mCherry ColN lowers cell viability of
Escherichia Coli cells
Naam student: Nigel Kroone Studentnummer: 11637773
Naam dagelijkse begeleider: Jolanda Verheul Naam stage begeleider: Tanneke Den Blauwen
Inhoud
mCherry ColN lowers cell viability of Escherichia Coli cells...1
Abstract...3
Introduction...3
Experimental procedures...5
Bacterial strains, growth media, and antibiotics...5
Molecular cloning of sfTq2ox-TolB and mCh-ColN...5
Microscopy and image analysis...6
SDS-PAGE and Western blot...6
Growth curve...6
Results...7
sfTq2ox-TolB localizes at the division sites of E. coli cells...7
Localization of mCh-ColN in LMC500 mut2GFP-TolA...9
sfTq2ox-TolB in ΔTolB cells restores normal cell function...11
Interaction between TolA and mCh-ColN...13
SDS-PAGE and Western Blot of mCh-ColN clones confirms integrity of mCh-ColN protein...15
mCh-ColN lowers cell viability in BW25113, ΔTolA and ΔTolB cells...16
Discussion...18
Colicin N has intrinsic toxic properties in the TolA binding domain hampering the possibility of a comparison assay using this molecule...18
Appendix 1...20
Appendix 2...21
Appendix 3...24
Abstract
Colicin N binds to TolA, which inhibits proper invagination of the outer membrane causing eventual cell death. This TolA binding property can be used as the basis of a comparison assay that could help to identify other molecules with this function. However, this thesis revealed the intrinsic toxic properties of the TolA binding domain of Colicin N by the use of phase contrast and fluorescence microscopy and comparissons between the growth curves of different E. Coli samples.
Introduction
Colicins are bacteriocins produced by some strains of Escherichia coli to kill related strains of E. coli. Colicins do this by binding to receptors on the outer membrane, from there translocator proteins are recruited and used to cross the outer membrane (Cascales et al., 2007; Lazdunski et al., 1998). Once in the periplasm colicins deliver their toxic domain into or through the inner membrane. Colicin N is one such colicin, its toxic domain kills the cell by forming a ion channel in the inner membrane (Baty et al., 1990). Colicins are classified into two groups depending on the translocater protein used to cross the outer membrane. Group A uses Tol proteins A, B, Q, and R, while Group B uses the Ton system. ColN is a part of Group A and binds to TolA (Ragget et al., 1998). TolA is part of the trans-envelope Tol-Pal complex, this complex is responsible for the invagination of the outer membrane during cell division (Gerding et al., 2007). The Tol system coordinates the recruitment of Pal at division sites after which Pal binds the peptidoglycan and causes the invagination of the outer membrane (Sczepaniak et al., 2020). If parts of this complex are not present during cell division the outer membrane will not invaginate properly. ColN competes with TolB for the binding of TolA and can thus cause improper cell invagination, a weakening of the outer membrane and eventually cell death (Gerding et al., 2007). ColN is a peptide, which makes it difficult to use as an antibiotic because it can only be administered intravenously as it would degrade too fast otherwise.
The TolA binding property of ColN could however be used as the basis of a comparison assay for other TolA binding molecules. Similar molecules that have the same TolA binding properties can be used to weaken the outer membrane and allow antibiotics that would normally be stopped by this barrier to pass it and kill the cell. The outer membrane of gram negative bacteria is a large obstacle for most antibiotics as it acts as a selective barrier using specialized pore forming proteins (Omps) (Ghai et al., 2017). Weakening this barrier could allow for antibiotics to bypass the problem. If The comparison assay will make use of a fusion protein consisting of mCherry genetically linked to the TolA binding domain of the ColN peptide to identify these molecules. Only this part of ColN will be used as this is the part of interest for the assay. Image analysis of living E. coli cells will then be performed to analyse the localization of the constructed protein. It is expected that the ColN is localized at the division site as this is also where TolA is localized during cell division. The co-localization of ColN and TolB will also be analysed using a fluorescence protein (FP) linked TolB. Here both ColN and TolB will be localized near TolA. However, as these molecules compete in their interaction with TolA some fluorescence is expected away from the division site in the periplasm. Furthermore, both proteins will be analysed in TolA and TolB cells to analyse their localization without TolA and TolB, respectively. If the TolB cells transformed with the FP-TolB protein regain the functions of the Tol-Pal system the FP-TolB protein is confirmed to have a functional TolB and has not been modified during transformation. The TolA cells transformed with mCh-ColN will confirm whether or not the interaction of ColN and TolA is toxic or if the ColN protein has intrinsic toxic properties. When only the interaction between ColN and TolA is toxic the TolA cells are not expected to show any
sign of this toxicity. If the comparison assay is successful in finding alternative TolA binding molecules the FP linked ColN and TolB will not be localized near TolA but spread across the periplasm. This would mean another molecule is outcompeting both ColN and TolB in their interaction with TolA. The eventual goal of this comparison assay is to find these alternative TolA binding molecules as these could be used together with antibiotics that cannot cross the outer membrane. The TolA binding molecules will then weaken the outer membrane which will open the gates for the antibiotic to kill the cell.
A) B) C)
Figure 1: schematic rendition of ColN Comparison assay. (A): Tol-Pal complex during cell division with a TolB linked to a
fluorescent protein (FP). The TolB is able to link to TolA and thus recruit Pal and cause proper outer membrane invagination. (B): Tol-Pal complex during cell division with fluorescently linked ColN and TolB. ColN binds to TolA which causes TolB to be present in the periplasm but not restricted to the division site while the fluorescence of the ColN molecule will be present at the division sites. (C): Tol-Pal complex during cell division with fluorescently linked ColN and TolB and new TolA binding molecule X. If a new TolA molecule is found with this comparison assay that can outcompete both ColN and TolB for their interaction with TolA, both the fluorescence of ColN and TolB will be found in the periplasm and not at the division site as TolA will bind this new molecule.
Experimental procedures
Bacterial strains, growth media, and antibiotics
Escherichia coli LMC500, LMC500 mut2GFP-TolA, BW25113, BW25113 ΔTolA and BW25113 ΔTolB cells were grown at 37˚C in rich glucose medium containing 10 g of bactotryptone, 5 g of yeast extract, 5 g of NaCl, 15 mmol NaOH per litre (TY). 1:1000 chloramphenicol was added to the media of strains transformed with the pSAV057-dsba-LEGPAGL-mCherry-GSGGS-40-74ColN camR plasmid while 1:1000 spectinomycin was
added to the media of strains transformed with pECS-dsbA-sfTq2ox-ToLB specR. Molecular cloning of sfTq2ox-TolB and mCh-ColN
For this experiment only the 40-74 amino acids of ColN were used as this is the part of the protein that interacts with TolA. ColN was cloned the plasmid pSAV057-dsba-LEGPAGL-mCherry-GSGGS-PBP5 camR from the DH5α E. coli. TolB was cloned as a whole using the
plasmid pECS-dsbA-sfTq2ox-PBP5 specR from XL1Blue E. coli. Both fusions were placed
under the control of the lac promoter (Plac). 40-74ColN was produced using a PCR with the
forward primer and reverse primers of
5’-atatatgaattcggatctggaggatctaattccaatggatggtcatggagtaataagcctcataaaaatgatggcttccacagtgatgg-3’ and 5’- tatataaagcttctaaggctttgaattattgtccccatgaaatgtaatatggtaagaaccatcactgtggaagccatc-3’, respectively. The PCR mix contained 1 µl dNTPs, 10 µl 5x phusion buffer, 0,3 µl phusion polymerase 10 µl Fw and Rev primer and 1 µl MgCl2, the solution was filled to 50 µl with
MQ. The PCR program is shown in Table 1 below. The TolB insert was produced using cPCR of LMC500 E. coli. The forward and reverse primers that were used for this PCR were, respectively, gcataggaattcggttctggaggttctgctgaagtccgcattgtgatcg-3’ and
5’-gcatagaagctttcacagatacggcgaccagg-3’. The PCR mix contained 1 LMC500 colony, 2 µl dNTPs, 2 µl Fw and Rev primer, 0,5 µl PFU polymerase, 5 µl PFU 10x buffer, 1 µl MgCl2
and MQ to 50 µl. The PCR program is shown in Table 1 below. Both the plasmids and the PCR products were digested for 2 hours at 37˚C using 0,25 µl EcoRI and 0,25 µl HindIII and 1,5 µl Cutsmart buffer. The digested 40-74ColN, TolB inserts and their respective plasmids were ligated overnight using 1 µl insert, 8 µl vector, 1 µl T4 ligase, 1,5 µl T4 buffer 10x and MQ to 15 µl. Competent LMC500 cells were transformed with either
pSAV057-dsba-LEGPAGL-mCherry-GSGGS-40-74ColN camR or pECS-dsbA-sfTq2ox-TolB specR and plated
on medium containing the antibiotic for which the plasmid contained resistance. Plasmid DNA of successful colonies was purified and sequenced using the Fw primer
5’-gcgatcacatggtcctg-3’ and Rev primer 5’-gtcttgcgtcttcgccagacta-3’ for TolB and the Fw primer 5’-catctacaaggtgaagctgc-3’ and Rev primer 5’-cccagtctttcgactgagcc-3’ for ColN (Eurofins Scientific, Luxembourg, Luxembourg).
A. B.
Temperature Time Cycles Temperature Time Cycles
98˚C 2” 1x 98˚C 3” 1x 98˚C 30’ 1x 98˚C 30’ 30x 54˚C 1” 1x 54˚C 1” 30x 72˚C 5” 1x 72˚C 2,5” 30x 72˚C 10” 1x 72˚C 10” 1x 4˚C ∞ ∞ 4˚C ∞ ∞
Microscopy and image analysis
Microscopy images were analyzed using phase contrast and fluorescence microscopy, the results were further analysed using the Coli inspector program for the image processing software “ImageJ” with the “ObjectJ” plugin (Schneider et al., 2012). ObjectJ allows for the organization of all image-analysis tasks and provides multiple output formats for the gathered data. Coli-Inspector was used to determine the division cycle cell age and analyze the localization of the fluorescent protein. The fluorescence channels that were used had the following cut-offs: mCherry Excitation: 540-580 nm, Emission: 593-667 nm; GFP Excitation: 450-490 nm, Emission: 500-550 nm and CFP Excitation: 426-446 nm, Emission: 460-500 nm. When induced, 15 µM IPTG was added to the cells for 1 hour.
SDS-PAGE and Western blot
The mCh-ColN protein was examined using a SDS-PAGE and subsequent western blot. For this experiment LMC500, LMC500 ColN, LMC500 dsbA-mCherry, BW25113 mCh-ColN, BW25113 ΔTolA mCh-ColN and BW25113 ΔTolB mCh-ColN cells were examined. The LMC500 cells served as a negative control and the LMC500 dsbA-mCherry served as a positive control for the western blot. The cells were first boiled in laemli buffer to release the proteins and coat them in SDS (Laemmli, 1970). The OD was then used as an approximation of protein concentration in the samples. A 12% acrylamide Tris gel was used for this
experiment as this best suited the mass of the mCh-ColN protein, which is about 31 kDa. The gel was stained using Ponceau S staining solution. For the western blot anti-mCherry
antibodies were used.
Growth curve
LMC500, LMC500 mCh-ColN, LMC500 dsbA-mCherry, BW25113 mCh-ColN, BW25113 ΔTolA mCh-ColN and BW25113 ΔTolB mCh-ColN cells were grown in GB4 minimal medium at 28˚C. Cultures were grown until an OD of 0.06-0.13 was achieved as this is the log phase. Cultures were then diluted back to lag phase and analyzed in a plate reader at 600 nm in 30˚C for 40 hours. Absorbance was measured every 5 minutes.
Results
sfTq2ox-TolB localizes at the division sites of E. coli cells
The localization of ColN and TolB in E. Coli was determined in E. coli LMC500 cells transformed with mCh-ColN or sfTq2ox-TolB. The resulting microscopy images in Figure 1A
visualized that sfTq2ox-TolB localizes around the division sites of dividing LMC500 cells. Figure 1B illustrates the localization of chromosomal TolA in LMC500
mut2GFP-TolA cells. However, these images do not share the same brightness and contrast as in Figure
1A and 1C because the mut2GFP-TolA protein that causes the fluorescence has a lower
quantum yield than sfTq2ox-TolB (Tsien, 1998; Meiresonne et al., 2019). This is also because
mut2GFP-TolA is produced from a chromosomal gene while sfTq2ox-TolB comes from a
plasmid, which causes less mut2GFP-TolA proteins to be produced than sfTq2ox-TolB (Li et
al., 2014). In the LMC500 mut2GFP-TolA cells that were transformed with sfTq2ox-TolB, the
same localization of fluorescence can be seen around the division sites (Figure 1C), this indicates that TolA and TolB localize at the same site in the cells. However, it does not prove that the sfTq2ox-TolB is functional and interacting with the chromosomal TolA. Appendix 1
shows the mean total fluorescence and mean total concentration of sfTq2ox-TolB and confirms
_________________________________________________________________________________
A. LMC500 with sfTq2ox-TolB
B. LMC500 mut2GFP-TolA
C. LMC500 mut2GFP-TolA with sfTq2ox-TolB
_________________________________________________________________________________
Figure 1: Microscopy images. A. microscopy images of LMC500 with sfTq2ox-TolB samples: These images show the localization of fluorescence coming from the sfTq2ox-TolB at the division site of the cells.
B. microscopy images of mut2GFP-TolA samples: These images show a fainter localization of
fluorescence around the division site of the cells. This can be explained by the difference between both proteins. Mut2GFP-TolA is a chromosomal protein while sfTq2ox-TolB is a plasmid bound protein, which
creates a difference in the amount of protein produced by the cell (Li et al., 2014). When this is coupled with the lower quantum yield of the mut2GFP protein compared to sfTq2ox it is clear why this signal is fainter. C.
microscopy images of mut2GFP-TolA with sfTq2ox-TolB samples: These images show a line of
fluorescence at the division site. The localization of fluorescence is caused by the combination of mut2GFP-TolA and sfTq2ox-TolB proteins. Grey arrows point to the localization of fluorescent protein. Scale bar = 5
Localization of mCh-ColN in LMC500 mut2GFP-TolA
The microscopy images of mCh-ColN expressed in LMC500 mut2GFP-TolA in Figure 2A do not show the same pattern of fluorescence around the division site as in the sfTq2ox-TolB
samples in Figure 1. The mCh channel has spots of fluorescence over the entirety of the cell, with cells having just some fluorescence around the division sites. This would indicate that ColN might not be localizing around the division site with TolA. However, in Figure 2C the cells do appear to form string like structures which is typically seen when ColN inhibits the Tol-Pal system (Gerding et al., 2007). This indicates that ColN is indeed present at the division sites of these cells. Figure 2B shows the localization of mut2GFP-TolA at the division sites of the LMC500 cells. This indicates that mut2GFP-TolA is present at these sites and that ColN could be interacting with TolA. This interaction can be tested by expressing mCh-ColN in ΔTolA cells. In these cells ColN will likely not show similar localization patterns as the TolA is lacking, which would normally localize around division sites. If TolA is not present in the cell ColN will not be able to interact with it and no localization of ColN will take place at the division site. Appendix 2 shows the mean fluorescence and mean concentration of mCh-ColN in the samples.
A. LMC500 mut2GFP-TolA with mCh-ColN in mCh/GFP channel
B. Mut2GFP-TolA GFP channel
C. LMC500 with mCh-ColN Ph channel
Figure 2: Microscopy images. A. microscopy images of LMC500 mut2GFP-TolA mCh-ColN samples, mCh channel left and GFP channel right: The images show some cells have fluorescence around the
division site in the GFP channel and even some fluorescence in the mCh channel. This would indicate that both ColN and TolB localize at these sites. A spotted pattern of fluorescence is visible in the mCh channel.
B. microscopy images of GFP-TolA samples in GFP-Channel: The localization around the division site
can also be seen in these cells. This is the TolA localizing at the division sites. C. Microscopy images of
LMC500 with mCh-ColN in Ph channel: The cells appear to form string like structures typical to cells
were the Tol-Pal system is inhibited by ColN implying the presence of ColN at the division sites of these cells. Grey arrows point to the localization of fluorescent protein. Scale bar = 5 µm
sfTq2ox-TolB in TolB cells restores normal cell functionΔ
The functionality of the sfTq2ox-TolB protein was tested by transforming E. coli BW25113
ΔTolB cells with the plasmid. Furthermore BW25113 ΔTolA cells were also transformed to examine the localization of sfTq2ox-TolB without TolA. Untransformed BW25113 cells
served as a negative control while transformed BW25113 cells were the positive control. In
Figure 3B the localization of the sfTq2ox-TolB protein can be seen around the division site of
cells in a similar fashion to Figure 1. When comparing Figure 3C and 3D the morphology of the cell seems to change when expressing the sfTq2ox-TolB protein. It appears that the cells
form long strings of cells unable to properly divide. The cells also appear to form “forks” indicated by the yellow arrow in Figure 3D. These changes in morphology are a sign of possible toxicity to the ΔTolA cells. This toxicity could be caused by a different cellular concentration of TolB when compared to untransformed BW25113 ΔTolA cells. If the transformed cells of Figure 3F are compared to the BW25113 ΔTolB cells in Figure 3E, the normal function of the cells appears to be restored. The transformed ΔTolB cells are no longer as small as the cells of Figure 3E but look similar to the BW25113 cells of Figure 3B. This means the sfTq2ox-TolB protein has not been modified during transformation and could allow
it to be used as a replacement for TolB in ΔTolB cells. For all images in Figure 3 the brightness and contrast was set at the same level to be able to compare the images correctly and without bias.
Figure 3: Microscopy images of the sfTq2ox-TolB samples. A: BW25113 samples. These cells do
not have any localization since they lack the fluorescent protein. B: BW25113 sfTq2ox-TolB samples.
Localization of sfTq2ox-TolB can be seen around the division site of dividing cells. C: BW25113
ΔTolA samples. No localization pattern is visible in these cells. However, these cells do appear to be
smaller than the BW25113 cells. D: BW25113 ΔTolA sfTq2ox-TolB samples. The cells appear to have
formed string like structures of cells with multiple division sites. sfTq2ox-TolB seems to localize at all
these sites, which would explain the pattern of fluorescence found in the sorted maps. There also appears to be signs of toxicity as some of the cells show “fork” like structures. E: BW25113 ΔTolB
samples. No localization pattern is visible in these cells. The cells do appear to have a different shape
than BW25113 cells as they seem to stick together in pairs. F: BW25113 ΔTolB sfTq2ox-TolB
samples. There is localization visible at the division site and the cells seem to have the same
morphology as BW25113 cells. This would mean that the plasmid has restored the normal cell function. Grey arrows point to localization of sfTq2ox-TolB, yellow arrow points to the “fork” structure and scale
Interaction between TolA and mCh-ColN
The interaction between TolA and mCh-ColN can be visualized in BW25113 ΔTolA cells. ColN interacts with TolA, which inhibits the Tol-Pal system. This interaction will not happen if TolA is not present in the cells. In these cells ColN will not interact with TolA and thus not localize around the division sites as can be seen in the BW25113 ΔTolA cells of Figure 4C. In this image mCh-ColN does not localize at any particular site in the cell. This is the case in
Figure 4A were mCh-ColN localizes in spots at the tips of the BW25113 cells. In the
BW25113 ΔTolB cells in Figure 4E there is no localization similar to Figure 4C, this implies that the lack of ToLB has an effect on localization of ColN. Another explanation of this lack of localization could be that these cells have modified the mCh-ColN protein by cleaving the signal sequence. When comparing the untransformed BW25113 ΔTolA cells (Figure 4B) with their transformed counterparts (Figure 4C), the transformed cells appear to be longer. The opposite is true for the BW25113 ΔTolB cells (Figure 4D and 4E) were the transformed cells appear to be smaller. This implies that mCh-ColN changes the morphology of both ΔTolA and ΔTolB cells, which would mean ColN still has a toxic effect on these cells. If mCh-ColN is still toxic to these cells it would mean ColN has intrinsic toxic properties and the toxicity of the molecule is not solely caused by its interaction with TolA. The microscopy images of the transformed cells share the same brightness and contrast with the untransformed cell images having a separate brightness and contrast as the fluorescence of the mCh-ColN is high as seen in Appendix 3. If this distinction was not made the BW25113 ΔTolA, ΔTolB cells would not be visible.
Figure 4: Microscopy images of the mCh-ColN samples. A: BW25113 mCh-ColN samples. There is no
apparent localization at the mid-cell site visible in these cells. However, there do appear to be spots of localization at the tips of the cells. Furthermore the localization of the protein is visible in the periplasm, which means that the signal sequence of this construct is intact. B: BW25113 ΔTolA samples. There is no localization visible in these cells as they lack the fluorescent protein. C: BW25113 ΔTolA mCh-ColN
samples. There is no localization of the mCh-ColN protein visible. The cells are also longer than the
BW25113 ΔTolA cells without plasmid. This would mean ColN has a toxic effect on the cells without its interaction with TolA. D: BW25113 ΔTolB samples. There is no localization visible in these cells as they lack the fluorescent protein. E: BW25113 ΔTolB mCh-ColN samples. These cells do not show any localization either. Additionally the cells are noticeably smaller which would imply that mCh-ColN is having a toxic effect on these cells. Scale bar is shown in grey= 5µm.
SDS-PAGE and Western Blot of mCh-ColN clones confirms integrity of mCh-ColN protein
To examine the integrity of the mCh-ColN fusion a SDS-PAGE and western blot was performed. In Lane 2 of Figure 5 the LMC500 without mCh-ColN is visible. Because of the lack of mCherry there is no immunolabeling, which is to be expected. The third and fourth lane of Figure 5 show two LMC500 mCh-ColN clones. However, only the third lane has any bands. This means the LMC500 clone of lane four does not contain the correct protein sequence on the plasmid it has been transformed by. The first LMC500 clone has a band at a height of 30 kDa, which is similar to the weight of the mCh-ColN protein, namely 31 kDa. This means this band is the ColN. The lower bands are most likely caused by mCh-ColN that has been cleaved as the height of these bands (25 kDa and 20 kDa) coincides with the weight of mCherry, which is 26 kD. The same bands at 31 kDa can be seen in lane 6,7 and 8 of Figure 5, which represent BW25113 mCh-ColN, BW25113 ΔTolA mCh-ColN and BW25113 ΔTolB mCh-ColN, respectively. This means these cells did indeed contained the intact mCh-ColN protein and it has not been cleaved by the cells. These lanes also have lower bands at 25 kD and 20 kD similar to lane 3, likely also caused by some of the mCh-ColN protein being cleaved as well. However, the presence and size of the band at 30 kD indicates most of the mCh-ColN protein in these cells is indeed intact. Lane 5 of Figure 5 contains proteins from a LMC500 dsbA-mCherry clone, which acts as positive control for the anti-mCherry antibodies. This lane has bands between the markers of 37 kDa and 25 kDa, at around 30 kD. Another band is visible below the first two at a height of 20 kDa. The dsbA signal sequence has a weight of 2 kDa, which makes the weight of dsbA-mCherry around 28 kD. This is in line with the height of the bands found in this lane, which confirms the presence of dsbA-mCherry and with that the specificity of the anti-mCherry antibodies.
Figure 5: The western blot of mCh-ColN colonies. Lane 1 is the Precision Plus Protein™ Dual Color
Standards marker. Lane 2 contains a LMC500 colony without vector and thus shows no bands. Lane 3 and 4 are two LMC500 mCh-ColN clones. Only lane 3 contains any bands at around 30 kD. Lane 5 contains a LMC500 dsbA-mCherry colony and has a band at 30 kD and at 20 kD. Lane 6,7 and 8 contain BW25113 mCh-ColN, BW25113 ΔTolA mCh-ColN and BW25113 ΔTolB mCh-ColN samples, respectively. These three lanes have bands at 30 kD. Lane 3 and 6 also contain 2 lower bands at around 25 kD and 20 kD, while lane 7 and 8 contain one lower band at 25 kD.
mCh-ColN lowers cell viability in BW25113, TolA and TolB cellsΔ Δ
The growth rate of the different transformants was examined to analyze the effect the mCh-ColN protein has on cell growth. The results of Figure 4 indicate that the mCh-mCh-ColN protein inhibits cell growth of ΔTolA and ΔTolB cells. When comparing the growth curves of untransformed LMC500 cells with 2 LMC500 mCh-ColN clones in Figure 6A, the transformed cells reach the log phase of the growth curve later when not induced or induced with 5 µM IPTG. It appears that the LMC500 mCh-ColN cells reach the log phase faster when induced with higher amounts of IPTG like 10 or 20 µM. The growth curve during the log phase differ for the 10 µM IPTG and 20 µM IPTG samples as there appears to be a decrease in growth halfway through the log phase. When comparing the growth curves of the BW25113, ΔTolA and ΔTolB cells with the untransformed LMC500 cells, the growth curve of the ΔTolA and ΔTolB cells reach the log phase later than the BW25113 and LMC500 cells (Figure 6B). Moreover, the absorbance of both of the delta BW25113 types remains lower during the stationary phase than in the LMC500 and BW25113 samples. The lower amount of absorbance in the stationary phase indicates that there are less viable bacterial cells in this phase for the ΔTolA and ΔTolB cells than the BW25113 and LMC500 cells. A lower viability could be caused by the mCh-ColN protein as the microscopy images of Figure 4 indicate that it has toxic properties on the ΔTolA and ΔTolB cells. Figure 6C visualizes the decrease in absorbance with the increase of induction in the BW25113 cells. This means that more mCh-ColN protein causes less viable cells. The growth curve of LMC500 dsbA-mCherry cells appears to be mostly similar to untransformed LMC500 cells (Figure 6D). The only difference is the lower absorbance in the stationary phase. This would mean there is a slight decrease in viable cells caused by the dsbA-mCherry protein.
B: LMC500 and BW25113, ΔTolA and ΔTolB mCh-ColN growth curves
D: LMC500 and LMC500 dsbA-mCherry growth curves
Discussion
Colicin N has intrinsic toxic properties in the TolA binding domain hampering the possibility of a comparison assay using this molecule
The results confirm the localization of sfTq2ox-TolB at the division sites of the cells and the
colocalization of TolB and TolA. The same localization cannot be seen as clearly in the fluorescence microscopy images of the mCh-ColN samples, However, the morphology of the cells indicates that mCh-ColN is indeed present at the division sites of the cells and is inhibiting proper cell division. When expressing the sfTq2ox-TolB protein in ΔTolB cells the
normal cell function appears to be restored. The expression of the protein also seems to have a toxic effect on ΔTolA cells. There was no localization of mCh-ColN in ΔTolA or ΔTolB cells. However, mCh-ColN appeared to have a toxic effect on both of these cell types. This toxicity means the TolA binding domain of the ColN protein has intrinsic toxic properties. The mCh-ColN protein was proven to be intact and unaltered using a SDS-PAGE and western blot. The toxicity was confirmed by comparing the growth curves. The growth curves of the transformed ΔTolA and ΔTolB cells had lower absorbances, mainly in the stationary phase, which indicates less viable cells in the samples. Furthermore, mCh-ColN appeared to affect the growth curves of transformed LMC500 cells when induced with higher amounts of IPTG. The intrinsic toxicity of the TolA binding domain of the ColN protein might prove an obstacle for the eventual comparison assay. The use of ColN might hamper the findings of one such assay as it affects the cells negatively itself. It will not be clear if it is ColN or the tested TolA binding molecule that is having a toxic effect on the cell. However, the toxicity of the TolA
Figure 6: Growth curves of mCh-ColN compared with untransformed LMC500 cells. A: LMC500 and LMC500 ColN growth curves. The growth curves of the uninduced and 5 µM IPTG induced
mCh-ColN transformants have a later log phase start while the 10 and 20 µM IPTG induced mCh-mCh-ColN transfomants have an altered log phase with a “pause” halfway through. B: LMC500 and BW25113, ΔTolA
and ΔTolB mCh-ColN growth curves. The ΔTolA and ΔTolB cells start their log phase later and have a
lower absorbance in the stationary phase for all amounts of induction. C: BW25113, ΔTolA and ΔTolB
mCh-ColN growth curves. The absorbance of all cell types decreases with higher amounts of induction by
IPTG. D: LMC500 and LMC500 dsbA-mCherry growth curves. The growth curves of the dsbA-mCherry cells have a lower absorbance in the stationary phase for all amounts of induction. Y-Axis is the absorbance and X-Axis is time in minutes.
binding domain of the ColN protein could lead to other methods of fighting gram negative bacteria. Furthermore, the localization of mCherry-ColN has not been as pronounced as the localization of sfTq2ox-TolB. This is partly caused by the quantum yield of mCherry being
22% and the quantum yield of sfTq2ox being 94% (Schaner et al., 2004; Meiresonne et al.,
2017). However, this assay requires a fusion protein capable of fluorescence in oxidizing environment of the periplasm. mCherry is the red fluorescent protein of choice in the periplasm as it is the brightest (Meiresonne et al., 2019). Another explanation could be the toxicity of the ColN domain of the mCherry-ColN fusion protein, which might disrupts the localization of the protein. However, the mCherry-ColN protein is proven to be intact in the cells, meaning the protein had to have been located in the periplasm. The growth curve of the LMC500 dsbA-mCherry cells indicate a slight decrease in viability for these cells. This could have contributed to the toxicity of the mCherry-ColN protein yet mCherry is known to have a low amount of cytotoxicity and the decrease in viable cells is greater than the drop seen for the dsbA-mCherry cells (Shemaikina et al., 2012).
Earlier research has shown that Colicin N kills cells by forming an ion channel in the inner membrane (Baty et al., 1990). Moreover, Colicin N has a TolA binding domain, which the protein uses to cross the outer membrane of the cell (Ragget et al., 1998). However, it was not yet known that this domain of the Colicin N protein has intrinsic toxic properties separate from its interaction with TolA. The toxicity of this domain of Colicin N needs to be better understood as it might introduce new methods of fighting gram negative bacteria. Future research could focus on the apparent toxic properties of the TolA binding domain of Colicin N. Furthermore, as the possibility of using Colicin N for the proposed comparison assay might prove difficult a new protein needs to be found that shares similar TolA binding properties but does not have the same toxic effect on the cells. This could be accomplished by using another member of the Colicin family like for example Colicin A as previous research has indicated that this protein has TolA binding properties (Penfold et al., 2012). The toxicity of the TolA binding domain of Colicin N might lead to new ways of combatting gram negative bacteria while a comparison assay using other members of the Colicin family is still a definite possibility.
Appendix 1
GFP-TolA sfTq2ox-TolB
A. Mean fluorescence total in all samples
0 5 10 15 20 25 GFP-TolA TolB LMC500 TolB GFP-TolA LMC500
B. Mean concentration total in all samples
0 1 2 3 4 5 6 GFP-TolA TolB LMC500 TolB GFP-TolA LMC500
When comparing the mean total fluorescence between the samples it is clear that sfTq2 is more visible in the samples than GFP. This is visible in Figure 7A as it shows that samples with only sfTq2 have about the same total fluorescence as samples with both GFP and sfTq2. Furthermore it shows that the total fluorescence of GFP is about as much as the background fluorescence of LMC500 cells. Figure 7B shows that there is little GFP-TolA present in the cells when compared to sfTq2-TolB which has a higher concentration. When comparing the fluorescent protein concentration of the sample LMC500 mut2GFP-TolA sfTq2ox-TolB to the
other samples it appears that the concentration is as high as the concentration of LMC500 mut2GFP and LMC500 sfTq2ox-TolB combined.
Appendix 2
A. Mean fluorescence total in all samples mCherry channel
Figure 7: Mean total fluorescence and Mean concentration of fluorescence protein of the samples. A. Mean fluorescence of all samples in GFP channel: The mean fluorescence of the GFP-TolA sfTq2-TolB
samples is similar to that of the LMC500 sfTq2-TolB samples. This is a clear sign that sfTq2 added more fluorescence to the samples than GFP. GFP-TolA samples show a similar fluorescence as LMC500. This would mean that the fluorescence caused by GFP is just slightly above the background fluorescence of LMC500 cells. B. Mean concentration of all samples: it is clearly visible that there is less GFP-TolA present in the samples when compared to sfTq2-TolB. This is because TolB is produced from a construct while TolA is chromosomal.
B. Mean concentration total in all samples mCherry channel 0 10 20 30 40 50 60 70 TolA ColN TolA Coln LMC500 0 2 4 6 8 10 12 14 TolA ColN TolA ColN LMC500
C. M ea n
fluorescence total in all samples GFP channel
D. Mean concentration total in all samples GFP channel 0 2 4 6 8 10 12 14 TolA ColN TolA ColN LMC500 0 2 4 6 8 10 12 14 16 18 TolA ColN TolA ColN LMC500
Figure 8A and 4B show that there is a difference in the amount of fluorescence present in the
mCh-ColN sample compared to the concentration of the protein. This difference could be caused by the low quantum yield of mCherry. However, the mut2GFP-TolA mCh-ColN samples show a high amount of fluorescence and fluorescent protein. The same high fluorescence of the mut2GFP-TolA mCh-ColN samples can be found in Figure 8C and 8D For the GFP channel of the mut2GFP-TolA mCh-ColN samples. The samples also show a decrease in fluorescence in the mut2GFP-TolA samples. This would mean that the fluorescence of the proteins would only be visible in mut2GFP-TolA mCh-ColN cells. However, the results of the TolB experiment in Figure 1 have shown that mut2GFP-TolA is present but just in a smaller amount.
Figure 8: Mean total fluorescence and Mean concentration of fluorescence protein of the samples in both channels. A. Mean fluorescence of all samples in mCherry channel: The fluorescence of the
LMC500 mut2GFP-TolA mCh-ColN sample is noticeably higher than the other 3 samples. This is unexpected as the LMC500 mCh-ColN channel does not appear to have a noticeably high fluorescence. The fluorescence in for this sample was expected to be higher because only mCherry is visible in this channel B.
Mean concentration of all samples in mCherry channel: the concentration of fluorescence proteins in the
cells is highest in the GFP-TolA mCh-ColN samples. This explains the higher fluorescence in these samples. However, the higher concentration of fluorescent protein in the LMC500 mCh-ColN samples is not displayed in the mean total fluorescence of the LMC500 mCh-ColN samples. This would mean there are fluorescent proteins present, which do not add to the total fluorescence of the sample. C. Mean
fluorescence of all samples in GFP channel: These samples show a similar difference in fluorescence
between the LMC500 mut2GFP-TolA mCh-ColN samples compared to the other samples with this sample having a noticeably increased fluorescence. D. Mean concentration of all samples in GFP channel: The concentration of fluorescent proteins is in line with the fluorescence observed in the samples. However, for the LMC500 mut2GFP-TolA samples this increased concentration is not shown in the fluorescence.
Appendix 3
A. Mean total fluorescence of all samples
0 50 100 150 200 250 dTolA dTolA ColN dTolB dTolB ColN
B. Mean total concentration of fluorescent protein of all samples
0 5 10 15 20 25 30 35 40 45 dTolA dTolA ColN dTolB dTolB ColN
Figure 9 indicates that the concentration of fluorescent protein and fluorescence of the
samples containing the mCh-ColN plasmid is as expected. The samples containing the plasmid have higher fluorescence and concentration of fluorescent protein. This will allow for better comparisons to be made between the microscopy images of the samples.
Figure 9: Mean total fluorescence and concentration of fluorescent protein of all ColN samples. A: Bar graph of mean total fluorescence of all samples. The total
fluorescence of the samples containing the mCh-ColN plasmid is higher than the samples without the plasmid. This is in line with what would be expected as the samples without plasmid do not contain any fluorescent protein. B: Bar graph of mean total
concentration of fluorescent protein of all samples. The concentration of fluorescent
protein is also higher in the samples containing the plasmids. This was expected as these samples have more fluorescence and the samples without the plasmid do not contain any fluorescent protein.
References:
Baty, D., Pattus, F., Parker, M., Benedetti, H., Frenette, M., Bourdineaud, J. P., ... & Lazdunski, C. (1990). Uptake across the cell envelope and insertion into the inner membrane of ion channel-forming colicins in E coli. Biochimie, 72(2-3), 123-130.
Cascales, E., Buchanan, S. K., Duché, D., Kleanthous, C., Lloubes, R., Postle, K., ... & Cavard, D. (2007). Colicin biology. Microbiol. Mol. Biol. Rev., 71(1), 158-229.
Durrani, F. G., Gul, R., Mirza, M. U., Kaderbhai, N. N., Froeyen, M., & Saleem, M. (2019). Mutagenesis of DsbAss is Crucial for the Signal Recognition Particle Mechanism in Escherichia coli: Insights from Molecular Dynamics Simulations. Biomolecules, 9(4), 133.
Gerding, M. A., Ogata, Y., Pecora, N. D., Niki, H., & De Boer, P. A. (2007). The trans‐envelope Tol– Pal complex is part of the cell division machinery and required for proper outer‐membrane invagination during cell constriction in E. coli. English (U.K.) Molecular microbiology, 63(4), 1008-1025.
Ghai, I. (2017). Quantifying the Flux of Charged Molecules through Bacterial Membrane
Proteins (Doctoral dissertation, IRC-Library, Information Resource Center der Jacobs University
Bremen).
Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. nature, 227(5259), 680-685.
Lazdunski, C. J., Bouveret, E., Rigal, A., Journet, L., Lloubès, R., & Bénédetti, H. (1998). English
(U.K.) Colicin import into Escherichia coli cells. Journal of bacteriology, 180(19), 4993-5002.
Li, G. W., Burkhardt, D., Gross, C., & Weissman, J. S. (2014). Quantifying absolute protein synthesis rates reveals principles underlying allocation of cellular resources. Cell, 157(3), 624-635.
Meiresonne, N. Y., Consoli, E., Mertens, L. M., Chertkova, A. O., Goedhart, J., & den Blaauwen, T. (2019). Superfolder mTurquoise2ox optimized for the bacterial periplasm allows high efficiency in vivo FRET of cell division antibiotic targets. Molecular microbiology, 111(4), 1025-1038.
Penfold, C. N., Li, C., Zhang, Y., Vankemmelbeke, M., & James, R. (2012). Colicin A binds to a novel binding site of TolA in the Escherichia coli periplasm.
Raggett, E. M., Bainbridge, G., Evans, L. J., Cooper, A., & Lakey, J. H. (1998). Discovery of critical Tol A‐binding residues in the bactericidal toxin colicin N: a biophysical approach. Molecular
microbiology, 28(6), 1335-1343.
Schneider, C. A., Rasband, W. S., & Eliceiri, K. W. (2012). NIH Image to ImageJ: 25 years of image analysis. Nature methods, 9(7), 671-675.
Shaner, N. C., Campbell, R. E., Steinbach, P. A., Giepmans, B. N., Palmer, A. E., & Tsien, R. Y. (2004). Improved monomeric red, orange and yellow fluorescent proteins derived from Discosoma sp. red fluorescent protein. Nature biotechnology, 22(12), 1567-1572.
Shemiakina, I. I., Ermakova, G. V., Cranfill, P. J., Baird, M. A., Evans, R. A., Souslova, E. A., ... & Chepurnykh, T. V. (2012). A monomeric red fluorescent protein with low cytotoxicity. Nature
communications, 3(1), 1-7.
Szczepaniak, J., Holmes, P., Rajasekar, K., Kaminska, R., Samsudin, F., Inns, P. G., ... & Kleanthous, C. (2020). The lipoprotein Pal stabilises the bacterial outer membrane during constriction by a
mobilisation-and-capture mechanism. English (U.K.) Nature Communications, 11(1), 1-14. Tsien, R. Y. (1998). The green fluorescent protein.
