Subscriber access provided by Kaohsiung Medical University
Letter
The fate of nascent APP in hippocampal neurons: a live cell imaging study Claire E. DelBove, Xian-zhen Deng, and Qi Zhang ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.8b00226 • Publication Date (Web): 05 Jun 2018 Downloaded from http://pubs.acs.org on June 5, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 27
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Neuroscience
Title: The fate of nascent APP in hippocampal neurons: a live cell imaging study
Authors: Claire E. DelBove1, Xian-zhen Deng1, Qi Zhang1*
Affiliation: 1
Department of Pharmacology, Vanderbilt University, Nashville, TN 37232-6600
* Corresponding author
Contact Information: Department of Pharmacology Vanderbilt University 23rd Ave. S. at Pierce Ave. Nashville, TN 37232-6600 Phone: (615)-875-7620 Email:
[email protected] 1
ACS Paragon Plus Environment
ACS Chemical Neuroscience
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 28 13 29 30 14 31 32 15 33 34 16 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 27
Abstract: Amyloid precursor protein (APP) is closely associated with Alzheimer’s disease (AD) because its proteolytic products form amyloid plaques and its mutations are linked to familial AD patients. As a membrane protein, APP is involved in neuronal development and plasticity. However, it remains unclear how nascent APP is distributed and transported to designated membrane compartments to execute its diverse functions. Here, we employed a dual-tagged APP fusion protein in combination with a synaptic vesicle marker to study the surface trafficking and cleavage of APP in hippocampal neurons immediately after its synthesis. Using long-term time-lapse imaging, we found that a considerable amount of nascent APP was directly transported to the somatodendritic surface, from which it propagates to distal neurites. Some APP in the plasma membrane was endocytosed and some was cleaved by α-secretase. Hence, we conclude that surface transportation of APP is a major step preceding its proteolytic processing and neuritic distribution.
Keywords: Alzheimer’s disease, Amyloid precursor protein, Secretase, dynamin, plasma membrane, Endocytosis
2
ACS Paragon Plus Environment
Page 3 of 27
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 21 47 48 22 49 50 23 51 52 53 24 54 55 25 56 57 26 58 59 60
ACS Chemical Neuroscience
Introduction APP is well-known for its association with Alzheimer’s disease (AD), an incurable from of neurodegeneration prevalent in the elderly. In addition to its pathological connection, APP has been implicated in the development of neurons
1-2
, the maturation of synapses 3, the regulation of synaptic
plasticity5, and even the metabolism of cholesterol in the central nervous system (CNS)
4-7
. APP’s
various proteolytic pathways and different cleavage products underlie its diverse functions. One of the proteolysis pathways generates Aβ peptides, the major constituent of the amyloid plaques commonly found in AD patients’ brains. Through decades of research, it is now clear that the different secretases and their proteolytic processing of APP are compartmentalized in different parts of cells, including the plasma membrane and the endosomal/lysosomal membrane. However, how newly synthesized APP arrives at those membrane compartments in synapses, neurites, and somatodendritic areas of neurons remains understudied.
Early investigations using non-polarized cells and biochemistry methods reported that APP was first transported to the plasma membrane, where the majority was cleaved by α-secretase (α-Sec), and the remaining was internalized and cleaved by β-secretase (β-Sec)
8-10
. However, this model of APP
trafficking and cleavage has not been investigated as thoroughly in the neurons of the central nervous system (CNS). Some reported that only a small fraction of neuronal APP reaches the plasma membrane 11. Even if most APP is first transported to the plasma membrane before being cleaved, as shown in non-neuronal cells, there are still two different trafficking routes to follow: it can be sent to the somatodendritic plasma membrane right after synthesis and propagated along the neurite surface or it can be internalized and intracellularly transported to distal neurites before surface presentation. Evidence for both surface and intracellular transportation has been observed, but it remains unclear which is more prominent and more relevant to amyloidogenesis
12-14
. It is well demonstrated that a
considerable amount of APP is cleaved by α-Sec, which is predominantly localized in the plasma membrane
9, 15
. Consequently, a shift of APP distribution from cell surface to intracellular membranes 3
ACS Paragon Plus Environment
ACS Chemical Neuroscience
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 21 47 48 22 49 50 23 51 52 53 24 54 55 25 56 57 26 58 59 60
Page 4 of 27
may result in an increase of amyloidogenic processing, a potential etiological factor in AD
16
.
Accordingly, it has been shown that irregular APP trafficking, specifically an increase in intracellular retention
11
and a decrease in exportation to the plasma membrane
17-18
, cause increased Aβ
production and therefore could contribute to more amyloidogenesis. Therefore, it is crucial to know if and how nascent APP is transported between surface and intracellular membranes and where it is processed by α-Sec.
Recently, Das et al. utilized fluorescent APP fusion proteins to illustrate the intracellular trafficking of nascent APP
19
, which, along with a few other recent reports
20
, has demonstrated the power of
fluorescence live-cell imaging for studying APP in CNS neurons. Interestingly, they have shown that while a considerable amount of newly synthesized APP co-localizes with BACE1 (β-site APP cleaving enzyme 1) in the endoplasmic reticulum and Golgi at somatodendritic areas19, which is in conflict with the reports that surface cleavage by α-Sec is the major proteolytic route
10, 21
. However, it has also
been shown that APP can be co-transported with BACE1 without being cleaved
21
. All of these
illustrate the complexity of APP trafficking and processing in CNS neurons and demonstrate the need for further investigation of the APP trafficking and processing between surface and intracellular membranes.
To solve those problems, live-cell fluorescence imaging is the most practical option as it offers adequate spatiotemporal resolution to deal with morphologically complex neurons. We started with an APP fusion protein with a pH-sensitive fluorescent protein at its N-terminal. It could measure surface and intracellular APP because of the pH gradient between extracellular space and intracellular compartments like endosomes and lysosomes. We added a pH-insensitive BFP2 at its C-terminal to obtain an independent measurement of APP and its C-terminal fragments (CTFs). Using it, we discovered that a considerable amount of nascent APP is trafficked to the surface at somatodendritic areas and propagates to distal neurites along the neuronal surface. The APP in the plasma 4
ACS Paragon Plus Environment
Page 5 of 27
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 21 47 48 22 49 50 23 51 52 53 24 54 55 25 56 57 26 58 59 60
ACS Chemical Neuroscience
membrane is subject to removal by α-Sec cleavage and dynamin-dependent endocytosis. Our findings demonstrate that surface presentation and cleavage are major routes for newly synthesized APP in neurons.
Results APP is a single transmembrane protein with its N-terminal in extracellular/luminal space and Cterminal in cytosolic space. Adding pHluorin, a pH-sensitive green fluorescent protein, to its Nterminal (pH-APP) allowed Groemer et al to visualize the surface turnover of APP in live neurons
20
.
However, pH-APP alone cannot determine if decrease of pHluorin fluorescence is due to APP internalization or cleavage by α-Sec in the plasma membrane. To monitor APP proteolysis, especially N-terminal cleavage by α-Sec or β-Sec, we added a monomeric pH-insensitive blue fluorescent protein 2 (BFP2) to APP’s C-terminal to make pH-APP-BFP2 (Figure 1a), which enables the ratiometric measurement (i.e. pHluorin vs. BFP2 ratio) of APP’s N-terminal cleavage (Figure 1a). For the purpose of comparison, we co-expressed SypHTm
22
, a pH-sensitive mCherry (a.k.a. pHTomato,
pHTm) fused to the ectodomain of Synaptophysin which is abundant in synaptic vesicles (SVs). We chose SypHTm because APP is believed by some to be enriched in SVs
23
. To achieve better co-
expression, we used a viral T2A linker that can cause ribosome skipping and thus produces two separate proteins from one piece of mRNA. T2A provides better stoichiometry of co-expression than the commonly used IRES (internal ribosome entry site) 24. We inserted the SypHTm coding sequence in front of T2A and pH-APP-BFP2 after (i.e. SypHTm:T2A:pH-APP-BFP2) because SypHTm overexpression was less of a concern than that of pH-APP-BFP2. To further reduce overexpression artifacts, the whole construct was driven by human Synapsin I promoter, known to have a moderate and neuron-specific expression in rat hippocampal culture25. Due to pH-sensitivity, changes in pHluorin or pHTm fluorescence can report the surface turnover of APP and Synaptophysin, respectively, whereas changes in BFP2 fluorescence reflect only APP synthesis or degradation (e.g. by γ-secretase cleavage) (Figure 1a). To validate the predicted fluorescence changes, we applied 5
ACS Paragon Plus Environment
ACS Chemical Neuroscience
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 21 47 48 22 49 50 23 51 52 53 24 54 55 25 56 57 26 58 59 60
Page 6 of 27
Tyrode’s solution with normal pH7.35, 50mM NH4Cl (neutralizing intracellular luminal pH) and pH5.5. As expected, 50mM NH4Cl increased pHluorin and pHTm fluorescence whereas pH5.5 solution reduced both (Figure 1b). With 50mM NH4Cl, we were able to easily identify individual processes of transfected neurons. BFP2’s fluorescence intensity exhibited no response to pH manipulation (Figure 1b). Notably, SypHTm is not fully quenched even in acidic subcellular compartments or when exposed to low pH solutions (Figure 1b), so along with the pH-insensitive BFP2, it can be used to identify transfected cells without the need of pH manipulation. Notably, the BFP2 signal was significantly weaker than pHluorin because BFP2 fluoresces about 53% less than pHluorin, its excitation and emission were less transmittable through microscope optics, and its blue fluorescence was ~40% less detectable by EMCCD (electron multiplying charge coupled device) or sCMOS (scientific complementary metal-oxide-semiconductor) camera we use (Figure S1).
Next, we tested if pH-APP-BFP2 recapitulates endogenous APP in terms of its surface and intracellular distribution and cleavage. To do so, we conducted live-cell imaging of pHluorin, BFP2, and pHTm and immunostaining of endogenous APP in parallel. To study nascent APP, we focused on somatodendritic areas where APP is synthesized. The total APP (i.e. pHluorintotal) is calculated by the difference in pHluorin fluorescence between NH4Cl-containing and pH5.5 Tyrode’s solutions, the surface APP (i.e. pHluorinout) by the difference between normal and pH5.5 Tyrode’s solutions, and the intracellular APP (pHluorinin) by the difference between NH4Cl-containing and normal Tyrode’s solutions. To mitigate the variation of protein expression among different cells, we calculated the surface and intracellular APP as fractions of the total APP (i.e. pHluorin out/total and pHluorin in/total respectively). Furthermore, we estimated the ratio of holo-APP vs. both APP and its C-terminal fragments (CTFs) by pHluorintotal/BFP2, which is influenced by the secretase-mediated cleavage of pH-APP-BFP2. Our data showed that a considerable amount of pH-APP-BFP2 resides on the surface membrane at somatodendritic areas, as shown by sample images in Figure 2a and plots in Figure 2b. Next, we used three separate anti-APP primary antibodies to detect endogenous APP. We 6
ACS Paragon Plus Environment
Page 7 of 27
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 21 47 48 22 49 50 23 51 52 53 24 54 55 25 56 57 58 59 60
ACS Chemical Neuroscience
labeled surface APP with one antibody selectively recognizing APP’s N-terminal epitope before cell membrane permeabilization, detected total APP with a second antibody specifically against APP’s ectodomain different from the previous one, and measured total APP plus CTFs with a third antibody highly selective to a C-terminal epitope. All antibody information can be found in Supplementary Table 1, and the latter two antibodies were applied after cell membrane permeabilization in order to label intracellular APP. Consistently, our triple immunostaining and quantitative fluorescence imaging of endogenous APP supported our live-cell imaging results (Figure 2c&d). Notably, the drastic difference between pHluorintotal/BFP2 (via live cell imaging) and APP N/C (via immunostaining) was most caused by technological differences like BFP2 fluorescence and immunolabeling efficacy, which makes it inappropriate to directly compare to the fraction of pHluorin out/total. Nevertheless, together these two sets of results suggest that at least one third of APP is directly transported to the plasma membrane after synthesis. To confirm if the pH-APP-BFP2 faithfully represents the trafficking and processing of endogenous APP, we blocked α-Sec using a selective inhibitor (GI 254023X)
26-27
. As
expected, this manipulation significantly increased holo-APP measured by both live-cell and immunofluorescence imaging (i.e. augments in pHluorintotal/BFP2 and APP-Ntotal/APP-C) (Figure 2), confirming that pH-APP-BFP2 is cleaved like endogenous APP. While one would think that blocking α-Sec would increase surface APP, we observed decreases in both the surface fractions of pH-APPBFP2 and the ratio of out vs total endogenous holo-APP (Figure 2), although the result was not significant in the former (likely due to the large variance of autofluorescence from neuronal somas). Nevertheless, the similarity in the direction and the degree of change (28% vs. 46% decrease via live cell imaging and immunostaining, respectively) vindicated our conclusion. In fact, previous studies about APP trafficking in non-neuronal cells have suggested that the amount of cell surface APP is strictly regulated to prevent aberrant increase28-31. Our results from the two different methods suggest that surface APP spared by α-SI was mostly internalized, which is in good agreement with the idea that cell surface APP is homeostatically regulated. Furthermore, the consistency between the two
7
ACS Paragon Plus Environment
ACS Chemical Neuroscience
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 21 47 48 22 49 50 23 51 52 53 24 54 55 25 56 57 26 58 59 60
Page 8 of 27
different sets of results validated that pH-APP-BFP2 offers a reliable representation of endogenous APP.
Dual-fluorescence tagging of APP combined with long-term live-cell imaging provides us a unique opportunity to investigate the fate of nascent APP. This is an important research topic because APP has been widely implicated in neuronal survival2, axon growth and pruning32, and synaptogenesis3, and because the disruption of these functions is linked to AD33. Here, we mounted hippocampal cultures in a weather-controlled station immediately after transfection, and continuously monitored the fate of newly expressed pH-APP-BFP2 as well as SypHTm by time-lapse imaging of pHluorin, BFP2, and pHTm for up to 24 hours (Figure 3a&b and Supplementary Video 1). Notably, pHluorin signal largely represented pH-APP-BFP2 residing in the plasma membrane because pHluorin is nearly completely quenched in acidic intracellular membrane compartments34. We observed that: (1) pHluorin and BFP2 signals appeared nearly simultaneously at 5 hours after transfection, whereas pHTm signal came up a couple of hours later likely due to the longer maturation time for pHTm
22, 35-36
(Figure 3c-e); (2) while neuronal nuclei remained relatively pHluorin- and pHTm-free during imaging, BFP2 showed up in nuclei a few hours after its cytosolic appearance, which likely represents the nuclear entrance of the AICD (APP intracellular domain) (Supplementary Video 1); (3) only the pHluorin signal exhibited a transient increase followed by a partial decrease at somatodendritic areas (Figure 3c-e); (4) BFP2 fluorescence began to level off after 10 hours whereas pHTm fluorescence continued to rise for at least 20 hours (Figure 3d&e); (5) the ratio of pHluorin to BFP2 peaked, then declined, first in the soma, then in the immediately proximal stretch of neurites, and then further down to distal neurites (Figure 3f); (6) in a larger, separate data set (DMSO controls in Figure 4b), at the somatodendritic areas, pHluorin/BFP2 was 1.662 ± 0.284 (n = 7) during the 8-10th hour period, significantly higher than that of 4-6th hour period (0.967 ± 0.071, n = 7; Tukey's multiple comparisons test, p = 0.0205) and that of 16-18th hour period (0.853 ± 0.134, n = 4; Tukey's multiple comparisons test, p = 0.0214). These observations indicate that, unlike synaptic vesicle proteins such as 8
ACS Paragon Plus Environment
Page 9 of 27
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 21 47 48 22 49 50 23 51 52 53 24 54 55 25 56 57 58 59 60
ACS Chemical Neuroscience
Synaptophysin (i.e. SypHTm), (a) a substantial amount of nascent APP is first transported to the surface of somatodendritic areas; (b) APP transportation to the plasma membrane eventually reaches an equilibrium; (c) APP in the plasma membrane propagates to distal neurites via lateral diffusion, although the possibility of additional transportation via intracellular cargo vesicles cannot be excluded; and (d) APP in the plasma membrane is continuously removed while moving along the plasma membrane. Given pHluorin’s N-terminal position and pH-sensitivity, two mechanisms can be in play for such removal: endocytosis-mediated APP internalization, N-terminal cleavage by α-Sec, or both.
To evaluate the contribution of those two mechanisms, we pharmacologically inhibited endocytosis or α-Sec activity. For the former, we inhibited dynamin because almost all endocytosis is dynamindependent. We chose Dyngo-4a, which is six times more potent than the popular Dynasore and does not interfere with dynamin-independent membrane trafficking
37
. Using long-term time-lapse imaging,
we found that Dyngo-4a eliminated the decrease of pHluorin/BFP2 ratio with little effect on the transient increase and that the ratio reached a steady state at around 12 hours after transfection (Figure 4a). Notably, this ratio during the steady state was significantly higher than that of DMSO control (Figure 4b). This result suggested that endocytosis-mediated APP internalization contributes to the homeostatic regulation of APP in the plasma membrane. To study the contribution of α-Sec, we applied the previously tested α-SI. In contrast to Dyngo-4a, the inhibitor reduced the transient increase of pHluorin/BFP2 ratio, and it reached equilibrium earlier than the other two conditions (Figure 4a). Notably, the ratio at the steady state was similar to that of the DMSO control (Figure 4b). To test if either of the two mechanism(s) is generic to any neuronal membrane proteins, we analyzed the change of SypHTm. Unlike pH-APP-BFP2, SypHTm signal exhibited continuous increase up to the end of the long-term imaging, and this increase was not affected by either drug (Figure 4c). Based on the outcomes of the two pharmacological interventions, we conclude that both dynamin-dependent endocytosis and α-Sec cleavage contribute to the homeostatic control of surface
9
ACS Paragon Plus Environment
ACS Chemical Neuroscience
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 21 47 48 22 49 50 23 51 52 53 24 54 55 25 56 57 26 58 59 60
Page 10 of 27
APP. Between the two mechanisms, the former is responsible for the internalization of excessive APP whereas α-Sec has little effect in determining the amount of plasma membrane APP at steady state.
Discussion In this study, we generated a dual-fluorescence APP fusion protein and utilized long-term time-lapse imaging to study the fate of nascent APP in live neurons. Our results illustrate that a substantial amount of newly synthesized APP is directly transported to the plasma membrane at the somatodendritic areas and propagates towards the distal direction in the plasma membrane. Moreover, neuronal surface APP is regulated by α-Sec cleavage and endocytosis together. Methodologically, we demonstrate that our dual-fluorescence APP fusion protein is a useful tool for studying APP trafficking and processing. Its combination with long-term time-lapse imaging is a powerful tool to investigate the fate of APP in polarized and morphologically complicated neurons, which are the most relevant cell type for AD research. Furthermore, the dual-tagged pH-APP-BFP2 can be a useful probe for other assays like fluorescence recovery after photobleaching (FRAP) because only unquenched (i.e in the plasma membrane) pHluorin will be bleached.
Accumulated evidence suggests that APP’s surface and intracellular distributions are tightly regulated and they influence APP’s proteolytic products because of differentially compartmentalized secretases9, 38. Furthermore, APP localization is also essential for its stipulated functions, e.g. being a receptor ligand requires its surface presentation. Therefore, when and how APP is transported to different membrane compartments are both physiologically and pathologically crucial. The abilities to distinguish surface and intracellular APP by relative pHluorin signals and to estimate its cleavage by pHluorin/BFP2 ratio make pH-APP-BFP2 a versatile reporter for multiple aspects of APP behavior such as trafficking and processing. The best example in our results is the finding that α-Sec cleavage, commonly occurring at the neuronal surface, caused a decrease instead of an increase in surface APP fraction, implying that the control of surface APP concentration is stricter than that of intracellular 10
ACS Paragon Plus Environment
Page 11 of 27
1
1 2 32 4 53 6 74 8 9 5 10 11 12 6 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 28 13 29 30 14 31 32 15 33 34 35 16 36 37 17 38 39 18 40 41 42 19 43 44 20 45 46 21 47 48 22 49 50 51 23 52 53 24 54 55 25 56 57 58 59 60
ACS Chemical Neuroscience
APP. Using this probe, it is now possible to study how changes in the activities of other secretases such as β- and γ-secretases affect APP distribution in different membrane compartments. Combined with reporters for synaptic activity (i.e. SypHTm), it is also possible to evaluate if and how APP trafficking and processing are associated with neuronal firing at synaptic and non-synaptic regions.
Here, we focused on the fate of nascent APP. Long-term time-lapse imaging enabled us to track pHAPP-BFP2 immediately after its synthesis, which is a less studied aspect of APP. Available reports offered opposite theories: APP is either cleaved right after synthesis or is transported to distal neurites before cleavage12. While it is possible that nascent APP directly encounters β-Sec in intracellular compartments
11, 21
, our data have shown that a large amount of APP clearly follows the
regular route of plasma membrane proteins first being transported to the plasma membrane at the somatodendritic areas. In addition to the well-studied anterograde transportation via cargo vesicles (e.g. polarized axonal transport along microtubules
39
40
), we have found that it also laterally diffuses
along the neuronal surface to the processes. Furthermore, APP in the plasma membrane is also partially consumed by α-Sec and/or internalized through endocytosis, which are likely two of the constitutive mechanisms regulating the amount of APP in the plasma membrane. Based on our observations and its global expression, we speculate that APP is a housekeeping protein important for the maintenance of cell membranes. Given that APP has a cholesterol-binding motif partially overlapping with its transmembrane domain
41
, given that neuronal surface membrane contains more
cholesterol than most cells in the body do, and given that cholesterol has been implicated in AD through genetic risk factors like ApoE4, we speculate that APP may be part of the regulatory mechanism for neuronal membrane cholesterol. While our study mostly used young neurons, major routes of APP trafficking and processing likely remain the same in aging neurons 42.
11
ACS Paragon Plus Environment
ACS Chemical Neuroscience
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 21 47 48 22 49 50 23 51 52 53 24 54 55 25 56 57 26 58 59 60
Page 12 of 27
Methods DNA recombination The mammalian expression construct containing pH-APP and driven by human synapsin1 promoter was from Dr. Jürgen Klingauf Yulong Li
20
. The plasmid containing SypHTm, T2A linker and BFP2 was from Dr.
22
. The SypHTm-encoding cDNA sequence and the T2A linker were amplified using 2
primers,
5’-
CGTGCCTGAGAGCGCAGTCGAATTAGCTTGGTACCATGGACGTGGTGAATCAGCTGGTGG-3’ and 5’-CCAGGCTGGGCAGCATGGTGGCGGCGGATCCAGGGCCGGGATTCTCCTCCACGTCAC3’ (forward and reverse respectively). The amplicon was inserted to the C-terminal of pHluorin-APP before the stop codon using the Gibson Assembly Cloning Kit (NEB)
43
. The resulting plasmid,
SypHTm:T2A:pH-APP-BFP2 was verified by DNA sequencing.
Cell Culture and Transfection All animal procedures were approved by the Vanderbilt University Animal Care and Use Committee. Rat postnatal hippocampal cultures were prepared from P0 to P1 Sprague-Dawley rats using an established protocol
44
with slight modifications. Dissociated cells were plated onto 12mm-ø glass
coverslips (~200,000 cells/mL) coated with Matrigel (Life Technologies) and all coverslips were placed in 24-well plates (ThermoScientific). with plating media consisting of Minimal Essential Medium (MEM, Life Technologies) and (in mM) 27 glucose, 2.4 NaHCO3, 0.00125 transferrin, 2 Lglutamine, 0.0043 insulin and 10%/vol fetal bovine serum (FBS, Gibco). After 1-2 day recovery and growth, an equal volume of media containing (in mM) 27 glucose, 2.4 NaHCO3, 0.00125 transferrin, 0.5 L-glutamine, 2 Ara-C, 1 %/vol B27 supplement (Life Technologies) and 5 %/vol FBS was added. Astroglia proliferation was inhibited by Ara-C, a mitosis inhibitor. Typically, Ca3(PO4)2 transfection was performed at 8-9 days. In all time lapse imaging experiments, dissociated cells were plated onto glass bottom, 4-compartment 35/10 mm cell culture dishes (Greiner Bio-one) and were transfected at 11-16 days at the start of the experiment. 12
ACS Paragon Plus Environment
Page 13 of 27
ACS Chemical Neuroscience
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 21 47 48 22 49 50 23 51 52 53 24 54 55 25 56 57 26 58 59 60
Immunocytochemistry In order to isolate different populations of APP using immunocytochemistry, three separate APP antibodies were applied to the same samples. Two different N-terminal APP antibodies with no overlapping epitopes were selected for surface and total N-terminal labeling. After treatments, coverslips were fixed using 4% paraformaldehyde in PBS for 30 minutes and blocked with horse serum and BSA in PBS for at least one hour in the absence of detergents. Then, the coverslips were treated with goat anti-APP-N overnight at 4°C, washed, and treated with an anti-goat secondary antibody for one hour at room temperature. After at least 5 washes to eliminate the anti-goat secondary entirely, the cells were permeabilized with 0.25% Triton X-100 for 10 minutes. The coverslips were blocked for one additional hour in goat serum, BSA, and Triton X-100 solution at room temperature. Next, they were incubated with the two other APP antibodies (22C11 and Y188) and guinea pig anti-Synaptophysin at room temperature for at least one hour, washed, and treated with three fluorophore-labeled goat secondary antibodies (see Supplementary Table 1, 1: 1000 dilution for all, Life Technologies or Biotium) at room temperature for one hour before mounting.
Imaging and analysis Except for long term time lapse imaging, most samples were imaged on a Nikon Eclipse Ti inverted microscope with a 100X Plan Apo VC objective (N.A. 1.40) and an EMCCD camera (Andor). For Alexa 488, 568, and 647 fluorescence, we used the following filter sets (Semrock) respectively: Ex 405/20X, DiC 425LP and Em 460/50; Ex 460/50, DiC 495LP and Em 535/25; Ex 565/25, DiC 585LP and Em 644/90; Ex 644/10 DiC 660LP and Em 710/50. All imaging conditions, including the intensity of input light source, exposure time and EM gain, were kept the same among different treatment groups. For live-cell imaging, individual coverslips were mounted in an RC-26G imaging chamber (Warner Instruments) with a 24x40 mm size 0 cover glass (Fisher Scientific) sealing the bottom. The chamber 13
ACS Paragon Plus Environment
ACS Chemical Neuroscience
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 21 47 48 22 49 50 23 51 52 53 24 54 55 25 56 57 26 58 59 60
Page 14 of 27
was inserted into a PH-1 platform (Warner Instruments) which was mounted on a motorized stage. The perfusion bath had a constant rate of ~50 µL/sec, and solution exchange was controlled by a VC6 valve control system and a 6-channel manifold (Warner Instruments). Imaging and solution exchange were controlled via Micro-manager. In each experiment, the same imaging settings were used for the same fluorophore, regardless of treatments and sample batches. For BFP2, pHluorin and pHTm, we used the following filter sets (Semrock) respectively: Ex 405/20X, DiC 425LP and Em 460/50; Ex 480/20X, DiC 495LP and Em 535/40; Ex 560/40M, DiC 585LP and Em 610/20nm BP. Samples were perfused with pH 7.35, 50mM NH4Cl-containing and pH5.5 Tyrode’s solutions sequentially. Tyrode’s saline is made of (in mM): 150 NaCl, 4 KCl, 2 MgCl2, 2 CaCl2, 10 N-2 hydroxyethyl piperazine-n-2 ethanesulphonic acid (HEPES), 10 glucose, pH 7.35 or pH 5.5. The 50 mM NH4Cl solutions was made by substituting for NaCl equimolarly, pH 7.35. All solutions contained 10 µM NBQX and 20 µM D-AP5. The live cell soma imaging experiments (Figure 3a-b) were performed in a similar manner using a spinning disk confocal microscope equipped with the same perfusion system as the Eclipse Ti and a LUMPlanFl 40x/0.80w objective. 405, 480 and 561nm lasers paired with the emission filters Et 460/50m, Et525/50m and Et605/52m were used to image BFP2, pHluorin and pHTomato, respectively. For long-term time-lapse imaging, we used an Olympus IX-81 microscope equipped with an ASI motorized xy stage, Mightex LED light source, an Olympus 10X Ph1/Fl (N.A. 0.10) objective and a Flash 4.0 sCMOS camera (Hamamatsu). Cells were grown in 4 compartment glass bottom 35/10 mm dishes (Greiner Bio-one), transfected and treated immediately prior to the experiment. The dish was placed in a sealed microscope stage chamber. A customized weather station enclosure maintained a constant temperature of 37°C. 5%CO2 air was supplied to the sealed stage after bubbling through a H2O tank inside the weather station in order to maintain 5% CO2 and saturated humidity in the sealed imaging chamber. The image acquisition started approximately 1 hour after the replacement of regular media after transfection and addition of the drugs. 4-5 fields of view were chosen per quadrant in areas of high neuronal density without the knowledge of fluorescent protein expression. Image acquisition was controlled via Micro-Manager45 14
ACS Paragon Plus Environment
Page 15 of 27
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 10 22 23 11 24 25 12 26 27 13 28 29 30 14 31 32 15 33 34 16 35 36 37 17 38 39 18 40 41 19 42 43 44 20 45 46 21 47 48 22 49 50 23 51 52 53 24 54 55 25 56 57 58 59 60
ACS Chemical Neuroscience
(UCSF). Phase contrast images as well as blue, green and red fluorescence images for every field of view were taken every 30 minutes. For image analysis, Fiji
46
, a distribution of ImageJ
47
, was used. For immunofluorescence and live-
cell imaging, soma ROIs were hand drawn in areas with a relatively low background based on APP. Nearby regions, either relatively cell free or on top of glia as appropriate for each individual soma, were selected as background regions. Mean values were exported from Fiji and background subtraction was performed using Microsoft Excel. Prism 7.03 was used to generate all plots and perform all statistical analysis. For long-term time-lapse imaging, image stacks were corrected for stage drift using the FIJI plugin called StackReg48. For Figure 3, one soma with three processes (divided into 100uM segments) with very few morphology changes was chosen, and the same ROIs were used for every frame. Using the MultiStackReg v1.45 plug-in (written by Brad Busse), pHTm images were used to align all stacks for the same field of view because of the higher background in the other channels. For Figure 4, soma ROIs were hand-drawn frame by frame based on phasecontrast images because morphology changes to the soma shape and size was too common to be a basis for exclusion. For these experiments, the background was defined by a single ROI near the cell to account for uneven illumination and, in some cases, auto fluorescence from untransfected glia. Due to the high variability in background and fluorescence interference by Dyngo4a, the pHluorin images only were corrected before being quantified (“Float the stack” macro, Kenton Arkill). In all figures, the same contrast settings were used for each fluorophore in each experiment across conditions. However, some images were processed by simple subtraction of the average background (e.g. Figure 2) so that the use of the same contrast reflects the differences between conditions more accurately. In Figure 2A, outlier pixels were removed from the BFP2 images in FIJI for presentation purposes only but never for quantification.
Statistics
15
ACS Paragon Plus Environment
ACS Chemical Neuroscience
1
1 2 32 4 53 6 74 8 9 5 10 11 6 12 13 14 7 15 16 8 17 18 9 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 16 of 27
Normalization methodology was determined based on the experimental question. In Figure 3, pHluorin/BFP2 ratio was normalized to the maximum and minimum because the purpose of the experiment was to compare the kinetics of the rise and fall at different regions of the same cells, whereas in Figure 4, the ratio was normalized to the baseline for the purpose of comparing the magnitude of the change in the ratio itself after different treatments. Statistical tests were performed using Graphpad Prism 7.03. If not specified, two-tailed t-tests were used to compare two conditions, one-way ANOVA and Dunnett's multiple comparisons test were used to compare the control condition to multiple other conditions, and two-way ANOVA and Sidak’s multiple comparisons test was used to compare results within one variable in experiments with two variables.
16
ACS Paragon Plus Environment
Page 17 of 27
1
1 2 32 4 53 6 74 8 95 10 11 12 6 13 14 7 15 16 17 8 18 19 9 20 21 10 22 23 24 11 25 26 27 12 28 29 30 13 31 32 33 14 34 15 35 36 37 16 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Neuroscience
Acknowledgements We thank K. E. Kitko and R.M. Lazarenko for technical help, sharing cell cultures, and discussion. We thank C.E. Strothman and all other members of Zhang lab for comments and discussions. The pHluorin-APP was generously provided by J. Klingauf lab. This work is funded by the National Institutes of Health (OD00876101 and NS094738 to Q.Z.).
Author Contributions C.E. DelBove and Q. Zhang conceived this project, designed experiments, analyzed images and wrote the manuscript. C.E. DelBove conducted all experiments and analyzed the data. X. Deng participated in some experiments.
Declaration of Interests All authors declare no competing financial interests.
Supplementary Information: The supplementary materials including supplementary table and video.
17
ACS Paragon Plus Environment
ACS Chemical Neuroscience
1
1 2 32 4 53 6 74 8 95 10 11 6 12 13 7 14 15 8 16 17 9 18 19 20 10 21 22 11 23 24 12 25 26 13 27 28 14 29 30 31 15 32 33 16 34 35 17 36 37 18 38 39 19 40 41 20 42 43 21 44 45 22 46 47 23 48 49 50 24 51 52 25 53 54 26 55 56 57 27 58 59 60
Page 18 of 27
Figure Legends Figure 1. Multi-fluorescence reporter for imaging APP in live neurons. a) The upper schematic shows the linear structure of the SypHTm:T2A:pH-APP-BFP2 plasmid introduced to cultured neurons for imaging. hpSyn1, the human Synapsin I promoter; SypHTm, Synaptophysin-pHTomato; T2A, thosea asigna virus 2A peptide; SP, APP signal peptide; pHluorin, pH-sensitive green fluorescent protein; APP 695, rat neuronal amyloid precursor protein isoform with 695 amino acid residues; Aβ, β-amyloid peptide; BFP2, blue fluorescent protein 2. The lower diagram indicates the simplified trafficking and processing routes of the fluorescent reporters and the impact of pH on their fluorescence. b) Sample images of pHluorin, pHTm, BFP2, and overlay at normal, 50 mM NH4Cl-containing, and pH5.5 Tyrode’s solutions, demonstrating the impact of pH on their fluorescence. Scale bar, 10 µm.
Figure 2. Fluorescence analysis of surface and intracellular membrane APPs. a) Sample pHluorin images for two pH conditions, and BFP2 images. Cultures were treated by DMSO control or α-secretase inhibitor (αSI). Images depicted are background-subtracted. Scale bars, 50 µm. b) The average somatodendritic surface fraction of pHluorin (pHluorin out/total), and total pHluorin vs. BFP2 (pHluorintotal/BFP2) after DMSO and α-SI treatments. Surface vs total pHluorin was decreased by 28.1% in the α-SI group. However, an unpaired, twotailed t-test failed to detect significance (p = 0.2966). Total pHluorin vs. BFP2 was increased by 32.0% in α-SI group, which also fails to reach significance (unpaired, two-tailed t-test, p = 0.6137). For DMSO, n = 6 and for α-SI, n = 9, where n is the number of somas. c) Sample images of triple immunofluorescence labeling for surface APP (APP-Nout), total APP (APP-Ntotal) and APP C-terminal (APP-C) after DMSO and α-SI treatments. Images depicted are background-subtracted. Scale bars, 50 µm. d) The average somatodendritic immunofluorescence ratios of surface vs. total APP (APP-N out/total), and total APP N vs. C-terminals (APP N/C) after DMSO and α-SI treatments. Surface vs total APP was significantly decreased by application of α-SI according to an unpaired, two-tailed t-test (*, p = 0.0152). Total APP-N vs. APP-C was significantly increased by application of α-SI (unpaired, two-tailed t-test, *, p = 0.0397). For DMSO, n = 6 and for α-SI, n = 5, where n is the number of somas.
18
ACS Paragon Plus Environment
Page 19 of 27
1
1 22 3 4 3 5 6 74 8 95 10 6 11 12 7 13 14 15 8 16 17 9 18 19 20 10 21 22 11 23 24 12 25 26 27 13 28 29 14 30 31 15 32 33 16 34 35 17 36 37 18 38 39 19 40 41 20 42 43 21 44 45 46 22 47 48 23 49 50 24 51 52 25 53 54 26 55 56 57 27 58 59 60
ACS Chemical Neuroscience
Figure 3. Surfacing and propagation of nascent APP at somatodendritic areas. a) Snapshots of pHluorin fluorescence from a time-lapse video started after transfection. Scale bar, 100 µm. b) Sample images of pHluorin, BFP2, pHTomato and overlay at 16 hours after transfection. Scale bar, 100 µm. c-e) Normalized fluorescence changes of pHluorin, BFP2, and pHTomato at the soma and at 1-100 and 101-200 µm from the neuronal soma over time. f) Normalized ratio of pHluorin vs. BFP2 fluorescence at the soma and at 1-100 and 101-200 µm from the soma over time. The gray solid line illustrates that the peak is earlier in ROIs closer to the soma. For c-f, 3 processes divided into two segments, which are 1-100 and 101-200 µm away from the soma, are averaged).
Figure 4. The effects of α-SI and Dyngo4a on nascent APP trafficking. a) Changes in pHluorin vs. BFP2 ratio at representative somas over time after treatments of DMSO, α-secretase inhibitor (α-SI) and Dyngo4a (a dynamin inhibitor). Each cell is normalized to the 6th hour, when all fluorescent signals are significantly and reliably above cell autofluorescence and background. b) Average pHluorin vs. BFP2 ratios at 4-6, 8-10 and 1618 hour time periods after treatments of DMSO, α-SI and Dyngo4a. Ordinary two-way ANOVA detected effects based on time (F (2, 46) = 4.932, p = 0.0115), treatment (F (2, 46) = 4.412, p = 0.0177) and an interaction (F (4, 46) = 3.316, p = 0.0181). At time 4-6 hours, when all fluorescent reporters are reliably identifiable, Tukey’s multiple comparisons test detected no significant differences between treatments (Dyngo4a vs. α-SI, ns, p > 0.9999; Dyngo4a vs. DMSO, ns, p = 0.9687; α-SI vs. DMSO, ns, p = 0.959). By 8-10 hours, there is a significant difference between DMSO and the α-SI (*, p = 0.0401) only (Dyngo4a vs. α-SI, ns, p = 0.439; Dyngo4a vs. DMSO, ns, p = 0.6036). At 16-18 hours, there is a significant difference between the Dyngo4atreated cells and the other cells (Dyngo4a vs. DMSO, **, p = 0.009, Dyngo4a vs. α-SI, **, p = 0.0049) but there is no longer any difference between DMSO and α-SI (ns, p = 0.9698). c). Changes of normalized pHTomato over time after treatments of DMSO, α-SI and Dyngo4a. The fluorescence of every cell is normalized to the minimum and maximum signal. Shadows represent s.e.m. At 4-6 and 8-10 hours: for DMSO, n = 7, for α-SI, n = 10, and for Dyngo4a, n = 5, where n is the number of somas. By 16-18 hours, for DMSO, n = 4, for α-SI, n = 4, and for Dyngo4a, n = 3.
19
ACS Paragon Plus Environment
ACS Chemical Neuroscience
1
1 2 32 43 54 6 5 7 6 8 97 8 10 9 11 12 10 13 11 14 12 15 13 16 14 17 15 18 16 19 17 20 18 21 22 19 23 20 24 21 25 22 26 23 27 24 28 25 29 26 30 31 27 32 28 33 29 34 30 35 31 36 32 37 33 38 34 39 35 40 41 36 42 37 43 38 44 39 45 40 46 41 47 42 48 43 49 50 44 51 45 52 46 53 47 54 48 55 49 56 50 57 58 59 60
Page 20 of 27
References: 1. Han, K.; Müller, U. C.; Hülsmann, S., Amyloid-precursor Like Proteins APLP1 and APLP2 Are Dispensable for Normal Development of the Neonatal Respiratory Network. Frontiers in Molecular Neuroscience 2017, 10 (189). 2. Wang, S.; Bolós, M.; Clark, R.; Cullen, C. L.; Southam, K. A.; Foa, L.; Dickson, T. C.; Young, K. M., Amyloid β precursor protein regulates neuron survival and maturation in the adult mouse brain. Molecular and Cellular Neuroscience 2016, 77, 21-33. 3. Southam, K. A.; Stennard, F.; Pavez, C.; Small, D. H., Knockout of Amyloid β Protein Precursor (APP) Expression Alters Synaptogenesis, Neurite Branching and Axonal Morphology of Hippocampal Neurons. Neurochemical Research 2018. 4. Igbavboa, U.; Sun, G. Y.; Weisman, G. A.; He, Y.; Wood, W. G., Amyloid beta-protein stimulates trafficking of cholesterol and caveolin-1 from the plasma membrane to the Golgi complex in mouse primary astrocytes. Neuroscience 2009, 162 (2), 328-38. 5. Liu, Q.; Zerbinatti, C. V.; Zhang, J.; Hoe, H. S.; Wang, B.; Cole, S. L.; Herz, J.; Muglia, L.; Bu, G., Amyloid precursor protein regulates brain apolipoprotein E and cholesterol metabolism through lipoprotein receptor LRP1. Neuron 2007, 56 (1), 66-78. 6. Pierrot, N.; Tyteca, D.; D'Auria, L.; Dewachter, I.; Gailly, P.; Hendrickx, A.; Tasiaux, B.; Haylani, L. E.; Muls, N.; N'Kuli, F.; Laquerrière, A.; Demoulin, J.-B.; Campion, D.; Brion, J.-P.; Courtoy, P. J.; Kienlen-Campard, P.; Octave, J.-N., Amyloid precursor protein controls cholesterol turnover needed for neuronal activity. EMBO Molecular Medicine 2013, 5 (4), 608-625. 7. Wang, W.; Mutka, A. L.; Zmrzljak, U. P.; Rozman, D.; Tanila, H.; Gylling, H.; Remes, A. M.; Huttunen, H. J.; Ikonen, E., Amyloid precursor protein alpha- and beta-cleaved ectodomains exert opposing control of cholesterol homeostasis via SREBP2. FASEB J 2014, 28 (2), 849-60. 8. Thinakaran, G.; Koo, E. H., Amyloid Precursor Protein Trafficking, Processing, and Function. Journal of Biological Chemistry 2008, 283 (44), 29615-29619. 9. Andrew, R. J.; Kellett, K. A.; Thinakaran, G.; Hooper, N. M., A Greek Tragedy: the Growing Complexity of Alzheimer Amyloid Precursor Protein Proteolysis. J Biol Chem 2016. 10. Haass, C.; Kaether, C.; Thinakaran, G.; Sisodia, S., Trafficking and Proteolytic Processing of APP. Cold Spring Harb Perspect Med 2012, 2 (5), a006270. 11. Plácido, A. I.; Pereira, C. M. F.; Duarte, A. I.; Candeias, E.; Correia, S. C.; Santos, R. X.; Carvalho, C.; Cardoso, S.; Oliveira, C. R.; Moreira, P. I., The role of endoplasmic reticulum in amyloid precursor protein processing and trafficking: Implications for Alzheimer's disease. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease 2014, 1842 (9), 1444-1453. 12. Yamazaki, T.; Selkoe, D. J.; Koo, E. H., Trafficking of cell surface beta-amyloid precursor protein: retrograde and transcytotic transport in cultured neurons. J Cell Biol 1995, 129 (2), 431-42. 13. Niederst, E. D.; Reyna, S. M.; Goldstein, L. S. B., Axonal amyloid precursor protein and its fragments undergo somatodendritic endocytosis and processing. Molecular Biology of the Cell 2014, 26 (2), 205-217. 14. Muresan, V.; Ladescu Muresan, Z., Amyloid-beta precursor protein: Multiple fragments, numerous transport routes and mechanisms. Experimental Cell Research 2015. 15. Lammich, S.; Kojro, E.; Postina, R.; Gilbert, S.; Pfeiffer, R.; Jasionowski, M.; Haass, C.; Fahrenholz, F., Constitutive and regulated alpha-secretase cleavage of Alzheimer's amyloid precursor protein by a disintegrin metalloprotease. Proceedings of the National Academy of Sciences 1999, 96 (7), 3922-3927. 16. Choy, R. W. Y.; Cheng, Z. L.; Schekman, R., Amyloid precursor protein (APP) traffics from the cell surface via endosomes for amyloid beta (A beta) production in the trans-Golgi network. Proceedings of the National Academy of Sciences of the United States of America 2012, 109 (30), E2077-E2082. 17. Cossec, J.-C.; Simon, A.; Marquer, C.; Moldrich, R. X.; Leterrier, C.; Rossier, J.; Duyckaerts, C.; Lenkei, Z.; Potier, M.-C., Clathrin-dependent APP endocytosis and Aβ secretion are highly sensitive to the level of plasma membrane cholesterol. Biochimica et Biophysica Acta (BBA) - Molecular and Cell Biology of Lipids 2010, 1801 (8), 846-852. 18. Cossec, J.-C.; Marquer, C.; Panchal, M.; Lazar, A. N.; Duyckaerts, C.; Potier, M.-C., Cholesterol changes in Alzheimer's disease: methods of analysis and impact on the formation of enlarged endosomes. Biochimica et Biophysica Acta (BBA) - Molecular and Cell Biology of Lipids 2010, 1801 (8), 839-845.
20
ACS Paragon Plus Environment
Page 21 of 27
1
1 2 2 3 3 4 4 55 66 77 88 99 10 10 11 11 12 12 13 13 14 14 15 15 16 17 16 18 17 19 18 20 19 21 20 22 21 23 22 24 23 25 26 24 27 25 28 26 29 27 30 28 31 29 32 30 33 31 34 32 35 36 33 37 34 38 35 39 36 40 37 41 38 42 39 43 40 44 45 41 46 42 47 43 48 44 49 45 50 46 51 47 52 48 53 54 49 55 50 56 51 57 52 58 59 60
ACS Chemical Neuroscience
19. Das, U.; Scott, David A.; Ganguly, A.; Koo, Edward H.; Tang, Y.; Roy, S., Activity-Induced Convergence of APP and BACE-1 in Acidic Microdomains via an Endocytosis-Dependent Pathway. Neuron 2013, 79 (3), 447-460. 20. Groemer, T. W.; Thiel, C. S.; Holt, M.; Riedel, D.; Hua, Y.; Huve, J.; Wilhelm, B. G.; Klingauf, J., Amyloid precursor protein is trafficked and secreted via synaptic vesicles. PLoS One 2011, 6 (4), e18754. 21. Das, U.; Wang, L.; Ganguly, A.; Saikia, J. M.; Wagner, S. L.; Koo, E. H.; Roy, S., Visualizing APP and BACE-1 approximation in neurons yields insight into the amyloidogenic pathway. Nature Neuroscience 2015, 19 (1), 55-64. 22. Li, Y.; Tsien, R., pHTomato, a red, genetically encoded indicator that enables multiplex interrogation of synaptic activity. Nature Neuroscience 2012, 15 (7), 1047-53. 23. Wilhelm, B. G.; Mandad, S.; Truckenbrodt, S.; Krohnert, K.; Schafer, C.; Rammner, B.; Koo, S. J.; Classen, G. A.; Krauss, M.; Haucke, V.; Urlaub, H.; Rizzoli, S. O., Composition of isolated synaptic boutons reveals the amounts of vesicle trafficking proteins. Science 2014, 344 (6187), 1023-8. 24. Osborn, M. J.; Panoskaltsis-Mortari, A.; McElmurry, R. T.; Bell, S. K.; Vignali, D. A. A.; Ryan, M. D.; Wilber, A. C.; McIvor, R. S.; Tolar, J.; Blazar, B. R., A picornaviral 2A-like sequence-based tricistronic vector allowing for high-level therapeutic gene expression coupled to a dual-reporter system. Molecular Therapy 2005, 12 (3), 569-574. 25. Wienisch, M.; Klingauf, J., Vesicular proteins exocytosed and subsequently retrieved by compensatory endocytosis are nonidentical. Nature Neuroscience 2006, 9 (8), 1019-1027. 26. Fridman, J. S.; Caulder, E.; Hansbury, M.; Liu, X.; Yang, G.; Wang, Q.; Lo, Y.; Zhou, B. B.; Pan, M.; Thomas, S. M.; Grandis, J. R.; Zhuo, J.; Yao, W.; Newton, R. C.; Friedman, S. M.; Scherle, P. A.; Vaddi, K., Selective Inhibition of ADAM Metalloproteases as a Novel Approach for Modulating ErbB Pathways in Cancer. Clinical Cancer Research 2007, 13 (6), 1892-1902. 27. Krämer, A.; Mentrup, T.; Kleizen, B.; Rivera-Milla, E.; Reichenbach, D.; Enzensperger, C.; Nohl, R.; Täuscher, E.; Görls, H.; Ploubidou, A.; Englert, C.; Werz, O.; Arndt, H.-D.; Kaether, C., Small molecules intercept Notch signaling and the early secretory pathway. Nature Chemical Biology 2013, 9 (11), 731-738. 28. Burns, M. P.; Igbavboa, U.; Wang, L.; Wood, W. G.; Duff, K., Cholesterol distribution, not total levels, correlate with altered amyloid precursor, protein processing in statin-treated mice. Neuromolecular Medicine 2006, 8 (3), 319328. 29. Burns, M.; Gaynor, K.; Olm, V.; Mercken, M.; LaFrancois, J.; Wang, L.; Mathews, P. M.; Noble, W.; Matsuoka, Y.; Duff, K., Presenilin redistribution associated with aberrant cholesterol transport enhances beta-amyloid production in vivo. J Neurosci 2003, 23 (13), 5645-9. 30. Willnow, T. E.; Andersen, O. M., Sorting receptor SORLA - a trafficking path to avoid Alzheimer disease. Journal of Cell Science 2013, 126 (13), 2751-2760. 31. Cai, D.; Leem, J. Y.; Greenfield, J. P.; Wang, P.; Kim, B. S.; Wang, R.; Lopes, K. O.; Kim, S. H.; Zheng, H.; Greengard, P.; Sisodia, S. S.; Thinakaran, G.; Xu, H., Presenilin-1 regulates intracellular trafficking and cell surface delivery of betaamyloid precursor protein. J Biol Chem 2003, 278 (5), 3446-54. 32. Allinquant, B.; Hantraye, P.; Mailleux, P.; Moya, K.; Bouillot, C.; Prochiantz, A., Downregulation of amyloid precursor protein inhibits neurite outgrowth in vitro. The Journal of Cell Biology 1995, 128 (5), 919-927. 33. Selkoe, D. J.; Hardy, J., The amyloid hypothesis of Alzheimer's disease at 25 years. EMBO Mol Med 2016, 8 (6), 595-608. 34. Sankaranarayanan, S.; De Angelis, D.; Rothman, J. E.; Ryan, T. A., The Use of pHluorins for Optical Measurements of Presynaptic Activity. Biophysical Journal 2000, 79 (4), 2199-2208. 35. Tsien, R. Y., The Green Fluorescent Protein. Annual Review of Biochemistry 1998, 67 (1), 509-544. 36. Piatkevich, K. D.; Verkhusha, V. V., Guide to Red Fluorescent Proteins and Biosensors for Flow Cytometry. Methods in cell biology 2011, 102, 431-461. 37. McCluskey, A.; Daniel, J. A.; Hadzic, G.; Chau, N.; Clayton, E. L.; Mariana, A.; Whiting, A.; Gorgani, N. N.; Lloyd, J.; Quan, A.; Moshkanbaryans, L.; Krishnan, S.; Perera, S.; Chircop, M.; Kleist, L. v.; McGeachie, A. B.; Howes, M. T.; Parton, R. G.; Campbell, M.; Sakoff, J. A.; Wang, X.; Sun, J. Y.; Robertson, M. J.; Deane, F. M.; Nguyen, T. H.; Meunier, F. A.; Cousin, M. A.; Robinson, P. J., Building a Better Dynasore: The Dyngo Compounds Potently Inhibit Dynamin and Endocytosis. Traffic 2013, 14 (12), 1272-1289. 38. Sun, J.; Roy, S., The physical approximation of APP and BACE-1 - a key event in Alzheimer's disease pathogenesis. Developmental Neurobiology 2017. 39. Winckler, B.; Mellman, I., Trafficking Guidance Receptors. Cold Spring Harbor Perspectives in Biology 2010, 2 (7), a001826-a001826.
21
ACS Paragon Plus Environment
ACS Chemical Neuroscience
1
1 2 2 3 3 44 55 66 77 88 99 10 10 11 11 12 12 13 13 14 14 15 15 16 17 16 18 17 19 18 20 19 21 20 22 23 21 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 22 of 27
40. Nakata, T.; Hirokawa, N., Microtubules provide directional cues for polarized axonal transport through interaction with kinesin motor head. The Journal of Cell Biology 2003, 162 (6), 1045-1055. 41. Barrett, P. J.; Song, Y.; Van Horn, W. D.; Hustedt, E. J.; Schafer, J. M.; Hadziselimovic, A.; Beel, A. J.; Sanders, C. R., The amyloid precursor protein has a flexible transmembrane domain and binds cholesterol. Science 2012, 336 (6085), 1168-71. 42. Dawkins, E.; Small, D. H., Insights into the physiological function of the beta-amyloid precursor protein: beyond Alzheimer's disease. Journal of Neurochemistry 2014, 129 (5), 756-769. 43. Gibson, D. G.; Young, L.; Chuang, R.-Y.; Venter, J. C.; Hutchison, C. A.; Smith, H. O., Enzymatic assembly of DNA molecules up to several hundred kilobases. Nature Methods 2009, 6 (5), 343-345. 44. Liu, G.; Tsien, R. W., Synaptic transmission at single visualized hippocampal boutons. Neuropharmacology 1995, 34 (11), 1407-1421. 45. Edelstein, A.; Amodaj, N.; Hoover, K.; Vale, R.; Stuurman, N., Computer Control of Microscopes Using µManager. 2010. 46. Schindelin, J.; Arganda-Carreras, I.; Frise, E.; Kaynig, V.; Longair, M.; Pietzsch, T.; Preibisch, S.; Rueden, C.; Saalfeld, S.; Schmid, B.; Tinevez, J.-Y.; White, D. J.; Hartenstein, V.; Eliceiri, K.; Tomancak, P.; Cardona, A., Fiji: an opensource platform for biological-image analysis. Nature Methods 2012, 9 (7), 676-682. 47. Schneider, C. A.; Rasband, W. S.; Eliceiri, K. W., NIH Image to ImageJ: 25 years of image analysis. Nature Methods 2012, 9 (7), 671-675. 48. Thevenaz, P.; Ruttimann, U. E.; Unser, M., A pyramid approach to subpixel registration based on intensity. IEEE Transactions on Image Processing 1998, 7 (1), 27-41.
22
ACS Paragon Plus Environment
Page 23 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Neuroscience
Figure 1 341x548mm (150 x 150 DPI)
ACS Paragon Plus Environment
ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Figure 2 564x579mm (150 x 150 DPI)
ACS Paragon Plus Environment
Page 24 of 27
Page 25 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Neuroscience
Figure 3 416x499mm (150 x 150 DPI)
ACS Paragon Plus Environment
ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Figure 4 620x150mm (150 x 150 DPI)
ACS Paragon Plus Environment
Page 26 of 27
Page 27 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Chemical Neuroscience
ACS Paragon Plus Environment