[Babcock_Hall]

We are making compounds with a pyridinium nitrogen, an alpha-amino nitrogen and two carboxylate groups (compound 8 in the attached document).   We have tried to purify similar compounds on Dowex 1 (which has a quaternary ammonium group), using HCl to elute, which protonates the carboxylate group to a carboxylic acid.  I am not certain, but I don't think that our recovery has been good.  I am considering switching to cellulose- or to dextran-based matrices because some aromatic compounds are retained preferentially when polystyrene resins are used, according to the manufacturer's literature.  We tried purifying tryptophan (Trp) using Dowex 1 to practice our technique, and our recovery was poor.  We have DE-52 (diethylaminoethyl) in the lab but not QE-52 (which is a quaternary ammonium exchanger).  I would like to avoid cation exchangers, because the pyridinium nitrogen does not have a proton to lose; therefore, our usual method of adding ammonia to deprotonate the ammonium group of most amino acids will not work in this case.  For the time being, I would like to avoid triethylammonium bicarbonate as a volatile salt eluent, because undergraduates might have a tough time working with it (although it remains an option).

We have seen a small amount of leftover phenol after extracting most of it away with ethyl acetate, and that is one of the compounds I want to get rid of.  Our product should carry two positive charges at this point, and I don't expect it to be very soluble in the organic phase.  I am not sure whether or not there will be other compounds present besides phenol.  What sort of purification schemes can we accomplish with DE-52?  I might start with this gel in the acetate form.  I think that we might elute the compound of interest 8 with HCl, because a carboxylate will no longer bind once it becomes protonated to a carboxylic acid.  However, I could also envision eluting compound 8 with ammonia, which might deprotonate the DE-52.  Any thoughts or ideas would be greatly appreciated.

[discodermolide]

What's the counter ion to the pyridinium nitrogen in compound 8? I assume the amino acid end is zwitterionic.

[OC pro]

Had purified pyridine carboxylates and also pyridine substituted amino acids in my former company. Was pain in the ass. However, cation exchange (Amberlite) worked always (conditioning with ethanol, water, 1M HCl, water). I eluted with aqueous ammonia or trimethylamine. After lyophyllization you will have a clean product. But be aware of the fact that you can loose a significant amount of product (in one case ~50% sticked onto the column).
You will end up with inner salts of course so don´t expect something crystalline. I often observed like sticky oils.

[Babcock_Hall]

Discodermolide, In the pentultimate step the counterion to the pyridinium nitrogen has to be iodide, from the leaving group.  Once the trifluoroacetic acid (TFA) and phenol are added, I would say that the alpha-amino nitrogen has trifluoroacetate as the counterion.  I am also guessing that both carboxylic acids are protonated by TFA.  Once the TFA is removed by rotary evaporation, I am less sure about what the protonation state is.  When we have done similar purifications in the past, we have prepared Dowex-1 in either the hydroxide form or the acetate form.  My assumption has been that our product will lose protons and bind.

OC Pro, My understanding is that when people purify derivatives of ADP and ATP on ion-exchange, they avoid Dowex (which I think is pretty similar to Amberlite) for the reason that you indicate: low recovery due to hydrophobic interactions.  That is why I was considering switching to carbohydrate-based ion exchangers.  Might be more trouble than it is worth, however.  Is any of your work from your former company published?  I would like to read it if it is.

[OC pro]

Unfortunately nothing is published since we had worked in contract for big pharmaceutical companies. They took also all the electronic labjournals away. So I am still not allowed to go into details.
Maybe you can use resins based on Polystyrene. I am doing a lot of SPE cartridge stuff now and the PS cartridges with ion exchange materials show the lowest binding overall.

[Babcock_Hall]

I have heard that some people use alcohol (methanol or ethanol?) to reduce the hydrophobic interactions with Dowex.  Does anyone here have any experience with this?

[angryman]

I hope this isn't too simple, but Reverse phase? or I'd try SP-sephadex, depending on solubility
let us know if you improve the process

[Babcock_Hall]

Someone else also suggested reverse phase, but with what stationary phase?

[Arkcon]

The simplest reverse phase media to get is C18 derivitized silica.  Since your product is aromatic, you might get good use out of phenyl derivitized silica, but in that case, even better is to use polymeric reverse phase -- porous divinylbenzene-styrene.  But if you're going this far, you might want to look into prep-scale HPLC or a low pressure LC system, such as the ones made by Pharmacia.