A few weeks ago I posted a request for information on peoples practical
experiences correlating binary vectors and Agrobacterium strains with
T-DNA integration patterns (when infiltration was used for
transformatio). Unfortunately I only got two replies to the request from
Drs. Mark Aarts and Alan Rose whom I thank very much for their trouble and
the info provided. On the other hand, I did get quite a number of requests
to share the information (and I did promise I would). I do apologize for
the delay in passing down the messages [but better later than never] which
can be found below. If anyone else would like to add to the list, please
do not be shy.
[my] original message
'The integration patterns of T-DNAs (namely getting "single copy" versus
"repeats" at a locus) depend on a variety of factors. Included are the
VECTOR used, Agrobacterium strain used and the transformation method.
I am seeking binary vectors (and strains of bacteria) that when used for
transformation via infiltration will tend to yield transformants carrying
a single T-DNA copy at the integration site (aren't we all?). If you have
any tips on this, please share them. I would also be very happy to hear
about vectors/strain combinations that upon transformation by this method
would give the"undesirable" T-DNA repeats at the integration sites.
I will collate and post back any replies as this information might be of
general utility. Thanks in advance for your help.'
>From Dr. Mark Aarts
"I have used pBINPLUS derived vectors in GV3101 or C58 A. tum., which
frequently give multiple copies, including tandem and inverted repeats,
transformed to Arabidopsis (Ws) by dipping. I am very interested in the
responses you get.
Dr. M.G.M. Aarts
Plant Research International
>From Dr. Alan Rose
Great question. I am very interested to hear what you learn from the
For all of the following data, Arabidopsis ecotype Columbia was vacuum
transformed using Agrobacterium strain C58C1(pMP90), also known as
GV3101(pMP90). Primary transformants (T1) were selected on kanamycin
plates, their progeny scored for those giving a 3:1 ratio of KanR:KanS,
and the putative single locus transformants screened by Southern to
determine single vs multiple inserts.
Using the vector pEND4K, a total of 29 very similar constructs were
tested. Of the total of 723 T1 lines screened, 440 (60%) apparently had a
single locus of insertion. Of those, 87 (12% of transformants, 20% of
single locus lines) were shown to be single copy and the rest had two or
I had much worse luck with 5 very similar pBI101-based comstructs. Out of
142 T1 lines, 72 (51%) gave a 3:1 segregation ratio, but just 3 lines (2%
of transformants, 4% of single locus lines) had a single copy insert. The
majority of the lines had multiple inserts, many more inserts in each line
than obtained using the pEND4K-based plasmids.
University of California, Davis"
Dept. of Genetics
U. of Cambridge
E-mail: pscfr2 at cam.ac.uk