Understanding the 15-PGDH study: methods and results

Male C57BL/6J mice were purchased from Daehan BioLink Co., Ltd (Korea) and maintained at the Animal Center for Pharmaceutical Research of Seoul National University under temperature, light, and humidity-controlled conditions with free access to food and water. Hpgd +/- males and females were bred and maintained under the same conditions as C57BL/6J mice. All animal work was approved by the Institutional Animal Care and Use Committee (IACUC, SNU-221026-6-5) of Seoul National University.

Here, we establish the therapeutic potential of 15-PGDH inhibition in PD, a mechanistically different neurodegenerative disorder. We observed significant 15-PGDH upregulation in the substantia nigra of postmortem human PD brains as well as three different mouse models of PD: systemic 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) [13], intranigral lipopolysaccharide (LPS) [14], intrastriatal AAV-α-synuclein with intra-ventral tegmental area α-synuclein preformed fibrils (PFFs) [15]. We utilized both pharmacologic inhibition (SW033291 treatment) and genetic partial reduction of 15-PGDH. SW033291 has high specificity for inhibiting 15-PGDH, with IC50 < 1 nM and no interaction with other structurally related short-chain dehydrogenases [6]. Nonetheless, to control for any previously unidentified off-target drug effects, we compared 15-PGDH inhibition with SW033291 versus genetic inhibition of 15-PGDH via knockout of one Hpgd allele. Both drug and genetic inhibition of 15-PGDH showed complete concordance in protecting from motor deficits, dopaminergic cell loss, BBB degradation, and induction of multiple molecular markers of inflammation and ROS-induced tissue damage. This identification of SW033291 and genetic reduction of Hpgd in protecting the substantia nigra is consistent with our prior findings of complete concordance of SW033291 and genetic deletion of Hpgd in AD and TBI models [12].

Postmortem human substantia nigra tissue from PD subjects displayed elevated mRNA for Hpgd (the gene encoding 15-PGDH) (Fig. 1A; Table S1). Similarly, exposure of mice to the neurotoxin MPTP, which selectively targets the substantia nigra dopaminergic neurons implicated in PD [[17], [18], [19], [20]], also increased Hpgd mRNA levels, 15-PGDH protein levels, and 15-PGDH enzymatic activity in the substantia nigra (Fig. 1B–D). Hpgd mRNA was similarly elevated in mice receiving intranigral LPS, intrastriatal AAV-α-synuclein with intra-ventral tegmental area α-synuclein PFFs, and intrastriatal PFF (Fig. 1E–G). Lastly, Hpgd mRNA induction was observed in SH-SY5Y cells that were transduced to overexpress wild-type α-synuclein, a cellular model of PD (Fig. 1H). Collectively, our findings establish 15-PGDH elevation as a consistent feature across human PD and preclinical models.Fig. 115-PGDH is elevated in Parkinson's disease (PD) in cellular and animal models and human brains(A) HPGD mRNA is increased in the substantia nigra of human PD brain, relative to control subjects (n = 5-8 per group, ∗p < 0.05, unpaired t-test).(B) Hpgd mRNA is increased in the substantia nigra of MPTP-exposed mice, relative to vehicle-treated mice (n = 3 per group, ∗p < 0.05, unpaired t-test).(C) 15-PGDH protein expression is increased in the substantia nigra of MPTP-exposed mice, relative to vehicle-treated mice (n = 3 per group, ∗p < 0.05, unpaired t-test).(D) 15-PGDH activity is elevated in the substantia nigra of MPTP-exposed mice, relative to vehicle-treated animals (n = 5 per group, ∗∗p < 0.01, unpaired t-test).(E) Hpgd mRNA expression is elevated in mice expose to intranigral LPS administration, relative to vehicle-treated mice (n = 3 per group, ∗p < 0.05, unpaired t-test).(F) Hpgd mRNA expression is increased in mice with intrastriatal AAV-ɑ-synuclein combined with intra-ventral tegmental area injection of ɑ-synuclein preformed fibrils (PFF), relative to AAV-eGFP and PBS-exposed mice (n = 5-6 per group, ∗p < 0.05, unpaired t-test)(G) Hpgd mRNA expression is increased in mice with intrastriatal fibrillar alpha-synuclein administration, relative to PBS-exposed mice (n = 3 per group, ∗p < 0.05, unpaired t-test).(H) HPGD mRNA is increased in the SH-SY5Y α-syn BiFC cell line, which stably expresses wild-type alpha-synuclein, relative to control SH-SY5Y cells (n = 6 per group, ∗∗∗∗p < 0.0001, unpaired t-test).

Previous studies have demonstrated that 15-PGDH inhibition confers protection from liver, colon, bone marrow, and kidney injury through stimulating resident stem and stem-like cells, and protects in TBI and AD through preventing ROS-mediated damage to the BBB [[9], [10], [11], [12],16]. To explore a potential role for 15-PGDH in PD, we analyzed 15-PGDH expression in the substantia nigra and striatum in human PD tissues and rodent PD models. The substantia nigra houses the dopaminergic neurons that project axons to the striatum to control movement, and loss of these cells is a principal pathophysiology of PD.

The enzyme 15-hydroxyprostaglandin dehydrogenase (15-PGDH) occupies a central regulatory role in bioactive lipid metabolism, catalyzing nicotinamide adenine dinucleotide (NAD+)-dependent degradation of prostaglandins and autocoids, with prostaglandin E2 (PGE2) representing its prototypical substrate [[5], [6], [7], [8]]. Our prior work identified SW033291 as a potent small-molecule 15-PGDH inhibitor that enhances tissue regeneration by activating stem cell populations in peripheral organs [6]. Subsequent studies revealed that 15-PGDH inhibition attenuates oxidative stress and inflammation, conferring protection against renal and hepatic injury, albeit by incompletely understood mechanisms [[9], [10], [11]]. More recently, we demonstrated that 15-PGDH suppression counteracts neuroinflammation and blocks generation of myeloid-derived reactive oxygen species (ROS) in the brain, thereby preserving blood-brain barrier (BBB) integrity and preventing neurodegeneration and cognitive impairment in mouse models of traumatic brain injury (TBI) and Alzheimer's disease (AD) [12]. Notably, this neuroprotection occurred without altering amyloid pathology in an amyloid-driven mouse AD model, revealing a paradigm-shifting therapeutic axis for AD [12].

Here, we establish the therapeutic potential of 15-PGDH inhibition in PD, a mechanistically different neurodegenerative disorder. We observed significant 15-PGDH upregulation in the substantia nigra of postmortem human PD brains as well as three different mouse models of PD: systemic 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) [13], intranigral lipopolysaccharide (LPS) [14], intrastriatal AAV-α-synuclein with intra-ventral tegmental area α-synuclein preformed fibrils (PFFs) [15]. We utilized both pharmacologic inhibition (SW033291 treatment) and genetic partial reduction of 15-PGDH. SW033291 has high specificity for inhibiting 15-PGDH, with IC50 < 1 nM and no interaction with other structurally related short-chain dehydrogenases [6]. Nonetheless, to control for any previously unidentified off-target drug effects, we compared 15-PGDH inhibition with SW033291 versus genetic inhibition of 15-PGDH via knockout of one Hpgd allele. Both drug and genetic inhibition of 15-PGDH showed complete concordance in protecting from motor deficits, dopaminergic cell loss, BBB degradation, and induction of multiple molecular markers of inflammation and ROS-induced tissue damage. This identification of SW033291 and genetic reduction of Hpgd in protecting the substantia nigra is consistent with our prior findings of complete concordance of SW033291 and genetic deletion of Hpgd in AD and TBI models [12].

We next investigated the neuroprotective potential of 15-PGDH inhibition in two PD models, systemic MPTP administration and intranigral LPS injection. In the MPTP study, mice underwent a one-week acclimation period prior to receiving intraperitoneal injections of the 15-PGDH inhibitor SW033291 (0.5 or 5 mg/kg, twice daily) for two days, after which treatment was continued alongside daily MPTP injections (30 mg/kg) for seven more days. SW033291 significantly reduced 15-PGDH activity and increased PGE2 levels in the substantia nigra without affecting body weight (Fig. S1A–C). Behavioral assessments were conducted 2–3 h after the final MPTP injection (Fig. 2A). MPTP administration induced significant motor deficits, evidenced by reduced rotarod latency (Fig. 2B), increased hind-limb clasping (Fig. 2C and Fig. S1D and E), and prolonged pole test completion times (Fig. 2D). Notably, SW033291 treatment dose-dependently prevented all these impairments, preserving motor performance to near non-MPTP exposed levels (Fig. 2B–D and Fig. S1E).Fig. 2Pharmacological and genetic inhibition of 15-PGDH ameliorates motor deficits and dopaminergic neuronal loss in MPTP- and intranigral LPS-induced PD models(A) Experimental procedure for evaluating neuroprotective efficacy of SW033291 in the MPTP-induced mouse PD model.(B) SW033291 protects MPTP-exposed mice from motor deficits in the rotarod test (n = 10-11 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(C) SW033291 dose-dependently protects MPTP-exposed mice from hindlimb clasping behavior (n = 10-11 per group, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(D) SW033291 protects MPTP-exposed mice from increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 10-11 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(E) Experimental procedure for evaluating neuroprotective efficacy of 15-PGDH haploinsufficiency (Hpgd+/−) in the MPTP-induced mouse PD model.(F) Male and female Hpgd heterozygous mice are protected from MPTP-induced motor deficits in the rotarod test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(G) Hpgd heterozygous mice are protected from MPTP-induced hindlimb clasping behavior (n = 15-22 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(H) Hpgd heterozygous mice are protected from MPTP-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(I) Experimental procedure for evaluating the neuroprotective efficacy of SW033291 in the intranigral LPS-injected mouse PD model.(J) SW033291 protects mice from intranigral LPS-induced motor deficits in the rotarod test (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(K) SW033291 protects mice from mice from intranigral LPS-induced hindlimb clasping behavior (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(L) SW033291 protects mice from intranigral LPS-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 6-12 per group, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(M) Representative TH-stained images of the substantia nigra of mice treated with vehicle or MPTP in the absence or presence of SW033291 (scale bar = 100 μm) (N) Quantification of TH fluorescence signal shows that SW033291 treatment dose-dependently protects MPTP-exposed mice from loss of TH intensity (n = 4 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(O) Representative TH-stained images of the substantia nigra from Hpgd heterozygous mice and their wild-type littermates treated with vehicle or MPTP (scale bar = 100 μm) (P) Quantification of TH fluorescence signal shows that Hpgd heterozygous mice are protected from MPTP-induced loss of TH intensity (n = 4 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(Q) Representative TH-stained images of the substantia nigra from mice treated with vehicle or LPS, in the absence or presence of SW033291 (scale bar = 200 μm)(R) Quantification of TH fluorescence signal shows that SW033291 treatment protects mice from intranigral LPS-induced loss of TH intensity (n = 3 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).

As α-synuclein–driven models most closely recapitulate key features of human PD, we further tested the efficacy of 15-PGDH inhibition in this mouse model. Seven-week-old C57BL/6J mice were acclimated for one week and then pretreated with either vehicle or the 15-PGDH inhibitor SW033291 (5 mg/kg, intraperitoneally, twice daily) for two days prior to surgery. Mice subsequently received intranigral injections of either AAV-GFP (comparator) or AAV–α-synuclein (disease model), followed by intra-VTA injection of PBS (comparator) or α-synuclein preformed fibrils (PFFs) (disease model). Behavioral assessments were performed at designated time points, and brains were collected for biochemical and histological analysis at 33 days post-injection (Fig. 4A). Consistent with our previous reports [39,40], the combination of AAV-α-synuclein and PFF administration induced pronounced motor dysfunction, reflected by shortened rotarod latency (Fig. 4B), elevated hind-limb clasping scores (Fig. 4C), and prolonged pole test times (Fig. 4D). SW033291 treatment prevented these impairments and maintained motor performance comparable to AAV-eGFP/PBS controls, with no change in body weight (Fig. 4B–D and Fig. S8A).Fig. 415-PGDH inhibition ameliorates motor deficits, dopaminergic neuronal loss, and oxidative stress in the ɑ-synuclein mouse PD model(A) Experimental procedure for evaluating the efficacy of SW033291 in the α-synuclein mouse model.(B) SW033291 protects mice from AAV-ɑ-synuclein/PFF-induced motor deficits in the rotarod test (n = 5-6 per group, ∗∗p < 0.01, one-way ANOVA and Tukey's post hoc analysis).(C) SW033291 protects mice from AAV-ɑ-synuclein/PFF-induced from hindlimb clasping behavior (n = 5-6 per group, ∗p < 0.05, ∗∗p < 0.01, one-way ANOVA and Tukey's post hoc analysis).(D) SW033291 protects mice from AAV-ɑ-synuclein/PFF-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 5-6 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(E) Representative images and quantification of TH immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced loss of TH intensity (scale bar = 800 μm, n = 5-6 per group, ∗p < 0.05, one-way ANOVA and Tukey's post hoc analysis).(F) Representative images and quantification of 4-HNE immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in 4-HNE (scale bar = 100 μm, n = 5-6 per group, ∗p < 0.05, one-way ANOVA and Tukey's post hoc analysis).(G) Representative images and quantification of 3-NT immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in 3-NT (scale bar = 100 μm, n = 5-6 per group, ∗p < 0.05, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(H) RT-qPCR analysis shows that SW033291 treatment protected mice from administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in Cybb mRNA expression (n = 3 per group, ∗∗p < 0.01, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(I) Representative images and quantification of Iba1 immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in Iba1 (scale bar = 200 μm, n = 5-6 per group, ∗∗p < 0.01, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(J) Representative images and quantification of GFAP immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in GFAP (scale bar = 300 μm, n = 5-6 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(K) RT-qPCR analysis shows that SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in Lcn2 mRNA expression (n = 3 per group, ∗p < 0.05, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(L) Representative images and quantitative analysis of p-α-synuclein immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 treatment had no effect on p-α-synuclein accumulation in AAV-ɑ-synuclein/PFF mice (scale bar = 20 μm, n = 5-6 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).

We next investigated the neuroprotective potential of 15-PGDH inhibition in two PD models, systemic MPTP administration and intranigral LPS injection. In the MPTP study, mice underwent a one-week acclimation period prior to receiving intraperitoneal injections of the 15-PGDH inhibitor SW033291 (0.5 or 5 mg/kg, twice daily) for two days, after which treatment was continued alongside daily MPTP injections (30 mg/kg) for seven more days. SW033291 significantly reduced 15-PGDH activity and increased PGE2 levels in the substantia nigra without affecting body weight (Fig. S1A–C). Behavioral assessments were conducted 2–3 h after the final MPTP injection (Fig. 2A). MPTP administration induced significant motor deficits, evidenced by reduced rotarod latency (Fig. 2B), increased hind-limb clasping (Fig. 2C and Fig. S1D and E), and prolonged pole test completion times (Fig. 2D). Notably, SW033291 treatment dose-dependently prevented all these impairments, preserving motor performance to near non-MPTP exposed levels (Fig. 2B–D and Fig. S1E).Fig. 2Pharmacological and genetic inhibition of 15-PGDH ameliorates motor deficits and dopaminergic neuronal loss in MPTP- and intranigral LPS-induced PD models(A) Experimental procedure for evaluating neuroprotective efficacy of SW033291 in the MPTP-induced mouse PD model.(B) SW033291 protects MPTP-exposed mice from motor deficits in the rotarod test (n = 10-11 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(C) SW033291 dose-dependently protects MPTP-exposed mice from hindlimb clasping behavior (n = 10-11 per group, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(D) SW033291 protects MPTP-exposed mice from increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 10-11 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(E) Experimental procedure for evaluating neuroprotective efficacy of 15-PGDH haploinsufficiency (Hpgd+/−) in the MPTP-induced mouse PD model.(F) Male and female Hpgd heterozygous mice are protected from MPTP-induced motor deficits in the rotarod test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(G) Hpgd heterozygous mice are protected from MPTP-induced hindlimb clasping behavior (n = 15-22 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(H) Hpgd heterozygous mice are protected from MPTP-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(I) Experimental procedure for evaluating the neuroprotective efficacy of SW033291 in the intranigral LPS-injected mouse PD model.(J) SW033291 protects mice from intranigral LPS-induced motor deficits in the rotarod test (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(K) SW033291 protects mice from mice from intranigral LPS-induced hindlimb clasping behavior (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(L) SW033291 protects mice from intranigral LPS-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 6-12 per group, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(M) Representative TH-stained images of the substantia nigra of mice treated with vehicle or MPTP in the absence or presence of SW033291 (scale bar = 100 μm) (N) Quantification of TH fluorescence signal shows that SW033291 treatment dose-dependently protects MPTP-exposed mice from loss of TH intensity (n = 4 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(O) Representative TH-stained images of the substantia nigra from Hpgd heterozygous mice and their wild-type littermates treated with vehicle or MPTP (scale bar = 100 μm) (P) Quantification of TH fluorescence signal shows that Hpgd heterozygous mice are protected from MPTP-induced loss of TH intensity (n = 4 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(Q) Representative TH-stained images of the substantia nigra from mice treated with vehicle or LPS, in the absence or presence of SW033291 (scale bar = 200 μm)(R) Quantification of TH fluorescence signal shows that SW033291 treatment protects mice from intranigral LPS-induced loss of TH intensity (n = 3 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).

We next evaluated whether these same interventions also conferred neuroprotection to dopaminergic neurons in these models. Immunohistochemical analysis revealed a significant reduction in tyrosine hydroxylase (TH), a marker of dopaminergic neurons, in the substantia nigra of MPTP-treated mice compared to vehicle-treated controls (Fig. 2M and N). Strikingly, administration of the 15-PGDH inhibitor SW033291 enhanced dopaminergic neuron survival in a dose-dependent manner (Fig. 2M and N). MPTP-injected animals also exhibited pronounced neuronal loss in the striatum, which was dose-dependently attenuated by 15-PGDH inhibition (Fig. S2A and B). Western blot analysis confirmed these findings, showing that 15-PGDH inhibition significantly preserved TH expression in both the substantia nigra and striatum of MPTP-exposed mice (Fig. S2C and D). Hpgd +/- mice were also protected from MPTP-induced dopaminergic neuronal loss (Fig. 2O and P, Fig. S2E and F). Western blot analysis confirmed TH levels, which declined in MPTP-exposed Hpgd+/+ mice and remained at baseline in Hpgd +/- mice (Fig. S2G and H). Lastly, 15-PGDH inhibition also maintained TH levels in the intranigral LPS model (Fig. 2Q and R).

We next investigated the neuroprotective potential of 15-PGDH inhibition in two PD models, systemic MPTP administration and intranigral LPS injection. In the MPTP study, mice underwent a one-week acclimation period prior to receiving intraperitoneal injections of the 15-PGDH inhibitor SW033291 (0.5 or 5 mg/kg, twice daily) for two days, after which treatment was continued alongside daily MPTP injections (30 mg/kg) for seven more days. SW033291 significantly reduced 15-PGDH activity and increased PGE2 levels in the substantia nigra without affecting body weight (Fig. S1A–C). Behavioral assessments were conducted 2–3 h after the final MPTP injection (Fig. 2A). MPTP administration induced significant motor deficits, evidenced by reduced rotarod latency (Fig. 2B), increased hind-limb clasping (Fig. 2C and Fig. S1D and E), and prolonged pole test completion times (Fig. 2D). Notably, SW033291 treatment dose-dependently prevented all these impairments, preserving motor performance to near non-MPTP exposed levels (Fig. 2B–D and Fig. S1E).Fig. 2Pharmacological and genetic inhibition of 15-PGDH ameliorates motor deficits and dopaminergic neuronal loss in MPTP- and intranigral LPS-induced PD models(A) Experimental procedure for evaluating neuroprotective efficacy of SW033291 in the MPTP-induced mouse PD model.(B) SW033291 protects MPTP-exposed mice from motor deficits in the rotarod test (n = 10-11 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(C) SW033291 dose-dependently protects MPTP-exposed mice from hindlimb clasping behavior (n = 10-11 per group, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(D) SW033291 protects MPTP-exposed mice from increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 10-11 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(E) Experimental procedure for evaluating neuroprotective efficacy of 15-PGDH haploinsufficiency (Hpgd+/−) in the MPTP-induced mouse PD model.(F) Male and female Hpgd heterozygous mice are protected from MPTP-induced motor deficits in the rotarod test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(G) Hpgd heterozygous mice are protected from MPTP-induced hindlimb clasping behavior (n = 15-22 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(H) Hpgd heterozygous mice are protected from MPTP-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(I) Experimental procedure for evaluating the neuroprotective efficacy of SW033291 in the intranigral LPS-injected mouse PD model.(J) SW033291 protects mice from intranigral LPS-induced motor deficits in the rotarod test (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(K) SW033291 protects mice from mice from intranigral LPS-induced hindlimb clasping behavior (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(L) SW033291 protects mice from intranigral LPS-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 6-12 per group, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(M) Representative TH-stained images of the substantia nigra of mice treated with vehicle or MPTP in the absence or presence of SW033291 (scale bar = 100 μm) (N) Quantification of TH fluorescence signal shows that SW033291 treatment dose-dependently protects MPTP-exposed mice from loss of TH intensity (n = 4 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(O) Representative TH-stained images of the substantia nigra from Hpgd heterozygous mice and their wild-type littermates treated with vehicle or MPTP (scale bar = 100 μm) (P) Quantification of TH fluorescence signal shows that Hpgd heterozygous mice are protected from MPTP-induced loss of TH intensity (n = 4 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(Q) Representative TH-stained images of the substantia nigra from mice treated with vehicle or LPS, in the absence or presence of SW033291 (scale bar = 200 μm)(R) Quantification of TH fluorescence signal shows that SW033291 treatment protects mice from intranigral LPS-induced loss of TH intensity (n = 3 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).

We next investigated the neuroprotective potential of 15-PGDH inhibition in two PD models, systemic MPTP administration and intranigral LPS injection. In the MPTP study, mice underwent a one-week acclimation period prior to receiving intraperitoneal injections of the 15-PGDH inhibitor SW033291 (0.5 or 5 mg/kg, twice daily) for two days, after which treatment was continued alongside daily MPTP injections (30 mg/kg) for seven more days. SW033291 significantly reduced 15-PGDH activity and increased PGE2 levels in the substantia nigra without affecting body weight (Fig. S1A–C). Behavioral assessments were conducted 2–3 h after the final MPTP injection (Fig. 2A). MPTP administration induced significant motor deficits, evidenced by reduced rotarod latency (Fig. 2B), increased hind-limb clasping (Fig. 2C and Fig. S1D and E), and prolonged pole test completion times (Fig. 2D). Notably, SW033291 treatment dose-dependently prevented all these impairments, preserving motor performance to near non-MPTP exposed levels (Fig. 2B–D and Fig. S1E).Fig. 2Pharmacological and genetic inhibition of 15-PGDH ameliorates motor deficits and dopaminergic neuronal loss in MPTP- and intranigral LPS-induced PD models(A) Experimental procedure for evaluating neuroprotective efficacy of SW033291 in the MPTP-induced mouse PD model.(B) SW033291 protects MPTP-exposed mice from motor deficits in the rotarod test (n = 10-11 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(C) SW033291 dose-dependently protects MPTP-exposed mice from hindlimb clasping behavior (n = 10-11 per group, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(D) SW033291 protects MPTP-exposed mice from increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 10-11 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(E) Experimental procedure for evaluating neuroprotective efficacy of 15-PGDH haploinsufficiency (Hpgd+/−) in the MPTP-induced mouse PD model.(F) Male and female Hpgd heterozygous mice are protected from MPTP-induced motor deficits in the rotarod test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(G) Hpgd heterozygous mice are protected from MPTP-induced hindlimb clasping behavior (n = 15-22 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(H) Hpgd heterozygous mice are protected from MPTP-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(I) Experimental procedure for evaluating the neuroprotective efficacy of SW033291 in the intranigral LPS-injected mouse PD model.(J) SW033291 protects mice from intranigral LPS-induced motor deficits in the rotarod test (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(K) SW033291 protects mice from mice from intranigral LPS-induced hindlimb clasping behavior (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(L) SW033291 protects mice from intranigral LPS-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 6-12 per group, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(M) Representative TH-stained images of the substantia nigra of mice treated with vehicle or MPTP in the absence or presence of SW033291 (scale bar = 100 μm) (N) Quantification of TH fluorescence signal shows that SW033291 treatment dose-dependently protects MPTP-exposed mice from loss of TH intensity (n = 4 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(O) Representative TH-stained images of the substantia nigra from Hpgd heterozygous mice and their wild-type littermates treated with vehicle or MPTP (scale bar = 100 μm) (P) Quantification of TH fluorescence signal shows that Hpgd heterozygous mice are protected from MPTP-induced loss of TH intensity (n = 4 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(Q) Representative TH-stained images of the substantia nigra from mice treated with vehicle or LPS, in the absence or presence of SW033291 (scale bar = 200 μm)(R) Quantification of TH fluorescence signal shows that SW033291 treatment protects mice from intranigral LPS-induced loss of TH intensity (n = 3 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).

We next evaluated whether these same interventions also conferred neuroprotection to dopaminergic neurons in these models. Immunohistochemical analysis revealed a significant reduction in tyrosine hydroxylase (TH), a marker of dopaminergic neurons, in the substantia nigra of MPTP-treated mice compared to vehicle-treated controls (Fig. 2M and N). Strikingly, administration of the 15-PGDH inhibitor SW033291 enhanced dopaminergic neuron survival in a dose-dependent manner (Fig. 2M and N). MPTP-injected animals also exhibited pronounced neuronal loss in the striatum, which was dose-dependently attenuated by 15-PGDH inhibition (Fig. S2A and B). Western blot analysis confirmed these findings, showing that 15-PGDH inhibition significantly preserved TH expression in both the substantia nigra and striatum of MPTP-exposed mice (Fig. S2C and D). Hpgd +/- mice were also protected from MPTP-induced dopaminergic neuronal loss (Fig. 2O and P, Fig. S2E and F). Western blot analysis confirmed TH levels, which declined in MPTP-exposed Hpgd+/+ mice and remained at baseline in Hpgd +/- mice (Fig. S2G and H). Lastly, 15-PGDH inhibition also maintained TH levels in the intranigral LPS model (Fig. 2Q and R).

We next investigated the neuroprotective potential of 15-PGDH inhibition in two PD models, systemic MPTP administration and intranigral LPS injection. In the MPTP study, mice underwent a one-week acclimation period prior to receiving intraperitoneal injections of the 15-PGDH inhibitor SW033291 (0.5 or 5 mg/kg, twice daily) for two days, after which treatment was continued alongside daily MPTP injections (30 mg/kg) for seven more days. SW033291 significantly reduced 15-PGDH activity and increased PGE2 levels in the substantia nigra without affecting body weight (Fig. S1A–C). Behavioral assessments were conducted 2–3 h after the final MPTP injection (Fig. 2A). MPTP administration induced significant motor deficits, evidenced by reduced rotarod latency (Fig. 2B), increased hind-limb clasping (Fig. 2C and Fig. S1D and E), and prolonged pole test completion times (Fig. 2D). Notably, SW033291 treatment dose-dependently prevented all these impairments, preserving motor performance to near non-MPTP exposed levels (Fig. 2B–D and Fig. S1E).Fig. 2Pharmacological and genetic inhibition of 15-PGDH ameliorates motor deficits and dopaminergic neuronal loss in MPTP- and intranigral LPS-induced PD models(A) Experimental procedure for evaluating neuroprotective efficacy of SW033291 in the MPTP-induced mouse PD model.(B) SW033291 protects MPTP-exposed mice from motor deficits in the rotarod test (n = 10-11 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(C) SW033291 dose-dependently protects MPTP-exposed mice from hindlimb clasping behavior (n = 10-11 per group, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(D) SW033291 protects MPTP-exposed mice from increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 10-11 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(E) Experimental procedure for evaluating neuroprotective efficacy of 15-PGDH haploinsufficiency (Hpgd+/−) in the MPTP-induced mouse PD model.(F) Male and female Hpgd heterozygous mice are protected from MPTP-induced motor deficits in the rotarod test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(G) Hpgd heterozygous mice are protected from MPTP-induced hindlimb clasping behavior (n = 15-22 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(H) Hpgd heterozygous mice are protected from MPTP-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(I) Experimental procedure for evaluating the neuroprotective efficacy of SW033291 in the intranigral LPS-injected mouse PD model.(J) SW033291 protects mice from intranigral LPS-induced motor deficits in the rotarod test (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(K) SW033291 protects mice from mice from intranigral LPS-induced hindlimb clasping behavior (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(L) SW033291 protects mice from intranigral LPS-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 6-12 per group, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(M) Representative TH-stained images of the substantia nigra of mice treated with vehicle or MPTP in the absence or presence of SW033291 (scale bar = 100 μm) (N) Quantification of TH fluorescence signal shows that SW033291 treatment dose-dependently protects MPTP-exposed mice from loss of TH intensity (n = 4 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(O) Representative TH-stained images of the substantia nigra from Hpgd heterozygous mice and their wild-type littermates treated with vehicle or MPTP (scale bar = 100 μm) (P) Quantification of TH fluorescence signal shows that Hpgd heterozygous mice are protected from MPTP-induced loss of TH intensity (n = 4 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(Q) Representative TH-stained images of the substantia nigra from mice treated with vehicle or LPS, in the absence or presence of SW033291 (scale bar = 200 μm)(R) Quantification of TH fluorescence signal shows that SW033291 treatment protects mice from intranigral LPS-induced loss of TH intensity (n = 3 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).

As α-synuclein–driven models most closely recapitulate key features of human PD, we further tested the efficacy of 15-PGDH inhibition in this mouse model. Seven-week-old C57BL/6J mice were acclimated for one week and then pretreated with either vehicle or the 15-PGDH inhibitor SW033291 (5 mg/kg, intraperitoneally, twice daily) for two days prior to surgery. Mice subsequently received intranigral injections of either AAV-GFP (comparator) or AAV–α-synuclein (disease model), followed by intra-VTA injection of PBS (comparator) or α-synuclein preformed fibrils (PFFs) (disease model). Behavioral assessments were performed at designated time points, and brains were collected for biochemical and histological analysis at 33 days post-injection (Fig. 4A). Consistent with our previous reports [39,40], the combination of AAV-α-synuclein and PFF administration induced pronounced motor dysfunction, reflected by shortened rotarod latency (Fig. 4B), elevated hind-limb clasping scores (Fig. 4C), and prolonged pole test times (Fig. 4D). SW033291 treatment prevented these impairments and maintained motor performance comparable to AAV-eGFP/PBS controls, with no change in body weight (Fig. 4B–D and Fig. S8A).Fig. 415-PGDH inhibition ameliorates motor deficits, dopaminergic neuronal loss, and oxidative stress in the ɑ-synuclein mouse PD model(A) Experimental procedure for evaluating the efficacy of SW033291 in the α-synuclein mouse model.(B) SW033291 protects mice from AAV-ɑ-synuclein/PFF-induced motor deficits in the rotarod test (n = 5-6 per group, ∗∗p < 0.01, one-way ANOVA and Tukey's post hoc analysis).(C) SW033291 protects mice from AAV-ɑ-synuclein/PFF-induced from hindlimb clasping behavior (n = 5-6 per group, ∗p < 0.05, ∗∗p < 0.01, one-way ANOVA and Tukey's post hoc analysis).(D) SW033291 protects mice from AAV-ɑ-synuclein/PFF-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 5-6 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(E) Representative images and quantification of TH immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced loss of TH intensity (scale bar = 800 μm, n = 5-6 per group, ∗p < 0.05, one-way ANOVA and Tukey's post hoc analysis).(F) Representative images and quantification of 4-HNE immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in 4-HNE (scale bar = 100 μm, n = 5-6 per group, ∗p < 0.05, one-way ANOVA and Tukey's post hoc analysis).(G) Representative images and quantification of 3-NT immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in 3-NT (scale bar = 100 μm, n = 5-6 per group, ∗p < 0.05, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(H) RT-qPCR analysis shows that SW033291 treatment protected mice from administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in Cybb mRNA expression (n = 3 per group, ∗∗p < 0.01, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(I) Representative images and quantification of Iba1 immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in Iba1 (scale bar = 200 μm, n = 5-6 per group, ∗∗p < 0.01, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(J) Representative images and quantification of GFAP immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in GFAP (scale bar = 300 μm, n = 5-6 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(K) RT-qPCR analysis shows that SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in Lcn2 mRNA expression (n = 3 per group, ∗p < 0.05, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(L) Representative images and quantitative analysis of p-α-synuclein immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 treatment had no effect on p-α-synuclein accumulation in AAV-ɑ-synuclein/PFF mice (scale bar = 20 μm, n = 5-6 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).

We next examined neuroprotective effects on dopaminergic neurons and reduced oxidative stress in this model. TH, 4-HNE, and 3-NT immunostaining demonstrated that combined administration of AAV-α-synuclein and PFF resulted in decreased TH intensity and elevated oxidative stress markers in the substantia nigra, with SW033291 treatment markedly attenuating these alternations (Fig. 4E–G) along with markedly reducing Cybb mRNA levels (Fig. 4H).

Notably, SW033291 treatment did not significantly reduce phosphorylated α-synuclein accumulation (Fig. 4L), indicating that 15-PGDH inhibition primarily alleviates oxidative stress and neuroinflammation independent of α-synuclein pathology, and that this is sufficient to eliminate the motor impairment associated with PD. This finding draws a notable parallel with our recent report demonstrating the ability of 15-PGDH inhibition to prevent pathological features and cognitive impairment in an amyloid-driven mouse model of AD without affecting amyloid pathology [12].

We next assessed efficacy in a second PD model: intranigral LPS administration [14,21,22]. Mice pretreated with SW033291 (5 mg/kg, twice daily) or vehicle received bilateral substantia nigra LPS injections, with continued SW033291 treatment (Fig. 2I). LPS administration did not affect body weight (Fig. S1L). LPS impaired motor function one week later, indicated by reduced latency to fall on the accelerating rotarod (Fig. 2J), increased hind-limb clasping (Fig. 2K and Fig. S1M), and prolonged pole test duration (Fig. 2L). Notably, SW033291 prevented these deficits (Fig. 2J–L).

Here, we establish the therapeutic potential of 15-PGDH inhibition in PD, a mechanistically different neurodegenerative disorder. We observed significant 15-PGDH upregulation in the substantia nigra of postmortem human PD brains as well as three different mouse models of PD: systemic 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) [13], intranigral lipopolysaccharide (LPS) [14], intrastriatal AAV-α-synuclein with intra-ventral tegmental area α-synuclein preformed fibrils (PFFs) [15]. We utilized both pharmacologic inhibition (SW033291 treatment) and genetic partial reduction of 15-PGDH. SW033291 has high specificity for inhibiting 15-PGDH, with IC50 < 1 nM and no interaction with other structurally related short-chain dehydrogenases [6]. Nonetheless, to control for any previously unidentified off-target drug effects, we compared 15-PGDH inhibition with SW033291 versus genetic inhibition of 15-PGDH via knockout of one Hpgd allele. Both drug and genetic inhibition of 15-PGDH showed complete concordance in protecting from motor deficits, dopaminergic cell loss, BBB degradation, and induction of multiple molecular markers of inflammation and ROS-induced tissue damage. This identification of SW033291 and genetic reduction of Hpgd in protecting the substantia nigra is consistent with our prior findings of complete concordance of SW033291 and genetic deletion of Hpgd in AD and TBI models [12].

We next examined whether genetic partial reduction of 15-PGDH activity could also provide neuroprotection against MPTP-induced behavioral deficits. Eight-week-old male and female Hpgd+/+ (wild type: WT) and Hpgd+/− (haploinsufficient) mice received daily MPTP injections (30 mg/kg) for seven days, followed by behavioral assessments 2-3 h after the final injection (Fig. 2E). MPTP did not affect body weight in either genotype (Fig. S1F). After MPTP exposure, heterozygous Hpgd ±/- mice exhibited superior motor performance compared to WT Hpgd+/+ littermates, demonstrating longer rotarod retention (Fig. 2F), reduced hindlimb clasping (Fig. 2G and Fig. S1G), and faster pole test completion (Fig. 2H).

We next investigated the neuroprotective potential of 15-PGDH inhibition in two PD models, systemic MPTP administration and intranigral LPS injection. In the MPTP study, mice underwent a one-week acclimation period prior to receiving intraperitoneal injections of the 15-PGDH inhibitor SW033291 (0.5 or 5 mg/kg, twice daily) for two days, after which treatment was continued alongside daily MPTP injections (30 mg/kg) for seven more days. SW033291 significantly reduced 15-PGDH activity and increased PGE2 levels in the substantia nigra without affecting body weight (Fig. S1A–C). Behavioral assessments were conducted 2–3 h after the final MPTP injection (Fig. 2A). MPTP administration induced significant motor deficits, evidenced by reduced rotarod latency (Fig. 2B), increased hind-limb clasping (Fig. 2C and Fig. S1D and E), and prolonged pole test completion times (Fig. 2D). Notably, SW033291 treatment dose-dependently prevented all these impairments, preserving motor performance to near non-MPTP exposed levels (Fig. 2B–D and Fig. S1E).Fig. 2Pharmacological and genetic inhibition of 15-PGDH ameliorates motor deficits and dopaminergic neuronal loss in MPTP- and intranigral LPS-induced PD models(A) Experimental procedure for evaluating neuroprotective efficacy of SW033291 in the MPTP-induced mouse PD model.(B) SW033291 protects MPTP-exposed mice from motor deficits in the rotarod test (n = 10-11 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(C) SW033291 dose-dependently protects MPTP-exposed mice from hindlimb clasping behavior (n = 10-11 per group, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(D) SW033291 protects MPTP-exposed mice from increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 10-11 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(E) Experimental procedure for evaluating neuroprotective efficacy of 15-PGDH haploinsufficiency (Hpgd+/−) in the MPTP-induced mouse PD model.(F) Male and female Hpgd heterozygous mice are protected from MPTP-induced motor deficits in the rotarod test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(G) Hpgd heterozygous mice are protected from MPTP-induced hindlimb clasping behavior (n = 15-22 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(H) Hpgd heterozygous mice are protected from MPTP-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(I) Experimental procedure for evaluating the neuroprotective efficacy of SW033291 in the intranigral LPS-injected mouse PD model.(J) SW033291 protects mice from intranigral LPS-induced motor deficits in the rotarod test (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(K) SW033291 protects mice from mice from intranigral LPS-induced hindlimb clasping behavior (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(L) SW033291 protects mice from intranigral LPS-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 6-12 per group, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(M) Representative TH-stained images of the substantia nigra of mice treated with vehicle or MPTP in the absence or presence of SW033291 (scale bar = 100 μm) (N) Quantification of TH fluorescence signal shows that SW033291 treatment dose-dependently protects MPTP-exposed mice from loss of TH intensity (n = 4 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(O) Representative TH-stained images of the substantia nigra from Hpgd heterozygous mice and their wild-type littermates treated with vehicle or MPTP (scale bar = 100 μm) (P) Quantification of TH fluorescence signal shows that Hpgd heterozygous mice are protected from MPTP-induced loss of TH intensity (n = 4 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(Q) Representative TH-stained images of the substantia nigra from mice treated with vehicle or LPS, in the absence or presence of SW033291 (scale bar = 200 μm)(R) Quantification of TH fluorescence signal shows that SW033291 treatment protects mice from intranigral LPS-induced loss of TH intensity (n = 3 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).

We next investigated the neuroprotective potential of 15-PGDH inhibition in two PD models, systemic MPTP administration and intranigral LPS injection. In the MPTP study, mice underwent a one-week acclimation period prior to receiving intraperitoneal injections of the 15-PGDH inhibitor SW033291 (0.5 or 5 mg/kg, twice daily) for two days, after which treatment was continued alongside daily MPTP injections (30 mg/kg) for seven more days. SW033291 significantly reduced 15-PGDH activity and increased PGE2 levels in the substantia nigra without affecting body weight (Fig. S1A–C). Behavioral assessments were conducted 2–3 h after the final MPTP injection (Fig. 2A). MPTP administration induced significant motor deficits, evidenced by reduced rotarod latency (Fig. 2B), increased hind-limb clasping (Fig. 2C and Fig. S1D and E), and prolonged pole test completion times (Fig. 2D). Notably, SW033291 treatment dose-dependently prevented all these impairments, preserving motor performance to near non-MPTP exposed levels (Fig. 2B–D and Fig. S1E).Fig. 2Pharmacological and genetic inhibition of 15-PGDH ameliorates motor deficits and dopaminergic neuronal loss in MPTP- and intranigral LPS-induced PD models(A) Experimental procedure for evaluating neuroprotective efficacy of SW033291 in the MPTP-induced mouse PD model.(B) SW033291 protects MPTP-exposed mice from motor deficits in the rotarod test (n = 10-11 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(C) SW033291 dose-dependently protects MPTP-exposed mice from hindlimb clasping behavior (n = 10-11 per group, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(D) SW033291 protects MPTP-exposed mice from increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 10-11 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(E) Experimental procedure for evaluating neuroprotective efficacy of 15-PGDH haploinsufficiency (Hpgd+/−) in the MPTP-induced mouse PD model.(F) Male and female Hpgd heterozygous mice are protected from MPTP-induced motor deficits in the rotarod test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(G) Hpgd heterozygous mice are protected from MPTP-induced hindlimb clasping behavior (n = 15-22 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(H) Hpgd heterozygous mice are protected from MPTP-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 15-22 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis, males and females are shown as circles and triangles, respectively).(I) Experimental procedure for evaluating the neuroprotective efficacy of SW033291 in the intranigral LPS-injected mouse PD model.(J) SW033291 protects mice from intranigral LPS-induced motor deficits in the rotarod test (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(K) SW033291 protects mice from mice from intranigral LPS-induced hindlimb clasping behavior (n = 6-12 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(L) SW033291 protects mice from intranigral LPS-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 6-12 per group, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(M) Representative TH-stained images of the substantia nigra of mice treated with vehicle or MPTP in the absence or presence of SW033291 (scale bar = 100 μm) (N) Quantification of TH fluorescence signal shows that SW033291 treatment dose-dependently protects MPTP-exposed mice from loss of TH intensity (n = 4 per group, ∗p < 0.05, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(O) Representative TH-stained images of the substantia nigra from Hpgd heterozygous mice and their wild-type littermates treated with vehicle or MPTP (scale bar = 100 μm) (P) Quantification of TH fluorescence signal shows that Hpgd heterozygous mice are protected from MPTP-induced loss of TH intensity (n = 4 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(Q) Representative TH-stained images of the substantia nigra from mice treated with vehicle or LPS, in the absence or presence of SW033291 (scale bar = 200 μm)(R) Quantification of TH fluorescence signal shows that SW033291 treatment protects mice from intranigral LPS-induced loss of TH intensity (n = 3 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).

We next assessed efficacy in a second PD model: intranigral LPS administration [14,21,22]. Mice pretreated with SW033291 (5 mg/kg, twice daily) or vehicle received bilateral substantia nigra LPS injections, with continued SW033291 treatment (Fig. 2I). LPS administration did not affect body weight (Fig. S1L). LPS impaired motor function one week later, indicated by reduced latency to fall on the accelerating rotarod (Fig. 2J), increased hind-limb clasping (Fig. 2K and Fig. S1M), and prolonged pole test duration (Fig. 2L). Notably, SW033291 prevented these deficits (Fig. 2J–L).

As α-synuclein–driven models most closely recapitulate key features of human PD, we further tested the efficacy of 15-PGDH inhibition in this mouse model. Seven-week-old C57BL/6J mice were acclimated for one week and then pretreated with either vehicle or the 15-PGDH inhibitor SW033291 (5 mg/kg, intraperitoneally, twice daily) for two days prior to surgery. Mice subsequently received intranigral injections of either AAV-GFP (comparator) or AAV–α-synuclein (disease model), followed by intra-VTA injection of PBS (comparator) or α-synuclein preformed fibrils (PFFs) (disease model). Behavioral assessments were performed at designated time points, and brains were collected for biochemical and histological analysis at 33 days post-injection (Fig. 4A). Consistent with our previous reports [39,40], the combination of AAV-α-synuclein and PFF administration induced pronounced motor dysfunction, reflected by shortened rotarod latency (Fig. 4B), elevated hind-limb clasping scores (Fig. 4C), and prolonged pole test times (Fig. 4D). SW033291 treatment prevented these impairments and maintained motor performance comparable to AAV-eGFP/PBS controls, with no change in body weight (Fig. 4B–D and Fig. S8A).Fig. 415-PGDH inhibition ameliorates motor deficits, dopaminergic neuronal loss, and oxidative stress in the ɑ-synuclein mouse PD model(A) Experimental procedure for evaluating the efficacy of SW033291 in the α-synuclein mouse model.(B) SW033291 protects mice from AAV-ɑ-synuclein/PFF-induced motor deficits in the rotarod test (n = 5-6 per group, ∗∗p < 0.01, one-way ANOVA and Tukey's post hoc analysis).(C) SW033291 protects mice from AAV-ɑ-synuclein/PFF-induced from hindlimb clasping behavior (n = 5-6 per group, ∗p < 0.05, ∗∗p < 0.01, one-way ANOVA and Tukey's post hoc analysis).(D) SW033291 protects mice from AAV-ɑ-synuclein/PFF-induced increased time to orient downward (time to turn) and total time to descend to the floor (time to go down) in the pole test (n = 5-6 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(E) Representative images and quantification of TH immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced loss of TH intensity (scale bar = 800 μm, n = 5-6 per group, ∗p < 0.05, one-way ANOVA and Tukey's post hoc analysis).(F) Representative images and quantification of 4-HNE immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in 4-HNE (scale bar = 100 μm, n = 5-6 per group, ∗p < 0.05, one-way ANOVA and Tukey's post hoc analysis).(G) Representative images and quantification of 3-NT immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in 3-NT (scale bar = 100 μm, n = 5-6 per group, ∗p < 0.05, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(H) RT-qPCR analysis shows that SW033291 treatment protected mice from administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in Cybb mRNA expression (n = 3 per group, ∗∗p < 0.01, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(I) Representative images and quantification of Iba1 immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in Iba1 (scale bar = 200 μm, n = 5-6 per group, ∗∗p < 0.01, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(J) Representative images and quantification of GFAP immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in GFAP (scale bar = 300 μm, n = 5-6 per group, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).(K) RT-qPCR analysis shows that SW033291 administration protected mice from AAV-ɑ-synuclein/PFF-induced increase in Lcn2 mRNA expression (n = 3 per group, ∗p < 0.05, ∗∗∗p < 0.001, one-way ANOVA and Tukey's post hoc analysis).(L) Representative images and quantitative analysis of p-α-synuclein immunoreactivity in the substantia nigra of mice treated with vehicle or SW033291. SW033291 treatment had no effect on p-α-synuclein accumulation in AAV-ɑ-synuclein/PFF mice (scale bar = 20 μm, n = 5-6 per group, ∗∗∗∗p < 0.0001, one-way ANOVA and Tukey's post hoc analysis).

Notably, SW033291 treatment did not significantly reduce phosphorylated α-synuclein accumulation (Fig. 4L), indicating that 15-PGDH inhibition primarily alleviates oxidative stress and neuroinflammation independent of α-synuclein pathology, and that this is sufficient to eliminate the motor impairment associated with PD. This finding draws a notable parallel with our recent report demonstrating the ability of 15-PGDH inhibition to prevent pathological features and cognitive impairment in an amyloid-driven mouse model of AD without affecting amyloid pathology [12].

Postmortem human substantia nigra tissue from PD subjects displayed elevated mRNA for Hpgd (the gene encoding 15-PGDH) (Fig. 1A; Table S1). Similarly, exposure of mice to the neurotoxin MPTP, which selectively targets the substantia nigra dopaminergic neurons implicated in PD [[17], [18], [19], [20]], also increased Hpgd mRNA levels, 15-PGDH protein levels, and 15-PGDH enzymatic activity in the substantia nigra (Fig. 1B–D). Hpgd mRNA was similarly elevated in mice receiving intranigral LPS, intrastriatal AAV-α-synuclein with intra-ventral tegmental area α-synuclein PFFs, and intrastriatal PFF (Fig. 1E–G). Lastly, Hpgd mRNA induction was observed in SH-SY5Y cells that were transduced to overexpress wild-type α-synuclein, a cellular model of PD (Fig. 1H). Collectively, our findings establish 15-PGDH elevation as a consistent feature across human PD and preclinical models.Fig. 115-PGDH is elevated in Parkinson's disease (PD) in cellular and animal models and human brains(A) HPGD mRNA is increased in the substantia nigra of human PD brain, relative to control subjects (n = 5-8 per group, ∗p < 0.05, unpaired t-test).(B) Hpgd mRNA is increased in the substantia nigra of MPTP-exposed mice, relative to vehicle-treated mice (n = 3 per group, ∗p < 0.05, unpaired t-test).(C) 15-PGDH protein expression is increased in the substantia nigra of MPTP-exposed mice, relative to vehicle-treated mice (n = 3 per group, ∗p < 0.05, unpaired t-test).(D) 15-PGDH activity is elevated in the substantia nigra of MPTP-exposed mice, relative to vehicle-treated animals (n = 5 per group, ∗∗p < 0.01, unpaired t-test).(E) Hpgd mRNA expression is elevated in mice expose to intranigral LPS administration, relative to vehicle-treated mice (n = 3 per group, ∗p < 0.05, unpaired t-test).(F) Hpgd mRNA expression is increased in mice with intrastriatal AAV-ɑ-synuclein combined with intra-ventral tegmental area injection of ɑ-synuclein preformed fibrils (PFF), relative to AAV-eGFP and PBS-exposed mice (n = 5-6 per group, ∗p < 0.05, unpaired t-test)(G) Hpgd mRNA expression is increased in mice with intrastriatal fibrillar alpha-synuclein administration, relative to PBS-exposed mice (n = 3 per group, ∗p < 0.05, unpaired t-test).(H) HPGD mRNA is increased in the SH-SY5Y α-syn BiFC cell line, which stably expresses wild-type alpha-synuclein, relative to control SH-SY5Y cells (n = 6 per group, ∗∗∗∗p < 0.0001, unpaired t-test).

Further investigation into the regulatory mechanisms governing Hpgd expression will be important for understanding the upstream processes that drive 15-PGDH elevation in PD. While in vitro experiments have demonstrated that PGE2 can reduce LCN2 and NOX2 expression, in vivo studies will be necessary to fully interrogate the interactions among these signaling pathways. Additional experiments, including in vivo EP4 knockout and inhibition, NOX2 silencing, and LCN2 neutralization, will further validate these findings and clarify their mechanistic and therapeutic implications. In addition, further validation in additional human patient samples will be important in future studies.

In conclusion, our findings establish 15-PGDH as a robust therapeutic target in PD. 15-PGDH levels are elevated in the brains of PD mice and humans, and both genetic and pharmacologic inhibition produces marked therapeutic efficacy in mouse PD models. Previous work from our team demonstrated high CNS penetration of SW033291, with sustained drug levels in both brain and plasma sustained for up to 6 h, and, as shown in the present study (Fig. S1A), near-complete ablation of 15-PGDH enzyme activity in the brain. The clinical safety of 15-PGDH inhibition is supported by the absence of toxicity in a recent human phase 1 trial of the 15-PGDH inhibitor MF-300 [56], as well as by findings from humans with biallelic inactivating mutations of 15-PGDH, in whom the only consistently observed phenotype is congenital digital clubbing [57]. Encouragingly, both pharmaceutical and biotechnology companies have initiated development of 15-PGDH inhibitors for peripheral indications, and inhibitor MF-300 has already completed human phase 1 trials. Our results now provide the rationale to repurpose such agents for the treatment of PD.