No abstract

N1 is the first residue in an ␣-helix. We have measured the contribution of all 20 amino acids to the stability of a small helical peptide CH 3 CO-XAAAAQAAAAQAAGY-NH 2 at the N1 position. By substituting every residue into the N1 position, we were able to investigate the stabilizing role of each amino acid in an isolated context. The helix content of each of the 20 peptides was measured by circular dichroism (CD) spectroscopy. The data were analyzed by our modified Lifson-Roig helix-coil theory, which includes the n1 parameter, to find free energies for placing a residue into the N1 position. The rank order for free energies is Asp − , Ala > Glu − > Glu 0 > Trp, Leu, Ser > Asp 0 , Thr, Gln, Met, Ile > Val, Pro > Lys + , Arg, His 0 > Cys, Gly > Phe > Asn, Tyr, His + . N1 preferences are clearly distinct from preferences for the preceding N-cap and ␣-helix interior. pK a values were measured for Asp, Glu, and His, and protonation-free energies were calculated for Asp and Glu. The dissociation of the Asp proton is less favorable than that of Glu, and this reflects its involvement in a stronger stabilizing interaction at the N terminus. Proline is not energetically favored at the ␣-helix N terminus despite having a high propensity for this position in crystal structures. The data presented are of value both in rationalizing mutations at N1 ␣-helix sites in proteins and in predicting the helix contents of peptides.

N2 is the second position in the ␣-helix. All 20 amino acids were placed in the N2 position of a synthetic helical peptide (CH 3 CO-[AXAAAAKAAAAKAAGY]-NH 2 ) and the helix content was measured by circular dichroism spectroscopy at 273K. The dependence of peptide helicity on N2 residue identity has been used to determine a free-energy scale by analysis with a modified Lifson-Roig helix-coil theory that includes a parameter for the N2 energy (n2). The rank order of ⌬⌬G (relative to Ala) is Glu − , Asp − > Ala > Glu 0 , Leu, Val, Gln, Thr, Ile, Ser, Met, Asp 0 , His 0 , Arg, Cys, Lys, Phe > Asn, > Gly, His + , Pro, Tyr. The results correlate very well with N2 propensities in proteins, moderately well with N1 and helix interior preferences, and not at all with N-cap preferences. The strongest energetic effects result from interactions with the helix dipole, which favors negative charges at the helix N terminus. Hydrogen bonds to side chains at N2, such as Gln, Ser, and Thr, are weak, despite occurring frequently in protein crystal structures, in contrast to the N-cap position. This is because N-cap hydrogen bonds are close to linear, whereas N2 hydrogen bonds have poor geometry. These results can be used to modify protein stability rationally, help design helices, and improve prediction of helix location and stability.

Chronic lymphocytic leukemia (CLL) is a common hematopoietic malignant disease with variable outcome. CLL has been divided into distinct groups based on whether somatic hypermutation has occurred in the variable region of the immunoglobulin heavy-chain locus or alternatively if the cells express higher levels of the CD38 protein.We have analyzed the proteome of 12 cases of CLL (six mutated (M-CLL) and six unmutated (UM-CLL) immunoglobulin heavy-chain loci; seven CD38-negative and five CD38-positive) using two-dimensional electrophoresis and mass spectrometry. Statistical evaluation using principal component analysis indicated significant differences in patterns of protein expression between the cases with and without somatic mutation. Specific proteins indicated by principal component analysis as varying between the prognostic groups were characterized using mass spectrometry. The levels of F-actin-capping protein ␤ subunit, 14-3-3 ␤ protein, and laminin-binding protein precursor were significantly increased in M-CLL relative to UM-CLL. In addition, primary sequence data from tandem mass spectrometry showed that nucleophosmin was present as several protein spots in M-CLL but was not detected in UM-CLL samples, suggesting that several post-translationally modified forms of nucleophosmin vary between these two sample groups. No specific differences were found between CD38-positive and -negative patient samples using the same approach. The results presented show that proteomic analysis can complement other approaches in identifying proteins that may have potential value in the biological and diagnostic distinction between important clinical subtypes of CLL.