Microsequence analysis of peptides and proteins: trimethylsilylisothiocyanate as a reagent for COOH-terminal sequence analysis

A reinvestigation of the isothiocyanate-based chemistry for cyclic degradations of peptides and proteins revealed that the reagent trimethylsilylisothiocyanate (TMS-ITC) gives superior results in terms of coupling efficiency and lack of complicating side reactions. Acetic anhydride (10 min at various temperatures) was used to activate the carboxyl terminus, and 6 N HCl (30 min at room temperature) was used for cleavage as originally described by G.R. Stark (Biochemistry 8, 4735, 1968). Reaction conditions for efficient coupling were explored using subtractive chemistry on bradykinin, a nonapeptide, and separation of the reaction products by reverse-phase high-performance liquid chromatography. The products were analyzed by fast atom bombardment-mass spectrometry and shown to be the N-acetylated starting material and the N-acetylated des-Arg9 derivative of bradykinin. The pseudo-first-order rate constants measured at 50, 70, and 90 degrees C were 5.6 X 10(-5), 5.1 X 10(-4), and 8.6 X 10(-4) s-1, respectively. In order to obtain complete couplings within 30-40 min at 50 degrees C, the effect of pyridine catalysis was studied. The addition of 0.225 M pyridine resulted in roughly doubling the rates at 50 and 70 degrees C. In the case of bradykinin, the reaction with TMS-ITC in the presence of the pyridine catalyst at 50 degrees C was complete in 15 min. In order to apply this methodology to the analysis of proteins, the thiohydantoin derivatives of amino acids were synthesized and separated by reverse-phase HPLC. The derivatives were also characterized by mass spectrometry. The above reaction conditions were tested on 3 nmol of sperm whale apomyoglobin for three cycles of degradation. The sample was first coupled to p-phenylene diisothiocyanate-derivatized aminopropyl glass with a 90% yield. The approximate initial yield of glycine at cycle one was 30%. The first three cycles corresponded exactly to the predicted carboxy-terminal sequence of myoglobin. These results demonstrate the feasibility of a new Stark reagent for automated carboxy-terminal chemistry.