dimtry at BSCR.UGA.EDU dimtry at BSCR.UGA.EDU
Thu Apr 6 20:29:26 EST 1995

Many thanks to all who responded to my Kan question. Responses
follow; in brief, the options to try are:
1. Drop Kan conc to 20-30 mg/l
2. Lower light intensity
3. Increase Kan in initial selection of primary transformants
4. Some people suggested that media (GM vs PNS) and density of plating
are also important.
I also have to point out that my problem with selection of T2, T3 etc
on Kan-50 concerns transformants of RLD ecotype; it may not be the case
for other ecotypes  - further comments on that will be appreciated.

Dima Belostotsky
U of Georgia

Try lower levels of kanamycin.  Even when transformed tissue is selected on
Kan50, we find that Kan at 20 ug/ml is best for testing transformed

Brian Keith


we have seen a similar situation in my lab with a construct in which the 
kanamycin resistance gene converged (transcriptionally) with the other 
gene on the T-DNA. Virtually all transgenic events expressed KanR poorly, 
although the expressed the other marker satisfactorily. We interpreted 
this effect as caused by run-through transcription from the marker gene 
(35S promoter and SSU terminator) antisensing the npt message. We have no 
direct evidence that this hypothesis is correct. In many other 
T-DNAs in which the genes are not converging we saw good KanR.
Luca Comai


Hi!  Have you ever tried Sigma K4378?  Good Luck on your experiment!

Tsai-Yun Lin


I would try a couple of things:
1) Lower the kanamycin concentration in the MS media (e.g. 20ug/ml)
2) Lower the light intensity in the growth chamber you use for the 
germination test or cover the plates with paper or cheescloth.
Good luck. Let me know if it helps.
James Zhang
Carnegie Institution of Washington


I assume your parental plants were original transformants selected on 
kanamycin at 50mg/l. I originally selected for transformants on 
kanamycin at 50 mg/l and also had similar problems in the next 
generation. This was overcome by increasing the initial selection to 
kanamycin 100mg/l. This presumably prevented the growth of 
transformants not expressing the NPTII gene at a high enough level, 
or prevented chimeric tissues growing in which the L2 layer is not 
transgenic. The L2 layer gives rise to the next generation.

Cameron Johnson
Genetics and Developmental Biology Dept.
Monash University, Clayton


Hi Dima,

I don't tkink I can be much help to you re: high susceptibility to kanamycin,
but I am interested in what you learn and about your kan PCR primers.  I am
also following heritability of kan selectable marker in transgenic Landsberg
ecotype.  I find that if anything, my plants WITHOUT the selectable marker
show some slight resistance to kan plates (on Somerville minimal nutrient
agar plates), such that I germinate my seeds on 250 
ug/mL kan, which makes
the resistant plants grow slowly and become somewhat chlorotic. It takes the
susceptible plants along time to die, and so my correlation for correct 
selection is not absolute.  Hence, your PCR experiment sounded good to me,
and any info you might pass along about the source of primers or availability
would be appreciated.  Good luck,
Chris Rock
Quatrano lab


In our hands, it is very common and annoying to see kan resistance fade in 
subsequent generations.  I do not know what to do about it.


Dr. Belostotsky,  I was just reading a paper that describes loss of kan

Deroles, S.C. and R.C. Gardner (1988) Expression and inheritance of
kanamycin resistance in a large number of transgenic petunias generated by
Agrobacterium-mediated transformation.  Plant Molecular Biology 11:

If I remember the contents of that paper right, you may be able to select
on a lower conc of kan.

Also, it has been shown that the level of expression of two genes linked in
one T-DNA construct is not related one to the other.  That is, just because
your reporter is expressed at a high level doesn't mean a linked gene will
be also.  That was work by Dunsmuir's lab published around 1987-9.

Sincerely,  Jill Deikman


Der Dr. Belostotsky,

The kan-resistance conferred by pBIN19 derivatives (nos promoter-nptII) is
often weak.  Also the kan-resistance is somewhat dependent on density of
seedlings on plates.  When I select transformants after vacuum infiltration
transformation, I use kan-50 (4,000 seeds/15 cm plate).  For the next
generation with a much lower density (50-100 seeds/9 cm plate), I use
kan-30.  Some transgenic lines are affectted even with this concentration
of kan (usually apical meristems are the most sensitive part).

Fumi Katagiri
Ausubel lab


Dima, ...   I should
point out that the best amount of Kan may depend on what medium you're
using.  We usually score transformed seedlings on PNS (somerville's medium,
recipe below) containing 20 ug/ml Kan.  If you're using GM from the
Valveekens protocol, you may need to use a bit more - I'm afraid it's been
somewhat variable in our hands.  The PNS Kan-20 has been a reliable test,
though.  As always, root growth is the most informative marker for



Per liter

Place about 700 ml ddH20 in a beaker, then add:

250 mM KPO4 (pH 5.5)    10 ml

1M KNO3                 5 ml

1M MgSO4                2 ml

1M Ca(NO3)              2 ml

FERRIC EDTA             0.018 gm    (or, dissolve 0.073 gm in 10 ml H20,
add 2.5 ml)

Micronutrients (1000X)  1 ml

Sucrose                 5 gm

Stir until sucrose is dissolved, then bring volume to 1 liter

Place 500 mls in each of two bottles containing one of the following
(depending on your needs):

3.5 gm Bacto Agar (final 0.7%)

Autoclave 25 minutes on liquid cycle, cool to 550Cand add any supplements
before pouring plates

1000X Micronutrient solution:

Compound                        per liter               Concentration

H3BO3                           4.33 gm                 70 mM
MnCl2.4H20                      2.77 gm                 14 mM
CuSO4.5H20                      0.125 gm                0.5 mM
ZnSO4.7H20                      0.288 gm                1.0 mM
Na2MoO4                         0.048 gm                0.2 mM
NaCl                            0.584 gm                10 mM
CoCl.6H20                       0.0024 gm               0.01 mM


Would you mind post ing your responses on the net, I would be interested in 
seeing what others are finding. I am in the middle of a large screen of 
transgenics on kan50 plates and my number of kan-resistant plants is also 
quite low. 

Thanks in advance

Sharon Regan


I am having the same problem with kanamycin selection.  The transgenic 
Arabidopsis were selected with kanamycin but the resulting seeds either do not 
survive selection or only a small percentage do.  I am not having problems with 
germination.  I have done southern blots to confirm the presence of the 
transgene. However, I have other transgenic lines that respond to kanamycin as 
expected.  I therefore do not think that this problem is due to a contaminant in
the kanamycin.  I have used a callus selection procedure to analyze some of my 
problem lines and I have found that the leaves of some problematic plants do 
generate callus tissue on callus inducing medium supplemented with 50ug/ml 
kanamycin. Perhaps the expression of npt II is very low in these lines and 
seedlings may be more sensitive to kanamycin.  I don't have the answer to your 
problem but I would be interested in receiving the responses to your question.
Jed Doelling

Jed Doelling
Biology Department, Washington University


Dear Dima,
I had very similar problems in selecting progeny of tobacco transformed
plants after plating them in MS medium plates. The Km concentration
which is most often used that is 100 microgram/mL was not adequate
and therefore I had to test different Km concentrations till I saw
a difference between transformed and not transformed plants in a
Km concentration of 500 microgram/mL. Currently I am testing wether
the same concentration of km to the agar overlay will make any difference
or not. I am using the plant tissue tested Km from Sigma.
I hoppe that this info is usefull for you. Please forward me if you
have any information from successful and unsucessful trials.Thanks,
Elias Anastassopoulos, IMBB, CRETE, HELLAS.

More information about the Arab-gen mailing list