Before fully test the STEP system, we first divided the whole transducer into two sections: the extracellular domain of ligand receptor & transmembrane domain, and the intracellular domain of transcription factor & protease. After testing both parts we were able to examine and optimise the whole system in both HEK293T and HeLa cell lines.

A critical factor for the system to work is the ability of the transducers to correctly locate onto the cell membrane. To verify that we designed a test of bimolecular fluorescent complementation (BiFC), split the N- and C-terminal of ECFP (BBa_E0020) at the A155 site and directly attached them to transmembrane linkers of the two VEGF-STEP chains.

After confirming the validity of our split ECFP in E. coli BL21 (see our parts page), we transfected HeLa cells using the two chains and added VEGF 24 hours later. After 12 hours of induction and 4 hours under low temperature for maturation, we successfully observed cyan fluorescence under microscope (Figure 1).
It is obvious that a complementary ECFP would not form if the two chains failed to locate to cell membrane or if they couldn’t bind under the effect of VEGF. Thus we can infer from the results that our VEGF-scFv together with the signal sequence and the transmembrane domain is ready for further tests.

Another important part of our system is the transcription factor on the intracellular domain. After viewing information of different transcription factors in previous parts, we finally decided to use the advanced tTA as our transcription factor, together with a pTight promotor.

We first co-transfected HEK293T cells with a plasmid encoding tTA with an EGFP protein and another plasmid with pTight promotor and a mCherry reporter, and observed both strong green and red fluorescence after 48 hours of expression (Figure 2).
Then, for further tests of the whole system, we constructed stable cell lines of both HEK293T and HeLa, with downstream plasmid of pTight and mCherry.

Now that we had successfully tested both sections of STEP system and built stable cell lines, we were able to carry out further experiments concerning the whole system.

To see if the whole STEP system works we need to co-transfect the two plasmids encoding TC and PC chains. We want to find a best TC/PC ratio and total quantity of transfection.

At the beginning we transfected HEK293T cells with different TC/PC ratio. We used VEGF-STEP as an example. We controlled the final concentration of TC and PC respectively and changed it of the other chain (Figure 3). Cells were transfected 12 h after seeding and VEGF was added 24 h later. Photos were taken 24 h later under 561 nm exciting light.

From the results above we could conclude that TC/PC = 12 is a relatively ideal ratio. So we chose this ratio for the following experiments. Our model also confirmed the results and suggested that better dynamic range may be achieved by reducing the total quantity of plasmids. Thus we changed the total quantity, and obtained best effect at 0.75 fold of the original quantity (Figure 4).
To conclude the results of the overall VEGF test in HEK293T, the best transfection ratio is around 12, and the best quantity of plasmids encoding the two chains is around 2.44 μg per ml.

We then use HEK293T cells to examine our D-Dimer-STEP (Figure 5a). It’s quite surprising that this system has extremely low background noise (data not shown), and reached a 3.17-fold dynamic range at the TC/PC ratio of 12. This results suggests that D-Dimer-scFv may has a better characteristics for STEP system than VEGF-scFv, and we plan to quantify related features of them and further examine their influence to the quality of the whole system.

Apart from tests in HEK293T cells, we also tested the TC/PC = 12 ratio in HeLa cells, but didn’t get expected outcome (Figure 5b). This means that transfection ratios and quantities must be measured and modelled separately in different cell types.

As is suggested by our model, there must be a certain set of conditions that will lead to the largest dynamic range as well as the expression level. These conditions will be further explored in our future experiments, and be applied in real clinical cases.