Integrative Transcriptomic Analysis Identifies GPX3 and CSF3R as Candidate Markers of Diabetes-Associated Myocardial Injury

Q: What are the key findings of this study?

Integrative transcriptomics identified GPX3 and CSF3R as diabetes-induced myocardial injury markers (LASSO AUC 0.701), with lower GPX3 (0.55 vs 1.06, P<0.01) and higher CSF3R (1.94 vs 0.86, P<0.01) in diabetic mice.

Q: What is the clinical evidence from this study?

Population: Diabetes mellitus and diabetes-associated myocardial injury (n=79). Intervention: Diabetes mellitus (and streptozotocin 30 mg/kg/day for 3 days plus high-fat diet in T2DM-AS mice) vs. Healthy / non-diabetic controls (or atherosclerosis-only ApoE-/- mice). Primary outcome: Diagnostic discrimination of LASSO-selected candidate biomarkers (including GPX3 and CSF3R) for diabetes mellitus and validation of myocardial protein expression in T2DM-AS mice (AUC 0.701, 95% CI 0.556-0.833, p=<0.01).

Q: What does this research mean for the field?

Diabetes-associated myocardial injury is characterized by significant downregulation of myocardial GPX3 and marked upregulation of CSF3R, identifying them as candidate markers and mediators of diabetic cardiac damage. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.NEUTRAL.

Q: What question did this study set out to answer?

To investigate the connection between extracellular antioxidant defense mechanisms and inflammatory signaling pathways in diabetes mellitus-induced myocardial injury across multiple organs and tissues.

Jiayi Song; Yiyin Long; W Zhang; Longkun Liu

doi:10.2147/dmso.s639059

October 1, 2026Diabetes Metabolic Syndrome and ObesityOpen Access

Integrative Transcriptomic Analysis Identifies GPX3 and CSF3R as Candidate Markers of Diabetes-Associated Myocardial Injury

View Full Paper

Ask AI

Bookmark

Share

Key result

Transcriptomic markers GPX3 and CSF3R identify diabetic myocardial injury with an AUC of ~0.70.

AUC 0.701
95% CI 0.556-0.833
P<0.01
n=79

Key Points

To investigate the connection between extracellular antioxidant defense mechanisms and inflammatory signaling pathways in diabetes mellitus-induced myocardial injury across multiple organs and tissues.
Integrated six single-cell RNA-seq datasets (21 diabetic samples; 122,801 cells) with the bulk transcriptomic dataset GSE184050 (113 samples).
Screened essential genes via weighted gene co-expression network analysis (WGCNA) and LASSO regression with five-fold cross-validation (lambda.min = 0.0021; cross-validated AUC = 0.701).
Conducted histological (H&E) and Western blot analyses in male ApoE-/- mice modeled with streptozotocin (30 mg/kg/day) and high-fat diet (n = 3 per group for validation assays).
Identified GPX3 (enriched in antioxidant and defense pathways) and CSF3R (enriched in PI3K-Akt signaling and linked to decreased cGMP-PKG activity) as candidate diabetic myocardial injury markers.
Western blot showed significant downregulation of myocardial GPX3 and marked upregulation of CSF3R in diabetic mice versus controls, specific to diabetes rather than atherosclerosis alone.
Histological evaluation confirmed widespread myocardial injury in diabetic mice, displaying disorganized cardiomyocyte architecture and widespread myofiber rupture.

Structured PICO

P

Population

Integrative bioinformatic analysis of 21 diabetic single-cell RNA-seq samples (122,801 cells) and 58 participants (113 samples) from bulk RNA-seq dataset GSE184050, validated in vivo in 18 male mice over 8 weeks.

E

Exposure

Intraperitoneal streptozotocin (30 mg/kg/day for 3 consecutive days; cumulative dose 90 mg/kg) plus high-fat chow for 8 weeks in ApoE-/- mice to induce T2DM and atherosclerosis (T2DM-AS), alongside bioinformatic transcriptomic screening (DEG, WGCNA, LASSO regression).

C

Comparator

ApoE-/- mice injected intraperitoneally with an equal volume of purified water and fed high-fat chow (atherosclerosis alone, n=6), and C57BL/6J mice injected with purified water and fed regular chow diet (non-diabetic healthy control, n=6); for human transcriptomics, 33 non-transitioners vs 25 T2D transitioners.

O

Outcome

Identification and validation of candidate gene biomarkers (GPX3 and CSF3R) and histological myocardial injury score (0-3 scale) and myocardial protein expression via Western blot at 8 weeks.surrogate

Main Result

Effect estimate: AUC 0.701 (95% CI 0.556-0.833)

Absolute Event Rate: 0.55% vs 1.06%

p-value: p=<0.01

Integrative transcriptomic analysis and preclinical mouse modeling identify myocardial GPX3 downregulation and CSF3R upregulation as candidate biomarkers linking impaired antioxidant defense and inflammatory signaling to diabetes-induced myocardial injury.

Limitations

No gene remained statistically significant after Benjamini-Hochberg false discovery rate adjustment in the bulk RNA-seq DEG analysis (minimum adjusted p = 0.069) due to modest sample size (n = 58).
Candidate selection on the full dataset introduced optimistic bias, with nested cross-validation reducing the AUC from 0.701 to 0.613.
Terminal histological and Western blot validations were conducted in a small subset of animals (n = 3 mice per group).
Findings require validation in independent human cardiac datasets and direct mechanistic experiments.
Modest sample size in bulk RNA-seq dataset (n = 58 participants), resulting in no differentially expressed genes remaining significant after Benjamini–Hochberg false discovery rate correction (minimum adjusted p = 0.069)
Histological and Western blot confirmatory analyses were conducted in a small subset of n = 3 mice per group
Requires validation in independent human cardiac datasets and mechanistic experiments

Ask about this paper

Answers are grounded in this paper and cite their sources.

Implication

Hypothesis-generating for GPX3 and CSF3R in diabetic myocardial injury; requires human validation before any clinical consideration.

Abstract

Purpose: The clinical relationship between diabetes mellitus (DM) and myocardial injury (MI) is well established, yet the antioxidant protection mechanisms involving glutathione peroxidase that link DM to MI remain insufficiently characterized. Here, across multiple organs and tissues, we explored the relationship between extracellular antioxidant defense and inflammatory signaling in DM-induced MI. Methods: The analyses integrated six single-cell RNA-seq datasets (21 diabetic samples; 122,801 cells) with the bulk transcriptomic dataset GSE184050 (113 samples); LASSO regression yielded a cross-validated area under the curve (AUC) of 0.701, and GPX3 and CSF3R were further validated in a T2DM-AS mouse model (n = 3 per group for histological and Western blot analyses). Results: We identified glutathione peroxidase 3 (GPX3) and colony-stimulating factor 3 receptor (CSF3R) as candidate markers associated with DM-induced MI. CSF3R was enriched in injury- and inflammation-related pathways including PI3K-Akt signaling, whereas GPX3 was enriched in antioxidant and defense-related pathways. CSF3R enrichment was also associated with reduced activity of the cardioprotective cGMP-PKG pathway. Endothelial dysfunction and collagen remodeling, potentially involving monocyte–endothelial crosstalk, are proposed as candidate intermediates in this cascade. Histological examination confirmed extensive MI in DM mice, characterized by disordered cardiomyocyte arrangement and widespread myofiber rupture. Western blot analysis showed that myocardial GPX3 expression was significantly downregulated in DM mice compared with non-diabetic controls, whereas CSF3R was markedly upregulated, changes that were specific to the diabetic milieu and not observed in atherosclerosis alone. Conclusion: GPX3 and CSF3R were identified as candidate markers associated with diabetes-related cardiac injury and warrant validation in independent human cardiac datasets and mechanistic experiments.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Song et al. (2026) studied Diabetes mellitus and diabetes-associated myocardial injury (n=79). Diabetes mellitus (and streptozotocin 30 mg/kg/day for 3 days plus high-fat diet in T2DM-AS mice) vs. Healthy / non-diabetic controls (or atherosclerosis-only ApoE-/- mice) was evaluated on Diagnostic discrimination of LASSO-selected candidate biomarkers (including GPX3 and CSF3R) for diabetes mellitus and validation of myocardial protein expression in T2DM-AS mice (AUC 0.701, 95% CI 0.556-0.833, p=<0.01). Integrative transcriptomics identified GPX3 and CSF3R as diabetes-induced myocardial injury markers (LASSO AUC 0.701), with lower GPX3 (0.55 vs 1.06, P<0.01) and higher CSF3R (1.94 vs 0.86, P<0.01) in diabetic mice.

synapsesocial.com/papers/6adc966ca704d151585a5e21 https://doi.org/10.2147/dmso.s639059

View Full Paper

Ask AI

Bookmark

Share

Integrative Transcriptomic Analysis Identifies GPX3 and CSF3R as Candidate Markers of Diabetes-Associated Myocardial Injury

Key Points

Structured PICO

Main Result

Limitations

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Integrative Transcriptomic Analysis Identifies GPX3 and CSF3R as Candidate Markers of Diabetes-Associated Myocardial Injury

Key Points

Structured PICO

Main Result

Limitations

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading