Expression of cyclooxygenase-1 and cyclooxygenase-2, syndecan-1 and connective tissue growth factor in benign and malignant breast tissue from premenopausal women

S-nitrosylated and non-nitrosylated COX2 have differential expression and distinct subcellular localization in normal and breast cancer tissue

Immunohistochemical (IHC) staining in breast cancer shows both gain and loss of COX2 expression with disease risk and progression. We investigated four common COX2 antibody clones and found high specificity for purified human COX2 for three clones; however, recognition of COX2 in cell lysates was clone dependent. Biochemical characterization revealed two distinct forms of COX2, with SP21 recognizing an S-nitrosylated form, and CX229 and CX294 recognizing non-nitrosylated COX2 antigen. We found S-nitrosylated and non-nitrosylated COX2 occupy different subcellular locations in normal and breast cancer tissue, implicating distinct synthetic/trafficking pathways and function. Dual stains of ~2000 breast cancer cases show early-onset breast cancer had increased expression of both forms of COX2 compared to postmenopausal cases. Our results highlight the strengths of using multiple, highly characterized antibody clones for COX2 IHC studies and raise the prospect that S-nitrosylation of COX2 may play a role in breast cancer biology.

Cytoplasmic PPARγ is a marker of poor prognosis in patients with Cox-1 negative primary breast cancers

Background

The aim of this study was to investigate the expression of the nuclear receptor PPARγ, together with that of the cyclooxygenases Cox-1 and Cox-2, in breast cancer (BC) tissues and to correlate the data with several clinicobiological parameters including patient survival.

Methods

In a well characterized cohort of 308 primary BC, PPARγ, Cox-1 and Cox-2 cytoplasmic and nuclear expression were evaluated by immunohistochemistry. Correlations with clinicopathological and aggressiveness features were analyzed, as well as survival using Kaplan–Meier analysis.

Results

PPARγ was expressed in almost 58% of the samples with a predominant cytoplasmic location. Cox-1 and Cox-2 were exclusively cytoplasmic. Cytoplasmic PPARγ was inversely correlated with nuclear PPARγ and ER expression, but positively with Cox-1, Cox-2, and other high-risk markers of BC, e.g. HER2, CD133, and N-cadherin. Overall survival analysis demonstrated that cytoplasmic PPARγ had a strong correlation with poor survival in the whole cohort, and even stronger in the subgroup of patients with no Cox-1 expression where cytoplasmic PPARγ expression appeared as an independent marker of poor prognosis. In support of this cross-talk between PPARγ and Cox-1, we found that Cox-1 became a marker of good prognosis only when cytoplasmic PPARγ was expressed at high levels.

Conclusion

Altogether, these data suggest that the relative expression of cytoplasmic PPARγ and Cox-1 may play an important role in oncogenesis and could be defined as a potential prognosis marker to identify specific high risk BC subgroups.

Cyclooxygenase-1 (COX-1) and COX-1 Inhibitors in Cancer: A Review of Oncology and Medicinal Chemistry Literature

Prostaglandins and thromboxane are lipid signaling molecules deriving from arachidonic acid by the action of the cyclooxygenase isoenzymes COX-1 and COX-2. The role of cyclooxygenases (particularly COX-2) and prostaglandins (particularly PGE₂) in cancer-related inflammation has been extensively investigated. In contrast, COX-1 has received less attention, although its expression increases in several human cancers and a pathogenetic role emerges from experimental models. COX-1 and COX-2 isoforms seem to operate in a coordinate manner in cancer pathophysiology, especially in the tumorigenesis process. However, in some cases, exemplified by the serous ovarian carcinoma, COX-1 plays a pivotal role, suggesting that other histopathological and molecular subtypes of cancer disease could share this feature. Importantly, the analysis of functional implications of COX-1-signaling, as well as of pharmacological action of COX-1-selective inhibitors, should not be restricted to the COX pathway and to the effects of prostaglandins already known for their ability of affecting the tumor phenotype. A knowledge-based choice of the most appropriate tumor cell models, and a major effort in investigating the COX-1 issue in the more general context of arachidonic acid metabolic network by using the systems biology approaches, should be strongly encouraged.