Alzheimer's Disease and the Neuroscience of Aging

Alzheimer’s disease, the most common cause of dementia, is the result of abnormal changes in the brain that lead to a precipitous decline in intellectual abilities and changes in behavior and personality. As the primary Federal agency responsible for research on Alzheimer’s disease, the NIA leads national efforts to gain greater understanding of the biological mechanisms underlying Alzheimer’s disease and to develop preventive measures and treatments based on research findings. Tragically, as many as four million Americans now suffer from Alzheimer’s disease,³ and the predicted explosive growth in the number of people living to 85 years and older, persons most at risk for dementia, lends an urgency to this research. Although the early signs of Alzheimer’s disease involve mild forgetfulness, the progressive dementia ultimately leaves patients incapable of caring for themselves. Behavior changes may cause patients to become agitated, sometimes to the point of causing harm to themselves or others. Alzheimer’s disease devastates its victims and profoundly affects the millions of family members and other loved ones who provide most of the care for people with this disease. Alzheimer’s disease also necessitates formal services at substantial cost to individuals and public programs, estimated at greater than $100 billion per year.⁴ While much remains to be done, research progress has been accelerating rapidly, bringing the field to the threshold of prevention trials.

Story of Discovery—Progress in Understanding Alzheimer’s Disease

In 1906, Dr. Alois Alzheimer described a patient, Auguste D., who experienced a four-year progressive decline into dementia and then died at age 55. Alzheimer’s postmortem study of Auguste D.’s brain revealed two striking pathological findings—neuritic plaques and neurofibrillary tangles. Decades later, these lesions, recognized as hallmarks of Alzheimer’s disease (AD), are the focus of a vigorous research effort to understand the underlying causes of dementia in late life and to develop compounds that will prevent the disease or block its progress.

For years, evidence has burgeoned about the protein fragment beta-amyloid in AD pathology. Beta-amyloid is the primary component of neuritic plaques, along with inflammatory cells and other insoluble filamentous proteins that can contribute to neuronal damage. An early and consistent feature in the AD brain, beta-amyloid surrounds brain cells (neurons) in regions of the brain involved in memory and cognition. Beta-amyloid peptide is derived from a much larger protein called the amyloid precursor protein (APP). The discovery in 1990 of a genetic mutation on chromosome 21 that ties APP to AD was the first real indication of a link between beta-amyloid production and the pathology of the disease. This mutation is associated with a rare form of early-onset, familial AD. Two other genes that cause early-onset AD were identified in 1995, presenilin-1 on chromosome 14 and presenilin-2 on chromosome 1. The proteins produced by the presenilin genes are tied to beta-amyloid by the discovery that the AD-linked mutations in these genes are associated with increased production of beta amyloid. A defect in any one of these three genes can cause AD, accounting for approximately 50% of the inherited early-onset cases, or as few as 10% of persons with AD.

For most individuals, however, susceptibility to AD is more complex, and its genetic component probably involves more than one gene. The only accepted risk factor for the common, late-onset form of AD is ApoE4, a variant of the ApoE gene on chromosome 19. ApoE4 accounts for approximately 50% of the genetic effect in the development of late-onset AD. Recent findings suggest that ApoE is critical for promoting beta-amyloid deposition in neuritic plaques. This and other evidence help define beta-amyloid as a prime target for intervention in the cascade of events that initiate neuronal degeneration.

An important recent advance in AD research was the generation of transgenic mice expressing the mutant forms of human beta-amyloid associated with early onset AD. These mice develop amyloid plaques with similarities to those observed in AD patients, providing for the first time a candidate mouse model of this disease. This year, scientists at the Elan Corporation immunized young beta-amyloid transgenic mice with a synthetic form of beta-amyloid found in plaques and succeeded in preventing almost entirely the deposition of beta-amyloid in mouse brains and reducing other features of disease compared with controls. Older transgenic mice vaccinated at 11 months also had a considerable reduction in amyloid deposition at 15 and 18 months when compared with controls. Although the relevance of this model to human disease remains uncertain, these results raise the possibility of immunization as a treatment or perhaps a prophylactic measure against AD.

The role of neurofibrillary tangles, the second characteristic lesion of AD, has also been the subject of recent research advances. Found in the same areas of diseased brains as plaques, but inside neurons, tangles are the wreckage of the cell’s internal structural support and nutrient transportation system. In healthy cells, microtubules are formed like train tracks—long, parallel rails stabilized by "railroad ties" consisting of the protein tau. In AD and in some other dementias, the altered tau can no longer hold the microtubule "tracks" together, causing the transport system to collapse. The tau itself twists into paired helical filaments, like two threads wound around each other. Disruption of the microtubule assembly can lead to cell death. Tau also has long been associated with nerve cell destruction. Although evidence correlates the formation of tangles and the loss of neurons in the part of the brain most affected by AD with increased severity of dementia, until recently there was no evidence that changes in tau protein could directly initiate neuronal degeneration. This changed radically in 1998 when teams of researchers linked several tau mutations on chromosome 17 to inherited dementias characterized by AD-like brain tangles and nerve cell destruction. Now, after years of multinational research on families affected by rare dementias, there is hard evidence that tau can play a primary role in causing at least some cases of neurodegenerative disease. These findings confirmed that mutations in tau alone can lead to dementia. These advances offer new directions for exploring treatments for these dementias, perhaps using drugs that mimic the properties of normal tau or drugs that halt aggregation of tau into brain tangles. Further research is also needed to reassess the relationship between tau and beta-amyloid in dementing disease and to understand how abnormal protein filaments cause cell death in Alzheimer’s disease and other neurodegenerative diseases, including Parkinson’s disease, Huntington’s disease, and prion diseases.

Science Advances—Alzheimer’s Disease and the Neuroscience of Aging

Age-Associated Memory Loss Might Be Reversible. Researchers identified a process by which the normal primate brain degenerates with aging, and were able to show that this degeneration can be reversed by gene therapy. They found that control neurons in a specific area of the brain are most dramatically affected by aging. An actual count of brain cells in rhesus monkeys showed that very few cells are actually lost in the cerebral cortex with advancing age. In contrast, control neurons in another part of the brain (the basal forebrain) were found to shrink in size and to stop making regulatory chemicals, a change that seriously affects the ability to reason and store memories. Using skin cells from each individual monkey, researchers inserted a gene that makes human nerve growth factor (NGF) and then injected the modified cells into the brains of these monkeys. After three months, the brains of the monkeys with the NGF injections had an almost youthful appearance. The number of cells detected was restored to about 92 percent of normal for a young monkey, and the size of the cells was restored to within three percent of normal young values. Such gene transfer approaches to recover cellular function have important implications for the treatment of chronic age-related neurodegenerative disorders, such as AD.

Brain Atrophy Measured by Imaging Techniques Predicts Progression From Mild Cognitive Impairment to AD. Mild cognitive impairment (MCI) is characterized by a memory deficit, but not dementia. Compared to normal memory changes associated with aging, memory loss associated with MCI is more persistent and troublesome. Each year, 12–20 percent of people over age 65 with MCI develop AD, compared with 1–2 percent of people in this same age group without MCI. A recently completed study found that MCI can reliably be clinically defined and diagnosed. The ability to differentiate patients with mild cognitive impairment (MCI) from healthy control subjects and persons with very mild AD hopefully will lead to useful, practical, and cost-effective means to test drug interventions for AD. To help make these distinctions, researchers recently used magnetic resonance imaging (MRI) to determine volume measurements of a region of the brain known as the hippocampus in patients with a clinical diagnosis of MCI. The hippocampus was selected for imaging because this brain structure plays a central role in memory function. Patients were assessed annually for approximately three years using both clinical and cognitive assessments. In older individuals with MCI, the smaller the hippocampus at the beginning of the study, the greater the risk of developing AD later. Imaging studies such as these can actually identify deviations from normal cerebral function or normal anatomy before a clinical diagnosis can be made. This is true for both structural imaging measures such as MRI, and for functional measures such as positron emission tomography (PET). The ability to detect early disease will enable researchers to test the effectiveness of treatments or interventions designed to stop brain changes before clinical deterioration sets in.

Normal Cellular Enzyme Becomes a Marker for AD. Researchers examining the brains of people who had died from AD found abnormally large amounts of a normal enzyme called casein kinase-1 (CK-1) in nerve cells inside cellular sacs (vacuoles) called granulovacuolar degeneration (GVD) bodies. Previous research had shown that these vacuoles tended to accumulate in the hippocampus, a region of the brain important for learning and memory. Looking for an enzyme that adds phosphate to tau molecule, a key protein in the development of dementia, the investigator found a 30-fold increase in one form of CK-1 inside GVD bodies in the hippocampus. This finding enables researchers to use CK-1 as a molecular label for studying the vacuoles and forges a link between them and the plaques and tangles commonly studied in AD brains. Analysis of GVD bodies could provide valuable clues useful both for the diagnosis of AD and for gaining a better understanding of the disease.

Study Results Show Promise for Developing Treatment of Early-Onset AD. Most early-onset AD is the result of mutations in one of two human presenilin genes, PS-1 and PS-2. Mutations in PS-1 are found in about 40% of people with familial (early onset) AD. Every known presenilin mutation affects the processing of amyloid precursor protein (APP) into smaller fragments, such as beta-amyloid peptide, the primary constituent of the distinctive plaques that accumulate in the brains of Alzheimer’s patients. When scientists altered the amino acid sequence of the presenilin protein from its normal sequence in two critical locations, amyloid formation was reduced. Evidence indicates that mutated PS-1 protein may be able to clip the beta-amyloid fragment from APP. If true, the identification of the long-sought enzyme involved in producing neuritic plaques associated with AD should hasten development of drugs that inhibit the enzyme, blocking production of amyloid-beta in much the way cholesterol-lowering drugs work. These studies have implications for the treatment of AD and related disorders of amyloid accumulation. The challenge will be to develop drugs that reduce, but do not eliminate, presenilin since complete elimination of presenilin is lethal in mice, and presenilin is likely to have a similar essential function in humans.

Gene Causing a Form of Familial Dementia May Yield Clues to AD. A form of dementia that spans seven generations of members of the same family in England has been linked to a newly discovered, dominant gene, BRI, on chromosome 13. Familial British dementia (FBD), which has an onset at approximately age 50, is characterized by progressive dementia, muscle spasticity, and loss of muscle tone due to disease of the cerebellum. The predominant pathological lesions are abnormal protein deposits in the brain, plaques in the vicinity of blood vessels, and neurofibrillary tangles. FBD is similar to AD because in both disorders the production of a small insoluble protein is a key feature. Further, the neurofibrillary pathology observed in both FBD and AD is identical. While much remains unknown about the BRI gene and the function of the protein that it produces, understanding how the gene defect causes the disease will lead to insights into the pathogenesis of other neurodegenerative diseases characterized by amyloid "deposition." Understanding how the genetically distinct disorder FBD develops will contribute to efforts to understand the development and progression of the more prevalent Alzheimer’s disease. Further, insights gained in FBD may aid the design and development of treatments intended to disrupt peptide aggregation and prevent the ensuing neurodegeneration not only in FBD and AD but also in other diseases such as those caused by infectious particles call prions.

One Form of the ApoE Gene Protects Brain Cells From Injury. The protein apolipoprotein E (ApoE) participates in the transport of serum lipids (fats) and the redistribution of lipids among cells. Although the mechanism through which it works is unknown, the only accepted risk factor for sporadic late-onset AD is the ApoE4 structural variant of the ApoE gene. To test the hypothesis that ApoE3, but not ApoE4, protects against age-related neurodegeneration, researchers analyzed mice expressing similar levels of human ApoE3 or ApoE4 in the brain. It was determined that ApoE3 protected the brain against excitotoxic injury but that ApoE4 did not. ApoE3, but not ApoE4, also protected against age-dependent neurodegeneration. This study presents compelling evidence to suggest that the presence or absence of a particular ApoE structural variant or isoform affects the way neurons respond to injury. These differences in the effects of ApoE isoforms on neuronal integrity may relate to the increased risk of AD and to the poor outcome after head trauma and stroke in humans. The significance of this finding is that it may help to explain how ApoE4 functions as a risk factor for the development of AD, and, if confirmed, might suggest useful therapeutic strategies that could be started in advance of any cognitive impairment in at-risk individuals.

New Mouse Model Produces Tangles Similar to Those in AD. Developing mouse models with features of human AD is vital in helping researchers gain insights into the etiologies, mechanisms, and progression of AD. Mice implanted with human genes for beta-amyloid, the precursor to neuritic plaques, were developed in 1997. Now, for the first time, researchers have developed a transgenic mouse strain that expresses human tau genes and develops AD-like tau tangles. Unlike their litter-mates that lack the tau gene, these genetically altered mice developed masses of abnormal tau filaments in nerve cells within the spinal cord, cerebral cortex, and brainstem, and in three other critical regions of the central nervous system, as well as undergoing nerve cell degeneration as they aged. While this new strain of transgenic mice does not completely model AD, they closely resemble human diseases that accumulate AD-like tau deposits in the brain. The development of this mouse model will help researchers understand how tau produces disease in the brain, and together with other partial models of AD will move closer to developing effective preventive or treatment interventions against AD.

Study Finds That the Hormone Melatonin Does not Decrease With Age. Melatonin, a natural sleep inducer, is secreted by the pineal gland located deep within the brain. The hormone is produced at high levels during a person’s normal sleeping hours and is lowest during the day. A number of factors, including light and many common medications, such as aspirin, ibuprofen, and beta-blockers, can affect melatonin secretion. In the past two decades, more than 30 reports have suggested that the level of night-time melatonin peak declines progressively with age. These reports have led to a proliferation of over-the-counter supplements aimed at augmenting melatonin levels in the elderly. A five-year study was recently completed that measured serum melatonin levels in 120 healthy men and 24 women aged 18–81. The analysis found no statistically significant difference in night-time melatonin concentrations between the younger and older study participants. This means that in most healthy people, concentration of melatonin probably does not decline with age, and aging probably does not affect the regulation of melatonin secretion.

The Circadian Clock and Aging. Circadian rhythms are pervasive throughout plants and animals, and are closely related to the 24-hour solar day. However, early observations led to the conclusion that regulation of the human circadian clock was different from that of other species, with a median 25.2 hours for a complete cycle. The concept of a shortening of the period with age was used to explain the common observation that many older people go to sleep earlier in the evening and awake earlier in the morning. Recent research using precise measures of naturally occurring melatonin, core body temperature, and cortisol levels in healthy young and older individuals has revealed that the intrinsic period of the human circadian pacemaker averages 24hr 11 min in both age groups. The circadian rhythm is controlled by a small group of neurons deep within the hypothalamus. This "free-running" period is genetically determined and varies very little within individuals of the same species. This counters the belief that the circadian clock speeds up (shortens in duration) as we age, and indicates that the human circadian clock is as stable and precise as that of other animals. This study changes some fundamental assumptions about the causes of sleeplessness among the elderly. Poor sleep is not simply a function of being old. Other factors associated with aging, such as disease, changes in environment, changes in activities, or concurrent age-related processes may contribute to problems of sleep in older persons. The similarity of circadian periods across the animal kingdom suggests that the findings of basic cellular and molecular mechanisms in these model systems will be applicable to solving the problems of sleep and wakefulness in humans.

3. Small, GW, Rabine, PV, Barry, PP, et al. Diagnosis and Treatment of Alzheimer Disease and Related Disorders. JAMA 16: 1363-1371, 1997.
4. Ernst, RL, Hay, JW, Fenn, C, Tinklenberg, J and Yesavage, J. Cognitive Function and the Costs of Alzheimer's Disease, Arch Neurol 54:687-693, 1997.