Nanoliposomes Reduce Stroke Injury Following Middle Cerebral Artery Occlusion in Mice

Abstract

Background and Purpose:

Neuroprotective strategies for stroke remain inadequate. Nanoliposomes comprised of phosphatidylcholine, cholesterol and monosialogangliosides (NL) induced an antioxidant protective response in endothelial cells exposed to amyloid insults. We tested the hypotheses that NL will preserve human neuroblastoma (SH-SY5Y) and brain microvascular endothelial cell (HBMVECs) viability following oxygen-glucose deprivation (OGD)-reoxygenation and will reduce injury in mice following middle cerebral artery occlusion (MCAO).

Methods:

SH-SY5Y and HBMVECs were exposed to OGD-reoxygenation (3-hours 0.5–1% oxygen and glucose-free media followed by 20-hour ambient air/regular media) without or with NL (300 μg/ml). Viability was measured (calcein-acetoxymethyl fluorescence) and protein expression of antioxidant proteins heme oxygenase-1 (HO-1), NAD(P)H quinone dehydrogenase 1 (NQO1) and superoxide dismutase 1 (SOD1) were measured by Western blot. C57BL/6J mice were treated with saline (N=8) or NL (10 mg/ml lipid, 200 μl, N=7) while undergoing 60-minute MCAO followed by reperfusion. Day 2 post-injury neurologic impairment score and infarction size were compared.

Results:

SH-SY5Y and HBMVECs showed reduced viability post-OGD-reoxygenation that was reversed by NL. NL increased protein expressions of HO-1, NQO1 in both cell types and SOD1 in HBMVECs. NL-treated mice showed reduced neurologic impairment and brain infarct size (18.8±2% versus 27.3±2.3%, p=0.017) versus controls.

Conclusions:

NL reduced stroke injury in mice subjected to MCAO likely through induction of an antioxidant stress response. NL is a candidate novel agent for stroke.

Stroke remains a major cause of death and long-term disability¹. Prolonged ischemia and delayed reperfusion exacerbate injury, highlighting need to develop adjuvant therapies addressing reperfusion injury while extending the therapeutic window². Nanoliposomes (<100 nm-sized phospholipids) comprised of cholesterol, phosphatidylcholine and monosialoganglioside (NL) protected endothelial cells against oxidative stress induced by amyloidogenic light chain³ and medin proteins⁴ through nuclear factor erythroid 2-related factor 2 (Nrf2)-mediated activation of antioxidant protective responses, with increased heme oxygenase-1 (HO-1), NAD(P)H quinone dehydrogenase 1 (NQO1) and superoxide dismutase-1 (SOD1). We aim to test the efficacy of NL in mitigating hypoxic injury by testing the hypotheses that NL will preserve viability of human neuroblastoma (SH-SY5Y) and brain microvascular endothelial cells (HBMVECs) exposed to oxygen-glucose deprivation (OGD)-reoxygenation and that NL will reduce brain damage in mice subjected to middle cerebral artery occlusion (MCAO) stroke.

Methods

Data are available upon reasonable request.

Nanoliposomes

NL was prepared from phosphatidylcholine:cholesterol:monosialoganglioside (70:25:5% molar ratios, Avanti, Alabaster AL) using lipid film hydration^{3, 4} (Supplementary Material). Hydrodynamic diameters were measured by dynamic light scattering and zeta potential by electrophoretic light scattering.

Oxygen-glucose deprivation-reoxygenation

SH-SY5Y (ECACC, Public Health England, passages 16–18) and HBMVECs (CellBiologics, Chicago IL, passages 5–8) were exposed to one of the following conditions: (1) ambient air and regular media (control), (2) OGD-reoxygenation⁵ (3-hours of 0.5–1% oxygen inside BioTek-Cytation 5 (Fisher-Scientific, Waltham MA) and glucose-free media (GibcoA1443001, Invitrogen, Carlsbad CA) followed by 20-hours of ambient air/regular media), or (3) OGD-reoxygenation but treated with NL (300 μg/ml, dose previously reported to confer endothelial cell protection^{3, 4}) starting at 1-hour prior to OGD and continuing until harvest. In separate experiments, cells without or with NL treatment were exposed to 20-hour hypoxia (1% oxygen) followed by 3-hours ambient air (Supplement 1). Cells were given calcein-acetoxymethyl (10 nmol/L, Life Technologies, measures cell viability through detection of active/intact esterases⁴) and viability was measured using flow cytometer (Beckman-Coulter FC500, Indianapolis IN, 494/517 nm excitation/emission)⁴. Separate cells were exposed to OGD treatment conditions and HO-1, NQO1 and SOD1 protein expressions measured (Supplemental Material)⁴.

NL Injection

Experiments were approved by the Institutional Animal Care and Use Committees (University of Arizona College of Medicine-Phoenix Protocol#2021–0721, Barrow Neurological Institute/St. Joseph’s Hospital Protocol#540) and procedures followed institutional guidelines. To test delivery and persistence of NL we produced a fluorophore-containing NL modification (phosphatidylcholine, monosialoganglioside, 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-carboxyfluorescein 85:5:10% molar ratios). Three C57BL/6J male mice (8-weeks old, Jackson Laboratory, Bar Harbor ME) underwent surgery to secure a V3 Miniscope to the skull (details in Supplemental Material)⁶. Baseline images were acquired and then 100 μl of NL (10 mg/ml lipid) intravenous (IV) injection was given through tail vein. After 1-day recovery, the process was repeated post-injection of 100 μl of NL via intraperitoneal (IP) injection. Fluorescent signals from 2 regions of interest (cerebral artery and brain parenchyma) were measured using ImageJ (National Institutes of Health, Bethesda MD).

Middle cerebral artery occlusion injury

20-week-old male C57BL/6J mice underwent transient MCAO (details in Supplement 1)⁷. A silicon-coated 6–0 nylon suture was introduced to occlude the middle cerebral artery for 60-minutes followed by filament removal to restore perfusion. Mice were randomly assigned to saline (N=8) or NL (N=7, 100 μl of 10 mg/ml lipid; phosphatidylcholine:cholesterol:monosialoganglioside 70:25:5% molar ratio) intraperitoneally (IP) 1-hour prior to occlusion and a second (100 μl) dose intravenously (IV) immediately before reperfusion.

Neurological deficit scoring (NDS) used modified Bederson scale 48-hours post-stroke (Supplemental Material)⁷ followed by sacrifice. Brain coronal slices (2 mm) were stained with 1% 2,3,4-triphenyl tetrazolium chloride (Sigma-Aldrich). Using ImageJ, corrected infarct volume was calculated⁷. An algorithm described by McBride⁸ was applied to indirectly calculate infarct volumes, corrected for edema, and brain edema. Measurements were performed by an investigator blinded to treatment allocation (AK).

Data Analyses

Cell data were compared using one-way repeated measures analysis of variance (RM-ANOVA) with Holm-Sidak pairwise comparison (normally distributed following log-normal transformation), or RM-ANOVA on ranks with Tukey pairwise comparison (non-normal distribution) (Sigma Stat 3.5, Systat, San Jose CA). Animal outcomes were compared using unpaired Student’s t-test. Significant p-value was set at 0.05 (two-sided). Data are presented as mean±standard error of means.

Results

Nanoliposomes were 38.83±1.23 nm with polydispersity index of 0.32 and Z-potential of −8.76 mV.

SH-SY5Y and HBMVECs exposed to OGD-reoxygenation (Figure 1A) or 20-hour hypoxia (Supplement Figure I) had reduced viability versus control, while treatment with NL restored cell viability. There was no difference in HO-1, NQO1 or SOD1 protein expression between SH-SY5Y and HBMVECs exposed to OGD-reoxygenation versus control, but treatment with NL increased HO-1, NQO1 in both cell types and SOD1 in HBMVECs (Figure 1B–E).

Figure 1. Oxygen-glucose deprivation-reoxygenation in SH-SY5Y and HBMVECs.

A. SH-SY5Y and HBMVECs exposed to OGD-reoxygenation showed reduced viability that was restored by treatment with NL (300 μg/ml). B-D. OGD-reoxygenation did not elicit any change in protein expression of HO-1, NQO1 and SOD1, but treatment with NL showed significant increase in HO-1 and NQO1 in both cell types, and SOD1 in HBMVECs.

p<0.05/**p<0.01/***p<0.001

There was persistent brain fluorescent NL signal up to 2 hours post-IV and more than 4 hours post-IP NL injections (Figure 2).

Figure 2. In vivo administration of NL.

A. Mouse with Miniscope camera and GRIN lens. B. Time course of IV-injected fluorophore-labelled NL. C-E shows the time course of fluorescent signal in circulation (red) and parenchymal brain regions (black) following IV and IP administration. Signal saturates at 255 arbitrary units (A.U.).

For mice randomized to saline versus NL treatment, 2 animals per treatment group met criteria for exclusion based on lack of sufficient (60%) cerebral blood flow reduction⁹ by laser speckle contrast imaging. Mice given NL showed improved neurologic impairment score versus saline control (Figure 3). They also had significantly smaller infarcts and a trend towards decreased brain edema.

Figure 3. NL reduced stroke injury post-MCAO.

Mice treated with NL during MCAO showed less neurological impairment versus saline-treated controls (A). B shows representative brain sections (TTC staining). C-D show MCAO mice treated with NL had smaller infarcts, and trend towards reduced brain edema versus controls.

Discussion

Neutralizing oxidative stress in stroke is a promising adjuvant strategy to reperfusion as the ischemic brain is susceptible to oxidative damage due to high oxygen consumption and low antioxidant capacity¹. NL prevented endothelial dysfunction and cell death induced by amyloidogenic light chain and aging-associated medin proteins^{3, 4}. Our results show that NL also preserved viability of SH-SY5Y and HBMVECs exposed to OGD-reoxygenation. In previous work, NL’s protective effect was attributed to increased gene and protein expression of endogenous antioxidants HO-1, NQO1 and SOD1 through Nrf2-mediated signaling^{3, 4}. Similarly, in the current study, NL-treated SH-SY5Y and HBMVECs exposed to OGD-reoxygenation increased HO-1 and NQO1, although interestingly SOD1 was increased only in HBMVECs. In vivo, NL treatment reduced functional impairment and infarct size post-MCAO-reperfusion. The protective effect of NL following stroke is likely related, at least in part, to its ability to protect neuronal and endothelial cells against hypoxic insult, as demonstrated in vitro, although this has to be empirically validated in vivo in future follow-up studies. Miniscope imaging showed that NL persists in cerebrovascular circulation 2 hours or longer post-IV or post-IP administration, suggesting sufficient exposure time within the therapeutic window for stroke.

The data are consistent with prior reports of mitigation of stroke injury by monosialoganglioside in preclinical models¹⁰, although monosialoganglioside had equivocal results in human clinical trials^{11, 12}. Our study differed by injecting phospholipids in nanoliposomal structures, and our NL was composed of 3 phospholipids, with monosialoganglioside comprising only 5% molar weight. The prior study¹⁰ (150 mg/kg of monosialoganglioside) resulted in 6.8% reduction whereas our NL with ~8.4 mg/kg monosialoganglioside resulted in 8.5% reduction in infarct volume. The prior study injected free monosialoganglioside in saline; due to its low solubility in aqueous medium this glycolipid forms micellar aggregates the sizes of which were not reported. The anticipated uncontrolled aggregate formation can potentially limit bioavailability and reduce efficacy. This might explain why we required less monosialoganglioside to produce a protective effect.

Limitations include testing only males. The NL timing employed would be translatable only to strokes with prodromal transient ischemic attacks wherein NL could be given before complete occlusion onset. NL efficacy when administered solely post-occlusion needs to be tested in the future. Follow-up studies should further tease out mechanistic signaling including possible protection of blood brain barrier function, validate results in primary neurons, and confirm NL-induced antioxidant increase and neuronal viability in brain tissue in vivo.

Phosphatidylcholine, cholesterol and monosialoganglioside NL prevented hypoxia-induced injury in neuroblastoma and endothelial cells and reduced functional and structural brain damage in mice exposed to MCAO stroke.

Non-standard Abbreviations:

AU

arbitrary units

HBMVECs

human brain microvascular endothelial cell

HO-1

heme oxygenase-1

IP

intraperitoneal

IV

intravenous

MCAO

middle cerebral artery occlusion

NDS

neurologic deficit scoring

NL

nanoliposomes

NQO1

NAD(P)H quinone dehydrogenase 1

OGD

oxygen-glucose deprivation

RM-ANOVA

repeated measures analysis of variance

SOD1

superoxide dismutase 1