Biology Reference
In-Depth Information
dynamical semi-stable states or pluripotent attractors
[20,21,43,55]
. In extreme cases, stochastic variations in gene
expression may lead to a loss of one or more of the key self-
renewal factors. The remaining skewed and/or incomplete
combinations of the core factors may result in commitment
to a specific cell lineage
[10,55]
. For instance, alternative
expression of Oct4 or Sox2 in mouse ESCs may result in
commitment of cells to either mesendodermal (ME, Oct4) or
neuroectodermal (NE, Sox2) fates
[22, 56, 57]
.
Major efforts in the stem cell field are currently focused
on finding optimal factor combinations or 'cocktails' that
will ensure stem cell differentiation towards desired cell
types or tissues
[1,58
to dynamic attractors (see
Figure 22.1
A). Three potential
scenarios were revealed, all producing bimodal statistical
distributions for Nanog concentrations, observed in ESC
culture in vivo
[43]
. In the first scenario, highly cooperative
interactions between the core components produced a classic
bistable switch with two attractors, the first corresponding to
the high and the second corresponding to the low concen-
tration of Nanog (see
Figure 22.3
A)
[20,21]
. It has been
assumed that stochastic variations in gene expression may
lead to transition between the states without loss of pluri-
potency
[18]
. Addition of a transcriptional repressor to the
core pluripotency network produced a second scenario,
oscillation (
Figure 22.3
B)
[21]
. Interestingly, in the context of
the oscillatormodel, no stochastic noise is actually required to
achieve transition between the 'high' and the 'low' states. The
addition of inhibition to the core circuit may reflect known
antagonistic interactions between the transcriptional regula-
tors Cdx2 and Oct4
[61]
, Zfp281 and Nanog
[62]
or Tcf3 and
Nanog
[63]
. Some of the core factors, such as Oct4, may be
involved in both activation and repression
[3]
; the dual
functions of these pluripotency factors lead to a third possible
scenario, involving mutual repression (
Figure 22.3
C)
[64]
.
While the unidirectional repressive interactions considered
above may produce oscillatory behavior of the system,
mutual repression results in a scenario similar to the classic
phage lambda switch
[65]
, but rather different from the
cooperative switch considered above.
60]
. Hopefully, future quantitative
biological models explaining stem cell behavior will help
to carry out these studies in a more systematic and
predictive manner.
e
EXAMPLES OF QUANTITATIVE MODELS
EXPLAINING STEM CELL BEHAVIOR
Models for Embryonic Stem Cells
Embryonic stem cells are becoming a major subject for
quantitative modeling owing to the relatively high level of
understanding of this particular system. Earlymodels focused
on the structure of the core pluripotency network and tran-
sitions between alternative pluripotent states, corresponding
FIGURE 22.3
Known models and scenarios for embryonic and hematopoietic stem cells. (A
e
C) Models for pluripotency in ESC. Dynamic
solutions (phase space) are shown on the right side of each panel. Cooperative interactions between the core factors (A) may produce a bistable system
with two point attractors; unilateral repression, combined with self activation produces oscillatory behavior (limit cycle attractor, B); mutual repression
results in bistability again (C), but with different distribution of the point attractors. (D) Model describing alternative cell fate commitment in ESC. Two
loosely linked bistable switches ensure the fate choice between endoderm (Gata6 expression) and trophectoderm (Cdx2 expression). (E) Two parallel
bistable switches reinforce each other and govern alternative commitment of hematopoietic progenitors (here HSC) to erythroid or myeloid cell fates.
