(M1530-01-04) Heat Shock Protein-Loaded Mitochondria Containing Microvesicles Protect the Blood-Brain Barrier in Ischemic Stroke

Purpose: Oxygen-glucose deprivation (OGD) during ischemic stroke leads to mitochondrial damage and disruption of tight junctions in brain endothelial cells (BECs) lining the blood-brain barrier (BBB) (1). Restoration of blood flow (reperfusion) upon first-line/acute stroke treatment further damages the BBB integrity leading to the infiltration of inflammatory mediators from blood to the brain parenchyma (2). Therefore, protecting the BBB integrity is a promising approach to alleviate stroke-induced long-term neurovascular unit damage. Preclinical studies have demonstrated that overexpression of 27 kDa heat shock protein (HSP27) in BECs elicits protection against stroke-induced BBB disruption and allows long-term protection of neurological functions (3). Cell-derived microvesicles (MVs) are intriguing carriers due to the presence of innate mitochondrial components and their membrane affinity properties (4, 5). Our previous studies demonstrated that human BEC-derived MVs contain mitochondria and MV-mitochondria transferred into recipient BECs (Figure 1). MV-mediated mitochondria transfer increased recipient BEC ATP levels and mitochondrial respiration. MV-injected stroke mice showed a reduction in brain infarct volume compared to the vehicle-treated group (Figure 2). In the present work, we hypothesize that HSP27 loaded-MVs (HSP27-MVs) may concurrently increase BEC survival (due to MV mitochondria) and protect BBB integrity (due to HSP27 effects) for treating ischemic stroke (Figure 2).

Methods: We isolated MVs from a human BEC cell line using a sequential ultracentrifugation protocol. We used an incubation method to load human recombinant HSP27 in MVs (HSP27-MVs) either in the presence or absence of a cationic polyethylene glycol-b-poly diethyltriamine (PEG-DET). We determined the loading of HSP27 using native polyacrylamide gel electrophoresis (PAGE) and particle physicochemical characteristics using dynamic light scattering. The cytocompatibility of HSP27-MVs was determined using a Cell Titer glo-based ATP assay. We measured the paracellular permeability of 4.4 and 65-85 kDa Dextran tracers across primary BEC monolayers in transwell culture inserts to evaluate the effect of HSP27-MVs on tight junction integrity.

Results: The densitometry analysis of native PAGE revealed that incubation of HSP27 with MV resulted in about 57% of total HSP27 loading in MV (Figure 3), whereas PEG-DET-based HSP27-MVs showed about 45% HSP27 loading (Figure 3). Importantly, our MV-based HSP27 loading is three-fold higher than published reports (6, 7) on therapeutic protein loading in EVs. HSP27-MVs did not show any change in particle diameters or surface charge compared to naïve MVs. In contrast, PEG-DET-based HSP27-MVs showed a significant (p < 0.05) increase in particle diameters compared to naïve MVs. HSP27-MVs were cytocompatible with recipient BECs in vitro, confirming the safety of HSP27-MVs. Importantly, HSP27-MVs showed a prolonged (24 h) decrease in the paracellular permeability of small and large molecular mass fluorescent tracers during reperfusion compared to untreated and free HSP27 protein-treated cells.

Conclusion: Incubation of mitochondria-containing MVs with HSP27 showed about 57% HSP27 loading efficiency. No prior work has demonstrated concurrent mitochondria and exogenous HSP27 loading in a single MV carrier. MV mediated an exceptionally higher amount of HSP27 loading, suggesting that HSP27-MVs can be a potent treatment strategy for ischemic stroke. HSP27-MVs were cytocompatible and protected BBB integrity in vitro, suggesting their potential for BBB protection in treating stroke. We will determine the therapeutic efficacy of HSP27-EVs treatment in reducing brain infarct volume and BBB permeability in the middle cerebral artery occlusion mouse model of ischemic stroke.

References: 1. Lee MJ, Jang Y, Han J, Kim SJ, Ju X, Lee YL, et al. Endothelial-specific Crif1 deletion induces BBB maturation and disruption via the alteration of actin dynamics by impaired mitochondrial respiration. J Cereb Blood Flow Metab. 2020;40(7):1546-61.
2. Shi Y, Zhang L, Pu H, Mao L, Hu X, Jiang X, et al. Rapid endothelial cytoskeletal reorganization enables early blood-brain barrier disruption and long-term ischaemic reperfusion brain injury. Nat Commun. 2016;7:10523.
3. Shi YA-O, Jiang X, Zhang L, Pu H, Hu X, Zhang W, et al. Endothelium-targeted overexpression of heat shock protein 27 ameliorates blood-brain barrier disruption after ischemic brain injury. (1091-6490 (Electronic)).
4. Manickam DS. Delivery of mitochondria via extracellular vesicles – A new horizon in drug delivery. Journal of Controlled Release. 2022;343:400-7.
5. Dave KM, Stolz DB, Venna VR, Quaicoe VA, Maniskas ME, Reynolds MJ, et al. Mitochondria-containing extracellular vesicles (EV) reduce mouse brain infarct sizes and EV/HSP27 protect ischemic brain endothelial cultures. J Control Release. 2023;354:368-93.
6. Yerneni SS, Adamik J, Weiss LE, Campbell PG. Cell trafficking and regulation of osteoblastogenesis by extracellular vesicle associated bone morphogenetic protein 2. Journal of Extracellular Vesicles. 2021;10(12):e12155.
7. Haney MJ, Klyachko NL, Zhao Y, Gupta R, Plotnikova EG, He Z, et al. Exosomes as drug delivery vehicles for Parkinson's disease therapy. J Control Release. 2015;207:18-30.

Figure 1: Transmission electron microscopy of cross-sectioned MVs showed presence of mitochondria in MV lumen. Mitochondria-labelled MVs transferred in recipient BECs and colocalized with host mitochondria.

Figure 2: Schematic illustration of HSP27-MVs mediated increase in BEC mitochondrial functions and BBB integrity. MV showed neuroprotection in a mouse model of ischemic stroke.

Figure 3: Native polyacrylamide gel electrophoresis for HSP27 loading in HSP27-MV formulations. HSP27-MVs without lysis buffer indicates HSP27 loading on MV surface. HSP27-MVs were lysed with lysis buffer to determine total HSP27 loading in MVs. PEG-DET-based HSP27-MVs showed about 45% HSP27 loading efficiency. native HSP27 was used as a control to calculate HSP27 loading in HSP27-MVs.