|Home | About | Journals | Submit | Contact Us | Français|
Dimeric analogs of Anthracenyl Isoxazole Amides (AIMs) (the designation AIM is in honor of the memory of Professor Albert I. Meyers) were prepared and dimer 6 exhibited the highest efficacy to date for this class of anti-tumor compounds against the human glioma Central Nervous System cell line SNB-19.
We recently reported the anti-cancer activity of a new series of Anthracenyl Isoxazole Amides (AIMs),1 which exhibited significant activity in the 60 Cell Line protocol at the National Cancer Institute (NCI60).2 Our working hypothesis is that this novel class of compounds exert their effect by stabilization of quadruplex (G4) DNA,3 either at the telomere4 and/or specific oncogenes.5
Recently it has been postulated that the telomeric overhang of the human chromosomes possibly form multiple G4 conformers,4a,b if this argument is correct, one classic6 test of this hypothesis would involve the tethering of G4 binding moieties, with the expectation of enhanced biological effect.
Two methods were compared for the preparation of the AIM dimers 4-6. Method A utilized our previously reported lanthanide assisted Weinreb amidation,7 or double activation methodology.1b Method B is a straightforward Schotten-Baumann process.
The advantage of the double activation method is its directness, and it generally proceeds in overall synthetically useful yields after isolation and purification, however, in the case of AIM 4 a 2 to 1 ratio of dimer to mono-AIM incorporation was observed, though the material balance was near quantitative after the recovery of starting material. The double activation procedure was originally developed for amide couplings which were both sterically hindered and either contained functional groups incompatible with thionyl or phosphorus halides (i.e. acridines) or could be potentially chlorinated (i.e., C-10 H anthryl).1b,9 Our recent observation that the C-10 chloro derivative 7 actually possessed higher anticancer activity1a lead us to reexamine the more conventional Schotten-Baumann route. The three step Schotten-Baumann procedure first necessitates a hydrolysis, which due to the steric hindrance of the anthryl moiety proceeded in a very sluggish manner: 60 hours refluxing was required for complete conversion to the carboxylic acid. The next two steps, acid chloride and amide formations, proceed without complication. The overall yield for the three step Schotten-Baumann process, for AIM 4, was 24% after isolation and purification and utilizes reagents amenable to scale-up that are not pyrophoric. The overall yields from Method A and B are comparable.
There is an indication of folding of AIM 4 in solution as evidenced by magnetic anisotropy in the proton NMR. No evidence for folding of the dimers containing pyrrole moieties, AIMs 5 and 6, was observed by NMR.
The anticancer activity of the synthetic dimers assayed against a human glioma cell line, Central Nervous System (CNS) SNB-19, for AIM 4-7 are shown in Table 1. The effect of exposure time was examined for dimer 4, and optimal anticancer activity was observed for 24 hour exposure time. The anticancer activities of dimers 4-6 were benchmarked against monomeric AIM 7, which was among the most active of the AIM series recently reported, as measured against the NCI60, and represents the positive control.1a The anticancer activity of known 7 against SNB-19 cells is in good agreement with the Mean graph mid-point (MG-MID) as examined in the NCI60. The AIMs which are protonated at physiological pH, 4 and 6, were significantly more active than AIM 5, which would be expected to be neutral. AIM 5, while not active against SNB-19 cells, is not devoid of activity. In the NCI 60 cell line one dose study, it exhibited 66.5% tumor cell kill against the SNB-75 cell line (full NCI60 data is presented in the Supplementary Material), as well as activity against non-small cell lung cancer cell HOP-92 (34.04%), ovarian cancer cells lines IGROV1 and OVCAR-4 (35.62 and 32.16%, respectively) and renal cancer cell line UO-31 (39.36%). We attribute the lower activity of 5 in our assay to its high logP (calculated to be ca. 7) and lack of protonation at physiological pH.
AIM 6 exhibited single digit micromolar inhibition of SNB-19, and at 4.75 μM represents the most efficacious analog in this class of compounds to date. Inhibition of the tumor cells lines by AIM 6 in the one dose NCI60 panel also was quite robust: a mean growth inhibition across all cell lines of 81.8% and complete inhibition of tumor cell growth was observed for 25 of the cell lines. In the five dose NCI60 the mid graph mean point for AIM 6 was -5.8, with several cells lines having nanomolar anticancer activity. Thus, the GI50 against Leukemia cells lines SR, MOLT-4, K-562 were 394, 615 and 822 nM respectively.
The AIM 6 dimer, as most AIMs studied to date,1a exhibit useful fluorescence emission at 423 nm in ethanol solution upon excitation at 385 nm, with an approximate extinction coefficient of 20,000. The emission spectrum is shown in Figure 2. The intensity was observed to be enhanced both in buffer at pH 7.5 and in the presence of bovine serum albumin (BSA), and a slight wavelength shift was observed (to 430 and 431 nm respectively. The observations are consistent with restriction of conformation movement in buffer and the presence of BSA.10
The pharmacokinetic properties of the AIM dimers are not ideal for potential therapeutics, in light of the fact that their high molecular weight and lipophilicity both violate Lipinski's rule of five. Lipophilicity was calculated both as ALogP using Symyx Draw, and CLogD using Chem Axon software (Supplementary Material). Even for AIM 6, which would be expected to be protonated at physiological pH, there was only a modest difference in AlogP verses CLogD (7.4) of 7.0 and 6.0, respectively. However, given their robust antitumor activity and fluorescence properties, the bis-AIMs may be useful as probes and potential tools for mechanism of action studies.
In conclusion, tethering AIM moieties increases the anticancer activity of the resulting dimmer, and the most active analog, AIM 6, is - in fact - the most active AIM prepared to date. This is consistent with the hypothesis that multiple G4 conformers may be present in the telomere, and potentially represents a route to distinguish between telomeric and oncogenic G4 DNA. Future studies will be directed towards more detailed structural exploration of this unique target for anticancer drug discovery, and application of the AIMs as fluorescent probes of the mechanism of action. Our progress will be reported in due course.
The authors are grateful to the National Institutes of Health for NINDS 7R15-NS038444-04 and P20 RR015583. We thank Dr. J. B. A. (Sandy) Ross, director of the BioSpectroscopy Core at the Center for Structural and Functional Neuroscience, University of Montana, and Ayesha Sharmin for their assistance with fluorescence spectroscopy, funded by P20 RR015583. SMS acknowledges summer research support from NSF REU CHE 0243760, while NRN was at the University of Idaho. KCR thanks P30 NS055022. We would like to thank Dr. Keith Taylor for helpful discussions on the implementation of Symyx Draw. We thank ChemAxon for use of their software. We also acknowledge Dr. David J. Burkhart of GSKBio for the use of their NMR.
Supplementary Material: Full experimental details for preparation of synthetic dimers 4-6, procedure for Cell Culture and Growth Inhibition Assays. NSC numbers and one-dose data of for 5 and 6 and five-dose data for 6, in the NCI60 protocol. Calculated logD graph for 6. HSQC 2D NMR of dimer 5. Pharmacokinetic computations for AIMs 4-7.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.