Desmosome architecture derived from molecular dynamics simulations and cryo-electron tomography.

Desmosomes are cell-cell junctions that link tissue cells experiencing intense mechanical stress. Although the structure of the desmosomal cadherins is known, the desmosome architecture-which is essential for mediating numerous functions-remains elusive. Here, we recorded cryo-electron tomograms (cryo-ET) in which individual cadherins can be discerned; they appear variable in shape, spacing, and tilt with respect to the membrane. The resulting sub-tomogram average reaches a resolution of ∼26 Å, limited by the inherent flexibility of desmosomes. To address this challenge typical of dynamic biological assemblies, we combine sub-tomogram averaging with atomistic molecular dynamics (MD) simulations. We generate models of possible cadherin arrangements and perform an in silico screening according to biophysical and structural properties extracted from MD simulation trajectories. We find a truss-like arrangement of cadherins that resembles the characteristic footprint seen in the electron micrograph. The resulting model of the desmosomal architecture explains their unique biophysical properties and strength.

RNA-binding and prion domains: the Yin and Yang of phase separation.

Proteins and RNAs assemble in membrane-less organelles that organize intracellular spaces and regulate biochemical reactions. The ability of proteins and RNAs to form condensates is encoded in their sequences, yet it is unknown which domains drive the phase separation (PS) process and what are their specific roles. Here, we systematically investigated the human and yeast proteomes to find regions promoting condensation. Using advanced computational methods to predict the PS propensity of proteins, we designed a set of experiments to investigate the contributions of Prion-Like Domains (PrLDs) and RNA-binding domains (RBDs). We found that one PrLD is sufficient to drive PS, whereas multiple RBDs are needed to modulate the dynamics of the assemblies. In the case of stress granule protein Pub1 we show that the PrLD promotes sequestration of protein partners and the RBD confers liquid-like behaviour to the condensate. Our work sheds light on the fine interplay between RBDs and PrLD to regulate formation of membrane-less organelles, opening up the avenue for their manipulation.

Assembly of Proteins by Free RNA during the Early Phase of Proteostasis Stress.

At any stage of their lifecycle, mRNAs are coated by specialized proteins. One of few circumstances when free mRNA appears in the cytosol is the disassembly of polysomes during the stress-induced shutdown of protein synthesis. Using quantitative mass spectrometry, we sought to identify the free RNA-interacting cellular machinery in heat-shocked mammalian cells. Free RNA-associated proteins displayed higher disorder and larger size, which supports the role of multivalent interactions during the initial phase of the association with RNAs during stress. Structural features of the free RNA interactors defined them as a subset of RNA-binding proteins. The interaction between these assembled proteins in vivo required RNA. Reconstitution of the association process in vitro indicated a multimolecular basis for increased binding to RNA upon heat shock in the cytosol. Our study represents a step toward understanding how free RNA is processed in the cytosol during proteostasis stress.

Recognition of enzymes lacking bound cofactor by protein quality control.

Protein biogenesis is tightly linked to protein quality control (PQC). The role of PQC machinery in recognizing faulty polypeptides is becoming increasingly understood. Molecular chaperones and cytosolic and vacuolar degradation systems collaborate to detect, repair, or hydrolyze mutant, damaged, and mislocalized proteins. On the other hand, the contribution of PQC to cofactor binding-related enzyme maturation remains largely unexplored, although the loading of a cofactor represents an all-or-nothing transition in regard to the enzymatic function and thus must be surveyed carefully. Combining proteomics and biochemical analysis, we demonstrate here that cells are able to detect functionally immature wild-type enzymes. We show that PQC-dedicated ubiquitin ligase C-terminal Hsp70-interacting protein (CHIP) recognizes and marks for degradation not only a mutant protein but also its wild-type variant as long as the latter remains cofactor free. A distinct structural feature, the protruding C-terminal tail, which appears in both the mutant and wild-type polypeptides, contributes to recognition by CHIP. Our data suggest that relative insufficiency of apoprotein degradation caused by cofactor shortage can increase amyloidogenesis and aggravate protein aggregation disorders.

Polylysine is a Proteostasis Network-Engaging Structural Determinant.

C-terminal polylysine (PL) can be synthesized from the polyadenine tail of prematurely cleaved mRNAs or when a read-though of a stop codon happens. Due to the highly positive charge, PL stalls in the electrostatically negative ribosomal exit channel. The stalled polypeptide recruits the Ribosome-associated quality control (RQC) complex which processes and extracts the nascent chain. Dysfunction of the RQC leads to the accumulation of PL-tagged proteins, induction of a stress response, and cellular toxicity. Not much is known about the PL-specific aspect of protein quality control. Using quantitative mass spectrometry, we uncovered the post-ribosomal PL-processing machinery in human cytosol. It encompasses key cytosolic complexes of the proteostasis network, such as chaperonin TCP-1 ring complexes (TRiC) and half-capped 19S-20S proteasomes. Furthermore, we found that the nuclear transport machinery associates with PL, which suggests a novel mechanism by which faulty proteins can be compartmentalized in the cell. The enhanced nuclear import of a PL-tagged polypeptide confirmed this implication, which leads to questions regarding the biological rationale behind it.

The structural and functional roles of the flavin cofactor FAD in mammalian cryptochromes.

The importance of circadian rhythms in human health and disease calls for a thorough understanding of the underlying molecular machinery, including its key components, the flavin adenine dinucleotide (FAD)-containing flavoproteins cryptochrome 1 and 2. Contrary to their Drosophila counterparts, mammalian cryptochromes are direct suppressors of circadian transcription and act independently of light. Light-independence poses the question regarding the role of the cofactor FAD in mammalian cryptochromes. The weak binding of the cofactor in vitro argues against its relevance and might be a functionless evolutionary remnant. From the other side, the FAD-binding pocket constitutes the part of mammalian cryptochromes directly related to their ubiquitylation by the ubiquitin ligase Fbxl3 and is the target for protein-stabilizing small molecules. Increased supplies of FAD stabilize cryptochromes in cell culture, and the depletion of the FAD precursor riboflavin with simultaneous knock-down of riboflavin kinase affects the expression of circadian genes in mice. This review presents the classical and more recent studies in the field, which help to comprehend the role of FAD for the stability and function of mammalian cryptochromes.

Mammalian Flavoproteome Analysis Using Label-Free Quantitative Mass Spectrometry.

Human flavin cofactor-containing enzymes constitute a small, but highly important flavoproteome. Its stability is required to ensure key metabolic functions, such as oxidative phosphorylation and beta-oxidation of fatty acid. Flavoproteome disfunction due to mutations of individual proteins or because of the lack of FMN and FAD precursor riboflavin (vitamin B2) results in clinically relevant abnormal cellular states and diseases. Current technical possibilities in the field of the quantitative mass spectrometry of proteins allow studying the flavoproteome changes under different stress conditions, including the deficiency of vitamin B2. The biological readouts of flavoenzyme destabilization, such as protein degradation and aggregation, provide important insights into the molecular mechanisms of metabolic adaptation to nutrient deficiency. The proteomic-scale studies of protein stability have significant novelty potential in basic and applied biomedical research.

The protective role of m1A during stress-induced granulation.

Post-transcriptional methylation of N6-adenine and N1-adenine can affect transcriptome turnover and translation. Furthermore, the regulatory function of N6-methyladenine (m6A) during heat shock has been uncovered, including the enhancement of the phase separation potential of RNAs. In response to acute stress, e.g. heat shock, the orderly sequestration of mRNAs in stress granules (SGs) is considered important to protect transcripts from the irreversible aggregation. Until recently, the role of N1-methyladenine (m1A) on mRNAs during acute stress response remains largely unknown. Here we show that the methyltransferase complex TRMT6/61A, which generates the m1A tag, is involved in transcriptome protection during heat shock. Our bioinformatics analysis indicates that occurrence of the m1A motif is increased in mRNAs known to be enriched in SGs. Accordingly, the m1A-generating methyltransferase TRMT6/61A accumulated in SGs and mass spectrometry confirmed enrichment of m1A in the SG RNAs. The insertion of a single methylation motif in the untranslated region of a reporter RNA leads to more efficient recovery of protein synthesis from that transcript after the return to normal temperature. Our results demonstrate far-reaching functional consequences of a minimal RNA modification on N1-adenine during acute proteostasis stress.

Differential substrate specificity of group I and group II chaperonins in the archaeon Methanosarcina mazei.

Chaperonins are macromolecular machines that assist in protein folding. The archaeon Methanosarcina mazei has acquired numerous bacterial genes by horizontal gene transfer. As a result, both the bacterial group I chaperonin, GroEL, and the archaeal group II chaperonin, thermosome, coexist. A proteome-wide analysis of chaperonin interactors was performed to determine the differential substrate specificity of GroEL and thermosome. At least 13% of soluble M. mazei proteins interact with chaperonins, with the two systems having partially overlapping substrate sets. Remarkably, chaperonin selectivity is independent of phylogenetic origin and is determined by distinct structural and biochemical features of proteins. GroEL prefers well-conserved proteins with complex alpha/beta domains. In contrast, thermosome substrates comprise a group of faster-evolving proteins and contain a much wider range of different domain folds, including small all-alpha and all-beta modules, and a greater number of large multidomain proteins. Thus, the group II chaperonins may have facilitated the evolution of the highly complex proteomes characteristic of eukaryotic cells.

Flavin dependency undermines proteome stability, lipid metabolism and cellular proliferation during vitamin B2 deficiency.

Tumor cells adapt their metabolism to meet the energetic and anabolic requirements of high proliferation and invasiveness. The metabolic addiction has motivated the development of therapies directed at individual biochemical nodes. However, currently there are few possibilities to target multiple enzymes in tumors simultaneously. Flavin-containing enzymes, ca. 100 proteins in humans, execute key biotransformations in mammalian cells. To expose metabolic addiction, we inactivated a substantial fraction of the flavoproteome in melanoma cells by restricting the supply of the FMN and FAD precursor riboflavin, the vitamin B2. Vitamin B2 deficiency affected stability of many polypeptides and thus resembled the chaperone HSP90 inhibition, the paradigmatic multiple-target approach. In support of this analogy, flavin-depleted proteins increasingly associated with a number of proteostasis network components, as identified by the mass spectrometry analysis of the FAD-free NQO1 aggregates. Proteome-wide analysis of the riboflavin-starved cells revealed a profound inactivation of the mevalonate pathway of cholesterol synthesis, which underlines the manifold cellular vulnerability created by the flavoproteome inactivation. Cell cycle-arrested tumor cells became highly sensitive to alkylating chemotherapy. Our data suggest that the flavoproteome is well suited to design synthetic lethality protocols combining proteostasis manipulation and metabolic reprogramming.

RNA structure drives interaction with proteins.

The combination of high-throughput sequencing and in vivo crosslinking approaches leads to the progressive uncovering of the complex interdependence between cellular transcriptome and proteome. Yet, the molecular determinants governing interactions in protein-RNA networks are not well understood. Here we investigated the relationship between the structure of an RNA and its ability to interact with proteins. Analysing in silico, in vitro and in vivo experiments, we find that the amount of double-stranded regions in an RNA correlates with the number of protein contacts. This relationship -which we call structure-driven protein interactivity- allows classification of RNA types, plays a role in gene regulation and could have implications for the formation of phase-separated ribonucleoprotein assemblies. We validate our hypothesis by showing that a highly structured RNA can rearrange the composition of a protein aggregate. We report that the tendency of proteins to phase-separate is reduced by interactions with specific RNAs.

Structure and dynamics of a partially folded protein are decoupled from its mechanism of aggregation.

A common strategy to study the mechanism of amyloid formation is the characterization of the structure and dynamics of the precursor state, which is in most cases a partially folded protein. Here we investigated the highly dynamic conformational state formed by the protein domain HypF-N at low pH, before aggregation, using fluorescence, circular dichroism, and NMR spectroscopies. The NMR analysis allowed us, in particular, to identify the regions of the sequence that form hydrophobic interactions and adopt an alpha-helical secondary structure in the pH-denatured ensemble. To understand the role that this residual structure plays in the aggregation of this protein, we probed the mechanism of aggregation using protein engineering experiments and thus identified the regions of the sequence of HypF-N that play a critical role in the conversion of this dynamic state into thioflavin T-binding and beta-sheet containing protofibrils. The combination of these two complementary approaches revealed that the aggregation of pH-denatured HypF-N is not structure-dependent, meaning that it is not driven by the regions of the protein that are either less or more protected in the initial partially folded state. It is, by contrast, promoted by discrete protein regions that have the highest intrinsic propensity to aggregate because of their physicochemical properties.

CHIP as a membrane-shuttling proteostasis sensor.

Cells respond to protein misfolding and aggregation in the cytosol by adjusting gene transcription and a number of post-transcriptional processes. In parallel to functional reactions, cellular structure changes as well; however, the mechanisms underlying the early adaptation of cellular compartments to cytosolic protein misfolding are less clear. Here we show that the mammalian ubiquitin ligase C-terminal Hsp70-interacting protein (CHIP), if freed from chaperones during acute stress, can dock on cellular membranes thus performing a proteostasis sensor function. We reconstituted this process in vitro and found that mainly phosphatidic acid and phosphatidylinositol-4-phosphate enhance association of chaperone-free CHIP with liposomes. HSP70 and membranes compete for mutually exclusive binding to the tetratricopeptide repeat domain of CHIP. At new cellular locations, access to compartment-specific substrates would enable CHIP to participate in the reorganization of the respective organelles, as exemplified by the fragmentation of the Golgi apparatus (effector function).

The regions of the sequence most exposed to the solvent within the amyloidogenic state of a protein initiate the aggregation process.

Formation of misfolded aggregates is an essential part of what proteins can do. The process of protein aggregation is central to many human diseases and any aggregating event needs to be prevented within a cell and in protein design. In order to aggregate, a protein needs to unfold its native state, at least partially. The conformational state that is prone to aggregate is difficult to study, due to its aggregating potential and heterogeneous nature. Here, we use a systematic approach of limited proteolysis, in combination with electrospray ionisation mass spectrometry, to investigate the regions that are most flexible and solvent-exposed within the native, ligand-bound and amyloidogenic states of muscle acylphosphatase (AcP), a protein previously shown to form amyloid fibrils in the presence of trifluoroethanol. Seven proteases with different degrees of specificity have been used for this purpose. Following exposure to the aggregating conditions, a number of sites along the sequence of AcP become susceptible to proteolytic digestion. The pattern of proteolytic cleavages obtained under these conditions is considerably different from that of the native and ligand-bound conformations and includes a portion within the N-terminal tail of the protein (residues 6-7), the region of the sequence 18-23 and the position 94 near the C terminus. There is a significant overlap between the regions of the sequence found to be solvent-exposed from the present study and those previously identified to be critical in the rate-determining steps of aggregation from protein engineering approaches. This indicates that a considerable degree of solvent exposure is a feature of the portions of a protein that initiate the process of aggregation.

Comparison of the folding processes of distantly related proteins. Importance of hydrophobic content in folding.

The N-terminal domain of HypF from Escherichia coli (HypF-N) is a 91 residue protein module sharing the same folding topology and a significant sequence identity with two extensively studied human proteins, muscle and common-type acylphosphatases (mAcP and ctAcP). With the aim of learning fundamental aspects of protein folding from the close comparison of so similar proteins, the folding process of HypF-N has been studied using stopped-flow fluorescence. While mAcP and ctAcP fold in a two-state fashion, HypF-N was found to collapse into a partially folded intermediate before reaching the fully folded conformation. Formation of a burst-phase intermediate is indicated by the roll over in the Chevron plot at low urea concentrations and by the large jump of intrinsic and 8-anilino-1-naphtalenesulphonic acid-derived fluorescence immediately after removal of denaturant. Furthermore, HypF-N was found to fold rapidly with a rate constant that is approximately two and three orders of magnitudes faster than ctAcP and mAcP, respectively. Differences between the bacterial protein and the two human counterparts were also found as to the involvement of proline isomerism in their respective folding processes. The results clearly indicate that features that are often thought to be relevant in protein folding are not highly conserved in the evolution of the acylphosphatase superfamily. The large difference in folding rate between mAcP and HypF-N cannot be entirely accounted for by the difference in relative contact order or related topological metrics. The analysis shows that the higher folding rate of HypF-N is in part due to the relatively high hydrophobic content of this protein. This conclusion, which is also supported by the highly significant correlation found between folding rate and hydrophobic content within a group of proteins displaying the topology of HypF-N and AcPs, suggests that the average hydrophobicity of a protein sequence is an important determinant of its folding rate.

DnaK functions as a central hub in the E. coli chaperone network.

Cellular chaperone networks prevent potentially toxic protein aggregation and ensure proteome integrity. Here, we used Escherichia coli as a model to understand the organization of these networks, focusing on the cooperation of the DnaK system with the upstream chaperone Trigger factor (TF) and the downstream GroEL. Quantitative proteomics revealed that DnaK interacts with at least ~700 mostly cytosolic proteins, including ~180 relatively aggregation-prone proteins that utilize DnaK extensively during and after initial folding. Upon deletion of TF, DnaK interacts increasingly with ribosomal and other small, basic proteins, while its association with large multidomain proteins is reduced. DnaK also functions prominently in stabilizing proteins for subsequent folding by GroEL. These proteins accumulate on DnaK upon GroEL depletion and are then degraded, thus defining DnaK as a central organizer of the chaperone network. Combined loss of DnaK and TF causes proteostasis collapse with disruption of GroEL function, defective ribosomal biogenesis, and extensive aggregation of large proteins.

Low-level expression of a folding-incompetent protein in Escherichia coli: search for the molecular determinants of protein aggregation in vivo.

Aggregation of peptides and proteins into insoluble amyloid fibrils or related intracellular inclusions is the hallmark of many degenerative diseases, including Alzheimer's disease, Parkinson's disease, and various forms of amyloidosis. In spite of the considerable progress carried out in vitro in elucidating the molecular determinants of the conversion of purified and isolated proteins into amyloid fibrils, very little is known on factors governing this process in the complex environment of living organisms. Taking advantage of increasing evidence that bacterial inclusion bodies consist of amyloid-like aggregates, we have expressed in Escherichia coli both wild type and 21 single-point mutants of the N-terminal domain of the E. coli protein HypF. All variants were expressed as folding-incompetent units in a controlled manner, at low and comparable levels. Their solubilities were measured by quantifying the protein amount contained in the soluble and insoluble fractions by Western blot analysis. A significant negative correlation was found between the solubility of the variants in E. coli and their intrinsic propensity to form amyloid fibrils, predicted using an algorithm previously validated experimentally in vitro on a number of unfolded peptides and proteins, and considering hydrophobicity, beta-sheet propensity, and charge as major sequence determinants of the aggregation process. These findings show that the physicochemical parameters previously recognized to govern amyloid formation by fully or partially unfolded proteins are largely applicable in vivo and pave the way for the molecular exploration of a process as complex as protein aggregation in living organisms.

Conformational properties of the aggregation precursor state of HypF-N.

The conversion of specific proteins or protein fragments into insoluble, ordered fibrillar aggregates is a fundamental process in protein chemistry, biology, medicine and biotechnology. As this structural conversion seems to be a property shared by many proteins, understanding the mechanism of this process will be of extreme importance. Here we present a structural characterisation of a conformational state populated at low pH by the N-terminal domain of Escherichia coli HypF. Combining different biophysical and biochemical techniques, including near- and far-UV circular dichroism, intrinsic and 8-anilinonaphthalene-1-sulfonate-derived fluorescence, dynamic light scattering and limited proteolysis, we will show that this state is largely unfolded but contains significant secondary structure and hydrophobic clusters. It also appears to be more compact than a random coil-like state but less organised than a molten globule state. Increase of the total ionic strength of the solution induces aggregation of such a pre-molten globule state into amyloid-like protofibrils, as revealed by thioflavin T fluorescence and atomic force microscopy. These results show that a pre-molten globule state can be, among other possible conformational states, one of the precursor states of amyloid formation. In addition, the possibility of triggering aggregation by modulating the ionic strength of the solution provides one a unique opportunity to study both the initial precursor state and the aggregation process.

Sequence and structural determinants of amyloid fibril formation.

Amyloid fibril formation is a process that represents an essential feature of the chemistry of proteins and plays a central role in human pathology and the biology of living organisms. In this Account, we shall describe some of the recent results on the sequence and structural determinants of protein aggregation. We shall describe the factors that govern aggregation of unfolded peptides and proteins. We shall then try to summarize the factors that pertain to the aggregation of partially structured states and will show that even fully folded states of proteins have an ability to aggregate into at least early oligomers with no need to undergo substantial conformational changes.

Investigating the effects of mutations on protein aggregation in the cell.

The conversion of peptides and proteins into highly ordered and intractable aggregates is associated with a range of debilitating human diseases and represents a widespread problem in biotechnology. Protein engineering studies carried out in vitro have shown that mutations promote aggregation when they either destabilize the native state of a globular protein or accelerate the conversion of unfolded or partially folded conformations into oligomeric structures. We have extended such studies to investigate protein aggregation in vivo where a number of additional factors able to modify dramatically the aggregation behavior of proteins are present. We have expressed, in Escherichia coli cells, an E. coli protein domain, HypF-N. The results for a range of mutational variants indicate that although mutants with a conformational stability similar to that of the wild-type protein are soluble in the E. coli cytosol, variants with single point mutations predicted to destabilize the protein invariably aggregate after expression. We show, however, that aggregation of destabilized variants can be prevented by incorporating multiple mutations designed to reduce the intrinsic propensity of the polypeptide chain to aggregate; in the cases discussed here, this is achieved by an increase in the net charge of the protein. These results suggest that the principles being established to rationalize aggregation behavior in vitro have general validity for situations in vivo where aggregation has both biotechnological and medical relevance.