I have recently worked with amplifying a large vector construct based on pcDNA3 which is 18 kB in size.

You could perhaps try the following:

(i) Avoid the double culture method you are employing. Once you have a mini prep DNA and it is correct (size and sequence), use this to retransform a fresh competent bacteria on a plate. I found that STBL3 are great for stability. Then pick a new fresh colony, spike in a starter culture in the morning with Ampicillin (5 mL), then in the afternoon, spike this into 150 mL with Ampicillin.

and/or

(ii) Use your method with left over 200 uL culture. But just leave at RT or at 4 degree until the evening so that is does not overgrow in the small volume. Then once you have done your diagnostic test, spike this entire 200 uL into a large 150 mL volume.

and/or

(iii) When you elute your DNA from the column - try to add pre warmed to 50 degree or so. Helps get the DNA off the column for better yield.

and/or

(iv) -Check the size limit of your column! I was stung before where some midi columns could only bind plasmids up to maximum size of 14 kB and so my 18kB vector was just either getting stuck or just not binding.

and/or

(v) You could also use Carbenicillin as an alternative to Ampicillin.

Hope this helps a bit but can't be sure without knowing the size of your vector.

Yes, large plasmids are a pain to work with. What antibiotic are you using? If ampicillin, bear in mind that those 0.2 mL of culture you are using to inoculate the second 2 mL culture have enough B-lactamase to eliminate the selective pressure of the antibiotic after aprox. 20 min. If you then use that second culture to inoculate a larger culture, you could be inviting trouble.

Also, are you using equivalent mini and midi prep kits? Some Qiagen kits use hydrophobicity, others use anion exchange. Which ones are you using in either case?

It would be great if you could post a picture of those smears you are getting.

I have recently worked with amplifying a large vector construct based on pcDNA3 which is 18 kB in size.

You could perhaps try the following:

(i) Avoid the double culture method you are employing. Once you have a mini prep DNA and it is correct (size and sequence), use this to retransform a fresh competent bacteria on a plate. I found that STBL3 are great for stability. Then pick a new fresh colony, spike in a starter culture in the morning with Ampicillin (5 mL), then in the afternoon, spike this into 150 mL with Ampicillin.

and/or

(ii) Use your method with left over 200 uL culture. But just leave at RT or at 4 degree until the evening so that is does not overgrow in the small volume. Then once you have done your diagnostic test, spike this entire 200 uL into a large 150 mL volume.

and/or

(iii) When you elute your DNA from the column - try to add pre warmed to 50 degree or so. Helps get the DNA off the column for better yield.

and/or

(iv) -Check the size limit of your column! I was stung before where some midi columns could only bind plasmids up to maximum size of 14 kB and so my 18kB vector was just either getting stuck or just not binding.

and/or

(v) You could also use Carbenicillin as an alternative to Ampicillin.

Hope this helps a bit but can't be sure without knowing the size of your vector.

Thank you very much Alejandro, Iwan and Charles! Let me add the missing bits here. I use ampicillin for my final gateway clones as well as for my destination vectors. If i'm working with entry clones for the Gateway, then I use kanamycin but the entry clones are a piece of cake. The average size of these gateway clones ranges between 20kb and 25kb. The vector alone is 17.8kb.

Alejandro, I must add that I did overlook the aspect of selective pressure and so if I understand correctly, you are suggesting that the small volume of the remainder culture is probably a detriment to my construct as there isn't enough B-lactamase which in turn makes it difficult for the construct to propagate, right? Since these constructs appear to be low copy, I am using a lower conc. of ampicillin (20ug/ml as opposed to the standard 50ug/ml). As for the columns, I am using the anion exchange variety. I will also try to upload an image of the smears. Thank you very much.

Iwan, I am already following your suggestion. After I have inoculated the 2ml LB media with the remainder 200ul for about 6hours, I then divide this 2ml of culture in 10 tubes, each with 2ml LB. This is iffy though and does not always work and I can imagine why after having read Alejandro's answer.

Charles, I have tried transforming afresh and trying the midi preps directly instead of going via this process but I guess the size of the big constructs really makes things difficult and so I thought of using the mini to midi procedure. Coincidentally, today itself I have done what you suggest about leaving the remainder culture at 4C and directly inoculating the 25mL of LB. I'll know tomorrow. Also, I have been using pre-warmed elution buffer for cleaning mini preps of these large plasmids and followed it up with diagnostics. The columns which technically can only be used for upto 10kb constructs, have so far not failed me to clean constructs as large as 23kb! However, I will employ your suggestion and pre-warm the QF buffer for eluting during the midi. How does carbenicillin score over ampicillin?

Thank you everyone for your great contributions!

Wow - a very large plasmid indeed.

-Carbenicillin is a bit more stable over time.

-These large vectors can be a burden on the bacteria for sure. So to get over this, for plating out, I used to culture my large constructs at 30 degrees rather than 37 degrees for two days rather than just overnight. You can do the same for liquid cultures. This did help for me.

Good luck and when your method works be sure to let us all know your optimum method!

alejandro martin is right...the wast amount of b lactamase present in the medium will get rid of fresh ampicillin...and therfore compromise the stability of the construct in the bacteria.

from my expt,when facing plasmid instability in iquid culture, i use an old trick described by W studier (j mol biol paper in the 80's) for protein expression in ecoli.

shortrly, instead of inoculating bacteria in liquid LB, I spread a starter liquid culture (I spin this starter, remove the SN) et resustepnd the pellet in LB, and spread it on solid lB agar +amp plate. this way, there is less plasmid loss.

this is usefull when expressing toxic prot in ecoli.

In general I agree with the comments above. Especially with the need to re-transform the plasmid from the mini-prep and to work with the fresh transformants. Some time the cells after the ligation have more than one plasmid and of course they will favor to keep the shortest possible. So, I will use the plasmid from the mini-prep that have the correct size and correct restriction map to transform back into E. coli and then will use alkaline lysis method for large plasmids. There are several available and also will help you to save you on MIDI columns. They could be very expensive.

For instance check these two:

http://www.ncbi.nlm.nih.gov/pubmed/2848541

http://www.bmbq.uma.es/fmp/pdfs/PMBR04.pdf

Good luck,

Stefan

Hi Kunal,

The midi columns from Qiagen (as well as the mini and maxi) can handle up to 150kb plasmids, it is the PCR purification and Gel extraction columns that don't really like >10kb plasmids.

I have cloned and purified 18.6 and 19kb clones and also had problems where the miniprep would be fine but the plasmid was no longer around when I used larger volume cultures (for maxi prep in my case). I found that the E.coli strain was really important for my plasmids. Stbl3, XL-10 gold and DH5a all had the problem you faced but TOP10 cells did not. Also, rather than reducing the amount of ampicilin I doubled the concentration to 200ug/mL (100ug/mL is standard and ampcilin doesn't kill bacterial cells at least the these concentrations so doubling the concentration just slows down the problems with accumulating b lactamase though probably not hugely). I also grew the cells overnight at 30oC not 37oC which reduces your yield but at least gives you a product! My plasmids with kanamycin resistance (~19kb) were actually easier to work with than the 18.6kb ampiclin resistant plasmids.

Following the above I would grow an overnight starter culture from a single colony in 5mL LB at 30oC and then use this at 1000x to innoculate a maxi or giga cultures (500mL or 2L so 500uL or 2mL respectively) which I then grew overnight again at 30oC. I know this is not the method really used for standard plasmids but 18-19kb plasmids are far from standard and the bugs grow far slower so using the above innoculation methods helped improve my yields without clogging up the columns.

Hope this helps!

Good luck

Mustafa

Hi Kunal

I am just wondering if you have had any success with the above suggestions.

I too have had extreme difficulty with my plasmids. In fact mine isn't that large (14kb) however like you, everytime I re-transform (into DH5a as well as Stbl3) from positive minipreps, I get a smear.

I have been told to grow at 30oC, which I attempted however this too has been unsuccessful.

I wish I had read this post before I attempted my O/N cultures again. Perhaps I could have used TOP-10 as suggested by Mustafa to transform my ligations, rather than repeat my cloning via QUIKCHANGE.

Anyway, what I have done now is inoculate a 200ml SOB culture with an aliquot from my positive minipreps (XL-gold). As suggested by Charles, I may grow them for 2 days at 30oC. ( I have actually done this in 200ml LB-broth at 37oC O/N but the pellet after the isopropanol-ppt step, there appeared to be no DNA at all).

Please post what you have found from your troubleshooting I am very interested to know what parameters you used to get successful results

Thanks to everyone all that info was super helpful!

Hi, 

I face similar problems with the entry clone on gentamicin......! Did you overcome these problem, if so How? thanks for the suggestions in advance.

Hello, everyone! Thank you for your professional insights.... I have equally been having similar problems with my constructs, transformed with DH5-alpha cells; but the most baffling thing to me is that my construct size is just about 7kb, and each time i try to retrieve the plasmids from plasmid prep, the concentration is usually very low (33-40ng/microlitr) compared to the original vector, which is an high copy (500-600ng/microlitrE). Any helpful suggestions, please?

Thank you,

Kunle.

I just phoned Qiagen about a method for my 20Kb plasmid. I get good yields from a mini prep but very poor yields from a maxi prep, 130 ng/uL. They suggested grow 500mL culture but double the volume of P1, P2 and P3 to ensure the pellet is dissolved. Then warm the elution buffer to 50 deg.  I'll start with a fresh transfection of my mini prep.

I was wondering how any of you got on as I am having a similar issue myself.

Speaking to the plasmid company, they suggested having many repeat sequences in the plasmid is detrimental to the bacteria. Also to lower the LB volume when preparing the maxi.

Would you recommend to lower the Amp (from 100ug/ml)? And how would you select the clone from the miniprep? I'm assuming copy number plays a part in this... Thanks