Contribution of Cotranslational Folding Defects to Membrane Protein Homeostasis

Artificial Homeostasis Systems Based on Feedback Reaction Networks: Design Principles and Future Promises

Feedback-controlled chemical reaction networks (FCRNs) are indispensable for various biological processes, such as cellular mechanisms, patterns, and signaling pathways. Through the intricate interplay of many feedback loops (FLs), FCRNs maintain a stable internal cellular environment. Currently, creating minimalistic synthetic cells is the long-term objective of systems chemistry, which is motivated by such natural integrity. The design, kinetic optimization, and analysis of FCRNs to exhibit functions akin to those of a cell still pose significant challenges. Indeed, reaching synthetic homeostasis is essential for engineering synthetic cell components. However, maintaining homeostasis in artificial systems against various agitations is a difficult task. Several biological events can provide us with guidelines for a conceptual understanding of homeostasis, which can be further applicable in designing artificial synthetic systems. In this regard, we organize our review with artificial homeostasis systems driven by FCRNs at different length scales, including homogeneous, compartmentalized, and soft material systems. First, we stretch a quick overview of FCRNs in different molecular and supramolecular systems, which are the essential toolbox for engineering different nonlinear functions and homeostatic systems. Moreover, the existing history of synthetic homeostasis in chemical and material systems and their advanced functions with self-correcting, and regulating properties are also emphasized.

General trends in the effects of VX-661 and VX-445 on the plasma membrane expression of clinical CFTR variants

Cystic fibrosis (CF) is caused by mutations that compromise the expression and/or function of the cystic fibrosis transmembrane conductance regulator (CFTR) chloride channel. Most people with CF harbor a common misfolded variant (ΔF508) that can be partially rescued by therapeutic "correctors" that restore its expression. Nevertheless, many other CF variants are insensitive to correctors. Using deep mutational scanning, we quantitatively compare the effects of two correctors on the plasma membrane expression of 129 CF variants. Though structural calculations suggest corrector binding provides similar stabilization to most variants, it's those with intermediate expression and mutations near corrector binding pockets that exhibit the greatest response. Deviations in sensitivity appear to depend on the degree of variant destabilization and the timing of misassembly. Combining correctors appears to rescue more variants by doubling the binding energy and stabilizing distinct cotranslational folding transitions. These results provide an overview of rare CF variant expression and establish new tools for precision pharmacology.

Molecular Basis for Variations in the Sensitivity of Pathogenic Rhodopsin Variants to 9-cis-Retinal

Over 100 mutations in the rhodopsin gene have been linked to a spectrum of retinopathies that include retinitis pigmentosa and congenital stationary night blindness. Though most of these variants exhibit a loss of function, the molecular defects caused by these underlying mutations vary considerably. In this work, we utilize deep mutational scanning to quantitatively compare the plasma membrane expression of 123 known pathogenic rhodopsin variants in the presence and absence of the stabilizing cofactor 9-cis-retinal. We identify 69 retinopathy variants, including 20 previously uncharacterized variants, that exhibit diminished plasma membrane expression in HEK293T cells. Of these apparent class II variants, 67 exhibit a measurable increase in expression in the presence of 9-cis-retinal. However, the magnitude of the response to this molecule varies considerably across this spectrum of mutations. Evaluation of the observed shifts relative to thermodynamic estimates for the coupling between binding and folding suggests underlying differences in stability constrains the magnitude of their response to retinal. Nevertheless, estimates from computational modeling suggest that many of the least sensitive variants also directly compromise binding. Finally, we evaluate the functional properties of three previous uncharacterized, retinal-sensitive variants (ΔN73, S131P, and R135G) and show that two of these retain residual function in vitro. Together, our results provide a comprehensive experimental characterization of the proteostatic properties of retinopathy variants and their response to retinal.

Towards generalizable predictions for G protein-coupled receptor variant expression

Missense mutations that compromise the plasma membrane expression (PME) of integral membrane proteins are the root cause of numerous genetic diseases. Differentiation of this class of mutations from those that specifically modify the activity of the folded protein has proven useful for the development and targeting of precision therapeutics. Nevertheless, it remains challenging to predict the effects of mutations on the stability and/ or expression of membrane proteins. In this work, we utilize deep mutational scanning data to train a series of artificial neural networks to predict the PME of transmembrane domain variants of G protein-coupled receptors from structural and/ or evolutionary features. We show that our best-performing network, which we term the PME predictor, can recapitulate mutagenic trends within rhodopsin and can differentiate pathogenic transmembrane domain variants that cause it to misfold from those that compromise its signaling. This network also generates statistically significant predictions for the relative PME of transmembrane domain variants for another class A G protein-coupled receptor (β₂ adrenergic receptor) but not for an unrelated voltage-gated potassium channel (KCNQ1). Notably, our analyses of these networks suggest structural features alone are generally sufficient to recapitulate the observed mutagenic trends. Moreover, our findings imply that networks trained in this manner may be generalizable to proteins that share a common fold. Implications of our findings for the design of mechanistically specific genetic predictors are discussed.

Untangling the complexity of membrane protein folding

Delineating the folding steps of helical-bundle membrane proteins has been a challenging task. Many questions remain unanswered, including the conformation and stability of the states populated during folding, the shape of the energy barriers between the states, and the role of lipids as a solvent in mediating the folding. Recently, theoretical frames have matured to a point that permits detailed dissection of the folding steps, and advances in experimental techniques at both single-molecule and ensemble levels enable selective modulation of specific steps for quantitative determination of the folding energy landscapes. We also discuss how lipid molecules would play an active role in shaping the folding energy landscape of membrane proteins, and how folding of multi-domain membrane proteins can be understood based on our current knowledge. We conclude this review by offering an outlook for emerging questions in the study of membrane protein folding.

Molecular basis for the evolved instability of a human G-protein coupled receptor

Membrane proteins are prone to misfolding and degradation. This is particularly true for mammalian forms of the gonadotropin-releasing hormone receptor (GnRHR). Although they function at the plasma membrane, mammalian GnRHRs accumulate within the secretory pathway. Their apparent instability is believed to have evolved through selection for attenuated GnRHR activity. Nevertheless, the molecular basis of this adaptation remains unclear. We show that adaptation coincides with a C-terminal truncation that compromises the translocon-mediated membrane integration of its seventh transmembrane domain (TM7). We also identify a series of polar residues in mammalian GnRHRs that compromise the membrane integration of TM2 and TM6. Reverting a lipid-exposed polar residue in TM6 to an ancestral hydrophobic residue restores expression with no impact on function. Evolutionary trends suggest variations in the polarity of this residue track with reproductive phenotypes. Our findings suggest that the marginal energetics of cotranslational folding can be exploited to tune membrane protein fitness.

Integral membrane proteins are prone to misfolding, especially mammalian gonadotropin-releasing hormone receptors (GnRHRs). Chamness et al. show that the evolved instability of mammalian GnRHRs stems from adaptive modifications that disrupt translocon-mediated membrane integration, suggesting that membrane protein misfolding can be exploited to tune fitness.

Systematic Profiling of Temperature- and Retinal-Sensitive Rhodopsin Variants by Deep Mutational Scanning

Membrane protein variants with diminished conformational stability often exhibit enhanced cellular expression at reduced growth temperatures. The expression of “temperature-sensitive” variants is also typically sensitive to corrector molecules that bind and stabilize the native conformation. There are many examples of temperature-sensitive rhodopsin variants, the misfolding of which is associated with the molecular basis of retinitis pigmentosa. In this work, we employ deep mutational scanning to compare the effects of reduced growth temperature and 9-cis-retinal, an investigational corrector, on the plasma membrane expression of 700 rhodopsin variants in HEK293T cells. We find that the change in expression at reduced growth temperatures correlates with the response to 9-cis-retinal among variants bearing mutations within a hydrophobic transmembrane domain (TM2). The most sensitive variants appear to disrupt a native helical kink within this transmembrane domain. By comparison, mutants that alter the structure of a polar transmembrane domain (TM7) exhibit weaker responses to temperature and retinal that are poorly correlated. Statistical analyses suggest that this observed insensitivity cannot be attributed to a single variable, but likely arises from the composite effects of mutations on the energetics of membrane integration, the stability of the native conformation, and the integrity of the retinal-binding pocket. Finally, we show that the characteristics of purified temperature- and retinal-sensitive variants suggest that the proteostatic effects of retinal may be manifested during translation and cotranslational folding. Together, our findings highlight several biophysical constraints that appear to influence the sensitivity of genetic variants to temperature and small-molecule correctors.

Disease-linked supertrafficking of a potassium channel

Gain-of-function (GOF) mutations in the voltage-gated potassium channel subfamily Q member 1 (KCNQ1) can induce cardiac arrhythmia. In this study, it was tested whether any of the known human GOF disease mutations in KCNQ1 act by increasing the amount of KCNQ1 that reaches the cell surface-"supertrafficking." Seven of the 15 GOF mutants tested were seen to surface traffic more efficiently than the WT channel. Among these, we found that the levels of R231C KCNQ1 in the plasma membrane were fivefold higher than the WT channel. This was shown to arise from the combined effects of enhanced efficiency of translocon-mediated membrane integration of the S4 voltage-sensor helix and from enhanced post-translational folding/trafficking related to the energetic linkage of C231 with the V129 and F166 side chains. Whole-cell electrophysiology recordings confirmed that R231C KCNQ1 in complex with the voltage-gated potassium channel-regulatory subfamily E member 1 not only exhibited constitutive conductance but also revealed that the single-channel activity of this mutant is only 20% that of WT. The GOF phenotype associated with R231C therefore reflects the effects of supertrafficking and constitutive channel activation, which together offset reduced channel activity. These investigations show that membrane protein supertrafficking can contribute to human disease.

C-terminal tail length guides insertion and assembly of membrane proteins

A large number of newly synthesized membrane proteins in the endoplasmic reticulum (ER) are assembled into multiprotein complexes, but little is known about the mechanisms required for assembly membrane proteins. It has been suggested that membrane chaperones might exist, akin to the molecular chaperones that stabilize and direct the assembly of soluble protein complexes, but the mechanisms by which these proteins would bring together membrane protein components is unclear. Here, we have identified that the tail length of the C-terminal transmembrane domains (C-TMDs) determines efficient insertion and assembly of membrane proteins in the ER. We found that membrane proteins with C-TMD tails shorter than ∼60 amino acids are poorly inserted into the ER membrane, which suggests that translation is terminated before they are recognized by the Sec61 translocon for insertion. These C-TMDs with insufficient hydrophobicity are post-translationally recognized and retained by the Sec61 translocon complex, providing a time window for efficient assembly with TMDs from partner proteins. Retained TMDs that fail to assemble with their cognate TMDs are slowly translocated into the ER lumen and are recognized by the ER-associated degradation (ERAD) pathway for removal. In contrast, C-TMDs with sufficient hydrophobicity or tails longer than ∼80 residues are quickly released from the Sec61 translocon into the membrane or the ER lumen, resulting in inefficient assembly with partner TMDs. Thus, our data suggest that C-terminal tails harbor crucial signals for both the insertion and assembly of membrane proteins.

Predicting changes in protein thermodynamic stability upon point mutation with deep 3D convolutional neural networks

Predicting mutation-induced changes in protein thermodynamic stability (ΔΔG) is of great interest in protein engineering, variant interpretation, and protein biophysics. We introduce ThermoNet, a deep, 3D-convolutional neural network (3D-CNN) designed for structure-based prediction of ΔΔGs upon point mutation. To leverage the image-processing power inherent in CNNs, we treat protein structures as if they were multi-channel 3D images. In particular, the inputs to ThermoNet are uniformly constructed as multi-channel voxel grids based on biophysical properties derived from raw atom coordinates. We train and evaluate ThermoNet with a curated data set that accounts for protein homology and is balanced with direct and reverse mutations; this provides a framework for addressing biases that have likely influenced many previous ΔΔG prediction methods. ThermoNet demonstrates performance comparable to the best available methods on the widely used Ssym test set. In addition, ThermoNet accurately predicts the effects of both stabilizing and destabilizing mutations, while most other methods exhibit a strong bias towards predicting destabilization. We further show that homology between Ssym and widely used training sets like S2648 and VariBench has likely led to overestimated performance in previous studies. Finally, we demonstrate the practical utility of ThermoNet in predicting the ΔΔGs for two clinically relevant proteins, p53 and myoglobin, and for pathogenic and benign missense variants from ClinVar. Overall, our results suggest that 3D-CNNs can model the complex, non-linear interactions perturbed by mutations, directly from biophysical properties of atoms.

Cotranslational folding stimulates programmed ribosomal frameshifting in the alphavirus structural polyprotein

Viruses maximize their genetic coding capacity through a variety of biochemical mechanisms, including programmed ribosomal frameshifting (PRF), which facilitates the production of multiple proteins from a single mRNA transcript. PRF is typically stimulated by structural elements within the mRNA that generate mechanical tension between the transcript and ribosome. However, in this work, we show that the forces generated by the cotranslational folding of the nascent polypeptide chain can also enhance PRF. Using an array of biochemical, cellular, and computational techniques, we first demonstrate that the Sindbis virus structural polyprotein forms two competing topological isomers during its biosynthesis at the ribosome-translocon complex. We then show that the formation of one of these topological isomers is linked to PRF. Coarse-grained molecular dynamics simulations reveal that the translocon-mediated membrane integration of a transmembrane domain upstream from the ribosomal slip site generates a force on the nascent polypeptide chain that scales with observed frameshifting. Together, our results indicate that cotranslational folding of this viral protein generates a tension that stimulates PRF. To our knowledge, this constitutes the first example in which the conformational state of the nascent polypeptide chain has been linked to PRF. These findings raise the possibility that, in addition to RNA-mediated translational recoding, a variety of cotranslational folding or binding events may also stimulate PRF.

Probing biophysical sequence constraints within the transmembrane domains of rhodopsin by deep mutational scanning

Membrane proteins must balance the sequence constraints associated with folding and function against the hydrophobicity required for solvation within the bilayer. We recently found the expression and maturation of rhodopsin are limited by the hydrophobicity of its seventh transmembrane domain (TM7), which contains polar residues that are essential for function. On the basis of these observations, we hypothesized that rhodopsin's expression should be less tolerant of mutations in TM7 relative to those within hydrophobic TM domains. To test this hypothesis, we used deep mutational scanning to compare the effects of 808 missense mutations on the plasma membrane expression of rhodopsin in HEK293T cells. Our results confirm that a higher proportion of mutations within TM7 (37%) decrease rhodopsin's plasma membrane expression relative to those within a hydrophobic TM domain (TM2, 25%). These results in conjunction with an evolutionary analysis suggest solvation energetics likely restricts the evolutionary sequence space of polar TM domains.

Lipid-Assisted Membrane Protein Folding and Topogenesis

Due to the heterogenous lipid environment in which integral membrane proteins are embedded, they should follow a set of assembly rules, which govern transmembrane protein folding and topogenesis accordingly to a given lipid profile. Recombinant strains of bacteria have been engineered to have different membrane phospholipid compositions by molecular genetic manipulation of endogenous and foreign genes encoding lipid biosynthetic enzymes. Such strains provide a means to investigate the in vivo role of lipids in many different aspects of membrane function, folding and biogenesis. In vitro and in vivo studies established a function of lipids as molecular chaperones and topological determinants specifically assisting folding and topogenesis of membrane proteins. These results led to the extension of the Positive Inside Rule to Charge Balance Rule, which incorporates a role for lipid-protein interactions in determining membrane protein topological organization at the time of initial membrane insertion and dynamically after initial assembly. Membrane protein topogenesis appears to be a thermodynamically driven process in which lipid-protein interactions affect the potency of charged amino acid residues as topological signals. Dual topology for a membrane protein can be established during initial assembly where folding intermediates in multiple topological conformations are in rapid equilibrium (thus separated by a low activation energy), which is determined by the lipid environment. Post-assembly changes in lipid composition or post-translational modifications can trigger a reorganization of protein topology by inducing destabilization and refolding of a membrane protein. The lipid-dependent dynamic nature of membrane protein organization provides a novel means of regulating protein function.

Harmonizing Experimental Data with Modeling to Predict Membrane Protein Insertion in Yeast

Membrane proteins must adopt their proper topologies within biological membranes, but achieving the correct topology is compromised by the presence of marginally hydrophobic transmembrane helices (TMHs). In this study, we report on a new model membrane protein in yeast that harbors two TMHs fused to an unstable nucleotide-binding domain. Because the second helix (TMH2) in this reporter has an unfavorable predicted free energy of insertion, we employed established methods to generate variants that alter TMH2 insertion free energy. We first found that altering TMH2 did not significantly affect the extent of protein degradation by the cellular quality control machinery. Next, we correlated predicted insertion free energies from a knowledge-based energy scale with the measured apparent free energies of TMH2 insertion. Although the predicted and apparent insertion energies showed a similar trend, the predicted free-energy changes spanned an unanticipated narrow range. By instead using a physics-based model, we obtained a broader range of free energies that agreed considerably better with the magnitude of the experimentally derived values. Nevertheless, some variants still inserted better in yeast than predicted from energy-based scales. Therefore, molecular dynamics simulations were performed and indicated that the corresponding mutations induced conformational changes within TMH2, which altered the number of stabilizing hydrogen bonds. Together, our results offer insight into the ability of the cellular quality control machinery to recognize conformationally distinct misfolded topomers, provide a model to assess TMH insertion in vivo, and indicate that TMH insertion energy scales may be limited depending on the specific protein and the mutation present.