HILIC Help Analyte not showing up.

[ChemJay]

So I am doing some research into Glucoraphanin. I was trying out Phenomenex's Application to see how well it would work out for me. http://phx.phenomenex.com/lib/TN14810713_W.pdf I used the exact same parameters as in Figure 2. of the paper with exceptions, I am using a polyhydroxyethyl A column (100 X 4.6mm 3um 100-A). However If a peak does show up it usually comes in around 8-9 minutes retention time and looks like a half circle.... (peak starts at ~8mins and ends at about ~11 mins) Also I am not sure of the condition of the column as it was bought by a predocessor of mine, but I am getting sharp peaks for other Analytes and the column does seem to be functioning well. I'm not opposed to getting a new column I just want to exhaust all other options before I spend that kind of money. Any help is greatly appreciated.

[Andy Alpert]

I hope that that's not the same column that John Troyer et al. used for their paper from 2001. If so, then it would be 16 years old. That said, if peaks for other compounds look good, then it's possible that the problem lies with your sample. What solvent is the glucoraphanin dissolved in? Also, what's your source of your glucoraphanin standard? It's possible that the standard actually consists of several components that coeluted during purification but which are being partially resolved by HILIC and so eluting in a poorly-separated envelope.

The Phenomenex mobile phases are pretty good. The main difference from Troyer's 2001 paper is their use of 15 mM ammonium formate instead of 30 mM.

Feel free to contact me off-list.

Andy Alpert
PolyLC Inc.
(410) 992-5400
aalpert@polylc.com

[ChemJay]

I appreciate the info. It got me thinking about the sample prep. I did obtain a pure sample of glucoraphanin (96-97% pure) but I think my error was the solvent I used. I used 70% ACN 30% water, which I believe the water denatured the GR. I think next time I will make a stock solution using pure DMSO for storage and then make my samples from that.

[Andy Alpert]

To denature something usually means to cause it to lose secondary or tertiary structure that's necessary for its properties. A molecule as small as glucoraphanin doesn't have any elaborate secondary or tertiary structure to lose. Maybe you mean that it precipitated from the solvent you describe.

The best solvent to use would be one that matches the HPLC mobile phase as closely as possible. For that matter, why not use the mobile phase itself? Let me point out that glucosinolates are electrolytes because of the sulfate group that they all have. In order to get a sharp, symmetrical peak in chromatography, all of the molecules of a single substance should have the same counterion. Otherwise, different ion pairs will differ in polarity and will elute at different times in HILIC. Result: tailing or even multiple peaks for a single compound (cf. your dome-shaped peak for glucoraphanin). If you have ammonium acetate or formate in the mobile phase, then have it in the sample solvent too.

While we're on the subject: What is the counterion in the glucoraphanin sample that you received? I'd ask the supplier.

[ChemJay]

Potassium is the counter ion.

http://www.enzolifesciences.com/ALX-350 ... sium-salt/

I'm hesitant to use mobile phase directly for the stock solution because it does contain water and they warn that GR is very hygroscopic. That's why I was thinking of putting the stock solution in DMSO because it is partially soluable (1mg/ml). Then when I am ready to test put say 100ul of stock in the sample vial and like 900ul of mobile phase.

[Andy Alpert]

I see. Well, in order to get a single sharp peak for glucoraphanin here, it will have to exchange its potassium counterion for ammonium immediately at the start of the chromatography. The best way to make that happen would be to put a lot of ammonium ion there, both in the mobile phase and in the sample solvent. Try doubling or tripling the ammonium acetate/formate concentration and let's see how that affects your peak shape.

By "hygroscopic", the manufacturer merely means that the solid material will readily adsorb water from the air, lowering its titer. That's no reason not to dissolve the material in a mobile phase that contains some water. It will inevitably come into contact with some anyway during the chromatography.

[James_Ball]

Water will not harm glucoraphanin, I use it as my solvent when making standards all the time. Methanol is also good.

We used the Phenomenex HILIC method for one client and found it gave the same results as our C18 method, but it used too much Acetonitrile so we now only use RP C18 for glucoraphanin analysis. The difference is that with HILIC glucoraphanin elutes before sulforaphane (which is the R group without the glucose) and with C18 the order is reversed.

How did you prepare your samples? If you are extracting from plant materials, you can't extract with water alone. You have to use organics to denature the enzyme that breaks the glucoraphanin down to sulforaphane or you will have very rapid conversion during extraction.

I have been doing this analysis for the last ten years and it took us about a year to get a method that worked reliably and accurately. Extraction was the hard part, analysis by C18 on UV or MS was the easy part.

[ChemJay]

I was kind of following a paper I found for isolating Glucoraphanin from broccoli seed. I boil the broccoli seed (somewhere in the temperature range of 160°F-170°F). I assume this is to kill off the myrosinase enzyme. Then I dry the seed and crush it, mix it with my solvent of choice (which I have tried different methods here, TNTC) Centrifuge the sample and filter it into a HPLC sample vial. I've been thinking to get a more pure sample, I should run it through a sep funnel and remove the organic layer and just sample the aqueous layer.

[ChemJay]

I see. Well, in order to get a single sharp peak for glucoraphanin here, it will have to exchange its potassium counterion for ammonium immediately at the start of the chromatography. The best way to make that happen would be to put a lot of ammonium ion there, both in the mobile phase and in the sample solvent. Try doubling or tripling the ammonium acetate/formate concentration and let's see how that affects your peak shape.

By "hygroscopic", the manufacturer merely means that the solid material will readily adsorb water from the air, lowering its titer. That's no reason not to dissolve the material in a mobile phase that contains some water. It will inevitably come into contact with some anyway during the chromatography.

Ok I'll give this a try, I'll bump up the ammonium to something like 60mM of ammonium in both sample and mobile phase and see where that gets me. Thanks once again for all the help here guys this is really helpful stuff!

[ChemJay]

Alright, I've gone through and made sure that everything is as follows for HILIC. Solvent A is at 70% and is 85:15 ACN/ H2O w/ 200mM Ammonium Formate (ph-7). Solvent B is 30% ACN. I'm also running the same parameters as in the paper above. Now I've tried boiling the seed, then crushing and boiling a second time. I sent a portion of a sample to a third party lab (who uses this exact extraction/HPLC method for GR concentration.) They sent me back a chromatogram with a clear and nice GR peak at about the 10min mark. I however do not have any peaks after about 3 mins. (They also had a few peaks between 0-3mins, I assume some impurities in the sample.) Just looking for some new routes to try, I've exhausted all the one's I know of.

[lmh]

If you really get stuck, there is a nice alternative method for glucosinolates in which you bind them to an ion-exchange column (think Dowex in a pipette-tip), wash off the neutrals, add a bit of myrosinase to the top of the column, which cleaves the sulphate, and then wash off the neutrals again (which are now basically the glucosinolates). The desulphated glucosinolates are very easy to analyse by conventional reverse phase on any C18 column. It's more work preparing the samples, but the chromatography becomes very reliable.