OUP user menu

Relationship Among α-Synuclein Accumulation, Dopamine Synthesis, and Neurodegeneration in Parkinson Disease Substantia Nigra

Fumiaki Mori PhD, Makoto Nishie MD, Akiyoshi Kakita MD, Makoto Yoshimoto PhD, Hitoshi Takahashi MD, Koichi Wakabayashi MD
DOI: http://dx.doi.org/10.1097/01.jnen.0000230520.47768.1a 808-815 First published online: 1 August 2006

Abstract

The histologic hallmark of Parkinson disease (PD) is loss of pigmented neurons in the substantia nigra (SN) and locus ceruleus (LC) with accumulation of α-synuclein (αS). It has been reported that tyrosine hydroxylase (TH)-negative pigmented neurons are present in these nuclei of patients with PD. However, the relationship between TH immunoreactivity and αS accumulation remains uncertain. We immunohistochemically examined the SN and LC from patients with PD (n = 10) and control subjects (n = 7). A correlation study indicated a close relationship among decreased TH immunoreactivity, αS accumulation, and neuronal loss. In addition, 10% of pigmented neurons in the SN and 54.9% of those in the LC contained abnormal αS aggregates. Moreover, 82.3% of pigmented neurons bearing αS aggregates in the SN and 39.2% of those in the LC lacked TH immunoreactivity, suggesting that pigmented neurons in the SN have a greater tendency to lack TH activity than those in the LC. Recent studies have shown that this decrease of TH activity leads to a decrease of cytotoxic substances and that decreased dopamine synthesis leads to a reduction of cytotoxic αS oligomers. Therefore, the decrease of TH immunoreactivity in pigmented neurons demonstrated here can be considered to represent a cytoprotective mechanism in PD.

Key Words
  • α-Synuclein
  • Neuromelanin
  • Neuronal cell death
  • Parkinson disease
  • Substantia nigra
  • Tyrosine hydroxylase

Introduction

Parkinson disease (PD) is the most common neurodegenerative movement disorder in the elderly, characterized by progressive degeneration of pigmented neurons in the substantia nigra (SN) and locus ceruleus (LC). Neuromelanin pigment is a reliable marker for dopaminergic neurons in the SN and for noradrenergic neurons in the LC (1, 2). Neuromelanin synthesis results from excess cytosolic catecholamines that are not accumulated into synaptic vesicles (3). The concentration of neuromelanin in the SN increases throughout life and is dramatically decreased in PD (4). Although dopaminergic neurons in the midbrain express tyrosine hydroxylase (TH), the rate-limiting enzyme for catecholamine biosynthesis, there is no correlation between the degree of pigmentation and TH immunoreactivity (5). Hirsch et al were the first to observe that TH-negative pigmented neurons are present in the SN and LC of patients with PD and progressive supranuclear palsy, but not in normal controls (6). They speculated that these TH-negative pigmented neurons may represent dying neurons that once contained TH. However, it is still uncertain whether TH-negative pigmented neurons die or survive.

The histologic hallmark of PD is the presence of neuronal intracytoplasmic filamentous inclusions referred to as Lewy bodies (LBs). Recent studies have shown that α-synuclein (αS) is a major component of LBs and LB-related neuritic structures (Lewy neurites) in the brain of patients with PD and dementia with LBs (7-9). In addition, αS immunoreactivity is also present in pale bodies (PBs), the precursor of LBs (10-12). Moreover, in the SN and LC in PD, pigmented neurons that are apparently morphologically normal often demonstrate diffuse, faint staining in their cytoplasm. αS deposited in these lesions is extensively phosphorylated (13). Thus, αS is involved even in the early stage of LB formation. Recent in vitro studies have shown that αS plays a role in the regulation of dopamine biosynthesis, acting to reduce the activity of TH (14).

Despite the increase in our understanding of αS pathology in PD, there are some important questions that remain unanswered: What proportion of SN and LC neurons contains abnormal αS aggregates? What is the progression pattern of αS accumulation in the SN in PD? Is there a relationship between TH-negative pigmented neurons and αS accumulation? To answer these questions, we have performed an immunohistochemical and morphometric investigation of the SN and LC from patients with PD and control subjects using antibodies against TH and αS.

Materials and Methods

Subjects

Ten autopsied patients with PD (aged 60-85 years; average age, 72.7 years) were selected on the basis of their clinical histories and neuropathologic findings. Patient material was included in the study when 1) at least 2 of the 3 classic clinical features of parkinsonism (resting tremor, rigidity, and bradykinesia) had been present; 2) 1-dopa had been effective during life; and 3) neuropathologic examination had confirmed neuronal degeneration with LBs in the SN, LC, and other subcortical nuclei. None fulfilled either the Braak (15) or CERAD (16) histologic criteria for the diagnosis of Alzheimer disease. Disease duration ranged from 2 to 12 years (average, 6.5 years). In all cases, parkinsonism was the initial and cardinal clinical feature, and dementia appeared later in 5 cases. "Extra pyramidal motor only" symptoms had been present for more than 12 months in these demented patients. We also examined the brains from 7 age-matched control subjects having no extra pyramidal signs, dementia, or psychologic change (aged 53-86 years; average age, 71.0 years). Patients complicated by anoxic or ischemic episodes, severe liver dysfunction, or intracranial mass lesion were excluded.

For routine histologic examination, the brain of each subject was fixed with 10% buffered formalin for 3 weeks and then embedded in paraffin. Four-micrometer-thick sections were cut and stained with hematoxylin and eosin (H&E) and by the Klüver-Barrera method. Neuronal loss in the SN and LC was semiquantified on an arbitrary scale (mild, moderate, or severe).

Immunohistochemistry

Four-micrometer-thick sections of the SN were cut at a level through the outlet of the oculomotor nerve, red nucleus, and the top of the superior colliculus. The LC was cut at the level of the upper pons (the entry zone of the trigeminal nerve). The sections were subjected to immunohistochemical processing using the avidin-biotin-peroxidase complex (ABC) method with a Vectastain ABC kit (Vector, Burlingame, CA). Monoclonal antibodies against TH (TH-16; Sigma, St. Louis, MO; 1:3,000) and phosphorylated αS (#64; Wako, Osaka, Japan; 1:5,000) (11) were used as primary antibodies. For TH immunohistochemistry, the sections were pretreated in a microwave oven for 15 minutes in 10 mM citrate buffer (pH6.0). Diaminobenzidine was used as the chromogen. The sections were counterstained with hematoxylin.

In each case of PD, sections were double immunolabeled with combinations of monoclonal anti-TH (TH-16; 1:3,000) and polyclonal anti-αS antibodies (NACP-6; 1:5,000) (17). The immunoproducts of the monoclonal antibody were detected by the ABC method with diaminobenzidine as the chromogen. The immunoproducts for the polyclonal antibody were detected by the ABC method with streptavidin-alkaline phosphatase as the tertiary reagent and Vector Blue (Vector) as the chromogen for alkaline phosphatase.

Midbrain Cell Groups and Cell Populations

As described previously (18), the SN of each case was divided into 4 morphometric subregions: the paranigral nucleus (PN), intermediate group (IG), ventrolateral group (VL), and dorsal group (DG). Digital images of sections stained with H&E and sections immunostained with anti-αS or anti-TH under a 4× objective lens were captured on a Macintosh personal computer (Apple Computer, Inc., Cupertino, CA) with a Pro 600ES digital camera system (Pixera Co., Los Gatos, CA). Pigmented neurons were plotted on the enlarged prints (80×), which were reconstructed to determine the positions of each neuron in the SN. Similarly, neurons containing αS aggregates and TH-negative pigmented neurons were plotted in the SN.

Cell Counts

In each case of PD and controls, the numbers of TH-positive and TH-negative pigmented neurons in the SN and LC were counted on one unilateral section of the midbrain and upper pons immunostained with anti-TH. In PD, the numbers of TH-positive and TH-negative pigmented neurons with or without LBs and PBs in the SN and LC were counted on one unilateral section of the midbrain and upper pons double immunolabeled with anti-TH and anti-αS. Counting was performed at 200x original magnification using an eyepiece graticule and parallel sweeps of the microscope stage.

Morphometric Analysis

Digital images of midbrain sections immunostained with anti-TH under a 40x objective lens were captured at the highest resolution (2776 × 2074) on a Macintosh personal computer (Apple Computer, Inc.) with a Pro 600ES digital camera system (Pixera Co.). NIH Image 1.62 software was used to calculate the area of nuclei that contained nucleoli in the SN. In each section, more than 50 nuclei of TH-positive pigmented neurons and more than 5 nuclei of TH-negative pigmented neurons were measured. Comparison of the cross-sectional nuclear area was performed between TH-positive and TH-negative pigmented neurons in each case using the Student t test. All morphometric analyses were performed without knowledge of the case number by coding the specimens. Because the number of nucleated pigmented neurons in the LC was less than 3 in 6 of the 10 cases of PD, we excluded the LC from this analysis.

Statistical Analysis

Calculations were performed using STATVIEW II (Abacus Concepts Inc., Berkley, CA) software on a Macintosh personal computer (Apple Computer, Inc.). Statistical comparison between PD and control groups was performed with Student or Welcht t-test. Values were expressed as mean ± standard error mean. Correlations at p < 0.05 were considered significant.

Results

Incidence and Topography of Tyrosine Hydroxylase-Negative Pigmented Neurons

In PD, TH-negative pigmented neurons were found in the SN in all 10 cases (Fig. 1A) and in the LC in 6 of them (Fig. 1B). Pigmented neurons containing LBs often lacked TH immunoreactivity in the cytoplasm (Fig. 1C), although a few pigmented neurons showed TH immunoreactivity in the cytoplasm and LBs (Fig. 1D). In controls, TH-negative pigmented neurons were also found in the SN in all cases and in the LC in 5 of 7 cases.

FIGURE 1.

Tyrosine hydroxylase (TH) immunoreactivity in the substantia nigra (SN) and locus ceruleus (LC) in patients with Parkinson disease (PD). (A) TH-negative pigmented neurons (arrows) in the SN. (B) TH-negative pigmented neurons (arrows) in the LC. (C) TH-negative pigmented neuron containing a Lewy body (arrow). (D) TH-positive pigmented neuron containing a Lewy body (arrow). Scale bar = 10 μm.

The proportion of TH-negative pigmented neurons relative to the total number of pigmented neurons in the SN was significantly higher in PD (22.4%) than in the controls (2.1%) (p < 0.01) (Fig. 2A). In the LC, the proportion was also significantly higher in PD (12.0%) than in the controls (1.3%) (p < 0.05) (Fig. 2B).

FIGURE 2.

The proportion of tyrosine hydroxylase-negative pigmented neurons relative to the total number of pigmented neurons in the substantia nigra (A) and locus ceruleus (B) in subjects with Parkinson disease and controls.

Figure 3 shows the distribution of TH-negative pigmented neurons in the SN. In controls, a small number of TH-negative pigmented neurons were restricted almost entirely to the VL (Fig. 3A). In cases of PD with mild neuronal loss in the SN, a significant number of TH-negative pigmented neurons were found in the SN, mainly in the VL (Fig. 3B). In cases of PD with moderate or severe neuronal loss, many TH-negative pigmented neurons were distributed throughout the SN (Fig. 3C, D).

FIGURE 3.

Schematic distribution of tyrosine hydroxylase (TH)-negative pigmented neurons in the substantia nigra in a (A) control case and Parkinson disease cases with (B) mild, (C) moderate, and (D) severe neuronal loss. One clear circle indicates one TH-negative pigmented neuron.

In PD, the average percentage of TH-negative pigmented neurons was highest in the VL (31.0%), followed by the IG (18.6%) and DG (17.4%), and was lowest in the PN (12.9%) (Table 1). In all of the controls, TH-negative pigmented neurons were most frequent in the VL, the average percentage being 4.6% (Table 1).

View this table:
TABLE 1.

Incidence and Topography of αS Aggregates

As reported previously (10, 11), abnormal αS aggregates in pigmented neurons in the SN and LC were classified into 3 morphologically defined structures: diffuse cytoplasmic staining (Fig. 4A), irregularly shaped focal cytoplasmic aggregates (corresponding to PBs) (Fig. 4B), and positive donut-shaped structures with negative cores (corresponding to LBs) (Fig. 4C).

FIGURE 4.

α-Synuclein aggregates in pigmented neurons in the substantia nigra in Parkinson disease. (A) Diffuse cytoplasmic staining. (B) Irregularly shaped cytoplasmic staining corresponding to pale body. (C) Typical Lewy body. Phosphorylated α-synuclein immunohistochemistry. Scale bar= 10 μm.

Table 2 gives the proportion of αS aggregates relative to the total number of pigmented neurons in the SN and LC in PD cases and controls. In PD, 10% of pigmented neurons in the SN contained abnormal αS aggregates (diffuse cytoplasmic staining, PBs, and LBs); diffuse cytoplasmic staining (5.8%) was detected more often than PBs (2.5%) or LBs (1.7%). The average percentage of pigmented neurons containing αS aggregates was highest in the VL (18.9%), followed by the IG (9.6%) and PN (8.2%), and was lowest in the DG (7.8%). In the LC, 54.9% of pigmented neurons contained αS aggregates; diffuse cytoplasmic staining (32.6%) was seen more frequently than PBs (9.5%) or LBs (12.8%). In controls, only a few pigmented neurons contained PBs in the LC in one case, and no αS aggregates were found in the SN.

View this table:
TABLE 2.

Relationship Between Tyrosine Hydroxylase-Negative Pigmented Neurons and αS Aggregates

Double immunolabeling revealed that 82.3% of pigmented neurons containing αS aggregates in the SN lacked TH immunoreactivity. In addition, 39.2% of those in the LC showed no TH immunoreactivity. Pigmented neurons with αS aggregates usually showed less intense TH immunoreactivity than those without (Fig. 5).

FIGURE 5.

Double immunostaining for tyrosine hydroxylase (TH) (brown) and α-synuclein (blue) in the pigmented neurons in the substantia nigra of patients with Parkinson disease. (A) A pigmented neuron without α-synuclein aggregates shows strong cytoplasmic immunoreactivity for TH. (B-D) Pigmented neurons containing α-synuclein aggregates show weaker cytoplasmic immunoreactivity for TH. (B) A pigmented neuron with focal cytoplasmic staining for α-synuclein shows moderate cytoplasmic immunoreactivity for TH. (C) Another pigmented neuron with irregularly shaped cytoplasmic staining corresponding to pale body shows slight cytoplasmic immunoreactivity for TH. (D) The other pigmented neuron with a typical Lewy body has no immunoreactivity for TH. The more mature α-synuclein aggregates grow in the pigmented neurons, the weaker TH immunoreactivity becomes. Scale bar = 10 μm.

Topography of Neuronal Loss in the Substantia Nigra in Parkinson Disease

There was a significant reduction in both the total and regional neuronal cell counts in the SN in PD compared with those in the controls (Table 3). In the SN, the total cell count in PD was 36.7% of that in the controls. The regional cell count in PD was lower in the VL (28.5% of controls) than in the IG (36.2%), DG (42.5%), and PN (53.5%)(Table 3). In the LC, the total cell count in PD was 24.2% of that in the controls (p < 0.01).

View this table:
TABLE 3.

Mean Nuclear Area of Tyrosine Hydroxylase-Positive and Tyrosine Hydroxylase-Negative Pigmented Neurons

In the SN, the mean nuclear area of TH-negative pigmented neurons was significantly smaller than that of TH-positive neurons in 9 of the 10 cases of PD (Table 4). No fragmented or karyorrhectic nuclei were noted in TH-negative pigmented neurons. In controls, there was no significant difference in the mean nuclear area between TH-positive and TH-negative pigmented neurons (Table 4).

View this table:
TABLE 4.

Discussion

Incidence and Topography of Tyrosine Hydroxylase-Negative Pigmented Neurons

The present study showed that the proportion of TH-negative pigmented neurons relative to the total number of pigmented neurons in the SN (22.4%) was higher than that in the LC (12.0%) in patients with PD. This is consistent with the results of a previous study showing that in PD, 17% of SN dopaminergic neurons and 12% of LC noradrenergic neurons were TH-negative (6).

Hirsch et al reported that no TH-negative pigmented neurons were found in the SN and LC in normal controls (6). In the present study, however, TH-negative pigmented neurons were also found in the SN (2.1%) and LC (1.3%) in controls. Moreover, TH-negative pigmented neurons had a tendency to be localized in the VL in both PD and controls. Considering that the VL is relatively spared (15% loss) in comparison with the ventromedial (38%) and dorsal groups (48%) in brain aging (19), the decrease of TH immunoreactivity in pigmented neurons in the SN, which was most prominent in the VL, may represent a cytoprotective mechanism (discussed subsequently).

Incidence and Topography of αS Aggregates

Previous studies using H&E-stained sections have demonstrated that in PD, the incidence of LB-containing neurons in the SN and LC is 2.4% to 3.4% and 11.0% to 32.1%, respectively (20-22). Ubiquitin immunohistochemistry has shown that the proportion of pigmented neurons with LBs in the SN was 2.5% to 7.6% in PD (23). In the present study, phosphorylated αS immunohistochemistry confirmed that 10% of pigmented neurons in the SN and 54.9% of those in the LC contained abnormal αS aggregates (diffuse cytoplasmic staining, PBs, and LBs). Thus, the incidence of pigmented neurons with αS aggregates in the LC is much higher than that in the SN. This may be explained by the finding that, in PD, accumulation of αS in the LC neurons is earlier than that in the SN neurons (24, 25).

Gibb and Lees examined the topographic distribution of LBs in the SN in incidental LB disease (18). In mild cases, LBs were mostly restricted to the VL. In moderate cases, some LBs were also found in the DG and PN, whereas in severe cases, LBs were distributed throughout the SN. In the present study, αS aggregates were most numerous in the VL, followed by the IG and PN, and were least numerous in the DG. These findings suggest that in PD, αS pathology appears first in the VL, then spreads to the IG and PN, and finally to the DG.

Distribution of Neuronal Loss in the Substantia Nigra

The SN is an anatomically heterogeneous nucleus with regional variations in striatal projections and distribution of histochemical markers. In the human SN pars compacta, a ventral tier contains pigmented neurons with low melanin content, a dorsal tier contains pigmented neurons with high melanin content, and neurons in the PN contain relative amounts of melanin intermediate between those in the ventral and dorsal tiers (18). Fearnley and Lees (19) have reported that in PD, neuronal loss in the SN is greatest in the VL (91% loss) followed by the ventromedial group (71%) and least in the DG (56%). Gibb and Lees (18) have reported that in PD, the VL was affected more severely than the dorsal part and that damage to the PN was of intermediate severity. We showed that in PD, loss of pigmented neurons in the SN was more severe in the VL (71.5% loss) than in the IG (63.8%), DG (57.5%), and PN (46.5%). The VL of the SN is also vulnerable in autosomal-recessive juvenile parkinsonism (PARK2) (26, 27) as well as in striatonigral degeneration (19). Neuromelanin synthesis may thus be a cytoprotective mechanism in PD.

Relationship Among Tyrosine Hydroxylase-Negative Pigmented Neurons, αS Aggregates, and Neuronal Loss

The present study clearly demonstrated that there is a close topographic relationship among TH-negative pigmented neurons, αS aggregates, and neuronal loss. Interestingly, the proportion of TH-negative pigmented neurons relative to the total number of pigmented neurons with αS aggregates in the SN (82.3%) was much higher than that in the LC (39.2%). These findings indicate that pigmented neurons in the SN have a greater tendency to lack TH activity than those in the LC. We further demonstrated that TH-negative pigmented neurons in the SN in PD exhibited a significant reduction in the mean nuclear area. Morphometric studies of the nuclear area of affected neurons in several neurodegenerative diseases have suggested that the decrease in the nuclear area leads to neuronal degeneration (28, 29). It is likely that TH negativity in the SN is a manifestation of a diseased cell with decreased functional ability to synthesize dopamine.

Dopamine and αS are 2 key molecules associated with PD. Recent studies have indicated that oligomers and protofibrils of αS are cytotoxic (30-36)and that αS aggregates might represent less toxic byproducts or even a cellular strategy to inactivate or isolate more toxic oligomers (37) (Fig. 6A). Moreover, it is known that dopamine induces αS to form soluble toxic oligomers (38-40). In the present study, the majority of pigmented neurons with αS aggregates in the SN in PD lacked TH immunoreactivity. Considering that overexpression of soluble αS in a dopaminergic cell line dramatically reduces TH activity and dopamine synthesis (14), it appears that, in PD, the amount of soluble αS is increased in the TH-negative pigmented neurons in the SN. The decrease of TH activity causes a decrease of cytotoxic substances, i.e. free cytoplasmic dopamine, dopamin equinones, and reactive oxygen species (41)(Fig. 6B). Furthermore, the decrease of dopamine synthesis causes a reduction of cytotoxic αS oligomers (42). These findings suggest that TH activity is reduced to provide neuroprotection by decreasing toxic substrates in SN pigmented neurons in PD.

FIGURE 6.

Schematic illustration of presumable α-synuclein (αS) aggregation process (A) and protective role of the decrease of tyrosine hydroxylase (TH) immunoreactivity in pigmented neurons (B). Products and effects in red are cytotoxic. (A) αS oligomers are promoted from soluble αS by excessive dopamine. αS protofibrils are promoted from αS oligomers by oxidative stress, i.e. dopamine-quinones (DA-Q) and reactive oxygen species (ROS). (B) The decrease of TH immunoreactivity reduces cytoplasmic dopamine to decrease cytotoxic DA-Q, ROS, αS oligomers, or αS protofibrils in TH-negative pigmented neurons.

In conclusion, we have demonstrated that decreased TH immunoreactivity in pigmented neurons in the SN and LC is closely associated with αS aggregation in patients with PD. The decrease of TH immunoreactivity indicates reduction of dopamine synthesis, leading to a decrease of cytotoxic αS oligomers. The decrease of TH immunoreactivity in pigmented neurons may represent a cytoprotective mechanism in PD.

Acknowledgments

The authors thank M. Nakata (Institute of Brain Science, Hirosaki University School of Medicine), C. Tanda, J. Takasaki, N. Kaneko, Y. Ota, and S. Egawa (Brain Research Institute, University of Niigata) for their technical assistance.

Footnotes

  • This study was supported in part by a grant from the Fund for Promotion of International Scientific Research, Hirosaki University; and a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
View Abstract