HPLC Methods for Lipophilicity Determination

The partition coefficient logP or the distribution coefficient logD between water and n-octanol are determined traditionally by the shake flask method (see chapter about physicochemical parameters). The shake flask method is a rather labour intensive method and not suited for high throughput analysis needed in pharmaceutical research. 

The use of hydrophobic stationary phases and aqueous mobile phases in chromatography on so-called reversed phase (RP) materials is widely established. Reversed phase materials typically consist of alkyl chains (mainly C18 in length) immobilised on silica material. The chromatographic retention of a solute on such a system directly depends on its partition between the aqueous mobile and the hydrophobic solid phase (Berthod et al. 2004). Therefore lipophilicity (expressed as partition coefficient logP between stationary and mobile phase, can be estimated from the retention behaviour of the solute of interest on reversed phase material as reviewed by Nasal et al. (2003). 

Vm and Vs are the volumes of the stationary and mobile phase respectively k is the retention factor that is derived from the retention time (tR) taking the dead time 

The lipophilicity of compounds can be measured with any commercially available HPLC equipment most commonly UV absorption is used as detection system for the chromatography. The retention factor (retention time) of a compound is the analytical parameter needed to measure lipophilicity. To measure the retention factor, the retention time of the compound under investigation has to be determined reproducibly together with the dead time. The dead time is determined using the retention time of a not retained substance on reversed phase columns like buffer salts. 

As mobile phase, different buffers at different pH values (mainly 6.8 or 7.4) with organic modifiers are used. Results obtained with methanol as co-solvent were shown by Baczek et al. (2000) to correlate better to octanol-water partitioning data than results using acetonitrile as organic co-solvent. The H-bonding capabilities of the alcohol seem to be the reason for that. 

---

Du, C. M., Valko, K., Bevan, C., Reynolds, D., Abraham, M. H. Rapid gradient RP-HPLC method for lipophilicity determination a solvation equation based comparison with isocratic methods. Anal. Chem. 1998, 70, 4228-4234. 

DeLeenheer et al. (1991) contributed a review on the TLC of lipophilic vitamins. They noted that although the 1980s witnessed a proliferation of HPLC methods for the determination of fat-soluble vitamins, the advent of newer HPTLC-densitometric techniques may allow for a resurgence of interest in the 1990s for the analysis of lipophilic vitamins by TLC. Fried (1991) reviewed the literature on the TLC of hydrophilic vitamins from about 1965 to 1988. 

The great advantage of chromatography for lipophilicity determination is its ability to estimate log P or log D with a relative accuracy for non-pure samples, with a limited quantity of compound. Furthermore, these methods are relatively easily to automate and they only require an HPLC or UPLC apparatus. Many approaches have been described and this topic was recently reviewed [5, 27]. That is why the focus of the following section is on strategies developed to increase the throughput and the assay dynamic range. 

Krass et al. [45] have also reported a novel method for the determination of lipophilicity using a simple HPLC protocol based on gradient elution chromatography. They compared the gradient retention times with the traditional isocratic logkw param-... 

Technical details, including solid support, coating and column filling techniques, eluents, and factors affecting the reproducibility have been reviewed [173, 235, 236]. Lipophilicity values from HPLC measurements are not on a unique scale, but the log k values can be converted to -octanol/water partition coefficients with the help of a set of HPLC calibration standard compounds for which classical shake-flask partition coefficients are known. Experience shows once the HPLC method is successfully established in a laboratory, it will remain the method of choice for lipophilicity determinations. 

Mirrlees MS, Moulton SJ, Murphy CT et al. (1976) Direct measurement of octanol-water partition coefficients by high-pressure liquid chromatography. J Med Chem 19 615-619 Pagliara A, Khamis E, Thrinh A et al. (1995) Structural properties governing retention mechanisms on RP-HPLC stationary phases used for lipophilicity measurements. J Liquid Chromatography 18 1721-1745 Slater B, McCormack A, Avdeef A et al. (1994) pH-Metric log P. 4. Comparison of Partition Coefficients Determined by Shake-Flask, HPLC and Potentiometric Methods. J Phar-maceut Sci 83 1280-1283... 

With the slow-stirring method a precise and accurate determination of Kow of compounds with log Kow up till 8.2 is allowed (De Bruijn et al, 1989). For highly lipophilic compounds the shake-flask method is prone to produce artefacts (formation of microdroplets), and with the HPLC method Kow needs to be extrapolated beyond the calibration range to obtain estimates of Kow-... 

Most improved or new methods for OA group toxins that have been reported in the last few years involve the nse of HPLC-mass spectrometry (LC-MS). One reason that LC-MS methods have been so popular is the fact that a single LC-MS run can determine all the lipophilic toxins currently required by most official agencies, including OA group toxins. The topic of general toxin methods has been reviewed elsewhere (Quilliam, 2003). LC-MS has also become popular as electrospray ionization (ESI) sources have become more robust and more LC-MS systems are now in... 

In a recent contribution of ours, the sorption of selected imidazolium ionic liquid cations with different alkyl side chain lengths (from C3 to Cg) was determined and compared [16]. This enabled us to follow the changes in sorption behavior due only to the variation in lipophilicity of the cations under scrutiny. Three types of soils differing in their physical and chemical properties were chosen for this experiment. A batch-equihbrium technique was employed to determine the sorption of the ionic liquid cations. The experiments were performed according to OECD guidelines [17]. The determinations of the residual concentrations of ionic liquids were performed using HPLC methods developed by our group [18-21]. 

After hydrolysis of triglycerides, denaturation (ethanol) and extraction (hexane) of milk samples are similar to those procedures used in preparation of plasma samples. EHie to high concentrations of coextracted lipophilic compounds, a semipreparative HPLC cleanup step often is used in these preparations. Indyk et al. (92) and Lambert et al. (99) used adsorption HPLC and hexane/isopropanol mixtures for cleanup and a reversed-phase HPLC for detection, whereas Isshiki et al. (94) used a Cig reversed-phase system with a methanol/acetonitrile mixture for cleanup and a C2 or C3 reversed phase for detection. Canfield et al. (95,96) worked with two open-column chromatography systems (silica) to isolate vitamin K compounds prior to HPLC detection, whereas Schneiderman et al. (98) introduced a completely different method for isolation of these compounds. They used supercritical fluid extraction (SFE) with CO2 as solvent to determine VKl in powdered infant formulas. Details of this particular method will be discussed later in this chapter. 

The described method is a typical HPLC fluorescence assay for drug level determination for toxicokinetic purposes. However, the conditions of sample extraction, the choice of the internal standard substance, the choice of the HPLC stationary and mobile phase and the combination of excitation and emission wavelength has to be adjusted specifically to the properties of the analytes. Particularly, the lipophilicity, the pKa value and the pH stability of the analytes have to be considered. 

---