Revised: Cell cycle kinetics of hematopoietic stem cells

by Alexey Bersenev on March 10, 2011 · 0 comments

in quiescence

Cell cycle kinetics of the hematopoitic stem cell (HSC) population is very important in normal physiology of hematopoietic system and in pathological conditions. The current understanding of cell cycle kinetics of adult stem cell populations is based on two models:
(1) The vast majority of adult stem cells in the population (up to 90-95% for pure HSC) are non-dividing or a slow cycling cell population, which reside in the organs in G-0 phase of cell cycle (quiescent);
(2) The dormancy model of the somatic stem cell population implies the co-existence of phenotypically similar – “deeply quiescent” (dormant) and active stem cells (mostly cycling and metabolically active), residing in distinct niches. Dormant stem cells are virtually non-dividing in steady state organism and get activated only in case of “emergency”. When hibernating dormant stem cells are awakening, their self-renewal and regenerative potential is remarkable. Primed (active) stem cells are mostly responsible for maintenance of homeostasis in steady state condition (normal blood cells turnover).

In order to study cell cycle status of stem cell populations we need to separate phases G1 (enter into the cycle) from G-0 (quiescent) of cell cycle. There are few techniques based on labeling DNA versus RNA content combined with surface phenotype, which allow us to do it. To estimate the turnover rate and dynamical changes, so-called label-retaining assays are widely used. The most common assay is BrdU labeling. BrdU labels DNA and gets diluted equally with each cell division. The withdrawal of BrdU followed by detection (chasing), allows quantification of label retention, indicating the rate of cycling (slow dividing or quiescent). The disadvantages of these techniques are the following:

    (1) done on dead (fixed) cells;
    (2) presence of BrdU (or other pyrimidine analogs) induces cells to cycle due to some toxicity;
    (3) BrdU labeling doesn’t allow clear separation of slow dividing stem cells versus non-dividing.

Three years ago – for the first time – an in vivo method (H2B mice) of measuring quiescence was used for hematopoietic stem cells. Using this methodology on other adult stem cells (for example hair and intestine stem cells) permitted the develop of the “dormancy model”.

Some unresolved questions:
Even after using such sophisticated techniques, as H2B mice, proposing “dormancy model” and “co-existence of 2 stem cell populations” model, some unresolved questions remain. Most of them indicated in the study, which I’ll discuss below.

    (1) Because dormant HSC appear to be the most significant contributers in long-term repopulation after serial transplantation into irradiated mice, it raises a question: how could it be coupled with extremely slow division rate (~ 5 divisions per mouse lifespan)?
    (2) How do dormant HSC rescue hematopoiesis under “emergency conditions” – via differentiation only or acceleration of self-renewal or both?
    (3) How do dormant and active HSC interact and contribute to steady state hematopoiesis – simultaneously, sequentially or repetitively?

All of these questions underlie rationale of the study, which I’ll discuss below.

The challenge:
Implementing a new model, Hitoshi Takizawa from Markus Manz’s group have challenged a “dormant model” for HSC in their recently published study. The authors use CFSE labeling assay in order to track HSC divisional history. Methodological approaches made this study interesting and nicely done. I’d like to summarize the methodological advances as the following:

    (1) using non-irradiated recipients in order to assess steady state contribution of HSC with different divisional dynamics;
    (2) comparative analysis of CFSE versus BrdU techniques unveil mitogenic effect of BrdU on the cells and low resolution and sensitivity after 2nd-3rd divisions made labeling non-linear. Moreover, BrdU (as any other DNA labeling agent) doesn’t label all HSC, unlike CFSE, missing deeply quiescent ones with DNA replication “off”;
    (3) studying a link between divisional history and self-renewal and repopulating ability of individual HSC.

The most important conclusions from the study:

    (1) Majority of HSC (phenotypically LSK) divide actively and only small fraction is quiescent.
    (2) HSC required at least 2-3 divisions prior to starting a lineage commitment program. All of classical HSC surface markers were retained during 0-3 divisions.
    (3) Both frequently dividing and quiescent HSC contribute equally in life-long multilineage hematopoietic turnover in steady state organism.
    (4) In contrast to other studies, the authors demonstrated that self-renewal and repopulation potential of HSC is irrespective to cell cycle status. Some quiescent HSC can give no multilineage repopulation, but some cycling HSC do.
    (5) Frequently cycling HSC with extensive proliferative history can slow down and go into quiescence. So, high proliferative rate does not neccesserily lead to HSC functional exhaustion.
    (6) For the first time authors demonstrated an increased proliferation of HSC upon demand (model of infection) coupled with increased self-renewal rate of HSC rather than differentiation.
    (7) Increased dormancy of aged and expanded HSC was demonstrated.

New model and significance:
The authors proposed a new model of hematopoiesis maintenance by HSC:

Based on our data, we suggest a “dynamic repetition” model, where some HSCs dominate blood formation for a time, subsequently enter a quiescent state in which other HSCs increase hematopoietic contribution, and get reactivated again and contribute to blood formation in repetitive cycles.

So, there is a cell cycle plasticity on a single stem cell level, when each quiescent HSC can start rapidly dividing on demand and then go into quiescence again in order to protect itself from functional exhaustion. I like the concept of “dynamic equilibrium” between quiescent fraction and cycling fraction withing the HSC population.

The authors don’t dismiss the existence of dormant HSC, but they disagree about previously shown dormant cell division dynamics (~5 divisions per mouse lifespan), pointing out to methodological flaws. Even though they don’t criticise the H2B in vivo cell cycle labeling model, they suggest not to use BrdU labeling for assessment of stem cell quiescence. The authors indicated that due to HSC cell cycle heterogeneity, this population can’t be simply split for two permanent subsets with different cycling kinetics.

And of course, because HSC studying is always ahead of other adult stem cells:

The findings reported in this paper might represent a biological principle that could hold true for other somatic stem cell–sustained organ systems and might have developed during evolution to ensure equal distribution of work load, efficient recruitment of stem cells during demand, and reduction of risk to acquire genetic alterations by alternating fractions of stem cells in quiescence at any given time.

The authors give an interesting suggestion about aging of the hematopoietic system. Because quiescence could be a “rehabilitation center” for previously extensively proliferating HSC, we can see an increase of quiescent fraction with age and under influence of “stressful stimuli” for hematopoiesis.

Finally I’d like to speculate about the same possible mechanisms in cancer stem cells. Because recently we have seen cancer stem cell phenotypic and genotypic plasticity, we would expect to see cell cycle heterogeneity as well. It makes selective targeting of cancer stem cells even more difficult, because at any given time point there will be dynamic equilibrium between cycling and quiescent cells. I can’t wait to see the first study investigating this issue in detail.

Recently, researchers were trying to identify the correlation between stem cell engraftment and cell cycle kinetics in human. Interestingly, in bone marrow and mobilized peripheral blood only quiescent CD34+ engrafted, but in cord blood stem cell engraftment didn’t correlate with cell cycle phases G-0 or G-1. Applying some surrogate techniques and simulation, Sandra Catlin was able to calculate the replication rate of human HSC. They replicate once every 40 weeks on average. This is very interesting and unique stimulation study with very intriguing conclusions. But this is a topic of next essay.

{ 0 comments… add one now }

Leave a Comment

Previous post:

Next post: