AU2002331424A1 - Regulation of cytotrophoblast cell differentiation and cell migration - Google Patents

Description

Regulation Of Cytotrophoblast Cell Differentiation And Cell Migration

FIELD OF THE INVENTION

The present invention relates to the use of a method to adjust the behaviour of cytotrophoblast cells of the placenta and that of stem cells. In particular the invention is concerned with the regulation of their differentiation and migration.

BACKGROUND OF THE INVENTION

During embryo implantation and placeπtation, cytotrophoblast cells of the conceptus proliferate and migrate. Growth factors induce cytotrophobiasts to replicate, and to increase the number of cytotrophobiasts that migrate into the uterine endometrium. During this process cytotrophobiasts sequentially attach (adhere) and detach (de- adherβ) from surrounding cells and extracellular matrix. Migrating cytotrophobiasts also secrete extracellular matrix to which they can adhere, as well as proteolytic enzymes that degrade the extracellular matrix from which they detach and through which they migrate. Similar processes occur during differentiation of stem cells in the embryo, fetus, chiid, and adult.

In many biological processes, cells change or differentiate from one cell type to another in response to growth factor polypeptides and glycoproteins. In mammals, these growth factors may either originate from the progenitor cell undergoing differentiation (autocrine mechanism) or from neighbouring cells (paracrine mechanism). These biological processes include those that occur during normal mammalian development in which cells of different types in the conceptus change into other cell types that form the placenta, tissues and organs of the embryo, the fetus, and eventually the adult.

The differentiation pathway of cytotrophobiasts and embryonic and adult stem cells includes migration from one site to another. This can include invasion by the migrating cells into a tissue or organ comprised of or constructed by cells of other types or lineages.

Cells of all types proliferate and migrate during organogenesis and to increase the size of the tissue during normal development and growth in embryonic/fetal life and throughout childhood. Embryonic stem cells are pluήpotent cells derived from the inner cell mass of the early mammalian embryo from which all cell types of the embryo and all endodermal and mesodermal cells in the extra-embryonic tissues are derived. In vivo they differentiate in a low oxygen environment during the first third of pregnancy in humans and other mammals.

Adult stem cells are pluripotent cells found in all mammalian tissues from which many or all cell lineage types of the body may differentiate.

Cytotrophoblast cells are derived from extra-embryonic ectodermal cells of the conceptus, which comprise the trophectoderm of the blastocyst. Cytotrophoblast cells are therefore an epithelial cell type. Cytotrophoblast cells migrate into the endometrium of the maternal uterine decidua to form the placenta. Within the exchange region of the placenta cytotrophoblast cells, known as villous cytotrophobiasts, retain their epithelial phenotype. However, invasive cytotrophobiasts, also known as extravillous cytotrophobiasts, have undergone epithelial-mesenchymal transition, a process which allows them to assume a migratory phenotype. invasion of the maternal decidua by cytotrophoblast cells is terminated by fusion of cytotrophoblast cells to form multinucieate cells called placental bed giant cells.

Under specific conditions cytotrophobiasts and stem ceils, both embryonic and adult, are able to synthesise and secrete ϊnsulin-like growth factor II (IGF-11). These processes occur in a low oxygen environment

Since the IGF-II gene is imprinted and expressed by the paternal allele, the paternal genotype is important in determining the capacity of the placenta, the genotype of which is a combination of the paternal and maternal genotypes, to synthesise IGF-II. It is known that there are polymorphisms in the IGF-II gene which may determine the capacity of tissues to synthesis© IGF-II. Since the placenta synthesises abundant IGF-II it was postulated that mutations in the IGF-II gene may lead to pre-eclampsia. A common Apa I restriction fragment length polymorphism in exon 9 of the IGF-II gene was investigated as a possible mutation which causes pre-eclampsia (Bermingham et al. 2000). It was unequivocally shown that this polymorphism in the IGF-II gene is not involved in pre-eclampsia. The authors concluded that IGF-II does not play a role in the aetiology of pre-eclampsia. As there are other polymorphisms in this gene and neighbouring genes which determine the capacity of the placenta to synthesise IGF-II, and hence its capacity to invade the uterine decidua and establish optimal placental function, we claim that IGF-II plays a determining role in pre-eciampsia.

Expression of (GF-II is affected by a functional polymorphism of the insulin (INS) variable number of tandem repeats (VNTR) locus in humans. In Caucasians, the INS VNTR micro satellite divides into two classes of alleles which vary in size. Class 1 alleles (26-63 repeat units) have been strongly associated with insulin dependent diabetes mellitus (Bennett & Todd 1996), while Class HI alleles protect against IDDM but are associated with non-insulin dependent diabetes mellitus (Ong et al. 1999). The INS VNTR has been shown to be a long range control element for both insulin and IGF- II.

All nucleated cells in the body have the capacity to detect oxygen concentration and respond accordingly. Chronic reductions in oxygen concentration within the cell result in new gene expression which is mediated by several transcription factors. Hypoxia inducible factor -1α (HIF-1α) is an oxygen-sensitive transcription factor which regulates gene expression in response to low cellular oxygen concentration. HIF-1 α activates the transcription of a variety of target genes whose protein products are involved in angiogenesis, ceil proliferation and viability, and vascular remodelling. HIF-1β, also known as ARNT. is constitutively expressed and must bind to HIF-1 α, which is regulated by hypoxia (Huang et al. 1996), before binding to DNA as a heteromeric complex (Wang et al. 1995). iGF-ll is a target gene for HIF-1 a and is thought to be both regulated by, and a regulator of, HIF-1α (Feldser et al. 1999). Feldser et al. reported that in πormoxic conditions insulin, IGF-I and IGF-ll all induce HlF-1α protein expression resulting in transcription of its target genes including IGF-II, IGFBP-2 and IGFBP-3. When human chorionic villous explants were cultured in 2% O≤ and compared with those cultured in 20% O≥ cytotrophoblast proliferation as increased nearly 3-fold (Genbacev et al. 1997). In cultured human villous explants HIF-1 α transcription was stimulated by low oxygen tension (3% O≤) and cytotrophoblast proliferation ensued (Caniggia et al. 2000).

SUMMARY OF THE INVENTION

The present invention is predicated on the discovery of certain interactions between cellular growth factors and opposing actions that control differentiation and migration or invasion of cytotrophobiasts into the uterine endometrium which, during pregnancy, is called the decidua. In addition, this discovery may be applied to embryonic or adult stem cells to control their differentiation and migratory behaviour.

Therefore, according to a first aspect of the present invention, although this need not be the broadest nor indeed the only aspect of the invention there is provided a method of regulating cytotrophoblast and stem ceil differentiation and migration consisting of adjusting levels of IGF-II available for binding to the cation-independent mannose-6- phosphate (CIM6P) receptor.

We have discovered that IGF-II and latent transforming growth factor beta (TGFβ), the inactive precursor of TGFβ, compete for binding to the CIM6P receptor. IGF-II prevents latent TGFβ binding to the CIM6P receptor. It is known that binding of latent TGFβ to the C1M6P receptor leads to the production of active TGFβ by the urokiπase plasminogen activator (uPA) system (Godar et al. 1999). It is suggested that IGF-ll prevents activation of latent TGFβ. Cells that produce sufficient amounts of IGF-ll, or cells that are exposed to sufficient amounts of IGF-II, therefore cannot convert latent TGFβ into its active form by the uPA system that forms a complex with the CIM6P receptor on the surface of the cell.

The present invention therefore offers a method of regulating and directing cytotrophoblast differentiation and function based on the interaction between IGF-II, latent TGFβ and the C1M6P receptor.

During pregnancy in humans, IGF-ll is produced by cytotrophoblast cells and is most abundantly expressed at the invasive front. It is known that TGFβ promotes terminal differentiation in these cells, thus inhibiting their migratory behaviour.

We have discovered that IGF-II binding to the CIM6P receptor prevents local activation of TGFβ. In cytotrophoblast cells, IGF-ll thereby prevents migratory or invasive mesenchymal-type cytotrophoblast cells from differentiating into non-migratory or non- invasive giant cell types.

Competition for binding the CIM6P receptor is dependent on the concentration of IGF-li and the concentration of latent TGFβ in the vicinity of cytotrophobiasts. The CIM6P receptors on the surface of cytotrophobiasts that produce sufficient quantities of IGF-II or are exposed to sufficient amounts of IGF-ll are unable to bind latent TGFβ. In this way IGF-ll prevents cytotrophobiasts from activating latent TGFβ. In the presence of sufficient IGF-ll these cells therefore maintain their mesenchymal type and migratory activity.

Removal of IGF-ll permits these cells to activate latent TGFβ. The action of TGFβ converts these cells into non-replicating and non-migratory types.

It will be appreciated that in the placenta the promotion of cytotrophoblast migration is highly desirable. It will also be appreciated that the application of stem cell technology to treat disease may require promotion of epitheiial-mesenchymal transition or its inhibition depending on the direction of stem cell differentiation required.

Thus, in a first preferred form of the invention there is provided a method of promoting invasive and migratory cell behaviour by exposing cytotrophobiasts or stem cells (embryonic or adult) to elevated levels of IGF-ll, such that the cell CIM6P receptors are unable to bind latent TGFβ.

In reproduction treatment with insulin-like growth factor II (IGF-II) promotes implantation and placentation:

Insulin-like growth factor II treatment of embryos promotes conversion of trophectoderm cells into cytotrophoblast cells thereby increasing the success of Implantation of embryos into the uterine deciduai endometrium, thereby increasing the success of formation of a viable placenta, and thereby improving the rate of successful pregnancy.

Addition of insulin-like growth factor II to embryos produced by in vitro fertilisation techniques can be used to treat infertility. Thus, treatment of pregnant women or their embryos with insulin-like growth factor II can be used to prevent recurrent spontaneous miscarriage and to prevent pre-eclampsia.

Further, administration of IGF-II to pregnant women or their embryos may prevent iπtrauterine growth restriction or be used to treat placental abruption.

Treatment of embryonic stem cells or adult stem cells with insulin-like growth factor II or culture in a hypoxic environment (1% oxygen) which increases their synthesis of insuiin-like growth factor If, promotes their differentiation into mesodermal/mesenchymal cell types. In a further form of the invention there is provided a method of inhibiting stem cell division, stem cell migration and promoting terminal differentiation behaviour by exposing said stem cell to reduced levels of IGF-II, such that the stem ceil CIM6P receptors are able to bind latent TGFβ and thereby promote the activation of TGFβ.

Thus, treatment of embryonic stem cells or adult stem cells with inhibitors of IGF-ll action on CIM6P receptor (for example soluble CIM6P receptor or fragments thereof) will promote their differentiation into epithelial cell types, including neurones or their precursors.

The invention is, in a further aspect exemplified by the use of not only IGF-II in its basic form, but also to naturally occurring precursors of and isomers of IGF- II. Similarly, the treatments identified above may be carried out also using synthetic analogues of IGF-ll that have altered ability to bind to type-1 IGF receptors, altered ability to bind to insulin receptors, altered ability to bind to IGF-binding proteins and increased ability to bind to cation-independent maππose-6-phosphate receptors (also known as CIM6P receptors, tyρe-2 IGF receptors and IGF-ll receptors).

Still further aspects of the invention are concerned with diagnostic uses of IGF-II and the recognition that variation in the capacity of the placenta to produce IGF-II allows predictions to be made concerning the differentiation/migration behaviours to be expected from cytotrophobiasts and therefore the capacity of the placenta to sustain a healthy pregnancy.

Thus, in accordance with a still further aspect of the invention measurement of the composition of and the sequence of πucleotides in the deoxyribonucleic acid near the insulin-like growth factor II gene in human embryos can be used to diagnose their ability to establish successful healthy pregnancy. In particular, the sequence of nucleotides known as the insulin (INS) variable number of tandem repeats (VNTR), which is known to determine placenta! production of IGF-II, specifically determines the capacity of cytotrophobiasts to synthesize IGF-ll and hence determines their capacity to migrate into the Uterine decidua and therefore determines the capacity of the placenta to transport substrates to the embryo/fetus.

Alternatively, determination of the nucleotide sequence of the same gene locus INS

VNTR on biological specimens (for example btood) from both the mother and the father can be used to diagnose their ability to establish successful healthy pregnancy. Similarly, measurement of the amount of messenger ribonucleic acid transcribed from the insulin-like growth factor II gene in human embryos can be used to diagnose their ability to establish successful healthy pregnancy.

Measurement of the amount of insulin-like growth factor II protein secreted by human embryos can also be used to diagnose their ability to establish successful healthy pregnancy.

The measurements as described above may be performed on a specimen obtained from either or both biological parents of a human embryo and can be used to diagnose their ability to establish successful healthy pregnancy. This information can thus be used to assist in identifying those individuals who are likely to derive most benefit from treatment according to the methods of the invention.

It should be noted that the methods of the invention may be used in humans as well as in other mammalian species, such as the horse, cow, sheep, goat and pig.

The action of IGF-II in regulating cytotrophoblast differentiation/migration is provided below. IGF-II is produced by cytotrophoblast cells. These cells are known to have receptors able to bind IGF-II (Rebourcet et al. 1998). These receptors include type-1 IGF receptors, which bind IGF-I, IGF-II and insulin. In many cell types, binding of IGF-I, IGF-II or insulin to the type-1 IGF receptor promotes cell division, also alternatively known as mitosis, replication, multiplication or proliferation.

Cytotrophoblast cells have another receptor type that binds IGF-ll very well, binds IGF-I extremely poorly and does not bind insulin. This receptor has been called the type-2 IGF receptor (Rebourcet et al. 1998). This receptor also binds certain glycoproteins that contain mannose-6-phosphate and has also been called the cation-independent mannose-6-phσsphate (CIM6P) receptor. We have discovered that competition between IGF-ll and latent TGFβ for binding to the CIM6P/type-2 IGF receptor underlies the action of IGF-II in regulating the ability of cytotrophobiasts to undergo migratory or non-migratory behaviours in any environment.

In the human placenta, IGF-II is most abundantly produced by cytotrophoblast cells that have migrated furthest into the maternal decidual endometrium (Guidice et al. 1998; Irwin et al. 1999). IGF-II promotes migration of cytotrophoblast cells by a mechanism that has hitherto been unknown (Hamilton et al. 1998). Cytotrophoblast cells have TGFβ receptors able to bind TGFβ (Schilling & Yeh 2000). Treatment of cytotrophoblast cells with TGFβ promotes their fusion to form non- migratory multinucleatθ giant cells (Morrish et al. 1998).

Active TGFβ, capable of binding to TGFβ receptors, is derived by proteolytic conversion from inactive latent TGFβ, which does not bind to TGFβ receptors. Latent TGFβ is produced by maternal uterine decidual cells and by cytotrophoblast cells (Graham & Lala 1991).

Latent TGFβ is known to bind to the CIM6P receptor. Latent TGFβ can also bind to CIM6P receptors that are associated in a complex with plasminogen and urokinase plasminogen activator (uPA) receptors.

It is known that uPA converts latent TGFβ into active TGFβ by the catalytic action of uPA when bound to the complex formed by the simultaneous association of uPA receptor, plasminogen and latent TGFβ with the CIM6P receptor (Godar et al. 1 99). Cytotrophoblast cells are known to have uPA receptors (Floridon et al. 1999).

It has been suggested that binding of IGF-ll to the CIM6P receptor is a mechanism that removes and degrades IGF-II. The CIM6P receptor competes with the type-1 IGF receptor for IGF-ll binding. IGF-II binding to the type-1 IGF receptor promotes cell division and replication. According to the prior art, the CIM6P receptor reduces the amount of IGF-II able to bind to the type-1 IGF receptor and thus reduces the ability of cells to proliferate or replicate in response to IGF-II. Thus, the CIM6P receptor by competing with the type-1 IGF receptor prevents IGF-II from promoting cell replication. According to the prior art, IGF-II binding to the CIM6P receptor does not produce a direct biological response within the ceil upon whose surface the C1M6P receptor is located (Odeli δ Day 1998).

There is an additional body of evidence that shows that activation of latent TGFβ occurs on CIM6P receptors located on the cell surface and that this is dependent on simultaneous binding of the uPA/piasmϊnogen system (Godar et al. 1999).

However, it has hitherto not been recognised that IGF-II and latent TGFβ compete for binding at the same site on CI 6P receptors and that their competition for binding regulates the amount of active TGFβ formed by this receptor complex and therefore the amount of active TGFβ available to its specific receptors on the cytotrophoblast cell surface. We have discovered that competition between IGF-II and latent TGFβ for the CIM6P receptor occurs and that IGF-II inhibits activation of latent TGFβ and that this is therefore a mechanism for the regulation of cytotrophoblast differentiation and migration.

DESCRIPTION OF DRAWINGS

Figure 1 shows schematically the mechanism by which IGF-ll binding to the CIM6P receptor permits cytotrophoblast invasion of the decidua;

Figure 2 illustrates the effect of IGFs on activation of TGFβ by human TF1 monocytes;

Figure 3 illustrates the effect of oxygen on the number of outgrowing tips from human placental villous explants which formed cytotrophoblast cell columns;

Figure 4 the effect of oxygen on IGF-II mRNA expression by human placental villous explants at 7.5-8 weeks gestation;

Figure 5 illustrates the effect of treatment with IGF-II between days 2 and 10 of pregnancy in the mouse on placental weight at day 18;

Figure 6 illustrates the effect of treatment with IGF-I! between days 2 and 10 of pregnancy in the mouse on the distribution of placental weights at day 8;

Figure 7 illustrates the effect of treatment with IGF-ll between days 2 and 10 of pregnancy in the mouse on fetal weight at day 18;

Figure 8 illustrates the effect of treatment with IGF-ll between days 2 and 10 of pregnancy in the mouse on the distribution of fetal weights at day 18;

Figure 9 illustrates the effect of treatment with iGF-ll between days 2 and 18 of pregnancy in the mouse on placental weight at day 18;

Figure 10 illustrates the effect of treatment with IGF-II between days 2 and 18 of pregnancy in the mouse on placental thickness at day 18;

Figure 11 illustrates the effect of treatment with IGF-ll between days 2 and 18 of pregnancy in the mouse on the distribution of fetal weights at day 18; Figure 12 illustrates the effect of treatment with IGF-II between days 2 and 18 of pregnancy in the mouse on the distribution of placental weights at day 18; and

Figure 13 illustrates the effect of treatment with IGF-ll between days 2 and 18 of pregnancy in the mouse on the ratio of fetal to placental weights at day 18.

DESCRIPTION OF THE PREFERRED EMBODIMENT

The discovery of the interaction of IGF-ll and TGFβ with the CIM6P receptor can be applied by the use of IGF-II in early pregnancy to treat both implantation failure and recurrent spontaneous miscarriage.

During pregnancy cytotrophoblast cells that are originally derived from the trophectoderm of the blastocyst and which subsequently form the placenta remodel the endometrium. The placenta is the active interface between maternal and fetal tissues and is the organ through which exchange of soluble materials between fetus and mother occurs.

The embryo establishes its attachment with the endometrium through the actions of migratory cytotrophoblast cells known as extravillouε cytotrophobiasts. They secrete metalloproteinases that degrade the extracellular matrix of the endometrium, permitting their invasion. These invasive cells modify the endometria! spiral arterioles to provide access for the placenta to the maternal blood supply, the source of all matter for embryonic and fetal growth. Cytotrophoblast cells form columns that extend deep into the endometrium. The first wave of invasion of the endometrium by cytotrophoblast cells occurs between 6 and 12 weeks of pregnancy in humans (Pijnenborg et al. 1983). This is the period when most miscarriages occur. It is also the period in which there is no maternal blood flow into the placenta such that cytotrophobiasts proliferate and differentiate in a hypoxic environment. IGF-ll is abundantly produced early in pregnancy by cytotrophoblast cells at the leading edge of the invasive f ont of the cell column. IGF- ll promotes the invasive behaviour of cytotrophoblast cells. Latent TGFβ is produced by decidual cells of the maternal endometrium as well as by cytotrophoblast cells. TGFβ promotes fusion of cytotrophoblast cells, an essential step in their differentiation to form mature components of the placenta (Morrish et al. 1998). TGFβ also induces synthesis and secretion of tissue inhibitors of metalloproteinases (TlMPs) by cytotrophoblast cells (Lala & Graham 1990), thus preventing cytotrophoblast cells from degrading the extracellular matrix necessary for their migration. We hypothesise that early pregnancy loss due to failure to establish a viable placenta is caused by IGF-II deficiency in the IGF-lI/TGFβ/ClM6P receptor system of cytotrophobiasts. Further, it appears that failure of the embryo to successfully implant into the uterine endometrium of the mother is due to insufficient production of IGF-II by cytotrophoblast cells derived from the trophectoderm of the blastocyst, leading to reduced autocrine stimulation of invasion of the endometrium by cytotrophoblast cells. Thus, removal or lack of IGF-II induces terminal differentiation and fusion of cytotrophobiasts.

Spontaneous miscarriage (diagnosed between the 6th and 12th weeks of pregnancy) may thus be caused by deficiency in IGF-ll production of cytotrophoblast cells leading to premature fusion before sufficient colonisation of the endometrial decidua has been achieved to establish a healthy placenta. Accordingly, IGF-II can be used to treat both implantation failure and recurrent spontaneous miscarriage.

Impaired cytotrophoblast invasion of the uterine decidua is associated with pre- eclampsia, a common hypertensive disorder of pregnancy which is life-threatening in 3% of pregnancies in developed nations. It follows that deficiency of iGF-ll synthesis by extravillous cytotrophobiasts results in reduced invasive behaviour, poor placentation and pre-eclampsia. TGFβ has also been associated with pre-eclampsia as women who develop this disease have elevated concentrations of active TGFβl in their blood (Djurovic et al. 1997). We claim that this is a result of insufficient IGF-ll to maintain activation of TGFβ at normal levels.

We have discovered that IGF-II and latent TGFβ, the precursor of TGFβ, compete for binding to C1M6P receptors in human placenta. Latent TGFβ binding to CIM6P receptor is known to lead to its proteolytic conversion by plasmiπ into active TGFβ by the complex formed between urokinase plasminogen activator (uPA), uPA receptor and plasminogen (Godar et al. 1999), It is known that cytotrophoblast cells have CIM6P receptors (Rebourcet et al. 1998) and uPA receptors (Florldon et al. 1999). It follows from our discovery that secretion of IGF-II by cytotrophoblast cells maintains their invasive state by preventing local activation of TGFβ, the initiator of their terminal differentiation into πoπ-replicatiπg and non-migratory cells. It is known that invasive cytotrophoblast cells produce IGF-ll in abundance (irwiπ et al. 1999). We show herein that IGF-II and latent TGFβ compete for binding to the placental CIM6P receptor. It is known that activation of TGFβ from its latent form is a prerequisite for its promotion of cytotrophoblast cell fusion and terminal differentiation. It has been discovered that IGF-ll inhibits terminal differentiation of cytotrophoblast cells and, that this is the means by which IGF-I! promotes cytotrophoblast cell invasion of the decidua.

In figure 1 the mechanism by which IGF-II promotes cytotrophoblast invasion of the decidua is shown. In (A) IGF-ll inhibits binding of latent TGFβ to the receptor and permits the simultaneous activation of matrix metalloproteinases by plasmin produced by the action of uPA on plasminogen secreted by cytotrophobiasts to facilitate their migration into the uterine decidua. (B) When latent TGFβ binds the CIM6P receptor, plasmin that is generated as above is available to cleave the latency associated peptide from latent TGFβ resulting in activation of TGFβ which may then bind its specific receptors at the cell surface and induce cytotrophoblast terminal differentiation.

In accordance with our discovery that IGF-II and TGFβ compete for CIM6P binding, we have also discovered that during placental development, CIM6P receptor-bound IGF-I! maintains cytotrophoblast cell type and promotes degradation of extracellular matrix essential for invasion of the endometrium by cytotrophoblast cells. Binding of latent TGFβ to the same CIM6P receptor, in preference to IGF-II, activates TGFβ which initiates cytotrophoblast differentiation into a non-migratory cell type and promotes fusion into multinucieate placental bed giant cells. Thus, it is suggested that early pregnancy loss is characterised by inadequate IGF-ll synthesis or premature reduction or cessation of IGF-II synthesis and secretion by extravillous cytotrophoblast cells, which allows binding of latent TGFβ to their CIM6P receptors. Deficiency in IGF-ll results in premature activation of TGFβ by the uPA system complexed to the CIM6P receptor. We claim that premature activation of TGFβ is a consequence of IGF-II deficiency in cytotrophoblast cells. Active TGFβ then promotes premature fusion of cytotrophoblast cells. This reduces the numbers of invading extravillous cytotrophoblast cells below critical levels required to form a placenta that can sustain adequate extraction of oxygen and nutrients from the utero-placental blood supply for use by the placenta and the embryo and fetus. The placenta necroses and miscarriage ensues. We claim that IGF-II delivered to the implantation site promotes extravillous cytotrophoblast cell invasion. As normal pregnancy advances, cytotrophoblast cells become progressively exposed to increasing amounts of both oxygen and maternal adrenal glucocorticoids. It is known that both oxygen and glucocorticoids inhibit production of IGF-II by many cell types. This allows more latent TGFβ to bind to the C1M6P receptor and therefore increases the amount of active TGFβ produced at the cell surface. Increased exposure to active TGFβ causes terminal differentiation of cytotrophobiasts into non-replicating, non- invasive cells. The timing of premature terminal differentiation of cytotrophobiasts determines the severity of the defect in implantation and placeπtation.

Up to 50% of miscarriages have been attributed to inadequate invasion of the decidua by cytotrophoblast cells (Khong et al. 1987). We claim that reduced IGF-ll production by cytotrophoblast cells during embryo implantation and in the first weeks of pregnancy causes implantation failure and miscarriage.

Pre-eclampsia, a hypertensive disorder that occurs later in pregnancy and is also characterised by impaired cytotrophoblast cell invasion of the decidua (Khong et al. 1986), is, according to our discovery, caused by IGF-II deficiency of cytotrophoblast cells. The effects of impaired cytotrophoblast invasion in the first trimester of pregnancy are not apparent symptomatically in the mother until later. The most severe form of pre- eclampsia is diagnosed during the middle trimester of pregnancy, while less severe cases are not diagnosed until later in the third trimester. We^'claim that the more severe forms of pre-eclampsia are characterised by a poor capacity of cytotrophoblast cells to synthesise IGF-II and a poor response of these cells to synthesise IGF-II in hypoxic conditions as occur in the first trimester of pregnancy. Less severe forms of pre- eclampsia are also due to inadequate cytotrophoblast production of IGF-I! but their IGF- II response to hypoxia is between that in severe cases and normal pregnancy.

A diagnostic test based on cytotrophoblast capacity for IGF-il synthesis conducted early in pregnancy to identify those at risk of symptoms of pre-eclampsia later in pregnancy would allow early treatment and monitoring by health care providers. It is our discovery that IGF-II can be used to prevent and treat pre-eclampsia by elevating circulating IGF-ll concentrations to the normal pregnancy range which has been previously shown to be about 2_J200ng/ml in pooled serum compared to about 1500- 1600ng/ml in the non-pregnant state (Gargosky et al. 1990).

Convenient methods of treatment with IGF-II would include the use of subcutaneous delivery devices or extended release vagina! pessaries. A range of other conventional delivery systems would be suitable. Treatment with IGF-II is concerned with raising the maternal blood level of IGF-II to 'normal' levels throughout the pregnancy, that is the treatment is a generally remedial action preferably to be applied in situations where diagnosis has indicated a predicted or measured reduction in IGF-II levels. The duration of treatment will therefore be specific to each individual concerned. In some cases the treatment may be continued throughout the pregnancy and in other treatment may be required only during the first half of the pregnancy.

A less severe form of impaired cytotrophoblast cell invasion leads to placental abruption (Dommisse & Tiltman, 1992), in which premature separation of the placenta from the decidua occurs in the last weeks of pregnancy, and which requires emergency delivery of the baby, is also due to moderate IGF-II deficiency of cytotrophoblast cells. IGF-ll can be used to prevent placental abruption.

IGF-II deficiency impairs the ability of cytotrophoblast cells to invade the decidua. The diseases of pregnancy discussed above form a continuum of related disorders that depend on the degree of IGF-li deficiency and consequently the timing and extent of the reduced invasion of the decidual endometrium by cytotrophoblast cells.

Example 1 IGF-II competes with latent TGFβ for binding to cation-independent mannose-6-phosphatθ receptors on human placental microsomal membranes

The competition between iGF-ll and latent TGFβ for the CIM6P binding site can be demonstrated as follows:.

Human placenta was obtained from the Women's and Children's Hospital, North Adelaide after delivery of a normal baby following an uncomplicated pregnancy, and used as a source of CIM6P receptors (Owens et al. 1980). A suspension of microsomal membranes was prepared from the placenta by homogenisation and differentia! ultra- ceπtrifugation (Owens et al. 1980). Human IGF-ll and IGF-I (receptor grade) were obtained from GroPep Pty. Ltd. Human latent TGFβl was purchased from R&D Systems, Inc. IGF-II was radioactively labelled with iodine-125 to a specific activity of 70 Ci/g using chloramine-T (Francis et al. 1989). IGF-II labelled in this manner does not bind to type-1 IGF receptors (Francis et al. 1989). Replicates of 10μg of human placenta! microsomal protein were simultaneously incubated at 4°C for 20 h in 150 μl of 0.05 mol/litre Tris-HCI buffer pH 7.4 containing 1 picomoie of iodo-125-IGF-ll, 0.01 mol/litre calcium chloride and 5 gram litre bovine albumin (Owens et al. 1985). Different replicates also contained either IGF-II, IGF-I or latent TGFβl at a number of different concentrations. Radioactivity bound to human placental microsomal membranes was recovered by centrifugatibn (Owens et al. 1985). Blank binding of iαdo-125-IGF-II was measured in the absence of human placental microsomes. Non-specific binding of iodo-125-IGF-lf was measured in the presence of unlabelled IGF-II at a concentration of 133 nanomol litre (Owens et al. 1985).

IGF-I, IGF-II and latent TGFβl inhibited the binding of iodo- 25-IGF-II to human placental microsomes (Table 1).

Unlabelled IGF-II when present at a concentration of 1.6 nanomol/litre inhibited specific binding of iodo-125-IGF-ll by 40% and by 87% when present at a concentration of 15 nanomol/litre. Unlabelled IGF-l at a concentration of 1,600 nanomol/litre inhibited the specific binding of iodo-125-IGF-ll by 68%. Unlabelled latent TGFβl inhibited specific binding of iodo-125-IGF-ll by 25% when present at a concentration of 115 nanomol/litre (Table 1).

Table 1 Effects of unlabelled insulin-like growth factors and unlabelled latent transforming growth factorβ on binding of radioactiveiy labelled insulin-like growth factor II to human placenta! microsomal membranes.

Radioactivity added 40,709 counts per minute (cpm)

Total radioactivity bound 7,733 cpm (19.0% of the radioactivity added)

Blank radioactivity bound 2,606 cpm (6.4% of the radioactivity added)

Radioactivity bound non-specifϊcally 3,558 cpm (8.7% of the radioactivity added)

Radioactivity bound specifically = 7,733 cpm - 3,588 cpm

= 4,145 cpm

= 10.2% of the radioactivity added

= 53.6% of the total radioactivity bound

Radioactivity bound in the presence of the following:

unlabelled IGF-II at 1.6 nanomol/litre 6,059 cpm unlabelled IGF-II at 15 nanomol/litre 4,120 cpm

unlabelled IGF-I at 1 ,600 nanomol litre 4,905 cpm

unlabelled latent TGFβl at 116 6,717 cpm nanomol/litre

^'Radioactivity bound specifically is calculated to be the arithmetic difference between the total radioactivity bound and the radioactivity bound πon-specifically.

Example 1 demonstrates that latent TGFβl and iodo-125-IGF-ll compete for binding to the same site in suspended microsomal membranes prepared from human placenta. IGF-II also inhibits binding of iodo-125-IGF-ll to the same cell membranes.

The specific binding of iodo-125-IGF-ll to human placental microsomes in this example is not due to binding to either insulin receptors ortype-l IGF receptors, because more than 100-times more unlabelled IGF-l is required to inhibit binding of iodo-125-IGF-ll by as much as achieved by unlabelled IGF-II (Table 1). It is known that IGF-I and IGF-ll have similar affinities for their binding of both insulin receptors and type-1 IGF receptors, it is known that IGF-I has a very poor affinity for binding the CIM6P receptor.

The example above demonstrates that iodo-125-IGF-ll binds to CIM6P receptors in human placenta and that latent TGFβl inhibits this binding. Accordingly, latent TGFβ and IGF-II compete for binding to C1M6P receptors.

Example 2 IGF-ll inhibits activation of latent TGFβ

The result of competition between IGF-II and latent TGFβ for the CIM6P binding site and its effect on activation of latent TGFβ can be demonstrated as follows:

Human TF1 cells, a monocyte cell line, which express essential components of this system described above ie. CIM6P receptors, urokinase plasminogen activator receptors and secrete abundant latent TGFβl into the culture media, were cultured in serum-free RPM1 1640 media (JRH Biosciences) with 2% serum replacement (Sigma), 3.5 x 10^% β-mercaptoethanol (Sigma), 4mM l-glutamine (CSL), 1 μg ml plasminogen (Sigma), 5μg/ml urokinase plasminogen activator (uPA) (Sigma) and 2ng/ml recombinant human granulocyte-macrophage colony stimulating factor (GM-CSF) (Leucomax©, Schering-Plough). In replicate wells, monocytes were incubated with either 0, 1 , 12.5, 25, 50 or 100nM receptor grade IGF-II or 25nM receptor grade Leu²⁷IGF-H (GroPep, Adelaide). Approximately 1-2 x 10⁶ monocytes were placed in 500μl media and cultured for 24-26 hours at 37°C in 20% oxygen, 5% carbon dioxide, 75% nitrogen. The cell suspension was then gently aspirated and spun in a centrifuge at 200g for 10 mins at room temperature. The supernatant was collected and frozen at - 20^DC until assay for active and total TGFβ-1. Cells were resuspended in Hank's Balanced Salt Solution (HBSS, JRH Biosciences) and counted to determine the effect of culture on cell proliferation. Supematants were thawed and assayed forTGFβ-1 using a commercially available human TGFβ-1 enzyme-linked irnmunosorbent assay (ELISA) kit (R&D Systems, Inc.) according to the manufacturer's instructions. Color development was quantified using a BioRad Benchmark microplate reader set at a wavelength of 450nm and corrected to 550πm. The concentration of TGFβ-1 was calculated automatically by comparing with a standard curve of known concentrations of TGFβ on the same microplate using BioRad Microplate Manager 5.0 PC Software. Active and total TGFβ-1 were then calculated per million cells at the end of the culture period. The following table of results, table 2, shows thatTF monocytes secrete abundant latent TGFβ-1 into the culture medium and that iGF-ll inhibits activation of latent TGFβ-1. This is shown graphically in figure 2. When uPA is omitted from the culture media there is no activation of latent TGFβ-1. This provides further evidence that it is the uPA/uPAR/CIM6P receptor system which is responsible for activation of latent TGFβ on the cell surface.

Table 2 The effect of IGF-II on activation of latent TGFβ by 1x10^s TF1 monocytes in vitro.

IGF-ll or Leu²⁷-IGF-ll Total TGFβ-1 Active TGFβ-1 as Concentration (pg/ml) %Control

0 πmo!/L IGF-II 238.49±33.09 100.0

1 nmol/L IGF-II 229.73+6.33 4.25±2.60

12.5 nmol/L IGF-ll 252.67+-10.41 10,28+6.33

25 nmol/L IGF-ll 274.70+25.13 5.37±3.55

50 nmol/L IGF-ll 335.95 0.00 00 nmol/L IGF-ll 321.08 0.00

25 nmol/L Leu²⁷-I.GF-ll 366.64 12.72

AH data is Mean ±SEM

Hence, we have shown that IGF-II inhibits the activation of latent TGFβ in vitro.

Example 3 Effect of oxygen on cytotrophoblast column formation in vitro

It is known that binding of latent TGFβ to CIM6P receptors present on monocytes that also have uPA receptors results in proteolytic conversion of latent TGFβ into active TGFβ (Godar et al. 1999). It is known that cytotrophoblast cells have CIM6P receptors (Rebourcet et al. 1998) and uPA receptors (Floridon et al. 1999). TGFβ is known to change cytotrophoblast cells from the invasive type to the fused, terminally differentiated type resulting in formation of ultinucleate trophoblast cells called placental bed giant ceils. It therefore follows that IGF-ll inhibits activation of latent TGFβ by cytotrophoblast cells by preventing latent TGFβ from binding to CIM6P receptors of cytotrophoblast cells.

We also show herein that during eariy human pregnancy (7.5-8 weeks) when placentation is being established and cytotrophobiasts are known to be exposed to low oxygen tensions, they are most invasive and both IGF-ll protein and RNA expression is increased. The effect of oxygen on IGF-ll mRNA expression and the association of IGF-II mRNA with extravillous cytotrophoblast column formation can be demonstrated thus:

Human first trimester placentas were obtained from the Women's and Children's Hospital, North Adelaide, from legal elective terminations of normal pregnancies of 7.5- 8 weeks gestation. Placental villous tissue was obtained immediately after vaginal curettage. Vilious tissue was washed to remove blood, using cold, sterile PBS, then transported to the laboratory on ice, in HEPES buffered DMEM/F12 supplemented with 10% heat-inactivated fetal bovine serum (HIFBS) and 40 ng/ l gentamycin (GIBCO BRL, Grand Island, NY). The tissue was then rinsed and dissected into 2-3mm pieces in villous culture media (described below) at 37°C, under aseptic conditions. Villous culture media consisted of DMEM F12 (GIBCO BRL, Grand Island, NY) supplemented with: 0.04 M l-glutamine; 20 ng/ml plasminogen; 20 μl/m! antibiotic mix (GIBCO BRL, Grand Island, NY), comprised of 200 units/ml penicillin, 200 μg/ml streptomycin and 0.5 μg/ml amphotericin; with 10% HIFBS (CSL, Parkville, Vic).

Falcon 24-well tissue culture plates (Becton Dickinson, Kranklin Lakes, NJ) were coated with 20 μl of 1 :1 Matrigehculture media (Matrigel from GIBCO BRL, Grand Island, NY), set in a humid incubator at 37°C for at least 30 minutes. Villous pieces were weighed, one piece carefully plated per well, and covered with 15 μl of 1:1 MatrigeLculture media. The Matrigel was set in an incubator for 20 minutes, before 400μl of culture media was added and the explants cultured overnight, at 37°C in 5% C0₂ and either 20% or 1% 0₂ (equivalent to 98 mmHg or 7-10 mrnHg).

The following morning (« 15 hours), by which time villous tips were adherent to the bottom of the well, 600 μl of culture media was added per well, and explants were returned to the same culture conditions. Cultures were maintained for 6-8 days, with media changed every second day.

Villous cultures were digitally photographed (on an Olympus DP1 camera) immediately before RNA extraction was performed. From these photographs, cytotrophoblast outgrowths were visually scored for their pattern of outgrowth. Absolute counts were recorded for: 1 ) number of adherent villous tips with outgrowth and 2) number of cell column formations. The number of celi columns per explant were then divided by the total number of tips with outgrowth for that explant.

TRIZOL (GIBCO BRL, Grand Island, NY) was used to extract RNA from villous cultures, according to the manufacturer's protocol for cells grown in a monolayer. For placental villous cultures, 300μl TRIZOL per well was used. RNA concentration was determined by measuring UV light absorbance at 260 nm on a DU-50 Spectrophotometer (Beckman Instruments, Irvine, CA).

Aliquots containing 2 μg of RNA from each sample were reverse transcribed. The RNA was incubated with random hexamer primers at 65°C for 10 minutes. Tubes were cooled on ice for 5 minutes and ExpandRT, 5x buffer, DTTs (Roche Diagnostics, Indianapolis, IN), and Ultrapure dNTPs (Amersham Pharmacia Biotech, Piscataway, NJ) were added, according to the manufacturer's instructions. The tubes were then incubated at 30°C for 5 minutes, 42°C for 45 minutes, and 95°C for 2 minutes, cooled on ice and the cDNA was stored at -20°C.

Using PrimerExpress software (Applied Biosystems, Foster City, CA), IGF-ll oligonucleotide primers were designed to span exoπs 3 and 4 of the human IGF-II gene (GenBank accession number XQ3562), giving a predicted arnplicon of 90-100 bp. Primers were: forward, 5'-CCC CTC CGA CCG TGC T-3'; reverse, 5'-TGG ATG GAC TGC TTC CAG GTG TCA T-3'. Standard purity primers were made by GensetOligos (Genset Pacific, Usrnore, Australia).

Aliquots of 2 μl of oDNA were amplified in a reaction mixture (20 μl) containing 1 μ! of each IGF-ll primer at 10 μM/μl, 1 μl of SYBR Green Master Mix (including AmpliTaq^® Gold DNA Poly erase, Applied Biosystems, Foster City, CA) and 6 μl of molecular grade water (Fluka BioChemika, Messerschmϊttstr, Germany). For the 18s rRNA endogenous control, 2 μl of a 1/10 cDNA dilution was amplified in a reaction mixture (20 μl) containing 2 μl of Universal 18s primer pair mix (Ambioπ, Austin, TX), 10 μl of SYBR Green Master Mix and 7 μl of molecular grade water. Three replicates of each cDNA sample were amplified, and negative controls in Which cDNA was omitted (water substituted) were included to test for contamination.

Using a GeπeAmp 5700 thermal cycler (Applied Biosystems, Foster City, CA), the samples were amplified in 1 cycle of 50°C for 2 minutes, 95°C for 10 minutes, followed by 40 cycles of 95°C for 15 seconds and 60°C for 1 minute. The critical threshold (C_τ) value for each sample was determined with the fluorescence threshold set at 0.150 runs.

Aliquots of 7 μl of PCR product were fractionated by eleσtrophorβsis, in parallel with pUC19 molecular weight markers, in a 2% agarose gel (Sigma Chemical Co, St Louis, MO), stained with ethidium bromide. Bands were visualised under UV light, and digitally photographed.

Figure 3 shows graphically that the number of cytotrophoblast cell columns per growing placental villous tip is increased by culture in hypoxic conditions. Figure 4 shows graphically the ranked ΔCT medians were transformed to show IGF-II mRNA expression for each treatment group relative to the median for the calibrator (control) group for the hypothesis being tested, given as 100%.

Thus exposure of extravillous explants to hypoxic culture conditions increases the number of cytotrophoblast cell columns and increases extravillous cytotrophoblast invasive behaviour. In addition, culture in 1% oxygen induces a 7-fold increase in IGF-ll mRNA synthesis in placentas of 7-8.5 weeks which will greatly increase their capacity for invasion of the uterine decidua and subsequent placental differentiation. IGF-II mRNA was positively correlated with the proportion of outgrowing villous tips that formed cytotrophoblast cell columns (r=-0.296, n=65, p=0.01 ).

Treatment with IGF- II thus promotes and maintains the invasive state of cytotrophoblast cells by inhibiting their fusion due to preventing activation of latent TGFβ by binding to CIM6P receptors of cytotrophoblast cells. Alternatively, culture in low oxygen conditions (less than 3%) stimulates IGF-ll production, thus adjusting the level of IGF-ll in the cellular environment to thereby maintain the mesenchymal migratory phenotype.

Example 4 IGF-II enhances placental development in vivo

Enhanced invasion of cytotrophobiasts during pregnancy results in a more functional placenta because the circumferential expansion of the placenta as it grows, as well as acquisition of an adequate utero-placental blood supply, depend on trophoblast invasion of the decidua, It has previously been shown that in pregnant guinea pigs the concentration of IGF-II in maternal blood during pregnancy is a predictor of the structural development of the placenta, parameters of which correlate with placental function and fetal weight and that this is a particularly strong association in early to mid- gestation when the placenta is in its hyperpiastϊc growth phase (Roberts et al., 2001). Therefore we also show herein the effect of treatment of the pregnant mother with exogenous IGF-II on placental growth and development.

We obtained 30 eight week old C57BI/6 female mice from the University of Adelaide Medical School Animal House and divided them into three groups of similar weight. Females were mated with 9-10 week Balb/C male mice and checked for the presence of a vaginal copulatory plug the next morning. The day of the copulatory was designated as day 1 of pregnancy. At midday on day 2 of pregnancy mice were treated with either 0, 12.5 or 25μg/day IGF-II, which translates to 0, 0.5 or 1.0mg/kg/day IGF-ll based on weight at mating, in 0.1 mmol/L acetic acid delivered by a subcutaneous osmotic minipurnp (Alzet 1007D) from days 2-10 of pregnancy. Osmotic minipumps were inserted subcutaπeously on the back under anaesthesia and strict aseptic conditions. Anaesthesia was induced with 3% halothane in oxygen using an anaesthetic machine and maintained with 2% halothane. The tower back was shaved and swabbed and a 1.0cm transverse incision was made. An artery clamp was used to make a long subcutaneous pocket higher up on the back such that when the minipurnp was inserted it lay well away from the incision which was then closed with 2 wound clips. Mice were given one or two puffs of pure oxygen after which they regained consciousness within one minute. Pumps were not primed so as to delay the initiation of pumping by 4-6 hours. Osmotic minipumps delivered 0, 12.5 or 25μg/day IGF-II, at a flow rate of 0.52μl/h. Mice were killed by ether overdose on day 18 of pregnancy (term =19 days). Blood was taken in heparinized tubes and centrifuged. Plasma was frozen at -20"C. Placentas and fetuses were excised from the uterus and weighed. The mother's carcass minus the uterus and minipurnp was weighed.

The results, as depicted in figures 5- 3 indicate that treatment of pregnant mice between days 2 and 10 of pregnancy increased placental weight at day 18 of gestation by 9.6% (ρ<0.05) and 7.5 % (ρ<0.05) by treatment with.12.5 or 25μg/day IGF-ll, respectively (figure 5). The distribution of placental weights across litters was skewed to the right by treatment with IGF-II. The percentage of placentas that weighed more than 120mg was 3.1% in control mice, 25.8% in mice treated with the 12.5μg/day IGF-II, (p<0.0001) and 29.8% in mice treated with the 25μg/day IGF-li (p<0.0001) (figure 6).

Thus treatment of the mother with slow release IGF-II during the first half of pregnancy enhances placental growth. Treatment with 25μg/day IGF-ll increased fetal weight by 4.1 % (p<0.05) (figure 7). The distribution of fetal weights across litters was skewed to the right by treatment with IGF-II. The percentage of fetuses weighing more than 1100mg was 27.5% in control mice, 27.4% in mice treated with 12.5μg/day IGF-ll (NS) and 53.6% in mice treated with the 25μg/day IGF-II (p<0.0001) (figure 8).

There was no difference in maternal weight gain or net carcass weight following treatment with IGF-l! during pregnancy. Maternal weight gain was only related to litter size. Maternal net carcass weights were 24.1±0.4g, 24.2±0.4g and 24.9+0.4g for control, 0.5 mg/kg/day IGF-ll and l .0mg/kg/day IGF-ll (p=0,262), respectively. This suggests that treatment with IGF-II at these doses does not increase cell proliferation in the mother.

In a second replicate of 30 pregnant mice, treatments were identical to the first repiicate except that the osmotic minipumps used (Alzet 1002) delivered the growth factor or vehicle at 0.22μl/h from days 2-18 of pregnancy when the animals were killed and fetal and placental weights recorded. The daily dose delivered was identical to that of the first replicate, either 0, 12.5 or 25μg/day IGF-II, which translates to 0, 0.5 or 1.0mg/kg/day IGF-II based on weight at mating.

Treatment of pregnant mice with IGF-ll between days 2 and 18 of pregnancy reduced placental weight in the high dose group by 4.8% (p=0.001) (figure 9), increased the thickness of the placenta by 5.6% (p=0.026) (figure 10). The distribution of fetal weights across litters was altered by treatment of the mother with IGF-ll with a significant skew to the right in the 25μg/day IGF-ll group (p<0.0001) (figure 11). The distribution of placental weights across litters was also altered by treatment with IGF-ll with a significant skew to the right in both treatment groups (p<0.0001) (Figure 12)

in addition treatment of pregnant mice with 25μg/day IGF-II increased the ratio of fetal weight to placental weight by 6.9% (p=0.026) (figure. 13). This is an indicator of placental functional capacity. A higher ratio suggests that more grams of fetus are produced for every gram of placenta. Thus the placenta is more efficient at extracting nutrients from the mother and transporting them to the fetus. Thus we have shown that treatment with IGF-II of the pregnant mouse improves placenta! and fetal development,

It will be appreciated that the mice used in this experiment had normal placenta! IGF-ll production. It is known that when the iGF-ll gene is ablated in mice there is a considerable reduction in growth of the placenta and the fetus. It could be expected in embryos and/or their parents diagnosed with lower than normal capacity for IGF-ll synthesis, the restoration of IGF-ll to normal levels would have a still greater effect on placental development and hence fetal growth than is observed herein.

There was no difference in maternal weight gain or net carcass weight following treatment with lGF-il during pregnancy. Maternal weight gain was only related to litter size. Maternal net carcass weights were 24.1±0.4g, 24.2±0.4g and 24.9±0.4g for control, 0.5 mg/kg/day IGF-ll and 1.0mg/kg/day IGF-II (p=0.262), respectively. This suggests that treatment with IGF-ll throughout pregnancy at these doses does not increase cell proliferation in the mother.

The regulation of cytotrophoblast behaviour by IGF-II provides support for the notion that control of the cell exposure to IGF-II and latent TGFβ will affect stem cell (embryonic and adult) differentiation referred to hereinabove.

The invention has been described by way of example. The examples are not, however, to be taken as limiting the scope of the invention in any way. Modifications and variations of the invention such as would be apparent to a skilled addressee are deemed to be within the scope of the invention.

Publications Cited

Bennett ST & Todd JA (1996) Human type I diabetes and the insulin gene: principles of mapping polygenes. Annu Rev Genetics 30:343-370.

Berrniπgham J, Jenkins D, McCarthy T, O'Brien M (2000) Genetic analysis of insulin- like growth factor II and HLA-G in pre-eclampsia. Biochθm Soc Trans 28:215-9.

Caniggia 1, Mostachfi H, Winter J, Gassmanπ M, Lye SJ, Kuliszewski & Post M (2000) Hypoxia-inducible factor-1 mediates the biological effects of oxygen on human trophoblast differentiation through TGFbeta(3). J Cliπ Invest 105, 577-587,

Djurovic S, Schjetlein R, Wisloff F, Haugen G, Husby H, Berg K (1997) Plasma concentrations of Lp(a) lipoprotein and TGF-beta1 are altered in preeclampsia. Clin Genet 52:371 -376

Dommisse J & Tiltman AJ (1992) Placental bed biopsies in placental abruption. BrJ Obstet Gynaecol 99:651 -4

Feldser D, Agani F, Iyer NV, Pak B, Ferreira G, Semenza GL (1999) Reciprocal positive regulation of hypoxia-inducible factor 1 π and insulin-ϋke growth factor 2. Cancer Research 59:3915-3918.

Floridon C, Nielsen O, Holund B, Sunde L, Westergaard JG, Thorns SG & Teisner B (1999) Localization and significance of urokinase plasminogen activator and its receptor in placental tissue from intrauterine, ectopic and molar pregnancies. Placenta 20: 711-721

Francis GL, McNeil K, Wallace JC, Ballard FJ & Owens PC (1989) Sheep insulin-like growth factors I and II: sequences, activities and assays. Endocrinology 124:1173-1183

Gargosky SE, Moyse KJ, Walton PE, Owens JA, Wallace JC, Robinson JS, Owens PC 1990 Circulating levels of insuliπ-like growth factors increase and molecular forms of their serum binding porteins change with human pregnancy. Biochemical and Biophysical Research Communications 170:1157-1163

Genbacev O, Zhou Y, Ludlow JW & Fisher SJ (1997) Regulation of human placental development by oxygen tension. Science 277, 1669-1672.

Godar S, Horesji V, Weidle UH, Binder BR, Hans ann C & Stockinger H (1999) M6P/IGFII-receptor complexes urokinase receptor and plasminogen for activation of transforming growth factor betal . EurJ /tn.'Tπ/rco/ 29: 1004-1013

Graham CH & Lala PK (1991) Mechanism of control of trophoblast invasion in situ. J Cell Physiol 148:228-234

Guidice LC, Mark SP & irwin JC (1998) Paracrine actions of insulin-like growth factors and IGF binding protein-1 in non-pregnant human endometrium and at the decidual- trophoblast interface. J Reprod Immunol 39, 133-148.

Hamilton GS, Lysiak JJ, Han VKM & Lala PK (1998) Autocrine-paracrine regulation of human trophoblast invasiveness by insuliπ-like growth factor (IGF)-ll and iGF-binding protein (IGFBP)-1. Exp Cell RΘS 244:147-156

Huang LE, Arany Z, Livingston DM ^'& Bunn HF (1996) Activation of hypoxia-inducible transcription factor depends primarily upon redox-sensitive stabilization of its alpha subunit. J Biol Chem 271 , 32253-32259.

Irwin JC, Suen LF, Martina NA, Matk SP & Guidice LC (1999) Role of the IGF system in trophoblast invasion and pre-eclampsia. Human Reprod 14:90-96 Khong TY, De Wolf F, Robertson WB & Broseπs I (1986) Inadequate maternal vascular response to placentation in pregnancies complicated by pre-eclampsia and by small- for-gestational age infants. Br J Obstet Gynaecol 93: 1049-1059

Khong TY, Liddell HS, Robertson WB ( 987) Defective haemochorial placentation as a cause of miscarriage: a preliminary study. Br J Obstet Gynaecol 94(7):649-55.

Lala PK & Graham CH (1990) Mechanisms of trophoblast invasiveness and their control: the role of proteases and protease inhibitors. Cancer Metastasis flev9(4):369- 79

Morrish DW, Dakour J & Li H (1998) Functional regulation of human trophoblast differentiation. J Rβprod Immunol 39:179-195

Odell SD & Day INM (1998) Molecules in focus: insulin-like growth factor II. IntJ Biochem Cell Biol 30:767-771

Oπg KK, Phillips Dl, Fall C, Poulton J, Bennett ST, Golding J, Todd JA, Duπger DB (1999) The insulin gene VNTR, type 2 diabetes and birth weight. Nat Genet 21 :262-3.

Owens PC, Brinsmead MW, Waters MJ & Thorburπ GD (1980) Ontogenic changes in multiplication-stimulating activity binding to tissues and serum somatomedin-like receptor activity in the ovine fetus. Biochem Biophys Res Commun 96:1812-1820

Owens PC, Waters MJ, Thorbum GD & Brinsmead MW (1985) Insulin-like growth factor receptor in fetal lamb liver: characterization and developmental changes, Endocrinology 117:982-92

Pijnenborg R, Bland JM, Robertson WB & Brosens I (1983) Uteroplacental arterial changes related to interstitial trophoblast migration in early human pregnancy. Placenta 4:397-414

Rebourcet R, de Ceuninck F, Deborde S, Willeput J & Ferre F (1998) Differential distribution of binding sites for ¹²⁵l-insulin-like growth factor II on trophoblast membranes in human term placenta. Biol Rβprod 58:37-44. Schilling B & Yeh J (2000) Transforming growth factor beta(1), -beta(2), -beta(3) and their type I and II receptors in human term placenta. Gynecol Obstet Invest 50:19-23.

Wang GL, Jiang BH, Rueand EA & Semenza GL (1995) Hypoxia-inducible factor 1 is a basic-helix-loop-heϋx-PAS heterodimer regulated by cellular 0₂ tension. Proc NatlAcad Sci USA 92, 5510-5514.

Zelzer E, Levy Y, Kahana C, Shilo BZ, Rubinstein M, Cohen B (1998) Insulin induces transcription of target genes through the hypoxia-inducible factor HIF-1 alpha/ARNT. EtT7iJθ J l7(17):5085-94.

Claims (18)

1. A method of regulating cytotrophoblast and stem cell differentiation and migration characterized by adjusting levels of insulin-like growth factor II (IGF-II) available for binding to the cation-independent mannose-6-phosphate (CIM6P) receptor.

2. A method according to claiml characterized in that treatment of a subject in need thereof with IGF-II promotes implantation of the cytotrophoblast and thereby promotes placentation.

3. A method according to claim 2 characterized promoting invasive and migratory ceil behaviour by exposing cytotrophoblast or sterri cells (embryonic or adult) to elevated levels of IQF-H, such that the cell CIM6P receptors are unable to bind latent TGFβ.

4. A method according to claiml characterized in that treatment of embryos with IGF-ll promotes conversion of trophectoderm cells into cytotrophoblast cells thereby promoting the implantation of embryos into the uterine decidual endometrium, thereby increasing the success of formation of a viabie placenta.

5. A method according to claiml characterized in that IGF-II is added to embryos produced by in vitro fertilisation.

6. A method according to claiml characterized in that IGF-II is administered to a pregnant female subject in need thereof.

7. A method of according to claiml characterized in that that IGF-II is administered to embryonic stem cells, or adult stem cells, in an hypoxic environment to thereby promote differentiation into mesodermal/mesenchymal cell types, including haematopoietic precursor cells and muscle precursor cells, or cardiac muscle cells.

8. A method according to claim 1 , characterized in that there is provided a method of inhibiting stem cell division, stem cell migration by exposing said stem cell to reduced levels of IGF-ll, such that the stem cell C1M6P receptors are able to bind latent TGFβ and thereby promote the activation of TGFβ.

9. A method according to claiml characterized in that there is provided a method of promoting terminal differentiation of stem cells by exposing said stem cell to reduced levels of IGF-ll, such that the stem cell CIM6P receptors are able to bind latent TGFβ and thereby promote the activation of TGFβ.

10. A method of regulating differentiation and migration according to claiml, characterized in that embryonic stem cells or adult stem cells are exposed to inhibitors of IGF-II action on CIM6P receptor such that the stem cell C1M6P receptors are able to bind latent to thereby promote their differentiation into epithelial cell types, including neurones or their precursors.

11. A method according to claiml 0 characterized in that said inhibitors of IGF-II action on CIM6P receptor include soluble CIM6P receptors or fragments thereof, specific antibodies or specific oligonucleotide sequences.

12. A method of regulating cytotrophoblast and stem cell differentiation and migration characterized according to any one of the preceding claims wherein adjusting levels of insulin-iike growth factor II (IGF-ll) available for binding to the cation-independent manπose-6-phosphate (CIM6P) receptor, is achieved by regulation of levels of naturally occurring precursors of and isomers of IGF- II.

13. A method of regulating cytotrophoblast and stem cell differentiation and migration characterized according to any one of the preceding claims wherein adjusting levels of insulin-like growth factor II (IGF-II) available for binding to the cation-independent mannose-6-phosphate (CIM6P) receptor, is achieved by use synthetic analogues of iGF-ll that have altered ability to bind to type-1 IGF receptors, altered ability to bind to insulin receptors, altered ability to bind to

IGF-binding proteins and increased ability to bind to CIM6P receptors.

14. A method according to claiml characterized in that the cells are mammalian ceils including human, horse, cow, pig, goat or sheep cells.

15. A method of determining the ability of cytotrophoblast cells to differentiate and migrate characterized by determining in the mother, father or embryo the sequence of nucleotides in the deoxyribonucleic acid near the insuliπ-like growth factor I! gene known as the insulin (INS) variable number of tandem repeats (VNTR), to thereby determine the capacity of the cytotrophoblast to synthesize IGF-II and thus the capacity of the cytotrophoblast to migrate into the uterine 5 decidua and the capacity of the placenta to transport substrates to the embryo/fetus.

16. A method of determining the ability of cytotrophoblast to differentiate and migrate, characterized by measurement of the amount of messenger ribonucleio acid transcribed from the insuliπ-like growth factor II gene in embryos.

10 17. A method of determining the ability of cytotrophoblast to differentiate and migrate characterized by measurement of the amount of insuliπ-like growth factor II protein secreted by mammalian embryos.

18. A method of determining the ability of cytotrophoblast to differentiate and migrate characterized by measurement of the amount of insulin-like growth Ϊ5 factor II protein circulating in maternal and paternal blood.

0