p62/SQSTM1 differentially removes the toxic mutant androgen receptor via autophagy and inclusion formation in a spinal and bulbar muscular atrophy mouse model

Abstract

Polyglutamine (polyQ) diseases are inherited neurodegenerative disorders that are caused by the expansion of trinucleotide CAG repeats in the causative genes. Spinal and bulbar muscular atrophy (SBMA) is an inherited motor neuron disease that is caused by the expansion of a polyQ tract within the androgen receptor (AR). p62 is a ubiquitin- and light-chain 3-binding protein that is known to regulate the degradation of targeted proteins via autophagy and inclusion formation. In this study, we examined the effects of p62 depletion and overexpression on cultured cells and in a transgenic mouse model that overexpressed the mutant AR. Here, we demonstrate that depletion of p62 significantly exacerbated motor phenotypes and the neuropathological outcome, whereas overexpression of p62 protected against mutant AR toxicity in SBMA mice. Depletion of p62 significantly increased the levels of monomeric mutant AR and mutant AR protein complexes in an SBMA mouse model via the impairment of autophagic degradation. In addition, p62 overexpression improved SBMA mouse phenotypes by inducing cytoprotective inclusion formation. Our results demonstrate that p62 provides two different therapeutic targets in SBMA pathogenesis: (1) autophagy-dependent degradation and (2) benevolent inclusion formation of the mutant AR.

Figure 1.

AR is degraded in a p62-dependent autophagic pathway. A, PC12 cells that express wild-type (10Q) and mutant (112Q) AR were treated with bafilomycin A1 (20 nm) and/or lactacystin (2.5 μm) for 16 h after DHT treatment for 3 or 7 d. Immunoblot analysis was performed. B, Neuro2A cells that were transfected with wild-type (24Q) and mutant (97Q) ARs were treated with bafilomycin A1 (20 nm) and lactacystin (2.5 μm) for 16 h. C, Immunofluorescence for AR10Q and 112Q (green), p62 (blue), LC3 (red), and an overlay (white) of the three signals in PC12 cells for no treatment (control) and a treatment of bafilomycin A1 (20 nm). Scale bars, 10 μm. These experiments were repeated in five sets of cells, and equivalent results were obtained. All of the values are expressed as the means ± SEM (n = 5). *p < 0.05; **p < 0.01.

Figure 2.

p62 interacts with AR. A, Neuro2A cells transfected with wild-type (24Q) and mutant (97Q) AR were treated with bafilomycin A1 (20 nm) for 16 h. Immunoprecipitation was then performed with non-immune rabbit IgG (NI) or an antibody for endogenous p62 (I). B, PB1 mutant p62 (M0) or wild-type p62 (Wt) and AR were cotransfected in Neuro2A cells, followed by immunoprecipitation with FLAG M2 agarose beads. Mutant p62 coprecipitated with AR, whereas the control did not show an interaction. C, PB1 mutant p62 (M0) and AR were cotransfected in Neuro2A cells. Immunoprecipitation was then performed using FLAG M2 agarose beads. D, p62 with an HA tag and ARs were cotransfected in Neuro2A cells. Immunoprecipitation was then performed using an HA antibody. E, A schematic representation of p62 deletion mutants fused with an N-terminal 3xFLAG generated by mutagenesis. F, AR–97Q and p62 mutants (M0–M7) were cotransfected, and immunoprecipitation was performed. G, Mutant p62 (M406V) or wild-type p62 (Wt) and AR–97Q were cotransfected in Neuro2A cells, followed by immunoprecipitation using FLAG M2 agarose beads. IPs, Immunoprecipitations; C, control; K7A D69A, Phox and Bem1p (PB1) domain with mutations for each amino acid; ZZ, ZZ-type zinc finger; LRS, LC3 recognition sequence; Ub, ubiquitin.

Figure 3.

Depletion of p62 promotes the accumulation of AR. A, PC12 cells expressing wild-type (10Q) and mutant (112Q) AR were transfected with either control or p62 siRNA. The cells were treated with DHT for 3 or 7 d. B, Neuro2A cells transfected with wild-type (24Q) and mutant (97Q) AR were cotransfected with control or p62 siRNA. C, Pulse-chase analysis of two forms of AR in PC12 cells. Data from one representative experiment for wild-type and mutant AR. D, Pulse-chase assessment of half-life of the wild-type (left) and mutant AR (right). The percentages of wild-type (10Q) and mutant (112Q) remaining in the presence (●) and absence (○) of p62 siRNA are indicated. E, Real-time RT-PCR for wild-type (10Q) and mutant (112Q) AR mRNA normalized to GAPDH levels in PC12 cells. F, Real-time RT-PCR for wild-type (24Q) and mutant (97Q) AR mRNA normalized to GAPDH levels in Neuro2A cells. These experiments were repeated in five sets of cells, and equivalent results were obtained. All of the values are expressed as the means ± SEM (n = 5). *p < 0.05; **p < 0.01.

Figure 4.

Depletion of p62 impairs behavioral and visible phenotypes in male AR–97Q mice. A–D, p62 immunohistochemistry in the spinal anterior horn (A, B) and skeletal muscle (C, D) of 25-week-old AR–97Q/p62^+/+ (A, C) and 13-week-old AR–97Q/p62^−/− (B, D) mice counterstained with Mayer's hematoxylin. p62 immunoreactivity was localized to the nuclei and cytoplasm, with NIs (arrow) in the anterior horn cells and skeletal muscle. Scale bars, 20 μm. E, Western blotting analysis of p62 expression in the total spinal cord and muscle protein lysates from the indicated mice immunolabeled with antibodies against p62. F–J, Rotarod task (F), cage activity (G), grip strength (H), body weight (I), and survival rate (J) of the AR–97Q/p62^+/+(●; n = 20), AR–97Q/p62^+/−(▾; n = 28), and AR–97Q/p62^−/− (■; n = 20) mice. Although none of the parameters tested at 15 weeks revealed significant differences between AR–97Q/p62^+/− and AR–97Q/p62^+/+ mice, the AR–97Q/p62^+/− mice performed more poorly than the AR–97Q/p62^+/+ mice in all of the parameters. K, Footprints of representative 13-week-old AR–97Q/p62^+/+ and AR–97Q/p62^−/− mice. The front paws are indicated in blue, and the hindpaws are indicated in red. Values are expressed as the means ± SEM. *p < 0.05.

Figure 5.

Depletion of p62 induces the accumulation of AR in an SBMA mouse model. A–F, PolyQ immunohistochemistry (1C2) in the spinal anterior horn (A–C) and muscle (D–F) of 13-week-old AR–97Q/p62^+/+, AR–97Q/p62^+/−, and AR–97Q/p62^−/− mice. Scale bars, 50 μm. G, H, Quantitative assessment of 1C2 staining in the spinal ventral horn (G) and muscle (H). Bars represent the density of 1C2-positive cells in AR–97Q/p62^+/+, AR–97Q/p62^+/−, and AR–97Q/p62^−/− mice. The results are expressed as the means ± SEM (n = 6). I, J, Western blotting analysis of the total tissue homogenates from the spinal cord (I) and muscle (J) of AR–24Q and AR–97Q mice (13 weeks of age, n = 6), probed with anti-AR. Values are expressed as the means ± SEM (n = 6). K, L, Immunohistochemical staining with anti-GFAP antibody in the spinal anterior horn. Scale bars, 30 μm. M, Western blotting analysis of the total tissue homogenates from the spinal cord of AR–97Q mice (13 weeks of age, n = 6) probed with anti-GFAP. N, O, Hematoxylin and eosin staining of the muscles. P, The gastrocnemius muscles from AR–97Q/p62^+/+ and AR–97Q/p62^−/− mice were dissected and weighed. Q, Western blotting analysis of the total tissue homogenates from the spinal cord and muscle of AR–24Q and AR–97Q mice (13 weeks of age) probed with anti-ubiquitin (Ub). Scale bars, 50 μm. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 6.

p62 overexpression ameliorates behavioral and visible phenotypes in male AR–97Q mice. A, A schematic view of the transgene construct. The microinjected fragment was composed of a cytomegalovirus enhancer (E), a chicken β-actin promoter (Pro), full-length human p62 with an HA tag, and a rabbit β-globin polyadenylation signal sequence (polyA). B, Western blotting analysis of total spinal cord and muscle protein lysates from wild-type (Wt) and p62^(tg/+) mice immunolabeled with antibodies against p62 and HA. C, D, HA immunohistochemistry in the spinal anterior horn (C) and skeletal muscle (D) of 13-week-old wild-type (Wt) and p62^(tg/+) mice counterstained with Mayer's hematoxylin. p62 immunoreactivity localized to the nuclei and cytoplasms of anterior horn cells (C) and skeletal muscle (D). HA staining was absent in the wild-type mice. Scale bars, 50 μm. E–I, Rotarod task (E), cage activity (F), grip strength (G), body weight (H), and survival rate (I) of AR–97Q/p62^(+/+) (●; n = 25) and AR–97Q/p62^(tg/+)(▾; n = 21) mice. J, Footprints of representative 25-week-old AR–97Q/p62^(+/+) and AR–97Q/p62^(tg/+) mice. The front paws are indicated in blue, and the hindpaws are indicated in red. Values are expressed as the means ± SEM. *p < 0.01.

Figure 7.

p62 promotes inclusion body formation of ARs in male AR–97Q mice. A, Immunohistochemical staining with 1C2 antibody revealed DNS (arrowhead) in AR–97Q/p62^(+/+) mice and many NIs (arrow) in the AR–97Q/p62^(tg/+) mice in the spinal anterior horn and skeletal muscle at 25 weeks of age. The region of interest for the DNS is defined as the ROI (inset; open square). Scale bars, 30 μm. B, The ratio of the number of cells with NIs to the total number of IC2-positive cells. C, The ratio of the number of cells with DNS to the total number of IC2-positive cells. D, The R.S.I. of the DNS was significantly reduced in cells with NIs compared with those without NIs in spinal anterior horn neurons and skeletal muscle. E, F, Double-immunofluorescence staining with 1C2 (red) and anti-HA (green) antibodies in the spinal anterior horn (E) and skeletal muscle (F) of AR–97Q/p62^(tg/+) mice revealed the complete colocalization of HA-tagged p62 and mutant AR in NIs (arrow), but no colocalization was observed in the DNS (arrowhead). Scale bars, 20 μm. G, H, p62 immunohistochemistry in the spinal anterior horn neurons (G) and skeletal muscle (H) of AR–97Q/p62^(tg/+) mice. Arrows indicate NIs. Scale bars: G, 30 μm; H, 50 μm. Values are expressed as the means ± SEM (n = 6). I, 1C2 (red) and anti-HA (green) double immunofluorescence in the spinal cord of 25-week-old AR–97Q/p62^(tg/+) mice. *p < 0.01; **p < 0.001.

Figure 8.

Colocalization of nuclear-localized p62 with mutant AR. A–C, 1C2 (red) and anti-HA (green) double immunofluorescence in the spinal cord (A) and skeletal muscle (B) of 16-week-old AR–97Q mice and in the spinal anterior horn cells (C) of SBMA patients. Double-immunofluorescence staining revealed p62 and mutant AR colocalization in NIs (shown in yellow, arrow), but no colocalization was observed in the DNS (arrowhead) in AR–97Q mice and SBMA patients. Scale bars, 20 μm. D, Immunohistochemistry for the anti-p62 antibody in SBMA patients. p62 was localized in the NIs. Scale bar, 10 μm.

Figure 9.

An increase in p62 reduces mutant AR expression. A, B, Western blotting analysis of total tissue homogenates from the spinal cord (A) and muscle (B) of AR–24Q and AR–97Q mice (25 weeks of age) probed with anti-AR. Values are expressed as the means ± SEM (n = 5). C, D, Western blotting analysis of the pellet lysed in 8 m urea solution from the spinal cord (C) and muscle (D) of AR–24Q and AR–97Q mice (25 weeks of age) probed with anti-AR. Values are expressed as the means ± SEM (n = 5). E, Real-time RT-PCR of wild-type and mutant AR mRNA normalized to GAPDH levels. F, G, Immunohistochemical staining with anti-GFAP antibody in the spinal anterior horn. Scale bars, 30 μm. H, Western blotting analysis of the total tissue homogenates from the spinal cord of AR–97Q mice (25 weeks of age, n = 5) probed with anti-GFAP. I, J, Hematoxylin and eosin staining of the muscle. Scale bars, 50 μm. K, The gastrocnemius muscles from AR–97Q/p62^(+/+) and AR–97Q/p62^(tg/+) mice were dissected and weighed. L, Western blotting analysis of the total tissue homogenates from the spinal cord and muscle of AR–24Q and AR–97Q mice (25 weeks of age) probed with anti-ubiquitin (Ub). M, Ubiquitin immunohistochemistry revealed NIs (arrow) in the spinal anterior horn and muscle of 25-week-old AR–97Q/p62^(+/+) and AR–97Q/p62^(tg/+) mice. The inset shows a magnified image in the square box. Scale bars, 50 μm. N, Quantitative assessment of ubiquitin-positive NIs in the spinal ventral horn and muscle. Bars represent the density of ubiquitin-positive NIs in AR–97Q/p62^(+/+) and AR–97Q/p62^(tg/+) mice. The results are expressed as the means ± SEM (n = 6). *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 10.

LC3-II expression in AR–97Q/p62^(tg/+) mice. A, Western blotting analysis of total spinal cord protein lysates from the AR–97Q/p62^(+/+) and AR–97Q/p62^(tg/+) mice immunolabeled with antibodies against LC3. B, Quantitative analysis for the amounts of LC3-II. LC3-II levels were slightly elevated in the AR–97Q/p62^(tg/+) mice. This experiment was repeated in four sets of mice with equivalent results. All of the values are expressed as the means ± SEM (n = 5). *p < 0.05; **p < 0.01.

Figure 11.

p62 overexpression does not affect AR expression. A, Full-length wild-type (24Q) or mutant (97Q) AR and HA-tagged p62 (pCAGGS–p62–HA) were cotransfected into Neuro2A cells for 48 h. B, Full-length wild-type (24Q) or mutant (97Q) AR and FLAG-tagged p62 (pIRESpuro3–3XFLAG–p62) were cotransfected into NSC34 cells for 48 h. All of the cells were cultured in DMEM/10% FCS at 37°C and with 5% CO₂. Immunoblots revealed that the expression of AR was similar in the presence or absence of HA–p62 or FLAG–p62.