Long Non-Coding RNA KCNQ1OT1 Regulates Protein Kinase CK2 Via MiR-760 in Senescence And Calorie Restriction

Keywords: KCNQ1OT1; long non-coding RNA; senescence; calorie restriction; protein kinase CK2; miR-760 

Click Here To Get More Info About Cistanche Anti-Aging Effects

1. Introduction 

Long non-coding RNAs (lncRNAs) are transcripts with a size greater than 200 nucleotides that are not translated into proteins. LncRNAs are responsible for diverse functions, including transcriptional regulation in the cis or trans, organization of nuclear domains, and interaction with microRNAs (miRNAs) [1–3]. In particular, understanding the interaction network of lncRNAs and miRNAs provides a new perspective on the regulatory mechanisms of genes. The lncRNA KCNQ1OT1, the full name of which is KCNQ1 overlapping transcript 1, is a 91 kb transcript that is transcribed by RNA polymerase II in an antisense orientation, relative to KCNQ1 [4,5]. KCNQ1OT1 is present in intron 10 of the KCNQ1 gene. The KCNQ1 locus is located on the short arm of human chromosome 11 (11p15.5). Through recruiting G9a and the H3K27 histone methyltransferase PRC2 (polycomb repressive complex 2), KCNQ1OT1 interacts with chromatin to form a complex folding structure and then silences multiple target genes [6]. Additionally, KCNQ1OT1 is involved in several disorders, including heart disease, cerebral ischemic stroke, atherosclerosis, pyroptosis, and various cancers (gastric, ovarian, and colorectal) by binding to various miRNAs (for example, miR-760, miR-200a, miR-452-3p, miR-320a, miR-701-3p, and miR-2054), all of which regulate the expression of their target genes [7–13]. Nevertheless, the biological roles of KCNQ1OT1 in senescence and calorie restriction (CR) remain unclear. 

Cellular senescence is the terminal arrest of proliferation, stimulated by several cellular stresses such as telomere shortening, oncogenic activation, and oxidative stress [14,15]. Previous studies identified the protein kinase CK2 (CK2), composed of two catalytic (α and/or α 0 ) subunits and two regulatory β subunits, as a senescence regulator. CK2 inhibition triggers the expression of several senescence markers, including senescence-associated β-galactosidase (SA-β-gal) activity [16], p53–p21Cip1/WAF1 axis activation [17], reactive oxygen species (ROS) production [18], senescence-associated heterochromatin foci (SAHF) formation [19], and senescence-associated secretory phenotype (SASP) expression [20]. miR-186, miR-216b, miR-337-3p, and miR-760 promote cellular senescence by inhibiting CK2α [21,22]. Calorie restriction (CR), consisting of a chronic reduction in total calorie intake without malnutrition, is the most successful strategy to delay cellular senescence [23]. It has recently been reported that CK2 is upregulated by CR and induces autophagy [24]. However, the molecular mechanism underlying CR-mediated CK2 upregulation remains unclear. 

In this study, the potential role of KCNQ1OT1 in senescence and CR was assessed. KCNQ1OT1 knockdown promoted a senescence phenotype via downregulating CK2α in human cancer MCF-7 and HCT116 cells and lung fibroblasts IMR-90 cells. This research indicates that KCNQ1OT1 upregulates CK2α expression through interaction with miR-760 during senescence and CR, suggesting that KCNQ1OT1 may be a novel therapeutic target for aging-associated diseases. 

2. Results 

2.1. KCNQ1OT1 Knockdown Induced Activation of SA-β-gal Staining, the p53-p21Cip1/WAF1 Pathway, and H3K9 Trimethylation Via CK2α Silencing in Human Cancer Cells 

MCF-7 and HCT116 cells were transfected with KCNQ1OT1 siRNA to investigate the involvement of KCNQ1OT1 in senescence. KCNQ1OT1 knockdown upregulated SA-β-gal activity (Figure 1A). Additionally, immunoblot results indicated that KCNQ1OT1 knockdown upregulated the levels of p53 and p21Cip1/WAF1, and hallmarks of SAHF (increased H3K9 trimethylation (H3K9me3) and decreased H3K9 acetylation (H3K9Ac)) (Figure 1B). It was previously reported that CK2 downregulation induces these senescence markers (activation of SA-β-gal staining, the p53-p21Cip1/WAF1 pathway, and SAHF) [16,17,19]. Therefore, it was examined whether KCNQ1OT1 interacts with CK2. Interestingly, ectopic CK2α expression abrogated the induction of SA-β-gal activity, p53, p21Cip1/WAF1, and H3K9me3, mediated by KCNQ1OT1 downregulation (Figure 1A,B). Furthermore, KCNQ1OT1 knockdown reduced the protein level of CK2α (Figure 1B). These results collectively suggest that KCNQ1OT1 knockdown induces cellular senescence by downregulating CK2. 

2.2. KCNQ1OT1 Knockdown Induced SASP Factor Expression and ROS Generation Via CK2α Silencing in Human Cancer Cells

It was investigated whether KCNQ1OT1 downregulation increased the expression of SASP factors due to reports that senescent cells secrete pro-inflammatory factors [14,15]. KCNQ1OT1 knockdown induced the expression of SASP factors, including interleukin (IL)-1β, IL-6, and matrix metalloproteinase (MMP) 3 (Figure 2A). KCNQ1OT1 downregulation increased the amount of intracellular ROS because oxidative stress is a major cause of senescence [14,15]. For this purpose, HCT116 and MCF-7 cells were transfected with KCNQ1OT1 siRNA and stained with CM-H2DCFDA. KCNQ1OT1 knockdown increased ROS production, as indicated by the right shift in FL fluorescence during flflow cytometry (Figure 2B). It was previously reported that CK2 downregulation induces SASP expression [20] and ROS generation [18]. Ectopic expression of CK2α abrogated the induction of SASP factor expression and ROS generation, mediated by KCNQ1OT1 downregulation (Figure 2A,B). Collectively, these results indicate that KCNQ1OT1 knockdown induces ROS generation and inflflammation through downregulating CK2.

Figure 1. KCNQ10T1 knockdown induced activation of SA--gal staining, the p53-p21CipI/WAFpathway, and H3K9 trimethylation via CK2x silencing in human cancer cells. HCT116 and MCF-7 cells were transfected with KCNO10T1 siRNA for two days in the absence or presence of pcDNA3.1.HA-CK2x. (A) Cells were stained with 5-bromo-4-chloro-3-indolyl--D-galactoside, and representative images were obtained at 20x magnification (upper). Scale bar = 100 um. Representative data from three independent experiments are shown. The graphs represent the percentage of blue-stained cells (bottom). (B) Immunoblotting was used to determine the level of each protein using specific antibodies (upper). 3-Actin was used as a control, The graphs represent the quantitation of each protein relative to B-actin (bottom). Data are reported as mean  SEM.* p < 0.05; ** p < 0.01, ** p < 0.001.H3K9me3, histone H3 Lys9 trimethylation; H3K9AC, H3 Lys9 acetylation.

Figure 2. KCNQ10T1 knockdown induced senescence-associated secretory phenotype (SASP) factor expression and reactive oxygen species (ROS) generation via CK2x silencing in human cancer cellsHCT116 and MCF-7 cells were transfected with KCNO10T1 siRNA for two days in the absence or presence of pcDNA3.1-HA-CK2a. (A) The level of each mRNA was determined by reverse transcription-polymerase chain reaction (RT-PCR) using specific primers (upper). Representative data from three independent experiments are shown. The graphs represent the quantitation of each mRNA relative to 6-actin (bottom). (B) The cells were incubated with 10 UM CM-HDCEDAFluorescence intensity was determined by flow cytometry analysis (upper). Representative data from three independent experiments are shown. The graphs show the relative fluorescence level (bottom)Data are reported as mean 士 SEM. * p < 0.05; ** p < 0.01; *** p < 0.001.

2.3. KCNQ1OT1 Was Involved in Lipopolysaccharide (LPS)-Mediated SASP Factor Expression Via Silencing CK2α in Human Cancer Cells

Because treatment with LPS causes cellular senescence by downregulating CK2α [20], it was tested whether LPS downregulated KCNQ1OT1 expression. Treatment with LPS (6 µg/µL) reduced the transcript levels of CK2α and KCNQ1OT1 in human cancer cells (Figure 3A). Furthermore, the effect of KCNQ1OT1 on SASP factor expression in cells treated with LPS was investigated. However, treatment with LPS (6 µg/µL) increased SASP factor (IL-1β, IL-6, and MMP3) expression in human cancer cells; additional treatment with pcDNA3.1-KCNQ1OT1 (36,181–37,140) abrogated the LPS-mediated induction of SASP factors (Figure 3B). It was previously shown that the concerted action of miR-760, miR- 186, miR-337-3p, and miR-216b stimulated premature senescence through silencing the CK2α protein in HCT116 cells [21], and that miR-760 and miR-186 were upregulated in replicative senescent IMR-90 cells [22]. The expression patterns of these miRNAs affected by treatment with LPD were determined. Quantitative real-time polymerase chain reaction (qPCR) analysis revealed that the amount of miR-760 increased by more than 200% in both HCT116 and MCF-7 cells treated with LPS (6 µg/µL), in comparison with the control cells. miR-186 was not upregulated by LPS treatment, and miR-337-3p and miR-216b were differently regulated in these cells (Figure 3C). Thus, these results collectively indicate that LPS increased miR-760 amounts via downregulating KCNQ1OT1, resulting in CK2α downregulation-mediated senescence.

Figure 3. KCNO10T1 was involved in lipopolysaccharide (LPS)-mediated senescence-associated secretory phenotype (SASP) factor expression via silencing CK2x in human cancer cells. (A) HCT116 and MCF-7 cells were treated with LPS (6 ug/uL) for two days. The level of each mRNA was determined by RI-PCR using specific primers (top). Representative data from three independent experiments are shown. B-Actin was used as a control. The graphs represent the quantitation of each mRNA relative to B-actin (bottom). (B) Cells were treated with LPS (6 ug/uL) in the absence or presence of pcDNA3.1-KCNO10T1 (36,181-37,140) for two days. The level of each mRNA was determined by RI-PCR using specific primers (left). Representative data from three independent experiments are shown that 3-Actin was used as a control. The graphs represent the quantitation of each mRNA relative to -actin (right). (C) Cells were treated with LPS (6 ug/uL) for two days total RNA was isolated from cells and subjected to PCR analysis to determine the relative levels ofmiR-760, miR-186, miR-337-3p, and miR-216b using RNU48 for normalization. Data are shown as themeans 士 SEM.*p < 0.05; ** p < 0.01; *** p < 0.001.

2.4. KCNQ1OT1 Upregulated CK2α by Sponging miR-760 in Human Cancer Cells 

Next, it was examined whether KCNQ1OT1 regulates CK2α expression via miR-760. HCT116 and MCF-7 cells were transfected with pcDNA3.1-KCNQ1OT1 (36,181–37,140) or KCNQ1OT1 siRNA, along with a miR-760 mimic or inhibitor (sequences shown in Supplementary Table S1). Ectopic expression of KCNQ1OT1 (36,181–37,140) increased the mRNA level of CK2α, whereas additional treatment with the miR-760 mimic suppressed KCNQ1OT1-mediated CK2α upregulation (Figure 4A). In contrast, KCNQ1OT1 knockdown reduced the mRNA level of CK2α, whereas additional treatment with a miR-760 inhibitor suppressed KCNQ1OT1 knockdown-mediated CK2α downregulation (Figure 4B). Both the miR-760 mimic and inhibitor could not change the amount of KCNQ1OT1, indicating that miR-760 was not an upstream regulator of KCNQ1OT1. Altogether, these results indicate that KCNQ1OT1 increases the amount of CK2α mRNA by sponging miR-760. Supplementary Figure S1 shows the sequences and binding sites of KCNQ1OT1, miR-760, and CK2α mRNA, determined using TargetScan and Miranda. 

Figure 4. KCNQ10T1 upregulated CK2x by sponging miR-760 in human cancer cells. (A) HCT116 and MCF-7 cells were transfected with pcDNA3.1-KCNO10T1 (36,181-37,140) for two days in the absence or presence of miR-760. (B) HCT116 and MCF-7 cells were transfected with KCNO10T1 siRNA for two days in the absence or presence of a miR-760 inhibitor. The level of each mRNA was determined by RT-PCR using specific primers (top). Representative data from three independent experiments are shown. The graphs represent the quantitation of each mRNA relative to B-actin (bottom). Data arereported as mean 士 SEM.* p < 0.05; ** p < 0.01; *** p < 0.001.

2.5. CR Condition Upregulated CK2x by miR-760 Downregulation Via KCNQ1OT1 Upregulationin Human Cancer Cells

It was supposed that KCNQ1OT1 expression was upregulated in the CR condition because the level of CK2x mRNA was upregulated in the CR condition [24]. HCT116 and MCF-7 cells were incubated in CR conditions to test this hypothesis. As shown in Figure 5A, the expression of KCNO10T1 and CK2x was induced by CR in these cells where additional treatment with KCNQ10T1 siRNA suppressed CR-mediated CK2ainduction, indicating KCNQ10T1 as a positive regulator of CK2x in CR conditions. Additionally, analysis with real-time qPCR indicated that the level of miR-760 decreased by70% in CR conditions, whereas the levels of miR-186, miR-216b, and miR-337-3p in CRconditions were unchanged or increased (Figure 5B). Next, the effect of miR-760 on CRmediated CK2x upregulation was examined. Treatment with a miR-760 mimics suppresseoCR-mediated CK2x upregulation, indicating miR-760 as a major negative regulator of CK2ain CR conditions (Figure 5C). Finally, real-time gPCR analysis revealed that treatment with KCNO10T1 siRNA increased the levels of miR-760, whereas CR suppressed KCNO10T1 knockdown-mediated miR-760 induction (Figure 5D). Altogether, these results indicate that CR downregulated miR-760 by upregulating KCNQ10T1, resulting in CK2x upregulation in human cancer cells. Because it has been reported that the lncRNA SNHG6 also acts as a sponge of miR-760 in CRC cells [25], it was tested whether CR upregulates SNHG6 expression. However, SNHG6 expression was unchanged in the CR condition (data not shown)

Figure 5. Calorie restriction (CR) condition upregulated CK2x by miR-760 downregulation via KCNO10T1 upregulation in human cancer cells. (A,C and D) HCT116 and MCF-7 cells were transfected with KCNO10T1 siRNA (A and D) or miR-760 (C) for 1.5 days and then incubated under CR conditions for seven hours. (B) HCT116 and MCF-7 cells were incubated under CR conditions for seven hours. (A,C) The level of each mRNA was determined by RT-PCR using specific primers (top)Representative data from three independent experiments are shown. 3-Actin was used as a control graph to represent the quantitation of each RNA relative to B-actin (bottom). (B,D) Total RNA was isolated from cells and subjected to analysis by qPCR using RNU48 for normalization to determine the relative levels of the indicated miRNAs Data are shown as the means + SEM * y < 0.05.*p < 0.01;*** p < 0.001.

2.6. KCNQ1OT1 Was Downregulated during Replicative Senescence in Human Lung FibroblastCells, Which CR Conditions rescued

Since CK2x is downregulated in replicative senescent cells and aged tissues [16]it was examined whether KCNO1OT1 is downregulated during senescence in human lung fibroblast IMR-90 cells. KCNO10T1 knockdown decreased the mRNA level of CK2xin IMR-90 cells (Figure 6A). Conversely, KCNO1OT1 knockdown upregulated SA-P-gal activity in IMR-90 cells, and ectopic expression of CK2a abrogated the induction of SA-3-gal activity mediated by KCNO1OT1 downregulation (Figure 6B). To determine how KCNQ10T1 expression decreased by replicative senescence, IMR-90 cells were repeatedly passed until a senescence-like state was observed. Most cells at PDL 47 stained positively for SA-B-gal, whereas only a few stained positively for SA-B-gal among early-passage(PDL 34) cells (data not shown). The transcript levels of KCNO10T1 decreased by 60% in replicative senescent cells (PDL 47), compared to early-passage (PDL 34) cells, indicating that KCNO10T1 was downregulated during replicative senescence. Finally, CR conditions increased KCNO1OT1 by 150% in early-passage (PDL 34) cells compared with normal calorie conditions. However, CR conditions increased KCNO10T1 more strongly (by 250%) in replicative senescent cells (PDL 47) compared with normal calorie conditions. indicating that CR can rescue the decreased expression of KCNO10T1 and CK2x, mediated by replicative senescence (Figure 6C). Collectively, these data indicate that replicative senescence decreases CK2x expression via downregulating KCNQ10T1, and that CR can suppress replicative senescence via a KCNO1OT1-CK2 axis

Figure 6. KCNQ1011 was downregulated during replicative senescence in human lung fibroblast cells, which was able to be rescued by calorie restriction (CR) conditions. (A) IMR-90 cells (PDI36) were transfected with KCNO10T1 siRNA. The level of each mRNA was determined by RT-PCR using specific primers (top). Representative data from three independent experiments are shown3-Actin was used as a control. The graphs represent the quantitation of each mRNA relative to 6-actin (bottom). (B) IMR-90 cells (PDL 36) were transfected with KCNO10T1 siRNA for two days in the absence or presence of pcDNA3.1-HA-CK2a. Cells were stained with 5-bromo-4-chloro3-indolyl--D-galactoside, and representative images were obtained at 20x magnification (left)Scale bar = 100 um. Representative data from three independent experiments are shown. The graphs represent the percentage of blue-stained cells (right). (C) IMR-90 cells of PDL 34 and PDL 47 were incubated under CR conditions for seven hours. The level of each mRNA was determined by RT-PCR using specific primers (left). Representative data from three independent experiments are shown3-Actin was used as a control. The graphs represent the quantitation of each mRNA relative to 6-actin (right). Data are reported as mean 土 SEM.* p < 0.05, ** p < 0.01; *** p < 0.001. (D) Possible model illustrating the roles of KCNO10T1 for senescence and CR. PDL, population doubling level