Enzymatically Quiescent Heparanase Augments T Cell Interactions with VCAM-1 and Extracellular Matrix Components under Versatile Dynamic Contexts

Reagents

The following reagents and chemicals were purchased from the sources indicated: RPMI 1640 (Life Technologies, Paisley, U.K.), FCS, HEPES buffer, antibiotics, and sodium pyruvate (Kibbutz Beit-Haemek, Israel); CO type IV (CO-IV; Sigma-Aldrich, Rehovot, Israel); FN (Chemicon International, Temecula, CA); hyaluronic acid (HA; Sigma, Rehovot, Israel); ECM-coated 96-well plates (NovaMed, Jerusalem, Israel); and recombinant 7-domain human soluble VCAM-1 (sVCAM-1), kindly provided by R. Lobb (Biogen, Cambridge, MA). Recombinant human SDF-1α and IL-2 were purchased from PeproTech Asia (Rehovot, Israel). The FN peptides RGDS (Arg-Gly-Asp-Ser), RGES (Arg-Gly-Glu-Ser), and LDV (Leu-Asp-Val) were purchased from Sigma-Aldrich. mAbs directed against human β₁ integrins (CD29, clone 3SS) and against CD44 were obtained from Serotec (Oxford, U.K.). Function-blocking anti-VLA4 integrin mAb HP1/2 was the kind gift of T. Kishimoto (Boehringer-Ingelheim Pharmaceuticals, Ridgefield, CT). Polyclonal Abs against phosphorylated Pyk-2 (py881) were obtained from BioSource International (Camarillo, CA); anti-total Pyk-2 (clone N-19) and mAbs against c-myc (clone 9E10) were from Santa Cruz Biotechnology (Santa Cruz, CA), and against phosphorylated extracellular signal-regulated kinase-2 (ERK-2) were from BioSource International; and anti-ERK-2 polyclonal Ab was from Sigma-Aldrich.

Recombinant human heparanase

HEK 293 cells, stably transfected with the human heparanase gene construct in the mammalian pSecTag vector (Invitrogen, San Diego, CA), were kindly provided by ImClone (New York, NY). This plasmid vector provides the IgG signal peptide needed to ensure efficient protein secretion, together with Myc and His tags at the protein C terminus to enable easy detection and purification. The cells were grown in DMEM supplemented with 10% FCS, glutamine, sodium pyruvate, and antibiotics. For the purification of the Myc-tagged 65-kDa latent heparanase, the cells were grown overnight in serum-free DMEM, and the conditioned medium (∼1 L) was purified on a Fractogel EMD SO₃⁻ (Merck, West Point, PA) column. The bound material was eluted with 1 M NaCl and was further purified by affinity chromatography on an anti-Myc (Santa Cruz Biotechnology) column. Using this procedure, we obtained a >95% pure 65-kDa nonenzymatically active heparanase preparation.

Human T cells

T cells were purified from healthy human donor peripheral blood. Whole blood was incubated with RosetteSep human T cell enrichment mixture (StemCell Technologies, Vencouver, British Columbia, Canada) for 20 min at room temperature. Next, the T cells were loaded on lymphocyte separation medium (ICN Biomedicals, Asse-Relegem, Belgium), isolated by density centrifugation, and washed. The purified cells obtained (>97% CD3⁺ T cells) were cultured in RPMI 1640 containing antibiotics, sodium pyruvate, glutamate, and 10% heat-inactivated FCS (13).

Binding of recombinant heparanase to immobilized ECM moieties

Flat-bottom microtiter well plates were precoated with FN or CO-IV (10 μg/ml), and the remaining binding sites were blocked with 0.1% BSA. Recombinant human heparanase was diluted in PBS containing 0.01% BSA and incubated on immobilized ECM proteins for 1 h at 37°C, 7% CO₂ in a humidified atmosphere, and then gently washed. The remaining binding sites were blocked with 0.1% BSA. Bound protein was detected with anti-Myc tag mAb, followed by anti-mouse HRP-labeled rabbit polyclonal Ab. Peroxidase substrate (tetramethylbenzidine) was then added, and the reaction was stopped by 1 M H₂SO₄. The readouts represent the mean OD of triplicate samples obtained at 450 nm.

Shear flow assays

The flow chamber assays have been described in detail elsewhere (17, 18). Purified sVCAM-1 was mixed in coating medium (PBS buffered with 20 mM sodium bicarbonate, pH 8.5) with a fixed amount of carrier (2 μg/ml human serum albumin (HSA)) and adsorbed as 10-μl spots on polystyrene plates (BD Biosciences, Erembodegem, Belgium) for 18 h at 4°C, alone or with recombinant heparanase (2 μg/ml) with or without SDF-1α (2 μg/ml). Experiments with immobilized heparanase (2 μg/ml) in the absence of sVCAM-1 were also performed. Plates were washed and blocked with HSA (20 mg/ml). The flow chamber was mounted on the stage of an inverted phase contrast microscope (Diaphot 300; Nikon, Melville, NY). All flow experiments were conducted at 37°C. Cells were perfused at 10⁶ cells/ml through the chamber at the desired flow rate generated with an automated syringe pump (Harvard Apparatus, Holliston, MA). The entire periods of cell perfusion were recorded on a videotape with a high-resolution video camera (Applitech, Holon, Israel) and a Time Lapse SVHS-Video recorder (AG-6730; Panasonic, Tokyo, Japan). All cellular interactions with the adhesive substrates were determined by manually tracking the motions of individual cells along 0.9-mm field paths for 1 min. Categories of tethered cells were defined according to their subsequent motion (transient, rolling, rolling followed by arrest, and immediate arrest), as previously described (17, 18, 19). In each experiment, all events were normalized to a constant cell population flowing in the immediate proximity with the substrate. The frequency of each category of tethers was expressed in the percentage of units (event × cell⁻¹ × 10²). To block the integrin α₄β₁ chain on the perfused T cells, the cells were preincubated for 5 min in binding medium with 5 μg of blocking mAb HP1/2 (10 μg/ml) and perfused unwashed over the substrates. To abrogate T cell interactions with immobilized heparanase, we pretreated the cells with an excess of soluble heparanase (2 μg/ml) and then placed in the flow chambers without washing.

T cell adhesion and migration assays

Analysis of T cell adhesion to ECM components was determined, as previously described (20, 21). Briefly, flat-bottom tissue culture-treated microtiter plates were precoated with FN, LN, CO-I, CO-IV (10 μg/ml), or HA (50 μg/ml), and the remaining binding sites were blocked with 0.1% BSA. Next, ⁵¹Cr-labeled T cells were resuspended in RPMI 1640 medium supplemented with 1% HEPES buffer and 0.1% BSA (adhesion medium). Then the indicated activators were added to the cells, and they were seeded onto the wells (10⁵ cells/well). The plates were further incubated (30 min, 37°C in a 7% CO₂-humidified atmosphere) and then gently washed. The adherent cells were lysed (1 M NaOH, 0.1% Triton X-100 in H₂O), removed, and counted using a gamma counter (Packard, Downers Grove, IL). The results are expressed as the mean (±SD) percentage of bound T cells from quadruplicate wells. Where indicated, the purified human T cells were pretreated (1 h under tissue culture conditions in serum-free medium) with Flavobacterium heparinum-derived heparinase II (1 U/ml; Sigma-Aldrich). The cells were then extensively washed, radioactively labeled, and placed onto FN-coated microtiter wells in the presence of the recombinant human heparanase or PMA. Where indicated, the following inhibitors of intracellular signal transduction pathways were used: wortmanin (a phosphatidylinositol 3-kinase (PI3K) inhibitor), GF109203X (a protein kinase C (PKC) inhibitor), PD98059 (an ERK inhibitor), and pertussis toxin (a G protein inhibitor), all obtained from Calbiochem (San Diego, CA).

The migration of ⁵¹Cr-labeled T cells was examined in a 48-well Transwell chemotaxis apparatus (6.5 mm diameter; Corning Glass, Corning, NY), consisting of two compartments separated by polycarbonate filters (5 μm pore size) pretreated (1 h, 37°C) with FN (25 μg/ml), as previously described (22). ⁵¹Cr-labeled T cells (2 × 10⁵ in 100 μl of RPMI 1640 containing 0.1% BSA, antibiotics) were added to the upper chambers with or without the indicated activators. The bottom chambers contained 0.6 ml of the same medium, with or without human SDF-1α (200 ng/ml). After incubation at 37°C for 3 h, cells that had transmigrated into the lower wells were collected, centrifuged, and lysed (in 100 μl of distilled water containing 1 M NaOH and 0.1% Triton X-100), and the radioactivity in the resulting supernatants was determined with a gamma counter. The percentage (±SD) of cell migration was calculated as the radioactivity counts in the lysates of the lower chambers divided by the total counts.

Western blot analysis of T cell lysates

Human T cells were incubated in starvation medium (RPMI 1640 medium without serum) for 48 h. Before testing, 5 × 10⁶ cells per sample were treated with soluble heparanase (usually 10 ng/ml) for 10 or 40 min (37°C in a 7% CO₂-humidified atmosphere) washed or left in the same medium, with or without subsequent incubation with SDF-1α (100 ng/ml, 10 min). The reaction was terminated by freezing the plates at −70°C for 10 min. The thawed cells were lysed (60 min, 4°C) in buffer containing EDTA (0.5 mM), NaCl (150 mM), NaF (10 nM), Tris (25 mM), Triton X-100 (1%), PMSF (200 μg/ml), and phosphatase inhibitor mixture (1%; Sigma-Aldrich), and cleared by centrifugation (30 min, 14 × 10³ rpm), and the supernatants were analyzed for total protein content (using the Bradford assay). Reducing sample buffer was then added, the samples were boiled, and aliquots containing equal amounts of protein were separated on 10% SDS-PAGE gel and transferred to nitrocellulose membranes. The membranes were blocked (TBST buffer containing low-fat milk (5%), Tris (pH 7.5, 20 mM), NaCl (135 mM), and Tween 20 (0.1%)) and probed with appropriate Abs. Immunoreactive protein bands were visualized using HRP-conjugated secondary Ab and the ECL system (22).

Intact or VCAM-1-associated recombinant heparanase regulates resting or SDF-1α-activated human T cell rolling and arrest

We first wished to identify possible nonenzymatic functions of heparanase under conditions resembling those found in small blood vessels (possibly adjacent to inflamed areas) in vivo. To this end, we analyzed T cell interactions with nonezymatically active heparanase, under low shear flow conditions (0.5 dyn/cm²). The results, shown in Fig. 1⇓A, indicate that the freshly isolated and purified human CD3⁺ T cells did not respond (e.g., adherence or rolling) to the HSA-coated surface (as shown in the left-hand side bar). This control group also indicates that the components of the serum, present in the medium, did not affect T cell behavior in this system. However, the T cells responded markedly (p < 0.01) to the plastic-bound heparanase by increasing their propensity to interact either transiently or stably with the enzyme-coated surfaces. This was manifested by the ability of a large fraction of the flowing cells to rapidly tether and either roll immediately or become arrested after encountering the immobilized heparanase. Note that the addition of soluble heparanase completely (p < 0.01) abrogated the effect of the bound enzyme on T cell arrest under flow conditions, but only moderately inhibited the effect of the bound enzyme on the T cells’ transient arrest. This is probably due to the fact that the cumulative influence of these forms of heparanase affects the occupancy and distribution of its putative cell surface receptor(s), as was suggested in explaining the rapid effects of SDF-1α on CXCR4-mediated T cell arrest and rolling measured under identical experimental conditions (17). This also suggests that only the bound (immobilized) form of the enzyme mediates the T cells’ arrest under flow conditions. Thus, intact heparanase alone can capture flowing T cells and induce their rolling and subsequent firm adhesion under shear flow conditions, even in the absence of selectin and integrin ligands. This suggests that such rapid T cell interaction with heparanase does not involve the activation of T cell adhesion receptors, such as integrins or selectins. Thus, heparanase may serve as a ligand for T cell adhesion by itself.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Regulation of T cell tethering events by heparanase. A, Recombinant human heparanase (65 kDa) was immobilized on plastic for 18 h with HSA as a carrier protein, and the unbound protein was then washed. Purified T cells were resuspended at 10⁶ cells/ml, and placed in the flow chambers in the presence or absence of soluble heparanase (2 μg/ml). B, Heparanase was coimmobilized with sVCAM-1 (2 μg/ml) alone or with sVCAM-1 and SDF-1α (2 μg/ml). The involvement of integrin activation in heparanase-induced tethering events was assessed by treating T cells with mAb anti-β₁ integrin (designated HP1/2; 15 μg/ml) for 5 min before the placement of the cells in the flow chambers. T cell interactions with the immobilized substrates under shear flow conditions were calculated, as described in Materials and Methods. The results shown here are from one experiment representative of three independent experiments.

Next, we measured the ability of VCAM-1-associated heparanase to modulate T cell arrest and rolling under shear flow conditions mediated by α₄β₁ (VLA-4), in the absence or presence of the proadhesive chemokine SDF-1α (CXCL12). The results, shown in Fig. 1⇑B, indicate that VCAM-1-associated heparanase can synergize with SDF-1α to promote both the rolling and arrest of human T cells interacting with VCAM-1. Interestingly, when the cells were exposed to both VCAM-1-complexed heparanase and coimmobilized SDF-1α, the proportion of cells completely arrested, as well as rolled/arrested, on VCAM-1 was significantly increased (p < 0.05), whereas their ability to roll over the immobilized ligands was decreased. This suggests that the juxtaposition of both VCAM-1-bound heparanase and the CXC chemokine induced a firm, rather than transient, T cell adhesion to the matrix. It is noteworthy that a nonspecific protein, HSA, which was used as a carrier protein for heparanase and SDF-1α, had no effect on T cell/VCAM-1 interactions. This also indicates that components of the serum, present in the cell medium, did not regulate this mode of interaction of T cells with VCAM-1 (Fig. 1⇑B).

We also assessed the involvement of the α₄β₁ integrin in these interactions by pretreating the cells with anti-VLA4 mAb, designated HP1/2, which was shown to inhibit the effects of SDF-1α in T cell interactions with endothelial cell ligands under shear flow conditions (17). We found that similar concentrations of the mAb (e.g., 15 μg/ml) abrogated, to some extent, the heparanase- and SDF-1α-induced transient arrest and rolling of T cells. This partial effect of the mAb indicates that, although such T cell responses were mediated primarily by the α₄β₁ integrin, other receptors, such as the putative heparanase-binding moieties, may concomitantly contribute to the heparanase- and SDF-1α-induced T cell behavior under such shear flow conditions. Thus, the intrinsic capturing ability of heparanase (Fig. 1⇑A) can potentially augment both constitutive and VLA-4-mediated T cell interactions with vessel wall-associated VCAM-1 (Fig. 1⇑B), as well as cooperate with chemokine-induced tethering of T cells under shear flow. Our results indicate that under such circumstances, heparanase can act in concert with the SDF-1α present on inflamed endothelial cells and aid in recruiting T cells to sites of inflammation.

Heparanase induces T cell adhesion to ECM, HA, CO-IV, and FN in a substrate-specific receptor manner

T cell adhesion to components of the ECM is a fundamental step in their migration from blood vessels into inflamed tissue. Therefore, we examined the ability of the recombinant heparanase to induce T cell adhesion to immobilized ECM ligands in a static adhesion assay. T cell adhesion in the absence of immobilized ECM components, but in the presence of the immobilized control substrate HSA, was always 2 ± 2% (data not shown). The results, shown in Fig. 2⇓A, indicate that heparanase induced significant levels of adhesion not only to ECM, but also to its immobilized intact components, including HA, which can also be found on blood vessel walls, in a dose-dependent manner, reaching a maximal effect at 10 ng/ml. Note that heparanase induced a significantly higher level of adhesion to FN (30% at 10 ng/ml), but had a lesser effect on unfractionated ECM, HA, and CO-IV: 20, 18, and 14% T cell adhesion, respectively. The parabolic dose-response curve of adhesion indicates a specific threshold effect of heparanase (i.e., 1–100 ng/ml), in which higher concentrations may have induced inhibition in heparanase effects in a receptor-desensitization manner. We also compared the adhesive effect of heparanase with that induced by SDF-1α and found that heparanase (at 10 ng/ml) induced a marked level of T cell adhesion to FN, comparable to that induced by SDF-1α (100 ng/ml): 28% compared with 24%, respectively, and that these effects did not involve the modulation of expression of β₁ integrins (or CXCR4; data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Heparanase induces T cell adhesion to ECM and ECM constituents in a HSPG-dependent manner. A, Human T cells were labeled with ⁵¹Cr and seeded onto microtiter wells, coated with ECM, FN, HA, or CO-IV in the presence of different amounts of heparanase. The fraction of adherent cells was determined after 40 min. Mean ± SD of quadriplicate wells is shown. The results shown here are from one experiment representative of three. Value of p < 0.01 for the heparanase-induced T cell adhesion to FN, HA, and ECM, when the enzyme’s concentrations were 10–100 ng/ml. B, T cells were pretreated (1 h, tissue culture conditions, serum-free medium) with heparinase II (1 U/ml). The cells were then extensively washed, radioactively labeled, and placed onto FN-coated microtiter wells in the presence of the human heparanase (100 ng/ml) or PMA (50 ng/ml), and their adhesion was measured as above. For both A and B, T cell adhesion obtained in one experiment representative of three is shown. ∗, p < 0.01.

It is conceivable that the T cells’ surface-expressed HSPG play a role in mediating the proadhesive functioning of heparanase. To examine this possibility, the cells were pretreated with heparinase (1 h, 1 U/ml), a heparin- and HS-degrading bacterial enzyme, and their FN-adhesive capacities were examined after removing the enzyme from the culture medium. The results indicate that the heparinase pretreatment did not affect the basal responses of T cells to FN, nor did it affect the PMA-induced T cell adhesion. In contrast, the T cell HSPG appears to be involved in mediating the effects of heparanase, as the T cell-adhesive response to the enzyme was markedly (p < 0.05) suppressed (Fig. 2⇑B).

Next, we examined the role of CD29 (the common β₁ integrin chain) and CD44 in the heparanase-induced T cell adhesion to FN and HA. We compared the effects of heparanase with that of IL-2, which served as a control proinflammatory cytokine. IL-2- and heparanase-induced T cell adhesion to FN was markedly inhibited by mAb anti-CD29 (and not by mAb anti-CD44), as well as peptides representing FN α₄β₁ and α₅β₁ integrin-specific LDV- and RGD-containing amino acid sequences, respectively (Fig. 3⇓A). In parallel, T cell adhesion to HA induced by either IL-2 or heparanase was specifically inhibited by mAb anti-CD44 (Fig. 3⇓B). Note that although we used an optimal inhibitory concentration of the mAb anti-CD44 (e.g., 15 μg/ml) (21), it did not fully abolish the heparanase-induced T cell adhesion to HA, suggesting that in addition to its effect on CD44, heparanase induces the functioning of other, putative HA-specific receptors. Taken together, these results indicate that even in the absence of HSPG in the substrates, the enzymatically quiescent heparanase can induce marked T cell adhesion to ECM and its immobilized ligands in an integrin, as well as CD44-dependent manner.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Heparanase-induced T cell adhesion to FN and HA is dependent on CD29 and CD44, respectively. A, T cells were radiolabeled, pretreated (1 h) with mAb specific to the β₁ integrin chain (anti-CD29, 10 μg/ml), or the RGD-, LDV-, or RGE-containing peptides (20 μg/ml). Next, heparanase (100 ng/ml) was added to the cell suspensions, which were then seeded onto FN-coated microtiter wells, and the amount of cell adhesion was assessed 40 min later. Mean ± SD of quadruplicate wells of one experiment representative of five is shown. B, Labeled T cells were incubated with mAb anti-CD44 (15 μg/ml) or anti-CD29, and then placed on microtiter plates that were precoated with HA (50 μg/ml), in the presence of soluble heparanase (100 ng/ml) for 3 h. The adherent cells were then lysed, and the amount of cell-associated radioactivity was measured. For A and B, ∗ and ∗∗, p < 0.01 comparing the mAb- or peptide-treated cells with the untreated cells. Mean ± SD of one experiment representative of four experiments is shown.

Analysis of the signaling pathways involved in the heparanase-induced T cell-FN interactions

We examined the putative functional role of diverse intracellular signaling pathways that regulate the heparanase-mediated T cell adhesion to FN. To this end, T cell adhesion to FN was assessed following exposure of the cells to SDF-1α, heparanase (both at 10 ng/ml), or PMA (25 ng/ml), while also being exposed to active concentrations of specific inhibitors of cell signaling. The results, shown in Fig. 4⇓, indicate that similar to the effect of PMA, the heparanase-induced T cell adhesion appears to involve specific intracellular signaling pathways, including the PI3K, PKC, and ERK pathways, which are associated with leukocyte cytoskeletal rearrangement, integrin functioning, and cell adhesion (23), because signaling was inhibited by the wortmanin, GF109203X (GF), and PD98059 (PD), respectively, but not by pertussis toxin (PTX), which inhibits G protein-coupled receptor signaling (Fig. 4⇓, A and B). We could not observe significant changes in cell morphology under these conditions (data not shown). Because these results show the net outcome of T cell adhesion to FN (after 30 min in a static adhesion assay), we cannot demonstrate, at the present time, which particular step in T cell adhesion is sensitive to the inhibitory compounds. In contrast, and as expected, SDF-1α-induced T cell adhesion was also inhibited by PTX (Fig. 4⇓C). Thus, although the involvement of heparanase in inducing these particular T cell activatory pathways should be examined also in the shear flow assay, our findings suggest that the enzyme is capable of affecting the PI3K, PKC, and ERK intracellular signaling pathways, which are associated with T cell-ECM interactions.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Signaling elements involved in the PMA-, heparanase-, and SDF-1α-induced T cell adhesion to FN. Purified and ⁵¹Cr-labeled human T cells were pretreated with the indicated intracellular signal transduction inhibitors wortmannin (5 nM), GF109203X (GF; 20 nM), PD98059 (PD; 20 μM), and PTX (2 μg/ml), or left untreated. The cells were then activated with PMA (100 ng/ml; A), heparanase (100 ng/ml; B), or SDF-1α (150 ng/ml; C), and seeded, in quadriplicates, onto FN-precoated microtiter wells. The level of T cell adhesion was determined 30 min later by measuring the amount of radioactivity in the lysates of the bound cells. Mean ± SD representative of three separate experiments. Except for PTX in A and B, all the other compounds, including PTX in C, significantly (p < 0.01) inhibited T cell adhesion to FN.

Heparanase induces Pyk-2 and ERK phosphorylation and enhances SDF-1a-induced phosphorylation of these moieties

To gain further insight into the pathways through which heparanase exerts its proadhesive functions, we examined the effect of heparanase, with or without SDF-1α, on the actual activation (phosphorylation) of the cytoskeletal and integrin-associated components, ERK and Pyk-2. The 120-kDa proline-rich cytoplasmatic tyrosine kinase 2, also called Pyk-2, is a member of the focal adhesion kinase family, which upon the phosphorylation of tyrosine, links β₁ integrins to multiple signaling pathways involved in cell adhesion and migration (24). Such intracellular processes are also involved in the regulation of mitogen-activated protein kinases, such as the ERK-2 and Jun-NH₂ pathways (23, 25). As shown in Fig. 5⇓, heparanase (10 ng/ml) induced a marked phosphorylation of Pyk-2, comparable to that induced by SDF-1α alone (100 ng/ml for 10 min). When both mediators were simultaneously present, the amount of Pyk-2 phosphorylation was higher than that induced by either one of them alone (Fig. 5⇓A). Note that these intracellular effects of heparanase are probably β₁ integrin independent, because these studies were performed in the absence of immobilized ECM ligands. Thus, similar to its proadhesive effect, the biologically active concentration of heparanase induces Pyk-2 phosphorylation, and augments this activatory pathway upon cotreatment of the cells with SDF-1α.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Heparanase enhances SDF-1α-induced phosphorylation of Pyk-2 and ERK-2. A, T cells (5 × 10⁶ per group) were treated (10 min) with heparanase (10 ng/ml) and/or SDF-1α (100 ng/ml), frozen at −70°C, and lysed. The lysates were run on 10% SDS-PAGE, transferred to nitrocellulose membranes, and immunoblotted with peroxidase-labeled anti-phospho-Pyk-2 (pPYK2) and anti-total Pyk-2 (tPYK2) Abs, followed by ECL visualization and densitometry. One experiment representative of three is shown. B and C, T cells were incubated with heparanase (100 ng/ml) for 10 or 40 min, and then left in the same medium (B), or washed and resuspended in fresh medium (C) and exposed to SDF-1α (250 ng/ml) for 10 min. T cell lysates were then immunoblotted with the indicated Ab. One experiment representative of five is shown. The histograms represent a ratio between band density of phosphorylated and total protein for each sample.

We also measured the ability of heparanase to affect the phosphorylation of ERK. The results demonstrate that exposure of T cells to heparanase markedly elevated the SDF-1α-induced ERK phosphorylation, whether both mediators were simultaneously present in the T cell cultures, or whether the cells were first exposed to heparanase, and then washed and exposed to SDF-1α (Fig. 5⇑, B and C). Thus, heparanase, which induces Pyk-2 phosphorylation, also augments the SDF-1α-induced activation of this protein kinase, as well as that of ERK (which is not phosphorylated in response to heparanase alone; data not shown). It is noteworthy that although Fig. 5⇑ depicts the results of one experiment of three to five done on T cells from different donors, we always observed the same pattern of effects of heparanase on Pyk-2 and ERK. Interestingly, analysis of T cell from five different donors revealed that both the p42 and p44 isoforms of ERK were similarly affected (e.g., phosphorylated) by heparanase, although, as shown in this work, the level of phosphorylation of the p44 isoform was always lower than that of the p42 isoform. These findings may explain the ability of heparanase to activate ECM-specific receptors (i.e., β₁ integrins), and its ability to rapidly stimulate T cell responses (e.g., adhesion, chemotaxis) to SDF-1α, induced by its T cell surface receptor, CXCR4.

Heparanase binds FN, retains its proadhesive activity, and regulates SDF-1α-induced chemotaxis

We have previously demonstrated that at pH 7.2, the placental-extracted heparanase can bind, but not degrade, HSPG, and thus, noncovalently associate with ECM (13). We therefore questioned whether the highly purified recombinant heparanase can also bind FN, assuming that such molecular interactions with this major cell-adhesive glycoprotein of the ECM may serve to preserve heparanase, in its enzymatic or nonenzymatic active forms, within the context of migrating lymphocytes. To this end, the microtiter wells were precoated with FN and CO-IV, and the remaining uncoated binding sites were blocked with BSA. Next, heparanase binding to immobilized FN (and CO-IV) was examined using specific anti-c-Myc tag Ab. The results, depicted in Fig. 6⇓A, show that the values of heparanase binding to the ECM substrates are above those obtained with BSA alone. Moreover, they indicate that the human heparanase binds FN in a dose-dependent manner, reaching a saturable amount of binding at 1–4 μg/ml; calculation of the amount of the FN-bound enzyme revealed that ∼1–2.5% of the input protein actually remained bound to FN (data not shown). In addition, heparanase bound, although to a much lesser degree, to CO-IV.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Heparanase binds to FN and promotes T cell chemotaxis toward SDF-1α. A, Different concentrations of heparanase were incubated (1 h, 37°C) on FN- or CO-IV-coated microtiter wells, in the presence of 0.01% BSA as a carrier. The unbound material was removed, and bound heparanase was detected using mAb anti-c-Myc tag, followed by HRP-labeled secondary Ab, and substrate. The amount of bound protein was measured (at 450 nm) using a spectrophotometer. The binding values of heparanase shown were calculated as those obtained above the binding of the enzyme to immobilized BSA alone. Mean ± STD of triplicate wells of one experiment representative of two is shown. B, Polycarbonate membranes, separating the upper and lower chambers of the Transwell apparatus, were precoated with FN (1 h, 37°C), and further pretreated with heparanase or PBS (1 h, 37°C, pH 7.4). Next, the membranes were washed extensively, and the labeled T cells were placed in the upper wells, while SDF-1α was added to the lower chambers. T cell migration along the SDF-1α gradients was assayed after 2.5 h under tissue culture conditions. ∗, p < 0.01 compared with the other groups. C, T cells were placed in the upper chambers, and heparanase (100 ng/ml) was added to the upper or lower chambers, with or without SDF-1α. After 2.5 h, the cells in the lower chambers were collected by centrifugation and lysed, and their radioactivity was measured. ∗, p < 0.01 vs control, and ∗∗, p < 0.01 of the SDF-1α- and heparanase-treated groups was compared with the other groups. Values shown are mean ± SD, of a single experiment representative of three.

Next, we examined the biological ramifications of the FN-heparanase complexes by measuring T cell chemotaxis, through the immobilized ECM glycoprotein, in the absence or presence of SDF-1α. In the absence of SDF-1α, FN-bound heparanase did not affect T cell migration. However, the FN-bound heparanase markedly augmented (p < 0.01) the promigratory capacity of SDF-1α, from 30% in the absence of heparanase to 74% in the presence of the FN-heparanase complex (Fig. 6⇑B). Thus, the FN-bound heparanase enhances T cell responses to SDF-1α, as manifested in increased chemotaxis.

Finally, we examined whether soluble heparanase can induce T cell migration when it forms a concentration-dependent gradient in the Transwell system. The results indicate that, in contrast to the SDF-1α-induced T cell chemotaxis, soluble heparanase had no apparent promigratory effect on T cells (Fig. 6⇑C). However, when both mediators were placed in the bottom chambers, thus creating a combination of two chemotactic gradients, a marked elevation in T cell chemotaxis occurred (27% in the presence of SDF-1α vs 44% migration in the presence of both mediators; p < 0.01). Thus, soluble heparanase synergizes with SDF-1α to promote enhanced T cell chemotaxis through FN.