Recombineering (short for homologous recombination-mediated genetic engineering) can be used to make an assortment of DNA modifications: insertion and deletion of selectable and non-selectable sequences, point mutations or other small base pair changes, or assembly of multiple sequences together. It also has the flexibility to modify the E. coli chromosome, plasmids, or bacterial artificial chromosomes (BACs, a kind of plasmid). 

Recombineering in Escherichia coli is mediated by bacteriophage proteins, either RecE/RecT from Rac prophage ^[1] or Redαβδ from bacteriophage lambda.^[2][3] The λ red recombination system is now most commonly used and the first demonstrations of  λ red in vivo genetic engineering were independently made by Kenan Murphy^[4] and Francis Stewart.^[1][2] However, Murphy's experiments required expression of RecA and also employed long homology arms. Consequently the implications for a new DNA engineering technology were not obvious. The Stewart Lab showed that these homologous recombination systems mediate efficient recombination of linear DNA molecules flanked by homology sequences as short as 30 bp into target DNA sequences in the absence of RecA. Now the homology could be provided by oligonucleotides made to order, and standard recA⁻ cloning hosts could be used, greatly expanding the utility of recombineering.

To use the λ red recombineering system to modify target DNA, linear donor dsDNA _{(ssDNA usage is not covered here)} that includes 5′ and 3′ "homology arms" matching the target DNA is electroporated into E. coli expressing the λ red enzymes, usually encoded on a temporary plasmid. These enzymes then catalyze the homologous recombination of the donor DNA with the target DNA sequence. The homology arms of the donor DNA must match the region immediately 5′ and 3′ to the target sequence to be inserted into. Sequence in between the homology arms of the donor DNA replaces the sequence between the matching homology sequences in the target DNA. A selection, counterselection, or screening gene marker is usually included to select or screen for recombinants, and these markers are often designed to be removed by a second round of recombineering coupled with counterselection of a counterselectable marker, or using site-specific recombinases like the Flp-FRT system which requires FRT sequences around the region to be removed by Flp recombinase expressed from a second plasmid.

dsDNA Procedure

1. Construct integration cassette (no known hard size limit, but probably < 5 <5 kb is most efficient). It needs to consist of:
   1. 5’ and 3’ regions that overlap with genomic sequence flanking the genomic position of insertion. For example, you could use the 5′ UTR and 3′ UTR of uidA, if you planned on deleting/replacing uidA CDS. According to the original paper, these overlapping regions need only be 40–50 bp long, allowing addition to amplified DNA via oligos <60 bp.
      
    For cassettes longer
   1.  True or not? For cassettes larger than around 3 kb, much longer homology region around 500 bp are needed since it must use a different recombination mechanism.
   2. A selection/screening marker, commonly an antibiotic resistance cassette (that functions at a known, typically lower, c_antibiotic at genomic copy#). A fluorescent protein (gfp) or dye-producing enzyme (lacZα or uidA) cassette can be included or even tagged to an antibiotic resistance CDS to allow secondary visual confirmation of integration. Ideally, you have FRT (Flp recombinase target) sites around the marker to allow later removal of the marker with the site-specific Flp recombinase, without which a secondary λred integration with a different resistance gene or CRISPR/Cas is needed to remove the resistance marker. FRT-flanked Chl^R and Kan^R markers can be amplified (even using primers with genomic homologies) from the R6Kγ conditional origin plasmids pKD3 and pKD4, respectively.
   3. Optionally, other genes/DNA of interest, which remain at the locus after excision of the region between FRTs.
      [5′ genomic homology] – [
   Opt
   1. opt. other DNA] – [FRT] – [selection/screening marker(s)] – [FRT] – [Opt. other DNA] – [3′ genomic homology]
2. Transform strain of interest with pKD46[P_ara-λ.redori_pSC101ts Amp^R] , which or a similar inducible λ.red, temp-sensitive plasmid. Grow plates at 30°C to maintain the temp-sensitive plasmid.
   1. pKD46 has the phage λ
   recombination gene
   1. recombinase operon,
    λ 
   1.  λ.red, encoding Exo, Beta, and Gam, controlled by an arabinose-inducible promoter and on temperature-sensitive pSC101 replicon
   (Amp^R for pKD46). Grow plates at 30°C to maintain the temp-sensitive plasmid.
   1. . 
3. Make a ( small-scale - see below) electrocompetent cell prep of the pKD46 transformed cells: grow a seed culture at 30°C in 2×YT to an OD≈1, then induce 2×YT to OD₆₀₀≈1. Then induce λred with 1–50 mM (0.05–0.7% ) arabinose (m/V) arabinose and grow for 1 hr before harvesting the cells and washing two to four times in 4° 10%_V/V glycerol, concentrating it into ≈1/100th the culture volume.
   1. Full induction is 0.32%
   , 16 µL/mL of 20% arabinose,
   1. _m/V arabinose = 1.6%_V/V of 20%_m/V  arabinose stock; though less, say 1 mM
   /
   1. ~ 0.05%_m/V , also works
   ) and grow for 1 hr. (
   1. .
   2. The 2×YT and high OD are important
   - 3 hrs growth if you start off
   1. . 3 hr pre-induction growth seems to work well if you inoculate the culture with a
   1:
   1. 30
   dilution of overnight seems to work well).
   1. -fold dilution of saturated preculture.
   2.  Why can't this be done with growth in SOB or LB as many protocols stipulate? Is the λ.red induction or recombination less efficient?
4. Transform the induced pKD46 competent cells Transform pKD46 competent strain with linear dsDNA (as from PCR), ; allow to recover for an one hour at 37°C, ; plate on antibiotic used in integration cassetteLB agar + antibiotic_{int cassette}**, and grow at 37°C (or 42°C).37–42°C.
   
   1. ** Use single-resistance-copy antibiotic concentrations (e.g. kan²⁵ or chl¹⁵), typically half the multi-copy concentration (e.g. kan⁵⁰ or chl^25/34). 
5. Confirm integration Confirm presence of desired sequence by appropriate colony PCRs and presence of desired sequence at targeted locus. (Choose primers across integration boundaries.)
6. Grow a few colony-PCR-verified isolate in liquid culture to saturation at 37°C to cure away the λred plasmid.
7. genotyping PCRs, using a fraction of resuspensions of integrant colonies in 10 µL PBS/LB. Use PCR primers whose product extends from the integration cassette, through the integration homology, into the genomic context.
8. Cure the temp-sensitive pKD46
   Inoculate the remainder of a PCR-verified integrant colony suspension in 1 mL LB (+ opt. antibiotic_{int cassette}) to saturation at 37–42°C. Then plate on LB agar (+ optional antibiotic_{int cassette}), growing at 37–42° to obtain single colonies.
9. Confirm curing of pKD46
   To test for lack of growth in +amp medium and to use the corresponding –amp culture to make a frozen stock, inoculate both 1 mL LB amp and 1–2 mL LB (+ opt. antibiotic_{int cassette}) with each of a few colonies. If proceeding to remove the resistance marker, you can use some of the –amp culture as a preculture for preparing competent cells, or you can remove some of the culture in log phase for direct use in preparing competent cell (may require increasing the culture volume).
   
   1. To split a colony for separate inoculations, either resuspend it in a small volume PBS/LB and inoculate using fractions of this resuspension, or resuspend the colony in a combined volume of medium which you split appropriately and then add different antibiotics.
   Confirm loss (curing) of pKD46 plasmid by inoculating amp solid or liquid medium with 
10. At this point, phage P1 transduction can optionally be used to transfer the desired genotype to a clean genetic background (as genomic rearrangements may have been induced by the λred).
    
    Site-specific recombinatorial removal of resistance marker
11. Make competent cells with a Make a competent prep with pKD46-cured cells, then transform with pCP20 which has FRTs, is Amp^R, Chl^R, and temperature sensitive. For best cassette integrant colony/culture, making sure to use single-copy antibiotic concentrations (or none at all).
    1. The TSS method or even lower efficiency chemical competent cell prep methods are fine to use. 
12. Transform with pE-FLP[P_con-FLPori_pSC101ts Amp^R] with recovery at 30°C. Plate on LB amp , and grow up at 30°C.
13. Pick 4–16 colonies, grow it up at 37°C with NO antibiotics.
    1. DO NOT SKIP THIS STEP: Always grow up in media at 37°C first. Otherwise you will not enrich for the right pCP20 clones. (Confirmed by Josh)
14. Plate some of the culture at 42°C with no antibiotics.
15. 37°C to obtain colonies. Though typically not needed, you may test for loss of the resistance marker (and plasmid) as in Step 7, but with colonies tested for individual sensitivity to ampicillin (to ensure loss of pE-FLP) and to antibiotic_{int cassette} (to ensure cassette excision), reserving some colony cells for preparing a frozen stock.
    1. Strong, constitutive expression of Flp from pE-FLP should result in 100% resistance cassette excision without a culturing step at 30°C for prolonged plasmid maintenance.
    2. If using the classical but inferior pCP20[λ.p_R-FLP cI857ori_pSC101ts Amp^R Chl^R], transform a cassette integrant with pCP20, selecting on LB amp agar at 30°C. To induce FLP expression and subsequent pCP20 curing, grow a pCP20-transformed colony in LB (no antibiotic) at 37° and replate after a few hours' growth on LB agar (no antibiotic) growing at 42°C. Grow 4–16 colonies at 37–42°C in LB, expecting <5% of colonies to have successfully excised resistance markers. Test individual colonies for loss of all antibiotics via plating tests.
16. Optionally confirm integration of desired sequence by appropriate genotyping PCRs, using a fraction of resuspensions of integrant colonies in 10 µL PBS/LB. Use PCR primers targeting the genomic context flanking the integration homologies (not inside the integration homologies) to test for successful removal of the resistance marker / sequence between the FRT sites (leaving a single FRT). You may additionally use PCR primers targeted inside of the resistance cassette to ensure the cassette is indeed no longer in the genome at any locus, possibly silenced or mutated.

References

1. Zhang, Y., F. Buchholz, J. P. Muyrers & A. F. Stewart, (1998) A new logic for DNA engineering using recombination in Escherichia coli. Nature Genetics 20: 123-128.
2. Muyrers, J. P., Y. Zhang, G. Testa & A. F. Stewart, (1999) Rapid modification of bacterial artificial chromosomes by ET- recombination. Nucleic Acids Res. 27: 1555-1557.
3. Yu, D., H. M. Ellis, et al. (2000). "An efficient recombination system for chromosome engineering in Escherichia coli." Proceedings of the National Academy of Sciences of the United States of America 97(11): 5978-5983.
4. Murphy, K. C., (1998) Use of bacteriophage λ recombination functions to promote gene replacement in Escherichia coli. J. Bacteriol. 180: 2063-2071.