Secondary structures of peptides and proteins via NMR chemical-shielding anisotropy (CSA) Parameters

Eszter Czinki, A. Császár, G. Magyarfalvi, Peter R. Schreiner, Wesley D. Allen

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

Complete nuclear magnetic resonance (NMR) chemical-shielding tensors, σ, have been computed at different levels of density-functional theory (DFT), within the gauge-including atomic orbital (GIAO) formalism, for the atoms of the peptide model For-L-Ala-NH2 as a function of the backbone dihedral angles φ and ψ by employing a dense grid of 10°. A complete set of rigorously orthogonal symmetric tensor invariants, {σ iso, ρ, τ}, is introduced, where σiso is the usual isotropic chemical shielding, while the newly introduced ρ and τ parameters describe the magnitude and the orientation/shape of the chemical-shielding anisotropy (CSA), respectively. The set {σ iso, ρ, τ} is unaffected by unitary transformations of the symmetric part of the shielding tensor. The mathematically and physically motivated {ρ, τ} anisotropy pair is easily connected to more traditional shielding anisotropy measures, like span (Ω) and skew (κ). The effectiveness of the different partitions of the CSA information in predicting conformations of peptides and proteins has been tested throughout the Ramachandran space by generating theoretical NMR anisotropy surfaces for our For-L-Ala-NH2 model. The CSA surfaces, including Ω(φ, ψ), κ(φ, ψ), ρ(φ, ψ), and τ(φ, ψ) are highly structured. Individually, none of these surfaces is able to distinguish unequivocally between the α-helix and β-strand secondary structural types of proteins. However, two- and three-dimensional correlated plots, including Ω versus κ, ρ versus τ, and σiso versus ρ versus τ, especially for 13Cα, have considerable promise in distinguishing among all four of the major secondary structural elements.

Original languageEnglish
Pages (from-to)1568-1577
Number of pages10
JournalJournal of the American Chemical Society
Volume129
Issue number6
DOIs
Publication statusPublished - Feb 14 2007

Fingerprint

Secondary Protein Structure
Anisotropy
Shielding
Peptides
Magnetic Resonance Spectroscopy
Nuclear magnetic resonance
Proteins
Tensors
Protein Conformation
Dihedral angle
Gages
Density functional theory
Conformations
Atoms

ASJC Scopus subject areas

  • Chemistry(all)

Cite this

Secondary structures of peptides and proteins via NMR chemical-shielding anisotropy (CSA) Parameters. / Czinki, Eszter; Császár, A.; Magyarfalvi, G.; Schreiner, Peter R.; Allen, Wesley D.

In: Journal of the American Chemical Society, Vol. 129, No. 6, 14.02.2007, p. 1568-1577.

Research output: Contribution to journalArticle

@article{6c6b3bdc9df743c389197736be02b329,
title = "Secondary structures of peptides and proteins via NMR chemical-shielding anisotropy (CSA) Parameters",
abstract = "Complete nuclear magnetic resonance (NMR) chemical-shielding tensors, σ, have been computed at different levels of density-functional theory (DFT), within the gauge-including atomic orbital (GIAO) formalism, for the atoms of the peptide model For-L-Ala-NH2 as a function of the backbone dihedral angles φ and ψ by employing a dense grid of 10°. A complete set of rigorously orthogonal symmetric tensor invariants, {σ iso, ρ, τ}, is introduced, where σiso is the usual isotropic chemical shielding, while the newly introduced ρ and τ parameters describe the magnitude and the orientation/shape of the chemical-shielding anisotropy (CSA), respectively. The set {σ iso, ρ, τ} is unaffected by unitary transformations of the symmetric part of the shielding tensor. The mathematically and physically motivated {ρ, τ} anisotropy pair is easily connected to more traditional shielding anisotropy measures, like span (Ω) and skew (κ). The effectiveness of the different partitions of the CSA information in predicting conformations of peptides and proteins has been tested throughout the Ramachandran space by generating theoretical NMR anisotropy surfaces for our For-L-Ala-NH2 model. The CSA surfaces, including Ω(φ, ψ), κ(φ, ψ), ρ(φ, ψ), and τ(φ, ψ) are highly structured. Individually, none of these surfaces is able to distinguish unequivocally between the α-helix and β-strand secondary structural types of proteins. However, two- and three-dimensional correlated plots, including Ω versus κ, ρ versus τ, and σiso versus ρ versus τ, especially for 13Cα, have considerable promise in distinguishing among all four of the major secondary structural elements.",
author = "Eszter Czinki and A. Cs{\'a}sz{\'a}r and G. Magyarfalvi and Schreiner, {Peter R.} and Allen, {Wesley D.}",
year = "2007",
month = "2",
day = "14",
doi = "10.1021/ja065461k",
language = "English",
volume = "129",
pages = "1568--1577",
journal = "Journal of the American Chemical Society",
issn = "0002-7863",
publisher = "American Chemical Society",
number = "6",

}

TY - JOUR

T1 - Secondary structures of peptides and proteins via NMR chemical-shielding anisotropy (CSA) Parameters

AU - Czinki, Eszter

AU - Császár, A.

AU - Magyarfalvi, G.

AU - Schreiner, Peter R.

AU - Allen, Wesley D.

PY - 2007/2/14

Y1 - 2007/2/14

N2 - Complete nuclear magnetic resonance (NMR) chemical-shielding tensors, σ, have been computed at different levels of density-functional theory (DFT), within the gauge-including atomic orbital (GIAO) formalism, for the atoms of the peptide model For-L-Ala-NH2 as a function of the backbone dihedral angles φ and ψ by employing a dense grid of 10°. A complete set of rigorously orthogonal symmetric tensor invariants, {σ iso, ρ, τ}, is introduced, where σiso is the usual isotropic chemical shielding, while the newly introduced ρ and τ parameters describe the magnitude and the orientation/shape of the chemical-shielding anisotropy (CSA), respectively. The set {σ iso, ρ, τ} is unaffected by unitary transformations of the symmetric part of the shielding tensor. The mathematically and physically motivated {ρ, τ} anisotropy pair is easily connected to more traditional shielding anisotropy measures, like span (Ω) and skew (κ). The effectiveness of the different partitions of the CSA information in predicting conformations of peptides and proteins has been tested throughout the Ramachandran space by generating theoretical NMR anisotropy surfaces for our For-L-Ala-NH2 model. The CSA surfaces, including Ω(φ, ψ), κ(φ, ψ), ρ(φ, ψ), and τ(φ, ψ) are highly structured. Individually, none of these surfaces is able to distinguish unequivocally between the α-helix and β-strand secondary structural types of proteins. However, two- and three-dimensional correlated plots, including Ω versus κ, ρ versus τ, and σiso versus ρ versus τ, especially for 13Cα, have considerable promise in distinguishing among all four of the major secondary structural elements.

AB - Complete nuclear magnetic resonance (NMR) chemical-shielding tensors, σ, have been computed at different levels of density-functional theory (DFT), within the gauge-including atomic orbital (GIAO) formalism, for the atoms of the peptide model For-L-Ala-NH2 as a function of the backbone dihedral angles φ and ψ by employing a dense grid of 10°. A complete set of rigorously orthogonal symmetric tensor invariants, {σ iso, ρ, τ}, is introduced, where σiso is the usual isotropic chemical shielding, while the newly introduced ρ and τ parameters describe the magnitude and the orientation/shape of the chemical-shielding anisotropy (CSA), respectively. The set {σ iso, ρ, τ} is unaffected by unitary transformations of the symmetric part of the shielding tensor. The mathematically and physically motivated {ρ, τ} anisotropy pair is easily connected to more traditional shielding anisotropy measures, like span (Ω) and skew (κ). The effectiveness of the different partitions of the CSA information in predicting conformations of peptides and proteins has been tested throughout the Ramachandran space by generating theoretical NMR anisotropy surfaces for our For-L-Ala-NH2 model. The CSA surfaces, including Ω(φ, ψ), κ(φ, ψ), ρ(φ, ψ), and τ(φ, ψ) are highly structured. Individually, none of these surfaces is able to distinguish unequivocally between the α-helix and β-strand secondary structural types of proteins. However, two- and three-dimensional correlated plots, including Ω versus κ, ρ versus τ, and σiso versus ρ versus τ, especially for 13Cα, have considerable promise in distinguishing among all four of the major secondary structural elements.

UR - http://www.scopus.com/inward/record.url?scp=33846999342&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33846999342&partnerID=8YFLogxK

U2 - 10.1021/ja065461k

DO - 10.1021/ja065461k

M3 - Article

C2 - 17284001

AN - SCOPUS:33846999342

VL - 129

SP - 1568

EP - 1577

JO - Journal of the American Chemical Society

JF - Journal of the American Chemical Society

SN - 0002-7863

IS - 6

ER -