Single-Cell RNA Sequencing: Opportunities and Challenges for Studies on Corneal Biology in Health and Disease

Abstract

The structure and major cell types of the multi-layer human cornea have been extensively studied. However, various cell states in specific cell types and key genes that define the cell states are not fully understood, hindering our comprehension of corneal homeostasis, related diseases, and therapeutic discovery. Single-cell RNA sequencing is a revolutionary and powerful tool for identifying cell states within tissues such as the cornea. This review provides an overview of current single-cell RNA sequencing studies on the human cornea, highlighting similarities and differences between them, and summarizing the key genes that define corneal cell states reported in these studies. In addition, this review discusses the opportunities and challenges of using single-cell RNA sequencing to study corneal biology in health and disease.

Keywords: RNA sequencing; cornea; limbal stem cells; single-cell.

Figure 1

Schematic overview of the corneal tissue layers. The cellular layers are highlighted, with the limbus acting as the border between the conjunctiva and the central cornea.

Figure 2

Upset plots [47] of intersecting marker genes between studies. Marker genes for cells in (A) Central epithelium; (B) Conjunctiva; (C) Corneal stroma; (D) Endothelium are summarized from studies that reported marker genes for these regions. The gene lists were obtained from supplementary data tables from corresponding studies. The vertical intersection size represents the number of identified marker genes to define cell states in each study or the intersection with other studies specified below the graphs. The connected dots indicate which studies share intersection of marker genes. The horizontal set size indicates the total number of marker genes identified in each study. Note that only Collin et al. provided the complete list of limbal marker genes in the format of supplementary table, and therefore, no comparison between studies could be performed for the limbus. The number of marker genes was counted based on selected genes with adjusted p-value smaller than 0.05 and a log2 (fold change) greater than zero [28,29,30,34].