A Novel Presenilin-1 Mutation (Leu85Pro) in Early-Onset Alzheimer Disease With Spastic Paraparesis FREE

Early-onset familial Alzheimer disease (AD) is caused by mutations in the amyloid precursor protein (APP), presenilin-1 (PSEN1), or presenilin-2 (PSEN2) genes. Phenotypic diversity has been reported to be associated with various mutations in PSEN1. Studies of early-onset AD with spastic paraparesis (SP) have previously identified 8 missense mutations, 2 deletions, and 1 insertion in PSEN1.^1- 12 It remains unclear how the mutation is involved in the pathological cascade, particularly in the production of amyloid-β peptide 42 (Aβ42).

In this article, we report a case exhibiting early-onset AD with SP accompanied by a novel PSEN1 mutation (Leu85Pro). Cells cultured to express PSEN1 with this mutation produced a 2-fold elevation of Aβ42 production and an increase in the Aβ42/40 ratio.

A 27-year-old Japanese man was admitted to our hospital for evaluation because of the difficulties he was experiencing in coping with his daily life. He had lived in the United States since the age of 12 years. He was a good basketball player and captain of his high school team. After he entered a university in California, he gradually became withdrawn and unmotivated. Eventually, he gave up basketball and other forms of exercise. Despite his condition, he finally managed to graduate from the university at the age of 25 years, taking 2 years longer than usual to complete his degree. At the age of 26 years, he had difficulty driving a car and sometimes became lost while driving. Around this time, he also became aware of hand tremor and memory impairment. By the age of 27 years, he found that he was unable to cope with living alone, and he returned to Japan. His parents noticed that he could not cook, use the telephone, or bathe himself. He did not have a history of alcohol or drug abuse. No other family members showed similar signs or symptoms.

On hospital admission, he was alert and oriented. Neurological examination showed that he was neither interested in his condition nor aware of its deterioration. He did not have aphasia, ideomotor apraxia, ideational apraxia, or dressing apraxia. Intellect and memory were relatively intact. Other neuropsychological tests showed that he had visual constructional apraxia, visuospatial agnosia, simultaneous agnosia, and optic ataxia. He became lost while driving because of topographic disorientation. His score on the Mini-Mental State Examination was 23 points. His total IQ score on the Wechsler Adult Intelligence Scale-Revised was 57. While his verbal IQ score was 57, his performance IQ score was below 40, reflecting visuospatial agnosia. Examination of the cranial nerves showed no significant abnormality. Deep tendon reflexes were hyperreactive, and the Babinski sign was bilaterally extensor. Muscle tone was spastic in both lower limbs. There was no weakness, sensory disturbance, or autonomic disturbance. He had an action tremor in his fingers but no apparent ataxia or myoclonus. His gait was markedly spastic, and he was unable to run. Laboratory test values, including analysis of blood and urine, were within normal limits.

An electroencephalogram showed intermittent generalized slow delta-theta activity. Nerve conduction, sensory evoked potential, and visual evoked potential were normal. Brain magnetic resonance imaging showed diffuse mild cortical atrophy by the standards for his age (Figure 1A). Technetium Tc 99m–hexamethyl-propylene amine oxine single-photon emission computed tomography showed bilateral hypoperfusion in the occipital and temporal lobes (Figure 1B). A positron emission tomography study using 2–fluorodopa F 18–fluoro-2-deoxy-D-glucose as ligand demonstrated bilateral hypometabolism in the occipital and temporal lobes (Figure 1C). Ophthalmologic examination revealed no abnormality of optic fundi, visual acuity, visual field, or color identification.

Genomic DNA was examined in the coding regions of the PSEN1, PSEN2, and APP genes, using an ABI PRISM model 310 sequencer (PerkinElmer, Fremont, Calif ). APOE genotypes were determined as described previously.¹³ All materials were obtained with informed consent after an appropriate consultation. This study received prior approval from the institutional ethics committee of Osaka City University Medical School, Osaka, Japan. To examine the effect of the novel mutation (described later) on Aβ production by APP, we compared 2 PSEN1 complimentary DNA constructs with or without the mutation. The mutant PSEN1 complimentary DNA (T254C) was prepared by site-directed mutagenesis and introduced into the pCI mammalian expression vector (Promega, Madison, Wis). Wild-type and mutant PSEN1 complimentary DNA were cotransfected with mutant APP (Swedish) complimentary DNA into human embryonic kidney 293 (HEK293) cells by lipofection (Lipofectamine; Life Technologies, Gaithersburg, Md). Two days after transfection, conditioned media were collected and assayed for Aβ40 and Aβ42 by enzyme-linked immunosorbent assay (BioSource International, Camarillo, Calif ). To examine the protein expression level of each transfected cell, cells were lysed in 1% Triton-X100 in tris-buffered saline containing protease inhibitors (Sigma, St Louis, Mo) and centrifuged at 10 000 g for 15 minutes at 4°C. The supernatant was subjected to Western blotting for APP and actin and also immunoprecipitated with anti-PSEN1 C-terminal fragment antibody¹⁴ (a rabbit polyclonal antibody for the C-terminal fragment of PSEN1) for Western blotting of PSEN1. The samples were electrophoresed on NuPAGE 4% to 12% Bis-Tris gels (Invitrogen, Purchase, NY) and transferred onto polyvinylidene difluoride membranes. The membrane was blocked and subsequently incubated in the primary antibody solution (a rabbit polyclonal antibody that recognizes the C-terminus of APP, a mouse monoclonal antibody that recognizes the C-terminal fragment of PSEN1, or an antiactin antibody purchased from Sigma). Bound antibody was visualized using horseradish peroxidase–conjugated secondary antibody and ECL Plus (Amersham Pharmacia Biotech Inc, Piscataway, NJ). Amyloid-β40 and Aβ42 in cerebrospinal fluid and serum of the patient and the control were assayed by enzyme-linked immunosorbent assay as described earlier.

Sequence analysis showed that the patient had a novel PSEN1 mutation in exon 4, at nucleotide position 254 (T to C) in 1 PSEN1 allele, indicating an amino acid change from leucine to proline at position 85 (Leu85Pro) (Figure 2A). There was no other mutation detected in the coding regions of PSEN1, PSEN2, or APP. No other family members, including the patient’s parents and 2 siblings, had any mutations in PSEN1, PSEN2, or APP, although the other genetic analyses were not performed. All of the family members, including the patient, exhibited APOE allele ε3/3. The mean ± SD ratio of Aβ42/Aβ40 in control cerebrospinal fluid was 0.12 ± 0.02 (n = 5), and the ratio in our patient was reduced to 35.4% of the control. The ratio of Aβ42/Aβ40 in the blood of our patient increased to 143.0% of the control. These results are consistent with previous reports of Aβ levels in cerebrospinal fluid and blood in patients with AD and familial AD.^15- 17

Western blotting was used to show that APP was expressed equally in wild-type PSEN1, mutant PSEN1, and in mock-transfected cells (Figure 2B, upper and lower panel). Expression of full-length PSEN1 (so-called PSEN1 holoprotein) was detected in mutant PSEN1 cells and in wild-type PSEN1 cells but was not seen in mock-transfected cells because endogenous PSEN1 is rapidly degraded to its N-terminal fragment and C-terminal fragment¹⁸ (Figure 2B, middle panel). Secretion of Aβ42 in mutant PSEN1 cells was twice that of wild-type PSEN1 cells, whereas secretion of Aβ40 did not differ significantly. As a result, the ratio of Aβ42/Aβ40 in mutant PSEN1 cells was more than twice that of wild-type PSEN1 cells. The mean ± SD secretion values of Aβ40 and Aβ42 from cells transfected with wild-type and mutant PSEN1 were as follows: wild-type PSEN1, Aβ40, 4594 ± 110 pg/mL, Aβ42, 444 ± 14 pg/mL, and Aβ42/40 ratio, 9.7% ± 0.1%; mutant PSEN1, Aβ40, 4914 ± 312 pg/mL, Aβ42, 896 ± 48 pg/mL (P<.001 vs wild-type PSEN1), and Aβ42/40 ratio, 18.2% ± 0.6% (P<.001 vs wild-type PSEN1). The conditioned media were examined for Aβ concentration by enzyme-linked immunosorbent assay.

We found a novel heterozygous mutation in PSEN1 (T254C, resulting in Leu85Pro within the first transmembrane domain) of a patient with very early-onset AD with SP. Because the mutation detected in this case resulted in a marked increase in Aβ42 production and in the Aβ42/40 ratio, it is almost certainly causative. Several previous studies have shown other PSEN1 mutations to be causative in familial AD with SP. Only 1 mutation is associated with SP in the same domain.³ Unfortunately, we could not examine the paternity, because the family members refused consent to test it. The spontaneous mutation rates in the disease range from 10⁻⁴ to 10⁻⁷ per locus per generation,¹⁹ although the rates would vary according to their gene sizes. Ifthe mutation of our case were de novo, this would be a rarity.

Neuropsychological evaluation of the patient revealed a complex visual problem in addition to impairment of intelligence and memory. Single-photon emission computed tomography and positron emission tomography examinations showed bilateral hypoperfusion and hypometabolism in the occipital and temporal lobes. These findings are compatible with the diagnosis of the visual variant of AD, as previously described by Levine et al,²⁰ rather than typical AD. Although neuropathological confirmation was not available in this patient, his symptoms and neuroimaging data were consistent with the visual variant form of AD. We are able to exclude other neurological diseases leading to juvenile dementia. The present case is noteworthy because the patient was younger than other patients with visual variant AD, and this is the first report, to our knowledge, of early-onset AD with SP and with the visual variant form of the disease. The relationship between the Leu85Pro PSEN1 mutation and the unique visuospatial cognitive disorder requires further study.

Correspondence: Hiroshi Mori, PhD, Department of Neuroscience, Osaka City University Medical School, 1-4-3 Asahimachi, Abenoku, Osaka 545-8585, Japan (mori@med.osaka-cu.ac.jp).

Accepted for Publication: October 7, 2003.

Author Contributions:Study concept and design: Tomiyama and Mori. Acquisition of data: Ataka, Tomiyama, Yamashita, Shimada, Tsutada, Kawabata, and Miki. Analysis and interpretation of data: Tomiyama and Takuma. Drafting of the manuscript: Ataka, Takuma, Yamashita, Shimada, Tsutada, and Kawabata. Critical revision of the manuscript for important intellectual content: Tomiyama, Mori, and Miki. Statistical analysis: Kawabata. Obtained funding: Mori. Administrative, technical, and material support: Ataka, Tomiyama, Takuma, and Mori. Study supervision: Mori and Miki.

Funding/Support: This study was supported by a grant-in-aid for Scientific Research on Priority Areas (C)-Advanced Brain Science Project from the Ministry of Education, Culture, Sports, Science, and Technology, Tokyo, Japan, and by grants-in-aid from the Uehara Memorial Foundation and the Nakatomi Foundation, Tokyo, Japan.

Acknowledgment: We thank Akiko Fukushima, BA, and Rie Teraoka, BA, for technical assistance.