amino acids

[upamaniah]

hi,
amino acid by hplc is always is a problem.
i have l-valine, in which i have to quantify leucine and isoleucine content at 0.1% level each.
any suggestion of using a right column and mobilephase without derivatisation technique.
thanks in advance.
vijay

[Rafael Chust]

Dear Vijay:

I think AA analysis without derivatisation or other complicated sample prep is a myth!

Check Phenomenex site at www.phenomenex.com and look their EZ:faast kits. They proved to be very easy to use, quite sensitive and price effective.

[SIELC_Tech]

Check our applications for amino acids: no derivatization and no ion-pairing reagents. LC/MS/ELSD/UV compatible methods.
http://www.sielc.com/compound_184.html (leucine and isoleucine)
http://www.sielc.com/compound_002.html (other amino acids)

You can use phosphoric or sulfuric acid for UV (corresponsding buffers too)
ammonium formate/acetate/TFA for ELSD/LC/MS detection.

P.S. Rafael,
couple people from this board already are using our columns for AA (even for prep and semiprep applications). It si the easiest approach-simple mobile phase and simple method development

[Srinivas SK]

Hello All

Thought I'd throw in my two-bit's worth of advice... having done a lot of HPLC and GC-MS of amino acids in serum. Serum is perhaps the most difficult matrix to tackle. Really raises your blood pressure !

So, an in-vitro sample of L-valine containing 0.1% of other amino acids would be a welcome change.

Rafael is quite right - you can't escape derivatisation so easily. The alternative to derivatisation, would be to use an Ion-ex column and electrochemical detection. Not an easy technique, but a good one, if done right.

Else, you derivatise. Take your choice - precolumn or post-column derivatisation. There are vehement proponents for both techniques.

I'd prefer a precolumn method any day, but that's just my choice.

You have a choice of using a ready-made reagent kit from a good vendor or making your own reagents, as I did. Depends on your skill and your budget.

To give you an idea - I developed a validated HPLC method for amino acids in serum, using pre-column OPA derivatisation, done on-line using an autosampler, a phosphate buffer gradient, fluoresence detection and a C18 Phenomenex column. And a complex sample prep protocol, using a home-made SPE device.

Hope this helps.

Warm rgds,

[Uwe Neue]

since several ones of my colleagues are advertising, I thought that I might do the same thing...

Waters offers several amino acid derivatization kits with tools for doing the separation. For example, the ACCQTag method is a derivatization method that works in seconds and has a high sensitivity, both with UV and fluorescence. Check on our webside!

[Rafael Chust]

Although my original idea was not advertising, I must admit my sympathy for Phenomenex solution.

I have been involved in HPLC for a while (this month I commemorate 20 years of my post-graduation specialization in HPLC) and I have settled several AA analysis methods since.

The method described by Srinivas is quite similar to Waters AutoOPA solution. It works quite well, but requires a very adavanced autosampler to sample and mix stuff from 3 different autosampler vials (sample, reagent and IS). Also requires fluorescence detection.

I am not familiar with AccqTag but used PicoTag for some time. You can make all sample prep outside the system and PITC AA can be analysed with UV, which makes everything easier. The problem is that PITC AA are not very stable and you must prepare your sample a couple of hours before injection - you can not store it for long.

I have also worked with post-column derivatisation - it was such a mess that my selective memory even does not remember the derivatisation agent and I would not recomend this stuff (unless if I do hate my colleague )!

Why I liked the Phenomenex solution? Because all other methods took between 30 to 90 minutes run, while EZ:faast falls under 10 minutes run - both for GC-FID, GC-MS and LC-MS. The major drawback is that you need a MS detector to do it by HPLC.

Cost-wise, Phenomenex and Waters kits are quite similarly priced, although OPA would be cheaper if you prepare all the stuff yourself!

[Kostas Petritis]

Well, I am strongly opinionated towards amino acid analysis methods that do not use any derivatization (it is not a myth Rafael!) so I will write my personal view in detail (so I can in the future refer to this message for similar debates...). In short, personally I do not see why someone would still use derivatization at all for this class of molecules.

Before I go on, for the original poster, if your matrix is not complicated you can use a validated LC-ELSD method, that separates these 3 amino acids and several other at the same time in less than 20 min. If you are interested only in these 3 amino acids you can further increase the speed of the analysis.
Look at Petritis et al: Chromatographia 60 (2004) 293-298.

Now, generaly speaking the problem of analyzing amino acids in their underivatized form was a) their high polarity which made it difficult to separate, b) lack of chromophore that made it difficult to detect, c) complex matrix and detection limits.

There are several solutions developed, I will only mention my contributions to the subject.

1) For the detection part have a look at:

J. Chromatogr. A 961 (2002) 9-21
J. Sep. Sci. 25 (2002) 593-600
LC-GC Europe 14 (2001) 389-395

...where I compare all the possible detectors that you can use for underivatized amino acid analysis by using the same LC separation. The first paper compares, RI, CLND, ELSD, NMR, MS, MS/MS, CD, UV. The only detector ommited was the amperometric from Dionex as at that time we didn't have it in the laboratory and Dionex was not willing to give it to us for evaluation.

2) For the separation part:

In addition to SIELC-Tech contribution, there are several papers on the chiral or achiral amino acid separation. Some of the methods separate completely the 20 proteinogenic amino acids (something that is not a requirement when you are using mass spectrometry).

Have a look at:
J. Chromatogr. A 833 (1999), 147-155.
J. Chromatogr. A 855 (1999) 191-202
J. Chromatogr. A 870 (2000) 245-254
J. Chromatogr. A 913 (2001) 331-340

3) The presence of matrix and detection limits:

Very complex matrices can be a challenge but this is not an issue if you have mass spectrometry. Actually, you can find in the literature examples where you can analyze relatively complex mixtures like fermentation matrices with amperometric detectors etc. Now with mass spectrometry you can do virtually every matrix and pretty fast as a bonus.

For MS/MS optimization for amino acids analysis you can look at:
J. Chromatogr. A 896 (2000) 253-263.
Rapid Commun. Mass Spectrom. 17 (2003) 1297-1311

For analysis in phytochemical matrices you can look at:

J. Chromatogr. A 896 (2000) 335-341
Phytochem. Anal. 14 (2003) 347-351

For analysis in agricultural/food matrices you can look at:

J. Agr. Food Chem. 52 (2004) 4318-4322

For analysis of biological fluids (i.e. urine and plasma) you can look at:

Rapid Commun. Mass Spectrom. 19 (2005) 1587-1602.
Rapid Commun. Mass Spectrom. 19 (2005) 3287-3297.

Especially these last two can simultaneously do up to 76 amino acids in a single run at less than 20 min. Actually you can do even more if needed... Limits of detection are pretty low as well...

As a result, I personally think that any amino acid analysis can be done without derivatization. For very complex matrices you need mass spectrometry as the detector...

Hope the above helps,

Kostas

[domenger]

Hi,

We have a very nice application on our Hypercarb column ( porous graphitic carbon) concerning these compounds: 20 AA separated without any derivatization. You need to have a ELSD detector... and a hypercarb column.
Do you want to see the application?
Thierry

[Kostas Petritis]

Sure!

PS: Although I have a suspicion...

[SIELC_Tech]

I would like to see this too.

Kostas,

Have you ever tried Hypercarb? Is there a problem with mismatch between solubility, surface property, etc.?

regards,

Vlad

[Kostas Petritis]

Vlad,

Yes I have tried Hypercarb and it behaves like a reverse phase packing, no special precautions to take. The only thing I can think of is that the PGC is much easier to modify permanetetly due to high affinities for flat compounds and some other molecules... The solvent of choice for it's regeneration... THF...

I will let Thierry respond and I'll might elaborate more...