Home
Search
Collections
Journals
About
Contact us
My IOPscience
Prokaryotic transport in electrohydrodynamic structures
This article has been downloaded from IOPscience. Please scroll down to see the full text article. 2012 Phys. Biol. 9 026006 (http://iopscience.iop.org/1478-3975/9/2/026006) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 128.103.149.52 The article was downloaded on 09/06/2012 at 13:01
Please note that terms and conditions apply.
IOP PUBLISHING
PHYSICAL BIOLOGY
doi:10.1088/1478-3975/9/2/026006
Phys. Biol. 9 (2012) 026006 (11pp)
Prokaryotic transport in electrohydrodynamic structures A H Paulitsch-Fuchs 1 , E C Fuchs 1 , A D Wexler 1 , F T Freund 2 , L J Rothschild 3 , A Cherukupally 4 and G J W Euverink 5 1 2 3 4 5
Wetsus, Centre of Excellence for Sustainable Water Technology, Leeuwarden, The Netherlands SETI Institute, Carl Sagan Center, Mountain View, CA, USA NASA Ames Research Center, Moffett Field, Mountain View, CA, USA University of Arizona, Department of Mining and Geological Engineering, Tucson, AZ, USA Institute for Technology and Management (ITM), University of Groningen, The Netherlands
E-mail:
[email protected]
Received 7 October 2011 Accepted for publication 22 February 2012 Published 4 April 2012 Online at stacks.iop.org/PhysBio/9/026006 Abstract When a high-voltage direct-current is applied to two beakers filled with water, a horizontal electrohydrodynamic (EHD) bridge forms between the two beakers. In this work we study the transport and behavior of bacterial cells added to an EHD bridge set-up. Organisms were added to one or to both beakers, and the transport of the cells through the bridge was monitored using optical and microbiological techniques. It is shown that Escherichia coli top10 (Invitrogen, Carlsbad, CA, USA) and bioluminescent E. coli YMC10 with a plasmid (pJE202) containing Vibrio fischeri genes can survive the exposure to an EHD liquid bridge set-up and the cells are drawn toward the anode due to their negative surface charge. Dielectrophoresis and hydrostatic forces are likely to be the cause for their transport in the opposite direction which was observed as well, but to a much lesser extent. Most E. coli YMC10 bacteria which passed the EHD bridge exhibited increased luminescent activity after 24 h. This can be explained by two likely mechanisms: nutrient limitation in the heavier inoculated vials and a ‘survival of the strongest’ mechanism.
1. Introduction
behavior in a reduced gravity environment (Fuchs et al 2008, 2009, 2010a, 2010c, 2011, Woisetschl¨ager et al 2010, Del Guidice et al 2010, Fuchs 2010b). The macroscopic stability of the water bridge has been explained in the framework of electrohydrodynamic (EHD) theory (Widom et al 2009, Marin and Lohse 2010, Aerov 2010) from the balance between the induced polarization forces at the surface, capillary forces and gravity. According to these findings, the most important properties necessary for liquid bridge formation are high dielectric permittivity, low electric conductivity and a permanent molecular dipole moment. Thus the phenomenon is not water specific, but can be obtained with any liquid of similar properties like methanol (Fuchs 2010b) or glycerol (Marin and Lohse 2010). A recent publication (Eisenhut et al 2011) reported the transport of phenol and ethylene glycol solutions in both directions through the water bridge. Woisetschl¨ager et al (2010) investigated the motion of microscopic tracer particles
In 1893 Sir William Armstrong (Armstrong 1893) placed a cotton thread between two wine glasses filled with chemically pure water. After applying a high voltage, a watery connection formed between the two glasses, and after some time, the cotton thread was pulled into one of the glasses, leaving, for a few seconds, a rope of water suspended between the lips of the two glasses. Over the next century scientists studied related effects, like electrowetting (Mugele and Baret 2005) or the Sumoto effect (Sumoto 1956), but the water bridge itself was not investigated until its recent rediscovery (Fuchs et al 2007). The presence of high electric fields (few kV/cm) and instabilities caused by the experimental probes make an experimental investigation challenging. At present, reports discuss the macroscopic properties of the water bridge like its mass and charge transport, its density and temperature gradients, the potential presence of microdomains and its 1478-3975/12/026006+11$33.00
1
© 2012 IOP Publishing Ltd
Printed in the UK & the USA
Phys. Biol. 9 (2012) 026006
A H Paulitsch-Fuchs et al
and found a bidirectional transport as well. In this work we study the transport and behavior of bacterial cells added to a floating water bridge set-up. Organisms were added to one or to both beakers, and the transport of the cells through the water bridge was monitored using optical and microbiological techniques. In an investigation of the possible effects of a dc electric field on growth processes of E. coli, the presence of certain group VIIIb transition metal compounds in concentrations of about 1–10 parts per million of the metal in the culture medium was found to cause an inhibition of the cell division process (Rosenberg et al 1965), whereas no effect of the electric field itself could be found. However, E. coli strains carry a negative net-charge on their cell walls (Dickson and Koohmaraie 1989, Li and McLandsborough 1999, Ukuku and Fett 2002); thus, from a physical point of view they promised to be interesting candidates for the presented experiments. There are several research areas dealing with the interaction of living cells with electric fields including, e.g., the following:
Table 1. Time, voltage and current of the experiments performed at Wetsus. Series 1 2 3 4
Inoculated beaker
min
kV
mA
Medium
Both Anode Cathode Both Anode Cathode Both Anode Cathode Both Anode Cathode
2 2 2 3 3 3 3 3 3 6 6 6
14.6 14.3 15.0 10.0 10.0 10.0 10.1 10.1 10.1 16.25 16.25 16.25
2.3–4.2 0.65–0.85 1–1.25 0.2 0.16 0.2 0.16 0.14 0.15 0.6–1.6 0.6–1 0.7–1.6
Glucose Glucose Glucose Glycerol Glycerol Glycerol Glycerol Glycerol Glycerol Glycerol Glycerol Glycerol
After the experiment 4 mL of the solutions from each beaker and the blank were added to 15 mL Greiner tubes containing 4 mL of 2xLB-medium and 1 mL of sterile filtered media of the overnight culture containing the autoinducer necessary for the fluorescence of the E. coli YMC10 strain. The fluorescence of the bacteria was measured with a Victor spectrophotometer (Perkin–Elmer, Waltham, MA, USA) in black 96 well plates (200 μL sample volume per well) resulting in absolute numbers in CPS (counts per second). This measurement was repeated one day after the experiment and once every day up to a total of three days. Additionally dilutions of the solution were plated for the count of colony forming units (CFU) directly after the experiments. Dilutions were made in phosphate buffered solution (PBS) and colonies were counted after 24–48 h at 30 ◦ C. The experiments were performed with three different configurations:
(1) Electrophoresis and dielectrophoresis for the means of manipulating and/or sorting cells in microfluidic devices which have been studied quite extensively during the last decades (e.g. reviews on this topic: Andersson and van der Berg 2003, Yi et al 2006, Tsutsui and Ho 2009, Gossett et al 2010). (2) Electroporation of cells both in microfluidic devices and for cloning purposes (e.g. Calvin and Hanawalt 1988, Chen et al 2006, Fox et al 2006, Krassowska Neu and Neu 2009). (3) High-voltage disinfection of liquids and (liquid) food using pulsed electric fields (ac or dc). (e.g. Johnstone and Bodger 1997, Mosqueda-Melgar et al 2008, Gusbeth et al 2009, Hwang et al 2010).
(a) bacterial solution in both beakers (66 g of bacterial solution in each beaker); (b) bacterial solution in the anode beaker only, MilliQ water was used for the other beaker (66 g of bacterial solution/anode beaker, 66 g of MilliQ water in the cathode beaker); (c) bacterial solution in the cathode beaker only, MilliQ water was used for the other beaker (66 g of bacterial solution/cathode beaker, 66 g of MilliQ water in the anode beaker).
The extensive literature in these fields does however not include EHD liquid bridging; therefore (to the best knowledge of the authors), the present study is the first on the behavior of living cells in an EHD liquid bridge system.
2. Microbiological methods and experimental set-up Escherichia coli (YMC10) with a plasmid (pJE202) containing Vibrio fischeri genes specifying the luminescence enzymes (luxICDABEG operon consisting of the luxI, luxC, luxD, luxA, luxB, luxE and luxG genes) and encoding regulatory functions for luminescence (Engelbrecht et al 1983) was used as the organism for the experiments carried out at Wetsus. The cells of an overnight culture (30 ◦ C) in LB medium (10 g L−1 tryptone, 5 g L−1 yeast extract, 10 g L−1 NaCl; pH7) were harvested and washed in MilliQ water containing 2.5% glucose or 5% glycerol (see table 1). After washing, the cell density was adjusted to 1–2 · 108 cells mL−1. Just before the experiment, the stock was diluted 1:20 with MilliQ water to reduce the conductivity of the solution. Cell solutions and the MilliQ water used for diluting were kept on ice until the experiment with the exception of the experiments of series 1. The blank was also taken from the cell solution prepared for the experiment.
To understand the correlation of cell number, OD and CPS dilution series of E. coli YMC10 in LB starting with 1 · 107 cells mL−1 (10 dilutions, 1:10) have been prepared and inoculated for three days, OD 490 and CPS have been measured at 0, 24 and 48 h. To estimate the influence of destroyed cells on the conductivity of solutions E. coli YMC10 cells in MilliQ water (1 × 108 cells mL−1 and 1 · 107 cells mL−1 respectively) were treated in the microwave (700 W, 20 s). After the treatment, the solutions were allowed to cool down to room temperature. The conductivity was measured before the treatment, 1 h and 24 h after the treatment. The solutions were checked microscopically for intact cells, and after the microwave treatment no more active cells could be seen. 2
Phys. Biol. 9 (2012) 026006
A H Paulitsch-Fuchs et al
Table 2. Conductivity and temperature measured during the experiments. ◦
μS cm−1
C
Inoculated
Anode beaker
Cathode beaker
Anode beaker
Cathode beaker
Series
beaker
Before
After
Before
After
Before
After
Before
After
1
Both Anode Cathode Both Anode Cathode Both Anode Cathode Both Anode Cathode
n/ a n/ a n/a 0.79 0.88 0.88 1.18 1.06 1.00 1.42 1.29 0.88
n/a n/a n/a 1.24 1.19 1.11 1.39 1.23 1.13 2.77 2.02 1.85
n/a n/a n/a 0.79 0.88 1.07 1.17 0.90 1.15 1.38 0.61 1.46
n/a n/a n/a 0.94 0.75 1.04 1.05 0.84 1.10 1.47 0.89 1.46
n/a n/a n/a 10.0 10.7 7.9 17.3 14.1 10.9 24.4 24.5 23.8
21.4 22.2 25.9 19.1 16.2 13.1 19.9 17.2 14.6 32.8 30.9 31.1
n/a n/a n/a 10.0 7.9 9.7 16.8 15.1 13.6 24.5 25.0 23.8
21.0 23.8 22.6 19.0 14.4 14.0 19.3 16.3 17.2 33.0 30.9 31.4
2 3 4
3. Results
Due to the astonishing outcome of the experiments and in order to check whether the results obtained at Wetsus are reproducible with different bacteria in a different location, the authors chose to reassess their results in a renowned biological lab. Thus a series of comparable experiments was carried out at the NASA Ames lab in California using E. coli top10 (Invitrogen, Carlsbad, CA, USA). Cells of an overnight culture (37 ◦ C) in LB medium were harvested and washed in 5% glycerol solution. A cell density of 0.15 at OD 490 was measured for the stock solution of the cells in 5% glycerol. Cell solutions and the double distilled water used for diluting were kept on ice until the experiment. After the experiment 5 mL of the solutions from each beaker and the blank were added to 15 mL Greiner tubes containing 5 mL of 2xLB-medium. For the blank for each experiment 32 mL of water were mixed with 0.5 mL of bacterial stock solution. The OD 490 of the bacteria was measured with a spectrophotometer in transparent 96 well plates (200 μL sample volume per well). This measurement was repeated one day after the experiment. Additionally dilutions of the solution were plated for the count of CFU directly after the experiments. Dilutions were made in phosphate buffered solution (PBS) and colonies were counted after 24–48 h at 37 ◦ C. The experiments were performed with three different configurations:
3.1. Experiments at Wetsus Run duration, voltage and current were measured during bridge operation (table 1), conductivity and temperature of the bacterial solutions and the water were recorded before and after the experiment with the exception of series 1 (table 2). A comparison of voltage and current of the four series of experiments shows a different behavior of series 1 and series 4 (see table 1) compared to series 2 and 3. In series 1 the current was significantly higher than in series 2 and 3; and it increased steadily. Electrochemical reactions involving glucose in the solution during bridge operation required a greater input of electrical energy and resulted in oscillations in the drive current. For this reason glucose was replaced in the following series by glycerol, resulting in more stable conditions. The higher voltage and current flow in series 4 can be explained by the increased conductivity (see table 2) of the bacterial solution in the beginning of the experiments. Conductivity measurements of the solutions before and after the bridge runs showed an increase in conductivity in the anode beaker for all experiments in series 2–4. The conductivity in the cathode beaker after the experiments was found to be either slightly higher or lower when compared to the starting conductivity. Temperature in anode and cathode beakers increased in all experiments of series 2–4. The temperature increases were between 1.2 ◦ C and 10 ◦ C in the different experiments. Within one experiment the lowest increase difference between anode and cathode was 0.1 ◦ C (experiments 2a, 3a and 3c) and the highest difference was 3 ◦ C (experiment III, NASA). Representative results of the CFU/mL measurements and the CPS counts are shown in figures 1 and 2. For all experiments where the bacteria had been added to both beakers (1a, 2a, 3a and 4a) the data show an increase of the number of bacteria in the anode beaker compared to the cathode beaker (in CFU/mL and CPS measurements). When adding the bacteria to the anode beaker (1b, 2b, 3b, 4b) almost no transport to the cathode beaker could be measured (CFU/mL and CPS). In the cases where the bacteria were only present in the cathode
(I) bacterial solution in both beakers (64 mL water +0.5 mL bacterial stock solution/beaker); (II) bacterial solution in the anode beaker only, double distilled water was used for the other beaker (64 mL water +1 mL of bacterial solution/anode beaker, 64 mL of water in the cathode beaker); (III) bacterial solution in the cathode beaker only, water was used for the other beaker (64 mL water +1 mL of bacterial solution/cathode beaker, 64 mL of water in the anode beaker). 3
Phys. Biol. 9 (2012) 026006
A H Paulitsch-Fuchs et al
Figure 3. CPS values (blank, anode and cathode) after 0, 24, 48 and 72 h for series 4, experiment a, both beakers contained bacteria, bridge operated for 6 min.
Figure 1. CFU/mL of the experiments from series 1, experiments a–c, bridge duration 2 min. The titles along the abscissa indicate from which beaker the culture inoculum was taken: A—anode beaker, C—cathode beaker, blank—diluted directly from stock culture.
Figure 4. CPS values (blank, anode and cathode) after 0, 24, 48 and 72 h for series 4, experiment b, bacteria in anode beaker only, bridge operated for 6 min. Figure 2. CPS of the experiments from series 2, experiments a–c, bridge duration 3 min. The titles along the abscissa indicate from which beaker the culture inoculum was taken: A—anode beaker, C—cathode beaker, blank—diluted directly from stock culture.
beaker at the start of the experiment (1c, 2c, 3c, 4c) transport from the cathode to the anode beaker was measured (CFU/mL and CPS). The CPS measurements have been repeated for each sample of the series 2–4 after 24, 48 and 72 h. Representative examples for each experiment are shown in figures 3–5. Although the signal for the cathode beaker samples was the lowest in all a- and b-configuration experiments from all series (1–4) directly after the experiments the CPS signals after 24 h reached the same or higher level as those samples from the anode beaker or blank, respectively. The c-configuration experiments showed approximately the same increase for the anode beaker bacteria after 24 h of incubation. Time-dependent development in dilution series of E. coli YMC10 in LB were made and measured with CPS and OD 490 to be able to compare the results from Wetsus and NASA Ames. Figure 6 shows the results of CPS measurements and figure 7 gives the results for the OD measurements. Due to the smaller population of bacteria in higher dilutions, the CPS
Figure 5. CPS values (blank, anode and cathode) after 0, 24, 48 and 72 h for series 3, experiment c, bacteria in cathode beaker only, bridge operated for 3 min.
counts of those dilutions are peaking later (on day 3 or 4) compared to the CPS counts of the lower dilutions which are showing the same behavior as the dilutions in our experiments. The OD measurements show a clear development to higher optical densities from day 1 to day 4 for all the dilutions, 4
Phys. Biol. 9 (2012) 026006
A H Paulitsch-Fuchs et al
Figure 6. CPS values after 0, 24, 48 and 72 h for a 1:10 dilution series of E. coli YMC10 in LB media. The color bars in each category progress from vibrant color at 0 h to faded color at 72 h.
Figure 7. OD490 values after 0, 24, 48 and 72 h for a 1:10 dilution series of E. coli YMC10 in LB media. The color bars in each category progress from vibrant color at 0 h to faded color at 72 h.
with the higher dilutions showing a delayed increase in optical densities. The major difference between CPS and OD values is the fact that CPS values show a peak and then a decrease whilst the OD values are continuously increasing. As shown in table 3 the destruction of the cells gives an increase in conductivities between ∼21% (at a concentration of 1 · 108 cells mL−1) and ∼35% (at a concentration of 1 · 107 cells mL−1), whilst the MilliQ water shows a slight increase only after 24 h.
Table 3. Conductivities of cell solutions before the treatment, 1 h and 24 h after the microwave treatment (700 W, 20 s).
5
cells mL−1
μS cm−1 (before)
μS cm−1 (1 h after)
μS cm−1 (24 h after)
0 (MilliQ) 1 · 107 1 · 108
0.75 1.11 11.9
0.75 1.5 14.39
0.8 1.52 14.39
Phys. Biol. 9 (2012) 026006
A H Paulitsch-Fuchs et al
Figure 8. Colony formation from the NASA experiments after 24 h of incubation. CFU/mL values given for cultures inoculated from A—anode beaker, C—cathode beaker, and blank—stock culture dilution without bridge exposure.
Figure 9. OD 490 measurements from the NASA experiments after 24 h of incubation. Values given for cultures inoculated from A—anode beaker, C—cathode beaker, and blank—stock culture dilution without bridge exposure.
Table 4. Time, voltage and current of the experiments performed at NASA Ames. Experiment
min
kV (begin)
kV (end)
mA
I II III
6.5 6.5 6.5
13.3 14.3 14.1
N/A 15.8 15.5
0.4 0.6–0.2 0.8–1
3.3. Visualization In order to visualize the transport through the bridge an additional run with a high bacterial load (1–2 · 108 cells mL−1) was performed. Bacteria were added to both beakers and the bridge run was monitored using a Panasonic HDC SD-600 HDTV video camera. Figure 10 shows the timedependent transport of the bacterial cells from the anode to the cathode beaker. The transport from the cathode to the anode side can be clearly seen due to the development of a clear (darker) zone starting to appear in the top layer of the cathode beaker and proceeding to grow downwards with ongoing time. Consequently the density of the bacterial solution in the anode beaker becomes higher (the turbidity increases) which can be seen on the pictures as a progression in brightness.
Table 5. Temperature measured during the experiments. ◦
C
Anode beaker
Cathode beaker
Experiment
Before
After
Before
After
I II III
22 20 19
27.5 25 26
22 20 19
25.5 24 29
3.4. Electric field 3.2. Experiments at NASA Ames
Figure 11 shows the result of a numerical simulation of the electric field within the bridge using the electrostatics package of the ac/dc module in Comsol 4.2a multiphysics software (Comsol Inc., Palo Alto, CA), rendered as wire frame. The setup was that for a stable bridge under equilibrium conditions, the free floating portion of the bridge was approximated by a cylinder intersecting with spheres (at the beaker rims) and frustums (water connection between beaker rim and water surface). The triangular grid in front and in the back is the wire frame marking the boundaries of the 50 cm spherical air bubble within which the set-up was placed. The color code shows low field strengths within the beakers (
