Omega-3 fatty acid diglyceride emulsions as a novel injectable acute therapeutic in neonatal hypoxic-ischemic brain injury

Abstract

Hypoxic-ischemic encephalopathy (HIE), resulting from a lack of blood flow and oxygen before or during newborn delivery, is a leading cause of cerebral palsy and neurological disability in children. Therapeutic hypothermia (TH), the current standard of care in HIE, is only beneficial in 1 of 7-8 cases. Therefore, there is a critical need for more efficient treatments. We have previously reported that omega-3 (n-3) fatty acids (FA) carried by triglyceride (TG) lipid emulsions provide neuroprotection after experimental hypoxic-ischemic (HI) injury in neonatal mice. Herein, we propose a novel acute therapeutic approach using an n-3 diglyceride (DG) lipid emulsions. Importantly, n-3 DG preparations had much smaller particle size compared to commercially available or lab-made n-3 TG emulsions. We showed that n-3 DG molecules have the advantage of incorporating at substantially higher levels than n-3 TG into an in vitro model of phospholipid membranes. We also observed that n-3 DG after parenteral administration in neonatal mice reaches the bloodstream more rapidly than n-3 TG. Using neonatal HI brain injury models in mice and rats, we found that n-3 DG emulsions provide superior neuroprotection than n-3 TG emulsions or TH in decreasing brain infarct size. Additionally, we found that n-3 DGs attenuate microgliosis and astrogliosis. Thus, n-3 DG emulsions are a superior, promising, and novel therapy for treating HIE.

Keywords: diglycerides; gliosis; hypoxic-ischemic encephalopathy; lipid emulsion; neuroprotection; omega-3 fatty acids.

Graphical abstract

Fig. 1

Experimental design using the Rice–Vannucci model for hypoxic-ischemic encephalopathy (HIE) in mice and rats. (A) Ten-day-old (P10) C57BL/6 mice or (B) seven-day-old (P7) Wistar rat pups underwent surgery for ligation of the carotid artery (CCL), allowed to recover for 1.5 h with their dam, then placed in a hypoxia chamber at 8% oxygen for either 15 min (mice) or 120 min (rats), and returned to their dam to recover. Treatments in mice: two doses of either saline, n-3 DG or n-3 TG, 1 h apart, were IP injected immediately after HI. Sample collection in mice: at 24 h after hypoxic-ischemic (HI) injury, neurofunctional outcomes were performed. Immediately after, animals were euthanized, and brains were collected for TTC staining. Treatments in rats: single dose of either saline, Omegaven® or n-3 DG were IP injected immediately after HI. One group received TH for 5 h immediately after HI. Sample collection in rats: at 7 days after HI injury, brains were collected for H&E staining and immunofluorescence analyses. Abbreviations: diglyceride, DG; triglyceride, TG; therapeutic hypothermia, TH.

Fig. 2

n-3 DG emulsions have smaller particle size and are more stable than n-3TG. (A) Mean particle size (z-average) of diglyceride (DG), triglyceride (TG) and Omegaven®️ emulsions after the preparation procedure; n = 3–10. ****p<0.0001 (one-way ANOVA followed by Tukey's multiple comparisons test); ns: nonsignificant. (B) polydispersity index (PDI) values of DG, TG and Omegaven®️ emulsions; n = 3–10. (C) TLC assay of n-3 DG and TG emulsions; STD = standard: CE, cholesterol ester; TG; FFA, free fatty acids; DG; PL, phospholipids. (D) Mean particle size (z-average) values of DG emulsions stored over 6-month period at 25 °C after an accelerated stress stability study; n = 3–10. (E) p-anisidine values of n-3 DG and TG emulsions; n = 3–5.

Fig. 3

n-3 DGs incorporate in membrane systems more efficiently and are hydrolyzed better than n-3 TG.(A) Representative NMR spectra (carbonyl region) of vesicles with phosphatidylcholine (PC) and 20 mol% DG or TG molecules. Spectra are represented in the following order: PC – DG-DHA – TG-EPA – TG-DHA. (B) and (C) LpL-mediated hydrolysis of n-3 DG vs n-3TG. Experiments were performed with increasing amounts of LpL (0–20 μL of 1:100 dilution) over a fixed time (30 min); data shown are mmol FFA release on the left (B), and FFA areas under the curve on the right (C). n = 4–5. ****p<0.001.

Fig. 4

n-3 DG emulsions enter the systemic circulation more rapidly than n-3 TG after IP injection in neonatal mice. P10 non-fasting mice were IP injected with either n-3 DG or n-3TG emulsion (0.375 g/kg). Blood was collected either at 0, 1, 2 or 4 h after injection of the lipid emulsions. Ordinary two-away ANOVA (mixed-effects model) followed by uncorrected Fisher’s LSD was performed (n =4- 5 in each group); *p<0,05, **p<0.01, ***p<0.001, ****p<0.0001. Each bar is the mean ± SEM with the representation of individual values.

Fig. 5

n-3 DG emulsions show stronger neuroprotection than n-3 TG and TH. (A) Individual values and percentage of tissue loss in saline (n=18), n-3 TG (n=10) and n-3 DG (n=11) treated mice after hypoxic-ischemic injury. Values are mean ± SEM. *p<0.05; ****p<0.0001 (ordinary one-way ANOVA followed by Tukey's multiple comparisons test). Representative images of TTC staining for saline, n-3TG, and n-3 DG treated groups. (B) Individual values and percentage of tissue loss in saline (n=32), n-3TG, Omegaven® (n=12), TH (n=11), and n-3 DG (n=14) treated rats after hypoxic-ischemic injury. Values are mean ± SEM. *p<0.05; **p<0.01. Representative images of H&E staining for saline, TH, n-3TG, and n-3 DG treated groups.

Fig. 6

n-3 DG emulsions preserve neurofunctional outcomes in mice. (A) Righting reflex and (B) Negative geotaxis performances in neonatal mice subjected to ischemic injury and acutely treated with n-3 DG emulsions. Each bar represents the mean ± SEM with representation of individual values. Naïve (n=3); Saline (n=18); n-3 DG (n=11). * p<0.05, ** p<0.01; *** p<0.001 (ordinary one-way ANOVA followed by Tukey's multiple comparisons test).

Fig. 7

n-3 DG emulsions modulate astrogliosis and microgliosis after HI injury in rats. (A) Representative immunostaining of astrocyte cells, GFAP/DAPI+ cells, for naïve, saline, and n-3 DG treated groups in cortex, hippocampus, and thalamus. (B-D) Quantitative analysis (n=7) of GFAP that co-localized with DAPI+ cells in cortex, hippocampus, and thalamus; scale bar = 20 µm. (E) Representative immunostaining of microglia cells, IBA-1/DAPI+ cells, for naïve, saline, and n-3 DG treated groups in cortex, hippocampus, and thalamus. (B-D) Quantitative analysis (n=7) of IBA-1 that co-localized with DAPI+ cells in cortex, hippocampus, and thalamus; scale bar = 20 µm. Values are means ± SEM. Mann–Whitney U tests was performed, with * p<0.05, ** p<0.01; **** p<0.0001.