Design and Construction of Circuits
Characterization of U24
Characterization of U24 in HEK293T Cells
Determination of Optimal Concentration of Dox
Expression Time Course
Protein Half-Life of U24
The potential toxicity of U24
Tests in HEK293T cells
Expression of CARs in HEK293T cells
Downregulation of CARs in HEK293T cells
Tests in Jurkat cells
Characterization in Jurkat cells
Expression of CAR in Jurkat cells
Downregulation of TCR/CD3 complex in Jurkat cells
Reversible Effect of U24 on TCR/CD3 complex in Jurkat cells
Effect of U24 on CAR in Jurkat cells
Team:SYSU-CHINA/Demonstrate
Demonstrate
In order to achieve our goal, we first designed and constructed our circuits. Since we do not have a specific antibody for U24, to better detect the expression of U24, we added a HA epitope at the N-terminal of the U24 protein, which was demonstrated not to interfere with its functions (Sullivan and Coscoy, 2010) . To better visualize the expression of U24 under the control of tet-ON promoter, we positioned the HA-U24 downstream of a coGFP with a T2A peptide in between for bicistronic expression of coGFP as a fluorescence marker and HA-U24 as our protein of interest. For our control, we used a mutant of U24 whose proline 7-9 was replaced with alanine, which is not functional (Sullivan and Coscoy, 2010) . The entire coGFP-T2A-HA-U24 coding sequence was put downstream of tet-ON promoter. To constitutively express rtTA protein, the rtTA coding sequence was positioned downstream of a constitutive promoter, human ubiquitin C promoter (UbC promoter), and put into the same vector as GFP-U24 construct.
We also designed a vector for high expression of GFP-U24, with The entire coGFP-T2A-HA-U24 coding sequence positioned downstream of EF1α promoter.
To express CAR molecules, we positioned the CAR sequence downstream of EF1α promoter.
The vector backbone we used for construction is pHAGE and pEF1α, both of which are suitable for lentiviral packaging.
After constructing our vectors, we first tested our device in HEK293T cells for they are easy to transfect, and they possess some of the key components necessary for TCR internalization and recycling (Liu et al., 2000) .
In order to determine the optimal concentration of dox for induction, equal amount of HEK293T cells were seeded in 6-well plates and
transfected with ptetON-GFP-T2A-U24. Different concentration of dox was added to each well immediately after transfection. For
comparison, pEF1α-GFP-T2A-U24 was used as a positive control. 3 fluorescence images of each well were captured and analyzed with
ImageJ for fluorescence intensity 24h after introduction of dox. And equal amount of cells were harvested for western blots analysis.
In the absence of Dox, the transient transfected cells exhibited low level of fluorescence, the fluorescence intensity increased as
the concentration increase and reached a plateau. In agreement with the fluorescence data, the result from western blot analysis also
indicated a rapid increase on expression level at low level of dox and soon reached a plateau after the concentration exceeded
100ng/ml. Thus we reasoned that 100ng/ml is the lowest concentration needed for optimal induction. However, the basal expression
level was high, since only 3-fold increase at optimal condition compared to no induction.
Interestingly, although the predicted molecular weight of U24 protein is approximately 10kDa, two bands were detected at 20kDa,
consistent with previous research (Sullivan and Coscoy, 2010) . This suggested that U24
undergoes extensive post-translational modifications. However, the types and functions of these modifications remain unknown.
Figure 1 Optimal concentration of Dox for treating cells.
B: Relative fluorescence intensity of the images in A(3 replicates, data not show) after analyzed by ImageJ.
C: Western blot result of the cells harvested in A. D: Normalized protein level of the images in C after analyzed by ImageJ.
Since our project aims to construct a safe switch for CAR T cells, the response time for our device is critical in order to achieve
effective management of adverse effects. Because our device relies on transcription regulation, the delay between addition of drugs
and the U24 reaching its effective concentration may severely limit its application.
In order to determine the expression time course after adding dox, equal amount of HEK293T cells were seeded in a 6-well plate and
transfected with ptetON-GFP-T2A-U24. Dox was added at the same time in all wells and equal number of cells in each well were
harvested at the indicated time for western blot analysis. The result was analyzed with ImageJ and the amount of protein of interest
was normalized using β-actin as internal controls.
We observed an increase of HA-tagged protein level over the time of 24h and reached a plateau at around 6h, confirming a relatively
fast response. However, the fold change was small that a less than 2-fold change was observed, indicating insufficient induction. Due
to the time limits, we were unable to conduct additional experiments and more repeats are needed to further validate the results.
Figure 2. Expression time course of U24 in HEK293T cells
harvested at different time points: 0h, 2h, 4h, 8h, 12h, 24h.
B: Normalized protein level of the images in A after analyzed by ImageJ.
In order to determine the degradation rate of U24, equal amount of HEK293T cells were seeded in a 6-well plate and transfected with
ptetON-GFP-T2A-U24. Dox was added at the same time in all wells and cycloheximide (CHX), a protein synthesis inhibitor, was added to
each well 24h post transfection. Equal number of cells in each well were harvested at indicated time pointe for western blot
analysis. The result was analyzed with ImageJ and the amount of protein of interest was normalized using β-actin as internal
controls.
A decrease in immunoblotting signal was observed, indicating the degradation of U24 protein after protein synthesis inhibition. The
results were then analyzed with Mathmetica, and a fitting curve was drawn using the half-life exponential decay model. The half-life
was determined to be around 2 hours, indicating fast turnover of the protein, which is critical for our project that the T cells
functions can be restored soon after the drug is removed.
Figure 3. Protein half-Life of U24
B: Normalized protein level of the images in A after analyzed by ImageJ.
Previous research on U24 indicated that U24 targets the endosomal recycling pathway. Although the effects are relatively specific,
transferrin is shown to be downregulated by U24. Thus it is possible that U24 may induce cytotoxicity when expressed due to
downregulation of transferrin or other unknown surface protein. To test this hypothesis, we transfected HEK293T cells with ptetON-
GFP-T2A-U24 and Dox was added after transfection. The viable cell count and dead cell count were determined using typhan blue
staining. As a control, we transfected HEK293T cells with ptetON-GFP-T2A-U24(P7-9A) or Lipofectamine 3000 only with Dox added after
transfection, to rule out toxicity induced by transfection, Dox or the expression vector itself. Three parallel experiments were
conducted.
Growth curves were drawn for cells transfected with each vector. No significant different was seen in the viable cells count and dead
cells count at each time point (Student's T test, double tailed, P>0.05), indicating that U24 has little, if any, toxic effect on
cells and can potentially be safely used as an add-on for CAR T cells. However, direct tests on T cells are needed in order to draw a
persuasive conclusion.
Figure 4. The potential toxicity of U24 is neglectable
To detect the expression of CARs in HEK293T cells, we first transfected the cells with pEF1α-CAR and the expression was detected using biotin-protein L and R-phycoerythrin(PE)-conjugated streptavidin staining as previously described (Zheng et al., 2012) and flow cytometry. Untransfected HEK293Ts were used as control.
Figure 5. Successful Expression of CARs in HEK293T cells
As indicated by the flow cytometry, HEK293T expresses CAR at cell surface. The protein L staining method is a relatively new method for staining of a wide variety of CARs, utilizing the binding property of protein L to light Kappa chain to the immunoglobulin, which also presents in the scFv. Since this method only detect surface CARs, it is suitable for our subsequent experiments that intracellular and surface CARs need to be distinguished.
To determine whether U24 has an effect on CARs, we co-transfected HEK293T cells with pEF1α-CAR and ptetON-GFP-T2A-U24. Dox was added at 0h post transfection and not added as control. The cells were stained and subject to flow cytometry analysis 24h post transfection.
Figure 6. No significant effect of U24 on downregulation of CARs in HEK293T cells
B: The distribution of fluorescence intensity of both PE and GFP. The first picture stands for sample001, the second one stands for sample010 and the last one stands for sample 011.
However, we did not observe a significant change of the surface level of CARs comparing GFP-positive and negative cells in the ptetON- GFP-T2A-U24 population, nor comparing the ptetON-GFP-T2A-U24 with ptetON-GFP-T2A-U24(added Dox). A possible reason is that the transfection rate is lower in co-transfection experiment, as indicated in the large population of double negative cells. Since it is impossible to distinguish CAR-negative cells for CAR-internalized cells, the large number of untransfected cells may mask any effect of U24. Stable cell lines, preferably from a single colony, expressing CARs may be needed for this experiment, due to the time limits, however, we were yet unable to establish such cell lines.
Jurkat cell line is a T cell leukemia cell lines commonly used in immunology research. It was used in both the original U24 experiments (Sullivan and Coscoy, 2008, 2010) and research for CAR T cells (Wu et al., 2015) . Thus we also characterize our device in this cell line, providing more convincing evidence that U24 can be used to improve CAR T safety.
To detect the expression of CARs in Jurkat cells, we first packaged lentiviral vector pEF1α-CAR due to low transfection efficiency of Jurkat using Lipofectamine 3000. Transduced Jurkat cells with pEF1α-CAR were stained using biotin-protein L and PE-conjugated streptavidin and subject to flow cytometry. As control, untransduced Jurkat cells were stained with the same method.
Figure 7. Successful Expression of CARs in HEK293T cells
As indicated by the flow cytometry, HEK293T expresses CAR at cell surface.
To verify the ability of U24 to downregulate TCRs, and its potential use in TCR T therapy, we transduced Jurkat cells with lentiviral vector ptetON-GFP-T2A-U24 and ptetON-GFP-T2A-U24(P7-9A) as control. Dox was added 48h prior to flow cytometry. The cells were stained with PE-conjugated anti-CD3 monoclonal antibody (UCHT-1, LifeTechnologies, Thermo Scientific) and subject to flow cytometry.
Figure 8. U24 downregulates TCR/CD3 complex in Jurkat cells
B: The distribution of fluorescence intensity of both PE and GFP. The first picture stands for sample003, the second one stands for sample009 and the last one stands for sample 010.
In consistent with previous research (Sullivan and Coscoy, 2008, 2010) , a clear pattern was observed for the ptetON-GFP-T2A-U24 transduced cells that the CD3 level decreased as GFP level increased, with only small fraction of cells were double positive. In contrast, ptetON-GFP-T2A-U24(P7-9A) transduced cells exhibited no such tendency. Our result demonstrated that our device can be potentially used in TCR T therapy as a safe switch.
After verifying the ability of U24 to downregulate TCRs, we next determined whether the effect of U24 is reversible and confirming the notion that our device is indeed a reversible safe switch. We transduced Jurkat cells with lentiviral vector ptetON-GFP-T2A-U24 and ptetON-GFP-T2A-U24(P7-9A) as control. Dox was added 48h prior to flow cytometry. Dox was removed from 1 well of ptetON-GFP-T2A- U24 transduced cells and continued in the other well. The cells were stained with PE-conjugated anti-CD3 monoclonal antibodyand subject to flow cytometry.
Figure 9. Reversible effect of U24 on TCR/CD3 complex in Jurkat cells
B: The distribution of fluorescence intensity of both PE and GFP. The upper left picture stands for sample018, the upper right one stands for sample019,
the lower left one stands for sample020 and the lower right one stands for sample 021.
In consistent with our previous result, a clear pattern was observed for the ptetON-GFP-T2A-U24 transduced cells that the CD3 level
decreased as GFP level increased, with only small fraction of cells were double positive. In contrast, ptetON-GFP-T2A-U24(P7-9A)
transduced cells exhibited no such tendency.
Interestingly, when dox was removed, the GFP level was decreased and CD3 level was restored compared to the experiment group with
continued Dox. Our result demonstrated that our device is reversible, and T cells functions can be potentially restored after
removing drugs.
After verifying the ability of U24 to downregulate TCRs, we next determined To determine whether U24 has an effect on CARs. We cotransduced Jurkat cells with lentiviral vector pEF1α-CAR and lentiviral vector ptetON-GFP-T2A-U24 or ptetON-GFP-T2A-U24(P7-9A) as control. Dox was added 48h prior to flow cytometry.. The cells were stained with biotin-protein L and PE-streptavidin and subject to flow cytometry.
Figure 10. Effect of U24 on CAR in Jurkat cells
B: The distribution of fluorescence intensity of both PE and GFP. The first picture stands for sample003,
the second one stands for sample004 and the last one stands for sample 005.
We only observed a modest change in the surface level of CARs comparing ptetON-GFP-T2A-U24 with ptetON-GFP-T2A-U24(P7-9A). A possible reason is that the transduction rate is lower in co-transfection experiment, as indicated in the large population of double negative cells and lower fluorescence cell counts comparing co-transfected cells and single transduced cells (data not shown). Since it is impossible to distinguish CAR-negative cells for CAR-internalized cells, the large number of untransduced cells may mask any effect of U24. Stable cell lines, preferably from a single colony, expressing CARs may be needed for this experiment, due to the time limits, however, we were yet unable to establish such cell lines.
Liu, H., Rhodes, M., Wiest, D.L., and Vignali, D.A. (2000). On the dynamics of TCR:CD3 complex cell surface expression and
downmodulation. Immunity 13, 665-675.
Sullivan, B.M., and Coscoy, L. (2008). Downregulation of the T-cell receptor complex and impairment of T-cell activation by human
herpesvirus 6 u24 protein. Journal of virology 82, 602-608.
Sullivan, B.M., and Coscoy, L. (2010). The U24 protein from human herpesvirus 6 and 7 affects endocytic recycling. Journal of
virology 84, 1265-1275.
Wu, C.Y., Roybal, K.T., Puchner, E.M., Onuffer, J., and Lim, W.A. (2015). Remote control of therapeutic T cells through a small
molecule-gated chimeric receptor. Science 350, aab4077.
Zheng, Z., Chinnasamy, N., and Morgan, R.A. (2012). Protein L: a novel reagent for the detection of chimeric antigen receptor (CAR)
expression by flow cytometry. Journal of translational medicine 10, 29.
