Interactive effects of chronic stress and a high-sucrose diet on nonalcoholic fatty liver in young adult male rats

Abstract

Glucocorticoids have been implicated in nonalcoholic fatty liver diseases (NAFLD). The influence of a palatable diet on the response to stress is controversial. This study explored whether a high-sucrose diet could protect from hepatic steatosis induced by chronic restraint stress in young adult rats. Male Wistar rats aged 21 days were allocated into four groups (n = 6–8 per group): control, chronic restraint stress, 30% sucrose diet, and 30% sucrose diet plus chronic restraint stress. After being exposed to either tap water or sucrose solution during eight weeks, half of the rats belonging to each group were subject or not to repeated restraint stress (1 h per day, 5 days per week) during four weeks. Triacylglycerol (TAG), oxidative stress, activity of 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD-1), infiltration of immune cells, and glycogen amount in the liver were quantified. Serum concentrations of corticosterone and testosterone were also measured. The stressed group showed normal serum concentrations of corticosterone and did not have hepatic steatosis. However, this group showed increased glycogen, inflammation, mild fibrosis, oxidative stress, and a high activity of 11β-HSD-1 in the liver. The group exposed to the high-sucrose diet had lower concentrations of corticosterone, hepatic steatosis and moderate fibrosis. The group subject to high-sucrose diet plus chronic restraint stress showed low concentrations of corticosterone, hepatic steatosis, oxidative stress, and high concentrations of testosterone. Thus, restraint stress and a high-sucrose diet each generate different components of nonalcoholic fatty liver in young adult rats. The combination of both the factors could promote a faster development of NAFLD.

Keywords:

• 11β-HSD-1
• corticosterone
• hepatic steatosis
• hepatic fibrosis
• NAFLD
• oxidative stress

Introduction

Psychosocial stress can aggravate inflammation and fibrosis in a cirrhotic liver (Chida, Sudo, & Kubo, 2006; Vere, Streba, Streba, Ionescu, & Sima, 2009). Thus, an excess of glucocorticoid has been associated with the development of nonalcoholic fatty liver diseases (NAFLD; Ahmed et al., 2012). The local synthesis of glucocorticoids in the liver is mediated by the enzyme 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD-1), which interconverts the active cortisol (in humans) or corticosterone (in rats), and the inert cortisone (in humans) and 11-dehydrocorticosterone (in rats). This is associated with the development of metabolic syndrome, signs of inflammation, and oxidative stress in patients with NAFLD (Staab & Maser, 2010; Tarantino & Finelli, 2013). However, only 20% of patients diagnosed with Cushing’s syndrome show hepatic steatosis (Rockall et al., 2003), suggesting that other factors may explain fat accumulation.

Evidence about the influence of a hypercaloric and palatable diet on the attenuation of the response to stress is controversial. Some authors have shown that a high-carbohydrate diet reduces the serum concentration of adrenocorticotropic hormone (ACTH) but not corticosterone, as promoted by chronic stress in adult rats (Rho, Kim, Choi, & Lee, 2014; Zeeni et al., 2013). In contrast with these findings, it has been shown that a chronic high-carbohydrate diet can or cannot increases the serum corticosterone concentration (Ochoa, Lallès, Malbert, & Val-Laillet, 2015; Zeeni et al., 2013), enhances 11β-HSD-1 protein expression in the liver (Vasiljević et al., 2014), and down-regulates glucocorticoid receptors in hepatic and adipose tissue (Kovaćević, Nestorov, Matić, & Elaković, 2014; Vasiljević et al., 2014) of adult rats. Notably, a high-sucrose diet reduces the serum concentration of corticosterone, the amount of triacylglycerol (TAG), and the 11β-HSD-1 activity in the liver promoted by chronic stress in infant rats (Corona-Pérez et al., 2015). This differential effect of a palatable diet on the corticosterone concentration in stressed infant or adult subjects could be related to: (1) a different activity of adrenal glands between young and old rats (Meikle et al., 2007); (2) an interaction between the hypothalamic-adrenal-axis (HPA) and the 11β-HSD-1 activity in the liver (Maniam, Antoniadis, & Morris, 2014); or (3) an influence of testosterone on the corticosterone concentration that affects the response to stressors (Ajdžanovic et al., 2015; Green & McCormick, 2016; Meikle et al., 2007). Thus, the present study aimed to explore whether a high-sucrose diet could protect against the hepatic steatosis induced by chronic restraint stress in young adult rats by exploring the influence of a high-sucrose diet on the serum concentration of corticosterone, hepatic activity of 11β-HSD-1, hepatic steatosis, and liver inflammation. In addition, the potential relationship between corticosterone and testosterone was analyzed.

Methods

Animals

The experiment was carried out in the offspring from pregnant Wistar rats (Rattus norvegicus) living in the breeding colony maintained at the Centro Tlaxcala de Biología de la Conducta from the Universidad Autónoma de Tlaxcala, México. Dams were housed alone in polypropylene cages (65 cm ×25 cm ×15 cm) and maintained under a 12 h light-12 h dark cycle (with the lights off from 08:00 h to 20:00 h). This dark condition was used because rats are nocturnal, and we decided it was appropriate to apply stress and to make key measurements in this active phase. We maintained a constant temperature in the range of 20 ± 2 °C in our experimental room. A standard diet (Purina Laboratory chow 5001) and water were provided ad libitum. Size of litters was adjusted to 10 offspring soon after birth. Pups were weaned at postnatal day 21; male pups weighing 60–80 g were housed individually in polypropylene home cages (37 cm ×27 cm ×16 cm), and randomly assigned among four experimental groups. Only one rat per litter was included in each group. According to the experimental group the rats belonged to, they had access to either tap water or sucrose solution for eight weeks, plus a standard diet (Purina Laboratory chow 5001). Rats were randomly allocated to the following four groups (Figure 1): Control-standard chow (C, n = 6–8), chronic restraint stress (St, n = 6–8), 30%-sucrose diet (S30, n = 6–8), and 30%-sucrose diet plus chronic restraint stress (S30 + St, n = 6–8). The Research Ethics Committee from the Universidad Autónoma de Tlaxcala approved all the experimental procedures according to the National Guide for the Production, Care and Use of Laboratory Animals (Norma Oficial Mexicana NOM-062-200-1999, México).

Figure 1. Experimental design. Control-standard chow (C), chronic restraint stress (St), 30%-sucrose diet (S30), and 30%-sucrose diet plus chronic restraint stress (S30 + St).

Dietary protocol

All rats were fed with a standard diet. Rats from the control and stressed groups had continuous access to tap water, whereas those rats subject to high-sucrose diet and high-sucrose diet plus chronic restraint stress drank exclusively from a solution of 30% sucrose during eight weeks. Body weights were recorded daily to calculate the difference between initial and final weights (Δ weight). Food and water intake were also recorded daily. Individual caloric intake was calculated from the consumption of food (g/100 g body weight × 3.4 kcal/g of food) and sucrose consumption (mL of water/100 g body weight × g of sugar/mL of water × 4 kcal/g of sugar) (Corona-Pérez et al., 2015).

Restraint stress procedure

An earlier study reported the protocol to generate stress by motility restraint (Corona-Pérez et al., 2015). Briefly, rats were introduced into a plastic tube (16 × 6 cm); the wide end of the tube was sealed with adhesive tape to limit rat movements; the other end of the tube (1.8 cm external diameter) was kept open to allow rats to breath. Half of the rats that drank tap water and half of the rats that drank the 30% sucrose solution were daily exposed for 1 h to the restraint stress procedure at 11:00 h (i.e. in the dark phase) on five consecutive days per week for the last four weeks before euthanasia. No stressful stimulus was provided during weekends. Rats were returned to their home cages immediately after the stressful treatment. See experimental design (Figure 1).

Serum glucose concentration

On the last day of exposure to stress, rats were fasted overnight (12 h). Next morning (between 08:00 and 09:00 h), a blood sample was obtained by making a small incision at the end of the tail. Glucose test strips (Accutrend) were used to collect a drop of blood and glucose concentration was measured using an Accutrend GCT analyzer (Roche Diagnostics, USA).

Hormone analysis

Unanesthetized rats were killed by decapitation using a rodent guillotine-like device. Trunk blood was collected and serum was obtained by centrifugation (2,500 g for 15 min) at room temperature. Serum samples were divided into 200 µL aliquots and then stored at −70 °C (Corona-Pérez et al., 2015). Concentrations of serum corticosterone and testosterone were measured using commercially available ELISA kits in accordance with the manufacturer directions (corticosterone: Enzo Life Sciences Inc, USA; testosterone: Cayman Chemical, Ann Arbor, USA). Sensitivities of assays were 27 pg/mL and ∼6 pg/mL, respectively. Intra-assay and inter-assay variability was estimated as 2.2% and 5.3% for corticosterone and 7.7% and 4.9% for testosterone, respectively. 

Tissue collection

The liver was removed and weighed (g tissue/100 g body weight). Fat depots from the pericardial, gonadal and peritoneal cavity were excised. Adiposity index was calculated as the [total tissue weight (g) divided by the body weight (g)] × 100 (Corona-Pérez et al., 2015).

Hepatic 11β-HSD-1 activity

The activity of 11β-HSD-1 in the liver was determined as previously reported (Corona-Pérez et al., 2015). Briefly, approximately 3 g of the middle portion of the lower lobe of the liver was homogenized. Subcellular fractions (nucleus mitochondria, microsomes and cytosol) were obtained by centrifugation. Microsomal protein (30 mg) was transferred to glass tubes containing 700 μL of phosphate-buffered saline (PBS, pH 7.4). For the assay, each sample was mixed with 4 mM-nicotinamide adenine dinucleotide phosphate (NADPH; Sigma, USA) and 100 μL of PBS containing 3700 Bq (0.1 μCi) [1,2-³H]-cortisone (Perkin Elmer, USA) and unlabeled cortisone (Sigma, USA) for a final steroid concentration of 100 nM. Tubes were heated in a water-bath at 37 °C for 60 min and reactions were stopped by the addition of 2 mL of ice-cold chloroform (J. T. Baker, México). To partition the organic and aqueous phases, samples were centrifuged at 1000 × g for 30 min. After aspirating the aqueous supernatant fraction, the organic extracts were evaporated overnight at room temperature. Lipid and steroid residues were re-suspended in 20 μL of ethyl acetate containing 1 mM-cortisone (Sigma, USA) and resolved by thin layer chromatography (TLC), using Silica 60 TLC plates (Merck, México) in an atmosphere of 92:8 (v/v) chloroform 95% (v/v) ethanol (Merck, México). The spots corresponding to cortisol were scraped off and the [³H]-cortisol was measured using a Bioscan 200 TLC radiochromatogram scanner (LabLogic, USA). The concentration of protein was measured in duplicate using the Bradford method.

Liver glycogen

The amount of glycogen was determined by an acid-hydrolytic method (Passonneau & Lauderdale, 1974; Zhang, 2012) with slight modifications. For each sample, 200 μL of 2 M hydrochloric acid (HCl; J.T. Baker, México) in a 1.5 mL Eppendorf tube with a locking lid were added. As the control to measure free glucose, 2 M HCl was substituted by 2 M sodium hydroxide (NaOH; Sigma-Aldrich, USA). Tubes were heated in boiling water for 5 min. Frozen liver samples (0.1 g each of the middle portion of the lower lobe, approximately) were prepared on dry ice. Samples were transferred into tubes with hot HCl or NaOH to be homogenized. Tubes were tightly sealed and the samples were boiled in water for 1 h and then the tubes were shaken vigorously every 10 min during the whole process. At the end, the samples were centrifuged at 10,000 × g during 5 min. The supernatant containing hydrolysis product was assayed using a commercial kit (STANBIO Glucosa Liquicolor, México). The amount of liver glycogen is reported as mmoles of glucosyl units/kg wet liver weight.

Liver TAG

The hepatic TAG amount was measured as previously reported (Corona-Pérez et al., 2015). Frozen liver samples (0.5 g each, approximately) from the middle lobe of the lower lobe were disrupted in 5.4 mL of chloroform–methanol (2:1; v/v; J. T. Baker, México) using an electronic homogenizer (Tissue Tearor, Biospec Products Inc, México). The organic and inorganic layers were separated by adding 0.7% sodium chloride (NaCl; J. T. Baker, México). The samples were vortexed and centrifuged at 1200 × g at 4 °C for 15 min. The organic extracts were carefully transferred to a new tube and evaporated. The residuals were suspended in a mixture of isopropanol (J. T. Baker, México) and 10% Triton X-100 (PBST; Sigma, USA). The hepatic TAG amount was assayed in duplicate using a commercial kit (Elitech Clinical Systems, México). Sensitivities of assays was 25 mg/dL. The hepatic TAG amount was reported as mg of triacylglycerol/g of liver tissue. Intra-assay and inter-assay variability was estimated as 0.21% and 0.23%, respectively.

Liver histology

The lower lobe of the liver was immersed in Bouin Duboscq fixative for 24 h at room temperature. Afterwards, the tissue was embedded in Paraplast X-tra (Sigma-Aldrich, St. Louis, MO, USA) and 7 µm thick longitudinal sections were obtained with a microtome (Leica RM2135, Germany), mounting four sections per slide. One series of slides was stained with Masson´s trichrome to evaluate the histological organization and the presence of connective tissue in the liver of each rat. Photomicrographs were obtained at 100 and 400 magnification (10× and 40× objectives, respectively) using an optical microscope (Zeiss Axio Imager A1, USA) equipped with a digital camera (Tuczon Olympus, USA). The degree of fibrosis was semiquantitatively scored by estimating the presence and location of collagen in the hepatic parenchyma. The grade of fibrosis was estimating as described elsewhere (Tiniakos et al., 2010). The fibrosis score was graded using a scale from 0 to 4, scoring none (0) when 1a, mild (delicate) zone 3 perisinusoidal fibrosis; 1 b, moderate (dense zone 3 perisinusoidal fibrosis; 1c, portal/periportal fibrosis only; 2, zone 3 perisinusoidal fibrosis with portal/periportal fibrosis; 3, bridging fibrosis; and 4, Cirrhosis.

Oxidative stress markers

Lipid peroxidation in the liver was measured in vitro using 2-thiobarbituric acid reactive substances (TBARS) and a colorimetric assay as previously reported in homogenate and subcellular fractions (Luna-Moreno, Vázquez-Martínez, Báez-Ruiz, Ramírez, & Díaz-Muñoz, 2007). A sample of the homogenate (approximately 1 mg protein) was incubated for 30 min at 37 °C in 1 mL of 0.15 M Tris, pH 7.4; incubation was ended by adding 1.5 mL of 20% acetic acid (J.T. Baker, México) (adjusted to pH 2.5 with potassium hydroxide (KOH) and 1.5 mL of 0.8% of TBARS (Sigma-Aldrich, USA). The samples were kept for 45 min in a boiling water bath, and then 1 mL of 2% potassium chloride (KCl) was added to each sample. The colored complex formed was extracted with butanol–pyridine (J.T. Baker, México) (15:1, v/v) and quantified at 532 nm. Malondialdehyde (J.T. Baker, México) was used as standard (extinction coefficient: 1.56 × 105 cm⁻¹ M⁻¹) and expressed as nmol/mg protein. Lipid peroxidation in vivo was determined by measuring the presence of conjugated dienes by light absorption at 233 nm in Folch-extracts (chloroform–methanol 2:1, v/v J. T. Baker, México) of homogenates (Klaassen & Plaa, 1969).

Immune cell counts

Further series of slides were stained with hematoxylin and eosin to quantify immune cells within sinusoids. Slides were covered with the mounting medium and a coverslip, observed under a light microscopy (Olympus BH-2, Japan). Using as reference the hepatic central vein, photographs at 1000× magnification (100× oil-immersion objective) from ten histological fields were taken with a digital camera for counting all types of immune cells in the lumen of blood vessels per rat. A specific counting of neutrophils was done by identifying a polymorphic nucleus. The number of immune cells per field was then calculated (Rodríguez-Castelán et al., 2016). Additionally, other series of slides were stained with alcian blue pH 2.5 and counterstained with neutral red to demonstrate the presence of mast cells (Rodríguez-Castelán et al., 2016).

Statistical analysis

The measurements were made on from six to eight rats per group, depending on the parameter and technical difficulties with sample processing. All measured variables are expressed as the mean ± standard error (SEM). In accordance with the normality of the data, two-way ANOVA followed by Newman–Keuls tests were carried out to determine significant differences between groups. Factors considered for the former analysis were stress, the 30%-sucrose diet, and their interaction. For all cases, p ≤ .05 was accepted as statistically significant. All statistical analyses were carried out using the program GB STAT 6.0 (Dynamic Microsystems, San Diego, CA).

Results

Metabolic measurements

Rats subject to restraint stress and high-sucrose diet plus chronic restraint stress weighed less (∼22%) in comparison with those rats assigned to the control and high-sucrose diet groups (F_(1,1,31)= 21.35, p < .05). The body weight was unaffected by the high-sucrose diet or the interaction with chronic stress (Table 1). Rats from the stressed group ate less solid food (∼15%) compared to the rats from the control group. However, rats that drank the 30% sucrose solution ingested lesser solid food (∼60%) compared to those from the control and stressed groups (chronic stress, F_(1,1,31)= 25.57, p < .05; high-sucrose diet, F_(1,1,31)= 639.32, p < .05). Consequently, rats from the high-sucrose diet plus chronic restraint stress group reduced their intake of solid food compared to those from the other groups (interaction, F_(1,1,31)= 4.05, p < .05; Table 1). Water consumption was similar between the four groups. Rats under restraint stress showed a reduction (∼15%) in their total caloric intake in comparison with the control group (chronic stress, F_(1,1,31)= 10.78, p < .05). Consequently, total caloric intake was clearly higher in rats subject to the high-sucrose diet and the high-sucrose diet plus chronic restraint stress (∼21%) than in rats belonging to the control and stressed groups (high-sucrose diet, F_(1,1,31)= 64.79, p < .05). The interaction between caloric intake and chronic stress was not statistically significant. Glycaemia was similar between groups. No significant differences were also found in the liver weight and adiposity index (Table 1).

Table 1. Assessment of delta weight, food consumption, tap water or 30% sucrose, and caloric intake, blood glucose, liver weight, and adiposity index at the end of study.

	C		St		S30		S30 + St		F value
									St	S30	IN
Delta weight (g)	212.0	±11.2^a	169.9	±6.2^b	209.4	±9.2^a	156.6	±13.2^b	21.3	–	–
Food consumption (g/day)	25.8	±1.1^a	18.6	±0.7^b	10.9	±0.8^c	7.6	±0.5^d	43.9	271.5	6
Food consumption (g/day/100g body weight)	9.1	±0.2^a	7.7	±0.2^b	3.8	±0.2^c	3.2	±0.1^d	25.6	639.3	4.1
Equivalent in Kcal in food	30.9	±0.7^a	26.3	±0.7^b	13.0	±0.7^c	11.0	±0.5^d	25.5	636	4
Tap water or 30% sucrose consumption (mL/day)	54.3	±1.8^a	45.4	±2.9^a	62.6	±3.1^b	49.9	±3.1^a	–	5.3	–
Tap water or 30% sucrose consumption (mL/day/100g body weight)	19.3	±0.7	18.8	±0.9	22.0	±0.9	21.2	±1.1	–	–	–
Equivalent in Kcal in drinking water	0^a		0^a		26.4	±1.2^b	25.5	±1.3^b	–	911.6	–
Total Kcal/day/100 g body weight	30.9	±0.7^a	26.3	±0.7^b	39.4	±1.5^c	36.5	±1.5^c	10.8	64.8	–
Blood glucose (mg/dL)	105.6	±8.0	108.4	±5.5	110.4	±6.8	111.8	±12.4	–	–	–
Liver (g/100g body weight)	3.1	±0.1	3.1	±0.4	3.3	±0.1	3.3	±0.1	–	–	–
Adiposity index (g/100g body weight)	2.3	±0.1	2.1	±0.3	2.6	±0.4	2.6	±0.4	–	–	–

Date are represented as mean ± S.E.M. (n = 6–8 per group). C: control group receiving standard chow alone; St: control + chronic stress; S30: 30% sucrose diet; S30 + St: 30% sucrose diet + chronic stress. Post-hoc tests: p ≤ .05 for different letters. IN: interaction.

Hormone measurements

Serum corticosterone concentrations were similar between the control and stressed groups. Rats belonging to the high-sucrose diet and high-sucrose diet plus chronic restraint stress groups showed lower corticosterone concentrations (∼44%) than those included in the control and stressed groups (chronic stress, F_(1,1,31)= 0.0002, p > .05; high-sucrose diet, F_(1,1,31)= 15.81, p < .01; interaction, F_(1,1,31)= 0.15, p > .05; Figure 2(A)). Serum testosterone concentrations were unaffected by the chronic stress and high-sucrose diet. The interaction of high-sucrose diet plus chronic restraint stress increased circulating testosterone (∼153%) as compared to the control and stressed groups (F_(1,1,26)= 4.05, p > .05; F_(1,1,26)= 3.76, p > .05; F_(1,1,26)= 4.06, p < .05; Figure 2(B)).

Figure 2. Effect of chronic restraint stress and high-sucrose intake on serum concentrations of corticosterone (A) and testosterone (B). Data are expressed as means ± S.E.M. Data were analyzed by two-way ANOVA followed by Newman–Keuls tests. For the control (C) and stress (St) groups, rats were fed with a standard chow alone. Other rats were fed with the standard diet plus a high-sucrose diet without (S30 group) or with stress (S30 + St group). n = 6–8. Corticosterone: **p < .01 C vs. S30 and St vs. S30 + St groups; Testosterone: *p < .05 St vs. S30 + St groups.

Hepatic 11β-HSD1 activity, glycogen, and triacylglycerol concentration

Chronic stress increased hepatic 11β-HSD-1 activity: this was ∼155% higher in the stressed group than in the control group (chronic stress, F_(1,1,23)= 6, p < .05). However, this elevation was mitigated by drinking 30% sucrose solution, hence, 11β-HSD-1 activity was similar between the control, high-sucrose diet and high-sucrose diet plus chronic restraint stress groups (high-sucrose diet, (F_(1,1,23)= 0.67, p > .05; interaction, F_(1,1,23)= 2.66, p > .05; Figure 3(A)).

Figure 3. Effect of chronic restraint stress and high-sucrose intake on hepatic 11β-HSD1 activity (A), glycogen amount (B), and triacylglycerol amount (C). Data are expressed as means ± S.E.M. Data were analyzed by two-way ANOVA followed by Newman–Keuls tests. For the control (C) and stress (St) groups, rats were fed with a standard chow. Other rats were fed with the standard diet plus a high-sucrose diet without (S30 group) or with stress (S30 + St group). n = 6–8. 11β-HSD-1: *p < .05 C vs. St; Glycogen: #p < .05 C vs. St, and ##p < .01 St vs. S30 + St; Triacylglycerol: $p < .05 C vs. S30, and $$p < .01 St vs. S30 + St.

The amount of liver glycogen was affected by chronic stress, as stressed rats showed a significant reduction (∼70%) in liver glycogen in comparison with rats belonging to the control group. This effect was prevented by sucrose intake, as we did not observe differences in the liver glycogen between the control, high-sucrose diet and high-sucrose diet plus chronic restraint stress groups (chronic stress, F_(1,1,23)= 0.98, p < .05; high-sucrose diet, F_(1,1,23)= 8.27, p > .05; interaction, F_(1,1,23)= 5.76, p < .01; Figure 3(B)).

The amount of hepatic triacylglycerol was unaffected by chronic stress (F_(1,1,27)= 0.55, p > .05), but it was increased by the high-sucrose diet (∼84%; F_(1,1,27)= 19.08, p < .05). The amount of hepatic triacylglycerol was higher in rats subject to high-sucrose diet with chronic restraint stress than in rats exposed to chronic stress alone (interaction, F_(1,1,27)= 1.11, p > .05; Figure 3(C)).

Hepatic fibrosis

Absence of fibrosis as well as the presence of well-defined nuclei and cytoplasm were commonly observed in the control group (Figures 4(A,B)). In the stressed group, collagen was detected in the perisinusoidal and pericentral zone 3 of the hepatic parenchyma, and this was considered as a mild fibrosis grade 1 (Figures 4(C,D)). The high-sucrose diet group showed ballooned hepatocytes, fat microvesicles within hepatocytes, and more collagen in the perisinusoidal zone 3, which we considered as indicators of moderate grade 1 fibrosis (Figures 4(E,G)). The combination of chronic stress and high-sucrose diet induced a greater fat accumulation in the cytoplasm and ballooned hepatocytes; moreover, we also detected a reduction in collagen deposits (Figures 4(H,I)).

Figure 4. Effect of chronic restraint stress and high-sucrose intake on hepatic fibrosis. Masson’s trichrome stained liver sections from control (group C: panels A, B) that received a standard diet alone, chronic stress (St: C, D), 30% sucrose diet (S30: E–G), and 30% sucrose diet + chronic stress (S30 + St: H, I) groups. White arrows show the presence of collagen around the central vein and in the lumen of blood vessels. Black arrows define the presence of fat globules within hepatocytes. Scale bar 50 µm (A, C, E, H) and 10 µm (B, D, F, G, I). S: sinusoids; CV: central vein.

Hepatic pro-oxidant reactions

Chronic stress showed a dual effect in pro-oxidant reactions. It promoted a reduction in conjugated dienes, being significant for the high-sucrose diet plus chronic restraint stress group (∼33%), whereas the TBARS assay showed an increase (∼60%) in the liver of rats exposed and unexposed to the 30% sucrose treatment (chronic stress, conjugated dienes, F_(1,1,31)= 11.99, p < .01 and TBARS, F_(1,1,31)= 19.68, p < .05; Figures 5(A,B)). Conjugated diene amount was not affected by the high-sucrose diet or the interaction between high-sucrose diet plus chronic restraint stress (Figures 5(A,B)). A similar result was found for TBARs (Figures 5(A,B)).

Figure 5. Effect of chronic restraint stress and high-sucrose intake on oxidative damage in the liver. Data are expressed as means ± S.E.M. Data were analyzed by two-way ANOVA followed by Newman–Keuls tests. Folch extract optical density for conjugated dienes (O.D., A), 2-thiobarbituric acid (TBARS, B) for the control group receiving a standard diet alone (C); control + chronic stress (St); 30% sucrose diet (S30); and chronic stress +30% sucrose diet (S30 + St), n = 6–8. Conjugated dienes: *p < .05 C vs. St; TBARS: #p < .05 C vs. St, and ##p < .01 S30 vs. S30 + St.

Immune cells infiltration in the liver

The total number of immune cells in the lumen of the blood vessels was similar between groups (Figure 6(A)). However, the number of neutrophils (∼200%; Figure 6(B)) was higher in rats under chronic stress than in rats included in the control and high-sucrose diet plus chronic restraint stress groups (chronic stress, F_(1,1,23)= 7.17, p < .05; high-sucrose, F_(1,1,23)= 3.44, p > .05; and interaction, F_(1,1,23)= 2.94, p > .05; Figure 6(B)). Mast cells were more abundant in the lumen of blood vessels in the chronic stress groups (Figure 6(C)).

Figure 6. Effect of chronic restraint stress and high-sucrose intake on the hepatic immune cells infiltration (A), and neutrophils (B), and mast cells (C). Data are expressed as means ± S.E.M. Data were analyzed by two-way ANOVA followed by Newman–Keuls tests. The photomicrographs show the presence of mast cells around the central vein, stained with Alcian Blue. C, control group receiving standard chow alone; St, control + chronic stress; S30, 30% sucrose diet; S30 + St, 30% sucrose diet + chronic stress, n = 6–8. Scale bars =10 µm. Neutrophils: *p < .05 C vs. St, and #p < .05 St vs. S30 + St.

Discussion

The stressed group showed normal terminal serum concentrations of corticosterone and did not have hepatic steatosis. Nonetheless, stress promoted a high amount of glycogen, inflammation, mild fibrosis, oxidative stress, and a high activity of 11β-HSD-1 in the liver. The group treated with high-sucrose diet plus chronic restraint stress showed low concentrations of corticosterone, hepatic steatosis, oxidative stress, and high concentrations of testosterone. Moreover, a high-sucrose diet did not mitigate the effects promoted by stress in young adult rats, but even favored hepatic steatosis and a moderate fibrosis, while it reduced serum concentrations of corticosterone. Thus, corticosterone does not appear to be directly associated with the effects of stress or high-sucrose diet on liver damage. In addition, testosterone seems to be associated with the inhibition of inflammation in stressed rats with a high-sucrose diet.

An increase in the corticosterone concentration at the first day was observed for the exposure of stress, which return to normal values at days 15 and 30, indicating that the lack of high corticosterone concentrations in the stressed groups could be related to an adaptive process of adult rats to stress (Zardooz, Zahedi Asl, Gharib Naseri, & Hedayati, 2006). In this way, liver inflammation, oxidative stress, and mild fibrosis observed in the stressed group were independent of the corticosterone concentration. However, these signs could be associated with a great hepatic 11β-HSD-1 activity (Altuna, Lelli, San Martín de Viale, & Damasco, 2006; Chida et al., 2006; Corona-Pérez et al., 2015; Vasiljević et al., 2014). In agreement with our findings it has been reported that psychosocial stress aggravates inflammation and fibrosis in cirrhotic livers (Chida et al., 2006; Vere et al., 2009). In addition, hepatic 11β-HSD-1 activity has been related to signs of inflammation, and oxidative stress in patients with NAFLD (Staab & Maser, 2010; Tarantino & Finelli, 2013).

The augmentation of ACTH secretion by stress promotes an increase in dehydroepiandrosterone (DHEA) synthesis in the adrenal gland (Bertagna, 2017). However, high corticosterone concentrations seem to reduce the synthesis of testosterone in the testis (Dong et al., 2004). However, as the corticosterone concentration was normal in the stressed rats, the testosterone concentration was also unaffected (Hardy et al., 2005). Possibly, a stress induction model in which corticosterone is increased may help to test whether testosterone concentrations could also be affected. A limitation to interpretation in our study is that the stress protocol did not induce sustained increases in circulating corticosterone concentrations. Although the type and duration of stress used in the present study did not promote hepatic steatosis, it did give rise to inflammation signs and oxidative stress in the liver. Considering that both inflammation and oxidative stress are related to NAFLD (Harada et al., 2017), we suggest that stress could be associated with the development of NAFLD.

The high-sucrose diet reduced serum concentrations of corticosterone in the young adult rats in this study. A similar result was obtained in pubertal rats (Corona-Pérez et al., 2015). This finding could be related to the decrease in the corticotropin-releasing hormone (CRH) mRNA level in the paraventricular nucleus of the hypothalamus induced by sucrose (Ulrich-Lai et al., 2007), thus attenuating the HPA axis activity. In contrast to pubertal rats (Corona-Pérez et al., 2015), the high-sucrose diet in young adult rats favored hepatic steatosis and fibrosis. Although in this study rats consumed sucrose for two months, liver damage has been reported in adults exposed to sucrose diet for shorter periods of time (Fernandes-Lima et al., 2016). The effect of sucrose on the liver was independent of the hepatic 11β-HSD-1 activity. Although liver from rats exposed to a high-sucrose diet for two months may exhibit an important antioxidant defense, fructose may affect lipogenesis in the liver (Crescenzo et al., 2013; Oliveira, Santos, Barbosa-da-Silva, Mandarim-de-Lacerda, & Aguila, 2014). Serum testosterone concentration was unaffected by sucrose diet. However, further studies are necessary to analyze the possible effect of this diet on hepatic steroid metabolism and its participation in the impairment of liver function (Krawczyńska et al., 2017).

The influence of the high-sucrose diet on the reduction of serum concentrations of corticosterone and the hepatic 11β-HSD-1 activity induced by stress (Corona-Pérez et al., 2015; Mitra, Guèvremont, & Timofeeva, 2016; Green & McCormick, 2016), was confirmed in the high-sucrose diet plus chronic restraint stress group. However, in infant rats (Corona-Pérez et al., 2015), this diet induced hepatic steatosis in young adult rats. The combination of chronic restraint stress and high-sucrose diet increased serum testosterone concentrations. In contrast, a Western type diet modifies the hepatic steroid metabolism affecting the circulating concentration of these hormones (Krawczyńska et al., 2017), and patients with NAFLD have a low serum concentration of testosterone (Li, Liu, Wang, Chen, & Wang, 2015). In agreement, a high-fructose (Lê et al., 2006) or sucrose (Raben et al., 2011) consumption can augment lactate, which increases the testosterone concentration (Chen et al., 2017). Although testosterone has protective effects on oxidative stress (Das, Maiti, & Ghosh, 2005) and can ameliorate NAFLD (Kelly et al., 2014; Nikolaenko et al., 2014), we observed an inhibition of inflammation but not mitigation of the hepatic steatosis and oxidative stress. Glycogen has been associated with inflammation (Mulligan et al., 1998; Yamashita et al., 1982). Hence, the inhibition of inflammation could be related to the amount of glycogen in the liver, which was decreased by stress but increased by sucrose consumption with stress. The activation of androgen receptors, expected here as a result of increased testosterone production with stress combined with sucrose ingestion, increases the expression of insulin receptors, cholesterol storage, and glycogen synthesis in the liver of males, but decreases glucose uptake and lipogenesis (Shen & Shi, 2015).

Contrary to our hypothesis, a high-sucrose diet did not protect against hepatic injury promoted by stress in young adult rats. A combination of both stress and a high-sucrose diet could promote an accelerated development of NAFLD. In addition, it is clear that corticosterone is not the unique mechanism involved in the development of NAFLD promoted by stress. Thus, changes in the activity of hepatic 11β-HSD-1 protein seems be involved in the development of NAFLD (Staab & Maser, 2010; Tarantino & Finelli, 2013). Further studies are necessary to scrutinize the modulation of stress and carbohydrate-diet in hepatic steroid metabolism and its regulation of the hepatic steatosis.

Acknowledgements

The authors are grateful for the expert technical assistance of Laura García Rivera, and Sergio Ancona for the critical review of our manuscript.

Disclosure statement

The authors declare no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Additional information

Funding

This study was supported by the Consejo Nacional de Ciencia y Tecnología as a pre-doctoral fellowship (Reg. 417844) to Adriana Corona Pérez.