• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of embojLink to Publisher's site
EMBO J. Jan 24, 2007; 26(2): 578–588.
Published online Jan 11, 2007. doi:  10.1038/sj.emboj.7601521
PMCID: PMC1783457

NO and CO differentially activate soluble guanylyl cyclase via a heme pivot-bend mechanism

Abstract

Diatomic ligand discrimination by soluble guanylyl cyclase (sGC) is paramount to cardiovascular homeostasis and neuronal signaling. Nitric oxide (NO) stimulates sGC activity 200-fold compared with only four-fold by carbon monoxide (CO). The molecular details of ligand discrimination and differential response to NO and CO are not well understood. These ligands are sensed by the heme domain of sGC, which belongs to the heme nitric oxide oxygen (H-NOX) domain family, also evolutionarily conserved in prokaryotes. Here we report crystal structures of the free, NO-bound, and CO-bound H-NOX domains of a cyanobacterial homolog. These structures and complementary mutational analysis in sGC reveal a molecular ruler mechanism that allows sGC to favor NO over CO while excluding oxygen, concomitant to signaling that exploits differential heme pivoting and heme bending. The heme thereby serves as a flexing wedge, allowing the N-terminal subdomain of H-NOX to shift concurrent with the transition of the six- to five-coordinated NO-bound state upon sGC activation. This transition can be modulated by mutations at sGC residues 74 and 145 and corresponding residues in the cyanobacterial H-NOX homolog.

Keywords: carbon monoxide, guanylyl cyclase, heme sensors, nitric oxide, signal transduction

Introduction

Sensing gaseous molecules such as nitric oxide (NO), carbon monoxide (CO), and oxygen is a key trait present in virtually all life forms and predominantly mediated by heme-based sensors (Rodgers, 1999). In animals, differential sensing of NO, CO, and oxygen is critical for a wide range of physiological processes such as vascular homeostasis, platelet aggregation, host defense, neuronal signaling, and stress responses (Waldman and Murad, 1987; MacMicking et al, 1997; Ignarro, 2002; Lahiri et al, 2006; Moncada and Higgs, 2006; Ryter et al, 2006). Bacteria need this gas sensing capability for, for example, aerotaxis, to combat the host immune response (Roberts et al, 2004; D'autreaux et al, 2005), to sense CO as the sole energy source (Ascenzi et al, 2004), or to induce certain genes under anaerobic conditions (Bailey-Serres and Chang, 2005).

In animals, the major NO sensor is the soluble guanylyl cyclase receptor (sGC), which, when activated by NO, produces the second messenger cGMP (Garbers et al, 1994; Koesling et al, 2004; Padayatti et al, 2004; Cary et al, 2006). In addition to NO binding while excluding oxygen, there is now substantial evidence that CO can also regulate sGC, in particular in olfactory neurons in which heme oxygenase produces CO (Boehning and Snyder, 2003; Ryter et al, 2006). The NO–cGMP signaling pathway is involved in cardiovascular and neurological processes, and recent efforts lead to the development of sGC activators for potentially treating hypertension, erectile dysfunction, congestive heart failure, and chronic renal disease (Behrends, 2003; Evgenov et al, 2006). Despite this strong pharmaceutical interest, the molecular details of sGC activation are not well understood.

sGC's NO/CO sensing and catalytic abilities are contained within a single heterodimeric protein comprised of an α and β subunit. Four isoforms of these subunits are found, with α1/β1 being the most ubiquitous heterodimer (Koesling et al, 2004). The sGC subunits have three distinct regions: sGCβ1 containing an N-terminal NO sensing heme domain, a central dimerization region, and a C-terminal catalytic guanylyl cyclase domain (Figure 1A) (Russwurm and Koesling, 2004a; Pyriochou and Papapetropoulos, 2005). sGCα1 has a similar subunit arrangement, except its N-terminal domain does not contain a heme. The NO sensing domain belongs to the heme nitric oxide oxygen binding (H-NOX) domain family, which also includes bacterial members, some with ligand binding characteristics similar to sGC (Boon et al, 2006).

Figure 1
Structure and domain organization of H-NOX and H-NOXA-containing proteins. (A) Schematic diagram of the H-NOX and H-NOXA domains present in proteins from animals and Nostoc cyanobacteria. The sGC subunits contain additional coiled-coil (CC) and GC domains. ...

NO binds to the heme of sGC and forms a transient six-coordinated NO-bound state that progresses, upon heme-His bond breakage, to a five-coordinated NO-bound activated state (Pyriochou and Papapetropoulos, 2005; Cary et al, 2006). Whether this final step involves a second NO molecule and whether the transient six-coordinated NO-bound state is physiologically important as a partially active tonic signaling state is contentious (Cary et al, 2006; Roy and Garthwaite, 2006). The conversion of the six- to five-coordinated state of sGC can be positively regulated by the substrate GTP, whereas it is inhibited by ATP (Ruiz-Stewart et al, 2004; Russwurm and Koesling, 2004b; Cary et al, 2005). Similarly, NO stimulation of sGC can be sensitized by YC-1 (Koesling et al, 2005; Cary et al, 2006). These molecules are thought to bind to the cyclase domain, suggesting that there is crosstalk between both ends of the sGC protein (Lamothe et al, 2004; Chang et al, 2005; Yazawa et al, 2006).

In contrast to NO, CO only reaches the six-coordinated CO-bound state and has a much lower affinity and weak stimulatory properties for sGC compared with NO (Stone and Marletta, 1994; Martin et al, 2006). As the activity of CO-bound sGC is also drastically enhanced by YC-1 to NO activated levels (Friebe and Koesling, 1998; Stone and Marletta, 1998), it is speculated that one or more endogenous compounds mimicking YC-1 might exist to account for this discrepancy in CO-mediated sGC signaling/activation (Boehning and Snyder, 2003). The synergistically CO/YC-1-activated sGC is predominantly in the six-coordinated state (Friebe and Koesling, 1998; Stone and Marletta, 1998).

The differential regulation of sGC by NO and CO in terms of affinity, activity, and coordination states demonstrates an intriguing complexity of sGC. Although some insights have been gained from structures of the adenylyl cyclase catalytic domain (~30% sequence identical) (Tesmer et al, 1997; Zhang et al, 1997; Tews et al, 2005) and oxygen-bound H-NOX domain (18% sequence identical) from Thermoanaerobacter tengcongensis (Tt H-NOX) (Nioche et al, 2004; Pellicena et al, 2004), the molecular details of NO/CO binding to H-NOX and sGC activation are poorly understood. H-NOX domains are observed in a number of proteins found in bacteria and animals as either stand-alone proteins or fused to other domains (Figures 1A and B) (Iyer et al, 2003). Remarkably, the stand-alone H-NOX domains in Nostoc cyanobacteria are adjacent to genes containing domains that are homologous to the central dimerization of sGC (Ohmori et al, 2001), including a putative PAS domain (Iyer et al, 2003) and a coiled-coil region (Figure 1A). It was therefore suggested that the stand-alone H-NOX domains could work in concert with these proteins transcribed from the adjacent genes (Iyer et al, 2003). The H-NOX domain of Nostoc punctiforme (Np H-NOX) has similar ligand binding properties as sGC (Boon et al, 2006), and the homologous H-NOX domains from Nostoc cyanobacteria are therefore good models for sGC. Furthermore, phylogenetic analysis even suggested that the bacterial H-NOX domains were probably transferred to animals, resulting in the mammalian NO-responsive sGC (Iyer et al, 2003). These bacterial H-NOX domains were therefore speculated to act as sensors for gaseous molecules such as NO (or CO) (Iyer et al, 2003), probably as part of a bacterial chemotaxis system owing to H-NOX's close proximity to genes involved in the two-component system (Figure 1A). This could perhaps allow these cyanobacteria to possibly respond to the presence of (toxic) NO, which can be produced by for example other bacteria and plants.

To gain molecular insights into the NO and CO activation mechanism of sGC, we targeted the homologous H-NOX domain of Nostoc sp (Ns H-NOX) for crystallographic studies, as the sGC H-NOX domain has been refractory to crystallization. Here we describe the crystal structures of the free, NO, and CO liganded Ns H-NOX domains at 2.1, 2.6, and 2.5 Å resolution, respectively. Our structures, together with structure-guided mutational and biophysical characterization, provide evidence that the NO activation mechanism in sGC involves a stepwise heme pivot to facilitate an ~20° rotational shift of the N-terminal helical region in H-NOX, causing the heme to bend in the activated five-coordinated state. CO signals differ from NO by having a larger heme pivot shift without proceeding to the heme-bend five-coordinated state.

Results and discussion

Structure of Ns H-NOX heme domain

We determined the structure of Ns H-NOX heme domain at 2.1 Å resolution using single wavelength anomalous diffraction (SAD) (Table I). There are two Ns H-NOX molecules in the asymmetric unit and each of the two similar molecules (r.m.s.d. of 0.40 Å for 182 Cα atoms between molecules A and B) contains one heme moiety sandwiched between a small α-helical subdomain at its distal side and a larger mixed-α/β subdomain at its proximal side (Figure 2A). The proximal side of Ns H-NOX provides the conserved histidine (H105), which is the fifth ligand for a five-coordinated heme iron (NE2-Fe distance is 2.2 Å in both molecules). The Ns H-NOX absorption spectra is consistent with a five-coordinated iron (Figure 2B, left-hand panel), and is very similar to the spectrum of free sGC yet different from that of Tt H-NOX, as the latter binds oxygen under aerobic conditions (Karow et al, 2004; Pellicena et al, 2004). Correspondingly, Ns H-NOX shares a much higher sequence identity with sGC (33% identity) than Tt H-NOX with sGC (18% identity) (Figure 1B). This similarity extends to the heme pocket, in which 17 of 27 residues are identical between Ns H-NOX and sGC, compared with only eight of 27 between Tt H-NOX and sGC. The heme interactions are predominantly hydrophobic, except for the proximal heme ligand H105, and interactions with both propionate groups of the heme involving the main-chain nitrogen of Y2 and the conserved Y134-X-S136-X-R138 sequence (Figure 2C). The latter sequence was found to be key for heme binding in sGC (Schmidt et al, 2004, 2005) and Tt H-NOX (Nioche et al, 2004; Pellicena et al, 2004). In addition, the Ns H-NOX structure reveals an aromatic residue, W74 (F74 in sGC), oriented perpendicular to the heme at its distal face with the W74 NE1 atom located 3.3 Å from the heme NB atom (Figure 2C). Such an aromatic residue is absent in Tt H-NOX, which contains an N at this position. A possible explanation for Ns H-NOX's lack of oxygen binding is that it does not have Y140 found in Tt H-NOX (M in Ns H-NOX), which interacts with the bound oxygen (Nioche et al, 2004; Pellicena et al, 2004). Oxygen binding to heme model compounds, such as Fe(II)PP(1-MeIm), is much weaker compared with NO and CO binding (Martin et al, 2006), and heme proteins therefore require additional polar/electrostatic interactions for oxygen to bind (Jain and Chan, 2003). The absence of such polar interactions provides an explanation for Ns H-NOX's and sGC's lack of oxygen binding.

Figure 2
Crystal structure of unliganded Ns H-NOX. (A) Schematic diagram of the structure of Ns H-NOX. The three N-terminal helices αA–αC (red), the heme (blue), and H105 (green) are highlighted. The heme propionate groups attached to pyrrole ...
Table 1
Data collection, phasing, and refinement statistics for Ns H-NOX

Although similar in fold, a striking difference between Tt and Ns H-NOX is the relative orientation of the N-terminal subdomain, in particular helices αA–αC although some of the other nearby helices have shifted somewhat as well. Excluding αA–αC from the superposition lowers the r.m.s.d. from 2.5 to 1.6 Å and reveals an 18° difference in the orientation of this N-terminal helical region (Figure 2D). This N-terminal shift in oxygen-bound Tt H-NOX molecule A causes a strong heme distortion, as indicated by the saddling and ruffling distortions, both around −1 Å (Pellicena et al, 2004). The heme in Ns H-NOX is considerably less bent, with only moderate ruffling (~−0.35 Å) and dome (~0.5 Å) distortions. One of the four refined oxygen-bound Tt H-NOX molecules (monoclinic molecule B), however, contains an intermediate N-terminal shift of 11° with respect to Ns H-NOX, resulting in a moderate heme distortion (Figure 2D) (Pellicena et al, 2004). The H-NOX domain can thus adopt a range of N-terminal subdomain shifts and concomitant heme distortions.

NO and CO binding to H-NOX

In order to investigate the structural effects of NO and CO binding, we measured electronic absorption spectra of purified Ns H-NOX. Our results indicate that the NO:Ns H-NOX complex is predominantly a six-coordinated NO-bound heme species, with a small amount of five-coordinated NO-bound heme, as indicated by the small shoulder at 400 nm near the main 416 nm peak (Figure 2B, left-hand panel). This is similar to previous spectroscopic results for Np H-NOX (~60% sequence identical to Ns H-NOX) (Boon et al, 2006). As expected, CO binding leads to a six-coordinated state of Ns H-NOX (Figure 2B, left panel). In order to obtain NO/CO complex structures, we soaked Ns H-NOX crystals in solutions containing NO or CO. The resulting crystal structures of Ns H-NOX at 2.6 and 2.5 Å resolution, respectively, show strong difference density peaks corresponding to NO or CO in both molecules positioned at the distal face of the heme (Figures 3A and B). Both ligands are situated in a hydrophobic pocket, lined by V5, F70, W74, M144, and L148. In NO-bound H-NOX, the nitrogen atom of NO forms a 1.8 Å bond with Fe and the Fe–N–O angle is 143 and 157° for molecules A and B, respectively. These angles are more perpendicular (163 and 168°) for the CO-bound complex. These observed angle differences conform to the known angular preferences for NO and CO heme interactions (Kachalova et al, 1999; Carlsson et al, 2005; Copeland et al, 2006).

Figure 3
Structural consequences of NO and CO binding to Ns H-NOX. (A) NO binding to the distal face of the heme in the NO-bound Ns H-NOX structure. Omit [mid ]Fo[mid ][mid ]Fc[mid ] electron density (contoured blue at 5σ) and 2[mid ] ...

Conformational changes upon NO and CO binding to H-NOX

Unlike the five-coordinated NO-bound state, the six-coordinated state of sGC has low activity, and major conformational changes, compared with unbound H-NOX, are therefore not expected. Nevertheless, superposition of the free, NO-bound, and CO-bound Ns H-NOX structures reveals a striking heme pivot Fe shift of 0.2–0.3 Å for NO and a larger 0.8–0.9 Å Fe shift upon CO binding observed in both independently refined molecules in the asymmetric unit (Figures 3C and and4A).4A). The heme pivots while keeping the propionate group attached to pyrrole ring D (PD) in place serve as the pivot point. Shifts over 1 Å are observed for heme atoms farthest away from the pivot point in the CO-bound structure. The heme Fe-center pivots away from W74, which appears to be necessary to provide the steric space for NO and CO to bind to the distal face of the heme (the distance between NO/CO and the side chain of W74 is 3.0–3.4 Å). CO's larger shift may be due to the larger atomic radius of its carbon atom and its preference to bind perpendicular to the plane of the heme (Kachalova et al, 1999) compared with NO binding at an angle. These larger changes to accommodate CO suggest that the perpendicularly oriented aromatic residue at position 74 close to the Fe serves as a molecular ruler; its presence could explain why sGC and the model heme compound bind NO with similar affinity, whereas CO binds to sGC ~5 orders of magnitude more weakly than the model heme compound (Martin et al, 2006). Steric hindrance, and the relieve thereof, is also used by myoglobin to decrease its affinity for CO (Kachalova et al, 1999). The heme shift is accompanied by a shift of the H105 side chain towards the heme (Figure 3C). The NO-bound H-NOX structure yields minimal additional changes, yet the larger CO-induced heme shift causes several structural changes in Ns H-NOX (Figures 3C–E). First, the αF residue H105 and its adjacent residues shift with the heme. Second, the αA region near Y2, whose main chain interacts with the heme's PA propionate, shifts with the heme by about 0.5 Å. Third, the flanking αB–αC loop near residue E41 shifts about the same amount in concert with αA. Fourth, the interacting αF–β1 loop region near F112 and C-terminal part of helix αF becomes substantially disordered, displaying poor electron density and increasing its refined temperature factors from ~45 Å2 in the unbound structure, to ~75 Å2 in the CO-bound structure (Figure 3D). This increased flexibility is caused by heme shifting away, leading to decreased van der Waals interactions between the heme and the side chains of F112 and V108. These shifts and increased flexibility are observed in both independently refined H-NOX molecules in the asymmetric unit. The CO-induced shifts involving helix αA and loop αB–αC are in the same direction as the shifts in the oxygen-bound Tt H-NOX structure (Figure 2D), suggesting a conserved directional path of movement for this αA–αC region. These CO-induced structural changes likely provide the molecular basis for the four-fold activation of sGC by CO, yet the even smaller induced changes by NO potentially question the contentious, tonic signaling importance (Cary et al, 2006; Roy and Garthwaite, 2006) of the transient six-coordinated NO-bound intermediate before Fe–His bond breakage.

Figure 4
Evidence for heme pivoting and heme bending in the activation mechanism of H-NOX. (A) Top (distal) view of the heme groups of the superimposed H-NOX structures. Ns H-NOX residues M144 and W74 are shown, as well as the residues interacting with the propionate ...

Heme pivoting and heme-bending activation mechanism

Our Ns H-NOX structures, together with the Tt H-NOX structure (Nioche et al, 2004; Pellicena et al, 2004), likely represent the entire spectrum of activation: The oxygen-bound Tt H-NOX structure is thought to represent the activated state, whereas our new free and NO/CO-bound Ns H-NOX structures represent the basal and low-activity H-NOX state, respectively. During activation, the N-terminal αA helix may progressively push against the heme, causing a drastic heme bend in the final step; αA shifts range from about 0.5 Å for the CO-activated state to over 4 Å for oxygen-bound Tt H-NOX (Figures 2D, ,3C,3C, 4As and B; see Supplementary Video 1a for the large-scale subdomain movement and Supplementary Video 1b for a close-up view of heme pivoting and bending). The superposition reveals that the heme and heme iron follow a trajectory in the order of presumed increasing cyclase activity (Figure 4A and Supplementary Video 1b). In addition to this heme pivot, we speculate that two heme bending motions are key for the final activation step: the heme bending at the side of the pyrrole A ring with attached propionate PA provides the space to accommodate the large shift of the N-terminal αA–αC helices, whereas the pyrrole D ring attached to the propionate PD group tilts in the opposite (distal) direction towards M144 (Figure 4B and Supplementary Video 1b). Although structurally conceivable as a signaling mechanism for H-NOX domains in general, Ns H-NOX stays predominantly in the six-coordinated NO-bound state (Figures 2B and and3A)3A) and might perhaps be more sensitized for this transition once it interacts with its target output domain. Our activation mechanism offers a structural explanation for Ns H-NOX's limited ability to reach the acute state: the helical shift of αA would require, in addition to the heme bending, space in the direction of residue W74 as the αA residue A8 is postulated to shift about 3–4 Å towards W74 (Figure 4B). Owing to the size of W74 (which is the smaller F74 in rat sGCβ1) and the close 3.7 Å van der Waals interaction between the side chains of A8 and W74, the αA helix has limited space for this large shift in Ns H-NOX. In addition, the tilting of the pyrrole D ring with attached propionate PD of the heme involves a 0.7 Å shift of this propionate group when comparing the unliganded Ns H-NOX to the activated oxygen-bound Tt H-NOX structure (Figure 4B). However, the CE atom of M144 forms a snug 3.4 Å van der Waals interaction with the CAD atom of this propionate PD group in the unliganded Ns H-NOX structure (Figure 4B), thereby causing steric hindrance for such a heme bending at this side as well.

Mutagenesis studies to test H-NOX activation mechanism

To test this hypothesis about the importance of heme bending and N-subdomain movement during activation, we substituted Ns H-NOX residues W74 and M144 with the smaller amino acids found in rat sGCβ1 (F and I, respectively). The smaller residues at these positions are speculated to more easily facilitate the N-terminal subdomain shift and heme bending. The W74F and M144I mutations in Ns H-NOX both resulted in considerable increases in the soret peak near 400 nm, representing increased levels of the five-coordinated NO-bound state (Figure 2B). These results are in agreement with our activation model, indicating that reaching the activated five-coordinated NO-bound state is directly correlated with heme bending and the N-terminal subdomain shift. To further validate this model, we carried out the reverse mutagenesis experiment by introducing the larger Ns H-NOX residues into rat sGCβ1 (F74W and I145M mutations) with the intent of restricting the heme bending and N-terminal subdomain shift. These mutations caused a significant loss of NO-stimulated guanylyl cyclase activity: from 21-fold stimulation in wild-type (wt) sGC to only two- and five-fold stimulation for the F74W and I145M sGC mutants, respectively (Table II and Figure 4C). This observed drop is likely owing to the increased size of these side chains, thereby restricting the N-terminal subdomain shift and heme bending. In addition to this experimental validation, our activation model also provides a new explanation for the decreased activity of the previously characterized sGC mutants, where larger residues were introduced at the 74 and 145 positions. First, the sGC mutation I145Y exhibits decreased activity and diminished levels of the five-coordinated NO-bound state (Martin et al, 2006). Second, the F74Y mutation was also observed to cause a decrease in NO-stimulated activity (Rothkegel et al, 2006). Taken together, these results suggest that the two sGC activation events, reaching the five-coordinated activated NO-bound state and the N-terminal helical subdomain movement concomitant with heme bending, are strongly correlated. These latter events are preceded by the ligand-induced heme pivoting event described earlier and together comprise a new activation model for sGC (Figure 5). A bent heme as a key feature of the activated state in sGC also provides an explanation for the ability of the elongated and flexible heme mimetic BAY58-2667 (Schmidt et al, 2004) to activate heme-free sGC. The strong heme bending motion in H-NOX is unprecedented and is to our knowledge not observed in other heme proteins (Jain and Chan, 2003). The presence of an energetically strained heme in the activated state might, in part, be key for the ability of sGC to be rapidly deactivated (Margulis and Sitaramayya, 2000) upon relaxing the heme. This could serve a beneficial physiological purpose to rapidly control blood pressure, neuronal signaling, and other processes.

Figure 5
Schematic diagram of NO- and CO-dependent activation of sGC. NO and CO bind to the distal face of the heme and cause the heme to pivot, with varying degrees, away from F74 (W74 in Ns H-NOX). The larger pivot for CO is likely a result of the need to accommodate ...
Table 2
GC activity measurements of wt and mutants of rat sGC under basal and NO-stimulated conditions

Communication of the H-NOX domain to transmit the NO/CO signal to the catalytic guanylyl cyclase domain likely involves a direct interaction (Winger and Marletta, 2005). Such domain–domain communication likely involves H-NOX residues that shift upon ligand binding (as discussed in Figures 2D, ,3C,3C, and and4B),4B), as the catalytic domain would need to sense NO-induced conformational differences. Our results therefore point to a potential role for residues in/near αF (containing H105) and αA–αC, which includes the region near residue D45 previously suggested to be a switch region (Pellicena et al, 2004; Rothkegel et al, 2006). Although speculative, a subset of these could perhaps interact on a similar site on the guanylyl cyclase domain, where Gsα modulates the adenylyl cyclase activity (Tesmer et al, 1997), and our insights provide a framework for future structure-based mechanistic studies.

In conclusion, we provide evidence for a molecular ruler mechanism in sGC favoring NO over CO binding, using steric hindrance involving an aromatic residue at position 74. Both ligands induce differential heme pivoting and heme bending, which appear to correlate with differences in sGC activation levels. CO induces a larger initial heme pivot shift compared with NO, whereas NO only proceeds to the five-coordinated fully activated state concomitant with a strongly bent heme and large N-terminal subdomain shift. These H-NOX heme pivoting/bending signaling insights for sGC activation offer new opportunities for sGC to be therapeutically exploited.

Materials and methods

Cloning and mutagenesis of Ns H-NOX

The genomic DNA of Nostoc sp PCC 7120 (Anabaena sp PCC 7120) was used for PCR of the 189-amino-acid H-NOX gene using the forward primer 5′-ggaattccatatgtatggtttagtgaacaaag cc-3′ and the backward primer 5′-ccggaattctcagtcgtcatataaattcgagt c-3′. The amplified fragment was inserted into the pET22b (Novagen) expression vector. The W74F and M144I mutants were introduced using the QuickChange kit (Stratagene) and were confirmed by DNA sequencing.

Expression and purification of Ns H-NOX

The Ns H-NOX protein was expressed in Escherichia coli BL21(DE3) Star cells (Invitrogen). After overnight 0.1 mM IPTG induction at 18°C, cells were pelleted and lysed in buffer A containing 20 mM Tris–HCl (pH 7.5), 100 mM NaCl, and 2 mM β-mercaptoethanol. The cleared cell lysate was loaded onto a Q-Sepharose FF 5 ml column and the H-NOX protein was eluted using an NaCl gradient. The 420 nm peak fractions containing the H-NOX protein were concentrated and loaded onto a Sephadex 200 column with buffer A as the running buffer. The fractions with OD420>OD280 were loaded onto a high-resolution MonoQ anion exchange column and an NaCl gradient was used for elution. Finally, the H-NOX protein was cleaned up using a high-resolution prepacked Sephadex75 column.

UV/Vis absorption spectroscopic analysis

A SHIMADZU UV-1700 UV–visible spectrophotometer was used for all spectral analysis performed at room temperature. The concentration of the wt and the mutant H-NOX proteins was 1 μM in a buffer containing 10 mM Tris–HCl (pH 7.5), 100 mM NaCl, and 1 mM β-mercaptoethanol. The NO complex was prepared by adding the NO donor S-nitroso-N-acetylpenicillamine (SNAP) to a final concentration of 200 μM. The CO complex was obtained by using a saturated CO solution that was prepared by bubbling CO gas (Praxair) through the above Tris buffer.

Ns H-NOX crystallization

Ns H-NOX crystals were grown at 20°C by the sitting-drop vapor diffusion method. The protein was concentrated to 15 mg/ml in 5 mM Tris (pH 7.5), 100 mM NaCl, and 1 mM β-mercaptoethanol, and was mixed with an equal volume of the reservoir solution containing 1.4 M tri-sodium citrate dihydrate (pH 5.6–6.6). Native Ns H-NOX crystals were prepared for data collection by fast transfer into cryo-protectant solution containing 1.36 M tri-sodium citrate dihydrate with 15% ethylene glycol. Selenomethionine-substituted protein crystals (produced similar to Van Duyne et al, 1993) were used for phasing. The F74W and M145I Ns H-NOX mutants unfortunately resisted crystallization.

Ns H-NOX structure determination and refinement

The native 2.1 Å resolution data set was collected at the Brookhaven NSLS synchrotron (beamline X29) and processed by HKL2000 (Otwinowski and Minor, 1997). For phasing, a SAD data set was collected from a selenomethionine-substituted crystal at the Advanced Light Source ALS (beamline 4.2.2.) and processed by D*trek (Pflugrath, 1999). SOLVE/RESOLVE (Terwilliger and Berendzen, 1999) was used to calculate initial phases and density modification, resulting in locating 10 selenium atoms yielding a figure of merit of 0.7 (Table I). A partial model of two molecules in the asymmetric unit generated by RESOLVE was used as a starting point for ARP/wARP (Perrakis et al, 2001) for additional automatic model building, using the 2.1 Å native data set. Iterative cycles of model building in Coot (Emsley and Cowtan, 2004) and refinement using REFMAC (Murshudov et al, 1997) yielded a model comprising H-NOX residues 1–182 for both molecules A and B, two heme moieties, and 230 water molecules (if above 3σ observed in omit FoFc map and kept if peak remained above 1σ in 2 [mid ]Fo[mid ][mid ]Fc[mid ] map) with a final R-factor of 18.1% and Rfree of 22.2% (see Table I for additional refinement statistics). One high-temperature factor water molecule was observed in the general vicinity of Fe in only molecule B, but was located at >4.5 Å from this Fe, weakly bound, and therefore was not included in the analysis.

The 2.6 Å NO:H-NOX complex data set was collected at the Advanced Photon Source APS (beamline 19ID) and processed by HKL2000 (Otwinowski and Minor, 1997). The NO complex crystals were prepared by soaking wt H-NOX crystals in the similar cryo-protectant solution with 20 mM of the NO donor SNAP (Cayman) while varying the soaking time from 1, 2, 4, 6, 8, 10 min to several hours before crystal freezing by dunking it in liquid nitrogen for X-ray data collection. Tens of data sets were collected and analyzed. We found the 2 min soaking time point to be the most structurally informative regarding resolution of diffraction and NO electron density. The 2.1 Å native H-NOX structure was used as a starting point for crystallographic refinement with REFMAC. Strong NO density in the vicinity of Fe at the distal side of the heme was observed (7.7σ peak for molecule A and 4.5σ for molecule B) (Figure 3A). One NO molecule for each heme was positioned in the electron density and further refined using REFMAC. The final refined model contained residues 1–182 for both molecules A and B, two heme molecules, two NO molecules, and 64 waters. CO-bound crystals were obtained in a similar manner. The cryo-protection solution was bubbled with CO (Praxair) for an extended period of time to achieve a saturated CO soaking solution. The soaking time was varied and the most optimal time for H-NOX crystals was 45 min, and a 2.5 Å resolution data set was collected at APS. Initial refinement revealed strong [mid ]Fo[mid ][mid ]Fc[mid ] difference density, representing CO at the distal side of the heme (6.3σ and 7.0σ for molecules A and B, respectively). Both CO molecules were refined at 100% occupancy. The final refined model contained residues 1–182 of both molecules, two heme molecules, two CO molecules, and 122 water molecules. Heme distortion calculations were calculated using normal-coordinate structural decomposition (Jentzen et al, 1998). Owing to the 2.6/2.5 Å resolution of the NO/CO H-NOX complex structures, the angles are likely only accurate to within about 10° of the refined values, yet these refined angles are within the range of the Fe–N/C–O angles observed for other NO and CO heme-protein complexes (Carlsson et al, 2005; Copeland et al, 2006). The program ESCET (Schneider, 2000) was used in the analysis of conformational changes among the different H-NOX structures.

Mutagenesis of rat sGC and transfection in COS-7 cells

Templates were cDNAs encoding the α1 and β1 subunits of rat sGC cloned into the mammalian expression vector pCMV5 (Yuen et al, 1994). The F74W and I145M mutations were introduced by PCR (Quickchange, Stratagene) and checked by DNA sequencing. COS-7 cells were grown in DMEM supplemented with 10% heat-inactivated fetal bovine serum (FBS), penicillin, and streptomycin (100 U/ml). Cells were transfected for 48 h with Superfect reagent, according to the supplier's protocol (Qiagen).

Cytosol preparation and Western blot analysis

COS-7 cells were washed twice with ice-cold PBS and then scraped off the plate in cold lysis buffer: PBS buffer contained protease inhibitors, 50 mM HEPES (pH 8.0), 1 mM EDTA, and 150 mM NaCl. Cells were broken by sonication (three pulses of 3 s). The resulting lysate was centrifuged at 16 000 g for 10 min at 4°C to collect the cytosol. To determine the expression levels for wt sGC or sGC mutants in COS-7 cells, 15 μg of cytosol was resolved on 10% SDS–PAGE and analyzed by immunoblotting with anti-sGC (anti-α1 subunit and anti-β1 subunit; Cayman Chemicals).

sGC activity assay

GC activity was determined by the formation of [α-32P]cGMP from [α-32P]GTP, as previously described (Lamothe et al, 2004). Reactions were performed for 5 min at 33°C in a final volume of 100 μl, in 50 mM HEPES (pH 8.0) reaction buffer containing 500 μM GTP, 1 mM DTT, and 5 mM MgCl2. Typically, 40 μg of COS-7 cytosol transfected with either wt or mutants was used in each assay reaction. All assays were performed in duplicate and each experiment was repeated twice. Enzymatic activity was stimulated with the NO donor SNAP (Calbiochem) at 100 μM. sGC activity is expressed in pmol/min mg and as mean±s.e.

Supplementary Material

Supplementary video 1a

Supplementary video 1b

Acknowledgments

We thank Dr T Kaneko (Kazusa DNA Research Institute, Japan) for the genomic DNA of Nostoc sp PCC 7120, Dr R Bonomo, Dr E Jankowski, Dr P Pattanaik, Dr J Qin, Dr M Snider, and Dr V Yee for helpful comments, Dr D Lodowski and Dr K Palczewski (CWRU) for their generosity in sharing beamtime, and KP Ng and J He for the initial H-NOX studies. We thank the beam line support personnel at NSLS, APS, and ALS. This work was supported in part by grants from the NIH to FvdA (R01 HL075329) and AB (R01 GM067640) and from the American Heart Association to FvdA. Coordinates and structure factors of the free, NO-bound, and CO-bound Ns H-NOX structures have been deposited with the PDB (PDB identifiers 2O09, 2O0C, and 2O0G, respectively).

References

  • Ascenzi P, Bocedi A, Leoni L, Visca P, Zennaro E, Milani M, Bolognesi M (2004) CO sniffing through heme-based sensor proteins. IUBMB Life 56: 309–315 [PubMed]
  • Bailey-Serres J, Chang R (2005) Sensing and signalling in response to oxygen deprivation in plants and other organisms. Ann Bot (London) 96: 507–518 [PubMed]
  • Behrends S (2003) Drugs that activate specific nitric oxide sensitive guanylyl cyclase isoforms independent of nitric oxide release. Curr Med Chem 10: 291–301 [PubMed]
  • Boehning D, Snyder SH (2003) Novel neural modulators. Annu Rev Neurosci 26: 105–131 [PubMed]
  • Boon EM, Davis JH, Tran R, Karow DS, Huang SH, Pan D, Miazgowicz MM, Mathies RA, Marletta MA (2006) Nitric oxide binding to prokaryotic homologs of the soluble guanylate cyclase beta 1 H-nox domain. J Biol Chem 281: 21892–21902 [PubMed]
  • Carlsson GH, Nicholls P, Svistunenko D, Berglund GI, Hajdu J (2005) Complexes of horseradish peroxidase with formate, acetate, and carbon monoxide. Biochemistry 44: 635–642 [PubMed]
  • Cary SP, Winger JA, Derbyshire ER, Marletta MA (2006) Nitric oxide signaling: no longer simply on or off. Trends Biochem Sci 31: 231–239 [PubMed]
  • 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 USA 102: 13064–13069 [PMC free article] [PubMed]
  • Chang FJ, Lemme S, Sun Q, Sunahara RK, Beuve A (2005) Nitric oxide-dependent allosteric inhibitory role of a second nucleotide binding site in soluble guanylyl cyclase. J Biol Chem 280: 11513–11519 [PubMed]
  • Copeland DM, Soares AS, West AH, Richter-Addo GB (2006) Crystal structures of the nitrite and nitric oxide complexes of horse heart myoglobin. J Inorg Biochem 100: 1413–1425 [PubMed]
  • D'autreaux B, Tucker NP, Dixon R, Spiro S (2005) A non-haem iron centre in the transcription factor NorR senses nitric oxide. Nature 437: 769–772 [PubMed]
  • Emsley P, Cowtan K (2004) Coot: model-building tools for molecular graphics. Acta Crystallogr D 60: 2126–2132 [PubMed]
  • Esnouf RM (1999) Further additions to MolScript version 1.4, including reading and contouring of electron-density maps. Acta Crystallogr D 55: 938–940 [PubMed]
  • Evgenov OV, Pacher P, Schmidt PM, Hasko G, Schmidt HH, Stasch JP (2006) NO-independent stimulators and activators of soluble guanylate cyclase: discovery and therapeutic potential. Nat Rev Drug Discov 5: 755–768 [PMC free article] [PubMed]
  • Friebe A, Koesling D (1998) Mechanism of YC-1-induced activation of soluble guanylyl cyclase. Mol Pharmacol 53: 123–127 [PubMed]
  • Garbers DL, Koesling D, Schultz G (1994) Guanylyl cyclase receptors. Mol Biol Cell 5: 1–5 [PMC free article] [PubMed]
  • Ignarro LJ (2002) Nitric oxide as a unique signaling molecule in the vascular system: a historical overview. J Physiol Pharmacol 53: 503–514 [PubMed]
  • Iyer LM, Anantharaman V, Aravind L (2003) Ancient conserved domains shared by animal soluble guanylyl cyclases and bacterial signaling proteins. BMC Genomics 4: 5. [PMC free article] [PubMed]
  • Jain R, Chan MK (2003) Mechanisms of ligand discrimination by heme proteins. J Biol Inorg Chem 8: 1–11 [PubMed]
  • Jentzen W, Ma JG, Shelnutt JA (1998) Conservation of the conformation of the porphyrin macrocycle in hemoproteins. Biophys J 74: 753–763 [PMC free article] [PubMed]
  • Kachalova GS, Popov AN, Bartunik HD (1999) A steric mechanism for inhibition of CO binding to heme proteins. Science 284: 473–476 [PubMed]
  • Karow DS, Pan D, Tran R, Pellicena P, Presley A, Mathies RA, Marletta MA (2004) Spectroscopic characterization of the soluble guanylate cyclase-like heme domains from Vibrio cholerae and Thermoanaerobacter tengcongensis. Biochemistry 43: 10203–10211 [PubMed]
  • Koesling D, Mullershausen F, Lange A, Friebe A, Mergia E, Wagner C, Russwurm M (2005) Negative feedback in NO/cGMP signalling. Biochem Soc Trans 33: 1119–1122 [PubMed]
  • Koesling D, Russwurm M, Mergia E, Mullershausen F, Friebe A (2004) Nitric oxide-sensitive guanylyl cyclase: structure and regulation. Neurochem Int 45: 813–819 [PubMed]
  • Kraulis PJ (1991) MOLSCRIPT: a program to produce both detailed and schematic plots of protein structures. J Appl Crystallogr 24: 946–950
  • Lahiri S, Roy A, Baby SM, Hoshi T, Semenza GL, Prabhakar NR (2006) Oxygen sensing in the body. Prog Biophys Mol Biol 91: 249–286 [PubMed]
  • Lamothe M, Chang FJ, Balashova N, Shirokov R, Beuve A (2004) Functional characterization of nitric oxide and YC-1 activation of soluble guanylyl cyclase: structural implication for the YC-1 binding site? Biochemistry 43: 3039–3048 [PubMed]
  • Lawson DM, Stevenson CE, Andrew CR, George SJ, Eady RR (2003) A two-faced molecule offers NO explanation: the proximal binding of nitric oxide to haem. Biochem Soc Trans 31: 553–557 [PubMed]
  • MacMicking J, Xie QW, Nathan C (1997) Nitric oxide and macrophage function. Annu Rev Immunol 15: 323–350 [PubMed]
  • Margulis A, Sitaramayya A (2000) Rate of deactivation of nitric oxide-stimulated soluble guanylate cyclase: influence of nitric oxide scavengers and calcium. Biochemistry 39: 1034–1039 [PubMed]
  • Martin E, Berka V, Bogatenkova E, Murad F, Tsai AL (2006) Ligand selectivity of soluble guanylyl cyclase: Effect of the hydrogen bonding tyrosine in the distal heme pocket on binding of oxygen, nitric oxide and carbon monoxide. J Biol Chem 281: 27836–27845 [PubMed]
  • Merritt EA, Bacon DJ (1997) Raster3D photorealistic molecular graphics. Methods Enzymol 277: 505–524 [PubMed]
  • Moncada S, Higgs EA (2006) The discovery of nitric oxide and its role in vascular biology. Br J Pharmacol 147 (Suppl 1): S193–S201 [PMC free article] [PubMed]
  • Murshudov GN, Vagin AA, Dodson EJ (1997) Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D 53: 240–255 [PubMed]
  • Nioche P, Berka V, Vipond J, Minton N, Tsai AL, Raman CS (2004) Femtomolar sensitivity of a NO sensor from Clostridium botulinum. Science 306: 1550–1553 [PubMed]
  • Ohmori M, Ikeuchi M, Sato N, Wolk P, Kaneko T, Ogawa T, Kanehisa M, Goto S, Kawashima S, Okamoto S, Yoshimura H, Katoh H, Fujisawa T, Ehira S, Kamei A, Yoshihara S, Narikawa R, Tabat S (2001) Characterization of genes encoding multi-domain proteins in the genome of the filamentous nitrogen-fixing Cyanobacterium anabaena sp. strain PCC 7120. DNA Res 8: 271–284 [PubMed]
  • Otwinowski Z, Minor W (1997) Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol 276: 307–326
  • Padayatti PS, Pattanaik P, Ma X, van den Akker F (2004) Structural insights into the regulation and activation mechanism of mammalian guanylyl cyclases. Pharmacol Ther 104: 83–99 [PubMed]
  • 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 USA 101: 12854–12859 [PMC free article] [PubMed]
  • Perrakis A, Harkiolaki M, Wilson KS, Lamzin VS (2001) ARP/wARP and molecular replacement. Acta Crystallogr D 57: 1445–1450 [PubMed]
  • Pflugrath JW (1999) The finer things in X-ray diffraction data collection. Acta Crystallogr D 55: 1718–1725 [PubMed]
  • Pyriochou A, Papapetropoulos A (2005) Soluble guanylyl cyclase: more secrets revealed. Cell Signal 17: 407–413 [PubMed]
  • Roberts GP, Youn H, Kerby RL (2004) CO-sensing mechanisms. Microbiol Mol Biol Rev 68: 453–473 [PMC free article] [PubMed]
  • Rodgers KR (1999) Heme-based sensors in biological systems. Curr Opin Chem Biol 3: 158–167 [PubMed]
  • Rothkegel C, Schmidt PM, Stoll F, Schroder H, Schmidt HH, Stasch JP (2006) Identification of residues crucially involved in soluble guanylate cyclase activation. FEBS Lett 580: 4205–4213 [PubMed]
  • Roy B, Garthwaite J (2006) Nitric oxide activation of guanylyl cyclase in cells revisited. Proc Natl Acad Sci USA 103: 12185–12190 [PMC free article] [PubMed]
  • Ruiz-Stewart I, Tiyyagura SR, Lin JE, Kazerounian S, Pitari GM, Schulz S, Martin E, Murad F, Waldman SA (2004) Guanylyl cyclase is an ATP sensor coupling nitric oxide signaling to cell metabolism. Proc Natl Acad Sci USA 101: 37–42 [PMC free article] [PubMed]
  • Russwurm M, Koesling D (2004a) Guanylyl cyclase: NO hits its target. Biochem Soc Symp 71: 51–63 [PubMed]
  • Russwurm M, Koesling D (2004b) NO activation of guanylyl cyclase. EMBO J 23: 4443–4450 [PMC free article] [PubMed]
  • Ryter SW, Alam J, Choi AM (2006) Heme oxygenase-1/carbon monoxide: from basic science to therapeutic applications. Physiol Rev 86: 583–650 [PubMed]
  • Schmidt PM, Rothkegel C, Wunder F, Schroder H, Stasch JP (2005) Residues stabilizing the heme moiety of the nitric oxide sensor soluble guanylate cyclase. Eur J Pharmacol 513: 67–74 [PubMed]
  • Schmidt PM, Schramm M, Schroder H, Wunder F, Stasch JP (2004) Identification of residues crucially involved in the binding of the heme moiety of soluble guanylate cyclase. J Biol Chem 279: 3025–3032 [PubMed]
  • Schneider TR (2000) Objective comparison of protein structures: error-scaled difference distance matrices. Acta Crystallogr D 56: 714–721 [PubMed]
  • 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 [PubMed]
  • Stone JR, Marletta MA (1998) Synergistic activation of soluble guanylate cyclase by YC-1 and carbon monoxide: implications for the role of cleavage of the iron-histidine bond during activation by nitric oxide. Chem Biol 5: 255–261 [PubMed]
  • Terwilliger TC, Berendzen J (1999) Automated MAD and MIR structure solution. Acta Crystallogr D 55: 849–861 [PMC free article] [PubMed]
  • Tesmer JJ, Sunahara RK, Gilman AG, Sprang SR (1997) Crystal structure of the catalytic domains of adenylyl cyclase in a complex with Gsalpha. GTPgammaS. Science 278: 1907–1916 [PubMed]
  • Tews I, Findeisen F, Sinning I, Schultz A, Schultz JE, Linder JU (2005) The structure of a pH-sensing mycobacterial adenylyl cyclase holoenzyme. Science 308: 1020–1023 [PubMed]
  • Van Duyne GD, Standaert RF, Karplus PA, Schreiber SL, Clardy J (1993) Atomic structures of the human immunophilin FKBP-12 complexes with FK506 and rapamycin. J Mol Biol 229: 105–124 [PubMed]
  • Waldman SA, Murad F (1987) Cyclic GMP synthesis and function. Pharmacol Rev 39: 163–196 [PubMed]
  • Winger JA, Marletta MA (2005) Expression and characterization of the catalytic domains of soluble guanylate cyclase: interaction with the heme domain. Biochemistry 44: 4083–4090 [PubMed]
  • Yazawa S, Tsuchiya H, Hori H, Makino R (2006) Functional characterization of two nucleotide-binding sites in soluble guanylate cyclase. J Biol Chem 281: 21763–21770 [PubMed]
  • Yuen PS, Doolittle LK, Garbers DL (1994) Dominant negative mutants of nitric oxide-sensitive guanylyl cyclase. J Biol Chem 269: 791–793 [PubMed]
  • Zhang G, Liu Y, Ruoho AE, Hurley JH (1997) Structure of the adenylyl cyclase catalytic core. Nature 386: 247–253 [PubMed]

Articles from The EMBO Journal are provided here courtesy of The European Molecular Biology Organization
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links