The Role of EAAT4 in Epidermal Differentiation and Calcium Homeostasis during Aging

A calcium concentration gradient across the layers of the epidermis is essential for keratinocyte differentiation, barrier formation, hydration, skin pH, cell adhesion and microbial defense (Rinnerthaler and Richter, 2018Rinnerthaler M. Richter K. The Influence of Calcium on the Skin pH and Epidermal Barrier During Aging.Curr Probl Dermatol. 2018; 54: 79-86Crossref PubMed Scopus (12) Google Scholar, Tu and Bikle, 2013Tu C.L. Bikle D.D. Role of the calcium-sensing receptor in calcium regulation of epidermal differentiation and function.Best Pract Res Cl En. 2013; 27: 415-427Crossref PubMed Scopus (41) Google Scholar). This gradient is altered during aging, with the result that calcium is almost equally distributed across all epidermal layers (Denda et al., 2003Denda M. Tomitaka A. Akamatsu H. Matsunaga K. Altered distribution of calcium in facial epidermis of aged adults.J Invest Dermatol. 2003; 121: 1557-1558Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar, Rinnerthaler et al., 2013Rinnerthaler M. Duschl J. Steinbacher P. Salzmann M. Bischof J. Schuller M. et al.Age-related changes in the composition of the cornified envelope in human skin.Experimental Dermatology. 2013; 22: 329-335Crossref PubMed Scopus (56) Google Scholar). In a previous age specific transcriptomic analysis of human foreskins (Breitenbach et al., 2015Breitenbach J.S. Rinnerthaler M. Trost A. Weber M. Klausegger A. Gruber C. et al.Transcriptome and ultrastructural changes in dystrophic Epidermolysis bullosa resemble skin aging.Aging (Albany NY). 2015; 7: 389-411Crossref PubMed Scopus (28) Google Scholar) we identified genes that are involved in calcium homeostasis of keratinocytes and were either up- or down-regulated with age. In the present study, the corresponding genes were reevaluated by qPCR: Foreskin samples of five young males (1 to 7 years of age), five middle-aged (18 to 39) and five old individuals (61 to 74) were analyzed. The expression of sixteen of the 86 regulated genes analyzed was reduced during aging (supplementary information 1). The most down-regulated transporter gene showed a significant 19-fold decrease during aging when comparing young to old skin and a 9-fold down-regulation in old skin compared to middle-aged skin. This gene termed SLC1A6 (Solute Carrier Family 1 Member 6) encodes the sodium-dependent glutamate/aspartate transporter EAAT4 (Excitatory Amino Acid Transporter 4) (Fairman et al., 1995Fairman W.A. Vandenberg R.J. Arriza J.L. Kavanaugh M.P. Amara S.G. An excitatory amino-acid transporter with properties of a ligand-gated chloride channel.Nature. 1995; 375: 599-603Crossref PubMed Scopus (1019) Google Scholar). Although many types of calcium channels have been described that mediate the calcium flux within keratinocytes (Lee and Lee, 2018Lee S.E. Lee S.H. Skin Barrier and Calcium.Ann Dermatol. 2018; 30: 265-275Crossref PubMed Scopus (62) Google Scholar) none of genes encoding these channel proteins were down regulated to the same extent as SLC1A6 during aging. Similar to our results, SLC1A6 has also been shown to be one of the most strongly repressed genes in actinic lentigines (Warrick et al., 2022Warrick E. Duval C. Nouveau S. Piffaut V. Bourreau E. Bastien P. et al.Actinic lentigines from Japanese and European volunteers share similar impaired biological functions.J Dermatol Sci. 2022; 107: 8-16Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar). The down-regulation of SLC1A6 could be linked to the altered calcium gradient since SLC1A6 was shown to cotransport glutamate together with sodium in the brain (Vandenberg and Ryan, 2013Vandenberg R.J. Ryan R.M. Mechanisms of glutamate transport.Physiol Rev. 2013; 93: 1621-1657Crossref PubMed Scopus (240) Google Scholar). The sodium is then available for a sodium/calcium exchanger that contributes to the build-up of the calcium gradient. The validity of this model in the skin is supported by our results. The combination of 0.5 mM CaCl2 with glutamate/sodium bicarbonate increases differentiation in primary keratinocytes compared to 0.5mM CaCl2 alone as shown by a qPCR analysis. The age dependent keratinocyte differentiation proteins LOR (loricrin) (Rinnerthaler et al., 2013Rinnerthaler M. Duschl J. Steinbacher P. Salzmann M. Bischof J. Schuller M. et al.Age-related changes in the composition of the cornified envelope in human skin.Experimental Dermatology. 2013; 22: 329-335Crossref PubMed Scopus (56) Google Scholar) and CALML5 (Calmodulin Like 5) (Mehul et al., 2001Mehul B. Bernard D. Schmidt R. Calmodulin-like skin protein: A new marker of keratinocyte differentiation.Journal of Investigative Dermatology. 2001; 116: 905-909Abstract Full Text Full Text PDF PubMed Google Scholar) were used as markers (Figure S1). The addition of 0.5 mM CaCl2 led to a more than an 8-fold (Mean: 8.3 +/- SD: 3.3; N=7; p=0.0471; unpaired t-test) increase in SLC1A6 mRNA levels. Confirming our data, it was recently shown that a combined treatment of cells with CaCl2 and glutamine boosts the expression of the differentiation markers IVL (involucrin) and KRT1 (Jang, 2018Jang H.H. Effects of Acetyl Glutamine on the Improvement of Skin Barrier and Inhibition of Senescence.Asian J Beauty Cosmetol. 2018; 16: 579-589Crossref Google Scholar). To demonstrate the involvement of EAAT4 in epidermal differentiation, eyelid skin explants were incubated with two different EAAT4 inhibitors: (+/-)-threo-3-methylgluatmic acid (T3MG) and ZnCl2. T3MG specifically blocks EAAT2 as well as EAAT4 (Eliasof et al., 2001Eliasof S. McIlvain H.B. Petroski R.E. Foster A.C. Dunlop J. Pharmacological characterization of threo-3-methylglutamic acid with excitatory amino acid transporters in native and recombinant systems.J Neurochem. 2001; 77: 550-557Crossref PubMed Scopus (24) Google Scholar), whereas ZnCl2 interferes with the anion conductance of most glutamate transporters (Mitrovic et al., 2001Mitrovic A.D. Plesko F. Vandenberg R.J. Zn2+ inhibits the anion conductance of the glutamate transporter EAAT4.J Biol Chem. 2001; 276: 26071-26076Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar). Skin explants (N=5) were maintained for five days with either CaCl2 alone or CaCl2 and one EAAT4 inhibitor before qPCR analysis was performed. T3MG especially had profound effects on the expression of several keratinocyte specific differentiation markers (Table S1). In immortalized human keratinocytes, the observed effects upon addition of glutamate and EAAT4 inhibitors were less pronounced. In addition, the supplementation of cells with glutamate induced only moderately and partially significantly the expression of differentiation genes (Figure S2). To address the question of whether EAAT4 affects differentiation directly or indirectly by influencing calcium homeostasis, the calcium gradient in the epidermis was investigated: After 5 days in culture in the presence of calcium and EAAT4 inhibitors (in various combinations), eyelid skin explants were cut and stained with the calcium specific dye Calcium Green-1 AM. Both inhibitors showed a clear impact on the epidermal calcium gradient, especially on the calcium peak in the stratum granulosum (Figure 1a). The effect of T3MG was much more severe. Using ImageJ, the fluorescence intensity in the different epidermal layers (stratum spinosum and stratum granulosum) was measured and normalized to stratum basale. In all samples the relative brightness was highest in the stratum granulosum. Treatment of skin explants with 0.5 mM CaCl2 resulted in a much steeper gradient that flattens after addition of either ZnCl2 or T3MG (Figure 1b). To confirm the specificity of the EAAT4 inhibitors and their effect on keratinocyte differentiation, a “Reconstructed Human Epidermis” (RHE) was created. This RHE model established a calcium gradient, with a CaCl2 dependent expression of SLC1A6 and expressed age-specific epidermal differentiation markers FLG (filaggrin) and KRT2 in the outermost layers. A knockdown of SLC1A6 was achieved by lentiviral particles (MISSION® pLKO.1-puro; Merck) and was confirmed by qPCR analysis. The observed SLC1A6 knockdown was accompanied by a significant reduction of both LOR and FLG expression levels (Figure 1c). The SLC1A6 knockdown itself did not lead to an abnormal morphology (Figure S3a and S3b) but had a strong influence on the establishment of the epidermal calcium gradient (Figure S3c-e). Furthermore, IHC (immunohistochemistry) staining of RHEs showed a clear reduction of FLG, KRT2 and LOR (Figure S4) after lentiviral mediated SLC1A6 knockdown. Skin biopsies obtained after blepharoplasty (6 middle-aged women (30-50 years) and 7 elderly women (64-70 years)) were analyzed by IHC for the presence of EAAT4. Representative images of a middle aged as well as an old individual clearly show EAAT4 expression and its decrease with age (Figure 2). EAAT4 also resides in stratum granulosum, the most calcium-rich region of the epidermis indicating an involvement of this transporter in keratinocyte differentiation. This result is in accordance with a recent publication that showed of all five human sodium-dependent glutamate/aspartate transporters (EAAT1-5) only SLC1A3/EAAT1 (localized to stratum basale) and SLC1A6 (in stratum granulsoum) was expressed in skin (Jara et al., 2021Jara C.P. de Andrade Berti B. Mendes N.F. Engel D.F. Zanesco A.M. Pereira de Souza G.F. et al.Glutamic acid promotes hair growth in mice.Sci Rep. 2021; 1115453Crossref Scopus (10) Google Scholar). In summary, we confirmed by qPCR and histology array studies that EAAT4 expression was most altered during aging. Keratinocyte cultures, RHEs and ex vivo studies suggest that EAAT4 was involved in calcium responses and establishment of the epidermal calcium gradient. We propose a model in which the loss of EAAT4 with age is of particular importance in the stratum granulosum. Decreased EAAT4 expression leads to reduced import of glutamate and sodium and directly affects the Na+/Ca+ exchange and as a consequence the epidermal calcium gradient is altered. This is also associated with a reduced keratinocyte differentiation and a reduced skin barrier function as it is actually observed during the aging process. Microarray Datasets related to this article are available at https://www.aging-us.com/article/100755. The use of skin samples obtained from routine surgery for scientific project was approved by the ethic committee of the “Land Salzburg” (415-EP/73/548-2015). All included patients (in case of children their legal representatives/parents) gave their written, informed consent. Conceptualization: KR, MR, AD; Formal analysis: MR, AD. Writing-original draft: MR; Writing – review & editing: AD, KR; Investigation: JB, GR, VW, MKS, AD, MR; Methodology: JB; Funding acquisition: MR, AD; Visualization: MR. The authors state no conflict of interest. We want to thank Episkin Lyon for making Reconstructed Epidermis, Christine Collin for culturing and preparing modified keratinocytes and helpful criticisms, Philippe Bastien for helpful criticisms on statistics and Mark Donovan for the critical revision of the manuscript. Albert Duranton and Mark Rinnerthaler are co-corresponding authors for this paper. We are grateful to the Austrian Science Fund FWF for the grant P33511 to MR. Download .xlsx (.03 MB) Help with xlsx files Supplementary information 1: qPCR analysis of genes involved in calcium transport during aging. Five young foreskin samples (age range: 1 to 7 years of age), five middle-aged foreskin samples (age range:18 to 39 years) and five old foreskins (age range: 61 to 74 years) were compared. Old skins were either normalized to young or middle-ages samples. As a reference gene the acidic ribosomal protein ARP (RPL0) was chosen. The fold-changes were calculated using the formula 2−(ΔΔCt). The Student t-test was used for the calculation of the p-values.Figure S2Glutamate import and keratinocyte differentiation. The expression levels of the keratinocyte specific differentiation markers LOR, IVL und FLG were analysed via qPCR upon addition of either 0.09 mM CaCl2, 0.09 mM CaCl2 and 2 mM glutamate/sodium bicarbonate, 0.09 mM CaCl2 and 2 mM glutamate/sodium bicarbonate, 0.09 mM CaCl2 and 200 μM T3MG, 0.09 mM CaCl2 and 2 mM glutamate/sodium bicarbonate and 200 μM T3MG, 0.2 mM CaCl2 or 0.2 mM CaCl2 and 200 μM T3MG in immortalized human keratinocytes. As a reference gene RPL0 was chosen. All data were normalized to the expression values of either LOR (a and c), IVL (d and f) or FLG (g and i) upon addition of minimal CaCl2 levels (0.09 mM). In (a-c) data for LOR are presented, in (d-f) data for IVL and in (g-i) data for FLG. In (a), (d) and (f) data are presented for 0.09 mM CaCl2, in (c), (f) and (i) data after addition of 0.2 mM CaCl2. Statistics for (a), (d) and (g) was calculated using Kruskal Wallis followed by post-hoc Mann Whitney U and the Bonferroni correction method yielding p-values of 0.08018 (loricrin), 0.02547 (IVL) and 0.5647 (FLG) (N=4). The results of the post-hoc analysis is presented in (b) loricrin, (e) IVL and (h) FLG. Statistics for (c), (f) and (i) was calculated using an unpaired t-test. N=4; ***: p<0.0001. Because in all experiments performed, SLC1A6 inhibition/knockdown has a bigger effect on keratinocyte differentiation than glutamate, it cannot be excluded that EAAT4 fulfils its role independently of glutamate transport.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure S3SLC1A6 knockdown in RHEs. Hematoxylin and eosin staining of representative RHEs transduced with either pLKO.1-puro (a) or (b) pLKO.1-SLC1A6-shRNA. In (c) and (d) RHEs are shown after a Calcium Green-1 AM staining and a cultivation in the presence of 0.5 mM CaCl2. In (c) the RHE is transduced with pLKO.1-puro and in (d) and (e) with pLKO.1--SLC1A6-shRNA. The SLC1A6 knockdown strongly affects the establishment of the epidermal calcium gradient, albeit to varying degrees. Arrows point to the membranes on which the RHEs were cultivated. Scale bar represents 10 μm.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure S4Differentiation markers in RHEs. RHEs either transduced with the control vector (pLKO.1-puro) or a shRNA knockdown construct (pLKO.1-shRNA) were sectioned using a cryomicrotome and subjected to an IHC analysis. In (A) an anti-filaggrin antibody (CSB-PA235658; Cusabio) was used, in (B) an anti-KRT2-antibody (Progen; 65177) and in (c) an anti-LOR antibody (Ozyme; PRB-145P). Each skin section was counterstained with Hoechst. In the shRNA panels a decrease of the differentiation markers LOR, FLG and KRT2 are observable. Scale bar: 10 μm.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Tabled 1Table S1: EAAT4 inhibition and keratinocyte differentiation. After surgery skin explants (aged 48-53) were cultivated for 5 days in KGMTM-GOLD (Lonza) supplemented with either (1) 0.5 mM calcium, (2) 0.5 mM CaCl2 and 200 μM T3MG or (3) 0.5 mM CaCl2 and 200 μM ZnCl2. A qPCR analysis was performed using RPL0 as a reference gene. T3MG and ZnCl2 treatments were normalized with samples cultured with CaCl2 only. N=5; One-way ANOVA.GeneInhibitorRegulationFold changep-valueCASP14T3MGdown4.10.01ZnCl2down1.70.12CALML5T3MGdown6.60.01ZnCl2down1.150.41BLMHT3MGdown2.90.00ZnCl2down1.10.67AQP3T3MGdown2.00.01ZnCl2down1.10.89 Open table in a new tab