Childhood and puberty are life stages characterised by transitions and the formation of dietary patterns. Unhealthy food intake patterns among children and adolescents, as well as insufficient physical activity, contribute to overweight and obesity in this life stage and increase the risk of non-communicable diseases later in life. Although obesity is easily diagnosed by the well-standardised BMI, some individuals are known to have skinny fat, which shows metabolic obesity but an average weight(Reference Williams and Periasamy1,Reference Foulis, Hughes and Friedl2) .
An unhealthy diet is one of the factors contributing to the development of obesity, commonly due to an overload of sugar or fat in the diet. The consumption of ultra-processed foods is rapidly increasing in middle- and high-income countries(Reference Monteiro, Moubarac and Cannon3). Studies using animal models have shown that rats fed a diet rich in carbohydrates such as sucrose experience an increase in body adiposity(Reference Schultz, Barbosa-Da-Silva and Aguila4). Variations in diet composition change the way energy is used, which can lead to an increase in fat deposits(Reference Rippe, Berger and Böiers5). Tenk e Felfeli(Reference Tenk and Felfeli6) observed that similar to humans, rats also show a higher intake of processed foods with a high fat and/or sugar content due to the high palatability of the food.
Excess dietary sodium is linked to several pathologies, including heart damage and kidney injury, leading to morphological alterations in these tissues. Recently, it has been demonstrated that high-sodium intake during puberty leads to hypertension and cardiac remodelling in adulthood(Reference Rosa, Cavalet and De Melo7). In addition, high dietary sodium levels during puberty induce short-term cardiac and renal effects(Reference Moreira, da Silva and Silveira8).
Excessive carbohydrate intake promotes lipogenesis, a process by which TAG are formed from excess glucose and processed by the liver and adipose tissues(Reference Westin, Nascimento and Fontes9,Reference Schutz10) . Thus, in addition to the consequences of these changes in glucose metabolism, the liver participates in pathophysiological mechanisms of fat accumulation(Reference Taskinen and Borén11,Reference Zheng, Ley and Hu12) . An increase in sugar intake leads to obesity and the development of ectopic lipid storage in the liver(Reference Dai, Harada and Takamatsu13). Increased lipogenesis and hypertriglyceridaemia result in greater adiposity and hypertrophy of white adipose tissue (WAT)(Reference Legeza, Marcolongo and Gamberucci14–Reference Braun, Oeckl and Westermeier16). Brown adipose tissue (BAT) plays a crucial role in thermogenesis; thus, its adipocytes have a different morphology than WAT adipocytes. However, during the onset of obesity, BAT adipocytes change their morphology and metabolism, assuming a white adipocyte phenotype, among which we can highlight a more significant deposition of TAG and cell hypertrophy(Reference Braun, Oeckl and Westermeier16–Reference De Almeida, Fabrício and Trombini18). In BAT, uncoupled protein (UCP-1) is expressed as a thermogenic function, and the adrenergic β-3 receptor pathway is essential for thermogenesis(Reference Braun, Oeckl and Westermeier16,Reference Šefčíková and Raček19) . The adrenergic β-3 receptor stimuli trigger the cAMP-dependent protein kinase signalling cascade that results in lipolysis of TAG from BAT and mitochondrial β-oxidation, resulting in thermogenesis(Reference Jocken, Goossens and Van Hees20). PPAR-α plays an essential role in BAT function, with potential effects on thermogenesis and browning, as well as HDL-cholesterol and TAG blood levels(Reference Souza-Tavares, Miranda and Vasques-Monteiro21,Reference Jahansouz, Xu and Hertzel22) .
Although the effects of high-salt/sugar diets during adulthood are well-defined in the literature, little is known about the role of the early stages of life in the development of diseases in adulthood. In the present study, we aimed to investigate whether high-sodium/sucrose intake during the first stages of life affects whole-body development and leads to end-organ injury.
Material and methods
All protocols were approved by the Ethics Committee on the Use of Animals of the Federal University of Goiás (protocol no. 023/15) and carried out in accordance with the standards recommended by the Brazilian Animal Experimentation Control Council and Animal Research: Reporting of In Vivo Experiments guidelines for scientific research.
Animals and diets
The 21-day-old weaned male Wistar rats were housed and cared for at the Animal Facility of the Department of Physiological Sciences, Federal University of Goiás. The rats were maintained under controlled conditions, with a 22 ± 2°C temperature and a 12/12 h light/dark cycle. Each rat was housed in polypropylene cages with environmental enrichment appropriate for rodents. The cages were cleaned regularly, and the bedding, composed of wood shavings valued for their absorbent properties and low dust content, was replaced twice a week to maintain a clean and comfortable environment. The animals were allocated into two experimental groups: (i) the control group (CO; n 14), which received tap water filtered through activated carbon, and (ii) the high-sodium/sucrose diet group (SS; n 14), which was administered a solution containing 1·8 % (w/v) NaCl (Synth) and 20 % (w/v) sucrose (Synth), diluted in tap water filtered through activated carbon. This protocol was based on previous studies utilising high-sodium solutions(Reference Rosa, Cavalet and De Melo7,Reference Moreira, da Silva and Silveira8) and high-sucrose diets(Reference Sharma, Okere and Barrows23,Reference Ritze, Bárdos and D’Haese24) . Animals were fed their diets for 60 d from weaning (21-day-old) until 81-day-old and had unlimited access to standard chow (Nuvilab) and water. At the end of the experimental period, all the groups received tap water and were fed a standard diet. Body weight (BW) and food intake were measured every week. Lee index was calculated, according to previous work, briefly as a ratio between the cubic root of BW and the nasoanal length(Reference Zhou, Han and Xu25). Here, we only used male rats to avoid sex hormone-induced confounding factors. Energy intake of carbohydrates, sodium, protein and fat was estimated using the information provided by the standard chow manufacturer and the concentration of sodium and sucrose in the solution offered to the SS group.
Glucose homeostasis assessment
At the end of the experimental period, a batch of animals from both groups (n 6 per group) was fasted for 12 h to perform the oral glucose tolerance test (GTT). Initially, a blood sample was collected for basal glycaemia at 0 min. A glucose solution (2 g/kg of BW) was injected orally. Blood samples were collected at 60 and 120 min after glucose injection. Blood samples were collected through a small puncture at the tip of the tail, and blood glucose levels were measured using a commercial glucometer (On Call Plus).
The next day, the animals used in the oral GTT were subjected to an intraperitoneal insulin tolerance test. The food was removed 2 h before the protocol was performed. As in the oral GTT, a blood sample was collected to measure basal glycaemia at time 0. Fast-acting insulin solution (2 U/kg of the BW, intraperitoneal) (Eli Lilly) was administered intraperitoneally. Caudal blood samples were collected 10, 20 and 30 min after insulin injection. Blood glucose levels were measured using a commercial glucometer (On Call Plus).
The animals used for the glucose homeostasis assessment were euthanised after the experiments and were not used for further analysis.
Renal and cardiac assessments, euthanasia and sample collection
At the end of the experimental period, another batch of animals (n 8 per group) was placed in a metabolic cage for 12 h to collect the urine. After 12 h of acclimation, urine of 12 h was collected. Tap water and standard rodent chow were used in this study. The animals were then moved to their cages, and the next day, they were dressed in an acquisition jacket, and electrocardiography was performed. A period of 10 min of acclimatisation was followed by 5 min of acquisition. Data from the electrocardiography were amplified (1000×) and acquired at 1 kHz (ADInstruments) and used to calculate the heart rate from RR intervals (Powerlab, ADInstruments).
After electrocardiography, the animals were fasted overnight and anaesthetised with sodium thiopental (40 mg/kg of the BW, intraperitoneal; Thiopentax, Cristália). Blood samples were collected from the inferior vena cava using a sterile needle and syringe, centrifuged for plasma collection (3000 g , 15 min) and stored at –20°C. The pancreas, liver, kidney, heart, adrenal glands, BAT, retroperitoneal, epididymal and inguinal subcutaneous (sc) WAT were dissected and weighed.
Biochemical assays and calculations
Levels of aspartate aminotransferase, alanine aminotransferase and plasmatic/urinary creatinine were measured by enzymatic-colorimetric methods using a commercial kit (DOLES) following the manufacturer’s instructions.
The glomerular filtration rate was calculated according to a previous study(Reference Hinojosa-Laborde, Jespersen and Shade26) using the following formula: ((Urinary Creatinine × Urinary Flow)/Plasma Creatinine). Urinary flow was calculated as the ratio of the total volume of urine collected at the time of collection (in min).
Western blot
BAT samples (n 4 per group) were homogenised in ice-cold lysis buffer and centrifuged (12 000 g , 4°C) to collect the supernatant. Protein concentrations were determined using the Bradford protein quantification assay. The samples were denatured in Laemmli buffer and heated at 95°C for 3 min. SDS-PAGE separated aliquots of 15 µg protein from each sample and then transferred them to nitrocellulose membranes. The membranes were incubated with blocking solution (5 % skim powdered milk, 10 mM Tris, 150 mM NaCl and 0·02 % Tween 20) under mild agitation for 90 min at room temperature. Subsequently, the membranes were washed with Tris-buffered saline and incubated overnight at 4°C with the primary antibodies anti-β3AR (#sc-515763), anti-UCP-1 (#sc-518024), anti-PPARα (#sc-398394), anti-PPARγ (#sc-7273), anti-AKT (#sc-81434) and anti-pAKT (#sc-7985), following manufacturer instructions. The membranes were then gently washed (3 × 5 min; 10 mM Tris, 150 mM NaCl and 0·02 % Tween 20) and incubated with the appropriate horseradish peroxidase conjugated secondary antibody for 90 min. The protein bands were visualised using chemiluminescence (ECL kit; Amersham Biosciences) and exposed to ImageQuant LAS (GE Healthcare). The area and density of the bands were quantified using ImageJ software (NIH). β-Actin was used as a load control.
Histological analysis
The pancreas, liver, kidney, heart and BAT samples were fixed in 10 % formaldehyde for 24 h and stored in 70 % alcohol. The samples were then dehydrated, diaphanised in xylol and embedded in histological paraffin (Synth, Diadema). Subsequently, the material was sectioned at a thickness of 5 µm using a microphone (RM2245; Leica Microsystems). The sections were stained with haematoxylin and eosin (pancreas, liver, kidney and BAT) or picrosirius red for collagen staining (heart).
Measurement of the cardiomyocyte diameter, pancreatic islets, brown adipocyte and glomerular areas
Photomicrographs of the cross-sections of the left ventricular cardiomyocytes were analysed according to a previous study(Reference Ferreira-Junior, Cavalcante and Costa27). The distance between the upper and lower parts of the cell membrane was measured at the height of the nucleus in each cardiomyocyte. Measurements were made at four different points in the middle layer of each cut and in three different cuts to assess the thickness of the intima-media layer. The mean cardiomyocyte diameter for each animal was calculated, and the results were compared between groups.
Pancreatic islets, brown adipocytes and glomerular areas were measured using at least 20 measurements from three different cuts per animal. The mean values were calculated for each animal, and the results were compared between the groups.
For these image analyses, the ICY Image Analysis software was used (Institut Pasteur) (<http://icy.bioimageanalysis.org/>).
Interstitial fibrosis measurement
Interstitial fibrosis was analysed by macro-processing using FIJI (v1.54f, ImageJ). Briefly, the images were processed using colour deconvolution to split the haematoxylin and picrosirius red staining components. The picrosirius red channel was subjected to a threshold for evidence of actual staining areas. The marked areas were then quantified. The percentage of interstitial fibrosis was estimated as the ratio of the marked area to the total area of the image. The mean and standard error of the mean of the percentage of interstitial fibrosis were calculated for each animal, and the results were compared between the groups.
Liver lipid inclusion
To evaluate liver lipid inclusion, stereological analysis was performed on haematoxylin and eosin-stained sections (1000× magnification), with a mesh of 594 points. At least ten slices from each animal were used. The mean value of lipid inclusion for each animal was calculated, and the results were compared between groups.
Stereological analysis was performed using Image-Pro Plus software (version 6.0; Media Cybernetics).
Statistical analysis
The sample size was calculated a priori using G * Power (v3.1). Based on previous reports(Reference Rosa, Cavalet and De Melo7,Reference Moreira, da Silva and Silveira8,Reference Ritze, Bárdos and D’Haese24,Reference dos Santos Moreira, de Carvalho Nunes and Lopes28) , our first outcome was increasing dietary sodium and sucrose intake. For this purpose, a minimum variation of 25 % in the intake of the offered solution was expected. With a power of 80 % and a significance level of 5 %, this allowed the detection of a difference of 20 % between the groups, with a minimum sample size of eight animals/group.
Data are expressed as mean (standard error of the mean) (sem). Unpaired Student’s t test was used to compare differences between groups. Two-way ANOVA was used to compare the differences in BW, food intake, intravenous GTT and intraperitoneal insulin tolerance test curves, followed by the Sidak post hoc test. Simple linear regression was used to perform the correlation analysis. Prism version 9.01 software (GraphPad) was used for the data analysis and graphical construction. Differences were considered statistically significant at P< 0·05.
Results
High-sodium/sucrose diet impacts body development and increases adipose tissue weight
The SS group exhibited delayed BW gain throughout the experimental period (Fig. 1(a)). In addition, owing to the increased intake of high-sucrose and sodium mixtures (Fig. 1(b)), food intake was reduced in these animals (Fig. 1(c)) than in the CO group. At the end of the experimental period, the SS group had increased carbohydrate levels (CO 0·384 (sem 0·012) v. SS 0·707 (sem 0·029) g/b. wt./week; < 0·0001; Fig. 1(d)) and sodium intake (CO 0·18 (sem 0·007) v. SS 6·85 (sem 0·250) g/week; P< 0·0001; Fig. 1(e)); however, due to the decreased chow intake, they reduced the weekly intake of proteins (CO 0·0698 (sem 0·003) v. SS 0·0511 (sem 0·005) g/b. wt./week; P= 0·0051; Fig. 1(f)) and fat (CO 0·0222 (sem 0·0009) v. SS 0·0163 (sem 0·0016) g/b.wt./week; P= 0·0051; Fig. 1(g)). Overall energy intake did not differ between the groups (Fig. 1(h)).
The high-sodium/sucrose (SS) diet affects body development and leads to an increase in adipose tissue weight. The SS group presented lower body weight gain (a), high fluid intake (b) and lower food intake (c) during the experimental period compared with the control (CO) group. In the last week, the SS group presented increased carbohydrate and sodium intake (d)–(e) and reduced protein and fat intake (f)–(g) despite no alterations in overall energy intake (h). This dietary pattern affected body weight (i), nasoanal length (j) and tibial length (k). However, SS animals presented increased retroperitoneal (rp), epididymal (e) and inguinal subcutaneous (sc) white adipose tissue (WAT) (l)–(n) compared with CO animals. However, the Lee index (o), mainly used to indicate obesity in rats, was not altered. Two-way ANOVA was used for time-dependent analysis, and other analyses were performed using the Student’s t test (n 8 animals per group). *P< 0·05, **P< 0·01, ***P< 0·001 and ****P< 0·0001.
At the end of the experimental period, the BW (CO 315·3 (sem 11·3) v. SS 196·4 (sem 8·95) g; P< 0·001; Fig. 1(i)), nasoanal length (CO 20·4 (sem 0·27) SS 17·5 (sem 0·39) cm; P< 0·001; Fig. 1(j)) and tibia length (CO 3·77 (sem 0·034) v. SS 3·27 (sem 0·0348) cm; P< 0·0001; Fig. 1(k)) were reduced in SS group. However, despite lower BW, the adiposity in the SS group was higher than in the CO group, indicated by retroperitoneal (CO 466·3 (sem 60·46) v. SS 1125·0 (sem 87·01) mg/100 g b.wt.; Fig. 1(l), epididymal (CO 547·8 (sem 62·32) v. SS 932·8 (sem 103·70) mg/100 g b.wt.; Fig. 1(m)) and sc (CO 282·7 (sem 38·32) v. SS 583·0 (sem 63·11) mg/100 g b.wt.; Fig. 1(n)) fat pads increased the relative weight. In contrast, the Lee index was similar between groups (Fig. 1).
High-sodium/sucrose diet correlates with increased white and brown adipose tissue weight
Similar to the CO group, we found that the SS group had a positive correlation between adipose tissue weight and BW. However, despite a positive correlation, the retroperitoneal, epididymal and sc fat pads presented left-displaced values in the SS group compared with those in the CO group (P< 0·001; Fig. 2(a)–(c)). Regarding BAT, there was a strong correlation with the BW in SS animals, with BAT weights higher than those in CO animals (P= 0·002; Fig. 2(d)). Similarly, the adrenal gland presented a positive correlation with BW in both CO and SS animals; however, SS animals were more prone to show a slightly increased adrenal weight (P= 0·028; Fig. 2(e)). There was a weak correlation between tibia length and BW in SS animals (P= 0·032; Fig. 2(f)) compared with that in CO.
The high-sodium/sucrose (SS) diet correlated with increased white and brown adipose tissue weights. Visceral (a)–(b) and subcutaneous (c) WAT and interscapular BAT (d) were correlated with BW, as observed in the control group, although the SS group had left-displaced points. Inversely, as with adipose tissue, the adrenal gland (e) and tibia length (f) were weakly correlated with body weight. A simple linear regression was used (n 8 animals per group). BW, body weight; rpWAT, retroperitoneal white adipose tissue; eWAT, epididymal white adipose tissue; scWAT, subcutaneous white adipose tissue; BAT, brown adipose tissue. *P< 0·05, ***P< 0·001 and ****P< 0·0001.
High-sodium/sucrose diet does not affect glucose homeostasis but leads to hepatic lipid inclusion
There was no significant difference between the groups during the oral GTT (Fig. 3(a) and (e)) and insulin tolerance test (Fig. 3(b) and (f)). In addition, we observed that the basal blood glucose levels are similar under long (CO 3·59 (SEM 0·13) v. SS 3·58 (SEM 0·27) mM; Fig. 3(a)) and short fasting (CO 5·16 (SEM 0·32) v. SS 5·06 (SEM 0·37) mM; Fig. 3(b)). Despite the higher values of normalised pancreas weight (CO 168·7 (sem 6·202) v. SS 226·4 (sem 14·13) mg/100 g b.wt.; P= 0·0015; Fig. 3(c)), SS animals showed no difference in pancreatic islet area compared with the control group (Fig. 3(d)).
The high-sodium/sucrose (SS) diet does not affect glucose homeostasis but leads to hepatic lipid inclusion. At the end of the experimental period, no differences were detected during the oral insulin tolerance test (oGTT) (a), (e) or intraperitoneal insulin tolerance test (ipITT) (b), (f). Despite the high pancreas:body weight (BW) ratio in the SS group compared with the control (CO) group, no difference was detected in the pancreatic islet area (d). The liver:BW ratio (g) and hepatic lipid content were higher in SS animals than in CO. However, no differences were detected in the plasma markers of liver injury (i)–(j). Two-way ANOVA was used for time-dependent analysis, and other analyses were performed using Student’s t test (n 6–8 animals per group). *P< 0·05 and **P< 0·01. Representative photomicrographs of haematoxylin and eosin-stained slices of the pancreas and liver are shown at the bottom (black arrows indicate microvesicular lipid inclusion, while black bars indicate 50 μm).
The SS diet also affected the normalised liver weight (CO, 3720 (sem 167·7) v. SS 4297 (sem 155·7) mg/100 g; P< 0·05; Fig. 3(g)) and hepatic lipid accumulation (CO 0·4 (sem 0·04) % v. SS 2·5 (sem 0·25) %; P< 0·0001; Fig. 3(h)) compared with the CO diet; however, no difference was detected in plasma markers of hepatic injury (Fig. 3(i) and (j)).
High-sodium/sucrose diet affects renal function and increases heart rate
Despite increased normalised kidney weight (CO 347·1 (sem 6·63) v. SS 419·2 (sem 17·77) mg/100 g b.wt.; P< 0·01; Fig. 4(a)), the SS animals had lower glomerular (CO 5555 (sem 266·1) v. SS 3567 (sem 189·2) μm2; P< 0·001; Fig. 4(b)) and capsular space (CO 1136·0 (sem 124·10) v. SS 779·8 (sem 68·27) μm2; P< 0·05; Fig. 4(c)) areas compared with CO animals, as well as pericapsular collagen marking in the glomeruli (CO 6762 (sem 342·1) v. SS 4411 (sem 237·1) μm2; P< 0·01; Fig. 4(d)). In parallel to those morphologic changes, the SS group exhibits impaired urinary function, marked by lower creatinine clearance (urinary; CO 6·34 × 103 (SEM 0·651 × 103) v. SS 1·97 × 103 (SEM 0·444 × 103) μM; P < 0·0001; Fig. 4(e)/plasmatic; CO 37·14 (SEM 6·54) v. SS 57·47 (SEM 4·598) μM; P < 0·05; Fig. 4(f)), lower glomerular filtration rate (CO 15·160 (sem 5·717) v. SS 0·775 (sem 0·4305) ml/min; P= 0·0360; Fig. 4(g)) and urinary flow (CO 0·050 (sem 0·0051) v. SS 0·023 (sem 0·0079) ml/min; P< 0·05; Fig. 4(h)).
The high-sodium/sucrose (SS) diet affects renal function and increases heart rate. Despite a higher kidney:body weight (BW) ratio (a), the SS group showed a reduced glomerular area (b), capsular space (c) and pericapsular collagen (d) compared with the control (CO) group. In addition, the SS diet shifts urinary-to-plasmatic creatinine levels (e)–(f), which represent an impaired glomerular filtration rate (g) and reduced urinary flow (h). Despite a lower heart:BW ratio (i), the SS animals showed an increased heart rate (j). However, no differences were observed in the cardiomyocyte diameter (k) or interstitial collagen deposition (l). The Student’s t test was used (n 8 animals per group). *P< 0·05, **P< 0·01, ***P< 0·001 and ****P< 0·0001. Representative photomicrographs of picrosirius red–stained slices of the kidney and heart are at the bottom (collagen in red; black bars measure 50 μm).
In addition to renal effects, the SS diet also affects the heart. The SS group present lower normalised heart weight (CO 316·1 (sem 14·64) v. SS 242·3 (sem 15·06) mg/100 g b.wt.; P< 0·05; Fig. 4(i)) and consequently higher heart rate (CO 358·8 (sem 9·56) v. SS 419·5 (sem 13·91) bpm; P< 0·05; Fig. 4(j)), without morphological alterations on cardiomyocytes (Fig. 4(k)), or interstitial collagen deposition (Fig. 4(l)).
High-sodium/sucrose diet induces interscapular brown adipose tissue hypertrophy but not whitening
The SS group exhibited increased interscapular BAT normalised weight (CO 65·47 (sem 5·632) v. SS 194·6 (sem 11·17) mg/100 g b.wt.; P< 0·0001; Fig. 5(a)), as well as hypertrophy of brown adipocytes (CO 381·9 (sem 17·97) v. SS 500·3 (sem 41·11) μm2; P< 0·0001; Fig. 5(b)). Molecular analysis showed that BAT of the SS group had a strong trend of increasing β3 adrenoreceptors (P= 0·0943; Fig. 5(c)), although no changes in the thermogenic protein UCP-1 (Fig. 5(d)). In addition, an increase in PPARα expression (CO 1·00 (sem 0·1311) v. SS 1·51 (sem 0·1238) % of control; P< 0·05; Fig. 5(e)) without changes in PPARγ, AKT or the pAK:AKT ratio (Fig. 5(f)–(h)) was observed in SS animals compared with that in CO.
The high-sodium/sucrose (SS) diet-induced interscapular brown adipose tissue (BAT) hypertrophy but not whitening. The SS group showed both whole BAT (a) and brown adipocyte hypertrophy (b), despite a slight increase in the adrenergic β-3 receptor (c), without UCP-1 (d) changes. Interestingly, PPARα (e) levels were higher in the SS group than in the CO group regarding protection against whitening. No differences were observed in PPARγ (f), AKT (g) and pAKT:AKT ratio (h). The Student’s t test was used (n 4–8 animals per group). *P< 0·05 and ****P< 0·0001. Representative photomicrographs of haematoxylin and eosin-stained slices of BAT are at the bottom (brown adipocytes highlighted with a cyan dotted line; black bars measure 50 μm). UCP-1, uncoupled protein; β-3-AR, adrenergic β-3 receptor.
Discussion
Here, we evaluated the early effects of high-sodium and high-sucrose intakes during puberty and early adulthood on the health status of Wistar rats. Surprisingly, we found that the SS diet affected energy intake, leading to lower protein intake, lower lean mass and increased fat accumulation in the adipocytes and liver due to high carbohydrate intake. In addition to metabolic impairment, glucose homeostasis was not affected. There was impaired creatinine clearance in the kidney due to a reduced glomerular filtration rate and urinary flux. In the heart, the SS diet affected heart rate but not morphology. Like WAT, BAT was hypertrophied; however, PPARα was upregulated, supporting the tissue morphology.
In our study, the SS group had a lower BW and food intake than the CO group, in contrast to the higher SS consumption. Notably, the forced consumption of a solution containing a high concentration of sodium and sucrose instead of tap water will increase the liquid consumption owing to the high osmolarity of the solution and consequent increase in plasma osmolarity, which activates thirst mechanisms. Interestingly, we observed a slight reduction in liquid consumption and hypothesised that this is a mechanism related to momentary aversion to elevated sodium and sucrose levels in the diet, as the same event tends to repeat itself in the last week of treatment. Moreira et al. (Reference Moreira, da Silva and Silveira8) conducted a similar experimental design using the same sodium concentration in drinking water and found that the experimental groups did not have reduced food intake. This suggests that, in our study, the reduction in food intake was caused by the high carbohydrate intake from the SS diet rather than sodium intake. In addition, owing to the lower food intake and consequent reduction in protein intake, the SS animals probably had sarcopenia, which decreased the BMR and favoured fat accumulation, as observed in our study. In contrast, animals fed a low-protein diet for 1 month during puberty presented lower fat than those in the control groups(Reference de Oliveira, de Moura and Miranda29). In the study by Oliveira et al. (Reference de Oliveira, de Moura and Miranda29), despite high carbohydrate intake, the animals did not show adipose tissue hypertrophy, indicating that the source of carbohydrates can modulate fat accumulation because the source used by the authors was a mixture of sucrose and maize starch. Surprisingly, both groups had similar Lee index, which mimics individuals with normal BMI but increased metabolic impairments(Reference Williams and Periasamy1,Reference Foulis, Hughes and Friedl2,Reference Zhou, Han and Xu25) . Here, we analysed the correlation between fat pad weight and BW and found that the adipose tissue of the SS animals was positively correlated with BW, similar to the CO group. Notwithstanding, the point of interception of the curves differed between the groups, showing that despite the lower BW of SS animals, their adipose tissues were heavier than those of the CO groups. This result highlights fat accumulation in the SS group independent of BW gain, revealing a skinny fat phenotype.
We did not observe any differences between the groups during the insulin test and GTT, meaning the SS diet could not alter glucose homeostasis in the SS group. When we analysed the pancreas morphology, we observed an increased normalised weight without pancreatic islet hypertrophy. We did not observe any signs of ectopic fat accumulation in the pancreas of the SS animals, and we attribute the increased relative pancreatic weight to slight variability in gross pancreatic weight among the animals. Although the SS diet did not affect glucose homeostasis, several studies have reported normoglycemia in the context of hyperinsulinemia(Reference Chen, Hosokawa and Bumbalo30–Reference Alwahsh, Dwyer and Forbes32). While the combination of sodium and sucrose does not appear to disrupt glycaemic balance directly, one study examining the association between soft drink consumption, weight gain and DM reported that excessive consumption of these products contributes to increased body mass, raising blood glucose levels(Reference Schulze, Manson and Ludwig33).
The liver is one of the organs most injured during high carbohydrate consumption. In our model, we observed an increase in the normalised weight of the liver and microvesicular lipid inclusion. However, no differences in plasma markers of liver injury were observed, showing that despite the beginning of fat accumulation in the liver, there was no extensive damage to the tissue. These results indicate a liver de novo lipogenesis alteration, which needs to be assessed in future studies. A study carried out in mice showed that an overload of 50 % sucrose or 50 % fructose for 15 weeks contributed to increased steatosis and elevated plasma levels of alanine aminotransferase and aspartate aminotransferase(Reference Schultz, Barbosa-Da-Silva and Aguila4). In addition to liver changes, a sucrose-rich diet leads to increased adiposity and weight gain compared with a fructose-rich diet(Reference Schultz, Barbosa-Da-Silva and Aguila4).
The kidney is most injured during sodium or glucose overload(Reference Kitada, Daub and Zhang34–Reference Rosca, Mustata and Kinter37). In our study, we observed increased normalised weight of the kidney; however, a decrease in both glomerular and capsular space and pericapsular collagen decrease highlights the morphological effects of the SS diet in the experimental group. Similar to the morphological impact of the SS diet, kidney function was impaired, as shown by the lower creatinine clearance due to the reduced glomerular filtration rate and urinary flow. We observed a lower heart:BW ratio in the heart, although no alterations in cardiomyocyte diameter and collagen deposition were observed. However, the SS animals had higher heart rates than the CO animals. Moreira et al. (Reference dos Santos Moreira, de Carvalho Nunes and Lopes28) showed that a high-sodium diet increases collagen deposition and impairs cardiac relaxation.
The higher the nutrient availability, the higher the input for adipocyte energy accumulation(Reference Breton38). However, BAT regulates energy expenditure because of its high metabolic activity under certain stimuli(Reference Jahansouz, Xu and Hertzel22). Adrenergic stimulus increases thermogenesis in the BAT through β3 receptor activation and the upregulation of uncoupling protein type 1 (UCP-1)(Reference Collins39). Our study observed that the SS group had a higher normalised weight of BAT and brown adipocyte area. Even so, the adrenergic β3 receptor showed increased expression in the SS group, although no differences in UCP-1 expression were noted. In this context, hypertrophy of BAT is commonly linked to tissue whitening, which means the loss of mitochondrial function or number. Whitening leads to impaired thermogenesis and lipid handling in brown adipocytes, which function as a stock of lipids in unilocular vesicles instead of multilocular vesicles that facilitate lipid metabolism(Reference Souza-Tavares, Miranda and Vasques-Monteiro21). In our study, we did not observe any characteristics of BAT whitening in the SS group. PPARα is the primary protector against BAT whitening, mainly in animals fed high-fat diets(Reference Lehnig and Stanford17,Reference Chen, Tan and Wan40) . PPARα is the primary regulator of adipose tissue energy expenditure and regulates genes involved in thermogenesis and catecholamine response(Reference Pawlak, Lefebvre and Staels41). In our study, SS animals had an increased amount of PPARα, which probably drives the maintenance of the brown adipocyte phenotype. However, the SS diet did not affect PPARγ, the primary regulator of UCP expression. To the best of our knowledge, this is the reason for hypertrophied brown adipocytes(Reference Magliano, Bargut and de Carvalho42).
Conclusion
Altogether, our results show that an SS diet early in life affects body development by impairing WAT metabolism. In addition, the increase in de novo lipogenesis by the liver was partly buffered by BAT metabolism, highlighting the protective role of BAT metabolism under the study conditions. Although BAT could not increase lipid consumption, the maintenance of the brown phenotype shows a protective role in the maintenance of sympathetic tone in this tissue. This experimental model resembles obesity-related symptoms that occur in people with an average BMI and is a non-apparent obese model (skinny fat) for studying markers of metabolic syndrome, such as increased heart rate, liver lipid inclusion, impaired renal function and increased adiposity.
Acknowledgements
This research was funded by Fundação de Amparo à Pesquisa do Estado de Goiás (FAPEG grant nos.: 201710267000517-GRP, 201810267001636-GRP, 2022151000060-GRP), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq grant nos.: 447496/2014-0-GRP, 309989/2016-7-GRP, 312130/2019-8-GRP, 315335/2023-8-GRP; FAPESP/FAPEG grant no.: 2019/24154-3) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES).
K. V. N. C.: conceptualisation, methodology, investigation, writing – original draft; M. D. F-J.: conceptualisation, methodology, investigation, writing – review and editing; M. C. S. M.: conceptualisation, methodology, investigation, writing – review and editing; S. M. M.: writing – review and editing; J. O. F.: writing – review and editing; R. A. M.: resources, writing – review and editing; P. C. L.: resources, writing – review and editing; E. G. M.: resources, writing – review and editing; C. H. X.: writing – review and editing; E. C.: supervision, writing – review and editing; R. M. G.: conceptualisation, methodology, writing – review and editing; G. R. P.: conceptualisation, methodology, writing – review and editing. All authors have read and approved the final version of the manuscript.
The authors declare that they have no conflicts of interest. Data can be accessed upon request by the interested party to the corresponding author.
