James Till, University Health Network, Toronto, Canada
29 September 2009
The model shown in Figure 1 is an interesting one. It would have been helpful if the author had included comments about the particular context where purified stem cells have been studied. See, for example, the publications on purified hematopoietic stem cells by Spangrude, Heimfeld and Weissman, 1988 and Benveniste, Cantin, Hyam and Iscove, 2003.
Competing interests
No competing interests.
Response to comment by Professor James Till
Arthur D Lander, University of California, Irvine
29 September 2009
I believe that the question being asked is, "how does the model presented in Figure 1 bear on published results in which purified hematopoietic stem cells have been studied in reconstitution assays, and particularly in those studies in which quantitative analysis of engraftment efficiency over time has been measured?" This is an interesting question. Benveniste et al. (2003), for example, provide evidence that purified hematopoietic stem cells (HSCs) engraft with near 100% efficiency, but then clones are subsequently lost; this was taken as evidence in favor of the model of stochastic renewal (and extinction) of HSCs originally developed by Till and colleagues. This interpretation is not challenged by the possibility of feedback regulation of renewal probability (my Figure 1). What feedback regulation does is make renewal (and extinction) probabilities variable, rather than constant. For example, in the bone marrow of an irradiated host, in which mature hematopoietic cells are depleted, one might expect renewal probabilities of HSCs to be high shortly after transplantation, falling as restoration of normal mature cell numbers is achieved. This could contribute to the observation that short-term clonal engraftment efficiencies are so high (near 100%) at the outset, falling a great deal later on. It is certainly possible to build stochastic models of lineages that incorporate feedback, but to make precise predictions one would need to know the degrees to which feedback regulation is achieved through changes in the proportion of divisions that are asymmetric, the proportion of symmetric divisions that are renewal divisions, or both. It would also be necessary to know whether there are feedback effects on cell cycle speed as well. At the moment we don’t have this information for HSCs (in part because we don’t know with certainty what the feedback molecules are). Interestingly, a paper published a few months ago by Marciniak-Czochra et al. (Stem Cells and Development 18(3), 377-385 doi: 10.1089/scd.2008.0143) argues, based solely on the dynamics of hematopoietic repopulation following bone marrow transplantation, that feedback regulation just like that in my Figure 1 must occur in the HSC system (I apologize for not citing this paper in my piece; I only just learned of its existence).
Thanks, Professor Till for your interesting comment.
What about purified stem cells?
29 September 2009
The model shown in Figure 1 is an interesting one. It would have been helpful if the author had included comments about the particular context where purified stem cells have been studied. See, for example, the publications on purified hematopoietic stem cells by Spangrude, Heimfeld and Weissman, 1988 and Benveniste, Cantin, Hyam and Iscove, 2003.
Competing interests
No competing interests.
Response to comment by Professor James Till
29 September 2009
I believe that the question being asked is, "how does the model presented in Figure 1 bear on published results in which purified hematopoietic stem cells have been studied in reconstitution assays, and particularly in those studies in which quantitative analysis of engraftment efficiency over time has been measured?" This is an interesting question. Benveniste et al. (2003), for example, provide evidence that purified hematopoietic stem cells (HSCs) engraft with near 100% efficiency, but then clones are subsequently lost; this was taken as evidence in favor of the model of stochastic renewal (and extinction) of HSCs originally developed by Till and colleagues. This interpretation is not challenged by the possibility of feedback regulation of renewal probability (my Figure 1). What feedback regulation does is make renewal (and extinction) probabilities variable, rather than constant. For example, in the bone marrow of an irradiated host, in which mature hematopoietic cells are depleted, one might expect renewal probabilities of HSCs to be high shortly after transplantation, falling as restoration of normal mature cell numbers is achieved. This could contribute to the observation that short-term clonal engraftment efficiencies are so high (near 100%) at the outset, falling a great deal later on. It is certainly possible to build stochastic models of lineages that incorporate feedback, but to make precise predictions one would need to know the degrees to which feedback regulation is achieved through changes in the proportion of divisions that are asymmetric, the proportion of symmetric divisions that are renewal divisions, or both. It would also be necessary to know whether there are feedback effects on cell cycle speed as well. At the moment we don’t have this information for HSCs (in part because we don’t know with certainty what the feedback molecules are). Interestingly, a paper published a few months ago by Marciniak-Czochra et al. (Stem Cells and Development 18(3), 377-385 doi: 10.1089/scd.2008.0143) argues, based solely on the dynamics of hematopoietic repopulation following bone marrow transplantation, that feedback regulation just like that in my Figure 1 must occur in the HSC system (I apologize for not citing this paper in my piece; I only just learned of its existence).
Thanks, Professor Till for your interesting comment.
Competing interests
None declared