Isotopic filters

Gemmecker, G. Isotope filter and editing techniques. In BioNMR in Drug Research (Methods and Principles in Medicinal Chemistry), Zerbe, O. (eds.), Wiley-VCH, Weinheim, 2003, Vol. 16, 373-390. 

Petros AM, Kawai M, Luly JR, Fesik SW. Conformation of two nonimmunosuppressive FK506 analogs when bound to FKBP by isotope-filtered NMR. FEBS Lett 1992 308 309-314. 

DPFGSE sequence with double tuned filters has been first proposed by Ogura et al. These filters demonstrate very high filtering efficiency for isotope-filtered, isotope-edited NOESY spectra.40... 

Isotope filtering/editing NMR techniques make use of differential isotopic labeling to simplify spectra and thus more easily extract information from complex systems. In the case of biomolecular NMR these will generally be intermolecular complexes between one biomacromolecule (for example, a protein) and a second species either another protein, a nucleic acid, or a ligand (generally a small organic molecule). 

The basic idea of isotope filtering/editing techniques is to selectively observe the subspectra of the labeled or unlabeled components only. A prerequisite for the application is of course the different isotopic composition of the different components. In an ideal case, we will have one compound which is completely labeled with the NMR active iso-tope(s) of one (or several) elements, and the other compound completely devoid of these isotopes (see Fig. 17.1). 

Fig. 17.1 The concept of isotope filtering and isotope editing, in an idealized way (for a more realistic picture, see Fig. 17.21). In a complex consisting of two (or more) differentially labeled compounds (left), the NMR spectra could be consider-...

Isotope filtering suppresses these signals, leaving only the signals of protons bound to the NMR inactive carbon and nitrogen isotopes (1H-12C, 1H-14N), as well as protons not bound to carbon or nitrogen at all (mainly -SH, -OH). 

Fig. 17.2 A realistic picture of the possibilities and limitations of isotope filtering and isotope editing, shown using the example of 15N. The selection of the isotope-labeled moiety is not significantly perturbed by the low level of 15N natural abundance (center). However, in the 15N-filtered case (right), only the 15N-bound protons of the...

Of course, isotope filtering is not restricted to such binary systems, but can also be applied to multicomponent systems such as multiprotein complexes, or one or more ligands bound to proteins or protein complexes. It is also conceivable to construct single molecules from sections with different isotopic labeling in order to selectively observe one part by isotopic filtering. However, in these cases a specific synthetic approach has to be designed to allow for efficient incorporation of the isotope labels into the appropriate parts only (for example, by fragment condensation or inteins see also Chapt. 1) [5]. 

In the isotope edited/ filtered spectra of a protein-ligand complex, the species actually observed is generally the complex itself. This is an important difference from transferred NOE or saturation difference techniques, where the existence of an equilibrium between free and bound species - and a certain rate of exchange between them - is essential (Chapts. 13 and 16). The general conditions for isotope filtering/editing are therefore identical to those required for standard protein NMR sample concentrations are usually limited by availability and solubility of the components to the order of 1 mM. Considerably lower concentrations will reduce the sensitivity of the experiments to unacceptable levels,... 

Generally the sensitivity of the isotope filtered/edited version of an NMR experiment will be comparable to that of the corresponding standard experiment. However, some reduction in signal intensity will occur caused by the additional pulses (due to pulse imperfections and Bi inhomogeneity) and delays (due to relaxation) of the filter elements. These losses can become significant in the case of large molecular weight complexes. 

When combining isotope filtering/editing with coherence transfer steps to multidimensional experiments, then further size restrictions apply. For example, isotope edited / filtered H TOCSY or COSY experiments are generally limited to systems of <10 kDa, because of their sensitivity to T2 relaxation. In larger systems, heteronuclear correlation spectroscopy can be used for the correspondingly labeled component, while structural information about both the labeled and unlabeled moiety can be extracted from isotope edi-ted/filtered NOESY spectra, respectively. 

However, it is possible to add the non-observed component (usually the ligand) in excess (solubility permitting) in order to cope with weak (i.e., pM-mM) binding affinities. Naturally, this requires an even higher performance of the isotope filtering technique used to suppress the (excess) component. 

In case (2), isotope filtering will (ideally ) remove all protons bound to the chosen hetero-nuclear isotope(s). In the case of 13C and/or 15N filtering (requiring 13C and/or 15N labeling of one component), there will be approximately the following selectivity for the unlabeled component ... 

Also, imperfections of the isotope filters (e.g., from compromises in the delay settings, see Sect. 17.3.2) will lead to additional leakage of signals from the (unwanted) labeled component, thus further lowering the selectivity. 

Obviously, this approach cannot be used for selecting the nonisotope-labeled components. In the following we will consider isotope filtering/editing techniques that do not use heteronuclear chemical shift evolution. 

An alternative way of realizing an isotope filter is shown in Fig. 17.4b, where the 90° phase difference between the two proton magnetizations is exploited [18]. A second 90° j1 ) pulse (of same phase as the excitation pulse) at the end of the period r =l/2j leaves the heteronuclear antiphase magnetization of the X-bound protons unaffected, while the other protons are converted to z magnetization ... 

The performance of the isotope filter is a critical step for the acquisition of interpretable NMR spectra of mixed labeled/unlabeled complexes. Therefore several additional variations with promising results have been proposed in the literature, but cannot be discussed within the scope of this chapter [26-28]. 

With another immunophilin, FK binding protein (FKBP), experiments were performed using isotope editing of the [U-13C]-labeled inhibitor ascomycin (bound to unlabeled FKBP) [34], as well as by isotope filtering with unlabeled ascomycin derivatives (bound to labeled FKBP) [35],... 

The idea of back transformation of a three-dimensional NMR experiment involving heteronuclear 3H/X/Y out-and-back coherence transfer can in principle be carried to the extreme by fixing the mixing time in both indirect domains. Even if one-dimensional experiments of this kind fall short of providing any information on heteronuclear chemical shifts, they may still serve to obtain isotope-filtered 3H NMR spectra. A potential application of this technique is the detection of appropriately labelled metabolites in metabolism studies, and a one dimensional variant of the double INEPT 111/X/Y sequence has in fact been applied to pharmacokinetics studies of doubly 13C, 15N labelled metabolites.46 Even if the pulse scheme relied exclusively on phase-cycling for coherence selection, a suppression of matrix signals by a factor of 104 proved feasible, and it is easily conceivable that the performance can still be improved by the application of pulsed field gradients. 

Lee, W., Revington, M. J., Arrowsmith, C. and Kay, L. E. (1994). A pulsed field gradient isotope-filtered 3D 13C HMQC-NOESY experiment for extracting intermolecular NOE contacts... 

Breeze, A. L. (2000). Isotope-filtered NMR methods for the study of biomolecular structure and interactions. Prog. Nucl. Magn. Reson. Spectrosc. 36, 323-372. 

All the spectroscopic approaches applied for structural characterization of mixtures derive from methods originally developed for screening libraries for their biological activities. They include diffusion-ordered spectroscopy [15-18], relaxation-edited spectroscopy [19], isotope-filtered affinity NMR [20] and SAR-by-NMR [21]. These applications will be discussed in the last part of this chapter. As usually most of the components show very similar molecular weight, their spectroscopic parameters, such as relaxation rates or selfdiffusion coefficients, are not very different and application of these methodologies for chemical characterization is not straightforward. An exception is diffusion-edited spectroscopy, which can be a feasible way to analyze the structure of compounds within a mixture without the need of prior separation. This was the case for the analysis of a mixture of five esters (propyl acetate, butyl acetate, ethyl butyrate, isopropyl butyrate and butyl levulinate) [18]. By the combined use of diffusion-edited NMR and 2-D NMR methods such as Total Correlation Spectroscopy (TOCSY), it was possible to elucidate the structure of the components of this mixture. This strategy was called diffusion encoded spectroscopy DECODES. Another example of combination between diffusion-edited spectroscopy and traditional 2-D NMR experiment is the DOSY-NOESY experiment [22]. The use of these experiments have proven to be useful in the identification of compounds from small split and mix synthetic pools. 

Gonnella N, Lin M, Shapiro MJ, Wareing JR, Zhang X, Isotope-Filtered affinity NMR, J. Magn. Reson., 131 336-338, 1998. 

---