The arterial and cardiac epicardium in development, disease and repair

The importance of the epicardium covering the heart and the intrapericardial part of the great arteries has reached a new summit. It has evolved as a major cellular component with impact both in development, disease and more recently also repair potential. The role of the epicardium in development, its differentiation from a proepicardial organ at the venous pole (vPEO) and the differentiation capacities of the vPEO initiating cardiac epicardium (cEP) into epicardium derived cells (EPDCs) have been extensively described in recent reviews on growth and transcription factor pathways. In short, the epicardium is the source of the interstitial, the annulus fibrosus and the adventitial fibroblasts, and differentiates into the coronary arterial smooth muscle cells. Furthermore, EPDCs induce growth of the compact myocardium and differentiation of the Purkinje fibers. This review includes an arterial pole located PEO (aPEO) that provides the epicardium covering the intrapericardial great vessels. In avian and mouse models disturbance of epicardial outgrowth and maturation leads to a broad spectrum of cardiac anomalies with main focus on non-compaction of the myocardium, deficient annulus fibrosis, valve malformations and coronary artery abnormalities. The discovery that in disease both arterial and cardiac epicardium can again differentiate into EPDCs and thus reactivate its embryonic program and potential has highly broadened the scope of research interest. This reactivation is seen after myocardial infarction and also in aneurysm formation of the ascending aorta. Use of EPDCs for cell therapy show their positive function in paracrine mediated repair processes which can be additive when combined with the cardiac progenitor stem cells that probably share the same embryonic origin with EPDCs. Research into the many cell-autonomous and cell-cell-based capacities of the adult epicardium will open up new realistic therapeutic avenues.

MicroRNA-23 Restricts Cardiac Valve Formation by Inhibiting Has2 and Extracellular Hyaluronic Acid Production

Rationale:

Since their discovery almost 20 years ago, microRNAs have been shown to perform essential roles during tissue development and disease. Although roles for microRNAs in the myocardium during embryo development and cardiac disease have been demonstrated, very little is know about their role in the endocardium or during cardiac valve formation.

Objective:

To study the role of microRNAs in cardiac valve formation.

Methods and Results:

We show that zebrafish dicer mutant embryos, lacking mature miRNAs, form excessive endocardial cushions. By screening miRNAs expressed in the heart, we found that miR-23 is both necessary and sufficient for restricting the number of endocardial cells that differentiate into endocardial cushion cells. In addition, in mouse endothelial cells, miR-23 inhibited a transforming growth factor-β–induced endothelial-to-mesenchymal transition. By in silico screening of expression data with predicted miR-23 target sites combined with in vivo testing, we identified hyaluronic acid synthase 2 ( Has2) , Icat , and Tmem2 as novel direct targets of miR-23. Finally, we demonstrate that the upregulation of Has2 , an extracellular remodeling enzyme required for endocardial cushion and valve formation, is responsible for the excessive endocardial cushion cell differentiation in dicer mutants.

Conclusions:

MiR-23 in the embryonic heart is required to restrict endocardial cushion formation by inhibiting Has2 expression and extracellular hyaluronic acid production.

Human mesenchymal stem cell-conditioned medium improves cardiac function following myocardial infarction

Recent studies suggest that the therapeutic effects of stem cell transplantation following myocardial infarction (MI) are mediated by paracrine factors. One of the main goals in the treatment of ischemic heart disease is to stimulate vascular repair mechanisms. Here, we sought to explore the therapeutic angiogenic potential of mesenchymal stem cell (MSC) secretions. Human MSC secretions were collected as conditioned medium (MSC-CM) using a clinically compliant protocol. Based on proteomic and pathway analysis of MSC-CM, an in vitro assay of HUVEC spheroids was performed identifying the angiogenic properties of MSC-CM. Subsequently, pigs were subjected to surgical left circumflex coronary artery ligation and randomized to intravenous MSC-CM treatment or non-CM (NCM) treatment for 7 days. Three weeks after MI, myocardial capillary density was higher in pigs treated with MSC-CM (645 ± 114 vs 981 ± 55 capillaries/mm(2); P = 0.021), which was accompanied by reduced myocardial infarct size and preserved systolic and diastolic performance. Intravenous MSC-CM treatment after myocardial infarction increases capillary density and preserves cardiac function, probably by increasing myocardial perfusion.

Modulation of TGF-β/BMP-6 expression and increased levels of circulating smooth muscle progenitor cells in a type I diabetes mouse model

Background

Diabetic patients experience exaggerated intimal hyperplasia after endovascular procedures. Recently it has been shown that circulating smooth muscle progenitor cells (SPC) contribute to intimal hyperplasia. We hypothesized that SPC differentiation would be increased in diabetes and focused on modulation of TGF-β/BMP-6 signaling as potential underlying mechanism.

Methods

We isolated SPC from C57Bl/6 mice with streptozotocin-induced diabetes and controls. SPC differentiation was evaluated by immunofluorescent staining for αSMA and collagen Type I. SPC mRNA expression of TGF-β and BMP-6 was quantified using real-time PCR. Intima formation was assessed in cuffed femoral arteries. Homing of bone marrow derived cells to cuffed arterial segments was evaluated in animals transplanted with bone marrow from GFP-transgenic mice.

Results

We observed that SPC differentiation was accelerated and numeric outgrowth increased in diabetic animals (24.6 ± 8.8 vs 8.3 ± 1.9 per HPF after 10 days, p < 0.05). Quantitative real-time PCR showed increased expression of TGF-β and decreased expression of the BMP-6 in diabetic SPC. SPC were MAC-3 positive, indicative of monocytic lineage. Intima formation in cuffed arterial segments was increased in diabetic mice (intima/media ratio 0.68 ± 0.15 vs 0.29 ± 0.06, p < 0.05). In GFP-chimeric mice, bone marrow derived cells were observed in the neointima (4.4 ± 3.3 cells per section) and particularly in the adventitia (43.6 ± 9.3 cells per section). GFP-positive cells were in part MAC-3 positive, but rarely expressed α-SMA.

Conclusions

In conclusion, in a diabetic mouse model, SPC levels are increased and SPC TGF-β/BMP-6 expression is modulated. Altered TGF-β/BMP-6 expression is known to regulate smooth muscle cell differentiation and may facilitate SPC differentiation. This may contribute to exaggerated intimal hyperplasia in diabetes as bone marrow derived cells home to sites of neointima formation.

Stem cells: the building blocks to repair the injured heart

Myocardial infarction is the major cause of death in western countries due to impaired function of the heart, which is the result of cardiomyocyte death and fibrotic scar formation. The endogenous regenerative capacity of the heart is unable to replenish this significant loss of tissue and conventional medical management cannot correct the underlying defects in cardiac muscle cell number. Recently, tremendous effort is being put into the development of cell transplantation protocol for heart repair, which has been put forward as an alternative therapy to reduce cell damage, cardiomyocyte death and improve tissue contraction. Unfortunately the ideal stem cell population for heart repair has not been identified to date, but several characteristics are defined which the ideal population should have namely, reduce cell damage, reduce cardiomyocyte death, induce differentiation into cardiomyocytes and endothelial cells, and improve tissue contraction. It is unclear whether this will be possible in one optimal population. Therefore the research focus is shifting towards improving the characteristics of the stem cell populations that are identified to date. In this review, we will give an overview of the different stem/progenitor cell populations and their application in cardiac repair and discuss current knowledge on issues like differentiation capacity, paracrine secretion profile, genetic modification of progenitor cells and their influence on cardiac remodeling.

A new direction for cardiac regeneration therapy: application of synergistically acting epicardium-derived cells and cardiomyocyte progenitor cells

BACKGROUND

Adult human epicardium-derived cells (EPDCs), transplanted into the infarcted heart, are known to improve cardiac function, mainly through paracrine protection of the surrounding tissue. We hypothesized that this effect might be further improved if these supportive EPDCs were combined with cells that could possibly supply the ischemic heart with new cardiomyocytes. Therefore, we transplanted EPDCs together with cardiomyocyte progenitor cells that can generate mature cardiomyocytes in vitro.

METHODS AND RESULTS

EPDCs and cardiomyocyte progenitor cells were isolated from human adult atrial appendages, expanded in culture, and transplanted separately or together into the infarcted mouse myocardium (total cell number, 4x10(5)). Cardiac function was determined 6 weeks later (9.4T MRI). Coculturing increased proliferation rate and production of several growth factors, indicating a mutual effect. Cotransplantation resulted in further improvement of cardiac function compared with single cell-type recipients (P<0.05), which themselves demonstrated better function than vehicle-injected controls (P<0.05). However, in contrast to our hypothesis, no graft-derived cardiomyocytes were observed within the 6-week survival, supporting that not only EPDCs but also cardiomyocyte progenitor cells acted in a paracrine manner. Because injected cell number and degree of engraftment were similar between groups, the additional functional improvement in the cotransplantation group cannot be explained by an increased amount of secreted factors but rather by an altered type of secretion.

CONCLUSIONS

EPDCs and cardiomyocyte progenitor cells synergistically improve cardiac function after myocardial infarction, probably instigated by complementary paracrine actions. Our results demonstrate for the first time that synergistically acting cells hold great promise for future clinical regeneration therapy.

Stem cell sources for cardiac regeneration

Cell-based cardiac repair has the ambitious aim to replace the malfunctioning cardiac muscle developed after myocardial infarction, with new contractile cardiomyocytes and vessels. Different stem cell populations have been intensively studied in the last decade as a potential source of new cardiomyocytes to ameliorate the injured myocardium, compensate for the loss of ventricular mass and contractility and eventually restore cardiac function. An array of cell types has been explored in this respect, including skeletal muscle, bone marrow derived stem cells, embryonic stem cells (ESC) and more recently cardiac progenitor cells. The best-studied cell types are mouse and human ESC cells, which have undisputedly been demonstrated to differentiate into cardiomyocyte and vascular lineages and have been of great help to understand the differentiation process of pluripotent cells. However, due to their immunogenicity, risk of tumor development and the ethical challenge arising from their embryonic origin, they do not provide a suitable cell source for a regenerative therapy approach. A better option, overcoming ethical and allogenicity problems, seems to be provided by bone marrow derived cells and by the recently identified cardiac precursors. This report will overview current knowledge on these different cell types and their application in cardiac regeneration and address issues like implementation of delivery methods, including tissue engineering approaches that need to be developed alongside.

Controlling cardiomyocyte survival

Gradually the distinction between signalling pathways originally believed to be specific for either hypertrophy, cell cycle control, apoptosis and cell survival are fading. The subtle variations in stimuli to a cell and the microenvironment will determine cell fate. In cardiomyocytes the entrance into the cell cycle is efficiently blocked. Therefore attention has focused on pathways involved in hypertrophy to assess effects in ischaemic models and vice versa. Interventions at different levels have been shown to be cardiomyocyte protective. Various growth factors (including IGF1 and FGF1,2) have shown to prevent or delay cardiomyocyte loss in and ex vivo. Similar results have been reported for downstream interventions in the signalling pathways. Strong effects after MAPK activation have been shown in gene targeted mice. Especially constitutive activation of the ERK proteins prevents ischemic damage of the heart with conservation of left ventricular function. Evidence for a key role of nuclear Akt in preventing apoptosis is accumulating from various genetic and pharmacological sources. Development of techniques to measure the level of cardiomyocyte death depends on further improvements in molecular imaging in mouse and human. In addition to studying cardiomyocyte cell death, it is crucial to measure myocardial function. Whether hypertrophy following ischaemia is adaptive or maladaptive and whether all apoptosis is detrimental will have to be determined by assessment of left ventricular function through invasive and noninvasive methods.

New and viable cells to replace lost and malfunctioning myocardial tissue

The use of stem cells for cardiac repair is a promising opportunity for developing new treatment strategies as the applications are theoretically unlimited and lead to actual cardiac tissue regeneration. Human embryonic stem cells were only recently cloned and their capacity to differentiate into true cardiomyocytes makes them in principle an unlimited source of transplantable cells for cardiac repair, although practical and ethical constraints exist. Also, the study of embryonic stem cells and their differentiation into cardiomyocytes will bring forth new insights into the molecular processes involved in cardiomyocyte-development and -proliferation, which could lead to the development of other strategies to augment in vivo cardiomyocyte numbers. On the other hand, somatic stem cells are alternative cell sources that can be used for cell transplantation purposes. They do not evoke ethical issues and bear less ethical constraints. However, they also appear to be much more restricted in their differentiation potential than the embryonic stem cells. Here we discuss the use of both cell types, embryonic and somatic stem cells, in relation with their importance for the clarification of cardiomyocyte-development and their possible usefulness for clinical therapy.

BMP signaling components are expressed in human fracture callus

Of the various growth factors involved in the healing response after a fracture, bone morphogenetic proteins (BMPs) are emerging as key modulators. BMPs exert their effects by binding to a complex of type I and type II receptors leading to the phosphorylation of specific downstream effector proteins called Smads. The current study examined the presence of BMP signaling components in human callus obtained from five nascent malunions undergoing fracture fixation. These callus samples represented various stages of bone healing and a mixture of endochondral and intramembraneous bone healing. We performed immunohistochemistry on the callus, using antibodies for BMP (BMP-2,-3,-4,-7), their receptors (BMPR-IA, -IB, -II), and phosphorylated BMP receptor-regulated Smads (pBMP-R-Smads). Active osteoblasts showed fairly consistent positive staining for all BMPs that were examined, with the immunoreactivity most intense for BMP-7 and BMP-3. Immunostaining for BMPs in osteoblasts appeared to colocalize with the expression of BMPR-IA, -IB, and -II. Positive immunostaining for pBMP-R-Smads suggests that the BMP receptors expressed in these cells are activated. Staining for BMPs in cartilage cells was variable. The immunostaining appeared stronger in more mature cells, whereas staining for BMP receptors in cartilage cells was less ubiquitous. However, the expression of pBMP-R-Smads in cartilage cells suggests active signal transduction. Fibroblast-like cells also had a variable staining pattern. Overall, our findings indicate the presence of BMPs, their various receptors, and activated forms of receptor-regulated Smads in human fracture callus. To the best of our knowledge, this is the first study that documents the expression of these proteins in human fracture tissue. Complete elucidation of the roles of BMP in bone formation will hopefully lead to improved fracture healing care.

Signaling of transforming growth factor-beta family members through Smad proteins

Smads are pivotal intracellular nuclear effectors of transforming growth factor-beta (TGF-beta) family members. Ligand-induced activation of TGF-beta family receptors with intrinsic serine/threonine kinase activity trigger phosphorylation of receptor-regulated Smads (R-Smads), whereas Smad2 and Smad3 are phosphorylated by TGF-beta, and activin type I receptors, Smad1, Smad5 and Smad8, act downstream of BMP type I receptors. Activated R-Smads form heteromeric complexes with common-partner Smads (Co-Smads), e.g. Smad4, which translocate efficiently to the nucleus, where they regulate, in co-operation with other transcription factors, coactivators and corepressors, the transcription of target genes. Inhibitory Smads act in most cases in an opposite manner from R- and Co-Smads. Like other components in the TGF-beta family signaling cascade, Smad activity is intricately regulated. The multifunctional and context dependency of TGF-beta family responses are reflected in the function of Smads as signal integrators. Certain Smads are somatically mutated at high frequency in particular types of human cancers. Gene ablation of Smads in the mouse has revealed their critical roles during embryonic development. Here we review the latest advances in our understanding of the Smad mechanism of action and their in vivo functions.

Expression of TGF-beta stimulated clone-22 (TSC-22) in mouse development and TGF-beta signalling

TSC-22 is a highly conserved member of a novel family of transcription factors, that is a direct target of transforming growth factor-beta (TGF-beta) in osteoblastic cells. We have investigated the expression of TSC-22 in detail during mouse development using in situ hybridization. We detected strong expression of TSC-22 in the embryo proper first at embryonic day 8.5 (E8.5), in the primitive heart, intermediate mesoderm and the neural tube. The dynamics of the TSC-22 distribution in the neural tube was particularly striking, with ubiquitous expression rostrally and restriction to neural tissue nearer the floor plate more caudally; between E8.5 and E9.5 the zone of restricted expression extended rostrally. At later stages of development, TSC-22 was detected in the mesenchymal compartment of many tissues and organs, including the lung, trachea, kidney, stomach, intestine, tooth buds, and in precartilage condensations. Furthermore, TSC-22 was highly expressed in the floor plate itself and notochord, and the endothelium lining the blood vessels, in particular the major arteries. Many of these sites have been proposed previously as possible TGF-beta target tissues; the results imply that TSC-22 may also be a direct TGF-beta target gene during mouse embryogenesis. Experiments on TSC-22 expression in embryoid bodies of embryonic stem (ES) cells expressing dominant negative TGF-beta binding receptors initially supported this hypothesis. However, examination of somatic chimeras derived from these same mutant ES cells at nominal E9.5 showed that TSC-22 expression in the heart and neural tube was still detectable despite obvious phenotypic abnormalities. We therefore propose that although TSC-22 may be a direct target of TGF-beta in late development, other factors are likely to be major regulators of expression at earlier stages.

Expression of the inhibitory Smad7 in early mouse development and upregulation during embryonic vasculogenesis

SMAD proteins are downstream targets of serine/threonine kinase receptors of the transforming growth factor beta (TGF beta) superfamily. Ligands activating these receptors regulate cell growth, differentiation and development in many tissues of various organisms. In mammals eight different Smad genes are known, each with different roles in mediating signalling between plasma membrane and nucleus. Smad6 and Smad7 are inhibitors of TGF beta family signalling. They are both expressed in human adult vascular endothelial cells, particularly after these cells have been subjected to shear stress (Topper et al. [1997] Proc Natl Acad Sci USA 94:9314-9319). Here we show by reverse transcriptase polymerase chain reaction and in situ hybridization that Smad7 mRNA is highly expressed in the developing vascular system of the mouse embryo but is also detectable much earlier in preimplantation embryos and during gastrulation. We also demonstrate by transient transgenesis that overexpression of Smad7 in mouse zygotes inhibits development beyond the 2-cell stage. This confirms earlier conclusions of similar, but complementary, experiments using a dominant negative type II TGF beta receptor demonstrating that TGF beta signalling is required for normal preimplantation development.

Expression of TGF-beta stimulated clone-22 (TSC-22) in mouse development and TGF-beta signalling

TSC-22 is a highly conserved member of a novel family of transcription factors, that is a direct target of transforming growth factor-beta (TGF-beta) in osteoblastic cells. We have investigated the expression of TSC-22 in detail during mouse development using in situ hybridization. We detected strong expression of TSC-22 in the embryo proper first at embryonic day 8.5 (E8.5), in the primitive heart, intermediate mesoderm and the neural tube. The dynamics of the TSC-22 distribution in the neural tube was particularly striking, with ubiquitous expression rostrally and restriction to neural tissue nearer the floor plate more caudally; between E8.5 and E9.5 the zone of restricted expression extended rostrally. At later stages of development, TSC-22 was detected in the mesenchymal compartment of many tissues and organs, including the lung, trachea, kidney, stomach, intestine, tooth buds, and in precartilage condensations. Furthermore, TSC-22 was highly expressed in the floor plate itself and notochord, and the endothelium lining the blood vessels, in particular the major arteries. Many of these sites have been proposed previously as possible TGF-beta target tissues; the results imply that TSC-22 may also be a direct TGF-beta target gene during mouse embryogenesis. Experiments on TSC-22 expression in embryoid bodies of embryonic stem (ES) cells expressing dominant negative TGF-beta binding receptors initially supported this hypothesis. However, examination of somatic chimeras derived from these same mutant ES cells at nominal E9.5 showed that TSC-22 expression in the heart and neural tube was still detectable despite obvious phenotypic abnormalities. We therefore propose that although TSC-22 may be a direct target of TGF-beta in late development, other factors are likely to be major regulators of expression at earlier stages.

Functional analysis of the TGFbeta receptor/Smad pathway through gene ablation in mice

During recent years, our understanding of TGFbeta signalling through serine/threonine kinase receptors and Smads has increased enormously. Activation of R-Smads by receptor induced phosphorylation is followed by complex formation with co-Smads and translocation to the nucleus, where the transcription of specific genes is affected and ultimately results in changes in cell behaviour. Experimental analysis primarily of epithelial cells in culture has revealed that a number of members of the TGFbeta family are interchangeable in the effect they have on growth and differentiation. On the other hand, different ligands of the TGFbeta superfamily can result in different responses because of cell type specific expression of other components of the signalling pathway. The relative expression levels of receptors and Smads within the cell is an important determinant of TGFbeta induced responses. Functional analysis of genes in the TGFbeta superfamily signal transduction cascade in vivo in mice either lacking entire genes, or expressing dominant negative forms of particular proteins, are providing profound new insights into the signalling cascades, their interaction and their specificity (Table 3). For example, by phenotypical comparison and intercrossing different heterozygous mutants, it has become clear that nodal, until recently an orphan protein without receptor/signal complex, probably signals through the activin type II receptor, ALK-4 and Smad2 (Nomura and Li, 1998; Song et al., 1999). Many of the genes of this cascade that have been targeted in the mouse result in early embryonic lethal phenotypes, demonstrating an important function for the BMP and TGFbeta/activin-activated pathways in mesoderm formation and differentiation, but masking a possible role in later events. For example mutations in BMP2 and 4 are lethal at or soon after gastrulation so that their putative role in skeletogenesis cannot be studied in mice lacking these genes. The difference in severity of the phenotypes between ligand, receptor and Smad deficient mice suggest that other receptors and ligands may partially compensate for the loss of one protein. Chimeric analysis provides one tool for analysing later developmental functions. By rescuing the early defects it was demonstrated that TGFbeta family members have an important function in anterior development and left/right asymmetry. Temporal and spatial specific gene targeting will be a powerful tool for analysing the function of TGFbeta family members in for example, bone formation, angiogenesis and carcinogenesis. Isolation of cells from the different gene targeted mice provides a unique source of material to gain more insight in the biochemical mechanisms of specific pathways. For example, use of cells deficient in Smad2 for biochemical and cell biological assays could give a better view of the function of Smad3. Smad3 deficient mice already demonstrate that there is a clear difference between Smad2 and Smad3 during development. Full descriptions of the remaining gene ablation studies of this signal transduction cascade, namely those for ALK-5, BMPR-II and Smad1 and -7 are eagerly awaited to complete the puzzle. As more of these superfamily of ligands and their signalling pathways have been functionally dissected, it has become evident that this superfamily of growth factors plays a pivotal role in epiblast formation and gastrulation, signalling from both the epiblast as well as the extraembryonic tissues. Furthermore, it becomes clear that TGFbeta is indeed important for proper vessel formation and that it might use endoglin, as well as ALK-1, ALK-5 and Smad5 to mediate this function. Further analyses of these mice should provide a clearer understanding of the mechanism of TGFbeta action in vascular development and remodelling.

Transforming growth factor-beta signalling in extraembryonic mesoderm is required for yolk sac vasculogenesis in mice

We have analysed the function of transforming growth factor beta (TGF-beta) in yolk sac development in mice by generating somatic chimaeras in which the extraembryonic mesoderm, which gives rise to the endothelial and haematopoietic cells of the yolk sac vasculature, is derived from embryonic stem (ES) cells. The ES cells were stably transfected and express either the full-length type II binding receptor or a kinase-deficient mutant of this receptor. Examination of yolk sacs from chimaeras between E8.5 and 9.5, and analysis of marker expression in embryoid bodies from these mutant ES cell lines in prolonged suspension culture demonstrated that (1) a major function of TGF-beta in yolk sac mesoderm is to regulate production and deposition of fibronectin in the extracellular matrix that maintains yolk sac integrity, (2) TGF-beta signalling is not required for differentiation of extraembryonic mesoderm into endothelial cells but is necessary for their subsequent organisation into robust vessels, and (3) TGF-beta signalling must be tightly regulated for the differentiation of primitive haematopoietic cells to take place normally. Together, these results show that defective TGF-beta signalling in the extraembryonic mesoderm alone is sufficient to account for the extraembryonic phenotype reported previously in TGF-beta1(−/−) mice (Dickson, M. C., Martin, J. S., Cousins, F. M., Kulkarni, A. B., Karlsson, S. and Akhurst, R. J. (1995) Development 121, 1845–1854).

Expression of the alpha 6A integrin splice variant in developing mouse embryonic stem cell aggregates and correlation with cardiac muscle differentiation

Mouse embryonic stem (ES) cells grown in aggregates give rise to several different cell types, including cardiac muscle. Given the lack of cardiac muscle cell lines, ES cells can be a useful tool in the study of cardiac muscle differentiation. The laminin-binding integrin alpha 6 beta 1 exists in two different splice variant forms of the alpha chain (alpha 6A and alpha 6B), the alpha 6A form having been implicated as possibly playing a role in cardiac muscle development, based on its distribution pattern [4, 53]. In this study we characterise the ES cell model system in terms of the expression of the two different alpha 6 splice variants. We correlate their expression with that of muscle markers and the transcription factor GATA-4, using the reverse transcription-polymerase chain reaction (RT-PCR). We confirm that alpha 6B is constitutively expressed by ES cells. In contrast, alpha 6A expression appears later and overlaps in time with a period when the muscle marker myosin light chain-2V (MLC-2V) is expressed, but no MyoD is present, which indicates the presence of cardiac muscle cells in the aggregates. We further show that GATA-4 is present at the same time. Culturing the aggregates under conditions that stimulate (transforming growth factor beta 1 supplement) or inhibit (TGF beta 1 plus 10(-9) M retinoic acid supplement) cardiac muscle differentiation does not lead to any qualitative differences in the timing of expression of these genes, but quantitative changes cannot be excluded. The TGF beta 1 supplement does, however, lead to a relatively greater expression of alpha 6A compared to alpha 6B than the TGF beta 1 plus 10(-9) M RA supplement after 6 days in culture, suggesting that alpha 6A expression is favoured under conditions that stimulate cardiac muscle differentiation. The switch towards alpha 6A expression in ES cell aggregates is paralleled by expression of the binding receptor for TGF beta (T beta RII). Stable expression of a mutated (dominant negative) T beta RII in ES cells, however, still resulted in (TGF beta-independent) upregulation of alpha 6A, demonstrating that these events were not causally related and that parallel or alternative regulatory pathways exist. The initial characterisation of differentiating ES cell aggregates in terms of alpha 6A integrin subunit expression suggests that this model system could be a valuable tool in the study of the role of the alpha 6A beta 1 integrin in cardiac muscle differentiation.

Ectopic expression of the transforming growth factor beta type II receptor disrupts mesoderm organisation during mouse gastrulation

Transforming growth factor beta (TGFbeta) regulates the cell cycle and extracellular matrix (ECM) deposition of many cells in vitro. We have analysed chimaeric mouse embryos generated from embryonic stem cells with abnormal receptor expression to study the effect of TGFbeta on these processes in vivo and the consequences for normal development. The binding receptor for TGFbeta, TbetaRII, is first detected in the embryo proper around day 8.5 in the heart. Ectopic expression of TbetaRII from the blastocyst stage onward resulted in an embryonic lethal around 9.5 dpc. Analysis of earlier stages revealed that the primitive streak of TbetaRII chimaeras failed to elongate. Furthermore, although cells passed through the streak and initially formed mesoderm, they tended to accumulate within the streak. These defects temporally and spatially paralleled the expression of the TGFbeta type I receptor, which is first expressed in the node and primitive streak. We present evidence that classical TGFbeta-induced growth inhibition was probably the cause of insufficient mesoderm being available for paraxial and axial structures. The results demonstrate that (1) TGFbeta mRNA and protein detected previously in early postimplantation embryos is present as a biologically active ligand; and (2) assuming that ectopic expression of TbetaRII results in no other changes in ES cells, the absence of TbetaRII is the principle reason why the embryo proper is unresponsive to TGFbeta ligand until after gastrulation.

Identification of two distinct functions for TGF-beta in early mouse development

In this study the function of transforming growth factor-beta (TGF-beta) in preimplantation mouse embryos was examined. By RT-PCR, mRNA for the signalling type I (T beta R-I) and type II (T beta R-II) receptors for TGF-beta was shown to be present in two distinct time windows: in fertilized oocytes and at the blastocyst stage. The function of TGF-beta at these times was analysed in two ways. Firstly, the TGF-beta signalling pathway was blocked by injecting a DNA construct encoding a truncated T beta R-II, that acts as a dominant-negative receptor, in fertilized oocytes, and the effect on development was determined. Secondly, inner cell masses isolated at the blastocyst stage were cultured in vitro with and without TGF-beta under conditions that favour the outgrowth of parietal endoderm. The results show that TGF-beta signalling mediated by maternally expressed receptors is important for development of preimplantation embryos beyond the two-cell stage, and suggest a regulatory role for TGF-beta in the outgrowth of parietal endoderm.

Mouse embryonic stem cells with aberrant transforming growth factor beta signalling exhibit impaired differentiation in vitro and in vivo

Embryonic stem (ES) cells are resistant to transforming growth factor beta (TGF beta). We have shown previously that they lack type-II binding receptors (T beta RII) and in this respect resemble the inner cell mass and ectoderm cells of mouse embryos 4.5-7.5 days post coitum (dpc); they do however express type-I (alk-5) signalling receptors. Here we show that in contrast to several tumour cell lines, stable transfection of wtT beta RII is not sufficient for ES cells to become biologically sensitive to TGF beta. We analysed the expression of several down-stream molecules known to be involved in TGF beta signalling (Smads) and TGF beta-mediated cell cycle regulation (cyclins D) during the differentiation of control and wtT beta RII-expressing ES cells and showed that upregulation of these molecules correlated with (i) an increase in plasminogen activator inhibitor-1 (PAI-1) synthesis and (ii) growth inhibition, following addition of TGF beta 1. These TGF beta responses were reduced in an ES cell line expressing a dominant negative (truncated) T beta RII (delta T beta RII). The differentiation pattern of control and wtT beta RII-expressing ES cells was indistinguishable in monolayer culture and as embryoid bodies, but in delta T beta RII ES cells, the capacity to form mesodermal derivatives in monolayer cultures in response to the addition of retinoic acid (RA) and removal of leukemia inhibitory factor (LIF) was lost, and only endoderm-like cells formed. The T beta RII and delta T beta RII ES cells were, however, both distinguishable from control ES cells when allowed to differentiate in chimaeric embryos following aggregation with morula-stage hosts. Conceptuses containing mutant cells, recovered from pseudopregnant females at the equivalent of 9.5 dpc, exhibited highly defective yolk sac development; most strikingly, no blood vessels were present and in addition the yolk sacs with derivatives of ES cells containing wtT beta RII were blistered and lacked haematopoietic cells. The implications for understanding TGF beta signalling in early mouse development are discussed.

Knockout and knockin of the beta1 exon D define distinct roles for integrin splice variants in heart function and embryonic development

The beta1D integrin is a recently characterized isoform of the beta1 subunit that is specifically expressed in heart and skeletal muscle. In this study we have assessed the function of the beta1D integrin splice variant in mice by generating, for the first time, Cre-mediated exon-specific knockout and knockin strains for this splice variant. We show that removal of the exon for beta1D leads to a mildly disturbed heart phenotype, whereas replacement of beta1A by beta1D results in embryonic lethality with a plethora of developmental defects, in part caused by the abnormal migration of neuroepithelial cells. Our data demonstrate that the splice variants A and D are not functionally equivalent. We propose that beta1D is less efficient than beta1A in mediating the signaling that regulates cell motility and responses of the cells to mechanical stress.

Differential expression of BMP receptors in early mouse development

Bone morphogenetic proteins (BMPs) are members of the transforming growth factor-beta family of polypeptide signaling molecules. They function via binding to two types of transmembrane serine/threonine kinase receptors, type I and type II receptors, that are both necessary for signaling. The expression patterns of the type II BMP receptor (BMPR-II) and three type I BMP receptors (ActR-I, BMPR-IA and BMPR-IB) were examined in preimplantation embryos by means of heminested reverse transcription-polymerase chain reaction (RT-PCR). BMPR-II mRNA was detected in one-cell, two-cell and blastocyst stage embryos. ActR-I exhibited a similar expression pattern. BMPR-IA mRNA however was only detected in blastocysts, whereas BMPR-IB transcripts were detected at all stages from the one-cell zygote to the uncompacted morula, but not in the compacted morula and blastocyst. If translated into proteins, this suggests that different receptor complexes can be formed at different developmental stages. Transcripts for BMPs were not detected in preimplantation embryos, but were detected in the maternal tissues surrounding the embryos. BMPR-II, BMPR-IA and BMPR-IB mRNAs were also detected in undifferentiated and differentiated embryonal carcinoma and embryonic stem cells. In postimplantation embryos BMPR-II transcripts were first detected from 6.0 days post coitum. In situ hybridization analysis revealed that BMPR-II mRNA is ubiquitously expressed in the entire embryo at least until midgestation.