ESI Porter et al. 2019 p.1
Supplemental Information
Interplay of motility and polymer‐driven depletion forces in th
e initial
stages of bacterial aggregation
Michael K. Porter,
a
Asher Preska Steinberg
a
and Rustem F. Ismagilov
a,b*
a.
Division of Chemistry & Chemical Engineering
b.
Division of Biology & Biological Engineering
California Institute of Technolo
gy, 1200 E. California Blvd. Pa
sadena, CA 91125
* rustem.admin@caltech.edu
Supplemental Experimental section
Figures S1‐S2
Contributions of non‐corresponding authors
Electronic
Supplementary
Material
(ESI)
for
Soft
Matter.
This
journal
is
©
The
Royal
Society
of
Chemistry
2019
ESI Porter et al. 2019 p.2
Supplemental Experimental section
Viability of
E. coli
K12 after treatment with sodium azide was tested by plating ce
lls onto lysogeny broth (LB) with 1.5%
agar. Cells were prepared as described in the Experimental sect
ion, and were left to incubate at room temperature for 1
h before plating. Duplicate agar plates were incubated at 35 °C
overnight and then colonies were counted. None of the
cells in the azide‐treated group
survived, whereas the cells wa
shed in motility buffer
maintained viability.
Fig S1. Confirming the deactivation of
Escherichia coli
via plating.
Escherichia coli
washed (a) with sodium azide in motility
buffer and (b) motility buffer (n
o sodium azide) which were the
n plated on LB and 1.5% agar.
ESI Porter et al. 2019 p.3
Fig S2. A comparison of the aggr
egation of motile and nonmotile
E. coli
K12 in a range of concentrations of 10 kDa and
100 kDa PEG.
Volume‐weighted average aggregate sizes (Vol Wt Avg Size) of no
nmotile and motile
E. coli
K12 for serial
dilutions of (a) 10 kDa PEG and (b) 100 kDa PEG. Aggregate size
s were measured 10 min after cells were mixed with PEG
using a cell concentration of
10
ଽ
푐푒푙푙푠/푚퐿
. Volume‐weighted average sizes in terms of bacteria per aggreg
ate (N) are
plotted against polymer mass concentration (
푐
) normalized by overlap concentration (
푐
∗
). Overlap concentration was
estimated to be 85 mg/mL for 10 kDa PEG and 8.5 mg/mL for 100 k
Da PEG. Vertical error bars a
re 95% empirical bootstrap
confidence intervals using the bootstrapping protocol described
in the “Imaging Analysis” section of the Methods in Ref.
1
Data for the 10 kDa PEG concentration were compiled from one bi
ological replicate and data for the 100 kDa PEG
concentration were compiled from two biological replicates (whe
re each replicate is a separate bacterial culture). For each
concentration of PEG, each repli
cate was obtained from one z‐st
ack that was comprised of at least 120 slices.
Aggregation of motile and nonmotile
E. coli
was measured in a serial dilution of 10 kDa and 100 kDa PEG ar
ound
their respective overlap concentrations. In 10 kDa PEG, neither
motile nor nonmotile bacteria aggregated at the
concentrations tested (c/c* = 1/32 to 4). A similar observation
was made using PEG‐coated particles instead of bacteria in
a similar MW PEG solution in our
previous publication (Preska S
teinberg
et al
. 2019).
In the 100 kDa PEG solutions, nonmotile
E. coli
aggregated in a manner qualitatively consistent with polymer‐
driven depletion aggregation, as reported with PEG‐coated parti
cles tested with the same MW PEG in the same
concentration range (c/c* = 1/32 to 4). The nonmotile bacteria
aggregation measured in the 100 kDa PEG is similar but
smaller in magnitude compared to the aggregation measured in th
e 1 MDa PEG, indicative of the smaller depletion
potential exerted from the smaller MW PEG. Minimal aggregation
was measured for the motile
E. coli
in 100 kDa PEG,
suggesting that at this MW and concentration range, the swim fo
rce is strong enough to overcom
e the depletion potential.
ESI Porter et al. 2019 p.4
Contributions of non‐corresponding authors
M.P.
1.
Designed experiments, interpreted results, and analyzed data
2.
Optimized aggregation measuring experiment for bacteria
3.
Performed experiments to generate all data for figures 2, 3, 4,
S1, and S2
4.
Drew schematic in figure 1 and obtained images for figure 1
5.
Performed numerical calculations
for figure 3 using code adapte
d from A.P.S.
6.
Contributed to writing abstract,
introduction, results/discussi
on, and conclusion sections of the manuscript
7.
Contributed to writing of supplemental information
A.P.S.
1.
Contributed to experimental desi
gn, interpretation of results,
and analysis of data.
2.
Contributed to adaptation of imag
e analysis pipeline for use in
this system.
3.
Co‐developed numerical methods a
nd contributed to adaptation of
theory for use in figure 3.
4.
Contributed to writing abstract,
introduction, results/discussi
on, and conclusion sections of the manuscript.
5.
Contributed to writing of supplemental information.