Part:BBa_K1766010

BAR construct 2

The bacterial antigen receptor (BAR) was created in the hope of making a receptor able to signal binding to an antigen. By making a chimera between a histidine kinase (EnvZ) and a affibody molecule we would have a receptor capable of detecting protein biomarkers. The construct would then use the use the EnvZ-OmpR two-component regulatory system to signal biomarker binding and activation of the BAR.

Structure

The osmoregulator EnvZ was chosen a the scaffold for the BAR as it is a well characherized protein that has been used in other many iGEM projects. One particularly interesting aspect of EnvZ is that functional chimeras between it and other histidine kinases have been made. Specifically protein containing a HAMP domain responsible for signal progression seem to be suitalble for fusion proteins which is refered to as a “control cable”[1]. For example, Cph8 (BBa_K1017301 and BBa_I15010), a Cph1-EnvZ chimera is already available in the registry. Also in a reasent article [2] the activation and inactivation of EnvZ by moving aromatic residues was studied and later employed on a Tar-EnvZ chimera [3].

One troubling aspect on EnvZ is that the periplasmic domain has not been properly characherized or the structure determined. Structure prediction software recoognized what seems to be a PAS domain that could be involved in the signaling of EnvZ. In this PAS domain we found regions that seemed suitable for inserting a biomarker binding protein.

For binding the portein biomarker we choosed the small and stable affibody molecule. Affibody molecules consists of three helixes containging changeable residues for creating affinity to a certain protein. This has been used to create affibodies towards manny different proteins such as the HER2 (human epidermal growth factor receptor) protein. This variablilty and the stability of the structure would mean that the affibody could be switched to detect other proteins as well.

Construct 2 was created by replacing part of the PAS domain with the affibody Z:HER2:342. Unlike construct 2 a GSGGGSG linker was first added to the C-terminal of the affibody. This resulted in 62 residues in EnvZ were exchanged for 66 residues in the affibody. The PAS Beta sheet/Coil I region was kept as this was thought to effect dimerization between constructs.

Vision

BAR construct 3 was created to respond to presence of human epidermal growth factor receptor (HER2). This would lead to activation of the responce regulator OmpR inherent in E.coli. In practice this has not yet been proven.

Sequence and Features

Assembly Compatibility:

• 10
  COMPATIBLE WITH RFC[10]
• 12
  INCOMPATIBLE WITH RFC[12]

  Illegal NheI site found at 387
  
• 21
  INCOMPATIBLE WITH RFC[21]

  Illegal BglII site found at 236
  
• 23
  COMPATIBLE WITH RFC[23]
• 25
  COMPATIBLE WITH RFC[25]
• 1000
  COMPATIBLE WITH RFC[1000]

Experiments

Expression - Western blot

Method: To check protein expression BAR3_V5 was cloned into a pRD400 backbone and transformed in E.coli (TOP10). Western blot was performed on lysates. Antibodies targeted the V5 tag (anti-V5) and affibody molecules (polyclonal anti-affibody). The membrane were also stripped and incubated with anti-DnaK to compare expression.

Figure 1:Western blot on EnvZ wildtype and BAR constructs. pEB5 used as negative control.

Result: The construct can be observed in figure 1 as a homodimer with an approximate size of 100kDa (black triangle) and monomer with an approximate size of 50kDa (black circle). Further analysis, using an anti-Affibody antibody, confirmed that our construct strains are expressing the desired BAR constructs (black star) (Figure 1B). The faint bands at around 25kDa (black square) are presumably caused by some unspecific binding. The loading control DnaK shows in all WB lanes proving that all wells have been loaded equally with whole cell lysates of the different strains.

Is our construct in the inner-membrane?

As our chimeric receptor derives from the osmoregulator EnvZ which is situated in the inner membrane, we are suspecting that our BARs are going to have the same subcellular location. To confirm its localisation in the inner membrane we performed a FACS analysis comparing spheroplast (gram-negative cells with removed outer membrane and cell wall) to untreated cells.

Method:The FACS analysis, used the following E. colistrains: TOP10-pRD400-EnvZ_V5 and the TOP10-pRD400-BAR2_V5 which were induced in log phase (OD_{600 nm} ~0.3) with 25 µM IPTG and incubated overnight. One sample of each strain was treated in order to become a spheroplast whereas one sample of each strain was left untreated. All samples were then incubated with a fluorescently-labeled Alexa 488 anti-affibody antibody. Finally, the samples were analysed using Beckman Coulter Gallios FACS machine. Machine was set to a 11 V (FSC), 993 V (SSC) and 354 V for the direct Alexa488 anti-affibody antibody.

Figure 2: FACS analysis of pRD400-BAR2_V5 and pRD400-EnvZ_V5 expressing TOP10 strains and their corresponding spheroplasts.

Result: Cells have been resuspended in PBS and then gated accordingly for the E. coli population. Interestingly, the spheroplast populations seems to be differentiate from the broader untreated E. coli population (Figure 2B). Therefore, the gate setting could be kept the same for all samples test. The gated population has been consequently analyzed for their Alexa488 levels. Hereby, we used the untreated strains as control to adjust the voltage accordingly to have signal peak on the left side of the histogram. We observe that only the TOP10-pRD400-EnvZ Spheroplast samples express a distinct positive colony, whereas the TOP10-pRD400-BAR2 spheroplast population show in a perfect overlay with the negative population for Alexa488 (Figure 2A).

For this observation, we do have two plausible explanations: (1) We created only a spheroplast with the pRD400-EnvZ-V5 strain, hence, the positive population is due unspecific binding or (2) we have mixed up the samples between the spheroplast pRD400-EnvZ and pRD400-BAR2. We tried to reproduce the results but would require more test to verify the sub-cellular location.

Is our construct responsive to changes in osmolarity?

EnvZ is sensible to osmolarity. In our three EnvZ-Affibody chimeras (BAR1 (K1766009), BAR2(K1766010) and BAR3(K1766011)) a part of the periplasmic region has been replaced by the affibody. Therefore it was expected that our constructs would not be reacting to changes in osmolarity. EnvZ was therefore used as a control for the constructs.

Method: For this experiment BAR1 was put in the pRD400 backbone and transformed into E. coli EPB30 (strain with genomic OmpR regulated CFP expression + constitutive YFP expression). pEB5 which is an empty plasmid in the same pRD400 backbone was used as a negative control and EnvZ (K1766008) as a positve control. The samples were induced in log phase with 25 µM IPTG and cultured overnight in high and low osmolarity media (0% versus 15% sucrose). Tests were performed with triplicates.

Figure 3:CFP fluorescence of the different samples in low osmolarity medium (0% sucrose) and high osmolarity medium (15% sucrose). ‘*’ signifies significant P value. ns : P > 0.05 (not significant), ‘*’ : P ≤ 0.05, ‘**’ : P ≤ 0.01, ‘***’ : P ≤ 0.001.

Results:In figure 3, the graph shows that BAR2 has a non-significant increase in CFT fluorescence. This might prove that our construct has become insensitive to changes in osmolarity. YFP values obtained during this experiment were off and inconclusive.

Can BAR bind to SK-BR-3, a HER-2 expressing breast cancer cell line?

BARs are situated in the inner membrane as they derive from the inner membrane standing osmoregulator EnvZ. However, the integration of an Affibody molecule in the periplasmic region should allow it that the BAR expressing bacteria can bind to HER-2 expressing cancer cells when the bacterial outer membrane and cell wall has been removed. We were really gratefull that our collaborator, the iGEM Team ETH Zurich, took up on this task. The findings from these experiment represent an important checkpoint towards a functional system.

Method: They expressed all three EnvZ-Affibody chimeras (BAR1 (K1766009), BAR2(K1766010) and BAR3(K1766011)) and incubated them with a HER-2 expressing cell line (SK-BR-3). BAR expression was intiated in all three strains using ITPG induction. In order to efficiently visualize our bacteria at the cell surface, they wanted to introduce GFP into our bacteria strains which unfortuntely failed. Therefore, they continued with our unstained bacteria.

Figure 4:Testing our three BAR expressing constructs (BAR1 (BBa_K17009), BAR2 (BBa_K1766010) and BAR3 (BBa_K1766011)) for their binding capacity towards HER-2 expressing breast cancer cells. BAR expression was initiated by ITPG induction.

Results: The results represented in figure 3 show no clear interaction between the SK-BR-3 cells and our BAR-expressing bacteria. This might be due to too low contrast between bacteria and background. Therefore, we cannot conclude whether our bacteria is capable of binding membrane-bound HER-2. However, we see a stronger retainment of bacteria in all ITPG-induced cells which might hint towards a certain interaction of HER-2 and our chimeric protein. To sum it all up, we could not show an interaction between HER-2 expressing cancer cells and BAR-expressing bacteria. Other techniques will be needed to investigate on this question in order to make a final conclusion.

References

[1] Parkinson, J. S. (2010) Signaling mechanisms of HAMP domains in chemoreceptors and sensor kinases. Annu. Rev. Microbiol. 64, 101− 122
[2] Nørholm M. H. H., von Heijne G., and Draheim R. R. (2014) Forcing the Issue: Aromatic Tuning Facilitates Stimulus-Independent Modulation of a Two-Component Signaling Circuit. ACS Synth. Biol. 2015, 4, 474−481
[3] Yusuf R. and Draheim R. R. (2015) Employing aromatic tuning to modulate output from two-component signaling circuits. Journal of Biological Engineering, 9:7