July 3rd, 2024
After mulling over some ideas, I think I came up with a solid counterselection system to compliment sucrose positive selection via cscA — CITRATE! This idea stems from the need to have two different selection markers to be able to use the iterative Golden Braid System I am so fond of.
Addgene: Orzaez Lab GoldenBraid 2.0 Kit
In short, the system uses four plasmids and each pair (alpha and omega) can recursively clone together into one of the plasmids of the opposite pair. Two alpha plasmids’ transcription units can be chained together when cloned into an omega plasmid. The resulting omega plasmid carrying the two alpha transcription units can then be cloned along with a second omega plasmid into a chain in an alpha plasmid. This cycle can iterate forever for as long as your plasmid can replicate and DNA prep process can maintain plasmid size without shearing.
Alpha 1 + Alpha 2 → Omega 1 Omega 1 + Omega 2 → Alpha 1
This system relies on the ability to counter select between the alpha and omega plasmid pairs. Typically this is done by alternating between ampicillin and kanamycin antibiotic markers, but this not only goes against the core thesis of the project but also doubles the exotic/regulated chemicals one would need to source to do this sort of iterative cloning themselves. The need for a second selection system is now critical. Enter citrate!
E. coli Citrate Use Cell Biology - EvoEd : EvoEd (evo-ed.org)
In the Citric Acid Cycle we all suffered to learn in our school days, citrate ions play a crucial role keeping our cells alive by cycling to replenish ATP, NADH, and FADH2 molecules via bond energy harvesting. This cycle is absolutely vital for the survival of many living things, but in E. coli can only import citrate from the environment in anaerobic conditions.
E. coli Citrate Use Cell Biology - EvoEd : EvoEd (evo-ed.org)
The way in which citrate comes into the E. coli cell is via the citT citrate/succinate antiporter. It trades a succinate ion (produced downstream in the Citric Acid cycle) with exogenous citrate and is the only way citrate can enter. The citT gene is expressed in anerobic conditions and repressed in the presence of oxygen. The idea I want to try is simple:
Express citT constitutively via a plasmid to maintain selection of cells carrying said plasmid on media where citrate is the sole carbon source. This has two likely constraints:
Citrate metabolism in aerobic conditions has been shown to spontaneously evolve (thanks Richard Lenski for your diligent 30+ year experiment).
Strongly expressing transmembrane proteins which hook into cell wall may be bad.
The first can be mitigated by completely knocking out the native citT gene and the second constraint could be solved by using a weaker promoter. The latter situation may already be solved since I am using the kanR promoter as the driver for the selection marker anyway and I don’t think it’s extremely strong; need to do some lit review on this but worth trying regardless.
To construct this citrate selection plasmid, I first need to identify the citT coding sequence. In the Escherichia coli NEB Turbo Chromosome (Genbank Accession NZ_CP053605.1) there is a region annotated as citT and a UniProt Accession matches this coding sequence perfectly.
I then designed primers in Primer3Plus using the following template:
tatGGTCTCtgctt-YOUR GENE HERE-gctttGAGACCata
where the Bsa1 cut site is in capital letters. The trailing TAT is just to add space to the primers. The resulting primers to make a L0 part of citT is:
L0-citT-F 5’-TATGGTCTCtaATGTCTTTAGCAAAAGATAATATATGGAAAC-’3
L0-citT-R 5’-TATggtctcaAAGCTTAGTTCCACATGGCGAGAATCGGC-’3
The resulting PCR amplicon will then have Bsa1 sits flanking them using the AATG and GCTT MoClo/GoldenBraid overhangs which turns this into an L0 part. The native gene has the Nco1 and Kpn1 sites inside which wont be aproblem but muddies up some compatibility. If this works well I’ll have the citT gene properly resynthesized with some codon scrambling to avoid recombination events with native gene later….but we’ll cross that bridge when we get to it.
Golden Braid, like many other modular cloning standards, has a grammar of overhangs that coalesce to form seamless cloning.
This figure is overwhelming at first but it makes a lot of sense once you get used to it. At the very top we have the subdivisions of parts and there are short 4bp motifs between each of the smallest sub-parts. Those are the overhangs we add next to our Bsa1 cut sites to make the part syntactically sound. In our citT part, we add the AATG overhang (note the ATG start codon is part of this overhang so we only add an additional A) on the 5’ end and GCTT on the 3’ end. This formats the part as a basic coding sequence, the 13-14-15-16 (CDS) white bar at the top of the graph, otherwise known as an L0 (L-zero) part in MoClo grammar. This is the simplest expression part with no further subdivisions added or accounted for and will represent the bulk of my library.
As an example, a promoter compatible with such a part would have the GGAG 5’ overhang and the AATG 3’ overhang. Note how the 3’ of the promoter part matches the 5’ of the CDS part. A simple terminator would have GCTT on the 5’ and CGCT on the 3’. Any backbone plasmid, regardless of alpha or omega, once cut with Bsa1 or BsmB1, will have GCTT on the 5’ and CGCT on the 3’ end, which are the leftmost and rightmost overhangs on that Golden Braid grammar figure above. You can clone in any number of parts as long as they have overhangs that match the next part in the grammar and the final Transcription Unit is flanked with the CGCT and GCTT overhangs. This allows for massive amount of mix and match potential and the creation of an entire library. The basic part syntax matches other established standards for somewhat easy onboarding if need be. This was your crash course in Golden Braid Cloning!
Returning to the previous conversation, the above figure is the game plan:
PCR the citT gene from NEB Turbo cells, adding the Bsa1 cut sites to either end along with the appropriate overhangs as mentioned earlier.
Amplify the eforCP and ori section of my red chromoprotein plasmid, adding matching overhangs and Bsa1 cutsites as step 1 but facing inward so it can seamlessly clone.
Digest both fragments with Bsa1 and ligate with T4 Ligase overnight.
Transform and select on M9-Citrate (after establishing transformation protocols)
MAKE LEMONADE (cloning)!!!
Can’t wait for primers to arrive!!!
1 comment