A novel fluorescent probe reveals starvation controls the commitment of amyloid precursor protein to the lysosome

    A novel fluorescent probe reveals starvation controls the commitment of amyloid precursor protein to the lysosome Leanne K. Hein, Pirjo M. Apaja, Kathryn Hattersley, Randall H. Grose, Jianling Xie, Christopher G. Proud, Timothy J. Sargeant PII: DOI: Reference:

S0167-4889(17)30163-5 doi:10.1016/j.bbamcr.2017.06.011 BBAMCR 18124

To appear in:

BBA - Molecular Cell Research

Received date: Revised date: Accepted date:

15 February 2017 3 June 2017 17 June 2017

Please cite this article as: Leanne K. Hein, Pirjo M. Apaja, Kathryn Hattersley, Randall H. Grose, Jianling Xie, Christopher G. Proud, Timothy J. Sargeant, A novel fluorescent probe reveals starvation controls the commitment of amyloid precursor protein to the lysosome, BBA - Molecular Cell Research (2017), doi:10.1016/j.bbamcr.2017.06.011

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT A novel fluorescent probe reveals starvation controls the commitment of amyloid precursor protein to the lysosome

Affiliations:

Lysosomal Diseases Research Unit, Nutrition and Metabolism Theme, South

NU

a

IP

SC R

Xiec, Christopher G. Prouda, Timothy J. Sargeanta*

T

Leanne K. Heina, Pirjo M. Apajaa,d, Kathryn Hattersleya, Randall H. Groseb, Jianling

MA

Australian Health and Medical Research Institute, PO Box 11060, Adelaide 5001, South Australia, Australia

ACRF Innovative Cancer Imaging Facility, Cancer Theme, South Australian Health

D

b

TE

and Medical Research Institute,

c

CE P

Australia

Cell Signalling Group, Nutrition and Metabolism Theme, South Australian Health

Australia

Adelaide 5001, South Australia,

AC

and Medical Research Institute,

d

Adelaide 5001, South Australia,

EMBL Australia

*Corresponding author Dr Tim Sargeant Lysosomal Diseases Research Unit Nutrition and Metabolism Theme

1

ACCEPTED MANUSCRIPT South Australian Health and Medical Research Institute PO Box 11060, Adelaide 5001, South Australia, Australia

IP

T

Telephone: +61 8 81284940

SC R

E-mail: [email protected]

NU

Email addresses:

MA

Leanne K. Hein: [email protected] Pirjo M. Apaja: [email protected]

D

Kathryn Hattersley: [email protected]

TE

Randall H. Grose: [email protected]

CE P

Jianling Xie: [email protected] Christopher G. Proud: [email protected]

1

AC

Timothy J. Sargeant: [email protected]

1

Abbreviations: Aβ, amyloid-β; APP, amyloid precursor protein; APP-CTF, APP-carboxy-terminal fragment; β-CTF, amyloid precursor protein β-carboxy-terminal fragment; BSA, bovine serum albumin; CQ, chloroquine; DMEM, Dulbecco’s modified Eagle’s medium; DPBS, Dulbecco’s phosphate-buffered saline; EBSS, Earle’s balanced salt solution; EGFP, enhanced green fluorescent protein; FACS, fluorescence activated cell sorting; FBS, foetal bovine serum; FITC, fluorescein isothiocyanate; FRIA, fluorescence ratio image analysis; mRFP1, monomeric red fluorescent protein; mTOR, mechanistic/mammalian target of rapamycin; mTORC1, mechanistic/mammalian target of rapamycin complex 1; PBS, phosphate buffered saline; pen/strep, penicillin/streptomycin; PFA, paraformaldehyde; R/G, red fluorescence/green fluorescence ratio; TBS, trisbuffered saline; TBST, tris-buffered saline with Tween 20; tf-APP, tandem-fluorescent APP; tf-LC3, tandemfluorescent LC3.

2

ACCEPTED MANUSCRIPT ABSTRACT Alzheimer’s disease is the most important cause of dementia but there is no therapy

T

that has been demonstrated to stop or slow disease progression. Amyloid precursor

IP

protein (APP) is the source of amyloid-β (Aβ), which aggregates in Alzheimer’s

SC R

disease to form toxic oligomeric species. The endo-lysosomal system can clear APP and Aβ from the cell if these molecular species are trafficked through to the

NU

lysosome. Currently, there are no easy methods available for the analysis of lysosomal APP trafficking. We therefore generated a fusion protein (tandem-

MA

fluorescent, or tf-APP) that allows detection of changes in APP trafficking using accessible techniques such as flow cytometry. This permits rapid analysis or

D

screening of genes and compounds that alter APP processing in the cell. Using our

TE

novel molecular probe, we determined that starvation induces trafficking of APP and APP-carboxy-terminal fragments (APP-CTFs) to the degradative endo-lysosomal

CE P

network. In line with this finding, suppression of mTOR signalling using AZD8055 also strongly induced trafficking of APP to the endo-lysosomal system. Remarkably,

AC

activation of mTOR signalling via RHEB over-expression inhibited the starvationinduced autophagy but did not affect trafficking of tf-APP. These results show tf-APP can be used to determine how APP is trafficked through the lysosomal system of the cell. This molecular probe is therefore useful for determining the molecular mechanism behind the commitment of APP to the degradative pathway or for screening compounds that can induce this effect. This is important as clearance of APP and APP-CTF provides an important potential therapeutic strategy for Alzheimer’s disease.

3

ACCEPTED MANUSCRIPT Keywords: Alzheimer’s disease, lysosome, endocytosis, mTOR, starvation, amyloid precursor

AC

CE P

TE

D

MA

NU

SC R

IP

T

protein

4

ACCEPTED MANUSCRIPT 1. Introduction Dementia is a major and growing health problem for aging populations. The

T

most common cause of dementia is late-onset Alzheimer’s disease, a progressive

IP

neurodegenerative condition that currently has no therapy capable of slowing or

SC R

stopping the disease process. The amyloid cascade hypothesis (or variants thereof) is a widely-accepted model of Alzheimer disease pathogenesis that places

NU

accumulation and aggregation of the amyloid-β (Aβ) peptide as the cause of the disease process [1].

MA

Aβ is produced by the sequential proteolytic processing of amyloid precursor protein (APP). APP can be cleaved by β-secretase (BACE1) to produce membrane-

D

bound APP β-carboxy-terminal fragments (β-CTF) [2]. When followed by γ-secretase

TE

cleavage, this results in generation of Aβ [3]. The importance of Aβ accumulation to

CE P

the development of Alzheimer’s disease is underscored by the fact that early-onset Alzheimer’s disease can be caused by mutations in APP that result in either increased production of Aβ or its enhanced aggregation [4]. Furthermore, APP

[5].

AC

mutations that prevent β-secretase cleavage are protective from Alzheimer’s disease

Proteolytic cleavage of APP that produces potentially amyloidogenic fragments can occur in the early endosome. This organelle is also a part of the degradative endo-lysosomal system that is responsible for the final degradation of APP and its amyloidogenic fragments [3, 6, 7]. These observations led to the hypothesis that even minor perturbation of degradative capacity in this pathway could lead to the gradual accumulation of Aβ seen in Alzheimer’s disease [7].

5

ACCEPTED MANUSCRIPT Although trafficking of APP into the lysosome for hydrolysis was identified as an important process over two decades ago, very little information exists that

T

explains how this process works. For example, we still do not know which proteins

SC R

which signalling systems control these mechanisms.

IP

are responsible for recruiting APP from recycling to degradative pathways, and

The lysosomal destruction of APP and β-CTF is regulated at several different

NU

steps that include endocytosis, ubiquitination, and ESCRT-mediated sorting into the lumen of the multivesicular body before lysosomal hydrolysis [8]. Recent research

MA

has shown genetic variation associated with late onset Alzheimer’s disease may cause a defect in internalisation of APP into the multivesicular body and its

D

subsequent lysosomal degradation [9]. In keeping with these observations,

TE

repression of ESCRT activity causes an increase in intracellular Aβ [10]. Commitment of APP to lysosomal degradation also directly involves Beclin1,

CE P

providing an intriguing link with autophagy [11]. Furthermore, up-regulation of lysosomal activity caused by KO of an endogenous lysosomal protease inhibitor [12,

AC

13], or over-expression of transcription factor EB (TFEB), reduces the burden of disease in mouse models of AD [14]. The mechanisms that initiate the lysosomal trafficking of APP are poorly defined, and there are currently no tools to easily and quickly measure flux of APP and related fragments through the lysosomal system. To be useful, such tools should be suitable for use in screening strategies as well as in hypothesis-driven research for the discovery of regulators of APP and β-CTF lysosomal degradation. The aim of this study was to develop a method that could be used for this purpose. To do this, we adapted previously validated ratiometric tandem-fluorescent tag technology used for measuring autophagic flux with LC3 [15], which is an important autophagic 6

ACCEPTED MANUSCRIPT molecule that associates with the phagophore via phosphatidylethanolamine. The intracellular trafficking of tf-APP was measured using flow cytometry to produce a

T

simple and sensitive measure of recruitment of APP to the degradative endo-

SC R

IP

lysosomal system.

2. Materials and methods

NU

2.1 Materials

MA

AZD8055 was purchased from Selleckchem. Dynasore hydrate (Dynasore), chloroquine diphosphate salt (chloroquine, CQ), bafilomycin A1, Earle’s balanced

D

salt solution (EBSS), bovine serum albumin (BSA), penicillin-streptomycin solution

TE

(pen/strep), poly-l-lysine, monoclonal anti-FLAG M2 antibody produced in mouse (F3165) and anti-amyloid precursor protein, C-terminal antibody (rabbit polyclonal,

CE P

A8717) were purchased from Sigma. The WestFemto ECL blotting system and micro BCA protein assay kit were purchased from Thermo Scientific. Goat anti-rabbit IgG

AC

antibody (H+L) HRP conjugate (AP307P) and anti-Alzheimer precursor protein A4 clone 22C11 (mouse monoclonal, MAB348) were purchased from Merck Millipore. Foetal bovine serum (FBS), Dulbecco's phosphate-buffered saline (DPBS) and Dulbecco's modified Eagle's

medium

(DMEM)

were

purchased from Life

Technologies. LC3B/MAP1LC3B antibody (rabbit polyclonal, NB100-2220) and rabbit anti-TGN46 (polyclonal, NBP1-49643) antibodies were purchased from Novus Biologicals. Ribosomal protein S6 antibody (mouse monoclonal, sc-74459) was purchased from Santa Cruz Biotechnology. Phospho-S6 ribosomal protein (Ser240/244, rabbit polyclonal, CST 2215) antibody was purchased from Cell Signaling Technology. Anti-green fluorescent protein rabbit serum (A6455) was 7

ACCEPTED MANUSCRIPT purchased from Invitrogen. Polyclonal sheep anti-human LAMP1 was produced inhouse. Sheep anti-mouse IgG, HRP-linked whole antibody (NA931) was purchased

and

Alexa-647

conjugated

donkey

anti-rabbit

(711-605-152)

IP

(713-605-003)

T

from GE Healthcare Life Sciences. Alexa-647 conjugated donkey anti-sheep IgG

SC R

secondary antibodies were purchased from Jackson ImmunoResearch Laboratories. VectaShield Mounting Media with DAPI was purchased from Vector Laboratories.

NU

2.2 Generation of DNA constructs

All work involving recombinant DNA and lentivirus conducted in this study was

D

2.2.1 Construction of lenti-tf-LC3

MA

approved by the SAHMRI Institutional Biosafety Committee.

TE

tf-LC3 coding sequence was PCR-amplified from ptfLC3, a gift from Tamotsu Yoshimori (Addgene plasmid #21074; [15]), using the following primers: forward 5’and

ccggagccggatcctctagaTCACAAGCATGGCTCTCTTC-3’.

pUltra-hot,

CE P

tgcaggtccgatccaccggtATGGCCTTCTCCGAGGAC-3’

reverse a

gift

5’from

AC

Malcolm Moore (Addgene plasmid #24130), was cut using AgeI and XbaI. This created a vector backbone that no longer possessed the mCherry coding sequence contained in the original vector. PCR-amplified tf-LC3 was ligated into the cut pUltra vector backbone using Gibson Assembly Master Mix (New England Biolabs; #E2611S). The new vector, lenti-tf-LC3, expresses tf-LC3 downstream of a human ubiquitin C promoter. 2.2.2 Construction of lenti-tf-APP 1) Generation of pcDNA3 APP-mCherry: APP coding sequence was PCRamplified from pCAX APP 695, a gift from Dennis Selkoe and Tracy Young-Pearse (Addgene plasmid #30137; [16]), using the following primers: forward 5’8

ACCEPTED MANUSCRIPT and

tacttccaatccaatgccaccATGCTGCCCGGTTTGGCACT-3’

reverse

5’-

ctcccactaccaatgccGTTCTGCATCTGCTCAAAGA-3’. Amplified PCR product was

T

digested using the exonuclease activity of T4 DNA polymerase (New England

IP

Biolabs; #M0203S) in the presence of dCTP. pcDNA3 mCherry LIC cloning vector

SC R

(6B), a gift from Scott Gradia (Addgene plasmid #30125) was cut using SspI and digested using T4 DNA polymerase in the presence of dGTP. Amplified and digested APP sequence was annealed with SspI cut and T4 DNA polymerase-digested vector

NU

backbone, and this mixture was transformed into competent bacteria. This created

MA

pcDNA3 APP-mCherry, where mCherry was fused to the carboxy-terminus of APP. 2) Generation of tf-APP: APP-mCherry coding sequence was PCR-amplified the

following

D

using

primers: and

reverse

5’5’-

TE

tgcaggtccgatccaccggtGCCACCATGCTGCCCGGT-3’

forward

gtggccatCTTGTACAGCTCGTCCATGCCG-3’. EGFP sequence from LeGO-iG2, a

CE P

gift from Boris Fehse (Addgene plasmid #27341; [17]), was PCR-amplified using the following primers: forward 5’-ctgtacaagATGGCCACAACCATGGTG-3’ and reverse ccggagccggatcctctagaTTACTTGTACAGCTCGTCC-3’.

pUltra-hot

vector

AC

5’-

backbone was prepared as described above. Each of the three fragments of DNA were simultaneously ligated together using Gibson Assembly Master Mix. This created Lenti-tf-APP, where the fusion protein APP-mCherry-EGFP is expressed from a human ubiquitin C promoter. All PCR amplification was performed using Q5 High-Fidelity 2X Master Mix (New England Biolabs; #M0492S). Plasmid DNA was purified using the GeneJET plasmid midiprep kit (Thermo Scientific). All plasmids were verified by restriction digests and DNA sequence analysis.

9

ACCEPTED MANUSCRIPT 2.3 Generation of lentiviral particles Lentiviral particles that expressed tf-APP and tf-LC3 constructs were

T

produced by transfection of HEK 293T cells using Lipofectamine 3000 (Thermo-

IP

Fisher Scientific; #L3000015), as per the manufacturer’s instructions. Three plasmids

SC R

were used to make lentiviral particles, which included psPAX2 (a gift from Didier Trono, Addgene plasmid #12260), pCMV-VSV-G (a gift from Bob Weinberg,

NU

Addgene plasmid #8454; [18]) and lenti-tf-APP or lenti-tf-LC3 viral plasmids. Viruscontaining media was collected from the HEK 293T cells and filtered through a 0.45

MA

µM filter; polybrene was then added at a final concentration of 4 µg/ml. The viruscontaining media was then placed on 80% confluent HeLa cells overnight. The

D

following day the virus-containing media incubating with the HeLa cells was removed

TE

and normal growth media (DMEM with 10% FBS) was added. Cells were maintained

CE P

until sorted into high and low fluorescent populations by FACS. 2.4 Fluorescence Activated Cell Sorting (FACS) and production of monoclonal tf-LC3

AC

and tf-APP expressing HeLa cell populations

Cells used in this study routinely tested negative for mycoplasma infection. tfLC3 and tf-APP HeLa cells were maintained in normal growth media (DMEM with 10% FBS) and then harvested by trypsinisation and centrifugation. The cell pellet was resuspended in 500 µl of DMEM with 1% FBS and pen/strep, put through a 70 µM cell strainer into a FACS tube and then sorted. Cells were sorted at a moderate pressure (40 psi) using an 85 µm nozzle on a BD FACSAria Fusion cell sorter (BD Bioscience, USA) with BD FACSDiva software version 8.0 (BD Bioscience, USA). Both cell sample and collection tubes were maintained at 4°C. Cells were gated on SSC-H v FSC-H and FSC-H v FSC-A, respectively. Red mRFP1 or mCherry positive

10

ACCEPTED MANUSCRIPT single cells were further divided into mRFP1 or mCherry low or high fluorescent populations. Sorted cells were plated out at one or two cells per well in 96 -well cell

T

culture plates. Cells were maintained in DMEM with 10% FBS and pen/strep, and

IP

wells containing single colonies were identified before passaging up into 24-well

SC R

plates followed by 6-well plates. Once confluent, cells were harvested and the cell pellet (resuspended in 500 µl of DMEM with 1% FBS and pen/strep) was put through a 70 µM cell strainer into a FACS tube. Cells were analysed by flow cytometry on the

NU

BD LSR Fortessa X20 Analyser (BD Bioscience, USA) for red and green fluorescent

MA

intensities and variation around median fluorescence quality. Clonal cell lines that showed fluorescence and good peak shapes with low variation were grown in 6-well

D

plates in normal growth medium (basal) or under starved conditions by washing cells

TE

three times with DPBS before incubation in 4 ml of EBSS. After starvation for 4 h, cells were harvested and the cell pellet analysed by flow cytometry as described

CE P

above. Analysis included examination of the ratio of red fluorescence to green fluorescence (mRFP1/EGFP for tf-LC3 HeLa cells and mCherry/EGFP for tf-APP

AC

HeLa cells). The monoclonal lines that showed the greatest shift in the red to green fluorescent ratio between the basal cells and starved cells was then designated as the best clone and selected for future experiments. All experiments using flowcytometric analysis were performed in technical duplicates.

2.5 Preparation of cell lysates and western blot analysis Cells were harvested and washed twice with DPBS before preparing cell homogenates. HeLa cells were included as a negative control. Cell pellets were resuspended in 200 µl of 0.5 M NaCl, 0.02 M Tris (pH 7.0) and 0.1% nonidet P-40 (if detecting phosphorylated proteins 2.5 mM sodium pyrophosphate, 1 mM sodium

11

ACCEPTED MANUSCRIPT orthovanadate, 1 mM β-glycerophosphate (pH 7.4) and complete, EDTA-free protease inhibitor cocktail (Sigma) were included), and sonicated on ice for 20 s.

T

Total cell protein was determined by the micro BCA protein assay kit (Thermo-Fisher

IP

Scientific; #23235). Western blot analysis was performed using 10 µg of total protein

SC R

of each of the homogenates that were electrophoresed through 4-12% SDS-PAGE gels. The gels were transferred to a PVDF membrane at 35 V for 70 min. The membranes were incubated in block solution (Tris-buffered saline containing 0.1%

NU

(v/v) Tween 20 (TBST) and 5% (w/v) skim milk) for 1 h at room temperature, with

MA

rocking. Skim milk was replaced with 2% BSA if phospho-proteins were being measured. The membranes were washed for 5 min in TBST before incubating and

D

rocking overnight at 4oC with primary antibodies diluted 1:1000 in block solution. The

TE

following day the membranes were washed three times for 5 min in TBST and then incubated for 1 h at room temperature, with rocking, with HRP-conjugated goat anti-

CE P

rabbit immunoglobulin or sheep anti-mouse immunoglobin diluted 1:4000 in block solution. The membranes were washed three times for 5 min in TBST before a final

AC

5 min wash in TBS, and developed using the WestFemto ECL blotting system (34095 or 34077, Thermo Scientific) and detected using the LAS4000 Luminescent Image Analyser (Fujifilm Life Science). 2.6 Immunocytochemistry Two hundred thousand tf-APP HeLa or tf-LC3 HeLa cells were seeded into 12-well plates containing poly-l-lysine-coated coverslips. Cells were left to grow for 24 h. Cells were washed with PBS, then fixed with 1% PFA/PBS for 10 min. Cells were washed twice with PBS, then permeabilised with PBS containing 0.1% saponin. Cells were blocked with PBS containing 5% BSA and 0.01% saponin for 30 min. Primary antibodies were incubated in blocking solution for 1 h at 1:200 dilution for 12

ACCEPTED MANUSCRIPT anti-LAMP1 and anti-TGN46 antibodies or 1:50 for anti-APP antibodies. Cells were washed with PBS three times for 2 min each. Secondary antibodies were incubated

T

in blocking solution for 40 min at 1:500 dilution. Cells were washed three times for 2

IP

min each in PBS and then mounted onto slides using VectaShield with DAPI (Vector

SC R

Laboratories). Images were taken using a Leica TCS SP8X/MP confocal microscope. Images were imported into ImageJ for analysis.

NU

2.7 Starvation and treatment with pharmacological reagents

tf-LC3 and tf-APP HeLa cells were plated out in 6-well cell culture plates at a

MA

density of 2x105 cells per well in DMEM containing 10% FBS with pen/strep for 48 h. For starvation, cells were washed three times in DPBS and then incubated for 4 h in

D

EBSS. Normal growth media (DMEM containing 10% FBS) was replaced in the

TE

control wells. Cells were treated (as indicated in Results) with 50 µM chloroquine,

CE P

500 nM bafilomycin or 80 µM dynasore. Control conditions used the appropriate vehicle. After 4, cells were washed with DPBS and harvested. Cell pellets were resuspended in 0.5 ml media and placed through a 70 µM cell strainer into a FACS

AC

tube, then analysed by flow cytometry (BD LSR Fortessa X20 Analyser), as described above. HeLa cells were also analysed as a negative control in normal growth media and after treatment with the pharmacological reagents to confirm the absence of auto-fluorescence. The ratio of red fluorescence to green fluorescence (mRFP1/EGFP for tf-LC3 HeLa cells, and mCherry/EGFP for tf-APP HeLa cells) and the shift in this ratio between the basal cells and cells after treatment with pharmacological reagents was analysed using FlowJo software. Graphs depicting quantification of flow cytometry results were generated by quantifying test samples relative to their controls. This was done by calculating the percentage of cells that fell within a gate extending from the right hand side of the peak of the reference 13

ACCEPTED MANUSCRIPT population indicated in the figure legend. All treatments were done in duplicate on three separate occasions.

IP

T

2.8 Fluorescence ratiometric image analysis The methodology for FRIA of endocytic vesicles has been described in detail

SC R

[19]. Briefly, Hela cells expressing tf-LC3 or tf-APP were starved and/or treated with pharmacological reagents as described above. Cells were grown on glass cover

NU

slips and allowed to attach for 48 h before treatments and analysis. FRIA was performed on a Nikon TI-E inverted fluorescence microscope equipped with

MA

Lumencor Spectra X light source and electron-multiplying charge-coupled device (Photometrics). To detect plasma membrane originating APP, cells were transiently

D

transfected with APP-mCherry or pcDNA3.1. Plasma membrane APP was labelled

TE

with anti-APP antibody (22C11) and FITC-conjugated goat anti-mouse secondary

CE P

antibody (Jackson ImmunoResearch) on ice. Synchronized endocytosis was performed for indicated times at 37°C. At least 300 vesicles were analysed from each independent experiment. The acquisition was carried out at 495 ± 5–nm and

AC

570 ± 10–nm excitation wavelengths with a 535 ± 20–nm and 610 ± 20–nm emission filter using triggered acquisition. Results were analysed using NIS-Elements (Nikon, Japan). 2.9 Statistical analysis Statistical analysis was performed using two-tailed t-test, paired t-test or ANOVA with multiple comparison testing with the means of at least three independent experiments, as indicated in figure legends. GraphPad Prism software package was used to calculated p-values. 2.10 Image manipulation 14

ACCEPTED MANUSCRIPT Images were manipulated using ImageJ or Inkscape v0.91 software.

IP

T

3. Results

SC R

3.1 Generation and selection of cell lines

To generate a molecular probe that could be used to determine mechanisms for enhancing the lysosomal hydrolysis of APP, we fused tandem mCherry and

NU

EGFP to the carboxy-terminus of APP695 (tf-APP). The tandem fluorescence tag

MA

was placed on the carboxy-terminus as the amino terminus is lost during α- or βcleavage by the enzymes ADAM10 [20] and BACE1 [2], respectively. This coding

D

sequence was sub-cloned downstream of the human ubiquitin C promoter in a

TE

lentiviral construct (Fig. 1A). This eukaryotic promoter was chosen to promote stable, low expression [21], which can be desirable when expressing constructs that

CE P

measure lysosomal flux [22].

To compare the characterisation of flux through lysosome-dependent

AC

pathways in this study, we sub-cloned the well-characterised tf-LC3 construct into the same lentiviral vector (Fig. 1A). As previously reported, the tf-LC3 construct fluoresces both green and red under neutral pH conditions [15]. However, when sequestered by an autophagosome and exposed to an acidic environment such as one produced upon lysosomal fusion, EGFP quenches (pKa = 6.0 [23]). Red fluorescent proteins such as mRFP1 and mCherry have lower pKa values (pKa = 4.5 [23, 24]) and do not quench as easily in the low pH of the autophagic or endolysosomal systems. Therefore, the selective quenching of EGFP results in a higher red fluorescence/green fluorescence ratio (R/G) as the pH decreases (Fig. 1B). By fusing mCherry and EGFP to the carboxy-terminus of APP to make tf-APP, we 15

ACCEPTED MANUSCRIPT produced a probe that will report the passage of APP through the degradative pathway in the late endosome (Fig. 1C). When sequestered by the ESCRT complex

T

and trafficked into the lumen of the late endosome on an intraluminal vesicle, the low

IP

pH of this environment will be reflected by an increase in R/G in a similar way to the

SC R

tf-LC3 construct. Although there is some limited evidence that intraluminal vesicles are acidic [25], it is likely that the increase in R/G for tf-APP occurs when the intraluminal vesicle membrane loses integrity during the degradative process and the

NU

intraluminal vesicle contents mix with the acidic interior of the late endosome or

MA

lysosome.

Following the guidelines of Gump and Thorburn [22], we generated

D

monoclonal HeLa cell lines that expressed either tf-APP (58 monoclonal lines) or tf-

TE

LC3 (51 monoclonal lines) by lentiviral transduction. This allows us to create a cell line that could be used for quantitation using flow cytometry. Each monoclonal cell

CE P

line was subjected to a first screen using flow cytometry where R/G was quantified (data not shown). The cell lines that showed the least variation in R/G under

AC

unstimulated conditions were selected for further analysis. A second screen was performed where cell lines from the first screen were starved in Earle’s Balanced Salt Solution (EBSS) This revealed lines with the largest increases in R/G in response to stimulation (data not shown). The best tf-LC3 and tf-APP clones were selected for further analysis using microscopy- and flow cytometry-based methods. 3.2 Analysis of stable tf-APP expression in HeLa cells Clones were processed for immunocytochemistry to show localisation of the tf-APP/tf-APP-CTF. To determine whether fluorescence did in fact correlate to localisation of APP immunoreactivity, tf-APP-expressing cells were immuno-stained

16

ACCEPTED MANUSCRIPT using two different APP antibodies (Fig. 2A). Both antibody clone 6E10 (Aβ sequence) and an antibody that targets the carboxy-terminus of APP showed APP-

T

like immunoreactivity that co-localised with mCherry fluorescence. The cellular

IP

distribution of endogenous APP (seen in cells that did not express the tf-APP

fusion protein didn't change APP localization. blotting

of

HeLa

cells

expressing

tf-APP

revealed

bands

NU

Western

SC R

construct; Fig. 2B) was similar to that of transgenic APP, showing the transgenic

corresponding to tf-APP (160 kDa) and suspected tf-APP-CTFs (70 kDa) when

MA

probed with an antibody directed towards the carboxy-terminus of APP (Fig. 2C). Antibodies directed towards the amino terminus of APP also showed full-length tf-

D

APP at approximately 160 kDa (Fig. 2D). On the same blot, endogenous APP is

TE

shown as a band at 100 kDa. Anti-LC3 antibodies were capable of detecting tf-LC3 in appropriate cells (70 kDa, Fig. 2E), and anti-GFP antibodies detected bands

CE P

consistent with tf-APP, tf-APP-CTFs and tf-LC3 (Fig. 2F).

AC

3.3 Starvation directs tf-APP/tf-APP-CTF toward lysosomes Previous research has shown that plasma membrane proteins are selectively degraded in the lysosome as a source of amino acids in response to starvation [26]. To determine whether tf-APP responded to cellular nutrient status, tf-APP and tf-LC3 cells were subjected to starvation in EBSS for 4 h and then analysed by flow cytometry. As expected, tf-LC3 showed an increase in R/G when cells were starved in EBSS (Fig. 3A, C). This starvation response was strongly inhibited by the lysosomotropic agent, chloroquine (Fig. 3B, C), which raises lysosomal pH. Interestingly, tf-APP also showed a strong response to starvation, shown by a large increase in R/G (Fig. 3D). This response was also reversed by raising lysosomal pH

17

ACCEPTED MANUSCRIPT using chloroquine (Fig. 3E, F). The starvation response also persisted in the presence

of

the

γ-secretase

inhibitor

L-685,458

(Supplementary

Fig.

1),

T

demonstrating γ-secretase cleavage did not impact on measurement of R/G.

IP

Starvation induced-increases in R/G were also replicated in monoclonal Cath. A

SC R

differentiated (CAD) murine neuroblastoma cells that stably expressed tf-APP (Supplementary Fig. 2).

NU

To determine whether tf-LC3 and tf-APP localisation in response to starvation could be monitored by confocal imaging, cells were stained for LAMP1 with and

MA

without starvation in EBSS for 4 h. Cells were treated with leupeptin to stabilise the fluorescence signal inside degradative LAMP1-positive compartments as previously

D

published [27]. Confocal microscopy and quantitative analysis of overlap between

TE

red fluorescence from tf-LC3 or tf-APP and immunofluorescence from LAMP1 was performed by an experimenter who was blind to the genotype and conditions used.

CE P

Quantification of this overlap showed tf-LC3 strongly trafficked to the LAMP1-positive compartment during starvation (Fig. 4A, C). tf-APP trended towards increased co-

AC

localisation with the LAMP1-positive compartment under these conditions (Fig. 4B, C). While tf-APP showed a punctate distribution in cells that partially overlapped with LAMP1 (Fig. 4B), in some planes of the cell, tf-APP clustered and co-localised with a marker for the Golgi apparatus (TGN46, Fig. 4D). The endo-lysosomal pathway has a characteristic decreasing pH gradient reaching an acidic (< pH 5) environment in lysosomes. Fluorescence ratiometric image analysis (FRIA) [19] of vesicles that carry tf-APP as a cargo was used to monitor trafficking and vesicular pH (Fig. 5 A-C). Calibration of pH against fluorescence ratios for FRIA was performed for both tf-LC3 and tf-APP constructs (Supplementary Fig. 3). FRIA showed that the mean vesicular pH for the carboxy18

ACCEPTED MANUSCRIPT terminal tandem fluorescent tag present on APP in basal conditions was an average pH of 6.00 ± 0.05 (Fig. 5A is a representative histogram showing a single

T

experiment, Fig. 5C shows the mean of multiple independent experiments). The

IP

average pH value is a consequence of all tf-APP vesicles, some of which possess

SC R

intraluminal tandem fluorescence tags (see Fig. 1C) because of ESCRT-mediated multivesicular budding. Starvation in EBSS medium for 4 h substantially facilitated tfAPP transfer to late endosomes and lysosomes (mean pH 5.21 ± 0.1). This effect

NU

could be inhibited by increasing endo-lysosomal pH with the lysosomotropic agent

MA

chloroquine, or the selective v-type ATPase inhibitor, bafilomycin A1 (Fig. 5C), as observed for tf-APP in basal conditions. Dynasore, a GTPase inhibitor for dynamin-

D

dependent endocytosis, had the same inhibitory effect (mean pH 6.00 ± 0.1) (Fig.

TE

5B, C).

FRIA of vesicles containing tf-LC3 as a cargo in basal nutrient replete

CE P

conditions showed a mean vesicular pH of 5.18 ± 0.2. Nevertheless, starvation for 4 h decreased the pH reported by the tf-LC3 probe (mean vesicular pH of 4.85 ± 0.02)

AC

(Fig. 5D). Chloroquine and bafilomycin, compounds that increase endo-lysosomal pH, raised vesicular pH as predicted (pH 6.15 ± 0.04 and pH 6.00 ± 0.04, respectively) (Fig. 5D). APP has been linked to lipid rafts at the plasma membrane [28]. However, confocal imaging showed no obvious localisation of tf-APP at the plasma membrane (Fig. 2 and 4). This either means that APP is not trafficked to the plasma membrane in HeLa cells, or that its presence at the plasma membrane is transient. To test this, cells transiently expressing APP-mCherry were labelled on ice with an anti-APPFITC antibody (clone 22C11) that binds to the (extracellular) amino terminus of APP (Fig. 6A). FRIA was performed on vesicles that were positive for both fluorophores 19

ACCEPTED MANUSCRIPT indicating cargo originating from the plasma membrane. Cells were labelled and chased for 1-4 h. In basal conditions, major APP pools were found in early and

T

recycling endosomes (pH 6.15 ± 0.13). However, starvation induced APP relocation

IP

to late endosomes (mean pH 5.63 ± 0.12) (Fig. 6B, C, F), similar to the effect that

SC R

was observed for expressed tf-APP. This effect could be reversed using chloroquine (pH 6.59 ± 0.08) or dynasore (pH 6.53 ± 0.15) (Fig 6D, E, F). Interestingly, starvation didn’t influence the transferrin receptor, which was still located in recycling

MA

plasma membrane proteins (Fig. 6F).

NU

endosomes (pH 6.20 ± 0.02). This showed that starvation does not modulate all

3.4 Inhibition of mTOR is sufficient to promote trafficking of tf-APP to the lysosomal

D

compartment

TE

Starvation induces trafficking of tf-APP to the lysosome. Given that mTORC1

CE P

is a signalling hub that integrates nutrient sensing and controls lysosomal network activity, we tested whether suppression of mTOR affected tf-APP trafficking. tf-APP and tf-LC3 cells (as a positive comparison) were treated with vehicle or 1 µM

AC

AZD8055 (an inhibitor of mTOR kinase activity) for 24 h. Cell lines were analysed by flow cytometry, which showed that suppression of mTOR significantly increased R/G for both cell lines (Fig. 7A-D). As suppression of mTOR mimicked the effect of starvation, we hypothesized mTORC1 activation should supress starvation-induced increases in R/G for both tfLC3- and tf-APP-expressing cell lines. To activate mTORC1 signalling, cells were transfected with pRK7 encoding a constitutively active mutant of RHEB (Y35N) [29], equipped with a FLAG tag. Control cells were transfected with empty pcDNA3.1(+). Under basal conditions, rpS6 is phosphorylated downstream of mTORC1. This

20

ACCEPTED MANUSCRIPT phosphorylation is lost upon starvation due to impairment of mTORC1 signalling. Overexpression of RHEB is sufficient to completely restore rpS6 phosphorylation to

T

basal physiological levels, even during starvation (Fig. 8A), indicating it activates

IP

mTORC1. RHEB over-expression significantly decreased R/G for tf-LC3 in both

SC R

basal and starved states, as measured by flow cytometry (Fig. 8B-D). Of note, the striking effect of RHEB on lysosomal flux of tf-LC3 was absent for

NU

tf-APP. As before, starvation induced an increase in R/G for the empty vector (pcDNA3.1(+)) control transfected tf-APP cells. However, in stark contrast to tf-LC3

MA

cells, RHEB overexpression to activate mTORC1 had little effect on the trafficking of tf-APP, particularly in starved cells (Fig. 8E-G). FRIA of transiently expressed APP-

D

mCherry labelled at the cell surface with FITC and chased for 4 hours also showed

TE

that in contrast to agents such as dynasore and chloroquine (Fig. 6), RHEB overexpression was unable to rescue starvation-induced lysosomal trafficking

CE P

(Supplementary Fig. 4). Taken together, these data show starvation must work primarily through an alternate signalling pathway to influence the degradative

AC

trafficking of APP.

4. Discussion Here, we have created and characterised a novel molecular tool that can measure trafficking of APP/APP-CTFs to the endo-lysosomal system. To our knowledge, no other method has been devised that can easily measure the passage of this important molecule to the degradative endo-lysosomal network. Tools that reliably measure the trafficking of APP as it is sorted through the cell are important for identifying molecular targets that can be used to enhance the clearance of 21

ACCEPTED MANUSCRIPT APP/APP-CTFs via lysosomal hydrolysis. To date, this process has been largely neglected in preference for studying the α-secretase pathway, which is thought to

T

occur on the plasma membrane. Although the α-secretase pathway is an important

IP

process for anti-amyloidogenic APP turnover, effective Alzheimer’s disease

SC R

therapies that modulate this protease have yet to be generated. Understanding other APP clearance pathways, such as the endo-lysosomal system, could therefore prove

NU

fruitful for Alzheimer’s disease research.

The tool presented in this study does have some limitations. The tf-APP

MA

fusion protein exists primarily as a cleaved fragment and not full length APP. This can be seen in western blots using anti-carboxy terminus APP (Fig. 2C) and anti-

D

GFP (Fig. 2F) by comparing the 70 kDa band (cleavage fragment) with the 160 kDa

TE

band (full length fusion protein). Although the 70 kDa fragment is consistent with tfAPP-CTF, its identity is not known with complete certainty and therefore the effect of

our analysis.

CE P

this fragment on the trafficking events reported in this study represents a limitation in

AC

The enhanced lysosomal trafficking of APP in response to stimuli described here is consistent with previous research on plasma membrane proteostasis. Starvation,

which

up-regulates

autophagic

and

lysosomal

activity

through

suppression of mTORC1-mediated phosphorylation of ULK1, ATG13 and TFEB [30, 31, 32, 33, 34] not only causes increased turnover of autophagic cargo but was recently shown to cause selective remodelling of the plasma membrane in yeast. In this system, endocytosis trafficked membrane proteins to the vacuole (the yeast equivalent of the lysosome) to be hydrolysed as a source of amino acids that provide material to fuel synthesis of more lysosomal proteins. These newly-synthesised lysosomal proteins enhance hydrolysis of a later wave of autophagic cargo [26]. 22

ACCEPTED MANUSCRIPT As starvation-induced autophagy is tightly regulated by mTOR signalling, we suppressed and activated this pathway to determine its effect on the tf-APP probe.

T

As expected, inhibition of mTOR activity using AZD8055 enhanced autophagic flux,

IP

as shown using tf-LC3. APP also displayed robust trafficking to the degradative

SC R

endo-lysosomal compartment upon suppression of mTOR activity. How starvation or suppression of mTOR causes the degradative trafficking of APP represents a large gap in our knowledge. However, mTORC1 signalling regulates autophagy, and

NU

because autophagy interacts with endo-lysosomal trafficking by the fusion of

MA

autophagosomes with late-endosomes to form amphisomes, we cannot exclude the possibility that enhanced lysosomal trafficking of APP is caused by enhanced

D

autophagic flux.

TE

Up-regulating mTOR activity by overexpressing active RHEB strongly reduced lysosomal flux of tf-LC3, as expected. This repression was evident in both starved

CE P

and non-starved conditions, indicating that even in nutrient replete medium with FCS, mTOR can be further stimulated. Contrary to predictions, and even though tf-

AC

APP was strongly influenced by mTOR suppression with AZD8055, activation of mTOR through RHEB over-expression did not inhibit the starvation-induced lysosomal trafficking of tf-APP. This means that, although mTOR inhibition is sufficient to increase endo-lysosomal trafficking of APP, an additional nutrientsensing signalling system is necessary for starvation-mediated degradation of APP.

5. Conclusions Dissection of the mechanisms that drive the lysosomal degradation of APP could be important for exploring the treatment of Alzheimer’s disease. Experiments 23

ACCEPTED MANUSCRIPT in mouse models have already shown that lysosomal degradation of APP and Aβ reduces disease burden [12, 13, 14]. Here, we have demonstrated that starvation

T

induces the trafficking of APP to the degradative endo-lysosomal system. Although

IP

inhibition of mTOR was sufficient to replicate this effect, starvation signals through

SC R

an alternate mechanism to mTORC1. The constructs described in this paper provide useful tools for quickly testing whether a compound (demonstrated by AZD8055) or a gene (demonstrated by over-expression of RHEB) can influence the lysosomal flux

AC

CE P

TE

D

MA

NU

of autophagic cargo or of dementia-related proteins such as APP.

24

ACCEPTED MANUSCRIPT Declarations Ethics approval and consent to participate

IP

T

Not applicable

SC R

Consent for publication Not applicable

NU

Availability of data and materials

The data sets used and/or analysed during the current study are available

MA

from the corresponding author on reasonable request.

D

Competing interests

TE

Dr Sargeant has a patent (2017901032) pending.

CE P

Funding

Funding for this work came from the Lysosomal Diseases Research Unit,

AC

Nutrition and Metabolism Theme, SAHMRI. Author contributions LKH generated the cell lines and performed transfection and western blotting. KH analysed and interpreted confocal microscopy images. RG and LKH analysed and interpreted flow cytometry data. PMA analysed and interpreted fluorescence ratiometric image analysis. TJS, LKH, PMA, JX and CGP designed experiments or provided significant expertise. All authors were involved in the drafting process, and read and approved the final manuscript. Acknowledgements Not applicable. 25

ACCEPTED MANUSCRIPT References

hypothesis, Science 256 (1992) 184-185.

IP

T

[1] J.A. Hardy, G.A. Higgins, Alzheimer's disease: the amyloid cascade

SC R

[2] R. Vassar, B.D. Bennett, S. Babu-Khan, S. Kahn, E.A. Mendiaz, P. Denis, D.B. Teplow, S. Ross, P. Amarante, R. Loeloff, Y. Luo, S. Fisher, J. Fuller, S.

NU

Edenson, J. Lile, M.A. Jarosinski, A.L. Biere, E. Curran, T. Burgess, J.C. Louis, F. Collins, J. Treanor, G. Rogers, M. Citron, Beta-secretase cleavage

MA

of Alzheimer's amyloid precursor protein by the transmembrane aspartic protease BACE, Science 286 (1999) 735-741.

D

[3] C. Haass, D.J. Selkoe, Cellular processing of beta-amyloid precursor protein

TE

and the genesis of amyloid beta-peptide, Cell 75 (1993) 1039-1042.

CE P

[4] K. Kamino, H.T. Orr, H. Payami, E.M. Wijsman, M.E. Alonso, S.M. Pulst, L. Anderson, S. O’dahl, E. Nemens, J.A. White, A.D. Sadovnick, M.J. Ball, J. Kaye, A. Warren, M. McInnis, S.E. Antonarakis, J.R. Korenberg, V. Sharma,

AC

W. Kukull, E. Larson, L.L. Heston, G.M. Martin, T.D. Bird, G.D. Schellenberg, Linkage and mutational analysis of familial Alzheimer disease kindreds for the APP gene region, Am. J. Hum. Genet. 51 (1992) 998-1014. [5] T. Jonsson, J.K. Atwal, S. Steinberg, J. Snaedal, P.V. Jonsson, S. Bjornsson, H. Stefansson, P. Sulem, D. Gudbjartsson, J. Maloney, K. Hoyte, A. Gustafson, Y. Liu, Y. Lu, T. Bhangale, R.R. Graham, J. Huttenlocher, G. Bjornsdottir, O.A. Andreassen, E.G. Jönsson, A. Palotie, T.W. Behrens, O.T. Magnusson, A. Kong, U. Thorsteinsdottir, R.J. Watts, K. Stefansson,

A

mutation in APP protects against Alzheimer's disease and age-related cognitive decline, Nature 488 (2012) 96-99. 26

ACCEPTED MANUSCRIPT [6] T.E. Golde, S. Estus, L.H. Younkin, D.J. Selkoe, S.G. Younkin, Processing of the amyloid protein precursor to potentially amyloidogenic derivatives,

T

Science 255 (1992) 728-730.

IP

[7] C. Haass, E.H. Koo, A. Mellon, A.Y. Hung, D.J. Selkoe, Targeting of cell-

SC R

surface beta-amyloid precursor protein to lysosomes: alternative processing into amyloid-bearing fragments, Nature 357 (1992) 500-503. [8] E. Morel, Z. Chamoun, Z.M. Lasiecka, R.B. Chan, R.L. Williamson, C.

NU

Vetanovetz, C. Dall‘Armi, S. Simoes, K.S. Point Du Jour, B.D. McCabe, S.A.

MA

Small, G. Di Paolo, Phosphatidylinositol-3-phosphate regulates sorting and processing of amyloid precursor protein through the endosomal system, Nat.

D

Commun. 2013; doi: 10.1038/ncomms3250.

TE

[9] F. Ubelmann, T. Burrinha, L. Salavessa, R. Gomes, C. Ferreira, N. Moreno, C. Guimas Almeida, Bin1 and CD2AP polarise the endocytic generation of

CE P

beta-amyloid, EMBO Rep. 18 (2017) 102-122. [10] J.R. Edgar, K. Willén, G.K. Gouras, C.E. Futter, ESCRTs regulate amyloid

AC

precursor protein sorting in multivesicular bodies and intracellular amyloid-β accumulation, J. Cell Sci. 128 (2015) 2520-2528. [11] G. Swaminathan, W. Zhu, E.D. Plowey, BECN1/Beclin 1 sorts cell-surface APP/amyloid β precursor protein for lysosomal degradation, Autophagy 12 (2016) 2404-2419. [12] D.S. Yang, P. Stavrides, P.S. Mohan, S. Kaushik, A. Kumar, M. Ohno, S.D. Schmidt, D. Wesson, U. Bandyopadhyay, Y. Jiang, M. Pawlik, C.M. Peterhoff, A.J. Yang, D.A. Wilson, P. St George-Hyslop, D. Westaway, P.M. Mathews, E. Levy, A.M. Cuervo, R.A. Nixon, Reversal of autophagy dysfunction in the

27

ACCEPTED MANUSCRIPT TgCRND8 mouse model of Alzheimer's disease ameliorates amyloid pathologies and memory deficits, Brain 134 (2011) 258-277.

T

[13] D.S. Yang, P. Stavrides, M. Saito, A. Kumar, J.A. Rodriguez-Navarro, M.

IP

Pawlik, C. Huo, S.U. Walkley, M. Saito, A.M. Cuervo, R.A. Nixon, Defective

SC R

macroautophagic turnover of brain lipids in the TgCRND8 Alzheimer mouse model: prevention by correcting lysosomal proteolytic deficits, Brain 137 (2014) 3300-3318.

NU

[14] Q. Xiao, P. Yan, X. Ma, H. Liu, R. Perez, A. Zhu, E. Gonzales, D.L. Tripoli, L.

MA

Czerniewski, A. Ballabio, J.R. Cirrito, A. Diwan, J.M. Lee, Neuronal-Targeted TFEB Accelerates Lysosomal Degradation of APP, Reducing Aβ Generation

D

and Amyloid Plaque Pathogenesis, J. Neurosci. 35 (2015) 12137-12151.

TE

[15] S. Kimura, T. Noda, T. Yoshimori, Dissection of the autophagosome maturation process by a novel reporter protein, tandem fluorescent-tagged

CE P

LC3, Autophagy 3 (2007) 452-460. [16] T.L. Young-Pearse, J. Bai, R. Chang, J.B. Zheng, J.J. LoTurco, D.J. Selkoe, A

AC

critical function for beta-amyloid precursor protein in neuronal migration revealed by in utero RNA interference, J. Neurosci. 27 (2007) 14459-14469. [17] K. Weber, U. Bartsch, C. Stocking, B. Fehse, A multicolor panel of novel lentiviral "gene ontology" (LeGO) vectors for functional gene analysis, Mol. Ther. 16 (2008) 698-706. [18] S.A. Stewart, D.M. Dykxhoorn, D. Palliser, H. Mizuno, E.Y. Yu, D.S. An, D.M. Sabatini, I.S. Chen, W.C. Hahn, P.A. Sharp, R.A. Weinberg, C.D. Novina, Lentivirus-delivered stable gene silencing by RNAi in primary cells, RNA 9 (2003) 493-501.

28

ACCEPTED MANUSCRIPT [19] H. Barrière, P. Apaja, T. Okiyoneda, G.L. Lukacs, Endocytic sorting of CFTR variants monitored by single-cell fluorescence ratiometric image analysis

T

(FRIA) in living cells, Methods Mol. Biol. 2011; doi: 10.1007/978-1-61779-117-

IP

8_20.

SC R

[20] P.H. Kuhn, H. Wang, B. Dislich, A. Colombo, U. Zeitschel, J.W. Ellwart, E. Kremmer, S. Rossner, S.F. Lichtenthaler, ADAM10 is the physiologically relevant, constitutive alpha-secretase of the amyloid precursor protein in

NU

primary neurons, EMBO J. 29 (2010) 3020-3032.

MA

[21] J.Y. Qin, L. Zhang, K.L. Clift, I. Hulur, A.P. Xiang, B.Z. Ren, B.T. Lahn, Systematic comparison of constitutive promoters and the doxycycline-

D

inducible promoter, PLoS One 2010; doi: 10.1371/journal.pone.0010611.

TE

[22] J.M. Gump, A. Thorburn, Sorting cells for basal and induced autophagic flux by quantitative ratiometric flow cytometry, Autophagy 10 (2014) 1327-1334.

CE P

[23] N.C. Shaner, P.A. Steinbach, R.Y. Tsien, A guide to choosing fluorescent proteins, Nat. Methods 2 (2005) 905-909.

AC

[24] R.E. Campbell, O. Tour, A.E. Palmer, P.A. Steinbach, G.S. Baird, D.A. Zacharias, R.Y. Tsien, A monomeric red fluorescent protein, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 7877-7882. [25] C.Y. Ho, C.H. Choy, C.A. Wattson, D.E. Johnson, R.J. Botelho. The Fab1/PIKfyve phosphoinositide phosphate kinase is not necessary to maintain the pH of lysosomes and of the yeast vacuole, J. Biol. Chem. 290 (2015) 9919-9928. [26] M. Müller, O. Schmidt, M. Angelova, K. Faserl, S. Weys, L. Kremser, T. Pfaffenwimmer, T. Dalik, C. Kraft, Z. Trajanoski, H. Lindner, D. Teis, The

29

ACCEPTED MANUSCRIPT coordinated action of the MVB pathway and autophagy ensures cell survival during starvation, Elife 2015; doi: 10.7554/eLife.07736.

T

[27] J.H. Lee, W.H. Yu, A. Kumar, S. Lee, P.S. Mohan, C.M. Peterhoff, D.M.

IP

Wolfe, M. Martinez-Vicente, A.C. Massey, G. Sovak, Y. Uchiyama, D.

SC R

Westaway, A.M. Cuervo, R.A. Nixon. Lysosomal proteolysis and autophagy require presenilin 1 and are disrupted by Alzheimer-related PS1 mutations, Cell 141 (2010) 1146-1158.

NU

[28] R. Ehehalt, P. Keller, C. Haass, C. Thiele, K. Simons, Amyloidogenic

MA

processing of the Alzheimer beta-amyloid precursor protein depends on lipid rafts, J. Cell Biol. 160 (2003) 113-123.

D

[29] A.P. Ghosh, C.B. Marshall, T. Coric, E.H. Shim, R. Kirkman, M.E. Ballestas,

TE

M. Ikura, M.A. Bjornsti, S. Sudarshan, Point mutations of the mTOR-RHEB pathway in renal cell carcinoma, Oncotarget 6 (2015) 17895-17910.

CE P

[30] N. Hosokawa, T. Hara, T. Kaizuka, C. Kishi, A. Takamura, Y. Miura, S. Iemura, T. Natsume, K. Takehana, N. Yamada, J.L. Guan, N. Oshiro, N.

AC

Mizushima, Nutrient-dependent mTORC1 association with the ULK1-Atg13FIP200 complex required for autophagy, Mol. Biol. Cell 20 (2009) 1981-1991. [31] C.H. Jung, C.B. Jun, S.H. Ro, Y.M. Kim, N.M. Otto, J. Cao, M. Kundu, D.H. Kim, ULK-Atg13-FIP200 complexes mediate mTOR signaling to the autophagy machinery, Mol. Biol. Cell 20 (2009) 1992-2003. [32] I.G.

Ganley,

D.H.

Lam,

J. Wang,

X.

Ding,

S.

Chen,

X.

Jiang,

ULK1.ATG13.FIP200 complex mediates mTOR signaling and is essential for autophagy, J. Biol. Chem. 284 (2009) 12297-12305. [33] C. Settembre, R. Zoncu, D.L. Medina, F. Vetrini, S. Erdin, S. Erdin, T. Huynh, M. Ferron, G. Karsenty, M.C. Vellard, V. Facchinetti, D.M. Sabatini, A.

30

ACCEPTED MANUSCRIPT Ballabio, A lysosome-to-nucleus signalling mechanism senses and regulates the lysosome via mTOR and TFEB, EMBO J. 31 (2012) 1095-1108.

T

[34] S. Peña-Llopis, S. Vega-Rubin-de-Celis, J.C. Schwartz, N.C. Wolff, T.A. Tran,

IP

L. Zou, X.J. Xie, D.R. Corey, J. Brugarolas, Regulation of TFEB and V-

AC

CE P

TE

D

MA

NU

SC R

ATPases by mTORC1, EMBO J. 30 (2011) 3242-3258.

31

ACCEPTED MANUSCRIPT Figure legends

IP

T

Fig. 1. Constructs used to measure endo-lysosomal trafficking of LC3 and APP. A) A lentiviral vector containing the human ubiquitin C promoter (hUC) was chosen for

SC R

expression of tf-APP and tf-LC3 fusion proteins. B) tf-LC3 is sequestered by autophagosomes and EGFP quenches on acidification. C) EGFP that is a part of

NU

endocytosed tf-APP quenches once the molecule is taken into the lumen of the late

MA

endosome.

D

Fig. 2. Fluorescence staining and western blot analysis of tf-APP-expressing HeLa

TE

cells. A) tf-APP, shown by mCherry fluorescence (red), co-localises with antibodies directed towards Aβ sequence (6E10) and the carboxy-terminus of APP (CTF)

CE P

(pseudo-coloured green, secondary antibody conjugated to Alexa-647 and imaged in far-red spectrum). B) HeLa cells that do not express the tf-APP construct display

AC

similar staining patterns with each antibody showing endogenous APP (green). Western blot analyses of non-transgenic HeLa, tf-APP HeLa and tf-LC3 HeLa for the carboxy-terminus of APP (C), the amino-terminus of APP (clone 22C11, D), LC3 (E), GFP (F) and β-actin (G). Scale bars = 10 µm.

Fig. 3. Starvation induces endo-lysosomal trafficking of APP. Histograms show single experiments where flow cytometry was used to quantify R/G in tf-LC3 (A, B) and tf-APP (D, E) expressing HeLa cells in nutrient-replete, or basal conditions, under EBSS-induced starvation or starvation with CQ for 4 h. Quantification of data from flow cytometry is shown (C, F). Graphs show mean ± SD, n = 3 individual 32

ACCEPTED MANUSCRIPT experiments. Flow cytometry data was quantified as percentage of cells in gate that extends to the right hand side from basal media population peak. One-way ANOVA

T

and Tukey’s multiple comparisons test were used to determine statistical

SC R

IP

significance: * = P < 0.05, ** = P < 0.01, *** = P < 0.001.

Fig. 4. Analysis of tf-APP subcellular localisation with and without starvation. A)

NU

Immunofluorescence images show overlay between LAMP1 (pseudo-coloured green, secondary antibody conjugated to Alexa-647 and imaged in far-red spectrum)

MA

and mRFP1 (red) in tf-LC3 expressing HeLa cells. Inhibition of lysosomal proteolysis (Leu, leupeptin, µg/ml) or starvation in EBSS (4 h) is indicated on the left. B) Co-

D

localisation between LAMP1 (green) and tf-APP (mCherry, red) is shown as for

TE

panel A. Overlap of red fluorescence from either tf-LC3 or tf-APP with LAMP1

CE P

immunoreactivity in the presence of leupeptin is expressed as mean ± SEM for 5 independent experiments. Starved cells were normalised to controls. Paired t-tests were performed on raw data and p value is expressed on each graph, which show

AC

data standardised to paired controls. D) Co-localisation between Golgi apparatus marker TGN46 (green) and tf-APP (mCherry, red) is shown. Scale bars = 10 µm.

Fig. 5. Starvation directs APP to traffic into late endosomes and lysosomes. (A) HeLa cells expressing tf-APP were starved in EBSS and vesicles were analysed using FRIA. Representative histogram of tf-APP positive vesicles shows APP to reside in pH 6.1 at basal condition, but moving toward lysosomes upon starvation (peaks in pH 5.6, 5.3 and 4.4). Number of total vesicles is indicated. (B) Representative histogram for dynamin inhibitor, dynasore, which reduces trafficking 33

ACCEPTED MANUSCRIPT of tf-APP into late endosomes/lysosomes upon starvation. Number of total vesicles is indicated. (C) Mean vesicular pH of different treatment conditions for tf-APP

T

measured using FRIA is presented. (D) Mean vesicular pH of different treatment

SC R

four to seven independent experiments. ***P < 0.001.

IP

conditions for tf-LC3 measured using FRIA is presented. Error bars indicate SEM of

NU

Fig. 6. Plasma membrane labelled transgenic APP recycles, but is trafficked to the late endosome upon starvation. (A) Illustration of FRIA for plasma membrane

MA

occupying APP-mCherry trafficking and labelling using the antibody against the amino terminus of APP. (B, C) Representative histogram of APP-mCherry-FITC

D

containing vesicles. Plasma membrane APP was labelled on ice. In basal conditions,

TE

the majority of APP molecules are in recycling endosomes (pH 6.3), but when

CE P

exposed to starvation are directed to late endosomes (peak pH of 5.8 and 5.4). (D, E) Plasma membrane FITC-labelled APP-mCherry targeting to late endosomes upon starvation is completely suppressed by dynamin inhibitor dynasore. Mean vesicular

AC

pH is indicated. (F) Mean vesicular pH of plasma membrane labelled APP-mCherry vesicles after 4h chase was measured using FRIA. Treatment conditions are indicated. Transferrin was used as a control for receptor recycling. Error bars indicate SEM of four independent experiments. Student’s t-test was used. *P < 0.05, **<0.01.

Fig. 7. Inhibition of mTOR is sufficient to induce tf-LC3 (A, B) and tf-APP (C,D) trafficking to lysosomes in HeLa cells treated with AZD8055 for 24 h. R/G was measured using flow cytometry. Quantified results are shown as mean ± SD. N = 3 34

ACCEPTED MANUSCRIPT individual experiments. Flow cytometry data was quantified as percentage of cells in gate that extends to the right hand side from vehicle population peak. Statistical

IP

T

significance was determined using Student’s t-test. ** = P < 0.01, *** = P < 0.001.

SC R

Fig. 8. Activation of mTOR is not sufficient to rescue starvation-induced lysosomal trafficking of APP. (A) tf-LC3- and tf-APP-expressing cells were maintained in

NU

nutrient-replete conditions or starved. Cells were also transfected with a control vector (empty pcDNA3.1 (+)) or with a plasmid that expressed FLAG-RHEB (Y35N).

MA

Western blots show expression of FLAG-RHEB (anti-FLAG M2) and p-rpS6 as a measurement of mTORC1 activity. R/G for tf-LC3- (B-D) or tf-APP-expressing (E-G)

D

HeLa cells was measured using flow cytometry. Histograms showing data from flow

TE

cytometry (B, C, E, F) was quantified (D, G) and shown as mean ± SD. N = 3

CE P

individual experiments. Flow cytometry data was quantified as percentage of cells in gate that extends to the right hand side from empty vector control population peak for every condition. Statistical significance was determined using Student’s t-test. ***

AC

= P < 0.001.

35

AC

Fig. 1

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

36

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Fig. 2

37

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Fig. 3

38

Fig. 4

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

39

AC

Fig. 5

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

40

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 6

41

AC

Fig. 7

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

42

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Fig. 8

43

ACCEPTED MANUSCRIPT Highlights

IP

T

 We developed a new tool that measures trafficking of APP/APP-CTF to the lysosome

SC R

 Starvation regulates endo-lysosomal trafficking of APP

 Starvation-mediated lysosomal trafficking of APP is phenocopied by mTOR inhibition

AC

CE P

TE

D

MA

NU

 An nutrient sensing mTOR-independent pathway also regulates APP trafficking

44