Solid-State NMR Techniques for Structural Investigation of Proteins

STUDIA UNIVERSITATIS BABES-BOLYAI, PHYSICA, SPECIAL ISSUE, 2003

SOLID-STATE NMR TECHNIQUES FOR STRUCTURAL INVESTIGATIONS OF PROTEINS

Claudiu Filip

National Institute for Research and Development of Isotopic and Molecular Technology, Cluj-Napoca, Romania

Abstract: Applications of Solid-State NMR spectroscopy to structural investigation of large proteins requires extending its techniques from selectively to uniformly labeled compunds. Here we present two novel MAS NMR pulse sequences for simultaneous measurement of carbon- nitrogen distances in uniformly 13C,15N-labeled solids.

Introduction Solid-state nuclear magnetic resonance (SS-NMR) spectroscopy currently undergoes a remarkable evolution from being mainly physics- oriented method for studying small molecules in inorganic or organic materials toward biology-oriented applications handling complex clusters of atoms for structure analysis of large bio-molecules. Recent advances in magic-angle-spinning (MAS) SS-NMR instrumentation and methodology [1] have greatly contributed to extending its applicability from selectively to multiply 13C,15N labeled peptides and proteins. In a first step of obtaining full 3D structures for large proteins, site-specific resonance assignments are a source of valuable structural information. Whereas NMR experiments for sequential resonance assignments generally exploit the strongest spin-spin interactions and are relatively straightforward to implement in uniformly 13C,15N labeled systems, long-range distance measurements in these systems are more complicated. The nuclei form a tightly coupled network via through-space (dipolar coupling) and through-bond (J-coupling) spin-spin interactions; consequently, the strongest couplings can interfere with the accurate determination of the weaker ones. In the following we present new MAS NMR pulse sequences for simultaneous measurement of multiple carbon-nitrogen distances in uniformly 13C,15N labeled solids. The experiments are demonstrated in model U-13C,15N labeled tri-peptides N-formyl-L- CLAUDIU FILIP

Met-L-Leu-L-Phe (MLF), where 26 13C-15N distances up to ~5-6 Å were determined.

Long-range distance constraints in uniformly 13C,15N-labeled polypeptides The pulse sequences employ transferred echo double resonance (TEDOR) for 13C- 15N coherence transfer and 15N and 13C frequency labeling for site-specific resolution, and as such build on several previously described 3D TEDOR experiments. The novel feature of the 3D TEDOR pulse sequences presented here is that they are specifically designed to circumvent the detrimental effects of homonuclear 13C-13C J-couplings on the measurement of weak 13C-15N dipolar couplings. Homonuclear J-couplings: (i) generate undesirable anti-phase and multiple-quantum (MQ) spin coherences, which lead to spurious cross-peaks and phase-twisted lines in the 2D 15N-13C correlation spectra, and thus degrade the spectral resolution and prohibit the extraction of reliable cross-peak intensities and (ii) significantly reduce cross-peak intensities for strongly J-coupled 13C sites (e.g., CO and C). The first pulse sequence (Fig. 1a) employs z-filter periods to suppress the anti-phase and MQ coherences and generates 2D spectra with pure absorption mode peaks for all TEDOR mixing times. The second sequence (Fig. 1b) uses band-selective 13C pulses to refocus J-couplings between 13C spins within the selective pulse bandwidth and 13C spins outside the bandwidth. The NMR spectra were recorded at 11.7 T (500.1 MHz 1H, 125.8 MHz 13C, 50.7 MHz 15N) using a triple-resonance MAS probe. The spinning frequency of 10.0 kHz was used in the all experiments and regulated to 5 Hz and the samples

Figure 1: Z-Filtered (ZF) and band-selective (BASE) 3D TEDOR pulse sequences. Narrow and wide solid rectangles represent /2 and  pulses, respectively. In (a) the modifications to the recently described 3D TEDOR experiment [] are highlighted by gray rectangles. The modifications consist of two z-filter periods, , which eliminate undesired multiple quantum and anti-phase spin coherences generated by 13C-13C J-evolution (see text for SOLID-STATE NMR TECHNIQUES FOR STRUCTURAL INVESTIGATIONS OF PROTEINS

details). In (b) band- selective 13C Gaussian  pulses refocus 13C-13C J-couplings to nuclei outside the  pulse bandwidth

were centerpacked in the rotors to minimize the effects of rf inhomogeneity. TheU- 13C,15N labeled tri-peptide N-formyl-L-Met-L-Leu-L-Phe (MLF) represents an excellent model system for the 3D TEDOR experiments as its three-dimensional structure (see Figure 2) has been already determined using X-ray crystallography [2]. Using a zero order approximation [3] we proceed to describe the 3D TEDOR experiments by neglecting all orientation dependent terms. In summary, the main features of the simulation model are: (i) for the ith 13C nucleus coupled to 15 13 15 Ni N spins all Ni C- N distances are determined by simultaneously fitting the

experimentally observed buildup of cross-peaks located at (Nk, Ci), k = 1,2,,Ni, using eqs (3)-(5) in Ref. [3] with Ni + 2 fit parameters (Ni dipolar coupling

constants, Dik, a relaxation rate, i, and an amplitude scaling factor, i), (ii) no

Figure 2: X-ray crystal structure of N-formyl-L-Met-L-Leu-L-Phe-OMe (MLF)

knowledge of dipolar tensor orientations and no powder averaging are required, and (iii) the cross-peak intensities at long mixing times are inversely 15 13 proportional to the number of N spins, Ni, simultaneously coupled to the C 1 Ni nucleus, Vik(tmix) 2  , which places a practical limit on the magnitude of the weakest 13C-15N couplings that can be determined using 3D TEDOR methods (for the tri-peptide investigated in this work 13C-15N dipolar couplings corresponding to distances in the 4-6 Å regime could be determined). We note that for ZF-3D TEDOR the buildup of cross-peaks for the ith 13C nucleus also depends on the 13 coupling topology of the mi J-coupled C nuclei. These parameters are held

 CLAUDIU FILIP constant in the simulations since 13C-13C J-couplings in peptides do not vary significantly for different types of couplings (i.e., CO-C, C-C, etc.) [4]. Figure 3 shows representative 2D 15N-13C correlation spectra in N-formyl- [U-13C,15N]L-Met-L-Leu-L-Phe for the ZF-3D TEDOR experiment. A large number of dipolar correlations (~20) are observed in the aliphatic 13C spectral region for longer mixing period (Figure 10c), and each cross-peak carries information about 13C-15N internuclear distances. The 13C and 15N resonance

Figure 3: Representative two-dimensional slices from broadband 3D TEDOR experiments in N-formyl-[U-13C,15N]L-Met-L-Leu-L-Phe. The 2D spectra shown in (a) and (c) (with the 1D slices corresponding to Met, Leu and Phe 15N resonances shown in (b) and (d)) were 1 recorded at 500 MHz H frequency and r/2 = 10 kHz ± 5 Hz using the pulse sequence in Figure 1a with the z-filter delays of  = 200 s. Additional experimental parameters are described in the text. The 3D data set consisted of 16 2D spectra acquired for TEDOR mixing times up to 18 ms and incremented in a non-linear fashion in order to maximize the information content and minimize the duration of the experiment. Each 2D spectrum was acquired as (19, 1024) complex points in (t1, t2), with time increments of (800, 20) s SOLID-STATE NMR TECHNIQUES FOR STRUCTURAL INVESTIGATIONS OF PROTEINS

resulting in acquisition times of (14.4, 20.5) ms. Each FID was 128 scans, with a 3 s recycle delay, yielding a total measurement time of ~67 h.

The 13C and 15N resonance assignments for MLF have been presented in previous studies [5]. The 3D TEDOR cross-peak buildup curves for selected 13C sites

Figure 4: 3D TEDOR cross-peak buildup curves in N-formyl-[U-13C,15N]L-Met-L-Leu-L- Phe. The buildup curves were acquired using the pulse sequence in Figure 1a and correspond to cross-peaks observed for the Met C (a) and Leu C (b) resonances in the 2D correlation spectra in Figure 3. For each 13C the experimental buildup curves corresponding to the Met (), Leu() and Phe () 15N sites were fit (—) using analytical expressions based on the zeroth order Bessel function approximation. The values of 13C-13C 1 J-coupling constants used in the fits were 55 Hz for JC O C  and 34 Hz for all other J- couplings. The cross-peak buildup curves shown in the figure correspond to carbon- nitrogen distances in the 2.5-6 Å regime. Similar simulations were performed for the remaining cross-peaks (c.f. Figure 3) and the 13C-15N internuclear distances are summarized in the table

corresponding to distances in the 2.5-6 Å range are shown in Figure 4. Similar experiments were performed using the BASE-3D TEDOR pulse sequence [3] (spectra not shown) and the results are presented in the Table.

Conclusions In conclusion, we have presented two 3D TEDOR NMR pulse sequences for the simultaneous measurement of multiple carbon-nitrogen distances in uniformly 13C,15N labeled in site-specific fashion. The novel feature of the sequences is that they are compensated for the detrimental effects of 13C-13C J- couplings on the measurements of long-range carbon-nitrogen distances. The experiments are robust and straightforward to implement in practice. Qualitatively, the presence of 13C-15N correlations in the 2D spectra in itself provides direct information about three-dimensional structure. More accurate distance measurements were performed using a simple analytical simulation model, which

 CLAUDIU FILIP accounts explicitly for multiple spin-spin couplings contributing to cross-peak buildup.

Table: InternuclearC-N distances N-formyl-L-Met-L-Leu-L-Phe-OMe (MLF)

rC–N (Å) BASE atoms ZF-3D TEDOR X-ray 3D TEDOR Met(N) Met(C’) — 2.5 2.40 Met(C) 1.5 — 1.46 Met(C 2.5 2.6 2.50 Met(C 3.8 — 3.04 Met(C 5.3 — 5.71 Leu(C2 6.1 — 6.28 Leu(N) Met(C’) — 1.3 1.33 Met(C) 2.6 — 2.44 Met(C 3.2 3.2 3.20 Met(C 5.0 — 4.56 Met(C 5.8 — 5.93 Leu(C’) — 2.4 2.46 Leu(C) 1.5 — 1.46 Leu(C 2.6 2.6 2.50 Leu(C2 3.9 — 3.63 Phe(N) Met(C’) — 3.4 3.41 Met(C 3.9 3.8 4.06 Met(C 5.6 — 5.43 Met(C 5.2 — 5.62 Leu(C’) — 1.3 1.34 Leu(C) 2.3 — 2.43 Leu(C 3.2 3.1 3.12 Leu(C2 6.0 — 5.38 Phe(C’) — 2.7 2.37 Phe(C) 1.4 — 1.47 Phe(C 2.6 2.6 2.53

Acknowledgment. The author acknowledges financial support from the Romanian Academy by the Research Grant No. 22 /2003.

References [1] A.E. Benett, R.G. Griffin, S. Vega, Solid State NMR IV: Methods and Applications of Solid-State NMR, Springer, Berlin (1994) [2] E. Gavuzzo, F. Mazza, A. Scatturin, Int. J. Pept. Protein Res., 34, 409 (1989) SOLID-STATE NMR TECHNIQUES FOR STRUCTURAL INVESTIGATIONS OF PROTEINS

[3] C.P. Jaroniec, C. Filip, R.G. Griffin, J. Am. Chem. Soc., 124, 10728 (2002) [4] J. Cavanagh, W. Fairbrother, A.G. Palmer, N. Skelton, Protein NMR spectroscopy: Principles and Practice, Accademic Press, San Diego (1996) [5] C.P. Jaroniec, B. Tounge, J. Hertzfeld, R.G. Griffin, J. Am. Chem. Soc., 123, 3507 (2001)