Laforin prevents stress-induced polyglucosan body formation and Lafora disease progression in neurons

Abstract

Glycogen, the largest cytosolic macromolecule, is soluble because of intricate construction generating perfect hydrophilic-surfaced spheres. Little is known about neuronal glycogen function and metabolism, though progress is accruing through the neurodegenerative epilepsy Lafora disease (LD) proteins laforin and malin. Neurons in LD exhibit Lafora bodies (LBs), large accumulations of malconstructed insoluble glycogen (polyglucosans). We demonstrated that the laforin-malin complex reduces LBs and protects neuronal cells against endoplasmic reticulum stress-induced apoptosis. We now show that stress induces polyglucosan formation in normal neurons in culture and in the brain. This is mediated by increased glucose-6-phosphate allosterically hyperactivating muscle glycogen synthase (GS1) and is followed by activation of the glycogen digesting enzyme glycogen phosphorylase. In the absence of laforin, stress-induced polyglucosans are undigested and accumulate into massive LBs, and in laforin-deficient mice, stress drastically accelerates LB accumulation and LD. The mechanism through which laforin-malin mediates polyglucosan degradation remains unclear but involves GS1 dephosphorylation by laforin. Our work uncovers the presence of rapid polyglucosan metabolism as part of the normal physiology of neuroprotection. We propose that deficiency in the degradative phase of this metabolism, leading to LB accumulation and resultant seizure predisposition and neurodegeneration, underlies LD.

FIG. 1

ER stress induces glycogen bodies. (A,B) Laforin associates with the glycogen metabolic complex. HEK293 and N2A cells were transiently transfected with laforin-Flag plasmid for 24 hrs before N2A transfectants were treated with or without 2 μM thapsigargin for 8 hrs. Coomassie blue staining of laforin-Flag eluate of HEK293 cells is shown in (A); MS analysis of Flag peptide-purified laforin-associated proteins is shown in (B). Only proteins involved in glycogen metabolism are shown. *denotes two thirds of the laforin band on Coomassie blue-stained gel were cut out and remaining gel was digested for MS analysis. (C) GS1-positive glycogen granules are induced by ER stressors. N2A cells were treated with or without 2.5 mM 2-DG, 2 μg/ml tunicamycin, or 2 μM thapsigargin for 8 hrs in MEM medium containing 2.5% FBS and 5mM (for 2-DG) or 25mM (for others) glucose. The cells were then fixed and stained with an antibody against GS1, doubly stained with antibodies against GS1 and glycogen, or stained with the PAS reagent. Nuclei were stained with DAPI and merged images of immunostainings are shown. Scale bars, 10μm. (D,E) Increased ER stress marker (D) and glycogen content (E) in stressed cells. ER stress markers and glycogen content were determined in the lysate of N2A cells treated as in (C).

FIG. 2

ER stressors elevate allosteric regulators and enzyme activities required for glycogen synthesis and then breakdown, in that sequence. (A) Time course of indicated components as induced by 2-DG. N2A cells were treated with 2-DG for indicated times in 5 mM glucose DMEM medium containing 2.5% FBS. The levels of G6P, glycogen, and AMP, the activity levels of GS1 and GPBB, and the values of untreated vehicle controls, shown at time zero, were monitored. (B) Same experiment as in (A) but using all three ER stressors and showing results at selected time points rather than the entire chase; N2A cells used; 2-DG (2.5 mM), tunicamycin (2 μg/ml), thapsigargin (2 μM); G6P at 4 hrs, GS at 4 hrs, AMP at 6 hrs, and GPBB at 6 hrs. The error bar represents the mean + SD of three separate experiments.

FIG. 3

GS1 synthesizes LB-like polyglucosans that resist α-amylase hydrolysis. (A) GS1 synthesizes α-amylase-resistant polyglucosan in stressed cells. After treatment with 2 μM thapsigargin for 8 hrs, N2A cells were fixed by methanol. The fixed cells were treated or not with 0.2U/ml α-amylase for 15 min at 37°C and then stained with PAS. (B,C) Overexpressed GS1 synthesizes polyglucosans. HEK293 cells were transiently transfected with Flag-tagged WT GS1, truncated forms of GS1, PTG, or GS2 for 24 hrs. Next, the transfected cells were fixed by methanol and stained with Cy3-Flag antibody. PAS reagent stained the fixed cells that were pretreated with or without 0.2U/ml α-amylase for 30 min at 37°C (B). Glycogen content in lysates of transfected HEK293 cells before or after α-amylase pretreatment was quantified (C). Scale bars, 10μm.

FIG. 4

GS1 is phosphorylated in synthesized polyglucosans. (A,B) GS1 is phosphorylated in PBs induced by stress. N2A cells treated with or without 2 μM thapsigargin for 8 hrs were fixed and doubly stained with antibodies to p-GS1 and glycogen (A). After treatment with thapsigargin as (A), N2A cells were lysed by 0.55% NP-40 detergent and PBs were fractioned and subjected to Western blot (B). (C,D) GS1 is phosphorylated in PBs at physiological conditions. HEK293 cells were transiently transfected with WT GS1, its truncated forms, PTG, or GS2. 24 hrs after transfection, the cells were fixed and stained with antibodies to p-GS1 and glycogen (C). Western blot detected p-GS1 in HEK293 cells transfected with GS1 or other plasmids as indicated (D). (E) Quantitation of glycogen content in polyglucosans isolated from HEK293 cells transfected with indicated plasmids for 24 hrs. Scale bars, 10μm.

FIG. 5

Laforin is a phosphatase of GS1 in PBs. (A,B) Examination of laforin phosphatase activity on GS1 present in PBs in vivo. GS1-ΔN10 was co-transfected with laforin or C265S into HEK293 cells for 24 hrs before the transfected cells were treated or not with 2μM thapsigargin for 8 hrs. The co-localization of laforin or C265S with p-GS1-ΔN10 in PBs was identified by double staining (A), and the dephosphorylation of p-GS1-ΔN10 by laforin was detected by Western blot (B). (C,D) Determination of laforin phosphatase activity in vitro. Isolated GS1-ΔN10 polyglucosans, at an amount equal to 10μg glycogen, was added to phosphatase buffer containing purified laforin or C265S. After a 30 min incubation at 37°C, dephosphorylation of GS1-ΔN10 was detected by binding of antibody to p-GS1 (C). The released phosphate content by laforin from the GS1-ΔN10 polyglucosans was monitored (D). Scale bars, 10μm.

FIG. 6

ER stress accelerates LD. (A,B,C) PB induction in cultured neurons. Postnatal cortical neurons from WT and Epm2a KO mice were stimulated or not by ER stressors (0.5 mM 2-DG, 0.5 μg/ml tunicamycin, or 0.25 μg/ml thapsigargin) in neuron culture medium for 6 hrs. The neurons were then fixed and doubly stained with antibodies to p-GS1 and the neuronal marker Tuj-1 (A). The treated neurons were lysed in buffer containing 1% Triton X-100 and 0.1% SDS for Western blot (B). Quantitation of glycogen content of the PBs isolated from stressed neurons of WT mice is shown (C). The result is expressed as total glycogen content divided by total protein content in the isolated polyglucosan deposit after it was completely digested by amyloglucosidase and amylase. (D) PB induction in mice by 2-DG. Age-matched 8-week-old WT and Epm2a KO mice were intraperitoneally administered 2g/kg 2-DG every other day for a period encompassing 8 injections. 24 hrs after the last injection, the hippocampal sections of each group were stained by antibodies to GS1, p-GS1, or PAS reagent (bottom). Selected areas of WT sections are enlarged and shown on the right. (E)Phosphorylated GS1 is highly abundant in PBs in brain as Epm2a KO mice age. Numerous PBs staining positive for GS1, p-GS1, and PAS are observed in the hippocampal CA1 regions of 9-month-old Epm2a KO mice. Only merged photos are shown. (F) Phosphorylated GS1 is abundant in LBs of hippocampus CA1 region of 6-month-old malin KO mouse. Scale bars, 10μm.

FIG. 6

ER stress accelerates LD. (A,B,C) PB induction in cultured neurons. Postnatal cortical neurons from WT and Epm2a KO mice were stimulated or not by ER stressors (0.5 mM 2-DG, 0.5 μg/ml tunicamycin, or 0.25 μg/ml thapsigargin) in neuron culture medium for 6 hrs. The neurons were then fixed and doubly stained with antibodies to p-GS1 and the neuronal marker Tuj-1 (A). The treated neurons were lysed in buffer containing 1% Triton X-100 and 0.1% SDS for Western blot (B). Quantitation of glycogen content of the PBs isolated from stressed neurons of WT mice is shown (C). The result is expressed as total glycogen content divided by total protein content in the isolated polyglucosan deposit after it was completely digested by amyloglucosidase and amylase. (D) PB induction in mice by 2-DG. Age-matched 8-week-old WT and Epm2a KO mice were intraperitoneally administered 2g/kg 2-DG every other day for a period encompassing 8 injections. 24 hrs after the last injection, the hippocampal sections of each group were stained by antibodies to GS1, p-GS1, or PAS reagent (bottom). Selected areas of WT sections are enlarged and shown on the right. (E)Phosphorylated GS1 is highly abundant in PBs in brain as Epm2a KO mice age. Numerous PBs staining positive for GS1, p-GS1, and PAS are observed in the hippocampal CA1 regions of 9-month-old Epm2a KO mice. Only merged photos are shown. (F) Phosphorylated GS1 is abundant in LBs of hippocampus CA1 region of 6-month-old malin KO mouse. Scale bars, 10μm.

FIG. 7

Stress-accelerated LD includes increased seizure susceptibility and neurodegeneration. (A) Tonic-clonic seizure induction by 2-DG. Ten weeks-old WT and Epm2a KO mice were given 2-DG, every other day, at doses of 2g/kg for the first 4 injections, and 2.5g/kg for the second 4 injections. The number of mice in each group with a seizure score of 5 induced at each injection was compared. (B,C,D) 2-DG induced PB formation and neurodegeneration in hippocampal neurons. 24 hrs after the last injection with 2-DG, hippocampal sections were prepared and stained with antibodies to glycogen only (B) and glycogen plus GFAP (C). In (D), hippocampal sections from the mice treated in (A) were first stained with Fluoro-Jade C dye and then with an antibody against the astrocyte marker GFAP. Similar regions and merged photos are shown. (E) Fluoro-Jade C positive neurons in a similar area section of each dentate gyrus of 5 individual mice were numerated. Scale bars, 10μm.

FIG. 8

Knockdown of key enzymes related to polyglucosan metabolism enhances cell apoptosis induced by 2-DG. (A) Gene knockdown efficiency. Scrambled control or each small hairpin (Sh) RNA in combination with Flag-tagged laforin, malin, GPBB, or GS1 was transiently co-transfected into HEK293 cells for 24 hrs. The cells were then lysed and subjected to Western blot using a antibody to Flag. (B) N2A cells were transfected with a plasmid expressing the reporter EGFP and Sh-RNA of indicated genes or scrambled control. Blasticidin-resistant cells were treated with 2.5 mM 2-DG in MEM medium containing 2.5% FBS and 25mM glucose, or left untreated, for 36 hrs. The cells were then stained with Annexin V and nuclear dye DAPI and analyzed by flow cytometry. The EGFP-positive population was gated to evaluate apoptosis (Annexin V-positive) and death (DAPI-positive), respectively. One representative of two separate experiments is shown.