Clock Protein Bmal1 and Nrf2 Work Together to Control Aging and Oxidative Stress by Regulating Daily Cycles of Prdx6

Two types of human lens epithelial cells (hLECs) were used: (1) a cell line (SRA01/04) immortalized with SV40 and (2) primary hLECs isolated from deceased persons of different ages. To avoid confusion, the remaining text will designate the immortalized LECs as SRA-hLECs, and the primary human LECs as hLECs.

The SRA-hLECs were generated from 12 infants who underwent surgery for retinopathy of prematurity [50] (a kind gift of late Dr. Venkat N. Reddy, Eye Research Institute, Oakland University, Rochester, MI, USA). These cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM, Invitrogen, Waltham, MA, USA) with 15% fetal bovine serum (FBS, Atlanta Biologicals, Atlanta, GA, USA), 100 µg/mL streptomycin, and 100 µg/mL penicillin in 5% CO₂ environment at 37 °C, as described previously [44,51]. Cells were harvested and cultured in 96-, 24-, 48-, or 6-well plates and 60- or 100-mm petri dishes according to the requirements of the experiments. To examine the effect of H₂O₂-induced oxidative stress on cell viability and level of oxidative load, cells were harvested and cultured in 96- or 48-well plates. After 24 h, these cells were exposed to different concentrations of H₂O₂ (0, 50, 100, or 200 µM) for different time intervals as indicated in figure legends and then processed for cell viability and intracellular redox state. Similarly, cells or cells over- or under-expressing Bmal1 or Prdx6 were harvested and cultured in 60 mm petri dishes or 6-well plates. 24 h later, these cells were treated with H₂O₂ as noted above and processed for expression analysis to examine the effects of H₂O₂-induced oxidative stress on expression levels of Nrf2 or Bmal1 or Prdx6, as indicated in figure legends.

Primary hLECs were isolated from normal eye lenses of deceased persons or healthy donors of different ages (16, 21, 24, 52, 56, 58, 60, 64, 68, 72, 76, and 78 years) obtained from the Lions Eye Bank, Nebraska Medical Center, Omaha, NE, USA. and National Development & Research Institute (NDRI), Inc., Philadelphia, PA, USA. According to regulation HHS45CFR 46.102(f), studies involving material from deceased individuals are not considered human subject research as defined at 45CFR46.102(f) 10(2) and do not require Institutional Review Board (IRB) oversight. Due to the limited sample size, eye lenses were divided into three groups by age: those age 16, 21, and 24 years, n = 6; age 52, 56, 58, and 60 years, n = 8; and age 64, 68, 72, 76, and 78 years, n = 10. Briefly, each capsule was trimmed before explanting in 35 mm culture dishes precoated with collagen IV containing a minimum amount of DMEM containing 20% fetal bovine serum (FBS), with a brief modification as described earlier [52,53,54,55]. Capsules were spread by forceps with cell layers upward on the surface of plastic petri dishes. Culture explants were trypsinized and re-cultured. Cell cultures attaining 90%–100% confluence were trypsinized and used for experiments [45,51,56]. Western analysis was used to validate the presence of αA-crystallin, a specific marker for LEC identity (data not shown).

Intracellular ROS (overall cellular oxidative load) were measured by use of fluorescent dye 2’-7’-dichlorofluorescein diacetate (H2-DCF-DA), a nonpolar compound that is converted to a polar derivative (dichlorofluorescein) by cellular esterase after incorporation into cells [32,38,40,57]. On the day of the experiment, the medium was replaced with Hank’s solution containing 10 µM H2-DCF-DA dye and cells were incubated. After 30 min, intracellular fluorescence was detected with excitation (Ex) at 485 nm and emission (Em) at 530 nm by using a Spectra Max Gemini EM (Mol. Devices, Sunnyvale, CA, USA).

ROS were measured according to the company’s protocol (CellROX^® Deep Red Oxidative Stress Reagent, Catalog No. C10422↗) and as described in our previously published protocol [39,40,41,58]. In brief, lentiviral (LV) short-haipin(Sh)-Control or LV Sh-Bmal1 SRA-hLECs and/or SRA-hLECs overexpressing green fluorescent protein (GFP)-Vector and/or GFP-Bmal1 plasmids were seeded in 96-well plates, and, 24 h later, these cells were exposed to different concentrations of H₂O₂ (0, 50, 100, and 200 µM), as indicated in figure legends. After 6 h, CellROX deep red reagent was added at a final concentration of 5 µM, and cells were incubated at 37 °C for 30 min. Media containing, CellROX deep red reagent was removed and fixed with 3.7% formaldehyde. Fifteen min later fluorescence signals were measured at Ex640nm/Em665nm with Spectra Max Gemini EM (Mol. Devices, Sunnyvale, CA, USA).

Total ribonucleic acid (RNA) from the cultured hLECs and SRA-hLECs was isolated using the single-step guanidine thiocyanate/phenol/chloroform extraction method (Trizol Reagent, Invitrogen). To examine the levels of Bmal1, Clock, Nrf2, Prdx6, NQO1, HO1, superoxide dismutase (SOD)1, and SOD2, 0.5 to 5 micrograms of total RNA was converted to complementary deoxyribonucleic acid (cDNA) using Superscript II RNAase H-reverse-transcriptase. Real-time quantitative PCR was performed with SYBR Green Master Mix (Roche Diagnostic Corporation, Indianapolis, IN, USA) in a Roche^® LC480 Sequence detector system (Roche Diagnostic Corporation). PCR conditions included 10 min hot start at 95 °C, followed by 45 cycles of 10 s at 95 °C, 30 s at 60 °C, and 10 s at 72 °C. The primer Sequence is shown in Table 1.

The relative quantity of the messenger (m) mRNA was obtained using the comparative threshold cycle (CT) method. The expression levels of target genes were normalized to the levels of β-actin as an endogenous control in each group.

Plasmid vector pEGFP-C1 for eukaryotic expression was purchased from Clontech (Palo Alto, CA, USA). Human aryl hydrocarbon receptor nuclear translocator-like (ARNTL)/pGFP-Bmal1 was purchased from OriGene (Cat No. RG207870; OriGene, Rockville, MD, USA). SRA-hLECs were transfected by using the Neon Transfection System (Invitrogen, Waltham, MA, USA).

Human LECs cDNA library was used to isolate Prdx6 cDNA having a full-length open reading frame. A full-length Prdx6-As construct was made by sub-cloning Prdx6 cDNA into a pcDNA3.1/NT-GFP-TOPO vector in reverse orientation. Plasmid was amplified following TOP 10 bacterial cells transformation, as described earlier [38,43,57].

CopGFP (green fluorescent protein 2 from the Copepod Pontellina plumata) control lentiviral particle (LV Sh-Control, sc-108084) and Bmal1 (Bmal1)/GFP shRNA (LV Sh-Bmal1, sc-38165-VS) were purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Human LECs were infected following the company’s protocol and as described in our recent paper [41]. Briefly, SRA-hLECs were cultured in 12-well plates in complete medium. After 24 h, media were removed and replaced with 1 mL medium containing polybrene (sc-134220, Santa Cruz Biotechnology, Dallas, TX, USA) at a final concentration of 5 µg/mL. Cells were infected by adding Sh-Control and Sh-Bmal1 lentiviral particles to the culture, mixed by swirling and incubated overnight. Twenty-four hours after infection, polybrene-containing medium was removed, and fresh complete medium (without polybrene) was added. Infected cells were split and incubated in complete medium for a further 24–48 h. For stable selection, the infectants were treated with puromycin dihydrochloride (sc-108071, Santa Cruz Biotechnology, Dallas, TX, USA) selection marker. These stably-infected SRA-hLECs with LV Sh-Control or LV Sh-Bmal1 were used for the present studies.

Total cell lysates of SRA-hLECs were prepared in ice-cold radioimmune precipitation buffer (RIPA buffer), and protein blot analysis was performed as described previously [37,40,41,59,60]. The membranes were probed with anti-Bmal1 (sc-365645, Santa Cruz Biotechnology, Dallas, TX, USA); anti-Clock (#5157S, Cell Signaling Technology, Danvers, MA, USA); Anti-Nrf2 (sc-365949 Santa Cruz Biotechnology); Anti-Nrf2 (ab62352, Abcam^®, Cambridge, MA, USA); Anti-Prdx6 antibody (LF-PA0011, Ab Frontier, South Korea), Anti-NQO1 (ab28947, Abcam^®, Cambridge, MA, USA), Anti-HO1 (Ab13248, Abcam^®, Cambridge, MA, USA), Anti-SOD1 (sc-515404, Santa Cruz Biotechnology, Dallas, TX, USA), Anti-SOD2 (sc-137254, Santa Cruz Biotechnology, Dallas, TX, USA), or β-actin (A2066, Sigma-Aldrich, St. Louis, MO, USA)/Anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH)(sc-365062, Santa Cruz Biotechnology, Dallas, TX, USA)/Tubulin (Abcam, Cambridge, MA, USA) as an internal control to monitor those protein expressions. After secondary antibody (sc-2354 and sc-2768, Santa Cruz Biotechnology, Dallas, TX, USA), protein bands were visualized by incubating the membrane with luminol reagent (sc-2048; Santa Cruz Biotechnology, Dallas, TX, USA), and images were recorded with a FUJIFILM-LAS-4000 luminescent image analyzer (FUJIFILM Medical Systems Inc., Hanover Park, IL, USA).

ChIP was performed using the ChIP-IT^® Express (Cat. No. 53008; Active Motif, Carlsbad, CA, USA) and ChIP-IT^® qPCR analysis kit (Cat. No. 53029; Active Motif, Carlsbad, CA, USA) following the manufacturer’s protocol and as described earlier [32,38]. The following antibodies were used: control immunoglobulin G (IgG) and antibody specific to Bmal1 (sc-365645, Santa Cruz Biotechnology, Dallas, TX, USA) and GFP (sc-69779, Santa Cruz Biotechnology, Dallas, TX, USA); RT-qPCR; 2 min at 95 °C, 15 s at 95 °C, 20 s at 58 °C, and 20 s= at 72 °C for 40 cycles in 20 μL reaction volume (RT-qPCR). Data obtained from RT-qPCR were presented as a histogram. RT-PCR amplification was carried out using 5 μL of DNA sample with primers, as indicated below. The program for quantification amplification was 3 min at 94 °C, 20 s at 95 °C, 30 s at 59 °C, and 30 s at 72 °C for 36 cycles in 25 μL reaction volume (RT-PCR). Data obtained with RT-PCR were run on 1% agarose gel, bands were visualized under UV, and image was captured using LAS-4000 Image analyzer (FUJIFILM). Primers were as follows: (1) Within Bmal1 binding site ranging from -400 to -305: Forward primer, 5′-CAGAGTCAAACCTGGCGCATC-3′; and Reverse primer: 5′-CATCCTTCAGACACTATAGGCC-3′ and (2) Beyond Bmal1 binding site ranging from -2048 to -1913: Forward primer: 5′-GTCTCTCATCCCACCTGACG-3′; and Reverse primer: 5′-GGCAATGCTTCTGCACTCTG-3′.

The 5′-flanking region spanning from −918 to +30 bp was isolated from human genomic DNA by using an Advantage^® Genomic PCR Kit (Cat. No. 639103 & 639104, Clontech Laboratories Inc., Mountain View, CA, USA). PCR product was cleaned and verified by sequencing as described previously [32,41,51]. A construct containing −918 to +30 bp was engineered by ligating it to the basic pCAT vector (Promega) using the SacI and XhoI sites. Primers were as follows:

Forward primer; 5′-GACAGAGTTGAGCTCCACACAG-3′; and

Reverse primer; 5′-CACGTCCTCGAGAAGCAGAC-3′.

PCR-based site-directed mutagenesis was carried out using the QuikChange^TM lightning Site-Directed Mutagenesis kit (Agilent Technologies; Catalog No. 210518), following the company’s protocol. Briefly, amino acid exchanges at the Bmal1 site (E-Box; -341/-336) mutant (T to A and A to T) and at Nrf2 site (ARE; -357/-349) mutants (TG to GT) were generated by point mutations in the human promoter of Prdx6-CAT plasmid. The following complementary primers were used (changed nucleotides are in red boldface type and underlined):

A colorimetric 3-(4, 5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2 to 4-sulphophenyl)-2H-tetrazolium salt, (MTS) assay (Promega, Madison, WI, USA) was performed as described earlier [38,53,61]. This assay of cellular viability uses MTS and an electron coupling reagent (Phenazine ethosulfate; PES). PES has enhanced chemical stability, which allows it to be combined with MTS to form a stable solution. Assays were performed by adding MTS reagent directly to culture cells, incubating for 1–4 h and recording the absorbance at 490 nm with a 96-well plate reader, Spectra Max Gemini EM (Mol. Devices, Sunnyvale, CA, USA). Results were normalized with an absorbance of the untreated control(s).

C57BL/6female mice (8–10 months old) were obtained from Charles River Laboratories, Wilmington, MA, USA. Mice were maintained at a stable temperature (22 ± 2 °C) and humidity (55% ± 5%) under a 12/12 h light/dark cycle (lights on at 7:00 a.m.; lights off at 7:00 p.m., where ZT0 and ZT12 indicate the times when lights were switched on and off, respectively). C57BL/6 female mice were sacrificed by cervical dislocation and lenses were collected at ZT2, ZT6, ZT10, ZT14, ZT18, and ZT22 for analysis of mRNA, protein, and ROS. Lenses were isolated under dim red light during ZT14, ZT18, and ZT22 times to avoid the effect of light. The University of Nebraska Medical Center Institutional Animal Care and Use Committee approved all animal care and handling protocols.

For all quantitative data collected, statistical analysis was conducted by Student’s t-test and/or one-way ANOVA when appropriate and was presented as mean ± S.D. of the indicated number of experiments. A significant difference between control and treatment groups was defined as p value of <0.05 and 0.001 for two or more independent experiments.

In an earlier report, we demonstrated that an increase of oxidative load in aging cells is a cause of progressive failure of Nrf2/Prdx6-mediated protective response [32]. In this study, using aging primary hLECs of different ages, we sought to establish a connection (if any) between the expression levels of Bmal1 and the Nrf2-driven antioxidant pathway. We quantified ROS amounts and expression levels of Bmal1, Clock, Nrf2, and Nrf2-targeted antioxidant genes in primary hLECs of variable ages, as described in Materials and Methods. H2-DCF-DA dye and qPCR quantitation assays revealed an age–dependent progressive elevation in ROS levels with significant reduction in Bmal1 and Clock, and Nrf2 and Nrf2 target genes mRNA, as shown in Figure 1. However, because of limited availability of aging/aged primary hLECs, we were only able to measure the levels of Bmal1, Nrf2 and Prdx6 protein expression. Western analysis revealed that, similar to the mRNA expression pattern, protein expression of Bmal1, Nrf2, and Prdx6 declined with advancing age, suggesting that dysregulation occurred at transcription and translation levels with aging (Figure 1J). We found a significant increase in ROS levels was associated with a dramatic decrease of transcripts of clock genes (Bmal1 and Clock) and Nrf2 and Nrf2 targeted antioxidant genes, Prdx6, NQO1, HO1, SODs in hLECs derived from the aged group (Figure 1, 52 y onward). Taken together, the results demonstrated a substantial inverse correlation between cellular ROS level and circadian clock gene and Nrf2-mediated antioxidant genes in aging cells, as well as provided a background for this study.

Based on the results illustrated in Figure 1B–J showing the age-dependent reduction in Bmal1-Clock and Nrf2-mediated antioxidant gene expression with a progressive increase in ROS levels, we next examined whether Nrf2 or Nrf2/ARE pathway is upregulated in cells overexpressing Bmal1. We extrinsically expressed hLECs with different concentrations of Bmal1 plasmids. Due to the limited availability of primary hLECs, we utilized the hLECs cell line SRA-hLECs and overexpressed them either with pGFP-Bmal1 or pGFP-empty vector plasmid. To avoid any effect of DNA concentration on cells, an equal amount of plasmid DNA was transfected in each experimental group. Total RNA and protein isolated from SRA-hLECs transfected with different concentrations of pGFP-Bmal1 or pGFP empty vector were processed for real-time quantitative PCR (RT-qPCR) and Western blot analysis, respectively. A careful analysis of results of qPCR (Figure 2A) and Protein blot (Figure 2B) revealed that cells overexpressing Bmal1 displayed increased Nrf2, Prdx6, NQO1, and SODs mRNA and protein, as well as that enhanced expression of these genes was directly linked to Bmal1 concentrations in cells (Figure 2A,B, violet vs. light orange and dark orange bars). These results indicated Bmal1 regulation of the antioxidant pathway in hLECs.

Results from aging hLECs showed (1) a progressive reduction of Bmal1 expression, along with Nrf2-mediated antioxidant pathway (Figure 1), and (2) Bmal1 overexpression-mediated increased expression of Nrf2 and antioxidant genes (Figure 2). Next, we examined whether dysregulation of the Nrf2 pathway occurs in Bmal1- depleted SRA-hLECs as observed in aging hLECs. We knocked down Bmal1 by stably infecting SRA-hLECs using lentiviral (LV) Sh-Control or LV Sh-Bmal1 as described in Materials and Methods. Figure 3A is a representative photomicrograph of LV Sh-Control and/or LV Sh-Bmal1 infected SRA-hLECs. Bmal1-depleted SRA-hLECs did not display phenotype changes and were indistinguishable from LV Sh-Control infected SRA-hLECs. In addition, Bmal1 depletion did not affect cell survival/growth significantly in vitro, but these cells had increased sensitivity to oxidative stress-induced cell death. mRNA and protein analyses using specific probes to Bmal1, Nrf2, and its target genes by qPCR and Western blot showed significantly reduced expression of Bmal1 mRNA (Figure 3Ba) and protein (Figure 3Ca). Deficiency of Bmal1 was directly associated with reduction in Nrf2 and its target antioxidant genes mRNA and protein expression (Figure 3B,C). These results indicate that, indeed, Bmal1 is a critical component in regulation of Nrf2 and Nrf2/ARE-mediated antioxidant pathway in hLECs. Additionally, the data from SRA-hLECs provided support for using those cells in further research rather than using aging primary hLECs, which are more difficult to obtain.

Previous experiments revealed a connection between Bmal1 regulation of Nrf2 and Nrf2-dependent antioxidant genes, but the mechanism of regulation in hLECs was not evident. However, the experiments described above showed that repression of Nrf2 or Nrf2-mediated antioxidant genes occurred at mRNA, indicating that dysregulation of these genes occurred at transcriptional level. It has been reported that Bmal1 regulates Nrf2 transcription by binding to E-Box present in its promoter [6]. In addition, the presence of the E-Box element has been predicted in the proximal promoter region of major antioxidant genes [64], suggesting that some antioxidant genes can be directly regulated by Bmal1. To examine whether Prdx6 also is regulated through E-Box response elements, we analyzed Prdx6 gene promoter using MatInspector (Genomatix), a web-based computer program, to identify transcription factor binding sites. This predicted presence of a Bmal1 responsive element, E-Box (5`--³⁴¹CACGTG³³⁶-3′) in the proximal promoter region of hPrdx6 gene, spanning from 5′ -918 to +1 bp (TSS) (Figure 4A). The presence of functional Nrf2/ARE binding site, (5′-³⁵⁷nTGACCGAGCn³⁴⁹-3‘) in Prdx6 gene promoter was reported previously by our laboratory and others [6,18,32,41].

To investigate the mechanism by which Bmal1 might regulate antioxidant genes like Prdx6 in aging hLECs, we performed ChIP-RT-qPCR as described in Materials and Methods and previously published protocols [32,38,41]. We used antibody specific to Bmal1 to demonstrate whether Bmal1 binds to E-Box (-341 to -336, CACGTG) sequences present in the hPrdx6 gene promoter in hLECs of different ages directly detached from the lens (to avoid cell culture effects). Cells were selected from the age groups of lenses used in the previous experiment (Figure 1). As shown in Figure 4B, we observed a progressive decline of Bmal1 binding in aged hLECs (56 years and 68 years) in comparison to younger subjects (Figure 4B, 24 years; orange bars). No amplicon was detected with negative control IgG (Figure 4B, black bars). As a whole, our results demonstrate that during aging, the binding of Bmal1 to its responsive element, E-box, is decreased, and maybe a plausible cause for dysregulation of Bmal1/Nrf2/ARE-mediated protective antioxidant pathway.

SRA-hLECs with Bmal1 overexpressed showed enhanced expression of antioxidant genes (Figure 2), while Bmal1 knockdown downregulated the genes involved in antioxidant mechanism pathways (Figure 3). To investigate the effect of Bmal1′s cellular abundance on its binding activity to E-Box of hPrdx6 gene promoter, we performed chromatin ChIP-RT-PCR in Bmal1 overexpressed and Bmal1 knockdown SRA-hLECs. SRA-hLECs were transfected with cytomegalovirus (CMV) empty vector or Bmal1 linked to GFP. After 48 h, transfected cells were processed for ChIP assay with Prdx6 promoter using ChIP grade anti-Bmal1 (Figure 5A), as well as anti-GFP (Figure 5B) antibodies. The hPrdx6 DNA with Bmal1 site was detected in the immunoprecipitated samples after RT-PCR. As shown in Figure 5A,B, E-Box sequence was occupied by Bmal1, and increased enrichment of Bmal1 to the E-Box sequences was concentration-dependent in the hPrdx6 gene promoter. We did not detect amplicon with primers ranging from -2048 to -1913 that was beyond the Bmal1 binding site or in control IgG, indicating the specificity of the Bmal1 and GFP antibodies. Results revealed that increased abundance of Bmal1 significantly enhanced the Bmal1 availability at E-Box site and defined the mechanism of Bmal1/E-Box-dependent enhanced Prdx6 transcription. To confirm Bmal1 binding to E-Box in Prdx6 promoter, we next performed ChIP-PCR in Bmal1-depleted SRA-hLECs. SRA-hLECs infected with lentiviral specific shRNA to Bmal1 or Control shRNA was used to perform the ChIP analysis with Bmal1 antibody or control IgG. Significantly, reduced Bmal1-DNA binding in Bmal1-depleted cells in comparison to control was observed in the immunoprecipitated samples after RT-PCR (Figure 5C). The PCR products of hPrdx6 E-Box region were undetectable in control IgG samples and/or with primers selectively designed beyond the Bmal1 binding site, validating the specificity of antibody and specific interaction of Bmal1 with E-Box sequences. Bmal1-DNA abundance at E-Box was significantly reduced in Bmal1 shRNA, indicating less availability of Bmal1 to bind at E-Box sequence in hPrdx6 gene promoter. Collectively, these results demonstrated that increased or decreased Bmal1/DNA binding activity to E-Box sequence present in the gene promoter region was dependent on the availability of Bmal1, thereby suggesting a plausible mechanism of Bmal1 regulation of Nrf2 or Nrf2 –mediated antioxidant response in aging or aged cells.

We next sought to examine if the gain and loss in the Bmal1/E-Box–DNA binding phenomenon that occurred following Bmal1 over- and under-expression was functional and would modulate Prdx6 transcription through Bmal1; we carried out Prdx6 promoter activity using Prdx6 gene promoter-linked CAT, bearing Bmal1 and Nrf2 response elements or their mutants in cells over- or under-expressing Bmal1. For assay, we transfected SRA-hLECs with pCAT-hPrdx6-wild type (WT) promoter construct with GFP-vector and/or GFP-Bmal1 plasmid. Seventy-two hours later, CAT assay was performed as described previously [32,41] and transfection efficiency were normalized with GFP O.D. reading. Transactivation assay with WT construct showed significant increased CAT activity in Bmal1 over-expressed construct in a dose-dependent fashion (Figure 6B).

Next, we examined whether increased Prdx6 transcription is dependent upon Bmal1/E-Box and/or Nrf2/ARE. SRA-hLECs were transfected with pCAT-hPrdx6 wild-type (WT) or Bmal1 specific mutants at E-Box (E-Box-mut), or Nrf2 specific mutant at ARE (ARE-mut), or at both Bmal1/E-Box and Nrf2/ARE (E-Box-mut + ARE-mut) constructs fused to empty CAT reporter vector as shown (Figure 6C). After 48 h, transfectants were treated or untreated with H₂O₂ as indicated. At normal physiological conditions (untreated control), activity of Prdx6 promoters containing mutations at only one site showed the following reductions: with a mutation at E-box, ~70% reduction; at ARE construct, ~ 50% reduction. In the construct mutated at both sites, only 5% activity was observed, and that activity was indistinguishable from CAT-vector values (Figure 6C). Because the magnitude of oxidative stress can modulate genes and reset their transcription [65], transfectants were treated with increasing concentrations of H₂O₂. As we expected from previous reports [32,41,66], the regulation of promoter activity was H₂O₂ concentration-dependent; lower concentrations of H₂O₂ significantly augmented the promoter activity compared to transfectants exposed to higher concentrations, which showed reduced promoter activity (Figure 6C). The promoter activity was increased even at both Bmal1/E-Box and Nrf2/ARE (E-Box-mut + ARE-mut) mutant constructs at lower concentrations of H₂O₂, but was significantly lower than in WT or either single-mutant construct. To our surprise, Prdx6 promoter containing mutation at both E-Box and ARE sites had some activity, suggesting that Prdx6 promoter may contain some responsive regulatory element(s) other than Nrf2 or Bmal1 which require further investigation. Data analysis revealed that Prdx6 transcription was cooperatively regulated by Bmal1 and Nrf2. We believe that this is the first report showing that both Nrf2 and Bmal1 are essential for optimum transcription of Prdx6 in eye lens epithelial cells.

However, to further verify transcriptional activation of Prdx6 promoter by Bmal1, we performed transient transfection experiments in SRA-hLECs by transfecting cells with the pCAT-hPrdx6-WT reporter constructs already infected either with Bmal1-specific shRNA or with control shRNA lentiviral. A significant reduction in Prdx6 transactivation was observed in LV Sh-Bmal1 in comparison to control (LV Sh-Control; Figure 6D). Taken together, data indicated that Bmal1 expression influenced the transcriptional activity of the Prdx6 gene, and both transcriptional molecules, Bmal1 and Nrf2, contributed to regulation of Prdx6 transcription.

As shown in Figure 2, Figure 5 and Figure 6, overexpressing Bmal1 enhanced expression of antioxidant genes, such as Prdx6. Mounting evidence demonstrates that expression of Prdx6 is essential for cell survival/protection against various internal and external stressors [18,32,38,41]. To examine if Bmal1 overexpression rescued hLECs from H₂O₂-evoked oxidative stress as reported for other cell types [48,65,67,68] and to determine the contribution of Prdx6 in Bmal1–mediated cytoprotection, SRA-hLECs with or without the antisense of Prdx6 (As-Prdx6) were transfected with GFP vector or GFP-Bmal1 plasmid as described in Materials and Methods. Equal amounts of each plasmid DNA were used to avoid transfection effects. For this experiment, we used the same batch of transfected cells that were used for the overexpression experiments with Bmal1 (Figure 2), but SRA-hLECs overexpressing GFP-vector or GFP-Bmal1 were transfected with/without As-Prdx6 and were exposed to H₂O₂ as indicated in Figure 7. Quantification of ROS (Figure 7A, purple bar vs. orange bars) by CellROX Deep Red dye and cell viability (Figure 7B, purple bar vs. orange bars) by MTS assay revealed that Bmal1 overexpression significantly reduced ROS intensity and enhanced cell viability. However, it was not clear if Prdx6 is a requisite for Bmal1-mediated cytoprotection against H₂O₂-induced oxidative stress. To examine if Bmal1 exerts its protective activity via upregulation of Prdx6, we used the As-Prdx6 to knock down Prdx6 expression in SRA-hLECs as reported previously [32,57]. Expression level of Prdx6 was validated using protein blot analysis (Figure 7C). These Prdx6 depleted SRA-hLECs were overexpressed with GFP-empty vector or GFP-Bmal1, and the transfectants were treated or untreated with H₂O₂. Quantification of ROS levels (overall oxidative load) in Prdx6 knockdown SRA-hLECs overexpressing Bmal1 by using CellROX Deep Red dye showed that Bmal1 overexpression did significantly lower ROS levels (Figure 7A, orange bars vs. red bars). Furthermore, these Prdx6-depleted SRA-hLECs were highly vulnerable to cell death as evidenced by viability assay (Figure 7B, orange bars vs. red bars), suggesting that, indeed, Bmal1 acted through Prdx6, and Prdx6 was a major component for Bmal1-mediated protection, at least for lens epithelial cells during oxidative stress. However, Prdx6 depletion did not eliminate Bmal1-mediated protection absolutely, indicating that other antioxidants induced by Bmal1 contributed to protecting SRA-hLECs.

Previous experiments in this study demonstrated that Bmal1 was a critical element for hLECs protection, and acted by upregulating transcription of the Nrf2-mediated antioxidant pathway. However, those results did not shed light on the fate of Nrf2/antioxidant genes like Prdx6 in Bmal1-deficient SRA-hLECs (like aging hLECs) and SRA-hLECs lacking Bmal1 during oxidative stress. To examine this, SRA-hLECs were infected with the lentiviral specific to Bmal1 shRNA. ROS intensity and cell viability of Bmal1-depleted cells were measured and are shown in Figure 8. Results demonstrated that Bmal1-deficient SRA-hLECs were highly sensitive to H₂O₂ and showed progressive increases in intracellular ROS levels. Cellular ROS levels were H₂O₂ concentration-dependent compared to respective controls (Figure 8A). Next, we assessed cell viability by treating cells with the same amount of H₂O₂ as in Figure 8A. As shown in Figure 8B, Bmal1-depleted SRA-hLECs exposed to increasing concentrations of H₂O₂ showed concentration-dependent cell death, which was directly linked to levels of ROS (Figure 8A), suggesting that Bmal1-deficiency led to dysregulation of Nrf2/ARE-mediated antioxidant pathway. In parallel experiments, we examined the status of Bmal1, Nrf2 and its target antioxidant gene, Prdx6, in hLECs infected with LV Sh-Control or LV Sh-Bmal1 under oxidative conditions. LV Sh-Bmal1 successfully knocked down Bmal1 as evidenced by qPCR (Figure 3). These Bmal1-depleted cells displayed reduced expression of Nrf2, as well as Prdx6, as observed in primary aging hLECs (Figure 1). Surprisingly, the levels of Nrf2 and Prdx6 were further reduced with reduction in Bmal1 expression in cells facing higher levels of H₂O₂-induced oxidative stress, suggesting that increased oxidative stress negatively affected Bmal1 and Bmal1 regulation of Nrf2 and Prdx6. However, a lower concentration of H₂O₂ (50 µM) increased the expression of Bmal1/Nrf2/Prdx6 cells (Figure 8C, black bar vs. blue bar), indicating that levels of ROS drove expression levels of clock protein Bmal1 signaling that resulted in Bmal1-dependent expression of Nrf2/Prdx6 within the cellular microenvironment.

Bmal1 control genes, such as Nrf2, show a circadian oscillation [12]. Because we found that a major clock transcriptional protein, Bmal1, regulated antioxidant pathways, we asked whether Nrf2 and Nrf2-mediated antioxidant pathways, including Prdx6, is rhythmically expressed in the eye lens, like Bmal1 controlled genes, in vivo [14,69]. We isolated lenses from a middle-aged (8–10 months) group of C57BL/6 mice, as mice of this age can, provide clues about the biological rhythm of clock genes/Nrf2-mediated antioxidant response in lenses that occurs when age-related changes begin (Figure 9). We examined mRNA and protein levels of clock genes, Nrf2, and its target genes, including Prdx6 at 4-h intervals, from mice kept in a cycle of 12 h light (L) and 12 h dark (D) (12L:12D) and denoted as Zeitgeber time (ZT) as shown in Figure 9. qPCR and Western blot analyses using their corresponding probes revealed a robust rhythmic pattern in the expression of selected genes including Prdx6 mRNA over the course of 48 h of light/dark cycle. The peak expression of clock genes was observed at ZT22-ZT2 (ZT0 was 7 a.m., when lights were turned on), and expression dipped at ZT6-ZT14 as shown in Figure 9. A similar pattern was observed with Nrf2 and its target antioxidant genes, including Prdx6. A significant and dramatic dip of clock genes, as well as Nrf2 and Nrf2/ARE-mediated antioxidant gene mRNA, was observed at ZT10 and ZT14 (i.e., 9 p.m.); thereafter, the levels began to rise (Figure 9).

Next, we examined temporal expression profiles of the above-mentioned molecules at protein levels to establish their circadian pattern. Protein expression analysis revealed a clear rhythmic expression of Nrf2 and Nrf2-mediated antioxidant genes. The expression pattern was directly linked to rhythmic expression of clock genes as shown in Figure 9B. Their expression also peaked at the ZT22 –ZT2, reaching a trough at ZT10-ZT14 (Figure 9B). Results also showed that reduction in the protein levels started to recover after ZT14. This was true for all molecules examined. However, we observed the similarity in the ZT peak between mRNA and protein. We also quantified the levels of ROS in the lenses as described in Materials and Methods. It was intriguing to observe an inverse relationship between levels of Bmal1/Nrf2/antioxidant pathway and ROS, suggesting ROS regulation of clock-mediated antioxidant pathway or vice versa.