A protein molecule can be regarded as a precision mechanical instrument in itself. Thousands of such instruments with di โ erent structures are functioning in a very systematic manner in a living cell. At the present time, we are excited almost every week by a newly determined molecular structure of one of such "instruments,ร and by its inferred functioning mechanism. The structures come mostly from x-ray crystallographic studies, and some from multi-dimensional NMR studies. So, what is the role of Raman spectroscopy in this รeld ? A good part of the answer is found in this special issue, made up of 10 topical papers.
Brieรy, Raman spectroscopy can be advantageous for a protein study, (i) because it is applicable not only to crystals and aqueous solutions, but also to virtually any sample morphology, (ii) because by use of an optical microscope the sample size can be only 1/100 of that for x-ray crystallography, (iii) because it can be used to highlight a special portion of the molecule by changing the exciting wavelength (1064ร200 nm) and (iv) because it is applicable to unstable intermediate species. In the รrst paper of this issue, P. Carey reviews in more detail such advantages of Raman spectroscopy for enzymology, and illustrates them using his own experimental data. One of his subjects is a serine protease, whose substrate links covalently with the enzyme to form an ester as an intermediate. He shows that Raman spectroscopy can provide unique information about such an intermediate, namely, some bond lengths in the active site, e โ ective hydrogen bond strengths there and the electric รeld causing polarization of substrate electrons.
R. Callender et al. report, in the second paper, their studies which also deal with protein complexes with small molecules. Their special emphasis is on Raman di โ erence spectroscopy. They have constructed a special instrument with very high "subtraction รdelityร (0.1%). They assert that "isotope editingร procedures are very powerful and general. Here, an atom within a bond of interest is labelled with a stable isotope (2H, 13C, 15N 18O, etc.). Subtraction of labelled and unlabelled protein spectra yields an "isotopically editedร di โ erence spectrum in which only vibrational modes involving the labelled atom show up.
Overman and Thomas also perform such an "isotope editingร directly for a protein molecule involved in an intact biological supramolecular assembly (such as a virus), through a microbiological method. On the basis of their di โ erence spectra of labelled and unlabelled fd (bactriophage), we can identify unambiguously many of the tryptophan, tyrosine, phenylalanine, alanine and main-chain bands in the complicated Raman spectrum of this virus. Thus, we do not need to refer any more to the Raman spectra of small "modelร compounds ; indeed, our thinking process is sometimes even the reverse of the traditional process. Overman and Thomas also present Raman spectra of site-speciรc mutants of the virus. These can be a valuable source of insight into a particular aspect of the protein structure. The interpretation, however, is not always straightforward, because a mutation may CCC 0377ร0486/98/010003ร03 $17.50