Here is a summary of the responses to my question
about designing a primer for 5' RACE with
Arabidopsis mRNA / cDNAs.
Thank you to those who replied.
russell at dogwood.botany.uga.edu
> We need to do some 5' RACE (or anchored) PCR with
> Arabidopsis cDNAs. In some protocols G or C tailing
> is used, in others, T or A tailing is used because of
> the lower Tm.
>> We wondered if the relatively high AT content of
> Arabidopsis might make it a bad idea to use T or A
> tailing of 5' ends of cDNAs. Could the high internal
> AT content cause spurious PCR priming?
>> Does anyone have suggestions about this?
A slightly different answer could be to use 5' RACE where one ligates
a primer to the 5' end of the RNA. I have done this using Lemna RNA
using a kit supplied by Clontech. Providing the RNA is in good
condition this works very well in my hands (expensive though).
University of Sheffield, UK
s.rolfe at sheffield.ac.uk
I have done 5'RACE via tailing with A with some success; although simce I did
this new protocols came out which employ ligation of defined sequence oligo
onto 5' end which (supposedly) eliminates spurious priming.
Ref is Liu and Gorovsky NAR 21(21)4954-4960'1993
dimtry at bscr.uga.edu
We had some problems RACEing out a couple of cDNAs from tomato. We got
bands, but they were shorter than predicted. This was due to the
adaptor-T oligo annealing to an A-rich region in our cDNA. So yes, it is
a potential problem.
We got round it by doing PCR on cDNA libraries using forward or reverse
primers and a primer for our gene. It worked really well, but it depends
on whether you have a cDNA library.
"James S. Keddie" <keddie at nature.Berkeley.EDU>
An empirical result that we obtained trying 5' RACE on jojoba RNA showed that
A tailing worked better than C tailing. The only difference was in the
experiment was that in one reaction we C tailed and in another reaction we A
tailed the 1st strand cDNAs. The 2nd strand primers were identical except
for the presence of Gs or Ts. The A tailing gave us clones while the C
tailing did not.
"Lassner" <lassner at maillink.calgene.com>
The solution to your RACE question could be SLIC. This is modification of RACE
that work very well (at least in my hands). I use the Clontech kit,
5'Amplifinder RACE kit.
Salinas at cit.inia.es
I have had good luck tailing with C's. The BRL kit was
recommended to us and worked very well. They have 5'
and 3' RACE kits. I've only used the 5' RACE kit and it
worked very well. Another tip if this is the first time
you are doing this-just because you don't see anything
with the first PCR amplication doesn't mean something
isn't there. A couple of times I couldn't see anything
with the first round of PCR but got a band with the
sheveld at rockvax.ROCKEFELLER.EDU
I don't know about Arabidopsis, but the following protocol has worked
fine for me on pea, which is also quite A/T rich. Two other labs here has
tried it as well, in fact, one of them did it on Arabidopsis, and they
seemed to be happy with the result.
I do recommend to make primer extension first (if you have not already
done this) so that you know exactly how much of the cDNA is missing.
Igor Kardailsky JII <igk at sys.uea.ac.uk>
John Innes Institute
5'RACE protocol (amplification of 5' ends of cDNAs)
after Frohman et al., PNAS,85 pp763-773 with modifications by
In this protocol two antisense oligos are used to avoid
amplifying and cloning artefacts. Oligo1 (upstream in anti-sense)
serves as primer for reverse transcription, oligo2 is for PCR;
you'll also need a pair of primers similar to ones used on the
poly(A) end in 3'race.
Let oligo3 be the one with homopolymeric tail (i.e., dT), and
oligo4 be the one containing MCS (multiple cloning sites only).
Because of its large size I purify oligo3 by reverse phase
chromatography (may be optional?).
Prepare polyA+RNA your favourite way. 1 ug is enough (i used
Dynal oligo(dT) beads on 100ug of total RNA for 1 reaction).
1. Anneal 1pmole of oligo1 and 1 ug of RNA in 15 ul of 1xRT
buffer. 5x RT buffer is 0.3M KCl/0.3M Tris-HCl pH 8.3. Heat to
90 C, let cool to 43 C over ~1 hr. Add 10 ul MDR buffer (20 mM
MgCl2, 20 mM DTT in 1xRT buffer - store at -20), 1 ul of 25 mM
dNTP, 10-20 u RNAsin, AMV RT (0.5 ul?u).
NB the reason to do it in two steps is that you don't want to
heat RNA to 90 C in the presense of Mg... (my thanks go to George
Lomonosoff for this hint).
Incubate at 42 1-3 hrs.
2. Do alkaline hydrolysis - make you reaction into 100 ul of 0.3
N NaOH/10 mM EDTA.
Keep at 65 C for 1 hr.
3. Neutralize with 2 ul conc HCl, 10 ul Tris 8.0 1M; do
4. Put the aq phase on some kind of size-select columns - I used
Pharmacia's S-400 spin columns equilibrated with TE/0.1M NaCl pH
8.0. It is very important not to contaminate your synthesized
cDNA with excess of primer1 - it will otherwise become a major
source of artefacts during PCR.
5. Collect the eluate, add NaAc, EtOH ppt. Wait for at least 1
6. Do dA-tailing: spin the cDNA, wash, diss in 10 ul TE. Add 2
ul of Pharmacia's one-for-all buffer 10x, 1 ul of 1mM dATP, water
to 20 ul, TdT 6 units (terminal transferase may work better in
specialized cacodilate buffer, but this one does the job ...).
Incubate 20 min 37 C, add 1 ul 0.5M EDTA, 5 min at 65 C to kill
For a control of this and subseq steps one can use linear DNA
with protruding 3'-ends, i.e. Pst-cut bluescript 0.1-0.5 ug in
a parallel reaction.
7. Add 25 ul (concentration as shipped) of Dynal oligo(dT) beads
in their 2x binding buffer (20 mM Tris 7.5, 1M LiCl, 2 mM EDTA,
0.4-1 % SDS). Let sit at RT for 15 min. Collect beads, save sn
in the control.
8. Wash the beads twice with 100 ul of 10 mM tris 7.6/0.15 M
LiCl/1 mM EDTA/0.3% LiDS - the washing buffer supplied with beads
to make poly(A) RNA, something similar would do.
9. Do final wash in STE (0.1 M NaCl in TE). Elute the dA-tailed
1st strand in 2x10 ul TE at 65 C.
10(optional). Check how TdT/beads worked by putting equivalent
amounts of control DNA from various stages/phases on a gel and
observing where it goes.
11. Second strand synthesis.
Mix 15 ul of tailed 1st strand, 1 ul 5uM oligo3, 2 ul 10x one-
for-all buffer, 2 ul 0.125 mM dNTP. Heat 1 min 65 C, put on ice,
add 10 u of Klenov fragment of DNA polI. Incubate at 37 C 1 hr,
12. Do PCR using 1 ul of Klenov reaction as a template and
oligos2 and 4 as primers. 35 cycles (30" 94, 1' 48, 1' 72, hot
start) were enough for me. Set up negative controls, i.e. using
oligo4 only. You may wish to have oligo1/oligo4 reaction as well.
13. See if you have expected PCR product in the right reaction.
Clone it your favourite way.
I do CHCl3 extr, EtOH ppt, T4 end repair, magic prep, BamHI
digest to produce a sticky end on oligo4, magic prep, ligation
in blunt-sticky cut vector (no gel purification).
my universal primers:
GCTCGAGGATCCGCGGC contain sites for
XhoI BamHI SacII NotI, Tm=50C.