, , ,

And so we come to the slowest part of SPPS, purifying your crude compounds. For large peptide sequences the only viable purification method seems to be HPLC [1], as the truncated N-acetyl sequences are otherwise too similar to the final product. But for short sequences and peptides with unusually good coupling efficiency you can use what I like to call the poor man’s HPLC.

For most short peptide sequences the “high performance” capability of an HPLC isn’t required, you just need the reverse phase silica. Functionalized silica is a little more expensive than standard flash grade, but if you treat it right the silica can be reused at least 50 times [2]. The stuff I use is C18-functionalized [3], and it’s packed and used just like a flash column–provided you run your flash columns in water/methanol mixtures.

Since this is reverse phase, the more polar your compound the faster it will go through, and the more polar your solvent the less it will move any given compound through the silica. If this is disconcerting you may find that wearing a fake goatee helps. I load my samples in a minimum of pure water add them directly to the column. The silica is very fine and it’s almost impossible to add solvent without disturbing the top layer, but that doesn’t seem to affect resolution. Water is polar enough that it’s almost like adding my compound to a normal column in pure hexanes.

Regardless, once the peptide is on the column I run a simple gradient mixture, moving from pure water to 50%, 75% and finally 100% methanol [4], switching every two column volumes (1 CV is roughly 45cm of solvent). For me the compound is then visualized by spotting each test tube onto a TLC plate and staining with ninhydrin, though a different approach may be required if you have no lysines or arginines. At the end of the day the bulk solvent can be removed with a lyophilizer, a rotary pump/rotovap setup or with a steady stream of nitrogen gas or air. The column can then be regenerated with three column volumes of methanol, with 1% TFA in MeOH and DCM used every five peptides or so to remove any trace non-polar impurities that may build up.

Well, that’s wraps up the solid phase synthesis series of posts.  If you’ve got any questions, or if there’s something I missed, let me know.


[1] If a HPLC is well outside of your budget range, you could try cobbling together a poor man’s MPLC, using HPLC grade silica and self-packed columns. I can’t vouch for the success rate though.

[2] Protip: The greatest damage to the column has happened when I lent it out. The first time you give it to a labmate for an afternoon, make sure you keep an eye on them.

[3] WAT020594, 125A, 55-105um.  Doesn’t seem to be available anymore, though you could give Silicycle a try.  Honestly, their stuff seems a little superior, and is quite reasonably priced.

[4] To prevent acid-base exchange it is vital that your solvents be slightly acidic. I make 1% TFA stock solutions of MeOH/H2O, diluted one part in ten prior to use. Without the TFA [5] I was having lysines deprotonate on the column, leading to highly non-polar groups that couldn’t be removed with even DCM washes. These peptides would then accept protons from the NEXT peptide that I tried to purify on the column, leading to hopelessly impure fractions.

[5] TFA was just a handy acid. Formic would work just as well, or I suppose acetic. The strong acid just made my life easier, as I didn’t have to worry about counterion exchange with the TFA salts created during cleavage from the column.