Nonalcoholic fatty liver disease (NAFLD) is highly correlated with obesity, and lifestyle changes to reduce weight remain the main therapeutic approach. The noncoding RNA miR‐22 has previously been reported to be highly abundant in the sera of NAFLD patients. In addition, miR‐22 directly targets peroxisome proliferative‐activated receptor, Pgc‐1α, peroxisome proliferator‐activated receptor α, and sirtuin 1 (Sirt1), which are important factors involved in fatty acid metabolism. Given that miR‐22 directly targets genes involved in the control of metabolism and obesity, we investigated whether miR‐22 contributes to metabolic alterations induced by obesity. We observed increased expression of miR‐22, decreased expression of Sirt1, and alterations in the expression of adipogenesis‐related genes in a mouse model of obesity and a human hepatocyte cell line. We identified that miR‐22 and the 3′‐UTR of Sirt1 are complementary. Mutation of the complementary fragment abolishes the ability of miR‐22 to regulate the Sirt1 gene. Furthermore, treatment of hepatic steatosis cells with miR‐22 mimics or inhibitors showed that miR‐22 can promote hepatic steatosis, and miR‐22 inhibitors effectively reduced triglyceride levels without affecting cell activity. Finally, we validated that miR‐22 has similar effects on downstream lipid metabolism‐related genes. Our data reveal the pathways and mechanisms through which miR‐22 regulates lipid metabolism and suggest that miR‐22 inhibitors may have potential as candidate drugs for NAFLD and obesity.

The current cytological evaluation technique of health food raw materials does not entirely meet the needs of evaluating health food. Our study adopted the microfluidic chip technique for the first time to construct a hepatocyte model of evaluating emodin, which was composed of a human hepatocellular carcinoma cell (HepG2) and microfluidic chip. The mixed glue of a model with rat tail collagen type I (1.3 mg/mL) + gelatin (7.5%) was used to simulate the microenvironment of a cell. The validity of this model was evaluated by cell proliferation activity and cell staining, and the toxicity of emodin was evaluated by a series of metabolic indicators on this model. The results indicated that the repeatability of the constructed hepatocyte model was favorable, with a coefficient of variation (CV) of 2.8%. After emodin continuously was exposed for 48 h, the cell inhibition was obvious at 100 and 200 μM, and the number of dead cells gradually increased with the increasing of emodin concentration, and the difference of BUN was significant between the emodin group and blank group (p < 0.05). The constructed model has a favorable applicability in evaluating emodin. This study provides an important platform and a potential in vitro alternative model for assessing and predicting the health effects of health food.

We aimed to explore the optimization and application of a bionic system of dynamic co-culture with hepatocytes and renal cells based on the microfluidic chip technique in evaluating emodin, which might replace the conventionally cytological evaluation technique of health food. After optimal experiments, the improved bionic system was composed of human hepatocellular carcinoma cells (HepG2), human renal glomerular endothelial cells (HRGECs), rat tail collagen type I, and gelatin with optimized concentrations (1.3 mg/mL + 7.5%). The applicability of the bionic system indicated that the growth stability was appropriate (CV: 7.36%), and the cell viability of that gradually decreased with the increasing of emodin concentration from 0–100 μM, which statistic significances were at 50 and 100 μM (p < 0.05), and the stained results of dead/live cells also showed the same trend. The LDH level appeared rising trend after decline between 0 μM and 100 μM emodi, and the level of that at 100 μM emodin was significantly higher than that at 25 μM and 50 μM emodin, respectively. The BUN level continuously and significantly declined with the increasing of emodin concentration (p < 0.05). Our research realized the application of this optimized bionic system in evaluating emodin, and provided a useful platform and reference for further in vitro alternative research with regard to evaluating the efficacies of health food in the future.

Cassiae semen is widely used as a raw material of health food. Anthraquinone compounds, the main components in cassiae semen, have been reported to show nephrotoxicity. Aurantio-obtusin (AO) is a major anthraquinone compound extracted from cassiae semen. This study investigates the effects of AO on the morphology and physiological function of human renal glomerular endothelial cells (HRGECs) on a microfluidic chip device for the first time. HRGECs were cultured on a microfluidic plate and exposed to a series of AO concentrations. Compared with traditional 96-well culture, HRGECs cultured on the microfluidic chip appeared to better mimic the glomerular microenvironment in vivo. AO induced different degrees of damage to cellular morphology and physiological function. The leakage of lactate dehydrogenase (LDH), as well as the secretion of interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), transforming growth factor-β1 (TGF-β₁), and monocyte chemoattractant protein 1 (MCP-1), increased in the AO treated groups. At the same time, cell viability and expression of ZO-1 in the AO treated groups decreased in a dose-dependent manner. The innovative device enables direct visualization and quantification to evaluate the cytotoxic effects of AO on HRGECs, and provides a useful visual in vitro model for studying health effect of health food.