Systems based analysis of human embryos and gene networks involved in cell lineage allocation.

BACKGROUND: Little is understood of the molecular mechanisms involved in the earliest cell fate decision in human development, leading to the establishment of the trophectoderm (TE) and inner cell mass (ICM) stem cell population. Notably, there is a lack of understanding of how transcriptional networks arise during reorganisation of the embryonic genome post-fertilisation. RESULTS: We identified a hierarchical structure of preimplantation gene network modules around the time of embryonic genome activation (EGA). Using network models along with eukaryotic initiation factor (EIF) and epigenetic-associated gene expression we defined two sets of blastomeres that exhibited diverging tendencies towards ICM or TE. Analysis of the developmental networks demonstrated stage specific EIF expression and revealed that histone modifications may be an important epigenetic regulatory mechanism in preimplantation human embryos. Comparison to published RNAseq data confirmed that during EGA the individual 8-cell blastomeres are transcriptionally primed for the first lineage decision in development towards ICM or TE. CONCLUSIONS: Using multiple systems biology approaches to compare developmental stages in the early human embryo with single cell transcript data from blastomeres, we have shown that blastomeres considered to be totipotent are not transcriptionally equivalent. Furthermore we have linked the developmental interactome to individual blastomeres and to later cell lineage. This has clinical implications for understanding the impact of fertility treatments and developmental programming of long term health.

Investigating the role of CD44 and hyaluronate in embryo-epithelial interaction using an in vitro model.

Implantation failure is an important impediment to increasing success rates in assisted reproductive technologies. Knowledge of the cascade of morphological and molecular events at implantation remains limited. Cell surface CD44 and hyaluronate (HA) have been reported in the uterus, but a role in intercellular interaction at implantation remains to be evaluated. Mouse embryos were co-cultured with human Ishikawa endometrial epithelial monolayers over 2 days. Attachment was tenuous during the first 24 h, after which it became stable, leading to breaching of the monolayer. The effects of enzymatically reducing the density of HA, or introducing a function-blocking antibody to CD44, were monitored during progression from weak to stable embryonic attachment. Hyaluronidase-mediated removal of surface HA from the epithelial cells enhanced the speed of attachment, while a similar treatment of embryos had no effect. The antibody to CD44 caused retardation of initial attachment. These results suggest that CD44-HA binding could be employed by embryos during initial docking, but the persistence of HA in epithelial cells might be detrimental to later stages of implantation by retarding attainment of stable attachment.

The generation and validation of a dual cardiac HAND1-Tomato NKX2-5-GFP human embryonic stem cell line UMANe002-A-3.

The transcription factor HAND1 is a critical regulator of cardiac development which is expressed in sub-populations of cardiac progenitors and cardiomyocytes. The transcription factor NKX2-5, in contrast, is expressed more widely in cardiac cells. Here we report the generation of a dual reporter hESC line where the expression of these genes can be simultaneously measured, enabling lineage analysis as well as studies of HAND1 and NKX2-5 gene regulation and protein function. This tool will have wide utility particularly for research on developmental biology and disease modelling.

The generation and validation of two NKX2-5 fluorescent reporter human embryonic stem cell lines: UMANe002-A-1 and UMANe002-A-2.

The transcription factor NKX2-5 is a highly conserved master regulator of heart development which is widely expressed in cardiac progenitors and cardiomyocytes. Fluorescent reporters of NKX2-5 that minimally perturb normal protein expression can enable the identification, quantification and isolation of NKX2-5-expressing cells in a normal physiological state. Here we report the generation of two new hESC lines with eGFP inserted upstream (5') or downstream (3') of NKX2-5, linked by a cleavable T2A peptide. These complementary reporters produce a robust fluorescent signal in cardiac cells and have wide utility particularly for research on developmental biology and disease modelling.

Early stages of implantation as revealed by an in vitro model.

Our limited understanding of the processes underlying steroid hormonal control of human endometrial receptivity is largely due to the lack of a relevant model system. To overcome scarcity of material, we have developed a model in which mouse embryos attach to human Ishikawa cells, which express functional steroid hormone receptors. Blastocysts flushed from day 4 pregnant superovulated mice were transferred to confluent Ishikawa cell monolayers. After 48 h of co-culture, 85% of the blastocysts had attached loosely, but only 40% attached stably to the epithelial cell surface. In contrast, 95% of the embryos attached stably to tissue culture plastic. Thus, weak attachment of a majority of the embryos was followed by stronger adhesion of a smaller proportion. Seventeen percent of the transferred blastocysts modified the epithelial cell surface with loss of MUC1 at the attachment site, extending variably to adjacent epithelial cells. Initially, stable attachment occurred without disruption to the integrity of the epithelial monolayer, but at later stages after the embryo had spread laterally, displacement of subjacent cells was observed. A modest increase in stable attachment, but no changes to MUC1 clearance, was observed after assisted hatching. After 24 h priming of Ishikawa cells by 17beta-oestradiol (OE(2)) followed by 72-h incubation with medroxyprogesterone acetate and OE(2), stable attachment increased from 40 to 70%. Initial attachment is efficient either in the presence or in the absence of hormone; steroid treatment increased the incidence of stable attachment. Implantation failure is predicted to occur in this model when embryos fail to progress from initial to stable attachment.

Determining the LIF-sensitive period for implantation using a LIF-receptor antagonist.

Uteri of Lif null mice do not support embryo implantation. Since deletion of some genes often prevents the survival of null mice to adulthood, we have used a proven inhibitor of leukaemia inhibitory factor (LIF) signalling to identify the precise window of time during which LIF is required in vivo, and assessed the cellular expression of several LIF-associated targets. On day 4 of pregnancy, mice were injected with hLIF-05 (inhibitor) into the uterine lumen, with corresponding volumes of PBS (vehicle) injected into the contralateral horn. On days 5 and 6, the number of implantation sites was recorded and the uteri processed for immunohistochemistry. Blockade of LIF on day 4 reduced embryo implantation by 50% (P

Expression of genes involved in early cell fate decisions in human embryos and their regulation by growth factors.

Little is understood about the regulation of gene expression in human preimplantation embryos. We set out to examine the expression in human preimplantation embryos of a number of genes known to be critical for early development of the murine embryo. The expression profile of these genes was analysed throughout preimplantation development and in response to growth factor (GF) stimulation. Developmental expression of a number of genes was similar to that seen in murine embryos (OCT3B/4, CDX2, NANOG). However, GATA6 is expressed throughout preimplantation development in the human. Embryos were cultured in IGF-I, leukaemia inhibitory factor (LIF) or heparin-binding EGF-like growth factor (HBEGF), all of which are known to stimulate the development of human embryos. Our data show that culture in HBEGF and LIF appears to facilitate human embryo expression of a number of genes: ERBB4 (LIF) and LIFR and DSC2 (HBEGF) while in the presence of HBEGF no blastocysts expressed EOMES and when cultured with LIF only two out of nine blastocysts expressed TBN. These data improve our knowledge of the similarities between human and murine embryos and the influence of GFs on human embryo gene expression. Results from this study will improve the understanding of cell fate decisions in early human embryos, which has important implications for both IVF treatment and the derivation of human embryonic stem cells.

A 31P-NMR study of mono- and dimagnesium complexes of adenosine 5'-triphosphate and model systems.

31P-NMR chemical shifts and spin-lattice relaxation times of ATP (adenosine 5'-triphosphate), ribose 5'-triphosphate and tripolyphosphate show closely similar behaviour in aqueous solution at pH 7.5 on titration with Mg2+. The results are interpreted in terms of formation of 1 : 1 and 2 : 1 (dimagnesium) complexes with Mg2+ bound exclusively to the triphosphate chain. Stability constants for these complexes are reported. It is suggested that the predominant form of the 1 : 1 complexes has Mg2+ bound in tridentate manner (via non-bridging oxygen) to the alpha, beta and gamma phosphorus atoms; whilst that of the 2 : 1 complexes has each Mg2+ bound in bidentate manner, one to the alpha and beta, and the other to the beta and gamma, phosphorus positions.

Evidence on the role(s) of ATP in the mechanism of nitrogenase, from proton NMR relaxation studies on metal and nucleotide binding to the molybdenum-iron protein.

Interactions between the molybdenum-iron protein (Kp1) of nitrogenase (reduced ferredoxin:dinitrogen oxidoreductase (ATP-hydrolysing,) EC 1.18.2.1) from Klebsiella pneumoniae and the divalent ions, Mn2+, Mg2+, Ca2+ and Ba2+, have been studied by monitoring the water proton NMR relaxation enhancement caused by the paramagnetism of Mn2+. We observed several binding sites for Mn2+, equivalent within experimental error (Kd = 209 +/- 23 microM), increasing in number from 1.0 to 2.9 per molecule in direct proportion to the specific activity of the protein. Metal binding sites on the MoFe protein are therefore essential to the enzymic function of nitrogenase. A maximum of four such sites is inferred for the fully active protein molecule. All manganese sites can alternatively bind the diamagnetic ions studied, the binding being one order-of-magnitude weaker (Kd = 2.2 +/- 0.3 mM for Mg2+; 1.6 +/- 0.2 mM for Ca2+; 3.4 +/- 0.3 mM for Ba2+), ATP and ADP form ternary complexes via Mn2+ with Kpl. The above data and other evidence on MgATP binding are discussed in terms of the site of hydrolysis of ATP during turnover and its possible bridging role between the two protein components of the enzyme.

Morphological evidence for a morphogenetic field in gastropod mollusc eggs.

Eggs of the marine gastropod Crepidula fornicata examined by confocal imaging of FITC-lectin binding to the surface, and cryoscopic-SEM both reveal a surface architecture of linear structures organized around the animal-vegetal axis, which is spatially related to the anterior-posterior (a-p) axis of the subsequent embryo. A series of structures is also orientated with reference to specific micromere quartets formed during spiral cleavage. Thus, the surface architecture may provide a visible marker for a morphogenetic field which generates the a-p axis and organizes the cleavage pattern. Moreover, this architecture is co-extensive with that found on the vegetal, polar lobe-bearing region of eggs, as described by others, and which varies between gastropod taxa with varied types of body form. Confocal imaging reveals a distinct localization of F-actin to the architecture of the lobe region. However, the integrity of this F-actin is not responsible for the maintenance of the surface architecture. The significance of these findings to our understanding of the generation of diversity within the Gastropoda and general ontogenic mechanisms is discussed.

Expression of N-CAM in fertilized pre- and periimplantation and parthenogenetically activated mouse embryos.

The expression of the cell adhesion molecule of the immunoglobulin family, neural cell adhesion molecule (N-CAM), in the pre- and periimplantation embryo was examined by immunocytochemistry. N-CAM is expressed on unfertilized ovulated oocytes, fertilized preimplantation embryos at all stages of development and parthenogenetically activated eggs and embryos. In fertilized embryos, expression from the 4-cell stage can be partially inhibited by blocking embryonic transcription before 38 h post human chorionic gonadotropin (hCG). Expression of N-CAM was reduced on the trophoblast of day 6 blastocysts in culture, weak on the trophoblast of embryonic outgrowths and disappears from invading trophoblast in utero. An antibody against alpha(2-8) linked polysialic acid, mAb2-2B, reacted with embryos from the 8-cell stage, and staining was similarly reduced on the trophoblast of blastocysts at the time of implantation. These results suggest a role for N-CAM in the interactions of cells of the preimplantation mammalian embryo which requires further investigation.

Role of leukemia inhibitor factor (LIF) in decidualisation of murine uterine stromal cells in vitro.

Decidualisation of uterine stromal cells is a prerequisite for implantation of the embryo in mice. Here we have used an in vitro culture system in which stromal cells decidualise as indicated by a number of markers, including an increase in alkaline phosphatase (ALP) activity. The latter was used as a quantitative marker of decidualisation in the presence of low (2%) fetal calf serum. Prostaglandin E(2) (PGE(2)), which is known to induce decidualisation, increased ALP activity, and this effect was blocked in a dose-dependent manner by indomethacin. Leukemia inhibitory factor (LIF) was then examined, but it had no effect on PGE(2) secretion. However, LIF suppressed ALP activity in a dose-dependent manner in the presence of 2% serum, while an inhibitor of LIF that competes for binding to its receptor reversed the effect of LIF and increased ALP activity above the control level. In serum-free cultures, stromal cells differentiated rapidly, and no differences were observed between LIF-treated and untreated cultures. Stromal cells produce LIF during in vitro culture, and this peaked at 48 h. Freshly collected stromal cells from both day-2 and -4 pregnant mice expressed mRNA for the LIF receptor, and the transcript level was higher in cells isolated on day 4. However, no differences were observed in the relative levels of transcripts in cells from day 2 and day 4 after culture, nor were there differences between the LIF-treated cultures and controls. Therefore, in this study, we have shown that LIF suppresses decidualisation of murine uterine stromal cells in the presence of serum, this is not due to the regulation of PGE(2) secretion by stromal cells.

Distribution of some Gal beta 1-3(4)GlcNAc related carbohydrate antigens on the mouse uterine epithelium in relation to the peri-implantational period.

Using monoclonal antibodies of defined carbohydrate specificity we have looked at the distribution of various Gal beta 1-3(4)GlcNAc related oligosaccharide determinants in the mouse uterus during the first 6 days of pregnancy. Frozen sections of uterus from B6D2F1, B6CBF1 or B6D2F1/BOM female mice were incubated with the monoclonal antibodies and then with a fluorescein isothiocyanate conjugate of goat anti-mouse IgM and viewed by epifluorescence illumination. None of the antibodies bound specifically to stroma cells but antibodies recognising difucosylated Gal beta 1-3(4)GlcNAc structures, the monofucosylated type II determinant (SSEA-1) and an H type I oligosaccharide bound to cells of the uterine luminal epithelium and glands and to the uterine secretions. Antibodies recognising the three different types of saccharide showed independent changes in staining intensity during early pregnancy. The antibody which recognises H type I structures (667/9E9) showed a change in distribution from binding to most cells of the uterine epithelium in the non-pregnant mouse and on day 3 of pregnancy to binding restricted to areas of epithelial cells interspersed with non-staining clumps of cells between days 4 and 5 of pregnancy.

Molecular interactions at the maternal-embryonic interface during the early phase of implantation.

In mammals the embryo must implant in the uterus and develop a placenta to gain nutrition and facilitate gas exchange. In this article, the earliest events in this process are reviewed. The embryo can implant only when it has reached the blastocyst stage. The blastocyst is composed of an inner clump of cells, the inner cell mass, that gives rise to the fetus and an outer layer of trophectoderm (TE), the precursor of the placenta. Both blastocyst and uterus must differentiate in parallel to reach the appropriate state of maturity (activated blastocyst and receptive uterus) at which implantation can occur. Interaction between TE and the luminal epithelium (LE) lining the uterus initiates implantation, and both soluble signals and association between molecules on apposed surfaces appear to be involved. A number of cell surface molecules have been implicated in the initial attachment between TE and LE. These include HSPG, Le-y and the H-type-1 antigen, HB-EGF, trophinin-tastin-bystin complex, integrins, and extracellular matrix molecules such as osteopontin and laminin. Others, such as mucins, may need to be removed or modified to allow adhesion to proceed. Evidence for the role of these components is discussed.

Immunofluorescent lectin binding patterns and glycoprotein co-localization in the developing murine molar tooth.

Fluorescein-conjugated lectins were used in conjunction with antibodies to laminin, tenascin and amelogenin to investigate saccharide expression in the developing tooth germ. At the bud stage, peanut agglutinin (PNA) binding demonstrated residues that may be D-galactose-(beta 1----3)DGalNAc, and this staining occurred after the expression of tenascin. Only the cap-stage enamel organ suprabasal cells and the enamel knot stained intensely with Ulex europeus agglutinin-I, but not Lotus tetragonolobus agglutinin, implying the transient presence of blood group H type I oligosaccharides. At the late stages of amelogenesis, enamel synthesis is preceded by en bloc loss of inner enamel basement membrane components. Before this, Bandeiraea (Griffonia) simplicifolia--I (BSL-I) staining was lost from postmitotic ameloblasts, suggesting that a glycosylated species is initially removed. Additionally, PNA was co-localized with amelogenin protein, suggesting that it may express beta-D-galactosyl sequences. These results indicate that the glycosylation patterns of matrix components during odontogenesis may be important as they vary in a manner similar to that of the well-known glycoproteins.

Interleukin 1 signaling is regulated by leukemia inhibitory factor (LIF) and is aberrant in Lif-/- mouse uterus.

This study addresses the regulation of the interleukin 1 (IL1) system in the murine uterine luminal epithelium (LE) and stroma by leukemia inhibitory factor (LIF). Using RT-PCR we compared expression of Il1a, Il1b, Il1rn, Il1r1, and Il1r2 during the pre- and peri-implantation periods of pregnancy in wild-type (WT) and LIF-null LE and stroma. In WT LE, Il1a transcripts were down-regulated on Day 4 of pregnancy (D4), with renewed expression by the evening of D4 (D4 pm). In Lif(-/-) LE there was a gradual decrease in expression on D2, and expression became undetectable by D6. Il1b and Il1r1 expression were similar in WT and null mice, but Il1rn expression was almost completely lost during the peri-implantation period in Lif(-/-) LE. In the stroma, Il1a was sharply down-regulated on D4 and reappeared on D4 pm but was only expressed from D3 to D5 in the null mice. Stromal Il1r1 and Il1r2 were also misregulated. Il1rn showed constitutive expression in null stroma in contrast to the loss of expression on D4 in the WT mouse. In Lif-deficient mice, immunostaining indicated a reduction of endometrial IL1A at the time of implantation and of IL1B in stroma. LE-stromal coculture revealed that LIF stimulated the apical secretion of both IL1A and PTGES2 by LE cells without affecting basal secretion of IL1A and with only a small effect on basal PTGES2 secretion. We conclude that Il1a and Il1rn in LE and Il1a, Il1rn, and Il1r1 in stroma are regulated by LIF, which stimulates apical secretion of IL1A by LE.

Characterization of the uterine phenotype during the peri-implantation period for LIF-null, MF1 strain mice.

Leukemia inhibitory factor plays a major role in the uterus and in its absence embryos fail to implant. Our knowledge of the targets for LIF and the consequences of its absence is still very incomplete. In this study, we have examined the ultrastructure of the potential implantation site in LIF-null MF1 female mice compared to that of wild type animals. We also compared expression of proteins associated with implantation in luminal epithelium and stroma. Luminal epithelial cells (LE) of null animals failed to develop apical pinopods, had increased glycocalyx, and retained a columnar shape during the peri-implantation period. Stromal cells of LIF-null animals showed no evidence of decidual giant cell formation even by day 6 of pregnancy. A number of proteins normally expressed in decidualizing stroma did not increase in abundance in the LIF-null animals including desmin, tenascin, Cox-2, bone morphogenetic protein (BMP)-2 and -7, and Hoxa-10. In wild type animals, the IL-6 family member Oncostatin M (OSM) was found to be transiently expressed in the luminal epithelium on late day 4 and then in the stroma at the attachment site on days 5-6 of pregnancy, with a similar but not identical pattern to that of Cox-2. In the LIF-null animals, no OSM protein was detected in either LE or stroma adjacent to the embryo, indicating that expression requires uterine LIF in addition to a blastocyst signal. Fucosylated epitopes: the H-type-1 antigen and those recognized by lectins from Ulex europaeus-1 and Tetragonolobus purpureus were enhanced on apical LE on day 4 of pregnancy. H-type-1 antigen remained higher on day 5, and was not reduced even by day 6 in contrast to wild type uterus. These data point to a profound disturbance of normal luminal epithelial and stromal differentiation during early pregnancy in LIF-nulls. On this background, we also obtained less than a Mendelian ratio of null offspring suggesting developmental failure.

The secretion of the eggshell of Schistocerca gregaria: ultrastructure of the follicle cells during the termination of vitellogenesis and eggshell secretion.

The secretion of the eggshell by the follicle cells in the desert locust, Schistocerca gregaria, was studied using the electron microscope. The 3 layers of the eggshell, the vitelline membrane, the endochorion, and the exochorion, are produced in sequence over a short period of about 30-36 h. The follicle cells contain little rough endoplasmic reticulum (RER) and small inconspicuous Golgi bodies during vitellogenesis. As eggshell secretion approaches there is an increase in the amount of RER and Golgi cisternae contain electron-dense product. At each stage of the 3-phase secretion cycle the follicle cells contain Golgi bodies and secretion vesicles with distinct morphology. The follicle cells increase in breadth and decrease in height between the beginning and end of eggshell secretion. The endochorion ridges arise at the junction between follicle cells and appear to be moulded by the microvilli formed at this position. In the ovary prior to ovulation, the eggshell consists of a thin (0.5 micrometer) electron-dense vitelline and an outer fibrillar exochorion layer, 20-30 micrometer thick. Further changes take place in the vitelline membrane and the endochorion after oviposition, and a layer of curly fibres, the extrachorion, is secreted in the oviduct.