Changes in total cell numbers of the basal ganglia in patients with multiple system atrophy — A stereological study

Neurobiology of Disease 74 (2015) 104–113

Contents lists available at ScienceDirect

Neurobiology of Disease journal homepage: www.elsevier.com/locate/ynbdi

Changes in total cell numbers of the basal ganglia in patients with multiple system atrophy — A stereological study Lisette Salvesen a,b,c,⁎, Birgitte H. Ullerup b, Fatma B. Sunay b,d, Tomasz Brudek b, Annemette Løkkegaard c, Tina K. Agander e, Kristian Winge b,c, Bente Pakkenberg a a

Research Laboratory for Stereology and Neuroscience, Bispebjerg University Hospital, Copenhagen, Denmark Bispebjerg Movement Disorders Biobank, Bispebjerg University Hospital, Copenhagen, Denmark Department of Neurology, Bispebjerg University Hospital, Copenhagen, Denmark d Faculty of Medicine, Balikesir University, Balikesir, Turkey e Department of Pathology, Rigshospitalet University Hospital, Copenhagen, Denmark b c

a r t i c l e

i n f o

Article history: Received 15 October 2014 Revised 8 November 2014 Accepted 12 November 2014 Available online 21 November 2014 Keywords: Basal ganglia Glial cells Multiple system atrophy Neuronal loss Red nucleus Stereology

a b s t r a c t Total numbers of neurons, oligodendrocytes, astrocytes, and microglia in the basal ganglia and red nucleus were estimated in brains from 11 patients with multiple system atrophy (MSA) and 11 age- and gender-matched control subjects with unbiased stereological methods. Compared to the control subjects, the MSA patients had a substantially lower number of neurons in the substantia nigra (p = 0.001), putamen (p = 0.001), and globus pallidus (p b 0.001), and, to a lesser extent in the caudate nucleus (p = 0.03). A significantly lower number of oligodendrocytes were only observed in the putamen (p = 0.04) and globus pallidus (p = 0.01). In the MSA brains the total number of astrocytes was significantly higher in the putamen (p = 0.04) and caudate nucleus (p = 0.01). In all examined regions a higher number of microglia were found in the MSA brains with the greatest difference observed in the otherwise unaffected red nucleus (p = 0.001). The results from the stereological study were supported by cell marker expression analyses showing increased markers for activated microglia. Our results suggest that microgliosis is a consistent and severe neuropathological feature of MSA, whereas no widespread and substantial loss of oligodendrocytes was observed. We have demonstrated significant neuronal loss in the substantia nigra, striatum, and globus pallidus of patients with MSA, while neurons in other basal ganglia nuclei were spared, supporting the region-specific patterns of neuropathological changes in MSA. © 2014 Elsevier Inc. All rights reserved.

Introduction Multiple system atrophy (MSA)1 is a sporadic, adult-onset neurodegenerative disorder. It is clinically characterized by a combination of autonomic failure, parkinsonism, cerebellar ataxia, and corticospinal dysfunction with varying severity, causing progressive disability and death, usually within 7–9 years (O'Sullivan et al., 2008). Patients with MSA are clinically classified into one of two subtypes (parkinsonian or cerebellar) based on their predominant motor presentation (Gilman et al., 2008). Multiple system atrophy is an important differential diagnosis for Parkinson's disease due to a considerable overlap between the clinical presentations (Poewe and Wenning, 2002).

⁎ Corresponding author at: Research Laboratory for Stereology and Neuroscience, Bispebjerg University Hospital, Bispebjerg Bakke 23, opg 11B, 2. sal, 2400 Copenhagen NV, Denmark. Fax: +45 35316434. E-mail addresses: [email protected], [email protected] (L. Salvesen). Available online on ScienceDirect (www.sciencedirect.com). 1 Multiple system atrophy (MSA), glial cytoplasmic inclusions (GCIs), quantitative realtime polymerase chain reaction (qRT-PCR), coefficient of error (CE), coefficients of variation (CVs).

http://dx.doi.org/10.1016/j.nbd.2014.11.008 0969-9961/© 2014 Elsevier Inc. All rights reserved.

The defining neuropathology in MSA is the deposition of insoluble alpha-synuclein-positive glial cytoplasmic inclusions (GCIs) located in oligodendrocytes together with the degeneration of striatonigral or olivopontocerebellar structures (Trojanowski and Revesz, 2007). The pathogenesis of MSA is unknown. However, previously reported neuropathological findings suggest that alpha-synuclein accumulation precedes neuronal degeneration (Fujishiro et al., 2008; Wenning et al., 1994). The formation of GCIs is presumed to cause oligodendroglial dysfunction and myelin degeneration, which lead to axonal damage and neurodegeneration (Ahmed et al., 2012; Ishizawa et al., 2008; Wakabayashi and Takahashi, 2006; Yoshida, 2007). Variable degrees of neuronal loss have been reported in previously published non-stereological studies of brains from patients with MSA. In general, marked neurodegeneration with astrogliosis was found in the substantia nigra, putamen, and olivopontocerebellar structures (Dickson, 2012; Jellinger et al., 2005; Ozawa et al., 2004; Wenning et al., 1997, 2002). Involvement of the caudate nucleus and globus pallidus has also been described (Dickson, 2012; Jellinger et al., 2005; Ozawa et al., 2004; Wenning et al., 1997, 2002). In contrast, the subthalamic nucleus is described as unaffected or relatively spared (Dickson, 2012; Wenning et al., 1997), and the red nucleus, which is a part of the cerebellar output system and located in the midbrain, is

L. Salvesen et al. / Neurobiology of Disease 74 (2015) 104–113

described as unaffected (Dickson, 2012). In addition, similar distributions of oligodendroglial apoptosis and microgliosis have been described (Ishizawa et al., 2004; Probst-Cousin et al., 1998). However, to our knowledge this study is the first to report the total cell numbers of the basal ganglia and red nucleus in MSA measured by methods based on unbiased stereological principles. Stereology is a reliable method for the quantitative estimation of biological structures providing unbiased and precise estimates of structural features (Boyce et al., 2010; West, 2012). A major advantage of stereology is obtaining total quantities and stereology plays an important role in comparative studies of the human brain (Fabricius et al., 2013; Karlsen and Pakkenberg, 2011; Pakkenberg et al., 1991; Pakkenberg and Gundersen, 1997; Pedersen et al., 2005). In this study, we used stereology to quantify the total numbers of neurons and glial cells (subdivided into oligodendrocytes, astrocytes, and microglia) in the basal ganglia (substantia nigra [only pigmented neurons], putamen, caudate nucleus, and globus pallidus) and red nucleus in brains from 11 patients with MSA and 11 age- and gender-matched control subjects. Moreover, we employed a quantitative real-time polymerase chain reaction (qRT-PCR) approach to compare mRNA levels of specific markers for neurons, oligodendrocytes, astrocytes, and activated microglia measured in tissue samples from the substantia nigra and the striatum of 10 MSA brains and 14 control brains. Materials and methods Patients We assessed brain hemispheres from 11 patients with MSA and 11 neurologically healthy control subjects. The included hemispheres were obtained from autopsied individuals following the Danish laws on autopsied human tissue and all individuals gave their informed consent. Several of the control brains have been included in previous studies (Karlsen et al., 2014; Karlsen and Pakkenberg, 2011; Walloe et al., 2014). The cause of death of one control subject was an aortic aneurysm, while the remaining 10 control subjects died from acute myocardial infarctions. Demographic and autopsy-related data for the two groups are summarized in Table 1, and clinical characteristics for the patients with MSA are listed in Table 2. All patients were followed by a movement disorder specialist at the Movement Disorders Clinic, Department of Neurology, Bispebjerg Hospital, Copenhagen, and were clinically diagnosed with a probable diagnosis of MSA (either the parkinsonian or the cerebellar subtype) according to accepted clinical consensus criteria (Gilman et al., 2008). The clinical data were retrospectively obtained from the patients' medical records. Tissue processing The protocol for tissue processing and stereological cell counting of whole hemispheres is previously described in details (Karlsen et al., 2014; Karlsen and Pakkenberg, 2011; Sigaard et al., 2014).

105

In short, the hemispheres were fixed for at least 5 months in 0.1 M sodium phosphate-buffered (4% formaldehyde, pH 7.2) before being sliced into 3-cm thick slabs starting from a random position within the first 3 cm. Each slab was embedded in paraffin using a Leica ASP300 S tissue processor (Leica, Wetzlar, Germany) followed by exhaustive coronal sectioning on a Leica SM2400 sliding microtome with a setting of 40-μm. A known and predetermined fraction of the sections were sampled and stained with Giemsa's azure eosin methylene blue solution (Merck, Darmstadt, Germany) and KHPO4 at a 1:4 ratio. Stereological design The sections were sampled according to systematic uniform random sampling (Gundersen et al., 1999; Gundersen and Jensen, 1987). The six different regions were outlined in each section based on the cytoarchitectonic landmarks using a ×1.25 objective (Fig. 1). Volume estimates were obtained by employing Cavalieri's principle (Gundersen and Jensen, 1987). The numerical densities of neurons and glial cells were estimated with the optical disector (Gundersen et al., 1988). The neurons and glial cells were counted in systematically randomly placed counting frames using the optical disector principle, and the numerical density for one defined region was estimated. The total number of particles was calculated by multiplying the reference volume (uncorrected for shrinkage) with the numerical density. Multiplication by two was performed to obtain bilateral numbers. The sampling scheme for each region is provided in the online Supplementary material. The counting was performed using a × 60 (caudate nucleus, putamen, and globus pallidus) or × 100 (substantia nigra, red nucleus, and subthalamic nucleus) oil immersion objective, resulting in final magnifications of ×2650 (caudate nucleus, putamen, and globus pallidus) and ×1650 (substantia nigra, red nucleus, and subthalamic nucleus). The upper and lower guard zones were set at 5 and ~ 15 μm, respectively. The section thickness was measured in every disector to an average of 40 μm. No systematic variance in cell density was observed in the z-axis. All sections were coded during the stereological quantification process, and the investigator was blind to the disease status. Cell identification The different cell types were differentiated by morphological criteria and spatial distribution as illustrated in Fig. 2a (Pelvig et al., 2008). Neurons were identified by their large nuclei containing a single darkstained nucleolus and their visible cytoplasm. Oligodendrocytes are often situated in groups and in close proximity to neurons or blood vessels. They have small, round, dark nuclei. Astrocytes have round pale nuclei with granulated heterochromatine and visible nuclear membranes. Microglial cell have comma-shaped nuclei and are visibly smaller than the other cell types. As the degenerative pathology of MSA may influence the morphology of cells, immunolabeling for astrocytes (anti-glial fibrillary acidic protein antibody, M0761, 1:100; DAKO, Glostrup, Denmark) and neurons

Table 1 Demographic and autopsy-related data.a Putamen Caudate nucleus Globus pallidus

Age (years) Male/female PMI (hours) Hemisphere (left/right) Hemisphere weight (g)

Controls n = 11 68 [60–75] 5/6 31.7 [10–96] 6/5 571.4 [466–730]

Substantia nigra Subthalamic nucleus Red nucleus MSA n = 11 66 [61–73] 5/6 46.4 [22–115] 5/6 611.6 [533–703]

Controlsa n = 10 69 [60–75] 4/6 19 [10–96] 6/4 578.8 [466–730]

Cell markers qRT-PCR MSAa n = 10 66 [61–73] 4/6 48 [22–115] 5/5 601.4 [533–688]

Controls n = 14 74 [63–86] 8/6 34.9 [21–73] NA NA

MSA n = 10 64 [54–74] 8/2 35.9 [12–80] NA NA

Values are mean and [range]. MSA = multiple system atrophy, n = number of subjects, PMI = postmortem interval, qRT-PCR = quantitative real-time polymerase chain reaction, NA = not available. a Some samples were omitted from different assessments due to technical artifacts or lack of tissue.

106

L. Salvesen et al. / Neurobiology of Disease 74 (2015) 104–113

Table 2 Clinical characteristics of the patients with multiple system atrophy. Case

1

2

3

4

5

6

7

8

9

10

11

Mean [range] or n

Age at disease onset (years) Disease duration (years) Parkinsonisma Cerebellar signsb Autonomic failurec Pyramidal signsd Clinical subtype (P/C)e

63 7 ++ + ++ – P No

57 5 ++ ++ ++ – P Poor

60 7 ++ + ++ + P Poor

54 8 + ++ ++ + C No

60 5 ++ + ++ + P Good

55 6 ++ – ++ + P Good

58 10 ++ + ++ – P Poor

62 7 ++ + ++ + P Good

55 6 + ++ ++ – C No

61 5 + ++ ++ + C No

67 7 ++ + ++ + P Poor

59 [54–67] 6.6 [5–10] 11 10 11 7 8/3 4/4/3

3

6 3 +

4 6 +

4 7 +

6 3 +

6 3 +

1 10 +

6 5 +

4 3 +

3 4 +

5 6 +

4.4 [1–6] 5 [3–10] 11

f L-DOPA response (no/poor/good) Positive red flag categoriesg Wheelchair sign (years)h Abnormal DAT SPECT

i

+

a

Parkinsonism: + = bradykinesia or rigidity, ++ = both. b Cerebellar signs: gait ataxia, limb ataxia, cerebellar dysartria, and nystagmus. + = 1 or 2 positive signs, ++ = 3 or 4 positive signs. c Autonomic failure: + = orthostatic hypotension or urinary incontinence, ++ = both. d Pyramidal signs: + = Babinski sign or hyperreflexia. e P/C = parkinsonian/cerebellar subtype. f A good L-dopa response was defined as a continuous good response ≥5 years. g Positive red flag categories: Early instability, rapid progression, abnormal postures, bulbar dysfunction, respiratory dysfunction, and emotional incontinence. A category is positive if one or more symptoms are present (Köllensperger et al., 2008). h Wheelchair sign = time from disease onset to wheelchair-bound state in years. i Case one never reached a wheelchair-bound state.

(anti-NeuN antibody, MAB377, 1:1000; Millipore, Billerica, MA, USA) of tissue samples taken from the striatum of the control and MSA brains were performed (Fig. 2b and d). They corroborated the morphological criteria. Cells that could not be clearly identified as neurons or glial cells were classified separately; they only constituted a small fraction comprising b 1% of the total cell counts in each region.

Klüver–Barrera staining and immunohistochemistry with antibodies against beta-amyloid (M0872, 1:1000, DAKO) and tau (A0024, 1:1000, DAKO). The definite MSA diagnoses were confirmed by a pathologist based on the presence of GCIs, gliosis, and neuronal cell loss (Trojanowski and Revesz, 2007). In all of the MSA brains GCIs were seen in the tissue samples from the substantia nigra and putamen but not in all of the cortical and cerebellar tissue samples (Fig. 2c).

Neuropathology qRT-PCR Samples of tissue were taken from the substantia nigra, striatum, cerebral cortex, dentate nucleus, and cerebellar cortex. The tissues were embedded in paraffin, sliced into 8-μm-thick sections, and processed for immunohistochemistry with antibodies against alpha-synuclein (sc-12767, 1:9000, Santa Cruz Biotechnology, CA, USA) and ubiquitin (Z0458, 1: 2000, DAKO). Additional hematoxylin and eosin staining was also performed. Furthermore, control brain sections were processed for

Brain tissue samples were collected from the substantia nigra and striatum of 6 of the included patients, another 4 patients with a definite diagnosis of MSA and 14 control subjects for qRT-PCR analyses of neural and glial cell expression markers. Demographic and autopsy-related data for the two groups are summarized in Table 1. The brain tissue samples were stored at − 80 °C prior to RNA extraction. RNA was

Fig. 1. Giemsa-stained 40-μm coronal sections through the brain at the levels of (a) the subthalamic nucleus and (b) the substantia nigra and red nucleus. (c) Close-up of the red nucleus (*) and substantia nigra (arrowheads) in a control brain. (d) Close-up of the red nucleus (*) and substantia nigra (arrowheads) in a multiple system atrophy brain. (c) and (d) illustrate the loss of pigmented neurons in the substantia nigra in multiple system atrophy. The bars in (a) and (b) are 10 mm and those in (c) and (d) are 5 mm. CN = caudate nucleus, GP = globus pallidus, Hc = hippocampus, MSA = multiple system atrophy, Pu = putamen, RN = red nucleus, SN = substantia nigra, STN = subthalamic nucleus, Th = thalamus.

L. Salvesen et al. / Neurobiology of Disease 74 (2015) 104–113

a

107

b N N

O

N

A

N

A

O

N O

c

d A

GCI

A

O GCI

A A

GCI

O

A

O

Fig. 2. The different cell types were identified by their morphology and spatial distribution as illustrated in a paraffin section from a control brain stained with Giemsa (a). Immunostaining for neurons with anti-NeuN antibody (b) and for astrocytes with anti-glial fibrillary acidic protein antibody (d) in 8-μm-thick sections from putamen of an MSA brain corroborated the morphological criteria. Immunostaining for alpha-synuclein in 8-μm-thick sections from putamen of an MSA brain shows glial cytoplasmic inclusions in oligodendrocytes (c). The bars are 20 μm in (a) and (c) and 50 μm in (b) and (d). N = neuron, A = astrocyte, O = oligodendrocyte, GCI = glial cytoplasmic inclusion.

extracted from 50-mg samples of frozen tissue using the GeneJET RNA Purification Kit (K0732 Thermo Fisher Scientific Inc., Waltham, MA, USA) according to the manufacturer's instructions. Prior to RNA extraction, frozen human brain tissues were diced, placed in a 350 μL lysis buffer supplied with the RNA isolation kit, and homogenized using the MagNA lyser instrument (3000 g, 30 s) and MagNA lyser green beads (03358941001 Roche Diagnostics, Basel, Switzerland). Homogenates were treated with proteinase K for 1 h and then centrifuged at 12,000 g for 5 min. Supernatants were mixed with ethanol and loaded on the GeneJET RNA purification column. Subsequently, impurities were removed by washing steps, as described in the manufacturer's instructions. The RNA was eluted with 100 μL nuclease-free water and subjected to DNAse treatment using the Turbo DNA-free (AM1907 Life Technologies, Carlsbad, CA, USA) according to manufacturer's instructions. RNA integrity and quantity were assessed using the Agilent

2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). All samples had an RNA integrity index ≥6. The RNA samples were stored at −80 °C until use. Sequences for each primer pair are given in Table 3. Briefly, qRTPCR reactions were carried out with One Step Brilliant II SYBR Green QRTPCR Low ROX Master Mix (600835, Agilent Technologies) on the Stratagene Mx3005P QPCR System. Each sample was run in duplicate. The final reaction volume was 10 μL with 0.3 μM of each corresponding primer pair and 2 μL total RNA (5 ng/μL). To compare the multiple samples between the assays, a calibrator (Human Reference Total RNA (636538, Clontech, Mountain View, CA, USA) 25 ng/reaction) and a negative control were included in each run. The thermal cycling parameters were as follows: 30 min at 50 °C (reverse transcription), 10 min at 95 °C followed by 40 cycles of 15 s at 95 °C, 25 s at 62 °C, 25 s at 72 °C, and an extra 5-second acquisition step at

Table 3 PCR primer pairs used for amplification of GFAP = glial fibrillary acidic protein, MBP = myelin basic protein, NeuN = neuronal nuclei protein, OLIG = oligodendrocyte transcription factor, S100B = S100 calcium binding protein B, TNF-α = tumor necrosis factor alpha, IL-1β = interleukin-1 beta, and GAPDH = glyceraldehyde 3-phosphate dehydrogenase. Gene

Accession no.

Primer

Fragment size (in bp)

Sense 5′–3′

Antisense 5′–3′

NEFL NeuN GFAP S100B OLIG1 OLIG2 MBP TNF-α IL-1B GAPDH β-Actin

NM_006158.3 NM_001082575.1 NM_002055.3 NM_006272.2 NM_138983.2 NM_005806.2 NM_001025081.1 NM_000594.2 NM_000576.2 NM_002046.3 NM_001101.3

CTTGGAAGGCGAGGAGACCCGACT TCAATAATGCCACGGCCCGAGTGA CTTCTCCAACCTGCAGATTCG ATGTCTGAGCTGGAGAAGGC CCGCCGGCCAGGTCCTATCA GTGCGCAAGCTTTCCAAGATCG CAAGAACATTGTGACGCCTCGC CTCTGGCCCAGGCAGTCAGATC TGTTGTGGCCATGGACAAGC GACATCAAGAAGGTGGTGAAGCAGG TGACATTAAGGAGAAGCTGTGCTAC

TGGTGTAGTAGGACGGGAAGGAGC ACTGCGCCGACCACTGGATTTAGC CACGGTCTTCACCACGATGTT GTAACCATGGCAACAAAGGC TGTCCGCCGAGGGTCCGTC CTTCATCTCCTCCAGCGAGTTG GCGACTATCTCTTCCTCCCAGC GGCACCACCAGCTGGTTATCT AGGTGCATCGTGCACATAAG CCTGTTGCTGTAGCCAAATTCGTTG ACTTCATGATGGAGTTGAAGGTAGT

158 91 91 245 113 84 217 166 154 201 224

108

L. Salvesen et al. / Neurobiology of Disease 74 (2015) 104–113

80 °C. Primer dimers generated no signal in fluorescence captured at 80 °C. To check for PCR contamination, a negative control of RNA without the reverse transcriptase enzyme was included for each sample analysis. The runs ended with a melting curve analysis between 55 °C and 95 °C for product verification. A comparative cycle of threshold fluorescence method was used, and the relative transcription level of the target gene was normalized to that of average for glyceraldehyde 3-phosphate dehydrogenase or β-actin and expressed as a relative quantity to the calibrator sample using the Pfaffl method (Pfaffl, 2001). Statistics The precision of the estimates was expressed as the coefficient of error (CE) (Gundersen et al., 1999). CE values for the estimates of total cell numbers are provided in the online Supplementary material. For volume estimation, the CE values ranged from 1–3%. The coefficients of variation (CVs), which describe the inter-individual variation for the group, are listed after the group means in Table 4. Sampling is considered optimal when the CE was half or less of the CV (Gundersen and Osterby, 1981). This is the case for all of the

above-mentioned CE values apart from the CE values for astrocytes in the red nucleus of MSA patients and for microglia in general. Thus, the number of astrocytes in the red nucleus and the microglial numbers must be evaluated considering this limitation. The stereological data were analyzed using SigmaPlot 11.0 (Systat Software Inc., San Jose, CA, USA). Differences in total numbers of cells, volumes, and densities between the two groups were tested with Student's t-tests. In a few cases, the test for normal distribution failed, and Mann–Whitney U tests were employed. Potential correlations were tested using the Pearson Product Moment Correlation Test. In the case of multiple pairwise correlations, the significance level was adjusted with the Bonferroni correction. The statistical analyses of the mRNA levels were conducted with nonparametric Mann–Whitney U tests using GraphPad Prism software (Version 5, GraphPad Software Inc., La Jolla, CA, USA). The significance level for all tests was set at p b 0.05 (two-tailed). Results The estimated bilateral quantities of the six nuclei are shown in Table 4. The statistically significant differences in cell numbers detected

Table 4 Numerical estimations of the different cell types in the basal ganglia and red nucleus of patients with MSA and control subjects.

Total cell numbers Neurons

Oligodendrocytes

Astrocytes

Microglia

Density/mm3 Neurons

Oligodendrocytes

Astrocytes

Microglia

Area

Controls

SN PU CN GP STN RN PU CN GP STN RN PU CN GP STN RN PU CN GP STN RN

0.9 188.4 136.9 3.3 0.9 1.9 436.0 289.8 222.6 8.8 19.7 125.6 105.3 37.2 2.8 3.5 7.6 4.9 1.8 0.09 0.1

× × × × × × × × × × × × × × × × × × × × ×

SN PU CN GP STN RN PU CN GP STN RN PU CN GP STN RN PU CN GP STN RN

6.2 38.7 41.2 1.7 8.0 10.8 89.2 86.9 117.8 76.9 104.0 25.9 31.4 19.3 27.6 18.2 1.6 1.5 1.0 0.8 0.5

× × × × × × × × × × × × × × × × × × × × ×

CV

MSA

106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106

0.19 0.21 0.19 0.22 0.39 0.29 0.23 0.19 0.27 0.34 0.23 0.20 0.18 0.30 0.33 0.33 0.51 0.52 0.64 0.38 0.59

0.4 111.6 104.0 1.9 0.8 1.9 345.1 258.0 157.5 10.8 21.4 181.3 137.0 38.2 3.3 4.2 17.4 11.1 3.3 0.2 0.3

× × × × × × × × × × × × × × × × × × × × ×

103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103

0.22 0.14 0.12 0.15 0.27 0.18 0.16 0.09 0.19 0.25 0.09 0.16 0.17 0.19 0.33 0.17 0.49 0.54 0.62 0.38 0.54

4.8 26.3 32.5 1.3 7.1 9.6 88.9 83.3 100.6 84.7 97.0 48.2 44.3 25.3 24.9 18.9 4.3 3.7 2.3 1.5 1.4

× × × × × × × × × × × × × × × × × × × × ×

CV

p-values

Percent change

106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106

0.33 0.41 0.41 0.24 0.19 0.25 0.23 0.28 0.22 0.19 0.23 0.26 0.25 0.31 0.43 0.20 0.32 0.42 0.36 0.33 0.38

0.001* 0.001* 0.03* b0.001* 0.68 0.68 0.04* 0.22 0.01* 0.10 0.41 0.01** 0.04* 0.83 0.41 0.19 0.002* b0.001* 0.02* 0.001* 0.001*

55%↓ 41%↓ 24%↓ 41%↓

103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103 103

0.32 0.26 0.28 0.25 0.23 0.18 0.21 0.19 0.20 0.18 0.13 0.43 0.44 0.46 0.29 0.17 0.31 0.43 0.49 0.33 0.23

0.08 b0.001* 0.02* 0.003* 0.30 0.18 0.96 0.49 0.07 0.33 0.18 b0.001* 0.002* 0.02* 0.492 0.595 b0.001* b0.001* 0.008* 0.001* 0.001**

21%↓ 29%↓

44%↑ 30%↑

128%↑ 129%↑ 145%↑ 104%↑ 222%↑

32%↓ 21%↓ 25%↓

86%↑ 41%↑ 46%↑

172%↑ 146%↑ 130%↑ 79%↑ 171%↑

CN = caudate nucleus, CV = coefficient of variation, GP = globus pallidus, MSA = multiple system atrophy, PU = putamen, RN = red nucleus, SN = substantia nigra, STN = subthalamic nucleus. p-Values represent the results of Student's t-tests (*) or Mann–Whitney U tests (**).

L. Salvesen et al. / Neurobiology of Disease 74 (2015) 104–113

109

Fig. 3. Total numbers of neurons (a–b), oligodendrocytes (c), astrocytes (d), and microglia (e–f) in the basal ganglia nuclei and the red nucleus in control subjects (CS, n = 11) and multiple system atrophy (MSA) brains (○ represents a brain from a patient with the parkinsonian subtype of multiple system atrophy [n = 8]; ● represents a brain from a patient with the cerebellar subtype of multiple system atrophy [n = 3]). Horizontal bars indicate the group means.

between the groups are illustrated in Fig. 3, with distinctions between the patients diagnosed with the cerebellar and parkinsonian subtypes of MSA.

(Table 4, Fig. 3b). The neuronal densities for the two groups are listed in Table 4. Putamen

Substantia nigra There was a mean neuronal loss of 0.5 × 10 6 neurons (~ 55%) in the MSA brains compared to the control brains (p = 0.001)

We found a mean neuronal loss of 76.8 × 106 neurons (~ 41%) in the MSA brains compared to the control brains (p = 0.001). The total number of oligodendrocytes was significantly lower in the MSA brains (p = 0.04), whereas significantly higher numbers of astrocytes

110

L. Salvesen et al. / Neurobiology of Disease 74 (2015) 104–113

(p = 0.01) and microglia (p = 0.002) were observed in the MSA brains compared to the control brains (Table 4, Fig. 3). The overall mean volume of the putamen was = 4.9 cm 3 (range = 3.8–5.8 cm 3) in control subjects and 4.0 cm 3 (range = 2.1–6.2 cm 3 ) in the MSA brains (paraffin-embedded tissue). The mean putaminal volume in the MSA brains was 0.9 mm3 (~ 17.5%) lower than the putaminal volume of the control brains, which was not statistically significant different (p = 0.19). The cell densities are listed in Table 4. When adjusting the p-value with Bonferroni corrections due to multiple pairwise comparisons, we did not find a correlation between the numbers of neurons and microglia (p = 0.02), the numbers of neurons and astrocytes (p = 0.08), or the numbers of astrocytes and microglia (p = 0.10) in the MSA brains. The correlation between the number of neurons and the number of oligodendrocytes was significant after adjustment with the Bonferroni correction (r = 0.821, p = 0.002). Caudate nucleus There was a mean neuronal loss of 32.9 × 106 neurons (~24%) in the MSA brains compared to the control brains (p = 0.03). There was no statistically significant difference in the total numbers of oligodendrocytes between the groups (p = 0.22). The total numbers of astrocytes (p = 0.04) and microglia (p b 0.001) were significantly higher in the MSA brains compared to the control brains (Table 4, Fig. 3). The overall mean volumes of the caudate nucleus were 3.3 cm3 (range = 2.3–4.2 cm3) in the control brains and 3.1 cm3 (range = 1.9–4.1 cm3) in the MSA brains (paraffin-embedded tissue). The difference between the two groups was not statistically significant (p = 0.41). The cell densities are listed in Table 4. No correlation was found between the numbers of neurons and astrocytes (p = 0.09), neurons and microglia (p = 0.85), or astrocytes and microglia (p = 0.28). Globus pallidus There was a mean neuronal loss of 1.4 × 106 neurons (~41%) in the MSA brains compared to the control brains (p b 0.001). Additionally, the total number of oligodendrocytes was significantly lower in the MSA brains compared to the control brains (p = 0.01). There was no statistical difference in the total number of astrocytes between the two groups (p = 0.83). The total number of microglia was higher in the MSA brains compared to the control brains (p = 0.02) (Table 4, Fig. 3). The overall mean volume of the globus pallidus was 1.9 cm3 (range = 1.2–2.6 cm3) in the control brains and 1.6 cm3 (range = 0.9–2.4 cm3) in the MSA brains (paraffin-embedded tissue). Although the mean volume of the globus pallidus in the MSA brains was 0.3 cm3 (~18%) lower than one of the control brains, no statistically significant difference was found between the two groups (p = 0.09). The cell densities are listed in Table 4. The density of astrocytes was significantly higher in the MSA brains compared to the control brains (p = 0.02), although no difference in the total number of astrocytes was found. The correlation between the numbers of neurons and oligodendrocytes was significant after Bonferroni correction (r = 0.887, p b 0.001). We did not observe a correlation between the numbers of neurons and microglia (p = 0.70). Subthalamic nucleus The total numbers of neurons in the subthalamic nucleus did not differ between the MSA brains and the controls brains (p = 0.68) (Table 4). No difference was found between the two groups with regard to the total number of oligodendrocytes (p = 0.10) or astrocytes (p = 0.41). However, a significantly higher number of microglia was found in the MSA brains compared to the control brains (p = 0.001) (Table 4, Fig. 3f).

The overall mean volume of the subthalamic nucleus was 114.4 mm3 (range = 84.7–192.6 mm3) in the control brains and 128.8 mm3 (range = 102.5–177.4 mm3) in the MSA brains (paraffin-embedded tissue). We did not find a significant difference between the two groups (p = 0.09). The cell densities are listed in Table 4. Red nucleus There was no significant differences between the two groups in the total numbers of neurons (p = 0.68), oligodendrocytes (p = 0.41), or astrocytes (p = 0.19). However, significantly more microglia were found in the MSA brains compared to the control brains (p = 0.001) (Table 4, Fig. 3f). The overall mean volume of the red nucleus was 193.2 mm3 (range = 128.8–275.8 mm3) in the control brains and 222.2 mm3 (range = 143.3–308.5 mm3) in the MSA brains (paraffin-embedded tissue). No significant difference between the two groups was found (p = 0.16). The cell densities are listed in Table 4. Correlations There was no correlation between the number of pigmented neurons in the substantia nigra and the total number of neurons in the putamen (p = 0.24), caudate nucleus (p = 0.82), or globus pallidus (p = 0.19). In contrast, a positive correlation was found between the numbers of neurons in the putamen and the globus pallidus (r = 0.751, p = 0.007). The correlations between the number of neurons in the putamen and caudate nucleus (p = 0.01) and between the neuronal numbers of the caudate nucleus and globus pallidus (p = 0.02) were not significant after Bonferroni correction. Three of the 11 patients with MSA were clinically classified as having the cerebellar subtype, and 8 of the patients had the parkinsonian subtype (Table 2). Due to the low number of patients, no statistical comparisons were performed between the two subtypes. As illustrated in Fig. 3, there was no difference in the numbers of neurons or glial cells between the two groups. No correlation between the total numbers of neurons in any of the regions and the disease duration or disease severity (expressed by number of positive red flag categories and time in years from onset to wheelchair-bound state) (Table 2) (Köllensperger et al., 2008) was found. Cell marker mRNA level The qRT-PCR analysis results are shown in Fig. 4. We observed decreased levels of neuronal and oligodendroglial markers in tissue samples from both the substantia nigra and striatum of the MSA brains compared to control brains (Fig. 4a and b). In contrast, the levels of astroglial and activated microglial markers were increased in the tissue samples from substantia nigra and striatum of MSA brains in comparison with control brains (Fig. 4c and d). Discussion To our knowledge, this is the first study to stereologically quantify the total numbers of neurons and glial cells in the basal ganglia and red nucleus of patients with MSA. The most consistent finding in this study was a higher number of microglia in MSA brains compared to control brains. We found a significant increase in the number of microglia in all of the examined nuclei independent of neuronal loss. This finding was specific for microglia, as the changes seen in the numbers of oligodendrocytes and astrocytes were restricted to areas with concomitant neuronal loss. We did not observe correlations between the total numbers of microglia and neurons in the putamen, caudate nucleus, or globus pallidus, where concomitant neuronal loss was observed. This

L. Salvesen et al. / Neurobiology of Disease 74 (2015) 104–113

b 15 p=0.02

p=0.03

5

0

CS

MSA

Substantia nigra

CS

40 p=0.05

0

MSA

CS

MSA

Substantia nigra

Relative expression levels

80

MSA

p=0.04

p=0.01

p=0.02

40 20 MSA HC Substantia nigra

CS

MSA

MSA HC Striatum

GFAP

CS

MSA

Substantia nigra

CS

MSA

Striatum

S100

CS

Substantia nigra

Striatum

p=0.02

p=0.01

MSA

Striatum

CS

CS

MSA

Substantia nigra

MSA

Striatum

OLIG2

OLIG1

d

60

0

CS

MBP

p=0.03

p=0.02

10

Striatum

100

p=0.05

20

NeuN

c

p=0.02

30

Relative expression levels

10

Relative expression levels

Relative expression levels

a

111

6

p=0.0004 p=0.01

4 p=0.02

2 p=0.05

0

CS

MSA

Substantia nigra

CS

MSA

Striatum

TNF- 

CS

MSA

CS

Substantia nigra

MSA

Striatum

IL-1

Fig. 4. Semi-quantitative real-time PCR analysis of (a) neuronal, (b) oligodendroglial, (c) astroglial, and (d) microglial markers in the substantia nigra and striatum from control brains (black bars, n = 14) and multiple system atrophy brains (white bars, n = 10). The depicted values are mean ± SEM. The p-values represent the results of Mann–Whitney U tests. CS = control subjects, MSA = multiple system atrophy.

is in agreement with the lack of correlation between activated microglia and lesion severity in gray matter regions of MSA brains reported by Ishizawa et al. (2008). Our results demonstrate that microglial proliferation is a substantial and widespread pathological feature. Further, our finding of increased microglia in areas without neuronal loss may indicate that microgliosis precedes neuronal loss. We found a substantial neuronal loss in the substantia nigra, caudate nucleus, putamen, and globus pallidus of patients with MSA compared to control subjects. Our results confirm that the loss of striatal neurons is more severe in the putamen compared to the caudate nucleus. However, a notable discrepancy between our results and those of earlier semi-quantitative studies is the involvement of globus pallidus. While neurodegeneration of the globus pallidus was previously described as less frequent and severe compared to putaminal neurodegeneration, we found a substantial loss of neurons in the globus pallidus equal to that observed in the putamen. Orofacial dystonia and abnormal postures, such as the Pisa syndrome, are features supporting a MSA diagnosis (Gilman et al., 2008), but the underlying pathology is unknown. However, lesions of the internal segment of globus pallidus have previously been shown to produce generalized or axial dystonia (Bucher et al., 1996), and reduced output of the internal segment of the globus pallidus may cause involuntary movements observed in dystonia (Nambu et al., 2011). The substantial neuronal loss in the globus pallidus described here may be associated with these characteristic features. Our results showed a smaller loss of pigmented neurons in the substantia nigra of patients with MSA compared to patients with other

parkinsonian syndromes (Hardman et al., 1997a; Pakkenberg et al., 1991). A possible explanation could be the inclusion of patients with the cerebellar subtype of MSA. Compared to patients with the parkinsonian subtype, they show milder or subclinical parkinsonism and less severe neuropathological damage in the substantia nigra (Gilman et al., 2008; Ozawa et al., 2004). Neuroimaging with visualization of in-vivo binding of pre-synaptic dopamine transporters in patients with the cerebellar subtype of MSA reveals less severe reductions or even normal values compared to patients with the parkinsonian subtype (Lu et al., 2004; Munoz et al., 2011). Oligodendrocytes are presumed to play a major role in MSA pathogenesis (Wenning et al., 2008), and a previous study has shown that apoptosis mainly involves oligodendrocytes (Probst-Cousin et al., 1998). It has been suggested that oligodendroglial cell death caused by the formation of GCIs may be an early disease event that contributes to the acceleration of reactive gliosis and demyelination and eventually leads to neuronal degeneration (Ahmed et al., 2012; Fellner and Stefanova, 2013; Probst-Cousin et al., 1998; Wakabayashi and Takahashi, 2006). However, due to evidence of potential repair mechanisms in GCI-bearing oligodendrocytes (Kato et al., 2000; Probst-Cousin et al., 1998), it has also been proposed that the presence of GCIs might only cause cellular dysfunction (Wenning et al., 2008; Yoshida, 2007). No significant difference was found between the MSA brains and the control brains in the total number of oligodendrocytes of the caudate nucleus, red nucleus or subthalamic nucleus. In the putamen and globus pallidus a moderate but not severe loss of oligodendrocytes was observed. This indicates that oligodendroglial cell loss may not be a

112

L. Salvesen et al. / Neurobiology of Disease 74 (2015) 104–113

widespread and substantial neuropathological feature of MSA supporting previous hypotheses of the formation of GCIs causing cellular dysfunction but not necessarily cell death (Wenning et al., 2008; Yoshida, 2007). In the putamen and globus pallidus a severe neuronal loss was seen concomitant to the oligodendroglial cell loss and in both nuclei the number of oligodendrocytes was positively correlated to the number of neurons. Our findings may indicate that oligodendrocytes degenerate in areas with severe neurodegeneration and presuming that some of the degenerated oligodendrocytes contained GCIs, our results may explain previous findings of a low density of GCIs in regions with severe neurodegeneration (Jellinger et al., 2005; Ozawa et al., 2004; Yoshida, 2007). Compared to the results of previous studies using cell-counting methods to examine the basal ganglia nuclei of control subjects, the estimates of total neuronal numbers in our study were generally higher (Hardman et al., 1997b; Hardman and Halliday, 1999a,b; Kreczmanski et al., 2007). This discrepancy might be a result of differences in tissue processing, staining methods, delineation, or stereological design. Due to a low number of microglia counted, the CE values were correspondingly high, and our microglia estimates must be considered with this limitation in mind. The main limitation of the present study is the small sample size, and our negative results must be interpreted with caution. Verification of our results in studies including larger sample sizes is needed to substantiate the present observations. Cell quantification was performed using Giemsa-stained sections, and our results are therefore based on cell morphology alone However, astroglial- and neuronal-specific immunohistochemistry validated our cell identification criteria. The results from the stereological study were supported by cell marker expression analyses using qRT-PCR. In both of the examined regions (substantia nigra and striatum), we observed significantly decreased mRNA levels for neuronal- and oligodendroglial-specific markers and significantly increased levels of markers for astroglia and activated microglia in MSA brains compared to control brains. The finding of increased markers for activated microglia is in agreement with previous studies (Brudek et al., 2013; Stefanova et al., 2007). Microglial overactivation can lead to the release of deleterious and neurotoxic factors, including pro-inflammatory cytokines such as tumor necrosis factor-α and interleukin-1β, which facilitate chronic neuroinflammation. Pro-inflammatory cytokines may induce astrogliosis and cause neuronal dysfunction and death (Balasingam et al., 1996; Qian et al., 2010; Tansey and Goldberg, 2010). Activated microglia also play a fundamental role in the clearance of α-synuclein and damaged or necrotic cells (Fellner and Stefanova, 2013). It is unknown if microglia activation and neuroinflammation in MSA lead to neurodegeneration or if it is a secondary response to neuronal loss (Ahmed et al., 2012). Conclusion In summary, we employed a robust stereological counting method to demonstrate substantial neuronal loss in the substantia nigra, striatum, and globus pallidus of patients with MSA, while neurons in other brain regions were spared, supporting the region-specific patterns of neuropathological changes in MSA. In animal models of MSA microgliosis has been consistent and severe (Ahmed et al., 2012). The present results demonstrate that this feature is also present in brains from patients with MSA, who had been followed by movement disorder specialists for years and were clinically and neuropathologically well characterized. To what extent microglia-dependent inflammatory processes underlie MSA pathogenesis remains to be elucidated. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.nbd.2014.11.008. Funding The study was supported by the Velux Foundation (VELUX26781), the Danish Parkinson's Disease Association, the Bispebjerg Hospital

Research Foundation, the Danish Multiple System Atrophy Foundation, the Augustinus Foundation (10-3820), The Lundbeck Foundation (2012-12359), and Danmodis. The Bispebjerg Movement Disorders Biobank is supported by the John and Birthe Meyer Foundation and the Jascha Foundation. Conflict of interest The authors declare that they have no conflict of interest. The Ethics Committee for the Copenhagen Regional Area approved the study (H-C-2008-048). Acknowledgments We thank Susanne Sørensen for providing technical assistance. References Ahmed, Z., Asi, Y.T., Sailer, A., Lees, A.J., Houlden, H., Revesz, T., Holton, J.L., 2012. The neuropathology, pathophysiology and genetics of multiple system atrophy. Neuropathol. Appl. Neurobiol. 38, 4–24. Balasingam, V., Dickson, K., Brade, A., Yong, V.W., 1996. Astrocyte reactivity in neonatal mice: apparent dependence on the presence of reactive microglia/macrophages. Glia 18, 11–26. Boyce, R.W., Dorph-Petersen, K.A., Lyck, L., Gundersen, H.J., 2010. Design-based stereology: introduction to basic concepts and practical approaches for estimation of cell number. Toxicol. Pathol. 38, 1011–1025. Brudek, T., Winge, K., Agander, T.K., Pakkenberg, B., 2013. Screening of Toll-like receptors expression in multiple system atrophy brains. Neurochem. Res. 38, 1252–1259. Bucher, S.F., Seelos, K.C., Dodel, R.C., Paulus, W., Reiser, M., Oertel, W.H., 1996. Pallidal lesions. Structural and functional magnetic resonance imaging. Arch. Neurol. 53, 682–686. Dickson, D.W., 2012. Parkinson's disease and parkinsonism: neuropathology. Cold Spring Harb. Perspect. Med. 2. Fabricius, K., Jacobsen, J.S., Pakkenberg, B., 2013. Effect of age on neocortical brain cells in 90+ year old human females—a cell counting study. Neurobiol. Aging 34, 91–99. Fellner, L., Stefanova, N., 2013. The role of glia in alpha-synucleinopathies. Mol. Neurobiol. 47, 575–586. Fujishiro, H., Ahn, T.B., Frigerio, R., DelleDonne, A., Josephs, K.A., Parisi, J.E., Eric, A.J., Dickson, D.W., 2008. Glial cytoplasmic inclusions in neurologically normal elderly: prodromal multiple system atrophy? Acta Neuropathol. 116, 269–275. Gilman, S., Wenning, G.K., Low, P.A., Brooks, D.J., Mathias, C.J., Trojanowski, J.Q., Wood, N.W., Colosimo, C., Durr, A., Fowler, C.J., Kaufmann, H., Klockgether, T., Lees, A., Poewe, W., Quinn, N., Revesz, T., Robertson, D., Sandroni, P., Seppi, K., Vidailhet, M., 2008. Second consensus statement on the diagnosis of multiple system atrophy. Neurology 71, 670–676. Gundersen, H.J., Jensen, E.B., 1987. The efficiency of systematic sampling in stereology and its prediction. J. Microsc. 147, 229–263. Gundersen, H.J., Osterby, R., 1981. Optimizing sampling efficiency of stereological studies in biology: or ‘do more less well!’. J. Microsc. 121, 65–73. Gundersen, H.J., Bagger, P., Bendtsen, T.F., Evans, S.M., Korbo, L., Marcussen, N., Moller, A., Nielsen, K., Nyengaard, J.R., Pakkenberg, B., 1988. The new stereological tools: disector, fractionator, nucleator and point sampled intercepts and their use in pathological research and diagnosis. APMIS 96, 857–881. Gundersen, H.J., Jensen, E.B., Kieu, K., Nielsen, J., 1999. The efficiency of systematic sampling in stereology—reconsidered. J. Microsc. 193, 199–211. Hardman, C.D., Halliday, G.M., 1999a. The external globus pallidus in patients with Parkinson's disease and progressive supranuclear palsy. Mov. Disord. 14, 626–633. Hardman, C.D., Halliday, G.M., 1999b. The internal globus pallidus is affected in progressive supranuclear palsy and Parkinson's disease. Exp. Neurol. 158, 135–142. Hardman, C.D., Halliday, G.M., McRitchie, D.A., Cartwright, H.R., Morris, J.G., 1997a. Progressive supranuclear palsy affects both the substantia nigra pars compacta and reticulata. Exp. Neurol. 144, 183–192. Hardman, C.D., Halliday, G.M., McRitchie, D.A., Morris, J.G., 1997b. The subthalamic nucleus in Parkinson's disease and progressive supranuclear palsy. J. Neuropathol. Exp. Neurol. 56, 132–142. Ishizawa, K., Komori, T., Sasaki, S., Arai, N., Mizutani, T., Hirose, T., 2004. Microglial activation parallels system degeneration in multiple system atrophy. J. Neuropathol. Exp. Neurol. 63, 43–52. Ishizawa, K., Komori, T., Arai, N., Mizutani, T., Hirose, T., 2008. Glial cytoplasmic inclusions and tissue injury in multiple system atrophy: a quantitative study in white matter (olivopontocerebellar system) and gray matter (nigrostriatal system). Neuropathology 28, 249–257. Jellinger, K.A., Seppi, K., Wenning, G.K., 2005. Grading of neuropathology in multiple system atrophy: proposal for a novel scale. Mov. Disord. 20 (Suppl. 12), S29–S36. Karlsen, A.S., Pakkenberg, B., 2011. Total numbers of neurons and glial cells in cortex and basal ganglia of aged brains with Down syndrome—a stereological study. Cereb. Cortex 21, 2519–2524. Karlsen, A.S., Korbo, S., Uylings, H.B., Pakkenberg, B., 2014. A stereological study of the mediodorsal thalamic nucleus in Down syndrome. Neuroscience 279C, 253–259.

L. Salvesen et al. / Neurobiology of Disease 74 (2015) 104–113 Kato, S., Shinozawa, T., Takikawa, M., Kato, M., Hirano, A., Awaya, A., Ohama, E., 2000. Midkine, a new neurotrophic factor, is present in glial cytoplasmic inclusions of multiple system atrophy brains. Acta Neuropathol. 100, 481–489. Kollensperger, M., Geser, F., Seppi, K., Stampfer-Kountchev, M., Sawires, M., Scherfler, C., Boesch, S., Mueller, J., Koukouni, V., Quinn, N., Pellecchia, M.T., Barone, P., Schimke, N., Dodel, R., Oertel, W., Dupont, E., Ostergaard, K., Daniels, C., Deuschl, G., Gurevich, T., Giladi, N., Coelho, M., Sampaio, C., Nilsson, C., Widner, H., Sorbo, F.D., Albanese, A., Cardozo, A., Tolosa, E., Abele, M., Klockgether, T., Kamm, C., Gasser, T., Djaldetti, R., Colosimo, C., Meco, G., Schrag, A., Poewe, W., Wenning, G.K., 2008. Red flags for multiple system atrophy. Mov. Disord. 23, 1093–1099. Kreczmanski, P., Heinsen, H., Mantua, V., Woltersdorf, F., Masson, T., Ulfig, N., SchmidtKastner, R., Korr, H., Steinbusch, H.W., Hof, P.R., Schmitz, C., 2007. Volume, neuron density and total neuron number in five subcortical regions in schizophrenia. Brain 130, 678–692. Lu, C.S., Weng, Y.H., Chen, M.C., Chen, R.S., Tzen, K.Y., Wey, S.P., Ting, G., Chang, H.C., Yen, T.C., 2004. 99mTc-TRODAT-1 imaging of multiple system atrophy. J. Nucl. Med. 45, 49–55. Munoz, E., Iranzo, A., Rauek, S., Lomena, F., Gallego, J., Ros, D., Santamaria, J., Tolosa, E., 2011. Subclinical nigrostriatal dopaminergic denervation in the cerebellar subtype of multiple system atrophy (MSA-C). J. Neurol. 258, 2248–2253. Nambu, A., Chiken, S., Shashidharan, P., Nishibayashi, H., Ogura, M., Kakishita, K., Tanaka, S., Tachibana, Y., Kita, H., Itakura, T., 2011. Reduced pallidal output causes dystonia. Front. Syst. Neurosci. 5, 89. O'Sullivan, S.S., Massey, L.A., Williams, D.R., Silveira-Moriyama, L., Kempster, P.A., Holton, J.L., Revesz, T., Lees, A.J., 2008. Clinical outcomes of progressive supranuclear palsy and multiple system atrophy. Brain 131, 1362–1372. Ozawa, T., Paviour, D., Quinn, N.P., Josephs, K.A., Sangha, H., Kilford, L., Healy, D.G., Wood, N.W., Lees, A.J., Holton, J.L., Revesz, T., 2004. The spectrum of pathological involvement of the striatonigral and olivopontocerebellar systems in multiple system atrophy: clinicopathological correlations. Brain 127, 2657–2671. Pakkenberg, B., Gundersen, H.J., 1997. Neocortical neuron number in humans: effect of sex and age. J. Comp. Neurol. 384, 312–320. Pakkenberg, B., Moller, A., Gundersen, H.J., Mouritzen, D.A., Pakkenberg, H., 1991. The absolute number of nerve cells in substantia nigra in normal subjects and in patients with Parkinson's disease estimated with an unbiased stereological method. J. Neurol. Neurosurg. Psychiatry 54, 30–33. Pedersen, K.M., Marner, L., Pakkenberg, H., Pakkenberg, B., 2005. No global loss of neocortical neurons in Parkinson's disease: a quantitative stereological study. Mov. Disord. 20, 164–171.

113

Pelvig, D.P., Pakkenberg, H., Stark, A.K., Pakkenberg, B., 2008. Neocortical glial cell numbers in human brains. Neurobiol. Aging 29, 1754–1762. Pfaffl, M.W., 2001. A new mathematical model for relative quantification in real-time RTPCR. Nucleic Acids Res. 29, e45. Poewe, W., Wenning, G., 2002. The differential diagnosis of Parkinson's disease. Eur. J. Neurol. 9 (Suppl. 3), 23–30. Probst-Cousin, S., Rickert, C.H., Schmid, K.W., Gullotta, F., 1998. Cell death mechanisms in multiple system atrophy. J. Neuropathol. Exp. Neurol. 57, 814–821. Qian, L., Flood, P.M., Hong, J.S., 2010. Neuroinflammation is a key player in Parkinson's disease and a prime target for therapy. J. Neural Transm. 117, 971–979. Sigaard, R.K., Kjaer, M., Pakkenberg, B., 2014. Development of the cell population in the brain white matter of young children. Cereb. Cortex [epub ahead of print], http:// dx.doi.org/10.1093/cercor/bhu178. Stefanova, N., Reindl, M., Neumann, M., Kahle, P.J., Poewe, W., Wenning, G.K., 2007. Microglial activation mediates neurodegeneration related to oligodendroglial alphasynucleinopathy: implications for multiple system atrophy. Mov. Disord. 22, 2196–2203. Tansey, M.G., Goldberg, M.S., 2010. Neuroinflammation in Parkinson's disease: its role in neuronal death and implications for therapeutic intervention. Neurobiol. Dis. 37, 510–518. Trojanowski, J.Q., Revesz, T., 2007. Proposed neuropathological criteria for the post mortem diagnosis of multiple system atrophy. Neuropathol. Appl. Neurobiol. 33, 615–620. Wakabayashi, K., Takahashi, H., 2006. Cellular pathology in multiple system atrophy. Neuropathology 26, 338–345. Walloe, S., Pakkenberg, B., Fabricius, K., 2014. Stereological estimation of total cell numbers in the human cerebral and cerebellar cortex. Front. Hum. Neurosci. 8, 508. Wenning, G.K., Quinn, N., Magalhaes, M., Mathias, C., Daniel, S.E., 1994. “Minimal change” multiple system atrophy. Mov. Disord. 9, 161–166. Wenning, G.K., Tison, F., Ben, S.Y., Daniel, S.E., Quinn, N.P., 1997. Multiple system atrophy: a review of 203 pathologically proven cases. Mov. Disord. 12, 133–147. Wenning, G.K., Seppi, K., Tison, F., Jellinger, K., 2002. A novel grading scale for striatonigral degeneration (multiple system atrophy). J. Neural Transm. 109, 307–320. Wenning, G.K., Stefanova, N., Jellinger, K.A., Poewe, W., Schlossmacher, M.G., 2008. Multiple system atrophy: a primary oligodendrogliopathy. Ann. Neurol. 64, 239–246. West, M.J., 2012. Introduction to stereology. Cold Spring Harb. Protoc. 8, 843–851. Yoshida, M., 2007. Multiple system atrophy: alpha-synuclein and neuronal degeneration. Neuropathology 27, 484–493.