Abstract
Parkinson’s disease (PD) is a neurodegenerative disease with movement disorders including resting tremor, rigidity, bradykinesia and postural instability. Recent studies have identified a new PD associated gene, TMEM230 (transmembrane protein 230). However, the pathological roles of TMEM230 and its variants are not fully understood. TMEM230 gene encodes two protein isoforms. Isoform2 is the major protein form (~95%) in human. In this study, we overexpress isoform2 TMEM230 variants (WT or PD-linked *184Wext*5 mutant) or knockdown endogenous protein in cultured SH-5Y5Y cells and mouse primary hippocampus neurons to study their pathological roles. We found that overexpression of WT and mutant TMEM230 or knockdown of endogenous TMEM230-induced neurodegeneration and impaired mitochondria transport at the retrograde direction in axons. Mutant TMEM230 caused more severe neurotoxicity and mitochondrial transport impairment than WT-TMEM230 did. Our results demonstrate that maintaining TMEM230 protein levels is critical for neuron survival and axon transport. These findings suggest that mutant-TMEM230-induced mitochondrial transport impairment could be the early event leading to neurite injury and neurodegeneration in PD development.
Introduction
Parkinson’s disease (PD) is a common neurodegenerative disease with movement disorders including resting tremor, rigidity, bradykinesia and postural instability. Loss of dopaminergic neurons in substantia nigra and presence of Lewy bodies (LBs) are the hallmarks of PD pathology. There are about 5–10% of the PD cases associated with genetic mutations or overexpression, yet the mechanisms underlying molecular pathogenesis of PD remains unclear (1). Recently, Deng et al. (2) identified a new PD-associated gene, TMEM230 (transmembrane protein 230), which mutations (Y92C, R141L, *184Wext*5 and *184PGext*5) are identified in PD cases. Following-up genetic studies suggest that some of TMEM230 variants are risk factors for PD development (3,4). However, the normal functions of TMEM230 and the pathological roles of its PD-related variants are not fully understood.
TMEM230 gene encodes two protein isoforms (2). Isoform1 is composed of 183 amino acids. Isoform2 lacks 63 amino acids at the N-terminal end of isoform1. In human, more than 95% of TMEM230 total protein isoforms is isoform2. Deng et al. identified four TMEM230 mutations, Y92C, R141L, *184Wext*5 and *184PGext*5, that are linked to PD. *184Wext*5 mutation replaces stop codon with six amino acids (WHPPHS), whereas *184PGext*5 mutations replace the stop codon with seven amino acids (PGHPPHS). Y92C mutation is found in sporadic PD, while R141L and *184Wext*5 are found in familial PD in North American cohorts. *184PGext*5 mutation is identified in Chinese PD families (2).
The physiological functions of TMEM230 are unclear (5). We and others previously found that overexpression of these four PD-linked mutations of TMEM230 induced cellular toxicity in cultured cells in vitro (6,7). TMEM230 isoform2 protein is predominantly expressed in cytosol, and localizes to vesicle structures in mouse neurons, including dopaminergic neurons in the substantia nigra (2,8). TMEM230 protein is enriched in rat brain synaptosome fractionation and displays a high co-localization with some trafficking-linked proteins (2,5). These findings suggest that TMEM230 may be involved in cellular trafficking or transport process (2,8). A functional axonal transport system maintains normal neuronal functions including mitochondrial energy balance, abnormal protein degradation and transportation of vesicles and other cargos. Neurons are polarized cells which use specific motor proteins to move in directions away from the cell body (anterograde) or toward the cell body (retrograde) along cytoskeletons composed of microtubules (9,10). Defects in axonal transport can lead to disruption in energy homeostasis, impaired protein clearance pathways, neurite shortening and eventual cell death (9,10). Impaired axonal transport has been shown to contribute to the pathogenesis of neuro-degenerative disorders, including Alzheimer's disease (ad), amyotrophic lateral sclerosis (ALS) and PD (11,12). However, whether TMEM230 alters axonal transport is unknown.
In this study, we are interested in understanding whether TEM230 plays a role in axon transport of mitochondria using human neuroblastoma SH-SY5Y cells and mouse primary hippocampus neurons. Mitochondrial dysfunction is a key pathological process in neurodegeneration and contributes to dopaminergic neuron loss in PD (13). We previously found that all four PD-linked TMEM230 variants induced apoptotic cell death in cultured HEK293 cells (7). Among these variants, *184Wext*5 induced the most severe cellular toxicity. Thus, in this study, we focused on investigating the roles of wild type and *184Wext*5 TMEM230 variants in axon transport of mitochondria. Genetic overexpression and siRNA knockdown approaches were used to assess the effects of upregulated and downregulated TMEM230 expression in neuronal transport and neurodegeneration. Our studies provide the novel insight into the mechanisms of TMEM230 in neurodegeneration underlying PD pathogenesis.
Results
Expression of WT and mutant-TMEM230 induced neurodegeneration
Our previous studies demonstrated that overexpression of WT-TMEM230 or PD-linked variants caused cell toxicity in HEK293T cells (7). Mutant *184Wext*-TMEM230 induced the severe toxicity than other TMEM230 variants (7). To further study the roles of TMEM230, human neuroblastoma SH-SY5Y cells were co-transfected with GFP and vector, WT-TMEM230 and PD-linked mutant TMEM230-*184Wext*5 at 1:10 ratio for 48 h. *184Wext*5 mutation replaces stop codon with six amino acids (WHPPHS) and is identified in PD cases in a genetic study (2). We found that expression of both WT and mutant TMEM230 reduced viability (Fig. 1A and B) and increased apoptosis (Fig. 1C and D) in SH-SY5Y cells compared with vector cells. Mutant TMEM230 induced more cell toxicity than those of WT-TMEM230.
TMEM230-induced neuronal degeneration in SH-SY5Y cells and primary neurons. A and B, SYSH5Y cells were transfected with GFP and WT or mutant TMEM230 at 1:10 ratio for 48 h. A. Representative pictures of SHSY5Y cells under fluorescent microcopy. Scale bar, 50 μm. B. SH-SY5Y cell viability from each experimental group. C and D. Apoptotic cell death in SH-SY5Y cells. C. Representative GFP positive health (top) and apoptotic (bottom) cells. Scale bar, 20 μm. E to H, Mouse primary neurons were transfected with GFP and WT or mutant TMEM230 at 1:10 ratio for 48 h. E. Representative pictures of GFP positive primary hippocampus neurons. Scale bar, 50 μm. F. The GFP positive hippocampus neuron viability. G. Hippocampus neurons with injury. H. Cortical neuronal death *P < 0.05, vs cells expressing vector and GFP. # P < 0.05 by ANOVA, vs cells expressing WT-TMEM230 and GFP.
To further validate these results, mouse primary neurons were co-transfected GFP with vector, WT or mutant TMEM230-*184Wext*5 at 1:10 ratio for 48 h. Consistent with our findings in SH-SY5Y cells, expression of WT or mutant TMEM230 reduced neuronal viability (Fig. 1E and F), increased neurite injury (Fig. 1E and G), and increased neuronal death (Fig. 1H) in mouse primary neurons compared with vector groups, while mutant TMEM230 caused severe neuronal degeneration than those of WT-TMEM230 (Fig. 1).
Knockdown of TMEM230 induced neurotoxicity
To further study the roles of TMEM230 in neuronal functions, SHSY5Y cells were transfected with specific siRNA targeting TMEM230 at 0, 5, 10 and 25 nM concentrations for 72 h. We found that TMEM230-siRNA at 25 nM could reduce endogenous TMEM230 up to 90% (Fig. 2A and B). Moreover, knockdown of TMEM230 in SH-SY5Y cells significantly reduced cell viability compared with random RNA (cRNA) control cells (Fig. 2C–E). Expression of recombinant WT-TMEM230 rescued the knockdown-induced neurotoxicity (Fig. 2C) by restoring TMEM230 protein level as to the vector control group (Fig. 2D and E), whereas expression of mutant TMEM230 had no rescuing effect (Fig. 2F). To further validate these findings, mouse primary hippocampus neurons were transfected with control RNA or mouse TMEM230-siRNA for 24 h, and followed by vector, WT or mutant TMEM230 transfection for 48 h. We obtained the similar results as we did in SH-SY5Y cells. Knockdown of mouse TMEM230 reduced neuronal viability and induced neuronal injury more than those of control neurons with random RNA (Fig. 2G to J). Expression of WT-TMEM230 rescued knockdown-induced neurodegeneration (Fig. 2G and H), but expression of mutant TMEM230 had no rescuing effect (Fig. 2I and J).
Knockdown of endogenous TMEM230-induced neurotoxicity. A and B, SHSY5Y cells were transfected with random RNA (cRNA) or specific siRNA targeting TMEM230 for 72 h. The cell lysates from each group were subjected to immunoblot analysis using anti-TMEM230 antibody. A. Representative blots from three separate experiments. B. Quantification of A. C to J, SHSY-5Y cells (C to F) or primary neurons (G to J) were transfected cRNA or TMEM230-siRNA for 24 h, followed by GFP and WT-TMEM230 transfection at 1:10 ratio for another 48 h. Cell or neuron viability and neurite injury were measured. C. Expression of recombinant WT-TMEM230 rescued the knockdown-induced neurotoxicity. D and E. TMEM230 protein levels in each experimental group as indicated were measured by western blot analysis using anti-TMEM230 (top) and anti-actin (bottom, loading control) antibodies. D. Representative blots from three separate experiments. E. Quantification of D. WT stands for expression of WT-TMEM230 cDNA plasmid. *P < 0.05 by ANOVA, vs control cells with cRNA and vector expression, # P < 0.05 by ANOVA, vs cells with WT-TMEM230. F. Expression of mutant TMEM230 did not alter the knockdown-induced neurotoxicity in SH-SY5Y cells by cell viability assays. G and I. The GFP positive hippocampus neuron viability. H and J. Hippocampus neurons with injury. *P < 0.05, vs control cells with cRNA and vector expression, # P < 0.05 by ANOVA, vs cells with TMEM230-siRNA and vector.
Mutant TMEM230 impaired mitochondrial transport at retrograde direction
To further study the roles of TMEM230, we used mouse primary hippocampus neurons to study the roles of TMEM230 in axon transport of mitochondria. Primary hippocampus neurons were transfected with GFP, Mito-DsRed and TMEM230 variants at a ratio of 1:10:10 for 12 h. The mid-segments of axons with both GFP and Mito-DsRed positive were recorded using living-time lapse imaging under fluorescent microcopy. As Fig. 3A illustrated, GFP (green) represented the tracking direction of an axon, and Mito-DsRed (red) reflected mitochondria within an axon, which could be captured by the time lapse living image system under the fluorescent confocal microscopy (Supplementary Material, Movie S1). The mitochondrial movement in axons was recorded as described in the method section and can be transferred to the streamlined image with NIH Fiji software as shown in Fig. 3B. The numbers of stationary and moving (anterograde or retrograde or both) mitochondria were counted by the auto-computer program. Comparing with vector control cells, expression of WT and mutant TMEM230 had a slightly increased trend in stationary mitochondria (Fig. 3C) and a slightly decreased trend in anterograde moving mitochondria (Fig. 3D). There were less moving mitochondria at retrograde directions in neurons expressing mutant TMEM230 than those of WT-TMEM230 or vector (Fig. 3E).
TMEM230 had no significant effect on stationary mitochondria. Mouse primary hippocampus neurons were transfected with GFP, Mito-DsRed and TMEM230 variants at a ratio of 1:10:10 for 12 h. The mid-segments of axons with both GFP and Mito-DsRed were recorded using living-time lapse imaging under fluorescent microcopy. GFP (green) represented the tracking direction of an axon, and Mito-DsRed (red) labeling mitochondria within axons. A. A representative living imaging of one mid-segment of a recoding axon with both GFP and Mito-DsRed florescence in vector group. The representative movie of Mito-DsRed-mitochondria recoding in 3 min time periods, with the 15-s playing time (Supplementary Material, Movie S1). Scale bar, 10 μm. B. Representative streamline pictures of mitochondrial movement trajectory in 1 min with 10 s intervals. C. The number of stationary mitochondria in each group was counted. There was a slightly increased trend but no statistical difference between the groups. D. Anterograde moving mitochondria. E. Retrograde moving mitochondria. *P < 0.05 by ANOVA, vs cells with vector.
To further monitor the moving mitochondria profiles at anterograde and/or retrograde directions, we used the NIH Fiji software and custom MATHLAB program that was developed in our lab to quantify the distance and speed of moving mitochondria in both directions. Figure 4A showed the representative kymographs and trajectory runs of Mito-DsRed-labeled mitochondrial in one mid-axon segment in each experimental group. There were over 10 axons from each group and three mid-axon segments in each axon that were included in the quantification of moving speed and distance. Interestingly, expression of mutant TMEM230 dramatically reduced the speed and distance of mitochondria moving at retrograde direction compared with those of the vector cells and WT-TMEM230 cells (Fig. 4B and C). WT-TMEM230 only had a slightly decreasing trend (but no statistical difference) in reducing speed and distance of mitochondria moving at the retrograde direction. In contrast, expression of both WT and mutant TMEM230 only slightly decreased (but no statistical significance) anterograde moving in speed and distance (Fig. 4D and E). These results indicated that mutant-TMEM230-induced mitochondrial transport impairment in the axon, which could be the early event contributing to neural injury and neurodegeneration.
Mutant TMEM230 impaired mitochondrial transport at retrograde direction. Mouse primary hippocampus neurons were transfected with GFP, Mito-DsRed and TMEM230 variants at a ratio of 1:10:10 for 12 h. The axons at middle region with both GFP and Mito-DsRed were recorded using living-time lapse imaging under florescent microcopy. A. Representative images of mitochondrial kymographs trajectory runs from cells expressing vector, WT or mutant TMEM230. Top panel: Representative kymographs of Mito-DsRed-labeled mitochondrial motility. Bottom panel: Representative trajectory runs of Mito-DsRed-labeled mitochondria in top panel. Red: anterograde, Green: retrograde, Blue: stationary. Horizontal bar, 10 μm. Vertical bar, 30 s. The scale bar is the same in each image. B. Retrograde moving speed. C. Retrograde moving distance. D. Anterograde moving speed. E. Anterograde moving distance. *P < 0.05 by ANOVA, vs cells with vector.
Knockdown of TMEM230 induced mitochondrial retrograde transport impairment
Knockdown of TMEM230 by siRNA impaired mitochondria transport in axons at retrograde direction compared with control neurons (Fig. 5). Both speed and distance of retrograde mitochondrial transport were impaired by knockdown of TMEM230. There was no statistical significance to the change of mitochondria transport in anterograde direction by knockdown of TMEM230. Moreover, expression of WT-TMEM230 rescued the knockdown-induced mitochondria transport impairment in the retrograde direction (Fig. 5). These findings suggest that normal levels of TMEM230 proteins play a critical role in axon transport.
Knockdown of endogenous TMEM230-induced mitochondrial transport impairment. Mouse primary hippocampus neurons were transfected with cRNA and TMEM230-siRNA for 24 h, and followed by transfection of GFP, Mito-DsRed and various TMEM230 constructs at a ratio of 1:10:10 for 12 h at indicated. The axonal transport of mitochondria was recoded as described in method section. A. Retrograde moving speed. B. Retrograde moving distance. *P < 0.05 by ANOVA, vs cells with control RNA and vector, # P < 0.05 by ANOVA, vs cells with TMEM230-siRNA and vector.
Discussion
The main findings of this study are that overexpression of WT and mutant TMEM230 or knockdown of endogenous TMEM230-induced neurodegeneration and impaired mitochondria transport at the retrograde direction in axons. Mutant TMEM230 caused more severe neurotoxicity and mitochondrial transport impairment than WT-TMEM230. Our results demonstrate that maintaining steady TMEM230 protein levels is critical for neuron survival and axon transport. These findings suggest that mutant-TMEM230-induced mitochondrial transport impairment could be the early event leading to neurite injury and neurodegeneration in PD development.
One of the key pathological features of PD is neuron injury and/or neurodegeneration in certain brain regions. Our results showed that overexpression of WT- and PD-linked *184Wext*-TMEM230 decreased SH-SY5Y cell viability and increased neurite injury in mouse neurons. Our result is consistent with previous studies showing that both WT and mutant TMEM230 induces cell toxicity in HEK 293 T and SN4741 cells (6,7). Moreover, knockdown of endogenous TMEM230 also induced neurodegeneration, while expression of WT-TMEM230 rescued the knockdown-induced degeneration. These results indicated that TMEM230 expression is tightly controlled in neurons and plays a critical role in neuron survival. These findings suggest that either high level of TMEM230 or mutant TMEM230 in local brain regions may trigger neurodegeneration.
Neuritic injury may be induced by various pathological abnormalities in neurons, such as disruption of mitochondrial energy homeostasis, intracellular transport processes, etc. (11,14–16). Axon injury is one of the early neurodegeneration events occurring before cell body degeneration in various neurodegenerative diseases, including PD (10,11). We found that expression of WT- or mutant TMEM230 significantly induced neurite injury and mitochondria transport impairment in axons at the retrograde direction in mouse primary neurons. Mutant TMEM230 caused severe mitochondria transport impairment than WT-TMEM230 did. We also found that knockdown of TMEM230 impaired mitochondria transport in axons, while expression of WT-TMEM230 rescued this impairment. These findings are consistent with previous studies of other PD-related genetic mutations in neural transport (17–19). For instance, mutant alpha-synuclein induces neural transport abnormalities which have implications in PD pathology (20–22). We previously found that mutant LRRK2 also impairs both mitochondria and lysosome transport thereby leading to neurodegeneration in a PD cell model (12). Taken together, these findings suggest that axonal transport impairment may be a common early event in PD pathology.
Mitochondrial dysfunction is one of the key mechanisms to induce neurodegeneration in PD (23–26). There are about two-thirds of axonal mitochondria in a stationary phase under normal physiological conditions, whereas the other third are moving mitochondria with either anterograde or retrograde moving (10,14,15). Polarized microtubes in neurons with orientation allow cargoes transport from soma to distal end (anterograde) or backwards (retrograde) (27). Anterograde transport of mitochondria can replenish the newly generated mitochondria from soma to axons. Retrograde transport of mitochondria can transport damaged or senescent mitochondria to the soma for repair or degradation (28). Stationary mitochondria stay at the local compartment of axon to support energy (29). Our results demonstrated that expression of TMEM230 (WT or mutant) only slightly increased (but with no statistical difference) the number of stationary mitochondria, but significantly decreased the number of the retrograde moving mitochondria in the axons. Moreover, overexpression TMEM230 (mutant) or knockdown of TMEM230 reduced the distance and speed of mitochondria being transported in retrograde directions but not in the anterograde direction. Expression of WT-TMEM230 rescued the knockdown-induced impairment of mitochondrial transport. This impairment of the axonal transport of mitochondria at retrograde direction could disrupt the repair and degradation pathway in maintaining normal neuron survival, though this needs further investigation. Nevertheless, axonal mitochondria-transport impairment could be one of the early events for TMEM230-inducing mitochondrial dysfunctions and thereby inducing neuritic injury and neurodegeneration.
In conclusion, our findings demonstrated that knockdown or overexpression of TMEM230 (WT or mutant) induced the impairment of axonal mitochondria transport with a corresponding neurite injury and neurodegeneration. These findings provide novel insight into the roles of TMEM230 in the pathogenesis of PD and other related disorders.
Materials and Methods
Reagents
Reagents were used in this study including Opti-MEM I medium (Gibco), 10% heat-inactive FBS (Fetal bovine serum), 1% penicillin–streptomycin, Poly-L-lysine (Gibco), B27 supplement (Gibco), 2 mM Glutamax (Gibco), LipofectAMINE PLUS Reagent (Invitrogen), 35 mm dishes cover-glass on the bottom (MatTek), Lipofectamine2000 (Invitrogen), low-fluorescence nutrient media (Hibernate E, Brain Bits), anti-TMEM230 Rabbit antibody (1:1000, Sigma, HPA009078) and ECL goat anti-rabbit IgG (GE Healthcare, NA934V) or goat anti-mouse IgG (GE Healthcare, NA931V).
Cell culture, Plasmids and transfection
Human neuroblastoma SH-SY5Y cells were grown in Opti-MEM I (Gibco) medium with 10% heat-inactive FBS and 1% penicillin–streptomycin (12). TMEM230 (WT and *184Wext*5) cDNA constructs with C-myc-tag were used as described previously (7). LipofectAMINE PLUS Reagent was used for SH-SY5Y cells transfection according to the protocol from Invitrogen.
Cell viability assay
SH-SY5Y cells were co-transfected with GFP and various pCMV-TMEM230 plasmids at a ratio of 1:10 for 48 h in Opti-MEM I media with 10% FBS and antibiotics as described previously (12). The Opti-MEM I media contain growth factors and supplements and promote the growth of neurite-like processes, which represents some neuronal features. Thus, they have low transfection efficiency for cDNA constructs (15–20%). Under these transfection conditions, above 90% of GFP positive neurons also expressed TMEM230 variants confirmed by co-immunostaining with antibodies against TMEM230 and GFP. Cell viability was measured by counting the healthy viable GFP positive cells that contained at least one smooth extension (neurite) that was twice the length of the cell body from 20 randomly selected fields under fluorescence microscopy as described previously (30). DAPI nuclei staining was used to detect apoptotic cells with fragmented and/or highly condensed nuclei under fluorescent microscopy (7). The counting of the apoptotic cells was performed by an auto-computer program. The experiments were repeated three times in duplicate.
Mouse primary neurons culture
Mouse primary cortical or hippocampus neurons were derived from embryonic day 17 as described previously (7). Neurons were cultured on 35 mm dishes with cover-glass on the bottom (MatTek) that were coated with Poly-L-lysine (Gibco). The neurobasal media were used containing 2% B27 (Gibco) supplement, 2 mM Glutamax (Gibco) and 100 U/ml penicillin–streptomycin. At day 6 in vitro (DIV), neurons were co-transfected with various TMEM230 plasmids and GFP in a ratio of 10:1 for 12 to 48 h using Lipofectamine 2000 (Invitrogen) with manufactural protocol. There were about 90% of GFP positive cells expressing TMEM230 variants confirmed by co-immunostaining using anti-GFP and anti-TMEM230 antibodies. GFP positive neuron viability assays were performed and calculated as the same methods as we did in SH-SY5Y cells, described above. Neuronal injury assays were performed at 48 h post transfection as described previously (7,31). Living hippocampus neurons were imaged at end of experiments under fluorescent microscopy. Neurons with injury neurites were blindly quantified using NIH ImageJ software. Injured GFP positive neurons were defined as they displayed loss of neuronal projections, disappearance of neurites, presence of soma fragmentation, neurites with beading morphology or soma shape change from oval to completely circular. Cortical neuronal death was measured by nuclear condensation assay, as described previously (31). After 48-h transfection, neurons were fixed for 30 min with 4% PFA and the nuclei were labeled using DAPI. Image acquisition was done automatically on an Axiovert 200 (Zeiss) and the intensity of every nucleus was measured using Volocity (Perkin Elmer). Results are presented in the percentage of neuronal death. Data analysis was performed using a one-way or two-way analysis of variance (ANOVA) followed by post hoc analysis.
TMEM230 siRNA knockdown
Human TMEM230 siRNA (D-001810-10-05) or mouse TMEM230 siRNA (L-044978-01-0005) and non-targeting control RNA (cRNA) (J-015123-20-0002) were ordered from Dharmacon and transfected to SH-SY5Y cells or mouse primary neurons one day before TMEM230 constructs transfection using Lipofectamine2000 (Invitrogen) with manufactural protocol. Cell viability and neuron injury assays were performed at 72 h post transfection of TMEM230-siRNA transfection. Knockdown of TMEM230 was confirmed using western blot analysis with the cell lysates from 2–4 days post siRNA transfection.
Western blot analysis
Cell lysates were diluted in protein Loading buffer (Invitrogen) with 3% βME and denatured in 85°C for 5 min. Samples were resolved using 4–12% NuPAGE Bis-Tris gels and transferred to PVDF membrane. The primary antibodies against TMEM230 and actin were used and followed by HRP-conjugated secondary antibodies (goat anti-Rabbit IgG, or goat anti-Mouse IgG). Protein bands were detected by ECL enhanced chemiluminescence reagents (PerkinElmer) as described previously (31,32).
Live-cell imaging and axonal transport analysis
Measure of mitochondria transport in axons was performed as described previously with slightly modification (33,34). Mouse primary hippocampus neurons at 6 DIV (days in vitro) were co-transfected TMEM230, Mito-DsRed and GFP at 10:10:1 ratio for 12 h using Lipofectamine2000 according to supplier’s protocol. For knockdown experiment, TMEM230 siRNA was transfected one day prior to transfection of various TMEM230 constructs. Neurons were imaged at 12 h post TMEM230 plasmid transfection in the low-fluorescence nutrient media (Hibernate E, Brain Bits). Live-cell imaging was performed under enhanced Marianis/Yokogawa 3i spinning-disk confocal microscopy with 100× oil objective lens. The segments of mid-axon from each GFP/DsRed-positive neuron were recorded at 1 s interval for 3 min. The segment of mid-axon was defined as over 30 mm from both ends of an axon. Three segments of mid-axon were recorded in each neuron and at least 10 neurons from each group were recorded.
Imaging analysis was performed by NIH Fiji, and custom program with MATHLAB from our group as described previously with modification (35). Streamline images and Kymographs were generated using the Multiple Kymograph plugin for Fiji (NIH) and analyzed using custom MATLAB software. A punctum (mitochondria) was counted as anterograde or retrograde if it traveled a net distance of 5 μm in the appropriate direction over the imaging period; puncta that were nonmotile or traveled a net distance less than 5 μm were counted as stationary. The axon path and direction (close to cell body part is define as proximal end) were traced with computer program in the Kymographs. The x-axis and y-axis were defined by the particle (mitochondria) location and time points, respectively. The x-axis was corresponding to the axon direction from proximal to distal. The average speed, average run lengths, cumulative distance and cumulative time of moving mitochondria were analyzed.
Statistical analysis
All experiments were repeated at least three times. Data are the mean ± standard error of the mean (SEM). Mitochondrial moving distance and speed was analyzed by Fiji with preinstalled Kymograph plugins from NIH. Data were analyzed by one-way or two-way ANOVA tests using the GraphPad Prism 5 software. A P < 0.05 was considered significant.
Acknowledgement
We would like to express our sincere thanks to Dr Erika L.F. Holzbaur’s group for providing the training on axonal transport techniques and Dr Hanxiang Deng’s helpful discussion during the study.
Conflict of Interest statement. All authors declare no conflict of interest.
Author contributions
Conception and design of the study: X.W., W.W.S.; Acquisition, analysis and interpretation of data: X. W., G. G., J. Z, Z. L., C. F. L, C. A. R., W. W. S. Drafting the manuscript: X.W., W.W.S.; Statistical analysis: X.W., W.W.S.
