Optimization of Model Parameters for Describing the Amide I Spectrum of a Large Set of Proteins

Abstract

A new simulation protocol for the prediction of the infrared absorption of the amide I vibration of proteins was developed. The method incorporates known effects on the intrinsic frequencies (backbone conformation, inter-peptide and peptide-solvent hydrogen bonding) and couplings (nearest neighbor coupling, transition dipole coupling) of amide I oscillators in a parametrized manner. Model parameters for the simulation of amide I spectra were determined through fitting and optimization of simulated spectra to experimentally measured infrared spectra of 44 proteins that represent maximum structural variation in terms of different folds and secondary structure contents. Prediction of protein spectra using the optimized parameters resulted in good agreement with experimental spectra and in a considerable improvement compared to a description involving only transition dipole coupling.

Application of two-dimensional infrared spectroscopy to benchmark models for the amide I band of proteins

“…64,66,67,[70][71][72][73] The electrostatic frequency mappings have been tested against ab initio calculations, 50,51,69,74 solvent experiments of single amide units, 52,60,73,75 gas phase experiments 76,77 of small peptides, and linear absorption of proteins. 62,64 So far, only one mapping for the amide unit preceding proline has been presented. 59 This amide unit is special as it is a tertiary amide group.…”

“…The couplings have been tested on the linear spectra of proteins, 62,64,84 using isotope edited 2DIR, 85,86 ab initio calculations, 50,51,74 and gas phase experiments. 76,77 In general, benchmarking the coupling models is the most challenging as the observables depending on the couplings inherently also depend on the vibrational frequencies.…”

Electrostatic frequency shifts in amide I vibrational spectra: Direct parameterization against experiment

“…11,13,18,22,26 Site energy maps are based on the observation that electrostatic interactions (particularly hydrogen bonding) between a solvated molecule and its environment are the primary predictors of local vibrational frequencies. 29 Early models for amide I frequencies employed simple geometric criteria such as hydrogen bond count and length; 9,[30][31][32][33] although these models have proven intuitively useful and reasonably accurate for simple calculations, 34 they are generally regarded as insufficient for detailed spectral simulations. 35 Most current vibrational frequency models are based on a linear mapping of frequency against various combinations of electrostatic variables evaluated at different positions within the peptide group…”

Vibrational solvatochromism: Towards systematic approach to modeling solvation phenomena

“…Another approach designed directly for dealing with very large systems like proteins relies on the construction of a set of parameters using experimentally measured protein spectra as reference data. 38 Although these empirical approaches were shown to be capable of reproducing the frequency shifts induced by solutesolvent interactions, the fitting procedures can be biased towards a chosen subset of environments. That could limit their a) Author to whom correspondence should be addressed.…”