Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 55 of the 55 references it cites were found in a registry or an index.

Insulin Resistance Promotes Parkinson’s Disease through Aberrant Expression of α-Synuclein, Mitochondrial Dysfunction, and Deregulation of the Polo-Like Kinase 2 Signaling

Open access

Abstract

Background: Insulin resistance (IR), considered a hallmark of diabetes at the cellular level, is implicated in pre-diabetes, results in type 2 diabetes, and negatively affects mitochondrial function. Diabetes is increasingly associated with enhanced risk of developing Parkinson’s disease (PD); however, the underlying mechanism remains unclear. This study investigated the probable culpability of IR in the pathogenesis of PD. Methods: Using MitoPark mice in vivo models, diabetes was induced by a high-fat diet in the in vivo models, and IR was induced by protracted pulse-stimulation with 100 nM insulin treatment of neuronal cells, in vitro to determine the molecular mechanism(s) underlying altered cellular functions in PD, including mitochondrial dysfunction and α-synuclein (SNCA) aberrant expression. Findings: We observed increased SNCA expression in the dopaminergic (DA) neurons of both the wild-type and diabetic MitoPark mice, coupled with enhanced degeneration of DA neurons in the diabetic MitoPark mice. Ex vivo, in differentiated human DA neurons, IR was associated with increased SNCA and reactive oxygen species (ROS) levels, as well as mitochondrial depolarization. Moreover, we demonstrated concomitant hyperactivation of polo-like kinase-2 (PLK2), and upregulated p-SNCA (Ser129) and proteinase K-resistant SNCA proteins level in IR SH-SY5Y cells, however the inhibition of PLK2 reversed IR-related increases in phosphorylated and total SNCA. Similarly, the overexpression of peroxisome proliferator-activated receptor-γ coactivator 1-alpha (PGC)-1α suppressed ROS production, repressed PLK2 hyperactivity, and resulted in downregulation of total and Ser129-phosphorylated SNCA in the IR SH-SY5Y cells. Conclusions: These findings demonstrate that IR-associated diabetes promotes the development and progression of PD through PLK2-mediated mitochondrial dysfunction, upregulated ROS production, and enhanced SNCA signaling, suggesting the therapeutic targetability of PLK2 and/or SNCA as potential novel disease-modifying strategies in patients with PD.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/cells9030740

Cites

55 of 55 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Patients’ genomic landscape indicates a nosological association between diabetes and Parkinson’s disease (PD), in silico. PD and insulin resistance (IR)-related transcript expression heatmaps from transcriptional analyses of (A) putamen (Homo sapiens, A-AFFY-33, AFFY_HG_U133A, E-GEOD-20291, 30 samples, 22,283 genes), (B) prefrontal area 9 (Homo sapiens, A-AFFY-33, AFFY_HG_U133A, E-GEOD-20168, 30 samples, 22,283 genes), and (C) whole substantia nigra (Homo sapiens, A-AFFY-33, AFFY_HG_U133A, E-GEOD-20292, 26 samples, 22,283 genes) in PD. Rows are centered; unit variance scaling is applied to rows. Both rows and columns are clustered using correlation distance and average linkage. (D) Visualization of the functional protein–protein interaction of PD and IR proteins as depicted by Search Tool for the Retrieval of Interacting Genes/Proteins database (STRINGdb).
Figure 1 Patients’ genomic landscape indicates a nosological association between diabetes and Parkinson’s disease (PD), in silico. PD and insulin resistance (IR)-related transcript expression heatmaps from transcriptional analyses of (A) putamen (Homo sapiens, A-AFFY-33, AFFY_HG_U133A, E-GEOD-20291, 30 samples, 22,283 genes), (B) prefrontal area 9 (Homo sapiens, A-AFFY-33, AFFY_HG_U133A, E-GEOD-20168, 30 samples, 22,283 genes), and (C) whole substantia nigra (Homo sapiens, A-AFFY-33, AFFY_HG_U133A, E-GEOD-20292, 26 samples, 22,283 genes) in PD. Rows are centered; unit variance scaling is applied to rows. Both rows and columns are clustered using correlation distance and average linkage. (D) Visualization of the functional protein–protein interaction of PD and IR proteins as depicted by Search Tool for the Retrieval of Interacting Genes/Proteins database (STRINGdb).CC BY · PubMed Central
Figure 2 Diabetes contributes to the progression of PD by amplifying the loss of tyrosine hydroxylase (TH)+ dopaminergic neurons and enhancing α-synuclein (SNCA) expression, in vivo. (A) Representative IHC images of TH immunoreactivity in the midbrain substantia nigra of high-fat diet-induced diabetes C57BL/6 mice compared to the control mice. (B) Representative IHC images of SNCA immunostaining in the substantia nigra of diabetic MitoPark mice, compared to their control non-diabetic counterparts. (C) Representative immunofluorescence images of α-syn/SNCA (red) and TH (green) expression in the midbrain of control and diabetic MitoPark mice, captured by deconvolution microscopy. 4′,6′-diamidino-2-phenylindole (DAPI) serve as nuclear staining (left). Graphical representation of the SNCA fluorescence intensity in TH-positive dopaminergic (DA) neurons from diabetic MitoPark mice (control: 100 ± 6.6%, diabetic: 137.7 ± 8.2%, nTH + neurons = 24, p < 0.01) (right). (D) Representative images (left) and graphical representation (right) of SNCA immunoreactivity in the midbrain tectum of control and high-fat diet-induced diabetes MitoPark mice (control: 100 ± 8.7%, diabetic: 126 ± 7.6%, n = 3). (E) Representative IHC images (left) and graphical representation (right) of SNCA expression in the substantia nigra of control and high-fat diet-induced diabetes MitoPark mice (control: 100 ± 15.7%, diabetic: 139.2 ± 10.9%, n = 3). *, p < 0.05, **, p < 0.01; control, non-diabetic mice. (F) Diabetic MitoPark mice took a longer time to walk the beam compared with control MitoPark mice (control: 5.43 ± 0.46, diabetic: 6.67 ± 0.39 s, n = 3). (G) Diabetic MitoPark mice stayed a shorter length of time on the rotarod tests (control: 8.3 ± 2.5, diabetic: 5.3 ± 1 s, n = 3).
Figure 2 Diabetes contributes to the progression of PD by amplifying the loss of tyrosine hydroxylase (TH)+ dopaminergic neurons and enhancing α-synuclein (SNCA) expression, in vivo. (A) Representative IHC images of TH immunoreactivity in the midbrain substantia nigra of high-fat diet-induced diabetes C57BL/6 mice compared to the control mice. (B) Representative IHC images of SNCA immunostaining in the substantia nigra of diabetic MitoPark mice, compared to their control non-diabetic counterparts. (C) Representative immunofluorescence images of α-syn/SNCA (red) and TH (green) expression in the midbrain of control and diabetic MitoPark mice, captured by deconvolution microscopy. 4′,6′-diamidino-2-phenylindole (DAPI) serve as nuclear staining (left). Graphical representation of the SNCA fluorescence intensity in TH-positive dopaminergic (DA) neurons from diabetic MitoPark mice (control: 100 ± 6.6%, diabetic: 137.7 ± 8.2%, nTH + neurons = 24, p < 0.01) (right). (D) Representative images (left) and graphical representation (right) of SNCA immunoreactivity in the midbrain tectum of control and high-fat diet-induced diabetes MitoPark mice (control: 100 ± 8.7%, diabetic: 126 ± 7.6%, n = 3). (E) Representative IHC images (left) and graphical representation (right) of SNCA expression in the substantia nigra of control and high-fat diet-induced diabetes MitoPark mice (control: 100 ± 15.7%, diabetic: 139.2 ± 10.9%, n = 3). *, p < 0.05, **, p < 0.01; control, non-diabetic mice. (F) Diabetic MitoPark mice took a longer time to walk the beam compared with control MitoPark mice (control: 5.43 ± 0.46, diabetic: 6.67 ± 0.39 s, n = 3). (G) Diabetic MitoPark mice stayed a shorter length of time on the rotarod tests (control: 8.3 ± 2.5, diabetic: 5.3 ± 1 s, n = 3).CC BY · PubMed Central
Figure 3 Insulin resistance is associated with increased expression of SNCA and mitochondrial dysfunction in human differentiated dopaminergic neurons, ex vivo. Photo images of (A) the fusiform fibroblast-like adipocyte-derived stem cells (ADSCs) obtained from the adipose tissue of non-diabetic, non-PD subjects, and (B) neuron-like morphology with many neurite outgrowths, on day 21 of ADSC neuronal differentiation. (C) Immunofluorescence images of d-hDA neurons characterized by TH (green) and FOXA2 (red) immunopositivity and outgrowth of neurites. (D) Representative images and (E) histograms of SNCA immunofluorescence intensity in d-hDA neurons from patients with or without IR (control: 100 ± 6.3%, IR: 118.4 ± 5.6%, p < 0.05, n = 12). (F) Representative images and (G) histograms of CellROX live staining of d-hDA neurons from patients with IR compared to non-IR control, captured by confocal microscopy (control: 100 ± 7%, IR: 146.6 ± 8.1%, p < 0.001, n = 10). (H) Representative images and (I) graph of JC-1 (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazol-carbocyanine iodide) aggregate/monomer live staining of d-hDA neurons from patients with IR compared to non-IR control. Reduced JC-1 aggregate (red)/monomer (green) ratio indicates the ΔΨm depolarization in the d-hDA neurons (control: 100 ± 4.1%, IR: 77.8 ± 3%, p < 0.001, n = 12). *, p < 0.05, ***, p < 0.001; d-hDA, differentiated human dopaminergic neuron; IR, insulin resistance; ΔΨm, mitochondrial membrane potential.
Figure 3 Insulin resistance is associated with increased expression of SNCA and mitochondrial dysfunction in human differentiated dopaminergic neurons, ex vivo. Photo images of (A) the fusiform fibroblast-like adipocyte-derived stem cells (ADSCs) obtained from the adipose tissue of non-diabetic, non-PD subjects, and (B) neuron-like morphology with many neurite outgrowths, on day 21 of ADSC neuronal differentiation. (C) Immunofluorescence images of d-hDA neurons characterized by TH (green) and FOXA2 (red) immunopositivity and outgrowth of neurites. (D) Representative images and (E) histograms of SNCA immunofluorescence intensity in d-hDA neurons from patients with or without IR (control: 100 ± 6.3%, IR: 118.4 ± 5.6%, p < 0.05, n = 12). (F) Representative images and (G) histograms of CellROX live staining of d-hDA neurons from patients with IR compared to non-IR control, captured by confocal microscopy (control: 100 ± 7%, IR: 146.6 ± 8.1%, p < 0.001, n = 10). (H) Representative images and (I) graph of JC-1 (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazol-carbocyanine iodide) aggregate/monomer live staining of d-hDA neurons from patients with IR compared to non-IR control. Reduced JC-1 aggregate (red)/monomer (green) ratio indicates the ΔΨm depolarization in the d-hDA neurons (control: 100 ± 4.1%, IR: 77.8 ± 3%, p < 0.001, n = 12). *, p < 0.05, ***, p < 0.001; d-hDA, differentiated human dopaminergic neuron; IR, insulin resistance; ΔΨm, mitochondrial membrane potential.CC BY · PubMed Central
Figure 4 Peroxisome proliferator-activated receptor gamma coactivator 1-alpha impedes IR-induced mitochondrial dysfunction. Graphs of the differential (A) p-IRS-1/IRS-1 ratio (control: 100 ± 1.7%, IR: 223 ± 25%, p < 0.05, n = 4), (B) DCFDA (2′,7′-dichlorodihydrofluorescein diacetate) oxidation (control: 100 ± 5.3%, IR: 152.1 ± 8.9%, p < 0.01, n = 4), and (C) JC-1 aggregate/monomer ratio (control: 100 ± 1.5%, IR: 88.9 ± 0.8%, p < 0.01, n = 4), in IR SH-SY5Y cells, compared to the non-IR control cells. (D) Representative Western blot images and (E) densitometry-based histograms showing differential proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α) protein expression in IR and non-IR control SH-SY5Y cells (control: 100 ± 0%, IR: 73.4 ± 4.1%, p < 0.05, n = 4). β-actin is the loading control. (F) Graph showing the effect of PGC-1α o/e on the differential DCFDA oxidation level in IR and non-IR control SH-SY5Y cells. ns, p > 0.05, *, p < 0.05, **, p < 0.01; IRS, insulin receptor substrate; o/e, overexpression.
Figure 4 Peroxisome proliferator-activated receptor gamma coactivator 1-alpha impedes IR-induced mitochondrial dysfunction. Graphs of the differential (A) p-IRS-1/IRS-1 ratio (control: 100 ± 1.7%, IR: 223 ± 25%, p < 0.05, n = 4), (B) DCFDA (2′,7′-dichlorodihydrofluorescein diacetate) oxidation (control: 100 ± 5.3%, IR: 152.1 ± 8.9%, p < 0.01, n = 4), and (C) JC-1 aggregate/monomer ratio (control: 100 ± 1.5%, IR: 88.9 ± 0.8%, p < 0.01, n = 4), in IR SH-SY5Y cells, compared to the non-IR control cells. (D) Representative Western blot images and (E) densitometry-based histograms showing differential proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α) protein expression in IR and non-IR control SH-SY5Y cells (control: 100 ± 0%, IR: 73.4 ± 4.1%, p < 0.05, n = 4). β-actin is the loading control. (F) Graph showing the effect of PGC-1α o/e on the differential DCFDA oxidation level in IR and non-IR control SH-SY5Y cells. ns, p > 0.05, *, p < 0.05, **, p < 0.01; IRS, insulin receptor substrate; o/e, overexpression.CC BY · PubMed Central
Figure 5 IR elicits increased total, Ser129-phosphorylated, and proteinase K-resistant SNCA through a polo-like kinase 2-dependent mechanism. (A) Representative Western blot images and (B) densitometry-based graph showing the effect of IR on the expression of total (upper) and Ser129-phosphorylated (lower) SNCA proteins in SH-SY5Y cells (SNCA, control: 100 ± 1.8%, IR: 132.1 ± 11.3%, p < 0.05, n = 6; p-SNCA (Ser129): control: 100 ± 3.8%, IR: 214.3 ± 26%, p < 0.01, n = 6). (C) Representative Western blot images and (D) densitometry-based histograms showing the effect of proteinase K on the expression of SNCA protein in IR and non-IR control SH-SY5Y cells. (E) Graphical representation of the effect of 0.0625 nM BI-2536 on polo-like kinase-2 (PLK2) activity in IR SH-SY5Y cells, compared to non-IR control cells (control: 100 ± 3.3%, IR: 138.9 ± 6.4%, IR + BI2536: 73.4 ± 10.9%, n = 6). (F) Representative Western blot images and histograms showing the effect of BI-2536 on the expression of SNCA (upper) and p-SNCA (Ser129) (lower) proteins in SH-SY5Y cells. β-actin was used as a loading control; ns, p > 0.05, *, p < 0.05, **, p < 0.01, ***, p < 0.001; BI-2536, selective PLK2 inhibitor.
Figure 5 IR elicits increased total, Ser129-phosphorylated, and proteinase K-resistant SNCA through a polo-like kinase 2-dependent mechanism. (A) Representative Western blot images and (B) densitometry-based graph showing the effect of IR on the expression of total (upper) and Ser129-phosphorylated (lower) SNCA proteins in SH-SY5Y cells (SNCA, control: 100 ± 1.8%, IR: 132.1 ± 11.3%, p < 0.05, n = 6; p-SNCA (Ser129): control: 100 ± 3.8%, IR: 214.3 ± 26%, p < 0.01, n = 6). (C) Representative Western blot images and (D) densitometry-based histograms showing the effect of proteinase K on the expression of SNCA protein in IR and non-IR control SH-SY5Y cells. (E) Graphical representation of the effect of 0.0625 nM BI-2536 on polo-like kinase-2 (PLK2) activity in IR SH-SY5Y cells, compared to non-IR control cells (control: 100 ± 3.3%, IR: 138.9 ± 6.4%, IR + BI2536: 73.4 ± 10.9%, n = 6). (F) Representative Western blot images and histograms showing the effect of BI-2536 on the expression of SNCA (upper) and p-SNCA (Ser129) (lower) proteins in SH-SY5Y cells. β-actin was used as a loading control; ns, p > 0.05, *, p < 0.05, **, p < 0.01, ***, p < 0.001; BI-2536, selective PLK2 inhibitor.CC BY · PubMed Central
Figure 6 PGC-1α overexpression represses IR-enhanced PLK2 activity and subsequent increase in SNCA. (A) Representative Western blot images and (B) densitometry-based graph showing the effect of IR on the expression of PLK2 protein in SH-SY5Y cells (control: 100.5 ± 8.4%, IR: 152.9 ± 6.9%, p < 0.01, n = 5). (C) Graphical representation of the effect of IR on the expression of PLK2 mRNA level in SH-SY5Y cells (control: 100.4 ± 6.3%, IR: 139.3 ± 13.9%, n = 4). (D) Histograms showing the effect of PGC-1α o/e on PLK2 activity in IR or non-IR SH-SY5Y cells (control cells, IR (−): 100 ± 9.5%, IR (+):146 ± 14%, p < 0.05; PGC-1α o/e cells: IR (−): 50.5 ± 0.5%, IR (+): 48 ± 6%, p > 0.05, n = 4). (E) Representative Western blot images and histograms showing the effect of PGC-1α o/e on the expression of SNCA (upper) and p-SNCA (Ser129) (lower) proteins in IR SH-SY5Y cells compared to their non-IR control counterparts. β-actin was used as loading control; o/e, overexpression; ns, p > 0.05, *, p < 0.05, **, p < 0.01, ***, p < 0.001.
Figure 6 PGC-1α overexpression represses IR-enhanced PLK2 activity and subsequent increase in SNCA. (A) Representative Western blot images and (B) densitometry-based graph showing the effect of IR on the expression of PLK2 protein in SH-SY5Y cells (control: 100.5 ± 8.4%, IR: 152.9 ± 6.9%, p < 0.01, n = 5). (C) Graphical representation of the effect of IR on the expression of PLK2 mRNA level in SH-SY5Y cells (control: 100.4 ± 6.3%, IR: 139.3 ± 13.9%, n = 4). (D) Histograms showing the effect of PGC-1α o/e on PLK2 activity in IR or non-IR SH-SY5Y cells (control cells, IR (−): 100 ± 9.5%, IR (+):146 ± 14%, p < 0.05; PGC-1α o/e cells: IR (−): 50.5 ± 0.5%, IR (+): 48 ± 6%, p > 0.05, n = 4). (E) Representative Western blot images and histograms showing the effect of PGC-1α o/e on the expression of SNCA (upper) and p-SNCA (Ser129) (lower) proteins in IR SH-SY5Y cells compared to their non-IR control counterparts. β-actin was used as loading control; o/e, overexpression; ns, p > 0.05, *, p < 0.05, **, p < 0.01, ***, p < 0.001.CC BY · PubMed Central
Figure 7 IR-driven SNCA-PLK2-reactive oxygen species (ROS) signaling is implicated in PD development and progression. Molecular docking showing direct interaction between (A) SNCA and PLK2 with shape complementarity/docking score of –268.90, and ligand root-mean-square deviation (RMSD) of 47.26Å and (B) SNCA-PLK2 complex and reactive oxygen species modulator 1 (ROMO1) with a docking score of –244.41, and ligand RMSD of 59.80Å. (C) Schematic abstract illustrating an association between IR and PD, with IR promoting PD through aberrant expression of SNCA, enhanced mitochondrial oxidative stress, mitochondrial dysfunction, and deregulation of the PLK2 signaling.
Figure 7 IR-driven SNCA-PLK2-reactive oxygen species (ROS) signaling is implicated in PD development and progression. Molecular docking showing direct interaction between (A) SNCA and PLK2 with shape complementarity/docking score of –268.90, and ligand root-mean-square deviation (RMSD) of 47.26Å and (B) SNCA-PLK2 complex and reactive oxygen species modulator 1 (ROMO1) with a docking score of –244.41, and ligand RMSD of 59.80Å. (C) Schematic abstract illustrating an association between IR and PD, with IR promoting PD through aberrant expression of SNCA, enhanced mitochondrial oxidative stress, mitochondrial dysfunction, and deregulation of the PLK2 signaling.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

1 retracted

Cited by

See all 133

Metadata

Authors
Hong, C.-T., Chen, K.-Y., Wang, W., Chiu, J.-Y., Wu, D., Chao, T.-Y., Hu, C.-J., Chau, K.-Y., & Bamodu, O.
Affiliations
  • Taipei Medical University-Shuang Ho Hospital (TW)
  • Taipei Medical University (TW)
  • Taipei Medical University Hospital (TW)
  • UCL Queen Square Institute of Neurology (GB)
Corresponding author
  • Hong, Chien-Tai
  • Bamodu, Oluwaseun
Venue
Cells
Publisher
MDPI AG
Published
2020-03-17
Volume
9
Issue
3
Pages
740
ISSN
2073-4409 (ISSN-L)
OpenAlex
W3011296966
Cited by
133
References
55
Citations by year
2026: 14 · 2025: 31 · 2024: 24 · 2023: 20 · 2022: 22 · 2021: 16 · 2020: 6
FWCI
6.92
Type
journal article
Language
en
Open access
Locations
  • Cells (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • UCL Discovery (University College London) (submittedVersion)
  • Taipei Medical University Repository (submittedVersion)
Funders
  • Ministry of Science and Technology, Taiwan (105-2314-B-038-017, 106- 2314-B-038-023 MY3)
  • Taipei Medical University (TMU105-AE1-B30)
Topics
  • Health Sciences › Medicine › Clinical Neurology › Parkinson's Disease Mechanisms and Treatments
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Nuclear Receptors and Signaling
  • Life Sciences › Neuroscience › Neurology › Neuroinflammation and Neurodegeneration Mechanisms
Keywords
insulin resistance, Parkinson's disease, alpha-synuclein, PLK2, mitochondrial dysfunction, reactive oxygen species, dopaminergic neurons, PGC-1α, diabetes mellitus, neurodegeneration
MeSH terms
Peroxisome Proliferator-Activated Receptor Gamma Coactivator 1-alpha, Animals, Diabetes Mellitus, Type 2, Humans, Insulin Resistance, Male, Mice, Inbred C57BL, Mitochondria, Models, Biological, Parkinson Disease, Phosphoserine, Tyrosine 3-Monooxygenase, Signal Transduction, Genome, Human, Protein Serine-Threonine Kinases, Reactive Oxygen Species, Disease Progression, alpha-Synuclein, Dopaminergic Neurons

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.