A vascular endothelial cell, neuron, and microglia tri-culture model to study hypertension-related depression

Hypertension-related Depression (HD) is a complex mental disorder that exerts a significant negative impact on patients' quality of life. Previous studies have demonstrated that damages to vascular endothelial and hippocampus are the primary pathological features in HD rats. Under hypertensive conditions, inflammatory cytokines in peripheral blood vessels can induce central nervous system inflammation through penetration of a damaged blood-brain barrier, peripheral immune cells, and neural pathways, damaging the brain and triggering HD. Therefore, interactions between vascular endothelial cells, neurons, and glial cells are critical for the understanding of HD. However, in vivo animal models are often limited by the complexity of intrinsic systems, high inter-individual variability, and stringent ethical regulations. A reliable model that could be easily manipulated is needed for investigating the mechanisms involved in communication between vascular endothelial cells, neurons, and glial cells in HD. We therefore aimed to create a composite tri-culture model consisting of rat aortic endothelial cells (RAECs), neurons, and microglia to study HD. First, RAECs were stimulated with lipopolysaccharide to mimic endothelial injury under hypertensive conditions. Vascular endothelial function and inflammatory levels were assessed using fluorescent probes and enzyme-linked immunosorbent assays. RAECs treated with 1 μg/ml LPS for 24 h had reduced levels of nitric oxide, increased levels of endothelin-1 and inflammatory mediators. These findings are consistent with the endothelial dysfunction and inflammatory responses observed in spontaneously hypertensive rats, which suggests that the lipopolysaccharide-induced RAECs model effectively mimics key pathological features of hypertension-related endothelial injury. Subsequently, the supernatants from lipopolysaccharide-induced RAECs were combined with 200 μM corticosterone and transferred to neuron-microglia co-cultures to simulate damages to hippocampal neuron under HD conditions. To evaluate the features of cells, neuronal viability was measured by CCK-8 and live-dead assays. Nissl staining was used to assess neuronal Nissl bodies, while the levels of inflammatory factors and monoamine neurotransmitters in the culture supernatants were evaluated by enzyme-linked immunosorbent assays. Reactive oxygen species in neurons were visualized by a fluorescent probe, apoptosis was detected using TUNEL assays, and immunofluorescence was used to assess microglial phenotypes and the levels of TLR4 and NF-κB. It was found that neurons in the tri-culture model had reduced viability, higher levels of apoptosis, fewer Nissl bodies, increased inflammation, and reduced levels of monoamine neurotransmitters. Additionally, the number of M1 microglia was increased, along with elevated levels of TLR4 and NF-κB proteins. These findings were similar to damages of hippocampal neuron, abnormal levels of monoamine neurotransmitters, microglia polarization, and hippocampal inflammatory response observed in the HD rat model. In conclusion, our findings indicate that the tri-culture model can effectively simulate the pathological characteristics of HD, especially in vascular endothelial damage, neuroinflammation, monoamine neurotransmitters disorders. Therefore, the tri-culture model would provides a reliable and invaluable experimental tool for further research on the pathogenesis and treatment of HD.