Background
Camptothecin analogues have been used clinically to treat cancer for almost 20 years. Irinotecan (also known as CPT-11) is used in combination with other antitumor agents as a first-line therapy for metastatic colorectal cancer [
1] and has a history of use as a second-line therapy in advanced gastric and non-small cell lung cancers (NSCLC) [
2,
3]. The second clinically used camptothecin analogue, topotecan, is approved for treatment of ovarian, cervical, and small cell lung cancers [
1]. It has been established in the literature that camptothecin analogues function through inhibition of the topoisomerase I (Top1) enzyme. Camptothecin-class compounds target the DNA-Top1 covalent complex, forming a ternary complex that prevents the dissociation of Top1. This ternary complex inhibits replication and transcription and leads to the formation of double-strand DNA breaks [
4,
5].
Unfortunately, resistance to irinotecan and topotecan is observed in the clinic. Failure of irinotecan- and topotecan-based regimens has been hypothesized to occur through a number of different mechanisms, though only a few are supported with clinical data.
In vitro evidence and limited clinical observations suggest mutations in the
Top1 gene decrease the affinity of the Top1 protein with clinically used camptothecin analogues [
6,
7]. However, based on the literature, likely a more common cause of resistance to irinotecan and topotecan is the increased expression of ATP-binding cassette (ABC), subfamily G, isoform 2 protein (ABCG2, also known as breast cancer resistance protein, BCRP), a drug efflux pump and a member of the ABC transporter superfamily [
8]. A number of clinical studies revealed that failure of irinotecan and topotecan often correlates with increased ABCG2 expression [
9,
10]. Multiple
in vitro studies have demonstrated that irinotecan, SN-38 (active metabolite of irinotecan), and topotecan are all substrates for ABCG2, and high expression of ABCG2 is associated with decreased intracellular accumulation of these compounds and consequentially a decrease in drug potency [
11,
12]. Additionally, many other anticancer agents are known ABCG2 substrates, including methotrexate [
13], many anthracyclines [
14], and a variety of tyrosine kinase inhibitors [
15,
16].
Our lab recently reported on a novel camptothecin derivative, designated FL118 [
17,
18]. The chemical name of FL118 is 10,11-methylenedioxy-20(S)-camptothecin, also known as 10,11-MD-CPT, MDCPT [
19], and 10,11-mCPT [
20] (Additional file
1: Figure S1). FL118 shows strong anticancer activity in several different cancer types
in vitro and
in vivo [
17,
18]. We have demonstrated that although FL118 is not a better Top1 inhibitor than clinically used camptothecin analogues [
17,
18], FL118 is able to selectively inhibit the expression of several members of the Inhibitor of Apoptosis family (survivin, XIAP, and cIAP2) and the Bcl-2 family (Mcl-1), which was demonstrated to contribute to FL118 function and anti-cancer activity [
18,
21]. More recent studies have further characterized the novel properties of FL118. Induction of cancer cell senescence and cell death by FL118 employs both p53-dependent and p53-independent signaling pathways, and rapid induction of wild type p53 accumulation by FL118 is largely independent of the ATM-dependent DNA damage signaling pathway but dependent on E3-competent Mdm2 [
22]. Our previous studies also revealed that, while mice showed continuing body weight loss after treatment with irinotecan, body weight rapidly recovers after the completion of FL118 treatment [
18,
21], suggesting that FL118 possesses a more favorable toxicity profile in comparison with irinotecan.
In the present study we found that, although SN-38 and topotecan are ABCG2 substrates and fail to overcome ABCG2-mediated drug resistance, FL118 is insensitive to ABCG2 expression and effectively bypasses ABCG2 resistance. FL118 also demonstrates better antitumor efficacy than irinotecan in human xenografts with high ABCG2 expression. Additionally, we found that the relatively nonpolar nature of FL118 plays a role in bypassing ABCG2-induced resistance.
Discussion
The present study expands the uniqueness of FL118 in mechanism of action to overcome drug resistance, and demonstrated that, in a panel of colon cancer and NSCLC cell lines, FL118 is more potent than SN-38. We found that, in contrast to the two clinically used camptothecin analogues (irinotecan, topotecan), which are ABCG2 substrates and unable to overcome ABCG2 resistance, FL118 potency is not affected by ABCG2 expression and can bypass ABCG2-mediated treatment resistance. This phenomenon was investigated and confirmed by multiple independent approaches to either inhibit (i.e., pharmacological and genetic inhibition) or enhance (i.e., overexpression) ABCG2 activity. These approaches demonstrated that high ABCG2 activity results in resistance to SN-38 and topotecan, but not FL118.
We also assessed the efficacy of FL118 in two distinct in vivo xenograft models of ABCG2-mediated drug-resistant cancer. In both models, FL118 exhibited a better ability to decrease tumor growth in comparison with irinotecan, while maintaining a tolerable toxicity profile similar to that of irinotecan. Most importantly, mice treated with FL118 showed a significant increase in time to progression (TTP) compared to mice treated with irinotecan.
Our group recently reported on the
in vivo efficacy of FL118 in other models of human cancer, using an intravenous (IV)-compatible, Tween/polysorbate 80-free formulation [
29]. In that study, we compared three schedules (every day for five injections, every other day for five injections and once weekly for 4 injections) of FL118 administration via IV administration. We found that the optimum administration of FL118 appears to be every other day. However, for this study, we opted for a weekly schedule that mimics irinotecan administration in the clinic. Therefore, while our results in this study revealed that the weekly administration of FL118 was able to significantly extend TTP in comparison with irinotecan, we predict that treating with FL118 every other day for five treatments will show even better efficacy. Additionally, in the current studies, we used IP routes instead of intravenous administration of FL118 for technical convenience. However, the Tween/polysorbate 80-free formulation of FL118 can be administrated via IV routes with increased maximum tolerated dose (MTD). This may also result in better
in vivo tumor inhibition outcomes for FL118. We intend to perform these experiments with optimal routes and schedules in our follow-up studies as part of a broader goal of optimizing FL118 administration to prepare for clinical applications.
It was reported that the affinity of camptothecin analogues to ABCG2 is influenced by polar additions to the A ring [
27,
28]. We hypothesize that the polarity of B ring-substituted functional groups of FL118 may also affect ABCG2 binding. To test this hypothesis, we modeled the distinct electrostatic potential of six FL118 analogues with different chemical groups on the 7-position of the B ring versus FL118 itself, and assessed their potency using cell proliferation assays. As before, we used potency in the presence and absence of ABCG2 inhibition as a surrogate for assessing the affinity of individual drugs for ABCG2. We saw a significant change in EC
50 in the presence of Ko143 for FL118 analogues with stronger electronegative groups, indicating that these FL118 analogues are substrates of ABCG2. We also noted that some structures with less polar functional groups (e.g., 7-ethyl-FL118) still have regions of localized electronegativity in those groups (indicated by the red gradient) (Additional file
1: Figure S5), while some structures with more polar groups lack such regions of moderate localized electronegativity (e.g., 7-chloro-FL118, Additional file
1: Figure S5). Comparing these electrostatic potential maps to our results in Table
2, we propose that these localized electronegative regions are less predictive of a compound’s affinity for ABCG2 than the Pauling electronegativity of the molecule as a whole.
We recently reported a novel IV-compatible Tween/polysorbate 80-free formulation for FL118, which could increase FL118 MTD 3-7 fold in comparison with FL118 formulated using the previous Tween/polysorbate 80-containing recipe [
29]. We found that in this new formulation, while FL118 showed solubility similar to 7-bromomethyl-FL118, 7-chloromethyl-FL118 and 7-hydroxymethyl-FL118, the solubility of 7-methyl-FL118, 7-ethyl-FL118 and 7-allyl-FL118 was much poorer than FL118. Similarly, FL118 analogues with other 7-substituted nonpolar groups (e.g., cycloalkyl, aryl) also showed poor solubility in our Tween/polysorbate 80-free formulation recipes with or without 5% DMSO. Therefore, future studies related to the structural modification of the FL118 scaffold should be based on the findings revealed from these compounds to further optimize potency and drug-like properties, for example by adding hydrophilic group on positions of 5, 9 or 12. Alternatively, we may generate the pro-drug for potent compounds, such as dipeptide derivatives to increase water solubility. Working in these directions may allow us to generate compounds with even better therapeutic index (TI, i.e. ratio of antitumor activity versus toxicity) than the favorable TI of FL118 [
29].
Here, it should be pointed out that the relationship between the affinity of camptothecin analogues to ABCG2 and the strength of electronegative charges of distinct chemical groups on the B ring has not been investigated previously, although the B ring of camptothecin is often modified in order to improve the anticancer potency and pharmacological properties [
25,
26]. While there was a positive correlation between χ and ABCG2-induced resistance, the large loss of potency for 7-bromomethyl-FL118 compared to 7-chloromethyl-FL118 suggests that functional group electronegativity may be only one characteristic that influences ABCG2 affinity. Since ABCG2-mediated resistance to anticancer drugs, including irinotecan and topotecan, is a recognized problem in the clinic, further understanding of the structure-activity relationship of FL118 analogues for ABCG2 binding may lead to rational design of better anticancer agents for clinical application.
Interestingly, cabazitaxel, a taxane derivative with poor affinity for another drug-efflux pump protein, P-glycoprotein 1 (P-gp, also known as multidrug resistance protein 1 and ABCB1), was recently approved for use in patients with castration-resistant prostate cancer who had previously failed docetaxel-based regimens [
30]. It is thought that cabazitaxel’s lack of affinity for P-gp plays an important role in its effectiveness in docetaxel-refractory cancer. In keeping with this rationale, the work presented here suggests that this strategy may be useful in other cancer types and with other classes of cytotoxic agents, including camptothecins. Due to FL118’s superior anticancer activity, favorable tolerability, and insensitivity to ABCG2, we posit that FL118 may become a better option for targeted cancer therapeutics to address the increasingly complex issue of drug failure by circumventing multiple mechanisms of drug resistance, including efflux pump-mediated resistance.
In conclusion, the present study demonstrated that FL118 has additional mechanistic features in terms of its superior anticancer efficacy, which further distinguish it from irinotecan, SN-38 and topotecan. These findings suggest that FL118 is a poor substrate for the drug efflux pump ABCG2, and thus FL118 is able to overcome ABCG2-mediated resistance to SN-38, irinotecan and topotecan in vitro and in vivo. Additionally, this study also indicated that polar chemical groups on the B ring of FL118 analogues can contribute to ABCG2-mediated resistance, which provides one principle for new FL118 analogue design. Together, the new features of FL118 revealed in this study plus the other FL118 unique features reported in our previous studies warrants FL118 further development toward clinical application.
Acknowledgements
This work was supported in part by a US Army DOD grant (PC110408) and NIH/NCI Grants (CA180764, CA176937) as well as by shared resources supported by a NCI Cancer Center Support Grant to Roswell Park Cancer Institute (CA016056).
We wish to express gratitude to Drs. Xiaojun Liu, Lili Tian, Pam Hershberger, Kevin Eng, and Ms. Wendy Swetzig for offering constructive suggestions, troubleshooting, discussion, and training relevant to this study. We also wish to thank Dr. Wendy Huss for providing the ABCG2 overexpressed HEK293 cells, which were originally a gift from Dr. Susan Hates at the National Cancer Institute (Rockville, MD). Finally we want to acknowledge the RPCI Department of Pharmacology and Therapeutics faculty writing group for their critical assessment of this manuscript.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
FL118 will be further developed in Canget BioTekpharma LLC, a Roswell Park Cancer Institute spinoff company. XL, MW and FL are initial investors in Canget BioTekpharma.
Authors’ contributions
DW participated in the conception and design of this study, performed and analyzed cell viability assays, lentiviral shRNA knockdown assays, and in vivo experiments, produced electrostatic potential maps, and participated in the preparation of this manuscript; XL conducted initial experiments which served as the basis for this study and provided training for in vivo experiments, and performed and analyzed cell viability assays; HL performed and analyzed cell viability and immunoblot assays; JW performed and analyzed cell viability and immunoblot assays and assisted in production of electrostatic potential maps; CJ provided experimental compounds, technical expertise, and critically evaluated early drafts of the manuscript; CG and MDR provided proprietary cell lines and offered technical advice on their use in this work; MW was involved in initially synthesizing most of the compounds used in this study, and provided overall technical expertise related to medicinal chemistry; FL participated in the conception and design of the study and preparation of the manuscript. All authors read and approved the final manuscript.
Hong Lam and Jacob Welch were interns in the Roswell Park Cancer Institute Summer Research Program.
Dr. Wani is an emeritus member of RTI International.