Smoke marijuana and you'll soon be hooked on harder drugs, society warned.  Smoke marijuana and you'll sooner be hooked on potato chips, the smoker discovered. The threat of more dangerous addictions aside, an attack of the munchies was unavoidable.

People have known about the connection between marijuana and appetite for centuries. In China and India, people have used it for millenniums to help coax the sick to eat. These days, doctors with the same aim can prescribe a pill containing a synthetic version of tetrahydrocannabinol (or THC)--marijuana's main active ingredient. Scientists, however, are just beginning to understand why the weed affects eating. They hope that such knowledge will lead to new drugs--legal ones. Drugs to help obese people lose weight; drugs to stimulate the appetites of people with cancer or AIDS without making them high. Drugs for nausea, too, which THC also helps suppress. Neuroscientists, in particular, are excited about marijuana research because of what it is revealing about the chemistry of the brain.

It is now known that THC mimics chemicals made naturally by our brains--chemicals that influence a smorgasbord of bodily functions including movement, thought and perception. Studying these brain chemicals (known as "endogenous cannabinoids") is increasing our understanding of an array of medical conditions--among them pain, Parkinson's disease, Tourette's syndrome and memory loss. Drug companies are working busily to develop new therapies based on this knowledge. "Today, I think we can fairly say that the endogenous cannabinoids are a major system by which brain cells communicate with one another," says Daniele Piomelli, professor of pharmacology at UC Irvine. "It's a whole new era of neurobiological research."

Until recently, scientists didn't think marijuana could teach them much of anything about the brain. The weed, they thought, wasn't going to be like morphine, which led researchers to discover crucial brain chemicals such as opiate receptors--key to the perception of pleasure and pain--and endorphins, the body's own natural painkillers. Instead, THC was thought to alter the mind in a messy and random kind of way that didn't involve any particular part of brain chemistry.

But just over a decade ago, researchers discovered a receptor in the brain that THC fits snugly into, like a key in a lock. The receptor resides in the membrane of certain brain cells--and when THC binds, those brain cells send signals, creating the effects of a "high": the munchies, alterations in mood, thought and perception as well as impaired coordination and memory.

Presumably, the receptor didn't evolve in brains to give people pleasurable sensations should they happen to smoke a certain plant. Indeed, its discovery immediately triggered a search for the brain's own version of THC--a naturally occurring chemical lurking somewhere in our brains--one that binds the receptor under normal, undrugged circumstances. The search commenced, and finally, in 1992, after years of sifting through extracts of pig brain, scientists in Israel found the chemical.

They called it "anandamide," from the Sanskrit word for "bliss."

Tracking Down the Effects of Cannabinoids

Today, the researchers, led by Raphael Mechoulam of the Hebrew University of Jerusalem, have found two more such chemicals in mammal brains. And scientists have so far found two receptors to which those chemicals bind. By studying where the cannabinoids and receptors are made, and what happens when they're interfered with, researchers are figuring out what jobs the chemicals do in the day-to-day running of our bodies.

The naturally occurring chemical has very different effects from inhaled THC, which floods the system, distorts perception and produces a "high," often interfering with the jobs that the brain's own chemical performs.

Cannabinoids, for instance, play a crucial part in fine-tuning signals that certain brain cells send to each other, according to three research papers published in March in the journals Nature and Neuron. Scientists had been searching for some time for chemicals that can do this fine-tuning trick, which is important, among other things, for controlling movement and forging memories.

The chemicals also play a part in blood pressure regulation. Cannabinoid receptors, it turns out, aren't just in our brains--they're also in tiny blood vessels that dilate when they're exposed to cannabinoids. And cannabinoids seem to contribute to the all-important decision of when to start eating and when to stop--choices that are decided by a bunch of chemicals that duke it out in a primitive part of our brain, the hypothalamus. On one side of the ring are the appetite stimulators--chemicals that prompt us to eat. There are a variety of these stimuli.

On the other side of the ring are molecules, such as leptin, that put the brakes on feeding. One only has to look at a rat or mouse lacking the ability to make or respond to leptin to see why such a brake is important: leptin-lacking animals become hugely obese.

Now, in an April paper in the journal Nature, scientists have provided evidence that the endogenous cannabinoids are a normal part of this process--helping to stimulate appetite

In the study, a research team led by Dr. George Kunos, scientific director of the National Institute on Alcohol Abuse and Alcoholism in Bethesda, Md., deprived mice of food for 18 hours. Then they watched to see how much the animals ate when they were finally given some rodent chow. Mice that had been engineered to lack the cannabinoid receptor didn't eat as much as the normal mice, the scientists found--presumably because an important part of their appetite machinery had been removed.

And mutant mice that don't make leptin produce extra amounts of appetite-enhancing cannabinoid in their hypothalamus--which could be partly why such mice become grossly fat. What's more, a drug that sticks tightly to the cannabinoid receptor and stops it from doing its job reduces feeding in mice--presumably because, once again, a portion of the appetite-stimulating machinery has been blocked.

Could this drug do something similar in humans and help overweight people shed flab? The French pharmaceutical company Sanofi-Synthelabo has been testing just that. In a four-month trial, 400 people were given doses of varying strength of the receptor blocker. By trial's end, those taking the highest dose had lost about 11 pounds after four months, says Gerard Le Fur, head of research and development for the company. There's still a great deal to be learned about cannabinoids, muses Mechoulam: For now, for instance, scientists have only the vaguest idea how these small, fatty chemicals are made.

And there is still a lot to learn about the part cannabinoids play in regulating so many aspects of our biology: nausea, immunity, stress, sleep, reproduction and more. Just last month, Mechoulam says, a research paper in the Lancet reported that abnormal anandamide levels may be linked to miscarriages. Kunos and colleagues suspect that the body's cannabinoids are involved in alcohol abuse. The pace of discovery is brisk. "It is a very exciting time," said Mechoulam. Marijuana Studies Studying marijuana's effects on the brain has led to the discovery of important brain chemicals called endogenous cannabinoids, as well as cannabinoid receptors that bind them. These substances are present in many parts of the brain and are involved in many brain functions. Here are some of the regions and some of the functions they help regulate.