Biomedical Engineering Reference
In-Depth Information
action of inositol phosphates, resulting in high thresholds and narrow peaks.
When L
f
< 2.2, even the maximum possible increment of S cannot abolish the
threshold at the leading edge, thus making it impossible to provoke a peak. Con-
versely, if L
f
is large, the inositol phosphate pool responds slowly, so that the
peak is wider and the threshold lower. At very large values of L
f
, the threshold is
zero and membrane phosphoinositide synthesis is much faster than its removal.
Hence, the minutest chemoattractant gradient results in propagation of phospho-
inositides throughout the membrane before inositol phosphates can exert the
inhibitory effect required for formation of the polarized steady state. Thus, the
phosphoinositides ultimately increase to a near uniform steady state.
Now, we could just as well assume that the three different distributions of
L
f
S shown in Figure 7 were obtained by keeping the parameter L
f
fixed, and
varying the active receptor distribution S. In this case, the simulations would
imply that the geometry of the phosphoinositide peak varies with the shape of
the external chemoattractant.
3
.
FUTURE WORK
We have shown above that the model captures all the dynamics except ad-
aptation. To be sure, we modeled adaptation in our earlier work by assuming
slow desensitization and resensitization of the receptors (39). In response to uni-
form increases of the chemoattractant concentration, the model displayed perfect
adaptation over several orders of magnitude of the chemoattractant concentra-
tions. However, the mechanism proposed in the model was inconsistent with
experiments, since adaptation occurs at some level below the receptors (46) and
G-proteins (24). This deficiency was resolved in more recent work, wherein we
have taken due account of the fact that adaptation acts below the receptors (40).
Simulations of this modified model are in good agreement with experiments.
Although our local-activator-global-inhibitor model captures the spatiotem-
poral dynamics observed in experiments, the identity of the local activator and
global inhibitor are subjects of considerable debate. Specifically,
1. Experimental data concerning the polarization of PIP
2
is not definitive.
It is difficult to discern spatial variations of the PIP
2
distribution because its
concentration in the plasma is relatively high. It has been argued that the ap-
pearance of PIP
2
polarization observed in earlier work (18) does not reflect a
localized increase in the concentration of PIP
2
—it is a consequence of the high
surface area created at the leading edge by the formation of membrane folds
(47). On the other hand, it has been shown that PI5K, the enzyme that synthe-
sizes PIP
2
, localizes to the ruffles immediately after chemoattractant stimulation
(29,48).
