A Single Neurotransmitter May Hold the Key to Alzheimer's Disease

The Molecule That Forgot Itself

In the 1970s, a quiet observation began to unsettle neuroscientists. When they autopsied the brains of people who had died with Alzheimer’s disease, they found something missing. Deep in the basal forebrain, a region responsible for attention and memory, the neurons that produce acetylcholine were dying off in staggering numbers. The loss was not subtle. It was catastrophic. And it was specific: other types of neurons appeared relatively unharmed.

For decades, this observation formed the backbone of Alzheimer’s research. The cholinergic hypothesis, as it came to be called, held that the disease was fundamentally a failure of acetylcholine signaling. But then something strange happened. In the 1990s and 2000s, the field pivoted hard toward two other suspects: amyloid plaques and tau tangles. Billions of dollars flowed into drugs targeting these proteins. And one by one, those drugs failed.

Now, a growing number of researchers are circling back to the old idea. Not because they were wrong before, but because they were incomplete. A 2022 review by Zhi-ru Chen, Jiabao Huang, Shu-Long Yang, and Fen-Fang Hong, published in the journal Molecules, makes a compelling case that acetylcholine is not just a bystander in Alzheimer’s disease. It may be the central player that connects everything else (Chen et al., 2022).

What Acetylcholine Actually Does That Matters

Acetylcholine is not a glamorous neurotransmitter. Dopamine gets the headlines for pleasure and reward. Serotonin gets the credit for mood. Noradrenaline gets the blame for fight-or-flight. Acetylcholine is the workhorse of attention. It is the molecule that says: Pay attention to this, not that.

Chen and colleagues describe cholinergic neurons as the brain’s “attentional gatekeepers.” These neurons project from the basal forebrain to the hippocampus and cortex, regions critical for forming new memories and retrieving old ones. When acetylcholine release is normal, you can focus. You can encode a conversation, remember where you parked, learn a new name. When acetylcholine drops, the gate stays open too wide or too narrow. Irrelevant noise floods in. Important signals slip away.

The authors note that this decline in cholinergic signaling correlates tightly with the earliest symptoms of Alzheimer’s disease: forgetfulness, confusion, difficulty concentrating. In fact, the loss of cholinergic neurons in the basal forebrain is one of the earliest detectable changes in the disease, preceding widespread amyloid plaque deposition (Chen et al., 2022). This is not a late-stage artifact. It is an opening move.

The Acetylcholine Tau Connection

Here is where the story gets more interesting. For years, the amyloid hypothesis dominated thinking about Alzheimer’s. The idea was simple: sticky beta amyloid peptides clump into plaques, those plaques trigger a cascade of damage, and tau proteins form tangles inside neurons. The field treated amyloid as the trigger and tau as the bullet.

But Chen and colleagues point to a different sequence. Their review synthesizes evidence that acetylcholine depletion itself can drive tau pathology. When cholinergic signaling is disrupted, an enzyme called glycogen synthase kinase 3 beta (GSK3B) becomes hyperactive. GSK3B, in turn, phosphorylates tau protein excessively, causing it to detach from microtubules and form those characteristic tangles (Chen et al., 2022).

This is a mechanistic link that flips the causal story. It suggests that cholinergic loss is not just a consequence of Alzheimer’s pathology. It may be a driver of tau pathology. The authors write that “abnormal central cholinergic changes can also induce abnormal phosphorylation of tau protein” (Chen et al., 2022). In plain language: when acetylcholine goes quiet, tau goes rogue.

If this is correct, it changes how we think about intervention. Instead of targeting amyloid plaques that may form downstream of cholinergic failure, we might need to protect or restore cholinergic neurons early, before tau tangles become irreversible.

The Inflammation Loop Nobody Saw Coming

The second piece of the puzzle involves neuroinflammation. Alzheimer’s brains are inflamed. Microglia, the brain’s immune cells, become chronically activated and release toxic chemicals that damage neurons. For a long time, inflammation was seen as a secondary response to amyloid plaques or neuronal death.

But Chen and colleagues present evidence that cholinergic signaling directly suppresses inflammation. Acetylcholine binds to receptors on microglia and inhibits their release of pro inflammatory cytokines. This is called the cholinergic anti inflammatory pathway. When acetylcholine levels drop, the brake on inflammation is released. Microglia become hyperactive. They start attacking synapses and neurons.

The authors describe a feedback loop: cholinergic loss leads to inflammation, inflammation damages more cholinergic neurons, and those neurons die off, causing more inflammation. This loop may explain why Alzheimer’s disease accelerates once it begins (Chen et al., 2022).

The implication is that restoring cholinergic signaling could break this loop. Not just by improving memory, but by calming the immune system in the brain. This is a radically different mechanism from the amyloid targeting antibodies that have dominated clinical trials.

The Drug That Almost Worked

There is already a class of drugs that targets acetylcholine: cholinesterase inhibitors. Drugs like donepezil, rivastigmine, and galantamine work by preventing the breakdown of acetylcholine, thereby boosting its levels in the synapse. These are the most widely prescribed medications for Alzheimer’s disease.

They work. Sort of.

The authors note that these drugs produce modest improvements in cognition and daily function for some patients. The effect is real but limited. On average, they delay symptom progression by about six to twelve months. They do not stop the disease. They do not reverse it. And they do not work for everyone (Chen et al., 2022).

The reason, Chen and colleagues argue, is that cholinesterase inhibitors are a Band Aid. They boost acetylcholine that is already being released, but they do not protect the neurons that produce it. As those neurons continue to die, the drug has less and less substrate to work with. Eventually, the brain runs out of cholinergic neurons, and the drug stops working entirely.

What the field needs, the authors suggest, are therapies that protect cholinergic neurons from dying in the first place. Or therapies that stimulate the growth of new cholinergic neurons. Or therapies that target the downstream pathways triggered by cholinergic loss, like tau phosphorylation and inflammation.

What the Research Does Not Prove

This is where honest science journalism requires a pause. The cholinergic hypothesis is compelling, but it is not proven. Chen and colleagues are careful to state that “the pathogenesis of AD is complex and remains unclear, being affected by various factors” (Chen et al., 2022). The review synthesizes correlational and mechanistic evidence, but it does not report a single definitive experiment that proves acetylcholine depletion causes Alzheimer’s.

There are open questions that the review does not fully resolve:

• ▸Does cholinergic loss occur before amyloid deposition in all patients, or only in a subset? The evidence is strongest for sporadic late onset Alzheimer’s, but less clear for early onset genetic forms.
• ▸Can restoring acetylcholine levels in midlife prevent the disease, or is the window of intervention narrow?
• ▸Are there subtypes of Alzheimer’s where cholinergic dysfunction is not the primary driver? Some patients with significant amyloid pathology show minimal cholinergic loss.

The authors also note that clinical trials of cholinergic boosting agents have been disappointing overall. The modest effects of existing drugs may reflect the fact that we are intervening too late. Or it may mean that acetylcholine is only one piece of a larger puzzle.

Why This Matters Right Now

The timing of this review is not accidental. In 2023, the FDA granted full approval to lecanemab, an antibody that clears amyloid plaques. The drug slows cognitive decline by about 27% over 18 months. This is a real effect, but it is small. Many patients still progress to moderate or severe disease.

The amyloid hypothesis has dominated Alzheimer’s research for three decades. It has produced exactly one drug with marginal efficacy and a handful of spectacular failures. The cholinergic hypothesis offers an alternative framework that is older, less funded, and arguably more consistent with the clinical reality of the disease.

Chen and colleagues are not arguing that amyloid and tau are irrelevant. They are arguing that cholinergic signaling sits at the center of a network that includes amyloid, tau, and inflammation. Pull on any one thread, and the whole system moves. But if you want to understand why the system fails, you have to start at the center.

How the Study Was Done

The review by Chen et al. (2022) is a systematic synthesis of existing literature. The authors searched PubMed, Web of Science, and other databases for studies on cholinergic signaling and Alzheimer’s disease published up to 2021. They included both preclinical studies in animal models and clinical studies in humans. They did not conduct new experiments or meta analyses. Instead, they evaluated the strength of evidence for each link in the cholinergic cascade: from acetylcholine depletion to tau phosphorylation, from cholinergic loss to inflammation, from inflammation to neuronal death.

This is a standard methodology for a review article. The strength of the paper lies not in new data but in the clarity with which it connects disparate findings into a coherent model. The authors cite 773 other papers, which gives a sense of the breadth of evidence they are synthesizing.

What This Actually Means

The practical takeaways from Chen and colleagues’ review are not about a single miracle drug. They are about reorienting our approach to a disease that has frustrated every attempt at a cure.

• ▸Early detection of cholinergic loss could change treatment timelines. If cholinergic decline precedes amyloid plaques, then biomarkers of cholinergic function (like PET scans of cholinergic receptors) could identify people at risk years before symptoms appear. This would allow intervention during a window when the brain is still plastic enough to respond.

• ▸Cholinesterase inhibitors are not the final answer, but they are not useless either. They buy time. The question is whether we can combine them with neuroprotective agents that slow the death of cholinergic neurons. Several compounds are in preclinical development, including nerve growth factor gene therapy and drugs that block GSK3B.

• ▸Inflammation is a target, not just a side effect. The cholinergic anti inflammatory pathway suggests that boosting acetylcholine could calm microglial activation. This is a testable hypothesis. Clinical trials of cholinergic drugs combined with anti inflammatory agents could yield synergistic effects.

• ▸The amyloid hypothesis is not wrong, but it may be incomplete. Chen and colleagues’ model suggests that amyloid and tau pathology may be downstream consequences of cholinergic failure in some patients. Targeting amyloid alone may be like treating a fever without addressing the infection.

• ▸Personalized medicine matters more than we thought. Not every Alzheimer’s patient has the same degree of cholinergic loss. Some may have dominant amyloid pathology. Others may have dominant cholinergic pathology. Treatments may need to be matched to the individual’s specific biological subtype.

The cholinergic hypothesis is old. But it is not outdated. It has been waiting, quietly, for the rest of the field to catch up. Chen and colleagues have made a strong case that the molecule we forgot about may be the one we need to remember.