Regulation of the Glycerol Transporter, Aquaporin-3, by Histone Deacetylase-3 and p53 in Keratinocytes  Vivek Choudhary, Lawrence O. Olala, Karen Kagha, Zhi-qiang Pan, Xunsheng Chen, Rong Yang, Abigail Cline, Inas Helwa, Lauren Marshall, Ismail Kaddour-Djebbar, Meghan E. McGee-Lawrence, Wendy B. Bollag  Journal of Investigative Dermatology  Volume 137, Issue 9, Pages 1935-1944 (September 2017) DOI: 10.1016/j.jid.2017.04.031 Copyright © 2017 The Authors Terms and Conditions

Figure 1 The pan-HDAC inhibitor, SAHA, dose- and time-dependently increases AQP3 expression in primary mouse epidermal keratinocytes. Mouse keratinocytes were treated with vehicle (DMSO) or SAHA for 24 hours or as indicated. (a) Representative Western blot showing AQP3 and GAPDH levels. The right panel shows quantification and cumulative values (mean ± standard error of the mean, n = 3). (b) Quantitative real-time reverse transcriptase–PCR analysis for AQP3 from at least three independent experiments was performed using the ΔΔCt method with GAPDH as the endogenous control; results represent the mean ± standard error of the mean of groups presented relative to the control value. (c) A representative (n = 3) Western blot for AQP3 levels in keratinocytes treated with SAHA for the indicated time points. (d) Keratinocytes were treated with 5 μmol/L SAHA or DMSO for 24 hours. Cells were fixed and immunostained using antibodies recognizing AQP3 (red) and β-actin (green) with nuclei counterstained with DAPI (blue). Scale bar = 10 μm; ∗∗P < 0.01, ∗∗∗P < 0.001 versus control. AQP, aquaporin; g, glycosylated; GAPDH, gylceraldehyde-3-phosphate dehydrogenase; HDAC, histone deacetylase; M, mol/L; ng, non-glycosylated; SAHA, suberoylanilide hydroxamic acid. Journal of Investigative Dermatology 2017 137, 1935-1944DOI: (10.1016/j.jid.2017.04.031) Copyright © 2017 The Authors Terms and Conditions

Figure 2 SAHA increases AQP3 levels in situ and AQP3 activity in vitro. (a) Neonatal mouse skin incubated with medium containing DMSO or SAHA for 24 hours. A representative (n = 3) Western blot for AQP3 levels is shown (left panel). In the right panel, AQP3 levels were normalized to β-actin and expressed as fold change over the DMSO-treated group; cumulative results from at least three separate skins using DMSO or 5 μmol/L SAHA are presented as mean ± standard error of the mean. ∗∗P < 0.01 versus the DMSO group. (b, c) Keratinocytes from (b) wild-type and (c) AQP3-knockout mice were treated with DMSO or 5 μmol/L SAHA for 24 hours. AQP3 functionality was assessed by [3H]glycerol uptake assay. Please note the change in x-axis (panel b versus c). The data represent mean ± standard error of the mean from three independent experiments. ∗∗∗P < 0.001 versus DMSO-treated keratinocytes. AQP, aquaporin; g, glycosylated; KC, keratinocyte; M, mol/L; ng, non-glycosylated; ns, not significant; SAHA, suberoylanilide hydroxamic acid. Journal of Investigative Dermatology 2017 137, 1935-1944DOI: (10.1016/j.jid.2017.04.031) Copyright © 2017 The Authors Terms and Conditions

Figure 3 HDAC3 regulates AQP3 levels. (a) Mouse or (b) human KCs were treated for 24 hours with vehicle (DMSO) or HDAC inhibitors targeting different HDACs: 5 μmol/L SAHA, 300 nmol/L TSA, 50 μmol/L TC-H106 (Tocris, Bristol, UK), 50 μmol/L SBHA (Tocris), 50 μmol/L CI994 (Selleck Chemicals, Houston, TX), 50 μmol/L PCI 34051, 50 μmol/L droxinostat (Selleck Chemicals), or 50 μmol/L CAY10398 as indicated. Glycosylated (g) and non-glycosylated (ng) AQP3 protein levels were analyzed by Western blotting. Blots shown are representative of three separate experiments. (c) Primary mouse keratinocytes were treated for 24 hours with the HDAC3 inhibitor RGFP966 or the control vehicle (DMSO) and analyzed by Western blotting for levels of AQP3 compared with the loading control, GAPDH. Results shown are representative of three separate experiments. (d) Primary mouse keratinocytes grown on cover slips were treated as above. At the end of 24 hours of treatment, cells were fixed and immunostained using an antibody-recognizing AQP3 (red), with nuclei counterstained with DAPI (blue). Scale bar = 20 μm. AQP, aquaporin; CI994, tacedinaline or 4-(acetylamino)-N-(2-aminophenyl)benzamide; g, glycosylated; GAPDH, gylceraldehyde-3-phosphate dehydrogenase; KC, keratinocyte; M, mol/L; ng, non-glycosylated; PCI-34051, N-hydroxy-1-[(4-methoxyphenyl)methyl]-1H-indole-6-carboxamide; SAHA, suberoylanilide hydroxamic acid; SBHA, suberoyl bis-hydroxamic acid; TC-H106, N1-(2-aminophenyl)-N7-(4-methylphenyl)heptanediamide; TSA, trichostatin A. Journal of Investigative Dermatology 2017 137, 1935-1944DOI: (10.1016/j.jid.2017.04.031) Copyright © 2017 The Authors Terms and Conditions

Figure 4 HDAC3 knockdown increases and HDAC3 overexpression decreases AQP3 expression. Keratinocytes from neonatal floxed HDAC3 mice were infected with Cre-recombinase (CRE)- or control (GFP)-expressing adenovirus for 24 hours. 57 hours after infection, the cells were treated with DMSO or 1 μmol/L SAHA for 15 hours. (a) Representative Western blots (upper panel) and quantitation (lower panels), with cumulative values expressed relative to the maximum response as mean ± standard error of the mean from three separate experiments. (b) Keratinocytes (wild-type) were infected with adenovirus expressing GFP (control), HDAC3, or HA-tagged HDAC1 for 12 hours and treated with DMSO or 2.5 μmol/L SAHA for 10 hours. Representative Western blots (upper panel) and quantitation of AQP3 levels (lower panel; n = 3) as in a are shown. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001 versus the control (GFP); fP < 0.05, ##P < 0.01, #P < 0.05 versus the indicated groups. AQP, aquaporin; g, glycosylated; GAPDH, gylceraldehyde-3-phosphate dehydrogenase; GFP, green fluorescent protein; HA, hemagglutinin; HDAC, histone deacetylase; ng, non-glycosylated; SAHA, suberoylanilide hydroxamic acid. Journal of Investigative Dermatology 2017 137, 1935-1944DOI: (10.1016/j.jid.2017.04.031) Copyright © 2017 The Authors Terms and Conditions

Figure 5 p53 mediates SAHA-increased AQP3 levels in keratinocytes. (a) Mouse keratinocytes were treated with DMSO or 5 μmol/L SAHA in the presence and absence of 30 μmol/L PFT for 24 hours. Representative Western blot for AQP3 (left panel) and quantitation of 3 experiments (right panel) are shown. AQP3 levels normalized to GAPDH and expressed relative to the maximal response are presented as mean ± standard error of the mean. ∗∗∗P < 0.001 versus DMSO and fffP < 0.001 as indicated. (b, c) Mouse keratinocytes were infected with adenovirus expressing wild-type p53 or vector for 12 hours and treated with 5 μmol/L SAHA for 12 hours. (b) AQP3 mRNA expression analyzed by quantitative PCR and (c) protein levels by Western blots (upper panel) with quantitation of three experiments (lower panel) are shown. Results presented as mean ± standard error of the mean are expressed relative to the maximal response. ∗∗P < 0.01 versus vector, ffP < 0.01 as indicated. AQP, aquaporin; g, glycosylated; GAPDH, gylceraldehyde-3-phosphate dehydrogenase; ng, non-glycosylated; PFT, pifithrin; SAHA, suberoylanilide hydroxamic acid; wt, wild type. Journal of Investigative Dermatology 2017 137, 1935-1944DOI: (10.1016/j.jid.2017.04.031) Copyright © 2017 The Authors Terms and Conditions