dissolving of RNA

David F. Spencer dspencer at is.dal.ca
Tue Mar 25 13:47:07 EST 1997

In article <333751FA.37D at unity.ncsu.edu>, Susan Hogarth
<sjhogart at unity.ncsu.edu> wrote:

> David F. Spencer wrote:
> <snip> 
> > As to this (apparently popular) notion that you can overdry an RNA
pellet, let
> > me state that this is pure hogwash.  If RNA is properly desalted then
you can
> > dry it for a week and it will redissolve just fine.  Note that this is quite
> > different from the situation for larger quantities of even well washed high
> > molecular weight (greater than 20-30 kb) _DNA_.  In this case it is
> > unwise to do more than a light drying if you wish to get the DNA
> > resolubilized in less than days or weeks.
> Again; why? Is it the residual salt?
> -- 
> Susan

High molecular weight DNA dried with salt present is a big problem to redissolve
but that can't be a general explanation.  It is certainly possible, with
judicious spins that leave the DNA fairly loose and thus accessible to the etol
washes, to produce a fairly salt-free DNA (there will still be the obligatory
counterion present of course) and if overdried this will redissolve very slowly.
The simplest explanation for this is simply that of size.  As you observed in
your original query, plasmid DNA can be dried completely and will redissolve
quite readily but normally one doesn't try to dissolve 1 mg of 20-30 kb plasmid
in say 0.5 - 1.0 ml; the supercoiling may complicate the issue a bit but I
would doubt that really big plasmids would behave much differently from large
genomic DNA as far as solubilizing. I must admit that I haven't confirmed that
and although I once generated a suitable test DNA (the 2 ends of HindIII-cut 
lambda ligated via their cos ends and that product cloned into HindIII-cut 
vector), I just didn't have any need for mg quantities of that DNA. [Well, OK,
I didn't have any particular use for microgram quantities either!]

I am quite confident as well that 20-30 kb single-stranded DNA would dissolve as
easily as similarly-sized RNA but that experiment will be a bit of a challenge 
as neither test material is readily available.

The dissolving of DNA or RNA requires the molecules to be hydrated (obviously)
and that is a long process when dealing with material which becomes highly
viscous in solution.  You can watch mg quantities of genomic DNA go through the
process and initially the white solid hydrates enough to become clear
(transparent) to the extent that you may be fooled into thinking the DNA has
dissolved. Because DNA binds substantial quantities of water, and won't dissolve
well until quite highly hydrated, there becomes an accessibility problem with
the water not being able to penetrate the increasingly viscous mass.  Another
way of looking at this is to compare a solution of 1 mg/ml genomic DNA to the
same concentration of RNA.  The DNA will be moderately viscous (and large DNA
is not easily dissolved above about 2-4 mg/ml), whereas the RNA will be similar
to water (and even large RNA can be dissolved to 20-40 mg/ml with a
slightly increased viscosity).  If you shear DNA down to 1 to 2 kb it loses
its high 
viscosity and presumably will be quite easily redissolved (although I haven't 
done that experiment, either).  [Supercoiled DNA is a different case because it
presents an unusual hydrodynamic profile which should have low viscosity].

It should be possible to speed up the dissolving of large DNA if it is carefully
lyophilized from solution (as a recent post suggested) rather than etol 
precipitated, particularly if the DNA powder is slowly added to the TE (or 
whatever); the finer particles are more easily hydrated than a large lump.

Now, in anticipation of the ultimate question, DNA is not salt precipitable
(at least with monovalent cations like the alkali metals) but many RNAs
(and some
fairly small) are; this is the basis for LiCl/NaCl salt fractionations.  Why is
that?  The RNAs that are salt-precipitable do have a higher proportion of 
single-stranded regions (the two large rRNAs are approximately 50% ss), and 
sRNAs (tRNAs and 5S) are biased the other way (as is ds DNA) but frankly I have
never seen an explanation for this and don't have a good one myself, nor do the
other veteran (let's say oldish) RNA people that I work with. Its tempting to 
invoke hydration/water-availability arguments, but both DNA and large RNAs are 
PEG precipitable and that has to be a result of the PEG tying up so much water
that the macromolecules, with heavily salt-shielded phosphates, fall out of
solution.  And DNA will easily precipitate from TE (i.e, low salt) if spermine,
hexamine cobalt(III) or CETAB is added and that is obviously phosphate-based.

So I end with a puzzle, one of the enigmas of experimental molecular biology.


David F. Spencer, PhD
Dept. of Biochemistry
Dalhousie University
Halifax, Nova Scotia, Canada

dspencer at is.dal.ca
dspencer at rsu.biochem.dal.ca

More information about the Methods mailing list