Frontotemporal dementia-associated N279K tau mutant disrupts subcellular vesicle trafficking and induces cellular stress in iPSC-derived neural stem cells

Frontotemporal dementia with parkinsonism related to chromosome 17 (FTDP-17) prevails as one of the most common form of early-onset dementia [1]. FTDP-17 presents as a fulminant progressive neurodegenerative dementia, with no curative treatment or effective palliative relief. Clinically, FTDP-17 displays a triad of symptoms, with parkinsonism, behavioral changes and personality dysfunctions, and later cognitive impairments [1, 2]. Two genetic causes have been identified that lead to distinct forms of FTDP-17; mutations in progranulin (GRN) cause FTDP-17U (ubiquitin) with TDP-43 pathology and mutations in the microtubule-associated protein tau (MAPT) gene cause FTDP-17T (tau) [3]. Approximately one-half of all FTDP-17 cases are caused by autosomal dominant mutations in MAPT . The pallido-ponto-nigral degeneration (PPND) family is the largest and most comprehensively studied kindred of tau N279K mutant carriers and of all FTDP-17 families, containing to date, 59 affected individuals [4].

The N279K substitution is encoded within exon 10 of the MAPT gene locus and is one of the most frequent mutations in FTDP-17T patients [2]; together with mutations P301L and intron 10 + 16. Together these three mutations account for up to 60 % of FTDP-17T cases [5]. Multiple MAPT transcripts exist due to the alternative splicing of the MAPT gene, and transcripts of MAPT can be classified based on the inclusion/exclusion of exon 10 [6]. Most MAPT mutations including the N279K mutation increase the transcript containing exon 10 resulting in the overproduction of tau protein isoforms containing four tandem microtubule-binding domain repeats (4R-tau); whereas levels of the other major isoform in human brain containing only three repeats (3R-tau) is believed to be unaffected [7]. Although it is not fully understood how the increase in 4R-tau contributes to disease pathogenesis, MAPT mutations have been shown to cause multiple tauopathies [8‐10]. It is believed that tauopathies are caused by aberrant hyperphosphorylation of tau, leading to the assembly of variable neurotoxic tau aggregates and deposition of insoluble tau fibers in both neurons and glia [11]. Progressive tau deposition in FTDP-17T patients is associated with severe neocortical atrophy of the frontal and temporal lobes, in association with the degeneration of medial temporal structures [12]. Destruction of the basal ganglia and depigmentation of the substantia nigra may also be present to some degree in MAPT mutant carriers, but are documented to be a consistent pathological hallmark in N279K kindreds [13, 14].

The exact mechanism owing to the cell death of specific neuronal populations in FTDP-17T remains to be identified; however, the characteristic dominant penetrance of the N279K tau mutant suggests a gain-of-toxic-function that results in specific subsets of degenerating cells. The induced pluripotent stem cell (iPSC) technology provides a novel and unparalleled approach to aid the study of molecular and cellular dynamics of disease pathogenesis by utilizing patient-specific cells with pathologically linked genetic mutations on the inherent genetic background [15], which has facilitated its exponential use as a novel tool to understand the pathogenesis of multiple neurodegenerative diseases [16]. Herein, we investigated the pathogenic mechanism underlying FTDP-17T with mutant N279K tau by using cells derived from two patients from the PPND family who are both clinically affected. Neural stem cells (NSCs) derived from the patient-specific iPSCs revealed abnormal amounts of cellular vesicle components in the presence of aberrantly spliced tau, accompanied by increases in several components of cellular stress. Correspondingly, altered cellular markers reflecting abnormal vesicle trafficking were observed in the autopsy brains of patients carrying the N279K mutation, supporting the results from the iPSC-based approach. Our findings demonstrate that patient-specific iPSCs can be used to model neurodegenerative diseases and uncover cellular dysfunctions associated with a disease-causing mutation.

Inheritance of one allele of MAPT with the N279K mutation is causal to the development of pre-geriatric dementia, in the form of FTDP-17 [14]. The mutation resides within an exonic splicing enhancer site, and increases the level of the isoform containing exon 10, resulting in enhanced 4R-tau mRNA production [22]. Whereas 4R-tau protein is shown to have a stronger interaction with microtubules than 3R-tau, 4R-tau is also more effective at assembling microtubules [23]; the N279K mutation does not appear to affect the binding of tau to microtubules or decrease the ability of tau to promote microtubule assembly when compared to wild-type 4R-tau [24]. Thus, it remains unclear how the N279K tau mutant leads to neurodegeneration and tauopathy in FTDP-17T patients. In this study, we demonstrate that NSCs carrying the N279K tau mutant exhibit higher levels of 4R-tau and markedly higher cellular stress compared with control NSCs using patient-specific iPSCs. Furthermore, we show that intracellular vesicle trafficking is significantly perturbed. Importantly, we show that N279K tau influences intracellular vesicle trafficking as early as the NSC stage, which could not be examined through previous in vitro experiments.

Endocytosed and intracellular products are packaged into membrane bound specialized vesicles that are trafficked along microtubules to reach their correct destination or perform their specialized function to ensure proper cell dynamics. Impairment in the orderly trafficking of proteins and lipids intracellularly has been documented to initiate and execute cell death processes. Furthermore, intra-organelle communication is also believed to play a crucial role in the initiation of cell death events [25]. Mounting evidence implicates trafficking alterations, in particular, de-regulation of the endosomal system in the pathogenesis of neurodegenerative diseases [26]. In this work, we find substantial increases in the levels of flotillin-1 in N279K NSCs. Flotillin proteins which represent scaffolds of lipid raft microdomains [21] are involved in the sorting of intracellular vesicle compartments toward Golgi apparatus, recycling endosomes, multivesicular bodies [27], late endosomes/lysosomes [28] and exosomes [29]. Thus, the accumulation of flotillin-1 in intracellular space is indicative of interruptions of proper vesicle trafficking or abnormal acceleration of endocytosis machinery. Furthermore, our results show markedly increased levels of an early endosome marker EEA1 and a late endosome marker CHMP2B in N279K NSCs. In contrast, N279K NSCs show marked depletion of lysosomal compartments and protein levels of lysosomal marker LAMP-1. Lysosomes are highly dynamic membrane bound organelles whose main role is to receive and terminally degrade components sequentially sent through the endocytic pathway and from autophagic processes [30]. Therefore, the possession of the tau N279K mutant may disrupt the delivery of waste products from endosomes to lysosomes in NSCs, resulting in enlarged intracellular vesicles such as endosomes and exosomes. The aberrant processing of intracellular vesicles eventually render NSCs more intrinsically vulnerable to internal stressors and initiate signaling of cell stress through their accumulation. This accumulation may saturate the endocytic pathway and lead to the theorized propagation of pathologic/harmful products through intercellular exosomal transfer. Whilst this is seen as a permissive and protective mechanism to relieve intracellular vesicular traffic, exosomal enrichment of pathogenic tau may be speculated as an underlying vector of disease spread in FTDP-17.

Consistent with our NSC data, we show the increased flotillin-1 in the frontal and temporal cortices in post-mortem brain tissue of N279K carriers, and increased CHMP2B levels in the frontal cortex of the PPND/FTDP-17 patients. Interestingly, mutations in CHMP2B are also causal for the development of FTD [31, 32] and amyotrophic lateral sclerosis [33]. CHMP2B is a subunit of endosomal sorting complex required for transport-III, which is involved in the degradation of proteins in the endocytic and autophagic pathways [34]. In fact, CHMP2B immunoreactive granules are observed as granulovacuolar degeneration in the brains of Parkinson’s disease, incidental Lewy body disease [35] and Alzheimer’s disease (AD) patients [36]. Flotillin-1 is also known to accumulate in tangle-bearing neurons in AD brains [37]. Thus, the alteration of vesicle trafficking is likely a common pathogenic event across a spectrum of neurodegenerative diseases, although further work is required to completely elucidate whether the trafficking disturbance is a cause or a consequence of neurodegeneration.

In FTDP-17T as well as other neurodegenerative diseases termed tauopathies, accumulation of tau pathology is the primary histological feature. Flotillin-1 is increased in the frontal and temporal cortex, but not in occipital cortex where tau accumulation and deposition is relatively mild. Since 4R-tau levels are increased in all three cortical regions from the patients, there might be specific molecular mechanisms to compensate for the harmful effects of increased 4R-tau due to N279K tau in occipital cortex. Mutant N279K tau may initiate or facilitate the aggregation of tau into filaments by affecting vesicle trafficking in NSCs and/or neurons later on in disease when compensatory mechanisms are expended. Exosomes provide a motile intercellular chamber for the delivery of endocytosed and intracellular molecules to the extracellular space, for the active uptake by neighboring cells, which has been recently hypothesized to be a major component in the phenomena of disease pathology propagation [38], especially involving tau [39]. Since flotillin-1 is abundantly localized and enriched in intraluminal vesicles/exosomes, it is possible that N279K tau mutation facilitates the formation of exosomes and potentiates the spreading of tau pathology, which exacerbates cellular viability later on in the course of the disease.

Our results show that NSCs harboring the mutant N279K tau display altered levels of multiple intracellular vesicle trafficking markers, suggesting disruptions in the transport of membrane-bound organelles, which may ultimately lead to enhanced cellular stress observed in tau mutant cells. These findings have important implications for our understanding in the pathogenic mechanism of N279K tau mutation and those findings might be partly shared with sporadic FTD and/or PSP. Further studies are needed to elucidate the precise pathways by which N279K mutant tau alters intracellular vesicle trafficking. Importantly, our findings provide evidence that iPSC technology can be used to successfully model early dysfunctions in somatic cells derived from affected human patients, which can be complemented by post-mortem brain studies and may underlie future advances in the translatability of research involving PPND/FTDP-17 patients.