Phytoceramides

Best-in-class: Phytoceramides

Plant-derived ceramides (from wheat/rice) that rebuild the skin's moisture barrier from within — 'oral skincare' for hydration and fine lines.

How Phytoceramides actually works

Ceramides are the lipids that fill the space between corneocytes in the stratum corneum — the mortar in the brick-and-mortar model of the skin barrier. Ceramide content falls with age and in atopic dermatitis. Oral plant-derived ceramides appear to be absorbed and incorporated, raising skin ceramide content from the inside.

Where to get Phytoceramides

The evidence for Phytoceramides

Graded by what exists behind each claim.

✅ Clinically validated

• RCTs show improved skin hydration and reduced dryness/fine lines with oral ceramides.
• Ceramides are core structural lipids of the skin barrier.

📊 Correlative data

• Skin ceramide content falls measurably with age and is lower in atopic dermatitis, which is the observational rationale. Both are observations about the skin rather than about eating ceramides.

🧪 Theoretical / extrapolated benefits

• Predicted to be digested to sphingoid bases, absorbed, and to signal increased endogenous ceramide synthesis in skin — not to be incorporated intact. That is a long chain of assumptions, and it is the reason the effect size should be expected to be small even if every step holds.

How to read these tiers: they say how much human evidence exists, not how well something works — and ✗ flags harm, never a disappointing trial. How the evidence tiers work →

What Phytoceramides actually does

The barrier this product is sold to rebuild is a real, named structure: the lipid matrix between the dead flattened cells of the stratum corneum, roughly half ceramide, a quarter cholesterol and the rest free fatty acid, arranged in stacked lamellae that water cannot cross easily. If that matrix is disordered, water leaves. That is what dry skin is at the level of physics.

Human skin ceramides are a family, and the family is specific. Twelve or more classes, built on three sphingoid backbones — sphingosine, phytosphingosine and 6-hydroxysphingosine — each carrying an amide-linked fatty acid that is unusually long, C24 to C26 rather than the C16 to C18 of the rest of the body. One class matters more than the rest: the acylceramide EOS, an esterified omega-hydroxy sphingosine carrying linoleic acid, which is covalently bound to the corneocyte protein envelope and rivets the lipid lamellae to the cell. Lose EOS and the barrier fails regardless of how much total ceramide is present.

The enzymes that build it, by name. ELOVL4 makes the very-long-chain fatty acids. Ceramide synthase 3, CERS3, is the skin-specific synthase that joins them to the sphingoid base — mutations in it cause a lamellar ichthyosis, which is how we know it is load-bearing. The final step happens outside the living cells: glucosylceramide is secreted from lamellar bodies and beta-glucocerebrosidase strips the glucose in the extracellular space to give the mature ceramide. Skin barrier lipid is therefore assembled in situ, from precursors the keratinocyte made itself.

The molecule you swallow is not the molecule your skin makes. Plant ceramides are glucosylceramides built on 4,8-sphingadienine or 8-sphingenine — sphingoid bases carrying a double bond at position 8 and, in many plants, a 9-methyl branch. Human skin makes neither. So even the most generous absorption story cannot end with a wheat ceramide sitting in your lamellae; it has to end with a signal, or with a recycled fragment.

Which leaves the only mechanism that survives the gut: signaling. Digestion yields free sphingoid bases, and sphingoid bases are not inert cargo — sphingosine and its phosphate are signaling lipids acting on S1P receptors, on protein kinase C, and on keratinocyte differentiation programs. A plausible chain is: eat glucosylceramide, absorb sphingoid base, alter epidermal differentiation and lipid synthesis. Every step of that is plausible and the chain has never been demonstrated end to end in a person.

Cell, rodent, human — and where it stops

The digestion step, measured, and it does not favor the product. Maize sphingolipids fed to rats were digested rather than absorbed intact, and when the resulting plant-specific base was offered to differentiated Caco-2 cells, accumulation of sphingoid bases was lower for each isomer of sphingadienine than for sphingosine (P < 0.05) Sugawara 2003. Read that twice: the plant base — the one in your capsule — was taken up less well than the mammalian one. The unusual chemistry that makes a phytoceramide a phytoceramide is a liability at the brush border.

Rodents, where the endpoint was the right one. Hairless mice fed glucosylceramide from rice bran and germ showed transepidermal water loss significantly reduced at 2 weeks and stratum corneum flexibility increased at 3 weeks against controls Tsuji 2006. Transepidermal water loss is the instrumented barrier measurement, which makes this a real mechanistic result rather than a coat-condition observation.

The human trial, with its numbers. Fifty-one women aged 20 to 63 with dry to very dry skin took 350 mg a day of a wheat extract oil or placebo for three months; skin hydration rose significantly from day 0 to day 84 on the arms (P < 0.001) and on the legs (P = 0.012) Guillou 2011. Note where the effect was measured — limbs, not face — and note the size of the trial.

The obstacle is an arithmetic problem hiding inside a dose figure. “350 mg” is 350 mg of wheat extract oil, and glucosylceramide is a minor percentage of that oil rather than all of it. So the actual ceramide-class dose is on the order of ten milligrams a day. Now the other side: an adult carries roughly two square meters of skin with a stratum corneum around twenty micrometers thick, which is tens of grams of tissue, of which about a tenth is lipid and about half of that is ceramide — call it a gram or two of ceramide, replaced every couple of weeks. That is on the order of a hundred milligrams of ceramide synthesized per day. Ten milligrams swallowed, digested to bases that are absorbed less efficiently than mammalian ones Sugawara 2003, cannot be a supply effect. Whatever moved hydration in that trial Guillou 2011 did it as a signal.

And that is why the honest expectation is a small effect. The chain has four places to lose an order of magnitude — hydrolysis, uptake, first-pass handling, and whether an epidermal keratinocyte ever sees the fragment — and the trial result is correspondingly modest and measured with an instrument rather than visible in a mirror.

Phytoceramides — which form, and does it matter

Wheat, rice, konjac and sweet potato are four different lipid profiles wearing one word. The human hydration result belongs to a wheat extract oil at 350 mg Guillou 2011; the mouse barrier result to rice bran and germ glucosylceramide Tsuji 2006; the digestion work to maize Sugawara 2003. They differ in the sphingoid base, in the fatty acid chain length and in the sugar head, and no trial has compared two of them against each other. Buying “phytoceramides” without a source is buying an unspecified one of four.

The 350 mg on the label is almost never 350 mg of ceramide. It is 350 mg of extract, matching the trial's dosing unit Guillou 2011, and the glucosylceramide content of that extract is what actually varies between products. A label that states a ceramide percentage or a milligram figure for glucosylceramide is telling you something; a label that states only the extract mass is telling you the weight of the powder.

Wheat-derived is the version with the human trial and the gluten problem. The extract is an oil fraction and highly refined, so gluten protein should be minimal — but “should be” is not a certificate, and celiac disease is not a preference. Rice-derived material sidesteps the question entirely and carries the rodent barrier data Tsuji 2006 rather than the human hydration data. That is a genuine trade-off between evidence and safety, and it is the buying decision on this page.

Oral against topical is not a close contest for the same molecule. A topical ceramide-cholesterol-fatty-acid blend is delivered to the exact compartment it belongs in, at a concentration you control, with no digestion step. Oral ceramides are worth considering as an addition to that, not as a substitute for it, and any page implying otherwise is selling the harder route as though it were the better one.

Take it with fat. Sphingolipids are lipids and their digestion depends on bile and on lipase activity, so a capsule swallowed with water on an empty stomach is the least favorable condition for the one step Sugawara 2003 shows is already inefficient. No trial has tested this; it is mechanism, and it costs nothing to act on.

What would have to be true, and how you would know it was not

1. Pick a site you neglect, and watch water rather than wrinkles. What to watch: dryness, flaking and tightness on one patch of forearm or shin, plus how many hours a moisturizer lasts before the skin feels tight again. The trial measured hydration on arms and legs, and those are the sites where it reached significance Guillou 2011. How long before it means anything: eight to twelve weeks at around 350 mg, because barrier lipid is replaced as the epidermis turns over. What will fool you: winter. Skin ceramide content falls with cold and low humidity and recovers by itself, so a January-to-April course runs with the weather.

2. The prediction that cuts against the product. Predict that fine lines do not measurably change in twelve weeks. Wrinkle depth is a dermal collagen and elastin property; the mechanism here is an epidermal barrier one, and the trial that worked measured hydration Guillou 2011. If a before-and-after photograph shows your lines improving over a season, the likelier explanations are hydration filling them slightly, better light, and the fact that you were looking.

3. Predict a rice-derived product performs like a wheat-derived one, and be prepared to be wrong. They carry different sphingoid bases and different uptake efficiencies Sugawara 2003, so a sequential twelve-week test of one then the other, same site, same season, is a real experiment that nobody has published. A clear difference would be more informative than another underpowered hydration trial.

4. Predict nothing on any blood marker. There is no plausible route from ten milligrams of digested plant sphingolipid to hs-crp, a lipid-panel or any other systemic number at twelve weeks. If you are hoping to see this product in bloodwork, the honest answer is that it is not a bloodwork product and a page that offered you a panel for it would be selling you a test.

What nobody has tested yet

Nobody has looked at the skin's own ceramides after oral dosing. The entire mechanism claims a change in epidermal lipid composition, and the way to see it is tape stripping plus mass spectrometry — a routine dermatology technique. Twelve weeks of dosing and a lipidomic profile of the stratum corneum before and after would say directly whether the mechanism is real, and it has not been run.

Nobody has given the absorbed species instead. If digestion to sphingoid bases is the route Sugawara 2003, then feeding the base directly should work better than feeding the intact glucosylceramide, because it skips the inefficient step. That is a cheap, obvious comparison and the result would either simplify the product or kill the mechanism.

Nobody has dosed by ceramide content. Every trial and every label doses by extract mass Guillou 2011, which means two products at “350 mg” can differ severalfold in the molecule of interest. Standardizing the dose to milligrams of glucosylceramide would make trials comparable for the first time.

And nobody has tested it where the deficiency is real. Stratum corneum ceramide content is genuinely reduced in atopic dermatitis, which makes it the population in which a supply or signaling effect would be easiest to detect. The trials recruited healthy women with dry skin Guillou 2011. The obvious study sits unrun, and the observation about atopic skin keeps being quoted as though it were the evidence.

Phytoceramides — its own safety story, not its category's

The wheat question is the one that matters, and it is not hypothetical. The human evidence belongs to a wheat-derived extract Guillou 2011. Refined oil fractions carry little protein, but celiac disease responds to trace gluten and a supplement label is not a gluten-free certification unless it says so. Rice, konjac and sweet potato versions exist precisely for this, and swapping to one means accepting a weaker evidence base in exchange for removing a real risk.

Sphingoid bases are signaling molecules, which is the argument for the product and also the reason to be curious about long-term use. Sphingosine-1-phosphate signaling regulates lymphocyte trafficking — it is the target of a class of multiple sclerosis drugs — as well as keratinocyte differentiation. Dietary sphingolipid at these doses has no demonstrated systemic effect and nobody has measured plasma sphingoid base levels during chronic supplementation. That is a gap, stated as a gap.

The direct tolerability is unremarkable. Gastrointestinal upset is the main report and there are no established drug interactions. Dietary sphingolipids are eaten daily in milk, eggs and soy, which is a fair argument that the class is familiar to the gut — though not that a concentrated plant extract is the same thing as a glass of milk.

Duration is the honest limit. Three months Guillou 2011 is the longest human exposure on record here. Nothing about that period licenses a statement about a decade, and the category's habit of implying one is worth naming.

And the failure mode: dry, itchy skin that is a diagnosis. Persistent dryness with itch that wakes you, thickened plaques, scaling in a defined pattern, or dryness that arrived with fatigue, weight change or hair loss is a reason to check thyroid function and see a clinician. A barrier supplement that makes the surface feel better is exactly the thing that makes waiting feel reasonable.

Sources read for this page

• Guillou S, et al. The moisturizing effect of a wheat extract food supplement on women's skin: a randomized, double-blind placebo-controlled trial. International Journal of Cosmetic Science, 2011 · PMID 20646083
• Sugawara T, et al. Digestion of maize sphingolipids in rats and uptake of sphingadienine by Caco-2 cells. The Journal of Nutrition, 2003 · PMID 12949364
• Tsuji K, et al. Dietary glucosylceramide improves skin barrier function in hairless mice. Journal of Dermatological Science, 2006 · PMID 17000082

How you would know if it worked

Phytoceramides — safety & side effects

Bloodwork to run alongside Phytoceramides

Baseline first, then again at 8–12 weeks.

The Basics — Start Here panel covers these in one order — 4 markers, $32.40 with the discount applied.

Check results you already have → · All 103 markers A–Z

Phytoceramides — frequently asked questions

What Phytoceramides is used for

Phytoceramides appears under 1 goal in the goal router.

Related Skin & Structural supplements

← Browse the full Supplement Vault

↑ Back to on this page