Obesity during Pregnancy in the Horse: Effect on Term Placental Structure and Gene Expression, as Well as Colostrum and Milk Fatty Acid Concentration

In horses, the prevalence of obesity is high and associated with serious metabolic pathologies. Being a broodmare has been identified as a risk factor for obesity. In other species, maternal obesity is known to affect the development of the offspring. This article is a follow-up study of previous work showing that Obese mares (O, n = 10, body condition score > 4.25 at insemination) were more insulin resistant and presented increased systemic inflammation during pregnancy compared to Normal mares (N, n = 14, body condition score < 4 at insemination). Foals born to O mares were more insulin-resistant, presented increased systemic inflammation, and were more affected by osteoarticular lesions. The objective of the present study was to investigate the effect of maternal obesity on placental structure and function, as well as the fatty acid profile in the plasma of mares and foals, colostrum, and milk until 90 days of lactation, which, to our knowledge, has been poorly studied in the horse. Mares from both groups were fed the same diet during pregnancy and lactation. During lactation, mares were housed in pasture. A strong heat wave, followed by a drought, occurred during their 2nd and 3rd months of lactation (summer of 2016 in the Limousin region, France). In the present article, term placental morphometry, structure (stereology), and gene expression (RT-qPCR, genes involved in nutrient transport, growth, and development, as well as vascularization) were studied. Plasma of mares and their foals, as well as colostrum and milk, were sampled at birth, 30 days, and 90 days of lactation. The fatty acid composition of these samples was measured using gas chromatography. No differences between the N and O groups were observed for term placental morphometry, structure, or gene expression. No difference in plasma fatty acid composition was observed between groups in mares. The plasma fatty acid profile of O foals was more pro-inflammatory and indicated an altered placental lipid metabolism between birth and 90 days of age. These results are in line with the increased systemic inflammation and altered glucose metabolism observed until 18 months of age in this group. The colostrum fatty acid profile of O mares was more pro-inflammatory and indicated an increased transfer and/or desaturation of long-chain fatty acids. Moreover, O foals received a colostrum poorer in medium-chain saturated fatty acid, a source of immediate energy for the newborn that can also play a role in immunity and gut microbiota development. Differences in milk fatty acid composition indicated a decreased ability to adapt to heat stress in O mares, which could have further affected the metabolic development of their foals. In conclusion, maternal obesity affected the fatty acid composition of milk, thus also influencing the foal's plasma fatty acid composition and likely participating in the developmental programming observed in growing foals.

Alteration of the embryonic microenvironment and sex-specific responses of the preimplantation embryo related to a maternal high-fat diet in the rabbit model

The maternal metabolic environment can be detrimental to the health of the offspring. In a previous work, we showed that maternal high-fat (HH) feeding in rabbit induced sex-dependent metabolic adaptation in the fetus and led to metabolic syndrome in adult offspring. As early development representing a critical window of susceptibility, in the present work we aimed to explore the effects of the HH diet on the oocyte, preimplantation embryo and its microenvironment. In oocytes from females on HH diet, transcriptomic analysis revealed a weak modification in the content of transcripts mainly involved in meiosis and translational control. The effect of maternal HH diet on the embryonic microenvironment was investigated by identifying the metabolite composition of uterine and embryonic fluids collected in vivo by biomicroscopy. Metabolomic analysis revealed differences in the HH uterine fluid surrounding the embryo, with increased pyruvate concentration. Within the blastocoelic fluid, metabolomic profiles showed decreased glucose and alanine concentrations. In addition, the blastocyst transcriptome showed under-expression of genes and pathways involved in lipid, glucose and amino acid transport and metabolism, most pronounced in female embryos. This work demonstrates that the maternal HH diet disrupts the in vivo composition of the embryonic microenvironment, where the presence of nutrients is increased. In contrast to this nutrient-rich environment, the embryo presents a decrease in nutrient sensing and metabolism suggesting a potential protective process. In addition, this work identifies a very early sex-specific response to the maternal HH diet, from the blastocyst stage.

Metabolomic differences in blastocoel and uterine fluids collected in vivo by ultrasound biomicroscopy on rabbit embryos†

The success of embryo development and implantation depends in part on the environment in which the embryo evolves. However, the composition of the uterine fluid surrounding the embryo in the peri-implantation period remains poorly studied. In this work, we aimed to develop a new strategy to visualize, collect, and analyze both blastocoelic liquid and juxta-embryonic uterine fluid from in vivo peri-implantation rabbit embryos. Using high-resolution ultrasound biomicroscopy, embryos were observed as fluid-filled anechoic vesicles, some of which were surrounded by a thin layer of uterine fluid. Ultrasound-guided puncture and aspiration of both the blastocoelic fluid contained in the embryo and the uterine fluid in the vicinity of the embryo were performed. Using nuclear magnetic resonance spectroscopy, altogether 24 metabolites were identified and quantified, of which 21 were detected in both fluids with a higher concentration in the uterus compared to the blastocoel. In contrast, pyruvate was detected at a higher concentration in blastocoelic compared to uterine fluid. Two acidic amino acids, glutamate and aspartate, were not detected in uterine fluid in contrast to blastocoelic fluid, suggesting a local regulation of uterine fluid composition. To our knowledge, this is the first report of simultaneous analysis of blastocoelic and uterine fluids collected in vivo at the time of implantation in mammals, shedding new insight for understanding the relationship between the embryo and its local environment at this critical period of development.

142 Nulliparity alters gene expression in inner cell mass and trophoblast of equine blastocysts in old mares

An increased incidence in early embryo loss has been observed in aged mares. Moreover, the first foal born to a mare is lighter than her subsequent foals, with reported impaired placental function at term. Because trophoblast function may be affected from the embryo stage, the aim of this project was to determine the effect of parity in aged mares on gene expression in Day-8.5 embryos. Middle-aged (13.5±2.2 years) nulliparous (never foaled) (ON) or multiparous (1.8±1.6 foals) (OM) Saddlebred, non-nursing mares were inseminated with the semen of one unique stallion. At 8 days post-ovulation (10 days post-hCG), embryos were recovered by uterine flushing and bisected to obtain samples of pure (trophectoderm, TE) or inner cell mass enriched (ICM) trophoblast. Paired end, non-oriented RNA sequencing was performed with Illumina (NextSEqn 500) on 5 and 6 TE and ICM collected from ON and OM, respectively. Differential expression was analysed with DESEqn 2. Embryo size was included in the model and a P&lt;0.05 cutoff was used after false discovery rate correction. Gene set enrichment analysis (GSEA) was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) databases. Out of the 13 007 and 12 706 genes expressed in ICM and TE, respectively, only 8 in ICM and 6 in TE were differentially expressed, with 2 genes in common. Nevertheless, 19 gene sets were enriched and 6 depleted in the ICM of ON, whereas 2 gene sets were enriched and 8 depleted in the TE of ON compared with OM. Gene sets involved in ribosomal activity and structure, proteasome, integral component of plasma membrane, and immune response were enriched in ICM from ON embryos, and gene sets linked to sphingolipid metabolism, nucleosome, and constituents of the extracellular matrix (ECM) were depleted. In TE from ON mares, enriched gene sets were involved with ribosomes and depleted gene sets were linked to extracellular matrix, focal adhesion, myosin complex, and sequence-specific DNA binding. Overall, 1 enriched (linked to extracellular matrix) and 1 depleted gene set (involved in ribosomal structure) were common to ICM and TE. Thus, embryos from aged nulliparous mares seem to have higher protein turnover and higher immune response compared with those of OM, whereas the depletion of gene sets associated with extracellular matrix and membrane may indicate differences in cellular organisation into lineages. More work is ongoing to study effects on subsequent development.

202 Proteomic, metabolomic and fatty acid composition in lactating and non-lactating mares&rsquo; uterine fluids

In the equine industry, horse breeders aim to produce one foal per mare per year. Thus, mares are bred while nursing. In high-yielding dairy cattle, concurrent lactation and conception may affect embryo quality, but effects on uterine fluid (UF) quality are unknown. No data are available in horses. The aim of our study was to analyse the effect of nursing on protein, metabolite and fatty acid (FA) compositions of UF at ~7.5 days post-ovulation. Anglo-Arab mares (multiparous (2.7±0.9 foals), 10.8±2.5 years old) that were either nursing (N; 105±12 days of lactation) or barren (B) were inseminated with the semen of the same stallion at induction of ovulation with human chorionic gonadotrophin. Ovulation was confirmed by ultrasound within 48h. At 7.4±0.7 days post-ovulation, UF was collected with a human tampon that was left for 10min in the uterus before embryo collection. Only mares for which no embryo was collected were selected (n=5 in both groups). Trypsin digestion followed by liquid chromatography-tandem mass spectrometry analysis was used for proteomic analysis, with subsequent characterisation with the PANTHER platform using an overrepresentation test (Fisher's exact type with false discovery rate correction) with the PANTHER pathway database. An untargeted approach based on liquid chromatography-tandem mass spectrometry was performed for metabolomics (normalisation to 50µg of protein), and data were analysed with the MS Peaks to Pathways tool of the MetaboAnalyst platform, using both the GSEA and mummichog algorithms (cut-off=0.05). Fatty acid composition (% of total FA) was analysed using gas chromatography. Differences between groups were analysed using a linear model with permutation using R software with the pgirmess package (P=0.05 for significance). Altogether, 2706 proteins with at least two peptides were identified in mares’ UF, with 164 being differentially expressed. Ubiquitin proteasome, involved in embryo-endometrium interactions, was the most enriched pathway in N mares (fold enrichment=15.12; P&lt;0.0001). For metabolomics, ubiquinone biosynthesis [MetaFishNet, P=0.02; NES (Enrichment Score for the variable)/(mean of all Enrichment Score in all permutations in the dataset)=1.84] was enriched in N mares. N-Glycans, mainly guanosine diphosphate mannose (Kyoto Encyclopedia of Genes and Genomes, P=0.04; NES=−1.33) and leukotriene biosynthesis (Biocyc; P=0.006; NES=−1.31), were enriched in B mares with differentially expressed leukotrienes C4/D4. The proportion of saturated FA was higher in N vs. B mares (38.4±3.6% vs. 33.2±3.6%; P=0.03), probably due to increased palmitic (P=0.08) and stearic (P=0.08) acid proportions. In conclusion, pathways involved in uterine receptivity and inflammation seem to be enriched in B mares. Fatty acids that are readily available in the diet were more present in N mares, possibly because more elaborate FA are exported to the mammary gland for milk production. Nursing could thus modify the inflammatory response in the uterine environment at ~7.5 days post-ovulation and could affect reproductive efficiency in horses.

Effects of dietary arginine supplementation in pregnant mares on maternal metabolism, placental structure and function and foal growth

Foals born to primiparous mares are lighter and less mature than those born to multiparous dams. Factors driving this difference are not totally understood. Using 7 multiparous and 6 primiparous standardbred mares, we demonstrated that, in late gestation, primiparous mares were less insulin resistant compared to multiparous mares, and that their foals had reduced plasma amino-acid concentrations at birth compared to foals born to multiparous mares. Vascular development, as observed through structure and gene expression, and global DNA methylation were also reduced in primiparous placentas. Another group of 8 primiparous mares was orally supplemented with L-arginine (100 g/day, 210d to term). L-arginine improved pregnancy-induced insulin resistance and increased maternal L-arginine and L-ornithine plasma concentrations but foal plasma amino acid concentrations were not affected at birth. At birth, foal weight and placental biometry, structure, ultra-structure and DNA methylation were not modified. Placental expression of genes involved in glucose and fatty acid transfers was increased. In conclusion, maternal insulin resistance in response to pregnancy and placental function are reduced in primiparous pregnancies. Late-gestation L-arginine supplementation may help primiparous mares to metabolically adapt to pregnancy and improve placental function. More work is needed to confirm these effects and ascertain optimal treatment conditions.

Maternal obesity increases insulin resistance, low-grade inflammation and osteochondrosis lesions in foals and yearlings until 18 months of age

INTRODUCTION

Obesity is a growing concern in horses. The effects of maternal obesity on maternal metabolism and low-grade inflammation during pregnancy, as well as offspring growth, metabolism, low-grade inflammation, testicular maturation and osteochondrotic lesions until 18 months of age were investigated.

MATERIAL AND METHODS

Twenty-four mares were used and separated into two groups at insemination according to body condition score (BCS): Normal (N, n = 10, BCS ≤4) and Obese (O, n = 14, BCS ≥4.25). BCS and plasma glucose, insulin, triglyceride, urea, non-esterified fatty acid, serum amyloid A (SAA), leptin and adiponectin concentrations were monitored throughout gestation. At 300 days of gestation, a Frequently Sampled Intravenous Glucose Tolerance Test (FSIGT) was performed. After parturition, foals' weight and size were monitored until 18 months of age with plasma SAA, leptin, adiponectin, triiodothyronine (T3), thyroxine (T4) and cortisol concentrations measured at regular intervals. At 6, 12 and 18 months of age, FSIGT and osteoarticular examinations were performed. Males were gelded at one year and expression of genes involved in testicular maturation analysed by RT-qPCR.

RESULTS

Throughout the experiment, maternal BCS was higher in O versus N mares. During gestation, plasma urea and adiponectin were decreased and SAA and leptin increased in O versus N mares. O mares were also more insulin resistant than N mares with a higher glucose effectiveness. Postnatally, there was no difference in offspring growth between groups. Nevertheless, plasma SAA concentrations were increased in O versus N foals until 6 months, with O foals being consistently more insulin resistant with a higher glucose effectiveness. At 12 months of age, O foals were significantly more affected by osteochondrosis than N foals. All other parameters were not different between groups.

CONCLUSION

In conclusion, maternal obesity altered metabolism and increased low-grade inflammation in both dams and foals. The risk of developing osteochondrosis at 12 months of age was also higher in foals born to obese dams.

65 Placental Function at Term is Altered in Broodmares Fed with Cereals from Mid Gestation

Recent data obtained in our laboratory suggest that feeding pregnant broodmares with cereal concentrates may affect both mare and foal metabolism in the short and long term. Here, we investigated feto-placental biometry and placental function at term in mares fed with cereals and forage or forage only. Twenty-two multiparous mares inseminated with the same stallion were allocated to 1 of 2 groups from 7 months of gestation: group F (n = 10) were fed forage only, whereas group B (n = 12) received forage and cracked barley until foaling. At 3 and 9 months of gestation, a glucose tolerance test (IVGTT) was performed to evaluate the insulin resistance of pregnant mares. At birth, placentas and foals were weighed and measured. Placental samples were collected above the umbilical cord insertion and snap frozen. An RNA sequencing (RNAseq) analysis was performed on 9 placentas of each group. After normalization, gene levels were analysed using the DESEqn 2 package of R software (https://www.r-project.org/). Enrichment of gene sets was analysed using the Gene Set Enrichment Analysis (GSEA) software using the Kyoto Encyclopaedia of Genes and Genomes (KEGG) and Gene Ontology [GO, biological processes (bp), molecular function (mf) and cellular components (cc)] databases. Gene analysis statistical results were considered significant for P-values < 0.05 after false rate discovery (fdr) correction. The IVGTT results were analysed using a type 3 ANOVA on a mixed linear model with group as fixed effect and age of the mare as random effect. At 3 months of gestation, maternal glucose metabolism was not different between groups. At 9 months, B mares had a higher insulin area under the curve (AUC) after glucose injection than F mares (P < 0.01), without any difference in glucose AUC, suggesting that B mares were more insulin resistant than F mares. At birth, no difference was observed for feto-placental biometry between groups. Gene-level analysis could not discern differences in gene expression between groups after fdr correction. The GSEA analysis, however, showed that 8 gene sets were down-regulated in C placentas (2 KEGG, 2 GObp, 3 GOmf, 1 GOcc) and 193 gene sets were up-regulated (15 KEGG, 144 GObp, 12 GOmf, 22 GOcc) in B placentas. The down-regulated gene sets were involved in neutral amino acids and anion transport, fatty acid oxidation, acetyl coA synthesis, cholesterol and folate degradation, and the up-regulated gene sets were involved in RNA expression, inflammation (activation and recruitment of immune cells, MAPK signalling, complement and coagulation cascades, pro-inflammatory cytokine production and signalling) and in vascularisation (vasculogenesis, angiogenesis and smooth muscle cells development). The results are consistent with the altered function observed in term placentas of women who suffer from gestational diabetes. In conclusion, feeding pregnant mares with cereal from mid gestation alters the placental function at term. The authors thank the GeT platform (Toulouse, France) for the sequencing of the samples.

Placental structure and function in different breeds in horses

Ponies and sometimes draft horses are often used as experimental models for horses although size and metabolic parameters are known to vary between horse breeds. So far, there is little information about differences of placental structure and no information about differences of placental function between breeds. The aim of this study was to investigate differences in placental size, structure and function at birth in relation to foal size and weight in ponies, Saddlebred and draft horses. Pony, Saddlebred and draft horse pregnancies were obtained by artificial insemination over 2 successive breeding seasons. Foals and total fetal membranes (TFM) were weighed and placentas measured for surface area at term. Placentas were sampled above the umbilical cord insertion. Surface density and volume fraction of the different cellular components of the placenta were measured on histological sections using stereology. The expression of genes involved in growth and development, nutrient transfer and vascularization was compared between groups. Foals and TFM were lighter at birth in ponies than Saddlebred horses, and both were lighter compared to draft horses. The surface density and volume fraction of microcotyledonary vessels was increased in pony compared to Saddlebred placentas. The relative expression of genes involved in growth and development was different between breeds and increased with maternal, fetal and placental weight. Primiparous dams produced lighter foals and smaller placentas, associated with a decreased volume fraction of microcotyledonary vessels and genes involved in growth and development and vascularization. Foal sex had little effect on placental structure and function as the expression of only one gene differed according to sex, with EGFR expression being decreased in placentas of females compared to males. In conclusion, foal and placental weight, as well as placental expression of genes involved in growth and development were correlated with maternal size. Placental structure also differed between breeds, with a stronger difference between ponies and both breeds of horses.

67 MATERNAL OBESITY AT CONCEPTION AND INSULIN SENSITIVITY IN LATE GESTATION ALTERS PLACENTAL STRUCTURE BUT NOT FETAL BIOMETRY AT BIRTH IN THE HORSE

Obesity is a major health issue in the horse industry, often associated with insulin resistance. This study aimed to analyse effects of maternal obesity at insemination and insulin resistance during late gestation on term placenta and foals at term. A total of 24 multiparous saddlebred mares were allocated to 1 of 2 groups at insemination: group obese (O) had a body condition score (BCS, French scale 1–5) >4 (N = 15), and group normal (N) had a BCS = 4 (N = 9). From insemination until the sixth month of gestation, all the mares had access to pasture. From wintering, they were housed in box stalls and fed the same amount of energy, proteins, and fibre as a percentage of their body weight. During all the gestation, BCS, basal blood glucose, insulin, triglycerides, and nonesterified fatty acid plasma concentrations were measured each month. At 300 days of gestation, a frequent sampling glucose tolerance test, enabling the simultaneous evaluation of insulin sensitivity (IS) and glucose tolerance, was performed. At birth, placentas and foals were measured. Placentas were sampled around the umbilical cord to perform structural and functional analyses by stereology and RT-qPCR. Results were analysed using a type 3 ANOVA taking into account mare group and foal sex. Effects were considered significant when P < 0.05. At 300 days of gestation, the frequent sampling glucose tolerance test indicated that half of the O mares were insulin resistant (IS <1; N = 8), whereas the other half were insulin sensitive (IS >1; N = 7). Based on these results, O mares were subdivided in 2 groups: obese resistant and obese sensitive (OS). Obese resistant mares were 61% less insulin sensitive than N mares and 59% less insulin sensitive than OS mares (P < 0.0001, P < 0.01, respectively). There was no difference for IS between N and OS mares. All analyses were thus performed comparing the 3 groups. In the N group, 3 mares were insulin resistant (one-third of the mares of the N group). Feed intake during wintering was not different between groups. Mares of both O groups maintained a high BCS (>4) during pregnancy, whereas N mares lost BCS down 2.75 at birth (<0.001). Basal glucose, insulin, triglycerides, and nonesterified fatty acid plasma were not different during gestation. At birth, no difference was observed for placental weight, surface and volume, nor for foal weight and withers height. The volume of allantoic vessels was reduced in placentas of OS mares compared with those of OR (P = 0.03; 78%) and N mares (P = 0.005; 65%). Moreover, placentas from OS mares had an increased volume of haemotrophic trophoblast (P = 0.03) and microcotyledonary vessels (P = 0.03) compared with N mares. No difference of expression was observed for 11 genes related to nutrient transfer, vascularization, growth, and development. Mare that are obese at insemination and insulin sensitive in late gestation appear to develop placental structural adaptations during gestation, possibly to increase fetoplacental exchanges, compared with obese, insulin resistant mares and lean mares. The monitoring of foal development (growth, metabolism, and osteoarticular status) is ongoing.

Context dependence of the olfactory perceptual range in the generalist land snail Cornu aspersum

Dispersal success in animals depends in part on their perceptual range, i.e., the distance from which they can acquire information about their environment. We studied how the olfactory perceptual range of a generalist species, the brown garden snail (Cornu aspersum (Müller, 1774)), varied under controlled conditions depending on the context in which stimuli were presented, whether alone or in the presence of another stimulus with opposite properties. Cornu aspersum preferentially orient themselves towards small nettle (Urtica dioica L.) patches, a highly palatable plant, and move away from repulsive plants if these stimuli are placed up to between 20 and 40 cm away from their starting point. A blend of palatable and repulsive plants, tested together, do not significantly influence the orientation of individuals in either direction. Cornu aspersum are thus capable of detecting and evaluating relatively small potential resource patches from a distance, enabling them to limit costly explorations, but this ability is context-dependent. These data could lead to a better understanding of the behaviour of C. aspersum in very heterogeneous landscapes in relation to this species’ ability to colonise a wide range of anthropised and fragmented habitats.

120 MATERNAL EXPOSURE TO DIESEL ENGINE EXHAUST DURING PREGNANCY AFFECTS EARLY EMBRYO DEVELOPMENT IN A RABBIT MODEL

Airborne pollution has been associated with various adverse effects on human reproductive health, especially intrauterine growth retardation and early pregnancy loss. However, few studies have analysed its effect on early development. Diesel exhaust particles (DEP) have been shown to alter blastocyst formation when diluted in embryo culture medium (2010 Toxicol Sci. 117, 200–208), but no data are available concerning the effect of maternal inhalation of diesel exhaust on early embryo development. Our study has been designed to answer this question using rabbit as a model and DEP doses mimicking daily exposure to traffic in large European cities. New Zealand female rabbits were superovulated by means of 5 subcutaneous administration of pFSH for 3 days before mating, followed 10 to 12 h later by an intravenous administration of 30 IU of hCG at the time of mating (natural mating). Dams were exposed to a representative air pollution mixture; that is, diluted diesel engine exhaust (1 mg m–3; N = 14) or clean air (N = 12), for 1 h every morning and afternoon, from Day 3 to Day 6 post-coitum (dpc). At 6 dpc, in vivo-developed embryos were collected from uteri perfused with PBS and counted; their diameter was measured on pictures using ImageJ software (NIH, Bethesda, MD, USA). Another group of female rabbits was exposed to the same inhalation conditions from 3 to 27 dpc without superovulation treatment. Measures by ultrasound were performed on these dams at 7 dpc. Data were analysed by Mann-Whitney test and ANOVA, including dams as cofactor. At 6 dpc, number of embryos per dams was higher in exposed group compared with control (P < 0.05). In contrast, embryo diameter was significantly lower in the DEP exposed group compared with the clean air exposed group (P < 0.01). Gene expression analysis is being performed in these embryos. At 7 dpc, ultrasound measurements evidenced a decrease in embryo diameter, perimeter, and volume in the exposed group compared with control (P < 0.01, P < 0.01, and P < 0.01, respectively). These data indicate that repeated exposure to airborne pollution even for daily short periods affects early development. Consequences of maternal DEP exposure on feto-placental development are under investigation.

119 A SHORT PERICONCEPTIONAL MATERNAL HYPERGLYCEMIA IS SUFFICIENT TO DISRUPT THE FETO-PLACENTAL PHENOTYPE IN A RABBIT MODEL

Pre-gestational type 1-diabetes (T1D) increases the risk of miscarriage and congenital malformations and programs the offspring to develop metabolic syndrome in adulthood. Management of maternal diabetes is essential during gestation but could be also highly important around the time of conception. Using a rabbit model, the effects of maternal T1D during the periconceptional period on pre-implantation blastocysts have been well documented, but the effects on feto-placental phenotype at 28 dpc (term = 31 days) has not been explored. Diabetes was induced by Alloxan in dams 7 days before mating. Glycemia was maintained at 15 to 20 mmol L–1 with exogenous insulin injections. At 4 dpc, embryos were collected and transferred into nondiabetic recipients. At 28 dpc, control (C) and diabetic (D) fetuses were collected for biometric records, placental analyses including stereology and gene expression, and lipid profiles of feto-placental tissues by gas chromatography. Lipid data were analysed by principal component analysis. D-fetuses were growth retarded, hyperglycemic, and dyslipidemic compared with C fetuses. Moreover, placental efficiency was much higher in D- than in C-fetuses. The volume density of fetal vessels was significantly decreased in D-placentas compared to C-placentas, whereas the volume density of trophoblast tended to increase (P = 0.051). This morphometric disruption was associated with a deregulation of the expression of genes related to nutrient supply and lipid metabolism. In fetal plasma, a specific fatty acid signature was observed in D- and C-groups. Moreover, the composition of placental and fetal liver membranes differed according to maternal status and fetal sex. Tissues from D-fetuses contained significantly more n-6 polyunsaturated fatty acids compared with C. Docosahexaenoic acid decreased whereas linoleic acid increased in the cardiac membranes of D-fetuses, indicating a higher risk of ischemia. This study demonstrates that exposure to high plasma glucose during the short periconceptional period is sufficient to adversely program fetal phenotype by reducing fetal growth, altering placental function and lipid profiles in all fetal tissues.

72 EFFECTS OF A PRECONCEPTIONAL AND GESTATIONAL MULTI-VITAMIN-MINERAL-OMEGA3 SUPPLEMENTATION ON FETOPLACENTAL DEVELOPMENT IN A RABBIT MODEL

The developmental origins of health and disease concept stipulates that nutritional imbalance in the preconceptional and gestational periods induces risks for the development and long-term health of the individual. In industrialized countries, most women take nutritional supplements before and during pregnancy, regardless of their diet. Potentially harmful embryonic and fetal effects of an excess multi-micronutrient supplementation, however, are not known. The objective of this study was to assess, using a rabbit model, the effects of a multi-vitamin-mineral-omega 3 supplementation (Gestarelle® G) administered in slight excess during preconceptional and gestational periods. Twenty-seven New Zealand does were used. Supplementation contained vitamins (B1, B2, B5, B6, B8, B9, B12, C, E), minerals (magnesium, iron, copper, iodine), and n-3 polyunsaturated fatty acids (n-3 PUFA; namely DHA and EPA) in the form of oral capsules administered daily from 10 days before mating until the end of pregnancy. Group 1X posology allometrically matched that used in humans (n = 10), representing an excess of 6 to 50% of rabbit needs. Group 3X rabbits received 3 times the recommended posology (n = 9; 18 to 150% of rabbit needs), whereas the control group received capsules containing only excipient (n = 8). Embryo and fetoplacental development was monitored by transabdominal two- and three-dimensional ultrasound and Doppler on Days 5, 7, 14, 21, and 28 with 5 conceptuses per doe examined each time. Does were killed on Day 28 (term = 31 days), 24 h after the last capsule administration. Maternal and fetal blood, as well as placenta and fetal organs, were collected. Data were analysed by ANOVA and Kruskall-Wallis as appropriate. The n-3 PUFA plasma concentrations increased in the 3X does (P < 0.01). Fetal plasma concentrations were increased only in males in the 1X group (P < 0.001) and both in females and males in the 3X group (P < 1e–15). No statistical differences were observed between treatments for any of the ultrasound or Doppler analyses. In total, 208 fetuses were collected with no difference in litter size, sex ratio, nor biometric measurements between groups. No structural anomaly was observed by histological analysis of fetal organs. In conclusion, multi-vitamin-mineral-omega 3 supplementation administered during the preconceptional and gestational periods in a rabbit model in the present study and within the measured parameters had no deleterious effect for dam and conceptuses at 1 and 3 times the recommended posology. Further work is on-going to study the effects on fetal brain development. Further studies are needed to evaluate putative post-natal effects. This study was funded by IPRAD Santé Laboratories, Paris, France.

159 EFFECTS OF EMBRYO TRANSFER IN A LARGER BREED ON POSTNATAL GROWTH AND GLUCOSE METABOLISM IN HORSES

In equids, the size of the uterus determines fetal intrauterine development, which in turn affects postnatal insulin sensitivity and growth rate. We induced intrauterine growth enhancement through embryo transfer using Pony (P), Saddlebred (S), and Draft (D) horses and studied growth and insulin sensitivity in foals from birth to one year of age. Control pregnancies of S-in-S (n = 14) and P-in-P (n = 10) were obtained by AI. Enhanced fetal growth was obtained by transferring S (S-in-D, n = 7) and P embryos (P-in-D, n = 5) into D mares. From birth to weaning (180 days), each foal was kept with its surrogate P, S, or D dam. At 3, 140, and 380 days, glucose clearance and pancreatic beta cell response to exogenous glucose were assessed with an intravenous glucose tolerance test (IVGTT). At 200 days, the euglycemic-hyperinsulinemic clamp method was used to determine the sensitivity and responsiveness of tissues to exogenous insulin. Plasma T3, T4, and IGF1 were assayed at 3 and 180 days. Data were analysed using one-way ANOVA and Tukey post hoc tests. S-in-S were heavier and taller than P-in-P from birth to 380 days (P < 0.001). Before weaning, plasma concentrations of several hormones involved in growth were lower in S-in-S than P-in-P (at 3 and 180 days, respectively, T3: P = 0.08 and P = 0.02, T4: P < 0.001 and P = 0.06, IGF1: P = 0.04 and P < 0.001). No difference was found in glucose regulation between these groups. In contrast, post-weaning insulin resistance was observed in P-in-P at 200 days (P < 0.001) and confirmed at 380 days where they exhibited slower glucose clearance (P = 0.03) associated with higher fasting glucose (P < 0.001) than S-in-S. Fetal growth was not enhanced in S-in-D with no difference in height and weight at birth. Although S-in-D grew faster from 30 to 140 days, growth rates were not different from S-in-S after weaning, weaning coinciding with lower T3 (P < 0.001) in S-in-D than in S-in-S. Glucose regulation was not different between the two groups, but insulin remains to be assayed at 140 and 380 days. Fetal growth was enhanced in P-in-D: at birth, they were heavier (P = 0.01) and taller (P < 0.001) than P-in-P. Growth of P-in-D was faster until weaning. No more difference, however, was observed between P-in-D and P-in-P at 380 days. Plasma concentrations of T3 (P = 0.03) and those of T4 (P < 0.001) were lower at 3 days and T3 was still lower at 180 days (P < 0.001) in P-in-D compared with P-in-P. Moreover, P-in-D developed early insulin resistance: insulin secretion was higher in P-in-D compared with P-in-P (P = 0.002) after IVGTT at 3 days. At 200 days, however, P-in-D and P-in-P had the same sensitivity to insulin. There was no difference in glucose clearance rates at 380 days, but P-in-D had lower fasting glucose (P < 0.001) than P-in-P. Insulin assays at 140 and 380 days are pending. In conclusion, these data indicate that transfer of a small breed embryo into a large breed mare and subsequent suckling by the recipient mare can enhance fetal and postnatal growth and affect the foal’s glycaemia and sensitivity to insulin at birth and in subsequent months. Ongoing work includes analyses of milk samples and effects on general health.

98 THE FETAL AND POSTNATAL EFFECTS OF PERICONCEPTIONAL HYPERGLYCEMIA USING A RABBIT MODEL

The objective of this study was to analyse the effects of maternal hyperglycemia during the periconception period on fetal and postnatal development using a rabbit model. Diabetes was induced in adult New Zealand female does by a single intravenous alloxan injection (group D). Glycemia was maintained between 3 and 5 g L–1 with 2 daily subcutaneous insulin injections. Does from group D and contemporary controls (group C) were naturally mated 1 wk after induction of diabetes in group D animals, without superovulation, and embryos collected after sacrifice on Day 4 post-coitum. In Expt. 1, embryos from D (n = 11) and C (n = 13) groups were transferred respectively to the right and left horns of 3 female recipients, which were killed on Day 28. In Expt. 2, 17 D and 16 C females were mated naturally. 68 D and 98 C embryos were collected and transferred to 26 non-diabetic recipients (6–7 embryos per doe). 17 females were pregnant with no difference between C and D recipients. Fetal development was monitored by ultrasound. At birth, litters were equilibrated in number. 15 D and 7 C pups (from 4 D and 2 C litters) were killed at weaning (1 month of age). The remaining 13 D and 26 C pups (4 D and 7 C litters) were allocated to 1 of 2 feeding groups: control or obesogenic diet. The obesogenic diet was based on the control diet supplemented with animal fat (suet, 200%) and glucose (200%). Bodyweight, adiposity, and glucose metabolism were monitored until sacrifice at 5 months of age. Data were analysed by ANOVA using litter (fetuses), litter and sex (weaning), and litter and diet (5 months) as co-factors. Sex effects at 5 months were not analysed due to small numbers of animals. Fetal development was not different between D and C groups. Fetal, placental, and fetal organ weights did not differ at Day 28, except for brain weight, which was significantly lower in D fetuses (0.86 ± 0.1 v. 1.05 ± 0.08 g; P < 0.05). There was no difference in litter size at birth (3.3 ± 0.3 v. 3.6 ± 0.5 pups for D and C groups, respectively; P = 0.5), but birthweight was significantly increased in D offspring (211 ± 6 v. 194 ± 6 g; P < 0.05). There was no difference in weight after 14 days. At 1 month of age, adiposity, plasma insulin and leptin concentrations were not different between groups. In contrast, in male D offspring, fasting glycemia was significantly lower (1.7 ± 0.2 v. 2.1 ± 0.02 g L–1; P < 0.01), plasma insulin-like growth factor 1 was significantly increased (P < 0.05) and kidney/bodyweight ratio was significantly reduced (0.41 ± 0.03 v. 0.45 ± 0.04; P < 0.01). From 12 wk of age, bodyweight became significantly different between D and C groups and according to diet (P < 0.005), with D individuals being lighter than C individuals for each dietary group. Finally, fasting glycemia was significantly higher in the animals fed the obesogenic diet (1.35 ± 0.05 v. 1.19 ± 0.05 g L–1; P < 0.04), regardless of group. These data suggest that maternal hyperglycemia during the periconceptional period affects glucose metabolism and organ development in offspring with sexual dimorphism.

84 DEVELOPMENTAL EFFECTS OF CHOLESTEROL-ENRICHED HYPERLIPIDIC DIETS IN RABBITS

The long-term effects of low birth weight on the onset of adult metabolic diseases have been described in several species. Low birth weight is usually due to maternal food restriction or placental insufficiency during pregnancy. In contrast, the effects of maternal high planes of nutrition or dietary unbalance have been less studied. In previous work, we have shown that the administration of a hypercholesterolemic (0.2%) hyperlipidic (8%) diet to rabbit does throughout puberty and subsequent pregnancy significantly affected fetal and postnatal development of their offspring. The aim of this study was to determine the relative role of cholesterol and of the n3/n6 fatty acid ratio on fetal development using the same model. Twenty-five does were fed from 10 weeks of age with 1 of 5 diets: S or SC (soy oil, 9% lipids, n3/n6 = 0.15 without or with 0.1% cholesterol), R or RC (rapeseed, 8% lipids, n3/n6 = 0.38 without or with 0.1% cholesterol), or Co (2% lipids, no cholesterol). Does were bred at 18 weeks of age. The growth of 4 fetuses per dam was monitored by weekly ultrasound scanning (Chavatte-Palmer P et al. 2008 Theriogenology). All pups were weighed at birth. Additionally, fetuses and placentas from a further 2 to 4 does per treatment group were collected at 15 days of pregnancy. Data were analyzed by ANOVA using group, number of fetuses, and dam as variables. Because different measurements were taken at different stages, the analysis of repeated measures could not be performed. Although maternal lipid intake was significantly increased in all treated groups, does adjusted their intake so that caloric intake and doe body weight were not statistically different between groups. Mean litter size was not different between groups. Group and, to a lesser extent, dam, but not litter size, significantly affected fetal development. Conceptus length was reduced on Day 9 of pregnancy in SC (–14%) and RC (–5%) compared with Co (P < 0.05). Fetal length was reduced on Day 15 in groups RC and S (both –9%, P < 0.05) but not in SC. Fetal abdominal perimeter measurements were smaller in SC, RC, and R at Day 22 and in SC at Day 28 (P < 0.05). Birth weight was decreased in all groups (–10, –18, –21, and –7% in SC, S, RC, and R, respectively compared with Co). Postmortem analyses did not show significant differences in fetal weight at Day 15, but placentas were significantly heavier in S compared with all other groups (P < 0.05). In conclusion, long-term cholesterol- and lipid-enriched diets induced intrauterine growth retardation in does with variable effects according to the n3/n6 fatty acid ratio and cholesterol dietary content, with more significant effects observed in the cholesterol-enriched diets. These effects are observed very shortly after implantation, indicating that oocyte and/or embryonic development may be affected. Further work is in progress to evaluate these early effects, using micro-array analysis of gene expression in the embryos.

Centrosome inheritance in sheep zygotes: centrioles are contributed by the sperm

The inheritance and duplication of the sperm centriole in the sheep zygote was studied by transmission electron microscopy. We found two centrioles at one pole and a single centriole at the opposite pole of the first mitotic spindle, in monospermic eggs, 20-21 hours postinsemination. This indicated both duplication and relocation of centrioles to opposite spindle poles during fertilization. The absence of centrioles in mature sheep oocytes was confirmed. Following activation by the calcium ionophore A 23187, mature oocytes entered mitosis and formed a bipolar spindle 18 hours later. Centrioles were not detected in the mitotic spindle of parthenogenotes. Androgenetic eggs were obtained by excision of the anaphase II/telophase II meiotic spindle of fertilized eggs. They were capable of undergoing mitosis and formed one or two bipolar spindle(s) in monospermic and dispermic eggs, respectively, 20-24 hours postinsemination. In two monospermic androgenetic eggs, two centrioles were found at one pole and a single centriole at the opposite pole of the first mitotic spindle. Three centrioles were also observed in another androgenetic egg in prometaphase of the first mitotic division, in close vicinity to the sperm neck-piece. These data provide evidence that the sperm centriole do reproduce and occupy a pivotal position on opposite spindle poles at syngamy. Altogether, the present findings suggest that centrioles of sheep zygotes are paternally derived.

Meiotic competence of in vitro grown goat oocytes

The objective of the present study was to grow meiotically incompetent goat oocytes from early antral follicles in vitro and to render them competent to undergo germinal vesicle breakdown. Cumulus-oocyte complexes with pieces of parietal granulosa cells were isolated from follicles 0.35-0.45 mm in diameter using both mechanical and enzymatic methods. The cumulus-oocyte complexes were divided into two groups according to oocyte diameter (group A: < 95 microm; group B: > 95 microm) and cultured for 8 or 9 days on granulosa cell monolayers. Within 8 days of culture, the mean oocyte diameter increased from 86 +/- 0.4 microm to 95 +/- 0.7 microm in group Aand from 106 +/- 0.2 microm to 109 +/- 0.5 microm in group B. After 9 days of culture, the mean diameter of oocytes from groups A and B were 99 +/- 0.5 microm and 112 +/- 0.4 microm, respectively. The meiotic competence of oocytes grown in vitro was evaluated by in vitro maturation. Within 8 days of culture, only 3% of oocytes from group A and 6% of oocytes from group B acquired the ability to undergo germinal vesicle breakdown. After 9 days of culture, 7% of group A oocytes and 42% of group B oocytes were competent to resume meiosis. The expression of p34(cdc2) in oocytes grown in vitro was analysed by the western blot technique. During 9 days of culture, p34(cdc2) accumulated in both groups of growing oocytes, but its concentration was lower than in fully grown oocytes used as controls. The results showed for the first time that goat oocytes from early antral follicles can grow, accumulate p34(cdc2) and acquire the ability to resume meiosis, when cultured for 9 days on granulosa cell monolayers.

Meiotic competence of in vitro grown goat oocytes

The objective of the present study was to grow meiotically incompetent goat oocytes from early antral follicles in vitro and to render them competent to undergo germinal vesicle breakdown. Cumulus-oocyte complexes with pieces of parietal granulosa cells were isolated from follicles 0.35-0.45 mm in diameter using both mechanical and enzymatic methods. The cumulus-oocyte complexes were divided into two groups according to oocyte diameter (group A: < 95 microm; group B: > 95 microm) and cultured for 8 or 9 days on granulosa cell monolayers. Within 8 days of culture, the mean oocyte diameter increased from 86 +/- 0.4 microm to 95 +/- 0.7 microm in group Aand from 106 +/- 0.2 microm to 109 +/- 0.5 microm in group B. After 9 days of culture, the mean diameter of oocytes from groups A and B were 99 +/- 0.5 microm and 112 +/- 0.4 microm, respectively. The meiotic competence of oocytes grown in vitro was evaluated by in vitro maturation. Within 8 days of culture, only 3% of oocytes from group A and 6% of oocytes from group B acquired the ability to undergo germinal vesicle breakdown. After 9 days of culture, 7% of group A oocytes and 42% of group B oocytes were competent to resume meiosis. The expression of p34(cdc2) in oocytes grown in vitro was analysed by the western blot technique. During 9 days of culture, p34(cdc2) accumulated in both groups of growing oocytes, but its concentration was lower than in fully grown oocytes used as controls. The results showed for the first time that goat oocytes from early antral follicles can grow, accumulate p34(cdc2) and acquire the ability to resume meiosis, when cultured for 9 days on granulosa cell monolayers.