Separation of RNA oligos from DNA oligos.

[ksharp]

Hello again,

I'm curious - has anyone ever used HPLC to separate RNA oligos (25-mer) from their deoxy analogues?

I've used my DEAE anion-exchange column with NaCl and NaClO4 gradients, separating based on total charge (length of the oligo), but these oligos are the exact same length. Is there perhaps another gradient that could exploit the differences between these DNA/RNA analogues...pH, %organic, etc?

Thanks!

[rhaefe]

You could try RP ion pairing.
A: 0.1mol/L triethyammonium acetate (TEAA), pH=7; B: 0.1mol/L TEAA, 25% acetonitrile, pH=7
Preferable column is a Trangenomic DNASep (non porous, 2µm particle size, reversed phase and polymer based) but for oligos a regular RP column with 100Å pore size should work. A Hamilton PRP-1 column (RP, polymer based, 100Å) will work pretty good. Hamilton does have quite a few application notes for DNA related separations.

Seperation temperature should be at least 50°C or higher.

The separation will be based upon size AND base compostion. Oligos with higher AT content will elute later than oligos with high GC content (if the size is identical). Bases towards the end of the oligo will have more influence than those located in the middle.
I am pretty sure you will be able to get at least some separation. Make sure that your LC system is clean. While ologos are not as critical as dsDNA the separation can be completely destroyed by small quantities of Fe, Cr in the system. If this is the case a passivation of the system with nitric acid is recommended. I can send you a generic passivation procedure if you are interested.
--

Robert

[ksharp]

Thanks for the suggestion Robert, we are well equipped for an ion-pair method. In fact, it works great on the pure oligos, but not on the samples.

Without bogging you down with the details...in addition to the oligos my samples contain lipids which require a liquid-liquid extraction). That extraction procedure introduces small amounts (~10% by volume) of 1-butanol into the aqueous/oligo-containing layer of the sample. The small amount of butanol is sufficient to upset the equilibrium of the ion-pair method. By contrast, the ion-exchange system is unaffected by the butanol.

The quality of the separation on ion-pair was superior, and I've explored different extraction procedures to keep the organic solvents out of the aqueous layer (tried less water-miscible solvents like CHCl3, hexanes, pet ether) but with little luck - alcohols (1-butanol is the least miscible that I've tried) seem to be the only way.

[rhaefe]

If the lipids are present in only small amounts you might get by using a guard column and shooting the sample directly. The lipids will accumulate on the guard column, the oligos will pass through. You can then either discard the guard column after so many (=?) injections or try to clean it by passing 100% acetonitrile, methanol or THF through it.

If have done similar things without guard columns: once he separation on the analytical column gets unacceptable you back flush it and flush the gunk out. High temperatures help in removing any contaminants.


--
Robert

[JR Thayer]

KSharp,

I have separated RNA from DNA in ss, ds, and DNA/RNA hybrid forms on a DNAPac PA200 column.

Typically, I use pH 7, and NaCl as eluent salt. On this column a gradient of 200 - 800 mM NaCl will serve as a good survey gradient.

10% butanol should not materially affect the separation, but may cause the oligos to elute a bit earlier than without solvent.

This should be pretty simple.

Cheers,
Jim Thayer