Fetal stem cells are nothing new. Umbilical cord blood haemopoietic stem cells (HSC) have been extensively investigated and widely utilized over the last 10 to 20 years and fetal neural tissue has already been used therapeutically in Parkinson's disease, with some evidence of clinical improvement. Recent interest in stem cell biology and its therapeutic potential has led to the search for fetal stem cells in fetal organs obtained at termination of pregnancy, as well as for accessible sources of fetal stem cells that might be collected for autologous use in ongoing pregnancies. Stem cells obtained from fetal blood and tissues are believed to have similar properties and immunophenotype to comparable adult tissue-derived stem cells, although their development potential is more restricted than pluripotent embryonic stem (ES) cells. Fetal stem cells are described as multipotent, meaning that they exist within specific fetal tissues and give rise to differentiated cells of that tissue only. Although this potential to develop into tissues derived from mesoderm is well established, the ability also to differentiate in vivo into ectoderm-or endoderm-derived tissues is still under investigation. The question of whether the stem cells located in adult tissues are the same ones that are present during fetal life is yet to be answered.

Haemopoeitic stem cells

The phenotype and properties of fetal liver and adult bone marrow HSC have been compared in a number of studies. The two populations show significant differences in both expression and activation of molecules involved in signal transduction and in activation of other haemopoietic cell populations. The expression of HLA class I and class II molecules in HSC increases during ontogeny, from virtually no expression at the yolk sac stage to high expression in adult life. Fetal liver cells have a huge competitive engraftment advantage relative to adult bone marrow, as shown in murine fetal recipients and fetal liver HSC produce much larger numbers and types of progeny after transplantation in mice than adult bone marrow HSC. In the same study, the numbers of progeny produced progressively decreased for ontologically older cells and only transplanted cells from fetal liver were able to repopulate a secondary recipient. For all these reasons, and not withstanding their lower absolute numbers, fetal HSC would seem preferable to adult bone marrow for transplantation in humans.

Mesenchymal stem cells

Non-haemopoietic fetal stem cells are widely accepted as having a higher developmental potential than their counterparts in adult tissues. For example, comparisons of fetal stem cells with those derived from defined developmental niches in the mature brain indicate that NSC plasticity decreases with developmental age. Fetal MSC can be induced to differentiate along osteogenic, chondrogenic and adipogenic lineages, but unlike adult bone-marrow-derived MSC, they appear to differentiate into neurons, skeletal muscle and possibly blood cells as well. In addition, the proliferative capacity of fetal MSC is higher than adult MSC; when expanded in culture the number of fetal MSC increases over 12 times, whereas adult MSC numbers increase three-fold. The prevalence of fetal MSC in bone marrow declines with advancing age; for example one MSC is found among 10 000 mid-trimester fetal bone marrow cells, as compared to one MSC per 250 000 cells in the adult bone marrow. The prevalence of fetal MSC in fetal bone marrow, liver and blood is significantly higher than in adult tissues; fetal MSC make up 0.4% of nucleated cells in first-trimester fetal blood, whereas MSC represent 0.001-0.0001% of nucleated cells in adult bone marrow.

In terms of immunogenicity, MSC from adult tissue express intermediate levels of HLA class I and low levels of HLA class II. Most investigators are agreed that fetal MSC do not express HLA class II antigens but are divided as to whether or not they express HLA class I. In a recent report, fetal MSC did not elicit alloreactive T-cell proliferative responses in lymphocytes derived from adult donors and were also able to suppress lymphocyte proliferation induced by mitogens. Therefore fetal MSC might possess all of the immunomodulatory properties found in adult MSC and, because of their higher proliferative and differentiation potential, might have significant advantages over adult MSC as a source of cells for transplantation and therapy.

Stem cell ageing

The quantitative and qualitative effects of ageing on stem cells are not well understood, although it is believed that stem cells from a younger individual should have greater potential. Thus, many investigators suggest that fetal stem cells should have an advantage over adult stem cells in cell replacement therapies. All stem cells invest heavily in self-protective mechanisms and can self renew, but whether or not they age over the lifetime of an individual is the subject of much debate. On the one hand, stem cells remain viable for the lifespan of a mammal, small numbers can repopulate the whole mammal and many life-ending diseases have mechanisms that do not involve stem cells. On the other hand, most stem cells, including those from fetal and adult tissues, are not immortal and show an increase in apoptotic mechanisms with age. There is some evidence from murine models that the homing efficiency of older HSC is less than that of younger HSC, and adult stem cells are at a competitive disadvantage when transplanted with fetal cells. Older HSC have diminished self-renewal capacity, less developmental potency and give rise to decreased numbers of progeny when subjected to haemopoietic demands, and this decline in function is even more apparent when older stem cells undergo increased stress. Similar qualitative effects of ageing are seen with MSC; older bone marrow stroma blunts haemopoietic responses after transplantation and increases post-transplant autoimmunity.

Most, if not all, stem cells produce telomerase, which lengthens telomeres, protects against genotoxic damage and correlates with cell immortality. The self-renewal and replicative potential of stem cells probably depends on telomerase to maintain stable telomeres, as most evidence indicates that telomere length is a biomarker of the replicative history of cells. Cells of the germline have very long telomeres, which do not shorten with ageing of the organism, and fetal stem cells could be expected to have an advantage over adult stem cells in this regard. Comparative studies of fetal liver and adult bone marrow HSC have confirmed that fetal liver HSC have higher telomerase activity and adult bone-marrow-derived HSC shorter telomeres, which again implies that the proliferative potential of HSC is limited and declines with age.

Potential Applications of Fetal Stem Cells

Stem cells from any source are exciting, both as models of developmental biology and for their promise in treating human disease. Their properties mean it is feasible to contemplate isolating stem cells from the body, expanding them under cell culture conditions, directing their proliferation with growth factors and then transplanting them or their progeny into patients of all ages for clinical gain. In this way stem cells might in future be used to alleviate degenerative disorders, replace diseased or failing tissues with engineered substitutes, and correct genetic disease. HSC have already been proven permanently and clonally to produce all cell types of blood and the immune system in engrafted hosts. MSC might be applied therapeutically for the correction of disorders of mesenchymal origin and further potential therapeutic application is based on their ability to enhance engraftment of HSC after co-transplantation. Whereas fetal stem cells could theoretically be used in place of stem cells derived from adult tissues for most applications, there are some areas where they may have specific advantages.

In utero transplantation

In utero transplantation of allogeneic stem cells is a novel approach to overcome the limitations of postnatal therapy: including severe treatment-associated morbidity and pre-existing organ damage that develops before birth. Stem cell transplantation in early intrauterine life to correct genetic defects has several advantages and using fetal rather than adult stem cells might provide further benefits as fetal cells have a distinct competitive advantage over adult cells. Early establishment of donor stem cell engraftment prenatally might prevent or reduce the pathology associated with the underlying disorder still further. It is therefore important to understand the relative engraftment efficiency of different sources of fetal cells, which probably depends on intrinsic differences in cytokine requirements as well as extrinsic signals that differ in the fetal versus the adult microenvironment.

The initial targets chosen for in utero stem cell transplantation were HSC, not only because of their well-established proliferative and multilineage potential but also because of the years of clinical experience in adult transplantation. To date, most attempts at HSC transplantation in utero have failed because the mid-trimester fetus is immunocompetent and rejects allogeneic cells. This problem might be obviated by earlier transplantation, before the loss of tolerance. Despite advances in cord sampling techniques for early fetal blood collection in ongoing pregnancies, the low yield of HSC in the late first trimester, together with their limited expandability, makes this approach problematic. By contrast, adult MSC engraft widely in animal in utero transplantation models and appear to have unique immunological characteristics, which could allow engraftment irrespective of gestational age and immune competence. When transplanted in utero into fetal sheep, fetal MSC engraft-albeit at low level-into multiple organ compartments and co-transplantation of adult MSC has been shown to enhance engraftment of HSC in utero in animal models, suggesting that co-transplantation of fetal MSC might also result in accelerated haemopoietic engraftment. Fetal liver cells have already been used successfully to treat a fetus with X-linked severe combined immunodeficiency in utero.

Gene therapy

Stem cells have considerable utility as vehicles for gene therapy because they are self-renewing, thus precluding the need for repeated administration of the gene. Ex vivo gene therapy uses autologous HSC that are obtained first from the fetus, transduced in vitro and then transplanted back to the fetus. Results from postnatal gene therapy trials now prove the clinical effectiveness of this approach, although gene therapy is under scrutiny since the recent reports of two children developing leukaemia following ex vivo postnatal gene therapy, presumably as a result of the viral vector disrupting an oncogene. Other concerns are transgene expression in tissues other than the target tissue and inadvertent transfer into germline cells. The biological advantages of targeting fetal MSC over HSC for autologous gene therapy include their higher proliferative capacity, greater transduction efficiency, ready expandability, reduced immunogenicity, ability to engraft and differentiate into most tissue types and ability to be targeted via ligands to specific tissue types. This approach has the potential to cure quite a number of genetic diseases such as the mucopolysaccharidoses, the cerebral gangliosidoses, the leucodystrophies, osteogenesis imperfecta and muscular dystrophy.

Non-invasive prenatal diagnosis

Fetal stem cells cross into maternal blood during pregnancy and thus represent a potential non-invasive source of fetal genetic material for prenatal diagnosis. Research in this area has been hampered by the lack of cell types unique to fetal blood and the low frequency of fetal cells trafficking across the placenta in early pregnancy. The disparate range of cell types that traffic into the maternal bloodstream include HSC and MSC. Fetal haemopoietic progenitors have been demonstrated in the maternal circulation from early gestation onwards; however, they are difficult to distinguish from maternal circulating progenitors and most groups have been unable to expand the fetal cells sufficiently in vitro for clinical application. Fetal MSC, which circulate in first-trimester fetal blood, have been proposed as an alternative target cell for non-invasive prenatal diagnosis because they appear to have no counterpart in adult blood and can be clonally expanded into a pure source of fetal cells. However, although fetal MSC are likely to cross the placenta based both on theoretical considerations and on our group's findings that fetal MSC are detectable in a small proportion of maternal blood samples, they appear to circulate at very low numbers, making any application in the field of non-invasive prenatal diagnosis unlikely.

Microchimerism

The persistence of fetal cells for years in maternal tissues, known as fetomaternal microchimerism, has been implicated in autoimmune disease through graft-versus-host (GVH)-like responses. However, the frequency of fetomaternal microchimerism after normal pregnancy and the cell type responsible is unknown. One explanation for the apparent low frequency of fetal MSC in circulating maternal blood during pregnancy is engraftment in maternal tissues such as bone marrow soon after transplacental passage. Expression of cell adhesion molecules, lack of expression of HLA class II antigens, along with the adherent properties of MSC in vivo, suggests that MSC can disperse widely and implant in connective tissues. Adult bone-marrow-derived MSC readily engraft in most organs in animal models and preferentially home to bone marrow after infusion, whereas fetal MSC engraft diffusely after xenotransplantation in utero. MSC therefore seem the most likely fetal cell type to persist in maternal tissues. We have recently identified male fetal MSC in post-reproductive female bone marrow up to 50 years after pregnancy in a group of women who had sons.