Identification of cell cycle-associated and -unassociated regulators for expression of a hepatocellular carcinoma oncogene cyclin-dependent kinase inhibitor 3.

Human cyclin-dependent kinase inhibitor 3 (CDKN3) is a known oncogene in hepatocellular carcinoma (HCC) and its expression is promoted during tumor development. CDKN3 serves as a cell cycle regulator and its dysregulation is considered to be a causal factor for tumor progression. However, the molecular basis of the regulation of CDKN3 expression remains largely elusive. Using in silico approach, we identified CDKN3SE, a super enhancer (SE), and enhancer RNA (eRNA) candidates transcribed from this SE. Among the eRNA candidates, the expression of CDKN3eRNA was detected in the human HCC model cell line HepG2, and was found to facilitate the expression of CDKN3 without affecting the cell proliferation rate. In silico screening revealed two DNA-binding transcription factors, upstream stimulatory factor (USF) 1 and 2, involved in the regulation of CDKN3eRNA expression on CDKN3SE. A knock-down of USF1/USF2 expression in the HepG2 cells did not affect CDKN3eRNA expression, while the expression of CDKN3 was down-regulated. In a USF2 dominant negative HepG2 cell line generated by genome editing, a drastically altered cell shape and lowered cell proliferation rate were found; however, the expression of CDKN3eRNA appeared unaffected. Thus, the present study illustrated two regulators for CDKN3 expression: USF2, as a cell cycle-associated protein regulator, and CDKN3eRNA, as a cell cycle-unassociated RNA regulator.

A long non-coding RNA as a direct vitamin D target transcribed from the antisense strand of the human HSD17B2 locus.

Vitamin D (VD) exerts a wide variety of actions via gene regulation mediated by the nuclear vitamin D receptor (VDR) under physiological and pathological settings. However, the known target genes of VDR appear unlikely to account for all VD actions. We used in silico and transcriptomic approaches in human cell lines to search for non-coding RNAs transcriptionally regulated by VD directly. Four long non-coding RNAs (lncRNAs), but no microRNAs (miRNAs), were found, supported by the presence of consensus VDR-binding motifs in the coding regions. One of these lncRNAs (AS-HSD17ß2) is transcribed from the antisense strand of the HSD17ß2 locus, which is also a direct VD target. AS-HSD17ß2 attenuated HSD17ß2 expression. Thus, AS-HSD17ß2 represents a direct lncRNA target of VD.

Antagonistic action of a synthetic androgen ligand mediated by chromatin remodeling in a human prostate cancer cell line.

The clinical use of androgen receptor (AR) antagonists has been successful in treating prostate cancer patients, inducing remission of androgen-dependent tumors. However, a couple of years after treatment, prostate tumors transition into an androgen-independent state with altered gene expression profiles, but the molecular basis is not understood. Since the AR antagonists trigger this transition, we assessed whether AR antagonists induce chromatin reorganization in an androgen-dependent prostate cancer cell line (LNCaP). Treatment of LNCaP cells with two clinically used AR antagonists (bicalutamide [Bic] and enzalutamide [Enz]) expectedly resulted in antagonistic effects on cell proliferation, AR transactivation, and dihydrotestosterone (DHT)-induced expression of AR target genes. Thus, the antagonists expectedly acted to antagonize the transactivation function of AR activated by androgen binding. By ChIP-qPCR assay, AR bound to Bic, but not Enz, was recruited to an endogenous consensus AR-binding site within the kallikrein-related peptidase 3 gene promoter after treatment with Bic, similar to the effect of DHT. By ATAC-seq analysis of the cells after long-term treatment for 5 days, Bic and dihydrotestosterone DHT induced different chromatin reorganization patterns and gene expression profiles, suggesting that Bic exhibited a distinct action from that by DHT. Thus, these results suggest that the action of a known AR antagonist is mediated by chromatin reorganization in a prostate cancer cell line.

Androgen-dependent and DNA-binding-independent association of androgen receptor with chromatic regions coding androgen-induced noncoding RNAs.

Androgen induces the binding of its receptor (AR) to androgen-responsive elements (AREs), while genome-wide studies showed that most androgen-induced AR binding sites on chromatin were unrelated to AREs. Enhancer RNAs (eRNAs), a class of noncoding RNAs (ncRNAs), are transcribed from superenhancers (SEs) and trigger the formation of large ribonucleoprotein condensates of transcription factors. By in silico search, an SE is found to be located on the locus of KLK3 that encodes prostate specific antigen. On the KLK3 SE, androgen-induced expression of ncRNAs was detected and designated as KLK3eRNAs in LNCaP cells, and androgen-induced association of AR and FOXA1 on the KLK3eRNA coding regions was detected. Such androgen-induced association of an AR mutant lacking DNA binding activity on the KLK3eRNA coding regions was undetectable on an exogenous ARE. Thus, the present findings suggest a molecular basis of androgen-induced association of AR with chromatin on ARE-unrelated sequences.

Transcriptional Regulation of 25-Hydroxyvitamin D-24-Hydroxylase (CYP24A1) by Calcemic Factors in Keratinocytes.

CYP24A1 regulates serum vitamin D (VD) levels by inactivating 25(OH)2D3, which is the precursor of the active form of VD [1α,25(OH)2D3], and CYP24A1 expression is controlled by multiple calcemic factors such as 1α,25(OH)2D3, calcium, and phosphate. A major phosphaturic factor, FGF23, has also been identified as a regulator of serum VD levels by affecting renal CYP24A1 gene expression; however, its effect on CYP24A1 in extrarenal cells remains largely unstudied. Therefore, the direct effect of FGF23 on CYP24A1 was examined in a human keratinocyte cell line (HaCaT). In this cell line, significant induction of CYP24A1 gene expression by 1α,25(OH)2D3 was seen within 4 h by qRT-PCR, and this was mediated by the VD receptor, as shown in a mutant cell line genetically deficient in this receptor. However, FGF23 treatment up to 12 h did not induce CYP24A1 expression, although the expected activation of the downstream MAPK signaling pathway was seen. High calcium and phosphate treatments were also ineffective in inducing CYP24A1 gene expression. Furthermore, a luciferase assay showed no activation of a VD-sensitive proximal CYP24A1 promoter in response to the calcium and phosphate treatments, suggesting that the effect of FGF23 on dermal CYP24A1 gene expression is indirect. From these findings, we speculate that CYP24A1 gene regulation by FGF23 occurs mainly in renal cells, but not in extrarenal cells, at least not in keratinocytes.

A parallel permeability assay of peptides across artificial membranes and cell monolayers using a fluorogenic reaction.

Here, we report a facile permeability assay to quantitatively evaluate the membrane permeability of multiple peptides in parallel. With a fluorogenic click reaction between azidocoumarin and a terminal alkyne tag introduced on a peptide, the peptide that crossed an artificial membrane or a cell monolayer was quantitatively detected. The method allows a rapid measurement of the permeability of multiple compounds on a plate reader even in the presence of a complex mixture of biological molecules.