Team:Imperial College London/Project Chemotaxis Testing

From 2011.igem.org

(Difference between revisions)
Line 40: Line 40:
</div>
</div>
-
<p>After initial attempts to image and analyze random biased walk up a concentration gradient, we have changed the strategy for observation of the bacterial response to malate. We decided to look for difference in probabilities of bacteria being in any of the two movements. During bacterial movement up a concentration gradient of the attractant, the probability of the smooth swimming is higher than that of tumbling. However we have changed the assay to observe difference in velocities, when exposing bacteria to saturated attractant concentration. We expected the velocity of the cells that are capable of the attractant recognition to be different to those that can not.  This observation has been performed with bacteria in mid-exponential phase (OD<sub>600</sub> 0.4-0.6) induced in motility buffer 2 hours prior to observation. Bacteria containing construct PA2652 (<a href="http://partsregistry.org/Part:BBa_K515102">BBa_K515102</a>) have been exposed to motility buffer (reference), 10 mM serine (positive control) & 10 mM malate (test).<i>Escherichia coli</i> DH5α without any construct have been exposed to 10 mM malate (negative control). Observation was performed using a Zeiss Axiovert 200 Inverted Fluorescent Microscope, with video collection Volocity software. ImageJ plug-in Manual Tracking was used to collect data and Chemotaxis Tool plug-in to analyze the data.</p>
+
<p>After initial attempts to image and analyze random biased walk up a concentration gradient, we have changed the strategy for observation of the bacterial response to malate. We decided to look for a difference in probabilities of bacteria being in either of the two movements. During bacterial movement up a concentration gradient of attractant, the probability of the smooth swimming is higher than that of tumbling. However we have changed the assay to observe difference in velocities, when exposing bacteria to saturated attractant concentration. We expected the velocity of the cells that are capable of the attractant recognition to be different to those that can not.  This observation has been performed with bacteria in mid-exponential phase (OD<sub>600</sub> 0.4-0.6) induced in motility buffer 2 hours prior to observation. Bacteria containing construct PA2652 (<a href="http://partsregistry.org/Part:BBa_K515102">BBa_K515102</a>) have been exposed to motility buffer (reference), 10 mM serine (positive control) & 10 mM malate (test).<i>Escherichia coli</i> DH5α without any construct have been exposed to 10 mM malate (negative control). Observation was performed using a Zeiss Axiovert 200 Inverted Fluorescent Microscope, with video collection Volocity software. ImageJ plug-in Manual Tracking was used to collect data and Chemotaxis Tool plug-in to analyze the data.</p>
<h3>2.2 Qualitative analysis<a href="https://2011.igem.org/Team:Imperial_College_London/Protocols_Chemotaxis"><img src="https://static.igem.org/mediawiki/2011/5/58/ICL_ProtocolIconDark.png" width="140px" align="right"/></h3></a>
<h3>2.2 Qualitative analysis<a href="https://2011.igem.org/Team:Imperial_College_London/Protocols_Chemotaxis"><img src="https://static.igem.org/mediawiki/2011/5/58/ICL_ProtocolIconDark.png" width="140px" align="right"/></h3></a>

Revision as of 22:50, 21 September 2011




Module 1: Phyto-Route

Chemotaxis is the movement of bacteria based on attraction or repulsion of chemicals. Roots secrete a variety of compounds that E. coli are not attracted to naturally. Accordingly, we engineered a chemoreceptor into our chassis that can sense malate, a common root exudate, so that it can swim towards the root. Additionally, E. coli are actively taken up by plant roots, which will allow targeted IAA delivery into roots by our system.






Testing

Collapse all | Expand all

1. Introduction

The assembled construct PA2652 (BBa_K515102) and non-codon optimised mcpS (in pRK415 backone vector), have been inserted and tested for functionality in E. coli DH5α obtained from New England Biolabs. We carried out several tests in an attempt to characterise the rewired chemotaxis towards L(-)malic acid. We separated testing of chemotaxis towards malate into behavioral, qualitative & quantitative analyses.

To test bacterial uptake into the roots of the plants, we worked with Arabidopsis thaliana to replicate the experiment by Paungfoo-Lonhienne et al. A. thaliana is a common plant model organism. Its genome has been almost completely sequenced and replicates quickly, producing a large number of seeds. It is easily transformed and many different mutant strains have been constructed to study its different characteristics[1]. While Arabidopsis may not represent plant populations naturally occurring in arid areas threatened by desertification, it is a handy model organism we will be using to study the effect of the auxin plant hormone, indole-3-acetic acid (IAA) on roots, observe chemotaxis towards them, and observe them taking up bacteria as nutrients.

Collapse

2. Bacterial chemotaxis towards malate

2.1 Behavioral analysis


Bacteria perform two types of movement, smooth swimming and tumbling. In the absence of attractant the result of the two movements lead to random walk with no directionality. In the presence of a concentration gradient of attractant, the probabilities of the two movements change and lead to the directional movement known as random biased walk.




Video 1: Mixed population of GFP-expressing control and non-GFP labelled PA2652 expressing E. coli. The bacteria have been placed in motility buffer without attractant and therefore there is no difference in behaviour between the two. Two types of bacterial movement can be observed. As no attractant is present, a majority of the bacteria perform tumbling, however there are individual bacteria that perform smooth swimming, a rapid movement. In this video, bacteria decide randomly between tumbling and smooth swimming, this behaviour is known as random walk. When bacteria perform chemotactic behaviour towards a concentration gradient(not pictured), the probability of the bacteria performing smooth swimming increases after a transient decrease due to saturation of chemoreceptors. This video has been taken as a set of frames, with one frame per four seconds. Imaging done by Imperial iGEM 2011.

After initial attempts to image and analyze random biased walk up a concentration gradient, we have changed the strategy for observation of the bacterial response to malate. We decided to look for a difference in probabilities of bacteria being in either of the two movements. During bacterial movement up a concentration gradient of attractant, the probability of the smooth swimming is higher than that of tumbling. However we have changed the assay to observe difference in velocities, when exposing bacteria to saturated attractant concentration. We expected the velocity of the cells that are capable of the attractant recognition to be different to those that can not. This observation has been performed with bacteria in mid-exponential phase (OD600 0.4-0.6) induced in motility buffer 2 hours prior to observation. Bacteria containing construct PA2652 (BBa_K515102) have been exposed to motility buffer (reference), 10 mM serine (positive control) & 10 mM malate (test).Escherichia coli DH5α without any construct have been exposed to 10 mM malate (negative control). Observation was performed using a Zeiss Axiovert 200 Inverted Fluorescent Microscope, with video collection Volocity software. ImageJ plug-in Manual Tracking was used to collect data and Chemotaxis Tool plug-in to analyze the data.

2.2 Qualitative analysis


Qualitative assays should inform us about the result of rewiring chemotaxis in E. coli with the engineered constructs mcpS or PA2652 (BBa_K515102). However it does not inform us about the cell count and the extent to which engineered bacteria can chemotax towards the attractant source.

A number of methods exist that show chemotaxis towards a source [2]. Most of them are based on the properties of semi - solid agar, which allows diffusion of molecules and bacterial movement. We have modified agar plug assay to observe chemotactic response. The bacteria used were in mid - exponential phase (OD600 0.4-0.6). Cells suspended in the semi solid agar were positioned 2 cm away from the attractant source and left overnight to grow and move. At the end of the assay the plates were imaged using Fujifilm LAS-3000 Imager.

DH5-α Escherichia coli cells were used as studied subject expressing our construct. Negative control used were cells without engineered construct, with selection marker for ampicillin and kanamycin. This was to show inability of the non-engineered cells to perform chemotaxis towards malate. Positive control cells were exposed to rising concentrations of serine, a chemical which is recognised by native chemoreceptors of E. coli. This was to show that cells we are using to conduct our experiments have functional chemotaxis mechanism and are capable of recognising an attractant gradient. The tested DH5α Escherichia coli contained construct (BBa_K515102) PA2652 malate chemoreceptor. We have also tested cells containing non-codon optimised mcpS gene carried on pRK415 plasmid with selectable marker tetracycline. However due to the lack of information about the construct, and the fact that it is non-standard biobrick format, with several illegal restriction sites within the sequence, we did not test this construct with further analyses.

Results expected from this assay should show clear differences in shape of the formed colony, since the bacteria attracted to source will move and therefore distort the shape of the colony into eliptical, directed shape towards the source, in comparison with the control, which are expected to look circular as bacteria are equally likely to swim into any direction.

Table 1: The concentrations of attractant tested

Molar range

0 mM

0.01 mM

0.1 mM

1 mM

10 mM

100 mM

Milimolar range

0 mM

5 mM

10 mM

15 mM

20 mM

25 mM


Positive control

Figure 1: Rising concentrations of serine were tested. a) 0 mM control - circular colony b) 0.01 mM - circular colony e) 0.1 mM - circular colony d) 1 mM - possible eliptical colony c) 10 mM - eliptical colony f) 100 mM - eliptical colony away from the attractant. Data collected by Imperial iGEM 2011.

Figure 2: Rising concentrations of serine were tested. a) 0 mM control - circular colony, b)5 mM - eliptical colony c) 10 mM - eliptical colony d) 15 mM - eliptical colony e) 20 mM - eliptical colony f) 25mM - circular colony.

Negative control

Figure 3: Rising concentrations of malate were tested. a) 0 mM control - circular colony b) 0.01 mM - circular colony c) 0.1 mM - circular colony d) 1 mM - circular colony e) 10 mM - circular colony f) 100 mM - circular colony.

Figure 4: Rising concentrations of malate were tested. a) 0 M control - circular colony b)5 mM - circular colony c) 10 mM - circular colony d) 15 mM - circular colony e) 20 mM - circular colony, f) 25mM - circular colony.

McpS - pRK415

Figure 5: Rising concentrations of malate were tested. a) 0 mM control - circular colony b) 0.01 mM - circular colony c) 0.1 mM - circular colony d) 1 mM - circular colony e) 10 mM - circular colony f) 100 mM - circular colony.

Figure 6: Rising concentrations of serine were tested. a) 0 mM control - circular colony b)5 mM - circular colony c) 10 mM - eliptical colony formed not in the direction expected d) 15 mM - possible eliptical colony e) 20 mM - possible eliptical colony f) 25 mM - circular colony.

PA2652 - BBa_K515102

Figure 7: Rising concentrations of malate were tested. a) 0 M control - circular colony b) 0.01 mM - possible eliptical colony the shape is hard to analyze c) 0.1 mM - strange shape of colony observed, however this was not a result of mishandling with semi - solid agar d) 1 mM - colony shape is not perfectly circular however not eliptical either e) 10 mM - circular colony f) 100 mM - circular colony.

Figure 8: Rising consentrations of serine were tested. a) 0 mM control - circular colony b)5 mM - colony shape was rendered void due to mishandling with semi-solid agar c) 10 mM - possible eliptical shape colony d) 15 mM - circular colony e) 20 mM - colony shape was rendered void due to mishandling with semi - solid agar f) 25mM - circular colony.

The data obtained from this assay were not easily analysed. This assay is qualitative and therefore should provide us with positive or negative result. However upon analysis of the data a conclusion can not be drawn. This is because, set up of the assay gives space for ambiguity. This is due to a number of factors. Attractant is localised, and it diffuses to all directions, however the plate is not infinite and therefore loss of concentration gradient occurs over time. Another factor is the semi-solid agar itself. This medium is relatively difficult to manipulate with and a number of samples were ruined during the handling. Even considering that set up of this assay lead to vague results a number of points can be drawn. Positive control have shown that E. coli DH5α are capable of chemotaxis. It has also shown that when added attractant is of too high concentration (10-1 M) bacteria do not swim directly towards the attractant source since the medium around the source is saturated. Instead they perform chemotaxis in a direction, where the attractant concentration gradient is set up, so that it is in the range for sensing by chemoreceptors. This occurs even if it means that the bacteria effectively move away from the attractant source. Negative control have shown that at any tested concentration of L(-)malic acid E. coli DH5α without PA2652 or mcpS construct do not perform chemotaxis towards attractant show. The assay have, however failed to conclude a result for mcpS or PA2652 construct as the colonies observed have a range of shapes at different attractant concentrations, that do not allow us to conclude positive or negative result.

2.3 Quantitative Analysis


In comparison to qualitative assays, quantitative assays are more informative as they provide cell count based on different attractant concentrations and therefore allow identification of the optimal attractant concentration for bacterial chemotaxis. Our analyses were based on the high throughput capillary assay [3].

We have modified this assay to obtain cell count through flow cytometer BD FACScan in contrast to commonly used CFU count. The assay itself is based on a number of capillary tubes filled with different concentrations of attractant placed into bacterial suspension for a period of 30 minutes. Even though the assay itself sounds simple, the set up was proven to be very difficult to lead us to obtain any useful data.

Table 2: The concentrations of tested serine (control) & malate attractant.

Molar range

0 mM

0.001 mM

0.01 mM

0.1 mM

1 mM

10 mM

100 mM

Milimolar range

0 mM

5 mM

10 mM

15 mM

20 mM

25 mM

30 mM

The capillaries, however did not function in a way due to a number of issues, which we have not managed to fully address in our set up. First is the suspension of the liquid in the capillary, while capillary is not suspended in liquid. We have tried to parafilm the top of the glass capillary tubes but it did not seem to create full vaccuum. For this reason we have minimised our choice to capillaries that can suspend liquid,so we have continued testing with syringes and 10 µL BioRobotix™ tips with ART barrier. Second main problem we have found, was the generation of surface tension, when we tried to remove the capillaries at the end of the assay, this then influenced the result greatly.

Figure 9: Failed set up of high-throughput capillary assay, using syringes loaded with 100 µL of attractant. Possible reason for non-functionality of this assay is surface tension upon removal of the syringes from bacterial suspension.

Figure 10: Failed set up of high-throughput capillary assay, using 10 µL BioRobotix™ tips. Possible reason for non-functionality of this assay was the use of multichannel pippette.

To get around this we have tried to build a grid for the capillary tubes and perform the experiment on perfectly leveled surface, however removal of the rack was impossible without the result being influenced by surface tension. The surface tension influenced syringe capillaries to greater extent than 10 µL BioRobotix™ tips. The BioRobotix™ tips however seemed to have problem with releasing equal amounts of attractant with bacteria after the assay was finished and some of the tips did not release any of the liquid. One reason for this can be usage of multichannel pippette. These problems have been observed upon analysis of data. Data has been analyzed with CyflogicTM software, CyFlo Ltd, Finland.

Figure 11: Negative control are cells without construct PA2652 exposed to rising malate concentrations. Positive control are cells containing PA2652 construct exposed to rising serine concentrations. PA2652 cells is the test of the construct, exposed to rising malate concentrations. Blank consists of empty motility buffer exposed to rising attractant concentrations. Due to an inability to find any statistically significant trends in the data, we concluded that the assay´s set up had failed.

Collapse

3. Uptake of bacteria into roots

Figure 11. GFP-expressing E. coli cells inside Arabidopsis roots (data and imaging by Imperial College iGEM 2011).

One important part of our project is the uptake of our bacteria into plant roots. The observation that this occurs in both Arabidopsis and tomatoes and that both E. coli and the yeast Saccharomyces cerevisiae can be taken up by roots (albeit under controlled lab settings) is new and was only published last year [4].

As the amount of auxin needed for enhancing plant growth depends on whether our bacteria are producing the compound outside or inside the plants, we attempted to replicate these findings.

In preparation for confocal imaging, we met with Dr Martin Spitaler and Mark Scott who advised us on how to prepare samples and image them. Confocal microscopy is much more precise than conventional light field microscopy as it eliminates background light by focusing the laser through a pinhole (Mark Scott, oral communication). The confocal microscopy we conducted was focused on imaging GFP expressing bacteria inside Arabidopsis roots to show that uptake of the bacteria takes place.

In an imaging trial run we found that natural fluorescence can be measured in roots in a spectrum that does not interfere with measuring the superfolder GFP expressed by our E. coli cells. In addition, the autofluorescence is strong enough to enable us to identify individual cells (Video 2). We did therefore not need to dye the roots before imaging.







Video 2. Stack of wt Arabidopsis root. The root can be imaged at around 488nm. Imaging carried out by Dr Martin Spitaler for Imperial College iGEM 2011.

In order to prepare the GFP-expressing bacteria for plant uptake, they were spun down and media was exchanged prior to incubation at 37°C to reach exponential phase. Bacteria were then spun down and resuspended in wash buffer (5mM MES) to reach OD 30. 8 ml, 4 ml and 2 ml were added to separate flasks, containing 100ml of half-MS media each. 4ml and 6ml of wash buffer were added to the flasks containing 4ml and 2ml bacteria, respectively. 8 ml of wash buffer was added to the negative control. Ten Arabidopsis seedlings were distributed into each of the flasks. Incubation was carried out for 15 hours prior to imaging.

Prior to imaging, roots were washed in PBS to wash off bacteria and facilitate imaging. We imaged the plants incubated with 8ml of bacteria and were able to find bacteria inside one of the roots. A 3D picture was taken of uninfected roots and roots containing bacteria by taking a Z stack image using confocal microscopy (Video 3).

Video 3. Bacteria inside of Arabidopsis thaliana roots. These videos were put together using Z-stack images taken on a confocal microscope. These images can be converted into 3D pictures that allow us to verify that the bacteria can indeed be found inside the roots rather than on the surface (data and imaging by Imperial College iGEM 2011).

We repeated this experiment at a later date with plants that had been allowed to grow for a longer period of time. The bacteria were predominantly found in root hairs and inside of cells on the root surface (Figure 11). These plants were older than the ones previously used for uptake experiments and bacterial uptake was much more predominant.

These results show that we would be able to expose plants to auxin from inside the roots themselves. This is extremely important for assessing the secretion of auxin by our bacteria. Nevertheless, we will need to conduct further uptake experiments in soil and with other plant species to confirm that uptake would be possible in the projet's eventual implementation stage.

Collapse

4. References

[1] National Institute of Health. (no date) Model Organisms for Biomedical Research. [Online]. Available from: http://www.nih.gov/science/models/arabidopsis/index.html [Accessed 21st September 2011].

[2] Jain RK and Pandey J (2010) Chemotactic responses. Timmis VKN, ed. Handbook of hydrocarbon and lipid microbiology. Berlin, Heidelberg: Springer Berlin Heidelberg, pp. 3933-3955. Available at: http://www.springerlink.com/content/x458521h8420l478/ [Cited September 16, 2011].

[3] Bainer R, Park H and Cluzel P (2003) A high-throughput capillary assay for bacterial chemotaxis. Journal of Microbiological Methods 55: 315-319.

[4] Paungfoo-Lonhienne C et al. (2010) Turning the table: plants consume microbes as a source of nutrients.PLoS One 5(7): e11915.

Collapse