A 1 Main Characteristics of Infrared Spectra of Proteins

The backbone of a polypeptide chain absorbs infrared radiation, which excites vibrational modes of the peptide bonds. Infrared spectroscopy measures the amount of light absorbed by these vibrations over a range of frequencies of the incident light. The positions and intensities of the infrared absorbance bands of these vibrational modes are sensitive to the protein's secondary structure. Band frequencies are characteristic of specific structural units, and their areas are proportional to the amount of the structural unit present in the overall structure.

Two vibrational modes are of prime importance for secondary structural analysis. The amide I mode is caused mainly by stretching of the carbonyl or CO bond [5], Amide I vibrations appear in the general region around wave number 1660 cm1. The amide II mode is a combination of NH bend and CN stretch and appears in the region around 1540 cm"1.

The positions of the amide I and amide II bands in the infrared spectrum are influenced by the exact structural unit in the polypeptide within which the bond resides. This leads to a number of severely overlapped bands within the amide I and amide II regions [5], The individual bands in protein spectra can be modeled with the Gaussian peak shape, illustrated with its first and second derivatives in Figure 7.1. Torri and Tasumi [6-8] have calculated a theoretical FTIR amide I spectrum using the three-dimensional structure of lysozyme from X-ray crystallography and a Gaussian envelope of each peptide oscillator with a peak half-width at half-height of 3.0 cm-1. The amide I and amide II bands of lysozyme after Fourier deconvolution are illustrated in Figure 7.2.

Kauppinen and coworkers [2] and Susi and Byler [3] considered the amide I band as a sum of severely overlapped Gaussian peaks, or Gaussian peaks with a fraction of Lorentzian character (see Section 3.B.2). Susi and Byler adapted a methodology using second derivative spectra, Fourier deconvolution algorithms (Section 4.A.4), and nonlinear regression analysis

Figure 7.1 Shapes of a Gaussian peak and its first and second derivatives. The second derivative peak is plotted at twice the sensitivity of the other peaks.

for deconvoluting the amide I envelope into its individual component bands. Using the theoretical assignments for these bands [5, 9, 10], the fractions of various structural units such as a-helices, turns, extended structures, and so forth, in a protein's secondary structure can be estimated from the

Figure 7.2 Fourier deconvolution of FT-IR spectrum of lysozyme in Figure 7.4. Overlapping lines on outer envelope show agreements of experimental and computed results. Individual component peaks underneath are the results of the regression analysis. Inset shows second derivative of original spectrum. (Reprinted with permission from [10], copyright by the American Chemical Society.)

Figure 7.2 Fourier deconvolution of FT-IR spectrum of lysozyme in Figure 7.4. Overlapping lines on outer envelope show agreements of experimental and computed results. Individual component peaks underneath are the results of the regression analysis. Inset shows second derivative of original spectrum. (Reprinted with permission from [10], copyright by the American Chemical Society.)

fractional areas under the assigned component bands. Initially, the amide I band was considered only because experiments were done in D20, in which the amide II band does not appear. More recently, the amide I and amide II bands for spectra obtained in water and have been analyzed together [5, 9, 10]. As shown later, these analyses must be done in a way that is consistent with theoretical models [5-8] and statistical considerations.

0 0

Post a comment