Design and synthesis of new 7-(N-substituted-methyl)-camptothecin derivatives as potent cytotoxic agents

Abstract

A series of novel 7-(N-substituted-methyl)-camptothecin derivatives was designed, synthesized, and evaluated for in vitro cytotoxicity against four human tumor cell lines, A-549, MDA-MB-231, KB, and KBvin. All of the derivatives showed promising in vitro cytotoxic activity against the tested tumor cell lines, with IC₅₀ values ranging from 0.0023 to 1.11 μM, and were as or more potent than topotecan. Compounds 9d, 9e, and 9r exhibited the highest antiproliferative activity among all prepared derivatives. Furthermore, all of the compounds were more potent than paclitaxel against the multidrug-resistant (MDR) KBvin subline. With a concise efficient synthesis and potent cytotoxic profiles, especially significant activity towards KBvin, compounds 9d, 9e, and 9r merit further development as a new generation of camptothecin-derived anticancer clinical trial candidates.

Camptothecin (CPT, 1, Figure 1), a natural quinoline alkaloid isolated by Wall and Wani from the Chinese tree Camptotheca acuminata, showed potent antiproliferative activity against a broad spectrum of tumors.^1-3 Its antitumor activity is induced by directly binding to topoisomerase I (Topo I), which results in interference with the catalytic cycle of DNA-Topo I and stabilization of the DNA-Topo I binary complex.^4-6 Based on CPT’s remarkable anticancer activity and unique cytotoxic mechanism, numerous potent CPT analogs have been developed. Among them, two Topo I inhibitors, topotecan (2) and irinotecan (3), have been used successfully in the clinic as anticancer drugs, while several other analogs are the subjects of ongoing preclinical or clinical evaluation.^7-10

Figure 1.

Structures of camptothecin (1), topotecan (2), irinotecan (3), gimatecan (4), CKD-602 (5), and BNP-1350 (6).

Although CPT derivatives remain a promising class of antitumor agents, the highly electrophilic α-hydroxylactone of the E ring is intrinsically unstable and undergoes rapid hydrolysis to the biologically inactive carboxylate form under physiological conditions.^11,12 This chemical feature diminishes the efficacy of various CPT derivatives in vivo compared to the impressive results often obtained from in vitro studies. Thus, several synthetic strategies to overcome this challenge have been developed, resulting in a logical mapping of the structure–activity relationship (SAR) of CPT derivatives.^13,14 Substitutions at the 11- or 5-position are not well tolerated, whereas various substituents at the 7-, 9-, or 10-position can improve the antitumor activity, as well as increase E-ring stability.^11,15,16 In particular, previous studies documented that the introduction of lipophilic substituents at the 7-position provides favorable molecular interactions and improved pharmacological features that could have potential therapeutic advantages.¹⁷ A binding model of CPT with biological macromolecules also indicated that the C-7 molecular area could accommodate considerable structural diversity.^18-20 On the basis of these critical clues, various substitutions, such as ethyl, alkylsilyl, oxyiminoalkyl, and alkylsilylalkyl, were introduced at the 7-position of CPT to produce potent antitumor agents. To date, 7-substituted compounds constitute most of the second-generation CPT analogs that have reached preclinical or clinical development studies. Examples include gimatecan (4),²¹ CKD-602 (5),²² and BNP-1350 (6),²³ which contain highly lipophilic substituents intended to increase antitumor activity. These successful examples indicate the important role played by various C-7 substitutions in the activity profiles of CPT analogs and the feasibility of optimizing this compound class through rational C-7 modification.

In our continuing studies on the chemistry of CPT,^24-30 we recently reported a series of 7-ketone camptothecin derivatives with potent antitumor activity and significantly different drug-resistance profiles from those of the parent compound.³⁰ Some of the new compounds exhibited activity comparable to that of marketed CPTs, such as 2 and 3. Notably, some compounds displayed promising cytotoxicity against KBvin cells, while 3 lost activity completely. The encouraging preliminary results have prompted us to extend our investigation by synthesizing a novel series of 7-(N-substituted-methyl)-camptothecin derivatives. Herein, we describe our introduction of different nitrogen substituted groups into the 7-position of CPT via a coupling reaction and cytotoxic activity studies on the resulting compounds.

The synthetic route to target CPT derivatives 9a-s is depicted in Scheme 1. Briefly, treatment of 1 with hydrogen peroxide and ferrous sulfate in an aqueous methanol-sulfuric acid solution furnished 7-hydroxymethylcamptothecin (7) in 80% yield.³¹ Precursor 7 was converted into the key intermediate 7-bromomethylcamptothecin (8) in 66% yield by heating in hydrobromic acid.³² Intermediate 8 was coupled with various substituted amines in dry DMF to afford the desired derivatives 9a-s in 21–46% yields.³³ All synthesized target compounds were purified by column chromatography, and their structures were characterized by ¹H-NMR, MS, and elemental analysis.

Scheme 1.

Synthesis of target compounds 9a-s.

Target compounds 9a-s were evaluated for in vitro cytotoxicity against a panel of human tumor cell lines, including A-549 (lung carcinoma), MDA-MB-231 (triple-negative breast cancer), KB (originally isolated from nasopharyngeal carcinoma), and KBvin (MDR KB subline), using a sulforhodamine B colorimetric (SRB) assay with triplicate experiments.^34,35 Paclitaxel and 2 were used as positive controls and the screening results are shown in Table 1.

Table 1.

Antiproliferative activity of 9a-s against four human tumor cell lines

Compd	IC50 (μM) with SD
	A-549	MDA-MB-231	KB	KBvin
9a	0.0532±0.0226	0.0781±0.0178	0.0783±0.0272	0.0479±0.0126
9b	0.0587±0.0262	0.0927±0.0187	0.0817±0.0201	0.0795±0.0411
9c	0.0470±0.0191	0.0842±0.0062	0.0758±0.0107	0.0773±0.0255
9d	0.0023±0.0024	0.0264±0.0125	0.0046±0.0032	0.0149±0.0121
9e	0.0063±0.0027	0.0300±0.0163	0.0050±0.0013	0.0132±0.0055
9f	0.0391±0.0173	0.0644±0.0306	0.0397±0.0015	0.0129±0.0066
9g	0.0592±0.0252	0.114±0.0464	0.0821±0.0093	0.0633±0.0177
9h	0.0362±0.0211	0.0640±0.0025	0.0615±0.0119	0.0780±0.0853
9i	0.0504±0.0206	0.0961±0.0102	0.0713±0.0139	0.0688±0.0315
9j	0.0170±0.0115	0.0576±0.0514	0.0371±0.0031	0.0128±0.0081
9k	0.122±0.0406	0.337±0.0114	0.180±0.0662	0.117±0.0177
91	0.0514±0.0211	0.108±0.0197	0.0786±0.0058	0.0706±0.0217
9m	0.0188±0.0154	0.0532±0.0006	0.0362±0.0153	0.0165±0.0149
9n	0.0219±0.0107	0.0699±0.0043	0.0499±0.0074	0.0205±0.0141
9o	0.0155±0.0059	0.0609±0.0096	0.0371±0.0018	0.0208±0.0199
9p	0.0622±0.0168	0.110±0.0177	0.0895±0.0231	0.274±0.1592
9q	0.534±0.0701	1.11±0.0059	0.813±0.0556	1.06±0.0115
9r	0.0048±0.0042	0.0316±0.0026	0.0150±0.0102	0.0217±0.0157
9s	0.150±0.0275	0.378±0.0082	0.208±0.0209	0.183±0.0219
2	0.0452±0.0004	0.102±0.0055	0.0625±0.0042	0.396±0.0207
Paclitaxel	0.0057±0.0016	0.0066±0.0018	0.0039±0.0010	1.44±0.137

As illustrated in Table 1, all new compounds exhibited significant in vitro cytotoxic activity against the four tested tumor cell lines, with IC₅₀ values ranging from 0.0023 to 1.11 μM, and except for 9k, 9q, and 9s, were mostly more active than 2, a clinically used CPT-derived chemotherapeutic drug. Compounds 9d, 9e, and 9r were the most potent compounds in the series and were also superior to paclitaxel against the A-549 cell line, which was generally most sensitive to these CPT derivatives. Against the same cell line, many compounds, including 9d–9f, 9h, 9j, 9m–9o, and 9r, also showed better antiproliferative activity than 2. With few exceptions (i.e., 9k, 9q, and 9s), the tested compounds showed increased cytotoxic potency against the triple-negative breast cancer (MDA-MB-231) cell line compared with 2. With regard to the KB cell line, seven compounds exhibited significant cytotoxic activity comparable to that of 2. Remarkably, except for 9q, all of the compounds (IC₅₀ 0.0128–0.274 μM) were more potent than paclitaxel (IC_50: 1.44 μM) and 2 (IC₅₀ 0.396 μM) against the KBvin subline. These encouraging results suggested that these new derivatives could overcome the MDR phenotype overexpressing P-glycoprotein. Notably, the three most promising compounds 9d, 9e, and 9r showed broad in vitro antitumor spectra and were about 7- to 20-fold more potent than 2. Further pharmacological, toxicological evaluations and binding affinity analysis with DNA-Topo I target in the modeling of these promising compounds are in progress.

SAR analysis of the results from the synthesized compounds revealed several structural properties that could influence the in vitro cytotoxicity of the new CPT derivatives. Regarding compounds 9a to 9j with substituted (phenylamino)methyl groups at C-7, derivatives with m-/p-chlorophenyl rings (9f and 9i) exhibited lower activity than their o-substituted isomer 9j. Similarly, the relocation of an electron-donating methoxy group from the o-position (9d, IC₅₀ 0.0023μM) to the p-position (9c, IC₅₀ 0.0470 μM) also decreased the cytotoxicity remarkably. These results suggested that an o-substituted phenyl group is more favorable than m-/p-substitued rings for better activity. Moreover 9e and 9f with p-fluoro and p-chloro substitution, respectively, on the phenyl ring were more potent against A-549 tumor cells than 9g with p-bromo substitution, indicating that a substituent’s size, in addition to position, could affect the activity. However, the presence of a pyridine-2-yl rather than phenyl group on the amine did not substantially change the cytotoxic activity (9l vs 9a). Among compounds 9m–9s with non-aromatic amino substituents, the SAR analysis showed that the alkyl variant had an important effect on the in vitro cytotoxic activity. For example, the cytotoxic activity of 9r with a (propylamino)methyl group at C-7 was significantly greater than that of compounds with a (cyclohexylamino)methyl (9m) or (diisopropylamino)methyl (9q) moiety. These data suggested that the bulkier substituents in the two latter compounds might produce steric hindrance at the target level and thus lower the cytotoxic activity. The polarity and electron density in the C-7 side chain were also important, as changing the propylamino group in 9r to a 2-hydroxyethylamino group in 9s greatly decreased the cytotoxic activity (e.g., IC₅₀ 0.0048 μM for 9r vs 0.15 μM for 9s against A-549). Taken together, these results showed that both the identity and substitution pattern in the R group at C-7 could greatly influence the cytotoxicity of the new CPT derivatives

In summary, 19 novel 7-(N-substituted)-methyl-camptothecin derivatives were designed, synthesized, and evaluated for antiproliferative activity against four human tumor cell lines (A-549, MDA-MB-231, KB and KBvin) by using a sulforhodamine B colorimetric assay. Most of the new derivatives showed comparable or superior antiproliferative activity compared with 2. In particular, compounds 9d, 9e, and 9r were the most promising derivatives with 7- to 20-fold greater potency than 2 against the A-549 cell line and were selected as lead molecules for further development. Notably, with IC₅₀ values ranging from 0.0128 to 1.06 μM, all of the compounds also were more potent than paclitaxel (IC₅₀ 1.44 μM) against KBvin cells. Furthermore, SAR study indicated that both N-aromatic and N-aliphatic substituents at C-7 can produce potent activity, while selected variation of these substituents can greatly affect the activity. These findings support our further optimization of CPT to develop potential anticancer drug candidates. Continuing studies to substantiate and improve activity profiles are underway in our laboratories and will be reported in due course.