Usually when you ligate BamH1 with BglII you get a unique restriction site that is recognized by neither BamH1 or BglII.

If you want to try to cut at one BamH1 site and not the other, I usually do partial digests. The best protocol that I found is to start with ~5 ug of total plasmid in ~80 uL. Add 0.5 uL of enzyme and then at different time points (30s, 2 min, 5 min, 10, 30 min) remove 16 uL of the reaction and add it to an eppendorf already containing the gel loading buffer and then flash freeze the tube (use dry ice, liquid nitrogen, or an isopropnol:water (50:50) mix that has been in the -80). After the last time point, thaw out the samples and load directly onto an agarose gel. You don't want the sample to thaw out and sit around for too long or else the enzyme can start to cut again. I usually start loading the longest time point (30 min) first and then end with the first time point (30 s). You should see a ladder where the DNA will be mostly uncut at the early time points and mostly 100% cut at the later time points. You can then extract the band that you want. Alternately, you can then repeat the reaction but stop the whole thing at the optimal time point (that you discovered during the first time course).

Without knowing more about your strategy I'm not sure i can offer much more help (e.g. are you cutting with a second enzyme?).

Good luck,

J

Thanks Jonathan,

seems an interesting procedure..I've tried to do sequential digestion, but the idea of snap freeze the different time points is useful...however, two problems:

1-my sequence is small , 60bp, so checking on the agarose gel could be difficult

2- i need to digest with another enzyme (HindIII) to excise this sequence...

Hmm...the small fragment does make it difficult. You could try the partial digest, then extract the band representing the linear fragment (this will be combination of fragments cut at either one of the BamH1 sites), then do a complete HindIII digest. If you run that on a gel you should be able to separate the correct fragment from the one cut at the wrong BamH1 site. It really depends on how far away the two BamH1 sites are and the size of the fragments.

I have done this type of strategy a lot and it works pretty well. Partial digests can be intimidating but with the freezing method they are pretty easy and it can really open up the range of things you can do.

Alternately, I would try and use gene synthesis to make a fragment that you can use to clone into two unique restriction sites to give you the plasmid that you want. Gene synthesis is pretty cheap now (~30-40 cents per base pair) and it is usually my go-to approach for difficult clones.

Good luck!

J

I get the impression you want to remove the 60 base sequence. Do inverse PCR around the full circle of the plasmid, (use NEB's Q5) introducing any sequence or retriction site you like in the tails, cut and religate. By this method you can also achieve site direct mutagenesis

thanks Jonathan and David for your valuable suggestion... I think I'll try for site-directed mutagenesis, re-introducing the nucleotide changed...

Actually, 60 bp can even be ordered as an oligonucleotide. Order both strands with overhangs to match your restriction sites, anneal, phosphorylate if you are using a dephosphorylated vector (use PNK) and off you go! Quick and cheap. Beware: you will have to sequence a few colonies as long oligos are less reliable.

Good luck!

Hi Fulvio,

I have done both the procedures, as described by Jonathan and Anna. I would suggest start with digesting 15-20 ug DNA when doing partial digestions so that you have enough for downstream processes. I got right clone using this procedure but lots of nonspecific clones.

The Anna's procedure worked best for me if you are planning to clones little fragment...I routinely perform this technique and get almost 90-95% good clones (you can follow the procedure described for cloning of shRNA in in siRNA expression vector on page 11...... http://tools.lifetechnologies.com/content/sfs/manuals/fm_5760.pdf). The only drawback in this technique is the quality of oligos....the longer the oligos there are more chances of deletions (in my experience mostly one or two nt deletions if you have strech of poly T or C). You can request gel purified oligos, but they cost little more. Also ligation efficiencies are increased by 10-20 fold if you phosphorylate the oligos before ligation. Hope one of these would help.

Remember that BamHI is very stable. I would think that you might want chill then immediately purify, by phenol extraction or with a DNA clean up column

The only reason why I would advise against site directed mutagenesis is that you end up doing PCR of the entire plasmid and thus have the chance of introducing random mutations. Usually when I do mutagenesis I then re-clone the majority of the plasmid backbone from the original plasmid.

Note that the chance of introducing a mutation is very low, but I would hate to start working with the plasmid and then down the road find out that there was a point mutation somewhere that was affecting your results.