The Alzheimer's disease risk gene BIN1 regulates activity-dependent gene expression in human-induced glutamatergic neurons.

Bridging Integrator 1 (BIN1) is the second most important Alzheimer's disease (AD) risk gene, but its physiological roles in neurons and its contribution to brain pathology remain largely elusive. In this work, we show that BIN1 plays a critical role in the regulation of calcium homeostasis, electrical activity, and gene expression of glutamatergic neurons. Using single-cell RNA-sequencing on cerebral organoids generated from isogenic BIN1 wild type (WT), heterozygous (HET) and homozygous knockout (KO) human-induced pluripotent stem cells (hiPSCs), we show that BIN1 is mainly expressed by oligodendrocytes and glutamatergic neurons, like in the human brain. Both BIN1 HET and KO cerebral organoids show specific transcriptional alterations, mainly associated with ion transport and synapses in glutamatergic neurons. We then demonstrate that BIN1 cell-autonomously regulates gene expression in glutamatergic neurons by using a novel protocol to generate pure culture of hiPSC-derived induced neurons (hiNs). Using this system, we also show that BIN1 plays a key role in the regulation of neuronal calcium transients and electrical activity via its interaction with the L-type voltage-gated calcium channel Cav1.2. BIN1 KO hiNs show reduced activity-dependent internalization and higher Cav1.2 expression compared to WT hiNs. Pharmacological blocking of this channel with clinically relevant doses of nifedipine, a calcium channel blocker, partly rescues electrical and gene expression alterations in BIN1 KO glutamatergic neurons. Further, we show that transcriptional alterations in BIN1 KO hiNs that affect biological processes related to calcium homeostasis are also present in glutamatergic neurons of the human brain at late stages of AD pathology. Together, these findings suggest that BIN1-dependent alterations in neuronal properties could contribute to AD pathophysiology and that treatment with low doses of clinically approved calcium blockers should be considered as an option to slow disease-onset and progression.

Amyloid-Beta Peptides Trigger Premature Functional and Gene Expression Alterations in Human-Induced Neurons.

Alzheimer's disease (AD) is the most prevalent cause of dementia in the elderly, characterized by the presence of amyloid-beta (Aß) plaques, neurofibrillary tangles, neuroinflammation, synapse loss and neurodegeneration in the brain. The amyloid cascade hypothesis postulates that deposition of Aß peptides is the causative agent of AD pathology, but we still lack comprehensive understanding of the molecular mechanisms connecting Aß peptides to neuronal dysfunctions in AD. In this work, we investigate the early effects of Aß peptide accumulation on the functional properties and gene expression profiles of human-induced neurons (hiNs). We show that hiNs acutely exposed to low concentrations of both cell-secreted Aß peptides or synthetic Aß1-42 exhibit alterations in the frequency of calcium transients suggestive of increased neuronal excitability. Using single-cell RNA sequencing, we also show that cell-secreted Aß up-regulates the expression of several synapse-related genes and down-regulates the expression of genes associated with metabolic stress mainly in glutamatergic neurons and, to a lesser degree, in GABAergic neurons and astrocytes. These neuronal alterations correlate with activation of the SEMA5, EPHA and NECTIN signaling pathways, which are important regulators of synaptic plasticity. Altogether, our findings indicate that slight elevations in Aß concentrations are sufficient to elicit transcriptional changes in human neurons, which can contribute to early alterations in neural network activity.

The Alzheimer susceptibility gene BIN1 induces isoform-dependent neurotoxicity through early endosome defects.

The Bridging Integrator 1 (BIN1) gene is a major susceptibility gene for Alzheimer's disease (AD). Deciphering its pathophysiological role is challenging due to its numerous isoforms. Here we observed in Drosophila that human BIN1 isoform1 (BIN1iso1) overexpression, contrary to human BIN1 isoform8 (BIN1iso8) and human BIN1 isoform9 (BIN1iso9), induced an accumulation of endosomal vesicles and neurodegeneration. Systematic search for endosome regulators able to prevent BIN1iso1-induced neurodegeneration indicated that a defect at the early endosome level is responsible for the neurodegeneration. In human induced neurons (hiNs) and cerebral organoids, BIN1 knock-out resulted in the narrowing of early endosomes. This phenotype was rescued by BIN1iso1 but not BIN1iso9 expression. Finally, BIN1iso1 overexpression also led to an increase in the size of early endosomes and neurodegeneration in hiNs. Altogether, our data demonstrate that the AD susceptibility gene BIN1, and especially BIN1iso1, contributes to early-endosome size deregulation, which is an early pathophysiological hallmark of AD pathology.

The "sacral parasympathetic": ontogeny and anatomy of a myth.

We recently defined genetic traits that distinguish sympathetic from parasympathetic neurons, both preganglionic and ganglionic (Espinosa-Medina et al., Science 354:893-897, 2016). By this set of criteria, we found that the sacral autonomic outflow is sympathetic, not parasympathetic as has been thought for more than a century. Proposing such a belated shift in perspective begs the question why the new criterion (cell types defined by their genetic make-up and dependencies) should be favored over the anatomical, physiological and pharmacological considerations of long ago that inspired the "parasympathetic" classification. After a brief reminder of the former, we expound the weaknesses of the latter and argue that the novel genetic definition helps integrating neglected anatomical and physiological observations and clearing the path for future research.

Proteomic analysis reveals novel ligands and substrates for LNX1 E3 ubiquitin ligase.

Ligand of Numb protein X1 (LNX1) is an E3 ubiquitin ligase that contains a catalytic RING (Really Interesting New Gene) domain and four PDZ (PSD-95, DlgA, ZO-1) domains. LNX1 can ubiquitinate Numb, as well as a number of other ligands. However, the physiological relevance of these interactions in vivo remain unclear. To gain functional insights into the LNX family, we have characterised the LNX1 interactome using affinity purification and mass spectrometry. This approach identified a large number of novel LNX1-interacting proteins, as well as confirming known interactions with NUMB and ERC2. Many of the novel interactions mapped to the LNX PDZ domains, particularly PDZ2, and many showed specificity for LNX1 over the closely related LNX2. We show that PPFIA1 (liprin-α1), KLHL11, KIF7 and ERC2 are substrates for ubiquitination by LNX1. LNX1 ubiquitination of liprin-α1 is dependent on a PDZ binding motif containing a carboxyl terminal cysteine that binds LNX1 PDZ2. Surprisingly, the neuronally-expressed LNX1p70 isoform, that lacks the RING domain, was found to promote ubiquitination of PPFIA1 and KLHL11, albeit to a lesser extent than the longer RING-containing LNX1p80 isoform. Of several E3-ligases identified in the LNX1 interactome we confirm interactions of LNX1 with MID2/TRIM1 and TRIM27. On this basis we propose a model whereby LNX1p70, despite lacking a catalytic RING domain, may function as a scaffold to promote ubiquitination of its ligands through recruitment of other E3-ligases. These findings provide functional insights into the LNX protein family, particularly the neuronal LNX1p70 isoform.

Decreased Anxiety-Related Behaviour but Apparently Unperturbed NUMB Function in Ligand of NUMB Protein-X (LNX) 1/2 Double Knockout Mice.

NUMB is a key regulator of neurogenesis and neuronal differentiation that can be ubiquitinated and targeted for proteasomal degradation by ligand of numb protein-X (LNX) family E3 ubiquitin ligases. However, our understanding of LNX protein function in vivo is very limited. To examine the role of LNX proteins in regulating NUMB function in vivo, we generated mice lacking both LNX1 and LNX2 expression in the brain. Surprisingly, these mice are healthy, exhibit unaltered levels of NUMB protein and do not display any neuroanatomical defects indicative of aberrant NUMB function. Behavioural analysis of LNX1/LNX2 double knockout mice revealed decreased anxiety-related behaviour, as assessed in the open field and elevated plus maze paradigms. By contrast, no major defects in learning, motor or sensory function were observed. Given the apparent absence of major NUMB dysfunction in LNX null animals, we performed a proteomic analysis to identify neuronal LNX-interacting proteins other than NUMB that might contribute to the anxiolytic phenotype observed. We identified and/or confirmed interactions of LNX1 and LNX2 with proteins known to have presynaptic and neuronal signalling functions, including the presynaptic active zone constituents ERC1, ERC2 and LIPRIN-αs (PPFIA1, PPFIA3), as well as the F-BAR domain proteins FCHSD2 (nervous wreck homologue) and SRGAP2. These and other novel LNX-interacting proteins identified are promising candidates to mediate LNX functions in the central nervous system, including their role in modulating anxiety-related behaviour.

Tight, cell type-specific control of LNX expression in the nervous system, at the level of transcription, translation and protein stability.

LNX1 and LNX2 are E3 ubiquitin ligases that can interact with Numb - a key regulator of neurogenesis and neuronal differentiation. LNX1 can target Numb for proteasomal degradation, and Lnx mRNAs are prominently expressed in the nervous system, suggesting that LNX proteins play a role in neural development. This hypothesis remains unproven, however, largely because LNX proteins are present at very low levels in vivo. Here, we demonstrate expression of both LNX1 and LNX2 proteins in the brain for the first time. We clarify the cell-type specific expression of LNX isoforms in both the CNS and PNS, and identify a novel LNX1 isoform. Using luciferase reporter assays, we show that the 5' untranslated region of the Lnx1_variant 2 mRNA, that generates the LNX1p70 isoform, strongly suppresses protein production. This effect is mediated in part by the presence of upstream open reading frames (uORFs), but also by a sequence element that decreases both mRNA levels and translational efficiency. By contrast, uORFs do not negatively regulate LNX1p80 or LNX2 expression. Instead, we find some evidence that protein turnover via proteasomal degradation may influence LNX1p80 levels in cells. These observations provide plausible explanations for the low levels of LNX1 proteins detected in vivo.

Molecular evolution of the LNX gene family.

BACKGROUND: LNX (Ligand of Numb Protein-X) proteins typically contain an amino-terminal RING domain adjacent to either two or four PDZ domains - a domain architecture that is unique to the LNX family. LNX proteins function as E3 ubiquitin ligases and their domain organisation suggests that their ubiquitin ligase activity may be targeted to specific substrates or subcellular locations by PDZ domain-mediated interactions. Indeed, numerous interaction partners for LNX proteins have been identified, but the in vivo functions of most family members remain largely unclear. RESULTS: To gain insights into their function we examined the phylogenetic origins and evolution of the LNX gene family. We find that a LNX1/LNX2-like gene arose in an early metazoan lineage by gene duplication and fusion events that combined a RING domain with four PDZ domains. These PDZ domains are closely related to the four carboxy-terminal domains from multiple PDZ domain containing protein-1 (MUPP1). Duplication of the LNX1/LNX2-like gene and subsequent loss of PDZ domains appears to have generated a gene encoding a LNX3/LNX4-like protein, with just two PDZ domains. This protein has novel carboxy-terminal sequences that include a potential modular LNX3 homology domain. The two ancestral LNX genes are present in some, but not all, invertebrate lineages. They were, however, maintained in the vertebrate lineage, with further duplication events giving rise to five LNX family members in most mammals. In addition, we identify novel interactions of LNX1 and LNX2 with three known MUPP1 ligands using yeast two-hybrid asssays. This demonstrates conservation of binding specificity between LNX and MUPP1 PDZ domains. CONCLUSIONS: The LNX gene family has an early metazoan origin with a LNX1/LNX2-like protein likely giving rise to a LNX3/LNX4-like protein through the loss of PDZ domains. The absence of LNX orthologs in some lineages indicates that LNX proteins are not essential in invertebrates. In contrast, the maintenance of both ancestral LNX genes in the vertebrate lineage suggests the acquisition of essential vertebrate specific functions. The revelation that the LNX PDZ domains are phylogenetically related to domains in MUPP1, and have common binding specificities, suggests that LNX and MUPP1 may have similarities in their cellular functions.