This is an experiment associated with the Sporenet 3 memo. It was originally available on github, but is now also available here. I never completed this experiment, mainly because of synthesis costs. I am continuing this effort with Sporenet experiment 4, which will be online soon!
The goal is to put a landing pad for any plasmid (in particular, pOpen_v3) into Bacillus subtilis. We will accomplish that using the Bacillus subtilis REG120 strain, which has inducible supercompetence, manAP knockout for negative selection, and a degU knockout for more effective transformations [1]. To disrupt the current antibiotic resistance gene, ermC, we have to figure out where it came from to get the sequence. In the REG120 paper, there is mention of the original strain being derived from a transformation of pJOE6577.1 into B.subtilis 168[2]. pJOE6577.1 was derived from pSUN346.1, and the pSUN series plasmids ermC marker was derived from pDG1730[3]. Fortunately, the sequence for pDG1730 is on genbank under the accession number U46199.1.
In order to get the ermC marker in ref3, the gene was amplified with the following primer set:
s5069,AAAAAAGAATTCGATATCAGATCTACGCGTTAACCCGGGC s5070,AAAAAACAATTGAATCGATTCACAAAAAATAGG
The CDS of that region we can safely assume will be present in the strain REG120, and will be our target for integration. Dr. Zeigler from the BGSC recommnends at least 500bp of homology when doing transformations, so we will use that number as well. We don't want to have multiple resistance markers in our end strain, so we will delete a 275bp region of the ermC gene during recombination. This could potentially be repaired by environmental DNA, but this would cause the excision of our recombinant DNA, so we won't worry about repair of that resistance marker. 275bp was deleted because some of the ermC gene had extremely low GC, which would cause problems with synthesis.
pOpen_v3's ampicillin resistance marker (bla), with it's promoter, is 966bp long. So, we will cheat a little and have 483bp of homology for integration of pOpen_v3 into the *B.subtilis REG120* genome.
In ref2, pJOE6743.1 is the plasmid used for creating clean and marker-free deletions (also used by the Mini Bacillus project for this use case). For the Mini Bacillus project, plasmid pJOE8739.1 was created to allow for easy insertion of DNA in Escherichia coli using ccdB prior to being transformed into Bacillus subtilis.
A selection marker is necessary to integrate the landing pad, but it is undecided whether we want to excise this positive selection marker during integration of plasmid DNA or not. On one hand, it is extremely attractive to distribute strains with no antibiotic resistance marker, but on the other hand, this possibly increases the chance of environmental contamination by a huge amount in under-resourced labs. We plan on using Kanamycin since it was removed from the WHO Model List of Essential Medicines and the WHO Model List of Essential Medicines for Children.
WHO Model List of Essential Medicines for Children
We are going to use the Kanamycin resistance gene aph3 from pBSc291K. Most integration vectors with kanamycin resistance seem to be derived from pDG780, but the sequence for that plasmid isn't available, so this one will have to do.
To review, we have 4 different moving parts:
1. The Bacillus subtilis REG120 (ermC) homologous recombination sites (where the construct will enter our strain)
2. The ampicillin resistance (bla) homologoous recombination sites (where any plasmid will get integrated into our strain)
3. The manP negative selection marker
4. The kanamycin positive selection marker
Generally, we are going to be looking for two different constructs:
1. ermC_1 - bla_1 - kanR - manP - bla_2 - ermC_2
2. ermC_1 - kanR - bla1 - manP - bla_2 - ermC_2
The first will allow for completely antibiotic-free distribution of DNA. However, if while testing the system we find that the antibiotic-free methods are difficult for labs to handle, we will use the second construct, where plasmid integration will not remove the kanamycin resistance site.
Both constructs come out to 5147bp, which is larger than our Twist synthesis budget. In order to get around this limitation, we are synthesizing 3 different parts which recombine to create our two constructs:
1. prefix_1 : ermC_1 - bla_1 - kanR
2. prefix_2 : ermC_1 - kanR - bla_1
3. suffix : manP - bla_2 - ermC_2
In order to physically put the parts together, flanking the prefixs and the suffix are BbsI sites, which will allow prefix - suffix combinations. There are also sticky sites that allow for "suffix - prefix" combination, which allows for concatemerization of the constructs in a "- prefix - suffix - prefix - suffix - ..." pattern, which has been shown to increase transformation efficiency in Bacillus subtilis.
increase transformation efficiency in Bacillus subtilis through concatemerization
[1] Construction of a Super-Competent Bacillus subtilis 168 Using the PmtlA-comKS Inducible Cassette
[3] Characterization of a Mannose Utilization System in Bacillus subtilis
These are genbank files for the plasmids, synthesis, and constructs to complete this project. They were originally in .dna format, but converted to .gb format using TeslaGen's Open Vector Editor.