An improved human skin explant culture method for testing and assessing personal care products

Cultured human skin models have been widely used in the evaluation of dermato‐cosmetic products as alternatives to animal testing and expensive clinical testing. The most common in vitro skin culture approach is to maintain skin biopsies in an airlifted condition at the interface of the supporting culture medium and the air phase. This type of ex vivo skin explant culture is not, however, adequate for the testing of cleansing products, such as shampoos and body washes. One major deficiency is that cleansing products would not remain confined on top of the epidermis and have a high chance of running off toward the dermal side, thus compromising the experimental procedure and data interpretation.


| INTRODUC TI ON
Skin is the human body's largest organ and accounts for approximately 15% of the total body weight. 1 In particular, the skin represents a crucial barrier between the environment and the internal tissues and organs, 2 protecting the human body against external threats, such as infectious agents, chemicals, and allergens, and maintaining homeostasis and body hydration. 3 Skin care products are developed not only to promote good appearance but also to enhance the natural functions of the skin, such as blocking transepidermal water loss (TEWL), increasing hydration, and preventing flaking and itching caused by potentially harmful agents. 4 As such, over the last decades, research and development for personal care products have been focused on designing new products that can provide better cleansing outcomes while preserving essential skin barrier functions.
Despite the accelerated development of skin care products, there are only limited models suitable for the assessment of the effectiveness and safety of new active ingredients and new formulations. Personal care products, such as body washes, are usually tested in clinical studies involving the recruitment and management of volunteers. [5][6][7] These studies are costly, and only limited biological assessments can be evaluated due to ethical considerations. To overcome these problems, different experimental models have been developed and employed. For example, in vivo animal models, such as rats, mice, and pigs, have been used to test the dermal absorption of toxins and the safety of coal tar. 8,9 Nevertheless, there are fundamental differences in the structure, composition, and biological functions between the skin of animal species and human skin. Thus, the results generated using in vivo animal models often fail to translate to the actual effects on human skin. 10 Moreover, animal testing is limited or banned in the European Union, India, Israel, Norway, and Switzerland, raising the need to develop alternative methods for product evaluation if a company intends to commercialize its products in these countries.
In vitro cell-free assays can be used to evaluate the safety or to predict the effectiveness of active molecules in products by assessing their pH values, osmotic pressure, and other physicochemical changes, 11,12 but they cannot provide information revealing product interactions with skin components and properties. Various reconstructed human skin models developed by research groups and commercial sources, such as human skin constructs, [13][14][15] skinon-chip, 16,17 and more recently, three-dimensional (3D) bioprinted skin equivalents, 18 have been used in dermatology and cosmetic research. Companies, such as Sterlab and MatTek, also produce a wide range of ready-to-use reconstituted skin constructs for dermato-cosmetic research. Although these in vitro reconstructed skin models provide a reproducible biological environment for the assessment of a variety of biochemical activities and the toxicity of products, none of them was able to accurately reproduce the microanatomy of native human skin tissue. 19 In addition, unlike human skin, these models are more fragile, making the testing of topically applied products very challenging.
Ex vivo models of native human skin were introduced to resolve the deficiencies of reconstructed skin models. In 2012, Xu et al. reported a method of suturing a small piece of partial-thickness human skin on the surface of a cell strainer to study the wound healing process. 20 Human skin explants maintained at an air-liquid interface, in which culture medium only contacts the dermal side of the skin, are available for a variety of applications to evaluate the efficacy and percutaneous absorption of personal care products as well as their effects on skin infection, metabolism, immune responses, melanogenesis, and irritation. [21][22][23][24] Rush et al. used ex vivo cultured human skin to compare the absorption of the antimicrobial agent zinc pyrithione (ZnPT) in different vehicle formulations. 25 Commercially, Cutech Technology, XENOMETRIX, and some other companies supply frozen or fresh full-and split-thickness skin disc samples for pharmaceutical and cosmetic research. NativeSkin is a patented human skin model from GENOSKIN. 26 In the NativeSkin model, round human skin biopsies are embedded in a solid matrix supplemented with culture medium for cosmetic testing, which can maintain and nourish the skin alive for 7 days (https://www.genos kin. com/en/tissu e-sampl es/human -skin-model s-nativ eskin/). In almost all the existing ex vivo human skin culture models, culture medium could easily overflow onto the epidermal surface of the skin during transportation or experimental procedures since pieces of skin are set up flat at the air-liquid interface. Unwanted moisture is retained on the skin surface. Furthermore, due to the size of the skin samples, only a small amount of cleansing product can be applied to the skin surface for testing. As such, current ex vivo models are not suitable for the assessment of cleansing products, such as body wash and shampoo, because the washing and subsequent rinsing steps cannot be easily done for an extended period.
To address these challenges, we developed an improved ex vivo skin culture method, which allows us to perform a wide range of tests assessing the effectiveness and safety of cleansing products on the human skin surface in a semi-sealed system and obtain meaningful results. Furthermore, the method enables us to mimic the routine procedure of using cleansing products on the human body by applying the same wash-off protocol previously used in clinical studies.

| Establishment and evaluation of a new skin explant setup for ex vivo culture
To mimic the routine procedure of using cleansing products on the human body, namely applying them directly on the skin surface and then washing them off daily for a minimum of 2 weeks while maintaining skin viability, we tested and compared two different skin setups. As shown in Figure 1A, B, in both setups, skin explants are laid on top of the permeable membrane of trans-well inserts. In setup A ( Figure 1A on the left), each skin sample is cut in a size that is large enough to allow the peripheral edges to lay on the sidewalls of the insert, thus creating a well that can hold up to 1.5 ml of liquid, such as water or cleansing product suspension. In setup B ( Figure 1B on the right), the size of the skin explant is the same as the size of the insert membrane; therefore, it is laid flat on the insert. Both setups ensure that the dermis remains in constant contact with the culture medium underneath the insert while the epidermis is exposed to the air phase and ready for treatment. However, only setup A allows the washing procedure to be easily performed without running the product toward the dermal side and culture medium.
Although the sidewall of the insert was covered by the skin in setup A, we did not observe increased moisture accumulation on the skin surface. However, the color of the skin surrounding the sidewall appeared darker than the color of the skin that was laid directly on the membrane after 14 days in culture ( Figure 1A). The change in skin color suggested that there could be a change in skin viability because the skin overhanging on the sidewalls of the insert in setup A gradually dried out since they were not in direct contact with the culture medium. Therefore, we performed the MTT assay to assess the viability of collected skin punch biopsies. 27 Skin samples collected 1 day after culture without treatment were used as the baseline control for comparison. As shown in Figure 1C, the centerpieces collected from skin explants cultured using both setups had similar viability, which corresponded to approximately 65% of the viability of the skins used as the baseline control. As expected, the edge pieces collected from skin explants cultured using setup A showed a significant decrease in tissue viability, which was less than 35% of that of the baseline skin samples, potentially due to the lack of nutrients supplied by culture medium and dryness.
The skin tissue sections were also evaluated histologically for the integrity of skin structures by hematoxylin and eosin (H&E) staining. Figure 1D-G, the histological characteristics of the skin explants after 14 days in culture were consistent with the tissue viability assessed using the MTT assay. The disruption of normal skin anatomy, especially in the epidermis, was visible in the tissue sections of the edge piece collected from setup A, including the loss of nuclear staining in epidermal keratinocytes and increased space between the epidermis and dermis ( Figure 1F). The central pieces collected from skin explants cultured using both setups appeared to have intact and healthy anatomical conditions. Based on these observations, we decided that setup A would be our choice of skin explant culture method because it allows us to apply enough amount of cleansing products to treat the skin safely. However, only the central piece of skin explants was collected for further analysis.

| Comparative evaluation between body washes A and B in clinical testing
It is well known that body wash products can have a significant impact on skin conditions due to the effects of cleansing surfactants and skin conditioning agents. [28][29][30] For example, anionic and nonionic detergents present in the formulation are thought to have adverse effects on normal skin barrier functions and cause skin irritation and dry skin conditions. 31,32 On the other hand, some of the advanced body wash products are designed to deposit a meaningful amount of skin conditioning agents onto the skin during the cleansing process, which leads to significant improvement of dry skin conditions. To the best of our knowledge, there is no reliable ex vivo method that can assess the collective impact of the body wash product containing cleansing surfactants and skin conditioning lipids. The purpose of developing the new ex vivo skin culture method is to establish an in vitro system that could be used to evaluate the efficacy and safety of new cleansing formulations. Thus, to validate the new ex vivo skin explant method, we selected two different commercially available body wash products to treat the explants daily and compare their effects on skin appearance and hydration. Body wash A is an advanced body wash with dual lipids of petrolatum and glyceryl monooleate, and body wash B is an alternative moisturizing body wash with glycinate and soybean oil. First, to obtain a reference point, in vivo clinical testing using body wash A, body wash B, and water control was evaluated using the standard leg-controlled application test (LCAT) methodology. 33 The visual dryness result is plotted in Figure 2A, which shows that body wash A had significantly lower visual skin dryness after both 14 and 21 days of treatment than both body wash B and water. The data were further confirmed by measuring the biophysical data, including skin hydration and TEWL, using a Corneometer CM825 (Courage+Khazaka, Köln) ( Figure 2B) and a Dermalab Evaporimeter (Cortex Technologies, Hadsund) ( Figure 2C).

| The explant results correlate with the in vivo clinical data
The new ex vivo skin explant culture method (setup A) was then used to evaluate body wash A and body wash B. The main goal was to validate that the data generated using the new ex vivo culture method would be consistent with the in vivo clinical data. As shown in Figure 3 The observed appearance of skin viability reduction after 7 and 14 days of daily washes using body wash B was consistent with the data obtained using the MTT assay. To exclude the skin variance caused by the donor difference, the MTT data were normalized by the readings of the baseline skin explants for each batch. As shown in Figure 4A, no significant differences in tissue viability among groups were observed on day 7. However, the

F I G U R E 2 In vivo clinical studies on body wash A and B. (A). Skin dryness comparison by visual inspection. (B).
Skin hydration was assessed using a Corneometer CM825 (Courage+Khazaka, Köln). (C). Transepidermal water loss was measured using a Dermalab evaporimeter (Cortex Technologies). The data represent the changes after using body wash A (n = 31) and body wash B (n = 28) when compared to water-only control (n = 31) on days 1, 5, 12, and 21. Mean ± SEM. relative viability of the skin explants treated with body wash B was significantly decreased to 54.7 ± 1.3% of that of the baseline skin explants after 14 days of treatment, while the viability of the skin explants treated with body wash A was only reduced to 67.7 ± 1.2%.
Skin barrier function is essential for the maintenance of healthy and hydrated skin. In our experiments, the hydration levels of the skin explants were determined using a DermaLab single instrument with a hydration probe, which is a non-invasive method that allows the skin conductance response to be assessed in a fast and highly reproducible matter. 34 As shown in Figure 4B, the results clearly showed that skin explants treated with body wash A retained the highest degree of hydration among all treatments at both time points. Skin hydration in the body wash B group did not appear significantly different from the explants treated with water on day 7 but was significantly reduced compared with the explants in the water and body wash A groups on day 14. Overall, the data suggest that body wash B was the least effective in maintaining skin hydration.

| Biomarker evaluation correlates with skin explant viability and barrier function assessment
To further confirm the data gathered from the histological examination, the MTT assay, and hydration probe measurement, we assessed the expression of two biomarkers in the skin explants, 35 including  with water (39 ± 3/mm 2 on day 7 and 82 ± 4/mm 2 on day 14) and body wash A (40 ± 3/mm 2 on day 7 and 88 ± 4/mm 2 on day 14).
Barrier function and skin hydration are significantly influenced by the structural integrity of the stratum corneum, the outermost layer of the epidermis. The stratum corneum is formed by interlocked corneocytes containing cross-linked filaggrin, lipids, and natural moisturizing factors and helps to retain water and defend against harmful environmental agents. 38,39 Filaggrin is known to play an important role in skin barrier function. 40 As shown in Figure 5R

| CON CLUS ION
While there is an increased need for optimized cultured human skin models in dermatocosmetic research as an alternative to animal testing, current ex vivo skin culture models are not adequately set up for the testing of cleansing products with the wash-off procedure. Here, we report a new ex vivo skin culture method that is suitable for the testing of cleansing products by topical application without allowing the products to run off into the culture medium and contaminate the dermis. This new model allows for multiple endpoint analyses, including essential biomarker assessment, and provides reproducible results. Furthermore, we have demonstrated that this new skin explant method is highly useful for evaluating the safety, effectiveness, and cytotoxicity of cleansing products. In summary, our skin explant culture method constitutes a powerful in vitro system for both academic and industry-applied research.

| Skin explant culture system
All experimental procedures were performed under aseptic conditions inside a biological safety cabinet ( Figure S1A). Subcutaneous adipose tissue was removed using a scalpel and curved surgical scissor ( Figure S1B). Full-thickness skin was then immersed in povidone-iodine solution (Fisher Scientifi) for approximately 30 seconds and subsequently rinsed twice with sterile phosphate buffer solution (PBS). Next, the skin was cut into discs with a diameter of 4.5 cm using a sharp scalpel and carefully trimmed to a thickness of 2-3 mm, still retaining the full papillary dermis and partial reticular dermis ( Figure S1C,D). The skin discs were kept in PBS containing 1% antibiotic-antimycotic (Thermo Fisher) before they were assembled for explant culture.
To assemble the skin explant culture system, trimmed skin discs were placed with the dermal side facing down on top of cell culture inserts ( Figure S1E, ThinCert™ translucent polyethylene terephthalate (PET) membrane with a pore size of 8 μm) using sterile forceps. As shown in Figure S1F, the diameter of the skin disc was approximately 1 cm larger than the diameter of the insert membrane so that the skin nearly covered the sidewall of the insert.
The combined skin disc and insert were then placed into standard CELLSTAR® 6-well plates (Greiner Bio-One) ( Figure S1G,H). Skin maintenance medium was prepared using DMEM/F12 without phenol red (Thermo Fisher) supplemented with 0.4% (v/v) bovine pituitary extract (BPE), 0.5% (v/v) FBS (Thermo Fisher) 500 ng/ ml hydrocortisone, 10 ng/ml epidermal growth factor (EGF), 1 ng/ml basic fibroblast growth factor (bFGF), 1 ng/ml vitamin E, 1X insulin-transferrin-selenite (Thermo Fisher) and 1% (v/v) antibiotic-antimycotic. 1 milliliter of culture medium was added to the bottom of each well in a 6-well plate. In this assembly system, only the dermis had direct contact with the maintenance medium through the porous membrane of the insert. The skin explants were kept at 34°C in a 5% CO 2 incubator (Thermo Fisher) with reduced humidity (30%). 41 The culture medium was replaced every day throughout the experiment.

| Body wash procedure
The skin explants were first stabilized for 36 hours in the incubator to reduce the potential interference of inflammatory factors generated in the skin due to the surgical procedure. The timeline of sample treatment and collection is summarized in Figure S2A. Because skin health and viability are affected by the age and general health conditions of the donors, duplicate, or triplicate skin samples at the beginning of the treatment were collected to serve as the internal baseline reference. The skin explants were randomly separated into three groups: (1) water control (sterile tap water); (2) body wash A (ingredients: water, petrolatum, sodium trideceth sulfate, sodium chloride, cocamidopropyl betaine, trideceth-3, fragrance, guar hydroxypropyltrimonium chloride, sodium benzoate, xanthan gum, glycerate, etc.) and (3) Body wash B (ingredients: water, cocamidopropyl betaine, sodium hydroxypropyl starch phosphate, lauric acid, sodium lauroyl glycinate, sodium lauroyl isethionate, hydrogenated soybean oil, glycine soja (soybean) oil or helianthus annuus (sunflower seed) Oil, sodium chloride, glycerin, fragrance). Two time points, one after 7 days of daily washing and the other after 14 days of daily washing, were selected for skin sample collection. To closely mimic the shower routine and the procedures of previous clinical studies, we prepared a "small puff" for the skin wash procedure using the mesh of common body wash luffa ( Figure S2B). The "small puff" tool was sterilized by immersing it in 70% EtOH for at least 1 hour and then air-dried inside the biosafety cabinet under UV light exposure. Figure S2C shows the diagram of the daily in vitro body wash procedure. Briefly, skin explants assembled with the inserts were first transferred to a new 6-well culture plate without the medium.

| Skin hydration measurement
A Dermalab Single equipped with a hydration probe from Cortex Technology was used to determine the moisture levels of the stratum corneum. 42 Briefly, at the end of 7 and 14 days of treatments, skin discs were removed from the inserts and allowed to dry at room temperature for at least 30 min to eliminate any residual moisture condensation deposited on the skin surface during the tissue incubation. Measurements of skin hydration on the inner side of the arms were obtained as the reference readings. For each skin disc, at least eight measurements covering the entire center surface of the disc were taken, and the results were expressed as the average values.

| MTT skin viability assay
The MTT (3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide, Acros Organics, Geel, Belgium) assay was used to assess skin viability. 43    with VECTASHIELD antifade mounting medium with DAPI (Vector Laboratories). Images were acquired using a Nikon Eclipse 80i fluorescence microscope. The number of positively stained cells in each high-power field (40×) was counted using the particle analysis function of ImageJ software (NIH). A total of 4.5 high-power microscopic fields (40X) were combined for calculation as being equal to the size of one square millimeter.

| In situ cell death assay
TUNEL assay was used to identify apoptotic cells in the cultured skin explants using the In-Situ Cell Death Detection Kit, TMR red (Roche Applied Science) according to the manufacturer's instructions. Briefly, deparaffinized and rehydrated skin sections were first incubated in 10 μg/ml proteinase K (Thermo Fisher) in 10 mM Tris buffer at 37°C for 30 min. After washing with PBS three times, each skin section was incubated with 50 μl TUNEL reaction mixture (5 μl enzyme solution +45 μl of labeling solution) for 1 hour at 37°C in the dark. Finally, the slides were mounted with VECTASHIELD antifade mounting medium with DAPI (Vector Laboratories). Images were acquired using a Nikon Eclipse 80i fluorescence microscope.
The number of positively stained cells in each high-power field (40×) was counted using the particle analysis function of ImageJ software (NIH). A total of 4.5 high-power microscopic fields (40×) were combined for calculation as being equal to the size of one square millimeter.

| In vivo clinical testing of body wash
The standard Leg Controlled Application Test (LCAT) methodology was used to assess the performance of two body washes (A and B) through in vivo clinical testing. 33 Healthy females aged 18 to 65 years were recruited for the study during the winter season.
Subjects were given a commercial synthetic cleansing bar to use at home for 7 days as a prewash. At the end of the prewash, the skin conditions of the panelists' legs were evaluated by a qualified grader.
The dryness entrance criteria were 2.0-4.0 (0-6 scale, 0 = no dryness). The following products were applied once a day for a total of 21 days: (a) no treatment (water only) as a control; (b) body wash A (advanced body wash with dual lipids of petrolatum and glyceryl mono-oleate) and (c) Body wash B (alternative moisturizing body wash with glycinate and soybean oil). Skin hydration and TEWL were evaluated using a Corneometer CM825 (Courage+Khazaka, Köl) and a Dermalab Evaporimeter (Cortex Technologie) after one treatment, 12th treatment, and 21st treatment.

| Statistical analysis
All data were analyzed using the Prism 8 software package (GraphPad Software Inc) and expressed as the mean ± SEM. Differences were considered significant at p < 0.05. Differences between means were determined by Student's t tests and were considered statistically significant at p < 0.05.

AUTH O R CO NTR I B UTI O N S
K.W., AL.K., and Y.Z. designed the experiments; L.Z. and W. J. performed the experiments and collected the data; D.S., T.D., and D.F. provided either the skin specimens or experimental protocols, L.Z., AL.K., and Y.Z. analyzed the data; L.Z., AL.K., and Y.Z. wrote the main manuscript text, and L.Z. and Y.Z. prepared the figures; S.B., D.S., K.W., L.Z., AL.K., and Y.Z. reviewed and edited the manuscript.
All authors reviewed and approved the final version to be published.
All authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved and declare to have confidence in the integrity of the contributions of their coauthors.

ACK N OWLED G M ENTS
The authors thank Dr. W. John Kitzmiller and the surgeons, staff, and patients of the University of Cincinnati Medical Center for assistance with the collection of donated skin tissue samples.

FU N D I N G I N FO R M ATI O N
The work was supported by Procter & Gamble company (#406).

CO N FLI C T O F I NTE R E S T
The authors declare the reported work was funded by Procter & Gamble company.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing is not applicable to this article as no new data were created or analyzed in this study.

E TH I C A L A PPROVA L
Authors declare that human ethics approval was not needed for this study.

I N FO R M ED CO N S ENT
Not Applicable.