Novel Cyclopropylindole Conjugates and Dimers: Synthesis and Anti-Cancer Evaluation

Sanjay K. Sharma, Guofeng Jia and J. William Lown*

Department of Chemistry, University of Alberta, Edmonton, Alberta, Canada T6G 2G2

*Address  correspondence to this author at the Department of Chemistry, University of Alberta, Edmonton, Alberta, AB, Canada T6G 2G2. Tel. 780-492-3646; Fax: 780-492-8231. Email. annabelle.wiscman@ualberta.ca

Abstract: There is a considerable interest currently in the development of DNA sequence specific or selective agents for genetic targeting for the control of gene expression, for application in diagnosis or ultimately in therapy. In this context CC-1065 is one of the most impressive lead compounds isolated in trace quantities from the culture of Streptomyces zelensis at Upjohn in 1978. The unique structure was confirmed by single X-ray in 1981. However CC-1065 cannot be used in humans because it was found that it caused delayed deaths in experimental animals. In the search for compounds with better antitumor selectivity and DNA sequence specificity many CC-1065 analogs have been synthesized in an attempt to avoid the undesired side effects while retaining its potency against tumor cells. Two successful attempts in the modification in the active moiety of the parent natural product 1,2,8,8a-tetrahydro-7-methylcyclopropa[3,2-e]indole-4-one (CPI) and 1,2,9,9a-tetrahydrocyclopropa[c]benz[e]indole-4-one (CBI) have been made. We review here recent progress with the analogs of CPI and CBI and their conjugates both by solution and solid phase, also the progress and development of CPI and CBI conjugates with polyamides (information reading molecules in the minor groove of DNA). Since CPI-CPI dimers are significantly more potent than CC-1065 in vitro and in vivo, a large number of CBI-CBI dimers with varying linkers lengths and positions synthesized in our group and their pharmacological properties have been reviewed.

Introduction

     Extensive research has been reported on CC-1065 (1) an antitumor antibiotic isolated from the culture of Streptomyces zelensis [1]. It is one of the most potent anticancer compounds having a wide spectrum of activities against tumor cells in vitro and in vivo as well as against microbial organisims [2,3]. Cellular investigations have shown that CC-1065 alkylates B-DNA reversibly with a high sequence selectivity at AT regions of the minor groove sites of 5'-d(A/GNTTA)-3' and 5'-d(AAAAA)-3', by the N-3 position of the 3'-adenine by the cyclopropylindole unit present in the molecule [4,5]. Despite its high potency CC-1065 cannot be used in humans because it was found that it caused delayed death in experimental animals. In the search for compounds with better antitumor selectivity and DNA sequence specific binding property, many CC-1065 analogs have been synthesized in an attempt to avoid its unwanted side effects but to retain its potency against tumor cells [6,7]. Besides CC -1065, duocarmycin 2 and 3 are also potent examples of this family exceptionally potent antitumor antibiotics. Certain synthetic analogs of CC-1065 and duocarmycin, including dimeric structures, show promise and are currently undergoing clinical trials [8] indicating that synthetic efforts in this area can be useful.The structure of CC-1065 consists of three pyrrolo[3,2-e]dihydroindole moities of which the A unit ( refered to as CPI) is the pharmacophore, the B and C units are identical and have a strong influence on the binding specification and the biological potency [9]. Because it is such a highly potent pharmacophore there is great interest in analogs of the CPI subunit of CC-1065. A new class of 1,2,9,9a-tetrahydrocyclopropa[c]benz[e]indole-4-one (CBI) pharmacophore has been synthesized. In this review we examine the most recent analogs of CPI and CBI and also their in vitro cytotoxicities. CPI-polyamide conjugates have shown good cytotoxic potency [10,11] therefore we also review various new CPI and CBI polyamide conjugates .Since CPI-CPI dimers were very active against tumor cells a large number of CBI-CBI dimers with varying lengths of the linkers and positions of attachment were synthesized in our group their cytotoxic properties against tumor cells has been reviewed.

Modifications in the pharmaco-phore unit

     An isomeric structural modification in the CC-1065 and duocarmycins alkylation subunits was incorporated by synthesizing 2- (tert-butyloxy-carbonyl)-1,2,9,9a-tetrahydrocyclopropa [c]benzo [f]indol-8-one (4 (N-Boc-iso-CBI) and 1-(tert-butyloxycarbonyl)-4-hydroxy-3[[(methanesulfo-nyl)oxy]methyl]-2,3-dihydroindole (5 seco-N-Boc-iso CI) [12]. The evaluation of the iso-CBI based agents revealed a significant increase in stability comparable to that of CC-1065 and duocarmycin A but that it is more reactive than duocarmycin SA (6-7x) or by direct comparision CBI based agents (5x). Synthesis of a full set of the natural product analogs was reported (4-13). From the structural studies of DNA-agent adducts [13,14] it was evident that the C-4 carbonyl of the natural product projects out of the minor groove lying on the outer face of the complexes potentially accessible to the phosphate backbone. In contrast the relocated carbonyls of the iso CI and iso CBI would be required to project into the minor groove inaccessible to the phosphate backbone if participating in an analoguous adenine N3 alkylation reaction (Fig. 1). The relocation of the C-4 carbonyl (the most substantial structural modification to the alkylation subunit) permitted reaction at comparable rates and retained identical and characteristic sequence selectivity of CC-1065 and duocarmycins. This observation is inconsistent with the proposal that a sequence dependent C-4 carbonyl protonation by strategically located DNA backbone phosphates controls the DNA alkylation selectivity but is consistent with the proposal that it is determined by the AT rich noncovalent binding selectivity of the agents and the steric accessibility of the N-3 alkylation sites.

 

FIG.1

    

 In vitro cytotoxic activity of these structurally modified analogs of CC-1065 demonstrates that the (-) enantiomer of the analogues possessing the (S) configuration analogous to the natural product is more potent than the enantiomer by 10-50x. The exception to this generalization is N-Boc-iso-CBI where the two enantiomers were not readily distinguishable and the unnatural enantiomer was consistently slightly more potent (1-2x). The seco precursors, which lack the preformed cyclo-propane but possess the capability of ring closure, were found to possess cytotoxic activity that was indistinguishable from the final ring closed agent. Consistent with the unique importance of the C-5 methoxy group of the duocarmycins iso-CBI-TMI 8 and 9 were found to be equipotent, confirming that the C6 and C7 methoxy groups of 8 are not contributing to its cytotoxic potency. The cinnamate derivative 10 was found to be substantially less potent (40-50x) suggestive of the requirement for a rigid N2DNA binding subunit [15]. The agent exhibited a smooth trend of increasing cytotoxic potency as the size and length of the DNA binding subunits increases with iso-CBI-CDPI1(12) and iso-CBI-CDPI2(13) dis-playing the most potent cytotoxic activity in the series exhibiting IC50 values of 200 and 50 pM respectively.

     A carbocyclic C-ring analogue 1,2,9,9a-tetrahydro-1H-cyclopropa[c]benz[e]inden-4-one (CBIn 15) of the alkylation subunit of CC-1065 and duocarmycin was synthesized [16]. A study of the CBIn solvolysis reactivity, regioselectivity and mechanism of action revealed that removal of the nitrogen and resulting vinylogous amide stabilization increased the reactivity 3200x (pH 3) and reversed the inherent regioselectivity but did not alter the Sn2 reaction mechanism. The vinylogous amide found in the naturally occurring alkylation subunits is responsible for their unusual stability and, significantly, imparts the regio-selectivity. Solvolysis reactivity proved independent of pH throughout the range of 4-12 including the physiologically relevent range of 5-8, where CBI is completely stable.These obser-vations have important implications on the source of catalysis for the CC-1065/ duocarmycin DNA alkylation reaction indicating that it is not derived from acid catalysis and C-4 carbonyl protonation but rather a DNA binding induced conformational change that disrupts the cross conjugated vinylogous amide stabilization.

     Boger et al. [17] synthesized a new class of DNA alkylating agent that incorporates the quinone of the mitomycin (16 and 17). One compelling feature of mitomycin C is its effective use in the clinic, either as a single agent or in combination with other agents [18] and it is especially useful in the treatment of solid tumors. Following reductive activation, mitomycin C is known to cross link DNA through bifunctional alkylation [19] and it is believed that this interstrand mitomycin/DNA adduct inhibits DNA replication, resulting in inhibition of cell division [20]. The extended hybrid analogues 18-20 and the Boc-substituted agent 21-24 were synthesized along with the naphthoquinone analogues 25-27.

     In vitro cytotoxic activity was studied on three different cell lines L1210, H460 and H596. These studies have shown that the enantiomer with the natural (3S) configuration corresponding to that of the natural product is the more potent enantiomer by about 1-10x for the extended agents 18-20 and little to no difference in the enantiomers was obtained for the simple Boc-substituted agents 21-24. The p-quinones bearing the mitomycin A substitution (18 and 22) exhibited cytotoxic potencies comparable with the corresponding air stable hydroquinones (23 and 24). Consistent with this interpretation the intrinsically more stable naphthoquinone analogues (25-27) were found to be more potent. It has been reported that a variety of tumor types have elevated DT-diaphorase (NQO1) activities relative to normal tissues [21]. DT-Diaphorase is a two electron reductase that uses either NADH or NADPH as a cofactor [22] to catalyze the two electron reduction of quinones and can protect cells against their toxic effects. DT-Diaphorase is also involved in the reductive activation of mitomycin C(4) and related analogues [23]. Increased potency in the H460 cell lines verses H596 cell lines would be consistent with their preferential activation by DT-Diaphorase. The result indicates a modest (3-6x) increase in cytotoxic potency for the quinone 18 and 22 against the H460 cell line which was high in DT-diaphorase activity compared to the H596 cell line which has no measurable DT-diaphorase activity. Naphthoquinone analogues 25-27 display even greater levels of selectivity with 27 exhibiting 24x increased potency against H460 consistent with reductive activation by DT-diaphorase. The results suggest that these analogues are good substrates for human recombinant DT-diaphorase, indicating the potential for these and related agents to be tumor selective DNA-alkylating agents subject to bioreductive alkylation.

     A small series of CBI-indole2 prodrug analogs of CC-1065 was synthesized by Boger et al. [24]. (+)-CBI-indole2 (28) a simplified agent not only exhibited cytotoxic potency comparable with that of (+) CC-1065 and greater (4x) than that of (+)-CPI-indole2 (U71,184) [25] but it also exhibited marked in vivo antitumor activity [26]. O-substituted derivatives of seco-CBI-indole2 (29) was synthesized by the treatment of 29 with 4-nitrophenyl chloroformate to afford the carbonate as intermediate, followed by the addition of dimethylamine (30) or 1-methylpiperazine (31) or phenyl isocyanate (32) or methyl isocyanate (33) or acetic anhydride (34) or by esterification with the corresponding acid to give 35. Each of these prodrugs exhibited an in vitro cytotoxic activity in L1210 assay that was essentially equivalent to that of 28 and 29. The cytotoxicities of the derivatives 30-35 indicate that each prodrug is effectively hydrolyzed to release 29 which serves as the precurssor to the active constituent 28. The agents were further examined with a human colon carcinoma cell line (HCT 116) and two resistent variants HCT116/VM46 which are resistent to topoisomerase II inhibitors. In these cell lines the carbamate prodrugs incorporating a tertiary amine (30 and 31) were inactive, presumably because of the lack of hydrolysis, but the remainder exhibited potent cytotoxic activity essentially equipotent or more potent against both the wild type and resistent type. These agents may prove especially effective against resistant variants of tumor cells lines and be particularly useful in combination therapy or in relapse chemotherapy.

     The synthesis of 1,2,8,8a-tetrahydro-cyclo-propa[c]pyrrolo[3,2-e]indol-4(5H)- one (CPI) 36, the parent alkylation subunit of both duocarmycin SA and CC-1065, which lacks the C6 methoxy carbonyl and C7 methyl group of the respective natural products, was achieved by Boger et al. [27].The evaluation of its properties revealed that the duocarmycin SA C6 methyl ester substantially increases stability (6x) while the CC-1065, C7 methyl group has no or only a relatively small effect on the alkylation subunit reactivity. Neither substituent has any impact on the intrinsic reaction regioselectivity of the activated cyclopropane. The duocarmycin SA C6 metho-xycarbonyl group substantially increases both the efficiency and rate of DNA alkylation, despite its intrinsic greater stability without altering the inherent sequence selectivity, while the CC-1065, C7 methyl group slows the rate and lowers the efficiency of DNA alkylation. N-Boc-CPI (37), N-Boc-C7-MeCPI (38) and N-Boc-DSA(39) were also synthesized [28,29] for comparision studies. The seco-CPI conjugates 40-44 were synthesized which on debenzoylation, followed by the base catalyzed ring closure, gave the desired products 45-49. These compounds were evaluated for their cytotoxic activity. The results are summarized in (Table 1) indicating that the natural enantiomers exhibiting IC50 values in the range of 0.1-1 nM and that the unnatural enantiomers were again less potent (20-100 nM).

     The substitution of the fused pyrrole C ring of CPI, the alkylating subunit of CC-1065 with the six membered benzene ring in CBI has been to increase relative stability (4x) and biological potency (4x) without affecting DNA alkylation selectivity [30]. Synthesis of methyl-1,2,9,9a-tetrahydrocyclopropa[c]pyrido[3,2-e]indol-4-one-7-carboxylate (CPyI) 51, which contains a similar one carbon expansion of the C ring pyrrole found in the DSA alkylation subunit with the incorporation of pyridine ring, was achieved by Boger et al. [31] and subsequently the conjugates 52-56 were synthesized for the biological studies. The in vitro cytotoxic studies against L1210 cell line revealed that the (+) enantiomer of the analogues possessing the configuration of the natural products is more potent than the enantiomer by 3-30x. The exceptions to this trend are the simple alkylation subunits 50 and 51. The seco precursors which lack the preformed cyclopropane but possess the capabilities of ring closure were found to possess cytotoxic activity that was indistinguishable from the final ring closed agents. Consistent with the unique importance of the C5 methoxy group in the binding subunit of the duocarmycin, CPyI-TMI (52) and 53 were found to be equipotent indicating that the C6 and C7 methoxy groups of 52 are not contributing to its cytotoxic potency. The indole derivative 54 is less potent (10x) than 52 and 53. The extended agents 55 and 56 displayed the most potent cytotoxic activities reflecting their longer length and greater adduct stability.

     A series of A ring pyrrole derivatives of duocarmycin bearing b-(5',6',7',- trimethoxy-2'-indolyl)acryloyl group 57-66 was synthesized and evaluated for in vitro anticellular activity against HeLa S3 cells and in vivo antitumor activity against murine sarcoma 180 in mice [32] the results are summarized in (Table 2).

     The A-ring pyrrole derivatives bearing the b-(5',6',7',- trimethoxy-2'-indolyl)acryloyl group and containing a double bond as spacer to Seg B of the natural type showed weaker peripheral blood toxicity than derivatives having the Seg B of the natural type. Moreover most of them exhibited potent antitumor activity against in vivo murine tumor models.

     CC-1065 analogues possessing a biologically active CBI functional group and amide substituted indole and benzofuran 67-72 were synthesized by Wang et al. [33]. Consistent with previous observation, the major factors determining the potency of CC-1065 class of compounds were determined to include hydrophobic interactions and van der Waals contacts between the drug and DNA. The compound 68, bearing two indole moieties, and 67, bearing only one indole unit have IC50 values 0.4 and 3nM respectively against U937 leukemia cells in vitro. The IC50 values of compound 70, bearing a butyramino group, and 69, bearing an acetamino group, are 0.008 and 0.4 nM respectively against U937 leukemia cells in vitro. Compound 71, bearing a double bond linker, is about 4 fold more potent than 67 bearing no double bond linker. Compound 68 is highly potent against all cell lines tested in the NCI in vitro screening with IC50 values 0.1-5nM range for most cell lines. Compounds 68 and 72 are highly active against L1210 leukemia in mice. Compound 68 is also active against B16BL6 melanoma in mice. Most importantly 68 and 72 are not myelosuppressive at therapeutically effective doses. The mechanism of tumor cell death is considered to be through induction of apoptosis and is accompanied by DNA fragmentation.

     The subunit linking amide of duocarmycin and CC-1065 has been replaced with an amidine and thioamide 75 and 76 [34]. The cytotoxic activities of the thioamide and amidine analogues and their seco derivatives were established and are summarized in (Table 3).

 

Table 1.

    

 The seco and ring closed agents possess identical activities consistent with the previous studies. The natural enantiomers exhibited more potent cytotoxic activities (100x

 

Table 2.

 

(Table 2). cont….

 

 

Table 3.

 unnatural counterparts. Consistent with their relative DNA alkylation effectiveness and relative stabilities, the modified analogues 75 and 76 were less potent than (+)-DSA (100x) and (+)-CC-1065 (50x) establishing the amide as the optimum linkage unit.

     The remarkable stability of 74 [35], its 104-105x reduction in cytotoxic potency and its inability to alkylate DNA even under extraordinary conditions (37 oC, 2 weeks, 10-1M versus 4 oC, 1-6h, 10-7M of 73) highlights the essential role of the linking N2 amide. Compound 74 was >106 less effective than 73. These observations have clear implications for the source of catalysis for the DNA alkylation reaction. They are inconsistent with exceptions of catalysis derived from C4 carbonyl protonation (acid catalysis) but are fully consistent with catalysis derived from DNA binding induced conformational changes that disrupt the cross conjugated vinylogous amide stabilization activating the agents for nucleophilic attack.

Polyamide Conjugates

     In a novel approach to explore the properties of cyclopropylindole (CPI) Lown et al.[10] envisaged certain CC-1065 analogues which retain the unique CPI unit but bearing N-methyl pyrrole amide groups, a DNA sequence recognizing moiety coupled to it. They postulated a hypothesis that the positively charged functionality of the protonated dimethylamino group in the oligopeptide might contribute to reduced cytotoxic potency by reducing intracellular accessibility and therefore should be replaced by an uncharged moiety. It was also considered likely that the linker between the CPI unit and the oligopeptide would be important to the binding potency of the conjugate molecule to DNA in order to maintain proper contact over the whole of the molecule [36].

     Sugiyama et al. [37] synthesized novel hybrid molecules 77 and 78 between segment A of duocarmycin and pyrrole/imidazole diamide and studied their DNA alkylation properties. These hybrid molecules primarily alkylate the 3' end of A in AT rich sequences as does the parent duocarmycin. More significantly these hybrids alkylate G residues of predetermined DNA sequences effectively and with high specificity by formation of a heterodimer with distamycin.

     In continuation of these studies , Sugiyama et al. [38] synthesized a novel conjugate between segment A of duocarmycin A and N-methyl imidazole (Im)-N-methyl pyrrole (Py) hairpin polyamide 79 and 80.

The conjugates PyPyPygImPyPyDu (79) and ImPyPygImPyPyDu (80) were designed to alkylate the target sequence (A/T)G(A/T)2N(A/G) and (A/T)G(A/T)CN(A/G) respectively according to Dervan's ring pairing rule. High resolution denaturing gel electrophoresis indicated that 79 exclusively alkylated the A of the 5'-TGTAAAA-3' within a  400bp DNA fragment. Similarly alkylation of 80 occurred exclusively at the G of the 5'-AGTCAGA-3' sequence with efficiency at nanomolar concentration (Fig. 2).

 

Fig. (2).

    

 Lown and coworkers [39,40] synthesized a bis functionalized precursor of CBI 81 and with this precursor synthesized bis-polyamide CBI conjugates of symmetrical (82) and unsymmetrical (83) types by incorporating pyrrole rings on both side for AT specificity and pyrrole on one side and imidazole on the other side of CBI for mixed sequence recognition.

     Lown et al. [41] further explored solid phase techniques for the synthesis of seco-CBI (85) and a conjugate which bears two polyamides moieties on either side of the pharmacophore 84 through an intermediate 86.

     A new type of diamide CPI conjugate possessing a vinyl linker 87 was synthesized by Sugiyama et al. [42]. Molecular modelling suggested that the insertion of a vinyl linker (L) between polyamide and CPI adjusts the location of the cyclopropane ring of the conjugate and in this way improves the alkylation efficacy of the conjugates. Lown et al. [10,11] have demonstrated that incorporation of a vinyl linker dramatically enhances the efficiency of DNA alkylation as well as cytotoxicity. Sequence selective alkylation of double stranded DNA by 87 was investigated by high resolution denaturing gel electrophoresis using 400 bp DNA fragments. Highly efficient alkylation predominantly occurs simultaneously at the purines of 5'-PyG(A/T)CPu-3' site on both strands at nanomolar concentration of 87. These results suggest that the homodimer of conjugate 87 dialkylates both strands according to Dervan's pairing rule [43] together with a new mode of recognition in which the Im-vinyl linker (L) pair target G/C base pairs. In addition to the major dialkylation sites a minor alkylation site was also observed at 5'-GT(A/T)GC-3'.This alkylation can be explained by an analogous slipped homodimer recognition mode in which the L-L pair recognize the A/T base pair. HPLC analysis revealed that the conjugate 87 simultaneously alkylates GN3/AN3 of the target sequences on both strands of ODNs.

     Eight ring hairpin polyamides conjugated at the hairpin trun to both enantiomers of seco-CBI 88 and 89 were synthesized by Dervan et al. [44]. Alkylation yields and specificity were determined on a restriction fragment containing six base pair match and mismatch series. Alkylation was observed at a single adenine flanking the polyamide binding site and strand selective cleavage could be achieved based on the enantiomer of seco-CBI chosen. At 1nM concentration of polyamide seco-CBI conjugate, near quantitative cleavage was observed after 12h.

Bisalkylators (Dimers)

     The CPI bisalkylators U-77,100% (bizelesin) 90 in which the two alkylating moieties are connected with a rigid linker 1,3-bis(2-carbonyl-1H-indol-5-yl)urea group showed promising antitumor activity [45]. This compound is under development and is in clinical trials. Recently Fukuda et al. [46] have reported the synthesis and antitumor activity of novel 3-methoxy carbonyl-2-trifluoromethyl CPI (MCTF CPI) 91. It is reported that for bisalkylators bearing flexible methylene linkers, that their cytotoxicity highly depends upon the length of the linker [45a]. Although the studies on the length and type of rigid linkers have been quite limited, it appeared that in vivo antitumor activity of 90 carrying a rigid linker is superior to that of bisalkylators bearing flexible methylene linkers. Following these observations Fukuda et al.[47] synthesized and evaluated novel bisalkylators of the MCTF CPI group by varying the length and type of rigid linkers (92 a-f). These 92 a-f bisalkylators were investigated for their cytotoxicity (in vitro) against HeLaS3 human uterine cervix carcinoma assay and antitumor activity (in vivo) against colon 26 murine adeno-carcinoma. The results are shown in (Table 4).

     The results show that 92a, having a urediyl group as linker similar to 90, exhibited cytotoxicity and antitumor activity comparable to 90, while the cytotoxicity and the antitumor activity of 92f were superior to those of 90. These results showed that the length of a rigid linker has a significant influence on the cytotoxicity and antitumor activity rather than the type of rigid linker.

 

Table 4.

   

  Further exploring the activity of bisalkylators bearing different rigid linkers, which were shorter than that of 92f, Fukuda et al.[48] synthesized 93 a-f bearing a 3,3'-(1,4-phenylene)diacryloyl group as linker with various CPI units. The cytotoxicity and antitumor activity of these compounds are shown in (Table 5).

     The results show that 93a (AT-760), having two MCTF CPI (91) rings, exhibited cytotoxicity and antitumor activity superior to 90. The therapeutic index (MTD/TGI50) of 93a was superior to those of 90 and 92f. It is apperent that 93a shows less toxicity than 90 and 92f respectively.

      The synthesis of a series of four dimers 94-97 derived from head to tail coupling of the two enantiomers of duocarmycin SA alkylation subunits was achieved by Boger et al.[48]. All agents proved to be potent in cytotoxicity assays against the L1210 cell lines with (+)(+)-DSA2 (94, 3.0pM) and (+)(-)-DSA2 (95, 5.0pM) and dis-played a 2-3 fold higher activity than duocarmycin SA (10 pM). The properties of the agents appear to follow well defined and, in retrorespect, predictable potency orders. Those possessing a natural enantiomer left hand alkylation subunit (i,e. 94 and 95) were found to 10x more potent than the corresponding unnatural enantiomer (i.e. 96 and 97). Thus the dimerization head to tail linkage of the duocarmycin SA alkylation subunit provides compounds which are 2-3x more potent than duocarmycin SA and approximately 2000x more potent than N-Boc-DSA.

      CPI dimers have shown good cytotoxic and antitumor activities, and now it is well documented that the activity of the dimeric drugs is strongly related to the length and the position of the linker. In order to investigate the structure activity relationship systematically Lown and coworkers [49,50] have designed three types of seco-CBI dimers viz. C7-C7 dimers 98a-98d, N3-N3 dimers 99a-99d and N3-C7 dimers 100a-100d which contain two racemic CBI moieties linked from the two position by a flexible methylene chain of variable length. All compounds 98-100 were active against almost all cell lines with MGM values from 41.6 mM (98c) to 0.0120 mM (99a)

 

Table 5.

.

In general the activity sequence is C7-C7 dimers < C7-N3 dimers < N3-N3 dimers.The cytotoxic sequence of these seco-CBI dimers implies that the linking N3 amide plays a critically important role in DNA alkylation. Among the C7-C7 dimers 98a is the most potent compound and the potency decreases with the increasing length of the linker (98c and 98d have almost the same activities). For C7-N3 dimers (100 a-c), compound 100d, which possesses the longest linker (n = 6) in the series, proved to be the most potent with potency decreasing in order of 100d (n = 6) > 100a (n = 3) > 100b (n = 4) > 100c (n =5). Interestingly, compound 100d (n = 6, GMG = 0.0891mM) was almost 30x as potent as compound 100c (n= 5, GMG = 2.63 mM) with the only structural difference of one carbon decrease in the linker. Among N3-N3 dimers, it was evident that this series of compounds is the most potent with MGM values ranging from 0.116 mM for 99b to 0.0120 mM for 99a. Compound 99a displayed striking potency in the leukemia, CNS cancer, melanoma, prostate and breast cancer cell panels with GI50 values lower than 0.01 mM in the cell lines. Compound 99c and 99d selectively inhibited leukemia, CNS cancer and melanoma with GI50 values lower than 0.01 mM in all the cell lines. Some of these seco-CBI dimers are the most promising candidates and have been selected for further in vivo testing by NCI.

Conclusions

     The present understanding of the origin of the properties of the natural products emerged principally through the design and evaluation of synthetic substrates and agents containing systematic structural modifications, in addition to through extensive studies conducted solely on the natural product themselves. As a consequence certain derivatives have proven much more efficacious and potent than the parent natural products. This illustrates that their useful properties may be enhanced by well founded and designed structural modifications. The active units when coupled together with certain appropriate linkers as dimers have proven to be even more potent and are in clinical trials. Conjugating the pharmacophore unit of CC-1065 with the DNA binding polyamides suggests a promising additional approach for developing a new type of tailor made sequence specific DNA alkylating agent. Such agents should prove valuable and find utility in further development of many diagnostic and therapeutic applications, especially ones involving the targeting of genomic DNAs.

Acknowledgments

     The authors wish to thank the Natural Sciences and Engineering Research Council of Canada and the Department of Chemistry, University of Alberta, for generous financial support.

Abbreviations

CBI           =     1,2,9,9a-Tetrahydrocyclopropa[c]-benz[e]indol-4-one

iso-CBI     =     1,1a,2,3-Tetrahydrocyclopropa[c]-benza[f]indol-9-one

CDPI        =     3-Carbamoyl-1,2-dihydro-3H- pyrrolo[3,2-e]indole-7-carboxylate

CI             =     1,2,7,7a-Tetrahydrocyclopropa[c]-indol-4-one

CPI           =     1,2,8,8a-Tetrahydro-7-methylcyclo-propa[c]pyrrolo[3,2-e]indol-4-one

CpyI         =     1,2,9,9a-Tetrahydrocyclopropa[c]-pyrido[3,2-e]indol-4-one

DNA         =     Deoxyribonucleic acid

DSA         =     Duocarmycin

IC50           =     The concentration of agent required to inhibit by 50% the growth of the indicated cell                                        line

MMI         =     5-Methoxyindole-2-carboxylate

TMI          =     5,6,7-Trimethoxyindole-2-carboxylate

References

[1]            Hanka, L.J.; Dietz, A.; Gerpheide, S.A.; Kuentzel, S.L.; Martin, D.G. Journal of Antibiotics 1978, 31, 1211.

[2]            Bhuyan, B.K.; Newell, K.A.; Crampton, S.L.; Vonhoff, D.D. Cancer Research 1982, 42, 3532.

[3]            Reynolds, V.C.; McGrovern, J.P.; Hurley, L.H. Journal of Antibiotics 1986, 33, 319.

[4]            Reynolds, V.L.; Molineux, I.; Kaplan, D.J.; Swenson, D.H.; Hurley, L.H. Biochemistry 1985, 24, 6228.

[5]            Hurley, L.H.; Reynolds, V.L.; Swenson, D.H.; Petzold, G.L.; Scahill, T.A. Science 1984, 226, 843.

[6]            Boger, D.L.; Johnson, D.S. Angew. Chem. Int. Ed. Engl. 1996, 36, 1438.

[7]            Boger, D.L.; Boyce, C.W.; Garbaccio, R.M.; Goldberg, J.A. Chem. Rev. 1997, 97, 787.

[8]            Mitchell, M.A.; Kelly, R.C.; Wicnienski, N.A.; Matzenbuhler, N.T.; Williams, M.G.; Petzold, G.L.; Slightom, J.L.; Siemeniak, D.R. J. Am. Chem. Soc. 1991, 113, 8995.

[9]            Borman, S. Chem. Eng. News 1997, 75, No. 46, 37. (b) Boger, D.L.; Zarrinmayeh, H.; Ishizaki, T.; Kitos, P.A.; Suntornwat, O. J. Am. Chem. Soc. 1990, 112, 8961. (c) Hurley, L.H.; Warpehoski, M.A.; Lee, C.S.; McGroven, J.P.; Mitchell, M.A.; Kelly, R.C.; Aristoff, P.A. Biochemistry 1988, 27, 3886.

[10]         Wang, Y.; Gupta, R.; Huang, L.; Luo, W.; Lown, J.W. Anti-Cancer Drug Design 1996, 11, 15.

[11]         Iida, H.; Lown, J.W. Synth. Org. Chem. 1998, 1, 17.

[12]         Boger, D.L.; Garbaccio, R.M.; Jin, Q. J. Org. Chem. 1997, 62, 8875.

[13]         Lin, C.H.; Patel, D.J. J. Mol. Biol. 1995, 248, 162. (b) Lin, C.H.; Patel, D.J. J. Am. Chem. Soc. 1992, 114, 10658.

[14]         Eis, P.S.; Smith, J.A.; Rydewski, J.M.; Case, D.A.; Boger, D.L.; Chazin, W.J. J. Mol. Biol. 1997, 272, 237.

[15]         Boger, D.L.; Garbaccio, R.M. Bioorg. Med. Chem. 1997, 5, 263.

[16]         Boger, D.L.; Turnbull, P. J. Org. Chem. 1998, 63, 8004.

[17]         Boger, D.L.; Garbaccia, R.M. J. Org. Chem. 1999, 64, 8350.

[18]         Carter, S.K.; Crooke, S.T. Mitomycin C: Current Status and New Developments: Academic Press: New York 1979. (b) Mitomycin C in Cancer Chemotherapy Today; Taguchi, T.; Aigner, K.R. Eds Selected Proceedings of the 15th International Cancer Congress, Hamburg, Germany, Aug 16-22, 1990. Excerpta Med. 1991.

[19]         Iyer, V.N.; Szybalsiki, W. Proc. Natl. Acad. Sci. USA, 1963, 50, 355.

[20]         Iyer, V.N.; Szybalsiki, W. Science 1964, 145, 55. (b) Zwelling, L.A.; Anderson, T.; Kohn, K.W. Cancer Res. 1979, 39, 365.

[21]         Schlager, J.J.; Powis, G. Intl. J. Cancer 1990, 45, 403.

[22]         Ernster, L. Methods Enzymol. 1967, 10, 309.

[23]         Beal, H.D.; Hudnott, A.R.; Winski, S.; Siegel, D.; Swann, E.; Ross, D.; Moody, C.J. Bioorg. Med. Chem. Lett. 1998, 8, 545. (b) Beal, H.D.; Mulcahy, T.; Siegel, D.; Traver, R.D.; Gibson, N.W.; Ross, D. Cancer Res. 1994, 54, 3196. (c) Ross, D.; Siegel, D.; Beal, H.D.; Prakash, A.S.; Mulcahy, R.T.; Gibson, N.W. Cancer Metastasis Rev. 1993, 12, 83. (d) Ross, D.; Beal, H.; Traver, R.D.; Siegel, D.; Phillips, R.M.; Gibson, N.W. Oncol. Res. 1994, 6, 493.

[24]         Boger, D.L.; Boyce, C.W.; Garbaccio, R.M.; Seareey, M.; Jin, Q. Synthesis 1999, 1505.

[25]         Warpehoski, M.M. Tetrahedron Lett. 1986, 27, 4103.

[26]         Boger, D.L.; Ishizaki, T.; Sakya, S.M.; Munk, S.A.; Kitos, P.A.; Jin, Q.; Besterman, J.M. Bioorg. Med. Chem. Lett. 1991, 1, 115. (b) Boger, D.L.; Yun, W.; Hat, N. Bioorg. Med. Chem. 1995, 3, 1429.

[27]         Boger, D.L.; Santillan, A.J.; Searcey, M.; Brunette, S.R.; Walkenberg, S.E.; Hedrick, M.P.; Jin, O. J. Org. Chem. 2000, 65, 4101.

[28]         Warpehoski, M.A.; Harper, D.E. J. Am. Chem. Soc. 1994, 116, 7573. (b) Warpehoski, M.A.; Harper, D.E. J. Am. Chem. Soc. 1995, 117, 2951.

[29]         Boger, D.L.; Goldberg, J.; Mckie, J.A. Bioorg. Med. Chem. Lett. 1996, 6, 1955. (b) Boger, D.L.; Mckie, J.A.; Nishi, T.; Ogiku, T. J. Am. Chem. Soc. 1997, 119, 311.

[30]         Boger, D.L.; Ishizaki, T.; Kitos, P.A.; Suntornwat, O. J. Org. Chem. 1990, 55, 5823. (b) Boger, D.L.; Ishizaki, T.; Wysocki, R.J.; Munk, S.A.; Kitos, P.A.; Suntornwat, O. J. Am. Chem. Soc. 1989, 11, 6461.

[31]         Boger, D.L.; Boyce, C.W. J. Org. Chem. 2000, 65, 4088.

[32]         Amishiro, N.; Nagamura, S.; Kobayashi, E.; Okamoto, A.; Gomi, M.; Saito, H. Bioorg. Med. Chem. 2000, 8, 1637.

[33]         Wang, Y.; Yuan, H.; Ye, W.; Wright, S.C.; Wang, H.; Larrick, J.W. J. Med. Chem. 2000, 43, 1541.

[34]         Boger, D.L.; Santillan, A.; Searcey, M.; Jin, Q. J. Org. Chem. 1999, 64, 5241.

[35]         Boger, D.L.; Santillan, A.; Searcey, M.; Jin, Q. J. Am. Chem. Soc. 1998, 120, 11554.

[36]         Reddy, B.S.P.; Sharma, S.K.; Lown, J.W. Curr. Med. Chem. 2001, 8, 475.

[37]         Tao, Z.F.; Fujiwara, T.; Saito, I.; Sugiyama, H. Angew. Chem. Int. Ed. 1999, 38, 650.

[38]         Tao, Z.F.; Fujiwara, T.; Saito, I.; Sugiyama, H. J. Am. Chem. Soc. 1999, 121, 4961.

[39]         Jia, G.; Iida, H.; Lown, J.W. Heterocyclic Comm. 1998, 4, 557.

[40]         Jia, G.; Iida, H.; Lown, J.W. Chem. Comm. 1999, 119.

[41]         Jia, G.; Iida, H.; Lown, J.W. Synlett. 2000, 603.

[42]         Tao, Z.F.; Saito, I.; Sugiyama, H. J. Am. Chem. Soc. 2000, 122, 1602.

[43]         Wemmer, D.E.; Dervan, P.B. Curr. Opin. Struct. Biol. 1997, 7, 355.

[44]         Chang, A.Y.; Dervan, P.B. J. Am. Chem. Soc. 2000, 122, 4856.

[45]         Mitchell, M.A.; Johnson, P.D.; Williams, M.G.; Aristoff, P.A. J. Am. Chem. Soc. 1989, 111, 6428. (b) Mitchell, M.A.; Kelly, R.C.; Wicnienski, N.A.; Hatzenbuhler, N.T.; Williams, M.G.; Petzold, G.L.; Slighton, J.L.; Siemieniak, D.R. J. Am. Chem. Soc. 1991, 113, 8994.

[46]         Fukuda, Y.; Furuta, H.; Shiga, F.; Oomori, Y.; Kusama, Y.; Ebisu, H.; Terashima, S. Bioorg. Med. Chem. Lett. 1997, 7, 1683.

[47]         Fukuda, Y.; Furuta, H.; Kusama, Y.; Ebisu, H.; Oomori, Y.; Terashima, S. Bioorg. Med. Chem. Lett. 1998, 8, 1387.

[48]         Boger, D.L.; Searcey, M.; Tse, W.C.; Jin, Q. Bioorg. Med. Chem. Lett. 2000, 10, 495.

[49]         Jia, G.; Iida, H.; Lown, J.W. Heterocyclic Comm. 1999, 5, 497.

[50]         Jia, G.; Lown, J.W. Bioorg. Med. Chem. 2000, 8, 1607.