A switch in retrograde signaling from survival to stress in rapid-onset neurodegeneration

Abstract

Retrograde axonal transport of cellular signals driven by dynein is vital for neuronal survival. Mouse models with defects in the retrograde transport machinery, including the Loa mouse (point mutation in dynein) and the Tg(dynamitin) mouse (overexpression of dynamitin), exhibit mild neurodegenerative disease. Transport defects have also been observed in more rapidly progressive neurodegeneration, such as that observed in the SOD1(G93A) transgenic mouse model for familial amyotrophic lateral sclerosis (ALS). Here, we test the hypothesis that alterations in retrograde signaling lead to neurodegeneration. In vivo, in vitro, and live-cell imaging motility assays show misregulation of transport and inhibition of retrograde signaling in the SOD1(G93A) model. However, similar inhibition is also seen in the Loa and Tg(dynamitin) mouse models. Thus, slowing of retrograde signaling leads only to mild degeneration and cannot explain ALS etiology. To further pursue this question, we used a proteomics approach to investigate dynein-associated retrograde signaling. These data indicate a significant decrease in retrograde survival factors, including P-Trk (phospho-Trk) and P-Erk1/2, and an increase in retrograde stress factor signaling, including P-JNK (phosphorylated c-Jun N-terminal kinase), caspase-8, and p75(NTR) cleavage fragment in the SOD1(G93A) model; similar changes are not seen in the Loa mouse. Cocultures of motor neurons and glia expressing mutant SOD1 (mSOD1) in compartmentalized chambers indicate that inhibition of retrograde stress signaling is sufficient to block activation of cellular stress pathways and to rescue motor neurons from mSOD1-induced toxicity. Hence, a shift from survival-promoting to death-promoting retrograde signaling may be key to the rapid onset of neurodegeneration seen in ALS.

Figure 1.

Axonal transport is significantly inhibited in a mouse model of familial ALS. A, Double ligation assays were performed on sciatic nerves of 85-d-old SOD1^G93A (mSOD) and age-matched nontransgenic littermate (n.Tg) mice, and transgenic mice overexpressing wild-type SOD1 (wtSOD). Segments immediately proximal (P) and distal (D) to the ligation site were subjected to Western blot analysis for accumulation of KHC, p150^Glued subunit of dynactin, and p50-dynamitin subunit of dynactin; GAPDH served as a control. We observed significant differences in both retrograde (DIC, p150, p50) (p < 0.01) and anterograde (KHC) (*p < 0.05) transport compared with n.Tg and wtSOD1 controls. The average of distal and proximal intensity relative to GAPDH from three different experiments is plotted along with SEM. B, D, Live-cell imaging demonstrates slowing of axonal transport in neurons expressing mSOD1. Live imaging using phase optics was used to monitor vesicle transport. The average velocity of retrograde transport in primary motor neurons expressing mSOD1 (B) (n > 40) compared with neurons expressing wtSOD1 or noninfected (n.Inf) neurons and in DRG cultures (D) from 85-d-old mSOD1 mice compared with neurons cultured from littermate controls (n.Tg) or from mice expressing wtSOD1 (n > 30) shows a significant (p < 0.01) reduction. The average velocity of anterograde transport was also significantly reduced (p < 0.05) in MNs expressing mSOD1 (n > 39) and DRGs (n > 17) compared with controls. C, Comparisons of the distribution profiles for instantaneous velocities (n > 2000) show that expression of mSOD1 does not affect the maximum velocity of organelle motility but does induce an overall shift toward lower velocities.

Figure 2.

In vitro analysis of dynein motor function. A, B, Microtubule gliding assays are shown schematically (A) and in a time series (B) using dynein purified from n.Tg (top) or mSOD1 (bottom) mice. No difference in average velocity was observed (n > 3000). Scale bar, 3 μm. C, D, In vitro vesicular motility assays are shown schematically (C) and in a time series (D) from mSOD1/M30 (mSOD1) and M30 (n.Tg) control mice visualized using total internal reflection fluorescence microscopy. A significant difference in the number of vesicles (green; arrowheads) undergoing processive movement (arrows) along the microtubules (red) was observed for vesicles purified from mSOD1 compared with n.Tg mice (*p < 0.05). The graph represents average ± SEM of three independent experiments. Scale bar, 5 μm.

Figure 3.

Inhibition in the retrograde transport of trophic factors. A–E, Double ligation assays of sciatic nerves of 85-d-old SOD1^G93A (mSOD), 6-month Loa, and 1-year Tg^dynamitin overexpressing mice (p50) and age-matched control mice were performed, and segments immediately proximal (P) and distal (D) to the ligation site were subjected to Western blot analysis. Significant inhibition of the retrograde transport of survival factors such as P-Trk (A), BDNF (B), NGF (C), Erk1/2 (D), and Erk5 (E) was observed in all three models. *p < 0.01. F, NGF-Qdot transport was imaged in DRG neurons cultured from 85-d-old mSOD1 and n.Tg control mice. The arrows in the phase image show the direction of movement, the arrowheads track representative Q-dots, and the neurites have been pseudocolored in blue (n.Tg) or red (mSOD). Scale bar, 5 μm. G, A representative kymograph shows more pauses (arrows) in the transport of NGF-Qdots in neurons expressing mSOD1. H, The relative frequency distribution for instantaneous velocities (n > 120) shows a significant shift to lower velocities. The Qdot-NGF changes direction of movement many more times in neurons expressing mSOD1 compared with control (*p < 0.01). I, The average retrograde velocity of NGF-Qdots and the total run length were significantly decreased (n ≥ 5; *p < 0.01) (average velocity ± SEM).

Figure 4.

Proteomic analysis of dynein-associated proteins demonstrates a change in the balance of survival and stress signaling in mSOD1 mice. A, Schematic for dynein pull-down assays followed by protein microarray analysis. B, Dynein pull-downs isolate proteins interacting with dynein either directly, such as dynactin (p150 and p50), or indirectly, such as signaling endosome components synaptotagmin, Rab5, P-Erk1/2, and P-Trk. In contrast, the cytosolic protein GAPDH was not enriched in DIC pull-downs. Input, Load fraction; PD, pull-down fraction from DIC or control (Ctrl) beads; Synapto, synaptotagmin. C, D, Summary of the proteomics analysis of dynein pull-downs. Fewer survival factors (C) and more stress/death signals (D) were found to bind to dynein from mSOD1 axoplasmic extracts compared with control.

Figure 5.

Differential interaction of signaling proteins with dynein observed in SOD1^G93A but not in Loa mice compared with controls. A–C, Sciatic nerve extracts from 85-d-old SOD1^G93A (S), 12-month Loa (L), and age-matched n.Tg and wild-type (W) mice were used in DIC pull-down assays followed by Western blot analysis for different signaling proteins. Positive controls (Post. CTRL) for the pull-down assay were the p150^Glued and p50 (dynamitin) subunits of dynactin; GAPDH was used as a negative control (Neg. CTRL). D–F, P-Trk, P-Erk1/2, P-Erk5, and Bim all showed a decreased association with dynein in SOD1^G93A but not in Loa mice. G–I, Increased association with dynein was observed for activated caspase-8, P-JNK, and p75 cleavage fragment in SOD1^G93A mice, but no significant changes in these dynein cargos were observed in Loa mice. PD, DIC pull-down; CTRL, control beads; CLV, cleavage; FL, full length; CTF, C-terminal fragment; ICD, intracellular domain; Casp8, caspase-8. J, K, Differential association of vesicular signaling proteins observed in SOD1^G93A but not in Loa mice compared with wild-type controls (J). Vesicle extracts from mSOD1, Loa, and wild-type mice were subjected to Western blot analysis for differences between pTrk and p75. Synaptotagmin was used as loading control. Whereas there was a decrease in P-Trk levels (>75%; p < 0.01) and an increase in p75 cleavage fragments in the mSOD1 mice vesicles compared with control, the levels of P-Trk and p75 were the same in Loa mice compared with WT control. Blots are from representative experiments (J) and the graphs are averages of relative changes (1 = no relative change) of dynein association with specific cargo from ≥3 independent experiments ± SEM, compared with control mice (K). S, Soup; P, pellet; 0.6, 0.6 m sucrose; V, vesicle extracts; 1.5, 1.5 m sucrose; Syn, synaptotagmin.

Figure 6.

Alterations in the retrograde transport and localization of signaling molecules in neurons expressing mSOD1. A–C, Double ligation assays of sciatic nerves of 85-d-old SOD1^G93A (mSOD), 6-month Loa, and 1-year Tg^dynamitin overexpression mice (p50) and age-matched control mice were performed, and segments immediately proximal and distal to the ligation site were subjected to Western blot analysis. Stress/death factors like p75 cleavage fragments (A), JNK (B), and activated caspase-8 (actv-casp8) (C) showed alterations in retrograde axonal transport only in the mSOD1 mice. P, Proximal segment; D, distal segment. The bars represent the average intensity ± SEM relative to GAPDH from three different experiments. D–G, Rat E14 embryonic motor neurons were infected with either HSV-wtSOD or HSV-SOD1^G85R and stained with antibodies to P-Trk (D) or the intracellular domain of p75 (E). Motor neuron cultures expressing mSOD1 displayed less P-Trk and more p75 localization to processes compared with neurons expressing wtSOD1 (boxed regions are shown at higher magnifications in the insets). Scale bar: (in D) D, E, 5 μm. G, Quantification of staining indicates significantly higher levels of p75 and lower levels of pTrk (n > 30; *p < 0.01) in mSOD1 neurites than in processes from neurons expressing wtSOD1. The number of immunoreactive puncta per unit length was also quantitated. There were fewer P-Trk-positive and more p75-positive puncta along the neurites of mSOD1-expressing cells (n > 30; *p < 0.01). The graphs shown represent average ± SEM. Line scan analyses along neurites in the indicated regions are also shown (F). H, I, In vivo mouse immunostaining of p75 and P-Trk motor neurons from the sternomastoid muscles additionally shows less P-Trk expression and more p75 in mSOD1 axons. Scale bars: H, 50 μm; I, 5 μm. J, Localization along the axon. Representative line scans along axons from mSOD1 and wtSOD1 mice confirm the altered patterns of P-Trk and p75. The experiment was repeated twice with similar results.

Figure 7.

Non-cell-autonomous changes in retrograde signaling lead to cell death. E14 embryonic rat motor neurons were grown in the presence or absence of glial cells and infected with either HSV-wtSOD or HSV-SOD1^G85R. A–F, Staining by P-Trk decreased (A, C, D) and p75 reactivity along neuronal processes expressing mSOD1 increased (B, E, F) in neurons cocultured with glia. Line scans compare relative signal intensities along processes of neurons expressing mutant or wild-type SOD1 in the absence (green) or presence (orange) of glia. NF, neurofilament. Scale bar, 5 μm. G, H, Expression of mSOD1 led to enhanced reactivity with antibodies to caspase-3 compared with neurons expressing wtSOD1. The bar graph represents average ± SEM of percentage cells containing activated caspase-3 from ∼100 cells from three independent experiments. Activation of caspase-3 was more pronounced in neurons cocultured with glia also expressing mSOD1 (*p < 0.01). W, wtSOD1; S, mSOD1; Casp3, caspase-3. Scale bar, 8 μm.

Figure 8.

Inhibition of retrograde stress signaling in motor neurons is sufficient to block activation of P-c-Jun and to rescue motor neurons from mSOD1-induced toxicity. A, E14 embryonic rat motor neurons were grown for 3 weeks in compartmentalized Campenot chambers. CM-Dill tracer was added to the distal chambers to label the cell bodies of neurons that extend axons into the distal compartment. Glial cells were added to the distal compartments; both MN and glia cells were infected with HSV-mSOD1 or HSV-wtSOD1. Viable CM-Dill-positive MN cell bodies were counted after 2 (for basal number) and 6 d and percentages of survival after 6 d were scored. B, In coculture, expression of mSOD1 leads to significant cell death (∼50%) after 6 d. The addition of specific inhibitors to JNK, caspase-8, and p75 to the distal chambers in the presence of dynamitin overexpression to inhibit retrograde stress signaling led to a significant increase in cell survival (*p < 0.01). C, D, Immunostaining of MN cell bodies against the downstream target of JNK activation, the transcription factor P-c-Jun, shows activation in mSOD1-expressing cells in vitro. Partial inhibition of dynein transport by overexpression of the dynamitin (p50) subunit of dynactin prevents the upregulation of P-c-Jun (*p < 0.001) after 3 d but not after 5 d (C). However, adding a mixture of inhibitors (to JNK, p75, and caspase-8) prevents the upregulation of P-c-Jun (*p < 0.001) after 6 d (D). The experiments were repeated twice each time with four chambers. A total of more than 1000 cells for each condition was scored. MT, Microtubule. Scale bar, 20 μm. E, Western blot analysis of 85 d spinal cord extracts shows upregulation of P-c-Jun in mSOD1 mice (S) compared with n.Tg control (n.Tg). The Western blot analysis was repeated three times with similar results and quantified using ImageJ.

Figure 9.

Model correlating changes in axonal transport balance with disease severity and time course: slowing of retrograde transport is an early step that leads to slowly progressive motor neuron degeneration, because of decreased trophic factor signaling, as seen in the Loa and Tg^dynamitin models, whereas changes in the nature of the signals undergoing retrograde transport by 50 d (presymptomatic) lead to more severe and more rapidly progressive degeneration by 120 d (severe symptoms and late stage of disease), as seen in the SOD1^G93A model. Disease stages were classified based on symptoms like weight loss, walking difficulty, and muscle atrophy. Time scale for the SOD1^G93A is shown. Sym, Symptomatic.