The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pello, O. M.
Right arrow Articles by Mellado, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pello, O. M.
Right arrow Articles by Mellado, M.
The Journal of Immunology, 2006, 176: 1675-1685.
Copyright © 2006 by The American Association of Immunologists

Opioids Trigger {alpha}5beta1 Integrin-Mediated Monocyte Adhesion1

Oscar M. Pello*, Béatrice Duthey*, David García-Bernal{dagger}, José Miguel Rodríguez-Frade*, Jens V. Stein{ddagger}, Joaquín Teixido{dagger}, Carlos Martínez-A.* and Mario Mellado2,*

* Department of Immunology and Oncology, Centro Nacional de Biotecnología/Consejo Superior de Investigaciones Científicas, and {dagger} Centro de Investigaciones Biologícas/Consejo Superior de Investigaciones Científicas, Madrid, Spain; and {ddagger} Theodor-Kocher-Institut, Universität Bern, Bern, Switzerland


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Inflammatory reactions involve a network of chemical and molecular signals that initiate and maintain host response. In inflamed tissue, immune system cells generate opioid peptides that contribute to potent analgesia by acting on specific peripheral sensory neurons. In this study, we show that opioids also modulate immune cell function in vitro and in vivo. By binding to its specific receptor, the opioid receptor-specific ligand DPDPE triggers monocyte adhesion. Integrins have a key role in this process, as adhesion is abrogated in cells treated with specific neutralizing anti-{alpha}5beta1 integrin mAb. We found that DPDPE-triggered monocyte adhesion requires PI3K{gamma} activation and involves Src kinases, the guanine nucleotide exchange factor Vav-1, and the small GTPase Rac1. DPDPE also induces adhesion of pertussis toxin-treated cells, indicating involvement of G proteins other than Gi. These data show that opioids have important implications in regulating leukocyte trafficking, adding a new function to their known effects as immune response modulators.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The opioids are a family of small peptides produced in the nervous system, where they mediate important physiological functions including thermal regulation and control of pain sensation. Opioids are also expressed by immune system cells, such as neutrophils, monocytes, macrophages, and T cells, and participate in modulating the immune response (1).

Opioids act by binding to seven transmembrane-spanning G protein-coupled receptors (GPCR),3 which are coupled to pertussis toxin (PTX)-sensitive G proteins. There are three classical opioid receptors, µ, {kappa}, {delta} (MOR, KOR, DOR, respectively), as well as the structurally related nociceptin receptor (2). Agonists that bind to µ, {kappa}, {delta} opioid receptors induce similar effector responses, including inhibition of adenylyl cyclase, increase in potassium conductance, and inhibition of voltage-sensitive Ca2+ channels. Opioids also activate protein kinase C, Ca2+ mobilization from intracellular stores, phosphatidylinositol 3,4,5-trisphosphate (PIP3) formation, recruitment of membrane proteins such as RAS-GRF, and the MAPK transduction cascade (3, 4). Although PTX blocks most opioid-induced effects, numerous studies show that opioid receptors can interact with various G proteins (5) including Gz, which is expressed predominantly in neuronal tissues (6).

Immune system cells migrate from the circulation to inflamed tissue in a sequence that includes rolling, adhesion, and transmigration through the endothelium. This process is orchestrated by a number of molecules, including selectins, integrins, adhesion molecules, chemokines, and cytokines. In inflamed tissue, immune system cells also generate opioid peptides, which contribute to analgesia by acting on specific peripheral sensory nerves (7). Opioids also modulate cytokine production, Ab responses, cell-mediated immunity, phagocytic activity, and chemotaxis of immune system cells (1, 8, 9). Evidence suggests that opioids affect cell movement by triggering migration or by modifying chemoattractant-mediated movement (10, 11).

Based on in vitro and in vivo results, we describe an additional role for opioids in the immune system because by binding to DOR, opioids promote {alpha}5beta1-mediated monocyte adhesion, which in turn modulates cell migration. We analyzed the opioid-activated signaling cascade responsible for this effect and show that it includes G proteins, PI3K{gamma}, and the Vav-1/Rac1 axis. These data reflect the extensive role of opioids in the immune system and define a new target for intervention and design of drugs with therapeutic potential.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Cell lines and Abs

MonoMac-1 (DSM ACC252) cells used were from the German Collection of Microorganisms and Cell Cultures. Abs used include anti-DOR (Oncogene Research Products), anti-G{alpha}i1, anti-G{alpha}i2 (Calbiochem), anti-p110{gamma}, anti-GRK2, anti-G{alpha}t (Santa Cruz Biotechnology), anti-phospho-AKT (Ser473) and anti-Vav-1 (Cell Signaling Technology), anti-JNK and anti-phospho-JNK (Tyr185, Tyr223) (R&D Systems), anti-paxillin (BD Pharmingen), anti-G{alpha}z (Gramsch Laboratories), and anti-phospho-Tyr (Promega). mAbs include anti-RhoA, anti-ERK, and anti-phospho-ERK (E-4), anti-phospho-cofilin (Ser3) (Santa Cruz Biotechnology), anti-Rac1 (BD Pharmingen), anti-Src (Upstate Biotechnology), anti-beta-actin AC-15 (Sigma-Aldrich), anti-beta1 integrin LIA 1/2.1 (Cell Signaling Technology), FITC-CD62L, PE-CD106, PE-CD49e, PE-CD18 (BD Pharmingen), FITC-CD54, FITC-CD31, FITC-CD29 (Immunotech/Beckman Coulter), anti-{alpha}5beta1 integrin P1D6 (Invitrogen Life Technologies), and anti-CD61 (Chemikon). Neutralizing anti-beta2 LIA 3/2.1 and anti-{alpha}vbeta3 integrin ABA6D1 mAb were donated by Dr. F. Sánchez-Madrid (Hospital de la Princesa, Madrid, Spain) and Dr. R. Lacalle (Centro Nacional de Biotecnología, Madrid, Spain), respectively. Calcein-AM and BCECF-AM (2',7'-bis(carboxyethyl)-4(or 5)-carboxyfluorescein diacetoxymethyl ester) were from Molecular Probes; DPDPE (D-Pen2,5, p-Cl-Phe4)-Enkephalin, naltrindole, and PTX were from Sigma-Aldrich; fibronectin (FN), PP2 (4-amino-5-(4-chlorophenyl)-7-(t-butyl)pyrazolo[3,4-D]pyrimidine), and the inactive analog PP3 (4-amino-7-phenylpyrazolo[3,4-D]pyrimidine), LY294002, and wortmannin were from Calbiochem. CXCL12 was from PeproTech.

Human primary monocytes

Human PBMC were isolated from buffy coats from normal donors on a Lymphoprep (Nycomed) gradient according to standard procedures. Monocytes were purified by magnetic cell sorting using CD14 microbeads (Miltenyi Biotec).

Flow cytometry

Flow cytometry was as described (12), using biotin-labeled anti-DOR mAb (1 µg/50 µl/well, 60 min, 4°C), and FITC-labeled streptavidin (Southern Biotechnology Associates). Cell-bound fluorescence was determined at 525 nm (XL-MCL; Beckman Coulter). To evaluate the effect of DPDPE on membrane CD31, CD54, CD29, CD49e, CD106, CD18, and CD61 levels, cells were stained using commercial prelabeled specific mAb.

Short-interfering (si)RNA experiments

siRNA duplexes specific for human Rac1 were targeted to bases 229–249, according to sequences described (13). We designed siRNA for Rac1 (Rac Mut) with a U-to-A point mutation at position 242 and a siRNA duplex for human Vav-1 (targeted to bases 2134–2154), with sense strand CGUCGAGGUCAAGCACAUUdTdT. We also designed siRNA for Vav-1 (Vav-1 Mut) with a U-to-A point mutation in position 2145. The siRNA sequence sense strand used as negative control for siRNA activity was AUUGUAUGCGAUCGCAG ACdTdT. Control, Rac1, and Vav-1 siRNA duplexes were purchased from Dharmacon and Ambion. All 21 nucleotide siRNA duplexes were verified as target-specific by BLAST search against the human genome.

Transfection experiments

MonoMac-1 cells were transiently transfected with expression vectors coding for dominant negative PI3K{gamma} (PI3K{gamma}-KR) or constitutively active PI3K{gamma} (PI3K{gamma}-CAAX) forms, with a beta-adrenergic receptor kinase C terminus (betaARK-CT) or the transducin G protein {alpha} subunit (G{alpha}T), with control vectors (pCDNA3) or siRNA duplexes targeted to human Rac1 and Vav-1 sequences. Cells were resuspended in Nucleofector Solution V (5 x 106 cells/100 µl; Amaxa) and mixed with plasmids (2 µg/106 cells) or siRNA. Nucleofection was performed using an Amaxa Nucleofector, and transfectants were transferred to culture medium and assayed after 5 h. Transfection experiments of <80% efficiency were discarded.

cAMP level determination

Untreated or PTX-treated (0.1 µM, 2 h) MonoMac-1 cells (2 x 105 cells/ml) in RPMI 1640 containing 1 mg/ml BSA, 10 mM HEPES, and 1 mM IBMX (3-isobutyl-1-methylxanthine) were challenged with 10 µM forskolin (37°C, 3 min), then with DPDPE (10–7 M) for 10 min. The reaction was terminated by removing the medium and resuspending cells in 0.1 N HCl. cAMP production was measured with the cAMP Direct Immunoassay kit (Calbiochem).

Static adhesion assay

BCECF dye-labeled (Molecular Probes) human primary monocytes or MonoMac-1 cells were seeded on a 96-well plate precoated with FN, alone or with 10–7 M DPDPE or 50 nM CXCL12, then incubated and luminescence quantified as described (12). Where appropriate, cells were incubated with LY294002 (20 µM), wortmannin (20 µM), PTX (0.1 µg/ml), PP2 (5 µM), or PP3 (5 µM) for 120 min at 37°C or incubated with anti-{alpha}5beta1 integrin (CD49e, clone P1D6, 10 µg/ml), anti-beta1 integrin (CD29, LIA 1/2.1, 10 µg/ml), anti-beta2 integrin (LIA 3/2.1, 10 µg/ml), or anti-{alpha}vbeta3 mAb (ABA6D1, 10 µg/ml) for 60 min at 37°C. When necessary, cells were pretreated with 10–7 M naltrindole (60 min, 37°C) before stimulation. Luminescence was directly proportional to the number of cells per well from the 2.5 to 25 x 103 cell range.

Flow chamber cell adhesion assays

Plates were coated (37°C, 3 h) with 20 µg/ml FN alone or with 10–7 M DPDPE or 100 nM CXCL12, then blocked (1 h) with 2% BSA. Petri dishes were incorporated as the lower wall of a parallel flow chamber (IQuum) and mounted on an inverted microscope (IX-70; Olympus) connected to a CCD camera (Cohu Electronics). MonoMac-1 cells (2 x 106 cells/ml) were infused at a 0.5 dyn/cm2 flow rate in serum-free RPMI 1640 (37°C, 5 min). Flow was stopped and cells allowed to settle for different time periods. Total cells from several fields were counted before flow was restored (2 and 4 dyn/cm2), and cells remaining tightly bound for 3 min were counted. Data are presented as the percentage of bound cells compared with total cells in each field before reestablishing flow.

Immunoprecipitation and Western blot

After lysis of DPDPE-stimulated MonoMac-1 cells (2 x 107), immunoprecipitation, Western blot analysis, membrane stripping, and protein loading controls were as reported (12).

Pull-down assay

MonoMac-1 cells (5 x 106 cells/ml), plated on FN 24 h before the experiment, were activated with DPDPE (10–7 M). Cells were washed, lysed (30 min at 4°C with continuous rocking) in bacterial lysis buffer (20% sucrose, 10% glycerol, 50 mM Tris (pH 8.0), 0.2 mM Na2S2O5, 2 mM MgCl2, 2 mM DTT) containing 1 mM PMSF, 10 µg/ml aprotinin, and 10 µg/ml leupeptin, then centrifuged (15,000 x g for 15 min). Cell extracts (500 µg) were precipitated (1 h at 4°C with continuous rocking) with GST-PAK (GST-p21-activated kinase bound to glutathione-coupled Sepharose beads) or GST-C21 (GST-fused Rho-binding domain from rhotekin bound to glutathione-coupled Sepharose beads) constructs to detect Rac or Rho activity, respectively. Precipitates were washed, separated in SDS-PAGE, transferred to nitrocellulose membranes, and analyzed in Western blot for Rac or Rho using specific mAb. Protein loading was controlled with a protein detection kit (Pierce) before precipitation and by developing total Rho and Rac in the lysate. When necessary, cells were pretreated with PTX (0.1 µg/ml, overnight, 37°C), LY294002 or wortmannin (both at 20 µM, 120 min, 37°C), or PP2 or PP3 (5 µM, 120 min, 37°C) before stimulation.

In vitro PI3K activity

In vitro PI3K activity was determined using DOR immunoprecipitates from DPDPE-activated (10–7 M) cells, as described (12).

PCR analysis of cDNA

RNA was isolated from MonoMac-1 cells using TRIreagent (Sigma-Aldrich), and cDNA were thereafter obtained by using SuperScript First-Strand (Invitrogen Life Technologies) and manufacturer’s protocols. Up- and downstream oligonucleotide primers for DOR are 5'-CCCTGGCAATCGCCATCAC and 3'-TTGTAGTAGTCGATGGAGAGC. Primers for CXCR4 are 5'-AGTAGCCACCGCATCT GGAG and 3'-GAGCCCATTTCCTCGGTGT. The DOR gave rise to a 247-bp fragment, and CXCR4 a 252-bp fragment. PCR was performed as described (12); reaction products were separated on 3% NuSieve GTG agarose gel, and DNA bands stained by ethidium bromide.

Intravital microscopy

Intravital microscopy of mouse cremaster venules has been described (14). Briefly, BALB/c mice were anesthetized by i.p. injection of 5 mg/ml ketamine/1 mg/ml xylasine (10 ml/kg) and surgically prepared under a stereomicroscope (Leica Microsystems) to expose the cremaster muscle. The contralateral femoral artery was catheterized to permit retrograde injection of fluorescent cell suspensions. The mouse was transferred to an intravital microscope (INM 100; Leica) and body temperature maintained at 37°C using a heating lamp. Cremaster vessel and fluorescent cells were observed using x10 or x20 water immersion objectives by transillumination or epifluorescence illumination. Transilluminated and fluorescent events were visualized using a silicon-intensified target camera (Hamamatsu Photonics) and recorded for offline analysis (DSR-11 Sony; IEC-ASV). In some experiments, calcein-labeled human monocytes were injected first and their behavior in cremaster microcirculation recorded for at least 30 min. To evaluate the effect of opioids, 10–6 M DPDPE was added directly to the cremaster muscle 30 min after injection of labeled monocytes. For analysis, we examined seven venules per mouse in five independent animals. The rolling fraction was determined in each visible cremaster venule as the percentage of monocytes interacting with the endothelial lining divided by the total labeled cells entering the venule during the observation period. When necessary, cells were pretreated with 10–7 M naltrindole (60 min, 37°C) before injection.

Statistical analysis

For flow chamber assays and intravital microscopy experiments, we applied the paired Student’s t test. All statistical analyses were performed using GraphPad Prism software (GraphPad) resulting in values for p < 0.05, p < 0.01, and p < 0.001.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Opioids trigger rapid {alpha}5beta1 integrin-mediated monocyte adhesion

Human monocytes and T cells express opioid receptors (1), cross-desensitization processes have been described between opioid and chemokine receptors (11), and morphine alters leukocyte/endothelial cell interactions (15), all of which suggest that opioids modulate leukocyte trafficking. As an experimental model, we used the MonoMac-1 human monocytic cell line, which expresses DOR as assessed by flow cytometry (Fig. 1A) and confirmed by PCR analysis (Fig. 1B). These receptors are functional, as shown by the ability of DPDPE, the DOR-specific ligand, to inhibit the forskolin-mediated increase in intracellular cAMP levels (Fig. 1C). As predicted, this DOR-mediated response is blocked by PTX treatment, indicating Gi protein mediation (Fig. 1C).


Figure 1
View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 1. DPDPE triggers MonoMac-1 cell adhesion. A, MonoMac-1 cells were stained with anti-DOR Ab, followed by FITC-labeled goat anti-rabbit Ab. A representative experiment is shown of five performed. B, Amplification of gene fragments corresponding to DOR (lane 2) and CXCR4 (lane 4) with specific primers (see Materials and Methods), using cDNA from MonoMac-1 cells. Controls for the absence of genomic DNA are shown (lanes 3 and 4). C, cAMP levels were quantitated in forskolin-pretreated MonoMac-1 cells after DPDPE stimulation; the effect of PTX treatment is also shown. Data show the mean ± SD of five independent experiments. **, p < 0.01 calculated by Student’s two-tailed t test. D, Static adhesion assay of MonoMac-1 cells on FN, alone or with DPDPE. CXCL12 was used as a control. Also shown are the effects of naltrindole pretreatment on DPDPE-mediated MonoMac-1 adhesion. Results are expressed as number of adhered cells (Materials and Methods). The mean ± SD of three independent experiments is shown. **, p < 0.01; ***, p < 0.001 by Student’s two-tailed t test. E, Dynamic adhesion assay using MonoMac-1 cells after DPDPE stimulation and two shear-stress forces (2 and 4 dyn). Resistance to detachment is given as mean percentage ± SD (three independent experiments) of cells remaining bound after 3 min of washing. Data are shown from at least four fields counted, relative to total cells in each field before washing. *, p < 0.05; **, p < 0.01 calculated using two-tailed Student’s t test.

 
As cell adhesion is a crucial step in immune cell extravasation during inflammatory processes, and opioids can alter monocyte migration, we analyzed the effect of DPDPE on monocyte adhesion. MonoMac-1 cells were added to wells coated with FN alone or coimmobilized with DPDPE. Maximum adhesion was observed in the range from 10–7 to 10–5 M 5 min after ligand stimulation; as control, monocyte adhesion was determined on FN coimmobilized with the chemokine CXCL12 (Fig. 1D). DPDPE-mediated adhesion was blocked in cells pretreated with naltrindole, a highly selective DOR antagonist (16), indicating a critical role for this receptor in adhesion. DPDPE addition increased MonoMac-1 cell resistance to shear stress force (Fig. 1E), confirming the static adhesion results. PTX treatment affected neither opioid-induced adhesion (Fig. 2A, left) nor shear resistance (data not shown). As a control, PTX completely abrogated CXCL12-triggered MonoMac-1 cell adhesion (Fig. 2A, right). The results suggest that opioids not only modify chemokine-induced responses, but also trigger monocyte adhesion, even when Gi is blocked.


Figure 2
View larger version (42K):
[in this window]
[in a new window]
  		FIGURE 2. DPDPE-triggered MonoMac-1 cell adhesion depends on {alpha}5beta1 integrin activation. A, Static adhesion assay of untreated or PTX-treated MonoMac-1 cells on FN, alone or with DPDPE. Results are expressed as the number of adhered cells (Materials and Methods). The mean ± SD of four independent experiments is shown. CXCL12-mediated adhesion of PTX-treated MonoMac-1 cells was included as a control. Results are expressed as described (see Materials and Methods); data represent the mean ± SD of quadruplicate determinations. ***, p < 0.001 determined by Student’s two-tailed t test. B, Untreated or DPDPE-treated (10–7 M, 60 min, 37°C) MonoMac-1 cells were stained with indicated fluorescence-labeled mAb or an isotype-matched control and evaluated by flow cytometry. A representative experiment is shown of four performed. Values shown (inset) represent the percentage of positive cells (top) and the mean fluorescence intensity (bottom) indicated in each case. C, Static adhesion assay of MonoMac-1 cells untreated or treated with neutralizing anti-{alpha}5beta1, anti-beta1, anti-beta2, or anti-{alpha}vbeta3 mAb on FN, alone or with DPDPE. Results are expressed as in A. The mean ± SD of four independent experiments is shown. *, p < 0.05; **, p < 0.01 by Student’s two-tailed t test.

 
In flow cytometry analyses, DPDPE treatment did not alter MonoMac-1 cell membrane levels of L-selectin, {alpha}5beta1, beta1, beta2, and beta3 integrins or adhesion molecules such as ICAM-1, VCAM-1, or PECAM-1, even at 60 min postactivation (Fig. 2B). We tested whether DPDPE-induced MonoMac-1 adhesion to FN is integrin-dependent. Complete integrin neutralization is required to abrogate cell adhesion (17); we thus treated MonoMac-1 cells with saturating concentrations of neutralizing anti-integrin Abs before DPDPE stimulation. We found reduced DPDPE-triggered adhesion in cells pretreated with neutralizing anti-beta1 or anti-{alpha}5beta1 mAb (Fig. 2C). Anti-beta2 treatment had no effect on cell adhesion and anti-beta3 mAb treatment did not modify DPDPE-induced cell adhesion markedly. As control, none of the neutralizing mAb affected basal adhesion to FN (Fig. 2C). The results indicate that DPDPE induces {alpha}5beta1 integrin-mediated monocyte adhesion on FN, even in the absence of Gi protein activation.

Opioid-induced monocyte adhesion requires PI3K{gamma} activation

To characterize the signaling molecules involved in opioid-triggered monocyte adhesion, we pretreated MonoMac-1 cells with the PI3K inhibitors LY294002 or wortmannin, and determined DPDPE-induced adhesion. Pretreatment with either inhibitor drastically reduced DPDPE-triggered MonoMac-1 adhesion (Fig. 3A), indicating a role for PI3K in this process. Opioid-mediated PI3K activation through µ opioid receptor was reported (18), although it was restricted to JNK activation. In this study we show that PI3K activation is also needed for DOR-promoted monocyte adhesion.


Figure 3
View larger version (38K):
[in this window]
[in a new window]
  		FIGURE 3. DPDPE-triggered MonoMac-1 cell adhesion requires PI3K activation. A, Static adhesion assay of untreated, LY294002-treated, or wortmannin-treated MonoMac-1 cells on FN, alone or with DPDPE. Results are expressed as in Fig. 1C. The mean ± SD of four independent experiments is shown. **, p < 0.01 determined by Student’s two-tailed t test. B, Serum-starved MonoMac-1 cells, untreated or treated with PTX or LY294002, were stimulated with 10–7 M DPDPE (37°C). Cell lysates were immunoprecipitated with anti-DOR Ab, and an in vitro kinase assay was performed (see Materials and Methods). A representative experiment is shown of five performed. C, Serum-starved MonoMac-1 cells were incubated with 10–7 M DPDPE (37°C). Cell lysates were immunoprecipitated with anti-DOR Ab and Western blot developed with anti-p110{gamma} mAb (top). Protein loading was confirmed by reprobing membranes with anti-DOR Ab (bottom). Arrow indicates p110{gamma} position. D, MonoMac-1 cells transiently transfected with the PI3K{gamma} dominant negative construct (PI3K{gamma}-KR), constitutively active PI3K{gamma} (PI3K{gamma}-CAAX), or mock-transfected were lysed and analyzed in Western blot with anti-p110{gamma} mAb (left). The same cells were serum-starved and DPDPE-treated before lysis, then analyzed in Western blot with anti-phospho-Akt (right). We confirmed protein loading as mentioned, using anti-beta-actin mAb. E, Static adhesion assay of cells as in D on FN, alone or with DPDPE. Results are expressed as the fold increase compared with cells adhered to BSA (Materials and Methods). The mean ± SD of four independent experiments is shown. **, p < 0.01 by Student’s two-tailed t test. F, Untreated (top) or PTX-treated (bottom) MonoMac-1 cells were incubated with 10–7 M DPDPE (37°C). Cell lysates were immunoprecipitated with anti-DOR Ab and Western blot developed with anti-G{alpha}i (left) or G{alpha}z (right) mAb. As control, an unstimulated, unprecipitated MonoMac-1 cell lysate was developed with G{alpha}i or G{alpha}z mAb. G, MonoMac-1 cells, transiently transfected with the betaARK-CT or G{alpha}t constructs or mock-transfected, were lysed and analyzed in Western blot with anti-GRK2 (top) or anti-G{alpha}t (bottom) Ab as indicated. We confirmed protein loading as in D. H, Static adhesion assay of cells as in G, on FN alone or with DPDPE. Results are expressed as the fold increase compared with cells adhered to BSA (Materials and Methods). The mean ± SD of three independent experiments is shown. ***, p < 0.001 calculated by Student’s two-tailed t test.

 
The GPCR activate the class I PI3K isoform, PI3K{gamma} (19), and several reports show a role for PI3K in modulating lymphocyte motility (20). We thus analyzed PI3K{gamma} activity following DPDPE activation. Untreated, LY294002-treated, or PTX-treated MonoMac-1 cells were DPDPE-stimulated, lysed, and cell extracts immunoprecipitated with anti-DOR Ab before measurement of associated kinase activity. DPDPE induced rapid PI3K{gamma} activation (maximum at 10 s), which was blocked by LY294002 but was unaffected by PTX pretreatment (Fig. 3B).

To evaluate whether PI3K{gamma} activation requires its association to the DOR, DPDPE-treated cells were lysed and immunoprecipitated with anti-DOR Ab. PI3K{gamma} was analyzed by Western blot with anti-p110{gamma} Ab. DPDPE triggered rapid p110{gamma} association to the DOR (Fig. 3C). We controlled protein loading by reprobing the membrane with the immunoprecipitating Ab.

The results indicate that DPDPE triggers p110{gamma} recruitment to the DOR, and suggest that PI3K{gamma} activation participates in controlling monocyte adhesion to FN. To confirm these findings, we evaluated DPDPE-triggered cell adhesion in MonoMac-1 cells transfected with a dominant negative (PI3K{gamma}-KR) or a constitutively active (PI3K{gamma}-CAAX) PI3K{gamma} construct. Correct expression of PI3K{gamma} mutants was confirmed in Western blot (Fig. 3D, left). We lysed unstimulated or DPDPE-stimulated cells and blotted extracts with anti-phospho-AKT Ab to measure PI3K activity. As predicted, we detected constitutive phospho-AKT in PI3K{gamma}-CAAX-transfected cells, but not in unstimulated or DPDPE-activated PI3K{gamma}-KR transfectants (Fig. 3D, right). DPDPE induced adhesion of PI3K{gamma}-CAAX-transfected, but not of PI3K{gamma}-KR-transfected MonoMac-1 cells (Fig. 3E). These data also suggest that PI3K{gamma} activation alone is insufficient to mediate DPDPE effects; DPDPE activation is required even for PI3K{gamma}-CAAX-transfected cells.

DPDPE thus induces both monocyte adhesion and PI3K{gamma} activation; as PI3K{gamma} associates specifically to GPCR and is activated by G protein beta{gamma} subunits (21), a G protein other than Gi must be activated by DPDPE in PTX-pretreated cells to supply Gbeta{gamma}. Untreated and PTX-pretreated MonoMac-1 cells were DPDPE-stimulated before lysis, cell extracts were precipitated with anti-DOR Ab, and Western blots developed with anti-Gi (Fig. 3F, left), anti-Gz, and anti-Gq/11 Ab (Fig. 3F, right). Although G{alpha}i associated to DOR in untreated cells, G{alpha}z did not. G{alpha}z associated to DOR in PTX-treated cells, whereas G{alpha}i was not found in DOR immunoprecipitates in these conditions (Fig. 3F, bottom). G{alpha}q/11 did not associate to DOR in either untreated or PTX-treated cells (data not shown). The results show that Gi and Gz can activate PI3K{gamma}, and that under conditions in which G{alpha}i cannot associate to DOR, G{alpha}z binds the receptor and supplies the Gbeta{gamma} required for PI3K{gamma} activation.

To confirm the role of G protein beta{gamma} subunits in DPDPE-mediated MonoMac-1 adhesion, we transfected cells with the betaARK-CT or G{alpha}t, two plasmids encoding a peptide (betaARK-CT) or a protein (G{alpha}t) that associate efficiently with free Gbeta{gamma} dimer, suppressing Gbeta{gamma}-dependent responses. Correct betaARK-CT and G{alpha}t expression was confirmed in Western blot (Fig. 3G). In these conditions, DPDPE did not induce adhesion in betaARK-CT- or G{alpha}t-transfected MonoMac-1 cells (Fig. 3H). The results indicate that Gbeta{gamma} activation is a critical step in DPDPE-induced monocyte adhesion.

Opioids trigger PI3K{gamma}-mediated Rac1 activation

Rho GTPases RhoA, Cdc42, and Rac1 participate in the control of cell adhesion (22). The primary consequence of PI3K activation is PIP3 production, which facilitates Rho GTPase activation (23). Rac and Cdc42 dominant negative mutants block opioid-induced JNK activation (24). To study the link between DPDPE-mediated PI3K activation and cell adhesion, we analyzed DOR-induced activation of Rho GTPases. MonoMac-1 cells were plated on FN before DPDPE stimulation. Cells were lysed, cell extracts precipitated with GST-PAK or GST-C21 bound to Sepharose beads, and Rac1 and RhoA levels, respectively, were measured in Western blot using specific mAb. DPDPE promoted rapid Rac1 activation (Fig. 4A, left), but not that of RhoA (Fig. 4B, left). As a control, an aliquot of unprecipitated cell extract was probed on a separate membrane with anti-Rac1 or anti-RhoA mAb (Fig. 4, A and B, right).


Figure 4
View larger version (38K):
[in this window]
[in a new window]
  		FIGURE 4. DPDPE triggers PI3K-mediated Rac1 activation. A, Serum-starved MonoMac-1 cells were DPDPE-treated at the times indicated. Cell lysates were precipitated with GST-PAK and analyzed in Western blot with anti-Rac1 mAb (left). Protein loading was confirmed by developing total Rac in lysates on a separate membrane (right). B, Cells were activated as in A, lysates precipitated with GST-C21 and analyzed in Western blot with anti-RhoA mAb (left). Protein loading was confirmed by developing total RhoA in lysates (right). C, LY294002-treated (top) and PTX-treated (bottom) MonoMac-1 cells were treated and lysed as in A. Lysates were precipitated with GST-PAK and analyzed in Western blot with anti-Rac1 mAb (left). Loading was controlled by developing total Rac in lysates (right).

 
To define the relationship between PI3K activity and Rac1 activation, we performed a similar experiment in LY294002-pretreated cells, and found no DPDPE-mediated Rac1 activation (Fig. 4C, top left). DPDPE nonetheless activated Rac1 in PTX-treated MonoMac-1 cells (Fig. 4C, bottom left), indicating that PI3K, but not Gi, is required for opioid-mediated Rac1 activation. As G{alpha} subunits are unable to activate PI3K{gamma} (21), the results show that DPDPE induces rapid PI3K{gamma} activation through Gbeta{gamma}. Under conditions in which Gi is blocked, G{alpha}z binds the receptor ensuring Gbeta{gamma} availability, which facilitates PI3K{gamma} and subsequent Rac1 activation.

Opioid-mediated Rac1 activation requires Src and Vav activity

Vav-1 is a guanine nucleotide exchange factor (GEF) for Rho family proteins; it is expressed predominantly by cells of hemopoietic origin, and has GEF activity mainly on Rac1. Vav is phosphorylated on tyrosine residues in response to TCR stimulation, cytokines, growth factors, or chemokines. Through this mechanism, and facilitated in part by PI3K activation products, Vav Src homology 2 domains mediate association with the tyrosine kinase receptor itself or with cytoplasmic tyrosine kinases such as lyn or Jak (25).

We evaluated Vav-1 activation in DPDPE-stimulated monocytes. We lysed unstimulated and DPDPE-stimulated MonoMac-1 cells, immunoprecipitated cell extracts using anti-phospho-Tyr, and blotted with anti-Vav-1 Ab. The results showed rapid Vav-1 phosphorylation (Fig. 5A). Protein loading was controlled with a protein detection kit (Pierce). Although opioid receptors have no intrinsic tyrosine kinase activity, they activate kinases such as Src and the focal adhesion kinase (FAK) (26). To analyze the mechanism involved in DPDPE-mediated Vav-1 activation, we lysed unstimulated and DPDPE-stimulated MonoMac-1 cells, immunoprecipitated cell extracts with anti-DOR Ab, followed by Western blot with anti-Src Ab (Fig. 5B), and observed DPDPE-mediated Src association to DOR. This association coincided with DPDPE-mediated Src activation kinetics, as indicated by its presence in phospho-Tyr precipitates of DPDPE-stimulated MonoMac-1 cells (data not shown). Our findings thus suggest that Src kinases participate in DPDPE-mediated Vav-1 activation.


Figure 5
View larger version (56K):
[in this window]
[in a new window]
  		FIGURE 5. DPDPE triggers Src and Vav-1 activation. A, Serum-starved MonoMac-1 cells were incubated with 10–7 M DPDPE (37°C). Cell lysates were immunoprecipitated with anti-phospho-Tyr mAb and Western blot developed with anti-Vav-1 mAb. Protein loading was controlled with a protein detection kit. B, Serum-starved MonoMac-1 cells, treated and lysed as in A, were immunoprecipitated with anti-DOR Ab and analyzed in Western blot with anti-Src mAb. As control, an unstimulated, unprecipitated MonoMac-1 cell lysate was analyzed in Western blot with anti-Src mAb. C, Serum-starved MonoMac-1 cells were incubated with 10–7 M DPDPE (37°C). Cell lysates were analyzed in Western blot with anti-phospho-JNK mAb (left). We controlled protein loading by developing the membrane with anti-JNK Ab (right). D, Untreated (left) or PTX-treated (right) MonoMac-1 cells were incubated with 10–7 M DPDPE (37°C). Cell lysates were analyzed in Western blot with anti-phospho-Erk1/2 mAb (top). To control protein loading, the membrane was developed with anti-Erk 1/2 Ab (bottom).

 
As {kappa} opioid receptor signals through Src and FAK to stimulate JNK kinases in human monocytic cells (26), we evaluated JNK activity and found that DPDPE stimulation did not trigger JNK phosphorylation in this system (Fig. 5C). As control, the membrane was reprobed with anti-JNK Ab (Fig. 5C). Through Gi and Gz, DPDPE activates the MAPK ERK1/2, a response that is abolished by chronic opioid treatment in vitro (27). We lysed unstimulated and DPDPE-stimulated MonoMac-1 cells and blotted cell extracts with anti-phospho-ERK1/2 mAb. DPDPE mediated ERK1/2 activation (Fig. 5D, top left), which was not abolished by PTX treatment (Fig. 5D, top right). The data suggest Gi or Gz participation in DOR-mediated ERK1/2 activation in MonoMac-1 cells. To control protein loading, membranes were reprobed with anti-ERK1/2 mAb (Fig. 5D).

To evaluate the role of the Src/Vav-1/Rac1 axis in DPDPE-triggered cell adhesion, we pretreated MonoMac-1 cells with the Src inhibitor PP2, and measured adhesion. PP2 completely abrogated DPDPE-mediated MonoMac-1 adhesion (Fig. 6A). As control, treatment with a nonfunctional PP2 analog, PP3, did not alter DPDPE-induced cell adhesion. We thus tested Rac1 activation by DPDPE in PP2-pretreated cells. MonoMac-1 cells were plated on FN and PP2 pretreated before DPDPE stimulation. After pull-down with Sepharose-bound GST-PAK, we measured Rac1 levels in Western blot using specific mAb. DPDPE did not promote Rac1 activation (Fig. 6B, top). As control, an aliquot of unprecipitated cell extract was probed on a separate membrane with anti-Rac1 mAb (Fig. 6B) and PP3 treatment did not alter DPDPE-mediated Rac1 activation (Fig. 6B, bottom). MonoMac-1 cells were transiently transfected with Vav-1 or Rac1 siRNA probes, and DPDPE-mediated cell adhesion evaluated. Although cells transfected with control siRNA showed a notable DPDPE-promoted increase in adhesion to FN, DPDPE did not trigger adhesion in MonoMac-1 cells lacking Vav-1 or Rac1 (Fig. 6C). To confirm Vav-1 and Rac1 involvement, siRNA for these proteins with a single-base substitution were transfected in MonoMac-1 cells, and transfectants tested in FN adhesion assays. DPDPE-induced cell adhesion was unaffected in these transfectants (Fig. 6C). Western blot confirmed Vav-1 and Rac1 expression levels in the transfected cells (Fig. 6D). The results confirm the role of the Vav-1/Rac1 pathway in DPDPE-mediated MonoMac-1 adhesion, and suggest a mechanism involving G protein, PI3K{gamma}, and Src/Vav-1/Rac1 activation.


Figure 6
View larger version (33K):
[in this window]
[in a new window]
  		FIGURE 6. Src/Vav-1/Rac1 blockade disrupts DPDPE-mediated MonoMac-1 cell adhesion. A, Static adhesion assay of untreated or PP2-treated MonoMac-1 cells on FN, alone or with DPDPE. The PP2 solvent (DMSO) and PP3 were used as control. Results are expressed as in Fig. 1C. The mean ± SD of four independent experiments is shown. **, p < 0.01 was calculated by Student’s two-tailed t test. B, PP2-treated (top) and PP3-treated (bottom) MonoMac-1 cells were treated with DPDPE at the times indicated. Cell lysates were precipitated with GST-PAK and analyzed in Western blot with anti-Rac1 mAb (left). Loading was controlled by developing total Rac in lysates (right). C, Static adhesion assay of MonoMac-1 cells transiently transfected with Rac1 or Vav-1 siRNA on FN, alone or with DPDPE. As control, Rac1-mismatched (Rac1-M) or Vav-1-mismatched (Vav-1-M) probes were included (see Materials and Methods). Results are expressed as in Fig. 1C. The mean ± SD of three independent experiments is shown. **, p < 0.01; ***, p < 0.001 determined by Student’s two-tailed t test. D, MonoMac-1 cells transiently transfected as in C were lysed and analyzed in Western blot with anti-Vav-1 (top left) or anti-Rac1 mAb (top right). Protein loading was confirmed by reprobing membranes with anti-beta-actin mAb (bottom). E, Serum-starved MonoMac-1 cells were incubated with 10–7 M DPDPE (37°C). Cell lysates were immunoprecipitated with anti-DOR Ab and Western blot developed with anti-paxillin mAb (top). We confirmed protein loading as described, using anti-DOR Ab (bottom). F, Serum-starved MonoMac-1 cells as in E were lysed and immunoprecipitated with anti-phospho-Tyr mAb, and the Western blot developed with anti-paxillin mAb. Protein loading was controlled using a protein detection kit. The arrow shows the position of paxillin. G, Serum-starved MonoMac-1 cells, untreated (left) or treated with LY294001 (right), were incubated with 10–7 M DPDPE (37°C). Cell lysates were analyzed in Western blot with anti-phospho-cofilin mAb (top). We controlled protein loading as in D (bottom).

 
Adhesion sites are associated to an array of actin cytoskeleton-linked cytoplasmic proteins, including FAK and paxillin, which are critical for actin cytoskeleton remodeling (28) and might thus be involved in opioid-induced adhesion responses. We lysed untreated or DPDPE-treated MonoMac-1 cells and precipitated extracts with anti-DOR Ab. Western blot analysis showed paxillin association to the receptor, starting at 3 min poststimulation (Fig. 6E). To control protein loading, we reprobed membranes with the precipitating mAb. As paxillin, a downstream target of FAK/Src (29), is activated by tyrosine phosphorylation, we analyzed paxillin levels in phospho-Tyr immunoprecipitates of MonoMac-1 cells treated as described. DPDPE-triggered paxillin tyrosine phosphorylation kinetics coincided with that of paxillin association to DOR (Fig. 6F). As a positive control, CXCL12-stimulated MonoMac-1 cells were lysed, immunoprecipitated with anti-phospho-Tyr Ab and analyzed in Western blot with anti-paxillin Ab (Fig. 6F).

Actin assembly and disassembly are essential for cell movement, and cofilin is a major Ca2+-independent regulator of these processes (30). Cofilin is an actin-depolymerizing protein; its inactivation by serine phosphorylation permits actin polymerization (31). We lysed untreated and DPDPE-treated MonoMac-1 cells and measured phosphorylated cofilin in Western blot using anti-phospho-cofilin mAb. DPDPE triggered sustained cofilin phosphorylation, starting at 5 min and maintained at 30 min poststimulation (Fig. 6G, left). To control protein loading, the membrane was reprobed with anti-beta-actin mAb. To confirm the dependence of this pathway on PI3K activity, we tested DPDPE-mediated cofilin phosphorylation in LY294002-pretreated MonoMac-1 cells. DPDPE did not trigger cofilin phosphorylation (Fig. 6G, right). Protein loading was confirmed as previously described.

In situ analysis of DPDPE-treated human primary monocytes in mouse cremaster microcirculation

To validate these observations in a physiological setting, we studied the in vivo effect of DPDPE on monocyte homing. Human primary monocytes were isolated from healthy donors, and DOR expression was confirmed by flow cytometry (data not shown). We found that DPDPE triggered human primary monocyte adhesion on FN (Fig. 7A, left), which was unaffected by PTX treatment, and was reduced in LY294002-treated cells (Fig. 7A, right). As for MonoMac-1 cells, treatment with neutralizing anti-beta1 or anti-{alpha}5beta1 integrin mAb inhibited the DPDPE-triggered effect (Fig. 7A, right). In primary human monocytes, DPDPE thus also triggers {alpha}5beta1 integrin-mediated adhesion to FN, independent of Gi protein activation.


Figure 7
View larger version (21K):
[in this window]
[in a new window]
  		FIGURE 7. DPDPE triggers {alpha}5beta1 integrin-mediated primary monocyte adhesion. A, Static adhesion assay of primary monocytes, untreated or treated with PTX, LY294001, or neutralizing anti-{alpha}5beta1 or anti-beta1 mAb on FN, alone or with DPDPE (37°C). Results are expressed as in Fig. 1C. The mean ± SD of five independent experiments is shown. As control, CXCL12-mediated MonoMac-1 adhesion to FN was tested. *, p < 0.05; **, p < 0.01 calculated using Student’s two-tailed t test. B, DPDPE-stimulated calcein-AM-labeled of untreated and naltrindole-pretreated primary human monocytes were observed directly in cremaster muscle microcirculation (see Materials and Methods). The percentage of the rolling fraction of monocytes in cremaster arterioles is shown before (none) and after DPDPE stimulation. **, p < 0.01 determined by paired Student’s t test.

 
Human primary monocytes were injected into mice, and their behavior in cremaster microcirculation analyzed by intravital microscopy. Calcein-AM-labeled human primary monocytes were injected into the femoral artery of anesthetized mice, permitting immediate real-time analysis of monocytes in situ. Under basal conditions, circulating monocytes showed a low interaction ratio with the microvascular endothelium of the cremaster muscle (21.46 ± 2.54%). Exogenous addition of DPDPE elicited monocyte rolling (48.35 ± 3.6%) (Fig. 7B). This effect was specific, as it was abolished by naltrindole pretreatment of cells before injection (Fig. 7B). These data confirm the in vitro findings, indicating that DPDPE also modifies monocyte mobility in vivo, inducing slow rolling that may in turn affect the extravasation capacity of these cells.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
Polymorphonuclear cells recruited in early inflammation stages secrete opioids that bind to receptors on peripheral sensory neurons and mediate antinociception (32), and on immune system cells such as monocytes and lymphocytes (1), where they contribute to immune response modulation. Recent evidence shows cross-desensitization between opioid and chemokine receptors (10, 11), suggesting that opioid receptors can participate in modulating leukocyte trafficking.

We show that agonist binding to DOR triggers monocyte adhesion in a process involving {alpha}5beta1 integrin activation. Although these cells express beta1, beta2, and beta3 integrins, anti-beta2 treatment had no effect on cell adhesion to FN, as beta2 integrins do not bind FN (33). Anti-beta3 Ab treatment did not modify DPDPE-induced cell adhesion, possibly due to low beta3 integrin expression in these cells. Monocytic cell lines and isolated primary monocytes adhere firmly to FN after DPDPE activation. The process involves rapid DOR-mediated PI3K activation because adhesion was abrogated by specific PI3K inhibition using wortmannin or LY294002. Opioid receptors interact with Gi proteins to regulate a wide range of effectors such as adenylyl cyclase (5). We found that acute stimulation of opioid receptors in MonoMac-1 cells leads to PTX-sensitive inhibition of adenylyl cyclase activity, as described (27). DPDPE-mediated monocyte adhesion was unaffected by PTX treatment, indicating that adhesion also takes place in the absence of Gi activation. We observed DPDPE-mediated G{alpha}z association to DOR in PTX-treated cells, concurring with reports that DOR activates both Gi and Gz proteins (5). Acute activation of opioid receptors in Gz-transfected HEK293 cells resulted in PTX-insensitive adenylyl cyclase hypersensitization and induction of ERK1/2 phosphorylation (27).

G proteins signal through their {alpha} and beta{gamma} subunits; although G{alpha} subunits are specific for each G protein, small differences are reported among the beta{gamma} complexes (21). We found that even in the presence of PTX, DPDPE activates PI3K{gamma}, the isoform that associates GPCR specifically and is activated by Gbeta{gamma} subunits. It is thus possible that, in the absence of Gi activation, Gz may serve as a source of Gbeta{gamma} complexes. We observed that DPDPE-mediated monocyte adhesion was abrogated in cells transfected with betaARK-CT or G{alpha}t, two molecules that abolish Gbeta{gamma}-dependent responses by associating the free subunits (34). Acute morphine treatment activates the PI3K{gamma}/phospholipase C{gamma}1 pathway directly (35) and, unlike DOR, µ opioid receptor requires PI3K activity to activate JNK (18). This process involves Gbeta{gamma}, Src kinase, Rac, and Cdc42, and is abolished in cells treated with wortmannin or transfected with a PI3K{gamma} dominant negative mutant. We found that DPDPE triggers not only PI3K{gamma} activation, but also association of its catalytic subunit, p110{gamma}, to DOR.

PI3K lipid kinase activity increases PIP3 in specific cell membrane areas, facilitating recruitment of other signaling molecules (20); these include the GEF proteins that promote GDP-GTP exchange in Rho GTPases, which are required to regulate actin cytoskeleton rearrangement and cell adhesion (36). Some evidence suggests that small GTPases translocate to the plasma membrane to activate downstream effectors (37). Vav-1, a GEF that acts predominantly on Rac, has an N-terminal PH region that binds PIP3; DOR activation of PI3K{gamma} may thus facilitate Vav-1 recruitment to the membrane. Vav-1, a substrate for Src-related tyrosine kinases, is activated by tyrosine phosphorylation (38). We also observed DPDPE-mediated Src kinase activation. Our data suggest that DPDPE binding to DOR links Src kinase activation, Vav-1 phosphorylation, and Rac1 activity. This possibility was confirmed by the absence of DPDPE-mediated Rac1 activation in PP2-pretreated cells. siRNA interference with Vav-1 abolished DPDPE-mediated cell adhesion; this result was not anticipated, as there is a degree of redundancy among the three Vav proteins (39). This result may be due to differences in the ability of these proteins to activate small GTPases, as Vav-1 is reported as a GEF primarily for Rac (40), and Vav-2 and Vav-3 act on Rho (41, 42). DPDPE-mediated Rac1 activation has rapid effects on signaling. Opioid receptors require Ras GTPases to activate ERK via Raf and MEK1/2, and the pathway from DOR to JNK requires Rac or Cdc42 (23). DOR activation of JNK and ERK is a well-characterized process (3, 24). Stimulation of JNK activity is a slow, transient process compared with opioid-induced ERK stimulation (3). The pathway connecting DOR with JNK involves Gbeta{gamma} subunits, Src kinases, and the Rho GTPases Rac or Cdc42, and is independent of PI3K activity (24), as shown using neuroblastoma X glioma hybrid NG108-15 cells and transfected COS-7 cells. ERK activation is a Gi-dependent process involving PI3K{gamma}, Src, and Ras (3). Although DOR activates Gbeta{gamma}, PI3K{gamma}, Src, and Rac in MonoMac-1 cells, we found that DPDPE does not activate JNK, but triggers ERK1/2 activation. Additional experiments are under way to clarify this difference. Nonetheless, cross-talk between GPCR and receptor tyrosine kinases is an incredibly complex process, and the specific signaling molecules involved depend largely on cell type and the type of receptor that is activated (43).

Our data are compatible with a model in which DPDPE activates G{alpha}i, G{alpha}z, and their Gbeta{gamma} subunits through DOR. Certain functions such as inhibition of adenylyl cyclase activity require several factors, including G{alpha}I activation, but others such as cell adhesion require Gbeta{gamma} complex activity, and take place only after Gi or Gz activation. Gbeta{gamma} complexes activate PI3K{gamma}, increasing membrane PIP3 levels. This result might facilitate recruitment of Vav-1, which is then phosphorylated in tyrosines, probably by DOR-activated Src. Rac1, the key regulatory molecule for actin cytoskeleton reorganization, is consequently activated and cells adhere firmly. In the case of DPDPE, the presence of PIP3 alone at the cell membrane is insufficient to promote adhesion, as DPDPE activation is also required in PI3K{gamma}-CAAX cells. PIP3 at the surface of PI3K{gamma}-CAAX-transfected cells facilitates Vav-1 recruitment; nonetheless, Vav-1 must be activated in a process that requires ligand-mediated Src activation. We thus hypothesize that DPDPE activation of Vav-1 is the limiting factor in this signaling cascade, which would explain why PI3K{gamma}-CAAX-transfected cell adhesion is a ligand-mediated and ligand-restricted process. In support of this proposal, we observed that Src inhibitors blocked this adhesion process.

Paxillin participates in actin cytoskeleton remodeling at adhesion sites. We observed sustained, DPDPE-mediated paxillin activation, which correlates with the strong cell adhesion and the high shear stress resistance promoted by DPDPE. DPDPE also triggers serine phosphorylation of cofilin, an actin-depolymerizing protein (31). In some models, cofilin and the kinase LIM2 connect Rho GTPase activation with actin cytoskeleton reorganization (31). DPDPE-mediated inactivation of cofilin is a PI3K-dependent process, as it is abrogated by LY294002, confirming a role for PI3K in DPDPE-triggered monocyte adhesion.

Opioids secreted by leukocytes recruited in response to stress could control inflammatory pain by activating receptors in peripheral sensory nerves. Morphine administration alters leukocyte/endothelial cell interactions via stimulation of NO production (15). NO also attenuates leukocyte adherence to endothelium by down-regulating endothelial cell adhesion molecule expression (44). Our data indicate that DPDPE increased monocyte adhesion, without altering expression of surface adhesion molecules on these cells, and endothelial cells were not used in our in vitro experiments. Cross-talk was recently reported between NO via its major cytosolic receptor and the Src/Fyn tyrosine kinase signaling pathways (45). As we found that opioid-mediated MonoMac-1 adhesion involves Src, Vav-1, and Rac1, NO-mediated Src recruitment may interfere with the DPDPE signaling pathway, and thus with cell adhesion.

Opioids also modulate cytokine and chemokine production and trigger cross-desensitization with chemokine ligands. We show that by triggering cell adhesion, opioids may directly affect many of the complex interactions required for efficient immune cell function. Migration of immune system cells depends largely on their adhesive interactions and recognition of chemoattractant gradients; breakdown in this control contributes to immune response dysfunction. Indeed, anti-inflammatory effects of opioids have been reported in rheumatoid arthritis, based on their ability to alter cell adhesion and trafficking and to modulate expression of certain inflammatory mediators (46). Our intravital microscopy results show opioid effects on monocytes in an in vivo model. We observe DPDPE-mediated monocyte rolling rather than cell adhesion. Our in vitro results are nonetheless compatible with a DPDPE-mediated increase in leukocyte adhesion to the endothelium. In vitro and in vivo analysis conditions, however, differ; FN is not found on the endothelial lumen in contact with blood, and the FN-mediated adhesion observed in vitro may thus not take place in vivo. Although rolling usually requires selectins rather than integrins, integrins can also contribute to leukocyte rolling (47, 48), and DPDPE-mediated integrin activation could participate in in vivo monocyte rolling. Nonetheless, we cannot rule out a DPDPE effect on selectin function.

Adhesion mechanisms are involved not only in immune responses, but also participate in pain control. Interruption of cell rolling by selectin blockade attenuates intrinsic opioid analgesia (49), and blockade of ICAM-1 on vascular endothelium induces a substantial decrease in migration of opioid-bearing leukocytes to inflamed tissue (50). Cell adhesion also has an important role in tumor development, as metastasis formation requires tumor cell contact with surrounding cells and matrix elements (51).

Our findings concur with those of others (7, 50), supporting powerful opioid action in the periphery via specific receptors. By promoting cell adhesion and/or rolling, opioids might modulate various physiopathological processes such as immune response, pain sensation, or tumorigenesis. Opioids that act in the periphery should thus be considered regulatory units and potential targets for therapeutic intervention.


   Acknowledgments
 
We thank Dr. M. Wymann (University of Fribourg, Fribourg, Switzerland) for pcDNA3-PI3K{gamma}-KR and PI3K{gamma}-CAAX vectors, Dr. J. L. Rodríguez (Centro de Investigaciones Biológicas, Madrid, Spain) for pcDNA3-betaARK-CT and G{alpha}t vectors, Dr. A. C. Carrera and Dr. D. F. Barber for helpful discussion, M. C. Moreno-Ortíz for flow cytometry analysis, L. Gómez for expert animal care, and C. Bastos and C. Mark for secretarial and editorial assistance, respectively.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by grants from the Lilly Foundation, Comunidad de Madrid, the Spanish MEyC, and the European Union. It was also supported by a grant from the Fundación Ramón Areces (to O.M.P.). The Department of Immunology and Oncology, Centro Nacional de Biotecnología, was founded and is supported by Consejo Superior de Investigaciones Científicas (CSIC) and by Pfizer. Back

2 Address correspondence and reprint requests to Dr. Mario Mellado, Department of Immunology and Oncology, Centro Nacional de Biotecnología-CSIC, Darwin 3, Campus de Cantoblanco, E-28049 Madrid, Spain. E-mail address: mmellado{at}cnb.uam.es Back

3 Abbreviations used in this paper: GPCR, G protein-coupled receptor; PTX, pertussis toxin; DOR, {delta} opioid receptor; PIP3, phosphatidylinositol 3,4,5-trisphosphate; FN, fibronectin; FAK, focal adhesion kinase; siRNA, short-interfering RNA; betaARK-CT, beta-adrenergic receptor kinase C terminus. Back

Received for publication July 22, 2005. Accepted for publication November 15, 2005.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Gaveriaux, C., J. Peluso, F. Simonin, J. Laforet, B. Kieffer. 1995. Identification of {kappa}- and {delta}-opioid receptor transcripts in immune cells. FEBS Lett. 369: 272-276. [Medline]
  2. Waldhoer, M., S. E. Bartlett, J. L. Whistler. 2004. Opioid receptors. Annu. Rev. Biochem. 73: 953-990. [Medline]
  3. Belcheva, M. M., Z. Vogel, E. Ignatova, T. Avidor-Reiss, R. Zippel, R. Levy, E. C. Young, J. Bar, C. J. Coscia. 1998. Opioid modulation of extracellular signal-regulated protein kinase activity is Ras-dependent and involves Gbeta{gamma} subunits. J. Neurochem. 70: 635-645. [Medline]
  4. Law, P. Y., Y. H. Wong, H. H. Loh. 2000. Molecular mechanisms and regulation of opioid receptor signaling. Annu. Rev. Pharmacol. Toxicol. 40: 389-430. [Medline]
  5. Standifer, K. M., G. W. Pasternak. 1997. G proteins and opioid receptor-mediated signalling. Cell. Signal. 9: 237-248. [Medline]
  6. Bartlett, S. E., I. A. Hendry. 1997. Identifying the G protein, Gz{alpha}, and its associated proteins in nervous tissue using mass spectrometry and microsequencing techniques. Int. J. Dev. Neurosci. 15: 267-274. [Medline]
  7. Stein, C., M. Schafer, H. Machelska. 2003. Attacking pain at its source: new perspectives on opioids. Nat. Med. 9: 1003-1008. [Medline]
  8. Sheridan, P. A., J. A. Moynihan. 2005. Modulation of the innate immune response to HSV-1 following acute administration of morphine: role of hypothalamo-pituitary-adrenal axis. J. Neuroimmunol. 158: 145-152. [Medline]
  9. Peng, X., D. M. Mosser, M. W. Adler, T. J. Rogers, J. J. Meissler, Jr, T. K. Eisenstein. 2000. Morphine enhances interleukin-12 and the production of other pro-inflammatory cytokines in mouse peritoneal macrophages. J. Leukocyte Biol. 68: 723-728. [Abstract/Free Full Text]
  10. Suzuki, S., L. F. Chuang, P. Yau, R. H. Doi, R. Y. Chuang. 2002. Interactions of opioid and chemokine receptors: oligomerization of µ, {kappa}, and {delta} with CCR5 on immune cells. Exp. Cell Res. 280: 192-200. [Medline]
  11. Steele, A. D., I. Szabo, F. Bednar, T. J. Rogers. 2002. Interactions between opioid and chemokine receptors: heterologous desensitization. Cytokine Growth Factor Rev. 13: 209-222. [Medline]
  12. Mellado, M., J. M. Rodríguez-Frade, A. J. Vila-Coro, S. Fernández, A. Martín de Ana, D. R. Jones, J. L. Torán, C. Martínez-A. 2001. Chemokine receptor homo- or heterodimerization activates distinct signaling pathways. EMBO J. 20: 2497-2507. [Medline]
  13. Deroanne, C., V. Vouret-Craviari, B. Wang, J. Pouysségur. 2003. EphrinA1 inactivates integrin-mediated vascular smooth muscle cell spreading via the Rac/PAK pathway. J. Cell Sci. 116: 1367-1376. [Abstract/Free Full Text]
  14. Coxon, A., P. Rieu, F. J. Barkalow, S. Askari, A. H. Sharpe, U. H. von Andrian, M. A. Arnaout, T. N. Mayadas. 1996. A novel role for the beta2 integrin CD11b/CD18 in neutrophil apoptosis: a homeostatic mechanism in inflammation. Immunity 5: 653-666. [Medline]
  15. Ni, X., K. R. Gritman, T. K. Eisenstein, M. W. Adler, K. E. Arfors, R. F. Tuma. 2000. Morphine attenuates leukocyte/endothelial interactions. Microvasc. Res. 60: 121-130. [Medline]
  16. Quock, R. M., T. H. Burkey, E. Varga, Y. Hosohata, K. Hosohata, S. M. Cowell, C. A. Slate, F. J. Ehlert, W. R. Roeske, H. I. Yamamura. 1999. The {delta}-opioid receptor: molecular pharmacology, signal transduction, and the determination of drug efficacy. Pharmacol. Rev. 51: 503-532. [Abstract/Free Full Text]
  17. Arnaout, M. A.. 1990. Leukocyte adhesion molecules deficiency: its structural basis, pathophysiology and implications for modulating the inflammatory response. Immunol. Rev. 114: 145-180. [Medline]
  18. Kam, A. Y., A. S. Chan, Y. H. Wong. 2004. Phosphatidylinositol-3 kinase is distinctively required for µ-, but not {kappa}-opioid receptor-induced activation of c-Jun N-terminal kinase. J. Neurochem. 89: 391-402. [Medline]
  19. Hirsch, E., V. L. Katanaev, C. Garlanda, O. Azzolino, L. Pirola, L. Silengo, S. Sozzani, A. Mantovani, F. Altruda, M. P. Wymann. 2000. Central role for G protein-coupled phosphoinositide 3-kinase {gamma} in inflammation. Science 287: 1049-1053. [Abstract/Free Full Text]
  20. Ward, S.. 2004. Do phosphoinositide 3/kinases direct lymphocyte navigation?. Trends Immunol. 25: 67-74. [Medline]
  21. Kerchner, K. R., R. L. Clay, G. McCleery, N. Watson, W. E. McIntire, C. S. Myung, J. C. Garrison. 2004. Differential sensitivity of phosphatidylinositol 3-kinase p110{gamma} to isoforms of G protein beta{gamma} dimers. J. Biol. Chem. 279: 44554-44562. [Abstract/Free Full Text]
  22. Vicente-Manzanares, M., F. Sánchez-Madrid. 2004. Role of the cytoskeleton during leukocyte responses. Nat. Rev. Immunol. 4: 110-122. [Medline]
  23. Vivanco, I., C. L. Sawyers. 2002. The phosphatidylinositol 3-kinase AKT pathway in human cancer. Nat. Rev. Cancer 2: 489-501. [Medline]
  24. Kam, A. Y., A. S. Chan, Y. H. Wong. 2003. Rac and Cdc42-dependent regulation of c-Jun N-terminal kinases by the {delta}-opioid receptor. J. Neurochem. 84: 503-513. [Medline]
  25. Turner, M., D. D. Billadeau. 2002. VAV proteins as signal integrators for multi-subunit immune-recognition receptors. Nat. Rev. Immunol. 2: 476-486. [Medline]
  26. Kam, A. Y., A. S. Chan, Y. H. Wong. 2004. {kappa}-opioid receptor signals through Src and focal adhesion kinase to stimulate c-Jun N-terminal kinases in transfected COS-7 cells and human monocytic THP-1 cells. J. Pharmacol. Exp. Ther. 310: 301-310. [Abstract/Free Full Text]
  27. Tso, P. H., L. Y. Yung, Y. H. Wong. 2000. Regulation of adenylyl cyclase, ERK1/2, and CREB by Gz following acute and chronic activation of the {delta}-opioid receptor. J. Neurochem. 74: 1685-1693. [Medline]
  28. Giancotti, F. G.. 2000. Complexity and specificity of integrin signaling. Nat. Cell Biol. 2: E13-E14. [Medline]
  29. Turner, C. E.. 2000. Paxillin interactions. J. Cell Sci. 113: 4139-4140. [Abstract/Free Full Text]
  30. Bamburg, J. R., O. P. Wiggan. 2002. ADF/cofilin and actin dynamics in disease. Trends Cell Biol. 12: 598-605. [Medline]