Problems with Restriction Digest Results - (Apr/22/2012 )

Hi

I recently ligated a ~900bp insert into a P-Gem T easy Vector (3015bp) then transformed the vectors into DH5-alpha cells. After miniprep I cut the plasmid with EcoRI then ran the samples on a 1% agarose gel. Upon viewing the gel under a UV lamp I found the expected bands at the 900bp and 3000bp mark but I also found bands forming at the ~2500bp and ~700bp mark in the same lane. This promblem occured in Lane 5 and 10, as can be seen in the uploaded picture.

I am finding it difficult to explain what caused this.

I thought perhaps star activity occured on some level, as it causes cleavages to occur at any AATT site opposed to the GAATTC site. Though I am not sure whether that is the case, again because the expected bands also formed on the same lane.

The EcoRI enzyme I used went out of date sometime in 2005, was the only available enzyme in the lab, and I thought perhaps this had some strange effet upon the plasmid. However, the other samples worked out fine by only showing bands at the expected 900 bp and 3000 bp mark, leading me to think that this was not the problem.

I ruled out methylation because excision of the insert clearly occured. However, when I pasted the sequence in NEBcutter it stated that the EcoRI sites may be affected by CpG methylation.

After sequencing the sample it was found the plasmid had mutations. For example, one of the EcoRI cleavages sites was GATTC instead of GAATTC and the insert was determined to be ~500bp in length instead of the expected ~900bp in length.

I have spent a considerable amount of time trying to figure this one out and am having a very hard time. Any help would be greatly appreciated.

-0702311-

It could be uncut plasmid. Have you run an uncut control lane?

-leelee-

I didn't use uncut plasmid as a control. is it possible that uncut plasmid is in lanes 5 and 10? I don't see how that could happen after exposing the samples to EcoRI, as it would cause it to be linearised.

-0702311-

I'd re-stain the gel and look for small fragments (<200 bp) You may have cloned in concatemers. Is there for sure no Eco site in your product?

-Do1ngP3e-

Do1ngP3e on Mon Apr 23 18:20:53 2012 said:

I'd re-stain the gel and look for small fragments (<200 bp) You may have cloned in concatemers. Is there for sure no Eco site in your product?

I sequenced the plasmid samples and there seemed to be no evidence of concatemers, only one gene insert per sample. Yes there are EcoRI sites in most of the samples, as there should be as that is what the pGEM-T easy plasmid contains near the gene insert site. However, sequencing revealed that in some samples one of the two EcoRI cleavage sites had mutated slightly (i.e. GATTC instead of GAATTC), none the less the expected bands were gained in the gel although with some additional bands at the ~700bp and ~2500bp marks.

Another thing i forgot to mention was that I incubated the samples in a water bath overnight and the samples volume was very small (10

μl). I thought that perhaps the use of a water bath would cause evaporation and condensation formation which would result in a lower water volume but higher salt concentration in the samples, this would only have been worsened by the unnecessarily long incubation time. This would cause non-optimum conditions for the restriction mixtures, which may point to star activity occuring in some samples. Does this sound like a plausible/likely cause?

-0702311-

0702311 on Mon Apr 23 09:35:19 2012 said:

Sorry, I misread your original post and got confused about which bands were the unexpected ones, so my suggestion is somewhat irrelevant.

But yes- you absolutely can have uncut plasmid in a sample that has been digested with an RE. It is not uncommon for digestion to fail to go to completion in sub-optimal conditions (for example, too much DNA for the amount to RE added).

-leelee-

If it was uncut plasmid would it not be displayed as a higher band than the 2500bp and 700bp bands found on the gel? as pGEM-T easy is 3015bp in length and with the DNA insert it would be ~3900bp. Would an uncut plasmid therefore not show a band at around the 3000bp or 3900bp mark? or would supercoiled uncut plasmid possibly be the cause of the 2500bp band, as I have heard supercoiled plasmid moves faster in the gel than linearised or circular.

Thank You. I am really appreciating your help with this.

-0702311-

Do NOT attempt restriction digests in small volume! They rarely work, because the usual reason people do them is that they want to keep DNA concentrations high, and use their DNA as a major volume component of the reaction. This leads to problems, such as inhibitors in their DNA prep and high concentrations of glycerol from the enzymes in the reaction. There is a reason the manufacturers want you to do 50 ul reactions -- it is to keep the volume of your DNA sample (and of the restriction enzyme) small relative to the volume of the reaction. Lots of water is the secret ingredient in making your digests work. Uncut plasmid is a typical result of low volume reactions.

-phage434-