Abstract
Regenerative medicine aims to repair and regenerate damaged cells, tissues, and organs in order to restore function. Regeneration can be obtained either by cell replacement or by stimulating the body's own repair mechanisms. Importantly, a favorable environment is required before any regenerative signal can stimulate resident stem/stromal cells, and regeneration is possible only after the resolution of injury-induced inflammation. An exacerbated immune response is often present in cases of degenerative, inflammatory-based diseases. Here we discuss how amniotic membrane cells, and their derivatives, can contribute to the resolution of many diseases with altered immune response by acting on different inflammatory mediators.
Keywords
Introduction
Many adult organs and tissues are known to harbor various populations of stem cells, and among these the human placenta has demonstrated to be a rich resource of different cell populations 1 , such as those from fetal membranes (amnion and chorion)2-5, chorionic villi6,7, umbilical cord8-10, and maternal decidua basalis4,11 (Fig. 1).

The figure depicts human placenta tissues. The placental disc is enlarged to show the fetal membranes (amnion and chorion), which are subdivided into different layers. The amnion is subdivided into epithelial, compact, and mesodermal layers. The spongy layer is at the interface between the amnion and the chorion. The chorion is subdivided into mesodermal and trophoblast layers. The top table inset (A) provides an indication of several markers expressed by human amniotic epithelial cells (hAECs) and human amniotic mesenchymal stromal cells (hAMSCs) after isolation (p0) and between p2 and p4. Expression is presented as follows: 0%-2% (-); 2%-15% (-/+); 15%-30% (+); 30%-70% (++); over 70% (+++). The bottom table inset (B) shows the production of well-known anti-inflammatory factors. EpCAM, epithelial cell adhesion molecule; HLA, human leukocyte antigen; TGF-β, transforming growth factor-β; HGF, human growth factor; PGE-2, prostaglandin E2; IDO, indoleamine 2,3-dioxygenase.
There is a great deal of discussion on how stem/stromal cells isolated from the placenta, or other tissues, can contribute to the regeneration of damaged tissues. One mechanism is by means of cell differentiation into tissue-specific cell types in order to replace damaged tissue12,13. A more recent but widely accepted mechanism is that these cells can act via paracrine signaling, thus releasing bioactive mediators that may stimulate resident target cells to proliferate, or may induce resident progenitor cells to differentiate. In the context of diseases whereby an exacerbated inflammatory activation status persists, switching inflammation off is necessary for the resolution of injury. Since the bioactive mediators secreted by stem/stromal cells could modulate the immune response, a new proposed mechanism is that these cells could favor the repair and regeneration of damaged tissues by suppressing the immune response that is activated following the injury itself.
In this review, we will focus on the amniotic membrane (AM) of human term placenta and its derivatives, such as cells and their produced bioactive factors. The AM is the innermost layer of the amniotic sac in which the fetus develops and is a thin, avascular sheet where epithelial and stromal layers can be distinguished (Fig. 1). Two major cell types can be isolated from the AM, namely, human amniotic epithelial cells (hAECs) and human amniotic mesenchymal stromal cells (hAMSCs) 1 . The consensus from the First International Workshop on Placenta-Derived Stem Cells established the nomenclature and characteristics for the identification of cells from the fetal membranes of placenta 1 . These comprise the specific surface antigen expression observed for hAMSCs, such as the positive expression of CD73, CD90, and CD105 and the low or absent expression of hematopoietic markers and human leukocyte antigen-antigen D related (HLA-DR), and for hAECs, such as the positive expression of CD73, epithelial cell adhesion molecule (EpCAM), and CD166 (Fig. 1, table inset A) 1 . A comprehensive analysis of hAMSC and hAEC phenotypes from isolation (passage 0) up to passage 4 has been recently reported by our group (Fig. 1, table inset A)14,15. We, and others, have contributed to identifying factors produced by hAMSCs and hAECs that are potentially implicated in tissue regeneration. These include immune-modulatory factors important for the resolution of inflammation [e.g., interleukin-10 (IL-10), transforming growth factor-β (TGF-β), hepatocyte growth factor (HGF), prostaglandin E2 (PGE2), human leukocyte antigen (HLA)-G, and indoleamine 2,3-dioxygenase (IDO)] (Fig. 1, table inset B) and growth and angiogenic factors important for tissue remodeling [e.g., epidermal growth factor, angiogenin, growth-regulated oncogene (GRO), vascular endothelial growth factor (VEGF), tissue inhibitor metallopeptidase 1 (TIMP-1), platelet-derived growth factor (PDGF), insulin-like growth factor-1 (IGF-1), IL-6, IL-8, granulocyte colony-stimulating factor (G-CSF), and granulocyte-macrophage colony-stimulating factor (GM-CSF)]16-21.
Evidence supporting the beneficial properties of amniotic cells has increased exponentially in the past decade, and more recently, the notion that these cells act through paracrine mechanisms has been progressively substantiated. This review will attempt to describe molecular and cellular events that could support the beneficial, regenerative-promoting properties observed after transplantation of amniotic cells and/or their bioactive factors. Therefore, we will focus on how amniotic cells could contribute to triggering the regeneration of injured tissue by promoting the resolution of inflammation consequent to injury, both by taming the magnitude of proinflammatory signals and also by favoring anti-inflammatory immune components.
Evidence of the Therapeutic Effects of Amniotic Cells and their Derivatives
Transplantation of human amniotic cells has repeatedly shown that they can favor tissue repair and regeneration in rodent models of inflammatory-based diseases, such as liver fibrosis22,23, lung fibrosis22,24-29, collagen-induced arthritis 30 , inflammatory bowel disease 30 , dextran-induced severe colitis 31 , experimental autoimmune encephalomyelitis [EAE; an animal model for multiple sclerosis (MS)] 30 , traumatic brain injury (TBI)32,33, and cardiac ischemia34-36. Others have shown beneficial effects after transplantation of amniotic cells in fetal models of lung injury, such as bronchopulmonary dysplasia (BPD)-like injury induced by exposure to hyperoxic conditions 37 , ventilation-induced fetal lung injury 38 , or inflammation-induced fetal lung injury generated by intra-amniotic lipopolysaccharide (LPS) injection 39 . Moreover, ovine AECs allotransplanted into sheep with experimentally induced tendon lesions have been reported to stimulate tissue regeneration40,41.
Paracrine actions have been proposed to explain the benefits observed after amniotic cell transplantation despite the absence of transplanted cells in injured tissue. In such cases, it has been suggested that amniotic cells produce factors able to act on surrounding/adjacent resident cells, improving their survival, proliferation, differentiation, and functionality.
For example, we have shown that despite lack of hAMSCs in the brain after systemic injection, hAMSC administration in mice with TBI increased neuronal rescue and vascular density in the injured cortex 32 . In a model of ischemic stroke, others have shown that hAMSCs, besides modulating peripheral and local inflammation, produce factors with antiapoptotic, neurogenic, and angio genic effects 42 . Furthermore, epithelial cells of the AM have been shown to promote cutaneous wound healing by enhancing the migration and proliferation of keratinocytes43,44.
These studies widely demonstrate that the therapeutic mechanisms attributed to amniotic cells are not directly related to the presence of cells within the injury site, and the benefit observed is mainly due to paracrine/secretory actions. In support of this, others demonstrated that the beneficial effects were also achieved when using cell-free treatments, such as conditioned media (CM) containing factors secreted by amniotic cells during their in vitro culture. CM-hAMSC has shown beneficial effects in preclinical models of lung fibrosis45,46 and cardiac ischemia 47 . Recently, CM-hAMSC was shown to accelerate the healing of ulcers in diabetic mice 48 . Moreover, the application of CM-hAMSC on spontaneous tendon and ligament horse injuries was able to significantly decrease the rate of subsequent injuries compared to untreated animals 49 . Others have reported therapeutic effects using CM containing factors secreted from amniotic epithelial cells (CM-hAEC) in corneal alkali injuries in rabbits 50 or in dogs with induced corneal ulcers 51 . Finally, we have previously shown that the release of soluble factors from amniotic cells is associated with the therapeutic effects observed after treatment with patches of the AM in models of liver fibrosis52,53 and cardiac ischemia 54 .
The Contribution of Amniotic Cells and Their Derivatives in Tissue Regeneration
Tissue injury induces the onset of several overlapping molecular and cellular events leading to an inflammatory response. The inflammatory process involves the recruitment of immune cells [including monocytes, macrophages, neutrophils, dendritic cells (DCs), T cells, and B cells], under the control of extracellular molecular regulators (cytokines and chemokines), to the site of injury.
Inflammation is an important defensive response that plays a critical role in the regeneration of injured tissues. Acute, self-limiting, and resolving inflammation is the first step of repair and is essential for a proper restorative process. On the contrary, unresolved, chronic inflammation can lead to excessive tissue damage and deregulated tissue healing, leading to a series of pathologies, including fibrosis and autoimmune diseases.
Amniotic cells and derivatives can participate in the resolution of inflammation by acting on different inflammatory mediators.
The Effects of Amniotic Cells and Their Derivatives on the Inflammatory Phase
Cytokines and Chemokines
Cytokines and chemokines are early modulators of inflammation. They are produced by tissue-damaged cells and by numerous immune cells (subsequently recruited into injury site) and provide molecular cues that orchestrate immune cell migration to the sites of injury and trigger their activation. Reduction in cytokine and chemo kine production could lead to a reduced recruitment of inflammatory cells to sites of inflammation and, consequently, to a constrained/inhibited propagation of inflammation. Administration of amniotic cells has been shown to decrease the levels of proinflammatory cytokines in damaged tissues. For example, in mice with bleomycin-induced lung injury, administration of hAECs, hAMSCs, and CM from hAMSCs significantly reduced fibrosis and chronic inflammation25,26,29,45,55. This was associated with decreased levels of proinflammatory monocyte chemoattractant protein-1 (MCP-1), tumor necrosis factor-α (TNF-α), IL-1, interferon-γ (IFN-γ), and IL-6 and profibrotic TGF-β, PDGF-α, and PDGF-β cytokines/growth factors in the lungs25,26,29,45,55. In line with this, hAEC treatment reduced inflammatory cytokines TNF-α, IL-1β, and IL-6 in inflamed lungs of fetal sheep exposed to intrauterine LPS 39 . Furthermore, topical application of CM-hAEC was associated with inhibition of IL-1β and nitric oxide (NO) production and alleviated inflammation in dogs with corneal ulcers 51 .
Neutrophils
Neutrophils are the first to arrive at the site of repair and are crucial in detecting injury and clearance of the injured cells. Subsequent abrogation of neutrophil influx and activation has a relevant role for the resolution of inflammation since neutrophil-released oxidants and pro teases could result in host tissue damage and perpetuation of inflammation. hAEC administration has been shown to significantly reduce neutrophil infiltration into the lungs27,56. A decreased neutrophil infiltration has also been reported after hAMSC treatment in the joints of mice with collagen-induced arthritis 30 and in the lungs of mice with bleomycin-induced lung injury24,27,56. Furthermore, CM from the AM has been shown to reduce neutrophil survival in vitro by accelerating their apoptosis 57 . Since apoptotic neutrophils release microparticles containing annexin 1, a protein with an antimigratory effect on neutrophils 58 , acceleration of neutrophil apoptosis could also serve as a positive feedback loop further inhibiting neutrophil migration and recruitment.
Antigen-Presenting Cells
Antigen-presenting cells (APCs), such as macro phages and DCs, and lymphocytes from the adaptive immune response are also recruited to the site of injury and contribute to the inflammatory process. Evidence suggests that amniotic cells can counteract the recruitment, maturation, and proliferation of these immune cells. Specifically, transplantation of hAECs in a mouse model of MS has been shown to reduce monocyte/macrophage and T-cell infiltration in the central nervous system with reduction in the progression of clinical disease 59 . A reduction of inflammatory microglia/macrophages has also been observed after hAEC infusion in fetal sheep brains after injury induced by LPS 60 . In vitro evidence has shown that amniotic cells can directly target monocytes and APCs. For example, hAECs, hAMSCs, and their CM can prevent the differentiation and maturation of monocytes toward DCs14,61,62.
T Lymphocytes
Amniotic cells and their derivatives also have an important role in inhibiting T-cell activation/proliferation both in vitro and in vivo. Indeed, it is now widely recognized that when in vitro, hAECs21,63,64, hAMSCs18,64-67, and amniotic cell populations3,68 suppress T-lymphocyte proliferation induced by a variety of stimuli. CM from hAMSCs19,66 and hAECs 69 has been shown to possess antiproliferative effects on activated T lymphocytes, similar to those possessed by their cellular counterpart, thus supporting the fundamental role of paracrine, bioactive factors secreted by these cells in their immunosuppressive activity.
The Contribution of Amniotic Cells and Their Derivatives on the Resolution Phase
Not only is the resolution of inflammation dependent on the removal of inflammatory stimuli, abrogation of cytokine/chemokine production, and prevention of further leukocyte recruitment, but it is also an active process mediated by molecules that promote the return to homeostasis 70 . Anti-inflammatory repair mediators can favor the maturation/induction of immune cells with regulatory activities, including anti-inflammatory macrophages (M2) and regulatory T cells (Tregs), able to promote the switch from the inflammatory phase to the tissue repair phase.
Macrophages
In general terms, polarized macrophages are often referred to as M1 and M2, referring to the two major and opposing activities of macrophages. M1 macrophages orchestrate the inflammatory phase by producing pro inflammatory cytokines, chemokines, and reactive oxygen species to eliminate injury-initiating factors. Afterward, M2 macrophages promote the progression from the inflammatory phase to the tissue repair phase, restoring the damaged tissue architecture and integrity71,72. M2 cells are generally characterized by a low production of proinflammatory cytokines and high production of anti-inflammatory IL-10 73 and exhibit high levels of fibrogenic and angiogenic factors that serve to resolve inflammation and promote wound healing.
Interestingly, different findings provide evidence that macrophages are key players for the downstream immunosuppressive effects of amniotic cells and, ultimately, for the improvement of therapeutic outcome. For example, Murphy et al. demonstrated that hAECs are not able to exert reparative effects when administrated in fibrotic lungs of mice with impaired macrophage function 27 . Moreover, hAEC administration reduced bleomycin-induced lung injury, with a significant decrease in macrophage infiltration into the lungs27,56. Furthermore, while the infiltrating macrophages in untreated disease animals were predominantly M1, the predominant macrophage phenotype in hAEC-treated animals was M228,56. The important role played by macrophages and their switching toward the M2 phenotype in the resolution of the fibrotic process following chronic inflammation have been shown also in other in vivo studies. In fact, transplantation with hAECs was reported to cause a reduction of established hepatic fibrosis induced by a wound healing M2 macrophage phenotype in mice with carbon tetrachloride (CCl4)-induced hepatic fibrosis 23 . More recently, CM from hAMSCs was demonstrated to accelerate wound closure when subcutaneously injected in a mouse model of skin wound healing. Again, the CM-mediated tissue repair was associated with the modulation of resident myeloid populations, characterized by reduction of the inflammatory environment and M1 macrophages and induction of M2 macrophages 48 . In line with this, ovine AECs allotransplanted into sheep with experimentally induced tendon lesions have been reported to stimulate tissue regeneration 40 , which correlated with modulation of the M1/M2 macrophage balance, whereby the presence of proregenerative M2 macrophage phenotype subpopulations correlated to the ovine AEC ability to induce tissue regeneration 41 .
These in vivo immunomodulatory actions of amniotic cells on macrophages have been confirmed in in vitro studies showing the potential of amniotic cells to educate macrophages to adopt an anti-inflammatory/immunosuppressive phenotype. For example, hAMSCs have been reported to skew monocyte differentiation toward anti-inflammatory M2 macrophages when monocytes were differentiated in vitro into DCs or M1 macrophages, even in the absence of any differentiation/polarization stimuli14,48,74. The macrophages generated in the presence of hAMSCs, or their CM, showed increased production of the anti-inflammatory cytokine IL-10 and reduced secretion of different proinflammatory factors. These cells were also poor inducers of T-cell proliferation and inflammatory T helper (Th) 1 cells14,48, and induced the emergence of Tregs with potential suppressive activity 48 .
Regulatory T Cells (Tregs)
Tregs have emerged to be important players in the res olution of inflammation. Tregs are a subset of forkhead box P3 (FOXP3)-expressing CD4+CD25+ T lymphocytes that regulate immune response through the secretion of immunosuppressive and proresolving cytokines, such as IL-10 and TGF-β 75 . Beneficial roles of Tregs have been described in preclinical studies of different inflammatory diseases, such as MS 76 , rheumatoid arthritis 77 , colitis 78 , graft-versus-host disease 79 , and atherosclerosis 80 . A remarkable property of amniotic cells is their ability to induce generation of Tregs, both in vitro and in vivo. In a mouse model of collagen-induced arthritis, hAMSC treatment significantly increased peripheral generation of antigen-specific Tregs with immunosuppressive functions able to prevent arthritis progression when transferred to mice with collagen-induced arthritis 30 . Tregs have also been reported to be required for hAEC-mediated macrophage polarization and subsequent attenuation of bleomycin-induced lung injury 28 . Moreover, in a mouse model of MS, splenocytes from hAEC-treated mice showed increased numbers of Tregs, Th2 (IL-5) cells, and naive CD4+ T cells59,81.
Anti-Inflammatory Factors
Amniotic cells can improve the resolution of the inflammatory process through the modulation of the production of immune-modulatory factors that attenuate inflammatory responses and are important in reestablishing tissue integrity, such as the anti-inflammatory cytokine IL-10. Indeed, gene expression of IL-10 was reported to increase after hAEC infusion in the livers of mice with CCl4-induced hepatic fibrosis22,23. Moreover, draining lymph node cells from a mouse model of collagen-induced arthritis was reported to secrete high levels of IL-10 after amniotic cell infusion 30 . In addition, IL-10 has been shown to be secreted by hAMSCs per se14,19. Other factors produced by amniotic cells with immune-modulatory activity or that have been suggested to enhance restoration of normal organ function include PGE2, HGF, IDO enzyme, HLA-G, TGF-β1 and -β2, and PDGF17-19,59,63,67,82-86. Of particular interest are PGE2, HGF, IDO activity, and HLA-G, which have been described to mediate, at least in part, the immunosuppressive activity of amniotic cells. Indeed, these factors possess pleiotropic immune-modulatory properties often associated with resolution of inflammation, such as the capacity to moderate the immune functions of B cells and proinflammatory macrophages and suppress and restrain the generation and activity of cytotoxic T cells, proinflammatory neutrophils, and natural killer (NK) cells while promoting the development of tolerogenic DCs, M2 macrophages, and Tregs87-94.
Clinical Translation of Amniotic Cells and Derivatives in Immune-Based Disorders
Validation of the therapeutic potential of amniotic cells is featured by the currently ongoing clinical trials investigating their use in immune-based disorders, such as pulmonary fibrosis, liver cirrhosis, Crohn's disease, type 2 diabetes, MS, limb ischemia, systemic lupus erythematosus, and ophthalmologic disorders 95 . These studies are mostly early phase (Phases 0/I/II) and have suggested their safety in a low number of patients. At the time this review was written, the FDA had just approved a Phase III trial for the use of placenta-derived stem cells (PLX-PAD; Pluristem Therapeutics) in patients with critical limb ischemia.
Interestingly, there are several clinical trials studying amniotic cell-secreted factors, such as amnion-derived cellular cytokine solution (ACCS; Stemnion), in conditions such as dry eye, skin burns, and gingivitis. A couple of clinical trials have recently concluded using AM extract eye drops (AMEED) in corneal defects. These trials underline the important role that secreted factors from amniotic cells have in the therapeutic effects 95 . These studies will help define their molecular mechanisms and undoubtedly unveil additional applications in the field of regenerative medicine.
Concluding Remarks
The regenerative process is a complex, dynamic continuum of events in which inflammation plays a crucial role. If properly tamed and switched off, inflammation can lead to resolution; if unresolved, it could evolve into chronic inflammation, which could cause a continuous degeneration of the affected tissues and organs. Alternatively, chronic inflammation could result in the attempt to trigger tissue repair processes that ultimately induce fibrosis, with subsequent loss of tissue function. In this context, the modulation of inflammation is a key element to trigger the restoration of tissue integrity, and amniotic cells and their bioactive factors could foster resolution by dampening proinflammatory signals (cytokines and cells) and enhancing anti-inflammatory immune components.
Footnotes
Acknowledgments
The authors thank the physicians and midwives of the Department of Obstetrics and Gynaecology of Fondazione Poliambulanza-Istituto Ospedaliero, Brescia, Italy, and all the mothers who donated their baby's placenta for the research. This work was supported by Fondazione Poliambulanza-Istituto Ospedaliero, Brescia, Italy; the Cariplo Foundation (Grant No. 2012-0842); Comunità Bresciana (5° Bando 2015 Sostegno ai Giovani Ricercatori); and Contributo MIUR 5×1000 (2013, 2014). A.R.S., M.M., A.C., and O.P. contributed to the writing of the manuscript. O.P. gave the final approval of the version to be published. All authors read and approved the manuscript. The authors declare no conflicts of interest.
