|Home | About | Journals | Submit | Contact Us | Français|
The frequency and clinical and pathological characteristics associated with the Gly206Ala presenilin 1 (PSEN1) mutation in Puerto Rican and non-Puerto Rican Hispanics were evaluated at the University of Pennsylvania’s Alzheimer’s Disease Center. DNAs from all cohort subjects were genotyped for the Gly206Ala PSEN1 mutation. Carriers and non-carriers with neurodegenerative disease dementias were compared for demographic, clinical, psychometric, and biomarker variables. Nineteen (12.6%) of 151 unrelated subjects with dementia were discovered to carry the PSEN1 Gly 206Ala mutation. Microsatellite marker genotyping determined a common ancestral haplotype for all carriers. Carriers were all of Puerto Rican heritage with significantly younger age of onset, but otherwise were clinically and neuropsychologically comparable to those of non-carriers with AD. Three subjects had extensive topographic and biochemical biomarker assessments that were also typical of non-carriers with AD. Neuropathological examination in one subject revealed severe, widespread plaque and tangle pathology without other meaningful disease lesions. The PSEN1 Gly206Ala mutation is notably frequent in unrelated Puerto Rican immigrants with dementia in Philadelphia. Considered together with the increased prevalence and mortality of AD reported in Puerto Rico, these high rates may reflect hereditary risk concentrated in the island which warrants further study.
Some studies of Hispanics (from various countries of origin) in the U.S. have described a significantly younger age of onset of Alzheimer’s disease (AD) dementia, greater cognitive impairment at the time of presentation, and relatively lower frequency of apolipoprotein E ε4 compared to other ethnoracial groups [1, 2]. Explanations for these features include genetic mutations and polymorphisms associated with early onset and familial AD [3, 4], recruitment and/or self-selection bias among the patients who agreed to research assessment, language and cultural bias in assessment measures, lower education and socioeconomic status conferring reduced resilience to the clinical expression of AD, and greater psychiatric morbidity .
As we began to address these factors, we discovered a high frequency of the previously described [5, 6] Gly206Ala (glycine to alanine amino acid substitution at codon 206 in exon 7) mutation in the presenilin 1 gene (PSEN1) among Puerto Rican Hispanics with neurodegenerative dementias assessed at our Alzheimer’s Disease Center (ADC). Here we describe this frequency as well as clinical, biomarker, and neuropathological features in patients with the mutation and compare these to Hispanic patients without known mutations.
Informed consent for collection and use of clinical, psychometric, genetic, and biomarker data was obtained from all persons, in accord with University IRB-approved protocols. Genotyping for the PSEN1 Gly206Ala mutation was conducted in all 283 self-identified Hispanic unrelated subjects for whom DNA was banked. The ADC conducts outreach recruitment in several primary care settings including North Philadelphia, a neighborhood with the largest concentration of Hispanics, chiefly of Puerto Rican heritage, in the city . Assessments included histories, physical and neurological examinations conducted by clinicians experienced in the evaluation of dementia, and review of neuroimaging and laboratory data. On the basis of these data, a consensus diagnosis was established using standardized clinical criteria. Some subjects were further recruited into neuroimaging and biofluid biomarker studies and for autopsy in the event of death.
All clinical and biomarker data presented here were collected proximal to the initial evaluation at our ADC.
Variables included age at first visit, gender, self-reported race and ethnic affiliation, years of formal education, age of onset of cognitive impairment, interval between symptom onset and first visit, and documented history of diabetes mellitus, hypertension, hyperlipidemia, and/or cardiovascular disease (including cerebrovascular or peripheral vascular disease).
The Mini-Mental State Examination  was used as a summary measure of cognition in all subjects and most also had more extensive neuropsychological data available as well which included the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) cognitive battery for which a total score was calculated . The Dementia Severity Rating Scale [10, 11], an informant based rating scale for dementia severity, was obtained in all subjects as well. All tests were administered in the participants’ preferred language, most commonly Spanish.
Cerebrospinal fluid had been collected in 40 subjects in their initial evaluations. Amyloid-β, total tau, and phospho181-tau were measured in the University of Pennsylvania Alzheimer’s Disease Neuroimaging Initiative (ADNI) Biomarker Core laboratory using INNO-BIA AlzBio3, xMAP technology immunoassay (Innogenetics) on the Luminex analytical platform, as previously described [14, 15]. Volumetric measurements of brain magnetic resonance imaging had been conducted in 44 subjects. These were used to calculate the SPARE-AD (Spatial Pattern of Abnormalities for Recognition of Early AD), an established volumetric pattern classification index that has demonstrated good sensitivity and specificity for differential diagnosis of AD from other neurodegenerative dementias, as well as prediction of conversion of normal cognition to mild cognitive impairment (MCI) and MCI to AD [16–18]. 18F-FDG PET had been performed in 47 subjects according to the ADNI protocol  and as previously described .
DNA was extracted from peripheral blood following the manufacturer’s protocols (Flexigene (Qiagen) or QuickGene DNA whole blood kit L (Autogen). After initially identifying two proband cases with the Gly206Ala mutation by DNA sequencing of the coding region of PSEN1, the entire cohort was screened for the targeted mutation, along with APOE genotype, using real-time allelic discrimination with Applied Biosystem (ABI) TaqMan probes. The corresponding ABI assay by design product numbers are: APOE (rs7412, C_904973_10 and rs429358, C_3084793_20) and PSEN1 Gly206Ala (custom assay AHAA2BH). Genotyping was performed on an ABI 7500 real-time instrument using standard conditions. Data were analyzed using ABI 7500 Software v2.0.l. To assess for a common founder, 3 microsatellite markers ((CA)13 (near rs57281649), (GT)18 (near rsl48098788), rs112183078 (DS14S53)) were examined by fragment analysis on an ABI 3130 as described [5, 21] in 19 cases with Gly206Ala and 54 Hispanic individuals lacking the mutation.
One subject with the PSEN1 Gly206Ala died and had autopsy with full gross and microscopic diagnostic neuropathological examination as previously described [22, 23] with routine stains and immuno-histochemistry for amyloid-β, PHFtau, α-synuclein, ubiquitin, and TDP-43.
Descriptive statistics were generated for demographic, clinical, and biomarker data in PSEN1 Gly206Ala carriers and non-carriers with dementia, as well as normal control Hispanic participants. We examined differences between groups in continuous variables using t-tests or Wilcoxon rank sum tests (for variables with small sample size or non-normality), and χ2 tests were used for categorical variables. All statistical tests were two-sided. Data were analyzed with the JMP9 (SAS, Cary, NC). For the haplotype analysis, we used PHASE 2.1 to phase haplotypes of the four markers: microsatellite CA repeat (physical location = 73571770 bp on NC_000014.8), microsatellite GT repeat (73581862 bp), rs63750082 (Gly206Ala mutation, 73659420bp), and rs112183078 (microsatellite, D14S53, 76922382 bp) . Default parameters were used (100 iterations for burn-in followed by 100 additional iterations, thinning interval = 1).
Of the 283 Hispanic subjects (72.6% of Puerto Rican heritage, 27.4% from 16 other countries of origin), for whom DNA had been collected, 20 were found to carry the PSEN1 Gly206Ala mutation. Nineteen of these were diagnosed with dementia (17 AD, 1 vascular dementia, 1 frontotemporal dementia), while one was diagnosed as having normal cognitive function at age 58 but had major depression and cognitive complaints (and was subsequently lost to follow-up). Of the 263 non-carriers, 111 were diagnosed with AD, 19 with other neurodegenerative dementias, 56 with MCI, 33 with miscellaneous other non-neurodegenerativc disease conditions affecting cognition (depression, head trauma, etc.), and 44 were normal control recruits.
Demographic, clinical, and biomarker data for the PSEN1 Gly206Ala dementia subjects and non-carrier AD subjects are compared in Table 1 along with data from normal control subjects for contrast. An illustrative case is presented in Fig. 1. Compared to non-carriers with AD, PSEN1 Gly206Ala carriers had younger age of onset, lower frequency of APOE ε4 genotype, and lower rates of hypertension. The mean age of onset for carriers was 59.6 (range 46–67 with one outlier whose onset was 81), ten years younger than that of non-carriers whose mean age of onset was 69.2 (range=45–90). Among the 19 PSEN1 carriers, only 3/19 (16%) were also apolipoprotein E ε4 carriers and their ages of onset were 46 (ε4/ε4), 59 (ε3/ε4), and 65 (ε3/ε4). In contrast, 41 % of PSEN1 non-carriers were apolipoprotein E ε4 carriers and their mean age of onset was 68.8 (range 47–94).
Analyses of domain-specific psychometric test performances in the CERAD battery revealed no differences between carriers and non-carriers with AD in memory, language, visuoconstructional abilities, speed of processing or executive functions, except for marginally worse word-list recall in the PSEN1 Gly206Ala carriers (t[1,94] = 1.98, p<0.05). No differences were observed in rates of head injury, alcohol/other substance abuse, or use of medications including antidepressants, antipsychotics, anxiolytics, cholinesterase inhibitors, memantine, or statins.
Quantitative AD biochemical biomarkers were obtained in three carriers and 13 non-carriers. All three carriers exhibited abnormally elevated total tau and phospho-tau levels and two of the three had abnormally low amyloid-β levels. SPARE-AD, an index of AD-pattern cortical atrophy measured with volumetric MRI, was abnormally elevated in all three. 18FDG-PET was abnormal in all three with a characteristic temporo-parietal hypometabolism pattern. One carrier died with autopsy, which confirmed the diagnosis of AD with severe plaque and tangle pathology, as portrayed in Fig. 1.
Genotyping of 3 microsatellite markers was performed in 19 cases with Gly206Ala and 54 individuals lacking the mutation to determine whether there may have been a common ancestor of the Gly206Ala mutation by haplotype analysis. The following alleles were identified among the samples tested: 11 to 27 CA repeats, 15 to 25 GT repeats, and 9 to 24 repeats at D14S53. The CA(19) (allele frequency (AF) = 0.55), GT(22) (AF=0.5), and D14S53 14 repeat (AF=0.31) alleles were the most common among the mutation cases. PHASE 2.1 analysis for microsatellite markers segregating with the Gly206Ala identified the haplotype containing the Gly206Ala mutation: (CA)19, and (GT)22. This was present in all Gly206Ala carriers, while none of the non-carriers had both (CA)19 and (GT)22 alleles at the same time and hence did not carry the (CA)19-(GT)22 haplotype. This supports the hypothesis that all Gly206Ala mutations originated from the same common ancestral haplotype inherited by Puerto Rican Gly206Ala carriers .
Autosomal dominant mutations account for less than 1% of all AD, but 40–50% of people with early onset familial AD. Mutations in PSEN1 on chromosome 14 are by far the most common and at least 185 mutations in 405 families have been documented to date (http://www.molgen.ua.ac.be/ADMutations). Penetrance is nearly 100% and phenotypic expression is heterogeneous, ranging from more common amnesia-predominant progressive AD to rarer frontotemporal dementia-like symptoms, psychosis, spasticity, spastic paraparesis, myoclonus, seizures, ataxia, and/or extrapyramidal signs . Neuropathologically, most reports have documented typical plaque and tangle pathology of AD, although Lewy bodies have been described in some families  and cotton wool plaques in others .
The Gly206Ala missense mutation was first described in 2001 in two reports on individuals from 18 families of Caribbean Hispanic ethnicity (all Puerto Rico except one family from Dominican Republic) with early onset AD [5. 6]. While the families were unrelated, polymorphic microsatellite markers flanking PSEN1 demonstrated a common founder. The description of clinical phenotype in these reports was scant and to our knowledge, only one subsequent case report  on the PSEN1 GJy206Ala mutation has been published describing a single Puerto Rican man presenting with new onset persecutory delusions at age 58, subsequent recognition of progressive cognitive deficits and abnormal cerebrospinal fluid amyloid-β and tau levels consistent with AD, and distinctive neuroleptic hypersensitivity with severe extrapyramidal symptoms. However, a complicated past medical history (head trauma, syphilis, hepatitis C with interferon therapy, alcohol abuse, sarcoidosis, combat-related PTSD) confounds attribution of neuropsychiatric symptoms to the mutation. In our cohort, neuropsychiatric or neurological symptoms were not more common in PSEN1 Gly206Ala carriers than in the non-carriers with dementia, although we have noted greater depression in Hispanics overall than in other ethnic groups of our ADC cohort .
The Gly206Ala mutation was identified in ~13% of Hispanic individuals, all from Puerto Rico, with a neurodegenerative dementia at our ADC, a surprisingly high rate. Of these, 17/111 (15%) meeting strict criteria for AD were carriers. One case with the mutation was diagnosed clinically with vascular dementia and another was diagnosed with frontotemporal dementia, but many cases with both of these diagnoses have been shown to have AD or mixed pathology if autopsied [29, 30]. Although the mutation is from a common founder as determined by haplotype analysis, none of the carriers were known to be related.
We previously reported a younger age of onset of dementia in Hispanics compared with other ethnoracial groups. While this single mutation may explain some of this observation, we note that the mean age of onset of 69.0 in our non-carrier Hispanic subjects with AD is still younger than white non-Hispanics (mean 72.4) or African Americans (73.5) .
While the PSEN1 Gly206Ala mutation’s origin is not known, all cases derive from a common founder and have Puerto Rican heritage. Originally settled by the Taino Native Americans, Puerto Rico was colonized by Spaniards in 1508, following which the indigenous Taino population was severely diminished by disease, warfare, and enslavement. West Africans were brought to the island as slaves, but the total island census population was about 2,500 in 1530 and there was little new immigration until the late 19th century’s major influx of Spaniards from Europe and former colonies in South America as well as African slaves. Recent mitochondrial and autosomal ancestry studies have found most Puerto Ricans have highly admixed European, West African, and Native American DNA .
How representative our findings in Philadelphia are for non-emigrant Puerto Ricans (or those in other immigration destinations in the mainland U.S.) is not clear, given various selection biases that may confound our convenience sample. Nonetheless, some Puerto Rican health statistics find significantly greater prevalence of dementia among Puerto Rican veterans  and an almost two-fold greater mortality due to AD in Puerto Rico compared to the U.S. . Thus, given the relative genetic isolation of Puerto Rico for much of its recorded history and observations in emigrant Puerto Ricans in Philadelphia and elsewhere in the northeastern U.S., it is possible that these high rates may reflect notable hereditary risk concentrated in the island, including the PSEN1 Gly206Ala mutation, which warrant expanded study. As with the Colombia PSEN1 Glu280Ala kindred that is a focus of the Alzheimer Prevention Initiative , further epidemiological characterization of the PSEN1 Gly206Ala mutation in Puerto Rico may define an additional population facilitating analogous prevention and treatment studies.
The authors express appreciation to the research staff of the Penn Memory Center and above all to the volunteer participants in the University of Pennsylvania ADC. John Robinson conducted histology and microscopy. Supported by a grant from the NIH (AG 10124) and the Marian S. Ware Alzheimer Program.
Authors’ disclosures available online (http://www.j-alz.com/disclosures/view.php?id=1536).