Best Enzymes for 5' RACE? (II)

Jon Nakamoto jnakamot at
Sat Feb 4 18:43:56 EST 1995

In article <3guu45$cd8 at>
rmauk at (Rob Mauk) writes:

>   1.  What is the enzyme of choice for 1st strand cDNA synthesis for 5' 
>        RACE, given that my 5' end is probably 3kb or so away, and also 
>        that there is likely some nasty secondary strux to be crossed? 

Most people I talk to like RNase H- MMLV reverse
transcriptase.That's what's used in the Marathon cDNA kit you're
considering (using 32-P in a "lock-docking" oligo-dT primed 1st
strand synth, followed by 2nd strand synth, I got strong signals
from 700-4500 bp, with progressively weaker signals up to 7000 or
so; with a gene-specific primer, I got a dominant band around 3400
bp, with weaker signal surrounding it). I had to cross some pretty
"nasty secondary strux" (about 90% G-C, with lots o' loops), too, so
I think the enzyme did OK. I have in the past used Retrotherm RT, a
thermostable RTase from Epicentre, to plow through some G-C rich
areas, but I don't think this is processive enough for your purposes
(best used < 1000 bp). Gibco's Superscript II (another RNase H- MMVL
RTase) is supposedly modified so that you can run the reaction at 50
degrees C, which might help. Sorry I can't comment on rTth -- I only
tried it once, without success (probably my fault, not the
>    2.  Would a gene-specific primer or random hexamers be preferable?

Both can work (gene-specific primers have an edge for more optimal
annealing temps; hexamers can be useful for getting a tough 5' end;
gene-specific primers usually require some additional nested primers
for the 5' RACE reactions; etc, etc). 

>    3.  What is the preferred combo of polymerases for long distance PCR?

Clontech recommends either a Taq/Pwo mix (e.g., Boehringer's Expand
PCR system) or Takara's LA PCR kit. They claim to get slightly less
efficient rxns with Taq/Pfu or Taq/Vent combos, and do not recommend
Tth, which they say gives lower efficiency and higher background. 

> I have made and screened two cDNA libraries of the same tissue, one of 
> which is dT-primed, and the other random-primed.  All the clones I have 
> pulled out have the same 5' end, which I am certain is not the true 5' 
> end.  I plan to try the Clontech marathon cDNA RACE kit, perhaps 
> substituting a thermostable reverse trancriptase.  (Anyone have 
> experience with rTth?) 

I was pleased at the performance of the enzyme in the kit. If you
substitute a thermostable RTase, I think you'll lose some of the
advantage of the kit, due to the need to go through buffer changes
(gotta get rid of that Mn++). 

 5% DMSO in the RT rxn has also been suggested to 
> me.  Curiously, despite this kit's name, it does not come with any 
> enzymes for the PCR, hence question #3 above.  

I think this is a license issue. Boehringer's Taq/Pwo mix costs
somewhere in the range of $72/100 U (you'll use 2.5 U per 50 ul RACE

 Regarding first strand 
> primers, would the use of random primers be compatible with elevated RT 
> rxn temperature?  Is part of the logic in using random primers that they 
> aid in cutting down on RNA secondary strux?

My understanding is that they simply bypass the area of secondary
structure by binding upstream. 

E-mail in a week or so when I find out how the Marathon kit works
(for those who saw my original moaning & groanings about the price
of this kit, I'm eating crow -- but I need my complete cDNA
YESTERDAY). So far, so good. 

I've no connections with any of the companies mentioned above, and
views are my own, etc. etc....

Jon Nakamoto
Clinical Instructor of Pediatrics/Endocrinology, UCLA
jnakamot at

More information about the Methods mailing list