Doped materials for reverse phase chromatography

Abstract

A material for reverse phase chromatography comprises surface modifying apolar and charged groups bound to a solid support, said charged groups being present in amounts of about 0.25 to about 22% of the surface modifying groups, or in amounts of about 0.01 mol/m.sup.2 to 0.8 mol/m.sup.2 referred to the surface of the solid support for a material with a total amount of surface modifying groups of 3.6 mol/m.sup.2. Such material and suitable purification conditions for active pharmaceutical ingredients (APIs) like peptides can be evaluated by (a) determining the isoelectric point (pI) of the API of interest, (b) choosing a pH in a range where the solid phase material is stable, (c) determining the difference pIpH and (d) if the difference pIpH is positive, choosing an anion exchange (AIEX) material, or if the difference pIpH is negative, choosing an cation exchange (CIEX) material.

Claims

1. A material for reverse phase chromatography comprising surface modifying groups bound to a solid support that is silica gel, wherein said surface modifying groups are not horizontally polymerized, said surface modifying groups comprising apolar uncharged groups and polar groups, and wherein said polar groups being charged groups, said charged groups being either positively charged or negatively charged, and said charged groups being present in amounts of about 0.25 to about 22% of the surface modifying groups, or in amounts of about 0.01 mol/m.sup.2 to 0.8 mol/m.sup.2 referred to the surface of the solid support for a material with a total amount of surface modifying groups of 3.6 mol/m.sup.2, the surface modifying groups being bound to the surface via a reaction of monofunctional silanes carrying the surface modifying groups or trifunctional silanes carrying the surface modifying groups, the positively charged groups being NR.sub.3.sup.+ groups wherein R is selected from hydrogen or C1-C2 alkyl and the negatively charged groups being SO.sub.3.sup. groups, the charged groups are terminal substituents of aliphatic C1 to C18 groups, the apolar uncharged groups are linear C4 to C18 alkyl groups, and the aliphatic groups of the charged groups are shorter by up to 6 CH.sub.2 groups compared to the length of the apolar groups.

2. A method for a reversed phase chromatographic separation of active pharmaceutical ingredients (APIs), said method comprising evaluating a reversed phase material of claim 1, and API purification conditions by (a) determining the isoelectric point (pI) of the API of interest, (b) choosing a pH in a range where the solid support material is stable, (c) determining the difference of pIpH and (d) if the difference pIpH is positive (pI>pH), choosing an anion exchange (AIEX) material, or if the difference pIpH is negative (pI<pH), choosing a cation exchange (CIEX) material.

3. The method of claim 2 further comprising the step of (e) evaluating a desired retention time by adapting at least one eluting condition.

4. The method of claim 3 wherein the at least one eluting condition comprises the ionic strength of a buffer.

5. The method of claim 3 wherein the at least one eluting condition comprises an eluent system, an initial eluent composition and a gradient.

6. The method of claim 5 wherein the eluent system is an aqueous buffer and acetonitrile.

7. The method of claim 2 wherein the pH is between 3 and 9.

8. A method for producing a material for reverse phase chromatography of claim 1, comprising attaching all surface modifying groups simultaneously from a liquid solution comprising the apolar groups and the polar charged groups.

9. A method for producing a material for reverse phase chromatography of claim 1, comprising a first step, wherein the apolar groups are attached followed by a second step wherein the charged groups are attached or vice versa, wherein the first step is performed such that some available silanol groups are not reacted.

10. The material of claim 1, wherein the apolar uncharged groups comprise C8 groups and the charged groups comprise C3 groups.

11. The material of claim 1, wherein the charged groups are NR3+ and wherein R is methyl.

12. The material of claim 1, wherein the silica gel has a porosity of 50 to 300 , and particle sizes from 5 to 100 m.

13. The material of claim 12, wherein the silica gel has a porosity of about 100 , and particle sizes from 10 to 20 m.

14. The method of claim 2, wherein the active pharmaceutical ingredients (APIs) comprise peptides.

15. The method of claim 7, wherein the pH is between 4 and 7.

16. The method of claim 8, wherein the method comprises determining the amount of apolar and ionic groups attached to the material.

17. The method of claim 9, wherein the attached ionic groups are determined.

18. The material of claim 1, wherein the aliphatic groups of the charged groups are shorter than the apolar groups by up to 5 CH.sub.2 groups.

19. The material of claim 1, wherein the aliphatic groups of the charged groups are shorter than the apolar groups by 3 to 5 CH.sub.2 groups.

20. The material of claim 1, wherein the monofunctional silanes carrying the surface modifying groups are monochlorosilanes.

21. The material of claim 1, wherein the trifunctional silanes carrying the surface modifying groups are trichlorosilanes or trimethoxysilanes.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention will be better understood and objects other than those set forth above will become apparent when consideration is given to the following detailed description thereof. Such description makes reference to the annexed drawings, wherein:

(2) FIG. 1 shows a correlation between anionic functional group (SO.sub.3.sup.) in the reaction mixture and anionic functional group attached to the silica gel.

(3) FIG. 2 shows a correlation between cationic functional group (NR.sub.3.sup.+) in the reaction mixture and cationic functional group attached to the silica gel.

(4) FIG. 3 shows the effect of buffer concentration (FIG. 3a for 300 mM acetate, FIG. 3b for 500 mM acetate, FIG. 3c for 800 mM acetate) on the retention time of a positively charged API on various columns.

(5) FIG. 4 shows the effect of modified acetonitrile gradient on retention time, wherein FIG. 4a shows the retention time of the specific columns using identical elution conditions while FIG. 4b shows that the retention times can be brought together by varying the initial composition and/or the gradient.

(6) FIG. 5 shows the selectivity for the separation of insulin/desamido-insulin (insulin pI=5.3) on different column materials and at two different pH.

(7) FIG. 6 shows preparative purification of synthetic peptide 1 according to Example 5.

(8) FIG. 7: shows preparative purification of synthetic peptide 2 according to Example 6.

(9) FIG. 8: shows preparative purification of synthetic peptide 3 according to Example 7.

MODES FOR CARRYING OUT THE INVENTION

(10) According to the presently preferred method for surface modification, all surface modifying groups are attached simultaneously from a liquid solution comprising the apolar groups and the polar charged groups. Dependent on the substituents used it has been found that the reaction may more or less favor the charged or the apolar groups. Therefore the actually obtained material has to be determined by e.g. ionic titration.

(11) It is also possible to perform the reaction in several steps, i.e. attachment of the apolar groups first, followed by the charged groups or vice versa. In this procedure, however, the reaction conditions have to be selected such that in the first step some available silanol groups are not reacted. In any case, also in a two (or more) step procedure the actually attached ionic groups have to be determined, e.g. by ionic titration.

(12) The preferred and/or obtainable ranges of charged groups may vary from charged group to charged group. For SO.sub.3.sup. the usual range is from 0.01 mol/m.sup.2 to 0.04 mol/m.sup.2 and for NR.sub.3.sup.+ the usual range is from 0.01 mol/m.sup.2 to 0.8 mol/m.sup.2 although higher amounts can easily be produced (see examples below). Higher NR.sub.3.sup.+-doped materials with e.g. RCH.sub.3, however, have been found to be less good (see FIGS. 3 and 4). The lower limit of the amount of ionic groups is the minimal number of doping ionic groups that has to be present for a visible effect. The upper limit is given by the strength or weakness, respectively, of the adsorption. Thus, higher than the indicated amounts of ionic groups might be used but with possibly worse separation and/or inadequate retention times.

(13) The effect of the doping is that in case of a API that is positively charged (isoelectric point (pIpH=positive, or pI>pH, respectively) the adsorption on an AIEX material, e.g. NR.sub.3.sup.+ doped material, is weak (repulsion) and on a CIEX material, e.g. SO.sub.3.sup. doped material, strong (attraction) and vice versa.

(14) As indicated above, the apolar groups are preferably at least C4 groups and usually not larger than C18 groups with C8 groups being presently preferred. The charged groups encompass any charged substituent, preferably a charged substituent attached to a C1 to C18 hydrocarbon, more preferred a charged substituent attached to a hydrocarbon having slightly smaller or about the same length as the apolar chain. Due to the fact that the substituent is larger than a hydrogen, the charged substituent carrying group usually and preferably have a slightly sorter chain, e.g. C3 compared to C8 of the apolar group. Without wanting to reduce the scope of the invention by any interpretation, the inventors assume that attaching the ionic groups via a shorter hydrocarbon chain results in a weak shielding of the ionic groups and thus in a lower attraction or repulsion of the molecules in the analyte.

(15) The invention is now further described by means of some examples:

Example 1

(16) Several CIEX and AIEX materials have been prepared by subjecting them to compositions comprising the C8-compound octyltrimethoxysilane and (0%), 5%, 10%, 15% or 50% of SO.sub.3.sup.-charged or NR.sub.3.sup.+-charged C3-compound according to standard procedures.

(17) The NR.sub.3.sup.+ materials (R=methyl) were produced by combining the desired amount of silicagel, octyltrimethoxysilane, N-trimethoxysilylpropyl-N,N,N-trimethylammonium chloride, and p-toluensulfonic acid (catalyst) and boiling under reflux in toluene for 6 hours.

(18) The desired amounts were as follows: Octyltrimethoxysilane in a concentration of
(1x)*5.6*10.sup.6 mol/m.sup.2*S*M*MWt, wherein

(19) x is the desired weight percentage of NR.sub.3.sup.+ groups,

(20) S the specific surface of the silicagel [m.sup.2/g],

(21) M the mass of the silicagel used [g] and

(22) MWt the molecular weight of octyltrimethoxysilane [g/mol].

(23) N-trimethoxysilylpropyl-N,N,N-trimethylammonium chloride was used as a solution of 50% by weight of the silane in methanol and the amount solution used was
x*2g/g*Mx*2*M, wherein

(24) x is the percentage of the NR.sub.3.sup.+ groups as described above and

(25) M is the mass of silicagel [g].

(26) p-Toluensulfonic acid (catalyst) was used in a concentration of 8.6*10-3 g/g of octyltrimethoxysilane and N-trimethoxysilylpropyl-N,N,N-trimethylammonium chloride

(27) After 6 hours the material was endcapped, i.e. not reacted, still accessible silanol groups, were reacted with trimethylchloro silane by a standard technique and washed with a series of solvents with different polarities.

(28) The SO.sub.3.sup.-materials were produced by combining the desired amount of silicagel, octyltrimethoxysilane, mercaptopropyltrimethoxysilane, and p-toluensulfonic acid, in toluene and boiling under reflux for 6 hours.

(29) The desired amounts were as follows:

(30) Octyltrimethoxysilane in a concentration of
(1x)*5.6*10.sup.6 mol/m.sup.2*S*M*MWt, wherein

(31) x is the desired weight percentage of SO.sub.3 groups,

(32) S is the specific surface of the silicagel [m.sup.2/g],

(33) M is the mass of silicagel used [g] and

(34) MWt the molecular weight of the silane [g/mol].

(35) Mercaptopropyltrimethoxysilane was used in a concentration of
x*5.6*10.sup.6 mol/m.sup.2*S*M*MWt, wherein

(36) x is the desired weight percentage of SO.sub.3.sup. groups,

(37) S the specific surface of the silicagel [m.sup.2/g],

(38) M the mass of silicagel used [g] and

(39) MWt the molecular weight of the silane [g/mol].

(40) p-Toluensulfonic acid was used in a concentration of 8.6*10.sup.3 g/g of the total amount of octyltrimethoxysilane and mercaptopropyltrimethoxysilane used.

(41) The material was then endcapped by the use of standard techniques and washed with a series of solvent with different polarity.

(42) SO.sub.3.sup. groups were created by oxydizing the mercapto group as follows: The material was put in a solution of 30% by volume acetone in H.sub.2O.sub.2 (30% by weight in water) and kept in the solution for 12 hours at room temperature. Finally the material was washed with water and methanol.

(43) The amount of charged groups attached as a function of the concentration in solution and determined by ionic titration is shown in FIG. 1 for the anionic functional group SO.sub.3.sup. and in FIG. 2 for the cationic functional group NR.sub.3.sup.+.

(44) The CIEX columns thus prepared have been termed C5 (5% anionic groups in solution), C10 (10% anionic groups in solution), C15 (15% anionic groups in solution), C50 (50% anionic groups in solution), and C100 (100% anionic groups in solution), the neutral column RP (0% ionic groups in solution), and the AIEX columns A5 (5% cationic groups in solution), A10 (10% cationic groups in solution), A15 (15% cationic groups in solution), and A50 (50% cationic groups in solution). The amount of attached charged groups (found by titration) can be derived from FIGS. 1 and 2.

Example 2

(45) A peptide (synthetic peptide 1) with pI between 11 and 12 was subjected to separation with different columns and different ionic strength of the used acetate buffer of pH 4.8 and identical acetonitrile (AcN) gradient of 0.51% by volume/minute, starting at about 19% AcN.

(46) As can be seen from each of the FIGS. 3a to 3c, the adsorption of a positively charged peptide on NR.sub.3.sup.+/C8 materials is low (repulsion) while it is strong on SO.sub.3.sup./C8 materials (attraction). Several columns were tested under identical conditions and the retention times compared with each other. As can be seen, the column A15 has almost no adsorption or strong repulsion, respectively, while the material C15 has strong adsorption and thus long retention time. The A15 material is not usable since the peptide of interest does not adsorb while the C15 material is not so good in economic view because the purification process is too slow and requires substantial amount of solvent.

(47) As can be seen from a comparison of FIGS. 3a to 3c, column A15 with about 1 mol/m.sup.2 has not only an undesirably short retention time of less than 5 minutes but it does also not react to the changes in the ionic strength (buffer concentration). This is another piece of evidence that this material is not suitable for the intended use.

(48) Contrary thereto, the less doped materials reacted quite sensitive to the changed reaction conditions making all of them valuable reversed phase materials. By enhancing the buffer concentration from 300 mM acetate to 800 mM acetate the retention time on C15 could be reduced from more than 30 minutes to less than 25 minutes. In addition, at high buffer concentration the retention times of all columns came closer indicating that the salt (acetate) shields the charges such that the doped materials behave similar to the not doped RP material.

(49) As already indicated above, only for column A15 no effect was found. Without wanting to be limited by any interpretation, the inventors assume that this unexpected behaviour may indicate that this material is so repulsing that the buffer concentrations used are not sufficient to overcome this repulsion.

Example 3

(50) In this example the effect of a changed starting concentration and gradient on the retention times of the different columns was investigated. In order to minimize the effect of the buffer salt, only 100 mM acetate buffer was used. Also in this example the API was synthetic peptide 1 with pI of 11 to 12.

(51) The AcN gradient was 0.25% by volume/minute.

(52) In one experiment, the AcN initial concentration was 22% AcN (see FIG. 4a), in a second experiment the AcN initial concentration was changed to insure retention time t.sub.R=303 min (FIG. 4b).

(53) As can be seen form a comparison of FIG. 4a with FIG. 4b, the retention times are highly dependent from the initial AcN concentration.

(54) The conditions leading to the results shown in FIG. 4b for all columns are listed in Table 1 below:

(55) TABLE-US-00001 TABLE 1 AcN Start AcN End gradient gradient retention Column [v %] [v %] time [min] [v %/min] time [min] C15 31.7 47.3 61.2 0.25 28.5 C10 29.5 45.1 61.2 0.25 29.8 C5 24.2 39.8 61.2 0.25 32.8 RP 21.9 37.5 62.4 0.25 28.0 A5 14.8 30.4 61.3 0.25 27.2 A10 10.8 26.4 61.2 0.25 28.6 A15 6.3 21.9 61.3 0.25 dead time

Example 4

(56) The selectivity of the different columns in the separation of insulin/desamido-insulin (insulin pI=5.3) at pH 4.8 and pH 6.8 was investigated. The results are shown in FIG. 5.

(57) The experimental conditions were: eluent: 240 mM acetate (pH 4.8 or 6.8, respectively) and AcN in an amount to ensure a retention time t.sub.R of insulin of about 32 minutes; and isocratic elution.

(58) The insulin/desamido-insulin selectivity at pH 4.8 was S<1 on a CIEX material S>1 on the RP S=1 on the CIEX-RP materials S>>1 on the AIEX-RP materials

(59) As can be seen from FIG. 5, at pH 4.8 (.diamond-solid.) the NR.sub.3.sup.+/C8 materials show much enhanced selectivity compared to the pure C8 material. At higher pH () the selectivity is inversed. Thus, the repulsing materials are better than the attracting materials, i.e. SO.sub.3.sup./C8.

Example 5

(60) This example describes the purification of a synthetic peptide 1 that was selected for a pI between 11 and 12. The pH of the mobile phase was below the peptide pI, i.e. the peptide was positively charged.

(61) The mobile phase was pH=4.8, 500 mM sodium acetate buffer+acetonitrile

(62) The starting concentration and the gradient information is listed in Table 2:

(63) TABLE-US-00002 TABLE 2 AcN Start AcN End gradient time Column [v %] [v %] [min] C50 28.6 33 80 C5 17 33 80 RP 15.9 32.2 80 A5 13.9 30 80 A10 11.7 26.4 80

(64) Result: As can be seen from FIG. 6, by using an AIEX-RP material (A10) the yield could be raised for 25% at a fixed purity of 94%.

Example 6

(65) This example describes the purification of a synthetic peptide 2 that was selected for a pI between 9 and 10. The pH of the mobile phase below the peptide pI, i.e. the peptide was positively charged.

(66) The mobile phase was pH=4.8, 120 mM sodium acetate buffer+acetonitrile

(67) The starting concentration and the gradient information are listed in Table 3.

(68) TABLE-US-00003 TABLE 3 AcN Start AcN End gradient time Column [v %] [v %] [min] C5 28.6 48.6 80 RP 21.9 39.3 80 A5 17.5 35.7 80 A10 14.8 32.2 80

(69) Result: As can be seen from FIG. 7, by using an AIEX-RP material (A10) the yield could be raised for 7.4% at a fixed purity of 94%.

Example 7

(70) This example describes the purification of a synthetic peptide 3 that was selected for a pI between 4.5 and 5.5. The pH of the mobile phase was above the peptide pI, i.e. the peptide was negatively charged.

(71) The mobile phase was pH=6.5, 100 mM ammonium acetate buffer+acetonitrile.

(72) The starting concentration and the gradient information are listed in Table 4.

(73) TABLE-US-00004 TABLE 4 AcN Start AcN End gradient time Column [v %] [v %] [min] 15C 20.2 34.7 60 10C 20.2 34.7 60 5C 20.7 35.2 60 RP 21.7 36.2 60 A5 24.6 39.1 60 A10 28.4 42.9 60

(74) Result: As can be seen from FIG. 8, in this case, due to the negative charge of the API the SO.sub.3.sup./C8 materials show better performance. Also here the repulsing materials are better. By using a CIEX-RP material (C10) the yield could be raised for 5.9% at a fixed purity of 87%.

CONCLUSION

(75) All experiments show that repulsingly doped materials have improved performance. This is in contradiction to the general opinion that for improved performance additional adsorption positions have to be created. The present results show that the performance of the materials is improved by creating repulsing positions.

(76) While there are shown and described presently preferred embodiments of the invention, it is to be distinctly understood that the invention is not limited thereto but may be otherwise variously embodied and practiced within the scope of the following claims.