DEAD-box helicases are essential for gene expression and RNA metabolism. However, the mechanisms regulating their activity remain largely elusive. The DEAD-box helicase DDX39B/UAP56 forms a 2:1 complex with the C-terminal domain (CTD) of RNA-binding protein Tho1, but the functional relevance of this interaction is still elusive. Here, we show that the Tho1-CTD stimulates the helicase activity of Sub2, the yeast homologue of DDX39B/UAP56, by acting as a rigid scaffold that promotes Sub2 oligomerization on RNA. The Tho1-CTD has two conserved α-helical motifs, each interacting with one Sub2, and we demonstrate that both motifs are essential for the stimulation. This scaffolding mechanism is shared across species, as the Tho1 ortholog MOS11 from Arabidopsis thaliana stimulates A. thaliana UAP56. Interestingly, MOS11 has five of the conserved α-helical motifs, which are connected by flexible linkers. We show that the number and spatial separation of these motifs are critical for stimulation and that MOS11 stimulates unwinding on a broader range of substrates than the Tho1-CTD. The cofactor-mediated helicase oligomerization is reminiscent of the self-oligomerization observed for other DEAD-box helicases. Furthermore, our data illustrate how cofactor architecture affects substrate specificity and provide a comprehensive mechanistic framework for cofactor-mediated helicase activation.

DNA replication errors that escape the proofreading activity of the replicative DNA polymerase are repaired by DNA mismatch repair (MMR). The initiation of MMR in Escherichia coli involves the recognition of the mismatch by MutS, binding of MutL, and activation of the endonuclease MutH which incises DNA at a hemi-methylated GATC site. MutS exists in a dimer-tetramer equilibrium, but the function of the tetramer during MMR remains unknown. Here, we used in vitro MutH activation assays to examine the role of MutS in the reaction steps that couple mismatch recognition to daughter strand incision. To study the behavior of different MutS oligomers, we used obligate dimers and tetramers and quantified GATC site incision on circular and linear DNA substrates. Especially in the presence of free DNA ends, MutS tetramers mediate more efficient incision than MutS dimers, likely due to tetramers diffusing slower and therefore being more successful in assembling the active incision complex before dissociating from the DNA at the ends. Likewise, we observed that MutS tetramers have a higher preference for nicking the GATC site close to the mismatch than dimeric MutS. Through probabilistic modeling, we show that this increased preference is consistent with a fourfold decrease in diffusion constant for the tetramer compared to the dimer. We propose that during mismatch repair, DNA excision tracts resulting from MutS tetramer-mediated incision will be shorter than those mediated by the dimer, and that this explains the reported higher repair efficiencies of wild-type MutS compared to the dimer.

UvrD is a helicase vital for DNA replication and quality control processes. In its monomeric state, UvrD exhibits limited helicase activity, necessitating either dimerization or assistance from an accessory protein to efficiently unwind DNA. Within the DNA mismatch repair pathway, MutL plays a pivotal role in relaying the repair signal, enabling UvrD to unwind DNA from the strand incision site up to and beyond the mismatch. Although this interdependence is well-established, the precise mechanism of activation and the specific MutL-UvrD interactions that trigger helicase activity remain elusive. To address these questions, we employed site-specific crosslinking techniques using single-cysteine variants of MutL and UvrD followed by functional assays. Our investigation unveils that the C-terminal domain of MutL not only engages with UvrD but also acts as a self-sufficient activator of UvrD helicase activity on DNA substrates with 3’-single-stranded tails. Especially when MutL is covalently attached to the 2B or 1B domain the tail length can be reduced to a minimal substrate of 5 nucleotides without affecting unwinding efficiency.

In previous works, we demonstrated that tertiary 3-chloropiperidines are potent chemotherapeutics, alkylating the DNA through the formation of bicyclic aziridinium ions. Herein, we report the synthesis of novel secondary 3-chloropiperidine analogues. The synthesis incorporates a new procedure to monochlorinate unsaturated primary amines utilizing N-chlorosuccinimide, while carefully monitoring the temperature to prevent dichlorination. Furthermore, we successfully isolated highly strained bicyclic aziridines by treating the secondary 3-chloropiperidines with a sufficient amount of base. We conclude this work with a DNA cleavage assay as a proof of principle, comparing our previously known substrates to the novel compounds. In this, the secondary 3-chloropiperidine as well as the isolated bicyclic aziridine, proved to be more effective than their tertiary counterpart.

DNA mismatch repair detects and corrects mismatches introduced during DNA replication. The protein MutS scans for mismatches and coordinates the repair cascade. During this process, MutS undergoes multiple conformational changes in response to ATP binding, hydrolysis and release, but how ATP induces the various MutS conformations is incompletely understood. Here we present four cryogenic electron microscopy structures of Escherichia coli MutS at sequential stages of the ATP hydrolysis cycle that reveal how ATP binding and hydrolysis induce closing and opening of the MutS dimer, respectively. Biophysical analysis demonstrates how DNA binding modulates the ATPase cycle by prevention of hydrolysis during scanning and mismatch binding, while preventing ADP release in the sliding clamp state. Nucleotide release is achieved when MutS encounters single-stranded DNA that is produced during removal of the daughter strand. The combination of ATP binding and hydrolysis and its modulation by DNA enables MutS to adopt the different conformations needed to coordinate the sequential steps of the mismatch repair cascade. Cryogenic electron microscopy structures of Escherichia coli MutS at sequential stages of the ATP hydrolysis cycle reveal how ATP binding and hydrolysis and its modulation by DNA enable MutS to adopt different conformations during mismatch repair.

Covalent protein capture (cross-linking) by reactive DNA derivatives makes it possible to investigate structural features by fixing complexes at different stages of DNA–protein recognition. The most common cross-linking methods are based on reactive groups that interact with native or engineered cysteine residues. Nonetheless, high reactivity of most of such groups leads to preferential fixation of early-stage complexes or even non-selective cross-linking. We synthesised a set of DNA reagents carrying an acrylamide group attached to the C5 atom of a 2′-deoxyuridine moiety via various linkers and studied cross-linking with MutS as a model protein. MutS scans DNA for mismatches and damaged nucleobases and can form multiple non-specific complexes with DNA that may cause non-selective cross-linking. By varying the length of the linker between DNA and the acrylamide group and by changing the distance between the reactive nucleotide and a mismatch in the duplex, we showed that cross-linking occurs only if the distance between the acrylamide group and cysteine is optimal within the DNA–protein complex. Thus, acrylamide-modified DNA duplexes are excellent tools for studying DNA–protein interactions because of high selectivity of cysteine trapping.

Overproduction of the extracellular Serratia marcescens nuclease in Escherichia coli results in aggregation and sequestration of a large amount of the protein in inclusion bodies. Only a relatively small amount is secreted into the medium from which it can be purified following established procedures. The cell-associated insoluble protein can be solubilized in 6 M urea after breaking up the cells by sonication. Renaturation is achieved by dilution or dialysis. Subsequent phosphocellulose chromatography yields a homogeneous protein preparation which is shown by a variety of biochemical and biophysical analyses to be indistinguishable from conventionally prepared material. The high yield (>10 mg/500-ml culture) and the ease of preparation (2 to 3 days) make this an attractive alternative to previously described procedures.

DNA mismatch repair detects and removes mismatches from DNA reducing the error rate of DNA replication a 100-1000 fold. The MutS protein is one of the key players that scans for mismatches and coordinates the repair cascade. During this, MutS undergoes multiple conformational changes that initiate the subsequent steps, in response to ATP binding, hydrolysis, and release. How ATP induces the different conformations in MutS is not well understood. Here we present four cryo-EM structures of Escherichia coli MutS at sequential stages of the ATP hydrolysis cycle. These structures reveal how ATP binding and hydrolysis induces a closing and opening of the MutS dimer, respectively. Additional biophysical analysis furthermore explains how DNA binding modulates the ATPase cycle by preventing hydrolysis during scanning and mismatch binding, while preventing ADP release in the sliding clamp state. Nucleotide release is achieved when MutS encounters single stranded DNA that is produced during the removal of the daughter strand. This way, the combination of the ATP binding and hydrolysis and its modulation by DNA enable MutS to adopt different conformations needed to coordinate the sequential steps of the mismatch repair cascade.

ABSTRACTWe developed a sensitive, homogeneous fluorescence assay for the detection of DNA mismatches and DNA damage based on the mismatch repair (MMR) protein MutS. The assay is based on Förster resonance energy transfer (FRET) between SYBR Green I (SG), non-covalently bound to DNA, and Alexa Fluor 647 (AF647) conjugated to MutS. In contrast to previous assays using only the mismatch binding activity of MutS, we exploited the ATP-dependent loading of multiple MutS sliding clamps provoked by mismatch/damage to the DNA, which increases the overall sensitivity of the assay. The assay was validated using a well-characterized 3 kb circular DNA containing a single G/T mismatch. We also demonstrate that treatment of long (multiple kb) DNA with various chemical or physical agents including non-denaturing bisulfite conversion of cytosine to uracil, cisplatin modification or ultraviolet light (UVC) results in changes in the DNA that can be detected by the FRET-based MutS biosensor.

DNA mismatch repair detects and removes mismatches from DNA by a conserved mechanism, reducing the error rate of DNA replication by 100- to 1,000-fold. In this process, MutS homologs scan DNA, recognize mismatches and initiate repair. How the MutS homologs selectively license repair of a mismatch among millions of matched base pairs is not understood. Here we present four cryo-EM structures of Escherichia coli MutS that provide snapshots, from scanning homoduplex DNA to mismatch binding and MutL activation via an intermediate state. During scanning, the homoduplex DNA forms a steric block that prevents MutS from transitioning into the MutL-bound clamp state, which can only be overcome through kinking of the DNA at a mismatch. Structural asymmetry in all four structures indicates a division of labor between the two MutS monomers. Together, these structures reveal how a small conformational change from the homoduplex- to heteroduplex-bound MutS acts as a licensing step that triggers a dramatic conformational change that enables MutL binding and initiation of the repair cascade.

Noncovalent interactions between DNA and proteins are vital for cells. DNA-protein crosslinking may occur due to various endogenous and exogenous factors, as well as chemotherapeutic agents, and often affects the performance of DNA-binding proteins. Oligodeoxyribonucleotides with a reactive group can serve as a tool to investigate the structure and action of DNA-binding proteins. Here, we report the crosslinking of MutS, which is a sensor protein of DNA mismatch repair (MMR) in Escherichia coli, via the protein’s cysteine. This was realized via DNA that harboured a 2′-deoxy-2′-[3-(2-pyridyldithio)propionamide] group as part of a nucleoside at a given position, and two fluorophores. The proposed DNA duplex consisted of complementary oligonucleotides with breaks in the top and bottom strands. This feature allowed us to introduce the reactive group at different positions and made it possible to synthesize only the central fragment of the DNA with a minimal length. The MutS-DNA conjugate was obtained by thiol-disulfide exchange with quantitative yield and separated from the unreacted DNA by size-exclusion chromatography. Fluorescence resonance energy transfer was used to show that MutS remained functionally active within the conjugate, since the protein was able to change its conformation and DNA conformation as well.

As no crystal structure of full-size MutL bound to DNA has been obtained up to date, in the present work we used crosslinking and Förster resonance energy transfer (FRET) assays for probing the putative DNA-binding center of MutL from Escherichia coli. Several single-cysteine MutL variants (scMutL) were used for site-specific crosslinking or fluorophore modification. The crosslinking efficiency between scMutL proteins and mismatched DNA modified with thiol-reactive probes correlated with the distances from the Cys residues to the DNA calculated from a model of MutS–MutL–DNA complex. FRET-based investigation of DNA binding with different scMutL variants clearly showed that the highest signals were detected for the variants MutL(T218C) and MutL(A251C) indicating closeness of the positions 218 and 251 to DNA in the MutL–DNA complex. Indeed, the Cys218 and Cys251 of scMutL were crosslinked to the reactive DNA with the highest yield demonstrating their proximity to DNA in the MutL–DNA complex. The presence of MutS increased the yield of conjugate formation between the MutL variants and the modified DNA due to tighter MutL–DNA interactions caused by MutS binding to MutL.

DNA mismatch repair (MMR) maintains genome stability through repair of DNA replication errors. In Escherichia coli , initiation of MMR involves recognition of the mismatch by MutS, recruitment of MutL, activation of endonuclease MutH and DNA strand incision at a hemimethylated GATC site. Here we studied the mechanism of communication that couples mismatch recognition to daughter strand incision. We investigated the effect of catalytically-deficient Cas9 as well as stalled RNA polymerase as roadblocks placed on DNA in between the mismatch and GATC site in ensemble and single molecule nanomanipulation incision assays. The MMR proteins were observed to incise GATC sites beyond a roadblock, albeit with reduced efficiency. This residual incision is completely abolished upon shortening the disordered linker regions of MutL. These results indicate that roadblock bypass can be fully attributed to the long, disordered linker regions in MutL and establish that communication during MMR initiation occurs along the DNA backbone.

Maintenance of genome stability is dependent on correct action of DNA mismatch repair (MMR), to prevent the incorporation of polymerase errors during replication. In MMR, a MutS ATPase (or eukaryotic MutSalpha) recognizes the mismatch or short insertion/deletion loops and activates a MutL (or eukaryotic MutLalpha ATPase by changing its conformation. This enables MutL to activate an endonuclease for cleavage of the newly synthesized strand, so that it can be excised and resynthesized. MMR communicates between replication and other repair processes and has important roles in recombination, triplet repeat expansion, somatic hypermutation and signaling. Errors in MMR proteins lead to a mutator phenotype and a cancer predisposition in humans, called Lynch syndrome. How information is transmitted from the mismatch to repair action is unclear due to a lack of knowledge about the interplay between the different components. Structures of MMR complexes could resolve this, but the transient nature of the interactions has complicated their analysis. Recently, we managed to trap a first partial MutS/MutL complex, by site-specific crosslinking, which allowed structure solution by X-ray crystallography and single particle cryo-electron microscopy (cryo-EM). Here we show how cryo-EM can now be used to study MMR complexes by enriching for specific states. This provides detailed insight in multiple states along this transient pathway.

mRNA is the "hermes" of gene expression as it carries the information of a protein-coding gene to the ribosome. Already during its synthesis, the mRNA is bound by mRNA-binding proteins that package the mRNA into a messenger ribonucleoprotein particle (mRNP). This mRNP assembly is important for mRNA stability and nuclear mRNA export. It also often regulates later steps in the mRNA lifetime such as translation and mRNA degradation in the cytoplasm. Thus, mRNP composition and accordingly the assembly of nuclear mRNA-binding proteins onto the mRNA are of crucial importance for correct gene expression. Here, we review our current knowledge of the mechanism of co-transcriptional mRNP assembly and nuclear mRNA export. We introduce the proteins involved and elaborate on what is known about their functions so far. In addition, we discuss the importance of regulated mRNP assembly in changing environmental conditions, especially during stress. Furthermore, we examine how defects in mRNP assembly cause diseases and how viruses exploit the host's nuclear mRNA export pathway. Finally, we summarize the questions that need to be answered in the future.

How multidomain RNA-binding proteins recognize their specific target sequences, based on a combinatorial code, represents a fundamental unsolved question and has not been studied systematically so far. Here we focus on a prototypical multidomain RNA-binding protein, IMP3 (also called IGF2BP3), which contains six RNA-binding domains (RBDs): four KH and two RRM domains. We establish an integrative systematic strategy, combining single-domain-resolved SELEX-seq, motif-spacing analyses, in vivo iCLIP, functional validation assays, and structural biology. This approach identifies the RNA-binding specificity and RNP topology of IMP3, involving all six RBDs and a cluster of up to five distinct and appropriately spaced CA-rich and GGC-core RNA elements, covering a >100 nucleotide-long target RNA region. Our generally applicable approach explains both specificity and flexibility of IMP3-RNA recognition, allows the prediction of IMP3 targets, and provides a paradigm for the function of multivalent interactions with multidomain RNA-binding proteins in gene regulation.

We have purified the MutL protein from Rhodobacter sphaeroides mismatch repair system (rsMutL) for the first time. rsMutL demonstrated endonuclease activity in vitro, as predicted by bioinformatics analysis. Based on the alignment of 1483 sequences of bacterial MutL homologs with presumed endonuclease activity, conserved functional motifs and amino acid residues in the rsMutL sequence were identified: five motifs comprising the catalytic site responsible for DNA cleavage were found in the C–terminal domain; seven conserved motifs involved in ATP binding and hydrolysis and specific to the GHKL family of ATPases were found in the N–terminal domain. rsMutL demonstrated the highest activity in the presence of Mn2+. The extent of plasmid DNA hydrolysis declined in the row Mn2+ > Co2+ > Mg2+ > Cd2+; Ni2+ and Ca2+ did not activate rsMutL. Divalent zinc ions inhibited rsMutL endonuclease activity in the presence of Mn2+ excess. ATP also suppressed plasmid DNA hydrolysis by rsMutL. Analysis of amino acid sequences and biochemical properties of five studied bacterial MutL homologs with endonuclease activity revealed that rsMutL resembles the MutL proteins from Neisseria gonorrhoeae and Pseudomonas aeruginosa.

DNA mismatch repair (MMR) is necessary to prevent incorporation of polymerase errors into the newly synthesized DNA strand, as they would be mutagenic. In humans, errors in MMR cause a predisposition to cancer, called Lynch syndrome. The MMR process is performed by a set of ATPases that transmit, validate, and couple information to identify which DNA strand requires repair. To understand the individual steps in the repair process, it is useful to be able to study these large molecular machines structurally and functionally. However, the steps and states are highly transient; therefore, the methods to capture and enrich them are essential. Here, we describe how single-cysteine variants can be used for specific cross-linking and labeling approaches that allow trapping of relevant transient states. Analysis of these defined states in functional and structural studies is instrumental to elucidate the molecular mechanism of this important DNA MMR process.

DNA mismatch repair (MMR) is an evolutionarily-conserved process responsible for the repair of replication errors. In Escherichia coli, MMR is initiated by MutS and MutL, which activate MutH to incise transiently-hemimethylated GATC sites. MMR efficiency depends on the distribution of these GATC sites. To understand which molecular events determine repair efficiency, we quantitatively studied the effect of strand incision on unwinding and excision activity. The distance between mismatch and GATC site did not influence the strand incision rate, and an increase in the number of sites enhanced incision only to a minor extent. Two GATC sites were incised by the same activated MMR complex in a processive manner, with MutS, the closed form of MutL and MutH displaying different roles. Unwinding and strand excision were more efficient on a substrate with two nicks flanking the mismatch, as compared to substrates containing a single nick or two nicks on the same side of the mismatch. Introduction of multiple nicks by the human MutLα endonuclease also contributed to increased repair efficiency. Our data support a general model of prokaryotic and eukaryotic MMR in which, despite mechanistic differences, mismatch-activated complexes facilitate efficient repair by creating multiple daughter strand nicks.

The DNA mismatch repair (MMR) system plays a crucial role in the prevention of replication errors and in the correction of some oxidative damages of DNA bases. In the present work the most abundant oxidized pyrimidine lesion, 5,6-dihydro-5,6-dihydroxythymidine (thymidine glycol, Tg) was tested for being recognized and processed by the E. coli MMR system, namely complex of MutS, MutL and MutH proteins. In a partially reconstituted MMR system with MutS-MutL-MutH proteins, G/Tg and A/Tg containing plasmids failed to provoke the incision of DNA. Tg residue in the 30-mer DNA duplex destabilized double helix due to stacking disruption with neighboring bases. However, such local structural changes are not important for E. coli MMR system to recognize this lesion. A lack of repair of Tg containing DNA could be due to a failure of MutS (a first acting protein of MMR system) to interact with modified DNA in a proper way. It was shown that Tg in DNA does not affect on ATPase activity of MutS. On the other hand, MutS binding affinities to DNA containing Tg in G/Tg and A/Tg pairs are lower than to DNA with a G/T mismatch and similar to canonical DNA. Peculiarities of MutS interaction with DNA was monitored by Forster resonance energy transfer (FRET) and fluorescence anisotropy. Binding of MutS to Tg containing DNAs did not result in the formation of characteristic DNA kink. Nevertheless, MutS homodimer orientation on Tg-DNA is similar to that in the case of G/T-DNA. In contrast to G/T-DNA, neither G/Tg- nor A/Tg-DNA was able to stimulate ADP release from MutS better than canonical DNA. Thus, Tg residue in DNA is unlikely to be recognized or processed by the E. coli MMR system. Probably, the MutS transformation to active "sliding clamp" conformation on Tg-DNA is problematic.