Prediction of New Stabilizing Mutations Based on Mechanistic Insights from Markov State Models

Abstract

Protein stabilization is fundamental to enzyme function and evolution, yet understanding the determinants of a protein's stability remains a challenge. This is largely due to a shortage of atomically detailed models for the ensemble of relevant protein conformations and their relative populations. For example, the M182T substitution in TEM β-lactamase, an enzyme that confers antibiotic resistance to bacteria, is stabilizing but the precise mechanism remains unclear. Here, we employ Markov state models (MSMs) to uncover how M182T shifts the distribution of different structures that TEM adopts. We find that M182T stabilizes a helix that is a key component of a domain interface. We then predict the effects of other mutations, including a novel stabilizing mutation, and experimentally test our predictions using a combination of stability measurements, crystallography, NMR, and in vivo measurements of bacterial fitness. We expect our insights and methodology to provide a valuable foundation for protein design. ■ INTRODUCTIONStudying the evolution of antibiotic resistance has provided many insights into how proteins acquire new functions, but the mechanistic basis for how mutations alter a protein's activity and stability often remains unclear. For example, studying how bacteria evolve variants of TEM β-lactamase that confer resistance to new antibiotics by degrading these drugs has revealed that many of the mutations that give rise to new functions are destabilizing. Therefore, it is common for proteins to acquire one or more mutations that alter their function and then to acquire additional mutations that restore stability. 1M182T is one such stabilizing mutation in TEM, and it has appeared in numerous clinical isolates and directed evolution experiments.2−4 This substitution occurs far from the active site ( Figure 1A) and, on its own, has little effect on TEM's activity. It is often called a global suppressor because of its ability to counterbalance the destabilizing effects of a wide variety of other substitutions that do alter TEM's activity.3 Despite over two decades of work on this variant, the mechanism of stabilization by M182T is not understood well enough to predict new stabilizing mutations. Elucidating the mechanism underlying this stabilization would provide a basis for predicting other global suppressors and eventually developing quantitative design principles.A mechanistic understanding of how M182T stabilizes TEM remains elusive because of a lack of methods that provide both a detailed structural model of the relevant species and their relative populations. Spectroscopic studies have revealed that TEM-1, which we will refer to as wild-type TEM, populates at least three states at equilibrium: a native state (N), an intermediate state (I), and an unfolded state (U).5 Introducing the M182T substitution appears to reduce the number of equilibrium states to two. 4 However, there is debate over whether this results from M182T stabilizing the native state or destabilizing the intermediate. Moreove...

A structure-based deep learning framework for protein engineering

“…Using our model, we interrogated the microenvironment surrounding Met 182 in TEM-1 β-lactamase where a Met to Thr substitution results in global stabilization. We identified key backbone atoms which favor the Thr substitution in agreement with the findings of previous MD work at this locus 16 (see Supporting Results and SI Fig. 12).…”

Dynamics Rationalize Proteolytic Susceptibility of the Major Birch Pollen Allergen Bet v 1

“…The resulting cleavage data in turn can be used to refine the models and enhance prediction accuracy. Similar strategies have already demonstrated the reliability of predictions from MD simulations combined with MSM analysis on the impact of point mutations on protein stability (Zimmerman et al, 2018). Given the tremendous clinical relevance of allergen proteins we thus suggest physicsbased computational models as promising tools for the design recombinant proteins for allergen-specific immunotherapy (Curin et al, 2018).…”

“…The construction of the MSM allows to quantify thermodynamic and kinetic properties of each ensemble without the intrinsic bias resulting from the seeding process (Bowman et al, 2013;Kohlhoff et al, 2014). Similar workflows have already been proven to be extremely efficient and highly reliable (Noe et al, 2009;Nedialkova et al, 2014;Biswas et al, 2018;Fernandez-Quintero et al, 2018;Sun et al, 2018;Zimmerman et al, 2018).…”

pH-Induced Local Unfolding of the Phl p 6 Pollen Allergen From cpH-MD

“…This approach removes any bias stemming from a different number of starting clusters in the seeding process and allows us to capture accurate thermodynamic and kinetic properties of the sampled dynamic processes (Bowman et al, 2013;Kohlhoff et al, 2014). Similar approaches have been reported previously (Noé et al, 2009;Nedialkova et al, 2014;Biswas et al, 2018;Sun et al, 2018;Zimmerman et al, 2018;Fernández-Quintero et al, 2019a,b, 2020Kahler et al, 2020;Kamenik et al, 2020).…”