, neuropeptide Y, somatostatin, macrophage inflammatory protein-1b, and PMA. Inhibition of T cell adhesion wasobtained both by the intact SP peptide and by its 1–4 N-terminal and its 4–11, 5–11, and 6–11 C-terminal fragments, used atsimilar nanomolar concentrations. The inhibitory effects of the parent SP peptide and its fragments were abrogated by an SPNK-1 receptor antagonist, suggesting they all act through the same SP NK-1 receptor. These findings suggest that neuropeptides,by activating their specific T cell-expressed receptors, can provide the T cells with both positive (proadhesive) and negative(antiadhesive) signals and thereby regulate their function. Thus, neuropeptides may influence diverse physiologic processesinvolving integrins, including leukocyte-mediated migration and inflammation. The Journal of Immunology, 1998, 160: 993–1000.

T he neuropeptides somatostatin (SOM),3 calcitonin gene-related peptide (CGRP), neuropeptide Y (NPY), and sub-stance P (SP) are localized in the central and peripheral

nervous systems, where they exert influence at many levels, in-cluding the modulation of activity in sensory neurons and the reg-ulation of endocrine function (1, 2). These neuropeptides also ap-pear in nerve terminals that innervate both fenestrated andnonfenestrated blood capillaries. They act as vasodilators and af-fect vascular permeability as well as the behavior of various celltypes, including lymphocytes (3–6). Specific G protein-coupledreceptors for CGRP, SOM, and SP have been detected, mainly bybinding techniques on monocytes and B and T cells (7–11). How-ever, the physiologic functions of these neuropeptide receptors, thespecific receptor subtypes involved, and the significance of theiractivation under normal conditions and/or pathologic states of theimmune system are currently unknown.

Neuropeptides, unlike classical immunologic signals, are se-creted from nerve endings in transient bursts and induce signalingin target T cells over a time frame of milliseconds to minutes.Neuropeptides act as conventional neurotransmitters, transducingsignals from the environment, which can then be communicated tospecific targets, including the immune system. Since neuropeptidesare released from nerve endings present in lymphoid tissues andextravascular tissues (1), we examined whether neuropeptides suchas SOM, CGRP, NPY, and SP could modify the T cell adhesive-ness of to extracellular matrix (ECM) ligands, a prerequisite pro-cess for T cell extravasation and migration that involves activa-tion-dependent modulation of the avidity of ECM binding tob1

(VLA) integrins (12–15).

Materials and MethodsReagents

The following were obtained from the sources indicated: BSA, fibronectin(FN), PMA, Gly-Arg-Gly-Asp-Ser, Gly-Arg-Gly-Glu-Ser, SOM, CGRP,NPY, SP, SOM antagonist (cyclo-[7-aminoheptanoyl-Phe-Trp-Lys-Thr-(bzl)]), CGRP antagonist (CGRP8–37), haloperidol, SP antagonist (also re-ferred to as spantide 1; [D-Arg1,D-Trp7,9,Leu11]SP), SP fragments (1–4,4–11, 5–11, 6–11, 7–11, 8–11, and 9–11), genistein, staurosporine, andpertussis toxin (Sigma Chemical Co., St. Louis, MO); NPY amino acidsequence 18–36 (Peninsula Laboratories, Belmont, CA); wortmannin (Bi-omol Research Laboratories, Plymouth, PA); GF109203X (bisindolyma-leimide I; a gift from Dr. Y. Zick, The Weizmann Institute of Science,Rehovot, Israel); recombinant human macrophage inflammatory pro-tein-1b (MIP-1b; PeproTech, Inc., Rocky Hill, NJ); HEPES buffer, anti-biotics, sodium pyruvate, and RPMI 1640 (Beit-Haemek, Israel);Na2

51[Cr]O4 (Amersham, Aylesbury, U.K.); and mAb to the human CD29molecule (b1 integrin), LFA-1, anda2-, a4-, anda5-chains of the VLAintegrins (Serotec, Oxford, U.K.).

T cells

Human T cells were purified from the peripheral blood of healthy donorsas follows. The leukocytes were isolated on a Ficoll gradient, washed, andincubated on petri dishes (37°C, humidified 10% CO2 atmosphere). After2 h, the nonadherent T cells were removed and incubated on nylon-woolcolumns (Novamed Ltd., Jerusalem, Israel). Nonadherent T cells were

*Department of Immunology, The Weizmann Institute of Science, Rehovot, Is-rael; and †Department of Neurology and Neurological Sciences, Stanford Uni-versity, Beckman Center, Stanford, CA 94305

Received for publication March 25, 1997. Accepted for publication October6, 1997.

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported in part by The National Multiple Sclerosis Society,The Abisch-Frenkel Foundation (Basel, Switzerland) for the Promotion of LifeSciences, and the National Institutes of Health.2 Address correspondence and reprint requests to Dr. Lawrence Steinman, De-partment of Neurology and Neurological Sciences, Beckman Center, B002, Stan-ford University, Stanford, CA 94305–5429. E-mail address: steiny@leland.stanford.edu3 Incumbent of the Weizmann League Career Development Chair in Children’sDiseases.4 Abbreviations used in this paper: SOM, somatostatin; CGRP, calcitonin gene-related protein; NPY, neuropeptide Y; SP, substance P; ECM, extracellular ma-trix; VLA, very late antigen; FN, fibronectin; GF109203X (bisindolymaleimide I;MIP-1b, macrophage inflammatory protein-1b; PTK, protein tyrosine kinase;PKC, protein kinase C.

Copyright © 1998 by The American Association of Immunologists 0022-1767/98/$02.00

by guest on Decem

ber 12, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

eluted, washed, and passed through human CD31 cell purification columns(Cedar-Lane, Willowbrook, Ontario, Canada). The resulting cell popula-tion was.92% T cells (15). Myelin basic protein (amino acid sequence87–99)-specific CD41 T cell lines of the Th2 phenotype were obtainedfrom SJL/J mice.

Adhesion assay

Adhesion of these T cells to FN-coated microtiter flat-bottom wells (1mg/well; Sigma) was assayed as previously described (15). Briefly, T cellswere labeled with Na2[

51Cr]O4, washed, resuspended in adhesion medium(RPMI 1640 supplemented with 2% BSA, 1 mM Ca21, 1 mM Mg21, 1%sodium pyruvate, 1% glucose, and 1% HEPES buffer), pretreated (30 min,37°C) with neuropeptides (10216-1025 M), and added to the FN-coatedwells. The microtiter plates were then incubated (37°C, 30 min, humidified10% CO2 atmosphere) and washed with PBS to remove nonadherent Tcells. The adherent T cells were lysed with 1% Tween-20 in 1 N NaOH,and the radioactivity in the resulting supernatants was determined in agamma counter. For each experimental group, results were expressed as themean percentage (6SD) of bound T cells from quadruplicate wells. Neu-ropeptide-treated T cell adhesion to BSA-coated wells and untreated T celladhesion to FN-coated wells were always,6%. The percentage of cellsthat adhered was calculated as follows: (counts per minute of residual cellsin the well/(total counts per minute of cells added to the well2 sponta-neous release of51Cr)) 3 100.

Blocking neuropeptide-induced T cell adhesion by specificantagonists51Cr-labeled T cells were treated with neuropeptide antagonists (1026 M)and 2 min later also with SOM, CGRP, or NPY (1028 M). The treated cellswere suspended in adhesion medium and incubated (30 min, 37°C) in ahumidified 10% CO2 incubator. The cells were seeded in the FN-coatedmicrotiter plates, and the plates were then returned to the incubator for anadditional 30-min incubation. The amount of T cell adhesion wasdetermined.

Involvement of specific integrins in neuropeptide-inducedT cell adhesion to FN51Cr-labeled T cells were treated (30 min) either with the RGD- or theRGE-containing peptides (50mg/ml) or with mAb (15–25mg/ml) specificto the human integrins (CD29, LFA-1, anda2, a4, anda5 chains of theVLA integrins). The T cells were then treated (30 min) with SOM, CGRP,or NPY (1028 M) and incubated (30 min, 37°C, humidified 10% CO2

incubator). The treated cells were seeded in FN-coated microtiter plates.The plates were returned to the incubator for an additional 30-min incu-bation, and T cell adhesion was determined as previously described.

Modulation of neuropeptide-induced T cell adhesion to FNby inhibitors of intracellular signaling pathways

T cells were exposed (10 min, 37°C) to genistein (100 nM), staurosporine(10 nM), pertussis toxin (2 mg/ml), GF109203X (20 nM), or wortmannin(100 nM), and then to SOM, CGRP, or NPY (1028 M) or PMA (25 ng/ml;30 min in a 37°C, humidified 10% CO2 incubator). These T cells were thenseeded in FN-coated microtiter plates. The plates were returned to theincubator for an additional 30-min incubation, after which T cell adhesionwas determined.

Inhibition of T cell adhesion to FN by SP or its fragments

T cells were treated (30 min) with SP (10214-1026 M) or SP C-terminusamino acid fragments (10210 M: peptides 4–11, 5–11, 6–11, 7–11, 8–11,or 9–11). The T cells were then exposed to PMA, CGRP (1028 M), orMIP-1b (20 ng/ml). In a parallel set of experiments, T cells were treatedwith 1026 M [D-Arg1,D-Trp7,9,Leu11]SP (spantide 1, a SP NK1 receptorantagonist) and, 2 min later, with SP (10210 M). Thirty minutes later, thesecells were exposed to PMA (30 min, 25 ng/ml) or CGRP (1028 M) andseeded onto FN-coated microtiter wells. The plates were returned to theincubator for an additional 30-min incubation, and then treated as previ-ously described.

Statistical analysis

Statistical significance was analyzed by Student’st test.

ResultsSOM, CGRP, and NPY induce T cell adhesion to FN

To investigate whether SOM, CGRP, NPY, and SP can influencethe adhesion of T cells to FN, freshly purified T cells obtainedfrom the peripheral blood of healthy human donors were radioac-tively labeled; treated with 1024 to 10216 M SOM, CGRP, NPY,or SP; and seeded on FN-coated surfaces. Thirty minutes later,their adhesion to FN-coated surfaces was assessed. The resultsindicated that physiologic concentrations (Refs. 1, 2, 5, 16) ofSOM, CGRP, or NPY (Fig. 1,A and B, respectively) inducedmarked levels of T cell adhesion to FN, at a magnitude of 10- to

FIGURE 1. Induction of T cell adhesion to FN by SOM, CGRP, andNPY. A, Human T cells were labeled with 51Cr; washed; resuspendedin adhesion medium; pretreated (30 min, 37°C) with SOM, CGRP, andNPY (1028 M); and added to FN-coated wells. Nonadherent T cellswere removed by washing after incubation, the adherent T cells werelysed, and the radioactivity in the resulting supernatants was deter-mined. Neuropeptide-treated T cell adhesion to BSA-coated wells anduntreated T cell adhesion to FN-coated wells were ,6%. One exper-iment representative of five is depicted. B, A mouse CD41 T cell linespecific for myelin basic protein peptide87–99 was labeled with 51Crand seeded onto FN-coated microtiter wells. T cell adhesion, inducedby SOM, CGRP, and NPY (10211, 1027, and 1028 M respectively) wasdetermined as described above. BG, background, untreated groups ofcells. One experiment representative of four is depicted.

994 NEUROPEPTIDES AND T CELL ADHESION TO FIBRONECTIN

by guest on Decem

ber 12, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

30-fold more than the background adhesion (adhesion of neu-ropeptide-treated T cells to BSA and of untreated T cells to FN-coated wells). Calculation of the percentage of T cells that adhereto FN of the total T cell population present in the assay (taking intoaccount the radioactivity level of the cells added to the well, theradioactivity of the residual cells in the well, and the backgroundradioactivity) showed that SOM, CGRP, and NPY induced adhe-sion of 25 to 50% of the T cells, a level comparable to T cell-FNinteractions induced by chemokines (14, 15). Figure 1A shows thatSOM, CGRP, and NPY at 1028 M induced 49, 47, and 33% T celladhesion, respectively. The adhesion induced by these neurotrans-mitters was dose dependent with several peaks. The maximalproadhesive effects of SOM, CGRP, and NPY were evident with10211, 1028, and 1025 M for SOM; 10213, 10210, and 1027 M forCGRP; and 10212 and 1028 M for NPY (data not shown).

A similar (but not identical) pattern of dose response appearedin all experiments, and the SD between experiments comparing Tcell adhesion level induced by a given concentration of neuropep-tide was in the range of 10 to 20%. Such a multiple peak patternof dose response was previously observed for neuropeptide-in-duced cytokine secretion by T cells (5) and for chemokine-in-duced, in vitro T cell adhesion or migration through FN (14).

In addition to being assayed on human T cells, SOM and CGRPwere tested for their effects on the adhesion of the anti-myelinbasic protein CD41 murine T cell line. The unstimulated T cellswere treated with SOM and CGRP exactly as described for thehuman cells. The results (Fig. 1B) indicated that SOM (10211 M),CGRP (1027 M), and NPY (1028 M) induced a 10-fold increaseover the background level of the murine T cell line, correspondingto adhesion of 35, 44, and 38% of the total T cell population,respectively. Thus, the neuropeptides tested induced the adhesionof resting T cells of human as well as murine origin. In contrast, SPdid not increase the level of T cell adhesion to FN beyond thebackground level (Fig. 5A).

SOM, CGRP, and NPY act directly on T cells to induce theiradhesion to FN

Certain proadhesive mediators exert their effects while acting intheir soluble or matrix-bound forms (17, 18). Therefore, we ex-amined whether SOM, CGRP, and NPY (at 1028 M) induce T celladhesion by interacting with immobilized FN, by direct effecting Tcells, or both. Significant adhesion of T cells to FN was evidentonly if the T cells were pretreated with the neuropeptides, regard-less of whether the neuropeptides were removed by washing be-fore the seeding of the cells on immobilized FN (Fig. 2A). Pre-treatment of FN with a similar concentration of the neuropeptidesdid not affect T cell adhesion, implying that the neuropeptidesexert their proadhesive role on T cells. Pre-exposure of T cells toSOM, CGRP, and NPY followed by the removal of these neu-ropeptides is sufficient to activate the integrins mediating the sub-sequent adhesion to FN. Hence, after activating their respectivereceptors, the neuropeptides do not have to be present at the timeof T cell adhesion.

Activation of specific T cell-expressed receptors for SOM,CGRP, and NPY leads to T cell adhesion

Specific neuropeptide receptor antagonists (at 1026 M) were usedto test whether the proadhesive effects of SOM, CGRP, and NPY

FIGURE 2. Analysis of SOM, CGRP, and NPY-induced T cell adhe-sion to FN. A, Interactions between neuropeptides and adherent T cellsduring neuropeptide-induced T cell adhesion to FN. T cells were in-cubated with SOM, CGRP, and NPY (1028 M). The neuropeptideswere then either added directly (unwashed) or washed from the T cellsbefore their incubation on FN-coated wells (T cell pretreatment andwashed). Alternatively, FN-coated wells were pretreated with the neu-ropeptides, which were removed before cell seeding (FN pretreatedand washed). One experiment representative of three is shown. B, Tcells were treated with the indicated neuropeptide antagonists (1026

M) and with the respective neuropeptides (1028 M). After incubation,the cells were seeded in FN-coated microtiter well plates. T cell ad-hesion was then determined. One experiment representative of four isshown. C, Analysis of the proadhesive effects of intact NPY and its

C-terminus fragment. T cells were pretreated with the indicated com-pounds, and their adhesion to FN was determined thereafter. * indi-cates p , 0.05 compared with neuropeptide treatment only. One ex-periment representative of four is shown.

995The Journal of Immunology

by guest on Decem

ber 12, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

were indeed due to their interactions with specific receptors ex-pressed on T cells. The results (Fig. 2B) demonstrate that theproadhesive effects of SOM and CGRP on T cells were specificallyand significantly (p , 0.05) inhibited in the presence of theirrespective antagonists (7, 8, 11). Thus, cyclo-[7-aminoheptanoyl-Phe-Trp-Lys-Thr(bzl)] (19), an antagonist of SOM receptor, andCGRP8–37, an antagonist of the CGRP receptor (20) inhibitedSOM- and CGRP-induced T cell adhesion to FN, respectively.NPY-induced T cell adhesion was specifically inhibited by halo-peridol, a dopaminergic receptor antagonist previously describedas having the ability to interfere with NPY-induced effects (21).None of the antagonists alone influenced the background levelsof T cell adhesion to FN. Therefore, the proadhesive effects ofSOM, CGRP, and probably NPY are functionally linked to di-rect interactions with their specific surface-expressed T cellreceptors.

We further investigated the involvement of the NPY receptor inNPY-induced T cell adhesion to FN as well as its subtype speci-ficity. In the absence of an available NPY-specific antagonist, wetested the proadhesive effect of an NPY18–36C-terminal fragment,a selective NPY receptor agonist for the Y2 receptor subtype (22).Figure 2Cshows that, similar to the effects of the intact NPYmolecule, the NPY18–36 fragment, which is highly active as aninducer of histamine release from mast cells (23), markedly in-duced a significant adhesion of T cells to FN. This finding suggeststhat T cells express a functional NPY receptor of the Y2 subtype,which upon activation may provide the T cells with a proadhesivesignal.

Neuropeptide-induced T cell adhesion to FN is mediated bythe a4b1 and a5b1 integrins

T cell recognition and adhesion to FN are mediated primarily bythea4b1 anda5b1 integrins (12, 13). Whether SOM-, CGRP-, andNPY-induced T cell adhesion was regulated by these integrinswas analyzed using mAb specific fora4, a5, and b1 integrin

moieties and a peptide containing the cell binding motif of FNand related ECM and plasma proteins, Arg-Gly-Asp (RGD),that is recognized bya5b1 integrin. Figure 3 shows that adhe-sion to FN of resting T cells induced by SOM, CGRP, and NPYwas specifically and significantly inhibited by the presence ofan mAb against CD29 (theb1 integrin chain), by an mAb spe-cific to thea4 anda5 integrin chains, and by the RGD-contain-ing peptide, but not by the RGE-containing peptide. The adhe-sion was not influenced by the mAb anti-VLA-2 (a2b1) andanti-LFA-1 (aLb2) integrins.

The level of T cell expression of both VLA-4 and VLA-5 inte-grins following 1-h treatment with SOM, CGRP, or NPY was notchanged significantly, as determined by the fluorescence intensityof FITC-conjugated anti-VLA-4 and anti-VLA 5 Abs binding tothe cells (data not shown). In fact, the neurotransmitters caused aninsignificant elevation of VLA-4 and VLA-5 expression that waswithin the range of elevation previously observed following 18-htreatment of the cells with either IL-2 or RANTES, which induceT cell adhesion (data not shown). Hence, neuropeptide-inducedadhesion of T cells to immobilized FN involves specific recogni-tion and binding of FN by thea4b1 anda5b1 integrins. The neu-ropeptide-induced proadhesive effect is not associated with ele-vated expression of these integrins, but probably with induction ofconformational changes.

The activation of SOM, CGRP, and NPY receptors leading toT cell adhesion is mediated through diverse intracellularsignaling pathways

For SOM, CGRP, and NPY to induce T cell adhesion to FN, twodistinct processes must take place: 1) the activation of their spe-cific T cell-expressed, G protein-coupled receptors and their char-acteristic signal transduction pathways, and 2) the translation ofthe specific receptor signaling into a chain of events culminating inthe activation of specific integrins mediating the subsequent ad-herence of T cells to their ECM ligands. These processes probablyinvolve propagation of conformational changes from the cytoplas-mic domains of the integrins to their extracellular ligand bindingsites by rearranging of the cytoskeleton and forming cell-ECMfocal adhesion sites (12, 23). To examine the putative signal trans-duction pathways involved in this biphasic process, we used spe-cific signal transduction inhibitors and tested their effects on theSOM, CGRP, and NPY-induced T cell adhesion to FN. PMA-induced T cell adhesion to FN served as a control. Figure 4 showsthat pretreatment of the T cells with pertussis toxin, a specificinhibitor of G1a-coupled signaling (G protein-coupled receptor)(15), abolished T cell adhesion subsequently induced by SOM,CGRP, and NPY (all at 1028 M).

Furthermore, the neuropeptide-induced T cell adhesion wasblocked by several kinase inhibitors, including 1) genistein, a pro-tein tyrosine kinase (PTK) inhibitor (23); 2) wortmannin, whichinhibits phosphoinositide-3 kinase (PI-3 kinase) activity (and alsoother kinases, including myosin light chain kinase and PI-4 kinase,at 100-fold higher concentrations than those required for the inhi-bition of PI-3 kinase) (24–26); 3) staurosporine, a potent, broadspectrum inhibitor of protein kinases including myosin light chainkinase, protein kinase A, protein kinase C (PKC), and protein ki-nase G; and 4) GF109203X, a selective cell-permeable PKC in-hibitor (27, 28). All signal transduction inhibitors blocked neu-ropeptide-induced T cell-FN adhesiveness (Fig. 4,A–C) in a dose-dependent manner (data not shown). Note that wortmannin and, toa lesser degree, genistein and pertussis toxinonly partially inter-fered with the PKC-activating PMA-induced T cell adhesion toFN (Fig. 4D). Together, these results suggest an interaction

FIGURE 3. Involvement of integrins in neuropeptide-mediated bind-ing of T cells to FN. The T cells were treated with peptides (Gly-Arg-Gly-Asp-Ser or Gly-Arg-Gly-Glu-Ser) or with mAb to the human inte-grins (CD29, LFA-1, VLA-2, VLA-4, and VLA-5). The T cells thustreated were then exposed to SOM, CGRP, and NPY (all at 1028 M). Tcell adhesion to FN was then determined. * indicates p , 0.05 com-pared with neuropeptide treatment only. One experiment representa-tive of five is shown.

996 NEUROPEPTIDES AND T CELL ADHESION TO FIBRONECTIN

by guest on Decem

ber 12, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

between SOM, CGRP, and NPY and their specific T cell-ex-pressed G protein-coupled receptors, which leads to up-regula-tion of a4b1 anda5b1 affinities and to subsequent binding to FNmediated through diverse intracellular signaling pathways in-volving characteristic G protein, PTK, PKC, and PI-3 kinasesignaling.

SP, through activation of its NK1 receptor, abrogates T celladhesion induced by CGRP, SOM, NPY, PMA, and MIP-1b

SP, an undecapeptide, failed to induce T cell adhesion to FN. Nev-ertheless, SP has been found at sites of inflammation and withinlymphoid organs where it is frequently colocalized in perivascularas well as nonvascular nerve fibers together with other neuropep-tides, mainly CGRP (1, 11, 29, 30). In view of these findings, wedecided to examine the possible functional interaction between SPand CGRP as well as that between SP and SOM or NPY regardingtheir effect on T cell adhesion. We first tested the modulatory effectof SP on T cell adhesion to FN induced by PMA and found that SPinhibited the proadhesive effect of PMA in a dose-dependent man-ner, with an apparent maximal inhibitory effect occurring at 1028

M (Fig. 5A). To examine whether SP exerted its effect on PMA-induced adhesion through specific interaction with its T cell-ex-pressed receptors, we used the SP-derivative [D-Arg1,D-Trp7,9,Leu11]SP, referred to as spantide 1, a specific receptorantagonist for the NK1 receptor subtype. Figure 5A shows thatspantide 1 abrogated the inhibitory effect of SP on PMA-inducedT cell adhesion to FN, thus suggesting that the inhibitory effect ofSP is indeed mediated through a functional T cell-expressed SPreceptor of the NK1 subtype.

The physiologic relevance of SP-induced inhibition of T celladhesion was examined by testing neuropeptide ability to interferewith the proadhesive effect of CGRP, SOM, NPY, and MIP-1b.The results indicated that SP (10210 M) inhibited the proadhesiveeffects of each of the three neuropeptides (Fig. 5B). Moreover, SPinhibited the proadhesive effect of MIP-1b, a chemokine that playsa role in directing the migration of leukocytes from blood vessels

to inflamed sites and induces T cell adhesion to ECM moieties (14,15, 31). Exposure of T cells to alternate sequential combinations ofSOM, CGRP, or NPY revealed that none of these molecule inter-fered with the adhesive effects induced by the other, nor did thesemediators affect MIP-1b-induced T cell adhesion to FN (data notshown). Interestingly, the proadhesive effects of SOM, CGRP,NPY, and MIP-1b were not synergistic, since neither combinationinduced a higher adhesion level than that observed by any of theseeffectors alone (data not shown).

SP inhibits T cell adhesion either through its six-amino acidcarboxyl terminus or via its N-terminal fragment

The carboxyl-terminal amino acid sequence of SP, which is con-served in all members of the tachykinin family, is involved invasodilation, smooth muscle contraction, saliva secretion, and paintransmission. In contrast, the naturally occurring NH2-terminalfragments of SP are active in stimulating histamine release frommast cells, modulation of catecholamine release, and induction ofantinociception (11). To determine whether the C-terminus portionof SP can inhibit T cell adhesion to FN, and if so, which aminoacids within it are required to exert the inhibitory potential, wetested the SP peptides 4–11, 5–11, 6–11, 7–11, 8–11, and 9–11.The results showed that SP C-terminal fragments 4–11, 5–11, and,to a slightly lesser degree, 6–11 at a concentration of 1028 Minhibited PMA- and CGRP-induced T cell adhesion to FN as ef-ficiently as the intact SP molecule (Fig. 6A). Shorter C-terminal SPfragments, even at a concentration of 1023 M, failed to inhibit Tcell adhesion to FN (data not shown). The inhibitory effect of the4–11, 5–11, and 6–11 SP fragments was abrogated by the spantide1 receptor antagonist. Thus, the six-amino acid long carboxyl ter-minal of SP, through amino acids 4, 5, and 6, can inhibit T celladhesion at the same concentration range and through the samereceptor subtype as the intact SP peptide.

In parallel, we examined the ability of the N-terminal fragmentof SP, amino acid sequence 1–4, to inhibit CGRP-induced T celladhesion to FN. Unexpectedly we found that the SP1–4 peptide

FIGURE 4. Modulation of neuropeptide-induced T cell adhesion to FN by inhibitors of intracellular signaling pathways. T cells were untreated(none) or treated with genistein (100 nM), staurosporine (10 nM), pertussis toxin (2 mg/ml), GF109203X (20 nM), or wortmannin (100 nM). TheT cells were then exposed to SOM, CGRP, or NPY (1028 M), or PMA, and the adhesion to FN of the labeled cells thus treated was determined.* indicates p , 0.05 compared with neuropeptide treatment only. One experiment representative of four is shown.

997The Journal of Immunology

by guest on Decem

ber 12, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

was able to block (.95%) T cell adhesion at concentration as lowas 10210 M (Fig. 6B). The inhibitory effect of the SP1–4 peptidewas completely abrogated in the presence of spantide-1 (Fig. 6B),suggesting that the suppressive effect of the SP1–4peptide was alsomediated through the SP NK1 receptor subtype.

DiscussionWe have described the capacity of SOM, CGRP, and NPY, atphysiologic concentrations, to induce T cell adhesion to FN, amajor glycoprotein component of the ECM and that of SP to in-hibit it. Our results indicated that T cells express functional recep-tors to SOM, CGRP, and SP, since specific receptor antagonists toeach of these neuropeptides specifically antagonized the effect in-duced by the respective neuropeptide. We postulate that CGRPexerts its proadhesive effects via interaction with CGRP1 receptor

subtype expressed on the T cells, rather than CGRP2, since itsreceptor antagonist, CGRP8–37, which has higher affinity to theCGRP1 than to the CGRP2 receptor subtype (29), blocked theCGRP-induced proadhesive effect.

In addition to the indications regarding SOM, CGRP, and SPreceptors, we have provided evidence for a functional response ofT cells to NPY and suggest that it is mediated through an NPYreceptor of the Y2 receptor subtype on T cells, since T cell adhe-sion could be also induced by the NPY carboxyl fragment, NPY18–

36, a selective agonist for the NPY-Y2 receptor subtype. Receptorsof the NPY Y1 subtype, but not that of the Y2, were previouslyreported to be expressed, in low levels, on rat splenic lymphocytes(31, 32). The NPY-Y2 receptor subtype is widely distributed inthe brain and in the periphery, where it is localized at prejunc-tional sites at the sympathetic neuro-effector junctions, suppressingthe release of neurotransmitters (33, 34). It is also localized onother nerve fibers, such as the parasympathetic and sensory Cfibers (34, 35).

The T cell adhesion to FN induced by SOM, CGRP, and NPYwas found to be mediated by thea4b1 anda5b1 integrins involvedin T cell-FN interactions (12, 13). Moreover, the neuropeptide-induced T cell adhesion probably involves diverse intracellularsignal transduction pathways, including characteristic G proteinsignaling, PTK, PKC, and PI-3 kinase, since all the relevant in-hibitors blocked the effect.

Integrin activation and subsequent T cell adhesion to ECM gly-coproteins occurs after cell activation, since integrins expressed onresting T cells do not mediate strong adhesion to counter-receptorsand ligands (36, 37). T cells may be activated by one of variouspossible mechanisms, such as activation with phorbol esters, che-moattractants, or cross-linking of functionally relevant surface re-ceptors (e.g., Ag receptor/CD3 complex, CD2, Ig, or MHC class IImolecules) (36). For any given cell type, multiple activation stim-uli can up-regulate the functional activities of integrins. The acti-vation-dependent regulation of integrin adhesiveness does not re-quire an increase in the amount of integrins on the cell surface but,rather, qualitative changes in the integrin-receptor affinity or cy-toskeleton-dependent clustering of integrins that serve to increasethe overall avidity of these receptors (38–40). Our results implythat the binding of SOM, CGRP, and NPY to their respective Tcell-expressed receptors induces T cell activation that subse-quently leads to up-regulation of integrin functional activity. Suchactivation of T cells by these neuropeptides may lead to other Tcell functions in addition to adhesion to ECM components.

In contrast to the proadhesive effect of SOM, CGRP, and NPY,SP blocked the adhesion of T cells to FN by activating its NK1,rather than the NK2 or NK3, receptor subtype, since a specificNK1 receptor antagonist abrogated the SP-induced effect. Ourfindings with the SP receptor contradict the claim that SP receptorsare absent on human PBL (35). Inhibition of T cell adhesion wasalso induced by SP 4–11, 5–11, and 6–11 fragments (but not byshorter C-terminus peptides) and by its 1–4 amino-terminus por-tion. Interestingly, both the C- and the N-SP fragments could begenerated in vivo by enzymatic cleavage of the intact molecule(41). The inhibitory effect induced by the intact SP (i.e., full-length) and its N-terminal and six- to eight-amino acid long C-terminal fragments were blocked by spantide-1, an SP NK-1 re-ceptor antagonist. These results suggest that the parent SP peptideas well as its fragments mediate their inhibitory effect (at physio-logic concentrations) through a T cell-expressed SP NK1 receptorand raise the possibility the human T cells harbor an SP receptordisplaying an extended binding site to which various SP fragmentscan bind to induce its activation (11). Indeed, recent studies havedemonstrated that both the parent SP molecule and its N- and

FIGURE 5. Inhibition by SP of T cell adhesion to FN. A, T cells weretreated with the indicated concentrations of SP and then with PMA.Alternatively, the T cells were pretreated with spantide 1 (1026 M),then with intact SP, and finally with PMA. The labeled T cells werethen incubated (37°C, 30 min) and seeded in FN-coated microtiterplates, and T cell adhesion following a 30-min incubation was deter-mined. * indicates p , 0.01 compared with PMA-induced adhesiononly. One experiment representative of four is shown. B, T cells, whichwere untreated (none; 2) or pretreated with SP (10210 M; 1), werethen exposed to NPY, CGRP, or SOM (1028 M) or MIP-1b (20 ng/ml).The adhesion to FN of T cells thus treated was determined after anadditional incubation. * indicates p , 0.01 compared with T cell ad-hesion induced by SOM, CGRP, NPY, or MIP-1b. One experimentrepresentative of four is shown.

998 NEUROPEPTIDES AND T CELL ADHESION TO FIBRONECTIN

by guest on Decem

ber 12, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

C-terminal fragments at 1 nM can modulate striatal dopamineoutflow (42, 43).The precise mechanism(s) by which SP blocks T cell adhesion iscurrently under investigation. Nevertheless, the ability of SP toblock (at physiologic concentrations) T cell adhesion induced byneuropeptides, MIP-1b, and PMA suggest that this neuropeptidecould inhibit T cell migration into inflamed sites. Neuropeptideswith antagonistic functions may colocalize in nerve fibers inner-vating lymphoid organs, just as antagonistic neurotransmitters maycolocalize in nerve fibers within the central nervous system (1).Previous studies have shown that SP and CGRP are cotransmittedfrom peripheral endings of sensory nerves, including those inner-vating most of the lymphoid tissues (1, 29). Our finding that thesetwo neurotransmitters have opposing effects on T cell adhesion(CGRP induces adhesion, while SP blocks it) suggests that colo-calizing neuropeptides may provide T cells with both positive andnegative information and thereby regulate their function.In conclusion, we suggest that neuropeptides, usually found andactive in the sensory nervous system, can also function in lym-

phoid organs and in inflamed sites via binding and activating theirrespective T cell-expressed receptors. Neuropeptides may, thereby,play roles in T cell activation, adhesion, and migration. The phys-iologic relevance of these observations should be backed up by invivo studies.

References1. Weihe, E., D. Nohr, S. Michel, S. Muller, H. J. Zentel, T. Fink, and J. Krekel.

1991. Molecular anatomy of the neuro-immune connection.Int. J. Neurosci.59:1.

2. Wilder, R. L. 1995. Neuroendocrine-immune system interactions and autoimmu-nity. Annu. Rev. Immunol. 13:307.

3. Payan, D. G., D. R. Brewster, A. Missirian-Bastian, and E. G. Goetzl. 1984.Substance P recognition by a subset of human T lymphocytes.J. Clin. Invest.74:1532.

4. Goebeler, M., U. Henseleit, J. Roth, and C. Sorg. 1994. Substance P and calci-tonin gene-related peptide modulate leukocyte infiltration to mouse skin duringallergic contact dermatitis.Arch. Dermatol. Res. 286:341.

5. Nio, D. A., R. N. Moylan, and J. K. Roche. 1993. Modulation of T lymphocytefunction by neuropeptides: evidence for their role as local immunoregulatoryelements.J. Immunol. 150:5281.

FIGURE 6. Analysis of SP amino acid sequences required for optimal inhibition and of the receptor subtype involved. A, The adhesion toFN-coated surfaces of T cells that were treated with either intact or C-terminus amino acid fragments of SP was determined. B, Same as in A, exceptthat the T cells were treated with SP1–4 N-terminal fragment (10210 M) in the presence or the absence of spantide-1. One experiment representativeof three is shown.

999The Journal of Immunology

by guest on Decem

ber 12, 2015http://w

ww

.jimm

unol.org/D

ownloaded from

6. Johnston, J. A., D. D. Taub, A. R. Lloyd, K. Conlon, J. J. Oppenheim, andD. J. Kelvin. 1994. Human T lymphocyte chemotaxis and adhesion induced byvasoactive intestinal peptide.J. Immunol. 153:1762.

7. Hiruma, K., T. Koike, H. Nakamura, T. Sumida, T. Maeda, H. Tomioka,S. Yoshida, and Fujita, T. 1990. Somatostatin receptors on human lymphocytesand leukaemia cells.Immunology 71:480.

8. Sreedharan, S. P., K. T. Kodama, K. E. Peterson, and E. J. Goetzl. 1989. Distinctsubsets of somatostatin receptors on cultured human lymphocytes.J. Biol. Chem.264:949.

9. McGillis, J. P., S. Humphries, and S. Reid. 1991. Characterization of functionalcalcitonin gene-related peptide receptors on rat lymphocytes.J. Immunol. 147:3482.

10. Christian, J., K. S. Wiedermann, and B. P. Candace. 1986. Substance P receptorsin rat spleen: characterization and autoradiographic distribution.Blood 68:1398.

11. Regoli, D., A. Boudon, and J.-L. Fauchere. 1994. Receptor and antagonists forsubstance P and related peptides.Pharmacol. Rev. 46:551.

12. Springer, T. A. 1994. Traffic signals for lymphocyte recirculation and leukocyteemigration: the multistep paradigm.Cell 76:301.

13. Shimizu, Y., G. A. van Seventer, K. J. Horgan, and S. Shaw. 1990. Regulatedexpression and binding of three VLA (b1) integrin receptors on T-cells.Nature345:250.

14. Lloyd, A. R., J. J. Oppenheim, D. J. Kelvin, and D. D. Taub. 1996. Chemokinesregulate T-cell adherence to recombinant adhesion molecules and extracellularmatrix proteins.J. Immunol. 156:932.

15. Gilat, D., R. Hershkoviz, Y. A. Mekori, I. Vlodavsky, and O. Lider. 1994. Reg-ulation of adhesion of CD41 T lymphocytes to intact or heparinase-treated sub-endothelial extracellular matrix by diffusible or anchored MIP-1b and RANTES.J. Immunol 153:4899.

16. Wang, X., R. R. Fiscus, L. Yang, and H. L. Mathews. 1995. Suppression of thefunctional activity of IL-2-activated lymphocytes by CGRP.Cell. Immunol. 162:105.

17. Nathan, C., and M. Sporn. 1991. Cytokine in context.J. Cell Biol. 113:981.18. Gilat, D., L. Cahalon, R. Hershkoviz, and O. Lider. 1996. Counter-interactions

between tissue-infiltrating T lymphocytes, pro-inflammatory mediators, and en-zymatically- modified extracellular matrix.Immunol. Today 17:16.

19. Fries, J. L., W. A. Murphy, J. Sueiras-Diaz, and D. H. Coy. 1982. Somatostatinantagonist analog increases GH, insulin, and glucagon release in the rats.Pep-tides 3:811.

20. Chiba, T., A. Yamaguchi, T. Yamatani, A. Nakamura, T. Morishita, T. Luni,M. Fukase, T. Noda, and T. Fujita. 1989. Calcitonin gene-related peptide receptorantagonist, human CGRP 8–37.Am. J. Physiol. 256:331.

21. Pascaud, X. B., M. Chovet, C. Roze, and J. L. Junien. 1993. Neuropeptide Y andsigma receptor agonist act through a common pathway to stimulate duodenalalkaline secretion in rats.Eur. J. Pharmacol. 231:389.

22. Mousli, M., A. Trifilieff, J. T. Pelton, J. P. Gies, and Y. Landry. 1995. Structuralrequirements for neuropeptide Y in mast cells and G protein activation.Eur.J. Pharmacol. 289:125.

23. Juliano, R. L., and S. Haskill. 1993. Signal transduction from the extracellularmatrix. J. Cell Biol. 120:577.

24. Zell, T., S. W. Hunt III, J. L. Mobely, L. D. Finkelstein, and Y. Shimizu. 1996.CD-28-mediated up-regulation ofb1-integrin adhesion involves phosphatidylino-sitol 3-kinase.J. Immunol. 156:883.

25. Cross, M. J., A. Stewart, N. M. Modkin, and M. J. Wakelam. 1995. Wortmanninand its structural analogue demethoxyviridin inhibit stimulated phospholipase A2

activity in Swiss 3T3 cells: wortmannin is not a specific inhibitor of phosphati-dylinositol 3-kinase.J. Biol. Chem. 270:25352.

26. Nakamura, I., N. Takahashi, T. Sasaki, S. Tanaka, N. Udagawa, H. Murakami,K. Kimura, Y. Kabayama, T. Kurogawa, and T. Suda. 1995. Wortmannin, aspecific inhibitor of .phosphatidylinositol 3-kinase, blocks osteoclastic bone re-sorption.FEBS Lett. 361:79.

27. Toullec, D., P. Pianetti, H. Coste, J. Kirilovsky, P. Bellevergue, T. Grand-Perret,M. Ajakane, V. Baudet, P. Boissin, E. Boursier, and F. Loriolle. 1991. The bisin-dolylmaleimide GF109203X is a potent and selective inhibitor of protein kinaseC. J. Biol. Chem. 266:15771.

28. Calvo, C.-F., G. Chavanel, and A. Senik. 1992. Substance P enhances IL-2 ex-pression in activated human T-cells.J. Immunol. 148:3498.

29. Maggi, C. A. 1995. Tachykinins and calcitonin gene-related peptides (CGRP) asa co-transmitter released from peripheral endings of sensory nerves.Prog. Neu-robiol. 45:1.

30. Adams, D. H., and S. Shaw. 1994. Leukocyte-endothelial interactions and reg-ulation of leukocyte migration.Lancet 343:831.

31. Petitto, J. M., Z. Huang, and D. B. McCarthy. 1994. Molecular cloning ofNPY-Y1 receptor cDNA from rat splenic lymphocytes: evidence for low levelsof mRNA expression and [125I]-NPY binding sites.J. Neuroimmunol. 54:81.

32. Wahlestedt, C., N. Yanaihara, and R. Hakanson. 1993. Evidence for different pre-and post-junctional receptors for neuropeptide Y and related peptides.Regul.Pept. 13:307.

33. Westfall, T. C., X. L. Chen, a. Ciarlegio, K. Henderson, K. Del Valle, M. Curf-man-Falvey, and L. Naes. 1990. In vitro effects of neuropeptide Y at the vascularneuroeffector junction.Ann. NY Acad. Sci. 611:145.

34. Stjernquist, M., and C. Owman. 1990. Further evidence for a prejunctional actionof neuropeptide Y on cholinergic motor neurons in the rat uterine cervix.ActaPhysiol. Scand. 138:95.

35. Roberts, A. I., J. Taunk, and E. C. Ebert. 1992. Human lymphocytes lack sub-stance P receptors.Cell. Immunol. 141:457.

36. Shimizu, Y., and S. W. Hunt III. 1996. Regulating integrin-mediated adhesion:one more function for PI 3-kinase?Immunol. Today 17:565.

37. Clark, E. A., and S. J. Brugge. 1995. Integrins and signal transduction pathways:the road taken.Science 268:233.

38. Faull, R. J., N. L. Kovach, J. M. Harlan, and M. H. Ginsberg. 1994. Stimulationof integrin-mediated adhesion of T lymphocytes and monocytes: two mechanismswith divergent biological consequences.J. Exp. Med. 179:1307.

39. Jakubowski, A., M. D. Rosa, S. Bixler, R. Lobb, and L. C. Burkly. 1995. Vascularcell adhesion molecule (VCAM)-Ig fusion protein defines distinct affinity statesof the very late antigen-4 (VLA-4) receptor.Cell Adhes. Commun. 3:131.

40. Stewart, M. P., C. Cabanas, and N. Hogg. 1996. T-cell adhesion to intracellularadhesion molecule-1 (ICAM-1) is controlled by cell spreading and the activationof integrin LFA-1.J. Immunol. 156:1810.

41. Andren, P. E., and Caprioli, R. M. 1995. In vivo metabolism of substance P in ratstriatum utilizing microdialysis/liquid chromatography/micro-electrospray massspectrometry.J. Mass Spectrom. 30:817.

42. Khan, S., E. Grogan, R. Whelpton, and A. T. Michael-Titus. 1996. N- and C-terminal substance P fragments modulate striatal dopamine outflow through acholinergic link mediated by muscarinic receptors.Neuroscience 73:919.

43. Khan, S., R. Whelpton, and A. T. Michael-Titus. 1996. Evidence for modulatoryeffects of substance P fragments (1–4) and (8–11) on endogenous dopamineoutflow in rat striatal slices.Neurosci. Lett. 205:33.

1000 NEUROPEPTIDES AND T CELL ADHESION TO FIBRONECTIN

by guest on Decem

ber 12, 2015http://w

ww

.jimm

unol.org/D

ownloaded from