Bortezomib is the first in class, proteasome inhibitor that has demonstrated significant anticancer activity in patients with lymphoid malignancies especially multiple myeloma [38, 39]. However, growing studies indicated the potential effectiveness of bortezomib in treatment of patients with solid tumor including colon-gastric cancer [1–3], breast cancer [4–9], prostate cancer [10–14] and lung cancer [15–18]. However, despite its impressive single agent clinical activity in patients with either hematopoietic or solid malignancy, most patients either fail to respond or develop resistance to bortezomib treatment. Therefore, resistance to bortezomib is a challenging problem in the clinic. Identifying mechanism of bortezomib resistance not only can help identify novel therapeutic targets but will also contribute to better utilization of this important therapeutic agent.
In the present study, we focus on the role of survivin and p53 in bortezomib effectiveness as well as their functional relationship in solid tumor cell lines. We found that cancer cells with wild type p53 express much less survivin in comparison with cancer cells with either mutant or null p53. Moreover, bortezomib significantly increased survivin expression in the HCT116 colon or other cancer cell lines with p53 null, while it only showed a minimal effect on survivin expression in HCT116 and other cancer cells with wild type p53. Consistent with these findings, while bortezomib effectively inhibited cell growth and induced cell death in cancer cells with wild type p53, bortezomib showed ineffectiveness to inhibit cell growth and induce cell death for the cancer cells with abnormal p53 (null or mutated). We recognized that our experiment in Fig. 7 will be more convincing, if pairs of cancer cell lines as we have for the HCT116 line (HCT116p53+/+ vs. HCT116p53-/-) could be available to us for these experiment. Nevertheless, the role of survivin in bortezomib resistance was directly demonstrated in the study by silencing of survivin in several cancer cell lines with mutant p53 using survivin mRNA-specific siRNA/shRNA technology previously set up in our laboratory [35, 36]. Finally, our investigations in three different pair of cancer cell lines (originating from breast, lung and prostate) with different p53 status demonstrated that the p53 status-associated survivin expression is an essential parameter to predict bortezomib resistance irrespective of the origin of the cancer cell. Cancer cells having a wild type p53 were sensitive while those with abnormal p53 (mutated or null) were resistant to bortezomib treatment. Consistent with these findings, previous studies found that wild type p53 transcriptionally inhibits survivin expression in various cancer cell types [27–29] and bortezomib can stabilize wild type p53 in prostate cancer cells . Here, we would like to point out that the bortezomib concentration used affects the results, suggesting the dose used in the clinic should be carefully considered. When high dose may kill cancer cells better in a short term, high dose will increase the possibility to generate bortezomib resistance, suggesting that in addition to p53 status-associated survivin expression, other factors, such as other protein members in the IAP and Bcl-2 families may also play important roles in bortezomib resistance. Nevertheless, we have confirmed a role for survivin in bortezomib resistance by direct silencing of survivin expression using survivin-specific siRNA/shRNA. This finding is significant because our recent studies indicated that survivin may be a superior cancer stem cell marker and possibly plays critical role in cancer stem cell expansion . In this regard, cancer cells appear to have a higher percentage of subpopulation cells that are tumorigenic (cancer initiating/cancer stem cells) in xenograft mouse models .
Therefore, consideration of both survivin expression and p53 status as interconnecting biomarkers and targets in cancer cells may not only be useful for predicting the outcome of bortezomib treatment, but may also provide pivotal criteria for rational drug combination. For example, bortezomib likely induces survivin expression in cancer cells with mutated or null p53 (this study), and it is known that paclitaxel rapidly induces survivin expression . Thus, combination of bortezomib and paclitaxel likely obtained no good results in many cancer types with such as the mutated p53 background. Accordingly, a recent Phase II study in patients with metastatic esophageal, gastric, and gastroesophageal cancer showed poor results in the drug combination . However, it is also possible that the poor results derived from such a drug combination involve other mechanisms of drug resistance in these tumors that are notoriously difficult to treat with chemotherapy.
An important question that needs be answered for better application of the findings is the mechanism underlying bortezomib-mediated induction of survivin expression in mutated or null p53 cancer cells, while it showed downregulation of or minimal effect on survivin expression in wild type p53 cancer cells. Although answering this critical question will need further research efforts, based on the current available information, the potential p53 and NF-κB functional crosstalk could provide a plausible explanation, although need to be further confirmed. As reviewed before, the survivin gene is a potential downstream target for p53 and NF-κB transcriptional regulation . Alternatively, the previous finding that bortezomib stabilizes active form of p53 in human LNCaP-Pro5 prostate cancer cells may provide another explanation . Nevertheless, while survivin expression is inhibited by wild type p53 [27–29], survivin and NF-κB appear to be co-expressed in cancer such as in peripheral T-cell lymphoma , and inhibition of NF-κB activity using NF-κB-specific inhibitors decreased survivin expression . Consistent with these observations, bortezomib resistance requires NF-κB activity in mantle cell lymphoma . Therefore, the potential connection of these factors provide an interesting underlying mechanism, which is likely similar to the mechanism we recently discovered for the p53 and ERα on the survivin gene control in the breast cancer .
Finally, the p53 status in RPMI-8226 and Kms11 is not fully consistent in literature. Our literature search indicates that RPMI-8226 has mutant p53 , while Kms11 has wild type p53. However, some publication indicated that Kms11 is p53 null. This is likely due to the hypermethylation of the p53 gene to make p53 expression extremely low . Consistently, our results (Li and Chanan-Khan, unpublished observation) indicated that the expression of p53 in Kms11 was barely detected. Consistent with this, we found that the expression of survivin in Kms11 is comparable with its level in RPMI-8226 (Fig. 3C).