Biomedical Engineering Reference
In-Depth Information
swimming speed (~30 μm/s in
E. coli
) thus converts the spatial gradient into a temporal one and
determines the steepness of the gradients that can be sensed.
he irst bacterial chemotaxis assay on a chip is credited to Michael Manson's group from
Texas A&M University, who in 2003 designed a simple T-mixer (with an extra central inlet
for the bacteria) to measure
E. coli
motility under gradients of various chemoefectors (
Figure
6.32
). As in any T-mixer (see Section 3.9.1), the chemoefector gradient forms gradually down
the channel as the streams mix by difusion (
Figure 6.32a
). he 22 outlets allow for counting
cells and obtaining transverse distribution of bacteria across the chamber. For example, it was
found that wild-type
E. coli
responded to 3.2 nM gradients of l-aspartate (cells accumulate max-
imally in outlet 11, where the concentration is maximal), a concentration three orders of magni-
tude lower than the detection limit in the standard capillary assay (
Figure 6.32b
). he response
to bufer (0 nM) is not a Gaussian distribution because (a) the initial distribution of cells is a
151-μm-wide stream (not a point), (b) the cells low at diferent speeds (due to the parabolic low
proile, see Section 3.2.4), and (c) there is a distribution of swim speeds (given by
E. coli
's difer-
ent run/tumble biases). Surprisingly, l-leucine was sensed (by the receptor
Tar
) as an attractant
at low concentrations and (by
Tsr
) as a repellent at higher concentrations (
Figure 6.32c
).
Obviously, the mixer part of the gradient generator does not need to be a T-mixer, which is limited
to very simple monotonic gradients. Michael Manson's group has collaborated with Arul Jayaraman's
group, also at Texas A&M, to incorporate a Dertinger mixer (see Section 3.9.2) into Manson's low-
based bacterial chemotaxis device (
Figure 6.33
). his improvement has allowed for the investiga-
tion of
E. coli
chemoattraction to the quorum-sensing molecule autoinducer-2, chemorepulsion by
indole, as well as competing simultaneous gradients of both. (By using cells that express GFP or RFP,
the cells could be counted by luorescence instead of by collection into outlet channels.)
1
2
3
a
4
Inlet for
chemoeffectors
18 mm
21 mm
Inlet for
bacteria
Inlet for
buffer
Wild type
0.3
b
c
Chemotaxis towards
L
-aspartate
0.2
300
0.1
1 nM
0
Serine-blind
200
10 nM
-0.1
100
-0.2
Aspartate-blind
Chemotaxis
towards
L
-leucine
3.2 nM
-0.3
0
0 nM
-0.4
-0.5
-100
Channel number
Log concentration
FIGURE 6.32
First.bacterial.chemotaxis.assay.on.a.chip..(From.H..B..Mao,.P..S..Cremer.and.M..
D..Manson,.“A.sensitive,.versatile.microluidic.assay.for.bacterial.chemotaxis,”.
Proc. Natl. Acad.
Sci. U. S. A.
.100,.5449-5454,.2003..Copyright.(2003).National.Academy.of.Sciences,.U..S..A.)
