Reconciling conformational heterogeneity and substrate recognition in cytochrome P450

Kinetics of Intermediate Release Enhances P450 11B2-Catalyzed Aldosterone Synthesis

The mitochondrial enzyme cytochrome P450 11B2 (aldosterone synthase) catalyzes the 3 terminal transformations in the biosynthesis of aldosterone from 11-deoxycorticosterone (DOC): 11β-hydroxylation to corticosterone, 18-hydroxylation, and 18-oxidation. Prior studies have shown that P450 11B2 produces more aldosterone from DOC than from the intermediate corticosterone and that the reaction sequence is processive, with intermediates remaining bound to the active site between oxygenation reactions. In contrast, P450 11B1 (11β-hydroxylase), which catalyzes the terminal step in cortisol biosynthesis, shares a 93% amino acid sequence identity with P450 11B2, converts DOC to corticosterone, but cannot synthesize aldosterone from DOC. The biochemical and biophysical properties of P450 11B2, which enable its unique 18-oxygenation activity and processivity, yet are not also represented in P450 11B1, remain unknown. To understand the mechanism of aldosterone biosynthesis, we introduced point mutations at residue 320, which partially exchange the activities of P450 11B1 and P450 11B2 (V320A and A320V, respectively). We then investigated NADPH coupling efficiencies, binding kinetics and affinities, and product formation of purified P450 11B1 and P450 11B2, wild-type, and residue 320 mutations in phospholipid vesicles and nanodiscs. Coupling efficiencies for the 18-hydroxylase reaction with corticosterone as the substrate failed to correlate with aldosterone synthesis, ruling out uncoupling as a relevant mechanism. Conversely, corticosterone dissociation rates correlated inversely with aldosterone production. We conclude that intermediate dissociation kinetics, not coupling efficiency, enable P450 11B2 to synthesize aldosterone via a processive mechanism. Our kinetic data also suggest that the binding of DOC to P450 11B enzymes occurs in at least two distinct steps, favoring an induced-fit mechanism.

3site Multisubstrate-Bound State of Cytochrome P450cam

We identified a multisubstrate-bound state, hereby referred as a 3site state, in cytochrome P450cam via integrating molecular dynamics simulation with nuclear magnetic resonance (NMR) pseudocontact shift measurements. The 3site state is a result of simultaneous binding of three camphor molecules in three locations around P450cam: (a) in a well-established "catalytic" site near heme, (b) in a kink-separated "waiting" site along channel-1, and (c) in a previously reported "allosteric" site at E, F, G, and H helical junctions. These three spatially distinct binding modes in the 3site state mutually communicate with each other via homotropic allostery and act cooperatively to render P450cam functional. The 3site state shows a significantly superior fit with NMR pseudo contact shift (PCS) data with a Q-score of 0.045 than previously known bound states and consists of D251 free of salt-bridges with K178 and R186, rendering the enzyme functionally primed. To date, none of the reported cocomplex of P450cam with its redox partner putidaredoxin (pdx) has been able to match solution NMR data and controversial pdx-induced opening of P450cam's channel-1 remains a matter of recurrent discourse. In this regard, inclusion of pdx to the 3site state is able to perfectly fit the NMR PCS measurement with a Q-score of 0.08 and disfavors the pdx-induced opening of channel-1, reconciling previously unexplained remarkably fast hydroxylation kinetics with a koff of 10.2 s-1. Together, our findings hint that previous experimental observations may have inadvertently captured the 3site state as an in vitro solution state, instead of the catalytic state alone, and provided a distinct departure from the conventional understanding of cytochrome P450.

Identifying Selectivity Filters in Protein Biosensor for Ligand Screening

Specialized sensing mechanisms in bacteria enable the identification of cognate ligands with remarkable selectivity in highly xenobiotic-polluted environments where these ligands are utilized as energy sources. Here, via integrating all-atom computer simulation, biochemical assay, and isothermal titration calorimetry measurements, we determine the molecular basis of MopR, a phenol biosensor's complex selection process of ligand entry. Our results reveal a set of strategically placed selectivity filters along the ligand entry pathway of MopR. These filters act as checkpoints, screening diverse aromatic ligands at the protein surface based on their chemical features and sizes. Ligands meeting specific criteria are allowed to enter the sensing site in an orientation-dependent manner. Sequence and structural analyses demonstrate the conservation of this ligand entry mechanism across the sensor class, with individual amino acids along the selectivity filter path playing a critical role in ligand selection. Together, this investigation highlights the importance of interactions with the ligand entry pathway, in addition to interactions within the binding pocket, in achieving ligand selectivity in biological sensing. The findings enhance our understanding of ligand selectivity in bacterial phenol biosensors and provide insights for rational expansion of the biosensor repertoire, particularly for the biotechnologically relevant class of aromatic pollutants.

Resolving Protein Conformational Plasticity and Substrate Binding via Machine Learning

A long-standing target in elucidating the biomolecular recognition process is the identification of binding-competent conformations of the receptor protein. However, protein conformational plasticity and the stochastic nature of the recognition processes often preclude the assignment of a specific protein conformation to an individual ligand-bound pose. Here, we demonstrate that a computational framework coined as RF-TICA-MD, which integrates an ensemble decision-tree-based Random Forest (RF) machine learning (ML) technique with an unsupervised dimension reduction approach time-structured independent component analysis (TICA), provides an efficient and unambiguous solution toward resolving protein conformational plasticity and the substrate binding process. In particular, we consider multimicrosecond-long molecular dynamics (MD) simulation trajectories of a ligand recognition process in solvent-inaccessible cavities of archetypal proteins T4 lysozyme and cytochrome P450cam. We show that in a scenario in which clear correspondence between protein conformation and binding-competent macrostates could not be obtained via an unsupervised dimension reduction approach, an a priori decision-tree-based supervised classification of the simulated recognition trajectories via RF would help characterize key amino acid residue pairs of the protein that are deemed sensitive for ligand binding. A subsequent unsupervised dimensional reduction of the selected residue pairs via TICA would then delineate a conformational landscape of protein which is able to demarcate ligand-bound poses from unbound ones. The proposed RF-TICA-MD approach is shown to be data agnostic and found to be robust when using other ML-based classification methods such as XGBoost. As a promising spinoff of the protocol, the framework is found to be capable of identifying distal protein locations which would be allosterically important for ligand binding and would characterize their roles in recognition pathways. A Python implementation of a proposed ML workflow is available in GitHub https://github.com/navjeet0211/rf-tica-md.

A deep autoencoder framework for discovery of metastable ensembles in biomacromolecules

Biomacromolecules manifest dynamic conformational fluctuation and involve mutual interconversion among metastable states. A robust mapping of their conformational landscape often requires the low-dimensional projection of the conformational ensemble along optimized collective variables (CVs). However, the traditional choice for the CV is often limited by user-intuition and prior knowledge about the system, and this lacks a rigorous assessment of their optimality over other candidate CVs. To address this issue, we propose an approach in which we first choose the possible combinations of inter-residue Cα-distances within a given macromolecule as a set of input CVs. Subsequently, we derive a non-linear combination of latent space embedded CVs via auto-encoding the unbiased molecular dynamics simulation trajectories within the framework of the feed-forward neural network. We demonstrate the ability of the derived latent space variables in elucidating the conformational landscape in four hierarchically complex systems. The latent space CVs identify key metastable states of a bead-in-a-spring polymer. The combination of the adopted dimensional reduction technique with a Markov state model, built on the derived latent space, reveals multiple spatially and kinetically well-resolved metastable conformations for GB1 β-hairpin. A quantitative comparison based on the variational approach-based scoring of the auto-encoder-derived latent space CVs with the ones obtained via independent component analysis (principal component analysis or time-structured independent component analysis) confirms the optimality of the former. As a practical application, the auto-encoder-derived CVs were found to predict the reinforced folding of a Trp-cage mini-protein in aqueous osmolyte solution. Finally, the protocol was able to decipher the conformational heterogeneities involved in a complex metalloenzyme, namely, cytochrome P450.