Gt10 and Gt11

[J Preiss--Seq Anal]

        The comparison between the lambda cloning vectors gt10 and gt11 
provided by A. Hamel is essentially correct. His reason that gt10 is 
unsuitable for antibody screening begs for a simple question. The cI gene is 
transcribed in the intact phage. The repressor protein is an important 
determinant to direct the phage into the lysogenic portion of the lambda life 
cycle. Why should this transcription not occur simply because a piece of DNA 
which does not encode the cI repressor is placed downstream? The important 
distinction is the relative "strength" of the repressor. If the cI promoter is 
near the middle of the promoter scale, the lac promoter, after induction with 
IPTG, is off-scale. Therefore the amount of protein produced by the respective 
promoters is drastically different. 
        Also, the quantitation of successful orientations in gt11 is wrong. 
There are two promoters capable of driving the production of a fusion protein 
from an insert at the unique EcoR1 site. One is the lacZ promoter. The other 
is downstream of the lacZ gene and directed in the opposite direction. So 
either orientation of the insert can produce a fusion protein. The question 
might be differences of amount of fusion protein from the two promoters. I 
have a group of clones in gt11 with some lacZ fusions and some anti-lacZ 
fusions. I cannot distinguish between the two classes using the antigenic 
response of the plaque. Also, do you want an insert encoding the entire 
protein? How would you ascertain that fact unless you have 5' untranslated 
material from the mRNA? Does the 5' untranslated region of your specific 
message lack an in-frame translation stop codon? Many to most don't. 
Therefore, antibody screening intrinsically biases the recovered clones. 
        With respect to the original question of the differences between gt10 
and gt11, is this a history of science question or a serious query to decide 
which vector to use for a cDNA library. I have to wonder why someone is using 
these vectors at this date. Yes the cost is great - a gift from somebody down 
the hall as opposed to paying some biotech concern for their offering. But, to 
parphrase the cogent words of the advertisement long ago for FRAM auto 
filters: You can pay now, or you can pay later. Some dollars up front or some 
hours down the road (time is money sooner or later). i) The current vectors 
allow all but the illiterate to convert the 40 kb lambda entity into a 3kb 
plus insert size plasmid entity in an afternoon. Mapping of the insert just 
became a piece of cake. It is not a trivial matter to proceed from a tube of 
pure lambda phage from a single plaque to a tube of pure plasmid DNA of the 
insert in a vector. ii) The restriction sites that are supposed to delimit the 
insert sometimes become altered. Not a single isolate from the two classes 
described above have an intact EcoR1 site at the insert-vector joint. Don't 
ask me, I didn't make the library. With gt10 and gt11, you have to work to 
learn which of the nearby flanking sites, which may or may not be unique in 
the vector, are not also present in the insert. With the new generation 
vectors, there are numerous unique sites flanking the insert in a phagemid 
after conversion from a lambda genome to a plasmid genome. If I may be so 
rash, might I suggest checking the current catalog from either Novagen (in 
Madison, WI) or Stratagene (in California). They offer choices in the two 
methods of ready conversion from lambda to a phagemid (respectively using 
cre/lox of phage P1 or the encapsidated single-stranded intermediate). 
        Brian Smith-White
        preissj at clvax1.cl.msu.edu (this host is on life support)
        smithb at gaea.bch.msu.edu