Male and female C57BL/6 mice (6 weeks old) were purchased from Orient Bio Inc. and maintained in a specific pathogen-free animal facility at Ewha Womans University Medical College under a 12-h light/dark cycle (8:00 AM, lights on; 8:00 PM, lights off) at 22°C ± 2°C. Experiments were performed using young adult mice (2.5 months old), and aged mice (20 months old). The blood and brains of the mice were collected under terminal sodium pentobarbital anesthesia (120 mg/kg, intraperitoneal) at 10:00 and 22:00 h for the light and dark phases, respectively. For sampling in the dark phase, the mice were anesthetized in the dark, and the blood and brain were obtained under light. Following cardiac puncture, the blood was collected in heparin-coated tubes and centrifuged for 15 min at 4,000 rpm; the separated plasma layer was transferred into a new tube and stored at −80°C. Each hemispheric brain was obtained after removal of bilateral SCN regions, which were located just above the optic chiasm [23], in order to examine diurnal gene and protein expressions in extra-SCN sites, and then stored at −80°C until analyses.

All procedures were approved by the Institutional Animal Care and Use Committee of the Medical College of Ewha Womans University (approval number: EUM18-0419) and conformed to the international guidelines for the ethical use of experimental animals.

RNA isolation: Total RNA was purified using TRIzol reagent (Ambion) according to the manufacturer's instructions. RNA quality (greater than 8 by an RNA integrity value) was assessed using an Agilent 2100 bioanalyzer with the RNA 6000 Nano Chip (Agilent Technologies, Inc.), and RNA quantification (> 20 μg of total RNA and > 2 of 260/280) was achieved using ND-2000 Spectrophotometer (Thermo Fisher Scientific Inc.).

Library preparation and sequencing: For the control and test RNAs, the construction of library was performed using QuantSeq 3′ mRNA-Seq Library Prep Kit (Lexogen Inc.) according to the manufacturer’s instructions (Ebiogen Inc.). In brief, each 500 ng total RNA were prepared and an oligo-dT primer containing an Illumina-compatible sequence at its 5′ end was hybridized to the RNA and reverse transcription was then performed. After degradation of the RNA template, second strand synthesis was initiated by a random primer containing an Illumina-compatible linker sequence at its 5′ end. The double-stranded library was purified by using magnetic beads in order to remove all of the reaction components. The library was amplified to add the complete adapter sequences required for cluster generation. The finished library is purified from polymerase chain reaction components. High-throughput sequencing was performed as single-end 75 sequencing using NextSeq 500 (Illumina, Inc.).

Data analysis: QuantSeq 3′ mRNA Seq reads were aligned using Bowtie2 [24]. Bowtie2 indices were either generated from genome assembly sequence or the representative transcript sequences for aligning to the genome and transcriptome (mm10 mouse reference genome from University of California Santa Cruz). The alignment file was used for assembling transcripts, estimating their abundances and detecting differential expression of genes. Differentially expressed genes (DEGs) were determined based on counts from unique alignments using coverage in Bedtools [25]. The read count data were processed based on the trimmed mean of M values normalization in the calcNormFactors method using EdgeR within R (R development Core Team, 2016) using Bioconductor [26]. Gene classification was based on searches performed using by DAVID (http://david.abcc.ncifcrf.gov/) and Medline databases (http://www.ncbi.nlm.nih.gov/). Fold changes were analyzed as the ratio of normalized protein-coding gene expression values (log2) between brains in the light and dark phases or between young and aged brains in the corresponding phase (log2 > 4). Statistical significance was set at p < 0.05 using the t-test functions in R. Data mining and graphic visualization were performed using an Excel-based Differentially Expressed Gene Analysis (ExDEGA) tool (Ebiogen Inc.).

Changes in the levels of cytokines and chemokines in the peripheral blood during the light and dark phases were examined using a Proteome Profiler Mouse XL Cytokine Array Kit (R&D Systems), including 111 proteins, according to the manufacturer’s protocol. Proteins in the membranes were visualized using Western Blotting Luminol Reagent (Santa Cruz Biotechnology), and ImageJ (version 1.37, National Institutes of Health) was used to measure the pixel intensity of each captured antibody spot. Density was corrected for background intensity and normalized to the positive control on the membrane. Proteins with < 10-pixel density int both light and dark phases were excluded, given that the weak densities made it difficult to determine altered expression.

Proteins from the hemispheres excluding SCN regions were isolated, and 50 μg of protein was loaded, electrophoresed, and transferred to trans-Blot TurboTM Transfer membranes (Bio-Rad) as described previously [27]. Membranes were blocked for 1 h in Tris-buffered saline (TBS) containing 0.1% Tween-20 and 10% dry milk, followed by overnight incubation with primary antibodies against flavin-containing monooxygenase (FMO) 2 (1:1,000; Proteintech), hypoxia-inducible factor (HIF) 3α (1:1,000; Proteintech), indolamine N methyltransferase (INMT, 1:1,000; Proteintech), insulin-like growth factor binding protein (IGFBP) 1 (1:500; Santa Cruz Biotechnology) and chemokine (C-X-C motif) ligand (CXCL) 13 (1:500; LS Bio). After washing, membranes were incubated with horseradish peroxidase-conjugated secondary antibodies for 2 h. Protein bands were visualized using Western Blotting Luminol Reagent (Santa Cruz Biotechnology). For quantification, the density of each band was normalized to that of actin using ImageJ (version 1.378, National Institutes of Health).

Sections of brain tissues were obtained and prepared as previously described [28]. Sections were incubated in TBS containing 0.1% Triton X-100, 5% normal serum, and 1% bovine serum albumin for 1 h and then incubated overnight with primary antibodies against FMO2 (1:200) or IGFBP1 (1:500). The next day, sections were incubated in secondary antibodies conjugated to fluorescein isothiocyanate (1:1,000; Vector Laboratories, Inc.) for 1 h. Sections were washed with TBS for all steps. Sections were then mounted with the Vectashield mounting medium containing the nuclear stain 6′-diamidino-2-phenylindole (DAPI; Vector Laboratories, Inc.), and fluorescence images were obtained using a confocal microscope (LSM5 PASCAL; Carl Zeiss).

Data of protein expression are expressed as the mean ± standard error of the mean. The distribution of data was assessed using the Shapiro–Wilk normality test. Comparisons between the two groups were performed using the unpaired Student’s t-test (cytokine array). Comparisons between more than two groups were performed using one-way analysis of variance, followed by Tukey’s post-hoc test (Western blot analysis). Statistical analyses were performed using GraphPad Prism9.0 (GraphPad Software, Inc.). Statistical significance was set at p < 0.05.

To identify genes exhibiting altered diurnal expression with aging, the brains excluding SCN regions of young and aged mice of both sexes were subjected to RNA sequencing analyses. DEGs were identified by a ≥ 2-fold change in gene expression values between the dark and light phases. In male mice, we identified 27 DEGs (20 upregulated and 7 downregulated) in young and 18 DEGs (9 upregulated and 9 downregulated) in aged brains (Fig. 1A). In female mice, we identified 56 DEGs (36 upregulated and 20 downregulated) in young brains and 11 DEGs (5 upregulated and 6 downregulated) in aged brains (Fig. 1B). Diurnal expression patterns of genes related to molecular clock, including per1, dbp and ciart, detected in young brains of both sexes, disappeared in aged brains. However, the diurnal expression patterns of fmo2 (in both sexes), hif3a (in males), and inmt (in females), which increased during the dark phase, were maintained in aged brains, although these expression levels were higher in the aged brain than in the young brain (Fig. 1A, B). In addition, eight genes (acer2, arc, ciart, dbp, fmo2, hif3a, inmt, and per1) were common diurnal DEGs in the young brains of both sexes, but only fmo2 was conserved in the diurnal pattern in the aged brains of both sexes (Fig. 1C). To determine whether diurnal expression patterns in fmo2, hif3a, and inmt observed at the transcript level were reflected at the protein levels, we examined FMO2, HIF1α and INMT protein expression. In young male and female mice, FMO2 protein levels increased during the dark phase (1.40-fold in males and 1.56-fold in females; both p = 0.01 vs. light phase; Fig. 2A, B). Immunofluorescence staining revealed elevated FMO2 expression in the cortex (Fig. 2C), hippocampus (Fig. 2D), and striatum (Supplementary Fig. 1) of young female brains during the dark phase, implying that these areas contribute to its diurnal regulation. Similarly, HIF3α in young males (1.45-fold, p = 0.035) and INMT in young females (1.35-fold, p = 0.02) also showed diurnal variations consistent with their gene expression patterns (Fig. 2A, B). However, these diurnal patterns were absent in aged brains (Fig. 2A, B), suggest that the number of molecules exhibiting shared diurnal regulation across age and sex may be even more limited at the protein level. The discrepancy between mRNA and protein levels in the aged mice may resulted from altered translational efficiency or post-translational regulation [29]. Together, these findings suggest a disruption in the diurnal regulation of gene and protein expression in the aged brain.

Several DEGs have been identified in the brains of aged rodents when compared with those expressed in the brains of young rodents [30,31]. However, whether their expression levels vary during the day and night. Therefore, we analyzed gene expression levels in the brains of young and aged mice during the light or dark phase, respectively. In the aged male brain, we identified 40 DEGs (26 upregulated and 14 downregulated) in the light phase, 32 DEGs (21 upregulated and 11 downregulated) in the dark phase, and 21 DEGs (16 upregulated and 5 downregulated) in both phases (total: 93 DEGs; Fig. 3A). Gene ontology (GO) enrichment analysis revealed that the identified DEGs were mostly associated with inflammation and immune responses (Fig. 3A). Our findings are consistent with the previous study showing that upregulated DEG pathways at all time domains of day are the immune and inflammatory responses in the hypothalamus of aged male mice [32]. In the aged female brain, we identified 35 DEGs (23 upregulated and 12 downregulated) in the light phase, 33 DEGs (21 upregulated and 12 downregulated) in the dark phase, and 25 DEGs (22 upregulated and 3 downregulated) in both phases (total: 93 DEGs; Fig. 3B). Similar to the results observed in male mice, the GO enrichment analysis revealed that identified DEGs were mainly related to inflammation and immune responses (Fig. 3B).

Next, we further analyzed the diurnal patterns of genes involved in inflammation and immune responses (28 genes in both sexes) among the 93 DEGs. In the aged male brain, expression levels of 20 genes were phase-specifically upregulated (13 genes in the light phase and 7 genes in the dark phase), and the expression levels of eight genes (aim2, c4a, c4b, ccl21b, ccl21c, defb1, H2-Ab1, and slc7a5) were increased in both phases (Fig. 4A, B). In the aged female brain, expression levels of 18 genes were phase-specifically increased (8 genes in the light phase and 10 genes in the dark phase), and the expression levels of 10 genes (agpat2, c4a, c4b, ccl21b, ccl21c, clec7a, defb1, lyz2, slc11a1, and slc7a5) were increased in both phases (Fig. 4C, D). Six genes (c4a, c4b, ccl21b, ccl21c, defb, and slc7a5) were commonly increased in the aged brains of both sexes during the light and dark phases (indicated by upright grey triangles in Fig. 4), with the ccl21 gene (ccl21b and ccl21c) exhibiting the highest increase in fold changes in the aged brain.

These results show that genes involved in the inflammatory process are increased in the aged brain compared with those in the young brain, in both the light and dark phases, indicating a heightened inflammatory state in the aged brain with different molecules between phases and sexes.

Age-associated inflammation can be determined by detecting levels of inflammatory mediators in the peripheral blood, and these levels are known to exhibit diurnal variations [33,34]. To determine whether the expression changes of inflammatory molecules in the aged brain were systemic and general responses with aging, we analyzed plasma levels of cytokines in young and aged mice during the light and dark phases using a cytokine array. Differentially expressed proteins were identified by a ≥ 2-fold change in pixel density values between the dark and light phases of young or aged mice, respectively.

In young mice of both sexes, there were no diurnal changes in cytokine expression. In aged mice of both sexes, the expression of IGFBP1 was reduced in the dark phase (0.39-fold for males and 0.31-fold for females, p = 0.06 and p < 0.01 vs. light phase, respectively; Fig. 5A, B), and other cytokines did not exhibit more than a two-fold change in expression levels in aged male mice. In aged female mice, expression levels of CXCL13 (0.32-fold reduction, p < 0.01), IL-7 (2.11-fold increase, p = 0.15), and receptor for advanced glycation end products (2.03-fold increase, p = 0.08) were altered in the dark phase when compared to those detected in the light phase (Fig. 5B).

Next, cytokine levels were compared between young and aged mice in the light and dark phases, respectively. Compared with young male mice, aged male mice exhibited altered expression levels of CD105 (0.48-fold decrease, p = 0.01), CXCL5 (0.29-fold decrease, p < 0.05), CXCL13 (7.42-fold increase, p = 0.06), IL-28 (0.42-fold decrease, p = 0.13) in the light phase, while IGFBP1 (0.27-fold decrease, p < 0.05) and IGFBP3 (0.50-fold decrease, p = 0.03) levels were altered in the dark phase (Fig. 5A, C). During both the light and dark phases, aged male mice displayed a decrease of CD93 levels (0.34-fold in the light phase and 0.40-fold in the dark phase, both p < 0.05) and a marked increase in CXCL13 levels (7.42-fold with p = 0.06 in the light phase and 7.19-fold with p = 0.03 in the dark phase) compared with young male mice (Fig. 5A, D). Compared with young female mice, aged female mice displayed no change in cytokine levels during the light phase, while complement component C5/C5a (2.01-fold increase, p = 0.10), IGFBP1 (0.25-fold decrease, p = 0.01), and lipocalin 2 (2.05-fold increase, p < 0.01) levels were altered in the dark phase (Fig. 5B, C). Increased levels of CXCL13 in aged female mice in both phases (6.66-fold in the light phase and 1.97-fold in the dark phase, both p < 0.01 vs. the respective phase of young female mice; Fig. 5B, D), similar to those detected in aged male mice. Unlike the increased CCL21 gene expression in the aged brain, blood levels of CCL21 did not differ between young and aged mice (Supplementary Fig. 2). These findings indicate that the brain and peripheral blood exhibit distinct age-related alterations in cytokine expression during the light and dark phases.

Common findings in the peripheral blood of aged mice of both sexes include a decrease in IGFBP1 expression in the dark phase and an increase in CXCL13 expression in both phases. Finally, to determine whether these changes also occurred in the brain as general aging responses, we examined IGFBP1 and CXCL13 protein levels. Unlike the blood, IGFBP1 levels were increased in the aged brains of both sexes, independent of phase (more than 1.86-fold for both male and female, p < 0.05 vs. corresponding phases of young mice). CXCL13 levels also differed between the blood and brain. CXCL13 did not increase in the aged brain of either sex. Especially, its levels decreased in the aged male brain during the dark phase (0.47-fold, p < 0.01 vs. that of young male brains; Fig. 5E, F).

Given the marked increase of IGFBP1 levels in aged female mice during the dark phase, immunofluorescence staining was performed to identify brain regions showing elevated expression in aged female mice at this time point. IGFBP1 expression was increased in multiple regions, including the cortex, striatum and corpus callosum in the aged female mice, compared with young female mice (Fig. 6), indicating that these brain regions are, at least in part, responsible for the age-related increase in IGFBP1 expression throughout the day. Taken together, these results suggest that distinct inflammatory molecules exhibit different diurnal expression patterns in the brain and peripheral blood with aging.