(1) Skyline Peptides, 1901 Shoreline Dr. #210, Alameda, CA 94501, (2) Albert Einstein College Med., Bronx, NY 10461, (3) SmithKline Beecham Pharm., King of Prussia, PA 19406, (4) Cold Spring Harbor Lab., Cold Spring Harbor, NY 11724, (5) Stanford Univ., Stanford, CA 94305, (6) Genentech, Inc., South San Francisco, CA 94080,(7) USC, Los Angeles, CA 90033, (8) Yale Univ., New Haven, CT 06536, (9) Univ. of Minnesota, Minneapolis, MN 55455, U.S.A.
The ABRF presentation at the 13th American Peptide Symposium Workshop provided a summary (which is also in reference 1) of the 1991-1993 studies carried out by the ABRF Committee on Peptide Synthesis and Mass Spectrometry. This committee was formed to evaluate the quality of the synthetic methods utilized in its ~ 130 member laboratories that are engaged in the synthesis and structural analysis of peptides. For each of the last three years ABRF member laboratories have synthesized and/or cleaved a different peptide (2-4). Problems associated with peptide assembly and cleavage were evaluated by the Committee using amino acid analysis (AAA), reversed-phase high-performance liquid chromatography (RP-HPLC), capillary electrophoresis (CE), Edman degradation sequence analysis, and electrospray (ES), plasma desorption (PD), fast atom bombardment (FAB), and laser desorption (LD) mass spectrometry (MS).
The 1991 ABRF test peptide sequence was Val-Lys-Lys-ArgCys-Ser-Met-Trp-lle-lle-Pro-Thr-Asp-Asp-Glu-Ala.The sequence was designed to incorporate difficult couplings (Trp-lle-lle) and to provide potential cleavage side-reactions such as alkylation of Trp and oxidation of Cys and Met. A total of 36 crude products were submitted, l 8 synthesized by Boc chemistry and 18 by Fmoc chemistry. AAA indicated that all samples had correct compositions (i.e., there were no significant amounts of deletion or truncation products). The average content of the desired product was 44% by RP-HPLC and 53% by ESMS. As evaluated by ESMS, only 10 of the 18 Boc-synthesized crude products contained >25% of the desired product, while all 18 Fmoc-synthesized crude products fell into the same category. Examination of the nature of the non-desired products revealed 8 of the Boc-synthesized peptides contained a single dehydration and/or oxidation, while 11 of the Fmoc-synthesized peptides contained >10% of a tBu adduct. Thus, the respective cleavage conditions of Fmoc and Boc solid-phase peptide synthesis were the primary source of synthetic difficulties for the 1991 test peptide, since non-desired products were not deletions or truncations but rather the result of incomplete deprotections and/or other covalent modifications.
The 1992 ABRF test peptide sequence was Gly-Val-Arg-Gly-Asp-Lys-Gly-Asn-Pro-Gly-Trp-Pro-Gly-Ala-Pro-Tyr. The sequence incorporated numerous potential synthesis and cleavage side-reactions including dehydration of Asn and alkylation of Trp. A total of 58 crude products were supplied for the study, 16 synthesized by Boc chemistry and 42 by Fmoc chemistry. AAA showed 51 of 58 crude products to be compositionally correct (Trp was not quantitated), which was consistent with RP-HPLC analysis indicating 51 of 58 crude products had >25% of the apparent desired peptide. Deletion peptides detected by Edman degradation sequence analysis were des(Asn8), des(Pro12), des(Gly1, Gly7, Asn8), and des(Gly1, Gly4, Trp11). Assessment of product purity by ESMS, FABMS, and PDMS showed semiquantiative agreement with RP-HPLC analyses. As estimated by ESMS, 30 of the 58 crude samples (52%) and 26 of the 33 purified samples (79%) contained >75% of the desired product. This represents a considerable improvement over the 1991 study, where the percentage of crude and purified products containing >75% of the desired material were 28 and 65%, respectively. Possible reasons for these improved results were any combination of (i) a synthetic peptide sequence less susceptible to side reactions induced during peptide-resin cleavage, (ii) the greater fraction of peptides synthesized by Fmoc chemistry, where cleavage conditions are less harsh, and (iii) more rigor and care in laboratory techniques following the 1991 results.
Further examination of the 1992 samples showed 5 of the crude Fmoc syntheses and 2 of the crude Boc syntheses to contain poor yields (<25%) of the desired product. Five of the 7 poor yield crude products were the result of deletion peptides (4 from Fmoc syntheses, 1 from Boc syntheses). Of the 4 Fmoc-synthesized crude peptides that contained >10% des(Asn), 2 were assembled without Asn side-chain protection. The only Fmoc-synthesized peptide for which all three mass spectrometric techniques detected >10% dehydration was assembled without Asn side-chain protection. This dehydration was probably the result of a side reaction (nitrile formation) attributable to the use of unprotected Asn during coupling. Thus, of the 7 peptides synthesized using Fmoc-Asn,3 had >10% of a side reaction directly attributable to Asn incorporation. The other problem detected in Fmoc syntheses was the generation of ethanedithiol-thioanisyl or anisyltrifluoroacetyl adducts during peptide-resin cleavage.
Impurities in Boc-synthesized peptides were mostly due to cleavage problems. Of the 12 crude samples synthesized with Trp(For), 4 had >10% of the For group still attached. Variable success was seen when deprotection of Trp(For) was attempted during HF cleavage in the presence of anisole alone or anisole plus other scavengers. Deprotection of Trp(For) was complete following treatment of the peptide-resin with ~10% piperidine-DMF for 2 h prior to HF cleavage.
Cleavage conditions were more stringently reevaluated in the 1993 study using the test sequence Lys-His-Asp-Pro-Cys-Gly-TrpAsn-Gly-Pro-Arg-Pro-Met-Arg-Gly, which is susceptible to acidolysis of the Asp-Pro bond and incomplete Dnp removal from His in addition to the cleavage side-reactions documented in the 1991 and 1992 studies. The peptide was synthesized by the Committee in both Boc and Fmoc versions, with participating laboratories cleaving the peptide-resin and returning the crude product. A total of 46 crude products were submitted. RP-HPLC showed I out of the 12 Boc-based syntheses and 26 out of the 34 Fmoc-based syntheses to contain >25% of the desired product, which was consistent with the more qualitative findings of ESMS and LDMS. Most non-desired products were the result of incomplete deprotections and/or other covalent modifications, in similar fashion to the 1991 study. These results suggest that peptide-resin cleavages following Boc-based syntheses may require greater attention to detail to avoid deleterious side reactions.
The wide range of peptide quality allowed us to evaluate critically the strengths and weaknesses of the analytical techniques used here. RP-HPLC appeared to provide an accurate estimate of the complexity of the peptide samples and, in situations in which a sample of the desired product is available, an accurate estimate of the content of the desired product. The only exception was the failure to detect an Asn deletion peptide in the 1992 study which co-eluted with the desired product. Since a sample of the desired product is not usually available, additional analyses are required. AAA was the best technique for absolute amino acid quantitation, but was not helpful for detecting modifications of amino acid residues. Preview sequence analysis had the same strengths and weaknesses as AAA and the advantage of determining residue position. The mass spectrometric techniques examined here (ESMS, FABMS, PDMS, and LDMS) allowed for identification of peptide modifications, including residual protecting groups. Our experience in these studies suggests that peptide synthesis is susceptible to a variety of side-reactions, and that efficient characterization of synthetic peptides is best obtained by a combination of AAA, RP-HPLC, and MS, with sequencing by either Edman degradation or tandem MS being used to identify the positions of modifications and deletions.
1. Young et al, Peptides: Chemistry, Structure and Biology (1993) (Proceedings of the Thirteenth American Peptide Symposium, June 20-25,1993, Edmonton, Alberta, Canada) (R.S. Hodges and J.A. Smith, Eds.) Escom, Leiden, The Netherlands, in press.
2. Smith, A.J., Young, J.D., Carr, S.A., Marshak, D.R., Williams, L.C. and Williams, K.R., (1992) Techniques in Protein Chemistry 111 (Angeletti, R.H., Ed.), Academic Press, Orlando, FL, U.S.A., 219.
3. Fields, G.B., Carr, S.A., Marshak, D.R., Smith, A.J., Stults, J.T., Williams, L.C., Williams, K.R. and Young, J.D., (1993) Techniques in Protein Chemistry V (Angeletti, R.H., Ed.), Academic Press, Orlando, FL, U.S.A., 227.
4. Fields, G.B., Angeletti, R.H., Carr, S.A., Smith, A.J., Stults, J.T., Williams, L.C. and Young, J.D., (1993) Techniquesin Protein Chemistry V (Crabb, J., Ed.), Academic Press, Orlando, FL, U.S.A., in press.
Return to the The ABRF Home Page