The protocols were approved by the Bioethics Commission from the Universidad Austral de Chile (CBA: 352/2019). Animal handling was performed following the guidelines for the care and use of laboratory animals of the Institute for Laboratory Animals Research of the National Research Council.

We raised and maintained 32 female rats of Sprague-Dawley strain obtained from Charles River (CRL International Inc., Kingston NY). These rats were mated, and the pregnancy was determined by the presence of positive sperm, obtained by vaginal smears, calling that day “day 0 of gestation” (E0). All pregnant rats had water and standard food (Prolab^® RMH 3000, Lab diet, USA) ad libitum under controlled temperature (20°C–22°C) in standard cages inside a cabinet with filters and ventilation. From day 1 of gestation, pregnant rats were randomly separated into two groups of light/dark photoperiod:

A) Control light/dark (LD n = 16 pregnant female): From day 1 of gestation, a group of dams continued with the light/dark 12:12 photoperiod, with artificial white light ~400 lux at the head level, which is turned on at 0700 h and turned off at 1900 h.

B) Chronic phase shift of photoperiod (CPS n = 16 pregnant females): using a similar protocol reported by us (6). Briefly, pregnant females were exposed to lighting schedule manipulation every 3–4 days reversing completely the photoperiod. The photoperiod reversal occurred at the night of day 0 of gestation, so that lights, rather than going off at 1900 h, remained on until 0700 h of day 2. At 18 days of gestation, the mothers returned to a normal 24-h photoperiod (12:12, lights on at 0700 h) and continued in this photoperiod thereafter. Exposure of pregnant females to CPS had no impact on food consumption and maternal weight (6).

Exposure of pregnant females to CPS had no impact on daily food consumption (monitored every week) through pregnancy (LD: 21.2 ± 0.6 g/day, n = 16 vs. CPS: 22.3 ± 0.6, n = 16), maternal weight at the end of gestation (LD: 427.6 ± 6.8 g, n = 16 vs. CPS: 430.3 ± 10.6, n = 16), or maternal weight gain at the end of gestation (LD: 134.4 ± 2.8 g, n = 16 vs. CPS: 135.0 ± 7.5 g, n = 16) as reported previously (6).

At birth, mothers and their pups were maintained in control photoperiod 12:12, and the male offspring were studied. All the animals were housed individually after weaning and weighted weekly, and food consumption was measured by weighing the food placed in the cage and the amount left 3 days later, when bed and fresh food were replaced. For the current study, we generated two cohorts of animals separated and used them in the protocols described below ( Figure 1 ). At 100 days of age, two brothers from the same pre-natal condition were separated and one brother was exposed to a HFD (45% excess calories) for 12 weeks and paired with a brother fed with standard food (Prolab^® RMH 3000, Lab diet, USA). Intraperitoneal glucose tolerance test (iGTT, Cohort1) and intraperitoneal insulin tolerance test (iITT, Cohort 2) were performed at 100 days (basal) and 12 weeks after exposure to either a HFD or standard diet. In addition, only in those males exposed to HFD was an additional iGTT and iITT test performed after 6 weeks under HFD. One week after the last test, animals were euthanized and adipose tissue was collected.

Male rats (n = 8 per group-Cohort 1) at 0800 h (after 12-h fasting) were anesthetized (Isoflurane 2.5%–3.5%) and injected with intraperitoneal glucose (1 g/kg; Glucose, Sanderson laboratories; Chile). A blood drop was collected from the tail to measure glucose levels (Accu-Chek; Roche Diagnostics) at −15 and 0 min before glucose administration and 30, 60, 90, 120, and 180 min after glucose injection. A blood sample was collected at −15 min to measure fasting glucose, insulin, leptin and adiponectin.

Male rats (n = 8 per group-Cohort 2) at 1500 h were anesthetized and injected with insulin (1U/kg; Humalog^® #CAT: VL7510, Eli Lilly and Company, Indianapolis, USA) after 6-h fasting. A blood drop was collected from the tail to measure glucose levels (Accu-Chek; Roche Diagnostics) at 0, 15, 25, 35, 45, and 60 min after glucose injection. A blood sample was collected at 0 min to measure fasting glucose, insulin, leptin, and adiponectin.

At 200 days of age (after 12 weeks of HDF, n = 8 per group—Cohort 1), eight male rats per group were deeply anesthetized (Isoflurane 3%, Baxter Laboratories, Melbourne, Australia), a midline incision was done to expose the vena cava, and an overdose of sodium thiopental (150 mg/kg; Vetpharma, Buenos Aires, Argentina) was administered. Immediately after death was confirmed, whole white adipose depots (inguinal, epigonadal, and perirenal) and interscapular brown adipose depot were dissected and weighted. Pieces were stored at −80°C for proteomic analysis or fixed in 4% formalin for histological analysis. The remaining adipose tissue were stored in our tissue bank at −80°C.

(n = 8 per group—Cohort 2). Epigonadal white adipose tissue (eWAT) and interscapular brown adipose tissue (iBAT) were dissected from individual rats, cut in small explants (about 50 mg for eWAT and 25 mg for iBAT) and suspended in culture medium (D-MEM F12, Sigma-Aldrich, St. Louis, MO, USA). Explants (10 explants for each animal) were pre-incubated in culture medium for 6 h at 37°C and aerated with 95% air CO₂ and 5% O₂. Next, explants were incubated in duplicate for 6 h in 2 ml medium alone (basal) or containing 0.01, 0.1, 1, and 10 μM of norepinephrine [A7257 (−)-Norepinephrine, Sigma-Aldrich, St. Louis, MO, USA]. At the end of incubation, the supernatant was collected, total protein of explants was assayed by Bradford method, and lipolysis was determined measuring the glycerol present in the supernatant fraction with a commercial kit (FG0100 Free Glycerol Determination Kit, Sigma-Aldrich, St. Louis, MO, USA). Production of glycerol was calculated as microgram per milligram of total protein in the explant.

Blood samples were collected from the tail of male rats (n = 8 per group—Cohort 1) after 12-h fasting (n = 8 per group) at basal, 6 weeks, and 12 weeks after HFD exposure. The serum of each sample was obtained by centrifugation (4,000×g, 10 min, 22°C), and insulin, leptin, and adiponectin concentrations were measured by a commercial immunoassay kit according to the manufacturer’s instructions (MILLIPLEX^® MAP Kit; Merck, KGaA, Darmstadt, Germany). The inter- and intra-assay coefficients were less than 10%.

Adipose tissue (prWAT and iBAT, n = 5 CPS + SD and LD + SD; 90 days old), stored in our tissue bank from previous studies (7, 30), were used to evaluate global genomic DNA methylation using MethylFlash (methylated DNA quantification colorimetric kit; Epigentek Group Inc., Farmingdale, NY, USA), following the manufacturer’s instructions. Basically, our aim was to compare the effect of gestational chronodisruption in global methylation in adipose tissue since epigenetic changes have been reported previously by us in adrenal and kidney, organs in which such changes have been accompanied with changes in organ function already in fetal life.

Protein identification was performed by high-resolution mass spectrometry on a hybrid dual-pressure linear ion trap/orbitrap mass spectrometer (LTQ Orbitrap Velos Pro, Thermo Scientific, San Jose, CA, USA) equipped with an EASY-nLC Ultra HPLC (Thermo Scientific). Input was 50 mg of white adipose tissue for each sample (n = 8 per group). For analysis, peptide samples were adjusted to 10 μl 1% ACN/0.1% TFA and fractionated on a 75-μm (ID), 25-cm PepMap C18-column, packed with 2 μm resin (Dionex/Thermo Scientific). The separation was achieved through applying a gradient from 2% to 35% ACN in 0.1% formic acid over 150 min at a flow rate of 300 nl/min. An Orbitrap full MS scan was followed by up to 20 LTQ MS/MS runs using collision-induced dissociation (CID) fragmentation of the most abundantly detected peptide ions. Essential MS settings were as follows: full MS (FTMS; resolution 60,000; m/z range 400–2000); MS/MS (Linear Trap; minimum signal threshold 500; isolation width 2 Da; dynamic exclusion time setting 30 s; singly charged ions were excluded from selection). The normalized collision energy was set to 35%, and activation time was set to 10 ms. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (31) partner repository with the dataset identifier PXD026315.

Data are expressed as mean ± standard error of the mean (SEM). LD and CPS differences in weight gain, food consumption, glycerol release, hormone levels, iGTT, iITT, and global methylation were analyzed by two-way repeated measures ANOVA and Bonferroni post-hoc test when group factor or interaction factor was significant (32). Additionally, we calculated the area under the curve (AUC) for iGTT and area above the curve (AAC) for iITT. Statistical analyses were performed using GraphPad Prism (version 9; GraphPad Software Inc., San Diego, CA). Results were considered significant when p < 0.05. For adipose tissue proteomics analysis, raw data processing and protein identification were performed using PEAKS Studio V.8.0 (Bioinformatics Solutions, Waterloo, Canada). False discovery rate was set to <1%. Label-free quantification was performed using Progenesis QI for proteomics (Nonlinear Dynamics/Waters). Proteins with abundance ratios of >1.5 or <1/1.5 at p < 0.05 were considered as significantly regulated. We applied an integrative and unbiased analysis approach for functional analysis of the proteins with significant changes in expression against the control group (LD offspring). Ingenuity Pathways Analysis (IPA) computes a score for each protein network according to the fit of that network to the user-defined set of “focus protein.” The score is derived from a p-value and indicates the likelihood of the focus proteins in a network being found together due to random chance. A score of 2 indicates that there is a 1 in 100 chance that the focus proteins are together in a network due to random chance. Therefore, scores of 2 or higher have at least a 99% confidence of not being generated by random chance alone [for details, see (17)]. First, we identified the modified proteins for the LD + HFD, CPS and CPS + HFD conditions compared to LD (fold change > 1.5; p < 0.05). After this exploratory analysis, we decided to focus on the common proteins for the CPS and LD + HFD groups, based on our observed results. Under this criterion, we selected the 20 proteins shown in Figure 8 and were grouped to build a new dataset and analyzed by IPA to generate the interactomes.

Between weaning (21 days old) and 100 days of age, CPS and LD animals had similar food consumption expressed as kcal/day (LD: 137 ± 12 kcal/day, n = 16 versus CPS: 143 ± 14, n = 16; p = 0.197 unpaired t-test). However, CPS offspring was slightly heavier (LD: 602 ± 51 g, n = 16 versus CPS: 615 ± 50 g, n = 16; p = 0.5, unpaired t-test). As shown in Figure 2A , CPS animals fed a standard diet (CPS + SD) between 100 and 200 days became increasingly heavier than LD + ST animals. After the 4 weeks under HFD, LD + HFD and CPS + HFD animals showed a significant increase in body weight versus LD + SD (time × group factor p < 0.001; Figure 2A ). Thus, CPS animals with SD or HFD featured a similar weight gain profile, reaching an increase of about 50% versus LD + SD ( Figure 2A ). Food intake remains steady along the 12 weeks in animals receiving either standard or HFD ( Figure 2B ). Therefore, CPS animals fed with SD had a similar weight gain to those LD or CPS fed with HFD.

In line with a previous report (6), we found slight differences in intraperitoneal glucose tolerance test (iGTT) between LD and CPS adult offspring at 100 days of age, before the beginning of HFD ( Figure 3A ). After 60 min post glucose injection, glucose levels rapidly decreased to 173± 2 mg/dl in the LD group, remaining elevated in the CPS group (210 ± 16 mg/dl; p = 0.024; two-way ANOVA and Bonferroni test), but no interaction time × column was found (p = 0.172). In addition, analyzed as AUC, glucose response displayed a similar value to the CPS group without reaching statistical significance (p = 0.857, unpaired t-test; Figure 3A ). However, no difference was found at 200 days ( Figure 4A ). Important differences were found in iGTT after 6 weeks with HFD. At this time window, the adult offspring gestated under LD conditions maintained the glucose response similar to that observed at week 0. In contrast, a marked increase in glucose response was observed in the CPS group relative to control LD (time × group factor: p = 0.012, two-way ANOVA and Bonferroni; Figure 3C ). Analyzed as AUC, glucose response remained the same as week 0 (basal) for the LD group, while basal plasma glucose levels tended to increase in the CPS group relative to control LD group (p = 0.111; unpaired t-test). As anticipated, a significant increase in glucose response was found in LD animals after 12 weeks ( Figure 3C ) regarding 0 and 6 weeks of HFD. At this late time window, the adult CPS offspring presented significantly higher levels of glucose in basal conditions, raising up to 340 mg/dl levels after 30 min post intraperitoneal glucose injection ( Figure 3E ), but no interaction time × group was found (p = 0.622). Measured as AUC, the glucose response was similar to the CPS group compared to the LD group (p = 0.838; unpaired t-test), probably due to the fact that the CPS group begins with higher basal plasma levels.

Intraperitoneal insulin tolerance tests were carried out in the parallel cohort 2. One-hundred-day-old animals have shown no difference between LD and CPS offspring (time × group factor: p = 0.616, two-way ANOVA; Figure 3B ), expressed either as plasma glucose levels or area above the curve (AAC; p = 0.866 unpaired t-test). Similar results were found at 200 days of age ( Figure 4B ). After 6 weeks of HFD, the group that had been gestated under LD presented a glucose response to insulin alike to the one observed at week 0 (basal). However, the CPS group displayed a marked increase relative to both CPS and LD at week 0 (basal) and to LD at week 6 (time × group factor: p = 0.026, two-way ANOVA and Bonferroni; Figure 3D ). Glucose response expressed as AAC was at the same level in the LD group regarding week 0 (basal), while CPS showed a significant decrease of AAC (p = 0.033; unpaired t-test). Finally, a third iITT was performed after 12 weeks under HFD ( Figure 3F ). No significant increase in glucose response to insulin was found in LD animals regarding week 0 (basal). Meanwhile, CPS animals presented significantly higher levels of glucose in basal conditions, maintaining higher levels of glucose up to the end of the experiment (time × group factor: p = 0.019, two-way ANOVA and Bonferroni). However, AAC values for insulin response were lower in the CPS compared to the LD group, but not statistically significant (p = 0.325, unpaired t-test; Figure 3F ), probably due to the higher fasting glucose in CPS.

Next, we evaluated the potential role of the endocrine system to help to explain our findings. To this goal, insulin, leptin, and adiponectin plasma levels were measured before applying the glucose tolerance test in fasting conditions after HFD. First, steady levels of fasting glucose and circulating insulin, leptin, and adiponectin were found throughout 12 weeks under HFD challenge in the control LD group ( Figure 5 ). In contrast, adult CPS offspring displayed increased levels of fasting glucose (time × group factor: p = 0.025, two-way ANOVA and Bonferroni; Figure 5A ) at 6 and 12 weeks of treatment. In addition, they exhibited high insulin plasma levels at 6 weeks (time × group factor: p = 0.016, two-way ANOVA and Bonferroni; Figure 5B , i e, week 6 versus week 0/basal), dropping to minimal levels after 12 weeks of HFD challenge (i.e., week 12 versus week 6). On the other hand, when plasma levels of the two hormones produced by WAT from CPS animals were evaluated, leptin showed constantly increased levels along the 12 weeks under HFD, which is consistent with higher body weight gain ( Figure 5C ), without an interaction (time × group factor: p = 0.123, two-way ANOVA). Adiponectin displayed higher levels already in basal conditions in CPS offspring (i.e., before HFD), but like the levels measured in the control LD offspring, with stable values throughout HFD challenge (time × group factor: p = 0.149, two-way ANOVA; Figure 5D ).

White and brown adipose tissues were collected and analyzed 1 week after the last test described above was carried out. CPS animals fed standard diet had more iWAT and less BAT than animals gestated under control (LD) conditions, while pWAT and eWAT show no difference. HFD induced an increase in perirenal, epigonadal, and inguinal white adipose tissue pads ( Figures 6A–C ), matching findings from another report (33). Meanwhile, CPS offspring displayed increased subcutaneous iWAT depot with both standard diet and HFD, compared to LD without interaction (column factor: <0.001; time × group factor: p = 0.704, two-way ANOVA). Surprisingly, CPS iWAT levels under standard diet were similar to those of LD + HFD iWAT. We explored whether the treatments induced an increase in the number of white adipose cells in eWAT. Neither LD nor CPS with or without HFD was associated with differences in the number of cells per section, supporting the idea that the effects induced by HFD or CPS on eWAT total mass represent bigger fat pads. Next, we evaluated the effects of CPS and HFD on brown adipose tissue depot. As expected, HFD given to the control LD offspring did not affect the content of brown adipose tissue (interscapular depot). However, CPS offspring presented a significant decrease (about 50%; group factor: p < 0.001, two-way ANOVA and Bonferroni) in iBAT, which was not affected by HFD (time × group factor: p = 0.697, two-way ANOVA; ( Figure 6D ). Neither LD nor CPS with or without HFD displayed differences in the number of cells per section ( Figures 6E–H ).

The functional integrity of WAT was assessed under in vitro conditions by testing glycerol response to increasing doses of norepinephrine (NE). Explants of LD eWAT animals fed with standard diet had a significant glycerol response to 1 µM NE and reached a maximal response with 10 µM NE. In contrast, in CPS, eWAT explants’ glycerol response to NE was statistically significant only to 10 µM NE (group factor: p = 0.103, two-way ANOVA and Bonferroni; Figure 7A ). In animals fed with HFD, the glycerol response to NE in eWAT LD explants was similar to the one found under the standard diet (group factor: p = 0.64, treatment × group factor: p = 0.76, two-way ANOVA and Bonferroni; Figures 7A–C ). Therefore, HFD did not affect the in vitro response to NE in LD conditions ( Figures 7C ). In contrast, eWAT explants from adult animals that had been gestated under CPS conditions displayed a blunted glycerol response to all NE doses tested (group factor: p < 0.001, treatment × group factor: p = 0.024, two-way ANOVA and Bonferroni; Figure 7C ).

The functionality of BAT was also evaluated under in vitro conditions. First, BAT explants of adult offspring gestated under LD and CPS conditions and fed with standard diet showed increased glycerol response to 10 µM NE (group factor: p = 0.029, two-way ANOVA and Bonferroni; Figure 7B ). Of note, higher basal levels of glycerol were found in the incubation media from CPS animals than that of control (LD) animals. This difference was no longer present in BAT explants from CPS animals receiving HFD for 12 weeks. In addition, BAT explants from these animals showed a reduced glycerol response to NE and overall glycerol production was lower than that of BAT explants from control (LD) conditions also fed on HFD (group factor: p < 0.001, treatment × group factor: p = 0.440, two-way ANOVA and Bonferroni; Figure 7D ).

Some of the differences in white and brown adipose tissue between CPS and LD animals described above were present before the HFD challenge. Therefore, we explored potential epigenetic mechanisms involved by measuring global genomic DNA methylation status in both WAT and BAT depots in the offspring at 100 days. Perirenal WAT from LD offspring showed significant day/night methylation differences, with levels that were about 25% higher during daytime ( Figure 8A ). Conversely, lower levels of global genomic DNA methylation were detected during daytime in perirenal WAT from CPS offspring (time × group factor: p = 0.001, two-way ANOVA and Bonferroni). It should be noted that the mean day/night methylation level was similar for perirenal WAT from both LD and CPS adult offspring. Finally, no differences were found for global DNA methylation in BAT, either during daytime or nighttime (p = 0.948, two-way ANOVA; Figure 8B ). These early differences in global DNA methylation suggest that epigenetic programming has occurred in WAT from the CPS group.

Next, we evaluated the effects of prenatal chronodisruption and postnatal HFD on total protein expression in white adipose tissue using a quantitative proteomics analysis. (Data are available via ProteomeXchange with identifier PXD026315.)

Animals that had been gestated under LD conditions and received HFD during adult life displayed a significant increase in body weight and weight of white adipose tissue fat pads at 100 days. These exacerbated weight-gain pattern induced by HFD was accompanied by modification in the expression level of 91 proteins (fold change higher than 1.5) in white adipose tissue. We further analyzed the proteomics data using Ingenuity Pathway Analysis (IPA) to look for interactions among the differentially expressed set of proteins framed in functional pathways. As might be expected, the major effects at the proteome level were found in the functional pathway involving fatty acid metabolism. In addition, changes in the expression level of proteins related to cytoskeleton function and cellular differentiation were found.

WAT from adult CPS offspring at 200 days fed with SD displayed differential expression of 275 proteins with a fold change higher than 1.5, when compared to the control LD group fed with SD. Remarkably, 96% of the differentially expressed proteins were downregulated in WAT from adult CPS offspring. This significant proportion of downregulated proteins is consistent with the metabolic alterations described above for WAT from adult CPS relative to LD offspring.

Comparison of adult CPS offspring fed standard versus CPS fed HFD led to the identification of 235 proteins exhibiting significantly modified expression levels, that is to say, 114 more proteins than LD animals receiving HFD. Among these, 219 proteins were upregulated and 16 proteins were downregulated.

In contrast, comparison of WAT from CPS fed with SD and LD fed with HFD vs. both LD fed with SD show striking similarity and only four proteins were differentially expressed in CPS HFD. Taken together, these proteomics results support the notion that in terms of protein expression, WAT from adult CPS offspring is somewhat similar to WAT from adult LD offspring fed with HFD. This possibility was further investigated by running functional proteomic analyses to target common protein profiles by these means; 20 differentially expressed proteins were identified in both experimental groups as compared to the adult offspring that had been gestated under control LD conditions receiving SD postnatally ( Figure 9 ). Of note, a most important difference in WAT was found in MYADM, that was −9.7 times downregulated regarding LD + HFD. So far, the actual role of MYADM in adipose tissue function remains non-investigated, but recent evidence is in line with a potential role of this protein in local inflammation, as mediator of low-grade inflammation as deduced for its role in membrane (34). After this exploratory analysis, we decided to focus on the common proteins for the CPS and LD + HFD groups, based on our observed results. Under this criterion, we selected the 20 proteins shown in Figure 9 , grouped to build a new dataset and analyzed by IPA to generate the interactomes.

Detailed pathway analysis of WAT from adult CPS + SD showed an imbalance in the pathways related to inflammatory status, network 1 (TNF/IL4, score 14; Figure 10 , top panel), and network 2 (AKT/ERK, score 28; Figure 10 , lower panel) supporting the idea that WAT from adult CPS offspring fed with SD was undoubtedly affected to an important extent by gestational chronodisruption.