Lipoprotein Lipase Is a Feature of Alternatively-Activated Microglia and May Facilitate Lipid Uptake in the CNS During Demyelination

Abstract

Severe demyelinating disorders of the central nervous system (CNS) such as multiple sclerosis (MS), can be devastating for many young lives. To date, the factors resulting in poor remyelination and repair are not well understood, and reparative therapies that benefit MS patients have yet to be developed. We have previously shown that the activity and abundance of Lipoprotein Lipase (LPL)-the rate-limiting enzyme in the hydrolysis of triglyceride-rich lipoproteins-is increased in Schwann cells and macrophages following nerve crush injury in the peripheral nervous system (PNS), suggesting that LPL may help scavenge myelin-derived lipids. We hypothesized that LPL may play a similar role in the CNS. To test this, mice were immunized with MOG_35-55 peptide to induce experimental allergic encephalomyelitis (EAE). LPL activity was increased (p < 0.05) in the brain at 30 days post-injection, coinciding with partial remission of clinical symptoms. Furthermore, LPL abundance and activity was up-regulated (p < 0.05) at the transition between de- and re-myelination in lysolecithin-treated ex vivo cerebellar slices. Since microglia are the key immune effector cells of the CNS we determined the role of LPL in microglia. Lipid uptake was decreased (p < 0.001) in LPL-deficient BV-2 microglial cells compared to WT. In addition, LPL-deficient cells showed dramatically reduced expression of anti-inflammatory markers, YM1 (-22 fold, p < 0.001), and arginase 1 (Arg1; -265 fold, p < 0.001) and increased expression of pro-inflammatory markers, such as iNOS compared to WT cells (+53 fold, p < 0.001). This suggests that LPL is a feature of reparative microglia, further supported by the metabolic and inflammatory profile of LPL-deficient microglia. Taken together, our data strongly suggest that LPL expression is a novel feature of a microglial phenotype that supports remyelination and repair through the clearance of lipid debris. This mechanism may be exploited to develop future reparative therapies for MS and primary neurodegenerative disorders (Alzheimer's disease (AD) and Parkinson's disease).

Keywords: lipid metabolism; lipoprotein lipase (LPL); microglia; multiple sclerosis; myelination.

Figure 1

Lipoprotein lipase (LPL) is increased during de- and re-myelination. (A) Mean symptom scores of MOG treated (N = 10) and Control (N = 20) mice. (B) LPL activity in brain tissue at 20 and 30 days MOG post injection from Control (N = 20), MOG-treated (symptomatic; N = 10), and MOG-recovered (MOG-treated but asymptomatic following an initial bout of symptoms, N = 3). (C) LPL enzymatic activity in cerebellar slice cultures at 0, 1, 2, 7 and 14 days post LPC treatment (N = 3 per group, per time point). (D) LPL gene expression in cerebellar slice cultures at 0, 1, 2, 7 and 14 days post LPC treatment (N = 3 per group, per time point). (E–J) Lysolecithin-mediated demyelination of ex vivo cerebella brain slices (N = 3 per group, per time point). Blue—neurofilament (NFH), Green—PLPeGFP, Red—myelin associated glycoprotein (MAG). (E–G) No Lysolecithin control, 1, 4 and 7 days. (H–J) 1, 4 and 7 days post Lysolecithin treatment. *P < 0.05 vs. CONT (at corresponding time point). **P < 0.01 vs. CONT (at corresponding time point). ***P < 0.001 vs. CONT (at corresponding time point).

Figure 2

Characterization of LPL knock down (KD) cells. (A–D) Gene expression in microglial BV-2 KD (553) vs. control (202) cells (N = 4 per group). (E) Western blot analysis showing reduced LPL protein expression in BV-2 553 cells. ***P < 0.001 vs. CONT.

Figure 3

Chemokine and cytokine production in LPL KD microglia. Ray biotech mouse neuro array was used to determine differences in chemokine expression in media from (A) BV-2 202 or (B) BV-2 553 cells. (C–F) Comparative gene expression between control (BV-2 202) and LPL KD (BV-2 553) cells (N = 4 per group). (G) Ex vivo brain slices following incubation with conditioned media (CM) from either control (WT CM) or LPL KD (KD CM) for 24 h, characterizing Plp-eGFP MAG, neurofilament or Iba1 expression (N = 3 per group). ***P < 0.001 vs. CONT.

Figure 4

Lipopolysaccharide (LPS) activation reduces LPL expression and activity in both immortalized cells (N = 4 per group; A,B) and primary microglial cells from adult brain (C,D; N = 3 per group). *P < 0.05 vs. CONT. **P < 0.01 vs. CONT.

Figure 5

Glycolysis is increased and fatty acid oxidation (FAO) is decreased in BV-2 microglial cells lacking LPL. (A) Global metabolomics in control and LPL KD BV-2 microglia (N = 3 per group). (B) Schematic representation of metabolic reprogramming following depletion of LPL in microglia. (C) FAO in control vs. LPL KD BV-2 microglia (N = 6 per group). *P < 0.05 vs. CONT. **P < 0.01 vs. CONT. ***P < 0.001 vs. CONT.

Figure 6

LPL is involved in lipid and lipoprotein uptake in microglia. (A) Intracellular uptake of radiolabeled synthetic lipid vesicles containing 50 or 100 μM Triolein Tg. (B) Uptake/phagocytosis of 1′-dioctadecyl-3,3,3′3′-tetramethylindocarbocyanine perchlorate (DiI) labeled synthetic chylomicrons (SC) and phospholipid liposomes (PLs; N = 3 per group). (C–F) Expression of scavenger and lipoprotein receptors in BV-2 202 vs. 553 cells (N = 4 per group). **P < 0.01 vs. CONT. ***P < 0.001 vs. CONT.

Figure 7

CM from LPL KD cells prevents normal remyelination processes. Ex vivo cerebellar slices were demyelinated by overnight treatment (17 h) with lysolecithin. The next day slices were incubated with either standard slice media, or CM and allowed to recover for 7 days.