Team:Caltech/Project/Results

Testing Export of Signaling Ligands

To determine whether or not the signaling ligands in these quorum sensing systems are exported, we Western blotted both the supernatant and cell lysate from cell cultures expressing the constructs we named pTG004 and pTG005 (lam system with 3xFLAG at the N-terminus of the lamD prepeptide sequence, and fsr system with 3xFLAG at the N-terminus of the GBAP domain within the fsrB protein sequence, respectively). The results are shown in Figure 1 and Figure 2 below.

Figure 1. Western Blot – Cell Lysate Liquid cultures of cells expressing the lam and fsr export systems and induced with aTc were spun down and lysed before running the Western blot on the lysate samples. Only the samples originating from cells expressing the fsr system (pTG005) appeared to have even produced proteins with the 3xFLAG in the domain. Samples’ levels of aTc induction are indicated above their respective lanes.

Figure 2. Western Blot – Supernatant Liquid cultures of cells expressing the lam and fsr export systems and induced with aTc were spun down, and the Western blot was run on the supernatant. Only the samples originating from originating from cells expressing the fsr system (pTG005) appeared to have exported proteins with a 3xFLAG domain within them. The gel ran crooked, making the size of the smallest fragment uncertain. Samples’ levels of aTc induction are indicated above their respective lanes.

After running the Western blots, we observed multiple bands in both the supernatant and cell lysate for the samples originating from cells expressing the fsr quorum sensing system, offering promise that the GBAP signaling ligand is synthesized and exported. The bands observed appeared indicate the presence of proteins roughly 30 kDa, 25 kDa, and 5 kDa in size both in the supernatant and in the cell lysate. The 30 kDa fragment appears to correlate to entire, uncleaved fsrB membrane protein (see Figure 2 in "The Experiments"), while the 25 kDa fragment appears to correlate to the cleaved membrane-bound component of fsrB (the third fragment). Due to the faintness of the Western C ladder in the cell lysate blot and the crookedness of the supernatant blot, it was difficult to ascertain the size of the smallest band present in the fsr samples, but it is hypothesized to be close to 5 kDa, which, would suggest it is the GBAP signaling molecule with the 3xFLAG, and possibly the remainder of the fsrB membrane protein, attached to it (the last two fragments). While the presence of bands in the Westerns demonstrates that the fsr export system’s genes are being expressed in E. coli, the presence of large bands in the supernatant as well is troublesome, since those bands should correspond to fragments still embedded in the membrane and should have been pelleted and not present in the supernatant. It is possible that some cells were not fully spun down, and so the fsrB fragments embedded in their membranes were still present in the supernatant sample.

In contrast, no bands at all appeared on the Western blot for samples originating from cells expressing the lam system. This indicates that, barring some experimental error, none of the fragments detailed in Figure 1 (in "The Experiments") were synthesized by the cells, much less exported, suggesting that the ligand synthesis and export components of the lam QS system is incapable of being implemented in E. coli.

Additionally, to test for presence of the signaling ligands, supernatants of liquid cell cultures expressing our constructed plasmids were analyzed for the presence of signaling ligand using liquid chromatography mass spectroscopy. Due to time constraints, we were able to test only cell cultures expressing an unflagged variant of pTG004 (the lam QS system). We removed the 3xFLAG domain from pTG004 by linearizing pTG004 without the 3xFLAG via PCR with the appropriate primers and then recircularizing the plasmid via a round-the-horn reaction. Cell cultures carrying the unflagged pTG004 were then grown in MOPS media under 250 nM aTc induction and then spun down to collect the supernatant, which was run through a filtration column to isolate the signaling ligand. The resulting sample was then analyzed by LC/MS. As mentioned in the "The Experiment" section, strong peaks were expected to be seen at m/z ratios of 260.1, 345.2, 373.2, and 577.3 after roughly 27 minutes flowing through the column (at a flow rate of 0.2 mL/min). We did not observe these peaks in our data (see Figure 3 and Figure 4), and the only peaks observed in our experimental sample with 250 nM aTc induction (data in Figure 3) were also present in our negative control grown without aTc induction (data in Figure 4), corresponding to background signal probably from the growth media. Since there did not appear to exist any significant peaks present in the induced sample but not in the negative control (the LC/MS plots for the induced sample were nearly identical to those for the uninduced sample), these data support the conclusion obtained from our Western blot experiment that the lam system is unable to synthesize and export its signaling peptide in E. coli.

Figure 3. Unflagged lam-Expressing Cell Supernatant with 0 nM aTc Induction At charge-to-mass ratios of 577.3 and 260.1, no notable peaks were observed, and the overall signal was noisy. At m/z of 373.2, a significant peak was noticed at 3.31 minutes. At m/z of 345.2, significant peaks were observed at 1.13 and 3.31 minutes.

Figure 4. Unflagged lam-Expressing Cell Supernatant with 250 nM aTc Induction At charge-to-mass ratios of 577.3 and 260.1, no notable peaks were observed, and the overall signal was noisy. Significant peaks were noticed at 3.31 minutes for m/z of 373.2 and 1.12 and 3.32 minutes for m/z of 345.2. However, no notable peaks were noticed at ~13 minutes at any of the expected m/z ratios (within the marked red box), making this graph nearly identical to that for the negative control (Figure 3).

Testing Scaffold Protein System

We successfully tested the scaffold system on the original circuit by Whitaker et al and showed that it works by using a plate reader to detect GFP levels. The graph below represents our results. At 0% arabinose, we see the baseline fluorescence of E. coli. By using the scaffold, we see twice as much GFP. This means that the two-component system is activated two-fold above baseline in the presence of the scaffold proteins.

---

Testing Combinatorial Promoter Logic

To determine the expression logic of the combinatorial promoters that we used, DH5α-Z1 cells that were transformed with our test constructs were grown for about 8 hours in MOPS media with different inducer concentrations, and then GFP fluorescence of these cells were measured (see "The Experiments"). We found that the A90 promoter construct (BBa_K1369000) acted as an AND-gate, the B83 promoter construct (BBa_K1369001) acted as an asymmetric-AND gate, and the D46 promoter construct (BBa_K1369002) acted as a single-input gate. These results were consistent with the findings reported by Robert Sidney Cox et al.

Figures 1 and 2. A90 Combinatorial Promoter The heatmap and barchart both indicate that the A90 combinatorial promoter was only active when sufficient amounts of both aTc and IPTG were present. This indicates that the A90 promoter behaves as an AND-gate

Figures 3 and 4. B83 Combinatorial Promoter The behavior of the B83 combinatorial promoter is a bit more complex. The promoter is still somewhat active when IPTG is absent but aTc is present. On the other hand, the promoter is largely inactive in all cases of aTc absence. However, expression is still greatest when both aTc and IPTG are present. Therefore, the B83 promoter acts as an asymmetric-AND gate, as it's behavior deviates slightly from an ideal AND-gate.

Figures 5 and 6. D46 Combinatorial Promoter The fluorescence values show that activity of the D46 promoter seem to be independent of arabinose concentration. Rather, expression under the D46 promoter seems to be a function only of aTc concentration. Therefore, the D46 promoter acts as a single-input gate.