Page 53 - Read Online

Aydin et al. J Transl Genet Genom. 2025;9:406-26 https://dx.doi.org/10.20517/jtgg.2025.108 Page 420

proposed to hold great promise for DMD patients by alleviating effects on NF-κB signaling, inflammation,
and oxidative stress responses.

Emetine, by contrast, influences a complementary set of DMD-relevant pathways. Its ability to inhibit
[76]
protein synthesis under cellular stress, regulate apoptosis, and modulate the integrated stress response
intersects with core degenerative mechanisms observed in dystrophic muscle, including mitochondrial
dysfunction, excessive apoptotic signaling, and endoplasmic reticulum stress. Additionally, emetine has
been reported to exert anti-inflammatory effects and reduce fibrosis-related gene expression , processes
[77]
that are highly pertinent to the fibrotic remodeling and immune dysregulation that characterize late-stage
DMD pathology. Together, these drug-specific mechanisms map directly onto the molecular signatures
uncovered by our integrated PPI, miRNA, and TF networks - particularly through hubs such as SPP1,
CD44, MYCN, HLA-C, CIITA, and PML - highlighting how both compounds may modulate multiple
interconnected biological processes central to DMD progression.

Overall, by integrating curated pharmacological evidence with network-level molecular disruptions, our
findings provide a coherent mechanistic rationale linking celastrol and emetine to the inflammatory,
oxidative, regenerative, and proteostatic abnormalities that define DMD. These results not only strengthen
the biological plausibility of our computational predictions but also offer a focused set of hypotheses for
future experimental validation in DMD cellular and animal models.

DMD is a rare and clinically heterogeneous disorder for which comprehensive multi-omics datasets remain
limited, and this scarcity inevitably constrains the breadth and depth of computational investigations. In
this study, DMD was examined through the integration of transcriptome-derived molecular signatures and
network-level biomarkers, followed by in silico drug repositioning analyses targeting these diagnostic nodes.
Although this systems-based framework provides a valuable hypothesis-generating platform, several
important limitations should be acknowledged to contextualize the conclusions drawn from the findings.
The analyses depend entirely on publicly available transcriptome-level datasets, which, while enabling broad
molecular profiling, lack the granularity and biological control offered by experimental models. Publicly
archived datasets often differ in sample collection protocols, tissue processing methods, sequencing
platforms, and normalization strategies, all of which may introduce technical variability that cannot be fully
eliminated even with rigorous normalization and quality control. Despite employing variance-stabilized,
limma-normalized expression matrices to reduce such heterogeneity, the limited availability of matched
omics datasets for DMD restricts the capacity to capture the full molecular spectrum of the disease. The
absence of age- and sex-matched samples further complicates interpretation, as developmental stage and
biological sex are known to influence dystrophic muscle phenotypes and inflammatory responses. Even
though the combined sample size across datasets allows for statistically meaningful comparisons, the overall
omics landscape might be more accurately resolved through additional datasets encompassing broader
demographic and clinical diversity. The computationally predicted biomarker signatures and drug-target
interactions represent theoretical inferences grounded in transcriptomic and network-derived associations.
These predictions cannot be regarded as confirmatory in the absence of functional validation. DMD
patient-derived myoblasts, induced pluripotent stem cell (iPSC)-derived muscle models, and well-
established animal models such as the mdx mouse would provide the mechanistic context necessary to
validate the regulatory importance of the identified biomarkers and to test the biological efficacy of the
proposed drug candidates. The lack of experimental evaluation limits the capacity to fully infer whether the
predicted modulatory effects would manifest in physiological settings. The drug repositioning strategy
implemented here focuses predominantly on inhibitory interactions, which biases the analysis toward
regulators of inflammation, immune activation, and fibrosis that are typically upregulated in dystrophic