Higher cyclooxygenase 2 (COX-2) expression is often observed in aggressive colorectal cancers (CRCs). Here, we attempt to examine the association between COX-2 expression in therapy-refractory CRC, how it affects chemosensitivity, and whether, in primary tumors, it is predictive of clinical outcomes. Our results revealed higher COX-2 expression in chemoresistant CRC cells and tumor xenografts. In vitro, the combination of either aspirin or celecoxib with 5-fluorouracil (5-FU) was capable of improving chemosensitivity in chemorefractory CRC cells, but a synergistic effect with 5-FU could only be demonstrated with celecoxib. To examine the potential clinical significance of these observations, in vivo studies were undertaken, which also showed that the greatest tumor regression was achieved in chemoresistant xenografts after chemotherapy in combination with celecoxib, but not aspirin. We also noted that these chemoresistant tumors with higher COX-2 expression had a more aggressive growth rate. Given the dramatic response to a combination of celecoxib + 5-FU, the possibility that celecoxib may modulate chemosensitivity as a result of its ability to inhibit MDR-1 was examined. In addition, assessment of a tissue microarray consisting of 130 cases of CRCs revealed that, in humans, higher COX-2 expression was associated with poorer survival with a 68% increased risk of mortality, indicating that COX-2 expression is a marker of poor clinical outcome. The findings of this study point to a potential benefit of combining COX-2 inhibitors with current regimens to achieve better response in the treatment of therapy-refractory CRC and in using COX-2 expression as a prognostic marker to help identify individuals who would benefit the greatest from closer follow-up and more aggressive therapy.

SPARC, a matricellular protein with tumor suppressor properties in certain human cancers, was initially identified in a genome-wide analysis of differentially expressed genes in chemotherapy resistance. Its exciting new role as a potential chemosensitizer arises from its ability to augment the apoptotic cascade, although the exact mechanisms are unclear. This study further examines the mechanism by which SPARC may be promoting apoptosis and identifies a smaller peptide analogue with greater chemosensitizing and tumor-regressing properties than the native protein. We examined the possibility that the apoptosis-enhancing activity of SPARC could reside within one of its three biological domains (N-terminus (NT), the follistatin-like (FS), or extracellular (EC) domains), and identified the N-terminus as the region with its chemosensitizing properties. These results were not only confirmed by studies utilizing stable cell lines overexpressing the different domains of SPARC, but as well, with a synthetic 51-aa peptide spanning the NT-domain. It revealed that the NT-domain induced a significantly greater reduction in cell viability than SPARC, and that it enhanced the apoptotic cascade via its activation of caspase 8. Moreover, in chemotherapy resistant human colon, breast and pancreatic cancer cells, its chemosensitizing properties also depended on its ability to dissociate Bcl2 from caspase 8. These observations translated to clinically significant findings in that, in-vivo, mouse tumor xenografts overexpressing the NT-domain of SPARC had significantly greater sensitivity to chemotherapy and tumor regression, even when compared to the highly-sensitive SPARC-overexpressing tumors. Our results identified an interplay between the NT-domain, Bcl2 and caspase 8 that helps augment apoptosis and as a consequence, a tumor's response to therapy. This NT-domain of SPARC and its 51-aa peptide are highly efficacious in modulating and enhancing apoptosis, thereby conferring greater chemosensitivity to resistant tumors. Our findings provide additional insight into mechanisms involved in chemotherapy resistance and a potential novel therapeutic that specifically targets this devastating phenomenon.

Substrate use switches from fatty acids toward glucose in pressure overload-induced cardiac hypertrophy with an acceleration of glycolysis being characteristic. The activation of AMP-activated protein kinase (AMPK) observed in hypertrophied hearts provides one potential mechanism for the acceleration of glycolysis. Here, we directly tested the hypothesis that AMPK causes the acceleration of glycolysis in hypertrophied heart muscle cells. The H9c2 cell line, derived from the embryonic rat heart, was treated with arginine vasopressin (AVP; 1 M) to induce a cellular model of hypertrophy. Rates of glycolysis and oxidation of glucose and palmitate were measured in nonhypertrophied and hypertrophied H9c2 cells, and the effects of inhibition of AMPK were determined. AMPK activity was inhibited by 6-[4-(2-piperidin-1-yl-ethoxy)-phenyl]-3-pyridin-4-yl-pyrrazolo-[1,5-a]pyrimidine (compound C) or by adenovirus-mediated transfer of dominant negative AMPK. Compared with nonhypertrophied cells, glycolysis was accelerated and palmitate oxidation was reduced with no significant alteration in glucose oxidation in hypertrophied cells, a metabolic profile similar to that of intact hypertrophied hearts. Inhibition of AMPK resulted in the partial reduction of glycolysis in AVP-treated hypertrophied H9c2 cells. Acute exposure of H9c2 cells to AVP also activated AMPK and accelerated glycolysis. These elevated rates of glycolysis were not altered by AMPK inhibition but were blocked by agents that interfere with Ca 2ϩ signaling, including extracellular EGTA, dantrolene, and 2-aminoethoxydiphenyl borate. We conclude that the acceleration of glycolysis in AVP-treated hypertrophied heart muscle cells is partially dependent on AMPK, whereas the acute glycolytic effects of AVP are AMPK independent and at least partially Ca 2ϩ dependent. cardiac hypertrophy; energy metabolism; glucose utilization

No abstract