Drug-repurposing identified the combination of Trolox C and Cytisine for the treatment of type 2 diabetes

Ling Jin, Jian Tu, Jianwei Jia, Wenbin An, Huanran Tan, Qinghua Cui, Zhixin LiJournal of Translational Medicine 2014 · DOI

TL;DR

This paper introduces a novel computational method for drug repurposing, specifically designed to identify effective drug combinations for diseases. By applying this system to type 2 diabetes, they identified and experimentally validated a promising combination of Trolox C and Cytisine, demonstrating its potential for treating the condition.

What Problem Does This Paper Address?

Drug repurposing, the practice of finding new uses for existing drugs, offers a faster, cheaper, and less risky alternative to developing entirely new therapeutics. While computational methods for repurposing single drugs have advanced significantly, there's a notable gap in computational approaches for identifying effective combinations of drugs. This is a critical area because many diseases, including type 2 diabetes, are complex and often require multi-drug therapies for optimal management. The authors aim to bridge this gap by developing a system that can predict synergistic drug combinations for a given disease.

Methods

The core of this paper is a novel computational scoring system for drug repurposing of combined drugs. The system leverages the Connectivity Map (CMap) database, which contains gene expression profiles of cells treated with various drugs. Here's how it works:

  1. Disease Gene Identification: First, the researchers identify genes that are significantly up- or down-regulated in the target disease (in this case, type 2 diabetes).
  2. Drug-Induced Gene Signatures: For each drug in the CMap database, they determine which of the disease-associated genes are reversed (i.e., their expression pattern is normalized) and which are further exacerbated (i.e., their expression pattern mimics the disease).
  3. Scoring Drug Combinations: The novelty lies in how they score pairs of drugs. For a given disease, they calculate a score for a drug pair (Drug A + Drug B) based on how effectively each drug individually reverses the disease's deregulated genes and how many genes they both reverse. Specifically, they use a formula that multiplies the difference between reversed and exacerbated genes for each drug in the pair. A higher score indicates a greater potential for the drug combination to counteract the disease's genetic signature.
  4. Experimental Validation: After computationally identifying candidate combinations, the researchers selected one pair (Trolox C and Cytisine) for in vivo validation using a mouse model of type 2 diabetes.

Datasets

  • Connectivity Map (CMap) database: This is the primary dataset used for identifying drug-induced gene expression signatures. The paper mentions it initially contained profiles of 164 drugs and was later expanded to around 1400 FDA-approved small molecules.
  • ArrayExpress database: Used to obtain the deregulated genes (up- and down-regulated) specific to type 2 diabetes.

Key Insights and Ideas

  • Computational prediction of drug combinations is feasible: The paper demonstrates that a CMap-based approach can be adapted to predict synergistic drug combinations, not just single drugs.
  • Trolox C and Cytisine as a potential type 2 diabetes treatment: The computational prediction led to the identification of this specific combination, which was then experimentally validated. This highlights the power of the method to uncover unexpected therapeutic pairings.
  • Novelty in scoring: The scoring mechanism, which considers both the reversal and exacerbation of disease genes by individual drugs within a pair, is a key innovation for combination drug repurposing.
  • Type 2 diabetes complexity: The paper underscores that type 2 diabetes is a complex metabolic disorder with no single perfect treatment, necessitating exploration of novel therapeutic strategies like drug combinations.

Key Results

  • The computational scoring system identified a list of top ten candidate drug combinations for type 2 diabetes.
  • The combination of Trolox C and Cytisine was selected for further investigation.
  • In a mouse model of type 2 diabetes (HF/STZ model), the combination of Trolox C and Cytisine significantly reduced hyperglycemia.
  • Neither Trolox C nor Cytisine alone showed significant effectiveness in treating the diabetic mice, emphasizing the synergistic effect of the combination.
  • The combination treatment did not negatively affect the body weight of the diabetic mice.
  • The combination also showed partial alleviation of glucose tolerance issues in the diabetic mice.

Limitations and Future Work

The authors acknowledge several limitations:

  • Optimal Fraction Identification: The current method does not identify the optimal ratios or fractions of the two combined drugs, which can be crucial for efficacy.
  • Limited Genomic Information: The method doesn't incorporate all relevant genomic information, such as gene importance.
  • Qualitative Gene Expression Changes: The current method is qualitative and doesn't consider the magnitude (fold change) of gene expression changes or their statistical significance.
  • Animal Model Limitations: The animal model used for type 2 diabetes may not perfectly mimic the human condition.

Future work suggested includes:

  • Improving the computational method to incorporate the above limitations, potentially using random sampling for better prediction accuracy.
  • Validating the findings in other type 2 diabetes animal models (e.g., db/db mice, high-fat-diet-induced models) to strengthen the evidence for potential human application.

Summary updated

Abstract

The presented scoring system represents one novel method for drug repurposing, which would provide helps for greatly extended the space of drugs.
BibTeX
@article{jin2014drug,
  author = {Ling Jin and Jian Tu and Jianwei Jia and Wenbin An and Huanran Tan and Qinghua Cui and Zhixin Li},
  title = {{Drug-repurposing identified the combination of Trolox C and Cytisine for the treatment of type 2 diabetes}},
  journal = {Journal of Translational Medicine},
  year = {2014},
  doi = {10.1186/1479-5876-12-153},
}
Figure