Design of helices that are stable in vacuo
Design of helices that are stable in vacuo
复制标题
DOI:
10.1021/ja983021q
复制
发表时间:
1998-12-16
影响因子:
15
通讯作者:
Jarrold, MF
中科院分区:
文献类型:
--
作者:
Hudgins, RR;Ratner, MA;Jarrold, MF
The R-helix is the most common short-range structural motif in proteins, 1 and understanding its properties is central to understanding protein folding2 and enzyme function. 3 Different amino acids have different helix propensities. 2, 4, 5 Alanine has one of the highest helix propensities, and alanine-rich peptides have been widely studied in solution. However, the factors responsible for the different helix propensities certainly include the effects of the solvent. 6-8 In principle, intrinsic helix propensities can be obtained from gas-phase measurements where solvent interactions are absent. Here we report studies of alanine-based peptides in a solvent-free environment. Protonated polyalanines with up to 20 residues do not form extended helices in a vacuum. However, the addition of a single lysine at the C terminus (to give Ac-Alan-LysH+) results in the formation of very stable, monomeric, polyalanine helices. The design of the Ac-Alan-LysH+ peptide optimizes hydrogen bonding of the charged lysine with the C-terminal backbone carbonyl groups, and the interaction of the charge with the helix dipole. Helices are observed for Ac-Alan-LysH+ peptides with as few as eight residues, which is significantly shorter than the shortest helical peptides found in solution. A variety of experimental techniques have recently been used to examine the gas-phase conformations of proteins and peptides. 9-17 In the work reported here we have used high-resolution ion mobility measurements. 18, 19 The mobility is a measure of how rapidly an ion moves through an inert buffer gas under the influence of a weak electric field. The mobility depends on the ion’s collision cross section with the buffer gas. Structural information is deduced by comparing measured cross sections to orientationally averaged cross sections calculated for conformations derived from molecular dynamics (MD) simulations. 12 Our experimental apparatus consists of an electrospray source, a 63-cm long drift tube containing helium buffer gas, and a quadrupole mass spectrometer. Details of the experimental technique are described in a recent publication. 20 MD simulations were performed with the CHARMM force field21 using the 21.3 parameter set, the united atom approximation, and SHAKE22 constraints on all explicit bonds. Multiple simulations of 0.25-1.0 ns were performed at 300 K for each peptide. Cross sections were calculated by a trajectory method23 and averaged over 50 structures taken from an MD simulation at regular time intervals. Figure 1a shows a mass spectrum obtained by electrospraying a solution of unpurified Ac-Ala19-Lys in formic acid. A progression of peaks at mass to charge (m/z) ratios corresponding to Ac-Alan-LysH+, n) 14-19, is apparent. The distribution of peptide sizes results from inefficient coupling during solid-phase Fmoc synthesis (Anaspec Inc., San Jose, CA). Figure 1b shows a drift time distribution measured with the mass spectrometer set to transmit Ac-Ala19-LysH+. There are two main peaks; the peak at a drift time of∼ 170 ms is assigned to the Ac-Ala19-LysH+ monomer while the one at∼ 155 ms is assigned to the (Ac-Ala19-LysH) 2 2+ dimer, which occurs at the same m/z. 24 The measured drift times are converted into average collision cross sections. 25 In this work we will use a relative cross section scale given by Ωav-14.50n where Ωav (the measured cross section) is in Å2 and 14.50 Å2 is the calculated average cross section per residue for an ideal polyalanine R-helix with the torsion angles fixed at φ)-57 and ψ)-47. With this scale, helical conformations have relative cross sections that are independent of the number of alanine residues, whereas other …