My battle with BamHI - (Mar/23/2010 )

Over the last several months I have been attempting to create a plasmid library for several clones I work with, and I do mean attempt. After reading the NEB manual/catalog I thought that I would be able to quickly troubleshoot any problems and produce my clones quickly, man was I wrong. After the usual problems came and went I continued to have trouble finding clones within the myriad of oddball constructs.

To make a long story short I will cut to the ending.

All of my cloning troubles stemmed from BamHI. Basically I was only partially cutting my DNA, both insert and vector. BamHI looks perfect on paper: fast cutting, efficient ligation record, established in cloning, etc... Not quite. BamHI has the ability to slide along DNA once bound until it finds it's recognition sequence. This is controlled by the salt concentration for the reaction, if the conc is too high then no sliding will occur resulting in a lowered efficiency. This doesn't sound like a big deal since the enzyme is sold with an accompanying buffer, but BamHI has one more trick up it's sleeve: ES complexes that do not result in E + P or E + S. The Km value for BamHI seems to plummet in reactions that have low water and/or high salt; and no, I have no data to back this up. To fix my problem I did something that I didn't foresee having to do, be redundant. For all ligations that use BamHI I digest the DNA, purify, digest a second time with BamHI, gel extract. I also make sure that there is an excess of water, typically I use 100ul total volume. I'm not sure why but this reduced my negative control transformation (digested vector w/o ligase) numbers from 300+ to zero. I was worried about trashing my DNA but adding 0.5ul T4 ligase (with buffer of course) gives transformation numbers similar to the positive control.

I write this because this forum helped me figure out my problem. Hopefully this post will help another who is as frustrated as I was.

After everything I am reminded of something a professor once told me: "don't believe everything you read and question everything."

Thanks, I'll try to do your way next time I clone as I usually have very low number of positive colonies

Sorry if there is something I am missing here (I am new, and did not see the posts on this forum where you got help with this), but, honestly, based on the information in this particular post, I think you must be doing something wrong.

BamHI is the RE I use most frequently and I have never, ever encountered problems with it. I get complete digestion in an hour or less at 37C (based on agarose gel visualization). Remember that digesting for too long can cause star activity, though this isn't usually a problem with BamHI. I don't get the greatest ligation efficiencies (which is probably my fault), but I also do not get many (or any) negative control plate colonies, nor do I get false positives on my ligation plates.

Are your DNA samples PURE (i.e. either mini-prepped DNA or purified PCR products)? What are the A260/A230 and A260/A230's?

I think you just need additional purification steps. Everything you said about salts seems irrelevant to me, because the only salts that should be present in your digest are the salts that are in the RE buffer, as well as any salts in your DNA purification elution buffer (i.e. Tris*Cl found in Qiagen's Buffer EB).

Another option is to try out NEB's BamHI-HF (high fidelity), which is now the same price as regular BamHI (at least for the larger volume they sell), which may provide better results for you.

I added responses in blue.

DyDx on Mar 23 2010, 08:00 PM said:

Yea, sorry, I just can't fathom why you would be having problems with BamHI, it should just work.

One thing I do is that I ALWAYS dilute my DNA sample to approximately 50 ng/uL (before adding to digest) and make up a 50 uL digest. Every time. This concentration seems to be fully digested quite well, and it is still high enough to be visualized on a gel so you do not need additional purification (unless you need to remove unwanted large fragments, of course).

The reason I chose to do this is that when I was learning the methods, the vector I was using mini-prepped to about 50 ng/uL and I always had good results digesting it, so I have carried this along and it has been successful. I don't know if this helps, but if you are digesting a high concentration of DNA it is something to consider. You really should not have to digest twice to have a successful ligation.

I do not doubt your observations, georgiadave, and I'm glad you found a method that works for you. I also appreciate you posting your observations for others that might be having similar problems -- that's what the BioForum is all about!

I am a bit puzzled, though. Like DyDx, we use BamHI for cloning in my lab far more often than any other restriction enzyme, and have had very little trouble with it. We use BamHI from Invitrogen, routinely digest for just an hour in 30 - 50 ul volumes at 37C, and always gel purify our digested vector and insert segments through an agarose gel with guanidine HCL added as UV protectant before ligating, but these don't seem to be related to the problems you're having...

How much enzyme are you using? What is your typical digestion reaction mixture composed of?

I understand that you've done digestions with other enzymes without issue, which makes this a very curious problem...

Have you tried BamHI from a different vendor?

Ha, I know. When I finally found this method I was very surprised and upset at the same time - it took a while. The concentration of my stock DNA is higher than 50ng/ul but I don't digest a set volume, I usually digest 1-2ug at a time.

A common setup would be:
xul DNA (2ug)
10ul Buffer (10x stock)
1ul BSA (100x stock)
xul water
2ul BamHI
Total volume: 100ul

All of my RE's are from NEB. This is something that has happened with every tube I have used. I have tried using less enzyme per reaction to get the same results. I have never used enzymes from a different vendor, but this problem only occurs when using BamHI. All other ligations are by the book (ex: KpnI, SalI, HindIII, etc...)

Have you tried using more enzyme? I usually use 2 ul of enzyme in a 30 ul reaction volume, the same amount you're using in a 100 ul volume.

Your main problem -- negative control transformation (digested vector w/o ligase) producing 300+ colonies -- seems to stem from incomplete digestion, which is improved by a second round of digestion. I suspect you're not using enough enzyme in your initial digestion, and if you upped the amount in the initial round (as long as the amount of enzyme is < 10% of the final volume), the situation would improve.

Here is my standard digest, I wonder if you'd see a successful digestion in one step with this formula:

39 uL DNA @ ~50 ng/uL
5 uL NEBuffer
5 uL H2O or 10x BSA (I forget if regular BamHI uses BSA...)
1 uL BamHI
------
50 uL

37C for 1 hour. I believe NEB calls for a 50 uL reaction and I would suggest always using this volume if possible. It's not intuitive, but the volume can have an effect beyond the salts problem you were talking about.


Another possibility, though unlikely, is that your vector is self-ligating even though you are using sticky ends. I have noticed that if I take the time to treat with CIAP, I get 0 control plate colonies, whereas I might get a handful if I don't. I know -- it doesn't make sense, sticky ends shouldn't ligate -- but the cells manage to do it at a low rate anyhow. This doesn't sound like the cause of your extremely high background, but it might help too.

@Homebrew: I have tried digesting with up to 5ul enzyme in 50ul with 1ug DNA to get the same results.

@DyDx: your setup adds 39ul of 50ng/ul DNA - this equals ~2ug DNA...the same amount that I routinely digest. The catalog does call for 50ul digests of 1ug DNA. I generally increase my total volume as my total DNA concentration increases; so for 1ug I use 50ul, 2ug I use 100ul, etc... I do this based on a post by pcrman - I liked the logic behind his theory.

Also, say I do a sequential digest with SalI followed by BamHI with only purifications, no gel extraction. I will get moderate levels (still too high!) of background colonies. Now, if I flip the sequence of enzymes with BamHI first I will get the same extremely high background.

I have done controls by only using SalI once w/o gel extraction to get only 2 negative control colonies but great ligation transformant numbers.

For all trials I grow individual colonies for a mini-prep to check the plasmid by RE digestion; I used BamHI, as well as other enzymes known to cut the plasmid, to make sure the recognition sequence wasn't trashed. This is to check the size of the resulting clone, basically to check for oddball stuff that can happen - all have proper size. The only ones that don't are the partially digested mixtures of vector and insert that I attempt to ligate. After the double digest of BamHI protocol I do not get the weirdo constructs that I do not want. I haven't done sequencing b/c it's expensive

Thanks for the replies, this is great!