- Email: [email protected]
Absence of neurodegeneration in the thalamus and caudate of elderly patients with schizophrenia
Psychiatry Research 93 Ž2000. 103]110
Absence of neurodegeneration in the thalamus and caudate of elderly patients with schizophrenia Eric Falke a , Li-Ying Hana , Steven E. Arnolda,U a
Laboratory for Cellular and Molecular Neuropathology, Center for Neurobiology and Beha¨ ior, Department of Psychiatry, Uni¨ ersity of Pennsyl¨ ania, Philadelphia, PA 19104, USA Received 1 September 1999; received in revised form 15 December 1999; accepted 29 December 1999
Abstract The cognitive and functional deterioration observed in many ‘poor-outcome’ patients with schizophrenia suggests an ongoing neurodegenerative process. Diagnostic neuropathologic studies have excluded known neurodegenerative diseases as the cause of this dementia, and in a previous quantitative investigation of neurodegeneration and neural injury in this population we found no abnormalities in the cerebral cortex. However, it is possible that the deterioration observed in these patients could be due to subcortical neurodegenerative processes. Neurodegeneration and neural injury in the caudate nucleus and mediodorsal nucleus of the thalamus were investigated in a postmortem study of 11 prospectively accrued, clinically well-characterized elderly people with schizophrenia, 11 elderly control subjects with no neuropsychiatric illness, and 12 subjects with Alzheimer’s disease. Traditional and immunohistochemical staining and unbiased computerized counting methods were used to quantify common markers of neurodegeneration and neural injury Žneuron loss, neurofibrillary tangles, astrocytosis, microgliosis.. No statistically significant differences were found between schizophrenia and control subjects for the densities of any markers. There is no evidence that abnormal neurodegeneration occurs in these two important subcortical structures. Q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Postmortem; Astrocyte; Microglia; Neurofibrillary tangle; Alzheimer’s disease
U
Corresponding author. 142 Clinical Research Building, 415 Curie Blvd., Philadelphia, PA 19104, USA. Tel.: q1-215-573-2840; fax: q1-215-573-2041. E-mail address: [email protected] ŽS.E. Arnold. 0165-1781r00r$ - see front matter Q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 1 7 8 1 Ž 0 0 . 0 0 1 0 4 - 9
104
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
1. Introduction While neurodevelopmental processes appear to play a defining role in the pathobiology of schizophrenia ŽHarrison, 1997., some patients exhibit a deteriorating course suggestive of a progressive, neurodegenerative process as well ŽWinokur et al., 1987; Carpenter and Kirkpatrick, 1988.. In particular, we and others have observed a high frequency of cognitive and functional impairments in elderly, chronically institutionalized patients ŽArnold et al., 1995; Davidson et al., 1995., with as many as two-thirds of these persons meeting DSM-IV criteria for dementia. However, diagnostic neuropathologic studies have failed to identify any abnormalities in the cerebral cortex to explain the dementia in the vast majority of these patients. In several quantitative neuropathological studies of elderly ‘poor-outcome’ individuals with schizophrenia, we found no evidence of neurodegeneration or neural injury in temporal, frontal, or occipital cortices ŽArnold et al., 1998.. However, it remains possible that neurodegeneration that principally affects subcortical regions could occur and be related to the dementia ŽPartlow et al., 1992; Savage, 1997.. In support of this are two earlier reports that described astrocytosis in subcortical and periventricular regions in schizophrenia ŽStevens, 1982; Bruton et al., 1990.. The present work extends our search into two subcortical areas: the mediodorsal nucleus ŽMD. of the thalamus and the caudate nucleus of the basal ganglia. These are crucial components of several frontal]limbic]subcortical circuits that play important roles in cognition, memory, and behavior ŽMega and Cummings, 1994.. Damage to these circuits can cause dementia as well as psychotic and other schizophrenia-like behaviors ŽPantelis et al., 1992.. For instance, limbic nuclei of the thalamus Žincluding the MD. have been reported to exhibit neuron and synapse loss as well as neurofibrillary changes in Alzheimer’s disease ŽBraak and Braak, 1991; Paskavitz et al., 1995.. This is likely to contribute to the neuropsychiatric dysfunction in Alzheimer’s disease. These regions also have been implicated in the pathoetiology of schizophrenia ŽPantelis et al., 1992;
Arnold and Trojanowski, 1996.. Reduced neuron density, decreased total neuron number, and smaller volume of the MD of institutionalized patients with schizophrenia have been reported ŽPakkenberg, 1990., as has fibrillary gliosis of periventricular regions, including the caudate and MD ŽStevens, 1982; Bruton et al., 1990.. In this study, we quantify the densities of neurons and immunohistochemically labeled neurofibrillary tangles ŽNFTs., astrocytes, and microglia in persons with schizophrenia, Alzheimer’s disease, and non-psychiatric controls. NFTs are one of the hallmark lesions of Alzheimer’s disease ŽAD. and correlate with severity of dementia in AD ŽArnold et al., 1991; Arriagada et al., 1992.. Astrocytosis Žgliosis. is a response to almost every type of injury or disease in the central nervous system, characterized by an up-regulation of glial fibrillary acidic protein ŽGFAP. within astrocytes as well as hypertrophy and hyperplasia ŽDuchen, 1992; Norton et al., 1992.. Similarly, microglia proliferate and enlarge in response to many pathological conditions ŽNakajima and Kohsaka, 1993..
2. Methods 2.1. Case materials Thalamic and basal ganglia blocks were dissected at autopsy from the brains of 11 patients with schizophrenia, 11 age-compatible control patients with no neuropsychiatric disease, and 12 patients with AD. Clinical information is provided in Table 1. An MD block was available from all of these subjects, except for one of the normal patients. Caudate was available in 10 subjects with schizophrenia, seven non-neuropsychiatric controls, and six AD subjects. Nuclei from only one hemisphere per case were examined. All subjects with schizophrenia had been participants in a prospective clinicopathological studies program ŽArnold et al., 1995. in which diagnosis was established according to DSM-III-RrDSM-IV criteria based on history obtained in the medical chart, interview with caregivers, and direct clinical examination. All had required chronic hospital-
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
105
Table 1 Case information
Age, years ŽS.D.. Sex, MrF PMI, h ŽS.D.. Brain wt., g ŽS.D..
Schizophrenia
Non-neuropsychiatric control
Alzheimer’s disease
80.5 Ž8.6. 3r9 10.7 Ž3.4. 1213 Ž169.
77.6 Ž13.2. 7r4 12.4 Ž5.8. 1283 Ž199.
79.4 Ž6.7. 2r10 10.9 Ž5.9. 1116 Ž124.
ization because of severe schizophrenia-related symptomatology including cognitive and functional impairment. Brain tissues from normal and AD controls were obtained through the University of Pennsylvania’s Alzheimer Disease Center Core. There were no significant differences between the three groups for age or post-morten interval ŽPMI.. Gross and microscopic diagnostic neuropathological examination results were conducted in all cases. The brains of four schizophrenia patients had minor neuropathologic findings that were unlikely to be related to their mental status Žthree with lacunar infarcts that did not involve the thalamus or caudate, and one with a posterior fossa meningioma.. The remaining seven were without neuropathologic abnormalities. Similarly, the brains of three normal subjects exhibited incidental abnormalities Žone with a hemorrhagic infarct, one with bitemporal contusions, and one with a small cerebellar adenocarcinoma metastasis., while the other eight subjects were normal. Except for one AD patient who had an incidental microinfarct, the brains of AD patients were without neuropathological abnormalities, aside from abundant senile plaques and NFTs. Analyses were conducted both including and excluding brains with incidental findings and results were identical. 2.2. Tissue processing and immunostaining Blocks including the thalamus and the caudate]putamen were dissected at autopsy and fixed in ethanol Žethyl alcohol, 70%; sodium chloride, 150 mmolrl. for 24 h. The blocks were then paraffin embedded, and cut into 20-mm thick sections. Kluver and Barrea’s luxol fast blue stain for myelin with a cresyl violet Ž1%. counter stain
ŽLowe and Cox, 1990. was used in one section for identification and delineation of nuclei according to their cytoarchitectural and myeloarchitectural appearances ŽAlheid et al., 1990; Armstrong, 1990.. For neuron density measurements, adjacent sections were stained with 1% cresyl violet for Nissl substances. Pathologic markers were immunohistochemically labeled with the following antibodies: PHF-1 for NFTs ŽGreenberg and Davies, 1990., 2.2B10 for glial fibrillary acidic protein ŽGFAP. in astrocytes ŽTrojanowski et al., 1984. and CD68 for resting and active microglia ŽDAKO Corp., Carpinteria, CA.. Immunohistochemistry was performed using a previously described peroxidase]antiperoxidase procedure with diaminobenzidine as the reporter ŽArai et al., 1991; Arnold et al., 1998.. Staining for each marker was performed in one slide per case in single runs which included all cases, with precisely timed DAB development. These sections from each region of interest were used in the quantitative density determinations described below. 2.3. Neuropathology marker density estimation The densities of neurons, PHF-1-immunoreactive Ž-ir. NFTs, GFAP-ir astrocytes, and CD68-ir microglia were determined with computer-assisted microscopy using a random systematic samplingrfractionator paradigm ŽWest, 1993; Hyman et al., 1998. with specialized software ŽStereoInvestigator, MicroBrightfield Inc., Colchester, VT.. Briefly, after cytoarchitecturally delimiting the region of interest at low magnification, a grid of predetermined size was placed randomly over the entire region. Magnification was raised to 400 = and the program stopped at each intersec-
106
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
tion point on the grid for counting. The operator focused through the section depth as the fields were visualized on the video monitor with a superimposed counting frame. All objects within the 100 = 100 mm frame that did not touch the exclusion lines of the box Žbottom and left borders. were counted. Depending on the density of the objects, counts were made in between 30 and 65 framesrsection. To decrease bias introduced by split-cell artifact in our 20-mm thick sections, an Abercrombie correction factor was determined and applied for all marker densities ŽAbercrombie, 1946; Clarke, 1992.. Briefly, using the 100 = oil-immersion objective, the thickness of sections was optically determined and the diameter of 15 randomly selected objects Žneurofibrillary tangles, neurons, astrocytes, microglia. was calculated after measuring the cross-sectional area of each object through its midsection. Diameter was estimated by the formula ds 2 6Žarearp .. The diameters of the 15 objects in each section were averaged. The Abercrombie-corrected densities of objects Ž N . were then determined according to N s nw t srŽ t s q d n ., where n s raw object density, t s s section thickness and d n s average object diameter. Slides were analyzed in random order by one of two operators ŽE.A.F., L.Y.H.. who were blind to case information. Preliminary studies indicated high interrater and intrarater reliability with an intraclass correlation coefficient greater that 0.80 for the markers. For neuron density measurements, neurons were identified based on their cresyl violet labeling, characteristic size, and heterogeneous nuclear staining. NFTs were labeled with PHF-1 and had a characteristic ‘flared’ appearance. The GFAP-ir astrocytes had a visible nucleus and characteristic astrocytic processes. CD68-labeled active and resting microglia were small relative to neurons and astrocytes and had a distinguishable nucleus. 2.4. Data analysis and statistics The effects of aging, sex, and PMI were assessed for each pathologic measure in both regions within the normal group using Pearson
product]moment correlations and chi-square analyses. Because age is known to be associated with each of our pathological markers, age was used as a covariate in subsequent analyses of covariance ŽANCOVAs. assessing between-group differences in marker densities. The effects of age, sex and antipsychotic drug dose were separately assessed for the schizophrenia group to investigate the possibility that there was an interaction between these factors, disease state and accumulation of neurodegenerative lesions. Neuropathological marker findings were analyzed both as raw densities Žper mm2 . and as a ratio of neuropathological marker density per neuron density in each case. Given that cell density is a function of the number of cells within a given region in relation to the amount of neuropil between them, it is possible to have changes in the total number of cells without any changes in density, or vice versa. In order to control for this, we calculated the ratio of neurodegeneration marker densities ŽNFTs, astrocytes, microglia. per neuron density in adjacent slides. This creates a dimensionless index, thus protecting against possible differences in MD or caudate size, differences in neuropil, or differences in shrinkage during tissue processing. Between-group differences in neuropathological markers were analyzed in separate ANCOVAs with post-hoc Scheffe ´ tests for individual group differences.
3. Results 3.1. Effects of age, PMI, and antipsychotic medication as potentially confounding ¨ ariables There were no significant correlations between age, PMI or antipsychotic medication dose and any neuropathological measure in the normal or schizophrenia groups in the MD. In the caudate, we observed a significant correlation between age and microglia density Ž R s 0.87, P- 0.01. in the normal group, but not the schizophrenia group. Incidentally, we also observed a positive correlation between age and NFTs in the schizophrenia group in the MD Ž R s 0.74, Ps 0.01. but a negative correlation in the caudate Ž r s y0.64, P-
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
0.05.. There were no other significant correlations between potential confounds and neuropathological indices. There were no sex differences for any neuropathologic measure in any group. Right and left hemispheres were equally represented in each group and there were no significant differences in marker values between hemispheres by group or for all cases together. 3.2. Markers of neurodegeneration and neural injury There were no differences between the normal control, schizophrenia, and AD groups in the densities of the pathologic markers in either nucleus except for neurofibrillary tangles ŽTable 2..
107
NFT density and the NFTrneuron ratio were significantly higher in the AD group than in both normal and schizophrenia groups in the MD, but not the caudate. While differences were not significant, Nissl-stained neuron density was the lowest in AD subjects and highest in the schizophrenia group in the MD, but in the caudate neuron density was highest in the AD subjects and lowest in schizophrenia. For the ratios of GFAP astrocytes per neuron, AD subjects again exhibited the highest values of this index in the MD, but lowest in the caudate. Similarly, the CD68 microgliarneuron ratio was highest in the AD group in the MD, but lowest in the caudate.
Table 2 Mean ŽS.D.. density values, ANCOVA and post-hoc Scheffe ´ S-test comparisons for measures of neurodegeneration in the mediodorsal nucleus of the thalamus ŽMD. and the caudate nucleus a Means ŽS.D.. Normal
SZ
AD
105.9 Ž21.5. 304.9 Ž57.8.
92.1 Ž17.0. 344.5 Ž43.8.
Ancova F, P
Scheffe ´ S-tests, P-values Nl vs. SZ
Nl vs. AD
SZ vs. AD
1.47, 0.24 3.60, 0.05
0.47 0.42
0.93 0.06
0.26 0.33
2
Nissl-stained neurons (rmm ) MD 95.3 Ž19.5. Caudate 273.2 Ž46.3.
PHF-tau neurofibrillary tangles (rmm2 ) MD 0.08 Ž0.17. 0.16 Ž0.43. Caudate 2.42 Ž3.11. 2.19 Ž1.33.
11.28 Ž10.36. 3.39 Ž3.10.
11.65, 0.0002 0.41, 0.67
0.99 0.98
0.001 0.78
0.001 0.65
PHF-tau neurofibrillary tanglesr neuron MD 0.00 Ž0.00. 0.00 Ž0.00. Caudate 0.01 Ž0.01. 0.01 Ž0.00.
0.13 Ž0.13. 0.01 Ž0.01.
9.54, 0.0007 0.29, 0.74
0.99 0.93
0.004 0.92
0.003 0.74
2.03, 0.15 0.78, 0.47
0.92 0.71
0.36 0.81
0.18 0.38
2.07, 0.14 0.78, 0.48
0.84 0.93
0.43 0.69
0.16 0.46
2.53, 0.09 0.03, 0.97
0.20 0.99
0.11 0.97
0.95 0.97
2.38, 0.11 1.15, 0.34
0.37 0.60
0.10 0.41
0.74 0.85
GFAP astrocytes (rmm2 ) MD 18.2 Ž19.2. Caudate 16.1 Ž17.7. GFAP astrocytesr neuron MD 0.22 Ž0.27. Caudate 0.06 Ž0.06. CD68 microglia (rmm2 ) MD 50.3 Ž41.8. Caudate 44.7 Ž16.8. CD68 microgliar neuron MD 0.51 Ž0.39. Caudate 0.17 Ž0.08.
13.0 Ž12.6. 24.2 Ž23.9.
0.13 Ž0.15. 0.07 Ž0.08.
80.8 Ž30. 4. 44.6 Ž22.5.
0.83 Ž0.46. 0.14 Ž0.07.
36.6 Ž44.2. 8.9 Ž8.5.
0.41 Ž0.48. 0.03 Ž0.02.
85.7 Ž41.5. 41.6 Ž21.8.
0.99 Ž0.61. 0.12 Ž0.05.
SZ, schizophrenia; AD, Alzheimer’s disease; Nl, Normal a Ratios given per neuron are based on the density of Nissl-stained neurons.
108
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
4. Discussion This study examined the possibility of neurodegeneration and neural injury in two important subcortical nuclei in a sample of elderly, pooroutcome subjects with schizophrenia by assessing the densities of neurons, PHF-tau-ir NFTs, GFAP-ir astrocytes, and CD68-ir microglia. There were no differences between patients with schizophrenia and control patients without neuropsychiatric disease in either of the regions on any of the four measures. These data indicate that neurodegenerative disease-related or other processes of neural injury beyond those observed with normal aging in these subcortical nuclei do not play a role in this population’s highly deteriorated state. These findings are similar to those of our previously reported investigation of neurodegeneration and neural injury in the hippocampus and several neocortical regions ŽArnold et al., 1998.. Some investigators have suggested that a higher incidence of AD in poor-outcome schizophrenia patients might explain their high frequency of dementia ŽBuhl and Bojsen-Moller, 1988; Prohovnik et al., 1993; Arnold and Trojanowski, 1996., but our data do not support this. AD patients exhibited a higher density of NFTs in the MD nucleus than did schizophrenia subjects with or without dementia, and there was no difference between normal and schizophrenia patients on this measure. This finding is consistent with most of the more recent studies showing an absence of AD-related pathology in elderly patients with schizophrenia ŽBruton et al., 1990; Powchik et al., 1993; Purohit et al., 1993; Arnold et al., 1994, 1998; Harrison, 1999.. Postmortem studies of the MD nucleus in schizophrenia have been controversial, with some researchers finding decreased neuron number, density, and volume, and others finding no change ŽHeckers, 1997; Jones, 1997.. Pakkenberg has reported a marginally significant decrease in neuron density, a 40% reduction in total neuron number, and a 25% reduction in the volume of the mediodorsal nucleus ŽPakkenberg, 1990.. We found a slightly higher mean neuron density in
schizophrenia, although this was not significant. Other studies of the thalamus have reported normal volumes for all major subnuclei but decreased thickness of periventricular gray ŽLesch and Bogerts, 1984. and decreased parvalbuminimmunoreactive neurons in the anteroventral nucleus ŽDanos et al., 1998.. Neither our study nor others’ findings indicate that there is abnormal neurodegeneration in schizophrenia during aging. If there were neurodegeneration-related neuron loss, one might expect the ratio of astrocytes to neurons to increase. However, there was no difference in this ratio between schizophrenia and normal groups. While there has been disagreement in the past, consensus is building that there is no noteworthy astrocytosis in schizophrenia ŽArnold and Trojanowski, 1996; Harrison, 1999.. Early postmortem studies using traditional histological stains Že.g. Holzer. reported evidence of fibrillary gliosis in periventricular areas ŽStevens, 1982; Bruton et al., 1990. while other, more recent studies using Nissl stains and GFAP immunohistochemistry have failed to find astrocytosis. Similar to our findings in the cerebral cortex, we now report no differences using GFAP immunohistochemistry in MD and caudate. GFAP expression in response to neural injury is dynamic ŽNorton et al., 1992. and indicates the state of astrocyte reactivity for a variable time prior to death, but does not rule out a previous astrocyte reaction to earlier injury. It is still possible that remote neural injury could leave fibrillary astrocytic changes that would not be accompanied by persistent up-regulation of GFAP. Our data do not support the hypothesis that abnormal neurodegenerative processes or neural injury are evident in poor-outcome schizophrenia patients compared with normal elderly controls. However, it is still possible that neurodevelopmental abnormalities in important cognition-related areas of the brain, including the MD and caudate, may make these areas especially vulnerable to the effects of normal aging, such as the small accumulations of neurodegenerative lesions that typically occur. Thus, an interaction between neurodevelopmental abnormalities and age-
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
related neurodegenerative changes could manifest in the clinical deterioration evident in so many of these poor outcome patients.
Acknowledgements This work was supported by NIH grants MH43880 and MH55199. The authors thank members of the Penn Schizophrenia Mental Health Clinical Research Center and the Department of Pathology and Laboratory Medicine, and the patients and families whose generosity made this research possible. References Abercrombie, M., 1946. Estimation of nuclear populations from microtome sections. Anatomic Record 94, 239]247. Alheid, G.F., Heimer, L., Switzer, R.C., 1990. Basal ganglia. In: Paxinos, G. ŽEd.., The Human Nervous System. Academic Press, San Diego. Arai, H., Lee, V.M.-Y., Messinger, M.L., Greenberg, B.D., Lowery, D.E., Trojanowski, J.Q., 1991. Expression patterns of beta-amyloid precursor protein Žbeta-APP. in neural and nonneural human tissues from Alzheimer’s disease and control subjects. Annals of Neurology 30, 686]693. Armstrong, E., 1990. Limbic thalamus: anterior and mediodorsal nuclei. In: Paxinos, G. ŽEd.., The Human Nervous System. Academic Press, San Diego. Arnold, S.E., Trojanowski, J.Q., 1996. Recent advances in defining the neuropathology of schizophrenia. Acta Neuropathologica 92, 217]231. Arnold, S.E., Hyman, B.T., Flory, J., Damasio, A.R., Van Hoesen, G.W., 1991. The topographical and neuroanatomical distribution of neurofibrillary tangles and neuritic plaques in the cerebral cortex of patients with Alzheimer’s disease. Cerebral Cortex 1, 103]116. Arnold, S.E., Franz, B.R., Trojanowski, J.Q., 1994. Elderly patients with schizophrenia exhibit infrequent neurodegenerative lesions. Neurobiology of Aging 15 Ž3., 299]303. Arnold, S.E., Gur, R.E., Shapiro, R.M., Fisher, K.R., Moberg, P.J., Gibney, M.R., Gur, R.C., Blackwell, P., Trojanowski, J.Q., 1995. Prospective clinicopathological studies of schizophrenia: accrual and assessment. American Journal of Psychiatry 152, 731]737. Arnold, S.E., Trojanowski, J.Q., Gur, R.E., Han, L.-Y., Choi, C., 1998. Absence of neurodegeneration and neural injury in the cerebral cortex in a sample of elderly patients with schizophrenia. Archives of General Psychiatry 55, 225]232. Arriagada, P.V., Growdon, J.H., Hedley-Whyte, E.T., Hyman, B.T., 1992. Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer’s disease. Neurology 42, 631]639.
109
Braak, H., Braak, E., 1991. Alzheimer’s disease affects limbic nuclei of the thalamus. Acta Neuropathologica 81 Ž3., 261]268. Bruton, C.J., Crow, T.J., Frith, C.D., Johnstone, E.C., Owens, D.G.C., Roberts, G.W., 1990. Schizophrenia and the brain: a prospective clinico-neuropathological study. Psychological Medicine 20 Ž2., 285]304. Buhl, L., Bojsen-Moller, M., 1988. Frequency of Alzheimer’s disease in a postmortem study of psychiatric patients. Danish Medical Bulletin 35, 288]290. Carpenter, W.T., Kirkpatrick, B., 1988. The heterogeneity of the long-term course of schizophrenia. Schizophrenia Bulletin 14 Ž4., 515]542. Clarke, P.G.H., 1992. How inaccurate is the Abercrombie correction factor for cell counts. Trends Neurosci 15, 211]212. Danos, P., Baumann, B., Bernstein, H.G., Franz, M., Stauch, R., Northoff, G., Krell, D., Falkai, P., Bogerts, B., 1998. Schizophrenia and anteroventral thalamic nucleus: selective decrease of parvalbumin-immunoreactive thalamocortical projection neurons. Psychiatry Research: Neuroimaging 82 Ž1., 1]10. Davidson, M., Harvey, P.D., Powchik, P., Parella, M., White, L., Knobler, H.Y., Losonczy, M.F., Keefe, R.S.E., Katz, S., Frecska, E., 1995. Severity of symptoms in chronically institutionalized geriatric schizophrenic patients. American Journal of Psychiatry 152, 197]207. Duchen, L.W., 1992. General pathology of neurons and neuroglia. In: Adams, J.H., Duchen, L.W. ŽEds.., Greenfield’s Neuropathology. Oxford University Press, New York. Greenberg, S.G., Davies, P., 1990. A preparation of Alzheimer paired helical filaments that displays distinct tau proteins by polyacrylamide gel electrophoresis. Proceedings of the National Academy of Sciences ŽUSA. 87, 5827]5831. Harrison, P.J., 1997. Schizophrenia: a disorder of neurodevelopment? Current Opinion in Neurobiology 7, 285]289. Harrison, P.J., 1999. The neuropathology of schizophrenia. A critical review of the data and their interpretation. Brain 122, 593]624. Heckers, S., 1997. Neuropathology of schizophrenia: cortex, thalamus, basal ganglia and neurotransmitter-specific projection systems. Schizophrenia Bulletin 23, 403]421. Hyman, B.T., Gomez-Isla, T., Irizarry, M.C., 1998. Stereology: a practical primer for neuropathology. Journal of Neuropathology and Experimental Neurology 57 Ž4., 305]310. Jones, E.G., 1997. Cortical development and thalamic pathology in schizophrenia wsee commentsx. Schizophrenia Bulletin 23 Ž3., 483]501. Lesch, A., Bogerts, B., 1984. The diencephalon in schizophrenia: evidence for reduced thickness of the periventricular grey matter. European Archives of Psychiatry and Neurological Sciences 234 Ž4., 212]219. Lowe, J., Cox, G., 1990. Neuropathological techniques. In: Bancroft, J.D., Stevens, A. ŽEds.., Theory and Practice of Histological Techniques. Churchill Livingtstone, Edinburgh. Mega, M.S., Cummings, J.L., 1994. Frontal]subcortical cir-
110
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
cuits and neuropsychiatric disorders. Journal of Neuropsychiatry and Clinical Neurosciences 6 Ž4., 358]370. Nakajima, K., Kohsaka, S., 1993. Functional roles of microglia in the brain. Neuroscience Research 17 Ž3., 187]203. Norton, W.T., Aquino, D.A., Hozumi, I., Chiu, F.-C., Brosnan, C.F., 1992. Quantitative aspects of reactive gliosis: a review. Neurochemical Research 17, 877]885. Pakkenberg, B., 1990. Pronounced reduction of total neuron number in mediodorsal thalamic nucleus and nucleus accumbens in schizophrenia. Archives of General Psychiatry 47, 1023]1028. Pantelis, C., Barnes, T.R., Nelson, H.E., 1992. Is the concept of frontal-subcortical dementia relevant to schizophrenia? British Journal of Psychiatry 160, 442]460. Partlow, G.D., delCarpio-O’Donovan, R., Melanson, D., Peters, T.M., 1992. Bilateral thalamic glioma: review of eight cases with personality change and mental deterioration. American Journal of Neuroradiology 13, 1225]1230. Paskavitz, J.F., Lippa, C.F., Hamos, J.E., Pulaski-Salo, D., Drachman, D.A., 1995. Role of the dorsomedial nucleus of the thalamus in Alzheimer’s disease. Journal of Geriatrics, Psychiatry and Neurology 8 Ž1., 32]37. Powchik, P., Davidson, M., Nemeroff, C.B., Haroutunian, V., Purohit, D.P., Losonczy, M., Bissette, G., Perl, D., Ghanbari, H., Miller, B., Davis, K.L., 1993. Alzheimer’s-diseaserelated protein in geriatric schizophrenic patients with cog-
nitive impairment. American Journal of Psychiatry 150 Ž11., 1726]1727. Prohovnik, I., Dwork, A.J., Kaufman, M.A., Wilson, N., 1993. Alzheimer-type neuropathology in elderly schizophrenia patients. Schizophrenia Bulletin 19, 805]816. Purohit, D.P., Davidson, M., Perl, D.P., Powchik, P., Haroutunian, V.H., Bierer, L.M., McCrystal, J., Losonczy, M., Davis, K.L., 1993. Severe cognitive impairment in elderly schizophrenic patients: a clinicopathological study. Biological Psychiatry 33, 255]260. Savage, C.R., 1997. Neuropsychology of subcortical dementias. Psychiatric Clinics of North America 20, 911]931. Stevens, J.R., 1982. Neuropathology of schizophrenia. Archives of General Psychiatry 39, 1131]1139. Trojanowski, J.Q., Gordon, D., Obrocka, M., Lee, V.M.-Y., 1984. The developmental expression of neurofilament and glial filament proteins in the human pituitary gland: an immunohistochemical study with monoclonal antibodies. Developmental Brain Research 13, 229]239. West, M.J., 1993. New stereological methods for counting neurons. Neurobiology of Aging 14, 287]293. Winokur, G., Pfohl, B., Tsuang, M., 1987. A 40-year follow-up of hebephrenic-catatonic schizophrenia. In: Miller, N., Cohen, G. ŽEds.., Schizophrenia and Aging. Guilford Press, New York.
Absence of neurodegeneration in the thalamus and caudate of elderly patients with schizophrenia Eric Falke a , Li-Ying Hana , Steven E. Arnolda,U a
Laboratory for Cellular and Molecular Neuropathology, Center for Neurobiology and Beha¨ ior, Department of Psychiatry, Uni¨ ersity of Pennsyl¨ ania, Philadelphia, PA 19104, USA Received 1 September 1999; received in revised form 15 December 1999; accepted 29 December 1999
Abstract The cognitive and functional deterioration observed in many ‘poor-outcome’ patients with schizophrenia suggests an ongoing neurodegenerative process. Diagnostic neuropathologic studies have excluded known neurodegenerative diseases as the cause of this dementia, and in a previous quantitative investigation of neurodegeneration and neural injury in this population we found no abnormalities in the cerebral cortex. However, it is possible that the deterioration observed in these patients could be due to subcortical neurodegenerative processes. Neurodegeneration and neural injury in the caudate nucleus and mediodorsal nucleus of the thalamus were investigated in a postmortem study of 11 prospectively accrued, clinically well-characterized elderly people with schizophrenia, 11 elderly control subjects with no neuropsychiatric illness, and 12 subjects with Alzheimer’s disease. Traditional and immunohistochemical staining and unbiased computerized counting methods were used to quantify common markers of neurodegeneration and neural injury Žneuron loss, neurofibrillary tangles, astrocytosis, microgliosis.. No statistically significant differences were found between schizophrenia and control subjects for the densities of any markers. There is no evidence that abnormal neurodegeneration occurs in these two important subcortical structures. Q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Postmortem; Astrocyte; Microglia; Neurofibrillary tangle; Alzheimer’s disease
U
Corresponding author. 142 Clinical Research Building, 415 Curie Blvd., Philadelphia, PA 19104, USA. Tel.: q1-215-573-2840; fax: q1-215-573-2041. E-mail address: [email protected] ŽS.E. Arnold. 0165-1781r00r$ - see front matter Q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 1 7 8 1 Ž 0 0 . 0 0 1 0 4 - 9
104
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
1. Introduction While neurodevelopmental processes appear to play a defining role in the pathobiology of schizophrenia ŽHarrison, 1997., some patients exhibit a deteriorating course suggestive of a progressive, neurodegenerative process as well ŽWinokur et al., 1987; Carpenter and Kirkpatrick, 1988.. In particular, we and others have observed a high frequency of cognitive and functional impairments in elderly, chronically institutionalized patients ŽArnold et al., 1995; Davidson et al., 1995., with as many as two-thirds of these persons meeting DSM-IV criteria for dementia. However, diagnostic neuropathologic studies have failed to identify any abnormalities in the cerebral cortex to explain the dementia in the vast majority of these patients. In several quantitative neuropathological studies of elderly ‘poor-outcome’ individuals with schizophrenia, we found no evidence of neurodegeneration or neural injury in temporal, frontal, or occipital cortices ŽArnold et al., 1998.. However, it remains possible that neurodegeneration that principally affects subcortical regions could occur and be related to the dementia ŽPartlow et al., 1992; Savage, 1997.. In support of this are two earlier reports that described astrocytosis in subcortical and periventricular regions in schizophrenia ŽStevens, 1982; Bruton et al., 1990.. The present work extends our search into two subcortical areas: the mediodorsal nucleus ŽMD. of the thalamus and the caudate nucleus of the basal ganglia. These are crucial components of several frontal]limbic]subcortical circuits that play important roles in cognition, memory, and behavior ŽMega and Cummings, 1994.. Damage to these circuits can cause dementia as well as psychotic and other schizophrenia-like behaviors ŽPantelis et al., 1992.. For instance, limbic nuclei of the thalamus Žincluding the MD. have been reported to exhibit neuron and synapse loss as well as neurofibrillary changes in Alzheimer’s disease ŽBraak and Braak, 1991; Paskavitz et al., 1995.. This is likely to contribute to the neuropsychiatric dysfunction in Alzheimer’s disease. These regions also have been implicated in the pathoetiology of schizophrenia ŽPantelis et al., 1992;
Arnold and Trojanowski, 1996.. Reduced neuron density, decreased total neuron number, and smaller volume of the MD of institutionalized patients with schizophrenia have been reported ŽPakkenberg, 1990., as has fibrillary gliosis of periventricular regions, including the caudate and MD ŽStevens, 1982; Bruton et al., 1990.. In this study, we quantify the densities of neurons and immunohistochemically labeled neurofibrillary tangles ŽNFTs., astrocytes, and microglia in persons with schizophrenia, Alzheimer’s disease, and non-psychiatric controls. NFTs are one of the hallmark lesions of Alzheimer’s disease ŽAD. and correlate with severity of dementia in AD ŽArnold et al., 1991; Arriagada et al., 1992.. Astrocytosis Žgliosis. is a response to almost every type of injury or disease in the central nervous system, characterized by an up-regulation of glial fibrillary acidic protein ŽGFAP. within astrocytes as well as hypertrophy and hyperplasia ŽDuchen, 1992; Norton et al., 1992.. Similarly, microglia proliferate and enlarge in response to many pathological conditions ŽNakajima and Kohsaka, 1993..
2. Methods 2.1. Case materials Thalamic and basal ganglia blocks were dissected at autopsy from the brains of 11 patients with schizophrenia, 11 age-compatible control patients with no neuropsychiatric disease, and 12 patients with AD. Clinical information is provided in Table 1. An MD block was available from all of these subjects, except for one of the normal patients. Caudate was available in 10 subjects with schizophrenia, seven non-neuropsychiatric controls, and six AD subjects. Nuclei from only one hemisphere per case were examined. All subjects with schizophrenia had been participants in a prospective clinicopathological studies program ŽArnold et al., 1995. in which diagnosis was established according to DSM-III-RrDSM-IV criteria based on history obtained in the medical chart, interview with caregivers, and direct clinical examination. All had required chronic hospital-
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
105
Table 1 Case information
Age, years ŽS.D.. Sex, MrF PMI, h ŽS.D.. Brain wt., g ŽS.D..
Schizophrenia
Non-neuropsychiatric control
Alzheimer’s disease
80.5 Ž8.6. 3r9 10.7 Ž3.4. 1213 Ž169.
77.6 Ž13.2. 7r4 12.4 Ž5.8. 1283 Ž199.
79.4 Ž6.7. 2r10 10.9 Ž5.9. 1116 Ž124.
ization because of severe schizophrenia-related symptomatology including cognitive and functional impairment. Brain tissues from normal and AD controls were obtained through the University of Pennsylvania’s Alzheimer Disease Center Core. There were no significant differences between the three groups for age or post-morten interval ŽPMI.. Gross and microscopic diagnostic neuropathological examination results were conducted in all cases. The brains of four schizophrenia patients had minor neuropathologic findings that were unlikely to be related to their mental status Žthree with lacunar infarcts that did not involve the thalamus or caudate, and one with a posterior fossa meningioma.. The remaining seven were without neuropathologic abnormalities. Similarly, the brains of three normal subjects exhibited incidental abnormalities Žone with a hemorrhagic infarct, one with bitemporal contusions, and one with a small cerebellar adenocarcinoma metastasis., while the other eight subjects were normal. Except for one AD patient who had an incidental microinfarct, the brains of AD patients were without neuropathological abnormalities, aside from abundant senile plaques and NFTs. Analyses were conducted both including and excluding brains with incidental findings and results were identical. 2.2. Tissue processing and immunostaining Blocks including the thalamus and the caudate]putamen were dissected at autopsy and fixed in ethanol Žethyl alcohol, 70%; sodium chloride, 150 mmolrl. for 24 h. The blocks were then paraffin embedded, and cut into 20-mm thick sections. Kluver and Barrea’s luxol fast blue stain for myelin with a cresyl violet Ž1%. counter stain
ŽLowe and Cox, 1990. was used in one section for identification and delineation of nuclei according to their cytoarchitectural and myeloarchitectural appearances ŽAlheid et al., 1990; Armstrong, 1990.. For neuron density measurements, adjacent sections were stained with 1% cresyl violet for Nissl substances. Pathologic markers were immunohistochemically labeled with the following antibodies: PHF-1 for NFTs ŽGreenberg and Davies, 1990., 2.2B10 for glial fibrillary acidic protein ŽGFAP. in astrocytes ŽTrojanowski et al., 1984. and CD68 for resting and active microglia ŽDAKO Corp., Carpinteria, CA.. Immunohistochemistry was performed using a previously described peroxidase]antiperoxidase procedure with diaminobenzidine as the reporter ŽArai et al., 1991; Arnold et al., 1998.. Staining for each marker was performed in one slide per case in single runs which included all cases, with precisely timed DAB development. These sections from each region of interest were used in the quantitative density determinations described below. 2.3. Neuropathology marker density estimation The densities of neurons, PHF-1-immunoreactive Ž-ir. NFTs, GFAP-ir astrocytes, and CD68-ir microglia were determined with computer-assisted microscopy using a random systematic samplingrfractionator paradigm ŽWest, 1993; Hyman et al., 1998. with specialized software ŽStereoInvestigator, MicroBrightfield Inc., Colchester, VT.. Briefly, after cytoarchitecturally delimiting the region of interest at low magnification, a grid of predetermined size was placed randomly over the entire region. Magnification was raised to 400 = and the program stopped at each intersec-
106
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
tion point on the grid for counting. The operator focused through the section depth as the fields were visualized on the video monitor with a superimposed counting frame. All objects within the 100 = 100 mm frame that did not touch the exclusion lines of the box Žbottom and left borders. were counted. Depending on the density of the objects, counts were made in between 30 and 65 framesrsection. To decrease bias introduced by split-cell artifact in our 20-mm thick sections, an Abercrombie correction factor was determined and applied for all marker densities ŽAbercrombie, 1946; Clarke, 1992.. Briefly, using the 100 = oil-immersion objective, the thickness of sections was optically determined and the diameter of 15 randomly selected objects Žneurofibrillary tangles, neurons, astrocytes, microglia. was calculated after measuring the cross-sectional area of each object through its midsection. Diameter was estimated by the formula ds 2 6Žarearp .. The diameters of the 15 objects in each section were averaged. The Abercrombie-corrected densities of objects Ž N . were then determined according to N s nw t srŽ t s q d n ., where n s raw object density, t s s section thickness and d n s average object diameter. Slides were analyzed in random order by one of two operators ŽE.A.F., L.Y.H.. who were blind to case information. Preliminary studies indicated high interrater and intrarater reliability with an intraclass correlation coefficient greater that 0.80 for the markers. For neuron density measurements, neurons were identified based on their cresyl violet labeling, characteristic size, and heterogeneous nuclear staining. NFTs were labeled with PHF-1 and had a characteristic ‘flared’ appearance. The GFAP-ir astrocytes had a visible nucleus and characteristic astrocytic processes. CD68-labeled active and resting microglia were small relative to neurons and astrocytes and had a distinguishable nucleus. 2.4. Data analysis and statistics The effects of aging, sex, and PMI were assessed for each pathologic measure in both regions within the normal group using Pearson
product]moment correlations and chi-square analyses. Because age is known to be associated with each of our pathological markers, age was used as a covariate in subsequent analyses of covariance ŽANCOVAs. assessing between-group differences in marker densities. The effects of age, sex and antipsychotic drug dose were separately assessed for the schizophrenia group to investigate the possibility that there was an interaction between these factors, disease state and accumulation of neurodegenerative lesions. Neuropathological marker findings were analyzed both as raw densities Žper mm2 . and as a ratio of neuropathological marker density per neuron density in each case. Given that cell density is a function of the number of cells within a given region in relation to the amount of neuropil between them, it is possible to have changes in the total number of cells without any changes in density, or vice versa. In order to control for this, we calculated the ratio of neurodegeneration marker densities ŽNFTs, astrocytes, microglia. per neuron density in adjacent slides. This creates a dimensionless index, thus protecting against possible differences in MD or caudate size, differences in neuropil, or differences in shrinkage during tissue processing. Between-group differences in neuropathological markers were analyzed in separate ANCOVAs with post-hoc Scheffe ´ tests for individual group differences.
3. Results 3.1. Effects of age, PMI, and antipsychotic medication as potentially confounding ¨ ariables There were no significant correlations between age, PMI or antipsychotic medication dose and any neuropathological measure in the normal or schizophrenia groups in the MD. In the caudate, we observed a significant correlation between age and microglia density Ž R s 0.87, P- 0.01. in the normal group, but not the schizophrenia group. Incidentally, we also observed a positive correlation between age and NFTs in the schizophrenia group in the MD Ž R s 0.74, Ps 0.01. but a negative correlation in the caudate Ž r s y0.64, P-
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
0.05.. There were no other significant correlations between potential confounds and neuropathological indices. There were no sex differences for any neuropathologic measure in any group. Right and left hemispheres were equally represented in each group and there were no significant differences in marker values between hemispheres by group or for all cases together. 3.2. Markers of neurodegeneration and neural injury There were no differences between the normal control, schizophrenia, and AD groups in the densities of the pathologic markers in either nucleus except for neurofibrillary tangles ŽTable 2..
107
NFT density and the NFTrneuron ratio were significantly higher in the AD group than in both normal and schizophrenia groups in the MD, but not the caudate. While differences were not significant, Nissl-stained neuron density was the lowest in AD subjects and highest in the schizophrenia group in the MD, but in the caudate neuron density was highest in the AD subjects and lowest in schizophrenia. For the ratios of GFAP astrocytes per neuron, AD subjects again exhibited the highest values of this index in the MD, but lowest in the caudate. Similarly, the CD68 microgliarneuron ratio was highest in the AD group in the MD, but lowest in the caudate.
Table 2 Mean ŽS.D.. density values, ANCOVA and post-hoc Scheffe ´ S-test comparisons for measures of neurodegeneration in the mediodorsal nucleus of the thalamus ŽMD. and the caudate nucleus a Means ŽS.D.. Normal
SZ
AD
105.9 Ž21.5. 304.9 Ž57.8.
92.1 Ž17.0. 344.5 Ž43.8.
Ancova F, P
Scheffe ´ S-tests, P-values Nl vs. SZ
Nl vs. AD
SZ vs. AD
1.47, 0.24 3.60, 0.05
0.47 0.42
0.93 0.06
0.26 0.33
2
Nissl-stained neurons (rmm ) MD 95.3 Ž19.5. Caudate 273.2 Ž46.3.
PHF-tau neurofibrillary tangles (rmm2 ) MD 0.08 Ž0.17. 0.16 Ž0.43. Caudate 2.42 Ž3.11. 2.19 Ž1.33.
11.28 Ž10.36. 3.39 Ž3.10.
11.65, 0.0002 0.41, 0.67
0.99 0.98
0.001 0.78
0.001 0.65
PHF-tau neurofibrillary tanglesr neuron MD 0.00 Ž0.00. 0.00 Ž0.00. Caudate 0.01 Ž0.01. 0.01 Ž0.00.
0.13 Ž0.13. 0.01 Ž0.01.
9.54, 0.0007 0.29, 0.74
0.99 0.93
0.004 0.92
0.003 0.74
2.03, 0.15 0.78, 0.47
0.92 0.71
0.36 0.81
0.18 0.38
2.07, 0.14 0.78, 0.48
0.84 0.93
0.43 0.69
0.16 0.46
2.53, 0.09 0.03, 0.97
0.20 0.99
0.11 0.97
0.95 0.97
2.38, 0.11 1.15, 0.34
0.37 0.60
0.10 0.41
0.74 0.85
GFAP astrocytes (rmm2 ) MD 18.2 Ž19.2. Caudate 16.1 Ž17.7. GFAP astrocytesr neuron MD 0.22 Ž0.27. Caudate 0.06 Ž0.06. CD68 microglia (rmm2 ) MD 50.3 Ž41.8. Caudate 44.7 Ž16.8. CD68 microgliar neuron MD 0.51 Ž0.39. Caudate 0.17 Ž0.08.
13.0 Ž12.6. 24.2 Ž23.9.
0.13 Ž0.15. 0.07 Ž0.08.
80.8 Ž30. 4. 44.6 Ž22.5.
0.83 Ž0.46. 0.14 Ž0.07.
36.6 Ž44.2. 8.9 Ž8.5.
0.41 Ž0.48. 0.03 Ž0.02.
85.7 Ž41.5. 41.6 Ž21.8.
0.99 Ž0.61. 0.12 Ž0.05.
SZ, schizophrenia; AD, Alzheimer’s disease; Nl, Normal a Ratios given per neuron are based on the density of Nissl-stained neurons.
108
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
4. Discussion This study examined the possibility of neurodegeneration and neural injury in two important subcortical nuclei in a sample of elderly, pooroutcome subjects with schizophrenia by assessing the densities of neurons, PHF-tau-ir NFTs, GFAP-ir astrocytes, and CD68-ir microglia. There were no differences between patients with schizophrenia and control patients without neuropsychiatric disease in either of the regions on any of the four measures. These data indicate that neurodegenerative disease-related or other processes of neural injury beyond those observed with normal aging in these subcortical nuclei do not play a role in this population’s highly deteriorated state. These findings are similar to those of our previously reported investigation of neurodegeneration and neural injury in the hippocampus and several neocortical regions ŽArnold et al., 1998.. Some investigators have suggested that a higher incidence of AD in poor-outcome schizophrenia patients might explain their high frequency of dementia ŽBuhl and Bojsen-Moller, 1988; Prohovnik et al., 1993; Arnold and Trojanowski, 1996., but our data do not support this. AD patients exhibited a higher density of NFTs in the MD nucleus than did schizophrenia subjects with or without dementia, and there was no difference between normal and schizophrenia patients on this measure. This finding is consistent with most of the more recent studies showing an absence of AD-related pathology in elderly patients with schizophrenia ŽBruton et al., 1990; Powchik et al., 1993; Purohit et al., 1993; Arnold et al., 1994, 1998; Harrison, 1999.. Postmortem studies of the MD nucleus in schizophrenia have been controversial, with some researchers finding decreased neuron number, density, and volume, and others finding no change ŽHeckers, 1997; Jones, 1997.. Pakkenberg has reported a marginally significant decrease in neuron density, a 40% reduction in total neuron number, and a 25% reduction in the volume of the mediodorsal nucleus ŽPakkenberg, 1990.. We found a slightly higher mean neuron density in
schizophrenia, although this was not significant. Other studies of the thalamus have reported normal volumes for all major subnuclei but decreased thickness of periventricular gray ŽLesch and Bogerts, 1984. and decreased parvalbuminimmunoreactive neurons in the anteroventral nucleus ŽDanos et al., 1998.. Neither our study nor others’ findings indicate that there is abnormal neurodegeneration in schizophrenia during aging. If there were neurodegeneration-related neuron loss, one might expect the ratio of astrocytes to neurons to increase. However, there was no difference in this ratio between schizophrenia and normal groups. While there has been disagreement in the past, consensus is building that there is no noteworthy astrocytosis in schizophrenia ŽArnold and Trojanowski, 1996; Harrison, 1999.. Early postmortem studies using traditional histological stains Že.g. Holzer. reported evidence of fibrillary gliosis in periventricular areas ŽStevens, 1982; Bruton et al., 1990. while other, more recent studies using Nissl stains and GFAP immunohistochemistry have failed to find astrocytosis. Similar to our findings in the cerebral cortex, we now report no differences using GFAP immunohistochemistry in MD and caudate. GFAP expression in response to neural injury is dynamic ŽNorton et al., 1992. and indicates the state of astrocyte reactivity for a variable time prior to death, but does not rule out a previous astrocyte reaction to earlier injury. It is still possible that remote neural injury could leave fibrillary astrocytic changes that would not be accompanied by persistent up-regulation of GFAP. Our data do not support the hypothesis that abnormal neurodegenerative processes or neural injury are evident in poor-outcome schizophrenia patients compared with normal elderly controls. However, it is still possible that neurodevelopmental abnormalities in important cognition-related areas of the brain, including the MD and caudate, may make these areas especially vulnerable to the effects of normal aging, such as the small accumulations of neurodegenerative lesions that typically occur. Thus, an interaction between neurodevelopmental abnormalities and age-
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
related neurodegenerative changes could manifest in the clinical deterioration evident in so many of these poor outcome patients.
Acknowledgements This work was supported by NIH grants MH43880 and MH55199. The authors thank members of the Penn Schizophrenia Mental Health Clinical Research Center and the Department of Pathology and Laboratory Medicine, and the patients and families whose generosity made this research possible. References Abercrombie, M., 1946. Estimation of nuclear populations from microtome sections. Anatomic Record 94, 239]247. Alheid, G.F., Heimer, L., Switzer, R.C., 1990. Basal ganglia. In: Paxinos, G. ŽEd.., The Human Nervous System. Academic Press, San Diego. Arai, H., Lee, V.M.-Y., Messinger, M.L., Greenberg, B.D., Lowery, D.E., Trojanowski, J.Q., 1991. Expression patterns of beta-amyloid precursor protein Žbeta-APP. in neural and nonneural human tissues from Alzheimer’s disease and control subjects. Annals of Neurology 30, 686]693. Armstrong, E., 1990. Limbic thalamus: anterior and mediodorsal nuclei. In: Paxinos, G. ŽEd.., The Human Nervous System. Academic Press, San Diego. Arnold, S.E., Trojanowski, J.Q., 1996. Recent advances in defining the neuropathology of schizophrenia. Acta Neuropathologica 92, 217]231. Arnold, S.E., Hyman, B.T., Flory, J., Damasio, A.R., Van Hoesen, G.W., 1991. The topographical and neuroanatomical distribution of neurofibrillary tangles and neuritic plaques in the cerebral cortex of patients with Alzheimer’s disease. Cerebral Cortex 1, 103]116. Arnold, S.E., Franz, B.R., Trojanowski, J.Q., 1994. Elderly patients with schizophrenia exhibit infrequent neurodegenerative lesions. Neurobiology of Aging 15 Ž3., 299]303. Arnold, S.E., Gur, R.E., Shapiro, R.M., Fisher, K.R., Moberg, P.J., Gibney, M.R., Gur, R.C., Blackwell, P., Trojanowski, J.Q., 1995. Prospective clinicopathological studies of schizophrenia: accrual and assessment. American Journal of Psychiatry 152, 731]737. Arnold, S.E., Trojanowski, J.Q., Gur, R.E., Han, L.-Y., Choi, C., 1998. Absence of neurodegeneration and neural injury in the cerebral cortex in a sample of elderly patients with schizophrenia. Archives of General Psychiatry 55, 225]232. Arriagada, P.V., Growdon, J.H., Hedley-Whyte, E.T., Hyman, B.T., 1992. Neurofibrillary tangles but not senile plaques parallel duration and severity of Alzheimer’s disease. Neurology 42, 631]639.
109
Braak, H., Braak, E., 1991. Alzheimer’s disease affects limbic nuclei of the thalamus. Acta Neuropathologica 81 Ž3., 261]268. Bruton, C.J., Crow, T.J., Frith, C.D., Johnstone, E.C., Owens, D.G.C., Roberts, G.W., 1990. Schizophrenia and the brain: a prospective clinico-neuropathological study. Psychological Medicine 20 Ž2., 285]304. Buhl, L., Bojsen-Moller, M., 1988. Frequency of Alzheimer’s disease in a postmortem study of psychiatric patients. Danish Medical Bulletin 35, 288]290. Carpenter, W.T., Kirkpatrick, B., 1988. The heterogeneity of the long-term course of schizophrenia. Schizophrenia Bulletin 14 Ž4., 515]542. Clarke, P.G.H., 1992. How inaccurate is the Abercrombie correction factor for cell counts. Trends Neurosci 15, 211]212. Danos, P., Baumann, B., Bernstein, H.G., Franz, M., Stauch, R., Northoff, G., Krell, D., Falkai, P., Bogerts, B., 1998. Schizophrenia and anteroventral thalamic nucleus: selective decrease of parvalbumin-immunoreactive thalamocortical projection neurons. Psychiatry Research: Neuroimaging 82 Ž1., 1]10. Davidson, M., Harvey, P.D., Powchik, P., Parella, M., White, L., Knobler, H.Y., Losonczy, M.F., Keefe, R.S.E., Katz, S., Frecska, E., 1995. Severity of symptoms in chronically institutionalized geriatric schizophrenic patients. American Journal of Psychiatry 152, 197]207. Duchen, L.W., 1992. General pathology of neurons and neuroglia. In: Adams, J.H., Duchen, L.W. ŽEds.., Greenfield’s Neuropathology. Oxford University Press, New York. Greenberg, S.G., Davies, P., 1990. A preparation of Alzheimer paired helical filaments that displays distinct tau proteins by polyacrylamide gel electrophoresis. Proceedings of the National Academy of Sciences ŽUSA. 87, 5827]5831. Harrison, P.J., 1997. Schizophrenia: a disorder of neurodevelopment? Current Opinion in Neurobiology 7, 285]289. Harrison, P.J., 1999. The neuropathology of schizophrenia. A critical review of the data and their interpretation. Brain 122, 593]624. Heckers, S., 1997. Neuropathology of schizophrenia: cortex, thalamus, basal ganglia and neurotransmitter-specific projection systems. Schizophrenia Bulletin 23, 403]421. Hyman, B.T., Gomez-Isla, T., Irizarry, M.C., 1998. Stereology: a practical primer for neuropathology. Journal of Neuropathology and Experimental Neurology 57 Ž4., 305]310. Jones, E.G., 1997. Cortical development and thalamic pathology in schizophrenia wsee commentsx. Schizophrenia Bulletin 23 Ž3., 483]501. Lesch, A., Bogerts, B., 1984. The diencephalon in schizophrenia: evidence for reduced thickness of the periventricular grey matter. European Archives of Psychiatry and Neurological Sciences 234 Ž4., 212]219. Lowe, J., Cox, G., 1990. Neuropathological techniques. In: Bancroft, J.D., Stevens, A. ŽEds.., Theory and Practice of Histological Techniques. Churchill Livingtstone, Edinburgh. Mega, M.S., Cummings, J.L., 1994. Frontal]subcortical cir-
110
E. Falke et al. r Psychiatry Research 93 (2000) 103]110
cuits and neuropsychiatric disorders. Journal of Neuropsychiatry and Clinical Neurosciences 6 Ž4., 358]370. Nakajima, K., Kohsaka, S., 1993. Functional roles of microglia in the brain. Neuroscience Research 17 Ž3., 187]203. Norton, W.T., Aquino, D.A., Hozumi, I., Chiu, F.-C., Brosnan, C.F., 1992. Quantitative aspects of reactive gliosis: a review. Neurochemical Research 17, 877]885. Pakkenberg, B., 1990. Pronounced reduction of total neuron number in mediodorsal thalamic nucleus and nucleus accumbens in schizophrenia. Archives of General Psychiatry 47, 1023]1028. Pantelis, C., Barnes, T.R., Nelson, H.E., 1992. Is the concept of frontal-subcortical dementia relevant to schizophrenia? British Journal of Psychiatry 160, 442]460. Partlow, G.D., delCarpio-O’Donovan, R., Melanson, D., Peters, T.M., 1992. Bilateral thalamic glioma: review of eight cases with personality change and mental deterioration. American Journal of Neuroradiology 13, 1225]1230. Paskavitz, J.F., Lippa, C.F., Hamos, J.E., Pulaski-Salo, D., Drachman, D.A., 1995. Role of the dorsomedial nucleus of the thalamus in Alzheimer’s disease. Journal of Geriatrics, Psychiatry and Neurology 8 Ž1., 32]37. Powchik, P., Davidson, M., Nemeroff, C.B., Haroutunian, V., Purohit, D.P., Losonczy, M., Bissette, G., Perl, D., Ghanbari, H., Miller, B., Davis, K.L., 1993. Alzheimer’s-diseaserelated protein in geriatric schizophrenic patients with cog-
nitive impairment. American Journal of Psychiatry 150 Ž11., 1726]1727. Prohovnik, I., Dwork, A.J., Kaufman, M.A., Wilson, N., 1993. Alzheimer-type neuropathology in elderly schizophrenia patients. Schizophrenia Bulletin 19, 805]816. Purohit, D.P., Davidson, M., Perl, D.P., Powchik, P., Haroutunian, V.H., Bierer, L.M., McCrystal, J., Losonczy, M., Davis, K.L., 1993. Severe cognitive impairment in elderly schizophrenic patients: a clinicopathological study. Biological Psychiatry 33, 255]260. Savage, C.R., 1997. Neuropsychology of subcortical dementias. Psychiatric Clinics of North America 20, 911]931. Stevens, J.R., 1982. Neuropathology of schizophrenia. Archives of General Psychiatry 39, 1131]1139. Trojanowski, J.Q., Gordon, D., Obrocka, M., Lee, V.M.-Y., 1984. The developmental expression of neurofilament and glial filament proteins in the human pituitary gland: an immunohistochemical study with monoclonal antibodies. Developmental Brain Research 13, 229]239. West, M.J., 1993. New stereological methods for counting neurons. Neurobiology of Aging 14, 287]293. Winokur, G., Pfohl, B., Tsuang, M., 1987. A 40-year follow-up of hebephrenic-catatonic schizophrenia. In: Miller, N., Cohen, G. ŽEds.., Schizophrenia and Aging. Guilford Press, New York.