αB-Crystallin Overexpression in Astrocytes Modulates the Phenotype of the BACHD Mouse Model of Huntington's Disease

Abstract

Huntington's disease (HD) is caused by an expanded polyglutamine (polyQ) tract in the huntingtin (htt) protein. The polyQ expansion increases the propensity of htt to aggregate and accumulate, and manipulations that mitigate protein misfolding or facilitate the clearance of misfolded proteins are predicted to slow disease progression in HD models. αB-crystallin (αBc) or HspB5 is a well-characterized member of the small heat shock protein (sHsp) family that reduces mutant htt (mhtt) aggregation and toxicity in vitro and in Drosophila models of HD. Here, we determined if overexpressing αBc in vivo modulates aggregation and delays the onset and progression of disease in a full-length model of HD, BACHD mice. Expression of sHsps in neurodegenerative disease predominantly occurs in non-neuronal cells, and in the brain, αBc is mainly found in astrocytes and oligodendrocytes. Here, we show that directed αBc overexpression in astrocytes improves motor performance in rotarod and balance beam tests and improves cognitive function in the BACHD mice. Improvement in behavioral deficits correlated with mitigation of neuropathological features commonly observed in HD. Interestingly, astrocytic αBc overexpression was neuroprotective against neuronal cell loss in BACHD brains, suggesting αBc might be acting in a non-cell-autonomous manner. At the protein level, αBc decreased the level of soluble mhtt and decreased the size of mhtt inclusions in BACHD brain. Our results support a model in which elevating astrocytic αBc confers neuroprotection through a potential non-cell-autonomous pathway that modulates mhtt aggregation and protein levels.

Figure 1.

αBc levels decrease over time in BACHD mice. (A, B) BACHD (n = 5) and WT (n = 5) littermates were sacrificed at 3, 6, 9 and 12 months of age. Protein levels of αBc were determined by western blot analysis with whole-brain homogenates of BACHD and WT mice. (A) Immunoblots showing αBc protein level is decreased in BACHD, in an age-dependent manner, specifically after 6 months (when the motor symptoms are more robust in this mice model). (B) Respective quantifications of protein levels of αBc normalized to anti-β-tubulin (to show relative loading levels). Error bars represent standard errors of the mean (means ± SEM). Data were tested for statistically significant differences using t-tests (*P < 0.05; **P < 0.01).

Figure 2.

αBc overexpression improves behavioral readouts in BACHD mice. Behavioral readouts of disease progression in BACHD were measured at 3, 7, 10 and 13 months of age using rotarod and balance beam. Data at each time point were tested for statistically significant differences using two-way ANOVA with multiple comparisons tests corrected by Tukey post-hoc tests. At 3 months of age, group sizes were as follows: WT (n = 20), Cryab Tg (n = 19), BACHD (n = 14), DTg (n = 17). Values are means ± SEM. *P < 0.05, **P < 0.01 or ***P < 0.001 for main effect of the BACHD transgene by two-way ANOVA. (A) Rotarod. BACHD mice fall off an accelerating rotarod sooner than WT littermates and Cryab tg mice (significant main effect of the BACHD transgene). αBc overexpression partially rescues the BACHD motor impairment on DTg mice, and this is significant at all different time points at 3, 7, 10 and 13 months. (B) BACHD mice take significantly more time to cross the balance beam than WT littermates and Cryab tg mice. (C, D) The BACHD group also showed significant higher number of slips and falls during this task. αBc overexpression reduces the number of slips (C) and falls (D) in BACHD mice during the balance beam test. αBc overexpression does only change the overall latency to cross in DTg mice at 7 months, however, reduces significantly the number of slips (7, 10 and 13 months) and falls (10 and 13 months) of BACHD mice in the balance beam transversal assay.

Figure 3.

αBc overexpression improves BACHD strategy shifting in the water-T maze test. (A) Water-T maze test diagram. BACHD mice were tested in the swimming T-maze with reversal. At 13 months of age, group sizes were as follows: WT (n = 18), Cryab Tg (n = 17), BACHD (n = 10), DTg (n = 15). Symptomatic BACHD show cognitive deficits in strategy shifting, and αBc overexpression rescues this impairment. (B) During the normal phase of the test, the four distinct groups learned to swim to the platform, and no differences were observed in the percentage of correct arm entries, having the four groups reaching ≥80% of correct entries, and (D) in the time to reach the platform. (C) During the reversal phase, when the location of the platform was switched, BACHD mice showed a significant lower number of correct entries in the reversal arm and (E) required significantly more time to find the platform than the other groups from Day 3 to Day 5. Speed (cm/s) and distance (cm) to locate the platform in either training and reversal phases were not different among the four genotypes during this test performance. A two-way ANOVA was performed, with multiple comparisons tests, corrected by Tukey test. *P < 0.05.

Figure 4.

αBc overexpression decreases soluble levels of mhtt and reduces the size of S830 inclusion bodies in BACHD mice. (A) Representative western blot of WT, Cryab Tg, DTg and BACHD (n = 4) cortical lysates with 4H7H7 and anti-β-tubulin as a loading control and (B) subsequent quantification of signal intensity. Values are based on the mean of three independent 4H7H7 blots of four mice per group compared across different blots. All value were first normalized for input using the anti-β-tubulin controls. (C) Immunohistochemistry with polyoclonal S830 revealed prominent immunoreactive inclusion bodies in BACHD and DTg mice cortex and striatum but not in the WT and Cryab Tg littermate controls (data not shown). S830 immunohistochemistry in (a, c) BACHD and (b, d) DTg representative mouse brains with average levels of aggregate counts and density for the two groups: (a, b) M1/M2 cortex layer III and (c, d) striatum/medial caudate/putamen, using Renyi/Entropy threshold and >1 µm diameter. The scale bar represents 100 µm, and the field size is 205 μm × 150 μm. (D) The number of mhtt nuclear inclusions >1 µm were significantly reduced in the DTg when compared with BACHD mice. Values are means ± SD; BACHD (n = 9) and DTg (N = 6), *P < 0.05 by Student's unpaired t-test.

Figure 5.

αBc overexpression rescues the neurodegeneration caused by expression of full-length mhtt at 13 months of age in BACHD. Measurement of striatal DARPP-32-positive cells and striatal cortical NeuN-positive cells in a: WT (n = 5), b: Cryab Tg (n = 5), c: BACHD (n = 8), d: DTg (n = 7). (A) Representative confocal micrographs of DARPP-32 immunostaining and (B) DARPP-32 positive cells quantification plots. (C, E) Representative confocal micrographs of NeuN immunostaining in the striatum/medial caudate/putamen (C) and M1/M2 cortex (E), followed by NeuN-positive cells quantification plots (D, F). Each field is 600 × 600 µm, and the scale bar represents 20 µm. Values represent means ± SEM. *P < 0.05, one-way ANOVA with multiple Tukey multiple comparisons correction.