SciELO - Scientific Electronic Library Online

vol.45 issue3Kinetics of the Water Gas Shift Reaction Catalyzed by [Rh(COD)(4-Picoline)2]PF6 Immobilized On Poly(4-vinylpyridine)CATALIZADORES Ru/SiO2 y Ru/Al2O3 PREPARADOS POR SOL-GEL PARA HIDROGENACION DE TOLUENO author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Boletín de la Sociedad Chilena de Química

Print version ISSN 0366-1644

Bol. Soc. Chil. Quím. vol.45 n.3 Concepción Sept. 2000 

The Syn-Anti Equilibrium of Guanosine Cyclic 3',5'-
Monophosphate and 8-Sustituted Derivatives: A
Theoretical Study.


Departamento de Química Orgánica, Universidad de Concepción, Casilla 3-C,
Concepción , Chile.
(Received: Frebuary 2, 2000 - Accepted: April 19, 2000)


The syn-anti conformational equilibrium for a series of nine cyclic nucleotides was studied at semiempirical AM1 level. The AM1 results indicate that for guanosine cyclic 3',5'-monophosphate and their derivatives substituted at 8-position, the syn-conformers are more stable than the anti ones. For purine derivatives without an amino group at position 2, which participate in a hydrogen bonding with the axial oxygen of the phosphate group, the anti conformation is preferred.

The AM1 results suggest that syn-anti equilibruim is a dominant factor for the binding of cyclic nucleotides to cyclic phosphodiesterases. Reinterpretation of experimental results for the competitive inhibition of c-GMP-stimulated phosphodiesterase was performed.

Keywords: AM1, cyclic nucleotides, syn-anti equilibrium, phosphodiesterases.


Se estudió el equilibrio syn-anti de una serie de nueve nucleótidos cíclicos utilizando el método AM1. Los resultados indican que para el 3',5'-monofosfato cíclico de guanosina y sus derivados sustituídos en la posición 8, el isómero syn es más estable que el isómero anti. Los derivados de purina que no poseen un grupo amino en la posición 2, el cual puede participar en enlace de hidrógeno con el oxígeno axial del grupo fosfato, prefieren la conformación anti.

Los resultados obtenidos por el método AM1 sugieren que el equilibrio syn-anti es un factor dominante para la union de nucleótidos cíclicos a fosfodiesterasas. Según esto se hizo un nuevo análisis de los resultados de la inhibición competitiva de la fosfosdiesterasa cíclica estimulada por c-GMP.

Palabras claves: AM1, nucleótidos cíclicos, equilibrio syn-anti, fosfodiesterasas.


Both guanosine cyclic 3’, 5’ –monophosphate, c-GMP, and adenosine cyclic 3’, 5’-monophosphate, c-AMP; are very important biological molecules which mediate a series of hormonal and regulatory cellular mechanism1. The ring opening of these cyclic nucleotides, enzymatic hydrolysis is performed by multiple cyclic nucleotide phosphodiesterases2. Recently, the existence of 20 distinct isozymes3 which present a great diversity, has been reported. The experimental data suggest that cyclic nucleotide can interact with the isozyme through specific interactions and that syn-anti equilibrium should play an important role in the enzymatic inhibition performed by cyclic nucleotides analogues. The knowledge of these interactions and topology of catalytic sites should have important applications in the development of selective inhibitors that can manipulate the intracellular levels of cyclic nucleotides and also in future pharmacological applications.

Beltman4 et al., have tested a series of cyclic nucleotide derivatives as competitive inhibitors of cyclic nucleotide phosphodiesterase isozymes in an attempt to define the interactions of c-GMP with the catalytic sites of cyclic nucleotide phosphodiesterases. These authors found that some isozymes show a preferential tendency to bind cyclic nucleotide analogues that present a particular conformation, for example the anti conformation. Then, the study of this conformational equilibrium is very important to interpret the experimental data along with the specific interactions due to purine bases. In the case of 8-Br and 8-OH c-GMP derivatives they presented IC50 values from 2.3 to 65 times greater than c-GMP and 12 to 230 times greater than c-GMP, respectively. In particular for the isozymes c-GMP binding phosphodiesterase and c-GMP inhibited phosphodiesterase theIC50 values for the 8-Br-derivative are 32 and 65 times greater than c-GMP while that of 8-OH-derivative the IC50 values are 140 and 300 times greater than c-GMP. Both isozymes show a preferential binding of the anti conformer of c-GMP. The shift of syn-anti equilibrium produced by substitution at C8 by bromide and hydroxyl group could explain the difference found in the IC50 values. However, other specific interactions can be modified by 8-substitution at purine bases. Also, in 8-hydroxy-derivative the tautomeric equilibrium should play an important role.

It is known that the tautomeric properties of puric bases have an important role in the modulation of their biological properties5. In particular, the solvent effects and intrinsic stabilities determine the predominance of some particular tautomer. Thus, for guanine6, thioguanine7 and hipoxanthine8, it has been shown that in solution, the keto and thione tautomers are the most stable species. For 6-thioguanine the thiol species are the predominant forms in gas phase9,10. Thus, the tautomeric preference of the bases should have an important role in the binding properties of the cyclic nucleotide. Also the tautomeric preference determines the conformational stability of the nucleotide.

In the present work we studied, by using AM1 semiempirical method, the syn-anti equilibrium of nine cyclic nucleotides (figure 1). It was also examined the effect of 8-Hydroxy substitution on tautomeric properties of the puric base. A series of molecular properties and reactivity indices were calculated for the model compounds, N9-metil-8-Hydroxyguanine at AM1, HF and DFT levels.

Fig. 1 Structures of cyclic nucleotides. All nucleotides are shown in anti conformation

Computational aspects

Calculations were performed by using AM1 method11 implemented on Gaussian 94W series of programs12. Ab initio HF13 and DFT14 calculations performed on the N9-methyl-8-hydroxyguanine were carried out by using optimized structures at HF/6-31G* level. All the optimized structures were minima on potential energy surface. The quantum mechanical descriptors, used in the multilinear regression analysis, were evaluated from AM1 molecular orbital calculation at optimized geometry of the cyclic nucleotides. Atomic charges were obtained from a Mulliken population analysis15,16.

Results and discussion

Tautomerism of 8-hydroxy-guanine.

In the study of protomeric tautomerism of 8-OH-cGMP only two tautomeric forms were considered; the hydroxy and oxo tautomers are shown in figure 2. The tautomers product of hydrogen migration in the pyrimidinic ring were not considered due to experimental and theoretical evidence which indicate that keto tautomer is the predominant tautomeric specie in the guanine derivatives6,8. Table 1 shows the relative energies for both tautomers in the anti and syn conformation. Due to AM1 tendency to stabilize some particular tautomer calculations at ab intio HF and DFT levels were carried out, on the model compound N9-Methyl-8-Hydroxyguanine. The results are also shown in the table 1, all methods: AM1, ab initio HF and DFT indicate that the keto tautomer is the more stable tautomer by 15 kcal mol-1. Some molecular properties of both tautomers of the N9-Methyl-8-hydroxyguanine are given in table 2. Table 2 shows the dipole moments, EHOMO, ELUMO, the first vertical ionization potential, the first vertical electronic affinity and chemical hardness for both tautomers, calculated at B3LYP/6-31G*//HF/6-31G* level of theory. From the molecular orbital analysis, a p-type character for the HOMO is deduced for both tautomeric species. It could be suggested that the electron donor properties of both tautomers are independent of the position of the interchangeable hydrogen atom, and that they are of p-type.

Fig. 2 The Hydroxy and oxo tautomers of N9-Methylguanine

On the other hand, the LUMO wave function is the p-type for both, the Hydroxy and oxo tautomers. In this case, there exists a minor dependence of the tautomeric form with a small modification of the electronic distribution of the imidazolinic ring, mainly at N7, C8 and N9 atoms.

Table I. Energy difference of Hydroxy Û Oxo protomeric equilibrium for 8-OH-c-GMP and N9-methyl-8-hydroxyguanine calculated at different levels of theorya.

a) The ab intio and DFT calculations were merfprmed at Hf/6-31G* optimized geometry
b) Hydroxy Û Oxo energy differences for ant conformers.
c) Hydroxy Û Oxo energy difference for syn conformers.

The calculated values for the first vertical ionization potential (IP) and the first vertical electron affinity (EA) are related with the following processes: Neutral species ® Monocationic radical and Neutral specie ® Monoanionic radical, respectively. The IP for the oxo-tautomer, i.e, the most stable tautomeric form, is greater than the hydroxy ones. While that of EA values for oxo-tautomer is less than that of the hydroxy-tautomer. In table 2 the chemical hardness for both tautomers of N9-Methyl-8-hydroxyguanine are also given, calculated at B3LYP/6-31G* level. Finite difference approximation17 was used to evaluate this molecular property. From table 2 it can be observed that oxo species has the great hardness. These results are in accordance with the principle of maximum hardness18 which states, that for a constant external potential, the system with maximum hardness is the most stable18,19. This principle has been verified for many systems20,21 and recently it was verified in a series of tautomeric acetyl derivatives22.

Table II. Diople moments, HOMO and LUMO energies, vertical ionization potential, vertical electron affinity and chemical hardness for hydroxy and oxo tautomers of N)-Methylguaninea.

Syn Û anti conformations of cyclic nucleotides.

The greater conformational flexibility of cyclic nucleotides analogues studied in this work is associated with the glycosyl bond, i.e, the torsional angle [O1'-C1'-N9-C8]. There are two conformational minima for rotation around this bond; the syn and anti conformations. The energetic and structural results for the cyclic nucleotide analogues are shown in table 3. The glycosyl torsion angle for the syn conformers have values between 850 and 950. The anti conformers have glycosyl torsion angle around 550 in all the cyclic nucleotides studied. The exception is 8-OH-cGMP derivative, which in the anti conformation has a value of 36.60. This is due to hydrogen bonding interactions between the hydrogen atom of hydroxyl group at position 8 and O1' atom of the ribose, which stabilize this particular conformation.

Table III. AM1 calculated energy difference for syn Û anti equilibria of cyclic nucleotide analogues and quantum chemical descriptors.

From AM1 calculations it can be inferred that c-GMP and its analogues prefer the syn conformation while for c-AMP and cyclic nucleotides without amino group at position 2 of the purinic base, the anti conformation is the most stable. Such changes in the conformational stabilities can be associated with intramolecular hydrogen bonding between the hydrogen atom of the amino group, at position 2 of the base, and the axial oxygen atom of the phosphate group (figure 3). Thus, the hydrogen bonding distance between these atoms in the cyclic nucleotides studied are: 2.22 A0 for c-GMP, 2.18 A0 for 8-Br-cGMP, 2.20 A0 for 8-OH-cGMP, 2.14 A0 for 8-oxo-cGMP and 2.21 A0 for c-TGMP. All these distances are less than the sum of oxygen and hydrogen van der Waals radii23. Then, in these compounds a potential hydrogen bonding can be assumed24. Other authors have also appealed to hydrogen bonding interactions to explain the preference of c-GMP by the syn conformation25. However, the energy associated with this interaction is not easily determined. The shift of syn Û anti equilibrium towards the syn conformer in c-GMP can be related to two factors: the oxo substitution at C6 of the purine and hydrogen bonding between the amino group of the purine with the axial oxygen of the phosphate group. These effects can be approximately evaluated comparing the DE values for c-PMP and c-IMP, for the oxo effect, and the DE values of c-PMP and 2-NH2-c-PMP, for the hydrogen bonding interactions. Likely, we have determined that the oxo substitution at C6 of the purine shifted the equilibrium towards the syn conformation by approximately 2.1 kcal mol-1. While, the hydrogen bonding interactions also shifted the equilibrium in direction to syn conformer by 2.0 kcal mol-1.

Fig. 3 Optimized structures, at AM1 level, of yhe syn and anti conformers of c-GMP

The syn Û anti equilibrium of cyclic nucleotide analogues should play an important role in the inhibition of cyclic nucleotide phosphodiesterases due to the conformational preference that these enzymes present for the binding of cyclic nucleotides. For example, the 8-substitution at purine should shift the equilibrium toward the syn conformer and consequently decrease the percentage of the cyclic nucleotide in the anti conformation. This fact would increase the amount of derivatives required for 50 % inhibition if the enzyme have a high preference towards the anti conformation. In table 3 it can be observed that the substitution at position 8 of c-GMP by Br and OH (lactam tautomer) shift the equilibrium towards the syn conformation. These tendencies are in agreement with the relative IC50 values reported for c-GMP-inhibited phosphodiesterase4; 1.0 for c-GMP, 65 for 8-Br-c-GMP and >230 for 8-Oxo-GMP, respectively. It is also possible to find a direct correlation between the relative IC50 values for the inhibition of c-GMP-binding phosphodiesterase and the shifted syn-anti equilibrium of the cyclic nucleotide derivative substituted at position 8.

The AM1 calculations also allow to reinterpret the experimental data for c-GMP-stimulated phosphodiesterase. For this isozyme, Beltman4 et al. have concluded that the enzyme prefer to bind the syn conformer of the cyclic nucleotide. However, from the values of IC50 displayed in the table 3, it can be observed that there exists a direct relationship between them and the values of DE for the syn Û anti equilibrium of the cyclic nucleotides. All these derivatives that have positive DE values and greater than DE for c-GMP (1.32 Kcal.mol-1) are poorer inhibitors. Moreover, derivatives that have negative DE are stronger inhibitors than c-GMP with exception of c-IMP and c-PMP. However, in the case of these nucleotides, other interactions should also play an important role, for example hydrogen bonding, due to the interaction of the amino group at position 2 and the carbonyl oxygen at position 6 with specific sites into the isozyme. The loss of amino group in c-IMP and the loss of both the oxygen atom and amino group in c-PMP should increase the C50 values for these nucleotides in spite of DE values for the syn-anti equilibrium that these nucleotides present. If these assumptions are valid, a correlation between the values of C50 with DE and hydrogen bonding descriptors should be observed.

A Multilinear Regression Analysis26 was performed to correlate the values of log IC50, of c-GMP-stimulated phosphodiesterase, with the energy difference of syn Û anti equilibrium of cyclic nucleotides and hydrogen bonding descriptors associated to specific sites of the bases. Many studies on hydrogen bonding have established that electrostatic and charge transfer interactions are important factors in the formation of a hydrogen bond. However, hydrogen bonding with participation of atoms such as nitrogen and oxygen are strongly electrostatic in nature, with a small contribution from charge transfer interaction27. Thus, atomic charges on the atoms directly bonded to C6 (QX) and C2 (QY) of the purine were used as potential hydrogen bonding descriptors. From a set of three descriptors, DE, QX, Qy, only DE showed a relatively strong linear correlation with log IC50, R2= 0.78, another parameters have correlation coefficient significantly less than 0.78. Due to small number of cyclic nucleotide analogues, we have limited the multilinear analysis to 3 parameters.

The best multilinear correlation found is given by the following equation:

log IC50 = 0.502 (± 0.273) + 0.273 (± 0.076) DE + 1.178 (± 0.249) Qy +
    1.027 (± 0.309) QX

R2 =0.969 s2= 0.151 F= 42.11


Where R2, s2 and F have the usual significance26.

This model evidently needs to be justified with a larger number of cyclic nucleotides and thus have a more solid quantitative structure activity relationship for the inhibition of c-GMP stimulated phosphodiesterase from cyclic nucleotides. However, the picture found from multilinear regression analysis, performed on limited number of cyclic nucleotides, should be qualitatively correct because the experimental results have shown that the syn Û anti equilibrium and hydrogen bonding interaction are key factors for the inhibition of phosphodiesterases by cyclic nucleotides analogues4,28.


The theoretical study on the protomeric tautomerism of N9-methyl-8-hydroxyguanine showed that the oxo tautomer is the most stable species. This tautomeric species also have a hardness greater than that of the hydroxy tautomer verifying the principle of maximun hardness18.

The AM1 results have allowed to interpret the experimental results, for the competitive inhibition of c-GMP-stimulated phosphodiesterase, in terms of syn-anti equilibrium of these nucleotide analogues. Multilinear regression analysis indicate that syn-anti equilibrium is a dominant factor in the inhibition of c-GMP-stimulated phosphodiesterase, and that the enzyme prefers to bind the anti conformer of cyclic nucleotides. This result is in opposition to the postulated by other authors, which have inferred that c-GMP-stimulated phosphodiesterase prefers to bind the syn conformer of cyclic nucleotides4.


This work was supported by Dirección de Investigación de la Universidad de Concepción Grant 97.023.015-1.1D.


1. P.Greengard (Ed). Advances in Cyclic Nucleotides Research (1970-1985)        [ Links ]

2. J.A.Beavo. Adv. Second Messenger Phosphoprotein Res. 22, 1 (1988)        [ Links ]

3. J.K.Bentley and J.A.Beavo. Curr. Opin. Cell Biol. 4, 233 (1992)        [ Links ]

4. J.Beltman, D.E.Becker, E.Butt, G.S.Jensen, S.D.Rybalkin, B.Jastorff and J.A.Beavo. Mol. Pharmacol. 47, 330 (1995)        [ Links ]

5. J.S.Kwiatkowski, T.J.Zielinski and R.Rein. Adv. Quantum Chem. 18, 85 (1986)        [ Links ]

6. W.B.Person, K.Szczepaniak, M.Szczesniak, J.S.Kwiatkowski, L.Hernandez and R.Czerminski. J. Mol.Struct. 194, 239 (1989)        [ Links ]

7. M.T.Chenon, R.J.Pugmire, D.M.Grant, R.P.Panzica and L.B.Townsed. J. Am.Chem. Soc. 97, 4636 (1975)        [ Links ]

8. G.W.Buchanan and J.Bell. Can. J. Chem. 61, 2445 (1983)        [ Links ]

9. J.G.Contreras and J.B.Alderete. J. Mol. Struct.(Theochem) 283, 283 (1993)        [ Links ]

10. J.Leszcsynski. J. Mol. Struct.(Theochem) 311, 37 (1994)        [ Links ]

11. M.J.S.Dewar, E.G.Zoebisch, E.F.Healy and J.J.P.Stewart. J.Am.Chem.Soc. 107, 3902 (1985)        [ Links ]

12. Gaussian 94, Revision B.2, M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill,B. G. Johnson, M. A. Robb, J. R. Cheeseman, T. Keith,G. A. Petersson, J. A. Montgomery, K. Raghavachari, M. A. Al- Laham, V. G. Zakrzewski, J. V. Ortiz, J. B. Foresman, C. Y. Peng, P. Y. Ayala, W. Chen, M. W. Wong, J. L. Andres, E. S. Replogle, R. Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley, D. J. Defrees, J. Baker, J. P. Stewart, M. Head-Gordon, C. Gonzalez, and J. A. Pople, Gaussian, Inc., Pittsburgh PA, 1995.        [ Links ]

13. W.J.Here, L.Random, P.V.R.Schleyer and J.A.Pople. "Ab initio molecular orbital theory" John Wiley and Sons, N.Y. (1986)        [ Links ]

14. R.G. Parr and W.Yang. "Density funtional theory of atoms and molecules". Oxford University Press. N.Y. (1989)        [ Links ]

15. R.Mulliken. J.Chem.Phys. 23, 1833 (1955)        [ Links ]

16. R.Mulliken. J.Chem.Phys. 23, 2343 (1955)        [ Links ]

17. R.G.Parr and R.G.Pearson. J.Am.Chem.Soc. 105, 7512 (1983)        [ Links ]

18. R.G.Pearson J.Chem.Educ. 64, 561 (1987)        [ Links ]

19. R.G.Parr and P.K.Chattaraj J.Am.Chem.Soc 113, 1854 (1991)        [ Links ]

20. S.Pal, N.Vaval and R.K.Roy. J.Phys.Chem. 97, 4404 (1993)        [ Links ]

21. R.G.Pearson and W.E.palke. J.Phys.Chem. 90, 328 (1992)        [ Links ]

22. P.Perez and A.Toro-Labbe. J.Phys.Chem. 104, 1557 (2000)        [ Links ]

23. A.Bondi. J.Phys.Chem. 68, 441 (1964)        [ Links ]

24. F.Hibbert and J.Emsley. Adv. Phys. Org.Chem. 26, 255 (1990)        [ Links ]

25. S.Topiol, T.K. Morgan, Jr, M.Sabio and W.C. Luma, Jr. J. AM. Chem. Soc. 112, 1452 (1990)        [ Links ]

26. F.J.Wall. "Statistical data analysis handbooks" Mc Graw-Hill, Inc. (1986)        [ Links ]

27. H.Umeyama and K.Morakuma. J.Am.Chem.Soc 99, 1316 (1977)        [ Links ]

28. E.Butt, J.Beltman, D.E.Becker, G.S.Jensen, S.D.Rybalkin, B.Jastorff and J.A.Beavo. Mol. Pharmacol. 47, 340 (1995)        [ Links ]

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License