Extracellular lipid composition and their functions

To begin with a description of the roles of lipids either in extracellular vesicle (EV) formation or exosome bioactivity, it is important to mention that current knowledge on those subjects is limited. Therefore, the following review will aim to present a state of the art of current research on EV lipid composition and the function of lipids in EVs.

Lipid constitution of exosomes

EVs are nano-size particles constituted of a lipid bi-layer encapsulating a variety of biomolecules such as proteins and nucleic acids. The current understanding of EVs’ characteristics led to a consensus allowing to discriminate different groups according to their size and biogenesis mechanism (Doyle and Wang, 2019). Exosomes are the smallest extracellular vesicles, with sizes ranging from 30 to 150 nm. They are believed to be formed during the maturation of multivesicular bodies (MVBs) and released in the extracellular environment when MVBs fuse with the plasma membrane. Microvesicles are particles with diameters ranging from 100 nm to 1 µm. 

They are formed and released by outward budding of the plasma membrane. Dying cells release apoptotic bodies, particles with sizes from 50 nm up to 5 µm. They form after plasma membrane gets detached from cytoskeleton. Improvements in analysis methods of the content and composition of extracellular vesicles represents a major issue in further understanding the different biogenesis of the different groups established and their biological activity in a clinical perspective. The lipid composition of extracellular vesicles is expected to reflect their biogenesis mechanism.

Lipidomics technics, an emerging discipline in biosciences, have greatly improved with the development of mass spectrometry (Yang and Han, 2016), allowing progress in the elucidation of EV membrane lipid composition. Despite those efforts, lipidomics applied to biosciences remains a whole of very specialized methods, with an efficacy depending on sample purity, with result uncertainties up to or even above 15% (Skotland, Sandvig and Llorente, 2017). However, challenges inherent to the technics used, such as resolving power for ion-peak quantification (Saito, Ohno and Saito, 2017) or processing the data generated by mass spectrometry (Schiffman et al., 2019) are being addressed.

Due to their biogenesis, EVs can be expected to have a lipid membrane composition similar to the plasma membrane of their parent cell. The following table proposes a summary of the most common lipid classes in biological membranes (Figure 1). 

However, further analysis of the compositions of cell plasma membrane and EVs reveals lipids specifically enriched in plasma membrane or in EVs (Figure 2). In particular, cholesterol has been reported to take part in the active regulation of the movement of multivesicular bodies on microtubules (Huotari and Helenius, 2011).

Cholesterol has also been shown to participate in MVBs formation, as two other lipids which seem to play a preponderant role : phosphatidylinositol 3-phosphate [PI(3)P] and BMP/LBPA (Piper and Katzmann, 2007). Cholesterol presence in membranes may help in promoting proper membrane curvature (Wang, Yang and Huang, 2007), and is involved in the formation of domains enriched in proteins such as tetraspanins (Silvie et al., 2006). 

Regarding the role of BMP/LBPA, it seems that it’s presence is related to a better membrane fusion (Kobayashi et al., 2002) and induces spontaneous membrane fusion under acidic conditions (Matsuo et al., 2004). All this evidence illustrates the critical role of some lipids in extracellular vesicle biogenesis. 

Figure 2 : Enrichment of lipid classes in PC-3 cells or exosomes released from these cells calculated as mol% of lipids (Skotland, Sandvig and Llorente, 2017)

Role of exosome lipids in health and diseases

Furthermore, lipid composition in EVs has been linked to diseases. In pathological conditions, a greater enrichment of sphingolipids in EVs has been observed. For instance, higher sphingomyelin levels in EVs have been linked to ST-segment-elevation myocardial infarction (Burrello et al., 2020). Studying lipid composition of EVs using new artificial intelligence methods can also lead to great improvements in diagnosis. 

As an example, lipid composition of blood plasma exosomes can be linked to early or late stage non-small cell lung cancer (Fan et al., 2018). Phosphatidylserine (PS) is another lipid under investigation for cancer diagnosis, tumor cells exposing Ps on their plasma membrane. It has been showed that PS detection can be used for early diagnosis of ovarian malignancies (Lea et al., 2017). 

Thus, lipidomic studies of EVs have the potential to lead to better understanding of development mechanisms of some pathologies and may provide new diagnosis methods for early detections of diseases. 

The understanding of the implications of EV lipids in pathological conditions is an opened door for the clinical use of EVs. More and more research groups are interested in elaborating EV formulations and EV engineering methods to treat a variety of diseases. Understanding how lipids are involved in metabolic mechanism allows to take advantage of their characteristics. For instance, it seems that exosomes presenting membranes enriched in lyso phospholipids and no phosphatidylserine are able to cross the blood-brain barrier (Jakubec et al., 2020a).

The literature is very rich in studies of specific lipids of EVs and their role in EV biogenesis and bioactivity. This profusion gives hope for future developments of treatments but must be contrasted with a striking fact: all these studies rely on EV separation and purification methods. However, these methods are not perfect, there is a great need of improvement to have even more reliable results of lipid analysis. 


BMP = bismonoacyl glycerophosphate

CE = cholesteryl ester

CHOL = cholesterol

Cer = ceramide

DAG = diacylglycerol

Gb3 = globotriaosylceramide

HexCer = hexosylceramide

HG = hexadecylglycerol

LacCer = lactosylceramide

LBPA = lysobisphosphatic acid

MVB = multivesicular body

PA = phosphatidic acid

PC = phosphatidylcholine

PC O/P = PC ethers (alkyl or alkenyl)

PE = phosphatidylethanolamine

PE O/P = PE ethers (alkyl or alkenyl)

PG = phosphatidylglycerol

PLD2 = phospholipase D2

PI = phosphatidylinositol

PS = phosphatidylserine

SM = sphingomyelin

SMase = sphingomyelinase

TAG = triacylglycerol


Burrello, J. et al. (2020) ‘Sphingolipid composition of circulating extracellular vesicles after myocardial ischemia’, Scientific Reports, 10(1), p. 16182. Available at: https://doi.org/10.1038/s41598-020-73411-7.

Doyle, L. and Wang, M. (2019) ‘Overview of Extracellular Vesicles, Their Origin, Composition, Purpose, and Methods for Exosome Isolation and Analysis’, Cells, 8(7), p. 727. Available at: https://doi.org/10.3390/cells8070727.

Fan, T.W.M. et al. (2018) ‘Exosomal lipids for classifying early and late stage non-small cell lung cancer’, Analytica Chimica Acta, 1037, pp. 256–264. Available at: https://doi.org/10.1016/j.aca.2018.02.051.

Huotari, J. and Helenius, A. (2011) ‘Endosome maturation: Endosome maturation’, The EMBO Journal, 30(17), pp. 3481–3500. Available at: https://doi.org/10.1038/emboj.2011.286.

Jakubec, M. et al. (2020a) ‘Plasma-derived exosome-like vesicles are enriched in lyso-phospholipids and pass the blood-brain barrier’, PLOS ONE. Edited by J.W. Ramos, 15(9), p. e0232442. Available at: https://doi.org/10.1371/journal.pone.0232442.

Jakubec, M. et al. (2020b) ‘Plasma-derived exosome-like vesicles are enriched in lyso-phospholipids and pass the blood-brain barrier’, PLOS ONE. Edited by J.W. Ramos, 15(9), p. e0232442. Available at: https://doi.org/10.1371/journal.pone.0232442.

Kobayashi, Toshihide et al. (2002) ‘Separation and Characterization of Late Endosomal Membrane Domains’, Journal of Biological Chemistry, 277(35), pp. 32157–32164. Available at: https://doi.org/10.1074/jbc.M202838200.

Lea, J. et al. (2017) ‘Detection of phosphatidylserine-positive exosomes as a diagnostic marker for ovarian malignancies: a proof of concept study’, Oncotarget, 8(9), pp. 14395–14407. Available at: https://doi.org/10.18632/oncotarget.14795.

Matsuo, H. et al. (2004) ‘Role of LBPA and Alix in Multivesicular Liposome Formation and Endosome Organization’, Science, 303(5657), pp. 531–534. Available at: https://doi.org/10.1126/science.1092425.

Piper, R.C. and Katzmann, D.J. (2007) ‘Biogenesis and Function of Multivesicular Bodies’, Annual Review of Cell and Developmental Biology, 23(1), pp. 519–547. Available at: https://doi.org/10.1146/annurev.cellbio.23.090506.123319.

Saito, K., Ohno, Y. and Saito, Y. (2017) ‘Enrichment of resolving power improves ion-peak quantification on a lipidomics platform’, Journal of Chromatography B, 1055–1056, pp. 20–28. Available at: https://doi.org/10.1016/j.jchromb.2017.04.019.

Schiffman, C. et al. (2019) ‘Filtering procedures for untargeted LC-MS metabolomics data’, BMC Bioinformatics, 20(1), p. 334. Available at: https://doi.org/10.1186/s12859-019-2871-9.

Silvie, O. et al. (2006) ‘Cholesterol contributes to the organization of tetraspanin-enriched microdomains and to CD81-dependent infection by malaria sporozoites’, Journal of Cell Science, 119(10), pp. 1992–2002. Available at: https://doi.org/10.1242/jcs.02911.

Skotland, T. et al. (2020) ‘An emerging focus on lipids in extracellular vesicles’, Advanced Drug Delivery Reviews, 159, pp. 308–321. Available at: https://doi.org/10.1016/j.addr.2020.03.002.

Skotland, T., Sandvig, K. and Llorente, A. (2017) ‘Lipids in exosomes: Current knowledge and the way forward’, Progress in Lipid Research, 66, pp. 30–41. Available at: https://doi.org/10.1016/j.plipres.2017.03.001.

Wang, W., Yang, L. and Huang, H.W. (2007) ‘Evidence of Cholesterol Accumulated in High Curvature Regions: Implication to the Curvature Elastic Energy for Lipid Mixtures’, Biophysical Journal, 92(8), pp. 2819–2830. Available at: https://doi.org/10.1529/biophysj.106.097923.

Yang, K. and Han, X. (2016) ‘Lipidomics: Techniques, Applications, and Outcomes Related to Biomedical Sciences’, Trends in Biochemical Sciences, 41(11), pp. 954–969. Available at: https://doi.org/10.1016/j.tibs.2016.08.010.