Hyaluronidase reduces human breast cancer xenografts in SCID mice
Abstract
A hyaluronan-rich environment often correlate with tumor progression. and may be one mechanism for the invasive behavior of malignancies. Eradication of hyaluronan by hyaluronidase administration could reduce tumor aggressiveness and would provide, therefore, a new anti-cancer strategy. Hyaluronan interaction with its CD44 receptor and the resulting signal transduction events may be among the mechanisms for hyaluronan-associated cancer progression. We have shown previously that hyaluronidase treatment of breast cancer cells in vitro not only eradicates hyaluronan but also modifies expression of CD44 variant exons of tumor cells. We now determine if such effects occur in vivo and if it is accompanied by tumor regression. SCID mice bearing xenografts of human breast carcinomas were given intravenous hyaluronidase. Tumor volumes decreased 50% in 4 days. Tumor sections showed decreased hyaluronan. Intensity of staining for CD44s was not affected, whereas staining for specific CD44 variant exon isoforms was greatly reduced in residual tumors. Necrosis was not evident. Hyaluronidase, used previously as an adjunct in cancer treatment, presumably to enhance penetration of chemotherapeutic drugs, may itself have intrinsic anti-cancer activity. Removing peritumor hyaluronan appears to cause an irreversible change in tumor metabolism. Continuous hyaluronan binding to CD44 variant exon isoforms may also be required to stabilize inherently unstable isoforms that participate perhaps in tumor progression. Further investigation is required to confirm a cause and effect relationship between loss of hyaluronan, changes in CD44 variant exon expression and tumor reduction. If confirmed, hyaluronidase may provide a new class of anti-cancer therapeutics and one without toxic side effects. © 2002 Wiley-Liss, Inc.
Hyaluronan, a non-sulfated glycosaminoglycan polymer, is prominent in the extracellular matrix surrounding cancer cells, as well as normal cells undergoing rapid turnover and movement. During embryonic development, hyaluronan is elevated early, during the phase of migration and expansive growth of pluripotential cells. Degradation of the hyaluronan-rich matrix by hyaluronidase corresponds to the cessation of cell movement and the onset of differentiation.1 Hyaluronan promotes growth. and spread of tumor cells. Levels on the surface of tumor cells often correlate with invasive and metastatic behavior.2, 3 Elimination of pericellular hyaluronan by hyaluronidase may be a mechanism for reducing tumors and slowing malignant progression, as has already been documented in some experimental tumor systems.4, 5 Hyaluronidase may function therefore as a new class of anti-cancer agent. This potential treatment modality would have few toxic side effects. An LD50 could not be achieved in cats and dogs despite administration of increasing doses by either oral or intravenous routes, as indicated by the specifications of the drug insert for a clinical preparation of hyaluronidase (Wydase®).
The large volume of water of hydration associated with hyaluronan creates spaces through which cells move, while simultaneously conferring motility upon cells. Mechanisms for the cell motility involve binding of hyaluronan to its CD44 receptor. The intracellular component of CD44 interacts with the cytoskeleton,6 stimulating motility and various signal transduction pathways.7, 8
CD44, a transmembrane glycoprotein, is the prominent receptor for hyaluronan.9 It exists in a number of isoforms, products of a single gene10 generated by alternative splicing of variant exons inserted into a single extracellular membrane-proximal site.11 Several splice variants of CD44 have been associated with malignant progression, presumably through their ability to enhance the invasive motility of tumor cells. Correlation of particular CD44 variant exons with clinical prognoses are claimed, though no consistent patterns have been observed.12 We postulate that hyaluronan acts in concert with a specific array of variant exons of CD44 in the ability to promote cancer growth and spread and that the putative anti-cancer effect of hyaluronidase might be coordinated with a loss of such CD44 exons together with the eradication of hyaluronan.
We and others have shown in vitro that degradation of cancer cell-associated hyaluronan by hyaluronidase can modulate CD44-splice variant expression.13, 14 Such studies are extended here to in vivo experiments. Hyaluronidase was administered to SCID mice bearing human breast cancer xenografts, to determine the effects on such tumors and to examine the effect on expression of CD44 and its exon splice variants.
MATERIAL AND METHODS
Preparation of hyaluronidase
Bovine testicular hyaluronidase (PH-20) was preparation Type VI-S, a product of the Sigma (St. Louis, MO). The testicular hyaluronidase was purified to near homogeneity using 2 sequential chromatographic steps: Concanavalin-A and Mono-S cation exchange. The purified enzyme had 2 bands on SPAGE and corresponded to 2 bands of activity on hyaluronan-substrate gel electrophoresis,15 as shown in Figure 1 (lane 1). The 2 bands may represent the 2 forms of the enzyme, related to the soluble and membrane-bound forms of PH-20.16, 17 Separation of the 2 forms of the enzyme, the 75 and 58 kDa bands, was achieved by tandem gel filtration on Superose 12. Single bands were obtained on SPAGE, indicating that there was not equilibrium between the 2 forms of the enzyme (Fig. 1, lanes 2,3). The preparation used for administration to animals contained both enzyme isoforms. The purified material had a range of specific activities of from 80,000–105,000 relative turbidity reducing units (rTRU)/mg protein, a unit of enzyme activity that has been described previously.18 In the ELISA-like assay for enzyme activity,19 a standard hyaluronidase preparation from bovine testes, Wydase® (Wyeth-Ayerst Labs., St. Davids, PA) was used as a reference.
thumbnail image
Figure 1. Hyaluronan-SDS substrate gel electrophoresis of purified testicular hyaluronidase. The partially purified enzyme preparation was subjected to substrate gel electrophoresis. Approximately 0.05–0.50 rTRU of hyaluronidase enzyme activity were electrophoresed in the presence of SDS under non-reducing conditions in a 10% polyacrylamide gel containing hyaluronan. After electrophoresis, gels were incubated overnight in Na formate buffer, pH 4.5, in 0.1% Triton X-100 detergent, to substitute the SDS. Gels were then stained for carbohydrate with Alcian blue followed by Coomassie blue. The cleared bands in lanes 1–3 represent the positions of the hyaluronidase that had hydrolyzed the hyaluronan within the gel. The hyaluronidase preparation, purified by sequential Concanavalin-A and Mono S cation exchange chromatographies (lane 1), was further separated into two peaks of activity by Superose 12 gel filtration, lanes 2 and 3, representing proteins of 75 and 58 kDa molecular size, respectively. Lane 4 contains prestained molecular weight markers. The enzyme preparation used in these experiments was material shown in Lane 1, containing the 2 peaks of enzyme activity. The two enzymes contained in the PH-20 preparation may represent PH-20 plus a second-neutral-active enzyme. HYALP1, a pseudogene, is one of the six hyaluronidase-like genes in the human, one of the three on 7q31.3. The mRNA transcription contains an aberrant stop codon, and is not translated into protein. However, active enzyme is present in the mouse, and possibly in other mammals, such as bovine. This observation may also explain why the mouse with a null mutation in the PH-20 gene is fertile. (Baba et al., Mouse sperm lacking cell surface hyaluronidase PH-20 can pass through the layer of cumulus cells and fertilize the egg. J. Biol. Chem, in press).
Cell culture
The human breast cell lines MDA435 was obtained from the American Type Culture Collection. The cells were grown routinely in RPMI-1640 in the presence of 10% FCS in a humidified chamber with 5% CO2. The breast cancer cell line had a hormone-containing mixture added. This culture media addition contained a final concentration of hydrocortisone, 0.5 μg/ml, β-estradiol and progesterone, each at 1 ng/ml, transferrin, 1 ug/ml, triiodo L-thyronine, 10 μM, insulin, 0.5 μg/ml and 2 mM glutamine. Flasks of cells were scraped with a rubber policeman and cells gently centrifuged, rinsed with media and recentrifuged before inoculation. An additional flask was used for cell counts, in which case, cells were trypsinized.
For experiments in which cells were pretreated with hyaluronidase, cells were incubated in this same medium in the presence and absence of purified testicular hyaluronidase, 100 rTRU/ml diluted in the culture medium.
In vivo experiments
For each inoculum, 5 × 106 tumor cells, concentrated by centrifugation, were suspended in media (1:1) with Matrigel (Collaborative Biomed, Bedford, MA). This suspension was inoculated orthotopically into the mammary fat pad of 6-week-old homozygous, female ICR SCID mice (Taconic Farms, Inc., York, ME). Tumors were measured in 3D by digital caliper. All experiments were carried out with IACUC approval.
Histochemistry of hyaluronan
For the histolocalization of hyaluronan, a biotinylated hyaluronan-binding peptide was utilized, derived from a trypsin digestion of bovine nasal cartilage. The peptide was isolated by affinity chromatography using a column of hyaluronan-Sepharose.20 Staining was similar to that described previously.21 Slides were incubated in 3% BSA for 30 min, rinsed with PBS-CMF and then 0.3 ml of the diluted biotinylated hyaluronan binding peptide solution (1:200 of a 0.5 mg/ml preparation) and incubated overnight at 4°C. Slides were then washed for 10 min with calcium- and magnesium-free PBS. Endogenous peroxidase activity was blocked using 0.6% hydrogen peroxide in methanol and incubating for 30 min at room temperature. Slides were rinsed for 20 min in PBS-CMF and then incubated for 45 min with the avidin-labeled horseradish peroxidase solution, prepared as specified by the manufacturer (Vecstatin ABC Peroxidase Kit PK-4000, Vector, Burlington, CA). Slides, washed for 15 min in PBS-CMF, were incubated for five min in peroxidase substrate solution (Peroxidase Substrate Kit, DAB SK-4100, Vector), washed for 5 min in tap water, counterstained with hematoxylin, cleared and mounted with a coverslip. For controls, slides were either preincubated with bacterial hyaluronidase in a solution containing 0.1 M sodium acetate, pH 5.0, 50 mM NaCl, 0.1 mg/ml bovine serum albumin and 100 rTRU/ml Streptomyces hyaluronidase (Calbiochem, San Diego, CA), or by the preincubation of the HABP with excess HA before application to the slides. Each of these controls gave comparable results.
Immunolocalization of CD44
For immunohistochemistry, slides were incubated in 3% goat serum (Vector) in PBS for 30 min at 37°C to block non-specific binding sites. Slides were then incubated with the primary monoclonal antibodies. Two antibodies were used, BM-CD44-UN [BMS 113] for anti-CD44s and BM-CD44-V7V8 [BMS 118] for anti-CD44 v7-8, both from BioSource International (Camarillo, CA), at a dilution of 1:30 in 3% goat serum in PBS-CMF for 1 hr at 37°C and then overnight at 4°C. Slides, washed for 10 min in PBS-CMF, were then incubated with biotinylated goat anti-mouse IgG (Vector) diluted 1:100 with 3% normal goat serum in PBS-CMF for 45 min at room temperature. Slides (were again washed, as above. Endogenous peroxidase activity was blocked as above using 0.6% hydrogen peroxidase in methanol, incubating for 30 min at room temperature. Slides were then processed as described above following the endogenous peroxidase blocking reaction. For negative controls for CD44 v7-8, mouse IgG2A from BioSource International was utilized at a dilution of 1:150 in 3% goat serum in PBS-CMF. Dilutions were used that provided comparable final protein concentrations.
Photographs were taken on an Olympus Vanox AHBT3 Microscope fitted with an integrated Olympus C-35 AD-4 camera, using Kodak Gold Plus 100 film.
Read More http://onlinelibrary.wiley.com/doi/10.1002/ijc.10668/full