REGULAR PAPERS

QSAR studies on urea and thiourea derivatives. Relationship between descriptors log P, p, MR and MV and antibacterial acvtivity in Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli.

Figueroa V.L.¹; Díaz C.F.²; Ceballos R.G.³; López R.M.¹; Maldonado V. G.¹, Camacho L. A.⁴

¹Universidad Autónoma de Campeche, Av. Agustín Melgar, Col Buenavista C.P.24030 Campeche Cam., México.
²Escuela de Ciencias Biológicas del IPN. Plan de San Luís y Díaz Mirón s/n Col. Santo Tomas, México, D.F. C.P. 11340.
³Escuela de Medicina del IPN. Plan de San Luís y Díaz Mirón s/n Col. Santo Tomas, México, D.F. C.P. 11340.
⁴Universidad Autónoma del Estado de Hidalgo, Instituto de Ciencias de la Salud, Area de Farmacia. Ex-Hacienda la Concepcion, Tilcuautla, Hgo, México, C.P. 42160.
E-mail: lauro_1999@yahoo.com; Phone: (981) 8119800 Ext. 73006;
Fax (981) 8119800 Ext. 73002.

Received May 30, 2008.
In final form February 2, 2009.

Abstract
In this work the structural requirements of both urea and thiourea derivatives were evaluated for optimal antibacterial activity on Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli, using several organic compounds (VI-X) as chemical tools. In order to delineate the structural chemical requirements of studied compounds, the descriptors Log P, p MR and MV were calculated. The results showed that bacterial growth of the studied microorganisms, in presence of IV and VI compounds was inhibited at low concentrations in comparison with I-III, V and VII-X. Other results indicate that Log P increase in II-V, VIII compounds and decrease in the VI-X substances in comparison with I thiourea-derivative. In addition, other data indicate that the p values are lower in the II-X substances in comparison with I compound. Other data showed an increase in both MR and MV values in the I-V and VIII compounds and a decrease in the VI, VII, IX and X substances. In conclusion, the results found indicate that substituents involved in the chemical structure of both urea and thiourea derivatives increase the lipophilicity and changes in the functional groups decrease the Log P. Therefore, this phenomenon can affect the antibacterial activity on Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli. In addition the results indicate that steric impediment, the molecular mechanism, conformational preferences and internal rotation of different compounds could affect the antibacterial effect of studied compounds.

Keywords: Thiourea; Antibacterial activity; Descriptors

Resumen
En este trabajo, los requerimientos estructurales de los derivados de urea y tiourea fueron evaluados para su actividad antibacterial sobre Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli, usando varios compuestos orgánicos (VI-X) como herramientas químicas. Para delinear los requerimientos estructurales químicos de los compuestos estudiados, fueron calculados los descriptores Log P, p, MR y MV. Los resultados mostraron que el crecimiento bacterial de los microorganismos estudiados, en presencia de los compuestos IV y VI fue inhibido a bajas concentraciones en comparación con I-III, V y VII-X. Otros resultados indican que el Log P aumenta en los compuestos II-V, VIII y disminuye en las sustancias VI-X en comparación con el derivado de tiourea I. Además otros datos indican que los valores de p son menores en las sustancias II-X en comparación con el compuesto I. Otros datos, mostraron un incremento en los valores de RM y VM en los compuestos I-V, VIII y un decremento en las sustancias VI, VII, IX y X. En conclusión, los resultados encontrados indican que los sustituyentes involucrados en estructura química de los derivados de urea y tiourea incrementan la lipofilicidad y los cambios en los grupos funcionales disminuyen el log P. Por lo tanto, este fenómeno puede afectar la actividad antibacterial sobre Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli. Además, los resultados indican que el impedimento estérico, el mecanismo molecular, las preferencias conformacionales y la rotación interna de los diferentes compuestos podrían afectar su efecto antibacterial.

Palabras clave: Tiourea; Actividad antibacterial; Descriptores.

Introduction
Antibiotic resistance can be considered as a serious threat for health, and it requires an international approach to its management, in this sense, new drugs have been developed for the control of bacterial resistance [1-3]. The synthesis and antibacterial activity of urea and several of its derivates has been the subject of numerous investigations [4-6] for example; Hackbarth and coworkers [7] demonstrated the antibacterial activity of urea derivates (N-Alkyl Urea Hydroxamic Acids) on Gram positive and Gram negative bacteria. Additionally, several thiourea derivatives were synthesized by Trani and coworkers [8], whom showed that these compounds had antibacterial activity on Staphylococcus aureus and Escherichia coli. Other studies showed that bacterial growth of Escherichia coli was inhibited in presence of nitrourea, thiourea y S-methyliso-thiourea [9]. In addition, studies made by Iwai and coworkers [10] showed that S-benzylisothiourea derivatives induced formation of spherical cells in both Escherichia Coli and Staphylococcus aureus inducing bacterial death. Other results found by Cunha and coworkers showed antibacterial activity of thiourea derivatives and suggest an influence of electronic nature of N²-group of N¹1-benzoyl-N²-substituted thioureas [11], nevertheless these data are not clear. In addition, the results reported by Kumar and coworkers [12] indicate that electron withdrawing groups in the ortho position of the phenyl ring contained in the chemical structure of both urea and thiourea derivatives, enhances the antibacterial activity on Staphylococcus aureus.
On the other hand, studies made by Struga and coworkers [13] showed that a series of nineteen urea and thiourea derivatives of 4-Azatricyclo[5.2.2.0^2,6]undec-8-ene-3,5-dione were completely inactive on Staphylococcus aureus (ATTC 25923, 6538P), and Escherichia Coli (ATTC 25922, 10538). Additionally, several experimental reports exist to determine the relationship between the lipophilicity and antibacterial activity of thiourea derivatives, for example, the works of Tokuyama and coworkers [14] showed a relation between the hydrophobic parameters (p and Log P) and antibacterial activity. All those data are controversial; therefore in this work our aim is to analyze the electronic nature of urea and thiourea derivatives in order to evaluate their antibacterial activity. In this sense, we made a quantitative structure activity relationship parameters (QSAR) study on urea and thiourea derivatives and the analyzed compounds were used to assess their antibacterial effect on Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli using the microbial minimal inhibitory (MIC) method described by Chiong and coworkers [15], in order to have new drugs that can be used for the treatment of infections diseases.

Experimental
QSAR. To estimate the logarithmic octanol-water partition coefficient (log P) of organic compounds the logKow method (atom/fragment contribution), introduced by Mannhold and Howard (method A) [16], available as the KOWWIN and KLogP (method B) [17] software's and the fragmental technique ACDLogP (method C) [18] were used.

Antibacterial evaluation
Strains. The microorganisms in this study belonged to the strain bank at the Departament of Pharmaco-Chemistry at the Faculty of Chemical Biological Sciences of the Universidad Autonóma de Campeche. The strains are certified by Center for Disease Control in Atlanta and were as follows. Staphylococcus aureus (ATCC 25923), Klebsiella pneumoniae (ATCC 700603) and Escherichia coli (ATCC 25922). The strains are kept under refrigeration at 4°C in special gel (BBL).
Antimicrobial agents. The urea and thiourea derivatives were obtained from Sigma-Aldrich Co. The compounds were dissolved in methanol and diluted with distilled water. Cefotaxime, gentamicin and ampicillin were used as the standar drug.
Antimicrobial activity. The evaluation of antimicrobial effect of the different compounds on the bacterial species was made by the method described by Chiong et al [15]. The bacterial species were incubated on Mc-Conkey (Escherichia coli and Klebsiella pneumoniae) and Staphylococcus 110 (Staphylococcus aureus) agars for 24 hours at 37°C, after such time, it could be determined whether growth had taken place or not.
On the other hand, a series of tubes were prepared, where the first of which contained 2 ml of culture medium (tripticase soya) at double concentration and the remainder (11 tubes), contained the same quantity of medium at simple concentrations. From the first tube (double concentration) an aliquot of 2 ml was added of the studied compound (1 mg/ml) and stirred, from this tube an aliquot of 2 ml was taken and added to the following tube (simple concentration) and the process was successively repeated until the last 2 ml of dissolution had been used up. After this process, each tube was inoculated with 0.1 ml of the bacterial suspension whose concentration corresponded to McFarland scale (9 x 10⁸ cells/ml) and all the tubes were incubated at 37°C for 24 hours. Subsequently, a loop was taken from each of them and inoculated into the appropriate cultures for different bacterial organisms, and were incubated for 24 hours at 37°C. After such time, the minimum inhibitory concentration (MIC) was evaluated to consider the antimicrobial effect of the both urea and thiourea-derivatives. In order to discard the effect of methanol on the bacterial species studied, a series of the same number of tubes was prepared in parallel, to which 2 ml of methanol at 60% was added to the first and corresponding successive dilutions were added in the same way as before. In addition a control series was also performed using distilled water to pH 7.0.
Statistical analysis
Statistical analysis was performed by Pearson's correlation coefficient. The differences were considered significant when p was equal or smaller than 0.05.

Results and Discussion
In this work the structural requirements of both, urea and thiourea derivatives were evaluated for antibacterial activity on Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli, using several organic compounds VI-X (Figures 1 and 2) as chemical tools. In addition, the bacterial activity of all compounds was compared with the antibacterial effect induced by cefotaxime, gentamicin and methicillin (controls) in such bacterial microorganism. The results showed (Table 1) that bacterial growth of Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli was inhibited with cefotaxime and gentamicin but not with ampicillin. Other results indicated that bacterial growth, in presence of I-III thiourea-derivatives (same dose) was inhibited. Here is important to mention that II and III substances has as chemical characteristic methyl groups as substituents in both rings of thiourea-derivatives in position ortho (II) and meta (III) to ring nitrogen. The results showed that these thiourea-derivatives do not modify the antibacterial activity of I substance, these experimental data suggest that both o-methyl (II) and m-methyl (III) substituents to ring nitrogen could be not essential to antibacterial-induced effect.

Figure 1. Chemical Structure of thioureas-derivatives.

Figure 2. Chemical Structure of acid-derivative (VII), urea-derivative (VIII), and phenylacetamide-derivatives (IX-X).

Table 1. Log of Minimum inhibitory concentration experimental (Log 1/MICEXP) of thiourea-derivatives.

In order to discard that orientation of substituent can affect the antibacterial activity it was used the thiourea-derivative IV, with a p-methyl substituent to B ring-nitrogen. The results indicate that this compound induced greater antibacterial effect with smaller dose in comparison with I-III, these experimental data suggest that methyl electron donating group (IV) can affect the antibacterial activity on the microorganism studied. This premise is supported by the studies of Warner and coworkers [19] whom showed that p-methyl to ring nitrogen of biguadines derivatives induce antibacterial activity on streptococcus mutans No. 6715.
On the other hand, it has been reported that electron withdrawing groups are essential for drugs with antibacterial activity therefore, in this work we made experimental approaches with the purpose of evaluate the role of electron withdrawing groups involved in the chemical structure of both urea and thiourea-derivatives, on their antibacterial effect on Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli. In this sense, the N,N'-bis(3-chlorophenyl)thiourea compound (V) was evaluated. The results indicate that bacterial growth of studied microorganisms was inhibited in its presence, at similar dose than the II and III thiourea derivatives; these data suggest that m-chloride substituent involved in the V compound do not modify the antibacterial effect of II and III compounds. This phenomenon could be because the m-chloride is an electron withdrawing group inductively but possibly it can act as an electron donating group through resonance on phenyl-ring and by this does not affect the antibacterial activity of thiourea-derivative.
On the other hand, we analyzed the possibility that phenyl groups could be important for the antibacterial effect using the VI and VII compounds. The results showed that these substances had the same antibacterial effect that I-III and V thiourea-derivatives on Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli. These experimental data suggest that functional phenyl groups possibly are not required for antibacterial effect, and indicate that antibacterial activity could be related with the functional thiourea group. This premise is supported by studies reported by Westland and coworkers [20] whom indicate that bacterial growth of Staphylococcus aureus was inhibited in presence of alky thiourea-derivatives. In addition, the studies reported by Bandelin & Tuschhoff showed that isothioureas are germicides on Gram-positive and Gram negative bacteria [21].
On the other hand, thinking about of role that has the sulfur atom in the antibacterial activity induced by thiourea-derivatives on bacterial microorganisms studied and analyzing the works done by Klotz & Mellody [22] whom showed that thiourea substances have greater antibacterial properties at low concentrations on Escherichia coli in comparison with the urea compound. In this sense, alternative experiments were made to evaluate this premise using as tool the VIII compound. The results showed that this substance increased the antibacterial activity on Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli in comparison with I, II, III, and V thiourea-derivatives. Nevertheless, the VIII urea-derivative showed the same antibacterial activity that IV compound. This experimental data suggest; 1) the substitution of oxygen by sulfur atom; and 2) p-chloride substituents can be specific for antibacterial effect, therefore, this data suggest that functional groups involved in the chemical structure of urea-derivatives can affect their antibacterial activity. To further evaluation the IX compound was evaluated, the results showed that bacterial growth of Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli was inhibited with high doses of IX substance in comparison with the I-VIII compounds. This data suggest that change in the position of amine group can affect the antibacterial activity of urea-derivative. Alternative experiments were made using the X substance, here is important to mention that this compound has in its structure a hydrazine group. The results showed antibacterial activity with greater dose than with the VIII urea-derivative, these data suggest that changes in the position of amino groups can affect the antibacterial activity of the compounds studied. However, the addition of a second amino group to form hydrazine group showed similar antibacterial effect in comparison with VIII. All data suggest that structural chemistry of compounds studied is specifically to their antibacterial activity.
To delineate the structural chemical requirements of urea, thiourea derivatives and phenylacetamide compounds (Figures 1 and 2) as inhibitors of Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli growth; we calculate other parameters such as, the descriptors Log P, p [23]. Log P estimates the logarithmic octanol-water partition coefficient, therefore the Log P represents the lipophilic effects of a molecule which includes the sum of the lipophilic contributions of the parent molecule and its substituent [24]. The difference between the substituted and unsubstituted Log P values is condicionated by the p value for the particular substituent. Hammett showed that p values measure the free energy change caused by particular substituent to relate to biological activity [25].
The Log P and p parameters were calculated by three different methods [16-18]. The results (Table 2) showed an increase in Log P and decrease of pvalues on the II-V and VIII compounds with respect to (I) thiourea derivative. This phenomenon is conditioned mainly, by the contribution of all substituent atoms involved in the chemical structure of the different compounds, as is showed in the Tables 3. The results showed that m-CH₃ aliphatic carbon substituents in the II-IV compounds contribute to high lipophilicity in comparison with I compound. The change of the m-CH₃ by a m-chloride substituent in A and p-chloride substituent B ring-nitrogen increases the lipophilicity in the V thiourea-derivative. This result is supported by the QSAR studies made by Tokuyama and coworkers [14] whom showed that halogen substituent involved in chemical structure of 5-thiourea oxazolidinones induced changes in both the lipofilicity and antibacterial activity in comparison with methyl-thiourea derivatives.

Table 2. Physicochemical parameters of thiourea-derivatives. w = mean; a = Method A; b = Method B; c = Method C.

Table 3. Log P of thiourea-derivates. Constant Equation = 0.2290 (I-X); Amino acid (alpha-position) correction = - 2.0238 (VII); Aromatic-CH(-CO-N)- {-N<,-OH,CO} correct = 0.7500 (IX-X).

The calculated data for VI and VII substances showed less solubility in comparison with the I-V compounds, this phenomenon is due to the loss of the aromatic carbons (3.528) contained in the chemical structural of thiourea-derivatives (see Table 3) that contribute to lipophilicity. It is important to mention that the three methods used to calculate the Log P differ, and although they give reasonable predictions on VI, VII and IX thiourea derivatives, the results showed opposite directions. These results are similar to reports showed by Leo and coworkers et al [23] whom found differences in the results calculated to benzotriazine-derivates using both ACDLog P and KOWWIN programs to calculate log P(oct).
On the other hand, to prove the existence of a correlation between the calculated log P and antibacterial activities of all studied compounds, the MIC was calculated using the method proposed by Hansch and compared it with experimental MIC values [26]. The results are showed in the Table 4, in addition the MIC observed and MIC calculated were evaluate using the data obtained in this study. The Statistical analysis showed a correlation between the MIC observed and MIC calculated of r = 0.588 (Method A), r = 0.637 (p = 0.005) (Method B) and a relationship of r = 0.555 with Method C on Staphylococcus aureus (Figure 3). The results on Klebsiella pneumoniae (Figure 4), show a relationship between the MIC observed and the MIC calculated of r = 0.398 (Method A), r = 0.409 (Method B) and a relationship of r = 0.453 with Method C. Experimental alternative shows a relationship between the MIC observed and the MIC calculated on Escherichia coli (Figure 5), r = 0.580 (Method A), r = 0.189 (Method B) and a relationship of r = 0.583 with Method C.

Figure 3. Correlation with between the MIC observed and MIC calculated in Staphylococcus aureus with different methods.
The results showed an correlation between the MIC observed and MIC calculated of r = 0.588 (Method A), r = 0.637 (p = 0.005) (Method B) and a relationship with Method C of r = 0.555. Method A = KOWWIN; Method B = KLogP; Method C = ACDLogP.

 

Figure 4. Relationship with between the MIC observated and MIC calculated on Klebsiella pneumoniae.
The results found an correlation between the MIC observed and MIC calculated of r = 0.398 (Method A), r = 0.409 (Method B) and a relationship of r = 0.453 with Method C. Method A = KOWWIN; Method B = KLogP; Method C = ACDLogP

Figure 5. Correlation between the MIC observed and MICs calculated on Escherichia coli. The results found an correlation between the MIC observed and MIC calculated of r = 0.580 (Method A), r = 0.189 (Method B) and a relationship of r = 0.583 with Method C. Method A = KOWWIN; Method B = KLogP; Method C = ACDLogP

Tabla 4. Minimum inhibitory concentration calculated (MIC_CALC) of thiourea derivatives on Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli.

We found some variability of experimental data, which can be possibly due to other chemical parameters involved in the antibacterial activity of the compounds studies. We calculate some steric constants (MV and MR, i.e., the molar volume and molar refractivity [27], these options are useful tool for correlation of different properties which depend on characteristics of substituents attached to a constant reaction center. The results are showed in table 5, there are increases in both, MR and MV values in the I-V and VIII compounds and a decrease in the VI-VII and IX-X substances, these data indicate that steric impediment could affect the antibacterial activity of the studied compounds. Different molecular mechanisms and conformational preferences and internal rotation of the different compounds, can influence their antibacterial activity on Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli. These data are supported by the studies reported by Bryantsev and coworkers [28] whom showed that the conformational differences between urea and thiourea groups have some important consequences in the union to biological receptors by conformational changes.

Table 5. Physicochemical parameters of thiourea-derivatives. MR = molar refractivity; MV = molar volume; b = Method B.

Conclusions
In conclusion, the results found indicate that substituents involved in the chemical structure of both urea and thiourea derivatives increase the lipophilicity and changes in the functional groups decrease the LogP. Therefore, this phenomenon can affect the antibacterial activity on Staphylococcus aureus, Klebsiella pneumoniae and Escherichia coli. In addition the results indicate that steric impediment, the molecular mechanism, conformational preferences and internal rotation of different compounds could affect the antibacterial effect of studied compounds.

Acknowledgements.
Lauro Figueroa Valverde is grateful to Angelica Leon Garcia for their helpful discussions .

References
1 Tenover, F.; Mohammed, J.; Stelling, J.; O'Brien, T.; Williams, R., J. Clin. Microbiol., 2001, 39, 241.         [ Links ]

2 Oteo, J.; Lázaro, E.; De Abajo, J.; Baquero, F.; Campos, J., J. Clin. Microbiol., 2004, 42, 5571.         [ Links ]

3 Gordon, E.; Barrett, R.; Dower, J., J. Med. Chem,. 1994, 37, 1385.         [ Links ]

4 Zhong, Z.; Xing, R.; Liu, S.; Wang, L.; Cai, S.; Li, P., Carbohydrate Research, 2008, 25, 566.         [ Links ]

5 Chen, S.; Wu, G.; Zeng, H., Carbohydrate Polymers, 2005, 60, 33.         [ Links ]

6 Kogen, H.; Tago, K.; Arai, M.; Minami, E.; Masuda, K.; Akiyama, T., Bioorg. Med. Chem. Lett., 1999, 9, 1347.         [ Links ]

7 Hackbarth, C.; Chen, D.; Lewis, J.; Clark, K.; Mangold, J.; Cramer, J.; Margolis, P.; Wang, W.; Koehn, J.; Wu, C.; Lopez, S.; Withers, G., Antim. Agents. & Chem., 2002, 46, 2752.         [ Links ]

8 Trani, A.; Ferrari, P.; Pallanza, R.; Ciabatti, R. J antibiotics, 1989, 42, 1268.         [ Links ]

9 Klotz, M.; Mellody, M., J. Bacteriol. 1949, 57, 477.         [ Links ]

10 Iwai, N.; Ebata, T.; Nagura, H.; Kitazime, T.; Nagai, K.; Wachi, M., Biosci. Biotechnol. Biochim., 2004, 68, 2365.         [ Links ]

11 Cunha, S.; Macedo, F.; Costa, G.; Rodriguez, M.; Verde, R.; De Souza, L; Lariucci, C., Monatsh Chem., 2007, 138, 511.         [ Links ]

12 Kumar. S.; Kanthi, M.; Prabhu, G.; Ganesh, S.; Vijayan, M., Chem Pharm Bull,, 2007, 55, 236.         [ Links ]

13 Struga, M.; Kossakowski, J.; Kedzierska, E.; Fidecka, S.; Stefanska, J., Chem Pharm Bull,, 2007, 55, 796.         [ Links ]

14 Tokuyama, R.; Takahashi, Y.; Tomita, Y.; Tsubouchi, M., Yoshida, T., Chem Pharm Bull., 2001, 49, 353.         [ Links ]

15 Chiong, R.; Betancourt, A. Inst Nal Higiene, Epidemiol Microbiol, La Habana Cuba, 1985, pp. 24-30.         [ Links ]

16 Mannhold, R.; Waterbeemd, H., J Comput-Aided Mol Design, 2001, 15, 337.         [ Links ]

17 Azencott, A. C.; Ksikes, A.; Swamidass, J. S.; Chen, H. J.; Ralaivola, L.; Baldi, P., J Che.m In.f Model., 2007, 47, 965.         [ Links ]

18 Thompson, J. S.; Hattotuwagama, K. C.; Holliday, D. J.; Flower, R. D., Bioinformation, 2006, 1, 237.         [ Links ]

19 Warner, V.; Lynch, D., J. Med. Chem., 1979, 22:359.         [ Links ]

20 Westland, D. R.; Kauger, R.; Green, B., Dice, R. J., J. Med. Chem., 1968, 11, 84.         [ Links ]

21 Bandelin, J. F.; Tuschhoff, V. J., J Am Chem Soc, 1952, 74, 4271.         [ Links ]

22 Klotz, M. I.; Mellody, M., J Bacteria,l, 1949, 57, 477.         [ Links ]

23 Leo, A.; Jow, P.Y.; Silipo, C., J, Med, Chem., 1975, 18, 865.         [ Links ]

24 Leo, A.; Hoekman, D., Persp. Drug. Discov. Design, 2000, 18, 19.         [ Links ]

25 Hansch, C.; Leo, A.; Taft, R. W., Chem. Rev., 1991, 91, 165.         [ Links ]

26 Hansch, C.A., Acct. Chem. Res., 1969, 2, 232.         [ Links ]

27 Yoshimoto, M.; Hansch, C., J. Med. Chem., 1976, 19, 71.         [ Links ]

28 Bryantsev, S. V.; Hay, P. B., J. Phys. Chem., 2006, 110, 4678.         [ Links ]