Commentary: Analysis of differentially expressed genes in bacterial and fungal keratitis.

Corneal infection is one of the very poorly understood disease in terms of pathogenesis and host immune response. The advancement in molecular methods enables us to study global gene expression patterns using microarray or next-generation sequencing technologies. These methods provide us with a huge volume of data and come up with the major challenges in critically analyzing the data to find the right significant genes. In this article, the author has analyzed already published microarray data of 30 samples from Gene Expression Omnibus (GEO) database and proposed two potential candidate genes, which can be explored as biomarkers to distinguish the bacterial or fungal keratitis.

In the present study, 451 genes from bacterial keratitis and 353 genes from fungal keratitis were found to be differentially expressed compared to the uninfected corneal tissue.[1] Upon further bioinformatic analysis, 117 co-expressed gene pairs were identified in bacterial keratitis and 87 pairs in fungal keratitis involved in nine biological pathways.[1] Among these differentially expressed genes, TLR4 (5-fold increase) and SOD2 (9-fold increase) gene standout significantly upregulated in bacterial keratitis and fungal keratitis, respectively.[1]

Toll-like receptor 4 (TLR4) is one of the pathogen-associated pattern recognition receptors (PRR). The increase in the TLR4 mRNA level was already reported in bacterial Pseudomonas aeruginosa infected corneal tissue[2] as well in Fusarium keratitis[3] and Aspergillus fumigatus keratitis.[4] Another candidate gene SOD2, which participates in the wounding and oxidation-reduction pathways, had found to be upregulated in mouse corneas with fungal (Candida albicans) keratitis[5] and also found to be upregulated during LPS treatment.[6] Based on the significant increase in the expression of these genes, the author suggested that it can be used as the diagnostic marker.

The limitation of the study, which is clearly addressed in the manuscript, includes the nonavailability of the clinical materials to verify the results and also a small sample size to provide strong conclusion. Although the approach towards developing a new candidate gene seems novel, in terms of the practical application, there are many confounding factors that needs to be addressed. The expression of these host genes depends on the nature of the pathogen, stage of the disease, and also the genetic background of the host. Further experiments are required to prove that selected candidate genes are far more superior in diagnosis, in terms of sensitivity and specificity than the already existing methods.