CA150 expression delays striatal cell death in overexpression and knock-in conditions for mutant huntingtin neurotoxicity

Abstract

Transcriptional dysregulation caused by expanded polyglutamines (polyGlns) in huntingtin (htt) may be central to cell-autonomous mechanisms for neuronal cell death in Huntington's disease (HD) pathogenesis. We hypothesized that these mechanisms may involve the dysfunction of the transcriptional regulator CA150, a putative modifier of onset age in HD, because it binds to htt and accumulates in an HD grade-dependent manner in striatal and cortical neurons. Consistently, we report herein that CA150 expression rescues striatal cell death in lentiviral overexpression (rats) and knock-in (mouse cells) conditions for mutant htt neurotoxicity. In both systems, rescue was dependent on the (Gln-Ala)38 repeat normally found in CA150. We excluded the possibility that rescue may be caused by the (Gln-Ala)38 repeat interacting with polyGlns and, by doing so, blocking mutant htt toxicity. In contrast, we found the (Gln-Ala)38 repeat is required for the nuclear restriction of exogenous CA150, suggesting that rescue requires nuclear CA150. Additionally, we found the (Gln-Ala)38 repeat was dispensable for CA150 transcriptional repression ability, suggesting further that CA150 localization is critical to rescue. Finally, rescue was associated with increased neuritic aggregation, with no reduction of nuclear inclusions, suggesting the solubilization and nuclear export of mutant htt. Together, our data indicate that mutant htt may induce CA150 dysfunction in striatal neurons and suggest that the restoration of nuclear protein cooperativity may be neuroprotective.

Figure 1.

CA150 overexpression protects primary striatal cells from mutant htt toxicity. A, Eight weeks after infection, neuronal dysfunction was measured by Neu-N immunohistochemistry. Neu-N expression was restored after htt171-82Q and CA150 infection compared with cells infected with htt171-82Q alone (∗∗p < 0.005; average rescue is 83.3%). Data are mean ± SD (n = 6) expressed as percentages of control, namely htt171-19Q (27.2 ± 9 Neu-N-positive cells/field of view). The mean rescue of htt171-82Q-infected cells by CA150 was 83.3%. B, CA150, and CA150 deleted for the (Gln-Ala)₃₈ repeat (CA150ΔQA), did not alter PGK-driven expression as assessed by GFP immunoblot analysis 6 weeks after infection. C, No noticeable difference of htt transgene expression by CA150 or CA150ΔQA overexpression was observed 3 weeks after infection, as tested by immunoblot for the myc tag of htt.

Figure 2.

The overexpression of CA150 or CA150ΔQA does not produce a striatal pathology in rats. Striatal toxicity was assessed using DARPP-32 and Neu-N immunostaining. Immunostaining for extragenous CA150 is shown (HA epitope). No lesion was observed at 8 or 12 weeks compared with animals injected with htt171-19Q lentiviruses. A, Simple infections analyzed at 8 weeks. B, Double infections analyzed at 12 weeks. C, Striatal sections incubated only with the secondary antibodies (left panel, no primary DARPP-32 antiserum; right panel, no Neu-N antibody) show no immunostaining.

Figure 3.

CA150 attenuates the striatal neuropathology induced by mutant htt overexpression in a (Gln-ala)₃₈-dependent manner. Rat striata injected with the indicated lentiviruses were examined for Neu-N and DARPP-32 expression 8 or 12 weeks after infection. A, Quantification of striatal lesions at 8 and 12 weeks. At 8 weeks, CA150 rescued 77.3% of the DARPP-32 downregulation and 70% of the Neu-N downregulation, whereas CA150ΔQA rescued 37.16% of the DARPP-32 downregulation and 34.1% of the Neu-N downregulation. No effect was observed at 12 weeks. Data are mean ± SD from seven to eight animals (∗∗∗p < 0.0001, ∗∗p < 0.001, and ∗p < 0.05 compared with rats injected with htt171-82Q alone). B, Examples of DARPP-32 immunostaining. The arrowheads indicate lesions. C, Examples of Neu-N immunostaining. The black arrowheads indicate lesions at 8 or 12 weeks. Also shown are high-magnification views (bottom panels) of the lesions at 12 weeks. The blue arrowheads indicate Neu-N-positive cells that survived in the lesion of rats coinjected with htt171-82Q and CA150. D, Examples of immunostaining (HA epitope) revealing the expression of extragenous CA150 species. Shown here are the lesioned areas. The most intense staining was observed in the presence of CA150 rescuing activity. E, Examples of human mutant htt expression revealed by 2B4 (8 weeks) and EM48 (12 weeks) immunostaining. The black arrowheads show the lesions produced by mutant htt.

Figure 4.

CA150 overexpression increases NA formation in the rat striatum. At 8 weeks, htt-containing aggregates were revealed by immunohistochemistry using the antibody 2B4 and nuclei stained using acridine orange. A, Number of NAs, NIs, and NA/NI ratio. The coexpression of htt171-82Q and CA150 led to a 5.65-fold increase in the number of NAs compared with rats injected with htt171-82Q alone (∗∗p < 0.001) without decreasing NIs, which corresponds to a 7.04-fold increase in the NA/NI ratio (∗∗∗p < 0.0001). The coexpression of htt171-82Q and CA150ΔQA did not modify the number of NAs or NIs compared with expression of htt171-82Q alone. Aggregate counts were performed on double-stained sections (2B4, acridine orange) for five fields of view/section at high magnification. Data are mean ± SD from six to eight animals. B, Examples of 2B4 immunostaining at 100× magnification. Top panels show 2B4 staining with gray arrowheads for NIs and blue arrowheads for NAs. Bottom panels show confocal images with 2B4 (red) and nuclear (green) staining. The NIs (white arrowheads) and NAs (yellow arrowheads) are also shown. C, Area and number of objects positive for either exogenous CA150 (HA epitope) or ubiquitin, or both, eight weeks after coinjection of htt171-82Q and CA150. D, Example of HA and ubiquitin coimmunostaining. The white arrowheads show ubiquitinated, HA-positive large objects likely to represent NIs. The blue arrowheads show ubiquitin-positive and HA-negative small objects likely to represent NAs.

Figure 5.

The (Gln-Ala)₃₈ repeat of CA150 influences the nuclear localization of the protein. Eight or 12 weeks after infection of rat striata, the localization of exogenous CA150 and CA150ΔQA was revealed using HA epitope staining. Nuclei were stained using acridine orange. A, Quantification of cytoplasmic and nuclear signals from confocal planes for CA150 compared with CA150ΔQA. Shown are nuclear intensity of HA signals and cytoplasmic area of HA staining. Rats injected with CA150ΔQA alone showed increased cytoplasmic signals and weaker nuclear signals compared with rats injected with CA150 alone, an effect not influenced by htt171-19Q or the time after infection (∗∗∗p < 0.0001). Data are mean ± SD from six fields of view. B, Confocal images of HA epitope staining (red) and nuclear counterstaining (green). The gray arrowheads show nuclear CA150 and white arrowheads show nucleo-cytoplasmic CA150ΔQA.

Figure 6.

Effect of overexpression of CA150 species in 109Q/109Q striatal lines. A, CA150 overexpression reduces cell death induced in 109Q/109Q striatal lines compared with cells transfected with empty vector (∗∗∗p < 0.0001). No difference in cell death was observed in cells overexpressing CA150ΔQA compared with cells transfected with the empty vector. Overexpression of a CA150 fragment containing the QA repeat and 16 aa N-terminal and 76 aa C-terminal to the repeat (CA150_162–331) was toxic for 7Q/7Q cells compared with cells transfected with empty vector (p < 0.0001). Data are mean ± SD of six independent experiments (100–150 cells counted per experiment). B, No noticeable difference of htt expression by CA150 species or (QA)₃₈ overexpression was observed as tested by Western blotting.

Figure 7.

Formation of amyloid-like aggregates from glutamine containing repeat peptides. Monomeric KKQ₄₀KK (▪, □) and KK(QA)₂₀KK (•, ○) were incubated in the 10–20 μm range at 37°C in PBS and amyloid formation monitored by the ThT reaction. Monomer alone (○, □), monomeric Q₄₀ peptide plus 10% by weight of (QA)₂₀ aggregate seeds (•), and monomeric (QA)₂₀ peptide plus 10% by weight of Q₄₀ aggregate seeds (▪) are shown.