The ABRF Peptide Synthesis Workshop featured three presentations: "Biological Results of Poorly Characterized Synthetic Peptides" by Dr. Ruth H. Angeletti, Albert Einstein College of Medicine, "Intramolecular Disulfide Bond Formation of Synthetic Peptides" by Dr. Gregg B. Fields, University of Minnesota, and "Synthesis and Characterization of Phosphorylated Peptides" by Dr. Daniel R. Marshak, Cold Spring Harbor Laboratories. The first presentation served as a warning to researchers as to the erroneous conclusions about biological activity that can be made when synthetic peptides are not well characterized. The other two presentations described useful synthetic and characterization methods for incorporating disulfide bonds and phosphorylated residues into peptides via Fmoc chemistry.
Biological results obtained with synthetic peptides may be subject to misinterpretation if several criteria are not carefully considered. For example, was the desired product obtained? Is the peptide pure? Is the peptide itself the biologically active component? Careful analysis of synthetic peptides, including mass spectra, is needed to ensure that subsequent activities are not due to contaminants. A recent example of this problem involved a 20 residue peptide that was deemed inhibitory to catecholamine release (1). Later analyses of the peptide product revealed a 455 Da contaminant, not the intact peptide, was responsible for the activity (2). In a similar vein, inhibition of HIV infection was found for a CD4(83-94) peptide byproduct containing a benzyl side-chain protecting group still attached to Cys, but not for the intact CD4(76- 94) peptide itself (3).
Even if the desired peptide is obtained, it must be determined if the peptide is soluble and stable under the assay conditions. A peptide that "doesn't work" may simply be insoluble, a problem that can often be resolved by using organic solvents to dissolve the peptide. Appropriate controls should be utilized to determine the effect of such solvents on biological activity. Finally, after having carefully characterized the peptide, biological results may be unexpected due to an incorrect hypothesis. Unexpected results, carefully interpreted, can provide insight into new biological mechanisms and hypotheses.
Intramolecular disulfide bond formation has achieved a new level of applicability with the development of conformationally constrained peptide libraries, where the constraint can result in increased affinity (4). The two most general approaches for the formation of single intramolecular disulfide bonds are (i) the removal of Cys side-chain protecting groups followed by air oxidation or (ii) concomitant removal of the Cys side-chain protecting groups and formation of the disulfide bond. The most convenient method is to use approach (ii) with resin-bound peptides, taking advantage of the solid-phase. Using octreotide (a somatostatin analog, sequence: D-Phe-Cys-Phe- D-Trp-Lys-Thr-Cys-Thr(ol))] as an example, disulfide bond formation was readily achieved with Cys(Acm) residues and thallium trifluoroacetate [Tl(Tfa)3] in dimethyl formamide (5). To avoid Trp modification by Tl(Tfa)3 during cyclization, Trp was side-chain protected by the Boc group (6). Characterization of the RP-HPLC purified product indicated that disulfide bond formation was highly efficient and that Trp modification had not occurred. This method is also applicable for Met-containing peptides, as the addition of thioanisole (to inhibit Met oxidation) during trifluoroacetic acid cleavage of the peptide-resin does not disrupt the disulfide bond.
Two methods were suggested for disulfide bond formation after cleavage of the peptide. The first is treatment of Cys(Acm) residues with mercuric acetate [Hg(OAc)2] and subsequent precipitation of the mercury with beta- mercaptoethanol (7). To avoid HgOAc modification of Trp, the reaction is performed in 50% acetic acid/water (8). The second option uses Cys(Trt) residues, where the Trt groups are removed during peptide-resin cleavage, followed by air oxidation of the free sulfhydryl groups in 10% dimethyl sulfoxide/water (9).
The related issue of Cys racemization was also discussed. For some peptides it will be desirable to form disulfide bonds that constrain the entire sequence, i.e. Cys residues are located at the N- and C-terminus of the peptides. However C-terminal esterified Cys residues are subject to racemization during Fmoc removal (10). Racemization can be inhibited by using sterically hindered linkers, such as 2- chlorotrityl, to form the desired ester bond (11). Alternatively, a peptide-amide linker could be used, as C- terminal amidated Cys residues do not racemize (10).
Synthetic phosphopeptides have utility as standards for peptide maps, antigens for phosphorylation site-specific antibodies, substrates for protein kinases and phosphatases, and probes of structure-activity relationships. The specific application discussed here was to study Cdc2 protein kinase, which is involved in cell division. Phosphorylation of the Cdc2 kinase catalytic subunit is important for cell-cycle regulation. The general approach used for phosphopeptide synthesis was (i) solid-phase peptide assembly by Fmoc chemistry using Ser, Thr, and/or Tyr without side-chain protection, (ii) derivatization of the unprotected hydroxyl group with phosphoramidite by treatment of the peptide-resin, (iii) oxidation of the peptide-resin phosphite to the phosphate, (iv) cleavage and deprotection of the peptide with trifluoroacetic acid, and (v) peptide purification and characterization. Step (ii) was achieved with a 1.5 h treatment of di-tert-butyl-(N,N- diisopropyl)phosphoramidite, while step (iii) utilized a 1.0 h application of tert-butyl hydroperoxide. It should be noted that the N-terminal amino acid was incorporated as a Boc derivative, thus avoiding a base treatment to remove the N-terminal protecting group after phosphorylation of the peptide-resin. Base treatment can dephosphorylate Thr via a ß-elimination mechanism (12).
The first synthetic phosphopeptide discussed was Cys- Lys-Lys-Ile-Arg-Leu-Glu-Ser(PO3)-Glu-Glu-Glu-Gly-Val-Pro- Ser-Thr-Ala-Ile-Arg. This Cdc2 kinase derived sequence utilizes Ser39 as a phosphorylation site. Synthesis was reasonably straightforward, and peptide purification achieved with RP-HPLC using water/acetonitrile, 0.1% TFA. Mass spectrometric characterization confirmed the desired product. The second synthetic phosphopeptide discussed was Cys-Arg-Val-Tyr-Thr(PO3)-His-Glu-Val-Val-Thr-Leu-Trp-Tyr- Arg. This Cdc2 kinase derived sequence utilizes Thr161 as a phosphorylation site. Initial RP-HPLC purification and analysis by mass spectrometry indicated the presence of the desired product, the non-oxidized peptide, and the non- phosphorylated peptide. Isolation of the desired product required sequential RP-HPLC, first using dilute TFA as the counter-ion to remove the non-oxidized peptide, followed by dilute HCl as counter-ion to separate the desired product from the non-phosphorylated peptide. The third synthetic phosphopeptide discussed was Cys-Lys-Ile-Gly-Glu-Gly- Thr(PO3)-Tyr(PO3)-Gly-Val-Tyr-Lys. This Cdc2 kinase derived sequence utilizes Thr14 and Tyr15 as phosphorylation sites. Initial RP-HPLC purification and analysis by mass spectrometry indicated the presence of the desired product, single and double non-oxidized peptides, and single and double non-phosphorylated peptides. Isolation of the desired product required iron-chelate affinity chromatography, where differentially phosphorylated peptide species could be separated by elution with acetate buffer. For all three peptides, mass spectrometric characterization was invaluable for isolation of the desired product.
References
Return to the The ABRF Home Page