EP0073828A1 - High performance peptide synthesis - Google Patents

Abstract

Method for synthesizing a polypeptide chain on an insoluble solid support, in which an amino acid is introduced into a reactor filled with an insoluble solid support containing substituent groups reacting with this amino acid, the method consisting in coupling this amino acid to the support by a condensation reaction with the reactive substituent groups, to introduce a second analogous or different amino acid into the reactor and to couple the second amino acid to the first amino acid and to repeat the process until the desired polypeptide is obtained, the improvement being constituted by the use of a pressurized flow system making it possible to obtain rapid reactions and high production yields of the desired products.

Description

HIGH PERFORMANCE PEPTIDE SYNTHESIS

Technical Field

This invention relates to the synthesis of peptides, polypeptides and proteins. More particularly, the invention relates to an improvement in peptide, polypeptide and protein synthesis wherein the peptide, polypeptide or protein is "grown" on an insoluble support or carrier by a series of stepwise coupling reactions. Background Art

Classically, sequential polypeptides have been prepared by extremely laborious techniques wherein the intermediates have been isolated after the addition of each amino acid moiety. This has made the synthesis complicated and the preparation of long chain polypeptides nearly impossible because of low yields and/or racemization or other side reactions. In 1963, Merrifield (J. Amer. Chem. Soc., 85, 2149) and Letsinger and Kornet (J. Amer. Chem. Soc., 85, 2045) suggested the use of an insoluble polymer support for the growing peptide chain. This process permitted the separation of the growing peptide chain, prepared by classical synthetic methods, from reagents and by-products, without isolating the intermediate. In such a process the insoluble support is provided a reactive substituent group as by chloromethylation, carboxylation, hydroxymethylation, etc. and a protected amino acid coupled thereto either via the amino or carboxyl group. Then using a series of deprotection and coupling reactions the peptide is synthesized in a stepwise manner on the insoluble support. Although in principle the peptide may be assembled either from the amine or the carboxyl terminus, in practice the latter is preferred, mainly because of lower extents of racemization when Nα-urethane-protected, carboxyl activated amino acids are used for coupling reactions. After each deprotection step the insoluble support and its appended peptide chain are neutralized if necessary and washed before the addition of the next amino acid residue. Finally, the polypeptide is removed from the solid support by use of a suitable cleaving reagent, and any necessary deprotections completed. The final peptide is then subjected to extensive purification.

Although the solid-phase method of peptide synthesis, originally developed by Merrifield, has revolutionized the field of synthetic peptide chemistry, the nethod is complicated by many problems, especially when applied to large peptides ( > 10 amino acides). For example, the reactive sites on the insoluble supports employed in the synthesis are located on and within the support at varying degrees or depths of accessibility. Because of shrinkage and swelling of the resin during a normal synthetic procedure, the number of "inaccessible sites" may change throughout the procedure. This means that a reactive site which is inaccessible during one particular cycle may in practice become accessible during a subsequent cycle. Because of the heterogeneous nature of the reaction sites on the support and their variability with cycles or steps, reactions in the Merrifield synthetic procedure are often incomplete and unpredictable even though a substantial excess of the amino acid reactant is used. When very long polypeptide chains are synthesized this failure to obtain 100% reaction during every step of the synthesis gives rise to a large variety of "contaminant" polypeptides with "failure sequences" (i.e. sequences containing deletion of one or more amino acid residues). Since these "contaminant" polypeptides are similar to the desired product they are difficult or impossible to separate and reduce the yield of desired peptide.

Another drawback of the Merrifield method is that long reaction times are ordinarily required to complete the coupling reaction. Depending on the amino acid reactant used, the addition of a single residue to the sequence may take anywhere from 3 to 24 hours or more. Consequently, it is difficult to do more than 2 or 3 couplings during a 24 hour period. These long coupling times may also lead to many undesirable side reactions which, after many coupling reactions, result in a marked decrease in yields and purity of the desired peptide. These problems generally tend to limit the usefulness of the Merrifield synthesis to peptides with chainlengths of about 10 to 15 amino acids (more or less, depending on the amino acid sequence). Scale-up of the conventional Merrifield method has also proved to be difficult and problematical. Disclosure of the Invention

It is an object of the invention therefore to provide a method for the synthesis of polypeptides and proteins which overcomes the aforementioned problems associated with the Merrifield method or approach.

More specifically, it is an object of the invention to provide a method for the synthesis of polypeptides and proteins which offers extraordinarily rapid reaction rates compared to conventional solid-phase reactions.

Yet another object of the invention is to provide an automatable method wherein long chain polypeptides of greater than 20 amino acids can be prepared on a large scale, with little, if any, side reactions.

A further object of the invention is to provide a method wherein coupling reactions are complete at every stage of the synthesis regardless of the sequence or structure of the peptide so that the polypeptide or protein product of desired chain length or size requires minimum purification or in many instances no purification after cleavage from the support.

These objects are obtained by an improvement in the method for the synthesis of polypeptide or protein chains on an insoluble solid support wherein an amino acid is passed in a continuous flow through a reactor packed with an insoluble solid support containing substituent groups reactive with said amino acid and coupled to said support by condensation reaction with said reactive substituent groups, a second similar or dissimilar amino acid is passed in a continuous flow through said reactor and coupled to said first amino acid and the process repeated until the desired polypeptide is obtained, which improvement comprises conducting the reaction in a pressurized flow reactor. By the term "pressurized" as used herein and in the appended claims is meant under a pressure of at least ambient pressure plus 25 psi or 20.0003625 dyne/cm².

Preferably the flow reactor is pressurized to at least 100 psi or 0.00145 dyne/cm ², up to 1,000 psi or 0-0145 dyne/cm² although pressures of up to 10,000 psi or 0.145 dyne/cm ² or more may be used. In general, the flow rate of the reactants and reagents employed will fall in the range of 4 to 50 ml per minute or more depending on the size of the reactor used.

In a preferred aspect of the invention the first amino acid passed through the reactor packed with the insoluble support containing substituent groups reactive with said amino acid, is a protected amino acid. In accordance with this aspect of the invention a protected amino acid is coupled to said support by condensation reaction with said substituent groups, said coupled first amino acid is deprotected by passing a deprotecting agent through the flow reactor, a second similar or dissimilar protected amino acid is passed through said reactor and coupled to said first coupled amino acid, said coupled second amino acid is deprotected and the process repeated until the desired polypeptide is obtained, the improvement being that during said synthesis the reactants and reagents are passed through the reactor in a continuous flow and the reaction pressure in said reactor is at the defined elevated pressures in order to obtain rapid reactions and quantitative yields of the desired products. It has been surprisingly found that conducting the steps of the synthesis in such a pressurized flow system ensures complete coupling reactions between the amino acids introduced and all of the available reactive sites on or within the insoluble support within a greatly reduced time period which is largely independent of the sequence of the polypeptide or protein. Similarly, it has been found that 100% deprotection is effected when this aspect of the invention is employed. In addition, any wash or neutralization operations that may be employed are greatly facilitated. As a consequence, formation of by-product polypeptides, that is, polypeptides of shorter chain lengths or polypeptides formed as a result of rearrangement of the activated amino acids or other side reactions is completely eliminated. Quantitative yields of desired longchain polypeptides and proteins are therefore made possible and the separation or purification problems that ordinarily plague conventional solid-phase synthesis in this regard are overcome. Also, the reaction rates achieved under the pressurized flow system of the invention compared to conventional solid-phase polypeptide coupling reactions are unexpectedly rapid. For example, coupling reactions which typically take, hours by the conventional procedures are completed in minutes. Such rapid rates of reaction result in the complete elimination of undesirable side reactions that occur because of the tendency of activated amino acids to decompose or rearrange during the long reaction times required for conventional synthetic procedures.

In a similar manner, a pressurized flow reactor may be used to increase rates and efficiencies of reactions in sequencing of polypeptides, proteins and nucleic acids. The use of a pressurized flow reactor in accordance with the teachings of the invention results in increased rates of reaction and also drives the reaction to completion, the both of which combine to result in rapid synthesis of pure materials in complicated, repetitive, sequential synthetic procedures. Modes for Carrying Out the Invention

It should be understood that the term "polypeptides" as used in the specification and the appended claims is meant to include peptides and proteins.

In the synthesis of the present invention, an insoluble solid support or matrix, advantageously in bead form such as any of the conventional solid-phase polymeric substrates conventionally employed for the synthesis of polypeptides can be utilized. Typical of such polymeric resins are crosslinked polystyrene resins, crosslinked polyacrylamides, glass beads, clays, celite, crosslinked dextran, and similar insoluble solid supports which either naturally contain reactive sites for coupling with the amino acid components or which can be provided with such reactive sites. Insoluble supports particularly preferred are derivatized, crosslinked polystyrene resins, such as chloromethylated or hydroxymethylated, crosslinked polystyrene resins, benzhydrylamine resins and the like. The crosslinked polystyrenes are normally copolymers of styrene and a crosslinking agent preferably formed by way of a pearl or bead polymerization process using an aqueous suspension system. Preferred crosslinking agents for preparation of the crosslinked polystyrene resins include divinyl compounds such as para-divinylbenzene, meta-divinylbenzene, divinylcyclohexane, butadiene, and the like. Ordinarily, use of resin supports containing higher than normal levels of crosslinking (e.g. 2% rather than 1% crosslinked polystyrene) or even non-swellable macroporous resins is preferred, since swelling of the resin and consequent variability in the reactivity of reactive sites is minimized. For this reason, the use of a single solvent system with relatively low concentrations (1-20% by volume) of reagents is generally preferred. The use of high pressures in the preferred aspect of this invention also minimizes swelling effects arising from the increase in the ratio of peptide support as the synthesis of the peptide proceeds, which can create problems, especially for larger peptides.

Several preliminary operations are necessary before the synthesis of a peptide can be started. First, the supporting resin containing the C-terminal amino acid of the proposed peptide chain must be prepared. This can be accomplished in the same reactor and under the pressurized flow conditions of the present invention. For example, hydroxymethylated resins may be esterified in the flow reactor by esterification with a protected amino acid mixed or symmetrical anhydride, catalyzed by 4-dimethylaminopyridine. However, since the attachment of the first (i.e. C-terminal) protected amino acid residue involves the formation of an ester linkage and therefore involves a relatively longer reaction time than the subsequent amino acid coupling reactions, it is often preferred to complete the first attachment beforehand and store the C- terminal amino acid-substituted resin until needed. The insoluble solid support containing the protected

C-terminal amino acid may be prepared for instance by esterifying a suitably protected amino acid with the reactive site or substituent group on the insoluble support such as chloromethy1ated or hydroxymethylated crosslinked polystyrene resins. The esterification reaction is accomplished directly with the chloromethylated resin or via suitable activation of the protected amino acid in the case of the hydroxymethylated resin. There are a considerable number of protecting groups for terminal reactive amino groups which have been employed in peptide synthesis. The protecting groups of choice in prior art syntheses have been the benzyloxycarbonyl and especially the t-butyl-oxycarbonyl groups but, because of the extremely strong acidic conditions required for their removal, more labile groups are preferred for high performance peptide synthesis. In order to minimize the formation of contaminants in the desired product careful choice of protecting groups and method of attachment of the first amino acid to the resin support are recommended. For example, it is important that sidechain protecting groups, as well as the linkage between the resin and the growing peptide chain be completely stable to the various conditions used throughout the synthesis. However, these protecting groups and the linkage to the resin should be sufficiently labile so that final deprotection of the peptide and its cleavage from the support may be accomplished under the mildest possible conditions without decomposition of sensitive amino acids or destruction of the peptide. Conventional solid-phase techniques use Nα-t-butyloxycarbonyl protection in conjunction with benzyl sidechain protection and attachment to the resin. t-Butyloxycarbonyl group removal requires relatively strongly acidic conditions (e.g. 50% trifluoroacetic acid:dichloromethane), while final deprotection and cleavage from the resin requires even more strongly acidic conditions (e.g. anhydrous hydrogen fluoride or hydrogen bromide/acetic acid), which can lead to extensive decomposition of the final product. Furthermore, the lack of complete selectivity during Nα-deprotection can lead to partial cleavage of the growing peptide chain and consequent reduction in yield, as well as the possibility of contamination by truncated sequences. Termination by trifluoroacetylation of the amine terminus during deprotection with trifluoroacetic acid has also been shown to be a serious side reaction in this synthetic scheme. Thus, while t-butyloxycarbonyl and benzyloxycarbonyl protecting groups can.be used in the present invention, it is preferred that Nα-protecting groups which are removed under the mildest possible conditions be used. Illustrative of such groups are biphenylisopropyloxycarbonyl, adamantylisopropyloxycarbonyl and nitrophenylsulfenyl groups which are removable under mild, acidic conditions, or the 9-fluorenyImetHyIoxycarbonyl or trifluoroacetyl groups which are cleaved under mild, basic conditions. For resin attachment, resins of the p-alkoxybenzyl alcohol type as [proposed by Wang, J. Amer. Chem. Soc. 95, 1328 (1973)], which are cleaved under relatively mild, acidic conditions (e.g. 50:50 trifluoroacetic acid: dichloromethane) are superior. A preferred combination for the invention utilizes Nα-fluorenylmethyloxycarbonyl in conjunction with t-butyl sidechain protection and the p-alkoxybenzyl alcohol resin. This results in complete selectivity, with Nα-deprotection under basic conditions and final cleavage and sidechain deprotection under acidic conditions. Use of this combination also reduces the danger of cleavage of the growing peptide chain from the resin during the synthesis and minimizes decomposition during final deprotection and cleavage of the product. Furthermore, coupling cycles are shortened substantially since no neutralization is required after the deprotection step.

After the first C-terminal amino acid is coupled to the support, the resulting product is commonly analyzed using standard procedures such as spectrophotometric or quantitative amino acid analysis to determine the amino acid content for the purpose of calculating the amounts of subsequent amino acid reactants and deprotecting agents to be used in the synthesis.

The thus-prepared C-terminal amino acid-containing, support is packed into a suitable continuous flow reactor. The reactor may take any desired shape or form so long as it is capable of withstanding the elevated pressures under which the synthesis of the present invention may be conducted. The preferred reactors, however, are column reactors having an inlet and an outlet so as to maximize the contact time between solvent or reactants and the growing peptide chains on the resin, thus maximizing the efficiency of the process.

The remaining synthesis to form the desired polypeptide sequence is carried out as follows. Before coupling of the second amino acid residue can take place, the first residue already on the support must be deprotected. Deprotection of the first amino acid residue on the resin as well as on each of the subsequently coupled amino acid residues can be carried out by pumping through the reactor an appropriate deprotecting agent. The deprotecting agents employed for this purpose are well known to those of ordinary skill in the peptide synthesis art and the particular deprotecting agent employed in any given instance will depend of course upon the protecting group on the amino acid/resin. If the protecting group is t-butyloxycarbonyl, trifluoroacetic acid, methanesulfonic acid or hydrochloric acid in a suitable solvent such as dioxane or dichloromethane may be used. On the other hand, if the protecting group is biphenylisopropyloxycarbonyl, mild acidic solvolysis (for example with 1% trifluoroacetic acid in dichloromethane) is the preferred method of deprotection. When the Nα-protecting group is 9-fluorenylmethyloxycarbonyl the preferred deprotecting agent is piperidine in dimethyformamide (DMF). The concentrations of the particular deprotecting agent in the solvent will vary depending again upon the particular protecting agent employed but will ordinarily range from about 5 to 50% by volume. A sufficient volume of deprotecting agent is pumped through the reactor over a period of time sufficient to effect complete removal of the protecting groups. Frequently, multiple reactor volumes (for example 2-10 column volumes over a period of 5-30 minutes) of the solution containing the deprotecting agent are passed through the reactor to ensure complete removal of the protecting group. After the deprotecting step, the resin is washed with a suitable solvent, normally the solvent in which the deprotecting agent was dissolved, in order to remove excess deprotecting agent. If the deprotecting agent is an acid the peptide on the resin must be neutralized by washing with an appropriate base such as triethylamine in a solvent such as dichloromethane. Any excess triethylamine and triethylammonium chloride, or trifluoroacetate, formed may be removed by repeated washings with a suitable solvent such as dichloromethane or dimethylformamide. The free α-amino group, thus prepared, is now ready for coupling with the next protected amino acid.

The next, Nα-protected amino acid is first activated, that is, converted into a reactive form, for instance, by converting the amino acid into an active ester or anhydride or by activation with dicychlohexylcarbodiimide, carbonyldiimidazole or other activating agents. In the high performance peptide synthesis method of the present invention the most preferred activated amino acid derivatives are the mixed anhydride, symmetrical anhydride or active ester derivatives (such as the p-nitrophenyl esters catalyzed by 1-hydroxybenzotriazole). These activated amino acids are preferred because of the short coupling times, quantitative yields and minimization or complete elimination of side reactions they provide. A solution of the activated protected second amino is then passed into and through the reactor packed with the support now containing an unprotected C-terminal amino acid. In general, an excess of the activated, protected amino acid per equivalent of the first amino acid on the resin is employed although the excess required may be limited by recycling solutions through the column. Again the common practice is to pass more than one reactor volume of the activated, protected amino acid through the reactor to ensure complete reaction. It should be understood that any of the conventional methods of activating amino acids for the purpose of coupling with another amino acid may be applied to this method. These procedures should be well-known to anyone skilled in the art of peptide chemistry.

After the coupling of the second protected amino acid to the first amino acid, the attached protected amino acid is then deprotected, neutralized (if necessary) and washed as described above before coupling of the next amino acid derivative is effected. This procedure is repeated until the desired sequence of amino acids has been assembled on the insoluble support.

While the above described methods progress from the carboxyl terminus end toward the amine terminus end of the peptide, it should be understood that the reverse direction of synthesis, that is, from amine terminus to carboxyl terminus can be employed. For instance, a prepared or selected insoluble support having sites such as alkyloxycarbonyl or aryloxycarbonyl chloride groups, activated carboxyl groups, etc., reactive with the amine group of an amino acid can be used. In practice synthesis of a peptide from its carboxyl terminus is generally preferred, mainly because of lower extents of racemization when Nα-urethane-protected, carboxyl activated amino acids are used for coupling reactions. Resins of the p-alkoxybenzyl alcohol type are preferred because of the relatively mild cleavage conditions (for example 50% trifluoroacetic acid: dichloromethane) which are required. These mild cleavage conditions minimize or eliminate completely the destruction of sensitive amino acids during the cleavage process. The entire series of coupling reactions in the abovedescribed methods, from the second amino acid, (and if desired the first amino acid) to the last is conducted in a pressurized reactor, preferably under a reactor pressure of at least 40 psi or 0.00058 dyne/cm up to 1,000 psi or 0.0145 dyne/cm², although pressures of up to 10,000 psi or 0.145 dyne/cm² or more may be used.

Reactor pressures of this level can be generated by use of commercial pressurizing equipment and methods. For example, any of the commercially available reciprocating pumps capable of generating the required pressures and flow rates can be used and the reactants, reagents and wash solvents pumped directly into and through the reactor. Alternatively, the reactants, reagents and wash solvents may be pumped through the reactor by means of pressurization with an inert gas such as nitrogen and the pressure in the reactor regulated by controlling the volume of inert gas released to transfer the reactants, reagents and wash materials into and through the reactor. The scale of the peptide syntheses in the pressurized flow reactor may vary widely, being limited solely by the capacity of the pressurized pun-ping means utilized. For example, reactors in the range 5-100 ml in volume may be used for the production of up to tens of grams of peptides, although larger reactors (up to 1000 ml or more) may be used if necessary.

The completed peptide sequence can be removed from the insoluble support by any of the standard methods as, for instance, by cleavage with anhydrous hydrogen fluoride, transesterification, acidolysis, aminolysis, etc. This cleavage is most conveniently accomplished by extruding the peptide-support from the reactor and treating it with the cleavage reagent at atmospheric pressure. However, the cleavage may also be carried out in the reactor and under the elevated reactor pressures of the invention provided the reactor is resistant to the cleavage reagent.

After cleavage, the resulting peptide is found to be remarkably homogeneous and to require no or minimal purification. Because of the very low contamination of by-products overall yields are found to be surprisingly high and whatever purification is necessary can be carried out with relative ease. Such purifications, if required, are preferably carried out by partition chromatography, ion exchange chromatography, gel permeation chromatography, countercurrent distribution and the like.

Illustrations of peptides, polypeptides and proteins which can be obtained by the method of the invention are enkephalins, angiotensin, oxytocin, vasopressin, leuteinizing hormone releasing hormone, somatostatin, gastrin, insulin, glucagon, ribonuclease, endorphins, etc.

The following examples are included to further illustrate the present invention. In all of the examples, the nitrogen-pres-surized system described in Example 2 of United States Patent No. 4,192,798, incorporated herein by reference, was employed.

Example 1 : Synthesis of L-Leucyl-L-alanyl-glycyl-L-valine using 9-Fluorenylmethyloxycarbonyl (Fmoc)- protected Amino Acid Mixed Anhydrides p-Benzyloxybenzyl alcohol polystyrene resin (2% crosslinked) was esterified with Fmoc-L-valine by the method of Wang [J. Amer. Chem. Soc. 95, 1328 (1973)] to give a substitution level of 0.6 moles/gram as determined by the spectrophotometric method of Meienhofer et al. [Int. J. Pept. Prot. Res. 13, 35 (1979)]. The resin (1.2 g) was swollen in dry DMF, packed into a stainless steel column reactor (1.0 × 5 cm) and treated with 10% (w.v) peperidine in D-MF at a flow-rate of 5 ml/minute for 10 minutes. The column reactor was washed to neutral pH with DIM-F at a flow-rate of approximately 20 ml/minute over a 5 minute period before treating with a solution of Fmoc-glycine mixed anhydride [prepared by treating a solution of Fmocglycine (3.5 mmoles) and N-methylmorpholine (3.5 mmoles) in DMF (40 ml) with isobutyl chloroformate (3.2 moles)] over a 10 minute period. The cleavage, washing and coupling cycles were repeated using equivalent amounts of Fmoc-L- alanine and Boc-L-leucine mixed anhydrides. The synthesis was conducted under 50 psi nitrogen pressure. After completion of the synthesis the Boc-tetrapeptide resin was transferred from the column reactor to a sintered funnel where it was washed copiously with DMF, methanol and dichloromethane and then suspended in 75% trifluoroacetic acid in dichloromethane (25 ml) at room temperature. After stirring for 30 minutes, the resin was filtered and washed with methanol. The combined filtrates were evaporated and the residue washed with ether to give a quantitative yield (0.34 g) of the tetrapeptide. Analysis of the product by thin-layer and high performance liquid chromatography showed the product to be homogeneous ( 99.9% pure) with no detectable contamination by "failure sequences" or "wrong-opening" products.

Example 2: Synthesis of L-Leucyl-L-alanyl-glycyl-L-valine using Fmoc-protected Amino Acid Symmetrical Anhydrides The synthesis described under Example 1 was repeated using 1 gram of Fmoc-valine resin (substitution = 0.52 mmoles/gram) and preformed symmetrical anhydrides (2.3 equivalents) for activation. The symmetrical anhydrides were prepared by treating the appropriate Fmoc-amino acid (2.4 mmoles) with dicyclohexylcarbodiimide (1.2 mmoles) in DMF (12 ml). The solution was filtered to remove dicyclohexylurea prior to use. Couplings were allowed to proceed for 10 minutes. Deprotection, cleavage and isolation of the product as described under Example I gave 0.242 g of the tetrapeptide (99%) yield) which was homogeneous ( 99% pure) by chromatographic analysis.

Example 3 : Synthesis of L-Leucyl-L-alanyl-glycyl-L-valine using Fmoc-protected Amino Acid p-Nitrophenyl Esters The synthesis described under Example 2 was repeated using p-nitrophenyl esters catalyzed by 1-hydroxybenzotriazole for activation. Couplings were allowed to proceed for 15 minutes using solutions of the appropriate Fmoc-amino acid p-nitrophenyl esters (1.2 mmoles, 2.3 equivalents) and 1-hydroxybenzotriazole (1.2 mmoles) in DMF (12 ml). Deprotection and cleavage of the product as described under Example ϊ gave 0.24 g of the product (98% yield) which contained only traces (<2%) of contaminant peptides on chromatographic analysis. Example 4: Synthesis of L-Leucyl-L-alanyl-glycyl-L- valine using Biphenylisopropyloxycarbonyl (Bpoc) - protected Amino Acid Mixed Anhydrides p-Benzyloxybenzyl alcohol polystyrene resin (2% crosslinked) was esterified with Bpoc-L-valine by the method of Wang, to give a substitution level, as determined by nitrogen analysis, of 0.49 mmoles/gram. This resin (1.2 g) was swollen in dichloromethane and packed into a stainless steel column reactor (1.0 × 5 cm). Cleavage of the Bpoc-group was effected by passing a 1% solution of trifluoroacetic acid in dichloromethane through the column for 10 minutes at a flow-rate of 5 ml/minute. The column was subsequently washed with dichloromethane (10 ml/min., 5 minutes), neutralized with 5% diisopropylethylamine in dichloromethane (10 ml/min., 5 minutes) and finally washed to neutrality with dichloromethane (10 ml/min., 5 minutes). Bpoc-glycine mixed anhydride (0.03 M, 80 ml) (generated as described under Example 1) was passed through the column over a 15 minute period. The resin was washed with dichloromethane (10 ml/min., 10 minutes) before repeating the total cycle using Bpoc alanine and Boc-leucine mixed anhydrides. The entire synthesis was carried out under 200 psi or 0.0029 dyne/ cm² nitrogen pressure. The tetrapeptide was isolated as described under Example 1 (0.273 g, 99% yield) and shown to be homogeneous (>99% pure) by chromatographic analysis.

Example 5 : Synthesis of L-Prolyl-L-prolyl-L-valine using Fmoc-protected Amino Acid Mixed Anhydrides Fmoc-L-valine p-benzyloxybenzyl ester polystyrene resin (1 g, subst. = 0.5 mmoles/gram) was swollen in dichloromethane and packed into a column reactor (1.0 × 5 cm) and coupled with Fmoc-proline and then with Boc-proline mixed anhydrides (8 equivalents of each) by the general procedure described under Example 1. The synthesis was carried out in dichloromethane as solvent, except for Fmoccleavage cycles which utilized 10% piperidine in DMF. The tripeptide was cleaved and deprotected as described in Example 1 and shown to contain 2% isobutyldxycarbonyl-L-prolyl-L-valine (the "wrong-opening" product) by reverse phase high performance liquid chromatography. This was a surprising result in that proline-to-proline couplings by the mixed anhydride method in solution syntheses of the prior art are known to be problematical and give substantial amounts (up to 10%) of the "wrong-opening" product [see for example, Katsoyannis and Schwartz in "Methods of Enzymology" 47, 555 (1977)].

Example 6: Synthesis of Methionine -Enkephalin (L-Tyrosylglycyl-glycyl-L-phenylalanyl-L-methionine) using Fmoc-Amino Acid Mixed Anhydrides Fmoc-L-methionine p-benzyloxybenzyl ester polystyrene resin (1.1 g, subst. = 0.59 mmoles/gram) was swollen in dimethylacetamide and packed into a stainless steel column (1.0 × 5 cm). After deprotection as described under Example 1, the mixed anhydrides (3.6 mmoles, 5.5 equivalents) of Fmoc-L-phenylalanine, Fmoc-glycine ( twice) and finally Boc-O-t-butyl-L-tyrosine were coupled as 0. 1M solutions in dimethylacetamide for 10 minutes per coupling. The entire synthesis was conducted under

1000 psi or 0.0145 dyne/cm² nitrogen pressure. The peptide-resin was removed from the reactor and cleaved by suspending in thioanisole:dimethylsulfide: trifluoroacetic acid (2:8:20) at room temperature under nitrogen for 1 hour. The resin was filtered, washed with methanol, the filtrate evaporated and the residue washed with ether to give methionine⁵-enkephalin in quantitative yield

(0.495 g). Analysis by reverse phase high performance liquid chromatography showed the product to be > 95% pure, with only trace contamination by peptide impurities. Example 7: Synthesis of Delta-Sleep Inducing Peptide (L-Tryptσphanyl-L-alanyl-glycyl-glycyl-L- aspartyl-L-alanyl-L-seryl-glycyl-L-glutamic acid) using Fmoc-protected Amino Acid Mixed Anhydrides Fmoc-α-t-butylglutamic acid p-benzyloxybenzyl ester polystyrene resin (8 g, substitution = 0.20 mmoles/gram) was swollen in DMF and packed into a stainless steel column reactor (1.5 × 15 cm). After deprotection (10 minutes) as described under Example 1 the resin was treated sequentially with the mixed anhydrides (16 mmoles/100 ml; 10 minutes) of Fmoc-glycine, Fmoc-O-t-butyl-L-serine, Fmoc-L-alanine, Fmoc-S-t-butyl-L-aspartic acid, Fmoc-glycine (twice), Fmoc-L-alanine, and finally Boc-L-tryptophan. The synthesis was carried out under 1000 psi or 0.0145 dyne/cm² nitrogen pressure. The peptide-resin was removed from the reactor and cleaved and deprotected by suspending in trifluoroacetic acid:ethanedithiol: dichloromethane (90:10:20) for 1.75 hours under nitrogen at room temperature. After filtration and evaporation of the solvent the residue was washed with ether to give delta-sleep inducing peptide (1.70 g, 100%). Analysis by reverse phase high performance liquid chromatography showed the peptide to be>95% pure, with only trace contamination by peptide impurities.

Claims

It is Claimed:

1. In the continuous flow method for the synthesis of a polypeptide chain on an insoluble solid support wherein a protected amino acid is passed in a continuous flow through a flow reactor packed with an insoluble solid support containing substituent groups reactive with said protected amino acid, and coupled to said support by condensation reaction with said reactive substituent groups, a second similar or dissimilar protected amino acid is passed through said reactor in a continuous flow and coupled to said first amino acid and the process repeated until the desired polypeptide is obtained, the improvement comprising conducting the reaction in said flow reactor under pressure.

2. The improvement of Claim 1 wherein the reaction pressure is at least about 40 psi or 0.00058 dyne/cm².

3. The improvement of Claim 1 wherein the reactive substituent groups on said insoluble support are protected amino acids.

4. The improvement of Claim 3 wherein the protected amino acids are Nα-fluorenylmethyloxycarbonyl amino acids.

5. The improvement of Claim 3 wherein the protected amino acids are Nα-biphenylisopropyloxycarbonyl amino acids.

6. The improvement of Claim 3 wherein the protected amino acids are Nα-adamantylisopropyloxycarbonyl amino acids.

7. The improvement of Claim 1 wherein said support is a crosslinked polystyrene.

8. The improvement of Claim 1 wherein said support is a crosslinked polyacrylamide.

9. The improvement of Claim 1 wherein the protected amino acids are employed as their mixed anhydrides.

10. The improvement of Claim 1 wherein the protected amino acids are employed as their symmetrical anhydrides.

11. The improvement of Claim 1 wherein the protected amino acids are employed as their active esters.

12. In the method for synthesis of a polypeptide chain on an insoluble solid support wherein a first protected amino acid is passed through a reactor packed with an insoluble solid support containing substituent groups reactive with said protected amino acid and coupled to said support by condensation reaction with said substituent groups; said amino acid is deprotected by passing through said reactor a deprotecting agent, a second similar or dissimilar protected amino acid is passed through said reactor and coupled to said first coupled amino acid, said coupled second amino acid is deprotected by passing through said reactor a deprotecting agent and repeating the process until , the desired peptide is obtained, the improvement wherein the reactants and reagents are passed through the reactor in a continuous flow under pressure.

13. The improvement of Claim 12 wherein the reaction pressure is at least 40 psi or 0.00058 dyne/cm ².

14. The improvement of Claim 13 wherein the reactive substituent groups on said insoluble support are protected amino acids.

15. The improvement of Claim 14 wherein the protected amino acids are Nα-fluorenylmethyloxycarbonyl amino acids.

16. The improvement of Claim 14 wherein the protected amino acids are Nα-biphenylisopropyloxycarbonyl amino acids.

17. The improvement of Claim 14 wherein the protected amino acids are Nα-adamantylisopropyloxycarbonyl amino acids.

18. The improvement of Claim 14 wherein the insoluble support is a crosslinked polystyrene.

19. The improvement of Claim 14 wherein the insoluble support is a crosslinked polyacrylamide.

20. The improvement of Claim 12 wherein the protected amino acids are employed as their mixed anhydrides.

21. The improvement of Claim 12 wherein the protected amino acids are employed as their symmetrical anhydrides.

22. The improvement of Claim 12 wherein the protected amino acids are employed as their active esters.