Background of the invention

1. Technical Field

The present invention relates to an improved process for the production of peptides by solid-phase synthe¬ sis. The invention also relates to an agent which is useful in solid-phase peptide synthesis.

Background Art

Solid-phase peptide synthesis (SPPS) is a highly suc¬ cessful method introduced by Merrifield in 1963 (Ref. 1) . Numerous peptides have been synthesized with this technique since then. An excellent review of the cur¬ rent chemical synthesis of peptides and proteins is provided by S.B.H. Kent (Ref. 2) which is incorpo¬ rated herein by reference.

In practice, two strategies for the assembly of pep¬ tide chains by solid-phase synthesis are used, viz. the stepwise solid-phase synthesis, and solid-phase fragment condensation.

In stepwise SPPS, the C-terminal amino acid in the form of an N-α-protected, if necessary side-chain protected reactive derivative is covalently coupled either directly or by means of a suitable linker to a "solid ^" support, e.g. a polymeric resin, which is swollen in an organic solvent. The N-α-protective group is removed, and the subsequent protected amino acids are added in a stepwise fashion.

When the desired peptide chain length has been ob _¬ tained, the side-chain protective groups are removed

and the peptide is cleaved from the resin, which might be done in separate steps or at the same time.

In solid-phase fragment condensation the target se¬ quence is assembled by consecutive condensation of fragments on a solid support using protected frag¬ ments prepared by stepwise SPPS.

Over the years, two coupling strategies have been de¬ veloped based on the use of different N-α-protective groups and matching side-chain protective groups.

Merrifield used tert . butyloxycarbonyl ( Boc ) as the N- α protective group, while 9-fluorenylmethyloxy- carbonyl (Fmoc) was introduced by Carpino and Han (Ref . 12) .

The operations involved in one cycle of chain exten- sion in stepwise SPPC using Boc- and Fmoc-chemistries are illustrated in the reaction-schemes below (taken from Ref . 2 ) .

B.

-b Boc-AA^-OO.ICH _j CONH-CH _j - RESIN t Fmoc A 0-B/l-0-CM ₂ RtjIN

Deprolecl Utp'Oleci

Tlo-O ^" NM.-AA -Oθ2lCH ₂ CONH-CM ₂ -RLSlN

eutralise

NM NH.-AΛ -O-Bϊl-O-CH - tSlN i.-AA ft -OG.ΠCM i,CONM-CM i,-RLSIN

(BocAA) _j O

Couple Couple

BocAA -OH ) Bo -A ,-AA -OB ONH-CH Fmoc-AA _n ft - i ft zlCH i,C j, -RESIN

The N-α-Boc-protected peptide coupled to a PAM-resin is N-α-deprotected with trifluoro-acetic acid (TFA) . The resulting amine salt is washed and neutralized with a tertiary amine. The subsequent peptide bond is formed by reaction with an activated Boc-amino acid, e.g. a symmetric anhydride. Generally, the side-chain protection is benzyl-based, and the deprotection is made with HF or a sulphonic acid.

The N-α-Fmoc protected peptide coupled to a resin is N-α-deprotected by treatment with a secondary amine, normally piperidine, in an organic solvent, e.g. N,N- dimethyl formamide (DMF) or dichloromethane (DCM) . After washing, the neutral peptide resin is reacted with an activated Fmoc-amino acid, e.g. a hydroxy- benzotriazole active ester.

The side-chain protection is tert . butyl, trityl and arylsulfonyl based, and for deprotection of the side- chains, preferably TFA is used.

While the Boc- and F oc-strategies are used for es- sentially all current practical peptide synthesis, other N-α protective groups have been proposed (Steward & Young, Ref .13) .

Boc forms an acid-labile urethane group, and other proposals of this category are biphenylisopropyloxy- carbonyl (Bpoc ) , 3, 5-dimethoxyphenylisopropyloxy- carbonyl (Ddz), phenylisopropyloxycarbony1 (Poc) and 2 , 3 , 5-tetramethylbenzyloxycarboxy] (Tmz) .

Other types of N-α protecting groups available in¬ clude nitrophenylsulfenyl (Nps) which can be removed by either very dilute anhydrous acid, e.g. HCl, or by nucleophilic attack, e.g. with methyl-3-nitro-4-

mercapto benzoate . Also dithiasuccinyl (Dts) , which is removable by nucleophilic attack, might be used.

SPPS has the general advantage that it lends itself to fully automated or semi-automated chain assembly chemistry. A system for SPPS under low pressure con¬ tinuous flow conditions was developed by Dryland & Sheppard (Ref. 7) and was further refined, see Cam¬ eron, Meldal & Sheppard (Ref. 14) Holm & Meldal (Ref.15) , and WO 90/02605.

While SPPS has now developed to be a cornerstone in protein and peptide synthesis, certain problems still remain to be solved. Since some of these problems might well be related to the peptide structure, a brief discussion is deemed proper.

Empirical predictions of protein conformations have been made by Chou & Fasman (Ref. 6) . It is well-known that protein architectures may be described in terms of primary, secondary, tertiary and quaternary struc¬ ture. The primary structure refers to the amino acid sequence of the protein. The secondary structure is the local spatial organization of the polymer back¬ bone without consideration of the side-chain confor¬ mation. As examples of secondary structures, α- helixes, β-sheets and β-turns, which are chain rever- sal regions consisting of tetrapeptides can be men¬ tioned. The tertiary structure is the arrangement of all the atoms in space, including disulphide bridges and side-chain positions, so that all short- and long-range interactions are considered.

The term quaternary structure may be used to denote the interaction between subunits of the protein, e.g. the α- and β-chains of hemoglobins.

Following a discussion of earlier attempts to corre¬ late protein secondary structure with amino acid com¬ positions, where e.g. Ser, Thr, Val, lie and Cys were classified as "helix breakers ^" and Ala, Leu and Glu as "helix formers ^" , while hydrophobic residues were classified as strong "β-formers ^" , and proline to¬ gether with charged amino acid residues as "β- breakers", Chou & Fasman made a statistical analysis of 29 proteins with known X-ray structure in order to establish prediction rules for α- and β-regions .

Based on these studies they determined so-called pro¬ pensity factors Pα, Pβ and Pt which are conforma- tional parameters expressing the positional prefer¬ ences as α-helix, β-sheet and β-turn, respectively, for the natural L-amino acids forming part of pro¬ teins .

For the sake of convenience, the Pα and Pβ are listed below. One-letter abbreviations for the individual amino acids are given in parenthesis.

α

Glu (E) 1.51 Val 1.70

Met M 1.45 He 1.60

Ala A 1.42 Tyr 1.47

Leu 1.21 Phe 1.38

Lys K 1.16 Trp 1.37

Phe 1.13 Leu 1.30

Gin Q 1.11 Cyr 1.19

Trp W 1.08 Thr 1.19

He 1.08 Gin 1.10

Val V 1.06 Met 1.05

Asp D 1.01 Arg 0.93

His 1.00 Asn 0.89

Arg R 0.98 His 0.87

Thr T 0.83 Ala 0.83

Ser 0.77 Ser 0.75

Cys 0.70 Gly 0.75

Tyr Y 0.69 Lys 0.74

Asn N 0.67 Pro 0.55

Pro 0.57 Asp 0.54

Gly 0.57 Glu 0.37

Generally speaking, values below 1.00 indicate that the amino acid in question must be regarded as unfa¬ vourable for the particular secondary structure.

As an example, the hydrophobic acids (e.g. Val, He, Leu) are strong β-sheet formers, while the charged amino acids (e.g. Glu, Asp, His) are β-sheet break¬ ers .

In the α-helix structure, the spiral configuration of the peptide is held rigidly in place by hydrogen bonds between the hydrogen atom attached to the ni¬ trogen atom in one repeating unit

O H

(C - C - N -)

and the oxygen atom attached to a carbon atom three units along the chain.

If a polypeptide is brought into solution, the α- helix can be made to unwind to form a random coil, by adjustment of the pH . The transition from α-helix to random coil occurs within a narrow pH . Since the hy- drogen bonds are all equivalent in bond strength in the α-helix, they tend to let go all at once. The change can also be induced by heat.

The β-sheet structure consists of fully extended pep¬ tide chains in which hydrogen bonds link the hydrogen atoms on one chain to the oxygen atoms in the adjoin¬ ing chain. Thus hydrogen bonds do not contribute to the internal organization of the chain as they do in the α-helix, but only bond chain to chain. Adjacent chains may be parallel or antiparallei .

β-turns are frequently observed in these parts of a peptide chain which connect antiparallel chains in a β-sheet structure.

In a β-turn, the CO- and NH-groups from amino acid No. n in the peptide chain form hydrogen bond to the corresponding groups in amino acid No. n+4.

α-helix and β-sheet constitute strongly variating parts of the peptide conformation of proteins (from 0 to 80 %) , and the remaining parts of the proteins are folded in other structures. In most proteins sections of the peptide chains appear as irregularly folded "random coils ^" .

Turning now to the general problems still prevailing in connection with SPPS, S.B.H. Kent (Ref. 2) high- lights the synthesis of "difficult sequences ^" .

Obviously, the whole rationale of SPPS is based on a complete N-α-deprotection prior to each of the cou¬ pling steps involved.

By the same token, ideally all of the N-α-deprotected amino groups should be coupled to the reactive amino acid derivative according to the desired sequence, i.e. a complete aminoacylation should take place.

Kent states that the most serious potential problem in stepwise SPPS is incomplete peptide bond formation giving rise to peptides with one or more amino acids missing (deletions) , but with properties similar to the target sequence.

Such incomplete couplings are more prevalent in some sequences than in others, hence the term "difficult sequences ^" , and are apparently also more predominant in Fmoc-chemistry than in Boc-chemistry .

A number of recognized "difficult sequences" have been studied by scientists including the present in¬ ventors .

During SPPS of homo-oligopeptides containing leucine or alanine using the Fmoc-strategy, ineffective N-α- deprotection with piperidine in a sequence dependent manner (Refs. 3 and 4) was observed. Investigations showed that this phenomenon was associated with sub¬ sequent slow or incomplete amino acid coupling and evidence for β-sheet aggregation of the growing pep¬ tide chain was presented as a cause for the difficult couplings and incomplete Fmoc-deprotections . This evidence was based on general physical-chemical ob¬ servations (Ref. 4) and on a detailed Raman Near In- frared Spectroscopic study (Ref. 5) . The physical- chemical observations referred to may be summarized as follows: In case of synthesis of H- (Ala ) _n -Lys-0H on a polyamide polymerized kieselguhr matrix ( PepSyn K), no problems were observed until n = 5, but ap- proximately 20 - 25% Fmoc-protected peptide was still present after standard deprotection with piperidine (20% piperidine in DMF) with n = 6. When continuing the synthesis to n = 10 a relatively complicated mix¬ ture was obtained comprising the target peptide (n = 10) as well as deletion peptides corresponding to n = 6, 7, 8, and 9 and deletion peptides with the Fmoc group still attached to the N-terminal where n = 6,7,8, and 9 respectively (Fig. 1) . This mixture was identified by FAB MS after hplc separation of the single components. Failure sequences or partial de¬ protection with n = 2 - 5 or incomplete deprotection of the target peptide (n = 10) was not observed, thus confining the problems to a given stretch of the homo-oligopeptide chain. This type of difficult ami- noacylations and incomplete deprotections can there-

1 ]

fore be referred to non-random in contrast to random difficulties which are due only to common steric problems (Ref. 2) . Although experimental conditions were varied, e.g. resin-type, deprotection time, sol- vents, addition of chaotropes, problems still per¬ sisted although heating to 50 °C had an optimal ef¬ fect on incomplete Fmoc-deprotection (Ref. 4) .

It is a most characteristic feature of the homo-oligo alanine chain, that the incomplete Fmoc-deprotection steps are followed by strikingly slow acylations with the next amino acid in the sequence. Thus effective acylation times for each of the first six alanines are less than 60 min, while complete acylation with Ala7, Alaβ; Alag, Alairj is 26, 28, 30 and 7 hours, respectively (Fig. 2) .

Kent (Ref. 2) proposes a number of solutions to the problem related to sequence-dependent coupling diffi¬ culties, viz. the use of heat in the coupling step and a quantitative conversion of residual unreacted resin-bound peptide chains to terminated species in a "capping" procedure.

Summary of the invention

The object of the present invention is to provide an improved SPPS according to which peptides, which are recognized as or prove to be "difficult sequences", can be synthesized in high yield and purity.

A further object of the invention is to provide an improved SPPS which provides for reduced coupling times, not only for difficult sequences, but also for otherwise uncomplicated sequences where it is desir¬ able to reduce the normally long coupling times.

1 ?

A still further object of the invention is to provide an agent or a kit for use in SPPS, whereby the above¬ mentioned process improvements are obtained.

The manner in which these and other objects and ad- vantages of the invention is achieved will appear more fully from the following description and accom¬ panying drawings showing hplc diagrams for various peptides obtained by solid phase synthesis.

Brief description of the drawings

Fig. 1 is an hplc of crude H-Alairj-Lys-OH showing a substantial amount of deletion peptides (peak 1, 2, 3 and 4) and incompletely Fmoc- deprotected peptides (peak 6, 7, 8 and 9) be¬ sides the target peptide (peak 5) .

Fig. 2 is a diagram showing sequential coupling times for each of the alanines in the synthesis of H-Ala ₁₀ -Lys-OH.

Fig. 3 is an hplc of H-Alaιo- ^Ij y ^s ₃ ^_OJ1 showing the tar¬ get peptide. No deletion peptides are ob- served.

Fig. 4 is an hplc of H-Ala ₂ o ^_ Ly ^s ₃ -OH showing deletion peptides (peak 1, 2, 3, 4 and 5) besides the target peptide (peak 6) .

Fig. 5 is an hplc of H-Alaιrj-Lys ₆ -OH . No deletion peptides are observed.

Fig. 6 is an hplc of H-Ala ₂₀ ^_ Lys ₆ -OH . No deletion peptides are observed.

Fig. 7 is an hplc of H-VQAAIDYING-OH, Acyl Carrier Protein (ACP) 65-74) . The target peptide (peak

3) is accompanied with deletion peptides (peak 2 is the des-Val peptide) .

Fig. 8 is an hplc of H-VQAAIDYING-K ₆ -OH, Acyl Carrier Protein (ACP) 65-74 coupled to the Lys(Boc) ₆ - pre-sequence with only a minor amount of des-

Val peptide (peak 1) .

Fig. 9 is an hplc of H-Alaιo-Ly ^s -° ^H prepared by in¬ troduction of an HMPA linker. Several deletion peptides are observed (from Ala ₁₀ -Lys ( tBOC) - HMPA-(Lys ( tBoc) ) ₆ -resin) .

Fig.10 is a similar hplc of H-Alaio-Lys-OH prepared by using an MMa-linker showing a significant reduction the amount of deletion peptides com¬ pared to Fig. 9 (from Ala ₁₀ -Lys-MMa- (Lys ( tBoc ) ) ₆ -resin) .

Detailed disclosure of the invention

To investigate the problems described above with ref¬ erence to Ref . 3 and 4 further the present inventors addressed themselves to the question to which degree the β-sheet formation of the homo oligo-alanine chain may be affected by inclusion of a shorter peptide se¬ quence, a pre-sequence, in the chain at the C- terminus . This question is associated with the fact that protein structures and polypeptide sequences may have stretches of well defined structures dependent on the amino acids in the sequence and of preceding amino acids. As mentioned earlier Chou and Fasmans investigations (Ref. 6) on protein structures have led to recognition of classes of amino acids which are defined as predominantly α-helix inducing, β- sheet or random coil inducing. Although it a priori may be assumed that similar predictions may apply for

homo oligo alanines or leucines it was pointed out, however, in Ref. 4 that Chou and Fas an rules do not predict homo oligo alanine or leucine peptides as β- sheet forming chains. Furthermore, Chou and Fasman' s rules cannot be expected to apply during peptide syn¬ thesis on a resin and do clearly not apply when side- chain protected amino acids are part of the synthe¬ sis .

In the basic studies the synthesis of H-(Ala) _n - (Lys) _m -OH, where (Lys)jη represent the pre-sequence with m equal to 1 , 3 and 6 was investigated. A con¬ tinuous-flow version of the polyamide solid-phase method (Ref. 7) on a fully automated peptide synthe¬ sizer developed was used as previously described (Ref. 14) with DMF as solvent and 3 times excess of Fmoc-alanine and Fmoc-lysine( tBoc ) -pfp esters, re¬ spectively and standard Fmoc-deprotection with 20% piperidine in DMF for 10 minutes. Coupling times were monitored with Dhbt-OH which is deprotonated to the yellow Dbht-O- anion when un-acylated amino groups are still present and disappearance of the yellow colour marks the end-point of the synthesis. After cleavage of the peptide from the resin with 95% aque¬ ous TFA, the product was washed with ether and ana- lyzed by hplc. The results with m = 1 are described above (Fig. 1) . In case of = 3, it is seen from the hplc trace shown in Fig. 3 that the synthesis may be continued to Alaio without detectable amounts of de¬ letion peptides or incomplete Fmoc-deprotection. How- ever, when continuing the synthesis to Ala20 the chromatogram (Fig. 4) shows the presence of deletion peptides . The results are even more striking with H- (Ala) _n - (Lys ) 6-OH where products without detectable deletion peptides are obtained with both Alaio (Fig. 5) and Ala20 (Fig- 6) . Furthermore, coupling times

are drastically reduced from up to 30 hours to stan¬ dard coupling times (< 2 hours) in the single steps. The H-Ala20-OH sequence has previously been attempted synthesized by the Boc-methodology but high levels of deletion and insertion peptides were obtained (Ref. 8) . Clearly the pre-sequence Lysς, which under the prevailing synthesis conditions is fully protected with the tBoc-group, has a most definitive and fa¬ vourable effect on the structure of the growing pep¬ tide chain eliminating the otherwise very severe syn¬ thetic problems due to incomplete deprotections and extremely slow couplings.

In accordance with these surprising findings the pre¬ sent invention is based on the incorporation of a particular pre-sequence in the C-terminal part at¬ tached to the solid support.

This is a fundamental breach with the prior art at¬ tempts to deal with difficult sequences, where the focus was on the reaction conditions and the nature of the solid support.

As further discussed below the C-terminal sequence by which the desired peptide is attached to the solid support might also include suitable linkers in order to provide for e.g. better attachment or cleavage conditions .

Thus in a first aspect the present invention relates to a process for the production of peptides

X-AAi-AA ₂ AA _n -Y

wherein AA is an L- or D-amino acid residue, X is hy- drogen or an amino protective group, Y is OH, NH ₂ or an amino acid sequence comprising from 3 to 9 amino

acid residues and n is an integer greater than 2 by solid phase synthesis wherein the C-terminal amino acid in the form of an N-α-protected , if necessary side chain protected reactive derivative is coupled to a solid support or a polymer optionally by means of a linker, subsequently N-α-deprotected, whereafter the subsequent amino acids forming the peptide se¬ quence are stepwise coupled or coupled as a peptide fragment in the form of suitably protected reactive derivatives or fragments, wherein the N-α-protective group is removed following formation of the desired peptide and the peptide is cleaved from the solid support, c h a r a c t e r i z e d in that the C- ter inal part attached to the support or polymer com- prises a pre-sequence comprising from 3 to 9 , pref¬ erably from 5 to 7 amino acid residues independently selected from native L-amino acids having a side chain functionality which is protected during the coupling steps and having a propensity factor Pα > 0,57 and a propensity factor Pβ > 1,10 or the corre¬ sponding D-amino acids and said pre-sequence is op¬ tionally cleaved from the formed peptide.

L-amino acids meeting the above-mentioned limits for the propensity factors Pα and Pβ are Lys, Glu, Asp, Ser, His, Asn, Arg, Met and Gin.

These amino acids all have a side chain functionality selected from a carboxy, carboxa ido, amino, hydroxy, guanidino, sulphide or imidazole group.

Presently preferred amino acids in the pre-sequence are Lys and Glu and combinations thereof, e.g. (Glu)q(Lys )p, where p + q is 3 to 9 , preferably 6 to 9, and the order of Lys and Glu is arbitrarily cho¬ sen .

The N-α amino group of the amino acids or peptide fragments used in each coupling step should be suita¬ bly protected during the coupling. The protective group may be Fmoc or Boc or any other suitable pro- tective group, e.g. those described above with refer¬ ence to Ref. 13 and 18. The presently preferred N-α protective group is Fmoc.

It is important that the side chain functionality in the pre-sequence is suitably protected during the coupling steps. Such protective groups are well-known to a person skilled in the art and the preferred groups are listed in claims 7-12.

Without wishing to be bound by any particular theory, it is assumed that the physical-chemical properties of the protected side-chain of the pre-sequence exem¬ plified by lysine are responsible for the observed "structural assisted peptide synthesis" (SAPS) by re¬ ducing or eliminating β-sheet formation in the poly alanine sequence.

In case of other homo-oligo pre-sequences it has been observed that (Glu(tBu))6, as well as the mixed se¬ quence (Glu(tBu)Lys ( tBoc ) ) 3 induces a favourable structure in the poly-alanine chain affording prod¬ ucts without deletion peptides.

To investigate whether SAPS is a more general phe¬ nomenon or if it is confined only to homo-oligo¬ peptides such as the poly alanine sequence, some mixed sequences reputedly known as difficult se¬ quences were investigated.

The synthesis of H-VQAAIDYING-OH, Acyl Carrier Pro¬ tein (ACP) 65 - 74, is a well known difficult synthe¬ sis which has been used as a model reaction in a num-

ber of cases (Ref. 9) . Deletion peptides are observed in standard synthesis with a des-Val peptide (peak 2) as a prominent by-product (see example 7 and figure

7) .

In accordance with the invention the sequence H- VQAAIDYING-K ₆ -0H was synthesized using the pre- sequence (Lys(tBoc) ) ₆ attached at the C-terminus on a pepsyn K. The synthesis proceeded to give a product with the correct molecular weight in high purity and a significantly reduced amount of des-Val peptide (see example 8 and figure 8) .

Another difficult sequence is reported to be H- VNVNVQVQVD-OH which has been synthesized on a variety of flow resins (Rapp polymer, PEG-PS, Pepsyn K, PEGA 1900/300, PEGA 800/130 and PEGA 300/130) in all cases accompanied by considerable glutamine preview already from Vail except when using a PEGA 1900/300 resin (Ref. 10) . In accordance with the invention the syn¬ thesis of H-VNVNVQVQVDK6-OH proceeded to give a prod- uct with the correct molecular weight. Deletion pep¬ tides were not detected in the spectrum (see example

9) .

In order to verify another important aspect of the present invention, viz. the reduced coupling times obtained by introducing a pre-sequence at the C- terminal part of desired peptide the coupling times of the individual amino acids in the synthesis of enkephalin, H-Tyr-Gly-Gly-Phe-Leu-OH , have been moni¬ tored with and without the pre-sequence (Lys(tBoc) )6 _as described in example 15-

The results of the measurements demonstrate that the coupling times in this otherwise uncomplicated syn-

thesis proceeds effectively in combination with the chosen pre-sequence.

In the above described cases the peptide sequences have been obtained with a hexa lysine pre-sequence which for certain purposes may be acceptable or even of advantage. Thus, a β-sheet forming sequence may cause severe solubility problems, but in case of H- Ala20( ^L y ^s ) 6-OH ^tne peptide turned out to be soluble in aqueous solutions while H-Alaιo-Lys-OH very rap- idly precipitated from TFA solutions when diluted with water. In case of (Glu) ₆ the solubility provided by the pre-sequence and the multitude of carboxylic acid groups may be utilized for e.g. ELISA, where site-directed attachment to a suitably amino-group- activated surface is possible because of facile acti¬ vation in aqueous solution with e.g. a carbodiimide . However, it has also been considered, how the obser¬ vations may be utilized for practical peptide synthe¬ sis without presence of the pre-sequence in the final product. It was therefore attempted to introduce a linker between the pre-sequence and the target pep¬ tide, e.g. resin- (Lys ( tBoc )) 6-linker-peptide , permit¬ ting cleavage of the peptide from the linker with standard reagents such as scavenger containing TFA solutions. Introduction of the commonly used HMPA linker as described below in example 10, resin- (Lys ( tBoc )) 6-HMPA-Lys ( tBoc)Alaio, has, however, a clear negative effect on the synthesis of H-Alaio- Lys-OH giving rise to formation of deletion peptides (Fig. 9) . Apparently, the effect of the pre-sequence on the poly alanine structure is lost, pointing to that the aromatic group in the linker breaks the structure effect in the peptide back-bone of (Ala) _n - (Lys)m- In order to conserve the structural effect the introduction of an optically active linker be-

tween (Ala) _n and (Lys) _m was investigated using 4- methoxymandelic acid (MMa) as a possible candidate. This type of linker has supposedly not been used hitherto. The presence of the methoxy group is as- su ed to facilitate the cleavage of the target pep¬ tide from the linker by means of standard scavenger containing TFA solutions. 4-Methoxymandelic acid may be resolved in its optically active forms of which the R-configuration ((+) -configuration) is identical to L-protein amino acids. Two types of experiments have been carried out: (a) synthesis using racemic 4- methoxymandelic acid; (b) synthesis using R- 4- methoxymandelic acid (R-MMa) . The synthesis scheme is shown below (Scheme 1) and further illustrated in ex- amples 11, 12, 13 and 14.

Scheme I

activation

(R)— CONHCH ₂ CH ₂ NH-(AA) _IU -NH ₂ + MeO— iTj— CH(OH)COOH

OMe

Fmoc-AA.-OH

@— CONHCH ₂ CH ₂ NH-(AA) _rn .NHCO.CH-OH activation

1. Fmoc-deprotection 2. Coupling with 1st

Fmoc-amino acid etc. Il ^t moc

In the first case the sequence resin- (Lys ( tBoc )) 6- MMa-Lys-Ala o was synthesized to give H-Lys-Alaio-OH where the lysine group is introduced to enhance solu¬ bility in aqueous solutions for hplc analysis. The results of the hplc analysis is shown in Fig. 10. A much better product was achieved than for the HMPA- linker (Fig. 9 and example 10) although deletion pep¬ tides are present. In case (b) the same peptide was synthesized under identical conditions and a peptide is formed without detectable deletion peptides or in¬ complete Fmoc-deprotection. These results may be com¬ pared to synthesis of H-Lys-Ala10-OH using the con¬ struct resin-MMa-Lys-Alaio where the pre-sequence Lys(tBoc))6 is omitted. Significant amounts of dele- tion peptides are noted.

In accordance with the above observations another as¬ pect of the present invention relates to a method for the solid phase synthesis of peptides as described above having the further feature that a linker is in- serted between the pre-sequence attached to the sup¬ port and the desired peptide sequence AAχ-AA _n , which enables a selective cleavage of said sequence. Pref¬ erably, the linker is optically active. An applicable group of linkers is α-hydroxy and α.-amino acids of the general formula

R ^J

COOH

R ²

wherein X is OH or NH ₂ , and R 1 and R2 are independ¬ ently selected from H, C . ₃ alkyl, phenyl and substi¬ tuted phenyl , where the substitutents are one or more electron donating substituents chosen among C ₁ . ₃ alk- oxy, Cι_ ₃ alkyl, or two vicinal substituent groups are joined to form a 5 or 6 membered ring together with the carbon atoms to which they are attached.

The most preferred linkers are racemic 4-methoxy- mandelic acid, ( + ) -4-methoxymandelic acid, diphenyl- glycin and glycolic acid.

It will be understood that when X is OH the product formed after cleavage will be a peptide AA -AA _n -OH, i.e. Y = OH, while in case of X being NH ₂ a peptide amide AA ₁ -AA _n -NH ₂ is formed, i.e. Y = NH ₂

In a still further embodiment of the invention a first linker is inserted between the pre-sequence at¬ tached to the support and the AAι-AA _n sequence and a second linker is inserted between the pre-sequence and the solid support with orthogonal cleavage condi- tions to the first linker enabling a selective cleav¬ age of the second linker, e.g. by means of trifluoro- acetic acid (TFA), trifluoromethanesulfonic acid (TF SA) , HBr, HCl, HF or a base such as ammonia, hy- drazine, alkoxide or hydroxide to give the desired peptide AAι-AA _n linked to the pre-sequence by means

of said first linker, which is then optionally cleaved from the pre-sequence

Another important embodiment of the invention relates to agents for use in solid phase synthesis incorpo- rating one or more of the above described features with relation to pre-sequences and linkers.

A first aspect of this embodiment relates to an agent for use in solid phase peptide synthesis having the general formula

X-AA' i - ...-AA' _m -Yι-R

wherein R is a solid support applicable in solid phase peptide synthesis, Yj is an amino acid sequence comprising from 3 to 9 , preferably from 5 to 7 amino acid residues independently selected from native L- amino acids having a side chain functionality which is protected during the coupling steps and having a propensity factor Pα > 0,57 and a propensity factor Pβ 1,10, e.g. Lys, Glu, Asp, Ser, His, Asn, Arg, Met or Gin, or the corresponding D-amino acid, AA' is an L- or D-amino acid residue, is zero or an inte¬ ger from 1 to 40 and X is hydrogen or an amino pro¬ tective group.

A second aspect of this embodiment relates to an agent for use in solid phase peptide synthesis having the general formula

X-AA'x-... -ΛΛ'rn-Li-Yi-R

wherein X, AA' , , Y^ and R are as defined above and Li is a preferably optically active linker which en¬ ables a selective cleavage of the bond to AA'rp.

Preferred linkers are α-hydroxy or α-amino acids as described above.

A third aspect of this embodiment relates to an agent for use in solid phase peptide synthesis having the general formula

X-AA'i-. • . -AA' _m -L3- ^v l-L2-R

wherein X, AA' , m, Y _lf R and i are as defined above and L ₂ is a linker which enables a selective cleavage from the solid support.

A fourth aspect of this embodiment relates to an agent for use in solid phase peptide synthesis having the general formula

X-AA' ₁ - • .. -AA' _rn -Y ₁ -L ₂ -R

wherein X, AA' , m, Y^, R and L ₂ is as defined above.

It will be understood that the above described agents may be in the form of polymers, gels or other solid phases which are prepared for solid phase peptide synthesis according to the invention by

a)containing a pre-sequence and optionally one or more amino acids from the desired sequence (the first aspect) or

b)containing a pre-sequence, a preferably optically active cleavage linker and optionally one or more amino acids from the desired sequence (the second aspect ) , or

c)containing a second linker enabling cleavage from the support, a pre-sequence, a first preferably op¬ tically active cleavage linker and optionally one

or more amino acids from the desired sequence (the third aspect) and

d)containing a linker enabling cleavage from the sup¬ port, a pre-sequence and optionally one or more amino acids from the desired sequence (the fourth aspect) .

The specific conditions used in the experiments on which this invention is based are stated below under general procedures .

Generally speaking, apart from the novel and charac¬ teristic features related to the pre-sequence and the novel cleavage linkers the method according to the invention may be carried out under traditional condi¬ tions for solid phase peptide synthesis described in the literature referred to in the background art.

However, a brief summary is deemed proper.

Fmoc groups may be deprotected by means of an amine such as piperidine or diazabicyclo[ 5 , 4 , 0 ]undec-7-ene (DBU) .

Side chain protective groups may be deprotected by means of an acid such as trifluoroacetic acid (TFA) , trifluoromethanesulfonic acid (TFMSA) , HBr, HCl or HF.

The solid support is preferably selected from func- tionalized resins such as polystyrene, polyacryla- mide, polyethyleneglycol , cellulose, polyethylene, latex or dynabeads .

If desired, C-terminal amino acids may be attached to the solid support by means of a common linker such as 2 , 4-dimethoxy-4 ' -hydroxy-benzophenone , 4- ( 4-hydroxy-

methyl-3-methoxyphenoxy) -butyric acid (HMPB), 4- hydroxymethylbenzoic acid, 4-hydroxymethylphenoxy- acetic acid (HMPA), 3- ( 4-hydroxymethylphenoxy)pro- pionic acid or p- [ (R, S)-a[ 1- ( 9H-fluoren-9-yl)- methoxyformamido]-2, 4-dimethoxybenzyl ] -phenoxyacetic acid (AM) .

The synthesis may be carried out batchwise or con¬ tinuously on an automated or semi automated peptide synthesizer .

The individual coupling steps may be performed in the presence of a solvent, e.g. selected from acetoni¬ trile, ,N-dimethylformamide (DMF), N-methylpyrro- lidone ( NMP) , dichloromethane (DCM) , trifluoroethanol (TFE), ethanol, methanol, water, mixtures of the men- tioned solvents with or without additives such as perchlorate or ethylenecarbonate .

The individual couplings between two amino acids, an amino acid and the earlier formed peptide sequence or a peptide fragment and the earlier formed peptide se- quence may be carried out according to usual conden¬ sation methods such as the azide method, mixed acid anhydride method, symmetrical anhydride method, car- bodiimide method, active ester method such as pen- tafluorophenyl (Pfp) , 3 , 4-dihydro-4-oxobenzotriazin-3- yl (Dhbt) , benzotriazol-1-yl (Bt), 7-azabenzotriazol- 1-yl (At) , 4-nitrophenyl , N-hydroxysuccinic acid imido esters (NHS) , acid chlorides, acid fluorides, in situ activation by O- ( 7-azabenzotriazol-l-yl ) - 1,1,3, 3-tetramethyluronium hexafluorophosphate (HATU), 0-(benzotriazol-l-yl )-l , 1 , 3, 3-tetramethyl- uronium hexafluorophosphate (HBTU), 0- (benzotriazol- l-yl)-l,l,3, 3-tetramethyluronium tetrafluoroborate

(TBTU), or benzotriazolyl-oxy-tris- (dimethylamio) - phosphonium hexafluorophosphate (BOP) .

The formed peptide may be cleaved from the support by means of an acid such as trifluoroacetic acid (TFA) , trifluoromethanesulfonic acid (TFMSA) , hydrogen bro¬ mide (HBr), hydrogen chloride (HCl) , hydrogen fluo¬ ride (HF) or a base such as ammonia, hydrazine, an alkoxide or a hydroxide.

Alternatively, the peptide is cleaved from the sup- port by means of photolysis.

In the embodiment, where a linker is inserted between the pre-sequence attached to the support and the AAι~ AA _n sequence which enables a selective cleavage of said sequence, said cleavage may be made by means of an acid such as trifluoroacetic acid (TFA), tri- fluoromethanesulfonic acid (TFMSA) , hydrogen bromide (HBr), hydrogen chloride (HCl), hydrogen fluoride (HF) or a base such as ammonia, hydrazine, an alkox¬ ide or a hydroxide.

Sequence Assisted Peptide Synthesis (SAPS).

EXPERIMENTAL PROCEDURES:

Peptide synthesis

General procedures

Apparatus and synthetic strategy

Peptides were synthesized either batchwise in a poly¬ ethylene vessel equipped with a polypropylene filter for filtration or in the continuous-flow version of the polyamide solid-phase method (Dryland, A. and Sheppard, R.C. , Ref. 7 ) on a fully automated peptide

synthesizer (Cameron et al . , Ref. 14) using 9- fluorenylmethyloxycarbonyl (Fmoc) or tert . -Butyloxy- carbonyl, (Boc) as N-α-amino protecting group and suitable common protection groups for side-chain functionalities.

Solvents

Solvent DMF ( N,V-dimethylformamide, Riedel de-Haen, Germany) was purified by passing through a column packed with a strong cation exchange resin (Lewatit S 100 MB/H strong acid, Bayer AG Leverkusen, Germany) and analyzed for free amines prior to use by addition of 3 , 4-dihydro-3-hydroxy-4-oxo-1 , 2 , 3-benzotriazine (Dhbt-OH) giving rise to a yellow color (Dhbt-O" an¬ ion) if free amines are present. Solvent DCM (dichloromethane, analytical grade, Riedel de-Haen, Germany) was used directly without purification.

Amino acids

Fmoc-protected amino acids and corresponding pen- tafluoropheny1 (Pfp) esters were purchased from Mil- liGen, UK, NovaBiochem, Switzerland and Bache , Swit¬ zerland, and the Dhbt-esters from NovaBiochem, Swit¬ zerland in suitable side-chain protected forms. Boc- protected amino acids were purchased from Bachem, Switzerland.

Coupling reagents

Coupling reagent diisopropylcarbodiimide (DIC) was purchased from Riedel de-Haen, Germany and distilled prior to use, dicyclohexylcarbodiimide ( DCC) was pur¬ chased from Merck-Schuchardt, Mϋnchen, Germany, and purified by destination. O-Benzotriazoly1-N,N, N ' ,N ' - tetramethyluronium tetrafluoroborate (TBTU) was pur-

chased from PerSeptive Biosystems GmbH Hamburg, Ger¬ many .

Linkers

Linkers HMPA, Novabiochem, Switzerland; 4-hydroxy- methylbenzoic acid, Novabiochem; 4-methoxymandelic acid, Aldrich, Germany; HMPB, Novabiochem; AM, Novabiochem; 3- ( 4-hydroxymet ylphenoxy)propionic acid, Novabiochem, was coupled to the resin as a pre¬ formed 1-hydroxybenzotriazole (HObt) ester generated by means of DIC. Racemic 4-methoxymandelic acid (98% pure, Aldrich, Germany) was used directly as linker or resolved by treatment with ( + ) -cinchonine (85% pure, Aldrich, Germany) giving the optical active linker ( + ) -4-methoxymandelic acid, [a]20 =+ 45 (water) in 95.8 % optical purity and ( - ) -4-methoxy- mandelic acid, [a] ²⁰ =- 128.6 (water) in 88.1 % op¬ tical purity.

Resolution of ( + / - ) -4-methoxymandelic acid (A. Mc Kenzie, D.J.C. Pirie, Ref. 16; E. Knorr, Ref. 17)

( +/-) -4-Methoxymandelic acid (10 g, 54.89 mmol; Ald¬ rich, 98%) was dissolved in 500 ml hot water (60-80 °C) and the solution decanted while still warm in or¬ der to remove insoluble impurities. (+) -Cinchonine (16,16 g, 54.89 mmol, Aldrich, 85%, [a)o ²⁰ = + 211 ° (litt. : + 228 °) ) was added to the hot solution in small portions. The solution became clear after 15 min stirring at 60-80 °C and was cooled in ice. After 1 h the precipitate was collected by filtration and dried in an exsiccator over night, yielding 9.9 g of the chinconine salt. The salt was recrystallized from boiling water (80 ml), the solution decanted while still warm and then cooled in ice. The precipitate was collected by filtration after 1 h, washed three

times with cold water, and dried in an exsiccator over night yielding 7.84 g ( 16.45 mmol) . The chin- conine salt (2 g, 4,2 mmol) was dissolved in 40 ml 2N HCl and immediately extracted with 3 x 30 ml dieth- ylether. The ether phase was dried over Na2S04 and evaporated to dryness yielding 0.55 g 4-methoxy- mandelic acid. The optical purity of the liberated 4- methoxymandelic acid was estimated to 18.5% ( fa]D = + 27 °) . After a second recrystal1isation of the chinconine salt followed by liberation of the man- delic acid as described above the optical purity was estimated to 69.0 % ( [a]o ²⁰ = ^ 100.8 °) . A third recrystallisation resulted in an optical purity of 95,8% ( [a]D ²⁰ = ^■ ♦ 140.0 °) .

Solid supports

Peptides synthesized according to the Fmoc-strategy were synthesized on three different types of solid support using 0.05 M or higher concentrations of F oc-protected activated amino acid in DMF. 1) PEG-PS ( polyethyleneg] ycol grafted on polystyrene; TentaGel S NH2 resin, 0.27 mmol/g, Rapp Polymere, Germany or NovaSyn TG resin, 0.29 mmol/g, Novabiochem, Switzer¬ land) ; 2) PepSyn Gel (polydimethylacrylamide resin functionalized with sarcosine methylester, 1.0 mmol/g; MilliGen, UK) . 3) PepSyn K (Kieselguhr sup¬ ported polydimethylacrylamide resin functionalized with sarcosine methylester 0.11 mmol/g; MilliGen, UK) .

Peptides synthesized according to the Boc-strategy were synthesized on a Merrifield-resin ( polystyrene- divinylbenzene) with the first amino acid attached (Novabiochem, Switzerland) .

Catalysts and other reagents

Diisopropylethylamine (DIEA) was purchased from Aid- rich, Germany, and ethylenediamine from Fluka, Swit¬ zerland, piperidine from Riedel-de Haen, Frankfurt, Germany. 4- (N,N-dimethylamino)pyridine ( DMAP) was purchased from Fluka, Switzerland and used as a cata¬ lyst in coupling reactions involving symmetrical an¬ hydrides . 3 , 4-dihydro-3-hydroxy-4-oxo-l ,2,3-benzo- triazine (Dhbt-OH) was obtained from Fluka, Switzer- land, and 1-hydroxybenzotriazole (HObt) from NovaBio¬ chem, Switzerland.

Coupling procedures

The first amino acid was coupled as a symmetrical an¬ hydride in DMF generated from the appropriate N-α- protected amino acid and DIC. The following amino ac¬ ids were coupled as Pfp- or Dhbt-esters or as pre¬ formed HObt esters made from appropriate N-α- protected amino acids and HObt by means of DIC or TBTU in DMF. In the case of Fmoc all acylations were checked by the ninhydrin test performed at 80 °C in order to prevent Fmoc deprotection during the test (Larsen, B. D. and Holm, A., Ref. 4)

Deprotection of the N-α-amino protecting group.

Deprotection of the Fmoc group was performed by treatment with 20% piperidine in DMF (1x3 and 1x7 min when synthesized batchwise) or by flowing the depro¬ tection solvent through the resin (10 min, flow rate 1 l/min using continuous flow synthesis) , followed by wash with DMF until no yellow color (Dhbt-O") could be detected after addition of Dhbt-OH to the drained DMF.

Deprotection of the Boc group was performed by treat¬ ment with 50% TFA in DCM (v/v) 1x1.5 min and 1x20 min followed by wash 6x9 min each with DCM, neutraliza¬ tion with 10% triethylamine in DCM (v/v) 2x1.5 min each, followed by 6x9 min wash with DCM.

Cleavage of peptide from resin with acid.

Peptides were cleaved from the resins by treatment with 95% triflouroacetic acid (TFA, Halocarbon Prod¬ ucts Corporation, U.S.A.; Biesterfeld & Co. Hamburg, Germany) -water v/v at r.t. for 2 h. The filtered res¬ ins were washed with 95% TFA-water and filtrates and washings evaporated under reduced pressure. The resi¬ due was washed with ether and freeze dried from ace¬ tic acid-water. The crude freeze dried product was analyzed by high-performance liquid chromatography (hplc) and identified by matrix assisted laser desorption ionization time of flight mass spectrome- try (MALDI TOF MS) or by electrospray ionization mass spectrometry (ESMS) .

Cleavage of peptide from resin with base .

The dried resin (1 g) was treated with IM sodium hy¬ droxide (10 ml) at 4 oC and left for 15 min at rt . The resin was filtered into a flask containing 10% aq. acetic acid. The peptide was isolated by lyophi- lization and submitted to gel filtration.

Cleavage of peptide from resin with TFMSA.

The dried resin (250 mg) was placed in a round- bottomed flask with a stirring bar. Thioani- εole/ethanedithiol (2:1, 750 μl) was added, the mix- ture chilled in ice, 5 ml TFA was added and the mix¬ ture was stirred for 5 - 10 min. TFMSA (500 μl ) was

added dropwise and the reaction continued at rt for 30 - 60 min. The peptide was precipitated after addi¬ tion of ether.

Deprotection of side chain protective groups

Preferably, the side chains are deprotected simulta¬ neously with the cleavage of the peptide from the resin .

Preformed HObt-ester

Method a . 3 eq . N-α-amino protected amino acid was dissolved in DMF together with 3 eq . HObt and 3 eq DIC. The solution was left at r.t. for 10 minutes and then added to the resin, which had been washed with a solution of 0.2% Dhbt-OH in DMF prior to the addition of the preactivated amino acid.

Method b . 3 eq . N-α-amino protected amino acid was dissolved in DMF together with 3 eq . HObt, 3 eq TBTU and 4,5 eq . DIEA. The solution was left at r.t. for 5 minutes and then added to the resin.

Preformed symmetrical anhydride

6 eq . N-α-amino protected amino acid was dissolved in DCM and cooled to 0 °C . DCC (3 eq . ) was added and the reaction continued for 10 min. The solvent was re¬ moved i n vacuo and the remanens dissolved in DMF. The solution was filtered and immediately added to the resin followed by 0.1 eq . of DMAP .

Estimation of the coupling yield of the first N-α- amino protected amino acid

3 - 5 mg dry Fmoc-protected peptide-resin was treated with 5 ml 20% piperidine in DMF for 10 min at r.t.

and the UV absorption for the dibenzofulvene- piperidine adduct was estimated at 301 nm. The yield was determined using a calculated extension coeffi¬ cient e oi based on a Fmoc-Ala-OH standard.

In case of Boc-protection, the coupling was estimated according to the ninhydrin-method after removal of the Boc-group (Sarin, V.K. et al . , Ref. 11)

Peptide synthesis on PepSyn K resin

Dry PepSyn K (ca 500 mg) , was covered by ethylenedia- mine and left at rt over night. The resin was drained and washed with DMF 10 x 15 ml, 5 min each. After draining the resin was washed with 10% DIEA in DMF v/v (2 x 15 ml, 5 min each) and finally washed with DMF until no yellow color could be detected by addi- tion of Dhbt-OH to the drained DMF. 3 eq . HMPA 3 eq. HObt and 3 eq . DIC was dissolved in 10 ml DMF and left for activation for 10 min, after which the mix¬ ture was added to the resin and the coupling contin¬ ued for 24 h. The resin was drained and washed with DMF (10 x 15 ml, 5 min each), and the acylation was checked by the ninhydrin test. The first amino acid was coupled as the side chain protected preformed symmetrical anhydride (see above), and the coupling yields estimated as described above. It was in all cases better than 70%. The synthesis was then contin¬ ued either as " con tinuous- fl ow " or as " ba tchwise " as described below.

Continued peptide synthesis on PepSvn K using con¬ tinuous-flow technique.

The resin (ca. 500 mg) with the first amino acid at¬ tached was placed in a column connected to the fully automated peptide synthesizer. The Fmoc group was

deprotected as described above. The remaining amino acids according to the sequence were coupled as Fmoc- protected, if necessary side chain protected, Pfp es¬ ters (3 eq. ) with the addition of Dhbt-OH (1 eq. ) . The end-point of each coupling was determined auto¬ matically by monitoring the disappearance of the yel¬ low color of the Dhbt-OH anion spectrophotometri- cally. After completed synthesis the peptide-resin was washed with DMF (10 min flow rate 1 ml/min), DCM (3x5 ml, 3 min each) and finally diethyl ether (3x5 ml each) , removed from the column and dried i n vacuo .

Continued batchwise peptide synthesis on PepSyn K.

The resin (ca. 500 mg) with the first amino acid at¬ tached was placed in a polyethylene vessel equipped with a polypropylene filter for filtration, and the Fmoc-group deprotected as described above. The re¬ maining amino acids according to the sequence were coupled as preformed Fmoc-protected , if necessary side chain protected, HObt esters (3 eq . ) in DMF (5 ml) prepared as described above. The couplings were continued for 2 h unless otherwise specified. Excess reagent was then removed by DMF washing (12 min, flow rate 1 ml/min) All acylations were checked by the ninhydrin test performed at 80 °C . After completed synthesis the peptide-resin was washed with DMF ( lo min, flow rate 1 ml/min), DCM (5x5 ml, 1 min each) and finally diethyl ether (5x5 ml, 1 min each) and dried in vacuo .

Batchwise peptide synthesis on PEG-PS.

TentaGel S NH2 or NovaSyn TG resin (250 mg, 0.27-0.29 mmol/g) was placed in a polyethylene vessel equipped with a polypropylene filter for filtration. The resin was swelled in DMF (5 ml) , and treated with 20%

piperidine in DMF to secure the presence of non- protonated amino groups on the resin. The resin was drained and washed with DMF until no yellow color could be detected after addition of Dhbt-OH to the drained DMF. HMPA (3 eq . ) was coupled as a preformed HObt-ester as described above and the coupling was continued for 24 h. The resin was drained and washed with DMF (5 5 ml, 5 min each) and the acylation checked by the ninhydrin test. The first amino acid was coupled as a preformed symmetrical anhydride as described above. The coupling yields of the first Fmoc-protected amino acids were estimated as de¬ scribed above. It was in all cases better than 60%. The following amino acids according to the sequence were coupled as preformed Fmoc-protected, if neces¬ sary side chain protected, HObt esters (3 eq . ) as de¬ scribed above. The couplings were continued for 2 h, unless otherwise specified. The resin was drained and washed with DMF ( 5 x 5 ml , 5 min each) in order to remove excess reagent. All acylations were checked by the ninhydrin test performed at 80 °C . After com¬ pleted synthesis the peptide-resin was washed with DMF (3x5 ml, 5 min each) , DCM (3x5 ml, 1 min each) and finally diethyl ether (3x5 ml, 1 min each) and dried in vacuo .

Batchwise peptide synthesis on PepSyn Gel.

Dry PepSyn Gel resin (500 mg, 1 mmol/g) was placed in a polyethylene vessel equipped with a polypropylene filter for filtration. The resin was swelled in eth- ylenediamine (15 ml) and gently agitated by shaking for 20 h. The resin was drained and washed with DMF (10 x 15 ml, 5 min each) . After draining the resin was washed with 10% DIEA in DMF v/v (2 x 15 ml, 5 min each) and finally washed with DMF (5 x 15 ml, 5

min each) until no yellow color could be detected af¬ ter addition of Dhbt-OH to the drained DMF. HMPA (3 eq. ) was coupled as a preactivated HObt-ester as de¬ scribed above (method a) and the coupling was contin- ued for 24 h. The resin was drained and washed with DMF (5 x 15 ml, 5 min each) . The acylation was checked by the ninhydrin test. The first amino acid was coupled as preformed side chain protected symmet¬ rical anhydride as described above. The coupling yields of the first Fmoc-protected amino acids were estimated as described above. It was in all cases better than 70%. The remaining amino acids according to the sequence were coupled as preformed Fmoc- protected, if necessary side chain protected, HObt esters (3 eq . ) as described above (method a _. ) . The couplings were continued for 2 h and, if necessary, double coupled over night. The resin was drained and washed with DMF (5 x 5 ml, 5 min each) in order to remove excess reagent. All acylations were checked by the ninhydrin test performed at 80 °C . The Fmoc group was deprotected as described above. After completed synthesis the peptide-resin was washed with DMF (3x15 ml, 5 min each) , DCM (3x15 ml, 2 min each) and fi¬ nally diethyl ether (3x15 ml, 2 min each) and dried in vacuo .

Hplc conditions.

Hplc was performed on a Waters 600 E instrument equipped with a Waters 996 Photodiode array detector with a Waters Radial Pak 8 x 100 mm Ciβ reversed- phase column. Buffer A was 0.1 vol % TFA in water and buffer B 90 vol% acetonitrile, 9.9 vol% water and 0.1 vol% TFA. The Buffers were pumped through the column at a flow rate of 1.5 ml/min using the gradient: 1. Linear gradient from 0% - 70% B (20 min), linear

gradient from 70 - 100% B (1 min), isocratic 100% B (5 min) . 2. Isocratic with 0% B (2 min) , linear gra¬ dient from 0 - 50% B (23 min), linear gradient from 50 - 100% B (5 min) , isocratic 100% B (5 min) .

Mass spectroscopy.

Matrix assisted laser desorbtion ionization time-of- flight (MALDI TOF) mass spectra were obtained on a Fisons TofSpec E instrument. Electrospray ionization mass spectra were obtained on a Finnigan Mat LCQ in- strument equipped with an electrospray (ESI) probe (ES-MS) .

Peptide syn thesi s of indi vi dual pept i des .

Example 1. Synthesis of H-Alaip-Lys-OH (comparison) .

500 mg Fmoc-Lys (Boc ) PepSyn KA resin (0.086 mmol/g) was used for synthesis according to "continuous-flow technique" .

The crude freeze dried product was analyzed by hplc and found to be a complicated mixture comprising the target peptide (n - 10) as well as deletion peptides corresponding to n = 6, 7, 8, and 9 and deletion pep¬ tides with the Fmoc group still attached to the N- terminal where n = 6,7,8, and 9 respectively. The identity of the individual peptides was confirmed by MALDI TOF MS.

Example 2. Synthesis of H-Ala p-Lys^-OH.

500 mg Fmoc-Lys ( Boc ) PepSyn KA resin (0.086 mmol/g) was used for synthesis according to "continuous-flow technique" .

The crude freeze dried product was analyzed by hplc and found to be homogeneous without deletion and Fmoc-protected sequences. Yield 91,0%. The purity was found to be better than 98% according to hplc (see Fig. 3) . The identity of the peptide was confirmed by MALDI TOF MS.

Example 3. Synthesis of H-AlaiQ-Lys -OH.

500 mg Fmoc-Lys ( Boc ) PepSyn KA resin (0.086 mmol/g) was used for synthesis according to "continuous-flow technique" .

The crude freeze dried product was analyzed by hplc and found to be homogeneous without deletion and Fmoc-protected sequences. Yield 90,9%. The purity was found to be better than 98% according to hplc (see Fig. 5) . The identity of the peptide was confirmed by ES MS.

4. Synthesis of H-Ala2Q-Lysj-OH.

500 mg Fmoc-Alaιo- (Lys (Boc ) ) 3 PepSyn KA resin (from the synthesis of H-Ala 0-Lys3-OH) was used for syn- thesis according to "continuous-flow technique" and the synthesis continued.

The crude freeze dried product was analyzed by hplc and found to comprise the target peptide H-Ala _n -Lys3- OH (n = 20) as well as deletion peptides correspond- ing to n - 19, 18, 17 and 16 (see Fig. 4) . Fmoc- protected sequences were not detected. The identity of the peptides were confirmed by ES MS.

Example 5. Synthesis of H-Ala2Q-Lys -0H .

500 mg Fmoc-Lys ( Boc ) PepSyn KA resin (0.086 mmol/g) was used for synthesis according to "continuous-flow technique " .

The crude freeze dried product was analyzed by hplc and found to be homogeneous without deletion and Fmoc-protected sequences. Yield 91.4%. The purity was found to be better than 98% according to hplc (see Fig. 6) . The identity of the peptide was confirmed by ES MS.

Example 6. Synthesis of H-Ala2Q-Lys- (GIu-Lvs )_τ_ -OH.

500 mg Fmoc-Lys ( Boc ) PepSyn KA resin (0.086 mmol/g) was used for synthesis according to "continuous-flow technique" .

The crude freeze dried product was analyzed by hplc. It was found to be better than 98% pure without dele¬ tion and Fmoc-protected sequences. Yield 97%. The identity of the peptide was confirmed by ES MS.

Example 7. Synthesis of Acyl Carrier Protein (ACP) 65-74 , H-Val-Gln-Ala-Ala-Ile-Asp-Tyr-lle-Asn-Gly-OH (comparison) .

500 mg F oc-Gly PepSyn KA resin (0.074 mmol/g) was used for synthesis according to "continuous-flow technique" .

The crude freeze dried product was analyzed by hplc and found to contain the target molecule accompanied by ca. 16% of the des-Val peptide. The identity of the peptides was confirmed by MALDI TOF MS.

Example 8. Synthesis of Acyl Carrier Protein (ACP)

65-74, H-Val-Gln-Ala-Ala-Ile-Asp-Tyr-Ile-Asn-Glγ-

Lys -OH. using the pre-sequence fLvs(Boc) )fi

500 mg Fmoc-Lys ( Boc ) PepSyn KA resin (0.086 mmol/g) was used for synthesis according to "continuous-flow technique" .

The crude freeze dried product was analyzed by hplc and found to contain the target peptide in high pu¬ rity (- 95%) with a significantly reduced amount of the des-Val peptide. The identity of the peptide was confirmed by MALDI TOF MS.

Example 9. Synthesis of H-Val-Asn-Val-Asn-Val-Gln-

Val-Gln-Val-Asp-Lys -OH, using the pre-sequence

(Lys(Boc) )&.

500 mg Fmoc-Lys (Boc ) PepSyn KA resin (0.086 mmol/g) was used for synthesis according to "continuous-flow technique" .

The crude freeze dried product was analyzed by hplc and found to be better than 90% pure without deletion and Fmoc-protected sequences. Yield 100%. The iden¬ tity of the peptide was confirmed by MALDI TOF MS.

Example 10. Synthesis of H-Alaip-Lys-OH using

(Lys(Boc) )^ as pre-sequence and HMPA as linker ( H- PepSyn K \ .

500 mg dry PepSyn K (0.1 mmol/g) was covered by eth- ylenediamine (5 ml) and left at rt over night. The resin was drained and washed with DMF 10 x 15 ml, 5 min each. After draining the resin was washed with 10% DIEA in DMF v/v (2 x 15 ml, 5 min each) and fi¬ nally washed with DMF until no yellow color could be

detected by addition of Dhbt-OH to the drained DMF. The derivatized resin was used for synthesis accord¬ ing to "continuous-flow technique".

The first 6 Lysines forming the pre-sequence were coupled as Fmoc-Lys (Boc) -Pfp esters (3 eq. ) with the addition of Dhbt-OH (1 eq. ) . The end-point of each coupling was determined automatically as described above. The Fmoc group was cleaved as described above. After finishing the pre-sequence 3 eq . HMPA coupled as a preactivated HObt ester as described above was introduced at the top of the column. The synthesizer was operated in recirculation mode for 2 h and excess of reagent was then removed by DMF washing (12 min, flow rate 1 ml/min) . A small resin-sample was removed in order to check the acylation by the ninhydrin test. The next amino acid according to the sequence was coupled as preformed side chain protected symmet¬ rical anhydride as described above and introduced at the top of the column together with (0.1 eq . ) DMAP and the synthesizer was operated in recirculation mode for 90 min. Excess reagent was then removed by DMF washing (12 min, flow rate 1 ml/min) . A small resin-sample was removed in order to check the cou¬ pling yield, which was estimated as earlier described and found to be 84%. The synthesis was then continued by cleavage of the Fmoc group as described above. The remaining amino acids according to the sequence were coupled as Fmoc-protected, if necessary side chain protected, Pfp esters (3 eq . ) with addition of Dhbt- OH ( 1 eq.) . The end-point of each coupling was deter¬ mined automatically as described above. After com¬ pleted synthesis the peptide-resin was washed with DMF (10 min, flow rate 1 ml/min), DCM (3x5 ml, 1 min each) and finally diethyl ether (3x5 ml, 1 min each) , removed from the column and dried i n vacuo .

The peptide was cleaved from the resin as described above .

The crude freeze dried product was analyzed by hplc and found to comprise the target peptide H-Ala _n -Lys- OH (n = 10) as well as deletion peptides correspond¬ ing to n = 9 , 8, 7 and 6 (see Fig. 9) . Fmoc-protected sequences were not detected. The identity of the pep¬ tides were confirmed by MALDI TOF MS.

Example 11. Synthesis of H-Ala p-Lys-OH using (+/-)- 4-methoxymandelic acid as linker (H-Alaip-Lys ( Boc ) - OCH-f 4-MeOPnCONHCH CH NH PepSvn K resin) .

500 mg dry PepSyn K (0.1 mmol/g) was treated with ethylenediamine as described above. The derivatized resin was used for synthesis according to "continuous-flow technique". 10 eq . ( +/- ) -4-methoxy- mandelic acid, 10 eq . HObt and 10 eq . DIC dissolved in 5 ml DMF, preactivated for 10 min was introduced at the top of the column and the synthesizer was op¬ erated in recirculation mode for 2 h. Excess of rea- gent was then removed by DMF washing (12 min, flow rate 1 ml/min) . A small resin-sample was removed in order to check the acylation by the ninhydrin test. The first amino acid according to the sequence was coupled as Fmoc protected and side chain protected preformed symmetrical anhydride as described above and introduced at the top of the column together with (0.1 eq . ) DMAP and the synthesizer was operated in recirculation mode for 90 min. Excess reagent was then removed by DMF washing (12 min, flow rate 1 ml/min) . A small resin-sample was removed in order to check the coupling yield, which was estimated as de¬ scribed above and found to be 75%. The synthesis was then continued by cleavage of the Fmoc group as ear-

lier described. The remaining amino acids according to the sequence were coupled as Fmoc-protected Pfp esters (3 eq. ) with the addition of Dhbt-OH (1 eq. ) . The end-point of each coupling was determined auto- matically as described above. After completed synthe¬ sis the peptide-resin was washed with DMF (10 min, flow rate 1 ml/min) , DCM (3x5 ml, 1 min each)and fi¬ nally diethyl ether, removed from the column and dried in vacuo .

The peptide was cleaved from the resin as earlier de¬ scribed and lyophilized from acetic acid-water.

The crude freeze dried product was analyzed by hplc and found to comprise the target peptide H-Ala _n -Lys- OH (n = 10) as well as deletion peptides correspond- ing to n = 9 , 8, 7 and 6. The identity of the pep¬ tides were confirmed by MALDI TOF MS.

Example 12. Synthesis of H-Alaip-Lys-OH using

(Lys (Boc ) ) as pre-sequence and ( +/- )-4-methoxy- andelic acid as linker (H-Alaip-Lys( Boc ) -OCH-( 4- MeOPh)CO-( ysfBoc) -NHCH CH2NH PepSyn K resin)

500 mg dry PepSyn K (0.1 mmol/g) was treated with ethylenediamine as described above. The derivatized resin was used for synthesis according to "continuous-flow technique". The first 6 Lysines forming the pre-sequence were coupled as Fmoc- protected and side chain protected Pfp esters (3 eq. ) with the addition of Dhbt-OH (1 eq . ) . The end-point of each coupling was determined automatically as de¬ scribed above. The Fmoc group was cleaved as de- scribed above. After completed synthesis of the pre- sequence 10 eq . ( +/- )-4-methoxymandelic acid was coupled as preactivated HObt-ester as earlier de¬ scribed and introduced at the top of the column. The

synthesizer was operated in recirculation mode for 2 h and excess reagent was then removed by DMF washing (12 min, flow rate 1 ml/min) . A small resin-sample was removed in order to check the acylation by the ninhydrin test. The next amino acid according to the sequence was coupled as Fmoc protected and side chain protected preformed symmetrical anhydride as earlier described and introduced at the top of the column to¬ gether with (0.1 eq. ) DMAP . The synthesizer was oper- ated in recirculation mode for 90 min and excess rea¬ gent was then removed by DMF washing (12 min, flow rate 1 ml/min) . A small resin-sample was removed in order to check the coupling yield, which was esti¬ mated as described above and found to be 68%. The synthesis was then continued by cleavage of the Fmoc group as earlier described. The remaining amino acids according to the sequence were coupled as Fmoc- protected Pfp esters (3 eq . ) with the addition of Dhbt-OH (1 eq . ) . The end-point of each coupling was determined automatically as described above. After completed synthesis the peptide-resin was washed with DMF (10 min, flow rate 1 ml/min) , DCM (3x5 ml min, flow rate 1 ml/min) and finally diethyl ether (3x5 ml min, flow rate 1 ml/min), removed from the column and dried in vacuo . The peptide was cleaved from the resin as earlier described and freeze dried from ace¬ tic acid-water. The crude freeze dried product was analyzed by hplc and found to comprise the target peptide H-Ala _n -Lys-OH (n = 10) as well as deletion peptides corresponding to n = 9 , 8, 7 and 6. The amount of deletion peptides was found to be signifi¬ cantly reduced compared to examples 10 and 11 (see fig. 10) . Fmoc-protected sequences were not detected. The identity of the peptides were confirmed by MALDI TOF MS.

Example 13. Synthesis of H-AlaiQ-Lys-OH using

(LγsfBoc )^ as pre-sequence and ( + ) -4-methoxymandelic acid as linker (H-Ala _j j^-Lys ( Boc ) -OCH- ( 4-MeOPh)C0- (Lys(Boc) -NHCH2CH H PepSvn K resin) .

Dry PepSyn K (ca 500 mg, 0.1 mmol/g) was placed in a polyethylene vessel equipped with a polypropylene filter for filtration and treated with ethylenedia- mine as earlier described. The first 6 Lysines form¬ ing the pre-sequence were coupled as Fmoc-protected and side chain protected Pfp esters (3 eq . ) with the addition of Dhbt-OH (1 eq. ) . The acylations were checked by the ninhydrin test performed at 80°C as described above. The Fmoc group was deprotected as described above. After finishing the pre-sequence the deprotected peptide-resin was reacted with 10 eq . ( + ) -4-methoxymandelic acid as a preactivated HObt- ester as described above (resolved as described above, 95.8% optical purity ) and the coupling was continued for 24 h. Excess reagent was removed by DMF washing (12 min flow rate 1 ml/min) . The acylation was checked by the ninhydrin test. The next amino acid according to the sequence was coupled as Fmoc protected and side chain protected preformed symmet¬ rical anhydride as described above and the reaction was continued for 2 h. Excess reagent was then re¬ moved by DMF washing (12 min flow rate 1 ml/min) . A small resin-sample was removed in order to check the coupling yield, which was estimated as described above, and found to be 66%. The synthesis was then continued by cleavage of the Fmoc group as described above .

The first alanine was coupled as a Fmoc-protected Pfp ester (3 eq. ) with the addition of Dhbt-OH (1 eq . ) in DMF (2 ml) for 2 h. Excess reagent was then removed

by DMF washing (12 min flow rate 1 ml/min) and the acylation was checked by the ninhydrin test performed at 80 °C as described above. The Fmoc group was then removed by treatment with 2% piperidine in DMF v/v (1 min, flow rate 1 ml/min) followed by flushing with DMF (10 sec, flow rate 10 ml/min), flowing with 0.2% Dhbt-OH in DMF (20 min, flow rate 1 ml/min) and fi¬ nally washing with DMF (2x5 ml, 1 min each) . The fol¬ lowing Fmoc-protected alanine was coupled immediately as a Pfp ester (3 eq. ) with the addition of 1 eq Dhbt-OH in DMF (2 ml) for 2 h. The acylation was checked by the ninhydrin test performed as described above. The remaining amino acids according to the se¬ quence were coupled as Fmoc-protected Pfp esters (3 eq. ) with the addition of Dhbt-OH (1 eq . ) in DMF (2 ml) . Excess reagent was removed by DMF washing (12 min flow rate 1 ml/min) and acylations were checked by the ninhydrin test performed 80 °C as described above. The Fmoc group was deprotected as described above. After completed synthesis the peptide-resin was washed with DMF (10 min, flow rate 1 ml/min), DCM (3x5 ml, 1 min each), diethyl ether (3x5 ml, 1 min each) and dried i n vacuo .

The peptide was cleaved from the resin as described above and freeze dried from glaciai acetic acid. The crude freeze dried product was analyzed by hplc and found to be homogeneous without deletion and Fmoc- protected sequences.

Example 14. Synthesis of H-Alaip-Lvs-OH using

( Glu ( OtBu) ) as pre-sequence and ( +) -4- methoxymandelic acid as linker ( H-Alaip-Lys ( Boc ) -OCH-

Dry PepSyn K (ca 500 mg) was placed in a polyethylene vessel equipped with a polypropylene filter for fil¬ tration and treated with ethylenediamine as described above. The first 6 glutamic acids forming the pre- sequence were coupled as Fmoc-Glu ( OtBu ) -Pfp esters (3 eq. ) with the addition of Dhbt-OH (1 eq . ) . The acyla¬ tions were checked by the ninhydrin test performed at 80°C as described above. The Fmoc group was depro¬ tected as described above. After finishing the pre- sequence 10 eq . ( + ) -4-methoxymandelic acid (resolved as described above, 95.8% optical purity) was coupled as a preactivated HObt-ester as described above. The coupling was continued for 24 h and excess reagent was then removed by DMF washing (12 min flow rate 1 ml/min) . The acylation was checked by the ninhydrin test. The next amino acid according to the sequence was coupled as Fmoc protected and side chain pro¬ tected preformed symmetrical anhydride as described above catalyzed by DMAP (0.1 eq . ) and the reaction was continued for 2 h. Excess reagent was removed by DMF washing (12 min flow rate 1 ml/min) . A small resin-sample was removed in order to check the cou¬ pling yield, which was estimated as described above and the yield found to be 75%. The synthesis was then continued by cleavage of the Fmoc group as described above .

The first alanine was coupled as a Fmoc-protected Pfp ester (3 eq. ) with the addition of Dhbt-OH (1 eq. ) in DMF (2 ml) for 2 h. Excess reagent was then removed by DMF washing (12 min flow rate 1 ml/min) and the

acylation was checked by the ninhydrin test performed at 80 °C as described above. The Fmoc group was then removed by treatment with 2% piperidine in DMF v/v (1 min, flow rate 1 ml/min) followed by flushing with DMF (10 sec, flow rate 10 ml/min), flowing with 0.2% Dhbt-OH in DMF (20 min, flow rate 1 ml/min) and fi¬ nally washing with DMF (2x5 ml, 1 min each) . The fol¬ lowing Fmoc-protected alanine was coupled immediately as a Pfp ester (3 eq . ) with the addition of 1 eq Dhbt-OH in DMF (2 ml) for 2 h. The acylation was checked by the ninhydrin test performed as described above. The remaining amino acids according to the se¬ quence were coupled as Fmoc-protected Pfp esters (3 eq.) with the addition of Dhbt-OH (1 eq . ) in DMF (2 ml) . Excess reagent was removed by DMF washing (12 min flow rate 1 ml/min) and acylations were checked by the ninhydrin test performed 80 °C as described above. The Fmoc group was deprotected as described above. After completed synthesis the peptide-resin was washed with DMF (10 min, flow rate 1 ml/min), DCM (3x5 ml, 1 min each) , diethyl ether (3x5 ml, 1 min each) and dried i n vacuo .

The peptide was cleaved from the resin as described above and freeze dried from glacial acetic acid. The crude freeze dried product was analyzed by hplc and found to be homogeneous without deletion and Fmoc- protected sequences .

Example 15. Peptide synthesis of Tyr-Gly-Gly-Phe-Leu- Lys -0H on NovaSyn TentaGel ■

Dry NovaSyn TG resin (0.29 mmol/g, 250 mg) was placed in a polyethylene vessel equipped with a poly¬ propylene filter for filtration and treated as de¬ scribed under "batchwise peptide synthesis on PEG-PS"

until finishing the pre-sequence Lys6- The following amino acids forming the Leu-enkephaline sequence were coupled as preformed Fmoc-protected HObt esters (3 eq . ) in DMF (5 ml) generated by means of DIC. Before each of the last five couplings the resin was washed with a solution of Dhbt-OH (80 mg in 25 ml), in order to follow the disappearance of the yellow color as the coupling reaction proceed. When the yellow color was no longer visible the couplings were interrupted by washing the resin with DMF (5 x 5 ml, 5 min each) . The acylations were then checked by the ninhydrin test performed at 80 °C as earlier described. After completed synthesis the peptide-resin was washed with DMF (3x5 ml, I min each), DCM (3x5 ml, 1 min each), diethyl ether (3x5 ml, 1 min each) and dried in vacuo .

The peptide was cleaved from the resin as described above and freeze dried from acetic acid. The crude freeze dried product was analyzed by hplc and found to be homogeneous without deletion and Fmoc-protected sequences. The purity was found to be better than 98% and the identity of the peptide was confirmed by ES- MS. Yield 84%.

Table 1.

Amino acid Peptide-resin Coupling time

Pre-sequence: (Lys (tBoc) ) 6 With pre-sq. Without pre-sq.

Fmoc-Leu-OH H-Lys (Boc) 6-HMPA-R < 2 min

Fmoc-Phe-OH H-Leu-Lys (Boc) 6-HMPA-R < 5 min

Fmoc-Gly-OH H-Phe-Leu-Lys (Boc) 6-HMPA-R < 2 min

Fmoc-Gly-OH H-Gly-Phe-Leu-Lys (tBoc) 6-HMPA-R < 2 min

Fmoc-Tyr-OH H-Gly-Gly-Phe-Leu-Lys (tBoc) 6-HMPA-R < 1 min

CO

H-Tyr-Gly-Gly-Phe-Leu-Lys (tBoc) 6-HMPA-R

R= NovaSyn TG derivatized with the linker 4-hydroxymethyl-phenoxy acetic acid (HMPA) (Subst.: 0.29 mmol/g) . Coupling times are based on the Dhbt-OH and the ninhydrin test

Materials and methods

Abbreviations :

AM, p-[ (R,S)-a[l-(9H-fluoren9-yl) -methoxyformamido]-

2 , 4-dimethoxybenzyl ] -phenoxyacetic acid At, 7-azabenzotriazol-1-yl

Boc, tert . butyloxycarbonyl

BOP, benzotriazolyl-oxy-tris-(di thylamino) -phospho- nium hexafluorophosphate

Bpoc , biphenylpropyloxycarbonyl Bt, benzotriazol-1-yl tBu, tert. butyl

DBU, diazabicyclo[5,4,0 ]undec-7-ene

Ddz , 3 , 5-dimethoxyphenylisopropyloxycarbonyl

DCC, dicyclohexylcarbodiimide DCM, dichloromethane

DIC, diisopropylcarbodiimide

DIEA, N,N-diisopropylethylamine

DMAP, 4- (N,N-dimethylamino)pyridine

Dhbt-OH , 3 , 4-dihydro-3-hydroxy-4-oxo-1,2,3-benzo- triazine

DMF, N,N-dimethylformamide

Dts , dithiasuccinyl

EDT, ethanedithiol

FAB, fast atom bombardment Fmoc, 9-fluorenylmethyloxycarbonyl

HATU, 0-( 7-azabenzotriazol-1-yl ) -I , 1 , 3, 3- tetra ethyluronium hexaflurophosphate

HBTU, 0-(benzotriazol-l-yl)-l, 1,3,3- tetramethyluronium hexafluorophosphate HMPA, 4-hydroxymethylphenoxyacetic acid

HMPB, 4-(4-hydroxymethyl-3-methoxyphenoxy) -butyric acid

HObt, 1-hydroxybenzotriazole

HOAt , l-hydroxy-7-azobenzotriazole

HPLC, high pressure liquid chromatography

MCPS, multiple column peptide synthesis

MHC, major histocompatibility complex

MMa, 4-methoxymandelic acid NMR, nuclear magnetic resonance

NHS , N-hydroxy-succinic acid imido ester

NMP, N-methylpyrrolidone

NPS , nitrophenylsulfenyl

Mtr, 4-methoxy-2,3, 6-trimethy1phenylsulfonyl PAM, phenylacetamidomethyl

Pbf, 2,2,4,6, 7-pentamehtyldihydrobenzofuran-5- sulfony1

PEG-PS, polyethyleneglycol grafted on polystyrene

PepSyn Gel, polydimethylacrylamide resin functional- ized with sarcosine methylester

PepSyn K, Kieselguhr supported polydimethylacrylamide resin functionalized with sarcosine methylester

Pfp, pentafluorophenyl

P c , 2,2,5,7, 8-pentamethylchorman-6-sυ1 fonyl Poc , phenylisopropyloxycarbonyl

TBTU, 0- (benzotriazol-1-yl ) -1 , 1 , 3 , 3- tetramethyluronium tetrafluoroborate

TFA, trifluoroacetic acid

TFE, trifluorethanol TFMSA, trifluoromethanesulfonic acid

Tmz , 2,4, 5-tetramethylbenzyloxycarboxyl

REFERENCES:

1. Merrifield, R.B. (1963) J. Amer. Chem. Soc . 85, 2149 - 2154.

2. Kent, S.B.H. (1988), Annu Rev. Biochem. 57, 957 - 989.

3. B.D. Larsen, C. Larsen, and A. Holm. , Pep¬ tides 1990; E. Giralt and D. Andreu, (Eds) . 1991 ESCOM Science Publishers B.V., p. 183 - 185.

4. Larsen, B.D. , and Holm, A. (1994) , Int. J. Peptide & Protein Res. 43 , 1 - 9.

5. Due Larsen, B. , Christensen, D.H. , Holm, A., Zillmer, R. , and Faurskov, 0. (1993), J.Amer.Chem. Soc . 115, 6247 - 6253.

6. Chou, P.Y. and Fasman, G.D. (1978) , Annu Rev. Biochem. 47, 251 - 276.

7. Dryland, A. and Sheppard, R.C. (1986) J.Chem. Soc. Perkin Trans. I, 125 - 137.

8. Merrifield, R.B., Singer, J. and Chait, B.T. (1988) Anal. Biochem. 174, 399 - 414.

9. Cameron, L.R. , Meldal, M and Sheppard, R.C. (1987), J. Chem. Soc. Chem. Co mun, 270-272.

10. Meldal, M. (1993) , Peptides 1993; Schneider, CH. and Eberle, A.N. (Eds) . 1993 ESCOM Science Pub¬ lishers B. V. , p. 61-62.

11. Sarin, V.K., Kent, S.B.H., Tarn, J.P. and Mer¬ rifield, R.B. (1981), Anal. Biochem, 117, 147-157.

12 Carpino, L.A. and Han, G.Y. (1972), J. Org. Chem. 37, 34P4-34P9

13 Stewart, J.M. and Young, J.D. , Solid phase peptide synthesis, Pierce Chemical Company (1984)

14 Cameron, L. , Meldal, M. and Sheppard, R.C. (1987) , J. Chem. Soc. Chem. Commun , 270-272.

15. Meldal, M. , Bisgaard Holm , C. , Boejesen, G., Havsteen Jakobsen, M. and Holm, A. (1993), Int. J. Peptide and Protein Res. 41, 250-260.

16. McKenzie, A. , Pirie, D.J.C. (1936), Berichte 69, 868.

17 Knorr, E. (1904), Berichte 37, 3172

18. Peptides: Synthesis, structures and applica¬ tions. Gutte, B. , Ed. Academic Press Inc. 1995.