Method for preparing glucagon-like peptides

Abstract

The present invention refers to a method for preparing a glucagon-like peptide, comprising precipitation of the peptide or of a precursor peptide by means of mixing with an anti-solvent comprising diisopropyl ether and acetonitrile. Further, the present invention also relates to a peptide conjugated to a solid phase and a pharmaceutical composition comprising a Liraglutide peptide obtainable from a method according to the present invention.

Claims

1. A method for preparing a Liraglutide peptide or a salt thereof, comprising: (i) providing a solution S comprising a peptide of formula I: TABLE-US-00011 His-Ala-Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser-Ser- Tyr-Leu-Glu-Gly-Gln-Ala-Ala-B.sup.1-Glu-Phe-Ile-Ala- Trp-Leu-Val-Arg-Gly-Arg-Gly, wherein B.sup.1 is Lys(palmitoyl-Glu-OH) or Lys(H-Glu-OH); (ii) precipitation of the peptide of step (i) by mixing solution S with an anti-solvent comprising at least 50% (v/v) of diisopropyl ether and acetonitrile, wherein the volume ratio (diisopropyl ether:acetonitrile) is in the range of from (3:1) to (5:1); and (iii) isolating the precipitate obtained from step (ii).

2. The method according to claim 1, wherein step (i) comprises: (i-a) providing a precursor peptide conjugated to a solid phase: TABLE-US-00012 His-Ala-Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser-Ser- Tyr-Leu-Glu-Gly-Gln-Ala-Ala-B.sup.2-Glu-Phe-Ile-Ala- Trp-Leu-Val-Arg-Gly-Arg-Gly-[resin], wherein B.sup.2 is Lys (palmitoyl-Glu-OR.sub.1) or B.sup.2 is Lys (R.sub.2-Glu-OR.sub.1), with R.sub.1 being a carboxylic acid protecting group and R.sub.2 being an amino protecting group; and wherein at least the side chains of Glu, Asp, and Lys bear protecting groups; and (i-b) cleaving the precursor peptide off the resin.

3. The method according to claim 2, wherein step (i-a) comprises Fmoc-based Solid Phase Peptide Synthesis using suitably protected amino acid derivatives or dipeptide derivatives, wherein said protected amino acid derivatives or dipeptide derivatives are activated by one or more coupling reagent/additive mixtures selected for each step independently from the group consisting of: (A) (benzotriazolyl)tetramethyluronium tetrafluoroborate (TBTU)/diisopropylethylamine (DIPEA); (B) diisopropylcarbodiimide (DIC)/cyano-hydroxyimino-acetic acid ethyl ester; (C) 3-(diethoxy-phosphoryloxy)-3H-benzo[d][1,2,3]triazin-4-one (DEPBT)/DIPEA; and (D) DIC/hydroxybenzotriazole (HOBt).

4. The method according to claim 2, wherein one or more pseudoproline dipeptides are introduced at a position selected from the group consisting of Gly.sup.4-Thr.sup.5, Phe.sup.6-Thr.sup.7, Thr.sup.7-Ser.sup.8, Val.sup.10-Ser.sup.11, and Ser.sup.11-Ser.sup.12 of the peptide of formula I.

5. The method according to claim 2, wherein the N-terminal histidine moiety is introduced into the precursor peptide conjugated to the solid phase using an amino acid derivative selected from the group consisting of Boc-His(Boc)-OH, Boc-His(1-Trt)-OH, and Fmoc-His(1-Trt)-OH and the coupling reagent/additive mixture DEPBT/DIPEA.

6. The method according to claim 3, wherein the Fmoc protecting group is cleaved off the growing peptide chain conjugated to the solid phase using a mixture selected from the group consisting of 5-50% (v/v) piperidine or 4-methyl piperidine in N,N-dimethylformamide (DMF), 5-50% (v/v) piperidine or 4-methyl piperidine in N-methylpyrrolidone (NMP), 1-5% (v/v) diazabicyclo[5.4.0]undec-7-ene (DBU) in DMF, and 50% (v/v) morpholine in DMF.

7. The method according to claim 1, further comprising a step of reacting an activated ester of palmitic acid, with the Lys(H-Glu-OH) moiety of the peptide of formula I or with a Lys(H-Glu-OR.sub.1) moiety obtainable by cleaving off the amino protecting group R.sub.2 from the precursor peptide provided in step (i-a).

8. The method according to claim 1, wherein the solution S obtained from step (i) further comprises trifluoroacetic acid (TFA) and one or more scavengers.

9. The method according to claim 1, wherein the anti-solvent used in step (ii) comprises at least 75% (v/v), of a mixture M of diisopropyl ether and acetonitrile, and wherein the volume ratio (diisopropyl ether:acetonitrile) in said mixture M is in the range of from (3:1) to (5:1).

10. The method according to claim 1, wherein step (ii) comprises mixing the anti-solvent with solution S obtained from step (i) by: (ii-a) pre-mixing diisopropyl ether and acetonitrile before mixing it with the solution S obtained from step (i); or (ii-b) first mixing diisopropyl ether with the solution S obtained from step (i) and subsequently mixing acetonitrile with the mixture comprising solution S and diisopropyl ether; or (ii-c) first mixing acetonitrile with the solution S obtained from step (i) and subsequently mixing diisopropyl ether with the mixture comprising solution S and acetonitrile.

11. The method according to claim 1, wherein step (ii) is carried out at a temperature in the range of −5° C. to 10° C.

12. The method according to claim 1, wherein in step (iii), the precipitate obtained from step (ii) is isolated by filtration, centrifugation, or a combination of filtration and centrifugation.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 shows the sequence alignment of selected glucagon-like peptides. Moieties sharing identity with the GLP-1 sequence are written in bold.

(2) FIG. 2 shows the particle size distribution of peptide precipitates as measured by FBRM technology (conducted on the following device: ParticleTrack G600L, Mettler Toledo, Settings: Macro V1.1.11, cube weight, normalized, mean (cube weight)). The chord length is the distance across each particle as calculated based on scan speed and the number and duration of distinct pulses of backscattered light. The percentage given relates to the number of detected particles and is normalized such that the maximal number of particles of a given size within one scan is 100%. Herein, the precipitate EOP26a of Example 5 (anti-solvent: diethyl ether mean particle size: 261 μm) is compared with the precipitate EOP22a of Example 2 (anti-solvent: IPE/ACN mean particle size: 422 μm)

EXAMPLES

Example 1: Solid Phase Peptide Synthesis of Glucagon-Like Peptides

(3) Stepwise Fmoc-SPPS of Liraglutide was performed on a 100-200 or 200-400 mesh H-Gly-2-chlorotrityl resin (Bachem no. 4092098 or 4026823) using the standard Fmoc amino acid derivatives Fmoc-Ala-OH, Fmoc-Arg(Pbf)-OH, Fmoc-Asp(OtBu)-OH, Fmoc-Asp(OMpe)-OH, Fmoc-Gly-OH, Fmoc-Gln(Trt)-OH, Fmoc-Glu(OtBu)-OH, Boc-His(Boc)-OH, Fmoc-His(1-Trt)-OH Fmoc-Ile-OH, Fmoc-Leu-OH, Fmoc-Phe-OH, Fmoc-Ser(tBu)-OH, Fmoc-Thr(tBu)-OH, Fmoc-Trp(Boc)-OH, Fmoc-Tyr(tBu)-OH, and Fmoc-Val-OH, as well as the previously described building block Fmoc-Lys(N.sup.ε-(γ-glutamyl(OtBu)-(N.sup.α-hexadecanoyl))) (=Fmoc-Lys(palmitoyl-Glu-OtBu)-OH, see WO 2013/171135) and at least one pseudoproline dipeptide selected from Fmoc-Gly-Thr(Psi(Me,Me)pro)-OH, Fmoc-Phe-Thr(Psi(Me,Me)pro)-OH, Fmoc-Thr(tBu)-Ser(Psi(Me,Me)pro)-OH, Fmoc-Val-Ser(Psi(Me,Me)pro)-OH, and Fmoc-Ser(tBu)-Ser(Psi(Me,Me)pro)-OH. Coupling reactions were executed either with DIC/OXYMAPURE® (cyano-hydroxyimino-acetic acid ethyl ester), TBTU/DIPEA or DEPBT/DIPEA with appropriate coupling (1.5-24.0 h) and Fmoc deprotection (0.5-4.0 h) times.

(4) When desired, 20% piperidine in NMP was used instead of 20% piperidine in DMF to enhance deprotection rates. After coupling steps, an acetylation was optionally performed using acetic anhydride. DMF and IPA were used as solvents for the washing steps after acetylation or Fmoc deprotection. Crude Liraglutide was typically obtained after cleavage and precipitation with a U-HPLC purity of about 50%.

(5) Alternatively, a stepwise protocol was carried out involving the use of Fmoc-Lys(Boc-Glu-OtBu)-OH as a building block and subsequent palmitoylation of the purified peptide as follows.

(6) Stepwise Fmoc SPPS of a fully protected peptide of the primary amino acid sequence:

(7) TABLE-US-00007 His-Ala-Glu-Gly-Thr-Phe-Thr-Ser-Asp-Val-Ser-Ser- Tyr-Leu-Glu-Gly-Gln-Ala-Ala-Lys(Boc-Glu-OtBu)-Glu- Phe-Ile-Ala-Trp-Leu-Val-Arg-Gly-Arg-Gly-[resin]
was carried out on a automated synthesizer using a H-Gly-2-chlorotrityl resin. Coupling reactions were carried out using the respective standard Fmoc amino acid derivative, DIC/OXYMAPURE® (cyano-hydroxyimino-acetic acid ethyl ester) or TBTU/DIPEA as coupling reagents and DMF as a solvent. As an exception, the coupling of Boc-His(Boc)-OH was carried out using DEPBT/DIPEA in DMF. The coupling time varied between 1.5 h and 4 h. The introduction of the Fmoc-Lys(Boc-Glu-OtBu)-OH building block was performed manually (DIC/OXYMAPURE® (cyano-hydroxyimino-acetic acid ethyl ester), 4 h coupling time) whereas the rest of the synthesis was conducted via fully automated peptide synthesis. Fmoc deprotection was performed using a solution of 20% piperidine in DMF (v/v). Deprotection time varied between 15 up to 90 min. After each coupling, an systematic capping step was performed using acetic anhydride.

(8) DMF and/or IPA were used as solvents for the washing steps after Fmoc deprotection and acetylation. The peptide was cleaved from the solid support with concomitant removal of all the side chain protecting groups by incubation with a cleavage composition consisting of TFA/H.sub.2O/EDT (90:5:5) v/v/v for 2.5 h, followed by precipitation. The resulting crude material was subjected to a two dimensional preparative reversed phase HPLC using C8-bonded silica as a stationary phase, a TEAP eluent system in the first, and a TFA eluent system in the second dimension. The purified precursor peptide had an U-HPLC purity of typically 95% and was reacted with N-succinimidyl palmitate in aqueous solution comprising 66% THF. The resultant product was precipitated by evaporation of the solvent, re-dissolved by the addition of 50% AcOH in 10% ACN/H.sub.2O and filtered.

Example 2: Precipitation of Glucagon-Like Peptides from the Cleavage Composition

(9) General Procedure:

(10) Peptide resin (5 g) was suspended in 50 mL of cleavage composition and stirred for 3 h at RT. The resin was then filtered off and washed with TFA (2×3.5 mL). The cleavage composition and washing fractions were pooled. 570 mL of the respective anti-solvent was added to initiate precipitation. The resulting suspension was stirred for a further 2 h before separating the peptide precipitate by the use of either a filter funnel or a standard pressurized filter equipment (“pocket filter”). Filtration time and gross yield based on peptide resin were determined. The precipitate was washed with the respective ether (3×15 mL) and dried under vacuum at RT to yield the crude peptide.

(11) Alternatively, peptide resin (3 g) was suspended in 30 mL cleavage composition and stirred for 3 h at RT. The resin was then filtered off and washed with TFA (2×2.1 mL). The cleavage composition and washing fractions were pooled. The solution was added to 342 mL of anti-solvent to initiate precipitation. The resulting suspension was stirred for a further 2 h before separating the peptide precipitate by the use of either a filter funnel or a standard pressurized filter equipment (“pocket filter”). Filtration time and gross yield based on peptide resin were determined. The precipitate was washed with the respective ether (3×18 mL) and dried under vacuum at RT to yield the crude peptide.

(12) Results:

(13) The dried peptide precipitate was analyzed for purity and TFA content by analytical reversed phase UHPLC, and the aggregate content was determined by analytical size exclusion UHPLC. The appearance/stickiness of the precipitate was assessed by visual inspection.

(14) As shown in Table 2, it was found that precipitation with mixtures of IPE/ACN in a range of 1:3-1:10 not only resulted in acceptable yields and filtration times but also improved the precipitate's purity as compared to pure ether anti-solvents (compare lines 5-7 with lines 1-3).

(15) The purity could further be improved to 63.1% when the anti-solvent was not pre-mixed, but divided in an ACN part and an IRE part and said parts were contacted subsequently with the peptide solution (data not shown). To further facilitate the comparison of the various precipitation methods, a performance score P was calculated as P P=P.sub.filtration time×P.sub.purity×P.sub.yield×P.sub.appearance×P.sub.aggregate content. As each sub-score ranged from 1-3, the maximal performance score is 243, the minimal score is 1. Also if assessed by this metric, mixtures of IPE/ACN were found to be clearly advantageous compared to the other anti-solvents tested.

(16) TABLE-US-00008 TABLE 2 Precipitation of Liraglutide from TFA/water/EDT/TIPS cleavage composit. Filtration HPLC Experiment Performance Anti- Yield Time Purity No. Score solvent Mode [%] [min] [%] Appearance EOP18a 54 IPE Classic 44 01:36 55.85 Precipitate (3 × 3 × 3 × 2 × 1) slightly sticky EOP19a 18 Et.sub.2O classic 42 00:47 52.52 precipitate (3 × 2 × 3 × 1 × 1) sticky EOP20a 72 MTBE classic 43 01:18 52.2  precipitate (3 × 2 × 3 × 2 × 2) slightly sticky EOP5i 18 IPE/ACN classic  1 00:31 74.29 Very fine (3 × 3 × 1 × 1 × 2) 1:1 particles, small sticky lumps EOP23a 108 IPE/ACN classic 36  1:47 60.52 Precipitate (3 × 3 × 2 × 3 × 2) 3:1 not sticky EOP22a 162 IPE/ACN classic 39 01:58 56.08 Precipitate (3 × 3 × 3 × 3 × 2) 5:1 not sticky EOP21a 18 IPE/ACN classic 42 05:13 57.12 Precipitate (2 × 3 × 3 × 1 × 1) 10:1 sticky EOP3a 36 IPE/hexane inverse 34 70:00:00 68.1  Fine (1 × 3 × 2 × 3 × 2) 1:2 precipitate EOP28a 1 IPE/hexane classic n/a n/a n/a Oiling out (1 × 1 × 1 × 1 × 1) 1:2 EOP24a 36 IPE/hexane classic 41 01:00 52.53 sticky (3 × 2 × 3 × 1 × 2) 5:1

Example 3: Composition of Peptide Solution

(17) A further experiment was carried out as in Example 2 above, but using a cleavage composition composed of TFA/thioanisole/anisole/EDT (90:5:3:2) v/v/v/v, which is disclosed in EP-A 2 757 107 for the preparation of Liraglutide.

(18) The observed performance score of 108 (3×2×3×3×2) illustrates that the advantageous effect of the IPE/ACN anti-solvent is not limited to one specific peptide containing solution.

Example 4: Precipitation Temperature

(19) Further experiments were carried out as in Example 2 above to analyze the preferred temperature range for the precipitation.

(20) TABLE-US-00009 TABLE 3 Precipitation temperature Filtration HPLC EXPERIMENT Anti- Temperature Yield Time Purity No. Solvent [° C.] [%] [min:sec] [%] Appearance EOP6a IPE/ACN 3:1  0 to 6° C. 28 5:55 64.84 Precipitate not sticky EOP7a IPE/ACN 3:1 −10 to −3° C.  37 4:51 60.52 Precipitate not sticky EOP8a IPE/ACN 3:1 10 to 17° C. 32 3:11 60.52 Precipitate not sticky

Example 5: Comparative Example

(21) Liraglutide was cleaved off the resin by incubation with TFA/thioanisole/anisole/EDT (90:5:3:2) v/v/v/v and precipitated using ice-cooled diethyl ether as is described in EP-A 2 757 107. The precipitate was analyzed as in Example 2 above.

(22) TABLE-US-00010 TABLE 4 Performance of comparative precipitation protocol HPLC EXPERIMENT Performance Yield Filtration Purity No. Score [%] time [min] [%] Appearance EOP26a 18 38 22:30 44.89 Fine (2 × 1 × Precipitate 3 × 3 × 1)

(23) Representative suspensions obtained from Example 5 (Experiment No. EOP26a), and EOP22a of Example 2 were analyzed by Focused Beam Reflectance Measurement (FBRM) technology using a ParticleTrack G600L device. FBRM technology allows determination of the particle size distribution after precipitation based on the detection of backscattered laser light.

(24) As shown in FIG. 2 it was found that the precipitate of the present invention was characterized by a shift in its particle size distribution to larger size. The mean particle size was 422 μm for the precipitate produced according to the present invention as compared to a mean particle size of 261 μm for the precipitate produced according to the protocol of EP-A 2 757 107. This translates to a significant difference in filtration rates with much faster filtration obtained by the present invention. Hence, it is concluded that the precipitates obtained according to the present invention differ from those disclosed in the prior art not only by their improved purity and macroscopic properties, but also in terms of their size distribution.

Example 6: Composition of Peptide Precipitate

(25) A crude liraglutide peptide precipitate obtained as in Example 2 above was analyzed by analytical reversed phase UHPLC and mass spectrometry for truncated Liraglutide variants. N-terminally truncated Liraglutide variants could be