Silencing Lysine-Specific Histone Demethylase 1 (LSD1) Causes Increased HP1-Positive Chromatin, Stimulation of DNA Repair Processes, and Dysregulation of Proliferation by Chk1 Phosphorylation in Human Endothelial Cells

Abstract

: The methylation of histone lysine residues modifies chromatin conformation and regulates the expression of genes implicated in cell metabolism. Lysine-specific demethylase 1 (LSD1) is a flavin-dependent monoamine oxidase that can demethylate mono- and dimethylated histone lysines 4 and 9 (H3K4 and H3K9). The removal of methyl groups from the lysine residues of histone and non-histone proteins was found to be an important regulatory factor of cell proliferation. However, its role has not been fully elucidated. In this study, we assessed LSD1-mediated cell cycle progression using a human endothelial cell model. The short hairpin RNA knockdown of LSD1 inhibits the G₂/M phase of cell cycle progression by checkpoint kinase 1 (Chk1) phosphorylation (S137). We observed elevated DNA damage, which was consistent with the increased detection of double-strand breaks as well as purines and pyrimidines oxidation, which accompanied the activation of ATR/ATRIP signaling by H2AXS139 phosphorylation. The irreversible pharmacological inhibition of LSD1 by 2-phenylcyclopropylamine (2-PCPA) inactivated its enzymatic activity, causing significant changes in heterochromatin and euchromatin conformation assessed by chromatin assembly factor 1 subunit A (CAF1A) and heterochromatin protein 1 isoform α and γ (HP1α/γ) immunofluorescence analysis. We conclude that the knockdown of LSD1 in endothelial cells leads to increased HP1-positive chromatin, the stimulation of DNA repair processes, and the dysregulation of proliferation machinery.

Keywords: Chk1; DNA damage and repair; LSD1; cell cycle; chromatin remodeling; histone posttranslational modifications.

Figure 1

Effect of pharmacological and transcriptional inhibition of lysine-specific demethylase 1 (LSD1) on cell cycle progression of human microvascular endothelial cells (HMEC-1). (A) The effect of LSD1 shutdown due to 2-phenylcyclopropylamine (2-PCPA) treatment or shRNA transfection on the mitotic index, the induction of aberrations in the M phase of cell cycle and apoptosis (MI, a percentage of mitotic cells in relation to all cells, both interphase and mitotic). The percentages were calculated based on 3000 cells per repeat; (B,D) Flow cytometry analysis of cell cycle. Cells were stained with propidium iodide. The obtained histograms were analyzed using FlowJo software 10.4.1. (“A” stands for the subpopulation of apoptotic cells). Presented histograms are representative of three independent experiments; (C,E) Gene expression profile of cell cycle regulators in HMEC-1 treated with 100 µM of 2-PCPA (E) or after transcriptional LSD1 silencing by specific shRNA (C). Presented data are average of three independent experiments. The level of significance was determined at * p < 0.05, ANOVA and post hoc analysis by Tukey’s test.

Figure 2

Consequences of LSD1 inhibition/silencing in human microvascular endothelial cells on checkpoint kinase 1 (Chk1) activation. Immunofluorescent analysis of the Chk1 level (A,B) and the Chk1 activation (ph-Chk1(S317) level (A,C). Cells were stained with anti-Chk1 or anti-ph-Chk1(S317) antibodies conjugated with AlexaFluor 488 fluorescent dye. The labeling index was calculated as the ratio of immunofluorescence-labeled cells to all cells in an HMEC-1 cell population. The intensity of fluorescence was visualized in ImageJ program (B a6, a7, b6, b7; C a6, a7, b6, b7). The presented pictures are representative of three independent experiments. (D), (E) Analysis at the protein level of the expression of Chk1 and ph-Chk1 in (D) pharmacological and (E) transcriptional experimental models of LSD1 activity inhibition. Scale bars are equal to 10 µm. Statistical significance was analyzed at * p < 0.05 using ANOVA and post hoc analysis by Tukey’s test.

Figure 3

Effect of pharmacological and transcriptional inhibition of lysine-specific demethylase 1 (LSD1) on the ‘topography’ of heterochromatin fraction. Immunofluorescent analysis of HP1 alpha and gamma localization (A–E) and CAF1A protein (F,G). The labeling index was calculated as the ratio of immunofluorescence-labeled cells to all cells in an HMEC-1 cell population. Subfigures D and E show exemplary nuclei from the series “not-treated” (nonT) versus “treated” (LSD1 KDs) for respectively: HP1 alpha protein and HP1 gamma, and enroll the most characteristic location of specific foci, labeling heterochromatin regions at the periphery of the interphase nuclei (double red star), which is usually associated with the nuclear lamina, and perinucleolar heterochromatin (single red star). The red arrows indicate the unlabeled region (the so-called “hole”) in the location of the nucleoli. The protein expression of HP1 aplha, HP1 gamma, and CAF1A by Western blotting in cells treated with 100 µM of 2-PCPA (H), and (I) shRNA-transfected HMEC-1. Presented images are representative, and the data are the average of three independent experiments. Scale bars are equal to 10 µm. * p < 0.05, ANOVA and post hoc analysis by Tukey’s test.

Figure 4

The tail moment (TM) of oxidatively damaged DNA in the nonT and LSD1 KDs HMEC-1 evaluated by alkaline and modified comet assay using lesion-specific enzyme endo III (endonuclease III) and FPG (formamidopyrimidine glycosylase) to differentiate DNA double-strand breaks (cells untreated) from oxidized purines (+ endo III) and pirymidines (+ Fpg). (A), (B) All the values are expressed as mean ± SD (the number of cells used per repeat, n = 100). The level of significance was determined at *** p < 0.001 and ** p < 0.01 nonT vs. LSD1 KDs (ANOVA and post hoc analysis by Tukey’s test).

Figure 5

Level of DNA damage markers: H2AXS139ph and ATRIP after shutting down LSD1 activity. (A–D) immunofluorescent localization of the antigens, (E,F) protein level of markers by Western blotting. Presented images are representative, and the data are the average of three independent experiments. The labeling index was calculated as the ratio of immunofluorescence-labeled cells to all cells in an HMEC-1 cell population. Scale bars are equal to 10 µm. * p < 0.05, ANOVA and post hoc analysis by Tukey’s test.

Scheme 1

Effect of pharmacological and transcriptional inhibition of lysine-specific demethylase 1 (LSD1) on induction of the cell cycle and DNA damage response (DDR) pathway in microvascular endothelial cells (HMEC-1).