BioMEMS for Cell Biology - Introduction to BioMEMS - page 351

(a) (b)

(c) (d)

FIGURE 6.57 Directed. neurite. outgrowth. on. aligned. Schwann. cell. monolayers.. The. images. depict.
neurite.outgrowth.on.(a).aligned.Schwann.cell.monolayer,.(b).micropatterned.laminin,.(c).unaligned.
Schwann.cell.monolayer,.and.(d).uniform.laminin..Scale.bar.is.50.μm..(From.Deanna.M..Thompson.and.
Helen.M..Buettner,.“Neurite.outgrowth.is.directed.by.Schwann.cell.alignment.in.the.absence.of.other.
guidance.cues,”. Ann. Biomed. Eng. .34,.161-168,.2006..Reprinted.with.permission.from.Springer.)

6.5.2 Neuronal Polarization
In vivo, neurons grow their axons in prespeciied directions because of the presence of gradient
signals that “polarize” the neurons and determine which of the nascent processes become an
axon. On traditional cell culture substrates, however, these directional signals are absent and
the axon is determined at random. his limitation is a serious obstacle for the development of
“neuronal networks on a chip,” which aim to reproduce and study some of the characteristics of
real networks on a reduced scale.
In 1998, a large collaborative group led by Carl Cotman at the University of California at
Irvine and David Stenger at the Naval Research Laboratory was able to bias the polarity of embry-
onic hippocampal neurons in a speciied direction given by a micropattern of aminosilane lines
on a luoroalkylsilane background ( Figure 6.58 ). Each somata, adhered to a 25-μm-diameter
aminosilane island, was “ofered” four 5-μm-wide aminosilane paths along which the cell was
able to extend processes; however, three of the aminosilane paths were broken in 10-μm-long
segments, each separated by 10-μm-long luoroalkylsilane spacings, whereas the remaining
aminosilane path connected to the soma was continuous. Approximately 76% of the island-
conined cells developed a process that was 100 μm or longer (along the continuous aminosilane
line) which was identiied as an axon, whereas the other processes were identiied as dendrites.
Identiication was possible by immunostaining of the microtubule-associated proteins MAP2
(depleted in developing axons) and MAP5 (concentrated in axons undergoing rapid elongation)
and of the neuroilament polypeptide NF150 (localized in axons late in postnatal development).
Despite the fact that the surface patterning might be considered “outdated,” this early work
remains an outstanding example of micrometer-scale engineering and control of cell function.
Recently, a team led by Sarah Heilshorn at Stanford University and Luke Lee at the University
of California (Berkeley) designed a microluidic device that presents hippocampal neurons with
biochemically diferent attachment sites and axonal guidance patterns, which direct neuronal
polarization ( Figure 6.59 ). To facilitate the illing of the intricate microluidic networks with