Abstract

The brain is expensive tissue. It must maintain electrical gradients, recycle transmitters, move cargo, remodel connections, regulate blood flow, and coordinate immune defense without stopping.

A brain can preserve basic function while losing the reserve needed for learning, stress, sleep loss, infection, or aging. Here, brain energy margin means the balance between network support and network demand—not ATP concentration alone. The Fructose Model proposes that systemic metabolic disease, fragile vessels, and locally produced fructose may converge on that shrinking margin. It also asks a more provocative question: could KHK help place the brain into an energy-triage state—temporarily favoring food-seeking and immediate survival over memory, restraint, and long-range thought?

This does not reduce Alzheimer's disease to sugar. It places fructose/KHK on a broader map of brain vulnerability that can now be tested.

1. The Brain's Energy Margin

Neurons signal; astrocytes manage substrates and transmitters; microglia respond to injury; oligodendrocytes maintain myelin; vessels regulate delivery. Brain function depends on all of them coordinating under changing demand.

Resting ATP can look adequate while synaptic reserve or recovery is poor. The more revealing question is how well the network responds when demand rises—and how completely it returns afterward. Energetic, vascular, trophic, immune, and signaling support must together keep pace with the work and stress imposed on the network.

Sleep, oxygen, blood flow, glucose regulation, mitochondrial function, and inflammation all shape that margin. Fructose biology can intersect with each.

2. Fructose Made in the Brain

Dietary fructose is handled substantially by the intestine and liver. The stronger brain hypothesis is local production from glucose through the polyol pathway.

In a small human spectroscopy experiment, a brain fructose-compatible signal increased during several hours of controlled hyperglycemia even though plasma fructose did not explain the rise. This supports the ability of the human brain to make fructose acutely from glucose. [NEURO-H2017]

In a diabetic mouse model, hippocampal sorbitol, fructose, and KHK rose alongside lower ATP. Reducing KHK in the hippocampus improved selected redox, mitochondrial-protein, neural, and cognitive outcomes. [NEURO-LI2023]

Together, these findings provide a credible bridge: high glucose can create local fructose, and local KHK can matter in a defined disease model. They do not yet tell us which human brain cells carry the most flux or how much it contributes to long-term neurodegeneration.

That is already a meaningful advance over a purely dietary argument. It suggests that years of poor glucose regulation could expose the brain to fructose biology from within, even when little dietary fructose crosses directly into brain tissue.

3. KHK as a Candidate Energy-Triage Switch

Dr. Richard Johnson's Alzheimer's hypothesis adds a crucial layer to this story. It does not treat cerebral fructose metabolism as random damage alone. It proposes that the pathway may help coordinate a short-term survival state. [NEURO-J2020] [NEURO-J2023]

The logic is simple. When food or water is scarce, the brain does not need every function operating at full power. It needs attention directed toward finding resources. A temporary program that heightens hunger, makes food cues harder to ignore, relaxes restraint, and reduces energy spent on less urgent work could improve the odds of getting through the immediate problem.

KHK is a plausible biochemical contributor to that program because it can change cellular energy state and behavior-facing signals. Its rapid use of ATP and phosphate lowers the cell's immediate energy state. AMP disposal and uric-acid production can make ATP recovery harder while adding redox, inflammatory, and mitochondrial pressure. Whether those cellular changes actually reallocate human brain function away from memory formation, sustained attention, and executive control and toward appetite, reward, and foraging remains an integrative hypothesis.

Human feeding studies point in the same direction. Compared with glucose, fructose produced less suppression of appetite-related brain activity and less satiety in one small crossover study. In another, it produced stronger responses to food cues, greater hunger and desire for food, and a greater willingness to trade a delayed monetary reward for immediate high-calorie food. [NEURO-P2013] [NEURO-L2015]

These studies did not manipulate brain KHK, so they do not prove that KHK caused the response. They show that fructose can shift human brain and appetitive behavior in the direction the triage hypothesis predicts. The direct KHK evidence remains preclinical.

The pathway is also unlikely to act as a single dimmer switch across the whole brain. Fructose-handling machinery has been mapped to selected regions, including the hippocampus and cortex. [NEURO-O2017] That raises a sharper possibility: the brain may be reprioritizing networks, preserving the circuits most useful for immediate resource-seeking while allowing higher-order functions to become temporarily less dominant.

4. When Triage Becomes a Trap

A brief shift in priorities is not Alzheimer's disease. The proposed danger is repeated activation without complete recovery.

Bredesen's 2026 network theory offers a broader frame: neurodegeneration may emerge when the support available to a neural network becomes insufficient for its combined functional and defensive demands. [NEURO-B2026] In that architecture, energetic supply, blood flow, trophic and hormonal signals, inflammation, toxins, infection, and proteostatic work can interact rather than compete as single causes. The paper is a hypothesis synthesis, not direct validation of this Fructose Model, and it does not test fructose, KHK, F1P, or ATP. KHK therefore enters here as one proposed pathway-specific amplifier within a broader support-to-demand problem.

If KHK repeatedly draws down ATP and phosphate while uric acid sustains mitochondrial and inflammatory pressure, the brain can become less able to restore the energy it has spent. The next challenge then arrives before the previous one has been fully repaid. What was useful as a temporary state can become a chronic operating condition.

The combined model predicts a three-stage continuum:

1. Compensation: support is still sufficient for ordinary function, but reserve is recruited more often and recovery takes longer.
2. Recurrent functional insufficiency: demand repeatedly exceeds available support, producing reversible or fluctuating network dysfunction before major structural loss.
3. Structural progression: persistent insufficiency, impaired repair, and interacting pathologies increasingly damage synapses, cells, and circuits.

This staging is a testable hypothesis, not an established natural history of Alzheimer's disease.

That is where the model gains explanatory reach. Early cerebral glucose hypometabolism, mitochondrial dysfunction, insulin resistance, reduced blood flow, neuroinflammation, amyloid, tau, and other disease processes may be mutually reinforcing contributors to a network support-to-demand imbalance rather than interchangeable signs of one cause. Postmortem Alzheimer's tissue has also shown higher glucose and polyol-pathway intermediates, including sorbitol and fructose, although this does not establish that the pathway initiated the disease or resolve how much KHK contributes. [NEURO-X2016]

The model does not ask us to choose between energy failure and familiar Alzheimer's pathology. A brain with less ATP-producing and recovery capacity may be less able to maintain synapses, regulate immune cells, clear damaged material, preserve blood flow, and withstand amyloid, tau, vascular injury, or infection. The energy state may help determine whether those stresses are tolerated, contained, or allowed to spread.

5. Two Routes to a Fragile Brain

Fructose biology may reach the brain by two routes.

The systemic route: liver fat, insulin resistance, hypertension, kidney disease, sleep apnea, and vascular dysfunction change the blood, oxygen, and signals arriving at the brain. Even modest local fructose metabolism could become important in a network already receiving poor support.

The local route: endogenous fructose reaches KHK, drawing down ATP and phosphate while generating uric-acid/redox pressure. In susceptible cells, that could narrow the energy available for signaling, repair, and inflammatory resolution.

The systemic route is likely broader. The local route is a more specific research target. Both fit the same cellular-energy model.

They can also reinforce one another. Metabolic disease can impair vessels and raise glucose; impaired delivery and high glucose can increase local stress; a less resilient brain may then regulate appetite, sleep, and autonomic function less effectively, feeding pressure back to the rest of the body.

6. Metabolic Disease and Neurodegeneration

Type 2 diabetes and metabolic dysfunction are associated with higher dementia risk. Vascular disease, inflammation, altered insulin signaling, and mitochondrial stress can all contribute. [NEURO-J2023]

The phrase “type 3 diabetes” is memorable but incomplete. Alzheimer's disease also involves amyloid, tau, immune responses, synaptic loss, vascular injury, and genetics. The Fructose Model offers a connecting question rather than a replacement diagnosis:

Does lower metabolic and vascular reserve make the brain less able to tolerate and repair these other disease processes?

If so, the model predicts that improving support and recovery could change the terrain on which neurodegeneration progresses, even when it does not remove every initiating cause.

This may help explain why apparently different risks—diabetes, hypertension, sleep disruption, inactivity, and vascular disease—converge on cognitive decline. They need not share one origin to subtract from the same reserve.

7. What Should Be Tested

The next research questions are concrete:

• Which human brain cells produce and metabolize fructose under hyperglycemia?
• Does local F1P or KHK activity track cognitive or vascular recovery?
• Can a preregistered support-to-demand index—combining perfusion, metabolic recovery, inflammatory load, and task demand—predict functional decline better than any single resting marker?
• Does blocking KHK improve recovery only in people, tissues, or models with demonstrably elevated pathway activity?
• After current conditions are matched, does prior KHK activation leave a slower recovery response than an otherwise equivalent unexposed state?
• Does blocking KHK change the predicted balance between executive-memory networks and appetite-reward networks?
• Do sleep, exercise, glucose control, or pathway inhibition change the recovery curve?
• Is the pathway important only in a metabolically defined subgroup?

These questions are narrower than saying fructose causes dementia—and much more useful.

They also preserve hope. A vulnerability factor is not destiny. If part of brain risk comes from the metabolic terrain rather than an irreversible initiating lesion, improving that terrain may extend the period in which neural networks can compensate and repair.

Conclusion

The brain sits at the end of every metabolic supply line. It inherits the health of the liver, vessels, kidneys, sleep, and whole-body energy system.

Fructose/KHK may narrow brain-energy margin both indirectly and locally. Johnson's hypothesis proposes that KHK could help triage brain energy toward immediate resource-seeking. Bredesen's broader network theory clarifies the surrounding terrain: functional and then structural failure may emerge when support becomes insufficient for combined network demands. These are overlapping hypotheses, and neither establishes that chronic cerebral KHK activation causes Alzheimer's disease.

That proposed function is not yet settled human biology. But it connects appetite, behavior, glucose hypometabolism, mitochondrial dysfunction, inflammation, and cognitive decline in a way that is specific enough to test. A brain with less reserve is more vulnerable, and restoring recovery may be as important as removing any single insult.

Sources are linked inline; full citations and evidence boundaries are available in the Master Bibliography.