The production of collagen, a protein crucial for skin elasticity and firmness, has long been considered primarily linked to the activity of fibroblasts. However, a study recently published in Nature Communications suggests that keratinocytes could play a significant role in collagen synthesis. This raises the following question: should pro-collagen cosmetics target keratinocytes instead? Let's explore this question together.

Collagen synthesis: Should skincare products target keratinocytes rather than fibroblasts?
- Collagen, fibroblasts, keratinocytes: what are we discussing?
- Could keratinocytes also synthesise collagen?
- Why is this discovery of interest to the skincare industry?
- Sources
Collagen, fibroblasts, keratinocytes: what are we discussing?
Highly prevalent in the skincare industry, collagen embodies the promise of firm, resilient skin that withstands the passage of time. However, before being one of the most utilised ingredients in skincare formulations, collagen is primarily the most abundant structural fibrous protein in the dermis. It forms a dense network, the true framework of the skin, providing both its mechanical resistance and flexibility. Among the many forms of collagen present in the body, types I and III are the most represented in the skin. More specifically, young skin is composed of about 80% type I collagen and approximately 15% type III collagen. With age, the ability to renew collagen naturally decreases and the fibres become thicker and shorter, which impairs their function.
Approximately 1.5%
of collagen naturally lost per year from the age of 25.
The synthesis of collagen relies on specialised cells in the dermis: fibroblasts. These not only produce collagen, but also other components of the extracellular matrix, such as elastin and glycosaminoglycans, including hyaluronic acid, thus contributing to the architecture and hydration of the skin. Closer to the skin's surface, in the epidermis, are the keratinocytes. Sometimes compared to bricks, keratinocytes ensure the renewal of the epidermis and contribute to its barrier function. They also interact with immune cells, nerve endings and the fibroblasts themselves. Indeed, the epidermis and dermis do not function as closed vessels but communicate via biochemical mediators (cytokines, neuropeptides...) that modulate their respective activities.

Could keratinocytes also synthesise collagen?
This is suggested by a recent study conducted by SATOH A. and his team. The researchers focused on collagen synthesis in axolotls, amphibians with transparent skin and significant regenerative capabilities. They used various imaging techniques on axolotls of different sizes (from 5 to 12 cm) to track the formation of the dermis as they grew, including collagen-specific fluorescent probes (DAR and DAF). These two probes differ in their structure and affinity: DAR is a small peptide molecule capable of recognising partially unfolded α-helices of collagen, making it an excellent probe for marking procollagen, the immature form still in the process of assembly. Conversely, DAF is a dye activated by structural affinity that preferentially targets mature fibres, already organised into a stable triple helix. By first injecting a red marker (DAR) and then a second green marker (DAF) a few days later, the researchers were able to distinguish old collagen fibres from new ones, while identifying their degree of maturity. This technique is called pulse-chase and allows us to see where and when collagen is produced.
Surprisingly, researchers observed that the collagen in the skin of young 5 cm axolotls was already well organised into lattice-structured fibres, even in the absence of fibroblasts. By analysing the expression of the gene Col1a1, which codes for type I collagen, they detected a strong signal in the keratinocytes of the basal layer of the epidermis, the layer in direct contact with the dermis. The presence of procollagen, an immature form of collagen, was also detected in this location using immunofluorescence. Finally, electron microscopy confirmed that these keratinocytes indeed contained procollagen ready to be secreted through the basal membrane. As the axolotl grew, the researchers observed through histological staining, notably Masson's trichrome, a popular method for highlighting collagen fibres, that the dermis transforms and becomes more complex. From a simple initial sheet, it becomes a structure in three well-organised layers:
Stratum baladachinum (SB): a layer located just beneath the epidermis, not very dense and synthesised by the keratinocytes.
Stratum spongiosum (SS): This is an intermediate layer where cells originating from the mesoderm, particularly fibroblasts, begin to appear.
Stratum compactum (SC): a deep, dense, and regular layer, with a highly structured network of orthogonal fibres.
It was only from 8 cm that scientists detected the arrival of mesenchymal cells, namely fibroblasts. These cells penetrate the existing collagen matrix through certain matrix metalloproteinases (MMPs), enzymes capable of locally digesting collagen to allow cell invasion. Once settled in the dermis, the fibroblasts deploy filopodia, extensions that insert between existing collagen fibres. Thanks to these structures, they modify, thicken and reorganise the initial network formed by the keratinocytes. The use of electron microscopy has allowed the tracking of the evolution of collagen fibres during the growth of axolotls. At 5 cm, the fibres are thin and isolated, while at 12 cm, they are significantly thicker and intertwined, particularly in the stratum compactum. This change bears witness to the work of the fibroblasts on the matrix initially produced by the keratinocytes. The figure below summarises the structural evolution undergone by the collagen.

In order to determine whether this phenomenon was unique to the axolotl or shared by other species, researchers repeated their analyses on zebrafish, chick embryos, and mouse embryos. In each case, they observed the expression of the gene Col1a2, which codes for another form of type I collagen, in the keratinocytes, above the basal membrane, suggesting that this mechanism of collagen production by the epidermis is conserved in evolution, at least during embryonic development.
The authors, however, remain cautious about their work. Indeed, even though the axolotl provides a good visualisation of developing skin, it remains an amphibian, with characteristics that differentiate it from human skin. Its skin is thinner, devoid of hair, sebaceous glands, and a thick horny layer, unlike human skin. These structural and functional differences limit the direct extrapolation of the results. It would thus be interesting for additional studies to be conducted on human keratinocyte cultures to test their ability to produce collagen under certain conditions (oxidative stress, healing...). Furthermore, the mechanisms that explain the triggering of collagen production and how these fibres cross the basal membrane are still unknown.
Furthermore, the study primarily relies on young or embryonic models. It is not yet established that adult keratinocytes, especially in humans, retain this ability to produce collagen. Therefore, it is possible that this function is transient, specific to certain stages of development or tissue regeneration. If this is confirmed, future skincare strategies could then primarily target young skin, to delay the natural loss of collagen associated with age as much as possible.
This study presents a more dynamic perspective on collagen formation, where keratinocytes initiate its structure and fibroblasts shape it.
Why is this discovery of interest to the skincare industry?
The notion that the epidermis can actively contribute to the formation of dermal collagen is not without implications for the cosmetic industry. Until now, products aimed at enhancing skin collagen synthesis have sought to target fibroblasts and stimulate their activity. However, if it turns out that keratinocytes can initiate collagen production, then the epidermis becomes a prime target, potentially more easily accessible.
Furthermore, it's noteworthy that the axolotl is capable of regenerating its skin without leaving any scars, a process that researchers have often compared to embryonic development. This study shows that the keratinocyte production of collagen precedes the arrival of fibroblasts and that this timeline appears necessary for an orderly organisation of the extracellular matrix. Applied to human medicine, this suggests that stimulating early collagen production by keratinocytes could promote cleaner healing that leaves no marks.
This discovery paves the way for a reassessment of traditional skincare targets. However, further studies on human skin are required to confirm these results and understand their concrete implications.
Sources
SIBILLA S. & al. An Overview of the Beneficial Effects of Hydrolysed Collagen as a Nutraceutical on Skin Properties: Scientific Background and Clinical Studies. The Open Nutraceuticals Journal (2015).
GENOVESE L. & al. An Overview of the Beneficial Effects of Hydrolysed Collagen as a Nutraceutical on Skin Properties: Scientific Background and Clinical Studies. The Open Nutraceuticals Journal (2015).
SATOH A. & al. Keratinocyte-driven dermal collagen formation in the axolotl skin. Nature Communications (2025).
Diagnostic
Understand your skin
and its complex needs.