| The Nobel Prize in physiology and medicine in1998was awarded to R. F. Furchgott, L. J. Ignarro, and F. Murad for their discovery of "the nitric oxide (NO) as a signaling molecule in the cardiovascular system". This stimulated intense interests of scientists in the study on soluble guanylate cyclase (sGC) mediated NO-signaling system. sGC, as a nitric oxide sensor, is a critical heme-containing enzyme in NO-signaling pathway of eukaryotes. Upon NO binding, sGC catalyzes the conversion of guanosine5’-triphosphate (GTP) to3’,5’-cyclic guanosine monophosphate (cGMP). cGMP, as an important second messenger, regulates several effector proteins and plays an important role in many physiological processes, for example, vasodilatation, smooth muscle relaxation, platelet aggregation, and neuronal transmission. Dysfunction of NO signaling results in many pathological disorders, ranging from several cardiovascular diseases, such as arterial hypertension, pulmonary hypertension, heart failure, atherosclerosis and restenosis, to neurodegenerative diseases.sGC in eukaryote is a heterodimeric hemoprotein, composed of a and β subunit. For human sGC, there are two isoforms for each subunit in vivo:α1/α2and (β1/β2, while the α1β1heterodimer is the more abundant. The prosthetic heme moiety, crucial for NO sensing, is located in the heme binding domain with His105of β1subunit as the axial ligand. Both α and β subunit can be divided into three domains:the N-terminal heme domain (H-NOX family), central domain (containing Per/Armt/Sim (PAS)-like region and Coiled-coil region) and the C-terminal catalytic domain. The heme domain, responsible for heme binding and NO/CO binding, is one of the most important and popular regions in sGC study.In the past decades, there were nearly5,000papers published in this area. However, several crucial problems are still unclear and controversial. For example, the exact heme binding region and environment is to be determined; the exact mechanism of NO activation and deactivation is unknown although several models have been proposed; the region for the heterodimerization regulation and related mechanism is ambiguous; all the above problems are largely due to the limited structural information for sGC structure. Limited availability of large quantities of highly pure sGC and its large size has impeded analysis of the crystal structure. Originally, sGC was isolated from a number of mammalian tissues, mainly bovine and rat lungs. In the case of human isoforms, direct isolation from native sources becomes virtually impossible. Although the functional expression of recombinant human sGC was achieved in the baculovirus/sf9cells, the yield of human sGC was relatively limited. To this end, the E. coli expression system is the most desired method to overexpress sGC. However, the overexpression and purification of eukaryotic sGC is one of the bottle neck for sGC study and hardly reported.This thesis is focused on the heme domain of human sGC (hsGC). We report, for the first time, the recombinant hsGC in full-length (hsGCβ619) and its truncated N-terminal fragments with195and384residues (hsGCβ195and hsGCβ384), which were overexpressed in E. coli and purified successfully with a yield of20mg/L cell culture. This highly efficient E. coli expression system of hsGC proteins is critical to the study of the structure, function and catalytic mechanism of human sGC and we have submitted the application of patent. As for sGC heme domain, there are only two published crystal structures of H-NOX domain from bacteria, Tt H-NOX and Ns H-NOX. Based on the crystal structure of Ns H-NOX domain (pdb entry:2O09), the homology model of hsGCβ195was constructed through energy optimization by NAMD program. After heme reconstitution, the three proteins in different forms (ferric, ferrous, NO-bound and CO-bound) were characterized via UV-vis, EPR, CD and fluorescence spectroscopy. The characterization results suggested that hsGCβ195, as an excellent tool, can be used to further study the sGC heme pocket and NO activation/deactivation mechanism. The midpoint temperature (Tm) of conformation transition for the apo-hsGCβ195and heme-hsGCβ195is56±1℃and54±3℃, respectively, indicating the similar and good stability for temperature. The pH titration results showed that the midpoint pH (pKa) for the acid and alkaline transition process is5.7±0.2and9.3±0.1, respectively. hsGCβ195exhibites desirable stability in alkaline condition, while hsGCβ195easily precipitates along with heme dissociation in acidic condition. Trp22, the only tryptophan residue in hsGCβ195, can be used to study the conformation change with the intrinsic fluorescence spectra measurement. The Trp22fluorescence spectra revealed that Trp22was at a relative hydrophobic environment with the maximum emission wavelength (λmax) at327nm for the apo-hsGCβ195protein. Heme reconstitution resulted in the λmax red-shifting and a largely concomitant decrease in the fluorescence intensity, revealing the large conformation change occurred, especially for the ferrous and NO-bound forms. We conferred that the heme binding induced the movement of the first a-helix (aA), leading to the change of relative position between Glu10(located at aA) and Trp22and further quench the Trp22fluorescence. ANS fluorescence is commonly to investigate the hydrophobic properties of proteins. We found that ANS could bind to some part of sGC hydrophobic pocket and complete with heme binding.To further explore the heme binding region and heme pocket microenvironment, the heme domain of hsGC al subunit (hsGCa259) was constructed, expressed and purified. The homology model of hsGCa259displayed a large hydrophobic pocket and the heme could be located at hsGCa259stably via heme reconstitution, which resulted in the speculation that the role of the pocket is heme binding. The EPR spectra of ferric hsGC identified that the heme of hsGCβ195is5-coordinate high spin with His105as the axial ligand, while both the axial ligands of the heme in hsGCa259are weak, probably H2O. We guessed that heme binds to hsGCa259probably with strong hydrophobic interaction according to heme transfer and ANS fluorescence measurements. This is the first direct evidence that both the heme domain of al and β1subunit contribute to the heme binding. The CO and NO binding equilibrium and kinetics of hsGCa259and hsGCβ195were carried out by UV-vis and stopped-flow measurement. One possible CO binding model is proposed. NO dissociation results suggested that NO dissociation is a more complex process and possibly two different but UV-vis spectroscopy indistinguishable NO-bound heme existed. Recently, the pocket of hsGCα1subunit was inferred to YC-1binding, which will lead to CO affinity increase. However, we found that YC-1had little effect on CO affinity, the isothermal titration calorimetry (ITC) also indicated no obvious interaction between YC-1and hsGCa259.In order to investigate the exact role of the heme axial ligand (His105), hsGCβ195H105G was constructed and fully characterized. EPR spectra proved that the axial ligand was removed in hsGCβ195H105G. The Trp22fluorescence indicated that the H105G mutation had little effect on the apo-hsGCβ195conformation, while exhibited large influence on the heme-hsGCβ195conformation. The heme transfer and reduction experiments suggested that H105mainly increase the heme reduced potential and protect the protein from heme oxidation and loss, which is critical for several physiological processes.On the purpose of studying the effect of heterodimerization on the heme binding properties, hsGCβ195-a259, the hybrid protein of hsGCβ195and hsGCa259linked with GSGSGG peptide, was constructed and detailed characterized. The CD and Trp22fluorescence spectra demonstrated that hsGCβ195-α259displayed tremendous conformation change compared to hsGCβ195or hsGCα259with excellent stability. The heme reconstitution and EPR spectra identified that hsGCβ195-α259could bind heme via the similar fashion with that of hsGCα259, in which His105did not bond to heme. We conferred that there are two possible modes of heme location in hsGCβ195-α259. The first possibility is that heme mainly resides at the hydrophobic pocket of β195. The addition of linker and α259induces a large conformation change through interaction with β195, resulting that H105is not close enough to bond with heme. The other situation is that the entrance of heme to β195is blocked by the addition of linker and α259, which leads to the impossible bonding between H105and heme. The exact heme binding pattern awaits further experiments, particularly the resolution of crystal structure. The CO and NO binding results showed that hsGCβ195-α259exhibited higher CO affinity (almost twice) and larger NO dissociation rates (approximate twice) compared with hsGCβ195or hsGCα259. Besides, we co-expressed hsGCβ195and hsGCα259by inserting β195and α259into the two multi-cloning sites (MCS) of pETDuet-1vector, respectively and demonstrated that hsGCβ195definitely interacted with hsGCα259in vivo through Amylose pull-down and Ni-NTA pull-down methods. In contrast, no observable heterodimer was detected by mixing purified hsGCβ195or hsGCα259in vitro. The above observations revealed that the heme domain of sGC indeed contributed to the sGC heterodimerization to a certain extent, which may be regulated by in vivo signaling.In addition, we devoted much effort to study the crystal structure of α1/β1heme domain. Recently and fortunately, we cultured several hsGCβ195-α259and hsGCβ195H105G-α259crystals, containing apo, heme-containing, CO-bound and NO-bound forms. The diffraction data have collected and the structure determination is undergoing. We believe that the related crystal structure will broaden our knowledge of the structure-property-reactivity relationships of sGC and be beneficial for understanding the overall structure of the heme binding site of hsGC and the NO/CO signaling mechanism. |
PDF Full Text Request