- Engineering Success -

Taxis

---

Design

Engineer in foreign chemotaxis under Nd upregulation =>

1. WT must not natively show positive or negative migration
2. Easily integrable => Use system from gram negative (same as M. extorquens).

1. => Test M. extorquens native taxis to the below E.coli chemoattractants.
2. => E. coli MCPs (ligand sensing component of taxis), which are:
   • Tar†: Aspartate, Maltose*
   • Tsr†: Serine
   • Tap: Ribose*, Galactose*
   • Trg: Dipeptides*
   • Aer: O₂ Experimental setup too complicated
† Abundant in E.coli, * bind MCP via periplasmic binding proteins

---

Build

No build for WT characterisation

---

Test

We were interested in chemotaxis, and therefore had to decide on a motility assay and optimise for M. extorquens. We chose a qualititive assay that gives a robust binary result within an hour of inoculation, and uses small plates such that it does not waste resources.

---

Learn

Unfortunately, we got no positive results, even from positive control. Therefore, we moved on to a protocol that i) can be troubleshot more easily, and ii) better matches the setup for our proof of concept.

---

Design

The new design facilitates taxis over the course of the growth timeframe, and also allows a quantitative measurement of the cells’ propensity to move towards the chemoattractants.

---

Test

---

Learn

We see colonies mostly ascribing to the third option – no motility. This was highly concerning, as it suggested that the strain of M. extorquens we had had been domesticated such that it had lost motility. We knew we needed to somehow reintroduce it, but that using genetic engineering would be a largely blind process.

---

Design

We opted to perform a series of successive swim plates, each time taking bacteria from the farthest point that it grew from inoculation, and reinoculating with that.

---

Build

This directed evolution was repeated until we saw sufficient motility to be characterisable. We grew up two motile colonies, made glycerol stocks of them and named them ‘Mex mot A’; and ‘Mex mot B’. We used Mex mot A for all future experimentation and transformation.

Test

We reattempted the assay in the previous cycle with our Mex mot A strain.

Learn

We successfully observed negative results for the E. coli chemoattractants. This gave us a choice of chemoattractants, so we were able to pick one that is:

• Easy to engineer
• Cheap at industrial scale

Design

We chose Tar for aspartate sensing, because:

• The MCP binds aspartate directly => only one component => simple.
• Abundant in E. coli => well characterised activity.
• Asp is a chemically simple amino acid => cheap and readily available.

For Nd regulation, we inserted the intergenic region upstream of M. extorquens' xoxF1, which is characterised as being robustly upregulated in lanthanides (La³⁺, Ce³⁺).

---

Build

---

Test

Using aspatate in plug, perform Ex (2) from previous cycle with:

1. Mex mot A
2. Mex mot A + P_xoxF1-tar

---

Learn

We saw some limited movement towards Asp in the transformant and not in the untransformed Mex mot A strain at 0.2% w/v agar.

=> Successful endowment of aspartate taxis in Nd-supplemented media.

Bioaccumulation

---

We wanted to optimise the capacity of M. extorquens for lanthanide uptake, so we designed a bioaccumulation strategy, as below.

Therefore, we planned to engineer in LutH and LanM under inducible, titratable regulation, to test for optimal expression levels.

---

Build

With our promoter choice, even leaky expression of lutH was observed to be lethal to cells.

---

Test: Uptake

1. Successive growths in Nd-supplemented media
   • Grow, then remove majority of cells by centrifugation, and save a sample of media.
   • Allow regrowth repeatedly until cells fail to grow.

   This assay measures the reduction in Nd concentration, which is in theory more sensitive to conditions where the majority of Nd is uptaken.

2. Overnight uptake
   • Grow to late exponential in absence of Nd, then induce overnight with Nd.
   • Centrifuge and wash cells repeatedly, collecting the cells for sampling after the last wash.

   This assay measures the uptake into cells directly, and allows standardisation by cell density. In theory it is more sensitive to conditions where cell uptake of Nd is limited.

Limited use of ICP-MS caused us to decide to run the assay 2) only.

---

Learn

We found WT uptake was extremely efficient at the concentrations of Nd that we used. Therefore, we did not redesign the expression construct for weaker expression of LutH, because it was no longer considered an effective optimisation. This choice was supported by our model by this stage.

lanM did result in increased uptake per cell, which supports our idea of using it for neodymium sequestration. Nevertheless, it was not deemed necessary to optimise the design of lanM for uptake, because the WT was sufficiently efficient.

---

Test: Stress

1. Growth curves at variable [Nd]
   • High throughout
   • Highly quantified
   • Limited access to plate reader as undergraduate students
2. Viability plates
   • Magnitudes of growth observable
   • No specialist equipment

We decided to go with viability plates, because we expected our results to be spread across several magnitudes of neodymium concentrations.

---

Learn

Inconsistent results across similar experiments cast doubt on results, but initial analysis is promising. More repeats are required for a robust claim of success, but LanM does appear to reduce stress manyfold, in line with predictions from our model.