"METHOD FOR A SOLID PHASE SYNTHESIS OF A LINEAR COMBINATION OF AMINO ACID RESIDUES" TECHNICAL FIELD
The present invention relates to a method for a solid phase synthesis of a linear combination of amino acid residues, linked via peptide bonds, starting with an amino acid residue covalently linked to a support and protected by an N-alpha-amino protecting group, comprising the following steps: (a) removing the N-alpha-amino protecting group to obtain an N-alpha-amino group,
(b) adding an amino acid residue protected by an N-alpha-amino protecting group, via a peptide bond, to the N-alpha-amino group obtained in step (a) by use of a reactive protected amino acid derivative and, where necessary, a catalyst,
(c) repeating steps (a) and (b) until the said linear combination has been obtained.
BACKGROUND ART Traditionally, the above solid phase peptide synthesis has been carried out using t-butoxycarbonyl (Boc) amino-acids activated in situ with an equivalent amount of dicyclohexyl-carbodiimide. A significant advance was the introduction of preformed Boc and fluorenylmethoxycarbonyl (Fmoc) amino-acid anhydrides in both polystyrene and polyamide-based solid phase synthesis, avoiding contact of the reactive resin-bound amino group with the activating reagent. Acylation reactions are rapid, especially in polar media such as dimethylformamide.
Activated esters (particularly p-nitrophenyl or trichlorophenyl derivatives) have also been used from time to time in solid phase synthesis but reaction rates may be low even in the presence of catalysts. Again polar reaction media are to be preferred. This last consideration is particularly relevant in polyamide-based synthesis since this resin support is totally compatible
with a wide range of aprotic polar organic solvents. SUMMARY OF THE INVENTION
The object of the present invention is to seek alternative activated Fmoc-amino-acid derivatives which would combine the high reaction rates and freedom from side reactions achievable with symmetric anhydrides with the crystallinity, stability, and ease of handling of activated esters in order to avoid manual pre-activation at each step in the synthesis and substantially simplify mechanisation.
This is attained by the method according to the invention in that the said reactive protected amino acid derivative has the acyl group used to form the peptide bond activated as a pentafluorophenyl ester. BRIEF DESCRIPTION OF DRAWING
The single figure is an HPLC diagram of a peptide assembled in accordance with the method of this invention.
DETAILED DESCRIPTION For a test case the very difficult acyl carrier protein 65-74 sequence
H. Val.Gln.Ala.Ala.Ile.Asp.Tyr.Ile.Asn.Gly.OH 1 10
was selected. Earlier attempts to assemble this sequence using p-nitrophenyl esters in the presence of catalyst 1-hydroxybenzotriazole were quite unsuccessful, although excellent syntheses were achieved using symmetrical anhydrides.
Fmoc-amino-acid pentafluorophenyl esters were prepared following the procedures of Kisfaludy and Schoen in Synthesis, 1983, 325.
The polydimethylacrylamide resin
was functicnalised with an internal reference norleucine residue and with the acid-labile p-alkoxybenzyl alcohol linkage agent as known per se.
Esterification of the C-terminal Fmoc-glycine residue utilised the preformed symmetric anhydride with dimethylaminopyridine catalysis.
The progress of all subsequent acylation reactions was monitored by sensitive ninhydrin and trinitrobenzene-sulphonic acid colour tests for residual amine and by later amino-acid analysis.
Fmoc-asparagine pentafluorophenyl ester (five-fold escess in dimethylformamide) had reacted completely with deprotected glycyl resin at the time of the first colour test (25 min). The reaction was allowed to continue for a total of 50 min. Sterically hindered Fmoc-isolueucine pentafluororphenyl ester reacted with the resulting asparaginyl-glycyl resin much more sluggishly. Positive colour tests were obtained after 49 min. After 60 min 1-hydroxybenzotriazole catalyst was added. Only very faint colour tests were obtained after an additional 45 min and the reaction was terminated after a total reaction period of three hours. O-t-butyl-Fmoc-tyrosine pentafluorophenyl ester reacted completely within 25 min. This reactivity pattern was confirmed with the succeeding residues. Complete acylation was indicated at the first colour test for t-butyl aspartate-6 (10 min), alanine-4 (13 min), and alanine-3 (5 min). Isoleucine-5 again required catalysis by 1-hydroxybenzotriazole which was added after 43 min and gave complete reaction after an additional 50 min. An anomalous result was obtained after addition of Fmoc-glutamine-2 pentafluorophenyl ester with very occasional resin beads giving positive colour tests for residual amine in a generally colourless bulk background. Some shrinkage of the resin also occurred at this stage. Reaction was complete 35 min after the addition of catalyst. The terminal valine residue was nearly
complete after 25 min and was left overnight without catalysis.
The completed decapeptide was cleaved from the resin with 95% trifluoroacetic acid. Residual resin analysis showed cleavage to be 96% complete. No significant peptide was lost from the resin during the course of the synthesis. The isolated yield of crude decapeptide determined by quantitative amino-acid analysis was 91%
(Found: Gly, 1.00; Asp, 1.89; lie, 1.80; Tyr , 0.93; Ala, 1.90; Gin, 0.98; Val, 0.97).
Analytical HPLC of total crude decapeptide on
Aquapore RP-300 is shown in the figure. Reservoir A contained 0.1% aq. trifluoroacetic acid; B contained 90% acetonitrile, 10% A. After 2 min isocratic elution with 5% B, a linear gradient of 5-60% B was developed over 40 min at a flow rate of 1.5 ml/min. On μ-Bondapack C18, reservoir A contained 0.01 M ammonium acetate, pH 4.5; B, 90% acetonitrile, 10% A. After 2 min at 15% B, a linear gradient of 15-35% over 40 min eluted the decapeptide at 11.5 min.
The retention time of the principal peak is identical with that of previous preparations.
From this it can be concluded that Fmoc-amino-acid pentafluorophenyl esters are valuable alternatives to symmetrical anhydrides for solid phase peptide synthesis. Their crystallinity and apparent stability greatly simplify and speed the conduct of solid phase synthesis. Reaction rates are substantially accelerated by catalyst 1-hydroxybenzotriazole and appear likely to be adequate for most sequences. The high purity (> 90%) of the crude reaction product obtained above shows that serious side reactions are not induced by the reagent.