In Drosophila, mitochondrial transport is mediated by milton and Miro, which attach mitochondria to microtubules via kinesin heavy chain (Guo et al., 2005; Glater et al., 2006; Figure 1A). It has been reported that expression of milton RNAi in neurons via pan-neuronal elav-GAL4 driver reduced milton protein levels in Drosophila head lysate to 40% and mito-GFP signals in axons to 50% (Iijima-Ando et al., 2012; Iijima-Ando et al., 2009).

To test how loss of axonal mitochondria affects proteostasis in neurons, we first examined the accumulation of ubiquitinated proteins. At 14 days old, more ubiquitinated proteins were deposited in the brains of milton knockdown flies than in those of age-matched control flies (Figure 1B, p<0.005 between control RNAi and milton RNAi). There was no significant increase in ubiquitinated proteins in milton knockdown flies at 1 day old, suggesting that the accumulation of ubiquitinated proteins caused by milton knockdown is age-dependent (Figure 1—figure supplement 1). We also analyzed the effect of the neuronal knockdown of Miro, a partner of milton, on the accumulation of ubiquitin-positive proteins. Since severe knockdown of Miro in neurons causes lethality, we used UAS-Miro RNAi strain with low knockdown efficiency, whose expression driven by elav-GAL4 caused 30% reduction of Miro mRNA in head extract (Iijima-Ando et al., 2012). Although there was a tendency for increased ubiquitin-positive puncta in Miro knockdown brains, the difference was not significant (Figure 1B, p>0.05 between control RNAi and Miro RNAi). These data suggest that the depletion of axonal mitochondria induced by milton knockdown leads to the accumulation of ubiquitinated proteins before neurodegeneration occurs.

It has been reported that ubiquitinated proteins accumulate with aging (Tonoki et al., 2009); thus, we analyzed the accumulation of ubiquitinated proteins in aged brains (30-day-old) with milton knockdown. The number of puncta of ubiquitinated proteins did not significantly differ between control and milton knockdown flies or between control and Miro knockdown flies (Figure 1C, p>0.05). These results suggest that depletion of axonal mitochondria may have more impact on proteostasis in young neurons than in old neurons.

We examined the ultrastructure of presynaptic terminals and cell bodies in photoreceptor neurons with milton knockdown by transmission electron microscopy in 27-day-old flies (Figure 1D). As previously reported (Iijima-Ando et al., 2012), the number of mitochondria in presynaptic terminals decreased in milton knockdown (Figure 1E). The swelling of presynaptic terminals, characterized by the enlargement and roundness, was not reported at 3-day-old (Iijima-Ando et al., 2012) but observed at this age with about 4% of total presynaptic terminals (Figure 1F, asterisks).

Some presynaptic terminals of milton knockdown neurons contained dense materials (Figure 1F and G, arrowheads). Dense materials are rarely found in age-matched control neurons, indicating that milton knockdown induces abnormal protein accumulation in the presynaptic terminals (Figure 1G and H). In milton knockdown neurons, dense materials are found in swollen presynaptic terminals more often than in presynaptic terminals without swelling, suggesting a positive correlation between the disruption of proteostasis and axonal damage (Figure 1G). In contrast, dense materials were not observed in cell bodies in the milton knockdown retina (Figure 1H). These results indicate that the depletion of axonal mitochondria induces protein accumulation in the axon.

Since abnormal proteins were accumulated in milton knockdown brains, we next examined if protein degradation pathways were suppressed. We analyzed autophagy via western blotting of the autophagy markers LC3 and p62 (Klionsky et al., 2021). During autophagy progression, LC3 is conjugated with phosphatidylethanolamine to form LC3-II, which localizes to isolation membranes and autophagosomes. LC3-I accumulation occurs when autophagosome formation is impaired, and LC3-II accumulation is associated with lysosomal defects (Klionsky et al., 2021; Bartlett et al., 2011). p62 is an autophagy substrate, and its accumulation suggests autophagic defects (Klionsky et al., 2021; Bartlett et al., 2011). We found that milton knockdown increased LC3-I, and the LC3-II/LC3-I ratio was lower in milton knockdown flies than in control flies at 14-day-old (Figure 2A). We also analyzed p62 levels in head lysates sequentially extracted using detergents with different stringencies (1% Triton X-100 and 2% SDS). Western blotting revealed that p62 levels were increased in the brains of 14-day-old milton knockdown flies (Figure 2B). The increase in the p62 level was significant in the Triton X-100-soluble fraction but not in the SDS-soluble fraction (Figure 2B), suggesting that depletion of axonal mitochondria impairs the degradation of less-aggregated proteins. Proteasome activity was also significantly decreased in brains with neuronal knockdown of milton (Figure 2C, p<0.005).

At 30 days old, LC3-I was still higher, and the LC3-II/LC3-I ratio was lower, in milton knockdown compared to the control (Figure 2D). At this age, milton knockdown increased p62 significantly in 1% Triton X-100 fraction and 2% SDS fraction (Figure 2E). Proteasome activities were also decreased in milton knockdown flies at 30-day-old (Figure 2F). These results indicate that depletion of axonal mitochondria impairs protein degradation pathways.

milton knockdown downregulates ATP in the axon (Oka et al., 2021). To examine whether the disruption of protein degradation pathways by milton knockdown is due to ATP deprivation, we investigated the effects of knocking down phosphofructokinase (Pfk), a rate-limiting enzyme in glycolysis, on protein degradation pathways. Neuronal knockdown of Pfk was reported to lower ATP levels in brain neurons (Oka et al., 2021). Pfk knockdown and milton knockdown decreased ATP to similar levels (Figure 3A–C). However, in contrast with milton knockdown, Pfk knockdown did not affect the levels of LC3-I, LC3-II, or the LC3-II/LC3-I ratio (Figure 3D). Pfk knockdown decreased p62 level (Figure 3E), suggesting that autophagy is promoted. On the other hand, proteasome activity was decreased by Pfk knockdown (Figure 3F). These results suggest that the downregulation of axonal ATP upon depletion of axonal mitochondria decreases proteasome activity, but not autophagy.

To identify the pathways that mediate the decrease in autophagy in milton knockdown brains, we performed proteome analysis to systematically detect differentially expressed proteins upon neuronal knockdown of milton. We analyzed flies at 7- and 21-day-old, the age before autophagic defects are detected and the age just before the onset of neurodegeneration, respectively (Figure 4A). 1039 proteins were detected by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Expression of 36 proteins was significantly increased (22 proteins) or decreased (14 proteins) by milton knockdown at 7-day-old (Figure 4B, Table 1 and Supplementary file 1). At 21 days old, the expression of 41 proteins (31 upregulated and ten downregulated proteins) was significantly altered in milton knockdown flies compared with control flies (Figure 4C, Table 1 and Supplementary file 1).

The ‘Interaction search’ algorithm using KeyMolnet showed that proteins whose expression was significantly altered in the brains of milton knockdown flies at both 7- and 21-day-old were closely associated with the autophagic pathway (Table 2). Proteins involved in pathways characteristics of aging, such as the immune response (transcriptional regulation by STAT), cancer (transcriptional regulation by SMAD, transcriptional regulation by myc), longevity (transcriptional regulation by FOXO, Sirtuin signaling pathway), and stress responses (HSP90 signaling pathway, MAPK signaling pathway; Zia et al., 2021; Haigis and Yankner, 2010), were enriched in the proteome profiles of milton knockdown flies compared with those of control flies at 7-day-old (Table 2), suggesting that depletion of axonal mitochondria accelerates aging in the brain.

Differentially expressed proteins at 7-day-old flies may reflect alterations that are causal for autophagic defects. We noticed that the expression level of eIF2β was 2.465-fold higher in the brains of milton knockdown flies than in those of control flies (Figure 4B and D). Upregulation of eIF2β in the brains of milton knockdown flies was confirmed by western blotting. milton knockdown increased eIF2β protein levels more than twice (Figure 4E), but did not change the level of eIF2β mRNA (Figure 4F).

We also investigated age-dependent changes in eIF2β by western blotting of control flies at 7-, 21-, 35-, 49-, and 63-day-old. eIF2β levels increased during aging until 49-day-old (Figure 4G). These results suggest that upregulation of eIF2β in milton knockdown fly brain reflects early an onset of age-dependent increase of eIF2β levels.

eIF2β is a subunit of the eukaryotic initiation factor 2 (eIF2) complex, which is critical for translation initiation and the integrated stress response (ISR; Kimball, 1999). eIF2 is a heterotrimer of α, β, and γ subunits, and eIF2α is phosphorylated during the ISR (Pakos-Zebrucka et al., 2016). As for the other subunits of the eIF2 complex, proteome analysis did not detect a significant difference in the protein levels of eIF2α and eIF2γ between milton knockdown and control flies at 7- and 21-day-old (Figure 4D). Western blotting of brain lysates showed that milton knockdown reduced eIF2α levels (Figure 5A), while p-eIF2α levels were not significantly affected (Figure 5B).

To analyze local changes of eIF2α and p-eIF2α, we carried out immunostaining. We focused on the mushroom body, where axons, dendrites, and cell bodies can be easily identified (Figure 5C). Both eIF2α and p-eIF2α were downregulated in the cell body (Kenyon cells) and dendritic (Calyxes) regions of the brains of milton knockdown flies (Figure 5D). In axons (lobe tips), milton knockdown did not affect eIF2α (Figure 5E, p=0.271) but significantly downregulated p-eIF2α (Figure 5E). The ratio of p-eIF2α to eIF2α was lower in the axon but not in the soma or dendritic region. These results suggest that axonal distribution of mitochondria regulates the level of overall eIF2α protein and local p-eIF2α.

Phosphorylation of eIF2α induces conformational changes in the eIF2 complex and inhibits global translation (Wek, 2018). To analyze the effects of milton knockdown on translation, we performed polysome gradient centrifugation to examine the level of ribosome binding to mRNA. Since p-eIF2α was downregulated, we hypothesized that milton knockdown would enhance translation. However, unexpectedly, we found that milton knockdown significantly reduced the level of mRNAs associated with polysomes (Figure 6A and B). We also compared the level of translation between the brains of control and milton knockdown flies by assessing the incorporation of puromycin (Figure 6C). Puromycin incorporation was lower in the brains of milton knockdown flies than in those of control flies, while it was not statistically significant (Figure 6C, indicated by a bracket). These data suggest that the depletion of axonal mitochondria suppresses global translation.

We were motivated to ask if eIF2β upregulation mediates autophagic defects caused by milton knockdown. If so, neuronal overexpression of eIF2β would also induce autophagy impairment. Neuronal overexpression of eIF2β increased LC3-II, while the LC3-II/LC3-I ratio was not significantly different (Figure 7A and B). Overexpression of eIF2β significantly increased the p62 level in the Triton X-100-soluble fraction (Figure 7C, fourfold vs. control, p<0.005 [1% Triton X-100]) but not in the SDS-soluble fraction (Figure 7C, twofold vs. control, p=0.062 [2% SDS]), as observed in brains of milton knockdown flies (Figure 2B). These data suggest that neuronal overexpression of eIF2β accumulates autophagic substrates.

Since the milton knockdown reduced the p-eIF2α level (Figure 5E), we asked whether an increase in eIF2β affects p-eIF2α. Neuronal overexpression of eIF2β did not affect the eIF2α level but significantly decreased the p-eIF2α level (Figure 7D and E).

Depletion of axonal mitochondria causes age-dependent decline in locomotor function (Iijima-Ando et al., 2012). We found that neuronal overexpression of eIF2β also caused locomotor dysfunction (Figure 7F). Locomotor functions were significantly impaired in those flies at 20 days old and worsened further during aging (Figure 7F, compare 4-, 20-, and 30-day-old). We asked if eIF2β overexpression causes neurodegeneration, as depletion of axonal mitochondria in the photoreceptor neurons causes axon degeneration in an age-dependent manner (Iijima-Ando et al., 2012). eIF2β overexpression in photoreceptor neurons tends to increase neurodegeneration in aged flies, while it was not statistically significant (p>0.05, Figure 7—figure supplement 1).

These data indicate that an increase of eIF2β in neurons phenocopies depletion of axonal mitochondria, including suppression of autophagy and age-dependent locomotor dysfunction, and suggest that increase of eIF2β mediates these phenotypes downstream of loss of axonal mitochondria.

Finally, we investigated whether suppression of eIF2β rescues autophagy impairment and locomotor dysfunction caused by neuronal knockdown of milton. Null mutants and flies with RNAi-mediated knockdown of eIF2β in neurons did not survive. Flies lacking one copy of the eIF2β gene survived without any gross abnormality, and the level of eIF2β mRNA in these flies was about 80% of that in control flies (Figure 8A). eIF2β heterozygosity did not affect the eIF2α and p-eIF2α levels (Figure 8—figure supplement 1A and B).

Neuronal knockdown of milton causes accumulation of autophagic substrate p62 in the Triton X-100-soluble fraction (Figure 2B), and we tested if lowering eIF2β ameliorates it. We found that eIF2β heterozygosity caused a mild increase in LC3-I levels and decreases in LC3-II levels, resulting in a significantly lower LC3-II/LC3-I ratio in milton knockdown flies (Figure 8B). eIF2β heterozygosity decreased the p62 level in the Triton X-100-soluble fraction in the brains of milton knockdown flies (Figure 8C). The p62 level in the SDS-soluble fraction, which is not sensitive to milton knockdown (Figure 2B), was not affected (Figure 8C). These results suggest that suppression of eIF2β ameliorates the impairment of autophagy caused by milton knockdown.

eIF2β heterozygosity also rescued locomotor dysfunction induced by milton knockdown. milton knockdown flies with eIF2β heterozygosity exhibited better locomotor function than milton knockdown alone (Figure 8D). The milton mRNA level was not increased in these flies, indicating that the rescue effect in the eIF2β heterozygous background was not mediated by an increase in the milton mRNA level (Figure 8—figure supplement 1). These data suggest that eIF2β upregulation mediates autophagy impairment and locomotor dysfunction caused by the depletion of axonal mitochondria.

Flies were maintained in standard cornmeal medium (10% glucose, 0.7% agar, 9% cornmeal, 4% yeast extract, 0.3% propionic acid, and 0.1% n-butyl p-hydroxybenzoate) at 25 °C under light–dark cycles of 12:12 hr. The flies were transferred to fresh food vials for every 2–3 days. UAS-milton RNAi (v41508) was from VDRC and outcrossed to [w1118] for five generations in our laboratory. Transgenic fly lines carrying UAS-Miro RNAi and UAS-luciferase RNAi (control for milton RNAi) were reported previously (Iijima-Ando et al., 2012). GMR-gal4, Elav-gal4, UAS-Pfk RNAi (Bloomington stock center #36782), UAS-luciferase RNAi (Bloomington stock center #31603) (control for Pfk RNAi), UAS-GFP (used for control for UAS-eIF2β), and UAS-eIF2β (eIF2β^EY08063, Bloomington stock center #17425) were from the Bloomington stock center. eIF2β loss-of-function strain (PBac{SAstopDsRed} LL07719, DGRC#142114) was from KYOTO Drosophila Stock Center. UAS-mitoGFP was a kind gift from Dr. W. M. Saxton (University of California, Santa Cruz). Fly genotypes used in this study are listed in Supplementary file 2.

Fly brains were dissected in PBS and fixed for 45 min in formaldehyde (4% v/v in PBS) at room temperature. After incubation in PBST containing 0.1% Triton X-100 for 10 min three times, samples were incubated for 1 hr at room temperature in PBST containing 1% normal goat serum (Wako, #143–06561) and then incubated overnight with the primary antibody (anti-ubiquitin antibody Ubi-1 (Thermo Fisher #13–1600) (1:50), anti-eIF2α (abcam #ab26197) (1:50) and anti-p-eIF2α (Cell signaling #3398 S) (1:50)) diluted in 1% NGS/PBST at 4 °C. Samples were then washed for 10 min with PBST including 0.1% Triton X-100 three times and incubated with the secondary antibody overnight at 4 ° C. Brains were mounted in Vectashield (Vectorlab Cat#H-1100) and analyzed under a confocal microscope (Nikon). Quantitative analysis was performed using ImageJ (National Institutes of Health) with maximum projection images derived from Z-stack images acquired with same settings. Puncta were identified with mean intensity and area using ImageJ. For eIF2α and p-eIF2α immunostaining, the mushroom body was detected by mitoGFP expression.

Proboscis was removed from decapitated heads, which were then incubated in primary fixative solution (2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M sodium cacodylate buffer) at R.T. for 2 hr. After washing heads with 3% sucrose in 0.1 M sodium cacodylate buffer, fly heads were post-fixed for 1 hr in secondary fixation (1% osmium tetroxide in 0.1 M sodium cacodylate buffer) on ice. After washing with H₂O, heads were dehydrated with ethanol and infiltrated with propylene oxide and Epon mixture (TAAB and Nissin EM) for 3 hr. After infiltration, specimens were embedded with an Epon mixture at 70 °C for 2–3 days. Thin sections (70 nm) of laminas were collected on copper grids. The sections were stained with 5% uranyl acetate in 50% ethanol and Reynolds' lead citrate solution. Electron micrographs were obtained with a CCD camera mounted on a JEM-1400 plus electron microscope (Jeol Ltd.). Quantitation was performed using ImageJ (National Institutes of Health).

Western blotting was performed as reported previously (Iijima-Ando et al., 2012). Briefly, heads of 10–20 Drosophila were homogenized with SDS-Tris-Glycine sample buffer (0.312 M Tris, 5% SDS, 8% glycerol, 0.0625% BPB, 10% β-mercaptoethanol, 10 μg/mL leupeptin, 0.4 μM Pefabloc, 10 mM β-glycerophosphate, 10 mM NaF) and after boiling at 95 °C for 2 min, it was centrifuged at 13,200 rpm, and the supernatant was used as a sample. For p62 western blot, fly heads were homogenized with 1% PBST and after centrifugation at 13,200 rpm, the supernatant was mixed 1:1 SDS-Tris-Glycine sample buffer, and boiled at 95 °C for 2 min. The pellet was dissolved with 2% SDS in PBS, then centrifuged again at 13,200 rpm. The supernatant was mixed 1:1 SDS-Tris-Glycine sample buffer and then boiled at 95 °C for 2 min. SDS–PAGE for western blotting was performed using 15%(w/v) (LC3), 7.5%(w/v) (p62), 10% (w/v) (eIF2α, β, and p-eIF2α) polyacrylamide gels. After electrophoresis, they were transferred to PVDF membrane (Merck Millipore) using a transfer device (BIO-RAD). After transfer, the membrane was blocked with 5% skim milk/TBST (50 mM Tris (pH 7.5), 0.15 M NaCl, 0.05% Tween20) for 1 hr and incubated with primary antibody listed below overnight at 4 °C. Membranes were rinsed twice with TBST containing 0.65 M NaCl and once with TBST containing 0.15 M NaCl. After incubation with the secondary antibody at room temperature for 1 hr, membranes were rinsed twice with TBST containing 0.65 M NaCl and once with TBST containing 0.15 M NaCl. After incubation with Immobilon Western Chemiluminescent HRP Substrate (Merck Millipore), chemiluminescent signals were detected with Fusion FX (Vilber). Experiments were repeated at least three times with independent cohorts of flies.

Heads from ten flies were homogenized in 150 µl of buffer B (25 mM Tris-HCl [pH 7.5], 2 mM ATP, 5 mM MgCl2, and 1 mM dithiothreitol). Proteasome peptidase activity in the lysates was measured with a synthetic peptide substrate, succinyl-Leu-Leu-Val-Tyr-7-amino-4-methyl-coumarin (Suc-LLVY-AMC; Cayman). Luminescence was measured on a multimode plate reader 2300 Enspire (PerkinElmer). Experiments were repeated at least three times with independent cohorts of flies.

Heads from the 10 flies were homogenized in 50  μl of 6  M guanidine-HCl in extraction buffer (100  mM Tris and 4  mM EDTA, pH 7.8) to inhibit ATPases. Samples were boiled for 5  min and centrifuged. The supernatant was diluted 4% with extraction buffer and mixed with a reaction solution (ATP Determination kit, Invitrogen). Luminescence was measured on a multimode plate reader 2300 Enspire (PerkinElmer). The relative ATP levels were calculated by dividing the luminescence by the total protein concentration, which was determined by the Bradford method. Experiments were repeated at least three times with independent cohorts of flies.

Heads from more than 25 flies were mechanically isolated, and total RNA was extracted using ISOGEN (NipponGene) followed by reverse-transcription using PrimeScript RT reagent kit (Takara). The resulting cDNA was used as a template for PCR with THUNDERBIRD SYBR qPCR mix (TOYOBO) on a Thermal Cycler Dice real-time system TP800 (Takara). Expression of genes of interest was standardized relative to rp49. Relative expression values were determined by the ∆∆CT method. Experiments were repeated three times, and a representative result was shown.

Primers were designed using DRSC FlyPrimerBank (Harvard Medical School). Primer sequences are shown below:

30 heads were homogenized in 150 µl of lysis buffer (25 mM Tris pH 7.5, 50 mM MgCl2, 250 mM NaCl, 1 mM DTT, 0.5 mg/ml cycloheximide, 0.1 mg/ml heparin). The lysates were centrifuged at 13,200 rpm at 4 °C for 5 min, and the supernatant was collected. The samples containing 38 µg of RNA were layered gently on top of a 10–50% w/w sucrose gradient (50 mM Tris pH 7.5, 50 mM MgCl2, 250 mM NaCl, 0.1 mg/ml heparin, 0.5 mg/ml cycloheximide in 5 ml polyallomer tube) and centrifuged at 37,000 rpm at 4 °C for 150 min in a himac CP-NX ultracentrifuge using a P50AT rotor. Samples were fractionated from top to bottom, and absorbance at OD260 nm was analyzed by a Plate reader (EnSpire). Experiments were repeated at least three times with independent cohorts of flies.

13-day-old flies were starved for 6 hr and fed 600 μM puromycin (Sigma) or 600 μM puromycin/35 mM cycloheximide (Sigma) in 5% sucrose solution for 20 hr. Incorporated puromycin was quantified by western blot with anti-puromycin antibody (Enzo # CAC-CAC-PEN-MA001) and normalized with actin. Experiments were repeated at least three times with independent cohorts of flies.

Fly heads were fixed in Bouin’s fixative solution for 48 hr at room temperature, incubated for 24 hr in 50 mM Tris/150 mM NaCl, and embedded in paraffin. Serial sections (7 μm thickness) through the entire heads were stained with hematoxylin and eosin and examined by bright-field microscopy. Images of the sections that include the lamina were captured with Keyence microscope BZ-X700 (Keyence), and the vacuole area was measured using ImageJ (National Institutes of Health).

The climbing assay was performed as previously described (Iijima-Ando et al., 2012). Flies were placed in an empty plastic vial (2.5 cm in diameter ×10 cm in length). The vial was gently tapped to knock the flies to the bottom, and the number of flies that reached the top, middle, and bottom areas of the vials in 10 s was counted. Experiments were repeated 10 times, and the mean percentage of flies in each area and standard deviations were calculated. Experiments were repeated with independent cohorts more than three times, and a representative result was shown.

The number of replicates, what n represents, precision measurements, and the meaning of error bars are indicated in Figure Legends. Data are shown as means ± SEM. For pairwise comparisons, Student’s t-test was performed with Microsoft Excel (Microsoft). For multiple comparisons, data were analyzed using one-way ANOVA with Tukey’s HSD multiple-comparisons test in the GraphPad Prism 6.0 software (GraphPad Software, Inc, La Jolla, CA). Results with a p-value of less than 0.05 were considered to be statistically significant.