Functional analysis of stress-dependent RNA-enzyme interactions

Lead Research Organisation: University of Surrey
Department Name: Microbial & Cellular Sciences

Abstract

Cells have to immediately respond to changing environmental conditions. In particular, unicellular organisms such as baker's yeast Saccharomyces cerevisae must immediately react to cope with an altered environment. Since yeast is also important for production of food, wine and other goods, there is great interest to understand how the cells can deal with stress to optimise industrial applications. Specifically, the response to oxidative stress, which is imposed by the imbalance in the levels of so-called reactive oxygen species (ROS) generated during normal aerobic metabolism but particularly through exposure to certain toxic chemicals or irradiation, is of wider interest because it is connected to the development of diverse pathological processes in humans, such as neurodegenerative disorders, diabetes, arthritis and cancer.

Cells have developed a variety of mechanisms to adapt to stress, including immediate changes in the levels of key metabolites and by altering gene expression. Regarding the latter, it has become particularly recognised that stress response is considerably affecting the synthesis of new proteins, and the stability of RNAs. RNA represents an essential intermediate step in gene expression, where DNA is copied into RNA to serve as template for protein synthesis. Importantly, RNA is never naked in cells but covered by a host of proteins, so-called RNA-binding proteins (RBPs). These proteins can remove or rearrange parts of RNA, store, or deliver it to particular locations within the cell, and ultimately degrade it. They also control when and how messenger RNAs (mRNAs; refers to the class of RNAs that encode proteins) are translated into proteins.

Due to the tremendous importance of RPPs in gene expression control, we and other researchers have developed an experimental approach to catalogue all of the RBPs that interact with mRNAs in cells. Besides detecting many of the previously known RPBs, it was found that proteins with other well-established functions, such as enzymes, could interact with mRNAs (enzymes are proteins that perform essential chemical reactions in the cell). Furthermore, we have investigated which RBPs change mRNA associations upon oxidative stress in yeast. Interestingly, we found that many of the ones that changed mRNA associations were metabolic enzymes acting in carbon metabolisms, which is central for energy production and storage in cells; and provides building blocks for the synthesis of complex biological molecules.

Therefore, we wish to understand the function of these "enigmatic" enzyme-RNA interactions and whether they could play a role in coping with oxidative stress. We will first test whether the observed changes in RNA-enzyme interactions are specific to oxidative stress or apply to other stress conditions as well. We will then comprehensively identify the RNA targets for selected enzymes acting in central carbon metabolisms, and investigate where and how they bind to RNAs. Finally, we wish to explore whether the stress-dependent interactions with RNA affects their fate, or conversely, whether it modulates enzymatic activity and plays a role in the cell's adaptation to stress.

With our research, we expect to discover previously unrecognised links between RNA regulation, metabolism and cellular stress response. If so, this knowledge will likely have impact on diverse important aspects of our society, from food-production towards a better understanding of components that contribute to today's most prominent diseases including cancer.

Technical Summary

RNA-binding proteins (RBPs) play essential roles in the post-transcriptional control of gene expression. The recent introduction of proteome-wide approaches has dramatically expanded the repertoire of proteins interacting with RNA, revealing many "unconventional" RBPs with other well-established functions, such as metabolic enzymes. Nevertheless, while the repertoire of RBPs is steadily increasing, very little is known about the reconfiguration of the RNA-protein interactions upon stress and RNA-related functions of "unconventional" RBPs.

To address this lack of knowledge, we monitored the changes of the mRNA-binding proteome (mRBPome) upon oxidative stress in the yeast Saccharomyces cerevisiae. Intriguingly, we observed prime changes in the RNA associations among enzymes acting in carbon metabolism, which is reminiscent to previously reported metabolic reconfigurations.

In our proposed research, we wish to investigate whether the observed changes of enzyme-mRNA interactions are stress-dependent, and undertake a functional analysis of selected enzymes. Therefore, we will i) profile the changes of the mRNA-protein interactome across distinct environmental stress conditions, ii) comprehensively profile the RNA targets and proteins interacting with selected enzymes in stressed and non-stressed cells; and iii) monitor the implications of specific enzyme-RNA interactions in gene expression, for enzyme activity and consequences in stress adaptation.
Our research is expected to elucidate "moonlighting" functions of key metabolic enzymes and likely uncovers new principles for cellular stress adaptation. Since both, enzymes acting in carbon metabolism and the cell's response to oxidative stress are of considerable interest - ranging from improving fermentation processes in yeast which are relevant in the food and biofuel industry to the development of new strategies for cancer treatment - our research could eventually generate economical and societal impact.

Planned Impact

Our project deals with fundamental aspects of cell biology, studying the reaction of yeast cells to stress, in particular oxidative stress, and lead to new understanding of the implications of enzyme-RNA interactions for stress adaptation. Therefore, it will capture the interest of several groups of beneficiaries outside of the academic research community (outlined in the academic beneficiaries section). In the following, we describe some of these groups, and define how they will benefit from our research (further information is also given in the "Pathways to Impact" attached to this proposal).

i) Industrial biotechnology: Baker's yeast is the most commonly used organism for industrial production of ethanol via fermentation (e.g. beer, wine and bioethanol). Yeast are exposed to manifold stresses during industrial processes which leads to unwanted cell damage and reduction in fermentation ability. Therefore, there is great interest from these industries to understand critical factors that contribute to stress tolerance for optimisation of industrial yeast, which could deliver substantial economic impact. Of note, members of our department (e.g. CAR) have established contacts to companies working in this field (see pathways to impact for details).

ii) Pharmaceutical companies: Oxidative stress is closely linked to chronic inflammation, cancer and ageing. Moreover, diverse diseases are associated with mutations in enzymes of the central carbon metabolism, e.g. the phosphate pentose pathway has been implicated in several human diseases including metabolic syndrome, neurodegeneration (Alzheimer's disease), cardiovascular disease, parasite infections and cancer. Thus, an understanding of additional activities for metabolic enzymes acting in carbon metabolism will be of great interest when devising new strategies for treatment of disease. In the long term, our research could therefore attract the interest from pharmaceutical industries. AG has established contacts to Pfizer and Novartis, two pharmaceutical companies developing cancer treatments and through his previous workplace at the Institute of Pharmaceutical Sciences, ETH Zurich.

iii) UK trained workforce: This proposal includes the training of a PDRA researcher who will acquire new skills and knowledge in RNA biology, enzymes, and metabolism as well as in bioinformatics (e.g. proteomics and next-generation sequencing data analysis). The PDRA will thus mature into a highly trained researcher who will be able to pursue a career in academic or industrial research. In addition, the PDRA will be in a position to teach high-level techniques to postgraduate students. This will impact in the area of training and delivery of highly skilled researchers.

iv) Undergraduate and postgraduate students: The proposed research will contribute to fundamental theories and concepts underpinning the regulation of gene expression, metabolism and stress response. We will impart this new knowledge to students, via teaching activities and research project supervision.

v) The general public: Since our research will have such widespread implications, ranging from medicine/health to the food/beverage/biofuel industry by connection two previously separated fields of research, (namely RNA biology and metabolism), we expect that our result will attract substantial interest from the media and the general public. Hence, our research will have impact in the broader areas of public engagement, public health and societal issues.

Finally, by completing this project, we will reinforce the UK's position in the field of RNA research and metabolism, contributing to the attraction of talented undergraduate students and postgraduate researchers to UK universities. It also enhances our collaborations with international leading scientists and thus, it will also impact in the area of international development.

Publications

10 25 50
 
Description In this project, we aimed to investigate changes of the RNA-protein interactions upon oxidative and other stress conditions in the yeast Saccharomyces cerevisiae (aim 1); identify the RNA targets for stress-responsive key metabolic enzymes (aim 2); and perform initial molecular characterization of RNA-binding properties of those enzymes (aim 3).

Aim 1: At first, we have optimised RNA-interactome capture (RIC) protocol to reduce background and to increase specificity and recovery of RNA-protein complexes. For instance, we have been comparing different crosslinking procedures. We observed that in vivo crosslinking of complexes with low concentrations of formaldehyde was more efficient than classical UV-crosslinking. We have also been recapitulating the changes in the RNA-associations of several conventional and unconventional RBPs upon oxidative stress. The analysis was then expanded to the proposed additional stress conditions, namely reductive stress (treating cells with 0.5 mM dithiothreitol for 15 min) and compared with cells grown in medium with acetate as carbon source instead of glucose. RIC was performed in triplicates across those stress conditions from non-crosslinked cells (negative control), as well as from UV-irradiated and formaldehyde-treated cells to crosslink RNA-protein interaction in vivo. In addition, total cell proteome analysis under the different treatments and growth conditions was performed for data normalization; and we used standard transcriptome analysis with RNA-seq. The MS analysis involving >80 samples was performed by our collaborator Alexander Schmidt (AS; Biozentrum, University of Basel). During the last year, the data was re-analysed in-depth. This was done as we realised that the initial analysis did occasionally not match with validation results obtained with standard immuno-blot analysis, mainly due to the handling with missing values from MS. Thus, we implemented modelling approach to cope with missing values (a missing value imputation approach), which has recently developed by our collaborator AS. While this reanalysis did not substantially change the overall outcome, we believe that we have obtained a very comprehensive high-quality data set.

Essentially, our analysis revealed distinct patterns of changes in RNA-protein protein interactions across the chosen conditions. Interestingly, we found strong links to metabolisms as many RNA-binding enzymes showed coordinated changes. As could be expected, the changes were most prominent in cells grown in acetate medium as compared to glucose. Here, more than 500 proteins changed associations (FDR < 5%), with a substantial overlap when cells were UV-irradiated to crosslink RNA-protein interactions as compared to chemically crosslinked cells. Interestingly, the differences can be allocated to certain protein complexes that are either preferentially captured with UV or formaldehyde, respectively; and it highlights a currently uncharacterised specificity associated with each methodological approach. Furthermore, our data further supports strong links to metabolic configurations as recently suggested by us with only one stress-data (oxidative stress; Matia-Gonzalez et al., 2021, iScience). Hence, our comprehensive study adds evidence for a strong-coordination between metabolism and RNA-binding, primarily through RNA-binding enzymes. We believe that this finding is remarkable and important as it suggests many more direct connections between different layers of cellular-control that is almost unknown and should be further characterised in the future.
Another key finding concerns a strong bias towards paralogue-specific alterations of RNA-protein interaction triggered by the different growth conditions. On this line, careful reanalysis of the MS data was in need as the distinction of paralogous proteins is not trivial due to relatively high overlapping peptides detected by MS. In addtion, the results can also provide new insights into evolution of enzymes, adding a new layer for specialization of the enzymatic vs. RNA-binding functions. It will be interesting to monitor such paralog specificity in "higher" eukaryotes in the future as we have obtained preliminary data for such specialisation.

Aim 2/3: We have established a simplified yeast CLIP protocol based on chemically crosslinked yeast cells. We prepared quality-controlled RNA-seq libraries for 3 metabolic enzymes plus a control sample (the well-characterised RNA-binding protein Puf3 in yeast). However, we repeatedly experienced problems with our in-house sequencing facility (Illumina Miniseq), possibly due to failures in machine/cartridge maintenance. Thus, we are currently looking for a possibility to outsource the sequencing for our prepared libraries. Nevertheless, we confirmed RNA-binding for several new metabolic enzymes with more standard RNA-binding assays, confirming new RNA-binding proteins. In this regard, we got particularly interested in several enzymes acting in fatty-acids metabolisms that show a paralog-specific RNA-associations depending on treatment of growth of cells. The proposed functional characterization is ongoing (aim 3), and we have obtained a series of single and double mutant yeast strains for further investigation, and we are further working on those aspects with project and master students.
Exploitation Route Besides a research paper describing the results of one stress treatment (Matia-Gonzalez et al., 2021, iScience), we recently published our standard and optimized RIC protocol open access to dissiminate the protocol to the wider research community (Matia-Gonzalez et al., 2021, Star Prot.). We also shared the protocol early on and advice other laboratories to perform likewise experiments in other organisms (e.g., Trypanosoma, Instituto Carlos Chagas, FIOCRUZ PR, Curitiba, Brazil). Furthermore, we have now aquired a substantial expertise in MS analysis that is beneficial for future projects. At this point, we are collating the data and prepare a manuscript mainly describing the results from aim 1 and confirming RNA-protein interactions for some new metabolic enzymes along aims 2 and 3. To do so, we have been developing visualization tools to present the data from all our three analysis levels (transcriptome, proteome, and mRBPome) in an 'intuitive' fashion that can be understood by the non-specialist in the field. A bioinformatician from University of Granada (Spain) joined us as an EMBO short-term fellow to do so and we are writing-up the results
Sectors Agriculture, Food and Drink,Education,Healthcare,Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology

 
Title Saccharomyces cerevisiae mRBPome in oxidative stress conditions 
Description Yeast S. cerevisiae proteomics data have been deposited in the ProteomeXchange Consortium database under accession code PXD005943. This data is associated with the following publication: Ana M Matia-González, Ibtissam Jabre, Emma E Laing, André P Gerber (2021)Oxidative stress induces coordinated remodeling of RNA-enzyme interactions. iScience 24(7), 102753 
Type Of Material Database/Collection of data 
Year Produced 2021 
Provided To Others? Yes  
Impact Data has been deposited and allows researchers to extract raw data for analysis of yeast mRNA binding proteins under mild oxidative stress conditions. 
URL http://www.ebi.ac.uk/pride/archive/projects/PXD005943
 
Description Collaborator - Dierk Niessing (University of Ulm) 
Organisation University of Ulm
Country Germany 
Sector Academic/University 
PI Contribution AG has a long-standing collaboration with Dierk Niessing. In the frame of the allocated grant projects, AG has contacted Dierk on technical issues for measureing RNA-protein interactions and whether there is an interest in solving the struture of enzyme-RNA complexes. AG has also advice Dierk in a related project on RIP-chip experiments and data interpretation. Both group are keen to embark into a long term collaboration to study biophysical properties of RNA-enzyme interactions.
Collaborator Contribution Dierk's lab has long-standing experties in protein cristallography and solved numerous RNA-protein complexes. His lab has also further optimized and produced good amounts of recombinant enzymes that are used in our lab for RNA-protein interaction and enzymatic assays. We hope to engage in a long term collaboration based on the results obtaiined from associated project grants.
Impact Output are to come. We though have initiated some collaborative work years ago and a paper describing the outcome was submitted during the last year. However, since the manuscript was rejected, further studies are planned to improve the impact and for resubmisssion soon.
Start Year 2020
 
Description Collaborator - Dr. Alexander Schmidt, Biozentrum, Basel (Yeast Proteomics) 
Organisation University of Basel
Department Biozentrum Basel
Country Switzerland 
Sector Academic/University 
PI Contribution We got in contact with Alex in the frame of a common grant application with Mihaela Zavolan, University of Basel. Alex heads the MS facility at the University of Basel and conducted several yeast proteomics analysis.
Collaborator Contribution Alex performed some yeast proteomics TMT analysis for pkf mutants for us. Data was processed by Alex and sent to us for further analysis.
Impact We submitted common grant applications (with Mihaela Zavolan) to HFSP and ERC. Alex adviced on and performed yeast proteomics analysis. He will be a co-author on upcoming publications that include MS analysis from his facility.
Start Year 2020
 
Description Collaborator - Fred Allain (ETH Zurich) 
Organisation ETH Zurich
Department Department of Biology
Country Switzerland 
Sector Academic/University 
PI Contribution Collaborated in the past in the frame of a Sinergia grant from Swiss National Science Foundation. In the frame of your enzyme-RNA project, we shall provide enzymes and RNA targets for further structural investigation by the collaborator.
Collaborator Contribution Collaborated in the past in the frame of a Sinergia grant from Swiss National Science Foundation. F.A. shall contribute for the analysis of enzyme-RNA complexes with NMR and MS in later stages of this project.
Impact Publication on past collaborations in the frame of a Sinergia project. On current project, no output yet.
Start Year 2009