VLC-PUFAs, the very long chain fatty acids the retina concentrates in its photoreceptors, are the longest members of the omega-3 family. Nwagbo and Bernstein classify them as fatty acids with greater than 24 carbons on the longest continuous carbon chain [1]. ALA has 18 carbons, EPA 20, DHA 22. In normal systems these longest ones are not eaten. They are made. The retina builds them in place using an enzyme called ELOVL4, and they are found in a short list of highly specialized tissues [1].
Key takeaways
- VLC-PUFAs are the longest members of the omega-3 family, classified by Nwagbo and Bernstein as fatty acids with greater than 24 carbons on the longest continuous carbon chain.
- The retina builds them in place with the enzyme ELOVL4 rather than absorbing them ready-made from food, which is why a concentrate is a recent development rather than an obvious one.
- They make up less than 2 percent of total fatty acids in the whole retina, yet the C28 to C36 species reach 10 mol percent of the phosphatidylcholine in photoreceptor outer segments. Despite being a small percentage, these lipids pack a punch, because they are bioactive.
- ELOVL2 slows with age, which makes it the bottleneck, two enzymes run the pathway and only the first one. ELOVL4 can only work on what ELOVL2 hands it.
- The strongest tissue evidence puts polyunsaturated production in retina and testis, with brain, skin and meibomian gland making mainly saturated very long chains. Why the same enzyme makes different products in different tissues is stated in the reviews as still unknown.
What are the types of omega-3 fatty acids?
Omega-3s are sorted by the length of their carbon chain. ALA (18:3n-3) is the plant one, in flax and walnuts and canola. EPA (20:5n-3) and DHA (22:6n-3) are the long chain omega-3s, the two that fish oil is built around. In 22:6n-3, the first number counts carbons, the second counts double bonds, and the n-3 names the family.
Consumer articles about omega-3s stop there. The research literature does not. Agbaga and colleagues describe the long chain PUFAs of 20 to 22 carbons, including DHA and arachidonic acid, as highly enriched in vertebrate retina, where they are elongated into very long chain PUFAs of 28 carbons or more by the ELOVL4 enzyme [4].
That extra step is what this article is about.
What is a VLC-PUFA?
A VLC-PUFA is a very long chain polyunsaturated fatty acid. Same chemical family as EPA and DHA, same kinked polyunsaturated shape, a much longer carbon chain. Berdeaux and Acar describe retinal VLC-PUFAs from the n-3 and n-6 series as having 24 to 36 carbon atoms and four to six double bonds [3]. Named species in the retina include C32:6n-3, C32:5n-6 and C34:5n-3 [1]. VLC-PUFAs with up to 13 double bonds have been identified in vertebrate tissue [1].
On the carbon count itself, the literature uses two thresholds. The broad chemical definition is greater than 24 carbons [1], and much of the retinal literature uses 28 or more, because that is where the species that actually accumulate in photoreceptors live [2][4]. Both are attributed here, and nothing in this article turns on which one you prefer.
Where do VLC-PUFAs sit in the omega-3 family?
They sit at the end of an assembly line, two enzymes run it, and only one of them is the bottleneck.
ELOVL2 is a fatty acid elongation enzyme. Deak and colleagues describe it as elongating PUFAs with 20 to 22 carbon chains [2]. Gao and colleagues describe the ELOVL2 gene as encoding a transmembrane protein that produces precursors to DHA and VLC-PUFAs [6].
ELOVL4 finishes the job. Nwagbo and Bernstein describe it as a 3-keto acyl-CoA synthase that catalyzes the rate-limiting condensation step of very long chain fatty acid synthesis from shorter chain precursors [1]. Rate-limiting means it is the slow step that sets the pace for everything downstream. Deak and colleagues report that ELOVL4 elongates long chain PUFAs and long chain saturated fatty acids of 24 carbon length into very long chain products of 26 carbons or more, and can extend them as far as 38 carbons [2].
Deak and colleagues also report that EPA is preferred as a substrate for elongation to VLC-PUFA over arachidonic acid and DHA [2]. The starting material for the longest omega-3s is not the omega-3 that gets the most attention.
Which enzyme matters more depends on which one runs short. ELOVL2 is the gatekeeper, because ELOVL4 can only work on what ELOVL2 hands it, and in aged mouse retina the two do not age together. ELOVL4 and ELOVL5 expression held steady between 5 months and 26 months while ELOVL2 expression fell [6]. The first half of the line slows down and the second half is left waiting.
Why does the retina concentrate VLC-PUFAs in its photoreceptors?
Why does the retina concentrate VLC-PUFAs in its photoreceptors?
VLC-PUFAs make up less than 2 percent of total fatty acids in the retina, Nwagbo and Bernstein report [1]. Berdeaux and Acar report that the C28 to C36 VLC-PUFAs represent 10 mol percent of total fatty acids in the phosphatidylcholine of bovine photoreceptor outer segments [3]. Those two figures are not in conflict. The first is measured against the whole retina, the second against one lipid class in one compartment inside it. Quote either without its denominator and you are off by about five-fold.
Despite being a small percentage, these lipids pack a punch, because they are bioactive. A tissue does not spend energy enriching a molecule it has no use for. When the busiest membranes in the eye go out of their way to collect something this rare, that is biology telling you where the molecule matters.
Berdeaux and Acar report that the majority of the phosphatidylcholine species containing VLC-PUFA are localized in photoreceptor outer segments, where the phototransduction reactions take place [3]. Deak and colleagues report that ELOVL4 is expressed exclusively in photoreceptors within the retina and produces VLC-PUFA products that are incorporated into phosphatidylcholine and enriched in the light sensitive membrane disks of the outer segments [2]. A rare molecule, made by an enzyme the retina expresses only in its photoreceptors, delivered into the membranes that catch light.
What it does there is hedged in the primary literature, and it stays hedged here. Nwagbo and Bernstein write that the unique hybrid structure of VLC-PUFAs has led researchers to believe they maintain the curvature of photoreceptor disks by their unique ability to occupy both leaflets of a phospholipid bilayer [1]. Ordinary fatty acids sit in one half of a membrane. These reach across both. The same review reports that the C32:6n-3 VLC-PUFA enriched in rod outer segments may promote lipid translocation across photoreceptor outer segment bilayers, and that VLC-PUFAs and ELOVL4 have been associated with rhodopsin, the main G protein coupled receptor involved in phototransduction [1].
Believe. May. Associated with. Those are the words the sources use, and they are the words I use. What is not hedged is the descriptive biology underneath, which is that these molecules exist, the retina builds them, ELOVL4 is the enzyme, and they concentrate in the membranes where light is caught.
Do rods and cones carry the same lipids?
Apparently not. Agbaga and colleagues found that whole retinas and outer segment membranes in rod-dominant animals had higher amounts of long chain PUFAs and very long chain PUFAs than cone-dominant animals, with cone-dominant animals showing about 2-fold lower levels of the di-DHA (22:6/22:6) molecular species [4]. Their reading, in their own words, is that because PUFAs are necessary for optimal G protein coupled receptor signaling in rods, these findings suggest that cones may not have the same lipid requirements as rods [4]. That comparison is across animal species.
Rods are the low-light cells. If the lipid demand really is concentrated there, that is a reasonable place to look when someone asks what changes in a healthy aging eye that makes low-light vision harder.
Which tissues make very long chain fatty acids, and which kind?
This is where a lot of confusion comes from, because the two enzymes are usually discussed as if they behaved the same way. They do not.
ELOVL2 is the broad one. Chen and colleagues describe Elovl2 expression as detectable in many tissues, with the highest levels observed in liver, testis, and the central nervous system including retina [7]. So ELOVL2 is not an eye gene. It is a general fatty acid elongation gene that happens to matter a great deal in the eye, and that distinction matters for anyone reading about it as an aging marker in blood.
ELOVL4 is the narrow one, and it is the enzyme that reaches very long chain length. Agbaga and colleagues, reviewing a decade of ELOVL4 work, describe the protein as expressed in retina, brain, meibomian glands, skin, testes and sperm [9].
The strongest evidence, stated as such. Retina and testis are where the polyunsaturated versions are made. Brain, skin and meibomian gland make mainly very long chain saturated fatty acids, incorporated into sphingolipids in brain, into the skin permeability barrier, and into omega-O-acylceramides in the meibomian gland [2][9]. Saturated is not polyunsaturated. Same enzyme, same extreme chain length, different chemistry and a different job.
Where it is less settled. Why the same enzyme makes polyunsaturated products in one tissue and saturated ones in another is stated plainly in the reviews as not yet understood [9]. Anyone who tells you the tissue split is a solved problem is ahead of the literature.
One more thing worth saying, because it is easy to assume otherwise. The reproductive side of this biology, as published, is male. Testis and sperm appear across the tissue lists, and a testis-specific Elovl4 knockout in mice reduced sperm count and motility and disrupted normal spermatogenesis [10]. I have not found a source placing ELOVL4-driven VLC-PUFA synthesis in the ovary, so this article does not claim one. If that evidence exists and I have missed it, I would rather be corrected than keep repeating the narrower version.
Where do you get VLC-PUFAs in food?
In normal systems you mostly do not, and until recently there was nothing to buy. They are built inside the tissues that use them, out of shorter fatty acids, by ELOVL2 and then ELOVL4 [1][2][6]. Fish do concentrate them, in eye, brain and gonad, which is how a fish-oil-derived concentrate became possible at all [7][10].
That is not a small point and it is worth sitting with. Your body was never set up to get these from a meal. It runs an assembly line to build them inside the one tissue that needs them most, which is an expensive thing for a body to choose to do.
It also explains why this is a new product category rather than an obvious one. For most of the history of this field there was nothing to buy. Researchers who wanted to test these molecules in animals had to make them synthetically, because no supply existed to evaluate [5]. In normal systems these are not eaten, they are made, and that was the end of the story until a concentrate rich in these chains became commercially available.
The nearest thing to a dietary connection in the verified literature is that EPA is the preferred substrate for elongation to VLC-PUFA [2]. That is a statement about enzyme chemistry, and it is not evidence that eating more of anything changes what is in a human retina.
Is this the same thing as the VLCFA blood test?
No. It is a common mix-up and worth clearing up.
Search that phrase and most of what comes back is clinical. A physician orders that blood panel to answer a metabolic question about how the body handles a particular class of fats, and the person who ordered it is the person who interprets it.
The molecules in this article are a different class and a different measurement. Very long chain saturated and very long chain polyunsaturated fatty acids are chemically distinct, and the same elongation enzyme makes different ones in different tissues [2]. A blood panel is also not a retina. Nothing on a routine lab report tells you what is in your photoreceptor outer segments.
Where the evidence actually stands
I run a company built on this molecule, so it is worth being precise about where the evidence sits today.
The functional work is preclinical. Gorusupudi and colleagues gavage fed a synthetic VLC-PUFA to mice, which is direct delivery into the stomach by tube, and reported a significant increase in retinal VLC-PUFA levels compared with controls, alongside improvement in the animals' visual acuity and electroretinography measurements [5]. The word significant attaches to the increase in retinal VLC-PUFA levels. It does not attach to the acuity or the electroretinography outcomes, and I am not going to attach it for them. What that study answers is the first question worth asking about any fatty acid supplement, which is whether the molecule can travel from the gut to the tissue that needs it. In mice it can, and visual function moved with it.
The 2025 result is the most striking and it came by a different route. Gao and colleagues reported that intravitreal supplementation, meaning injection directly into the eye, with 0.36 nmol of 24:5n-3, the direct product of ELOVL2, improved visual function in aged mice for up to 4 weeks [6]. Injection is not oral and mice are not people, and both of those belong in the same sentence as the finding. What it shows is that in an aged retina the decline was not fixed in place. Restore the fatty acid, and measured visual function came back.
The aging link, stated at its real strength. Gao and colleagues write that during aging, alterations in lipid metabolism lead to reduced content of very long chain PUFAs in the retina, which is associated with normal age-related reductions in contrast sensitivity, diminished photoreceptor function and delayed rod-mediated dark adaptation recovery [6]. Associated is not the same word as caused. Separately, Chen and colleagues showed in mouse retina that rising methylation at the Elovl2 promoter goes with falling expression, and that removing the methylation with a demethylating drug delivered into the eye raised expression again and rescued age-related visual function [7].
What we make
Vision lipids is our daily softgel built around this class of molecule. The design follows the bottleneck. Rather than trying to fix an aging enzyme, it supplies what that enzyme has stopped handing over, and the retina finishes the job itself. The ingredient is named on the label at a disclosed dose rather than hidden in a proprietary blend, so you can look it up and check it against anything written on this page. The label directs two softgels daily.
That is the whole product description, and it is deliberately short. This article exists to explain the molecule, not to sell you on it.
References
- Nwagbo U, Bernstein PS. Understanding the Roles of Very-Long-Chain Polyunsaturated Fatty Acids (VLC-PUFAs) in Eye Health. Nutrients. 2023;15(14):3096. DOI 10.3390/nu15143096. Open access. https://pmc.ncbi.nlm.nih.gov/articles/PMC10383069/
- Deak F, Anderson RE, Fessler JL, Sherry DM. Novel Cellular Functions of Very Long Chain-Fatty Acids: Insight From ELOVL4 Mutations. Frontiers in Cellular Neuroscience. 2019;13:428. DOI 10.3389/fncel.2019.00428. Open access. https://www.frontiersin.org/journals/cellular-neuroscience/articles/10.3389/fncel.2019.00428/full
- Berdeaux O, Acar N. Very-long-chain polyunsaturated fatty acids in the retina, analysis and clinical relevance in physiological and pathological conditions. OCL. 2011;18(5):284-290. DOI 10.1051/ocl.2011.0406. Open access. https://www.ocl-journal.org/articles/ocl/full_html/2011/05/ocl2011185p284/ocl2011185p284.html
- Agbaga MP, Merriman DK, Brush RS, et al. Differential composition of DHA and very-long-chain PUFAs in rod and cone photoreceptors. Journal of Lipid Research. 2018;59(9):1586-1596. DOI 10.1194/jlr.M082495. Open access. https://pmc.ncbi.nlm.nih.gov/articles/PMC6121944/
- Gorusupudi A, Rallabandi R, Li B, et al. Retinal bioavailability and functional effects of a synthetic very-long-chain polyunsaturated fatty acid in mice. Proceedings of the National Academy of Sciences USA. 2021;118(6):e2017739118. DOI 10.1073/pnas.2017739118. Open access via PMC8017942. https://pmc.ncbi.nlm.nih.gov/articles/PMC8017942/
- Gao F, Tom E, Rydz C, et al. Retinal polyunsaturated fatty acid supplementation reverses aging-related vision decline in mice. Science Translational Medicine. 2025;17(817):eads5769. DOI 10.1126/scitranslmed.ads5769. Paywalled at the publisher. Open data at https://datadryad.org/dataset/doi:10.5061/dryad.gxd25480h
- Chen D, Chao DL, Rocha L, et al. The lipid elongation enzyme ELOVL2 is a molecular regulator of aging in the retina. Aging Cell. 2020;19(2):e13100. DOI 10.1111/acel.13100. Open access. https://onlinelibrary.wiley.com/doi/full/10.1111/acel.13100
- Garagnani P, Bacalini MG, Pirazzini C, et al. Methylation of ELOVL2 gene as a new epigenetic marker of age. Aging Cell. 2012;11(6):1132-1134. Open PDF. https://air.unimi.it/bitstream/2434/208634/3/acel12005.pdf
- Dietary fish oil enriched in very-long-chain polyunsaturated fatty acid reduces cardiometabolic risk factors and improves retinal function. iScience. 2023. PMID 38047069. Open access: https://www.cell.com/iscience/fulltext/S2589-0042(23)02488-4
- Yeboah GK, Lobanova ES, Brush RS, Agbaga MP. Very long chain fatty acid-containing lipids, a decade of novel insights from the study of ELOVL4. Journal of Lipid Research. 2021;62:100030. DOI 10.1016/j.jlr.2021.100030. Open access: https://pmc.ncbi.nlm.nih.gov/articles/PMC8042400/
- Wen S, Wang Y, Yang S, Zhao X, Ju J, Yang H, Zhu G. Testis-specific knockout of Elovl4 reduces sperm motility and fertility in male mice. Scientific Reports. 2026;16:4344. DOI 10.1038/s41598-025-33558-7. Open access: https://www.nature.com/articles/s41598-025-33558-7