Distinct roles of Bdnf I and Bdnf IV transcript variant expression in hippocampal neurons.

Brain-derived neurotrophic factor (Bdnf) plays a critical role in brain development, dendritic growth, synaptic plasticity, as well as learning and memory. The rodent Bdnf gene contains nine 5' non-coding exons (I-IXa), which are spliced to a common 3' coding exon (IX). Transcription of individual Bdnf variants, which all encode the same BDNF protein, is initiated at unique promoters upstream of each non-coding exon, enabling precise spatiotemporal and activity-dependent regulation of Bdnf expression. Although prior evidence suggests that Bdnf transcripts containing exon I (Bdnf I) or exon IV (Bdnf IV) are uniquely regulated by neuronal activity, the functional significance of different Bdnf transcript variants remains unclear. To investigate functional roles of activity-dependent Bdnf I and IV transcripts, we used a CRISPR activation system in which catalytically dead Cas9 fused to a transcriptional activator (VPR) is targeted to individual Bdnf promoters with single guide RNAs, resulting in transcript-specific Bdnf upregulation. Bdnf I upregulation is associated with gene expression changes linked to dendritic growth, while Bdnf IV upregulation is associated with genes that regulate protein catabolism. Upregulation of Bdnf I, but not Bdnf IV, increased mushroom spine density, volume, length, and head diameter, and also produced more complex dendritic arbors in cultured rat hippocampal neurons. In contrast, upregulation of Bdnf IV, but not Bdnf I, in the rat hippocampus attenuated contextual fear expression. Our data suggest that while Bdnf I and IV are both activity-dependent, BDNF produced from these promoters may serve unique cellular, synaptic, and behavioral functions.

Rho Kinase Inhibition as a Therapeutic for Progressive Supranuclear Palsy and Corticobasal Degeneration.

Progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD) are neurodegenerative four-repeat tauopathies with no cure. Mitigating pathogenic tau levels is a rational strategy for tauopathy treatment, but therapeutic targets with clinically available drugs are lacking. Here, we report that protein levels of the Rho-associated protein kinases (ROCK1 and ROCK2), p70 S6 kinase (S6K), and mammalian target of rapamycin (mTOR) were increased in PSP and CBD brains. RNAi depletion of ROCK1 or ROCK2 reduced tau mRNA and protein level in human neuroblastoma cells. However, additional phenotypes were observed under ROCK2 knockdown, including decreased S6K and phosphorylated mTOR levels. Pharmacologic inhibition of Rho kinases in neurons diminished detergent-soluble and -insoluble tau through a combination of autophagy enhancement and tau mRNA reduction. Fasudil, a clinically approved ROCK inhibitor, suppressed rough eye phenotype and mitigated pathogenic tau levels by inducing autophagic pathways in a Drosophila model of tauopathy. Collectively, these findings highlight the Rho kinases as rational therapeutic targets to combat tau accumulation in PSP and CBD. SIGNIFICANCE STATEMENT: Studies of progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD) suggest that mitigating pathogenic tau levels is a rational strategy for tauopathy treatment. In this report, the Rho-associated protein kinases (ROCK1 and ROCK2) are identified as novel drug targets for PSP and CBD. We show that elevated insoluble tau levels are associated with increased ROCK1 and ROCK2 in PSP and CBD brains, whereas experiments in cellular and animal models identify pharmacologic inhibition of ROCKs as a mechanism-based approach to reduce tau levels. Our study correlates bona fide changes in PSP and CBD brains with cellular models, identifies drug targets, and tests the therapeutic in vivo.

Rho-associated protein kinase 1 (ROCK1) is increased in Alzheimer's disease and ROCK1 depletion reduces amyloid-ß levels in brain.

Alzheimer's disease (AD) is the leading cause of dementia and mitigating amyloid-ß (Aß) levels may serve as a rational therapeutic avenue to slow AD progression. Pharmacologic inhibition of the Rho-associated protein kinases (ROCK1 and ROCK2) is proposed to curb Aß levels, and mechanisms that underlie ROCK2's effects on Aß production are defined. How ROCK1 affects Aß generation remains a critical barrier. Here, we report that ROCK1 protein levels were elevated in mild cognitive impairment due to AD (MCI) and AD brains compared to controls. Aß42 oligomers marginally increased ROCK1 and ROCK2 protein levels in neurons but strongly induced phosphorylation of Lim kinase 1 (LIMK1), suggesting that Aß42 activates ROCKs. RNAi depletion of ROCK1 or ROCK2 suppressed endogenous Aß40 production in neurons, and Aß40 levels were reduced in brains of ROCK1 heterozygous knock-out mice compared to wild-type littermate controls. ROCK1 knockdown decreased amyloid precursor protein (APP), and treatment with bafilomycin accumulated APP levels in neurons depleted of ROCK1. These observations suggest that reduction of ROCK1 diminishes Aß levels by enhancing APP protein degradation. Collectively, these findings support the hypothesis that both ROCK1 and ROCK2 are therapeutic targets to combat Aß production in AD. Mitigating amyloid-ß (Aß) levels is a rational strategy for Alzheimer's disease (AD) treatment, however, therapeutic targets with clinically available drugs are lacking. We hypothesize that Aß accumulation in mild cognitive impairment because of AD (MCI) and AD activates the RhoA/ROCK pathway which in turn fuels production of Aß. Escalation of this cycle over the course of many years may contribute to the buildup of amyloid pathology in MCI and/or AD.

Distinct roles of Bdnf I and Bdnf IV transcript variant expression in hippocampal neurons.

Brain-derived neurotrophic factor (Bdnf) plays a critical role in brain development, dendritic growth, synaptic plasticity, as well as learning and memory. The rodent Bdnf gene contains nine 5' non-coding exons (I-IXa), which are spliced to a common 3' coding exon (IX). Transcription of individual Bdnf variants, which all encode the same BDNF protein, is initiated at unique promoters upstream of each non-coding exon, enabling precise spatiotemporal and activity-dependent regulation of Bdnf expression. Although prior evidence suggests that Bdnf transcripts containing exon I (Bdnf I) or exon IV (Bdnf IV) are uniquely regulated by neuronal activity, the functional significance of different Bdnf transcript variants remains unclear. To investigate functional roles of activity-dependent Bdnf I and IV transcripts, we used a CRISPR activation (CRISPRa) system in which catalytically-dead Cas9 (dCas9) fused to a transcriptional activator (VPR) is targeted to individual Bdnf promoters with single guide RNAs (sgRNAs), resulting in transcript-specific Bdnf upregulation. Bdnf I upregulation is associated with gene expression changes linked to dendritic growth, while Bdnf IV upregulation is associated with genes that regulate protein catabolism. Upregulation of Bdnf I, but not Bdnf IV, increased mushroom spine density, volume, length, and head diameter, and also produced more complex dendritic arbors in cultured rat hippocampal neurons. In contrast, upregulation of Bdnf IV, but not Bdnf I, in the rat hippocampus attenuated contextual fear expression. Our data suggest that while Bdnf I and IV are both activity-dependent, BDNF produced from these promoters may serve unique cellular, synaptic, and behavioral functions.

The SETD6 Methyltransferase Plays an Essential Role in Hippocampus-Dependent Memory Formation.

BACKGROUND: Epigenetic mechanisms are critical for hippocampus-dependent memory formation. Building on previous studies that implicate the N-lysine methyltransferase SETD6 in the activation of nuclear factor-κB RELA (also known as transcription factor p65) as an epigenetic recruiter, we hypothesized that SETD6 is a key player in the epigenetic control of long-term memory. METHODS: Using a series of molecular, biochemical, imaging, electrophysiological, and behavioral experiments, we interrogated the effects of short interfering RNA-mediated knockdown of Setd6 in the rat dorsal hippocampus during memory consolidation. RESULTS: Our findings demonstrate that SETD6 is necessary for memory-related nuclear factor-κB RELA methylation at lysine 310 and associated increases in H3K9me2 (histone H3 lysine 9 dimethylation) in the dorsal hippocampus and that SETD6 knockdown interferes with memory consolidation, alters gene expression patterns, and disrupts spine morphology. CONCLUSIONS: Together, these findings suggest that SETD6 plays a critical role in memory formation and may act as an upstream initiator of H3K9me2 changes in the hippocampus during memory consolidation.

Fasudil or genetic depletion of ROCK1 or ROCK2 induces anxiety-like behaviors.

Twenty-nine protein kinase inhibitors have been used to treat human diseases. Out of these, two are Rho-associated protein kinase (ROCK) 1 and 2 inhibitors. The ROCKs heavily influence neuronal architecture and structural plasticity, and ROCKs are putative drug targets for various brain disorders. While the pan-ROCK inhibitor Fasudil has been clinically approved to treat hypertension, heart failure, glaucoma, spinal cord injury, and stroke, a barrier to progress on this therapeutic avenue is the lack of experimental comparisons between pharmacologic and genetic manipulation of ROCKs. Our study begins to address this question using parallel approaches to study behavior in mice that were treated with Fasudil or were heterozygous for ROCK1 or ROCK2. Adult mice treated with Fasudil for thirty days displayed reduced time spent in the open arms of the elevated plus maze, whereas activity in the open field was more analogous to mock-treated animals. Both male and female adult ROCK1+/- and ROCK2+/- mice exhibited reduced time spent in open arms of the elevated plus maze compared to littermate controls. However, ROCK1 or ROCK2 heterozygosity did not alter performance in the open field or Y-maze. These results indicate that chronic treatment with Fasudil induces anxiety-like behaviors that are likely the consequence of ROCK1 and/or ROCK2 inhibition. Our findings may have implications for several ongoing clinical trials using Fasudil or other ROCK-based therapeutics.

Pharmacologic inhibition of LIMK1 provides dendritic spine resilience against ß-amyloid.

Alzheimer's disease (AD) therapies predominantly focus on ß-amyloid (Aß), but Aß effects may be maximal before clinical symptoms appear. Downstream of Aß, dendritic spine loss correlates most strongly with cognitive decline in AD. Rho-associated kinases (ROCK1 and ROCK2) regulate the actin cytoskeleton, and ROCK1 and ROCK2 protein abundances are increased in early AD. Here, we found that the increased abundance of ROCK1 in cultured primary rat hippocampal neurons reduced dendritic spine length through a myosin-based pathway, whereas the increased abundance of ROCK2 induced spine loss through the serine and threonine kinase LIMK1. Aß42 oligomers can activate ROCKs. Here, using static imaging studies combined with multielectrode array analyses, we found that the ROCK2-LIMK1 pathway mediated Aß42-induced spine degeneration and neuronal hyperexcitability. Live-cell microscopy revealed that pharmacologic inhibition of LIMK1 rendered dendritic spines resilient to Aß42 oligomers. Treatment of hAPP mice with a LIMK1 inhibitor rescued Aß-induced hippocampal spine loss and morphologic aberrations. Our data suggest that therapeutically targeting LIMK1 may provide dendritic spine resilience to Aß and therefore may benefit cognitively normal patients that are at high risk for developing dementia.

Distinct and complementary functions of rho kinase isoforms ROCK1 and ROCK2 in prefrontal cortex structural plasticity.

Rho-associated protein kinases (ROCK) 1 and 2 are attractive drug targets for a range of neurologic disorders; however, a critical barrier to ROCK-based therapeutics is ambiguity over whether there are isoform-specific roles for ROCKs in neuronal structural plasticity. Here, we used a genetics approach to address this long-standing question by analyzing both male and female adult ROCK1+/- and ROCK2+/- mice compared to littermate controls. Individual pyramidal neurons in the medial prefrontal cortex (mPFC) were targeted for iontophoretic microinjection of fluorescent dye, followed by high-resolution confocal microscopy and neuronal 3D reconstructions for morphometry analysis. Increased apical and basal dendritic length and intersections were observed in ROCK1+/- but not ROCK2+/- mice. Although dendritic spine densities were comparable among genotypes, apical spine length was decreased in ROCK1+/- but increased in ROCK2+/- mice. Spine head and neck diameter were reduced similarly in ROCK1+/- and ROCK2+/- mice; however, certain spine morphologic subclasses were more affected than others in a genotype-dependent manner. Biochemical analyses of ROCK substrates in synaptic fractions revealed that phosphorylation of LIM kinase and cofilin were reduced in ROCK1+/- and ROCK2+/- mice, while phosphorylation of myosin light chain was decreased exclusively in ROCK1+/- mice. Collectively, these observations implicate ROCK1 as a novel regulatory factor of neuronal dendritic structure and detail distinct and complementary roles of ROCKs in mPFC dendritic spine structure.

α-Synuclein fibril-induced paradoxical structural and functional defects in hippocampal neurons.

Neuronal inclusions composed of α-synuclein (α-syn) characterize Parkinson's Disease (PD) and Dementia with Lewy bodies (DLB). Cognitive dysfunction defines DLB, and up to 80% of PD patients develop dementia. α-Syn inclusions are abundant in the hippocampus, yet functional consequences are unclear. To determine if pathologic α-syn causes neuronal defects, we induced endogenous α-syn to form inclusions resembling those found in diseased brains by treating hippocampal neurons with α-syn fibrils. At seven days after adding fibrils, α-syn inclusions are abundant in axons, but there is no cell death at this time point, allowing us to assess for potential alterations in neuronal function that are not caused by neuron death. We found that exposure of neurons to fibrils caused a significant reduction in mushroom spine densities, adding to the growing body of literature showing that altered spine morphology is a major pathologic phenotype in synucleinopathies. The reduction in spine densities occurred only in wild type neurons and not in neurons from α-syn knockout mice, suggesting that the changes in spine morphology result from fibril-induced corruption of endogenously expressed α-syn. Paradoxically, reduced postsynaptic spine density was accompanied by increased frequency of miniature excitatory postsynaptic currents (EPSCs) and presynaptic docked vesicles, suggesting enhanced presynaptic function. Action-potential dependent activity was unchanged, suggesting compensatory mechanisms responding to synaptic defects. Although activity at the level of the synapse was unchanged, neurons exposed to α-syn fibrils, showed reduced frequency and amplitudes of spontaneous Ca2+ transients. These findings open areas of research to determine the mechanisms that alter neuronal function in brain regions critical for cognition at time points before neuron death.

Mutant human torsinA, responsible for early-onset dystonia, dominantly suppresses GTPCH expression, dopamine levels and locomotion in Drosophila melanogaster.

Dystonia represents the third most common movement disorder in humans with over 20 genetic loci identified. TOR1A (DYT1), the gene responsible for the most common primary hereditary dystonia, encodes torsinA, an AAA ATPase family protein. Most cases of DYT1 dystonia are caused by a 3 bp (ΔGAG) deletion that results in the loss of a glutamic acid residue (ΔE302/303) in the carboxyl terminal region of torsinA. This torsinAΔE mutant protein has been speculated to act in a dominant-negative manner to decrease activity of wild type torsinA. Drosophila melanogaster has a single torsin-related gene, dtorsin. Null mutants of dtorsin exhibited locomotion defects in third instar larvae. Levels of dopamine and GTP cyclohydrolase (GTPCH) proteins were severely reduced in dtorsin-null brains. Further, the locomotion defect was rescued by the expression of human torsinA or feeding with dopamine. Here, we demonstrate that human torsinAΔE dominantly inhibited locomotion in larvae and adults when expressed in neurons using a pan-neuronal promoter Elav. Dopamine and tetrahydrobiopterin (BH4) levels were significantly reduced in larval brains and the expression level of GTPCH protein was severely impaired in adult and larval brains. When human torsinA and torsinAΔE were co-expressed in neurons in dtorsin-null larvae and adults, the locomotion rates and the expression levels of GTPCH protein were severely reduced. These results support the hypothesis that torsinAΔE inhibits wild type torsinA activity. Similarly, neuronal expression of a Drosophila DtorsinΔE equivalent mutation dominantly inhibited larval locomotion and GTPCH protein expression. These results indicate that both torsinAΔE and DtorsinΔE act in a dominant-negative manner. We also demonstrate that Dtorsin regulates GTPCH expression at the post-transcriptional level. This Drosophila model of DYT1 dystonia provides an important tool for studying the differences in the molecular function between the wild type and the mutant torsin proteins.