Pharmacia Recomb. phage Ab kit (ScFv)

Thu Oct 26 17:32:01 EST 1995


We had similar problems with the kit.  We spent quite some time, money and 
efforts figuring out what the problem could be.  Pharmacia's tech-support 
was of no significant help.  Their answers were : 

1."it is very TRICKY to get the assembly of VH and VL fragments done"

2."send us a photograph of your gel so that we can confirm that you are 
using equal amounts of VH and VL PCR products in the assembly reaction"

3."different people have different judgments on how 50ng of VH or VL PCR 
product DNA looks on a gel"

The tech-support could not offer one useful suggestion.  On one fine day, 
they suggested undertaking three rounds of biopanning and amplification of 
ScFv phage library over antigen before testing in ELISA.  We did that, but 
without any success.

The problem lies in the degeneracy of the 5-prime VH and VL primers that 
Pharmacia provides.  We think that because of the degeneracy, these primers 
may anneal to the target Ig mRNA (from hybridoma) even though there is some 
mismatch.  Since the hybridoma produces only one mRNA species for VH and VL, 
the mismatched primers may yield a range of altered nucleotide sequences in 
the PCR product, thereby giving several altered protein sequences of the 
resultant ScFv.  Furthermore, the mismatch annealing may also shift the 
reading frame in the VH or VL fragment.

In any of the above conditions, the result would be identical : the 
assembled ScFv will not bind the antigen.  You will be able to amplify and 
clone the DNA, but you may not produce the CORRECT PROTEIN.

In case of mRNA prepared from splenocytes of immunized mice, multiple mRNA 
species for VH and VL will exist, and one of those WILL encode the V region 
protein that binds the desired antigen.

In our case, we ended up amplifying the VH and VL regions from hybridomas 
using Novagen's primers that bind outside the antibody framework regions, 
and then sequencing those regions to find the EXACT sequence of the first 
and the last framework regions.  We then synthesized our own primers that 
matched EXACTLY these sequences, re-amplified the VH, VL regions, and 
assembled ScFv.  We are testing these ScFv clones in ELISA at this time.

The conclusion from Pharmacia's kit is very simple : Don't take their word 
(or technical excuses) for it.  Change, adapt and succeed.  Good luck.

Rajendra V. Deshpande, Ph.D.
Molecular Biology Research
The Upjohn Company
Kalamazoo, MI 49007
e-mail : rvdeshp0 at

______________________________ Reply Separator _________________________________
Subject: Pharmacia Recomb. phage Ab kit (ScFv)
Author:  debbritt at (Deb Britt) at INTERNET
Date:    10/26/95 1:03 PM

We have been trying to construct ScFv antibodies using Pharmacia's 
Recombinant Phage Antibody system.  After hashing our way through various 
technical problems, we finally got phage that gave a decent reaction by 
ELISA.  We tried to cut the ScFv DNA insert back out of the positive 
clones, but didn't see the right size band on the gel.  Finally tried PCR 
using vector primers, and got a great product, but the wrong size.  It 
looks like we have cloned just the heavy or light chain variable region, 
rather than the complete ScFv.  This theoretically shouldn't be possible, 
since only the complete ScFv has the NotI and SfiI sites for cloning.  I 
can see where maybe a small percentage of clones might end up with only 
half the ScFv, but we have screened literally dozens without getting one 
good one. This is very frustrating (and expensive!), and we have gotten 
only one decent ScFv from three different hybridoma cell lines. Has anyone 
else had this experience?  I would also be happy to hear from anyone who 
has successfully used this system to produce ScFv's.  Did you follow the 
protocol, or make your own modifications?

Thanks for any input.


Deborah Britt, Ph.D.
Medical Oncology
RI Hospital/Brown University

More information about the Methods mailing list