Amyloid plaques and neurofibrillary tangles are two types of abnormal protein deposits that accumulate in the brains of people with Alzheimer’s disease. Plaques are clumps of a sticky protein fragment called amyloid-beta that build up between nerve cells, while tangles are twisted fibers of a different protein called tau that form inside them. First described in 1906 by Alois Alzheimer himself during a postmortem examination of a patient’s brain, these two structures remain the defining hallmarks of the disease more than a century later.
How Amyloid Plaques Form
Amyloid-beta starts out as part of a much larger molecule called amyloid precursor protein, or APP, which sits in the outer membrane of brain cells. Under normal circumstances, an enzyme called alpha-secretase snips APP in a way that prevents amyloid-beta from ever being released. But when two other enzymes get to APP first, the outcome is different. An enzyme called BACE (beta-site APP-cleaving enzyme) cuts APP at one specific location, and then a second enzyme called gamma-secretase makes a follow-up cut, releasing the amyloid-beta fragment into the space between neurons.1PubMed. Beta-secretase cleavage of Alzheimer’s amyloid precursor protein by the transmembrane aspartic protease BACE2PubMed Central. Amyloid-β production via cleavage of amyloid-β protein precursor is modulated by cell density
Not all amyloid-beta fragments are equal. Gamma-secretase trims the protein in a stepwise fashion, snipping off three amino acids at a time, and the final length of the fragment matters enormously. The two most common versions are 40 and 42 amino acids long. The 42-amino-acid form is far more likely to clump together, and it is the version most closely linked to plaque formation in Alzheimer’s.3eLife. The amyloid-beta forming tripeptide cleavage mechanism of γ-secretase
Once released, individual amyloid-beta molecules can stick to one another, first forming small clusters called oligomers, then longer chains called protofibrils, and eventually dense, insoluble deposits known as plaques. This aggregation cascade is not merely cosmetic damage. The smaller, intermediate forms of amyloid-beta, particularly oligomers and protofibrils, are thought to be the most toxic. They disrupt communication between neurons at the synapses and trigger inflammatory reactions long before large visible plaques appear.4PubMed. The aggregation of amyloid-β: from condensation, nucleation, and conformation to targeting therapy
How Neurofibrillary Tangles Form
Tau is a fundamentally different kind of protein from amyloid-beta. In a healthy neuron, tau stabilizes the internal scaffolding known as microtubules, which act like railway tracks along which nutrients and signals are transported from one end of the cell to the other. Tau binds to microtubules cooperatively, forming a kind of protective coating that helps regulate which other molecules can access the tracks and keeps intracellular transport running smoothly.5eLife. Tau hyperphosphorylation disrupts cooperative microtubule binding and dysregulates axonal transport
Problems begin when tau picks up too many phosphate groups, a process called hyperphosphorylation. Normal tau has some phosphate groups attached, but in Alzheimer’s, the balance tips: phosphate groups are added at sites where they should not be, or far more are added than the protein can handle. Hyperphosphorylated tau loses its grip on microtubules, dissociates faster, and distributes unevenly along the axon. The microtubule tracks lose their stability, and the transport of essential cargo inside the neuron breaks down.5eLife. Tau hyperphosphorylation disrupts cooperative microtubule binding and dysregulates axonal transport
Freed from its normal job, the excess hyperphosphorylated tau begins to stick to other tau molecules. The initial step is dimerization, in which two tau molecules pair up. These dimers then recruit additional tau monomers in a nucleation process that requires roughly eight to fourteen tau molecules to form a stable seed. Once a seed reaches this critical size, it grows rapidly into long twisted fibers called paired helical filaments, which bundle together into the neurofibrillary tangles visible under a microscope.6PubMed. A nucleated assembly mechanism of Alzheimer paired helical filaments7PubMed. Hyperphosphorylation induces self-assembly of tau into tangles of paired helical filaments/straight filaments
The process can also be seeded by pre-existing filaments, including fragments isolated from Alzheimer’s brains, which helps explain how tau pathology appears to spread from one brain region to the next.6PubMed. A nucleated assembly mechanism of Alzheimer paired helical filaments
Where Plaques and Tangles Appear First
Plaques and tangles do not appear everywhere in the brain at once. They follow remarkably predictable geographic patterns, which researchers have mapped out using staging systems named after the neuropathologists who described them. Amyloid plaques typically show up first in the outer layers of the cerebral cortex, especially in temporal and frontal regions, before gradually spreading into deeper structures and eventually reaching the cerebellum. Tau tangles follow a different route: they tend to start in areas related to memory, including the entorhinal cortex and hippocampus, and only later spread outward to involve the rest of the cortex.8Brain. Clinicopathologic and 11C-Pittsburgh compound B implications of Thal amyloid phase across the Alzheimer’s disease spectrum
These two staging systems, Thal phases for amyloid and Braak stages for tau, are not independent of each other. As amyloid pathology advances, the likelihood of tau pathology reaching the cortex increases sharply. Among people in the earliest Thal phase of amyloid deposition, about 97% still had tangles confined to limbic regions or below. By the most advanced amyloid phase, that number dropped to just 6%, meaning nearly everyone also had extensive cortical tau.8Brain. Clinicopathologic and 11C-Pittsburgh compound B implications of Thal amyloid phase across the Alzheimer’s disease spectrum
Research in people with Down syndrome, who are genetically predisposed to developing Alzheimer’s pathology early, has helped clarify the timing more precisely. In this population, diffuse amyloid plaques begin appearing in the temporal cortex as early as the late teens or twenties, while tau pathology typically starts after age 35, initially in the hippocampus and also in deep brainstem structures like the locus coeruleus.9PubMed Central. The age of onset and evolution of Braak tangle stage and Thal amyloid pathology of Alzheimer’s disease in individuals with Down syndrome
The Brain’s Immune Response
Plaques and tangles do not exist in a vacuum. The brain has its own immune cells, primarily microglia and astrocytes, that respond to these abnormal deposits. Early in the disease, this response is actually helpful. Microglia can engulf and break down amyloid-beta, acting like cellular garbage collectors. But as the disease progresses and the amyloid burden grows, these immune cells shift from protective mode to a chronically activated, inflammatory state. They begin releasing inflammatory signaling molecules, including compounds like interleukin-1-beta and tumor necrosis factor-alpha, which damage surrounding neurons.10PubMed. Activation of microglia and astrocytes: a roadway to neuroinflammation and Alzheimer’s disease
This creates a vicious cycle. Chronic inflammation impairs the ability of microglia to clear amyloid-beta, so amyloid accumulates further, which drives even more inflammation. The resulting neuroinflammation is now considered a major contributor to cell death in Alzheimer’s, not just a bystander effect.10PubMed. Activation of microglia and astrocytes: a roadway to neuroinflammation and Alzheimer’s disease
How Amyloid-Beta and Tau Damage Synapses Together
For a long time, amyloid plaques and neurofibrillary tangles were studied as somewhat separate problems. Increasingly, evidence points to a synergy between the two. Small oligomeric forms of both amyloid-beta and tau appear to work together to disrupt synaptic connections, the junctions where neurons communicate. Synaptic dysfunction shows up before neurons actually die, and the regions where synapses fail earliest tend to be the same regions that later lose the most cells.11PubMed Central. Tau Oligomers: The Toxic Player at Synapses in Alzheimer’s Disease
One of the more surprising findings is that tau may need to leave its normal territory for amyloid-beta to do its worst. Tau normally lives in the axon, the long projection a neuron sends out. But in Alzheimer’s disease, tau becomes mis-sorted and shows up in the dendrites, the branching input end of the neuron. A growing body of research suggests this mislocalization of tau into dendrites is actually required for amyloid-beta to cause its toxic effects at the synapse.11PubMed Central. Tau Oligomers: The Toxic Player at Synapses in Alzheimer’s Disease
Sleep and the Clearance of Brain Proteins
Your brain does not simply produce amyloid-beta and tau and leave them there. It has a clearance system, driven largely by what is called the glymphatic pathway, which flushes waste products from the brain into the bloodstream, primarily during sleep. A recent study found that the overnight clearance of amyloid-beta and tau into the blood was strongly influenced by deep sleep quality, cerebrovascular health, and the physical resistance of brain tissue to fluid flow. These factors accounted for more than half the variation in morning blood levels of amyloid-beta and tau in people with amyloid pathology, and an even higher proportion in those without it.12PubMed Central. The glymphatic system clears amyloid beta and tau from brain to plasma in humans
When sleep is disrupted, the consequences for these proteins are measurable. In a controlled study, one night of sleep deprivation increased certain forms of tau in cerebrospinal fluid by 30 to 50% compared to normal sleep, with the phosphorylation at one particular site jumping by 60 to 80%.13PubMed Central. Sleep deprivation affects tau phosphorylation in human cerebrospinal fluid Similarly, animal studies have shown that amyloid-beta levels in brain fluid rise roughly 90% during normal wakefulness compared to sleep, and even more during forced sleep deprivation. Chronic sleep deprivation in mice also accelerated the spread of tau pathology to new brain regions.14PubMed Central. The sleep-wake cycle regulates brain interstitial fluid tau in mice and CSF tau in humans
This does not mean that a few bad nights cause Alzheimer’s. But it suggests that chronically poor sleep could contribute to the gradual buildup of the very proteins that define the disease, creating a slow-burning problem that compounds over years.
How Doctors Detect Plaques and Tangles in Living People
Until recently, the only way to confirm amyloid plaques and neurofibrillary tangles was to examine brain tissue after death. That has changed substantially. Amyloid PET scans use radioactive tracers that bind to amyloid deposits in the brain, allowing doctors to estimate plaque density in a living person. Several tracers, including florbetapir, flutemetamol, and florbetaben, are approved for clinical use. Tau-specific PET tracers have also been developed, and both types show high accuracy when compared against autopsy findings.15PubMed Central. Role of Fluid Biomarkers and PET Imaging in Early Diagnosis and its Clinical Implication in the Management of Alzheimer’s Disease
Fluid-based biomarkers offer a less expensive alternative. Cerebrospinal fluid tests measuring amyloid-beta 42 and tau have been used clinically for years. The ratio of amyloid-beta 42 to amyloid-beta 40, or the ratio of phosphorylated tau to amyloid-beta 42, tends to predict Alzheimer’s pathology more accurately than measuring any single protein alone. Blood-based tests are catching up: plasma measurements of the amyloid-beta 42/40 ratio and phosphorylated tau at position 181 now show strong agreement with PET scan results, making broad screening far more feasible.16PubMed. Diagnostic Biomarkers of Amyloid and Tau Pathology in Alzheimer’s Disease: An Overview of Tests for Clinical Practice in the United States and Europe
Plaques and Tangles Without Dementia
One of the more humbling findings in Alzheimer’s research is that some people accumulate heavy loads of both plaques and tangles yet never develop cognitive problems during their lifetime. Postmortem studies consistently find individuals whose brains look, under a microscope, as though they should have had severe dementia, yet who were mentally sharp until the end.17Nature Reviews Neurology. Lesions without symptoms: understanding resilience to Alzheimer disease neuropathological changes A study of cognitively normal elderly people found that their Braak scores and plaque loads at death spanned a wide range, with some showing pathology comparable to people who had been diagnosed with Alzheimer’s.18PubMed Central. Braak staging, plaque pathology, and APOE status in elderly persons without cognitive impairment
This phenomenon is usually described as “resilience” or “cognitive reserve.” What allows these individuals to tolerate so much pathology remains one of the most important open questions in the field. Proposed explanations include greater synaptic density, more robust neuronal connections, less neuroinflammation, and differences in how the proteins interact in these brains. Whatever the mechanism, the existence of resilient individuals makes clear that plaques and tangles are necessary for an Alzheimer’s diagnosis but not always sufficient to cause symptoms on their own.
Drugs That Target Amyloid Plaques
The past few years have seen the first Alzheimer’s drugs designed specifically to clear amyloid-beta from the brain. Lecanemab targets early toxic forms of amyloid, particularly protofibrils, while donanemab goes after a chemically modified version of amyloid that is found in mature plaques. A third antibody, gantenerumab, aims at both soluble and fibrillar forms of amyloid-beta and promotes clearance through the brain’s own immune cells.19The Egyptian Journal of Neurology, Psychiatry and Neurosurgery. Amyloid solutions: lecanemab, gantenerumab, and donanemab in the treatment of Alzheimer’s disease
In clinical trials, both donanemab and lecanemab significantly reduced amyloid plaque levels on PET scans compared to placebo. Lecanemab also slowed the rate of clinical decline, with one phase III trial reporting that the drug roughly halved the worsening in a quality-of-life measure compared to placebo.20Equity Neuroscience. Donanemab and lecanemab in Alzheimer’s disease treatment: A narrative review of clinical trials and discussion of implications for patient access These are real but modest effects. The drugs slow cognitive decline rather than stopping or reversing it, and they carry a risk of brain swelling and microbleeds. No approved therapy currently targets neurofibrillary tangles directly, though several anti-tau approaches are in clinical trials.
APOE4 and the Genetic Connection
The strongest common genetic risk factor for Alzheimer’s disease is a variant of the apolipoprotein E gene called APOE ε4. Carrying one copy roughly triples the risk; carrying two copies raises it more than tenfold. The connection between APOE ε4 and amyloid-beta has been known for decades, but its relationship with tangles is also substantial. Postmortem studies show that people who carry two copies of APOE ε4 have more tau pathology than those with one copy or none. In laboratory models, neurons expressing the ε4 variant produce more hyperphosphorylated tau, and humanized mice carrying both the ε4 gene and a tau mutation develop worse brain shrinkage and inflammation.21Acta Pharmaceutica Sinica B. Apolipoprotein E and Alzheimer’s disease
Interestingly, some studies have also found increased tau pathology in mice carrying the APOE ε2 variant, which is generally considered protective against Alzheimer’s. The mechanisms behind this are not fully understood, but the finding highlights that the relationship between APOE and tau is more complicated than a simple “bad gene, more tangles” story.
Tau Tangles Beyond Alzheimer’s Disease
Neurofibrillary tangles are not exclusive to Alzheimer’s. A group of brain diseases called tauopathies all involve abnormal tau accumulation, though the specific shapes and locations of the deposits differ. Chronic traumatic encephalopathy, known as CTE, is one of the most publicly recognized examples. CTE is linked to repeated head impacts, and its defining feature is the accumulation of hyperphosphorylated tau in a pattern distinct from what is seen in Alzheimer’s.22PubMed Central. Cognitive, functional, and neuropsychiatric correlates of regional tau pathology in autopsy-confirmed chronic traumatic encephalopathy
Other tauopathies include progressive supranuclear palsy, corticobasal degeneration, and some forms of frontotemporal dementia. What distinguishes Alzheimer’s from these conditions is the co-occurrence of amyloid plaques alongside tangles. In most other tauopathies, the tau deposits appear without significant amyloid pathology, which is one reason why the interaction between plaques and tangles is considered so central to understanding Alzheimer’s specifically.
What Cryo-Electron Microscopy Has Revealed
A major advance in recent years has been the ability to image the actual molecular structures of amyloid and tau filaments extracted from human brains. Using cryo-electron microscopy, researchers have determined that amyloid-beta 42 filaments from Alzheimer’s brains adopt an S-shaped fold and come in two types. Type I filaments are found predominantly in sporadic Alzheimer’s cases, the common form not caused by a known inherited mutation. Type II filaments are found primarily in familial (inherited) Alzheimer’s and in certain other conditions.23PubMed Central. Cryo-EM structures of amyloid-β 42 filaments from human brains
Tau filaments from Alzheimer’s brains consistently adopt what has been called the “Alzheimer’s disease fold,” and this same fold appears in both sporadic and familial cases, as well as in a distinctive plaque type called cotton wool plaques found in certain inherited mutations.24PubMed Central. Cryo-EM structures of cotton wool plaques’ amyloid β and of tau filaments in dominantly inherited Alzheimer disease In other tauopathies, tau adopts entirely different folds, meaning the shape of the tau filament effectively acts as a molecular signature of each disease. This discovery has practical implications: it suggests that therapies designed to break up or prevent tau aggregation may need to be tailored to the specific fold involved, and a drug that works in Alzheimer’s might not work in CTE or frontotemporal dementia.
Normal Roles of Amyloid-Beta
Given how much attention is paid to its destructive potential, it is easy to forget that amyloid-beta is a normal brain product that appears to serve useful functions at low concentrations. Emerging research suggests monomeric amyloid-beta helps regulate synaptic plasticity, assists in recovery after brain injury, helps seal leaks in the blood-brain barrier, and may even suppress microbial infections. The pattern is hormetic: at low concentrations, amyloid-beta is neuroprotective, but at higher concentrations it becomes pathological.25PubMed Central. Physiological Roles of Monomeric Amyloid-β and Implications for Alzheimer’s Disease Therapeutics
This dual nature creates a genuine dilemma for treatment. Drugs that aggressively clear amyloid-beta from the brain might remove not only the harmful aggregated forms but also the low-level monomers that serve protective roles. Whether this matters clinically is still an open question, but it is one reason researchers are focused on targeting the toxic aggregated forms of the protein rather than eliminating all amyloid-beta production entirely.