Structural studies of a selective nitric oxide sensor by solution NMR
The SO2144 H-NOX domain from Shewanella oneidensis (So) is a homolog of the mammalian nitric oxide receptor soluble guanylate cyclase (sGC) and has similar spectroscopic and biochemical properties (3). In particular, it is stable in the ferrous oxidation state, not prone to rapid autooxidation, and forms complexes with the diatomic gases NO and CO, forming the analogous heme-CO and heme-NO complexes, but not O2 (3). More specifically, CO coordinates to the ferrous heme of SO2144 forming a six coordinate complex (Fe(II)-CO WT). When NO binds to the heme cofactor of the SO2144 H-NOX, the axial histidine iron bond is cleaved, and a five coordinate complex (Fe(II)-NO WT) is formed. The SO2144 H-NOX and the SO2145 histidine kinase have been shown to interact directly to form a complex (3). The autophosphorylation activity of SO2145 is suppressed by interaction with NO bound SO2144 H-NOX, but SO2144 H-NOX without ligand (Fe(II) WT) does not inhibit the activity of the kinase (Figure 2.1) (3). The Fe(II)-CO WT H-NOX inhibits the SO2145 autophosphorylation activity with an IC50 of 84 ± 5 mM (3). The Fe(II)-NO WT H-NOX inhibits kinase activity with an IC50 of 9 ± 2 mM and is, therefore, a more potent inhibitor than Fe(II)-CO WT (3).
While numerous biochemical studies have established relationships between the ligation of the H-NOX domain with diatomic molecules and the activity of corresponding signaling systems, no study to date has provided high-resolution structural insights into the corresponding protein structure-function relationships. In non-oxygen binding H-NOX domain containing proteins such as sGC or the SO2144 H-NOX/SO2145 histidine kinase pair displacement of the axial histidine from the iron correlates with altered signaling activity (3, 4). In sGC, axial histidine bond cleavage precedes activation of cGMP synthase activity (5-7). Analogously, in the SO2144 H-NOX/SO2145 histidine kinase pair, axial histidine bond cleavage is associated with kinase inhibition (3). To identify the conformational changes in the SO2144 H-NOX domain that initiate inhibition of the SO2145 kinase directly, we initiated the determination of its solution structure by NMR spectroscopy in different states of kinase inhibitory activity. NMR spectroscopy was selected as the method of structure determination because previous attempts to crystallize a non-oxygen binding H-NOX domain were unsuccessful (unpublished data).
The heme cofactor of the SO2144 H-NOX is paramagnetic in both the Fe(II) WT state which does not inhibit the kinase and the Fe(II)-NO WT state which does. These states have electronic spin of S = 2 and S = 1/2, respectively (8, 9). NMR structure determination of paramagnetic proteins is technically challenging because electron-nuclear spin relaxation broadens resonance lines beyond the limit of detection for protons within a ~10 Å radius around the heme iron (10). However, six coordinate Fe(II)-CO complexes are diamagnetic (11) allowing more detailed NMR analysis, so efforts were focused on diamagnetic CO derivatives.
2.2 Materials and Methods
2.2.1 Protein expression, purification, and mutagenesis
The WT SO2144 H-NOX domain was expressed and purified as previously described (3). The H103G mutant was prepared from a pET20b expression vector with the gene encoding the full length SO2144 H-NOX domain (residues 1-181) (3) by site-directed mutagenesis using pairs of complementary primers and the QuikChange kit (Stratagene). The forward and reverse primers were 5’-GTC ATG GGG ATC CAT GAT GTG ATC GGT TTA GAG GTG AAT AAG-3’ and 5’-CTT ATT CAC CTC TAA ACC GAT CAC ATC ATG GAT CCC CAT GAC-3’. Introduction of the desired mutation was confirmed by DNA sequencing (UC Berkeley, DNA Sequencing Facility). Wild type (WT) and H103G mutant H-NOX domains from S. oneidensis were expressed in E. coli BL21 (DE3) pLysS cells. Cultures were grown at 37 °C to an OD600 of 0.6 in M9 media isotopically enriched for 15N or 13C/15N and then cooled to 27 °C. Expression of WT protein was induced by addition of isopropyl b-D-thiogalactopyranoside (IPTG) to 1 mM and the medium was supplemented with aminolevulinic acid (ALA) to 1 mM to facilitate heme synthesis. Expression of H103G was induced by addition of IPTG to 1 mM. Immediately prior to IPTG induction the media was supplemented with ALA to 1 mM and imidazole to 10 mM. Specific incorporation of isotope labeled amino acids was achieved by the addition of the appropriately labeled amino acid during the induction of expression in M9 media. Following a previously reported protocol, expression of H-NOX in 15N isotopically enriched M9 medium in the presence of 13C enriched ALA resulted in 12C protein and 13C heme enrichment sufficient to selectively observe the heme methine 1H and 13C resonances (12). Cultures were grown for 5-6 hours and harvested. Purifications of wild type and H103G H-NOX were carried out as described previously (13, 14). H103G mutant samples were kept in 10 mM imidazole to maintain heme binding. Protein purity, identity, and molecular mass of H-NOX samples were verified by SDS-PAGE, electronic absorption spectroscopy, and electrospray ionization mass spectrometry (ESI-MS).
2.2.2 NMR sample preparations
NMR experiments were performed on a Bruker DRX spectrometers (500, 600, 800, and 900 MHz) equipped with 1H/15N/13C cryoprobes. Wild type H-NOX NMR samples contained 90 % H2O/10 % D2O or 100 % D2O and 50 mM K3PO4, 5 mM dithiothretol, 5 % glycerol at pH 7.4. H103G NMR samples were prepared in the same manner with the addition of 10 mM imidazole. The Fe(II)-CO and Fe(II)-NO complexes were prepared as described (3). The concentrations of NMR samples were 0.4-0.8 mM. The samples were loaded into sealed microcells (Shigemi, Allison Park, PA) in an anaerobic chamber (Coy Laboratory Products) prior to NMR measurement. Electronic absorption spectra of samples before and after data collection confirmed that the oxidation and ligation state of the heme did not change during the course of data collection. Aligned samples for the measurement of residual dipolar couplings (RDCs) were prepared by addition of pf1 phage (Asla Biotech, Riga, Latvia) to a final concentration of approximately 15 mg/mL (15). The magnitude of alignment was quantified by monitoring the D2O quadrupolar splitting.
2.2.3 NMR data analyses
NMR data were processed with the NMRPipe software package (16). The program nmrDraw, a component of NMRpipe, was used to assign and integrate RDC data. CARA was used for resonance assignments and analysis of NOE spectra (17). The software programs Mathematica and SFIT from James J. Chou (http://sbgrid.org/chou/index.html) were utilized to extract J coupling constants from quantitative-J RDC data sets (1). The program TALOS was used to estimate backbone dihedral angles from characteristic chemical shifts (18). Fitting of residual dipolar couplings (RDCs) to structures was done by singular value decomposition (SVD), using the program iDC (19, 20).
2.2.4 Assignment of protein backbone, sidechain, and heme resonances
1H, 13C, and 15N protein and 1H heme resonance assignments for Fe(II)-CO WT were obtained at 35 °C by using the following triple resonance experiments: 15N-1H HSQC (21, 22), 3D HNCA (23), 3D HNCO (24, 25), 3D HNCACB (26), 3D CBCA(CO)NH (27), 3D HNHA (28), 3D HBHA(CO)NH (29), 3D HN(CA)CO (24, 30), 3D HC(CO)NH (31), 3D C(CO)NH (31), 3D HCCH-TOCSY (32), 3D HCCH-COSY (32), 3D 13C-NOESY-HSQC (100 ms mixing time) (33), 3D 15N-NOESY-HSQC (100 ms mixing time) (33), 2D 13C-filtered [F1,F2] NOESY (100 ms mixing time) (34), 2D NOESY (35), 2D TOCSY (36, 37), 2D double quantum filtered COSY(38), and 2D constant time 1H-13C HSQC (39). 1H, 13C, and 15N protein and 1H heme resonance assignments for Fe(II)-CO H103G were obtained at 25 °C by using the following experiments: 15N-1H HSQC, 3D HNCA, 3D HNCO, 3D CBCA(CO)NH, 3D HBHA(CO)NH, 3D C(CO)NH, 3D HCCH-TOCSY, 3D HCCH-COSY, 3D 13C-NOESY-HSQC (100 ms mixing time), 3D 15N-TOCSY (40), 3D 15N-NOESY-HSQC (100 ms mixing time), and 2D constant time 1H-13C HSQC. 1H, 13C, and 15N resonance assignments for Fe(II)-NO wild type H-NOX were obtained at 35 °C by the following experiments: 15N-1H HSQC, 3D HNCA, 3D HNHA, 3D HCCH-TOCSY, 3D HCCH-COSY, 3D 13C-NOESY-HSQC (100 ms mixing time), 3D 15N-NOESY-HSQC (100 ms mixing time), and 2D constant time 1H-13C HSQC.
Amino acid specific 15N labeling was used to confirm the resonance assignments of L, E, D, N, Q, and V residues. 2D HNCO spectra recorded on a double-labeled sample (13C carbonyl proline/15N leucine) were utilized for the assignment of L115 (41). 5-13C labeled ALA was utilized to specifically enrich the heme methine carbon positions with the 13C isotope (12). 1H and 13C heme methine resonances were assigned by recording a 3D 13C-NOESY-HSQC (100 ms mixing time) on a 12C protein and specifically 13C labeled heme sample. Methine assignments could be made unambiguously by using initial structures to aid NOE assignment. Stereo-specific assignment of the g 13C methyl groups of valines and the d 13C methyl groups of leucine was accomplished by preparation of a 10% 13C-labeled protein sample and monitoring the presence or absence of Jcc in a 1H-13C HSQC spectra as previously described (42, 43).
2.2.5 Identification of distance restraints
Intramolecular protein distance restraints were obtained from 3D 15N-NOESY-HSQC, 2D NOESY, and 3D 13C-NOESY-HSQC all collected with 100 ms mixing times (33). Intermolecular protein-heme distance restraints were obtained from 2D 13C-filtered [F1] NOESY (34) and 3D 13C-NOESY-HSQC obtained with 100 ms mixing times. Hydrogen bond restraints for the protein were defined from slowly exchanging amide protons and analysis of characteristic NOE patterns indicative of elements of regular secondary structure (44).
2.2.6 Residual dipolar coupling measurements
Residual dipolar coupling (RDC) constants were obtained by calculating differences in the J couplings from unaligned and pf1 phage aligned samples (45). For both Fe(II)-CO WT and Fe(II)-CO H103G, HN and HaCa RDCs were measured by using 3D HNCO (1DNH) and CBCA(CO)NH (1DHaCa) type experiments. Additionally, 1H-13C heme methine RDCs were determined using a 2D 1H-13C HSQC quantitative-J experiment (2). The 1H-15N couplings were measured at 800 MHz by collecting a 3D HNCO without 1H decoupling during 50 ms of constant time 15N evolution for Fe(II)-CO WT (46) and by collecting a 2D IPAP-HSQC for the Fe(II)-CO H103G (47). The one-bond 1Ha13Cacouplings were obtained at 900 MHz with a modified quantitative-J CBCA(CO)NH optimized for measuring backbone HaCa RDCs only (1, 48).
2.2.7 Structure calculations
Initial structure refinements used iterative cycles of manual NOE assignment followed by seven cycles of automated NOESY spectra analysis with the CANDID algorithm embedded in the CYANA 2.1 software package (49). The structures were then refined with a three step protocol against RDCs using the program XPLOR-NIH (50). In XPLOR-NIH, the initial fold of the structure was first calculated from an initial random coil protein, with a random heme orientation, using NOE derived distance restraints and backbone dihedral restraints derived from chemical shifts (TALOS). This was done with a high-temperature simulated annealing (SA) protocol (51). The distance restraints were enforced with soft-square potentials with force constants of 50 kcal mol-1 Å-2. Other force constants were gradually increased during the course of each SA trajectory: k(dihed) = 5 à 200 kcal mol-1 rad-2, k(vdw) = 0.002 à 1.0 kcal mol-1 Å-2, k(impr) = 0.1 à 1.0 kcal mol-1 degree-2, and k(bond angle) = 0.4 à 1.0 kcal mol-1 degree-2. During the SA run, the thermal bath was first held constant at 1000 K and the system was annealed for 16.5 ps of Verlet molecular dynamics with time steps of 5 fs. The thermal bath was then cooled from 1000 to 100 K with temperature steps of 50 K, and the system was annealed with 2.78 ps of Verlet molecular dynamics at each temperature step, using time steps of 5 fs. A total of 200 structures were calculated using this protocol.
Each of these structures was then further refined against the distance restraints, dihedral angle restraints, and RDC restraints with a low temperature SA protocol similar to that previously described (52). RDCs were normalized to those of the amide proton (HN) and nitrogen (N) atom pairs (45). The force constants for NOE and experimental dihedral angle restraints were fixed at 20 kcal mol-1 Å-2 and 10 kcal mol-1 rad-2, respectively. No NOEs could be identified between the heme propionates and other atoms in the protein. Five Å distance restraints were utilized to restrain the positions of the conserved heme propionate side chain/YxSxR motif, based upon the observation that the YxSxR motif is: (i) strictly conserved in the H-NOX family of proteins and (ii) crystallographic data have shown heme-YxSxR interactions in two different prokaryotic H-NOX domains (53)(54). Addition of the heme proprionate-YxSxR restraints did not result in the violation of any experimental distance restraints. The conformation of the heme cofactor was restrained by methine 1H-13C RDC values. Heme planarity restraints were enforced during the first stage of simulated annealing and were removed during the low temperature simulated annealing runs. The 20 structures with the lowest energy, no NOE violations > 0.3 Å, and no angle violations > 5o were chosen to represent the Fe(II)-CO WT and Fe(II)-CO H103G solution structures and were deposited in the Protein Data Bank. Structure quality was assessed with the protein structure validation software suite (PSVS). (55).
2.3 Results and Discussion
2.3.1 Expression and purification of the SO2144 H-NOX domain for NMR analysis
The SO2144 H-NOX domain was prepared via a previously described protocol in E. coli (3, 13). As shown in Figure 2.2, the SO2144 H-NOX domain is composed of 181 amino acids. For NMR spectroscopy, E. coli transformed with a plasmid encoding SO2144 (3) were grown on minimal M9 media enriched in 15N or both 15N and 13C. The protein was purified to >95% homogeneity as verified by SDS-PAGE (Figure 2.3). By SDS-PAGE, the expressed protein has a molecular weight of ~20 kDa. Mass spectrometry of unlabeled protein was utilized to verify that recombinant expression of SO2144 in E. coli produced the desired protein product. The molecular mass of SO2144 measured by electrospray ionization mass spectrometry (ESI-MS), 20510.0 Da, agreed with the theoretical molecular mass, 20,510.6 Da, calculated from the amino acid sequence (Figure 2.4) to within ±1 Da. Also, matrix-assisted laser desorption/ionization (MALDI) mass spectrometry analysis of the product peptides from a tryptic digest of an isotopically labeled untagged protein sample following expression in 15N labeled M9 minimal media revealed peptides with molecular masses consistent with the SO2144 H-NOX domain with ~95% 15N isotope enrichment (Figure 2.5). Therefore, recombinant expression of SO2144 in E. coli yields the desired protein product with suitable isotope enrichment for structural analysis by NMR spectroscopy.
2.3.2 Oxidation and ligation state of the SO2144 H-NOX domain for NMR analysis
After aerobic purification, the electronic absorption spectrum (UV-Vis) of the SO2144 H-NOX domain displays a sharp Soret absorbance at 429 nm and a single broad peak in the a/b region at 561 nm, consistent with a Fe(II), pentacoordinate, high-spin (S = 2) heme (Figure 2.6) (6). Upon addition of nitric oxide (NO), the Soret shifts to 399 nm with the appearance of a shoulder at 481 nm and peaks at 539 nm and 570 nm (Figure 2.6), consistent with a pentacoordinate nitrosyl-heme (Fe(II)-NO) complex. Addition of carbon monoxide (CO) to the Fe(II) SO2144 H-NOX domain shifts the Soret to 423 nm and the single broad a/b peak is replaced by two distinct peaks at 567 and 539 nm (Figure 2.6), consistent with a hexacoordinate complex with CO and imidazole axial ligands (Fe(II)-CO). Together, these data indicate that the SO2144 H-NOX domain binds heme and forms stable complexes with NO and CO, displays UV-Vis spectra nearly identical to sGC, and has no measureable affinity for O2, confirming previous observations (3, 6).
2.3.3 Assignments of chemical shifts in SO2144 H-NOX domain NMR spectra in the Fe(II)-CO WT state
The diamagnetic Fe(II)-CO complex of WT SO2144 H-NOX domain was prepared and analyzed by 1H-15N HSQC spectroscopy (Figure 2.7). This protein gave a high quality spectrum as evidenced by the homogeneous cross peak line widths and the appearance of ~180 individual amide cross-peaks. 1H, 15N, and 13C resonance assignments for Fe(II)-CO WT were made using standard 2D and 3D techniques (Appendix 1) (44, 56). A representative set of assigned 3D-HNCA data is shown in Figure 2.8 illustrating the resonance assignments of peptide backbone resonances (HN, N, and CA) of residues 148-152. Specific 15N labeling, in conjunction with 1H-15N HSQC spectroscopy, was utilized to confirm resonance assignments made by 3D NMR techniques as outlined in Materials and Methods. Heme resonance assignments were made through the use of isotope filtered NOESY experiments (34). Further heme assignments were made by 3D 13C-NOESY following the initiation of structure calculations described below. The protein backbone, protein side chain, and heme resonance assignments provided a foundation for the interpretation of NMR spectra that report on the geometry of the Fe(II)-CO WT SO2144 H-NOX domain (BMRB Code: 16276).
2.3.4 Structural analysis of the SO2144 H-NOX domain in the Fe(II)-CO ligation state:
distance and dihedral angle restraints
The solution structure of Fe(II)-CO WT was solved using information from a combination of 2D, 3D 15N-edited, 3D 13C-edited, and 2D 13C-filtered NOESY experiments. NOESY cross-peaks were assigned with a semi-automated protocol consisting of iterative rounds of manual and automated NOESY assignment via the CYANA automated NOESY analysis software package (49) and CARA (17). Following the completion of this procedure 3112 total, heme-protein (70) and protein-protein (3042), NOE cross-peaks were unambiguously assigned. Representative strips of assigned 3D 13C-edited NOESY data are shown in Figure 2.9. Following assignment, the volumes of NOE cross-peaks were obtained by integration and distance restraints were calculated from the corresponding peak intensities using CYANA (49). At the end of the NOE assignment procedure 150 hydrogen bond restraints, between backbone amide and carbonyl moieties in elements of regular helical and b-sheet secondary structure were identified by NOE analysis and hydrogen/deuterium exchange (44). Protein backbone dihedral angle information was provided from amide hydrogen (HN), alpha hydrogen (HA), alpha carbon (CA), beta carbon (CB) chemical shift values by the chemical shift index method, TALOS (18). In total, 277 φ and ψ backbone dihedral angle restraints were obtained through this protocol. These three sets of distance and dihedral angle restraints were then collected and formatted for use in structure calculations with the XPLOR-NIH NMR structure calculation software package (Appendix 2) (50).
2.3.5 Residual dipolar coupling restraints
Following the completion of distance and dihedral angle restraint measurements, three sets of residual dipolar coupling (RDC) restraints were obtained. Protein alignment was introduced through the use of pf1 phage (Figure 2.10) (15). First, 129 amide HN-N RDCs (1DNH) were measured using a modified 3D HNCO experiment (46). A representative segment of these data are shown in Figure 2.11. The RDC for residue 168 is seen in Figure 2.11 as a -7.1 Hz difference between the unaligned and aligned SO2144 H-NOX multiplet components. Second, 133 HA-CA RDCs (1DHaCa) were measured with a modified CBCACONH experiment by quantitative-J spectroscopy (Figure 2.12) (1)(48). In this protocol, the experiment is run three separate times with different amounts of evolution (1.93, 3.73, and 7.22 ms) under the 1JHaCa (unaligned) and 1JHaCa + 1D HaCa (aligned) coupling constant(s) during a dephasing delay (Figure 2.13). Integration of the resulting peak intensities and fitting evolution curves to the data provides the value of the 1JHaCa (unaligned) and 1JHaCa + 1D HaCa (aligned) coupling constants. Three representative HN-CA planes from this experiment are shown in Figure 2.14 to illustrate the modulation of the peak intensities by 1JHaCa (unaligned) and 1JHaCa + 1D HaCa (aligned). Last, the heme methine 13C atoms were labeled via 5-13C labeled ALA as shown schematically in Figure 2.15 (12). The 1H and 13C chemical shifts of the heme methine positions were assigned by 2D 1H-13C HSQC and 3D 1H-13C edited NOESY (Figure 2.16). Heme methine 1H-13C RDCs (Figure 2.18) were then measured by a quantitative-J 2D 1H-13C HSQC experiment illustrated in Figure 2.17 (2). The three sets of residual dipolar couplings were collected and utilized in structure calculations (Appendix 3).
2.3.6 Structure calculations of the SO2144 H-NOX domain in the Fe(II)-CO ligation state
A three step simulated annealing protocol outlined in Materials and Methods was utilized to calculate 200 structures of Fe(II)-CO WT consistent with all the distance, dihedral angle, and residual dipolar coupling geometric restraints. The three step XPLOR-NIH protocol is provided in Appendix 4. The 20 lowest energy structures were selected as described in Materials and Methods: Structure Calculations (PDB Code: 2kii). A detailed description of the results of these structure calculations are provided in chapter 3.
2.3.7 Chemical shift studies of the SO2144 H-NOX domain in the Fe(II)-NO ligation state
Following completion of the solution structure of the SO2144 H-NOX domain in the Fe(II)-CO ligation state, a preliminary series of chemical shift difference studies were performed to identify regions of the protein that undergo conformational changes associated with the binding of nitric oxide to the heme, formation of the Fe(II)-NO WT state, and cleavage of the axial-histidine iron bond. The SO2144 H-NOX domain was prepared in the paramagnetic (S = ½) Fe(II)-NO WT ligation state and analyzed by 1H-15N HSQC spectroscopy (Figure 2.19). The Fe(II)-NO protein gave an excellent spectrum as evidenced by the homogeneous cross peak line widths and the appearance of ~160 individual cross-peaks. An overlay of the 1H-15N HSQC spectra from the Fe(II)-CO WT and Fe(II)-NO WT ligation states is shown in Figure 2.20. As is apparent from the HSQC overlay, significant chemical shift changes are seen between these two ligation states. The chemical shift changes between the Fe(II)-CO WT and Fe(II)-NO WT states are caused by both protein conformational differences and heme electronic differences. 1H, 15N, and 13C resonance assignments for the NO-bound SO2144 H-NOX domain were made using standard 2D and 3D techniques (Appendix 5) (44, 56). However, resonances belonging to spins within a ~10 Å radius of the heme iron were not visible due to paramagnetic broadening (Figure 2.21). The protein backbone amide H and HN chemical shift differences were calculated and plotted as a function of residue number (Figure 2.22). Residues 75-100 and residue 120 showed the greatest chemical shift changes. No chemical shift differences could be identified for any amino acid residue in helix aF because paramagnetic broadening precluded sequential assignment of this region of the protein. Together, these observations suggest that a conformational change may occur between the Fe(II)-CO and Fe(II)-NO ligation states of the SO2144 H-NOX domain in the region surrounding the helix proximal to the heme cofactor, aF, (Figure 2.22), but they do not illuminate the structural nature of this change.
2.3.8 Preparation of a diamagnetic analog of the Fe(II)-NO ligation state: H103G mutant
The paramagnetic heme cofactor of the Fe(II)-NO WT SO2144 H-NOX domain prevented the complete assignment of amino acids required for structure determination. We anticipated that the H103G mutant would structurally mimic the Fe(II)-NO state regardless of the bound diatomic ligand (Figure 2.23). The H103G SO2144 mutant protein was generated by site-directed mutagenesis and purified from E. coli via a previously described protocol (3, 13). The presence of imidazole in the protein buffer was required at all times to maintain heme binding (14). For NMR spectroscopy E. coli cells, transformed with a plasmid encoding the SO2144 H103G mutant, were grown on minimal media M9 enriched in 15N or 15N and 13C. The protein was purified to >95% homogeneity as verified by SDS-PAGE (Figure 2.24). By SDS-PAGE, the overexpressed protein has a molecular weight of ~20 kDa. Mass spectrometry of non-isotope labeled protein was utilized to verify that recombinant expression of SO2144 in E. coli produced the desired protein product. The molecular mass of SO2144 measured by electrospray ionization mass spectrometry (ESI-MS), 20430.5 Da, agreed with the theoretical molecular mass, 20429.8 Da, calculated from the amino acid sequence (Figure 2.25) to within ±1 Da. Therefore, recombinant expression of the H103G SO2144 mutant in E. coli yields the desired protein product for structural analysis by NMR spectroscopy.
2.3.9 Assignments of chemical shifts in H103G SO2144 H-NOX domain NMR spectra in the Fe(II)-CO state
The H103G SO2144 H-NOX domain mutant was prepared in the diamagnetic Fe(II)-CO ligation state (Fe(II)-CO H103G) and analyzed by UV-Vis (Figure 2.26). The UV-Vis spectrum of Fe(II)-CO H103G is identical to that of the wild type (Fe(II)-CO WT) protein, indicating a hexacoordinate heme with CO and imidazole as the two axial ligands. By 1H-15N HSQC spectroscopy (Figure 2.27), the protein gave an excellent spectrum as evidenced by the homogeneous cross peak line widths and the appearance of ~180 individual amide cross-peaks. When the 1H-15N HSQC spectra from Fe(II)-CO H103G are superimposed onto the spectra of Fe(II)-CO WT (Figure 2.28) a large number of chemical shift changes are observed, indicating the conformation of the protein is significantly different between these two states. The NMR analysis of Fe(II)-CO H103G progressed in a very similar manner to that of the wild type protein. 1H, 15N, and 13C resonance assignments for the Fe(II)-CO H103G SO2144 H-NOX domain were made using standard 2D and 3D techniques (Appendix 6) (44, 56). Many of the chemical shift assignments for Fe(II)-CO H103G were similar in magnitude to that of wild type. However, chemical shift differences between Fe(II)-CO WT and Fe(II)-CO H103G were seen for residues within and in the vicinity of helix aF (Figure 2.29). Heme chemical shift assignments were made through the use of NOESY experiments. The protein backbone, protein side chain, and heme chemical shift assignments provided a foundation for the interpretation of NMR spectra that report on the geometry of the SO2144 H-NOX domain (BMRB Code: 16278).
2.3.10 Structural analysis of the SO2144 H-NOX domain in the Fe(II)-CO ligation state
The solution structure of Fe(II)-CO H103G was solved using a combination of 2D, 3D 15N-edited, and 3D 13C-edited NOESY experiments similar to Fe(II)-CO WT. NOESY cross-peaks were assigned with a semi-automated protocol consisting of iterative rounds of manual and automated NOESY assignment via the CYANA automated NOESY analysis software package (49) and CARA (17). Following the completion of this procedure 2222 total, heme-protein (48), protein-protein (2172), and imidazole-protein (2), NOE cross-peaks were unambiguously assigned. Two NOEs to a unique proton resonance (assigned to the HD1 atom of heme bound imidazole) indicate that imidazole binds to the heme on the proximal face, the same side as helix αF (Figure 2.30). Following assignment, the volumes of NOE cross-peaks were obtained by integration and distance restraints were calculated from the corresponding peak intensities using CYANA (49). At the end of the NOE assignment procedure 148 hydrogen bond restraints, between backbone amides and carbonyls in elements of regular helical and b-sheet secondary structure, were identified by NOE analysis (44). Protein backbone dihedral angle information was provided from HN, HA, CA, CB chemical values by the chemical shift index method, TALOS (18). In total, 250 φ and ψ backbone dihedral angle restraints were obtained through this protocol. These three sets of distance and dihedral angle restraints were then used in structure calculations with the XPLOR-NIH NMR structure calculation software package (Appendix 7) (50). Similar to Fe(II)-CO WT, pf1 phage was utilized to acquire three sets of RDCs as described in section 2.2 Materials and Methods for Fe(II)-CO H103G (Appendix 8). The following numbers of RDCs were measured and tabulated for Fe(II)-CO H103G: 1DNH (114), 1DHaCa (114), and heme methine 1H-13C RDCs (3). The three step simulated annealing protocol outlined in section 2.2 Materials and Methods was utilized to calculate 200 structures consistent with all the distance, dihedral angle, and residual dipolar coupling geometric restraints. The three step XPLOR-NIH protocol utilized for Fe(II)-CO H103G structure calculations is provided in Appendix 9. The 20 lowest energy structures were selected as described in Material and Methods: Structure Calculations (PDB Code: 2kil). A detailed description of the results of these structure calculations is provided in chapter 3.
NMR spectroscopy is a powerful method to interrogate the structure and dynamics of proteins in solution. In this chapter, the results of structural studies of the SO2144 H-NOX domain by NMR spectroscopy leading to the identification of geometric restraints for structure calculation are reported for one state with an intact axial histidine iron bond, Fe(II)-CO WT, and one state without a protein derived heme axial ligand, Fe(II)-CO H103G. Comparison of chemical shifts shows that the SO2144 H-NOX domain undergoes a conformational change following the release of the proximal histidine ligand from the heme iron. NMR data that reports on the structural properties of the SO2144 H-NOX domain were then obtained and analyzed. Two NMR structures of the SO2144 H-NOX domain were then calculated. In chapter 3, the results of these structure calculations are described.
1. Chou JJ & Bax A (2001) Protein side-chain rotamers from dipolar couplings in a liquid crystalline phase. Journal of the American Chemical Society 123:3844-3845.
2. de Alba E & Tjandra N (2006) Interference between cross-correlated relaxation and the measurement of scalar and dipolar couplings by Quantitative J. J Biomol NMR 35:1-16.
3. Price MS, Chao LY, & Marletta MA (2007) Shewanella oneidensis MR-1 H-NOX regulation of a histidine kinase by nitric oxide. Biochemistry 46:13677-13683.
4. Poulos TL (2006) Soluble guanylate cyclase. Curr Opin Struct Biol 16:736-743.
5. Ignarro LJ, Wood KS, & Wolin MS (1982) Activation of purified soluble guanylate cyclase by protoporphyrin IX. Proc Natl Acad Sci U S A 79:2870-2873.
6. Stone JR & Marletta MA (1994) Soluble guanylate cyclase from bovine lung: activation with nitric oxide and carbon monoxide and spectral characterization of the ferrous and ferric states. Biochemistry 33:5636-5640.
7. Cary SP, Winger JA, & Marletta MA (2005) Tonic and acute nitric oxide signaling through soluble guanylate cyclase is mediated by nonheme nitric oxide, ATP, and GTP. Proc Natl Acad Sci U S A 102:13064-13069.
8. La Mar GN & Walker FA (1979) The Porphyrins, ed Dolphin D (Academic Press, New York), Vol IV, pp 61-157.
9. Praneeth VK, Nather C, Peters G, & Lehnert N (2006) Spectroscopic properties and electronic structure of five- and six-coordinate iron(II) porphyrin NO complexes: Effect of the axial N-donor ligand. Inorg Chem 45:2795-2811.
10. Bertini I, Luchinat C, Parigi G, & Pierattelli R (2005) NMR spectroscopy of paramagnetic metalloproteins. Chembiochem 6:1536-1549.
11. Lukin JA & Ho C (2003) Nuclear magnetic resonance of hemoglobins. Methods Mol Med 82:251-269.
12. Alontaga AY, Bunce RA, Wilks A, & Rivera M (2006) 13C NMR spectroscopy of core heme carbons as a simple tool to elucidate the coordination state of ferric high-spin heme proteins. Inorg Chem 45:8876-8881.
13. Karow DS, et al. (2004) Spectroscopic characterization of the soluble guanylate cyclase-like heme domains from Vibrio cholerae and Thermoanaerobacter tengcongensis. Biochemistry 43:10203-10211.
14. Zhao Y, Schelvis JP, Babcock GT, & Marletta MA (1998) Identification of histidine 105 in the beta1 subunit of soluble guanylate cyclase as the heme proximal ligand. Biochemistry 37:4502-4509.
15. Hansen MR, Mueller L, & Pardi A (1998) Tunable alignment of macromolecules by filamentous phage yields dipolar coupling interactions. Nat Struct Biol 5:1065-1074.
16. Delaglio F, et al. (1995) Nmrpipe – a multidimensional spectral processing system based on Unix pipes. Journal of Biomolecular Nmr 6:277-293.
17. Keller R (2004) Optimizing the process of nuclear magnetic resonance spectrum analysis and computer aided resonance assignment. (Swiss Federal Institute of Technology Zurich).
18. Cornilescu G, Delaglio F, & Bax A (1999) Protein backbone angle restraints from searching a database for chemical shift and sequence homology. J Biomol NMR 13:289-302.
19. Losonczi JA, Andrec M, Fischer MW, & Prestegard JH (1999) Order matrix analysis of residual dipolar couplings using singular value decomposition. J Magn Reson 138:334-342.
20. Wei Y & Werner MH (2006) iDC: A comprehensive toolkit for the analysis of residual dipolar couplings for macromolecular structure determination. J Biomol NMR 35:17-25.
21. Mori S, Abeygunawardana C, Johnson MO, & Vanzijl PCM (1995) Improved sensitivity of HSQC spectra of exchanging protons at short interscan delays using a new fast HSQC (Fhsqc) detection scheme that avoids water saturation. J Magn Reson Series B 108:94-98.
22. Sklenar V, Piotto M, Leppik R, & Saudek V (1993) Gradient-tailored water suppression for 1H-15N HSQC experiments optimized to retain full sensitivity. J Magn Reson Series A 102:241-245.
23. Grzesiek S & Bax A (1992) An efficient experiment for sequential backbone assignment of medium-sized isotopically enriched proteins. J Magn Reson 99:201-207.
24. Kay LE, Xu GY, & Yamazaki T (1994) Enhanced-sensitivity triple-resonance spectroscopy with minimal H2O saturation. J Magn Reson Series A 109:129-133.
25. Stonehouse J, Clowes RT, Shaw GL, Keeler J, & Laue ED (1995) Minimization of sensitivity losses due to the use of gradient pulses in triple-resonance NMR of proteins. Journal of Biomolecular Nmr 5:226-232.
26. Wittekind M & Mueller L (1993) Hncacb, a high-sensitivity 3D NMR experiment to correlate amide-proton and nitrogen resonances with the alpha-carbon and beta-carbon resonances in proteins. J Magn Reson Series B 101:201-205.
27. Grzesiek S & Bax A (1992) Correlating backbone amide and side-chain resonances in larger proteins by multiple relayed triple resonance NMR. Journal of the American Chemical Society 114:6291-6293.
28. Kuboniwa H, Grzesiek S, Delaglio F, & Bax A (1994) Measurement of HN-H alpha J couplings in calcium-free calmodulin using new 2D and 3D water-flip-back methods. Journal of Biomolecular Nmr 4:871-878.
29. Grzesiek S & Bax A (1993) Amino acid type determination in the sequential assignment procedure of uniformly 13C/15N-enriched proteins. J Biomol NMR 3:185-204.
30. Clubb RT, Thanabal V, & Wagner G (1992) A constant-time 3-dimensional triple-resonance pulse scheme to correlate intraresidue 1H(N), 15N, and 13C(‘) chemical-shifts in 15N-13C-labeled proteins. J Magn Reson 97:213-217.
31. Grzesiek S, Anglister J, & Bax A (1993) Correlation of backbone amide and aliphatic side-chain resonances in 13C/15N-enriched proteins by isotropic mixing of 13C magnetization. J Magn Reson Ser B 101:114-119.
32. Kay LE, Xu GY, Singer AU, Muhandiram DR, & Formankay JD (1993) A gradient-enhanced Hcch TOCSY experiment for recording side-chain 1H and 13C correlations in H2O samples of proteins. J Magn Reson Series B 101:333-337.
33. Davis AL, Keeler J, Laue ED, & Moskau D (1992) Experiments for recording pure-absorption heteronuclear correlation spectra using pulsed field gradients. J Magn Reson 98:207-216.
34. Zwahlen C, et al. (1997) Methods for measurement of intermolecular NOEs by multinuclear NMR spectroscopy: Application to a bacteriophage lambda N-peptide/boxB RNA complex. Journal of the American Chemical Society 119:6711-6721.
35. Jeener J, Meier BH, Bachmann P, & Ernst RR (1979) Investigation of exchange processes by 2-Dimensional NMR-spectroscopy. J Chem Phys 71:4546-4553.
36. Braunschweiler L & Ernst RR (1983) Coherence transfer by isotropic mixing – application to proton correlation spectroscopy. J Magn Reson 53:521-528.
37. Bax A & Davis DG (1985) Mlev-17-based two-dimensional homonuclear magnetization transfer spectroscopy. J Magn Reson 65:355-360.
38. Shaka AJ & Freeman R (1983) Simplification of NMR-spectra by filtration through multiple-quantum coherence. J Magn Reson 51:169-173.
39. Vuister GW & Bax A (1992) Resolution enhancement and spectral editing of uniformly 13C-enriched proteins by homonuclear broad-band 13C decoupling. J Magn Reson 98:428-435.
40. Driscoll PC, Clore GM, Marion D, Wingfield PT, & Gronenborn AM (1990) Complete resonance assignment for the polypeptide backbone of interleukin 1 beta using three-dimensional heteronuclear NMR spectroscopy. Biochemistry 29:3542-3556.
41. Takeuchi K, Ng E, Malia TJ, & Wagner G (2007) 1-13C amino acid selective labeling in a 2H15N background for NMR studies of large proteins. J Biomol NMR 38:89-98.
42. Neri D, Szyperski T, Otting G, Senn H, & Wuthrich K (1989) Stereospecific nuclear magnetic resonance assignments of the methyl groups of valine and leucine in the DNA-binding domain of the 434 repressor by biosynthetically directed fractional 13C labeling. Biochemistry 28:7510-7516.
43. Szyperski T, Neri D, Leiting B, Otting G, & Wuthrich K (1992) Support of 1H NMR assignments in proteins by biosynthetically directed fractional 13C-labeling. Journal of Biomolecular Nmr 2:323-334.
44. Wuthrich K (1986) in NMR of Protein and Nucleic Acids (John Wiley & Sons, New York) pp 162-175.
45. Bax A, Kontaxis G, & Tjandra N (2001) Dipolar couplings in macromolecular structure determination. Methods in enzymology 339:127-174.
46. Permi P, Rosevear PR, & Annila A (2000) A set of HNCO-based experiments for measurement of residual dipolar couplings in 15N, 13C, (2H)-labeled proteins. J Biomol NMR 17:43-54.
47. Ottiger M, Delaglio F, & Bax A (1998) Measurement of J and dipolar couplings from simplified two-dimensional NMR spectra. J Magn Reson 131:373-378.
48. Douglas SM, Chou JJ, & Shih WM (2007) DNA-nanotube-induced alignment of membrane proteins for NMR structure determination. Proc Natl Acad Sci U S A 104:6644-6648.
49. Herrmann T, Guntert P, & Wuthrich K (2002) Protein NMR structure determination with automated NOE assignment using the new software CANDID and the torsion angle dynamics algorithm DYANA. J Mol Biol 319:209-227.
50. Schwieters CD, Kuszewski JJ, Tjandra N, & Clore GM (2003) The Xplor-NIH NMR molecular structure determination package. J Magn Reson 160:65-73.
51. Nilges M, Clore GM, & Gronenborn AM (1988) Determination of three-dimensional structures of proteins from interproton distance data by dynamical simulated annealing from a random array of atoms. Circumventing problems associated with folding. FEBS Lett 239:129-136.
52. Chou JJ, Li S, & Bax A (2000) Study of conformational rearrangement and refinement of structural homology models by the use of heteronuclear dipolar couplings. J Biomol NMR 18:217-227.
53. Pellicena P, Karow DS, Boon EM, Marletta MA, & Kuriyan J (2004) Crystal structure of an oxygen-binding heme domain related to soluble guanylate cyclases. Proc Natl Acad Sci U S A 101:12854-12859.
54. Ma X, Sayed N, Beuve A, & van den Akker F (2007) NO and CO differentially activate soluble guanylyl cyclase via a heme pivot-bend mechanism. EMBO J 26:578-588.
55. Bhattacharya A, Tejero R, & Montelione GT (2007) Evaluating protein structures determined by structural genomics consortia. Proteins 66:778-795.
56. Ferentz AE & Wagner G (2000) NMR spectroscopy: a multifaceted approach to macromolecular structure. Q Rev Biophys 33:29-65.