Image Map


SUBGROUP MEMBERS: Kathryn Brink, Andrew Chen, Erik Ersland, Christian Richardson, Alex Smith

Attributions: Lyla Atta (Experiments), Kathryn Brink (Experiments, Animations), Kathryn Brink (Animations),
Erik Ersland (Descriptions), Alexa Garcia (Parts)

B-cell receptors (BCRs) are naturally occurring, transmembrane protein complexes that consist of a membrane-bound antibody (IgM) and some associated proteins (CD79A and CD79B). Given that the variable region of the antibody can be specific for any of a large number of antigens, we designed a B-cell receptor to bind beta-amyloid plaques (a biomolecular hallmark of Alzheimer's disease). Once bound, activated receptors instigate intracellular signalling, which can then be manipulated to diagnose the disease.


B-cell receptors (BCRs) are multiprotein immune receptors found exclusively on the surface of B cells. The BCR multiprotein complex is centered around a membrane-bound IgM antibody. When the antibody binds to an extracellular antigen, receptors dimerize resulting in the phosphorylation of the intracellular tails of CD79A and CD79B by the tyrosine-protein kinase Lyn. In response, another cofactor, spleen tyrosine kinase (Syk), is recruited to the receptor and phosphorylated, initiating a signalling cascade that results in the proliferation of the activated B cells. This receptor is important in clonal selection of B cells during human immune response.

For this project, we engineered a BCR to respond to beta-amyloid plaques, the hallmark of Alzheimer's disease. This task was accomplished by using a beta-amyloid specific variable region [derived from Gantenerumab] in the membrane-bound IgM antibody. Our design was based on that of the Tango system [1], which capitalizes on the interaction between TEV protease (TEVp) and its cleavage site (TCS), an amino acid sequence for which the protease has a high affinity. A TEV cleavage site was used to link a transcriptional activator (Gal4VP16) to the intracellular tails of BCR accessory proteins CD79A and CD79B, and the receptor’s cofactor, Syk, was fused to TEV protease. Thus, when the modified receptor activates upon binding its antigen, beta-amyloid, Syk-TEVp fusion protein is recruited, bringing TEVp in close proximity to its cleavage site. This proximity of TEVp to TCS results in the cleavage of the transcriptional activator from the receptor releasing it to activate downstream gene circuits.

The engineered BCR we developed binds beta amyloid with high specificity and releases a transcriptional activator upon binding, making it an extremely valuable tool in the detection of Alzheimer’s Disease. More importantly, the IgM antibody that determines what the receptor binds can be easily swapped out as can the transcription factor the receptor releases. This means that the receptor we developed can bind to any molecule that an antibody can be produced against and it can release any transcription factor in response to the binding of the target molecule. This modularity allows this receptor to be generalized to almost any extracellular sensing making it an invaluable part of any synthetic biologists toolkit.

return to top


Experiment 1:

  Localization of receptor to the cell membrane

In our first experiment, we aimed to determine if the engineered B-cell receptor components (CD79A, CD79B, IgM Heavy Chain, and Kappa Light Chain) were able to assemble to form a receptor complex and then localize to the cell membrane. Since the beta-amyloid oligomers characteristic of Alzheimer's disease accumulate in the extracellular matrix of the brain, it is important that the receptor membrane localize so that it can detect these plaques outside the cell.

To determine the localization of the receptors, we immunostained using IgM specific antibodies. We analyzed the immunostained samples in two ways. The first was through flow cytometry analysis. This method enabled us to determine whether the antibodies bound to the outside of our cells, which would indicate that the B-cell receptor's IgM component had reached the membrane. We also used confocal microscopy to visualize the localization of our receptor inside our cells by permeabilizing the cells and incubating them with anti-IgM antibodies.

For samples that were analyzed using flow cytometry, we transiently transfected HEK293 cells with plasmids encoding constitutive expression (hEF1a promoter) of the engineered B-cell receptor components along with hEF1a:mKate2 (constitutive red fluorescent protein) as a transfection marker. The transfection marker provides an indication of approximately how many plasmids are uptaken by a particular cell, which helps to connect plasmid number to observed output levels. We then treated cells with anti-IgM antibodies conjugated to Alexa Fluor 488 (yellow fluorescent dye). By measuring yellow output relative to red output using the flow cytometer, we hoped to be able to compare plasmid number to anti-IgM antibody binding, where high levels of red fluorescence (many plasmids) would correspond to high levels of yellow fluorescence (high levels of antibody binding, meaning a high level of BCR surface expression).





Determining membrane localization was complicated by bleedthrough from mKate into the yellow fluorescence channel. Samples were stained with Alexa fluor 488 (yellow) conjugated anti-IgM antibodies and evaluated using flow cytometry. Red fluorescence refers to the amount of mKate (transfection marker) in the cells whereas yellow fluorescence measures the amount of anti-IgM binding (indicating the presence of B-cell receptors on the cell surface). (A) Untransfected, unstained HEK293; (B) HEK293 transfected with dummy DNA, stained; (C) HEK293 transfected with mKate, stained; (D) HEK293 transfected with mKate and BCR components, stained.

In our first trial of this experiment, we saw a significant increase in yellow fluorescence between untransfected cell populations (A,B) and transfected ones (C,D). However, we saw similar amounts of yellow fluorescence between cells that were transfected with just hEF1a:mKate2 and those transfected with both hEF1a:mKAte2 and the B-cell receptor DNA. Additionally, there was a very strong one-to-one correlation between yellow and red fluorescence - a tighter distribution than would generally be expected for this kind of experiment. This led us to believe that our results were actually stemming from bleedthrough of the mKate2 fluorescent protein into the FITC channel that we used to detect yellow fluorescence. If both the red and yellow channels registered signal from the same protein, this would explain the tight, one-to-one correlation.

To address this problem, we decided to repeat the transfection without using a transfection marker since we determined that all of the fluorescent proteins that we had available to us would produce the same, if not a greater, bleedthrough effect.

Flow cytometry demonstrates synthetic B-cell receptor membrane localization in HEK293 cells. Cells were stained with Alexa Fluor 488 conjugated anti-IgM antibodies. Typical HEK293 cells do not express B-cell receptors, whereas Ramos cells (a positive control) are derived from B cells and do express B-cell receptors. (A) HEK293 transfected with dummy DNA, stained; (B) HEK293 transfected with synthetic B-cell receptor, stained; (C) Ramos, unstained; (D) Ramos, stained

By transfecting without a transfection marker, we were able to identify a population of HEK293 cells that were expressing our B-cell receptor on the cell membrane. However, less than 2% of transfected cells showed this kind of expression, suggesting that either our transfection efficiency was low or that expressing our receptor resulted in harmful effects to the cells.

Our second method of determining membrane localization was using confocal microscopy. To do this we once again transfected HEK293 cells with plasmids encoding constitutive expression of the receptor and a constitutive color transfection marker (hEF1a:eYFP). Our choice of transfection marker here was not important since any fluorescence would be quenched when the cells were fixed. Instead, we used the transfection marker to determine if we had a high enough transfection efficiency to proceed with immunostaining. After transfecting, we then fixed the samples and stained them with the same antibodies that we used for the flow cytometry analysis and added DAPI to stain the nucleus for better visualization of the cells.

Fluorescent microscopy suggests membrane localization of synthetic B-cell receptor in HEK293 cells. Cells were stained with Alexa Fluor 488 conjugated anti-IgM antibodies and DAPI was used as a nuclear stain. (Left) Untransfected HEK293 control, stained; (Right) HEK293 transfected with synthetic B-cell receptor, stained

In the resulting microscopy images, we saw a clear increase in yellow fluorescence between cells that were transfected with the receptors and those that were not. We also saw halos of yellow around the blue nuclei in the transfected cells, suggesting that the receptor may have been localizing to the cell membrane. However, we also observed some cytosolic expression, potentially from receptor being held in endoplasmic reticulum during processing. Results of further experiments suggested that the receptors might have been getting overexpressed, given the large mass of receptor DNA that we were transfecting and the fact that we were using a strong constitutive promoter to express the receptors.

Based on the results from flow cytometry and microscopy, we concluded that the B-cell receptor was reaching the surface of the transfected HEK293 cells.

return to top

Experiment 2:

  Beta-amyloid binding to the receptor

In this experiment we aimed to determine whether or not the B-cell derived receptor in our system was in fact binding to beta-amyloid oligomers. To do this, we transfected HEK293 cells with plasmids encoding our receptor components and hEF1a:eBFP2 (a transfection marker). We then treated the cells with biotinylated beta-amyloid oligomers and red Alexa Fluor 594-conjugated streptavidin. If the receptor bound to the beta-amyloid, the streptavidin would, in turn, bind to the biotin on the beta-amyloid oligomers leading to a higher level of red fluorescence. Similar to the first experiment, we analyzed the cells using both flow cytometry and confocal microscopy, looking for increased red fluorescence in cell populations that were transfected with the receptors.

Flow cytometry was inconclusive about beta-amyloid binding. Cells were incubated with biotinylated beta-amyloid and Alexa fluor (red) conjugated streptavidin. Red fluorescence indicates beta-amyloid binding and blue fluorescence is the transfection marker (eBFP). (A) HEK293 transfected with eBFP; (B) HEK293 transfected with eBFP, stained with beta-amyloid and streptavidin; (C) HEK293 transfected with eBFP and synthetic B-cell receptor, stained with streptavidin; (D) HEK293 transfected with eBFP and synthetic B-cell receptor, stained with beta-amyloid and streptavidin

The flow cytometry results that we obtained through this experiment did not lead us to conclusive results as to whether our receptor was binding beta-amyloid oligomers. Though there were an increased amount of cells showing red fluorescence in after beta-amyloid/streptaviding staining, this increase did not correlate with an increase in plasmid count (as measured by the blue transfection marker). Additionally, the truncation in blue fluorescence observed in cell populations transfected with the B-cell receptor (C, D) suggested that at a certain level of expression our receptor was becoming toxic to the cells or inducing too high of a metabolic load.

Fluorescent microscopy to determine beta-amyloid binding to synthetic B-cell receptor was inconclusive. Cells were incubated with oligomerized biotinylated beta-amyloid and red Alexa fluor conjugated streptavidin. Blue indicates DAPI nuclear staining and red indicates Alexa fluor. (Left) Untransfected HEK293 control, stained with streptavidin and beta-amyloid; (Right) HEK293 transfected with synthetic B-cell receptor, stained with streptavidin and beta-amyloid

Like cytometry, microscopy results for beta-amyloid binding were also inconclusive. Both samples of cells expressing the receptor and untransfected cell samples showed some degree of red fluorescence (indicating the presence of streptavidin), but there was no clear difference between the two and no particular localization was observed.

Given the cytometry and microscopy results from our beta-amyloid binding experiment, it is unclear whether beta-amyloid does in fact bind our synthetic B-cell receptor. However, this inconclusive result does not necessarily keep us from testing activation of the receptor, since anti-IgM antibodies have also been shown to cause receptor dimerization, activating the BCR.

return to top

Experiment 3:

  Evaluating relative levels of Syk-TEVp and endogenous Syk

In this experiment, we wanted to compare the levels of endogenous cofilin and Syk-TEV protease (TEVp) expressed under an inducible promoter with different levels of induction. Different levels of Syk-TEVp expression were achieved using an rtTA/TRE system, where rtTA is a transcription factor activated by doxycycline (a small molecule) that activates genes under the regulation of a TRE promoter. High concentrations of doxycycline correspond to high levels of gene expression in this system. By examining differences in expression levels between endogenous and exogenous Syk, we hoped to gain insight into what level of Syk-TEVp expression would lead to the best signal:noise ratio of exogenous to endogenous Syk. We transfected HEK293 cells with DNA encoding inducible expression of our Syk-TEVp fusion construct and hEF1a:eYFP (a transfection marker) as well as hEF1a:rtTA, which is required for the doxycycline-inducible activation of our Syk-TEVp construct. We added different concentrations of doxycycline to various cell populations, and subsequently analyzed the cell lysates by probing for Syk in a Western blot analysis.



Exogenous Syk-TEVp under high levels of doxycycline induction is expressed at comparable levels to endogenous Syk in HEK293 cells. Cell lysates were examined from various cell populations: untransfected HEK293 cells, Ramos cells, HEK293 cells transfected with dummy DNA, HEK293 cells transfected with hEF1a:eYFP only, and HEK293 cells transfected with hEF1a:eYFP and TRE:Syk-TEVp under varying levels of doxycycline (dox) induction. Primary antibody probes against Syk, GAPDH (a loading control), and eYFP (a transfection efficiency control) were used along with IR dye conjugated secondary antibodies, and the blots were imaged using an IR scanner. The two copies represent different blocking conditions: (A) blocked in 5% BSA and (B) blocked using Odyssey Blocking Buffer.

We used an antibody specific to Syk to probe for both Syk and Syk-TEVp. The difference in size between the endogenous Syk (72.1 kDa) and the exogenous Syk-TEVp (100.5 kDa) allow us to distinguish between the two on the Western blot, and hence compare their relative quantities. We also probed for GAPDH (37 kDa) and eYFP (27 kDa). We used GAPDH as a loading control to allow us to normalize for the amount of protein that was loaded in each lane and we used eYFP to normalize for variations in transfection efficiency.

From this experiment, we were able to determine that HEK293 cells express Syk endogenously at a level similar to the level of exogenous Syk-TEVp expression we observe when our system is induced with high levels of doxycycline (1000-2000nM). Though Ramos cells were intended to provide a positive control for endogenous Syk expression, the amount of protein loaded for the Ramos samples in the blots was only sufficient to produce a very faint band.

return to top

Experiment 4:

  Quantifying cleavage levels with non-activated receptor




Non-activated cells were examined to determine basal TEVp cleavage levels. eBFP was used as a transfection marker in all samples, and TEVp cleavage was measured via Gal4UAS:mKate activation (Gal4UAS is a Gal4VP16-inducible promoter and mKate production results in red fluorescence). Black lines indicate a control containing Gal4UAS:mKate, TRE:Syk-TEVp (inducible Syk-TEVp), and all components of the BCR, where neither CD79A nor CD79B is fused to Gal4VP16. Cleavage in non-activated cells was examined for three different variations of transcription factor (Gal4VP16) placement: (A) CD79A-Gal4VP16 and CD79B, (B) CD79A and CD79B-Gal4VP16, and (C) CD79A-Gal4VP16 and CD79B-Gal4VP16. Red lines indicate controls for each of these variations where no Syk-TEVp was transfected. Blue lines indicate samples where Syk-TEVp was transfected (and its expression induced in the rtTA/TRE system using 2000nM doxycycline).

To determine the frequency of non-specific activation of our system, we tested our system's output in cells that were not activated (they were incubated with neither anti-IgM antibodies nor beta-amyloid oligomers). We used constitutive eBFP (blue fluorescence) as a transfection marker and Gal4UAS:mKate (red fluorescence) as a reporter for system activation. Gal4UAS is a mammalian promoter whose activation is dependent on binding by Gal4VP16, the transcription factor that we fused to CD79A and CD79B components of the BCR, meaning that the production of a red fluorescent signal should be related to cleavage of the transcription factor from the receptor by Syk-TEVp. Three different transcription factor arrangements were examined: CD79A-Gal4VP16 and CD79B, CD79A and CD79B-Gal4VP16, and CD79A-Gal4VP16 and CD79B-Gal4VP16.

As expected, across most variations of transcription factor placement (such as B, C), the highest levels of red fluorescence that we observed occurred in samples where Syk-TEVp was expressed at high levels (with 2000nM doxycycline induction under the regulation of rtTA and TRE). Also unsurprisingly, we observed almost no red fluorescence in cells where the BCR components did not contain any fusion proteins with our transcriptional activator (i.e., where CD79A and CD79B were used without any fusions to Gal4VP16). To our surprise, however, we saw some red fluorescent output in cells that were not transfected with Syk-TEVp (in A and C, at levels comparable to those observed under high Syk-TEVp induction). This suggests that, in some cases, TEV protease cleavage is not required for our system to produce an output. To explain this phenomenon, it may be possible that, rather than localizing to the cell membrane, fusion proteins of CD79A and CD79B with Gal4VP16 are recruited to activate Gal4UAS:mKate.

From this non-activated receptor experiment, we learned that high levels of Syk-TEVp result in system activation even in the absence of stimulus and that our system can produce high levels of output even in the absence of TEV protease. While investigating these phenomena in more detail could produce useful information for the optimization of our system, we were not able to pursue this line of inquiry given our limited time frame.

return to top

Experiment 5:

  Cleavage levels in active versus non-activated receptor





Cells activated with anti-IgM antibodies show lower levels of fluorescent output relative to non-activated cells. Red lines indicate cells activated with anti-IgM antibodies and blue lines indicate non-activated cells. Red fluorescent output was used to quantify system activation based on signaling between Gal4VP16 and Gal4UAS:mKate. tagBFP was used as a transfection marker. Various conditions were assayed which involved transfecting different masses of the B-cell receptor components and varying levels of Syk-TEVp (achieved by adding differing levels of doxycycline). Some of the conditions that were tested included: (A) 25ng of each receptor component, 10nM doxycycline; (B) 12.5ng of each receptor component, 1nM doxycycline; (C) 6.25ng of each receptor component, 0nM doxycycline; (D) 6.25ng of each receptor component, 0nM doxycycline.

Our last experiment compared output from activated (using anti-IgM antibodies) and non-activated versions of our system. This experiment used a similar design to that of Experiment 4 (including the same transfection marker (eBFP) and fluorescent output (mKate); however, the only variation of transcription factor placement that was tested was CD79A-Gal4VP16 and CD79B. To find the best signal to noise ratio for our system, we varied two different parameters: the mass of receptor components that was transfected and the amount of Syk-TEVp present (which was altered using different doxycycline concentrations). A sampling of these cases are presented here (A,B,C,D). In every case that we tested, the amount of system output was higher for the non-activated cells than for the activated cells, which was the opposite of what we were expecting. Though the exact mechanism behind this discrepancy remains unclear, it is possible that activation of our receptor causes secondary, unintended effects that affect the cell's ability to produce our output. Further investigation will be required to determine the mechanism behind this effect.


[1] Gilad Barnea, Walter Strapps, Gilles Herrada, Yemiliya Berman, Jane Ong, Brian Kloss, Richard Axel, Kevin J. Lee.The genetic design of signaling cascades to record receptor activation. PNAS (2007) Print

return to top


full parts list available here
  • pENTR Gantenerumab Variable Human Ig-M Heavy Chain
  • pEXPR hEF1a:Gmab Heavy
  • pENTR Gantenerumab Variable Human Ig-M Light Chain
  • pEXPR hEF1a:Gmab light
  • pENTR CD79A
  • pEXPR hEF1A:CD79A
  • pENTR CD79B
  • pEXPR hEF1a:CD79B
  • pENTR CD79A-TCS-Gal4VP16
  • pEXPR hEF1a:CD79A-TCS-Gal4VP16
  • pENTR CD79B-TCS-Gal4VP16
  • pEXPR hEF1a:CD79B-TCS-Gal4VP16
  • pENTR Syk
  • pEXPR TRE:Syk
  • pENTR Syk-TEVp
  • pEXPR TRE:Syk-TEVp
  • pENTR Lyn
  • pEXPR hEF1a:Lyn
  • pENTR Lyn-TEVp
  • pEXPR TRE:Lyn-TEVp

Team Members: Erik Ersland, Kathryn Brink, Christian Richardson, and Alex Smith