Synthesis and Cytotoxic Activity of Some New 2,6-Substituted Purines

A seriesof twenty four acyclic unsaturated 2,6-substututed purines 5a-20b were synthesized. These compounds were evaluated for cytotoxic activity against NCI-60 DTP human tumor cell line screen at 10µM concentration. N9-[(Z)-4'-chloro-2'-butenyl-1'-yl]-2,6-dichloropurine(5a), N9-[4'-chloro-2'-butynyl-1'-yl]-2,6-dichloropurine(10a), N9-[(E)-2',3'-dibromo-4'-chloro-2'-butenyl-1'-yl]-6-methoxypurine(14) and N9-[4'-chloro-2'-butynyl-1'-yl]-6-(4-methoxyphenyl)-purine(19) exhibited highly potent cytotoxic activity with GI50 values in the 1–5 µM range for most human tumor cell lines. Other compounds exhibited moderate activity.


Introduction
According to WHO report on cancer about 7.6 million people died in the year 2005 and the number is expected to raise to 9 million by the year 2015 and 11.5 million by 2030 [1]. Hence, development of new potent and selective anticancer agents has become one of most intensely pursued goals in drug development around the world. Neplanocin A, (1, Figure 1) is considered a carbocyclic analogue of a natural nucleoside and has shown potent antitumor and antiviral properties [2][3][4]. As a part of our research program on the synthesis of anti-cancer agents, we have synthesized some aromatic neplanocin-A analogues like 3a-3b, 4a-c [5][6][7].

OPEN ACCESS
The N 9 -hydroxymethyl analogues of adenine, guanine and 2,6-diaminopurine related to 3a-3b did not exhibit any anticancer activity, however, their N 9 -chloromethyl arylpurine intermediates, related to 4a-4c (Figure 1), were found to be potent in vitro growth inhibitors of several human tumor cell lines. These results prompted us to consider purines with an unsaturated N 9 -linker that has been terminated with a chloromethyl group. In the synthesis of anti-cancer purine compounds, many times either the purine base is modified or the sugar moiety is modified or replaced with a non-sugar linker or sometimes all these changes have been done by researchers simultaneously in an attempt to make an active compound. We have followed a very similar path in the present work.
Purine base selection: 2,6-Dichloropurine is selected where the chlorines are expected to serve as powerful electron withdrawing centers on the purine ring. 2-Chloro-6-methoxypurine is expected to serve a double role, with electron withdrawing and electron donating centers on the purine ring. The 6methoxy group was selected for its electron donating nature to the purine ring. 6-(4-Methoxy)phenyland 6-(4-fluoro)phenyl-substituted purines were selected to significantly alter the purine base properties and to improve the lipophilicity. It is interesting to note that purines with those substitution patterns were also reported to elicit wide range of anti-viral and anti-cancer activities. 6-Methoxypurine arabinoside was reported as a potent inhibitor of Varicella-Zoster virus [8]. 6-Methoxy group-containing Nelarabine and the 2-chloro group-containing compound Clofarabine elicit anticancer activities [9]. Further, 6-(4-methoxphenyl)purine and 6-(4-fluorophenyl)purine ribonucleosides were reported to elicit significant cytostatic activity [10].
Linker selection: We chose linkers like cis-1,4-dichlorobutene, trans-1,4-dichlorobutene and 1,4dichlorobutyne. All these linkers are acyclic five carbon length open chain liners analogous to the linker of acyclovir, with some degree of unsaturation. All these liners are common for each of the above purine bases selected, like 2,6-dichloropurine, 6-methoxyurine, 2-chloro-6-methoxypurine, 6-(4methoxyphenyl) purine and 6-(4-fluorophenyl)purine. Reaction of each of these purine bases with cis-a 1,4-dichlorobutene linker furnishes N 9 substituted purines with methylchloromethyl-cis-butene units, e.g., compounds 5a, 6, 7, 15a and 16a. Reaction with the trans-1,4-dichlorobutene linker furnishes N 9 substituted purines with methylchloromethyl-trans-butene units, e.g., compounds 8a, 17 and 18a. Reactionwith1,4-dichlorobutyne is expected to furnish N 9 substituted purines with methylchloro-methyl-butyne moieties, e.g. compounds 9a, 10a, 11, 12, 19 and 20a. Compounds 13 and 14 represent vinylic dibromides, a new class of purines, which were also synthesized in this work to assess their cytotoxic activity. This plan gives an opportunity for us to assess the cytotoxicity for a group of purine compounds ( Figure 2) and to understand how the activity is changing for a given linker with a change on the substitution pattern on the purine ring.   Here, we have focused primarily on the synthesis of N 9 substituted purines because they were found to be more active when compared to N 7 isomers at our laboratory. Furthermore the N 7 isomers are expected to be minor products in the synthesis.

Chemistry
The N 9 -alkylated compounds 5a-20b were prepared by the direct alkylation approach on the appropriately substituted purine bases in presence of K 2 CO 3 in dimethyl formamide (DMF) medium (Schemes 1 and 2). A 1-3 fold excess of the alkylating agent and anhydrous potassium carbonate were employed for one equivalent of the purine base taken to isolate N 9 -purine isomers as the major product in moderate to good yields. A ten equivalent excess is not required as previously reported [12]. Minor dimeric products ( Figure 3) have been isolated whenever formed during the synthesis for each linker. Increasing the reaction time and the molar ratio of the potassium carbonate favors the higher yields of the dimeric products. The UV maxima for N7 isomers were 10-15 nm higher (275-320 nm) than the N 9 -isomers (265-310 nm) [13].
Minor modifications were made to the Suzuki-Miyaura cross coupling procedure [10,14]. Reaction of appropriate phenylboronic acid with 9-(tetrahydropyran-2-yl)-6-chloropurine under Suzuki-Miyaura cross coupling methodology afforded 6-(4'-methoxyphenyl), 6-(4'-fluorophenyl)-purines (Scheme 2). The reported procedure of pyridiniumtribromide bromination of the acetyleneic compounds [11] was adopted with minor modifications to furnish the vinylicdibromides13, 14 (scheme 3). Acetyl chloride-mediated THP protection and deprotection of hydroxyl functional groups on a wide range of aliphatic and aromatic systems has been reported [15]. We have extended the concept to 6-chloropurine and found acetyl chloride to be a versatile clean inexpensive deprotective agent in methanol medium for the THP removal. No side products are found on the procedure we described, although Dowex 50W X 8 works [10]. During the protection of the NH of 6-chloropurine, acetyl chloride was found to form one side product, possibly an N-acetyl derivative, still the % yield was about 70 after column purification. The pyridinium-p-toluenesulfonate (PPTS) catalyzed THP protection of 6-chloropurine was found to be relatively very clean, no major side products formed and the yield was about 90%. Hence we chose PPTS catalyst for the THP protection of 6-chloropurine.
Reaction of sodium methoxide with 6-chloropurine and 2,6-dichloropurine in methanol medium furnished 6-methoxypurine and 2-chloro-6-methoxypurine.respectively. Further reaction of these purine bases with various alkenyl and akynyl linkers as explained above furnished the target compounds 5a-20b. The structures of all the purines 5a-20b were confirmed by 1 H-NMR, 13 C-NMR, LC-MS and satisfactory elemental (C, H, N) analysis within ± 0.4% of theoretical values.
Out of these compounds 6, 7, 8a, 11, 12, 15a, 15b, 16a, 16b, 20a, 20b exhibited growth %in the range of 70-100 plus and hence may be considered inactive. Table 1 summarizes the single dose 10 µM test results for the active compounds. Compounds 5a, 5b, 8b, 10a, 10b and14 elicited significant cytotoxicity on almost all the cell lines such as leukemia, non-small cell lung cancer, colon cancer, CNS cancer, melanoma, ovarian cancer, renal cancer, prostate cancer and breast cancer. Under the same single dose testing, compound 17 elicited cytotoxicity to MDA-MB-468 breast cancer cell line while compound 19 elicited cytotoxicity to LOXIMVI melanoma and UO-32 renal cancer cell lines. Table2 represent the five dose testing results GI 50 and LC 50 for compounds 5a, 10a, 14 and 19. Compound 5a was found very active with GI 50 values 1-2 µM for almost all the cell lines. Compound 10a was also found very active with GI 50 values under 2 µM for many cell lines. For leukemia HL-60TB and melanoma UACC-62 the GI 50 values are 3.5 and 3.7 µM respectively. Compound 14 displayed impressive activity, with GI 50 values of 2-3 µM for leukemia, melanoma, renal cancer and breast cancer. It also elicited significant activity on non-small cell lung cancer, colon cancer and CNS cancer. Compound 19 exhibited striking activity, with GI 50 values of 2-4 µM for breast cancer and 2-8 µM for leukemia, colon, renal and prostate cell lines. Replacing the chlorine at 6-position in compound 5a with a methoxy group results in compound 11. Similarly replacing6-chlorine in compound 5a with a methoxy group results in compound 12. These changes resulted in a total loss of cytotoxicity. When the N-9 cis-butene stereochemistry in 5a is changed to a trans form as in 8a also resulted in the loss of activity, although the corresponding dimer 8b with a trans stereochemistry elicited good cytotoxicity.
On the 6-phenyl substituted compounds, only 19 elicited good activity and all other compounds were inactive. The reported procedure [11] was employed to transform the triple bond compounds 9a and 11 in to the corresponding vinylicdibromides13 and 14(Scheme2).Indeed one of the vinylic dibromide with a 6-methoxy substituent,14, was found very active for leukemia, melanoma, renal cancer, breast cancer (GI 50 value 2-3 µM) and significant activity on all other cancers. The other vinylic dibromide with a 6-chlorine group 13 did not elicit any cytotoxicity. Compounds 18a and 18b were not tested.

Pharmacology
A total of 60 human cell lines, derived from nine cancer types (leukemia, lung, colon, brain, melanoma, ovarian, renal, prostate, breast) formed the basis of this NCI-60 DTP human tumor cell line screen [16,17].The tumor cells were cultured in RPMI1640 medium supplemented with 5% fetal bovine serum and 2 mM L-glutamine. The tumor cells were inoculated in to 96-well microtiter plates, 100 µL at plating densities ranging from 5,000 to 40,000 cells/well depending on the doubling time of the individual cell lines [16][17][18][19].After this cell inoculation, the microtiter plates are incubated at 37 °C, 5% CO 2 , 95% air and 100% relative humidity for 24 h. Two plates of each cell line are fixed in situ with TCA to represent a measure of the cell population (T0)before adding the target compounds. The target compounds are dissolved in DMSO and diluted in the test medium to obtain the desired concentration. 100 µL of each of the test compound solution is now added to the above appropriate cell line microtiter wells and incubated for 48 h at 37 °C, 5% CO 2 , 95% air and 100% relative humidity. A sulforhodamine B (SRB) protein assay was used to estimate cell viability or growth. The cytotoxic effects were evaluated and the assay results and dose response parameters calculated as previously described [17,[20][21][22]. At the present time at NCI the target compounds were tested initially at a single dose at 10 µM concentration and those promising target compounds are further tested at five dose testing concentrations 0.01, 0.1, 1.0, 10, 100 µM.
Concentration parameters GI 50 , TGI and LC 50 : The NCI re-named the IC 50 as GI 50 . GI 50 value represents the concentration of the target compound that causes 50% growth inhibition, that is derived from the formula 100 × (T -T0) / (C -T0) = 50, where T is the optical density of the target compound after 48 h exposure. T0 is the optical density at time 0 and C is the control optical density. TGI represents the concentration of the target compound where 100 × (T -T0) / (C -T0) = 0 and it is the cytostatic effect. LC 50 is the concentration of the target compound where 100 × (T -T0) / T0 = -50. LC 50 also signifies the cytostatic effect and the control optical density is not used in the calculation.

General
Unless otherwise stated, all chemicals and reagents were purchased from Sigma-Aldrich Chemical Co. Melting points were determined on an Electrothermal MEL-TEMP apparatus and are uncorrected. 1 H-NMR and 13 C-NMR spectra were recorded in DMSO-d 6 on a Bruker 500 MHz instrument and the chemical shift (δ) values are reported in parts per million (ppm)relative to TMS. A Thermo Scientific LTQ Linear Trap LC/MS/MS system was used for mass spectrometry. UV spectra were recorded on a Beckman-Coulter DU-800 spectrophotometer. Analytical TLC was carried out on Sigma-Aldrich (cat # Z122785-25EA), 0.2 mm percolated silica gel polyester sheets with UV indicator. Elemental analysis was carried out by M-H-W Laboratories, Phoenix, AZ. Analysis of C, H, N were within ± 0.4% of theoretical values. The carbon numbering was shown for representative monomer 20a and for one representative dimer 20b. All others were referred similarly on the 13 C-NMR assignments.

General Procedure-A for N 9 -[(Z)-4'-chloro-2'-butenyl-1'-yl]-6-methoxypurine (6)
Synthesis of 6-methoxypurine: To a stirred suspension of 6-chloropurine (5.0 g, 32 mmol) in anhydrous methanol (240 mL) was slowly added 30 W% sodium methoxide in methanol (17.3 g, 320 mmol) at room temperature and then the reaction mixture was refluxed for 18 h. It was cooled to room temperature, neutralized with glacial acetic acid to pH 7.5-8.0 and then evaporated in a rotary evaporator to remove the solvent. The residue was treated with cold water (5 °C) (100 mL), the resulting solid was filtered, thoroughly washed with DI water. The product was crystallized from methanol as brownish white solid (3.5 g), 72% yield. 1  To a suspension of 6-methoxypurine (1.51 g, 10 mmol), anhydrous potassium carbonate (2.1 g, 15 mmol) in DMF (50 mL) at room temperature, was added cis-1,4-dichloro-2-butene (1.25 g, 10 mmol) and the contents stirred at room temperature for 5 h. The reaction mixture was filtered to remove the potassium carbonate that was also washed with DMF (25 mL). The filtrate and washings combined and evaporated under vacuum. The residue was chromatographed on a column of silica gel and the product was eluted with ethyl acetate: hexane 1:1 v/v. Evaporation of the homogeneous fractions resulted in a residue that was further crystallized from ethyl acetate-hexane as cream white rosettes, (1.