Hyaluronan Oligosaccharides Induce CD44 Cleavage and Promote Cell Migration in CD44-expressing Tumor Cells*

CD44 is an adhesion molecule that serves as a cell surface receptor for several extracellular matrix components, including hyaluronan (HA). The proteolytic cleavage of CD44 from the cell surface plays a critical role in the migration of tumor cells. Although this cleavage can be induced by certain stimuli such as phorbol ester and anti-CD44 antibodies in vitro , the physiological inducer of CD44 cleavage in vivo is unknown. Here, we demonstrate that HA oligosaccharides of a specific size range induce CD44 cleavage from tumor cells. Fragmented HA containing 6-mers to 14-mers enhanced CD44 cleavage dose-dependently by interacting with CD44, whereas a large polymer HA failed to enhance CD44 cleavage, although it bound to CD44. Examination using uniformly sized HA oligosaccharides revealed that HAs smaller than 36 kDa significantly enhanced CD44 cleavage. In particular, the 6.9-kDa HA (36-mers) not only enhanced CD44 cleavage but also promoted tumor cell motility, which was completely inhibited by an anti-CD44 monoclonal antibody. These results raise the possibility that small HA oligosaccharides, which are known to occur in various tumor tissues, promote tumor invasion by enhancing the tumor cell motility that may be driven

CD44 is a widely distributed cell adhesion protein that serves as a cell surface receptor for several extracellular matrix components, including hyaluronan (HA) 1 (1). CD44 participates in various biological processes, such as lymphocyte rolling, tumor cell migration, and invasion (2). The cell surface CD44 is proteolytically cleaved at the extracellular domain by membrane-bound metalloproteases such as MT1-MMP (3); this cleavage has been suggested to play an important role in tumor cell migration along extracellular matrix components (4,5). Indeed, enhanced CD44 cleavage has been reported in invasive tumors such as gliomas, breast carcinomas, non-small cell lung carcinomas, colon carcinomas, and ovarian carcinomas (6). Although several reagents such as phorbol myristate acetate (PMA) (7) and an anti-CD44 monoclonal antibody (mAb), IM7 (8), can potently induce CD44 cleavage in vitro, the physiological inducer of CD44 cleavage in vivo remains uncharacterized.
HA is a nonsulfated linear glycosaminoglycan that consists of repeating disaccharide units of D-glucuronic acid and Nacetyl-D-glucosamine (9). Accumulating evidence indicates that HA not only acts as an extracellular matrix component but also participates in a number of physiological events such as cell adhesion, migration, and proliferation (10). HA also plays a role in pathological conditions, including cancer. An increased synthesis of HA has been reported in various malignant tumors (11,12). HA enhances tumor cell adhesion and migration (13) and activates the Ras-mitogen-activated protein kinase pathway as well as the phosphoinositide 3-kinase pathway (14). HA forms a protective barrier around tumor cells, which may help the tumor cells to evade immune surveillance (15). Although HA usually exists as a high molecular mass polymer (in excess of 1000 kDa) as a component of the extracellular matrix under physiological conditions (9), HA of a much lower molecular mass is detected in association with certain pathological conditions, such as inflammation (16) and tumors (17)(18)(19)(20). It is becoming increasingly clear that the low molecular mass HA fragments can induce a variety of biological events, such as chemokine gene expression (21)(22)(23), activation of transcription factors such as NF-B (24), cell proliferation (19,25), and angiogenesis (26 -29) and that they are more potent than the high molecular mass HA in mediating these biological processes. High levels of angiogenic HA fragments are detected in several human tumors, such as bladder cancers (19), prostate cancers (20), Wilms' tumors (18), and mesothelioma (17), and interestingly, high levels of hyaluronidase activity are also found in bladder and prostate cancer (20,30), raising the interesting possibility that the autoregulatory degradation of HA in tumor tissues may enhance tumor invasion and metastasis.
These reports have led us to speculate that HA and/or its degradation products may be involved in the CD44 cleavage that enhances tumor motility. We herein demonstrate that small HA oligosaccharides but not large HA polymers efficiently induce CD44 cleavage and also promote tumor cell migration in a CD44-dependent fashion. Our results suggest that HA fragments may serve as a physiological inducer of CD44 cleavage in vivo and reinforce the notion that HA degradation products play an important role in tumor invasion.

EXPERIMENTAL PROCEDURES
Reagents-A rabbit polyclonal antibody (pAb), anti-CD44cyto pAb, which is directed against the cytoplasmic domain of CD44, was raised as described previously (4). An anti-human CD44 mAb, BRIC235, was purchased from the International Blood Group Reference Laboratory * This work was supported in part by a grant-in-aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan. 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.
Cell Culture-The human pancreatic carcinoma cell line MIA PaCa-2 was obtained from the Cell Resource Center for Biomedical Research, Institute of Development, Aging and Cancer, Tohoku University (Sendai, Japan). The cells were grown in RPMI 1640 medium (Sigma-Aldrich) supplemented with 10% fetal calf serum, 1% (v/v) 100ϫ nonessential amino acids, 1 mM sodium pyruvate, 2 mM L-glutamine, 50 M 2-mercaptoethanol, 100 units/ml penicillin, and 100 g/ml streptomycin at 37°C in an atmosphere containing 5% CO 2 .
Fluorescein Isothiocyanate Conjugation to BRIC235-The BRIC235 mAb solution was first dialyzed against a 0.1 M sodium carbonate buffer, pH 9.2. A 400-l aliquot of 1.4 mg/ml BRIC235 was incubated with 28 g of fluorescein isothiocyanate (Dojindo, Kumamoto, Japan) for 4 h at room temperature. The solution was applied to a PD-10 column (Amersham Biosciences) equilibrated with PBS containing 0.05% sodium azide. The eluates were monitored by absorbance at 280 nm.
BRIC235 Fab Generation and Purification-BRIC235 Fab fragments were generated and purified using the ImmunoPure Fab preparation kit (Pierce) according to the manufacturer's instructions. Briefly, the BRIC235 antibody solution was extensively dialyzed against a 20 mM sodium phosphate, 10 mM EDTA buffer at pH 7.0. After concentration to ϳ20 mg/ml, the antibody was digested by incubation with immobilized papain for 5 h at 37°C. The Fab fragments were then separated using a protein A column and dialyzed against PBS overnight at 4°C.
Preparation of HA Fragments-The 200-kDa HA (200 mg) was digested with 3,600 units of sheep testicular hyaluronidase at 37°C for 19 h, and the mixture of HA fragments was separated on a Bio-Gel P-10 column (1.5 cm ϫ 100 cm; Bio-Rad) that had been equilibrated with 1 M NaCl containing 10% ethanol. The column fractions were applied to a Sephadex G-25 column (1 cm ϫ 50 cm; Bio-Rad) equilibrated with distilled water for desalting, and the eluates were analyzed by high performance liquid chromatography (HPLC). HPLC analysis was performed with the GULLIVER system (Nippon Bunko Engineering, Tokyo, Japan) on a YMC-Pack PA-03 column (YMC, Wilmington, NC; 4.6 mm ϫ 250 mm) using a programmed linear gradient elution from 16 mM to 1 M NaH 2 PO 4 over 70 min at a flow rate of 1.0 ml/min. The eluates were monitored by absorbance at 210 nm. The HA fragments (10 mg) ranging from HA 6-mers to 14-mers were digested with 0.1 unit of hyaluronidase SD at 37°C for 3 h and then applied to the Sephadex G-25 column equilibrated with distilled water for desalting. The eluates were analyzed by HPLC as described above. In some experiments, hyaluronidase SD was heat-inactivated by boiling before use.
CD44 Cleavage Assay-MIA PaCa-2 cells were seeded into 24-well plates at 5 ϫ 10 4 cells/well, cultured overnight at 37°C, and then incubated with 10 M MG132 for 30 min at 37°C to inhibit the secondary cleavage of the CD44 intracellular domain (32). The cells were then incubated with HA for 1 h or with 100 ng/ml PMA for 30 min at 37°C in the presence of 10 M MG132, and the culture supernatants were collected for analysis by enzyme-linked immunosorbent assay (ELISA). The cells were lysed with SDS sample buffer (2% SDS, 10% glycerol, 0.1 M dithiothreitol, 120 mM Tris-HCl, pH 6.8, 0.02% bromphenol blue) and boiled for 5 min. Samples extracted from equal numbers of cells were separated by electrophoresis on an SDS-polyacrylamide gel and transferred to a polyvinylidene difluoride filter. The filter was blocked in PBS containing 3% BSA and then incubated with anti-CD44cyto pAb or with anti-␤-tubulin mAb. The filters were then incubated with horseradish peroxidase-conjugated anti-rabbit IgG to detect the anti-CD44cyto pAb or with horseradish peroxidase-conjugated anti-mouse IgG to detect the anti-␤-tubulin mAb. The secondary Abs were detected using ECL Western blotting detection reagents (Amersham Biosciences). The densitometric analysis of the bands was performed on an image analysis program (NIH Image; National Institutes of Health, Bethesda, MD). For ELISA, the culture supernatants from the stimulated cells were filtered using a 0.22-m Millipore filter (Millipore Co., Bedford, MA) before analysis. Soluble human CD44 in the samples was quantified using a soluble CD44H ELISA kit (Bender MedSystems, Vienna, Austria) according to the manufacturer's instructions.
Flow Cytometry-The binding of FL-HA to the cell surface was determined as reported previously using an EPICS XL flow cytometer (Coulter, Hialeah, FL) (33). The binding of FL-HA to CD44 was verified by its inhibition with the anti-CD44 mAb BRIC235. The binding of the HAs to the cell surface was analyzed by observing the inhibition of FL-HA binding upon the addition of the HAs as follows. One million cells were incubated with serial dilutions of the HAs for 20 min at 4°C, then 1.0 g/ml of FL-HA was added, and the cells were incubated for another 30 min at 4°C, followed by two washes with cold PBS containing 0.1% BSA. The cells were then analyzed by flow cytometry.
Immunofluorescence Microscopy-MIA PaCa-2 cells were seeded at a concentration of 3 ϫ 10 5 cells/well on 1000-kDa HA-coated cover glasses placed in 6-well plates and incubated overnight at 37°C. The medium was changed to serum-free RPMI medium containing the same supplements as above, and then the cells were incubated with or without 100 ng/ml PMA for 30 min or 6.9-kDa HA for 1 h. In this assay, the cells were not pretreated with a proteasome inhibitor MG132. The cells were then fixed with 4% paraformaldehyde/PBS for 10 min followed by treatment with 0.2% Triton X-100/PBS for 5 min and blocked in PBS containing 1% BSA for 30 min at room temperature. After being washed with PBS, the cells were incubated with the anti-CD44cyto pAb for 1 h at room temperature and then with fluorescein isothiocyanate-conjugated goat anti-rabbit IgG (Jackson ImmunoResearch Laboratories, West Grove, PA) and rhodamine-conjugated phalloidin (Molecular Probes, Eugene, OR) for 1 h at room temperature. After being washed with PBS, the samples were mounted in Prolong Antifade (Molecular Probes) and visualized with an LSM 410 confocal laser scanning microscope (Carl Zeiss, Oberkochen, Germany).
Migration Assay-Cell migration was analyzed using 24-well Costar Transwell chambers (Corning Inc., Corning, NY) containing polycarbonate filters with an 8-m pore size. Both sides of the filter were coated with 100 g/ml of 1000-kDa HA. The lower compartment of the chamber was filled with 600 l of RPMI medium containing 0.1% BSA, and the filters were placed into the chamber. MIA PaCa-2 cells at a logarithmic phase of growth were detached by brief exposure to trypsin-EDTA and resuspended at 2 ϫ 10 5 cells/ml in RPMI medium containing 0.1% BSA. A 100-l aliquot of the cell suspension was added to the upper compartment. After incubation at 37°C for 3 h so that the cells become attached to the filter, the cells were incubated with or without 10 g/ml BRIC235 or mouse IgG in RPMI medium containing 0.1% BSA 20 min prior to and during the migration assay. Finally, HAs or PBS were added to the upper compartment of the wells at the final concentration of 50 g/ml. The chambers were subsequently incubated at 37°C in a 5% CO 2 atmosphere for 24 h. As in the immunofluorescence microscopy, the cells were not treated with MG132. After the cells on the upper side of the filters were gently wiped off, the filters were fixed in methanol, stained with hematoxylin and eosin, and mounted on glass slides. The cells that had migrated to the lower side of the filters were counted under a light microscope. The number of cells in five defined high power fields (ϫ200) was counted, and the average was determined. Each assay was performed five times.

HA Fragments Induce CD44 Cleavage in a Human
Pancreatic Carcinoma Cell Line, MIA PaCa-2-The degradation products of HA have been shown to have a variety of biological activities in CD44-expressing cells (21, 22, 24 -29, 34). We therefore examined their ability to induce CD44 cleavage using a human pancreatic carcinoma cell line, MIA PaCa-2, which has been reported to show substantial CD44 cleavage (3). As shown in Fig. 1, MIA PaCa-2 cells expressed CD44 abundantly (Fig. 1A) and bound FL-HA; this binding was completely blocked by an anti-CD44 mAb, BRIC235, indicating that these cells require CD44 for HA binding. In agreement with previous reports (4, 6, 7), Western blotting with anti-CD44cyto pAb showed that MIA PaCa-2 cells expressed CD44H, the 90-kDa standard form of CD44 (Fig. 1B), and that treatment of these cells with PMA significantly up-regulated the extracellular domain cleavage of CD44, as evidenced by an increase in the membrane-bound 25-kDa cleavage product (Fig. 1B). The en-hancement of CD44 cleavage upon stimulation with PMA was confirmed by ELISA (Fig. 1C).
When MIA PaCa-2 cells were treated with HA fragments that mainly consisted of 6 -14-mers (frHA) ( Fig. 2A), they showed enhanced CD44 cleavage in an HA concentration-dependent manner (Fig. 2B). In contrast, when the cells were treated with the undigested 200-kDa HA, which mainly consisted of polymers of 1000-mers or larger, no increase in CD44 cleavage was observed (Fig. 2C). Basically identical results were obtained using the human glioblastoma cell line U251MG and the human pancreatic carcinoma cell line Panc-1 (data not shown), indicating that small HA fragments, but not large HA polymers, have the ability to enhance CD44 cleavage in various tumor cell types.
Digestion of HA Fragments to Disaccharides Abolishes the HA Fragment-induced CD44 Cleavage-Various HA fragmentdependent phenomena hitherto reported, such as inflammatory cytokine gene expression (21) and inducible nitric-oxide synthase gene induction (35,36), cannot be seen when the HA fragments are digested into disaccharides. This was also the case with the HA fragment-induced CD44 cleavage, as shown in Fig. 3. When completely digested to disaccharides by hyaluronidase derived from S. dysgalactiae (Fig. 3A), the HA fragments up-regulated CD44 cleavage only marginally, if at all (Fig. 3C), whereas the HA preparation that had been treated with heat-inactivated hyaluronidase and hence still contained undigested HA fragments (Fig. 3B) up-regulated the CD44 cleavage to a level comparable with that induced by untreated HA fragments (Fig. 3C). These results clearly indicate that CD44 cleavage is induced by HA fragments but not by HA disaccharides and also exclude the possibility that non-HA molecules that might have been present in the frHA sample induced the CD44 cleavage.
Only HA Fragments of a Certain Size Range Can Induce CD44 Cleavage-To further examine the role of HA size in mediating CD44 cleavage, we used uniformly sized HA fragments ranging from 2-to 12-mers and also the 6.9-kDa HA preparation that contained mainly 36-mers. As shown in Fig. 4 3-7). The cells were lysed, and the samples were subjected to immunoblotting as described in Fig. 1. C, cells were treated as above except that 200-kDa HA was used instead of frHA.
(A and C), HA 2-and 4-mers had only marginal CD44 cleavage up-regulating activity, whereas HA 6-mers and larger fragments induced CD44 cleavage in a manner that was apparently dependent on the size of the fragment. In contrast, HA preparations of much larger sizes, i.e. 36, 200, and 1000 kDa, uniformly failed to up-regulate the CD44 cleavage (Fig. 4, B and C). These results show that only HA fragments of a certain size range can up-regulate the CD44 cleavage.
The HA-induced CD44 Cleavage Is Due to the Interaction between the HA Fragments and CD44 -We next investigated the possibility that HA fragments up-regulate CD44 cleavage by binding to CD44. To this end, we first examined the blocking effect of 6.9-kDa HA on FL-HA binding to MIA PaCa-2 cells, because the binding of small HA fragments to CD44 is not readily detectable by flow cytometry (37). As shown in Fig. 5, the binding of FL-HA to MIA PaCa-2 cells, which was completely blocked by the neutralizing anti-CD44 mAb BRIC235, was significantly blocked by the high molecular mass HAs, such as 36-, 200-, and 1000-kDa HA, as well as 6.9-kDa HA in a concentration-dependent manner, whereas the efficiency of the blocking was greater with the large molecular mass HAs than with 6.9-kDa HA. The fragmented HA that had been used to induce CD44 cleavage in Fig. 2 also inhibited FL-HA binding (data not shown). Consistently, our preliminary studies using surface plasmon resonance technology also confirmed the binding of various HA fragments to a recombinant soluble form of CD44 (CD44-Ig). 2 These results indicate that the HA fragments that can induce CD44 cleavage can bind to CD44.
We then examined whether the up-regulation of CD44 cleavage could be inhibited by blocking the HA binding ability of CD44. For this experiment, we used the Fab fragment of the anti-CD44 mAb BRIC235 to avoid causing CD44 cross-linking, which can itself induce CD44 cleavage (8). As shown in Fig. 6, the fragmented HA-induced CD44 cleavage was almost completely inhibited by the Fab fragment of the BRIC235 mAb, indicating that the HA-induced cleavage was induced by the interaction between CD44 and HA. The 6.9-kDa HA Fragments Promote Migration of MIA PaCa-2 Cells-CD44 cleavage has been reported to play a critical role in the CD44-mediated tumor cell migration that occurs through a highly dynamic interaction between CD44 and the extracellular matrix (3)(4)(5). Therefore, we examined whether 6.9-kDa HA, which was extremely potent in inducing CD44 cleavage, might affect tumor cell motility. MIA PaCa-2 cells stimulated with 6.9-kDa HA developed numerous filopodia and showed actin filament remodeling (Fig. 7, D-F) similar to cells stimulated with PMA (Fig. 7, J-L). In sharp contrast, MIA PaCa-2 cells treated with 1000-kDa HA showed only minimal changes in filopodia formation and actin rearrangement (Fig. 7, G-I), indicating that only the small HA fragments could affect the filopodia formation.
We next examined whether HA preparations of different molecular sizes might differentially affect the migration of MIA PaCa-2 cells using a Boyden-type chamber in which the upper and lower wells were separated by a filter coated with 1000-kDa HA. As shown in Fig. 8, MIA PaCa-2 cells that had been treated with the 6.9-kDa HA migrated efficiently; this migration was completely blocked by the addition of the neutralizing anti-CD44 mAb BRIC235 but not by mouse IgG, indicating that the enhanced migration was dependent on the CD44-HA interaction. In contrast, MIA PaCa-2 cells that had been treated with the 1000-kDa HA did not show any significant changes in 2 K. N. Sugahara, unpublished observation. migration. PMA that enhanced CD44 cleavage and cell motility (Figs. 1 and 7) failed to enhance cell migration in the Boydentype chamber assay (data not shown), indicating that CD44 cleavage itself does not necessarily result in CD44-dependent cell migration. These results collectively demonstrate that the low molecular mass HA can induce not only CD44 cleavage but also CD44-dependent cell migration in tumor cells. DISCUSSION In this report, we demonstrate a novel activity of small HA oligosaccharides. We showed that HA oligosaccharides, especially those ranging from 6-to 36-mers, could up-regulate CD44 cleavage in tumor cells and that neither smaller HA oligosaccharides nor high molecular mass HA had this ability. The CD44 cleavage was apparently induced by the interaction between the small HA oligosaccharides and CD44. In addition, we showed that 6.9-kDa HA, which contained mainly 36-mers, promoted tumor cell migration in a CD44-dependent manner. These results are consistent with the idea that HA of different molecular sizes has different biological activities and support the hypothesis that small HA oligosaccharides serve as a physiological inducer of CD44 cleavage in tumor tissues.
Although HA exists as a high molecular mass (Ͼ1000 kDa) polymer consisting of at least several thousand disaccharide repeats under physiological conditions, HA fragments can be generated in vivo by a variety of mechanisms, such as enzy-matic digestion by hyaluronidases and acid hydrolases, degradation by oxygen-derived free radicals (38), and de novo synthesis of low molecular mass HA (39). Accumulating evidence indicates that HA fragments have biological activities that are not shared by large HA polymers. For instance, HA with an average molecular mass of 250 kDa induce the expression of inflammatory cytokines (40) and chemokines (24) in macrophages, whereas high molecular mass HA molecules fail to do so. The 6.9-kDa HA that was used in the present study has also been shown to enhance Fas expression and Fas-mediated apoptotic cell death in synovial cells (41). The same HA fragments have been shown to activate integrins on colon carcinoma cells (42). Much smaller HA fragments of the tetra-and hexasaccharide size (4-and 6-mers) but not larger HA species induce inflammatory gene expression in dendritic cells (43). 44) and 10 -15-mers (19, 20) induce angiogenesis, and HA 4-mers induce heat shock protein expression (34). However, it has also been reported that HA fragments of comparable sizes can inhibit tumor growth (45) and tumor cell invasion (46). Currently, the reason for this discrepancy is unclear, but differences of the tumor types examined and/or assay types and conditions used may at least in part account for the differences in these observations. In the present study, we showed that HA 6-mers to 36-mers but not fragments larger than 36-kDa (ϳ200-mers) could significantly up-regulate CD44 cleavage (Fig. 4), indicating that HA fragments within a specific size range can induce this event. Complete digestion of the HA fragments to disaccharides abrogated the CD44 cleavage enhancing activity (Fig. 3), confirming the above notion and excluding the possibility that a non-HA contaminating molecule(s) that might have been present in the preparations was responsible for the biological activity. Thus, CD44 cleavage inducing activity can be added to an expanding list of novel biological activities attributed to small HA oligosaccharides.
Why only HA fragments within a specific size range can induce CD44 cleavage requires discussion. The MIA PaCa-2 cells used in the present study express CD44 apparently as a crucial cell surface receptor for HA, because the FL-HA binding by these cells was completely abrogated by the neutralizing anti-CD44 mAb BRIC235 (Fig. 1). In fact, for CD44 cleavage on MIA PaCa-2 cells to occur, HA binding to CD44 appears to be essential, because CD44 cleavage was almost completely inhibited by the Fab fragment of BRIC235 (Fig. 6). However, occupation of the HA-binding site of CD44 per se does not seem sufficient for the induction of CD44 cleavage, because high molecular mass HA that can bind to CD44 on MIA PaCa-2 cells failed to induce CD44 cleavage (Fig. 4). Given that large HA polymers bind to CD44 more avidly than do the small HA oligosaccharides (37), the sheer amount of HA binding is probably not important. In addition, the extent of CD44 crosslinking by HA does not seem to be essential either, because small HA fragments in the 6 -12-mer range, which presumably undergo only monovalent binding to CD44 (37), also induced CD44 cleavage. This speculation is supported by the observations that high molecular mass HA (Fig. 4) and a majority of the anti-CD44 mAbs (data not shown), which are expected to cause significant cross-linking of CD44, also failed to induce CD44 cleavage. Nevertheless, although the cross-linking of CD44 may not be essential, it may still play a role in CD44 cleavage, because HA fragments within a certain size range up-regulated the cleavage apparently in a manner dependent on their molecular size (Fig. 4). Other factors may also need to be considered. For instance, HA is taken up by cells via CD44 (47), and HA fragments of different sizes are likely to differ in the ease by which they are endocytosed, which may differentially affect intracellular signaling. In any case, the mechanism FIG. 8. The 6.9-kDa HA induces cell migration that is inhibited by a blocking anti-CD44 antibody. The effect of 6.9-kDa HA on the MIA PaCa-2 cell migration was assessed by a Boyden chamber-type migration assay. Both sides of the membranes separating the upper and lower chambers were coated with 1000-kDa HA. Cells (2.0 ϫ 10 4 cells/ well) were placed on the upper side of the filters and incubated in the absence of antibody (columns 1, 4, and 7) or in the presence of BRIC235 (columns 2, 5, and 8) or mouse IgG (columns 3, 6, and 9) for 20 min. The cells were then added with culture medium alone (columns 1-3), 6.9-kDa HA (columns 4 -6), or 1000-kDa HA (columns 7-9) and cultured for 24 h in the presence of the antibodies. Columns and bars represent the means and S.D. obtained from five independent experiments. Statistical differences were determined with Student's t test. *, p Ͻ 0.05. by which HA fragments within a specific size range can induce CD44 cleavage requires further investigation.
Okamoto et al. (32) reported that the CD44 extracellular domain cleavage induced by phorbol ester is followed by a CD44 intracellular domain cleavage. The extracellular domain cleavage of CD44 induced by the small HA fragments might also be accompanied by cleavage of the CD44 intracellular domain, because when the CD44 cleavage assay was carried out in the absence of MG132, we could detect a 15-kDa fragment (data not shown) that appeared to represent a CD44 intracellular domain fragment (32). Kajita et al. (3) reported that MIA PaCa-2 cells shed CD44 spontaneously by two different mechanisms, one mediated by MT1-MMP and the other mediated by a serine protease. Judging from the size of the cleavage product (ϳ25 kDa), MT1-MMP is likely to be at least in part involved in the CD44 cleavage. Consistent with this notion, the 6.9-kDa HA-induced CD44 cleavage was also inhibited by MMP inhibitors. 2 HA fragments may thus somehow activate MT1-MMP. In this regard, our observation that 6.9-kDa HA strongly enhanced cell motility (Fig. 8) is reminiscent of the observation by Kajita et al. (3) that the expression of MT1-MMP resulted in CD44 cleavage as well as stimulation of cell motility. The mechanism of the fragmented HA-induced CD44 cleavage requires further investigation.
The small HA oligosaccharide-induced CD44 cleavage may have significant bearing to tumor biology. Certain tumors have high activities of hyaluronidase (20,30), which generates HA fragments. Highly invasive bladder cancers have been shown to produce HA oligosaccharides ranging from 10-to 15-mers that are angiogenic (19). As shown in the present study, small HA oligosaccharides of a similar size can induce CD44 cleavage in tumor cells. In addition, 6.9-kDa HA, which was a potent inducer of CD44 cleavage and CD44-dependent cell migration in our study, can activate tumor cell integrins, resulting in the up-regulation of cell binding to ICAM-1 (42). Therefore, the small HA oligosaccharides may promote tumor invasion not only by inducing angiogenesis (26,44) but also by serving as a signaling molecule to CD44, enhancing the detachment of the tumor cells from the primary tissue, inducing the attachment of the tumor cells to endothelial cells, and increasing cell motility.