Your Brain Can Become Diabetic

Alzheimer's & Metabolic Science

Your Brain Can
Become Diabetic —
And That May Be What's
Driving Alzheimer's

For decades, scientists hunted a protein. The cures didn't come. A new framework says we may have been looking at the wrong problem entirely.

Rick · Entropy Lab, Bluewhale AI Inc. June 2026 · 8 min read

The problem

Why Can't We Cure Alzheimer's?

Every three seconds, someone in the world is diagnosed with dementia. Alzheimer's disease accounts for roughly 70% of those cases. Decades of research and billions of dollars later, we still don't have a single drug that stops the disease in its tracks.

For a long time, scientists thought they knew the culprit: sticky protein clumps called amyloid plaques building up in the brain and killing neurons. Dozens of pharmaceutical companies built drugs to dissolve those clumps. In clinical trials, the plaques disappeared. The patients' memories didn't come back.

"We caught the suspect. Why didn't the crime stop?"

A growing body of research now points to a different answer. Alzheimer's may not primarily be a problem of protein buildup — it may be a problem of the brain's entire energy system breaking down. And at the center of that breakdown is something called brain insulin resistance.

The mechanism

Insulin Isn't Just About Blood Sugar — Not in the Brain

When most people hear "insulin," they think of diabetes and blood sugar. The classic story: you eat, blood sugar rises, the pancreas releases insulin, insulin unlocks cells so they can absorb glucose. That's accurate — for the rest of your body.

The brain works differently. Brain neurons don't need insulin to absorb glucose. They have their own always-open glucose transporters (GLUT3) that work around the clock, independent of insulin. So if the brain can get its fuel without insulin, what is insulin actually doing up there?

It's managing the construction and maintenance of memory. Insulin in the brain supports the growth of new synapses — the connections between neurons that form the physical structure of memory and learning. It strengthens existing connections, protects neurons from stress, and keeps energy production running efficiently. Think of it less like a fuel deliverer and more like a site manager overseeing the brain's renovation and repair crew.

Brain insulin resistance is what happens when neurons stop listening to that site manager. Insulin shows up, knocks on the door, and gets no answer. The instructions for building memories, protecting cells, and maintaining energy never get through.

In 2005, neuroscientist Suzanne de la Monte gave this condition a name that raised eyebrows: "Type 3 Diabetes." Type 1 is when the body can't make insulin. Type 2 is when the body ignores insulin. Type 3 — her proposal — is when the brain specifically ignores insulin. She found this state widespread in Alzheimer's patients' brains.

Two disruptions

How the Problem Starts

When insulin docks onto a neuron's surface receptor, it triggers a chain of signals inside the cell — like pressing a doorbell that rings through the whole house. The main relay is a protein called IRS-1, and the key switch it controls is called Akt. When Akt is on, memory formation, cell survival, and energy production all function properly.

In Alzheimer's disease, this chain gets disrupted at two points simultaneously.

First, amyloid. The amyloid beta (Aβ) protein clusters physically attach to the insulin receptor and block it. Even when insulin is present in normal amounts, the receptor can't respond. It's like someone stuffing gum into a lock: the key is there, but the door won't open.

Second, tau. When tau misbehaves, it activates an enzyme (PTP1B) that immediately switches off the insulin signal the moment it starts — like cutting the power line to the doorbell right as someone answers it.

The result: neurons that are effectively deaf to insulin, even when insulin is present.

The self-amplifying loop

When the Power Plant Catches Fire

Disrupted insulin signaling sets off a cascade of damage. The most critical link in that chain is mitochondrial failure. Mitochondria are the cell's power plants — they convert glucose and oxygen into ATP, the energy currency that powers everything a cell does. The brain is only 2% of body weight but consumes 20% of the body's energy.

When insulin signaling drops, the signal that builds new mitochondria (PGC-1α) falls as well. Fewer power plants, lower output. Energy production in affected neurons drops by 40–50%.

But in 2026, a study by Li and colleagues added something new and alarming. They found that abnormal tau protein physically enters mitochondria and latches onto a core component of the main power-generating machinery — a protein called NDUFS3, part of Complex I, the first step in the cell's electron assembly line.

When tau grabs NDUFS3, the electron flow reverses direction. This — called Reverse Electron Transport (RET) — causes a sudden explosive release of free radicals: unstable molecules that attack proteins, membranes, and DNA. Think of it as sparks flying inside the power plant itself.

What makes this truly dangerous is that the sparks feed the fire that made them. Below is that loop — press Run to watch it unfold.

The Tau–NDUFS3–RET Self-Amplifying Loop

Tau protein misbehaves

Enters mitochondria · binds NDUFS3

Complex I reversal → RET triggered

Free radical explosion (ROS ↑↑)

Insulin resistance worsens further

More tau goes wrong → loop repeats

↺ Self-amplifying · independent of amyloid

This loop explains something that has puzzled clinicians for years: why Alzheimer's, once it reaches a certain point, seems to accelerate on its own — and why removing amyloid plaques doesn't stop it. The loop runs independently. It doesn't need the plaques to keep spinning.

Synapses

When the Connections Go Dark

Energy-starved neurons face one more critical problem: they can't maintain their synapses. A synapse is the junction between two neurons — the physical point where information passes from one cell to another. Memory is, at its most basic level, a pattern of synaptic connections.

Here's a detail that often surprises people: in Alzheimer's research, the amount of amyloid plaque in a patient's brain correlates only loosely with how severe their cognitive symptoms are. But the number of lost synapses correlates very strongly. Some patients with heavy plaque loads show surprisingly mild symptoms; others with fewer plaques are severely impaired. The difference often comes down to how many connections remain.

Plaque is not the whole story. Synapse loss is closer to the real story — and insulin resistance is one of its primary causes.

What to do

What Can Actually Be Done

Brain insulin resistance is not fixed fate. It develops over years, shaped by choices that are largely within our control.

🏃

Exercise

Aerobic exercise restores insulin sensitivity, builds new mitochondria, and raises BDNF — the brain's synaptic growth factor.

Strongest evidence

🥑

Diet & Ketones

Ketone bodies bypass the broken Complex I machinery entirely — running the backup generator when the main plant is damaged.

Metabolic bypass

😴

Sleep

Deep sleep flushes amyloid via the glymphatic network. Poor sleep raises cortisol and accelerates insulin resistance.

Clearance + repair

💉

Emerging Drugs

Intranasal insulin, Complex I RET inhibitors, and NAD⁺ precursors (NR/NMN) represent a fundamentally different approach than chasing amyloid.

In development

Final thought

A Different Way to See Alzheimer's

Alzheimer's is not simply bad luck — a protein misfiring in an otherwise healthy brain. It is, in many cases, the long-term consequence of how a brain was fueled, maintained, and protected over decades. The tau-mitochondria loop doesn't appear overnight. It is the culmination of years of compounding insults: chronic insulin resistance, mitochondrial stress, synaptic erosion, inflammation.

The sobering part: prevention has to start earlier than most people think — in midlife, perhaps even in one's thirties and forties. The hopeful part: the main risk factors are things we can actually do something about. Movement, food, sleep. None of these are exotic interventions. They are the ordinary architecture of a daily life.

Alzheimer's may not be curable yet. But it may be, at least in part, preventable — and that changes everything about how we should be thinking about it.

Based on: Talbot et al. (2012, J Clin Invest); de la Monte & Wands (2005); Li et al. (2026); Arnold et al. (2018, Nat Rev Neurol); Verdin (2015, Science).

Your Brain CanBecome Diabetic —And That May Be What'sDriving Alzheimer's