Fibroblasts were obtained under an MTA from the National Institute of Neurological Disorders and Stroke (NINDS) human and cell repository or the Parkinson’s Progression Markers Initiative (PPMI), which is in a partnership with multiple institutions that approved study protocols, ensured consent from donors, and deposited fibroblasts. All available lines were acquired from NINDS at the time of purchase. Fibroblast and HEK cell culture, immunoprecipitation (IP), and mitochondrial purification were described in Hsieh et al. (2016). Briefly, CCCP in DMSO or the same volume of DMSO treated fibroblasts were lifted by a cell scraper, and mechanically homogenized with a Dounce homogenizer in 750 μl isolation buffer (200 mM sucrose, 10 mM TRIS/MOPS, pH 7.4). After centrifugation at 500 g for 10 min, crude supernatant was spun at 10,000 g for 10 min to pellet intact mitochondria. Mitochondrial pellet was washed twice with isolation buffer. After this step, supernatant was named “cytosolic fraction (Cyto),” and pellet was resuspended in 50 μl lysis buffer (50 mM Tris pH 8.0, 150 mM NaCl, and 1% Triton X-100–T8787, Sigma-Aldrich) with 0.25 mM phenylmethanesulfonyl fluoride (P7626, Sigma-Aldrich) and protease inhibitors (Roche) named “mitochondrial fraction (Mito).” Samples were mixed 1:1 with 2 × laemmli buffer (4% SDS, 20% Glycerol, 120 mM Tris–HCl, 0.02% bromophenol blue, 700 mM 2-mercaptoethanol) and boiled for 5 min prior to being loaded (Mito:Cyto = 25:1) into an SDS-PAGE. pRK5-EGFP-tau (#46904, addgene), pRK5-EGFP-tau-P301L, pRK5-EGFP-tau-N279K, and pRK5-EGFP-tau-R406W (the latter three were custom-made by Synbio Technology) were used for transfection by the calcium phosphate transfection protocol (Wang and Schwarz, 2009). Transferred membranes were first blocked overnight in phosphate-buffered saline (PBS) containing 5% fat-free milk and 0.1% tween-20 at 4°C, and then incubated with the following primary antibodies: mouse anti-Miro1 (WH0055288M1, Sigma-Aldrich) at 1:1,000, rabbit anti-Miro1 (HPA010687, Sigma-Aldrich) at 1:1,000, rabbit anti-VDAC (4661S, Cell Signaling Technology) at 1:1,000, mouse anti-Mitofusin2 (H00009927-M01, Abnova) at 1:1,000, mouse anti-Parkin (sc32282, Santa Cruz Biotechnology) at 1:500, rabbit anti-LRRK2 (NB300-268, Novus Biologicals) at 1:500, rabbit anti-GAPDH (5174S, Cell Signaling Technology) at 1:3,000, rabbit anti-β-actin (4967S, Cell Signaling Technology) at 1:1,000, rabbit anti-GFP (A11122, Invitrogen) at 1:750, mouse anti-ATP5β (ab14730, Abcam) at 1:100, rabbit anti-Calreticulin (2891, Cell Signaling Technology) at 1:1,000, mouse anti-Golgi (NB600-412, Novus Biologicals) at 1:100, or mouse anti-Myc (9E10/sc-40, Santa Cruz Biotechnology) at 1:100, at 4°C overnight in blocking buffer. HRP-conjugated goat anti-mouse or rabbit IgG (Jackson ImmunoResearch Laboratories) were used at 1:5,000–10,000. West Dura ECL Reagents (34075, GE Healthcare) were used for ECL immunoblotting. Membranes were exposed to UltraCruz autoradiography films (Santa Cruz Biotechnology) and developed on a Konica Minolta SRX-101A developer or scanned using a Bio-Rad ChemiDoc XRS system.

All experiments were performed in a blinded format, and the identities of the lines were un-blinded after the experiments. For Figure 1C, the intensities of protein bands were measured by ImageJ (ver. 1.48V, NIH). The intensity of each band in the mitochondrial fraction was normalized to that of the mitochondrial loading control VDAC from the same blot, and expressed as a fraction of Mean of Healthy-1 with DMSO treatment (Hsieh et al., 2016); this control was included in every independent experiment. The band intensities of VDAC were not significantly different among all fibroblast lines and conditions (P > 0.8, One-Way ANOVA Post Hoc Tukey Test).

All experiments were performed as blinded tests. A 40 μM CCCP in DMSO or the same volume of DMSO alone was applied to fibroblasts for 6 h, and then cells were lysed in lysis buffer (100 mM Tris, 150 mM NaCl, 1 mM EGTA, 1 mM EDTA, 1% Triton X-100, 0.5% Sodium deoxycholate) with protease inhibitor cocktail (539134, Calbiochem). Cell debris was removed by centrifugation at 17,000 g for 10 min at 4°C. Details of ELISA are in Supplementary Methods.

Cells were grown on ∼10 mm Aclar discs in 37°C incubator with 5% CO₂ till 95–100% confluent. Each Aclar disc was then transferred separately into an Eppendorf tube with 1 ml of fixative solution (2% glutaraldehyde and 4% paraformaldehyde in 0.1 M sodium cacodylate buffer) and incubated for an hour. These cells were kept on ice until processed for imaging. Fixative was removed and 1% OsO₄ in ddH₂O was added. Samples were then gently shaken and incubated for 1 h at 4°C. Incubated samples were washed three times with cold ddH₂O for 5 min each. Samples were then stained with 1% uranyl acetate for 2 h at 4°C. Stained samples were dehydrated, first with 50% EtOH for 10 min, next 70% EtOH for 10 min, and lastly 95% EtOH for 10 min. Samples were allowed to warm up to room temperature before two more dehydration procedures with 100% EtOH for 10 min each. The final dehydration was performed with acetonitrile for 15 min. The dehydration mixture was then replaced with Embed 812 medium (44% Embed 812, 35% DDSA, 18% NMA, 3% BDMA) and acetonitrile (1:1 ratio) for 1 h. The ratio was then increased to 2:1 (Embed 812 medium to acetonitrile) for overnight incubation. Next, the mixture was replaced with 100% Embed 812 medium and incubated for 2 h before samples were placed in molds for overnight settling. Settled samples were polymerized in a 65°C oven for 24 h before being sectioned for imaging. Imaging was performed using a Jeol TEM 1400 microscope, and images were acquired at 5,000×. Images were saved as.dc3 files. Analysis of.dc3 images was performed using ImageJ.

Throughout the paper, the distribution of data points was expressed as box-whisker, or dot-plot with Mean ± SEM, except otherwise stated. Box center line is median and box limits are upper and lower quartiles. One-Way or Two-Way ANOVA Post Hoc Tukey Test was performed for comparing multiple groups. Mann-Whitney U or T Test was performed for comparing two groups. Chi-Square Test was performed for Figure 3B because the data was categorical. Statistical analyses were performed using the Prism software (ver. 8.01, GraphPad) or Excel (ver. 16.51). The number of experimental replications (n) can be found in Figure Legends. No statistical methods were used to predetermine sample sizes, but the number of experiments and biological replicates was chosen based on the nature of the experiments (it is usually difficult to assess an outcome that follows a normal distribution in our experiments), degree of variations, and published papers describing similar experiments. We did not exclude any data. ^∗P < 0.05, ^∗∗P < 0.01, ^∗∗∗P < 0.001, for all Figures.

We have identified two parallel molecular pathways essential for removing Miro from the OMM of depolarized mitochondria–LRRK2 and the PINK1-Parkin axis (Wang et al., 2011; Hsieh et al., 2016, 2019). In addition, the OMM protein Mitofusin2 is a target of the PINK1-Parkin pathway, but not of LRRK2, for depolarization-triggered degradation (Poole et al., 2010; Tanaka et al., 2010; Hsieh et al., 2016). We have previously found that in 12 healthy control fibroblast lines, CCCP treatment, which depolarizes the mitochondrial membrane potential (ΔΨm), for only 1 h triggers the recruitment of a small fraction of cytosolic LRRK2 and Parkin to mitochondria, prior to Miro1 and Mitofusin2 removal at 6 h detected by Western blotting (Figures 1A–C; Supplementary Figure 1A; Hsieh et al., 2016, 2019). Antibodies against LRRK2, Parkin, and Miro1 have been validated in human cells lacking the corresponding genes (Hsieh et al., 2016). We then examined whether LRRK2 and Parkin recruitment to damaged mitochondria was impaired in 7 fibroblast lines of FTLD and parkinsonism patients or at-risk individuals with MAPT mutations, obtained from the NINDS human and cell repository (Figure 1B; Supplementary Table 1). At-risk subjects are younger asymptomatic family members of probands and carry the same genetic mutations. In contrast to healthy controls, 4 MAPT lines failed to significantly recruit LRRK2 to mitochondria, and 2 lines failed to recruit Parkin (Figures 1A–C). One line (MAPT-7, R406W) failed to recruit both proteins (Figures 1A–C). Basal levels of LRRK2 and Parkin were comparable among all lines (P > 0.8). Collectively, our results provide evidence indicating that the LRRK2 and PINK1-Parkin pathways are affected in selective MAPT patients’ fibroblast lines.

Because the failure to relocate LRRK2 or Parkin to damaged mitochondria disrupts the following removal of Miro1 or Mitofusin2 (Hsieh et al., 2016), we next examined Miro1 and Mitofusin2 protein levels in the mitochondrial fractions of these cells upon depolarization. In healthy controls at 6 h following CCCP treatment, both Miro1 and Mitofusin2 are removed from damaged mitochondria detected by Western blotting (Figures 1A–C); this time point is earlier than the completion of mitophagy when multiple mitochondrial markers are degraded (Figure 1A; Hsieh et al., 2016, 2019). Notably, we discovered a unifying impairment in removing Miro1 from the mitochondrial fractions at 6 h after CCCP treatment in all 7 MAPT lines (Figures 1A–C). By contrast, Mitofusin2 was significantly degraded in the mitochondrial fractions after CCCP treatment in all MAPT lines, just as in control cells (Figures 1A–C). Basal protein levels of Miro1 and Mitofusin2 were largely comparable among all lines (Figure 1C; P > 0.09). For all experiments, the cell passaging numbers were within the range of 5–19 which had no influence on the phenotypes (Hsieh et al., 2019). We next validated the result from Western blotting by an alternative assay: ELISA (Hsieh et al., 2019; Nguyen et al., 2021). We tested 3 MAPT lines and consistently observed the resistance of Miro1 to degradation upon CCCP treatment (Figures 1D–G; Hsieh et al., 2019). Taken together, our results show that the failure to remove Miro1 from damaged mitochondria is a common molecular defect associated with pathogenic MAPT mutations.

In order to examine mitochondrial changes at the ultrastructural level, we performed TEM on a healthy and a MAPT fibroblast line used in Figure 1, with or without CCCP treatment for 6 h (Figure 2A). We found that the mitochondrial size, perimeter, and crista junction number were indistinguishable among all conditions (Figures 2B–D), whereas the aspect ratio (minor to major diameter) was increased in healthy control with CCCP treatment (Figure 2E), indicating mitochondrial rounding in this condition. Strikingly, we found that following CCCP treatment the ER-mitochondrial contact number, particularly for rough ER (RER), was significantly reduced in healthy fibroblasts; however, it was not altered at the same time point in MAPT fibroblasts (Figures 2A,F–I). This result suggests that mitochondrial depolarization causes the separation of mitochondria and ER contact, consistent with a previous report (McLelland et al., 2018). This study also implicates that the mitochondrial and ER separation is essential for the following mitophagy, and blocking their dissociation, as seen in MAPT fibroblasts (Figure 2), could lead to mitophagy impairment (McLelland et al., 2018), observed in other tau models as well (Hu et al., 2016; Cummins et al., 2019; Fang et al., 2019). In addition, we saw mitochondria frequently juxtaposed with cytoskeleton (actin or microtubule) tracks in healthy control fibroblasts and this association was significantly reduced following CCCP treatment (Figures 3A,B), showing that damaged mitochondria are sequestered from cytoskeleton networks. However, MAPT fibroblasts failed to exhibit this response (Figures 3A,B). Furthermore, MAPT fibroblasts displayed the widespread presence of structures resembling lamellar bodies (LB) and multivesicular bodies (MV) under both basal and depolarized conditions (Figure 3C), indicating possible imbalance in proteostasis and lipid homeostasis. The global mitochondrial network or ΔΨm visualized by TMRM staining was indistinguishable among healthy and MAPT fibroblasts at baseline (Figure 4A). Together, our results show that the dissociation of damaged mitochondria from ER and cytoskeleton networks is blocked in fibroblasts of a MAPT patient.

One hypothesis that could explain our TEM observations is that normal MAPT is essential to facilitate the separation of mitochondria from ER and cytoskeleton at the early stage of mitophagy. Interestingly, Miro can localize to the ER-mitochondrial contact sites (Kornmann et al., 2011; Stroud et al., 2011) and anchors mitochondria to microtubule and actin tracks (Stowers et al., 2002; Glater et al., 2006; Wang and Schwarz, 2009; Morlino et al., 2014; Lopez-Domenech et al., 2018; Oeding et al., 2018). Notably, Miro is quickly removed from depolarized mitochondria (Chan et al., 2011; Wang et al., 2011; Birsa et al., 2014), and our earlier results showed a unifying impairment in removing Miro1 from damaged mitochondria in all 7 MAPT lines (Figure 1), suggesting that Miro1 and tau may coordinate to ensure an efficient mitophagy process. We next determined whether tau could physically interact with Miro1. By performing co-immunoprecipitation (co-IP) from HEK cells transfected with tau and Miro1 constructs, we found that wild-type tau, but not tau-P301L or tau-R406W, interacted with Miro1 (Figures 4B,C; Supplementary Figure 1B). Tau-N279K protein was not well-expressed in HEK cells (Figure 4D) though the same amount of DNA was transfected as tau-P301L or tau-R406W, which did not allow reliable co-IP experiments. Therefore, mutant tau compromises its ability to bind to Miro1.

The failure to remove Miro1 from damaged mitochondria in fibroblasts of MAPT patients (Figure 1) is reminiscent of that observed in fibroblasts of PD patients (Hsieh et al., 2019). We have previously discovered a small molecule (named Miro1 Reducer or MR3) that binds to and destabilizes human Miro1 protein (Hsieh et al., 2019; Li et al., 2021). Treating fibroblasts of PD patients with Miro1 Reducer rescues the phenotype of Miro1 accumulation on damaged mitochondria, and applying Miro1 Reducer to human neuron and fly models of PD ameliorates Parkinson’s relevant phenotypes, without affecting Miro1’s overall GTPase activity or other mitochondrial proteins including Miro2, Mitofusin, OPA1, VDAC, and ATP5β (Hsieh et al., 2019). We examined whether Miro1 Reducer could also promote Miro1 degradation in MAPT fibroblasts. We administered Miro1 Reducer to MAPT-1 line which exhibited the phenotype of Miro1’s resistance to degradation upon CCCP treatment, shown earlier (Figure 1). We found that Miro1 Reducer treatment caused significant Miro1 degradation following CCCP treatment detected by ELISA (Figure 4E), just like the response of Miro1 in healthy control fibroblasts (Figure 1). These results suggest that pharmacologically targeting Miro1 could be tested as a therapeutic approach in tauopathy models.