Catálogo de publicaciones - libros

The phenotype of neurodegenerative diseases is defined by the distribution of cellular changes, which in turn depends on vulnerability to specific abnormalities of protein metabolism. Distinctive features have been identified that assist in the recognition of specific diseases or specific types of abnormalities of protein metabolism that are shared by several neurodegenerative disorders. Identification of the earliest manifestations of the phenotype will facilitate early implementation of therapy and improve understanding of protein metabolism abnormalities and their secondary consequences. This will assist in drug development and effective therapeutics for neurodegenerative disorders.

Pedigrees with familial Alzheimer's disease (FAD), caused by mutations in either the amyloid precursor protein (APP) or the presenilin 1 (PS1) or presenilin 2 (PS2) genes, show considerable phenotypic variability. Monogenic diseases typically exhibit variations in biological features, such as age of onset, severity, and multiple clinical and cellular phenotypes. This variation can be due to specific alleles of the disease gene, environmental effects, or modifier genes.

Spastic paraparesis (SP), or progressive spasticity of the lower limbs, is frequently hereditary, with over 20 loci being identified for uncomplicated (paraparesis alone) and complicated (paraparesis and other neurological features) disease subtypes. Moreover, over 10 different genes have been identified with mutations that lead to SP. While dementia is a common feature of complicated SP, a reciprocal observation has also been made since the earliest clinical reports of FAD: namely, that a number of AD families have been reported in which some individuals have SP. In 1997, the key observation was made that PS1 mutations were associated with the presence of SP, suggesting that there was a complex relationship between SP and AD. In addition, in 1998, it was also shown that PS1 AD/SP pedigrees frequently have variant, large, non-cored plaques without neuritic dystrophy, named cotton wool plaques (CWP). The PS1 mutations associated with CWP secrete unusually high levels of the amyloid β 42 peptide, suggesting a molecular basis for the formation of this distinctive plaque type.

The SP phenotype in PS1 pedigrees appears to be associated in some cases with a delayed onset of dementia, compared with affected individuals who present with dementia only. Some individuals who present with SP have remained dementia-free for up to 10 years. Variations seen in neuropathology and neurological symptoms in PS1 FAD suggest that modifier genes may underlie this phenotypic heterogeneity. As PS1 mutations are almost always associated with a particularly aggressive form of presenile dementia, these findings suggest the existence of a protective factor in some individuals with SP.

To provide better care for patients, physicians most often seek to determine the cause of symptoms by recognizing a clinical phenotype and then relating that phenotype to known disease mechanisms, including genetic variations and pathology. This approach has been particularly difficult with frontotemporal dementias (FTD), because of the diversity and continuous evolution of its behavioral, language and cognitive symptoms. Recent systematic clinical-pathological observations have resulted in proposed clinical criteria for FTD, but accurate diagnosis remains challenging and, in clinical practice, the phenotype of FTD can easily be confused with other disorders.

Positron emission tomography with [F]fluorodeoxyglucose (FDG-PET) can aid in the recognition of clinical phenotype and explore disease mechanisms. We reviewed the results of FDG-PET in 19 patients who subsequently had FTD confirmed at postmortem examination, including one with a known tau gene mutation. A comparison of scans from normal elderly controls with scans of patients with pathologically confirmed Alzheimer's disease (AD) demonstrates that FTD causes a distinctive pattern of hypometabolism, with considerable individual variability within this general pattern. FTD consistently causes predominant hypometabolism in frontal association, anterior cingulate, and anterior temporal regions, but the involvement of each region is variable and can be either symmetric or asymmetric. As the illness progresses, glucose hypometabolism becomes more pervasive, extending into regions characteristically affected early in AD. Although FDG-PET abnormalities accurately reflect clinical phenotype, neither pathological diagnosis nor genotype reliably predicts the variations observed in the pattern of glucose hypometabolism in FTD.

As people worldwide live to an older age, dementia, whose best-known risk factor is aging, has become a serious growing public health problem. While Alzheimer's disease (AD) is the most frequent cause of dementia, other progressive neurodegenerative disorders can be responsible for it. For these reasons, they are grouped as non-AD dementias. There is evidence that reactive oxygen species-mediated reactions, particularly of neuronal lipids, are present in brains affected by both types of dementing disorders. Traditional views have claimed that oxidative-mediated brain injury in these diseases is merely the result of the neurodegenerative processes. While numerous investigations have shown that oxidative stress is increased in AD, conflicting results exist for the heterogeneous group of non-AD dementias. The availability of specific and sensitive markers to monitor in vivo oxidative stress, in combination with studies performed in living patients, are helping us to elucidate these issues. This paper summarizes some of the most recent research on the relevance of oxidative stress and lipid peroxidation in AD and non-AD dementias. The evidence accumulated so far clearly indicates that oxidative stress is an early and specificaspect of AD pathogenesis but not of the pathogenesis of other dementias. This new concept implies that this phenomenon is not a general and common pathway of the neurodegenerative process, but it may play a more specific and important role in AD than in non-AD dementias.

In patients with familial Alzheimer's disease (FAD) with early onset, mutations have been identified in three genes: Presenilin 1 (PS1), Presenilin 2 (PS2) and amyloid protein precursor (APP). Although many different mutations have been recorded for each gene, in most cases the clinical picture is typical of AD, with earlier onset than in sporadic AD. The question of whether mutations in these genes invariably predict AD pathology is the subject of this review. Almost all mutations in APP, PS1 and PS2 promote the development of neuropathological findings of AD, with deposition of amyloid beta protein (Aβ) to form prominent amyloid pathology. Progress has been made in correlating the effects of mutations on species of Aβ with vascular versus parenchymal deposition of Aβ. There are some exceptions to what appears to be an otherwise inevitable relationship. First, a few innocent mutations in APP and PS1 have been described. Second, in rare families with PS1 mutations, there are mutation-bearing individuals who may have escaped clinical disease. Third, some APP mutations alter the sequence of Aβ and lead to severe angiopathy with hemorrhage, rather than the plaques and tangles of AD. Finally, some families with fronto-temporal dementia (FTD) have shown PS1 mutations. Although neuropathologic studies are limited at present, the FTD association raises questions about whether amyloidogenic disease pathways are the only mechanisms that lead to neuropathological changes in early onset AD with presinilin (PS) mutations.

We have identified more than 20 nuclear Alzheimer's disease (AD) families in Colombia that carry a single point mutation, E280A, in the presenilin 1 () gene. Genealogical and genetic studies have demonstrated that these families share a common founder more than 400 years ago. The mean age of onset of AD in these families is 45.2 years but the range is almost 30 years (35–62 years). The wide range in age of onset suggests that genetic and/or environmental risk factors modify the age of onset. We are using two complementary approaches to the identification of genetic modifiers: candidate gene analyses and a whole genome screen analysis.

We have genotyped polymorphisms in each of the known AD genes gene (APP), (1) gene, (2) gene and gene (). Several polymorphisms in these genes have previously been associated with risk for AD in some but not all studies. To determine whether these polymorphisms modify the age of onset in the Colombian kindreds, we initially used survival curve analysis. The only gene that modified age of onset with this methodology was . The allele was associated with an earlier age of onset whereas the presence of the 2 allele was associated with later age of onset. This result is consistent with the observations of the effect of alleles on sporadic AD and suggests that genes that influence the risk for late onset AD may also modify age of onset in familial early onset Alzheimer's disease (FAD).

We have also used survival analysis to examine several putative environmental risk factors. These studies demonstrated that both years of education and urban dwelling were associated with an earlier age of onset. Since these two factors were highly correlated in this study, it was not possible to determine which one was primarily the responsible risk factor.

To identify novel genetic risk factors for AD, we have used two complementary approaches: genetic analysis of a large series of late onset AD sibling pairs and a search for genes that modify age of onset in the Colombian families. Our genome screen in AD sibling pairs has provided evidence of AD susceptibility genes on chromosomes 9, 10, and 12 and suggestive evidence of an age of onset modifier gene on chromosome 12.

Pedigrees with familial Alzheimer's disease (FAD), caused by mutations in either the amyloid precursor protein (APP) or the presenilin 1 (PS1) or presenilin 2 (PS2) genes, show considerable phenotypic variability. Monogenic diseases typically exhibit variations in biological features, such as age of onset, severity, and multiple clinical and cellular phenotypes. This variation can be due to specific alleles of the disease gene, environmental effects, or modifier genes.

Spastic paraparesis (SP), or progressive spasticity of the lower limbs, is frequently hereditary, with over 20 loci being identified for uncomplicated (paraparesis alone) and complicated (paraparesis and other neurological features) disease subtypes. Moreover, over 10 different genes have been identified with mutations that lead to SP. While dementia is a common feature of complicated SP, a reciprocal observation has also been made since the earliest clinical reports of FAD: namely, that a number of AD families have been reported in which some individuals have SP. In 1997, the key observation was made that PS1 mutations were associated with the presence of SP, suggesting that there was a complex relationship between SP and AD. In addition, in 1998, it was also shown that PS1 AD/SP pedigrees frequently have variant, large, non-cored plaques without neuritic dystrophy, named cotton wool plaques (CWP). The PS1 mutations associated with CWP secrete unusually high levels of the amyloid β 42 peptide, suggesting a molecular basis for the formation of this distinctive plaque type.

The SP phenotype in PS1 pedigrees appears to be associated in some cases with a delayed onset of dementia, compared with affected individuals who present with dementia only. Some individuals who present with SP have remained dementia-free for up to 10 years. Variations seen in neuropathology and neurological symptoms in PS1 FAD suggest that modifier genes may underlie this phenotypic heterogeneity. As PS1 mutations are almost always associated with a particularly aggressive form of presenile dementia, these findings suggest the existence of a protective factor in some individuals with SP.

The vast majority of families with presenilin (PSEN) mutations have the clinical phenotype of Alzheimer's disease. However, there are reports of patients who carry PSEN mutations and have Alzheimer's disease with a variety of other clinical phenotypes including spastic paraplegia, seizures, myoclonus, parkinsonism, epilepsy and amyloid angiopathy. Remarkably, three of the studied families have frontotemporal dementia (FTD). The mutations associated with FTD are L113P (Raux et al. 2000), Ins R362 (Tang-Wai et al. 2002) and G183V (Dermaut et al. 2004). Raux and colleagues (2000) reported six members from four generations of the SAL family who had early onset FTD. Tang-Wai and colleagues (2002) reported three patients from three generations who had FTD in their 50s and 60s. Dermaut and colleagues (2004) reported a large family from two generations. Three were definitely affected and a total of 12 members were evaluated. Again the disease was of early onset. In all three families, the clinical phenotype was convincingly FTD in nature. In the first two families (L113P and Ins R362), no autopsy was available, but in the third family (G183V), one case had an autopsy and the pathology showed Pick's disease with Pick bodies and no Alzheimer pathology. Usually PSEN1 mutations enhance the γ-secretase effect on the amyloid precursor protein (APP), increasing Aβ42 protein, but a study by Amtul and colleagues (2002) found that the InsR 362 (but not L113P, which they also tested) caused a “dominant negative” effect on the metabolism of APP (and NOTCH), decreasing Aβ42 production. The G183V mutation does not have the same effect. In this family, there are two siblings without the mutation (II-3 age 67 and II-4 age 66) who had abnormal SPECT scans and mild dysexecutive function, and II-3 had anomia and mild MRI atrophy. All of these findings raise the possibility that FTD might not be linked to the G183V mutation. In conclusion, in the families described so far, there is suggestive but not conclusive evidence that PSEN1 mutations can cause FTD.

As people worldwide live to an older age, dementia, whose best-known risk factor is aging, has become a serious growing public health problem. While Alzheimer's disease (AD) is the most frequent cause of dementia, other progressive neurodegenerative disorders can be responsible for it. For these reasons, they are grouped as non-AD dementias. There is evidence that reactive oxygen species-mediated reactions, particularly of neuronal lipids, are present in brains affected by both types of dementing disorders. Traditional views have claimed that oxidative-mediated brain injury in these diseases is merely the result of the neurodegenerative processes. While numerous investigations have shown that oxidative stress is increased in AD, conflicting results exist for the heterogeneous group of non-AD dementias. The availability of specific and sensitive markers to monitor in vivo oxidative stress, in combination with studies performed in living patients, are helping us to elucidate these issues. This paper summarizes some of the most recent research on the relevance of oxidative stress and lipid peroxidation in AD and non-AD dementias. The evidence accumulated so far clearly indicates that oxidative stress is an early and specificaspect of AD pathogenesis but not of the pathogenesis of other dementias. This new concept implies that this phenomenon is not a general and common pathway of the neurodegenerative process, but it may play a more specific and important role in AD than in non-AD dementias.

Familial forms of frontotemporal dementia (FTD) are in 10–43% of patients, caused by mutations in the gene encoding the microtubule associated protein tau () located at chromosome 17q21. Neuropathologically, these patients are characterized by tau-positive depositions in brain. However, autosomal dominant forms of FTD without mutations have been reported, suggesting other tauopathy-related genetic defects. One such form is FTD linked to 17q21, with tau-negative but ubiquitine-positive neuronal inclusions or FTD-U. We previously reduced the candidate chromosomal region to 4.8 cM in a Dutch FTD-U family, 1083. A mutation in was excluded by genomic sequencing. More recently, we identified three Belgian FTD families of which two, DR2 and DR8, showed linkage to the 17q21 region. Both families shared a common haplotype in an 8.04 cM region, indicating that they are genetically related to a common founder. In the third family, DR7, we obtained an autopsy confirmation of the characteristic ubiquitin-positive, tau-negative neuronal inclusions. Currently, there are 11 FTD families linked to 17q21 that do not segregate a mutation, of which five are conclusively linked (LOD score > 3). Together the data suggest that FTD-U could represent an important subtype of FTD, and that identification of the underlying gene defect might significantly contribute to our understanding of the pathomechanism leading to neurodegeneration in this dementia subtype.