FIELD OF THE INVENTION

[0001] The invention relates to a method of separating the stereoisomers of N- phthaloyl -glutamic acid. A mixture of the L stereoisomer and the D stereoisomer of N-phthaloyl -glutamic acid can be purified by separating the stereoisomers from one another.

BACKGROUND

[0002] N-Phthaloyl-L-glutamic acid (CAS: 340-90-9) is a reagent that plays a central role in the synthesis of products in the class of gamma-glutamyl-4- azoanilides, which are substrates for the enzyme gamma-glutamyl transferase (GGT).

[0003] GGT is critical to amino acid transport in kidneys, as well as cellular detoxification and leukotriene biosynthesis processes. It has been implicated as being involved in many important physiological disorders, such as Parkinson’s disease and inhibition of apoptosis. GGT binds glutathione as a donor substrate and initially forms a gamma-glutamyl enzyme, which can then react with a water molecule to form glutamate or with an acceptor substrate, such as an amino acid or a dipeptide, to form a product containing new gamma-glutamyl isopeptide bonds, and hence regenerate the enzyme.

[0004] The activity measurement of gamma-glutamyl transferase (GGT) is known to play an essential role in clinical diagnostics as liver activity marker. The GGT enzyme only accepts the L-configuration of the glutamic acid residue when coupled to a dye, as in gamma-glutamyl-4-azoanilides.

[0005] The stereocenter is introduced and controlled when synthesising the product N-phthaloyl-L-glutamic acid. After this synthetic step, the stereocenter is no longer prone to isomerization and therefore only needs to be controlled in the preceding synthetic step.

[0006] The handling of the stereocenter, and its respective control and measurement, are influenced by the synthetic process used for preparing the products in the class of gamma-glutamyl-4-azoanilides. Quality control is normally performed by measuring the optical rotation of this product. The optical rotation that is measured is heavily dependent on the concentration of the solution, its temperature and the solvent that is used, as well as the purity of the product. It is therefore difficult to obtain reproducible values for the optical rotation when preparing batches of the product.

SUMMARY OF THE INVENTION

[0007] The invention provides a method of separating stereoisomers of N- phthaloyl -glutamic acid. The method comprises: passing a solution comprising an L/D stereoisomer mixture of N-phthaloyl- glutamic acid through a chiral chromatography column by normal phase chromatography, wherein the chiral chromatography column comprises, as a stationary phase, silica having groups represented by formula (1) covalently bonded to a surface thereof:

L - polysaccharide

(1) wherein:

L is a linker group covalently bonded to a surface of the silica, wherein the linker group has a length of no more than 8 atoms, and polysaccharide is represented by one of formulas (2), (3) or (4): each R ² is independently selected from H and a covalent bond to a linker group L, and wherein at least one R ² is the covalent bond to the linker group L; and each R ^1A , R ^2A and R ^3A is independently selected from H, methyl and chloro, provided that 3 of any R ^1A , R ^2A and R ^3A are H and 2 of any R ^1A , R ^2A and R ^3A are methyl and/or chloro; and wherein the dashed line in R ¹ indicates a covalent bond to an oxygen atom of the polysaccharide. Preferably, the method of separating stereoisomers of N- phthaloyl -glutamic acid, the method comprising: passing a solution comprising an L/D stereoisomer mixture of N-phthaloyl- glutamic acid through a chiral chromatography column by normal phase chromatography, wherein the chiral chromatography column comprises, as a stationary phase, silica having groups represented by formula (1) covalently bonded to a surface thereof:

L - polysaccharide

(1) wherein: L is a linker group covalently bonded to a surface of the silica, wherein the linker group has a length of no more than 8 atoms, and polysaccharide is represented by one of formulas (2), (3) or (4): each R ² is independently selected from H and a covalent bond to a linker group L, and wherein at least one R ² is the covalent bond to the linker group L; and each R ^1A , R ^2A and R ^3A is independently selected from H, methyl and chloro, provided that 3 of any R ^1A , R ^2A and R ^3A are H and 2 of any R ^1A , R ^2A and R ^3A are methyl and/or chloro; wherein the dashed line in R ¹ indicates a covalent bond to an oxygen atom of the polysaccharide, wherein the normal phase chromatography comprises eluting the solution with a non-polar solvent having a partition coefficient ( ) that satisfies log? > 0.5, wherein the non-polar solvent comprises heptane or hexane and heptane, and wherein the linker group L is selected from an oxyalkylene group having 2 to 8 atoms and a Ci-salkylene group, wherein the oxyalkylene group or the Ci-salkylene group may be substituted by one or more groups selected from OR ^C , Ci- ₆ alkyl, -OC(O)OR ^C ,-OC(O)NR ^C ₂ , -NHC(O)NR ^C ₂ and -NHC(O)OR ^c , and each R ^c is independently selected from H and Ci-6 alkyl.

[0008] The vast number of current materials for chiral HPLC rely on chiral adsorbent materials that do not provide sufficient separation of the L- and D- stereoisomers of N-phthaloyl -glutamic acid. The immobilized adsorbent materials are unable to obtain a high-resolution chiral separation of the L- and D- stereoi somers, specifically it has not been possible to achieve a baseline separation of the stereoisomers. This can be problematic from both an analytical and preparative standpoint. When using chiral chromatography analytically, the absence of a chiral separation means that it is not possible to accurately determine the purity of the stereoisomer mixture. Similarly, in preparative scale chromatography, it is not possible to fully purify the stereoisomers because of the incomplete baseline separation between them.

[0009] The inventors have unexpectedly found a chiral, active material for use as a stationary phase in chromatography, which can achieve excellent separation of the stereoisomers of N-phthaloyl -glutamic acid. In particular, it is possible to use this material to determine the purity of N-phthaloyl-L-glutamic acid in a mixture that may contain the L- and D-stereoisomers of N-phthaloyl -glutamic acid.

[0010] One aspect of the invention provides a method of determining a purity of N-phthaloyl-L-glutamic acid in a solution comprising N-phthaloyl -glutamic acid. The method comprises: passing a solution comprising N-phthaloyl-glutamic acid through a chiral chromatography column by normal phase chromatography, wherein the chiral chromatography column comprises, as a stationary phase, silica as defined herein (e.g. the silica has groups represented by formula (1) covalently bonded to a surface thereof as described above). [0011] Another aspect of the invention provides a method of purifying N- phthaloyl-L-glutamic acid from a solution comprising an L/D stereoisomer mixture of N-phthaloyl -glutamic acid. The method comprises: passing a solution comprising an L/D stereoisomer mixture of N-phthaloyl -glutamic acid through a chiral chromatography column by normal phase chromatography, wherein the chiral chromatography column comprises, as a stationary phase, silica as defined herein (e.g. the silica has groups represented by formula (1) covalently bonded to a surface thereof as described above).

[0012] Preferred, suitable, and optional features of any one particular aspect of the invention described below are also preferred, suitable, and optional features of any other aspect.

BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The invention is further described hereinafter with reference to the accompanying drawings.

[0014] Figures 1 and 2 show HPLC spectra obtained from solutions comprising N- phthaloyl -glutamic acid. The spectra were obtained in accordance with the method of the invention. The x-axes represent the time in minutes and the y-axes represents absorbance units (AU).

[0015] Figure 3 shows a HPLC spectrum obtained from a solution comprising N- phthaloyl -glutamic acid. The spectrum was not obtained in accordance with the method of the invention. The x-axis represents the time in minutes and the y-axis represents absorbance units (AU).

DEFINITIONS

[0016] The term “N-phthaloyl -glutamic acid” as used herein refers to a compound having the following structure.

The stereochemical configuration at the atom marked by the asterisk is undefined. A mixture or a solution of N-phthaloyl-glutamic acid may comprise N-phthaloyl-L- glutamic acid and/or N-phthaloyl-D-glutamic acid.

[0017] The expression “L/D stereoisomer mixture of N-phthaloyl-glutamic acid” as used herein refers to a mixture comprising N-phthaloyl-L-glutamic acid and N- phthaloyl-D-glutamic acid (e.g. a mixture comprising both the L- and D- stereoisomers of N-phthaloyl-glutamic acid). The mixture may have an enantiomeric excess of from 0% to 100% of either the L-stereoisomer or the D- stereoisomer, preferably an enantiomeric excess of from 0% to less than 100%, such as from 0% to 99%. Generally, it is preferred that the mixture has an enantiomeric excess of from 0% to 100% of the L-stereoisomer, preferably an enantiomeric excess from 0% to less than 100% of the L-stereoisomer, such as from 0% to 99% of the L- stereoisomer.

[0018] The term “N-phthaloyl-L-glutamic acid” or any reference to the “L- stereoisomer” of N-phthaloyl-glutamic acid refers to the following compound having the stereochemical configuration that is shown. The L-stereoisomer has the S- configuration, as denoted using the Cahn-Ingold-Prelog (CIP) priority rules.

[0019] The term “N-phthaloyl-D-glutamic acid” or any reference to the “D- stereoisomer” of N-phthaloyl-glutamic acid refers to the following compound having the stereochemical configuration that is shown. The D-stereoisomer has the Reconfiguration, as denoted using the Cahn-Ingold-Prelog (CIP) priority rules.

[0020] The term “normal phase chromatography” as used herein has its normal meaning in the art and includes normal phase high-performance liquid chromatography (HPLC). Normal phase chromatography refers to a chromatographic method where the separation of analytes is based on their affinity for a stationary phase, such as silica, which typically has a polar surface. Normal phase chromatography involves the use of a polar stationary phase and a non-polar mobile phase for the separation of polar compounds (i.e. the analytes), in this case the stereoisomers of N-phthaloyl -glutamic acid. Separation occurs because of the differing ability of the stereoisomers to form polar, diastereomeric interactions (e.g. hydrogen-bonding or dipole-dipole interactions) with the polar, chiral surface of the stationary phase. The use of more polar solvents in the mobile phase will decrease the retention time of analytes, whereas more hydrophobic solvents tend to induce slower elution (e.g. increase retention times).

[0021] The term “stationary phase” as used herein has its conventional meaning in the field of column chromatography, particularly HPLC. The stationary phase is the part of a column that interacts with the analyte, in this case the stereoisomers of N- phthaloyl -glutamic acid.

[0022] The stationary phase in the present invention is an inorganic carrier having groups represented by formula (1) covalently bonded to a surface of the inorganic carrier. The inorganic carrier may include silica, alumina, magnesia, glass, kaolin, titanium oxide, a silicate or a hydroxyapatite. Silica is the preferred inorganic carrier and the present disclosure describes the invention in the context of silica being the inorganic carrier. However, it is to be understood that any reference to “silica” below may refer to the inorganic carrier, particularly an inorganic carrier selected from alumina, magnesia, glass, kaolin, titanium oxide, a silicate and a hydroxyapatite, unless the context indicates otherwise. It is preferred that the stationary phase includes silica, i.e. silica having the groups represented by formula (1) covalently bonded to a surface of the silica. The groups represented by formula (1) are immobilised on the surface of the silica by either being directly covalently bonded to the silica surface (e.g. when L, the linker group, has 0 atoms) or by being covalently bonded to a relatively short linker group, represented by L, which is itself covalently bonded to a surface of the silica.

[0023] The various hydrocarbon-containing moi eties provided herein may be described using a prefix designating the minimum and maximum number of carbon atoms in the moiety, e.g. “Ca-b” or “Ca-Cb”. For example, Ca-balkyl indicates an alkyl moiety having the integer “a” to the integer “b” number of carbon atoms, inclusive. Certain moieties may also be described according to the minimum and maximum number of members with or without specific reference to a particular atom or overall structure. For example, the terms “a to b membered ring” or “having between a to b members” refer to a moiety having the integer “a” to the integer “b” number of atoms, inclusive.

[0024] The term “alkyl” or “alkyl group” as used herein, by itself or in conjunction with another term or terms, refers to a branched or unbranched saturated hydrocarbon chain. Unless specified otherwise, an alkyl group typically contains 1-6 carbon atoms, such as 1-4 carbon atoms or 1-3 carbon atoms, and can be substituted or unsubstituted. Representative examples include, but are not limited to, methyl, ethyl, n-propyl, i-propyl, n-butyl, i-butyl, s-butyl, t-butyl, n-pentyl, n-hexyl, isopropyl, tertbutyl, isobutyl, etc. The alkyl group is unsubstituted, unless the context indicates otherwise.

[0025] The term “alkylene” or “alkylene group” as used herein, by itself or in conjunction with another term or terms, refers to a linear (e.g. an unbranched) saturated hydrocarbon chain. Unless specified otherwise, alkylene groups typically contain 1-8 carbon atoms, such as 1-6 carbon atoms or 1-3 carbon atoms, and can be substituted or unsubstituted. Representative examples include, but are not limited to, methylene (-CH2-), ethylene (-CH2CH2-), and propylene (-CH2CH2CH2-). The alkylene group is unsubstituted, unless the context indicates otherwise.

[0026] The term “oxyalkylene” or “oxyalkylene group” as used herein, by itself or in conjunction with another term or terms, refers to a linear (e.g. unbranched) saturated hydrocarbon chain containing one or more oxy groups, preferably one or two oxy groups, which are interspersed within the hydrocarbon chain. Unless specified otherwise, oxyalkylene groups contain 2 to 8 atoms, such as 2 to 7 atoms or 2 to 6 atoms, and can be substituted or unsubstituted. Representative examples include, but are not limited to, oxyethylene (-OCH2CH2-), oxypropylene (-OCH2CH2CH2-), methyleneoxyethylene (-CH2OCH2CH2-), ethyleneoxyethylene (-CH2CH2OCH2CH2-), ethyleneoxypropylene (-CH2CH2OCH2CH2CH2-), propyleneoxy-ethylene (-CH2CH2CH2OCH2CH2-), propyleneoxy propylene

(-CH2CH2CH2OCH2CH2CH2-). The alkylene group is unsubstituted, unless the context indicates otherwise.

[0027] The term “about” when used herein in conjunction with a measurable value such as, for example, an amount or a period of time and the like, is meant to encompass reasonable variations of the value, for instance, to allow for experimental error in the measurement of said value.

[0028] The term “comprises” as used herein has an open meaning, which allows other, unspecified features to be present. This term embraces, but is not limited to, the semi-closed term “consisting essentially of’ and the closed term “consisting of’. Unless the context indicates otherwise, the term “comprises” may be replaced with either “consisting essentially of’ or “consists of’. The term “consisting essentially of’ may also be replaced with “consists of’.

DETAILED DESCRIPTION

[0029] The method of the invention involves passing a solution comprising N- phthaloyl -glutamic acid, preferably a solution comprising an L/D stereoisomer mixture of N-phthaloyl -glutamic acid, through a chiral chromatography column by normal phase chromatography. [0030] As a stationary phase, the chiral chromatography column comprises silica having groups represented by formula (1) covalently bonded to a surface thereof:

L - polysaccharide

(1).

[0031] It has surprisingly been found that the modified polysaccharide of the invention is able to form diastereomeric interactions with the stereoisomers of N- phthaloyl -glutamic acid, which provide good separation of the enantiomers. The modified polysaccharide mimics the hydrophobic nature of the product, which is further reflected in the normal phase behaviour of the solvent choice for performing chromatographic method.

[0032] The polysaccharide comprises a repeating unit represented by one of formulas (2), (3) or (4):

[0033] In each of formulas (2), (3) or (4), n is an integer. The value of n represents the number of repeating units of formula (2), (3) or (4) in the polysaccharide.

[0034] Typically, n is at least 3, such as at least 5, preferably at least 10.

[0035] In general, the polysaccharide has a number-average degree of polymerization of 5 or more, preferably 10 or more, such as 25 or more, and more preferably 50 or more. [0036] The polysaccharide may have a number-average degree of polymerization of 1,500 or less, preferably 1,000 or less, such as 750 or less, and more preferably 400 or less.

[0037] Each of the lower limits for the number-average degree of polymerization may be combined with each of the upper limits for the number-average degree of polymerization.

[0038] It is preferred that the polysaccharide has a number-average degree of polymerization of from about 5 to about 1,500, such as from about 10 to about 1,000, more preferably from about 25 to about 750, and even more preferably from about 50 to about 400.

[0039] The number-average degree of polymerization can be determined using conventional techniques known in the art, such as by measuring the Rayleigh light scattering of the polysaccharide.

[0040] The polysaccharide may be represented by formula (2). This polysaccharide is a cellulose-derived polysaccharide. The polysaccharide may be represented by formula (3). This polysaccharide is an amylose-derived polysaccharide. The polysaccharide may be represented by formula (4). This polysaccharide is an amylose-derived polysaccharide.

[0041] It is preferred that the polysaccharide is a polysaccharide represented by formula (2) or a polysaccharide represented by formula (3). More preferably, the polysaccharide is a polysaccharide represented by formula (2).

[0042] The polysaccharide includes a group represented by R ¹ , which is shown below.

[0043] The dashed line in R ¹ indicates a covalent bond to an oxygen atom of the polysaccharide. Thus, for example, when the polysaccharide is represented by formula (2) and the structure of R ¹ is shown, then the polysaccharide may be represented by formula (2’).

[0044] In R ¹ , each R ^1A , R ^2A and R ^3A is independently selected from H, methyl and chloro, provided that 3 of any R ^1A , R ^2A and R ^3A are H and 2 of any R ^1A , R ^2A and R ^3A are methyl and/or chloro. To achieve a baseline separation of the stereoisomers, it has been found that there must be two substituents on the phenyl group in R ¹ . For the avoidance of doubt, R ¹ is the same (e.g. has the same composition) in each repeating unit of the polysaccharide.

[0045] The phenyl group in R ¹ may be selected from one of the following structures, where the dashed line represents the bond to the amino group in R ¹ . The forward slash in “Me/Q” indicates that the group is Me or Cl.

2 x R ^1A = Me or Cl; _R 2A _R 3A _{= Me or} 2 x R ^2A = Me or Cl; and Cl; and and

2 x R ^2A and R ^3A = H. 2 x R ^1A and R ^2A = 2 x R ^1A and R ^3A = H.

H.

[0046] It is preferred that 3 of any R ^1A , R ^2A and R ^3A are H and 2 of any R ^1A , R ^2A and R ^3A are both methyl or both chloro. More preferably, both R ^2A are chloro or methyl. Thus, R ¹ has formula (5F) or formula (5G).

[0047] R ¹ may have formula (5F). R ¹ may have formula (5G).

[0048] In formulas (2), (3) and (4), X is O or NH. It may be preferably that X is NH. Alternatively, it may be preferably that X is O.

[0049] When the polysaccharide is represented by formula (2), then the polysaccharide may comprise a terminal group represented by formula (2a) and/or (2b). (2a) (2b)

[0050] In formulas (2a) and (2b), the dashed line represents the covalent bond to a repeating unit of the polysaccharide. In the repeating unit represented by formula (2), the dashed line is the covalent bond that passes through one of the parentheses shown in formula (2).

[0051] In formulas (2a) and (2b), R ¹ , R ² and X are as defined herein.

[0052] In formula (2a), R ¹¹ is selected from H and R ¹ . When R ¹¹ is R ¹ , then R ¹ is as defined herein. R ¹ is the same (e.g. has the same composition) as in the repeating unit of the polysaccharide (e.g. the same as R ¹ in formula (2)). It is preferred that R ¹¹ is R ¹ .

[0053] In formula (2b), R ¹² is selected from H and R ¹ . When R ¹² is R ¹ , then R ¹ is as defined herein. R ¹ is the same (e.g. has the same composition) as in the repeating unit of the polysaccharide (e.g. the same as R ¹ in formula (2)). It is preferred that R ¹² is R ¹ .

[0054] When the polysaccharide is represented by formula (3), then the polysaccharide may comprise a terminal group represented by formula (3a) and/or (3b).

[0055] In formulas (3a) and (3b), the dashed line represents the covalent bond to a repeating unit of the polysaccharide. In the repeating unit represented by formula (3), the dashed line is the covalent bond that passes through one of the parentheses shown in formula (3).

[0056] In formulas (3a) and (3b), R ¹ , R ² and X are as defined herein. [0057] In formula (3a), R ¹³ is selected from H and R ¹ . When R ¹³ is R ¹ , then R ¹ is as defined herein. R ¹ is the same (e.g. has the same composition) as in the repeating unit of the polysaccharide (e.g. the same as R ¹ in formula (3)). It is preferred that R^ is R ¹ . [0058] In formula (3b), R ¹⁴ is selected from H and R ¹ . When R ¹⁴ is R ¹ , then R ¹ is as defined herein. R ¹ is the same (e.g. has the same composition) as in the repeating unit of the polysaccharide (e.g. the same as R ¹ in formula (3)). It is preferred that R ¹⁴ is R ¹ .

[0059] When the polysaccharide is represented by formula (4), then the polysaccharide may comprise a terminal group represented by formula (4a) and/or (4b).

[0060] In formulas (4a) and (4b), the dashed line represents the covalent bond to a repeating unit of the polysaccharide. In the repeating unit represented by formula (4), the dashed line is the covalent bond that passes through one of the parentheses shown in formula (4).

[0061] In formulas (4a) and (4b), R ¹ , R ² and X are as defined herein. [0062] In formula (4a), R ¹⁵ is selected from H and R ¹ . When R ¹⁵ is R ¹ , then R ¹ is as defined herein. R ¹ is the same (e.g. has the same composition) as in the repeating unit of the polysaccharide (e.g. the same as R ¹ in formula (4)). It is preferred that R^ is R ¹ .

[0063] In formula (4b), R ¹⁶ is selected from H and R ¹ . When R ¹⁶ is R ¹ , then R ¹ is as defined herein. R ¹ is the same (e.g. has the same composition) as in the repeating unit of the polysaccharide (e.g. the same as R ¹ in formula (4)). It is preferred that R ¹⁶ is R ¹ .

[0064] In formulas (2), (3) and (4) (as well as formulas (2a), (2b), (3a), (3b), (4a) and (4b)), each R ² is independently selected from H and a covalent bond to a linker group L, and wherein at least one R ² is the covalent bond to the linker group L. R ² represents the point of attachment of the polysaccharide to the silica either directly or by the linker group represented by L.

[0065] At least one of the repeating units of the polysaccharide is bonded to the silica. Within each polysaccharide molecule, the polysaccharide may be bonded to the silica through R ² in one or more repeating units.

[0066] The nature of the specific bonding moiety that links the polysaccharide to the surface of the silica is not important. In principle, the polysaccharide can be bonded to the surface of the silica in any manner provided it is immobilized on the surface of the silica. Methods are known in the art for immobilizing compounds on the surface of silica.

[0067] The linker group (or the polysaccharide when the linker group has a length of 0 atoms) is covalently bonded to a surface of the silica by a silicon-containing group.

[0068] The silicon-containing group may, for example, be any silicon-containing group shown in (Bl) to (B3) or (Cl) to (C3) below.

(Cl) (C2) (C3)

[0069] The wavy line represents the surface of the silica and the dashed line represents the covalent bond to the linker group or to X when the linker group has a length of 0 atoms and R ² is a covalent bond.

[0070] The silicon-containing group may be selected from (Bl) and (Cl).

[0071] L represents a linker group covalently bonded to a surface of the silica. The linker group has a length of no more than 8 atoms.

[0072] The purpose of the linker group, L, is to immobilise the polysaccharide moiety on a surface of the silica. The linker group, L, should be relatively short to restrict the freedom of movement of the polysaccharide moiety. This is to ensure that the stereoisomers of N-phthaloyl -glutamic acid primarily interact with the polysaccharide moieties immobilised on the surface of the silica, rather than with the silica surface itself.

[0073] In principle, the chemical structure of the linker group L is not limited provided that it can covalently immobilise the polysaccharide moiety on a surface of the silica. In other words, the chemical composition of L does not affect the chromatographic separation of the stereoisomers of N-phthaloyl -glutamic acid because these stereoisomers predominantly interact with the polysaccharide moiety.

[0074] The linker group, L, has a length of no more than 8 atoms, preferably a length of no more than 7 atoms, more preferably a length of no more than 6, and even more preferably a length of no more than 5 atoms. [0075] L may have a length of from 0 to 8 atoms. When L has a length of 0 atoms, then the linker group is absent.

[0076] The linker group may have a length of 1, 2, 3, 4, 5, 6, 7 or 8 atoms. It is preferred that the linker group has a length of 1 to 8 atoms, such as 1 to 7 atoms, more preferably 2 to 7 atoms, and even more preferably 3 to 6 atoms.

[0077] The length of the linker group, L, refers to the length of the group that directly connects the polysaccharide, particularly a repeating unit of the polysaccharide, to a surface of the silica. The length of the linker group, L, does not include the silicon-containing group, such as the silicon-containing group on a surface of the silica as described herein. The length of the linker group also does not include substituents on the linker group that do not form part of the direct connection between the polysaccharide and the surface of the silica.

[0078] In one embodiment, the linker group, L, may be selected from an oxyalkylene group having 2 to 8 atoms and a Ci-salkylene group.

[0079] The oxyalkylene group or the Ci-salkylene group may be substituted by one or more groups selected from -OR ^C , Ci-ealkyl, -OC(O)OR ^c ,-OC(O)NR ^C 2, - NHC(O)NR ^C 2 and -NHC(O)OR ^c . The oxyalkylene group or the Ci-salkylene group may preferably be substituted by one group selected from -OR ^C , Ci-ealkyl, -OC(O)OR ^c , -OC(O)NR ^c ₂ , -NHC(O)NR ^c ₂ and -NHC(O)OR ^c .

[0080] Each R ^c is independently selected from H and Ci-ealkyl. It is preferred that R ^c is H.

[0081] When L is an oxyalkylene group, then it is preferred that the group selected from -OR ^C , Ci-ealkyl, -OC(O)OR ^c ,-OC(O)NR ^c ₂ , -NHC(O)NR ^c ₂ and -NHC(O)OR ^c is not bonded to a carbon atom of the oxyalkylene group that is directly adjacent to an oxy group of the oxyalkylene group.

[0082] The groups selected from -OR ^C , Ci-ealkyl, -OC(O)OR ^c , -OC(O)NR ^C 2, - NHC(O)NR ^C 2 and -NHC(O)OR ^c do not form part of the length of the linker group.

[0083] The oxyalkylene group or the Ci-salkylene group may be unsubstituted. Alternatively, it is preferred that the oxyalkylene group or the Ci-salkylene group is substituted by one or more groups, preferably one group, selected from -OH, Ci-ealkyl and -OC(O)NHR ^c . More preferably, the oxyalkylene group or the Ci- salkylene group is substituted by one or more groups, preferably one group, which is -OC(O)NHR ^c .

[0084] When L is an oxyalkylene group, it is preferred that a carbon atom of the oxyalkylene group is covalently bonded to the polysaccharide, particularly a repeating unit of the polysaccharide.

[0085] It is preferred that the oxyalkylene group is selected from oxyethylene, oxypropylene, ethyleneoxyethylene, ethyleneoxy propylene, propyleneoxyethylene and propyleneoxy-propylene. The length of each group is 3, 4, 5, 6, 6 and 7 atoms respectively.

[0086] More preferably, the oxyalkylene group is selected from oxyethylene, oxypropylene, ethyleneoxyethylene, and propyleneoxy propylene. Even more preferably, the oxyalkylene group is selected from ethyleneoxyethylene and propyleneoxy propylene. Most preferred is when the oxyalkylene group is propyl eneoxy propyl ene .

[0087] The linker group may be selected from -(CH2)3OCH2CH(OH)CH2- or -(CH2)3OCH2CH(OCONHR ^C )CH2-. These are substituted oxyalkylene groups, which each have a length of 7 atoms.

[0088] The Ci-salkylene group is preferably a C2-salkylene group, more preferably a C2-ealkylene group, and even more preferably a C2-4alkylene group.

[0089] In another embodiment, the linker group, L, is represented by formula (A): wherein:

Z is O or NH; m is an integer from 0 to 6; each R ^B is independently selected from H, OR ^C , Ci- ealkyl, -OC(O)OR ^c ,-OC(O)NR ^c ₂ , -NHC(O)NR ^c ₂ and -NHC(O)OR ^c ; and each R ^c is independently selected from H and Ci-ealkyl.

[0090] The dashed line at one end of formula (A) indicates a covalent bond to X in the polysaccharide and the other end of formula (A) is covalently bonded to a surface of the silica, such as by the silicon-containing group.

[0091] In formula (A) above, the length of the linker group is 2 + m, where m is as defined herein.

[0092] Generally, it is preferred that m is at least 2. More preferably m is from 2 to 4. Even more preferably m is 3.

[0093] It may be preferable that Z is O. Alternatively, it may be preferably that Z is NH. It is preferred that Z is the same as X. Thus, when X is O, then Z is O. Similarly, when X is NH, then Z is NH.

[0094] The linker group, L, is preferably represented by formula (Al): where Z, each R ^B and each R ^c are as defined herein; and p is an integer from

0 to 5.

[0095] As above, the dashed line at one end of formula (Al) indicates a covalent bond to X in the polysaccharide and the other end of formula (Al) is covalently bonded to a surface of the silica, such as by the silicon-containing group.

[0096] It is preferred that p is at least 1. More preferably p is from 1 to 3. Even more preferably p is 2. The integer p may be represented by m - 1, when m is greater than or equal to 1.

[0097] In formula (A) or formula (Al), each R ^B is independently selected from H, OR ^C , Ci-ealkyl, -OC(O)OR ^c ,-OC(O)NR ^c ₂ , -NHC(O)NR ^c ₂ and -NHC(O)OR ^c More preferably, each R ^B is independently selected from H, -OH, Ci-ealkyl and -OC(O)NHR ^c . Even more preferably, each R ^B is independently selected from H, Ci-ealkyl and -OC(O)NHR ^c . Most preferably, each R ^B is independently selected from H and Ci-ealkyl.

[0098] Generally, in formula (A) or formula (Al), it is preferred that at least one R ^B is hydrogen. More preferably, all of the R ^B are hydrogen. Thus, the linking group is preferably represented by formula (Aa) or formula (Ala), particularly formula (Ala).

(Aa) (Ala) wherein Z, m and p is each defined as described hereinabove.

[0099] In general, each R ^c is independently selected from H and Ci-ealkyl. It is preferred that R ^c is H. [0100] The linking group, L, is preferably -(CH2)3NHC(O)-. This is a group represented by any one of formulas (A), (Al), (Aa) or (Ala), which has a length of 4 atoms.

[0101] In general, it is preferred that L is -(CH2)3NHC(O)-, (CH ₂ )3OCH ₂ CH(OH)CH2- or -(CH ₂ )3OCH ₂ CH(OCONHR ^c )CH2-. [0102] In a further embodiment, the linker group, L, may be a siloxane group. The siloxane group may have the formula (SI): wherein q is an integer from 1 to 4, and each R’ is independently selected from hydrogen, hydroxy (-OH), Ci-ealkyl and -O-Ci-ealkyl.

[0103] In formula (SI), q may be 1, 2, 3 or 4. [0104] It is preferred that each R’ is independently selected from hydrogen, hydroxy, methyl, ethyl, methoxy and ethoxy.

[0105] In formula (SI), a silicon atom may be covalently bonded (e.g. directly bonded) to X in the polysaccharide.

[0106] The invention relates to the use of silica having groups represented by formula (1), which are covalently bonded to a surface of the silica. This is the stationary phase for performing the column chromatography.

[0107] The silica may be in the form of particles. Thus, the stationary phase may comprise particles of silica, wherein a surface of the particles of silica has groups represented by formula (1) covalently bonded to the surface.

[0108] Alternatively, the stationary phase may be pellicular (e.g. comprise particles having a core-shell structure). When the stationary phase is pellicular, then each particle may have a core and an outer shell comprising the silica having groups represented by formula (1), which are covalently bonded to a surface of the silica.

[0109] Typically, the composition of the core is different to the composition of the outer shell. The core may comprise an impermeable material.

[0110] The core may preferably comprise an inorganic carrier selected from alumina, magnesia, glass, kaolin, titanium oxide, a silicate and a hydroxyapatite.

[0111] Generally, the silica is porous. The silica may have a mean pore size of 10 A to 100 mm, and preferably 50 A to 50,000 A.

[0112] The silica gel has a particle size of from 0.1 pm to 300 pm, preferably from 1 pm to 100 pm, such as from 1 pm to 75 pm, more preferably from 1.0 pm to 20 pm, particularly from 1.0 pm to 10 pm.

[0113] The silica having groups represented by formula (1) may be prepared as described in EP1500430A2 or EP. In particular, the preparative methods and stationary phases of (a) Examples 1, 2 and 5 to 7 of EP 1500430 A2, and (b) Synthesis Examples 2 and 3, and Examples 1 to 4 of EP1632286A1, which are incorporated by reference herein. [0114] The method of the invention involves passing a solution through a chiral chromatography column by normal phase chromatography. It is preferred that the normal phase chromatograph is normal phase HPLC.

[0115] The solution may be passed through the chiral chromatography column at a flow rate of 0.05 to 500 mL/min, preferably 0.10 mL/min to 250 mL/min, such as 0.25 to 50 mL/min, more preferably 0.25 to 25 mL/min. The flow rate depends on the diameter of the chromatography column.

[0116] Typically, the normal phase chromatography, including the normal phase HPLC, comprises eluting the solution with a non-polar solvent, particularly a nonpolar solvent having a partition coefficient ( ) that satisfies logP > 0.5. The partition coefficient is a conventional parameter in the field of column chromatography, particularly HPLC. The partition coefficient is typically determined at the temperature at which chromatography will be performed, which is normally room temperature (e.g. about 20°C).

[0117] The non-polar solvent may comprise hexane, heptane or hexane and heptane. The hexane is preferably n-hexane. The heptane is preferably n-heptane. It has been found that good baseline separation of the stereoisomers can be achieved when hexane and/or heptane is used as the non-polar solvent. This has a safety advantage, e.g. in terms of workplace concentration.

[0118] It is preferred that the non-polar solvent comprises, or consists essentially of, n-heptane.

[0119] The normal phase chromatography, including the normal phase HPLC, comprises eluting the solution with a non-polar solvent mixture. The non-polar solvent mixture may comprise the non-polar solvent and a solvent polarity modifier. The solvent polarity modifier may be added to the non-polar solvent to increase the polarity of the eluting solution.

[0120] The solvent polarity modifier may have a partition coefficient (P’) that satisfies 0.5 > logP ’ > -1.2.

[0121] The non-polar solvent mixture may comprise, or consist essentially of, 70.0 to 99.9 by % volume of the non-polar solvent and 0.1 to 30.0 by % volume of the solvent polarity modifier, preferably 75.0 to 95.0 by % volume of the non-polar solvent and 5.0 to 25.0 by % volume of the solvent polarity modifier.

[0122] The solvent polarity modifier may comprise, or consist essentially of, an alcohol. The alcohol may be selected from ethanol, isopropanol or a combination of ethanol and isopropanol. It may be preferable that the solvent modifier or the alcohol comprises, or consists essentially of, ethanol. Alternatively, it may be preferable that the solvent modifier or the alcohol comprises, or consists essentially of, isopropanol.

[0123] The solvent polarity modifier may comprise, or consist essentially, of an alcohol, such as described herein, and a carboxylic acid. The carboxylic acid may be acetic acid or trifluoroacetic acid. It is preferred that the carboxylic acid is trifluoroacetic acid.

[0124] It is preferred that the solvent polarity modifier is selected from (i) isopropanol and trifluoracetic acid, (ii) ethanol and trifluoroacetic acid and (iii) isopropanol, ethanol and trifluoroacetic acid. In one embodiment, the non-polar solvent comprises isopropanol in combination with ethanol and trifluoroacetic acid, as a solvent polarity modifier.

[0125] In a preferred embodiment, the non-polar solvent mixture comprises 75.0 to 85.0 by % volume of the non-polar solvent and 15.0 to 25.0 by % volume of the solvent polarity modifier. The solvent polarity modifier comprises 0.05 to 0.5 by % volume of the carboxylic acid and 14.50 to 24.95 by % volume of the alcohol. It is preferred that the non-polar solvent is n-heptane, the alcohol is ethanol and the carboxylic acid is trifluoroacetic acid. More preferably the non-polar solvent mixture comprises, or consists essentially of, n-heptane, ethanol and trifluoroacetic acid in a ratio by volume of about 80:20:0.1. It has been found that excellent baseline separation of the stereoisomers can be achieved when using a non-polar solvent mixture as defined in this embodiment.

[0126] The chromatography column may be a preparative chromatography column or an analytical chromatography column. In general, preparative chromatography columns are physically larger than analytic chromatography columns. [0127] The method of the invention may further comprise: individually collecting each separate stereoisomer of N-phthaloyl -glutamic acid from the chromatography column. This involves collecting a fraction (e.g. from the chromatography column) containing N-phthaloyl -L-glutamic acid and separately collecting a fraction (e.g. from the chromatography column) containing N-phthaloyl-D-glutamic acid.

[0128] Thus, the method of the invention may further comprise: collecting a fraction comprising N-phthaloyl -L-glutamic acid (e.g. as the only stereoisomer ofN- phthaloy l-glutamic acid) from the chromatography column in a first container; and collecting a fraction comprising N-phthaloyl-D-glutamic acid (e.g. as the only stereoisomer of N-phthaloyl -glutamic acid) from the chromatography column in a second container. For the avoidance of doubt, the first container is different to the second container (e.g. the containers are not the same).

[0129] It is preferable that the method of the invention comprises the collection steps described above when the method is a method of purifying N-phthaloyl-L- glutamic acid from a solution comprising N-phthaloyl -glutamic acid, preferably a solution comprising an L/D stereoisomer mixture of N-phthaloyl -glutamic acid. The method of purifying N-phthaloyl-L-glutamic acid is preferably a preparative method of purifying N-phthaloyl-L-glutamic acid. The chromatography column is typically a preparative chromatography column.

[0130] The method of the invention may be a method of determining a purity of N- phthaloyl -L-glutamic acid, such as determining an enantiomeric excess of N- phthaloyl -L-glutamic acid in a solution comprising N-phthaloyl -glutamic acid, particularly a solution comprising an L/D stereoisomer mixture of N-phthaloyl- glutamic acid.

[0131] The method of the invention may comprise: passing a solution comprising a solution comprising N-phthaloyl-glutamic acid, particularly a solution comprising an L/D stereoisomer mixture of N-phthaloyl- glutamic acid, through a chiral chromatography column by normal phase chromatography to separate N-phthaloyl-L-glutamic acid from N-phthaloyl-D- glutamic acid. [0132] The separation is a baseline separation. The chiral chromatography column comprises, as a stationary phase, the silica having groups represented by formula (1) covalently bonded to a surface thereof, as described herein.

[0133] The method may further comprise: measuring a signal from N-phthaloyl-L-glutamic acid (e.g. in the solution passing through the chiral chromatograph column).

[0134] It is preferred that the method further comprises: measuring a signal from N-phthaloyl-L-glutamic acid (e.g. in the solution passing through the chiral chromatograph column); and measuring a signal from N-phthaloyl-D-glutamic acid (e.g. in the solution passing through the chiral chromatograph column).

[0135] In general, the method of the invention may further comprise: determining the purity of N-phthaloyl-L-glutamic acid in the solution comprising N-phthaloyl -glutamic acid, particularly the solution comprising an L/D stereoisomer mixture of N-phthaloyl -glutamic acid.

[0136] The purity may be the enantiomeric excess of N-phthaloyl-L-glutamic acid.

[0137] The purity of N-phthaloyl-L-glutamic acid may be determined from the signal measured for N-phthaloyl-L-glutamic acid and the signal measured for N- phthaloyl-D-glutamic acid. Methods for determining the purity of a substance using column chromatography, particularly HPLC, are known in the art. Such conventional methods may be used to determine the purity of N-phthaloyl-L- glutamic acid.

[0138] The signal from N-phthaloyl-L-glutamic acid may be detected by an ultraviolet/visible light detector (UV-Vis detector). Thus, the signal from N- phthaloyl-L-glutamic acid may be a peak in a UV and/or visible light absorbance spectrum.

[0139] Similarly, the signal from N-phthaloyl-D-glutamic acid may be detected by a UV-Vis detector. The signal from N-phthaloyl-D-glutamic acid may be a peak in a UV and/or visible light absorbance spectrum.

[0140] The method may involve: determining a first area under the curve (AUC) from the UV-Vis signal for N-phthaloyl-L-glutamic acid; determining a second area under the curve (AUC) from the UV-Vis signal for N phthaloyl-D-glutamic acid; and calculating the amount of N-phthaloyl-L-glutamic acid as a proportion of the total amount of the L- and D-stereoisomers of N-phthaloyl-L-glutamic in the solution using the first AUC and the second AUC.

[0141] When determining a purity of N-phthaloyl-L-glutamic acid, it is preferred that the chiral chromatography column is an analytical chiral chromatography column.

EXAMPLES

[0142] The following examples are provided to illustrate the invention and are not intended to limit the scope of the invention, as described herein.

Example 1

[0143] A solution comprising 10 mg/mL of an L/D stereoisomer mixture of N- phthaloyl -glutamic acid was passed through a chiral chromatography column (250 mm length, 4.6 mm inner diameter, and 5 pm diameter particle size) using a solvent mixture of n-heptane, ethanol and trifluoroacetic acid in a ratio by volume of 80:20:0.1.

[0144] The chiral chromatography column had a stationary phase in accordance with the invention (CHIRALPAK™ IC column), where a polysaccharide was immobilized on silica. The polysaccharide was a cellulose-derived polysaccharide, where R ¹ had the structure shown below. A flow rate of 0.40 mL/min was used. [0145] The HPLC spectrum from a HPLC instrument equipped with a UV-Vis detector set at 254 nm is shown in Figure 1. As can be seen from Figure 1, the spectrum shows two distinct peaks 10 and 20, which are well-separated at the baseline as shown by arrow 100. Example 2

[0146] The experiment in Example 1 was repeated using a different chiral chromatography column. The column had a stationary phase in accordance with the invention (CHIRALPAK™ IB-N column). The polysaccharide was immobilized on silica. The polysaccharide was a cellulose-derived polysaccharide in which R ¹ had the structure shown below.

[0147] The HPLC spectrum is shown in Figure 2. The UV-Vis detector set at 290 nm. The spectrum also shows two distinct peaks 10 and 20, which have excellent baseline separation (see 100).

Comparative Example

[0148] The experiment of Example 1 was repeated using a different chiral chromatography column, which is not in accordance with the invention (CHIRALPAK™ OJ-H column). The polysaccharide was immobilized on silica. The polysaccharide was a cellulose-derived polysaccharide in which R ¹ had the structure shown below. [0149] The HPLC spectrum is shown in Figure 3. The spectrum also shows two peaks 10 and 20, which were not separated at the baseline (see 50). It is not possible to achieve good separation of N-phthaloyl-L-glutamic acid from the mixture of stereoisomers.

Reference Numerals

[0150] The following reference numerals are used in the figures.

10 peak for a stereoisomer of N-phthaloyl -glutamic acid

20 peak for a stereoisomer of N-phthaloyl -glutamic acid 50 overlap between the peaks from the stereoisomers 100 separation at the baseline between the peaks for the stereoisomers

[0151] All references, including publications, patent applications, and patents, cited herein are hereby incorporated by reference in their entirety and to the same extent as if each reference were individually and specifically indicated to be incorporated by reference and were set forth in its entirety herein (to the maximum extent permitted by law).

[0152] All headings and sub-headings are used herein for convenience only and should not be construed as limiting the invention in any way.

[0153] The use of any and all examples, or exemplary language (e.g. “such as”) provided herein, is intended merely to better illuminate the invention and does not pose a limitation on the scope of the invention unless otherwise paragraphed. No language in the specification should be construed as indicating any non-paragraphed element as essential to the practice of the invention.

[0154] This invention includes all modifications and equivalents of the subject matter recited in the paragraphs appended hereto as permitted by applicable law.

[0155] This patent application claims the priority of the European patent application 22170440.6, wherein the content of this European patent application is hereby incorporated by references.