Making in-frame deletions using FRT sites is straight forward. However, designing the primers can often be confusing. Here are some rules which will ease the process.

1. Some definitions

FRT sequence: TGAAGTTCCTATACTTTCTAGAGAATAGGAACTTC

FRT' (complemented FRT) sequence: GAAGTTCCTATTCTCTAGAAAGTATAGGAACTTCA

2. Upstream (left) primer:

Modular structure: [left homology block]—ATG—[1+3N]—FRT—[drug left]

The "left homology block" is identical to sequence preceeding the start site of gene X, the knock-out target. It can be extended to include the ATG and [1+3N] blocks and may be exactly these.

"ATG" is the start codon.

"[1+3N]" is a block of containing one plus a multiple of 3 nucleotides, and must be chosen not to introduce a nonsense triplet into the reading frame defined by the start codon (i.e., the 0 reading frame).

"FRT" is the corresponding sequence given above.

"drug left" is identical to a tract at the left end of a drug cassette and will allow PCR to initiate. It is generally 20 bases long.

3. Downstream (right) primer:

Modular structure: [right homology block]'—[stop]'—[3M]—FRT'—[drug PCR right]

"[right homology block]'" is identical to the complement of sequence following the stop codon of gene X. It can be extended to include the stop codon complement and the [3M] block and may be exactly these.

"[stop]'" is the complement of the stop codon.

"[3M]" is a block containing a multiple of 3 nucleotides. As before, these must be chosen not to introduce any complemented nonsense triplets into the 0 reading frame, defined in this case by the preceeding complemented stop codon. N and M need not be the same.

"FRT'" is the complemented FRT sequence given above.

"drug right" is identical to the complement of a tract at the right end of a drug cassette and will allow PCR to initiate. It is generally 20 bases long.

4. Product:

Modular structure: …ATG—[1+3N]—FRT—[3M]—[stop]…

The length is thus 3 + 1 + 3N + 35 + 3M + 3 = 3(14 + M + N).

This places the stop codon in frame with the start codon, and introduces no terminators from the FRT region. If care was taken with the design of the [1+3N] and [3M] blocks, the result is a minigene of 13 + M + N codons. Again, if the filler blocks were properly made, the peptide expressed will be identical to the wild-type protein only in the first 1+N and last M-1 amino acids.

5. Examples.

These primers were used to make an in-frame deletion of eutE:

EE-FRT1

ATG—AATCAACAGGATATTGAACAGGTGGTGAAAGCGGTAC— TGAAGTTCCTATACTTTCTAGAGAATAGGAACTTC—AAGCCACTGGAGCACCTCAA

1-3 = eutE start codon
1-40 = eutE homology block
41-75 = FRT
76-95 = camRL-core sequence.
In this case, N = 12.

EE-FRT2

TTA—TACAATGCGAAACGCATCCACCAGCACGCATCGACGCAG— GAAGTTCCTATTCTCTAGAAAGTATAGGAACTTCA—CGGGGAGAGCCTGAGCAAA

1-3 = complemented eutE stop codon (TAA)
1-42 = complemented eutE homology block
43-77 = FRT'
77-96 = camRR-core sequence (note overlap with FRT').
In this case, M = 13.

Minigene (after FLP excision)

ATGAATCAACAGGATATTGAACAGGTGGTGAAAGCGGTACTGAAGTTCCT
ATACTTTCTAGAGAATAGGAACTTCCTGCGTCGATGCGTGCTGGTGGATG
CGTTTCGCATTGTATAA

where underlines indicate the start and stop codons.

Peptide product

MNQQDIEQVVKAVLKFLYFLENRNFLRRCVLVDAFRIV*

where underlines indicate regions common to EutE. In both cases, these are 1 amino acid longer than minimally expected, as the FRT seqeuence contributed leucine at each end. This residue is found, by chance, in the corresponding positions in EutE.