Abstract
Objective: To present the clinical, neuroimaging, and electrophysiologic characteristics of a variant AD phenotype.
Background: The authors have identified a large Finnish kindred with presenile dementia and spastic paraparesis due to deletion of exon 9 of presenilin 1. Neuropathologic analysis showed unusual cortical “cotton wool” plaques, immunoreactive for the beta-amyloid peptide but lacking congophilic cores.
Patients and Methods: Twenty-two affected individuals (16 men and 6 women) were identified in four successive generations. All surviving five patients were examined and subjected to molecular genetic analysis. In addition, the neurologic records of nine deceased patients were evaluated. Electrophysiologic investigations were available in eight cases. CT or MRI of the head had been performed on 11 patients and PET was performed on three patients. Result:— The mean age at onset (±SD) was 50.9 ± 5.2 years (range 40 to 61 years). Memory impairment was present in all patients. Memory impairment appeared simultaneously with or was preceded by walking difficulty due to spasticity of the lower extremities (10/14). Impaired fine coordination of hands (9/14) and dysarthria (6/14) in some patients suggested cerebellar involvement. EEG showed intermittent generalized delta-theta activity. Head MRI showed temporal and hippocampal atrophy; PET showed bilateral temporo-parietal hypometabolism.
Conclusion: Spastic paraparesis or brisk stretch reflexes of lower extremities or clumsiness of hands combined with dementia suggests this variant of AD.
Patients with early-onset familial AD (FAD) comprise about 5 to 10% of all individuals with AD. FAD segregates as an autosomal dominant trait and is genetically heterogeneous.1,2 Three genes have been shown to be involved in the etiology: presenilin 1 (PS-1) gene on chromosome 14,3-5 presenilin 2 (PS-2) gene on chromosome 1,6,7 and the amyloid precursor protein (APP) gene on chromosome 21.8 About 50 different mutations of the PS-1 gene, the majority of which are missense mutations, have been identified in more than 80 families of different ethnic origins.2,9 Estimates of the proportion of early-onset FAD due to mutations in the PS-1 gene vary from 18 to 70%,9-11 whereas pedigrees with mutations of PS-2 and APP genes constitute 1% and 5%, respectively.9,12 Consequently, defects in these three genes are not responsible for AD in all early-onset families,9 and additional FAD genes must exist.
In contrast to this marked genetic heterogeneity, there is little evidence of phenotypic heterogeneity in FAD, apart from variation in the age at onset.1,5,13 Early progressive aphasia, myoclonus, generalized seizures, and paratonia are apparently more common in PS-1–encoded FAD than in patients with APP (codon 717) mutation.14 In addition, there are a few reports on phenotypic heterogeneity between different mutations of the same gene. Mutations of the APP gene that lie on either end of the β-peptide (Aβ) sequence cause a typical presenile FAD, whereas a mutation of codon 693 within the Aβ sequence leads to hereditary cerebral hemorrhage of the Dutch type.15 Headache was reported to be a distinctive symptom in FAD due to the E280A mutation of the PS-1 gene,16 whereas early and prominent myoclonus mimicking Creutzfeldt-Jakob disease was found to be associated with the M146V mutation of the same gene.5,17
We recently reported a variant form of FAD, due to deletion of exon 9 of PS-1, with unusual cortical “cotton wool” plaques at neuropathologic analysis.18 We now describe the distinctive clinical phenotype of this FAD in which presenile dementia was combined with spastic paraparesis. We also report morphologic and functional imaging and neurophysiologic studies.
Patients and methods.
Patients.
The pedigree (figure 1) was ascertained through Patient III-18, who had paraparesis and memory impairment. The oldest known affected member of this family was born in 1860 in southern Finland, and his offspring were traced from parish registers. Twenty-two affected subjects were identified in four generations. However, 17 patients had died by 1997. All surviving five patients were personally examined (table 1) and subjected to molecular genetic analysis. Medical and neurologic records were available from nine deceased patients, including reports on neuropathologic autopsies in three cases.18 Two more patients died during the study and neuropathologic autopsy was performed. A total of nine neuropsychologic test results were obtained. Results of electrophysiologic investigation were available in nine cases and CSF investigations in nine cases (table 2).
Figure 1. Pedigree of the family with variant AD with spastic paraparesis. Individuals in the fourth generation, most of whom have not reached the age at onset of the disease, are omitted, except for one who is affected.
Clinical features and apolipoprotein E genotype in variant AD with spastic paraparesis
Laboratory, neuropsychologic, and electrophysiologic findings in variant AD with spastic paraparesis
Neuroimaging.
Head CT had been performed on eight patients and head MRI (1.5-T) on three patients. Myelography had been carried out on three patients with spastic paraparesis, and spinal MRI (1.0 to 1.5-T) on four patients. Results of SPECT were available for four patients. PET was performed on three patients.
PET imaging.
PET was performed using a GE (Milwaukee, WI) Advance scanner giving 35 transaxial planes at 4.3-mm intervals. 2-[18F]-fluoro-2-deoxy-d-glucose (FDG) was used as a ligand. FDG was prepared as described.19,20 The radiochemical purity exceeded 99% and the specific radioactivity at the time of injection was about 75 GBq/μmol. A dynamic 55-minute study was performed after IV injection of a bolus of 3.7 MBq/kg of FDG. Arterialized blood samples were drawn from the antecubital vein to measure plasma radioactivity concentration and glucose level.
Neuropathologic studies.
The brains and spinal cords of Patients III-14 and III-15 were fixed in 4% phosphate-buffered formaldehyde. Representative tissue samples were embedded in paraffin, and stained with hematoxylin-eosin, Luxol Fast Blue–cresyl violet, and modified Bielschowsky stains. Amyloid was identified after Congo red staining by red–green dichroism in polarized light and by fluorescence after thioflavin S staining. Selected specimens were studied by standard immunoperoxidase methods18 for the presence of Aβ peptide and for hyperphosphorylated tau protein.
Molecular genetic analysis.
Molecular genetic investigations, including analysis of lymphoblast-derived mRNA, were performed as described previously.18 In addition, the apolipoprotein E (apoE) genotypes of the five living patients were determined by standard PCR and restriction fragment length polymorphism (RFLP) methods.21 The following markers were used in the linkage analysis: D14S277, S268, S77, 2 cM gap, S71, 2 cM gap, S43, S273, S284, 12 cM gap, S256. The lod score was calculated using MLINK (http://linkage.rockefeller.edu/soft/list.html#m) set at a mutant allele frequency of 0.01, a disease frequency of 0.01, and fully penetrant by age 60 years.
Statistical analysis.
The level of significance of the differences between the mean ages at onset and ages at death of affected men and women were analyzed using the ANOVA test with SPSS/PC (Cary, NC) program.
Results.
Pedigree analysis.
Although it was previously thought that there were two separate Finnish families18 we have subsequently found that they descended from the same ancestor born in 1860. The distribution of affected individuals within the pedigree was compatible with an autosomal dominant mode of inheritance with full penetrance. Among the 22 affected subjects, the male/female ratio (16/6) was 2.7. This skewed ratio may be due to the fact that, by chance, there were more men than women both in the second (8/3) and third (14/9) generations.
Clinical characteristics.
Memory impairment was the main neurologic sign shared by all affected subjects. The patients had difficulties in immediate and delayed recall of recently presented material. In addition, all patients had other severe neuropsychological deficits such as impairment of spatial skills, apraxia, acalculia, and aphasia. Dementia was diagnosed by the age of 40 to 61 years (see table 1). Only three patients (two personally examined) of 14 maintained some insight into their cognitive decline.
In addition to dementia, the most characteristic clinical feature was gait disorder due to spastic paraparesis. It was often the first symptom and preceded dementia. Ten patients (two personally examined) had spastic paraparesis verified by medical examination. In addition, at least five deceased patients had had a history of spastic paraparesis. Neurologic examination of patients with gait impairment showed hyperreflexia or clonus of lower extremities. The muscle tone of both lower extremities was increased, and in most cases the Babinski response was abnormal bilaterally. However, paraparesis was not invariably present, dementia being the only symptom in four patients. Three of these patients were personally examined and two of these showed brisk tendon reflexes in the lower extremities. In general, sensory modalities of the trunk and extremities remained intact. Six patients with spastic paraparesis had had lower back pain about 1.5 to 5 years before gait disorder. Two patients had retired because of severe chronic back pain before the manifestation of other symptoms.
Nine patients (five personally examined) had clumsiness of hands. Neurologic examination showed that there was hyperreflexia in 10 (4 personally examined) and spasticity of upper extremities in 3 patients. Most patients had impaired fine coordination of hands. In addition, they had dysdiadochokinesia, intention tremor, and dysmetria of hands (see table 1). Three patients (two personally examined) had gaze palsy resembling that of supranuclear pathway type. In six patients, speech became dysarthric. Seven patients (four personally examined) had motor dysphasia that progressed to aphasia. Epileptic seizures of grand mal type were seen in four patients 5 to 10 years after the onset of the first symptoms.
Course.
The age at onset of symptoms ranged from 40 to 61 years. The mean age at onset (±SD) was 50.9 ± 5.2 years for all patients, 50.7 ± 5.4 years for men, and 51.4 ± 5.5 years for women. All patients had dementia, preceded in most patients by approximately 5 years by spastic paraparesis, leading to impaired gait at age 45 to 60. The course of spastic paraparesis was progressive. Five patients out of 10 with spastic paraparesis became wheelchair-bound after 6 years on average (range 5 to 8 years), whereas three further patients had walked with crutches. In the late stage of the disease all patients were bedridden, unable to cooperate, mute, and incontinent. They developed flexion contractures, and some patients had lower facial weakness and difficulties in swallowing. The duration of the disease from onset of symptoms to death ranged from 5 to 12 years (mean 9 years). The mean age at death (±SD) was 60.7 years ± 5.8 for all patients, 62.3 ± 4.5 years for men, and 58 ± 7.4 years for women. Men had a higher mean age at death than women, but the difference was not statistically significant (p = 0.257). The most common immediate cause of death was bronchopneumonia.
Illustrative case report.
A 54-year-old man (III-14) experienced walking difficulties and occasionally lower back pain. At neurologic examination, memory impairment typical of AD as well as spasticity of both legs were noted. The patient had hyperreflexia in upper extremities and clonus in lower extremities. The Babinski response was abnormal on both sides. The finger-nose and the heel-knee-shin tests were slightly inaccurate. The sensation of trunk and extremities was intact. Spastic paraparesis was diagnosed. The protein level of CSF was elevated at 589 mg/L (normal 150 to 450 mg/L). Head CT showed two small old infarcts on top of the right lateral ventricle and in the left precentral gyrus. MRI of the thoracolumbar region showed no sign of a spinal stenosis or other lesions that could cause paraparesis. The etiology of spastic paraparesis remained unknown. On neuropsychological examination, the patient was disoriented and had severe difficulties in immediate and delayed recall. Performance in visuoconstructional and spatial abilities had deteriorated. He also had difficulties in abstract reasoning and speech comprehension. The patient was unaware of his cognitive deficits. Two years later, the patient had acute cerebral hemorrhage in his left frontal lobe. After this his condition remained poor. He became bedridden, aphasic, and was permanently hospitalized for 5 years before death.
Laboratory and electrophysiologic investigations.
Five patients had increased protein concentration of CSF (see table 2). Four of them had spastic paraparesis. The cell count and glucose level of CSF remained normal in all cases. All eight EEG were abnormal, showing intermittent generalized slow delta-theta activity (3 to 4 Hz) that is frequently seen in the severe stage of AD. Nerve conduction and EMG studies indicated mild distal sensorimotor polyneuropathy of unknown etiology in two patients. Visual evoked potentials (VEP) were normal in the only tested patient, and somatosensory evoked potentials (SEP) (three patients) were normal with the exception of abnormal SEP in one patient with polyneuropathy.
Neuroimaging.
Head CT showed central and cortical atrophy. Some incidental old infarcts without clinical relevance were detected. Neither myelography nor spinal MRI (1.5-T) revealed any abnormalities. SPECT had remained normal. Three patients were subjected to both head MRI and FDG PET. All FDG PET scans demonstrated pronounced bilateral hypometabolism in the temporo-parietal areas (figure 2). Head MRI of Patient III-22, a 57-year-old woman, showed severe cortical atrophy especially in the temporal and parietal lobes. Furthermore, there was severe atrophy in both hippocampi. Head MRI of Patient III-23, a 54-year-old man, showed pronounced atrophy in the temporal, parietal, and frontal lobes. Head MRI of Patient IV-1, a 43-year-old man, showed mild atrophy of the temporal lobes, more pronounced on the left. There was mild atrophy in the right and moderate atrophy in the left hippocampus. All three patients also had signs of possible cerebellar origin but head MRI failed to show anything abnormal in the cerebellum.
Figure 2. FDG PET scan of a patient with variant AD shows bilateral hypometabolism in temporo-parietal areas.
Neuropathologic characteristics.
Patients III-14 and III-15 showed a profusion of cortical plaques, immunoreactive for Aβ, numerous neurofibrillary tangles, immunoreactive for hyperphosphorylated tau protein, and pronounced cerebral Aβ angiopathy. In addition, there was degeneration of the corticospinal tracts in the medulla oblongata and the spinal cord. The predominant plaques were of the “cotton wool” type: large, distinct, eosinophilic structures without a congophilic core and with only minor neuritic pathology (figure 3). Variable numbers of diffuse plaques but few plaques with amyloid cores were detected. These neuropathologic findings corresponded to those reported previously in Patients III-7, III-18, and III-21.18 Detailed neuropathologic description is the subject of a separate study.
Figure 3. Unusual “cotton wool” plaques in the temporal neocortex of Patient III-14, easily discernible in hematoxylin-eosin–stained sections as roundish eosinophilic structures (A). In Bielschowsky preparations (B), the plaques show minor neuritic changes, and lack dense immunoreactive amyloid cores when immunostained for the amyloid beta peptide (C). Paraffin sections, scale bar 100 μm.
Molecular genetic analysis.
The lod score at recombination fraction (theta) 0.00 was 2.25 (table 3). All five living patients had a deletion of exon 9 of PS-1, based on analysis of lymphoblast-derived mRNA. The deletion was not detectable in genomic DNA.18 ApoE genotypes (see table 1) did not explain the variation in the age at onset.
Linkage analysis (lod scores) between PS1 deletion and variant AD
Discussion.
The disease of our patients was inherited as an autosomal dominant trait, and their clinical picture was characterized by two salient features: spastic paraparesis and presenile dementia. This combination may occur in a number of familial neurologic conditions.
Hereditary spastic paraparesis is a heterogeneous condition with several forms showing autosomal dominant inheritance. Although one locus for spastic paraparesis has been mapped to chromosome 14q, it is at a long genetic distance from the location of the PS-1 gene,22 and dementia is a very rare feature of hereditary spastic paraparesis. Familial Creutzfeldt-Jakob disease is in most cases distinguished by a rapid course, myoclonus, and EEG that may show a typical pattern of periodic sharp waves or spikes.
The disease of our patients resembles familial British dementia,23 previously known as familial presenile dementia with spastic paralysis,24,25 or as familial cerebral amyloid angiopathy (British type).26,27 This autosomal dominant disease is characterized by progressive spastic tetraparesis, cerebellar signs, and dementia in the fifth decade.24-26 Neuropathologic findings include severe cerebral amyloid angiopathy, neurofibrillary tangles, and non-neuritic amyloid plaques of varying size.26,27 However, in contrast to the plaques of our patients, these plaques are not immunoreactive for beta-amyloid peptide but contain a 4K protein subunit named ABri. This highly insoluble peptide is a fragment of a putative type-II single-spanning transmembrane precursor that is encoded by a novel gene BRI located on chromosome 13.23
An early description of the unusual combination of dominantly inherited Alzheimer-type presenile dementia with spastic paralysis was published in 1940.28 An autopsied patient showed numerous neurofibrillary tangles as well as a multitude of various types of senile plaques not only in the cerebral but also in the cerebellar cortex. In addition, “drusenartige Entartung” of cerebral blood vessels as well as degeneration of the pyramidal tracts were observed. A 33-year-old Japanese woman with AD and unilateral spastic hemiparesis, cerebellar dysarthria, and ataxia was described in a previous article.29 Postmortem examination revealed severe AD changes, including cerebral cortical plaques with or without central amyloid cores, degeneration of the corticospinal tracts, and neuritic plaques in the cerebellum. Another study reported two Japanese sisters with presenile dementia, preceded by spastic tetraplegia.30 Postmortem examination showed degeneration of the corticospinal tracts in addition to numerous senile plaques and neurofibrillary tangles in the neocortex. “Cotton wool” plaques were not described. Molecular genetic data are not available on these Belgian and Japanese patients and their exact nosologic classification is not possible.
Recently, the combination of presenile AD type dementia and spastic paraparesis was described in three Australian families. However, neuropathologic features were not described in detail. One of these families carried a missense mutation (R278T) in exon eight and another a deletion of exon 9 (Δ290–319) of the PS-1 gene, due to a splice acceptor site mutation, whereas in the third pedigree no mutation was found.31 However, in this third pedigree, a mutation similar to ours cannot be excluded because our deletion is not detectable from genomic DNA.18 Furthermore, mutations causing splicing out of exon 9 have been identified in a Japanese and a British family.32,33 The Japanese patients were said to represent typical AD even though it was mentioned that they had bilateral “spastic paralysis with rigidity.”32
Most of our current patients carrying a deletion of exon 9 of the PS-1 gene had spastic paraparesis, and neuropathologic analysis showed degeneration of the corticospinal tracts at the level of the pyramids and the spinal cord. Numerous “cotton wool” plaques, neurofibrillary tangles, and amyloid angiopathy were also found in the precentral cortex that may have contributed to the spastic paraparesis.
It is becoming increasingly important to identify subtypes of AD clinically so that molecular genetic studies can focus on the right gene. The combination of spastic paraparesis and dementia should prompt molecular genetic studies of PS-1 gene mutations. All AD patients should be checked for gait, muscle tone, and tendon reflexes. When molecular genetic analysis is considered in the current type of mutation, it is important to realize that in genomic DNA only the normal allele over exon 9 will be amplified by PCR. Consequently, mRNA isolated from lymphoblasts and amplification of cDNA is needed.18,34 In the future, such analyses may be of therapeutic significance because treatment strategies may be different in different variants of AD.
Acknowledgments
Supported by the Academy of Finland (project 48173), Neurological Foundation, Uulo Arhio Foundation, Helsinki and Turku University Central Hospitals (EVO funding), and Päivikki and Sakari Sohlberg Foundation.
Acknowledgment
The authors thank Professor Aapo Lehtonen, University of Turku, and docent Aki Hietaharju, Tampere University Hospital, for patient information, and MSc Tove Grönroos and the personnel of the Turku PET Center for assistance in performing the PET scans. They also thank Dr. Timo Kurki for neuroradiologic consultation, Dr. Sari Rastas for statistical assistance, and registered nurse Raija Ahlfors for excellent secretarial work.
- Received June 28, 1999.
- Accepted November 13, 1999.
References
Disputes & Debates: Rapid online correspondence
NOTE: All authors' disclosures must be entered and current in our database before comments can be posted. Enter and update disclosures at http://submit.neurology.org. Exception: replies to comments concerning an article you originally authored do not require updated disclosures.
- Stay timely. Submit only on articles published within the last 8 weeks.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- 200 words maximum.
- 5 references maximum. Reference 1 must be the article on which you are commenting.
- 5 authors maximum. Exception: replies can include all original authors of the article.
- Submitted comments are subject to editing and editor review prior to posting.