2D and 3D dynamic imaging of live biofilms in a microchannel by
time-of-flight secondary ion mass spectrometry
Xin Hua,1,2 Matthew J. Marshall,3 Yijia Xiong,4 Xiang Ma,5 Yufan Zhou,6
Abigail E. Tucker,3 Zihua Zhu,6 Songqin Liu,1,a) and Xiao-Ying Yu2,a)
1
School of Chemistry and Chemical Engineering, Southeast University, Nanjing, Jiangsu Province, 211189,
PR China
2
Atmospheric Sciences and Global Climate Change Division, Pacific Northwest National Laboratory,
Richland, WA 99354, USA
3
Biological Sciences Division, Pacific Northwest National Laboratory, Richland, WA 99354, USA
4 College
of Osteopathic Medicine of the Pacific-Northwest, Western University of Health Sciences, Lebanon,
OR 97355, USA
5 Material
6
Sciences, Pacific Northwest National Laboratory, Richland, WA 99354, USA
W. R. Wiley Environmental Molecular Science Laboratory, Pacific Northwest National Laboratory,
Richland, WA 99354, USA
a)
Authors to whom correspondence should be addressed. Electronic addresses:
liusq@seu.edu.cn, Tel: (+86) 13851521532, and xiaoying.yu@pnnl.gov, Tel: (+1) 509-3724524.
1
Dynamic imaging of live biofilms by Hua et al.
EXPERIMENTAL DETAILS
Additional experimental details are provided below.
A.
SALVI fabrication
The fabrication and assembly of SALVI were reported in earlier papers1-3 and the
5
schematic of the SALVI and ToF-SIMS setup for imaging biofilms is depicted in SI Figure S1(a). The microfluidic channel dimension was 100 µm wide and 300 µm deep to reduce
potential biofouling. The SALVI fabrication process was described in our earlier papers.2,
4-6
Briefly, soft lithography was employed to make a 100 m wide by 300 m deep channel on a
10
PDMS block. A SiN window consisting of a 100 nm thick SiN membrane and a silicon frame
(frame size 7.5×7.5 mm2, window size 1.5×1.5 mm2) was bonded to the PDMS block by oxygen
plasma treatment. The two pieces are brought to each other by immediate contact to seal the
microchannel and form the detection area.
B.
15
Biofilm growth
A culture of Shewanella oneidensis strain MR-17 was used to initiate biofilm growth.7, 8
Biofilms were adherent to the SiN membrane in the microfluidic channel. Biofilm growth was
confirmed by in situ CLSM imaging of the fluorescence of the constitutively expressed
cytoplasmic green fluorescent protein (GFP) (488 nm excitation; 500-550 nm emission) across
the channel (Figure S-1(b)).7
20
All chemicals used in the chemically defined, modified M1
minimal medium were purchased from Sigma-Adrich Chemical Co. (St. Louis, MO, United
States) unless otherwise noted. The modified M1 medium solution consists of piperazine-N,N’bis(ethanesulfonic acid) (PIPES) buffer (30 mM in the starter culture or 3 mM in the
microfluidic reactor) at pH 7.2, 7.5 mM sodium hydroxide, 28 mM ammonium chloride, 1.34
2
Dynamic imaging of live biofilms by Hua et al.
mM potassium chloride, 4.35 mM monobasic sodium phosphate, 30 mM sodium chloride, 0.68
25
mM calcium chloride, 0.005 mM ferric nitrilotriacetic acid, and 0.001 mM sodium selenate.
Wolfe’s vitamins and minerals solutions were provided as described in Kieft et al.9, and the
amino acids L-glutamic acid, L-arginine, and D,L-serine were supplemented at final
concentrations of 2.0 mgL−1. In the starter culture, 30 mM sodium lactate was added as the
electron donor and atmospheric O2 was the terminal electron acceptor. The microfluidic reactor
30
contained 20 mM sodium lactate and 20 mM sodium fumarate as the electron donor and
acceptor, respectively.
Prior to inoculating the SALVI microfluidic reactor, the reactor was sterilized by flowing
a 70% ethanol solution through the system for a minimum of 3 hr. Filtered sterilized (0.22 µm)
ultrapure water was passed through the system for a minimum of five volume-changes and a
35
sterile medium solution was passed through the system overnight.
To grow the starter culture, 20 mL of modified M1 minimal medium was added to a 60
mL serum bottle and sealed with a thick butyl rubber stopper. The batch starter culture was
grown for 24 hours at 30° C with shaking (150 rpm). Cells were harvested by centrifugation for
10 minutes at 5000 x g at 23° C. The supernatant was decanted and the cell pellet was
40
resuspended in 10 mL of medium optimized for the SALVI microfluidic reactor.
The
resuspended bacterial culture was flown through the microfluidic reactor at 2 µL/min for 3 hr.
Two 10-mL syringes containing sterile growth medium and a drip tube flow break to prevent
back-contamination were aseptically attached to the manifold at the end of inoculation period.
The medium solution was run through SALVI at room temperature for six days at a flow rate of
45
2 µL/min, which was permissive for suboxic bacterial growth. In the microfluidic channel, the
3
Dynamic imaging of live biofilms by Hua et al.
biofilm was adherent to the SiN membrane and biofilm growth was confirmed by in situ CLSM
imaging of GFP (Figure S-1(b)).
C.
Agarose sample preparation
In experiments to determine ToF-SIMS sputtering rate in soft material, agarose was
50
infused in to the microfluidic channel for analysis. The SALVI device was sterilized using
identical conditions to the live biofilm experiments (described in the previous section). One
percent low-melting point agarose (Beckman Industries, Palo Alto, CA) (gelling temperature ≤20
˚C to 30 ˚C) in MR-1 minimal media was liquefied by autoclaving and pushed through SALVI
cell at a flow rate of 2 µL/min by a syringe pump (Harvard Apparatus) while keeping the agarose
55
temperature above 45 ˚C. During agarose infusion, the SALVI device was set up with the SiN
membrane facing down to ensure agarose would remain attached if any shrinking occurred
during agarose cooling and solidification. Once the agarose had made its way through the
device, the union was added to seal the device; and the agarose was allowed to solidify overnight
at room temperature before ToF-SIMS and AFM analysis.
60
D.
ToF-SIMS instrumentation
A ToF-SIMS V spectrometer (IONTOF GmbH, Münster, Germany) was used in this
study. A 25 keV Bi+ beam was used as the primary ion beam in all measurements. During
SIMS measurements, the Bi+ beam was focused to about ~250 nm diameter, and it was scanned
on a round area with a diameter of ~2 µm. To save the punch-through time of the SiN film, a
65
long pulse-width (1000 ns, beam current ~7.7 pA at a repeated frequency of 20 kHz) Bi+ beam
was used for pre-punching for ~115 s, and then the pulse width was reduced to 130 ns for data
collection (beam current ~1.0 pA at a repeated frequency of 20 kHz). Vacuum pressure during
measurements was 2.5-5.5×10-7 mbar in the analysis chamber, indicating that no spraying or fast
4
Dynamic imaging of live biofilms by Hua et al.
spreading of aqueous solutions from the aperture occurred.10 Before each measurement, a 500
70
eV O2+ beam (~40 nA) was scanned on the SiN window with a 400×400 µm2 area for ~30 s to
remove surface contamination. Also, an electron flood gun was used to compensate surface
charging during all measurements. All data were analyzed with IONTOF software (SurfaceLab,
version 6.3). Mass spectra were calibrated using C-, CH-, CH2-, O-, OH- and PO2- peaks,
respectively.11
75
The analytical capability of SALVI enabled liquid ToF-SIMS was detailed in our earlier
paper.12
Potential interference from liquid diffusion after punching-through in SIMS
measurement may not as significant as those cases of aqueous solutions,1, 13 because biofilm is
more similar to a soft solid material than a liquid. However, the biofilm still shows some
mobility, thus direct measurement of the crater depth using AFM after SIMS measurement was
80
not practical. Therefore, agarose was used to mimic the biological active soft material and AFM
measurement of the crater depth was feasible.
E.
AFM instrumentation
AFM imaging was performed in air using an MFP-3D AFM (Asylum Research)
instrument at room temperature. Silicon cantilevers (SSS-NCL, NanoWorld) with a tip radius <5
85
nm and a spring constant of ~48 N/m were used. AFM images were obtained in the tapping
mode using low set point amplitudes (<300 mV) to minimize damage to samples. Images were
analyzed using Igor Pro 6.34 software (WaveMetrics).
FIGURES
90
Additional figures and movies are provided to substantiate the content presented in the
manuscript. SI Fig. S-1 depicts the schematic set up of the SALVI microfluidic cell and the
5
Dynamic imaging of live biofilms by Hua et al.
ToF-SIMS imaging. SI Fig. S-2 shows the ToF-SIMS negative m/z spectrum of hydrated
biofilms. Spectral PCA analysis results from day 6 biofilm using all unit mass peaks ranging
from m/z 1-300 is shown in SI Fig. S-3. Results are in agreement with that using only a selected
95
range of m/z peaks that are considered to be main characteristic fragments relevant to
Shewanella biofilms. SI Fig. S-4 shows additional 2D images of fatty acid fragments. SI Fig. S5 shows additional reconstructed 3D images at day 5. SI Fig. S-6 shows AFM results of the
SIMS ionization crater obtained from the agarose sample.
Three movies showing the reconstructed 3D biofilm fatty acid components, C12 FA, C15
100
FA, and merged C12 FA and C15 FA, are also provided for ease of visualization.
6
Dynamic imaging of live biofilms by Hua et al.
SI Figure S-1
SI FIG. S-1. (a) Schematic illustration of biofilm imaging by SALVI and liquid ToF-SIMS.
105
Inset: a photo of SALVI. (b) Composite CLSM images showing time-dependent biofilm growth
(green) in the microfluidic channel. Channel width is 100 µm and the direction of fluid flow is
indicated in the black arrow at the bottom.
7
Dynamic imaging of live biofilms by Hua et al.
110
SI Figure S-2
SI Figure S-2. ToF-SIMS m/z spectra of the live Shewanella sp. biofilm in SALVI.
8
Dynamic imaging of live biofilms by Hua et al.
SI Figure S-3
115
SI Figure S-3. Spectra PCA score and loading plots of day 6 biofilm showing the differences
and similarities among the five regions using unit mass peaks ranging from 1-300 m/z. A 95%
confidence limit for each region was defined by an ellipse with the same color to the
corresponding region clusters. Five regions representing sample before SiN punch-through (I)
120
during punch-through (II) or within the biofilm region (III, IV, V) are illustrated.
9
Dynamic imaging of live biofilms by Hua et al.
SI Figure S-4
SI Figure S-4. C13, C14 and C16 2D false color images among five selected regions along the
125
depth profile time series in day 6 biofilm. Five regions representing sample before SiN punchthrough (I) during punch-through (II) or within the biofilm region (III, IV, V) are illustrated.
10
Dynamic imaging of live biofilms by Hua et al.
The analysis results from day 5 and day 6 are similar, because ToF-SIMS probes the
130
interface of the biofilm and attachment surface. We only chose to show data from day 6 in the
main text due to space constraints.
SI Figure S-5
SI Figure S-5. Reconstructed 3D biofilm images showing FA distributions within the entire
135
biofilm region (III to V, 302 seconds) at day 5. The time axis represents depth profiling from
near the SiN surface into the biofilm.
11
Dynamic imaging of live biofilms by Hua et al.
SI Figure S-6
(a)
140
(b)
SI Figure S-6. AFM measurements of an agarose sample showing the shape (a) and size/depth
(b) of the hole drilled by ToF-SIMS on the SiN membrane. The top diameter of the hole is 2.134
m, the depth is 285.2 nm.
This indicates that the 100 nm thick SiN membrane was drilled through; and the depth
profiling data in Figure 2 reflects the biofilm attached to the SiN membrane. The exact depth of
145
biofilm ionization by ToF-SIMS could not be directly measured using real-time correlative
atomic force microscopy (AFM) because the biofilm was hydrated and viable and its fluid
properties would recover itself after ionization ceased.
12
Dynamic imaging of live biofilms by Hua et al.
MOVIE CAPTIONS
150
SI Movie S-1. Reconstructed 3D biofilm image showing C12 FA distributions within the entire
biofilm region (III to V, 302 seconds) at day 6.
SI Movie S-2. Reconstructed 3D biofilm image showing C15 FA distributions within the entire
biofilm region (III to V, 302 seconds) at day 6.
SI Movie S-3. Reconstructed 3D biofilm image showing merged distributions of C12 and C15
155
FAs within the entire biofilm region (III to V, 302 seconds) at day 6.
REFERENCES
160
165
170
175
180
1.
L. Yang, Y. X.-Y., Z. H. Zhu, T. Thevuthasan and J. P. Cowin, J. Vac. Sci. Technol. A 29 (6),
article no. 061101 (2011).
2.
L. Yang, X. Y. Yu, Z. H. Zhu, M. J. Iedema and J. P. Cowin, lab chip 11 (15), 2481-2484 (2011).
3.
X. Y. Yu, B. W. Liu and L. Yang, Microfluid Nanofluid 15 (6), 725-744 (2013).
4.
X. Hua, X. Y. Yu, Z. Y. Wang, L. Yang, B. W. Liu, Z. H. Zhu, A. E. Tucker, W. B. Chrisler, E.
A. Hill, T. Thevuthasan, Y. H. Lin, S. Q. Liu and M. J. Marshall, Analyst 139 (7), 1609-1613 (2014).
5.
L. Yang, X. Y. Yu, Z. H. Zhu, T. Thevuthasan and J. P. Cowin, J Vac Sci Technol A 29 (6)
(2011).
6.
X.-Y. Yu, L. Yang, J. Cowin, M. Iedema and Z. Zhu, USA Patent No. 8,555,710 (Oct. 15., 2013
2013).
7.
J. S. McLean, G. E. Pinchuk, O. V. Geydebrekht, C. L. Bilskis, B. A. Zakrajsek, E. A. Hill, D. A.
Saffarini, M. F. Romine, Y. A. Gorby, J. K. Fredrickson and A. S. Beliaev, Environ Microbiol 10 (7),
1861-1876 (2008).
8.
X. Hua, X.-Y. Yu, Z. Wang, L. Yang, L. Bingwen, Z. Zhu, A. E. Tucker, W. B. Chrisler, E. A.
Hill, T. Thevuthasan, Y. Lin, S. Liu and M. J. Marshall, Analyst 139 (7), 1609-1613 (2014).
9.
T. L. Kieft, J. K. Fredrickson, T. C. Onstott, Y. A. Gorby, H. M. Kostandarithes, T. J. Bailey, D.
W. Kennedy, S. W. Li, A. E. Plymale, C. M. Spadoni and M. S. Gray, Appl. Environ. Microbiol. 65 (3),
1214-1221 (1999).
10.
L. Yang, X.-Y. Yu, Z. Zhu, M. J. Iedema and J. P. Cowin, Lab Chip 11 (15), 2481-2484 (2011).
11.
F. M. Green, I. S. Gilmore and M. P. Seah, Journal of the American Society for Mass
Spectrometry 17 (4), 514-523 (2006).
12.
L. Yang, Z. Zhu, X.-Y. Yu, E. Rodek, L. Saraf, T. Thevuthasan and J. P. Cowin, Surface
Interface Analysis (46), 224-228 (2014).
13.
B. Liu, X.-Y. Yu, Z. Zhu, X. Hua, L. Yang and Z. Wang, Lab Chip 14 (5), 855-859 (2014).
13