CIDNP method

Figure B 1.16.9 shows background-free, pseudo-steady-state CIDNP spectra of the photoreaction of triethylamine with (a) anthroquinone as sensitizer and (b) and (c) xanthone as sensitizer. Details of the pseudo-steady-state CIDNP method are given elsewhere [22]. In trace (a), no signals from the p protons of products 1 (recombination) or 2 (escape) are observed, indicating that the products observed result from the radical ion pair. Traces (b) and (c) illustrate a usefiil feature of pulsed CIDNP net and multiplet effects may be separated on the basis of their radiofrequency (RF) pulse tip angle dependence [21]. Net effects are shown in trace (b) while multiplet effects can...

In an extension of traditional CIDNP methods, Closs and co-workers developed time-resolved CIDNP (TR-CIDNP) m the late 1970s [24, 25 and 26]. The initial time-resolved experiments had a time resolution in the... 

The first application of the time-resolved CIDNP method by Closs and co-workers involved tire Norrish 1 cleavage of benzyl phenyl ketone [24, 25]. Geminate RPs may recombine to regenerate the starting material while escaped RPs may fonn the starting ketone (12), bibenzyl (3), or benzil (4), as shown below. 

CIDNP proved to be a very promising new technique in studjdng the reactions given by tnplet and singlet species. Whereas all other spectroscopic techniques merely provide information on the carbenes formed, CIDNP is the only technique for the investigation of the reaction mechanism of carbenes. Since the CIDNP method is relatively new the information collected so far is scanty. A simplified description of CIDNP is given below ... 

As with any other physical methods, the CIDNP method is not universal and not immune to misinterpretation. It has certain drawbacks The polarization is weak and hardly detected in reactions involving extremely short-lived radicals and, if so, the polarization disappears quickly. It is often difficult to attribute the polarization to products of the main conversion, rather than the side or reverse conversions. The latter threat is most serious for the reactions with participation of ion-radicals—the formation of end products often proceeds concurrently with the restoration of the initial neutral molecules, due to a reverse electron transfer as in Scheme 4.29. 

As discussed earlier, CIDNP method is of great help to chemists. However, it cannot always give straightforward information, because the CIDNP effect may be masked and errors may creep into its interpretation. CIDNP requires strong chemical and physical professional skills (which are useful for all the method considered here). However, using CIDNP, a researcher can be compensated by the reliability of the conclusions. 

The CIDNP method is an indirect method since the hyperfine pattern of a paramagnetic intermediate is derived from the unusual NMR intensities of a diamagnetic product derived from it. This method also has limitations and potential sources of misassignments. Similar to the EPR technique, the CIDNP method only documents the unpaired spin additional evidence for the nature of the paramagnetic intermediate, particularly for the presence of the charge, is derived typically on the basis of mechanistic considerations and from supporting secondary experiments. 

This conclusion raises the question whether the radical cation of 107 may be localized on a single cyclopropane entity rather than involving both strained rings. The CIDNP method does not lend itself to decide this issue, because the two cyclopropane groups are magnetically equivalent in the substrate/product. The ESR method, on the other hand, is well suited to elucidate the precise nature of... 

The applications of CIDNP to mechanistic studies of organic photochemical reactions are numerous, but only a few systems, such as the photoreduction of quinones, have been fully examined by both CIDEP and CIDNP methods. Instead of repeating some of the well-known CIDNP mechanistic studies summarized in other reviews, we shall go into a relatively new area of CIDNP studies involving metallorganic compounds. 

Of special note is the observation of proton nuclear spin polarization in certain exterior amino acids of protein molecules using a newly developed laser photo-CIDNP method (81). The method discriminates between exterior and interior amino acid chains. 

However, the instability of the simplest a-tin ketones precludes the application of CIDNP methods in studying their reaction mechanisms. Me3SnCOMe decomposes on attempted isolation if exposed to daylight. Therefore, to study the mechanisms of the photodecomposition of R3SnCOR it is reasonable to choose more steric hindered derivatives which are relatively stable under ambient conditions. Consequently, the regularities of the reaction mechanisms of a-stannyl ketones R3SnCOR have been studied with 2-methylpropanoyltripropylstannane" Pr3SnCOCHMe2 (9). 

Similar to the case of 7-heteronorbornadienes, the application of the CIDNP method has allowed us to identify the elementary stages of the photolytic decomposition of 26 . In addition to 27, the main reaction products also include oligogermanes and the product of photorearrangement of the initial 26, namely 6,6, 7,7 -tetramethyl-2,5-diphenyl-3,4-benzo-6,7-digermatricyclo[3.3.0.0]octane . Table 11 and Figure 17 show the polarization effects of methyl and aromatic 5,6-protons of the initial 26 observed in the photolysis in CsDs similar effects were also observed in c-C Di2 and in CCLj-CsDs mixtnres. 

A new technique for a determination of spin-multiplicities is the CIDNP-method. From emission signals in the nmr-spectrum a singlet or triplet state is deduced in insertion or addition reactions. This technique has been applied recently to 3a and 3g CIDNP-measurements in cyclohexane or CCI4 demonstrate that 3a reacts as singlet whereas 3g (di-f-butyl-derivative) involves a triplet intermediate. The latter experiment is in contrast with the results of stereo-specific addition of 3g to cis- or trans-huiene However the faster rate of addition versus insertion is thought to be responsible for this descrepancy... 

Electron transfer of cyclopropane systems—the CIDNP method... 

UDA, the plant lectin from rhizomes of the stinging nettle, is comprised of two covalently linked hevein domains. The interaction of UDA with chit-ooligosaccharides has been studied by multiple methods, including X-ray, NMR titration data, laser photo-CIDNP methods, titration microcalorimetry, and also fluorescence measurements. 

Figure 11 shows the result of this experiment on a solution of 5 mM N-acetyl tryptophan and 0.2 mM 3-N-carboxy-methyl lumiflavin, hereafter simply called flavin (see Figure 10). Positive enhancements can be observed for the aromatic C-2, C-4 and C-6 protons, while the CH2 group shows emission. This polarization pattern corresponds with a tryptophyl radical in which the electron spin is delocalized over the aromatic ring. It can further be noted that almost no flavin polarization is present in the difference spectrum. Figure 11c (weak lines are present at 2.6 and 4.0 ppm). This is due to cancellation of recombination and escape polarization as will be discussed in Section 5. The mechanism of the photoreaction undoubtedly involves triplet flavin (17). Since 1-N-methyl tryptophan shows similar CIDNP effects, the primary step most probably is electron transfer to the photo-excited flavin. This is also supported by a flash photolysis study by Heelis and Phillips (18). The nature of the primary step in the photoreactions with amino acids is important in view of the interpretation of "accessibility" of an amino acid side chain in a protein as seen by the photo-CIDNP method. This question is therefore the subject of further study.

Recently an X-ray structure became available for the bovine enzyme (24). This, combined with the results from numerous studies on this enzyme by De Haas and his coworkers has led to a mechanistic model for the catalytic activity. Another problem is the mode of activation of the pro-enzyme and the nature of the interface recognition site. The photo-CIDNP method has shed some light on this latter problem. 

Some features of the laser photo-CIDNP method will now be summarized. We have shown that hy employing a dye as sensitizer nuclear spin polarization can be generated in native proteins. When flavins are used as dyes nmr lines of Tyr, His, and Trp residues can be selectively enhanced when these residues are accessible to the photo-excited dye. Thus, the method constitutes a "surface probe" with the very high resolution of the nmr technique, which is capable of resolving lines due to individual residues. Large enhancements, often of the order of a magnitude or more, are translated in considerable time savings. 

Combined with a difference technique the method results in substantial simplifications of the complex protein nmr spectra and in an effective increase in the spectral resolution for the polarized groups. This permits detailed studies of various interactions of proteins, such as the enzyme-inhibitor interactions discussed in sections 5,2 and 6.2 for RNase A and lysozyme. For both enzymes the photo-CIDNP spectrum proved to be very sensitive to the presence of inhibitors, albeit in conpletely different ways Examples of applications to the study of protein-nucleic acid interactions will be treated by Hilbers et al. elsewhere in this volume. We believe that in spite of its brief existence the photo-CIDNP method is potentially a powerful tool for the study of protein structure in solution. 

Using the photo CIDNP method, resonances from particular amino acid side chains (tyrosine, histidine and tryptophan) can be selectively enhanced when these residues are situated at the surface of the protein. To this end a flavin dye, added to the sanq>le, is photo excited in the NMR probe by an argon laser. In this way triplet state flavin is generated, which in the case of tyrosine residues is able to abstract the phenolic hydrogen atom. Consequently a radical pair is formed (reaction 2 below) which reversibly yields flavin and tyrosine (reaction 3 below). 

The application of the photo-CIDNP method in conjunction with the introduction of deuterated amino acids holds great promise for the study of protein-protein and protein-nucleic acid interactions with NMR. In this contribution it has been shown that this allows partial assignment of the aromatic part of the protein spectrum in a convenient way. Binding of DNA fragments affects both the resonances of the phenylalanyl residues as well as the resonances of the tyrosines at the surface of the protein. The disappearance of the emission signals of the latter provides direct evidence for their Involvement in the interaction with DNA. 

---