ORBIT: FAQs

Why didn't I get any colonies?

Observing zero colonies from an ORBIT experiment can happen for many reasons, most of which are easily solved. Our first recommendation would be to try a positive control alongside your new targeting oligo (TO). If this is your first attempt you may not have an oligo that definitely works for your strain. If your E. coli strain has galK, we recommend our ∆galK as a positive control. If none of those are options, then we recommend designing a longer oligo (up to ~150 or 160 nt) that makes a very small deletion (<100bp) and order it with a high quality PAGE purification (this will be relatively expensive >$100). If the positive control works and your new oligo does not, then that tells you this is an issue specific to your oligo (poor efficiency, essential gene, bad synthesis etc.). If the positive control also fails, this is an issue that is not specific to your oligo - this is the most common scenario we see.

For general ORBIT failures that are not specific to the TO, we recommend checking the quality of the integrating plasmid  - run on a gel, nanodrop / qubit, nanopore sequence if there's any doubt about the identity. If the concentration is unexpectedly low, for example, caused by high amounts of contaminating genomic DNA, ORBIT will not work as well. We would also recommend remaking the induced electrocompetent cells. These are the two most common issues we have encountered, however, despite variations in efficiency we very very rarely have observed zero colonies (even with the no TO control we get a few colonies).

Another scenario is that ORBIT actually does work in the initial transformation, but for some reason recombinants are not recovered. If ORBIT works but is not highly efficient, more cells may need to be plated. The entire recovery culture can be pelleted and plated to ensure this is not the issue.

Are all cells dying after the electroporation? Poor competent cell prep or transformation conditions (high salt can cause arcing) can cause massive cell death following electroporation. Plate cells from the recovery culture on LB without antibiotics to assess if the cell number is near the expected number. Also make sure that antibiotics are not added to the intial recovery culture following electroporation - this will kill the cells. Transformations need at least 30 min - 1 hr to start expressing antibiotic resistance markers, depending on the marker. Another similar issue would be using the wrong antibiotic or too much antibiotic (resistance may be lower for single copy chromosomal insertion than high copy plasmid). Finally, if you are trying to counter select against the helper plasmid or another sacB marker using sucrose in the plate, you might observe no colonies if the helper plasmid has not been lost yet. In this case grow cells longer without the helper plasmid antibiotic or plate on the integrating plasmid antibiotic first and then counter select against the helper plasmid in a subsequent step.

In summary:

• Use a positive control targeting oligo to assess the issue

• Plate the entire recovery culture in case of low efficiency

• Assess the quality of the integrating plasmid

• Remake induced electrocompetent cells and make sure they aren't dying (plate on LB) and that you are using the correct induction scheme, antibiotics, and recovery conditions

Why is my ORBIT experiment less efficient than expected?

First, there truly is variability between strains and even between different experiments with the same strain. Second, ask yourself if higher efficiency is important for your purposes. This is certainly important for high throughput libraries, but for making individual strains in the lab it may or may not be worth diving into this issue. The most common issue we encounter is electrocompetent cell prep. ORBIT relies on transformation through electroporation, so this is absolutely critical to achieving high efficiency. Advice here for comp cell prep.

You may want to optimize ORBIT parameters for your strain / scenario. We recommend testing oligo / integrating plasmid concentrations, helper plasmid induction schemes, oligo homology arm length, and recovery conditions. 

For integrating a plasmid library into a single genomic locus, it may increase perceived efficiency to first make a "landing pad" strain with the attB site only and then transforming in the plasmid library with the attP sites. This decouples the two ORBIT steps and often the oligo recombineering step is the less efficient step. This strain can be made by using pInt_tsXis_sacB_kanR for integrating plasmid excision. See protocols for details.

How do I create markerless / scarless mutations?

The FLP-FRT system can be used to excise the pInt backbone using FRT sites that are already on the plasmid. This leaves a 172 nt scar.

The pInt_tsXis_attP_kanR plasmid can be used to excise the integrating plasmid, leaving only the 38 nt attB scar.

A second oligo recombineering step can be used to cleanly delete the integrating plasmid, leaving no scar. The oligo will be the same as the original targeting oligo, but without the attB site (homology arms only).

How do I create double / triple mutants?

Multi mutants can be made sequentially or in a single step if the modifications are relatively efficient. For single step multi mutants we suggest leaving cells for a longer recovery before plating on double / triple antibiotic plates. Alternatively, the recovery culture can be diluted into LB + antibiotics to enrich for mutants and then plated on antibiotic agar.

How do I make SNP mutations?

We recommend you do this simply by transforming an oligo without the attB site or the integrating plasmid. In this way, you can use the ORBIT helper plasmid just for its first inducible module: oligo recombineering. Efficiency should be relatively high and colonies will have to be screened by PCR. Design a PCR primer such that the 3' end of the fwd primer (or rev) is directly on the mutated base. The primer will not extend if that basepair is not formed. See [this resource] for details. The ORBIT plasmid was not designed for doing this type of oligo recombineering, but it is effective and unlike some other tools the helper plasmid can be removed with sacB.

Where should I put my construct on the integrating plasmid?

We recommend putting constructs immediately adjacent to the 20 bp spacers that flank the attP site.

How can I get a specific attP / resistance / sacB integrating plasmid combination that I don't see available?

with the starter kit you can make any of these with a two part gibson assembly or your favorite cloning scheme. There are annotated gibson sites. For attP variants, you can encode the mutation on a primer and split the plasmid at the resistance marker. Feel free to email the saunders lab to see if we have the specific combination you want, but most of what we have is on addgene.

How long of an oligo do I need?

We recommend 120 nt oligos from Sigma ($24), however 90 nt oligos (including the 38 nt attB site) seem to be sufficient for most applications. [oligo len figure]

Should I pay for oligo purification (HPLC / PAGE)?

We see that PAGE purification yields approximately 2x higher efficiency, but we almost never buy PAGE purified oligos due to the extra cost. For our purposes desalted oligos (typically the default w/ no charge) work perfectly fine and keep costs low. If you need extra high efficiency then pay for PAGE purification. [supp fig panel]

How do I design the oligo against the lagging strand?

If you are working with E. coli MG1655 then we recommend you use the targeting oligo design app, which will automagically find the lagging strand and design the oligo for your gene / region of interest. If you are working with a different strain, then at a minimum you will need to know the genomic coordinates for your region of interest and the origin of replication (ori) and the total length of the genome. The ori is often right at the start of the genome, but this is not guaranteed and is not true for E. coli MG1655. If the ori is not annotated, you may need to do some detective work to find it (there are online tools).

With the ori and genome length you can infer the replichores by assuming the replication terminus (ter) lies exactly opposite the origin (again not strictly true, but a reasonable approximation - feel free to do something more sophisticated to find ter). For example, if the ori starts at 1 and the genome is 4 Mb, assume ter lies at 2 Mb. If ori starts at 3 Mb, then assume ter lies at 1 Mb. Replichore 1 will be the segment to the right or greater than the ori position. Replichore 2 will be the segment to the left or less than the ori position.

For replichore 1, the leading DNA strand will be synthesized against the '-' strand, therefore the leading strand sequence will be the '+' direction. The lagging strand is the opposite, synthesized against the '+' strand it will therefore be '-' direction. So, for all genomic coordinates within replichore 1, use the '-' strand sequence for the targeting oligo and that will bind the template of the newly forming lagging strand. For replichore 2, it is the opposite and you should use the '+' strand sequence for your targeting oligo to bind the template of the lagging strand.

See the Oligo Design page for details. 

Which direction should I put the attB site?

The direction of the attB site and the attP site on the plasmid determine what direction the integrating plasmid inserts into the genome. 

How can I confirm my mutants with PCR?

Design primers flanking the locus of interest. These can be used to amplify the entire modification + pInt. Alternatively genomic primers can be paired with outward facing primers on pInt to amplify the genome:pInt junctions.

Will ORBIT work in organisms beyond E. coli MG1655?

Yes! First, the ORBIT concept should work in diverse organisms with the correct recombineering machinery. The original ORBIT paper was in Mycobacteria. 

More specifically, will the set of ORBIT plasmid tools provided from the paper work in other bacteria? Maybe. Anecdotally, colleagues have had success with other Enterobacteriaceae that are closely related to E. coli. It's worth a try!

For more distantly related bacteria it may be worth trying the available plasmids, but it is probably unlikely to work well, since it's known the oligo recombineering proteins are relatively host specific. That said, it should be possible to adapt ORBIT systems for diverse new species by modifying the helper plasmid. We're happy to advise and help - feel free to reach out to Scott via email (scott.saunders@utsouthwestern.edu).