| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Zoology Department, Oxford University, Oxford OX1 3PS, United Kingdom
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The association between high IGF levels and edema has been observed in a variety of situations. Infusion of IGF-1 into humans leads to a mild systemic edema that is accompanied by increased movement of substances from the circulation to extravascular space in both the skin and the retina [1, 2]. In the mouse conceptus, the ratio of wet to dry weight is greatly increased when IGF-2 levels are elevated after mutations in the insulin-like growth factor-2/mannose 6-phosphate receptor gene (Igf2r; [3, 4]). The embryo is also edematous, and the volume of the exocelomic and amniotic fluid is doubled in mice with the Thp deletion, which includes the Igf2r gene [5, 6]. In adult mice expressing Igf2 transgenes, local IGF-2 excess provokes unusual growth, with high ratios of wet weight to DNA content in the alimentary canal, skin, and uterus, in which the transgene is expressed [7]. In the adult, the uterus shows edema, whereas both wet and dry weight increase coordinately in the alimentary canal [8].
The fluid content of mouse embryonic tissues lacking an active Igf2 gene has not been described. The reduced wet weight in mice lacking IGF-2 has been attributed to a reduced cell number because cell size is not obviously altered in histological sections [9–12]. However, recent analysis of parts of these mice at 3 mo old shows that they differ in wet and dry weight from their normal litter mates whereas the content of soluble protein and DNA is not significantly changed [8].
We analyzed conceptus composition when IGF-2 production is limited to only some of its constituent parts. To discover which of the components of the conceptus regulate fluid content and growth in the uterus, inner cell masses (ICMs) and trophectoderm vesicles were exchanged between blastocysts that had (Igf2 +m/+p, normal) or lacked (Igf2 +m/-p, Igf2 ko) a functional Igf2 allele inherited from the father.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
All experimental procedures were conducted under licenses issued by the Home Office (UK) in accordance with the Animals (Scientific Procedures) Act 1986.
Male Igf2 +m/-p mice, kindly provided by Prof A. Efstratiadis (Columbia University, New York, NY), were bred for 6–8 generations with 129J/Sv females [13]. Igf2 +m/-p or Igf2 +m/+p males were mated by 129J/Sv or closed albino PO stock females. To identify the tissues that regulated growth, the ICMs and trophectoderm vesicles were interchanged, with the ICMs stripped of trophectoderm after immunosurgery, and the recombined blastocysts were returned to the uterus of PO females on Day 2.5 of pseudopregnancy [14–16]. The mating age of the normal pregnant or recipient pseudopregnant mouse was used to gauge the embryonic age at autopsy.
At embryonic day (E)16.5, the recipients were killed by cervical dislocation, and the conceptuses were dissected into PBS lacking Ca2+ and Mg2+. The "total placenta" consisted of the maternal decidua basalis (decidual base) and maternal and fetal cells in the spongiotrophoblast and labyrinthine trophoblast. Reichert's membrane was trimmed away, and the visceral yolk sac and umbilical cord were severed at their base to give the yolk sac and fetus parts. The fetal placenta consisted of the labyrinthine trophoblast, dissected free of most of the spongiotrophoblast, which is whiter and less vascular [17, 18]. The endodermal sinuses of Duval were also included in the fetal placenta: they derive from the ICM. The amnion was removed from around the embryo and discarded.
Measurements
In all experiments, the wet weights are the weights after partial removal of fluid from around the tissue with absorbent paper, with the tissue held on a plastic coffee spoon or dish. In the initial observations of nonmanipulated conceptuses, dry weight measurements were also made by drying parts at 68°C for eight days. The DNA content was measured with a fluorometric technique [19], modified by the addition of 2 M salt to the assay buffer in the initial observations but not in the blastocyst recombination work. This modification slightly increased the apparent DNA content. In the initial observations, the DNA content of the fetus plus total placenta was measured (the yolk sac was taken for polymerase chain reaction [PCR] typing). In the recombination experiments, the DNA content of the yolk sac and fetal placenta were estimated by taking fragments, weighing the fragment and the rest of the sample separately, measuring the fragment's DNA content, and calculating the content of the whole. The fetal DNA content was measured on the whole.
The fluid volume of the exocelomic and amniotic cavities was measured in two ways. Both methods used the assumption that the specific gravity of the fluid and the tissues equalled 1 g per ml. In the initial observations, the wet weights of the dissected total placenta, the yolk sac, and the fetus were subtracted from the total wet weight of the conceptus with intact membranes. In the recombination experiments, the volume bounded by the yolk sac was estimated by photographing the isolated whole conceptuses over graph paper soon after the death of the mother, taking the mean of the short and long axis of the yolk sac, and using this mean as the diameter of a sphere in calculating volume. These measurements were corrected for optical distortion caused by the different distances between the conceptus, the graph paper, and the lens. The volume contained in the yolk sac and fetus (from wet weight) was then subtracted from the volume bounded by the yolk sac.
Genotyping
DNA was extracted as described [20]. The transmission of the Igf2 ko allele was first followed by the PCR, amplifying a 2.1-kilobase (kb) fragment between Igf2 exon 3 and neomycin [13]. Some of the results were double-checked with the primers Neo-left and Neo-Right1, which amplify a 555-base pair (bp) region from neomycin [21]. The polymerase was from Thermus "icelandicus" (Advanced Biotechnologies, Epsom, Surrey, UK), with a hot start for 3 min at 94°C, and 35 cycles of 94°C for 1 min, 65°C for 1 min, and 70°C for 4 min, finishing with 10 min at 72°C. For the Neo primers, the 65°C step was left out.
Statistical Analysis
Litters and conceptuses were selected for statistical analysis, which was routinely performed with Student's t-test. The initial observations on nonmanipulated conceptuses concerned 62 conceptuses in 9 litters (wet and dry weight) and 66 conceptuses in 8 litters (wet weight and DNA content). In the blastocyst recombination and transplantation experiments, 5 litters with fewer than 4 viable conceptuses were excluded. The remaining 67 viable conceptuses were distributed among 13 litters. Conceptuses with placentae flattened against each other contributed to all the data except the effect of their genotype on their own weight (one conceptus apposed to a blighted twin and two apposed to each other). The following features were not scored or considered further: two early resorptions marked by dark spots, one small placenta without a yolk sac or fetus, and one small placenta with a yolk sac and traces of fetal debris.
The interaction between position and growth of the conceptus was analyzed by scoring position in two ways [22]. First, position numbers were given to each viable embryo by their order along the uterus, starting with 1 at the ovarian end. Second, position values were given by the proportion of the length they would occupy if they were evenly distributed: these values take into account the number of conceptuses in the horn.
The analysis of local interactions between conceptuses was mainly conducted with conceptuses in the body of the horn. The conceptuses at each end of the horn had a single neighbor and they were treated separately. First, the quantitative effect of neighbors was analyzed by adding the weight of the neighbors on either side and looking for an interaction with the weight of the central conceptus. In this analysis, no account was taken of the genotype. Second, the genotypes of the conceptus and its neighbors were considered by dividing neighbor types into categories. For instance, when the ICM was Igf2 +m/+p in a wild-type trophectoderm vesicle, then the neighbors might each be Igf2 +m/-p (category N:2ko), or be Igf2 +m/+p on one side and Igf2 +m/-p on the other side (category N:1ko), or each be wild-type (category N:0ko). The interactions between these neighbor categories and the weight of the central conceptus were studied. Last, the tendency of fetuses to be smaller towards the cervical end of the uterus [22] was analyzed by comparing the weight of each fetus with the weight of its neighbor on the cervical side and vice versa, thus including the conceptus at each end in the study.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Nonmanipulated Conceptuses
To provide a baseline for the recombination experiments, observations were made on nonmanipulated conceptuses from 129J/Sv matings. The wet weight, dry weight, and DNA content of conceptuses were compared at E16.5 (Table 1).
|
To reduce the variation between the size of conceptuses in different litters, litters were selected in which the wet weight of the normal conceptuses did not exceed 1 g in both the wet and dry weights and the DNA content series. For the wet and dry weight series, the Igf2 ko total conceptus weights were significantly decreased compared to the normal ones (35% wet weight, 37% dry weight, mean 3.78 viable conceptuses per horn). There was no evidence that the Igf2 ko conceptuses were unusually dry. In the DNA content litters, the Igf2 ko samples had 31% lower total conceptus weights and 32% lower total placenta + fetus wet weights (mean 3.67 viable conceptuses per horn). The estimated fluid volume of the exocelomic and amniotic cavities was reduced by 25%. However, tissue composition was altered because the DNA content was only reduced by 15%. These results showed that the Igf2 +m/-p genotype severely reduced wet and dry weight growth but that it had a minor effect on DNA content.
Recombination Experiments
The development of the recombined blastocysts to viable E16.5 conceptuses was recorded (Table 2). There was no evidence that an Igf2 ko component in either the trophectoderm or the ICM derivatives influenced the success of development after the operation and transplantation.
|
Further analysis was restricted to the 13 litters with 4–6 viable conceptuses in one horn of the uterus, with a mean of 5.15 viable conceptuses per horn (see Materials and Methods). The combinations of genotypes produced by microsurgery were ICM Igf2 ko + normal trophectoderm, normal ICM + trophectoderm Igf2 ko, and normal ICM + normal trophectoderm.
Volume of Fluid in the Exocelomic and Amniotic Cavities
When either the trophectoderm or the ICM was Igf2 ko, then the volume bounded by the yolk sac at E16.5 was reduced (Table 3). The fluid volumes in the exocelomic and amniotic cavities were estimated, and they were significantly reduced when the yolk sac and fetus were Igf2 ko (ICM series, Table 3). Thus IGF-2 deficiency in the whole conceptus (Table 1) or in ICM derivatives reduced fluid transport from the mother or fluid retention by the conceptus.
|
Wet Weight of Placenta, Yolk Sac, and Fetus
When the E16.5 conceptuses were derived from blastocysts that were Igf2 ko in either the trophectoderm or the ICM, then the wet weights of all three parts were significantly reduced (Table 4). With the ICM Igf2 ko, the weight losses were greatest in the yolk sacs and fetuses (26–27%). With the trophectoderm Igf2 ko, the weight losses were greatest in the placentas (21%). The severest effects of IGF-2 deficiency on wet weight were therefore localized to the regions that lacked active Igf2 genes. Note that in these recombination experiments, the fetal placenta fraction did not contain all the derivatives of the trophectoderm: the giant cells and much of the spongiotrophoblast were removed to clear most maternal components.
|
DNA Contents of Placenta, Yolk Sac, and Fetus
In general, the DNA contents of the three parts of the conceptus showed a pattern of reduction similar to that of wet weights (Table 5). However, in no case were the reductions significant.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
IGF-2 production by the conceptus begins to alter its size from E10.5. Substantial quantities of Igf2 mRNA can first be detected by autoradiography at E5.5 in trophectoderm derivatives, at E6.5–7.0 in the extra-embryonic endoderm and extra-embryonic mesoderm, and at E7.5 in the mesoderm of the embryo [23]. Immunohistochemistry suggests that at least some of this Igf2 mRNA is translated into protein by E8. The absence of Igf2 transcripts markedly reduces the wet weight of the fetus at about E11 and the wet weight of the placenta at about E13 [9]. The reduced size of mice lacking an active paternal Igf2 gene persists throughout life, despite the expression of the maternal allele in the exchange tissues around the central nervous system [10].
Subtle effects of IGF-2 levels on cell interactions might precede the large effects on wet weight. For instance, exogenous IGF-2 promotes trophoblast giant cell transformation in cell culture [24]. In the late blastocyst, the ICM orients and induces the proliferation of the trophectoderm [14, 25], and it has been argued that the mechanical properties of the growing extraembryonic ectoderm derivative of the trophectoderm shape the growing ICM descendants [26]. These early postimplantation interactions show that the growth of one part is dependent on the other. In addition, the growth of the whole conceptus is clearly subject to size regulation, because combined whole morulae develop into pups of normal size [27–29].
Later in development, it is likely that the size and functions of the placenta and yolk sac set upper limits on the growth of the fetus. However, recent studies show that there is no consistent relationship between the size of the placenta and fetus. For instance, the placenta may be of normal size and the fetus less than half-size (e.g., Igf1 ko and Igf1r ko; [9, 30]). The present paper explores the consequences of the combined effects of early and later interactions between trophectoderm and ICM derivatives.
Variables in the Recombination Experiments
Many genetic and environmental factors can influence the weight of the mouse conceptus. In the nonmanipulated conceptuses, no effects of litter size, uterine position, or nearest neighbors' genotype on the total wet weights of the Igf2 +m/-p conceptuses were observed (n = 61 viable Igf2 +m/-p conceptuses). In the recombination experiments, analysis was limited to E16.5 litters with 4–6 viable fetuses confined to one uterine horn, thus reducing or excluding the following variables: gestation length [31], litter size [32], systemic effects between conceptuses in different horns [33], and uterine position effects resulting from 8 or more conceptuses in a single horn [34]. In the present small experimental series of litters, the following additional variables did not have significant effects: number of viable embryos in the litter, position in the uterus, or the size or Igf2 status of nearest neighbors. In each recombination series, the comparisons were between manipulated normal conceptuses (normal ICM + normal trophectoderm) and manipulated experimental combinations in the same litters (e.g., ICM Igf2 ko + normal trophectoderm).
Placental Growth
When no IGF-2 was expressed in the derivatives of either the trophectoderm or the ICM, then the placenta + fetus had 68% normal wet weight at E16.5, confirming the original description of the phenotype (Table 1, [10]). When only one of these components lacked IGF-2 production, then growth retardation was less (Table 4). This partial alleviation of the syndrome could be due to either systemic endocrine effects of IGF-2 produced in one part, or to the near-normal size of one part providing additional trophic functions for the other. It is clear that neither derivative has absolute control over the growth of the other because the main wet weight deficiencies are localized to the sites of IGF-2 loss (Table 4).
Placental wet weight is increased by IGF-2 excess, as seen in Igf2r ko mice [35, 36], and decreased by IGF-2 deficiency, as observed in Igf2 ko mice ([9, 12] and this study). The mouse placenta is an anatomical unit in which maternal decidual tissue and conceptus components make contact and intermix; thus the total placenta fraction had a dual origin. As expected from its cellular composition, the total placenta had a significantly reduced wet weight when the trophoblast was Igf2 ko (data not shown). This reduction in wet weight was much lower than that seen when all parts of the conceptus lack active Igf2 genes [9, 12].
The fetal placenta was selected for study because it is the part of the placenta that is principally derived from the trophectoderm and that should, therefore, be most responsive to the genotype of the trophectoderm. The decidual base, invaded by extensive trophoblast giant cells, and the bulk of the spongiotrophoblast were removed from the total placenta to give the fetal placenta. The fetal placenta consisted of labyrinthine trophoblast decorated with small fragments of the spongiotrophoblast. This fetal placenta fraction was free of most maternal contamination, and similar dissections indicate that 80–90% is derived from the trophectoderm and 1–2% from maternal cells, with the rest made up from derivatives of the ICM, which include the sinuses of Duval [18].
Deficiency of IGF-2 in the trophoblast decreased the mean fetal placenta weight by 21%, and deficiency in the ICM derivatives decreased it by 14% (Table 4). There was no significant change in the DNA content in either case (Table 5), and thus wet weight and DNA content give different indicators of size. With IGF-2 in excess (Igf2r ko), the wet weight of the placenta is enlarged even if the type 1 IGF receptor is inactive (Igf1r ko; see [4]). It follows that this enlargement must depend on a novel receptor in the derivatives of the trophectoderm (XRp, see [9, 37]) or on direct interactions between IGF-2 produced in the conceptus and the maternal component of the placenta. For instance, local high IGF-2 levels may influence maternal capillary permeability and blood flow (see below).
Fluid Transport to the Conceptus and the Volume of the Exocoelom Plus Amniotic Cavities
During early postimplantation development, there is no defined site for fluid entry into the conceptus: fluid probably passes through the trophoblast layers, Reichert's membrane, the parietal endoderm, and the visceral endoderm before reaching the fetal precursor cells in the epiblast. Later, the degeneration of Reichert's membrane leaves the visceral yolk sac exposed as an extensive bounding barrier [38] and fluid exchange may occur here, as well as at the placenta. The fluid content of the normal exocelomic plus amniotic cavities is influenced by conceptus genotype, and the volume is normally greatest at E16.5 [6, 39]: the volume is under physiological regulation, and it is nearly halved by treatment of the mother with corticoids [40, 41].
When Igf2 gene activity is absent from the whole conceptus or the derivatives of either component of the blastocyst, then there are major reductions in the volume bounded by the visceral yolk sac at E16.5 (Tables 1 and 3). The reduced fluid volume of the exocelomic and amniotic cavities may provide another example of IGF-2 control of fluid content, for the volume of these cavities is enlarged when IGF-2 levels are unusually high (see Introduction). Our primary conclusion is that IGF-2 deficiency reduces fluid transport from the mother or fluid retention by the conceptus. The ICM derivatives have a major influence on these processes because IGF-2 deficiency in the ICM causes a decrease in fluid volume similar to that caused by IGF-2 deficiency in the whole conceptus (compare Tables 1 and 3).
It is not known if one or many exchange tissues are involved in regulating the extracellular fluid volume in the mouse conceptus. It is unlikely that the number of cells is a major determinant. There is only a 15% decrease in the DNA content of the placenta + fetus at a time when the cavities are reduced by 25% (Table 1). For a similar reason, it is also unlikely that the cell numbers of the fetal placentas are decisive in regulating these volumes. When there is a minor and nonsignificant reduction in the DNA content of this fraction (Table 5, ICM series), then the volume of the cavities is reduced by 25% (Table 3). We therefore favor an explanation of the effects of IGF-2 deficiency based on its physiological functions.
IGF-2 might control fluid uptake by a direct action on the maternal capillaries. Currently, only IGF-1 is known to increase fluid movement from the capillaries to the extra-vascular spaces, with increased capillary permeability and increased "filtration" following increased blood flow as the probable mechanisms [1, 2]. It is probable that IGF-2 has a similar effect because both the IGFs are potent inducers of vascular endothelial growth factor (VEGF) mRNA and protein, and VEGF increases capillary permeability as well as promoting angiogenesis [42, 43]. Further, IGF-2 stimulates angiogenesis in both the chorioallantoic membrane and rat cornea assays [43, 44]. The maternal and fetal circulatory systems in the placenta provide extensive surfaces for these local actions.
IGF-2 and the Control of Cell Population Size In Vivo
The main effect of Igf2 ko on the fetus and yolk sac results from low IGF-2 production by the derivatives of the ICM, with both parts suffering more than a 25% decrease in wet weight (Table 4). IGF-2 deficiency in the trophoblast decreased the wet weight of both the yolk sac and the fetus, but the effect was less, at 12%. The study of nonmanipulated conceptuses showed that both wet and dry weight were reduced by about 35% while the DNA content declined by only 15% (Table 1). These observations suggest that one major effect of IGF-2 deficiency is on tissue composition rather than cell number: there is a reduction in the amount of wet and dry weight associated with cells (as scored by DNA content). Similarly, the significant losses of wet weight in the recombination experiments were not accompanied by significant decreases in DNA content (Tables 4 and 5). This dissociation between biomass and DNA content has accompanied other changes in the IGF-2 system [8, 45].
It is currently probable that at least four interrelated processes are involved in the normal IGF-2 stimulation of mouse conceptus weight gain in the uterus. First, there may be changes in fluid dynamics, which could include acceleration of fluid and nutrient transport from the capillaries to the extravascular spaces and/or fluid retention in these spaces: these changes might be mediated by VEGF [43, 44]. Second, it is probable that increased cell survival occurs, as can be demonstrated both in vivo and in vitro in other systems [46, 47]. Third, there may be a slight increase in the rate of progression through the cell cycle, as found in cell culture and in vivo [3]; this action probably depends on collaboration with other growth factors [48]. Last, cell-associated biomass may increase [8, 45].
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: C.F. Graham, Zoology Department, South Parks Road, Oxford OX1 3PS, UK. FAX: 01865 271228; chris.graham{at}zoo.ox.ac.uk
Accepted: September 2, 1998.
Received: July 14, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. Palmieri, P. Loi, G. Ptak, and L.D. Salda REVIEW PAPER: A Review of the Pathology of Abnormal Placentae of Somatic Cell Nuclear Transfer Clone Pregnancies in Cattle, Sheep, and Mice Vet. Pathol., November 1, 2008; 45(6): 865 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Constancia, E. Angiolini, I. Sandovici, P. Smith, R. Smith, G. Kelsey, W. Dean, A. Ferguson-Smith, C. P. Sibley, W. Reik, et al. Adaptation of nutrient supply to fetal demand in the mouse involves interaction between the Igf2 gene and placental transporter systems PNAS, December 27, 2005; 102(52): 19219 - 19224. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. R. Ravelich, B. H. Breier, S. Reddy, J. A. Keelan, D. N. Wells, A. J. Peterson, and R. S.F. Lee Insulin-Like Growth Factor-I and Binding Proteins 1, 2, and 3 in Bovine Nuclear Transfer Pregnancies Biol Reprod, February 1, 2004; 70(2): 430 - 438. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Reik, M. Constancia, A. Fowden, N. Anderson, W. Dean, A. Ferguson-Smith, B. Tycko, and C. Sibley Regulation of supply and demand for maternal nutrients in mammals by imprinted genes J. Physiol., February 15, 2003; 547(1): 35 - 44. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Burns and A. B. Hassan Cell survival and proliferation are modified by insulin-like growth factor 2 between days 9 and 10 of mouse gestation Development, October 1, 2001; 128(19): 3819 - 3830. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Hill, R. C. Burghardt, K. Jones, C. R. Long, C. R. Looney, T. Shin, T. E. Spencer, J. A. Thompson, Q. A. Winger, and M. E. Westhusin Evidence for Placental Abnormality as the Major Cause of Mortality in First-Trimester Somatic Cell Cloned Bovine Fetuses Biol Reprod, December 1, 2000; 63(6): 1787 - 1794. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
