In leukocyte trafficking from the blood to peripheral tissues during normal immune surveillance and host defense, integrin activation should be spatially and temporally regulated (42
). Like many other cellular proteins, integrins are not fixed in a particular conformation. Instead, they reversibly equilibrate between the bent, low affinity and the extended, high affinity conformation (43
). Recently, dramatic advances in defining the structure of integrins by crystallography, nuclear magnetic resonance, and electron microscopy have demonstrated important molecular mechanisms for integrin activation and ligand binding, and provided useful information in drug discovery. Despite recent progress in defining the structure of integrins, a key question that remains unanswered is whether the conformational changes in integrins actually occur in living cells during migration, and if so, how they are regulated. Based primarily on studies with monoclonal antibodies that preferentially bind to active integrins, it was hypothesized that high affinity integrins are localized at the leading edge of spreading and migrating cells (45
). In this study, we have been able to directly test this hypothesis in live cells using a fluorescent VLA-4 peptide inhibitor and an activation dependent mAb in human T cells, and our novel TIRF/FRET-based assay for VLA-4 activation in GD25 cells.
As FRET is the only biological assay that measures the dynamics of protein structure in living cells, our novel FRET system in which to study the dynamic regulation of VLA-4 activation in living cells provides a unique opportunity to understand a fundamental aspect of inflammatory-based pathology. However, FRET analysis of VLA-4 in primary human T cells is not feasible because of the expression of endogenous α4 and β1 integrin subunits. Therefore fluorescence microscopy using FITC labeled BIO1211 and LIBS Ab B44, and FRET microscopy using VLA-4 FRET sensor should be considered complementary techniques. TIRF microscopy is advantageous in studying the molecular mechanisms involved in cell migration, since the technique allows us to monitor optical events within 100 nm of the plasma membrane, where cell-substrate traction occurs. Thus, by combining FRET with TIRF microscopy we were able to investigate the dynamic activation of VLA-4 at the cell-substrate interface during cell migration on VCAM-1.
The data presented here demonstrate that the activated form of VLA-4 is dynamically distributed to the leading edge of migrating cells and the spatial distribution of active VLA-4 is critical for T cell migration on VCAM-1. The low FRET signal of the VLA-4 FRET sensor at the leading edge indicates the separation of the α and β cytoplasmic domains of VLA-4. During lamellipodial protrusion at the leading edge, VLA-4 is activated at the extending front of the cell and subsequently maintains the active and extended conformation only when it encounters VCAM-1, suggesting that the active form of VLA-4 is induced by intracellular signaling and then further stabilized by ligand binding. However, at the inner region of the cell contact zone and more distal region from the leading edge, VLA-4 returns to a low affinity state, as it exhibits a lack of LIBS Ab staining on human T cells and higher FRET signals on GD25 ells. Consistent with our study, α4
subunit phosphorylation by protein kinase A (PKA) inhibits paxillin binding and paxillin-free phosphorylated α4
is mainly localized at the leading edge and promotes cell migration (5
). Recent evidence demonstrates that AKAP-Lbc, a specific A-kinase anchoring protein, is critical for generating a PKA activity gradient at the leading edge (46
), although regulation of AKAP-Lbc distribution and PKA activity gradients by chemokine stimulation is not known. Interestingly, the α4
subunit cytoplasmic domain can function as a Type I PKA-specific AKAP and the association is critical for α4
phosphorylation and persistent directional cell migration (47
). Our data indicates that chemokine receptor CXCR4 and Rap1 activation are restricted to the front of migrating T cells. Active Rap1 has been shown to interact with RIAM (Rap1-GTP-interacting molecule), and this Rap1-RIAM complex can target talin to the plasma membrane to activate integrins (48
). Therefore it is likely that, when the α4
subunit of VLA-4 is phosphorylated by PKA at the leading edge, VLA-4 switches to an intermediated-affinity state. After VCAM-1 binding, the β1
subunit of this intermediate affinity VLA-4 could again associate with Rap1/RIAM/talin complex and interact with the cytoskeleton. The interaction will then reinforce the interaction of VLA-4 and VCAM-1, thus stabilizing the high affinity state of VLA-4 at the leading edge and providing traction force for cell migration.
Our studies on VLA-4 do not define the mechanisms of active redistribution of other integrins during T cell migration; nonetheless, evidence indicates that on a given cell, activation of one integrin subset may be regulated by activation or ligation of another, allowing flexibility for specific leukocyte adhesion in different microenvironments. In monocytes, ligation of high affinity LFA-1 by ICAM-1 results in decreased binding of VLA-4 to VCAM-1 (49
). Although it is not known whether the negative crosstalk between LFA-1 and VLA-4 occurs in T cells, activated LFA-1 may not simultaneously localize with high affinity VLA-4 during migration. Indeed, unlike VLA-4, the high affinity LFA-1 is excluded from the leading edge and is restricted to the mid-cell focal zone, whereas intermediate affinity LFA-1 is expressed at the leading edge (50
Conventional anti-adhesion therapy is designed for non-selective inhibition of cell surface integrins by complete saturation. This might cause massive suppression of normal immune reactions and increase susceptibility to infections (51
). In addition, relatively high dose Ab administration is often required to produce clear clinical improvement due to the lack of the specificity or because of rapid consumption after systemic administration. As described in , we found that T cell migration on VCAM-1 + CXCL12 is inhibited by BIO1211, a selective antagonist against VLA-4. At a concentration of 4 nM, BIO1211 could bind to VLA-4 only when cells were treated with Mn2+
, suggesting that BIO1211 selectively blocks activated VLA-4 at this concentration. Interestingly, 4 nM BIO1211 could successfully block T cell migration on VCAM-1 and was as potent as 400 nM, a concentration at which BIO1211 binds to both inactive and active VLA-4. These data suggest that selective blocking of subpopulation of activated VLA-4 is sufficient to inhibit T cell migration on VCAM-1, and may prove to be an effective therapy for inflammatory diseases.
Our study also addresses the mechanisms of preferential distribution of active VLA-4 at the leading edge during lymphocyte migration in response to chemokine stimulation. Recent evidence demonstrates that chemokines induce redistribution of chemokine receptors to the leading edge of polarized lymphocytes in a PTX sensitive manner (35
), although its functional relation with dynamic distribution of leukocyte integrin activation is not known. Our results show that chemokine receptor CXCR4 is simultaneously localized at the leading edge with active VLA-4. Pretreatment of cells with PTX abolished the BIO1211 staining at the front of T cells, suggesting that CXCL12-induced activation of endogenous PTX-sensitive Gαi
heteromeric G proteins is critical for the redistribution of active VLA-4. These findings imply that chemokine stimulation can induce endogenous polarization of chemokine receptor and its downstream signals, which may be crucial for the localized activation of VLA-4. The small GTPase Rap1 has emerged as an important regulator of integrin-mediated cell adhesion and migration (37
), especially for VLA-4 activation (53
). During T cell activation, only a small subpopulation of activated intracellular Rap1 shows a restricted distribution at the plasma membrane, while the majority of total Rap1 is localized at the perinuclear endosome (54
). In addition, the dominant active form of Rap1 is redistributed to the leading edge of polarized lymphocytes together with CXCR4 (37
). Our TIRF/FRET analysis with the Rap1 FRET sensor indicates that activated Rap1 is restricted to the plasma membrane of the leading edge during T cell migration on immobilized VCAM-1 and CXCL12. This Rap1 activation pattern is in agreement with a previous report showing that chemokine stimulation triggers Rap1 activation, which was sensitive to PTX, and the activated Rap1 subsequently regulates VLA-4 activation (37
). Interestingly, inhibition of endosomal recycling by a dominant-negative Rap1 binding protein blocked both activation of Rap1 at the plasma membrane of Jurkat T cells and their adhesion to fibronectin or ICAM-1 (54
). This study suggests that activation of Rap1 at the plasma membrane might result from the transport of intracellular vesicles containing the activated GTPase to the cell surface. If this is indeed the case, the transportation of active Rap1 might also guide activated VLA-4 to the leading edge of migrating lymphocytes (55
). Alternatively, activation of VLA-4 at the leading edge might be caused by localized activation of Rap1 complex with RIAM/talin at the region.