Different protocols have been established for the isolation of hAEC. Current research is directed towards developing isolation protocols to introduce hAEC in clinics for therapeutic purpose [12–14].

The isolation of hAEC begins with peeling off the amniotic membrane from the underlying chorionic membrane, followed by extensive washing in an appropriate buffer [12, 15–17] and subsequent cutting of the membrane into small pieces. The washing step is crucial as residual blood clots tend to reduce the efficiency of the trypsin digestions. The basis for isolation includes a series of trypsin digestions to release the hAEC, filtration using filters of varying pore sizes (70–100 μm), centrifugation and thereafter suspending in an appropriate medium best suited to culture the adherent epithelial cells. The most commonly followed protocol constitutes treating the amnion pieces with trypsin (0.05 %) three times, with incubation intervals lasting 10–40 min, with a reported yield of 80–300 × 10⁶ hAEC from a single term placenta [18].

Various factors such as the trypsin concentration, duration of incubation, as well as the freshness of placenta influence the yield and viability of hAEC post isolation [17, 19]. Single 30 min incubation in trypsin has also been reported, highlighting that a reduction in the incubation time decreases the probability of causing cellular damage due to enzymatic exposure, preserving a greater cellular viability [17]. Therefore, attempts to improve the isolation protocol and to achieve better yield led to modification of trypsin concentrations (ranging 0.05–2.5 %) depending on which incubation periods have also been standardized [16, 17, 19–22]. Alternate to these isolation procedures, hAEC isolation post hAMSC removal from the amnion has also been reported [23, 24]. This is achieved by either manually scraping out the amniotic mesoderm without affecting the epithelial layer [17], or enzymatic removal of hAMSC following which the remaining tissue is subjected to trypsin digestion [23]. Furthermore, use of density gradient (Percoll) centrifugation for enrichment of state specific embryonic antigen (SSEA)-4-positive cell populations from the isolated hAEC prior to culture has been described [18].

hAEC are plastic adherent and on culturing, readily adhere to the culture dish without any feeder layer or any specific pretreatment of the culture substrate [25]. It was observed that hAEC do not proliferate well at low densities but replicate faster for longer duration when cultured in media with low calcium [25]. In addition, when cells are cultured in DMEM supplemented with 10 % serum and epidermal growth factor (EGF), the cells grow for 2–6 passages. Removal of the EGF results in a rapid decrease in proliferation and what appears to be terminal differentiation [26]. Proliferating hAEC display a normal karyotype [11], and have an average doubling time of 38.4 h [27].

Interestingly, it has been reported that after 5 days in culture, two fractions of cells become evident; first a population of adherent cells and second loosely attached cells over the adherent ones [11]. On collecting these two fractions by differential trypsinizations, it was observed that the clusters remaining in culture over the basal layer of adherent cells contained more cells with stem cell characteristics than in the adherent fraction [11].

In terms of morphological appearance, hAEC in culture form a confluent monolayer of cobblestone-shaped epithelial cells [21, 22]. In terms of marker expression, nearly 100 % of hAEC are positive for pan-Cytokeratin (CK) [28], markers generally used to differentiate epithelial from mesenchymal cell lineages. Interestingly, there are reports of hAEC expressing some of the mesenchymal markers, namely vimentin [28] and alpha smooth muscle actin (α-SMA) [29]. Molecular evidence of a bona fide epithelial to mesenchymal transition (EMT) undergone by cultured hAEC (i.e., increased expression of several genes associated with EMT such as snail homologue 1 (Drosophila) protein, matrix metalloproteinase (MMP)-9, plasminogen activator inhibitor 1 and α-SMA) was also provided, possibly due to hAEC autocrine production of transforming growth factor-β (TGF-β) [29]. In regard to the expression of other markers, freshly isolated/naive hAEC are not homogenously positive for all the markers analyzed. Reportedly, hAEC present with markers typically expressed by human embryonic stem cells, namely, SSEA-3, SSEA-4, tumor rejection antigen (TRA)1-60 and TRA1-81, but lack the expression of SSEA-1 [4]. Other markers expressed include crypto, FRL-1, criptic family 1, developmental pluripotency-associated protein 3, prominin 1, paired box gene (PAX)-6 [21] and GCTM2 [30], while expression of forkhead box D3, growth differentiation factor-3, and telomerase reverse transcriptase are not observed [22, 31]. Along with these markers, molecular markers of pluripotent stem cells like octamer-binding protein (OCT)-4, SRY (sex determining region Y)-box 2 (SOX-2), Nanog, Lefty-A, fibroblast growth factor (FGF)-4, reduced expression protein-1 (REX-1) and teratocarcinoma-derived growth factor-1 [32] are also observed. Among these molecular markers, OCT-4 which is one of the transcription factors that plays a critical role in maintaining pluripotency and self-renewal in undifferentiated cells, is found to be expressed in majority of hAEC [28].

Expression of mesenchymal and hematopoietic markers such as human leukocyte antigen (HLA)-A, HLA-B, and HLA-C, cluster of differentiation/designation (CD)10, CD13, CD29, CD44, CD49e, CD73, CD90 (Thy-1), CD105, CD117 (c-kit), CD166 and stromal cell surface marker (STRO)-1 are reported; however HLA-DQ, HLA-DR, CD14, CD34, CD45, CD49d are found to be negative [4, 22, 33]. Low expression of HLA-A, HLA-B, and HLA-C was reported immediately after isolation, which surprisingly increased to significant levels in culture [34]. Similarly, the expression of CD90 is reported to be low initially, but after 6 days of culture, approximately 50 % of the cells express CD90 [28].

In addition, hAEC show expression of CD9, E-cadherin (CD324), integrin α₆, integrin β₁, CD24, c-met and ATP-binding cassette transporter G2 (ABCG2/BCRP) [35]. The expression of ABCG2 in hAEC indicates that they may have some properties similar to the so-called side population of hematopoietic stem cells found in bone marrow [28].

Along with these markers, undifferentiated hAEC also show expression of neural (glutamic acid decarboxylase, myelin basic protein, neurofilament medium chain, neuron-specific enolase, 2′,3′-cyclic nucleotide 3′-phosphodiesterase, proteolipid protein/DM-20, microtubule-associated protein 2 (MAP2), MAP2 kinase, glial fibrillary acidic protein (GFAP), neurofilament proteins), lung (NK2 homeobox1, mucin, occludin, aquaporin-5, caveolin-1), hepatic (albumin, α-fetoprotein (FP), α-1 antitrypsin (AT), CK18, glutamine synthetase, carbamoyl phosphate synthetase-1, phosphoenolpyruvate carboxykinase, cytochrome (CYP)2D6, CYP3A4, CYP2C9, transthyretin, tyrosine aminotransferase, hepatic nuclear factor 3-α, CCAAT/enhancer binding protein-α), cardiomyogenic (Gata binding protein (GATA)-4, Nk2 homeobox5 (NKx 2.5), myosin light chain (MLC)-2A, MLC-2V, myosin regulatory light chain-7, atrial natriuretic peptide (ANP), calcium channel, voltage-dependent, L type, alpha 1C subunit, potassium voltage-gated channel, Shal-related family, member 3), and pancreatic (pancreatic and duodenal homeobox 1 or PDX-1) lineage associated markers [22]. Some differences in the percentage of cells expressing the reported markers have been observed, possibly depending on different gestational age [5], and variable culture conditions used by different groups and expansion in vitro [13, 36, 37]. Culturing of hAEC in a serum free medium led to the expression of hematopoietic and monocytic markers, high telomerase activity and elongated telomeres whereas hAEC normally cultured in media supplemented with fetal calf serum lacked these markers and also showed no telomerase activity [22].

It is evident that a standardization of isolation as well as culture protocols is necessary for obtaining uniformity in the cell populations, to guarantee accurate data interpretation of the experiments using these cell populations and enabling reliable comparison between results obtained from different research groups.

Since the discovery of hAEC, several groups have been keenly interested in exploring their differentiation potential with the hope of using them in clinical/translational medicine. Pioneering studies by Sakuragawa and colleagues showed that hAEC express neuronal and glial markers such as neurofilament and MAP2, or GFAP [38], oligodendrocyte markers [39], synthesize catecholamines from l-tyrosine [40] and have the ability to convert 3,4-dihydroxyphenylalanine (l-DOPA) into dopamine [41], suggesting therefore their potential use for the treatment of neurological disorders. Thereafter, the potential of hAEC to differentiate towards the neural (ectodermal) lineage has been investigated, both in vitro and in vivo.

hAEC cultured in a medium containing supplements, including –trans retinoic acid and FGF-4, differentiated in vitro towards glial- and neuronal-like cells [11, 21]. Niknejad and colleagues reported that the capability of hAEC to express neural cell markers upon induction of neural differentiation in vitro is affected by some factors including serum, noggin, basic-FGF and retinoic acid [19]. The neural and neurotrophic potential of hAEC under in vivo conditions have shown that they can facilitate neuroregeneration/repair in disorders like Parkinson’s disease, stroke, and spinal cord injury. For instance, hAEC grafted into the dopamine-denervated striatum of a rat immunosuppressed model of Parkinson’s disease survived and improved the neurobehavioral deficit [42]. Liu et al. transfected hAEC, previously infected with a recombinant lentivirus to express the glial cell line-derived neurotrophic factor, into the brains of rats with a transient middle cerebral artery occlusion, and observed that hAEC survive and migrate to the ischemic area of rats, significantly ameliorating behavioral dysfunction and reducing infarct volume [43]. In animal models with spinal cord injury the administration of hAEC promoted hind limb motor function recovery; the atrophy was ameliorated and the size of injured neurons partially restored [44]. Recent preliminary findings indicate the feasibility to potentially use hAEC to treat multiple sclerosis. Indeed, transplantation of hAEC in an experimental autoimmune encephalomyelitis (EAE) mouse model of multiple sclerosis ameliorated EAE and reduced infiltration of T lymphocytes and monocyte/macrophages, and demyelination within the central nervous system [45]. Similar findings, as well as decreased central nervous system inflammation, demyelination, and axonal degeneration in the mouse spinal cord and brain were reported on intraperitoneal injection of hAEC in a mouse model of multiple sclerosis [46]. Reportedly, the conditioned medium derived from cultured hAEC exhibits neurotrophic effects on rat cortical cells [47].

The differentiation potential of hAEC towards cells of the endodermal lineage has also been investigated. In terms of hepatic differentiation, after Sakuragawa and colleagues [48] first demonstrated that albumin and α-FP-producing hAEC are promising transgene carriers for allogeneic transplantation into liver, other works have encouraged the use of placental cells for restoring functionality of hepatic tissues. Takashima and colleagues reported that cultured hAEC, even in the absence of hepatic differentiating stimuli, express several hepatocyte-related genes and demonstrated albumin production, glycogen storage, and albumin secretion [49]. Further, reports of hAEC differentiation in vitro towards hepatocyte-like cells have been put forward by different groups using variable inducing factors [11, 21, 50, 51].

Even though there are no clear reports on in vivo differentiation, many in vivo studies provide evidences supporting the potential application of hAEC in the treatment of hepatic diseases. The anti-fibrotic effect of hAEC was demonstrated by transplanting them into CCl4-induced cirrhosis animal model which resulted in cell engraftment with reduction of hepatocyte apoptosis, inflammation, and fibrosis. The observed therapeutic effect of transplanted hAEC as reported was more likely due to hAEC-mediated reduction of pro-inflammatory and pro-fibrotic cytokines and induction of a collagen-degrading phenotype [52]. The same group, using mice chronically injured with long-term CCl₄ treatment, also reported that hAEC engrafted in injured livers lead to significant changes in hepatic macrophage numbers and phenotype, associated with a significant reduction in the extent of established fibrosis [53]. In another study, hAEC were transplanted into the liver of retrorsine-treated immunodeficient mice and post transplantation were found in the liver expressing mature liver genes, including cytochromes, plasma proteins, transporters, and other hepatic enzymes [50].

The differentiation potential of hAEC towards pancreatic cell type has also been investigated. Pancreatic differentiation was induced on culturing the hAEC in media containing nicotinamide [11]. The differentiated hAEC showed the expression of the downstream transcription factors PAX-6, Nk2 homeobox2, and the mature hormones insulin and glucagon [11] along with formation of masses of cells with a cyst like appearance, expressing the pancreatic exocrine cell marker amylase 2B [21]. Interestingly, transplantation of hAEC reported a reduction in the hyperglycemia levels in streptozotocin-induced diabetic mice, thereby indicating potent future therapeutic application for the treatment of type I diabetes mellitus [54, 55].

The capability of hAEC to differentiate towards other cell types of endodermal lineage has also been reported. Moodley et al. demonstrated that naive hAEC differentiate into cells with features of type II pneumocyte 2 weeks post parenteral injection into a severe combined immunodeficiency (SCID) mouse model of bleomycin-induced lung injury. The transplanted hAEC reduced inflammation and reduced fibrosis post-lung injury [30]. Meanwhile, our group found that transplantation of a mixture of amniotic and chorionic membrane-derived cells, including hAEC, into bleomycin-challenged immunocompetent mice, by different delivery routes, significantly reduces the severity of lung fibrosis; likely owing to paracrine actions exerted by the soluble molecules they secrete [56], rather than differentiation of the transplanted cells, as supported by the fibrosis-reducing action of conditioned medium generated from amniotic membrane-derived cells when delivered in the same lung fibrosis animal model [57].

Additionally, there are in vitro reports indicating that hAEC possess a mesodermal differentiation potential as well. For example, it is reported that naive hAEC can differentiate into cells with characteristics of mesodermal-derived adipocytes, osteocytes [5, 11, 21, 33, 58], and chondrocytes [22]. hAEC were also reported to differentiate towards the myogenic lineage comparable to bone marrow-MSC [5]. Differentiation towards cardiomyocyte-like cells was induced on using ascorbic acid supplemented media and after induction, the expression of cardiac-specific genes atrial and ventricular MLC 2 and the transcription factors GATA-4 and Nkx 2.5 was observed [11, 21]. The differentiated cells expressed Troponin T and contained T tubules, numerous myofilaments and myofibrils, and H bands characteristic of relatively mature cardiomyocytes [21].

Taken together the above evidences indicate the potential of hAEC to differentiate towards cells of all the three germ layers in vitro, while there are no strong evidences of their in vivo differentiation capacity. However, several preliminary studies mainly in the field of hepatic and neural diseases support their possible role as an effective therapeutic agent.

Various groups described the isolation of MSC from the decidua basalis as well as the parietalis [9, 63–67]. Isolation of decidua-MSC is usually initiated by mechanical mincing of the tissue followed by enzymatic digestion employing collagenase [68], or trypsin (with or without EDTA) [66, 67, 69], or collagenase in combination with trypsin [65, 70, 71]. Cocktails of enzymes such as collagenase and DNase with dispase or hyaluronidase and pronase [72] have also been employed. The enzymatically digested tissue is filtered through meshes of varying pore size (25–100 μm), centrifuged [63, 70] and the pellet is then suspended and cultured in an appropriate medium. Use of density gradient centrifugation for cell enrichment post enzymatic digestion has also been reported [65, 71].

Non-enzymatic method of isolation of MSC has also been reported [9]. The procedure briefly entails mechanical mincing of the desired tissue, following which cell suspension is prepared by extensively flushing the minced tissue with an appropriate washing media through a filter. The cell suspension is then cultured in an appropriate media and the adherent cells are selectively grown to confluence [9]. The non-enzymatically prepared decidual MSC usually results in a certain degree of contamination of fetal alleles [9]. However, other researchers clearly indicated that the isolated cells in culture at varying passages are largely of maternal origin, as demonstrated by fluorescence in situ hybridization (FISH), short tandem repeat analysis, and karyotypic analysis [66, 67].

The decidua-derived MSC exhibit high proliferative ability and can be continuously cultured until passage 20 after which they undergo senescence [63, 67, 73]. The telomere progressively shortens with each cell division and as telomere length is an indicator of replicative senescence, decidual MSC would not proliferate indefinitely thereby indicating the safety issues concerning telomerase-associated teratoma formation [67].

The regions of the fetal part of term placenta that have been of particular interest in terms of isolation of MSC include chorionic villi, chorionic membrane, and amniotic membrane. Irrespective of the placental region of interest, the basis for isolation of MSC broadly entails mechanical separation of the desired tissue, mincing and enzymatic digestion followed by filtration, centrifugation, and suspension in an appropriate medium to culture adherent stromal cells.

A few initial reports on isolation of MSC employed the whole unfractionated placenta, wherein all the different regions of the fetal part are used as a starting material [74, 75]. The tissue is minced, enzymatically digested using trypsin-EDTA [74–76] and finally processed and cultured. Contaminating erythrocytes are removed either by flushing the working medium through the arterial-vein circuit prior to any enzyme treatment or by lysing them using ammonium chloride solution after enzymatic treatment [75, 77].

Isolation of MSC from the chorionic villi popularly involves the explant culture method, in which the decidual layer is usually removed and a central cotyledon of the villous vascular bed of the fetal tissue is minced. The minced tissue, either treated with trypsin and DNase or untreated, is washed and the tissue pieces are allowed to adhere on a suitable surface enabling the cells to migrate out from the cut end of the tissue pieces which are then harvested [78]. Other enzymes reported to have been employed on tissue fragments for culture include collagenase and dispase [5]. Alternative to the explant procedure, the isolation of MSC from the villi may also employ enzyme treatment (trypsin or collagenase-DNase) and the tissue suspensions obtained are further processed and cultured [8, 79]. The duration of trypsin treatment may vary, that is, from as less as 10 or 20 min to as much as 24 h [8, 78, 80].

In terms of MSC isolated from the fetal membranes (the chorionic and amniotic membranes), the hCMSC are relatively less characterized compared to hAMSC and hence their descriptions are relatively limited [81, 82]. The hCMSC are isolated from the deflected part of the chorionic membrane through an initial mechanical and enzymatic removal of the trophoblastic layer, followed by a combination of enzymes to release the MSC [4]. The amniotic membrane on the other hand comprises of epithelial cells on one side of the membrane adjacent to the amniotic fluid with the mesodermal layer facing the other side. Thus the isolation of hAMSC entails either prior removal of contaminating epithelial cells or isolating without affecting the epithelial layer [23, 83, 84].

Various enzymes employed in the isolation of MSC from the amnion and the chorion have been reported; such as dispase and collagenase [82, 85] in combination with DNase [23, 86]. Some workers have reported on the use of trypsin solely [87] or in combination with collagenase [88–90]. The use of papain with dispase and DNase has also been described [91]. Trypsin employed in the isolation procedure helps in removal of the contaminating epithelial cells [25, 84] and the relatively purified tissue is then subjected to appropriate enzymatic digestion to release the corresponding hAMSC [83, 90]. Reports on scraping out of the amniotic mesoderm without affecting the epithelial layer have been described. The isolated mesoderm is then subjected to collagenase digestion to liberate the stromal cells [17, 92]. Other popular methods on selective isolation of hAMSC mostly entail enzymatic process [23, 35, 89, 91]. The digested tissues are generally filtered, centrifuged and the pellet is then suspended in an appropriate medium. Use of gradient ultracentrifugation of digested tissues for enrichment of the MSC prior to culture has also been reported [83]. The isolated cells are often accompanied with contaminating blood cells and removal of these contaminants requires lysis with ammonium chloride, KHCO₃, and EDTA [34], or employs anti-CD45 and glycophorin A immunomagnetic beads [86].

The MSC from amnion and chorion are usually of fetal origin as indicated by FISH and polymerase chain reaction analysis [23, 82, 87]. Nonetheless, presence of maternal cells in fresh as well as cultured preparations of hCMSC cannot be ruled out [23, 82].

It has been reported that the isolated hAMSC and hCMSC adhere and proliferate in culture and can be kept until passages 5–10 [4, 61]. However, recent report indicated that hCMSC could be propagated until 55 passages (more than 100 population doublings) and maintains telomere length and lack telomerase activity [82].

According to the above mentioned consensus reached by the first international workshop on placenta-derived stem cells [4], the minimum criteria for identifying hAMSC and hCMSC include adherence to plastic; formation of fibroblast colony-forming units; expression of mesenchymal markers CD90, CD105, and CD73 by greater than 90 %; absence or less than 2 % expression of hematopoietic markers such as CD45, CD34, CD14, MHC class II; fetal origin; and potential to differentiate towards at least one or more lineages such as osteogenic, adipogenic, chondrogenic, and vascular or endothelial. hAMSC and hCMSC isolated from bulk populations retained the mentioned phenotype for at least 15 passages despite the acquisition of morphological changes [23]. In addition expression of CD166 [93], CD29 (β1-integrin receptors) [4], CD106 or vascular cell adhesion molecule-1 [8, 87], CD54 [7, 23, 94], α-SMA, vimentin [73], CK7 [23] at varying levels have been reported. Conflicting reports on STRO-1 expression have also surfaced [63, 66, 67]. In addition expression of pluripotency markers such as OCT-3, OCT-4, SOX-2, Nanog, SSEA-3, and SSEA-4 have been inconsistently reported [93]. Moreover, OCT-4A, isoform of OCT-4, associated with pluripotency, may be a case of false positive expression and therefore warrants critical refinement in characterization scheme [95].

Detailed characterizations by researchers are revealing that marker expression varies at different time points of the culture and therefore necessitates passage specific marker profiling. Freshly isolated hAMSC expressed CD44, CD73, CD90, and CD105 at low to moderate levels as compared to up to 90 % of its expression after 4–5 passages [90, 96–98]. Hematopoietic and vascular cell related markers such as CD14 [90], CD45, and MHC class II [96–98] have been reported in relatively low but distinguishable amounts in freshly isolated hAMSC [96] which were lost or absent after subsequent passages [97, 98].

A plethora of markers in addition to these are being increasingly reported and hence the characterization of MSC isolated from different parts of the placenta is getting far more complex than previously imagined. A few freshly isolated hAMSC showed positive staining for epithelial makers such as CK5 and CK18, epithelial cell adhesion molecule and CD49f [84, 90] and low affinity nerve growth factor receptor (CD271). Increased surface expression of CD349 or frizzled family receptor 9 (FZD9) has been reported suggesting that it could be a suitable marker for MSC isolation and important for stem cell renewal [79]. CD51/61 and β3 integrins [87], CD146 (also known as the melanoma cell adhesion molecule or cell surface glycoprotein MUC18) [8], and REX-1 [67, 85] have also been noted. Other reported markers include organogenesis regulator GATA-4 [67, 99] and hepatic nuclear factor-α [22].

MSC isolated from the different parts of the placenta have successfully shown to differentiate towards cells of the classic mesodermal lineages, namely adipogenic, osteogenic, and chondrogenic [4, 9, 67], with reportedly different magnitude of commitment towards a specific lineage among cell types depending on the placental tissue source and gestational age, passages in culture and protocols to induce/evaluate differentiation [5, 23, 24, 63, 68, 72–74, 85, 87, 88, 90, 92, 94, 100]. Interestingly, MSC from placenta have also demonstrated increased osteogenic differentiation on microcarriers [101], nanofiber scaffolds [102] and slowly degradable polyurethane foams [103] indicating potent application for future in vivo bone regeneration and tissue engineering.

Successful reports of in vivo cartilage repair in rat and rabbit animal models have also been demonstrated [90, 104]. For example, hAMSC transplanted into rat with collagen-scaffold into defective cartilage underwent characteristic morphological changes concurrently with deposition of collagen type II, suggesting their differentiation into chondrocytes in vivo thereby alleviating ostochondral defect [104]. Likewise, hAMSC seeded onto poly lactic-co-glycolic acid blocks and transplanted into rabbit knee joints were detectable up to 8 weeks, without evident inflammatory response or tumorigenic proliferation, suggesting that they represent potent allo-transplantable cell resource for cartilage repair [90].

Induction of myogenic differentiation in vitro of MSC from amnion [5, 94] and from other placental regions [5] results in the formation of myotube precursors in the form of long multinucleated cells [5] coupled with the expression of transcription factor MyoD1 [5, 94], myogenin [67, 94], and desmin [94]. hCMSC and first trimester villous MSC exhibited more pronounced myogenic differentiation with respect to first trimester and term hAMSC [5].

Cardiomyocyte differentiation of decidual MSC led to the expression of ANP, cardiac-specific transcription factors homeobox Nkx [67] and GATA-4, genes such as atrial MLC-2a, ventricular MLC-2v, and the cardiac troponins cTn1 and cTnT. Further, integration of hAMSC into cardiac tissue and their differentiation into cardiomyocyte-like cells have been observed after transplantation into rat hearts following myocardial infarction [105]. It has been revealed that mixed ester of hyaluronan and butyric and retinoic acids (HBR) promoted cardiogenic/vasculogenic differentiation of fetal membrane-derived MSC and enhanced the expression of genes essential for cardiomyogenesis and cardiac markers such as the sarcomeric myosin heavy chain and the alpha sarcomeric actin [106]. HBR treated cells reflected expression of von Willebrand factor (vWF) and enhanced cardiac repair in infarcted rat hearts [106]. Moreover, hAMSC seeded onto an elastic combined scaffold, composed by a fibrin layer and by a microporous synthetic layer, have shown increased disposition to cardiac and vascular marker expression [107] indicating avenues for potent future therapeutic applications.

In addition to the traditional mesodermal lineages, placenta-derived MSC have shown to exhibit ectodermal as well as endodermal differentiation. Ectodermal (neural) changes result in formation of long bipolar, complex multipolar or branching processes [71, 91, 92]. Nestin, the widely accepted neuroectodermal maker is expressed in almost all placental MSC in both undifferentiated and differentiated states [71, 74, 80, 82, 86, 91, 108] at variable levels (20–87 %) [5, 80, 91, 109]. Neuronal markers post differentiation such as β-tubulin-III [80, 87], mushashi [91], NSE [68, 71, 74, 75], neurofilament-200 [67, 82], Tuj 1 [91, 92], MAP2 [86, 87, 92], Neu-N [92, 108] and NMDA receptor NR1 [87], GFAP [68, 71, 75, 91, 92] and CD133 [5] have been reported. Transcription factors related to midbrain dopaminergic neurons including expression of LIM homeobox transcription factor 1-beta, PAX-2 and nuclear receptor related-1 protein have also been documented [92]. Expression of dopamine secreting neuron marker tyrosine hydroxylase [71, 92, 108] and markers such as FZD9 further support neurodifferentiation potential [71]. Neural differentiation of chorionic villi MSC was observed when the same was co-cultured with astrocytes from newborn rats without the addition of exogenous differentiation factors [80]. The authors have demonstrated that the presence of astrocytes (i.e., neural environment) influenced the neural differentiation of villi MSC, indicating the importance of microenvironmental stimuli in regulating differentiation cues [80]. Reports by Park et al. on in vivo differentiation potential of placenta-derived MSC in a neonatal rat model of stroke as well as Parkinson’s disease rat model was reflected when it led to subsequent recovery of locomotor activity upon formation of dopaminergic cells in the host brain [109, 110]. Facilitation of neuronal differentiation with placenta-derived MSC seeded on to gold coated collagen nanofibers have also been demonstrated, indicating possible application in neuroregeneration for future therapeutic purposes [111].

Endodermal (hepatic, alveolar, and pancreatic) differentiation potential has also been reported [4, 99, 112]. In terms of hepatic differentiation, it has been observed that undifferentiated decidual MSC already express albumin and c-met [69] while undifferentiated hAMSC express α-FP, CK18, and α-1AT in addition to albumin [113] even in the absence of differentiating stimuli. Following differentiation, by means of various protocols, increased expression of α-1AT and albumin [113] and expression of CK19 and CK7 [84] were observed. However, contradictory results with α-FP expression were reported; from induction of expression post differentiation [113], to reduced expression within 48–72 h [69], to absolute undetected levels [84]. Urea synthesis, low density lipoprotein uptake [114], CYP450 mixed function oxidase activity [84, 114] as well as generation and storage of glycogen have also been shown [113]. Moreover, placental MSC transplanted into injured pig models exhibited differentiation resulting in hepatic regeneration and better survival rate [114].

In terms of alveolar differentiation, decidual MSC expressed prosurfactant protein C and surfactant protein B (SP-B) immunostaining and expression of SP-B and GATA-6 genes post differentiation [67].

Pancreatic differentiation has been exhibited by in vitro differentiation into pancreatic beta cells [70, 115]. Formation of epithelial-like cells and many islet-like clusters usually accompany the changes [115]. Pancreatic cell specific genes such as PDX-1, Islet-1, PAX-4, PAX-6, neurogenin 3, neurogenic differentiation-1, glucose transporter (Glut)-1, Glut-2, proprotein convertase (PC)1/3, PC2, glucokinase, pancreatic polypeptide, somatostatin, and guanylate cyclase are significantly expressed post differentiation [115, 116]. Further they are positive for insulin, glucagon, somatostatin, and c-peptide [115, 116]. Transplantation of placental MSC into kidneys of diabetic mouse model reflected improved recovery with normalized blood glucose owing to co-expression of human as well as mouse insulin and c-peptide [116]. Human cell specific gene hAlu and other pancreatic specific genes were also detected in the graft [116]. These findings indicate potent future application in diabetes.

The angiogenic differentiation of placental MSC has been controversial. Increased expression of vascular endothelial growth factor receptor (VEGFR)-1 (FLT-1) [82, 94, 117] and VEGFR-2 (KDR) [94, 117] along with the expression of intercellular adhesion molecule-1 [94], CD34 [94, 117], CD31 [68, 82, 117] and mature endothelial marker vWF post differentiation have been reported in hAMSC, hCMSC as well as decidual MSC. In addition, post differentiated expression of angiogenic factors such as vascular endothelial growth factor (VEGF), hepatocyte growth factor (HGF) [68, 118], angiopoietin [68, 118, 119], EGF [118, 119], interleukin (IL)-8, and insulin-like growth factor-1 [118] have also been demonstrated. The angiogenic potential is further reflected by the ability to form capillary tube-like structures in vitro [82, 94, 119]. In vivo differentiation with the ability to augment angiogenesis in chick chorioallantoic membranes has also been reported [117]. The authors suggest the involvement of α5β1 and fibronectin to mediate the key angiogenic steps through VEGF-A induced pathway [117]. In contrast to these findings, however, a report on the ability of hAMSC to exhibit angiogenic properties but resistance to mature endothelial cell differentiation has been demonstrated [119]. The authors reported that hAMSC exhibited significant downregulation of pro-angiogenic genes and proteins such as tenascin C, angiopoietin receptor Tie-2, VEGF-A, CD146, FGF-2, IL-8, MMP-1, urokinase type plasminogen activator receptor with concomitant up-regulation of anti-angiogenic factors endostatin, serpinF1, the FGF-2 signaling antagonist sprouty1 and angioarrestin. Interestingly, despite this observation, hAMSC exhibited angiogenic properties but resisted expression of vWF and vascular endothelial-cadherin [119]. It is speculated that the niche or the microenvironment of MSC may influence the observed differences in angiogenic differentiation. Castrechini et al. for example reported that MSC isolated from human chorionic villi reside in a vascular niche and are able to contribute to angiogenic features such as vessel maturation and stabilization. However, the stromal cells isolated from this region were negative for vWF [8].