The Specificity of Lipid Molecular Species Characterization

Posted on April 12, 2019

Have you ever wondered about claims there are perhaps 150,000 or more different lipid molecular species? Then, when you look in text books, our catalog, or even at your own data, do you get the sense there are only hundreds of species? This was an important step in my own study of lipids: to recognize the sheer diversity of lipids present in the biome and the analytical challenges we face as a consequence of that diversity. Plants, mammals and cells all have unique lipids, and bacteria are just down right goofy.

I thought it would be interesting to demonstrate just how complex the lipidome is by examining an ostensibly simple lipid—PE 36:1—and expanding on the number of lipid molecular species the sum composition potentially represents. It should be noted that identifying and quantitating lipids via sum composition is very common and is the data output from ‘Shotgun’ lipidomics and many untargeted strategies using HRAM.

The structural specifics of a lipid are shown in the figure below. Lipids can be described at multiple levels of increasing specificity, starting from the lipid class all the way to a chemically defined structure that includes fatty acid identification, fatty acid position, the number(s) and position(s) of double bonds and their respective stereochemistry. Multiple technologies are required to completely characterize endogenous lipids, so many of us have to settle at some level of specificity. At Avanti Analytical Services, we have the technology to identify lipids down to the fatty acid position, but generally we report our results at the fatty acid ID level.


Using this model, I determined the potential lipid molecular species for PE 36:1 at each level of specificity. Note, I did not have a particular matrix in mind when I did this—I just focused on literature-relevant fatty acids—so a particular matrix would likely be simpler. But, I was very surprised to see that the sum composition naming convention, which is widely used in the literature, is holding back a lot of potential secrets… Based on the assumptions I made, PE 36:1 could represent as many as 56 individual lipid molecular species. If you are doing biomarker research, this example makes it clear you must persist to fully characterize the structural identity of your target.