Shake-flask method

The shake-Lask method is the most commonly used method for the measurement of partition coef-Lcients (Leo et al., 1971). However, for many insoluble compounds, the solubility in the aqueous phase may be too low to be accurately determined. In these cases, an alternative method needs to be applied. 

The shake-flask method has some drawbacks, however. One problem that occurs occasionally is the formation of micro-emulsions which remain stable for hours or days, and prevent the two solvent layers from separating. In this case it is not possible to measure the concentration of analyte in each phase. More generally, the upper and lower range of log D values is limited by problems of solution handling. For very high log D values the sample is much more soluble in octanol than in water. For example, if log D is 4 the sample is 10,000 times more soluble in octa- 

Another source of high-quality measurement of log D values, though only for ion-izable drugs, is the pH-metric method. In this method, log P is calculated from the difference between the apparent pKa (p0Ka) measured in a dual-phase system, the volume ratio of the phases and the aqueous pKa. First described 50 years ago [8], this method was overlooked until the early 1990s, when it was developed at Sirius as the basis for log P measurement in the GLpKa instrument. The pH-metric method has been extensively described [9-13]. 

To use this method, the sample is dissolved in a system containing two phases (e.g., water and octanol) such that the solution is at least about 5 x 10-4 M. The solution is acidified (or basified) and titrated with base (or acid) under controlled conditions. The shape of the ensuing titration curve is compared with the shape of a simulated curve, which is created in silico. The estimated p0Ka values (together with other variables used to construct the simulated curve such as substance concentration factor, CO2 content of the solution and acidity error) are allowed to vary systematically until the simulated curve fits as closely as possible to the experimental curve. The p0Ka values required to achieve the best fit are assumed to be the correct measured p0Ka values. This computerized calculation technique is called refinement , and is described elsewhere [14, 15]. 

Equations have been published [16] which relate pKa and p0Ka values to partition coefficient (P) values for monoprotic acids and bases, and diprotic acids, bases and ampholytes. For example, P1 for a monoprotic acid is calculated from 

Lipophilicity profiles are readily calculated from pH-metric data, as pH, P and Ka are known and can be used in Eq. (1). 

The traditional shake-flask technique remains the method of reference for the direet measurement of log P ranging from 3 to - -6 with great accuracy and high precision if temperature control is maintained. In this method the pure solute is partitioned between two immiseible but mutually pre-saturated liquid phases sueh as distilled/deionized/Milli-Q water and highly pure 

1-octanol (note when the hydrophobie parameter at physiological pH is desired, the aqueous phase is usually 0.1 M phosphate buffer). The quantification of solute is carried out using various methods such as UV-based procedures, GC, and HPLC. Although this is the most popular and commonly used method for log Poet measurement, it is time consuming and requires an appreciable amount of solute in some cases. To overcome well-known limitations of this method, several other direct log Poet measurement methods have been developed, such as slow stirring and filter-probe extractor methods etc.  

A 100 mg sample of each coating was cut into small pieces, sterilised by UV light, and then dispersed in 9 ml of sterile saline water (0.85 wt%). 1 ml of bacterial (Escherichia coli or Staphylococcus aureus) culture (10 CFU/ml) was subsequently added to this solution and finally the concentration of polymer in the suspension was diluted to reach 10 mg/ml. The flasks were shaken at 90 rpm for 24 h and the temperature was maintained at 37 °C blanks without the coating were also run. The surviving bacteria before and after shaking were counted using the plate count method. 

As chronicled by Dearden [21], the association of compound lipophilicity with membrane penetration was first implied by Overton and Meyer more than a century ago. To enhance this understanding, lipophilicity measurements were initially performed using a variety of lipid phases [22], while the comprehensive review by Hansch et al. [23], with extensive data from literature and their own measurements, lent further support to the now accepted wide use of the octanol-water solvent system. 

A detailed descriphon of octanol-water distribuhon coefficient measurements by shake-flask can be found in publications by Dearden [2] and Hansch [24], The method usually involves the following solubilization of the compound in a mixture of mutually presaturated buffered water and octanol, agitation unhl equilibrium has been reached, careful separation of octanol and aqueous phases, and direct measurement of the solute concentration in both phases. Although seemingly simple, the method has a number of caveats making it inappropriate for some compounds. 

Gocan et al. [25] summarized the drawbacks of shake-flask determinations in a recent review and identified the following issues  

Some of the compUcations Usted above could be flagged and, sometimes, remedied in a manual shake-flask experiment, but that is unlikely to be the case in automated shake-vial procedures, especially if performed in a 96-well plate setting. Nevertheless, the demands of modern pharmaceutical discovery operations emphasize high-throughput measurements, low compound consumphon 

An automated log P workstation using a shake-flask method and robotic liquid handling in 96-well plate format is commercially available [30]. The system is equipped with a diode-array spectrophotometer and equimolar nitrogen detector. 

---

Distribution of benzodiazepines in I-octanol - water system was investigated by a direct shake flask method at the presence of the compounds used in HPLC mobile phases the phosphate buffer with pH 6,87 (substances (I) - (II)), acetic and phosphate buffer, perchloric acid at pH 3 (substances (III) - (VI)). Concentrations of substances in an aqueous phase after distribution controlled by HPLC (chromatograph Hewlett Packard, column Nucleosil 100-5 C, mobile phase acetonitrile - phosphate buffer solution with pH 2,5, 30 70 (v/v)). 

The octanol-water parhtion coefficient. Poet (often reported as log Poet), is a particularly useful parameter in quantitative structure-achvity relationships, apphed to predichon of properhes related to drug absorphon, distribution, metabohsm and excrehon [61, 62]. Although the traditional log Poet measurements have been done by the shake-flask method [63, 64], high-performance liquid chromatography-... 

In the traditional shake-flask method, the apparent partition coefficient, log D, is measured, usually at pH 7.4 (sometimes at pH 6.5). Different buffers are used to control each pH used in the determinations [70]. Usually, in a comprehensive study, several pH measurements are made, and values of log are plotted against the pH. This plot is often called the lipophilicity profile . One can determine the true partition coefficients (log P ) and the ionization constants from the features in such a curve. 

However, as stated above, the partition coefficients measured by the shake-flask method or by potenhometric titration can be influenced by the potenhal difference between the two phases, and are therefore apparent values which depend on the experimental condihons (phase volume ratio, nature and concentrahons of all ions in the solutions). In particular, it has been shown that the difference between the apparent and the standard log Pi depends on the phase volume raho and that this relationship itself depends on the lipophilicity of the ion [80]. In theory, the most relevant case for in vivo extrapolation is when V /V 1 as it corresponds to the phase ratio encountered by a drug as it distributes within the body. The measurement of apparent log Pi values does not allow to differentiate between ion-pairing effect and partihoning of the ions due to the Galvani potential difference, and it has been shown that the apparent lipophilicity of a number of quaternary ion drugs is not due to ion-pair partitioning as inihally thought [80]. 

In liquid-solid extraction (LSE) the analyte is extracted from the solid by a liquid, which is separated by filtration. Numerous extraction processes, representing various types and levels of energy, have been described steam distillation, simultaneous steam distillation-solvent extraction (SDE), passive hot solvent extraction, forced-flow leaching, (automated) Soxh-let extraction, shake-flask method, mechanically agitated reflux extraction, ultrasound-assisted extraction, y -ray-assisted extraction, microwave-assisted extraction (MAE), microwave-enhanced extraction (Soxwave ), microwave-assisted process (MAP ), gas-phase MAE, enhanced fluidity extraction, hot (subcritical) water extraction, supercritical fluid extraction (SFE), supercritical assisted liquid extraction, pressurised hot water extraction, enhanced solvent extraction (ESE ), solu-tion/precipitation, etc. The most successful systems are described in Sections 3.3.3-3.4.6. Other, less frequently... 

Niflumic acid, which has two pKa values, was studied both pH-metrically and spectroscopically using the shake-flask method [224]. The monoprotonated species can exist in two forms (1) zwitterion, XH 1 and (2) ordinary (uncharged) ampholyte, XH°. The ratio between the two forms (tautomeric ratio) was measured spectroscopically to be 17.4. On assuming that a negligible amount of zwitterion XH partitions into octanol, the calculated micro-log/1 for XH° was 5.1, quite a bit higher than the macro-log/1 3.9 determined pH-metrically in 0.15 M NaCl. It is noteworthy that the distribution coefficient D is the same regardless of whether the species are described with microconstants or macroconstants [275]. 

The pH-metric procedure has been validated against the standard shake-flask method [150,357], and many studies using it have been reported [56,149-151,153,161,162,224,225,229,246,250,268,269,275,276,280,281,324-363], Determinations of values of log P as low as —2 and as high as +8 have been documented [161,162,352]. The published literature clearly indicates that the Dyrssen technique is a reliable, versatile, dynamic, and accurate method for measuring logP. It may lack the speed of HPLC methods, and it cannot go as low in log P as the CV... 

Aqueous solubility values for the samples analyzed compared favorably with results obtained by traditional methods. The solubility values for amiodarone HC1, reserpine, and benzanthrone were lower than the LOQ of the, uPLC system used for the evaluation. Results of the evaluation of compound solubility employing no-filtration /iPLC were compared with those obtained by two traditional methods (1) multiscreen filtration followed by a UV plate reader, and (2) the shake flask method followed by a UV plate reader. As shown in Figure 6.31, the solubility values determined by the different methods are comparable for most compounds examined. Figure 6.32 shows the results of evaluations of aqueous solubility at four different pH levels for phenazopyridine and piroxicam samples. 

FIGURE 6.31 Comparison of solubility results obtained by no-filtration method followed by PLC detection and those obtained by two traditional methods multiscreen filtration followed by UV plate reader, and shake flask method followed by UV plate reader. (Data provided by Steven Hobbs, Courtney Coyne, and Gregory Kazan.)... 

The logarithm for the capacity factor correlates well with known log P values obtained by the shake flask method. In practice, the k values are determined isocratically from 70 to 30% organic mobile phase and then extrapolated to 0%. Prior to determining the log P for an unknown compound, a set of structurally related molecules (standards) are analyzed to construct a correlation model between the logarithm of the retention factor and known log P values. The process is then repeated for the test compounds and their log P values determined from the mathematical relationship established for the standard compounds. 

---