JP5219879B2 - Method for determining logP, and two-phase solvent used in the method - Google Patents

Description

  The present invention relates to a method for determining logP and a two-phase solvent used in the method.

Log P is a logarithmic value of the distribution coefficient (P _{o / w} ) of a compound in a 1-octanol / water two-phase solvent. Log P is an index value indicating the degree of hydrophobicity or hydrophilicity of a compound, and is useful for predicting behavior such as pharmacological activity and drug absorption / distribution / metabolism / excretion (ADME) in vivo. So it is well known especially in the pharmaceutical field.

  In the field of pharmaceutics, compounds with moderate hydrophobicity (+4> logP> +1) were generally easily selected in drug screening, but recently, due to advances in DDS (drug delivery system) and formulation technology, The types of polar drugs (0> logP) are increasing. Therefore, the range of log P of the target drug is greatly expanded.

As a method for measuring log P, (1) flask-shaking to measure P _{o / w} (= C _octanol / C _water ) from the drug concentration distributed in the upper and lower phases of 1-octanol / water two-phase solvent (shake-flask) (2) reverse phase (RP) HPLC method or micellar electrokinetic chromatography (MEKC) method for indirectly measuring _{Po / w} from the retention coefficient of solute with respect to the alkyl (or micelle) stationary phase, and (3) 1- There is a high-speed counter-current chromatography (HSCCC) method in which Po _{/ w} is measured from the retention time of a solute using the upper and lower phases of an octanol / water two-phase solvent as a stationary phase / mobile phase.

  However, in any of the methods, the range in which logP can be measured accurately is narrow, and in particular, the negative logP of a water-soluble (high polarity) drug cannot be accurately determined.

  For example, in the flask shaking method (JIS Z 7260-107, 2000), which is specified as an official method of the Japanese Industrial Standard, the test substance is distributed between the upper and lower phases of 1-octanol / water two-phase solvent and then the phase is mixed. After separating and diluting as necessary, it is necessary to quantify the substance concentration in each phase. In known analytical methods and analytical instruments used for quantification, the quantitative values are affected by sensitivity and error. For example, in the UV-Vis spectrophotometric method, which is the most common concentration quantification method, the absorbance value that can be accurately measured is in the range expressed as approximately 1 to 0.01, so it is accurately measured as the concentration ratio of the upper and lower phases. The range of possible log P was generally around -2 to +2. In order to expand this measurement range, it is necessary to dilute the solution when the solution obtained as a result of two-phase partitioning is high, and when the concentration is low, the measurement scale is increased to increase the absolute amount of the substance. However, it is necessary to measure the solution after further concentration, but the work becomes complicated and errors occur.

  Further, for example, in the reverse phase HPLC method using a C8 or C18 alkyl column as a stationary phase, the test substance moves and elutes in the column while being distributed between the alkyl stationary phase and the mobile phase solvent (aqueous buffer solution or methanol mixed solution). Therefore, it is known that the mass distribution ratio k ′ calculated from the retention time of the elution peak strongly correlates with the P value. Therefore, by creating a correlation equation between k ′ and P using a standard substance with a known P value in advance, the P value of the test substance can be estimated indirectly from the actually measured k ′ without quantifying the concentration. it can. The logP measurement range by the reverse phase HPLC method is generally about −1 to +6. However, since the retention of the solute is affected by the physicochemical interaction with the stationary phase support other than the partition, the correlation accuracy is There were difficulties. In particular, it has been known that a hydrophilic compound having a dissociating group interacts with the residual silanol group of the stationary phase, and thus the reliability of log P is lowered. In addition, regarding the correlation equation of k ′ and P, the correlation coefficient is different between substances having greatly different chemical structures, and thus the accuracy of the obtained value is lowered. That is, accurate log P may not be measured even within the above measurement range. Furthermore, in the reverse phase HPLC method, a negative log P value (-1 or less) of a highly polar compound hardly retained in the alkyl stationary phase could not be measured.

  Further, for example, in the HSCCC method which is liquid-liquid partition chromatography, one liquid phase of 1-octanol / water two-phase solvent can be a stationary phase and the other liquid phase can be a mobile phase. The P value of the test substance can be measured from the retention time of the elution peak without using. The HSCCC method is also a logP measurement method that does not require concentration determination. However, since the P value of a substance whose distribution ratio is extremely biased hardly appears in the difference in retention time, the substantial logP measurement range is approximately −2 It remained in the range of about ~ + 4.

  An object of the present invention is to provide a method capable of obtaining a wide range of logP by a flask shaking method and a two-phase solvent used in the method.

The present invention
A step of preparing a two-phase solvent by mixing and leaving a polar organic solvent, an aqueous solvent and a hydrophobic organic solvent;
measuring a partition coefficient (K) of the compound having a known log P in the two-phase solvent and deriving a correlation equation from the logarithmic value of the partition coefficient (log K) and the known log P;
measuring the partition coefficient (K) of the compound for which logP is to be determined in the two-phase solvent, and calculating logP from the logarithmic value of the partition coefficient (logK) based on the correlation equation;
The method of determining logP including

  The present invention also relates to a two-phase solvent used in the logP determination method.

According to the present invention, logP, which could not be measured by the conventional flask shaking method, can be determined in a wide range (−8 to +8) by the flask shaking method.
According to the present invention, the above log P can be determined even when a small amount of the compound that is the target of determining log P is used.

It is an example of the graph which shows the correlation of logK and logP.

  The logP determination method according to the present invention is characterized by using a specific two-phase solvent, at least a two-phase solvent preparation step, and a logK-logP correlation derivation step for a compound having a known logP. , And a logP calculation step for the compound for which logP is to be determined.

In the present specification, log P is one index value indicating the degree of hydrophobicity or hydrophilicity well known in the field of pharmacology, and is a value of a compound in a “1-octanol / water” two-phase solvent. It means the logarithmic value of the distribution coefficient (P).
logK means the logarithmic value of the partition coefficient (K) of the compound in the “polar organic solvent / aqueous solvent / hydrophobic organic solvent” two-phase solvent described in detail later.

(Two-phase solvent)
The two-phase solvent used in the present invention is a mixed solvent composed of a polar organic solvent, an aqueous solvent and a hydrophobic organic solvent. In the present invention, a hydrophobic organic solvent is added to and mixed with a mixed solvent containing a polar organic solvent and an aqueous solvent, and after sufficient mixing, the mixture is allowed to stand at room temperature (25 ° C.) for two-phase separation. The solvent separated into phases is separated and used as a two-phase solvent. Two-phase separation during fractionation is sufficiently confirmed by mixing and stirring a polar organic solvent, aqueous solvent and hydrophobic organic solvent in a solvent-resistant container, and then allowing to stand for 30 minutes or longer. Make sure that this is done. In the present specification, the two-phase solvent used in the present invention may be referred to as “polar organic solvent / aqueous solvent / hydrophobic organic solvent” two-phase solvent.

  The “polar organic solvent / aqueous solvent / hydrophobic organic solvent” two-phase solvent used in the present invention is compared with the conventional “1-octanol / water” two-phase solvent in which the target compound is mixed with the upper and lower phases. It can be distributed and dissolved without causing a significant concentration difference between the phases. Therefore, the distribution coefficient (K) of the compound in the two-phase solvent of the present invention can be determined with an accurate value of an appropriate size, and as a result, sufficiently accurate log P can be measured in a very wide range.

  As the polar organic solvent, an organic solvent having a dielectric constant higher than that of the hydrophobic organic solvent and compatible with water at room temperature and normal pressure to form a homogeneous system is used. The polar organic solvent mainly constitutes the upper phase in the two-phase solvent. If the two-phase solvent does not contain a polar organic solvent and is composed only of an aqueous solvent and a hydrophobic organic solvent, the polarity difference between the upper phase and the lower phase becomes large, and a wide range of log P values cannot be obtained finally.

  As the polar organic solvent, for example, acetonitrile, acetone or a mixture thereof can be used. A preferred polar organic solvent is acetonitrile.

  As the aqueous solvent, for example, water such as pure water, ion exchange water and tap water, or an aqueous buffer solution obtained by dissolving a pH buffer in these waters is used. The aqueous solvent mainly constitutes the lower phase in the two-phase solvent.

  The aqueous solvent is preferably an aqueous buffer. By approximating the pH of the aqueous buffer to the pH of the body fluid in the living body, it is possible to obtain logP that well predicts the behavior of drug absorption, distribution, metabolism, excretion (ADME), etc. in the living body. It is.

  As the aqueous buffer, an aqueous so-called pH buffer is used. The aqueous buffer is preferably one that can maintain the pH of the lower phase at a constant value according to the purpose even when measuring logK, which will be described later. The value of the retained pH varies depending on the action site and absorption site of the test drug in the living body. However, when assuming a blood environment or a general intracellular environment, the pH of the lower phase is set to 7.0 to 7.0. A buffer solution that can be maintained at 8.0, particularly 7.4 is useful. For example, a phosphate buffer solution, an acetate buffer solution, a carbonate buffer solution, or a Tris buffer solution can be used.

  The phosphate buffer is an aqueous solution containing phosphoric acid or a salt thereof. As the phosphate, for example, sodium dihydrogen phosphate, disodium hydrogen phosphate, potassium dihydrogen phosphate, dipotassium hydrogen phosphate and the like can be used.

  The phosphate buffer solution can be prepared by adding and mixing two kinds of conjugate phosphate aqueous solutions or phosphoric acid aqueous solutions. The concentration of each aqueous solution is preferably 500 mmol / L or less, particularly preferably 20 to 200 mmol / L.

  The acetic acid buffer is an aqueous solution containing an acetate and its conjugate acid or base. As the acetate, for example, sodium acetate, potassium acetate, ammonium acetate and the like can be used. As the conjugated acid or base, acetic acid, potassium hydroxide, sodium hydroxide, ammonia or the like can be used.

  The acetic acid buffer can be prepared by adding and mixing a conjugated acid or base to an aqueous acetate solution. The concentration of the aqueous acetate solution is preferably 500 mmol / L or less, particularly preferably 20 to 200 mmol / L.

  The carbonate buffer is an aqueous solution containing carbonic acid or a salt thereof. As the carbonate, for example, sodium hydrogen carbonate, potassium hydrogen carbonate, ammonium hydrogen carbonate, sodium carbonate, potassium carbonate, ammonium carbonate and the like can be used.

  The carbonate buffer can be prepared by adding and mixing two kinds of conjugated carbonate aqueous solution or carbonic acid aqueous solution. The concentration of each aqueous solution is preferably 500 mmol / L or less, particularly preferably 20 to 200 mmol / L.

  The Tris buffer is an aqueous solution containing 2-amino-2-hydroxymethyl-1,3-propanediol and an acid. For example, hydrochloric acid or sulfuric acid can be used as the acid.

  The Tris buffer can be prepared by adding and mixing an acid or an aqueous solution thereof to an aqueous 2-amino-2-hydroxymethyl-1,3-propanediol solution. The concentration of the 2-amino-2-hydroxymethyl-1,3-propanediol aqueous solution is preferably 500 mmol / L or less, particularly preferably 20 to 200 mmol / L.

  The amount of the aqueous solvent used is not particularly limited as long as the object of the present invention is achieved, and is usually 60 to 130 parts by volume, preferably 90 to 110 parts by volume with respect to 100 parts by volume of the polar organic solvent.

  Hydrophobic organic solvents have a lower dielectric constant than polar organic solvents and do not form a homogeneous system that is incompatible with water at room temperature and normal pressure. However, organic solvents that are compatible with polar organic solvents to form a homogeneous system are used. Is done. The hydrophobic organic solvent constitutes an upper phase together with the polar organic solvent in a two-phase solvent. If the two-phase solvent does not contain a hydrophobic organic solvent, the solvent will not separate into two phases.

Examples of such hydrophobic organic solvents include saturated aliphatic monoalcohols.
As the hydrophobic saturated aliphatic monoalcohol, those having 4 or more carbon atoms, particularly 4 to 10 carbon atoms, preferably 7 to 9 carbon atoms can be used, and linear ones are more preferable. Specific examples of such preferred hydrophobic linear saturated aliphatic monoalcohols include 1-butanol, 1-amyl alcohol, 1-hexanol, 1-heptanol, 1-octanol, 1-nonyl alcohol and 1-decyl alcohol. Etc.

  The amount of the hydrophobic organic solvent used is not particularly limited as long as the object of the present invention is achieved, and is usually 4 to 40 parts by volume with respect to 100 parts by volume of the polar organic solvent. From the viewpoint of shortening the standing time for two-phase separation, the amount is preferably 10 to 20 parts by volume. If the amount of the hydrophobic organic solvent used is too small, separation into two phases becomes difficult. If the amount of the hydrophobic organic solvent used is too large, the difference in solubility in the compound between the upper phase and the lower phase will increase, so it will be remarkable when the compound is distributed between the upper phase and the lower phase. Concentration differences occur. For this reason, the actually measured log K value is too close to the actual log P value, and thus a wide range of log P cannot be measured.

In the preparation of the two-phase solvent of the present invention, the mixing time is not particularly limited as long as sufficient mixing is achieved, and usually 1 minute or more, preferably 1 to 60 minutes.
The standing time may be 1 minute or longer, and usually 1 to 60 minutes is preferable.

(Step of deriving log K-log P correlation equation of compound with known log P)
In this step, first, a distribution coefficient (K) of a compound having a known log P in the above-mentioned “polar organic solvent / aqueous solvent / hydrophobic organic solvent” two-phase solvent is measured, and the logarithmic value (log K) of the distribution coefficient is measured. )

  A compound whose log P is known is a compound that can know log P, which is considered to be an accurate value from literatures and measurements, and is measured by a conventional flask shaking method using a 1-octanol / water two-phase solvent. A compound having a log P in the range of about -2 to +4 when measured by an HSCCC method using a 1-octanol / water biphasic solvent, And compounds having a log P in the range of about −1 to +6 as measured by the reverse phase HPLC method. In particular, for a compound having a log P in the range of about −1 to +6 as measured by the reverse phase HPLC method, a compound that is not affected by the physicochemical interaction from the stationary phase support is used.

  Specific examples of the compound with known log P are not particularly limited as long as the exact log P is known, and examples thereof include the compounds described in Table 1 and Table 2 used in the examples. It is not limited to.

  From the viewpoint of deriving a correlation equation for obtaining a more accurate log P, the number of compounds for measuring log K is preferably as large as possible, usually 3 or more, preferably 20 or more, more preferably 30 to 50. It is more preferable to select the compounds listed in Table 1 and Table 2 as the compounds for deriving the correlation equation.

  As a method for measuring log K, a predetermined compound is distributed in a two-phase solvent. Specifically, for example, 1 mL of each of the upper phase solution and the lower phase solution is added to 1 mg of a sample (compound), and the mixture is sufficiently stirred and mixed in a test tube, and then allowed to stand to perform two-phase separation. Even when the compound and the two-phase solvent are used on a small scale as described above when measuring log K, accurate log P can be obtained as a result. The compound concentration must be within a concentration range such that the compound is completely dissolved in the biphasic solvent.

  Next, the sample concentrations of the upper phase solution and the lower phase solution after the separation are measured. The sample concentration is quantified for the sample solution of the upper phase solution and the sample solution of the lower phase solution. For concentration quantification, an appropriate method can be selected according to the chemical structure, physicochemical properties, and chemical reactivity of the sample compound. From the viewpoint of versatility for various compounds and ease of measurement, UV-visible Spectrophotometry is preferred. In this method, since the absorbance range of the sample solution that can be measured with high accuracy is narrow (about 0.01 to 1.0), for example, about 7 types of dilution factor (for example, 1 time, 2 times, 5 times, 10 times) with methanol. , 20 times, 50 times, and 100 times), after preparing a sample dilution solution, the wavelength at which the sample compound absorbs light may be selected to measure the absorbance.

Next, the distribution coefficient (K) is obtained from the quantified concentration quantitative value by the following calculation formula, and the logarithmic value (log K) is calculated.
K = C _UP / C _LP = (absorbance _UP × dilution factor _UP ) / (absorbance _LP × dilution factor _LP )

In the above formula, C _UP is a quantitative concentration value before dilution of the sample solution using the upper phase solution.
C _LP is a quantitative concentration value before dilution of the sample solution using the lower phase solution. The concentration may be expressed in any unit, but the units of _CUP and _CLP need to be the same.
Absorbance _UP is the absorbance (unitless) of the sample diluent using the upper phase solution, and the dilution factor of the sample diluent is the dilution factor _UP .
The absorbance _LP is the absorbance (no unit) of the sample diluent using the lower phase solution, and the dilution factor of the sample diluent is the dilution factor _LP .
Note that it is preferable to use correction values for both the absorbance _UP and the absorbance _LP . A value obtained by subtracting the absorbance value of the diluent not containing the sample from the absorbance value of the predetermined sample diluent is used.

The absorbance _UP and dilution factor _UP used in the above calculation formula are the absorbance and dilution factor of a diluted sample solution having an absorbance value in the range of 0.1 to 1.0. The absorbance _LP and dilution factor _LP used in the above calculation formula are the absorbance and dilution factor of a diluted sample solution having an absorbance value in the range of 0.1 to 1.0.

  In this specification, the absorbance is a value measured by a microplate reader (manufactured by TECAN; XSafire Mini Version). However, it does not have to be measured by the device, and the same principle and principle as the device are used. Any device can be used as long as the absorbance can be measured by the above method.

  After obtaining log K for a compound with a known log P, a correlation equation is derived from the log K and the known log P. Specifically, log K and log P of a compound with known log P are plotted on the horizontal axis; log K-vertical axis: log P coordinates, for example, as shown in FIG. To derive.

The correlation equation is derived in the form of a linear function shown below.
logP = A × logK + B
[Where A represents the slope and B represents the intercept].

  The square value of the correlation coefficient R of the correlation equation varies depending on the compound selected for deriving the correlation equation, the literature value to be referred to, and the number of compounds, but in the present invention, it normally reaches 0.9000 or more. For example, when the compounds listed in Table 1 and Table 2 are selected, the square value of the correlation coefficient R of the correlation equation is 0.9300 or more in this method.

The square value of the correlation coefficient R is a value known in the field of statistics, and the closer to 1, the stronger the positive correlation. In the present invention, the square value of the correlation coefficient R is specifically an average index representing the degree of deviation from the correlation equation of the log K-log P plot of all the compounds used for deriving the correlation equation on the graph. The value is closer to 1, and the deviation is smaller, which means that the calculated value based on the correlation equation is closer to the value (reference value) considered to be accurate.
Therefore, it is clear that the correlation formula derived in the present invention correlates well with the known logP, and logP calculated based on the correlation formula is sufficiently significant.

(Log P calculation step of the compound for which log P is to be determined)
In this step, first, the distribution coefficient (K) of the compound for which logP is determined in the above-mentioned “polar organic solvent / aqueous solvent / hydrophobic organic solvent” two-phase solvent is measured, and the logarithmic value of the distribution coefficient ( logK).

  The “compound for which logP is to be determined” refers to a compound for which logP is to be known. For example, logP may be an unknown compound, or logP is known, but the known value is inaccurate. Therefore, it may be a compound for which an accurate log P is desired.

Examples of compounds with unknown log P include compounds with extremely high hydrophilicity and compounds with extremely high hydrophobicity.
Specific examples of the compound having extremely high hydrophilicity include nucleotides, nucleosides, polysaccharides and peptide antibiotics, polysaccharides, peptides, and drug conjugates.
Specific examples of the compound having extremely high hydrophobicity include, for example, fat-soluble vitamins, fatty acids, polysensitive aromatic compounds, cyclic peptides, steroids, terpenes and the like.

  The log P is known, but the compound for which the exact log P is desired is a value where the log P is less than −2 or more than +2 when measured by a conventional flask shaking method using a 1-octanol / water two-phase solvent. , Compounds whose log P was less than −2 or greater than +4 as measured by the HSCCC method using 1-octanol / water two-phase solvent, and reversed-phase HPLC or micellar electrokinetic chromatography ( When measured by the MEKC) method, the log P is close to the measurement limit (-1 or +6) or less than -1 or exceeds +6, and the log P by the reverse phase HPLC method or the micellar electrokinetic chromatography (MEKC) method is Even within the range of -1 to +6, the stationary phase support is affected by physicochemical interactions. Compounds, and the like.

  As a method for measuring log K of such a compound, a method similar to the method for measuring log K described in the previous step can be employed.

  After obtaining log K for the compound for which log P is to be determined, log P is calculated from the log K based on the correlation equation. Specifically, logP can be obtained by substituting logK into the correlation equation.

  The logP determination method of the present invention and the two-phase solvent used therein can be employed not only in the flask shaking method but also in the HSCCC method. For example, in the HSCCC method, chromatography is performed using one phase of the two-phase solvent of the present invention as a stationary phase and the other phase as a mobile phase, and a log K in the range of −2 to +2 is obtained from the retention time of the test compound. be able to. More specifically, log K of a highly polar hydrophilic compound is measured for a short time without quantifying the concentration by reverse phase distribution mode HSCCC in which the upper phase is a stationary phase and the lower phase is a mobile phase. Can be obtained. In addition, the log K of the hydrophobic compound can be similarly determined in a short time by HSCCC in the normal phase distribution mode in which the lower phase of the two-phase solvent system is the stationary phase and the upper phase is the mobile phase. After obtaining log K, log P is calculated from the log K based on the correlation equation. Specifically, logP can be obtained by substituting logK into the correlation equation.

<Experimental example A>
(Preparation process of two-phase solvent)
First, 10% aqueous ammonia was added to and mixed with a 50 mM aqueous ammonium acetate solution to adjust the pH to 7.4, thereby preparing an acetic acid buffer.
Next, 500 mL each of acetonitrile for HPLC and ammonia-based buffer was added to a 1 L glass separatory funnel, and further 80 mL of 1-octanol was added, followed by mixing by shaking and stirring for 5 minutes. . The mixed solution was allowed to stand at room temperature (25 ° C.) for 30 minutes, and then it was confirmed that the solution was completely separated into two phases, and a two-phase solvent A was obtained. The upper phase liquid (solution) and the lower phase liquid (solution) in the two-phase solvent A were separated and stored in separate containers.

(Step of deriving log K-log P correlation equation of compound with known log P)
The compounds with known logP listed in Table 1 and Table 2 were obtained from Sigma Aldrich Japan Co., Ltd. and Wako Pure Chemical Industries, Ltd. These compounds are compounds for which accurate log P is known. For example, compounds in which accurate log P can be directly measured with a 1-octanol / water two-phase solvent by a flask shaking method or an HSCCC method, and literature Is the compound whose exact log P was revealed. In this document, logP is measured by a reverse phase HPLC method.

  For each compound, the partition coefficient (K) in the two-phase solvent A was measured by a flask shaking method, and the logarithmic value (log K) of the partition coefficient was determined. The obtained log K and known log P are shown in Table 1 and Table 2. log K is an average value when the following measurement is performed three times, and SD is a standard deviation.

In detail, the following method was implemented.
logK measurement method;
1 mL of each of the upper phase solution and the lower phase solution was added to 1 mg of the sample (compound), and the mixture was sufficiently stirred with a vortex mixer, and then allowed to stand for 5 minutes for two-phase separation. 200 μL each of the upper phase solution and the lower phase solution after separation in a 96-well plate at 7 different dilution ratios (1 ×, 2 ×, 5 ×, 10 ×, 20 ×, 50 ×, 100 ×) with methanol Using a microplate reader (XSafire Mini Version manufactured by TECAN), the absorbance was measured at detection wavelengths of 260 nm and 230 nm. Similarly, the absorbance of the upper phase solution and the lower phase solution without a sample was measured. After correcting by subtracting the absorbance of the sample-free solution from the absorbance of the sample-containing solution, the absorbance ratio (distribution coefficient K) of the upper and lower phases was determined based on the formula for calculating K, and the logarithmic value thereof was determined. The absorbance _UP and dilution factor _UP used in the calculation formula were the absorbance and dilution factor of the diluted sample solution having an absorbance value in the range of 0.1 to 1.0. The absorbance _LP and the dilution factor _LP were the absorbance and the dilution factor of the diluted sample solution having an absorbance value in the range of 0.1 to 1.0.

Method for deriving the correlation equation;
Next, as shown in FIG. 1, the log K and the known log P of the various compounds are plotted on the horizontal axis; log K-vertical axis: log P coordinates, and linear regression is performed by the least square method, as shown below. The correlation formula was derived.
log P = 4.0358 × log K + 0.1617

  The square value of the correlation coefficient R in the above correlation equation was 0.9399.

From the slope and intercept of the above correlation equation, it was found that a relatively wide range of log P values (−8 to +8) can be measured with a relatively narrow range of log K values (−2 to +2). Such a log K range can be accurately measured on a small scale by a method similar to the conventional flask shaking method or HSCCC method except that the solvent of the present invention is used.
From these results, it is clear that sufficiently accurate log P can be measured in a very wide range even when the compound for which log P is determined is used on a small scale.

In Tables 1 and 2, known values of logP indicated by a to f are values shown below.
a is an average value of log P measured three times by the same method as the measurement method of log K except that 1-octanol / water two-phase solvent was used.
b is log P described in Lombardo, F .; Shalaeva, MY; Tupper, KA; Gao, F .; Abraham, MHJ Med. Chem. 2000, 43, 2922-2928.
c is logP described in Lombardo, F .; Shalaeva, MY; Tupper, KA; Gao, FJMed. Chem. 2001, 44, 2490-2497.
d is logP described in Takacs-Novak, K .; Avdeef, AJPharm. Biomed. Anal. 1994, 12, 1369-1377.
e is logP described in Avdeef, A. Absorption and Drug Development: Solubility, Permeability, and Charge State; John Wiley & Sons, Inc .: Hoboken, NJ, 2003; pp42-66.
f is logP described in Sugano, K .; Hamada, H; Machida, M .; Ushio, HJ Biomol. Screening 2001, 6, 189-196.

<Experiment B>
(Log P calculation step of the compound for which log P is to be determined)
The compounds with unknown log P listed in Table 3 were obtained from Sigma Aldrich Japan Co., Ltd. and Wako Pure Chemical Industries, Ltd.
These compounds are classified as nucleotides or nucleosides, and are very water-soluble (hydrophilic), so it was not possible to measure logP by the conventional flask shaking method, and there are literature values. It was a compound that did not.

  For each compound, the partition coefficient (K) in the two-phase solvent A was measured by a flask shaking method, and the logarithmic value (log K) of the partition coefficient was determined. The measurement of log K was carried out by the same method as in Experimental Example A. logK is an average value when logK is measured three times.

  Based on the correlation equation, logP was calculated from logK. The obtained log K and log P are shown in Table 3.

Claims (3)

A step of preparing a two-phase solvent by mixing and leaving a polar organic solvent, an aqueous solvent and a hydrophobic organic solvent;
measuring a partition coefficient (K) of the compound having a known log P in the two-phase solvent and deriving a correlation equation from the logarithmic value of the partition coefficient (log K) and the known log P;
measuring the partition coefficient (K) of the compound for which logP is to be determined in the two-phase solvent, and calculating logP from the logarithmic value of the partition coefficient (logK) based on the correlation equation;
A method for determining logP including: The polar organic solvent is acetonitrile, acetone or a mixture thereof;
The aqueous solvent is an aqueous buffer,
The method for determining logP according to claim 1, wherein the hydrophobic organic solvent is a saturated aliphatic monoalcohol having 4 to 10 carbon atoms.   A two-phase solvent used in the method for determining logP according to claim 1 or 2.

Images