Naturally occurring horse model of miscarriage reveals temporal relationship between chromosomal aberration type and point of lethality.

Chromosomal abnormalities are a common cause of human miscarriage but rarely reported in any other species. As a result, there are currently inadequate animal models available to study this condition. Horses present one potential model since mares receive intense gynecological care. This allowed us to investigate the prevalence of chromosomal copy number aberrations in 256 products of conception (POC) in a naturally occurring model of pregnancy loss (PL). Triploidy (three haploid sets of chromosomes) was the most common aberration, found in 42% of POCs following PL over the embryonic period. Over the same period, trisomies and monosomies were identified in 11.6% of POCs and subchromosomal aberrations in 4.2%. Whole and subchromosomal aberrations involved 17 autosomes, with chromosomes 3, 4, and 20 having the highest number of aberrations. Triploid fetuses had clear gross developmental anomalies of the brain. Collectively, data demonstrate that alterations in chromosome number contribute to PL similarly in women and mares, with triploidy the dominant ploidy type over the key period of organogenesis. These findings, along with highly conserved synteny between human and horse chromosomes, similar gestation lengths, and the shared single greatest risk for PL being advancing maternal age, provide strong evidence for the first animal model to truly recapitulate many key features of human miscarriage arising due to chromosomal aberrations, with shared benefits for humans and equids.

Does inbreeding contribute to pregnancy loss in Thoroughbred horses?

BACKGROUND: Excessive inbreeding increases the probability of uncovering homozygous recessive genotypes and has been associated with an increased risk of retained placenta and lower semen quality. No genomic analysis has investigated the association between inbreeding levels and pregnancy loss. OBJECTIVES: To compare genetic inbreeding coefficients (F) of naturally occurring Thoroughbred Early Pregnancy Loss (EPLs), Mid and Late term Pregnancy Loss (MLPL) and Controls. The F value was hypothesised to be higher in cases of pregnancy loss (EPLs and MLPLs) than Controls. STUDY DESIGN: Observational case-control study. METHODS: Allantochorion and fetal DNA from EPL (n = 37, gestation age 14-65 days), MLPL (n = 94, gestational age 70 days-24 h post parturition) and Controls (n = 58) were genotyped on the Axiom Equine 670K SNP Genotyping Array. Inbreeding coefficients using Runs of Homozygosity (FROH) were calculated using PLINK software. ROHs were split into size categories to investigate the recency of inbreeding. RESULTS: MLPLs had significantly higher median number of ROH (188 interquartile range [IQR], 180.8-197.3), length of ROH (3.10, IQR 2.93-3.33), and total number of ROH (590.8, IQR 537.3-632.3), and FROH (0.26, IQR 0.24-0.28) when compared with the Controls and the EPLs (p < 0.05). There was no significant difference in any of the inbreeding indices between the EPLs and Controls. The MLPLs had a significantly higher proportion of long (>10 Mb) ROH (2.5%, IQR 1.6-3.6) than the Controls (1.7%, IQR 0.6-2.5), p = 0.001. No unique ROHs were found in the EPL or MLPL populations. MAIN LIMITATIONS: SNP-array data does not allow analysis of every base in the sequence. CONCLUSIONS: This first study of the effect of genomic inbreeding levels on pregnancy loss showed that inbreeding is a contributor to MLPL, but not EPL in the UK Thoroughbred population. Mating choices remain critical, because inbreeding may predispose to MLPL by increasing the risk of homozygosity for specific lethal allele(s).

The equine umbilical cord in clinically healthy pregnancies.

BACKGROUND: Excessive umbilical cord length (UCL) is associated with equine pregnancy loss. However, a lack of UCL reference values makes it difficult to define excessive UCL. Further, factors associated with differences in UCL are poorly understood. OBJECTIVES: To (i) report the total, allantoic and amniotic UCL in healthy term pregnancies in Thoroughbreds, (ii) describe the relationship between gestational age and UCL, fetal weight and crown rump length (CRL) using clinically normal pregnancies (CNPs) from mares dying during gestation, and (iii) identify associations between UCL and maternal age and parity, paternal age, and fetal sex. STUDY DESIGN: Cross-sectional. METHODS: Data including UCLs, fetal weight, CRL and maternal age, parity, paternal age and fetal sex were taken from CNPs from Thoroughbred mares dying during gestation (n = 32), and placentas from HTPs (n = 34) in England. Correlations were assessed using Spearman's rank with significant correlations estimated by locally weighted scatter plot smoothing (LOWESS). Regression plots were fitted to highly correlated variables to further assess and quantify relationships. Differences in UCL between categorical variables were assessed using Kruskall Wallis and Mann-Whitney U tests. RESULTS: The median total, amniotic and allantoic HTP UCLs were 53.5 cm (interquartile range [IQR] 16), 29.5 cm (IQR 7) and 25.0 cm (IQR 8) respectively. Gestational age and amniotic UCL were moderately correlated (rho = 0.53, p = 0.04), with LOWESS estimating an exponential increase followed by plateauing at around Day 200. Nonlinear associations were observed between fetal weight and gestational age and CRL (adjusted r2 = 0.98 and 0.95 respectively). A linear association was observed between gestational age and CRL: predicted CRL = -17.60 + 0.38 × gestational age, p < 0.001. MAIN LIMITATIONS: Limited availability of CNPs from mares dying during gestation. Estimated relationships can only approximate growth. CONCLUSIONS: This study provides important UCL and fetal size reference values, which may aid in assessing abnormalities. For the first time, associations between UCL and gestational age have been described.