Sex differences in the blood-brain barrier and neurodegenerative diseases

Abstract

The number of people diagnosed with neurodegenerative diseases is on the rise. Many of these diseases, including Alzheimer's disease, Parkinson's disease, multiple sclerosis, and motor neuron disease, demonstrate clear sexual dimorphisms. While sex as a biological variable must now be included in animal studies, sex is rarely included in in vitro models of human neurodegenerative disease. In this Review, we describe these sex-related differences in neurodegenerative diseases and the blood-brain barrier (BBB), whose dysfunction is linked to neurodegenerative disease development and progression. We explain potential mechanisms by which sex and sex hormones affect BBB integrity. Finally, we summarize current in vitro BBB bioengineered models and highlight their potential to study sex differences in BBB integrity and neurodegenerative disease.

FIG. 1.

Overview of BBB degradation with age and disease. The BBB, which is formed by BMECs and maintained through interactions with pericytes and astrocytes, restricts cell and molecule movement from the blood into the brain. BMECs form impermeable intercellular junctions through tight junction proteins, including occludins, claudins, VE-Cadherin, JAMs, and ZO proteins. With the increasing age and disease, tight junction proteins degrade, leading to BBB opening. The leaky BBB allows cytokines, neurotoxins, and leukocytes to infiltrate the brain, which can cause downstream inflammation and neurodegeneration.

FIG. 2.

Mechanisms through which biological sex could affect BBB integrity. (a) Estrogen increases NO production. In the classic genomic pathway, estradiol (E2) binds to transmembrane estrogen receptors, which are then internalized and dimerize before binding to E2 response elements. The complex then regulates eNOS transcription and, thus, NO production. In the non-classic genomic pathway, E2 binds to either ER-associated GPCRs or GPR30, which triggers an intracellular signaling cascade including MAPK/ERK, PI₃K, and cAMP. This then leads to increased eNOS transcription and NO production through co-factors such as SP-1, AP-1, and NF-κB. In the non-genomic pathway, E2 binding to ER-associated GPCRs or GPR30 activates Akt, which phosphorylates eNOS at Ser1177 and enables NO production. (b) Increased MMP-9 and MMP-2 or decreased TIMP1 and TIMP2 could lead to collagen IV degradation in the BBB extracellular matrix and break down claudin-5, occludin, and ZO1 to decrease BBB integrity. (c), Estrogen may inhibit the RhoA/ROCK2 pathway to maintain BBB integrity. Inflammatory cytokines bind to CCR2, which activates RhoA/ROCK2 to inhibit MLCP. This leads to tight junction protein internalization and degradation as well as actin stress fiber formation, which contracts the cell and pulls apart the BBB. E2 binding inhibits RhoA/ROCK2 to maintain the BBB. (d) In the presence of inflammatory cytokines, downstream GPCR signaling from E2 binding to ERs leads to ANXA1 phosphorylation, which stabilizes tight junction proteins and inhibits NF-κB to downregulate VCAM-1 and ICAM-1 expression on the plasma membrane. This then reduces the inflammatory response and leukocyte transmigration.

FIG. 3.

Current 3D in vitro models of a functional BBB. (a) Transwell filter models include a BMEC monolayer on top of the semi-permeable membrane, pericytes on the bottom of the semi-permeable membrane, and pericytes, astrocytes, and neurons below the filter. (b) Hydrogel models incorporate BMEC lining a hollow channel, and pericytes, astrocytes, and neurons dispersed in the surrounding hydrogel. (c) Microfluidic BBB models feature BMECs lining the blood compartment and pericytes, astrocytes, and neurons in the brain channel. (d) Organoid brain models include self-organized capillary networks representing the BBB, surrounded by neurons and astrocytes.