Team:MIT/Design
Design
Overview
Background
Design Iterations
Basic Proof
Recombinase Switch
BCAA Sensor
Background
In order to determine how our probiotic should function, we conducted additional background research on the molecular basis of Maple Syrup Urine Disease (MSUD). We learned that mutations in the branched chain keto-acid dehydrogenase (BCKDH) enzyme complex, which is responsible for breaking down branched chain keto-acids (BCKAs), cause the buildup of branched chain amino acids (BCAAs) and branched chain ketoacids (BCKAs) to toxic levels in MSUD patients.
We decided to have our probiotic bacteria overexpress BCKDH to break down excess BCAAs before they are absorbed into the bloodstream. We were initially drawn to E. coli Nissle as a chassis, being a well studied probiotic strain of E. coli commonly used in synthetic biology and by iGEM teams. However, the human BCKDH enzyme is a very large complex consisting of multiple copies of each of 4 subunits encoded by different genes. E. coli lacks this enzyme complex or similar machinery to break down BCAAs, so we would need to import all these sequences into E. coli, which would be rather difficult and potentially unstable. As a result, we instead turned to looking at existing bacteria that already possess the BCKDH complex, and we found B. subtilis.
B. subtilis is a gram-positive bacterium commonly found in soil (1). B. subtilis is also a native inhabitant of the GI tract and a commonly used probiotic with proven health benefits (2). One unique feature of B. subtilis is that it is able to form spores and become metabolically dormant to survive in adverse environments, and many probiotics consist of B. subtilis in spore form. This allows it to survive the transit through the acidic environment of the stomach, and then germinate in the nutrient-rich small intestine.
The bkd operon in B. subtilis encodes the BCKDH enzyme complex. Combined with a transaminase and various other downstream enzymes, B. subtilis is able to fully metabolize BCAAs down to the basic substrates of central metabolism, such as acetyl CoA and succinyl CoA. As B. subtilis already has the bkd operon within its genome, we can simply use homologous recombination to replace the operon’s promoter with a constitutive one in order to upregulate it. We also identified various other genes in B. subtilis responsible for importing and exporting BCAAs in order to enhance B. subtilis’s ability to uptake and retain BCAA’s.
Below is a simplified model of the relevant genetic circuitry within B. subtilis. Our genes of interest are
- bcaP: a BCAA symporter
- ilvE and ilvK which encode BCAT: BCAA transaminase, which catalyzes the first step in the breakdown of BCAAs
- bkd operon: an operon that encodes branched chain keto-acid dehydrogenase (BCKDH), the enzyme complex that catalyzes the second step in the breakdown of BCAAs
- azlC and azlD: BCAA exporters in the azlBCDEF operon
Design Iterations
Our project had three main iterations:
Basic Proof of Concept
In this first iteration, we wanted to prove that overexpressing these critical genes could lead to a measurable increase in BCAA uptake, so we designed parts for constitutive expression of these genes.
Design The four main components of this system are
- Constitutive expression of the bcaP importer to import more BCAAs into the cell
- Constitutive expression of ilvE/ilvK transaminase to break down BCAAs into alpha keto-acids
- Constitutive expression of the bkd operon to break down alpha keto-acids
- Knockout of azlCD exporters to prevent BCAAs from leaving the cell
Testing To transform B. subtilis with our DNA, we planned to have our parts integrate directly into the genome through homologous recombination. To confirm that we were able to execute this, we also designed parts that expressed mApple, a red fluorescent protein, and KanR, a protein for kanamycin resistance. These same parts were also used in E. coli to check that our promoters were working as expected.
In total, our cloning and testing steps included the following experiments:
- Gibson cloning and traditional cloning to construct plasmid backbones
- Golden Gate to construct plasmids from our parts
- Colony PCR of both plasmid and genomic DNA to quickly check our Golden Gate products, as well as to check integration of our DNA into the B. subtilis chromosome
- Restriction digests to check our Golden Gate products prior to sequencing
- Transformation of E. coli to replicate our plasmid DNA
- Transformation into B. subtilis
Following successful transformation, we then assayed the ability of our B. subtilis to break down BCAAs, as well as B. subtilis's natural rate of BCAAs.
We performed the following experiments to assay B. subtilis:
- Test homologous recombination into B. subtilis with kanamycin selection marker
- BCAA concentration assay to determine the sensitive range of our assay and the best starting concentration of BCAAs with which to perform subsequent assays
- BCAA time point assay to compare the rate of BCAA breakdown between wild-type and probiotic B. subtilis
Recombinase switch for constitutive expression
This next design iteration keeps the basis of overexpressing genes but adds in additional regulatory controls. As we are engineering B. subtilis to overexpress many large protein complexes, it is critical to maintain fitness of the organism while it is being manufactured or traveling within the GI tract. After much discussion with our mentors, we decided to induce the expression of a recombinase to switch on constitutive expression of these proteins at the right time.
What is a recombinase? There are two types of recombinases: tyrosine recombinases (ex. Cre) and serine recombinases (ex. PhiC31). Recombinases recognize certain sequences in the DNA (ex. loxP and attP/B), and act on the DNA in between them. Generally, this takes the form of either an excision or a reversion of the intervening sequence, depending on the directions in which the flanking sites are facing. Excision is useful for permanently turning on or off a gene, while an inversion can function as a switch that can be adjusted depending on the condition of the cell.
Design We placed the Cre recombinase under the inducible control of a pTet/tetR/ATC system in order to turn on the constitutive expression of bcaP, ilvE/ilvK, and the bkd operon in the presence of doxycycline.
Cre recognizes specific loxP sites in DNA, in our case the lox71 and lox66 sites. There are two ways in which Cre can function, depending on how these sites are oriented:
- If the loxP sites face the same direction, the region of DNA between the sites is excised as a circular piece of DNA.
- If the loxP sites face opposite directions, the region of DNA in between the sites is inverted.
Using these properties, we have created a few different constructs for controlling constitutive expression: Constitutive expression of bkd operon:
- The native promoter is excised and replaced with a constitutive promoter
- The native promoter is initially in place, while the constitutive promoter is inverted. Inversion of both these promoters swaps their positions, placing the constitutive promoter in charge of the operon
Constitutive expression of bcaP and ilvE/ilvK Inversion of a reversed constitutive promoter results in constitutive expression
Testing In order to test the Cre recombinase switch, we first have to verify that Cre works as expected in B. subtilis. We designed a construct of mApple2 surrounded by lox71 and lox66 sites so we can test whether Cre excises mApple2 from the genome and note the time it takes for this process to occur. Additionally, we will continue to use the BCAA assay to test the rate of BCAA uptake into B. subtilis before and after adding dox to the system to induce the expression of Cre.
BCAA Sensor Using BRET and Optogenetic System
Our final conceptual design is a fully autonomous system capable of switching on constitutive expression of our genes of interest in response to fluctuating BCAA levels within B. subtilis. This is better than the recombinase approach as B. subtilis would be able to regulate its own gene expression and we would not need to induce it with dox or some other small molecule.
What is BRET? BRET is a system that utilizes one bioluminescent (donor) and one fluorescent protein (acceptor). Fluorescent proteins have defined excitation and emission wavelengths. After absorbing light with a wavelength within their excitation range, fluorescent proteins can then emit light in a separate wavelength, which can then be detected by a microscope. We can take advantage of this by coupling a fluorescent protein with a bioluminescent protein that emits light at the excitation wavelength. When these proteins are in close proximity, light emitted from the fluorescent protein may be detected.
Design While conducting background research on promoters and systems that respond to different levels of BCAAs, we came across oLIVe, a BCAA biosensor designed to exhibit FRET (BRET with a donor that is a fluorescent protein). oLIVe consists of livJ, a protein that binds BCAAs, with fluorescent proteins fused with linkers in regions of livJ that allow all three proteins to be folded properly and exhibit correct behaviours. The binding of a BCAA to livJ triggers a conformation shift that brings the fluorescent proteins in close enough proximity for FRET to occur.
In a human system, it is undesirable to have to excite the initial donor fluorescent protein at a similar wavelength via FRET, hence the decision to use BRET. In our theoretical system, RLuc8.6 luciferase (em 535 nm) and TurboFP635 fluorescent protein (em 635 nm) are inserted in opposite regions of livJ. Initially, only RLuc8.6 exhibits bioluminescence. When the livJ protein binds BCAA, livJ undergoes a conformational change, and the two fluorescent proteins are brought closer. This conformational change thus places RLuc8.6 and TurboFP635 in close proximity to each other and allows BRET to occur, and emission from RLuc8.6 excites TurboFP635, and allows TurFP635 to fluorescence instead of RLuc8.6.
Next, a separate optogenetic circuit may be used to detect the specific wavelength of light emitted by TurboFP635 and trigger gene expression. Here, we adopt the CcaSR system developed by the Voigt lab in E. coli, further adapted to B. subtilis by the Tabor lab at Rice University. CcaSR is able to turn on gene expression when induced by light at 535 nm and turn off expression at 670 nm. For our purposes, we need default activation at 535 nm to repress our genes of interest, and inactivation of CcaSR at 670 nm to cease repression of our genes, leading to expression. This is due to an inability to have a default ‘off’ system at 670 nm, as BRET cannot have a lower energy state FP (TurboFP635 emitting at 635 nm) excite a higher energy state (RLuc8.6 emitting at 535 nm).
Putting these two systems together, we would configure BRET with RLuc8.6 and TurboFP635 both fused to livJ. The default ‘on’ state with RLuc8.6 (em 535 nm) activity induces CcaSR expression of a repressor that represses constitutive expression of our genes of interest. When livJ binds BCAA, it undergoes a conformational change to excite TurboFP635 (em 635 nm) to inactivate expression of the repressor.
Testing To test that the optogenetic system works in B. subtilis, we must confirm multiple things:
- Whether CcaSR is inducible by correct wavelengths of light
- Whether a livJ fusion protein is capable of inducing BRET upon BCAA binding
- Whether the two systems can work in conjunction with one another, and how this system can be tuned to express our genes at the appropriate levels
References
- Castillo-Hair, S. M., Baerman, E. A., Fujita, M., Igoshin, O. A., & Tabor, J. J. (2019). Optogenetic control of Bacillus subtilis gene expression. Nat Commun, 10(1). 10.1038/s41467-019-10906-6
- Dallas, S. L., Xie, Y., Shiflett, L. A., & Ueki, Y. (2018). Mouse Cre Models for the Study of Bone Diseases. Curr Osteoporos Rep, 16(4), 466–477. 10.1007/s11914-018-0455-7
- Yoshida, T., Nakajima, H., Takahashi, S., Kakizuka, A., & Imamura, H. (2019). OLIVe: A Genetically Encoded Fluorescent Biosensor for Quantitative Imaging of Branched-Chain Amino Acid Levels inside Single Living Cells. 10.1021/acssensors.9b02067.s002