Multimerization after restriction digestion - (May/27/2008 )

I am facing a unique problem and even after exhaustive searches, I have not got an answer.

I have a PCR amplified product (~1 kb), which gives me a single band on the gel. I have gel-purified this fragment and re-checked on the gel to confirm it gives me a single band of ~1 kb. Subsequently I have restriction digested this fragment with NdeI and BamHI. When I load this digested product, I get 2 bands (one corresponding to ~1 kb and another of ~2 kb). I have never seen such a thing happening in the past, though I have cloned several fragments using the same enzymes. Further, recently, I am seeing such a pattern in 2 other inserts of different sizes as well. My guess is that the digested products have cohesive ends and hence forming dimers, but I am not sure. Further when I gel-eluted the lower band (~1 kb), I am not getting any transformants upon ligation.
Does this mean that
1. After RE digestion, the lower band (~1 kb) is undigested and therefore no transformants?
2. After RE digestion, the higher band (~2 kb) is digested and due to this their cohesive ends join and give a dimer?

I hope someone can shed light on this mysterious problem.
Thanking you in anticipation

---

You are thinking in the wrong direction: The kind o f digestion products you describe always have cohesive ends. Yours is not unique.

It is difficult to understand what exactly you are doing here, so I would suggest that you talk about this problem with a senior colleague and I am sure you will find out what is wrong.

---

You'll always have some dimers forming due to the cohensive ends joining with one another and this is what you are seeing. The reason for your lack of transformants is not related the dimers you are talking about here. If anything, the dimers show that your PCR product has the cohesive ends it needs for ligation into the vector. The ligation or transformation is failing due to other reasons. The evidence here suggests your insert is fine.

---

That is precisely my point. The higher molecular weight band is probably the one that has got digested and hence form dimers with their cohesive ends. But I have not used that fragment for ligation. I have used the lower molecular weight band for my ligation and probably, they are undigested and hence I am not getting any clones. And what I see as high molecular weight band is about 1/4th of the total insert present. And I have checked my ligase by ligating single enzyme digested vector molecule and it gave me ~350 transformants (And I have made sure that my vector was completely digested as restriction digested vector without ligation just gave me about 25 clones. So, if ligase, comp cells are not a problem, what else could be the problem?

---

There is nothing wrong with what I am doing and my problem is fairly straightforward. Upon restriction digestion of a PCR product of 1 kb, I am seeing two bands on the gel, one of 1 kb and the other of 2kb. So, my doubt is which one should I elute and set up for ligation - The one which gives 2 kb or the one with 1 kb. AND I KNOW IT WILL GIVE ME COHESIVE ENDS !!!!

---

Have a look at the detailed description of NdeI activity on the NEB website. DNA purified by standard miniprep procedures cuts at lower efficiency, and I suspect the same might be true for DNA from gel extractions. Try re-precipitating the PCR product and washing very well in 70% EtOH. Alternately, if your PCR product is clean, digest the reaction with some DpnI, which will get rid of the template. Then do an EtOH precipitation, or even just go straight to your NdeI/BamHI digest, adjusting salt etc as necessary. Fewer steps, better DNA.

Next, in order to reduce the chance of cohesive ends causing dimerisation, which I frankly doubt is what's happening, heat your sample before you run. Your ends are only going to be 4 nt each, so they should melt at a very low temp.

There are cases where sticky ends cause fragments to run higher, but the only example I know of is the cos termini of lambda, and they are 12 nt long. If 4nt overhangs were able to cause dimerisation, every RE gel would have dimers.

---

All because the 1 kb band is there that doesn't mean it hasn't got the required sticky ends, it just hasn't formed a dimer that's all. Dimers form when you have lots of DNA, it says nothing about the quality of the 1 kb band. If you're getting no colonies and your ligation and transformation reactions are working i would test each of the restriction enzymes to make sure they are working. The best way to do this is to setup a test digestion with each enzyme and another reliable enzyme and digest the circularised vector to produce specific bands. The presence of the bands will confirm both REs are working.

And clearly there is something wrong with what you're doing, you're on here asking about it. If it is straight forward then why haven't you figured it out? We're trying to help so just be a little more gracious.

Good luck, Rob

---

Hi Rob,

I am sorry if I have offended you (which certainly was not the point) and I have tried all the things you have mentioned above like check9ing the enzymes individually, ligases, comp cells, etc. I was only frustrated with the fact that so far whoever I have discussed this problem with is asking me to repeat the whole procedure again (which I have done several times). By basic question however remains unanswered. Are such dimers commonly seen, b'coz according to one of the reply posts, it is not. So, I am confused by this whole thing. What I meant by saying that I am not doing wrong is that the method I follow is the same for all my inserts, and I see this dimerization happening with only few of my inserts. And once again, appologies for sounding rude, which was certainly not the intention. All I wanted was to convey that this is a genuine problem not faced by a novice, but by someone who has been doing molecular biology for the past several years and I have even discussed the problem with my seniors, who couldn't help either. So I hope you will forgive me and not keep this against me. Actually this outburst was not against your reply at all. But still I appologise and thank you for your suggestions, which I will try once again. After all molecular biology does come with a little bit of voodoo magic

---

Rob made a very good point, and I am glad it was not your intention. I too was put off by your tone. See? science is not all nonhuman.

If you have been doing a lot of RE over the years, and if you have discussed it with other experts, there must be something uniquely wrong here. In that case, a more detailed analysis of the problem along with images and controls would help us in troubleshooting. Molecular biology is a bit of voodoo, but REs are quite standardized, and unless there is some unnecessary voodoo going on with your technique, I don't see how this can happen.

But you may have found the next blockbuster to bladder polyps here, so don't give up

---

I still think the insert is ok. I wouldn't say dimers are commonly seen, i don't see them that often. But if i have a lot of DNA, i see them every now and then just because there's always a small percent of them that occur. I still wouldn't be concerned about the dimers though, i don't think it's very important.

---