TRAP1 Is Expressed in Human Retinal Pigment Epithelial Cells and Is Required to Maintain their Energetic Status

Age-related macular degeneration (AMD) is the leading cause of severe vision loss and blindness in elderly people worldwide. The damage to the retinal pigment epithelium (RPE) triggered by oxidative stress plays a central role in the onset and progression of AMD and results from the excessive accumulation of reactive oxygen species (ROS) produced mainly by mitochondria. Tumor necrosis factor receptor-associated protein 1 (TRAP1) is a mitochondrial molecular chaperone that contributes to the maintenance of mitochondrial integrity by decreasing the production and accumulation of ROS. The present study aimed to evaluate the presence and the role of TRAP1 in the RPE. Here, we report that TRAP1 is expressed in human adult retinal pigment epithelial cells and is located mainly in the mitochondria. Exposure of RPE cells to hydrogen peroxide decreases the levels of TRAP1. Furthermore, TRAP1 silencing increases intracellular ROS production and decreases mitochondrial respiratory capacity without affecting cell proliferation. Together, these findings offer novel insights into TRAP1 functions in RPE cells, opening possibilities to develop new treatment options for AMD.


Introduction
Age-related macular degeneration (AMD) is the major cause of serious vision loss and blindness in seniors worldwide [1]. Accumulation of drusen in the macula and pigmentary abnormalities are hallmarks of the early stages of AMD [2]. Progression to late AMD is Antioxidants 2023, 12, 381 2 of 17 characterized by the loss of both retinal pigment epithelium (RPE) and photoreceptor cells (dry AMD) or abnormal proliferation of choroidal capillaries (wet AMD) [2,3]. The RPE is a monolayer of polarized and pigmented cells located between the photoreceptor cells and the choroid, and it is an essential component of the outer blood-retinal barrier [4,5]. The RPE plays an important role in the maintenance of overlying photoreceptor cells by performing crucial functions such as transepithelial transport, phagocytosis of photoreceptor outer segments, and protection against oxidative stress [4]. Degeneration and impaired clearance mechanisms of RPE contribute to increased accumulation of lipofuscin, which is a yellowish aggregate of oxidized proteins and lipids, the main constituent of drusen, and one of the hallmarks of AMD. Therefore, RPE has been suggested as a critical site of pathology [6]. Retinal pigment epithelial cells are particularly susceptible to oxidative stress due to exposure to intense focal light, high metabolic activity, unique phagocytic function, large oxygen gradient from the choroid, and accumulation of oxidized lipoproteins with ageing [6]. In the context of AMD, oxidative-stress-induced RPE damage results from the excessive accumulation of reactive oxygen species (ROS), which is produced mainly in the mitochondria under pathological conditions. Mitochondrial function homeostasis is critical for maintaining ROS at physiological levels, avoiding the metabolic dysfunction observed in AMD pathology [7]. Therefore, it is important to study novel proteins that might be involved in oxidative stress mechanisms in the RPE in the context of AMD. Tumor necrosis factor receptor-associated protein 1 (TRAP1), also known as heat shock protein 75 (HSP75), is a mitochondrial molecular chaperone that supports protein folding and contributes to the maintenance of mitochondrial integrity, even under cellular stress [8][9][10]. TRAP1 has attracted increasing interest for its homology to heat shock protein 90 (HSP90), and it has long been pursued as a potential therapeutic target for designing novel anticancer agents [9,[11][12][13]. Previous studies in several cell lines demonstrated that TRAP1 protects against mitophagy, mitochondrial apoptosis, and dysfunction by decreasing the production and accumulation of ROS [8,[14][15][16][17]. It has also been suggested that downregulation of TRAP1 expression impacts cellular function, resulting in lower mitochondrial membrane potential, increased intracellular ROS production and increased cell death [18][19][20]. TRAP1 inhibits mitochondrial oxidative phosphorylation (OXPHOS) via downregulation of cytochrome oxidase in the respiratory chain [9]. Considering the above-mentioned TRAP1 functions described in other cell types [8][9][10][11][12][13][14][15][16][17][18][19][20], we hypothesize that TRAP1 modulates mitochondrial metabolism in the RPE, and that it has a role in avoiding AMD onset and progression. The present study aimed to evaluate the presence and function of TRAP1 in the RPE and whether TRAP1 mediates protection against oxidative stress in RPE cells. Here, we report that TRAP1 is expressed in human adult RPE cells and located mainly in the mitochondria. Furthermore, TRAP1 silencing increased ROS production by human RPE cells, ultimately driving cells into a quiescent metabolic state. Taken together, these results provide novel insights into TRAP1 antioxidant functions in RPE cells and the susceptibility of RPE to oxidative stress.

Polymerase Chain Reaction
ARPE-19 cells were washed twice with ice-cold sterile PBS and total RNA was extracted using the Trizol ® reagent (Life Technologies, Carlsband, CA, USA) according to the manufacturer's protocol. RNA concentration and purity were determined using NanoDrop ® (Thermo Fisher Scientific, Waltham, MA, USA). One µg of total RNA was reverse-transcribed into cDNA using the NZY First-Strand cDNA Synthesis Kit (NZYTech, Lisbon, Portugal) according to the manufacturer's instructions. The cDNA was then treated with RNase-H for 20 min at 37 • C and stored at −20 • C until further analysis. Amplification of the cDNA was performed using a BioRad T100 Thermal Cycler (Bio-Rad, Hercules, CA, USA) in a total volume of 40 µL containing MYtaq Red mix (NZYTech, Lisbon, Portugal) and 0.25 mM of specific primers for TRAP1: HA29_TRAP1_FW 5 -GCTCTGGGAGTACGACATGG-3 and HA30_TRAP1_RV 5 -CTCCGAGGACACAGAAT TGGT-3 (129 bp amplicon). Cycling conditions were as follows: melting step at 95 • C for 1 min, annealing at 62 • C for 30 s and elongation at 72 • C for 30 s, for 35 cycles. Total RNA from iRPE, donor RPE and RPE spheroids (DD114) was harvested using the RNeasy kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. DNase I (Invitrogen, Waltham, USA) treatment was conducted to avoid genomic DNA contamination. The reverse transcription reaction was performed using the SuperScript™ III First-Strand Synthesis SuperMix (Thermo Fisher Scientific, Waltham, MA, USA) to generate cDNA. PCR was performed using Phire Hot Start II DNA Polymerase (Thermo Fisher Scientific, Waltham, USA) to determine gene expression. The PCR program was 25 • C for 10 min, 50 • C for 30 min, 85 • C for 5 min and 4 • C for 5 min. The primers used for PCR were HA29_TRAP1_FW 5 -GCTCTGGGAGTACGACATGG-3 and HA30_TRAP1_RV 5 -CTCCGAGGACACAGAATTGGT-3 . PCR product was separated using a 2% (w/v) agarose gel.

Western Blotting
ARPE-19 protein extracts were prepared in ice-cold radioimmunoprecipitation assay (RIPA) buffer with 1 mM of dithiothreitol (Sigma-Aldrich, St. Louis, MO, USA) and a protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO, USA). The protein concentration was measured by Bradford's assay (Invitrogen, Waltham, MA, USA). Forty µg of total protein lysates were resolved on a 12% SDS-PAGE gel and transferred onto a PVDF membrane. The membranes were then blocked in 5% skim milk for 1 h and incubated for 16 h at 4 • C with primary anti-TRAP1 (1:1000; BD Bioscience, Franklin Lakes, NJ, USA) and anti-β-actin (1:5000; Sigma-Aldrich, St. Louis, MO, USA). After washing 3 times with Tris Buffered Saline with 0.1% Tween 20 (TBST) for 10 min, the membranes were incubated for 1 h at room temperature with the respective secondary antibody (anti-mouse, 1:10,000; Invitrogen, Waltham, MA, USA) and washed 3 times with TBST for 10 min each. After that, membranes were imaged in ImageQuant LAS 500 (GE Healthcare Bio-Sciences, Chicago, IL, USA) using WesternBright Sirius™ (Advansta, San Jose, CA, USA). The protein band densitometry was quantified using the software Quantity One (Bio-Rad, Hercules, CA, USA), using β-actin as a loading control.

TRAP1 Silencing in ARPE-19 Cells
To silence TRAP1 expression, we used a siRNA anti-Human TRAP1 (siTRAP1, NM_016292, SI00115150 A60-SIRNA, Qiagen, Hilden, Germany) [30]. As a negative control, a scramble siRNA (siCTL, Unspecific_AllStars_1 A60-SIRNA SI03650318; Qiagen, Hilden, Germany) was used. On day 0, ARPE-19 cells were seeded in DMEM/F-12 with Glu-taMAX™ supplemented with 1% FBS and antibiotics (100 U/mL penicillin, 100 µg/mL streptomycin) at a density of 10,000, 60,000 or 115,000 cells/well in a 96-, 24-or 6-wellplate, respectively, to be approximately 80% confluent at the day of transfection. On day 1, 4 h before the transfection, the medium was replaced by Opti-MEM medium (Gibco, Waltham, MA, USA). Afterwards, lipofectamine RNAiMAX transfection reagent (Invitrogen, Waltham, MA, USA) was diluted in Opti-MEM medium. The master mix containing 0.1 µM of siCTL or siTRAP1 diluted in Opti-MEM medium was added to the diluted RNAiMAX in a 1:1 ratio and incubated for 5 min at room temperature. Thereafter, 10, 50 or 250 µL of the previous solution, depending on the well size, was added dropwise to the cells. Cells were analyzed at 24 or 72 h after transfection. In the cells cultured for 72 h, the culture medium was replaced by DMEM/F-12 with GlutaMAX™ supplemented with 1% FBS and antibiotics 48 h after transfection.

Resazurin Assay
ARPE-19 cells were seeded (10,000 cells/well) in flat-bottom 96-well microplates in DMEM/F-12 with GlutaMAX™ supplemented with 1% FBS and antibiotics (100 U/mL penicillin, 100 g/mL streptomycin), at a density of 10,000 cells/well in flat-bottom 96-well microplates. ARPE-19 cells were then incubated with 31.25, 62.5, 125, 500, 1000 or 2000 µM of H 2 O 2 (PanReac AppliChem, Ottoweg, Germany) for 24 h. In the experiments to assess the impact of TRAP1 loss in ARPE-19, the cells were transfected twenty-four hours after seeding, as previously described. Cell viability/metabolism was assessed using Resazurin (Invitrogen, Waltham, MA, USA) according to the manufacturer's instructions. Briefly, the medium was replaced with DMEM/F-12 with GlutaMAX™ containing antibiotics and resazurin (1 mg/mL) and incubated at 37 • C for 2 h. The fluorescence was measured at Ex/Em = 550/590 nm using Synergy Multi-Mode Reader (BioTek, Winooski, VT, USA). Cells were then washed twice with PBS and fixed with 4% PFA in PBS for 10 min. Thereafter, nuclei were counterstained with DAPI (1/5000; Invitrogen, Waltham, MA, USA) and counted using an automated high-content fluorescence microscope (INCell Analyzer 2000, GE Healthcare, Chicago, IL, USA). The values are presented as a percentage of the control (non-treated cells).

Sulforhodamine B Assay
Seventy-two hours after transfection, the culture medium was removed, and cells were washed with 1% phosphate-buffered saline (PBS). The cells were then fixed with ice-cold 1% acetic acid in 100% methanol for 16 h at −20 • C. The microplates containing fixed cells were dried at 37 • C. Thereafter, cells were incubated with 0.5% Sulforhodamine B (SRB) (Sigma-Aldrich, St. Louis, MO, USA) in 1% acetic acid for 30 min at 37 • C and subsequently washed three times with 1% acetic acid. Plates were then allowed to dry before adding 10 mM Tris (pH 10), followed by incubation at room temperature for 15 min with agitation. Lastly, 200 µL of the solubilized solution was added to each well and optical density was measured at 540 nm.

Dihydroethidium Probe
Superoxide anion was detected using the Dihydroethidium (DHE) probe. ARPE-19 cells were seeded in DMEM/F-12 with GlutaMAX™ supplemented with 1% FBS and antibiotics (100 U/mL penicillin, 100 µg/mL streptomycin) at a density of 60,000 cells/well in 12 mm glass coverslips placed on a 24-well plate. TRAP1 silencing was performed as previously described. Twenty-four or seventy-two hours after transfection, cells were incubated with 10 µM of Dihydroethidium (Invitrogen, Waltham, USA) diluted in PBS with 5% FBS for 30 minutes at 37 • C in a light-protected environment. Thereafter, cells were washed twice with PBS, fixed with 4% PFA with 4% sucrose for 10 min and permeabilized with PBS containing 1% Triton X-100 for 5 min. Cells were then counterstained with DAPI (1:5000; Invitrogen, Waltham, USA) diluted in PBS containing 10% normal goat serum, 3% bovine serum albumin and 0.3% Triton X-100. Samples were mounted with Glycergel mounting medium (Dako Cytomation, Glostrup, Denmark) and imaged using a confocal microscope (Zeiss LSM510, Carl Zeiss, Oberkochen, Germany). Mean fluorescence intensity (MFI) and the number of DAPI-positive nuclei were automatically calculated/counted using ImageJ v1.53C, using at least 4 images for each condition/experiment. The values are presented as MFI per cell.

Transmission Electron Microscopy
Seventy-two hours after transfection with siCTL or siTRAP1, ARPE-19 cells were fixed with 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.2) supplemented with 1 mM calcium chloride. Sequential post-fixation was performed using 1% osmium tetroxide (Sigma-Aldrich, St. Louis, MO, USA) for 1 h and contrast-enhanced with 1% aqueous uranyl acetate for 90 min. After rinsing in distilled water, samples were dehydrated in a graded ethanol series (50-100%), impregnated, and embedded using an Epoxy embedding kit (Fluka Analytical, Waltham, MA, USA). Ultrathin sections (70 nm) were mounted on copper grids (300 mesh) and stained with 0.2% lead citrate for 7 min. Observations were carried out using an FEI-Tecnai G2 Spirit Bio Twin at 100 kV.

Mitochondrial Network
Mitochondria were labelled with Image-iT™ TMRM Reagent (Invitrogen, Waltham, MA, USA). Briefly, 72 h after transfection, ARPE-19 cells were incubated with 75 nM TMRM Reagent diluted in Hank's Balanced Salt Solution (HBSS) containing Ca 2+ and Mg 2+ for 30 min at 37 • C. Thereafter, the TMRM Reagent solution was removed and replaced by Opti-MEM medium. Cell images were acquired using a Zeiss LSM710 confocal microscope (Carl Zeiss, Oberkochen, Germany) equipped with an environmental chamber. Analysis of the images was performed using ImageJ v1.53K (National Institute of Health, Bethesda, MD, USA) software. Parameters were calculated using the Mito-Morphology ImageJ macro [31]. The average area/perimeter ratio was employed as an index of mitochondrial interconnectivity and inverse circularity used to assess mitochondrial elongation [31].

Oxygen Consumption and Extracellular Acidification Rate
The oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) of ARPE-19 cells were determined using Seahorse XFe96 Extracellular Flux Analyzer Technology (Agilent, Santa Clara, CA, USA). Cells (10,000 per well) were seeded in XF96-well plates (Agilent, Santa Clara, CA, USA). Cells were transfected 24 h later with siCTL or siTRAP1 and cultured for another 48 h. The cell culture medium was then replaced by 175 µL/well of pre-warmed low-buffered serum-free minimal DMEM medium (D5030; Sigma-Aldrich, St. Louis, USA) supplemented with glucose (17.5 mM), L-glutamine (2 mM) and sodium pyruvate (0.5 mM). The pH was adjusted to 7.4 and the cells incubated at 37 • C in a non-CO 2 incubator for 1 h to allow the temperature and pH of the medium to reach equilibrium before the first-rate measurement. Oligomycin, FCCP, rotenone, and antimycin A solutions were prepared in DMSO, diluted in low-buffered serum-free DMEM medium, and then pre-loaded (25 µL) into the ports of each well in the XFe96 sensor. The sensor cartridge and the calibration plate were loaded into the XFe96 Extracellular Flux Analyzer for calibration. Then, 2 µM oligomycin was injected into reagent delivery port A, 1 µM FCPP was injected into port B, and 1 µM rotenone plus 1µM antimycin A were injected into reagent delivery port C and mixed. For oxygen consumption rate (OCR) measurements, three baseline rate measurements of OCR and extracellular acidification rate (ECAR) of the cells were made using a 3 min mix and a 5 min measure cycle. The total amount of protein was determined using the Sulforhodamine B (SRB) assay, and each OCR's raw data was normalized to the total protein content in each well. Results were analyzed by using the Software Version Wave Desktop 2.6 (Agilent, Santa Clara, CA, USA).

Statistical Analysis
The normality of the distribution was tested by the Kolmogorov-Smirnov test. Statistical significance was calculated by using an unpaired t-test or by using the Mann-Whitney U test if the data did not follow a normal distribution. All statistical analyses were performed using GraphPad Prism, version 9.0 (GraphPad Software, Boston, MA, USA). All values are expressed as mean ± standard error of the mean (SEM). For all tests, p values < 0.05 were considered statistically significant. At least three independent experiments were used for analysis (n = 3), unless otherwise stated.

TRAP1 Localizes in the Mitochondria of Human RPE Cells
The expression of TRAP1 mRNA in ARPE-19 cells was confirmed by PCR analysis using specific primers ( Figure 1A), while TRAP1 protein was also detected in these cells ( Figure 1B-D). Furthermore, we showed that TRAP1 protein is located in the mitochondrial compartment since its expression co-localized with MitoTracker ( Figure 1C), a dye that accumulates specifically in mitochondria [32], and with SAM50 ( Figure 1D), a mitochondrial protein. We further confirmed the presence of TRAP1 mRNA in the human RPE cells using samples from human induced pluripotent stem cells (hiPSCs)-derived RPE (iRPE), RPE isolated from hiPSC-derived retinal organoids (RPE spheroids) and human donor RPE ( Figure 1E). Similar to the results obtained with ARPE-19 cells, TRAP1 protein was also detected at the basal side in RPE spheroids ( Figure 1F,G), where it co-localizes with TOM20, a mitochondrial marker ( Figure 1G). This result is in accordance with previous electron microscopy studies on RPE spheroids in which mitochondria were localized at the basal side [26].

Impact of Oxidative Stress on TRAP1 Levels
Exposure to H 2 O 2 is commonly used as a model to assess the oxidative stress susceptibility and antioxidant activity of RPE cells [6,7,33]. To define the optimal dose and incubation time, we exposed ARPE-19 cells to different concentrations of H 2 O 2 for 2, 4 and 24 h and assessed the cell viability using the resazurin assay (Figure 2A and Figure S1). As expected, H 2 O 2 decreased the viability of ARPE-19 cells in a dose-dependent manner (Figure 2A). For subsequent experiments, 500 µM H 2 O 2 (24 h incubation) was used as these conditions decreased the cell viability of ARPE-19 cells by 53.5% ± 6.6% compared with the control (non-treated cells; p = 0.0286). To assess if the levels and localization of TRAP1 in ARPE-19 cells changed upon oxidative stress, cells were incubated with 500 µM of H 2 O 2 for 24 h (Figure 2). Our results showed that the localization of TRAP1 was similar in both non-treated ( Figure 2B) and treated cells ( Figure 2C). However, the western blotting analysis demonstrated that TRAP1 levels decreased (61.4% ± 9.1%) upon exposure to 500 µM H 2 O 2 for 24 h (Figure 2D,E).

Impact of Oxidative Stress on TRAP1 Levels
Exposure to H2O2 is commonly used as a model to assess the oxidative stress susceptibility and antioxidant activity of RPE cells [6,7,33]. To define the optimal dose and incubation time, we exposed ARPE-19 cells to different concentrations of H2O2 for 2, 4 and 24 h and assessed the cell viability using the resazurin assay (Figures 2A and S1). As expected, H2O2 decreased the viability of ARPE-19 cells in a dose-dependent manner ( Figure  2A). For subsequent experiments, 500 μM H2O2 (24 h incubation) was used as these conditions decreased the cell viability of ARPE-19 cells by 53.5% ± 6.6% compared with the control (non-treated cells; p = 0.0286). To assess if the levels and localization of TRAP1 in ARPE-19 cells changed upon oxidative stress, cells were incubated with 500 μM of H2O2 for 24 h (Figure 2). Our results showed that the localization of TRAP1 was similar in both non-treated ( Figure 2B) and treated cells ( Figure 2C). However, the western blotting analysis demonstrated that TRAP1 levels decreased (61.4% ± 9.1%) upon exposure to 500 µ M H2O2 for 24 h (Figure 2D,E). . The values are presented as a percentage of the control (non-treated cells). Cell metabolism was assessed using the resazurin assay. Four independent experiments were performed (n = 4). For the graph (A), statistical significance was calculated by using the Mann-Whitney U test. For the graph (E), statistical significance was calculated by using an unpaired t-test. Statistically significant values: * p < 0.05, ** p < 0.01, *** p < 0.001. In the graphs, all values are expressed as mean ± standard error of the mean (SEM).

TRAP1 Silencing Decreases Cell Metabolic Activity and Increases Reactive Oxygen Species Production
We wondered if TRAP1 silencing affects cell metabolism (measured as dehydrogenase activity), production of reactive oxygen species (ROS), proliferation and survival of RPE cells. The levels of TRAP1 were reduced by approximately 50% at 72 h upon silencing . The values are presented as a percentage of the control (non-treated cells). Cell metabolism was assessed using the resazurin assay. Four independent experiments were performed (n = 4). For the graph (A), statistical significance was calculated by using the Mann-Whitney U test. For the graph (E), statistical significance was calculated by using an unpaired t-test. Statistically significant values: * p < 0.05, *** p < 0.001. In the graphs, all values are expressed as mean ± standard error of the mean (SEM).

TRAP1 Silencing Decreases Cell Metabolic Activity and Increases Reactive Oxygen Species Production
We wondered if TRAP1 silencing affects cell metabolism (measured as dehydrogenase activity), production of reactive oxygen species (ROS), proliferation and survival of RPE cells. The levels of TRAP1 were reduced by approximately 50% at 72 h upon silencing TRAP1 expression using specific siRNAs (siTRAP1) ( Figure 3A,C-E) [30]. In contrast, the transfection of a control scramble RNA (siCTL), used as a control, did not affect TRAP1 levels ( Figure 3A,B,D,E). Seventy-two hours after TRAP1 silencing, we detected a decrease of 33.8% in cell metabolic activity as measured by the resazurin assay ( Figure 3F). Moreover, TRAP1 silencing did not exacerbate the decrease in cell metabolic activity in ARPE-19 cells exposed to 500 µM of H 2 O 2 for 24 h ( Figure 3G). The total number of cell nuclei ( Figure 3H), cell mass (measured by Sulforhodamine B (SRB) assay ( Figure 3I)), and the number of proliferating cells (measured by the number of Ki67-positive cells ( Figure 3J)) remained similar to the control (siCTL). The data suggest that TRAP1 silencing promotes a decrease in cell metabolic activity in live cells, rather than an increase in cell death or a reduction in proliferation.
Antioxidants 2023, 12, x FOR PEER REVIEW 11 of 18 TRAP1 expression using specific siRNAs (siTRAP1) ( Figure 3A,C-E) [30]. In contrast, the transfection of a control scramble RNA (siCTL), used as a control, did not affect TRAP1 levels ( Figure 3A,B,D,E). Seventy-two hours after TRAP1 silencing, we detected a decrease of 33.8% in cell metabolic activity as measured by the resazurin assay ( Figure 3F). Moreover, TRAP1 silencing did not exacerbate the decrease in cell metabolic activity in ARPE-19 cells exposed to 500 μM of H2O2 for 24 h ( Figure 3G). The total number of cell nuclei ( Figure 3H), cell mass (measured by Sulforhodamine B (SRB) assay ( Figure 3I)), and the number of proliferating cells (measured by the number of Ki67-positive cells ( Figure 3J)) remained similar to the control (siCTL). The data suggest that TRAP1 silencing promotes a decrease in cell metabolic activity in live cells, rather than an increase in cell death or a reduction in proliferation.  To assess the effect of TRAP1 silencing on free radical levels, we evaluated the levels of superoxide anion production by dihydroethidium (DHE) assay at 24 and 72 h after transfection (Figure 4). Upon TRAP1 silencing, we observed an increase in DHE signal 24 and 72 h after transfection ( Figure 4A-C and Figure 5D-F), suggesting that TRAP1 may regulate superoxide anion levels in these cells.
To assess the effect of TRAP1 silencing on free radical levels, we evaluated the levels of superoxide anion production by dihydroethidium (DHE) assay at 24 and 72 h after transfection (Figure 4). Upon TRAP1 silencing, we observed an increase in DHE signal 24 and 72 h after transfection ( Figures 4A-C and 5D-F), suggesting that TRAP1 may regulate superoxide anion levels in these cells.  crease in basal extracellular acidification rate (ECAR) ( Figure 6D). Measurement of the two major energy-producing pathways of the cell, mitochondrial respiration (measured as the OCR) and glycolysis (indirectly measured as the ECAR) can reveal the four bioenergetic phenotypes that a cell can display: (a) quiescent; (b) energetic; (c) aerobic; and (d) glycolytic. Interestingly, by plotting the average data of basal OCR ( Figure 6B) against basal ECAR (Figure 6D), we observed that TRAP1 silencing shifted the cell's metabolic profile towards a more quiescent phenotype compared with untreated cells (Figure 6B,D).  . Representative mitochondria are presented. Scale bars in A and B: 100 nm. Analysis of confocal microscopy images obtained using Image-iT™ TMRM Reagent (TMRM) for mitochondrial labelling in ARPE-19 cells transfected with siCTL (C) or siTRAP1 (D). Scale bars in C and D: 50 µm. The mitochondrial interconnectivity was similar between the two experimental groups (E). siTRAP1 cells showed increased mitochondrial elongation (F). Statistical significance was calculated by using an unpaired t-test. Statistically significant values: *** p < 0.001. In the graphs, all values are expressed as mean ± standard error of the mean (SEM).

TRAP1 Silencing Impairs Mitochondrial Structure
We have characterized the outcome of TRAP1 silencing on mitochondrial physiology. We started by assessing the impact on the mitochondrial ultrastructure by transmission electron microscopy (TEM). TRAP1 silencing in ARPE-19 cells had no major effect on the mitochondrial ultrastructure, in which clear cristae and membranes were observed ( Figure 5A,B). To complement the ultrastructural analysis, mitochondria were stained with Image-iT™ TMRM Reagent (TMRM), and morphological parameters such as mitochondrial interconnectivity and elongation were quantitated [27]. No fragmentation of mitochondria was observed in siTRAP1 cells compared with the control (Figure 5C,D). Moreover, silencing of TRAP1 did not affect the mitochondrial interconnectivity ( Figure 5E) but increased mitochondrial elongation ( Figure 5F) compared with siCTL cells. The effects of TRAP1 silencing on ARPE-19 cells mitochondrial oxygen consumption rates (OCR) and extracellular acidification rate (ECAR) were also evaluated ( Figure 6A). TRAP1 silencing significantly reduced the basal (40.1%) and maximal (42.8%) OCR (Figure 6 B,C) as well as the ATP production-linked (36.2%) and proton leak-related (14.6%) OCR ( Figure 6E,F). Furthermore, the decrease in mitochondrial OCR was paralleled by a slight decrease in basal extracellular acidification rate (ECAR) ( Figure 6D). Measurement of the two major energy-producing pathways of the cell, mitochondrial respiration (measured as the OCR) and glycolysis (indirectly measured as the ECAR) can reveal the four bioenergetic phenotypes that a cell can display: (a) quiescent; (b) energetic; (c) aerobic; and (d) glycolytic. Interestingly, by plotting the average data of basal OCR ( Figure 6B) against basal ECAR ( Figure 6D), we observed that TRAP1 silencing shifted the cell's metabolic profile towards a more quiescent phenotype compared with untreated cells (Figure 6B,D).

Discussion
Tumor necrosis factor receptor-associated protein 1 (TRAP1) is expressed in several tissues and cell types [6,20,25,[34][35][36][37][38][39]. TRAP1 mRNA is reported to be variably expressed Figure 6. TRAP1 silencing drives a quiescent metabolic state. Representative images of oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) measurements using a Seahorse XFe96 Extracellular Flux Analyzer (A-F). siTRAP1 cells presented a decreased basal and maximal OCR compared with the siCTL cells (B,C). A shift towards a more quiescent phenotype was observed when the average mitochondrial basal OCR was plotted against the average basal ECAR (D). A decrease in mitochondrial OCR was paralleled by a slight decrease in ECAR (D). siTRAP1 silencing significantly reduced the ATP production-linked (E) and proton leak-related (F) OCR. Data are the mean ± SEM of 3 independent experiments and show the effects of mitochondrial inhibitors (a) oligomycin (2 µM), (b) FCCP (1 µM) and (c) antimycin A (1 µM) plus rotenone (1 µM) injected as indicated. The results are expressed as pmol O 2 /min/cell mass for OCR and mpH/min/cell mass for ECAR. Statistical significance was calculated by using an unpaired t-test. Statistically significant values: ** p < 0.01. In the graphs, values are expressed as mean ± standard error of the mean (SEM).
In the present study, we showed that TRAP1 is present in the mitochondria of human RPE cells and that TRAP1 silencing in ARPE-19 cells did not affect the number of cells. However, others have demonstrated that TRAP1 knockdown inhibits cell survival in different lung cancer cells, H1299 (human non-small cell lung carcinoma cell) and A549 (adenocarcinomic human alveolar basal epithelial cells) [20,25]. In our experiments, TRAP1 silencing also did not decrease cell proliferation. This is in contrast with results reported by Agorreta et al., and Palladino et al., who showed that downregulation of TRAP1 arrests cell proliferation in the A549 lung cancer cell line and thyroid carcinoma cells [36,37]. The discrepancy might be related to the levels of TRAP1 ablation or to the cell type itself.
After demonstrating the expression of TRAP1 in the RPE, we also analyzed if hydrogen peroxide-induced damage changed the TRAP1 levels and localization. We observed that TRAP1 levels were decreased 24 h after exposure to 500 µM of H 2 O 2 , without affecting the localization of the protein (Figure 2). However, others have demonstrated that TRAP1 expression is up-regulated following treatment with H 2 O 2 (300 µM) for 24 h in a neuroblastoma cell line (SH-SY5Y) [38]. The discrepancy in the effect of H 2 O 2 treatment in TRAP1 levels might be related to the cell type itself, since several studies show that the role of TRAP1 might differ depending on the cell type [20,36,39].
Our data showed that TRAP1 silencing results in increased ROS production (Figure 4), suggesting that TRAP1 plays an important role in regulating ROS production in RPE cells. Increased ROS production upon TRAP1 silencing was also previously reported in colon cancer cells (HCT116) [6], lung cancer cells [20] and HeLa cells upon exposure to H 2 O 2 [39]. Finally, we also assessed the impact of TRAP1 silencing on mitochondria morphology. TRAP1 silencing led to increased mitochondrial elongation without affecting mitochondrial interconnectivity ( Figure 5). Previous studies performed in MRC-5 cells demonstrated that cells with reduced levels of TRAP1 did not display alterations in mitochondrial morphology compared with the control cells and presented a complex interconnected mitochondrial network [20]. However, the same authors demonstrated that in A549 cells, TRAP1 silencing reduced mitochondrial elongation and interconnectivity [20]. It is known that TRAP1 plays a role in the regulation of mitochondrial energy metabolism, including oxidative phosphorylation (OXPHOS) processes, by interacting with proteins of the mitochondrial electron transport chain (ETC) complex II and complex IV [10]. Furthermore, we assessed the role of TRAP1 in the metabolic pattern and abilities of ARPE-19 cells. Our data showed that decreased levels of TRAP1 led to an overall decrease in respiratory capacity, driving cells into a quiescent state ( Figure 6). Similarly, others demonstrated that A549 siTRAP1 cells presented decreased oxygen consumption without altered glycolytic rates [20]. In contrast, TRAP1 knockout murine adult fibroblast cells displayed a higher basal oxygen consumption rate and a significantly higher maximum respiratory capacity than control cells [36]. Recently, Xiao et al. also demonstrated that TRAP1 overexpression, but not TRAP1 ablation, increased maximal respiration and ATP production in cancer-associated fibroblasts [39].
In summary, our study demonstrates that TRAP1 silencing increases ROS production by human RPE cells and drives RPE cells into a quiescent metabolic state (Figure 7). Since oxidative-stress-induced RPE damage and mitochondrial dysfunction in RPE are hallmarks of AMD [7], it is plausible to speculate that TRAP1 might play a role in AMD pathology, opening new avenues for the development of therapeutical approaches based on the modulation of TRAP1. Therefore, future experiments to test the efficacy of TRAP1 gene augmentation therapy in animal models of AMD are needed [43]. the modulation of TRAP1. Therefore, future experiments to test the efficacy of TRAP1 gene augmentation therapy in animal models of AMD are needed [43].

Figure 7.
Graphical representation of the highlights of our study. TRAP1 is present in retinal pigment epithelial cells and is located mainly in the mitochondria. TRAP1 silencing increases reactive oxygen species (ROS) production and decreases mitochondrial respiratory capacity.

Supplementary Materials:
The following supporting information can be downloaded at: www.mdpi.com/xxx/s1, Figure S1:  Graphical representation of the highlights of our study. TRAP1 is present in retinal pigment epithelial cells and is located mainly in the mitochondria. TRAP1 silencing increases reactive oxygen species (ROS) production and decreases mitochondrial respiratory capacity.