Ancient and modern mechanisms compete in progesterone receptor activation.

The progesterone receptor (PR) belongs to the steroid receptor family of ligand-regulated transcription factors, controlling genes important for development, metabolism, and reproduction. Understanding how diverse ligands bind and modulate PR activity will illuminate the design of ligands that control PR-driven signaling pathways. Here, we use molecular dynamics simulations to investigate how PR dynamics are altered by functionally diverse ligands. Using a library of 33 steroidal ligands that range from inactive to EC50 < 0.1 nM, we reveal an unexpected evolutionary basis for the wide gamut of activation. While other oxosteroid receptors employ an evolutionarily conserved mechanism dependent on a hydrogen bond between the receptor and ligand, extant PR has evolved a preference for activation that is not reliant on this polar interaction. We demonstrate that potent ligands utilize the modern PR mechanism while weaker ligands coopt the defunct ancestral mechanism by forming hydrogen bonds with Asn719. Based on their structures and dynamic signatures, ligands partition into four classes (inactive, weak, moderate and high potency) that interact distinctly with the PR binding pocket. Further, we use luciferase reporter assays and PR mutants to probe the roles of pocket residues in mediating distinct PR mechanisms. This combination of MD simulations and in vitro studies provide insight into how the evolutionary history of PR shapes its response to diverse ligands.

Identification of an Evolutionarily Conserved Allosteric Network in Steroid Receptors.

Allosteric pathways in proteins describe networks comprising amino acid residues which may facilitate the propagation of signals between distant sites. Through inter-residue interactions, dynamic and conformational changes can be transmitted from the site of perturbation to an allosteric site. While sophisticated computational methods have been developed to characterize such allosteric pathways linking specific sites on proteins, few attempts have been made to apply these approaches toward identifying new allosteric sites. Here, we use molecular dynamics simulations and suboptimal path analysis to discover new allosteric networks in steroid receptors with a focus on evolutionarily conserved pathways. Using modern receptors and a reconstructed ancestral receptor, we identify networks connecting several sites to the activation function surface 2 (AF-2), the site of coregulator recruitment. One of these networks is conserved across the entire family, connecting a predicted allosteric site located between helices 9 and 10 of the ligand-binding domain. We investigate the basis of this conserved network as well as the importance of this site, discovering that the site lies in a region of the ligand-binding domain characterized by conserved inter-residue contacts. This study suggests an evolutionarily importance of the helix 9-helix 10 site in steroid receptors and identifies an approach that may be applied to discover previously unknown allosteric sites in proteins.

Interactions governing transcriptional activity of nuclear receptors.

The key players in transcriptional regulation are transcription factors (TFs), proteins that bind specific DNA sequences. Several mechanisms exist to turn TFs 'on' and 'off', including ligand binding which induces conformational changes within TFs, subsequently influencing multiple inter- and intramolecular interactions to drive transcriptional responses. Nuclear receptors are a specific family of ligand-regulated TFs whose activity relies on interactions with DNA, coregulator proteins and other receptors. These multidomain proteins also undergo interdomain interactions on multiple levels, further modulating transcriptional outputs. Cooperation between these distinct interactions is critical for appropriate transcription and remains an intense area of investigation. In this review, we report and summarize recent findings that continue to advance our mechanistic understanding of how interactions between nuclear receptors and diverse partners influence transcription.

Three-dimensional structure of a mycobacterial oligoribonuclease reveals a unique C-terminal tail that stabilizes the homodimer.

Oligoribonucleases (Orns) are highly conserved DnaQ-fold 3'-5' exoribonucleases that have been found to carry out the last step of cyclic-di-GMP (c-di-GMP) degradation, that is, pGpG to GMP in several bacteria. Removal of pGpG is critical for c-di-GMP homeostasis, as excess uncleaved pGpG can have feedback inhibition on phosphodiesterases, thereby perturbing cellular signaling pathways regulated by c-di-GMP. Perturbation of c-di-GMP levels not only affects survival under hypoxic, reductive stress, or nutrient-limiting conditions but also affects pathogenicity in infection models as well as antibiotic response in mycobacteria. Here, we have determined the crystal structure of MSMEG_4724, the Orn of Mycobacterium smegmatis (Ms_orn) to 1.87 Å resolution to investigate the function of its extended C-terminal tail that is unique among bacterial Orns. Ms_orn is a homodimer with the canonical RNase-H fold of exoribonucleases and conserved catalytic residues in the active site. Further examination of the substrate-binding site with a modeled pGpG emphasized the role of a phosphate cap and "3'OH cap" in constricting a 2-mer substrate in the active site. The unique C-terminal tail of Ms_orn aids dimerization by forming a handshake-like flap over the second protomer of the dimer. Our thermal and denaturant-induced unfolding experiments suggest that it helps in higher stability of Ms_orn as compared with Escherichia coli Orn or a C-terminal deletion mutant. We also show that the C-terminal tail is required for modulating response to stress agents in vivo. These results will help in further evaluating the role of signaling and regulation by c-di-GMP in mycobacteria.

Ligand-induced shifts in conformational ensembles that describe transcriptional activation.

Nuclear receptors function as ligand-regulated transcription factors whose ability to regulate diverse physiological processes is closely linked with conformational changes induced upon ligand binding. Understanding how conformational populations of nuclear receptors are shifted by various ligands could illuminate strategies for the design of synthetic modulators to regulate specific transcriptional programs. Here, we investigate ligand-induced conformational changes using a reconstructed, ancestral nuclear receptor. By making substitutions at a key position, we engineer receptor variants with altered ligand specificities. We combine cellular and biophysical experiments to characterize transcriptional activity, as well as elucidate mechanisms underlying altered transcription in receptor variants. We then use atomistic molecular dynamics (MD) simulations with enhanced sampling to generate ensembles of wildtype and engineered receptors in combination with multiple ligands, followed by conformational analysis and correlation of MD-based predictions with functional ligand profiles. We determine that conformational ensembles accurately describe ligand responses based on observed population shifts. These studies provide a platform which will allow structural characterization of physiologically-relevant conformational ensembles, as well as provide the ability to design and predict transcriptional responses in novel ligands.

Structural and thermodynamic characterization of a highly stable conformation of Rv2966c, a 16S rRNA methyltransferase, at low pH.

Rv2966c is a highly specific methyltransferase that methylates G966 at the N2 position in 16S rRNA of mycobacterial ribosome and can be secreted inside the host cell to methylate host DNA. However, how the secreted protein retains its structure and function in the harsh environment of host cell, remains unclear. In this work, we investigate structural features of Rv2966c at pH 4.0 and pH 7.5 by far-UV- and near-UV-circular dichroism (CD) and fluorescence spectroscopy, to gain insights into its folding and stability at the acidic pH, that it is likely to encounter inside the macrophage. We show that Rv2966c exists in a compact, folded state at both pH 7.5 and pH 4.0, a result corroborated by molecular dynamics simulations as a function of pH. In fact, Rv2966c was found to be more stable at pH 4.0 than pH 7.5, as evidenced by thermal-induced CD and nanodifferential scanning fluorimetry, and urea-induced denaturation measurements. Interestingly, unlike pH 7.5 (two-state unfolding), denaturation of Rv2966c at pH 4.0 occurs in a biphasic (N ↔ X ↔ U) manner. Further spectroscopic characterization of 'X' state, identifies characteristics of a molten globule-like intermediate. We finally conclude that Rv2966c maintains a compact folded state at pH 4.0 akin to that at pH 7.5 but with higher stability.

Effects of natural mutations (L94I and L94V) on the stability and mechanism of folding of horse cytochrome c: A combined in vitro and molecular dynamics simulations approach.

Known crystal structures of 10 cytochromes (cyts) c from different sources led to the conclusion that natural mutations in these proteins does not affect their 3D structure, hence evolution preserved structure for function. A sequence alignment of horse cyt c with all other 284 cyts c led to two important conclusions: (i) Leu at position 94 is conserved in all 30 mammalian known sequences, and (ii) there are 14 other species which have either Val or Ile at 94th position. We asked a question: Is the avoidance of substitution by Val or Ile at position 94 in the mammalian cyts c by design or by chance? To answer this question, we introduced natural substitutes of Leu94 by Val and Ile in horse cyt c using site-directed mutagenesis. Here, from our in vitro and molecular dynamic simulation studies on L94V and L94I mutants, we concluded that (i) although the natural mutations destabilize the wild type cyt c, it does not significantly affect the mechanism of folding of the protein, (ii) urea-induced denaturation of WT cyt c and its mutants is a two-state process, and (iii) denaturation of WT cyt c and its mutants by guanidinium chloride is not a two-state process.

Protein folding: Molecular dynamics simulations and in vitro studies for probing mechanism of urea- and guanidinium chloride-induced unfolding of horse cytochrome-c.

Urea- and guanidinium chloride (GdmCl)-induced denatured states of horse cytochrome-c (cyt-c) are structurally identical. It is then expected that estimates of ∆G0N→U (Gibbs free energy change in the absence of denaturants) from GdmCl- and urea-induced denaturation curves should be identical, if denaturation induced by them follows a two-state mechanism. That denaturation of cyt-c by urea or GdmCl follows a two-state mechanism is reported by some in vitro studies while other in vitro studies reported contradictory observations. Molecular dynamic (MD) simulation is a technique that could reveal the mechanism of unfolding/folding of proteins in the absence and presence of chemical denaturants at the amino acid residue level. We therefore performed multiple unconstrained MD simulations of cyt-c (PDB ID: 1HRC) in water and aqueous mixtures of GdmCl and urea for the period of 0-500 ns at 300, 400 and 450 K, which showed that denaturation of cyt-c by urea and GdmCl is a two-state and three-state process, respectively. To corroborate these findings, we measured urea- and GdmCl-induced denaturation curves of different optical properties (circular dichroism at 222, 405 and 416 nm and absorbance at 405 nm) and analyzed them for ∆G0N→U. These studies supported conclusions reached from MD simulation studies.

Unravelling the unfolding mechanism of human integrin linked kinase by GdmCl-induced denaturation.

Integrin-linked kinase (ILK) is a ubiquitously expressed Ser/Thr kinase which plays significant role in the cell-matrix interactions and growth factor signalling. In this study, guanidinium chloride (GdmCl)-induced unfolding of kinase domain of ILK (ILK193-446) was carried out at pH 7.5 and 25 °C. Eventually, denaturation curves of mean residue ellipticity at 222 nm ([θ]222) and fluorescence emission spectrum were analysed to estimate stability parameters. The optical properties maximum emission (λmax) and difference absorption coefficient at 292 nm (Δε292) were analysed. The denaturation curve was measured only in the GdmCl molar concentration ranging 3.0-4.2 M because protein was aggregating below 3.0 M of GdmCl concentrations. The denaturation process of ILK193-446 was found as reversible at [GdmCl] ≥ 3.0 M. Moreover, a coincidence of normalized denaturation curves of optical properties ([θ]222, Δε292 and λmax) suggesting that GdmCl-induced denaturation of ILK193-446 is a two-state process. In addition, 100 ns molecular dynamics simulations were performed to see the effects of GdmCl on the structure and stability of ILK193-446. Both the spectroscopic and molecular dynamics approaches provided clear insights into the stability and conformational properties of ILK193-446.

Mechanistic insights into the urea-induced denaturation of kinase domain of human integrin linked kinase.

Integrin-linked kinase (ILK), a ubiquitously expressed intracellular Ser/Thr protein kinase, plays a major role in the oncogenesis and tumour progression. The conformational stability and unfolding of kinase domain of ILK (ILK193-446) was examined in the presence of increasing concentrations of urea. The stability parameters of the urea-induced denaturation were measured by monitoring changes in [θ]222 (mean residue ellipticity at 222nm), difference absorption coefficient at 292nm (Δε292) and intrinsic fluorescence emission intensity at pH7.5 and 25±0.1°C. The urea-induced denaturation was found to be reversible. The protein unfolding transition occurred in the urea concentration range 3.0-7.0M. A coincidence of normalized denaturation curves of optical properties ([θ]222, Δε292 and λmax, the wavelength of maximum emission intensity) suggested that urea-induced denaturation of kinase domain of ILK is a two-state process. We further performed molecular dynamics simulation for 100ns to see the effect of urea on structural stability of kinase domain of ILK at atomic level. Structural changes with increasing concentrations of urea were analysed, and we observed a significant increase in the root mean square deviation, root mean square fluctuations, solvent accessible surface area and radius of gyration. A correlation was observed between in vitro and in silico studies.

Effect of conservative mutations (L94V and L94I) on the structure and stability of horse cytochrome c.

A sequence alignment of horse cytochrome c (cyt c) with all known cyts c shows that Leu at position 94 is conserved, except in 14 species which have either Val or Ile at this position. It is also known that Leu94 of the mammalian cyt c plays an important role in folding and stability. The important question here is as to what will happen in terms of folding and stability if Leu94 of the mammalian cyt c is substituted by Val or Ile. To answer this question, we introduced natural substitutes of Leu94 by Val and Ile in horse cyt c. The purified L94V and L94I mutants under native condition (pH 6.0, 25 °C) were characterized using far-UV, near-UV and Soret- circular dichroism, visible absorbance, Trp and ANS (1-anilino-8-napthaline sulphonate) fluorescence and dynamic light scattering measurements. Furthermore, stability parameters Tm (mid-point of denaturation) and ΔGD0 (Gibbs free energy change at 25 °C) were also determined using spectroscopic and differential scanning calorimetric methods. All these measurements led us to conclude that both mutants exist as molten globule and are less stable than the wild-type protein. These observations are supported well by examining the structure of horse cyt c (PDB ID, 1HRC).

Structural and thermodynamic characterisation of L94F mutant of horse cytochrome c.

Mammalian mitochondrial cytochromes c (cyts c) has a conserved Leu94 which is replaced by valine/isoleucine in some lower eukaryotes and prokaryotes. It is expected that nature substituted Leu94 with Val/Ile, for they have similar van der Waals volume and hydrophobicity with a difference in side chain branching only. Reports also suggested the presence of phenylalanine at position 94, which leads to questions: (i) How bulky aromatic amino acid residue fitted at position 94 in cyt c family proteins? (ii) What is the effect of L94F mutation on protein stability and folding? Here, we selected horse cyt-c as a model to answer the second question. We generated L94F mutant of horse cytochrome c and subsequently characterised using far-UV, near-UV and Soret circular dichroism, absorbance, intrinsic and extrinsic ANS (8-anilino-1-napthalenesulfonic acid) fluorescence and dynamic light scattering measurements. We observed that this mutation affects the native state and arrests the protein folding at the molten globule state. Thermal stability of L94F mutant is also measured by spectroscopic techniques and differential scanning calorimetry. The midpoint of thermal denaturation of L94F mutant is 17°C less than wild type. Molecular dynamics simulation study also supports our in vitro observation that this mutant has stable backbone conformation.