Mark D. Garfinkel
mg16 at ellis.uchicago.edu
Wed Mar 16 10:57:51 EST 1994
futers at bpxtal.leeds.ac.uk writes:
> I have used NM1149. The host cells I used were as follows:-
> Selective host - C600
> Non selective host - SM32 (for growing up non-recombinant NM1149)
> I hope this helps.
This is absolutely 100% wrong!!
NM1149, gt10, and other cI-insertional-inactivation vectors rely
upon the hflA host mutation for lysogenic elimination of non-recombinant
vector phage. Standard E. coli C600 is hflA+ and *will not* select
against non-recombinant vector!
For the cI-insertional-inactivation phage cloning do-it-yourselfers:
C600 is the *non*-selective host one uses to grow your own preps
of NM1149, gt10 or equivalent vector phage. One begins by plating a
portion of phage stock at low density using C600, and then picking a dozen
single-plaque isolates. Then you replica-plate these phage substrains on
C600 and on a hflA-mutant strain in order to measure the frequency of
spontaneous cI-mutant phage; dicker around with the dilutions you plate
(think about it...). Of the dozen single-plaque isolates, the spontaneous
clear-plaque-mutant frequency can vary quite widely (up to several orders
of magnitude; this is a Luria-Delbruck fluctuation test of mutation
induction). Choose the lowest-frequency isolate (ca. 10e-4) for large-
scale phage growth & DNA purification prior to library construction.
Plating of a ligated-packaged library for hybridization screening
is done on a hflA-mutant host in order to eliminate the non-recombinant
vector-onlies that had arisen during the ligation & packaging steps. If
you're meticulous about your library construction experiment, then you did
the following ligation, packaging & plating controls before you attempted
to screen the library: (i) vector DNA that's uncut, ligated & packaged,
plated on both C600 and hflA; (ii) vector DNA that's cut, unligated &
packaged & dual-plated; (iii) vector DNA that's been cut, ligated &
packaged & dual-plated; (iv) vector DNA that's been cut, ligated to insert
DNA & packaged & dual-plated. I leave it as an exercise to the reader to
consider *why* one should do all these controls, and *what* the expected
results will be.
PS - In this case the Sambrook, Fritsch & Maniatis (1989) cloning book
got the strain genotypes right (see page A.9 in volume 3).
Mark D. Garfinkel (e-mail: garfinkl at iitmax.acc.iit.edu)
My views are my own, which is why they're copyright 1994 (c)
Ignore the header; I post from here only if I can't post from there.
More information about the Methods