Cranberry extracts have shown in vitro
inhibitory effects in multiple human cancer cell lines (Gutherie et al., 2000
; Fergusan et al., 2004
; Neto et al., 2005
). However, the structural features of proanthocyanidins as related to their cytotoxic activity are poorly defined. Furthermore, the higher sensitivity of oncogenic cell lines relative to non-oncogenic cell lines to PACs is particularly noteworthy and has not been reported. Understanding the mechanisms underlying these observations is essential to ascertain the true therapeutic potential of cranberry PACs. Our interest in the identification of novel therapies from cranberry plant products specifically for human ovarian cancer arises from the potential consumption-linked health benefits to treat or manage this cancer. In the United States, epithelial ovarian cancer (EOC) is the leading cause of death from gynecologic malignancies and the fourth-most-common cause of death due to cancer among women. In 2007, there were an estimated 22 000 new cases and an estimated 15 000 deaths secondary to ovarian cancer (American Cancer Society, 2007
). Progress is being made in treating this devastating cancer but the discovery of new treatments with increased therapeutic activities and decreased dose-related toxicities is needed.
Currently, the development of adjuvant therapies offering additive or synergistic effects is attracting much attention. Dietary phytochemicals could be such agents. They have the potential to increase drug efficacy by either modulating the disease process itself, or affecting specific cellular pathways known to cause resistance to standard cytotoxics. In addition, by acting as chemosensitizers, dietary phytochemicals can reduce the side effects of cytotoxic drugs by simply reducing the dose of drug needed for effect.
The SKOV-3 cell line was chosen for the initial cytotoxicity screening and expanded studies because it is a well-characterized ovarian tumor cell model derived from a highly malignant human ovarian cancer that is platinum-resistant and possesses several key oncogenic characteristics, for example, EGFR over-expression, p53 mutation (Husain et al., 1998
; Anderson et al., 2001
). The other two cell lines were chosen because they are also highly resistant to known therapies. SMS-KCNR cells represent a resistant phenotype and other features of highly therapy refractory neuroblastoma (Singh et al., 2007
). The PC-3 cell line is a well-studied cell line derived from human prostate cancer that is hormone-refractory and resistant to multiple chemotherapeutic agents (Hart et al., 2005
Our in vitro experiments have shown that PAC-1 and PAC-2 display partially overlapping cytotoxic effects in all cell lines tested. SKOV-3 cancer cells were more sensitive to PAC-1 than to PAC-2 or other flavonoid fractions. Even though other fractions (F1–F5) contained oligomers with type-A structural linkages similar to those present in PAC-1, the PAC-1 composition has relatively greater proportions of higher molecular weight polymers (over PAC-2). Additionally, other structural differences may affect the kinetic behavior of PAC-1 in different cell lines, as suggested by differences in the relative sensitivity to PAC-1 versus other fractions in SKOV-3, SMS-KCNR, PC-3, and lung fibroblasts cells. The altered kinetic behavior of PAC-1 could also partially account for its decreased activity in non-cancerous lung fibroblasts that have a similar metabolic rate than the malignant cell lines tested.
The cytotoxicity profiles of the PAC oligomers tested (PAC-1 and PAC-2) also appeared to be different from previously described PAC activities. These differences may be the result of the specific oligomers in our PAC-1/2 extracts due to differences in our material source and processing vs. PAC oligomeric compositions tested by others. Furthermore, the type of linkages in PACs strongly influences their biological activity. A-type linkage containing dimers and trimers were more cytotoxic than dimers and trimers with only B-type linkages (Kolodziej et al., 1995
) against GLC4 lung and COLO 320 colon carcinomas.
Our work also suggests that PACs are effective as chemosensitizers by significantly reducing the dosages of chemotherapeutic agents required to achieve similar or improved efficacy. Pretreatment with PACs allowed lowering of paraplatin dosages while maintaining efficacy. Ultimately, lowering chemotherapeutic drug concentrations should, at the very least, improve toxicity profiles of these drugs.
The unique stereospecific properties of specific components in PAC-1 may allow for more effective targeting of membrane proteins compared with other PAC fractions lacking these properties. Stereospecific properties may account for the apparent synergism with paraplatin in SKOV-3 cells. Qualitatively, PAC-1 appears to impede certain key resistance factors (e.g., affecting platinum drug cellular efflux, decreasing the inactivation of platin by GSH by affecting GSH metabolism, or reversing the inhibition of apoptosis) (Alia et al., 2006
; Perez et al., 1998
; Siddick, 2003
) in SKOV-3 cells. Indeed, even sublethal concentrations of paraplatin achieved significant cytotoxicity and pronounced cell proliferation reduction in SKOV-3 cells in the presence of PAC-1.
The mechanisms by which cranberry proanthocyanidins inhibit tumor cell growth remain unknown. While we were studying PAC-1-induced apoptosis in ovarian cancer cells, published reports revealed that whole cranberry extracts inhibit the expression of matrix metalloproteinases (MMPs) associated with tumor migration and proliferation in the DU-145 prostate cancer cell line (Neto et al., 2005
). The authors attributed the observed cytotoxicity partly to PACs but it is likely that other cranberry components, such as quercetin and ursolic acid, contribute to the observed MMP inhibition. Our data indicate that PAC treatment of ovarian cancer cells results in significant morphological changes consistent with apoptosis. It is possible that MMP inhibition may be involved in the mechanism(s) of action of PAC-1/2
in SKOV-3 cells.
Finally, our experiments also demonstrated that subcytotoxic concentrations of paraplatin may alter the absorption, adherence or intracellular metabolism of PAC-1 in SKOV-3 cells. We noted a significant increase in DP-8, and very significant decrease in DP-11 ( and ), suggesting that DP-11 may be rapidly hydrolyzed, possibly to DP-8 and other PACs but in a non-stoichiometric fashion vs. non-paraplatin-treated cells. Moreover, DP-9 and DP-14 were detected only in cells co-treated with paraplatin/PAC-1, indicating altered oligomer specificity of absorption, which is induced only by combination treatment ( and ). Thus it would appear that sublethal concentrations of paraplatin have a significant impact on the interaction of the cell membrane with PAC oligomers. The recovery of specific oligomers in SKOV-3 cells following co-treatment suggests that the mechanism(s) of action involve specific drug–drug interactions.
As of now it is unknown to us if the consumption of DP-11 and/or the increased production of DP-8 and the apparent differential formation of DP-9 and DP-14 are essential for cytotoxicity. To answer these questions would require the isolation and/or the chemical synthesis of DP-8, DP-9, and DP-14 PACs and assessing their biological activities. This is beyond the scope of this descriptive study but is the interest of future investigations that will specifically examine the structural differences or key requirements and exact mechanism(s) of action of PACs.