Scientists discover metformin controls blood sugar through the brain, not just the liver

Overview

Updated May 30

More than 150 million people take metformin every year to manage Type 2 diabetes. Doctors prescribed it starting in 1957, and for most of that time, assumed it worked primarily in the liver and gut. A Baylor team found that at clinically relevant low doses, the drug deactivates Rap1 in the ventromedial hypothalamus—a brain region controlling metabolism.

Here's why: a brain pathway governs metformin's effect at far lower doses than the liver requires. Scientists can now design drugs targeting that pathway, bypassing the gastrointestinal side effects that cause patients to quit. This also confirms a trend in diabetes research: the brain controls blood sugar far more than was previously thought.

Why it matters

The most-prescribed diabetes drug on Earth works through a mechanism nobody knew about, opening a new class of potential treatments.

Questions about this story

Free account needed to ask — your question is kept and asked for you right after sign-up. Answers are public.

No questions yet — be the first to ask.

Key Indicators

150M+

Annual metformin prescriptions worldwide

Metformin is the most widely prescribed oral diabetes medication globally

69 years

Years in clinical use before brain pathway identified

First prescribed in France in 1957; brain mechanism published in 2025

~537M

People living with Type 2 diabetes worldwide

The global diabetes population for whom this discovery is relevant

Lower

Drug concentration needed via brain pathway vs. liver

The brain pathway responds to metformin at much lower doses than the liver or intestines require

Voices

Curated perspectives — historical figures and your fellow readers.

Ever wondered what historical figures would say about today's headlines?

People Involved

Makoto Fukuda

Lead researcher, Associate Professor at Baylor College of Medicine

Active researcher continuing work on brain-metabolism connections

Yong Xu

Co-author, Professor at Baylor College of Medicine

Active researcher in hypothalamic neuroscience

Jean Sterne

French physician who first used metformin to treat diabetes

Deceased

Organizations Involved

Baylor College of Medicine

Academic Medical Institution

Produced the landmark metformin-brain pathway study

A private medical school and research institution in Houston, Texas, whose Children's Nutrition Research Center houses the Fukuda lab that discovered metformin's brain pathway.

Timeline

January 1918 March 2026

9 events Latest: March 25th, 2026 · 6 months ago

Tap a bar to jump to that date

March 2026
Discovery gains widespread media attention
Latest Media Coverage

Major science outlets including ScienceDaily and SciTechDaily publish in-depth coverage of the Baylor team's findings, bringing the brain-pathway discovery to a broad public audience and sparking discussion about next-generation diabetes treatments.
March 25th, 2026
July 2025
Landmark paper connects metformin to brain Rap1 pathway
Publication

The study "Low-dose metformin requires brain Rap1 for its antidiabetic action" is published in Science Advances, demonstrating that metformin at clinically relevant low doses works by deactivating Rap1 in VMH neurons. Mice lacking brain Rap1 do not respond to low-dose metformin.
July 30th, 2025
June 2021
Rap1 in VMH shown to regulate glucose homeostasis
Research

Fukuda's team publishes in JCI Insight that deleting Rap1 specifically in SF1-positive neurons of the ventromedial hypothalamus markedly lowers blood glucose and improves insulin tolerance in mice — the first direct evidence that this brain protein controls blood sugar.
June 22nd, 2021
October 2016
Fukuda lab identifies brain protein Rap1 as obesity target
Research

Makoto Fukuda's team at Baylor College of Medicine publishes early work showing that the protein Rap1 in the hypothalamus plays a role in regulating body weight, laying groundwork for the later metformin connection.
October 1st, 2016
January 2001
AMPK activation identified as metformin's mechanism
Research

Researchers report that metformin activates the enzyme AMP-activated protein kinase (AMPK) in liver cells, establishing what becomes the dominant explanation for how the drug works for the next two decades.
January 1st, 2001
January 1998
UK study confirms metformin's cardiovascular benefits
Research

The UK Prospective Diabetes Study shows metformin reduces cardiovascular risk in overweight diabetic patients, cementing its status as the first-line treatment worldwide.
January 1st, 1998
January 1995
FDA approves metformin for use in the United States
Regulatory

The United States Food and Drug Administration approves metformin nearly four decades after its European debut. It rapidly becomes the most-prescribed oral diabetes drug in the country.
January 1st, 1995
January 1957
Jean Sterne introduces metformin as a diabetes drug in France
Clinical Milestone

French physician Jean Sterne publishes the first report of metformin treating diabetes, branding it "Glucophage." It enters clinical use in Europe but remains unknown in the United States for decades.
January 1st, 1957
January 1918
Guanidine's blood sugar effects discovered
Discovery

Researchers identify that guanidine, a compound found in the French lilac plant (Galega officinalis), lowers blood glucose in animals — the chemical ancestor of metformin.
January 1st, 1918

Scenarios

New drugs targeting brain Rap1 enter clinical trials for Type 2 diabetes

Possible

Discussed by: Baylor College of Medicine researchers; pharmacology commentators at Technology Networks and MedicalXpress

If the Rap1 pathway proves targetable with a purpose-built molecule, pharmaceutical companies could develop diabetes drugs that work at lower doses with fewer gastrointestinal side effects than metformin. The proof-of-concept — that deactivating a single brain protein at low drug concentrations controls blood sugar — makes this an attractive target. Early-stage drug discovery programs could emerge within one to two years, with clinical trials following. This would represent the first new class of brain-targeted diabetes drugs.

Metformin dosing guidelines revised to reflect brain-pathway mechanism

Possible

Discussed by: Diabetes treatment guideline bodies (American Diabetes Association, European Association for the Study of Diabetes); clinical pharmacologists

If follow-up human studies confirm that metformin's brain pathway is active at lower doses than currently prescribed, clinical guidelines could shift toward lower starting doses. This could reduce the gastrointestinal side effects that cause roughly 5-10% of patients to discontinue the drug. However, translating mouse-model findings to human dosing recommendations typically requires years of clinical validation.

Discovery reshapes metformin's position against GLP-1 drugs

Uncertain

Discussed by: American Diabetes Association (2026 Standards of Care); clinical endocrinologists; pharmaceutical analysts covering GLP-1 market

Metformin's first-line status is already under pressure from GLP-1 receptor agonists like semaglutide and tirzepatide, which offer cardiovascular and weight-loss benefits. Understanding exactly how metformin works could either bolster its position — by enabling optimized dosing or combination therapies that exploit the brain pathway — or accelerate its displacement if the brain mechanism proves less potent than GLP-1 effects. The competitive dynamic between a $0.05/day generic and $1,000/month GLP-1 drugs adds economic urgency to the scientific question.

Brain-metabolism research expands beyond diabetes to obesity and neurodegeneration

Likely

Discussed by: Neuroscience and metabolism researchers; GLP-1 investigators studying Alzheimer's applications

The finding that a widely used metabolic drug works through the brain strengthens the broader hypothesis that brain circuits are central to metabolic disease. This could accelerate research into brain-targeted treatments for obesity, metabolic syndrome, and even neurodegenerative diseases where metabolic dysfunction plays a role. Metformin is already under study for potential anti-aging and neuroprotective effects — a confirmed brain mechanism could sharpen those investigations.

Historical Context

3 moments from history that rhyme with this story — and how they unfolded.

1949–present

Lithium for bipolar disorder — still partially unexplained after 75 years (1949–present)

Australian psychiatrist John Cade discovered in 1949 that lithium calmed manic patients, and it was approved by the FDA for bipolar disorder in 1970. Despite 75 years of clinical use and extensive research, lithium's full mechanism of action remains incompletely understood. Proposed explanations include inhibition of glycogen synthase kinase-3 (GSK-3), effects on inositol signaling, and modulation of neurotransmitter systems.

Then

Lithium became the gold standard for bipolar disorder treatment despite the mechanistic mystery, prescribed purely on the basis of clinical effectiveness.

Now

The persistent knowledge gap has made it difficult to develop better-tolerated alternatives. Lithium's narrow therapeutic window and significant side effects remain clinical challenges that a clearer mechanism might help solve.

Why this matters now

Lithium shows what happens when a drug's mechanism stays unclear: it remains effective but difficult to improve upon. Metformin's story could have gone the same way — the Rap1 discovery breaks that pattern and provides a concrete molecular target for next-generation drugs.

1971

Aspirin's mechanism discovered 70 years after first use (1971)

Aspirin had been used since 1899 to reduce pain, fever, and inflammation, but nobody knew how it worked. In 1971, British pharmacologist John Vane showed that aspirin inhibits cyclooxygenase (COX) enzymes, blocking the production of prostaglandins — signaling molecules that drive inflammation and pain. The discovery earned Vane a share of the 1982 Nobel Prize in Physiology or Medicine.

Then

Vane's finding immediately explained aspirin's anti-inflammatory, antipyretic, and analgesic effects in a single mechanism, transforming it from an empirical remedy into a rationally understood drug.

Now

The discovery led directly to the development of COX-2 selective inhibitors (like celecoxib) in the 1990s, designed to reduce pain without aspirin's stomach-bleeding side effects. It also unlocked aspirin's use as a cardiovascular preventive by explaining its antiplatelet action.

Why this matters now

The closest parallel to metformin's story: a ubiquitous, decades-old drug whose mechanism was finally explained, immediately opening new drug-design possibilities. If the Rap1 discovery follows aspirin's trajectory, it could similarly spawn an entire class of targeted therapies.

1982–2005

Helicobacter pylori and ulcers — overturning decades of medical consensus (1982–2005)

For decades, doctors attributed stomach ulcers to stress and diet. In 1982, Australian researchers Barry Marshall and Robin Warren identified the bacterium Helicobacter pylori as the actual cause. The medical establishment resisted for over a decade — Marshall famously drank a petri dish of the bacteria to prove his point. Consensus shifted gradually through the 1990s.

Then

Marshall developed a painful gastric infection that proved his hypothesis, leading to publication and slow clinical adoption of antibiotic-based ulcer treatment.

Now

Marshall and Warren won the 2005 Nobel Prize. Ulcer treatment shifted from symptom management (antacids, surgery) to a curative antibiotic regimen, dramatically reducing ulcer recurrence and the need for stomach surgery.

Why this matters now

Like the metformin discovery, this was a case where the actual mechanism of a common condition turned out to be fundamentally different from what the medical establishment assumed. The lesson: when the real mechanism is found, treatment can shift from managing symptoms to targeting root causes.