Imprinted genes in mammals are expressed from only one of the parental chromosomes, and are crucial for placental development and fetal growth. The insulin-like growth factor II gene (Igf2) is paternally expressed in the fetus and placenta. Here we show that deletion from the Igf2 gene of a transcript (P0) specifically expressed in the labyrinthine trophoblast of the placenta leads to reduced growth of the placenta, followed several days later by fetal growth restriction. The fetal to placental weight ratio is thus increased in the absence of the P0 transcript. We show that passive permeability for nutrients of the mutant placenta is decreased, but that secondary active placental amino acid transport is initially upregulated, compensating for the decrease in passive permeability. Later the compensation fails and fetal growth restriction ensues. Our study provides experimental evidence for imprinted gene action in the placenta that directly controls the supply of maternal nutrients to the fetus, and supports the genetic conflict theory of imprinting. We propose that the Igf2 gene, and perhaps other imprinted genes, control both the placental supply of, and the genetic demand for, maternal nutrients to the mammalian fetus.

The interacting influences of maternal size and fetal genotype on placental and fetal development in the mare were assessed by comparing conventional within-breed Thoroughbred (Tb-in-Tb, n = 7) and Pony (P-in-P, n = 7) control pregnancies established by artificial insemination (AI) with between-breed (Tb-in-P, n = 8; deprived in utero condition and P-in-Tb, n = 7; luxurious in utero condition) experimental pregnancies established by embryo transfer. All foals were born spontaneously and the mean (+/- SEM) duration of gestation in the two groups of control mares was significantly different (P < 0.001) at 325 +/- 3.0 days for the P-in-P pregnancies and 339 +/- 3.0 days for the Tb-in-Tb pregnancies, whereas the durations of gestation for the two experimental groups were very similar and midway between those of the control pregnancies at 332 +/- 2.8 days for the Tb-in-P and 331 +/- 2.7 days for the P-in-Tb. Mean (+/- SEM) foal birth weight and the mean (+/- SEM) values for the mass, gross area and volume of the allantochorion were all highest in the seven Tb-in-Tb pregnancies (53.1 +/- 2.6 kg, 3.8 +/- 0.3 kg, 12.9 +/- 0.3 x 10(3) cm(2), 3.5 +/- 0.2 l, respectively) and lowest in the seven P-in-P control pregnancies (24.0 +/- 1.3 kg, 1.7 +/- 0.1 kg, 8.3 +/- 0.3 x 10(3) cm(2), 1.8 +/- 0.1 l, respectively). These parameters were higher in the seven P-in-Tb pregnancies (37.9 +/- 2.1 kg, 2.7 +/- 0.1 kg, 10.1 +/- 0.5 x 10(3) cm(2), 2.5 +/- 0.1 l, respectively) than in the eight Tb-in-P (33.0 +/- 2.4 kg, 2.3 +/- 0.2 kg, 9.0 +/- 0.5 x 10(3) cm(2), 2.1 +/- 0.1 l) experimental pregnancies. Foal birth weight was positively correlated with the mass (r = 0.84, P < 0.001), gross area (r = 0.87, P < 0.001) and volume (r = 0.91, P < 0.001) of the allantochorion, and maternal weight was also positively correlated with both the mass and gross area of the allantochorion (r = 0.64 and 0.69, respectively; both P < 0.001). Application of stereology to multiple random biopsies recovered from each placenta produced mean values for the surface density of microcotyledons on the allantochorion (S(v)). Values were higher in Thoroughbred than in Pony mares regardless of the breed of fetus being carried. Multiplication of S(v) by the volume of the allantochorion to give the total microscopic area of fetomaternal contact at the placental interface was also positively correlated with foal birth weight (r = 0.84, P < 0.001). Foal birth weight was determined by the microscopic area of fetomaternal contact of the placenta and there were no differences in foal weight per m(2) of placenta regardless of fetal or maternal genomes. Thus, the results indicate that in equids, maternal size interacts with both the maternal and fetal genotypes to control the rate and extent of fetal growth by influencing the gross area of the diffuse allantochorion, and the density, complexity and depth of the microcotyledons on its surface.

A tough, elastic glycoprotein capsule envelops the equine blastocyst between Days 6 and 23 after ovulation. It maintains the spherical configuration of, and provides physical support for, the embryo as it traverses the entire uterine lumen during Days 6-17, propelled by myometrial contractions that are stimulated by pulsatile release of prostaglandin F2alpha and prostaglandin E2. The capsule also accumulates constituents of the exocrine secretions of the endometrial glands ('uterine milk') as nutrients for the mobile embryo as it releases its antiluteolytic maternal recognition-of-pregnancy signal to the whole of the surface of the endometrium. Mobility ceases abruptly on Day 17 with a sudden increase in uterine tonicity that 'fixes' the conceptus at the base of one of the uterine horns. At Day 35, the trophoblast of the spherical conceptus has separated into its invasive and non-invasive components. The former, distinguished as the thickened, annulate chorionic girdle, invades the maternal endometrium to form the unique endometrial cups. These secrete a chorionic gonadotrophin that synergizes with pituitary follicle-stimulating hormone to induce secondary luteal development in the maternal ovaries. The cup cells express foreign fetal antigens that stimulate strong maternal humoral and cell-mediated immune responses, which curtail their lifespan. The non-invasive trophoblast of the allantochorion establishes a stable microvillous contact with the endometrial epithelium around Day 40 and, over the next 100 days, develops a complex multibranched interdigitation with the endometrium to form the microcotyledonary haemotrophic exchange units that cover the entire surface of the diffuse epitheliochorial placenta. Reduction in the effective total area of fetomaternal contact at this placental interface, by competition between twin conceptuses for the limited area of available endometrium, by attachment of the allantochorion to an imperfect endometrium in a mare with endometrosis, or following cross-breeding or embryo transfer between a sire and dam of dissimilar size, will all induce intrauterine growth retardation of the fetus and runting of the foal, which persists into adult life. Over 40 years ago, Professor Roger Short and his colleagues determined that the high concentrations of conventional and unique ring B unsaturated oestrogens in the blood and urine of mares during the second half of pregnancy stem from placental aromatization of large quantities of C-19 precursor molecules secreted by the temporarily hypertrophic fetal gonads. Placental production of progesterone and 5alpha-reduced progestagens, on the other hand, depends on both maternal and fetal adrenal sources of pregnenelone.

Within-breed artificial insemination and between-breed embryo transfer were carried out in small pony (P) and large Thoroughbred (Tb) mares to create 4 types of horse pregnancy in which the fetus experienced spatial and nutritional deprivation (Tb-in-P; n=8), luxury (P-in-Tb; n=7) or normality (Tb-in-Tb; n=7 and P-in-P; n=7) in utero.Measurement of equine chorionic gonadotrophin (eCG), total conjugated oestrogens and progestagen concentrations in serial peripheral serum samples recovered from all the mares throughout gestation showed that the amount of eCG produced during the first half of gestation was dependent upon the breed of the mare rather than the breed of the fetus being carried. In contrast, the mean total amounts of oestrogens produced, as measured by area under the curve, were significantly greater (P=0·003) in the two types of pregnancy in which a Thoroughbred fetus was being carried (Tb-in-Tb and Tb-in-P) than those in which a pony fetus was gestated (P-in-P and P-in-Tb); the evidence suggests that the Tb fetus may have larger gonads than the P fetus and thereby secrete more C-19 precursor steroids for aromatisation to oestrogens by the placenta. In the final weeks of pregnancy mean plasma progestagen concentrations rose much earlier, and to significantly higher levels (P<0·001), in the Tb-in-P than in the P-in-Tb pregnancies, thereby reflecting the increased fetal stress in the former causing premature maturation of the fetal adrenal gland. This, in turn, resulted in increased secretion of pregnenolone by the adrenal cortex for conversion to progestagens by the placenta.

Specific radioreceptor assays for FSH and LH, which employ tissue receptors from rat testis and highly purified human FSH (LER 1575-C) and LH (Hartree IRC-2, 24/6/69) as standards, have been developed to determine the FSH-like and LH-like activities in pregnant mare serum gonadotropin (PMSG). Measurements of FSH and LH concentrations in the serum of six pregnant Pony mares showed that the ratio of these two activities did not vary significantly between mares and remained constant between days 40 and 80 of gestation with a value of 1-45 +/- 0-04 (S.E.M.). The FSH:LH ratio of PMSG produced by cultured equine trophoblast cells was found to be 0-72 +/- 0-03 (S.E.M.) and that of partially purified serum extracts of PMSG 1-08 (range 0-87-1-30).