Proteome wide identification of clinically-relevant carbon dioxide targets

Carbon dioxide is a bioactive gas whose altered partial pressure has clinical consequences. There is an emerging consensus that the beneficial or detrimental effects of CO₂ depend upon the specific clinical context. Therefore, exploring the therapeutic potential for CO₂ depends on identifying CO₂ sensors that explain clinical observations. We have identified chromatin associated proteins as CO₂ targets. We propose to characterise the impact of CO₂ on clinically significant cell processes dependent upon modification of chromatin-associated proteins. The studentship, therefore, will reveal specific clinical settings where increased PCO₂ might be beneficial or detrimental.

Background

Scientific consensus has considered CO₂, at best, a relatively inert metabolic by-product, and at worst, a toxic molecule with severe clinical consequences if dysregulated. However, clinical observations demonstrate elevated CO₂ partial pressure (PCO₂) might be beneficial to patients in certain circumstances.

There is clinical value in addressing the molecular mechanisms for the physiological effects of CO₂. A large cohort study (LUNG SAFE, NCT02010073) and retrospective analysis of the Acute Respiratory Distress Syndrome Network demonstrated the potential therapeutic benefit of elevated PCO₂. However, elevated PCO₂ can also have harmful pathophysiological effects on the lung (alveolar fluid clearance and epithelial cell repair), skeletal muscle and innate immunity and host defence.

Understanding the therapeutic potential for CO₂ requires knowledge of its sensors. We have established two methodologies for identifying CO₂ target proteins, experimental and computational.

We have deployed the experimental methodology to identify CO₂-binding proteins and investigate the therapeutic potential for CO₂. We have identified chromatin-associated proteins as CO₂ targets and demonstrated that CO₂ modifies lysine acetylation/methylation in transcriptional control. This studentship will

1. Investigate the biomedical consequences for chromatin modification through CO₂ to reveal the therapeutic potential for PCO₂ in the clinic.

2. Deploy computation to reveal further CO₂ targets in the human proteome that might impact on these processes

Research Plan

Hypercapnia is known to target the Wnt signalling pathway and Wnt dysregulation is linked to a range of diseases including cancer and cardiovascular disease.

The research plan is split an experimental stream and a computational stream.

The experimental stream will address the hypothesis that hypercapnia impacts Wnt signalling through modifying chromatin-associated targets thus altering the production of Wnt-target gene products.

The computational stream will use a computational approach to identify all CO₂ binding sites in the human proteome with an emphasis on targets in the Wnt signalling pathway.

Experimental stream

CO₂-binding in vitro

Use chemical proteomics and demonstrate CO₂-binding to chromatin-associated proteins over physiologically meaningful partial pressures.

Use ¹³C-NMR to demonstrate CO₂-binding under native conditions.

CO₂-binding in vivo

Use quantitative chemical proteomics to demonstrate CO₂-binding to chromatin-associated proteins in the cell over physiologically meaningful partial pressures.

Use a combination of ELISA and co-immunoprecipitation to demonstrate that CO₂-binding correlates with alterations in histone acetylation/methylation.

Biomedical consequences

Use whole-genome ChIP to investigate altered chromatin-binding by a site-specific histone interactor over physiologically meaningful PCO₂.

Knockout a site-specific histone interactor and examine the effect on transcriptomics over physiologically meaningful PCO₂.

Computational stream

CO₂ can form a carbamate post-translational modification on lysine to mediate CO₂ sensing. We have shown a link between structure-dependent lysine epsilon-amino pK_aH, solvent accessibility and CO₂-binding. We have developed software to calculate lysine pK_aH in Protein Data Bank and AlphaFold file ensembles. The software will download a user-defined file list, curate the structures and flag structures requiring manual curation. It will report, for each lysine in each structure, a set of descriptors including pK_aH at physiological conditions, solvent accessible surface, predicted flexibility and local sequence conservation.

We will apply the software to a training set of >200 structures known, from the literature and our own work, to feature both carbamylated and non-carbamylated lysines. Approximately half of the proteins will be carbamylated and derive from structurally diverse proteins. The lysine descriptors will constitute a training set for a Random Forests classifier, tasked to predict which are carbamylated. We will determine which combination of structural features provides the highest predictive power and apply our software and trained classifier to the human proteome.

Predicted carbamylated proteins will be analysed by GO term ontology for those in the Wnt signalling pathway and feed into the experimental stream.

Training & Skills

Interdisciplinarity is embedded into the proposal. The student will experience training in skills in Biosciences, Chemistry, and Computational Biology.

Biosciences-Training in molecular biology, protein expression and purification, cell biology, microscopy.

Chemistry-Training in chemical handling, potentiometry, NMR.

Interdisciplinary-Training in mass spectrometry, data handling and analysis, computational skills.

Computation-Python, Molecular dynamics. Prior familiarity with Python is desirable.

Entry Requirements:

Successful candidates will have an excellent Master’s level degree in a subject aligned with your chosen project. Typically, this will be synthetic chemistry, medicinal chemistry, theoretical chemistry, biochemistry, molecular biology, structural biology, cell biology, molecular modelling, computer science, or biophysics. First class Bachelor degrees plus relevant experience will also be considered.

Please note: this is a four year fully-funded PhD project available for October 2022

Please contact the Lead Supervisor Professor Martin Cann ([Email Address Removed]) for further information about the project. For enquiries regarding the application process for this project please contact the MoSMed CDT Manager (Durham): [Email Address Removed]

For Information on how to apply for this project please refer to the following webpage: https://research.ncl.ac.uk/mosmed/apply/ Please note application to this project is through the Durham University application portal to be found at: https://studyatdurham.microsoftcrmportals.com/en-US/