Proof of Concept
Bacillus subtilis Cholesterol Uptake Tests
In order for our pathway to work in our desired chassis, the chassis has to be able to uptake cholesterol to feed in.
To solve this, we took a two-pronged approach. First, it has been published in literature that Bacillus mojavensis, a phylogenetically close cousin of B. subtilis, has the capability of uptaking cholesterol1. Given this, we hypothesized that B. subtilis could also uptake cholesterol on its own. However, in case it did not, we cloned in a suspected cholesterol transporter (MFS transporter) gene from Eubacterium coprostanoligenes.
We then ran a cholesterol uptake assay by incubating B. subtilis, B. subtilis with MFS, and E. coli as a negative control with cholesterol over a 24 hour period (fig 1.). B. subtilis appeared to uptake cholesterol by itself without the need of a cholesterol transporter.
Figure 1. Cholesterol uptake assay
Cholesterol was incubated with B. subtilis, B. subtilis with MFS, and E. coli over a 24 hour period, and the difference between cholesterol remaining in media at 0 hours and 24 hours is taken.
Individual in vitro enzyme tests
In order to prove that our concept of this enzymatic pathway could turn cholesterol into the derivatives leading up to and including coprostanol, we tested each enzyme individually in vitro.
ismA
ismA is the first enzyme in our pathway, and has been already validated by Kenny et al. in a paper published in Cell 2. However, we wanted to validate this for ourselves as well.
Despite difficulties in purifying out the ismA protein, the protein was eventually isolated and tested in an in vitro setting. The signal was quite poor, but when ismA was incubated with cholesterol, a cholestenone peak was observed in the correct position. Under optimized buffer conditions, the signal would likely be considerably higher, but an in vivo test with the ismA expressed in a cholesterol-uptaking bacterial chassis would likely yield more important and desirable results.
Figure 2. ismA Reaction Tests in vitro.
Cholesterol was incubated with ismA for 16 hours overnight in pH 6.25 potassium phosphate buffer. Buffer contained 0.2% Triton X-100 and 5% ethanol to solubilize cholesterol.
Concentration of ismA protein was 0.5 mg/ml of each in the final reaction mix. 200 μM of cholesterol was added.
AKR1D1
AKR1D1 is the second enzyme in our pathway, and is actively involved in the bile acid synthesis pathway, catalyzing metabolism of cholestenone-like molecules (7a-hydroxy-4-cholesten-3-one) into coprostanone-like molecules (7a-hydroxy-5b-cholestan-3-one).
While AKR1D1 did not initially show promising catalysis efficiency and results, the ability of the protein to catalyze our desired cholestenone to coprostanone reaction became apparent after proper experimental process optimization.
Figure 3. AKR1D1 Reaction Tests in vitro.
Cholestenone was incubated with AKR1D1 for 1 hour in pH 6.5 potassium phosphate buffer. Buffer contained 0.2% Triton X-100 and 5% ethanol to solubilize cholestenone.
Concentration of AKR1D1 protein was 0.5 mg/ml of each in the final reaction mix. 200 μM of cholestenone was added.
AKR1C4
AKR1C4 is the third and final enzyme in our pathway, and is actively involved in the bile acid synthesis pathway, catalyzing metabolism of coprostanone-like molecules (7a-hydroxy-5b-cholestan-3-one) into coprostanol-like molecules (5b-cholestane-3a,7a-diol).
AKR1C4 initially showed excellent catalysis efficiency of coprostanone to coprostanol, which only increased after experimental process optimization. This gave us the confidence that we could do a one-pot reaction with AKR1D1 and AKR1C4 together to further verify their ability to complete the pathway.
Figure 4. AKR1C4 Reaction Tests in vitro.
Coprostanone was incubated with AKR1C4 for 1 hour in pH 6.0 potassium phosphate buffer. Buffer contained 0.2% Triton X-100 and 5% ethanol to solubilize coprostanone.
Concentration of AKR1C4 protein was 0.5 mg/ml of each in the final reaction mix. 200 μM of coprostanone was added.
One-pot in vitro enzyme tests
In order to prove that our repurposed enzymes worked together, we decided to run the enzymes we did not know would work (due to lack of literature on the subject) together in vitro incubating with cholestenone. This reaction would theoretically yield a coprostanol product (our end goal), in addition to coprostanone.
Our GCMS testing results indicated that coprostanol was indeed produced, validating our enzyme model and tests.
Figure 5. AKR1D1 and AKR1C4 One Pot Reaction Tests in vitro.
Cholestenone was incubated with AKR1D1 and AKR1C4 for 16 hours overnight in pH 6.25 potassium phosphate buffer. Buffer contained 0.2% Triton X-100 and 5% ethanol to solubilize cholestenone.
Concentration of AKR1D1 and AKR1C4 proteins were 0.5 mg/ml of each in the final reaction mix. 200 μM of cholestenone was added.
Explore
