Halauko Yu.S., Ivashkevich O.А., Matulis Vadim E.

Belarusian State University

DFT Investigation and QSAR in Series of 5-Nitroimidazoles

Diseases caused by different pathogenic microorganisms are very widespread and bring a lot of trouble to mankind. Discovery of natural and development of synthetic antibiotics in the middle of XX century have led to the illusion about complete victory in the battle with infectious disorders. Temporary decrease of pharmaceutical companies` interest to these substances, development of bacterial resistance and appearance of new diseases have pointed out new challenges for researches [1]. Introduction of nitroheterocyclic drugs in the middle of XX century have commemorated a new era in therapy of diseases, caused by protozoa and some bacteria. 5-Nitroimidazole derivatives represent an important group of effective antimicrobials which are used as therapeutic agents against a variety of anaerobic bacterial and protozoal infectious disorders. These compounds are widely used in medical practice from 1960-s till nowadays [2]. Metronidazole was the first substance approved for clinical use, later a few effective analogs of metronidazole were developed and synthesized [3]. All these substances have nitro group in position “5” of imidazole ring. Changes in structure leading to changes in physico-chemical properties and also in biological activity are mainly connected with substituents modification in position “1” and “2” of imidazole ring. Nitro group causes a high one-electron reduction potential connected with mechanism of nitroimidazoles action: in hypoxic cells and anaerobic organisms the reduction of nitro group occurs, leading to active intermediates, which interact with important biomolecules (such as DNA), causing death of the cell [2].

QSAR models development is an important stage of modern drug design process. Analysis of model obtained allows to recognize clearly essential structural features, responsible for specific type of activity for further recommendations on lead compounds optimization.

Undoubtedly, that although considerable proportion of drugs exhibit their activity not as a result of direct chemical transformations, but as a result of intermolecular interaction with definite target, this interaction can be described in scientific terms and can be treated as system perturbation (early stage of chemical reaction). One of possible ways to achieve this goal is the analysis of substances reactivity within DFT framework. Hence, molecules biological activity can be characterized through such descriptors as orbital energies, electrostatic potentials, atomic charges, hardness and softness and others [4].

Eight different 5-nitroimidazoles were investigated. All these compounds were used as drugs (some are in use nowadays). Considered nitroimidazoles were tested against protozoa in test with clear quantitative endpoint [5] chosen for further modelling. We assume for convenience the activity of metronidazole in the set of compounds equal to 1, for other substances activities expressed by relative values (BA_r). All quantum chemical calculations were performed using GAUSSIAN 03W package. Geometric parameters were calculated by the gradient technique using 6-31G* basis set. To verify that stationary point is true minimum, force constants and vibrational frequencies were determined (the absence of imaginary values). 6-311+G** Basis set have been employed for subsequent energies and other descriptors calculations. Solvation was treated using PCM (Polarizable Continuum Model) with standard parameters for water. The methodology of descriptors calculation, including hydration free energy (E_h), ionization energy (IE), electron affinity (EA), electronegativity (c), total hardness (h), electrophilicity index (w) and others, was described in [6] in details. The names, biological activities and some of quantum chemical descriptors for investigated compounds are given in Table.

Table. Nitroimidazole derivatives under investigation: biological activity and descriptors

compound		descriptor
nonproprietary name	systematic name	BAr [5]	IE, eV	EA, eV	Eh, kJ/mol	m, Db	mi, Db	c, eV	h, eV	w, eV
metronidazole	1-(2-hydroxyethyl)-2-methyl-5-nitroimidazole	1	9.32	-0.95	-5.57	3.60	0.98	4.19	5.14	1.71
nimorazole	1-[2-(4-morpholyl)-ethyl]-5-nitroimidazole	2.22	8.60	-0.99	-4.01	3.65	1.26	3.81	4.80	1.51
ornidazole	1-(3-chloro-2-hydroxypropyl)-2-methyl-5-nitroimidazole	1.88	9.43	-1.13	-4.06	4.07	1.31	4.15	5.28	1.63
panidazole	1-[2-(4-pyridyl)-ethyl]-2-methyl-5-nitroimidazole	0.09	8.94	-1.06	-2.34	4.36	1.49	3.94	5.00	1.55
ronidazole	1-methyl-2-(carbamoyloxymethyl)-5-nitroimidazole	6.01	9.29	-0.84	-7.64	4.78	1.39	4.23	5.07	1.76
secnidazole	1-(2-hydroxypropyl)-2-methyl-5-nitroimidazole	0.98	9.25	-0.88	-3.49	4.92	1.56	4.19	5.07	1.73
tinidazole	1-(2-(ethylsulphonyl)-ethyl)-2-methyl-5-nitroimidazole	1.41	9.40	-1.12	-5.40	7.89	2.61	4.14	5.26	1.63
flunidazole	1-(2-hydroxyethyl)-2-(4-fluorophenyl)-5-nitroimidazole	4.98	8.64	-1.24	-7.05	3.29	1.01	3.70	4.94	1.39

QSAR model was developed using multiple linear regression, choosing best model by the criteria of R² maximization and minimization of mean standard error. Analysis demonstrates, that the best model for compounds under consideration includes four descriptors, namely, hydration energies, dipole moments (m), its` induction during solvation (m_i) and electrophilicity indices.

Act(rel) = -18,46 – 1,84E_h – 8,78m + 23,6m_i + 10,8w (1)

R² = 0,925 F = 9,22 MSE = 0,76 N = 8

Figure represents the relationship between experimental antiprotozoal activity and the prediction 91.

Figure. Experimental versus predicted activities of nitroimidazoles

The influence of chosen descriptors on biocide activity of nitroimidazoles needs for some clarification. Hydration free energy calculated by means of DFT can be used as a measure of substance partition between two phases. Hence, this descriptor represents membrane transport under conditions close to quasi-static and can serve as an alternative to widely used empirical octanol/water partition coefficient logarithm (ClogP). One can conclude from negative coefficient attached to this descriptor in model equation, that transport of nitroimidazoles to biological target improves with hydrophilicity or decreasing of hydration energy. In much the same manner next descriptor, dipole moment, can be referred to pharmacokinetic phase of drug action. Activity of compounds decreases with this parameter increase, what can be related to difficulties in biological barriers penetration. Increase in activity with electrophilicity index increase can be easily explained, considering electron transfer process takes place in pharmacodynamic phase of nitroimidazoles action. The model developed was used to estimate the activity of a substance with a close activity spectrum – nitazolum (2-acetylamino-5-nitrothiazole). According to [7] the ratio of nitazolum to metronidazole activity equals 12, whereas predicted by our model value is 18, which can be judged as good estimation, regarding change of heterocycle and substituents nature.

A physically interpretable QSAR model, based on quantum chemical descriptors, for nitroimidazoles antiprotozoal activity was obtained. We introduced DFT-calculated hydration free energy as a descriptor, that can be a measure of pharmacokinetic properties of compounds and serve as alternative to applying empirical models. It`s potentialities and applicability limits are in need of subsequent specification. The model obtained can be useful for development of new drugs carrying nitroazole moiety – potentially with antibacterial or antiprotozoal action.

References

1. Ni, H.; Wendoloski, J. Ann. Rep. Comp. Chem. 2006, 2, 279.

2. Raether, W.; Haenel, H. Parasitol. Res. 2003, 90, S19.

3. Padeyskaya, Е. N. Infect. Antimicrob. Ther. 2000, 2, 110.

4. Parr, R. G.; Yang, W. Density Functional Theory of Atoms and Molecules; Oxford University Press: New York, 1989.

5. Boreham, P. F. L.; Phillips, R. E.; Shepherd, R. W. J. Antimicrob. Chemother. 1985, 16, 589.

6. Halauko Yu. S., Ivashkevich O. A., Matulis Vadim E. Doklady of the National Academy of Sciences of Belarus. 2008, 52, 79.

7. Blatun, L. A.; Lyapunov, N. A.; Kalinichenko, N. F.; Osolodchenko, T. P.; Svetukhin, A. M.; Pavlova, M. V.; Grishina, I. A.; Goncharova, Z. G. Antibiot. Khimioter. 1998, 43, 20.