Fearnley Lees BRAIN 1991
Fearnley Lees BRAIN 1991
Fearnley Lees BRAIN 1991
INTRODUCTION
The concept of neuronal loss occurring with advancing age was first proposed by Hodge (1894). For the substantia nigra this was confirmed by three morphometric studies, which found losses of 38%, 48% and 36% between the age of 20 and 90 yrs (Hirai, 1968; McGeer et al., 1977; Mann et al., 1984). This has also been verified by biochemical studies. Riederer and Wuketich (1976) found a 13% reduction of caudate dopamine per decade; Carlsson et al. (1984) observed a non-linear decline of striatal dopamine with little loss until the age of 60 yrs followed by a dramatic fall thereafter; Scherman et al. (1989) assessed striatal dopamine levels by measuring the binding of alpha-dihydrotetrabenazine, a ligand of the vesicular monoamine transporter and found a linear decline of under 10% per decade. Neuronal loss in normal ageing is not sufficient to cause PD. In PD, there is a presymptomatic phase and clinical signs do not appear until 50% of nigral neurons and 80% of striatal dopamine is lost (Marsden, 1990).
Correspondence to: Dr A. J. Lees, National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N 3BG, UK. Oxford University Press 1991
2284
The Lewy body is a hyaline neuronal inclusion body, which is always present in PD (Gibb and Lees, 1988). Forno (1969) has proposed that incidental Lewy body cases are representative of the presymptomatic phase of PD. These are individuals who die without a history of parkinsonism or dementia, but who are found to have Lewy bodies at autopsy. However, it has been suggested that Lewy bodies as an incidental finding are a feature of normal ageing without any additional nigral cell loss (Hansen et al., 1990). Similarities have been drawn between senility and PD on the grounds of shared clinical features (Teravainen and Calne, 1983; Evarts et al., 1981) and enhancement of the long latency stretch reflex (Tatton and Lee, 1975; Evarts et al., 1979). Moreover, ageing has been considered to play an important role in the pathogenesis of PD. It has been suggested that PD is a form of accelerated ageing (Mann and Yates, 1983; Barbeau, 1984) or alternatively that PD is a biphasic illness (Riederer and Wuketich, 1976; Calne et al., 1986; Mattock et al., 1988). The first phase, an acute illness early in life sharply depletes neuronal reserves, but not enough to cause clinical parkinsonism. This deficit is then unmasked by further losses due to normal ageing over the following 20 to 30 yrs. Evidence cited for this theory is the finding of subclinical nigral damage in methylphenyl-tetrahydropyridine (MPTP) toxicity (Calne et al., 1985) and the frequent lag between encephalitis lethargica and the onset of parkonsonism (Duvoisin and Yahr, 1965). The substantia nigra is an anatomically heterogeneous nucleus with regional variations in striatal projections and distribution of histochemical markers. Evidence suggests that degenerative diseases of the nigra demonstrate regional selectivity and this may be important in the pathogenesis of disease. In PD, neuronal loss is most severe in its caudal and ventrolateral parts (Hassler, 1938). In striatonigral degeneration (SND) the pattern of cell loss has been variously described as medial (Andrews et al., 1970), lateral (Jellinger and Danielczyk, 1968; Sharpe et al., 1973; Trotter, 1973; Buonanno et al., 1975; Bannister et al., 1981), central (Takei and Mirra, 1973; Sung et al., 1979), dorsolateral (Borit et al., 1975) and affecting the ventral and middle parts (Rajput et al., 1972). The nigral distribution of Steele-Richardson-Olszewski syndrome (SRO) has been described as lateral (Steele et al., 1964), ventromedial (Jellinger, 1971) and non-selective (Anzil, 1969). In ageing, the distribution of nigral cell loss has not been studied. We present the first regional semi-quantitative study to be based on division of the substantia nigra by defined cell groups. Using this technique we studied the regional selectivity of normal ageing, incidental Lewy body cases, PD, SND and SRO.
M A T E R I A L S AND METHOD Two hundred and five control cases without a history of parkinsonism or dementia were collected prospectively over a 5 yr period by several hospital pathology departments through the auspices of the UK Parkinson's Disease Society Brain Bank and for 1 yr by the Department of Pathology, Innsbruck, Austria. All cases were used to calculate the prevalence of incidental Lewy bodies, but only a proportion were suitable for morphometric study. During the same period, 50 PD cases, 22 SND cases and 6 SRO cases were also collected. The brains were divided midsagittally: one-half immersion fixed in 10% buffered formalin for a period of 5 wks and the other frozen at - 7 0 C . Multiple blocks were taken for microscopic examination. Pathological diagnosis of PD, SND and SRO was made on the distribution of cell loss and in the case of PD and SRO the presence and distribution of Lewy bodies and neurofibrillary tangles, respectively. A number of control cases had Lewy bodies in a distribution similar to PD and formed the incidental Lewy body group. The anatomy of the substantia nigra was comprehensively studied by Hassler (1937) who identified around
2285
50 neuronal groups. Hassler used the coronal plane perpendicular to the axis of Forel rather than the more contemporary transverse plane perpendicular to the axis of Meynert. For this study, nigra] anatomy was reappraised in 5 control cases by examining 20 n sections spaced every 250 /* to 500 n throughout the nigra and stained with luxol fast blue/cresyl violet and cresyl violet alone. Blocks were taken in the transverse plane, but in one whole brain, one side of the nigra was blocked in the coronal plane for comparison. Unlike Hassler who divided the nigra into separate oral and caudal parts, we found that the nigra is organized into three tiers: pars reticulata, ventral pars compacta and dorsal pars compacta, see Fig. 1. The tiers are staggered and the ventral tier is mainly confined to the caudal nigra arising orally within the pars reticulata. Within the two tiers of the pars compacta we were able to confirm Hassler's groups. However, we found that the oral and caudal groups were continuous forming parallel columns along the length of the tier, see Fig. 2. As a result of this parallel arrangement there was little difference in the cross-sectional configuration of the neuronal groups between the coronal and transverse planes, which are angled by 30 - 40. However, the levels of sampling within the nigra were different. In the transverse plane compared with the coronal plane, the medial groups were sampled at a higher level and the lateral and dorsal groups at a lower level.
Oral
Dorsal
An alternative system of classification of the pars compacta groups is proposed taking into account the concept of tiering and Hassler's original work. Each group is abbreviated to two letters corresponding to the tier and division and one letter in brackets corresponding to the group. The pars compacta is divided into two tiers: ventral v, and dorsald. The dorsal tier has three divisions: medial-dm, lateral-dl and pars lateralispi. The ventral tier has two divisions: medial-vm, and lateral-vl. Each division has 1 to 4 groups: medial(m), intermediate-(i), lateral (1) and eutopic(e). This classification is compared with Hassler's in Table 1. The group vl(e) corresponds to Hassler's eutopic group eJ and has been included with vl for the sake of simplicity. A number of Hassler's nigral groups correspond to the ventral tegmental area and the retrorubral area and have not been included. A single 7 \i section of the caudal nigra stained with haematoxylin and eosin was used for morphometry. At this level, the Ulrd nerve fascicles demarcate the ventral tegmental area from the nigra (Olszewski and Baxter, 1954). Criteria for selecting a consistent level were defined on the basis of neuronal groups and their complementary nerve fibre fields as described by Hassler and are as follows: medially, where the Ulrd nerve fascicles divide the nigra from the ventral tegmental area above the decussation of the superior cerebellar peduncle between the main bulk of the Hassler's oral nerve fibre field rab, and the caudal field v; centrally, at the level of Hassler's neuronal group hj and its distinctive nerve fibre field; laterally, at the point where the striatonigral pathway (Hassler's nerve fibre field K) is well formed and oral enough to be distant from the medial lemniscus and caudal enough to be below neuronal group vl(e). In the pathological cases, it was still possible to recognize neuronal groups by the remaining nerve cells,
2286
dm
FIG. 2. Substantia nigra groups. The substantia nigra in the transverse plane perpendicular to the axis of Meynert: A, oral nigra; B, mid nigra; c, caudal nigra. Medial nigra left of page and lateral nigra right of page. Tiers coded according to Fig. 1. Bold letters correspond to the tier and the division and regular letters to group, see text for explanation, mrd nerve fascicles. RN = red nucleus. *Putaminonigral pathway (Hassler's nerve fibre field K).
TABLE 1. COMPARISON PRELIMINARY ANATOMICAL STUDY WITH THE WORK OF HASSLER (1937) Feamley and Lees Hassler's groups Tier Division Group Caudal Oral v vm (m) Spvm Spvi (0 Spvl (1)
vl (m) ()
0) (e) d dm (m)
0)
dl
(m) (i) 0)
pl
Nomenclature (Fearnley and Lees): v = ventral, d = dorsal, m = medial, i = intermediate, 1 = lateral, e = eutopic. * Order of appearance from oral (1) to caudal (3).
nerve fibre fields, and condensation of glia. For example, even when vl has no pigmented neurons its outline is recognizable by a strip of closely packed astrocytes. As there was no landmark laterally when cutting the midbrain many of the blocks were tilted and unsatisfactory for morphometry. The proportion of cases that were suitable was: 36/192 controls without LBs, 6/13 controls with LBs (incidental LB cases), 20/50 PD cases, 15/22 SND cases and 4/6 SRO cases.
2287
Discrepancies between these proportions arose from chance and the fact that the brains had been cut by five different persons. Archival material at the Institute of Psychiatry, London, yielded a further 10 SRO cases and 1 incidental Lewy body case suitable for morphometry. The substantia nigra of each case was divided into 6 morphometric regions (Fig. 3), by marking the coverslip of the slide with a 0.13 mm Rotring pen under 40 and 100 x magnification. The morphometric regions (capital letters) differed from the anatomical groups in two instances. vl(m) was divided off from
Downloaded from brain.oxfordjournals.org at Leibniz-Institut fuer Altersforschung - Fritz-Lipmann-Institut on April 7, 2011
FIG. 3. Morphometric regions. Taken at the level of the caudal nigra, Fig. 2c. Dorsal tier: medial part (DM), lateral part (DL) and pars lateralis (PL). Ventral tier: medial part (VM) and lateral part (Vl-intermediate and VL).
vl and counted separately as VI. The purpose in selecting this area for particular attention was to test further whether pathological processes respected the boundaries of the major nigral divisions. dl(l) was included in DM for ease of marking. Slides were coded so that the morphometrist was unaware of the patient's age. Pigmented neurons in each region were counted twice using alternately the cell body and nucleolus as the counting unit for the purposes of comparison and validation. The nucleolus has the advantages of good contrast with the rest of the cell, small size and uniform shape. However, a potential disadvantage in the substantia nigra is obscuration by pigment. The total count (number of pigmented neurons in a single section of the caudal nigra) was calculated by summating the regional counts. The slide was then remarked and recounted. Counting was performed under 400 x magnification using an eyepiece graticule and parallel sweeps of the microscope stage. No corrections were made for split cell error. The following areas were not counted: pars reticulata; ventral tegmental area; retrorubral area and nucleus peripeduncularis. Regression equations were calculated for the control regional and total counts vs age. The pathological cases including the incidental Lewy body cases varied in age. In order to assess the neuronal loss due to the pathological process separate from losses due to ageing it was necessary to correct the regional and total counts for age. For each of the pathological cases predicted control regional and total counts were calculated from the corresponding regression equations (Table 4). The pathological count was then expressed as a percentage of the predicted control count. Alternatively, cell loss was estimated by expressing the pathological count as a percentage of an age-matched control count. Significance was tested using Student's t test. Figures preceded by represent one standard deviation.
RESULTS
The results of the nucleolus and cell body counts were comparable except that the cell body counts reached a greater level of statistical significance due to the larger numbers counted and only data from cell body counting will be presented. Obscuration of the nucleolus by pigment was not a problem at a section thickness of 7 n and using a magnification of 400 x . Variation between consecutive counts (including re-marking of the slide) was < 5 %. The average neuronal loss in the incidental Lewy body and parkinsonian groups was estimated by using age-adjusted counts expressed as a percentage of the predicted control and alternatively by comparing uncorrected counts directly with age-matched controls. Both methods gave similar results with no significant difference. No difference was found between VI and VL in normal ageing or disease. Uncorrected and age-adjusted cell body counts are listed in Tables 2 and 3, respectively.
2288
(A) Control 21M 22F 29M 31M 44M 47M 53F 54M 55M 56F 58M 58M 60F 61M 61F 65F 65M 69F 70F 70M 71M 75F 75M 77M 78M 79M 80F 81F 81M 84F 85F 86M 87M 89M 91F 91M
206 184 192 1 7 1 146 176 136 184 188 172 140 155 152 150 177 88 1 3 1 188 72 152 158 124 109 85 104 132 148 139 116 193 168 120 114 139 122 90
45 54 47 46 44 44 43 45 41 47 37 37 42 40 46 34 22 51 20 48 41 39 46 40 36 49 40 42 33 40 40 37 46 35 41 42
200 216 198 178 206 200 188 1 8 1 178 188 184 167 215 177 173 147 201 187 146 200 190 170 182 160 183 180 179 152 175 190 199 176 206 192 159 167
74 93 72 83 64 48 60 87 84 75 71 60 61 69 50 32 50 68 32 71 75 50 45 30 37 58 60 68 39 61 64 45 42 50 32 30
76 108 82 88 58 50 94 84 90 88 69 58 76 65 58 53 50 96 72 44 48 37 57 29 58 44 63 83 55 62 62 39 48 59 39 35
692 692 695 657 633 583 613 692 653 658 588 544 642 587 585 403 518 702 406 615 558 493 504 390 503 520 556 543 448 648 616 471 540 578 426 394
(B) Incidental Lewy bodies 65F 152 75M 156 78M 118 80F 132 80M 114 81F 143 82F 127 (C) Parkinson 's disease 68M \Vi 62F 2 74M 6 81M 6 73M 10 73F 73M 1 1 83F 12 133 67 36 123 70 132 8 21
18 22 1 1 20 17 24 17 24 5 8 17 16 5 2 2
99 113 76 94 63 72 81 78 30 22 49 23 13 1 1 3
55 55 50 51 41 53 50 67 34 39 43 38 54 20 29
55 48 48 47 44 51 41
443 459 359 401 308 404 364 403 234 192 295 212 291 97 92
56 32 38 32 3 1 34 23 23
45 66 49 31 34 53 33 14
AGEING AND PARKINSON'S DISEASE Age/sex 76F 80M 78M 80M 78M 71F 71M 87F 82F 64F 74M 77M Dur. VM VI 0 0 0 0 0 1 0 0 0 0 0 2 0 5 1 0 0 3 17 0 4 0 0 5 2 0 0 VL 0 10 0 8 2 13 4 0 0 0 0 7 0 11 2 0 3 4 14 2 1 1 11 3 4 0 1 DM 22 14 27 34 16 18 18 31 12 18 15 25 10 25 20 4 27 33 39 4 26 16 14 17 31 5 7 DL 16 12 41 30 13 19 22 17 11 9 15 17 9 15 12 2 27 11 19 5 5 20 34 10 29 4 5 PL 24 27 26 35 4 7 25 19 12 8 15 18 3 16 15 2 8 24 18 8 4 13 30 13 34 3 4 Total 78 83 124 140 68 66 78 70 37 47 57 75 36 138 76 18 107 119 189 32 88 69 97 105 174 26 20
2289
13 16 15 19 20 30 33 15 33 17 8 18 9 25 3 2 29 12 38 12 6 Striatonigral degeneration 68M 14 2 67F 3 66 51M 4 26 56F 10 5 61M 42 5 66M 44 5 72F 5 82 43M 13 6 62M 48 6 74F 6 19 65M 7 8 64F 57 8 78M 74 8 59F 14 10 61F 10 3
Steele-Richardson-Olszewski syndrome 70M 2 32 8 46 59F 4 34 66 9 92 59M 4 14 8 68M 4 32 10 39 79F 22 4 15 6 66F 8 2 20 5 75M 31 5 20 5 79F 2 13 5 3 55F 3 2 15 6 67F 10 32 6 18 71M 4 37 6 10 72F 11 10 63 8 66F 11 22 9 28 4 17 68F _ 9
15 11 10 11 4 9 11 3 6 7 10 7 12 15
14 24 12 15 7 8 3 5 8 6 14 12 8 9
37 34 16 20 25 2 26 5 15 29 25 32 4 28
Control cases and ageing The 36 control brains suitable for morphometry ranged in age from 21 to 91 yrs. Regression equations were calculated for the regional and total neuronal counts vs age, see Table 4. There was a linear decline with advancing age in both the total count of the caudal nigra (Fig. 4) and the regional counts. The regression equations were then used to calculate the neuronal loss between the age of 20 and 90 yrs for the caudal nigra and each region. The total count fell 33% (4.7% per decade). Regionally there was sparing of VI/VL with 2.5 to 3 times greater loss in the other regions (Fig. 5B). The
2290
(B) ParkinsonI'S disease Mild cases 68M 1V4 62F 2 74M 6 81M 6 73M 10 73F Moderate cases 73M 11 83F 12 76F 13 80M 15 78M 20 80M Severe cases 78M 15 71F 17 71M 18 87F 25 82F 29 64F 38 74M 77M (C) Striatonigral 68M 67F 51M 56F 61M 66M 72F 43M 62M 74F 65M 64F 78M 59F 61F
93 45 26 95 51 % 6 17 24 15 23 25 25 6 6 2 2 8 9 5
59 12 20 43 40 12 5 5 0 0 0 0 0 2 0 0 0 0 0 5
42 16 12 28 13 7 6 2 0 6 0 5 1 7 2 0 0 0 0 4
118 57 73 87 71 100 37 60 40 28 53 68 31 33
31
92 49 67 61 54 59 40 45 29 23 75 56 24 32 37 35 21 14 26 31
60 83 69 47 48 74 46 22 35 41 38 53 6 10 34
31
72 40 36 57 39 54 18 18 15 16 24 27 13 12 14 14 7 8 11 14
68 25 31 28 46
19 10 21 26
degeneration 10 2 46 3 16 4 6 5 28 5 30 5 61 5 8 6 32 6 14 6 7 6 39 8 56 8 9 10 2 10
0 12 2 0 0 7 42 0 10 0 0 12 5 0 0
0 6 1 0 2 2 8
1
0 1 6 2 2 0 0
18 44 30 6 44 57 70 6 43 30 24 29 60 8 12
15 23 17 3 41 18 32 7 8 35 54 16 53 6 8
4 20 17 2 10
31
25 9 5 18 39 17 50 4 5
6 25 12 3 18 21 35 5 15 13 17 18 33 4 4
27 18
25 36
50 42
27 30
2291
59M 68M 79F 66F 75M 79F 55F 67F 71M 72F 66F 68F
4 4 4 5 5 5 6 6 6 8 9 _
9 23 1 1 6 23 2 2 12 7 8 19 6
19 25 15 5 1 3 5 5 25 10 25 27 10
49 21 12 1 1 1 1 7 8 17 20 35 12 9
16 20 8 16 21 6 9 12 18 17 21 26
18 25 13 13 5 9 12 10 24 21 13 15
20 27 38 3 37 7 18 38 34 45 5 37
26 23 1 5 9 18 6 8 18 18 24 15 15
800 700
^ " ^ ^ ^
600
^ * *
500
r =-0.668
/MD.001
80
range of counts within each decade increased with consecutive decades suggesting variability in the effects of ageing. Qualitatively, DM, DL, PL and less frequently VM were the most heavily pigmented regions within the nigra. Incidental Lewy body cases Incidental Lewy bodies were found in 13 of the 205 control cases. In no case was there a history of parkinsonism or dementia. The incidental cases ranged in age from 46 to 90 yrs with a mean age of 76 12. The age specific prevalence increased with age to 12% over 80 yrs (Table 5). There was a significant reduction in the total count {P < 0.01) of the 7 incidental cases suitable for morphometry compared with age-matched controls. Regionally, there was only significant loss in VI/VL (P < 0.001). The average reduction in the total count was 2717% and in VI/VL 529% (Fig. 5c).
2292
Attrition** between 20 and 90 yrs 38% 18% 14% 48% 48% 46% 33%
Egression equation ^coml = b~m"e> b m 214.0 49.0 206.4 95.6 104.6 125.5 794.8 1.050 0.122 0.356 0.572 0.632 0.735 3.462
Correlation r P< Downloaded from brain.oxfordjournals.org at Leibniz-Institut fuer Altersforschung - Fritz-Lipmann-Institut on April 7, 2011 -0.580 -0.339 -0.387 -0.630 -0.619 -0.552 -0.668 0.001 0.05 0.02 0.001 0.001 0.001 0.001
*See Fig. 3 for the location of the region within caudal nigra. **Calculated from regression equation: attrition = (county - counts/county 100% = (70-m)/(b - 20-m) 100% b = intercept; m = slope; r = correlation coefficient. ***Caudal substantia nigra.
Parkinson's disease cases There was a significant reduction in both the total and regional counts of the PD cases compared with controls (P < 0.001). The average percent losses were maximal in VI/VL followed by VM and then the dorsal regions DM, DL and PL, see Fig. 5D. The 20 PD cases ranged in age from 62 to 87 yrs (mean 75 6). In 16 cases the onset of symptoms was well documented with an average age of 60 11 yrs and an average duration of 15 10 yrs. There was no correlation between age and symptom duration or between age and neuronal count. However, there was good correlation between symptom duration and neuronal loss (P < 0.001). There was an exponential decline in the total and regional age-adjusted counts with increasing symptom duration (Figs. 6, 7), so that losses were greatest in the beginning of the disease and then tailed off as it progressed. In the first decade from the onset of neuronal loss, there was a 45% fall in the age-adjusted total count, ten times greater than the rate of loss that could be accounted for by ageing. Similar results were found using uncorrected counts. Presymptomatic phase of Parkinson's disease The presymptomatic phase was taken as the period from the onset of neuronal loss to the onset of symptoms. It was found to be 4.7 yrs using the equation from Fig. 6 for the total age-adjusted count in PD. During this period neuronal loss due to PD, excluding age-related loss, was 31 % for the total count and 64% for the regional VL count. The loss due to ageing was calculated from the control regression equations of Table 4 over the interval from both birth to onset of neuronal loss (mean age 55 yrs) and birth to the onset of symptoms (mean age 60 yrs). The average loss due to ageing was 24 26% for the total count and 10 11 % for the regional VL count. The average combined loss at the onset of symptoms was 48 % for the total count and 68 % for the VL regional count. To test whether any of the incidental Lewy body cases might have had undetected
A G E I N G AND P A R K I N S O N ' S D I S E A S E
2293
FIG. 5. Substantia nigra regional losses in ageing and disease, A, normal anatomy; B, ageing*; c, incidental Lewy body cases**; D, Parkinson's disease**; E, striatonigral degeneration**; F, Steele-Richardson-Olszewski syndrome**. A continuous border between two regions indicates a significant difference in the percent neuronal loss. *Percentage loss between the age of 20 and 90 yrs. "Average percentage loss in cases studied.
-i 7 J. I
parkinsonism during life individual total counts were scrutinized. From each the symptom duration was calculated using the equation of Fig. 6. Using this analysis, 1 case might well have been symptomatic for a duration of 2 yrs. Another case may have been
2294
FIG. 6. Parkinson's disease. Total age adjusted count vs symptom duration. *****Presymptomatic phase: duration 4.7 yrs.
FIG. 7. Parkinson's disease. Logarithm of regional age adjusted counts vs symptom duration. Region VM VL DM DL PL Log Log Log Log Log Regression equation (count) 1.78-0.0393 duration (count) = 1.57-0.0685 duration (count) = 1.87-0.0128 duration (count) = 1.84-0.0169 duration (count) = 1.80-0.0211 duration r 0.740 0.787 0.618 0.742 0.637 < < < < < 0.01 0.001 0.02 0.001 0.01
marginally symptomatic with a symptom duration of 1 mth. Both these cases would have been aged 78 yrs at the onset of symptoms. Topographical progression of Parkinson's disease The topographical progression in the nigra of PD was studied by including the incidental Lewy body cases, as representative of the presymptomatic phase, and dividing the PD
2295
+ + + severe
> 35% < 35% < 15% P < 0.001 P < 0.02
Stage*,** 3.4 4.4 4.8 Downloaded from brain.oxfordjournals.org at Leibniz-Institut fuer Altersforschung - Fritz-Lipmann-Institut on April 7, 2011 P < 0.05*** n.s.
Hoehn and Yahr (1967). ** Where available. Wilcoxon rank sum test.
cases into mild, moderate and severe groups depending on the total count (see Table 6). Fig. 8 illustrates the staggering of regional neuronal loss, starting in VI/VL and then spreading to the other regions and in particular VM. Striatonigral degeneration cases The 15 SND cases ranged in age from 43 to 78 yrs (mean 63 9) and the average symptom duration was 6 2 yrs. Compared with age-matched controls there was a significant reduction in both the total and regional counts (P < 0.001). The average percent neuronal loss is shown in Fig. 5E. There was no significant difference between the average total count of the SND cases and the PD cases. However, loss in the dorsal tier (DM, DL and PL combined) was greater than PD by 21% (P < 0.01). When compared with the moderate/severe cases of PD (average total loss 85%), the SND cases had a smaller loss in VM of around 12% (P < 0.05). Steele-Richardson-Olszewski syndrome The 14 SRO cases ranged in age from 55 to 79 yrs (mean 68 7) and the average symptom duration was 5 2 yrs. Compared with age-matched controls there was a significant reduction (P < 0.001) in both the total and regional counts. The average
Normal
FIG. 8. Topographical progression of Parkinson's disease. Horizontal axis: phase of disease from no neuronal loss (normal) through presymptomatic disease (I: incidental Lewy body cases) to mild (+), moderate ( + + ) and severe ( + + +) symptomatic Parkinson's disease. Vertical axis: average age adjusted regional counts for each group.
2296
percent loss is shown in Fig. 5F. The difference between regions is not as distinct as occurs in PD and SND. There was no significant difference between the divisions of the ventral tier, VM, VI and VL. VM sustained greater losses than the dorsal tier divisions DM (P < 0.05), DL and PL (P < 0.001). There was no difference between the middle groups VI, VL, DM and DL. PL was relatively spared compared with all the other regions (P < 0.02). Compared with PD, losses in the dorsal tier were greater by 30% (P < 0.001). Losses in the lateral ventral tier (VI and VL combined) were smaller than in PD by 10% (/> < 0.01) and smaller than moderate/severe PD by 17% (P < 0.001).
DISCUSSION
The anatomy of the substantia nigra may be crucial to the understanding of its pathology. The pars compacta is divided into a ventral and dorsal tier. Gerfen et al. (1987) found that in the rat, the ventral tier projects to the striatal striosomes and the dorsal tier to the matrix. The striosomes project back to the cell bodies and proximal dendrites of the ventral tier and the matrix projects back to the pars reticulata and distal dendrites of the ventral tier. The striatal striosomes and matrix are histochemically distinct and have separate dendritic fields. The striosomal projecting nigral neurons of the cat have a high concentration of sigma receptors (Graybiel et al., 1989) and the dorsal tier is marked by the presence of calcium binding protein (Gerfen et al., 1985). Also there are developmental differences, the ventral tier forms connections with the striatum before the dorsal tier (Olson et al., 1972). In PD, the ventral tier and especially its lateral part bears the brunt of neuronal loss, which suggests that the cellular characteristics or connections of the ventral tier may be important in providing clues to the pathogenesis of PD. Each tier of the pars compacta has several neuronal groups. In the monkey, Parent et al. (1983) demonstrated alternating clusters of putamen and caudate projecting neurons forming columns through the nigra. It is possible that the anatomical groupings seen in the human correspond to these clusters and that there may be associated functional divisions. Topographical studies of the nigrostriatal projection in the monkey have established that the oral nigra projects mainly to the head of the caudate and the caudal nigra projects to the putamen and the body of the caudate (Carpenter and Peter, 1972; Szabo, 1980). The lateral ventral nigra projects to the dorsal putamen. In PD, neuronal loss begins in the lateral ventral tier and throughout the illness this remains the region most severely affected. Kish et al. (1988) in a post-mortem study of PD found severe dopamine depletion in the putamen with relative sparing of the caudate. Within the putamen, the dorsal and intermediate zones were more depleted than the ventral, but severe dopamine reduction occurred in all regions. The correlation between the regional selectivity of PD and its clinical expression is uncertain. The putamen is considered to be predominantly involved in motor programming and the caudate in cognition (DeLong et al., 1983). The motor cortex input to the putamen is somatotopically organized. The leg area of the motor cortex projects to the dorsal zone of the putamen in non-human primates, the arm area to the intermediate zone and the head area to the ventral zone (Kiinzle, 1975). In PD, cell loss starting in the lateral
2297
ventral nigra suggests that dopamine depletion starts in the dorsal putamen. If striatal function mirrors the motor cortex input then symptoms would be expected to start in the leg. However, it is more common for symptoms to start in the arm (Schelosky and Poewe, 1990). It can be argued that a small deficit of hand function is soon noted by the patient, but the same deficit in the leg goes unnoticed. Furthermore, the bipedal gait of the human may explain the difference. In contrast, dystonia in young onset PD usually begins in the leg and often does not involve the arm. In the past ageing has been considered to play an important role in the pathogenesis of PD, because of the considerable neuronal loss with advancing age in the substantia nigra. This study confirmed a linear fallout of pigmented neurons at a rate of 4.7% per decade in the caudal pars compacta. In the biphasic theory of ageing and PD, the rate of neuronal loss in the second phase including the period of symptomatic PD is equivalent to the rate of neuronal loss seen in normal ageing. However, the rate of loss in cases of established PD is not linear but exponential and in the first decade there is a 45% loss, 10 times greater than is seen in normal ageing. McGeer et al. (1988) have also found that the rate of neuronal loss in PD is greater than could be accounted for by ageing by estimating the numbers of degenerating neurons at the time of death. The alternative notion of PD being a form of accelerated ageing also seems improbable. Neuronal loss due to ageing shows marked regional variation in the nigra. Losses are greatest in the dorsal tier (6.9% per decade) followed by the medial ventral tier (5.4%) and least in the lateral ventral tier (2.1%). If PD is due to an acceleration of normal ageing then the pattern of cell loss should be similar to that seen as a result of normal ageing. In fact, the complete opposite applies. Neuronal loss in PD is greatest in the lateral ventral tier (average loss 91 % excluding ageing) followed by the medial ventral tier (71 %) and the dorsal tier (56%). These findings are inconsistent with the accelerated ageing theory. It seems likely that PD is a relatively acute monophasic illness. The exponential loss of neurons and greater losses early in the disease means that the presymptomatic phase is considerably shorter than previously suspected, lasting around 5 yrs. At the onset of symptoms, striatal dopamine is reduced by 80%, but the total caudal nigral count is only reduced by about 50% (lateral ventral tier by 70%). This suggests that the remaining neurons are dysfunctioning and Hirsch et al. (1988) have found that 17% of pigmented nigral neurons in PD failed to stain for tyrosine hydroxylase. It has been proposed that the presymptomatic phase of PD is represented by incidental Lewy body cases, which exhibit neuronal loss intermediate between control and PD cases (Forno, 1969). Hansen et al. (1990) have recently challenged this concept by suggesting that Lewy bodies may be a feature of normal ageing. They found in a study of 13 patients with dementia and Lewy bodies at autopsy that 4 had no cell loss in the nigra. However, this report was not supported by any quantitative evidence and no regional morphometry was carried out. In this study of 7 incidental cases, there was an average 27% loss of nigral neurons in excess of ageing (P < 0.01). Regionally, this loss was confined to the lateral ventral tier with an average loss of 52% (P < 0.001) similar to the pattern of cell loss seen in PD. This suggests that incidental Lewy bodies are indicative of presymptomatic PD rather than a feature of normal ageing. Furthermore in PD, neuronal loss appears to be confined to the lateral ventral tier early in the disease,
2298
but by the time symptoms appear it has spread to the other regions and in particular the medial ventral tier. There is a discrepancy between the prevalence of incidental Lewy body cases and the prevalence of clinical PD in the community. This could be explained by a long presymptomatic phase with the majority of incidental cases dying before the onset of parkinsonism (Gibb and Lees, 1988). However, if there is indeed as our results suggest a relatively short presymptomatic phase this explanation is untenable. It might be that PD is considerably under-diagnosed in the elderly. The incidence of new PD cases has been reported to fall after the age of 75 (Schoenberg, 1986). Two of the 7 incidental Lewy body cases in this study might have had undiagnosed PD with an onset of parkinsonism at the age of 78 yrs. Therefore, the difference between the prevalence of PD and incidental cases might be spurious, because of an underestimation of PD in the elderly and a consequent overestimation of the number of true incidental Lewy body cases. PD regional nigral selectivity is relatively specific when compared with SND and SRO. Both PD and SND share a predilection for the lateral ventral tier, which is not seen in SRO. Differences were seen in the involvement of the dorsal tier, which was greatest in SRO followed by SND and least in PD. Neuronal loss in the dorsal tier was greater in SND and SRO than in PD by approximately 21% and 30%, respectively. In the monkey, Parent et al. (1983) found that putaminal projecting clusters were more frequent in the caudal ventral nigra and that caudate clusters were more frequent in the oral dorsal nigra probably corresponding to the ventral and dorsal tiers, respectively. This would suggest that greater involvement of the dorsal tier would result in greater reduction of nigral terminals in the caudate. This would explain the differences in I8 Fdopa striatal uptake between PD, SND and SRO on positron emission tomography (Brooks et al., 1990). Reduction of uptake in the caudate was greatest in SRO followed by SND and least in PD. Since the advent of MPTP induced parkinsonism it has been mooted that PD is due to one or several environmental neurotoxins (Davis et al., 1978). However, epidemiological studies have failed to unearth any single conclusive exogenous factor although some association with exposure to herbicides and residence in rural areas has been implicated (Barbeau et al., 1986). Furthermore, despite close similarity between the MPTP lesion and that seen in PD, differences in the pathological lesion exist. Although the nigral lesion has a ventral predilection (German et al., 1988), the caudate and putamen are equally depleted of dopamine, unlike PD in which there is some sparing of the caudate (Pifl et al., 1988). There is also some question that MPTP in the human spares the nucleus basalis of Meynert and that Lewy bodies do not occur (Burns et al., 1984). However, it is possible that some of these histological differences are due to the acute administration of the toxin. Neuromelanin has been implicated in the pathogenesis of PD (Hirsch et al., 1988), but non-pigmented regions can be equally affected (Marsden, 1983). It has been speculated that because MPP + has a high affinity for neuromelanin it may be slowly released after the initial exposure and result in chronic toxicity (D'Amato et al., 1986). However, we found that the neuromelanin-containing cells with the least pigment to be the most affected in PD suggesting that neuromelanin accumulation is not a pathogenetic factor (Gibb et al., 1990). However, neuromelanin may be important in the pathogenesis of
2299
age-related neuronal attrition as ageing seems to be confined to the most heavily pigmented regions. It is known that the formation of neuromelanin is associated with the generation of free radicals and toxic quinones (Graham, 1978), which may be responsible for agerelated cell death.
ACKNOWLEDGEMENTS
Downloaded from brain.oxfordjournals.org at Leibniz-Institut fuer Altersforschung - Fritz-Lipmann-Institut on April 7, 2011
We wish to thank the following for providing incidental Lewy body cases: Dr Poewe and Professor Mickucz, University Departments of Neurology and Pathology, Innsbruck (2 cases) and Professor Lantos, MRC Alzheimer's disease Brain Bank, Institute of Psychiatry, London (1 case). REFERENCES ANDREWS JM, TERRY RD, SPATARO J (1970) Striatonigral degeneration. Clinical-pathological correlations and response to stereotaxic surgery. Archives of Neurology, 23, 319-329. ANZIL AP (1969) Progressive supranuclear palsy: case report with pathological findings. Ada Neuropathologica, Berlin, 14, 7216.
BANNISTER R, GIBSON W, MICHAELS L, OPPENHEIMER DR (1981) Laryngcal abductor paralysis in multiple
system atrophy: a report on three necropsied cases, with observations on the laryngeal muscles and the nuclei ambigui. Brain, 104, 351368. BARBEAU A (1984) Etiology of Parkinson's disease: a research strategy. Canadian Journal of Neurological Sciences, 11, 2 4 - 2 8 .
BARBEAU A, ROY M, CLOUTIER T, PLASSE L, PARIS S (1986) Environmental and genetic factors in the
etiology of Parkinson's disease. Advances in Neurology, 45, 299-306. BORTT A, RUBINSTEIN LJ, URICH H (1975) The striatonigral degenerations: putaminal pigments and nosology. Brain, 98, 101-112.
BROOKS DJ, IBANEZ V, SAWLE GV, QUINN N, LEES AJ, MATHIAS CJ et al. (1990) Differing patterns
of striatal 18F-dopa uptake in Parkinson's disease, multiple system atrophy, and progressive supranuclear palsy. Annals of Neurology, 28, 547555.
BUONANNO F, NARDELU E, ONNIS L, RIZZUTO N (1975) Striato-nigral degeneration: report of a case
with an unusually short course and multiple system degenerations. Journal of the Neurological Sciences, 26, 545-553.
BURNS RS, MARKEY SP, PHILLIPS JM, CHIUEH CC (1984) The neurotoxicity of l-methyl-4-phenyl-
1,2,3,6-tetrahydropyridine in the monkey and man. Canadian Journal of Neurological Sciences, 11, 166-168.
CALNE DB, LANGSTON JW, MARTIN WRW, STOESSL AJ, RUTH TJ, ADAM MJ et al. (1985) Positron
emission tomography after MPTP: observations relating to the cause of Parkinson's disease. Nature, London, 317, 246-248.
CALNE DB, EISEN A, MCGEER E, SPENCER P (1986) Alzheimer's disease, Parkinson's disease, and
motoneurone disease: abiotropic interaction between ageing and environment? Lancet, ii, 10671070. CARLSSON A, NYBERG P, WINBALD B (1984) The influence of age and other factors on concentrations of monoamines in the human brain. In: Brain Monoamines in Normal Aging and Dementia. Edited by P. Nyberg. Sweden: Umea Medical Dissertations, pp. 5 3 - 8 4 . CARPENTER MB, PETER P (1972) Nigrostriatal and nigrothalamic fibers in the rhesus monkey. Journal of Comparative Neurology, 144, 93 115. D'AMATO RJ, LIPMAN ZP, SNYDER SH (1986) Selectivity of the parkinsonian neurotoxin MPTP: toxic metabolite MPP + binds to neuromelanin. Science, 231, 987-989.
DAVIS GC, WILLIAMS AC, MARKEY SP, EBERT MH, CAINE ED, REICHERT CM et al. (1979) Chronic
parkinsonism secondary to intravenous injection of meperidine analogues. Psychiatry Research, 1, 249-254. DELONG MR, GEORGOPOULOS AP, CRUTCHER MD (1983) Cortico-basal ganglia relations and coding of motor performance. Experimental Brain Research, Supplement 7, 3040. DUVOISIN RC, YAHR MD (1965) Encephalitis and parkinsonism. Archives of Neurology, Chicago, 12, 227-239.
2300
EVARTS EV, TERAVAINEN HT, BEUCHERT DE, CALNE DB (1979) Pathophysiology of motor performance
in Parkinson's disease. In: Dopaminergic Ergot Derivatives and Motor Function. Edited by K. Fuxe and D. B. Calne. Oxford: Pergamon Press, pp. 4 5 - 5 9 . EVARTS EV, TERAVAINEN H, CALNE DB (1981) Reaction time in Parkinson's disease. Brain, 104, 167-186. ORNO LS (1969) Concentric hyalin intraneuronal inclusions of Lewy type in the brains of elderly persons (50 incidental cases): relationship to parkinsonism. Journal of the American Geriatrics Society, 17, 557-575. GERFEN CR, BAIMBRJDGE KG, MILLER JJ (1985) The neostriatal mosaic: compartmental distribution of calcium-binding protein and parvalbumin in the basal ganglia of the rat and monkey. Proceedings of the National Academy of Sciences of the USA, 82, 8780-8784. GERFEN CR, HERKENHAM M, THIBAULT J (1987) The neostriatal mosaic: n . Patch- and matrix-directed mesostriatal dopaminergic and non-dopaminergic systems. Journal of Neuroscience, 7, 39153934.
GERMAN DC, DUBACH M, ASKARJ S, SPECIALE SG, BOWDEN DM (1988) l-Methyl-4-phenyl-l,2,3,6-
tetrahydropyridine-induced parkinsonian syndrome in Macaca fascicularis: which midbrain dopaminergic neurons are lost? Neuroscience, 24, 161 174. GIBB WRG, LEES AJ (1988) The relevance of the Lewy body to the pathogenesis of idiopathic Parkinson's disease. Journal of Neurology, Neurosurgery, and Psychiatry, 51, 745-752. GIBB WRG, FEARNLEY JM, LEES AJ (1990) The anatomy and pigmentation of the human substantia nigra in relation to selective neuronal vulnerability. Advances in Neurology, 53, 3134. GRAHAM DG (1978) Oxidative pathways for catecholamines in the genesis of neuromelanin and cytotoxic quinones. Molecular Pharmacology, 14, 633643. GRAYBIEL AM, BESSON M-J, WEBER E (1989) Neuroleptic-sensitive binding sites in the nigrostriatal system: evidence for differential distribution of sigma sites in the substantia nigra, pars compacta of the cat. Journal of Neuroscience, 9, 326338.
HANSEN L, SALMON D, GALASKO D, MASUAH E, KATZMAN R, DETERESA R et al. (1990) The Lewy
body variant of Alzheimer's disease: a clinical and pathological entity. Neurology, Cleveland, 40, 18. HASSLER R (1937) Zur Normalanatomie der Substantia nigra. Versuch einer architektonischen Gliederung. Journal filr Psychologie und Neurologie, 48, 1 55. HASSLER R (1938) Zur Pathologie der Paralysis agitans und des postenzephalitischen Parkinsonismus. Journal far Psychologie und Neurologie, 48, 387 - 4 7 6 . HIRAI S (1968) Histochemical study on the regressive degeneration of the senile brain, with special reference to the aging of the substantia nigra. Advances in Neurological Science, Tokyo, 12, 848-849 (in Japanese). HIRSCH E, GRAYBIEL AM, AGID YA (1988) Melanized dopaminergic neurons are differentially susceptible to degeneration in Parkinson's disease. Nature, London, 334, 345 348. HODGE CF (1894) Changes in ganglion cells from birth to senile death. Observations on man and honeybee. Journal of Physiology, London, 17, 129134. HOEHN MM, YAHR MD (1967) Parkinsonism: onset progression, and mortality. Neurology, Minneapolis, 17, 427-442. JELLINGER K, DANIELCZYK W (1968) 'Striato-nigrale degeneration'. Acta Neuropathologica, Berlin, 10, 242-257. JELLINGER K (1971) Progressive supranuclear palsy (subcortical argyrophilic dystrophy). Acta Neuropathologica, Berlin, 19, 347-352. KJSH SJ, SHANNAK K, HORNYKIEWICZ O (1988) Uneven pattern of dopamine loss in the striatum of patients with idiopathic Parkinson's disease: pathophysiologic and clinical implications. New England Journal of Medicine, 318, 876-880. KONZLE H (1975) Bilateral projections from precentral motor cortex to the putamen and other parts of the basal ganglia. An autoradiographic study in Macaca fascicularis. Brain Research, Amsterdam, 88, 195-209. MCGEER PL, MCGEER EG, SUZUKI JS (1977) Aging and extrapyramidal function. Archives of Neurology, Chicago, 34, 3 3 - 3 5 . MCGEER PL, ITAGAKI S, AKIYAMA H, MCGEER EG (1988) Rate of cell death in parkinsonism indicates active neuropathological process. Annals of Neurology, 24, 574-576. MANN DMA, YATES PO (1983) Possible role of neuromelanin in the pathogenesis of Parkinson's disease. Mechanisms of Ageing and Development, 21, 193 203.
2301
MANN DMA, YATES PO, MARCYNIUK B (1984) Monoaminergic neurotransmitter systems in presenile Alzheimer's disease and in senile dementia of Alzheimer type. Clinical Neuropathology, 3, 199205. MARSDEN CD (1983) Neuromelanin and Parkinson's disease. Journal of Neural Transmission, 19, Supplement, 121-141. MARSDEN CD (1990) Parkinson's disease. Lancet, 335, 948-952. MATTOCK C, MARMOT M, STERN G (1988) Could Parkinson's disease follow intra-uterine influenza?: a speculative hypothesis. Journal of Neurology, Neurosurgery, and Psychiatry, 51, 753756. OLSON L, SEIGER A, FUXE K (1972) Heterogeneity of striatal and limbic dopamine innervation: highly fluorescent islands in developing and adult rats. Brain Research, Amsterdam, 44, 283-288. OLSZEWSKI J, BAXTER D (1954) Cytoarchitecture of the Human Brain Stem. Basel: S. Karger, pp. 131-132. PARENT A, MACKEY A, DE BELLEFEUILLE L (1983) The subcortical afferents to caudate nucleus and putamen in primate: a fluorescence retrograde double labeling study. Neuroscience, 10, 1137 1150. PIFL C, SCHINGNITZ G, HORNYKIEWICZ O (1988) The neurotoxin MPTP does not reproduce in the rhesus monkey the interregional pattern of striatal dopamine loss typical of human idiopathic Parkinson's disease. Neuroscience Letters, 92, 228-233. RAJPUT AH, KAZI KH, ROZDILSKY B (1972) Striatonigral degeneration response to levodopa therapy. Journal of the Neurological Sciences, 16, 331341. RJEDERER P, WUKETICH S (1976) Time course of nigrostriatal degeneration in Parkinson's disease. A detailed study of influential factors in human brain amine analysis. Journal of Neural Transmission, 38, 277-301. SCHELOSKY L, POEWE W (1990) Topographical onset and progression of motor symptoms in idiopathic Parkinson's disease. Movement Disorders, 5, Supplement 1, 13.
SCHERMAN D, DESNOS C, DARCHEN F, POLLAK P, JAVOY-AGID F, AGID Y (1989) Striatal dopamine
deficiency in Parkinson's disease: role of aging. Annab of Neurology, 26, 551557. SCHOENBERG BS (1986) Descriptive epidemiology of Parkinson's disease: disease distribution and hypothesis formulation. Advances in Neurology, 45, 277284.
SHARPE JA, REWCASTLE NB, LLOYD KG, HORNYKIEWICZ O, HILL M, TASKER RR (1973) Striatonigral
degeneration: response to levodopa therapy with pathological and neurochemical correlation. Journal of the Neurological Sciences, 19, 275-286. STEELE JC, RJCHARDSON JC, OLSZEWSKI J (1964) Progressive supranuclear palsy: a heterogeneous degeneration involving the brain stem, basal ganglia and cerebellum with vertical gaze and pseudobulbar palsy, nuchal dystonia and dementia. Archives of Neurology, Chicago, 10, 333359. SUNG JH, MASTRI AR, SEGAL E (1979) Pathology of Shy-Drager syndrome. Journal of Neuropathology and Experimental Neurology, 38, 353368. SZABO J (1980) Organization of the ascending striatal afferents in monkeys. Journal of Comparative Neurology, 189, 307-321. TAKEI Y, MIRRA SS (1973) Striatonigral degeneration: a form of multiple system atrophy with clinical parkinsonism. In: Progress in Neuropathology, Volume II. Edited by H. M. Zimmerman. New York and London: Grune and Stratton, pp. 217251. TATTON WG, LEE RG (1975) Evidence for abnormal long-loop reflexes in rigid parkinsonian patients. Brain Research, Amsterdam, 100, 671676. TERAVAINEN H, CALNE DB (1983) Motor system in normal aging and Parkinson's disease. In: The Neurology of Aging. Edited by R. Katzman and R. Terry. Philadelphia: F. A. Davis, pp. 85-109. TROTTER JL (1973) Striatonigral degeneration, Alzheimer's disease, and inflammatory changes. Neurology, Minneapolis, 23, 1211-1216. (Received August 1, 1990. Revised March 14, 1991. Accepted April 9, 1991)