Chronic, low-dose rotenone reproduces Lewy neurites found in early stages of Parkinson's disease, reduces mitochondrial movement and slowly kills differentiated SH-SY5Y neural cells

The neuronal phenotype of differentiated SH-SY5Y cells was confirmed by labeling with antibodies to the neuron specific beta-tubulin (TUJ1), the 68 kD neurofilament subunit (NF68) and with the microtubule associated proteins MAP-2 and tau (Figure 1). Both MAP-2 and tau were uniformly distributed to the cell body and neurites. Tau was not segregated to the axon-like processes. Synaptophysin and α-synuclein (αSYN) were also detected by immunocytochemistry in differentiated SH-SY5Y cells and exhibited punctuate staining consist with their localization in synaptic vesicles (Figure 1). The dopaminergic phenotype of differentiated SH-SY5Y cells was confirmed by immunostaining with tyrosine hydroxylase (TH), dopamine transporter (DAT), the vesicular monoamine transporter (VMAT) and the D2 dopamine receptor (D2) (Figure 1). The dopaminergic markers also exhibited punctuate staining in the cell body and processes of SH-SY5Y cells in culture. As part of a separate study we have also confirmed that staurosporine-differentiated SH-SY5Y cells express choline acetyltransferase (data not shown), indicating that these cells continue to express multiple neurotransmitter types.

Chronic treatment with 50 nM rotenone induced the formation of neuritic swellings beginning at day 3 and continuing for up to 20 days. The number of neuritic swellings per culture was low but they were readily identifiable. Cultures of staurosporine-differentiated, rotenone-treated SY5Y neural cells were stained with antibodies to α-synuclein and ubiquitin (Figure 2). These neuritic swellings morphologically resemble the α-synuclein and ubiquitin immunoreactive Lewy neurites seen in increasing numbers in PD brain samples [1‐4]. Neuritic swellings generated by rotenone exposure also contained concentrations of bioenergetically competent mitochondria, based on their capacity to accumulate the potentiometric dye MitoTracker CMX-Ros (MtRed) and LysoTracker positive lysosomes (Figure 2). The dynamic movement of mitochondria in rotenone-treated neurites exhibiting neuritic swellings was reduced in comparison with intact neurites (Figure 3). Reduction in mitochondrial movement was statistically significant by 8 days of rotenone treatment and was greatest at 16 days, the longest period studied.

50 nM rotenone produced a biphasic survival curve with ~40% loss by 6 days and a second decline to ~60% loss between 18 and 21 days (Figure 4). The differentiated cells surviving in rotenone exhibited progressive loss of processes (Figure 5). General activation of caspases, monitored with zVAD.fmk-FITC staining, was visible by 2 days of rotenone at a low level with small intense foci (Figure 6). By 11 days of rotenone more generalized caspase activation was noted that was still visible at 21 days (Figure 6). We did not observe any evidence for loss of mitochondrial membrane potential, monitored with JC-1 staining (Figure 7).

SH-SY5Y cells were routinely cultured in growth medium (GM), consisting of high glucose DMEM with 10% FBS, 100 μg/ml pyruvate, 50 μg/ml uridine, and antibiotic-antimycotic: 100 Units/ml penicillin G, 100μg/ml streptomycin, 0.25 μg/ml amphotericin β. Cells were fed every 2–3 days and passed once a week. Cells were differentiated by culturing in Neurobasal medium and adding staurosporine at 6–10 nM (differentiation medium, DM).

At the end of differentiation, samples were collected for Day 0 data and medium was changed to begin 50 nM rotenone treatment on selected flasks and dishes. Both rotenone and staurosporine were added, fresh, to DM at each medium change.

To immunostain differentiated SH-SY5Y neural cells, cells were grown in 35 mm glass bottom culture dishes [21], fixed with 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS), rinsed with PBS and incubated in blocking solution (0.05% bovine serum albumin and 0.2% triton X-100 in PBS) for 30 min at room temperature. Then primary antibodies for each antigen were diluted in PBS containing a 1:10 dilution of blocking buffer and incubated overnight at 4°C. The primary antibodies used for this study include-α-synuclein (1:500, AB5038), dopamine transporter (1:1000, MAB369), neurofilament 68 (1:100, AB1983), synaptophysin (1:100, MAB368), tau (1:100, AB1512), tyrosine hydroxylase (1:1000, AB151), ubiquitin (1:250, MAB1510) and vesicular monoamine transporter (1:200, AB1767) from Chemicon International, Inc. (now owned by Millipore, Temecula, CA). The antibody for the D2DR dopamine receptor (1:100, sc-9113) was obtained from Santa Cruz Technology Inc. (Santa Cruz, CA), the microtubule associated protein 2 was purchased from (MAP-2, 1:62, 13–1500) Zymed now owned by Invitrogen, (Carlsbad, CA), and the Neuronal Class III ^®-Tubulin was obtained from Dr. Anthony Frankfurter (University of Virginia). After 3 washes in PBS, the cells were incubated in appropriate fluorescein or Texas-red labeled secondary antibodies (1:100, Vector Labs, Burlingame, CA) for 30 min at room temperature and mounted with Vectashield (Vector Labs, Burlingame, CA). The fluorescently-labeled anti-rat secondary antibodies needed to image the dopamine transporter primary antibody were purchased from Jackson ImmunoResearch Labs, Inc (West Grove, PA). Images of stained cybrid neurons were created using an Olympus Fluoview laser scanning confocal microscope system.

To label mitochondria, differentiated cybrid cells were incubated with 15 nM MitoTracker CMXRos (Invitrogen, Carlsbad, CA) for 10 min at 37°C. After a brief wash in GM, the cells were imaged using an Olympus IX70 inverted microscope equipped with epifluorescence and Nomarski optics, a Lambda 10-2 filter wheel, a Photometrics CoolSnap HQ progressive scan CCD camera and MetaMorph software (Molecular Devices). Mitochondrial movement was visualized and quantitated as described [22]

Differentiated SH-SY5Y cells were grown and treated with rotenone and staurosporine in 35 mm dishes as described above. Labeling solutions and labeled cells were protected from light during the following steps. To label suspended cells, 40–50 ml of "spent" medium was collected from 20 dishes, centrifuged at 250 × g for 8 min, the pellet was resuspended in 400 ul Dulbecco's Modified Eagle Medium with high glucose, HEPES buffer and without phenol red (Gibco-Invitrogen, Grand Island, NY) and 200 ul plated on the coverslip area of two, 1% porcine skin gelatin-coated, 35 mm dishes. After 30 min at 37°C, 100 ul of HEPES medium containing 30 uM CaspACE FITC-zVAD-FMK (Promega, Madison, WI) and 240 nM MitoTracker Red (Molecular Probes, Eugene, OR) was added and the cells incubated for another 30 min at 37°C. To label attached cells, maintenance media was removed and cells incubated for 30 min. at 37°C in HEPES medium containing 10 uM CaspACE FITC-z VAD-FMK and 80 nM MitoTracker Red. All labeled dishes were washed with Hank's Balanced Salt Solution (Gibco-Invitrogen, Grand Island, NY), fixed for 20 min in 4% phosphate buffered paraformaldehyde with 4% sucrose, pH 7.4, washed with PBS and coversliped using Vectashield mounting medium (Vector Laboratories, Burlingame, CA). Single plane confocal images of labeled cells were collected using an Olympus Confocal Laser Scanning System mounted on an Olympus IX70 microscope.

We examined gene expression in differentiated SH-SY5Y after 3, 6, 10, 13, 17 and 20 days of exposure to 50 nM rotenone. Our approach to microarray studies was to emphasize both technical and biological replication. To achieve those goals, we carried out three temporally consecutive, independent studies of rotenone exposure. SH-SY5Y were differentiated as above and then carried forward in culture with or without 50 nM rotenone. At each time point of rotenone exposure, RNA was harvested from rotenone-exposed and non-exposed companion cultures. This procedure attempted to control for any effects of the differentiation procedure and "aging" in culture, leaving rotenone exposure as the only variable.

RNA samples were reverse transcribed to fluorescein labeled (CTL cells) or biotin labeled (ROT cells) cDNA using the Micromax TSA™ Labeling and Detection Kit (Perkin Elmer Life and Analytical Sciences, Inc., Waltham, MA). ROT samples for each time point were each assayed against the corresponding CTL SH-SY5Y cells.

The two labeled cDNAs were then mixed and hybridized to a 19.2K human gene array (University Health Network Microarray Center, Toronto, Ontario) overnight at 40°C in a humidified chamber. Detection and amplification of signal were accomplished using a series of antibody-enzyme conjugates catalyzing the deposition of Cyanine 3 tyramide (Cy3) to CNT or UNT cDNA and Cyanine 5 tyramide (Cy5) to PD or ROT cDNA. Arrays were hybridized in duplicate and imaged in a scanning confocal slide reader Scanarray^® 4000, (Packard Biochip Technologies, LLC, Billerica, MA). Random dye-swap experiments with arrays and labeling reagents from the above manufacturers always produced > 90% concordance for expression levels.

Duplicates of background-subtracted, normalized to total array fluorescence transcript fluorescence ratios (Rot/Ctl) were first averaged then processed for significance using B.A.G.E.L. (Bayesian Analysis of Gene Expression Levels, [23]). The B.A.G.E.L. algorithm makes no assumptions about what expression levels are "biologically significant" and instead calculates significance for a given transcript based on its variance compared to the overall population variance. In the B.A.G.E.L. analyses performed, all three independent rotenone experiments at each time point of rotenone exposure were analyzed together.

We used the B.A.G.E.L. statistical output to devise a liberal approach to filter transcripts used for subsequent analysis. To analyze transcript expression patterns, we first collected all transcripts at each rotenone treatment time point or in the PD brain samples that achieved p < 0.05 significance in the B.A.G.E.L. analysis and were not corrected for multiple comparisons. The GenBank transcripts were then annotated to gene symbols using D.A.V.I.D. (Database for Annotation, Visualization and Integrated Discovery, http://​david.​abcc.​ncifcrf.​gov/​) and analyzed using the Cluster [24] algorithm to create visual patterns of gene changes across time of rotenone exposure compared to those observed in the PD brain samples. We utilized the latest version of Cluster 3.0 to perform hierarchical clustering and JavaTreeView (v. 1.0.12, http://​jtreeview.​sourceforge.​net) to view the Cluster output.

Primers for mtDNA and nDNA-encoded ETC genes were designed with Beacon Designer software and used with SyberGreen detection to assay relative gene expression in cDNA's prepared by reverse transcription of the RNA samples. All assays were carried out in an iQ5 instrument (BioRad). Relative gene expression was calculated with the method of Pfaffl [25].