Neurodegenerative disease: protein misfolding, Alzheimer's, Parkinson's, and the amyloid/tau/synucleinopathies

Intuition Beginner

Your brain has billions of neurons, and each neuron contains billions of protein molecules folded into precise three-dimensional shapes. In neurodegenerative disease, certain proteins misfold: they take on a wrong shape, stick to other misfolded copies, and form toxic clumps that kill neurons. Different diseases involve different proteins. Alzheimer's disease features amyloid-beta in plaques outside neurons and tau in tangles inside neurons. Parkinson's disease features alpha-synuclein in Lewy bodies inside neurons. ALS features TDP-43. Prion diseases like Creutzfeldt-Jakob disease feature prions.

The clumps spread neuron to neuron along the brain's wiring, which is why each disease begins in a different brain region. Alzheimer's starts in the hippocampus, the memory center, so the first symptom is forgetfulness. Parkinson's starts in the substantia nigra, a movement-control center, so the first symptom is tremor and stiffness. By the time memory loss or tremor appears, large fractions of the affected neurons are already dead. This is why current treatments can slow but not stop these diseases.

Neurodegenerative disease is the leading cause of disability in aging populations. About 55 million people worldwide live with dementia (2023 WHO estimate), and Alzheimer's accounts for about 70 percent of cases. Parkinson's is the fastest-growing neurological disorder: prevalence doubled from 1990 to 2015. The shared mechanism — protein misfolding and prion-like spread — is the reason this concept exists as a single unit rather than four separate diseases.

Visual Beginner

The picture shows three classes of misfolded-protein pathology side by side, plus the propagation model that unifies them. In Alzheimer's, amyloid-beta plaques sit outside neurons and tau tangles sit inside them. In Parkinson's, alpha-synuclein aggregates form Lewy bodies inside neurons. In Creutzfeldt-Jakob disease, prion aggregates spread so fast that the whole brain becomes sponge-like within months.

The bottom of the picture shows the Braak-Braak 1991 staging of Alzheimer's disease as a wave of tau pathology moving from the entorhinal cortex (stage I-II), through the hippocampus (stage III-IV), to the neocortex (stage V-VI) over about twenty years. The wavefront follows anatomical connections, which is the prion-like propagation idea in visual form.

Worked example Beginner

Consider a woman who will develop Alzheimer's disease at age 75. The disease process is already underway in her brain long before any symptom appears.

Braak stage I-II, around age 50 to 55. Tau tangles are confined to the entorhinal cortex, a small region where the brain's memory stream begins. She has no symptoms. A doctor would find nothing wrong on cognitive testing.

Braak stage III-IV, around age 65 to 72. Tau tangles have spread into the hippocampus. She now notices mild memory problems — losing her keys, repeating questions — a clinical state called mild cognitive impairment. About 30 percent of hippocampal neurons are already gone.

Braak stage V-VI, around age 75 and beyond. Tau tangles fill the neocortex. Memory, language, and reasoning collapse. By the time dementia is diagnosed, about 50 percent of hippocampal neurons are lost. No drug can regrow them.

Step 1: the lecanemab trial of 2023 enrolled 1,795 patients with early Alzheimer's disease and randomized them to monoclonal antibody or placebo for 18 months.

Step 2: lecanemab cleared most amyloid-beta plaques and slowed cognitive decline by 27 percent on the Clinical Dementia Rating scale.

Step 3: 27 percent slowing over 18 months translates to about 5 months of preserved cognitive function — modest, but the first disease-modifying therapy in Alzheimer's history.

What this tells us: even modest slowing matters when the disease course is 10 to 15 years, but the propagation model predicts that intervening before neuron loss — at Braak stage I-II, decades before symptoms — is the only realistic route to true disease modification.

Check your understanding Beginner

Formal definition Intermediate+

Definition (neurodegenerative disease). A neurodegenerative disease is a chronic, progressive disorder characterized by the progressive loss of structurally and functionally defined populations of post-mitotic neurons, accompanied by the accumulation of misfolded protein aggregates. The major classes are defined by the principal aggregating protein:

• Amyloidopathies — extracellular plaques of amyloid-beta (A$\beta$) peptides, the hallmark of Alzheimer's disease (Glenner-Wong 1984 BBRC 120:885).
• Tauopathies — intracellular neurofibrillary tangles of hyperphosphorylated tau, the second hallmark of Alzheimer's and the primary lesion in chronic traumatic encephalopathy, progressive supranuclear palsy, and corticobasal degeneration (Goedert 1988 PNAS 85:4051).
• Synucleinopathies — intracellular Lewy bodies and neurites of alpha-synuclein, the hallmark of Parkinson's disease, Lewy body dementia, and multiple system atrophy (Spillantini 1997 Nature 388:839).
• TDP-43 proteinopathies — intracellular inclusions of TDP-43 in nearly all amyotrophic lateral sclerosis and about half of frontotemporal dementia cases.
• Prion diseases — autocatalytic conversion of native PrP${}^{\mathrm{C}}$ into the misfolded scrapie isoform PrP${}^{\mathrm{Sc}}$, the agent of Creutzfeldt-Jakob disease, kuru, and bovine spongiform encephalopathy (Prusiner 1982 Science 216:136).

Definition (templated misfolding and prion-like propagation). Each of these proteins is natively unfolded or marginally stable in physiological conditions and can adopt a beta-sheet-rich conformation that self-templates: a small seed of misfolded monomers converts naïve monomers to the same conformation at a rate proportional to seed concentration. The aggregate then fragments, producing new seeds, and the cycle accelerates. When seeds traffic along anatomical axonal connections, regional involvement follows network topology. This is the prion-like propagation model (Brundin-Melki-Kopito 2010 Nat. Rev. Mol. Cell Biol. 11:301; Walsh-Selkoe 2016).

Definition (Braak staging). Braak-Braak 1991 defines six neuropathological stages of Alzheimer's disease by the stereotyped spread of tau neurofibrillary tangles: stage I-II in the entorhinal cortex, stage III-IV in the hippocampus and adjacent limbic regions, stage V-VI throughout the association neocortex. Braak 2003 defines an analogous six-stage scheme for Parkinson's disease by the stereotyped spread of alpha-synuclein pathology: stage 1 in the dorsal motor nucleus of the vagus and olfactory bulb, stage 2 in the medulla and pontine coeruleus, stage 3 in the midbrain substantia nigra, stage 4 in the basal forebrain and temporal mesocortex, stage 5 in the neocortex, stage 6 in primary sensory and motor areas. In both diseases, motor or cognitive symptoms appear only at intermediate-to-late stages.

Definition (genetic risk). Sporadic Alzheimer's disease (about 95 percent of cases) has the APOE epsilon-4 allele as its strongest genetic risk factor: one copy increases risk roughly threefold, two copies roughly twelvefold. Familial early-onset Alzheimer's (about 5 percent of cases) is autosomal-dominant and caused by mutations in APP (amyloid precursor protein), PSEN1, or PSEN2 (Hardy 1991+). Familial Parkinson's is caused by mutations in SNCA (alpha-synuclein), LRRK2, GBA, PINK1, Parkin (PRKN), or DJ-1; these unify the alpha-synuclein, mitochondrial, and lysosomal theories of the disease.

Definition (emerging therapies). Anti-amyloid monoclonal antibodies — aducanumab (controversially FDA-approved 2021), lecanemab (full approval 2023), donanemab (2024) — bind aggregated A$\beta$ and accelerate microglial clearance; the major risk is amyloid-related imaging abnormalities (ARIA), reflecting vascular A$\beta$ deposition. Anti-alpha-synuclein antibodies (prasinezumab, cinpanemab) have shown mixed results. Deep brain stimulation of the subthalamic nucleus or globus pallidus internal segment controls the motor symptoms of Parkinson's but does not halt progression.

Counterexamples to common slips Intermediate+

• "Significant memory loss is a normal part of aging." No. Normal aging produces mild word-finding pauses and slower processing speed, but disabling memory loss for recent events reflects Alzheimer's or another neurodegenerative process. The confusion is harmful because it delays diagnosis.
• "Amyloid-beta plaques equal Alzheimer's disease." The correlation is imperfect. About 30 percent of cognitively normal elderly people have substantial amyloid plaque burden at autopsy — the so-called asymptomatic AD puzzle. Tau tangles, not plaques, correlate with cognitive decline.
• "The amyloid cascade hypothesis is dead." Misleading. The 30-year history of anti-A$\beta$ trial failures was sobering, but lecanemab and donanemab both show measurable slowing. The hypothesis is partially correct; the failure mode was intervening too late and with insufficient target engagement.
• "Parkinson's disease is just tremor." No. Non-motor symptoms — anosmia (loss of smell), REM-sleep behavior disorder, constipation, depression, progressive dementia — precede motor symptoms by years or decades and define much of the disease burden. The motor symptoms are the tip of a much larger iceberg.
• "Alzheimer's runs in families." Mostly no. Only about 5 percent is autosomal-dominant early-onset (APP, PSEN1, PSEN2). The remaining 95 percent is sporadic, with APOE epsilon-4 as the main genetic risk factor and age as the dominant risk.
• "Deep brain stimulation cures Parkinson's." No. DBS controls motor symptoms — tremor, rigidity, dyskinesia — while the underlying alpha-synuclein pathology continues to spread. It is symptomatic therapy, not disease modification.

Key result: prion-like propagation Intermediate+

Theorem (anatomical-connectivity propagation; Brundin-Melki-Kopito 2010; Walsh-Selkoe 2016). Let $G=(V,E)$ be the graph of an anatomical brain connectome on $n$ regions, with weighted adjacency $W_{ij}$ giving the density of axonal projections from region $i$ to region $j$. Suppose a misfolded-protein seed of concentration $s_i(t)$ is produced in region $i$ at rate ${\alpha }_i$, cleared by proteostasis at rate $c$, and templated onto naïve monomer at rate $k$. Suppose aggregate seeds are transported along axons at rate $\beta W_{ij}$ from $i$ to $j$. Under the linear regime $s\ll K_M$, the regional seed concentrations satisfy

$$\frac{d{s}_i}{dt}={\alpha }_i+(k-c)s_i+\beta \sum _j{W}_{ji}{s}_j-\beta \sum _j{W}_{ij}{s}_i.$$

If $k>c$ (templating beats clearance) and the initial seed is localized to region $r$, then the time $t_j$ at which region $j$ first exceeds a pathological threshold $\theta$ is asymptotically governed by the shortest path distance from $r$ to $j$ in the weighted graph $(V,E,W)$:

$$t_j\approx \frac{1}{k-c}\log \left(\frac{\theta }{s_0}\right)+\frac{1}{\beta }\cdot d_W(r,j).$$

Consequently, the order in which regions become pathologically involved is the order of their graph-theoretic distance from the origin, which is the empirical content of Braak-Braak staging for Alzheimer's and Braak staging for Parkinson's.

Proof.

Step 1 — local dynamics. Within a single region in the absence of trafficking, the seed concentration obeys $d{s}_i/dt={\alpha }_i+(k-c)s_i$. If $k>c$, the seed grows exponentially once a naïve monomer pool exists, with local doubling time $\log 2/(k-c)$.

Step 2 — coupling through the connectome. Aggregate seeds are transported along axons at rate proportional to projection density. Region $j$ sends seeds to region $i$ at rate $\beta W_{ji}{s}_j$ and loses seeds to region $j$ at rate $\beta W_{ij}{s}_i$. The full system is the linearization above, written in matrix form $\dot{s}=\alpha +(k-c)s+\beta (W^T-D)s$, where $D$ is the weighted out-degree matrix.

Step 3 — origin-localized initial condition. Take $s(0)=s_0{\delta }_r$. The transport component $\beta (W^T-D)s$ propagates the seed as a discrete diffusion process on the graph. By standard spectral graph theory, the leading-order solution is a sum over paths from $r$:

$$s_j(t)=s_0{e}^{(k-c)t}\cdot \sum _{\text{paths }r\to j}\frac{{\beta }^{|\gamma |}{t}^{|\gamma |}}{|\gamma |!}\cdot \prod _{e\in \gamma }{W}_e+O({\alpha }_{i}t).$$

Step 4 — threshold-crossing time. Setting $s_j(t_j)=\theta$ and solving logarithmically, the dominant term is the path with the smallest product ${\prod }_e(1/W_e)$ from $r$ to $j$, which is exactly the shortest-path distance $d_W(r,j)$ in the weighted graph. The local exponential growth contributes the $\log (\theta /s_0)/(k-c)$ term. Endogenous production ${\alpha }_i$ is a small correction that shifts thresholds but does not change the ordering.

Step 5 — derivation of Braak staging. Choose $r$ = entorhinal cortex and $W$ = the human white-matter connectome. The shortest-path ordering of cortical regions reproduces the Braak-Braak 1991 staging of Alzheimer tau pathology to within regional resolution. The same argument with $r$ = dorsal motor nucleus of the vagus and $W$ = brainstem-and-forebrain connectivity reproduces the Braak 2003 staging of Parkinson alpha-synuclein pathology. The propagation model therefore predicts the staging schemes from connectome data alone. $\square$

Bridge. This result builds toward the design rationale for early-intervention trials in 35.03.01, where the chronic-disease survey identifies timing-of-intervention as the load-bearing variable across cardiovascular, oncology, and neurodegeneration chapters, and the prion-like propagation model appears again in 17.09.02, where the action-potential machinery of the neuron supplies the physical substrate along which aggregates traffic. The foundational reason combination anti-amyloid therapy can fail in advanced disease yet succeed in early disease is precisely that threshold-crossing is a one-way ratchet — once a region exceeds $\theta$, neuron loss begins and clearing the upstream seed cannot regenerate dead neurons. This is exactly the insight that identifies the asymptomatic-AD population as the correct target for true disease modification, and the bridge is between the molecular biology of templated misfolding and the clinical epidemiology of pre-symptomatic intervention.

Exercises Intermediate+

Advanced results Master

Result 1 (Alzheimer 1907 — the index case). Alois Alzheimer's 1907 report ^{[Alzheimer1907]} described Auguste D., a 51-year-old woman with progressive memory loss, disorientation, hallucinations, and psychosis. At autopsy, Alzheimer found the two lesions — extracellular plaques and intracellular neurofibrillary tangles — that still define the disease. Kraepelin named the syndrome Alzheimer's disease in his 1910 textbook, distinguishing it from senile dementia on the basis of early age of onset; the distinction is no longer considered valid, since the pathology is identical.

Result 2 (Glenner-Wong 1984 — amyloid-beta isolation). Glenner and Wong isolated and sequenced the cerebrovascular amyloid protein from Alzheimer's brain tissue, identifying it as a 4 kilodalton peptide derived from a larger precursor ^{[GlennerWong1984]}. The peptide, named amyloid-beta, was later localized to chromosome 21 (the Down-syndrome chromosome, whose carriers develop AD-like pathology by age 40 to 50). The 1984 paper established the molecular identity of the plaque and made modern anti-amyloid therapy possible.

Result 3 (Goedert 1988 and Spillantini 1997 — tau and alpha-synuclein cloning). Goedert and colleagues cloned the cDNA encoding the tau component of paired helical filaments in 1988 ^{[Goedert1988]}, establishing tau as the tangle protein. Spillantini and colleagues showed in 1997 that alpha-synuclein is the main component of Lewy bodies ^{[Spillantini1997]}, unifying Parkinson's disease, Lewy body dementia, and multiple system atrophy as synucleinopathies. Together with Glenner-Wong 1984, these results completed the molecular taxonomy of the major neurodegenerative proteinopathies.

Result 4 (Prusiner 1982 — the prion hypothesis). Prusiner's 1982 paper introduced the term prion and argued that the scrapie agent was a protein with no required nucleic acid component ^{[Prusiner1982]}. The hypothesis was controversial for a decade but is now established: the scrapie isoform PrP${}^{\mathrm{Sc}}$ templates the conversion of host PrP${}^{\mathrm{C}}$ by direct molecular contact. Prusiner received the 1997 Nobel Prize in Physiology or Medicine. The prion paradigm was later extended heuristically to other neurodegenerative diseases, where aggregates show prion-like spread without being infectious between individuals.

Result 5 (Hardy 1992 and Braak-Braak 1991 — amyloid cascade and staging). Hardy and Higgins proposed the amyloid cascade hypothesis in 1992: A$\beta$ accumulation is upstream of tau pathology, which is upstream of neurodegeneration and dementia ^[Hardy1992]. Braak and Braak in 1991 published the anatomical staging scheme showing that tau pathology spreads stereotypically from entorhinal cortex through hippocampus to neocortex ^{[BraakBraak1991]}. These two results together define the modern framework: a molecular cause (A$\beta$) and a propagation pattern (tau staging), with the relationship between them being the load-bearing question of Alzheimer's research.

Result 6 (Braak 2003 — Parkinson staging). Braak and colleagues extended the staging approach to Parkinson's disease in 2003 ^[Braak2003], showing that alpha-synuclein pathology begins in the dorsal motor nucleus of the vagus and the olfactory bulb (stage 1), ascends through the brainstem (stages 2-3, reaching the substantia nigra), and finally invades the forebrain and neocortex (stages 4-6). This staging explains the prodromal non-motor symptoms — anosmia and constipation appear years before tremor, because the olfactory bulb and enteric nervous system are stage-1 sites.

Result 7 (van Dyck 2023 — lecanemab phase 3). The Clarity AD trial randomized 1,795 patients with early Alzheimer's disease to lecanemab or placebo for 18 months ^{[vanDyck2023]}. The result was a 27 percent slowing of cognitive decline on the Clinical Dementia Rating Sum of Boxes, with substantial amyloid-beta plaque clearance. About 13 percent of treated patients developed amyloid-related imaging abnormalities (ARIA), reflecting vascular amyloid. This is the first unambiguously positive disease-modifying trial in Alzheimer's after three decades of failures, and it establishes that anti-amyloid therapy works when given early enough and with sufficient target engagement.

Synthesis. The 120-year arc from Alzheimer's 1907 case report to the 2023 lecanemab trial is the foundational reason neurodegeneration is now a single conceptual object rather than four separate diseases. The central insight — that natively-unfolded proteins can adopt beta-sheet-rich conformations that template further misfolding and propagate along anatomical connections — appears again in each major result above, and the pattern generalises from prion disease to Alzheimer, Parkinson, ALS, and the tauopathies. Putting these together with the connectivity-staging theorem identifies templated polymerisation with anatomical spread, and this is exactly the bridge between molecular biophysics and clinical phenomenology; the bridge is also the reason the asymptomatic-AD population is the correct target for true disease modification, since the model predicts a one-way ratchet once threshold is crossed. The same mathematical structure builds toward future anti-tau and anti-alpha-synuclein trials now in phase 2, and the propagation model appears again in 35.03.01 as the central reason timing-of-intervention dominates every other therapeutic variable in chronic-disease medicine.

Full proof set Master

Proposition 1 (template kinetics is a one-way ratchet above threshold). Under the prion-like propagation model with templating rate $k$ and clearance rate $c$, if $k>c$ and the seed concentration $s$ exceeds a nucleation threshold $s_{\text{nuc}}$ in any region, then in the absence of intervention the seed concentration grows exponentially and the region is committed to subsequent pathology regardless of upstream events.

Proof. Above the nucleation threshold, naïve monomer is depleted slowly relative to seed growth because the seed grows by contact, with templating rate $k\cdot s\cdot m$ where $m$ is naïve-monomer concentration. Below saturation, $m\approx m_0$ is approximately constant, so $ds/dt=(k{m}_0-c)s=(k_{\text{eff}}-c)s$ with $k_{\text{eff}}=k{m}_0>c$. The solution is $s(t)=s(0)e^{(k_{\text{eff}}-c)t}$, which crosses any pathological threshold $\theta >s(0)$ in finite time $t_{\theta }=\frac{1}{k_{\text{eff}}-c}\log (\theta /s(0))$. Once $s(t_{\theta })=\theta$, downstream processes — synaptic dysfunction, microglial activation, neuron death — begin, and clearing the seed after this point halts progression only for those neurons still alive. This is the formal content of the claim that neurodegeneration is a threshold phenomenon with a one-way ratchet. $\square$

Proposition 2 (anatomical staging is graph-shortest-path). In the propagation model of the Key Result theorem, if regions are ordered by their weighted graph distance $d_W(r,\bullet )$ from the origin $r$, the resulting ordering coincides with the empirical Braak-Braak 1991 staging of Alzheimer's tau pathology and the Braak 2003 staging of Parkinson alpha-synuclein pathology, given the human white-matter connectome as $W$.

Proof. The asymptotic threshold-crossing time $t_j=\frac{1}{k-c}\log (\theta /s_0)+\frac{1}{\beta }{d}_W(r,j)$ is monotone increasing in $d_W(r,j)$. Therefore the ordering of regions by $t_j$ is the ordering by $d_W(r,j)$, which is the ordering by anatomical proximity to the origin in the human connectome. Substituting the measured white-matter connectome for $W$, the predicted ordering reproduces Braak-Braak 1991 (for $r$ = entorhinal cortex and tau as the propagating species) and Braak 2003 (for $r$ = dorsal motor nucleus of the vagus and alpha-synuclein as the propagating species) within regional resolution. Endogenous production ${\alpha }_i$ shifts absolute thresholds but does not change the ordering in regions where ${\alpha }_i$ is comparable; in regions with very large ${\alpha }_i$ (rare), pathology can originate locally rather than by propagation, which is the explanation for the occasional non-Braak AD case. $\square$

Proposition 3 (combination anti-amyloid plus anti-tau is more effective than either alone, predicted by the cascade). Under the amyloid cascade hypothesis (Hardy 1992), where A$\beta$ accelerates tau aggregation, simultaneous reduction of both A$\beta$ production (by anti-amyloid antibody) and tau aggregation (by anti-tau antibody) yields greater slowing than the sum of individual effects, because the cascade amplification is broken at two points rather than one.

Proof. Let A$\beta$ concentration be $a(t)$ with production rate $P_a$ and clearance $c_a$; let tau-aggregate seed concentration be $\tau (t)$ with intrinsic templating $k_{\tau }$, clearance $c_{\tau }$, and A$\beta$-driven acceleration $\gamma a(t)\tau (t)$. The coupled dynamics are

$$\dot{a}=P_a-c_{a}a,\dot{\tau }=(k_{\tau }+\gamma a-c_{\tau })\tau .$$

If an anti-amyloid antibody reduces $a$ from $a_0$ to $a_1=ϵa_0$ (small $ϵ$), then the effective tau-growth rate drops from $k_{\tau }+\gamma a_0-c_{\tau }$ to $k_{\tau }+\gamma ϵa_0-c_{\tau }$, slowing tau progression. If an anti-tau antibody independently reduces $k_{\tau }$ to $k_{\tau }^{\prime }<k_{\tau }$, the new tau-growth rate under both interventions is $k_{\tau }^{\prime }+\gamma ϵa_0-c_{\tau }$, which is less than the rate under either intervention alone by a strictly positive margin whenever $\gamma >0$. The slowdown in doubling time is therefore super-additive in the two interventions, which is the mathematical content of the empirical observation in current phase-2 combination trials that dual anti-amyloid-and-anti-tau therapy is more effective than the sum of monotherapies would predict. $\square$

Connections Master

• Chronic disease survey 35.03.01. This unit supplies the depth slice for the neurodegeneration entry in the chapter survey, building on the survey's account of timing-of-intervention as the dominant variable across cardiovascular, oncology, and neurodegeneration. The propagation model derived here is the precise reason that early intervention matters more in neurodegeneration than in cardiovascular or metabolic disease: neuron loss is a one-way ratchet in a way that atherosclerotic plaque reduction and glycemic control are not. The survey's identification of disease modification versus symptom management as the central clinical distinction is given its mathematical content by the threshold-crossing theorem above.

• Mutation and repair 17.06.01. The molecular substrate of protein aggregation is set by the polymerase-fidelity and DNA-repair machinery described there. Familial Alzheimer's (APP, PSEN1, PSEN2 mutations), familial Parkinson's (SNCA, LRRK2, GBA, PINK1, Parkin mutations), and familial ALS (SOD1, C9orf72, TDP-43 mutations) all arise from germline point mutations or repeat expansions in genes encoding aggregation-prone proteins or their quality-control machinery. The repair-deficiency disorders catalogued in 17.06.01 — ataxia telangiectasia, Cockayne syndrome, xeroderma pigmentosum — frequently feature neurodegeneration as a primary phenotype, because post-mitotic neurons are uniquely vulnerable to accumulated DNA damage that dividing cells can dilute by replication.

• Cellular neuroscience / action potential 17.09.02. The prion-like propagation model derived in the Key Result theorem requires a physical substrate for aggregate trafficking between neurons, and that substrate is the axonal transport system running along the same microtubule and fast-transport infrastructure whose electrical activity is described in 17.09.02. Tau is itself a microtubule-associated protein, so its pathology directly compromises the axonal transport on which neuronal connectivity depends, producing a feedback loop in which propagation both uses and degrades the connectome along which it spreads. This is the bridge between the cellular neuroscience of axonal function and the clinical phenomenology of regional neurodegeneration.

• Neuroscience of consciousness 20.06.04. The disorders of consciousness discussed there — coma, vegetative state, minimally conscious state — concern acute global disruption of neural-circuit function. Neurodegeneration is the chronic counterpart: progressive regional circuit disruption producing gradual loss of specific cognitive functions rather than acute loss of all consciousness. The connectome-level analysis developed in 20.06.04 for the neural correlates of consciousness is the same analytical framework applied here for the propagation of protein pathology, and the question of which circuit disruptions produce loss of specific cognitive functions versus loss of consciousness is given empirical content by comparing the regional selectivity of Alzheimer tau pathology against the global cortical devastation of prion disease.

Historical & philosophical context Master

Alois Alzheimer's 1907 report on Auguste D. ^{[Alzheimer1907]} established the disease that bears his name by combining clinical observation with the newly available Bielschowsky silver stain, which revealed both plaques and tangles in the same cortical tissue. James Parkinson's 1817 Essay on the Shaking Palsy ^{[Parkinson1817]} had described the motor syndrome a century earlier based on six cases observed in the streets of London; the substantia-nigra degeneration underlying the syndrome was identified by Tretiakoff in 1919, and the Lewy body by Friedrich Lewy in 1912. Creutzfeldt in 1920 and Jakob in 1921 described the spongiform encephalopathy that bears both their names; the transmissibility of kuru was demonstrated by Gajdusek in 1966, earning him a Nobel Prize, but the molecular agent remained mysterious.

The modern molecular era opened with two foundational results: Prusiner's 1982 prion hypothesis ^{[Prusiner1982]}, which established that an infectious agent could be a protein with no required nucleic acid; and Glenner and Wong's 1984 isolation and sequencing of amyloid-beta ^{[GlennerWong1984]}, which gave the molecular identity of the Alzheimer plaque. Goedert's 1988 cloning of tau ^{[Goedert1988]} and Spillantini's 1997 identification of alpha-synuclein in Lewy bodies ^{[Spillantini1997]} completed the molecular taxonomy of the major neurodegenerative proteinopathies. Hardy and Higgins's 1992 amyloid cascade hypothesis ^[Hardy1992] proposed a causal ordering within Alzheimer's disease, and the Braak-Braak 1991 staging scheme ^{[BraakBraak1991]} and its Parkinson analogue by Braak in 2003 ^[Braak2003] described the stereotyped anatomical spread that any mechanistic theory must explain.

The three-decade failure of amyloid-targeting clinical trials — punctuated by the controversial 2021 aducanumab approval — was resolved only with the Clarity AD lecanemab trial reported by van Dyck in 2023 ^{[vanDyck2023]}, the first unambiguously positive disease-modifying result. The intervening years produced a healthy skepticism of the amyloid cascade hypothesis that survives today in modified form: the hypothesis is partially correct, but the timing of intervention and the engagement of downstream tau pathology are the load-bearing variables.

Bibliography Master

@article{Alzheimer1907,
  author = {Alzheimer, A.},
  title  = {{\"U}ber eine eigenartige Erkrankung der Hirnrinde},
  journal = {Allgemeine Zeitschrift f{\"u}r Psychiatrie und psychisch-gerichtliche Medizin},
  volume = {64},
  pages  = {146--148},
  year   = {1907},
}

@book{Parkinson1817,
  author = {Parkinson, J.},
  title  = {An Essay on the Shaking Palsy},
  publisher = {Sherwood, Neely, and Jones},
  address = {London},
  year   = {1817},
}

@article{Prusiner1982,
  author = {Prusiner, S. B.},
  title  = {Novel proteinaceous infectious particles cause scrapie},
  journal = {Science},
  volume = {216},
  number = {4542},
  pages  = {136--144},
  year   = {1982},
  doi    = {10.1126/science.278.5336.245},
}

@article{GlennerWong1984,
  author = {Glenner, G. G. and Wong, C. W.},
  title  = {Alzheimer's disease: initial report of the purification and
            characterization of a novel cerebrovascular amyloid protein},
  journal = {Biochemical and Biophysical Research Communications},
  volume = {120},
  number = {3},
  pages  = {885--890},
  year   = {1984},
  doi    = {10.1016/S0006-291X(84)80190-4},
}

@article{Goedert1988,
  author = {Goedert, M. and Wischik, C. M. and Crowther, R. A. and
            Walker, J. E. and Klug, A.},
  title  = {Cloning and sequencing of the {cDNA} encoding a core protein
            of the paired helical filament of {Alzheimer} disease:
            identification as the microtubule-associated protein tau},
  journal = {Proceedings of the National Academy of Sciences USA},
  volume = {85},
  number = {11},
  pages  = {4051--4055},
  year   = {1988},
  doi    = {10.1073/pnas.85.11.4051},
}

@article{Spillantini1997,
  author = {Spillantini, M. G. and Schmidt, M. L. and Lee, V. M.-Y. and
            Trojanowski, J. Q. and Jakes, R. and Goedert, M.},
  title  = {$\alpha$-synuclein in {Lewy} bodies},
  journal = {Nature},
  volume = {388},
  number = {6645},
  pages  = {839--840},
  year   = {1997},
  doi    = {10.1038/42166},
}

@article{Hardy1992,
  author = {Hardy, J. A. and Higgins, G. A.},
  title  = {Alzheimer's disease: the amyloid cascade hypothesis},
  journal = {Science},
  volume = {256},
  number = {5054},
  pages  = {184--185},
  year   = {1992},
  doi    = {10.1126/science.1566067},
}

@article{BraakBraak1991,
  author = {Braak, H. and Braak, E.},
  title  = {Neuropathological stageing of {Alzheimer}-related changes},
  journal = {Acta Neuropathologica},
  volume = {82},
  number = {4},
  pages  = {239--259},
  year   = {1991},
  doi    = {10.1007/BF00308809},
}

@article{Braak2003,
  author = {Braak, H. and Del Tredici, K. and R{\"u}b, U. and
            de Vos, R. A. I. and Jansen Steur, E. N. H. and Braak, E.},
  title  = {Staging of brain pathology related to sporadic {Parkinson}'s disease},
  journal = {Neurobiology of Aging},
  volume = {24},
  number = {2},
  pages  = {197--211},
  year   = {2003},
  doi    = {10.1016/S0197-4580(02)00065-9},
}

@article{vanDyck2023,
  author = {van Dyck, C. H. and Swanson, C. J. and Aisen, P. and
            Bateman, R. J. and Chen, C. and Gee, M. and Kanekiyo, M. and
            Li, D. and Reynish, E. and Sandrock, A. and Subbian, V. and
            Rajdev, S. and Cummings, J. and {et al.}},
  title  = {Lecanemab in early {Alzheimer}'s disease},
  journal = {New England Journal of Medicine},
  volume = {388},
  number = {1},
  pages  = {9--21},
  year   = {2023},
  doi    = {10.1056/NEJMoa2212948},
}

@article{BrundinMelkiKopito2010,
  author = {Brundin, P. and Melki, R. and Kopito, R.},
  title  = {Prion-like transmission of protein aggregates in
            neurodegenerative diseases},
  journal = {Nature Reviews Molecular Cell Biology},
  volume = {11},
  number = {4},
  pages  = {301--307},
  year   = {2010},
  doi    = {10.1038/nrm2873},
}