a comparison of theoretical methods of calculation of partition ...

Acta Poloniae Pharmaceutica ñ Drug Research, Vol. 63 No. 3 pp. 159ñ167, 2006

ISSN 0001-6837 Polish Pharmaceutical Society

ANALYSIS

A COMPARISON OF THEORETICAL METHODS OF CALCULATION OF PARTITION COEFFICIENTS FOR SELECTED DRUGS ALINA PYKA*, MAGDALENA BABUåKA, and MAGDALENA ZACHARIASZ Silesian Academy of Medicine, Faculty of Pharmacy, Department of Analytical Chemistry, 4 JagielloÒska Str., 41-200 Sosnowiec, Poland Abstract: Hundred ninety three drugs of different pharmacological activity were studied. Lipophilicity of a drug is one of the parameters, which influence its biological activity. The n-octanol-water partition coefficients were calculated for these compounds by use of different theoretical procedures (AlogPs, IAlogP, miLogP, ClogP, logPKowwin, and xlogP). Particular theoretical partition coefficients were compared with experimental noctanol-water partition coefficients (logPexp) for all studied drugs. It was shown that experimental partition coefficients correlate the best with theoretical partition coefficients calculated by use of logPKowwin and AlogPs methods. It was shown that it exists the possibility of the prediction of experimental n-octanol-water partition coefficients on the basis of logPKowwin, AlogPs, and ClogP for fifteen drugs (adrenalin, clobazam, 5,5-dimethylbarbituric acid, ethyl nicotinate, fluphenazine, ibuprofen, methyllorazepam, pimozide, prednisolone, promethazine, spironolactone, surital, theophylline, triamterene, and trimethoprim). Keywords: Drug, lipophilicity, theoretical partition coefficient, experimental logP, QSAR

shaking. The compatibility of experimental and theoretical approaches to determination of lipophilicity of organic compounds, including drugs, remains a focus of scientific interest (10,11,18). The aim of this work was the comparison of the experimental n-octanol-water partition coefficients with theoretical partition coefficients calculated by use of different theoretical procedures for 193 drugs with different pharmacological activity.

Lipophilicity of a substance, including drugs, is one of the parameters, which influence its biological activity (1-9). The n-octanol-water partition coefficients, usually expressed as logP values, are used as a measure of lipophilicity. LogP values can be computed or determined experimentally. LogP can be determined experimentally by a classical method of measuring partition of organic compounds between a non polar phase and water (10). The importance of the use of partition coefficients in quantitative structure activity relationships (QSAR) is well established for prediction of biological or pharmacological activity of compounds (11-20). Hansch et al. initiated the study of QSAR in 1962, with a publication on plant growth regulation in which bioactivity was successfully coupled with lipophilicities and electronic character of the phenyl substituents (10,21). QSAR play a crucial role in designing new drugs. The possibility to predict drugs biological properties due to their lipophilicity allows the optimization of new drugs structure designs. Measurement of partition coefficients is not as easy as one would expect from their simple definition. The measurement of partition coefficients by equilibration methods is frequently difficult, or even impossible, due to the impurity or instability of compounds, by a strong preference of the compounds for one of the two phases of the system, or by the formation of stable emulsions after

EXPERIMENTAL The n-octanol-water partition coefficients were calculated for the hundred ninety three drugs by use of different theoretical procedures (22-28). Numerical values of the partition coefficients were obtained using the following methods: AlogPs ñ this method, for the assessment of noctanol-water partition coefficient, was developed on the basis of neutral network ensemble analysis of 12908 organic compounds. The atom and bond-type E-state indices, the number of hydrogen and nonhydrogen atoms were considered in calculations. A first selection of indices was performed by multiple linear regression analysis, and 75 input parameters were selected. Some of the parameters combined several atom-type or bond-type indices with similar physicochemical properties (23,25).

* Corresponding author: e-mail ñ [email protected]

159

160

ALINA PYKA et al.

IAlogP ñ this method bases on the set of 238 MolconnZ molecular indices generated from 13000 organic structures with accurately measured logP values. Interactive Analysis, using neutral networks technology to derive a set of 10 fold cross-validated networks, developed the IAlogP predictor (22,23). ClogP ñ this method estimates interaction parameters for new fragments; ClogP bases on estimation of these interactions on the well-accepted principles defined by Hammett, so not only is ClogP the most accurate calculation available. but it produces results that are based on established chemical interactions, unlike other programs which are based solely on statistics (23,26) logPKowwin ñ this is an atom-fragment contribution method for n-octanol-water logP calculation. This method logPKowwin bases on data for 13058 organic compounds, for which accurate experimental partition coefficients logP are published (23,27). xlogP ñ this is an atom-additive method for noctanol-water logP calculation. XlogP method includes correction factors to account for some

intramolecular interactions; 1831 organic compounds were analyzed by multivariate regression to derive the parameters. After including these correction factors, the final logP is described as: logP = Σi aiAi+Σj bjBj

[1]

where ai and bj are regression coefficients, Ai is the number of occurrences of the ith atom type, and Bj is the number of occurrences of the jth correction factor identified by this program (23,28) MILOGP ñ method for logP prediction developed at Molinspiration (miLogP 2.2) is based on group contributions. These have been obtained by fitting calculated logP with experimental logP for a training set of more than twelve thousand, mostly drug-like molecules (24). RESULTS AND DISCUSSION Hundred ninety three drugs with different pharmacological activity were studied. The theoret-

Figure 1. Relationships between the experimental n-octanol-water partition coefficients (logPexp) and theoretical AlogPs values.

Figure 2. Relationships between the experimental n-octanol-water partition coefficients (logPexp) and theoretical logPKowwin values.

161

A comparison of theoretical methods of calculation of partition...

Table 1. Experimental (10,19,23,29-31) and theoretical partition coefficients (22-28) for examined drugs.

No.

Drug

1 2 3

Adrenalin Allopurinol 5-Allyl-5-(1-methyl-1butyl)barbituric acid Alprazolam Alprendol Amantadine Amfetamine 4-Aminosalicylic acid Amitriptyline Amobarbital Amoxicillin Ampicillin Androstenedione Androsterone Aprobarbital Atenolol Atropine Azoperone Barbital Barbituric acid Betamethasone Benorylat Benperidol Benzyl nicotinate Bifanazole Bromazepam Buthabarbital Butyl nicotinate Captopril Carbamazepine Chloramphenicol Chloropropamide Chlorothiazide Chlorpromazine Chlorprothixene Cimetidine Clobazam Clofibric acid Clomipramine Clonazepam Clozapine Codeine Corticosterone Cortisone 3-Cyanopyridine Cyclopal Dehydroepiandrosterone N-Demethyldiazepam Demoxepam Desipramine Desmethyldiazepam

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51

logPexp (10, 19, 23, 29-31) -1.37 -0.55

AlogPs

IAlogP ClogP miLogP logPKowwin xlogP

-0.60 -0.18

-0.99 -0.06

-0.69 0.63

-0.06 -0.08

-0.69 -0.55

0.55 0.91

1.97

2.20

2.26

2.16

2.08

2.36

2.28

2.12 3.10 2.44 1.76 0.89 4.92 2.07 0.87 1.35 2.75 3.69 1.15 0.16 1.83 3.30 0.65 -1.47 1.83 2.15 3.91 2.40 4.77 2.05 1.65 2.27 0.34 2.45 1.14 2.27 -0.24 5.35 5.18 0.40 2.12 2.57 5.19 2.41 3.23 1.19 1.94 1.47 0.36 1.51 3.23 2.93 1.41 4.90 2.93

2.23 2.38 2.53 1.85 0.62 5.10 1.85 0.99 1.05 3.02 3.71 1.21 0.57 2.20 2.76 0.74 -1.12 1.95 2.60 3.52 2.39 5.29 2.09 1.70 2.24 1.02 2.10 1.15 2.15 0.01 5.26 5.42 0.48 2.13 2.81 5.33 2.73 3.80 1.20 2.09 1.99 0.36 1.58 3.53 2.79 0.99 4.07 2.79

3.20 2.81 2.24 1.85 0.66 5.15 2.06 -2.21 -1.10 3.25 3.46 1.15 0.23 1.76 2.35 0.57 -1.27 1.82 2.34 2.79 2.00 5.56 2.22 1.38 2.29 0.25 2.76 0.61 2.24 0.34 5.40 5.07 0.41 2.20 2.93 5.09 2.59 3.20 1.03 3.47 2.46 0.57 1.19 2.29 2.90 2.35 4.10 2.90

2.56 2.65 2.00 1.74 1.06 4.85 2.11 -1.87 -1.20 2.82 3.55 1.10 -0.11 1.30 3.34 0.65 -1.44 1.78 2.05 3.80 2.60 4.99 1.70 1.58 2.35 0.89 2.38 1.28 2.35 -0.29 5.30 5.48 0.38 2.44 2.82 5.92 2.38 3.71 0.98 2.32 1.29 0.27 1.16 3.07 3.02 1.07 4.47 3.02

2.29 2.58 2.65 1.32 0.92 4.19 1.78 -1.35 -0.87 3.06 3.43 1.02 0.72 1.77 2.88 0.74 -1.48 2.06 2.61 3.41 2.13 4.97 2.41 1.49 1.97 -1.09 2.84 0.73 2.21 0.02 5.03 5.43 0.14 2.55 2.73 4.82 2.77 4.14 1.41 1.88 1.43 0.46 1.19 3.24 2.86 1.23 3.92 2.86

3.07 2.81 2.43 1.76 0.98 4.95 2.00 0.97 1.45 2.99 3.07 1.38 -0.03 1.91 3.23 0.60 -1.25 1.72 1.75 3.55 2.35 5.71 1.93 1.51 2.11 0.84 2.25 0.92 2.01 5.20 5.14 0.57 1.82 2.84 5.65 2.53 2.84 1.28 1.99 1.81 0.35 2.03 2.98 2.87 1.23 4.80 2.8

4.89 2.84 2.31 1.76 1.61 4.93 2.08 0.03 0.43 3.06 4.30 1.14 0.46 1.83 3.35 0.43 -1.63 1.14 2.56 3.00 2.42 5.49 1.79 1.51 2.06 0.64 2.30 0.69 1.80 -0.50 4.92 5.19 0.74 2.28 2.58 4.65 2.67 3.74 1.08 1.67 0.10 0.49 1.42 3.04 2.78 2.42 3.79 2.78

162

ALINA PYKA et al.

Table 1. cont.

No.

Drug

52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102

Dexamethasone 5,5-Diallylbarbituric acid Diazepam N,N-Diethylnicotinamide Diltiazem hydrochloride 5,5-Dimethylbarbituric acid Diphenhydramine Disopyramide Dopamine Dothiepin Dozepine Droperidol Erythromycin Estradiol Estriol Estrone Ethopropazine Ethyl nicotinate 5-Ethyl-5-heptyl barbituric acid 5-Ethyl-5-methyl barbituric acid 5-Ethyl-5-nonyl barbituric acid 5-Ethyl-5-octylbarbituric acid 5-Ethyl-5-pentylbarbituric acid 5-Ethyl-5-propyl barbituric acid 5-Ethylbarbituric acid Fluanisone Flufenamic acid Flumazenil Flunitrazepam Fluoxymesterone Flupenthixol Fluphenazine Flurazepam Flurbiprofen Fluspirilene Furosemide Glipizide Haloperidol Hexachlorophene Hexobarbital Hexyl nicotinate Hydrochlorothiazide Hydrocortisone Hyoscine Hyoscyamine Ibuprofen Imipramine Indomethacin Isoniazid Ketoconazole Ketoprofen

logPexp (10, 19, 23, 29-31) 1.83 1.15 2.82 0.33 2.70 -0.44 3.27 2.58 -0.98 4.49 4.29 3.50 3.06 4.01 2.45 3.13 4.77 1.32 3.37 0.19 4.43 3.78 2.24 1.25 -0.35 3.60 5.25 1.00 2.06 2.38 4.51 4.36 2.35 4.16 5.86 2.03 1.91 4.30 7.54 1.98 3.51 -0.07 1.61 0.98 1.83 3.97 4.80 4.27 -0.70 4.35 3.12

AlogPs 1.95 1.07 2.60 0.83 3.09 -0.21 3.44 3.21 -0.40 4.98 4.25 3.92 2.34 3.57 2.54 4.03 5.88 1.17 3.18 0.19 4.12 3.60 2.15 1.25 -0.34 3.82 4.74 1.52 2.20 2.50 4.56 4.40 3.81 3.57 5.17 2.35 1.83 3.70 6.77 1.80 3.06 -0.16 1.71 1.40 2.20 3.50 4.58 4.25 -0.71 4.30 3.28

IAlogP ClogP miLogP logPKowwin xlogP 2.58 0.92 2.95 1.58 2.65 0.34 3.39 2.97 -0.48 4.83 4.08 2.61 2.15 2.73 1.67 2.08 5.13 1.33 3.16 0.19 4.18 3.66 2.09 1.02 -0.41 3.00 4.89 1.23 1.92 2.41 4.53 4.05 4.02 3.59 4.81 1.76 2.47 3.41 7.06 1.35 3.27 0.05 2.91 0.90 1.76 3.54 4.62 3.58 -0.58 4.11 2.78

1.78 0.75 2.96 0.56 3.65 -0.40 3.45 2.58 0.17 4.53 4.09 3.06 1.61 3.78 3.20 3.38 5.46 1.30 3.30 0.13 4.36 3.83 2.24 1.18 -0.39 4.22 5.53 1.29 1.78 2.65 4.34 4.12 4.22 3.75 5.42 1.90 2.57 3.85 7.03 1.63 3.41 -0.37 1.70 0.29 1.30 3.68 5.04 4.18 -0.67 3.63 2.76

2.06 0.81 2.74 0.42 3.34 0.07 3.50 2.78 -0.05 4.23 3.86 3.40 2.28 3.43 2.51 3.24 5.19 0.91 3.32 0.40 4.33 3.82 2.31 1.30 -0.50 3.79 4.84 0.86 2.13 2.76 4.91 4.51 3.64 4.05 5.54 1.77 2.31 4.30 7.10 2.10 2.98 -0.06 1.62 1.05 1.77 3.46 4.16 3.99 -0.97 3.77 3.59

1.72 1.31 2.70 0.52 2.79 -0.38 3.11 2.96 0.38 4.51 3.99 3.46 2.48 3.94 2.81 3.43 5.47 1.13 3.06 0.11 4.04 3.55 2.08 1.10 -0.34 3.86 5.15 1.03 1.91 2.49 4.07 4.13 3.02 3.81 5.59 2.32 3.35 4.20 6.92 2.02 3.10 1.62 0.39 1.91 3.79 5.01 4.23 -0.81 4.45 3.00

1.14 0.84 2.92 1.16 2.84 -0.70 3.16 3.49 0.92 4.56 3.99 2.75 1.14 4.23 3.63 3.63 5.25 1.13 3.28 -0.14 4.42 3.85 2.14 1.00 -0.85 3.79 4.42 1.92 2.35 1.96 4.42 4.16 3.78 3.76 6.09 1.41 2.53 3.98 6.27 1.18 3.19 --0.47 0.52 0.79 1.83 3.64 4.03 4.18 -0.82 3.96 3.22

163

A comparison of theoretical methods of calculation of partition...

Table 1. cont.

No.

Drug

103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153

Librium Lidocaine Lorazepam Metamphetamine Mebendazole Medazepam Metharbital Methotripremazine Methyl nicotinate Methyl salicylate 5-Methylbarbituric acid Methyllorazepam 17a-Methyltestosterone Metronidazole Mexiletine Naproxen Nembutal Nicotinamide Nicotinic acid Nifedipine Nimodipine Nitrazepam Nitrofurantoin Norepinephrine Norethindrone 19-Nortestosterone Orphenandrine Oxaproxin Oxazepam Oxprendol Pentobarbital Pentoxifylline Pericyazine Phenobarbital Phenylbutazone Phenylephrine Phenytoin Perphenazine Pimozide Pindolol Pipamperone Pirimidone Prazepam Prednisolone Prednisone Procainamide Prochlorperazine Progesterone Promethazine Propafenone Propanolol

logPexp (10, 19, 23, 29-31) 2.44 2.26 2.39 2.07 2.83 4.41 1.15 4.68 0.83 2.55 -0.84 2.58 3.36 -0.02 2.15 3.18 2.10 -0.37 0.36 2.20 3.05 2.25 -0.47 -1.24 2.97 2.62 3.77 4.19 2.24 2.10 2.10 0.29 3.52 1.47 3.16 -0.31 2.47 4.20 6.30 1.75 2.02 0.91 3.73 1.62 1.46 0.88 4.88 3.87 4.81 4.63 3.48

AlogPs 2.01 1.81 2.98 2.23 2.67 4.08 1.18 4.97 0.47 2.07 -0.82 2.92 3.62 -0.15 2.17 3.29 2.17 -0.45 0.29 2.32 3.08 1.95 0.03 -1.40 2.72 2.60 3.55 3.33 2.01 2.50 2.17 0.08 3.78 1.41 2.81 -0.62 2.26 4.16 6.37 2.13 2.32 0.62 3.68 1.64 2.07 1.42 4.67 3.58 4.52 3.11 3.03

IAlogP ClogP miLogP logPKowwin xlogP 0.43 2.21 2.53 2.15 2.72 3.93 0.82 4.93 0.82 1.98 -0.87 3.10 2.41 -0.79 2.23 3.32 1.92 -0.16 0.59 1.52 2.93 1.79 -0.37 -1.65 3.45 2.68 3.69 3.56 2.64 2.07 1.92 0.17 3.53 1.80 3.39 -0.13 2.52 4.31 4.92 1.41 2.71 1.15 3.82 1.23 1.28 1.34 4.99 3.33 4.70 3.25 2.96

3.79 1.95 2.37 1.89 3.08 4.12 1.14 4.83 0.77 2.34 -0.92 2.40 3.74 -0.46 2.57 2.82 2.11 -0.21 0.80 3.13 4.00 2.32 -0.47 0.99 2.78 2.70 3.90 2.95 2.31 2.09 2.11 0.12 3.23 1.37 3.38 -0.09 2.09 3.81 6.40 1.67 2.24 0.88 3.93 1.42 1.66 1.42 4.38 3.78 4.40 3.64 2.75

2.12 2.13 2.47 2.23 2.89 3.42 1.21 4.89 0.53 2.13 -1.00 2.71 3.69 -0.47 2.00 3.38 2.04 -0.48 0.27 3.07 4.10 2.14 -0.05 -1.04 3.23 3.00 3.90 3.75 1.84 2.07 2.04 0.65 3.13 0.80 4.56 0.41 2.18 4.29 5.62 1.98 2.33 0.92 3.61 1.60 1.41 0.99 4.92 3.81 4.44 3.46 2.97

2.42 1.66 2.41 2.22 2.71 4.43 0.82 5.06 0.64 2.60 -0.83 2.23 3.72 0.00 2.61 3.10 2.00 -0.45 0.69 2.50 3.13 2.45 -0.17 -1.16 2.99 2.82 3.65 4.04 2.32 1.83 2.00 0.56 3.93 1.33 3.52 -0.21 2.16 3.82 6.23 1.48 2.28 0.73 3.99 1.40 1.59 0.97 4.79 3.67 4.49 3.37 2.60

3.39 2.10 3.47 2.15 2.79 4.40 0.58 4.51 0.71 2.54 -1.41 3.61 4.01 -0.14 2.07 2.84 2.08 -0.34 0.39 2.37 3.07 2.05 -0.09 0.16 3.21 3.19 3.59 3.67 2.84 2.02 2.08 0.03 3.86 1.32 3.71 0.52 2.22 3.86 5.60 1.92 1.73 1.57 3.66 1.02 0.61 1.31 4.57 3.52 4.40 3.20 3.03

164

ALINA PYKA et al.

Table 1. cont.

No.

Drug

154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193

Pseudoephedrine 3-Pyridinemethanol 1-(3-Pyridinyl)etanone Pyrilamine Quinidine Rutinal Salicylamide Salicylanilide Spiperone Spironolactone Sulfadiazine Sulfadimidine Sulfaguanidine Sulfamethoxazole Sulfanilamide Sulfathiazole Sulfisoxazole Sulfoxyphenyl pyrazolidine Sultrin Surital Talbutal Temazepam Tertbutaline Testosterone Tetracaine Tetracycline Tetrazepam Theophylline Thiobarbital Thioridazine Thiothixene Thymol Tolbutamide Triamterene Triazolam Trifluoperazine Trifluperidol hydrochloride Trimeprazine Trimethoprim Verapamil

logPexp (10, 19, 23, 29-31) 0.89 -0.02 0.43 3.27 3.44 0.91 1.28 3.27 3.03 2.78 -0.09 0.89 -1.22 0.89 -0.62 0.05 1.01 2.30 -0.96 3.23 1.47 2.19 0.90 3.32 3.51 -1.30 3.20 -0.02 1.50 5.90 3.78 3.30 2.34 0.98 2.42 5.54 3.03 4.71 0.91 3.79

AlogPs 1.00 -0.13 0.45 2.89 2.82 0.99 0.74 2.65 2.86 3.18 0.25 0.43 -0.55 0.79 -0.16 0.94 1.14 2.92 0.15 3.11 1.79 2.46 0.55 2.99 3.54 -0.68 3.53 -0.32 1.72 5.93 4.00 3.16 2.04 1.21 2.94 4.96 2.86 4.99 1.26 5.23

IAlogP ClogP miLogP logPKowwin xlogP 1.03 0.18 0.65 2.46 3.26 1.35 1.01 3.26 2.45 2.44 0.29 0.79 -1.37 0.75 -0.41 0.85 0.95 2.12 -0.25 3.18 1.50 2.20 0.77 2.39 3.10 -1.73 3.34 -0.49 1.49 5.69 3.81 3.06 2.05 -0.03 3.56 4.94 2.19 5.08 1.21 6.93

0.89 0.06 0.48 3.22 2.79 0.84 1.28 3.27 2.82 2.65 0.10 1.10 -1.24 0.56 -0.57 0.73 0.22 1.65 -0.98 3.03 1.63 2.34 0.48 3.22 3.83 -0.91 3.68 -0.03 1.52 6.00 3.23 3.20 2.50 1.61 2.62 5.61 2.82 4.80 0.98 4.47

1.24 0.04 0.60 2.63 3.06 0.47 1.25 3.32 3.34 3.03 -0.04 0.83 -0.84 0.61 -0.29 0.83 0.99 3.35 -0.56 2.62 1.52 2.08 1.07 3.24 3.43 -0.69 3.84 0.00 1.28 5.68 3.95 3.34 2.54 0.82 2.92 5.25 3.34 4.85 1.00 4.55

0.68 -0.11 0.49 2.81 3.29 0.84 1.03 3.30 2.92 2.88 -0.34 0.76 -1.07 0.48 -0.55 0.72 1.03 2.14 -0.60 3.23 1.87 2.15 0.67 3.27 3.02 -1.33 3.67 -0.39 1.47 6.45 3.14 3.52 2.41 0.80 3.96 5.52 2.92 4.98 0.73 4.80

1.39 -0.04 0.61 3.18 2.60 0.75 0.93 3.02 3.48 3.41 -0.17 0.46 -0.45 0.68 -0.84 -0.03 0.90 2.40 -0.56 2.53 1.71 2.99 1.42 3.60 2.81 -0.93 2.75 -0.48 0.68 5.94 3.30 3.25 2.18 0.28 5.51 5.23 3.48 4.60 0.65 4.70

Table 2. Matrix of correlation coefficients of linear dependences between particular partition coefficients (n = 193).

logPexp

AlogPs

IAlogP

ClogP

miLogP

logPKowwin*

xlogP

1

0.9717

0.9315

0.9546

0.9607

0.9777

0.9295

1

0.9435

0.9523

0.9623

0.9740

0.9350

1

0.9328

0.9462

0.9380

0.9085

1

0.9626

0.9574

0.9387

1

0.9553

0.9376

1

0.9465

logPexp * AlogPs IAlogP ClogP miLogP logPKowwin xlogP * for logPKowwin n =191

1

Adrenalin

Clobazam

5,5-Dimethylbarbutiric acid

Ethyl nicotinate

Fluphenazine

Ibuprofen

Methyllorazepam

Pimozide

Prednisolone

Promethazine

Spironolactone

Surital

Theophylline

Triamterene

Trimethoprim

1

37

57

69

83

97

114

141

146

151

163

173

181

187

192

0.91

0.98

-0.02

3.23

2.78

4.81

1.62

6.30

2.58

3.97

4.36

1.32

-0.44

2.12

-1.37

logPexp

Agreement with experimental logP(exp) ñ correlation coefficient (r) and determination coefficient (r2, %)

Omitted drug

Drug No.

1.21

1.16

-0.39

3.09

3.16

4.52

1.60

6.40

2.90

3.49

4.40

1.12

-0.28

2.09

-0.68

logPpred

0.9895 97.91%

-0.30

-0.18

0.37

0.14

-0.38

0.29

0.02

-0.10

-0.32

0.48

-0.04

0.20

-0.16

0.03

-0.69

∆logP= logPpred-AlogPs

logP predicted by use of eq. [6]

0.73

0.80

-0.37

3.20

2.85

4.44

1.40

6.16

2.21

3.75

4.09

1.13

-0.36

1.81

-0.67

logPpred

0.9928 98.57%

0.18

0.18

0.35

0.03

-0.07

0.37

0.22

0.14

0.37

0.22

0.27

0.19

-0.08

0.31

-0.70

∆logP= logPpred-logPKowwin

logP predicted by use of eq. [7]

Table 3. Experimental n-octanol-water partition coefficients of drugs omitted from eqs. [6]-[9], and predicted by these equations.

1.07

1.66

0.12

3.00

2.64

4.29

1.48

6.17

2.40

3.61

4.02

1.37

-0.23

2.44

-0.50

logPpred

0.9908 98.17%

-0.16

-0.68

-0.14

0.23

0.14

0.52

0.14

0.13

0.18

0.36

0.34

-0.05

-0.21

-0.32

-0.87

∆logP= logPpred-ClogP

logP predicted by use of eq. [8]

1.05

0.88

0.09

2.62

3.02

4.39

1.64

5.53

2.71

3.44

4.45

0.97

0.15

2.56

0.03

-0.14

0.10

-0.11

0.61

-0.24

0.42

-0.02

0.77

-0.13

0.53

-0.09

0.35

-0.59

-0.44

-1.40

∆logP= logPpred-miLogP

0.9781 95.67%

logPpred

logP predicted ] by use of eq. [9

A comparison of theoretical methods of calculation of partition...

165

166

ALINA PYKA et al.

ical partition coefficients were calculated by use of different methods (22-28). Numerical values of experimental n-octanol-water partition coefficients for all studied drugs were reported previously (10,19,23, 29-31) and together with theoretical partition coefficients (22-28) are presented in Table 1. The experimental values of n-octanol-water partition coefficients (logPexp) with the theoretical partition coefficients (AlogPs, IAlogP, ClogP, miLogP, logPKowwin, and xlogP) for the studied drugs were compared. The linear dependencies were received between experimental n-octanol-water partition coefficients (logPexp) and particular theoretical partition coefficients (AlogPs, IAlogP, ClogP, miLogP, logPKowwin, and xlogP. The obtained correlation coefficient values for the linear dependencies between particular partition coefficients are presented in Table 2. They indicate, that theoretical partition coefficients calculated using IAlogP and xlogP methods are the least agreed with the experimental partition coefficients (logPexp) (r=0.9315, r2=86.77% and r=0.9295, r2=86.40%, respectively). The remaining theoretical partition coefficients (AlogPs, ClogP, miLogP, and logPKowwin) show high agreement with experimental partition coefficients (logPexp). For those dependencies correlation coefficients have larger values than 0.954. We have been affirmed that experimental values of n-octanol-water partition coefficients for all drugs correlate the best with logPKowwin and AlogPs values. The relationships between experimental n-octanol-water partition coefficients (logPexp) and AlogPs as well as logPKowwin are presented in Figs. 1 and 2, respectively. The computer program logPKowwin based on atom-fragmental contributions did not calculate the partition coefficients for chlorothiazide (33) and hydrochlorothiazide (93). The values of partition coefficients for these compounds calculated on the basis of equation [3] are equal ñ0.26 and ñ0.08, respectively. For flurazepam (84), propafenone (152) , and verapamil (193), the Studentized residual for use of eq. [2] were ñ3.72, 3.95 and ñ3.76, respectively (Fig.1). For dopamine (60), glipizide (88), propafenone (152), and triazolam (188), the Studentized residual for use of eq. [3] were ñ4.00, 4.15, 3.69, and ñ4.46, respectively (Fig. 2). These points are particularly distant from straight lines (Figs. 1 and 2). The experimental values of n-octanol-water partition coefficients for all studied drugs correlate also best with theoretical ClogP and miLogP: logPexp =0.1148(±0.0614) + 0.9539(±0.0215)◊ClogP [4] n=193; r=0.9546; r2=91.13%; F=1963; s=0.51; p