Design, Synthesis, and Biological Evaluation of a Conjugate of 5-Fluorouracil and an LSD1 Inhibitor

Abstract

Prodrug approaches are useful for enhancing the efficacies and reducing the side effects of anticancer drugs. Previously, we proposed a prodrug strategy for targeting cancers overexpressing lysine-specific demethylase 1 (LSD1), namely, conjugates of trans-2-phenylcyclopropylamine (PCPA, an LSD1 inhibitor) and anticancer drugs. In this study, we applied this prodrug strategy to the anticancer agent 5-fluorouracil (5-FU). In vitro assays showed that the PCPA-5-FU conjugate (1) released 5-FU upon the inhibition of LSD1. Furthermore, the conjugate (1) exerted an antiproliferative effect on colon cancer HCT116 cells. Thus, the PCPA-5-FU conjugate (1) was able to function as a prodrug of 5-FU, activated by LSD1 inhibition, and provided a useful new lead structure for further development.

Introduction

The controlled release of an anticancer drug is a key issue of targeted therapy. Various types of drug delivery systems for cancer targeting have been reported; most contain macromolecular drug carriers, such as nanoparticles and antibodies.^1,2) However, they are not without problems, including high production costs and limited administration routes. Small molecule-based prodrugs have also proven useful for achieving enhanced efficacies and reduced side effects of anticancer drugs and represent an effective method to avoid the aforementioned problems.³⁾ Thus, the development of a novel prodrug strategy using small molecules may yield improvements in targeted cancer therapies.

We recently reported a prodrug strategy using trans-2-phenylcyclopropylamine (PCPA), which is a representative lysine-specific demethylase 1 (LSD1) inhibitor⁴⁾ (Fig. 1). LSD1 is a FAD-dependent histone demethylase⁵⁾ known to be overexpressed in various cancer cells, such as breast cancer,⁶⁾ neuroblastoma,⁷⁾ and colon cancer.⁸⁾ Thus, LSD1 is considered to be an important molecular target for cancer therapy, with LSD1 inhibitors expected to be anticancer drugs.^9–11) To develop a novel prodrug strategy, we focused on the proposed mechanism of LSD1 inhibition by PCPA^12,13) (Fig. 1a). PCPA irreversibly inhibits LSD1 enzymatic activity through the formation of PCPA-FAD adducts (Fig. 1a). When LSD1 is inhibited by PCPA, the nitrogen atom of PCPA is released in an ammonia molecule through the hydrolysis of an imine intermediate (Fig. 1a). Based on this mechanism, we proposed that a PCPA-drug conjugate (PDC) would be a suitable prodrug for LSD1-overexpressing cancers⁴⁾ (Fig. 1b). The PDC would act as an inhibitor of LSD1 and lead to the release of the drug in the cancer cells in which LSD1 was overexpressed (Fig. 1b). We previously performed a proof-of-concept study using PCPA-tamoxifen conjugates as a prototype PDC⁴⁾ and believe this strategy has broad applicability. In this study, we applied this PDC strategy to 5-fluorouracil (5-FU).

Fig. 1. (a) The Proposed Mechanism of LSD1 Inhibition by PCPA; (b) Conceptual Method of Action of PCPA-Drug Conjugate (PDC)

(Color figure can be accessed in the online version.)

5-FU is an antimetabolite, and 5-FU and its prodrugs are used widely for chemotherapy.¹⁴⁾ It is reported that in combination with LSD1 inhibitors, such as PCPA or pargyline, 5-FU exerts a potent antiproliferative effect on oral cancer cells.¹⁵⁾ Therefore, we hypothesized that the PCPA-5-FU conjugate may be an interesting prodrug as well as a bifunctional molecule for the therapy of LSD1-overexpressing cancers.

Here, we have reported the design, synthesis, and biological evaluation of a conjugate of 5-FU and PCPA. Given the stability and reactivity in water, we designed the PCPA-5-FU conjugate 1, which was composed of PCPA, a linker with an N-hydroxymethyl group, and 5-FU (Fig. 2a). The expected mechanism of 5-FU release triggered by LSD1 inhibition by 1 is shown in Fig. 2b. Conjugate 1 was predicted to inhibit LSD1 activity through the formation of a PCPA-FAD adduct, thereby releasing an amine compound. The amine would then lead to intramolecular cyclization and the release of an N-hydroxymethylated 5-FU, generating 5-FU through deformylation. Thus, conjugate 1 was expected to function as a prodrug targeting LSD1-overexpression cancers. As a reference compound, we also designed tert-butoxycarbonyl (Boc)-5-FU conjugate 2, in which the PCPA moiety of 1 was replaced with a Boc group (Fig. 2a). Conjugate 2 was not expected to release 5-FU, even in cancer cells overexpressing LSD1, owing to the absence of the PCPA moiety, which is the pharmacophore for LSD1 inhibition (Fig. 1a).

Fig. 2. (a) Design of PCPA-5-FU Conjugate (1) and Boc-5-FU Conjugate (2); (b) The Expected Mechanism of 5-FU Release from 1 Triggered by LSD1 Inhibition

The synthetic routes of 1 and 2 are shown in Chart 1. Amines 3 were prepared by using reported methods.^4,16) Amines 3 were reacted with chloromethyl chloroformate 4 to afford carbamates 5. Nucleophilic substitution reactions between carbamates 5 and 5-FU in the presence of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) afforded the desired conjugates 1 and 2.

Chart 1. Synthesis of PCPA-5-FU Conjugate (1) and Boc-5-FU Conjugate (2)^a)

a) Reagents and conditions: (a) N,N-diisopropylethylamine, CH₂Cl₂; (b) DBU, N,N-dimethylformamide, 25% for 1 and 17% for 2 (two steps from 3).

First, we performed an in vitro study to examine whether conjugates 1 and 2 were able to inhibit the LSD1 enzymatic activity. To evaluate the LSD1 inhibitory activities of the conjugates, we used a LSD1 fluorometric drug discovery kit (Enzo, BML-AK544-0001), and found that the LSD1 inhibitory activity of 1 was five times greater than that of PCPA, a positive control (1, IC₅₀ = 5.06 µM; PCPA, IC₅₀ = 24.2 µM) (Table 1). In contrast, 100 µM compound 2, a negative control compound without a PCPA moiety, did not inhibit LSD1. These results showed that conjugate 1 was a potent LSD1 inhibitor, but 2 was not.

Table 1. LSD1 Inhibitory Activities of PCPA and Conjugates 1 and 2

Compound	IC50 (µM)a)
PCPA	24.2 ± 2.8
1	5.06 ± 0.25
2	>100

a) Values are the mean ± standard deviation (S.D.) of at least four experiments.

Next, we used HPLC analysis to investigate whether conjugate 1 released 5-FU dependently on LSD1. First, we analyzed a mixture of authentic 5-FU samples and 1, and confirmed that the peaks at 2.6 min and 13.3 min corresponded to 5-FU and 1, respectively (Fig. 3a). In addition, we confirmed that conjugate 1 was stable in the assay buffer (Figure S1a). We also confirmed that the retention time of FAD was 11.5 min (Figure S1b). We then analyzed the reaction mixtures after a 24 h incubation of conjugate 1 in the presence of LSD1 or FAD. In the mixture of 1 and LSD1, the peak at 2.6 min was observed, but the peak intensity at 13.3 min was relatively low (Fig. 3b). In contrast, in the mixture of 1 and FAD, the peak at 2.6 min was hardly visible and the peak intensity at 13.3 min was strong (Fig. 3c). These results suggested that conjugate 1 was degraded by LSD1, which was followed by the release of 5-FU, and that conjugate 1 did not release 5-FU when LSD1 was replaced by FAD.

Fig. 3. Identification of 5-FU Generated from PCPA-5-FU Conjugate (1) in the Presence of LSD1

(a) Mixtures of authentic samples: 5-FU (2.6 min) and 1 (13.3 min); (b) mixture of 1 and LSD1; (c) mixture of 1 and FAD.

Finally, we evaluated the antiproliferative effect of 1 on colon cancer HCT116 cells overexpressing LSD1⁸⁾ by using alamar Blue reagent. Conjugate 1 exerted an antiproliferative effect on HCT116 cells, with an IC₅₀ of 61.2 µM; the activity of 1 was twelve times stronger than that of PCPA (Table 2). In contrast, 100 µM reference compound 2 did not inhibit the proliferation of HCT116 cells (Table 2). These results suggested that conjugate 1 was a prodrug of 5-FU activated by LSD1 inhibition. However, the antiproliferative effect of 1 on HCT116 cells was 2.8-time lower than that of its parent compound 5-FU (Figure S2). This result indicates that 1 could not completely release 5-FU in the cells. This might be because 1 still has some problems to be improved such as insufficient LSD1-inhibitory activity, low membrane permeability and so on.

Table 2. Antiproliferative Effect of PCPA and Conjugates 1 and 2 in HCT116 Cells

Compound	Viability IC50 (µM)a)
PCPA	755 ± 96
1	61.2 ± 2.8
2	>100

a) Values are the mean ± S.D. of at least four experiments.

In summary, to develop PDC-based prodrug strategies, we have studied a PCPA-5-FU conjugate. In in vitro studies, conjugate 1 showed potent LSD1 activity and released 5-FU in the presence of LSD1. Furthermore, conjugate 1 exerted an antiproliferative effect on HCT116 cells, whereas the reference compound 2 did not. Thus, the novel PDC 1 was an effective prodrug of 5-FU. Although the biological activities of 1 were moderate in the in vitro and cellular assays, there is potential for their improvement through future optimization studies of the PCPA moiety and/or the linker of 1.^17,18) Further studies on PDCs are currently in progress in our laboratory.

Acknowledgments

The authors would like to thank Dr. Miki Suzuki and Dr. Yasunao Hattori for their technical support. This work was supported in part by a Grant-in-Aid from the Japan Society for the Promotion of Science (Y.O., T.S.) (JP14J10025, JP15K14982), the JST CREST program (T.S.) (JPMJCR14L2), the Takeda Science Foundation, and the Research Foundation for Pharmaceutical Sciences (T.S.).

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

The online version of this article contains supplementary materials.

References

• 1) Petros R. A., DeSimone J. M., Nat. Rev. Drug Discov., 9, 615–627 (2010).
• 2) Chari R. V. J., Miller M. L., Widdison W. C., Angew. Chem. Int. Ed., 53, 3796–3827 (2014).
• 3) Rautio J., Kumpulainen H., Heimbach T., Oliyai R., Oh D., Järvinen T., Savolainen J., Nat. Rev. Drug Discov., 7, 255–270 (2008).
• 4) Ota Y., Itoh Y., Kaise A., Ohta K., Endo Y., Masuda M., Sowa Y., Sakai T., Suzuki T., Angew. Chem. Int. Ed., 55, 16115–16118 (2016).
• 5) Shi Y., Lan F., Matson C., Mulligan P., Whetstine J. R., Cole P. A., Casero R. A., Shi Y., Cell, 119, 941–953 (2004).
• 6) Lim S., Janzer A., Becker A., Zimmer A., Schüle R., Buettner R., Kirfel J., Carcinogenesis, 31, 512–520 (2010).
• 7) Schulte J. H., Lim S., Schramm A., Friedrichs N., Koster J., Versteeg R., Ora I., Pajtler K., Klein-Hitpass L., Kuhfittig-Kulle S., Metzger E., Schüle R., Eggert A., Buettner R., Kirfel J., Cancer Res., 69, 2065–2071 (2009).
• 8) Ding J., Zhang Z.-M., Xia Y., Liao G.-Q., Pan Y., Liu S., Zhang Y., Yan Z.-S., Br. J. Cancer, 109, 994–1003 (2013).
• 9) Suzuki T., Miyata N., J. Med. Chem., 54, 8236–8250 (2011).
• 10) Wang X., Huang B., Suzuki T., Liu X., Zhan P., Epigenomics, 7, 1379–1396 (2015).
• 11) Fu X., Zhang P., Yu B., Future Med. Chem., 9, 1227–1242 (2017).
• 12) Schmidt D. M. Z., McCafferty D. G., Biochemistry, 46, 4408–4416 (2007).
• 13) Szewczuk L. M., Culhane J. C., Yang M., Majumdar A., Yu H., Cole P. A., Biochemistry, 46, 6892–6902 (2007).
• 14) Longley D. B., Harkin D. P., Johnston P. G., Nat. Rev. Cancer, 3, 330–338 (2003).
• 15) Wang Y., Zhu Y., Wang Q., Hu H., Li Z., Wang D., Zhang W., Qi B., Ye J., Wu H., Jiang H., Liu L., Yang J., Cheng J., Cancer Lett., 374, 12–21 (2016).
• 16) Zheng J.-S., Yu M., Qi Y.-K., Tang S., Shen F., Wang Z.-P., Xiao L., Zhang L., Tian C.-L., Liu L., J. Am. Chem. Soc., 136, 3695–3704 (2014).
• 17) Miyamura S., Araki M., Ota Y., Itoh Y., Yasuda S., Masuda M., Taniguchi T., Sowa Y., Sakai T., Suzuki T., Itami K., Yamaguchi J., Org. Biomol. Chem., 14, 8576–8585 (2016).
• 18) Ota Y., Miyamura S., Araki M., Itoh Y., Yasuda S., Masuda M., Taniguchi T., Sowa Y., Sakai T., Itami K., Yamaguchi J., Suzuki T., Bioorg. Med. Chem., 26, 775–785 (2018).