Finding the Statistical Rules of Blood Regeneration
On September 26, as part of the Fields Centre for
Mathematical Medicine Seminar Series, Professor Sidhartha Goyal from the University of Toronto presented a phenomenological model of blood regeneration to explain how 100 billion new blood cells are made every day from a much smaller pool of blood stem cells.
Using data from a primate experiment that tracked the offspring of blood stem cell clones over time, Goyal studied the clone size distribution, or how much each initial stem cell contributes to the final blood pool. Two interesting results emerged: first, the distribution was very broad, with some clones contributing as much as 10 % of the blood and others contributing very little; second, the size distribution remained constant even as the various clones ebbed and surged in numbers.
“ The stability is in this distribution rather than the individual actors,” says Goyal.
To explain results such as these, the idea of cellular heterogeneity has taken centre stage in blood research – is there some fundamental cell-level difference that determines which clone will overtake the others?
Using a simple, three-compartment model of stem cells, progenitor cells, and somatic cells, Goyal found that, in fact, the same broad distributions can be reproduced without any differences in the starting population of blood stem cells. That is, the results obtained in the primate experiment could be explained purely by chance.
“ This model allowed us to generate a new hypothesis, which right now is completely counterintuitive to anybody who thinks about these things. That is the real reason to do math models in my mind."
“ The truth is probably somewhere in the middle,” he explains, with some clone success due to cell-specific differences and some due to chance."
But if chance is a major driver as Goyal’ s results suggest, then the model makes some interesting predictions that could have profound implications in the treatment of blood diseases like leukemia. Currently leukemia is treated with chemotherapy and bone marrow transplant in order to rid the body of the defective blood stem cell pool and replace it with a healthy one. But if the presence of a large number of cancerous blood cells is not due to their superiority over other normal cells, but simply because of statistical probability, then perhaps the defective cells can be replaced in a different way.
Goyal reasons that if the clone size distribution could be reset and allowed to broaden again, the results may be different, with the cancerous clone now contributing very little to the blood pool. This could be done, Goyal suggests, by adding growth factors, increasing the differentiation and proliferation rate of all clones and transiently evening the playing field.
“ This model allowed us to generate a new hypothesis, which right now is completely counterintuitive to anybody who thinks about these things. That is the real reason to do math models in my mind – that’ s the real contribution.”
Goyal is now looking for collaborators to test these predictions either in vitro or in animal models. The results could change the way we think about treating blood cancers. �
— Malgosia Ip
20