# Purification of primary amines using Schiff base immobilization

I want to purify all primary amines containing compounds in a complex mixture (cell-extracted metabolites). I'm unaware of any published methods to do so. So I am thinking of making my own using a reversible Schiff base immobilization.

My input is a 1 mL metabolite extract that contains hundreds of compounds with a total molarity of ~1 mM. Of this ~50% contain a primary amine. Glyoxal agarose typically has 40–50 μmol aldehydes per mL, so capacity is not an issue. Impurities that I want to get rid of are, among others, organic acids like malate, pyruvate and succinate, sugars like glucose and glucose-phosphate, lipids and all non-volatile salts.

The idea is to use Schiff base chemistry:

$$\ce{R^1CHO + R^2NH2 <=>[k_1][k_2] R^1CH=NR^2 + H2O}$$

An immobilized aldehyde is used to capture all compounds containing primary amines. A Schiff base is generally stabilized under high pH and destabilized at low pH. It can be further stabilized by dehydration to shift the equilibrium towards the Schiff base side. Therefore, I would add my input at pH = 10 in 95% MeOH, wash under dehydrated conditions e.g. using 100% MeOH, and elude at pH = 2 using 0.1 M aqueous formic acid. Alternatively, adding glycine in the elution buffer to compete off remaining amines.

My questions are:

1. Are there alternative ways to purify primary amines from a complex mixture which I'm unaware of?

2. Did anyone publish a paper that used Schiff base chemistry to purify primary amines?

3. What would be your odds that this could work? Is it worth pursuing?

4. Any ideas for improvements? Eluate has to be run on mass-spectrometer, so my requirements are:
a. no non-volatile salts;
b. no solvents with MW above 150 Da.

• Might be easier for a first pass to pass the solution through an SCX cartridge which will retain all basic species then elute them off with ammonia soln. Apr 22 '20 at 20:11
• Thank you @Waylander. I have been thinking along those lines too but it's great to be reminded. My two biggest reservations are: 1) many metabolites are charged but not containing the primary amines that I am looking to purify (e.g. a slew of phosphatidylcholines), and 2) very low pH would be required to push all primary amine containing metabolites into positive charge (e.g. glutathione with pK1 = 2.12, and probably even lower for S-Succinylglutathione). Apr 26 '20 at 15:30
• I think it would be better to reduce this post to 1 and 2. Then hope for an answer and ask a followup question. I think it currently is too open-ended, and especially point 3 invites for too much speculation and is primarily opinion based. Apr 30 '20 at 15:46

Let's take your questions one at a time.

1. Are there any alternative ways to purify primary amines from a complex mixture that I am unaware of?

Your end application is LC-MS. There are plenty of papers on amine-specific derivatization methods for metabolome analysis. For example, there is one by Zhao et al. [1]. Instead of purifying the primary amine metabolites and isolating the purified mixture as a separate step, they just rely on the analytical instrumentation to be able to identify primary amines from their derivatization pattern.

Alternately, as suggested in the comments by @Waylander, you could acidify the sample to form ammonium ions, and then pass the sample through a cation exchange column, where only positively charged ions will bind. As you noted, this will not be perfectly selective for primary amines, because species like quaternary ammonium salts etc. will also be retained.

You could combine the derivatization approach with the SCX approach, of course. If you chose a primary-amine-selective derivatization, you might get a selective method. At some point, however, the complexity of the method will become a drawback.

1. Have anybody else, published or unpublished, used Schiff base chemistry to purify primary amines?

Well this question is impossible to answer for the unpublished case. I couldn't find any published methods in a quick search, which makes me suspect it is not the best approach to your problem. But I could be wrong of course!

1. What would be your odds that this could work? Is it worth pursuing?

Unanswerable without data. I suspect that secondary amines, at least some times, will also react with the aldehydes to form Schiff-base-like iminium cations, and may also be selected by your proposed method.

1. Any ideas for improvements?

I think derivatization, without separation or isolation of the amines, would be an attractive direction for you to consider. The paper I linked, its references, and the papers that reference it might be good starting places.

### Reference

1. Zhao, S.; Li, H.; Han, W.; Chan, W.; Li, L. Metabolomic Coverage of Chemical-Group-Submetabolome Analysis: Group Classification and Four-Channel Chemical Isotope Labeling LC-MS. Anal. Chem. 2019, 91 (18), 12108–12115. DOI: 10.1021/acs.analchem.9b03431.