Control Systems in the Living World - Building the Most Complex Structure on Earth - page 39

In a model by Glade (2008) , the elementary units of the biological micromachine
(cytoskeleton) are individual fibers (microtubules, actin filaments, and intermediate
filaments) whose monomers (molecules of tubulin, actin, etc.) may act as molecular
bits of information. Individual microtubules are connected by microtubule-associated
proteins (MAPs) and other proteins, thus making communication between the fibers
of the cytoskeleton network possible. In the process of their growth and depolym-
erization, microtubules leave chemical trails that enable information to be shared by
neighboring microtubules. As a result, the dynamic skeleton is enabled to perform
functions of a primitive brain and act “as an autonomous system sensible to exter-
nal stimuli, conferring very complex behaviours to cells.” ( Glade, 2008 ). It seems to
instruct the cell to react adaptively to external and internal challenges.
Microtubule functions and structure are closely related to MAPs. Axons and
dendrites of nerve cells are very rich in microtubules. The growth of microtubules
results from the polymerization of tubulin dimmers (α- and β-tubulin molecules) that
is influenced by the biochemical environment, and especially by MAPs bound to
microtubules.
In a model of the regulation of microtubule networks in the cell, MAPs convey
cytoplasmic signals to microtubules, which also receive signals from the whole net-
work of fibrous elements of cytoskeleton. MAPs in turn receive the output of the
processing of cytoplasmic information from microtubules.
The growth of microtubules depends on the microtubule network's MAP-binding
affinity. When the processing of signals in the network is good, MAPs bind to the
microtubules. Increased binding activity in the network increases the stability of the
network and signals that no change is needed. On the contrary, decreased stability of
the network serves as a signal for change in the network structure. The adaptive self-
stabilization ( Figure 1.24 ) is based on a feedback mechanism on the structure of the
Growth dynamics
Signal processing
Adaptive
self-stabilization
Figure 1.24 Cyclic flow of control. The growth dynamics and signal-processing modules
share a common microtubule network representation. The adaptive self-stabilization module
couples the growth dynamics to signal-processing performance ( Pfaffmann and Conrad, 2000 ).