Ligand-induced shifts in conformational ensembles that describe transcriptional activation
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
The manuscript will be of broad interest to readers in the fields of biochemistry, structural, molecular, and evolutionary biology. It outlines a systematic approach in characterizing nuclear receptor ligands based on the conformational ensemble of the receptor, further exploring the idea that perturbation of the ensemble orchestrates function. The results from the combined use of experiments and simulation are promising, suggesting that the change in the ensemble is responsible for function.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #3 agreed to share their name with the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Biochemistry and Chemical Biology (eLife)
Abstract
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.
Article activity feed
-
-
Evaluation Summary:
The manuscript will be of broad interest to readers in the fields of biochemistry, structural, molecular, and evolutionary biology. It outlines a systematic approach in characterizing nuclear receptor ligands based on the conformational ensemble of the receptor, further exploring the idea that perturbation of the ensemble orchestrates function. The results from the combined use of experiments and simulation are promising, suggesting that the change in the ensemble is responsible for function.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #3 agreed to share their name with the authors.)
-
Reviewer #1 (Public Review):
In this study, the authors compare computational MD simulations with functional activity data to determine if ligand activity can be predicted from simulations. As a test case, the authors use the ligand-binding domain (LBD) of an ancestral steroid receptor (AncSR2) that they and others have previously studied, providing a well-characterized system for their analyses. The studies include wild-type (WT) AncSR2 as well as four mutant proteins where a single methionine residue that contacts the steroid hormone within the pocket (Met75) was mutated (to Ala, Phe, Ile, or Leu). Computational analyses are performed to assess the stability of the complexes and determine whether the conformational ensembles generated show similarities or differences between the WT vs. mutant forms, or apo vs. ligand-bound forms …
Reviewer #1 (Public Review):
In this study, the authors compare computational MD simulations with functional activity data to determine if ligand activity can be predicted from simulations. As a test case, the authors use the ligand-binding domain (LBD) of an ancestral steroid receptor (AncSR2) that they and others have previously studied, providing a well-characterized system for their analyses. The studies include wild-type (WT) AncSR2 as well as four mutant proteins where a single methionine residue that contacts the steroid hormone within the pocket (Met75) was mutated (to Ala, Phe, Ile, or Leu). Computational analyses are performed to assess the stability of the complexes and determine whether the conformational ensembles generated show similarities or differences between the WT vs. mutant forms, or apo vs. ligand-bound forms (aromatic vs. 3-keto non-aromatic A-ring, EST/estrogen vs. progesterone/PROG). Simulations included conventional and accelerated methods. Clustering analysis of the accelerated simulations revealed some similarities and differences, which the authors then compare to luciferase reporter assay data (Gal4-fusion + WT vs. mutant LBDs) for the mutants where they performed dose-response experiments (up to 1 µM ligand added). One of the mutants studied did not show any activity (M75I); however, M75I and M75L both showed increased basal transcriptional activity (constitutively active) vs. WT without an exogenously added ligand. The authors developed a fluorescent ligand binding assay and showed the M75I mutant does not bind ligands (at least up to 1 µM added ligand). Next, hydrogen/deuterium exchange mass spectrometry data are provided to inform how the M75L mutant is constitutively active. The HDX results indicate that several regions display higher deuterium uptake in the M75L mutant and PROG binding has a larger destabilizing effect on WT vs. M75L. Finally, some structural snapshots from the MD simulations are shown (Fig 6A-C) that the authors claim to explain the altered transcriptional response of the M75 mutants vs. WT.
This study may be one of the first to attempt to make qualitative correlations between computational simulations of ligand-bound/free nuclear receptor LBDs and functional outcome. One could see a future where many different ligands are docked and a more quantitative, streamlined pipeline is used to predict functional outcome-this study takes the important first step in trying to determine if there are simulation-function correlations.
-
Reviewer #2 (Public Review):
This paper investigates the effects of mutation of a key Helix 5 residue of an ancestral glucocorticoid receptor (AncGR) on an AncGR simulation-generated structural ensemble and how effects on this ensemble correlate with experimentally measured receptor function. Utilizing simulations, the authors observe that ligand binding shifts the structural ensemble of select AncGR mutants. Using wet experiments, they find that non-ligand-responsive mutants, either don't bind the tested ligands, bind them poorly, or are not affected by ligand binding. These results indicate that enhanced sampling simulations are capable of capturing biologically relevant alternative states. Many of the conclusions drawn by the authors are supported by the data, particularly the correlation between co-clustering of apo with ligand …
Reviewer #2 (Public Review):
This paper investigates the effects of mutation of a key Helix 5 residue of an ancestral glucocorticoid receptor (AncGR) on an AncGR simulation-generated structural ensemble and how effects on this ensemble correlate with experimentally measured receptor function. Utilizing simulations, the authors observe that ligand binding shifts the structural ensemble of select AncGR mutants. Using wet experiments, they find that non-ligand-responsive mutants, either don't bind the tested ligands, bind them poorly, or are not affected by ligand binding. These results indicate that enhanced sampling simulations are capable of capturing biologically relevant alternative states. Many of the conclusions drawn by the authors are supported by the data, particularly the correlation between co-clustering of apo with ligand bound AncGR and wt with mutant AncGR simulation structures and cell-based activity of the receptors; however, the relationship between simulation and HDX-MS data is not clear.
-
Reviewer #3 (Public Review):
This manuscript details a methodological approach for the characterisation of ligands based on nuclear receptor conformational ensembles. Using ancestral steroid receptor AncSR2 and atomistic MD simulations, the authors generated ensembles of the WT and mutants of the conserved Methionine residue at position 75. The mutation, as well as the ligands (3-ketosteroid hormones and estradiol), shifted the populations into distinct conformational clusters. These clusters were then well correlated to ligand activation, making use of the cell-based luciferase assay. Next, the binding affinities of the ligands to the WT, M75L, and M75I were probed by fluorescence polarization assay to understand the extraordinary activation of M75L by estradiol (inactive ligand). The decreased binding affinity of M25L for the ligands …
Reviewer #3 (Public Review):
This manuscript details a methodological approach for the characterisation of ligands based on nuclear receptor conformational ensembles. Using ancestral steroid receptor AncSR2 and atomistic MD simulations, the authors generated ensembles of the WT and mutants of the conserved Methionine residue at position 75. The mutation, as well as the ligands (3-ketosteroid hormones and estradiol), shifted the populations into distinct conformational clusters. These clusters were then well correlated to ligand activation, making use of the cell-based luciferase assay. Next, the binding affinities of the ligands to the WT, M75L, and M75I were probed by fluorescence polarization assay to understand the extraordinary activation of M75L by estradiol (inactive ligand). The decreased binding affinity of M25L for the ligands was further investigated using differential hydrogen-deuterium exchange (HDX). The deprotection pattern observed for the M25L mutant compared to WT and decreased binding affinity of the ligands for this mutant led to the conclusion that this specific mutation shifts the ensemble conformation to a ligand-bound state.
This approach can be useful for the prediction of ligand responses, understanding underlying mechanisms, and their detailed characterisation based on the population shifts of the nuclear receptor conformational ensembles. It is commendable that the results obtained from computational techniques are well supported by a range of biochemical and biophysical techniques. Logical correlation is established between the results and light is shed on the underlying molecular mechanism through in-depth discussion. The control of the mutants based on secondary structure, melting temperature, and purity through SDS-PAGE is appreciable. The techniques are well chosen and appropriate to reach the conclusions.
-
