Diagnostic criteria for adult-onset leukoencephalopathy with axonal spheroids and pigmented glia due to CSF1R mutation.

BACKGROUND AND PURPOSE: To establish and validate diagnostic criteria for adult-onset leukoencephalopathy with axonal spheroids and pigmented glia (ALSP) due to colony-stimulating factor 1 receptor (CSF1R) mutation. METHODS: We developed diagnostic criteria for ALSP based on a recent analysis of the clinical characteristics of ALSP. These criteria provide 'probable' and 'possible' designations for patients who do not have a genetic diagnosis. To verify its sensitivity and specificity, we retrospectively applied our criteria to 83 ALSP cases who had CSF1R mutations (24 of these were analyzed at our institutions and the others were identified from the literature), 53 cases who had CSF1R mutation-negative leukoencephalopathies and 32 cases who had cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) with NOTCH3 mutations. RESULTS: Among the CSF1R mutation-positive cases, 50 cases (60%) were diagnosed as 'probable' and 32 (39%) were diagnosed as 'possible,' leading to a sensitivity of 99% if calculated as a ratio of the combined number of cases who fulfilled 'probable' or 'possible' to the total number of cases. With regard to specificity, 22 cases (42%) with mutation-negative leukoencephalopathies and 28 (88%) with CADASIL were correctly excluded using these criteria. CONCLUSIONS: These diagnostic criteria are very sensitive for diagnosing ALSP with sufficient specificity for differentiation from CADASIL and moderate specificity for other leukoencephalopathies. Our results suggest that these criteria are useful for the clinical diagnosis of ALSP.

Clinical and genetic characterization of adult-onset leukoencephalopathy with axonal spheroids and pigmented glia associated with CSF1R mutation.

BACKGROUND AND PURPOSE: The clinical characteristics of colony stimulating factor 1 receptor (CSF1R) related adult-onset leukoencephalopathy with axonal spheroids and pigmented glia (ALSP) have been only partially elucidated. METHODS: Clinical data from CSF1R mutation carriers who had been seen at our institutions or reported elsewhere were collected and analysed using a specific investigation sheet to standardize the data. RESULTS: In all, 122 cases from 90 families with CSF1R mutations were identified. The mean age of onset was 43 years (range 18-78 years), the mean age at death was 53 years (range 23-84 years) and the mean disease duration was 6.8 years (range 1-29 years). Women had a significantly younger age of onset than men (40 vs. 47 years, P = 0.0006, 95% confidence interval 3.158-11.177). There was an age-dependent penetrance that was significantly different between the sexes (P = 0.0013). Motor dysfunctions were the most frequent initial symptom in women whose diseases began in their 20s. Thinning of the corpus callosum, abnormal signalling in pyramidal tracts, diffusion-restricted lesions and calcifications in the white matter were characteristic imaging findings of ALSP. The calcifications were more frequently reported in our case series than in the literature (54% vs. 3%). Seventy-nine per cent of the mutations were located in the distal part of the tyrosine kinase domain of CSF1R (102 cases). There were no apparent phenotype-genotype correlations. CONCLUSIONS: The characteristics of ALSP were clarified. The phenotype of ALSP caused by CSF1R mutations is affected by sex.

Effects of short-chain fatty acids on Actinomyces naeslundii biofilm formation.

Actinomyces naeslundii is an early colonizer and has important roles in the development of the oral biofilm. Short-chain fatty acids (SCFA) are secreted extracellularly as a product of metabolism by gram-negative anaerobes, e.g. Porphyromonas gingivalis and Fusobacterium nucleatum; and the SCFA may affect biofilm development with interaction between A. naeslundii and gram-negative bacteria. Our aim was to investigate the effects of SCFA on biofilm formation by A. naeslundii and to determine the mechanism. We used the biofilm formation assay in 96-well microtiter plates in tryptic soy broth without dextrose and with 0.25% sucrose using safranin stain of the biofilm monitoring 492 nm absorbance. To determine the mechanism by SCFA, the production of chaperones and stress-response proteins (GrpE and GroEL) in biofilm formation was examined using Western blot fluorescence activity with GrpE and GroEL antibodies. Adding butyric acid (6.25 mm) 0, 6 and 10 h after beginning culture significantly increased biofilm formation by A. naeslundii, and upregulation was observed at 16 h. Upregulation was also observed using appropriate concentrations of other SCFA. In the upregulated biofilm, production of GrpE and GroEL was higher where membrane-damaged or dead cells were also observed. The upregulated biofilm was significantly reduced by addition of anti-GroEL antibody. The data suggest biofilm formation by A. naeslundii was upregulated dependent on the production of stress proteins, and addition of SCFA increased membrane-damaged or dead cells. Production of GroEL may physically play an important role in biofilm development.

Autosomal recessive hereditary spastic paraplegia with thin corpus callosum: a novel mutation in the SPG11 gene and further evidence for genetic heterogeneity.

BACKGROUND AND PURPOSE: Autosomal Recessive Hereditary Spastic Paraplegia with Thin Corpus Callosum (AR-HSPTCC) is a clinically and genetically heterogeneous complicated form of spastic paraplegia. Two AR-HSPTCC loci have been assigned to chromosome 15q13-15 (SPG11) and chromosome 8p12-p11.21 respectively. Mutations in the SPG11 gene, encoding the spatacsin protein, have been found in the majority of SPG11 families. In this study, involvement of the SPG11 or 8p12-p11.21 loci was investigated in five Italian families, of which four consanguineous. METHODS: Families were tested for linkage to the SPG11 or 8p12-p11.21 loci and the SPG11 gene was screened in all the affected individuals. RESULTS: Linkage was excluded in the four consanguineous families. In the only SPG11-linked family the same homozygous haplotype 4.2 cM across the SPG11 locus was shared by all the three affected siblings. A novel c.2608A>G mutation predicted to affect the splicing was found in exon 14 of the SPG11 gene. DISCUSSION: This collection of families contributes to highlight the intra and inter locus heterogeneity in AR-HSPTCC, already remarked in previous reports. In particular, it confirms heterogeneity amongst Italian families and reports a new mutation predicted to affect splicing in the spatacsin gene.

Silver syndrome variant of hereditary spastic paraplegia: A locus to 4p and allelism with SPG4.

OBJECTIVE: To perform a clinical and genetic study of two large Italian families (RM-36 and RM-51) showing the cardinal clinical features of Silver syndrome (SS), a rare dominantly inherited form of hereditary spastic paraplegia (HSP) complicated by amyotrophy of the small hand muscles. METHODS: Clinical assessment including neurophysiologic, neuropsychological, and neuroimaging evaluations. Genetic studies included linkage and sequence analyses. RESULTS: Using a genome-wide survey in the RM-36 family, a novel locus (SPG38) has been identified and mapped within the 13.1-cM region on chromosome 4p16-p15 between markers D4S432 and D4S1599. The RM-51 family was linked to the SPG4 locus at 2p21-p24 and sequence analysis of SPG4 showed a novel frameshift mutation p.Asp321GlyfsX6. Clinical examination of the affected members carrying the mutation showed high frequency of additional clinical features including decreased vibration sense, pes cavus, temporal lobe epilepsy, and cognitive impairment. CONCLUSIONS: This study demonstrates evidence of a novel locus SPG38 for Silver syndrome (SS) and suggests that genetic defects in SPG4 might lead to broad clinical features overlapped with those of SS.

Heterogeneity within a large kindred with frontotemporal dementia: a novel progranulin mutation.

BACKGROUND: Frontotemporal dementia (FTD) in several 17q21-linked families was recently explained by truncating mutations in the progranulin gene (GRN). OBJECTIVE: To determine the frequency of GRN mutations in a cohort of Caucasian patients with FTD without mutations in known FTD genes. METHODS: GRN was sequenced in a series of 78 independent FTD patients including 23 familial subjects. A different Calabrian dataset (109 normal control subjects and 96 FTD patients) was used to establish the frequency of the GRN mutation. RESULTS: A novel truncating GRN mutation (c.1145insA) was detected in a proband of an extended consanguineous Calabrian kindred. Segregation analysis of 70 family members revealed 19 heterozygous mutation carriers including 9 patients affected by FTD. The absence of homozygous carriers in a highly consanguineous kindred may indicate that the loss of both GRN alleles might lead to embryonic lethality. An extremely variable age at onset in the mutation carriers (more than five decades apart) is not explained by APOE genotypes or the H1/H2 MAPT haplotypes. Intriguingly, the mutation was excluded in four FTD patients belonging to branches with an autosomal dominant mode of inheritance of FTD, suggesting that another novel FTD gene accounts for the disease in the phenocopies. It is difficult to clinically distinguish phenocopies from GRN mutation carriers, except that language in mutation carriers was more severely compromised. CONCLUSION: The current results imply further genetic heterogeneity of frontotemporal dementia, as we detected only one GRN-linked family (about 1%). The value of discovering large kindred includes the possibility of a longitudinal study of GRN mutation carriers.

The effects of APOE and tau gene variability on risk of frontotemporal dementia.

Frontotemporal dementia (FTD) is a complex dementing syndrome whose genetic/non genetic risk factors are mostly unknown. Aim of the present work was to investigate whether APOE and/or tau gene variability does affect the risk of FTD. A sample of FTD cases (sporadic: n = 54; familial: n = 46, one subject per family) was collected in a genetically homogeneous population (Calabria, southern Italy) and analyzed in comparison with an age- and sex-matched control group (n = 180) extracted from the same population. Logistic regression analysis showed that APOE gene variability affects the probability of disease, with allele epsilon4 increasing (exp(beta1) = 2.68 with [1.51-4.76] 95% confidence interval; p = 0.001) and allele epsilon2 decreasing (exp(beta1) = 0.28 with [0.12-0.66] 95% confidence interval; p = 0.003) the risk of FTD. On the contrary, tau gene variability was ineffectual (exp(beta1) non significantly different from 1 for either H1 or H2 haplotypes), although a small effect was observed by the H1 haplotype in increasing the protective effect of the epsilon2 allele (p = 0.007).

LRRK2 gene in Parkinson disease: mutation analysis and case control association study.

BACKGROUND: In addition to the four well-confirmed genes linked to early-onset Parkinson disease (PD) (SNCA, PARKIN, DJ-1, and PINK1), mutations in the leucine-rich repeat kinase 2 gene (LRRK2) have recently been identified in families with autosomal dominant late-onset PD. OBJECTIVE: To perform mutation analysis of LRRK2 in probands of families showing dominant inheritance of PD and to conduct a case control association study to test the hypothesis that common coding variations might be associated with increased susceptibility to PD. METHODS: All 51 LRRK2 coding exons were sequenced in 23 probands and the mutation frequencies were evaluated in 180 neurologically normal control subjects. For the association study the authors genotyped four coding LRRK2 polymorphisms in 250 normal control subjects and 121 patients with PD (predominantly white patients of Canadian origin), 84% of whom had age at onset before 50 years and 42% had a positive family history. RESULTS: The authors identified three probands with heterozygous LRRK2 mutations: two of them have the known G2019S substitution and one proband has a novel I1371V substitution. Mutation analysis of a large family demonstrated complete segregation of the G2019S with PD. However, there was no association between PD and any of the four polymorphisms at the allelic or genotypic levels (p > 0.17). Furthermore, the authors did not detect a modifying effect for any genotype or of APOE genotypes upon the age at onset in the PD group (p > 0.20). CONCLUSIONS: The results support the prior suggestion that LRRK2 mutations cause PD. The disease in the families reported here presents a phenotype indistinguishable from typical PD. All three families demonstrate a very variable age at onset that is not explained by APOE genotypes. The common coding variations in the LRRK2 gene neither constitute strong PD risk factors nor modify the age at onset; however, the possibility of a modest risk effect remains to be assessed in large datasets.

Familial Alzheimer disease: decreases in CSF Abeta42 levels precede cognitive decline.

CSF amyloid beta-peptide 42 (Abeta42) levels in presymptomatic subjects with pathogenic mutations in the PS1 gene are significantly lower than in an age-matched control group. Consequently, in these subjects, there is a window of opportunity estimated as at least 4 to 12 years to evaluate the ability of any putative prophylactic therapy to decrease, arrest, or reverse abnormalities in Abeta42 metabolism many years before clinical symptoms of Alzheimer disease are otherwise likely to occur.

[Familial fronto-temporal dementia with brain stem ubiquitin-positive neuronal inclusions]. / Démence fronto-temporale familiale avec inclusions marquées par l'anti-ubiquitine dans le tronc cérébral.

INTRODUCTION: Fronto-temporal dementias (FTD) were described a century ago on the macroscopic basis of frontal and/or temporal lobe atrophy. Progress in neuropathology, immunohistochemistry, biochemistry and genetics has since shown that they are heterogeneous entities, encompassing many different diseases with similar clinical presentations. A few, such as tauopathies due to mutations of the gene coding for tau protein (MAPtau form a well-defined group. Definition and grouping of other types of FTD is still problematic. MATERIAL AND METHOD: We studied a family where the mother and 4/8 children were affected with FTD. Clinical presentation was typical of FTD. Onset was ill-defined with early (at age 40 years or less) personality changes. The clinical course was protracted (about 30 years). For a long period, the patients were able to live in the community in spite of obvious signs such as hyperorality and loss of verbal initiative; operative orientation as to place was preserved for a long time: a mute patient was still able to drive. Signs of extrapyramidal or motoneuron involvement were not observed. RESULTS: The genetic study failed to detect any mutation in MAPtau; the lod score for flanking markers was positive but not significant. Biochemical study showed no qualitative abnormality in tau protein. Neuropathological study of one affected subject showed brain atrophy (962 g), with elective frontal lobe involvement. Cortical nerve cell loss was more marked in superficial layers and in frontal areas; glia was inconspicuous; pseudolaminar spongiosis was present in the more severely affected zones. No argentophilic "Pick bodies" were seen; ubiquitin-positive, tau-negative round inclusions were present in the cytoplasm of fascia dentata neurones. "Tangles" were mostly restricted to the entorhinal cortex, partly correlated with tau immunoreactivity, but better with ubiquitin immunoreactivity. Large, ovoid or reniform, moderately dense, spongy, granular or filamentous argentophilic cytoplasmic nerve cell inclusions were observed. They were ubiquitin-positive, but did not react with other antibodies, particularly anti-tau. They were present in swollen nerve cells in the deeper cortical layers but were most conspicuous in the brain stem: in the magnocellular reticular nuclei (e.g. nucleus centralis pontis), in the pes pontis, in the inferior olive and in motor nuclei, especially in the trigeminal motor nucleus. They were not associated with nerve cell loss, atrophy nor pycnosis. Cerebellar relay nuclei neurones were swollen, and their cytoplasm contained argentophilic filaments. CONCLUSION: In our opinion, "ubiquitinopathy" would be non-specific and "Motor Neuron Disease-Inclusion Dementia" (MNDID) would not be satisfactory as a diagnosis for the present cases of FTD. Hopefully, progress in genetics may allow a causal, and thence definitive, classification.

Fine mapping of 10q and 18q for familial Alzheimer's disease in Caribbean Hispanics.

Familial Alzheimer's disease (AD [MIM 104300]) has been a focus of intense investigation, primarily in Caucasian families from Europe and North America families. Although the late-onset form of familial AD, beginning after age 65 years, has been linked to regions on chromosomes 10q and 12p, the specific genetic variants have not yet been consistently identified. Using a unique cohort of families of Caribbean Hispanics ancestry, we screened the genome using 340 markers on 490 family members from 96 families with predominantly late-onset AD. We observed the strongest support for linkage on 18q (LOD=3.14). However, 17 additional markers (chromosomes 1-6, 8, 10, 12, and 14) exceeded a two-point LOD score of 1.0 under the affecteds-only autosomal dominant model or affected sibpair model. As we previously reported the fine-mapping effort on 12p showing modest evidence of linkage, we focused our fine-mapping efforts on two other candidate regions in the current report, namely 10q and 18q. We added 31 family members and eight additional Caribbean Hispanic families to fine map 10q and 18q. With additional microsatellite markers, the evidence for linkage for 18q strengthened near 112 cM, where the two-point LOD score for D18S541 was 3.37 and the highest NPL score in that region was 3.65 (P=0.000177). This narrow region contains a small number of genes expressed in the brain. However, at 10q (134-138 cM), the NPL score decreased from 3.15 (P=0.000486) to 2.1 (P=0.0218), but two broad peaks remained overlapping with previously reported peaks. Our results provide modest support for linkage on 10q and 12p in this cohort of Caribbean Hispanic families with familial Alzheimer's disease, and strong evidence for a new locus on 18q.

A new SPG4 mutation in a variant form of spastic paraplegia with congenital arachnoid cysts.

The clinical and genetic findings are described for 16 patients from a large Italian family with a variant form of hereditary spastic paraplegia and congenital arachnoid cysts inherited as an autosomal dominant trait. A molecular study has revealed a novel missense mutation, T614I, in exon 17 of SPG4, which may play a role in both focal cortical dysgenesis and neurodegeneration of the motor neurons in the corticospinal tract.

Pleasant feeling from watching a comical video enhances free radical-scavenging capacity in human whole saliva.

OBJECTIVE: Free radicals have been implicated in aging, mutagenesis, inflammation and other pathological conditions. We conducted a study to clarify the relation between a pleasant feeling as a psychological eustress and free radical-scavenging capacity (FRSC) in saliva. METHODS: Saliva was collected from 27 healthy volunteers. FRSC before, during and after watching a cheerful comical video for 30 min was measured by using 1,1-diphenyl-2-picrylhydrazyl (DPPH). RESULTS: The median values of FRSC (micromol/ml) before, at 10 and 20 min during and after watching the video were 54.5, 66.8, 66.6 and 69.4, respectively. The FRSC values obtained after watching the video were significantly higher than those before watching it (P<.001). When the FRSC before watching was taken as 1, the value for the group that felt "Very good," "Good" or "Ordinary+Dull" while watching the video for 30 min was 1.38, 1.20 or 0.98, respectively (P<.01). CONCLUSION: Watching a comical video enhanced the FRSC in saliva; in addition, a pleasant feeling boosted it even more.

Effect of high pressure gaseous carbon dioxide on the germination of bacterial spores.

Effect of high pressure gaseous carbon dioxide treatment (HGCT) at 6.5 MPa, 35 degrees C on the germination of bacterial spores was investigated. Germination of bacterial spores was estimated by the decrease of heat tolerance. Approximately, 40% of Bacillus coagulans and 70% of Bacillus licheniformis were germinated by HGCT for 120 min at 35 degrees C, respectively. Germination was confirmed by phase contrast microscopy. The effect of hydrostatic pressure treatment (HPT) at 6.5 MPa, 35 degrees C on the germination of B. coagulans and B. licheniformis spores were also investigated. Spores did not germinate by HPT alone at 6.5 MPa for 120 min.

SulA-independent filamentation of Escherichia coli during growth after release from high hydrostatic pressure treatment.

To improve the efficiency of sterilization by high hydrostatic pressure treatment (HPT), it is desirable to know the biochemical process of bacteria most sensitive to the treatment. We investigated growth properties after release from HPT of exponentially growing Escherichia coli K-12 cells. We observed growth retardation after treatment (30 min at 37 degrees C) above 75 MPa. Long filamentous cells of about eight times normal cell length were observed at 90 min growth after treatment at 75 MPa. In the subsequent period the filamentous cells divided into normal-sized cells. recA and sulA mutant strains also formed filamentous cells, indicating that filamentation was SulA-independent. Nucleoids segregated normally in the filamentous cells. Only one FtsZ ring (or none) was detected at possible division sites in the elongated cells. Western blotting analysis demonstrated that the amount of FtsZ protein was not affected by the treatment. GTP-dependent in vitro polymerization of either FtsZ protein in E. coli crude extract or purified FtsZ protein, however, was sensitive to HPT. These facts suggest that HPT at 75 MPa denatures a fraction of FtsZ molecules, and that these denatured molecules interfere with the polymerization of functional FtsZ, resulting in the significantly reduced number of FtsZ rings.

PS1 Alzheimer's disease family with spastic paraplegia: the search for a gene modifier.

PS1 mutations are associated with classic Alzheimer's disease (AD); however, some families develop AD and spastic paraplegia (SP) with brain pathology characterized by Abeta cotton wool plaques. The authors report a variant AD family with the E280Q PS1 mutation. The fact that the same PS1 mutation can be found in patients with either variant or classic AD argues in favor of the presence of a genetic modifier. The authors have excluded that this modifier effect originates from coding sequence variations in three SP genes or from a second mutation in the other AD genes.

Clinical and genetic study of a large Italian family linked to SPG12 locus.

BACKGROUND: Seven loci for autosomal dominant hereditary spastic paraplegia (ADHSP) have been mapped. To date, two families of SPG12 (chromosome 19q13) have been analyzed; however, there is not enough clinical information on SPG12 to establish locus-phenotype correlations. METHODS: The authors studied 60 individuals from a large Italian family with ADHSP, in which 16 members in four generations were affected. They performed genetic linkage analysis with DNA markers from currently known ADHSP loci. After database searching, one candidate gene for SPG12 was analyzed by sequencing. RESULTS: The patients in this family showed an early onset and rapid progression of symptoms, resulting in severe disability, with a large proportion of affected members requiring use of a wheelchair. By age 16, most patients had sensory disturbance. Evidence for linkage to the SPG12 locus was obtained. Obligate recombination events observed in this family have narrowed the SPG12 locus from the 16.1 cM to 11.3 cM region between markers D19S416 and D19S412. In combination with previous genetic studies, the SPG12 locus was further narrowed to the 3.3 cM region between D19S416 and D19S220. A homologue of the AAA (ATPases associated with a variety of cellular activities) protein family, proteasome 26S subunit ATPase mapped near D19S220, was excluded by sequencing. CONCLUSIONS: This study refined the SPG12 region between D19S416 and D19S220 and revealed several clinical characteristics-early onset, rapid progression, and involvement of sensory disturbance-that may be unique to SPG12. Suggestive evidence of genetic anticipation was obtained, but should be confirmed in other SPG12 families.

A large Calabrian kindred segregating frontotemporal dementia.

Frontotemporal dementia (FTD) displays significant neuropathological and genetic heterogeneity among and within affected families. An early diagnosis is often difficult because cognitive symptoms are manifest only at a late stage of the disease. We have been studying a large pedigree segregating frontotemporal dementia (FTD) to which belong 34 identified affected persons, 11 of whom were personally examined. The kindred has been genealogically reconstructed; all FTD patients have been linked to the same ancestors who lived in the early 18(th) century (11 generations before the present one). Autosomal dominant transmission was evident. Clinical features were uniform within the kindred and met the Lund-Manchester criteria. Personality changes with absence of insight, lack of empathy and of social awareness manifested up to 5 years before medical advice was sought. Loss of fluency was the earliest neuropsychological sign, in the absence of memory, orientation and praxis deficits, which evolved late, together with hyperorality. Akinesia was observed early, rigidity appeared late, tremor was absent. Two patients showed myoclonus late in their evolution. No ALS signs were observed in this kindred. Mutations of the MAPt gene, coding for the Tau protein, were not detected in affected family members. Linkage studies excluded chromosomes 3 and 9 and gave indeterminate results that were model dependent for chromosome 17.

A founder mutation in presenilin 1 causing early-onset Alzheimer disease in unrelated Caribbean Hispanic families.

CONTEXT: Genetic determinants of Alzheimer disease (AD) have not been comprehensively examined in Caribbean Hispanics, a population in the United States in whom the frequency of AD is higher compared with non-Hispanic whites. OBJECTIVE: To identify variant alleles in genes related to familial early-onset AD among Caribbean Hispanics. DESIGN AND SETTING: Family-based case series conducted in 1998-2001 at an AD research center in New York, NY, and clinics in the Dominican Republic. PATIENTS: Among 206 Caribbean Hispanic families with 2 or more living members with AD who were identified, 19 (9.2%) had at least 1 individual with onset of AD before the age of 55 years. MAIN OUTCOME MEASURE: The entire coding region of the presenilin 1 gene and exons 16 and 17 of the amyloid precursor protein gene were sequenced in probands from the 19 families and their living relatives. RESULTS: A G-to-C nucleotide change resulting in a glycine-alanine amino acid substitution at codon 206 (Gly206Ala) in exon 7 of presenilin 1 was observed in 23 individuals from 8 (42%) of the 19 families. A Caribbean Hispanic individual with the Gly206Ala mutation and early-onset familial disease was also found by sequencing the corresponding genes of 319 unrelated individuals in New York City. The Gly206Ala mutation was not found in public genetic databases but was reported in 5 individuals from 4 Hispanic families with AD referred for genetic testing. None of the members of these families were related to one another, yet all carriers of the Gly206Ala mutation tested shared a variant allele at 2 nearby microsatellite polymorphisms, indicating a common ancestor. No mutations were found in the amyloid precursor protein gene. CONCLUSIONS: The Gly206Ala mutation was found in 8 of 19 unrelated Caribbean Hispanic families with early-onset familial AD. This genetic change may be a prevalent cause of early-onset familial AD in the Caribbean Hispanic population.