Study protocols were approved by the Bioethical Committee of the University of the Balearic Islands (Ref. 3513, 26/03/2012). International standards for the use and care of laboratory animals were followed. Animals were housed at 22°C, with a 12 h light-dark cycle (lights on at 08:00), and free access to tap water and standard chow diet (type A03; 3.3 kcal/g, 8% calories from fat; Panlab, Barcelona, Spain). Virgin female NMRI mice (Charles River Laboratories, Barcelona, Spain) were mated. Each female was single-caged after mating. The morning in which newborn litters were found was designated as day 0. At day 2, litter size was adjusted to 12 pups per dam. Pups in four litters were randomly assigned to control or RSV group; in a separate experiment, pups in four additional litters were assigned to control or NR group (six pups per litter per group in both experiments). From postnatal day 2 (P2) to P20, the pups received daily orally 10–15 μL of either vehicle (water, control groups), a solution of RSV (Sigma, Madrid, Spain) or a solution of NR (Chemical Point, Diessenhofen, Germany), using a pipette. The amount of supplemented RSV was adjusted along the suckling period to meet a daily dose of 2 mg/kg body weight. RSV at 10–30 mg/kg animal/day decreases adiposity in adult rodents (Rivera et al., 2009; Alberdi et al., 2013) and we further applied a security factor considering the young age of the animals. The NR supplied ranged from 24 μg on P2 to 45 μg on P20, and was equivalent to ∼15-fold the total vitamin B3 ingested daily from maternal milk, considering the vitamin B3 content found in milk of in-home mouse dams [2378 ppb (Serrano et al., 2018)] and the daily milk intake of pups throughout lactation (Kojima et al., 1998). Animals were weaned and separated by sex on P21. The studies were performed separately in male and female mice. Body weight was periodically recorded from birth. Body composition was analyzed at P34, using an Echo MRI body composition analyzer (EchoMRI, LCC, Houston, TX, United States). Naso-anal length was also measured on P34, and used to calculate Lee’s obesity index (body weight in g^0.33 × 1000/naso-anal length in cm). Energy intake from weaning until euthanization was estimated on a per-cage basis, from the actual amount of food consumed by the animals and its caloric equivalence. The animals were euthanized on P35, by decapitation, under fed conditions, within the first 2 h of the light cycle; half of the animals in each sex and treatment group (n = 5–6, from four different litters) were used to establish sex-separated adipose tissue primary cultures (see below), while the other half were used for tissue sampling for molecular analyses. Interscapular BAT and WAT depots (inguinal, epididymal, and retroperitoneal) were dissected in their entirety, weighed, snap-frozen in liquid nitrogen and stored at -80°C until processed. Glucose was measured from trunk blood using the Accu-Chek Aviva system (Roche Diagnostics). Adiposity index was calculated as the sum of all WAT depot mass divided per body weight and per cent.

The SVF containing preadipocytes was obtained from iWAT depot and pooled (interscapular, cervical, and axillary) BAT depots of 35-day-old mice, following a previously described protocol (Petrovic et al., 2010). The cell pellet was suspended in culture medium (0.5 mL/animal for brown preadipocytes, 0.4 mL/animal for white preadipocytes) and seeded onto 6-well culture plates (2 mL culture medium and 0.2 mL of cell suspension per well). The culture medium was Dulbecco’s modified Eagle’s medium with 10% (v/v) newborn calf serum (Gibco, Thermo Fisher Scientific, Waltham, MA, United States), 4 nM insulin, 25 μg/mL sodium ascorbate, 10 mM HEPES, 4 mM glutamine, 50 units/mL penicillin, and 50 μg/mL streptomycin. From day 0 of culture, all wells were supplemented with 1 μM rosiglitazone (BioVision, Milpitas, CA, United States) to favor the recruitment of brown-like adipocytes (Petrovic et al., 2010). Medium was changed on day 1 and then every second day, except the day the cells were harvested. The cells were grown for 7 days at 37°C in an atmosphere of 8% CO₂ in air. Adipogenic differentiation of the cells was regularly monitored through phase contrast microscopical examination. At harvesting, the percentage of cells showing intracellular lipid accumulation was ∼80% and 60% for the primary cultures established from, respectively, BAT and iWAT, as in previous reports from our group using the same primary culture protocol (Petrov et al., 2016) and without apparent differences between neonatal treatments (see representative microphotographs in Supplementary Figures S1, S2). Half of the cultures were exposed to 1 μM NA for the 2 h prior harvesting, to stimulate thermogenically competent cells and UCP1 expression (Petrovic et al., 2010).

Total RNA was extracted using Tripure Reagent (Roche, Barcelona, Spain) according to the manufacturer’s instructions, followed by a sodium acetate/ethanol precipitation to purify nucleic acids. Isolated RNA was quantified using NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies Inc., Wilmington, United States) and its integrity confirmed by agarose gel electrophoresis. Total RNA (0.25 μg/reaction) was reverse-transcribed using MuLV Reverse Transcriptase and random hexamers priming (Applied Biosystems, Madrid, Spain). Diluted cDNA template was used to perform PCR amplification of selected genes, along with specific forward and reverse primers (Sigma, Madrid, Spain), and Power SYBER Green PCR Master Mix (Applied Biosystems). Primers sequences are available on request. For Real-Time PCR the Applied Biosystems StepOnePlus Real-Time PCR Systems (Applied Biosystems) was used, with the following profile: 10 min at 95°C, followed by a total of 40 two-temperature cycles (15 s at 95°C and 1 min at 60°C). To verify the purity of the products, a melting curve was produced after each run. The threshold cycle was calculated by the instrument’s software (StepOne Software v2.2.2, Applied Biosystem) and the relative expression of each mRNA was calculated according to the 2^-ΔCt method, using 18S rRNA as reference gene.

Data are expressed as mean ± SEM. Statistical analyses were conducted separately by treatment (RSV and NR supplementation) and sex (male and female mice). Student’s t-test was used for single comparisons. Two-way ANOVA was used for analysis of treatments effects in primary adipocytes under basal and NA-stimulated conditions; post hoc comparisons were included when two-way ANOVA indicated an interactive effect between neonatal treatment and NA exposure of cultures. Threshold of significance was set at p < 0.05. IBM SPSS Statistics for Windows, version 19.0 (IBM Corp., Armonk, NY, United States) was used for the analyses.

To study the influence of treatments on the commitment of WAT preadipocytes toward the beige adipogenesis transcriptional program, expression levels of markers of brown/beige adipocyte determination and thermogenic function (Ucp1, Cpt1b, Prdm16, Ppargc1a, Ppargc1b, Ppara), and of purported beige adipocyte-selective transcriptional markers (Tmem26, Hoxc9, Slc27a1) were compared in differentiated iWAT primary cultures from RSV and NR mice and their respective controls. Pparg expression was also assessed, since the encoded product, PPARγ, is a master adipogenic factor that is also central to WAT browning (Ohno et al., 2012).

Adipocytes differentiated from the iWAT SVF of RSV female mice showed a generalized decreased expression of transcripts of brown and beige adipocyte marker genes (namely: Ucp1, Cpt1, Prdm16, Ppargc1b, Ppara, Tmem26, Slc27a1), as well as of Pparg, when compared to primary adipocytes from control female mice (Figure 1A). On the contrary, primary adipocytes from RSV male mice showed increased expression levels of Ucp1 and Slc27a1 compared with corresponding controls (Figure 1B). Ucp1 expression is the hallmark of the brown and beige adipocyte phenotype, and Slc27a1 encodes a fatty acid transport protein (FATP1) that is highly expressed in BAT and other tissues actively oxidizing fatty acids and that is required for BAT thermogenesis (Wu et al., 2006). Furthermore, Slc27a1 has been related to mitochondrial function in clonal white (3T3-L1) adipocytes (Wiczer and Bernlohr, 2009), and it is the only beige adipocyte transcriptional marker (out of 6 studied) found to be induced together with classical brown adipocyte marker genes in mouse WAT primary cultures following rosiglitazone exposure (Petrov et al., 2016). Additionally, trends to increased expression levels of most other genes related to oxidative metabolism/fatty acid oxidation assayed were apparent in the primary adipocytes derived from RSV male mice vs. controls under NA stimulated conditions (Figure 1B).

The overall picture was similar for the NR treatment. Ucp1 mRNA levels were increased in primary iWAT adipocytes derived from NR male mice but not female mice (Figure 2). Moreover, primary iWAT adipocytes from NR female mice showed a downregulated expression of several of the brown and beige transcriptional markers assessed (Ppargc1a, Ppargc1b, Tmem26, Hoxc9), whereas those from NR male mice showed an upregulated expression of Slc27a1 – as for the RSV treatment – and also of Prdm16 and Pparg (the latter with p = 0.073, two-way ANOVA) compared to corresponding controls (Figure 2). Prdm16 encodes a transcriptional coactivator critical for WAT browning in postnatal life (Seale et al., 2011; Cohen et al., 2014), and agonism of PPARγ favors WAT browning in vivo and beige adipogenesis in primary adipose cultures (Fukui et al., 2000; Petrovic et al., 2010), possibly through stabilization of the PRDM16 protein (Ohno et al., 2012).

Half of the cultures were adrenergically stimulated prior harvesting to potentiate the emergence of thermogenically competent cells (Petrovic et al., 2010). Expected transcriptional responses to NA stimulation were found. In particular, Ppargc1a expression (encoding PGC1α) was induced by NA exposure in all iWAT primary cultures, regardless of sex of origin or neonatal treatment (Figure 1, 2). Ucp1 expression was induced by NA exposure in the iWAT cultures from control and RSV mice regardless of sex, but not further induced by NA in the iWAT cultures from NR mice of either sex, which already overexpressed Ucp1 compared to controls under the basal, non-stimulated, conditions (Figure 2). Apart from Ppargc1a and Ucp1, other brown/beige marker genes analyzed in iWAT primary cultures were not induced following NA exposure, but were rather unaffected or downregulated (Figure 1, 2). Down-regulation of genes in this class upon NA exposure was especially evident in the cultures derived from control male mice, and was attenuated in the cultures derived from RSV male mice (see results for Cpt1b, Prdm16, Ppargc1b, Ppara, Hoxc9, and Slc27a1 expression in Figure 1B).

Altogether, these results indicated that the neonatal RSV and NR treatments had sex-dependent effects on the brown/beige transcriptional signature of iWAT primary cultures from young mice.

In adult animals, it is common that the same stimuli that induce WAT browning also induce BAT recruitment (Bonet et al., 2013, 2017). Thus, effects of the in vivo neonatal treatments on thermogenic gene expression in primary BAT adipocytes from young mice were studied, to check for potential parallelisms between results in iWAT- and BAT-derived cultures. Compared with levels in cultures from sex-matched controls, the mRNA levels of Ucp1, Cpt1b, Ppargc1b, and Ppara were downregulated in the BAT primary cultures from RSV female mice, but not (except for Ppargc1b expression under basal conditions) in those from RSV male mice (Figure 3A,B). On the contrary, BAT cultures from RSV male mice more readily induced Ucp1 and Ppargc1a mRNA levels following NA stimulation, and showed increased Ppargc1a and Ppara mRNA levels under NA-stimulated conditions compared with cultures from control male mice. Furthermore, upon NA stimulation, Ppargc1b expression levels decreased in the BAT cultures from control male mice but not in those from RSV male mice. Thus, the overall effect of neonatal in vivo RSV treatment on BAT primary adipocyte cultures from young mice resembled that in iWAT primary adipocyte cultures, with thermogenic gene expression being potentiated in the cultures derived from male mice and attenuated in the cultures derived from female mice. Somewhat oppositely, neonatal NR treatment associated with higher Ucp1 expression selectively in the female mice-derived BAT cultures, whereas expression of other genes assayed (Cpt1b, Prdm16, Ppargc1a, Ppargc1b, Ppara, Pparg) was not significantly affected by neonatal NR treatment in either sex (Figure 4A,B). Expected responses to NA, namely induction of Ucp1 and Ppargc1a expression, were present in all BAT primary cultures regardless of sex of origin or treatment (Figure 3, 4).

Eventual effects of treatments on early postnatal growth were evaluated by monitoring body weight evolution from birth; body length (nasal-to-anal distance), Lee index and body composition (by EchoMRI) at P34; and tissue/organ weights at euthanization at P35. RSV treatment had no significant effect on any of these parameters, neither in male nor in female animals (Table 1). As for the NR treatment, treated male mice, but not females, were slightly but significantly heavier (by ∼6%) and had higher (by ∼20%) body fat content and fat/lean ratio at P34 compared to sex-matched controls (Table 2). Consistent with these body composition results, NR male mice, but not females, displayed after dissection a trend to higher gonadal WAT mass (p = 0.090) (Table 2). Cumulative energy intake from weaning until P30 was significantly lower (by ∼4%) in the NR animals of both sexes relative to sex-matched controls (Table 2), and showed a trend to be lower (also by ∼4%, p = 0.076) in the RSV male mice, but not the RSV female mice. Fed blood glucose levels at P35 were unaffected by treatments in either sex (not shown). Collectively, these results indicate that the neonatal RSV and NR supplementations applied did not compromise the animals’ growth during early postnatal life, and had only a minor impact on it.

The same panel of brown and beige adipocyte transcriptional markers analyzed in the iWAT primary cultures was analyzed in the iWAT depot of 35-day-old RSV and NR male mice and their respective control littermates (Figure 5). Genes related to mitochondria biogenesis and function were additionally analyzed in the iWAT tissue, since expression of genes in this category (Nrf1, Nrf2, Tfam, Tfb2m, Mfn2) is increased at adult age in iWAT of RSV and NR neonatally treated male mice compared to control littermates (Serrano et al., 2018). Of the genes analyzed, the only significant difference from expression levels in controls was a greater than 2-fold increase in Prdm16 mRNA levels in the iWAT of young NR male mice (Figure 5). Lep, Mest, and Lpl mRNA levels in iWAT, which were used as indicators of adipose tissue mass and expandability, were unaffected by treatments (Figure 5), in good concordance with the lack of effect of treatments on iWAT depot mass (Tables 1, 2). mRNA expression levels of a selection of genes in the above categories (Ucp1, Slc27a1, Mfn2, Lpl) were analyzed in iWAT of young (P35) female mice and found to be unaffected by treatments (data not shown).

Finally, because RSV and NR can both enhance energy metabolism through the activation of SIRT1 (Canto et al., 2012; Park et al., 2012; Price et al., 2012), transcriptional markers of SIRT1 activity were assessed in iWAT of young animals. SIRT1 represses Ucp2 gene transcription (Bordone et al., 2006) and deacetylates FOXO1 transcription factor, resulting in higher expression levels in cells of FOXO1 target genes such as Gadd45, Sod2, and Atgl (Calnan and Brunet, 2008; Chakrabarti et al., 2011). Consistent with neonatal RSV and NR treatments activating SIRT1 specifically in iWAT of male mice, Ucp2 mRNA levels were significantly decreased in iWAT of both groups of treated male mice relative to controls (Figure 5), but not in the iWAT depot of treated female mice (controls, 100 ± 14 and 100 ± 21; treated 134 ± 63 and 92 ± 21, for the RSV and NR treatments, respectively). However, none of the three FOXO1 target genes analyzed was significantly induced by the experimental treatments in iWAT of young male mice, and expression of Gadd45 was even significantly downregulated in the RSV male mice. Sirt1 expression was also measured, and a trend to down-regulation of its mRNA levels was apparent in the RSV male mice (p = 0.053). Overall, from the transcriptional analysis performed, we conclude that there is little evidence of iWAT browning or SIRT1 activation in iWAT of young male mice receiving neonatal RSV or NR supplementation.