Female Holstman rats (n = 40) from the animal facility of the University of São Paulo State (UNESP, Araraquara, Brazil) weighting ~ 220 g upon arrival were housed in collective cages. The animals had ad libitum access to food and tap water. All animals were fed with commercial rodent chow. Their environment was controlled in a reverse light cycle (12 h dark starting at 08:00 AM). Food intake was monitored daily over the entire experimental period. Body mass was measured 3 times/week at the same time of day. All experiments described in the present report were conducted according to the Guide for Care and Use of Laboratory Animals¹⁹ and approved by the Ethics Committee on Animal Use from the Federal University of São Carlos (CEUA-UFSCar) Protocol n° 005/2013.

Rats were randomly distributed into five experimental groups (n = 8/group). Rats first underwent a bilateral ovariectomy (n = 24) or a bilateral sham operation (Sham, n = 16). Thereafter, one group of Ovx and one group of Sham rats remained sedentary while one group of Ovx and one group of Sham rats were submitted to a 12-week resistance training program. The third group of Ovx rats remained sedentary and were given 17β-estradiol supplementation. Altogether, five groups were compared: Sham-sedentary (Sed), Sham-resistance trained (RT), Ovx-Sed, Ovx-RT, and Ovx-Sed with hormone replacement (Ovx-E2).

Ovariectomy and sham surgeries were performed when the rats reached 250 g of body mass, according to the technique described by Kalu²⁰. To perform this procedure, rats were first anesthetized with a mixture of ketamine-zylazine (61.5 - 7.6 mg/kg, ip). A bilateral incision (1.0 - 1.5 cm) was made through the skin and muscle layers. The ovaries were removed and the skin and muscles were sutured. The sham surgery was performed using the same procedure but the ovaries were not removed. The antibiotic tramadol hydrochloride (20 mg/kg, sc) was injected every 24 h for five days after surgery. The rats were allowed to recover from surgery for seven days, which also permitted us to ensure the systemic effects of Ovx. At the end of this period, rats in the Ovx-E2 group underwent a second surgical procedure to insert silastic capsules under the skin of the neck to provide hormonal replacement.

Silastic capsules (15 mm) with internal and external diameter of 1.02 mm and 2.16 mm, respectively (Dow Corning VWR International, Buffalo Grove, IL, USA) were used for E2 replacement. Sunflower oil was used as the vehicle, with a 5% concentration of 17β-estradiol (50 mg/ml oil). Ten μL of this solution was pipetted into each capsule and then both sides were sealed with silastic glue. The capsules were stored for 24 h to allow the glue to dry. After drying, the capsules were kept in saline (0.9%) solution²¹.

The training sessions consisted of climbing a vertical wooden and metal ladder with the following dimensions: 80° of inclination, 110 cm height x 18 cm width, and 2 cm spacing between the ladder rungs. At the top of the ladder, the animals reached a 20 x 20 x 20 cm cage, which allowed them to rest between climbing sessions. The first three training sessions were used to allow the animal to adapt to the process of climbing and to determine the maximum possible weight load. The load consisted of fishing sinkers that were previously weighed, labeled, and stored in falcon tubes. The falcon tubes were tied to proximal portion of the rats’ tails. The present RT protocol is an adaptation of the program described by Hornberg & Farrar²².

The animals were first familiarized with the training procedures over two consecutive days. The overload apparatus was attached to the rat’s tail without any weight. The rats were placed at the bottom of the ladder and familiarized with the climb. Once they reached the cage at the top of the ladder, the rats were allowed a 2 min rest period. If necessary, a physical stimulus with fingers or tweezers was applied to the animal’s tail to initiate climbing movements. The rats were considered to have adapted to the ladder protocol when voluntary climbing was performed three consecutive times without any stimulus.

The maximal load determination was initiated with a load equivalent to 75% of the rat’s body mass. After each successful climb, 30 g were added to the apparatus until the rat was unable to climb the entire length of the ladder. The highest carried load was considered the maximum carrying capacity of each rat.

The RT program consisted of training sessions held every 3 days for 12 weeks, for a total of 27 sessions. The training sessions consisted of 4 to 9 climbs with progressively increasing weights. The size of the ladder required the animals to perform 8 - 12 movements per climb. The first four climbs were carried out with loads corresponding to 50, 75, 90, and 100% of the rat’s maximal carrying capacity as determined in the last training session. For the 5 subsequent climbs, an additional 30 g load were progressively added until the rat was unable to climb the entire ladder. The new maximal load determined during the training program was used for the next session.

Rats were euthanized between 9:00 AM and 12:00 PM after removal of food at 7:00 AM. Rats in the trained groups were euthanized 48 h following the last training session. Immediately after decapitation, blood samples were collected and centrifuged at 3000 rpm for 10 minutes at 4°C, and then stored at -20°C. The liver was weighed and then washed with saline (0.9%). The median lobe was used for all molecular analyses.

Measurement of 17β-estradiol plasma levels was performed by the ELISA technique²³ using commercial kits (ADI-900-174, Enzo Life Sciences, Farmingdale, NY, USA) according to the manufacturer’s instructions. Liver triglyceride concentrations were determined from glycerol released after KOH hydrolysis according to the technique described by Frayn and Maycock²⁴. Liver glycogen content was determined using phenol-sulfuric acid reactions as described by Lo et al.²⁵.

Total RNA was extracted from frozen livers with the use of the RNA extraction Mini kit (Invitrogen) according to the manufacturer’s protocol. Then, the RNA was treated with DNase (Invitrogen) in order to avoid genomic contamination. Total RNA (2 pg) was reverse-transcribed into complementary DNA (cDNA) using high-capacity cDNA reverse transcription kits (Applied Biosystems). Reverse-transcribed samples were stored at -20°C. GAPDH gene expression was determined using a pre-validated Taqman Gene Expression Assay (Applied Biosystems, Rn01462d61, Foster City, CA). The gene expression levels of target genes were determined using assays designed with the Universal Probe Library from Roche. The primer sets and UPL probe numbers are presented in Table 1. To validate the efficiency of the qPCR assays, we used a mix of the samples tested in the study. An ABI PRISM® 7900HT (Applied Biosystems) was used to detect amplification level and was programmed with an initial step of 3 min at 95°C, followed by 40 cycles for 5 s at 95°C and 30 s at 60°C. All reactions were run in triplicate and the average values of threshold cycle (C_T) were used for quantification. GAPDH was used as endogenous control. The relative quantification of target genes was determined using the ΔΔC_T method²⁶. Briefly, the C_T values of target genes were normalized to an endogenous control gene (GAPDH) (ΔC_T = C_{T Target} - C_{T GAPDH}) and compared with a calibrator: (ΔΔC_T = ΔC_{T Sample} - ΔC_{T Calibrator}). Relative expression (RQ) was calculated using the Sequence Detection System (SDS) 2.2.2 software (Applied Biosystems) with the formula RQ = 2^-ΔΔC_T.

Protein analyses were performed by the Western blotting method, described by Kurien and Scofield²⁷. Aliquots of hepatic lysate (~ 25 mg) were separated by SDS- PAGE gel and transferred to a nitrocellulose membrane. Before the procedure, the membrane was blocked with 5% albumin solution. After this, each membrane was incubated overnight at 4°C with appropriate dilutions of the primary antibodies, including GLUT2 (Sigma-aldrich) and β-actin (Cell Signaling Technology). Each membrane was washed in TBST (3 x 5 min) and then incubated with the appropriate secondary antibody conjugated to HRP for 120 min at room temperature. Antibody binding was detected by enhanced SuperSignal® West Pico Chemiluminescent Substrate (PIERCE, IL, USA), as described by the manufacturer. Blots were scanned and the densitometry of protein bands was determined by pixel intensity using Molecular Imager® ChemiDoc™ XRS+ (Bio-Rad Laboratories Inc., CA, USA) with Image Lab™ 2.0 Software.

Values are expressed as means ± standard error (SE). The data were first analyzed by the Kolmogorov-Smirnov normality test (SPSS 22.1 software) and all results were considered parametric. The one-way analysis of variance (ANOVA) test was used to quantify the RT or E2 variables. The Tukey post hoc test was used in the event of a significant (P < 0.05) F ratio.

The results of the present study first indicated that the well-documented fat accumulation in livers of Ovx rats is accompanied by an increase in the expression of GLUT2, the major glucose transporter in the plasma membranes of hepatocytes, along with an increase in gene expression of the transcription factor PPARγ, which is known to modulate GLUT2 gene expression¹¹. These results suggest that alterations in liver fat metabolism, known to ocur in Ovx rats, such as increased de novo lipogenesis, also affect hepatic glucose uptake. In addition, the present data indicate that these disturbances in liver fat and glucose transport may be overcome by a regular resistance training program. These results reinforce the concept that exercise training may alleviate some of the metabolic consequences of low estrogenic status.

Liver fat content was increased by about 35% in Ovx rats in this study. Our hypothesis that Ovx in rats may lead to an increase in gene expression of the GLUT2 transporter was based on previous findings that fat accumulation in the livers of Ovx animals is associated with an increase in de novo lipogenesis^7,8. Increased lipogenesis logically requires more glucose molecules to be available as substrates, thus, hypothetically, leading to increased glucose uptake through GLUT2 transport. Studies in GLUT2 knockout mice confirm that, indeed, glucose uptake by hepatocytes is a major substrate for lipogenesis²⁸. Although the basic function of GLUT2 is to catalyze the passive transport of glucose across the plasma membrane, it is required for glucose uptake, whereas it is dispensable for glucose output²⁹. Also supporting the link between liver fat accumulation and GLUT2 transport is the finding that glucose-induced SREBP- 1c upregulates GLUT2 mRNA in primary hepatocytes of mouse liver^30,31. SREBP-1c is a well-documented regulator of lipogenesis and has been repeatedly reported to be upregulated in Ovx animals^8,32. Altogether, the present data are in line with the concept that an increase in fat content in livers of Ovx rats is linked to increased glucose uptake through an increased expression of GLUT2 protein.

The increase in gene expression of the PPARγ transcription factor supports the increase in GLUT2 synthesis in the livers of Ovx rats. PPARγ expression is low in liver tissue¹⁰. Nevertheless, it has been reported that PPARγ is able to modulate gene expression of GLUT2 in liver¹¹. Ankaflavin, a natural PPARγ agonist, has been shown to enhance insulin sensitivity by increasing hepatic GLUT2 expression and glucose uptake in rats³³. There is little doubt that the increase in fat content in the liver cells of Ovx animals is related to a reduction in plasma estradiol levels^18,34. The present data, therefore, support the existence of a link between liver fat accumulation in Ovx rats at least partially through increased de novo lipogenesis and GLUT2 transport and an increase in PPARγ gene expression.

From a mechanical point of view, it is possible that the present increase in GLUT2 gene expression in Ovx animals is linked to a transient decrease in intracellular glucose level secondary to increased lipogenesis. Glucose has been reported to induce a dose-dependent increase in GLUT2 mRNA levels in cultured hepatocytes, while insulin had no effects³⁵. To go one step further into hepatic glucose metabolism in Ovx animals, we measured glycogen levels and PEPCK mRNA levels. Both of these parameters remain unchanged following estrogen withdrawal. The absence of an effect on glycogen levels in Ovx rats indicate that if indeed glucose uptake is increased in the present Ovx rats, it was not used for glycogen synthesis. Actually, an increase in liver fat, as seen in the present Ovx rats, is associated with a reduction in glycogen storage³⁶. There is also no indication that gluconeogenesis is increased for estrogen withdrawal, indicating no shortage of glucose.

E2 replacement in Ovx rats resulted, as would be expected, in a reduction in liver fat levels but also in a large increase in liver glycogen levels and a reduction in gene expression of PEPCK. The 5% hormone concentration used in the present study appears to be an effective replacement over a three-week period²¹. However, the present replacement over a 12-weeks period has led to supraphysiological plasma levels of estrogen, as judged from the high estrogen plasma concentrations and the increase in liver weight. Nevertheless, the present E2 replacement data are in line with a previous report indicating that treatment with estradiol in Ovx mice increases liver glycogen and reduces gluconeogenesis³⁷.

Training has been reported to be effective in reducing liver fat accumulation in Ovx animals^17,18,38. The present study extended these findings by showing that GLUT2 protein and PPARγ transcript levels were restored almost to normal levels in the livers of Ovx rats submitted to resistance training. This at first glance supports the link between liver fat accumulation through increased de novo lipogenesis and GLUT2 transporters. The present RT program is an interesting alternative to classical endurance training. Similar RT programs have been used successfully in previous studies, leading to similar reductions in liver fat accumulation^17,38,39. It is intriguing that RT and E2 replacement in the present study led to similar metabolic improvements in the livers of Ovx animals. The same phenomenon has been previously observed on the reduction of subclinical inflammation in Ovx rats¹⁶. The present data are, therefore, in line with the concept that exercise training has estrogenic-like effects.