The orphan drug dichloroacetate reduces amyloid beta-peptide production whilst promoting non-amyloidogenic proteolysis of the amyloid precursor protein.

The amyloid cascade hypothesis proposes that excessive accumulation of amyloid beta-peptides is the initiating event in Alzheimer's disease. These neurotoxic peptides are generated from the amyloid precursor protein via sequential cleavage by β- and γ-secretases in the 'amyloidogenic' proteolytic pathway. Alternatively, the amyloid precursor protein can be processed via the 'non-amyloidogenic' pathway which, through the action of the α-secretase a disintegrin and metalloproteinase (ADAM) 10, both precludes amyloid beta-peptide formation and has the additional benefit of generating a neuroprotective soluble amyloid precursor protein fragment, sAPPα. In the current study, we investigated whether the orphan drug, dichloroacetate, could alter amyloid precursor protein proteolysis. In SH-SY5Y neuroblastoma cells, dichloroacetate enhanced sAPPα generation whilst inhibiting β-secretase processing of endogenous amyloid precursor protein and the subsequent generation of amyloid beta-peptides. Over-expression of the amyloid precursor protein partly ablated the effect of dichloroacetate on amyloidogenic and non-amyloidogenic processing whilst over-expression of the β-secretase only ablated the effect on amyloidogenic processing. Similar enhancement of ADAM-mediated amyloid precursor protein processing by dichloroacetate was observed in unrelated cell lines and the effect was not exclusive to the amyloid precursor protein as an ADAM substrate, as indicated by dichloroacetate-enhanced proteolysis of the Notch ligand, Jagged1. Despite altering proteolysis of the amyloid precursor protein, dichloroacetate did not significantly affect the expression/activity of α-, β- or γ-secretases. In conclusion, dichloroacetate can inhibit amyloidogenic and promote non-amyloidogenic proteolysis of the amyloid precursor protein. Given the small size and blood-brain-barrier permeability of the drug, further research into its mechanism of action with respect to APP proteolysis may lead to the development of therapies for slowing the progression of Alzheimer's disease.

Introduction

The amyloid cascade hypothesis [1] states that the deposition of amyloid beta (Aβ)-peptides in the brain is the central event in the pathogenesis of the neurodegenerative condition, Alzheimer’s disease (AD). These neurotoxic peptides are generated from the larger amyloid precursor protein (APP) via the ’amyloidogenic’ pathway in which the protein is cleaved initially by β-secretase (β-site APP cleaving enzyme 1; BACE1) to liberate an N-terminal ectodomain termed sAPPβ and a residual beta C-terminal fragment (βCTF) [2]. This latter fragment is then cleaved by a multi-subunit protease complex known as γ-secretase in which the catalytic site resides on presenilin-1 or -2 [2, 3]. In the alternative ’non-amyloidogenic’ pathway, the zinc metalloproteinase, ADAM10 (a disintegrin and metalloproteinase 10), cleaves APP within the Aβ sequence, thereby precluding the formation of intact Aβ-peptides and releasing a soluble, neuroprotective, N-terminal ectodomain termed sAPPα [4] whilst leaving behind in the membrane an alpha C-terminal fragment (αCTF). ADAM 10 also mediates the proteolytic cleavage of a range of additional cell surface integral membrane proteins within their juxtamembrane region releasing a soluble protein ectodomain into the extracellular space; a process known as ’ectodomain shedding’ [5].

Dichloroacetate (DCA) is used for alleviating the symptoms of lactic acidosis associated with a number of congenital mitochondrial diseases in children [6]. Over the past few decades, chronic DCA administration has been clinically tested in adults and children in a number of disease states associated with lactic acidosis including diabetes mellitus and pulmonary arterial hypertension [7-9]. The beneficial effects of the drug in these conditions stem from its ability to activate the pyruvate dehydrogenase complex (PDC) which catalyzes the irreversible oxidative phosphorylation of pyruvate to acetyl-Coenzyme A (acetyl-CoA) thereby reducing the amount of pyruvate available for conversion to lactate [10]. This activation of the PDC is achieved through the ability of DCA to inhibit the phosphorylation and associated inactivation of the complex by pyruvate dehydrogenase kinase (PDK) [10]. More recently, and perhaps more controversially, DCA has been assigned a potential role in the treatment of various cancers [11] due to its ability to reverse the ’Warburg effect’ [12] whereby cancer cells are proposed to utilize cytosolic glycolysis as a means of energy generation as opposed to the more efficient mitochondrial conversion of pyruvate to acetyl-CoA and the subsequent tricarboxylic acid cycle. The activation of the PDC by DCA is thought to reinstate the latter as the main energy generating pathway in cancer cells thereby promoting mitochondrial-mediated apoptosis.

It was the putative anti-neoplastic effect of DCA that led us to initially investigate whether the drug might impact on additional pathways linked to the proliferation of cancer cells. During these studies we remarked that the ADAM-mediated shedding of the Notch ligand, Jagged1 was enhanced by DCA. As this protein was known to be shed in a cell-specific manner by both ADAM10 and ADAM17 [13, 14], we postulated that additional substrates of these enzymes might also be affected. Therefore, in the current study, we investigated whether DCA might impact on the proteolysis of the amyloid precursor protein. When SH-SY5Y neuroblastoma cells were treated with DCA a dramatic increase in the generation of sAPPα was observed consistent with more of the APP holoprotein being cleaved via the non-amyloidogenic pathway. Conversely, the generation of sAPPβ by β-secretase activity was severely impaired with a concomitant reduction in the subsequent generation of both Aβ1–40 and Aβ1–42. Notably, these effects were distinct from any effects that DCA had on culture cell number; in this respect the drug exhibited an anti-proliferative rather than cytotoxic effect. APP over-expression in SH-SY5Y cells partly ablated the effects of DCA on both the non-amyloidogenic and amyloidogenic processing of the protein. In contrast, the over-expression of BACE1 in SH-SY5Y cells only partly ablated the effects of DCA on amyloidogenic but not non-amyloidogenic APP processing. We were also able to demonstrate DCA-enhanced shedding of APP in two unrelated cell lines along with that of an unrelated ADAM substrate, Jagged1, indicating that the effect is not cell- or substrate-specific. Finally, we show that DCA did not regulate the expression of the secretases ADAM10, BACE1 or presenilin-1 or ADAM10 and BACE1 activity. These data suggest that, as DCA is a small molecule able to traverse the blood-brain-barrier, further research into its mechanism of action with respect to APP proteolysis may lead to the development of therapies for slowing the progression of Alzheimer’s disease.

Materials and methods

Materials

The generation of the APP695 and BACE1 constructs in the mammalian expression vector pIREShyg (Clontech-Takara Bio Europe, Saint-Germain-en-Laye, France) along with their expression and characterisation have been reported previously [15]. Anti-actin monoclonal (Cat. No. A4700), anti-BACE1 rabbit polyclonal (Cat. No. SAB2100200), anti-ADAM10 rabbit polyclonal (Cat. No. AB19026) and anti-APP C-terminal rabbit polyclonal (Cat. No. A8717) antibodies were from Sigma-Aldrich Company Ltd. (Poole, U.K.). Anti-APP 6E10 monoclonal (Cat. No. AB2564653) and anti-sAPPβ rabbit polyclonal (Cat. No. AB2564769) antibodies were from Biolegend (San Diego, U.S.A.). Anti-Jagged1 C-terminal goat polyclonal antibody (Cat. No. SC6011) was from Santa Cruz Biotechnology Inc. (California, U.S.A.) and anti-Jagged1 N-terminal (Cat. No. AF1277) antibody was from R & D Systems Ltd. (Abingdon, U.K.). All other materials, unless otherwise stated, were purchased form Sigma-Aldrich Company Ltd. (Poole, U.K.).

Cell culture

Cell culture reagents were purchased from Scientific Laboratory Supplies (Nottingham, U.K.) and Thermofisher Scientific (Waltham, U.S.A.). The human colon cancer cell line SW480, HEK293 and human neuroblastoma SH-SY5Y cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 25 mM glucose, 4 mM L-glutamine, 10% (vol/vol) foetal bovine serum (FBS), penicillin (50 U/mL) and streptomycin (50 μg/mL). Cells were maintained at 37°C in 5% CO2 in air.

Stable DNA transfections

Plasmids (8 μg) were linearized using AhdI before being subjected to ethanol precipitation and subsequent introduction into SH-SY5Y cells by electroporation. Recombinant cells were selected using 150 μg/ml of Hygromycin B (Invitrogen, Paisley, U.K.).

Treatment of cells and protein extraction

SH-SY5Y cells were treated with 0, 10 or 20 mM final concentrations of DCA either at the point of seeding or once the cells had reached confluency. In the former case, the growth medium and DCA were replaced every two days until the control cells reached confluence. At this point, the growth medium was removed and cells were washed in situ with 10 mL of UltraMEM (Scientific Laboratory Supplies, Nottingham, U.K.) before adding a fresh 10 mL of UltraMEM containing the required DCA concentration and culturing for a further 24 h. Cells grown to confluence before treating with DCA were similarly washed with UltraMEM and then cultured in the same medium with or without DCA for a further 24 h. Note that a 24 h period was chosen as UltraMEM is a low serum medium (facilitating protein analysis by immunoblotting without the distortion of gels by large amounts of albumin derived from FBS in complete medium) and, therefore, maintenance of cells in this medium becomes increasingly poor after 24 h. Following the final 24 h incubations, medium was harvested, centrifuged at 10,000 g for 10 min to remove cell debris, and then equal volumes were concentrated 50-fold using Amicon Ultra-4 centrifugal filter units (Merck Millipore, Watford, U.K.). For analysis of cell-associated proteins, cells were washed with phosphate-buffered saline (PBS; 20 mM Na2HPO4, 2 mM NaH2PO4, 0.15 M NaCl, pH 7.4) and scraped from the flasks into fresh PBS (10 mL). Following centrifugation at 500 g for 5 min, cell pellets were lysed in 0.1 M Tris, 150 mM NaCl, 1% (vol/vol) Triton X-100, 0.1% (vol/vol) Nonidet P-40, pH 7.4 containing a protease inhibitor cocktail (Sigma-Aldrich Company Ltd., Poole, U.K.).

Protein assay, sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblot analysis

Protein levels in cell lysates were quantified using bicinchoninic acid [16] in a microtitre plate with bovine serum albumin as a standard. Equal quantities of lysate protein and equal volumes of concentrated conditioned medium samples were resolved by SDS-PAGE using 5–20% polyacrylamide gradient gels and transferred to Immobilon P polyvinylidene difluoride (PVDF) membranes [17] before blocking in 5% (w/v) powdered milk in PBS containing 0.1% (vol/vol) Tween 20 (PBS-Tween) and incubating with primary antibody. Bound antibody was detected using peroxidase-conjugated secondary antibodies (Sigma-Aldrich Company Ltd, Poole, U.K. and R&D Systems Ltd., Abingdon, U.K.) in conjunction with enhanced chemiluminescence detection reagents (Perbio Science Ltd., Cramlington, U.K.). For re-probing with anti-actin antibody, the original antibodies were stripped from membranes by heating at 50°C for 30 min in 100 mM β-mercaptoethanol, 2% (wt/vol) SDS, 62.5 mM Tris pH 6.7 with occasional agitation. The membranes were then washed twice for 10 min in PBS-Tween before blocking and re-probing with primary antibody as described above.

For the resolution of APP C-terminal fragments, samples were run on 16% Tris/tricine gels before transferring to 0.2 micron nitrocellulose (Fisher Scientific, Loughborough, U.K.) and subsequently boiling for 5 min in PBS. Membranes were then further processed as described above.

Cell viability assays

For trypan blue assays, cells were harvested by trypsinisation and an aliquot from the resultant resuspended cell pellet was mixed with an equal volume of 0.4% (wt/vol) trypan blue solution and loaded onto a haemocytometer. Cell counts of live cells were obtained and an average of four squares taken. For MTS (methanethiosulfonate) assays, cells were incubated with CellTiter 96® AQueous One Cell Proliferation Assay solution (Promega, Wisconsin, U.S.A.) for 2 h at 37°C. Absorbance readings at 490 nm were then taken using a Victor2 1420 microplate reader (Perkin Elmer, Waltham, U.S.A.). Note that the two types of viability assays were performed separately from each other using independent cultures.

Aβ-peptide quantification

Aβ-peptides in unconcentrated conditioned medium samples were quantified using the Mesoscale Discovery (MSD) platform. Aβ1–40 and Aβ1–42 were measured using the V-Plex Aβ peptide panel (6E10) kit according to the manufacturer’s instructions (MSD, Maryland U.S.A.).

Real-time quantitative polymerase chain reaction (qPCR)

RNA was extracted from SH-SY5Y cells using Trizol (Thermofisher Scientific, Waltham, U.S.A.) following the manufacturer’s instructions and yield was determined by spectroscopy on a Nanodrop 2000 (Thermofisher Scientific, Waltham, U.S.A.). cDNA for real-time qPCR analysis was synthesized using SuperScript III (Thermofisher Scientific, Waltham, U.S.A.) following the manufacturer’s instructions. Typically 1000 ng of RNA was used in each synthesis reaction. Real-time qPCR analysis was performed on a CFX96 thermal cycler (Bio-Rad Laboratories, Watford, U.K.) using SYBR green Jumpstart Taq ready mix (Sigma-Aldrich Company Ltd., Poole, U.K.) and reaction conditions were as follows: initial denaturation at 95°C for 5 min, followed by 40 cycles of 94°C for 1 min, 63°C for 1 min and 68°C for 15 s. Reactions were performed in triplicate in volumes of 20 μl. The analysis was then repeated on triplicate independent biological samples. Relative expression was based on ΔΔCt methodology analysed in CFX software version 3.1 (Bio-Rad Laboratories, Watford, U.K.). Reaction primers for BACE 1 were 5’-CAGTCATCCACGGGCACTGT-3’ (forward) and 5’-CTGAACTCATCGTGCACATGGC-3’ (reverse). Primers for ADAM10 were 5’-GGCTTCACAGCTCTCTGCCCA-3’ (forward) and 5’-CCTGCACATTGCCCATTAATGCA-3’ (reverse). Primers for presenilin-1 were 5’-GCCAGAGAGCCCTGCACTCAA-3’ (forward) and 5’-GCATGGATGACCTTATAGCACC-3’ (reverse). Primers for ribosomal protein L15 (RPLO) were 5’-GCAATGTTGCCAGTGTCTG-3’ (forward) and 5’-GCGTTGACCTTTTCAGCAA-3’ (reverse).

Fluorimetric secretase activity assays

ADAM10 activity in cell lysates was assayed using a SensoLyte® 520 ADAM10 activity fluorimetric assay kit (Anaspec, Fremont, U.S.A.) according to the manufacturer’s instructions. 5-FAM fluorescence was monitored at excitation/emission = 490 nm / 520 nm using an Infinite M200 Pro Tecan plate reader (Tecan Trading, Männedorf, Switzerland). Controls using the ADAM10 inhibitor GM6001 were incorporated and results adjusted to compensate for non-specific substrate degradation. BACE1 activity in cell lysates was measured using a Fluorometric Beta Secretase activity assay kit (Abcam, Cambridge, U.K.) according to the manufacturer’s instructions. Fluorescence was monitored at excitation/emission = 335 nm / 495 nm using the Tecan plate reader described above. Controls using the β-secretase inhibitor supplied as part of the kit were incorporated and results adjusted to compensate for non-specific substrate degradation. Results from both fluorimetric assays were also adjusted to account for differences in total protein concentrations between lysate samples.

Statistical analysis

Data are presented as means ± S.D. and were subjected to statistical analysis via one-way analysis of variance (ANOVA) with Tukey’s post hoc tests (analysed using IBM SPSS software). No outliers were excluded. Levels of significance are indicated in figure legends.

Results

Growth of untransfected SH-SY5Y cells in the presence of DCA enhances non-amyloidogenic and inhibits amyloidogenic processing of endogenous APP

Initially, human SH-SY5Y cells (untransfected) were seeded in the absence or presence of DCA and cultured until the control flasks reached confluence (medium and DCA were replaced every two days). At this point the growth medium was replaced with UltraMEM containing the same DCA concentrations (see Materials and methods) and the cells were cultured for an additional 24 h. Trypan blue assays of cultures at this point demonstrated 25.02 ± 1.37 and 44.93 ± 7.89% decreased viable cell numbers in 10 mM and 20 mM DCA-treated cultures, respectively, relative to controls (Fig 1A). Note that few if any non-viable cells were seen in any of these assays indicating a growth inhibitory rather than cytotoxic effect of DCA. Similarly, MTS cell viability assays demonstrated a 40.00 ± 6.02% decrease in cell number in the 20 mM DCA-treated cultures relative to controls although no significant decrease could be determined at a 10 mM concentration of the drug (Fig 1A).

Fig 1

Growth of untransfected SH-SY5Y cells in the presence of DCA enhances non-amyloidogenic APP processing.

Cells were seeded and cultured in the absence or presence of the indicated DCA concentrations and grown until the control cultures reached confluence (medium and DCA were replaced every two days). At this point the growth medium was replaced with UltraMEM containing the same DCA concentrations and the cells were cultured for an additional 24 h. Viability assays were then performed or cell lysate and conditioned medium samples were prepared as described in the Materials and methods section. Equal amounts of protein (lysates) or equal volumes (medium) from samples were subjected to SDS-PAGE and immunoblotting (Materials and methods). (A) Trypan blue and MTS cell viability assays. (B) Lysates were immunoblotted with anti-APP C-terminal and anti-actin antibodies. Full-length APP (FL-APP) was quantified from multiple immunoblots and the results expressed relative to control values. (C) Lysates were resolved on Tris/tricine gels and immunoblotted with anti-APP C-terminal antibody to detect APP C-terminal fragments (APP-CTFs). The far right lane is a sample from APP over-expressing SH-SY5Y cells and the dashed line shows where lanes from different time exposures of the same immunoblot have been rearranged for illustrative purposes. AICD, APP intracellular domain. C83 was quantified from multiple immunoblots and the results expressed relative to control values. (D) Conditioned medium samples were immunoblotted with anti-APP 6E10 antibody in order to detect sAPPα. Multiple immunoblots were then quantified and results expressed relative to control values without correcting for cell number. (E) The results in (D) were corrected to account for relative changes in cell number using data from the trypan blue assays in (A). All results are expressed relative to control (0 mM DCA) values and are means ± S.D. (n = 3). * = significant at p < 0.05; *** = significant at p < 0.005.

When cell lysates prepared from cultures grown in the presence of DCA were subjected to immunoblotting using anti-APP C-terminal antibody, no significant changes were observed in levels of full-length APP (FL-APP) (Fig 1B). Equal protein loading between lysate samples was confirmed by re-probing immunoblots with anti-actin antibody (Fig 1B). Lysate samples were also resolved on Tris/tricine gels and immunoblotted with the same anti-APP C-terminal antibody in order detect APP C-terminal fragments (APP-CTFs). The results (Fig 1C) demonstrated highly significant increases specifically in the levels of α-secretase-generated C83 of 2.44 ± 0.18- and 2.14 ± 0.12-fold, respectively, in lysates prepared from 10 and 20 mM DCA-treated cells (relative to untreated controls). Levels of C99 and the APP intracellular domain (AICD) were below the limits of detection in untransfected SH-SY5Y cells.

In contrast to lysate samples, conditioned medium samples (from the final 24 h UltraMEM incubation) were resolved on gels on an equal volume (as opposed to equal protein) basis. Immunoblotting of medium samples with anti-APP 6E10 antibody revealed increases of 59.17 ± 27.12 and 42.13 ± 23.98%, respectively, in the levels sAPPα derived from APP695 (lower band) and APP751/770 (combined in the upper band on immunoblots) shed from 10 mM DCA-treated relative to control cells (Fig 1D). The data presented in Fig 1D were then adjusted (using the trypan blue assay data) in order to compensate for the reduction in cell number observed following DCA treatment (i.e. to provide information on the amount of protein shed per unit cell number). The resultant data (Fig 1E) revealed actual 111.70 ± 36.08 and 89.03 ± 31.89% increases, respectively, in the levels of sAPP695α and sAPP751/770α shed from cells treated with 10 mM DCA compared to control cells. Similarly, at 20 mM DCA, the levels of these fragments shed were increased by 96.23 ± 34.51 and 121.80 ± 34.87%, respectively, compared to the controls.

The same conditioned medium samples were then immunoblotted with anti-sAPPβ antibody and the subsequent quantification of multiple immunoblots revealed that sAPPβ generation from APP695 was almost completely inhibited (unquantifiable in some immunoblots) in the presence of either 10 or 20 mM DCA (Fig 2A). Similarly, sAPPβ generated from APP751/770 was reduced by 61.88 ± 12.16 and 84.82 ± 15.67%, respectively, in 10 mM and 20 mM DCA-treated cultures compared to controls. Even when these data were adjusted in order to account for decreased cell numbers following DCA treatment the results still revealed 49.30 ± 16.17 and 72.37 ± 28.52% reductions, respectively, in the levels of sAPPβ generated from APP751/770 following 10 and 20 mM DCA treatments (Fig 2B). When unconcentrated conditioned medium samples were analysed in terms of Aβ-peptide content, the results (Fig 2C and 2D) showed that Aβ1–40 was reduced by 69.46% (59.60% corrected for cell number) and 94.72% (90.40% corrected), respectively, following treatment with 10 and 20 mM DCA. Similarly, Aβ1–42 levels were reduced by 58.93% (45.24% corrected) and 89.29% (80.36% corrected), respectively, at the same drug concentrations.

Fig 2

Growth of untransfected SH-SY5Y cells in the presence of DCA inhibits amyloidogenic APP processing.

Cells were seeded and cultured in the presence of the indicated DCA concentrations and grown until the control cultures reached confluence (medium and DCA were replaced every two days). At this point the growth medium was replaced with UltraMEM containing the same DCA concentrations and the cells were cultured for an additional 24 h. Conditioned medium samples were either used unconcentrated (Aβ-peptide analysis) or concentrated before subjecting equal volumes to SDS-PAGE and immunoblotting (Materials and methods). (A) Conditioned medium samples were immunoblotted with anti-sAPPβ antibody and multiple immunoblots were quantified and results expressed relative to control values without correcting for cell number. (B) The results in (A) were corrected to account for relative changes in cell number using data from the trypan blue assays in Fig 1A. (C) and (D) Aβ-peptide quantification in conditioned medium before (C) and after (D) correcting for changes in cell number. The sAPPβ results are expressed relative to control (0 mM DCA) values whereas the Aβ-peptide results are absolute values (pg/mL). All results are means ± S.D. (n = 3). * = significant at p < 0.05; ** = significant at p < 0.01; *** = significant at p < 0.005; **** = significant at p < 0.001; ***** = significant at p < 0.0005.

The effects of DCA on endogenous APP proteolysis in SH-SY5Y cells are distinct from any effects of the drug on cell number

In the preceding experiments we observed what appeared, given the lack of appreciable non-viable cell numbers, to be a growth inhibitory (as opposed to cytotoxic) effect of DCA. In order to circumvent the need to correct results for changes in cell number, we next grew untransfected SH-SY5Y cells to confluence before replacing the growth medium with UltraMEM and treating them with DCA for just 24 h. Indeed, trypan blue and MTS cell viability assays showed no appreciable effects of the drug on cell number following this treatment (Fig 3A).

Fig 3

Treatment of confluent untransfected SH-SY5Y cells with DCA enhances non-amyloidogenic and inhibits amyloidogenic APP processing.

Cells were grown to confluence before replacing the growth medium with UltraMEM containing the indicated DCA concentrations and culturing for an additional 24 h. Viability assays were then performed or cell lysate and conditioned medium samples were prepared as described in the Materials and methods section. Equal amounts of protein (lysates) or equal volumes (medium) from samples were subjected to SDS-PAGE and immunoblotting (Materials and methods). (A) Trypan blue and MTS cell viability assays. (B) Cell lysates were immunoblotted with anti-APP C-terminal and anti-actin antibodies. Full-length APP (FL-APP) was quantified from multiple immunoblots and the results expressed relative to control values. (C) and (D) Conditioned medium samples were immunoblotted with anti-APP 6E10 antibody in order to detect sAPPα (C) or anti-sAPPβ antibody (D). Multiple immunoblots were then quantified and results expressed relative to control values. (E) Aβ-peptide quantification in conditioned medium; results are absolute values (pg/mL). All results are means ± S.D. (n = 3). * = significant at p < 0.05; ** = significant at p < 0.01; *** = significant at p < 0.005; **** = significant at p < 0.001; ***** = significant at p < 0.0005.

When cell lysates from 24 h DCA-treated cells were immunoblotted using anti-APP C-terminal antibody (Fig 3B) a slight but significant increase in APP expression was observed only in the 10 mM drug-treated cell lysates; equal protein loading between lysate samples was confirmed by re-probing immunoblots with anti-actin antibody (Fig 3B). When the conditioned medium from these cells was immunoblotted with anti-APP 6E10 antibody the results (Fig 3C) revealed increased sAPP695α shedding from 10 mM and 20 mM DCA-treated cells but these changes could not be quantified due to the absence of a detectable signal in the samples from control cells. However, the shedding of sAPP751/770α was quantifiable and showed 5.63 ± 0.38- and 6.23 ± 0.56-fold increases, respectively, in medium from 10 and 20 mM DCA-treated cells relative to controls.

The same conditioned medium samples were then immunoblotted with anti-sAPPβ antibody (Fig 3D) and quantification of the results revealed 79.87 ± 13.17 and 100%, respectively, decreases in the shedding of sAPPβ751/770 from 10 mM and 20 mM DCA-treated cells relative to controls (no signal for sAPPβ695 was detected in these samples). When unconcentrated conditioned medium was analysed in terms of Aβ-peptide content, the results (Fig 3E) showed that Aβ1–40 was reduced by 33.71% and 60.79%, respectively, in medium from 10 mM and 20 mM DCA-treated cells relative to controls. Similarly, Aβ1–42 was reduced by 24.87% and 52.13% at the same drug concentrations.

Collectively, these data show that DCA is able to enhance the non-amyloidogenic and inhibit the amyloidogenic processing of endogenous APP in untransfected SH-SY5Y cells. Of note, we also tested the effect of DCA on endogenous APP processing in two additional unrelated cell lines (SW480 colon cancer and HEK293 cells) and demonstrated similar enhancements in the non-amyloidogenic processing of endogenous APP (although neither of these cell lines generate quantifiable endogenous levels of amyloidogenic processing proteolytic fragments) (S1 Fig).

Stable over-expression of APP in SH-SY5Y cells partially ablates the effects of DCA on both non-amyloidogenic and amyloidogenic proteolysis

The results in the preceding section suggested that DCA might impair amyloidogenic APP processing through a simple reciprocal enhancement of non-amyloidogenic processing leaving less substrate available for cleavage by BACE1. We, therefore, hypothesized that saturating the ADAM-mediated shedding of APP by over-expressing the latter protein should, at least in part, ablate the effects of DCA. Consequently, we repeated the 24 h DCA treatment experiments using SH-SY5Y-APP695 stable transfectants which dramatically over-express (approximately 50-fold) this smaller isoform of the protein [15]. No appreciable effects of DCA on cell number were observed over the treatment period (Fig 4A). Similarly, immunoblotting of cell lysates using anti-APP C-terminal antibody revealed no significant changes in the levels of FL-APP following DCA treatment (Fig 4B); equal protein loading between lysate samples was confirmed by re-probing immunoblots with anti-actin antibody (Fig 4B).

Fig 4

Stable over-expression of APP in SH-SY5Y cells partially ablates the effects of DCA on non-amyloidogenic and amyloidogenic APP processing.

Cells were grown to confluence before replacing the growth medium with UltraMEM containing the indicated DCA concentrations and culturing for an additional 24 h. Viability assays were then performed or cell lysate and conditioned medium samples were prepared as described in the Materials and methods section. Equal amounts of protein (lysates) or equal volumes (medium) from samples were subjected to SDS-PAGE and immunoblotting (Materials and methods). (A) Trypan blue and MTS cell viability assays. (B) Cell lysates were immunoblotted with anti-APP C-terminal and anti-actin antibodies. Full-length APP (FL-APP) was quantified from multiple immunoblots and the results expressed relative to control values. (C) and (D) Conditioned medium samples were immunoblotted with anti-APP 6E10 antibody in order to detect sAPPα (C) or anti-sAPPβ antibody (D). Multiple immunoblots were then quantified and results expressed relative to control values. (E) Aβ-peptide quantification in conditioned medium; results are absolute values (pg/mL). All results are means ± S.D. (n = 3). * = significant at p < 0.05; ** = significant at p < 0.01; **** = significant at p < 0.001.

When conditioned medium samples from the same experiments were immunoblotted using the anti-APP 6E10 antibody (Fig 4C), levels of sAPP695α shed from 10 mM and 20 mM DCA-treated cells were shown to be enhanced by only 1.83 ± 0.16- and 2.26 ± 0.34-fold, respectively, relative to controls (compare this with the 5.63 ± 0.38- and 6.23 ± 0.56-fold increases observed for endogenous sAPP751/770α using untransfected SH-SY5Y cells; Fig 3C). When the same samples were immunoblotted with the anti-sAPPβ antibody it was clear that DCA had no significant effect on the generation of this fragment by SH-SY5Y-APP695 cells (Fig 4D). Similarly, no significant effect of 10 mM DCA on Aβ-peptide levels in conditioned medium could be detected and, at a 20 mM concentration of the drug, Aβ1–40 and Aβ1–42 levels were reduced by 32.97% and 25.01%, respectively (Fig 4E) (compare this with the equivalent 60.79% and 52.13% reductions observed in untransfected cells; Fig 3E).

Collectively, these data indicate that saturating the non-amyloidogenic pathway of APP proteolysis reduces the impact that DCA has on the proportion of substrate processed through this route. Concomitantly, the amount of APP substrate available for cleavage by BACE1 would not be as greatly affected by DCA as would be the case in an unsaturated system (i.e. endogenous levels of APP).

Stable over-expression of BACE1 in SH-SY5Y cells partially ablates the effect of DCA on amyloidogenic but not non-amyloidogenic proteolysis

Assuming that DCA simply shifts competition for the APP substrate in favour of α-secretase activity, over-expression of BACE 1 might be expected to reverse this competition and, therefore, negate the effects of the drug on Aβ-peptide generation. Consequently, we repeated the 24 h DCA treatment experiments using SH-SY5Y-BACE1 stable transfectants which dramatically over-express (approximately 50-fold) the enzyme [15]. Cell viability assays demonstrated no change in cell number following a 24 h treatment with 10 mM DCA (Fig 5A). At 20 mM DCA, whereas the trypan blue assay showed no change in cell number, the MTS assay revealed a significant 29.82 ± 1.71% decrease in viability.

Fig 5

Stable over-expression of BACE1 in SH-SY5Y cells partially ablates the effect of DCA on amyloidogenic but not non-amyloidogenic APP processing.

Cells were grown to confluence before replacing the growth medium with UltraMEM containing the indicated DCA concentrations and culturing for an additional 24 h. Viability assays were then performed or cell lysate and conditioned medium samples were prepared as described in the Materials and methods section. Equal amounts of protein (lysates) or equal volumes (medium) from samples were subjected to SDS-PAGE and immunoblotting (Materials and methods). (A) Trypan blue and MTS cell viability assays. (B) Cell lysates were immunoblotted with BACE1 antibody and re-probed with anti-actin antibody. Multiple anti-BACE1 immunoblots were quantified and the results expressed relative to control values. (C) Cell lysates were immunoblotted with anti-APP C-terminal antibody and re-probed with anti-actin antibody. Full-length APP (FL-APP) was quantified from multiple immunoblots and the results expressed relative to control values. (D) and (E) Conditioned medium samples were immunoblotted with anti-APP 6E10 antibody in order to detect sAPPα (D) or anti-sAPPβ antibody (E). Multiple immunoblots were then quantified and results expressed relative to control values. (F) Aβ-peptide quantification in conditioned medium; results are absolute values (pg/mL). All results are means ± S.D. (n = 3). * = significant at p < 0.05; ** = significant at p < 0.01; *** = significant at p < 0.005.

Next, lysates from the SH-SY5Y-BACE1 cells were immunoblotted with anti-BACE1 antibody and multiple immunoblots were quantified. The results (Fig 5B) showed an intriguing dose-related increase in BACE1 expression (1.17 ± 0.13- and 1.62 ± 0.23-fold, respectively, at 10 and 20 mM DCA). Immunoblots were re-probed with anti-actin to confirm equal protein loading (Fig 5B). In contrast, no change in levels of endogenous FL-APP were detected when the same samples were immunoblotted with anti-APP C-terminal antibody (Fig 5C); again equal lysate protein loading was confirmed by immunoblotting with anti-actin antibody (Fig 5C). When the conditioned medium from these cells was immunoblotted with anti-APP 6E10 antibody the results (Fig 5D) revealed increased sAPP695α shedding from DCA-treated cells (5.57 ± 0.22- and 4.46 ± 0.67-fold, respectively, at 10 and 20 mM). Similarly, the shedding of sAPP751/770α was enhanced 5.32 ± 0.28- and 5.16 ± 0.34-fold. However, when the conditioned medium was immunoblotted with the anti-sAPPβ antibody, no DCA-induced changes in secretion of this fragment could be detected (Fig 5E). Finally, quantification of Aβ-peptides in unconcentrated conditioned medium samples showed no changes in medium from 10 mM DCA-treated cells relative to controls and, at the 20 mM drug concentration, Aβ1–40 and Aβ1–42 levels (Fig 5F) were reduced by 33.44% and 31.11%, respectively (compare this with the equivalent 60.79% and 52.13% reductions observed in untransfected cells; Fig 3E).

Collectively, these data show that increasing BACE1 levels in cells can partly ablate the effect of DCA on amyloidogenic APP processing whilst only having a very minor impact on the ability of the drug to enhance non-amyloidogenic proteolysis.

DCA does not alter secretase expression or activity

As DCA appeared to promote non-amyloidogenic and inhibit amyloidogenic endogenous APP processing, we hypothesized that the drug might simply alter the expression or activity of one or more secretases. To this end, untransfected SH-SY5Y cells were grown to confluence, washed in situ with UltraMEM and then incubated in the same medium with or without DCA for a further 24 h. Following harvesting, cell pellets were split and used for protein and RNA extraction (Materials and methods). Immunoblotting of cell lysates using anti-ADAM10 antibody revealed no change in the expression or processing of this physiological α-secretases (Fig 6A). Furthermore, real-time qPCR performed on the RNA extracted from the same cell pellets showed no significant differences in the relative normalized expression of either ADAM10, BACE1, or presenilin-1 following DCA treatment (Fig 6B). We also conducted parallel experiments in which the resultant cell pellets were lysed and subjected to commercial fluorimetric assays for ADAM10 and BACE1 according to the manufacturer’s instructions (see Materials and methods). However, the results (Fig 6C) also showed no significant effect of DCA on the catalytic activity of these two key secretases.

Fig 6

DCA does not alter secretase expression/activity in untransfected SH-SY5Y cells.

Cells were grown to confluence before replacing the growth medium with UltraMEM containing the indicated DCA concentrations and culturing for an additional 24 h. Harvested cell pellets were split and used for protein and RNA extraction (Materials and methods). Parallel experiments were performed in which the resultant cell pellets were lysed and subjected to commercial fluorimetric assays for ADAM10 and BACE1 activity (see Materials and methods). (A) Equal amounts of protein from cell lysates were immunoblotted for ADAM10 and the blots were reprobed for actin. The positions of prodomain containing (pADAM) and mature, prodomain lacking (mADAM) forms of the enzyme are indicated. (B) Real-time qPCR was performed according to the Materials and methods section. Results are relative values normalized to RPLO and are means ± S.D. (n = 3). (C) ADAM10 and BACE1 activities in cell lysates. Results presented are non-substrate limited end-point assays determined following subtraction of ADAM10 and BACE1 inhibitor-treated control assay values (inhibitors were supplied as part of the commercial kits) and are corrected for protein levels before being expressed relative to control lysates prepared from cells incubated in the absence of DCA. Results are means ± S.D. (n = 6).

The enhancement of ADAM-mediated shedding by DCA is not unique to APP

Our data show that endogenous APP shedding from SW480 cells was enhanced by DCA (S1 Fig). As these cells endogenously express the Notch ligand, Jagged1, which is also known to be shed by a member(s) of the ADAM family [13, 14], we used them to investigate whether or not the effect of DCA on protein shedding was unique to APP. SW480 cells were grown to confluence, washed in situ with UltraMEM, and then incubated for a further 24 h in the absence or presence of DCA. This length of incubation caused no decrease in cell viability (S1 Fig). When cell lysates from these incubations were prepared and immunoblotted with anti-Jagged1 C-terminal antibody, the results showed no significant changes in the expression level of full-length Jagged1 (Fig 7A). However, levels of the Jagged1 CTF generated by ADAM cleavage were enhanced 1.49 ± 0.26- and 1.98 ± 0.48-fold, respectively, in 10 and 20 mM DCA-treated cells compared to controls (Fig 7A); immunoblotting with anti-actin antibody confirmed equal protein loading (Fig 7B). Similarly, immunoblotting of conditioned medium from the same cultures using anti-Jagged1 N-terminal antibody (Fig 7B) showed that shedding of the Jagged1 N-terminal fragment (NTF) from cells was enhanced 2.21 ± 0.13- and 2.45 ± 0.27-fold, respectively, in cultures treated with 10 and 20 mM DCA.

Fig 7

DCA promotes the ADAM-mediated proteolysis of endogenous Jagged1 in SW480 cells.

Cells were grown to confluence before replacing the growth medium with UltraMEM containing the indicated DCA concentrations and culturing for an additional 24 h. Cell lysate and conditioned medium samples were prepared as described in the Materials and methods section. Equal amounts of protein (lysates) or equal volumes (medium) from samples were subjected to SDS-PAGE and immunoblotting (Materials and methods). (A) Cell lysates were immunoblotted with ant-Jagged1 C-terminal antibody, multiple immunoblots were the quantified in terms of both FL-Jagged1 and Jagged1 CTFs and results were expressed relative to control values. Equal protein loading was confirmed by re-probing with anti-actin antibody. (B) Conditioned medium samples were immunoblotted with anti-Jagged1 N-terminal antibody in order to detect the soluble Jagged1 NTF, multiples immunoblots were then quantified and results were expressed relative to control values. All results are means ± S.D. (n = 3). * = significant at p < 0.05; ** = significant at p < 0.01.

Discussion

In the current study, we report that the orphan drug dichloroacetate can simultaneously enhance non-amyloidogenic and inhibit amyloidogenic proteolysis of endogenous APP. One caveat is that plasma DCA concentrations in patients treated with the drug have previously been reported as being approximately ten-fold lower than the lowest concentration employed here [18]. However, it should be noted that, in vitro, higher concentrations of DCA have been shown to elicit a range of cellular effects not related to PDK inhibition e.g. effects on mitochondrial polarisation [19] and mitophagy [20]. As such, we wanted to employ drug concentrations likely to encompass a range of mechanisms and, therefore, elected to adopt higher concentrations more commonly used in studies of an in vitro nature [21-24]. The future identification of these mechanisms with respect to the effects of DCA on APP proteolysis may lead to the development of novel therapeutics for the treatment of AD.

Initially, when untransfected SH-SY5Y cells were cultured in the presence of DCA we observed a decrease in the number of viable cells (Fig 1A) in agreement with a previous study [20]. This study used DCA concentrations up to 60 mM and the authors observed cells in suspension following drug treatment suggesting cell death. In the current study, we did not observe appreciable numbers of cells in suspension but our highest DCA concentration was only 20 mM. Furthermore, the trypan blue assay did not demonstrate appreciable numbers of non-viable cells following DCA treatment. Additionally, when cells were grown to confluence and then treated for 24 h with the drug, no differences in the number of viable cells were detected over the duration of the treatment (Fig 3A). Therefore, our data indicate that, at least at the concentrations employed in the current study, DCA was growth inhibitory rather than cytotoxic. These aspects aside, the effects of the drug on APP proteolysis were clearly distinct from changes in cell number as the former but not the latter was altered in the 24 h incubation experiments. However, DCA is clearly cytotoxic to cells derived from a range of cancers [11] and, given the neuroblastoma origin of SH-SY5Y cells, it would not be at all surprising if the drug was cytotoxic to these cells at higher concentrations. Whether the drug is cytotoxic or growth inhibitory in primary neurons remains to be tested although the fact that chronic DCA administration, whilst being associated with peripheral neuropathy, has not been shown to exert any negative cognitive effects in patients treated for mitochondrial disease, would suggest that toxicity to neurons in the brain would not be problematic despite the compound freely traversing the blood-brain-barrier [25, 26]. In fact, chronic oral administration of DCA in a mouse model of amyotrophic lateral sclerosis has been shown to improve survival and motor performance [27].

The non-amyloidogenic processing of endogenous APP was enhanced in several unrelated cell lines (SH-SY5Y, HEK293 and SW480) demonstrating that the effect of DCA in this respect was not specific to a single cell type. Furthermore, the process was enhanced in untransfected SH-SY5Y cells regardless of whether the cells were grown in the presence of DCA (Fig 1C–1E) or merely incubated with the drug for 24 h once they reached confluence (Fig 3C). In the latter experiments endogenous APP shedding was enhanced 5.63 ± 0.38- and 6.23 ± 0.56-fold, respectively, in medium from cells treated with 10 and 20 mM DCA compared to only 1.83 ± 0.16- and 2.26 ± 0.34-fold in APP over-expressing SH-SY5Y cells (Fig 4C). As the ADAM-mediated shedding of APP in the latter cell line is already massively enhanced due to increased expression of FL-APP [15], the ability of DCA to further stimulate shedding might well have been limited by the substrate saturation of ADAM activity. The fact that the ability of DCA to promote non-amyloidogenic APP processing is reduced following over-expression of the protein also questions whether AD drug action is actually best studied in animal models over-expressing APP. In this respect drugs which might effectively be used to regulate the proteolysis of APP expressed at levels more physiologically relevant to the human brain might well be unintentionally overlooked if studied in such animal models. Notably, DCA also enhanced the non-amyloidogenic shedding of endogenous APP from BACE1 over-expressing cells to a similar level as it did the endogenous protein in the untransfected cells (Fig 5D). Of course, it might be argued that DCA is simply enhancing the trafficking/secretion of sAPPα. Unfortunately, the detection of this fragment in cell lysates using anti-APP antibody 6E10 is confounded by the presence of the epitope in full-length APP. However, we have also demonstrated that DCA enhances the level of α-secretase-derived C83 in cell lysates (Fig 1C) thereby proving unequivocally that non-amyloidogenic APP processing is indeed enhanced by the drug.

In terms of amyloidogenic APP processing, both sAPPβ and Aβ-peptide generation were impaired in untransfected SH-SY5Y cells treated with DCA (Figs 2 and 3) indicating that decreased β-secretase cleavage of APP was the initial event in this respect (as opposed to decreased γ-secretase cleavage of APP βCTF). This hypothesis is further supported by the fact that over-expression of BACE1 in SH-SY5Y cells ablated the effect of DCA on sAPPβ generation and, subsequently, partly ablated the effect of the drug on Aβ-peptide generation (Fig 5E and 5F). Notably this restoration of amyloidogenic processing only caused a slight reciprocal reduction of non-amyloidogenic APP processing (compare Fig 5D to Fig 3C) probably because, even in the presence of over-expressed BACE1, only a minor proportion of the total APP pool is processed via the former pathway.

The complete mechanisms underlying the effects of DCA on APP proteolysis remain to be elucidated. However, from our results it is apparent that the drug does not simply alter the relative expression levels of amyloidogenic and non-amyloidogenic secretases nor does it impact on their cognate catalytic activities (Fig 6). However, the fact that the shedding of another secretase substrate, Jagged1, is also enhanced by DCA (Fig 7) does implicate the enhanced ADAM-mediated processing of these substrates through an, as yet, undetermined mechanism such as enhanced co-localisation of substrates and enzyme (e.g. enhanced enzyme/substrate trafficking to the cell surface or impaired re-internalisation/degradation). Note that it is currently unclear as to whether DCA enhances notch signalling as enhanced shedding of a notch ligand (Jagged1) does not necessarily translate into enhanced pathway signalling. Canonical notch signalling requires a mechanical force exerted between a receptor and ligand both attached to the surface of cells [28] and, therefore, one might, if anything, expect notch signalling to be impaired by DCA.

As to how an inhibitor of pyruvate dehydrogenase kinase might regulate any of these processes (if indeed such a mode of action is of any relevance at all in the current context) is also unclear although it is apparent that there is a connection between glucose metabolism and Alzheimer’s disease. Decreased brain glucose metabolism is a well-documented event in AD as too is the accumulation of glucose and other sugars in the AD brain [29-31]. Mitochondrial dysfunction has also been documented in clinical and experimental AD studies [32] and, recently, decreased glucose oxidation in BACE1 over-expressing SH-SY5Y cells has been attributed to the impairment of tricarboxylic acid (TCA) cycle enzymes including the PDC [33]. As such, it is apparent that there is a link between increased BACE1 activity and decreased mitochondrial energy generation that can be overcome by treatment of cells with compounds capable of stimulating the TCA cycle. Whether a reciprocal mechanistic relationship between DCA-enhanced TCA activity and elevated ADAM function exists remains to be determined.

Conclusions

In conclusion, our data show that DCA, a small molecule PDK inhibitor which traverses the blood-brain-barrier, is capable of enhancing non-amyloidogenic endogenous APP processing at a reciprocal cost to the amyloidogenic pathway and Aβ-peptide generation. These data, therefore, justify further research into the molecular mechanisms involved in these events which may lead to the development of therapies for slowing the progression of Alzheimer’s disease.

DCA enhances amyloidogenic processing of endogenous APP in HEK293 (A-C) and SW480 (D-F) cells.

Cells were grown to confluence before replacing the growth medium with UltraMEM containing the indicated DCA concentrations and culturing for an additional 24 h. Viability assays were then performed or cell lysate and conditioned medium samples were prepared as described in the Materials and methods section. Equal amounts of protein (lysates) or equal volumes (medium) from samples were subjected to SDS-PAGE and immunoblotting (Materials and methods). (A) and (D) Trypan blue and MTS cell viability assays. (B and E) Cell lysates were immunoblotted with anti-APP C-terminal antibody and re-probed with anti-actin antibody. Full-length APP (FL-APP) was quantified from multiple immunoblots and the results expressed relative to control values. (C and F) Conditioned medium samples were immunoblotted with anti-APP 6E10 antibody to detect sAPPα. Multiple immunoblots were quantified and results were expressed relative to control values. All results are means ± S.D. (n = 3). * = significant at p < 0.05; ** = significant at p < 0.01; *** = significant at p < 0.005.

(TIF)

Click here for additional data file.

Original immunoblot images.

Whole blot images are shown from which the lanes marked ‘X’ are excluded from cropped images shown in figures.

(TIFF)

Click here for additional data file.

Minimal data set.

Means and standard deviations for individual figure panels along with the statistical testing method and levels of significance.

(DOCX)

Click here for additional data file.

30 Sep 2021

PONE-D-21-23666The orphan drug dichloroacetate reduces amyloid beta-peptide production whilst promoting non-amyloidogenic proteolysis of the amyloid precursor proteinPLOS ONE

Dear Dr. Parkin,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

In addition to the comments raised by two Reviewers, please address the following in the revised version.

1. Although DCA did not regulate the expression of the secretases ADAM10, ADAM17, BACE1 or presenilin-1, did it increase ADAM10 enzyme activity, without increase in protein amount?,

2. The catalogue numbers of all used antibodies should be listed.

3. 10 and 20 mM final concentrations of DCA were used in this study. Is it not too high concentration?. What concentration of DCA is seen in patients receiving this compound?. Comparisons should be made and discussed.

4. sAPPα levels in Fig. 1 and sAPPβ levels in Fig. 2 as well as in other figures should be shown from lysates also in addition to the medium. Also, why was the actin shown in different panels in all the figures?. Actin should be re-probed from the same blots and should be shown under FL-APP.

5. Does DCA alter sAPPα generation from the mutant APP?. Effect on few mutations such as Swedish or Indiana should be tested and included in the revised version.

6. It is also critical to detect levels of CTFα and CTFβ using more sensitive antibodies or by concentrating the samples and the data included in the revised version.

7. If DCA also enhances notch signaling, the resulting adverse effects if any should be discussed.

Please submit your revised manuscript by Nov 14 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Madepalli K. Lakshmana, Ph.D

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf.

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

3. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data.

4. PLOS ONE now requires that authors provide the original uncropped and unadjusted images underlying all blot or gel results reported in a submission’s figures or Supporting Information files. This policy and the journal’s other requirements for blot/gel reporting and figure preparation are described in detail at https://journals.plos.org/plosone/s/figures#loc-blot-and-gel-reporting-requirements and https://journals.plos.org/plosone/s/figures#loc-preparing-figures-from-image-files. When you submit your revised manuscript, please ensure that your figures adhere fully to these guidelines and provide the original underlying images for all blot or gel data reported in your submission. See the following link for instructions on providing the original image data: https://journals.plos.org/plosone/s/figures#loc-original-images-for-blots-and-gels.

In your cover letter, please note whether your blot/gel image data are in Supporting Information or posted at a public data repository, provide the repository URL if relevant, and provide specific details as to which raw blot/gel images, if any, are not available. Email us at plosone@plos.org if you have any questions.

Additional Editor Comments (if provided):

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In this manuscript, the authors validated the ability of orphan drug dichloroacetate (DCA) to enhance non-amyloidogenic proteolysis of the amyloid precursor protein (APP) in SH-SY5Y neuroblastoma and other cell lines. However, there are some questions/suggestions listed below that would help to clarify this work.

1. Authors tried to see the CTFs on total cell lysates/CMs, but they couldn’t detect it. Why authors did not tired of Immunoprecipitate methods to detect CTFs (PMID: 32514053; PMID: 17463224).

2. There is a slight variation in cell viability of 10 um DCA treatment (Fig. 1A) between Trypan blue and MTS data. Needs clarification regarding how authors carried these two experiments; are these cells from 2 independent experiments.

3. In all the figures of Aβ quantification Aβ1-42 express relatively lower than Aβ1-40 but merging of these bars in a single y-axis scale could not able to see the differences among Aβ1-42.

4. On what basis authors selected 24 h DCA treatment, why they did not do more than 24 h.

5. In the methods sections, the authors mentioned statistical analysis was carried out either by Student’s t-test or by one-way ANOVA. But in the legend section, does not mention clearly in which data they used student’s t-test / one-way ANOVA.

Reviewer #2: This is a very intriguing study that, if verified and extended by in vivo experiments, has significant translational potential. It is disappointing, therefore, that the authors presume to know the mechanism of action of DCA in affecting the reported changes in amyloid beta-peptide production and precursor protein. In my opinion, they do not, and this is the major concern about the submission.

The presumed mechanism of DCA's effects reported here is activation of the pyruvate dehydrogenase complex by inhibition of 1 or more pyruvate dehydrogenase kinase isoforms, but this requires direct testing and validation; otherwise, the findings are largely phenomenological and not sufficiently mechanistically-oriented. Are glucose oxidation and PDC activity suppressed due to up-regulation of a PDK? Can other specific PDK inhibitors exert the same changes in amyloid metabolism as DCA? Would genetic silencing of the E1 alpha subunit render DCA ineffective under these experimental conditions? Confirming DCA's MOA would significantly enhance the probative value of the paper.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

15 Nov 2021

Rebuttal to editorial comments:

1. Although DCA did not regulate the expression of the secretases ADAM10, ADAM17, BACE1 or presenilin-1, did it increase ADAM10 enzyme activity, without increase in protein amount?

This is a very good question and one which we have investigated previously using fluorometric secretase assay kits in relation to both ADAM10 and BACE1 activity. In the interests of scientific openness though, we have to state that we do not subscribe entirely to the accuracy of these kits as they rely on peptides that are subject to a degree of non-specific proteolysis even when the purported ‘specific’ inhibitors are incorporated as controls as we have done (hence the reason for not incorporating these data in the original submission). However, not withstanding this caveat, we have incorporated data into Fig. 6 which show no change in either ADAM10 or BACE1 activity in the presence of DCA and added accompanying text changes in the Materials and methods, results and other necessary sections.

2. The catalogue numbers of all used antibodies should be listed.

We have now listed the relevant catalogue numbers in the ‘Materials’ subsection of the ‘Materials and methods’.

3. 10 and 20 mM final concentrations of DCA were used in this study. Is it not too high concentration?. What concentration of DCA is seen in patients receiving this compound?. Comparisons should be made and discussed.

Plasma concentrations of individuals receiving DCA therapy are approximately 10-fold lower than the 10 mM lower concentration used in the current study and we have not been able to observe any change in APP processing at a DCA concentration of 1 mM in vitro. However, it should be noted that, in vitro, higher concentrations of DCA have been shown to elicit a range of cellular effects not related to pyruvate dehydrogenase kinase (PDK) inhibition e.g. effects on mitochondrial polarisation [1] and mitophagy [2]. As such, we wanted to employ drug concentrations likely to encompass a range of mechanisms and, therefore, elected to adopt higher concentrations more commonly used in in vitro studies such as those previously published in PLOS ONE [3-6] which utilised concentrations of up to 50 mM. However, we do agree that this might limit the direct therapeutic application of DCA for the treatment of Alzheimer’s disease and have, therefore, toned down statements pertaining to this possibility in the abstract, introduction and discussion. Furthermore, we have inserted additional text at the start of the discussion pertaining to the choices made in relation to drug concentrations. We would like to emphasise that, whilst the concentrations of DCA used in the current study exceed those previously observed in plasma, this does not negate the fact that determining the mode of action of DCA with respect to APP proteolysis may lead eventually to the identification of new AD therapies.

4. sAPPα levels in Fig. 1 and sAPPβ levels in Fig. 2 as well as in other figures should be shown from lysates also in addition to the medium. Also, why was the actin shown in different panels in all the figures?. Actin should be re-probed from the same blots and should be shown under FL-APP.

We suspect that the rationale for immunoblotting lysates for sAPP� and sAPPβ may be based on the supposition that DCA may simply be affecting the secretion of these fragments? However, unfortunately, the sAPPβ antibody does not detect any of this fragment at all in lysates. Furthermore, using the anti-APP antibody 6E10 to immunoblot lysates for sAPP� is confounded by the fact that the antibody detects an epitope also present in full-length APP which is far more abundant in lysates than soluble fragments and would, therefore, mask any changes in these latter fragments.

However, we appreciate this concern but believe that the increases in APP-CTF production following DCA treatment demonstrates firmly that the drug enhances alpha-secretase-mediated proteolysis of APP rather than solely stimulating secretion of sAPPalpha (note that we have also now demonstrated that the alphaCTF/C83, specifically, is increased following drug treatment; Fig.1). Additionally, we have added this point to the Discussion of the manuscript.

Actin was reprobed after stripping blots and, therefore, is a separate exposure of the same blot. We feel that it would be a little confusing to place the actin panel immediately under the FL-APP panel as the quantification in the same panel is for APP and not the actin (e.g. Fig 1B – the quantification here correlates directly with the FL-APP blot; if the actin panel were placed between the FL-APP blot and the quantification of FL-APP it would confuse matters). By way of compromise and to make more clear that the FL-APP and actin blots are associated, we have simply combined them as a single lettered panel and altered any relevant descriptions in the results commentary where necessary. Furthermore, we have added in the blot stripping methodology to the Materials and Methods section.

5. Does DCA alter sAPPα generation from the mutant APP?. Effect on few mutations such as Swedish or Indiana should be tested and included in the revised version.

Whilst this would be interesting there are two solid reasons as to why these suggested experiments are beyond the scope of the current study or would not necessarily yield meaningful results:

(i) 95% of AD cases are sporadic and not associated with a known mutation in APP or related secretases. Our study, in examining wt-APP, is representative of this 95% of AD cases. Familial Alzheimer’s disease-associated APP mutations, in this context, are not of direct relevance.

(ii) We have shown in the current study that over-expressing APP largely ablates the observed changes in amyloidogenic/non-amyloidogenic processing that are induced by DCA in the case of endogenously expressed APP. Therefore, generating more cell lines over-expressing FAD-associated APP mutations is likely to yield artefactual results. Indeed we discuss why the use of systems over-expressing APP are not likely to be suitable for the study of AD, generally and in the current context, within the Discussion section of the current document. The only way this could be done would be to generate a series of gene-edited cell lines which is not realistic or hugely relevant in the context of the current submission.

6. It is also critical to detect levels of CTFα and CTFβ using more sensitive antibodies or by concentrating the samples and the data included in the revised version.

See rebuttal to point 4 above. In short, we completely agree with this point as, without these data, there remains the possibility that DCA might simply be altering the secretion of APP fragments. The original CTF blots were derived from the same immunoblots as the FL-APP so are not able to tell us anything other than the fact that APP-CTFs per se increase following DCA treatment. Unfortunately, this work was done some time ago now and we no longer have access to the original samples to rerun them. However, we have now grown up untransfected SH-SY5Y cells and treated afresh with DCA subsequently preparing lysates of a higher protein concentration and resolving on Tris/tricine gels (details added to the Materials and methods section) in order to resolve individual CTF species (Fig.1). Whilst we could still not detect C99 in untransfected cells, we were able to unequivocally determine that the increase in CTFs in DCA-treated cells was specifically the consequence of C83 accumulation. We have removed the APP-CTF data from most of the other figures as, given that these samples were not run on Tris/tricine gels, the data only serves to confuse matters. We believe that our data now demonstrate unequivocally that DCA leads to an increase in alpha-secretase-mediated APP processing and does not simply alter the secretion of soluble APP fragments.

7. If DCA also enhances notch signaling, the resulting adverse effects if any should be discussed.

There is actually no evidence of this (enhanced ligand shedding does not necessarily translate into enhanced notch signalling) but it is an interesting comment. We have incorporated a very brief comment on this in the Discussion of the document.

Rebuttal to reviewer comments:

Reviewer 1:

1. Authors tried to see the CTFs on total cell lysates/CMs, but they couldn’t detect it. Why authors did not tired of Immunoprecipitate methods to detect CTFs (PMID: 32514053; PMID: 17463224).

See response to editorial comment 6.

2. There is a slight variation in cell viability of 10 um DCA treatment (Fig. 1A) between Trypan blue and MTS data. Needs clarification regarding how authors carried these two experiments; are these cells from 2 independent experiments.

Whilst the trypan blue data did show a decrease in viability at 10 mM DCA, ANOVA testing did not reveal a significant difference between the 10 mM-treated cells analysed with trypan blue and those analysed by MTS (these were, indeed, two independent experiments). We have inserted a ‘cover all’ statement at the end of the Materials and methods section pertaining to the cell viability assays which, hopefully, clarifies that these types of experiments were performed independently.

3. In all the figures of Aβ quantification Aβ1-42 express relatively lower than Aβ1-40 but merging of these bars in a single y-axis scale could not able to see the differences among Aβ1-42.

We have altered all of the Aβ quantification graphs such that they have double Y-axes. It is now much easier to see changes in Aβ1-42 levels.

4. On what basis authors selected 24 h DCA treatment, why they did not do more than 24 h.

A 24 h period was chosen as UltraMEM is a low serum medium (facilitating protein analysis by immunoblotting without the distortion of gels by large amounts of albumin derived from foetal bovine serum in complete medium) and, therefore, maintenance of cells in this medium becomes increasingly poor after 24 h. We have incorporated this explanation into the ‘Treatment of cells and protein extraction’ subsection of the Materials and methods section.

5. In the methods sections, the authors mentioned statistical analysis was carried out either by Student’s t-test or by one-way ANOVA. But in the legend section, does not mention clearly in which data they used student’s t-test / one-way ANOVA.

Actually, this is a good point – all the statistical analysis in this paper is ANOVA. We have removed the reference to student’s t-test in the Materials and methods. This was a cut and paste carry over from methods in an earlier manuscript.

Reviewer 2:

This is a very intriguing study that, if verified and extended by in vivo experiments, has significant translational potential. It is disappointing, therefore, that the authors presume to know the mechanism of action of DCA in affecting the reported changes in amyloid beta-peptide production and precursor protein. In my opinion, they do not, and this is the major concern about the submission.

The presumed mechanism of DCA's effects reported here is activation of the pyruvate dehydrogenase complex by inhibition of 1 or more pyruvate dehydrogenase kinase isoforms, but this requires direct testing and validation; otherwise, the findings are largely phenomenological and not sufficiently mechanistically-oriented. Are glucose oxidation and PDC activity suppressed due to up-regulation of a PDK? Can other specific PDK inhibitors exert the same changes in amyloid metabolism as DCA? Would genetic silencing of the E1 alpha subunit render DCA ineffective under these experimental conditions? Confirming DCA's MOA would significantly enhance the probative value of the paper.

It is somewhat disappointing to see that Reviewer 2 answered only ‘Partly’ to the question ‘Is the manuscript technically sound, and do the data support the conclusions?’ which was, presumably, on the basis of the comments above. We cannot see how the Reviewer arrived at the conclusion that ‘the authors presume to know the mechanism of action of DCA’ or that ‘The presumed mechanism of DCA’s effects reported here is activation of the pyruvate dehydrogenase complex by 1 or more pyruvate dehydrogenase kinase isoforms…’. We actually went to considerable efforts in the writing of the manuscript to avoid these presumptions instead trying to cover several possibilities particularly in the discussion. As such, we consider these comments to have arisen from insufficient scrutiny of the manuscript and, therefore argue that, in fact, the manuscript is both technically sound and the data do indeed fully support the conclusions made. To add further to this, we have tested another PDK inhibitor which actually does not exert the same effect as DCA with respect to APP proteolysis even at concentrations known to inhibit PDK. As such, we were aiming (and think we actually achieved this) to leave options open as to the complete mechanism of DCA action whilst making informed suggestions based on our data but without extrapolating our conclusions too far.

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming.

We have consulted the recommended templates and gone through the manuscript in detail checking for erroneous style/formatting and we think we have named submission files correctly but would be happy to make any further necessary changes. Specifically:

(i) We have capitalized proper nouns in the title.

(ii) We have indicated author affiliations with numbers only.

(iii) We have been through the entire text and ensured that symbols are inserted using the ‘Insert symbol’ function in word rather than by changing the font of existing text.

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

The funding body is a philanthropic organisation who donate by cheque payment to research at Lancaster University. There is no specific grant number associated with these payments.

3. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data.

The data are not a core part of the research as the enhanced shedding of Jagged1 is already incorporated in our figures – the comment on prion protein was merely a third example of a protein whose shedding is enhanced by DCA so the comment has been removed.

4. PLOS ONE now requires that authors provide the original uncropped and unadjusted images underlying all blot or gel results reported in a submission’s figures or Supporting Information files. This policy and the journal’s other requirements for blot/gel reporting and figure preparation are described in detail at https://journals.plos.org/plosone/s/figures#loc-blot-and-gel-reporting-requirements and https://journals.plos.org/plosone/s/figures#loc-preparing-figures-from-image-files. When you submit your revised manuscript, please ensure that your figures adhere fully to these guidelines and provide the original underlying images for all blot or gel data reported in your submission. See the following link for instructions on providing the original image data: https://journals.plos.org/plosone/s/figures#loc-original-images-for-blots-and-gels.

In your cover letter, please note whether your blot/gel image data are in Supporting Information or posted at a public data repository, provide the repository URL if relevant, and provide specific details as to which raw blot/gel images, if any, are not available. Email us at plosone@plos.org if you have any questions.

We have provided the images as requested which show the whole blots rather than cropped lanes/rows – we hope this is sufficient as it’s the only record we have of these blots but does show the whole blot area. Note that, previously, we no longer had access to the uncropped blots for ADAM17. As these blots neither showed any change following DCA treatment nor were central to the manuscript (with ADAM10 being the physiological alpha-secretase) we considered a sensible course of action here to simply remove the ADAM17 blot information from the manuscript. We have submitted our image data and minimal data set as Supporting Information.

1. Stockwin, L.H., et al., Sodium dichloroacetate selectively targets cells with defects in the mitochondrial ETC. Int J Cancer, 2010. 127(11): p. 2510-9.

2. Pajuelo-Reguera, D., et al., Dichloroacetate stimulates changes in the mitochondrial network morphology via partial mitophagy in human SH-SY5Y neuroblastoma cells. Int J Oncol, 2015. 46(6): p. 2409-18.

3. Dai, Y., et al., Dichloroacetate enhances adriamycin-induced hepatoma cell toxicity in vitro and in vivo by increasing reactive oxygen species levels. PLoS One, 2014. 9(4): p. e92962.

4. Harting, T.P., et al., Dichloroacetate affects proliferation but not apoptosis in canine mammary cell lines. PLoS One, 2017. 12(6): p. e0178744.

5. Klose, K., et al., Metformin and sodium dichloroacetate effects on proliferation, apoptosis, and metabolic activity tested alone and in combination in a canine prostate and a bladder cancer cell line. PLoS One, 2021. 16(9): p. e0257403.

6. Sun, W., et al., Chronic CSE treatment induces the growth of normal oral keratinocytes via PDK2 upregulation, increased glycolysis and HIF1alpha stabilization. PLoS One, 2011. 6(1): p. e16207.

Submitted filename: Response to Reviewers.doc

Click here for additional data file.

20 Dec 2021

The orphan drug Dichloroacetate reduces Amyloid beta-peptide production whilst promoting non-amyloidogenic proteolysis of the Amyloid Precursor Protein

PONE-D-21-23666R1

Dear Dr. Parkin,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Madepalli K. Lakshmana, Ph.D

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

4 Jan 2022

PONE-D-21-23666R1

The orphan drug Dichloroacetate reduces Amyloid beta-peptide production whilst promoting non-amyloidogenic proteolysis of the Amyloid Precursor Protein

Dear Dr. Parkin:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Madepalli K. Lakshmana

Academic Editor

PLOS ONE