Strategies to make human skin models based on cellular senescence for ageing research

Human skin ageing is closely related to the ageing of the whole organism, and it ’ s a continuous multisided process that is influenced not only by genetic and physiological factors but also by the cumulative impact of environmental factors. Currently, there is a scientific community need for developing skin models representing ageing processes to (i) enhance understanding on the mechanisms of ageing, (ii) discover new drugs for the treatment of age-related diseases, and (iii) develop effective dermo-cosmetics. Bioengineers worldwide are trying to reproduce skin ageing in the laboratory aiming to better comprehend and mitigate the senescence process. This review provides details on the main ageing molecular mechanisms and procedures to obtain in vitro aged skin models.


Introduction
The largest human organ is the skin.In most adults, the weight of the skin is 15-20 % of the total body mass and the area of the skin is around 1.5-2 m 2 (Mescher, 2018).Skin functions include protection from the environment, infectious agents and viruses; maintaining body temperature and humidity; and shielding internal organs; and, immune and regulatory capacity (Chiang et al., 2012;Gould, 2018;Pasparakis et al., 2014).
Over time, the skin and other human organs undergo many changes that lead to a deterioration of their basic functions (Nakamura et al., 2018).Skin ageing is a multifaceted process influenced by both intrinsic and extrinsic factors (Lowry, 2020;Quan, 2023).Intrinsic ageing, also known as chronological ageing, involves genetically predetermined mechanisms that lead to the gradual decline in skin function over time (Ho and Dreesen, 2021).Extrinsic factors, such as ultraviolet (UV) radiation (Rittié and Fisher, 2002), pollution (Lecas et al., 2016), and lifestyle choices (Guyuron et al., 2009), exacerbate these effects by accelerating the ageing process through mechanisms such as oxidative stress, inflammation, and extracellular matrix degradation (Kammeyer and Luiten, 2015).Several observable manifestations of skin ageing include evident features such as wrinkle formation, pigmentation changes, thinning, desiccation, coarseness, and the apparent visibility of the dermal vasculature (Georgakopoulou et al., 2021;Ghadially et al., 1995;Gilchrest, 1989).In addition to the impact on human health, obvious changes in appearance bring several mental and social complications like diminished self-esteem and psychological discomfort (Farage et al., 2015;Gupta and Gilchrest, 2005;Koblenzer, 1996).On the other hand, reduction in epidermal thickness, loss of rete pegs and ridges, loss of melanocytes and Langerhans cells, fragility of skin capillaries, reduction in collagen and elastic fibres, and xerosis etc. (Georgakopoulou et al., 2021;Giangreco et al., 2010;White-Chu and Reddy, 2011) can cause several skin conditions in older skin and require medical support (Volzer et al., 2023).
As population ageing becomes a global reality, with United Nations (UN) data on elderly population showing a steady increase from 727 million people over the age of 65 years in 2020 to a projected 16.0 % of the world's population by 2050 (United Nations Department of Economic and Social Affairs, 2020), the vulnerability of the elderly to skin disease escalates.As individuals age, the role of the skin is compromised due to a decline in epidermal barrier function, wound healing capacity, immune response efficiency, lipogenesis, vitamin D synthesis, and sweat production (Chang et al., 2013).This is associated with increased susceptibility to skin disorders such as cutaneous fungal infections and herpes zoster (Chang et al., 2013), tinea unguium (e.g., onychomycosis), xerosis cutis (e.g., eczema craqueleé), seborrheic keratosis (e.g., dermatosis papulosa nigra), pruritus, tinea pedis, nail disorders (e.g., onychorrhexis, onychauxis), cutis laxa senilis (e.g., elastosis), actinic keratosis, seborrhoea capitis, other nonthrombocytopenic purpura (e.g., senile purpura), irritant contact dermatitis, varicose veins of the lower extremities, xanthelasma of the eyelid, granuloma annulare, vitiligo (Blume-Peytavi et al., 2016), hemangioma (e.g., cherry angioma), melanoma and other malignant neoplasms of skin (Garcovich et al., 2017;Nakamura et al., 2023;Sinikumpu et al., 2022).In the geriatric population, research indicate that around 80 % of people over the age of 70 have one or more skin diseases requiring medical intervention (Sinikumpu et al., 2020).Age-related skin conditions also include skin pressure ulcers, which are a common and significant concern for immobilized residents in long-term care facilities (Ni et al., 2024), chronic wounds like venous, pressure, or diabetic foot ulcers (Alam et al., 2021;Gould et al., 2015), as well as chronic and relapsing inflammatory skin conditions namely but are not only psoriasis, atopic dermatitis, contact dermatitis, seborrheic dermatitis, rosacea, and Grover disease (Hessler-Waning and Heinecke, 2024).
This review explores various methodologies for replicating skin ageing in vitro, leveraging established molecular pathways associated with ageing.It covers diverse exposure conditions, ageing biomarkers, and inducing agents, while also examining different skin models, such as monolayers, reconstructed human epidermis (RHE), reconstructed human skin (RHS), skin biopsies, and human skin organ cultures (hSOC).By providing insights into the strengths and limitations of each model, this review aims to assist researchers in selecting the most appropriate model for their specific inquiries, thus advancing our understanding of skin ageing and facilitating the development of targeted interventions.With an ageing population on the rise, there is an urgent need to explore innovative approaches to address the challenges posed of skin ageing and improve the quality of life of individuals worldwide.

Molecular mechanisms of skin ageing based on cellular senescence
Cellular senescence and organismal ageing are intertwined processes characterized by a gradual decline in cellular and overall organismal functions, leading to chronic diseases such as cardiovascular disease, cerebrovascular accident, diabetes etc. (Ho and Dreesen, 2021;van Deursen, 2014).Leonard Hayflick's pioneering work in 1965 revealed that human lung fibroblasts undergo a finite number of divisions before entering a state of irreversible growth arrest known as cellular senescence (Hayflick, 1965).This phenomenon is associated with several hallmarks, including the inability to proliferate, resistance to apoptosis, and secretion of pro-inflammatory factors that contribute to tissue degradation (Gensler and Bernstein, 1981;Wang and Dreesen, 2018).Telomere shortening, first proposed by Watson and Olovnikov in the 1970s, emerged as a key molecular mechanism underlying cellular senescence, where the gradual erosion of telomeres triggers a persistent deoxyribonucleic acid (DNA) damage response, ultimately leading to growth arrest (Olovnikov, 1973;Victorelli and Passos, 2017;Watson, 1972;Yaar et al., 2002).In addition, various stressors, including oncogene activation (Serrano et al., 1997), epigenetic changes (Munro et al., 2004), oxidative stress (Yaar et al., 2002), mitochondrial dysfunction (Wiley et al., 2016), and continuous activation of DNA damage checkpoints (d'Adda di Fagagna et al., 2003) can induce cellular senescence and contribute to ageing (Wang and Dreesen, 2018).Importantly, around 20-50 % of cells in the skin of the elderly people are senescent, imparing natural physiological functions such as regeneration and exacerbating overall organismal ageing (Dimri et al., 1995).Concurrent with cellular ageing, organismal ageing manifests as a progressive decline in life expectancy and physiological function, driven by both intrinsic and extrinsic factors (Costello et al., 2022;Franco et al., 2022;Zhang and Duan, 2018).
Nowadays, plenty of biomarkers are investigated taking place inside and outside living organisms to detect ageing (Gorgoulis et al., 2019;Mohamad Kamal et al., 2020).However, the detection and quantification of senescent cells, especially within living organisms, presents a considerable problem in ageing studies.While these cells typically display specific signs like altered morphology and increased senescence-associated-β-galactosidase (SA-β-gal) activity, relying only on enzymatic assays has limitations, particularly with fixed or frozen tissue samples (van Deursen, 2014).Moreover, the presence of non-specific SA-β-gal activity complicates accurate identification, as it occurs in various unrelated cell types (Mohamad Kamal et al., 2020).Efforts to improve detection involve combining SA-β-gal with other biomarkers using techniques such as flow cytometry, and applying these methods directly to tissue samples remains a challenge.Additionally, while SA-β-gal activity is a known indicator of senescent cells, its absence doesn't necessarily mean that cellular senescence is absent, highlighting the complexity of ageing mechanisms (Lee et al., 2006).Addressing these challenges is crucial for advancing our knowledge of ageing and developing interventions targeting senescent cells.
Senescent cells stop dividing permanently, mainly during a specific stage of the cell cycle called G1/S transition.This is mainly caused by increased levels of proteins like p16 INK4a , p21 CIP1 and p53, which block cell division (Toutfaire et al., 2017;Wang and Dreesen, 2018).In skin biopsies from people of different ages, p16 INK4a levels rise with age in both the top and deeper layers of skin, making it a strong marker of skin ageing (Ressler et al., 2006).Moreover, p16 INK4a levels seem linked to how fast our bodies age biologically, not just how many years we've lived.In a study comparing middle-aged people with long-lived relatives to others, those with long-lived relatives had fewer p16 INK4a cells in their skin, suggesting slower biological ageing (Waaijer et al., 2012).Another indicator of cellular senescence is Lamin B1, an intermediate filament protein constituting part of the nuclear lamina.The nuclear lamina acts as a structural framework within the nuclear envelope, establishing multiple connections with essential proteins of the nuclear membrane and chromatin (Dobrzynska et al., 2016).Its reduction in senescent cells occurs at both the genetic and protein levels and is regulated by various pathways (Dreesen et al., 2013).Downregulation of Lamin B1 in cells during UV exposure, replicative senescence and oncogene activation (Ivanov et al., 2013;Wang et al., 2017).Chromatin remodelling is also a key factor in senescence.During normal cellular ageing, there is a notable depletion of global heterochromatin, indicated by markers H3K9me3 and H3K27me3 (Tsurumi and Li, 2012).Similarly, individuals with Hutchinson-Gilford progeria syndrome, a disorder characterized by accelerated ageing, display a marked reduction in heterochromatin (Chojnowski et al., 2020;Lamis et al., 2022).Moreover, the decline in chromatin organization linked with senescence, there is a decrease in overall DNA methylation and expression of the DNA methyltransferase DNMT1 during cellular ageing.These changes in DNA methylation profiles can differentiate cells based on their developmental potential and may serve as indicators of chronological and biological age (Horvath, 2013).DNA damage foci can be identified by staining DNA damage response proteins such as 53-binding protein 1 (53BP1) or phosphorylated histone H2AX.When these foci are specifically located on telomeric DNA, they are termed telomere-associated foci.However, these biomarkers are also present in non-senescent cells experiencing temporary DNA damage, indicating a lack of specificity for senescence identification (Sharpless and Sherr, 2015;Wang and Dreesen, 2018).Thus, it is recommended to combine their presence with other senescence biomarkers.Persistent DNA damage foci, known as "DNA segments with chromatin alterations reinforcing senescence", can be detected via immunocytochemistry by their positivity for 53BP1 and the tumor suppressor promyelocytic leukemia protein.In human senescent cell types like fibroblasts, epithelial, and endothelial cells, the presence of four or more positive foci per cell indicates persistent DNA damage (Rodier et al., 2011;Toutfaire et al., 2017).
Age-related changes in mitochondrial DNA (mtDNA) include mutations and the common mtDNA4977 deletion detected in various human tissues and cells (Passos et al., 2007;Wiley et al., 2016) more prevalent in photoaged skin (Koch et al., 2001).UV exposure correlates directly with the common deletion, persisting up to 16 months post-exposure (Berneburg et al., 2004).This deletion serves as a marker for skin ageing.Additionally, age-dependent accumulation of histone variant H2A.J occurs in mouse and human skin, especially in irradiated mouse skin.Elevated H2A.J levels, alongside p16 INK4a , are found in carcinogen-induced preneoplastic lesions but decrease in advanced papillomas in mouse skin (Azazmeh et al., 2020).H2A.J upregulation promotes inflammatory gene expression and increases senescence-induced hyperinflammation (Contrepois et al., 2017).
The senescence-associated secretory phenotype (SASP) is the one of important feature profiles which includes various biological components secreted by senescent cells such as chemotactic cytokines (e.g., chemokine (C-X-C motif) ligand 1; inflammatory cytokines (e.g., interleukin-1 (IL-1), IL-6, IL-8); matrix metalloproteinases (MMP-1, MMP-3, MMP-9), pro-oxidant reactive oxygen species (ROS) and growth factors (insulinlike growth factor-binding protein and hepatocyte growth factor) (Coppe et al., 2010;Franco et al., 2022).ROS is a natural side product of aerobic metabolism, which is important for the immune system and the regulation of cell activities (Ponnappan and Ponnappan, 2011).However, an increased level of ROS can cause oxidative imbalance, damage mtDNA, and make alterations in biological molecules including lipids, proteins, and DNA.ROS is the central player of the senescence acceleration (Kammeyer and Luiten, 2015).The main pathways to elevate ROS levels are acute sun exposure, air pollution and its intrinsic formation (Costello et al., 2022).The keratinocytes and melanocytes reveal the strongest response to oxidative stress and antioxidant capabilities, which tend to weaken with age (Dai et al., 2023).Oxidative stress and the consequent oxidative damage can intensify skin pigmentation and the ageing process, leading to alterations in the uniformity of skin tone, the development of wrinkles, sagging, dryness, and increased skin roughness (Chen et al., 2021).Keratinocytes and fibroblasts secret MMPs, which represent calcium-dependent endopeptidases containing zinc.Their principal function is the hydrolysis of the peptide bonds in collagen and elastin, which lead normal skin to wound healing, dermal reorganization, and cell migration.When the endogenous tissue inhibitors of MMPs bind with the catalytic zinc atom of MMPs, their downregulation occurs.During ageing, MMP levels increase, and ROS act as secondary messengers in that process (Gragnani et al., 2014).Moreover, in the Mitogen-Activated Protein Kinase pathway occurs the formation of the global transcription factor namely activator protein 1, which activates MMPs.It results in diminished collagen type I and III synthesis (Costello et al., 2022).Another ageing-related mechanism is glycation, which is described by elevated concentrations of reducing sugars binding to proteins.In fact, the generation of advanced glycation end products (AGEs) is a characteristic feature of the ageing process (Dyer et al., 1993).AGEs have a cumulative effect, an increase of around 33 % from 20 to 85 years noticed, which outcomes in skin elasticity impairment (Dyer et al., 1993).The accumulation of AGEs, increased MMPs activity, oxidative stress and senescence of cells are driven by intrinsic and extrinsic components, these mechanisms work together and lead to skin ageing.Fig. 1 presents a scheme of the described skin ageing molecular mechanisms and their interplay.
Aged cells stay metabolically active and produce SASP factors (e.g., IL-1, IL-6, IL-8, MMP-1, MMP3, and ROS), that modify tissue homeostasis and might stimulate ageing in nearby cells (Biran et al., 2015;da Silva et al., 2019).Another proposed ageing process includes the communication and signalling between cells through exosome transport, involving the transfer of microribonucleic acid or proteins via cytoplasmic bridges (Terlecki-Zaniewicz et al., 2018).Thus, the accumulation of dysfunctional senescent cells harms ageing tissues due to their diminished capacity to promote tissue neogenesis.SASP produced by senescent cells gives an inflammatory signal to tissue and provokes a chronic state of inflammation, which could increase the chances of getting cancer with metastasis (Ho and Dreesen, 2021).Furthermore, the senescence mechanism has significant importance in pre-malignant tumours and restricts the proliferation of cells with irreparable DNA damage (Collado and Serrano, 2010).High mobility group box-1 (HMGB1), an inflammatory mediator (Yamada and Maruyama, 2007), relocates from the nucleus to the cytoplasm and extracellular space in senescent cells, triggering the release of SASP factors like IL-1β, IL-6, and MMP3 (Biran et al., 2017).UVB exposure prompts HMGB1, IL-1, and IL-6 release in keratinocytes and reduces nuclear HMGB1 expression in mouse skin (Johnson et al., 2013).While HMGB1 is used as a marker for senescent cells, its efficacy in human skin remains unexplored (Wang and Dreesen, 2018).
Considering the established and previously described molecular mechanisms and biomarkers, generating experimental models for investigating skin ageing becomes feasible.The following section will reveal human skin models designed for studying skin ageing, categorized according to various molecular mechanisms, ranging from the most widely utilized to the less commonly employed model.

Using ageing molecular mechanisms to develop skin models
The design of reliable models for aged skin is an important tool for studying aged skin behaviours and developing effective therapies.Nowadays, various skin models are developed starting from the simple monolayer cell culture of primary or immortalized skin cells (Valerio et al., 2021) to the advanced three-dimensional (3D) reconstructed models including RHE with fully differentiated keratinocytes (Lowenau et al., 2017), RHE with the addition of melanocytes (Bernerd et al., 2012), RHS with epidermal and dermal layers, which represent skin equivalent (Rasmussen et al., 2010).There is ex vivo skin from hospitals and plastic surgeries, which can be used directly (Wedel et al., 2020) or transferred into the cell culture to obtain hSOC (Zhou et al., 2018).The hSOC is human skin segments completely submerged in the culture medium (Brennan et al., 2007) or cultured on an air-liquid interface with a collagen bed (Fernandez-Martos et al., 2021) or without (Fischer et al., 2013).
The known healthy skin models can be modified to achieve an appropriate skin ageing model.Since one of the extrinsic factors of ageing is exposure to UV radiation (UVA, UVB and UVC), it was applied as a first-line effect on cells in the aged skin models (Lowenau et al., 2017).Moreover, peroxides or mitomycin C (MMC) treatments, glycation of the collagen, air pollution exposure and extended time of cell culture are other possibilities for creating the in vitro aged skin model (Asselineau et al., 2015;Diekmann et al., 2016;Dos Santos et al., 2015;Kiyoshima et al., 2012;Lecas et al., 2016).Besides the different ageing agents, the models can be accomplished using three ways of exposure: the treatment of initial cells before establishing the model (Wedel et al., 2020), the processing during the growth of the model (Chen et al., 2006) and the exposure of the final skin model (Bernerd andAsselineau, 1998, 2008).

UV radiation and blue light induce skin ageing models
The development and use of radiation-induced skin ageing models represent a valuable avenue for scientific research.These models are easy to setup and provide a controlled environment to study the effects of various wavelengths of radiation on the skin, allowing researchers to investigate molecular mechanisms, cellular responses, and potential interventions related to skin ageing.Radiation primarily arises through sunlight exposure (Farage et al., 2008) but can also occur from artificial light sources as electronic devices (Mathes et al., 2014).
Solar radiation comprehends a broad electromagnetic spectrum of UV, visible and infrared radiation.Most of the UVAs rays (315-400 nm) reach the land surface at midday, whereas most of the UVBs (280-315 nm) are absorbed by the ozone layer, and UVC (100-280 nm) never attains the earth (Amaro-Ortiz et al., 2014).The damages to sun-exposed skin may vary from tanning and sunburn erythema to permanent conditions like extrinsic ageing, hyperpigmented areas, basal or squamous cell carcinoma and melanoma (Khan et al., 2018).These physiological and biological consequences occur in the epidermis and the dermis and have been associated with the influence of UV radiation involving UVA and UVB exposure (Bernerd and Asselineau, 2008).UVB radiation has more energy when compared with UVA light and can instantly provoke DNA lesions (Lee et al., 2020).However, UVA rays are distinguished by their greater penetration depth compared to UVB rays.Without proper sunscreen protection, these differences contribute to diverse effects on skin cells, often resulting in ROS formation in the tissue following exposure in all cases (Fig. 2).
At present, there is insufficient research on the influence of visible light on the skin photoaging (Cios et al., 2021).Visible light consists of the wavelength the human eye detects, ranging from 400 to 700 nm.In particular, blue light (400-500 nm) falls within the high-energy visible spectrum and accounts for a substantial proportion of visible light (Ge et al., 2023).Blue light sources comprise natural sunlight and artificial light from electronic devices and light-emitting diode fixtures that are used routinely (Coats et al., 2021a).Consequently, individuals are exposed to prolonged periods of blue light from both natural and digital sources (Kumari et al., 2023).Research indicates that blue light can trigger oxidative stress in the skin, and excessive exposure to light-emitting diode-blue light may lead to skin inflammation, skin barrier damage, and skin pigmentation issues (Coats et al., 2021a;Ge et al., 2023).So, there has been a significant increase in worries about the possible harmful impact of blue light on the skin in the past 15 years.Yet, the accurate influence of blue light on skin damage, ageing, and its underlying molecular mechanisms remains predominantly unexplored and further investigation is required (Coats et al., 2021b;Kumari et al., 2023).
Understanding the mechanisms activated by radiation exposure is crucial for both fundamental research and the development of photoprotective strategies.A crucial aspect in studying light-induced alterations involves understanding the concept of biologically efficient dose (BED), that is the minimal dose of radiation required to induce changes in cell structure and morphology, without the total dath of the tissue (Bernerd and Asselineau, 1997).The exposure to a specified BED of irradiation could lead to sunburn cells (SBC), cyclobutane pyrimidine dimers (CPD), and/ or apoptotic keratinocytes (Bernerd and Asselineau, 1998).Radiation-based models of skin ageing have been studied using in vitro cellular models, ex vivo or in vivo models.In vitro models include 2D monolayer cultures, where skin cells such as keratinocytes and fibroblasts are grown on flat surfaces (these are useful for studying cellular responses to UV radiation, but lack the complexity of actual skin), and for 3D skin equivalents, which are more advanced models that can mimic the architecture of human skin.These usually consist of multiple cell types, including keratinocytes, fibroblasts, and sometimes melanocytes, layered to form a structure similar to the epidermis and dermis.3D skin equivalents provide a more realistic environment to study the effects of UV radiation on skin.Ex vivo skin models or also denominated Human skin explants, use actual human skin samples maintained in culture.This model retains the complex structure and cell types found in vivo, making it highly relevant for studying skin aging.However, it has limitations in terms of sample availability and lifespan in culture.Animal models have mainly been developed in mice.These models allow whole organism studies, including systemic effects of UV radiation and the testing of protective interventions.Yet, there are differences between human and animal skin that need to be considered when translating findings.Upon selecting the type of model, it is also necessary to choose the UV rayswavelength, exposure and analysis time.So far, there is not a set of conditions widely accepted and applied for establishing a radiation-induced skin ageing.Various experimental conditions can be considered depending on the study's objectives.Examples of these models will be discussed and detailed in Table 1.Looking at the simplest strategy, studies investigated the effects of UVB radiation on skin monolayer models (Archambault et al., 1995;Jo et al., 2021).
Archambault's study focuses on UV irradiation's impact on melanocytes, while Jo and collaborators research examines UVB-induced skin ageing.
Both explore mechanisms that protect skin cells from UV-induced damage.The study using melanocytes monolayer-based cell model emphasizes the supportive role of keratinocytes and fibroblasts in enhancing melanocyte survival and melanin synthesis (Archambault et al., 1995), while the Human dermal fibroblasts-based model highlights a potential anti-ageing intervention through the use of Rosa gallica and its impact on the c-Raf signalling pathway (Jo et al., 2021).These findings collectively enhance our understanding of skin biology and the development of protective strategies against UV-induced damage.
Additionally, the use of 3D skin models validated these results, offering insights into both intrinsic and extrinsic protective mechanisms against UV-induced skin damage.In fact, using 3D skin models provides a more realistic environment to study the effects of UV radiation compared to traditional 2D cultures.Human primary keratinocytes were used by Tan and co-workers for both 2D and 3D organotypic epidermal models to simulate skin ageing models.Exposure to UVB radiation induced oxidative stress and photoaging (Tan et al., 2022).The study demonstrated that UVB and oxidative stress induced premature terminal differentiation and senescence in keratinocytes.In 3D models, this was also characterized by a thickened granular layer but an overall thinner epidermis.This research also evidenced the potential of nicotinamide as a protective agent against UVB and oxidative stress-induced skin ageing by maintaining cell health and function through DNA repair and metabolic support mechanisms.Bernerd and co-authors ( 2012) used an in vitro reconstructed pigmented epidermis exposed to solar-simulated radiation containing both UVA and UVB rays to assess biological alterations (Bernerd et al., 2012).This study highlighted the importance of comprehensive sun protection strategies to mitigate skin damage.UVB irradiation effect on the permeability of human RHE focusing on changes induced in keratinocytes was also evaluated.Lowenau and colleagues demonstrated that UVB irradiation led to increased permeability of the reconstructed epidermis, indicating a compromised skin barrier (Lowenau et al., 2017).The study provided insights into how UVB exposure affects skin barrier function, with implications for understanding UV-induced skin damage and developing protective measures.Primary melanocytes closely mimic the physiological conditions of human skin and were exploited by Martic (2020) to produce a photoaging model induced by UVB radiation (Martic et al., 2020).In this study, upon exposure melanocytes presented signs of cellular senescence and alteration in the pigmentation, replicating the effects of chronic sun exposure.The research revealed changes in cell morphology, cell cycle arrest, DNA damage, and oxidative stress, providing a comprehensive understanding of how UVB radiation affects melanocyte ageing and pigmentation.Detrimental effects of UVA radiation, namely on apoptosis of dermal fibroblasts, and on proteome remodelling and senescence in keratinocytes were described by Bernerd and Asselineau (1998) and Valerio (2021), respectively.These findings collectively underscore the role of UVA in inducing cell death and aging processes in different skin cell types.Bai et al. (2021) also applied a UVA radiation to induced a photoaging based on primary human dermal fibroblasts.This model exhibited characteristics such as skin roughness, dryness, and a significant reduction in dermal collagen (Bai et al., 2021).The study evidenced that rapamycin could protect skin fibroblasts from UVA-induced photoaging by enhancing autophagy and inhibiting key pathways involved in cellular stress responses and collagen degradation.
Recently, Ge and co-workers ( 2023) provided insights into the molecular pathways activated by blue light, specifically TGF-beta, JNK, and EGFR.This complements the understanding of UVA and UVB-induced ageing mechanisms by highlighting similar or additional pathways that might be involved in response to different wavelengths of light.
Girardi and co-authors ( 2024) implemented an ex vivo skin ageing model using skin biopsies cultured at an air-liquid interface and exposed to solar-simulated radiation (SSRincludes UVA and UVB radiation) to mimic natural sunlight exposure, which includes UVA and UVB radiation (Girardi et al., 2024).These exposure increased production of ROS, inflammatory markers in the skin as well as significant DNA damage, a hallmark of photoaging.This study highlights the importance of using ex vivo models for studying skin responses and evaluating potential photoprotective agents, as L-Carnosine.Table 1 summarises representative publications of UV and blue light exposure on skin models, and was structured according to the wavelengths of irradiation, starting from the lowest (UVC) and concluding with the highest (UVA and blue light).The incorporated models encompassed hSOC, RHS, pigmented RHS, and monolayers of melanocytes, fibroblasts, and keratinocytes.The diversity in the spectrum of wavelengths used for model development introduces complexity when attempting to compare results.The difficulty arises from the inherent variations in experimental conditions adopted by different laboratories.These conditions may include factors such as equipment differences, technical approaches, or environmental influences, all of which can impact the outcomes of experiments.In addition to irradiation, factors such as total BED, exposure time/periodicity, and biomarkers and biological assays also exhibit variability.Through a brief analysis of Table 1 is observed that the total BED spans from 2.5 mJ/cm 2 (Archambault et al., 1995) to 70 J/cm 2 (Bernerd et al., 2012), typically administered through single exposures.Only two reports applied total BED over a period of time with several exposures (Archambault et al., 1995;Jo et al., 2021).The most commonly used evaluation methods include histology, MMP-1, ROS, and collagen expression, but timeframes for analysis vary from hours to days and weeks, introducing complexity to results interpretation.Remarkably, many experiments overlooked a crucial parametercell viabilitywhich could impact study outcomes.The predominant focus of research efforts spins around investigating the impact of irradiation on various models to establish skin ageing models by assessing histology and biochemical markers.Only a limited number of articles employed aged skin models to explore the protective effects of sunscreens (Bernerd et al., 2012) and the anti-ageing properties of gallic acid (Jo et al., 2021).Summing up, while radiation-induced skin ageing models offer valuable insights, their optimization through standardized methodologies and inclusion of relevant parameters can further enhance the robustness and applicability of research findings in this critical area of dermatological research.Efforts to align methodologies across different laboratories would enhance the comparability and reliability of findings, ultimately advancing our understanding of radiation-induced skin ageing.

Application of air pollution components to produce skin ageing models
Environmental pollution is globally increasing every year and has a notable influence on health (Vierkotter et al., 2010).The skin's capacity to protect itself is limited, and exposure to environmental pollutants can impair its barrier function.This can result in a range of alterations including reduced skin elasticity and thickness, and changes to the structure of both epidermal barrier and dermal extracellular integrity (Ali et al., 2020;Dijkhoff et al., 2020).Air pollution includes tobacco or cigarette smoke (CS) (Prieux et al., 2020); particulate matter (PM) (Magnani et al., 2016), containing diesel PM (Nguyen et al., 2019), concentrated air particles (CAPs) (Magnani et al., 2016) and urban dust (Kim et al., 2019), ozone (O 3 ) (Muresan et al., 2018), dioxins (Fernandez-Martos et al., 2021) and heavy metals (As, Cd, Cr) (Fernandez-Martos et al., 2021) etc.An extensive review has comprehensively assessed the impact of recognized environmental pollutants on the skin, presenting insights into the utilization of scientific skin models (Fitoussi et al., 2022).The underlying mechanisms driving the impact of pollutants on the skin involve inflammation, oxidative stress, and apoptosis (Chen et al., 2016;Fitoussi et al., 2022).Thus, air pollution is considered one skin ageing factor (Chen et al., 2016;Markiewicz and Idowu, 2018).
Numerous research investigations were conducted concerning environmental pollution utilizing 3D skin models.In these studies, PM models were applied by dissolving CAPs in the media (Muresan et al., 2018).Furthermore, there are various ways to administer tobacco smoke to skin models (Fig. 3).
Muresan and co-workers (2018) evaluated how exogenous pollutant stressors affect cutaneous scavenger receptor B1 (SR-B1) levels and subsequently impair wound closure in vitro using the monolayer model with keratinocytes or fibroblasts (Muresan et al., 2018).The monolayer model was exposed to cigarette smoke, CAPs and ozone to observe their effects on SR-B1 levels and wound closure mechanisms.By using a monolayer model, the study provides findings into early molecular events in skin ageing induced by exogenous stressors, potentially  guiding strategies for skin protection and rejuvenation.However, the complexity of skin tissue architecture and cellular interactions could not be observed, and in vivo or 3D models are needed.Thus, researchers have also conducted the same study using a 3D model including both epidermal and dermal components, with keratinocytes forming the epidermis and fibroblasts providing the dermal support.It was possible to identify SR-B1 is involvement in lipid metabolism and cellular response to oxidative stress, and to monitor cell migration and proliferation across the wound area (Muresan et al., 2018).In vitro reconstructed models and hSOC showed sensitivity to tobacco smoke, ozone, diesel PM and concentrated air particle treatments.Airborne pollutants enter the deeper skin layers, initiating the activation of ROS and depleting antioxidants.They also activate the aryl hydrocarbon receptor activation, subsequently affecting matrix MMPs (Fitoussi et al., 2022).Exposure to PM increases the production of IL-1α and CYP1A1, leading to lipid oxidation and DNA alterations, inflammatory responses, and apoptotic process (Magnani et al., 2016).Cigarette smoke induces the release of pro-inflammatory cytokines, MMP-1, and MMP-3, leading to notable skin changes such as cornification and desquamation (Lecas et al., 2016).Ozone, noted for its potent oxidative properties, induces the expression of pro-inflammatory cytokines, leading to matrix damage and contributing to the ageing process of the skin (Gentili et al., 2020).Skin models are also useful tools to investigate the impact of environmental agents and the effectiveness of protectors against harmful effects of pollution, namely Feverfew or Deschampsia antarctica extract (Rasmussen et al., 2010), Camellia japonica flower extract (Kim et al.,

Table 2
Effect of the in vitro air pollution exposure on skin models.2019), and ethyl 2,4-dicarboethoxy pantothenate, a derivative of pantothenic acid (Yokota et al., 2018).For example, Deschampsia antarctica extracts exhibited protective effects against the noxious effects of common air pollutants (e.g., cigarette smoke, cadmium, chromium) on human skin.By focusing on antioxidants, anti-inflammatory agents, and barrier-supporting compounds, the study by Fernandez-Martos (2021) reveals the potential of natural extracts in skincare formulations to mitigate the adverse impacts of environmental pollutants on skin health.This research provides insights into novel strategies for developing skincare products aimed at protecting and maintaining skin integrity in urban and industrialized environments (Fernandez-Martos et al., 2021).
Establishing skin ageing models with exposure to air pollutants is a valuable, relevant, and frequently utilized approach in dermatological research.This approach can significantly contribute to the comprehension of how environmental factors influence the processes of skin ageing and which physiological mechanisms are involved in signalling pathways under different stress conditions.Table 2 presents some of the advances found in the literature.
Mimicking exposure to air pollutants in skin ageing models reflects the real-world conditions faced by people living in urban environments.This approach allows to study the specific effects of pollutants on the skin, including oxidative stress, inflammation, and structural changes.Such experimental model provides a controlled environment to explore the underlying mechanisms of pollutant-induced ageing.This includes studying the impact on collagen, elastin, and other key components of the skin's extracellular matrix, as well as understanding the role of oxidative stress and inflammatory pathways.Utilizing CAPs presents a more manageable approach in contrast to the challenges associated with the production of ozone or the acquisition of cigarette smoke fumes.However, researchers need to carefully design these skin ageing models, considering factors such as the types and concentrations of pollutants, exposure durations, post-exposure time, and potential synergies with other environmental stressors.Understanding the effects of air pollutants on skin ageing can lead to develop protective strategies.This may involve the formulation of skincare products with ingredients that counteract the impact of pollutants or lifestyle recommendations to minimize exposure.

Glycation of the collagen and its effect on reconstructed skin models
Glycation is a non-enzymatic chemical reaction of the attachment of sugar to protein or lipids with covalent bonds.Collagen glycation is an undesirable reaction, where free reducing sugars adhere to collagen and elastin, as it limits normal cellular and tissue functions, and is a confirmed factor of intrinsic ageing which is highly augmented in patients with diabetes (Hausmann et al., 2020).After the glycation, AGEs form in the dermis and act as cross-linkers between macromolecules, and may be accountable for the elasticity loss or variation of other features of the aged skin (Asselineau et al., 2015).Moreover, AGEs can be eliminated only through protein turnover and can accumulate in dermal collagen or elastin, which are noticed as long half-life proteins.Correlations between tumour formation and an elevated amount of AGEs have been reported and are related to higher cancer risk in older people (Hausmann et al., 2020).
To produce collagen glycation in vitro sodium glyoxylate or ribose are usually applied (Hausmann et al., 2020).Therefore, the creation of RHS with an aged appearance can be based on either glycated collagen or a combination involving untreated collagen.Briefly, collagen is incubated with a glycation inducer followed by extensive dialysis to remove unreacted components.Later, glycated collagen is mixed with untreated collagen in a 1:1 ratio to receive a uniform solution and to reach sufficient polymerization of the dermal layers (Pennacchi et al., 2015).The AGEs, formed during glycation influence all layers in the reconstructed skin (Fig. 4) (Pageon et al., 2008;Pennacchi et al.).In this study, aminoguanidine was employed as an antiglycation agent, known for its prominent inhibitory effects on glycation processes (Edelstein and Brownlee, 1992;Pageon et al., 2008).Aminoguanidine primarily acts by interacting with Amadori-derived fragmentation products present in the solution (Edelstein and Brownlee, 1992).Furthermore, the authors explored the anti-glycation properties of blueberry extract, rich in anthocyanins and recognized for its antioxidant activity (Bucci et al., 2018), which yielded promising findings.
Methylglyoxal is another glycation agent, a highly reactive compound implicated in the formation of AGEs (Sadowska-Bartosz et al., 2014).The methylglyoxal disturbs the redox balance of extracellular matrix-modified collagen leading to apoptosis and endoplasmic reticulum stress in human fibroblasts (Nowotny et al., 2018).Methylglyoxal also generates hydrogen peroxide and ROS, further contributing to disturbed redox balance (Nowotny et al., 2018).As a result, methylglyoxal-derived AGEs are implicated in reducing fibroblast function during the ageing process (Markiewicz et al., 2022;Nowotny et al., 2018).The glycation caused by methylglyoxal can be applied to a commercially available full-thickness skin model (Markiewicz et al., 2022).
Glycation can be combined with UVA radiation is a procedure with cumulative effects leading to the production of AGEs and elastosis in the dermal layer (Table 3) (Pageon et al., 2021).The RHS model treated with the glycated matrix can be used to study glycation and AGEs and to analyse the biological effects of anti-glycation agents such as blueberry extract, aminoguanidine etc.The changes induced by glycation imitate conditions seen in ageing skin.Overall, glycation in reconstructed skin models provides a controlled environment to investigate the specific effects of skin ageing process.The glycation of collagen has been associated with several negative effects on the skin, including increased stiffness, reduced elasticity, and altered matrix structure.Studying these effects in RHS models can provide insights into the molecular and cellular changes associated with glycation, helping researchers develop a better understanding of the underlying mechanisms of skin ageing.As RHS exhibit limitations in fully representing the complexity of native human skin, researchers should carefully design experiments, considering factors such as the choice of glycation inducers, concentrations, and exposure durations, to ensure the relevance and applicability of their findings to the natural ageing process.In some studies (Pageon et al., 2008(Pageon et al., , 2021)), glycation processes necessitate 30 days for implementation, thereby extending the overall experimental timeline.This prolonged period further contributes to the length of assays involved in the growth of RHS.

Drug-induced accelerated senescence as a tool to establish skin ageing models
The term "Drug-induced accelerated senescence" (DIAS) refers to a phenomenon first introduced by Lirija Alili, which describes cells exhibiting prolonged expression of senescence markers (Alili et al., 2014).This method involves treating fibroblasts alone, or 3D skin models or presented in ex vivo skin samples with peroxides or DNA-damaging drugs.
There are two well-described protocols for studying cellular senescence manifestations.One approach focuses on subjecting cells to constant stress factors without allowing time for cell recovery.This method has been applied to various cell types including human keratinocytes, dermal fibroblasts, and melanocytes.However, a limitation of this protocol is the inability to distinguish between immediate and permanent stress consequences (Dumont et al., 2000).The second procedure involves allowing cells to recover for 2 or 3 days after exposure to single or serial stresses.This approach enables the differentiation between immediate cell reactions such as cycle arrest and long-term consequences on cellular senescence, with human fibroblasts being primarily used in this protocol (Dumont et al., 2000).
Augmentation of intracellular ROS followed by an oxidative stress disturbance is one of the basic processes in extrinsic skin ageing.Contact with low doses of hydrogen peroxide was reported as a promising tool to study senescence-like changes in the human skin (Chen and Ames, 1994;Hahn et al., 2017), and mouse cells (Kiyoshima et al., 2012).As high concentrations of H 2 O 2 are toxic for human diploid fibroblasts, using dosages lower than 300 μM allows cell survival (Chen and Ames, 1994).
Exposure of monolayer skin models to H 2 O 2 confirmed the induction of cellular oxidative damage, with a dose-dependent decrease in DNA synthesis, and consequently, the characteristic aged profile (Chen and Ames, 1994).Besides hydrogen peroxide, tert-butyl hydroperoxide (tBHP) is also a well-known player in cellular damage and oxidative stress, partially acting by decreasing glutathione content and glutathione reductase activity (Kim et al., 1998).The tBHP reduces cellular antioxidant protection abilities, generates radicals that begin lipid peroxidation and decreases mitochondrial membrane potential in neuronal cells (Wedel et al., 2020).Dumont and colleagues reported that oxidative stress induced by tBHP exposure led human diploid fibroblast to a state close to the replicative senescence using the protocol with long-term consequences evaluation (Dumont et al., 2000).Wedel and co-workers studied the tBHP treatment and characterized the three systems: fibroblast monolayers, RHS, and skin samples after the biopsy (Wedel et al., 2020).According to the type of skin models the treatment duration varied, being for 4 days in the case of monolayers and 3 days for tissue-based models (RHS and biopsies).The effects caused by the tBHP were different in both tissue-based models with decrease of epidermal thickness observed only for RHS, while for the fibroblasts monolayer several biochemical parameters were also altered (Table 4).The effect of H 2 O 2 exposure on fibroblast monolayer models obtained with primary cells is discussed by Alili and collaborators (Alili et al., 2014).The results showed that H 2 O 2 exposure is promoting short-term senescence in fibroblasts, which is maintained for 3 days after treatment with increased ROS levels, and amplification of the cell size.After two weeks cell morphology recovered and showed no signs of ageing (Alili et al., 2014).It is noteworthy that an increased ROS level was not sufficient to maintain the cells in a prolonged senescence state.
RHS models and biopsies were also employed to investigate ageassociated changes caused by peroxide exposure.Briefly, the dermis layer in both models was immersed in tBHP solution, but the epidermis was not in contact with peroxide.The treated RHS demonstrated a thinner epidermis layer compared with the untreated control.Biopsy tissues otherwise respond to the tBHP treatment without affecting the epidermis thickness, but this layer was uneven with the presence of thin areas and thick regions.In summary, tBHP treatments thin the epidermis in models and increase the thickness variance of the epidermis in skin biopsies (Wedel et al., 2020).Consequently, RHS models demonstrated the characteristic signs of skin ageing such as constant cell damage and generation of the thinner epidermis and collagen fibres.Thus, peroxides are the potential ageing agents in the in vitro skin models for extrinsic skin research.
To imitate the DNA-damaging mechanism of skin ageing various compounds are used.Cellular ageing is a physiological reaction when the cell has contact with any DNA-damaging agent.This can be noticed in chemotherapeutic patients or victims of ionizing radiation as their skin appears older (Probin et al., 2007).There are at least three examples of drugs (MMC, busulfan and doxorubicin) used in cancer treatment that found their appliance as an ageing agent in human skin models.MMC is a well-known agent used to treat diseases related to cell growth and spreading.Two main functions of MMC have been described: as an antitumor drug due to the cross-linking of DNA, inhibiting RNA and protein synthesis; and, as an antiproliferative medication in the topical application by decreasing fibroblast activity (Chen et al., 2006).MMC is approved for use in chemotherapy combined with other drugs for disseminated adenocarcinoma, localized in the stomach or pancreas, as well as other solid tumors (Yoneda and Cross, 2010).Busulfan is a low-cost antineoplastic drug with a similar mechanism of action used to treat chronic myeloid leukaemia, although nowadays its effectiveness is inferior to new drugs (Probin et al., 2007).Besides, it is used in the therapy of specific blood disorders like polycythemia vera and myeloid metaplasia, and as a pre-treatment for bone marrow transplantation.Probin and colleagues have studied the response of human embryonic lung diploid fibroblasts after busulfan exposure (Probin et al., 2007).The study confirms that busulfan is introducing an alkyl radical that damages DNA by cross-linking DNAs with proteins.Additionally, it causes cell senescence and elevates the levels of SA-β-gal and ROS.Doxorubicin is widely used in chemotherapy due to its efficacy in fighting a wide range of cancers such as carcinomas, sarcomas and hematological cancers (Carvalho et al., 2009).The exact mechanisms of action of doxorubicin remain not fully understood.However, recent findings suggest its broad anticancer effects, encompassing DNA damage, production of ROS, induction of apoptosis, senescence, autophagy, ferroptosis, and pyroptosis, along with its role in modulating the immune system (Kciuk et al., 2023).Recently, doxorubicin was described to prompt senescent fibroblasts to generate a 3D senoskin model, which exhibited a characteristic reduction in epidermal thickness (Weinmullner et al., 2020).
The behaviour of fibroblast monolayer models after exposure to busulfan and MMC has been evaluated (Alili et al., 2014).Treatments with busulfan and MMC provoked and kept enlarged cell size, which is a characteristic marker of fibroblast senescence.ROS level was elevated on the first day of treatment and after 2 weeks became almost normal with these treatments.Nevertheless, MMC-treated cells have worse division ability in contrast to busulfan, which is characteristic of early and replicative senescence (Alili et al., 2014).
A 3D skin equivalent model prepared upon MMC treatment simulates in vivo aged skin (Diekmann et al., 2016).It was based on the patented technology EP 296078 1991 that uses collagen-glycosaminoglycan-chitosan scaffold, on which the skin fibroblasts were seeded and then, co-cultured with keratinocytes following air-liquid interface exposure.The ageing step was the pre-treatment with MMC of fibroblasts before seeding into the matrix.This comparative study of treated and untreated 3D skin models showed that aged models have decreased levels of filaggrin expression, type-1-procollagen amino-terminal propeptide (corresponding to collagen synthesis), collagen-I and collagen-III production, expression of protein Ki67 and increased levels of SA-β-gal activity, MMP-1, tumour suppressor protein p53, and ROS.From a histological point of view, treated models showed a smaller SC layer and a reduction of collagen in the aged model.Thus, MMC exposure to the skin fibroblasts resulted in mimicking ageing conditions in the 3D skin model (Diekmann et al., 2016).Moreover, this aged 3D full-thickness skin model was patented and is commercially available under the name Phenion® FT AGED Skin Model (Hausmann et al., 2020).
The DIAS approach entails the application of agents to fibroblasts or the dermis layer to simulate oxidative or DNA damage mechanisms, depending on the selected agent.From the literature review, it is apparent that this method is not extensively employed in laboratory practice.This section elucidates the diverse impacts of DIAS on various skin models, encompassing changes in cell morphology, gene expression, and cell viability.Table 4 depicts the diversity in treatment conditions and outcomes, underscoring the intricate nature of studying skin ageing in a laboratory setting.In the administration of treatments across diverse models, there seems to be inconsistency, particularly in exposure durations (e.g., 1×2 hours, 2×24 hours) and treatment periodicity.These variations pose a challenge in directly comparing results and decide on conclusive findings.Wedel and co-authors assessed the application of tBHP in three models of varying complexity, demonstrating that this repetitive treatment is conducive to establishing skin ageing models (Wedel et al., 2020).Nevertheless, the DIAS approach, despite being time-consuming, proves to be both feasible and cost-effective, bringing about significant alterations in human skin models and allowing its application across models of different complexities.In fact, one important advantage of this approach is the ability to speed up the ageing process, allowing researchers to observe and analyze cellular changes within a relatively short timeframe.This acceleration can be particularly useful for studying age-related conditions and testing anti-ageing interventions more efficiently.Yet, it is crucial to consider the relevance and accuracy of such developed models in reflecting natural ageing processes.DIAS may not precisely replicate the complex and multifaceted aspects of physiological ageing, presenting questions about the translatability of findings to the ageing human skin.Moreover, the selection of drugs, concentrations, and exposure durations requires careful consideration to ensure that the induced senescence closely mimics the characteristics of natural ageing.There may be limitations and potential side effects associated with the drugs used, impacting the model's applicability and interpretability.Overall, DIAS skin models present a promising path for research, offering a controlled and expedited means to study ageing.Nevertheless, researchers should have attention to the limitations and carefully validate the relevance of their findings to natural ageing processes to ensure the robustness and applicability of the strategy.

Prolonging culture time for investigating chronological epidermal ageing
The cell culture for long duration or replicative senescence (RS) can be applied to cells alone (keratinocytes and fibroblasts) or full-grown reconstructed skin.This aged model is based on multiple-passaged cells demonstrating characteristics of cellular ageing.Usually, the cells obtained from young donors after plastic surgery can be replicated by serial sub-cultivation until they become closer to the Hayflick limit, which refers to the number of times a normal human cell population will divide until cell division stops, and until complete depletion of the proliferative capacity.Janson and colleagues from Leiden University Medical Center (The Netherlands) have studied outcomes of serially passaged fibroblasts in monolayer models and the RHS (Janson et al., 2013).Late (passage 30) compared with early (passages 1-2) passages demonstrated in the monolayer model that fibroblasts have bigger cell size, increased SA-β-gal, p16 and myofibroblast marker α-smooth muscle actin (Janson et al., 2013).
Reconstructed skin equivalents were obtained by growing early and late passage fibroblast in the matrix and coculturing with early keratinocytes in both models.Decreased fibroblast-derived matrix, elevated expression MMP-1 and keratin 6, and slightly decreased keratin 10 are present in late fibroblast passaged RHS in contrast to RHS with early passaged fibroblast (Janson et al., 2013).Also, no significant changes are found in the epidermal layer, which could be explained by the presence of early keratinocytes (Janson et al., 2013).
Replicative senescence was also used by Lirija Alili and colleagues to make comparable models with the DIAS caused by MMC (Alili et al., 2014).The dermal fibroblasts (passages 2-15) were serially subcultured for 250 days under the required conditions to obtain replicative senescent cells (passages 55-60).The cell doubling assays show that normal human fibroblasts come to the replicative senescence stage after about 140 days and MMC decreased cell doubling significantly after exposure (Alili et al., 2014).These led to the enlargement of fibroblasts, and increased MMP-1 and SA-β-gal which are in good agreement with the previously published data (Janson et al., 2013).Additionally, ROS level, p53 and p21 activities increased significantly in the replicative senescence model (Alili et al., 2014).Dos Santos and French colleagues have investigated extending the time of in vitro 3D skin model culture (Dos Santos et al., 2015).This model was created using the glycosaminoglycan matrix where chitosan is cross-linked to collagen, and fibroblasts and keratinocytes were obtained from donors two-and one-year-old, respectively.The samples of RHS for the required experiments were collected for up to 120 days (Fig. 5).The epidermis thickness began to decrease after 42 days (Dos Santos et al., 2015).Histology images show that the remarkable reduction of the epidermal depth agreed with prolonged cell culture and correlated with a decrease in a cellular marker for proliferation (protein Ki-67).A wide range of senescent markers was analyzed and researchers show that obtained model imitates several essential aspects of ageing and could be useful to recognize new anti-ageing agents (Dos Santos et al., 2015).Based on the described procedure of 120 days in cell culture to prepare 3D human skin models, LabSkin Creations is producing and selling 3D CHRONO ageing skin models (Dos Santos et al., 2015;Kurfurst et al., 2021).
Accelerated ageing of a full-thickness skin equivalent was achieved by applying periodic mechanical stimulation, replicating the circadian rhythm for 28 days (Jeong et al., 2021).Briefly, mechanical compression stimulation that mimics circadian rhythms was administered to a 3D skin equivalent to create a flexible skin-on-a-chip ageing model.Ageing was demonstrated by reduced contraction of the full-thickness skin equivalent under mechanical stimulation, thinner epidermal layer, and elevated expression of the β-galactosidase gene.In dermal samples, collagen IV and fibronectin expression decreased upon mechanical stimulation, while epidermal filaggrin, keratin 10, involucrin, and integrin β-1 genes increased approximately 7 days earlier, a trend also reflected in protein expression.This in vitro ageing skin model can be applied in drug development for skin diseases.
A clinically relevant "speed-ageing" assay has recently been described as a simple research tool for investigating skin ageing and rejuvenation.Using a defined serum-free medium, healthy human skin of different age groups undergoes highly accelerated ageing (van Lessen et al., 2024).Several age-related biomarkers were quantitatively employed to validate this ex vivo phenotype, including reduced epidermal rete ridges and keratinocyte proliferation.Van Lessen and collaborators also demonstrated the feasibility of this model for identifying innovative anti-ageing agents directly in the human target organ.
Extended culture times can be resource-intensive, requiring more time, equipment, and maintenance.This might limit the scalability of experiments or increase the overall cost of research.In addition, longer culture times may introduce variability in experimental conditions, making it challenging to reproduce results consistently.So, standardization becomes crucial to ensure reproducibility across different studies.Of notice, is the ethical considerations when using human-derived cells or tissues.Researchers need to assure the experiment does not compromise the well-being of cell donors or participants.Overall, prior to preparing prolonged cell culture researchers must carefully balance the advantages and limitations to ensure the relevance, reproducibility, and ethical integrity of their studies.

From cellular monolayers to 3D reconstructed skin ageing modelsan overview
From the current literature review is clear that monolayers of skin cells (cell lines or primary cells) are commonly used in skin research and may contribute to preliminary studies regarding the effect of exposure, cell morphology changes, and cumulative population doublings (Chen and Ames, 1994;Dumont et al., 2000).However, lack of cell complexity and layers stratification hamper a deeper understanding of skin ageing mechanisms and anti-ageing treatment effects, motivating the use of 3D models.This could lead to an overestimation of the actual effect.For instance, when screening 451 compounds in 2D models, 46 compounds were initially identified as cytotoxic, but only seven exhibited cytotoxicity in a RHE model, and just one in a RHS model (Wei et al., 2020).
RHS are 3D model widely used in skin ageing research, as it mimics realistically layered skin and easily responds to exposure or stimuli agents.Preparation of this model is quite flexible as production can use cell lines or primary cells from juvenile or adult donors, and even pretreated cells, and or use glycation of collagen or prolonged cell culture to induce ageing phenotype (Asselineau et al., 2015).If needed, the skin structure complexity can be improved by the addition of melanocytes, which play a role in photoageing research and other cell lines.Of notice, RHS models have some limitations namely increased permeation and insufficient barrier properties as opposed to human skin (Bouwstra et al., 2021;Niehues et al., 2018;Van Gele et al., 2011).In terms of lipid composition, the reconstructed models contain all the lipids present in human skin, but the composition in triglycerides varies and the polar ceramides are under-represented in relation to the non-polar ceramides.Variance in the RHS lipid profile could be a reason for the elevated permeability (Asbill et al., 2000).When choosing RHS models few logistic constrains need to be considered as high variability among in-house batches, reagents and plastic materials irregularity, inconsistent experimental conditions to induce ageing, in-house models are time-consuming, and commercially available models expensive and with a short time period of use.
Aged skin models can also be obtained using hSOC, which is a simple, inexpensive tool with easy cell culture, but less popular than RHS in the ageing research (Zhou et al., 2018).These models are mainly used in photoageing induced by UV exposure (Brennan et al., 2007;Fischer et al., 2013;Portugal-Cohen et al., 2011), tobacco smoke exposure (Fernandez-Martos et al., 2021) and anti-ageing research (Boisnic et al., 1999).Besides, in terms of physiological relevance, hSOC has superiority to RHS in complex skin composition containing all resident cell populations and native extracellular matrix.Even highly advanced RHS with melanocytes and other cells cannot compete with hSOC in the clinical reliability (Zhou et al.).
Each ageing-promoting strategy developed so far corresponds to one of the known molecular senescence mechanisms caused by internal or external factors (Costello et al., 2022).Some strategies can be used only in RHS models, such as glycation, replicative senescence and intrinsic ageing.Also, in RHS the moment of treatment (cell line, final model or during growing) and conditions can be selected in contrast to hSOC and/or skin biopsies.Other strategies like DIAS, UV radiation, and pollution exposure are more flexible and can be used with cell lines, with RHE and RHS, hSOC and skin biopsies.This approach of introducing only one agent/stress factor allows us to investigate the occurring processes independently.But ageing is a complex process affecting internal and external factors; and, these factors are inseparable in real life.Only one research paper was found with a combination of two strategies: glycation and UVA radiation (Pageon et al., 2021).The results confirm the biological cumulative effect of both treatments, with significant AGEs formation and elastosis-like effect in the dermal compartment (Pageon et al., 2021).
In 3D models, synthetic materials like collagen, fibrin, or alginate are chosen as scaffolds for embedding fibroblasts to mimic the dermis, while keratinocytes are seeded on top to mimic the epidermis.Nonetheless, challenges such as scaffold contraction, degradation, uncontrolled growth of dermal fibroblasts, and the absence of native components from human dECM decellularized dermal extracellular matrix (dECM) are typical constraints in accurately replicating the human dermis.Recently, studies utilizing extracellular matrix (ECM) hydrogels sourced from various tissues have demonstrated the retention of native components.In particular, dECM extracted from adult human samples can provide a more realistic microenvironment in equivalent skin aging models.Fernandez-Carro and collaborators have applied human dermal dECM hydrogels as scaffolds for 3D in vitro skin aging-on-a-chip model (Fernandez-Carro et al., 2024).In human dECM, fibroblast behavior mimics that of natural adult skin, as opposed to collagen type I hydrogels where cell activity triggers biomaterial shrinkage, attributed to controlled fibroblast proliferation.These findings prompt substantial implications for advancing the development of novel in vitro skin aging models, increasing the efficacy of drug evaluation in dermatopathological research.
As observed in Fig. 6, significant differences are found among the 3D skin models, in particular in the interface between the epidermis and dermis layers, and the presence of distinct cells and cell distribution between in vitro and ex vivo models.The epidermis thickness tends to decrease with age, which is clearly shown in ex vivo samples and, also, observed after some treatments namely DIAS, pollution exposure and prolonged cell culture (Hausmann et al., 2020).

Conclusions
This literature revision illustrates the complexity of preparing an aged-skin model (cell sources, complexity in terms of stratified layers and in terms of types of cells, type of inducing ageing technique, time/ dosage/exposure periodicity of the agent, time after exposure to assess biomarkers, …), and consequently the difficulties in selecting the most appropriate experimental model.Another relevant aspect is related to the mechanism of ageing and the use of a treatment with a direct effect (e.g., cell cycle arrest stimulated by stress agents) or a condition with continuous consequences on cellular ageing (e.g., morphological changes, DNA damage, elevated oxidative stress, expression senescencerelated markers and genes).Guidelines are needed among the scientific community to allow comparisons between data obtained from different research groups.Until now, there is no approved tool by the Organisation for Economic Co-operation and Development for aged skin models.Bioengineered skin models afford a useful tool for human skin ageing research.Yet, most of the existing human skin models are obtained upon treatment or exposure to agents/conditions leading to partially mimetic ageing characteristics.Given the well-known ageing mechanisms and the strategies to obtain aged skin models developed so far, is clear the promising application of these models in anti-ageing and cosmetics research.

Declaration of Competing Interest
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests.Sofia A Costa Lima reports financial support was provided by Fundação para a Ciência e a Tecnologia

Fig. 1 .
Fig. 1. -Multiple molecular mechanisms of skin ageing are caused by internal and external factors.Adapted from (Costello et al., 2022) Copyright© 2022 The Authors.

Fig. 2 .
Fig. 2. -The schematic representation of UVB and UVA exposure effects on unprotected and sunscreen-covered RHS models.

Table 1
Radiation-induced skin ageing models.

Table 1
(continued ) (continued on next page)

Table 3
Glycation in reconstructed human skin models.

Table 4
Drug-induced accelerated senescence in skin models.