Selenium preserves keratinocyte stemness and delays senescence by maintaining epidermal adhesion

Effect of Sodium Selenite supplementation on keratinocyte viability and clonogenic potential

The effects of different concentrations of NaSe on cell viability were assessed using an MTT test on fibroblasts and keratinocytes (Figure 1A and B, respectively). Concerning fibroblasts, the first concentration that shows no significant toxicity is 0.0633 µM. The highest concentration tested (63.3 µM) is highly cytotoxic with 2% cell viability, but between 6.3 µM and 0.127 µM, cell viability ranges from 76% to 94% (***p<0.001). Concerning keratinocytes, they are more sensitive to the strong dose of NaSe, and at 63.3 µM there is no cell viability. Additionally, between 6.3 µM and 0.316 µM, the cell viability ranges from 40% to 94% (***p<0.001 and *p<0.05, respectively). Then, the effect of 0.253 µM and 0.127 µM of NaSe allows cell viability of approximately 98%, and the cytotoxic effect is not significant. One hundred percent viability is reached with the lowest dose (0.0633 µM). Interestingly, keratinocytes are also more resistant to the low dose of NaSe compared to fibroblasts.

Then, we performed a clonogenic assay with three selected concentrations of NaSe: 6.3 µM, 0.253 µM and 0.0633 µM. With a long-term exposure of low-density keratinocytes to NaSe, the results are comparable to short-term exposure. No clones are observed with 6.3 µM, confirming the toxic concentration, whereas for both concentrations (0.253 µM and 0.0633 µM), we observed a significant increase in the number of holoclones compared to the control condition (Figure 1C). The number of holoclones allows for the calculation of CFE (colony forming efficiency), clearly demonstrating that NaSe can improve the CFE ratio of keratinocytes from 1.5 to 2 times (Figure 1D **p<0.01 and *p<0.05).

Those results prompted us to explore lower concentrations of NaSe to obtain the lowest dose possible to maximize the effect on keratinocytes. We selected 8 concentrations in a range of from 0.633 µM to 1.875 nM.

Figure 2 shows the CFE results obtained after the treatment of keratinocytes from 3 independent donors with these selected concentrations. Because of the well-known inter-individual variation and strong variation of CFE, the results are expressed as the ratio of the treated keratinocyte CFE to the control keratinocyte CFE. The ratio is significantly increased in the range from 0.253 μM to 30 nM. Regardless of the donor, the concentration of 30 nM NaSe leads to the highest CFE ratio of 1.563 ± 0.255 (***p<0.001).

Since this concentration improves the capacities of keratinocytes to give rise to holoclones, 30 nM NaSe was selected for constant supplementation of the culture medium for all the following experiments.

Sodium selenite supplementation delays the replicative senescence

The proliferative capacity of primary keratinocytes measured via subculture of the cumulative population doubling (PD; Table in Figure 3) demonstrates in all cases that NaSe supplementation extended the replicative lifespan of human keratinocytes. Regardless of the donor, the cumulative PD that leads to holoclones appears to always increase by 156.074 ± 14.597% after 30 nM NaSe supplementation. NaSe treatment supports the presence of holoclones at least until passage 8 despite the donor (corresponding to 88.106 ± 15.118 cumulative PD), whereas in the control group, the holoclones are absent since passage 4 (corresponding to 57.666 ± 15.489 cumulative PD). Associated with the proliferative phenotype of the non-supplemented culture, the morphological aspects of the colonies are dramatically changed, as shown by a loss of smooth shape and a decrease in size, and are composed of differentiated keratinocytes with senescence morphology (Figure 3B and C, left panel).

Sodium selenite supplementation activates adhesion to type IV collagen and laminin 332 of human primary keratinocytes

Figure 4 demonstrates that NaSe supplementation improves keratinocyte adhesion during the replicative life span. Regardless of the donor, the passage and the adhesion markers, type IV collagen and laminin 332, which are ligands of β1 integrin and α6 integrin, respectively, the percentage of adhered keratinocytes is significantly increased from 110% to 200% when they are supplemented with NaSe compared to control keratinocytes. Indeed, the increase in adhesion tends to be higher in terms of laminin 332 (157.4% ± 22.9) than in terms of type IV collagen (133.1% ± 12.4). Finally, the adhesion seems to be maximal in the first passages (p1 and p2) and decreases as the subculture continues.

Sodium selenite supplementation results in delayed skin equivalent senescence

The histological images of SEs cultured for 45 and 60 days clearly demonstrate that the thickness of the epithelium decreases as the time of culture without treatment increases. A long-term application (17 and 32 days) of 30 nM NaSe supports an SE with a thicker epidermis compared to the control with preservation of terminal differentiation and a decrease in the stratum corneum (Figure 5). At day 45, we can observe a significant increase (108%; *p<0.05) in the epidermis thickness, calculated as 200 µm for the treated SE versus 185 µm for the control SE, and at day 60 we observe an increase of 121% (**p<0.01) in the epidermis thickness, calculated as 112 µm for the treated SE versus 92 µm for the control SE. Moreover, the thickness decreases occur with the time of culture in both the treated and control groups.

Regardless of the time of culture (45 or 60 days), NaSe supplementation allows for the maintenance of SE homeostasis. Therefore, to clarify, Figure 5 D, E, F and G presents the representative results at day 60, which is the time when the senescence is higher. Figure 5D demonstrates a significant increase of 3.2 times in the number of basal keratinocytes expressing the proliferating marker Ki67, which is responsible of epithelium thickness, after NaSe treatment. If we consider the expression of cytokeratin 15 (Figure 5E), a marker of KSCs, the distribution appears to be conserved in the basal layer in both groups. However, the relative intensity of labeling is increased by 2 time in the treated SE compared to the control SE (200 RU and 100 RU, respectively; *p<0.05). These data confirm that NaSe is able to maintain the proliferative capacity of KSCs to ensure the renewal of the epidermis and the maintenance of its thickness during skin aging.

Concerning the late differentiation marker (Figure 5F), the filaggrin expressing area in living epidermis (define as all the layers of the epidermis excepted for the stratum corneum) is 2.5 times higher in the treated SE versus the control SE (approximately 4.6% versus 11.4%, respectively; *p<0.05). These results demonstrate that NaSe preserves the homeostasis of the epidermis with a balance between proliferation and differentiation, with a thicker living epithelium.

Finally, in our SE a p16INK4a is detected in both nuclear and cytoplasmic part of the cells (Figure 5G). NaSe decreases senescence due to long-term culture because number of cells expressing p16INK4a, a senescence marker, is decreased by 2 times in the treated SE compared to the control SE (***p<0.01). All these data allow us to confirm our hypothesis that NaSe treatment delays SE senescence.

Focusing on adhesion markers, NaSe supplementation during SE development induces the increase of β1 and α6 integrin patterns (Figure 6). The α6 integrin expression appears to be increased by 1.6 times in the treated SE than in the control SE (180.3 ± 65.8 and 101.9 ± 31.5, respectively; **p<0.01). In the same manner, β1 integrin expression is increased by 1.8 times in the treated SE compared to the control SE (376.4 ± 202.2 and 225.9 ± 94.3, respectively; *p<0.05).

Tissue harvest

Normal human skin tissue explants were obtained from the surgical discard of anonymous healthy patients with informed consent in accordance with ethical guidelines and declared to the French research ministry (Declaration no. DC-2008-162 delivered to the Cell and Tissue Bank of Hospices Civils de Lyon).

Fibroblast and Keratinocyte cultures

Fibroblasts were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal calf serum (FCS), antibiotics (20 mg/mL gentamicin (Panpharma, Fougères, France), 100 IU/mL penicillin (Panpharma), and 1 mg/mL amphotericin B (Panpharma). Keratinocytes were grown on a feeder layer of irradiated human fibroblasts pre-seeded at 4000 cell/cm², as previously described [23,43], in a mixture of 3:1 DMEM and Ham’s F12 (Invitrogen, Carlsbad, USA), respectively, supplemented with 10% FCS, 10 ng/mL epidermal growth factor (EGF; R&D systems, Minneapolis, MN, USA), 0.12 IU/mL insulin (Lilly, Saint-Cloud, France), 0.4 mg/mL hydrocortisone (UpJohn, St Quentin en Yvelines, France), 5 mg/mL triiodo-L-thyronine (Sigma, St Quentin Fallavier, France), 24.3 mg/mL adenine (Sigma), isoproterenol (Isuprel, Hospira France, Meudon, France) and antibiotics as above. For each cell type, the medium was changed every 2 days until confluence was reached. At confluency, cells were re-suspended with trypsin-EDTA 0.05% (Thermo Fisher Scientific, Waltham, MA USA) and used for the in vitro 3D model or the colony forming unit and toxicity assays.

Cytotoxicity assay

Keratinocytes or fibroblasts were seeded in 6 -well plates and grown until 80% confluency. Then, NaSe, provide by Labcatal Pharmaceuticals as pharmaceutical liquid solution in sterile water, was diluted in keratinocyte or fibroblast cultures at a concentration ranging from 63.3 µM to 0.0633 µM for 72 h. After incubation, a solution of tetrazolium dye MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) in phosphate buffered saline (PBS) at 1 mg/ml was added to the supernatant of the cells for 2 h. The formazan precipitate was solubilized using dimethyl sulfoxide and transferred to 96-well plates to measure absorbance at 560 nm using a spectrophotometer. For each concentration, 3 wells and 12 measurements per well were obtained for 3 independent donors. Viability is calculated as the “(mean of treated cells–mean of blank) / (mean of untreated cells–mean of blank)” ±SD (n=3 donors and n=12 measurements per conditions). The stars illustrate statistical differences between untreated condition (100%) and mean of experimental points.

Clonogenic assay

Colony forming efficiency (CFE) was evaluated with a large-scale treatment of low-dose NaSe at concentrations ranging from 6.3 µM to 1.875 nM. Keratinocytes were seeded at clonal density, 10 to 20 cells/cm², onto a feeder layer and cultivated for 10 to 14 days. Three flasks were fixed and colored using rhodamine B (Sigma) at 0.01 g/mL in 4% paraformaldehyde for 30 min. The CFE was calculated according to Barrandon et al. [43] as CFE = CFU (meroclones + holoclones number) × 100)/(cellular seeding density). In our manuscript the term “holoclone” was used to refer to large (≥5 mm) and homogeneous clones with regular and smooth contours whereas meroclones are intermediate sized, heterogeneous and display contours. For the populating doubling calculation, cells from 3 flasks were re-suspended by a trypsin treatment and used for cell numeration. Population doubling was determined as PD = ln (number of harvested cells/number of cells seeded)/ln2 (with ln for natural logarithm).

For the analysis of the extended life span, cells from each donor were separated into two groups: treated (KT) or control (KC). The supplementations began at the seeding of the primary culture (just after keratinocytes extraction) and persisted throughout along passages. CFU/CFE was performed at each passage both for KC and KT. The experiments were performed on six independent donors: 3 for the low-dose scale of NaSe and 3 for the extended life span (n=6 flasks for each condition).

Adhesion assay

Because KSCs are characterized by their strong adhesion to the ECM in “niches” through β1 and α6 integrins [28], we tested the keratinocyte adhesion to their respective ligands, laminin 332 and type IV collagen, the main components of DEJs [44]. For that, multi-well tissue culture plates (Costar, Dutscher, Brumath, France) were coated with the indicated concentrations of laminin 332 [45] or collagen IV (Thermo Fisher Scientific, Waltham, MA, USA) substrates by overnight adsorption at 4°C. Cells were detached by incubation with trypsin and rinsed in serum-free medium prior to the experiment. After saturating the wells with 1% BSA, the plates were used immediately for cell adhesion assays (30 000 cells/well) in serum-free medium containing selenium or not at the concentration used for cell culture. The same amount of cells was also plated on poly-D-lysine coated plates (Biocoat Cell Environments, 6 wells per condition) to evaluate the maximal cell attachment capacity as previously described [46]. After 30 min to 1 h, non-adherent cells were washed with PBS, and the extent of adhesion was determined after fixation of adherent cells, followed by staining with 0.1% crystal violet and absorbance measurements at 570 nm. A blank value corresponding to BSA-coated wells was subtracted. In all experiments, each assay point was derived from triplicate measurements (three wells per assay point) and each adhesion score was normalized to the mean of adhesion obtained from poly-D-lysine coated plates.

In vitro 3D human skin equivalent (SE) culture

SE cultures were prepared as described previously [23]; fibroblasts from young donors were seeded at a final density of 250 000 cells/cm² onto a dermal substrate made of chitosan-cross-linked collagen-glycosaminoglycan matrix [47]. This SE was grown in fibroblast medium supplemented with 50 mg/mL L-ascorbic acid (Sigma) and 10 ng/ml of EGF at 37°C in a 5% CO2 atmosphere, and the medium was changed every day for 21 days. For the preparation of SE from keratinocytes, keratinocytes from young donor were seeded onto the SE on day 21. These submerged SEs were cultured for 7 days in keratinocyte medium and then raised at the air-liquid interface and cultured in a simplified keratinocyte medium containing DMEM supplemented with 10% FCS, 10 ng/mL EGF, 0.12 IU/mL insulin, 0.4 mg/mL hydrocortisone, and antibiotics for 17 days or 32 days. The NaSe was added extemporaneously to the culture medium for a final concentration of 30 nM. Samples were systematically harvested after 45 and 60 days of total cell culture and were immediately fixed in neutral buffered formalin 4% (Alphapath, Mudaison, France) for 24 h then embedded in paraffin or directly in OCT compound (Euromedex, Strasbourg, France). For each cell culture condition and analysis, SEs were produced in triplicate.

Histology and immunohistological (IH) analyses

Paraffin-embedded formalin-fixed samples were cut into 5-μm sections. After dewaxing and rehydration, sections were stained with hematoxylin phloxine saffron (HPS staining). For immunohistochemistry, after heat-mediated antigen retrieval treatment, tissue sections were incubated in 5% H₂O₂/3% normal goat serum (NGS; Jackson Immunoresearch, Suffolk, UK) to inactivate endogenous peroxidases. Non-specific binding was blocked in PBS containing 4% bovine serum albumin (BSA, Sigma, St Quentin Fallavier, France) and 5% NGS. Sections were then incubated with the following primary antibodies (Table1) diluted in PBS/4% BSA/5% NGS overnight at 4°C. After incubation for 1 h with a peroxidase-conjugated secondary antibody (EnVision, Dakocytomation, Glostrup, Denmark), the antigen was detected with 3,3-diaminobenzidine tetrahydrochloride as the substrate. Tissue sections were subsequently counterstained using Harris hematoxylin (25%, Sigma). For immunofluorescence, labeling was performed on air-dried 5-μm cryosections. Air-dried cryosections were incubated with monoclonal-mouse primary antibodies overnight at 4°C. Then, sections were incubated with secondary Alexa 488- or 568-conjugated anti-mouse or anti-rabbit antibodies (Molecular Probes, Invitrogen, Carlsbad, USA) for 1 h at room temperature. Nuclear counterstaining using Hoechst stain was carried out according to a routine protocol. Image acquisition was performed using an Eclipse 50i microscope (Nikon, Champigny sur Marne, France) for light microscopy and a Zeiss LSM 510 confocal laser scanning microscope for fluorescence.

Antibodies used and operating conditions are displayed in Table 1.

Image analysis and processing

Image processing and analysis were performed using the software Image J with the MBF plugin for microscopy (http://www.macbiophotonics.ca/imagej/, Research Service Branch, US National Institute of Health, United States).

The living epidermal thickness, consider as all layers from basal layer to stratum granulosum excepted the stratum corneum, was measured as the distance between the basement membrane to the stratum corneum and expressed in µm on at least 20 fields per sample. Ki67 and P16INK4a epidermal positive cells were automatically separated from the background and, counted. Data are expressed in number of positive cells per epidermal field.

The results are express as the surface area of filaggrin, and the relative expression for cytokeratin 15, integrin α6 and β1. Positively stained-tissue areas were automatically detected and segmented from other pixels and automatically measured. Data were normalized by the area of living epidermis for filaggrin expression, and by the length of basement membrane for cytokeratin 15, integrin α6 and β1. Measurements are performed on at least 3 different fields of 3 different SE per experimental condition.

Statistical analysis

For all data, the statistical significance was assessed using Mann Whitney test with the software GraphPad Prism 4 (GraphPad Software Inc., La Jolla, CA, USA), and statistically significant differences are indicated by asterisks as follows: *p<0.05, **p<0.01 and ***p<0.001 (or otherwise indicated on the figure, depending on the experiment).