Home Archive Vol 36, No.2, 2010 For practitioner Oxicams Structural Characteristics Determined by Molecular Modeling Methods

Oxicams Structural Characteristics Determined by Molecular Modeling Methods

Denisa-Constantina Amzoiu(1), Emilia Amzoiu(2), Florica Popescu(3)

 (1)Department of Pharmaceutical Chemistry;(2)Department of Physical Chemistry;(3)Department of Pharmacology,Faculty of Pharmacy, University of Medicine and Pharmacy of Craiova


ABSTRACT Classical nonsteroidal anti-inflammatory drugs (acetylsalicylic acid, ibuprofen, indomethacin) produce side effects like gastrointestinal ulcerations or renal failure. Current studies are trying to find drugs exerting analgesic, antipyretic and anti-inflammatory activity without these side effects. Compounds from oxicam class are included in this category. Physico-chemical properties of these drugs were evaluated with molecular modeling models in order to achieve a characterization of these compounds. This study aims to analyze electronic, atomic or molecular structural factors contributing to the biological action manifested by the compounds of oxicam class. The study showed the contribution of some parameters characteristic of the substances chemical structures on the partition coefficient. To evaluate these molecular descriptors is to show their importance in the partition of the studied substances in both aqueous and lipid phase (1 octanol).

KEY WORDS oxicams, descriptors, partition coeffi-cient


Introduction

The nonsteroidal anti-inflammatory drugs (NSAIDs) are derivatives of carboxylic acids (N phenylanthranylic acid (mefenamic and flufenamic acid), phenylacetic acid (diclofenac), heteroarylacetic acid (indomethacin), propionic acid (ibuprofen, ketoprofen)) and enolic acids (derivatives from the oxicam class) exerting analgesic, antipyretic and anti-inflammatory activity [1–3]. From the NSAIDs category, classic aspirin, ibuprofen and indomethacin are still the most prescribed drugs to treat diseases such as osteoarthritis, rheumatoid arthritis, postoperative pain, orthopedic injuries, severe myalgia [4–6].

These drugs exert anti-inflammatory action by inhibiting cyclooxygenase (COX) enzyme that catalyzes the conversion of arachidonic acid to prostaglandin (PG) H2 and thromboxane and therefore of other prostaglandins that are potential mediators of inflammation. Since conventional NSAIDs (acetylsalicylic acid, ibuprofen, naproxen) inhibit both isoforms of COX (COX-1 and COX 2) producing effects such as gastrointestinal ulceration or renal failure [7, 8], numerous studies have tried to find selective inhibitors for COX-2 [9, 10].

The purpose of this paper is to analyze the structure of compounds of oxicam class, which some are selective COX-2 inhibitors.

Method

Prediction of physicochemical properties of oxicams class

Chemical substances studied are 14 com-pounds of the oxicam class with analgesic, anti-pyretic and anti-inflammatory activity (Figure 1, and Tables 1a and 1b).

Physicochemical properties and molecular des-criptors for the chemical structures of oxicams were obtained covering the following steps:

▪ molecular modeling of chemical structures for the 14 studied substances;

▪ quanto-molecular calculations of molecular geometries;

▪ estimation of physicochemical properties and molecular descriptors.

The first stage of chemical structures modeling was performed using the program Hyperchem 6.0 (trial version) using the MM+ method (Molecular Mechanics).

Quanto-molecular calculations were performed using MOPAC 6.0 (Molecular Orbital Package) software package [13], which includes a series of approximations of molecular orbital calculations of RHF type (Restricted Hartrees Fock).

Figure 1 – General chemical structure of the oxicam class.

 

 

Table 1a – The chemical structures of the studied compounds from the oxicam class (1–11)

Substance

R1

R2

1.

H

2.

H

3.

H

4.

H

5.

–CH(CH3)O–COOCH2CH3

6.

–CO–CH=CH–C6H5

7.

H

–(CH2)3N(CH3)2

8.

H

9.

–CH(CH3)2

10.

–CH3

11.

–COC6H5

1. Piroxicam, 2. Meloxicam, 3. Sudoxicam, 4. Isoxicam, 5. Ampiroxicam, 6. Cinnoxicam, 7. Propoxicam, 8. N-6M2P-Piroxicam (N-(6-methyl-2-pyridyl)-4-hydroxy-2-methyl-2H-1,2-benzothiazine-3-carboxamide 1,1-dioxide), 9. 4-IPO-Piroxicam (4-Isopropyloxy-2-methyl-N-2-pyridinyl-2H-1,2-benzothiazine-3-carboxamide 1,1-dioxide), 10. 4-MO-Piroxicam (4-Methoxy-2-methyl-N-(2-pyridyl)-2H-1,2-benzothiazine-3-carboxamide 1,1-dioxide, 11. 4-BO-Piroxicam (4-Benzoyloxy-2-methyl-N-(2-pyridyl)-2H-1,2- benzothiazine-3-carboxamide 1,1-dioxide) [12].

Table 1b – The chemical structures of the studied compounds from the oxicam class (12–14)

1.     Substance

1.     Structure

2.     12.

2.     

3.     13.

3.     

4.     14.

4.     

12. Tenoxicam, 13. Droxicam, 14. Lornoxicam [12].

PM3 (Parameter Model 3) procedure was used regarding the semiempirical parameters used in calculations of quantum molecular structures. Thus were obtained data on the electronic structure of studied substances, data that helped us  to calculate the physicochemical properties and molecular descriptors using ChemOffice 7.0 soft-ware package (evaluation copy).

Connolly sizes [14–16] namely CSAA (Connolly Accessible Surface Area) and CSEV (Connolly Solvent Excluded Volume), and ovality index [17] were evaluated using ChemOffice 7.0 software package (evaluation copy), using as cartesian coordinates of input data on the position of atoms in space, coordinates found for the molecular geometries optimized by semiempirical quanto-molecular calculations MOPAC 6.0 (PM3) [13].

QSPR study of the partition coefficient water/octanol

In order to obtain more information on interactions with biological membranes of compounds from oxicam class with analgesic, antipyretic and anti-inflammatory activity, the partition coefficient was correlated with different structural data (descriptors). Partition coefficient describes substances hydrophobicity or, rather, lipophilicity, which play an important role in their adsorption and absorption by biological membranes.

Molecular geometries obtained for the ana-lyzed compounds were used for advanced study as input data in the software package MOPAC 7.0 [13]. The output data contain molecular electronic levels, electronic population, net atom charges, bond orders and free valences, dipole moments, moments of inertia, polarizability and different energy partition established by type of interactions and chemical bonds. Correlation was performed using CODESSA 1.0 Program (Comprehensive Descriptors for Structural and Statistical AnalysisUniversity of Florida) [18]. This program calculates from the MOPAC output data different structural descriptors (e.g. topo-logical, electrostatic, geometrical and quanto-molecular descriptors), making their selection through the cross-validation process and then correlates them through multi-linear regression with the desired property.

Results and Discussion

In Table 2 are presented the values for physico-chemical properties of the studied compounds.

Structural descriptors, which characterize the shape of molecular form of chemicals, calculated for these compounds and their values are listed in Table 3.

In QSPR-type correlation of partition coeffi-cient with various structural descriptors, from the 335 descriptors that CODESSA 1.0 program can calculate using the output data MOPAC 7.0, relevant in our case seem to be the quanto-molecular descriptors shown in Table 4 (R2 – correlation coefficient).

As can be seen, molecular form descriptors are important in the partition between aqueous phase and lipid phase (biological membrane) characterized by partition coefficients log P, i.e. such a correlation indicates the potential role of descriptors describing the chemical structure of substances.

This becomes more evident in the case where less descriptors are taking into consideration. Thus, the correlation of partition coefficient with 1 descriptor indicates that this coefficient values depend of polarizability and reactivity index, which ultimately depend on the form of each molecule in space and are directly implicated in the ligand–receptor interaction.


Table 2 – Values for physicochemical properties of the studied compounds

Substance

log P

I

α × 1024 [cm3]

Pc [bar]

Tc [K]

Vc [cm3/mol]

∆Hf,298 [kcal/mol]

1.

0.29

8.56

30.32

33.335

1005.42

761.5

-185.76

2.

0.35

8.66

31.68

32.728

1032.19

774.5

-58.456

3.

0.27

8.72

29.84

37.317

1027.13

718.5

-50.782

4.

-0.77

9.04

29.32

32.616

1007.6

749.5

-100.82

5.

1.81

8.66

40.86

18.499

1044.93

1078.5

-172.98

6.

2.76

8.87

45.38

18.904

1107.6

1144.5

-43.695

7.

-0.65

8.84

31.90

23.428

964.75

838.5

-92.627

8.

0.78

8.56

32.16

29.441

1012.48

817.5

-52.072

9.

1.31

8.69

35.83

22.229

995.33

922.5

-55.676

10.

0.65

8.76

32.16

26.653

988.90

816.5

-44.548

11.

2.42

8.71

41.90

21.433

1082.32

1052.5

-61.845

12.

0.27

8.69

29.84

38.723

1018.75

717.5

-36.05

13.

1.1

9.19

31.47

28.323

1015.11

795.5

-53.525

14.

0.64

8.71

31.77

36.333

1035.38

766.5

-42.555

log P – partition coefficient; I – ionization potential; α – polarizability; Pc, Tc, Vc – critical pressure, critical temperature and critical volume; ∆Hf,298 – heat of formation.

Table 3 – Values of structural descriptors

Substance

CSAA [Å2]

CSEV [Å3]

OV

RM [cm3/mol]

μ [D]

Et [kcal/mol]

1.

477.627

235.396

1.4034

85.04

5.857

-88055.6

2.

504.143

248.064

1.4286

90.91

5.259

-89634.6

3.

468.985

232.254

1.3877

85.09

4.684

-86183.8

4.

489.076

233.496

1.4329

85.80

4.048

-92053.9

5.

642.430

343.564

1.5350

111.21

3.957

-124880

6.

589.542

363.715

1.4001

124.61

6.090

-121500.6

7.

552.443

269.530

1.4892

89.99

4.830

-90954.5

8.

494.828

249.573

1.4145

90.08

5.952

-91388.67

9.

528.910

286.950

1.4160

98.96

2.516

-98175.5

10.

486.050

253.813

1.3763

89.79

3.624

-91273.05

11.

602.177

319.212

1.5024

114.37

3.368

-115159.5

12.

459.013

227.915

1.3689

83.91

5.746

-86119.4

13.

510.471

239.158

1.4739

87.92

3.580

-96847.6

14.

484.901

241.179

1.4013

88.68

5.930

-93062.5

CSAA – Connolly surface accessible area; CSEV – Connolly solvent-excluded volume; OV – ovality index; RM – molar refractivity; μ – dipole moment; Et – total energy.

 


Table 4 – Log P correlation: log P = a0 + , where Xi = descriptors

1.     Three descriptors:

2.     best correlations (F):

3.     1: R2=0.9694  F=105.6348 (three descriptors) 331 79 141

4.     2: R2=0.9616  F= 83.3638 (three descriptors) 331 79 122

5.    3: R2=0.9611  F= 82.2747 (three descriptors) 331 79 59

6.     Two descriptors:

7.     best correlations (F):

8.     1: R2=0.9486  F=101.4772 (two descriptors) 331 79

9.     2: R2=0.9246  F= 67.4424 (two descriptors) 331 81

10. 3: R2=0.9156  F= 59.6897 (two descriptors) 331 141

11.  One descriptor:

12.  best correlations (F):

13.  1: R2=0.8776  F= 86.0022 (one descriptor) 331

14. 2: R2=0.7663  F= 39.3455 (one descriptor) 48

15.  Descriptors involved:

16.  331 – ALFA polarizability (DIP);

17.    79 – Total point-charge component of the molecular dipole;

18.  141 – Minimum valency of a N-atom;

19.  122 – Minimum atomic orbital electronic population;

20.    59 – Maximum atomic one-electron react. index for a C-atom;

21.    81 – Total dipole of the molecule;

22.   48 – Average atomic electrophilic reaction index for a C-atom.

With these descriptors can be obtained a Hänsch-type equation:

log P = a0 + a1 X1 + a2 X2 + a3 X3,

linking the descriptors considered significant in Table 4 (X1 – ALFA polarizability, X2 – Total point-charge component of the molecular dipole, X3 – Minimum valency of a N-atom) with the partition coefficient of the studied substances, the regressional correlation coefficient being in this case close to unity (R2=0.9694, standard error of predicted values being 0.0408).

Values of regression coefficients of Hänsch-equations estimated by multiple linear regression method (CODESSA) are:

▪ a0 = -2.1337 ∙ 10+1 (interception or free term);

▪ a1 = 3.2353 ∙ 10-2;

▪ a2 = -3.0078 ∙ 10-1;

▪ a3 = 5.6162.

The degree of trust between observed and predicted values obtained by using this equation is represented in Figure 2. As can be seen in the diagram of the degree of confidence of observed and predicted values, can be concluded that Hänsch-type equation established in this study can be used to predict the biological activity of other new derivatives of the series.


Figure 2 – The degree of trust for Hänsch-equation.


Conclusions

1. Based on structural models, different physico-chemical properties and structural descriptors for a series of compounds from oxicam class were assessed in order to characterize these compounds.

2. Correlation of chemical structure–biolo-gical property for this substances showed the potential role of descriptors dependent of the substances chemical structures, i.e. the partici-pation of these descriptors to partition in the aqueous/lipid phases (biological membrane) characterized by partition coefficients log P.

3. Substances with positive lipophilicity (log P>0) penetrate more easily through membra-nes, lipophilicity being favorable for the transfer of the molecule in the aqueous phase and showing its ability to get in contact with the receptor/target cell.

4. Hänsch-equation established in this study allows calculation of the partition coefficient of a new derivative by chemical modulation using the computer, before this new substance is tested and its biological activity is determined experimentally.

References

1.     Demertzi-Kovala D – (2006) Recent advances on non-steroidal anti-inflammatory drugs, NSAIDs: Organotin complexes of NSAIDs, J Organomet Chem, 691(8):1767–1774.

2.     Bloemink MJ, Reedijk J – (1996) Cisplatin and derived anticancer drugs: mechanism and current status of DNA binding. In: Sigel H, Sigel A (eds), Metal ions in biological systems. Volume 32: Interactions of metal ions with nucleotides, nucleic acids, and their constituents, Marcel Dekker, New York, 641–685.

3.     Østensen ME, Skomsvoll JF – (2004) Anti-inflam-matory pharmacotherapy during pregnancy, Expert Opin Pharmacother, 5(3):571–580.

4.     Botting JH – (1999) Nonsteroidal antiinflammatory agents, Drugs Today (Barc), 35(4–5):225–235.

5.     Reitz DB, Isakson PC – (1995) Cyclooxygenase-2 inhibitors, Curr Pharm Des, 1:211–220.

6.     Vane JR, Botting RM – (1996) Mechanism of action of anti-inflammatory drugs, Scand J Rheumatol Suppl, 102:9–21.

7.     Marnett LJ, Kalgutkar AS – (1998) Design of selective inhibitors of cyclooxygenase-2 as nonulce-rogenic anti-inflammatory agents, Curr Opin Chem Biol, 2(4):482–490.

8.     Hawkey CJ – (1999) COX-2 inhibitors, Lancet, 353(9149):307–314.

9.     Kurumbail RG, Stevens AM, Gierse JK, McDonald JJ, Stegeman RA, Pak JY, Gildehaus D, Miyashiro JM, Penning TD, Seibert K, Isakson PC, Stallings WC – (1996) Structural basis for selective inhibition of cyclooxygenase-2 by anti-inflammatory agents, Nature, 384(6610):644–648.

10. Klein T, Nüsing RM, Pfeilschifter J, Ullrich V – (1994) Selective inhibition of cyclooxygenase 2, Biochem Pharmacol, 48(8):1605–1610.

11. Nicolas C, Verny M, Giraud I, Ollier M, Rapp M, Maurizis JC, Madelmont JC – (1999) New quarter-nary ammonium oxicam derivatives targeted toward cartilage: synthesis, pharmacokinetic studies, and anti-inflammatory potency, J Med Chem, 42(25):5235–5240.

12. *** – (2010) Molecular Modeling ProTM Plus, website http://www.chemsw.com/13071.htm.

13. *** – (2010) MOPAC 7.0 for UNIX, Quantum Chemistry Program Exchange, Project 688.

14. Lee B, Richards FM – (1971) The interpretation of protein structures: estimation of static accessibility, J Mol Biol, 55(3):379–400.

15. Connolly ML – (1983) Solvent-accessible surfaces of proteins and nucleic acids, Science, 221(4612):709–713.

16. Richmond TJ – (1984) Solvent accessible surface area and excluded volume in proteins. Analytical equations for overlapping spheres and implications for the hydrophobic effect, J Mol Biol, 178(1):63–89.

17. Karelson M – (2000) Molecular descriptors in QSAR/QSPR, John Wiley & Sons, New York.

18. Katritzky AR, Petrukhin R, Yang H, Karelson M – (2003) Comprehensive descriptors for structural and statistical analysis. Codessa Pro reference manual, University of Florida.


 

 

Correspondence Adress: Denisa-Constantina Amzoiu, University Assistant, PhD candidate, Department of Pharmaceutical Chemistry, Faculty of Pharmacy, University of Medicine and Pharmacy of Craiova, 2–4 Petru Rareş Street, 200349 Craiova, Romania; Phone +40251–524 442, e-mail: damzoiu@yahoo.com


All articles in this issue