Abstract
Purpose:
Cornea consists of anatomically distinct three layers, corneal epithelium (CEp), corneal stroma (CS), corneal endothelium (CEn), that have different developmental lineages and molecular functions. To reveal their transcriptional program, we performed spatial gene expression analysis of human cornea using next generation sequencer.
Methods:
100 ng of total RNA was extracted from the macroscopically dissected tissue fractions; CEp, CS, CEn, limbal epithelium (LEp) and conjunctiva (Cj) of the human donor corneoscleral tissue. For each sample, approximately 150 million of 100-bp, paired-end reads were sequenced (HiSeq2000, Illumina) and mapped against transcriptome database using BWA aligner. Quantity of 86,817 transcripts was corrected by gene/exon length (RPKM/FPKM), and then compared using regression analysis.
Results:
24,601 Transcript (28.3%) were expressed in common with all cell fractions. Global comparison of each spatial transcriptome showed that CS shows the closest pattern against CEn (correlation coefficient (R^2); CEn vs. CS 0.844, vs. CEp 0.807, vs. LEp 0.811, vs. Cj 0.821). We identified the candidate transcripts significantly upregulated in CEn against other tissue fractions; CEp (n=1,760), LEp (n=1,513), CS (n=1,482), Cj (n=1,322), respectively. As commonly up-regulated transcripts in the CEn, we listed up more than 300 coding genes in addition to previously reported CEn-specific markers such as COL8A2, CA2, SLC4A4, CDH2.
Conclusions:
We successfully demonstrated the layer-specific pattern of transcriptome and identified the novel up-regulated genes in CEn. This resource information is useful for understanding the spatial difference in transcriptional program to modulate cell lineages.
Keywords: 481 cornea: endothelium •
480 cornea: basic science •
533 gene/expression