Characterization of an Ex Vivo Equine Endometrial Tissue Culture Model Using Next-Generation RNA-Sequencing Technology

Abstract

Persistent mating-induced endometritis is a major cause of poor fertility rates in the mare. Endometritis can be investigated using an ex vivo equine endometrial explant system which measures uterine inflammation using prostaglandin F_2α as a biomarker. However, this model has yet to undergo a wide-ranging assessment through transcriptomics. In this study, we assessed the transcriptomes of cultured endometrial explants and the optimal temporal window for their use. Endometrium harvested immediately post-mortem from native pony mares (n = 8) were sampled (0 h) and tissue explants were cultured for 24, 48 and 72 h. Tissues were stored in RNALater, total RNA was extracted and sequenced. Differentially expressed genes (DEGs) were defined using DESeq2 (R/Bioconductor). Principal component analysis indicated that the greatest changes in expression occurred in the first 24 h of culture when compared to autologous biopsies at 0 h. Fewer DEGs were seen between 24 and 48 h of culture suggesting the system was more stable than during the first 24 h. No genes were differentially expressed between 48 and 72 h but the low number of background gene expression suggested that explant viability was compromised after 48 h. ESR1, MMP9, PTGS2, PMAIP1, TNF, GADD45B and SELE genes were used as biomarkers of endometrial function, cell death and inflammation across tissue culture timepoints. STRING assessments of gene ontology suggested that DEGs between 24 and 48 h were linked to inflammation, immune system, cellular processes, environmental information processing and signal transduction, with an upregulation of most biomarker genes at 24 h. Taken together our observations indicated that 24-48 h is the optimal temporal window when the explant model can be used, as explants restore microcirculation, perform wound healing and tackle inflammation during this period. This key observation will facilitate the appropriate use of this as a model for further research into the equine endometrium and potentially the progression of mating-induced endometritis to persistent inflammation between 24 and 48 h.

Keywords: RNA-seq; endometritis; endometrium; equine; explant; gene expression; reproduction; tissue culture; transcriptome.

Figure 1

PCA plot of the gene expression profiles for the eight horses. Biopsies taken at 0 h and explants cultured up to 24, 48 and 72 h (n = 8). The analysis demonstrates clustering for the 0 h (control, representing the whole mare) and another cluster for the cultured samples (24, 48 and 72 h time points). The paler ellipses represent 95 % confidence intervals for each experimental class.

Figure 2

Comparison of all genes expressed at all time points. Pairwise comparison of the number of genes expressed at the different time points (0, 24, 48 and 72 h) after mapping RNA-sequencing reads to the equine reference genome.

Figure 3

Comparison of differentially expressed genes at all time points. Comparison of up and downregulated genes between all-time points (p < 0.05; Log₂FC ≥ ± 2). * The comparison between 48 and 72 h of culture did not retrieve any DEG.

Figure 4

Heatmap featuring the gene expression pattern of genes of interest between 0, 24, 48 and 72 h of culture.