The accumulation of lipofuscin in the retinal pigment epithelium (RPE) is associated with the normal process of aging, but can also be an element in the pathological process associated with several retinal degenerative diseases [1
]. There is also evidence that lipofuscin accumulation can lead to RPE and photoreceptor cell death [4
]. Many aspects of this important biological process are not well understood [5
]. In particular, little is known about the cellular mechanisms that control the effective degradation of significant amounts of photoreceptor outer segment material by RPE during the process of photoreceptor outer segment renewal. In most mammals, each RPE cell opposes many dozens of rod photoreceptors and must process thousands of outer segment discs every day [7
]. For example, in the primate perifovea, each RPE cell services about 40 rod cells [8
] and the volume of the processed outer segment material on a daily basis (~10% of the length of each outer segment, phagocytized usually in a matter of few hours), which can be estimated to be equal to about 20% of the volume of one RPE cell. However, relatively little lipofuscin material accumulates in the RPE of normal subjects, especially during the first decade of life [9
]. Therefore, despite the heavy metabolic overload imposed by the considerable volume of rod outer segment (ROS) material to be phagocytized, degraded, and recycled, a very efficient and tightly regulated mechanism exists to keep the level of undigested material very low, which is an essential condition for the proper functioning of an RPE cell.
However, some accumulation of lipofuscin material still occurs with normal aging and even more is seen in some pathological conditions, like age-related macular degeneration (AMD) [10
], Stargardt’s disease [12
], and even retinitis pigmentosa [14
]. Therefore, it is likely that even small inefficiencies in the process of ROS degradation can compound over time and interfere with the functioning of the RPE cell layer and the overlying retina. The realization of this RPE vulnerability has stimulated an intensive research effort to understand the mechanisms which control the relatively efficient degradation under normal conditions and the pathological processes associated with their defects.
It has long been accepted that during the normal functioning of macrophages, different components of phagocytized material (lipids, amino acids, and carbohydrates) accumulate in the lysosomes [16
]. Similarly, for a long time, the RPE lysosomes have been suspected to play a central role in the degradation of photoreceptor outer segment material [18
]. The lysosomal complex participates in autophagy and there is significant indirect evidence to suggest that this process may play a central role in maintaining relatively low lipofuscin levels in normal RPE cells [19
]. Theoretically, if the process of autophagy is suppressed, a gradual accumulation of lipofuscin material occurs over time. Conversely, if stimulation of this process occurs, the results include an increase in the speed and efficiency of the degradation of newly-formed lipofuscin, less lipofuscin accumulation in the RPE cells over time, and possibly some degradation of existing undegraded material.
Since the degradation of RPE lipofuscin is a complicated, multi-step process and autophagy is the most important intracellular catabolic process involved in protein and organelle degradation via the lysosomal pathway, manipulating the process of autophagy in RPE cells is being explored as a potential way to regulate early lipofuscin formation [22
]. However, there is still a lack of direct evidence on whether inhibiting or stimulating autophagy increases or reduces the amount of lipofuscin-like material in the RPE cell. The goal of this work was to investigate some important aspects of this process. By adding different pharmacologic agents and conducting various experiments, we investigated several aspects of the role of lysosomes, autophagy, and the proteasome system in the fate of RPE lipofuscin degradation. Thus, this work could be a step forward towards defining new targets for drug development in AMD and other lipofuscin-related diseases.
4. Materials and Methods
4.1. RPE Cell Culture, ROS Isolation and Modification
ARPE-19 cells were cultured and maintained as previously described [51
]. Briefly, ARPE-19 cells were procured from the ATCC (Manassas, VA, USA) and grown in high-glucose DMEM (Cellgro/Mediatech Inc., Manassas, VA, USA) supplemented with 10% heat-inactivated fetal calf serum FCS (Sigma-Aldrich, St. Louis, MO, USA) and 1% penicillin/streptomycin (Gibco, Grand Island, NY, USA), at 37 °C in the presence of 5% CO2
. Cells were routinely subcultured or harvested for experiments using TrypLE Express (Gibco). ROS were prepared from cattle eyes following a described method [26
]. The yield was normally 10–20 nmol of rhodopsin per retina, with an OD280/OD500 ratio of 2.3 to 2.6. Oxidized ROS using 4-HNE were prepared as previously described [62
]. The protein content of ROS preparations was measured by a BioRad BC kit (Bio-Rad Laboratories, Hercules, CA, USA). The concentrations of protein modifications resulting from this procedure have been previously reported [62
]. Modified ROS were stored at −80 °C until use.
4.2. RPE Cell Treatment
Post confluent, stationary ARPE-19 cells cultures in 10 cm plates were trypsinized, plated in 24-well or 8-well chamber slides at a confluent density of 1.66 × 105/cm2. After an additional culturing for seven days, two different types of ROS were added every day for seven days. The types of ROS were: native ROS or HNE-modified ROS (see above). Unless stated otherwise, the daily dose of ROS was always 4 µg per cm2 growth area. The lysosomal inhibitors NH4Cl (10 or 20 mM) and chloroquine (20 µM), the proteosomal inhibitor MG-132 (10 µM), the autophagy inhibitor 3-methyladenine (3-MA) (10 mM), and the mTOR pathway inhibitors rapamycin (1 or 10 µM), PI-103 (1 µM), PIK-90 (1 µM), and Ku-0063794 (1 µM) were added separately on two consecutive days after seven days of feeding with ROS. All non-internalized ROS were washed out before the addition of inhibitors/stimulators and this procedure did not interfere with the binding or internalization of ROS.
4.3. Flow Cytometry
Flow cytometry was used to evaluate the change in RPE autofluorescence with different treatment. Cells were cultured in 24-well plates and incubated with different components, as described in the section RPE cell culture, ROS isolation and modification. Cells were repeatedly washed, detached with trypsin, and analyzed on a C6 flow cytometer (Accuri Cytometers, Inc. Ann Arbor, MI, USA). A gate was set to exclude cell debris and cell clusters, and 10,000 gated events were recorded. Experiments were performed in triplicate. Two channels were used, with an excitation wavelength of 488 nm: FITC/GFP channel (533/30 nm) and the PE/PI channel (585/40 nm).
4.4. RNA Interference
Small interfering RNA (siRNA) oligonucleotides against mTOR, ATG5, and ATG7 were purchased from Cell Signaling Technology (Danvers, MA, USA). For siRNA transfection, cells were seeded at 300,000/well in 24-well tissue dishes, and were transfected with 100 nM of the pooled oligonucleotide mixture by using Lipofectamine (Invitrogen, Carlsbad, CA, USA), following the manufacturer’s protocols. The transfection media were removed after 6 h, and cells were allowed to recover in complete growth media for 36–48 h, before being used in experiments. Western blots were used after three days of transfection, to confirm the effect of RNA interference. Then, the time point at which the decrease in the protein level was most pronounced was determined as three days and, therefore, ROS were supplemented with RPE cells from that time point for another three days, to explore the effects of the suppression of mTOR, ATG5, and ATG7 through RNA interference on RPE cell autofluorescence.
4.5. Lentivirus Vectors
Lentiviral vectors were used as a complementary approach to achieve an mRNA inhibition of proteins associated with autophagy. Two lentiviral constructs: shATG5 D6 and shATG5 D9 (Open Biosystems, Huntsville, AL, USA), expressing small hairpin RNAs against Atg5, and two lentiviral constructs: shATG7 84 and shATG7 87 (Sigma-Aldrich), expressing small hairpin RNAs against Atg7, were constructed at the Gene Therapy Center UMASS Medical School. Lentiviral shRNA constructs of mTOR were purchased from Addgene (Cambridge, MA, USA).
pLenti-mCherry-LC3, a construct that produces lentivirus expressing a fusion of the fluorescent protein mCherry (excitation 587 nm; emission 610 nm) and the autophagic marker microtubule-associated protein light chain 3 (LC3) (NM-026160.4), was constructed by cutting out an mCherry-LC3 fragment with SnaBI and EcoRI from pBebe-mCherry-LC3, and then blunting the ends with T4-polymerase, finally ligating into AfeI/XhoI-cut and T4-polymerase-blunted lentivirus construct backbone pLenti. Lentivirus was produced at the Gene Therapy Center of UMASS Medical School following the standard protocol for the transient co-transfection of 293 T cells with the vector plasmid, packaging plasmid, and VSVG envelope plasmid. Briefly, using phosphate calcium methods, the lentivirus vector plasmid (5 μg) and helper plasmids pCMV 8.91 (5 µg)/pCMV-VSVG (3 µg) were co-transfected into 293T cells with 60–70% confluence growing in 100 mm plate. We replaced the culture medium into 7 mL serum-free Opti-MEM one day after transfection, and collected the medium containing lentivirus one day later, then refilled the plate with serum-free Opti-MEM and recollected the medium after another day. The typical lentivirus titer from one 100 mm plate was 105 virus particles/mL ARPE cells in ~40% confluence. Cells were infected with lentivirus for one day by replacing half of the culture medium with lentivirus medium. ARPE cells infected with shRNA lentivirus were seeded at 300,000/well in 24-well tissue dishes. After three days of culture (based on the Western blot’s result), the cells were fed with ROS for three days.
The target sequences of each construct were as follows (5′ to 3′):
shATG5 D6 (TRCN0000151963): GGATGAGATAACTGAAAGG;
shATG5 D9 (TRCN0000151474): GGCATTATCCAATTGGTTT;
shATG7 84 (TRCN0000007584): GCCTGCTGAGGAGCTCTCCAT;
shATG7 87 (TRCN0000007587): CCCAGCTATTGGAACACTGTA;
mTOR 2: CCGCATTGTCTCTATCAAGTT.
4.6. Analysis of Punctate Cherry-LC3
ARPE-19 cells transfected with the mCherry-LC3 construct were seeded at 1.66 × 105/cm2 in an 8-well chamber slide. After three days of cell culture, test drugs (rapamycin, Ku0063794, PI 103, PIK90, NH4Cl and chloroquine) were twice added every 24 h and subsequently fixed with 4% PFA, washed three times with 1× PBS, mounted on slides with Vectashield mounting media (Vector Laboratories, Burlingame, CA, USA), and analyzed by confocal microscopy at the UMass Medical School Digital Light Microscopy Core Facility.
4.7. Immunoblot Analysis
Cells were lysed in RIPA Lysis Buffer System (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA). Samples were lysed using five thaw-freeze cycles. Lysates were cleared by centrifugation for 15 min at 4 °C, boiled in SDS sample buffer, resolved using SDS-polyacrylamide gel electrophoresis (S-PAGE), and transferred to the polyvinylidenedifluoride membrane. The membranes were blocked in Odyssey Blocking buffer (LI-COR Biosciences, Lincoln, NE, USA), incubated with the primary antibodies indicated overnight at 4 °C, washed, incubated with conjugated secondary antibodies for one hour, and analyzed by Odyssey Imaging Systems (LI-COR Biosciences). Membranes were blotted with antibodies directed against mTOR, LC3, (all from Cell Signaling) and GAPDH (Millipore, Billerica, MA, USA). Bound antibodies were detected with a goat anti-mouse or rabbit secondary antibody (Odyssey Infrared Image System, LI-COR Biosciences).
4.8. Confocal Microscopy
Cells were cultured in 8-well microscopy glass slides (Lab-Tek Chamber Slide; Nunc, Langenselbold, Germany) and treated with different components, as described. After seven days of feeding, cells were repeatedly washed to remove non-internalized ROS, fixed with 4% paraformaldehyde (PFA), stained with 1 mg/mL DAPI (AppliChem, Darmstadt, Germany), and mounted in Vectashield mounting medium (Vector Laboratories, Burlingame, CA, USA). Intracellular lipofuscin granules were documented on a Leica DM 6000 fluorescence microscope (Leica Microsystems, Wetzlar, Germany), using a fluorescein filter set with an excitation of 480/40 nm and an emission of 535/50 nm. Confocal microscopy was performed with a Leica TCS SP5 Spectral point scanning confocal microscope with a Leica DMI 6000 CFS (Confocal Fixed Stage, Leica Microsystems, Wetzlar, Germany). Cells were labeled with 1 µg/mL Hoechst stain (Sigma-Aldrich) for 5–7 min and visualized with the 405, 488, and 561 nm lasers. At the UMass Medical School Digital Light Microscopy Core, confocal microscopy was performed with a Solamere Technology Group (Salt Lake City, UT, USA) CSU10B Spinning Disk Confocal System, which consisted of a CSU10B spinning disk confocal scan head (Yokogawa Electric Corporation, Tokyo, Japan), with high efficiency dichroic mirrors and laser blocking filters attached to a Nikon TE2000-E2 motorized inverted fluorescence microscope equipped with a Nikon’s Perfect Focus System (PFS) (Nikon Instruments, Melville, NY, USA) and a custom acousto-optical tunable filter (AOTF) controlled laser launch, with 405, 488, 561, and 636 nm lasers.
4.9. Live Cell Imaging and Fluorescence Quantification
ARPE-19 cells were cultured on 35-mm dishes containing a central 14-mm #1.5 glass coverslip (MatTek, Ashland, MA, USA; P35G-1.5-14-C). Dishes containing cell cultures were transported to the imaging facility and maintained at 37°C by a 20/20 Technology Inc., (Wilmington, NC, USA) objective heater (OHXX) and stage micro-incubator (INC-2000), which also provided a humidified 5% CO2–95% air atmosphere. Live cell confocal microscopy was performed on the CSU10B confocal system (see above). Cells were illuminated with an argon ion laser (~4 mWat 488 nm) using a Nikon VC Plan Apo 60× oil objective (NA = 1.4) and the fluorescence emission passed through a 610/60 nm bandpass filter, before entering a RoleraMGi EMCCD 14-bit camera (Qimaging, Surrey, BC, Canada). MetaMorph V7.6.3 (Molecular Devices, Sunnyvale, CA, USA) software was used for equipment control, image acquisition, and image analysis. Cells were imaged before the addition of 50 µL rapamycin (10 µM) or the vehicle (1× PBS), to establish the baseline intensity values and time-lapse imaging continued for 6 h at 30-min intervals. At each time point, a stack of either 16 z-slices (∆z = 0.5 µm) for rapamycin treatment, or 10 z-slices (∆z = 1.0 µm) for the controls (image capture parameters, t = 500 ms, EM gain = 4095, gain = 2× for each slice), was acquired. Focus was maintained between time points with Nikon’s PFS, which was turned off during Z-series acquisition.
4.10. Image Analysis of Live Cell Imaging
A MetaMorph review of the multidimensional data application was used to convert the Z-series time-lapse data to a maximum intensity projection at each time point. Stacks which contained “hot pixels” were replaced by a local average (Median filter). These pixels did not occur in the areas where the intensity measurements were made. The stack align feature of MetaMorph was used to align subsequent time-lapse images with the first image in the time series, to ensure that measured particles did not move in or out of the measurement regions. The intensity of the aligned stacks was corrected by subtracting the background fluorescence from a non-particle region of each image plane. MetaMorph’s region measurement tool was used to measure the average intensity in four circular regions (diameter = 24.84 µm, area = 480.99 µm2).
4.11. Statistical Analysis
Al data are presented as means and the error bars indicate the standard deviations. The results from the flow cytometry and the quantification of the results from the Western blots were analyzed using ANOVA and post-hoc Dunnett’s multiple comparisons tests. The analysis was performed using Prism 6 for Windows (GraphPad Software Inc., La Jolla, CA, USA).