Live-Cell Imaging of Single Neurotrophin Receptor Molecules on Human Neurons in Alzheimer’s Disease

Neurotrophin receptors such as the tropomyosin receptor kinase A receptor (TrkA) and the low-affinity binding p75 neurotrophin receptor p75NTR play a critical role in neuronal survival and their functions are altered in Alzheimer’s disease (AD). Changes in the dynamics of receptors on the plasma membrane are essential to receptor function. However, whether receptor dynamics are affected in different pathophysiological conditions is unexplored. Using live-cell single-molecule imaging, we examined the surface trafficking of TrkA and p75NTR molecules on live neurons that were derived from human-induced pluripotent stem cells (hiPSCs) of presenilin 1 (PSEN1) mutant familial AD (fAD) patients and non-demented control subjects. Our results show that the surface movement of TrkA and p75NTR and the activation of TrkA- and p75NTR-related phosphoinositide-3-kinase (PI3K)/serine/threonine-protein kinase (AKT) signaling pathways are altered in neurons that are derived from patients suffering from fAD compared to controls. These results provide evidence for altered surface movement of receptors in AD and highlight the importance of investigating receptor dynamics in disease conditions. Uncovering these mechanisms might enable novel therapies for AD.


Introduction
Changes in lateral diffusion of receptors on the plasma membrane are crucial in determining their functional status. The altered surface movements can result in the formation of membrane protein complexes that activate signaling molecules [1]. For instance, single-molecule tracking experiments showed that the neurotrophin receptors, such as the tropomyosin receptor kinase A receptor (TrkA), also known as high-affinity nerve growth factor receptor, have two distinct motility states on the membrane surface, defined as mobile and immobile phases [1]. The membrane recruitment of downstream intracellular signaling proteins occurs when TrkA is in the immobile phase [1]. It has also been 2 of 15 shown that the diffusion coefficient (D) that is calculated from the cell surface movement of receptors provides a good estimate for the effect of a ligand on G protein-coupled receptors [2]. Consequently, the functions of receptors can be predicted by measuring the surface movement parameters of receptor molecules. A large number of diseases are initiated by dysregulation of the receptor-related signaling pathways such as Parkinson's disease, schizophrenia, and cancer [3][4][5]. Although the expression level and pattern of various receptors have been extensively studied in disease models [6,7] the surface movement of receptors in different diseases has been scarcely investigated. It has recently been published that the surface diffusion of AMPA receptors is disturbed in rodent models of Huntington's disease and the restoration of the AMPA receptor diffusion rescues memory dysfunction [8]. These data indicate that revealing receptor dynamics in different diseases may contribute to developing new treatments for neurodegenerative diseases such as Alzheimer's disease (AD).
Alzheimer's disease is the most common cause of dementia which is characterized by extracellular deposition of beta-amyloid (Aβ) and intracellular neurofibrillary tangles [9]. The familial form of AD (fAD) is caused by genetic mutations, such as mutations in genes for encoding presenilin (PSEN1 or PSEN2) proteins [10]. In AD the unbalanced signaling through the low-affinity binding p75 neurotrophin receptor (p75 NTR ) versus TrkA related signaling [6] plays a pivotal role in changes of altered cellular functions. TrkA and p75 NTR activate signaling pathways that are essential to decide the fate of neurons [6,11].
Our understanding of AD pathogenesis is currently limited by difficulties in obtaining live neurons from patients, especially in the earlier stages of the disease. Furthermore, the culture of primary human neuronal cells is particularly challenging because of their limited life span [19]. Also, biopsies are often taken from patients at the end-stage of the disease and suitable control tissues that are taken from healthy individuals might be inaccessible due to ethical concerns and potential health risks [20]. Human-induced pluripotent stem cell (hiPSCs) based technology overcomes many of these limitations [21], providing a suitable cellular model for AD. Human iPSCs can be cultured for an unlimited time and differentiated into any type of cell of the human body, such as neurons of the central nervous system (CNS) [22]. We have successfully established and applied AD patient-derived hiPSC lines for modeling AD neuropathology [23,24].
Here, we applied these hiPSCs and performed single-molecule tracking approaches to examine the surface movement of p75 NTR and TrkA in living human neurons from PSEN1 mutant patients and non-demented control subjects. We also examined how downstream signaling pathways, such as ERK1/2, AKT, and JNK1/2 that play essential roles in fAD, could be linked to TrkA and p75 NTR activation change in PSEN1 mutant neurons.

Characterization of Neuronal Properties of hiPSC-Derived Cultures
hiPSC-derived seven week old neurons were tested to see if they showed neuronal characteristics. All of the cell lines possessed neuron-like morphology and were positively stained for the neuronal markers, β-III tubulin (TUBB3) and microtubule-associated protein 2 (MAP2) ( Figure 1A), representing the neuronal differentiation of iPSC derived cell cultures.
The identification of the neuronal phenotype of the hiPSC-derived neurons was conducted by immunolabeling them for glutamatergic (VGLUT1/2), GABAergic (GAD65/67), and cholinergic (VAChT) markers. Our results showed that hiPSC-derived neurons represent a phenotypically mixed population ( Figure 1B) in agreement with our earlier published results [24].
In the next step, we examined whether our neuronal differentiation protocol resulted in cells with electrophysiological properties similar to mature neurons. Using the wholecell patch-clamp technique, the passive electrophysiological parameters of 12 neurons were measured in the cultures. The resting membrane potential and the input resistance of the patched neurons were −64.75 ± 3.9 mV and 742.18 ± 64.59 MΩ, respectively (values are represented as mean ± SEM). Action potentials and voltage-gated Na + and K + currents were also recorded from these cells. As evidence for maturation, the cells exhibited either single or repetitive action potentials in response to positive current injection ( Figure 1C). Furthermore, a series of 10 mV depolarization voltage steps from −90 mV resulted in the opening of Na + and K + channels, indicating fast inward and slow outward components, respectively ( Figure 1D). Tetrodotoxin (TTX) blocked the inward, while 4-AP blocked the outward currents representing the activation of voltage-gated Na + and K + channels ( Figure 1E,F). Overall, the seven week old hiPSC-derived cultures resembled mature neuronal cultures with mixed phenotypes.

Expression and Single-Molecule Detection of TrkA and p75 NTR in Control and fAD hiPSC-Derived Neurons
As the balance of TrkA and p75 NTR signaling is disturbed in fAD [6], the total TrkA and p75 NTR levels in hiPSC-derived neuronal cultures that were obtained from nondemented controls and fAD patients with the PSEN1 mutation were examined by Western blot. Our results demonstrated that the expression of TrkA significantly decreased in PSEN1 fAD hiPSC neurons (Figure 2A,B). In contrast, the p75 NTR signal was below the detection limit in the cultures of control lines but was abundant in PSEN1 fAD hiPSC neurons (Figure 2A,B). Consequently, the ratio of p75 NTR /TrkA was elevated in PSEN1 mutant neurons ( Figure 2C). Next, immunocytochemical stainings for TrkA and p75 NTR on fixed hiPSC-derived neuronal cultures that were obtained from non-demented control and PSEN1 mutant cell lines were performed. The results showed that both neurites and somata expressed TrkA and p75 NTR ( Figure 2D). In contrast to fixed immunocytochemistry, TrkA and p75 NTR molecules were only observed on processes of live cells (Supplementary Movies S1 and S2). To reveal that the moving TrkA and p75 NTR molecules are only detected on the neurites, we carried out correlated live-cell single-molecule (TrkA or p75 NTR ) imaging and fixed cell immunocytochemistry (TUBB3) experiments (Supplementary Movie S3).

Surface Movement Characteristics of TrkA and p75 NTR in Control hiPSC Neurons
Live-cell imaging experiments were performed to determine the surface movement parameters of TrkA and p75 NTR molecules in the plasma membrane of neurons using TIRF microscopy. After fluorescent labeling of TrkA and p75 NTR receptors, 10 s long videos were recorded, single molecule movements were tracked, and then analyzed. Before comparing the diffusion parameters of TrkA and p75 NTR in the control and fAD hiPSC neurons, we analyzed the surface dynamics of TrkA and p75 NTR in hiPSC neurons that were derived from healthy subjects. Based on the MSD functions, TrkA and p75 NTR molecules exhibited two main types of movements on hiPSC neurons that were obtained from non-demented control cell lines: Brownian diffusion, when receptors are moving freely between plasma membrane barriers, and confined motion when the surface movement of receptors are restricted to a small area ( Figure 3A,B). On average, p75 NTR molecules showed confined movements, while TrkA molecules exhibit Brownian diffusion ( Figure 3C). The evaluated average diffusion coefficients of TrkA and p75 NTR did not change significantly during the measurement either in the control or PSEN1 mutant samples ( Figure 3D). However, the average diffusion coefficient of TrkA and p75 NTR on healthy individualderived neurites indicated that p75 NTRs were moving significantly faster on neurites than TrkAs ( Figure 3D; Table S1).

Surface Movement Characteristics of TrkA and p75 NTR in Control hiPSC Neurons
Live-cell imaging experiments were performed to determine the surface movement parameters of TrkA and p75 NTR molecules in the plasma membrane of neurons using TIRF microscopy. After fluorescent labeling of TrkA and p75 NTR receptors, 10 s long videos were recorded, single molecule movements were tracked, and then analyzed. Before comparing the diffusion parameters of TrkA and p75 NTR in the control and fAD hiPSC neurons, we analyzed the surface dynamics of TrkA and p75 NTR in hiPSC neurons that were derived from healthy subjects. Based on the MSD functions, TrkA and p75 NTR molecules exhibited two main types of movements on hiPSC neurons that were obtained from non-demented control cell lines: Brownian diffusion, when receptors are moving freely between plasma membrane barriers, and confined motion when the surface movement of receptors are restricted to a small area ( Figure 3A,B). On average, p75 NTR molecules showed confined movements, while TrkA molecules exhibit Brownian diffusion ( Figure 3C). The evaluated average diffusion coefficients of TrkA and p75 NTR did not change significantly during the measurement either in the control or PSEN1 mutant samples ( Figure 3D). However, the average diffusion coefficient of TrkA and p75 NTR on healthy individual-derived neurites indicated that p75 NTRs were moving significantly faster on neurites than TrkAs ( Figure 3D; Table S1). The diffusion area of p75 NTR molecules was more extended on neurites that were derived from healthy individuals than that of TrkA ( Figure 3E; Table S1). Accordingly, the length of p75 NTR trajectories was significantly longer ( Figure 3F; Table S1).
To identify the diffusion states of TrkA and p75 NTR , we used Variational Bayesian-Hidden Markov model (VB-HMM) clustering analysis [25] on the trajectories for each neurite. The analysis showed that the diffusion of both TrkA and p75 NTR molecules can be classified into three phases: slow, medium, and fast. Our results suggested that the percentage of the fast fraction increased, while the proportion of slow and medium phases did not change significantly when comparing p75 NTR to TrkA molecules (Figure 5C,E; Table S2). Nevertheless, the average diffusion coefficient of the slow and medium fractions of p75 NTR significantly increased compared to TrkA (Figure 5C,E; Table S2). The diffusion area of p75 NTR molecules was more extended on neurites that were derived from healthy individuals than that of TrkA ( Figure 3E; Table S1). Accordingly, the length of p75 NTR trajectories was significantly longer ( Figure 3F; Table S1).
To identify the diffusion states of TrkA and p75 NTR , we used Variational Bayesian-Hidden Markov model (VB-HMM) clustering analysis [25] on the trajectories for each neurite. The analysis showed that the diffusion of both TrkA and p75 NTR molecules can be classified into three phases: slow, medium, and fast. Our results suggested that the percentage of the fast fraction increased, while the proportion of slow and medium phases did not change significantly when comparing p75 NTR to TrkA molecules ( Figure 5C,E; Table S2). Nevertheless, the average diffusion coefficient of the slow and medium fractions of p75 NTR significantly increased compared to TrkA ( Figure 5C,E; Table S2).

Comparison of the Surface Movement of TrkA and p75 NTR in Control and PSEN1 fAD hiPSC-Derived Neurons
As surface movement of receptors determine their activation state [26,27], we compared the diffusion parameters of TrkA and p75 NTR in healthy individual-derived and PSEN1 fAD hiPSC neurons. MSD-∆t plots of the trajectories showed that the surface movement of TrkA molecules were less confined in the PSEN1 fAD hiPSC neurons when compared with the non-demented control samples ( Figure 4A). On the other hand, p75 NTR molecules exhibited a more confined diffusion mode in the PSEN1 fAD hiPSC neurons ( Figure 4B).   Table S1). In addition, the trajectory length that was normalized to steps was also smaller in the case of both TrkA and p75 NTR molecules in fAD neurons ( Figure 4E,F; Table S1).
The average diffusion coefficient of TrkA molecules did not change in PSEN1 fAD hiPSC neurites compared to the non-demented control neurites ( Figure 5A; Table S1). In contrast, the surface movement of p75 NTR receptor molecules significantly decreased in the PSEN1 fAD hiPSC neurites ( Figure 5B; Table S1).

Expression of the ERK1/2, AKT and JNK1/2 Signaling Pathways in Non-Demented Control and PSEN1 fAD hiPSC-Derived Neurons
Since changes in surface movements of receptors lead to functional changes via the modification of signaling transduction [1], the activation of signaling pathways were examined. To investigate whether TrkA and p75 NTR -related downstream signaling were altered in PSEN1 fAD hiPSC neurons, we examined the ERK1/2, AKT, and JNK1/2 signaling pathways. Our Western blot analysis revealed that the expression levels of ERK1/2 and AKT did not change, while the JNK1/2 levels significantly decreased in the cultures of PSEN1 fAD samples ( Figure 6A-C). The activation of ERK1/2 and JNK1/2 pathway was not altered but the AKT signaling pathway was shown to be more active in the PSEN1 fAD cultures (Figure 6A-C).  Table S2).
In contrast to TrkA, the proportion of slow trajectories of p75 NTR increased while the percentage of fast trajectories decreased in the PSEN1 fAD hiPSC neurites ( Figure 5E,F; Table S2). The average diffusion coefficient of the slow phase of p75 NTR decreased in fAD, but the average diffusion coefficients of the other two fractions remained unchanged in the neurites of PSEN1 mutant hiPSC neurons ( Figure 5E,F; Table S2).

Expression of the ERK1/2, AKT and JNK1/2 Signaling Pathways in Non-Demented Control and PSEN1 fAD hiPSC-Derived Neurons
Since changes in surface movements of receptors lead to functional changes via the modification of signaling transduction [1], the activation of signaling pathways were examined. To investigate whether TrkA and p75 NTR -related downstream signaling were altered in PSEN1 fAD hiPSC neurons, we examined the ERK1/2, AKT, and JNK1/2 signaling pathways. Our Western blot analysis revealed that the expression levels of ERK1/2 and AKT did not change, while the JNK1/2 levels significantly decreased in the cultures of PSEN1 fAD samples ( Figure 6A-C). The activation of ERK1/2 and JNK1/2 pathway was not altered but the AKT signaling pathway was shown to be more active in the PSEN1 fAD cultures ( Figure 6A-C).

Discussion
Our results showed that the surface trafficking of TrkA and p75 NTR are altered in hiPSC-derived neurons that are differentiated from PSEN1 mutant fAD patients. The surface movement of TrkA molecules was less confined in PSEN1 mutant neurites. Contrarily, the trafficking of p75 NTR molecules was more confined in the fAD neurites. The movement of TrkA and p75 NTR both covered smaller areas. The average diffusion coefficient of TrkA did not change, while the average diffusion coefficient of p75 NTR decreased in fAD neurites. We revealed that the diffusion state of both TrkA and p75 NTR molecules can be classified into three diffusion states: slow, medium, and fast phases. While the percentage of these diffusion states of TrkA was unaffected in the neurons that were derived from

Discussion
Our results showed that the surface trafficking of TrkA and p75 NTR are altered in hiPSC-derived neurons that are differentiated from PSEN1 mutant fAD patients. The surface movement of TrkA molecules was less confined in PSEN1 mutant neurites. Contrarily, the trafficking of p75 NTR molecules was more confined in the fAD neurites. The movement of TrkA and p75 NTR both covered smaller areas. The average diffusion coefficient of TrkA did not change, while the average diffusion coefficient of p75 NTR decreased in fAD neurites. We revealed that the diffusion state of both TrkA and p75 NTR molecules can be classified into three diffusion states: slow, medium, and fast phases. While the percentage of these diffusion states of TrkA was unaffected in the neurons that were derived from PSEN1 mutant patients, the percentage of the slow component of p75 NTR increased and its fast component decreased. In addition, we found that the AKT signaling pathway was more active in PSEN1 fAD cultures.
We have characterized the neuronal phenotype of the hiPSC-derived cultures. Our results demonstrated that the seven week old cultures expressed neuronal markers such as TUBB3 and MAP2 and showed electrophysiological characteristics of mature neurons. In addition, we tested their neuronal phenotype, revealing that hiPSC-derived neuronal cultures form a mixed population of glutamatergic, GABAergic, and cholinergic neurons when dual-SMAD inhibition-based neuronal differentiation is used [22,24]. We have also validated the fAD patient's iPSC-based cellular pathology model that was used in this study. Our results showed that PSEN1 mutant hiPSC neurons produced more Aβ 1-40 and Aβ 1-42 and that Aβ 1-42 /Aβ 1-40 ratio increased compared to the control cells. Previously, we have confirmed that hyperphosphorylation of TAU protein and an increased level of active glycogen synthase kinase 3 beta (GSK3B), a physiological kinase of TAU, is detected in PSEN1 mutant patient groups [24]. In our previous findings we also demonstrated that γ-secretase inhibitor DAPT treatment on control and PSEN1 mutant iPSC-derived neurons resulted in reduced endogenous amyloid levels and intracellular accumulation of an AβPP-C-terminal fragment [23].
Next, we have examined the expression pattern of TrkA and p75 NTR on the control and PSEN1 mutant iPSC-derived neurons as TrkA and p75 NTR play a critical role in the progress of AD. Although the expression levels of TrkA and p75 NTR are differently affected depending on the examined brain area and the progression stage of AD [28], their ratio is essential in determining the functional outcome [6,29]. Our data show that the TrkA levels are decreased, while p75 NTR levels are elevated in the PSEN1 mutant neurites. Consequently, the ratio of p75 NTR /TrkA is higher in PSEN1 mutant neurons compared to neurons that were derived from non-demented control cell lines suggesting that the balance is shifted toward the p75 NTR activated intracellular signaling in PSEN1 mutant neurons.
Although the expression pattern of TrkA and p75 NTR , their interaction, and stimulated signaling pathways leading to AD progression are being extensively studied [6,30], their surface movements in AD have remained unexplored so far. Therefore, in the current study, we investigated the surface trafficking of TrkA and p75 NTR on control and PSEN1 mutant iPSC-derived neurons.
Our data demonstrate that TrkA and p75 NTR exhibited different diffusion modes in the control hiPSC-derived cultures. MSD-∆t plots of receptor trajectories indicated that TrkA molecules show Brownian diffusion, while p75 NTR molecules display confined motion, and these diffusion modes are altered in PSEN1 mutant fAD neurons. The lateral diffusion of TrkA molecules was less confined in fAD neurites compared to the non-demented controls, while the diffusion of p75 NTR was significantly more confined in fAD neurites. Our finding also shows that the diffusion area of both neurotrophin receptors is smaller in the PSEN1 mutant neurites. It is now evident that various mechanisms, including the corralling and tethering effect of membrane skeleton [31], lipid complexity [32], lipid-protein interaction [33], and crowding effect of membrane proteins [32,34,35] regulate the dynamics of membrane receptors and ultimately the cell functions. The actin cytoskeleton forms barriers making the membrane compartmentalized. The membrane domains that are generated by the actin fence deviate the dynamics of membrane proteins away from Brownian diffusion [36,37]. Axons and dendrites have a unique membrane cytoskeleton, a periodic cortical actin-spectrin network [38][39][40][41][42] that plausibly also creates special membrane domains causing anomalous diffusion [43]. AD is associated with a disruption of membrane properties such as alterations in membrane lipid composition [44] or increased spectrin proteolysis [45] that may contribute to changes that are seen in the altered diffusion mode and area of TrkA and p75 NTR in PSEN1 mutant neurites.
The surface movements of receptors also indicate their activation state [26,27]. Signal transduction starts with receptor activation, which, in turn, can be precisely described by the changes of their surface trafficking in the plasma membrane [26,27]. Here we show that p75 NTR moves faster along the neurites than TrkA in control hiPSCs. In PSEN1 mutant neurites the average diffusion coefficient of TrkA does not change while p75 NTR moves slower. As described earlier, TrkA is immobilized upon ligand binding, which is related to the start of signal transduction [1]; p75 NTR has also been shown to slow down and start signalization upon ligand binding [46]. Based on these results our data indicate that TrkA might be less active, while p75 NTR might be more active in PSEN1 mutant neurons as it is also suggested by the increased expression of p75 NTR /TrkA. The clustering analysis that was based on VB-HMM showed that the percentage of slow, medium, and fast fractions of TrkA did not change, whereas the proportion of slow components of p75 NTR molecules increased, their fast fraction decreased, and there was no change in the percentage of the medium fraction. As the average diffusion coefficients of the three fractions did not differ from the values that were calculated for p75 NTR in healthy individual-derived neurites, we assumed that the decrease in the average diffusion coefficient of p75 NTR in fAD was due to a shift of fast-moving molecules to the slow state.
TrkA and p75 NTR are associated with downstream signaling molecules that are important in determining the fate of neurons [6,11,47]. TrkA is generally responsible for signaling survival in response to ligand by inducing ERK1/2 and AKT phosphorylation [12]. On the other hand, p75 NTR activation can lead to survival via the AKT signaling pathway [18] but can stimulate apoptotic pathways as well via the JNK pathway [18]. As receptor dynamics determine signaling pathways activation, we examined the changes in the activation of the above-mentioned most relevant TrkA-and p75 NTR -related signaling pathways in fAD. We found that the AKT pathway is more active in PSEN1 mutant neurons. This finding is in accordance with a study that performed global transcriptomic analysis of human iPSC lines carrying PSEN1 mutations and reported that the genes of (PI3K)-AKT signaling pathway are among the most enriched [48].
Even though the activation of the ERK1/2 and JNK pathway did not show any differences between the control and PSEN1 mutant neurons, we uncovered a significant decrease in the expression levels of JNK and pJNK in fAD. The actual changes of the intracellular downstream actions and finding the connection between the receptor dynamics of TrkA/p75 NTR and the intracellular signaling pathways, however, require further investigation and are beyond the scope of this study.
In summary, our data provide evidence that the surface trafficking of TrkA and p75 NTR is altered in PSEN1 mutant hiPSC-derived neurons compared to those of non-demented control hiPSC-derived neurons. These results draw attention to the significance of the investigation of receptor dynamics in disease conditions. Understanding these mechanisms may provide novel therapeutic strategies for the prevention and/or treatment of AD.

Generation of hiPSC-Derived Neurons
A total of three Alzheimer's disease (AD) patient-derived iPSC lines were used in our experiments. The first one, BIOT-7183-PSEN1 (referred as BIOTi001-A in hP-SCreg; https://hpscreg.eu/, accessed on 7 December 2021), bearing a mutation in PSEN1 gene (PSEN1 c.265G > C, p.V89L; was already thoroughly characterized (fAD-1; female, age: 55) [49]. The other two iPSC lines were generated from two male siblings bearing the same PSEN1 c.449T > C, p.L150P mutation (fAD-3 and fAD-4; males, age: 58) as published previously [50]. The patients were clinically diagnosed and characterized by the Institute of Genomic Medicine and Rare Disorders, Semmelweis University, Budapest, Hungary or at the Danish Dementia Research Centre, Rigshospitalet, University of Copenhagen, Denmark, as described previously [24,49,50]. Non-demented volunteers (assessed by clinical evaluation) were used as controls (3 individuals: Ctrl-1, female, age:33; Ctrl-2, female, age:36; Ctrl-5, female, age:56) and all iPSC lines were established, characterized, and maintained under the same conditions, as we published earlier [24]. Neural progenitor cells (NPCs) were generated from each of the hiPSCs by dual inhibition of the SMAD signaling pathway [22] (See details in Supplementary Materials). NPCs then were terminally differentiated into cortical neurons (see details in Supplementary Materials). The extracellular Aβ 1-40 and Aβ 1-42 levels were measured in the cultures (Supplementary Figure S1) and the neuronal differentiation was validated with immunocytochemistry and electrophysiology (see Supplementary Materials, Supplementary Figure S2 and Table S3 for details). Expression of TrkA and p75 NTR on hiPSC-derived neurons was identified with immunocytochemistry ( Figure 2D, Supplementary Movies S4 and S5). Live-cell immunofluorescent labeling was performed to detect TrkA and p75 NTR molecules in the plasma membrane of the neurons. Compared to conventional epifluorescence imaging where high background and out-of-focus light is observable, total internal reflection fluorescence (TIRF) microscopy provides a better signal, lower light intensity, and a higher resolution which makes the precise superficial membrane detections and live-cell imaging possible. After the measurement, the diffusion parameters such as the diffusion coefficients; the percentage of slow, medium, and fast fractions; the diffusion area; and the trajectory length of TrkA and p75 NTR molecules were calculated. Maximum likelihood estimation [51] was applied to obtain the corresponding diffusion coefficient for each trajectory. Clustering analysis, that was based on a Variational Bayesian for Hidden Markov model, was used to determine the percentage of the slow, medium, and fast fractions of trajectories per neurites [25]. The diffusion area and the length of a molecule trajectory were calculated by a MATLAB script (see Supplementary Materials and Supplementary Figure S3 for details).

Investigation of the Activation of Signaling Pathways in hiPSC-Derived Neuronal Cultures
The activation of TrkA and p75 NTR -related downstream signaling pathways (ERK1/2, AKT, JNK) were examined in representative samples of a PSEN1 mutant and a nondemented individual (control) using Western blot (see Supplementary Materials and  Table S4 for details).

Statistical Analysis
Values are expressed as mean ± S.E.M. Statistical differences were considered significant at * p < 0.05; ** p < 0.01; *** p < 0.001. All results were analyzed using Statistica 13.3 for Windows (TIBCO) (see details in the Supplementary Materials).