Final Report Summary - POSTBIOMIN (Program developing interdisciplinary research POtential for the STudies of BIOMolecular INteractions)
Project context and objectives:
National Centre for Biomolecular Research (NCBR), Masaryk University Brno in the Czech Republic has become one of the leading laboratories for structural biology in the Central Europe. The institute is very well equipped by modern laboratory equipment and instruments. Project POSTBIOMIN strengthened our research potential in theoretical and experimental research that focuses on the structural characterisation of biomacromolecules and especially biomolecular interactions study through the scheme of support measures like recruitment of foreign experts, organisation of international PhD courses and lectures given by top international scientists. Series of workshops, seminars, postdoctoral and expert inter-ships and visits will be organised. Our technology transfer capabilities were strengthened during the project as well. POSTBIOMIN complemented investments in research capacity co-financed through the European Union (EU) cohesion policy programmes for 2007-2013, i.e. CEITEC, a large project of Central European Institute of Technology focused on life sciences and advanced materials and technologies. Project POSTBIOMIN contributed to the creation of human potential for modern research institutions and companies dealing with bio-medical research and biotechnologies within the region of Brno.
The global objective of this project is to further develop research potential of our institute (NCBR), which is important part of regional network of life sciences research institutions. This project will help us to gain knowledge and exchange experience with other European regions. We will work on setting up strategic partnerships with other research teams elsewhere in Europe as well. One of the greatest challenges to be achieved is the return of Czech scientist who would come back to pursue her/his independent career in our institute.
Apart from global objective we can identify technical and scientific objectives in POSTBIOMIN project. Due to the character of REGPOT-1 call, technical objectives are crucial and the most substantial for the project. However achievement of scientific goals and objectives is even more important, despite scientific objectives are rather hidden behind main content of work packages.
Technical objectives, i.e. organisation of post-doc and PhD courses, appointment of foreign experts and scientists, organisation of lectures, workshops, trainings both in Masaryk University in Brno and abroad (EU-15 countries) are described in details below within the description of particular work packages. Technical objectives of the project are translated into work packages (WPs) as follows:
- WP1 - Return of experienced Czech scientist
- WP2 - Incoming post-docs
- WP3 - PhD course
- WP4 - Lectures
- WP5 - Incoming experts
- WP6 Workshops
- WP7 - Training at foreign institutes
- WP8 - Technology transfer
- WP9 - Project management.
Scientific objectives will be targeted by all WPs. Structural biology is relatively young research area aiming at understanding the three-dimensional (3D) structures of biological macromolecules-proteins, nucleic acids and sugars. In recent years, the techniques of structural biology have significantly advanced and generated many structures of biomolecules. However, the visualisation of isolated molecules is not the only goal. Biological molecules work in the context of cells and tissues, interacting with other proteins, nucleic acids and carbohydrates. Detailed knowledge of mutual interactions provides fundamental insights into molecular biology and feeds into the development of medical applications. We aim to enhance our research capacities to study biomolecular interactions and their structures with the ultimate goal to find the molecular basis behind a biological event. This requires interdisciplinary knowledge in physics, chemistry and biology and to be able to develop and apply the state-of-the-art techniques in this area. This program will enable us to obtain this interdisciplinary research experience through the scheme of support measures such as recruitment of foreign experts and post-docs, organisation of international postdoctoral courses and lectures given by internationally recognised scientists. In general, we can describe scientific objectives as three research topics (see below).
Project results:
The project was concentrated on the following three research topics:
Research topic 1 (RT1) - Repair of damaged DNA
Responsible scientists: Dr Krejci, Dr Stefl, Prof. Sklenar, Prof. Koca
Research topic 2 (RT2) - RNA quality control
Responsible scientists: Dr Stefl, Prof. Koca, Prof. Sklenar
Research topic 3 (RT3) - Pathogen-host recognition and interaction
Responsible scientists: Dr Wimmerova, Prof. Koca, Prof. Sklenar
Achievement in the respective research topics:
Research topic 1 (RT1) - Repair of damaged DNA
First project period
The integrity of a cell genome is constantly challenged by DNA lesions such as base modifications, nicks and double-strand breaks. A single unrepaired DNA double-strand break (DSB) is lethal and if repaired improperly may lead to loss of heterozygosity, mutations, deletions, genomic rearrangements and chromosome loss. DSBs are caused by a vast number of endogenous and exogenous agents including genotoxic chemicals or ionising radiation, as well as through replication of a damaged template DNA or the replication fork collapse. In most organisms, DNA DSBs can be repaired in an error- prone and error-free manner by non-homologoues end-joining (NHEJ) and homologous recombination (HR), respectively. This proposal focuses on HR that plays a vital role in DNA metabolic processes including meiosis, DNA repair, DNA replication, maintenance of rDNA homeostasis, and genome plasticity. The importance of HR in maintaining genome integrity is underscored by its high degree of evolutionary conservation from yeast to human. Several models were proposed to understand the mechanism of double-strand break repair by HR, identify separate steps and role of the individual proteins involved.
The aim of this topic is to determine the molecular details of homologous recombination and repair of DSB by characterising the functional and physical interactions various protein factors. Since defects in HR are often related to human genomic instability, the results from these studies will provide insight into not only the mechanism of DNA repair, but could potentially be used for the diagnosis, prevention, and treatment of human diseases including cancer.
Scientific progress
During the first six months we have identified new DNA binding domain within Rad52 protein and showed that this domain is responsible also for loading Rad51 protein on ssDNA and mediating strand exchange. We have also demonstrated the ability of central region of Rad52 protein to bind RPA, ssDNA binding protein. Interestingly this interaction is mediated only when RPA is ssDNA bound status. The results of this work were published.
Furthermore, we were able to identify and characterise the effect of Rad54 protein on the nuclease activity of structure-specific nuclease Mus81/Mms4. We showed that this activation is evolutionary conserved does not require ATP binding nor hydrolysis. In addition, Rad54 is capable of targeting Mus81/Mms4 complex to DNA substrates, which has been confirmed by genetic analysis of Mus81 foci formation in the absence of RAD54 gene. As a part of collaboration, we have studied the effect of Mph1 and Srs2 helicase on the ability to disrupt D-loop intermediated during the process of homologous recombination. The results of this work were also published.
We have mapped the interaction domain within Srs2 and Rad51 proteins and tested isolated mutants in biochemical as well as genetic experiments to define the biological significance of this interaction. The manuscript is now in the preparation. In addition, we have also mapped the sites of SUMO-modifications on Srs2 and Rad52 proteins and are characterising the effect of Sumoylation on Rad52 in great details.
In next year, we plan to further characterise the effect of SUMOylation on the biochemical properties of Rad52 protein. Within the first six months we plan to characterise the wild-type and mutants defective in their sumoylation for Rad52-mediated biochemical activities as well their eight genetic properties. In the rest of the year we will test the effect of Sumoylated for of rad52 on its activities. We will also focus on localisation of Srs2 protein during replication and DNA repair and its ability to serve as quality control mechanism together with Rad52 protein. Next we plan to characterise the molecular events leading to disassembly of Rad51 filament via the action of Srs2 helicase.
Second project period
Short overview:
DNA double-strand breaks (DSBs) are one of the most toxic kinds of damage, implicated in cancer and many other diseases. Just a single un-repaired break can lead to aneuploidy, genetic aberrations or cell death. DSBs are caused by a vast number of both endogenous and exogenous agents including genotoxic chemicals or ionising radiation, as well as through replication of a damaged DNA templates or the replication fork collapse. To maintain the integrity of the genome it is essential to recognise and process DSBs as well as other toxic intermediates and launch most appropriate repair mechanism. How the cells achieve specificity, efficiency and cooperativity in the complex protein-protein interaction networks remains unclear. Therefore, we have designed specific aims that should uncover parts of this complexity. The link between defects in the processes that maintain genomic integrity and pathological disorders could shed light onto the knowledge of the molecular mechanisms behind it and potentially be targeted for therapeutic intervention.
During the past year, we have been involved in biochemical characterisation of the role of post-translational modification by SUMO on the biochemical properties of protein involved in homologous recombination. First we looked at effect of sumoylation on biochemical properties of the Saccharomyces cerevisiae mediator protein Rad52. Interestingly, binding of ssDNA enhances sumoylation of Rad52, and we demonstrated that modification of Rad52 inhibits its biochemical activities. It not only inhibited the ability of Rad52 to bind DNA but also catalyze annealing of complementary strands. On the other hand the Rad52 sumoylation did not have any effect on its protein-mediated interactions (oligomerisation, interaction with Rad51 and RPA). In addition, we also mapped the modification sites and generated sumo-deficient mutants that were characterised both in vitro and in vivo.
Furthermore, we have been also characterising the possible role of Srs2 in regulation of homologous recombination. Despite the role of HR contributing to the elimination of DSBs, it must be tightly regulated to prevent untimely events that could interfere with other DNA repair or replication. We have been collaborating with Dr Ellenberger on characterisation of Srs2 anti-recombinase activity. Biochemical characterisation revealed that led that a physical interaction between the C-terminal region of Srs2 and Rad51 protein and triggers ATP hydrolysis within the Rad51 filament, resulting in Rad51 dissociation from DNA. Durand last year we have also focused on characterisation of possible quality control mechanism that use both positive (Rad52) as well as negative (Srs2) regulators to shift the equilibrium of homologous recombination. Based on cell biology data we have suggested that Srs2 antagonises Rad52 in the formation of Rad51 filaments. This was further supported by the ability of Rad52 protein to suppress the inhibitory effect of Srs2 on Rad51-mediated strand exchange. Furthermore, the Rad51 mutants, originally isolated as Rad52-interaction deficient mutants, appeared also defective in the interaction with Srs2, suggesting that these proteins compete for the same interaction region. As expected, these Rad51 mutants show resistance to the action of Srs2 as well as inability to overcome the RPA inhibition on Rad51-mediated strand exchange. Accordingly, an Srs2 mutant protein, that fails to interact with Rad51, is not sufficient to disrupt the Rad51 presinaptic filament in vitro as well as in vivo. To further elucidate the mechanism that can negatively regulate recombination we have been involved in the identification of another regulatory mechanism requiring Mph1 helicase. In this pathway, Mph1 uses its translocase activities to unwind intermediate of homologous recombination, D-loop.
Next year, we plan to characterise the role of post-translational modification of Srs2 by SUMO peptide. Similarly to Rad52 we plan both biochemical as well as genetic characterisation of sumo-deficient mutants. We also plan to determine the role of Srs2 in replication and DNA repair synthesis. For this purpose we are developing D-loop assay linked with the extension of incorporated primer. We will test know replication factor that might be also involved during repair synthesis. These results should shed light in to the downstream steps of homologous recombination.
Third project period
DNA double-strand breaks (DSBs) are one of the most highly toxic kinds of damage, implicated in cancer and many other diseases. DSBs are caused by a vast number of both endogenous and exogenous agents including genotoxic chemicals or ionising radiation, as well as through replication of a damaged template DNA or the replication fork collapse. It is essential for cell survival to recognise and process DSBs as well as other toxic intermediates and launch most appropriate repair mechanism. How the cells achieve specificity, efficiency and cooperativity in the complex protein-protein interaction networks remain unclear. Therefore we have designed three specific aims that should uncover parts of this complexity:
1) resection of DNA breaks;
2) processing of other intermediates during various DNA metabolic processes as well as DNA and protein crosslinks;
3) high throughput screen identifying potential inhibitors of nucleases involved in these processes, and investigation of their effect on genome stability.
To achieve our goals, we plan to utilise challenging multi- as well as inter-disciplinary approaches combining biochemical, genetic, biophysical and cell biological characterisation combined with low- and high- resolution structural approaches and small molecule analysis. Our studies should contribute towards understanding how the cell responds to DNA damage and maintain its genomic integrity as well as for cancer therapeutics and gene targeting, as they could open the new horizons and opportunities for the predetermined control of homologous recombination and other types of repair.
During past six months, we have tested the effect of Rad52 sumoylation on its in vivo activities. To address the effects of Rad52 SUMOylation in vivo, we examined the SUMO-deficient mutants (rad52-K43,44R, -K253R or -K43,44,253R) using several recombination assays. We found that these mutations did not affect the rate of intra- or inter-chromosomal recombination during mitosis and meiosis. However, they lead to a shift from single-stranded annealing to gene conversion events. In addition, these mutations resulted in slight hyper-recombination phenotype in rDNA recombination. Taken together, our data suggested that SUMOylation of Rad52 facilitates certain recombination sub-pathway and disfavours the others. Finally, we also tested the in vivo localisation of Rad52 SUMO-deficient mutant by cytological analysis. The fusion of wild type as well as the rad52-K43,44,253R proteins to YFP revealed that the mutant protein forms foci similarly to wild type, but the duration of the foci is significantly shorter. In addition, we have also showed that also Rad59 protein is sumoylated both in vitro as well as in vivo after the DNA damage. We have mapped the residue that are modified and confirmed the mutation within this residues abolish sumoylation in vivo. Biochemical characterisation did not show any effect on its activities, including DNA binding and annealing of complementary strands. Interestingly, we have observed that Rad59 dramatically inhibited sumoylation of Rad52 in vitro and this inhibition was specific. This was oppose effect of Rad59 on Rad52 activity was later on confirmed using Rad59-sumodificent mutant that was able to suppress most of the rad52-K43,44,253R phenotype. The genetic experiments were done in close collaboration with Dr Michael Lisby with him we started to collaborate in genetic as well as cell biological characterisation of sumo-deficient mutant of recombination proteins. The results of this study are now in preparation for publication.
We were also able to successfully reconstitute a DNA repair synthesis step of homologous recombination. This assay is based on extension of D-loop structure formed by coordinated action Rad51 and Rad54 proteins. The actual extension is dependent on PCNA, its loader (RFC) and polymerase delta. RPA protein was found to stimulate the extension of longer products suggesting that it prevents formation of secondary structures on ssDNA that would otherwise block the DNA extension. The length of extension product depends on salt concentration and reaches up to 1 kb, which is in good agreement with genetic studies. Furthermore, pl epsilon was also capable of extension of D-loop substrate, but resulting on shorter extension products. In addition, we have also tested the effect of Srs2 and Mph1 helicases to unwind newly extended DNA. While Srs2 failed to do so, Mph1 was fully capable of this unwinding, suggesting its biological role in vivo. These results shed light into the downstream steps of homologous recombination and were published in journal DNA Repair. Next, we have used this assay to test the effect of SUMOylated PCNA that was show to play major role in regulation of recombination during the repair of stalled replication forks. Our preliminary data show the ability of Srs2 to block the extension of D-loop substrate. Interestingly, this activity does not require the helicase / translocase activity of Srs2, but only the SUMO interaction motif on the C-terminus of Srs2. The detail analysis of the mechanism revealed the ability of Srs2 to disrupt the direct interaction between Pol delta and SUMO-PCNA, which is required for primer extension. Manuscript with this work was recently submitted.
Finally, we have performed high throughput screening of chemical library of 30 000 compounds available to our collaborator Dr Bartunek (Institute of Molecular Genetics, Prague). For the purpose of the screen we have developed a novel liquid fluorescent-based assay with high detection limits that allows us to use nM concentrations of proteins and substrates. Among multiple nucleases available in our laboratory we have screened inhibitors against Mus81/Eme1 complex. We have identified several lead compounds that will be further characterised. Several other human nucleases from the same protein families as well as yeast homologues are now being tested to determine the specificity and selectivity of particular inhibitors.
Research topic 2 (RT2) - RNA quality control
First project period
Expression, labeling and purification of the entire TRAMP complex and individual recombinant subunits
In order to investigate the specific affinity of the TRAMP complex for the RNA substrates, additional DNA constructs were prepared for the expression of recombinant wild type and mutant and deletion forms of the two subunits of the TRAMP4 complex, the Trf4 and Air2 proteins. These proteins are currently being tested for RNA binding by gel shift assay in order to identify the protein signature for RNA binding. Subsequently, several tests were performed to obtain the reconstituted homogeneous population of the minimal PAP TRAMP4 complex. Currently, experiments with additional deletion versions and are in progress.
TAP tag purification of the TRAMP complex from yeast and MS-MS analysis of yeast TRAMP complex
We have optimised the conditions for complex affinity purification from yeast in order to analyze the subunits for potential posttranslational modifications. Since the amounts of the affinity purified complexes were low for the MS analysis of posttranslational modifications, we have been testing different approaches to scale up the purification procedures.
Mapping of interaction between RNA-quality-control-related proteins and RNA using NMR spectroscopy
Structure determination of protein-RNA complexes is challenging and involves many biochemical and biophysical issues. Only recently, the structure determination of protein-RNA complexes became possible by NMR spectroscopy. Often, truncated forms of the studied protein-RNA complexes have to be generated in order to obtain a well-behaving system for NMR structure determination. We started to map interactions between the proteins that are involved in RNA quality control. We always work with several protein and RNA constructs for each system. The designed constructs do not rely only on the computational alignment strategies but also on limited proteolysis and functional assays. This careful strategy excludes the possibility of truncating important elements out of the studied constructs. Using this approach, we identified several systems that are suitable for the structure determination by NMR. It is the minimal Nrd1-Nab3 heterodimer that has same RNA-binding properties as the full-length Nrd1-Nab3 heterodimer; the RNA-binding domains of both Nrd1 and Nab3. These constructs are currently used not only for the mapping of interaction with RNA by also for the structure determination by NMR.
Second project period
Protein preparations and purifications
Our laboratories continued in preparation of proteins that are responsible for the studied mechanism of RNA quality control. We have produced several constructs of Trf4 that are used for functional assays and crystallisation trials. For NMR structural studies, we produced PIN domain of Rrp44 (catalytic subunit of the exosome), Nab3 and Nrd1 proteins. We experienced problems with expression and solubility of some protein constructs. For example, none of the expression conditions, bacterial host strains, or solubility enhancement tags, provided soluble and folded Air2 protein. This problem was solved using SUMO-tag that helped as a folding chaperone during the refolding procedure of Air2 protein. We also prepared Nrd1-Nab3 heterodimer complex using bacterial co-expression system.
Preparation of TRAMP mutant strains for in vivo studies
We have prepared a series of deletion mutants of the catalytical subunit Trf4 to screen for the role of each of the protein domains in vivo. We have confirmed the constructs by sequencing. In parallel we have prepared a series of mutants of the RNA-binding subunit Air2 by site-directed mutagenesis.
Structural determination of Nab3 and mapping of its interaction with RNA using NMR spectroscopy
We have determined the three dimensional structure of RNA binding protein Nab3 that is involved in processing of functional RNAs such as sn/snoRNAs. The three-dimensional structure of Nab3 RRM adopts a compact fold with an 1-1-2-3-2-4 topology. The fold is composed of the two helices that are packed along a face of a four-stranded antiparallel sheet (4-1-3-2). We have identified the RNA interaction interface of Nab3 and the residues that are important for the interaction with RNA
Development of NMR methods and their application to biological problems
Development of NMR methods for studies of proteins and their complexes continued in two directions. First, an attention has been paid to the accuracy of determination of small residual dipolar couplings (RDCs), serving as a source of structural and motional information. An earlier developed method of experimental error correction has been implemented into a software S3E.py which is currently available to public and was published in a peer-reviewed journal. Second, application of high-resolution NMR methods to unstructured proteins has been successfully tested. Sufficient resolution for a complete assignment of highly degenerated frequencies in a partially disordered RNA polymerase delta subunit from Bacillus subtilis was achieved using 5D non-uniformly sampled NMR spectra.
The TRAMP complex analysis
We have started a detailed analysis of the TRAMP mutant strains. The phenotypic tests identified 5 Trf4p point and deletion mutants, respectively with strongly impaired growth when expressed in the double deletion trf4trf5 background. The same mutants expressed in trf4 delta showed defects in pre-rRNA processing as tested by Northern blot analysis by accumulating 23S precursor of the 18S rRNA. The wild type and mutant forms of Trf4p were subsequently affinity purified via the N-terminal fusion protein A tag by using IgG beads. Purified proteins were assays for polyadenylation activities. The mutants that showed growth and rRNA processing defects also displayed no polyadenylation activities. In parallel, we have expressed and purified the same recombinant deletion mutants of Trf4p from bacteria and their polyadenylation activity was tested upon the addition of recombinant Air2 protein. Similar studies were performed with the Air mutant strains and mutant recombinant proteins. Wild type and mutant forms of Air2p were episomally expressed in the double deletion strain air1air2. Surprisingly only one of the mutants showed only minor growth defects. The same strain revealed defects in rRNA processing. However, in vitro complementation assays of polyadenylation activity revealed two mutants that lost the ability to activate Trf4p in vitro. Currently, we are performing additional experiments to confirm these results.
Recognition of phosphorylated RNA polymerase II C-terminal domain by protein factors involved in the Nrd1 termination pathway
Transcription is coupled to RNA processing and RNA quality control machineries by the carboxyl-terminal domain (CTD) of RNA polymerase II, which consists of up to 52 repeats of the sequence Tyr1-Ser2-Pro3-Thr4-Ser5-Pro6-Ser7 in mammals. The CTD heptapeptide repeat is dynamically phosphorylated and dephosphorylated on serines in the course of transcription cycle and promotes the recruitment of various enzymes and RNA processing factors to the transcription site. In addition, cis-to-trans isomerisation of Pro3 and Pro6 promoted by prolyl isomerases serves a conformational switch, which further enhances the CTD code that controls binding of transcription and processing factors involved in RNA biogenesis.
We have started a detailed analysis of the CTD recognition by the Nrd1 processing factor that is modulated by phosphorylation and proline isomerisation. We prepared and labeled the Nrd1 protein. Using nuclear magnetic resonance spectroscopy, we determined the three-dimensional structure of Nrd1 in the free form. Currently, we are working on the structure of Nrd1 bound to the CTD peptide phosphorylated at serine 5. We also investigated how the Nrd1 recognition process is affected by another factors, a prolyl isomerase called Ess1, and a phosphatase, Ssu72. We found out that Ess1 contributes to the disassembly of the Nrd1-CTD complex and that Ess1 stimulates dephosphorylation mediated by Ssu72. Future studies will enable us to fully understand the structural basis behind the 'CTD code' and how it is used for the recruitment and dissociation of Nrd1 complex in the mechanism of poly(A) independent termination.
Third project period
Development of NMR methods and their application to biological problems
Development of NMR methods was focused on techniques for characterisation of unstructured proteins. The recently developed methodology of 5D non-uniform sampled NMR spectroscopy and the resulting complete assignment of highly degenerated frequencies of Bacillus subtilis RNA polymerase delta subunit has been published in a peer-reviewed journal. The methodology was further improved by introducing direct C-13 detection to enhance resolution in the direct dimension. The developed protocol allowed us to obtain a complete sequential assignment of the above mentioned testing sample from a single 5D spectrum. Two 5D NMR experiments based on the introduced technique were designed and a manuscript describing the approach and its applications has been submitted.
The previously published methodology of 5D non-uniform sampling was further developed. Direct C-13 detection and recycle delay optimisation resulted in a set of 5D experiments allowing an easy assignment of unstructured proteins with very low signal dispersion. The method has been published a peer-reviewed scientific journal and tested on difficult samples of unfolded proteins, including a protein with a high degree of sequence repetition and a proline-rich protein. The development continued by designing versions of the mentioned experiments with improved sensitivity.
Recognition of termination signals in poly(A)-independent RNA processing
We significantly progressed in our long-term goal to reveal the structural and mechanistic bases for the termination and processing of functional RNA molecules. In our work we focused on the structural determination of the Nrd1 complex. The Nrd1 complex includes two RNA-binding proteins, the nuclear polyadenylated RNA-binding (Nab)3 and the nuclear pre-mRNA down-regulation (Nrd)1 that bind their specific termination elements. We determined the solution structure of the RNA-recognition motif (RRM) of Nab3 in complex with UCUU RNA representing the Nab3 termination element. The structure shows that the first three nucleotides of UCUU are accommodated on the beta-sheet surface of Nab3 RRM, but reveals a sequence specific recognition only for the central cytidine and uridine. The specific contacts we identified are important for binding affinity as well as for proper transcription termination in vivo. In our mechanistic studies, we found that both RNA-binding motifs of Nab3 and Nrd1 alone bind their termination elements with a weak affinity. Interestingly, when Nab3 and Nrd1 form heterodimer, the affinity is significantly increased, most likely in a cooperative manner. These finding are in accordance with the model of their function in the poly(A) independent termination, in which the binding to the combined and/or repetitive termination elements elicits efficient termination. A manuscript describing the structure of the complex has been submitted for publication.
Recognition of C-terminal domain of RNA Polymerase II by Nrd1
RNA processing factors bind and process only the nascent RNA transcript but also interact and travel with RNA polymerase II during transcription. This phenomenon is called co-transcriptional processing in which a number of processing machineries associate with RNA polymerase II, creating a large processing assembly. Nrd1-dependent termination apparatus that is under investigation in this project also associates with RNA Polymerase II. To reveal this interaction, we studied how Nrd1 binds RNA Polymerase II. Recruitment of Nrd1 the site of transcription is controlled by phosphorylation of the C-terminal domain (CTD) of RNA Polymerase II. We determined the solution structure of Ser5 phosphorylated (pSer5) CTD bound to the CTD-interacting domain (CID) of Nrd1. The structure reveals a direct recognition of pSer5 by Nrd1 CID upstream of the CTD region usually recognised by other CID-containing proteins. The specific phosphoserine recognition requires the cis conformation of the upstream pSer5-Pro6 prolyl bond of the CTD peptide. Mechanistic studies by NMR show that Ess1 prolyl isomerase disassembles the Nrd1 CID-pSer5 CTD complex by isomerisation of the phosphoserine-proline prolyl bond. Mutations at the complex interface diminish binding affinity and impair processing of small nucleolar RNA, suggesting that the Nrd1-pSer5 CTD interactions are important for processing of non-coding RNAs. Our findings underpin the importance of both covalent and non-covalent changes in the CTD structure that constitute the CTD code. Furthermore, we suggest that Ess1 regulates directly association of Nrd1 with RNAP II, in addition to its role in the indirect regulation via stimulation of dephosphorylation by Ssu72. A manuscript describing the structure of the complex and the functional data is in preparation.
The TRAMP complex analysis
We have followed up the studies of the strains mutant in individual subunits. The point mutants of AIR2 that previously showed lack of Trf4p activation and growth defects in vivo were expressed in double deletion strain of air1 and air2 and affinity purified on IgG agarose. Because they failed to strongly copurify the other tw components of the TRAMP complex, we conclude that the regions mutated are responsible for intermolecular contacts within TRAMP4. Thus we extended the spectrum of Air2p mutants. We prepared deletion forms lacking region at the N- and/or C- termini, respectively. The deletion versions of Air2p were again expressed in the double deletion strain and assayed for growth. We have observed new motifs at the C- and N- termini that are likely responsible for Trf4p and Mtr4p interactions. In agreement with the growth defects, additional Northern blot analyses showed accumulation of cryptic unstable transcripts in mutants with major growth impairment.
To complete or analysis of the TRAMP4 complex, we prepared a set of deletion mutants in the third subunit Mtr4p. These mutants were introduced into the strain where endogenous MTR4 was deleted and the growth was rescued by wild type Mtr4p expressed from a URA3 plasmid. Growth test analysis on 5-FOA plates at different temperatures confirmed the lethality of most of these mutants, as most of them contained deletions in conserved helicase domains. Next, we will perform pull-down experiments to find out which parts of Mtr4p are crucial for interactions with Trf4p and Air2p in vivo.
Functional importance of association between Nrd1 and RNA polymerase II
We further studied the functional relevance of interaction between RNA Polymerase II and Nrd1-depedent termination apparatus that is involved in RNA processing. Specific association of pSer5 CTD with Nrd1 CID was further tested in a quantitative in vitro solution binding assay using FA experiments. We titrated the wild-type and mutants of Nrd1 CID against fluorescently labeled pSer5 CTD. Alanine or aspartate (charge swapping) substitutions at positions Ser25, Arg28, Ile29, and Lys30 significantly decreased the binding affinity with pSer5 CTD. These FA data strongly support the phosphorylation-specific recognition observed in the structure of Nrd1-pSer5 CTD complex and that the interaction relies upon the presence of the invariant basic residues in the CID domain. To determine the importance of these individual residues for Nrd1 function in vivo, we monitored cell viability and snoRNA expression in single amino acid mutants. The Nrd1 strain lacking CID (Nrd1-1-150) was used as a reference for the CID-related function. Deletion of CID is not lethal, however, we observed that deletion of CID did not support viability at 37 °C. Similar, to a lesser extent temperature sensitive growth defect was observed in mutants of Ser25, Arg28, Ile29 and Arg74 mutants. As both Ser25 and Arg28 are contacting the phosphorylated serine, we prepared Ser25+Arg28 double mutant that exhibited the same growth phenotype as CID Nrd1. Although CID Nrd1 is viable, the D70R mutant caused lethality upon the depletion of endogenous wt Nrd1. The lethality of D70R is not due to the absence of the mutant protein, but it destabilises the CID fold that likely interferes with the Nrd1 functions (D70R displays read-through for snR33). Nrd1 CID deletion causes an accumulation of snR33 precursors. We have observed that point mutants with ts growth phenotype showed an accumulation of pre-snR33. Importantly, the Ser25+Arg28 double mutant showed virtually identical processing defects of snR33 as for CID Nrd1. These data demonstrate that specific recognition of pSer5 CTD by Nrd1 CID is important for snoRNA processing in vivo.
Research topic 3 (RT3) - Pathogen-host recognition and interaction
First project period
Short overview
Saccharides are known to have a variety of more sophisticated and complex functions in cell-cell interaction and communication. The glycorecognition by carbohydrate-binding proteins, lectins, is important in the mediation of numerous physiological processes including fertilisation, pathogencell adhesion and recognition, inflammatory response and others. A number of pathogen microorganisms utilise lectin-carbohydrate interaction to recognise and infect host organisms.
Lectins are proteins of non immune and non catalytic origin which specifically bind carbohydrates. Understanding of host / pathogen interaction is the first step in designing inhibitor molecules as potent therapeutics. The key goal of the project is to understand pathogen-host recognition based on protein-carbohydrate recognition and to find a way how to inhibit both the pathogen adhesion and the formation of biofilms, which are highly resistant to classical antibiotic therapy.
Main results obtained during the first year of the project
We have focused on the cloning homologues of PA-IIL lectin, one of the virulence factors of Pseudomonas aeruginosa, which we identified in the genome of Burkholderia cenocepacia. Four lectins named BclA, BclB, BclC and BclD were cloned and prepared in the recombinant form. Since oligosaccharide specificity could be correlated to the possible function of the bacterial lectin in tissue recognition, binding to a broad spectrum glycan array was performed at the Consortium for Functional Glycomics. The screening results showed BclA to have a strict specificity for oligomannose type N-glycans, only binding to oligosaccharides with a mannoside capped branch. The glycan array results obtained on the other lectins were non-clear. All three lectins showed a mixed specificity toward human fucosylated as well as mannosylated epitopes. From the prediction methods developed in our laboratory the BclB protein should display fucose-binding preference whereas BclC and BclD mannose-binding preference, respectively. All three mentioned lectins contain in its amino acid sequence an additional N-terminal domain of unknown function that can be responsible for such behaviour.
Therefore, we separated these proteins on individual domains to decipher the function of the Nterminal part to clarify specificity of the lectin parts of the proteins. Important results were obtained with BclC protein. Glycan array of individual domains showed that C-terminal part of the protein BclC, that we expected to be a mannose-recognising lectin part, certainly recognises only human mannosylated epitops. To our surprise, N-terminal domain of unknown function can also bind sugar epitopes and it is specific towards fucosylated oligosaccharides. Thus, we identified a novel fucosebinding lectin that doesn't have any homologue in current sequence databases.
In collaboration with the group of Dr Anne Imberty, we characterised the specificity and affinity of the galactophilic lectin PA-IL for Gal1-4-Gal and Gal1-3-Gal epitopes by cell surface labeling combined with glycan array analysis and titration microcalorimetry. The crystal structure of PA-IL complexed with Gal1-3-Gal1-4Glc trisaccharide establishes the atomic basis of the specificity and reveals how the second galactose residue makes specific contacts with the protein surface (Blanchard et al., 2008). Following study was designed to evaluate the respektive contribution of PA-IL and PA-IIL lectins in the pathogenesis of P. aeruginosa-mediated acute lung injury. Using an in vitro model with A549 cells, and an experimental murine model of acute lung injury, comparison of the parental strain to mutants' strains with inactivated PA-IL and PA-IIL gene were performed.
In addition to mutagenesis in vitro that we have been performing on PA-IIL protein, we have been working on development of reliable mutagenesis of protein in silico to speed up the process of protein engineering of lectins. The graphical interface TRITON is under development in our laboratory. Program TRITON is a user oriented software with graphical interface that enables visualisation of molecular structures, preparation of input files for computational software and analysis of output data. It interfaces, for example, external software Modeler. Unavoidable part of the in silico protein engineering is a quantitative prediction of binding properties newly constructed lectins. Programme AutoDock was implemented into TRITON interface and the system is being parameterised for the work with saccharides. The results were published. By combination of experimental and computational methods, the effect of terminal GLY114* deletion on the binding affinity of the PA-IIL lectin towards sugars was investigated. Both have shown that the deletion mutation decreases the affinity of the mutant towards saccharides and confirmed that PA-IIL saccharide binding affinity is influenced by the dimerisation of the lectin. Corresponding manuscript has been accepted for publication.
Second project period
Short overview
Saccharides are known to have a variety of more sophisticated and complex functions in cell-cell interaction and communication. The glycorecognition by carbohydrate-binding proteins, lectins, is important in the mediation of numerous physiological processes including fertilisation, pathogen-cell adhesion and recognition, inflammatory response and others. A number of pathogen microorganisms utilise lectin-carbohydrate interaction to recognise and infect host organisms. Lectins are proteins of non immune and non catalytic origin which specifically bind carbohydrates.
Understanding of host / pathogen interaction is the first step in designing inhibitor molecules as potent therapeutics. The key goal of the project is to understand pathogen-host recognition based on protein-carbohydrate recognition and to find a way how to inhibit both the pathogen adhesion and the formation of biofilms, which are highly resistant to classical antibiotic therapy.
Key results obtained in the second year
We have focused on the cloning homologues of PA-IIL lectin, one of the virulence factors of Pseudomonas aeruginosa, which we identified in the genome of Burkholderia cenocepacia. Four lectins named BC2L-A, BC2L-B, BC2L-C and BC2L-D were cloned and prepared in the recombinant form. All three later mentioned lectins contain in its amino acid sequence an additional N-terminal domain of unknown function that can be responsible for such behaviour. As already reported, we separated these proteins on individual domains to decipher the function of the N-terminal part.
The N-terminal part of the BclC lectin was completely characterised. The recombinant Nterminal BC2L-C domain is a new lectin with specificity for fucosylated human histo-blood group epitopes H-type 1, Lewis B and Lewis Y, as determined by glycan array and isothermal titration calorimetry. Methyl selenofucoside was used as ligand to solve the crystal structure of the N-terminal BC2L-C domain. Additional molecular modelling studies rationalised the preference for Lewis epitopes. The structure reveals a trimetric jelly-roll arrangement with striking similarity to TNF-like proteins (Sulak et al., 2010).
In collaboration with Prof. Miguel Volvano and Dr Anne Imberty, the presence of BC2L-A, BC2L-B, BC2L-C on the outer membrane of the wild-type bacteria B. cenocepacia was confirmed, albeit with unknown secretion mechanism. This suggests that studied lectins of B. cenocepacia can play important role in host recognition or biofilm stabilisation.
Several native ligands have been studied to characterise binding properties of individual proteins and their domains. We continue also in mutagenesis of selected lectins to characterise key amino acids for affinity and/or specificity of binding. The mutagenesis is performed both in vitro (experimentally) and in silico (using in house software Triton being developed in our laboratory).
A site-directed saturating mutagenesis of amino acids in the specificity binding loop of PA IIL showed that any substitution of Ser at position 22 leads to fucose preference declining. Further characterisation of the interactions is being performed by molecular dynamics simulations.
Third project period
Short overview
Saccharides are known to have a variety of more sophisticated and complex functions in cell-cell interaction and communication. The glycorecognition by carbohydrate-binding proteins, lectins, is important in the mediation of numerous physiological processes including fertilisation, pathogen-cell adhesion and recognition, inflammatory response and others. A number of pathogen microorganisms utilise lectin-carbohydrate interaction to recognise and infect host organisms. Lectins are proteins of non-immune and non-catalytic origin which specifically bind carbohydrates.
Understanding of host / pathogen interaction is the first step in designing inhibitor molecules as potent therapeutics. The key goal of the project is to understand pathogen-host recognition based on protein-carbohydrate recognition and to find a way how to inhibit both the pathogen adhesion and the formation of biofilms, which are highly resistant to classical antibiotic therapy.
Main results obtained during the third year time
Our laboratories have continued in key lectins characterisation from Pseudomonas aeruginosa and Burkholderia cenocepacia pathogens. As already mention in the previous report, the very interesting features were connected with one of the B. Cenocepacia lectin BC2L-C.
The N-terminal part of the BclC lectin was structurally and functionally characterised as a novel fucose binding domain with a TNF-like fold (Sulak et al, 2010) while the C-terminal part is similar to a superfamily of calcium-dependent bacterial lectins. Recently, we have demonstrated that the C-terminal domain displays strict specificity for mannose- and heptose-containing oligosaccharides. Thus, BC2L-C is a 'superlectin' with mixed specificity for mannose / heptose glycocompounds and fucosylated human histo-blood group epitopes. The overall structure, determined by electron microscopy and small angle x-ray scattering, consists of a ring of three mannose / heptose-specific dimers flanked by two fucose-specific TNF-like trimers. We demonstrated that BC2L-C is immunogenic, binds to the bacterial surface and can be released into the medium. The unique architecture of this first superlectin correlates with multiple functions including bacterial cross-linking, adhesion to human epithelia, and stimulation of inflammation. Results were recently published in Plos Pathogen.
With new genome sequencing, several potential lectins were identified in the Aspergillus fumigatus genome. We focused on one of the predicted protein, and prepared it in a recombinant form. The recombinant protein binding activities were determined by hemagglutination studies, glycan array, surface plasmon resonance (SPR) and isothermal titration calorimetry (ITC). We confirmed its preference for fucose and fucosylated oligosaccharides. The 3D structure of AFL with fucosylated ligands was described using X-ray diffraction.
We continue also in mutagenesis of selected lectins to characterise key amino acids for affinity and/or specificity of binding. The mutagenesis is performed both in vitro (experimentally) and in silico (using in house software Triton being developed in our laboratory).
As stated in a previous report, a site-directed saturating mutagenesis of amino acids in the specificity binding loop of PA IIL showed that any substitution of Ser at position 22 leads to fucose preference declining. The mutations in the position 23 generally turned out to have similar effect. The mutations of the position 24 were therefore considered to be most promising. First and foremost, the side chains in this position have lesser potential to sterically hinder the binding and greater potential to actually create new interactions with the atoms of the ligand, basically being positioned not directly into the binding site. Except for significantly lower energies in cases of the detrimental mutations, there was not any mutation that would be clearly more favourable than the others. This is caused mainly by the fact that the side chain of the amino acid 24 usually sticked out of the binding site after Modeller perform the in silico mutagenesis. In order to explore the possibilities of the binding site actually accommodating for the sugar binding and forming new interactions, molecular dynamics simulations were performed at the results provided by the docking programs. The best docked structures of methyl-fucoside and methyl-mannoside were taken as the starting structures for molecular dynamics simulation. Besides the aspartate, glutamine and glutamate mutants, a tryptophan mutant was chosen as a representative of the aromatic bulky side chains. The trajectories were analysed both from energetic and geometric point of view, searching for correspondence between changes of energy and distances of chosen atoms. The atoms whose distances were analysed were chosen on the basis of visual analysis of the trajectory.
As already mentioned, we had proved that the specificity of the lectins in PA-IIL family is governed by the composition of a triad of amino acids at positions 22-23-24, and that the change of serine 22 (present in PA-IIL) to alanine (present in RS-IIL) caused the mutant to change the preference from fucose to mannose as a binding partner. To better understand the binding, we focused on other member of the PA-IIL family, CV-IIL, which binds fucose and mannose with similar affinities. Therefore, it appeared as an ideal molecule for testing the changes of specificity during mutagenesis of other amino acids.
Two approaches were selected, i.e. mutagenesis in silico and in vitro. We were mainly interested in performance of in silico methods once we employ more mutations in the same time. The performance of the Modeller software was therefore tested in relation to the different possibilities of creating the mutant proteins. Two main features were studied: the order of the mutations being performed on the protein, and the presence of the saccharide in binding site during in silico mutagenesis. As binding partners, the two 'border' saccharides, methyl-L-fucoside and methyl-D-mannoside were docked into the resulting mutants. The computational part has shown several important facts. In a surface binding site, where the positions of side chains important for binding are not spatially hindered thanks to close quarters, the way of performing the mutation was crucial for the overall result. The resulting structures can significantly differ, especially when mutation were done simultaneously and not sequentially. Also a presence of a saccharide in binding site has a beneficial effect on the results.
Potential impact:
South Moravia is a perspective region with a good infrastructure of human resources, functional institutional foul territory as well as entrepreneurial sphere. Main regional city Brno (400 000 inhabitants) is judged as having the best economic potential in the Central Europe. It is predicted that the city's rapid economic growth will continue (see http://www.fdimagazine.com for further details) due to a favourable structure of undergraduates, especially in biotechnological and technical fields. Major universities and several institutes of the Czech Academy of Science in Brno as well as industry oriented on life sciences and the local authorities were accordingly to their position involved in the project.
NCBR and POSTBIOMIN project were part of the regional activities in research, development and innovation. Being an excellent a very well networked research centre in the field of life sciences, NCBR actively contributed to the emergence of new research centres in South Moravia, which are financed by the EU Structural Funds. The research team of NCBR is the nucleus of structural biology research programme of an emerging European Centre of Excellence, CEITEC (see http://www.ceitec.cz for details).
CEITEC is a joint project of universities and research organisation from Brno, Czech Republic, coordinated by the Masaryk University. With a budget of more than EUR 200 million, it strives to establish a new European Centre of Excellence in the field of life sciences and advanced materials and technologies.
NCBR also actively participated in the following activities in the regional and national context during POSTBIOMIN project:
- participation on formulation and implementation of regional innovation strategy of South Moravian Region
- preparation of Czech position on future Eight Framework Programme (FP8) - active participation in working group on future of the European Research Center (ERC)
- advisory role in the SOMOPRO programme (research fellowships funded by FP7-COFUND, see http://www.jcmm.cz/en/somopro.html online) of South Moravian Centre for International Mobility
A. Impact of knowledge base
POSTBIOMIN project had a wide impact at different levels:
- Developing existing and emerging research potential
The huge importance of proper training for efficiency gains on the research site. Undergraduates not only Masaryk University bring high proficiency knowledge into the sector they work in, including further research. The benefits of training greatly outweigh costs by improving productivity and increasing equipment life trough proficient knowledge and modern research and scientific techniques.
During the project a new generation of post-docs emerged at NCBR. Most of them are staying at NCBR also after the end of POSTBIOMIN project. This new generation of researchers is therefore vital for the future of NCBR and its scientific excellence.
- Improvement of the capacity and increase visibility
POSTBIOMIN provided for a long term financing of some of the activities done by NCBR. This gave sufficient space for strategic decisions to be made and implement long term strategies. The fact that NCBR was able to attract such an interesting project (in terms of the budget size) gave a lot of additional visibility and also credibility when dealing with various Czech and international funding agencies. This can be proved also by the international evaluation of CEITEC project by an independent international panel on behalf of the Czech Ministry of Education. Structural biology research programme, where NCBR plays a crucial role was evaluated as the best research programme in CEITEC overall.
A number of FP7 proposals were submitted as a result of POSTBIOMIN project. They include also the following proposals:
1. EASTNMR, FP7-Infrastructures - selected for funding, NCBR as project partner, start - February 2009
2. BIONMR, FP7-Infrastructures - selected for funding, NCBR as project partner, start - autumn 2010
3. ERC Starting Grant - Lumír Krejí - submitted, not selected for funding
4. Richard Stefl - considered for ERC starting grant submission
5. Stepanka Vanacova - considered for ERC starting grant submission.
- New treatment
The findings linked to various human diseases including cancer in this project may well be of clinical value in the long term.
B. Contribution to community societal objectives
The project clearly contributed to several societal objectives of the EU:
- Training and preparation of young scientists and researchers
Training within the project took part in several ways. First of all postdoctoral training took place in WP2. Furthermore, given the fact that one of the WPs dealt with technology transfer a number of people from NCBR received training in technology transfer and management essentials.
- Cohesion and cooperation in the EU
The project was an excellent example of cooperation among EU Member States. During the project life time, a whole range and number of cooperation with institutes from all over the EU. This included mainly on the one hand visits of senior researchers from around the EU in Brno for lectures, PhD courses, workshops and activities in WP5 and on the other hand trips and trainings of NCBR staff to foreign institutes.
C. Economic impact
POSTBIOMIN had also direct scientific, technological and indirect economic impact in following terms:
- R&D
New and innovative methods and techniques were brought to the Czech Republic by the means of trainings at foreign institutes. This is for the benefit of the whole Czech life sciences research community.
- Increased collaboration between industry and researchers
Technology transfer aspect of the project proved to be beneficial. A number of researchers were trained in technology transfer and management essentials which will further improve the contact and real cooperation with industry. During the project, new partnerships with industry (e.g. Siemens, DHL, IBM) were established not only in the life sciences arena but also in the closely related field of information and communication technologies. One new spin-off company was established during the project.
- Innovation image of South Moravian Region, attraction of investors
South Moravian Region is regarded as the most innovative and research oriented region of the Czech Republic. This is very much supported also by regional government. Having excellent research teams (such as NCBR) based in the region help to further develop the vision of the region to be innovative. This image also helps to attract further investors into the region and especially investors with a strong focus on research and development. Over the last years the following companies established or expanded their R&D centres in the region: Honeywell (engineering), Tescan, FEI (scientific instrumentation - electron microscopes), RedHat (ICT).
Project website: http://ncbr.chemi.muni.cz/POSTBIOMIN.html
National Centre for Biomolecular Research (NCBR), Masaryk University Brno in the Czech Republic has become one of the leading laboratories for structural biology in the Central Europe. The institute is very well equipped by modern laboratory equipment and instruments. Project POSTBIOMIN strengthened our research potential in theoretical and experimental research that focuses on the structural characterisation of biomacromolecules and especially biomolecular interactions study through the scheme of support measures like recruitment of foreign experts, organisation of international PhD courses and lectures given by top international scientists. Series of workshops, seminars, postdoctoral and expert inter-ships and visits will be organised. Our technology transfer capabilities were strengthened during the project as well. POSTBIOMIN complemented investments in research capacity co-financed through the European Union (EU) cohesion policy programmes for 2007-2013, i.e. CEITEC, a large project of Central European Institute of Technology focused on life sciences and advanced materials and technologies. Project POSTBIOMIN contributed to the creation of human potential for modern research institutions and companies dealing with bio-medical research and biotechnologies within the region of Brno.
The global objective of this project is to further develop research potential of our institute (NCBR), which is important part of regional network of life sciences research institutions. This project will help us to gain knowledge and exchange experience with other European regions. We will work on setting up strategic partnerships with other research teams elsewhere in Europe as well. One of the greatest challenges to be achieved is the return of Czech scientist who would come back to pursue her/his independent career in our institute.
Apart from global objective we can identify technical and scientific objectives in POSTBIOMIN project. Due to the character of REGPOT-1 call, technical objectives are crucial and the most substantial for the project. However achievement of scientific goals and objectives is even more important, despite scientific objectives are rather hidden behind main content of work packages.
Technical objectives, i.e. organisation of post-doc and PhD courses, appointment of foreign experts and scientists, organisation of lectures, workshops, trainings both in Masaryk University in Brno and abroad (EU-15 countries) are described in details below within the description of particular work packages. Technical objectives of the project are translated into work packages (WPs) as follows:
- WP1 - Return of experienced Czech scientist
- WP2 - Incoming post-docs
- WP3 - PhD course
- WP4 - Lectures
- WP5 - Incoming experts
- WP6 Workshops
- WP7 - Training at foreign institutes
- WP8 - Technology transfer
- WP9 - Project management.
Scientific objectives will be targeted by all WPs. Structural biology is relatively young research area aiming at understanding the three-dimensional (3D) structures of biological macromolecules-proteins, nucleic acids and sugars. In recent years, the techniques of structural biology have significantly advanced and generated many structures of biomolecules. However, the visualisation of isolated molecules is not the only goal. Biological molecules work in the context of cells and tissues, interacting with other proteins, nucleic acids and carbohydrates. Detailed knowledge of mutual interactions provides fundamental insights into molecular biology and feeds into the development of medical applications. We aim to enhance our research capacities to study biomolecular interactions and their structures with the ultimate goal to find the molecular basis behind a biological event. This requires interdisciplinary knowledge in physics, chemistry and biology and to be able to develop and apply the state-of-the-art techniques in this area. This program will enable us to obtain this interdisciplinary research experience through the scheme of support measures such as recruitment of foreign experts and post-docs, organisation of international postdoctoral courses and lectures given by internationally recognised scientists. In general, we can describe scientific objectives as three research topics (see below).
Project results:
The project was concentrated on the following three research topics:
Research topic 1 (RT1) - Repair of damaged DNA
Responsible scientists: Dr Krejci, Dr Stefl, Prof. Sklenar, Prof. Koca
Research topic 2 (RT2) - RNA quality control
Responsible scientists: Dr Stefl, Prof. Koca, Prof. Sklenar
Research topic 3 (RT3) - Pathogen-host recognition and interaction
Responsible scientists: Dr Wimmerova, Prof. Koca, Prof. Sklenar
Achievement in the respective research topics:
Research topic 1 (RT1) - Repair of damaged DNA
First project period
The integrity of a cell genome is constantly challenged by DNA lesions such as base modifications, nicks and double-strand breaks. A single unrepaired DNA double-strand break (DSB) is lethal and if repaired improperly may lead to loss of heterozygosity, mutations, deletions, genomic rearrangements and chromosome loss. DSBs are caused by a vast number of endogenous and exogenous agents including genotoxic chemicals or ionising radiation, as well as through replication of a damaged template DNA or the replication fork collapse. In most organisms, DNA DSBs can be repaired in an error- prone and error-free manner by non-homologoues end-joining (NHEJ) and homologous recombination (HR), respectively. This proposal focuses on HR that plays a vital role in DNA metabolic processes including meiosis, DNA repair, DNA replication, maintenance of rDNA homeostasis, and genome plasticity. The importance of HR in maintaining genome integrity is underscored by its high degree of evolutionary conservation from yeast to human. Several models were proposed to understand the mechanism of double-strand break repair by HR, identify separate steps and role of the individual proteins involved.
The aim of this topic is to determine the molecular details of homologous recombination and repair of DSB by characterising the functional and physical interactions various protein factors. Since defects in HR are often related to human genomic instability, the results from these studies will provide insight into not only the mechanism of DNA repair, but could potentially be used for the diagnosis, prevention, and treatment of human diseases including cancer.
Scientific progress
During the first six months we have identified new DNA binding domain within Rad52 protein and showed that this domain is responsible also for loading Rad51 protein on ssDNA and mediating strand exchange. We have also demonstrated the ability of central region of Rad52 protein to bind RPA, ssDNA binding protein. Interestingly this interaction is mediated only when RPA is ssDNA bound status. The results of this work were published.
Furthermore, we were able to identify and characterise the effect of Rad54 protein on the nuclease activity of structure-specific nuclease Mus81/Mms4. We showed that this activation is evolutionary conserved does not require ATP binding nor hydrolysis. In addition, Rad54 is capable of targeting Mus81/Mms4 complex to DNA substrates, which has been confirmed by genetic analysis of Mus81 foci formation in the absence of RAD54 gene. As a part of collaboration, we have studied the effect of Mph1 and Srs2 helicase on the ability to disrupt D-loop intermediated during the process of homologous recombination. The results of this work were also published.
We have mapped the interaction domain within Srs2 and Rad51 proteins and tested isolated mutants in biochemical as well as genetic experiments to define the biological significance of this interaction. The manuscript is now in the preparation. In addition, we have also mapped the sites of SUMO-modifications on Srs2 and Rad52 proteins and are characterising the effect of Sumoylation on Rad52 in great details.
In next year, we plan to further characterise the effect of SUMOylation on the biochemical properties of Rad52 protein. Within the first six months we plan to characterise the wild-type and mutants defective in their sumoylation for Rad52-mediated biochemical activities as well their eight genetic properties. In the rest of the year we will test the effect of Sumoylated for of rad52 on its activities. We will also focus on localisation of Srs2 protein during replication and DNA repair and its ability to serve as quality control mechanism together with Rad52 protein. Next we plan to characterise the molecular events leading to disassembly of Rad51 filament via the action of Srs2 helicase.
Second project period
Short overview:
DNA double-strand breaks (DSBs) are one of the most toxic kinds of damage, implicated in cancer and many other diseases. Just a single un-repaired break can lead to aneuploidy, genetic aberrations or cell death. DSBs are caused by a vast number of both endogenous and exogenous agents including genotoxic chemicals or ionising radiation, as well as through replication of a damaged DNA templates or the replication fork collapse. To maintain the integrity of the genome it is essential to recognise and process DSBs as well as other toxic intermediates and launch most appropriate repair mechanism. How the cells achieve specificity, efficiency and cooperativity in the complex protein-protein interaction networks remains unclear. Therefore, we have designed specific aims that should uncover parts of this complexity. The link between defects in the processes that maintain genomic integrity and pathological disorders could shed light onto the knowledge of the molecular mechanisms behind it and potentially be targeted for therapeutic intervention.
During the past year, we have been involved in biochemical characterisation of the role of post-translational modification by SUMO on the biochemical properties of protein involved in homologous recombination. First we looked at effect of sumoylation on biochemical properties of the Saccharomyces cerevisiae mediator protein Rad52. Interestingly, binding of ssDNA enhances sumoylation of Rad52, and we demonstrated that modification of Rad52 inhibits its biochemical activities. It not only inhibited the ability of Rad52 to bind DNA but also catalyze annealing of complementary strands. On the other hand the Rad52 sumoylation did not have any effect on its protein-mediated interactions (oligomerisation, interaction with Rad51 and RPA). In addition, we also mapped the modification sites and generated sumo-deficient mutants that were characterised both in vitro and in vivo.
Furthermore, we have been also characterising the possible role of Srs2 in regulation of homologous recombination. Despite the role of HR contributing to the elimination of DSBs, it must be tightly regulated to prevent untimely events that could interfere with other DNA repair or replication. We have been collaborating with Dr Ellenberger on characterisation of Srs2 anti-recombinase activity. Biochemical characterisation revealed that led that a physical interaction between the C-terminal region of Srs2 and Rad51 protein and triggers ATP hydrolysis within the Rad51 filament, resulting in Rad51 dissociation from DNA. Durand last year we have also focused on characterisation of possible quality control mechanism that use both positive (Rad52) as well as negative (Srs2) regulators to shift the equilibrium of homologous recombination. Based on cell biology data we have suggested that Srs2 antagonises Rad52 in the formation of Rad51 filaments. This was further supported by the ability of Rad52 protein to suppress the inhibitory effect of Srs2 on Rad51-mediated strand exchange. Furthermore, the Rad51 mutants, originally isolated as Rad52-interaction deficient mutants, appeared also defective in the interaction with Srs2, suggesting that these proteins compete for the same interaction region. As expected, these Rad51 mutants show resistance to the action of Srs2 as well as inability to overcome the RPA inhibition on Rad51-mediated strand exchange. Accordingly, an Srs2 mutant protein, that fails to interact with Rad51, is not sufficient to disrupt the Rad51 presinaptic filament in vitro as well as in vivo. To further elucidate the mechanism that can negatively regulate recombination we have been involved in the identification of another regulatory mechanism requiring Mph1 helicase. In this pathway, Mph1 uses its translocase activities to unwind intermediate of homologous recombination, D-loop.
Next year, we plan to characterise the role of post-translational modification of Srs2 by SUMO peptide. Similarly to Rad52 we plan both biochemical as well as genetic characterisation of sumo-deficient mutants. We also plan to determine the role of Srs2 in replication and DNA repair synthesis. For this purpose we are developing D-loop assay linked with the extension of incorporated primer. We will test know replication factor that might be also involved during repair synthesis. These results should shed light in to the downstream steps of homologous recombination.
Third project period
DNA double-strand breaks (DSBs) are one of the most highly toxic kinds of damage, implicated in cancer and many other diseases. DSBs are caused by a vast number of both endogenous and exogenous agents including genotoxic chemicals or ionising radiation, as well as through replication of a damaged template DNA or the replication fork collapse. It is essential for cell survival to recognise and process DSBs as well as other toxic intermediates and launch most appropriate repair mechanism. How the cells achieve specificity, efficiency and cooperativity in the complex protein-protein interaction networks remain unclear. Therefore we have designed three specific aims that should uncover parts of this complexity:
1) resection of DNA breaks;
2) processing of other intermediates during various DNA metabolic processes as well as DNA and protein crosslinks;
3) high throughput screen identifying potential inhibitors of nucleases involved in these processes, and investigation of their effect on genome stability.
To achieve our goals, we plan to utilise challenging multi- as well as inter-disciplinary approaches combining biochemical, genetic, biophysical and cell biological characterisation combined with low- and high- resolution structural approaches and small molecule analysis. Our studies should contribute towards understanding how the cell responds to DNA damage and maintain its genomic integrity as well as for cancer therapeutics and gene targeting, as they could open the new horizons and opportunities for the predetermined control of homologous recombination and other types of repair.
During past six months, we have tested the effect of Rad52 sumoylation on its in vivo activities. To address the effects of Rad52 SUMOylation in vivo, we examined the SUMO-deficient mutants (rad52-K43,44R, -K253R or -K43,44,253R) using several recombination assays. We found that these mutations did not affect the rate of intra- or inter-chromosomal recombination during mitosis and meiosis. However, they lead to a shift from single-stranded annealing to gene conversion events. In addition, these mutations resulted in slight hyper-recombination phenotype in rDNA recombination. Taken together, our data suggested that SUMOylation of Rad52 facilitates certain recombination sub-pathway and disfavours the others. Finally, we also tested the in vivo localisation of Rad52 SUMO-deficient mutant by cytological analysis. The fusion of wild type as well as the rad52-K43,44,253R proteins to YFP revealed that the mutant protein forms foci similarly to wild type, but the duration of the foci is significantly shorter. In addition, we have also showed that also Rad59 protein is sumoylated both in vitro as well as in vivo after the DNA damage. We have mapped the residue that are modified and confirmed the mutation within this residues abolish sumoylation in vivo. Biochemical characterisation did not show any effect on its activities, including DNA binding and annealing of complementary strands. Interestingly, we have observed that Rad59 dramatically inhibited sumoylation of Rad52 in vitro and this inhibition was specific. This was oppose effect of Rad59 on Rad52 activity was later on confirmed using Rad59-sumodificent mutant that was able to suppress most of the rad52-K43,44,253R phenotype. The genetic experiments were done in close collaboration with Dr Michael Lisby with him we started to collaborate in genetic as well as cell biological characterisation of sumo-deficient mutant of recombination proteins. The results of this study are now in preparation for publication.
We were also able to successfully reconstitute a DNA repair synthesis step of homologous recombination. This assay is based on extension of D-loop structure formed by coordinated action Rad51 and Rad54 proteins. The actual extension is dependent on PCNA, its loader (RFC) and polymerase delta. RPA protein was found to stimulate the extension of longer products suggesting that it prevents formation of secondary structures on ssDNA that would otherwise block the DNA extension. The length of extension product depends on salt concentration and reaches up to 1 kb, which is in good agreement with genetic studies. Furthermore, pl epsilon was also capable of extension of D-loop substrate, but resulting on shorter extension products. In addition, we have also tested the effect of Srs2 and Mph1 helicases to unwind newly extended DNA. While Srs2 failed to do so, Mph1 was fully capable of this unwinding, suggesting its biological role in vivo. These results shed light into the downstream steps of homologous recombination and were published in journal DNA Repair. Next, we have used this assay to test the effect of SUMOylated PCNA that was show to play major role in regulation of recombination during the repair of stalled replication forks. Our preliminary data show the ability of Srs2 to block the extension of D-loop substrate. Interestingly, this activity does not require the helicase / translocase activity of Srs2, but only the SUMO interaction motif on the C-terminus of Srs2. The detail analysis of the mechanism revealed the ability of Srs2 to disrupt the direct interaction between Pol delta and SUMO-PCNA, which is required for primer extension. Manuscript with this work was recently submitted.
Finally, we have performed high throughput screening of chemical library of 30 000 compounds available to our collaborator Dr Bartunek (Institute of Molecular Genetics, Prague). For the purpose of the screen we have developed a novel liquid fluorescent-based assay with high detection limits that allows us to use nM concentrations of proteins and substrates. Among multiple nucleases available in our laboratory we have screened inhibitors against Mus81/Eme1 complex. We have identified several lead compounds that will be further characterised. Several other human nucleases from the same protein families as well as yeast homologues are now being tested to determine the specificity and selectivity of particular inhibitors.
Research topic 2 (RT2) - RNA quality control
First project period
Expression, labeling and purification of the entire TRAMP complex and individual recombinant subunits
In order to investigate the specific affinity of the TRAMP complex for the RNA substrates, additional DNA constructs were prepared for the expression of recombinant wild type and mutant and deletion forms of the two subunits of the TRAMP4 complex, the Trf4 and Air2 proteins. These proteins are currently being tested for RNA binding by gel shift assay in order to identify the protein signature for RNA binding. Subsequently, several tests were performed to obtain the reconstituted homogeneous population of the minimal PAP TRAMP4 complex. Currently, experiments with additional deletion versions and are in progress.
TAP tag purification of the TRAMP complex from yeast and MS-MS analysis of yeast TRAMP complex
We have optimised the conditions for complex affinity purification from yeast in order to analyze the subunits for potential posttranslational modifications. Since the amounts of the affinity purified complexes were low for the MS analysis of posttranslational modifications, we have been testing different approaches to scale up the purification procedures.
Mapping of interaction between RNA-quality-control-related proteins and RNA using NMR spectroscopy
Structure determination of protein-RNA complexes is challenging and involves many biochemical and biophysical issues. Only recently, the structure determination of protein-RNA complexes became possible by NMR spectroscopy. Often, truncated forms of the studied protein-RNA complexes have to be generated in order to obtain a well-behaving system for NMR structure determination. We started to map interactions between the proteins that are involved in RNA quality control. We always work with several protein and RNA constructs for each system. The designed constructs do not rely only on the computational alignment strategies but also on limited proteolysis and functional assays. This careful strategy excludes the possibility of truncating important elements out of the studied constructs. Using this approach, we identified several systems that are suitable for the structure determination by NMR. It is the minimal Nrd1-Nab3 heterodimer that has same RNA-binding properties as the full-length Nrd1-Nab3 heterodimer; the RNA-binding domains of both Nrd1 and Nab3. These constructs are currently used not only for the mapping of interaction with RNA by also for the structure determination by NMR.
Second project period
Protein preparations and purifications
Our laboratories continued in preparation of proteins that are responsible for the studied mechanism of RNA quality control. We have produced several constructs of Trf4 that are used for functional assays and crystallisation trials. For NMR structural studies, we produced PIN domain of Rrp44 (catalytic subunit of the exosome), Nab3 and Nrd1 proteins. We experienced problems with expression and solubility of some protein constructs. For example, none of the expression conditions, bacterial host strains, or solubility enhancement tags, provided soluble and folded Air2 protein. This problem was solved using SUMO-tag that helped as a folding chaperone during the refolding procedure of Air2 protein. We also prepared Nrd1-Nab3 heterodimer complex using bacterial co-expression system.
Preparation of TRAMP mutant strains for in vivo studies
We have prepared a series of deletion mutants of the catalytical subunit Trf4 to screen for the role of each of the protein domains in vivo. We have confirmed the constructs by sequencing. In parallel we have prepared a series of mutants of the RNA-binding subunit Air2 by site-directed mutagenesis.
Structural determination of Nab3 and mapping of its interaction with RNA using NMR spectroscopy
We have determined the three dimensional structure of RNA binding protein Nab3 that is involved in processing of functional RNAs such as sn/snoRNAs. The three-dimensional structure of Nab3 RRM adopts a compact fold with an 1-1-2-3-2-4 topology. The fold is composed of the two helices that are packed along a face of a four-stranded antiparallel sheet (4-1-3-2). We have identified the RNA interaction interface of Nab3 and the residues that are important for the interaction with RNA
Development of NMR methods and their application to biological problems
Development of NMR methods for studies of proteins and their complexes continued in two directions. First, an attention has been paid to the accuracy of determination of small residual dipolar couplings (RDCs), serving as a source of structural and motional information. An earlier developed method of experimental error correction has been implemented into a software S3E.py which is currently available to public and was published in a peer-reviewed journal. Second, application of high-resolution NMR methods to unstructured proteins has been successfully tested. Sufficient resolution for a complete assignment of highly degenerated frequencies in a partially disordered RNA polymerase delta subunit from Bacillus subtilis was achieved using 5D non-uniformly sampled NMR spectra.
The TRAMP complex analysis
We have started a detailed analysis of the TRAMP mutant strains. The phenotypic tests identified 5 Trf4p point and deletion mutants, respectively with strongly impaired growth when expressed in the double deletion trf4trf5 background. The same mutants expressed in trf4 delta showed defects in pre-rRNA processing as tested by Northern blot analysis by accumulating 23S precursor of the 18S rRNA. The wild type and mutant forms of Trf4p were subsequently affinity purified via the N-terminal fusion protein A tag by using IgG beads. Purified proteins were assays for polyadenylation activities. The mutants that showed growth and rRNA processing defects also displayed no polyadenylation activities. In parallel, we have expressed and purified the same recombinant deletion mutants of Trf4p from bacteria and their polyadenylation activity was tested upon the addition of recombinant Air2 protein. Similar studies were performed with the Air mutant strains and mutant recombinant proteins. Wild type and mutant forms of Air2p were episomally expressed in the double deletion strain air1air2. Surprisingly only one of the mutants showed only minor growth defects. The same strain revealed defects in rRNA processing. However, in vitro complementation assays of polyadenylation activity revealed two mutants that lost the ability to activate Trf4p in vitro. Currently, we are performing additional experiments to confirm these results.
Recognition of phosphorylated RNA polymerase II C-terminal domain by protein factors involved in the Nrd1 termination pathway
Transcription is coupled to RNA processing and RNA quality control machineries by the carboxyl-terminal domain (CTD) of RNA polymerase II, which consists of up to 52 repeats of the sequence Tyr1-Ser2-Pro3-Thr4-Ser5-Pro6-Ser7 in mammals. The CTD heptapeptide repeat is dynamically phosphorylated and dephosphorylated on serines in the course of transcription cycle and promotes the recruitment of various enzymes and RNA processing factors to the transcription site. In addition, cis-to-trans isomerisation of Pro3 and Pro6 promoted by prolyl isomerases serves a conformational switch, which further enhances the CTD code that controls binding of transcription and processing factors involved in RNA biogenesis.
We have started a detailed analysis of the CTD recognition by the Nrd1 processing factor that is modulated by phosphorylation and proline isomerisation. We prepared and labeled the Nrd1 protein. Using nuclear magnetic resonance spectroscopy, we determined the three-dimensional structure of Nrd1 in the free form. Currently, we are working on the structure of Nrd1 bound to the CTD peptide phosphorylated at serine 5. We also investigated how the Nrd1 recognition process is affected by another factors, a prolyl isomerase called Ess1, and a phosphatase, Ssu72. We found out that Ess1 contributes to the disassembly of the Nrd1-CTD complex and that Ess1 stimulates dephosphorylation mediated by Ssu72. Future studies will enable us to fully understand the structural basis behind the 'CTD code' and how it is used for the recruitment and dissociation of Nrd1 complex in the mechanism of poly(A) independent termination.
Third project period
Development of NMR methods and their application to biological problems
Development of NMR methods was focused on techniques for characterisation of unstructured proteins. The recently developed methodology of 5D non-uniform sampled NMR spectroscopy and the resulting complete assignment of highly degenerated frequencies of Bacillus subtilis RNA polymerase delta subunit has been published in a peer-reviewed journal. The methodology was further improved by introducing direct C-13 detection to enhance resolution in the direct dimension. The developed protocol allowed us to obtain a complete sequential assignment of the above mentioned testing sample from a single 5D spectrum. Two 5D NMR experiments based on the introduced technique were designed and a manuscript describing the approach and its applications has been submitted.
The previously published methodology of 5D non-uniform sampling was further developed. Direct C-13 detection and recycle delay optimisation resulted in a set of 5D experiments allowing an easy assignment of unstructured proteins with very low signal dispersion. The method has been published a peer-reviewed scientific journal and tested on difficult samples of unfolded proteins, including a protein with a high degree of sequence repetition and a proline-rich protein. The development continued by designing versions of the mentioned experiments with improved sensitivity.
Recognition of termination signals in poly(A)-independent RNA processing
We significantly progressed in our long-term goal to reveal the structural and mechanistic bases for the termination and processing of functional RNA molecules. In our work we focused on the structural determination of the Nrd1 complex. The Nrd1 complex includes two RNA-binding proteins, the nuclear polyadenylated RNA-binding (Nab)3 and the nuclear pre-mRNA down-regulation (Nrd)1 that bind their specific termination elements. We determined the solution structure of the RNA-recognition motif (RRM) of Nab3 in complex with UCUU RNA representing the Nab3 termination element. The structure shows that the first three nucleotides of UCUU are accommodated on the beta-sheet surface of Nab3 RRM, but reveals a sequence specific recognition only for the central cytidine and uridine. The specific contacts we identified are important for binding affinity as well as for proper transcription termination in vivo. In our mechanistic studies, we found that both RNA-binding motifs of Nab3 and Nrd1 alone bind their termination elements with a weak affinity. Interestingly, when Nab3 and Nrd1 form heterodimer, the affinity is significantly increased, most likely in a cooperative manner. These finding are in accordance with the model of their function in the poly(A) independent termination, in which the binding to the combined and/or repetitive termination elements elicits efficient termination. A manuscript describing the structure of the complex has been submitted for publication.
Recognition of C-terminal domain of RNA Polymerase II by Nrd1
RNA processing factors bind and process only the nascent RNA transcript but also interact and travel with RNA polymerase II during transcription. This phenomenon is called co-transcriptional processing in which a number of processing machineries associate with RNA polymerase II, creating a large processing assembly. Nrd1-dependent termination apparatus that is under investigation in this project also associates with RNA Polymerase II. To reveal this interaction, we studied how Nrd1 binds RNA Polymerase II. Recruitment of Nrd1 the site of transcription is controlled by phosphorylation of the C-terminal domain (CTD) of RNA Polymerase II. We determined the solution structure of Ser5 phosphorylated (pSer5) CTD bound to the CTD-interacting domain (CID) of Nrd1. The structure reveals a direct recognition of pSer5 by Nrd1 CID upstream of the CTD region usually recognised by other CID-containing proteins. The specific phosphoserine recognition requires the cis conformation of the upstream pSer5-Pro6 prolyl bond of the CTD peptide. Mechanistic studies by NMR show that Ess1 prolyl isomerase disassembles the Nrd1 CID-pSer5 CTD complex by isomerisation of the phosphoserine-proline prolyl bond. Mutations at the complex interface diminish binding affinity and impair processing of small nucleolar RNA, suggesting that the Nrd1-pSer5 CTD interactions are important for processing of non-coding RNAs. Our findings underpin the importance of both covalent and non-covalent changes in the CTD structure that constitute the CTD code. Furthermore, we suggest that Ess1 regulates directly association of Nrd1 with RNAP II, in addition to its role in the indirect regulation via stimulation of dephosphorylation by Ssu72. A manuscript describing the structure of the complex and the functional data is in preparation.
The TRAMP complex analysis
We have followed up the studies of the strains mutant in individual subunits. The point mutants of AIR2 that previously showed lack of Trf4p activation and growth defects in vivo were expressed in double deletion strain of air1 and air2 and affinity purified on IgG agarose. Because they failed to strongly copurify the other tw components of the TRAMP complex, we conclude that the regions mutated are responsible for intermolecular contacts within TRAMP4. Thus we extended the spectrum of Air2p mutants. We prepared deletion forms lacking region at the N- and/or C- termini, respectively. The deletion versions of Air2p were again expressed in the double deletion strain and assayed for growth. We have observed new motifs at the C- and N- termini that are likely responsible for Trf4p and Mtr4p interactions. In agreement with the growth defects, additional Northern blot analyses showed accumulation of cryptic unstable transcripts in mutants with major growth impairment.
To complete or analysis of the TRAMP4 complex, we prepared a set of deletion mutants in the third subunit Mtr4p. These mutants were introduced into the strain where endogenous MTR4 was deleted and the growth was rescued by wild type Mtr4p expressed from a URA3 plasmid. Growth test analysis on 5-FOA plates at different temperatures confirmed the lethality of most of these mutants, as most of them contained deletions in conserved helicase domains. Next, we will perform pull-down experiments to find out which parts of Mtr4p are crucial for interactions with Trf4p and Air2p in vivo.
Functional importance of association between Nrd1 and RNA polymerase II
We further studied the functional relevance of interaction between RNA Polymerase II and Nrd1-depedent termination apparatus that is involved in RNA processing. Specific association of pSer5 CTD with Nrd1 CID was further tested in a quantitative in vitro solution binding assay using FA experiments. We titrated the wild-type and mutants of Nrd1 CID against fluorescently labeled pSer5 CTD. Alanine or aspartate (charge swapping) substitutions at positions Ser25, Arg28, Ile29, and Lys30 significantly decreased the binding affinity with pSer5 CTD. These FA data strongly support the phosphorylation-specific recognition observed in the structure of Nrd1-pSer5 CTD complex and that the interaction relies upon the presence of the invariant basic residues in the CID domain. To determine the importance of these individual residues for Nrd1 function in vivo, we monitored cell viability and snoRNA expression in single amino acid mutants. The Nrd1 strain lacking CID (Nrd1-1-150) was used as a reference for the CID-related function. Deletion of CID is not lethal, however, we observed that deletion of CID did not support viability at 37 °C. Similar, to a lesser extent temperature sensitive growth defect was observed in mutants of Ser25, Arg28, Ile29 and Arg74 mutants. As both Ser25 and Arg28 are contacting the phosphorylated serine, we prepared Ser25+Arg28 double mutant that exhibited the same growth phenotype as CID Nrd1. Although CID Nrd1 is viable, the D70R mutant caused lethality upon the depletion of endogenous wt Nrd1. The lethality of D70R is not due to the absence of the mutant protein, but it destabilises the CID fold that likely interferes with the Nrd1 functions (D70R displays read-through for snR33). Nrd1 CID deletion causes an accumulation of snR33 precursors. We have observed that point mutants with ts growth phenotype showed an accumulation of pre-snR33. Importantly, the Ser25+Arg28 double mutant showed virtually identical processing defects of snR33 as for CID Nrd1. These data demonstrate that specific recognition of pSer5 CTD by Nrd1 CID is important for snoRNA processing in vivo.
Research topic 3 (RT3) - Pathogen-host recognition and interaction
First project period
Short overview
Saccharides are known to have a variety of more sophisticated and complex functions in cell-cell interaction and communication. The glycorecognition by carbohydrate-binding proteins, lectins, is important in the mediation of numerous physiological processes including fertilisation, pathogencell adhesion and recognition, inflammatory response and others. A number of pathogen microorganisms utilise lectin-carbohydrate interaction to recognise and infect host organisms.
Lectins are proteins of non immune and non catalytic origin which specifically bind carbohydrates. Understanding of host / pathogen interaction is the first step in designing inhibitor molecules as potent therapeutics. The key goal of the project is to understand pathogen-host recognition based on protein-carbohydrate recognition and to find a way how to inhibit both the pathogen adhesion and the formation of biofilms, which are highly resistant to classical antibiotic therapy.
Main results obtained during the first year of the project
We have focused on the cloning homologues of PA-IIL lectin, one of the virulence factors of Pseudomonas aeruginosa, which we identified in the genome of Burkholderia cenocepacia. Four lectins named BclA, BclB, BclC and BclD were cloned and prepared in the recombinant form. Since oligosaccharide specificity could be correlated to the possible function of the bacterial lectin in tissue recognition, binding to a broad spectrum glycan array was performed at the Consortium for Functional Glycomics. The screening results showed BclA to have a strict specificity for oligomannose type N-glycans, only binding to oligosaccharides with a mannoside capped branch. The glycan array results obtained on the other lectins were non-clear. All three lectins showed a mixed specificity toward human fucosylated as well as mannosylated epitopes. From the prediction methods developed in our laboratory the BclB protein should display fucose-binding preference whereas BclC and BclD mannose-binding preference, respectively. All three mentioned lectins contain in its amino acid sequence an additional N-terminal domain of unknown function that can be responsible for such behaviour.
Therefore, we separated these proteins on individual domains to decipher the function of the Nterminal part to clarify specificity of the lectin parts of the proteins. Important results were obtained with BclC protein. Glycan array of individual domains showed that C-terminal part of the protein BclC, that we expected to be a mannose-recognising lectin part, certainly recognises only human mannosylated epitops. To our surprise, N-terminal domain of unknown function can also bind sugar epitopes and it is specific towards fucosylated oligosaccharides. Thus, we identified a novel fucosebinding lectin that doesn't have any homologue in current sequence databases.
In collaboration with the group of Dr Anne Imberty, we characterised the specificity and affinity of the galactophilic lectin PA-IL for Gal1-4-Gal and Gal1-3-Gal epitopes by cell surface labeling combined with glycan array analysis and titration microcalorimetry. The crystal structure of PA-IL complexed with Gal1-3-Gal1-4Glc trisaccharide establishes the atomic basis of the specificity and reveals how the second galactose residue makes specific contacts with the protein surface (Blanchard et al., 2008). Following study was designed to evaluate the respektive contribution of PA-IL and PA-IIL lectins in the pathogenesis of P. aeruginosa-mediated acute lung injury. Using an in vitro model with A549 cells, and an experimental murine model of acute lung injury, comparison of the parental strain to mutants' strains with inactivated PA-IL and PA-IIL gene were performed.
In addition to mutagenesis in vitro that we have been performing on PA-IIL protein, we have been working on development of reliable mutagenesis of protein in silico to speed up the process of protein engineering of lectins. The graphical interface TRITON is under development in our laboratory. Program TRITON is a user oriented software with graphical interface that enables visualisation of molecular structures, preparation of input files for computational software and analysis of output data. It interfaces, for example, external software Modeler. Unavoidable part of the in silico protein engineering is a quantitative prediction of binding properties newly constructed lectins. Programme AutoDock was implemented into TRITON interface and the system is being parameterised for the work with saccharides. The results were published. By combination of experimental and computational methods, the effect of terminal GLY114* deletion on the binding affinity of the PA-IIL lectin towards sugars was investigated. Both have shown that the deletion mutation decreases the affinity of the mutant towards saccharides and confirmed that PA-IIL saccharide binding affinity is influenced by the dimerisation of the lectin. Corresponding manuscript has been accepted for publication.
Second project period
Short overview
Saccharides are known to have a variety of more sophisticated and complex functions in cell-cell interaction and communication. The glycorecognition by carbohydrate-binding proteins, lectins, is important in the mediation of numerous physiological processes including fertilisation, pathogen-cell adhesion and recognition, inflammatory response and others. A number of pathogen microorganisms utilise lectin-carbohydrate interaction to recognise and infect host organisms. Lectins are proteins of non immune and non catalytic origin which specifically bind carbohydrates.
Understanding of host / pathogen interaction is the first step in designing inhibitor molecules as potent therapeutics. The key goal of the project is to understand pathogen-host recognition based on protein-carbohydrate recognition and to find a way how to inhibit both the pathogen adhesion and the formation of biofilms, which are highly resistant to classical antibiotic therapy.
Key results obtained in the second year
We have focused on the cloning homologues of PA-IIL lectin, one of the virulence factors of Pseudomonas aeruginosa, which we identified in the genome of Burkholderia cenocepacia. Four lectins named BC2L-A, BC2L-B, BC2L-C and BC2L-D were cloned and prepared in the recombinant form. All three later mentioned lectins contain in its amino acid sequence an additional N-terminal domain of unknown function that can be responsible for such behaviour. As already reported, we separated these proteins on individual domains to decipher the function of the N-terminal part.
The N-terminal part of the BclC lectin was completely characterised. The recombinant Nterminal BC2L-C domain is a new lectin with specificity for fucosylated human histo-blood group epitopes H-type 1, Lewis B and Lewis Y, as determined by glycan array and isothermal titration calorimetry. Methyl selenofucoside was used as ligand to solve the crystal structure of the N-terminal BC2L-C domain. Additional molecular modelling studies rationalised the preference for Lewis epitopes. The structure reveals a trimetric jelly-roll arrangement with striking similarity to TNF-like proteins (Sulak et al., 2010).
In collaboration with Prof. Miguel Volvano and Dr Anne Imberty, the presence of BC2L-A, BC2L-B, BC2L-C on the outer membrane of the wild-type bacteria B. cenocepacia was confirmed, albeit with unknown secretion mechanism. This suggests that studied lectins of B. cenocepacia can play important role in host recognition or biofilm stabilisation.
Several native ligands have been studied to characterise binding properties of individual proteins and their domains. We continue also in mutagenesis of selected lectins to characterise key amino acids for affinity and/or specificity of binding. The mutagenesis is performed both in vitro (experimentally) and in silico (using in house software Triton being developed in our laboratory).
A site-directed saturating mutagenesis of amino acids in the specificity binding loop of PA IIL showed that any substitution of Ser at position 22 leads to fucose preference declining. Further characterisation of the interactions is being performed by molecular dynamics simulations.
Third project period
Short overview
Saccharides are known to have a variety of more sophisticated and complex functions in cell-cell interaction and communication. The glycorecognition by carbohydrate-binding proteins, lectins, is important in the mediation of numerous physiological processes including fertilisation, pathogen-cell adhesion and recognition, inflammatory response and others. A number of pathogen microorganisms utilise lectin-carbohydrate interaction to recognise and infect host organisms. Lectins are proteins of non-immune and non-catalytic origin which specifically bind carbohydrates.
Understanding of host / pathogen interaction is the first step in designing inhibitor molecules as potent therapeutics. The key goal of the project is to understand pathogen-host recognition based on protein-carbohydrate recognition and to find a way how to inhibit both the pathogen adhesion and the formation of biofilms, which are highly resistant to classical antibiotic therapy.
Main results obtained during the third year time
Our laboratories have continued in key lectins characterisation from Pseudomonas aeruginosa and Burkholderia cenocepacia pathogens. As already mention in the previous report, the very interesting features were connected with one of the B. Cenocepacia lectin BC2L-C.
The N-terminal part of the BclC lectin was structurally and functionally characterised as a novel fucose binding domain with a TNF-like fold (Sulak et al, 2010) while the C-terminal part is similar to a superfamily of calcium-dependent bacterial lectins. Recently, we have demonstrated that the C-terminal domain displays strict specificity for mannose- and heptose-containing oligosaccharides. Thus, BC2L-C is a 'superlectin' with mixed specificity for mannose / heptose glycocompounds and fucosylated human histo-blood group epitopes. The overall structure, determined by electron microscopy and small angle x-ray scattering, consists of a ring of three mannose / heptose-specific dimers flanked by two fucose-specific TNF-like trimers. We demonstrated that BC2L-C is immunogenic, binds to the bacterial surface and can be released into the medium. The unique architecture of this first superlectin correlates with multiple functions including bacterial cross-linking, adhesion to human epithelia, and stimulation of inflammation. Results were recently published in Plos Pathogen.
With new genome sequencing, several potential lectins were identified in the Aspergillus fumigatus genome. We focused on one of the predicted protein, and prepared it in a recombinant form. The recombinant protein binding activities were determined by hemagglutination studies, glycan array, surface plasmon resonance (SPR) and isothermal titration calorimetry (ITC). We confirmed its preference for fucose and fucosylated oligosaccharides. The 3D structure of AFL with fucosylated ligands was described using X-ray diffraction.
We continue also in mutagenesis of selected lectins to characterise key amino acids for affinity and/or specificity of binding. The mutagenesis is performed both in vitro (experimentally) and in silico (using in house software Triton being developed in our laboratory).
As stated in a previous report, a site-directed saturating mutagenesis of amino acids in the specificity binding loop of PA IIL showed that any substitution of Ser at position 22 leads to fucose preference declining. The mutations in the position 23 generally turned out to have similar effect. The mutations of the position 24 were therefore considered to be most promising. First and foremost, the side chains in this position have lesser potential to sterically hinder the binding and greater potential to actually create new interactions with the atoms of the ligand, basically being positioned not directly into the binding site. Except for significantly lower energies in cases of the detrimental mutations, there was not any mutation that would be clearly more favourable than the others. This is caused mainly by the fact that the side chain of the amino acid 24 usually sticked out of the binding site after Modeller perform the in silico mutagenesis. In order to explore the possibilities of the binding site actually accommodating for the sugar binding and forming new interactions, molecular dynamics simulations were performed at the results provided by the docking programs. The best docked structures of methyl-fucoside and methyl-mannoside were taken as the starting structures for molecular dynamics simulation. Besides the aspartate, glutamine and glutamate mutants, a tryptophan mutant was chosen as a representative of the aromatic bulky side chains. The trajectories were analysed both from energetic and geometric point of view, searching for correspondence between changes of energy and distances of chosen atoms. The atoms whose distances were analysed were chosen on the basis of visual analysis of the trajectory.
As already mentioned, we had proved that the specificity of the lectins in PA-IIL family is governed by the composition of a triad of amino acids at positions 22-23-24, and that the change of serine 22 (present in PA-IIL) to alanine (present in RS-IIL) caused the mutant to change the preference from fucose to mannose as a binding partner. To better understand the binding, we focused on other member of the PA-IIL family, CV-IIL, which binds fucose and mannose with similar affinities. Therefore, it appeared as an ideal molecule for testing the changes of specificity during mutagenesis of other amino acids.
Two approaches were selected, i.e. mutagenesis in silico and in vitro. We were mainly interested in performance of in silico methods once we employ more mutations in the same time. The performance of the Modeller software was therefore tested in relation to the different possibilities of creating the mutant proteins. Two main features were studied: the order of the mutations being performed on the protein, and the presence of the saccharide in binding site during in silico mutagenesis. As binding partners, the two 'border' saccharides, methyl-L-fucoside and methyl-D-mannoside were docked into the resulting mutants. The computational part has shown several important facts. In a surface binding site, where the positions of side chains important for binding are not spatially hindered thanks to close quarters, the way of performing the mutation was crucial for the overall result. The resulting structures can significantly differ, especially when mutation were done simultaneously and not sequentially. Also a presence of a saccharide in binding site has a beneficial effect on the results.
Potential impact:
South Moravia is a perspective region with a good infrastructure of human resources, functional institutional foul territory as well as entrepreneurial sphere. Main regional city Brno (400 000 inhabitants) is judged as having the best economic potential in the Central Europe. It is predicted that the city's rapid economic growth will continue (see http://www.fdimagazine.com for further details) due to a favourable structure of undergraduates, especially in biotechnological and technical fields. Major universities and several institutes of the Czech Academy of Science in Brno as well as industry oriented on life sciences and the local authorities were accordingly to their position involved in the project.
NCBR and POSTBIOMIN project were part of the regional activities in research, development and innovation. Being an excellent a very well networked research centre in the field of life sciences, NCBR actively contributed to the emergence of new research centres in South Moravia, which are financed by the EU Structural Funds. The research team of NCBR is the nucleus of structural biology research programme of an emerging European Centre of Excellence, CEITEC (see http://www.ceitec.cz for details).
CEITEC is a joint project of universities and research organisation from Brno, Czech Republic, coordinated by the Masaryk University. With a budget of more than EUR 200 million, it strives to establish a new European Centre of Excellence in the field of life sciences and advanced materials and technologies.
NCBR also actively participated in the following activities in the regional and national context during POSTBIOMIN project:
- participation on formulation and implementation of regional innovation strategy of South Moravian Region
- preparation of Czech position on future Eight Framework Programme (FP8) - active participation in working group on future of the European Research Center (ERC)
- advisory role in the SOMOPRO programme (research fellowships funded by FP7-COFUND, see http://www.jcmm.cz/en/somopro.html online) of South Moravian Centre for International Mobility
A. Impact of knowledge base
POSTBIOMIN project had a wide impact at different levels:
- Developing existing and emerging research potential
The huge importance of proper training for efficiency gains on the research site. Undergraduates not only Masaryk University bring high proficiency knowledge into the sector they work in, including further research. The benefits of training greatly outweigh costs by improving productivity and increasing equipment life trough proficient knowledge and modern research and scientific techniques.
During the project a new generation of post-docs emerged at NCBR. Most of them are staying at NCBR also after the end of POSTBIOMIN project. This new generation of researchers is therefore vital for the future of NCBR and its scientific excellence.
- Improvement of the capacity and increase visibility
POSTBIOMIN provided for a long term financing of some of the activities done by NCBR. This gave sufficient space for strategic decisions to be made and implement long term strategies. The fact that NCBR was able to attract such an interesting project (in terms of the budget size) gave a lot of additional visibility and also credibility when dealing with various Czech and international funding agencies. This can be proved also by the international evaluation of CEITEC project by an independent international panel on behalf of the Czech Ministry of Education. Structural biology research programme, where NCBR plays a crucial role was evaluated as the best research programme in CEITEC overall.
A number of FP7 proposals were submitted as a result of POSTBIOMIN project. They include also the following proposals:
1. EASTNMR, FP7-Infrastructures - selected for funding, NCBR as project partner, start - February 2009
2. BIONMR, FP7-Infrastructures - selected for funding, NCBR as project partner, start - autumn 2010
3. ERC Starting Grant - Lumír Krejí - submitted, not selected for funding
4. Richard Stefl - considered for ERC starting grant submission
5. Stepanka Vanacova - considered for ERC starting grant submission.
- New treatment
The findings linked to various human diseases including cancer in this project may well be of clinical value in the long term.
B. Contribution to community societal objectives
The project clearly contributed to several societal objectives of the EU:
- Training and preparation of young scientists and researchers
Training within the project took part in several ways. First of all postdoctoral training took place in WP2. Furthermore, given the fact that one of the WPs dealt with technology transfer a number of people from NCBR received training in technology transfer and management essentials.
- Cohesion and cooperation in the EU
The project was an excellent example of cooperation among EU Member States. During the project life time, a whole range and number of cooperation with institutes from all over the EU. This included mainly on the one hand visits of senior researchers from around the EU in Brno for lectures, PhD courses, workshops and activities in WP5 and on the other hand trips and trainings of NCBR staff to foreign institutes.
C. Economic impact
POSTBIOMIN had also direct scientific, technological and indirect economic impact in following terms:
- R&D
New and innovative methods and techniques were brought to the Czech Republic by the means of trainings at foreign institutes. This is for the benefit of the whole Czech life sciences research community.
- Increased collaboration between industry and researchers
Technology transfer aspect of the project proved to be beneficial. A number of researchers were trained in technology transfer and management essentials which will further improve the contact and real cooperation with industry. During the project, new partnerships with industry (e.g. Siemens, DHL, IBM) were established not only in the life sciences arena but also in the closely related field of information and communication technologies. One new spin-off company was established during the project.
- Innovation image of South Moravian Region, attraction of investors
South Moravian Region is regarded as the most innovative and research oriented region of the Czech Republic. This is very much supported also by regional government. Having excellent research teams (such as NCBR) based in the region help to further develop the vision of the region to be innovative. This image also helps to attract further investors into the region and especially investors with a strong focus on research and development. Over the last years the following companies established or expanded their R&D centres in the region: Honeywell (engineering), Tescan, FEI (scientific instrumentation - electron microscopes), RedHat (ICT).
Project website: http://ncbr.chemi.muni.cz/POSTBIOMIN.html