When using high-fidelity enzymes for site-directed mutagenesis, what should I do if the mutation efficiency remains low?

A high number of colonies on the plate after transformation does not necessarily indicate success. It is recommended to set up two controls: one with the PCR product without DpnI added, transformed directly to assess background; and one with no primers, only template, subjected to the same procedure. If the background control also yields numerous colonies, it indicates a problem with DpnI or antibiotic selection. During screening, do not pick only two or three colonies; pick at least 12-24 single colonies, first perform colony PCR or restriction digestion for preliminary screening, and send for sequencing only if differences are observed. When sequencing, perform both forward and reverse sequencing to avoid misjudgment from single-end reads.

If both enzymes have been tried, it is recommended to switch to a commercial site-directed mutagenesis kit, whose buffer and DpnI are already matched, and the success rate is often higher than assembling the system yourself. Alternatively, abandon the DpnI method and use a high-fidelity enzyme to amplify in two segments followed by seamless cloning (such as Gibson or homologous recombination), with the mutation introduced in the primers; the cost is slightly higher but the background is low. If you continue optimizing, extend the extension time by 1.5 times, use a hot-start enzyme, and perform a gradient for the annealing temperature at primer Tm+3°C. Do not change enzymes frequently; keep one set of variables fixed.

Site-directed mutagenesis primers are critical. It is recommended to use a pair of completely reverse complementary primers, 30-40 nt in length, with the mutation site placed in the middle, and at least 12-15 nt at both ends perfectly matching the template; calculate Tm using the nearest-neighbor method, preferably >=78 degreeC. If the primer ends have too little GC or secondary structure, the background will be high. When I encounter many clones but wild-type sequencing results, 90% of the time it is because the 5' end complementarity of the primers is insufficient or they were designed with partial overlap, causing the PCR product after gap repair to still be the original sequence. Resynthesize the primers with PAGE purification; do not use ordinary desalted primers.

A template of 50 ng is too high for a 6.5 kb plasmid, and DpnI digestion is difficult to complete. Try reducing it to 2-10 ng, or even 1 ng. Keep the cycle number at 10-15, and do not exceed 18; the more cycles, the more likely original plasmid amplification or concatemers occur, resulting in high background after transformation. For the extension time, 6.5 kb at 2 kb/min is about 3.5 min, but I usually give 5 min, especially for long plasmids with secondary structure. The product does not need to run to full intensity; try transformation with 10-12 cycles, and the success rate may be higher than with a bright band.

Confirm that your template plasmid is extracted from a dam+ strain (such as DH5α, TOP10). If it is not dam+, DpnI cannot recognize the methylation sites, and the wild-type template will not be cut. After template extraction, do not use prolonged elution at 65°C in the final step, as this can lead to loss of methylation. The DpnI amount can be increased to 1-2µL (20U/µL), digest at 37°C for 1.5-2h, or even add an additional 0.5µL after 1h of digestion. The digestion product can first be run on a gel to check for residual original plasmid; if the original band is still very strong, do not transform. When background is high, purify the digested product with a column before transformation.