Study: Discoveries in Non-Human Species Could Directly Benefit Man
28/04/10 12:37 Filed in: GenomeWeb Daily Scan
Submitted by S. Pelech - Kinexus on Wed, 04/28/2010 - 14:26.
The basic concept that investigations in yeast and other simple model organisms can lead to insights into human molecular physiology has been around for decades. In some cases, such as the basic processes of cell reproduction and metabolism, this activity has been fruitful and its success has widely expressed. However, it is a great leap to extrapolate from the functions of signalling proteins in these minute organisms to their roles in humans. Most of the major diseases confronting humans, including cancer and other diseases of aging, arise from defective cell signalling.
The most useful information from studies of less complex organisms that are more readily subjected to genetic manipulation is the identification of possible protein interactions to define rudimentary protein complexes and regulatory pathways. Critical regions within proteins, including phosphorylation sites, may also be identified through comparative studies of cognate proteins in diverse species. However, less than 2% of the estimated 500,000 phosphorylation sites in human proteins are located in cognate proteins in yeast.
Nature has been very creative in the deployment of protein kinases and other signalling proteins for very distinct and sometimes contrary purposes. For example, cAMP-dependent protein kinases facilitate yeast proliferation, but typically inhibit human cell growth. In yeast, cdc2 kinase regulates progression at multiple stages of the cell cycle, but in humans this is carried out by more than 7 different cyclin-dependent protein kinases. Fus3 and Kss1 MAP kinases in yeast are important for mating; their homologues in the fruit fly are critical for compound eye formation; the human counterparts ERK1 and ERK2 control a myriad of wide ranging functions. The more one critically examines the processes controlled by the 515 known human protein kinases in other species, it becomes clearer that their functions can change profoundly. Consequently, I am bemused with the regular announcements that yet another researcher has made a breakthrough in the understanding of a human disease from the study of a gene or protein in a species separated by a billion or more years of evolution.
The conservation of the amino acid sequence identities of human protein kinases in yeast is usually 40% or less, even if they possess the cognant kinases. Yeast have about 139 protein kinase-like genes out of about 6000 total genes; humans have about 620 protein kinase-like genes including pseudogenes out of around 23,000 genes. At Kinexus Bioinformatics Corporation, we estimate that the typical protein kinase targets about 1000 phospho-sites distributed over several hundred proteins. Each kinase on average appears to be phosphorylated at over 20 phospho-sites, and each of these sites are probably targeted by more than 20 protein kinases. It is obvious that the regulatory signalling pathways in mammals are orders of magnitude more complex than in yeast, which hardly feature any protein-tyrosine phosphorylation at all. Most of the known human oncogenes encode proteins that catalyze tyrosine phosphorylation, induce activation of tyrosine kinases or are substrates targeted for tyrosine phosphorylation.
The analysis of simple model organisms will likely continue to provide useful information about basic housekeeping cellular processes. However, a more comprehensive understanding of human disease processes will have to arise from studies of human and mammalian model systems.
Link to the original blog post
The basic concept that investigations in yeast and other simple model organisms can lead to insights into human molecular physiology has been around for decades. In some cases, such as the basic processes of cell reproduction and metabolism, this activity has been fruitful and its success has widely expressed. However, it is a great leap to extrapolate from the functions of signalling proteins in these minute organisms to their roles in humans. Most of the major diseases confronting humans, including cancer and other diseases of aging, arise from defective cell signalling.
The most useful information from studies of less complex organisms that are more readily subjected to genetic manipulation is the identification of possible protein interactions to define rudimentary protein complexes and regulatory pathways. Critical regions within proteins, including phosphorylation sites, may also be identified through comparative studies of cognate proteins in diverse species. However, less than 2% of the estimated 500,000 phosphorylation sites in human proteins are located in cognate proteins in yeast.
Nature has been very creative in the deployment of protein kinases and other signalling proteins for very distinct and sometimes contrary purposes. For example, cAMP-dependent protein kinases facilitate yeast proliferation, but typically inhibit human cell growth. In yeast, cdc2 kinase regulates progression at multiple stages of the cell cycle, but in humans this is carried out by more than 7 different cyclin-dependent protein kinases. Fus3 and Kss1 MAP kinases in yeast are important for mating; their homologues in the fruit fly are critical for compound eye formation; the human counterparts ERK1 and ERK2 control a myriad of wide ranging functions. The more one critically examines the processes controlled by the 515 known human protein kinases in other species, it becomes clearer that their functions can change profoundly. Consequently, I am bemused with the regular announcements that yet another researcher has made a breakthrough in the understanding of a human disease from the study of a gene or protein in a species separated by a billion or more years of evolution.
The conservation of the amino acid sequence identities of human protein kinases in yeast is usually 40% or less, even if they possess the cognant kinases. Yeast have about 139 protein kinase-like genes out of about 6000 total genes; humans have about 620 protein kinase-like genes including pseudogenes out of around 23,000 genes. At Kinexus Bioinformatics Corporation, we estimate that the typical protein kinase targets about 1000 phospho-sites distributed over several hundred proteins. Each kinase on average appears to be phosphorylated at over 20 phospho-sites, and each of these sites are probably targeted by more than 20 protein kinases. It is obvious that the regulatory signalling pathways in mammals are orders of magnitude more complex than in yeast, which hardly feature any protein-tyrosine phosphorylation at all. Most of the known human oncogenes encode proteins that catalyze tyrosine phosphorylation, induce activation of tyrosine kinases or are substrates targeted for tyrosine phosphorylation.
The analysis of simple model organisms will likely continue to provide useful information about basic housekeeping cellular processes. However, a more comprehensive understanding of human disease processes will have to arise from studies of human and mammalian model systems.
Link to the original blog post