PAAN/MIF nuclease inhibition prevents neurodegeneration in Parkinson's disease

Abstract

Parthanatos-associated apoptosis-inducing factor (AIF) nuclease (PAAN), also known as macrophage migration inhibitor factor (MIF), is a member of the PD-D/E(X)K nucleases that acts as a final executioner in parthanatos. PAAN's role in Parkinson's disease (PD) and whether it is amenable to chemical inhibition is not known. Here, we show that neurodegeneration induced by pathologic α-synuclein (α-syn) occurs via PAAN/MIF nuclease activity. Genetic depletion of PAAN/MIF and a mutant lacking nuclease activity prevent the loss of dopaminergic neurons and behavioral deficits in the α-syn preformed fibril (PFF) mouse model of sporadic PD. Compound screening led to the identification of PAANIB-1, a brain-penetrant PAAN/MIF nuclease inhibitor that prevents neurodegeneration induced by α-syn PFF, AAV-α-syn overexpression, or MPTP intoxication in vivo. Our findings could have broad relevance in human pathologies where parthanatos plays a role in the development of cell death inhibitors targeting the druggable PAAN/MIF nuclease.

Keywords: MIF; PARP1; Parkinson’s disease; cell death; neurodegeneration; parthanatos; synuclein.

Figure 1.. α-Syn PFF-induced pathology is reduced by deletion of PAAN/MIF’s nuclease in vivo

(A) Representative TH and Nissl staining of SNpc DA neurons of α-syn PFF injected WT and PAAN/MIF KO at 6 months after α-syn PFF or PBS injection. Scale bars, 400 μm. (B) Stereological counts of TH⁺ cells. Data are mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 5–6 mice per group). (C and D) (C) Dopamine and (D) DOPAC concentrations in the striatum of WT and PAAN/MIF KO at 6 months after α-syn PFF or PBS injection measured by HPLC. Bars represent mean ± SEM. *p < 0.05, two-way ANOVA followed by Tukey’s post hoc test (n = 5–7 mice per group). (E and F) 180 days after intrastriatal α-syn PBS or PFF injection, (E) pole test and (F) grip strength test was performed in WT or PAAN/MIF KO. Data are the mean ± SEM. *p < 0.05, **p < 0.005, two-way ANOVA followed by Tukey’s post hoc test (n = 7 mice per group). (G) Representative TH and Nissl staining of SNpc DA neurons of WT, E22Q, and P2G knockin mice at 6 months after intrastriatal α-syn PFF or PBS injection. Scale bars, 400 μm. (H) Stereological counts of TH⁺ cells. Data are mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 5–6 mice per group). ns, not significant. (I) Dopamine concentrations in the striatum of PAAN/MIF WT, E22Q, and P2G knockin mice at 6 months after α-syn PFF or PBS injection measured by HPLC. Bars represent mean ± SEM. **p < 0.005, ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 5–8 mice per group). ns, not significant. (J and K) 180 days after intrastriatal PBS or α-syn PFF injection in PAAN/MIF WT, E22Q, and P2G knockin mice, (J) pole test and (K) grip strength test was performed. Data are the mean ± SEM. *p < 0.05, **p < 0.005, two-way ANOVA followed by Tukey’s post hoc test (n = 7–10 mice per group). (L) Pulsed-field gel electrophoresis in the substantia nigra of PBS or α-syn-PFF-injected PAAN/MIF WT, E22Q KI, or P2G KI mice at 6 months. (M) Intensity of noncleaved genomic DNA is shown in the graph. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 3–4 mice per group). See also Figures S1 and S2.

Figure 2.. PAAN/MIF nuclease activity is required for prevention of α-syn PFF-induced neurotoxicity in neurons

(A) Nuclear translocation of AIF and PAAN/MIF after α-syn PFF treatment in the presence of the PARP inhibitor, ABT-888 (1 μM) in cortical neurons. Intensity of PAAN/MIF and AIF signal is shown in the graph. **p < 0.005 versus the PBS control group in the nuclear (N) fraction, two-way ANOVA followed by Tukey’s post hoc test (n = 3). (B) Immunoprecipitation (IP) of PAAN/MIF and AIF in PBS or α-syn-PFF-treated cortical neurons. Intensity of AIF-bound PAAN/MIF is shown in the graph. *p < 0.05, Student’s t test (n = 3). (C) Confocal images of nuclear translocation of PAAN/MIF (green) and AIF (red) after α-syn PFF treatment in primary cortical neurons. Scale bars, 20 μm. The white color in the merged images indicates the overlay of AIF, PAAN/MIF, and Hoechst dye in the nucleus. The percentage of cells with nuclear localization of PAAN/MIF and AIF is shown in the graph. ***p < 0.0005, Student’s t test (n = 5). (D) Pulsed-field gel electrophoresis of α-syn PFF-induced DNA cleavage in PAAN/MIF WT and KO neurons and KO neurons expressing PAAN/MIF WT, E22Q, E22A, or P2G. Intensity of noncleaved genomic DNA is shown in the graph. ***p < 0.0005, two-way ANOVA (n = 3). (E) Representative images of Hoechst and propidium iodide (PI) staining from primary cortical neurons transduced with lentivirus containing PAAN/MIF WT, E22Q, E22A, or P2G and further incubated with α-syn PFF. Scale bars, 20 μm. (F) Quantification of cell death. Bars represent mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 3). (G) Quantification of cell death from Hoechst and PI staining of primary cortical neurons from PAAN/MIF WT, KO, E22Q, and P2G KI mice. Bars represent mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 5). (H) Pulsed-field gel electrophoresis of α-syn PFF-induced DNA cleavage in PAAN/MIF WT, KO, E22Q, and P2G KI neurons treated with PBS or α-syn PFF. Intensity of noncleaved genomic DNA is shown in the graph. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 4). See also Figure S3

Figure 3.. Identification of PAAN/MIF nuclease inhibitors

(A) The schematic representation of macrocyclic screening of PAAN/MIF inhibitors based on PAAN/MIF nuclease DNA cleavage assay. Single-strand-amino-modified oligonucleotide (RF substrate) was immobilized on DNA-BIND plates and incubated in PAAN/MIF nuclease with or without inhibitors. After PAAN/MIF’s cleavage, the fragments were hybridized with biotin-labeled complementary oligonucleotides and detected by monitoring absorbance at 450 nm. (B) Scatter plot of percentage inhibition of PAAN/MIF cleavage from 45,000 compounds with 3,000 pools in 38 plates of the macrocyclic library. The top line (blue) is the incubation without PAAN/MIF and bottom line (green) is the incubation with PAAN/MIF. Right graph represents the histogram of the compounds tested. (C) The result of secondary screening from the primary screening pools that inhibited PAAN/MIF nuclease activity by at least 60%. Scatter plot of the percentage the inhibition of PAAN/MIF nuclease activity (x axis) and the inhibition of MNNG-induced cell death (y axis). (D) The result of the 90 individual compounds screening of 6 pools candidates from the secondary screening. Correlation analysis between the inhibition of PAAN/MIF nuclease activity and the inhibition of MNNG-induced cell death is shown in the graph. (E) Alamar blue cell viability assay from SH-SY5Y cells pre-incubated with PAAN/MIF inhibitors (C1–C12) selected by individual screening in a concentration-dependent manner (0.1, 0.2, 0.5, and 1 μM) in response to 50-μM MNNG for 15 min. A PARP inhibitor, ABT-888 (1 μM) is used for the positive control. Bars represent mean ± SEM. **p < 0.005, ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 3). (F) In vitro PAAN/MIF’s nuclease assay with PAAN/MIF inhibitors (C1–C12) using RF substrates. Quantification of noncleaved genomic DNA. Bars represent mean ± SEM. *p < 0.05, **p < 0.005, ***p < 0.0005, one-way ANOVA followed by Tukey’s post hoc test (n = 3). (G) Representative images of Hoechst and PI staining from human cortical neurons pre-incubated with 0.5 or 1 μM of C8, C12, or ABT-888 for 1 h and further incubated with human α-syn PFF for 14 days. Scale bars, 20 μm. Quantification of cell death. Bars represent mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 4). (H) Pulsed-field gel electrophoresis of α-syn-PFF-induced DNA cleavage in human cortical neurons treated with C8 or C12. Intensity of noncleaved genomic DNA is shown in the graph. **p < 0.005, ***p < 0.0005, two-way ANOVA (n = 4). See also Figure S4

Figure 4.. Identification of PAANIB-1 as a PAAN/MIF nuclease inhibitor

(A) The result of the structural-activity relationship (SAR) screening of C8. Scatter plot of the percent inhibition of PAAN/MIF nuclease activity (x axis) and the percent inhibition of MNNG-induced cell death (y axis). (B) Structure of PAANIB-1. (C) SH-SY5Y cells were pre-incubated C8 or PANNIB-1 with concentrations as indicated for 1 h, followed by 50-μM MNNG for 15 min. After 24 h, cell viability was measured by Alamar blue. Data represent mean ± SEM. *p < 0.05, **p < 0.005, Student’s t test (n = 3). The half maximal inhibitory concentration (IC₅₀) of C8 is 0.52 μM, and C8–31 is 0.28 μM. (D) Quantification of cell death from Hoechst and PI staining from primary cortical neurons pretreated with 0.2, 0.5, and 1 μM of C8 or PAANIB-1, followed by further incubation with α-syn PFF. Bars represent mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 3). ns, not significant. (E) C57BL/6 mice were orally administrated with vehicle or 10 mg/kg of C8, C12, or PAANIB-1 for 2 h. The concentrations of compounds in brain tissue were measured by LC/MS. *p < 0.05, one-way ANOVA followed by Tukey’s post hoc test (n = 3–5 per group). (F) Pulsed-field gel electrophoresis of α-syn PFF-induced DNA cleavage in mouse cortical neurons treated with PAANIB-1. Intensity of noncleaved genomic DNA is shown in the graph. Bars represent mean ± SEM. *p < 0.05, **p < 0.005, ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 4). (G) Representative traces of mEPSCs. Primary cortical neurons were pretreated with 1 μM of PAANIB-1, followed by further incubation with α-syn PFF. After 14 days, mEPSCs were recorded in the presence of TTX (0.25 μM), bicuculline (10 μM), and APV (20 μM). Quantification of amplitude and frequency for recordings. Bars represent mean ± SEM. *p < 0.05, **p < 0.005. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test. For each group, 10 neurons from three independent cultures were recorded and analyzed. (H) Quantification of noncleaved substrate DNA from in vitro MIF’s nuclease assay with indicated dose of rapamycin, FK506, or PAANIB-1 using the PS30 substrate. Bars represent mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 4). (I) In vitro PAAN/MIF’s nuclease assay using MIF tautomerase inhibitors (10 μM of p-425, 50 μM of ISO-1, or 10 μM of AV411) or a 10 μM of PAAN/MIF nuclease inhibitor, PANNIB-1. Quantification of noncleaved substrate DNA. Bars represent mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 5). (J) Alamar blue cell viability assay from SH-SY5Y cells pre-incubated with PAAN/MIF tautomerase inhibitors (10 μM of p-425, 50 μM of ISO-1, or 10 μM of AV411) or 1 mM of PANNIB-1 in response to 50-μM MNNG for 15 min. Bars represent mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 5). See also Figure S5

Figure 5.. Direct interaction of PAANIB-1 to PAAN/MIF as an action mechanism of inhibition

(A) Binding affinities of PAAN/MIF WT for PAANIB-1 determined by biolayer interferometry (ForteBio Octet) assay. Data are representative of three independent experiments. (B) In vitro PAAN/MIF’s nuclease assay using PAAN/MIF WT or mutants with 10 μM of PANNIB-1. Quantification of noncleaved substrate DNA. Bars represent mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 4). (C) WT or PAAN/MIF KO SH-SY5Y cells expressing Flag-PAAN/MIF WT or mutants were pre-incubated 1 μM of PANNIB-1 for 1 h, followed by 50-μM MNNG for 15 min. After 24 h, cell viability was measured by Alamar blue. Data represent mean ± SEM. *p < 0.05, ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 5). (D) Binding of PAAN/MIF WT or mutants for PANNIB-1 determined by biolayer interferometry (ForteBio Octet) assay. Data are average of three independent experiments. (E) Representative images of Hoechst and PI staining from WT or PAAN/MIF KO cortical neurons expressing PAAN/MIF WT, L62A, or N73A pre-incubated withPAANIB-1 and further incubated with α-syn PFF for 14 days. Scale bars, 20 μm. Quantification of cell death. Bars represent mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 5). See also Figure S6.

Figure 6.. PAANIB-1 protects against α-syn PFF-induced pathology in vivo

(A) Schematic diagram of the experimental design. PBS or α-syn PFF were injected into the striatum of 2-month-old WT mice. After 1 month, PAANIB-1 or vehicle was delivered by oral administration with two different doses (5 or 15 mg/kg) for 5 months. (B) Representative TH and Nissl staining of SNpc DA neurons of PBS or α-syn-PFF-injected WT mice treated with vehicle or PAANIB-1. Scale bars, 400 μm. (C and D) Stereological counts of (C) TH-positive cells and (D) Nissl-positive cells. Data are mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 6–8 mice per group). (E and F) (E) Dopamine and (F) DOPAC concentrations in the striatum of PBS or α-syn-PFF-injected WT mice treated with vehicle of PAANIB-1 as assessed by HPLC. Bars represent mean ± SEM. **p < 0.005, ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 5–6 mice per group). (G and H) 180 days after intrastriatal α-syn PFF or PBS injection in WT mice treated with vehicle or PAANIB-1, (G) pole test and (H) grip strength test was performed. Data are the mean ± SEM. *p < 0.05, ***p < 0.0005, two-way ANOVA. (I) Pulsed-field gel electrophoresis in the substantia nigra of PBS or α-syn-PFF-injected WT mice treated with vehicle or PAANIB-1. Intensity of noncleaved genomic DNA is shown in the graph. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 3 to 5 mice per group). See also Figure S7.

Figure 7.. PAANIB-1 protects against AAV-α-syn- or MPTP-induced pathology in vivo

(A) Schematic diagram of the experimental design. AAV-GFP or AAV-α-syn were injected into the substantia nigra of 2-month-old WT mice. After 14 days, PAANIB-1 or vehicle was delivered by oral administration with two different doses (5 or 15 mg/kg) for 2.5 months. (B) Representative TH and Nissl staining of SNpc DA neurons of AAV-GFP or AAV-α-syn injected WT mice treated with vehicle or PAANIB-1. Scale bars, 400 μm. (C and D) Stereological counts of (C) TH-positive and (D) TH- and Nissl-positive cells. Data are mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 5). (E–G) 90 days after intrastriatal AAV-GFP or AAV-α-syn injection in WT mice treated with vehicle or PAANIB-1, (E) amphetamine rotation test, (F) pole test, and (G) grip strength test was performed. Data are the mean ± SEM. **p < 0.005, ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test. (H and I) Representative immunoblots of TH, α-syn and β-actin in SNpc of AAV-GFP or AAV-α-syn-injected WT mice treated with vehicle or PAANIB-1. Quantification of TH levels (left) and α-syn level (right) normalized to β-actin. Bars represent the mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 3). ns, not significant. (J) Schematic diagram of the experimental design. Mice were pretreated with vehicle or PAANIB-1 (5 or 15 mg/kg) a week before MPTP injections and treatmentwas continued until experiments. (K) Representative TH and Nissl staining of SNpc DA neurons of saline or MPTP injected WT mice treated with vehicle or PAANIB-1. Scale bars, 400 μm. (L) Stereological counts of TH- and Nissl-positive cells. Data are mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test (n = 5). (M and N) 7 days after saline or MPTP injection, (M) pole and (N) grip strength tests were performed. Data are the mean ± SEM. ***p < 0.0005, two-way ANOVA followed by Tukey’s post hoc test. (O) Striatal MPP⁺ concentrations after a single injection of saline or MPTP treated with vehicle or PAANIB-1.