A novel target for therapeutic inhibitors of hyaluronan synthesis - altered membrane topology of HA synthase in breast cancer cells

4611

INTRODUCTION: The progression and invasion of breast cancer is often associated with the over-production of the glycosaminoglycan, hyaluronan (HA). HA is synthesized by a family of transmembrane glycosyltransferases termed HA synthases (HAS). Three eukaryotic HAS isoforms have been found and given the designations HAS1, HAS2 and HAS3. HA is produced by polymerization on the inner face of the plasma membrane; it is then extruded onto the extracellular surface of the cell. There is a direct correlation between HAS2 expression and the progression of breast cancer in vitro. The importance of HAS2 in both the initiation and progression of breast cancer was substantiated when antisense ablation of HAS2 in a highly invasive breast cancer cell line resulted in a 97% inhibition of cell invasiveness. When the HAS2 knock-out cancer cells were implanted into nude mice, tumors did not initiate highlighting the potential therapeutic value of inhibiting a functional hyaluronan synthase in the malignant state. METHODOLOGY: Therapeutic antibodies were developed by designing immunizing peptides to several separate regions of the proposed HAS tertiary structure. Two polyclonal antibodies were raised to predicted cytosolic domains (IN-1 & IN-2) and the third to a putative extracellular domain (EX-1) of the HAS protein. The effect of the HAS antibodies on cell proliferation and hyaluronan synthesis was evaluated in human dermal fibroblasts (DF) and 2 human breast cancer cell lines (MDA-MB-468 and MDA-MB-231) after the cells were cultured in the presence of varying concentrations of each affinity purified HAS antibody. The cellular location of each reactive epitope was determined by FACS and immunohistochemistry of both permeabilized (P) and non-permeabilzed (NP) DF and breast cancer cells. RESULTS: On NP cells antibodies against either IN-1 or IN-2 were non-reactive with dermal fibroblasts whereas anti-EX-1 reacted positively. In both breast cancer cell lines antibodies directed to EX-1 and IN-2 reacted with NP cells, indicating that the epitope for IN-2 localized to the extracellular domain of the HAS. In DF, anti-EX-1 inhibited HA synthesis and induced cell kill whereas antibodies to IN-1 or IN-2 did not. In the malignant state antibodies to EX-1 and also IN-2 inhibited HA synthesis and induced cell kill, whereas anti-IN-1 had no effect. CONCLUSIONS: The HA synthase of non-malignant cells appears to be localized consistent with the predicted topology. In contrast, breast cancer cells appear to extracellularly express what is predicted to be an intracellular epitope suggesting that the membrane topology of HAS is altered in the malignant state. These data warrants further evaluation of inhibitory antibodies specific to this novel epitope which could potentially act as therapeutic inhibitors of hyaluronan synthesis, thereby leading to new therapies for breast cancer.