The over-expression of HAS2, Hyal-2 and CD44 is implicated in the invasiveness of breast cancer

Experimental Cell Research 310 (2005) 205 – 217 www.elsevier.com/locate/yexcr

Research Article

The over-expression of HAS2, Hyal-2 and CD44 is implicated in the invasiveness of breast cancer Lishanthi Udabage a, Gary R. Brownlee a, Susan K. Nilsson b, Tracey J. Brown a,* a

Laboratory for Hyaluronan Research, Department of Biochemistry and Molecular Biology, Monash University, Wellington Road, Clayton, Victoria 3800, Australia b Stem Cell Laboratory, Level 1, Peter MacCallum Cancer Institute, St Andrew’s Place, East Melbourne 3002, Australia Received 18 February 2005, revised version received 18 July 2005, accepted 21 July 2005 Available online 26 August 2005

Abstract Within tumors there appears to be an intricate balance between hyaluronan (HA) synthesis and degradation where the invading edges display increased HA metabolism. The metabolism of HA has not been characterized in breast cancer cell lines; therefore, this study quantitatively identifies and characterizes the enzymes responsible for the synthesis and degradation of HA while correlating gene expression to cancer cell invasiveness and HA receptor status. In ten well-established breast cancer cell lines, the expression of the genes for each hyaluronan synthase (HAS) and hyaluronidase (Hyal) isoform was quantitated using real-time and reverse transcriptase polymerase chain reaction (PCR). The synthesis and degradation rates of hyaluronan were determined by ELISA, while quantitation of HA receptors, CD44 and RHAMM was performed by comparative Western blotting. The molecular weight of HA synthesized by each HAS isoform and the degradation products of each hyaluronidase were characterized by size exclusion chromatography. It was demonstrated that highly invasive cell lines preferentially expressed the HAS2 and Hyal-2 isoforms, while less invasive cells expressed HAS3 and Hyal-3. There was a correlation between elevated levels of HA synthesis, CD44 expression and cancer cell migration thereby highlighting the pivotal role that HA metabolism plays in the aggressive breast cancer phenotype. D 2005 Elsevier Inc. All rights reserved. Keywords: Hyaluronan; Hyaluronidase; Hyaluronan synthase; CD44; RHAMM and breast cancer

Introduction The extracellular environment of breast tumors consists of a matrix scaffold containing proteins [1,2] and glycosaminoglycans [3]. The major components of the tumoral extracellular matrix (ECM) include laminin, collagen IV, perlecan, entactin, HA, various growth factors and proteases, all of which promote malignancy and/or angiogenesis [4]. The amount and type of these components vary depending on the stage of development and the tissue type. Hyaluronan Abbreviations: Da, Daltons; DX, dextran sulphate; ECM, extracellular matrix; FCS, fetal calf serum; HA, hyaluronan; HABP, HA binding protein; HAS, hyaluronan synthase protein; HAS, hyaluronan synthase gene; Hyal, hyaluronidase protein; Hyal, hyaluronidase gene; Mr, molecular weight; PCR, polymerase chain reaction; Conc, concentration. * Corresponding author. Fax: +61 3 9905 3726. E-mail address: [email protected] (T.J. Brown). 0014-4827/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2005.07.026

(HA) is a ubiquitous ECM component of the tumor environment, especially in the stroma where its accumulation can be observed at the invading edge of breast carcinomas [5,6], in the extracellular environment [7] and as an integral component of the cell-associated matrix of aggressive cancer cells [8,9]. Within a tumor, HA appears to be multi-functional where it maintains hydration homeostasis, provides structural integrity and in conjunction with its receptors has been implicated in the intracellular signaling cascades associated with tumor cell proliferation and migration [10 – 12]. Most malignant solid tumors contain elevated levels of HA [12] where these high levels correlate with poor differentiation and decreased survival [13]. The increased concentration of HA in breast tumors is thought to be the result of fibroblasts being stimulated by the tumor cells to increase HA production [3,14,15]. However, the most tumorigenic and phenotypically aggressive breast carcinoma cell lines also

206

L. Udabage et al. / Experimental Cell Research 310 (2005) 205 – 217

synthesize large quantities of HA, unlike the less malignant cell lines [7]. The invasive potential created by the accumulation of HA may be further aggravated by changes in the expression of HA receptors, CD44 and RHAMM, which are both frequently observed in breast cancer cells [16,17]. The increased levels of HA in breast cancer indicate that HA metabolism is altered, and this perturbation of HA synthesis and/or degradation may play an important role in tumor initiation and progression. Hyaluronan is synthesized by a multi-isoform family of transmembrane glycosyltransferases termed the HA synthases [18] while HA is depolymerized by a combination of enzymic and non-enzymic mechanisms [19]. Hyaluronan polymerization occurs on the inner face of the plasma membrane where it is extruded onto the extracellular surface of the cell [20]. Three eukaryotic HAS isoforms have been identified, termed HAS1, HAS2 and HAS3. Sequence data of the HAS isoforms suggest that they contain seven membrane-associated regions and a central cytoplasmic domain possessing several consensus sequences that are substrates for phosphorylation by protein kinase C [18,21]. The catalytic rate for each HAS isoform is reported to be different [22]. HAS1 is supposedly the least active and drives the synthesis of high molecular weight (Mr) HA (2000 kDa), suggesting low constitutive levels of HA synthesis. HAS2 is more catalytically active and is associated with synthesis of high Mr HA (2000 kDa). HAS2 is implicated in developmental processes involving tissue expansion and growth. HAS3, the most active, drives the synthesis of short (100 –1000 kDa) HA chains. HAS3 expression may be activated to produce large amounts of low Mr HA to contribute to the pericellular matrix or may interact with cell surface HA receptors, triggering signaling cascades and profound changes in cell behavior [22]. Within the tumor environment, the production of high Mr HA is thought to provide a hydrated matrix which forces gaps in the ECM, enabling tumor cells to migrate and metastasize to other tissues. Manipulation of the HAS genes has enabled the overexpression or inhibition of the different HA synthase isoforms which has provided a preliminary insight into the role of HA synthesis in cancer. The upregulation of HAS1 in HA-deficient mouse carcinoma cells restored metastatic ability [23] while transfection of fibroblasts did not increase anchorage-independent growth or the rate of formation of a subcutaneous mass [24]. The over-expression or inhibition of HAS2 has generated more profound results where several studies demonstrated that in a variety of cancer cells, HAS2 is responsible for the generation of an HA pericellular coat, anchorage-independent growth and tumor formation [24 – 26]. The contrary has also been found where very high levels of HAS2 expression can inhibit tumor growth [26]. Similarly, the induction of HAS3 expression resulted in the formation of an HA pericellular coat and promoted the growth of TSU human prostate cancer cells without inducing a metastatic phenotype [27].

Hyaluronan is degraded by a group of enzymes known as the hyaluronidases where they exist in several isoforms, namely, Hyal-1, 2, 3 and PH-20 (for a review, see [19]). Hyal-1 and Hyal-2 are widely distributed and in collaboration with CD44 degrade high Mr HA [28]. It has been suggested that high Mr HA binds to CD44 and in cooperation with the GPI-anchored Hyal-2 [29]. The HA is internalized and degraded to 20-kDa HA fragments within unique acid endocytic vesicles [30]. The fragments are transported intracellularly and further digested by Hyal-1 together with two h-exoglycosidases, h-glucuronidase and h-N-acetyl glucosaminidase resulting in very low Mr oligosaccharides [29]. The bone-marrow-associated Hyal-3 has not been fully characterized. The sperm-associated HAase, PH-20, plays an important role in fertilization and differs from the other HAases by exhibiting enzymic activity at neutral pH [31]. Regulated uptake of tumor-associated HA via a CD44 receptor-mediated endocytosis pathway and subsequent degradation by HYAL-2 may be important for tumor growth and progression where it may play two roles; (i) induction of angiogenesis through the generation of small HA fragments (3 – 25 disaccharide units) that have been shown to promote angiogenesis, cell migration and differentiation of capillary endothelial cells [32,33]; and/or (ii) degradation of HA around blood vessels may also enhance tumor metastasis by enabling tumor cells to enter the circulation more readily [34]. There are contradictory reports about the role of hyaluronidases in cancer, where elevated levels of PH-20 are found in human melanoma, glioblastoma and colon cancer cell lines and in tumor biopsies from colorectal and laryngeal cancer [35,36]. An elevated level of Hyal-1 has been found in prostate cancer [37,38] even though Hyal-1 has been identified as a candidate tumor suppressor [38,39]. The expression of Hyal-1 has been demonstrated to suppress the growth of colon cancer [40] even though it enhances extravasation and metastasis of prostate cancer cells [37]. Hyal-2 has the ability to act as an oncogene where its over-expression in murine astrocytoma cells accelerated tumor formation [41]. Hyal-3 has not been implicated in cancer and to date very little is known about its activity or function. Despite the extensive evidence associating HA with cancer through its ability to promote experimental tumor progression, no direct relationship between the levels of HA synthesis and degradation has been established with respect to the invasiveness of the malignant phenotype, more specifically breast cancer. This study aims at identifying and establishing a relationship between the intricate and continual balance that may occur between the synthesis and degradation of HA in breast cancer while correlating this to the differential expression of the HAS and Hyal isoforms. Through establishing these relationships, it may be possible to highlight the causal role that HA metabolism may play in the initiation and progression of human breast cancer.

5VCCTGCATCAGCGGTCCTCTA 3V 5VCAGTCCTGGCTTCGAGCAG 3V 5VTTGCACTGTGGTCGTCAACTT 3V 5VAAGGTGAAGGTCGGAGTCAAC 3V 5VGCACAGGGAAGTCACAGATGTATGTGC 3V 5VGATGTGTATCGCC-GGTTATCACGCC 3V 5VGCACTGATGGAGGATACGCTGCG 3V

Reverse primer Sense primer

HAS1 HAS2 HAS3 GAPDH Hyal-1 Hyal-2 Hyal-3

Real-time and comparative reverse transcriptase polymerase chain reaction (PCR) were used, respectively, to quantitate the relative mRNA levels of the HA synthases (HAS1 – 3) in the ten human breast cancer cell lines by using gene-specific primers and an internal oligonucleotide probe (Table 1). Total RNA was extracted from triplicate cultures of cells grown to both exponential and plateau phase using RNeasy Mini Kits (QIAGEN, Basel, Switzerland). In brief, total RNA was purified from exponentially growing cells using TRI-reagent (Sigma), which was used to generate single stranded cDNA by incubating 2 Ag RNA with 0.5 Ag/Al random primers and superscript reverse transcriptase (Invitrogen, Carlsbad, CA, USA). For quantitative real-time PCR, gene-specific primers for each HAS isoform and an internal oligonucleotide probe were used. For HAS internal probes, the reporter dye 6-carboxylfluorescein (6-FAM) and quencher 6-carboxytetramethyl rhodamine (TAMRA) was labeled at the 5V and 3V, respectively. For GAPDH internal probes, the reporter 6-FAM was substituted with VICi (Applied Biosystems, Foster City, CA, USA). The PCR reaction was performed in a final volume of 30 Al and

Gene

Quantification of mRNA for HAS1, 2 and 3

Table 1 Primer sequences used for the amplification of the different isoforms of the HAS and Hyal genes

Aneuploid human breast adenocarcinoma cell lines, MDA-MB-231, MDA-MB-435, MDA-MB-468, MDAMB-453, MDA-MB-361, T47D, MCF-7A, BT-549, ZR-751 and Hs578T were obtained from the American Tissue Culture Collection, Rockville, USA. All cell culture propagation reagents were obtained from Sigma, St Louis, MO, USA. Cell lines, MDA-MB-231, MDA-MB-435, MDA-MB468, MDA-MB-453 and MDA-MB-361 were routinely grown and subcultured as a monolayer in 175 cm2 culture flasks in Leibovitz L-15 Medium supplemented with 10% fetal calf serum (FCS) at 37-C. The ZR-75-1 cell line was grown in RPMI Medium supplemented with 10% FCS, 2 mM l-glutamine, 1.5 g/l sodium bicarbonate, 4.5 g/l glucose, 10 mM HEPES, 1 mM sodium pyruvate at 37-C in humiditycontrolled incubator in 5% (v/v) CO2. The T47D cell line was maintained in a humidified incubator at 37-C in 5% CO2 in RPMI supplemented with 10% FCS, 4.5 g/l of glucose, 10 mM HEPES, 1 mM sodium pyruvate and 7.1 Ag/ml of insulin. BT-549 cell line was maintained in a humidified incubator at 37-C in 5% CO2 in RPMI supplemented with 10% FCS and 0.8 Ag/ml of insulin. The Hs578T cell line was cultivated in a humidified incubator at 37-C in 5% CO2 in DMEM supplemented with 10% FCS and 10 Ag/ml of insulin. The MCF-7A cell line was cultivated in a humidified incubator at 37-C in 5% CO2 in MEM supplemented with 10% FCS, 1 mM sodium pyruvate and 10 Ag/ml of insulin. All cell cultures were routinely maintained in media containing antibiotic/antimycotic reagents.

Hybridization probe

Culture of human breast cancer cells

5VGCCGGTCA-TCCCCAAAAG3V 5VTTGGGAGAAAAGTCTTTGGCT 3V 5VGTCGAGGTCAAACGTTGTGAG 3V 5VGAGTTAAAA-GCAGCCCTGGTG 3V 5VCCACTGGTCACGTTCAGGATGAAG-3V 5VCGTAGACTGGGAGTGCATGGTTGGC 3V; 5VGCTGGTGACTGCAGGCCATCGCTGC 3V

Materials and methods

207 5VAACCTCTTGCAGCAGTTTCTTGAGGCC 3V 5VCCATTGAACCAGAGACTTGAAACAGCCC 3V; 5VTCAAATCAAAAACAGGCAGGTACAGGTAGTGG 3V 5VTTTGGTCGTATTGGGCGCCTGG3V

L. Udabage et al. / Experimental Cell Research 310 (2005) 205 – 217

208

L. Udabage et al. / Experimental Cell Research 310 (2005) 205 – 217

consisted of 1 Taqman reaction mix, 6 AM of HAS forward and reverse primer, 1.5 AM of probe, 1 AM of each GAPDH primer and 500 nM of GAPDH probe. PCR amplification was performed by denaturation for 10 min at 95-C followed by annealing for 2 min at 50-C followed by 40 cycles of 15 s at 95-C and 1 min at 60-C. Thermocycling and fluorescence measurement were performed in an ABI Prism 7700 sequence detection system (Applied Biosystems). Relative quantitation was performed by normalizing threshold cycle (Ct) values of each sample gene with Ct values of the GAPDH. UCt corresponds to the difference between the Ct of the HAS genes of interest and the Ct of the GAPDH. Data are presented as fold-change difference relative to parental (arbitrarily set to 100) calculated according to the formula describing relative PCR quantitation 2 (DCtHAS DCtGAPDH). Characterization of mRNA expression for the hyaluronidase family Total RNA extracted from cells in both the exponential and growth-arrested phases were subjected to RT-PCR to determine the hyaluronidase gene expression for Hyal-1, 2 and 3 (Hyal-1– 3). The gene-specific primer sets were designed from sequences retrieved from GenBank (refer Table 1). Amplified sequences were visualized by agarose gel electrophoresis containing ethidium bromide and their identity confirmed by automated DNA sequencing. To quantitate the relative abundance of each PCR product, ethidium bromide-stained agarose gels containing amplified fragments were subjected to densitometric analysis using ProXpressi Imager (Perkin Elmer, Boston, MA, USA) and the data analyzed using Phoretix 1D software (Phoretic International, Newcastle, UK). Quantitation of liberated and cell-associated hyaluronan Triplicate cultures of the human breast cancer cell lines were seeded at 7.5  105 cells/75 cm2 culture flask and were grown with 400 Ag/ml of dextran sulfate (DX; 500 kDa Mr and 17% sulfur substituted; Pharmacia Fine Chemicals, Uppsala, Sweden) and 250 ACi d-[6-3H]glucosamine. This concentration of dextran sulfate inhibits endogenous hyaluronidase activity thus enabling an accurate quantitation and molecular weight characterization of HA production [42]. Cultures were grown for 24 h during which time cell cultures reached 85% confluence and then for a further 24 h until plateau phase was observed. At the conclusion of the incubation period, cells were harvested by trypsinization and counted using a Coulter counter. Media were used for quantitation of the liberated HA. Cell-associated extracellular HA was obtained by centrifugation of the cell/trypsin fraction at 400g av in a Beckman TJ-6 centrifuge where the supernatant was quantitated for HA. Intracellular HA concentration was determined by treating the cell pellet as follows: the cell pellet was lysed under hypotonic conditions by resuspending in 10 mM HEPES pH 7.2 followed by

mechanical disruption in a Dounce homogenizer. Cell lysis was confirmed by Giemsa stain and examination by light microscopy. To dissociate the HA from binding proteins, the cell lysate was heated to 37-C with 0.5% v/v Triton X-114 in 10 mM HEPES buffer pH 7.2 [43]. The HA/detergent micelles were centrifuged at 1500g av for 5 min and the upper aqueous phase was analyzed for HA. The individual analyses of the intra and extracellular HA fractions were not within the detection limits of the HA ELISA (>50 ng/ml); therefore, the extracellular and intracellular fractions were pooled and characterized for HA concentration and Mr. Hyaluronan production was quantitated using an enzymelinked HA binding protein (HABP) assay (Corgenix Inc., Colorado, USA). The assay was performed as directed by manufacturer’s instructions. In brief, duplicate 100 Al of samples and the HA standards (0 –800 ng/ml) were aliquoted into a 96-well plate coated with HABP, incubated for 60 min at room temperature (RT) followed by four washes with PBS. One hundred microliters of HABP conjugated to horseradish peroxidase was added and incubated at for 30 min at RT. After further PBS washes, the reaction was visualized with 100 Al of 3,3V,5,5V-tetramethylbenzidine (TMB) after a 30-min RT incubation. The reaction was sopped with 100 Al of 0.36 N sulfuric acid and read at 450 nm (650 nm reference) in a Bio-Rad 350 microplate reader. Growth media that had not been exposed to cells were used to determine the endogenous HA background, this figure was subtracted from all results. Quantitation of hyaluronan degradation The turnover of liberated and cell-associated HA in human breast cancer cell lines was quantitated by growing the cell lines as previously described with the modification of excluding the dextran sulfate in the growth media. The liberated and cell-associated HA fractions were harvested and quantitated as previously described in this publication. Subtracting the HA concentration in the cell cultures grown in the absence of dextran sulfate from the HA concentration of the identical cell cultures propagated in growth media containing DX (hyaluronidase inhibition growth conditions [42]) enabled the quantitation of the HA before and after enzymic degradation. The daily turnover rate of HA was derived by the following formula: HA Conc in DX growth conditions1 Cell Number 2 HA Conc in DX absent growth conditions1 Cell Number2

1

Media or cell associated fraction collected over a 24 h period. Cell number was determined at time of adding the test media and at time of removing the media enabling determination of cell expansion during the 24 h period. 2

L. Udabage et al. / Experimental Cell Research 310 (2005) 205 – 217

Characterization of the molecular weight of hyaluronan produced by human breast cancer cells Cells were seeded at 7.5  105 cells/75 cm2 culture flask and were grown for 24 h in growth media containing 400 Ag/ml DS and 2 ACi d-[6-3H]glucosamine hydrochloride (Perkin Elmer, Boston, MA, USA). At the conclusion of the 24-h incubation period, the medium was removed and exhaustively dialyzed (Mr exclusion of 6 kDa) against 10 mM Tris – HCl/0.15 M sodium chloride/0.02% sodium azide pH 7.4 at 4-C. The dialysate and dialysis fluid were chromatographically analyzed for the identification of [3H]HA and its degradation products. [3H]HA of >6 kDa was subjected to size exclusion chromatography in a Sephacryl S-1000 gel eluted in 0.15 M NaCl/phosphate pH 7.25, which contained 19 mM NaH2PO4, 38 mM Na2HPO4 and 94 mM NaCl at 13.6 ml/h. The dialysis fluid (molecules <6 kDa) was subjected to size exclusion chromatography in a Superose 12 gel eluted in the abovementioned buffer at an elution rate of 20 ml/h. Molecular weight estimations were calculated using calibration data for HA in Sephacryl S-1000 and Superose 12 data generated from commercially purchased HA fractions of high monodispersity ranging from 10 to 5000 kDa (CPN, Czech Republic and Pharmacia). To determine the percentage incorporation of d-[6-3H]glucosamine hydrochloride into HA macromolecules, the non-dialyzable (molecules >6 kDa) Dpm was subjected to digestion by 10 TRU of Streptomyces hyaluronidase at pH 6, 37-C for 24 h. Digested material was subjected to chromatography in both Sephacryl S-1000 and Superose 12 where profiles were compared to equivalent undigested sample. Any [3H] material not digested by hyaluronidase was excluded from the chromatography profiles. For the calculation of column recoveries, counts in each fraction were taken as significant when >3 SD above the mean background Dpm, with the background determined taking an equal number of sample points before and after Vo and V t , where the average number taken was 20. Visualization of the hyaluronan glycocalyx The HA-dependent pericellular matrix was visualized around the breast cancer cells by the addition of fixed human erythrocytes [44]. In brief, human erythrocytes were fixed overnight in 1.5% v/v formaldehyde in PBS at RT and were then washed exhaustively in PBS. In the last wash, sodium azide was added to a final concentration of 0.1% v/v and the cells stored at 4-C. Breast cancer monolayers were washed twice in PBS, 37-C and then incubated with 5 ml PBS to which 50 Al of fixed erythrocytes (¨108cells/ml) was added. The particles were allowed to settle for 15 –30 min after which the HA-dependent pericellular matrix was recorded by photography on a Nikon Optiflot inverted phase contrast microscope. The specificity of this method was demonstrated by incubation of the breast cancer cultures

209

with Streptomyces hyaluronidase, where cells were covered with 10 U/ml of hyaluronidase followed by incubation at 37-C for 15– 30 min. Monolayers were washed twice in PBS, 37-C and covered with a 5-ml suspension of fixed erythrocytes as previously described. The particles were allowed to settle for 15– 30 min, observed and photographed as described above. Evaluation of breast cancer cell line invasiveness: Boyden chamber migration assay Invasion assays were performed using modified Boyden chambers with polycarbonate nucleopore membrane (Corning, Corning, NY, USA). Pre-coated filters (6.5 mm in diameter, 12-Am pore size, Matrigel 100 Ag/cm2) were rehydrated with 100 Al of Leibovitz L-15 media supplemented with 0.1% w/v BSA (Sigma). Exponentially growing cells were harvested with trypsin/EDTA (Sigma), washed twice with serum-free growth medium containing 0.1% w/v BSA then added to the top chamber (3  105 cells/1 ml chamber). Normal growth media containing 10% v/v FCS were used as the chemoattractant. After incubation for 6 h at 37-C, non-invaded cells on the upper surface of the filter were wiped with a cotton swab, and migrated cells on the lower surface of the filter were fixed and stained with Diff-Quick kit. Invasiveness was determined by counting cells in five microscopic fields per well, and the extent of invasion was expressed as an average number of cells per microscopic field. Each experiment was performed in triplicate on two separate days where data are represented as percentage of migrating cells compared to the parental cell line. Quantitation of hyaluronan receptors, RHAMM and CD44 Cell extracts from exponentially growing cells were prepared by hypotonic lysis in 10 mM HEPES pH 7.2 followed by mechanical disruption in a Dounce homogenizer. Cell lysis was confirmed by Giemsa stain of cell lysate and examination by light microscopy. Cell lysate preparations were denatured at 65-C for 5 min and loaded (15 –30 Ag of protein per lane) onto a 10% polyacrylamide gel. Electrophoresis was performed on a Bio-Rad minigel apparatus. Proteins were transferred to nitrocellulose membranes and blocked for 1 h with Tris-buffered saline containing 5% nonfat dry milk and 0.1% Tween-20. Membranes were then washed and probed with the appropriate antibody diluted in Tris-buffered saline containing 5% bovine serum albumin (for polyclonal antibodies) or 5% non-fat dry milk (for monoclonal antibodies). The antibodies used for detection were 50 Ag of CD44s monoclonal antibody (Hybridoma Bank, USA) or 25 Ag RHAMM (kindly donated by R. Savani, University of Pennsylvania School of Medicine, USA). The secondary antibodies used were anti-rabbit IgG (New England Bio-labs) and rabbit anti-rat IgG (Bio-Rad), which were conjugated with horseradish peroxidase. Immu-

210

L. Udabage et al. / Experimental Cell Research 310 (2005) 205 – 217

Table 2 Quantitation of the HAS isoforms, hyaluronan production, CD44 expression and the invasiveness of human breast cancer cell lines Breast cancer Cell Line

Invasive potential CD44 expression HAS isoform mRNA expressiona Hyaluronan production (fg/cell/24 h) (% of migratory (densitometry HAS2 HAS3 Extracellular Cell associated cells) U/Ag protein) EXP. phaseb EXP. phase PLAT. phase EXP. phase PLAT. phase EXP. phase PLAT. phase

MDA-MB-453 1 MDA-MB-361 2 MDA-MB-468 23 ZRL-75-1 26 T47D 26 MCF-7A 31 MDA-MB-435 48 MDA-MB-231 80 BT-549 92 Hs578T 100

1 0 1.9 0 0.1 0.1 0.9 1.1 1.5 3.6

1 0 0.1 0 0 0.5 0.1 15 92 208

1 1 3 0.5 2 1 0.5 1 8 0.2

0 0 0 0 0 1 0.5 1 8 0.2

590 260 620 640 1520 1620 380 6450 13,090 12,710

140 50 0 630 0 310 90 1140 5280 4570

0 0 0 0 0 0 0 250 130 50

65 80 40 30 70 65 40 350 2790 100

0 indicates where mRNA for each respective gene, hyaluronan or CD44 was not detected. Quantitation of HA represents the average determination from triplicate determinations. % of variance between samples was less than 5%. a HAS expression as determined by real-time RT-PCR where figures are expressed as the fold difference relative to the least invasive cell line MDA-MB 453. Where mRNA for HAS3 was not detected in the plateau phase, differences in gene expression were compared to the HAS3 gene expression of exponentially growing MDA-MB 453 cells. b During plateau phase, the mRNA expression for HAS2 was comparable with that observed in the exponential growth (data not shown).

noreactive bands were detected by enhanced chemiluminescence, and the sizes of proteins were estimated using prestained molecular weight standards followed by densitometric quantitation using ProXpressi Imager (Perkin Elmer, Boston, MA, USA) and the data analyzed using Phoretix 1D software (Phoretic International, Newcastle, UK).

in all breast cancer cell lines, more particularly in the less invasive cell lines, suggests that this HAS isoform is primarily responsible for the synthesis of basal levels of HA production necessary for normal cell function and HAS2 is required for the rapid synthesis of large quantities of HA required for cancer invasion.

Results

The glycocalyx in exponentially growing breast cancer cells is generated by HAS2

Highly invasive breast cancer cells preferentially express HAS2 (Table 2) Endogenous levels of mRNA for the various HA synthase isoforms were quantitated in 10 different human breast cancer cell lines using real-time and comparative RT-PCR. HAS1 mRNA was not detected in any of the ten breast cancer cell lines. HAS2 mRNA was detected in all the breast cancer cell lines that demonstrated an invasiveness of >80% where the highly invasive BT-549 and Hs578T cell lines expressed up to 205 times more HAS2 mRNA than the non-invasive MDAMB 453 cell line. Negligible differences in HAS2 mRNA were observed between exponentially growing and growtharrested cells. All cell lines expressed low levels of HAS3 mRNA, but it was interesting to note that in cells with a low invasive potential (<30% of the cell population exhibited an invasive phenotype) no HAS3 mRNA was detected during the plateau phase, while in the highly invasive cell lines the transcription of this gene continued. The expression of HAS3

Utilizing the HA quantitation and particle exclusion assay, it was possible to uniquely demonstrate that when breast cancer cells were in exponential growth phase, cellassociated HA was only detected in HAS2 expressing cells (Table 2). In the exponentially growing, less invasive breast cancer phenotype that preferentially expressed HAS3, the synthesized HA was not retained as part of the glycocalyx. The retention of the HA as a component of the pericellular matrix only occurred after cells had reached growth arrest. This finding is contrary to studies in other cell types where it was suggested that HAS3 expression resulted in the retention of a pericellular matrix [22,27]. During the plateau phase, in general, the quantity of HA liberated into the media by HAS3 expressing cell lines decreased significantly or in some cases total inhibition of HA liberation was observed, a phenomenon that was concomitant with retention of the HA in the pericellular glycocalyx. During the plateau phase, highly invasive cell lines released 40– 60% less HA into the extracellular environment but retained

Fig. 1. HAS2 and HAS3 liberate high molecular weight HA that is rapidly depolymerized. Breast cancer cells were seeded at 7.5  105 cells/75 cm2 culture flask and were grown for 24 h in growth media containing T 400 Ag/ml DS and 2 ACi d-[6-3H]glucosamine. At the conclusion of the incubation, the media and cell-associated HA are removed and dialyzed (Mr exclusion of 6 kDa). After substantiation that the non-dialyzable Dpm was HA (as determined by Streptomyces hyaluronidase digestion), it was subjected to size exclusion chromatography in a Sephacryl S-1000 gel eluted in 0.15 M NaCl/phosphate buffer, pH 7.25, at 13.6 ml/h. Differences in the Mr of liberated (A, C, E and G) and cell-associated HA (B, D, F and H) was determined in the following cell line: (A, B) MDA-MB 453 cell line; (C, D) MDA-MB 231; (E, F) BT-549 and (G, H) Hs578T. To characterize the HA produced by the cell lines, cultures were treated with 400 Ag/ml dextran sulfate (?-?) and for the identification of HA degradation products the cultures did not contain dextran sulfate (>->).

L. Udabage et al. / Experimental Cell Research 310 (2005) 205 – 217

2- to 22-fold more HA in the pericellular matrix. The quantitation of the cell-associated HA was substantiated by the red cell exclusion assay which only demonstrated that

211

the presence of a pericellular coat was only observed in the exponentially growing MDA-MB 231, BT-549 and Hs578T cell lines (data not shown).

212

L. Udabage et al. / Experimental Cell Research 310 (2005) 205 – 217

HAS 2 and HAS3 liberate high molecular weight hyaluronan which is rapidly depolymerized (Table 3) Chromatographic characterization of the liberated HA (T Streptomyces hyaluronidase digestion) from cells which had undergone both exponential proliferation and growth arrest demonstrated that 80– 98% of the [3H]glucosamine was incorporated into [ 3H]HA, with the remaining [3H]Dpm identified as a pronase digestible macromolecule of approximately 50 kDa. All graphs represented in Figs. 1A – H have had any peaks associated with Streptomyces hyaluronidase-resistant material removed from the profile. The least invasive cell line, MDA-MB 453 which only expressed HAS3 liberated monodisperse HA of 10,000 kDa, while the equivalent sample from cells grown without inhibition of the endogenous hyaluronidase (DX culture), exhibited depolymerization of 22% of the liberated HA into a polydisperse mixture containing 60 –6000 kDa HA species, with 78% degraded to 30 kDa (Fig. 1A). The most invasive cell lines, BT-549 and Hs578T which both primarily expressed HAS2 and a very low expression of HAS3, both liberated large quantities of 10,000 kDa HA which in the presence of active hyaluronidases, was rapidly degraded into the HA fragments of 10, 20 and 40 kDa (Figs. 1E and G). The MDA MB-231 cells that expressed moderate levels of HAS2 and very low levels of HAS3 produced a polydisperse HA that ranged in modal Mr from 600 to 10,000 kDa and smaller fractions at

60 and 200 kDa, while after exposure to endogenous degradation processes these macromolecules were degraded to 20, 40 and 500 kDa (Fig. 1C). Analysis of the cell-associated HA demonstrated that when the normal HA degradation processes were inhibited by DX, a very high Mr HA could be detected as well as oligomers of intermediate Mr, ranging from 20 to 200 kDa. When the normal hyaluronidase-mediated and other potential degradative processes were active, only small fragments of HA ranging from 10 to 70 kDa were found associated with the cell fraction (Table 3). Heightened expression of Hyal-1 and Hyal-2 correlates with the invasive phenotype in human breast cancer cell lines Endogenous levels of mRNA were quantitated for select members of the hyaluronidase family using comparative RTPCR. Hyal-1, 2 and 3 were detected in varying quantities in all cell lines (Table 4). When comparing the expression of Hyal-1, 2 and 3 mRNA in the less invasive cell lines (¨30% of the cell population demonstrating migration), no significant differences in the level of gene transcripts were observed. In contrast, cells displaying a more aggressive invasive phenotype correlated with an increase in gene expression for Hyal-1 and 2. Hyal-2 mRNA was the most abundant transcript and was expressed 5- to 7-fold higher when compared with Hyal-1. Highly invasive cells lines

Table 3 Molecular weight analysis of the hyaluronan produced by the different HAS isoforms and characterization of the Hyal degradation products Breast cancer cell line

MDAMB-453

Invasive HAS mRNAa expression Hyal mRNAb expression potential (% of HAS2 HAS3 Hyal-1 Hyal-2 Hyal-3 migratory cells)

Characterization of extracellular HA

Characterization of cell-associated HA

Modal Mr (kDa)

% of Mr of HA % of Modal Mr % of Mr of HA % of HAc HA degradation HAc (kDa) HAc degradation products products (kDa) (kDa)

10,000

100

36 36 28 100

100

1

1

1

1

1

1

MDA80 MB 231

14

1

30

155

35

BT-549

92

92

8

28

180

5

60 200 600 – 10,000 10,000

Hs578T

100

208

0.2

29

201

0

10,000

25 70 800 2000 6000 20 40 500 10 20 40 10,000 20 40 100 1000 10,000

67 11 3 7 12 37 41 22 42 46 5 7 5 50 30 7 8

60 100 10,000

23 24 53

200 500 660 20 60 10,000

39 38 23 21 41 38

10,000

10 20 40

19 21 60

20 40 70 10 20 40 60 100 10 20 10,000

23 33 44 9 26 46 8 29 44 27

a HAS expression as determined by real-time RT-PCR where figures are expressed as the fold difference relative to the least invasive cell line MDA-MB 453. b Hyal expression as determined by RT-PCR and digitization of bands where figures are expressed as the fold difference relative to the least invasive cell line MDA-MB 453. c 80 – 98% of the [3H]-glucosamine was incorporated into [3H]HA. Remaining [3H]Dpm was identified as a pronase digestible macromolecule of ¨50 kDa.

L. Udabage et al. / Experimental Cell Research 310 (2005) 205 – 217

213

Table 4 Quantitation of the HYAL isoforms, hyaluronan production, CD44 expression and the invasiveness of human breast cancer cell lines Breast cancer Cell Line

MDA-MB-453 MDA-MB-361 MDA-MB-468 ZRL-75-1 T47D MCF-7A MDA-MB-435 MDA-MB-231 BT-549 Hs578T

Invasive Potential (% of migratory cells )

1 2 23 26 26 31 48 80 92 100

CD44 expression (densitometry U/Ag protein)

0 0 1.9 0 0.1 0.1 0.9 1.1 1.5 3.6

Hyaluronidase mRNA expression* Hyal-1

Hyal-2

Hyaluronan turnover** (fg/cell/24 h) Extracellular

Cell associated

EXP. phase

PLAT. phase

EXP. phase

PLAT. phase

Hyal-3 EXP. phase

PLAT. phase

EXP. phase

PLAT. phase

EXP. phase

PLAT. phase

1 1 5 10 16 5 14 30 28 29

0 0 0 0 0 0 16 30 30 35

1 1 5 11 17 5 103 155 180 201

0 0 0 0 0 0 105 158 192 205

1 5 10 14 5 25 75 35 5 0

1 2 10 0 30 5 76 0 0 0

590 260 620 590 1310 1620 380 2020 1480 2990

140 50 0 370 0 310 90 880 910 1760

0 0 0 0 0 0 0 0 60 35

6 45 6 0.8 20 40 20 50 1410 40

0 indicates where mRNA for each respective gene, hyaluronan or CD44 was not detected. Quantitation of HA represents the average determination from triplicate determinations. % variance between samples was less than 5%. * Hyal expression as determined by RT-PCR and digitisation of bands where figures are expressed as the fold difference relative to the least invasive cell line MDA-MB 453. ** Amount of HA degraded in 24 hours. Calculated from the difference between the accumulated HA T dextran sulphate, as previously described [42].

maintained this expression upon reaching plateau phase compared with the least invasive where transcription of Hyal1 and 2 could not be detected. This study has identified the expression of Hyal-3 in the breast cancer cell lines studied. The identification of Hyal-3 in breast cancer cells was unexpected as this gene has been reported in mammalian testis and bone marrow [28], but as yet it has to demonstrate activity in standard hyaluronidase assays [19]. The functional significance of this enzyme in these cell lines is uncertain but would appear not to correlate with cellular invasiveness. The cellular turnover of hyaluronan increases with increased cellular invasion To determine whether differences in the rates of degradation of HA could be correlated with invasive potential, the amount of HA degraded within 24 h was determined (Table 4). Figures in this table represent the net amount of HA degraded within 24 h when compared with the identical cultures grown in the presence of DX. Cells displaying a highly invasive phenotype were capable of degrading more HA than cells displaying a low invasive potential. When comparing the difference between the most aggressive cells with the least invasive, the difference between the degradation rates ranged in the order of 5- to 11-fold. Cells maintaining a low synthetic output of HA (i.e., MDA-MB453) appeared to be in equilibrium with the degradation processes resulting in 100% turnover of HA within 24 h. As the cells displayed a more aggressive invasive phenotype, the synthetic output of the HA potentially exceeded the maximum functioning capacity of the CD44/Hyal-2 degradative pathway thereby culminating in an extracellular environment rich in HA. Increased turnover of HA in the

more aggressive cell lines was consistent with the trends in increased CD44 and mRNA Hyal-2 expression. Heightened expression of CD44 epitope, HAS2, Hyal-1 and Hyal-2 correlates with increased cell invasiveness With the exception of the MDA-MB 468 cell line, when excluding CD44-negative cell lines from linear regression analysis, there was a significant correlation (r 2 = 0.73) between cell invasiveness and CD44 expression (Figs. 2A and B). Quantitation of the RHAMM receptor did not exhibit a strong correlation (r 2 = 0.13) with any particular HAS isoform or the prevalent expression of hyaluronidase or cell invasiveness (Figs. 2C and D). With the exception of the moderately invasive cell line, MDA-MB 468, there was a proportional relationship between HAS2, Hyal-1 Hyal-2, CD44 and breast cancer cell invasiveness (Tables 2 and 4). When examining the catabolic potential of the breast cancer cells, the higher the CD44 and Hyal-2 expression appeared to confer a greater ability to degrade large quantities of HA (Table 4).

Discussion Hyaluronan and hyaluronidases have been proposed to be involved in tumor angiogenesis and invasion where the production of high Mr HA is thought to provide a hydrated micro-niche that facilitates the invasion of tumor cells into the ECM [3], whereas the hyaluronidases degrade the larger polymers of HA into small angiogenesis-inducing oligomers allowing tumor neovascularization [45,46]. Based on the diverse functional differences that HA of differing Mr can exert on cancer cells and within the tumor

214

L. Udabage et al. / Experimental Cell Research 310 (2005) 205 – 217

Fig. 2. Characterization and quantitation of HA receptors and the evaluation of the invasive potential of human breast cancer cell lines. The invasive potential of the breast cell lines (?-?) were examined using the Boyden chamber chemoinvasion assay. The cells that had traversed the Matrigel and spread on the lower surface of the filter were expressed as a percentage of the cell count determined for the Hs578T cell line. The data presented represent the mean T SD average of triplicate experiments performed on two separate days. Note: percentage variance between triplicate determinations <2%. Quantitation of HA receptors (A) CD44 and (B) RHAMM was determined by immunoblotting where immunoreactive bands were quantified by densitometry analysis using ProXpressi Imager and the data analyzed using Phoretix 1D software.

environment, this study has demonstrated that breast cancer cells rapidly synthesize high Mr HA while simultaneously degrading it into fragments that are capable of initiating numerous cellular events such as migration, proliferation and differentiation. This is the first study of its kind to quantitate the different HAS isoforms where the differences in expression have been correlated with the aggressive phenotype of the cancer cell. Several studies have been able to demonstrate that both HAS2 and HAS3 are important in the initiation and progression of the malignant process [23 – 26,47], but no study has been able to compare these genes within a disease state, followed by correlation of their expression to the etiology of the disease. We have demonstrated that in breast cancer HAS3 appears to play a minor role, where all cells require this HAS isoform for the production of low

levels of extracellular HA where upon growth arrest the cell initiates the incorporation of the HA into the glycocalyx of the cell. Equating these observation to the in vivo situation, one could hypothesize that when tumorigenesis is initiated the cancer cells express HAS3, where all of the liberated HA is degraded into small molecular weight HA (<20 kDa) and intermediate sized polymers (200 –500 kDa), which act as a signaling molecules to induce neovascularization [48]. When tumor cells reach critical mass and cell proliferation decreases through hypoxia and nutrient deficiency, the tumor cells could initiate the retention of a pericullar matrix enabling it to become an integrated part of the ECM. Our findings have highlighted that even in non-genetically modified cell lines, the high expression of the HAS2 isoform correlates with the most aggressive forms of cancer, where the synthesized HA is

L. Udabage et al. / Experimental Cell Research 310 (2005) 205 – 217

both liberated in large quantities and retained as a very large pericellular matrix. To date, no studies have examined the invasive edges of tumors to determine which HAS isoform is preferentially expressed. Based on our observations, it would be fair to suggest that the invading front of a breast tumor may express HAS2 where it may play a crucial role in cancer cell proliferation via enhancing the cell’s progression into S and M phase [49] so resulting in cell detachment and enabling aggressive invasion. This in combination with the liberated and extruded HA could form a hydrated pathway for migration of the invasive cells subsequently enhancing metastasis. Several studies have characterized the Mr of HA produced by the three HAS isoforms. It has been reported that HAS2 synthesize larger polymers in the range of 200 to ¨2000 kDa, while HAS3 produces polydisperse HA of 100 –1000 kDa [22,50]. Our findings suggest that this data should be treated with caution for two reasons. Primarily all reports used transfected cell lines as the synthetic source of the HA, and secondly these studies did not inhibit the simultaneous degradation of the newly synthesized HA which could ultimately result in an underestimation of the molecular weight of the synthetic product This study has demonstrated that the MDA-MB 453 cell line which only expressed HAS3 produced a 10,000-kDa monodisperse fraction of HA, but as soon as it was liberated it was depolymerized into the polydisperse fractions which have been previously reported. When characterizing the product of HAS2, this study has once again demonstrated that the majority of the primary product is 10,000 kDa, which too is degraded into the reported smaller Mr species. Therefore, in breast cancer it would appear that the prevalent HAS isoforms generally produce the same Mr product, but it is the concerted action of the hyaluronidases or reactive oxygen intermediates [51] that generates the heterogeneity in the size of HA polymers. Even when the HA degradative processes were inhibited by dextran sulfate, intermediate size HA was detected in the cell-associated fraction. These small fragments of HA could be the result of interrupted chain elongation of HA that was attached to the synthase at the time of cell fractionation. In the least invasive cell lines (<30% of the population has an invasive phenotype), the expression of Hyal-1 and Hyal-2 was only detected during proliferation where 93 – 100% of the liberated HA was degraded or in the process of degradation, but as the cellular proliferation slowed and the plateau phase ensued mRNA for the Hyal-3 isoform continued to be expressed. We demonstrated that when the breast cancer cells were in plateau phase the only detectable hyaluronidase mRNA was Hyal-3, and in this instance the high Mr HA was rapidly degraded into oligomers which would normally be the expected result of Hyal-1 or Hyal-2 activity. Hyal-3 mRNA is widely expressed, but prior to this study its activity had not been demonstrated [19]. In this study, using well-published primer sequences and amplification conditions, we were unable to detect PH-20 even

215

though it has been previously found highly expressed in invasive and metastatic breast cancer [52]. There is evidence to suggest that when breast cancer cell lines become more invasive the balance between catabolic and anabolic processes is purposely altered. The potential modification of HA metabolism concurrent with cancer progression was supported in this study by the dramatic increase in the quantity of synthesized HA as well as the increased capacity for HA turnover and degradation (Table 4). In the absence of dextran sulfate, the more invasive cell lines rapidly degraded the normally synthesized HA into 10 – kDa polymers that persisted in both the extra- and cellassociated environment. The persistence of these low Mr polymers could contribute to an angiogenic switch [53]. The less invasive cell line (MDA-MB 453) did not completely depolymerize the high Mr HA, suggesting that a potential function of such cells is to provide the deposition of an HA matrix that could provide an attachment and hydration pathway enabling cellular nutrients to enter the primary tumor. Subsequently when the tumor phenotype becomes more aggressive, the presence of the small HA fragments that appear to be primarily generated by highly invasive cells commences the angiogenic switch to commence neovascularization and long-term maintenance of the tumor mass. The correlation of high CD44 expression with increased HA metabolism and invasiveness is not a new observation, several studies have demonstrated that CD44 plays a critical role in cancer cell migration [11,54], but this is the first investigation which clearly demonstrates a correlation between CD44, HAS isoform, hyaluronidase expression and HA metabolism. This data potentially add support to the innovative thinking of Stern [19] who proposed a mini-organelle called a hyaluronosome, which is a membrane bound structure, containing HA synthases, hyaluronidases and various HA receptors such as CD44 and RHAMM, an organelle which could regulate the deposition of HA into the ECM. A mini-organelle of this nature could explain the coordinated function of these individual components. The role of HA metabolism in breast cancer has been demonstrated in this study at both the genetic and functional levels, indicating that the fine balance between HA synthesis and degradation plays an integral role in the invasiveness of breast cancer, thereby highlighting potential future biological targets where their modulation may be of therapeutic value in the treatment of cancer.

References [1] J.A. Engbring, H.K. Kleinman, The basement membrane matrix in malignancy, J. Pathol. 200 (2003) 465 – 470. [2] J. Timar, K. Lapis, J. Dudas, A. Sebestyen, L. Kopper, I. Kovalszky, Proteoglycans and tumor progression: Janus-faced molecules with contradictory functions in cancer, Semin Cancer Biol. 12 (2002) 173 – 186.

216

L. Udabage et al. / Experimental Cell Research 310 (2005) 205 – 217

[3] B.P. Toole, Hyaluronan promotes the malignant phenotype, Glycobiology 12 (2002) 37 – 42. [4] D.H. Rohrbach, R. Timpl, Molecular and cellular aspects of basement membranes, Cell Biology: A Series of Monographs, Academic Press, New York, 1993. [5] P. Auvinen, R. Tammi, J. Parkkinen, M. Tammi, U. Agren, R. Johansson, P. Hirvikoski, M. Eskelinen, V.M. Kosma, Hyaluronan in peritumoral stroma and malignant cells associates with breast cancer spreading and predicts survival, Am. J. Pathol. 156 (2000) 529 – 536. [6] M. Jojovic, B. Delpech, P. Prehm, U. Schumacher, Expression of hyaluronate and hyaluronate synthase in human primary tumours and their metastases in scid mice, Cancer Lett. 188 (2002) 181 – 189. [7] K. Kimata, Y. Honma, M. Okayama, K. Oguri, M. Hozumi, S. Suzuki, Increased synthesis of hyaluronic acid by mouse mammary carcinoma cell variants with high metastatic potential, Cancer Res. 43 (1983) 1347 – 1354. [8] K. Ropponen, M. Tammi, J. Parkkinen, M. Eskelinen, R. Tammi, P. Lipponen, U. Agren, E. Alhava, V.-M. Kosma, Tumor cell-associated hyaluronan as an unfavorable prognostic factor in colorectal cancer, Cancer Res. 58 (1998) 342 – 347. [9] L. Seta¨la¨, M. Tammi, R. Tammi, M. Eskelinen, P. Lipponen, U. Agren, J. Parkkinen, E. Alhava, V.-M. Kosma, Hyaluronan expression in gastric cancer cells is associated with local and nodal spread and reduced survival rate, Br. J. Cancer 79 (1999) 1133 – 1138. [10] E.A. Turley, Hyaluronan and cell locomotion, Cancer Metastasis Rev. 11 (1992) 21 – 30. [11] L.Y. Bourguignon, CD44-mediated oncogenic signaling and cytoskeleton activation during mammary tumour progression, J. Mammary Gland Biol. Neoplasia 6 (2001) 287 – 297. [12] B.P. Toole, T.N. Wight, M.I. Tammi, Hyaluronan – cell interactions in cancer and vascular disease, J. Biol. Chem. 277 (2002) 4593 – 4596. [13] P. Lipponen, S. Aaltomaa, R. Tammi, M. Tammi, U. Agren, V.-M. Kosma, High stromal hyaluronan level is associated with poor differentiation and metastasis in prostate cancer, Eur. J. Cancer 37 (2001) 849 – 856. [14] D. Cedeno, R. Stern, Stimulation of hyaluronic acid synthesis in fibroblasts by cocultivation with human breast tumor cell lines, J. Cell Biol. 103 (1986) 101 – 108. [15] S.L. Schor, A.M. Schor, A.M. Grey, J. Chen, G. Rushton, M.E. Grant, I. Ellis, Mechanism of action of the migration stimulating factor produced by fetal and cancer patient fibroblasts: effect on hyaluronic acid synthesis, In Vitro Cell Dev. Biol. 25 (1989) 737 – 746. [16] D. Naot, R.V. Sionov, D. Ish-Shalom, CD44: structure, function, and association with the malignant process, Adv. Cancer Res. 71 (1997) 241 – 319. [17] C. Wang, A.D. Thor, D.H. Moore, Y. Zhao, R. Kerschmann, R. Stern, P.H. Watson, E.A. Turley, The overexpression of RHAMM, a hyaluronan-binding protein that regulates ras signaling, correlates with overexpression of mitogen-activated protein kinase and is a significant parameter in breast cancer progression, Clin. Cancer Res. 4 (1998) 567 – 576. [18] N. Itano, K. Kimata, Molecular cloning of human hyaluronan synthase, Biochem. Biophys. Res. Commun. 222 (1996) 816 – 820. [19] R. Stern, Devising a pathway for hyaluronan catabolism: are we there yet? Glycobiology 13 (2003) 105 – 115. [20] P. Prehm, A. Mausol, Isolation of streptococcal hyaluronate synthase, Biochem. J. 235 (1986) 887 – 889. [21] P.H. Weigel, V.C. Hascall, M. Tammi, Hyaluronan synthases, J. Biol. Chem. 272 (1997) 13997 – 14000. [22] N. Itano, T. Sawai, M. Yoshida, P. Lenas, Y. Yamada, M. Imagawa, T. Shinomura, M. Hamaguchi, Y. Yoshida, Y. Ohnuki, S. Miyauchi, A.P. Spicer, J.A. McDonald, K. Kimata, Three isoforms of mammalian hyaluronan synthases have distinct enzymatic properties, J. Biol. Chem. 274 (1999) 25085 – 25092. [23] N. Itano, T. Sawai, O. Miyaishi, K. Kimata, Relationship between

[24]

[25]

[26]

[27]

[28] [29]

[30]

[31] [32]

[33]

[34] [35]

[36]

[37]

[38]

[39]

[40]

[41]

hyaluronan production and metastatic potential of mouse mammary carcinoma cells, Cancer Res. 59 (1999) 2499 – 2504. N. Itano, F. Atsumi, T. Sawai, Y. Yamada, O. Miyaishi, T. Senga, M. Hamaguchi, K. Kimata, Abnormal accumulation of hyaluronan matrix diminishes contact inhibition of cell growth and promotes cell migration, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 3609 – 3614. R. Kosaki, K. Watanabe, Y. Yamaguchi, Overproduction of hyaluronan by expression of the hyaluronan synthase Has2 enhances anchorage-independent growth and tumorigenicity, Cancer Res. 59 (1999) 1141 – 1145. N. Itano, T. Sawai, F. Atsumi, O. Miyaishi, S. Taniguchi, R. Kannagi, M. Hamaguchi, K. Kimata, Selective expression and functional characteristics of three Mammalian hyaluronan synthases in oncogenic malignant transformation, J. Biol. Chem. 279 (2004) 18679 – 18687. N. Liu, F. Gao, Z. Han, X. Xu, C.B. Underhill, L. Zhang, Hyaluronan synthase 3 overexpression promotes the growth of TSU prostate cancer cells, Cancer Res. 61 (2001) 5207 – 5214. A.B. Csoka, G.I. Frost, R. Stern, The six hyaluronidase-like genes in the human and mouse genomes, Matrix Biol. 20 (2001) 499 – 508. S.K. Rai, F.M. Duh, V. Vigdorovich, A. Danilkovitch-Miagkova, M.I. Lerman, A.D. Miller, Candidate tumor suppressor HYAL2 is a glycosylphosphatidylinositol (GPI)-anchored cell-surface receptor for jaagsiekte sheep retrovirus, the envelope protein of which mediates oncogenic transformation, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 4443 – 4448. R. Tammi, K. Rilla, J.P. Pienimaki, D.K. MacCallum, M. Hogg, M. Luukkonen, V.C. Hascall, M. Tammi, Hyaluronan enters keratinocytes by a novel endocytic route for catabolism, J. Biol. Chem. 276 (2001) 35111 – 35122. G. Kreil, Hyaluronidases: a group of neglected enzymes, Protein Sci. 4 (1995) 1666 – 1669. P. Rooney, S. Kumar, J. Ponting, M. Wang, The role of hyaluronan in tumour neovascularization (review), Int. J. Cancer 60 (1995) 632 – 636. M. Rahmanian, P. Heldin, Testicular hyaluronidase induces tubular structures of endothelial cells grown in three-dimensional collagen gel through a CD44-mediated mechanism, Int. J. Cancer 97 (2002) 601 – 607. T.B. Csoka, G.I. Frost, R. Stern, Hyaluronidases in tissue invasion, Invasion Metastasis 17 (1997) 297 – 311. D. Liu, E. Pearlman, E. Diaconu, K. Guo, H. Mori, T. Haqqi, S. Markowitz, J. Willson, M.S. Sy, Expression of hyaluronidase by tumor cells induces angiogenesis in vivo, Proc. Natl. Acad. Sci. U. S. A. 93 (1996) 7832 – 7837. D.A. Godin, P.C. Fitzpatrick, A.B. Scandurro, P.C. Belafsky, B.A. Woodworth, R.G. Amedee, D.J. Beech, B.S. Beckman, PH20: a novel tumor marker for laryngeal cancer, Arch. Otolaryngol., Head Neck Surg. 126 (2000) 402 – 404. V.B. Lokeshwar, D. Rubinowicz, G.L. Schroeder, E. Forgacs, J.D. Minna, N.L. Block, M. Nadji, L.B. Lokeshwar, Stromal and epithelial expression of tumor markers hyaluronic acid and HYAL1 hyaluronidase in prostate cancer, J. Biol. Chem. 276 (2001) 11922 – 11932. S. Patel, P.R. Turner, C. Stubberfield, E. Barry, C.R. Rohlff, A. Stamps, E. McKenzie, K. Young, K. Tyson, J. Terrett, G. Box, S. Eccles, M.J. Page, Hyaluronidase gene profiling and role of Hyal-1 overexpression in an orthotopic model of prostate cancer, Int. J. Cancer 97 (2002) 416 – 424. T.B. Csoka, G.I. Frost, H.H. Heng, S.W. Scherer, G. Mohapatra, R. Stern, The hyaluronidase gene HYAL1 maps to chromosome 3p21.2p21.3 in human and 9F1-F2 in mouse, a conserved candidate tumor suppressor locus, Genomics 48 (1998) 63 – 70. A. Jacobson, M. Rahmanian, K. Rubin, P. Heldin, Expression of hyaluronan synthase 2 or hyaluronidase 1 differentially affects the growth rate of transplantable colon carcinoma cell tumors, Int. J. Cancer 102 (2002) 212 – 219. U. Novak, S.S. Stylli, A.H. Kaye, G. Lepperdinger, Hyaluronidase-2 overexpression accelerates intracerebral but not subcutaneous tumor

L. Udabage et al. / Experimental Cell Research 310 (2005) 205 – 217

[42]

[43] [44] [45]

[46]

[47]

formation of murine astrocytoma cells, Cancer Res. 59 (1999) 6246 – 6250. L. Udabage, G.R. Brownlee, R. Stern, T.J. Brown, Inhibition of hyaluronan degradation by dextran sulphate facilitates characterisation of hyaluronan synthesis: an in vitro and in vivo study, Glycoconjugate J. 20 (2004) 461 – 471. P. Prehm, Release of hyaluronate from eukaryotic cells, Biochem. J. 267 (1990) 185 – 189. B.J. Clarris, J.R. Fraser, On the pericellular zone of some mammalian cells in vitro, Exp. Cell Res. 49 (1968) 181 – 193. A. Sattar, P. Rooney, S. Kumar, D. Pye, D.C. West, I. Scott, P. Ledger, Application of angiogenic oligosaccharides of hyaluronan increase blood vessel numbers in rat skin, J. Invest. Dermatol. 103 (1994) 576 – 579. P. Rooney, M. Wang, P. Kumar, S. Kumar, Angiogenic oligosaccharides of hyaluronan enhance the production of collagens by endothelial cells, J. Cell Sci. 105 (1993) 213 – 218. M.A. Simpson, C.M. Wilson, L.T. Furcht, A.P. Spicer, T.R. Oegema, J.B. McCarthy, Manipulation of hyaluronan synthase expression in prostate adenocarcinoma cells alters pericellular matrix retention and

[48] [49]

[50]

[51] [52] [53] [54]

217

adhesion to bone marrow endothelial cells, J. Biol. Chem. 277 (2002) 10050 – 10057. P. Rooney, S. Kumar, J. Ponting, M. Wang, The role of hyaluronan in tumour neovascularization, Int. J. Cancer 60 (1995) 632 – 636. M. Brecht, U. Mayer, E. Schlosser, P. Prehm, Increased hyaluronate synthesis is required for fibroblast detachment and mitosis, Biochem. J. 239 (1986) 445 – 450. J. Brinck, P. Heldin, Expression of recombinant hyaluronan synthase (HAS) isoforms in CHO cells reduces cell migration and cell surface CD44, Exp. Cell Res. 252 (1999) 342 – 351. K. Mio, R. Stern, Inhibitors of the hyaluronidases, Matrix Biol. 21 (2002) 31 – 37. D.J. Beech, A.K. Madan, N. Deng, Expression of PH-20 in normal and neoplastic breast tissue, J. Surg. Res. 103 (2002) 203 – 207. J. Folkman, Role of angiogenesis in tumor growth and metastasis, Semin. Oncol. 29 (2002) 15 – 18. M. Culty, M. Shizari, E.W. Thompson, C.B. Underhill, Binding and degradation of hyaluronan by human breast cancer cell lines expressing different forms of CD44: correlation with invasive potential, J. Cell. Physiol. 160 (1994) 275 – 286.