Coffee, Cocaine and your Brain

This article covers neuroscience research for general interest. It is not medical advice, and none of the disease-related research described below is an established treatment for any condition.

My mother can’t touch caffeine after lunch without lying awake half the night. I can drink a triple-shot espresso drink at 9 p.m. and be asleep by midnight without issue. The gap between us likely comes down to how our individual brains handle a couple of specific systems, and the research into those systems has genuinely interesting implications well beyond just explaining why some people can drink coffee at bedtime.

What DARPP-32 actually does

DARPP-32 is a protein your brain produces, found in both mice and humans, that acts as a kind of switchboard for other brain chemicals affecting mood, alertness, cognitive function, and sleep. One of its roles involves regulating adenosine, a chemical that builds up in your brain as you stay awake and mentally active, and that eventually triggers drowsiness once it accumulates enough. Caffeine works by blocking adenosine’s effects at a specific receptor, the adenosine A2A receptor, which is the entire mechanism behind why coffee keeps you feeling alert rather than sleepy. Cocaine and nicotine interact with overlapping brain reward pathways in their own, separate ways.

What the animal research found

Researchers in Sweden, and separately at the University of Rouen, used mice genetically engineered to lack functioning DARPP-32 and compared how they responded to caffeine against mice with normal DARPP-32 function. The genetically altered mice needed roughly twice as much caffeine to get the same stimulant effect as mice with the gene working normally, which is a real, measurable finding about how central this one protein is to caffeine’s effects.

The same altered mice showed up in separate cocaine-addiction research, where they were notably less motivated to work for cocaine-linked rewards compared to normal mice. That’s the finding that connects DARPP-32 to the broader mechanism of stimulant addiction generally, not just caffeine specifically.

Where this gets genuinely interesting: real human data on Parkinson’s

Here’s the part the original coverage of this research never got to, because the human epidemiology has matured significantly since. Multiple large observational studies, pooled in dose-response meta-analyses, consistently find that people who drink coffee or consume caffeine regularly have a meaningfully lower risk of developing Parkinson’s disease, a condition that itself involves the loss of dopamine-producing neurons. One meta-analysis found roughly a 25 percent lower risk of Parkinson’s among caffeine consumers overall, with the strongest protection, about a 28 percent reduction, showing up around 3 cups of coffee a day. The effect follows a real dose-response pattern (more caffeine, more protection, up to a point) and appears stronger in men than in women, an asymmetry researchers haven’t fully explained.

The proposed mechanism lines up directly with the adenosine biology above: blocking the A2A receptor is thought to help protect the specific dopamine-producing neurons in a brain region called the substantia nigra, the same neurons that progressively die off in Parkinson’s disease, from a kind of overexcitation damage.

A real drug came out of this, just not caffeine itself

This is the part that actually validates the mechanism without overstating it. In 2019, the FDA approved istradefylline, a drug that works by blocking the same adenosine A2A receptor caffeine blocks, as an add-on treatment for people with Parkinson’s disease experiencing “off” periods, stretches where standard medication temporarily stops controlling their symptoms well. It took more than two decades of research and clinical trials involving over 4,000 Parkinson’s patients to get there, and istradefylline was the first non-dopamine-based drug approved for Parkinson’s in twenty years. That’s a genuinely real, FDA-approved outcome of adenosine-receptor science, concrete proof the mechanism matters clinically. It is not caffeine, and it is not a cure; it is a targeted, purpose-built drug approved for one specific symptom pattern in people who already have the disease.

An important distinction: preventing Parkinson’s is not the same as treating it

Here’s the honest, slightly deflating nuance that the original excitement around this research glossed over. The lower-risk finding above is about whether caffeine consumption over years is associated with a lower chance of ever developing Parkinson’s in the first place. That’s a different question from whether caffeine helps someone who already has Parkinson’s. A randomized controlled trial specifically testing caffeine as a treatment for existing Parkinson’s motor symptoms, the Café-PD trial, found no meaningful improvement in motor symptoms compared to placebo, and noted caffeine might even slightly worsen certain symptoms like dyskinesia in some patients, while offering a mild, temporary alerting effect. Caffeine was well tolerated with no unusual safety concerns, it just didn’t work as a treatment the way the earlier prevention research might have led you to hope.

In plain terms: the observational, years-long association between coffee drinking and lower Parkinson’s risk is real and well-replicated. Caffeine as a treatment once someone already has the disease has been tested directly and didn’t pan out. Both findings can be true at once, and conflating them is exactly the kind of overreach that made the original version of this article promise more than the science actually delivered.

Why caffeine’s effects last so long for some people

Normally, a separate set of brain proteins works to shut DARPP-32 activity back down once you’ve been stimulated for a while, which is part of what lets a caffeine buzz eventually fade. Caffeine’s presence interferes with that shutdown process, creating something of a feedback loop that keeps the alert feeling going longer than caffeine’s simple presence in your bloodstream would predict on its own.

That’s not the only reason people vary so much, though. A separate, genuinely distinct mechanism also matters: the gene CYP1A2 controls how quickly your liver actually clears caffeine from your blood in the first place, independent of how your brain’s receptors respond to it. People with the fast-metabolizing version of this gene clear caffeine quickly and generally feel its effects fade sooner; people with the slower-clearing variant keep meaningful caffeine levels in their system for hours longer. Our guide to caffeine and athletic performance covers the CYP1A2 research in more depth. The two systems, DARPP-32’s receptor-level sensitivity and CYP1A2’s metabolism speed, are genuinely separate mechanisms that both contribute to why the same cup of coffee hits two different people so differently, which is likely a real part of the answer for my mother and me specifically.

Frequently asked questions

Does this research mean caffeine or DARPP-32 could cure Parkinson’s or schizophrenia?

No, and there’s now a clear real-world example showing exactly why not: a drug (istradefylline) targeting the same adenosine receptor pathway was approved in 2019, but it took over two decades of dedicated drug development to get there, it treats a specific symptom in people who already have Parkinson’s, and it is not caffeine. Basic research pointing to a mechanism is a long way from a usable treatment, even when the mechanism turns out to be genuinely important.

If coffee drinkers have lower Parkinson’s risk, why doesn’t caffeine help treat it?

Because preventing a disease from developing and treating it once it’s present are different biological questions, and this is a documented case where the answers diverge. Long-term coffee consumption is consistently associated with lower risk of developing Parkinson’s in observational research. A randomized trial testing caffeine specifically as a treatment for people who already have Parkinson’s found no meaningful benefit for motor symptoms.

Why do some people handle caffeine so much better than others?

Two separate, real mechanisms both contribute: how sensitive your brain’s adenosine receptors and systems like DARPP-32 are, and how quickly your CYP1A2 gene metabolizes caffeine out of your bloodstream in the first place. Both vary genuinely from person to person, and either can explain why the same dose hits two people very differently.

Sources

• Dose-response meta-analysis on coffee, tea and caffeine consumption with risk of Parkinson’s disease. Geriatrics & Gerontology International, 2013.
• The belated US FDA approval of the adenosine A2A receptor antagonist istradefylline for Parkinson’s disease.
• Caffeine for treatment of Parkinson disease: a randomized controlled trial. Neurology, 2012.
• CYP1A2 genotype and caffeine’s effects on exercise performance.