Keywords

3.1 Mesenchymal Stem Cell Concept

Mesenchymal stem cells (MSCs) are multipotent non-hematopoietic progenitors located within the stroma of the bone marrow and other organs that are phenotypically characterized by the expression of several markers (e.g., CD73, CD90, and CD105) and the lack of expression of CD14 or CD11b, CD19 or CD79α, CD34, CD45, and HLA-DR surface molecules [1, 2]. According to a proposal of the International Society for Cellular Therapy [3], three criteria define all types of stem cells: self-renewal, multipotency, and the ability to reconstitute a tissue in vivo. Since there are no specific markers for MSCs, the main criteria for their identification are adherence to the plastic of the tissue culture flask, fibroblast-like morphology, prolonged capacity for proliferation, and the capacity to differentiate into cells of mesodermal lineage in vitro. MSCs are classified, according to the developmental stage from which they are obtained, into embryonic, fetal, or adult stem cells. hESCs are pluripotent and could give rise to all specialized cell types of the organism. However, the tumorigenicity of these cells and technical and ethical considerations limit their availability. In contrast, adult stem cells are rare cells thought to be present in all tissues and responsible for maintaining the homeostasis of the specific tissue [4]. These cells, previously thought to be limited in potential, have been shown to differentiate into multiple mesoderm-type lineages, including chondrocytes, osteoblasts, adipocytes, tenocytes, myotubes, astrocytes, and hematopoietic-supporting stroma [5–7], and also into cell types of ectodermal (e.g., neurons) and endodermal (e.g., hepatocytes) origin [8].

These cells have been isolated from several tissues such as bone marrow [2, 9], articular cartilage [10], synovial membrane [11, 12], perichondrium [13], periosteum [14], connective tissue of dermis and skeletal muscle [15], adipose tissue [16, 17], peripheral blood [18–20], liver [21], lung [22], placenta [5, 23–25], umbilical cord [26–28], umbilical cord blood [29], amniotic fluid [23, 25, 30], and amniotic membrane [31–33]. Moreover, the list of tissues with the potential for tissue engineering is increasing because of recent progress in stem cell biology [34].

Cell therapy using MSCs is a new clinical approach for the treatment of a large number of genetic and degenerative human diseases, including hematopoietic and immune system disorders, diabetes, heart failures, chronic liver injuries, and neurodegenerative disorders. The recent use of autologous or allogenic stem cells has been suggested as an alternative therapeutic approach for cartilage treatment [35, 36]. Human MSCs are probably responsible for normal tissue renewal as well as for response to injury [37–39]. Stem cell transplantation uses cells isolated from small tissue samples, proliferated in culture, to obtain the appropriate number for clinical applications. The use of autologous MSCs avoids immunological rejection problems and the ethical conflict of using hESCs. For these reasons, MSCs are a promising cell resource for tissue engineering and cell-based therapies [38]. The interest in MSCs and their possible application in cell therapy have resulted in a better understanding of the basic biology of these cells. Due to the low number of MSCs that can be isolated from a tissue sample, culture expansion is necessary to obtain adequate cell numbers for clinical purposes and for the analysis of molecular mechanisms.

The bone marrow is the traditional tissue source used for obtaining adult MSCs, but it has a number of disadvantages. The most important limitations are the accessibility and that the procedure required for obtaining this type of tissue is invasive, painful, and associated with morbidity. In addition, the number of cells obtained is low and the potential for proliferation and differentiation declines with donor age [40, 41]. Therefore, the identification of alternative sources of MSCs for both therapeutic and research purposes would be beneficial.

The HAM or amnion has recently emerged as another novel and alternative source of stem cell populations. The HAM is the innermost membrane surrounding the fetus. Because it arises from embryonic epiblast cells prior to gastrulation, it has been suggested that it may retain a reservoir of stem cells throughout pregnancy [42].

3.2 Human Amniotic Membrane or Amnion

The placenta is a structure of fetal-maternal origin with a round shape, 15–20 cm in diameter, and 2–3 cm in thickness [43]. The thickness of the full-term amnion varies between humans and depends on the location of the sample. HAM functions as a filter and preventive shock absorber that protects against infections, traumas, and toxins. This organ is involved in the maintaining fetal tolerance and allows nutrient uptake and gas exchange with the mother but also contains a high number of progenitor cells or stem cells. Moreover, the volume of term placenta makes it an attractive source of stem cells, since as an average human term placenta weighs more than 590 g [44]. HAM develops from extraembryonic tissue and consists of both a fetal component (the chorionic plate) and a maternal component (the deciduas) that are comprised of an epithelial monolayer, a thick basement membrane, and an avascular stroma [45, 46] (Fig. 3.1). The amnion is a thin (up to 2 mm), avascular, strong, elastic, translucent, and semipermeable fetal membrane attached to the chorionic membrane. Both the amnion and chorion form the amniotic sac filled with amniotic fluid, providing and protecting the fetal environment. The outer layer, the chorion, consists of trophoblastic chorionic and mesenchymal tissues. The inner layer, the amnion, consists of a single layer of ectodermally derived epithelium uniformly arranged on the basement membrane, which is one of the thickest membranes found in any human tissue, and a collagen-rich mesenchymal layer [47]. This mesenchymal layer can be subdivided into the compact layer, forming the main fibrous skeleton of the HAM, the fibroblast layer, and an intermediate layer, which is also called the spongy layer or zona spongiosa [45]. Resistance to rupture of HAM is provided by the collagen present in the basement membrane of the amnion. Spontaneous premature rupture of the fetal membranes complicates 1–4 % of the pregnancies. This is due to multiple factors such as infection and genetic predisposition. These premature ruptures are associated with elevated expression levels of relaxins, low expression levels of extracellular matrix (ECM) proteins synthesized by the fetal membranes, or to degradation of these proteins by induced matrix metalloproteinases (MMPs) and subsequent ECM remodeling [48].

Fig. 3.1
figure 00031

Structure of the fetal membrane at term stained with hematoxylin and eosin (HE). Original magnification: 40× (a) and 200× (b)

Full size image

The two layers of the amniotic membrane originate at day 8–9 after fertilization, when implantation of the blastocyst has occurred. The inner cell mass of the blastocyst differentiates into two layers, the epiblast and the hypoblast; both layers form the bilaminar embryonic disc. The epiblast gives rise to the three germ layers (ectoderm, mesoderm, and endoderm) and the amniotic epithelium [49].

3.3 Localization of Human Amniotic Membrane-Derived Cells

The localization of HAM-derived cells was examined by our group [31, 32]. We assessed the co-localization of different stem cell markers in histological sections of amniotic membrane by means of immunofluorescence assays. In particular, we studied the co-localization of the CD44, CD90, CD105, and CD271 markers.

Our group did not observe any cells in which co-localization of three and/or four stem cell markers occurred. However, we frequently observed co-localization of double markers, for example, we found CD105 co-located with CD90, CD44 co-located with CD90, and CD271 co-localized with CD44 (Fig. 3.2). Most cells labeled with the different stem cell markers were hAMSCs from the thick basement membrane, although in some membranes we observed hAECs, derived from the embryonic ectoderm, that were labeled only for the CD105 marker. hAMSCs are derived from embryonic mesoderm [50] and are sparsely distributed in the stroma underlying the amnion epithelium [51]. On the other hand, hAECs form a continuous monolayer of embryonic ectodermally derived epithelium uniformly arranged on the basement membrane in contact with the amniotic fluid.

Fig. 3.2
figure 00032

Localization of HAM-derived cells, hAMSCs and hAECs, in healthy HAMs. Representative section of the HAM stained with HE (a), indicating the epithelial cells from the extraembryonic ectoderm (EC) and the thick basement membrane (BM). Immunofluorescence analysis of stem cell marker expression of human amnion cells, nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) (b–d). Representative images of hAECs, positive for CD105, and hAMSCs, positive for CD44 (b). Representative photos of hAMSCs positive for CD105-CD90 (c) and CD44-90 (d). Original magnifications: 200× (Images taken from Díaz-Prado et al. [32])

Full size image

The immunofluorescence results of our group [31, 32] indicated that the HAM contains at least two different cell types having stem cell characteristics and that these cells are located in the basement membrane and in the single layer of ectodermally derived epithelium. These common and well-defined human MSCs markers were previously described for bone marrow MSCs. Moreover, we showed that hAECs are positive for the epithelial marker cytokeratin 7, which confirms its epithelial nature (Fig. 3.3).

Fig. 3.3
figure 00033

hAECs (a) and hAMSCs (b) were stained with CK7 antibody by means of immunohistochemistry

Full size image

3.4 Human Amniotic Membrane as a Source of Stem Cells

HAM expresses only moderate levels of major histocompatibility complex (MHC) class I antigens and MHC class II antigens on its surface. HAM-isolated cells have anti-inflammatory properties. Moreover, there was no evidence of tumorigenicity when isolated human amniotic cells were transplanted into human volunteers or into patients in an attempt to correct lysosomal storage diseases [52–54]. Therefore, hAECs and hAMSCs seem to be immune-privileged cells and suitable for allotransplantation and regenerative medicine [40, 55].

Because fetal tissues are routinely discarded postpartum, HAMs have proved to be abundant, inexpensive, and easily obtained with a virtually limitless availability [45, 47, 56–58]. Therefore, the HAM represents a very useful source of progenitor cells for a variety of applications. Because human embryos are not sacrificed for the isolation of progenitor cells from HAMs, the current controversies associated with the use of hESCs can be avoided [43, 55, 56, 58]. Given the minimal ethical and legal issues associated with HAM cell usage, further investigation into their functional potentials in vivo is warranted.

HAM is becoming appreciated as an alternative to bone marrow for adult MSCs for regenerative medicine. This tissue provides high efficiency in MSC recovery with no intrusive procedures [33]. Moreover, harvesting cells from the HAM is noninvasive and safe. A major advantage of cells isolated from the HAM is that they are harvested after birth and can be cryogenically stored to be available in a timely manner for patient therapy after being thawed and expanded for use in tissue engineering, cell transplantation, and gene therapy.

MSCs from first-, second- and third-trimester placental compartments, including the amnion, chorion, decidua parietalis, and decidua basalis, were isolated and represent less than 1 % of the cells present in the human placenta [22, 33, 59, 60]. HAM contains two different cell types from different embryological origin [33, 61]: hAECs, derived from embryonic ectoderm [50] which form a continuous monolayer that contacts the amniotic fluid, and hAMSCs, derived from embryonic mesoderm [50] which are sparsely distributed in the stroma that underlies the amnion epithelium [51]. Both hAECs and hAMSCs secrete various antiangiogenic and anti-inflammatory proteins such as interleukin (IL)-1 receptor antagonist; activin A; tissue inhibitors of metalloproteinases (TIMP-1, TIMP-2, TIMP-3, and TIMP-4); and IL-10 which are deposited within the amniotic membrane stroma [62]. Hyaluronic acid may act as a ligand for CD44 and may entrap inflammatory cells in the stroma.

Some papers reported the isolation of HAM-derived cells, from the mesenchymal and epithelial regions of the amnion, from the full-term amnion after its manual separation from the chorion. Bailo et al. [63] isolated and characterized amnion and chorion cells from human term placenta suggesting that both kinds of cells may represent an advantageous source of progenitor cells with potential applications in a variety of cell therapy and transplantation procedures. For this purpose, different methods to isolate HAM-derived cells have been published [33, 59, 63–67].

All these protocols start with a mechanical separation of the amniotic membrane from the underlying chorion through the spongy layer [43]. This step is followed by a digestion with trypsin, dispase, or other digestive enzymes, in different concentrations and for different periods of time, to release the hAECs from the basal membrane. hAMSCs can be subsequently released through subsequent digestion with collagenase [49], alone or combined with DNAase [23].

Regardless of the morphological features of human amnion-derived cells, hAMSCs show plastic adherence and fibroblast-like growth usually observed with MSCs from bone marrow (Fig. 3.4). After 3–4 weeks of hAMSCs culture, it is possible to obtain a population of adherent mesenchymal cells morphologically identical to MSCs isolated from bone marrow. These stromal cells are easy to expand in vitro for at least 9 passages without morphological changes. Furthermore, their immunophenotypic characterization demonstrates the presence of common well-defined human MSC markers previously described for bone marrow (CD90, CD44, CD73, CD166, CD105, CD29) with the absence of the hematopoietic markers CD34 and CD45 and the concomitant lack of fibroblast marker [68, 69]. The absence of hematopoietic or monocytic marker gene expression excludes the possibility that the observed plasticity of these cells is due to contamination with stem cells from fetal or cord blood or with embryonic fibroblasts. This antigen expression pattern is consistent with the data previously published in cells isolated from the amnion and other regions of the term placenta [5, 51, 63]. hAMSCs are also positive for pluripotency markers such as Oct4 (octamer-binding protein 4), NANOG, SOX2 (SRY-related HMG-box gene 2), and REX-1 [49], but positivity for embryonic stem cell markers, SSEA-3 or SSEA-4, remains debated [49]. hAMSCs may be considered as superior to adult MSCs in their differentiation and proliferation capacity due to their higher OCT4 mRNA levels [33]. Moreover, hAMSCs also express low levels of HLA-A, HLA-B, and HLA-C, but do not express HLA-DR, indicating that these stromal cells may be useful in clinical transplantation procedures [49].

Fig. 3.4
figure 00034

Morphology of cultured hAMSCs (a) and hAECs (b) isolated from healthy HAM. Original magnifications: 100×

Full size image

On the other hand, isolated hAECs are small-size cells that are easy to expand in vitro for at least three passages without morphological changes; they display epithelial morphologies and grow into a tightly packed, cobblestone monolayer in culture [70] (Fig. 3.4). These cells generally have a central or eccentric nucleus, one or two nucleoli, and abundant cytoplasm, usually vacuolated [66]. hAECs are positive for desmin and vimentin [58]. These epithelial cells also reveal an antigen expression profile characteristic of culture-expanded MSCs [51], since they are positive for the same markers as for hAMSCs. Primary hAECs seem to contain class IA and class II HLAs, consistent with a low risk of tissue rejection [42]. They do not express HLA-A, HLA-B, and HLA-C belonging to class I of the MHC and HLA-DR and HLA-DQ belonging to the class II MHC [45, 65]. when these cells follow pancreatic or hepatic differentiation, but not cardiogenic differentiation, express a significant percentage of class IA but not class II HLAs [71]. In addition, hAECs secrete a number of immunosuppressive factors that target the innate and adaptive immune systems, which may support survival following transplantation [70]. Evidence for long-term self-renewal is not still available for hAECs, probably may be due to the absence of telomerase that limits their ability to divide in culture.

Phenotypes of the two cell populations (Fig. 3.5), hAMSCs and hAECs, are maintained from passage 0 to passage 9 [32]. It is important to notice that although both populations show similar signature regarding cell surface receptor expression pattern, they show many differences with regard to cell shape and cell arrangement [32, 51]. These same findings were previously described by Bilic et al. [51]. These investigators isolated these two populations and concluded that hAECs and hAMSCs in culture exhibited and maintained a similar marker profile of mesenchymal progenitors. hAECs also express surface markers of undifferentiation normally present on embryonic stem and germ cells such as SSEA-4 and STRO-1. Both embryonic stem cell markers are present in more quantity in hAECs than in hAMSCs [32, 42, 51], possibly indicating that hAECs could be at a more early state of undifferentiation. In this regard, Ilancheran et al. [42] also showed that hAECs expressed SSEA-3 (stage-specific embryonic antigen 3), SSEA-4, TRA-1–60 (tumor rejection antigen) and TRA-1–81, and other antigens such as the ABCG 2/BCRP (a member of the ATP-binding cassette superfamily), CD9, CD24, CD90, CD117, E-cadherin, integrin α6 and β1, and c-met (receptor growth factor of the hepatocyte) [43, 45]. It has to be noted that initially isolated, hAECs are not homogenously positive for all these antibodies. Some surface markers such as CCR4- and CD117-positive cells are very rare, while others such as CD9 and integrin α6 and β1 are expressed on virtually 100 % of the cells, indicating that hAECs are a heterogeneous cell population with respect to cell surface profiling [67]. These epithelial cells also express Oct4, NANOG, SOX2, REX-1, FGF4, Lefty-A, and TDGF-1gene products associated with pluripotent embryonic stem cells [49, 67, 72, 73]. When hAECs are cultured as an adherent monolayer for several weeks, small spheroids are evidenced over the cobblestone pavement of epithelial cells. These cell clusters express SSEA-3, SSEA-4, TRA 1–60, and TRA 1–80 stem cell-specific cell surface antigens. Moreover, the stem cell molecular marker genes Oct4 and NANOG are also expressed in the small cell clusters, suggesting that hAECs form embryonic body-like structures that maintain their stem cell nature in culture [73].

Fig. 3.5
figure 00035

Analysis of hematopoietic and standard adult stem cell markers on hAECs and hAMSCs. *Means P value <0.05 (Mann–Whitney U-test)

Full size image

hAECs and hAMSCs can be grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 20 % fetal bovine serum (FBS) and 1 % penicillin-streptomycin (P/E) and seeded into culture flasks. Moreover, hAECs could be cultured with or without the addition of growth factors such as epidermal growth factor (EGF) or basic fibroblast growth factor (bFGF) and mostly in the absence of leukemia inhibitor factor (LIF) [48]. Both populations should be expanded in a humidified 5 % CO2  atm at 37 °C. After the isolation of both cell types, it is advisable to perform immunohistochemical stainings (e.g., for cytokeratin 7, CK7) to demonstrate the purity of both populations. In this regard, only hAECs may be positive for this or other epithelial markers. In monolayer cultures, these hAECs are positive for low molecular weight cytokeratins, confirming their epithelial nature. Moreover, although initially they are vimentin-negative, hAECs become vimentin-positive during cell culture. Vimentin-positive hAECs remain positive for cytokeratins, indicating that in vitro culture may induce dedifferentiation of these epithelial cells [67].

Recently, Parolini et al. [49] published a comparison of key features of HAM-derived cells and human BM-MSCs. These authors postulated that BM-MSCs have a higher cell doubling time than hAECs, while for the hAMSC this time was not reported yet. Regarding the maximum number of passages, it ranges from 5 to 10 for hAMSCs, 10–20 for BM-MSCs, and 30 for hAECs. But there is a contradiction with the passage number at which HAM-derived cells stop proliferation. Based on the literature, proliferation slows down with every passage and cells settle into senescence until proliferation ceases. For example, Miki et al. [66] and Parolini et al. [65] state that hAECs grow normally for 2–6 passages before proliferation ceases. On the contrary, Bilic et al. [51] confirmed that hAECs and hAMSCs proliferation almost stops beyond passage 5, whereas Toda et al. [58] postulated that hAECs senescence is reached at lower passages, P3 or P4. However, Alviano et al. [33] and Soncini et al. [59] indicated that hAMSCs are easily expanded in vitro for at least 15 passages without any visible morphological alterations, but they used cells not exceeding P4 for cell characterization and multilineage differentiation potential studies.

Another comparison between placental cells and BM-MSCs was the aim of the paper published by Barlow et al. [5]. These authors compared human placenta-derived MSCs (the placental tissue included amnion, chorion, and decidua) and human bone marrow-derived MSC in terms of cell characteristics, optimal growth conditions, mesodermal lineage differentiation, and in vivo safety specifically to determine if human placenta-derived MSCs could represent a source of human MSC for clinical trials. They demonstrated that both populations were similar in terms of growth condition requirements and in terms of subsequent biological characterization. However, both populations differed with respect to their proliferation capabilities at different seeding densities. In this regard, human bone marrow-derived MSCs proliferated more slowly than human placenta-derived MSCs in every experiment. Also the latter had greater long-term growth ability than the former. Moreover, MSCs from both sources exhibited similar morphology, size, and cell surface phenotype, and mesodermal differentiation ability with the exception that human placenta-derived MSC consistently appeared less able to differentiate to the adipogenic lineage. In line with the results obtained, these authors suggested that human placenta is an acceptable alternative source for human MSC.

All published protocols to obtain HAM-derived cells yield hAMSCs, but no studies have compared their efficacy in the isolation. Our group [31] compared two protocols, described in the literature by Alviano et al. [33] and Soncini et al. [59], for the isolation of hAMSCs from the HAM. Alviano’s protocol involved three digestions (one mechanical and two enzymatic), whereas Soncini’s protocol used only two enzymatic digestions. This study included the comparison of hAMSCs, isolated using both methodologies, in terms of their phenotypic characterization and their in vitro potential for differentiation toward osteogenic, adipogenic, and chondrogenic mesodermal lineages. Both protocols allowed the successful isolation and culture of cells attached to the culture flask with fibroblast-like cell morphology from full-term placenta. These cells showed similar immunophenotype but with differences in cell yield and in the in vitro differentiation potential into the main mesodermal lineages. In particular, quantitative studies showed that Soncini’s protocol typically showed an increase in the hAMSCs isolation yield of almost tenfold with regard to Alviano’s protocol. Also, the former protocol allowed the isolation and expansion of a larger number of cells in a very short time period. This ready and rapid availability of cells is one criterion required of a source of MSCs for it to be considered for cell transplantation. Therefore, the differences found using both protocols should be taken into account when using these cells for cell therapy.

3.5 Differentiation Potential of Human Amniotic Membrane-Derived Cells

Placental MSCs have been shown to differentiate into chondrogenic, osteogenic, endothelial, hepatocytic, myogenic, and neurogenic lineages, with some differences among cell types depending on the placental tissue sources [8, 33, 40, 43, 50, 56, 68, 69, 74, 75] (Figs. 3.6 and 3.7).

Fig. 3.6
figure 00036

Adipogenic (DIF Adipo) and osteogenic (DIF Osteo) differentiation of human amnion mesenchymal stromal cells (hAMSCs) and human amnion epithelial cells (hAECs) with their respective controls (C hAMSC and C hAEC) grown for 21 days in Dulbecco’s Modified Eagle Medium (DMEM 21). The presence of adipocytes was assessed by detection of lipid drops using Oil Red O (OR-O) stain (a). The presence of the calcium deposits characteristic of osteoblasts was detected using Alizarin Red (AR) stain (b). Original magnifications: 200× (a) and 100× (b) (Images taken from Díaz-Prado et al. [32])

Full size image
Fig. 3.7
figure 00037

Chondrogenic differentiation (DIF Chondro) of human amnion mesenchymal stromal cells (hAMSCs) and human amnion epithelial cells (hAECs) and their respective controls (C hAMSC and hAEC) grown for 21 days in Dulbecco’s Modified Eagle Medium (DMEM). Micropellets were stained with HE, Masson’s trichrome (MM), and toluidine blue (AT) for proteoglycans. Immunodetection of Agg (Ag-C20) and collagen type II (Col II) was performed to detect molecules characteristic of hyaline cartilage. Immunodetection for collagen type I (Col I) was also assessed. Original magnifications: 100× and 200× (Images taken from Díaz-Prado et al. [32])

Full size image

hAMSCs differentiation to neuronal lineage has been demonstrated by the fact that these cells express neuronal markers (nestin, Musashi 1, neuron-specific enolase, neurofilament medium, microtubule-associated protein [MAP]-2 and Neu-N) and glial (GFAP) markers, after their culture in specific neural-induction media [50, 60, 75, 76].

Tamagawa et al. [74] showed that hAMSCs were able to differentiate into cells with characteristics of hepatocytes. In this regard, native cells expressed typical hepatocytic mRNA such as albumin, CK (cytokeratin) 18, α-fetoprotein, α1-antitrypsin, and HNF-4α, but only glucose-6-phosphatase and ornithine transcarbamylase expression and glycogen storage were observed after in vitro hepatic induction.

Regarding hAMSC differentiation toward mesodermal lineage, In’t Anker et al. [22] demonstrated the potential of hAMSCs to differentiate into osteogenic and adipogenic cells. After osteogenic differentiation, hAMSCs suffered morphologic changes and showed calcium deposits when they were stained with von Kossa’s dye. On the other hand, and after adipogenic differentiation, hAMSCs become multi-vacuolated cells that were stained with Oil Red O stain. Later, Portmann-Lanz et al. [60] showed the capacity of these stromal cells for differentiation to chondrogenic and myogenic lineages. Chondrogenic differentiation of these cells was demonstrated by the presence of abundant collagen in the ECM by means of Alcino’s blue dye. Myogenic differentiation of hAMSCs has been determined by RT-PCR since Portmann-Lanz et al. [60] demonstrated the mRNA expression of myogenic transcription factors such as MyoD and myogenin and the protein expression of desmin in hAMSCs cultured in differentiation media. Alviano et al. [33] confirmed these results and also were the first to demonstrate the angiogenic differentiation potential of these cells. This latter study revealed that hAMSCs, after culture in induction media with VEGF, expressed endothelial-specific markers such as the receptors of the vascular endothelial growth factor 1 and 2 (FLT-1, KDR), ICAM-1, as well as the appearance of CD34 and von Willebrand Factor (vWF)-positive cells.

Regarding cardiomyogenic potential, it has been demonstrated that hAMSCs expressed cardiac-specific genes such as GATA4, MLC-2a (myosin light chain), MLC-2v, cTnI, and cTnT [77, 78] after cardiomyogenic induction. Zhao et al. [77] showed that after hAMSCs transplantation into the myocardial infarcts in rat hearts, these cells survived in the scar tissue for at least 2 months and differentiated into cardiomyocyte-like cells. On the other hand, spontaneous differentiation of hAMSCs toward myofibroblasts has also been observed after their culture in standard medium (DMEM/FBS) within 2 passages [79].

The ability of hAECs to differentiate into cardiomyocytic, myocytic, osteocytic, adipocytic (mesodermal), pancreatic, hepatic (endodermal), neural, and astrocytic (neuroectodermal) cells in vitro has been established [42, 43, 73, 80]. However, in contrast with embryonic stem cells, hAECs did not form tumors up to 7 months posttransplantation in SCID/Beige mice [42, 73]. The capacity of hAECs to differentiate into cell types from all three germ layers may be associated with the fact that the hAECs are directly derived from the epiblast and thus may retain the plasticity of pregastrulation embryonic stem cells.

The pluripotency of hAECs was supported by the study of Tamagawa et al. [81]. The ultimate approach to determine the pluripotency of amniotic epithelium-derived stem cells is to generate chimeric animals by injecting the single stem cell into a blastocyst. If the stem cell contributes all germ layer cells in the chimeric animal, pluripotency will be confirmed [67]. Tamagawa et al. [81] created a xenogeneic chimera with hAECs and mouse embryonic stem cells in vitro. This chimera gives rise to cells of all germ layers, confirming the in vitro pluripotency of hAECs. Later studies have corroborated the ability of hAECs to in vitro differentiate into cells from the three germ layers [42, 43, 73, 80].

hAECs have characteristics of neural progenitor cells since freshly epithelial cells constitutively express a number of neural genes, including neuron-specific enolase (NSE), NF-M, and myelin basic protein (MBP), perhaps suggesting a predilection for neural differentiation [70]. Exposure of hAECs to all-trans-retinoic acid and FGF4 resulted in adoption of an elongated, neural morphology and enhanced expression of some differentiation markers for neural stem such as nestin and GAD (glutamate decarboxylase). Differentiation to astrocyte-like and oligodendrocyte-like cells was also evidenced by expression of glial fibrillary acidic protein (GFAP) and cyclic nucleotide phosphodiesterase (CNP), respectively [73]. Kakishita et al. [82] and Elwan and Sakuragawa [83] demonstrated the differentiation of the epithelial cells to neural cells (ectodermal lineage) with capacity to synthesize and release acetylcholine, catecholamines, neurotrophic factors, activin, noggin, and dopamine, suggesting their possible utility in the treatment of neural degenerative diseases. In this regard, several studies have already been published showing promising results in animal models with Parkinson’s disease and mucopolysaccaridosis type VII. Studies of intracerebral grafting of hAECs for the treatment of a mouse model of Parkinson’s disease showed that these epithelial cells can synthesize and release catecholamine and neurotrophic factors such as nerve growth factor, neurotrophin-3, and brain-derived neurotrophic factor [82, 84, 85]. Kosuga et al. [86] suggested that transplantation of hAECs transduced with adenoviral vectors can be employed for the treatment of congenital lysosomal storage disorders.

Hepatic differentiation (endodermal lineage) of hAECs was reported by Sakuragawa et al. [87]. They demonstrated albumin and α-fetoprotein production from cultured hAECs, and when LacZ-labeled human hAECs were transplanted into the liver of an immunodeficient mouse, the transplanted cells were found to integrate into the hepatic plate. Some reports demonstrated that these epithelial cells also displayed other functional properties associated with hepatocytes, such as glycogen storage and expression liver-enriched transcription factors, such as hepatocyte nuclear factor (HNF)-3γ and HNF4α, CCAAT/enhancer-binding protein (C/EBP-α and C/EBP-β), and several of the drug-metabolizing genes (cytochrome P450) [73, 88, 89]. Some papers showed albumin production and induction of early markers of hepatic differentiation of hAECs after the addition of specific growth factors to the culture media such as FGF-2, hepatocyte growth factor, oncostatin M, and heparin sodium salt [88]. These findings suggest the potential utility of hAECs to restore hepatic tissues that have been diseased or injured.

Differentiation of hAECs to another endodermal lineage, pancreatic, was reported. Wei et al. [90] cultured these epithelial cells in the presence of nicotinamide to induce pancreatic differentiation, and they observed that the treated cells initiated the expression of multiple pancreatic genes, including the transcription factor Pax-6 and the hormones glucagon and insulin. Subsequent transplantation of these insulin-expressing cells in the spleen of diabetic SCID mice normalized the levels of serum glucose for several months after the transplant, indicating the therapeutic potential of hAECs to treat diabetes mellitus type I. Later, Miki et al. [73] showed by RT-PCR analysis that, after pancreatic differentiation, hAECs express pancreatic α- and β-cell markers such as the transcription factors PDX-1 (pancreatic duodenum homeobox 1), PAX-6 (paired box homeotic gene 6), and NKX2.2 (NK2 transcription factor-related locus 2) and the mature hormones insulin and glucagon.

The differentiation of hAECs to cardiac cells (mesodermal lineage) was first evaluated by Miki et al. [73]. They demonstrated by RT-PCR that cardiac-specific genes atrial and ventricular myosin light chain 2 (MLC-2A and MLC-2V, respectively) and the transcription factors GATA-4 and Nkx 2.5 are expressed or induced in hAECs cultured in media supplemented with ascorbic acid 2-phosphate for 14 days. The immunohistochemical analysis of alpha-actinin expression showed a staining pattern very similar to the one reported for hESC-derived cardiomyocytes.

Differentiation of hAECs to another mesodermal lineages was reported by Ilancheran et al. [42], who showed that native hAECs can differentiate into cells with a phenotype and marker characteristic of mesodermal-derived myocytes, osteocytes, and adipocytes.

3.6 Preclinical Studies of Amnion-Derived Cells Applications

There are a limited number of studies showing results of preclinical investigations using amnion-derived cells [91]. New research focusing on alternative therapeutic applications is currently in progress.

Some reports suggested the beneficial effects of primary hECs in lung fibrosis when they were transplanted into a mouse model. After the transplantation, these cells expressed surfactant proteins and displayed lamellar bodies indicating their differentiation into type II pneumocytes in vivo. hECS transplantation reduced collagen deposition, induced its degradation, and overall reduced fibrosis in the injured lungs [92]. Regarding liver fibrosis, hECs transplantation also showed a significant reduction in the number of hepatic cells producing collagen [93].

Cell therapy using hAECs was assessed for the treatment of pancreatic diseases. In rat models of insulin-dependent diabetes mellitus, transplanted hAECs were able to normalize blood glucose level, since they were able to differentiate into pancreatic β-cells in vivo [90]. On the other hand, and for the treatment of muscle diseases, when hAMSCs were transplanted into a mouse model of Duchenne muscular dystrophy, they underwent myogenic differentiation or fusion with host muscle cells [94].

Moreover, the differentiation potential of hAECs into neurons and glial cells was investigated by several groups for the treatment of neurological disorders which affect both the spinal cord and the brain. For example, for the treatment of Parkinson’s disease, hAECs transplanted into an immunosuppressed rat model of Parkinson’s disease produced dopaminergic and other diffusible molecules with trophic and beneficial activities on dopaminergic neurons [82, 84]. In case of ischemic stroke, hAECs transplanted into ischemic rats resulted in an improvement of behavioral dysfunction and reduction of infarct volume. These beneficial effects probably could be due to the hECs differentiation toward neurogenic lineage in vivo and to the paracrine actions of the neurotrophic factors secreted by these amnion epithelial cells [95]. Moreover, hAECs have been investigated to treat spinal cord injury. When these cells were transplanted into a monkey or rat models, hAECs prevented degeneration of axotomized neurons and exerted neurotrophic effects, in part due to the release of neurotrophic factors by hAECs [96].

3.7 Clinical Application of Human Amniotic Membrane as Scaffold

HAM has been reported for the first time as a biological dressing to heal skin wounds a century ago [97]. Davis was the first to report the use of fetal membranes as surgical materials in skin transplantations. Later, other surgery applications for HAM have been reported, such as its use as a biological dressing for skin wound treatment, chronic leg ulcers, and burn injuries. Since the 1940s, the use of de-epithelialized HAM has been well documented in ophthalmology for the treatment of Stevens-Johnson syndrome, cicatricial pemphigoid, acute thermal and alkali burns, pterygium surgery, and limbal stem cell transplantation among others [98–102]. HAMs have also been used as biologic dressings for plastic surgery, dermatology, and gynecology procedures [103–107]. In management of open wounds, HAM provides a clean and closed wound in the shortest time possible; it avoids fluid, nutrient, and heat loss; prevents wound infection and pain; and reduces mobility. The amnion adheres firmly to an exposed surface. Moreover, HAM can provide a healthy new substrate suitable for reepithelization and epithelial healing [47]. These properties enable surgeons to apply the graft on various tissue surfaces without need for suturing or application of secondary dressings. Immediately after grafting, the process of biodegradation begins and the membrane self-dissolves over a period of time from days to 3–4 weeks depending on the characteristics of the wound, the presence or absence of coexisting pathogens, the polarization of the applied graft, and the type of graft applied.

Importantly, full-term placentas are evaluated after the birth of the baby and are discarded at the hospital as medical waste. Therefore, HAMs are inexpensive and easily obtained with an availability that is virtually limitless, negating the need for mass tissue banking [45, 47, 57, 58]. The HAM possesses clinical considerable advantages to make it potentially attractive as a biomaterial. It is antimicrobial, anti-fibrosis, antiangiogenic, and antitumorigenic and has acceptable mechanical properties. It also reduces pain and inflammation, inhibits scarring, enhances wound healing and epithelialization, has analgesic properties, acts as an anatomical and vapor barrier, and modulates angiogenesis, all important requirements for tissue engineering [45]. Several growth factors, such as TGF-β, βFGF, EGF, TGF-α, keratinocyte growth factor, and hepatocyte growth factor, produced from amniotic membrane, are involved in some of these processes [4]. All these characteristics are not shared by other natural or synthetic polymers, highlighting the clinical advantages of HAM as a scaffold compared to other biocompatible products. Also, amnion shows little or no immunogenicity, and the immune response against the graft, if there is, is slight and ineffective, so it does not represent transplantation risks. On the contrary, chorion shows high immunogenicity, and for this reason, it is not used as biomaterial for transplantation purposes. It is important to note that HAM has been approved as a medical material by the Food and Drug Administration [67].

Nowadays, HAMs are used as allograft in general surgery for reconstructions, as an autograft in neonatal reconstruction surgery and as a scaffold in tissue engineering research [48]. The low cost of amnion graft preparation and the very good clinical results in multipurpose applications have made it a viable alternative to other natural (i.e., preserved human skin) and synthetic wound dressings [108]. Moreover, for all the clinical applications, HAM is usually preserved and stored using different methods such as cryopreservation, irradiation, air drying, lyophilization, or glycerol preservation.

3.8 Summary

The HAM, an abundant, inexpensive, and readily obtained tissue that is discarded postpartum, represents a valuable cell and tissue source of great interest in the field of cell therapy and regenerative medicine. Both cell populations isolated from HAM, hAMSCs and hAECs, show an antigen expression profile characteristic of culture-expanded MSCs and differentiation potential into ectodermal, mesodermal, and endodermal lineages. hAMSCs, hAECs, and HAM fragments were used in ­preclinical studies to treat pancreatic, muscle, vascular, lung, and liver diseases. However, more studies are needed to demonstrate the potential effects of either amnion-derived cells or amnion allografts in animal models of different diseases in the hope of increasing their future clinical applications.