Abstract
Stem cell transplantation is a potentially important means of treatment for a number of disorders. Two different stem cell populations of interest are mononuclear umbilical cord blood cells and menstrual blood-derived stem cells. These cells are relatively easy to obtain, appear to be pluripotent, and are immunologically immature. These cells, particularly umbilical cord blood cells, have been studied as either single or multiple injections in a number of animal models of neurodegenerative disorders with some degree of success, including stroke, Alzheimer's disease, amyotrophic lateral sclerosis, and Sanfilippo syndrome type B. Evidence of anti-inflammatory effects and secretion of specific cytokines and growth factors that promote cell survival, rather than cell replacement, have been detected in both transplanted cells.
Keywords
Introduction
There are currently no effective treatments for a number of neurological disorders and diseases. One possible avenue of research is stem cell transplantation. In this review we will be focusing on our current research with two different stem cells: human umbilical cord blood stem cells (hUCBCs) and human menstrual blood-derived stem cells (MenSCs). Both of these cell types have shown some degree of success in treating animal models of stroke and hUCBCs have been used in other related disorders with a single transplant. We will also be discussing the use of multiple injections, which has shown some degree of success in animal models of Alzheimer's disease (AD) and Sanfilippo Syndrome type B.
The advantage with these two stem cell types is that they can both be acquired relatively easily and ethically compared with other stem cells such as embryonic stem cells and neural stem cells. hUCBCs are limited to collection at the time of birth, whereas MenSCs could theoretically be collected once a month for approximately 40 years from reproductive women.
Human Umbilical Cord Blood Cells (hUCBCs) and Stroke
The first use of UCB cells to treat any disorder occurred in 1972, when Ende et al. successfully treated a 16-year-old male with acute lymphoblastic leukemia (24). Since then, UCB has been used to successfully treat a number of hematopoietic disorders in babies, children, and/or adults, including Fanconi's anemia, aplastic anemia, beta-thalassemia and sickle cell disease, hemoglobinopathies, and acute leukemia (32,43,54,63,89, 91). Examples of some of the potential disorders that UCB and MenSCs may treat are shown in Table 1. Ende et al. also transplanted large doses of UCBCs by an intraocular route for the treatment of animal models of neurodegenerative diseases such as AD, amyotrophic lateral sclerosis (ALS), Huntington's disease (HD), and Parkinson's disease (PD) (16,25—27,30). They observed a significantly increased survival rate for these animals.
Some of the Disorders That Could Be (or Already Are) Targets for Treatment by Umbilical Cord Blood or Menstrual Blood-Derived Stem Cells With Select References
In the last 10 years, the ability of hUCBCs to treat stroke, spinal cord injuries, and metabolic disorders such as Sanfilippo Syndrome Type B [Mucopolysaccharidosis (MPS) type III] has also been explored, along with further exploration of treatments for AD and ALS by a number of researchers.
The majority of stroke studies show limited entry of hUCBCs into the brain following an IV or intra-arterial transplant (9,105,107,112). Nevertheless, behavioral improvements and reduced infarct size were observed in most studies, although there are exceptions (66,80,112). It is worth noting that in the papers that did not observe any improvement, Wistar rats aged 3—4 months were used, compared with younger Sprague-Dawley rats (2 months old) in the majority of the other studies.
The limited presence of IV transplanted hUCBCs within the brain suggests that their mode of action does not relate to cell replacement. Instead, modification of inflammation, angiogenesis, and neurogenesis has been proposed by a number of studies (5,18,76,77,82,83, 85,88,97,103). We have recently shown that hUCBCs can directly affect cell survival and modify innate inflammation (44), as well as reduce microglial survival (50) and recruitment of proinflammatory cells (61). In addition, preliminary data from Womble et al. (106) suggest that the CD14+ monocytes and macrophages and the CD133+ stem cells may be the most important cell types within cord blood, with respect to the beneficial effects observed in the stroke model. Transplantation of hUCBCs with these fractions excluded had greatly reduced improvement. These cells possess anti-inflammatory properties and are proangiogenic. Therefore, these cells could play an important role in the repair of ischemic tissue (93).
Menstrual Blood Cells (MenSCs) and Stroke
In the last few years, the potential of menstrual blood and endometrial cells has been realized as a source for stem cells. There have been two approaches to obtain these cells, namely 1) collection of cells directly from intact endometrial tissue (i.e., hysterectomy) (14,81) and 2) collection of menstrual blood. The collection of menstrual blood to obtain stem cells has been implemented with unique collection and processing techniques by several other researchers over the last 3 years (17,21, 48,69,72,84,95). The collection and processing methods may render biologically distinct and dynamic cell populations in menstrual blood with stem cells being derived from the endometrial lining and the blood itself. These cells have been shown to be multipotent or even pluripotent cells, as they can be made to differentiate into a number of different cell types, including cardiomyocytes, adipocytes and osteogenic and neural cells. Due to their ability to become cardiomyocytes, Hida et al. (48) explored how menstrual blood-derived stem cells could treat heart disease and found a reduction in the size of the myocardial infarct in a nude rat model of myocardial infarction.
Because MenSCs can be relatively easily obtained and are a rich source of stem cells with the potential to become neurons, they may prove to be an ideal tool for the treatment of neurodegenerative disorders. Consequently, Borlongan et al. (10) used the cells characterized by Patel et al. (84) in in vitro and in vivo models of stroke. They first showed that these cells offered some protection to primary neuronal cultures against oxygen-glucose deprivation. Using the MenSC media, they showed that factors secreted by the cells were able to exert a similar neuroprotective effect to the cells themselves, and this may relate to the cells secreting vascular endothelial growth factor (VEGF), brain-derived growth factor (BDNF), and neurotrophin-3 (NT-3) into the media. Previous studies by us and other groups with other stem cell lines have shown release of one or more of these factors (31,57,76,77,86,109) and their potential benefits for the treatment of stroke (96,104).
In a transient MCAo animal model of stroke, Borlongan et al. (10) observed neuroprotection and behavioral improvements following either IV or intracerebral (IC) injection of these cells 2 h after lesioning. Using a human-specific antibody, transplanted cell survival was observed within the penumbra of both IC (~15%) and IV (~1%) transplanted animals 14 days later. These cells were also shown to have maintained their stem cell phenotype as they continued to express OCT4. This, along with the similar degree of neuroprotection in IC and IV transplanted animals despite the ~15-fold difference in cell survival observed between the two routes, again suggests that cell replacement is not the mode of action in this situation. The cells' ability to secrete specific growth factors (at least in vitro) may mean that this is their mode of action. The secretion of these growth factors promotes the survival of endogenous neural cells. This study certainly provides encouraging data for the potential use of these cells for the treatment of stroke. Elucidation of the optimum dose and time of delivery would provide further support for these cells, as well as determining their mode of action. This is good news for a sizeable proportion of the world's population as autogeneic transplants could theoretically be performed—although unless a later time period proves to be effective, the patient would need to have banked the cells prior to the stroke to ensure that a sufficient quantity of cells are available for transplant. Of note, no immunosuppression was used in this study, which suggests that these cells are immune immature in a similar way to hUCBCs and therefore may also be available for allogeneic transplants.
hUCBCs and Alzheimer's Disease (AD)
The detrimental role of inflammation in a number of neurodegenerative disorders has been proposed. Because hUCBCs have been shown to promote the anti-inflammatory T helper 2 response in a stroke model, along with neuroprotection (102), it is therefore possible that these cells may also prove to be effective in other disorders that involve inflammation. One of the major pathologies of AD, the deposition of amyloid β (Aβ), is a potent activator of the central nervous system's resident immunocompetent cells, the microglia, and astrocytes. This leads to an underlying proinflammatory state and neurotoxicity (8,67,92,101). A key component of this cascade is the interaction of the CD40 ligand (CD40L) with its CD40 protein (99). Increased expression of both CD40 and CD40L, particularly in association with plaques, has been detected in AD and animal models (13) and the soluble form of CD40L was found to be elevated in the bloodstream of AD patients (22).
Interestingly, interruption of CD40-CD40L interactions reduced Aβ accumulation and cognitive deficiencies in animal models of AD (100). Consequently it is likely that hUCB, which can modify the inflammatory response, could exert beneficial effects in animal models of AD.
Ende et al. showed that injection of massive doses (110 × 106) of hUCBCs prolonged the survival of animals transgenic for a human amyloid precursor protein mutation (tg2576 mice) (27). AD is a chronic neurodegenerative disorder; therefore it is likely that multiple injections would prove to be more successful than a single injection and pilot studies by Jun Tan's research group supported this idea. They therefore performed multiple injections of 100,000 cells into two different transgenic mouse models of AD [tg2576 and the double transgenic presenilin APP mouse (PSAPP)] from 7 to 13 months of age (79). For the first 2 months, IV injections were given every 2 weeks, and then once a month for the remaining 4 months, for a total of eight injections. They observed a significant decline in cerebral Aβ pathology in both transgenic animal models at 13 months of age following hUCBCs. Soluble and insoluble Aβ peptides were also shown to be decreased in brain extracts. In a separate study, cerebral amyloid angiopathy (Aβ deposition in the cerebral vasculature) was found to also be significantly reduced in mice aged 12—18 months. CD40+ microglia and reactive astrocytic association with Aβ deposits was depressed, suggesting a reduced inflammatory response, and plasma Aβ was increased in an inverse correlation with soluble CD40L levels.
Tan et al. (98) had previously shown that this was related to an increased brain-to-blood clearance of Aβ. An Aβ phagocytic assay of adult microglia from hUCBC- and vehicle-treated mice revealed a significantly elevated phagocytosis of Aβ by hUCBC-treated microglia, further supporting the suggestion of increased Aβ clearance. The ratio of IgM to IgG provided further evidence for functional suppression of the CD40 signaling pathway, because the CD40-CD40L interaction is necessary for IgM to IgG switching. CD40 knockout from PSAPP mice has previously been shown to ameliorate AD pathology (58), and so Nikolic et al. (79) used a CD40 knockout PSAPP transgenic mouse and demonstrated that no benefit of hUCBC transplantation was observed with respect to the plasma Aβ levels, confirming that this effect relates to the CD40-CD40L interaction. As expected, removal of the CD40-CD40L interaction was found to lead to increased anti-inflammatory cytokines within the brain and plasma, whereas proinflammatory cytokines were reduced. By using serum from hUCBC- and vehicle-treated mice to treat primary microglial cultures, they provided evidence of a secretory factor from hUCBCs, causing the CD40 inhibition. Nikolic et al. (79) also found that the sera from hUCBC-treated mice could promote Aβ phagocytosis by neonatal microglia and peripheral macrophages.
Interestingly, the investigators found no evidence of hUCBCs within the brain, suggesting that the cells do not need to cross the blood—brain barrier to exert their benefit, providing further support for a secretory element, which may relate to the observed shift from proinflammatory to anti-inflammatory cytokines within the brain and plasma. It is unclear whether it is the hUCBCs themselves or the recipient's response to the cells that results in the secretion of a factor that reduces CD40—CD40L interactions and enhances Aβ clearance, but determination of the factor(s) involved and their source could prove invaluable in finding treatments for AD.
Recent evidence suggests that while microglia are triggered to secrete proinflammatory factors by Aβ (19, 45,51,87), under certain conditions (such as adult stem cell transplantation) they can become “alternatively activated” and adopt an anti-inflammatory action (59,60, 81). IL-4 is associated with the alternative activation of microglia (20), and Nikolic et al. (79) saw increased IL-4 in the hUCBC-treated rats. This may suggest that not only was microglial phagocytosis of Aβ enhanced, but they may also have switched to an alternative activation whereby anti-inflammatory activity is dominant. Butovsky et al. (12) showed that the alternative activation of microglia enhanced Aβ clearance, reduction of plaques size, and promoted neurogenesis and this is therefore an interesting potential mechanism for hUCBCs to function by. In addition, the interaction between fractalkine, expressed by neurons, and its receptor CX3CR1 expressed by microglia, has been shown to modulate microglial activation and is generally anti-inflammatory, suppressing excessive activation (14). Microglial activation is normally neuroprotective, but under excessive activation it can become neurotoxic and suppress neurogenesis (3,7,42). However, there is an age-related loss of fractalkine, meaning that microglia are more likely to become activated with time and this could become excessive and eventually cause neural damage (6). In addition, the fractalkine—CX3CR1 signaling has been shown to regulate neurogenesis, with exogenous fractalkine being able to restore the age-related decrease in neurogenesis observed in aged rats (6). Modification of this signaling pathway has also been shown in APP transgenic mice (23) and in AD patients (55). Restoration of the fractalkine—CX3CR1 interaction could therefore be beneficial and it would be interesting to see if hUCBCs and other stem cell transplants can influence this pathway, as this could be a mode of action for the promotion of neurogenesis.
Further studies to elucidate whether hUCBCs, and which specific fraction of the cells contained therein, can elicit cognitive benefit and reduce the formation of hyperphosphorylated tau, the second hallmark of AD, could provide valuable information in creating potential treatments for AD.
hUCB and Amyotrophic Lateral Sclerosis (ALS)
ALS is another disorder that could benefit from hUCB transplantation. Again, there is considerable evidence that an inflammatory response may be involved, with several groups having shown autoimmunity in both animal models and patients of ALS (2,4,53,78). Ende's group published a couple of papers showing significant benefit in an animal model of ALS (the SOD1 transgenic mouse) with respect to survival following intraocular transplant of large numbers of hUCBCs on 2 consecutive days (16,30). Since then, Garbuzova-Davis et al. (39) demonstrated that IV administration of a much smaller single dose of the mononuclear cell fraction from hUCB into presymptomatic ALS mice increased survival and delayed symptom onset. Garbuzova-Davis's group performed additional studies to optimize the dose required in the G93A SOD1 transgenic mouse model of ALS and found that 25 × 106 cells given IV to 7—8-week-old mice was optimal when the animals were immunosuppressed. They found that some cell-transplanted animals survived to 25 weeks, compared with no animals surviving longer than 20 weeks in the media-injected group (36). Symptoms such as weight loss and motor deficits also developed 2—3 weeks later in the cell-treated mice. Cytokine profiles revealed reduced proinflammatory cytokine expression (e.g., IL-1, TNF-α) in the lumbar spinal cord. By looking at the presence of human Th1 and Th2 cytokines in the plasma of the mice, they observed more Th2 cytokines and hence an anti-inflammatory response in hUCB-treated animals. A significantly reduced number of microglia was detected in mice treated with the optimal number of hUCBCs compared with untreated mice and those given a different dose. Additional studies have shown that this relates to activated microglia, supporting the anti-inflammatory action of hUCBCs (70,71). As well as changes in microglia, the peripheral number of lymphocytes was increased, whereas neutrophils were decreased in mice given the optimal dose, suggesting that the “immune defenses” have been upregulated as well as an anti-inflammatory effect (36).
In contrast to the stroke and AD transplants, a considerable number of hUCBCs were detected within the spinal cord. These cells were nestin positive and therefore appeared to be undifferentiated and so, as with the stroke and AD studies, the cells do not appear to be acting by replacing the dead or dying motor neurons (36). The greater presence of administered hUCBCs within the spinal cord may relate to the 10-fold or greater number of cells administered IV in this study, or it suggests that there is greater homing in ALS than in stroke. Another possible explanation is the compromised blood—brain/spinal cord barrier in ALS (37), which may allow greater entry of these cells into the CNS. A high dose of cells, however, still appeared to be necessary to get the maximum benefit. However, because ALS is a chronic disease, smaller multiple injections may prove to be better than a large single one, and this is currently being explored.
The impairment of the blood—brain barrier provides another potential mode of action for hUCB, which contains a high number of endothelial cell precursors. In vitro, it has been shown that endothelial cells could potentially repair or replace the impaired endothelial cells and astrocytes that form the blood—brain barrier (65). This is an additional possible mode of action for hUCBCs, because endothelial precursors also express nestin and transplanted cells are frequently seen in association with blood vessels.
As with many disorders, astrocytic and microglial activation is a hallmark of ALS. By removing the mutated SOD1 gene from astrocytes via a Cre-lox expression system, Yamanaka et al. (108) demonstrated that disease onset was not affected, but microglial activation and disease progression were delayed, suggesting that astrocytes may have an important role in ALS, where they act to promote the inflammatory response of microglia and possibly mediate excitotoxicity. A likely mode of action for hUCBCs in ALS is therefore modulation of microglial activation and inflammation.
hUCB and Sanfilippo Syndrome Type B
Another disorder that we have shown to improve following hUCB is the genetic metabolic disorder, Sanfilippo syndrome type B (mucopolysaccharidosis III B). This is caused by a deficiency of the enzyme α-N-acetylglucosaminidase (Naglu), resulting in heparin sulfate accumulation that culminates in cerebral and multiple systemic organ abnormalities. The main focus of treatment is currently enzyme replacement therapy. A knockout mouse model of the Naglu gene exhibits similar characteristics to the disorder in humans (62). Garbuzova-Davis et al. have previously shown in vitro that mononuclear hUCBCs can secrete Naglu enzyme, suggesting that these cells could be an effective for enzyme replacement cell therapy (38). Following intracerebroventricular (ICV) transplantation of mononuclear hUCBCs into 1-month-old Naglu knockout mice, the cells were found to migrate into the brain tissue, differentiate into neurons, and they survived for at least 6 months. Normal neural architecture was observed within the brain, along with a reduced presence of glycoaminoglycans such as heparin sulfate in the liver and periphery (38).
However, ICV transplantation is not an optimal route of administration and, because this is a genetic disorder, prenatal transplantation was suggested. Pregnant heterozygous Naglu knockout mice were administered mononuclear hUCBCs 5 days into pregnancy and 7 days later the embryos were tested for the presence of hUCBCs (33). Cells were detected throughout the fetal brain as well as in the liver. Naglu expression in the embryos was similar to their heterozygote parents, which is half of the normal level. This therefore presents a potential new means to treat genetic-based disorders, although more long-term studies would be required before the full impact could be realized.
Alternatively, the effect of IV hUCBC transplantation into Naglu mutant mice was also investigated (34). Three-month-old mice were administered either 3 × 106 mononuclear hUCBCs or vehicle via the jugular vein and given immunosuppressants for the rest of the study. The mice were followed for 6 months and those given hUCBCs demonstrated improved cognitive function (by active avoidance testing) and decreased hyperactivity (open field locomotor activity). Grafted cells were found to have survived for 6 months and were spread throughout the brain, particularly around blood vessels, and also in peripheral organs. Cytokine analysis revealed reduced Th1 (proinflammatory) cytokines and increased Th2 (anti-inflammatory) cytokines within the brain. Reduced heparin sulfate accumulation was also seen in the liver (34). These cells are likely to act by reducing the inflammatory response as well as being a vector for enzyme replacement. A monocyte/macrophage-specific fraction of hUCBCs was also found to be effective, although there was some suggestion of greater effects in female mice (40). Decreased glial activation was also apparent, suggesting modulation of inflammation. A phagocytic effect on heparin sulfate by these cells can also not be ruled out. These findings support the previous studies showing that the monocyte and macrophage component of hUCBCs are an important fraction with respect to the beneficial effects normally observed by mononuclear hUCBC transplantation (93).
The advantages were found to be diminished with time and therefore multiple injections should be encouraged, so that enzyme replacement and the anti-inflammatory effects could be maintained. Studies are on-going to determine whether multiple injections result in a greater recruitment of hUCBCs into the brain and a greater reduction in heparin sulfate accumulation in peripheral organs such as the liver than with a single injection (41). Impairment of the blood—brain barrier occurs in ALS and it may also be present in the Sanfilippo mouse model (35,37). This may contribute to the ability of the cells to enter the brain, in conjunction with a homing mechanism triggered by cell damage, such as release of monocyte chemoattractant protein-1 and macrophage inflammatory protein as is seen in stroke (49).
Conclusions
Our recent studies using mononuclear human umbilical cord blood cells and human menstrual blood-derived cells to treat neurodegenerative disorders such as stroke, AD, ALS, and Sanfilippo syndrome provide hope for the potential of these methods to treat these disorders. The importance of the monocyte component and the cell's ability to modify inflammation and microglia activation are highlighted and clearly require further study, but provide encouraging results. Several of the studies involve using multiple transplantations instead of just one and this appears to be more effective, particularly in progressive disorders.
Footnotes
Disclaimer
A.E.W., S.G.-D., and C.V.B. are consultants and P.R.S. is cofounder of Saneron CCEL Therapeutics.