Abstract
Purpose:
We have previously shown that age related meibomian gland dysfunction occurs in aging mice, similar to that observed in MGD subjects. To further understand the molecular changes associated with aging glands we generated transcriptome profiles from meibomian glands and conjunctiva isolated from young, old, male and female mice.
Methods:
4 groups of C57Bl6 mice, aged 3 months or 2 years, male or fe male were sacrificed and the lower eyelids harvested. Tarsal plates containing the meibomian glands and overlying conjunctiva was frozen under liquid nitrogen, crushed and suspended in RLT buffer (Qiagen). RNA was then isolated over RNeasy columns and RNA quality and quantity assessed with an Agilent analyzer. Using TruSeq RNA Sample Prep kit (Illumina), mRNA was enriched and chemically fragmented. RNA fragments were then transcribed into cDNA using random primers and reverse transcriptase. After second strand synthesis, the cDNA fragments were blunt ended and ligated to adapters. The adapter ligated fragments were enriched by 11 PCR cycles using primers complementary to the adapters. This generated 4 non-stranded libraries with distinct bar codes. Libraries were then sequenced on an Illumina HiSeq 2500 using version 3 chemistry. Sequences were mapped to the mm10 reference genome using Casava software (Illumina). Expression results were gathered as RPKM values for each gene, exon and splice junction in the reference database. Gene expression analyses were performed using CyberT
Results:
Over 55 million reads were generated from each library and base call frequencies, base call quality, Kmer enrichment and per sequence quality indicated the data was highly reliable. The data indicate that ~16,000 genes are expressed in these tissues. Gene expression data indicated that 8 and 20 genes were exclusively expressed in young and old tissues, respectively. Of genes showing differences more than 2 fold in either young or old tissue, 50 genes were identified and by GO ontology, FGF, chemokine/cytokine, integrin, and G protein signaling pathways were represented. In addition, genes associated with Pparγ function and Wnt signaling genes were enriched in young tissue.
Conclusions:
RNA-seq data yields a wealth of data quantifying both gene expression values as well as exon usage and splice variants. In this data, a number of signaling pathways, including Pparγ and Wnt were identified as altered in meibomian glands and conjunctiva of aging mice.