Site directed mutagenesis oligo desi

[Ken Baker]

>>G'day
>>
>>Is there anyone out there that could offer advice as to the
>>design of an oligonucleotide to introduce a 12bp deletion
>>via site directed in vitro mutagenesis.
>>
>>Also I'd like to solicit opinions as to the easiest and most effective
>>method for doing a site directed in vitro mutagenesis.
>> 
>>Thanks,
>>BILL
>>------------
>>
>>^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^
>>William D. WARREN
>>Center for Agricultural Biotechnology     Email: ww40 at umail.umd.edu
>>University of Maryland at College Park    Phone: (301) 405-7681 
>>
>>^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^~^
>
>
>Depending upon what form DNA (cloned piece, presumably in plasmid vector),
>then how about PCRing away from the 12 bp site that you wish to eliminate?
>PCR from the 5' end of it towards the vector (hopefully a unique sequencing
>primer at that end will be useful) and in a seperate PCR reaction prime
>from the 3' end to the other end of the vector.
>
>
>vector                insert     12 bp       insert           vector
>===================xxxxxxxxxxxx!!!!!!!!!!!!xxxxxxxxxxxxxxxx===================
>              --->         <---            --->            <----
>               A             B              C                D
>
>
>              -----------------            ---------------------   
>              A              B              C                D
>            
>                  5' end                       3' end
>                  PCR product                  PCR product
>
>All primers would work best if already phosphorylated chemically when
>synthesized (purchased).
>With planning the restriction endo. cloning sites within about 5 bp from 
>ends of primers A and D will each be unique to vector and not in insert.
>After "polishing" ends of purified (phenol/chloroform, etc) PCR products
>with T4 DNA polymerase (plus all four dNTP's in reaction), they can be
>blunt end ligated (with 5% PEG, mw 6,000-8,000, see BRL cataloque) at
>conc. of about 50 ng/uL at 15oC for at least 4 hours. Then ethanol ppt'd,
>etc., then ends digested with appropriate rest. endo, subcloned, clones
>sequenced.
 
[...] Text Deleted

>Good luck
>
>Andre Hamel
>Manitoba Vet Virology
>Winnipeg, Manitoba
>Canada
>
>email: hamel at ccu.umanitoba.ca

Hi Andre, Hi Bill.
Andre's method is very similar to one we use here a lot, mutagenic inverse PCR.
Basically it is as above, however, omit primers A and D. We find it is ok to 
kinase the PCR product rather than the oligos, but the latter could give better
results. After PCR, run 10% of the reaction on a gel, to check that the band is
unique, then:
1] 85ul PCR product
   2ul Klenow
   13ul 5nM dNTP (in case the old stuff is used up)
   30 min 37C
   10 min 65C

2] 100ul 5M NH4OAc
   200ul Isopropanol
   spin

3] 14ul resuspended PCR product
   2ul Kinase buffer
   2ul T4K
   2ul 5mM ATP
   30 min 37C
   10 min 65C

4] 10ul PCR product
   1ul 5mM ATP
   2ul ligase buffer
   1ul ligase 
   6ul H2O
   O/N 15-20C

   Ensure you set up a control without ligase to check for template carryover.
   (You will always get a few.)
   PCR conditions are as follows:
   Template:  20ng
   Primers:   500ng each
   5ul x 5mM dNTP
   10ul x10 buffer
   1ul Taq
   qs MgCl2
   ->100 ul with H2O

   Denature 94C 1.5min
   25 cycles
   Denature 94 1 min
   Anneal 50C 1 min

After ligating, either transform with Ca++, or glassmilk then
electroporate. We make the primers about 25-30bp each, and you can
speed up selection of mutant clones by either introducing or deleting
a restriction site by using a missmatch oligo to introduce a silent
mutation. The method can handle a few missmatches, but try to keep
them away from the 3' end of the primers so as not to destabilise
them. We normally use this method (with back-to-back primers) to
introduce single changes into the aa sequence of the expressed
protein, and would put that mutation onto one primer and the
restriction site mutation onto the other: even 4-base frequent cutters
are useful as you can compare the pattern after digestion with that of
the template. Method should give a few dozen to a few hundred
transformants, BUT, you MUST sequence the mutant, and it is a good
idea to restrict out the minimum length possible bounding the ligation
site, and cassette it back into unPCR'd background, as you can get a
few random mutations. If you get low rates of transformation, try
treating the PCR product with protease K 
to clean it up prior to religation. 
Best of luck,

Ken.
*************************************************************************
Dr Ken Baker                              JANET : UK.AC.AFRC.FRIR::BAKERK
Department of Protein Engineering      INTERNET :  BAKERK at FRIR.AFRC.AC.UK
AFRC Institute of Food Research             TEL :        (+44) 734 357139
Reading                                     FAX :        (+44) 734 267917
Berks RG6 2EF
UK                          (AFRC = Agricultural & Food Research Council)
*************************************************************************