Distinct aggregation and cell death patterns among different types of primary neurons induced by mutant huntingtin protein

Aggregation of disease proteins is believed to be a central event in the pathology of polyglutamine diseases, whereas the relationship between aggregation and neuronal death remains controversial. We investigated this question by expressing mutant huntingtin (htt) with a defective adenovirus in different types of neurons prepared from rat cerebral cortex, striatum or cerebellum. The distribution pattern of inclusions is not identical among different types of primary neurons. On day 2 after infection, cytoplasmic inclusions are dominant in cortical and striatal neurons, whereas at day 4 the ratio of nuclear inclusions overtakes that of cytoplasmic inclusions. Meanwhile, nuclear inclusions are always predominantly present in cerebellar neurons. The percentage of inclusion-positive cells is highest in cerebellar neurons, whereas mutant htt induces cell death most remarkably in cortical neurons. As our system uses htt exon 1 protein and thus aggregation occurs independently from cleavage of the full-length htt, our observations indicate that the aggregation process is distinct among different neurons. Most of the neurons containing intracellular (either nuclear or cytoplasmic) aggregates are viable. Our findings suggest that the process of mutant htt aggregation rather than the resulting inclusion body is critical for neuronal cell death.     

Huntingtin associated protein 1 regulates trafficking of the amyloid precursor protein and modulates amyloid beta levels in neurons

J. Neurochem. (2012) 122, 1010-1022. ABSTRACT: Amyloid precursor protein (APP) is involved in the pathogenesis of Alzheimer's disease. It is axonally transported, endocytosed and sorted to different cellular compartments where amyloid beta (Aβ) is produced. However, the mechanism of APP trafficking remains unclear. We present evidence that huntingtin associated protein 1 (HAP1) may reduce Aβ production by regulating APP trafficking to the non-amyloidogenic pathway. HAP1 and APP are highly colocalized in a number of brain regions, with similar distribution patterns in both mouse and human brains. They are associated with each other, the interacting site is the 371-599 of HAP1. APP is more retained in cis-Golgi, trans-Golgi complex, early endosome and ER-Golgi intermediate compartment in HAP1-/- neurons. HAP1 deletion significantly alters APP endocytosis and reduces the re-insertion of APP into the cytoplasmic membrane. Amyloid precursor protein-YFP(APP-YFP) vesicles in HAP1-/- neurons reveal a decreased trafficking rate and an increased number of motionless vesicles. Knock-down of HAP1 protein in cultured cortical neurons of Alzheimer's disease mouse model increases Aβ levels. Our data suggest that HAP1 regulates APP subcellular trafficking to the non-amyloidogenic pathway and may negatively regulate Aβ production in neurons.     

The neurosteroids, allopregnanolone and progesterone, induce autophagy in cultured astrocytes

Recent studies have suggested that neurosteroids such as pregnenolone, progesterone (PG) and their derivatives, have a role in activating autophagy in addition to diverse other functions. In our previous studies, we demonstrated that cellular free Zn(2+) is involved in oxidative stress-induced autophagy and autophagic cell death in astrocytes. In the present study, we examined the possibility that neurosteroids, allopregnanolone (Allo) and PG, also activate autophagy in cultured mouse astrocytes through modulation of intracellular Zn(2+). Exposure of astrocytes to 250 nM Allo or 500 nM PG caused cytosolic vacuoles to appear within a few hours of treatment onset. Live-cell confocal microscopy of astrocytes transfected with red fluorescent protein-conjugated LC3 (RFP-LC3), a marker for autophagic vacuoles (AVs), as well as transmission electron microscopy, revealed that these vacuoles were AVs. In addition, Western blots showed increases in LC3-II levels. Interestingly, mTOR and Akt were concurrently activated, and their blockade further increased LC3-II levels and caused some cell death. These results indicate that co-activation of mTOR and Akt may act to limit neurosteroid-induced autophagy and thus inhibit autophagic cell death. As in other cases of autophagy, cellular Zn(2+) levels increased after treatment with neurosteroids. The neurosteroid-induced increase in LC3-II levels was inhibited by addition of the Zn(2+) chelator TPEN. Both the increase in LC3-II levels and activation of Akt and mTOR by neurosteroids were all mediated by PG receptors, as the effects were blocked by the addition of RU-486, a PG receptor antagonist. Moreover, mutant huntingtin (mHtt) aggregates in GFP-mHttQ74-transfected astrocytes were substantially reduced by neurosteroid treatment, indicating that neurosteroid-induced autophagy may be functional. Present results demonstrate that Allo and PG activate autophagy in astrocytes. Notably, unlike several other autophagy inducers that, in excess, may cause autophagic cell death, Allo and PG are relatively non-toxic, possibly because of concurrent Akt and mTOR activation. Thus, as natural endogenous brain substances, Allo and PG may have a potential as therapeutic agents in neurodegenerative conditions in which abnormal protein aggregates are involved.     

Ubiquitin ligase Hrd1 enhances the degradation and suppresses the toxicity of polyglutamine-expanded huntingtin

E3 ubiquitin ligases catalyze the conjugation of ubiquitin onto proteins, which acts as a signal for targeting proteins for degradation by the proteasome. Hrd1 is an endoplasmic reticulum (ER) membrane-spanning E3 with its catalytic active RING finger facing the cytosol. We speculated that this topology might allow Hrd1 to ubiquitinate misfolded proteins in the cytosol. We tested this idea by using polyglutamine (polyQ)-containing huntingtin (htt) protein as a model substrate. We found that the protein levels of Hrd1 were increased in cells overexpressing the N-terminal fragment of htt containig an expanded polyQ tract (httN). Forced expression of Hrd1 enhanced the degradation of httN in a RING finger-dependent manner, whereas silencing of endogenous Hrd1 expression by RNA interference stabilized httN. Degradation of httN was found to be p97/VCP-dependent, but independent of Ufd1 and Npl4, all of which are thought to form a complex with Hrd1 during ER-associated degradation. Consistent with its role as an E3 for httN, we demonstrate that Hrd1 interacts with and ubiquitinates httN. Subcellular fractionation and confocal microscopy revealed that Hrd1recruits HttN to the ER and co-localizes with juxtanuclear aggregates of httN in cells. Interaction of Hrd1 with httN was found to be independent of the length of the polyglutamine tract. However, httN with expanded polyglutamine tracts appeared to be a preferred substrate for Hrd1. Functionally, we found that Hrd1 protects cells against the httN-induced cell death. These results suggest that Hrd1 is a novel htt-interacting protein that can target pathogenic httN for degradation and is able to protect cells against httN-induced cell death.     

Iron dysregulation in Huntington's disease

Huntington's disease (HD) is one of many neurodegenerative diseases with reported alterations in brain iron homeostasis that may contribute to neuropathogenesis. Iron accumulation in the specific brain areas of neurodegeneration in HD has been proposed based on observations in post‐mortem tissue and magnetic resonance imaging studies. Altered magnetic resonance imaging signal within specific brain regions undergoing neurodegeneration has been consistently reported and interpreted as altered levels of brain iron. Biochemical studies using various techniques to measure iron species in human samples, mouse tissue, or in vitro has generated equivocal data to support such an association. Whether elevated brain iron occurs in HD, plays a significant contributing role in HD pathogenesis, or is a secondary effect remains currently unclear.

     

Diminished hippocalcin expression in Huntington's disease brain does not account for increased striatal neuron vulnerability as assessed in primary neurons

Hippocalcin is a neuronal calcium sensor protein previously implicated in regulating neuronal viability and plasticity. Hippocalcin is the most highly expressed neuronal calcium sensor in the medium spiny striatal output neurons that degenerate selectively in Huntington's disease (HD). We have previously shown that decreased hippocalcin expression occurs in parallel with the onset of disease phenotype in mouse models of HD. Here we show by in situ hybridization histochemistry that hippocalcin RNA is also diminished by 63% in human HD brain. These findings lead us to hypothesize that diminished hippocalcin expression might contribute to striatal neurodegeneration in HD. We tested this hypothesis by assessing whether restoration of hippocalcin expression would decrease striatal neurodegeneration in cellular models of HD comprising primary striatal neurons exposed to mutant huntingtin, the mitochondrial toxin 3-nitropropionic acid or an excitotoxic concentration of glutamate. Counter to our hypothesis, hippocalcin expression did not improve the survival of striatal neurons under these conditions. Likewise, expression of hippocalcin together with interactor proteins including the neuronal apoptosis inhibitory protein did not increase the survival of striatal cells in cellular models of HD. These results indicate that diminished hippocalcin expression does not contribute to HD-related neurodegeneration.     

Modulation of protease expression by the transcription factor Ptx1/PITX regulates protein quality control during aging

1. Download : Download high-res image (208KB)
2. Download : Download full-size image

     

Uninterrupted CAG repeat drives striatum-selective transcriptionopathy and nuclear pathogenesis in human Huntingtin BAC mice

1. Download : Download high-res image (160KB)
2. Download : Download full-size image

     

Polyglutamine expansion in huntingtin increases its insertion into lipid bilayers

An expanded polyglutamine (Q) tract (>37Q) in huntingtin (htt) causes Huntington disease. Htt associates with membranes and polyglutamine expansion in htt may alter membrane function in Huntington disease through a mechanism that is not known. Here we used differential scanning calorimetry to examine the effects of polyQ expansion in htt on its insertion into lipid bilayers. We prepared synthetic lipid vesicles composed of phosphatidylcholine and phosphatidylethanolamine and tested interactions of htt amino acids 1-89 with 20Q, 32Q or 53Q with the vesicles. GST-htt1-89 with 53Q inserted into synthetic lipid vesicles significantly more than GST-htt1-89 with 20Q or 32Q. We speculate that by inserting more into cell membranes, mutant huntingtin could increase disorder within the lipid bilayer and thereby disturb cellular membrane function.