I'd separate them preparatively and determine positions in individual fractions by NMR. One can not beat LC-NMR

Thanks Alex. Would you know a lab able to test a few glyceride samples for us to measure omega-3 on sn-2 compared to total omega-3?

Hello Hiskias,

Give us a ring at 443-791-7162. Let's talk!

I think this problem does not worth a damn, since the enzymatic hydrolysis of purified triacylglycerols is a reliable and accurate method to determine the fatty acids composition of sn-2 position of glycerol backbone. During the hydrolysis, TAGs will degrade with the yield of free fatty acids, sn-1,2(2,3)-diacylglycerols and sn-2-monoacylglycerols. All you have to do later, is to separate this mixture of acylglycerols and FFAs via preparative TLC plate preimpregnated with boric acid (to avoid the acyl migration during separation) and collect area containing sn-2-monoacylglycerols, convert them into fatty acids methyl esters and analyse the mixture via GC-MS or GC-FID equiped with DB-23 or HP-88 capillary column (even DB-5 will work ok). After, you can easily calculate the amount of omega-3-fatty acids in the sn-2-position.

I may assure that stereospecifity of pancreatic lipase does not matter at all for your purpose.

The full protocol mentioned here you may find in this paper:

Yoshida H, Tomiyama Y, Yoshida N, Kyoko Shibata K and Yoshiyuki Mizushina Y (2010). Regiospecific Profiles of Fatty Acids in Triacylglycerols and Phospholipids from Adzuki Beans (Vigna angularis). Nutrients, 2(1): 49–59

doi:10.3390/nu2010049

As for regiospecific analysis via Grignard's deacylation, I won't recommend this method for several reasons:

1) It needs to be a well proven procedure of deacylation followed the extraction of its products i.e. fatty alcohols, sn-1,2(2,3)(1,3)-diacylglycerols, sn-1(3)-monoacylglycerols, sn-2-monoacylglycerols, free fatty acids and so on;

2) This complex mixture is quite hard to separate in the preparative scale for further analysis of certain component;

3) Acyl migration may occur during deacylation;

4) For each experiment ethylmagnesiumbromide should be as fresh as possible. Ideally, it should be prepared prior to deacylation;

5) And last one, the final result of all derivatization procedures may puzzle you :) I spent three weeks in hard work, trying to understand why all saturated fatty acids were in the sn-2 position (I tried adopt the method working with freshly squeezed sunflower oil). I tried three methods of EtMgBr synthesis, distilled all the solvents, tried different mobile phases to separate sn-2-acylglycerols and did not get digestible result, whereas pancreatic lipase gave excellent result, which corresponds to published data, from the first time.

Now I do not think how to analyse fatty acids composition of sn-2-position in triacylglycerol and I do not try to reinvent a bicycle :)

Dear Alex,

preparative HPLC combined with NMR of course is a powerful method, but in connection of some properties of mixtures of triacylglycerols, this approach could be very laborious. As I know, unsaturated fatty acids in the sn-2 position of TAGs give good signals on 13C and 1H NMR spectra, so they content could be calculated from this data. For detailed information I would reccomend to find papers by Dr. Giovanna Vlahov and also some useful tips could be find here:

http://lipidlibrary.aocs.org/nmr/nmrregio/index.htm

I'd just want to say, that this problem could be solved by NMR without preliminary separation of complex mixture of position-species of TAGs, but, based on my humble experience in that area, the method of enzymatic hidrolysis is cheaper, easier and more accurate.

Thanks for the input, Roman.

You are unfortunately correct: Grignard's is a nightmare in microscale and yields are not consistent.

This looks like a good project for BS honors/MS thesis. To wit, compare at least three approaches: Enzymatic, Grignard's, NMR and some combination of the above.

We have been using a pancreatic lipase method in our most recent work and have been quite happy with it. We are investigating if the sn-2 fatty acid is more affected by our feeding treatments than the sn-1 and sn-3 fatty acids are.

Pancreatic lipase method is actually reliable when analyzing sn-2 position from triacylglycerols from plant origin, containing mostly C18-polyunsaturated fatty acids. For triacylglycerols from animal origin containing arachidonic and highly polyunsaturated fatty acids (i.e. fish), enzymatic hydrolisis can be not such a good option, being advisable 13C-NMR for that purpose.  

Dear Dr. Rincon-Cervera,

Just yesterday I have read your article "Positional distribution assessment of essential fatty acids in several fats and oils including plant, fish, and microbial sources

and subcutaneous fat of Galician horse"  I am impressed by your capabilities and found out that you actually did some measurements on one of our oil "Marinol". I have sent your co-author Dr. Guil-Guerrero some questions about that and I am looking forward to get in contact with your group.

With kind regards,

Hiskias Keizer