Abstract
Parkinson's disease (PD) is a common neurodegenerative disease and typically presents with tremor, rigidity, bradykinesia, and postural instability. The hallmark pathological features of PD are loss of dopaminergic neurons in the substantia nigra (SN) and the presence of neuronal intracellular Lewy body (LB) inclusions. In general, PD is sporadic; however, familial PD, while uncommon, can be inherited in an autosomal dominant (AD) or autosomal recessive (AR) manner. The molecular investigations of proteins encoded by PD-linked genes have clarified that ADPD is associated with α-synuclein and LRRK2, while ARPD is linked to Parkin, PINK1, DJ1, and ATP13A2. Understanding these genes can bring insights into this disease and create possible genetic tests for early diagnosis. Long-term pharmacological treatment is so far disappointing, probably due to unwanted complications and decreasing drug efficacy. Several strategies have been proposed and tested as alternatives for PD. Cellular transplantation of dopamine-secreting stem cells opens the door to new therapeutic avenues for restoration of the functions of degenerative and/or damaged neurons in PD.
Keywords
Prevalence
Parkinson's disease (PD), known as idiopathic parkinsonism, primary parkinsonism, or paralysis agitans, is a neurodegenerative disorder (145). Epidemiological data show that PD affects total 0.3% of the population in industrialized countries, more than 1% of 60-year-old individuals, and 4% of those ages over 80 (27). The overall age- and gender-adjusted incidence rate is 13.4 per 100,000, with a higher prevalence in male (male vs. female:19.0 per 100,000 vs. 9.9 per 100,000) (148). The incidence of PD is between 8 and 18 per 100,000 person-years (27). Early-onset PD is found in patients with onset before age 50 years. Late-onset PD is diagnosed in those over age 50 years. Epidemiological studies showed that most PD cases are sporadic (90%) and late-onset (144). Only 5–10% of cases are characterized by early onset (130), and these mostly occur in familial clusters (100). The age-adjusted incidence rate is 10.4 per 100,000 population for all age groups of PD in Taiwan (18).
Clinical Manifestations
Cardinal symptoms of PD are tremor, rigidity, bradykinesia, hypokinesia, akinesia, and postural instability. Reduced blinking rate and facial expressions, absent or reduced arm swinging, and absence of associated movements during daily life activities such as rising from a chair and waving are noted within PD families. The parkinsonian gait includes reduced stride length and decreased off-ground elevation of the feet, leading to short stepping and shuffling, which are associated with reduced arm swinging and predominance of flexor posture. Psychiatric problems contain executive dysfunction, learning problems, mood disorders, and visual hallucinations. Dementia occurs in at least 20% of cases (126).
Pathology
At autopsy, gross examination of the brain in a patient with PD reveals no remarkable abnormality, and the brain weight is typically normal for age, and only the substantia nigra (SN) and the locus ceruleus (LC) are generally pale due to loss of the pigmented dopaminergic neurons. Microscopically, SN and LC may show Lewy bodies (LBs) and selective neuronal degeneration and/or loss. In fact, before the onset of symptoms, at least 70% of neurons exhibit degenerative dysfunction or neuronal loss in PD patients (8).
Lewy Bodies (LBs)
LBs have two distinct morphological types, classical LBs and cortical LBs. Classical LBs are found in the brainstem and are eosinophilic cytoplasmic inclusions that consist of a dense core surrounded by a halo of 10-nm-wide radiating fibrils, the primary structural component of which is α-synuclein (113). Cortical LBs can be detected in the nucleus basalis of Meynert, dorsal motor nucleus of the vagus nerve, innominate substance, amygdale, hippocampus, spinal cord, sympathetic ganglia, and the cerebral cortex (41). The cortical LB is also made up of α-synuclein fibrils (30). LBs can be found in several neurodegenerative diseases, such as PD, Alzheimer disease, Pick's disease, and multiple system atrophy (61).
Selective Degeneration and Loss of Neurons
Slow and progressive degeneration and loss of dopaminergic neurons in the SN and degeneration of the nerve terminals in the striatum are the underlying pathophysiological mechanisms of PD (54), leading to typical PD symptoms, such as tremor, rigidity, and hypokinesia. Affected noradrenergic neurons of LC may cause dementia, depression, and disease progression (46). Degenerated serotonergic neurons of the raphe obscurus and medial raphe may result in depression (62). Causes of the selective degeneration and/or loss of specific populations of neurons in PD are not clear. Risks such as infectious agents (141), pesticides (131), heavy metals (100), and living in rural environments (122) have been reported to be linked to PD.
Genetics
Familial PD is an uncommon entity that affects patients younger than those affected by sporadic PD. Ten to thirty percent of familial PD patients reported a first-degree relative having PD (1). Familial PD can be inherited in an autosomal dominant (AD) or autosomal recessive (AR) manner. At least 13 loci and 9 genes have been proposed to be linked with PD. Six genes are linked to Mendelian forms of PD. α-Synuclein (SNCA) and leucine-rich repeat kinase 2 (LRRK2) are associated with ADPD, while Parkin, phosphatase and tensin homolog (PTEN)-induced kinase 1 (PINK1), DJ-1, and ATPase type 13A2 (ATP13A2) are linked to ARPD.
α-Synuclein (SNCA)
The SNCA gene has several other appellations including PD1 (Parkinson's Disease 1), NACP (non-amyloid component of senile plaques precursor protein), and PARK1 (Parkinson's 1). It has six exons, encoding an abundant 140 amino acids, including α-synuclein (70), which is expressed predominantly in the brain, heart, pancreas, skeletal muscles, megakaryocytes, platelets, lymphocytes, and monocytes (134). α-Synuclein is involved in neuronal plasticity, learning, development, cell adhesion, phosphorylation, cellular differentiation, and regulation of dopamine uptake (22,44). Gene rearrangements of α-synuclein, including duplications or triplications of the wild-type gene, may be related to ADPD (42,57). The deposition of α-synuclein aggregates in neurons or glial cells is a hallmark lesion in a subset of neurodegenerative disorders (3,138), including PD, dementia with Lewy bodies (DLB), and multiple system atrophy (MSA), collectively referred to as synucleinopathies. The aggregation of α-synuclein may be toxic to the host and cause the death of dopamine neurons. Overexpressing α-synuclein transgenic animals have been shown to mimic several aspects of PD, such as degeneration of the nigrostriatal system causing motor dysfunction, α-synuclein aggregation/accumulation, and neurodegeneration (68,95). Overexpression of a truncated α-synuclein, containing the non-amyloid component, enhanced aggregation into large inclusions bodies, leading to increasing dopaminergic neurotoxicity in flies (116). α-Synuclein is involved in the modulation of dopamine neurotransmission (1,104). Therefore, α-synuclein is not only one of the familial PD-causing genes but also a risk factor for sporadic PD. Structurally, α-synuclein contains an N-terminal amphipathic region, a central hydrophobic non-amyloid component domain, and a C-terminal region. Truncation and/ or phosphorylation at C-terminal sites is likely to influence α-synuclein affinity to proteins (39,63), metals, and other ligands (dopamines and polyamines) (14,15) and alter the biochemical and biological processes regulated by its interaction with these molecules. A missense G209A mutation (A53T) in the α-synuclein was discovered in a large family with ADPD (119). Mutated G209A stable transfection of HEK293 cells increased dopamine metabolism and caused selective oxidative damage to dopaminergic cells (109). G88C (A30P) was identified in a German family with ADPD (75), but these two mutations have not been detected in Taiwanese PD patients (56,86). α-Synuclein can translocate to mitochondrial membranes with increasing intracellular acidification, and it will become more acidic when phosphorylated, leading to translocation (23). The higher matrix metalloproteinases (MMPs) can cleave α-synuclein, generate several C-terminally truncated fragments, and inhibit α-synuclein aggregation concentration (83). Missense mutations may be associated with posttranslational modifications because hyperphosphorylated α-synuclein at S129 is found in pathological inclusions in postmortem brain samples, and S129 can help target the phosphorylated α-synuclein to the proteasome pathway (91). The accumulation of phosphorylated α-synuclein, especially phosphorylated at S129, may result in the generation of LBs in PD (91). It is therefore clear that phosphorylated α-synuclein needs to be rapidly dephosphorylated and degraded by the proteasome pathway for its toxicity. Moreover, α-synuclein is a substrate for several kinases, such as G-protein-coupled receptor kinases (GRKs, 1, 2, 5, and 6, phosphorylated at S129) (123), casein kinase I (CK I, phosphorylated at S87 and S129) and casein kinase II (CK II, phosphorylated at S129) (108), leucine-rich repeat kinase 2 (LRRK2, phosphorylated at S129) (124), and polo-like kinases (PLKs, 1, 2, and 3, phosphorylated at S129) (60,96). PLKs may be involved in the phosphorylation of aggregated α-synuclein. Activation or overexpression of CK2 and PLKs can significantly enhance the phosphorylation of α-synuclein at S129 (157). CK2 and PLK inhibitors reduced S129 phosphorylation but did not have a substantial effect on the propensity of α-synuclein to aggregate (157). PLK2 can enhance the phosphorylation of aggregated α-synuclein at S129, and the phosphorylation is also largely reduced in PLK2–/– transgenic mice (60). However, PLK2 is seemed to colocalize with α-synuclein when overexpressed, and PLK1 is significantly involved with phosphorylation of α-synuclein aggregates.
Mitochondrial complex I possibly plays a central role in the pathogenesis of PD because a significant increase of mitochondrial α-synuclein and a decrease of complex I activity were found in PD patients (29), and derangements in mitochondrial complex I may cause α-synuclein aggregation (26,132). Overexpressing α-synuclein disrupted endoplasmic reticulum (ER)–Golgi vesicular (24) and microtubule-dependent trafficking (79). Since microtubules are fundamental to vesicular movement, impairment within the microtubule complex increases α-synuclein aggregation and toxicity (66). Thus, agents modulating the microtubule system may provide therapeutic benefits in PD. NAPVSIPQ (NAP; also known as davunetide or AL-108) is a peptide derived from the activity-dependent neuroprotective protein (ADNP) that interacts with both neuronal and glial tubulin to modulate microtubule assembly (33,49) and can reach the brain by IV or intranasal administration. NAP can protect neuronal cells by against dopamine toxicity and severe oxidative stress (106). Mice overexpressing human α-synuclein and that were intranasally administered daily with NAP for 2 months showed a significant improvement in behavior tests and a reduction in α-synuclein inclusions in SN (40). Hence, the improvement of dysfunctional proteasomes to facilitate the dephosphorylation and degradation of phosphorylated α-synuclein unravels the role of inhibitors of α-synuclein aggregation/accumulation and kinases such PLKs, and restoration of mitochondrial complex I, modulation of MMPs, and administration of NAP may revolutionize the treatment of this debilitating disorder.
Leucine-Rich Repeat Kinase 2 (LRRK2)
LRRK2 gene (acronym PARK 8) consists of 51 exons encoding a protein, Dardarin, which has 2,527 amino acids and contains an N-terminal leucine-rich repeat, a GTPase ROC (Ras of complex proteins)/COR (C-terminal of Roc) domain, a mitogen-activated protein kinase kinase kinase (MAPKKK), and a C-terminal β-propeller (166). The expression rate of LRRK2 is low in the predominantly affected dopaminergic neurons of SN but relatively higher in the dopamine terminals innervating the caudate nucleus and putamen, suggesting a protective role of LRRK2 in PD (43). More than 50 variants in the LRRK2 gene have been identified, and the mutation frequencies range from 2% to 40% in different population (6,81,111). G2019S, the most common mutated point, has been found in approximately 5–7% of familial PD and 1–2% of idiopathic PD (47), less than 0.1% in East Asia, approximately 2% in European-descent patients, and about 15–40% in Ashkenazi Jews and North African Arabs (82). Mutations including R1441H/ C/G, G2019S, and I2020T are all located in the kinase domains (R1441H/C/G in the GTPase ROC domain; G2019S and I2020T in the MAPKKK domain), suggesting that these mutations cause an erratically increased kinase activity of the mutant LRRK2, and end in increased neuronal toxicity in cortical neurons. However, variations in neuropathologic findings in affected individuals with the same mutation are not consistent with this suggestion. LRRKs may be involved in the regulation of mRNA translation processes (59).
Parkin
Parkin gene (acronym PARK2) consists of 12 exons. Point mutations, deletions, and multiplications in Parkin gene account for 45% of ARPD with an age of onset of <45 years and for 77% of isolated cases of PD with an age of onset <20 years (90). These findings suggest that mutated Parkin gene may increase susceptibility for PD. Moreover, the number of heterozygous mutation carriers in Parkin and PINK1 genes was significantly higher than that of homozygous or compound heterozygous cases (69), suggesting that Parkin may be AR. Parkin protein is composed of three domains: an N-terminal ubiquitin-like domain (Ubl), a PDZ [postsynaptic density protein (PSD95), Drosophila disc large tumor suppressor (Dlg1), and zonula occludens-1 protein (zo-1)] linker region, and a C-terminal Really Interesting New Gene (RING) box (133). Parkin may localize to the mitochondrial outer membrane (25) or to the ER under stress conditions (142). Parkin KO (knock out) Drosophila displayed mitochondrial dysfunction and increased oxidative stress (50). Parkin and E2 protein (ubiquitin-conjugating enzyme) form a complex to ubiquitinate cellular proteins for degradation. Parkin may play a role in the neuroprotection because mutated Parkin may lead to an abnormal accumulation of non-ubiquitinated intracellular proteins, causing the loss of dopaminergic neurons.
Pten-Induced Kinase 1 (Pink1)
PINK1 gene (acronym PARK6) encodes a 581-amino acid protein and is predicted to encode a 34-amino acid mitochondrial targeting motif (58). The frequency of PINK1 mutations ranges from 1% to 9% in different ethnicities (145). Mutations in the PINK1 gene, a truncating nonsense mutation W437X and a G309D missense mutation (147), were originally discovered in three pedigrees with ARPD. siRNA-mediated depletion of PINK1 is shown to increase the susceptibility to apoptotic cell death (28). Additionally, PINK1 protein is found to localize in the mitochondria (98). A wild-type PINK1 can phosphorylate tumor necrosis factor (TNF) receptor-associated protein 1 (TRAP1), a mitochondrial molecular chaperone (heat shock protein 75; HSP75) (38). The phosphorylation of TRAP1 was significantly increased in response to oxidative stress induced by H2O2. This stress did not affect the PINK1-TRAP1 interaction (121). PINK1 can protect cells against the loss of mitochondrial function resulting from exposure to proteasome inhibitors (147). Therefore, TRAP1, which is phosphorylated by PINK1, can resist the oxidative-stress-induced apoptosis. PINK1 kinase activity is important in protecting nigral neurons, and PD-linked PINK1 mutants may attenuate/abolish this kinase activity, causing the phenotypes of PINK1-related PD. The phenotypes of Drosophila KO PINK1 and KO Parkin are very similar (51), and their mitochondria were shown to be abnormal in muscle and gonadal cells (120). The phenotypes of Drosophila KO PINK1 can be rescued by Parkin overexpression (115), indicating that Parkin and PINK1 may play a critical role in the same mitochondrial signaling pathway. The phenotypes of PD patients with PINK1 mutations are similar to those with Parkin mutations (58), but psychiatric features, such as affective and schizophrenia spectrum disorders, may be part of the phenotypic spectrum in PD patients with PINK1 mutations (12,58). PINK1-related PD cases have a slow progression and a good and sustained response to l-DOPA. α-Synuclein can induce mitochondrial fragmentations and can be rescued by the coexpression of PINK1, Parkin, or DJ-1 (64), which suggests that Parkin, PINK1, and DJ-1 may interact in the same signaling pathway. Any defect or disturbance of the balance in these factors may lead to PD.
DJ-1 (Acronym Park 7)
The DJ-1 gene contains eight exons and spans 24 kb on chromosome 1p36, encoding for a 189-amino-acid protein, which is conserved and ubiquitously expressed and is rarely associated with early-onset PD, and its phenotype is similar to the Parkin and PINK-1-linked types (112). Normally, wild-type DJ1 protein is a homodimer, distributes either in the intracellular nucleus or cytoplasm (165), and may increase in response to endogenously produced reactive oxygen species (ROS), probably through the activation of nicotinamide adenine dinucleotide phosphate (NADPH) oxidase (99). DJ1 KO cells are sensitized to H2O2-driven oxidative stress and rescued by overexpression of wild-type DJ1 (162). Moreover, when cells were exposed to hydrogen peroxide, DJ1 was able to conduct an acidic shift in isoelectric point value (from 6.2 to 5.8) mainly oxidizing its Cys 106 residue, suggesting that DJ-1 protein function is significantly involved in the redox, and this protein is delicately regulated by oxidation of Cys-106 (67). All these findings support that the oxidative stress is an important component in the pathogenesis of PD.
Atpase Type 13A2 (ATP13A2)
ATP13A2 encodes a lysosomal type 5 P-type ATPase of 1,180 amino acids, highly expressed in brain, especially in SN (125). ATP13A2 gene was identified in Kufor–Rakeb syndrome (KRS) patients (31), who manifested l-DOPA-responsive, juvenile-onset akinesia, rigidity, dyskinesia, visual hallucinations, pyramidal signs, supra nuclear gaze paresis, cognitive deterioration, and diffuse moderate atrophy, indicating that mutated ATP13A2 may be a cause of early-onset PD (103). α-Synuclein toxicity can be suppressed by ATP13A2 (48). GBA, a gene encoding glucocerebrosidase, is reported to be a risk factor for PD (2). Glucocerebrosidase is a lysosomal enzyme and is linked to LB disorders (105). Together, all these findings suggest that lysosomes may involve in the pathogenesis of PD.
Novel Genes
There are some novel genes related to PD, including
PLA2G6, encoding a calcium-independent group VI phospholipases A2 (PLA2s) (101), is reported as the causative gene for PARK14-linked ARPD (163). PLA2s are involved in reactions that result in the release of fatty acids in inflammation. A novel heterozygous PLA2G6 (2417C>G) mutation in exon 17, and heterozygous PLA2G6 (87G>A) mutation in exon 22 was discovered in atypical PD (143).
FBXO7, a member of the F-box protein family, acting in the protein degradation involving the ubiquitinproteasome system (135), was found in atypical PD.
Grb10-interacting GYF protein-2 (GIGYF2) is a candidate gene for the PARK11 locus, and 10 different missense mutations were identified in PD patients in France, Italy (76), and China (156).
Ubiquitin C-terminal hydrolase-L1 (UCHL1), a 223-amino-acid protein, abundantly and selectively expressed in brain, is a member of the UCH (ubiquitin C-terminal hydrolase) family of deubiquitinating enzymes (DUBs). UCHL1 is involved in regulation of ubiquitin pool, apoptosis, and learning and memory (13). A point mutation Ile93 to Met has been implicated in PD (80). Immunohistochemistry of midbrain sections of a PD patient showed α-synuclein and UCHL1 double-positive LBs in nigral DA neurons and abnormal accumulation of α-synuclein in the dopaminergic neurons in UCHL1-transgenic and spontaneous UCH-L1-null gracile axonal dystrophy mice. Overexpression of α-synuclein significantly enhanced the loss of nigral DA cell bodies in mutated UCHL1(Ile93Met)-transgenic mice, sup porting that UCHL1(Ile93Met) mutant may be involved in the pathogenesis of PD (160).
Nuclear receptor subfamily 4, group A, member 2 (NR4A2) protein, also known as nuclear receptor related 1 protein (NURR1), is encoded by the NR4A2 gene in humans. Mutations in exon 1 of NR4A2 have been identified in PD families (78) and a defect in NR4A2 causes increased susceptibility to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), leading to SN injuries (77).
Synphilin-1, encoded by SNCAIP gene, has been shown to interact directly with α-synuclein and is found in LBs (153,154). Synphilin-1 is potentially involved in the pathogenesis of PD. An investigation in a postmortem brain tissue showed this protein was very similar to α-synuclein (37). Overexpression of synphilin-1 was observed to decrease caspase 3 activation, increase beclin-1 and microtubule-associated light chain 3 (LC3) II expression, and promote formation of aggresome-like structures, suggesting that synphilin-1 can protect against neuronal degeneration (136). A novel C-to-T transition in position 1861 of the coding sequence, leading to an amino acid substitution from arginine to cysteine in position 621 (R621C), was identified in a group of 328 German PD patients (94). Cells transfected with the mutated R621C were more susceptible to apoptosis than cells expressing wild-type synphilin-1 (136).
Diagnosis and Treatment
The diagnosis of PD is clinical and there is no specific test for PD. A good response to dopamine agonists is generally regarded as supporting diagnostic features. PD invariably progresses with time, and so current therapies center on medical and surgical treatments for controlling symptoms. Pharmacologic replacements of dopamine, such as 3,4-dihydroxy-l-phenylalanine (l-DOPA), carbidopa, and monoamine oxidase-B inhibitors, are the mainstay of the treatment of PD. They are particularly effective during the early stages of the disease. However, long-term use of these medications may lead to reduced drug efficacy and the development of involuntary motor complications, such as wearing-off and sudden-off phenomena and troublesome hyperkinesias, which more or less impact patients' quality of life. This hampers the continuous use of these medical therapies. Surgical procedures, including pallidotomies and thalamotomies, are applied for treatment of refractory symptoms (73). Although symptoms of PD are better after these procedures, patients are often at risk of having severe side effects, such as dysarthria or hemiparesis. The deep brain stimulation (DBS) of the globus pallidus and the subthalamic nuclei is another choice. Only minimal tissue damages have been reported in DBS, and numerous studies have demonstrated that DBS not only improves troublesome symptoms but also reduces the daily dosages of medications (52). Even though several therapeutic approaches for this disease are available, there is no complete treatment for halting the degenerative process of this disease.
Stem Cell Transplantation: A Promising Therapy for PD
An alternative clinical approach to pharmacological or DBS treatment for PD is to implant actively synthesizing dopamine cells to replace, repair, or restore the degenerative and/or damaged neurons in the caudate and putamen of the striatum in PD patients. Animals with successful grafts showed significantly decreasing PD behaviors, indicating that grafted fetal dopaminergic neurons can provide therapeutic benefit in aged and dopamine-depleted rodents (10,11,55,117). Later, aborted human embryonic mesencephalic tissue was then implanted in the caudate and putamen of PD patients, and data showed that midbrain dopaminergic neurons can survive long-term in PD patients and improve the functions of dopaminergic neurons (34,87,118). Moreover, cell dissociation, suspension culture, and incubation in growth factors such as glial cell line-derived neurotrophic factor (GDNF) and epidermal growth factor (EGF) may enhance graft survival (16,45,127). However, in primary fetal ventral mesencephalon transplantation, a relatively high percentage of grafted cells die, and the majority of dopaminergic neurons originally grafted become necrotic or apoptotic within the first 24 hours (36) and grafted neurons may undergo a PD-like process (21,72). Additionally, the application of transplant to each PD patient at least six fetuses for sufficient dopaminergic neurons has limited the clinical use (9). Because stem cells can provide an extensive source for maintaining cell numbers in the growth of embryonic development and in injury, disease, and natural cell turnover of adult tissue. The discovery of stem cells in the embryonic and adult CNS and the successful isolation of these cells overcome these problems. There are several types of stem cells, including embryonic stem cells, neural stem cells, mesenchymal stem cells, induced pluripotent stem cells, and others from new approaches, which will be discussed in the below.
Embryonic Stem (ES) Cells
ES cells, derived from the inner cell mass of mammalian blastocysts, can proliferate extensively in vitro and can differentiate into all cell lineages, such as neurons, astrocytes, and oligodendrocytes (5,107). Thus, human ES cells are the best source for cell transplantation. The five-stage method, including ES cells in an undifferentiated status, embryonic body (EB) formation (4 days), selection of neural precursors (7–10 days), expansion of neural precursors (4 days), and differentiation into neurons (11–15 days), can help researchers harvest as much as 34% of tyrosine hydroxylase (TH)-positive neurons from mouse ES cells in the presence of signaling molecules such as sonic hedgehog (Shh), fibroblast growth factor (FGF)-8, and ascorbic acid. Nurr1 and pituitary homeobox 3 (Pitx3) can cooperatively promote the midbrain dopaminergic neuron phenotype maturation in murine and human ES cell cultures (93). The stromal cell-derived inducing activity (SDIA) method is faster, easier, and more efficient than the five-stage method (65). Human ES cells exposed to signaling molecules including Shh and FGF-8 with telomerase-immortalized human fetal midbrain astrocytes activate dopaminergic neurogenesis differentiating to dopaminergic neurons, leading to a significant improvement of motor function in PD rats (129). However, lack of genetic stability and long-term survival, risk of tumor formation, and the need of destroying a human embryo to obtain hES cell lines limit the research on the use of these cells.
Neural Stem Cells
Neural stem cells (NSCs) are the earliest uncommitted multipotent cells in the CNS and are able to differentiate into neurons, oligodendrocytes, and astrocytes. In addition, NSCs have the abilities of extensive proliferation and self-renewal. NSCs can be isolated from almost all parts of the embryonic/fetal brains, such as cortex, septum, thalamus, ventral mesencephalon, forebrain ventricular zone, hippocampal dentate gyrus, and spinal cord, from adult cadavers, and from pluripotent ESC and induced pluripotent stem cells (IPSCs) (139,140). However, only few NSCs can successfully differentiate into nigral dopaminergic neurons in vitro (20) or in vivo (155). Several factors contribute to this problem. First of all, NSCs in different regions of brains may possess variable proliferative, self-renewal, and differentiation abilities because NSCs, derived from the ventral mesencephalic region, may have a better result in the dopaminergic differentiation when compared with other regions of the brain (128). Secondly, a complex interplay of inductive factors and dopaminergic neurons specific transcription factors, such as Wnt1, Shh, and FGF8, strictly regulates the percentage of dopaminergic differentiation in the embryonic ventral mesencephalon (161). Thirdly, the quality and quantity of growth factors and serum-free medium will affect the functions of NSCs. A defined serum-free medium supplemented with a mitogenic factor (EGF and/or bFGF) is used to maintain the NSCs in vitro in an undifferentiated state in the form of clusters of cells, called neurospheres, and can be passaged effectively (88). Cells isolated from different regions of the rat and human CNS are shown to possess the ability to form neurospheres in similar serum-free culture conditions with mitogens, but the rate of success in continually generating neurospheres is variable (88).
Mesenchymal Stem Cells (MSCs)
Bone marrow is known to contain hematopoietic and mesenchymal stem cells (MSCs). Normally, MSCs can be expanded in vitro and differentiated into various mesoderm-type lineages, including bone, fat, cartilage, muscle, tendon, hematopoiesis-supporting stroma and vasculature. MSCs have been isolated from adipose tissue, umbilical cord blood, placental tissue, liver, spleen, testes, menstrual blood, amniotic fluid, pancreas, and periosteum and infant dental pulp (149). Under appropriate conditions, MSCs can differentiate into skeletal, cardiac muscle, lung, and liver cells and can migrate toward damaged tissues in an animal model of strokes and PD (17,53). Although the mechanisms are not clear, chemokine (C-X-C motif) receptor 4 (CXCR4), expressed on MSCs, can be navigated by a signal molecule, stromal cell-derived factor-1α (SDF-1α), which may direct MSCs to ischemic and degenerative lesions in the brain (4,92). Studies have shown that rat and human MSCs can differentiate into neuron-like cells and can enter the mouse CNS (92,139). MPTP-treated PD mice given an intrastriatal injection of MSCs exhibited significant improvement on the rotarod test, and the cells survived more than 4 months (84). The TH gene-engineered rat MSCs grafted to PD rat decreased the rounds of asymmetric rotation, and the TH gene expression efficiency was about 75%. Histological examination showed that TH gene was expressed around the transplantation points. The dopamine level in the lesioned striatum of rats injected with TH-MSCs was significantly higher (89). Another report showed that approximately 35.7% of MSCs in the SN were TH-positive and had a dopaminergic function (114). All these data suggest that cell therapy with MSCs for PD may be feasible, easy, and, most importantly, has no ethical problems.
Induced Pluripotent Stem (IPS) Cells
IPS cells are generated from skin fibroblasts through the activation of a combination of genes that are capable of dedifferentiating the fibroblasts into pluripotent cell (164) and neural lineages (152). IPS cells can be generated from animals, humans, and patients with diseases, such as amyotrophic lateral sclerosis (32), spinal muscular atrophy (35), and PD (137). These cells can therefore be used as a disease model for exploring underlying mechanisms and developing new therapeutic strategies. Transplantation of IPS-derived neurons with normal or better functional activity into patients with PD or other degenerative disorders or functional deficits may be a potential option of total cure.
Others
Umbilical cord blood (UCB) is another source of hematopoietic stem cells. Human UCB (hUCB) stem cells are considered as an effective agent for the treatment of spinal cord injury (19) or intrinsic sphincter deficiency (85) because these cells can be induced to differentiate into neuron-like cells, astrocytes, and oligodendrocytes by the treatment of retinoic acid (RA). MSCs can also be derived from the human umbilical cord (hUCMSCs) and they can be transplanted into the brains of hemiparkinsonian rats, where they could differentiate into dopaminergic neuron-like cells and ameliorate their rotations. When hUCMSCs were modified by adenovirus-mediated vascular endothelial growth factor (VEGF) gene transfer and subsequently transplanted into hemiparkinsonian rats, the VEGF expression significantly enhanced the dopaminergic differentiation of hUCMSCs in vivo and ameliorated apomorphine-evoked rotations and reduced the loss of dopaminergic neurons in the lesioned SN, suggesting that the implantation of UCMSCs not only can directly improve a recipient's impaired functions due to degenerated and/or damaged neurons in rat, but the benefits of implant these stem cells will be better when hUCMSCs are modified by adenovirus-mediated gene transfer (159). Embryonic germ (hEG) cells and amniotic fluid-derived stem (hAFDS) cells are also a source of stem cells. Implantation of differentiated hEG cells can replace the damaged neurons and oligodendrocytes in the forebrain of neonatal mice (102). hAFDS cells stably express Pitx3 and Nurr1, which are essential for induction and survival of midbrain dopaminergic neurons (97). Moreover, the level of dopamine is higher in the tissue with dopaminergic cells in the recipients after hAFDS cell implantation (146). Together, therapy with UCB, UCMSCs, hUCMSCs, hEG, and hAFDS cells has potential for the treatment of degenerative and/or damaged neurons.
Conclusion
Studies have unraveled a growing number of responsible loci/genes related to familial PD, such as α-synuclein (SNCA), LRRK2, PINK1, DJ-1, and ATP13A2. However, the functions of these genes have not been completely understood. Anecdotally, estrogen, vitamins C and D, smoking, nicotine, and caffeine intake may play a protective role against PD, but smoking and estrogen are hampered by such serious problems as the development of cardiovascular diseases and carcinogenesis. Mitochondrial DNA deletions in SN neurons are common in PD (7,74). Individuals with a specific variant in the NADH complex I enzyme have a significantly lower risk of PD, suggesting that complex I proteins are related to PD susceptibility. The accumulation of damaged mitochondria, caused by mutated PINK1 and Parkin, may increase the vulnerability of the neurons in the SN to oxidative stress. Neuroprotective agents, including restoration of complex I activity, decrease of oxidative stress and α-synuclein aggregations, and enhancement of protein degradations are promising in the treatment of PD. Moreover, novel therapeutic agents histone deacetylase (HDAC) inhibitors such as sodium butyrate or suberoylanilide hydroxamic acid (SAHA), and sirtuin 2 (SIRT2) inhibitors can protect against α-synuclein overexpression -induced neurotoxicity (71,110). Current therapies, including medical and surgical treatments, mainly control patients' symptoms. Pharmacologic treatments are initially effective, but drug efficacy decreases in the long-term treatment. Surgical interventions, including pallidotomies and thalamotomies, are effective but too invasive. The deep brain stimulation (DBS) of the globus pallidus and the subthalamic nuclei is another choice because it can reduce the daily dosages of medications and cause only minimal damage (52,158). However, the effectiveness of DBS is variable in several PD cases (150), and there are several reports of surgical and nonsurgical complications associated with this application (151). Nevertheless, there is no complete treatment for halting the degenerative process of this disease even though several novel therapeutic approaches for this disease are available. Implant actively synthesizing dopamine cells to replace, repair, or restore the degenerative and/or damaged neurons in the striatum where dopamine is no longer produced in patients with PD is an ultimate resolution. However, the low survival rate of implanted fetal dopaminergic neurons in PD patients, the occurrence of significant graft-mediated side effects, the PD-like degenerative process in grafted neurons, and ethical issues force researchers and clinicians to overcome these problems. The implantation of stem cells, including ES, NSCs, MSCs, IPS, UCB, UCMSCs, hUCMSCs, hEG, and hAFDS cells, which can specifically produce l-DOPA and/or dopamine or replace degenerative and/ or damaged neurons, is feasible and effective in the treatment of degenerative diseases and, most importantly, has no ethical problems.
Footnotes
Acknowledgement
The authors declare no conflict of interest.
