Characterization of Mechanisms Involved in Secretion of Active Heparanase*

Heparanase is an endo-β-d-glucuronidase involved in extracellular matrix remodeling and degradation and implicated in tumor metastasis, angiogenesis, inflammation, and autoimmunity. The enzyme is synthesized as a latent 65-kDa protein and is processed in the lysosomal compartment to an active 58-kDa heterodimer, where it is stored in a stable form. In contrast, its heparan sulfate substrate is localized extracellularly, suggesting the existence of mechanisms that trigger heparanase secretion. Here we show that secretion of the active enzyme is mediated by the protein kinase A and C pathways. Moreover, secretion of active heparanase was observed upon cell stimulation with physiological concentrations of adenosine, ADP, and ATP, as well as by the noncleavable ATP analogue adenosine 5′-O-(thiotriphosphate). Indeed, heparanase secretion was noted upon cell stimulation with a specific P2Y1 receptor agonist and was inhibited by P2Y receptor antagonists. The kinetics of heparanase secretion resembled the secretion of cathepsin D, a lysosomal enzyme, indicating that the secreted heparanase is of lysosomal origin. We suggest that secretion of active heparanase is initiated by extracellular cues activating the protein kinase A and C signaling pathways. The secreted enzyme(s) then facilitate cell invasion associated with cancer metastasis, angiogenesis, and inflammation.

Heparanase is an endoglycosidase that specifically cleaves heparan sulfate (HS) 2 side chains of heparan sulfate proteoglycans (HSPG) (1,2). HSPG consist of a protein core to which HS side chains are covalently attached. These complex macromolecules are highly abundant in the extracellular matrix (ECM) and are thought to play an important structural role, contributing to ECM integrity and insolubility (3). In addition, HS side chains can bind a variety of biological mediators, such as growth factors, cytokines, and chemokines, thus functioning as a readily available reservoir that can be liberated upon local or systemic cues. Moreover, HSPG on the cell surface participate in signal transduction cascades by potentiating the interaction between certain growth factors and their receptors (4 -6). HSdegrading activity is thus expected to affect several fundamental aspects of cell behavior under normal and pathological settings and should therefore be kept tightly regulated. Traditionally, heparanase activity was implicated in cellular invasion associated with angiogenesis, inflammation, and cancer metastasis (7)(8)(9)(10)(11). This notion gained further support by employing small interfering RNA and ribozyme technologies, clearly depicting heparanase-mediated HS cleavage and ECM remodeling as critical requisites for metastatic spread (12). Since the cloning of the heparanase gene and the availability of specific molecular probes, heparanase up-regulation was documented in an increasing number of primary human tumors, correlating with enhanced local and distant metastasis, increased microvessel density, and reduced postoperative survival of cancer patients. Collectively, these studies provide compelling evidence for the clinical relevance of the enzyme, making it an attractive target for the development of anti-cancer drugs (1,2,13).
Similar to several other classes of enzymes, heparanase is first synthesized as a latent enzyme that appears as a ϳ65-kDa protein when analyzed by SDS-PAGE. The 65-kDa latent enzyme is directed to the ER by a C terminal 35-amino acid signal peptide and is readily detected in the culture medium of transfected cells (14). The latent heparanase form does not accumulate extracellularly, however, due to an efficient cellular uptake (14,15), followed by intracellular proteolytic processing (15,16), yielding an 8-kDa polypeptide at the N terminus and a 50-kDa polypeptide at the C terminus that heterodimerize to form the active heparanase enzyme (17)(18)(19). Likewise, heparanase was noted to reside primarily intracellularly within endocytic vesicles identified as endosomes and lysosomes (20 -22). Applying a polyclonal antibody (number 733) that preferentially recognizes the 50-kDa heparanase subunit versus the 65-kDa latent enzyme, we have demonstrated that the 50-kDa active heparanase subunit similarly resides in endocytic vesicles, assuming a perinuclear localization (16). More recently, we have demonstrated heparanase processing by endosomal/lysosomal preparation (22), identified the lysosome as the heparanase-processing organelle (16), and identified cathepsin family members, mainly cathepsin D and L, as heparanase-activating proteases (23). Accumulation of heparanase in endocytic vesicles for a relatively long period of time (16,21) led us to hypothesize that this compartment may serve as an intracellular enzyme pool that can get secreted in response to a proper stimulus, ensuring a tightly regulated extracellular enzymatic function. We investigated this hypothesis by examining heparanase secretion and activity in the cell conditioned medium in response to exogenous stimuli. Here, we provide evidence that stimulation of tumor-derived cells with phorbol 12-myristate 13-acetate (PMA) and forskolin markedly enhances the secretion of active heparanase in a time-and dose-responsive manner. We further demonstrate that physiological concentrations of ATP and ADP similarly enhance the secretion of active heparanase and discuss the significance and possible implications of these findings.
Cell Culture and Transfection-Human MDA-MB-231, MDA-MB-435, and MDA-MB-468 breast carcinoma and HCT116 and HT29 colon carcinoma cells were purchased from the ATCC. Cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum and antibiotics. For stable transfection, subconfluent MDA-468, MDA-435, MDA-231, and HT29 cells were transfected with the pSecTag2 vector (Invitrogen) containing the full-length heparanase cDNA, using the FuGENE 6 reagent, according to the manufacturer's instructions (Roche Applied Science). The pSecTag2 vector is designed for efficient protein secretion driven by the IgG signal peptide and contains c-Myc and His tags at the protein C terminus. Transfection proceeded for 48 h, followed by selection with Zeocin (Invitrogen) for 2 weeks. Stable transfectant pools were further expanded and analyzed.
Heparanase Secretion and Immunoblotting-Cells were grown to confluence, followed by incubation for 20 h in serumfree medium. Fresh serum-free medium was added, and the cells were incubated without or with the indicated reagent for an additional 20 h, unless stated otherwise. Conditioned medium was collected and applied onto 35 S-labeled ECMcoated dishes to evaluate heparanase enzymatic activity (see below) or preabsorbed with concavalin A-Sepharose beads to concentrate the samples and reduce nonspecific reactivity, followed by SDS-PAGE under reducing conditions using 10% gels. After electrophoresis, proteins were transferred to polyvinylidene difluoride membrane (Bio-Rad) and probed with the appropriate antibody followed by horseradish peroxidase-conjugated secondary antibody (Jackson ImmunoResearch, West Grove, PA) and an enhanced chemiluminescent substrate (Pierce), as described (14,16,24).
Immunocytochemistry-Heparanase transfected HT29 colon carcinoma cells were left untreated or incubated with ATP (10 M) for 2 h. Indirect immunofluorescence staining was then performed essentially as described (16,25). Briefly, cells were fixed with cold methanol for 10 min, washed with phosphatebuffered saline, and subsequently incubated in phosphate-buffered saline containing 10% normal goat serum for 1 h at room temperature, followed by a 2-h incubation with monoclonal anti-heparanase antibodies. Cells were then extensively washed with phosphate-buffered saline and incubated with Cy2-conjugated secondary antibody (Jackson ImmunoResearch) for 1 h, washed, and mounted (Vectashield, Vector, Burlingame, CA). Nuclei were counterstained with propidium iodide (Vector), and staining was visualized by confocal microscopy.

PMA and Forskolin Induce Secretion of Active Heparanase-
Heparanase activity is strongly implicated in biological processes that require ECM remodeling, such as cancer metastasis, inflammation, and angiogenesis (1,2,13). Interestingly, however, heparanase was noted to reside primarily intracellularly within endocytic vesicles identified as endosomes and lysosomes (16,20,21). Thus, heparanase secretion from intracellular pools may be required in order to exert its enzymatic function extracellularly. We examined this possibility by exposing cells overexpressing heparanase to PMA, followed by immunoblot analysis of the culture medium. As demonstrated in Fig. 1, treatment of MDA-468 (Fig. 1A), MDA-435 (Fig. 1B), HT29 (Fig. 1C), and 293 cells (Fig. 1D) with PMA resulted in a marked increase of the 50-kDa active heparanase subunit in the culture medium. In contrast, the latent 65-kDa protein was readily detected in the culture medium of control untreated cells (Con) due to its secreted nature (1), and its levels were not significantly changed upon PMA treatment. In order to confirm the immunoblotting results, conditioned medium from control and PMA-treated HT29 cells was applied onto 35 S-labeled ECM, and heparanase enzymatic activity was evaluated. Heparanase activity was not detected in medium conditioned by nontransfected HT29 cells (data not shown) but was clearly evident in medium conditioned by heparanase-transfected cells (Fig. 1E, Con). Heparanase activity was markedly increased in the conditioned medium of heparanase-transfected HT29 cells in response to treatment with PMA ( Fig. 1E, PMA), in agreement with the elevated levels of the 50-kDa heparanase subunit detected by immunoblotting (Fig. 1C). Treatment of heparanase-transfected MDA-468, MDA-435, and HT29 cells with forskolin did not result in a significant increase of heparanase secretion (Fig. 1, A-C) and activity (data not shown). In contrast, treatment of heparanasetransfected 293 cells with forskolin stimulated a significant increase in secretion of the 50-kDa heparanase subunit (Fig. 1D), correlating with enhanced heparanase activity in the cell culture medium (Fig. 1F). Since PMA is a strong PKC inducer, we next examined the ability of PKC inhibitors to block heparanase secretion induced by PMA. To this end, heparanase-transfected HT29 (Fig. 1G, upper panel) and MDA-435 (Fig. 1G, lower panel) cells were left untreated (Ϫ) or stimulated with PMA in the absence or presence of PKC (Bis) or PKA (H89) inhibitors. Heparanase secretion was examined by immunoblotting. PMA treatment elicited a marked increase in secretion of the 50-kDa heparanase subunit, an increase that was practically blocked (Fig. 1G, top) or markedly reduced (Fig. 1G, bottom) in the presence of the PKC inhibitor Bis. In contrast, the PKA inhibitor H89 only slightly reduced the effect of PMA, indicating, as expected, activation of PKC rather than of PKA. The inverse situation was noted upon treatment of heparanase-transfected 293 cells with forskolin. In these cells, forskolin was found to effectively induce secretion of active heparanase (Fig. 1D), and this effect was significantly inhibited by H89 but not by Bis (not shown).
Next, we examined the kinetics of heparanase secretion elicited by PMA and forskolin (Fig. 2). Heparanase-transfected MDA-435 cells were incubated with PMA for the time indicated, conditioned medium was collected, and heparanase secretion was examined by immunoblotting. The 50-kDa heparanase subunit was first detected in the culture medium 1 h following PMA stimulation, peaked at 2 h, and gradually decreased ( Fig. 2A,  top). A similar kinetic was noted upon treatment of heparanasetransfected 293 cells with forskolin (Fig. 2B, top). The decline in extracellular heparanase levels at later time points following PMA or forskolin treatment is most probably due to HS-mediated heparanase uptake (14).
Since the active 50-kDa heparanase subunit was mainly detected in endocytic vesicles (16), we rationalized that the secreted heparanase found in the culture medium following treatment with PMA and forskolin originated from such an intracellular pool. In order to support this notion, we compared the secretion kinetics of the lysosomal protein cathepsin D with that of the 50-kDa heparanase subunit. Notably, secretion of the low molecular weight forms of cathepsin D, typically processed and residing in lysosomes, was not only induced by PMA and forskolin but also resembled the kinetics of heparanase secretion (Fig. 2, lower panels), suggesting that both proteins indeed originate from similar cellular compartments.
Nucleotides Induce Secretion of Enzymatically Active Heparanase-The ability of PMA and forskolin to induce the secretion of enzymatically active heparanase as well as of a typical lysosomal enzyme, such as cathepsin D, supports the notion that heparanase is stored in the lysosomal compartment in a stable form (16,21) and is secreted in response to local or systemic cues, thus maintaining its extracellular availability tightly regulated. Since both PMA and forskolin are not considered physiological, we sought inducers more relevant to biological systems in which heparanase activity is implicated, mainly cancer metastasis and inflammation. The intracellular role of ATP is well recognized as a key energy source utilized for cellular metabolism. In addition, purines and pyrimidines, mainly ATP, ADP, UTP, and adenosine, have the ability to function extracellularly and to initiate signal transduction cascades mediated by a family of purinergic receptor that appears to play important roles in development, differentiation, and cell proliferation (27,28). The P1 and P2Y subgroups of the purinergic receptors family are G-protein-coupled receptors that activate phospholipase C and modulate cAMP levels, leading to PKC-and PKAmediated signal transduction (28). Since PKA and PKC activation was noted to enhance heparanase secretion (Figs. 1 and 2), we examined the ability of nucleotides to elicit a similar response. All cell types included in this study responded to nucleotides by elevating the levels of active heparanase in the cell conditioned medium, yet each cell type exclusively responded to different nucleotides (Fig. 3). Thus, whereas 293 cells preferentially responded to ATP and ADP (Fig. 3A), MDA-435 cells responded to adenosine (ADO ; Fig, 3B), and MDA-231 cells responded primarily to ADP (Fig. 3C), possibly reflecting different receptor profiles expressed by each cell line. Importantly, elevation of the 50-kDa heparanase subunit in the cul-ture medium was accompanied by increased heparanase activity in the nucleotide-treated 293, MDA-435, and MDA-231 cells (Fig. 3, D-F, respectively). Next, we examined the effect of ATP on heparanase secretion as a function of concentration and incubation time. HEK293 cells stably overexpressing heparanase were left untreated (0) or stimulated with ATP at the indicated concentrations. Heparanase secretion into the culture medium was evaluated by immunoblotting. As shown in Fig. 4A, lowering the ATP concentration markedly enhanced secretion of the 50-kDa heparanase protein. A concentration of 1 M was applied in subsequent experiments. Interestingly, low ATP concentrations were also more effective in stimulating the secretion of cathepsin D (Fig. 4A, bottom), further supporting endocytic vesicles and the lysosomal compartment as the primary source of these enzymes. Next, we examine the kinetics of heparanase secretion in response to ATP. To this end, heparanase-transfected 293 cells were left untreated or stimulated with 1 M ATP for the indicated time, and heparanase secretion was evaluated. Whereas the 65-kDa latent heparanase was readily secreted by control, untreated cells (Fig. 4B, left, Con), secretion of the 50-kDa protein was restricted to cells treated with ATP (Fig. 4B, right, ATP). The 50-kDa heparanase subunit was first detected 20 min following ATP addition, peaked at 1 h, and declined thereafter to undetectable levels (16 h), possibly due to efficient cellular uptake of the secreted protein (14). Importantly, secretion of the 50-kDa subunit precedes the secretion of the 65-kDa latent heparanase. These differences in secretion kinetics rule out the possibility that the 50-kDa subunit found in the cell culture medium is the result of extracellular processing of the 65-kDa latent heparanase, thus strongly implying that the 50-kDa form originated from an intracellular pool.
Next, we examined heparanase localization in control untreated and ATP-stimulated cells by means of confocal microscopy. In control unstimulated cells, heparanase was noted to reside mainly in vesicle-like structures, assuming a polar, perinuclear localization (Fig. 4C, left), in agreement with previous studies (16,21,29). In contrast, upon ATP stimulation, heparanase-positive vesicles assumed a more diffused localization and appeared in close proximity to the cell periphery (Fig. 4C, right panel). In addition, heparanase-positive vesicles appeared bigger upon ATP treatment, possibly due to vesicle fusion, resulting in, perhaps, a more readily secretable organelle.
Heparanase Secretion Induced by ATP and ADP Involves the P2Y1 Receptor and PKC-Having demonstrated that heparanase secretion induced by PMA and forskolin is mediated by the PKC and PKA signaling pathways (Fig. 1), we questioned whether a similar signaling mechanism is also evoked by nucleotides. Treatment of heparanase-transfected 293 cells with ATP (1 M) stimulated secretion of the 50-kDa protein (Fig. 5A, top), in agreement with our previous results (Figs. 3 and 4), and correlated with a comparable induction of cathepsin D secretion (Fig. 5A, lower panel). Importantly, secretion of heparanase and cathepsin D was markedly attenuated by pretreatment with PKC and PKA inhibitors (Bis and H89, respectively) (Fig. 5A), whereas the phospholipase C inhibitor (U73122) appeared less effective. Similarly, treatment of these cells with ADP (10 M) markedly stimulated heparanase secretion that was practically blocked by the PKC inhibitor Bis, whereas PKA inhibition by H89 was less effective (Fig. 5B). These results were further corroborated by evaluating heparanase activity in the cell conditioned medium. Clearly, heparanase activity was mark-edly increased upon ADP treatment, whereas PKC inhibitor (Bis) reduced heparanase activity to the level of control untreated cells, and PKA inhibition resulted in an about 50% decrease in heparanase activity (Fig. 5C), nicely correlating with the immunoblotting results (Fig. 5B). Furthermore, ATP␥S, a nonhydrolyzed ATP analogue that cannot be consumed by cells as an energy source, was as effective as ATP in eliciting heparanase secretion (data not shown), supporting the notion that heparanase secretion induced by ATP, and probably by other nucleotides, is a receptor-mediated response.
In order to better characterize cell surface receptors that mediate heparanase secretion, we utilized more specific P2Y receptor agonists and antagonists. Treatment of heparanase-transfected 293 cells with 2Me-SADP (2Me), a P2Y1 receptorspecific agonist (30), elicited a marked elevation of heparanase secretion (Fig. 5D), comparable in magnitude with the stimulatory effect of ATP and ADP on heparanase secretion (Figs. 3 and 4). Moreover, the effect of 2Me-SADP was abolished by PPADS, an antagonist of P2Y1, -4, and -6 receptors, and even more so by MRS2179, an antagonist specific for P2Y1 receptors, clearly implicating the P2Y1 receptors and PKC as mediators of heparanase secretion.
Since heparanase activity has long been correlated with acquired cellular motility, we questioned whether elevated levels of extracellular heparanase activity upon nucleotide stimulation enhance cell migration. Significantly, ATP and ADP markedly enhanced 293 (Fig. 6, left panels) and MDA-231 cell migration (Fig. 6, right panels), ascribing a cellular effect to the above described molecular and biochemical alterations induced by nucleotides.

DISCUSSION
HSPG are thought to play key roles in numerous biological settings, including cytoskeleton organization, cell migration, wound healing, inflammation, cancer metastasis, and angio-  Heparanase-transfected HEK293 cells were serum-starved for 20 h. Serum-free medium was then replaced, and cells were left untreated (0) or stimulated with the indicated concentration (in M) of ATP in the absence of heparin. Conditioned medium was collected after 2 h and subjected to immunoblotting with anti-heparanase 1453 (top) and anti-cathepsin D (bottom) antibodies. B, kinetics. Heparanase-transfected HEK293 cells were serum-starved for 20 h, serum-free medium was replaced, and cells were stimulated with ATP (1 M) for the indicated time in the absence of heparin. The cell conditioned medium was analyzed by immunoblotting with anti-heparanase 1453 antibodies. C, immunofluorescence. Hepa-transfected HT29 cells were serum-starved for 20 h, serum-free medium was replaced, and the cells were left untreated as control (Con) or stimulated with ATP (10 M) for 2 h. Cells were then fixed and double-stained with monoclonal anti-heparanase antibody (green) and propidium iodide (red) and examined by confocal microscopy. Note the more diffused and larger heparanase-positive vesicles upon ATP stimulation.
genesis (3,4,31,32). HSPG exert their multiple functions via several distinct mechanisms, combining biochemical, structural, and regulatory aspects. The multitude of polypeptides sequestered and regulated by HS (33,34) and the ability of heparanase to convert these into biologically accessible active molecules (1,35) require that heparanase extracellular activity be kept tightly regulated. Mechanisms that dictate heparanase regulation are poorly understood but are expected to operate at several distinct levels. Heparanase induction noted in an increasing number of human tumors clearly indicates gene transcription as a major regulatory mechanism (36 -38). Regulation at the post-translational level, namely heparanase processing and extracellular retention, has also been implicated as a major regulatory mechanism (14,16,23). Heparanase is first synthesizing as a preprolatent enzyme containing an N terminus signal peptide (Met 1 -Ala 35 ) that directs the newly formed protein to the ER and ensures protein secretion. Indeed, the 65-kDa latent heparanase protein is readily detected in the culture medium of transfected cells (14,15), yet the protein does not accumulate to high levels extracellularly due to rapid and efficient cellular uptake mediated by cell surface HSPG, the mannose 6-phosphate, and the low density lipoprotein-related receptors (14,15). Subsequently, the 65-kDa latent heparanase protein is subjected to intracellular lysosomal processing (15,16,22,23) and assumes typical polar, perinuclear localization (16,20,21). Secretion mechanisms described in the current study provide an additional level of heparanase regulation. Lysosomal heparanase processing (16,22) and accumulation of the active enzyme in lysosomes, due to the relatively long halflife of the 50-kDa heparanase subunit (14,21), led us to hypothesize that this intracellular enzyme pool could be secreted in response to exogenous stimuli. Inflammatory cytokines, such as tumor necrosis factor-␣ and interleukin 1␤, as well as fatty acids were shown to stimulate secretion of heparanase by endothelial cells (39), although activity data were not provided. Notably, we could not demonstrate enhanced heparanase secretion by tumor necrosis factor-␣ in any of the tumor-derived cell lines included in the present study (data not shown), suggesting that effective stimuli vary among cell types and biological settings. PMA and forskolin effectively stimulated the secretion of enzymatically active heparanase in a PKC-and PKA-dependent manner, respectively (Fig. 1). Importantly, the 50-kDa heparanase subunit was detected in the culture medium by 1 h and peaked 2 h after PMA stimulation, followed by a decline. This elimination of extracellular active heparanase is mediated, among other receptors (15), by cell surface HS and was prevented by the addition of heparin (data not shown), in agreement with our previous identification of HSPG as mediators of heparanase uptake (14). Thus, the availability of extracellular heparanase is tightly regulated by at least two independent mechanisms, one that mediates secretion of the enzyme from intracellular pools and a second that limits accumulation of the secreted enzyme by efficient cellular uptake, together ensuring a limited retention of active heparanase in the extracellular environment.
The presence of the 50-kDa heparanase subunit can also potentially be explained by extracellular processing of the 65-kDa latent enzyme. This, however, does not seem to be the case, since accumulation of the 50-kDa subunit preceded the appearance of the 65-kDa protein, best demonstrated in the kinetics studies for ATP (Fig. 4B). Moreover, the addition of protease inhibitors, including cathepsin inhibitors, to cell cultures had no influence on the levels of the 50-kDa protein found in the medium (data not shown), in  were serum-starved for 20 h, and cells were collected by trypsin and applied onto fibronectin-coated inserts, without (Con; upper panels) or with ATP or ADP (lower panels). Growth medium containing 10% fetal calf serum was added to the lower compartment as chemoattractant, and migrating cells were visualized after 4 h, as described under "Experimental Procedures." agreement with the notion that heparanase processing primarily occurs intracellularly (15,16), by lysosomal enzymes (16,22,23). Several lines of evidence suggest that the secreted heparanase originated from such an intracellular pool, most likely endosomes and lysosomes. Secretion of the 50-kDa protein elicited by PMA and forskolin was rapid and became detectable biochemically within 1 h. In contrast, these reagents did not induce a parallel secretion of the latent 65-kDa protein, suggesting that this effect is not merely due to a general induction of protein secretion but rather to a tightly regulated event. The kinetics of heparanase secretion elicited by PMA and forskolin resembled the secretion kinetics of cathepsin D (Fig. 2), a lysosomal resident enzyme, supporting the notion that both enzymes were secreted from intracellular vesicles. Lysosomal secretion of heparanase is further persistent in terms of the amount being released. By applying iodinated heparanase, we estimated that ϳ4 and ϳ8% of total heparanase is released by 293 and MDA-231 cells, respectively, in response to PMA treatment (data not shown). Such a magnitude of lysosomal secretion is in agreement with Rodriguez et al. (40), who have found that secretion of ␤-hexosaminidase, a lysosomal marker, typically reaches ϳ10% of the total enzyme content of the cell in response to ionomycin treatment. Similarly, ionomycin elicited rapid secretion of the 50-kDa heparanase subunit (data not shown), as would be expected from PMA effectiveness. Moreover, immunofluorescence staining revealed a clear change in the localization of heparanase-positive vesicles toward the cell periphery upon stimulation with ATP (Fig. 4). Thus, although not regarded as typical secretory vesicles, lysosomes may function as such and secrete their content under certain conditions and in response to the proper stimuli (41), in agreement with elevated levels of cathepsins found in several human malignancies (42,43). Regarded as a universal source of metabolic energy, extracellular ATP as well as other nucleotides are capable of initiating signaling cascades through two classes of P2 receptors: P2X, which has an intrinsic activity of ion channel; and P2Y, a G-protein-coupled receptor (44). P2Y receptor activation is coupled to phospholipase C and adenylate cyclase activation, leading to PKC and PKA activation (28,44,45). Remarkably, each and every cell line examined in this study responded to nucleotides by a stimulated secretion of active heparanase, comparable in magnitude and kinetics with PMA and forskolin (Figs. 3 and 4). The differential responsiveness of cell lines to different nucleotides (ATP, ADP, and adenosine) is probably due to different receptor profiles, expressed by each cell type. Importantly, ATP exerts its maximal effect at a physiological concentration (1 M; Fig. 4B) (27), emphasizing the biological relevance of this mediator. Extracellular nucleotides at low micromolar concentrations can influence a number of biological processes, including vascular tone, neurotransmission, cardiac function, and muscle contraction (28,44,46), but their significance in tumor progression and cancer metastasis in unclear. High levels of ATP and ADP were documented during platelet aggregation and at sites of inflammation, where their plasma concentration can transiently reach up to 50 M (46), processes that appear to play important roles in cancer initiation and progression (47). Thus, nucleotides, by inducing P2Y-and PKC-dependent secretion of heparanase, cathepsins, and most likely other lysosomal resident enzymes, may facilitate cancer progression and tumor metastasis (Fig. 7).
Taken together, our results support the notion that heparanase resides and accumulates in the lysosomal compartment, where it may participate in normal turnover of HS. Lysosomal enzymes, including heparanase, can get secreted in response to a proper stimulus, thus maintaining the extracellular levels of ECM-degrading enzymes tightly regulated, preventing tissue damage that may result from an excess of proteolytic and endoglycosidic activities.