Solid phase peptide synthesis (SPPS)

The following information should allow the novice or casual reader to understand the basic concepts of SPPS without further need for searching. Other parties especially those already armed with SPPS knowledge are encouraged to search the literature elsewhere.

A short history

In the early 1900s, Emil Fischer introduced the concept of peptides and polypeptides and presented protocols for their synthesis. But because solution based techniques were laborious and slow and the demand for peptides began to rise with the realisation that the investigation of the relationship between chemical structure and biological activity could be more easily investigated using synthetic peptides rather than purified naturally occuring peptides, new techniques were needed to meet ever increasing needs. Coupled to this was the fact that yields were often low and contaminated with side products that gradually build up during the synthesis even in the most experienced of hands. However, in spite of the difficulties, several "difficult" peptides were synthesised throughout the 20s-60s that pushed the known chemistries to their limitations. In 1963, Bruce Merrifield published a landmark paper describing the development of solid-phase peptide synthesis (SPPS). This technique (for which he was awarded the Nobel Prize in Chemistry in 1984) was responsible, more than anything else, for opening the way to the widespread use of synthetic peptides as reagents in chemical and biomedical investigations. Nowadays it is not unreasonable to expect the commonplace synthesis of 30 residue peptides to be within reach of most laboratories that have some degree of specialisation in peptide research and synthesis. Peptides exceeding 100 residues in length have been successfully made.

An overview of the Merrifield technique.

The process of obtaining a synthetic peptide involves several discrete steps that can be represented diagramatically (as shown in figure 1 below) as individual chemical steps. Where chemical groups or functions are used that vary depending upon the exact procedure of synthesis they are depicted generically in highlighted boxes. The chemical groups that are shown are "backbone" structures for the lengthening peptide chain.

The basic principle for SPPS involves the stepwise addition of amino acids to the growing oligopeptide chain that is anchored to a chemically stable particle so that the peptide can be separated from solvents and reagents during its synthesis by simple filtration. Once the synthesis is complete the chain is cleaved from the support and purification takes place in solution. The advantages of this method can not be over emphasised. The separation of reagents can be fully automated, losses due to conventional chemistries (evaporation, recrystalisation, etc.) are, thus, totally avoided and time savings are enormous. However, although an elegant technique, it does have its caveats. For instance, amino acid side chains that have to be protected by stable groups (see figure 1) have to be able to withstand substantial chemical attrition throughout many synthetic cycles but also be fully removable post synthesis. A consequence of unstable protecting groups is unwanted side chain reactions during the synthesis that are carried through the cleavage and end up contaminating the final product. Whilst the SPPS reaction scheme shown in figure 1 is simple enough in principle, in practice the design, synthesis and purification of synthetic oligopeptides is compounded by choices that have to be made and these considerations make the task of synthesis less than straightforward. These considerations include i) the choice between batch and continuous flow, ii) the choice of temporary -amino protection (ie Fmoc or Boc), iii) the types of permanent

Figure 1. SPPS represented diagramatically

side group protection to use, iv) the type of solid support/resin/matrix to use, v) the type of linkage between the first amino acid and the resin to be made, vi) peptide bond formation conditions, vii) type of monitoring, viii) final cleavage conditions and ix) purification regime. Each of the above need individual consideration with respect to each peptide synthesis attempted. Each will be briefly described and elaborated in the following description which, whilst not comprehensive, should allow the uninitiated to comprehend the basics of SPPS. The remaining text will describe very briefly some of the applications that synthetic peptides can be used in and what has been achieved with them (under the final heading, x) applications of synthetic peptides).

SPPS considerations

i) Batch verses continuous flow

SPPS allows for either batchwise or continuous flow reactions to take place. Batch occurs in a filter reaction vessel and can be useful since reactions are accessible and can be added manually or automatically. However, continuous flow can allow for real-time spectrophotometric monitoring of the progress of coupling and deprotection throughout the whole of the synthesis without the cumbersome intervention and assays associated with batchwise synthesis by simply passing the flowing solution through a silica flow cell placed in a spectrophotometer. The majority of labs utilise continuous flow SPPS because of convienience and its compatibility with Fmoc temporary protection schemes. Excellent batch and continuous flow machines are both commercially available, though, and it is probably only through effective marketing that continuous flow machines are the machines of choice for the uninformed.

ii) Fmoc verses Boc protection (temporary protection)

Both Fmoc and Boc protecting groups protect the sensitive -amino group from deleterious reactions during the formation of a new peptide bond between the unprotected carboxy group of the free amino acid and the deprotected -amino of the growing peptide chain. Selective temporary Boc group deprotection is achieved by trifluoroacetic acid (TFA) treatment during peptide chain extension immediately prior to the addition of each new amino acid residue whilst the permanent protecting side groups are removed by stronger acidolysis (usually hydrogen fluoride (HF)) once the synthesis is complete (shown below in figure 2).

Fig. 2 Removal of temporary and permanent protecting groups by graduated acidolysis in Boc chemistry.

Boc chemistry has lost favour recently, however, since HF is extremely hazardous and requires very expensive laboratory equipment and it is now generally accepted that the repetitive TFA treatment during deprotection can lead to alteration of sensitive peptide bonds as well as acid catalysed side reactions. Fmoc protection was developed subsequently to avoid these problems and is now widely accepted throughout the world scientific community even though Fmoc reagents are still quite costly in comparison to their Boc relations. Selective temporary Fmoc-group deprotection is achieved by mild base treatment (usually piperidine) during chain extension and the permanent protecting side groups are removed by moderate acidolysis (usually TFA) once the synthesis is complete. This mild orthogonal deprotection scheme causes less problems than the equivalent graduated acidolysis of Boc deprotection (shown below in figure 3).

Fig. 3 Removal of temporary and permanent protecting groups by orthogonal base/acid deprotection in Fmoc chemistry

Boc and Fmoc groups have risen to the fore as the most widely used and commercially viable N -amino protecting groups for SPPS. Other N -amino protecting groups do exist such as 2-(4-biphenyl)propyl[2]oxycarbonyl (Bpoc) and 6-nitroveratyl-oxycarbonyl (Nvoc) that are removed by dilute TFA and photolysis respectively, and as such are designed to reduce the severity of the deprotection chemistry but such protecting groups will not be discussed here. In general, then, peptide synthesis on a routine basis (where speed, convenience, low cost and flexibility is called for) is carried out on a solid phase matrix, in a continuous flow reaction vessel utilising Fmoc temporary protection chemistry. The remaining text for this chapter describes the other considerations for peptide synthesis with regard to only SPPS, continuous flow, Fmoc chemistry.

iii) Permanent protecting groups for reactive amino acid side chains of Fmoc amino acids

Picking the best amino acid derivative is probably the most important and often the most difficult aspect of peptide synthesis. For certain amino acids (e.g. Cys, Asp, Glu and Lys), side-chain protection is absolutely essential, whereas for others, an informed decision should be made depending upon length and other considerations. Protection must be compatible with Fmoc N -amino protection, that is to say, not cleaved by base treatment. t-Butyl-based side-chain protecting groups cleavable by a single mildly acidic treatment are chosen where possible but these do not exist for some amino acids or have been found to be unsuitable so other side-chain protecting groups have arisen with similar acid labilities to t-butyl to ensure deprotection compatibility.

The table above lists the most commonly used protecting groups, the amino acids usually protected by the particular group with its common notation in brackets and the reagent of choice for deprotection during cleavage. It should be stressed that because of the way these protecting groups are eventually removed during peptide cleavage from the solid phase matrix they are not all mutually compatible with each other and careful selection of these groups is required to eliminate unwanted chemical modifications. Even with carefully chosen protection, they are still susceptible to modification. For instance, the tert-butyl ester (OtBu) protecting the carboxyl of Asp from cyclization to form succinimide and the subsequent reopening of the ring to yield undesirable aspartyl peptides can undergo an intramolecular elimination to form an aspartimide, which can then partition in water to the desired a-peptide and the undesired by-product with the chain growing from the b-carboxyl. Hundreds more protecting groups exist but are usually designed with very specific purposes in mind. Side-chain protecting groups also exist that can protect simultaneously both Fmoc and Boc derivatives thus avoiding the need for expensive duplication of derivatives in laboratories that utilise both Fmoc and Boc chemistries.

iv) Solid supports and matrices

Supports are often polystyrene suspensions (relatively polar in nature) cross-linked with 1% 1,3-divinylbenzene. Dry polystyrene beads have an average diameter of about 50mm, but with the commonly used solvents for peptide synthesis (i.e. DMF which is relatively polar), they tend to swell considerably in volume. Thus, the chemistry of SPPS takes place within a well-solvated gel containing mobile and reagent-accessible chains. DCM is often used since it apolar nature means maximal solvation of the polystyrene. Polymer supports have also been developed however that have comparable polarities (i.e. polar) with the peptide backbone (e.g. polyamide or Pepsyn) so that single solvent systems can be utilised. This is preferable since DMF has been shown to facilitate, for instance, acylation at faster reaction rates than less polar media such as DCM or chloroform. Under ideal conditions both polystyrene and polyamide beads have reaction rates approaching (but not reaching) those attainable in solution. Other supports exist (composites) whose characteristics try to combine the physical rigidity required for the column backpressures associated with continuous flow SPPS with the solvation and permeability required for reaction rates to remain high. Some manufacturers are so confident that their supports meet these requirements that terms such as transparent to soluble reactants?are now commonplace. All three types are still regularly used with Fmoc chemistry since the effectiveness of each is measured by the degree of substitution that the matrix is capable of for a given mass and this substitution level is generally picked by the user according to the length of the peptide to be synthesised. Generally the longer the sequence the lower the substitution should be since as steric hinderance increases with chain length the chances of incorrect chemistries and in practice the choice of resin type depends more on how the first amino acid is to be attached to the resin (see linking amino acids to resins and functional linkers).

v) Linking amino acids to resins and functional linkers

Syntheses are carried out in the C to N direction and, therefore, generally start with the intended C-terminal residue of the desired peptide being linked to the support either directly or by means of a suitable functional linker. These are designed so that eventual cleavage provides either a free acid or amide at the c-terminus. Linkers are the same irrespective of the resin used.

Peptide acids

The HMP/PAB linker generates a free acid either on resin or as a handle after cleavage in 50-100% TFA for 1-2hr at 25°C. Other supports and linkers that can be cleaved in less harsh conditions (i.e. dilute acids under certain specified circumstances) so that protected peptide segments retaining side chain tert-butyl protection can be generated e.g. Rink acid, Sasrin, HMPB and HAL. These fragments can be used to generate sequences that prove difficult to synthesise de novo. The latter are all esters; and rates and yields of reactions for ester bond formation are lower than those for corresponding methods for amide bond formation (see formation of peptide bond) so that compromises are needed to achieve reasonable loading reaction times and substitution levels, while ensuring that the extent of racemization remains acceptably low.

Peptide amides

Most anchoring linkages that ultimately provide C-terminal peptide amides in a useful and general manner are benzhydrylamide derivatives. Attachment is via direct coupling of an N -protected amino acid by means of it carboxyl to an appropriate benzhydrylamine resin with eventual cleavage at a different locus providing the desired carboxamide. This system is both Fmoc and Boc compatible. In Fmoc chemistry the system allows electron donating methoxy groups to create a TFA-sensitive Rink amide, Dod, Breipohl amide or SAMBHA linker compatible with Fmoc cleavage chemistry.

vi) Formation of peptide bonds

There are four major coupling techniques that allow the stepwise introduction of N -protected amino acids for SPPS. In essence all are designed to force reactions to completion since the presence of unreacted amino groups causes truncation of sequence.

1. In situ reagents

Classic example is N,N'-dicyclohexylcarbodiimide (DCC) as shown below.

N,N'-Diisopropylcarbodiimide (DIPCDI) a related agent is more often used because a urea coproduct that needs to be removed from the reaction is more soluble in DCM (when used as a solvent). Carbodiimide-mediated couplings are speeded up significantly by the addition of 1-hydroxybenzotriazole (HOBt) (see below) with the bonus that possible dehydration of Asn and Gln residues is avoided as well as suppressing racemization.

Other in situ coupling reagents have been developed recently to further enhance reaction rates, reduce side reactions and make their use easier. Most are based on phosphonium or uronium salts which in the presence of a tertiary base, can smoothly convert protected amino acids into a variety of activated species (eg BOP, PyBOP, HBTU and TBTU that all generate HOBt esters).

2. Active esters

The classical 2- and 4-nitrophenyl esters (ONo and ONp respectively) although allowing dehydration free introduction of Asn and Glu are relatively slow reactions. Extensive studies on other active esters have allowed the evolution of the following important reagents now routinely used during Fmoc coupling. HOBt esters of protected amino acids are easily formed (see above) and react extremely fast in most cases. Pentafluorophenyl (OPfp) esters (see below) are also efficient acylating agents that react slowly but their chemical structures provide little opportunity for side-reactions. One minor problem is that some amino acid OPfps are not available (since they do not crystallise) but the 1-oxo-2-hydroxy-dihydrobenzotriazine (ODhbt) esters (see below) provide suitable alternatives in such cases.

3. Preformed symmetrical anhydrides (PSA)

Although sometimes used in Fmoc chemistry they are generally consigned to Boc since intermediates formed during PSA generation can undergo rearrangement and the chemistry is very wasteful.

4. Acid halides

Na-protected amino acid chlorides have a long history of use in solution synthesis but their use in SPPS is limited because the Boc group is unstable in all solvents used to prepare acid chlorides. The Fmoc group does survive acid chloride preparation and has found limited use in continuous flow SPPS.

vii) Monitoring

A crucial issue for stepwise SPPS is the repetitive yield per deprotection/coupling cycle. There are numerous ways of monitoring these steps, including some with real time feedback based on the kinetics of appearance or disappearance of appropriate soluble chromophores measured in a flowthrough system. The following methods are all performed on aliquots taken from either batch/manual synthesis as it progresses or single-residue automatic synthesis and do not lend themselves well to flowthrough systems. They are all destructive but yield the best information (both qualitatively and quantitatively).

Qualitative colour methods

Ninhydrin, trinitrobenzene sulphonic acid and isatin are all used to detect resin bound amino groups i.e. test for incomplete acylation. These tests should be negative before synthesis is continued.

Quantitative amino acid analysis

In general, amino acid analysis is too slow to be used routinely for the analytical control of synthesis whilst in progress. Also total acidic hydrolysis can partly or nearly completely destroy some amino acids (e.g. serine and tryptophan respectively) so that it is barely adequate for control of SPPS. For final confirmation of sequence, however, this is an absolute must.

HPLC examination

Intermediate peptides can be rapidly cleaved from an aliquot of resin that contain acid-labile linkages, however for analytical purposes brief TFA treatment of normal linkers generally yields enough peptide for analysis. Used mostly to yield information at the end of synthesis, during cleavage or for reverse phase purification it can be very useful during runs since microbore columns can be run in a matter of minutes.

The following are suitable for the monitoring of flowthrough systems and both are used commonly in commercially available equipment. They are non-invasive, non-destructive techniques that provide information for immediate use by the user to assess coupling and deprotection efficiencies. This information can then be acted upon to correct an aberrant reaction rather than allowing the synthesis to proceed. In this way, a successful synthesis needs no user intervention (i.e. is fully automated) whilst glitches in a synthesis need only minor user intervention (e.g. the manual recoupling of an amino acid or further deprotection).

Counterion Distribution Monitoring

This system allows real-time monitoring of active ester coupling reaction by following at 600nm release of a reporter dye from the solid support. Figure 4, box A, shows a stylised monitoring trace from a NovaSyn Crystal automated peptide synthesizer. The end point of the coupling reaction is indicated as the absorbance level reaches a plateau.

Release of the cleaved Fmoc group (detectable at 300-320nm)

As deprotection occurs, the speed of release of the temporary Fmoc protecting group can be used to measure the efficiency of this reaction. A normal deprotection peak is shown in figure 4, box B, trace A. When deprotection is delayed for any reason this can be seen as a reduction in the rate of release of Fmoc as shown in figure 4, box B, trace B. The area under the curve also gives a crude estimate of the total Fmoc released and thus (given the known Fmoc concentration to start with) how efficient the deprotection procedure was and ultimately coupling efficiency.

Fig. 4 Flowtrough monitoring system traces. A shows an acylation trace whilst B shows a deprotection monitoring trace.

viii) Final cleavage conditions

Having successfully synthesised a protected peptide, one is confronted with a difficult task of having to simultaneously detach the peptide from the resin and remove all side-chain protecting groups to yield the desired peptide. This process is always hampered by deleterious side-reactions involving certain side-chain protecting groups which are liberated as stable cations during TFA cleavage and deprotection. The majority of these side-reactions involve modifications of sensitive amino acid residues such as Trp, Met, Tyr and Cys by TFA-liberated protecting groups like Mtr and Pmc of Arg, Tmob and Mbh of Asn and Gln and Trt of Cys, Asn and Gln. Many other reactions can and do occur not only between side-chains and protecting groups but also linkers. An ideal cleavage and deprotection protocol, thus, does not exist but there are generally accepted schemes. All cleavage mixtures can be divided, then, into two categories, i) for peptides which do not contain sensitive amino acid residues and protecting groups and ii) for complex peptides which contain one or more of the problem species detailed above. The latter mixtures contain a multiplicity of scavengers to quench all the liberated reactive carbonium ions originating from the protecting groups and/or the linkers on the resin. Most TFA cleavage protocols recommend cleavage at room temperature whilst times depend on sequence, resin type, and side chain protection. It is usual to monitor cleavage via direct HPLC using small aliquots removed from the cleavage vessel as it proceeds. Just prior to cleavage, the peptide-resin is usually washed with a mild acid (e.g. acetic) to remove residual basic DMF that might interfere with TFA-acidolysis then extensively dried overnight over KOH. The following chart shows a typical deprotection/cleavage scheme that deals with the aforementioned difficult residues/protecting groups.

Basic flowchart for Fmoc cleavage of peptide-resin. EDT is ethanedithiol, TIS is triisopropylsilane, TMSBr is bromotrimethylsilane and TA is thioanisole.

When cleavage is complete, the cleavage mixture is separated from the resin by filtration and the TFA is removed by evaporation (TFA is volatile and its removal from the cleavage mixture is said to improve the yield of crude peptide upon precipitation). Scavengers are then removed from the peptide by precipitation of the peptide in a suitable solvent (e.g. ether). After drying the crude product is stored dry at -18 C or subjected to clean-up procedures.

ix) Purification regimes

For some applications (e.g. antibody blocking prior to immune detection) the peptide can be used in its crude state. However it is usual to determine what this crude state?is via either HPLC and/or MS on a small aliquot of crude peptide. Regardless of ultimate usage the minimum requirement is usually some form of desalting to remove traces of scavenger (on Sephadex or its equivalent) then lyophilisation prior to storage. It is inevitable that the products of SPPS will be initially impure. For instance, a peptide 15 residues long usually requires 30 successive steps in its assembly. A 99% yield in every one of these steps will theoretically lead to a product containing only 74% of the desired sequence. Importantly, the crude peptide will contain 26% of very closely related impurities. By-products must be removed if they are not to effect the biological activity of the desired peptide in any assay system however they are usually so closely related physicochemically that purification becomes very difficult. Conventional separation procedures are still valid (e.g. gel filtration) however these are mostly superseded by reverse-phase chromatography which depends largely on hydrophobicity differences. Reverse-phase HPLC is more or less the minimum requirement within peptide synthesis laboratories with some ion-exchange chromatography complementing awkward separations. Reverse-phase HPLC can be used both qualitatively and quantitatively to achieve purities (in the best instances) of over 98%. Reverse-phase chromatography can be executed on low pressure systems with the benefit of low cost but with some loss of resolution. Depending on the peptide size and hydrophobicity, packings of C18, C8, C4 or diphenyl are recommended. A polar/non-polar solvent system such as water/acetonitrile should be used for gradient formation with an acidic ion pairing reagent like TFA. Solubility is a problem with peptides (caused by hydrophobic residues, aggregation, disulfide formation between Cys side-chains, secondary structure formation and high content of acidic residues) with most rarely dissolving completely in water but this can usually be overcome by the addition of acetic acid for basic peptides and ammonia for acidic peptides with other peptides being soluble in a variety of buffered systems or organic solvents. Purified peptides should be keep lyophilised, frozen, under nitrogen and with no residual acid components. Freeze-thaw cycles are detrimental.

x) Applications of synthetic peptides

Whilst by no means a complete list, the following detail some of the main ways that synthetic peptides have been and can be used.

Antigenic and immunogenic uses

Elucidation of antigenic peptide sequences in proteins by epitope mapping. Although powerful this technique is restricted to small proteins by the expense of producing all possible internal sequences and time. Prediction of protein antigenic regions via either tertiary structure information obtained from x-ray or NMR or information obtained from the primary structure and predicted expected surface properties.

Novel analogue mimicry of biological sequences for the synthesis of potent and selective analogues of peptide hormones and neurotransmitters.

Antigenic peptides in the isolation of gene products i.e. predicting the protein structure of an unknown gene product after the gene has been identified and using this structure to raise antibodies that would allow the proteins identification.

Mimotopes and peptide libraries i.e. the construction of synthetic peptide libraries that are screened for activities or binding to antibodies that by definition are not necessarily identical to native epitopes.

Antibody production.

Affinity chromatography.

Enzyme inhibitors

Potential therapeutic agents for either the replacement of expensive, difficult to purify and difficult to produce biotics or alternatives to common biotics such as antibacterials.

Antisense peptides

Interestingly either 5' to 3' or 3' to 5' reading of the complementary nucleotide sequence displays the same pattern in hydropathicity of the coded amino acid sequence with respect to the same sense peptide sequence. This stems from the fact that the middle base of the triplet codon specifies the hydropathic nature of the amino acid and this is the same whether 5' to 3' or 3' to 5' (termed anticomplementarity). Although not studied extensively, antisense peptides, in general, have shown some selectivity in binding with their sense counterparts. Potential applications could include an antisense peptide acting as a sink for an elevated level of an endogenous sense peptide thereby managing a disease caused due to elevated levels of the sense peptide.

Affinity labelling of receptors

The establishment of some type of irreversible, but specific, interaction between a ligand and its biological acceptor is central in attempts to identify, characterise and isolate receptors for hormones, neurotransmitters, signal transduction elements, etc. The establishment of a permanent covalent bond between the two elements via either chemical affinity labeling or photoaffinity labelling has found numerous applications in protein/peptide-hormone receptor systems.

Structure-function studies

Structure-function studies that aim to describe native ligand binding to a receptor or receptor class can use a variety of peptide agonist or antagonists to help study "binding sites" Such peptides include deletion peptides, stereochemical substitutions (i.e. D for L forms), isoteric substitutions (i.e. replacement of similarly sized but differently charged amino acids), constrained peptides (e.g. a-alkylated amino acids induce b-turn structures) and peptides with variable ring size or ring constraints.

Peptide-based vaccines

Most work has been done on antiviral agents, such as influenza and HIV. Some work has also been done on antibacterial peptide vaccines (e.g. diptheria and cholera toxins) as well as antiparasitic immunogens for malaria prevention. The peptide based vaccines are usually rendered immunogenic by conjugation to carriers such as BSA, tetanus toxoid, ovalbumin and KLH through either side chain groups or primary amino groups. In conjugating the peptide it is, however, unclear what conformational or topographical constraints are placed on antigen or whether the peptide is uniformly coupled. Also the carrier can induce hypersensitive responses. To circumvent such problems the multiple antigen peptide (MAP) system and the multivalent B and T cell epitope vaccine system have emerged (see figure 5).

Fig. 5 Carrier-free peptide vaccine vectors. A shows a cartoon of the basic elements of the multiple antigenic peptide (MAP) with eight copies of the antigenic peptide. B shows a cartoon of the basic elements of the multivalent B and T cell epitope vaccine.

References

1. The following references were all used to compile the above information. In many instances, text is close to the original and I fully credit the original authors. The reader should realise that all the above text is grossly over-simplified and is encouraged to consult the literature for more detail or recent developments.
2. Atherton E. and Sheppard R. C. - Solid phase peptide synthesis - a practical approach IRL Press Oxford New York Tokyo.
3. Gloor A. P., Hoare S. M., Lawless K., Steinauer R.A., White P., Yong C. W. Synthesis notes, Novabiochem.
4. Grant A. G. Synthetic peptides - a user抯 guide W. H. Freeman and Company New York.
5. Pennington M. W. and Dunn B. M. Peptide synthesis protocols.

若您有任何意见或建议，请及时E-mail通知我们，谢谢！