Robustness to Sub-optimal Temperatures of the Processes of Tsr
Cluster Formation and Positioning in Escherichia Coli
Teppo Annila, Ramakanth Neeli-Venkata and Andre S. Ribeiro
Department of Signal Processing, Tampere University of Technology, Korkeakoulunkatu 10, Tampere, Finland
Keywords: Protein Cluster, Tsr Proteins, Cluster Localization, Microscopy.
Abstract: Clustering and positioning of chemotaxis-associated proteins are believed to be essential steps for their
proper functioning. We investigate the robustness of these processes to sub-optimal temperatures by
studying the size and location of clusters of Tsr-Venus proteins in live cells. We find that the degree of
clustering of Tsr proteins is maximal under optimal temperature. The data further suggests that the
weakening of the clustering process in lower-than and higher-than optimal temperatures is not due to the
same cause. Meanwhile, the location of the clusters is found to be weakly temperature independent, within
the range tested. We conclude that while the clustering of Tsr is heavily temperature dependent, the
localization is only weakly dependent, suggesting that the functionality of the proteins responsible for
retaining Tsr-clusters at the cell poles, such as the Tol-Pal complex, is robust to suboptimal temperatures.
1 INTRODUCTION
Escherichia coli have evolved mechanisms to
respond to external stimuli, such as chemotaxis. This
process is based on clusters of chemoreceptors
(Sourjik and Berg, 2004; Wadhams and Armitage,
2004; Parkinson et al., 2005) that can perform
multiple tasks, including thermosensing (Lee et al.
1988) and aerotaxis (Rebbapragada et al. 1997).
One identified transmembrane chemotaxis
receptor protein is Tsr, a cytoplasmic double
membrane-spanning serine receptor (Lee et al.,
1988) that preferentially accumulates at the cell
poles, where it forms clusters (Thiem et al., 2007).
Clustering and positioning at the poles of the
chemotaxis-associated protein clusters are believed
to be essential for their proper functioning, namely,
they are expected to affect the signal processing
capabilities of the receptor system (Kentner and
Sourjik, 2006; Vaknin and Berg, 2006; Skidmore et
al., 2000; Lybarger and Maddock, 1999), even
though changes in receptor activity do not
necessarily require changes in the assembly process
of the clusters (Liberman et al., 2004).
Recent evidence suggests that the retention of the
clusters at the poles is made possible by proteins
such as the trans-envelope Tol-Pal complex, a
widely conserved component of the cell envelope of
Gram-negative bacteria (Santos et al., 2014). Little
is known about how robust this process to sub-
optimal conditions is.
Here, we use the Tsr-Venus construct (Yu et al.,
2006) to study, in live E. coli cells, the robustness of
the clustering process and spatial distributing of Tsr
clusters to sub-optimal temperatures. We chose this
construct since the tagging of Tsr with Venus does
not affect its spatial distribution, due to a linker
sequence that allows the cytoplasmic domain of Tsr
to freely interact with other signalling proteins,
similar to the natural system (Yu et al., 2006).
Meanwhile, the Venus protein is a YFP variant,
derived from GFP, that has a fast maturation time
(Nagai et al., 2002), allowing real time imaging by
fluorescence microscopy.
2 MATERIALS AND METHODS
2.1 Chemicals
For routine cultures, the M9 glucose media
components, isopropyl β-D-1-thiogalactopyranoside
(IPTG), agarose for microscopic slide gel
preparation and antibiotics were purchased from
Sigma-Aldrich. The amino acids and vitamins were
purchased from ThermoFisher.
Annila, T., Neeli-Venkata, R. and Ribeiro, A.
Robustness to Sub-optimal Temperatures of the Processes of Tsr Cluster Formation and Positioning in Escherichia Coli.
DOI: 10.5220/0005647001370141
In Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2016) - Volume 3: BIOINFORMATICS, pages 137-141
ISBN: 978-989-758-170-0
Copyright
c
2016 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
137
2.2 Bacterial Strain and Growth
Conditions
We used E. coli K-12 strain SX4, harbouring the
Tsr-Venus gene construct under the control of the
lac promoter (P
lac
) (Yu et al., 2006), a kind gift from
Sunny Xie, Harvard University, U.S.A. Overnight
liquid cultures were grown in M9 glucose (0.4%)
media, supplemented with amino acids along with
the appropriate antibiotics for 14 h, at 37°C with
shaking (250 rpm).
2.3 Induction of Tsr-Venus Expression
From overnight cultures, cells were inoculated into a
fresh media with the same antibiotics as above, at an
initial OD of 0.05 at 600nm, and grown until the
OD600 reached ~ 0.3, at 37°C with shaking (250
rpm). The induction of Tsr-Venus expression was
performed by adding 200µM IPTG to the culture,
which was left in the incubator at appropriate
temperature (10°C, 15°C, 24°C, 37°C and 43°C) for
1 hour before imaging.
2.4 Microscopy
To study the formation and localization of Tsr-
Venus clusters, 8 µL of cells left at appropriate
temperature were then placed on 1% agarose gel pad
prepared in M9 glucose media for image acquisition.
Cells were visualized in a Nikon Eclipse (Ti-E,
Nikon, Japan) inverted microscope with a C2
confocal laser scanning system using a 100x Apo
TIRF (1.49 NA, oil) objective. Images of cells were
taken with Nikon NIS-elements. The Tsr-Venus
proteins can be detected as fluorescent spots under
the fluorescent confocal microscope using a 488 nm
argon ion laser (Melles-Griot) and a 515/30 nm
detection filter. Images were acquired using a large
pinhole, gain 165 and 3.36 µs pixel dwell.
2.5 Image Analysis
Cell segmentation was performed by a custom-made
software that integrates MAMLE (Chowdhury et al.,
2013) and CellAging (Häkkinen et al., 2013), for
cell segmentation. Phase contrast images were
automatically segmented and the results were
manually corrected as needed. Finally, fluorescence
images were automatically aligned to phase-contrast
images. The fluorescence in each cell was then
extracted and analysed by a custom-made Matlab
script.
To assess the location of Tsr-Venus clusters
along the major axis of a cell, we first formally
defined a ‘cluster’ or ‘spot’ as a connected
component with each pixel having a light intensity
above a threshold. For this, we performed a
Laplacian of Gaussian based spot detection
combined with an adaptive local thresholding step
(Annila, 2015). It assumes that the background pixel
intensities follow a Gaussian distribution. The
threshold is then selected for each cell separately
based on the fitted distribution such that the
probability of mislabeling a pixel from this
distribution is smaller than 0.005. From the
segmented image, the number of clusters was
counted and the area of each cluster was calculated
by counting the number of pixels within.
To distinguish between midcell and poles, we
defined a boundary between them at 0.5 (with 0
being midcell and 1 being the cell extreme) (Santos
et al., 2014).
Finally, the intensity of the clusters was directly
acquired from background corrected cells, which
were obtained by subtracting the median cell
intensity from each pixel of the cell. Next, for each
cell, the total protein fluorescence was obtained by
summing the fluorescent intensity of the clusters.
3 RESULTS
First, after segmenting cells and clusters of Tsr-
Venus (see example Figure 1), for each temperature
condition, we extracted from each cell the total
fluorescence of the clusters, the number of clusters
and the area occupied by the clusters (Table 1).
Figure 1: Fluorescence microscopy image of Escherichia
coli cells expressing Tsr-Venus.
From Table 1, the production of Tsr-Venus proteins
under the control of P
lac
is heavily temperature
dependent, as expected (Kandhavelu et al., 2012),
BIOINFORMATICS 2016 - 7th International Conference on Bioinformatics Models, Methods and Algorithms
138
Table 1: Tsr-Venus clustering at different temperatures. Standard error of the mean (SEM) is shown in parenthesis.
10°C 15°C 24°C 37°C 43°C
No. cells 176 301 259 265 299
Mean total cluster
fluorescence (a.u.)
0.3x10
4
(0.02x10
4
)
0.8x10
4
(0.04x10
4
)
1.5x10
4
(0.06x10
4
)
1.5x10
4
(0.08x10
4
)
0.8x10
4
(0.08x10
4
)
Avg. no. clusters per cell 0.53 (0.05) 0.74 (0.03) 1.02 (0.03) 1.23 (0.03) 0.49 (0.04)
Avg. cluster area (µm
2
) 0.5 (0.03) 0.74 (0.02) 1.1 (0.03) 1.1 (0.04) 0.8 (0.04)
Clustering coefficient (%) 13.9 (1.4) 30.6 (1.5) 56.2 (1.6) 61.5 (1.2) 18.0 (1.4)
with the total fluorescence per cell from these
proteins being maximized at 37°C. Meanwhile, we
expect the decrease in mean fluorescence per cell
from 37°C to 43°C to likely be due to the increased
cell division rate with temperature not being
compensated sufficiently by an increase in protein
production.
Next, in order to assess whether the clustering
process is temperature dependent, we calculated the
‘clustering coefficient’ of Tsr-Venus. We define this
quantity as the ratio between the fluorescence from
clusters and the total cell fluorescence. Averages
from all cells of these three quantities for each
condition are shown in Table 1. Visibly, the
clustering coefficient maximizes at 37°C and, thus,
we conclude that, under optimal temperature, the Tsr
proteins are more efficiently clustered.
We also plotted the mean cluster fluorescence
against the mean number of clusters per cell, for
each condition (Figure 2). Visibly, there is a lack of
linear relationship between temperatures lower than
and higher than the optimal. This suggests, as
mentioned above, that the underlying causes for
weaker clustering at low temperatures differ from
the causes at high temperatures. Further studies are
needed to identify these causes. We expect division
rates and diffusion rates of the clusters to play a role
in these differences.
To test whether Tsr-Venus fluorescence is
affected by temperature in the range studied, we
plotted the mean cluster fluorescence against the
mean area of the clusters per cell, for each condition
(Figure 3).
From Figure 3, we conclude that bigger clusters
tend to be brighter and that this relationship between
size and brightness is fairly independent of
temperature.
Finally, we studied whether the intracellular
localization of the clusters is affected by
temperature. For that, we measured the fractions of
Figure 2: Mean total fluorescence from Tsr clusters as a
function of the average number of these clusters per cell.
Figure 3: Mean total fluorescence from Tsr-clusters as a
function of the average area of the clusters per cell.
‘polar’ and ‘lateral’ clusters (Table 2). The former
are defined as those inside the poles of the cell,
while the latter are at midcell, the region in between
the poles.
In all cases, the percentage of polar clusters was
over 94%. We performed a binomial test for the
proportions of lateral clusters with the null
Robustness to Sub-optimal Temperatures of the Processes of Tsr Cluster Formation and Positioning in Escherichia Coli
139
hypothesis that they are coming from the same
distribution. It is usually accepted that, for p-values
smaller than 0.01, the null hypothesis is rejected.
Table 2: Localization of Tsr-Venus clusters. Shown are
the percentages of clusters at the poles (‘polar clusters’),
and the percentages of clusters at midcell (‘lateral
clusters’) for cells at 10°C, 15°C, 24°C, 37°C, and 43°C.
10°C 15°C 24°C 37°C 43°C
Polar
clusters (%)
98.9 99.1 99.6 96.0 94.5
Lateral
clusters (%)
1.1 0.9 0.4 4.0 5.5
From the comparisons between all pairs of
conditions, we find a tangible difference only
between 24°C and 43°C (all other p-values were
above 0.05). From this, we conclude that the
localization process is only weakly dependent of
temperature, in the range tested.
4 CONCLUSIONS
Tsr proteins play a central role in the chemotaxis
mechanisms of Escherichia coli. For this, they
participate in large clusters of various proteins.
From microscopy measurements of cells
expressing Tsr proteins tagged with Venus proteins,
we compared the clustering process and the
behaviour of the clusters at different temperatures.
We found that, in all conditions, these proteins
are able to form clusters and that these preferentially
locate at the cell poles. Nevertheless, the clustering
process is temperature dependent in that, at 37 °C,
the clusters clearly harness Tsr-Venus proteins more
efficiently than in the other conditions.
At the moment, the cause(s) for the dependence
of the clustering process on temperature is unknown.
However, recent studies showed that both the
cytoplasm viscosity (Parry et al., 2014) as well as
the relative nucleoid size of these cells are heavily
temperature dependent, which could explain why
temperature affects the long-term spatial distribution
of the clusters. Another possibility is that the
clustering process is not energy-free, and thus
depends on how much energy the cell has available
(similarly to the clustering of unwanted protein
aggregates (Govers et al., 2014)).
On the other hand, we observed that the
localization of the clusters (at the pole or at midcell),
appears to be only weakly temperature dependent,
within the range of temperatures studied. From this,
we conclude that the functionality of the proteins
responsible for retention of Tsr-clusters at the poles,
such as the Tol-Pal complex, is robust to suboptimal
temperatures, or that other, non-energy dependent
mechanisms, contribute to this preference for polar
localization.
In the future, it would be of interest to investigate
the causes for the temperature dependence of the
clustering process and for the temperature
independence of the localization process of these
proteins.
ACKNOWLEDGEMENTS
Work supported by Academy of Finland (257603,
ASR) and Portuguese Foundation for Science and
Technology (PTDC/BBB-MET/1084/2012, ASR).
The funders had no role in study design, data
collection and analysis, decision to publish, or
preparation of the manuscript.
REFERENCES
Annila, T., 2015. Detection of fluorescently labeled
particles in Escherichia coli. Msc Thesis. Department
of Signal Processing. Tampere University of
Technology.
Chowdhury, S. et al., 2013. Cell segmentation by multi-
resolution analysis and maximum likelihood
estimation (MAMLE). BMC Bioinformatics, 14 Suppl
1(Suppl 10), p.S8.
Govers, S. K., Dutre, P. & Aertsen, A., 2014. In Vivo
Disassembly and Reassembly of Protein Aggregates in
Escherichia coli. Journal of Bacteriology, 196(13),
pp.2325–2332.
Häkkinen, A. et al., 2013. CellAging: a tool to study
segregation and partitioning in division in cell lineages
of Escherichia coli. Bioinformatics, 29(13), pp.1708–
9.
Kandhavelu, M. et al., 2012. Single-molecule dynamics of
transcription of the lar promoter. Physical Biology,
9(2), p.026004.
Kentner, D. & Sourjik, V., 2006. Spatial organization of
the bacterial chemotaxis system. Current Opinion in
Microbiology, 9(6), pp.619–24.
Lee, L., Mizuno, T. & Imae, Y., 1988. Thermosensing
properties of Escherichia coli tsr mutants defective in
serine chemoreception. Journal of Bacteriology,
170(10), pp.4769–74.
Liberman, L., Berg, H. C. & Sourjik, V., 2004. Effect of
Chemoreceptor Modification on Assembly and
Activity of the Receptor-Kinase Complex in
Escherichia coli. Journal of Bacteriology, 186(19),
pp.6643–6646.
Lybarger, S. R. & Maddock, J. R., 1999. Clustering of the
chemoreceptor complex in Escherichia coli is
BIOINFORMATICS 2016 - 7th International Conference on Bioinformatics Models, Methods and Algorithms
140
independent of the methyltransferase CheR and the
methylesterase CheB. Journal of Bacteriology,
181(17), pp.5527–9.
Nagai, T. et al., 2002. A variant of yellow fluorescent
protein with fast and efficient maturation for cell-
biological applications. Nature Biotechnology, 20(1),
pp.87–90.
Parkinson, J. S., Ames, P. & Studdert, C. A., 2005.
Collaborative signaling by bacterial chemoreceptors.
Current Opinion in Microbiology, 8(2), pp.116–21.
Parry, B.R. et al., 2014. The bacterial cytoplasm has glass-
like properties and is fluidized by metabolic activity.
Cell, 156(0), pp.183–194.
Rebbapragada, A. et al., 1997. The Aer protein and the
serine chemoreceptor Tsr independently sense
intracellular energy levels and transduce oxygen,
redox, and energy signals for Escherichia coli
behavior. Proceedings of the National Academy of
Sciences of the United States of America, 94(20),
pp.10541–6.
Santos, T. M. A. et al., 2014. Polar localization of
Escherichia coli chemoreceptors requires an intact
Tol-Pal complex. Molecular Microbiology, 92(May),
pp.985–1004.
Skidmore, J. M. et al., 2000. Polar clustering of the
chemoreceptor complex in Escherichia coli occurs in
the absence of complete CheA function. Journal of
Bacteriology, 182(4), pp.967–73.
Sourjik, V. & Berg, H. C., 2004. Functional interactions
between receptors in bacterial chemotaxis. Nature,
428(3), pp.437–41.
Thiem, S., Kentner, D. & Sourjik, V., 2007. Positioning of
chemosensory clusters in E. coli and its relation to cell
division. The EMBO Journal, 26(6), pp.1615–23.
Wadhams, G. H. & Armitage, J. P., 2004. Making sense of
it all: bacterial chemotaxis. Nature Reviews. Molecular
Cell Biology, 5(12), pp.1024–37.
Vaknin, A. & Berg, H. C., 2006. Osmotic stress
mechanically perturbs chemoreceptors in Escherichia
coli. Proceedings of the National Academy of Sciences
of the United States of America, 103(3), pp.592–6.
Yu, J. et al., 2006. Probing gene expression in live cells,
one protein molecule at a time. Science, 311(5767),
pp.1600–3.
Robustness to Sub-optimal Temperatures of the Processes of Tsr Cluster Formation and Positioning in Escherichia Coli
141