White light emitting diode induces autophagy in hippocampal neuron cells through GSK-3-mediated GR and RORα pathways

White LED light exposure induces autophagy in hippocampal neuron cells

White LED light-induced autophagy in hippocampal neuron cells was demonstrated by using different methods. Firstly, transmission electron microscopy (Figure 1A–1D) demonstrated that the number of autophagic vacuoles was significantly increased (p < 0.01) in Light group (Figure 1E). Secondly, autophagy-related genes, such as LC3B, Beclin and ATG7, were significantly (p < 0.01) up-regulated in Light group at both mRNA (Figure 1F) and protein (Figure 1G) levels. Thirdly, the GFP-LC3B immuofluorescence was significantly increased (Figure 1H–1I) in light-exposed cells transfected with the GFP-tagged MAP1LC3B expression plasmid (GFP-MAP1LC3B). Furthermore, blocking the autophagy with 3-MA, an early stage autophagy inhibitor, rescued light-induced decrease of cell viability, in a dose-dependent manner (Supplementary Figure 2).

Figure 1. White LED light induces activation of autophagy process in HT-22 cells. (A–B) TEM images of autophagic vacuoles in Dark and Light groups, respectively. Scale bars, 2 μm, n = 3; (C–D) Higher magnification of indicated regions in A and B (squares), showing autophagosomes with double membrane (red arrows). Scale bars, 0.5 μm; (E) The number of autophagic vacuoles. Values are means ± SEM, ** p < 0.01 compared with Dark group, n = 3. (F) Quantitative Real-time PCR analysis of autophagy-related gene lc3b, atg7, atg4c and beclin. Values are means ± SEM, ** p < 0.01 compared with Dark group, n = 6; (G) Western blot analysis of autophagy-related protein LC3B, Beclin and ATG7. Values are means ± SEM, ** p < 0.01 compared with Dark group, n = 6; (H–I) Representative fluorescence images of HT-22 cells transfected with GFP-LC3 plasmid. Cells were counterstained with Hochest (nuclei in blue) and more autophagosomes (green) were seen in Light group. Scale bar, 10 μm.

White LED light exposure enhance autophagy flux in hippocampal neuron cells

To detected the effect of white LED light on autophagy flux, we added 50 μM chloroquine (CQ) in the last 2 h of white LED light exposure. CQ significantly increased LC3II and p62 protein levels (p < 0.01) both in Dark and Light group (Figure 2B and 2E). Meanwhile, white LED light exposure significantly increased (p < 0.01) LC3II flux both in net flux and relative flux (Figure 2C and 2D). Concurrently, white LED light exposure significantly increased (p < 0.01) the relative flux of p62 (Figure 2G), yet not influence net flux of p62 (Figure 2F).

Figure 2. White LED light exposure enhance autophagy flux in hippocampal neuron cells. To detect the effect of white LED light on autophagy flux, we added 50 μM chloroquine (CQ) in the last 2 h of white LED light exposure. (A) Images of bands detected in Western blot analyses; (B) LC3B protein levels in Dark and Light group with or without CQ. Values are means ± SEM. Bars with different superscripts are significantly different from each other (p < 0.05, n = 3); (C) Net flux of LC3II protein. Values are means ± SEM, ** p < 0.01 compared with Dark group, n = 3; (D) Relative flux of LC3II protein. Values are means ± SEM, ** p < 0.01 compared with Dark group, n = 3; (E) p62 protein levels in Dark and Light group with or without CQ. Values are means ± SEM. Bars with different superscripts are significantly different from each other (p < 0.05, n = 3); (F) Net flux of p62 protein. Values are means ± SEM, n = 3; (G) Relative flux of p62 protein. Values are means ± SEM, ** p < 0.01 compared with Dark group, n = 3.

Surprisingly, we found that different CQ added mode lead to different autophagy flux results. LC3II and p62 protein levels were significantly increased (p < 0.01) both in Dark and Light group with added 50 μM CQ in 2 h after white LED light exposure (Supplementary Figure 3B and 3E). However, white LED light exposure significantly decreased LC3II flux both in the net autophagy flux (p < 0.01) and the relative autophagy flux (p < 0.05) (Supplementary Figure 3C and 3D). Concurrently, white LED light exposure significantly increased (p < 0.05) the relative flux of p62 (Supplementary Figure 3G), yet not influence the net flux of p62 (Supplementary Figure 3F).

White LED light activates GSK-3 and nuclear receptors GR and RORα in HT-22 cells

White LED light exposure significantly increased protein levels of GSK-3α and GSK-3β (p < 0.01), yet the serine-21 phosphorylated GSK-3α (Figure 3A) and serine-9 phosphorylated GSK-3β (Figure 3B) were significantly decreased (p < 0.01), indicating enhanced GSK-3 activity as phosphorylation at these particular serine residues inhibits the activity of GSK-3α and GSK-3β, respectively [30]. Concurrently, protein contents of GR and phospho-GR were also significant increased (p < 0.01). Moreover, RORα was significantly increased (p < 0.01), while REV-ERBα significantly decreased (p < 0.01) in light-exposed cells (Figure 3C), implicating possible disruption of clock genes and their circadian rhythms. Furthermore, SIRT1/AMPK pathway, which is known to regulate autophagy, is activated in response to white LED light exposure. The protein levels of SIRT1 and phospho-AMPK were significantly increased (p < 0.01) after LED light exposure (Figure 3D).

Figure 3. White LED light activates GSK3 and nuclear receptor GR, RORα in HT-22 cells. (A–B) Protein content of total and phospho-GSK-3α/β (Ser21/9). Significant decrease of p-GSK3/GSK3 ratio in Light group indicates activation of both GSK3α and GSK3β; (C) Protein content of GR, phospho-GR, RORα and REV-ERBα, showing up-regulation of both GR and RORα in Light group; (D) Protein content of SIRT1, AMPK and phospho-AMPK. Values are means ± SEM. **p < 0.01, compared with Dark group (n = 6).

White LED light disrupts the circadian rhythm of circadian clock genes in HT-22 cells

In order to elaborate the effect of white LED light on circadian rhythm of clock genes, we examined the mRNA expression profile of clock genes every 6 h for 48 h. In general, LED light exposure disrupted circadian rhythm of clock genes, in a gene-specific manner (Figure 4). Among 9 genes detected, clock (Figure 4A), bmal1 (Figure 4B), cry1 (Figure 4G) and cry2 (Figure 4H) genes displayed obvious time-phase shift and significantly suppressed (p < 0.01) oscillation amplitudes, while per1 (Figure 3D), per2 (Figure 4E) and per3 (Figure 4F) genes showed a delayed appearance of the 2^nd peak, indicating reduced frequency of oscillation. Interestingly, rorα (Figure 4C) completely lost its circadian rhythm and kept increasing (p < 0.01) over the period of examination.

Figure 4. White LED light influences the circadian rhythm of clock-related genes in HT-22 cells. (A–B) Clock and bmal1 mRNA, showing time-phase shift and suppressed oscillation amplitudes in Light group; (C) Rorα mRNA, showing diminished circadian rhythm and a continous increase in mRNA abundance over the period of examination in Light group; (D–E) Per1, per2 and per3 mRNA expression, showing reduced oscillation frequencies in Light group; (G–H) Cry1 and cry2 mRNA, displaying obvious time-phase shift and significantly suppressed oscillation amplitudes in Light group; (I) Rev-erbα mRNA, showing an obvious time-phase shift in Light group.

White LED light disrupts the circadian rhythm of autophagy genes in HT-22 cells

Compared to clock genes, autophagy-related genes showed less clear pattern of circadian rhythms (Figure 5). No significant effect of LED light exposure was observed for LC3B (Figure 5A), yet the fluctuation of ATG3 (Figure 5B) and ATG5 (Figure 5C) was significantly blunted (p < 0.05) in light-exposed cells. Interestingly, ATG7 (Figure 5D) was significantly up-regulated (p < 0.01), reaching a plateau at 42 h after light exposure.

Figure 5. White LED light influences the circadian rhythm of autophagy-related genes expression in HT-22 cells. (A) Lc3b mRNA, with no clear pattern of circadian rhythm and no significant alterations in Light group; (B–C) Atg3 and atg5 mRNA, showing significantly blunted fluctuation in light-exposed cells; (D) Atg7 mRNA, showing a continuous increase in mRNA abundance until a plateau was reached 42 h after light exposure.

RORα inhibitor rescues autophagy and normalizes the expression of autophagy-related proteins, but not GR or GSK-3

SR1001, a specific RORα inhibitor, significantly rescued the expression of autophagy-related proteins such as LC3B, Beclin (Figure 6A), ATG4c and ATG7 (Figure 6B) in light-exposed HT-22 cells. Accordantly, the light-induced increase in the protein content of SIRT1 and phospho-AMPK was also rescued by SR1001 (Figure 6C). However, SR1001 was not able to restore light-induced GR activation (Figure 6D) and GR nuclear translocation (Figure 6E–Figure 6G). Moreover, SR1001 also failed to rescue the activity of GSK-3α (Figure 6H) and GSK-3β (Figure 6I) in light-exposed cells. These results indicate that RORα plays a pivotal role in white LED light-induced cell autophagy downstream of GR and GSK-3 action.

Figure 6. RORα inhibitor, SR1001, rescues the expression of autophagy-related proteins, yet not GSK3α, GSK-3β and GR in light-exposed HT-22 cells. (A–D) Protein content of LC3B, Beclin, ATG4c, ATG7, SIRT1, phospho-AMPK, RORα, GR and phospho-GR. RORα inhibitor SR1001 completely rectified light-induced up-regulation of RORα and autophagy-related proteins, including LC3B, Beclin, ATG4c, ATG7, SIRT1 and phospho-AMPK, in HT-22 cells. However, increases in GR and phospho-GR protein expression were not rectified by SR1001. Values are means ± SEM, *p < 0.05, **p < 0.01, compared with Dark group, ^#p < 0.05, ^##p < 0.01, compared with Light group (n = 3 or 4); (E–G) Immunofluorescence of GR, showing that SR1001 was not able to restore light-induced GR activation and GR nuclear translocation. The nuclei were stained with Hoechst (blue) and GR was stained with GR antibody (red). Scale bars, 75 μm; (H–I) Protein content of total and phospho-GSK-3α/β (Ser21/9). Light-induced increase in total GSK-3α/β and decrease in phospho-GSK-3α/β were not rescued by SR1001. Values are means ± SEM, *p < 0.05, **p < 0.01, compared with Dark group (n = 3).

GR inhibitor rescues autophagy and the expression of autophagy-related proteins, but not GSK3

GR inhibitor RU486 significantly rescued the expression of autophagy-related proteins such as LC3B, Beclin (Figure 7A), ATG4c and ATG7 (Figure 7B), together with restored protein content of SIRT1 and phospho-AMPK (Figure 7C) in light-exposed HT-22 cells. Moreover, RU486 significantly prevented light-induced up-regulation of RORα protein expression (Figure 7D) and GR activation indicated by GR nuclear translocation (Figure 7E–Figure 7G). However, the increase of GSK-3α and GSK-3β protein expression and the decrease of their phosphorylations induced by LED light exposure were not affected by RU486 (Figure 7H and Figure 7I). These results indicate that GR and RORα play an important role in white LED light-induced autophagy downstream of GSK-3 action.

Figure 7. GR inhibitor, RU486, rescues the expression of autophagy-related proteins and nuclear receptor RORα expression, yet not GSK3α and GSK-3β in light exposure cells. (A–D) Protein content of LC3B, Beclin, ATG4c, ATG7, SIRT1, phospho-AMPK, RORα, GR and phospho-GR. GR inhibitor RU486 completely rectified light-induced up-regulation of GR, phospho-GR, RORα and autophagy-related proteins, including LC3B, Beclin, ATG4c, ATG7, SIRT1 and phospho-AMPK, in HT-22 cells. Values are means ± SEM, *p < 0.05, **p < 0.01, compared with Dark group, ^#p < 0.05, ^##p < 0.01, compared with Light group (n = 4); (E–G) Immunofluorescence of GR, showing that RU486 was able to restore light-induced GR activation and GR nuclear translocation. The nuclei were stained with Hoechst (blue) and GR was stained with GR antibody (red). Scale bars, 75 μm; (H–I) Protein content of total and phospho-GSK-3α/β (Ser21/9). Light-induced increase in total GSK-3α/β and decrease in phospho-GSK-3α/β were not rescued by RU486. Values are means ± SEM, *p < 0.05, compared with Dark group (n = 4).

GSK-3 inhibitor or knockdown rescues autophagy and activation of autophagy-related proteins as well as GR and RORα expression

CHIR-99021, a specific GSK-3 inhibitor, significantly rescued the expression of autophagy-related proteins such as LC3B, Beclin (Figure 8A), together with decreased SIRT1 and phospho-AMPK (Figure 8B) protein expression by CHIR-99021. Accordingly, GSK-3 inhibitor significantly suppressed nuclear translocation of GR in light-exposed cells (Figure 8C–Figure 8E), which was associated with normalized phospho-GR but not total GR protein content (Figure 7F). Moreover, GSK-3 inhibitor partially yet significantly alleviated light-induced increase of RORα protein expression, yet REV-ERBα protein content was further decreased (Figure 8G). These results indicate that GSK-3 serves as an upstream regulator of GR and RORα signaling in autophagy induced by white LED light exposure. This was further supported by shRNA-mediated GSK-3α knockdown (Figure 9A and 9B), which led to significantly blunted or diminished (p < 0.01) responses of phospho-GR (Figure 9A and 9C), RORα (Figure 9A and 9D) and LC3B (Figure 9A and 9E) protein expression in light-exposed cells.

Figure 8. GSK3 inhibitor, CHIR-99021, rescues the expression of autophagy-related proteins and nuclear receptor GR, RORα expression in light exposure cells. (A–B) Protein content of LC3B, Beclin, SIRT1 and phospho-AMPK. GSK3 inhibitor CHIR-99021 completely rectified light-induced up-regulation of autophagy-related proteins, including LC3B, Beclin, SIRT1 and phospho-AMPK, in HT-22 cells. Values are means ± SEM, **p < 0.01, compared with Dark group, ^##p < 0.01, compared with Light group (n = 4); (C–E) Immunofluorescence of GR, showing that CHIR-99021 was able to restore light-induced GR activation and GR nuclear translocation. The nuclei were stained with Hoechst (blue) and GR was stained with GR antibody (red). Scale bars, 75 μm; (F–G) Protein content of GR, phospho-GR, RORα and REV-ERBα. CHIR-99021 significantly alleviated light-induced increase of phospho-GR and RORα protein expression, yet the increase of total GR and the decrease of REV-ERBα in Light group were not restored. Values are means ± SEM, *p < 0.05, **p < 0.01, compared with Dark group, ^#p < 0.05, ^##p < 0.01, compared with Light group (n = 3 or 4).

Figure 9. Knockdown of GSK3α significantly suppresses protein content of GSK3α, p-GR, RORα and LC3B in HT-22 cells. (A) Images of bands detected in Western blot analyses; (B) Protein content of GSK3α; (C) Protein content of phospho-GR; (D) Protein content of RORα; (E) Protein content of LC3B. Values are means ± SEM. Bars with different superscripts are significantly different from each other (p < 0.05, n = 3).

Cell culture and LED exposure

Mouse hippocampal neuron cells (HT-22) purchased from Shanghai HuiYing Biological Technology Co. Ltd. (Shanghai, China) were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) (Lot AAH204831, SH30243.01, Hyclone) containing 10 % (v/v) FBS at 37°C under 5% CO₂. Cells were divided into Dark and Light groups and cultured in constant dark or light (2500 lux white LED) condition for 24 h. The white LED light was purchased from Wangdeng Technology Co., LTD (ShenZhen, China). The CCT of the light source was 6500 K (Light spectrums in Supplementary Figure 1) and the surface power density was 15 mW/cm². After cultured under respective condition for 24 h, cells were harvested and the culture media were collected.

Serum shock procedures

Approximately 5 × 10⁵ cells/well were seeded in 6-well plates and cultured in DMEM containing 10 % (v/v) FBS at 37°C under 5% CO₂ for 24 h before the medium was replaced with serum-rich medium (DMEM, supplemented with 50% horse serum, Hyclone) for serum shock. Two hours after the serum shock (time = 0), the medium was replaced with serum-free DMEM and the cells were divided into Dark and Light groups. Cells were collected every 6 h to 54 h after light treatment. At the indicated times, the cells were washed twice with ice-cold PBS, frozen on a layer of liquid nitrogen, and kept at -80°C until the extraction of whole cell RNA.

Total RNA isolation and real-time PCR

Total RNA was extracted from the HT-22 cells using 700 μL TRIzol reagents (Cat#3101-100, Invitrogen, USA). Two micrograms of total RNA were treated with RNase-Free DNase and reverse-transcribed to cDNA using the HiScript^® Q RT SuperMix (Vazyme, China). Four microliters of diluted cDNA (1:25, vol/vol) were used for real-time PCR performed in Mx3000P (Stratagene, USA). Peptidylprolyl isomerase A (PPIA) was used as an internal control to normalize the technical variations. Data were analyzed using the method of 2^–ΔΔCT and presented relative to the Dark group. Nucleotide sequences of specific primers are shown in Table 1.

Total protein extraction and Western blotting

Whole cell lysates were extracted and the protein concentration was determined using Pierce BCA Protein Assay kit (Rockford, IL, USA) according to the product instruction. Whole cell proteins (30 µg) were loaded on a 10% SDS-PAGE gel for electrophoresis. Western blot analysis for LC3B (AP0762, Bioworld, USA, diluted 1:500), Beclin (AP0769, Bioworld, USA, diluted 1:500), ATG4c (5262, Cell Signaling Technology, USA, diluted 1:1000), ATG7 (2631, Cell Signaling Technology, USA, diluted 1:1000), GSK-3α (9338, Cell Signaling Technology, USA, diluted 1:1000), phospho-GSK-3α (9316, Cell Signaling Technology, USA, diluted 1:1000), GSK-3β (9315, Cell Signaling Technology, USA, diluted 1:500), phospho-GSK-3β (9336, Cell Signaling Technology, USA, diluted 1:500), GR (ab2768, Abcam, USA, diluted 1:500), phospho-GR (4161, Cell Signaling Technology, USA, diluted 1:1000), RORα (ab60134, Abcam, USA, diluted 1:500), REV-ERBα (ab174309, Abcam, USA, diluted 1:500), SIRT1 (ab12193, Abcam, USA, diluted 1:1000), AMPK (BS1009, Bioworld Technology USA, diluted 1:1000), phospho-AMPK (2531, Cell Signaling Technology, USA, diluted 1:1000), SQSTM1/p62 (bs-2951R, Bioss, USA, diluted 1:1000), was carried out according to the recommended protocols provided by the manufacturers, and β-actin (AP0060, Bioworld, USA, diluted 1:10,000) was used as loading control. Images were captured by VersaDoc 4000MP system (Bio-Rad, USA) and the band density was analyzed with Quantity One software (Bio-Rad, USA).

Measurement of autophagosome formation

HT-22 cells were seeded on coverslips in 12-well plates, and grown to 70% to 80% confluency before a Lipofectamine 2000-mediated transfection with the green fluorescent protein (GFP)-tagged microtubule-associated protein 1 light chain 3 beta (MAP1LC3B) expression plasmid (GFP-MAP1LC3B), which was a gift from Prof. Honglin Liu (Nanjing Agricultural University). Cells were cultured for 48 h before exposed to light treatments, and the cytoplasmic GFP-MAP1LC3B puncta were then observed under a laser-scanning confocal microscope (Carl Zeiss, Zeiss LSM 710 META). Experiments were performed in triplicate, and the punctate GFP-MAP1LC3B was counted in 3 randomly selected fields in each coverslip.

Determination of autophagy flux

To detected the effect of white LED light on autophagy flux, HT-22 cells were treated with 50 μM chloroquine (CQ, C6628, sigma, USA), a specific protease inhibitor, in the last 2 h of white LED light exposure. Then the levels of autophagy related protein LC3B and p62, as well as both the net and relative autophagy flux of LC3B and p62 were detected.

Transmission electron microscopy

HT-22 cells were treated with typsin for 5 min at 37 °C and centrifuged at 200 ×g for 5 min. After the supernatant was removed, the cell pellets were fixed in 2% glutaraldehyde in 0.1 M sodium cacodylate (NaCac) buffer, pH 7.4, postfixed in 2% osmium tetroxide in NaCac, stained en bloc with 2% uranyl acetate, dehydrated in a graded ethanol series, and embedded in Epon-Araldite resin. Thin sections were made using a diamond knife on a Leica EM UC6 ultramicrotome (Leica Microsystems, Wetzlar, Germany), collected on copper grids, and stained with uranyl acetate and lead citrate. Cells were observed under a JEM 1230 transmission electron microscope (JEOL USA, Peabody, MA) at 110 kV and imaged with an UltraScan 4000 CCD camera and First Light Digital Camera Controller (Gatan, Pleasanton, CA). Autophagic vacuoles were counted in individual cells from multiple fields and nonserial sections.

Determination of GR nuclear translocation

Sub-cellular localization of GR protein in HT-22 cells was determined through immunofluorescence staining analysis. After 24 h culture, HT-22 cells were fixed with 4% paraformaldehyde for 10 min and treated with 0.3% TritonX-100 for 30 min on coverslips. After washing with PBS for three times and blocking with 10% Fetal Bovine Serum for 10 min, HT-22 cells were incubated with GR antibodies (ab2768, Abcam, USA, diluted 1:500) at 4°C overnight. Signal of primary antibody was amplified by TRITC-labelled goat anti-rabbit IgG (ab97200, Abcam, USA, diluted 1:1000) and were visualized and imaged with a fluorescence microscope.

Knockdown of GSK-3α by shRNA transfection

To analyze the function of GSK-3α, HT-22 cells were transfected with a shRNA plasmid constructed to specifically knockdown GSK-3α (pPLK/GFP+Puro-Gsk3a shRNA-1). HT-22 cell line was maintained at 37°C in a humidified incubator containing 5% CO₂. Cells were grown in medium supplemented with 10% FBS in 6-well plates 24 h prior to transfection and the medium was replaced by fresh DMEM immediately before transfection. The transfection was conducted following the protocols of Lipofectamine 2000 Transfection Reagent (11668-019, Invitrogen, US) and Opti-MEM (31985, Gibco, USA) provided by the manufacturers.

Statistical analysis

All data are presented as means ± SEM and the differences between groups were analyzed using independent-samples T-Test with SPSS 20.0 for Windows. The differences were considered statistically significant when P < 0.05.