Dynamic BTB-domain filaments promote clustering of ZBTB proteins.

The formation of dynamic protein filaments contributes to various biological functions by clustering individual molecules together and enhancing their binding to ligands. We report such a propensity for the BTB domains of certain proteins from the ZBTB family, a large eukaryotic transcription factor family implicated in differentiation and cancer. Working with Xenopus laevis and human proteins, we solved the crystal structures of filaments formed by dimers of the BTB domains of ZBTB8A and ZBTB18 and demonstrated concentration-dependent higher-order assemblies of these dimers in solution. In cells, the BTB-domain filamentation supports clustering of full-length human ZBTB8A and ZBTB18 into dynamic nuclear foci and contributes to the ZBTB18-mediated repression of a reporter gene. The BTB domains of up to 21 human ZBTB family members and two related proteins, NACC1 and NACC2, are predicted to behave in a similar manner. Our results suggest that filamentation is a more common feature of transcription factors than is currently appreciated.

Updated protein domain annotation of the PARP protein family sheds new light on biological function.

AlphaFold2 and related computational tools have greatly aided studies of structural biology through their ability to accurately predict protein structures. In the present work, we explored AF2 structural models of the 17 canonical members of the human PARP protein family and supplemented this analysis with new experiments and an overview of recent published data. PARP proteins are typically involved in the modification of proteins and nucleic acids through mono or poly(ADP-ribosyl)ation, but this function can be modulated by the presence of various auxiliary protein domains. Our analysis provides a comprehensive view of the structured domains and long intrinsically disordered regions within human PARPs, offering a revised basis for understanding the function of these proteins. Among other functional insights, the study provides a model of PARP1 domain dynamics in the DNA-free and DNA-bound states and enhances the connection between ADP-ribosylation and RNA biology and between ADP-ribosylation and ubiquitin-like modifications by predicting putative RNA-binding domains and E2-related RWD domains in certain PARPs. In line with the bioinformatic analysis, we demonstrate for the first time PARP14's RNA-binding capability and RNA ADP-ribosylation activity in vitro. While our insights align with existing experimental data and are probably accurate, they need further validation through experiments.

Structural insights into the regulation of the human E2∼SUMO conjugate through analysis of its stable mimetic.

Protein SUMOylation is a ubiquitylation-like post-translational modification (PTM) that is synthesized through an enzymatic cascade involving an E1 (SAE1:SAE2), an E2 (UBC9), and various E3 enzymes. In the final step of this process, the small ubiquitin-like modifier (SUMO) is transferred from the UBC9∼SUMO thioester onto a lysine residue of a protein substrate. This reaction can be accelerated by an E3 ligase. As the UBC9∼SUMO thioester is chemically unstable, a stable mimetic is desirable for structural studies of UBC9∼SUMO alone and in complex with a substrate and/or an E3 ligase. Recently, a strategy for generating a mimetic of the yeast E2∼SUMO thioester by mutating alanine 129 of Ubc9 to a lysine has been reported. Here, we reproduce and further investigate this approach using the human SUMOylation system and characterize the resulting mimetic of human UBC9∼SUMO1. We show that substituting lysine for alanine 129, but not for other active-site UBC9 residues, results in a UBC9 variant that is efficiently auto-SUMOylated. The auto-modification is dependent on cysteine 93 of UBC9, suggesting that it proceeds via this residue, through the same pathway as that for SUMOylation of substrates. The process is also partially dependent on aspartate 127 of UBC9 and accelerated by high pH, highlighting the importance of the substrate lysine protonation state for efficient SUMOylation. Finally, we present the crystal structure of the UBC9-SUMO1 molecule, which reveals the mimetic in an open conformation and its polymerization via the noncovalent SUMO-binding site on UBC9. Similar interactions could regulate UBC9∼SUMO in some cellular contexts.

Thiopurine Derivative-Induced Fpg/Nei DNA Glycosylase Inhibition: Structural, Dynamic and Functional Insights.

DNA glycosylases are emerging as relevant pharmacological targets in inflammation, cancer and neurodegenerative diseases. Consequently, the search for inhibitors of these enzymes has become a very active research field. As a continuation of previous work that showed that 2-thioxanthine (2TX) is an irreversible inhibitor of zinc finger (ZnF)-containing Fpg/Nei DNA glycosylases, we designed and synthesized a mini-library of 2TX-derivatives (TXn) and evaluated their ability to inhibit Fpg/Nei enzymes. Among forty compounds, four TXn were better inhibitors than 2TX for Fpg. Unexpectedly, but very interestingly, two dithiolated derivatives more selectively and efficiently inhibit the zincless finger (ZnLF)-containing enzymes (human and mimivirus Neil1 DNA glycosylases hNeil1 and MvNei1, respectively). By combining chemistry, biochemistry, mass spectrometry, blind and flexible docking and X-ray structure analysis, we localized new TXn binding sites on Fpg/Nei enzymes. This endeavor allowed us to decipher at the atomic level the mode of action for the best TXn inhibitors on the ZnF-containing enzymes. We discovered an original inhibition mechanism for the ZnLF-containing Fpg/Nei DNA glycosylases by disulfide cyclic trimeric forms of dithiopurines. This work paves the way for the design and synthesis of a new structural class of inhibitors for selective pharmacological targeting of hNeil1 in cancer and neurodegenerative diseases.

The nucleoid-associated protein HU enhances 8-oxoguanine base excision by the formamidopyrimidine-DNA glycosylase.

The nucleoid-associated protein HU is involved in numerous DNA transactions and thus is essential in DNA maintenance and bacterial survival. The high affinity of HU for SSBs (single-strand breaks) has suggested its involvement in DNA protection, repair and recombination. SSB-containing DNA are major intermediates transiently generated by bifunctional DNA N-glycosylases that initiate the BER (base excision repair) pathway. Enzyme kinetics and DNA-binding experiments demonstrate that HU enhances the 8-oxoguanine-DNA glycosylase activity of Fpg (formamidopyrimidine-DNA glycosylase) by facilitating the release of the enzyme from its final DNA product (one nucleoside gap). We propose that the displacement of Fpg from its end-DNA product by HU is an active mechanism in which HU recognizes the product when it is still bound by Fpg. Through DNA binding, the two proteins interplay to form a transient ternary complex Fpg/DNA/HU which results in the release of Fpg and the molecular entrapment of SSBs by HU. These results support the involvement of HU in BER in vivo.

Zinc finger oxidation of Fpg/Nei DNA glycosylases by 2-thioxanthine: biochemical and X-ray structural characterization.

DNA glycosylases from the Fpg/Nei structural superfamily are base excision repair enzymes involved in the removal of a wide variety of mutagen and potentially lethal oxidized purines and pyrimidines. Although involved in genome stability, the recent discovery of synthetic lethal relationships between DNA glycosylases and other pathways highlights the potential of DNA glycosylase inhibitors for future medicinal chemistry development in cancer therapy. By combining biochemical and structural approaches, the physical target of 2-thioxanthine (2TX), an uncompetitive inhibitor of Fpg, was identified. 2TX interacts with the zinc finger (ZnF) DNA binding domain of the enzyme. This explains why the zincless hNEIL1 enzyme is resistant to 2TX. Crystal structures of the enzyme bound to DNA in the presence of 2TX demonstrate that the inhibitor chemically reacts with cysteine thiolates of ZnF and induces the loss of zinc. The molecular mechanism by which 2TX inhibits Fpg may be generalized to all prokaryote and eukaryote ZnF-containing Fpg/Nei-DNA glycosylases. Cell experiments show that 2TX can operate in cellulo on the human Fpg/Nei DNA glycosylases. The atomic elucidation of the determinants for the interaction of 2TX to Fpg provides the foundation for the future design and synthesis of new inhibitors with high efficiency and selectivity.

An Atypical Mechanism of SUMOylation of Neurofibromin SecPH Domain Provides New Insights into SUMOylation Site Selection.

Neurofibromin (Nf1) is a giant multidomain protein encoded by the tumour-suppressor gene NF1. NF1 is mutated in a common genetic disease, neurofibromatosis type I (NF1), and in various cancers. The protein has a Ras-GAP (GTPase activating protein) activity but is also connected to diverse signalling pathways through its SecPH domain, which interacts with lipids and different protein partners. We previously showed that Nf1 partially colocalized with the ProMyelocytic Leukemia (PML) protein in PML nuclear bodies, hotspots of SUMOylation, thereby suggesting the potential SUMOylation of Nf1. Here, we demonstrate that the full-length isoform 2 and a SecPH fragment of Nf1 are substrates of the SUMO pathway and identify a well-defined SUMOylation profile of SecPH with two main modified lysines. One of these sites, K1731, is highly conserved and surface-exposed. Despite the presence of an inverted SUMO consensus motif surrounding K1731, and a potential SUMO-interacting motif (SIM) within SecPH, we show that neither of these elements is necessary for K1731 SUMOylation, which is also independent of Ubc9 SUMOylation on K14. A 3D model of an interaction between SecPH and Ubc9 centred on K1731, combined with site-directed mutagenesis, identifies specific structural elements of SecPH required for K1731 SUMOylation, some of which are affected in reported NF1 pathogenic variants. This work provides a new example of SUMOylation dependent on the tertiary rather than primary protein structure surrounding the modified site, expanding our knowledge of mechanisms governing SUMOylation site selection.

Comparing native and irradiated E. coli lactose repressor-operator complex by molecular dynamics simulation.

The function of the E. coli lactose operon requires the binding of the tetrameric repressor protein to the operator DNA. We have previously shown that gamma-irradiation destabilises the repressor-operator complex because the repressor gradually loses its DNA-binding ability (Radiat Res 170:604-612, 2008). It was suggested that the observed oxidation of tyrosine residues and the concomitant structural changes of irradiated headpieces (DNA-binding domains of repressor monomers) could be responsible for the inactivation. To unravel the mechanisms that lead to repressor-operator complex destabilisation when tyrosine oxidation occurs, we have compared by molecular dynamic simulations two complexes: (1) the native complex formed by two headpieces and the operator DNA, and (2) the damaged complex, in which all tyrosines are replaced by their oxidation product 3,4-dihydroxyphenylalanine (DOPA). On a 20 ns time scale, MD results show effects consistent with complex destabilisation: increased flexibility, increased DNA bending, modification of the hydrogen bond network, and decrease of the positive electrostatic potential at the protein surface and of the global energy of DNA-protein interactions.

Radiation-induced oxidative damage to the DNA-binding domain of the lactose repressor.

Understanding the cellular effects of radiation-induced oxidation requires the unravelling of key molecular events, particularly damage to proteins with important cellular functions. The Escherichia coli lactose operon is a classical model of gene regulation systems. Its functional mechanism involves the specific binding of a protein, the repressor, to a specific DNA sequence, the operator. We have shown previously that upon irradiation with gamma-rays in solution, the repressor loses its ability to bind the operator. Water radiolysis generates hydroxyl radicals (OH* radicals) which attack the protein. Damage of the repressor DNA-binding domain, called the headpiece, is most likely to be responsible of this loss of function. Using CD, fluorescence spectroscopy and a combination of proteolytic cleavage with MS, we have examined the state of the irradiated headpiece. CD measurements revealed a dose-dependent conformational change involving metastable intermediate states. Fluorescence measurements showed a gradual degradation of tyrosine residues. MS was used to count the number of oxidations in different regions of the headpiece and to narrow down the parts of the sequence bearing oxidized residues. By calculating the relative probabilities of reaction of each amino acid with OH. radicals, we can predict the most probable oxidation targets. By comparing the experimental results with the predictions we conclude that Tyr7, Tyr12, Tyr17, Met42 and Tyr47 are the most likely hotspots of oxidation. The loss of repressor function is thus correlated with chemical modifications and conformational changes of the headpiece.

Radiation damage to DNA-protein specific complexes: estrogen response element-estrogen receptor complex.

The exposure of a DNA-protein regulatory complex to ionising radiation induces damage to both partner biomolecules and thus can affect its functioning. Our study focuses on a complex formed by the estrogen response element (ERE) DNA and the recombinant human estrogen receptor alpha (ER), which mediates the signalling of female sex hormones, estrogens. The method of native polyacrylamide retardation gel electrophoresis is used to study the stability of the complex under irradiation by low LET radiation ((60)Co gamma rays) and the ability of the separately irradiated partners to form complexes. The relative probabilities of ERE DNA strand breakage and base damages as well as the probabilities of damages to the ER binding domain are calculated using the Monte Carlo method-based model RADACK.

Modification of DNA radiolysis by DNA-binding proteins: structural aspects.

Formation of specific complexes between proteins and their cognate DNA modulates the yields and the location of radiation damage on both partners of the complex. The radiolysis of DNA-protein complexes is studied for: (1) the Escherichia coli lactose operator-repressor complex, (2) the complex between DNA bearing an analogue of an abasic site and the repair protein Fpg of Lactococcus lactis. Experimental patterns of DNA damages are presented and compared to predicted damage distribution obtained using an improved version of the stochastic model RADACK. The same method is used for predicting the location of damages on the proteins. At doses lower than a threshold that depends on the system, proteins protect their specific binding site on DNA while at high doses, the studied complexes are disrupted mainly through protein damage. The loss of binding ability is the functional consequence of the amino-acids modification by OH* radicals. Many of the most probably damaged amino acids are essential for the DNA-protein interaction and within a complex are protected by DNA.

Model of a DNA-protein complex of the architectural monomeric protein MC1 from Euryarchaea.

In Archaea the two major modes of DNA packaging are wrapping by histone proteins or bending by architectural non-histone proteins. To supplement our knowledge about the binding mode of the different DNA-bending proteins observed across the three domains of life, we present here the first model of a complex in which the monomeric Methanogen Chromosomal protein 1 (MC1) from Euryarchaea binds to the concave side of a strongly bent DNA. In laboratory growth conditions MC1 is the most abundant architectural protein present in Methanosarcina thermophila CHTI55. Like most proteins that strongly bend DNA, MC1 is known to bind in the minor groove. Interaction areas for MC1 and DNA were mapped by Nuclear Magnetic Resonance (NMR) data. The polarity of protein binding was determined using paramagnetic probes attached to the DNA. The first structural model of the DNA-MC1 complex we propose here was obtained by two complementary docking approaches and is in good agreement with the experimental data previously provided by electron microscopy and biochemistry. Residues essential to DNA-binding and -bending were highlighted and confirmed by site-directed mutagenesis. It was found that the Arg25 side-chain was essential to neutralize the negative charge of two phosphates that come very close in response to a dramatic curvature of the DNA.

Alanine screening mutagenesis establishes the critical inactivating damage of irradiated E. coli lactose repressor.

The function of the E. coli lactose operon requires the binding of lactose repressor to operator DNA. We have previously shown that γ rradiation destabilizes the repressor-operator complex because the repressor loses its DNA-binding ability. It was suggested that the observed oxidation of the four tyrosines (Y7, Y12, Y17, Y47) and the concomitant structural changes of the irradiated DNA-binding domains (headpieces) could be responsible for the inactivation. To pinpoint the tyrosine whose oxidation has the strongest effect, four headpieces containing the product of tyrosine oxidation, 3,4-dihydroxyphenylalanine (DOPA), were simulated by molecular dynamics. We have observed that replacing Y47 by DOPA triggers the largest change of structure and stability of the headpiece and have concluded that Y47 oxidation is the greatest contributor to the decrease of repressor binding to DNA. To experimentally verify this conclusion, we applied the alanine screening mutagenesis approach. Tetrameric mutated repressors bearing an alanine instead of each one of the tyrosines were prepared and their binding to operator DNA was checked. Their binding ability is quite similar to that of the wild-type repressor, except for the Y47A mutant whose binding is strongly reduced. Circular dichroism determinations revealed small reductions of the proportion of α helices and of the melting temperature for Y7A, Y12A and Y17A headpieces, but much larger ones were revealed for Y47A headpiece. These results established the critical role of Y47 oxidation in modifying the structure and stability of the headpiece, and in reduction of the binding ability of the whole lactose repressor.

Oxidation-sensitive residues mediate the DNA bending abilities of the architectural MC1 protein.

The Methanosarcina thermophila MC1 protein is a small basic protein that is able to bend DNA sharply. When this protein is submitted to oxidative stress through gamma irradiation, it loses its original DNA interaction properties. The protein can still bind DNA but its ability to bend DNA is decreased dramatically. Here, we used different approaches to determine the oxidations that are responsible for this inactivation. Through a combination of proteolysis and mass spectrometry we have identified the three residues that are oxidized preferentially. We show by site directed mutagenesis that two of these residues, Trp74 and Met75, are involved in the DNA binding. Their substitution by alanine leads to a strong reduction in the protein capacity to bend DNA, and a total loss of its ability to recognize bent DNA. Taken together, these results show that oxidation of both these residues is responsible for the protein inactivation. Furthermore, the results confirm the strong relationship between DNA bending and recognition of DNA sequences by the MC1 protein.

NEMO trimerizes through its coiled-coil C-terminal domain.

NEMO/IkappaB kinase (IKK) gamma is the regulatory component of the IKK complex comprising the two protein kinases, IKKalpha and IKKbeta. To investigate the self-assembly properties of NEMO and to understand further the mechanism of activation of the IKK complex, we purified wild-type and mutant NEMO expressed in Escherichia coli. In the absence of its IKK partners, recombinant NEMO (rNEMO) is a metastable functional monomer correctly folded, according to its fluorescence and far-UV CD spectra, which is binding specifically to the IKK complex. A minor fraction of rNEMO was found tightly associated with DnaK (E. coli Hsp70). We also examined the interaction of NEMO with prokaryotic and eukaryotic Hsp70, and we showed that the Hsp70-NEMO complex forms a supramolecular structure probably corresponding to an assembly intermediate. In vivo cross-linking experiments indicate that native NEMO in association with IKK is in equilibrium between a dimeric and a trimeric form. Similarly to native NEMO, a NEMO mutant deleted from its IKK binding N-terminal domain (residues 242-388) forms a stable trimeric coiled-coil, suggesting that the association of NEMO with IKK or with Hsp70 prevents incorrect interdomain pairing reactions that could lead to aggregation or to an non-native oligomeric state of rNEMO. We propose a model in which the activation of the IKK complex occurs through the trimerization of NEMO upon binding to a not yet identified upstream activator.