AFS for prenatal and perinatal intervention. AFS collected from the amniotic fluid are expanded and either directly administered, or incorporated into scaffolds for tissue engineering, and subsequently transplanted, or stored for future use (LN2: liquid nitrogen). They could be used for treating birth defects in the fetus (orange) or neonate (not shown) from which they were derived (autologous transplantation), or for treating a different fetus (green) or neonate (not shown, allogeneic transplantation). Some of the tissues/structures that may be repaired in the fetus or neonate using AFS cells are indicated
Crucially, AFS cells display high proliferative activity and plasticity typical of fetal cells and can therefore be rapidly expanded to obtain sufficient number of cells for therapeutic purposes. As we will discuss later, they appear to maintain their differentiation potential in a variety of scaffolds that are needed to provide the required tissue shape and size according to the anatomical defect to be managed. Furthermore, AFS cells appear to lack tumorigenicity and have low immunogenicity following transplantation into adult animals . Hence they could be a valuable stem cell source both for autologous and allogeneic transplantation. It has been reported that cryopreservation of AFC cells does not affect their stemness and biological activity [25, 26]. Although the clinical and economical value of long-term stem cell banking is still a matter of debate, and much attention should be paid to stringent quality control for banking, the first private amniotic stem cell bank was opened by Biocell Center in October 2009 in the USA (www.biocellcenter.com/).
In summary AFS cells can offer several advantages for inducing tissue repair and tissue engineering prenatally and perinatally in comparison to other stem cell sources that include:
No major ethical issues attached
Ease of harvesting cells from the amniotic fluid
Rapid expansion time
Recently, the long-term safety and efficacy of tissues engineered using AFS cells has been validated, paving the way for clinical trials . Altogether, AFS cells seem on course to become one of the cell sources of choice for cell-based therapies in the fetus and perinatally.
4 AFS Cell Behavior in a Prenatal Environment
As AFS cells can be expanded relatively rapidly, autologous AFS cells can in principle be easily available for the treatment of a range of defects identified during routine antenatal screening. Furthermore, given the immunotolerance of the developing fetus and AFS cell low immunogenicity, allogeneic AFS cells might also be used. The pros and cons of the two approaches have yet to be fully evaluated, and they may differ depending on the pathology to be tackled and the stage of development at which medical intervention is deemed appropriate.
We have recently investigated the effect of AFS cell transplantation in an easily accessible embryonic injury model, the chick (Fig. 11.2a) . In this severe thoracic crush injury model that results in extensive damage of the early developing spinal cord and surrounding mesenchymal tissues, most embryos die within 24 h. Grafting AFS cells into the spinal cord at the time of injury significantly reduced tissue damage and increased cell survival assessed over 5 days. The effect of grafting different cell types or using a pharmacological treatment at the time of injury was also investigated. No significant difference in survival was observed in injured embryos grafted with either 3T3-NIH fibroblasts or C17.2 neural stem cells, or treated with the anti-haemorrhage drug, desmopressin. Therefore, the rescue response observed appeared to be specific to AFS cells.
Tissue damage following injury is reduced by AFS via a paracrine effect. (a) Upper panel: schematic representation of embryonic chick injury model. GFP-positive AFS cells (green) are injected at the injury site. Lower panel: 48 h after injury the embryo without AFS cells has died (left), whereas the one injected with AFS cells (right), is still alive and development has progressed; the insert from the region indicated by the box shows a fluorescent image of the GFP-labelled AFS grafted cells. (b) Upper panel: schematic representation of the co-culture system. E15 chick spinal cord organotypic slices in the upper chamber are cultured with or without AFS cells in the lower chamber. Lower panel: analysis of cell death using propidium iodide shows extensive cell death in a slice cultured without AFS cells (left), but not in the one maintained in their presence (right) 
The extensive reduction in tissue damage observed in live embryos grafted with AFS cells at 24 h after surgery could be due to integration and differentiation of grafted AFS cells at the injury site, cell–cell interaction between AFS and host cells, that would allow to fill the gap created by the injury, or by trophic effects. These mechanisms are not mutually exclusive. While there was clearly tissue sparing after injury in embryos injected with AFS cells, no evidence of neural differentiation was observed at the injury site . Significantly, however, there was a clear relationship between embryo survival over time and detection of grafted AFS cells. Evidence that tissue sparing and embryo survival was due to secreted factors was demonstrated in vitro. These experiments showed great reduction in cell death and a parallel increase in the presence of healthy neurons in organotypic spinal cords co-cultured, but not in direct contact, with AFS cells. This is consistent with a paracrine mechanism (Fig. 11.2b). A key paracrine role for AFS cells has been proposed also for the closure and re-epithelialization of skin wounds in mice .
In addition to increasing survival within the injured tissues, AFS cells may boost endogenous repair mechanisms that are known to occur spontaneously via the recruitment of endogenous stem cells in the injured spinal cord of lower vertebrates . Indeed, scaffolds containing human AFS cells, but not empty ones, implanted subcutaneously in nude mice have been reported to recruit host cells to the scaffold .
Damage to the fetal nervous system caused by a variety of insults, such as oxygen deprivation, can be greatly increased by concomitant infection and consequent inflammatory responses . Given the early stages of development used, it is unlikely that the AFS cells-dependent rescue observed in the chick spinal cord injury model is mainly due to an anti-inflammatory response. However, in later models of neural injury, AFS cells may provide additional significant beneficial effects because of their anti-inflammatory properties. For example, they may also help to reduce injury to the fetal brain associated with pathological exposure to maternal antigens [33, 34]. An anti-inflammatory role for AFS cells has been recently supported in perinatal and adult models of disease including inflammatory bowel diseases and vasculopathies that occur as a consequence of organ transplantation [35, 36].
The ability of AFS cells to reduce neural damage and stimulate epithelialization could also be of great value in developing therapies for spina bifida. This defect can be detected at early stages of gestation, when neural damage is believed to be still limited. Early covering of the exposed neural tissue using AFS cell-based constructs could have the dual advantage of being neuroprotective and stimulate epithelialization over the exposed spinal cords, as suggested in wound healing models [29, 37]. In a recent study AFS cells were injected in the amniotic fluid of mouse fetuses that had been treated with retinoic acid to induce neural tube defects mimicking spina bifida. AFS cell-derived neural cells were detected on the surface of the exposed neural tissues in most of the embryos examined . Therefore, for some fetal abnormalities intra-amniotic delivery of AFS cells could represent a convenient administration route worth additional investigation.
Molecules secreted by AFS cells have been proposed to be responsible for tissue repair also in non-neural tissues, such as following heart tissue ischemia , lung moderate hyperoxia , and lung hypoplasia . A paracrine effect may also underlie increased proliferation observed when collagen plugs with platelets and AFS cells are used, as compared to plugs without AFS cells, to seal an iatrogenic membrane defect in a fetal rabbit model . However, which of the factors secreted by AFS cells, that are likely to include cytokines and neurotrophic factors, are “protective/repair inducing” in different disease models has yet to be established.
Whatever the specific molecular mechanisms underlying the efficacy of AFS cells may be, in order for them to be useful therapeutic tools, they need to be able to home to the desired organ, ideally following simple administration protocols. As discussed above, in some cases intra-amniotic delivery might provide a suitable administration route . Recent studies have investigated the distribution of autologous AFS cells injected intraperitoneally in the fetal sheep. The injected GFP-labelled AFS cells were detected in several organs, including liver, heart, placenta, membrane, umbilical cord, adrenal gland, and muscle . Furthermore, injection of AFS cells in mouse heart ventricles within 24 h after birth, resulted in engrafting of the injected cells within the heart, airways, and lungs .
Studies in diseased fetuses and newborns to establish whether there is effective homing to the damaged organ of interest and a therapeutic effect following injection of AFS cells have yet to be carried out. Nonetheless, there is a least proof of principle that AFS cells injected in utero in the sheep model can reach a variety of tissues, and that intra-amniotic delivery and intraventricular and intraperitoneal injection perinatally also result in some AFS cell engraftment as indicated in Tables 11.1 and 11.2 [35, 38, 44, 45, 49].
Examples of studies where AFS cells (AFSC) were transplanted prenatally
Injury to spinal cord and surrounding tissues
Injection at the injury site
Decreased tissue damage and increased embryo survival
Neural tube defect
Undifferentiated AFSC homed to the neural placode
Widespread distribution of grafted cells
Heart valve defect
Construct implantation (synthetic material)
Intact functional valves/absence of thrombus formation
Circumferential tracheal defect
Construct implantation (synthetic material)
Remodeling to fibrous cartilage pattern
Circumferential tracheal defect
Construct implantation (decellularized rabbit trachea)
Enhanced remodeling, epithelialization and growth
Damaged fetal membrane
Construct implantation (synthetic material)
Increased local cell proliferation
Examples of studies where AFS cells (AFSC) were transplanted perinatally
Injection in heart ventricle
Engraftment in heart, lung, and airways
Migration, homing and integration into various organs
Reduced inflammation, improved function
Construct implantation (collagen hydrogel)
Improved tensile strength/structure decreased diaphragmatic hernia recurrence
Diploic nasal bone defect
Construct implantation (synthetic material)
5 AFS Cell-Based Tissue Engineering for the Management of Birth Defects
AFS cells are amenable to be used for tissue engineering Kaviani et al., . Postnatal transplantation of construct containing AFS cells for the management of various birth defects has been suggested to be of potential clinical relevance in the few studies currently available in animal models (Table 11.2). Examples of birth defects that could benefit from the use of AFS cell-based implants to repair or reconstruct the abnormal tissue for which some experimental evidence is available include enteric, cardiovascular, diaphragmatic, and skeletal defects.
5.1 Necrotising Enterocolitis
Necrotising enterocolitis is the most common gastrointestinal surgical emergency occurring in neonates, with high mortality rates ranging from 15 % to 30 %. However, there is no effective therapy for NEC, and surgery remains the treatment of choice for necrotic bowels . The most widely accepted hypothesis is that enteral feeding concomitant with intestinal hypoxia-ischemia-reperfusion and pathogen colonization stimulates an inappropriate inflammatory response by the immature intestinal epithelial cells . There is a growing body of evidence that stem cells can play a therapeutic role in inflammatory bowel diseases . Significantly, AFS cells showed therapeutic effects in a well-established neonatal rat model of necrotising enterocolitis. When grafted in the bowel wall of a diseased newborn rat, they could integrate, decrease apoptosis and bowel inflammation, increase enterocyte proliferation, and improve intestinal function and survival . Also in this model, paracrine effects on endogenous cells, rather than significant repopulation of the diseased gut, appear to underlie the beneficial effects of the grafted AFS cells.
5.2 Craniofacial Defects
Craniofacial defects include defects of skull bones (e.g., premature suture fusion), facial cartilages (e.g., incomplete, microtia, or absent, anotia, ears), palate, and soft tissues (e.g., cleft lip and palate). AFS cells have been shown to stably differentiate into all the major cell types present in craniofacial tissues, such as bone, cartilage, muscle, and blood vessels [17, 56, 57]. Notwithstanding the relative frequency of craniofacial birth defects and the skeletogenic potential of AFS cells, information on the potential use for repair of these tissues perinatally is still lacking. Allogeneic AFS cells mixed with electrospun biodegradable poly-l-lactic acid nanofibers were tested for their ability to repair full thickness diploic nasal defects in young adult rabbits, as grafting a bioengineered construct in neonates seemed to be difficult in this species . Histological examination of AFS-cellularized constructs showed evidence of bone formation with levels of extracellular calcium significantly higher in these constructs than in the acellular implants. Although some areas of AFS-constructs displayed fairly normal bone architecture, this was neither homogeneous nor consistently observed in all cellularized constructs. Nonetheless, the overall extent of bone formation was significantly higher in these constructs than in the acellular implants. This suggests that further studies on the use of AFS cells for perinatal nasal defect repair are worth pursuing, possibly selecting different scaffolds and using different species for the validation of this therapeutic approach.
5.3 Tracheal Defects
Abnormalities of the tracheal cartilage can cause narrowing of the airway, a birth defect known as congenital tracheal stenosis. This results in breathing problems that can be very severe and difficult to resolve, particularly when several segments of tracheal cartilage are defective. Successful replacement of defective airway segments with decellularized donor cartilage seeded with stem cells was first achieved in an adult female patient , and more recently in a paediatric patient . Comparison of AFS cells and other mesenchymal stem cells for cartilage tissue engineering in vitro showed an overall higher content of extracellular matrix proteins in cartilage derived from AFS cells .
Approaches to fetal cartilage engineering have involved the use of AFS cells combined either with synthetic scaffolds or decellularized donor scaffolds [47, 48]. In both cases, constructs containing AFS cells were shown to be superior to acellular scaffolds for airway reconstruction. The constructs consisting of decellularized rabbit trachea seeded with AFS induced to differentiate into cartilage and implanted in fetal sheep were of particular interest . Although some degrees of stenosis were reported in both the AFS-containing and acellular tracheal scaffolds, the extent of epithelialization was much greater in grafts with AFS cells. Significantly, these grafts were found to have grown in size post-implantation. This is a crucial feature for the treatment of a birth defect in a growing organism. In addition, AFS cells were shown to effectively enhance normal fetal wound healing, another highly desirable feature both when performing in utero and perinatal surgery . Altogether, there is some evidence that transplantation of cartilage tissue-engineered using AFS cells could be used to treat congenital tracheal abnormalities.
The tracheal transplantation field has been fast moving, and given the clinical experience rapidly accumulating into the use of cellularized scaffolds together with the encouraging results on the use of AFS cells emerging in this context from animal models, it is conceivable that their clinical use in paediatric patients with tracheal defects might not be too far down the line.