167
different turn geometries to Cd bands often lacks solid empirical verification. the
problem has been briefly addressed in earlier works (e.g. Ref. 7) and is investigated
by means of quantum chemical calculations [58].
Because of the near additivity of these signals, a far-uv Cd spectrum of a protein with a known structure can be relatively easily predicted (as discussed above),
given the typical structural similarity of crystal and water-dissolved proteins. From
a practical standpoint, a much more interesting and important is the opposite goal:
estimation of the content of various secondary structure elements based on an experimentally collected far-uv Cd spectrum. Achieving it permits using Cd spectroscopy as a very fast and quantitative means of analyzing protein secondary structure in solution. the study by greenfield and Fasman [59] is one of the earliest
and most straightforward approaches to this problem. the authors used far-uv Cd
spectra of poly-L-lysine—a homopolypeptide capable of adopting (depending on
pH and temperature) either α-helical, antiparallel β-sheet, or random coil (polyproline II) conformation—as a reference set for semi-quantitative deconvolution of
several globular proteins. Since that seminal work it has become clear that a more
accurate far-uv Cd- based estimation of ingredient secondary components would
require both more advanced computational algorithms and more expanded reference set of Cd spectra of proteins with well-defined (through X-ray diffraction or
solution NmR) 3d-structure. Several existing tools (including internet-accessible
online servers [60]) for quantitative analysis of the far-uv Cd spectra of proteins
have been described and reviewed earlier [61–63]. As was mentioned in the previous paragraphs, expanding the Cd spectral range to include vacuum-uv-detectable
transitions markedly improves accuracy of the method, however, this also requires
having a protein Cd reference set properly expanded in the range of low wavelengths [53, 64–66].
B. CD signals from aromatic rings In the absence of electronic transitions stemming from prosthetic groups (such as heme) Cd signals of proteins in the near-uv
region (which could be arbitrarily defined as 250–300 nm) originate from disulfide bridges and aromatic side chains (table 6.1). typically, these are weaker than
far-uv Cd by at least an order of magnitude. diagnostic value of broad and featureless disulfide bands is negligible. on the other hand, Cd of the aromatic bands
did find specific application in biophysics. the coupling of achiral chromophores
of phenylalanine, tyrosine, or tryptophan planar side-chains to nearest chiral carbon atom is weak and the corresponding low rotational strength in a fluctuating
disordered protein chain would be sufficiently low to make the Cd signal almost
undetectable. the situation changes when upon ’hydrophobic effect’-driven folding
the non-polar amino acid side chains (and the aromatic residues among them) are
clustered and buried in a protein hydrophobic core. transition dipole moments of
the individual aromatic chromophores interact strongly, forming a chiral quasi-united chromophore with a rather strong Cd signal [7, 67]. hence, the tertiary structure
of a stable unambiguously defined protein envelope surrounding the aromatic residues is a condition sine qua non of the occurrence of measurable near-uv Cd. Such
Cd signal would be only yet-another (next to the far-uv Cd) chiroptical aspect
6 Electronic Circular dichroism Spectroscopy in Structural Analysis …
different turn geometries to Cd bands often lacks solid empirical verification. the
problem has been briefly addressed in earlier works (e.g. Ref. 7) and is investigated
by means of quantum chemical calculations [58].
Because of the near additivity of these signals, a far-uv Cd spectrum of a protein with a known structure can be relatively easily predicted (as discussed above),
given the typical structural similarity of crystal and water-dissolved proteins. From
a practical standpoint, a much more interesting and important is the opposite goal:
estimation of the content of various secondary structure elements based on an experimentally collected far-uv Cd spectrum. Achieving it permits using Cd spectroscopy as a very fast and quantitative means of analyzing protein secondary structure in solution. the study by greenfield and Fasman [59] is one of the earliest
and most straightforward approaches to this problem. the authors used far-uv Cd
spectra of poly-L-lysine—a homopolypeptide capable of adopting (depending on
pH and temperature) either α-helical, antiparallel β-sheet, or random coil (polyproline II) conformation—as a reference set for semi-quantitative deconvolution of
several globular proteins. Since that seminal work it has become clear that a more
accurate far-uv Cd- based estimation of ingredient secondary components would
require both more advanced computational algorithms and more expanded reference set of Cd spectra of proteins with well-defined (through X-ray diffraction or
solution NmR) 3d-structure. Several existing tools (including internet-accessible
online servers [60]) for quantitative analysis of the far-uv Cd spectra of proteins
have been described and reviewed earlier [61–63]. As was mentioned in the previous paragraphs, expanding the Cd spectral range to include vacuum-uv-detectable
transitions markedly improves accuracy of the method, however, this also requires
having a protein Cd reference set properly expanded in the range of low wavelengths [53, 64–66].
B. CD signals from aromatic rings In the absence of electronic transitions stemming from prosthetic groups (such as heme) Cd signals of proteins in the near-uv
region (which could be arbitrarily defined as 250–300 nm) originate from disulfide bridges and aromatic side chains (table 6.1). typically, these are weaker than
far-uv Cd by at least an order of magnitude. diagnostic value of broad and featureless disulfide bands is negligible. on the other hand, Cd of the aromatic bands
did find specific application in biophysics. the coupling of achiral chromophores
of phenylalanine, tyrosine, or tryptophan planar side-chains to nearest chiral carbon atom is weak and the corresponding low rotational strength in a fluctuating
disordered protein chain would be sufficiently low to make the Cd signal almost
undetectable. the situation changes when upon ’hydrophobic effect’-driven folding
the non-polar amino acid side chains (and the aromatic residues among them) are
clustered and buried in a protein hydrophobic core. transition dipole moments of
the individual aromatic chromophores interact strongly, forming a chiral quasi-united chromophore with a rather strong Cd signal [7, 67]. hence, the tertiary structure
of a stable unambiguously defined protein envelope surrounding the aromatic residues is a condition sine qua non of the occurrence of measurable near-uv Cd. Such
Cd signal would be only yet-another (next to the far-uv Cd) chiroptical aspect
6 Electronic Circular dichroism Spectroscopy in Structural Analysis …