In Vitro Model of Blood-Brain Barrier Dysfunction and Chronic Mast Cell Activation in Autism Spectrum Disorder

Introduction:
Autism Spectrum Disorder (ASD) affects approximately 1 in 36 8-year-old children in the US.¹ A significant subset (26.7%) have profound ASD, and are generally non-/minimally verbal with substantial co-morbidities.² The development of effective treatments for profound autism remains a challenge due to the complex nature of ASD and its poorly understood etiology. Nevertheless, pathological features such as neuroanatomical alterations, neuroinflammation, and immune dysfunction have been consistently observed.^3–5 Importantly, concurrent neuroinflammation and blood-brain barrier (BBB) dysfunction – observed in numerous neuropsychiatric disorders – exacerbates neuroinflammatory and neuroanatomical aberrations via a self-propagating feedback loop.⁶ In ASD, the few available studies have indicated BBB alterations are present, with significant involvement of inflammatory mast cells (MCs).^5,7 However, the potential interplay between BBB dysfunction and MC-driven neuroinflammation in ASD requires further investigation. Herein, we first establish that human MCs and primary human brain microvascular endothelial cells (BMECs) triggered with pro-inflammatory cytokine TNF result in further production of pro-inflammatory factors in co-culture, indicating a positive feedback loop. We then investigate cellular dysfunction between BMECs and cortical neurons – both derived from induced pluripotent stem cells (iPSCs) – in ASD relative to typical development (TD). To gain insight into the role of altered immune landscapes characteristic of ASD on BBB dysfunction, factors released by neurotensin (NT)-activated MCs were applied to BMEC-neuron cocultures. Through this, we aim to evaluate differences in TD and ASD BMEC-neuron cocultures following insult with activated MC conditioned medium for understanding BBB dysfunction in ASD.




Materials and Methods::
hBMECs (BMECs) were seeded onto fibronectin-coated transwell inserts at 60,000 cells/cm² and maintained for 3 days before stimulation in growth medium containing 100 ng/mL TNFα. A human MC line HMC-1.2s were stimulated in growth medium containing 100 ng/mL TNFα. After BMECs and MCs were separately stimulated for 24 hrs, a coculture was established with BMECs and MCs seeded at a 1:1 ratio. Following stimulation, designated groups were treated for 24 hrs with suramin or rapamycin before sample collection. Next, two iPSC lines – one being from a profound ASD donor and one being from an age- and gender-matched TD donor – were used for differentiation of BMECs and cortical neurons. BMEC differentiations and characterizations were completed in accordance with the literature, specifically referencing the protocol outlined in Qian et al.⁸ Neuron differentiations were completed using commercially available kits through STEMCELL Technologies. A transwell coculture system was established with BMECs seeded on the permeable transwell insert and neurons seeded at the bottom of the well. Paracrine interactions between these cell types were investigated in the presence or absence of stimulation from medium conditioned by HMC-1.2s. These MCs were activated by NT, which triggers degranulation and sustained release of proinflammatory cytokines. Cocultures (hBMEC-MC and BMEC-neuron) were assessed for barrier integrity, neuronal dysregulation, and expression of proinflammatory markers through transendothelial electrical resistance (TEER), immunocytochemistry (ICC), and multiplex immunoassays. Results of multiplex immunoassays were analyzed with Partial Least Squares – Discriminant Analysis.




Results, Conclusions, and Discussions::
In examining the effects of pro-inflammatory MCs on TNFa stimulated BMECs, we found that BMECs exposed to TNFa had a marked increase in proinflammatory markers GM-CSF (~15.4-fold, p<0.001), CCL2 (~3.8-fold, p<0.01), and CCL5 (~4.8-fold, p<0.001) along with a decrease in anti-inflammatory IGFBP-2 (~1.4-fold, p<0.01) compared to unstimulated BMEC CTL. Incorporating stimulated MCs did not significantly increase BMEC proinflammatory markers compared to stimulated monocultures BMECs. However, coculture conditions significantly impacted intracellular MC expression as can be seen by a large shift in the expression profile from PLS-DA. Specifically, coculture resulted in increased IL-6 (~129.6-fold, p<0.01), IL-8 (~79.3-fold, p<0.001 ), and IL-1B (~9.2-fold, p<0.001) compared to stimulated monocultures of MCs. In BMECs, rapamycin and suramin decreased expression of GM-CSF (~1.35-fold, p<0.05 and ~1.36-fold, p<0.05, respectively) but they did not significantly decrease CCL2 or CCL5. With MCs playing a substantial role in BMEC dysregulation, and BMECs having a known impact on neuronal function, it is therefore reasonable to explore the role of MC secretory products in an iPSC derived BMEC-cortical neuron coculture to better understand potential feedback loops. While coculture of BMECs and cortical neurons did not significantly impact TEER, addition of MC conditioned media caused a significant decrease in TEER for TD BMECs, while FXS derived BMECs had consistently decreased TEER compared to TD. Further testing is required to understand inflammatory differences between experimental groups, however a clear dysfunction is seen in FXS compared to TD.




Acknowledgements (Optional): :
This work was supported by National Institute of Health Research Grant R03AG067140 to M.A.G. and M.S.H and R03AG067970 to M.S.H.




References (Optional): :
1. Maenner, M. J. et al. MMWR Surveill. Summ. (2023), 72, 1–14.




2. Hughes, M. M. et al. Public Health Rep. (2023), 003335492311635 (2023).




3. Lord, C. et al. Nat. Rev. Dis. Primer. (2020), 6, 5.




4. Croonenberghs, J., et al. Neuropsychobiology. (2002), 45, 1–6.




5. van Sadelhoff, J. H. J. et al. Front. Endocrinol. (2019), 10, 247.




6. Takata, F., et al. Front. Cell. Neurosci. (2021) 15.




7. Fiorentino, M. et al. Mol. Autism. (2016), 7.




8. Qian, T. et al. Sci. Adv. (2017), 3.




9. Autar, K. et al. Stem Cell Rep.(2022) 17, 96–109.