The contribution of the hydrogen bond acidity on the lipophilicity of drugs estimated from chromatographic measurements.

The influence of the hydrogen bond acidity when the 1-octanol/water partition coefficient (log P(o/w)) of drugs is determined from chromatographic measurements was studied in this work. This influence was firstly evaluated by means of the comparison between the Abraham solvation parameter model when it is applied to express the 1-octanol/water partitioning and the chromatographic retention, expressed as the solute polarity p. Then, several hydrogen bond acidity descriptors were compared in order to determine properly the log P(o/w) of drugs. These descriptors were obtained from different software and comprise two-dimensional parameters such as the calculated Abraham hydrogen bond acidity A and three-dimensional descriptors like HDCA-2 from CODESSA program or WO1 and DRDODO descriptors calculated from Volsurf+software. The additional HOMO-LUMO polarizability descriptor should be added when the three-dimensional descriptors are used to complement the chromatographic retention. The models generated using these descriptors were compared studying the correlations between the determined log P(o/w) values and the reference ones. The comparison showed that there was no significant difference between the tested models and any of them was able to determine the log P(o/w) of drugs from a single chromatographic measurement and the correspondent molecular descriptors terms. However, the model that involved the calculated A descriptor was simpler and it is thus recommended for practical uses.

Evaluation of fluorescence anisotropy to assess drug-lipid membrane partitioning.

PURPOSE: We evaluated fluorescence anisotropy measurements as an alternative technique to estimate drug-lipid membrane partitioning of fluorescent solutes. METHODS: The lipid bilayer partitioning of six drugs (logP in octanol/water between 2.6 and 5.4) was investigated by fluorescence anisotropy measurements with egg phosphatidylcholine liposomes at pH 7.4. Anisotropy was measured at about 5 and 50µM drug and varying lipid concentrations between 3 and 700µM. Fluorescence was corrected for light scattering and membrane affinities were estimated by non-linear regression analysis of the relative anisotropy as a function of lipid and solute concentrations. Liposome partitioning was in addition determined by equilibrium dialysis and potentiometric titration for comparison. RESULTS: Correction for light scattering by the liposomes was possible to some extent for two drugs. The estimated partition coefficients of three drugs were concentration-independent. For two drugs, membrane saturation was expected at the higher drug concentration. One drug showed significant differences between the parameters estimated at high and low drug concentrations, indicating measurement artifacts. Linear regression between the estimated logarithmic partition coefficients from anisotropy measurements and equilibrium dialysis revealed a slope of 1.05 and an intercept of 1.16 (n=5, r(2)=0.87). CONCLUSION: Anisotropy measurement with liposomes is hampered by light scattering and pH-dependent fluorescence properties of ionizable drugs. Taking these limitations into account, the technique may offer an alternative to established methods for the estimation of drug membrane partitioning, in particular when potentiometric titration or equilibrium dialysis are not applicable.

Extension of the liquid chromatography/quantitative structure-property relationship method to assess the lipophilicity of neutral, acidic, basic and amphotheric drugs.

A reported chromatographic method to determine the 1-octanol/water partition coefficient (logP(o/w)) has been used to estimate the lipophilicity of 33 drugs with diverse structures and functionalities, including neutral, acid, basic, and amphoteric compounds. The applicability of the chromatographic method has been extended to the UHPLC technique, and the results obtained were compared to those obtained from conventional HPLC. No significant difference between the results obtained by both techniques is noticed. Thus, the suitability of UHPLC, which involves shorter run times, for lipophilicity assessment is demonstrated. In order to show the consistency of this chromatographic method, the logP(o/w) values of those drugs which have acid-base properties have been also determined by potentiometry, and the final results have been compared with both values derived from the chromatographic method and the ones from the literature.

Lipophilicity assessment of basic drugs (log P(o/w) determination) by a chromatographic method.

A previously reported chromatographic method to determine the 1-octanol/water partition coefficient (log P(o/w)) of organic compounds is used to estimate the hydrophobicity of bases, mainly commercial drugs with diverse chemical nature and pK(a) values higher than 9. For that reason, mobile phases buffered at high pH to avoid the ionization of the solutes and three different columns (Phenomenex Gemini NX, Waters XTerra RP-18 and Waters XTerra MS C(18)) with appropriate alkaline-resistant stationary phases have been used. Non-ionizable substances studied in previous works were also included in the set of compounds to evaluate the consistency of the method. The results showed that all the columns provide good estimations of the log P(o/w) for most of the compounds included in this study. The Gemini NX column has been selected to calculate log P(o/w) values of the set of studied drugs, and really good correlations between the determined log P(o/w) values and those considered as reference were obtained, proving the ability of the procedure for the lipophilicity assessment of bioactive compounds with very different structures and functionalities.

Determination of the hydrophobicity of organic compounds measured as logP(o/w) through a new chromatographic method.

A new chromatographic method to determine the octanol-water partition coefficient (logP(o/w)) of organic substances is proposed in this paper. This method is based on a previously reported model that relates the retention factor in reversed-phase liquid chromatography with solute (p), mobile phase (P(m)(N)) and stationary phase (P(s)(N)) polarity parameters: logk=(logk)(0)+p(P(m)(N)-P(s)(N)). P(m)(N) values are calculated through expressions that depend only on the organic solvent fraction in the mobile phase. (logk)(0) and P(s)(N) parameters are characteristic of the chromatographic system and are determined from the retention of a selected set of 12 compounds. Then, the p value of a solute determined in a properly characterized system is easily derived from the retention factor data. Solute p values are slightly dependent on the chromatographic system but they are linearly related to those obtained in the reference system (Spherisorb ODS-2 column and acetonitrile as organic modifier). Therefore, they can be easily transferred from any experimental system to the reference one. A Quantitative Structure-Property Relationship study reveals that the p parameter in the reference chromatographic system depends, mainly, on the hydrophobicity of the compound, expressed as the n-octanol/water partition coefficient (logP(o/w)), and five additional structural descriptors which can be easily calculated through the CODESSA program from the chemical structure of the solute. In this work the p descriptors of a wide set of structurally different organic compounds have been determined in several chromatographic systems and transferred to the reference one from these and the CODESSA structural parameters. The logP(o/w) values have been determined. The obtained values agree with those determined from classical experimental techniques and validate the new method as a useful tool to determine the hydrophobicity of a wide variety of compounds in a broad logP(o/w) range.