Chronic sleep deprivation is associated with delayed puberty onset in rats, activation of proinflammatory cytokines and gut dysbiosis

Animal study

We conducted this study to investigate the impact of sleep deprivation on altering the puberty onset in juvenile Sprague-Dawley (SD) rats. Four pregnant SD rats were purchased from BioLASCO Taiwan Co. Ltd and housed at the Laboratory Animal Center at Taipei Medical University in a controlled environment (12-hour light-dark cycle, 22–24 °C, 40%–60% humidity). The juvenile SD rats were weaned and grouped at postnatal day 21 (PND 21). Then, they were randomly assigned to control female (CF) (n = 6), sleep-deprivation female (SDF) (n = 6), control male (CM) (n = 6), and sleep-deprivation male (SDM) (n = 6) groups, a total of 24 rats. We ensured their body weight and sex were balanced across groups. They were subjected to 15 h of sleep deprivation per day for 4 weeks after weaning. Sleep deprivation is a highly stressful condition; therefore, the body weights of rats were monitored every other day, and all rats survived until euthanization. The rats were euthanized by using cardiac puncture after 4 weeks of sleep deprivation, and blood samples and tissues were collected and stored in the −80 °C refrigerator until used.

CSD model

The rats in the SDF and SDM groups were subjected to sleep deprivation for 15 h (from 08:30 to 23:30) per day for 4 weeks continuously based on the modified inverted flowerpot method (Machado et al., 2004). Sleep deprivation for 4 weeks was considered as CSD (Poroyko et al., 2016; Zhu et al., 2015). The SD rats were placed inside a water tank containing eight circular platforms, each with a diameter of six cm. They were put in the tanks with the same sex and group-housed after they returned to the home cage. Water was added to within one cm of the upper surface of each platform. Each water tank had at most six rats. When the rats reached the rapid eye movement stage of sleep, muscle atonia caused them to fall into the water, at which point they had to climb up a platform to avoid being drowned. The control rats were housed in standard cages in the same animal room under identical light-dark cycles, temperature, humidity, and handling frequency. However, they were not placed in water tanks or exposed to the inverted flowerpot environment, which may have resulted in different levels of physical or psychological stress compared to the SD groups. This difference in environmental exposure is acknowledged as a limitation of the study.

Assessment of the onset of puberty

The onset of puberty in the female rats was started observed at PND25 and determined by using vaginal smears. Vaginal smears were performed by inserting a sterilized pipette tip filled with 10 µL of normal saline into the vagina and then taking the fluid. The fluid was placed in the glass slide and observed under microscope vision. Preestrous, estrous, metertrous, and diestrous of estrous cycle stages were identified by cell morphology (Marcondes, Bianchi & Tanno, 2002). In addition, the onset of puberty in the male rats was started at PND 35 and determined by preputial separation (Korenbrot, Huhtaniemi & Weiner, 1977). We pre-specified puberty-onset timing as our primary hypothesis; all other comparisons are exploratory and unadjusted for multiplicity.

Tissue preparation

Colon (cecum), ovary, and testes samples were dissected and washed immediately in 0.1 M phosphate buffer saline (PBS). They were homogenized on ice in PBS 1:2 (w/v; one g tissue with three mL PBS, pH 7.4) and centrifuged at 10,000× g for 15 min at 4 °C. The supernatants were collected to determine CAT, SOD, and glutathione GPx activity.

Protein determination

The protein levels of the colon, ovary, and testes homogenates were determined by using the Bradford method (Bradford, 1976).

Antioxidant enzyme activities

Antioxidant enzyme activities were determined by using CAT, SOD, and GPx activities. Antioxidant enzyme activities were analyzed by using assay kits from Cayman Chemical (Ann Arbor, MI, USA). Analyzed procedures followed the manufacturer’s protocols: CAT (item no. 707002), SOD (item no. 706002), and GPx (item no. 703102). Antioxidant enzyme activities were normalized to the total protein in the homogenates and expressed as units per mg of protein.

Circulating lipopolysaccharide-binding protein (LBP) levels and proinflammatory cytokines (IL-1β, IL-6, and TNF-α) were quantified in rat plasma samples. Plasma LBP levels were measured using an enzyme-linked immunosorbent assay (ELISA) kit (Cusabio; CSB-E11184r). Briefly, 100 µL of plasma samples and standards were added to a 96-well plate pre-coated with LBP-specific antibodies and incubated for 2 h at 37 °C. After washing, an HRP-conjugated secondary antibody was added and incubated for 1 h. TMB substrate was then applied, and the color reaction was stopped with a stop solution. Absorbance was measured at 450 nm using a microplate reader.

IL-1β, IL-6, and TNF-α levels were assessed using the LEGENDplex™ Multi-Analyte Flow Assay Kit (Cat. 741395 and 741396; BioLegend, San Diego, CA, USA) based on flow cytometry. In brief, 25 µL of each plasma sample was incubated with a mixture of capture beads coated with cytokine-specific antibodies. After a 2-hour incubation at room temperature, biotinylated detection antibodies were added, followed by streptavidin-PE. Samples were analyzed on a flow cytometer, and cytokine concentrations were calculated using standard curves generated with LEGENDplex™ Data Analysis Software Suite (BioLegend, San Diego, CA, USA).

RNA extraction and real-time quantitative reverse transcription-polymerase chain reaction

The RNA was extracted by using the RNeasy Mini Kit (Qiagen, Hilden, Germany) in accordance with the manufacturer’s protocol. Following RNA isolation, one µg of RNA was used for reverse transcription to cDNA by using the MMLV Reverse Transcription Kit (Protech Technology Enterprise, Co., Ltd., Taipei, Taiwan). cDNA was used to quantify the transcript levels on the Smart Quant Green Master Mix system (Protech Technology Enterprise, Co., Ltd., Taipei, Taiwan). GAPDH is used as an internal control (Ct of target gene−Ct of GAPDH = ΔCt), and ΔCt of control is used as the calibrator (ΔCt of sample−ΔCt of calibrator = ΔΔCt). Relative mRNA levels of target genes = 2^−ΔΔCT (fold change vs. control).

DNA sequences of rat-specific primers were summarized as follows:

Gut microbiome composition

Fecal samples were collected at vaginal opening days (PND 30∼PND 40) from female rats and were collected at preputial separation days (PND 39∼PND 49) from male rats during the experiment. In brief, collected fecal samples were transferred immediately to cold storage and remained stored at −80 °C until processing near days of vaginal opening. Fecal genomic DNA was extracted using the QIAamp DNA Stool Mini Kit (cat. no. 51504, Qiagen, Hilden, Germany) according to the manufacturer’s instructions, stored at −80 °C, and underwent processing including polymerase chain reaction (PCR) assays and 16S rRNA sequencing. The PacBio sequencing for full-length 16S genes (V1–V9 regions) was performed. The full-length 16S genes was amplified using barcoded 16S gene-specific primers. Subsequently, the PCR reaction was carried out by KAPA HiFi HotStart ReadyMix (Roche, Basel, Switzerland), and its products were purified using the AMPure PB Beads for SMRTbell library construction and sequencing processes. Consequently, multiple sequence alignment was performed by QIIME2 alignment MAFFT against the NCBI database to analyze the sequence similarities among the amplicon sequence variants (ASVs).

Operational taxonomic unit (OTU) clustering and taxonomic analysis were performed using Genomics workbench v.22.0 (CLC Bio, Denmark). The sequences were trimmed, merged, and clustered into OTUs at 97% sequence similarity based on the SILVA v.32 database using CLC Microbial Genomics Module. Within sample microbial diversity (alpha diversity) were assessed using the phyloseq package in R from normalized operational taxonomic unit (OUT) abundances. Between group diversity (beta diversity) was calculated by evaluating the unique taxa present in each group relative to the shared taxa across both. We used the principal coordinate analysis (PCoA) to explore the inter-sample variation, grounded in Bray–Curtis dissimilarity. For unsupervised grouping of the samples, hierarchical clustering was executed employing Bray–Curtis distance with a complete linkage method. LEfSe analysis was performed to detect bacterial taxa with significantly different abundance between the control and sleep deprivation groups; significance was indicated if the linear discriminant analysis value was >2.0 with p < 0.05.

Statistical analysis

Values are presented as mean ± standard error of the mean, and all data were examined the normality by using Shapiro–Wilk test. A student’s t-test was performed to compare two groups. ANOVA was performed for three groups, following Tukey’s multiple comparisons test via GraphPad Prism 8.0.1 software. Heatmaps were plotted via R version 4.0.3 (R Foundation for Statistical Computing, Vienna, Austria), which showed Spearman’s rank correlation coefficient between the abundance of bacterial taxa and gene/protein levels. Differences were considered significant at p < 0.05. Data processing and visualization were performed by using the heatmap package (version 1.0.12), along with phyloseq (1.34.0), ggplot2 (3.3.3), and dplyr (1.0.2). Correlation analyses were based on Spearman’s rank correlation using the cor.test() function in base R, and microbiome data structures were managed using the phyloseq package.

Ethical approval

The Taipei Medical University Institutional Animal Care and Use Committee (IACUC/IACUP) approved all animal procedures (approval no. LAC-2020-0048). All procedures were conducted in accordance with the Taiwan Code of Practice for the care and use of animals for scientific purposes.

CSD attenuated growth status and delayed onset of puberty in juvenile rats

Body weight and pubertal timing in rats were monitored during experiment. The results revealed that CSD for 4 weeks is associated with a decrease in body weight in female (SDF:151.90 ± 3.84 g vs. CF:168.6 ± 3.33 g) and male rats (SDM: 197.30 ± 2.72 g vs. CM: 218.90 ± 4.82 g) (Fig. 1A); in addition, pubertal timing also delayed in female rats (SDF: 36.33 ± 1.17 day vs. CF: 31.00 ± 0.89 day) and male rats (SDM: 45.17 ± 0.65 vs. CM: 40.33 ± 0.76 day) (Fig. 1B). Biochemical characteristics and organs/tissues weight analysis revealed that CSD significantly attenuated blood levels of total protein and albumin in female and male rats but significantly increased levels of triglyceride and aspartate aminotransferase (AST) in male rats (Table S1); in addition, CSD significantly attenuated weights of muscle in female rats and significantly attenuated brain, muscle epididymal white adipose tissue, liver, kidney, seminal vesicle, and epididymis in male rats (Table S2). Taken together, CSD for 4 weeks significantly attenuates growth status and delays the onset of puberty in both female and male rats. In our model, this delayed pubertal timing may be partly attributed to reduced somatic growth, systemic inflammation, and oxidative stress—factors known to influence the hypothalamic-pituitary-gonadal (HPG) axis.

CSD increases antioxidant enzyme activities in reproductive organs in female and male rats

To determine the effects of CSD on antioxidant responses, we investigated the antioxidant enzyme activities in the colon and reproductive organs. In the colon organ, we observed that the antioxidant enzymes as CAT, SOD, and GPx activities were not significantly changed between CF and SDF rats (Fig. 2A, superior section); however, CAT (102.50 ± 15.56 nmol/mg protein/min), SOD (22.27 ± 2.70 U/mg protein), and GPx (158.20 ± 10.31 nmol/mg protein/min) in SDM rats were significantly higher than those of CM rats (Fig. 2A, inferior section). In reproductive organs, we observed that CAT (20.66 ± 1.63 nmol/mg protein/min) and SOD (1.18 ± 0.05 U/mg protein) activities in SDF rats were significantly higher than those of CF rats (Fig. 2B, superior section); in addition, CAT (13.75 ± 0.77 nmol/mg protein/min), SOD (0.925 ± 0.03 U/mg protein) and GPx activities (114.4 ± 3.88 nmol/mg protein/min) in SDM rats were significantly higher than those of CM rats (Fig. 2B, inferior section). Taken together, CSD is associated with an increase in antioxidant enzyme activities in reproductive organs of both sexes; however, antioxidant enzyme activities in colon organs have only increased in male rats.

CSD leads to inflammation in the colon, reproductive organs, and circulatory system in female and male rats

Next, we investigated whether CSD is leading to an inflammatory response in female and male rats’ colon, reproductive organs, and circulatory systems. Therefore, we determined protein levels of lipopolysaccharide-binding protein (LBP), IL-1β, IL-6, and TNF-α in the circulatory system and mRNA levels of those in the colon and reproductive organs.

The results revealed that the protein levels of LPS (2,424 ± 153.2 pg/ml), IL1-β (0.132 ± 0.048 pg/ml), IL-6 (10.64 ± 2.85 pg/ml), and TNF-α (1.06 ± 0.11 g/ml) in the plasma of SDF rats were significantly higher than those of CF rats (Fig. 3A, superior section). The protein levels of IL-6 (13.32 ± 2.93 pg/ml) in SDM rats’ plasma were significantly higher than those of CM rats (Fig. 3A, inferior section). In the colon, mRNA levels of TNF-α in SDF rats were considerably higher than those of CF rats (2.86 ± 0.33 old vs. CF); mRNA levels of IL-1β in SDM rats were significantly higher than those of CM rats (0.98 ± 0.24 fold vs. CF) (Fig. 3B). In reproductive organs, mRNA levels of IL-1β in SDF rats were significantly higher than those of CF rats (1.62 ± 0.17 fold vs. CF); mRNA levels of TNF-α in SDM rats were significantly higher than those of CM rats (1.26 ± 0.07 fold vs. CM) (Fig. 3C). CSD leads to inflammation in the colon, reproductive organs and circulatory system, especially in female rats.

CSD alters gut microbiome composition

Next, we investigated the association between CSD and the composition of the gut microbiome. First, α-diversity (including Shannon and Simpson index) analysis indicated the sleep deprivation groups were significantly lower than control groups in both sexes (Fig. S1). Second, β diversity analysis indicated the distinct clustering of the microbiome compositions between the control and the sleep-deprived groups, and the result revealed a significant difference between CF and SDF groups (p = 0.013), as well as CM and SDM groups (p = 0.006) (Fig. 4A). Furthermore, the relations between specific bacterial taxa and sleep deprivation in both sexes were determined by using LEfSe analysis. The predominant bacteria at the genus level were Muribaculaceae, Prevotellaceae UCG-001, and Ruminococcaceae UCG-005 in the SDF group, and Prevotellaceae NK3B31, Ruminococcaceae UCG-010, Eubacterium coprostanoligenes, and Shuttleworthia in the SDM group (Fig. 4B).

Correlation among abundant genera, pubertal timing, antioxidant enzyme activity, and inflammatory cytokines

Abundant genera were involved in pubertal timing in female and male rats, and the results revealed that g_Ruminococcaceae_UCG-005 was positively correlated with vaginal opening day (Spearman’s rho = 0.633, p = 0.027) in SDF group, whereas g_Roseburia was negatively correlated with vaginal opening day (Spearman’s rho = −0.696, p = 0.012) in CF group (Table S3). In addition, g_Prevotellaceae_NK3B31_group was positively correlated with preputial separation day (Spearman’s rho = 0.587, p = 0.045) in SDM group, whereas g_Lachnospiraceae_A2, g_Ruminiclostridium_9, g_Clostridium_sensu_stricto_1 and g_Clostridiales_vadinBB60_group _Uncultured were negatively correlated with preputial separation day in CM group (Table S4).

The heat maps showed the correlations between abundant genera and antioxidant enzyme activity (Figs. 5A and 5B) and the correlations between abundant genera and inflammation (Figs. 5C and 5D). The results revealed the abundant genera in the SDF and SDM groups were positively correlated with antioxidant enzyme activity and inflammation; in contrast, the abundant genera in the CF and CM groups were negatively correlated with antioxidant enzyme activity and inflammation.

CONCLUSIONS

Our findings suggest that CSD is associated with delayed puberty onset and altered growth profiles in juvenile rats. These physiological changes coincided with increased antioxidant enzyme activities, elevated levels of proinflammatory cytokines (LBP, IL-1β, IL-6 and TNF-α), and significant shifts in gut microbiome composition. While these correlations support a potential mechanistic link between CSD, inflammation, and gut microbial dysbiosis, the data do not establish a direct causal pathway. Further mechanistic studies, particularly involving gut microbiome modulation or cytokine inhibition, are needed to confirm whether these pathways mediate the effects of CSD on pubertal timing. Future interventional or mechanistic studies in the context of sleep-deprivation should focus on treating oxidative stress and gut dysbiosis.