TL;DR
This paper introduces a novel network medicine approach using microRNA-disease associations to identify new therapeutic targets. They discovered that abdominal aortic aneurysm (AAA) shares significant similarities with autoimmune diseases, leading them to repurpose methotrexate (MTX), a common autoimmune drug, for AAA treatment. Animal studies confirmed MTX's efficacy in preventing and mitigating AAA formation.
What Problem Does This Paper Address?
Abdominal aortic aneurysm (AAA) is a life-threatening condition characterized by the permanent dilation of the aorta. Currently, there are no effective non-invasive medical therapies to prevent its enlargement or rupture, leaving surgery as the primary treatment option for large or symptomatic cases. Developing new drugs is a lengthy and costly process. Therefore, there's a critical need for novel, non-invasive therapeutic strategies for AAA, and drug repurposing offers a promising avenue. The paper aims to leverage network medicine and microRNA-disease associations to uncover these new therapeutic possibilities.
Methods
The core of this study lies in a novel network medicine approach.
- MicroRNA-Disease Association Data Curation: They began by collecting and curating a comprehensive dataset of human microRNA-disease associations, specifically focusing on those with tissue expression information from the HMDD database.
- Novel Disease Similarity Algorithm: A key innovation is their algorithm for quantifying disease similarity. Unlike previous binary representations, their method uses a vector where each dimension represents a miRNA, and the value reflects the strength and direction (up/down-regulation) of the miRNA's association with a disease. They employed the Tanimoto coefficient to calculate similarity, which accounts for both the direction of miRNA regulation and the research imbalance in the data.
- MicroRNA-based Disease Network (MRDN) Construction: Using the calculated disease similarities, they built a network where nodes represent diseases and edges represent their similarity (positive for similar regulation patterns, negative for opposite). They established a similarity threshold (0.05) to create a manageable network.
- Network Analysis: They analyzed the MRDN to identify patterns, such as disease clustering and coherence.
- Drug Repurposing: Based on the network analysis, they identified diseases with high similarity to AAA and explored potential drug repurposing.
- In Vivo Validation: Promising drug candidates were then tested in an animal model of AAA to validate their therapeutic effects.
Datasets
- Human miRNA-disease association dataset: This was downloaded from the HMDD database (version 3.0). The authors extracted associations with unambiguous tissue expression regulation information.
Key Insights and Ideas
- AAA as potentially autoimmune: The most striking finding is the strong similarity between AAA and autoimmune diseases like rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) within the MRDN. This challenges the traditional view of AAA as solely a cardiovascular or atherosclerotic disease and suggests a shared autoimmune etiology.
- Network medicine for drug repurposing: The study demonstrates the power of network medicine, specifically using microRNA-disease associations, to uncover non-obvious relationships between diseases and identify potential drug repurposing opportunities for conditions with unmet therapeutic needs.
- Tanimoto coefficient for robust similarity: The use of the Tanimoto coefficient for disease similarity is a significant methodological contribution, as it addresses limitations of previous methods by incorporating both the direction and strength of miRNA associations and mitigating the impact of data imbalance.
- Disease clustering and coherence: The MRDN revealed that diseases tend to cluster by class and exhibit coherent relationships, meaning that if A is similar to C and B is similar to C, then A and B are likely similar. This suggests underlying shared biological mechanisms.
Key Results
- MRDN Structure: The constructed MRDN comprised 304 nodes (diseases) and 2440 edges, with 1521 positive and 919 negative edges. The network showed a giant interconnected component, indicating that most diseases are interconnected at the miRNA level.
- AAA's Neighbors: AAA was found to be most similar to autoimmune diseases (RA and SLE) rather than other cardiovascular diseases.
- MTX Efficacy (Preventive): In an elastase-induced AAA mouse model, MTX administration (15 mg/kg, twice a week) significantly attenuated AAA expansion and elastin degradation, and reduced macrophage infiltration, indicating a protective effect against AAA formation.
- MTX Efficacy (Therapeutic): For existing AAAs, MTX at a higher dose (25 mg/kg, twice a week) significantly retarded aorta expansion and macrophage infiltration, suggesting a potential therapeutic effect by alleviating inflammation.
- HCQ Inefficacy: Hydroxychloroquine (HCQ), another drug for autoimmune diseases, did not show a significant effect on AAA development in the animal model.
Limitations and Future Work
- Data Completeness: The authors acknowledge that the current miRNA-disease association data is incomplete. Some diseases are associated with few miRNAs, and some are not linked to any diseases in the network. This incompleteness might affect the accuracy and comprehensiveness of the MRDN.
- Organ-Specific miRNA Information: The current dataset lacks organ-specific miRNA distribution information, which could provide a more accurate representation of disease mechanisms, as miRNAs can have different roles in different organs.
- Mechanism of MTX in AAA: While MTX showed efficacy, the study notes that it's unknown whether its mechanism of action in AAA is similar to its action in RA. Further research is needed to elucidate this.
- Integration of Multi-omics Data: The authors suggest that integrating multi-omics data (e.g., genomics, proteomics) into network medicine frameworks could further enhance future analyses and provide a more holistic understanding of complex diseases.
- Further Validation: While promising, the findings in mice need to be translated to human studies to confirm MTX's therapeutic potential for AAA.
Summary updated