Proteins are the fundamental building blocks of life: they form the basis for many cellular processes in the body. Changes in the abundance of a specific protein can lead to diseases like Alzheimer’s or Huntington’s.
Dr. Grant Brown and his team at U of T’s Donnelly Centre for Cellular and Biomolecular Research have quantified more than 90 per cent of the Saccharomyce cerevisiae yeast proteome. The proteome is the entire set of proteins expressed in a cell. The study estimated the total number of protein molecules in a S. cerevisiae cell to be approximately 42 million.
Knowing the abundance of a specific protein in a cell allows scientists to determine the function of the cell and offers insight into diagnosing diseases. “It’s more like a protein census of a cell,” said Brown. “If you’re doing a census of people in Toronto and you work your way north from the lake, and suddenly you find that there’s an increase of bankers and brokers, that might tell you something about the function of the location… From a disease sort of perspective, there [are] very few diseases that don’t have some basis in alteration of protein.”
The researchers did not need to quantify this number from scratch — the yeast cell is extensively studied, partly because it is less complicated than other eukaryotic cells, such as human cells. Instead, Brown and his team combined data from 21 other yeast protein abundance studies and converted these into one unifying measurement of protein molecules.
Because there is more than one method to quantify protein abundance, Brown’s team were working with results that ranged from methods like green fluorescence protein microscopy to mass spectrometry (MS).
Each method reports protein abundance in different units. After testing various statistical techniques to convert the datasets into a common number of protein molecules per cell, the result was simpler than expected: they were able to use the mean of the five MS-based studies, which already reported in molecules per cell, as a calibration tool to convert the other datasets into protein molecules per cell.
“The really nice thing is that the model is very simple,” said Brown.
Using their knowledge about protein abundance, the Brown lab plans to further investigate the function of each cell in the yeast genome. Brown said that this computational model can also be applied to data sets from mammalian cells with bigger proteomes that are more difficult to compute.
“I like to think of this study as a nice resource or tool for the community,” said Brandon Ho, lead author and PhD student in the Brown lab.