The microtubule cytoskeleton participates in control of ββββ 2 integrin avidity

of indicate is the dynamics the is for regulation of 2 integrin dependent adhesion. These results provide new insights into the mechanism of regulation of integrin mobility, and define an unforeseen role for microtubules.


Introduction
The β 2 integrin family of adhesion molecules plays an important role in cell adhesion and cell spreading in the immune system. An early event of cell adhesion is integrin activation (1).
Following adhesion, the assembly of a focal adhesion complex, consisting of many cytoskeletal proteins, is required for cells to spread (2). Many intracellular and extracellular signals regulate integrin activation (3). In the case of β 2 integrins, activation is tightly regulated by a number of intracellular signals, often originating from other receptors. For example, on T cells, adhesion is activated by crosslinking of T cell receptors (4). Activation of integrins by signals originating within the cell is termed "inside-out" signaling (5). Although the mechanism of inside-out signal transduction is not clear, both protein kinase C (PKC) and the cytoskeleton are involved (for review see (3;6;7).
Protein kinase C-mediated effects on the cytoskeleton play a key role in β 2 -integrin activation. In unactivated leukocytes, β 2 integrins are relatively immobile in the membrane because of cytoskeletal constraints. Activation of PKC with phorbol ester releases these constraints, allowing the integrins to diffuse and increasing their probability of encountering ligand (8). Although the exact mechanism of PKC's involvement is not clear, its kinase activity causes a relaxation of cytoskeleton. It has been shown by direct measurement of integrin mobility that not only phorbol esters, but also PKC-activating cytokines (9) stimulate the mobility of β 2 integrin molecules on the membrane. The PKC signal is most likely transduced by MacMARCKS protein (10;11), because either lack of this protein, or a mutation that prevents phosphorylation, prevents both PKC-stimulated molecular mobility of integrin (9) and β 2integrin-mediated cell adhesion (12)(13)(14). The actin cytoskeleton is important to activation of adhesion in at least three ways. First, the restriction of mobility of integrins before activation is mediated by the cytoskeleton. The role of the cytoskeleton in this process was confirmed previously by treating cells with low-dose (0.1-0.3 µg/ml) cytochalasin D to partially depolymerize the actin cytoskeleton. This not only stimulated β 2 integrin mobility, but also activated β 2 -mediated cell adhesion (8). Second, cytoskeletal relaxation results in integrin clustering (15;16). Both a conformational (affinity) change (17;18) and an avidity change (8;16;19;20) occur, and may be complementary in some cases (21;22). β 2 integrin clustering seems to be automatic once the cytoskeletal constraint is removed. Multivalent or clustered ligand may then promote further formation of integrin clusters. Integrin clustering results in "outside-in" signaling, activating signaling pathways within the cell. Third, cytoskeletal proteins participate in activation of adhesion by their involvement in outside-in signaling. In response to integrin clustering, paxillin (23;24) and other proteins (25) are tyrosine-phosphorylated. Subsequently, cytoskeletal proteins are recruited to link the cytoskeleton to integrins, forming focal adhesions. This stabilizes the clusters (26), enhancing cell adhesion and spreading. At this stage, an intact actin cytoskeleton is required for the assembly of focal adhesions, because at high concentrations of cytochalasin D, cells fail to spread. It is interesting that the actin filaments play opposite roles in cell adhesion and cell spreading. While partial disruption of actin results in activation of the β 2 integrin, an intact actin cytoskeleton is needed to establish adhesions for cell spreading.
It has been demonstrated that microtubules are also involved in cell spreading (27)(28)(29).
For example, depolymerization of microtubules with nocodazole causes spreading macrophages to retract quickly into a spherical shape (30). Similarly, in fibroblasts, cell spreading is impaired when microtubules are disassembled (27;28). Indeed, cell spreading closely correlates with by guest on March 23, 2020 http://www.jbc.org/ Downloaded from microtubule integrity, which can influence both the growth of actin stress fibers and the assembly of focal adhesions (32;33). Growing microtubule filaments are targeted toward the focal adhesion site (33), and such microtubule targeting seems to influence the turnover rate of focal adhesions (35). In addition to microtubules themselves, microtubule-associated proteins are also involved in cell spreading. For example, kinesin, the plus-end directed microtubuledependent motor protein, is involved, because anti-kinesin antibody decreases cell spreading when injected into cells (29). Dynamitin, a component of dynein/dynactin motor complex, is also involved in cell spreading (31) via its P62 subunit, which localizes to focal contacts (34). It has not yet been established, however, whether the microtubule cytoskeleton also plays a dual role, as actin does, affecting activation of adhesion and cell spreading in different ways. Such a dual role in leukocytes would be consistent with a recent study that showed differential effects of microtubule disrupting agents on adhesion and spreading in an adenocarcinoma cell line (36).
In this report, we investigated the roles of both microtubule disruption and stabilization on β 2 integrin mobility and its consequences for leukocyte adhesion. We used Single Particle Tracking, in which receptor movement was visualized by conjugating small particles  nm) to the integrins (8;37-41) and positions of single molecules were determined with nanometer-level spatial resolution and 30-ms time resolution (42;43). Restricted thermal motion of the integrins is then used as a readout for cytoskeletal connections (8,9,38). We showed that depolymerization of microtubules enhanced mobility of the β 2 integrin molecules, a very early step in β 2 integrin activation. Our data clearly indicate that although depolymerization of microtubules inhibits cell spreading, it releases β 2 integrins to diffuse freely, resulting in activation of cell adhesion. This is associated with increased tyrosine phosphorylation of paxillin, a biochemical marker of outside-in signaling (14;23). Further, similar effects of   Because the latex microbeads used here have autofluorescence, at the end of the experiments the beads were illuminated for their fluorescence so that they would not be confused with the subcellular organelles (9).

Video microscopy
The detailed microscopic procedure was previous described (9). Briefly, cells were plated on poly-L-lysine-coated coverslips. In some experiments, cells were incubated in 15 µg/ml C3 exoenzyme for 24 hours as previously described (53). Coverslips were then mounted in a steel chamber and placed on a water-jacketed heating stage on a Zeiss Axiovert 100 microscope. After coated with BSA or antibody against MHC II were used as controls as described previously (9).
Bead motion was observed using video-enhanced differential interference contrast microscopy.

Data analysis
The xy coordinates of a particle at any given time point were automatically recorded from the digitized image using a in-house written software created with Visual Basic on Window 95 platform. For each track of a bead, the mean square displacement (MSD) for each time interval was calculated from the xy coordinates of the particles according to the formulas [1], [2] and [3] (for review see (43)) using the same software # : The obtained MSD is the sum of the random and directed motion: MSD total = MSD random + MSD directed [4] because MSD random = 4Dt [5] MSD directed = (νt) 2 [6] thus, MSD total = 4Dt + (νt) 2 [7] By fitting the MSD calculated from the experimental data to the quadratic equation [7], the diffusion coefficient D can be extracted.

Colchicine stimulates β 2 integrin mobility on MP EBV-transformed B cells
The actin cytoskeleton is required for the cytoskeletal constraint on the molecular mobility of β 2 integrin, and, thus, regulation of β 2 integrin mediated adhesion (8;9). Here we integrin mobility (p < 0.01). This confirms that, in the MP cell system, microtubules also play a role in control of integrin mobility in these cells.

Nocodazole and taxol stimulate the lateral mobility of β 2 integrin on macrophage cell lines in a concentration-dependent manner
The molecular basis of the cytoskeletal constraints on β 2 lateral mobility has been shown recently to involve MacMARCKS (9). This was demonstrated in both Wehi and J774 macrophage cell lines. Since MacMARCKS binds to dynamitin, a microtubule-associated protein (31), we asked whether microtubules were also necessary for restriction of lateral mobility in the macrophage system. We measured the effect of microtubule depolymerization on β 2 integrins in J774 cells. Further, to confirm that the effects on integrin mobility (as shown in figure 1) were due to microtubule disruption, rather than a nonspecific effect of colchicine, we used nocodazole, another microtubule-depolymerizing reagent (47), for these experiments. When nocodazole was added to J774 macrophage, we observed a concentration-dependent increase in mobility of β 2 integrin (Fig. 2). The effect of nocodazole on integrin mobility reached a plateau at a concentration of 12.5 µM. At this concentration, nocodazole caused an 5.8-fold increase in integrin mobility, which was close to the previously reported effect of depolymerization of actin by cytochalasin D (8;9). As with the colchicine-treated lymphocytes, the increase in β2 integrin mobility was due to an increase in random motion. Thus, we conclude that microtubule cytoskeleton is involved in constraining integrin molecules.
Because the microtubule cytoskeleton is a dynamic system, constantly undergoing polymerization and depolymerization in living cells, we also tested taxol, a reagent that keeps microtubules in the polymerized state (48). We observed a concentration-dependent increase in mobility of β 2 integrin in J774 cells treated with taxol, similar to that caused by nocodazole (Fig.   3). This shows that, while taxol stabilizes microtubules, it also removes the cytoskeletal constraint on β 2 integrin mobility. Therefore, the dynamics of the microtubule system, and not merely the state of polymerization, are important for constraints on integrin mobility.
Microtubule-regulated integrin mobility is independent of PKC.

Microtubule depolymerization induces integrin-mediated signal transduction and cell-cell adhesion of J774 cells.
To determine whether the enhanced β 2 integrin mobility correlated with integrin- hours, J774 cells formed large aggregates (Fig. 6A). To confirm that this was β 2 integrinmediated cell aggregation, the aggregates were dispersed by repeated pipeting and the cells were recultured in the presence or absence of the Fab fragment of anti-β 2 antibody (2E6). Within 6 hours, cells cultured without anti-β 2 reaggregated, whereas those with antibody remained dispersed (Fig. 6B). Thus, we conclude that depolymerization of microtubules is sufficient to activate β 2 -integrin-mediated cell adhesion.
This β 2 integrin-mediated aggregation was also examined in J774-ED cells. Because of the MacMARCKS mutation, J774-ED cells did not aggregate in response to the addition of PMA (Fig. 6C). However, 25 µM nocodazole induced aggregation in these mutant cells (Fig. 6C).
This further supports the contention that the microtubule-mediated constraint on integrin mobility is independent of PKC and MacMARCKS.
Intact microtubules are required for cell spreading. In cell culture plates in media containing 10% FBS, J774 macrophages adhere to the culture dish surface but do not spread extensively. Treating these cells with PMA is sufficient to induce β 2 integrin-dependent spreading (Fig. 7). Although depolymerization of microtubules has the same effect as PMA in activating integrin-mediated adhesion, it does not stimulate cell spreading (Fig. 7). In fact, adding nocodazole (5 µM) inhibits macrophage spreading. This is a minimal nocodazole concentration, which only partially mobilizes integrins (c.f. figure 2), suggesting that cell spreading is exquisitely sensitive to microtubule disruption.

C3 exoenzyme, an inhibitor of Rho, prevents nocodazole-, but not taxol-induced increases
in integrin mobility. Rho regulates cytoskeletal rearrangement, and has effects on cell spreading and focal contact formation (32;55;58;59). Since microtubule inhibitors have been shown to affect Rho function, it is reasonable that Rho might be involved in microtubulemediated regulation of integrin mobility. To explore the molecular basis of regulation of integrin mobility, we tested the involvement of Rho in nocodazole-, taxol-and PMA-stimulated integrin diffusion. We found that while the C3 strongly inhibits the nocodazole-stimulated increase in integrin diffusion, it has little effect on the taxol-or PMA-mediated increase ( fig. 8)

Discussion
In this report, we demonstrated by direct diffusion measurements that the microtubule cytoskeleton participates in control of β 2 integrin avidity. This is the first report implicating microtubules in integrin rearrangement, and provides new insights into the regulation of activation of leukocyte adhesion. We demonstrated that disruption of the microtubule cytoskeleton causes an increase in the mobility of β 2 integrins. Further, not only the integrity, but also dynamics of microtubules are important for their role in controlling integrin mobility, since taxol, which stabilizes microtubules, also releases the integrins to diffuse.
It has been known for some time that microtubule integrity is necessary for cell spreading (54). It is generally believed that integrin activation precedes cell spreading. Thus, one might assume that microtubule integrity would be required for integrin activation. Interestingly, our data show that the opposite is the case. While we confirm that intact microtubules are required for cell spreading, it is interference with microtubule dynamics, either by depolymerization or stabilization, that results in integrin activation. This is true for both increased integrin mobility and enhanced cell-cell adhesion. Thus, microtubule dynamics are necessary to maintain β 2 integrins in the immobile, inactive state. Together with the result that partial disruption of the actin cytoskeleton is sufficient to mobilize β 2 integrins and activate adhesion (8) We prefer the latter, since a large body of evidence indicates that an intact cytoskeleton is an integral part of the spreading mechanism.
Regulation of both actin and tubulin polymerization in cells is complex, and the molecular details remain to be worked out. Clearly, the PKC pathway is involved, because activation of PKC achieves the same effect on integrin mobility as dissociating the cytoskeleton ((8) and this report). The PKC substrate MacMARCKS is also clearly involved, since mutation of MacMARCKS blocks increased β 2 integrin mobility (9). Microtubule involvement, however, is independent of PKC and MacMARCKS signals, because disruption of microtubules bypasses the mutation of MacMARCKS. This could indicate that the microtubule effect is downstream of PKC and MacMARCKS, or that the pathways are separate.
Our data also show that once the constraint on microtubule cytoskeleton is released, the It is also interesting that talin has been shown to bind to the cytoplasmic tail of The finding that the microtubule cytoskeleton is involved in control of integrin mobility provides important new information about regulation of integrin activation. This report demonstrates that the microtubule cytoskeleton has dual effects: while disruption of the microtubule cytoskeleton results in integrin activation, its integrity is required for cell spreading.
Our data, demonstrating that it is the dynamics, not the integrity of microtubules that maintains integrins in the cytoskeleton-restricted state indicates that microtubules probably regulate, rather that physically restrict, integrin motion. Evidence is accumulating that microtubules play unexpected roles in several adhesion-mediated cell functions, including adhesion, spreading and migration. However, the role of microtubules is often assumed to involve either structural effects or transport of intracellular vesicles. Our finding that microtubules are involved activation of integrins will have implications for understanding integrin function in general and its regulation in cell functions.   have been shown earlier to be unable to respond to PMA stimulation (13,14). However, in both cell types, 20 µM taxol and 20 µM Nocodazole each induced a significant increase in the diffusion coefficient of β 2 integrins (p < 0.01).