Subjects

Forty healthy adult women of European ancestry were enrolled in the study. They comprised 20 young women (age 20–29 years, mean ± standard deviation, 24.25 ± 2.45 years) and 20 older women (60–82 years, 68.60 ± 5.68 years). The study was confined to women because sex differences have been shown for oxidative stress levels [16], TNF-α [17] and cortisol secretion [18]. The women were recruited through flyers, by word-of-mouth and in different clubs. Women were included in the study if they were healthy, nonsmokers, had C-reactive protein < 5.0 mg/L, BMI between 18.5 and 29.0 kg/m2, sleep duration between 6 and 9 h, bedtimes between 21:00 and 01:00 h and a regular sleep–wake pattern. Exclusion criteria were acute and chronic illness, cancer, diabetes, psychiatric illness, oral diseases, canker sores, bleeding gums, dental treatment within the last month as well as anti-inflammatory, sleeping, and psychiatric medication. Additional exclusion criteria were physical injury, sunburn, current diet, high-performance sport, pregnancy, shift work, and jet-lag within the last three months. Inclusion and exclusion criteria were assessed by questionnaires, medical history and laboratory analysis. 38% of the women never drank alcohol and 62% only small amounts. None of the older women received hormone replacement therapy. Seven of the young women took monophasic oral contraceptives for birth control. All participants gave written informed consent. The ethics committee at the Christian-Albrechts-University of Kiel approved the study protocol (D 535/19).

Procedure

Data was collected from June to August 2022. Participants visited the Department of Human Biology at Kiel University two times. At the first appointment, each participant received personally oral and written instructions on the study procedure. The participants were told to follow a regular sleep–wake pattern for seven days prior to the study day to stabilize their biological rhythms. They noted bedtimes for each of the 7 days. The second appointment was on study day. The participants arrived after an overnight fast at 07:30 h at the university and completed several questionnaires. The concentrations of hs-CRP and vitamin D were measured before 08:00 h and anthropometric measurements were taken. The participants stayed in a separate room for saliva collection. During the 24-h sampling period, whole unstimulated saliva was collected at 4-h intervals at 08:00, 12:00, 16:00, 20:00, 24:00, 04:00 and 08:00 h into specific sampling devices (SaliCaps, IBL, Hamburg, Germany). Saliva was collected from 08:00 to 20:00 h by one of the authors (P.A.) and from 24:00 to 08:00 h by the participants themselves, who were carefully trained and equipped with collecting devices. Within one hour before each saliva sampling, the participants refrained from brushing teeth, drinking, eating and chewing gum. Saliva samples were cleared by centrifugation at 1000 rpm for 15 min to pellet bacteria, food debris and cells. The supernatant was immediately stored in aliquots at -80 °C until analysis. The aliquots were used for determining concentrations of OGG1, DNA/RNA oxidative damage, TNF-α and cortisol. Core body temperature was measured twice at the same time-points. During the study day, the participants were not allowed to do sports, nap, drink alcohol and caffeinated drinks. All participants had breakfast, lunch and dinner, respectively, between 08:30–10:00 h, 12:30–14:00 h and 18:00–19:00 h. From 19:00 h until bedtime, no meals were allowed. From 16:00 h until bedtime, the participants wore blue light blocking glasses (400–500 nm, Prisma blue-light protect Amber PRO, Innovative eyewear, Weilheim, Germany), because blue light exposure inhibits endogenous melatonin secretion, which is important for generating robust circadian 24-h rhythms [19].

Anthropometry, bioelectrical impedance analysis, questionnaires

Body weight, body height, body circumferences (waist circumference, abdomen circumference, hip circumference), and body composition, respectively, were determined using an electronic scale (TGF 302H, Rossmann, Burgwedel), a wall-mounted measuring device (Seca 206, Seca, Hamburg), a measuring tape, and a whole-body tetrapolar bioelectrical impedance analyzer (Nutriguard-M, Data Input, Pöcking, Germany). Body mass index (BMI) was calculated as body weight (kg) divided by the square of body height (m2). Morningness-eveningness preference, sleep quality and daytime sleepiness, respectively, was estimated using the Morningness-Eveningness-Questionnaire ([20], German validated version by [21]), the Pittsburgh Sleep Quality Index [22], and the Epworth Sleepiness Scale [23].

Measurement of hs-CRP and vitamin D

The C-reactive protein (CRP) is an acute-phase protein, whose concentration increases in blood in response to inflammation. CRP was quantified to rule out the presence of acute or chronic inflammation in the body of the participants. Any participant having CRP > 5.0 mg/L was excluded from the study. High-sensitive CRP (hs-CRP) was quantified in capillary blood of the finger using the hs-CRP assay on Eurolyser CUBE-S laboratory photometer (Hitado, Dreihausen, Germany). This quick test is controlled for hematocrit. The assay range is 0.8–20 mg/L blood.

Vitamin D was quantified because a deficiency may increase levels of OGG1, DNA damage and TNF-α [24]. Vitamin D was measured using the “Vitamin D quantitative test” assay on VHC reader (Hitado, Dreihausen, Germany). This immunochromatography-based one step quick test quantifies the total 25-hydroxy vitamin D (25-OH vitamin D) in finger capillary blood. It can be used for screening for vitamin D deficiency. The assay range was 4–100 ng/ml and its sensitivity was 3.3 ng/ml.

Oxoguanine glycosylase (OGG1) assay

The protein level of OGG1 was quantified in human saliva samples that were collected at seven time-points (08, 12, 16, 20, 24, 04, and 08 h). The OGG1 concentrations were measured using a human enzyme-linked immunosorbent assay (ELISA kit ELK3235, ELK Biotechnology, Wuhan, Hubei, China), as described by the manufacturer. The test principle is based upon a quantitative sandwich enzyme immunoassay technique. Streptavidin-HRP binds to biotinylated antibodies and provides enzyme activity for detection with a substrate system. The reaction between enzyme (HRP) and substrate (TMB) resulted in colour change. The optical density was measured with a microtiter plate reader (Multiskan FC, ThermoScientific, Darmstadt, Germany) at 450 nm. The concentrations of the samples were determined from a standard curve with seven standards, adapted to the software of the microtiter plate reader. All samples were analyzed in duplicate. The assay range was 0,16–10,00 ng/ml and the sensitivity of the assay was 0.058 ng/ml.

Oxidative DNA/RNA damage assay

The level of oxidative damage to DNA/RNA was measured in human saliva samples that were collected at seven time-points (08, 12, 16, 20, 24, 04 and 08 h) and stored at -80 °C. A competitive enzyme-linked immunosorbent assay (ELISA kit no. 589320, Cayman Chemical, Ann Arbor, MI, USA) was used to quantify all three oxidized guanine species from DNA and RNA (8-hydroxy-2’-deoxyguanosine, 8-OHdG; 8-hydroxyguanosine, 8-OHG; 8-oxoguanine, 8-OxoG), following the manufacturer’s instructions. This quantitative assay is based on the competition between the oxidatively damaged guanine species and an 8-OH-dG-acetylcholinesterase conjugate for an DNA/RNA oxidative damage monoclonal antibody. An enzymatic reaction using Ellman’s reagent lead to a yellow product whose optical density was measured with a microtiter plate reader (Multiskan FC, ThermoScientific, Darmstadt, Germany) at 405 nm. Sample concentrations were calculated by means of the manufacturer’s table in Microsoft Excel 2016. The sensitivity of the assay approximated 30 pg/ml.

Tumor necrosis factor alpha (TNF-α) high-sensitivity assay

The protein level of TNF-α was determined in human saliva samples that were collected at seven time-points (08, 12, 16, 20, 24, 04, and 08 h). The TNF-α concentration was measured using the high-sensitivity human Quantikine™ enzyme-linked immunosorbent assay (ELISA kit HSTA00E, R&D Systems, Minneapolis, USA), according to the manufacturer’s instructions. The test principle of the assay employs a quantitative sandwich enzyme immunoassay technique using a monoclonal antibody specific for human TNF-α, coated on a 96-well plate. After washing of the wells, a polyclonal antibody specific for TNF-α, was added to the plate. After adding enzyme-linked streptavidin and substrate solution, a colour reaction developed, which was proportional to the amount of TNF-α in the sample. Optical density was read at 570 nm and a reference wavelength of 620 nm with a microtiter plate reader (Multiskan FC, ThermoScientific, Darmstadt, Germany). The concentration of TNF-α was determined from a standard curve, which was created using the software of the microtiter plate reader. All samples were analyzed in duplicate. The sensitivity of the assay approximated 0.022 pg/ml.

Cortisol assay

The concentration of cortisol was determined in human saliva samples, which were collected at seven time-points (08, 12, 16, 20, 24, 04, and 08 h). After screening for blood contamination, concentrations of free cortisol in saliva were measured with a quantitative enzyme-linked immunosorbent assay (ELISA kit RE52611, IBL, Hamburg, Germany), following the manufacturer’s instructions. The test principle is based upon the competition between the unknown amount of cortisol in the sample and a fixed amount of enzyme-labelled cortisol for antibodies coated onto a 96-well plate. The cortisol concentration is inversely proportional to the intensity of color change. Optical density was measured at 450 nm with a reference wavelength of 620 nm using a microtiter plate reader (Multiskan FC, ThermoScientific, Darmstadt, Germany). The cortisol concentrations were calculated from a standard curve with the software of the microtiter plate reader. The assay range was 0.005–3.00 µg/dl.

Determination of core body temperature

Core body temperature (CBT) was measured in the right posterior sublingual pocket of the mouth with a digital thermometer (Thermoval, Hartmann, Heidenheim, Germany) at 08, 12, 16, 20, 24, 04 and 08 h. Temperature measurements were done in duplicate and the mean value was used for analysis.

Statistical analyses

Statistical analyses were performed by means of IBM SPSS software for MS Windows, release 29.0 (IBM, Armonk, NY, USA). Data are shown as mean ± standard deviation (SD) or standard error of the mean (SEM). Outliers beyond three standard deviations from the mean were excluded from data analysis following Snedecor and Cochran [25]. Normal distribution of data was analyzed by Shapiro–Wilk test. Differences between age groups were tested using the two-tailed Student’s t test or Mann–Whitney-U-test, depending on normal distribution of data. Friedman tests were applied to test for 24-h variation of levels for OGG1, oxidative DNA/RNA damage, TNF-alpha and cortisol, separately for young and older women. Differences between day levels and night levels of the circadian parameters were compared using t tests for dependent samples or Wilcoxon tests, were appropriate. Mixed linear models were applied to test the interaction between time (day vs. night) and age (young vs. older women). The dependent variables were the levels for OGG1, oxidative DNA/RNA damage, TNF-alpha and cortisol as well as CBT. Log-transformed values were used for OGG1, TNF-alpha and cortisol. The fixed factors were time and age. The amplitude of the circadian cycle of CBT was calculated as amplitude percent mean (A%Mean) as follows: highest CBT value of the 24-h day minus lowest CBT value of the 24-h day, divided by the mean CBT value of the 24-h day and multiplied with 100 [26]. Pearson or Spearman rank correlation coefficients were used to explore relationships between variables, where appropriate. Multiple regression analysis was applied to examine the influence of age, oxidative DNA/RNA damage, TNF-α, cortisol and anthropometric measures (independent variables) on the 24-h mean level of OGG1 (dependent variable). Log-transformed values for OGG1, oxidative DNA/RNA damage, TNF-alpha and cortisol levels were used in the regression analysis. A two-sided P value of < 0.050 was considered statistically significant.