The standard treatment for advanced breast cancer includes doxorubicin (DOX), administered either as monotherapy or in combination with other cytotoxic or targeted drugs [1
]. However, its use has been limited because of dose-limiting toxicities such as cardiotoxicity. Moreover, the generation of drug-resistant tumors after continued therapy is still inevitable, and this eventually narrows down the treatment armamentarium. It is imperative therefore, to keep looking for new drug combinations that can enhance or maintain efficacy, while minimizing toxicity and delaying the development of drug resistance.
Renieramycin M (RM) is a KCN-stabilized tetrahydroisoquinoline purified from the blue sponge Xestospongia
sp. (Figure 1
), with nanomolar IC50
s against the colon, lung, melanoma, and pancreatic cancer cells [2
]. RM induces apoptosis and inhibits invasion and migration in non-small cell lung cancer cells (NSCLC) in vitro, making it a potential antimetastatic agent [8
RM is structurally related to ecteinascidin-743 (Et-743; Trabectedin, Yondelis®
), an anticancer drug for advanced soft tissue sarcoma and recurrent platinum-sensitive ovarian cancer. The renieramycins and ecteinascidins are the two major categories of the 1,2,3,4-tetrahydroisoquinoline alkaloids that have an anticancer effect. This warrants further investigation on the potential clinical utility of RM. A transcriptional structure–activity relationship (SAR) study and molecular network profiling revealed that RM and the ecteinascidin class of compounds induce apoptosis via a common pathway in the colon, breast [2
], and glioblastoma cells [9
]. Et-743 was reported to have a sequence-dependent synergistic effect with paclitaxel in breast carcinoma [10
], and with doxorubicin in soft tissue sarcoma in vitro [11
]. In view of the similarities between RM and Et-743, we hypothesize that RM can act also synergistically with standard cytotoxic drugs and thus, may be potentially useful to improve the therapeutic outcome.
In this study, we investigated the effects of the combination of RM and DOX in estrogen receptor positive (ER+) MCF-7, an in vitro model for the most common type of breast cancer and determined the drug ratio and regimen that will yield a synergistic effect. We also determined the effects of the combination on the cell cycle, apoptosis, and transcriptome in order to gain insights on the mechanism of combinatorial synergy, which could suggest therapeutic strategies for the treatment of breast cancer.
The marine habitat proves to be a rich source of anticancer drugs [39
]. Here we report another marine compound, renieramycin M (RM), from the blue sponge Xestospongia
sp. that synergizes with DOX against ER+
MCF-7, thus offering potential for the treatment of the most common breast cancer subtype. This study serves as an initial attempt to assess the combination effects of RM with other standard cytotoxic agents, targeted, and immunotherapies.
The therapeutic outcome of drug combinations depends on the regimen, dosage or drug ratios, and mode of action of the compounds. For some class of compounds, certain ratios and schedule of administration are synergistic and others are antagonistic. Our results indicate that RM synergizes with pharmacologically achievable concentrations of DOX over a range of combination ratios (1:20–1:80, (RM:DOX)) when administered simultaneously and not sequentially. The synergistic effect was greatest at 1:40–1:50 drug ratio, particularly at 75%–95% kill level, which is most relevant in the clinics. Simultaneous administration of RM and DOX reduced the IC95 of both compounds by two- to eight-fold, hence toxicities are expected to be lower.
We attempted to explain the mechanism of RM and DOX synergy by integrating real-time cell kinetic profiling, cell cycle, and transcriptome analysis. Data showed that RM and DOX damaged the DNA, possibly via different but overlapping mechanism of action. Both compounds repressed genes relating to DSB repair. DOX can (1) intercalate in the DNA minor groove, resulting in positive torsion, thereby inhibiting topoisomerase II (topo II) or (2) directly poison topo II in its double-strand cleavage form and prevent ligation [40
]. RM also repressed topo II, but it was unlikely its main target. The basis of antiproliferative activity of RM is not well established, although mechanistic studies focusing on transcriptional profiling have shown similarities with the structurally related ecteinascidin compounds [2
]. Et-743 binds to the exocyclic N2 amino group of guanines in the DNA minor groove via two of its rings (subunits A and B), forming DNA adducts and bending DNA toward the major groove [42
]. Et-743 also interacts with transcription factors or DNA repair proteins via the third ring (subunit C) that protrudes from the DNA duplex [42
]. In lieu of its chemical structure, it is also possible for RM to have a DNA and a non-DNA target, which would eventually trigger apoptosis. This is supported by the real-time cell profile and the transcriptome analysis revealing that RM downregulated DNA repair proteins and tyrosine kinase signaling proteins involved in ErbB/PI3K-Akt and focal adhesion pathways. The differential regulation of these pathways could be integral to the cytotoxicity of RM. Growth arrest and DNA damage-inducible gene 45 (GADD45A
), a p53-inducible gene involved in the mitotic phase of the cell cycle, was induced by RM, corroborating the results of an earlier report [2
]. However, our results revealed no significant cell cycle arrest, suggesting that other events have led to apoptosis. Possible routes could be through a putative protein target or the non-specific oxidative stress, which can cause DNA-strand breaks, membrane damage, and eventually, cell death [37
]. Saframycin A (Saf A), another closely related tetrahydroisoquinoline with pyruvamide side chain at C-22 instead of an angelate ester [44
], was also shown to alkylate guanine residues in DNA duplexes [45
]. Both compounds offered compelling evidence supporting the involvement of iminium ion intermediate in the process, either by the expulsion of cyanide by Saf A or water by Et-743 [47
]. However, based on gene expression analysis, Saf A did not affect DNA repair genes, as might have been expected if the primary action is through covalent modification of DNA [48
]. Like Et-743, this suggests that Saf A also targets a non-DNA target, which was later on identified to be GAPDH via formation of DNA ternary complexes [49
]. Based on the SAR study, the C-22 angelate ester is important in the cytotoxicity of RM [3
] and may be key to the activation of DNA damage response. SAR studies designed to determine the effects on DNA, coupled with computational modeling will be important in unraveling the moieties that bind to the DNA and putative non-DNA targets.
There are two possible mechanisms therefore to explain the synergy between RM and DOX in MCF-7 breast cancer cells. First, RM and DOX may have acted on the same or related pathway and amplified DNA injury while simultaneously repressing DNA repair pathways. One possible scenario would be accumulation of lesions due to DSBs and SSBs, requiring DNA repair. We hypothesized that these lesions were not repaired due to downregulation of BRCA1 by both compounds. With the simultaneous inhibition of PARP1/2 by DOX, the number of SSBs increased, resulting in a greater replication-associated DSBs, which in turn produced chromatid breaks and exchange aberrations, leading to cell death. Moreover, DOX also downregulated BRCA2, resulting to additional BRCA2-mediated HR repair defects. The mechanism of RM and DOX synergy may be reminiscent of the synthetic lethality in HR-defective breast and ovarian cancer (with BRCA1/2 mutations) treated with PARP inhibitors [27
]. Another possible scenario would be the blocking of the mediators (ATM and ATR), and transducers (CHEK1 and CHEK2) of SSB and DSB repair. CHEK1/2 is responsible for controlling G1/S, S, and G2/M checkpoints. Simultaneous inhibition of CHEK1/2 and other DNA repair proteins may have propagated the DNA damage by permitting the replication of unrepaired DNA, causing genomic instability. Cells that accumulated DNA injury may have progressed through mitosis without arrest for repair and eventually underwent mitotic catastrophe or apoptosis [38
]. Inactivation of anti-apoptotic BCL2 and cytochrome C release upon combination treatment may also have enhanced cell death. Finally, the activation of interferon gamma signaling may indicate immune regulation that may have altered the expression of DNA damage repair proteins and cell cycle regulators [38
Second, RM and DOX may have acted on different targets or pathways that may either complement or reinforce drug action or neutralize compensatory mechanisms [50
]. An example is vinblastine, a tubulin inhibitor that works synergistically with DOX for non-Hodgkin lymphoma [53
] and triple-negative breast cancer [50
]. Here we show that RM is able to downregulate ErbB (possibly through ERBB4), PI3K-Akt, and focal adhesion pathways. Miller et al. [54
] reported that combined inhibition of CDK4 and IGF1R cooperatively suppresses the activation of proteins within the Akt pathway resulting in synergistic cytotoxicity. This might be implicated in the synergism of RM and DOX, as shown by the downregulation of IGF1R by RM, and CDK4 by DOX. Inhibition of PI3K-Akt by RM may also have utility in repressing DOX resistance in melanoma and colon cancer cells [55
]. RM was reported to inhibit the migration and invasion of H460 lung cancer cells, thus may be a potential anti-metastatic agent [8
]. We speculate that downregulation of focal adhesion and integrin signaling via protein tyrosine kinase 2 (PTK2) may be involved. PTK2 or focal adhesion kinase (FAK) is enriched in focal adhesions and together with Src kinase coordinates adhesion turnover, actin cytoskeleton dynamics and cell shape and regulates cancer cell migration and cell invasion [56
]. Another key protein that was recently reported to be involved in breast cancer metastasis is the Myb [57
]. A three- to five-fold increase in repression of MYB was observed after combination, thus providing another basis for its use in the treatment of advanced breast cancer. Currently, it is unclear whether perturbation of ErbB/PI3K-Akt and focal adhesion pathways occur upstream, downstream, or simultaneously with DNA damage response, and whether it may have impact to the synergistic effect of RM + DOX. Liu and Zhao [58
] reported that two compounds with different gene expression profiles may offset each other’s effects when applied together, thus are less likely able to ‘collaborate’ to generate synergistic effects. Further investigation on the cross-talk of these two pathways is needed.
The mechanism of additivity and antagonism after sequential administration is currently unknown. We suspect that pretreatment with DOX have locked the cells at G2/M and caused senescence. Senescent cells tend to be resistant to apoptotic signals and may have reduced sensitivity to RM. Another possibility is that RM action may require dividing cells. A number of DEGs after RM treatment were associated with mitotic cell cycle, hence growth arrested cells may have reduced sensitivity to RM.
Several studies on the individual effects of RM and DOX have been published, however its combination effects have not been explored. To our knowledge, this is the first report on the synergistic cytotoxicity of RM and DOX in a breast cancer cell line with proposed mechanisms based on real-time kinetic profiles, cell cycle effects, and transcriptome signatures. Our study confirms several findings of previous reports and provides new insights to the mechanism of RM alone and the combination with DOX. However, the main drug driving the synergy remains largely unknown and further testing is needed. Although, it is possible that the mechanism of synergy may be explained by more than one mechanism. We found that the combination of RM and DOX also induces synergistic cytotoxicity in mammalian normal canine kidney cells (MDCK). This can be circumvented by synthesizing analogues of RM with improved selectivity. Other strategies such as antibody-drug conjugation [59
] and controlled drug release in tumor tissues at specific ratios are being explored to maximize therapeutic index [60
]. Successful translation of ratio-dependent synergistic drug combinations in pre-clinical and clinical trials using nano-scale drug delivery liposomes has been reported [61
]. Currently, the efficacy of RM and DOX combination is being evaluated in vivo. Phenotypic profiling coupled with transcriptomic profiling of the compounds have aided us in understanding the synergism of RM and DOX. This study adds impetus to investigate the effects of RM with other anticancer drugs—cytotoxic agents, targeted, or immunotherapies, not only in ER+
but also in other subtypes of breast cancer and other solid tumors.