A variant of the autophagic receptor NDP52 counteracts phospho-TAU accumulation and emerges as a protective factor for Alzheimer's disease

Abstract

Selective elimination of early pathological TAU species may be a promising therapeutic strategy to reduce the accumulation of TAU, which contributes to neurodegeneration and is a hallmark of Alzheimer's disease (AD). Pathological hyper-phosphorylated TAU can be degraded through selective autophagy, and NDP52/CALCOCO2 is one of the autophagy receptors involved in this process. In 2021, we discovered a variant of NDP52, called NDP52^GE (rs550510), that is more efficient at promoting autophagy. We here anticipate that this variant could be a powerful factor that could eliminate pathological forms of TAU better than its WT form (NDP52^WT). Indeed, we provide evidence that in in vitro systems and in a Drosophila melanogaster model of TAU-induced AD, the NDP52^GE variant is much more effective than the NDP52^WT in reducing the accumulation of pathological forms of TAU through the autophagic process and rescues typical neurodegenerative phenotypes induced by hTAU toxicity. Mechanistically, we showed that NDP52^WT and NDP52^GE bind pTAU with comparable efficiency, but that NDP52^GE binds the autophagic machinery (LC3C and LC3B) more efficiently than NDP52^WT does, which could explain its greater efficiency in removing pTAU. Finally, by performing a genetic analysis of a cohort of 435 AD patients, we defined the NDP52^GE variant as a protective factor for AD. Overall, our work highlights the variant NDP52^GE as a resilience factor in AD that shows a robust effectiveness in driving pathological TAU degradation.

Fig. 1. NDP52^GE binds LC3C more efficiently than NDP52^WT in a human neuroblastoma cell line.

A Lysates of SH-SY5Y cells expressing the indicated FLAG- and HA-tagged proteins was immunoprecipitated with anti-HA beads. Samples were analyzed by Western blot using the indicated antibodies. The graph reports the amount of the indicated FLAG-NDP52 protein coprecipitated by the corresponding HA-LC3C protein. Data were expressed as percentage variation over FLAG-NDP52^WT. Images and data are representative of seven independent experiments. B Lysates of SH-SY5Y cells expressing the indicated GFP- and HA-tagged proteins were immunoprecipitated with anti-GFP beads. Samples were analyzed by Western blot using the indicated antibodies. The graph reports the amount of HA-LC3C coprecipitated by the corresponding GFP-NDP52 protein. Data were expressed as percentage variation over GFP-NDP52^WT. Images and data are representative of four independent experiments. C Representative confocal images of SH-SY5Y cells expressing the indicated GFP- and HA-tagged proteins, fixed and stained with anti-HA (magenta staining) antibodies and with DAPI (blue staining in the merge panels) to detect nuclei. Colocalization of GFP-NDP52 and HA-LC3C was quantified by measuring the Pearson’s Correlation Coefficient. The results are reported in the graph. Images are representative of three independent experiments, and at least 30 cells for each condition were analyzed. Scale bar: 10 μm. In (B, C) data are presented as means ± SEM. *p < 0,05; ***p = <0.001 (two-tailed unpaired t-test).

Fig. 2. The increased binding affinity of NDP52^GE towards LC3C strictly depends on the cLIR motif.

A Schematic representation of the human NDP52^WT protein, the human variant NDP52^GE, and the NDP52 mutants used in this study. The G140E aminoacidic substitution is highlighted, and the position relative to cLIR motif as well as the LIR-like motif are shown in the expanded dotted area. SKICH skeletal muscle and kidney-enriched inositol phosphatase carboxyl homology, cLIR non-canonical LC3-interacting region, LIR LC3-interacting region, UBZ ubiquitin-binding zinc finger. Lysates of SH-SY5Y cells expressing the indicated FLAG- and HA-tagged proteins were immunoprecipitated with anti-HA (B) or anti-FLAG (C) beads. Samples were analyzed by Western blot using the indicated antibodies. The graphs report the amount of the indicated FLAG-NDP52 protein coprecipitated by the corresponding HA-LC3C protein. Data were expressed as percentage variation over FLAG-NDP52^WT. Images and data are representative of three independent experiments. ****p-value < 0.0001; ***p-value < 0.001; **p-value < 0.01: *p-value < 0.05 (Ordinary One-way ANOVA, Turkey’s multiple comparisons test).

Fig. 3. NDP52^GE reduces pTAU levels better than NDP52^WT, in a human neuroblastoma cell line.

A Western blot (WB) on extracts of SH-SY5Y cells treated with 100 nM OkA for the indicated times. TAU bands were measured, summed, and normalized to the corresponding signal of VINCULIN. Results are expressed as percentage variation of the levels at time 0. Images and data are representative of eight independent experiments. B Lysates of SH-SY5Y cells treated as in (A) were split and incubated with (+) or without (−) Lambda Protein Phosphatase (λPP). Images are representative of three independent experiments. Lilac lines = non-phosphorylated TAU; purple lines = phosphorylated TAU. C SH-SY5Y cells expressing the indicated FLAG-tagged proteins were treated as in (A). Signals from TAU antibody were measured on WB and normalized as in (A). Results are expressed as percentage variation of the levels measured at time 0 for each FLAG-tagged protein expressing group. Images and data are representative of eight independent experiments and are reported as means ± SEM. * indicates comparison to time 0 of each FLAG-tagged protein expressing group; & indicates comparison between 1 h and 2 h of each group (two tailed unpaired t-test) D SH-SY5Y cells expressing FLAG-tagged NDP52^GE were pre-treated with NH₄Cl (20 mM, 24 h) or not, and then were treated with 100 nM OkA as indicated. Signals from TAU antibody were measured on WB and normalized as in (A). Images and data are representative of three independent experiments and are reported as means ± SEM. At each OkA time, NH₄Cl non-treated vs NH₄Cl treated results were tested for statistical significance (two-tailed unpaired t-test). E Representative confocal images of in situ PLA performed in SH-SY5Y overexpressing FLAG-NDP52^WT or FLAG-NDP52^GE treated with DMSO or with 100 nM OkA for 1.5 h at 37 °C. Dots were counted using CellProfiler software and results were shown in the violin plot graphs. At least 100 cells for each condition were analyzed. (Ordinary One-way ANOVA, Turkey’s multiple comparisons test). Scale bar: 10 μm.

Fig. 4. NDP52^GE degrades P301S-TAU via autophagy and is more efficient than NDP52^WT.

A WB analysis of human SH-SY5Y P301S-TAU inducible cells non-treated (TET −) or treated (TET +) with tetracycline for 24 h to induce P301S-TAU expression. To block the autophagosome–lysosome fusion, cells were treated with NH₄Cl (20 mM, 24 h). Images are representative of three independent experiments. Signals from LC3II, TAU, and NDP52 antibodies were measured and normalized to the corresponding signal of VINCULIN. Data are reported as means ± SEM. (two-tailed unpaired t-test). B WB analysis of human SH-SY5Y P301S-TAU inducible cells overexpressing GFP, GFP-NDP52^WT or GFP-NDP52^GE and treated with tetracycline to induce P301S-TAU expression. After 16 h, the medium was substituted with a TET-free medium for 8 h in order to evaluate and compare TAU degradation rate. Data are reported as means ± SEM. (two-tailed unpaired t-test). C Representative confocal images of human SH-SY5Y P301S-TAU inducible cells non-treated (TET −) or treated (TET +) with tetracycline for 24 h to induce P301S-TAU expression, fixed and stained with anti-TAU (magenta staining) antibody and with DAPI (blue staining in the merge panels) to detect nuclei. Intensity of anti-TAU signal was measured, and results are reported in the graph. Images are representative of three independent experiments and more than 70 cells for each condition were analyzed. (two-tailed unpaired t-test). Scale bar: 10 μm. D Representative confocal images of human SH-SY5Y P301S-TAU inducible cells overexpressing GFP-NDP52^WT or GFP-NDP52^GE and treated as in (B). Cells were left untreated (ctrl) or were treated with the autophagic inhibitor 3-Methyladenine (+3-MA) for 8 h. Cells were fixed and stained with anti-TAU antibody and with DAPI to detect nuclei. Intensity of anti-TAU signal was measured, and results are reported in the graph. Images are representative of three independent experiments and more than 200 cells for each condition were analyzed. (two-tailed unpaired t-test).

Fig. 5. hNDP52^GE improves hTAU phenotypes in Drosophila melanogaster better than hNDP52^WT.

A Lysates from heads of flies expressing the indicated transgenes in the eye, at 29 °C, under control of GMR-GAL4 driver were analyzed by WB using the indicated antibodies. + refers to non-transgenic flies used as control. Images are representative of four independent experiments. The ratio between dAtg8II/dAtg8I was calculated and reported in the graph as means ± SEM. * indicates comparison of dAtg8II/dAtg8I between flies expressing hNDP52^GE and control; § indicates comparison of dAtg8II/dAtg8I between flies expressing hNDP52^GE and hNDP52^WT (Ordinary One-way ANOVA, Turkey’s multiple comparisons test). B Representative images of eye from flies expressing the indicated human (h) transgenes at 25 °C under control of eyeless-GAL4 driver. Violin plot below reports eye areas. Median and quartiles are shown as dotted lines. More than 150 eyes from both males and females flies were measured for each condition. * indicates comparison to control (+); # indicates comparison to hTAU; § indicates comparison to hTAU;hNDP52^WT (Ordinary One-way ANOVA, Turkey’s multiple comparisons test). C Box and whiskers (min to max) plot showing the percentage of flies reaching the threshold. Female flies expressed panneuronally (elav-GAL4) the indicated human transgenes at 29 °C. More than 40 flies per genotype were analyzed (Control n = 54 flies; hTAU n = 44; hTAU, NDP52^WT n = 53; hTAU, NDP52^GE n = 48 flies). * indicates comparison to control (+); # indicates comparison to hTAU (Ordinary One-way ANOVA, Turkey’s multiple comparisons test). D Survival curve showing lifespan of female flies expressing panneuronally (elav-GAL4) the indicated transgenes at 29 °C. * indicates comparison to control (+); # indicates comparison to hTAU (Log-rank (Mantel-Cox) test). E Lysates from heads of flies expressing panneuronally (elav-GAL4) the indicated transgenes at 29 °C were analyzed by WB using the indicated antibodies. Images are representative of four independent experiments. Signals from the indicated TAU antibodies were measured and normalized to the corresponding signal of GIOTTO. Results are expressed as means ± SEM. # indicates comparison to hTAU; § indicates comparison to hTAU;hNDP52^WT (Ordinary One-way ANOVA, Turkey’s multiple comparisons test).

Fig. 6. NDP52^GE is a protective factor in AD and counteracts pTAU accumulation.

Illustration of the highlights of the work. By using two in vitro systems, we showed that the NDP52^GE variant counteracts the accumulation of pathological pTAU better than NDP52^WT, through the autophagy process. Furthermore, we showed in vivo, in a Drosophila melanogaster model of TAU-mediated AD that the NDP52^GE variant partially rescues different pathological phenotypes induced by hTAU expression. Finally, using a genetic association case-control analysis, we showed that NDP52^GE is a protective factor in AD. Created with BioRender.com.