|Home | About | Journals | Submit | Contact Us | Français|
We describe the synthesis of 8-heteroaromatic-2′-deoxyguanosine analogues using Suzuki-Miyaura or Stille conditions. Unprotected and protected 8-bromo-2′-deoxyguanosine was coupled with commercially available heteroarylboronic acids or the trialkyltin derivatives of 2-pyridylbromides either with or without microwave irradiation in good yields.
The development of efficient strategies for the synthesis of nucleoside analogues is of great importance due to their potential uses in biological applications. For example, many nucleoside analogues have found uses as antibiotics and antivirals,1 as probes for nucleic acid structure2 and as chain terminators for DNA sequencing.3 Our main interest is in guanine and its derivatives, which arises from the fact that they can self-associate to form macrocyclic, planar tetramers known as G-tetrads. These further associate through π-stacking interactions to form G-quadruplexes (GQs). These GQs can be formed from individual guanine subunits or from G-rich oligonucleotides that fold up into a variety of quadruplex structures.4 The cation templated self-assembly of guanosines into G-quadruplexes has been recognized since the 1960’s.5 More recent studies of G-quadruplex formation have been in diverse areas such as cancer research and nanotechnology.4,6 We have demonstrated the aggregation of 8-aromatic-2′-deoxyguanosine analogues in organic solvents7 and have shown their selective binding of cations in some cases.8 We are particularly interested in the synthesis of 8-heteroaromatic-2′-deoxyguanosine (8-(HetAr)dG) derivatives. Heteroatoms and functional groups in the heteroaryl moieties can be used to increase the number of H-bonds. We have shown that functional groups within an 8-phenyl ring can lead to supramolecular complexes with higher stability.7
Since Shaughnessy’s method9 of using the easily prepared 8-bromo-2′-deoxyguanosine (8-BrdG)10 and commercially available boronic acids had proven successful in the synthesis of 8-aromatic-2′-deoxyguanosine analogues (8-ArdG),7 we decided to apply it to the synthesis of 8-(HetAr)dG targets. However, although the method is useful in the coupling of a variety of heterocycles vide infra, we quickly realized that we needed to use a different method to couple the 2-pyridyl moiety. Because of the expense, reported instability11and relative scarcity of commercially available, functionalized 2-pyridyl boronic acids we decided to use Stille conditions as an alternative method.12 This is not without precedent since the Stille methodology has been used for 2-pyridylstannane couplings to various aromatic skeletons13 including a number of purine analogues.14 Nonetheless, to the best of our knowledge it has never been used to make 8-(HetAr)dG derivatives. In our hands, when we tried to couple heteroaromatics to 3′, 5′-isobutyric ester of 8-bromo-2′-deoxyguanosine (3′,5′-i-Bu-8-BrdG) using heteroaromatic tin compounds, tetrakistriphenylphosphinepalladium in refluxing toluene,14f the sole reaction product was that derived from replacement of the bromide in the substrate with a hydrogen atom. Under these conditions the reaction is extremely sluggish and on this substrate dehalogenation/ hydrogenation ensues. Recently the coupling of heteroaromatics to various guanine analogues has been reported using Stille conditions: refluxing in xylene at 150 °C, tetrakistriphenyl phosphinepalladium and triphenylbismuth as a co-ligand.14a However this work which parallels ours does not try the coupling reaction on 3′,5′-i-Bu-8-BrdG.
Herein we report two convenient methods for the direct functionalization of the C-8 position of 2′-deoxyguanosine with heteroaromatic substituents either on 8-BrdG using commercially available boronic acids (Suzuki-Miyaura conditions) or via 2-pyridylstannanes and (Stille conditions) upon 3′,5′-i-Bu-8-BrdG. The Stille reaction on this substrate is accompanied by a competitive side reaction, which by taking measures to increase the rate of Stille Coupling, can be significantly reduced. The methodology presented not only allows us to make a series of target molecules which we believe will have novel supramolecular properties but it should also be useful to others who wish to functionalise the C-8 position of guanine and its analogues.
Boronic acids (1a, b and c) were found to couple efficiently to 8-BrdG using the Shaughnessy method (Scheme 1, Table 1, Entries 1, 2 and 3) to yield the 8-(HetAr)dG derivatives 2a, b and c.15 It was found however that the coupling reaction was very inefficient with the boronic acids 1d, e, f. Since it has been reported that Suzuki couplings can be extremely efficient under microwave irradiation,16 we modified the methodology by treating 8-BrdG with boronic acids (1d, e, f) in aqueous acetonitrile/DMF mixtures under microwave irradiation, obtaining the 8-(HetAr)dG products (2d, e, f, entries 4, 5, 6).17 As shown in Scheme 1, these 8-(HetAr)dG compounds were converted to their isobutyric esters18 (3a-f) for future studies of their self-assembly in organic solvents.
We next turned our attention to the synthesis of target molecule 4a (Scheme 2). The starting secondary alcohol was protected as the benzyl ether 5a. This was converted to the tributyltin compound 6a in good yield. We applied a modification of the method reported by Brill and Riva-Toniolo,14b where they describe the solid phase reaction of 8-bromo-2,6-diaminopurines with various organostannanes. The initial coupling under homogeneous conditions used DMF as the solvent, copper(II)oxide as additive and much less of the ligand, bisdiphenylphosphinopropane (DPPP). We found that the reaction gave a mixture of the debrominated 7 and coupled product 4a in a 1:1 ratio (entry 2, Table 2). Furthermore, 4a turned out to be a 1:1 mixture of diastereoisomers as determined by proton and carbon NMR, indicating no diastereoselectivity.
However, upon changing the additive to Cu(I)oxide14b and the equivalents of additive to two, we observed a significant improvement in the overall rate of the reaction and on the ratio of the two products (7:4a, Entry 3). Changing the ligand to triphenylarsine, which has weak donicity, led to improved yields of 4a (Entry 4) with further improvement by changing the solvent to N-methylpyrrolidinone (NMP) (Entry 5). The overall optimal combination of ligand and solvent was DPPP with NMP (Entry 6). Under these conditions reducing the amount of additive (Cu(I)O), reduction in the initial amount of tin compound added or increasing the concentration of ligand (DPPP) had adverse effects on the rate of formation of the required product (data not shown). These results are in accord with those of other groups, who have sought to increase the rate of the Stille coupling reaction.14b-e,12b
These results encouraged us to attempt the synthesis of more challenging 8-(2Py)dG targets. The protected bromopyridines (5b-d) were converted to the tributyltin derivatives (6b-d). Each of these was coupled to 3′,5′-i-Bu-8-BrdG as outlined in Table 2 (Entries 7-9).19 As can be seen the coupling reaction proceeds with moderate to good yields in all cases.
We have applied a set of conditions that make the rate of the Stille coupling reaction competitive to a problematic side reaction, obtaining reasonable yields of the target molecules. Using this method or the previously reported Suzuki-Miyaura conditions and a modification of these, we are now able to synthesize a library of 8-(HetAr)dG derivatives. We expect that these methodologies should also be useful to others who wish to attach functional groups to the C-8 position of purines.
We thank the NIH-SCoRE program for financial support, the National Science Foundation EPSCoR program for a graduate fellowship to V.G. and the NASA Space Grant program for a graduate fellowship to M.C.R.S.