Dim Light at Night Impairs Daily Variation of Circulating Immune Cells and Renal Immune Homeostasis

Male Wistar rats were obtained at the age of 14 weeks from the breeding station of the Institute of Experimental Pharmacology and Toxicology, Slovak Academy of Sciences (Dobrá Voda, Slovak Republic). The animals were housed in plastic cages in groups of three to four rats at an ambient temperature of 21.5 ± 1.3°C and humidity of 55%–65%, and they were provided with a standard pelleted diet and water ad libitum. During the acclimation period of four weeks, rats were maintained on a LD cycle of 12:12 h with lights on at 10:00 am, designed as Zeitgeber time 0 (ZT0). At the level of animal cages, room lights emitted white light with an illumination of 150–200 lx and color temperature of 2,900 K.

After acclimation, rats (311 ± 28 g) were assigned to either the control group (CTRL, n=15) with the standard lighting regime described above, or to the experimental group (dLAN, n=18), which was exposed to low-illuminance levels of 2 lx during the entire night phase. The dim light conditions were provided by a shaded lamp with LED bulb Star Classic A60 10W (OSRAM GmbH, Germany), which emitted a broad-spectrum white light with a peak of 610 nm and color temperature of 2,700 K. Illuminance and color temperature was measured at the level of cages using an illuminance spectrophotometer CL-500A (Konica Minolta Sensing Europe BV, Germany).

To examine the numbers and daily pattern of circulating leukocyte populations, rats (CTRL, n=7; dLAN, n=9) were repeatedly blood sampled from the tail vein at ZT9 and ZT21 after 2 and 5 weeks of the experiment. At each time point, sampling was completed within 1 h for all animals. These time points were chosen to encompass night-time melatonin peak, which occurs in the second half of the dark phase in rats (26). Plasma melatonin levels from this experiment have already been reported in our recent paper (14). Rats were immobilized under isoflurane anesthesia, and blood was collected from a lateral tail vein into tubes with either ethylenediaminetetraacetic acid (for immunophenotypic analysis) or heparin (for analysis of the oxidative burst of neutrophils).

After 2 weeks (CTRL, n=7; dLAN, n=8) and 5 weeks (CTRL, n=8; dLAN, n=10) of the experiment, rats were sacrificed under isoflurane anesthesia during the first half of the light phase (between ZT3 and ZT6). Blood was collected into heparin tubes and centrifuged (2,500 g, 10 min, 4°C). The separated plasma was stored at –20°C until performing the corticosterone assay. The left kidney was dissected and cut into three pieces; the middle part was processed for immunofluorescence and both ends were immediately frozen in liquid nitrogen and stored at –76°C until molecular analyses.

A suspension of blood leukocytes in phosphate-buffered saline (PBS) was obtained after the lysis of red blood cells using ammonium–chloride–potassium buffer. Thereafter, leukocytes were fixed in 4% phosphate-buffered formaldehyde (Ultra Pure, Polysciences, USA; 87001-890) and washed, and the pellets were diluted to the original blood volume in PBS supplemented with 0.5% bovine serum albumin (BSA) and 0.1% sodium azide (staining buffer). Aliquots of the leukocyte suspension (50 µl) were quadruple or triple stained with a cocktail of antibodies, incubated for 30 min at 4°C in the dark, and analyzed on a BD Accuri C6 cytometer (BD Bioscience, USA). The following fluorochrome-conjugated monoclonal antibodies were used: PE-Cy7 anti-rat CD45 (clone OX-1; Sony Biotechnology, USA; 1611065), FITC anti-rat CD3 (clone G4.18; Thermo Fisher Scientific, USA; 11-0030), APC anti-rat CD4 (clone OX-35; Thermo Fisher Scientific; 17-0040), PE anti-rat CD8a (clone OX-8; Thermo Fisher Scientific; 12-0084), PE anti-rat CD45RA (clone OX-33; Thermo Fisher Scientific; MR6404), APC anti-rat CD161a (clone 10/78; Thermo Fisher Scientific; MR6805), FITC anti-rat granulocytes (clone HIS48; Thermo Fisher Scientific; 11-0570), and PE anti-rat CD43 (clone W3/13; Sony Biotechnology; 1614060). The following rat leukocyte subsets were identified on the basis of their surface markers in the gate for total leukocytes (CD45⁺): T-cells (CD3⁺), helper T cells (CD3⁺CD4⁺), cytotoxic T cells (CD3⁺CD8a⁺), B cells (CD45RA⁺), NK cells (CD3^–CD161a⁺), neutrophils, and monocytes (HIS48⁺ and side scatter gating). Reciprocal expression of CD43 and HIS48 markers was used to identify classical (CD43^loHIS48^hi) or non-classical monocytes (CD43^hiHIS48^lo) (27). The gating strategy for identifying individual populations of peripheral blood leukocytes is given in Figure S1. All cytometric data were analyzed using FlowJo software (TreeStar, USA).

The oxidative burst of blood neutrophils was evaluated in rats after 5 weeks of dLAN. In this procedure, 50 µl of whole blood was diluted 20 times with PBS containing 20 µM of 2′,7′-dichlorodihydrofluorescein diacetate (H₂-DCF-DA; MilliporeSigma, USA; D6883) and incubated for 30 min at 37°C in the dark. Thereafter, each sample was divided into two equal parts; one part was stimulated with phorbol-12-myristate-13-acetate (PMA; MilliporeSigma; P8139) at a final concentration of 1 µM and one part served as the unstimulated control (dimethyl sulfoxide was added at a final concentration of 0.1%). After incubation (for 30 min at 37°C in the dark), the cells were washed with PBS and red blood cells were lysed using eBioscience™ RBC Lysis Buffer (Thermo Fisher Scientific; 00-4333-57). After lysis of erythrocytes, the cells were washed, resuspended in 0.3 ml of PBS, and analyzed on a flow cytometer. Within the cell, H₂-DCF-DA is converted to fluorescent 2′,7′-dichlorofluorescein (DCF), and this fluorescence is a measure of H₂O₂ production and is linearly related to the oxidative burst of the stimulated neutrophils (28). The functional activity of the neutrophils was expressed as a fold increase in the median DCF fluorescence of the PMA stimulated samples over the unstimulated samples.

The kidney samples were washed in cold PBS and fixed in 4% phosphate-buffered formaldehyde overnight at 4°C. Then, the samples were cryoprotected in 1.1 M sucrose, embedded in OCT Cryomount (Histolab Products AB, Sweden; 45830), and stored at –76°C. Frozen kidney sections (8 µm) were prepared on adhesive slides, rehydrated, and incubated with 50 mM NH₄Cl for 30 min. Thereafter, sections were permeabilized with 0.25% Triton X-100 in PBS for 10 min and blocked with 5% goat serum in PBS for 1 h at room temperature. To identify macrophages, sections were stained with mouse anti-rat CD68 (Bio-Rad, USA; 1:100 dilution; MCA341R) overnight at 4°C. Next, the sections were incubated with an Alexa Fluor 660-conjugated secondary antibody (Thermo Fisher Scientific; 1:750 dilution; A-21055) for 1 h at room temperature. Cell nuclei were counterstained with DAPI (Roche, USA; 1:10,000 dilution; 10236276001). The CD68-positive cells were analyzed using a Zeiss Axioscope (Carl Zeiss, Germany) fluorescence microscope. The cells were counted in 30 randomly selected images of the renal cortex and averaged from 2 sections per individual. The cell numbers were calculated per mm².

The samples of the renal cortex were homogenized on ice in 0.05 M Tris-HCl buffer (pH 7.4) supplemented with protease inhibitors. The protein concentrations were determined with the BCA assay kit (MilliporeSigma; 71285). Total proteins (20 µg) were separated with 12% SDS-PAGE and transferred to a nitrocellulose membrane. Membranes were blocked with 5% BSA in Tris-buffered saline containing Tween 20 (TBS-T). Afterwards the membranes were incubated with mouse anti-3-nitrotyrosine antibody (Abcam, 1:1,500 dilution; ab61392) or mouse anti-rat CD68 antibody (Bio-Rad; 1:1000 dilution; MCA341R) overnight at 4°C. Following three washes (3×10 min) with TBS-T, the membranes were incubated with horseradish peroxidase-conjugated secondary antibody (Cell Signaling Technology, USA; 1:2,000 dilution; 7076) for 1 h at room temperature. Both primary and secondary antibodies were diluted in TBS-T containing 1% BSA. All blots were reprobed with mouse anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody (clone 6C5; MilliporeSigma; 1:2,000 dilution; MAB374) and incubated for 1 h at room temperature. The signals were visualized with Clarity Western ECL substrate (Bio-Rad; 1705061) using ChemiDoc™ Touch Imaging System (Bio-Rad) and quantified with Image Lab Software. The amount of target proteins was normalized relative to GAPDH and presented in arbitrary units (a.u.).

The total superoxide dismutase (SOD) activity was measured in 0.5% homogenates of renal cortex prepared in 0.05 M Tris-HCl buffer (pH 7.4) containing protease inhibitors with the SOD assay kit (MilliporeSigma; 19160-1KT-F), according to the manufacturer’s instructions. The tissue SOD activity was calculated using a SOD standard curve (MilliporeSigma; S7571-15KU) and expressed in units per milligram of proteins (U/mg).

Plasma corticosterone levels were determined by a radioimmunoassay (RIA) using the Corticosterone Rat ¹²⁵I RIA kit (DRG Instruments GmBH, Germany; RIA-1364), according to the manufacturer’s instructions.

Tissue samples of the renal cortex were homogenized using a FastPrep instrument (MP Biomedicals, Germany), and total RNA was isolated with an Rneasy Plus Universal Mini kit (Qiagen, Germany; 73404), according to the manufacturer’s instructions. Before reverse transcription, genomic DNA was eliminated by DNase I (Thermo Fisher Scientific, USA; EN0521). For synthesis of complementary DNA, a Maxima First Strand cDNA Synthesis Kit for RT-qPCR (Thermo Fisher Scientific; K1641) was used. The quantity and purity of the isolated RNA were measured with a NanoDrop One spectrophotometer (Thermo Fisher Scientific). The RNA integrity was verified with 1.2% agarose gel electrophoresis. Amplification of cDNA was performed with Maxima SYBR Green qPCR Master Mix (Thermo Fisher Scientific; K0223) and the CFX Connect real-time PCR detection system (Bio-Rad). The relative expression of the target and reference genes was calculated using a standard curve method. The expression of the target genes was normalized to the expression of the ribosomal protein S29 (Rps29). Primer sequences are listed in Table S1.

Data are presented as means ± standard error of mean (SEM). Statistical analyses were performed in R (R Core Team, Vienna, Austria). Two-way analysis of variance (ANOVA) with repeated measures and Tukey post hoc tests were used to analyze differences in the numbers of individual white blood cell (WBC) subsets between the CTRL and dLAN groups and between two ZT points. For other parameters, CTRL and dLAN groups were compared by the Student’s t-test or the Mann–Whitney U test, depending on the normal distribution. A p-value of <0.05 was considered to be statistically significant.

In blood, leukocytes exhibit a pronounced daily rhythmicity, which is, for most of the leukocyte types in rats and mice, characterized by a peak at the beginning of the passive (light) phase and a trough at the beginning of the active (dark) phase (29). To test whether dLAN can influence the number of immune cells in blood and their daily oscillations, we evaluated the immunophenotype of circulating WBCs at ZT9 and ZT21 in rats exposed to dLAN (~2 lx) for 2 and 5 weeks. Rats in the control regime had lower numbers of blood monocytes and T cells at ZT21 compared to ZT9, whereas the opposite pattern was observed for neutrophils (Figure 1A). After 5 weeks of dLAN, the daily variation was lost or disturbed for most types of WBCs, except neutrophils (Figure 1A).

Next, we analyzed whether dLAN affects the absolute number of leukocyte populations in the blood. No differences between the CTRL and dLAN rats were recorded for the total leukocyte count after either 2 or 5 weeks of dLAN (Figure 1A). However, 2 weeks of dLAN resulted in a lower number of monocytes and a higher number of T cells, though the increase in T cells was specific for ZT21 (Figure 1A). Both CD4⁺ helper and CD8⁺ cytotoxic T cells displayed the same significant changes as the total T cell numbers (Figure 1B). Moreover, there were no differences between the CTRL and dLAN rats in the CD4/CD8 ratio, indicating that both T cell subsets responded to dLAN in the same way. A day–night difference was also recognized for the CD4/CD8 ratio, with higher levels at ZT21 compared to ZT9, and this daily variation was impaired in rats exposed to dLAN for 5 weeks (Figure 1B).

Because the circulating monocytes are phenotypically and functionally distinguishable into several populations (30), we examined whether the dLAN-induced decrease in total blood monocyte numbers could be attributed to the changes in classical or non-classical monocytes. In rats, reciprocal expression of CD43 and HIS48 can be used to identify classical (CD43^loHIS48^hi) or non-classical (CD43^hiHIS48^lo) monocytes (27), which are analogous to murine Ly6C^hi and Ly6C^lo monocyte subsets (30). After 2 weeks of dLAN, the decline in total monocyte number was mainly due to a decrease in the population of classical CD43^loHIS48^hi monocytes (Figure 1C). On the other hand, the loss of daily variation following 5 weeks of the dLAN regime was mainly explained by the population of non-classical CD43^hiHIS48^lo monocytes, though the pattern was similar in both monocyte subsets (Figure 1C).

Collectively, these results indicate that chronic exposure of rats to dLAN can impair daily variation in the main leukocyte populations in the blood.

Based on the dLAN-induced changes in the immune cell numbers in the circulation, we expected the effects on the recruitment of immune cells into tissues. Kidney-resident and infiltrating immune cells are increasingly recognized to play a central role in tissue homeostasis as well as in the modulation of many pathophysiological conditions (31, 32), and recent data show that renal immune pathways can be activated by the disturbed circadian clock (23). Therefore, we quantified specific markers for macrophages and T cells, including chemokines and cell adhesion molecules, in the renal cortex of rats exposed to dLAN for 2 and 5 weeks. Following 2 weeks of the dLAN regime, the rats showed higher renal mRNA levels of a common macrophage marker, Cd68, compared to controls (Figure 2A), though no differences between groups were detected at the protein level analyzed by Western blot (Figures 2B, C) and in the CD68+ cell counts analyzed by immunofluorescence staining (Figures 2D, E) after both 2 and 5 weeks of dLAN. No differences between dLAN and CTRL groups were found in the gene expression of toll-like receptor 4 (Tlr4), C-C chemokine receptor type 2 (Ccr2) and the macrophage scavenger receptor (Cd36) (Figure 2A). Moreover, renal mRNA levels of a specific marker for T cells, Cd3d, were not affected by dLAN (Figure 2F).

Next, we analyzed the renal expression of chemokines and cell adhesion molecules, which mediate monocyte and lymphocyte trafficking into tissues. In both steady-state and disease conditions, the important monocyte-attracting chemokine is C-C motif chemokine ligand 2 (CCL2) (33, 34). We found that rats exposed to dLAN for 2 weeks displayed lower Ccl2 mRNA levels in the kidney compared to controls (Figure 2G). dLAN did not affect the gene expression of C-C motif chemokine ligand 5 (Ccl5), intercellular cell adhesion molecule-1 (Icam1), and vascular cell adhesion molecule-1 (Vcam1) (Figures 2G, H).

Because oxidative stress is an important factor that can influence the recruitment of immune cells into tissues (32), we examined whether dLAN can impact the redox balance in the kidney. Interestingly, rats exposed to dLAN for 2 and 5 weeks showed lower 3-nitrotyrosine (3-NT) protein levels in the renal cortex compared to controls (Figures 3A, B). Nitrotyrosine can be formed through nitration of tyrosine residues on proteins and is often measured as a potent indicator of endogenous nitrating agents, especially peroxynitrite, which is known to alter protein structure and function (35). Peroxynitrite is generated by the reaction between nitric oxide (NO) and superoxide anion, and therefore, we analyzed gene levels of enzymes involved in these redox pathways. After 5 weeks, dLAN-exposed rats displayed higher mRNA levels of endothelial NO synthase (eNos) in the kidney compared to controls, while renal gene expression of neuronal NO synthase (nNos) was not affected by dLAN (Figure 3D). No differences between CTRL and dLAN groups were found in anti-oxidative enzymes, such as total SOD activity in the kidney (Figure 3C) and renal mRNA levels of Sod3 and heme oxygenase-1 (Hmox1) (Figure 3D).

Next, we quantified the gene expression of an important redox-sensitive transcription factor and transcription coactivator in the renal cortex. Following 5 weeks of the dLAN regime, rats displayed higher mRNA levels of nuclear factor erythroid 2-related factor 2 (Nrf2) and peroxisome proliferator-activated receptor gamma coactivator-1 alpha (Pgc-1a) compared to controls (Figure 3E), suggesting a disturbance in the redox balance and up-regulation of the main endogenous antioxidant pathways. We found no effects of dLAN on other transcription factors involved in renal metabolism, peroxisome proliferator-activated receptor alpha (Pparα) and peroxisome proliferator-activated receptor gamma (Pparγ) (Figure 3E).

Oxidative burst in immune cells is an important mechanism that leads to the generation of reactive oxygen and nitrogen species (28). Therefore, we tested the capacity of the circulating neutrophils to induce oxidative burst in response to PMA stimulation in rats exposed to dLAN for 5 weeks. No differences between the dLAN and CTRL rats were detected in the oxidative burst of blood neutrophils (Figures 4A–C), indicating that NADPH oxidase, the enzyme responsible for the production of superoxide, was not affected by dLAN in blood neutrophils. Next, we recorded significant daily variation in the oxidative burst of neutrophils, with lower levels at ZT21 compared to ZT9, and this variation remained unaffected in the dLAN regime (Figure 4C).

In order to explore the link between dLAN and the circadian clock in the kidney, we analyzed molecular clock components (gene expression of Bmal1 and Rev-erbα) in the renal cortex of rats exposed to dLAN for 2 and 5 weeks. Bmal1 and Rev-erbα represent core clock components of two interconnected feedback loops, which are able to generate and regulate circadian periodicity (3). Moreover, both BMAL1 and REV-ERBα are directly involved in transcriptional control of prominent immune and redox regulators, including Ccl2 (20, 36) and Nrf2 (37). Following 2 and 5 weeks of the dLAN regime, the rats showed lower Bmal1 mRNA levels compared to controls (Figures 5A, B) and, simultaneously, 2 weeks of dLAN resulted in increased Rev-erbα mRNA levels in the kidney (Figure 5A). These results show that dLAN altered gene expression of important components of the molecular clockwork, implying changes in daily oscillations of renal genes that are under direct circadian control.

Next, we measured plasma corticosterone concentrations because glucocorticoids are important systemic humoral signals that represent not only potent synchronizers of the peripheral circadian clocks (38) but also regulators of immune cell trafficking into tissues in both homeostatic and inflammatory states (39). We found that rats exposed to dLAN for 2 and 5 weeks showed higher plasma corticosterone levels compared to controls (Figure 6), suggesting that glucocorticoid signaling is disturbed under dLAN conditions.

The datasets generated for this study are available on request to the corresponding author.