Glucose absorption in the duodenum is modulated by an estrogen receptor α-dependent regulation of glucose transporter functional expression.

Estrogen is a steroid hormone produced in both men and women. In premenopausal women, most of the estrogen is synthesized by the ovaries, while in postmenopausal women and in men, adipose tissue is the predominant tissue source of estrogen. 1 2 Estrogen has several physiological functions and plays a crucial role in human health. Additionally, it is involved in functions related to sex differentiation and reproduction, and in glucose homeostasis, lipid homeostasis, brain function, bone metabolism, and control of inflammation. High or low levels of estrogen are associated with various diseases, including breast cancer, polycystic ovary syndrome, endometriosis, osteoporosis, ovarian cancer, gastric cancer, pituitary cancer, Alzheimer’s disease, schizophrenia, and obesity. 3 4 Clinical and laboratory studies have revealed the mechanism of estrogen action under physiological and pathological conditions with well-established roles in glucose metabolism. However, fewer studies have focused on estrogen-mediated glucose absorption, 5 6 and the specific mechanism remains unclear. Glucose absorption involves the active transport across the luminal brush border membrane of small intestinal epithelial cells via sodium/glucose cotransporter 1 (SGLT1) and the facilitative diffusion of glucose across the basolateral membrane of enterocytes via glucose transporter 2 (GLUT2). This process is driven by the transmembrane Na + gradient and membrane potential generated by the Na + -K + -ATPase, which facilitates effective absorption of luminal glucose at concentrations lower than those in the blood. 7 Our previous studies have shown that the estrogen receptor (ER) is expressed in mucosal epithelial cells along the digestive tract. 8 However, no direct evidence exists to show whether estrogen is involved in regulating glucose absorption via glucose transporters in small intestinal mucosal epithelial cells. Therefore, this study aims to explore the role of estrogen in regulating duodenal glucose absorption and the underlying molecular mechanisms.

Estrogen levels change periodically throughout different phases of a normal menstrual cycle in women. The first 7 days of the menstrual cycle (early follicular phase) are typically characterized by low serum E2 levels. With the rise of the dominant follicle, E2 levels rapidly increase during the second week of the menstrual cycle (late follicular phase), continue to increase gradually through the mid-follicular phase, and then spike dramatically just before ovulation. 14 In our previous study, we reported that estrogen levels are lower during the premenstrual phase than during the preovulatory phase. 15 Here, we tested the blood glucose levels of young women volunteers with regular menstrual cycles using an OGTT. As shown in figure 1 , glucose tolerance was markedly less in the premenstrual phase than in the preovulatory phase (p<0.05), indicating that women have lower estrogen and less glucose tolerance in the premenstrual phase and vice versa in the preovulatory phase. Next, we established an OVX mouse model ( figure 2A ). To verify the success of the model, we used a radioimmunoassay to measure serum E2 levels. The results showed that serum E2 levels were significantly lower in OVX mice than in control mice ( figure 2B ). In addition, IHC and western blot analyses were performed to detect ERα and ERβ expression in the duodenal tissues of the two groups. ERα ( figure 2C,E ) and ERβ ( figure 2D,F ) expression was significantly reduced in the duodenal tissue of OVX mice than in that of the control mice. Furthermore, after estrogen deficiency, ERα and ERβ expression was also significantly downregulated in the duodenal tissue. These results indicated that the model was successfully established. We continuously observed and recorded the body weights of the mice postoperatively. The weights of mice in the two groups were not significantly different 1 week postoperatively but showed an upward trend 2 weeks after the operation. From day 8, the body weight of the OVX mice was significantly greater than that of the control mice ( figure 2G ). Two weeks postoperatively, randomly selected mice from each group were dissected to observe the fat in the abdominal cavity. The abdominal fat content of OVX mice was significantly increased ( figure 2I ). To explore whether OVX mice with lower estrogen levels could result in impaired glucose tolerance, we measured the blood glucose levels of mice from both groups using the OGTT. These results were consistent with our hypothesis that the glucose tolerance of OVX mice was significantly worse than that of control mice ( figure 2H ). The above results confirmed that with decreased estrogen levels ( figure 2B ) in OVX mice, ERα and ERβ expression in the intestinal mucosa was reduced ( figure 2C,F ), and the glucose tolerance was impaired ( figure 2H ). Next, we sought to determine how the intestinal absorption of glucose was altered in OVX mice with low estrogen levels or estrogen deficiency. The Ussing chamber technology was used to detect changes in the I sc of the duodenum. After glucose stimulation, the changes in I sc in OVX mice were significantly lower than those in the control group ( figure 3A ), suggesting that estrogen deficiency results in decreased glucose absorption. To identify the possible mechanisms underlying the decreased glucose absorption, we examined the effects of estrogen on duodenal glucose transporters in mice. First, SGLT1 and GLUT2 expressions in the murine duodenum were studied using IHC. The expressions of SGLT1 ( figure 3B ) and GLUT2 ( figure 3C ) were significantly lower in the duodenum of OVX mice compared with control mice. Furthermore, the effect of estrogen on duodenal glucose transport proteins was confirmed using in vitro cell experiments. We examined the expression of SGLT1 and GLUT2 in SCBN cells after treatment with 17β-E2 for 24 and 48 hours using western blotting. Compared with the control, E2 treatment significantly increased the expression of SGLT1 and GLUT2 in SCBN cells ( figure 4A,B ), indicating that estrogen promotes the expression of glucose transporters. To further screen for functional receptors binding to estrogen, ERα or ERβ receptor genes were silenced by lentivirus-based shRNA transfection in SCBN cells ( figure 4C,D ). Silencing ERα expression significantly reversed the estrogen-mediated upregulation of SGLT1 and GLUT2 protein expression, while silencing ERβ expression had minimal effect, and the difference was not statistically significant ( figure 4E ). These results indicated that ERα shRNA, but not ERβ shRNA, inhibited estrogen-induced expression of SGLT1 and GLUT2. Estrogen activates the PKC signaling pathway, with the Gα q -coupled Phospholipase C-PKC-Protein Kinase A pathway playing a critical role in energy homeostasis. 16 To determine whether PKC is involved in the estrogen-induced regulation of intestinal glucose absorption, we explored its role further. As shown in figure 5A , E2 treatment significantly reduced p-PKC expression in SCBN cells, with the effect becoming more pronounced over time, indicating that estrogen inhibits PKC. To examine PKC’s influence on duodenal glucose transporters, we used the PKC agonist PMA and the inhibitor Gö6976. PMA stimulation led to a significant decrease in SGLT1 and GLUT2 expression ( figure 5B,C ), whereas Gö6976 treatment resulted in a marked increase ( figure 5D,E ), suggesting that PKC negatively regulates these transporters. Additionally, we assessed the combined effects of estrogen with PMA or Gö6976. Compared with estrogen alone, cotreatment with PMA significantly reversed the stimulatory effect of estrogen on SGLT1 and GLUT2 expression, while cotreatment with Gö6976 enhanced this effect synergistically ( figure 5F,G ). Finally, we examined the effects of ERα and ERβ on PKC phosphorylation. In SCBN cells, estrogen failed to inhibit p-PKC expression when ERα was silenced but had no effect in ERβ-silenced cells. Moreover, PMA enhanced PKC phosphorylation in ERα-silenced SCBN cells, while the addition of Gö6976 significantly reduced PMA-induced PKC phosphorylation ( figure 5H ). These findings suggest that estrogen inversely regulates p-PKC expression, with ERα playing a critical role.

In this study, we investigated the effects of estrogen on glucose homeostasis in women aged 20–30 years, female mice, and SCBN cell lines. The lack of estrogen was associated with weight gain and increased abdominal fat. These results agree with previous findings that estrogen deficiency in postmenopausal women and OVX animals is associated with obesity. 17 Furthermore, the increase in abdominal fat was consistent with the role of estrogen in regulating fat distribution. 18 Although body weight was increased in OVX mice, glucose absorption in the duodenum was reduced. This finding seems paradoxical, given the common understanding that obesity is associated with excessive sugar intake. This apparent discrepancy may be explained by insulin resistance, which is the primary factor explaining why estrogen deficiency leads to weight gain in OVX mice. Preclinical and clinical studies have demonstrated that low estrogen levels or its absence can induce insulin resistance, while estrogen hormone replacement therapy can restore the insulin response to glucose. 19 20 Based on various studies, we identified two potential mechanisms of insulin resistance in OVX mice: primary insulin resistance mediated by estrogen deficiency and secondary insulin resistance resulting from fat deposition following estrogen reduction. A previous study reported that the nerve growth factor p75 NTR , a regulator of glucose uptake and insulin resistance, differentially regulates Rab5 and Rab31, resulting in decreased GLUT4 plasma membrane translocation, which leads to a decrease in insulin-stimulated glucose uptake. Furthermore, p75 NTR knockout mice show higher insulin sensitivity to a normal diet. 21 We further observed that p75 NTR , as measured by western blotting, was significantly upregulated in OVX mice ( online supplemental figure 1A ), indicating that ovariectomy caused insulin resistance, which is consistent with previous studies. 19 We detected the expression of CD36 in the duodenal mucosal tissue of the animal models using western blotting. CD36, a membrane glycoprotein on the cell surface, is highly regulated by estrogen and plays a crucial role in regulating metabolic phenotypes, particularly glucose and fatty acid metabolism. 22 Several studies have shown that CD36 is involved in the development of insulin resistance in adipose tissue, liver, skeletal muscle, and heart. Deficiency in CD36 function or expression can lead to dyslipidemia or insulin resistance. 23 24 We observed a significant decrease in CD36 expression in OVX mice compared with that in the control group ( online supplemental figure 1B ), indicating that estrogen deficiency in OVX mice may lead to dyslipidemia and insulin resistance. Recent studies on the effect of estrogen on CD36 expression in the liver and skeletal muscle of OVX mice support our conclusions. 25 26 Further, we found that OVX mice showed a significant increase in insulin and HOMA-IR, while showing no significant change in FBG levels as compared with the control group ( online supplemental table 1 ). Overall, our results show that insulin resistance plays an important role in glucose metabolism in OVX mice. In this study, we observed that estrogen regulates glucose absorption in the duodenum by inhibiting PKC, with ERα playing a crucial role in this process ( figure 6 ). ERα and ERβ belong to the nuclear receptor family of ligand-activated transcription factors, which regulate gene transcription through estrogen response elements and are widely expressed in reproductive and non-reproductive tissues. ERα and ERβ mRNAs and proteins are expressed in the gastrointestinal tract, providing a basis for the role of estrogen in regulating gastrointestinal function and pathophysiology. 12 27 Although ERα and ERβ are highly homologous, data suggest that they may have different functions. For example, in colon cancer, ERα upregulation is a risk factor, whereas ERβ upregulation predicts better survival outcomes. 27 28 In skeletal muscles, activation of ERα increases the expression of GLUT4 and stimulates muscle GLUT4 functionality, whereas activation of ERβ enlarges muscle fibers, which may lead to enhanced glucose utilization. 29 Early studies on the feeding behavior of ERα knockout(αERKO) mice showed that, in the absence of ERα signaling, increased body fat is mediated by decreased energy expenditure rather than increased energy intake. Chronic E2 treatment significantly increases food intake in OVX wild-type mice but not in OVX αERKO mice. 30 Our study indicates that ERα may be the key receptor involved in the estrogen regulation of intestinal glucose transporters, providing new insights into the regulation of intestinal energy intake and glucose metabolism by estrogen. Furthermore, impaired glucose tolerance and increased insulin resistance in ERα-deficient mice 31 32 support our findings that ERα plays a crucial role in glucose homeostasis and metabolism. Our findings indicate that inhibiting PKC downstream of estrogen contributes to the upregulation of SGLT1 and GLUT2. PKC may suppress transcription by inhibiting key transcription factors, thereby reducing gene expression. Additionally, PKC activation may promote the internalization and degradation of these transporters, limiting their presence on the cell surface. Furthermore, PKC can activate inflammatory pathways such as nuclear factor kappa B, 33 which may further downregulate glucose transporter expression. These mechanisms may individually or collectively contribute to PKC-mediated regulation of SGLT1 and GLUT2. Further studies are warranted to elucidate the underlying molecular mechanisms.

This study is the first to demonstrate that estrogen regulates duodenal glucose absorption through the influence of ERα on glucose transporters and its inhibition of PKC, as evidenced by research on women aged 20–30, female mice, and SCBN intestinal epithelial cells. These findings offer new insights into the role of estrogen in intestinal energy intake and glucose metabolism. However, further investigation is needed to explore the effects of estrogen on intestinal glucose absorption in males and non-mammalian species.

Healthy Chinese women aged 20–30 years with regular menstrual cycles (menstrual period within 3–7 days, menstrual cycle in the range of 25–32 days) and without lesions or medical conditions causing irregular menstruation were included. Research on healthy human volunteers was conducted in accordance with the principles of the Declaration of Helsinki. Oral glucose tolerance test (OGTT) was performed during the menstrual and ovulation periods, when the estrogen level in the body was the lowest and highest, respectively. Before the measurements, all participants were required to fast for 12 hours. After fasting and intake of 75 g oral glucose, peripheral blood samples were collected at 5, 15, 30, 45, 60, 75 and 90 min, and blood glucose levels were measured using a blood glucose meter (Accu-Chek, Roche Diagnostics, Germany). For each participant, the area under the curve (AUC) was calculated to quantify the response to an oral glucose load. C57BL/6 female mice, 6 weeks old (~17±2.2 g), were purchased from Beijing Huafukang Biotechnology, China, and housed in the experimental animal facility at Zunyi Medical University under standard care conditions. Sexually mature female mice (n=60) were randomly assigned to two groups and subjected to either ovariectomy (ovariectomized (OVX)) or a sham operation (control), following anesthesia induced by intraperitoneal injection of 50 mg/kg ketamine and 10 mg/kg xylazine. The backs and sides of the mice were shaved and cleaned with 70% ethanol and betadine. Ovaries were removed through a dorsal incision made between the dorsal hump and the base of the tail. A ligature was placed before excising the ovary. The muscle and the skin incisions were closed with sutures. The procedures were repeated for the second ovary. 9 Daily weight was recorded for all mice from the second postoperative day until 2 weeks after the operation. All animal experiments were approved by the Committee on Investigations Involving Animals at Zunyi Medical University, China (KLLY(A)-2019-105) in strict accordance with the Guidelines of the Committee on the Care and Use of Laboratory Animals and the Guidelines of Animal Research: Reporting of In Vivo Experiments. Two weeks after the operation, 10 mice from each of the OVX and the control groups were randomly selected for vena cava blood collection, and serum estradiol (E2) levels were detected using a rabbit anti-E2 antiserum radioimmunoassay kit (Beijing Northern Institute of Biotechnology; detection range 20–4000 pg/mL; sensitivity 15 pg/mL), according to the manufacturer instructions. Two weeks after surgery, 10 mice were randomly selected from each of the two groups (distinct from those used in the radioimmunoassay). Following a 12-hour fast, glucose (3 g/kg) was administered via gavage. Blood was collected from the tail vein at 0, 5, 15, 30, 45, 60, 75, and 90 min, and blood glucose levels and AUCs were determined as described for human participants. Following the experiment, mice were euthanized by cervical dislocation, and the duodenum was harvested for subsequent Ussing chamber experiments. Two weeks postoperation, six mice from both the OVX and control groups were randomly selected to assess insulin resistance. Fasting blood glucose (FBG) levels were determined in overnight-fasted animals using an automatic glucose meter (Accu-Chek, Roche Diagnostics) with blood samples collected from the tail tip. Overnight-fasted animals were anesthetized with ketamine/xylazine (100/10 mg/kg intraperitoneally), and blood samples were obtained from the cardiac ventricle. Serum was separated by centrifugation (800×g, 4°C, 20 min), and plasma insulin concentrations were determined using an ELISA kit (Abcam, Shanghai, China; detection range 6.25–400 µlU/mL, sensitivity 5 µlU/mL). The homeostatic model assessment of insulin resistance (HOMA-IR) index was calculated as: HOMA-IR=(fasting glucose [mmol/L]×fasting insulin [µlU/mL])/22.5. Glucose absorption was measured ex vivo using Ussing chamber experiments in mouse duodenum, as previously described. 10 Approximately 1.5 cm of proximal duodenum was placed in ice-cold iso-osmolar mannitol (10 mmol/L) solution and indomethacin (1 μM) solution (to suppress trauma-induced prostaglandin release) at 4°C. The intestinal segment was placed on a flat paraffin block, with a sealing membrane on its surface. The duodenum was opened along the mesenteric border, and the external serosal and muscle layers were removed by sharp dissection in the ice-cold iso-osmolar mannitol and indomethacin solution. The duodenal mucosae were mounted between two chambers (effective penetration area: 0.16 cm 2 ) and placed in an Ussing chamber. Parafilm O-ring was used to minimize edge damage to the tissue, securing between the chamber halves. The mucosa (top) was added to 10 mL of working fluid, with continuous and uniform oxygen (100%) perfusion. The serosal side (basal side) was perfused with 10 mL of working solution, and a mixture of 95% oxygen and 5% carbon dioxide was infused continuously and uniformly. Each bath contained 10.0 mL of the respective solution maintained at 37°C by a heated water jacket. Experiments were performed under continuous short-circuit conditions to maintain the electrical potential difference at zero, except during brief intervals (<2 s) when the open-circuit potential difference was measured. Luminal pH was maintained at 7.40 by the continuous infusion of 0.5 mM HCl under the automatic control of a pH stat system (PHM290, pH-Stat controller; Radiometer Copenhagen). After the assembly, transepithelial short-circuit current ( I sc ) was recorded every 5 min via an automatic voltage clamp (voltage-current clamp EVC-4000; World Precision Instruments). After stabilization for 15 min, 20 mM glucose solution was added to the mucosal side, and I sc was continuously observed and recorded for 30 min. The changes in I sc were compared. The difference in I sc was used to calculate the tissue absorption index (difference between the highest and lowest values). The mucosal solution used in Ussing chamber experiments contained the following (in mM): 140 Na + , 5.4 K + , 1.2 Ca 2+ , 1.2 Mg 2+ , 120 Cl − , 25 gluconate, and 10 mannitol. The serosal solution contained the following (in mM): 140 Na + , 5.4 K + , 1.2 Ca 2+ , 1.2 Mg 2+ , 120 Cl − , 25 HCO 3 − , 2.4 HPO 4 2− , 2.4 H 2 PO 4 − , and 10 glucose. Duodenal tissues from the two groups of mice were fixed in formalin and embedded in paraffin. Immunohistochemistry (IHC) was performed as previously described to stain ERα, ERβ, SGLT1, and GLUT2 in paraffin-embedded tissue blocks 11 using 1:50 dilutions of anti-ERα (Abcam, ab32063) and anti-ERβ (Abcam, ab288) and 1:100 dilutions of anti-SGLT1 (Abcam, ab321787) and anti-GLUT2 (Abcam, ab54460). All antibodies were purchased from Abcam. Data were collected from an average of four randomly selected areas in a random section of each tissue sample. The results were analyzed using Image Pro-Plus (Media Cybernetics, USA) and expressed as mean optical density (measured in arbitrary units) for ERα, ERβ, SGLT1, and GLUT2, representing the mean intensity of staining in the considered area. SCBN cells, a non-transformed duodenal epithelial cell line obtained from a human patient, were provided by Dr Hui Dong of the University of California, San Diego. The cells were frozen in liquid nitrogen. SCBN cells grow as a polarized confluent monolayer and express ERα and ERβ. 12 Cells were grown as previously described 2 13 in Dulbecco's Modified Eagle Medium(DMEM) supplemented with 10% fetal calf serum in an incubator with a 5% CO 2 at 37°C. To silence ERα or ERβ, lentivirus-based short hairpin RNA (shRNA) was used. SMART vector lentiviral human ERα shRNA (CTCTACTTCATCGCATTCCTT), human ERβ shRNA (CGGCAGACCACAAGCCCAAAT), and SMART vector non-targeting control particles (TTCTCCGAACGTGTCACGT) were purchased from Genechem (Shanghai, China). SCBN cells were transfected according to the manufacturer’s protocol, and western blotting was used to detect protein expression to demonstrate successful silencing. Murine duodenal mucosal tissues were collected at the time previously described. Normal or ERα or ERβ-silenced SCBN cells were treated with 17β-estradiol (10 nM), protein kinase C (PKC), agonist phorbol 12-myristate 13-acetate(PMA) (200 µM), or PKC inhibitor Gö6976 (200 µM). Duodenal mucosal tissue, pancreatic tissues, or SCBN cells were homogenized in lysis buffer at 4°C. Western blotting for protein expression was performed as previously described. 11 Anti-ERα (1:100, Abcam, ab32063), anti-ERβ (1:100, Abcam, ab288), anti-SGLT1 (1:500, Abcam, ab14686), anti-GLUT2 (1:1000, Abcam, ab54460), anti-PKC (1:500, Abcam, ab181558), anti-phosphorylated-PKC (anti-p-PKC) (1:500, Abcam, ab109539), anti-p75 neurotrophin receptor (anti-p75 NTR ) (1:1000, Abcam, ab52987), anti-cluster of differentiation 36 (anti-CD36) (1:1000, Abcam, ab252922), and anti-glyceraldehyde-3-phosphate dehydrogenase (anti-GAPDH) (1:5000, Abcam, ab8245) were used as primary antibodies and were purchased from Abcam. The results are expressed as a ratio relative to GAPDH. Experimental data were collected and organized per the requirements of a completely randomized controlled design. All data are expressed as the mean±SD. Student’s t-test was generally used to determine the significance of differences between two groups. For multiple comparisons, one-way analysis of variance was used. P<0.05 indicated statistically significant results.