Knockdown Delta-5-desaturase in Breast Cancer Cells That Overexpress COX-2 Results in Inhibition of Growth, Migration and Invasion Via a Dihomo-γ-linolenic Acid Peroxidation Dependent Mechanism

Background: Cyclooxygenase-2 (COX-2), the inducible COX form, is a bi-functional membrane-bound enzyme that typically metabolizes arachidonic acid (downstream ω-6 fatty acid) to form 2-series of prostaglandins known to be involved in cancer development. Overexpression of COX-2 has been found in a majority of breast carcinomas, and has also been associated with increased severity and the development of the metastasis. Our lab recently demonstrated that COX-2 can also metabolize dihomo-γ-linolenic acid (DGLA, a precursor of ω-6 arachidonic acid) to produce an anti-cancer byproduct, 8-hydroxyoctanoic acid (8-HOA) that can inhibit growth and migration of colon and pancreatic cancer cells. We thus tested whether our strategy of knocking down delta-5-desaturase (D5D, the key enzyme that converts DGLA to arachidonic acid) in breast cancer cells overexpressing COX-2 can also be used to promote 8-HOA formation, thereby suppressing cancer growth, migration, and invasion.

Methods: SiRNA and shRNA transfection were used to knock down D5D expression in MDA-MB 231 and 4 T1 cells (human and mouse breast cancer cell lines expressing high COX-2, respectively). Colony formation assay, FITC Annexin V/PI double staining, wound healing and transwell assay were used to assess the effect of our strategy on inhibition of cancer growth, migration, and invasion. GC/MS was used to measure endogenous 8-HOA, and western blotting was performed to evaluate the altered key protein expressions upon the treatments.

Results: We demonstrated that D5D knockdown licenses DGLA to inhibit growth of breast cancer cells via promoting formation of 8-HOA that can inhibit histone deacetylase and activate cell apoptotic proteins, such as procaspase 9 and PARP. Our strategy can also significantly inhibit cancer migration and invasion, associated with altered expression of MMP-2/- 9, E-cadherin, vimentin and snail. In addition, D5D knockdown and DGLA supplementation greatly enhanced the efficacy of 5-fluorouracil on breast cancer growth and migration.

Conclusions: Consistent to our previous studies on colon and pancreatic cancer, here we demonstrate again that the high level of COX-2 in breast cancer cells can be capitalized on inhibiting cancer growth and migration. The outcome of this translational research could guide us to develop new anti-cancer strategy and/or to improve current chemotherapy for breast cancer treatment.