34 replies to this topic

[jiajia1987]

Hi people,

I have been trying to cut pet22b(+) vector from novagen using Nde1 and BamH1 for my work, which requires ligation with Nde1 and BamH1.

I have tried the following way of cutting:

1) Cut with either enzyme first, gel purify, then cut with the other and purify again.

2) Double cut with both enzymes for a few hours and gel-purifying the product.

The above two gave me high background, so I got some high fidelity BamH1, which works in the same buffer as Nde1 and tried the following:

-  Cutting with both Nde1 and BamH1(HF) overnight before gel-purifying.

- Cutting with both Nde1 and BamH1(HF) overnight before gel-purifying and taking the purified product for another round of cutting with both enzymes overnight before gel-purifying it again. This would mean double of double digestion and double gel-purification to get the cut vector.

Despite what I did above, I still got back high background during a test ligation without any inserts. With CIP treatment, the background was reduced significantly but this will not help me since I need the cut vectors for cloning, which will eventually be used for selection. Plus, all CIP does is to increase the chances of insert to vector ligation and reduce the changes of vector self ligation.


Thus, as a last resort, I ordered primers flanking the Nde1 and BamH1 RE regions on the vector and amplified the vector with Pfu before cutting it overnight with Nde1 and BamH1(HF) and gel-purifying it. What amazed me (in a not-so-happy way) was the presence of a high background! I do understand that self ligation can occur if the vector was not cut at all (since Pfu produces blunt ends). A single cut would not be able to generate any self ligation since one end would be blunt and the other cohesive. A double cut would not ligate too since the ends are incompatible. The enzymes cant be that inefficient, right? Especially since they are new enzymes and I have tried them out previously and saw double cut vectors for myself on the gel.

Now this is confusing me, is it THAT HARD to get a double cut vector with all the methods that I have tried out? Gel purifying ensures that I cut out and purify what I want but that is not happening. Some of the single-cut or uncut vector still gets purified along somehow. I do soak my gel tank with 0.1M HCl overnight to remove any traces of DNA to get the best possible purified band.

Now I am thinking of incubating my amplified vector with Taq 15mins after PCR to get A hangovers so that self ligation would be minimized. But nevertheless, could someone give me more suggestions?

I would appreciate any insights or help on this matter! I need double cut vectors!!! ARGH!!!

[phage434]

PCR amplify the vector (not the insert) with primers leaving NdeI and BamHI sites (along with 5' overhangs).  Treat with DpnI in the PCR buffer to remove the template circular plasmid.   Purify with column PCR cleanup.  Cut with NdeI and BamHI to provide good cut vector without circular DNA.

[Denny]

I feel your pain and understand the frustration. Let's break this down a bit and see where things are going wrong.

First a few questions:

1) What temp are incubating the digestions at?

2) After your digestions (single or double) are you getting clean, sharp bands to excise? picture would be really helpful, With sizes of vector and insert. Uncut supercoiled vector will move down the gel differently than linearized vector. With a sharp band to excise you should not be getting uncut vector with it. BUT, one of your enzymes might not be working which will still produce a linear vector and depending on how close your restriction sites are, you might not be able to see a size difference if you cut with one enzyme run on gel then cut with second enzyme and run on gel.

1)You could run digestion with enzyme 1, gel purify, set aside sample for later gel.
2)Perform second  digestion with enzyme 2, gel purify and set aside a sample for later gel.
3)Treat remainder with CIP, gel purify.
Run all three samples and uncut vector on a gel in separate lanes and
on the same gel combine the two post digest samples and run in a single well.

Lane 1 - Uncut vector
Lane 2 - First digestion
Lane 3 - Second digestion
Lane 4 - Post CIP
Lane 5 - Combine 1st digestion and 2nd digestion

See what you can see!
Now just a couple of suggestions:

1)Use new NdeI and reduce your incubation times to 30 min @ 37C - for Nde I and to 15 - 20 min @ 37C for Bam HI (or just use your new BamHI(HF) and follow the NEB protocol.
- Standard BamHI will have star activity and cause problems. The longer it incubates, the more damage it can do.

2) A few years ago NEB changed their buffers for this double digestion. Previously, NdeI and BamHI did not play well together. (I still don't trust them to be alone together even with the new buffer, call me overprotective) Be sure to use new buffer for this double digest and not some extra's stored in the -20C for 10 years.

3) After cutting with BamHI, gel purify immediately and treat with CIP according to protocol and purify again.

4) Control ligation reaction with just your CIP treated vector:
- If after you've gel purified and verified by gel that your vector has been linearized you get

Little or no colonies - the vector has remained linear, does not tell you if both enzymes worked, but your CIP treatment is working.

Lot's of colonies - the vector has re-ligated, tells you that one enzyme did not work and your CIP treatment isn't working.

And to answer your question, No this should not be that difficult, but clearly you've got a problem somewhere with the protocol or reagents. If you could post pictures of your gels and explain whose who, we can probably pin point the problem

If after getting clean vector and insert you're having ligation problems, I can address that separately, but let's tackle one thing at a time.
Gotta run, hope this makes, sense

Edited by bioforum, 20 March 2010 - 11:05 PM.

[jiajia1987]

phage434, on Mar 10 2010, 11:25 PM, said:

PCR amplify the vector (not the insert) with primers leaving NdeI and BamHI sites (along with 5' overhangs).  Treat with DpnI in the PCR buffer to remove the template circular plasmid.   Purify with column PCR cleanup.  Cut with NdeI and BamHI to provide good cut vector without circular DNA.

Hi phage434,

Thanks for your reply. I am sorry if I didnt make myself clear enough, but what you just said was what I did and it still gave me a high background.

[phage434]

So, identify what is not working.  Transform the pcr product before and after dpnI treatment.  Transform before cutting with BamHI and NheI.  Transform before and after ligation (without an insert).  You can assess quality of the cut ends by cutting with each enzyme separately, then ligating, to form the double-length fragment.  The ratio of single to double length fragment shows how well the cut/ligation works with that end of the fragment.

[jiajia1987]

Hi there,

Thanks for the reply! My answers are as follows:


1) What temp are incubating the digestions at?

- I am incubating the digestions at 37degC.

2) After your digestions (single or double) are you getting clean, sharp bands to excise? picture would be really helpful, With sizes of vector and insert. Uncut supercoiled vector will move down the gel differently than linearized vector. With a sharp band to excise you should not be getting uncut vector with it. BUT, one of your enzymes might not be working which will still produce a linear vector and depending on how close your restriction sites are, you might not be able to see a size difference if you cut with one enzyme run on gel then cut with second enzyme and run on gel.

- Yes, I do get clean bands to excise. I do not have pictures since I always load everything on my gel for gel purification. I cut my vector from this vector containing an insert, so I will know that digestion has been successful when I see a band representing the insert on the gel. Sometimes, I do see a single cut with a double cut and the insert. When this happens, I will run the insert out of the gel so the single and double cut can be separated with the best possible resolution. It can take 6 hours of running at 60V just to separate these two bands to a sufficient distance to cut them nicely! At other times, I just see one band, which is the double cut and just purify it.



1)You could run digestion with enzyme 1, gel purify, set aside sample for later gel.
2)Perform second  digestion with enzyme 2, gel purify and set aside a sample for later gel.
3)Treat remainder with CIP, gel purify.
Run all three samples and uncut vector on a gel in separate lanes and
on the same gel combine the two post digest samples and run in a single well.

Lane 1 - Uncut vector
Lane 2 - First digestion
Lane 3 - Second digestion
Lane 4 - Post CIP
Lane 5 - Combine 1st digestion and 2nd digestion

See what you can see!

This sounds like a good idea, I will bring this up to my supervisor and try it out!


Now just a couple of suggestions:

1)Use new NdeI and reduce your incubation times to 30 min @ 37C - for Nde I and to 15 - 20 min @ 37C for Bam HI (or just use your new BamHI(HF) and follow the NEB protocol.
- Standard BamHI will have star activity and cause problems. The longer it incubates, the more damage it can do.

- The NEB protocol stated that Nde1 and BamH1(HF) can be incubated at 37degC overnight. Would 30 mins and 15-20mins be sufficient for a digestion? I cut my vector for 2 hours before and the digestion was incomplete.

2) A few years ago NEB changed their buffers for this double digestion. Previously, NdeI and BamHI did not play well together. (I still don't trust them to be alone together even with the new buffer, call me overprotective) Be sure to use new buffer for this double digest and not some extra's stored in the -20C for 10 years.

- Good point, yes, I do use new buffer!

3) After cutting with BamHI, gel purify immediately and treat with CIP according to protocol and purify again.

4) Control ligation reaction with just your CIP treated vector:
- If after you've gel purified and verified by gel that your vector has been linearized you get

Little or no colonies - the vector has remained linear, does not tell you if both enzymes worked, but your CIP treatment is working.

Lot's of colonies - the vector has re-ligated, tells you that one enzyme did not work and your CIP treatment isn't working.

And to answer your question, No this should not be that difficult, but clearly you've got a problem somewhere with the protocol or reagents. If you could post pictures of your gels and explain whose who, we can probably pin point the problem

If after getting clean vector and insert you're having ligation problems, I can address that separately, but let's tackle one thing at a time.
Gotta run, hope this makes, sense



And, I appreciate your help so much. I'm keeping my fingers crossed that everything will work out soon.

[phage434]

The method I suggested does not involve gel purification, which, in my experience, is the cause of many failures.  Have you checked, e.g. for UV exposure problems in your gel preparation?  Are you using long wave (365 nm) UV, or better, blue light?

[Dukey]

jiajia1987, on Mar 10 2010, 12:24 AM, said:

Now this is confusing me, is it THAT HARD to get a double cut vector with all the methods that I have tried out?

Well the short answer is yes it can be! I have cloned what seems like hundreds of constructs and sometimes it is more problematic than others. There are so many parameters that I could write all night, but my initial thought is that the enzyme combination is not going to work for you no matter what you do. Sometimes the sites are too close apart so that when you digest with the first enzyme, you do not leave enough 5' and/or 3' bases around the second site. You should check that. The second digest is always the problem. I always avoid NdeI, call me paranoid but I just do not like it for cloning.  

Also you probably do have a few positives on your plates and if you are desperate enough you could do a colony hybridization and label your PCR primer with P-32 and use it as a probe. Depending on your experience, this can be done easily. This has also worked for me in the past when the background just wont go away.

The bottom line is you could PCR clone this in frame and in the right direction using more "modern" methods and with only minimal digests. I like Clontechs InFusion system for difficult clones - in a word it is outstanding and works EVERY time, FIRST time if your primers are right. Check it out and save yourself heart ache. I have seen PhD students burn literally 6 months in the lab cloning!

[jiajia1987]

phage434, on Mar 11 2010, 10:09 AM, said:

The method I suggested does not involve gel purification, which, in my experience, is the cause of many failures.  Have you checked, e.g. for UV exposure problems in your gel preparation?  Are you using long wave (365 nm) UV, or better, blue light?

Why would gel purification cause many failures? I used to think that it would help me to get the band which I want.     

Yes, I used blue light, no UV used!  

[jiajia1987]

Dukey, on Mar 11 2010, 10:29 AM, said:

jiajia1987, on Mar 10 2010, 12:24 AM, said:

Now this is confusing me, is it THAT HARD to get a double cut vector with all the methods that I have tried out?

Well the short answer is yes it can be! I have cloned what seems like hundreds of constructs and sometimes it is more problematic than others. There are so many parameters that I could write all night, but my initial thought is that the enzyme combination is not going to work for you no matter what you do. Sometimes the sites are too close apart so that when you digest with the first enzyme, you do not leave enough 5' and/or 3' bases around the second site. You should check that. The second digest is always the problem. I always avoid NdeI, call me paranoid but I just do not like it for cloning.  

Also you probably do have a few positives on your plates and if you are desperate enough you could do a colony hybridization and label your PCR primer with P-32 and use it as a probe. Depending on your experience, this can be done easily. This has also worked for me in the past when the background just wont go away.

The bottom line is you could PCR clone this in frame and in the right direction using more "modern" methods and with only minimal digests. I like Clontechs InFusion system for difficult clones - in a word it is outstanding and works EVERY time, FIRST time if your primers are right. Check it out and save yourself heart ache. I have seen PhD students burn literally 6 months in the lab cloning!

Haha, looks like I am not the only one, which certainly makes me feel better!

THe reason why I am sticking to Nde1 and BamH1 is that I have a construct which has been made by inserting using Nde1-BamH1 ligation so all our work uses Nde1 and BamH1. With the construct (or otherwise called insert over here) in the vector, the distance between the two sites can span up to 1.5kb! Surely this space is enough to ensure that the second enzyme binds? And by the way, why do you dislike Nde1? Any reason?

[jiajia1987]

phage434, on Mar 11 2010, 09:32 AM, said:

So, identify what is not working.  Transform the pcr product before and after dpnI treatment.  Transform before cutting with BamHI and NheI.  Transform before and after ligation (without an insert).  You can assess quality of the cut ends by cutting with each enzyme separately, then ligating, to form the double-length fragment.  The ratio of single to double length fragment shows how well the cut/ligation works with that end of the fragment.

I am using Nde1, not Nhe1.  

Your method will certainly help me to identify what is wrong in the process. But I do not understand the part about forming double-length fragment. Where does that part come in?

[Dukey]

jiajia1987, on Mar 10 2010, 06:35 PM, said:

Dukey, on Mar 11 2010, 10:29 AM, said:

jiajia1987, on Mar 10 2010, 12:24 AM, said:

Now this is confusing me, is it THAT HARD to get a double cut vector with all the methods that I have tried out?

Well the short answer is yes it can be! I have cloned what seems like hundreds of constructs and sometimes it is more problematic than others. There are so many parameters that I could write all night, but my initial thought is that the enzyme combination is not going to work for you no matter what you do. Sometimes the sites are too close apart so that when you digest with the first enzyme, you do not leave enough 5' and/or 3' bases around the second site. You should check that. The second digest is always the problem. I always avoid NdeI, call me paranoid but I just do not like it for cloning.  

Also you probably do have a few positives on your plates and if you are desperate enough you could do a colony hybridization and label your PCR primer with P-32 and use it as a probe. Depending on your experience, this can be done easily. This has also worked for me in the past when the background just wont go away.

The bottom line is you could PCR clone this in frame and in the right direction using more "modern" methods and with only minimal digests. I like Clontechs InFusion system for difficult clones - in a word it is outstanding and works EVERY time, FIRST time if your primers are right. Check it out and save yourself heart ache. I have seen PhD students burn literally 6 months in the lab cloning!

Haha, looks like I am not the only one, which certainly makes me feel better!

THe reason why I am sticking to Nde1 and BamH1 is that I have a construct which has been made by inserting using Nde1-BamH1 ligation so all our work uses Nde1 and BamH1. With the construct (or otherwise called insert over here) in the vector, the distance between the two sites can span up to 1.5kb! Surely this space is enough to ensure that the second enzyme binds? And by the way, why do you dislike Nde1? Any reason?

No what is important is free bases EITHER side of the enzyme recognition site. This is the reason why you add poly-A tails to PCR primers that are to be used for cloning. Some enzymes are very picky and need lots of room around their binding site to work. When you digest the plasmid with one enzyme, depending on the closeness with the second site, you may leave only a few bases on one side of the recognition sequence. For the insert (I did my PhD in UK!), it is even worse as one side is completely without DNA IF you do not add a poly-A tail to the primer.

NdeI is just not an enzyme I use often and when I have I have always thought it to be inefficient. I have no data to back it up, just personal feeling.

[Denny]

jiajia1987, on Mar 10 2010, 09:02 PM, said:

Hi there,

Thanks for the reply! My answers are as follows:


1) What temp are incubating the digestions at?

- I am incubating the digestions at 37degC.

2) After your digestions (single or double) are you getting clean, sharp bands to excise? picture would be really helpful, With sizes of vector and insert. Uncut supercoiled vector will move down the gel differently than linearized vector. With a sharp band to excise you should not be getting uncut vector with it. BUT, one of your enzymes might not be working which will still produce a linear vector and depending on how close your restriction sites are, you might not be able to see a size difference if you cut with one enzyme run on gel then cut with second enzyme and run on gel.

- Yes, I do get clean bands to excise. I do not have pictures since I always load everything on my gel for gel purification. I cut my vector from this vector containing an insert, so I will know that digestion has been successful when I see a band representing the insert on the gel. Sometimes, I do see a single cut with a double cut and the insert. When this happens, I will run the insert out of the gel so the single and double cut can be separated with the best possible resolution. It can take 6 hours of running at 60V just to separate these two bands to a sufficient distance to cut them nicely! At other times, I just see one band, which is the double cut and just purify it.


Excellent, sounds good so far

1)You could run digestion with enzyme 1, gel purify, set aside sample for later gel.
2)Perform second  digestion with enzyme 2, gel purify and set aside a sample for later gel.
3)Treat remainder with CIP, gel purify.
Run all three samples and uncut vector on a gel in separate lanes and
on the same gel combine the two post digest samples and run in a single well.

Lane 1 - Uncut vector
Lane 2 - First digestion
Lane 3 - Second digestion
Lane 4 - Post CIP
Lane 5 - Combine 1st digestion and 2nd digestion

See what you can see!

This sounds like a good idea, I will bring this up to my supervisor and try it out!


Now just a couple of suggestions:

1)Use new NdeI and reduce your incubation times to 30 min @ 37C - for Nde I and to 15 - 20 min @ 37C for Bam HI (or just use your new BamHI(HF) and follow the NEB protocol.
- Standard BamHI will have star activity and cause problems. The longer it incubates, the more damage it can do.

- The NEB protocol stated that Nde1 and BamH1(HF) can be incubated at 37degC overnight. Would 30 mins and 15-20mins be sufficient for a digestion? I cut my vector for 2 hours before and the digestion was incomplete.


I very rarely run my digest more than an hour, usually stick close to the 30 minutes unless I'm having problems and consistently get high percentage (95%+) of cut product. The NdeI won't be a problem running longer, but I cringe whenever I have to use BamHI and keep the time short on that one.
I'll check out the NEB protocol, I'm surprised that they recommend overnight with BamHI, but when you've been using it for so long and used to doing things a certain way, you don't check to see if the recommendations have changed. These incubation times have not been a problem for me and I've cloned thousands of different genes and worked with a multitude of vectors. When I've forgotten a digestion reaction  , in the heat block over night and run it out on a gel the next day, I've gotten a mish mash mess of fragment sizes.

Here's my protocol just for reference:
For one Enzyme/20ul rx
10 ul DNA
2 ul Buffer
0.2 - 0.5 ul BSA (BSA is like butter, it makes a bitter batter better  )
1 ul Enzyme
6 ul dH20




I've used gel purification for all my cloning/subcloning work using blue light in the imaging box and it's worked very well for our lab.
This can be VERY frustrating, and there are things happening that don't make any sense.
Now you've got fresh template, but I wonder if you amplified a possible problem with your original vector?
Do you have stocks of the empty vector?
Maybe it's best to go back to the source and make fresh preps of the original uncut empty vector.
Just thinking out loud here...........

To see if both enzymes are currently working set up two reactions:
One with BamHI cutting first and One with NdeI cutting first then check by gel.
I've had NdeI go bad in the past and seem to quit working for no apparent reason - several times.
Good luck........let us know how it works out and if you find the problem.





2) A few years ago NEB changed their buffers for this double digestion. Previously, NdeI and BamHI did not play well together. (I still don't trust them to be alone together even with the new buffer, call me overprotective) Be sure to use new buffer for this double digest and not some extra's stored in the -20C for 10 years.

- Good point, yes, I do use new buffer!

3) After cutting with BamHI, gel purify immediately and treat with CIP according to protocol and purify again.

4) Control ligation reaction with just your CIP treated vector:
- If after you've gel purified and verified by gel that your vector has been linearized you get

Little or no colonies - the vector has remained linear, does not tell you if both enzymes worked, but your CIP treatment is working.

Lot's of colonies - the vector has re-ligated, tells you that one enzyme did not work and your CIP treatment isn't working.

And to answer your question, No this should not be that difficult, but clearly you've got a problem somewhere with the protocol or reagents. If you could post pictures of your gels and explain whose who, we can probably pin point the problem

If after getting clean vector and insert you're having ligation problems, I can address that separately, but let's tackle one thing at a time.
Gotta run, hope this makes, sense



And, I appreciate your help so much. I'm keeping my fingers crossed that everything will work out soon.

[b]

Edited by Denny, 10 March 2010 - 07:21 PM.

[phage434]

Sorry, NdeI not NheI.  If you cut a linear fragment (PCR product, e.g.) with NdeI only, then one end is cohesive to other NdeI sites, the other end is inactive.  If you ligate, you get double length fragments, if the cutting and religation works.  Same with the BamHI site at the other end.  No other side reactions occur, so you can easily determine the cutting and religation efficiency.

[jiajia1987]

Thanks for the recipe reference! Just a question here, you stated 10ul of DNA, how much of the DNA would that be? How many ng or ug?

And thanks for the advice, I will follow them and see what went wrong and post here again when everything gets right!