HELP!! transformation problems

Frank O. Fackelmayer fof1 at chclu.chemie.uni-konstanz.de
Wed Aug 5 07:51:49 EST 1998


For routine cloning, where maximum competence is not an issue, we always use
the TSS method of Chung et al. 
It is very reliable once you got it to work. Thats how I do it:

* grow bacteria in 10ml LB to OD=0.8 (we have better results at 0.8 than at 0.4)
* spin down for 5min at 8000g
* decant supernatant
* add 500ul of ICE-COLD LB, resuspend by vortexing
* add 450ul of ICE-COLD 2xTSS (see below), and, simultaneously, 50ul of (RT
warm) DMSO
* mix by very gentle shaking (NO VORTEX!!)
* add 150ul of suspension to the DNA in a prechilled microcentrifuge tube
* mix by gentle mixing (AGAIN NO VORTEX!!)
* let stand on ice for a minimum of 10min (usually 30min)
* plate onto LB-agar plate with appropriate antibiotic, e.g. amp
* grow overnight at 37C

Ok, there are some points to consider:
* to make 2xTSS, dissolve 4.1g MgCl2 and 40g PEG-8000 (other PEGs also work)
in 200ml LB. Sterilize by autoclaving. After autoclaving, there will be a
phase separation as PEG settles. Mix the phases by shaking vigorously. The
solution will be slightly turbid, but is ok. Turbidity will disappear after a
few days. We store the solution at 4C where it is stable forever.
* DO NOT make up the TSS solution 1x to directly resuspend bacteria. It simply
does not work for whatever reason.
THIS IS MOST PROBABLY THE REASON WHY IT DOES NOT WORK IN YOUR HANDS!!!!
* it is better to autoclave than to put thru a 0.2um filter, because some
filters contain detergents that will inhibit bacterial growth.
* always use ICE-COLD LB for resuspending bacteria, and ICE-COLD TSS. DMSO
must NOT be put on ice as it solidifies.
* never use a vortex to mix the bacteria with LB or TSS
* there is no need for a recovery period before plating, nor for glucose

Using this method, we routinely get a efficiency of 10e6 to 10e7. Never 10e8.
This is sufficient for routine cloning, but NOT sufficient e.g. for library
construction. Best method for high efficiency is electroporation, or the Inoue
protocol for chemical competence.

Hope this helps
Frank

PS: Do not try to freeze the bacteria. They will go down in competence at
least a factor of 100. Well, that might still be ok for transformation of
supercoiled plasmids, but does not work for ligations any more.


P.J. v. Santbrink wrote:
> 
> Dear fellow scientists,
> 
> I'm having trouble to get my Coli's competent (chemical method).
> 
> I follow the protocol  from Chung et al (PNAS Vol 86, 1989):
> 
> -    pick a single colony (DH5 alpha), grow overnight in 5 ml LB-medium.
> 
> -    add 310 µl overnight culture to 31 ml prewarmed LB-medium: grow
> 37°C
> -    control OD600 at time intervals till OD is between 0,3-0,4 (in my
> case OD was 0,38)
> -    spin bugs down (4000 rpm, 4°C, 20 minutes).
> -    remove sup, and what I do is resuspend the bugs in the few drops
> that are left from the sup
> -    add 1/10 of original volume of ice cold TSB (dissolve PEG-3350 in
> LB -pH 6.8- add MgCl2 to final concentration of
>      50 mM filter through 0,45 µm filter, add DMSO, 5 % v/v , place on
> ice) to the tube..
> -    swirl the tube gently: keep on ice for 15 minutes.
> -    add 100 µl of the cells to 1 µl pUC19 (100 ng) in precooled eppie.
> -    incubate on ice for another 25 minutes.
> -    add 0,9 ml LB + 20 mM glucose.
> -    grow for 1 hour add 37°C, while shaking
> -    plate 100 µl on LB-agar plate with 100 µg/ml AMP
> -    grow overnight at 37°C.
> 
> Result: 2 colonies (200 CFU/µg DNA should be like 10e7-10e8 )
> 
> Questions:
> 
> Can someone explain to me  why the cells aren't competent ?
> 
> I need at least 10e7 CFU/µg DNA; does someone have a protocol that can
> give me that kind of efficiency ?
> 
> Since we don't have the means to perform electroporation and don't have
> the money to buy those ridiculous priced competent cells sold
> commercially  I've to make them competent myself, but I've tried it
> several times (different incubation times, fresh solutions etc) but
> failed each time, so if someone out there can help me out I really would
> appreciate it.
> 
> Peter van Santbrink
> Division of Biopharmaceutics
> University of Leiden
> The Netherlands
> E-mail: santbrin at LACDR.leidenuniv.nl



More information about the Methods mailing list