Brain–skin axis: what the research says

[INDICE:h2]

"The skin is a mirror of what you are experiencing." It is a phrase you see everywhere, and it is usually not accompanied by anything that explains why. In recent years, however, research has reconstructed with considerable precision the biological processes that connect the brain to the skin, and they are worth explaining — including the points where the evidence is still weak.

This article is based primarily on a review published in October 2025 in JAAD International, which selected 159 studies out of 814 reviewed, and on research published in 2025 in Science Immunology. You will find all the references at the end.

A methodological premise that the review’s authors state explicitly: much of what we know comes from animal models and in vitro studies. Human studies are mostly small, observational, and cross-sectional, and controlled trials are few. The mechanisms that follow are therefore well characterized biologically, but not all of them have been confirmed with the same degree of certainty in people.

First of all: where the nerves are in the skin

The connection between the brain and the skin is not a metaphor; it is an anatomical structure — and knowing this makes everything else much more concrete.

Cutaneous innervation is of two types. That of the autonomic nervous system reaches the arrector pili muscles and blood vessels: it is responsible for goosebumps and cold-induced vasoconstriction, that is, phenomena we do not consciously control. Sensory innervation consists of afferent fibers that originate from two plexuses, one deep in the dermis and one just beneath the epidermis.

From these plexuses branch free nerve endings and encapsulated nerve endings — Pacinian corpuscles in the deep dermis, Meissner’s corpuscles in the dermal papillae, followed by Krause and Ruffini corpuscles. Encapsulated endings are far less numerous than free endings.

Two details that change how you read everything that follows.

The first: free nerve endings are also found in the epidermis, not just in the dermis. In other words, the nerve reaches the skin’s most superficial layer — and this is the anatomical foundation of neurocosmetics. You do not need to reach the dermis to encounter a nerve ending.

The second: the fibers are not distributed uniformly. They are extremely numerous on the face and extremities, and relatively sparse on the back. The face is not more reactive than the rest of the body because of suggestion: it is the most densely innervated area we have.

Finally, the basal layer contains Merkel cells, mechanoreceptors that respond to vibrational stimuli. They have extensions that infiltrate between keratinocytes and detect the slightest change in the epidermal structure, transmitting it to the nerve ending with which they form the so-called Merkel complex.

The skin has its own stress axis

The most interesting point is also the least known. There is the classic hypothalamic-pituitary-adrenal axis: under stress, the hypothalamus releases CRH, the pituitary produces POMC, from which ACTH and MSH are derived, and the adrenal glands respond with cortisol, catecholamines, and androgens.

But there is also a peripheral HPA axis within the skin itself. Keratinocytes, mast cells, sebocytes, and melanocytes locally produce the same hormones, replicating the central stress response on a smaller scale and regulating inflammation in an autocrine and paracrine manner.

In other words, the skin does not merely undergo stress coming from the brain: it produces its own version locally.

What CRH does when it reaches the skin

CRH is the most direct link between psychological stress and skin inflammation, and it acts on several fronts simultaneously.

• It activates mast cells, which release histamine, TNF-α, and IL-6, increasing vascular permeability and facilitating the infiltration of immune cells.
• It stimulates sebocytes to produce IL-6 and IL-8, a mechanism implicated in acne and seborrheic dermatitis.
• It increases matrix metalloproteinases, the enzymes that degrade the structural proteins of the dermis: hence the slower healing, greater skin fragility, and accelerated aging observed under chronic stress.
• It triggers a vicious cycle: it promotes the local synthesis of cortisol, which in turn reinforces the release of CRH.

This latter point explains why skin stress, once it has begun, tends to sustain itself.

Cortisol and the barrier

The review goes into detail on cortisol: it increases free radicals and reduces epidermal lipids, hyaluronic acid, type 1 collagen, and stratum corneum hydration. All of these factors weaken the skin barrier and increase vulnerability to irritation, dryness, and infections.

Catecholamines — adrenaline and noradrenaline — do a different job: they cause vasoconstriction, reducing blood flow to the skin. This leads to hypoxia, oxidative stress, and slower healing. They also stimulate sebum production, which is the other link between stress and acne.

The second point calls for an anatomical clarification, because it explains how the connection actually works: the sebaceous gland is not innervated. No nerve reaches it—it is surrounded by blood vessels, not nerve fibers. Stress does not "speak" to it through the nervous system; it arrives hormonally through the bloodstream. This is why the effect on sebum is observed over weeks rather than within a few hours, unlike redness, which can appear within seconds precisely because the blood vessels are innervated.

A detail that is relevant to many women: adrenal androgens such as DHEA are converted locally in the skin into testosterone and dihydrotestosterone, which act on sebocytes, keratinocytes, and fibroblasts. A systemic hormonal imbalance is not necessary for the skin to be exposed to more androgens: it is enough for them to be produced locally.

The 2025 discovery: when stress switches off the defenses

This brings us to the most recent and most surprising study, published in Science Immunology in 2025.

The researchers showed that psychological stress compromises the response to Staphylococcus aureus through an unexpected pathway: dermal fibroblasts. Under stress, adrenergic signaling activates TGFβ, which in turn blocks fibroblasts' ability to produce cathelicidin, an antimicrobial peptide that the skin uses as its first line of defense. The bacterium finds the way clear.

The confirmation is that blocking adrenergic signaling or TGFβ restores normal defense: the mechanism is not a correlation but a causal chain.

One important caveat: this is a study in a mouse model. It is an important result, but it has not yet been shown that the same process occurs in humans. I mention this because the distinction matters, and because in beauty communications, results in mice are often presented as though they directly applied to people.

The 2025 review nevertheless confirms the overall picture on a broader basis: chronic stress reduces the production of antimicrobial peptides through cortisol, catecholamines, and neuropeptides, and this deficiency is implicated in atopic dermatitis, psoriasis, and rosacea.

The reference to rosacea is not accidental. Cathelicidin is the same molecule involved in the mechanism of rosacea together with kallikrein 5, the pathway targeted by actives such as azelaic acid. In other words, stress and rosacea reach the same molecular target from two different directions.

Not just cortisol

For years, popular explanations focused almost exclusively on cortisol. More recent literature paints a more nuanced picture, and it is worth updating it.

Neuropeptides

Substance P and CGRP are released by cutaneous nerve endings and cause mast cell degranulation, vasodilation, and the recruitment of immune cells. They are the main drivers of what is known as neurogenic inflammation: inflammation that starts in the nerve rather than in the immune system. Both also amplify the perception of itch and pain, which explains why the same irritation feels worse under stress.

This brings us back to the anatomical point from the beginning: the nerve endings of the face and extremities are more numerous, and neurogenic inflammation is most evident in those areas. It is no coincidence that reactive skin is especially common on the face.

NGF

Nerve growth factor is produced under stress by keratinocytes, mast cells, and fibroblasts. It helps healing, but its chronic overexpression leads to hyperproliferation, overactive sebaceous glands, and barrier impairment, contributing to stress-induced skin aging.

Oxytocin and sleep

Two less predictable elements. Acute stress increases oxytocin, which has a protective effect; chronic stress reduces it, resulting in slower healing and more inflammation. And elevated nighttime cortisol suppresses melatonin, disrupting the circadian rhythm and increasing the production of pro-inflammatory mediators.

In other words: sleeping badly is not an incidental factor; it is part of the mechanism.

The skin talks to the brain

The newer aspect concerns the opposite direction. Cytokines produced by inflamed skin can reach the central nervous system: IL-17 and IL-22 weaken the tight junctions of the blood–brain barrier, facilitating the infiltration of immune cells and sustaining neuroinflammation.

This is the biological reason why people with a chronic skin condition are more likely to experience anxiety and depression—quite apart, of course, from the psychological burden of living with it. The two mechanisms compound and reinforce each other.

What a neurocosmetic can really do

Let's get to the question that matters to people who buy cosmetics, and I'll try to answer it without sugarcoating.

A review published in 2026 in the journal Cosmetics proposes a useful definition: neurocosmetics are products formulated to interact with cutaneous neurosensory, neuroimmune, or neuroendocrine pathways relevant to the brain–skin axis. It also adds a distinction rarely seen on packaging: this definition should be kept separate from generic marketing claims about psychological or emotional benefits, which may reflect expectation, sensory pleasure, or the context of use more than any actual modulation of the skin's nervous system.

In other words: when a cream “reduces stress,” much of the effect may come from the fragrance, the texture, and the act of taking five minutes for yourself. It isn’t deceptive — the perceived well-being is real — but it isn’t the same as acting on a receptor.

The same review explicitly addresses the ethical issues surrounding claims about mood and stress in cosmetic products. As a company, this is an area where I prefer to take the cautious approach: a cosmetic product can improve the appearance and comfort of the skin and make a moment of the day more enjoyable. It does not treat anxiety and does not replace medical or psychological support.

That said, there are actives with documented mechanisms. Tephrosia purpurea extract is among the most studied in this field — and here too, transparency matters: the available efficacy data come from the manufacturer’s tests, not independent studies. The general concept is explained in the article on neurocosmetics.

What this means in practice

If research says that stress affects the barrier, lipids, hyaluronic acid, collagen, and antimicrobial peptides, the practical consequences are less spectacular than marketing would have you believe, but more solid.

1. Protecting the barrier comes first, because it is the target on which cortisol’s effects are best documented.
2. Simplify during difficult periods. Skin with reduced antimicrobial defenses and a weakened barrier is not the right time to introduce acids and retinoids.
3. Sleep is part of the routine, not an afterthought.
4. Don’t expect a cosmetic product to solve the cause. If stress is chronic, the skin is the symptom.

The practical side — which products, in what order, and with which actives — can be found in the article on reactive skin, cortisol, and barrier-supporting routines. If, on the other hand, you want to understand why your skin has become reactive, the analysis of the causes of skin sensitivity is the starting point.

What we still don’t know

I’ll end where I began, because it seems to me the most honest part.

The authors of the review point out three limitations: most of the evidence comes from animal or in vitro models; human studies are small and observational, so they cannot establish causal relationships; and almost all of them have looked at one system at a time, providing little information about how neuroimmune and endocrine pathways actually interact with one another.

The direction of the research is clear, the mechanisms are plausible and partly demonstrated. But we are far from being able to say that a cream alters your stress axis. When someone promises that, they are going beyond what the evidence allows.