Abstract
The human umbilical cord is a promising source of mesenchymal stem cells (HUCMSCs). Unlike bone marrow stem cells, HUCMSCs have a painless collection procedure and faster self-renewal properties. Different derivation protocols may provide different amounts and populations of stem cells. Stem cell populations have also been reported in other compartments of the umbilical cord, such as the cord lining, perivascular tissue, and Wharton's jelly. HUCMSCs are noncontroversial sources compared to embryonic stem cells. They can differentiate into the three germ layers that promote tissue repair and modulate immune responses and anticancer properties. Thus, they are attractive autologous or allogenic agents for the treatment of malignant and nonmalignant solid and soft cancers. HUCMCs also can be the feeder layer for embryonic stem cells or other pluripotent stem cells. Regarding their therapeutic value, storage banking system and protocols should be established immediately. This review critically evaluates their therapeutic value, challenges, and future directions for their clinical applications.
Introduction
Mesenchymal stem cells (MSCs) are attractive cells due to their capacity for proliferation, multilineage differentiation, and immunomodulatory properties. These cells were first identified and isolated from bone marrow (BMMSCs) and have since emerged as important components in regeneration therapy (53). However, the process of isolating MSCs from bone marrow is complex and painful. As such, derived stem cells from dumped fetal tissue are preferred (9–11).
Embryonic stem cells (ESCs) can differentiate into almost all tissues in the human body and are thus labeled as pluripotent. Recently, induced pluripotent stem cells (iPSCs) have been developed (76) and have pluripotent properties like ESCs. Pluripotency is defined as the ability of these cells to produce tissues from all three germ layers (ectoderm, mesoderm, and endoderm) when transplanted into immunodeficient mice. However, the use of ESCs generated from surplus embryos has raised ethical concerns. Moreover, the clinical applications of iPSCs have been criticized because of the possibility of forming tumors by integrated oncogenes, particularly c-myc (51), by insertional mutagenesis (28), or by disrupting tumor suppressor genes (3). Epigenetic memories and genomic aberrations in the reprogrammed cells have also been noted (26). Thus, in the manufacturing of ESCs or iPSCs for clinical applications, precautions regarding safety and efficacy need to be taken. Finding another useful source of stem cells is imperative.
Recently, two papers have drawn attention to the so-called vampire therapy (63,70). They used blood from young mice and transplanted it into old mice. The young stem cells in the blood can help the aged mice recover their muscle and neuron functions. The growth differentiation factor 11 (GDF11) and cAMP response element binding protein (CREB) were shown to be responsible for this effect.
To date, younger stem cells include fetal stem cells obtained from amniotic fluid, the umbilical cord (UC), and placenta (4,9,66). Fetal stem cells are mostly MSCs, and fetal MSCs from birth-associated tissues are gaining popularity. MSCs derived from the umbilical cord can be obtained from the amniotic membrane, cord lining, Wharton's jelly, and perivascular region (Fig. 1). They are the focus of this review (Fig. 2).
Diagrammatic illustration of cross-section of the human umbilical cord showing different compartments (cord lining, Wharton's jelly, and perivascular region) from where stem cells can be derived.
Diagrammatic illustration of this review of human umbilical cord mesenchymal stem cells, including three germ layer differentiation, antitumor effect, immunomodulation, and clinical application.
Stem Cells Derived from Different Parts of Umbilical Cord
The human umbilical cord starts developing on the fifth week of gestation and continues to grow until 50 cm in length (75). Stem cells can be derived from various parts of the umbilical cord. All of these compartments have been described in the literature, including Wharton's jelly, cord lining, and the perivascular region.
Wharton's Jelly
Most studies use UC MSCs from Wharton's jelly (2,9,74). Regarding isolation, there are two kinds of methods: the explant method and the enzymatic digestion method (50). In the explant method, the Wharton's jelly is manually minced into 1–2-mm3 fragments after removal of UC vessels. The fragments are undisturbed for 7 days to allow the stem cells to come out (9). However, the downside of the explant method is that the fragments often float in the medium. Moreover, this method may not provide a consistent number of MSCs.
In the enzymatic digestion method, the enzymes used for digestion vary from collagenase to a combination of collagenase and hyaluronidase with or without trypsin (12,59,65). This method can provide more homogenous cell populations and more consistent cell numbers compared to the explant method.
Cord Lining
Umbilical cord lining cells have been isolated by the explant method (25,56,57). Two kinds of cells, MSCs (CLMCs) and epithelial cells (CLECs), can be isolated from cord lining. The CLMCs are isolated from the subamnion region by dissecting out the Wharton's jelly. Pieces of the outer envelope membranes are cultured with Connaught Medical Research Laboratories (CMRL) 1660 containing L-glutamine and 10% fetal bovine serum (FBS) (33). Around 20 million cells can be generated at passage 1 (38).
The CLECs can be used for treating persistent corneal epithelial defects and as a skin cosmetic improvement, whereas CLMCs have been used for burn and diabetic ulcer wound healing (38). Since preclinical studies reveal successful disease treatment, further exploration of the utility of these cells is warranted.
Biomarker of Hucmscs
Surface Markers
MSCs are positive for cluster of differentiation 29 (CD29), CD44, CD90, CD73, CD105, and human leukocyte antigen (HLA)-ABC. Conversely, MSCs are negative for the endothelial cell marker CD31; hematopoietic cell markers CD34, CD45, and CD117; and HLA-DR (14,62). The surface markers of HUCMSCs are similar to those of MSCs (31), but they are negative for CD133 (31). Nevertheless, phenotypic characterization of HUCMSCs may be influenced by the culture passage number, medium, and method.
Embryonic Stem Cell Markers
Octamer-binding transcription factor 4 (Oct4), Nanog, sex-determining region Y box 2 (Sox2), and Kruppel-like factor 4 (KLF4) are expressed only at low levels in HUCMSCs (27), suggesting that MSCs are primitive stem cells, between ESCs and adult mature cells. Nonetheless, the isolation of pluripotent MSCs using specific markers remains a challenge. HUCMSCs can also express stage-specific embryonic antigen 4 (SSEA4) in medium supplemented with FBS, whereas SSEA3 is conversely correlated (29). The gene profile of HUCMSCs is reported to be close to those of ESCs (30).
Differentiation Capability
Mesoderm
Adipocytes
HUCMSCs can produce small lipid vacuoles, whereas BMMSCs produce more mature adipocytes (48). HUCMSCs can undergo more than 40 passages (12), and they maintain their multipotency for longer periods compared to BMMSCs (18). Recently, HUCMSCs were shown to be able to differentiate into adipocytes using different induction chemical combinations (58). The presence of indomethacin greatly enhances their adipogenic potential beyond that of rosiglitazone.
Osteocytes
Regarding osteogenic differentiation, HUCMSCs show delayed and insufficient differentiation into osteocytes (30). HUCMSCs from Wharton's jelly present defective osteogenesis ability (31). In contrast, another work demonstrated that HUCMSCs have the best osteogenesis differentiation when obtained from different regions of the UC (48). Recently, HUCMSCs have shown better osteogenesis than stem cells derived from periodontal tissue (32,77).
Cartilage
Regarding chondrogenic differentiation, HUCMSCs reveal the same potential regardless of the portion of the UC from which they were isolated (48). Moreover, the chondrogenic potential of HUCMSCs is thrice that of BMMSCs in producing collagen (18). An unpublished study also reveals that HUCMSCs can effectively repair the monoiodoacetic acid (MIA)-injured cartilage via a decrease in the interleukin (IL)-1β levels induced by MIA treatment. Nevertheless, HUCMSCs possess chondrogenesis compared to stem cells derived from the infrapatellar fat pad (13).
Ectoderm
A previous study reveals that HUCMSCs can differentiate into neurons, astrocytes, and glial cells and can rescue the stroke rat model via an increase in β1-integrin and neurotrophic factors (9,41). When HUCMSCs are exposed to rat neuronal conditioned medium, they differentiate into microglial cells, generate neuronal proteins, and upregulate the astrocyte protein glial fibrillary acidic protein (GFAP) (21,42). Several neuronal differentiation protocols have been published (19,49,71). A 5–6-h exposure to neuronal induction chemicals like potassium chloride, valproic acid, forskolin, hydrocortisone, and insulin can result in long-term neuronal differentiation (19,20).
Endoderm
Liver
HUCMSCs for rescue of liver fibrosis has been previously reported (2,40,67). Firstly, HUCMSCs can express hepatic markers and differentiate into hepatocyte-like cells in vitro and in vivo (2). HUCMSCs transplanted into the CCl4-injured rat model have proven to be able to rescue liver fibrosis (67) and differentiate into hepatocytes in a chemically injured liver rat model (40). An unpublished study also reveals that HUCMSC-differentiated hepatocytes can engraft successfully into injured liver.
Islet Cells
HUCMSCs can differentiate into insulin-producing cells in vitro (68,72). Using a portal vein injection, HUCMSC-differentiated islet cells can alleviate hyperglycemia in diabetic rats (68) and mice (72).
Immunomodulation
The HUCMSCs that show an absence or low expression of major histocompatibility complex (MHC) class II and costimulatory molecules may be considered immunoprivileged cells (8). HUCMSCs can alter immune cell function by inhibiting T-cell, B-cell, and natural killer (NK) cell proliferation and by steering monocytes and dendritic cells to an immature state (16,69).
The immunomodulation of HUCMSCs on T-cells, B-cells, NK cells, and dendritic cells has been comprehensively reviewed previously (15).
Nonclassical type I HLA molecules are interesting but only partly explored in HUCMSC function. HUCMSCs express the HLA-G molecule, at both the mRNA and protein levels and in its soluble form, HLA-G5 (34,60,74). The HLA-G6 isoform is also reported to be involved in the HUCMSC immunosuppressive function (34). HLA-G5 is involved in the induction of regulatory cells (37) and suppression of NK cell production of interferon (IFN)-γ (61). Since fetal expression of high levels of HLA-G isoforms can inhibit maternal alloreactivity, the HLA-G isoforms need to be evaluated in detail (54).
In addition, HUCMSCs can express HLA-E and HLA-F, both of which are implicated in the tolerogenic process occurring in the fetal–maternal interface, along with HLA-G (35). A recent study also demonstrated HLA-G5 expression in all four HUCMSC cell lines responsible for decreasing lymphocyte proliferation during mixed lymphocyte reaction. HUCMSCs, but not BMSCs or placenta MSCs, can maintain low HLA-DR expression under IFN-γ stimulation (in revision).
Banking
Banking of umbilical cord blood is common worldwide, including public and private banking services (45). Similarly, banking of HUCMSCs can also fulfill the need for transplantation purposes. HUCMSCs are an alternative source of stem cells for patients seeking an unrelated donor. Advantages of HUCMSCs compared to other sources of stem cells include procurement, less stringent requirements for HLA matching, reduced graft-versus-host disease (GVHD), and improved access to transplantation. The differentiation of HUCMSCs is better than that of umbilical cord blood cells, not only due to restricted hematopoietic lineage but also for differentiation into other germ layer lineages.
There are several cord blood banks that include storage of MSCs derived from Wharton's jelly and placenta in Taiwan. The potential of banking MSCs is full of hope for the future. Nonetheless, building standard isolation and freezing–thaw protocols is necessary. Deriving stem cells from frozen umbilical cord is not possible (6). A fresh UC tissue fragment is necessary for deriving stem cells. For the isolation of stem cells in good manufacturing practice (GMP) laboratories, several tests should be performed before using these products. These include checking the sterility of cells by 14-day microbial culture protocol, Gram staining to rule out bacterial infection, culture method to rule out mycoplasma infection, and checking cell surface marker and differentiation ability of stem cells.
The endotoxin level and viability after thawing must also be checked. Cell counts would also need to be performed (24). A banking system using the above procedures has been established, with 70 cell lines and HLA typing in a GMP laboratory, for transplantation purposes.
Feeder for Human Embryonic Stem Cells
The HUCMSCs can be a feeder layer for ESCs, with a nontumorigenic effect (12). Such nontumorigenic effect may be caused by the downregulation of the c-myc signaling (12). There are also various fetal stem cells that can also act as a feeder layer, such as stem cells derived from the placenta and amniotic fluid (7,36). Human feeder cell layers would eliminate the xenoprotein (n-glycolyneuraminic acid) (46) caused by using mice fibroblast feeder layers.
Anticancer Effect of Hucmscs
Most studies using HUCMSCs for anticancer research are on solid cancers. One recent report is on soft cancer.
Solid Cancer
Human breast cancer cell line MDA MB-231 (derived from a 51-year-old female) is the most tested cell line (1,5,43,64). An in vivo study has revealed that three weekly intravenous injections of HUCMSCs can attenuate tumor growth (1). Another study used a rat breast cancer cell line to test the anticancer effect of rat umbilical stem cells (rUSCs) (22). Intravenous or intratumor injection of rUSCs can attenuate tumor growth. Moreover, complete tumor regression can be achieved after 1 month, and this was maintained for more than 3 months (22).
Chao and coworkers reported using HUCMSCs to treat MDA MB-231 cells. The tumor attenuation effect is dependent on cell–cell contact and internalization (5). This phenomenon may be present both in vitro and in vivo. Another study used umbilical cord blood (UCB) MSCs to treat the MDA MB-231 cell line (64). They found that dickkopf (DKK1) secreted by UCBMSCs can inhibit tumor growth via the phosphatase and tensin homolog (PTEN) pathway. Ma et al. reported the use of HUCMSCs to treat the breast cancer cell lines MDA MB-231 and MCF-7 (69-year-old female) (43).
Cancer stem cells (CSCs) are sorted using the surface markers of epidermal surface antigen positive (ESA+), CD44+, and CD24-/low. The CSCs reveal a high degree of apoptosis when cocultured with HUCMSCs. HUCMSCs also decrease the xenograft tumor size, and the phosphoinositide-3-kinase (PI3K) and Akt signaling pathways play roles in this anticancer effect. Another study has used MDA-231 and TOV-112D cells (latter is from an endometrioid carcinoma of a 42-year-old female) as a tumor formation model (23). Conditioned medium and cell lysates of HUCMSCs have been used for treating cancer cell lines. Both the conditioned medium and cell lysates can inhibit tumor cell proliferation and make them stay at the sub-G1 going to the apoptosis stage. Upregulation of BCL2-associated X protein (BAX) and downregulation of B-cell CLL/lymphoma 2 (BCL2) and survivin genes are observed. Autophagy genes are also upregulated upon treatment. The conclusion is that HUCMSCs possess tumor inhibitory properties via secretory agents.
Aside from breast cancer, lung cancer is the second most studied cancer. Maurya et al. reported using rUSCs to inhibit murine lung adenocarcinoma (47). rUSCs can attenuate the proliferation and colony formation of cancer. A coculture experiment revealed that most cancer cells stay at the G0/G1 phase, while cyclin A and cyclin-dependent kinase 2 (CDK2) downregulation is noted.
Treatment with rUSCs can decrease tumor size and weight. Homing of rUSCs to the tumor site is also noted. Another study reported by Rachakatla et al. describes the use of HUCMSCs modified to express the human interferon β gene to treat a MDA-231-formed lung tumor (55). Following intravenous injection of HUCMSCs, the cells can migrate to the lung tumor site. The engineered HUCMSCs significantly reduced the MDA-231 tumor burden.
The preliminary results of another study reveal that HUCMSCs have effects on SKOV3 cells (ovarian cancer cell line) via suppression of the cyclin-dependent kinase inhibitor 1/retinoblastoma (P21/Rb) signaling pathway (unpublished data).
Nonsolid Cancer
Recently, HUCMSCs have been used to treat Burkitt's lymphoma (39). Cell proliferation, viability, and death of lymphoma cells were significantly inhibited after 48 h of exposure to HUCMSCs or its extracts, suggesting that HUCMSCs secreted molecules that inhibited lymphoma cell growth via oxidative stress pathways.
Safety
In early phase MSC preclinical and clinical trials, the safety of transplanted MSCs is well documented in animal models and in human trials. However, in vivo efficacy is controversial in humans (52). Moreover, HUCMSCs have been injected intravenously into nonhuman primates to test safety (73). Cells have been injected once every 2 weeks for 6 weeks into cynomolgus monkeys. No stem cell transplantation-related toxicity has been reported, and all injection sites and organs studied are normal, with no tumor noted.
Furthermore, HUCMSCs injected into xenograft disease rat models result in good engraftment and functional outcome. No immunorejection or tumorigenesis have been noted (17). Further long-term in vivo studies must be conducted to assure the safety of MSCs and can increase the homing and therapeutic efficacy of transplanted MSCs derived from adult tissues.
Future Applications
Cell-Based Therapy
The HUCMSCs can be used for specific cell-based therapies (15,17). Preclinical validation of HUCMSCs or their derived tissue in disease models have been reviewed (17). In all of these studies, the HUCMSCs can be differentiated and engrafted with successful functional outcome in vivo in rat models for cerebral ischemia (9), Parkinson's disease, Alzheimer's disease, multiple sclerosis, retinal disease, type 1 and type 2 diabetes, and myogenic disease. Immunity is always a hazard for transplantation. The HUCMSCs have low immunity and immunomodulatory effects that can increase the survival of transplanted cells and decrease the risk of GVHD (44).
Anticancer Therapy
Many studies have proven that HUCMSCs have an anticancer effect. Therefore, it is necessary to perform clinical trials to determine the real effects on tumor abolishment. Dose–time studies on larger nonhuman primates are also warranted. The administration of GMP-grade HUCMSCs to shrink the tumor first, followed by chemotherapy or surgery, can be done. CSCs are currently a hot topic. Whether HUCMSCs can target CSCs is an interesting study for further investigation.
HUCMSC Banking
The HUCMSC banking system and protocol is important. The establishment of a banking protocol and various microbial testing warrants further studies. The simultaneous storage of UCBs and HUCMSCs has already been performed in some cord blood banks in Taiwan. Storage of HUCMSCs provides an opportunity for use in various applications.
The pros and cons of each application are listed in Table 1.
Pros and Cons of Each Application
Conclusions
The use of HUCMSCs has many attractive advantages, including a noninvasive collection procedure, low risk of infection, nontumorigenesis, multipotency, and low immunogenicity. But whether HUCMSCs are the best for clinical use is not yet known. Nevertheless, the era of clinical use of HUCMSCs has arrived and has full potential.
Footnotes
Acknowledgments
The authors sincerely thank the National Science Council of the Republic of China, Taiwan, for financially supporting this research (Contract No. NSC 98–2314-B-303 -009 -MY3) and the Buddhist Tzu Chi General Hospital for the financial support (Contract Nos. TCRD99–12 and TCSP98–07). The authors declare no conflicts of interest.