Synthesis and cytotoxic evaluation of new terpenylpurines

Departamento de Ciencias Farmacéutica Facultad de Farmacia, CIETUS-IBSAL, Uni de Unamuno, E-37007 – Salamanca, Spai 294515; Tel: +34 923 294528 Centro de Investigação de Montanha (CI Politécnico de Bragança, Campus de San Portugal. E-mail: iferreira@ipb.pt; Fax: +35 † Electronic supplementary information NMR data and assignments for the new c HMQC and HMBC spectra for severa 10.1039/c6ra24254e Cite this: RSC Adv., 2016, 6, 105412


Introduction
The purine heterocycle is one of the most widely distributed in Nature since it can be found in the nucleic acids and in other important primary metabolites as NADP or ATP. 1 Although unsubstituted purine does not exist in Nature, there are a great number of purine derivatives, particularly adenine derivatives, that are involved in numerous metabolic processes.Many structurally modied purine nucleosides and nucleotides are biologically signicant with activities ranging from antineoplastic and antiviral to antihypertensive, antiasthmatic or antituberculosis among others 2 and the scope of therapeutic applications seems to be far from being completed.
Among the purine derivatives, nucleosides and nucleotides, either from natural or synthetic origin, are the best known, but we have put our attention on other natural N-alkylpurines, 3 in which the alkyl chains can be found on any nitrogen atom at the purine moiety.Those alkyl chains can go from a simple methyl group in xanthines to large diterpenoids in agelasines or asmarines. 4ome representative examples of natural N-alkylpurines are shown in Fig. 1 and usually those natural products have served as models for the design and synthesis of a great number of derivatives and analogues for which a variety of biological activities have been described. 2,3n the course of our research towards new antitumour cytotoxics based on natural products, we were particularly interested on those secondary metabolites of marine origin formed by a diterpenoid attached to the 7-nitrogen atom of an adenine derivative as asmarines or agelasimines, some examples are shown in Fig. 1. 3 Those derivatives can be called terpeneadenine hybrids or terpenylpurines.These compounds have attracted the scientic interest because of their biological properties, 3 such as cytotoxic, antimicrobial, antiprotozoal, antifungal, antifouling, etc.
Our interest in this type of compounds is related mainly with the potent cytotoxicity described for some of them.Our research group has been involved for years in the design, synthesis and biological evaluation of cytotoxic compounds related to natural products.We have obtained very interesting results on the chemomodulation of cytotoxicity in podophyllotoxin related lignans, 5 and also in the chemoinduction of bioactivity on inactive terpenoids such as communic acids. 6In this sense, we have synthesized a large number of derivatives, named terpenylquinones that showed very interesting cytotoxicity. 6These terpenylquinones can be considered hybrids of a terpenoid rest and a quinone moiety and can also be considered analogues of other cytotoxic marine natural products such as avarone. 7ecently, we have also described the synthesis and cytotoxicity of a new family of hybrids, lignopurines, between lignans and purines 8 that revealed the importance of the purine core on their cytotoxicity.
This background, and the fact that the natural alkylpurines are usually isolated in very small quantities which limited their structure-activity relationship studies, 3 prompted us to design and prepare new terpenylpurine derivatives starting from natural monoterpenoids and diterpenoids, either commercially available or isolated by us from their natural sources, and to evaluate the inuence of the terpenoid size on their cytotoxic properties.

Chemistry
The N-alkylpurines described in this work have been prepared by the classical procedure of alkylation of purines with alkyl halides. 9As starting materials for the synthesis of terpenyl bromides we used the commercial monoterpenoid myrtenal and the diterpenoids trans-communic and cupressic acids isolated from their natural sources.We also used other commercially available alkyl halides such as 1-bromopentane, cynnamyl bromide, isoprenyl chloride and geranyl bromide.
Myrtenal was easily transformed into the myrtenyl bromide 1 through NaBH 4 reduction followed by substitution of the hydroxyl group with CBr 4 , 10 in this way, compound 1 was obtained in 87% overall yield from myrtenal (Scheme 1).trans-Communic and cupressic acids were isolated from the acid fraction of the n-hexane extract of Cupressus sempervirens L. cones (Cupressaceae).Both acids were quantitatively transformed into their corresponding methyl esters by treatment with trimethylsilyldiazomethane 11 and then transformed into the terpenyl bromides 3, 3 0 and 4 as shown in Scheme 1. Epoxidation of the trisubstituted double bond in methyl transcommunate, followed by oxidation with periodic acid 12 yielded the tetranorditerpenic aldehyde 2, which was transformed into the bromide 3 by reduction and substitution as described for myrtenal.Bromide 3 0 was prepared following the same procedure, previous isomerization of the exocyclic double bond.6b Diterpenyl bromide 4 was obtained in 77% yield by treatment of methyl cupressate with PBr 3 at À35 C. 13 Nucleophilic substitution and allylic isomerization took place at once and compound 4 was obtained as an unresoluble mixture of the E and Z isomers in a 5 : 1 ratio that was used for the alkylation step.The alkylation of purines was performed in DMF, using potassium carbonate or cesium carbonate as a base 14 (Scheme 2).In general, the reaction products were mixtures of alkylpurines, in which the major regioisomer was the corresponding 9-alkylated product (isomer "a").In most of the cases, the minor regioisomer, 7-alkylated purine (isomer "b"), was also separated by chromatographic procedures (Scheme 2, entries 1-9).When the diterpenyl bromide 4 was used (entries 16-19), not only isomers "a" and "b" were detected but also the 3-alkylated products (isomers "c") were isolated and even the Z and E stereoisomers were also separated and characterized on the bases of the chemical shi (d) observed in 13 C NMR spectra for the C-16 0 methyl group on the terpenyl moiety.In the Z isomers, this methyl signal appeared at z23 ppm whereas in the E isomers, d was at z16 ppm. 15For the assignments of the NMR data in the synthesized alkylpurines, the purine positions were numbered from 1 to 9, whereas prime numbers were used for the alkyl side chain, keeping the terpenoid numbering system stated in Scheme 1: from 1 0 to 10 0 in those alkylpurines derived from monoterpenes and from 1 0 to 20 0 in those derived from the diterpenes.In the case of the terpenylpurines obtained from the tetranor-derivatives 3 and 3 0 , the numbering shown in Scheme 1 with primes was used.
All the alkylpurines isolated are shown in Scheme 2. Some of the regioisomeric 7-and 9-pairs of alkylated purines showed characteristic chemical shi differences in their 1 H and 13 C spectra as those described for similar alkylpurines, 16 however we had several compounds in which the differences were not so evident and so that, the position of the radical was unequivocal assigned by two-dimensional HMBC correlations obtained for purines 6a, 7a, 7b, 13a, 13b, 17a, Z-20b, E-20b, E-20c, E-21a, E-  21b, E-21c, E-22a, E-22b, E-22c, Z-23a, E-23a, Z-23c and E-23c.Correlations observed between 2-H, 8-H and the CH 2 group of the side chain attached to N-7 (N-9), with the quaternary carbons C-4 (C-5) in the purine ring were the most determinant.NMR spectra and complete 1 H and 13 C NMR data assignments are included in the ESI.† As an example of the three isolated regioisomers, representative correlations experimentally observed for 13a, 13b and E-22c are shown in Fig. 2.

Cytotoxicity on tumour and normal cell lines
Most of the compounds prepared were evaluated in vitro, using the sulforhodamine B colorimetric assay, 17 to establish their cytotoxicity against the following human tumoural cell lines: NCI-H460 (non-small cell lung carcinoma), HeLa (cervical carcinoma), HepG2 (hepatocellular carcinoma) and MCF-7 (breast adenocarcinoma).The toxicity on non-tumour cell lines was also evaluated using a porcine liver primary cell culture designated as PLP2.The results were expressed as GI 50 , compound concentration in mM that inhibited 50% of the net cell growth and they are shown in Table 1.Only those derivatives that showed GI 50 values lower that 125 mM against one or more cell lines were considered active.In general, those compounds with a GI 50 value under 20 mM were considered good cytotoxic, while values over 70 mM were considered slightly cytotoxic.The starting purines (purine, 6-chloropurine, 6-methoxypurine and adenine) together with the anticancer drugs 6-mercaptopurine and ellipticine were included in the assays as references.
As it can be seen in Table 1, it is possible to establish some general considerations of structure-activity relationship.First,   it could be said that the presence of alkyl groups in a purine system promotes cytotoxicity compared to simple analogues as 6-chloropurine and adenine, which were inactive in the tests, and even better than 6-mercaptopurine which was only active in the HepG2 cell line.
The increasing of chain size led to a better cytotoxic activity.Thus, purines with linear and shorter side chains were inactive (5a, 5b and 6b) or slightly active (6a, 9a, and 9b) against any of the four tumour cell lines tested, only purines 6a and 9b showed an interesting cytotoxicity against HepG2 (19.1 mM) and MCF-7 cells (13.6 mM) respectively.The presence of a phenyl group improved the cytotoxicity as happened with 8a and 8b, which had a cinnamyl residue and showed moderate activity.
Among the monoterpenyl substituents, a linear geranyl substituent was better than the bulky pinene moiety (10a, 10b vs. 11a, 12b, 13b) with similar GI 50 values for all the tumour cell lines and the same applied to the evaluated tetranorditerpenyl derivatives 15a, 17a and 18a.It is interesting to note the activity of the purine 8a against MCF-7 line (15.3 mM) and the purine 17a against HepG2 (21.7 mM) without showing toxicity in the primary line PLP2.
Purines with a diterpenoid moiety included the most cytotoxic derivatives obtained, although the terpenoid is not as determinant as the substitution on the C-6 position of the purine ring: 6-chloropurine derivatives (21) were most potent than purine (20), 6-methoxypurine (22) and adenine (23) derivatives, independently if the terpenoid was at N-9, N-7 or N-3.Exceptionally some differences were observed between several Z/E isomers at the diterpenoid moiety, being more potent the E-isomers (E-20a, E-20b, E-21b and E-22a vs. Z-20a, Z-20b, Z-21b and Z-22a).
The most cytotoxic of all compounds tested on the different tumour cell lines were diterpenylpurines E-21a and E-21b (3.30-13.7 mM), being several times more potent against tumour lines than against non-tumour line PLP2.Particularly, compound E-21a was the most potent against NCI-H460 line (3.98 mM) and E-21b presented the best value cytotoxicity against HeLa cell line (3.30mM).Both compounds improved cytotoxicity values of ellipticine in NCI-H460 and HepG2 tumour lines and besides showed less toxicity to non-tumour cells.

Conclusions
As a conclusion, several new terpenylpurine derivatives were efficiently prepared through alkylation of different purines with halogenated reagents derived from natural terpenoids, commercially available or isolated from their natural sources.Thus, cupressic acid was easily isolated in a good yield from cones of C. sempervirens L. and further transformed into appropriate alkylated agents.Alkylation of the purines gave mixtures of 9-and 7-alkylpurines, being the 9-alkylpurines the major regioisomers.Sometimes the 3-alkylpurine derivatives were also isolated in low yield from the reaction product.The presence of the terpenyl residue induced cytotoxicity on simple purines and, in general, that activity improved as the substituent was larger, like those present in the marine terpenylpurines.Although more derivatives are necessary to obtain more signicant conclusions on structure-activity relationship, the fact that derivative E-21b was the most cytotoxic in the series, encourage us to continuous with our research towards the selective preparation of 7-alkylated purines, which could be considered analogues of agelasines and agelasimines, which were the marine natural terpenylpurines taken as models for this work.deg cm 2 g À1 .UV spectra were obtained on a Hitachi 100-60 spectrophotometer.IR spectra were obtained on a Nicolet Impact 410 spectrophotometer in NaCl lm.HRMS were run in a QSTAR XL Q-TOF (Applied Biosystems) using electrospray ionization (ES) at 5500 V with an HPLC Agilent 1100 chromatograph.Solvents and reagents were puried by standard procedures as necessary and the chlorinated solvents, including CDCl 3 , were ltered through NaHCO 3 prior its use, in order to eliminate acid traces.DMF was dried over molecular sieves and treated with K 2 CO 3 for the same reason.Column chromatography (CC) purications were performed using silica gel 60 (40-63 mm, 230-400 mesh, Merck).
4.1.2.10-Bromo-2-pinene 1.To a solution of myrtenal (500 mg, 3.3 mmol) in THF (8 mL) NaBH 4 (400 mg, 10.5 mmol) was added and kept stirring at room temperature for 8 h.Then, the reaction mixture was quenched with a saturated aqueous solution of NH 4 Cl and extracted with EtOAc.The combined organic layers were washed with brine and dried over anhydrous Na 2 SO 4 .Removal of the solvent gave a reaction product that was dissolved in CH 2 Cl 2 (10 mL); then, CBr 4 (1.58 g, 4.78 mmol) and PPh 3 (1.25 g, 4.78 mmol) were added at 0 C and stirred at this temperature for 30 min.The solvent was evaporated and ltered over celite and chromatographed on silica gel to give 1 (600 mg, 84%).

Cytotoxicity assays
Four human tumour cell lines were used: NCI-H460 (non-small cell lung cancer), HeLa (cervical carcinoma), HepG2 (hepatocellular carcinoma) and MCF-7 (breast adenocarcinoma) from DSMZ (Leibniz-Institut DSMZ -Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH).Cells were routinely maintained as adherent cell cultures in RPMI-1640 medium containing 10% FBS and 2 mM glutamine, 100 U mL À1 penicillin and 100 mg mL À1 streptomycin, at 37 C, in a humidied air incubator containing 5% CO 2 .The cell line was plated at an appropriate density (1.0 Â 10 4 cells per well) in 96-well plates and allowed to attach for 24 h.Cells were then treated for 48 h with the different diluted compound solutions.Aerwards, sulforhodamine B assay 17a was performed as follows: cold trichloroacetic acid (TCA 10%, 100 mL) was used in order to bind the adherent cells and further incubated for 60 min at 4 C. Aer the incubation period, the plates were washed with deionised water and dried and sulforhodamine B solution (SRB 0.1% in 1% acetic acid, 100 mL) was then added to each plate well and incubated for 30 min at room temperature.The plates were washed with acetic acid (1%) in order to remove the unbound SRB and then air dried; the bound SRB was solubilised with Tris (10 mM, 200 mL) and the absorbance was measured at 540 nm using an ELX800 microplate reader (Bio-Tek Instruments, Inc.; Winooski, VT, USA).The results were expressed as GI 50 values; compound concentration that inhibited 50% of the net cell growth.Ellipticine was used as positive control.
For hepatotoxicity evaluation, a cell culture was prepared from a freshly harvested porcine liver obtained from a local slaughterhouse, according to an established procedure 17b and it was designed as PLP2.Cultivation of the cells was continued with direct monitoring every two to three days using a phase contrast microscope.Before conuence was reached, cells were subcultured and plated in 96-well plates at a density of 1.0 Â 10 4 cells per well, and cultivated in DMEM medium with 10% FBS, 100 U mL À1 penicillin and 100 mg mL À1 streptomycin.Cells were treated with different concentrations of the compounds and SRB assay was performed as previously described.The results were expressed as GI 50 values; sample concentration that inhibited 50% of the net cell growth.Ellipticine was used as positive control.
For each compound, two independent experiments were performed, each one carried out in duplicate.The results are expressed as mean values and standard deviation (SD).

Scheme 2
Scheme 2 Preparation of the alkylpurines 5-23 by treatment with alkyl halides.