Skip to main content

Advances in blood-brain barrier modeling in microphysiological systems highlight critical differences in opioid transport due to cortisol exposure


Brown JAJacquelyn A , Faley SLShannon L , Shi YYajuan , Hillgren KMKathleen M , Sawada GAGeri A , Baker TKThomas K , Wikswo JPJohn P , Lippmann ESEthan S . Fluids and barriers of the CNS. 2020 6 3; 17(1). 38


BACKGROUND: The United States faces a national crisis involving opioid medications, where currently more than 130 people die every day. To combat this epidemic, a better understanding is needed of how opioids penetrate into the central nervous system (CNS) to facilitate pain relief and, potentially, result in addiction and/or misuse. Animal models, however, are a poor predictor of blood-brain barrier (BBB) transport and CNS drug penetration in humans, and many traditional 2D cell culture models of the BBB and neurovascular unit have inadequate barrier function and weak or inappropriate efflux transporter expression. Here, we sought to better understand opioid transport mechanisms using a simplified microfluidic neurovascular unit (NVU) model consisting of human brain microvascular endothelial cells (BMECs) co-cultured with astrocytes.

METHODS: Human primary and induced pluripotent stem cell (iPSC)-derived BMECs were incorporated into a microfluidic NVU model with several technical improvements over our previous design. Passive barrier function was assessed by permeability of fluorescent dextrans with varying sizes, and P-glycoprotein function was assessed by rhodamine permeability in the presence or absence of inhibitors; quantification was performed with a fluorescent plate reader. Loperamide, morphine, and oxycodone permeability was assessed in the presence or absence of P-glycoprotein inhibitors and cortisol; quantification was performed with mass spectrometry.

RESULTS: We first report technical and methodological optimizations to our previously described microfluidic model using primary human BMECs, which results in accelerated barrier formation, decreased variability, and reduced passive permeability relative to Transwell models. We then demonstrate proper transport and efflux of loperamide, morphine, and oxycodone in the microfluidic NVU containing BMECs derived from human iPSCs. We further demonstrate that cortisol can alter permeability of loperamide and morphine in a divergent manner.

CONCLUSIONS: We reveal a novel role for the stress hormone cortisol in modulating the transport of opioids across the BBB, which could contribute to their abuse or overdose. Our updated BBB model represents a powerful tool available to researchers, clinicians, and drug manufacturers for understanding the mechanisms by which opioids access the CNS.