Exploring the hidden small proteome of a unicellular eukaryote

Our bodies are made of very different types of cells: Skin cells are flat and protect our body, while brain cells have cables that pass messages around. Despite being so different, all our cells carry exactly the same information in their genes. What makes them special is what information they use, that is, which genes they switch on and off.

The information on how to make a cell is stored in the form of a DNA molecule. However, this information cannot be read directly: it first needs to be copied into another molecule called messenger RNA (mRNA), from which it can be 'translated' into a protein. Proteins are the components that directly build the cell and make it function. Cells also produce other RNAs that are not translated to make proteins (non-coding RNAs, or ncRNAs), which have other roles in the cell.

The identity of a protein can be predicted from the sequence of the RNA. Moreover, proteins can also be identified directly using specialized techniques. However, both approaches are very inefficient at identifying very small proteins. Thus, these proteins have been largely ignored by researchers, even though there are examples of small proteins with key biological functions.

A new experimental method has been recently developed that allows the detection of every RNA region that is actively translated in a cell. From these data, all proteins can be predicted regardless of their size. The method is called 'ribosome-profiling' after the ribosome, which is the cellular machine that carries out translation. The application of this approach to several organisms has revealed the existence of hundreds of previously unknown predicted short proteins. Many of these translated regions were in RNAs that were not thought to be translated (ncRNAs). In some organisms, these short may proteins represent as much as 20% of all previously known proteins.

Our aims are to identify small proteins systematically and to understand how they work. One way to study a complicated process of the human body is to use a model organism: this is a simpler creature, but similar enough to allow us to learn about ourselves. To study these questions we will use a simple yeast -made of a single cell- that can acquire different forms. We will use different methods to identify all small proteins produced by these cells. We will then remove individual proteins and study how this affects how cells grow and reproduce.

We expect this information will be useful to understand how human cells behave and, eventually, help us devise cures for disease.

The prediction and experimental identification of small proteins (<100 aminoacids) are challenging. Thus, they have frequently been overlooked and most remain unidentified. However, there is evidence from single-celled and multicellular eukaryotes that small peptides (as short as 11 aminoacids) have essential biological roles.

A recently developed approach, called ribosome profiling, has revolutionized the study of small proteins. This method allows the systematic identification of translated regions regardless of their length. The application of this technique to yeast and animals revealed hundreds of translated short open reading frames (sORFs), located in 5' leader sequences, genes annotated as non-coding RNAs and novel transcripts. If the proteins translated from these loci (sORF-encoded peptides, or SEPs) are stable, this would represent a major increase to the complexity of eukaryotic proteomes. For example, our ribosome profiling of the fission yeast Schizosaccharomyces pombe revealed >900 translated sORFs of 20 codons or longer, which correspond to ~ 20% of the known proteome of this well-studied organism (Duncan and Mata, Nature Stuctural Molecular Biology 2014).

Very little is known about the expression and function of SEPs. We will use the model organism S. pombe to study these questions. We will initially create a comprehensive list of sORFs by performing ribosome profiling in additional conditions, and complement these studies with targeted mass spectrometric approaches to identify stable SEPs. We will then systematically study the phenotype of cells in which sORFs have been inactivated or overexpressed. Finally, we will select a small number of SEPs for detailed characterisation.

These studies will provide an overview of the expression and functional importance of this novel class of proteins. We expect that this work will allow the discovery of general principles that may be applicable to human cells.

This project will contribute to the training of researchers in key areas of research, and will result in knowledge that may have long-term implications in various areas of medical research as described below.

The biotechnology and pharmaceutical industries are potential beneficiaries of this project, both through the training of highly qualified researchers (point 1) and the knowledge and expertise it will generate (points 2 and 3). In addition, the project may contribute to fighting human disease, which would benefit the general public (points 2 and 3). The 'Pathways to Impact' document discusses in detail how we will ensure that the potential beneficiaries of this project will be reached.

[1] Training and capacity building in functional genomics / systems biology. This project will provide an excellent opportunity for the training of the postdoctoral researcher in the analysis of large scale datasets, both proteomic and genomic. This will be done through the work carried out in the laboratory, as well as through courses and interactions with members of the Cambridge Systems Biology Centre. The provision of scientists trained in these multidisciplinary approaches will be beneficial for the UK industry, especially the biotechnology and pharmaceutical sectors. This is also a key objective of the BBSRC strategic plan 'Exploiting new ways of working', which aims to 'enhance skills and capacity to exploit new tools and approaches e.g. through training for researchers'.

[2] General understanding of human disease: Although very little is known about the function of small peptides, there are examples that demonstrate their function for key biological processes. For example, 30-aminoacid peptides regulate calcium uptake in the heart and have been have been implicated in cardiac pathologies. This proposal aims at identifying general principles of how small peptides regulate biological functions, which may be applicable to human cells.

[3] As described in more detail in the 'Academic Beneficiaries' section, recent work has shown similarities between pathogens of the Pneumocystis genus and fission yeast (in particular, in their meiotic pathways). These organisms cause pneumonia in patients with weakened immune systems (premature babies, AIDS and cancer patients). As Pneumocystis cannot be cultured in vitro, there is a need for model systems that allow the study of their basic biology. Therefore, our results on the function of small proteins might be useful to understand the biology of these pathogens and develop treatments against their infection. To make sure this information reaches the Pneumocystis research community, we will highlight these similarities in peer-reviewed publications, our website and relevant scientific conferences.