Neurobiology of Aging
Volume 33, Issue 1 , Pages 197.e11-197.e20, January 2012

Stereological quantification of the cerebellum in patients with Alzheimer's disease

  • Kjeld Andersen

      Affiliations

    • Research Laboratory for Stereology and Neuroscience, Bispebjerg University Hospital, Copenhagen, NV, Denmark
    • Corresponding Author InformationCorresponding author at: Research Laboratory for Stereology and Neuroscience, Bispebjerg University Hospital, Bispebjerg Bakke 23, DK-2400 Copenhagen NV, Denmark. Tel.: +45 35316421; fax: +45 35316434
    • ,
  • Birgitte Bo Andersen

      Affiliations

    • Research Laboratory for Stereology and Neuroscience, Bispebjerg University Hospital, Copenhagen, NV, Denmark
    • Memory Disorder Research Group, Department of Neurology, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark
    • ,
  • Bente Pakkenberg

      Affiliations

    • Research Laboratory for Stereology and Neuroscience, Bispebjerg University Hospital, Copenhagen, NV, Denmark

Received 3 November 2009; received in revised form 27 April 2010; accepted 21 June 2010. published online 23 August 2010.

Article Outline

Abstract 

Nonquantitative studies indicate that the cerebellum is neuropathologically affected in Alzheimer's disease; however, no quantitative studies on the subject have yet been conducted. Ten cerebella from elderly female subjects with severe Alzheimer's disease and 10 age- and gender-matched controls were examined. The cerebellum was divided into 5 regions and the Purkinje and granule cell number and density, cortical volume, molecular and granular layer volume and thickness, white matter volume, surface area, and the Purkinje cell gradient were stereologically estimated. There was no significant difference between the groups in Purkinje or granule cell number or density, and no overall difference in Purkinje cell gradient. However, there was a significant 12.7% reduction in total cerebellar volume in the Alzheimer's group and significant localized differences between the groups regarding other parameters. The relative lack of neuropathological changes in the cerebellum of severely demented Alzheimer's patients suggests that neuronal cell bodies on a global scale apparently still are intact.

Keywords: Cerebellum, Alzheimer's disease, Stereology, Cell gradient, Quantification, Cell number, Cell density

 

Back to Article Outline

1. Introduction 

Alzheimer's disease (AD) is characterized by extracellular amyloid plaques and intraneuronal neurofibrillary changes consisting of neuritic plaques (NP), neuropil threads (NT), and neurofibrillary tangles (NFT) in cerebral tissue. Because neurofibrillary changes are seen in nondemented individuals as well as in patients with AD and can be absent in demented individuals (Tomlinson et al., 1968, Tomlinson et al., 1970), the diagnosis of AD is given by the clinicopathological correlation between these changes and a clinical history of dementia (Alafuzoff et al., 2008, Tomlinson et al., 1968, Tomlinson et al., 1970). The neuropathological staging of the disease is described by Braak and Braak (1991), and based on the regional distribution of neurofibrillary tangles (NFT) and neuropil threads (NT). Initially, the transentorhinal region and hippocampus are the primary affected structures. As the disease progresses, changes are evident in the limbic system and subcortical nuclei, and finally neocortical regions are affected (Braak and Braak, 1991). No neurofibrillary tangles are observed in the cerebellar cortex (Aikawa et al., 1985, Azzarelli et al., 1985, Wegiel et al., 1999, Yamamoto and Hirano, 1985). Stereological studies have shown a reduction in the number of neurons in the hippocampus of AD patients (Simic et al., 1997, West et al., 1994, West et al., 2000) and a reduction of neocortical volume and neocortical thickness (Regeur, 2000), but no reduction of cortical surface area (Oster et al., 1993). Bundgaard et al. (2001) have shown an increasing size of neocortical neurons in patients with AD and stereological studies have so far not shown a global change of neocortical neurons (Regeur et al., 1994) and glial cells (Pelvig et al., 2003) in patients with AD. Semiquantitative studies have shown a reduced number of synapses (DeKosky and Scheff, 1990, Terry et al., 1991, Zhan et al., 1993) and dendritic spines (Buell and Coleman, 1979, Coleman and Flood, 1987, Flood, 1991) in the cerebral cortex of patients with AD, and others suggest that loss of cortical synapses and dendrites may be responsible for reduced cognitive function in AD patients, in the absence of a global cortical cell loss (Stark et al., 2005). However, Jørgensen et al. (2008) found no difference in white fiber length, fiber length density, or subcortical white matter volume in patients with AD compared with controls.

Traditionally, the cerebellum was considered a structure controlling motor regulation such as movement, gait, posture, balance, motor coordination, and learning (Albus, 1971, Bastian et al., 1998, Marr, 1969, Thach, 1996, Wegiel et al., 1999). However, in recent years it has been acknowledged that the cerebellum exerts nonmotor functions as well, such as cognitive, behavioral, and affective processing (Bower, 1997, Ito, 1997, Leiner et al., 1991, Timmann and Daum, 2007, Wegiel et al., 1999). Impairment of the cerebellum has been associated with cognition since the recognition of the cerebellar cognitive affective syndrome (CCAS). CCAS includes impairments in executive, visual-spatial, and linguistic abilities, with affective disturbance ranging from emotional blunting and depression to disinhibition and psychotic features. The cognitive and psychiatric components of the CCAS, together with the ataxic motor disability of cerebellar disorders, are conceptualized within the dysmetria of thought hypothesis. This concept holds that a universal cerebellar transform facilitates automatic modulation of behavior around a homeostatic baseline, and the behavior being modulated is determined by the specificity of anatomic subcircuits, or loops, within the cerebrocerebellar system. Damage to the cerebellar component of the distributed neural circuit subserving sensorimotor, cognitive, and emotional processing disrupts the universal cerebellar transform, leading to the universal cerebellar impairment affecting the lesioned domain. Cognitive and emotional disorders may accompany cerebellar diseases or be their principal clinical presentation, and this has significance for the diagnosis and management of patients with cerebellar dysfunction (Schmahmann and Sherman, 1998; Schmahmann, 1991; Schmahmann, 2004). Impairment of the cerebellum is also seen in psychiatric disorders (Andreasen et al., 1999, Baldaçara et al., 2008, Bugalho et al., 2006, Giedd et al., 2001, Hamilton et al., 1983, Konarski et al., 2005, Lane et al., 1997, Mackie et al., 2007, Nasrallah et al., 1981, Sassi and Soares, 2001, Schmahmann et al., 2007, Soares and Mann, 1997), including schizophrenia (Andreasen and Pierson, 2008, Andreasen et al., 1996, Ichimiya et al., 2001). So far only few studies have dealt with the quantitative neuropathological consequences of AD in the cerebellum. Sjöbeck and Englund (2001) found, in a semiquantitative study, a significant reduced linear Purkinje cell density in the vermis in AD cerebella compared with a control group. In another semiquantitative study, Wegiel et al. (1999) found a decreased volume of the molecular and granular layer and a loss of Purkinje and granule cells.

The cognitive and motor impairments in AD combined with the recently discovered cognitive function of the cerebellum might suggest that neuropathological changes are evident in this structure. However, it still remains unclear whether or to what extent the cerebellum is quantitatively affected in AD, as measured with unbiased quantitative methods. The current semiquantitative studies on the subject do not provide solid evidence of cerebellar affection in AD and because patients with AD typically do not present with symptoms of cerebellar dysfunction, it is proposed that AD is not quantitatively affecting the cerebellum, thereby rejecting a clinical-pathological correlation between the cerebellum and this disease.

Back to Article Outline

2. Methods 

2.1. Material 

The brains were collected from 1971 to 1987 following the Danish law on autopsied human tissue and have been used in previous studies of the neocortical nerve cell loss (Jørgensen et al., 2008, Pelvig et al., 2003, Regeur et al., 1994). Appropriate Institutional Review Board approval was obtained and procedures were used concerning human subjects. All the demented patients came from a chronic psychogeriatric ward in Copenhagen, Denmark, and all had been evaluated prospectively with a psychometric dementia test and neurological examinations once a year during their last years. The neuropsychological investigations were performed before the modern dementia rating scales were established, e.g., the MiniMental State Examination (MMSE) (Folstein et al., 1975), Consortium to Establish a Registry for Alzheimer's disease (CERAD) (Morris et al., 1989) evaluation, the Alzheimer's Disease Assessment Scale – cognitive section (ADAS-cog) (Mohs et al., 1983), and the Global Deterioration Scale (GDS) (Reisberg et al., 1982), but the tests assessed a wide range of cognitive functions such as attention, orientation, long-term memory, naming, word recall, word recognition, comprehension, constructional spoken language, and ideational praxis, word finding, and delayed recall. A row of subtests from different test batteries were used, among others subtests from Wais, Wechsler, Ranshburg, Kimura, Goldstein-Scheerer, Stroop-test and others (see Regeur, 2000). A slow progression in dementia was found from year to year. All patients had initially severe memory complaints and died with global cognitive disorder. All in all, the patients fulfilled the McKhann criteria from 1984 (McKhann et al., 1984). Performances on the different psychological subtests were scored on a 7-point scale, and a general score was obtained from 1 to 7 with a high score indicating severe dementia.

A total of 46 brains from demented patients were available. Of these some were excluded due to strokes, concomitant Parkinson's disease, or other serious illnesses while 15 were excluded because of a non-Alzheimer dementia diagnosis leaving 12 brains from AD patients. However, due to poor quality or missing tissue, 3 of the cerebella were excluded, and ultimately 10 cerebella from the control group and 10 cerebella from AD patients were included. The average age for the AD subjects was 82.3 years, (range 79–88 years), while the 10 control females were between 74 and 90 years of age (average age 82.9 years) and had died of nonneurological illness, excluding all subjects with diseases that might affect the central nervous system (CNS) (Table 1). No AD subjects had a score lower than 5, so the material comprised only brains from the most severely demented Alzheimer females with the subjects having an average score of 5.6. The pathoanatomical diagnosis of AD was based on the presence of numerous widespread senile plaques in the neocortex shown after application of a specific stains (see below), but the isocortex also showed tangles and amyloid angiopathy comparable to Braak stages V and VI (Braak and Braak, 1991). None were excluded due to white matter changes and none of the cases met with a history or neuropathology of frontal lobe dementia, or Lewy body disease.

Table 1. The age, body height, body weight, brain weight, cerebellar weight, and cause of death for Alzheimer and control subjects
nAge (years)Height (cm)Body weight (kg)Brain weight (g)Cerebellar weight (g)Cause of death
Alzheimer's patients
17916846103572Pneumonia
280156761125100Cardiac failure
380153401000107Cardiac failure
480156381165102Cardiac failure
581157431165113AMI
68115652124597Pulmonary edema
783152401180101Pneumonia
88515864117585Pneumonia
986159631180116Pneumonia
1088158641120107AMI
Mean82.3(0.04)157.3(0.03)52.6(0.26)1139(0.06)100.0(0.13)
Controls
174147531220117AMI
277162551330109AMI
38016247970105Cardiac failure
48115751109099AMI
58515939118096AMI
689153551190112AMI
790165601310114Peritonitis
884155671265120AMI
985159721410121AMI
1084160571236120AMI
Mean82.9(0.06)157.9(0.032)55.6(0.17)1220(0.10)111(0.08)

The CVs are given in parenthesis.

Key: AMI, acute myocardial infarction; CV, coefficient of variation.

In Denmark people do not move much and the local practitioners and hospital departments know their patients well. The hospital files of the donors, the autopsy report, as well as some files from the local practitioners were carefully studied to exclude individuals with diseases that might affect the central nervous system, such as dementia or neurodegenerative illnesses, cerebrovascular diseases, metastatic cancer, diabetes, hypertension, or abuse of alcohol or drugs. Although no cognitive assessment was performed, the hospital staff or their local practitioner regarded them cognitively normal, and a thorough neuropathological examination was performed. One of the control subjects had few scattered plaques in neocortex and 1 had both plaques and vascular amyloid deposits in neocortex but no tangles. This control subject was an 89-year-old female with no sign of dementia, dying suddenly of a myocardial infarction but it cannot be excluded that this subject was a case of early AD. In conclusion, none of the control subjects could be categorized as Alzheimer subjects according to Braak and Braak, but some had mild neuropathological changes as seen in normal aging (Berg et al., 1998).

2.2. Preparation 

All brains were fixed within 72 hours after death and have been fixated for at least 5 months in 4% formaldehyde buffered with phosphate. The identity of the cerebella was blinded to the investigator during the preparation and stereological quantification process.

The cerebella were separated from the brain stem by vertical incisions through the peduncles. The meninges and vessels were removed, and the cerebella weighed. Each cerebellum was cut in the midline of the vermis and according to a random systematic sampling protocol, and the right or left hemisphere was weighed and included in the study. The flocculus and nodularis were removed, painted, and processed separately. The flocculus, nodularis, vermis, and the anterior lobe were painted with distinctive colors (C.D.I.'s Tissue Marking Dye System, Cancer Diagnostics, Inc., Troy, MI, USA) in order to identify the region later on. The posterior lobe was left uncolored (Fig. 1).

  • View full-size image.
  • Fig. 1. 

    The human cerebellum divided in the vermical midline. The vermis is painted orange, the anterior lobe painted blue, and the posterior lobe left uncolored. The scale bar = 5 cm.

The selected hemisphere was embedded in 4% agar (Sigma Cas9002-18-0 A9915) and processed according to the vertical section principle (Baddeley et al., 1986). The tissue was cut into 4 mm thick slices parallel to the horizontal plane, with a random starting point according to systematic uniform random sampling (SURS). On average, 11.2 slices were obtained per specimen. The flocculus and nodularis were separately embedded in 2% agar, constituting a tube. A random starting point was marked at the end of the tube and the nodularis was rotated a random number of degrees along the vertical axis. The flocculus was rotated 90 degrees clockwise relative to the nodularis. A random starting point was chosen according to SURS and the flocculus and nodularis were cut into 2 mm thick slices. To estimate the volume, a test grid was applied to the slices, the slices were photographed, and the pictures were edited with a digital photo software (Magix Photo Clinic 4.5), using various filters to distinguish between the white- and gray matter. The slices were cut into 4 mm thick bars with a random starting point according to SURS. Every third bar was sampled with a random starting point to obtain an average of 4–6 bars (mean 5.1) in the flocculus, nodularis, vermis, and the anterior lobe, and 6–12 bars (mean 9.8) in the posterior lobe. White matter was removed from the bars before further processing. To ensure that the pieces of tissue with the greater length was represented to a greater extent and thus had a larger probability of being sampled, the posterior lobe was subsampled. The sampled bars were aligned end to end and the line was divided in 1 cm bars. Every n'th bar was subsampled to obtain 6–12 bars. In total 33–46 (mean 41.3) bars were sampled per cerebellum. Each bar was randomly rotated around the vertical axis, and a central 40 μm thick section was cut from each bar and incubated in a solution of a modified Giemsa (Merck 9204) stain, 67 mmol/L, pH 4.5 potassium dihydrogenphosphate and dehydrated in a 93%, 99%, and 99% ethanol solution. The position of the bars relative to the vermical midline was noted in order to measure the gradient of Purkinje cell density.

2.3. Neuropathology 

After sampling biopsies to be used in the stereological estimates, additional tissue blocks for making the pathological diagnosis were sampled from frontal, parietal, temporal, and occipital lobes, from insula, gyrus cinguli, hippocampus, and 1 or 2 tiers of mesencephalon. The cerebellum was not examined. The tissue was embedded in paraffin wax. Four-μm thick sections were cut from all blocks for hematoxylin and eosin stains, and for immunohistochemistry in selected areas. Eight-μm thick sections were cut from all blocks for Kluver-Barrera staining. Immunohistochemistry included beta-amyloid (DAKO M0872 1:200), tau (DAKO A0024 1:50,000), ubiquitin (DAKO Z0458 1:5000), and alpha-synuclein (Zymed zs 18-0215 1:10,000).

2.4. Stereological procedure 

2.4.1. Volume estimation 

The volume was estimated by the Cavalieri method. It states that the volume of any object is estimated from parallel sections of the object, a known distance apart, by summing up all areas associated with points of the applied test grid hitting the tissue of interest on each section and multiplying with the thickness of the sections:

where V is the volume of the object, t is the average thickness of the section, a(p) is the area associated with each point on the test grid and P is the number of points on the test grid hitting the tissue of interest (Gundersen and Jensen, 1987, Regeur and Pakkenberg, 1989). A test grid with a(p) = 16 mm2 was used for the white matter, vermis, anterior and posterior lobe (Fig. 2), and a test grid of a(p) = 4 mm2 was used for the flocculus and nodularis, in order to obtain a sufficient number of points in these relatively small structures. An average of 1385 points were counted per cerebellum (range 730–2730) and an average of 231 (range 101–4801) points were counted per region.

  • View full-size image.
  • Fig. 2. 

    The volume of the cerebellar subcompartments is estimated by randomly placing a point-counting grid on the cut surface of each slab. The area per point, a(p), is the area associated with each point on the test grid and the sum of P is the total number of points on the test grid hitting the tissue of interest. A test grid with a(p) = 16 mm2 was used for the white matter, vermis, and anterior and posterior lobe. The volume of a layer is the volume fraction of the layer multiplied by the cortical volume. The scale bar = 5 cm.

2.4.2. Number estimation 

The number of Purkinje cells and granule cells, N, was obtained by multiplying the estimate of the volume of the tissue of interest, V(ref), with the estimate of the numerical density of the cells, NV, as obtained separately by the optical disector.

where N is the total number of cells, NV is the numerical density of cells in the reference volume and V(ref) is the reference volume. The principle of the optical disector states that all 3-dimensional objects regardless of size, shape, and orientation are counted with the same probability (Gundersen, 1986). An Olympus microscope with high numerical aperture (NA = 1.40) and oil immersion objectives of 60× or 100× magnification was used. The field of vision was transmitted to a computer monitor and using software (CAST 2.0, Visiopharm, Denmark) a counting frame was applied on the monitor. The microscope stage was driven by stepping motors with steps of a known length in the x and y direction, and manually maneuvered in the z direction. Initially, the stage was positioned randomly outside the section and according to SURS, the microscope stage was driven stepwise through the slice in a meander pattern. The step length, counting frame, and height of the optical disector was adjusted in order to efficiently count the different cells. The disector used to count the Purkinje cells had a relatively large counting frame, the x–y step length was short and the disector height 20 μm with a guard zone of 15 μm. When counting the granule cells, the counting frame was relatively small, the step length short, and the disector height was 10 μm with a guard zone of 4 μm. Due to the relatively small number of Purkinje cells, 200–300 disectors had to be investigated in order to count 40 to 70 Purkinje cells, and due to the high density of granule cells only approximately 25–50 disectors had to be sampled in order to count 70–80 granule cells. The Purkinje cells were identified by their arrangement, location, and the irregular Nissl granules surrounding the large clear nucleus (Fig. 3). The deeply stained nucleolus was used as the counting item. An average of 73 (range 51–140) Purkinje cells were counted in the vermis, anterior, and posterior lobe, and an average of 58 (range 40–93) Purkinje cells were counted in the flocculus and nodularis. The granular layer was used as a reference space. The granule cells were identified by their location in the densely packed granular layer and by the large dark nucleus constituting the majority of the cytoplasm. On average 82 (range 51–182) granule cells were counted in all the regions.

  • View full-size image.
  • Fig. 3. 

    To the left an illustration of the molecular layer, the Purkinje cell layer, the granule layer, and the cerebellar white matter. The scale bar = 600 μm. In the center the granule cell layer with 2 counting frames. The scale bar = 10 μm. To the right the characteristically large Purkinje cells at 60× magnification. The Purkinje cells have a clear nucleus with a deeply stained nucleolus and irregular Nissl granules. Notice the multibranched dendrites projecting into the molecular layer. There are no qualitative histological differences between the Alzheimer's disease (AD) group and the control group. The scale bar = 10 μm.

2.4.3. Surface estimation 

In order to estimate the surface area of the cerebellum, the vertical section design was applied (Baddeley et al., 1986, Gundersen et al., 1988). It states that the surface area is obtained by estimating the surface density and multiply it by the volume of the cerebellum. The surface density is obtained by counting intersections (I) between the surface of the cerebellar cortex and the arcs of the cycloid test system and points hitting the cortical tissue.

where S is the surface of the cerebellum, 2 is a constant valid for SURS, I is the number of intersections between the outer surface and the molecular layer and V is the volume of the cortical layer. The sections were obtained from bars previously cut perpendicular to the horizontal surface and rotated around the vertical axis as required by the vertical section design. The sections were studied under a 15× magnification microscope and the vertical axis identified. The sine weighted cycloid test system was applied and the intersections and points counted. An average of 230 (range 112–302) intersections and 52 (range 29–76) points per cerebellum was obtained.

2.4.4. Volume estimation of the cortical layers 

The volume of the molecular and granular layer was estimated under a microscope with 15× magnification. A test grid was applied to the sections and the points hitting the molecular and granular layer counted in order to obtain the volume fractions of the layers. The volume of a layer is the volume fraction of the layer multiplied by the cortical volume.

where P is the number of points hitting the layer, VV is the volume fraction of the layer and Vref is the cortical volume.

2.4.5. Thickness estimation 

The thickness of the molecular and granular layer was calculated by the relationship of the volume of the respective layer and the surface area of the molecular layer.

where t is the thickness, V is the volume and S is the surface of the layer.

2.4.6. Shrinkage 

The shrinkage was measured as the ratio between the difference of volume prior and after incubation and the volume after incubation. The shrinkage was −11% with a high variation and an insignificant difference between the groups and therefore ignored. This is in accordance with prior stereological studies showing no significant shrinkage after incubation of cerebral tissue in glycolmethacrylate (Historesin) (Braendgaard et al., 1990, Dorph-Petersen et al., 2001, Pakkenberg et al., 1989, West and Gundersen, 1990).

2.4.7. Statistics 

The data were skewed and logarithmic transformation was applied to the results in order to obtain normal distributed values. Not all data could be transformed to normality, and hence both the Wilcoxon rank-sum test and the Student t test was applied in order to find any significant difference between the 2 groups, accepting a significance level of 0.05. For all data the coefficient of variation (CV) = SD/mean was calculated as estimation of the variability within the groups and shown in parenthesis throughout. Statistical analysis of the results was conducted using MyStat (Systat Software) statistical software package and Microsoft Excel 2000 on a Toshiba personal computer.

Back to Article Outline

3. Results 

The total volume of cerebellum was significantly reduced in the Alzheimer patients (100 grams, CV = 0.13) compared with controls (111 grams, CV = 0.13), p = 0.037. This was not reflected in the total number of Purkinje cells (p = 0.23) or granula cells, (p = 0.35) see Figs. 4 and 5. The overall density of Purkinje cells was (0.67 × 103 mm−3, CV = 0.19) in the AD group compared with (0.76 × 103 mm−3, CV = 0.13) in the control group which was not statistically significantly different, p = 0.11. The overall granula cell density was 2070 × 103 mm−3 (0.24) in the AD group and 1704 × 103 mm−3 (0.25) in the control group, which was not statistically significant, p = 0.13.

The total number of Purkinje and granula cells in the subdivisions of cerebellum (pars anterior, pars posterior, vermis, flocculus, and nodularis) was not statistically significant different between the 2 groups with p-values from 0.15 to 0.76. Further, there was no significant difference in the total white matter volume in the control subjects (22.5 cm3, CV = 0.14) compared with the AD group (19.8 cm3, CV = 0.21, p = 0.11). The total surface area was not different in the AD group (1088 cm2) compared with control subjects, 1027 cm2, p = 0.49.

The volume of the molecular layer in the AD group was 27.3 (0.35) cm3 for the AD group which was statistically significantly smaller than the 38.1 cm3 (0.15) for the control group, p = 0.012. No difference was found between the volume of the granula cell layer in the AD subjects (38.9 cm3, CV = 0.13), compared with control subjects (38.1 cm3, CV = 0.19), p = 0.80. There was no significant correlation between body weight and cerebellar weight in the AD group (r = 0.012) or in the control group (r = 0.89).

The correlation between increasing distance in 4-mm intervals from the midline, or medial part regarding the flocculus, and the Purkinje cell density in each of the groups was obtained by linear regression. There was a decreasing Purkinje cell density with increasing distance from the vermis midline in the AD group and the control group in the nodularis (r = −0.75 and −0.73), posterior (r = −0.041 and −0.22), and anterior lobe (r = −0.19 and −0.26). However, the opposite was the case in the AD group and control group in the vermis (r = 0.46 and 0.75). Overall, AD did not affect the Purkinje cell density medial or from the vermical midline, however, a significant difference was found at a 4 mm (p = 0.013) distance medial in the flocculus (data not shown).

Back to Article Outline

4. Discussion 

No major changes were seen in the cerebellum of severely demented AD patients, with the exception of a reduction of the total cerebellar volume. Especially the lack of difference in the total Purkinje and granula cell number was unexpected given that the patients were severely demented Alzheimer subjects who requested constant care around the clock and thus were in the end stage of the disease. However, it should be pointed out that some of our data are low of statistical power due to an unusual high biological differences found in some of the data in both groups. The negative results should therefore be interpreted in that light.

A reduction was seen in the total cerebellar volume and the cortical volume. The significant reduction in the total cerebellar volume in AD patients is in accordance with the significant 21% decrease in cerebellar volume found by Wegiel et al. (1999) in severe cases of AD. Furthermore, the reduction of the total cortical volume in the AD group is somewhat comparable to the significant decrease in the volume of the cerebellar cortex found by Wegiel et al. (1999). This consistency of difference in global volume between the groups suggests that a real atrophy is present. In contrast to Wegiel et al. (1999) who found a reduction in the volume of the granular layer in AD patients, this study found no change in the volume of the granular layer. The volume reduction is also in concordance with the known atrophy in AD in total cerebral volume (Baddeley et al., 1986) as well as atrophy in other cerebral areas, including a reduction of the neocortical volume (De la Monte, 1989, Hubbard and Anderson, 1981, Miller et al., 1980, Mouton et al., 1998, Regeur et al., 1994, Regeur, 2000). Because no stereological studies have found a significant global loss of cortical neurons in AD (Regeur et al., 1994), a possible explanation for the volume reduction could be loss of synaptic connectivity in terms of loss of synapses and dendritic spines, as shown by semiquantitative studies (DeKosky and Scheff, 1990, Terry et al., 1991, Zhan et al., 1993; and Buell and Coleman, 1979, Coleman and Flood, 1987, Flood, 1991).

In the AD group, Sjöbeck and Englund (2001) found a significantly reduced linear density of Purkinje cell per mm in the vermis. In the current study there was no change in the total Purkinje cell density but a significant reduction of Purkinje cell density in the nodularis in the AD group (data not shown), suggesting that this region may be particularly vulnerable in AD.

Wegiel et al. (1999) found a significant reduction of the total number of Purkinje cells in the AD group, whereas this study showed no change in the number of Purkinje cells in the AD group. The number of granule cells was not different in the AD patients compared with the control group which is in contrast to Wegiel et al. (1999) who found the numerical density of cerebellar granule cells to be significantly lower in the AD patient group compared with the nondemented subjects group, and a decrease in the total number of granule cells in the nondemented patients group compared with the AD patients group. There is a relatively large difference between cell numbers in the semiquantitative studies on the subject and cell numbers in the current study. However, in magnitude the cell numbers in the current study is in accordance with earlier stereologically-obtained estimates of cell numbers in the cerebellum (Andersen et al., 1992, Andersen et al., 2003, Andersen, 2004).

The general correlation between increasing distance from the midline and Purkinje cell density is interesting given the otherwise homogenous architecture of the cerebellar cortex. The overall lack of difference in density at various distances between the groups shows that no focal, within group, irreversible pathology is evident, except at a 4 millimeter distance medial in the flocculus. The interpretation of this is not clear given the normal function of the flocculus in balance and eye movement, which is not significantly dysfunctional in AD. However, subtle localized differences in density may be found when examining the various regions with shorter intervals.

The stereological methods used in this study are efficient, but more time-consuming compared with more conventional semiquantitative methods. However, despite the use of unbiased principles, subtle localized changes in neuron numbers can be still overlooked. Furthermore, counting Purkinje cells is very time-consuming due to the monolayer of Purkinje cells and the large number of disectors that has to be sampled to obtain an acceptable precision of the estimates. The high biological variability blurs the picture further. Attempts have been made to accelerate this process (Agashiwala et al., 2008), however, so far only the newly introduced stereological method, the proportionator, seems to offer a time-saving protocol (Gardi et al., 2008). The absence of existing stereological studies on the subject renders direct comparison of measurements between studies difficult.

The definitive measure of irreversible neuronal damage is loss of neurons, and the lack of difference of the numbers of granule and Purkinje cells in the 2 groups indicates that the neuronal cell number on a global scale is still intact. The results are in line with earlier studies of the central nervous system of AD patients showing no loss of neocortical neurons and glial cells and intact length of the axons of the subcortical white matter in AD patients compared with control subjects (Pelvig et al., 2003, Regeur et al., 1994).

This new stereological knowledge has contributed to the current knowledge of the role of the cerebellum in cognitive function and disorders. Furthermore, new knowledge on the normal anatomy of the cerebellum has been obtained. The current study has shown that major neuropathological change is absent in the cerebellum in AD, which is in accordance with previous studies. However, localized, significant quantitative differences were seen in AD compared with the controls. Further stereological studies must be conducted in order to examine these affected areas and the role of the cortico-cerebello-thalamic-cortical circuit and the cerebellar nuclei in AD in order to complete the picture.

Back to Article Outline

Disclosure statement 

There are no actual or potential conflicts of interest.

Appropriate Institutional Review Board approval was obtained and procedures were used concerning human subjects.

Back to Article Outline

Acknowledgements 

The authors are grateful to Susanne Sørensen for technical assistance in preparing the tissue and the staining. We thank H.J.G. Gundersen for excellent assistance in the experimental design of the study and The Danish Council for Independent Research, Medical Sciences for economical support, jr. number 271-07-0175.

The study was approved by the Danish Ethical Committee jr. number 2007-58-0015.

Back to Article Outline

References 

  1. Agashiwala RM, Louis ED, Hof PR, Perl DP. A novel approach to non-biased systematic random sampling: a stereologic estimate of Purkinje cells in the human cerebellum. Brain Res. 2008;1236:73–78
  2. Aikawa H, Suzuki K, Iwasaki Y, Iizuka R. Atypical Alzheimer's disease with spastic paresis and ataxia. Ann. Neurol. 1985;17:297–300
  3. Alafuzoff I, Arzberger T, Al-Sarraj S, Bodi I, Bogdanovic N, Braak H, et al. Staging of neurofibrillary pathology in Alzheimer's disease: a study of the BrainNet Europe Consortium. Brain Pathol. 2008;18:484–496
  4. Albus JS. A theory of cerebellar function. Math. Biosci. 1971;10:25–61
  5. Andersen BB, Korbo L, Pakkenberg B. A quantitative study of the human cerebellum with unbiased stereological techniques. J. Comp. Neurol. 1992;326:549–560
  6. Andersen BB, Gundersen HJ, Pakkenberg B. Aging of the human cerebellum: a stereological study. J. Comp. Neurol. 2003;466:356–365
  7. Andersen BB. Reduction of Purkinje cell volume in cerebellum of alcoholics. Brain Res. 2004;1007:10–18
  8. Andreasen NC, O'Leary DS, Cizadlo T, Arndt S, Rezai K, Ponto LL, et al. Schizophrenia and cognitive dysmetria: a positron-emission tomography study of dysfunctional prefrontal–thalamic–cerebellar circuitry. Proc. Natl. Acad. Sci. U. S. A. 1996;93:9985–9990
  9. Andreasen NC, O'Leary DS, Paradiso S, Cizadlo T, Arndt S, Watkins GL, et al. The cerebellum plays a role in conscious episodic memory retrieval. Hum. Brain Mapp. 1999;8:226–234
  10. Andreasen NC, Pierson R. The role of the cerebellum in schizophrenia. Biol. Psychiatry. 2008;64:81–88
  11. Azzarelli B, Muller J, Ghetti B, Dyken M, Conneally PM. Cerebellar plaques in familial Alzheimer's disease Gerstmann-Straussler–Scheinker variant?. Acta Neuropathol. 1985;65:235–246
  12. Baddeley AJ, Gundersen HJ, Cruz-Orive LM. Estimation of surface area from vertical sections. J. Microsc. 1986;142:259–276
  13. Baldaçara L, Borgio JG, de Lacerda AL, Jackowski AP. Cerebellum and psychiatric disorders. Rev. Bras. Psiquiatr. 2008;30:281–289
  14. Bastian AJ, Mink JW, Kaufman BA, Thach WT. Posterior vermal split syndrome. Ann. Neurol. 1998;44:601–610
  15. Berg L, McKeel DW, Miller JP, Storandt M, Rubin EH, Morris JC, et al. Clinicopathologic studies in cognitively healthy aging and Alzheimer's disease: relation of histologic markers to dementia severity, age, sex, and apolipoprotein E genotype. Arch Neurol. 1998;55(3):326–335
  16. Bower JM. Is the cerebellum sensory for motor's sake, or motor for sensory's sake: The view from the whiskers of a rat. Prog. Brain Res. 1997;114:463–496
  17. Braak H, Braak E. Neuropathological staging of Alzheimer-related changes. Acta Neuropathol. 1991;82:239–259
  18. Braendgaard H, Evans SM, Howard VC, Gundersen HJ. The total number of neurons in the human neocortex unbiasedly estimated using optical disectors. J. Microsc. 1990;157:285–304
  19. Buell SJ, Coleman PD. Dendritic growth in the aged human brain and failure of growth in senile dementia. Science. 1979;206:854–856
  20. Bugalho P, Correa B, Viana-Baptista M. Role of the cerebellum in cognitive and behavioural control: scientific basis and investigation models [in Portuguese]. Acta Med. Port. 2006;19:257–268
  21. Bundgaard MJ, Regeur L, Gundersen HJG, Pakkenberg B. Size of neocortical neurons in control subjects and in Alzheimer's disease. J. Anat. 2001;198:481–489
  22. Coleman PD, Flood DG. Neuron numbers and dendritic extent in normal aging and Alzheimer's disease. Neurobiol. Aging. 1987;8:521–545
  23. DeKosky ST, Scheff SW. Synapse loss in frontal cortex biopsies in Alzheimer's disease: Correlation with cognitive severity. Ann. Neurol. 1990;27:457–464
  24. De la Monte SM. Quantitation of cerebral atrophy in preclinical and end-stage Alzheimer's disease. Ann. Neurol. 1989;25:450–459
  25. Dorph-Petersen KA, Nyengaard JR, Gundersen HJ. Tissue shrinkage and unbiased stereological estimation of particle number and size. J. Microsc. 2001;204:232–246
  26. Flood DG. Region-specific stability of dendritic extent in normal human aging and regression in Alzheimer's disease. Subiculum. Brain Res. 1991;540:83–95
  27. Folstein MF, Folstein SE, McHugh PR. ‘Mini-Mental State’: a practical method for grading the cognitive state of patients for the clinician. J. Psychiatr. Res. 1975;12:189–198
  28. Gardi JE, Nyengaard JR, Gundersen HJ. Automatic sampling for unbiased and efficient stereological estimation using the proportionator in biological studies. J. Microsc. 2008;230:108–120
  29. Giedd JN, Blumenthal J, Molloy E, Castellanos FX. Brain imaging of attention deficit/hyperactivity disorder. Ann. N. Y. Acad. Sci. 2001;931:33–49
  30. Gundersen HJ. Stereology of arbitrary particles (A review of unbiased number and size estimators and the presentation of some new ones, in memory of William R. Thompson). J. Microsc. 1986;142:3–45
  31. Gundersen HJ, Jensen EB. The efficiency of systematic sampling in stereology and its prediction. J. Microsc. 1987;147:229–263
  32. Gundersen HJ, Bendtsen TF, Korbo L, Marcussen N, Mallei A, Nielsen K, et al. Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. APMIS. 1988;96:379–394
  33. Hamilton NG, Frick RB, Takahashi T, Hopping MW. Psychiatric symptoms and cerebellar pathology. Am. J. Psychiatry. 1983;140:1322–1326
  34. Hubbard BM, Anderson JM. A quantitative study of cerebral atrophy in old age and senile dementia. J. Neurol. Sci. 1981;50:135–145
  35. Ichimiya T, Okubo Y, Suhara T, Sudo Y. Reduced volume of the cerebellar vermis in neuroleptic-naive schizophrenia. Biol. Psychiatry. 2001;49:20–27
  36. Ito M. Cerebellar microcomplexes. Int. Rev. Neurobiol. 1997;41:475–487
  37. Jørgensen AM, Marner L, Pakkenberg B. No change in total length of white matter fibers in Alzheimer's disease. Neuroscience. 2008;157:878–883
  38. Konarski JZ, McIntyre RS, Grupp LA, Kennedy SH. Is the cerebellum relevant in the circuitry of neuropsychiatric disorders?. J. Psychiatry Neurosci. 2005;30:178–186
  39. Lane RD, Reiman EM, Ahern GL, Schwartz GE, Davidson RJ. Neuroanatomical correlates of happiness, sadness, and disgust. Am. J. Psychiatry. 1997;154:926–933
  40. Leiner HC, Leiner AL, Dow RS. The human cerebro-cerebellar system: Its computing, cognitive, and language skills. Behav. Brain Res. 1991;44:113–128
  41. Mackie S, Shaw P, Lenroot R, Pierson R, Greenstein DK, Nugent TF, et al. Cerebellar development and clinical outcome in attention deficit hyperactivity disorder. Am. J. Psychiatry. 2007;164:647–655
  42. Marr D. A theory of cerebellar cortex. J. Physiol. 1969;202:437–470
  43. McKhann G, Drachman D, Folstein M, Katzman R, Price DL, Stadlan EM. Clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA work group under the auspices of department of health and human services task force on Alzheimer's disease. Neurology. 1984;34:939–944
  44. Miller AK, Alston RL, Corsellis JAN. Variation with age in the volumes of grey and white matter in the cerebral hemispheres of man (Measurements with an image analyser). Neuropathol. Appl. Neurobiol. 1980;6:119–132
  45. Mohs RC, Rosen WG, Davis KL. The Alzheimer's disease assessment scale: an instrument for assessing treatment efficacy. Psychopharmacol. Bull. 1983;19:448–450
  46. Morris JC, Heyman A, Mohs RC, Hughes JP, van Belle G, Fillenbaum G, et al. The Consortium to Establish a Registry for Alzheimer's Disease (CERAD) (Part I. Clinical and neuropsychological assessment of Alzheimer's disease). Neurology. 1989;39:1159–1165
  47. Mouton PR, Martin LJ, Calhoun ME, Dal Forno G, Price DL. Cognitive decline strongly correlates with cortical atrophy in Alzheimer's dementia. Neurobiol. Aging. 1998;19:371–377
  48. Nasrallah HA, Jacoby CG, McCalley-Whitters M. Cerebellar atrophy in schizophrenia and mania. Lancet. 1981;1:1102
  49. Oster S, Christoffersen P, Gundersen HJG, Nielsen JO, Pakkenberg B, Pedersen C. Cerebral atrophy in AIDS: a stereological study. Acta Neuropathol. 1993;85:617–622
  50. Pakkenberg B, Evans SM, Møller A, Brændgaard H, Gundersen HJ. Total number of neurons in human neocortex related to age and sex estimated by way of optical disectors. Acta Stereol. 1989;8:251–256
  51. Pelvig DP, Pakkenberg H, Regeur L, Oster S, Pakkenberg B. Neocortical glial cell numbers in Alzheimer's disease. Dement. Geriatr. Cogn. Disord. 2003;16:212–219
  52. Regeur L, Pakkenberg B. Optimizing sampling designs for volume meassurements of components of human brain using a stereological method. J. Microsc. 1989;155:113–121
  53. Regeur L, Jensen GB, Pakkenberg H, Evans SM, Pakkenberg B. No global neocortical nerve cell loss in brains from patients with senile dementia of Alzheimer's type. Neurobiol. Aging. 1994;15:347–352
  54. Regeur L. Increasing loss of brain tissue with increasing dementia: a stereological study of post-mortem brains from elderly females. Eur. J. Neurol. 2000;7:47–54
  55. Reisberg B, de Ferris SH, Leon MJ, Crook T. The Global Deterioration Scale for assessment of primary degenerative dementia. Am. J. Psychiatry. 1982;139:1136–1139
  56. Sassi RB, Soares JC. [Structural magnetic resonance imaging in affective disorders]. Rev. Bras. Psiquiatr. 2001;23(Suppl 1):11–14Portuguese
  57. Schmahmann JD. An emerging concept (The cerebellar contribution to higher function). Arch Neurol. 1991;48(11):1178–1187
  58. Schmahmann JD, Sherman JC. The cerebellar cognitive affective syndrome. Brain. 1998;121(Pt 4):561–579
  59. Schmahmann JD. Disorders of the cerebellum: ataxia, dysmetria of thought, and the cerebellar cognitive affective syndrome. J. Neuropsychiatry Clin. Neurosci. 2004;16(3):367–378
  60. Schmahmann JD, Weilburg JB, Sherman JC. The neuropsychiatry of the cerebellum − insights from the clinic. Cerebellum. 2007;6:254–267
  61. Simic G, Kostovic I, Winblad B, Bogdanovic N. Volume and number of neurons of the human hippocampal formation in normal aging and Alzheimer's disease. J. Comp. Neurol. 1997;379:482–494
  62. Sjöbeck M, Englund E. Alzheimer's Disease and the Cerebellum: A Morphologic Study on Neuronal and Glial Changes. Dement. Geriatr. Cogn. Disord. 2001;12:211–218
  63. Soares JC, Mann JJ. The anatomy of mood disorders − review of structural neuroimaging studies. Biol. Psychiatry. 1997;41:86–106
  64. Stark AK, Pelvig DP, Jørgensen AM, Andersen BB, Pakkenberg B. Measuring morphological and cellular changes in Alzheimer's dementia: a review emphasizing stereology. Curr. Alzheimer Res. 2005;2:449–481
  65. Terry RD, Masliah E, Salmon DP, Butters N, DeTeresa R, Hill R, et al. Physical basis of cognitive alterations in Alzheimer's disease: synapse loss is the major correlate of cognitive impairment. Ann. Neurol. 1991;30:572–580
  66. Thach WT. On the specific role of the cerebellum in motor learning and cognition: Clues from PET activation and lesion studies in humans. Behav. Brain Sci. 1996;19:411–431
  67. Timmann D, Daum I. Cerebellar contributions to cognitive functions: A progress report after two decades of research. Cerebellum. 2007;6:159–162
  68. Tomlinson BE, Blessed G, Roth M. Observations on the brain of non-demented old people. J. Neurol. Sci. 1968;7:331–356
  69. Tomlinson BE, Blessed G, Roth M. Observation on the brain of demented old people. J. Neurol. Sci. 1970;11:205–242
  70. Wegiel J, Wisniewski HM, Dziewiatkowski J, Badmajew E, Tarnawski M, Reisberg B, et al. Cerebellar atrophy in Alzheimer's disease-clinicopathological correlations. Brain Res. 1999;818:41–50
  71. West MJ, Gundersen HJ. Unbiased stereological estimation of the number of neurons in the human hippocampus. J. Comp. Neurol. 1990;296:1–22
  72. West MJ, Coleman PD, Flood DG, Troncoso JC. Differences in the pattern of hippocampal neuronal loss in normal ageing and Alzheimer's disease. Lancet. 1994;344:769–772
  73. West MJ, Kawas CH, Martin LJ, Troncoso JC. The CA1 region of the human hippocampus is a hot spot in Alzheimer's disease. Ann. N. Y. Acad. Sci. 2000;908:255–259
  74. Yamamoto T, Hirano A. Hirano bodies in the perikaryon of the Purkinje cell in a case of Alzheimer's disease. Acta Neuropathol. 1985;67:167–169
  75. Zhan SS, Beyreuther K, Schmitt HP. Quantitative assessment of the synaptophysin immuno-reactivity of the cortical neuropil in various neurodegenerative disorders with dementia. Dementia. 1993;4:66–74

PII: S0197-4580(10)00278-2

doi:10.1016/j.neurobiolaging.2010.06.013

Neurobiology of Aging
Volume 33, Issue 1 , Pages 197.e11-197.e20, January 2012

Article Tools

* Image Usage & Resolution