Can a Lack of Sleep Cause Psychiatric Disorders?

Study shows that sleep deprivation leads to a rewiring of the brain's emotional circuitry

By Nikhil Swaminathan

There's no question that people need their sleep: studies have linked a lack of shut-eye to everything from disruptions in the immune system to cognitive deficits to weight control.

In fact, psychologist Matthew Walker of the University of California, Berkeley, says that "almost all psychiatric disorders show some problems with sleep.'' But, he says that scientists previously believed the psychiatric problems triggered the sleep issues. New research from his lab, however, suggests the reverse is the case; that is, a lack of shut-eye is causing some psychological disturbances.

Walker's team and collaborators from Harvard Medical School reached their conclusions, published in Current Biology, after studying 26 healthy students aged 24 to 31 after either an all-nighter or a full night's sleep.

Fourteen subjects spent 35 straight hours without getting a wink before being rolled into a functional magnetic resonance imaging (fMRI) scanners where their brains were observed while they viewed a set of 100 photos that became increasingly disturbing as they progressed. Early slides were snapshots of an empty wicker basket on a table; the scenes changed as the series progressed, however, to more shocking settings, such as a tarantula on a person's shoulder and finally pictures of burn victims and other traumatic portraits.

The researchers mainly monitored the amygdala, a midbrain structure that decodes emotion, and observed that both sets of volunteers had a similar baseline of activity when shown the innocuous images. But, when the scenes became more gruesome, the amygdalae of the sleep-deprived participants kicked up, showing 60 percent more activity relative to the normal population's response. In addition, the researchers noticed that more than five times more neurons in the area were transmitting impulses in the sleep-deprived brains.

Walker described the heightened emotional response in the weary as "profound," noting, "We've never seen a magnitude of increase between two groups that big in any of our studies before."

The team also checked the fMRI readings to determine whether any other brain regions had a similar pattern of activity, which would indicate that the brain networks were communicating with one another. In normal participants, the amygdala seemed to be talking to the medial prefrontal cortex, an outer layer of the brain that, Walker says, helps to contextualize experiences and emotions. But, in the sleep-deprived brain, the amygdala seemed to be "rewired," coupling instead with a brain stem area called the locus coeruleus, which secretes norepinephrine, a precursor of the hormone adrenaline that triggers fight-or-flight type reactions.

"Medial prefrontal cortex is the policeman of the emotional brain," Walker says. "It makes us more rational. That top-down, inhibitory connection is severed in the condition of sleep deprivation. … The amygdala seems to be able to run amok." People in this state seem to experience a pendulum of emotions, going from upset and annoyed to giddy in moments, he says.

"There seems to be a causal relationship between impaired sleep and some of the psychiatric symptomatology and disorders that we're seeing," says Robert Stickgold, an associate professor of psychiatry at Harvard Medical School who was not involved in this study. He cites research linking sleep apnea, in which breathing is disrupted, to attention deficit hyperactivity disorder and the evidence of a connection between depression and insomnia as examples. "It might be that those medial frontal regions tell the rest of the brain, 'You can chill,'" he says. "Those circuits become exhausted or altered after a lack of sleep."

Walker says the team now plans to examine the effects of disruption of certain types of sleep, such as REM sleep or slow-wave sleep. "I think we may start to think about a new potential function for sleep," says Walker. "It does actually prepare our emotional brains for next-day social and emotional interactions."

First Gene for Schizophrenia Discovered

By Kristin Leutwyler

No single genetic mutation can ever account for the complex range of symptoms that arise in devastating neuropsychiatric disorders such as schizophrenia. But scientists from the Julius-Maximilians-University of Wuerzburg in Germany have zeroed in on one mutation of a gene on chromosome 22 that appears to play an important role in catatonic schizophrenia particularly severe form of the disease characterized by acute psychotic breaks and disturbed body movements. They will describe their finding, based on an analysis of a large pedigree, in an upcoming issue of Molecular Psychiatry.

Earlier linkage studies lead the team of geneticists, psychiatrists and neuroscientists to examine chromosome 22 more closely and in particular, they focused on a gene encoding a protein called WKL1. This protein appears to share many features with ion channels, complexes that straddle a cell's membrane and help transport electric currents along neurons. (Mutations in one remotely related ion channel, the potassium channel KCNA1 cause a rare movement disorder called episodic ataxia.) Of significance, the researchers found the WKL1 gene transcript exclusively in brain tissue.

Further study is needed to determine if the same WKL1 mutation occurs in other families with a history of schizophrenia, or in uniherited cases of the disease. Still, the scientists are hopeful that their discovery may help elucidate some of the biological mechanisms behind schizophrenia and ultimately one day lead to better treatment options.

Instant Expert: The Human Brain

Instant Expert: The Human Brain
• NewScientist.com news service
• Helen Philips

The Human Brain - With one hundred billion nerve cells, the complexity is mind-boggling. Learn more in our cutting edge special report. The brain is the most complex organ in the human body. It produces our every thought, action, memory, feeling and experience of the world.

This jelly-like mass of tissue, weighing in at around 1.4 kilograms, contains a staggering one hundred billion nerve cells, or neurons. The complexity of the connectivity between these cells is mind-boggling. Each neuron can make contact with thousands or even tens of thousands of others, via tiny structures called synapses.

Our brains form a million new connections for every second of our lives. The pattern and strength of the connections is constantly changing and no two brains are alike. It is in these changing connections that memories are stored, habits learned and personalities shaped, by reinforcing certain patterns of brain activity, and losing others.

Grey matter

While people often speak of their "grey matter", the brain also contains white matter. The grey matter is the cell bodies of the neurons, while the white matter is the branching network of thread-like tendrils - called dendrites and axons - that spread out from the cell bodies to connect to other neurons.

But the brain also has another, even more numerous type of cell, called glial cells. These outnumber neurons ten times over. Once thought to be support cells, they are now known to amplify neural signals and to be as important as neurons in mental calculations. There are many different types of neuron, only one of which is unique to humans and the other great apes, the so called spindle cells.

Brain structure is shaped partly by genes, but largely by experience. Only relatively recently it was discovered that new brain cells are being born throughout our lives - a process called neurogenesis. The brain has bursts of growth and then periods of consolidation, when excess connections are pruned.

The most notable bursts are in the first two or three years of life, during puberty, and also a final burst in young adulthood. How a brain ages also depends on genes and lifestyle too. Exercising the brain and giving it the right diet can be just as important as it is for the rest of the body.

Chemical messengers

The neurons in our brains communicate in a variety of ways. Signals pass between them by the release and capture of neurotransmitter and neuromodulator chemicals, such as glutamate, dopamine, acetylcholine, noradrenalin, serotonin and endorphins.

Some neurochemicals work in the synapse, passing specific messages from release sites to collection sites, called receptors. Others also spread their influence more widely, like a radio signal, making whole brain regions more or less sensitive.

These neurochemicals are so important that deficiencies in them are linked to certain diseases.
For example, a loss of dopamine in the basal ganglia, which control movements, leads to Parkinson’s disease. It can also increase susceptibility to addiction because it mediates our sensations of reward and pleasure.

Similarly, a deficiency in serotonin, used by regions involved in emotion, can be linked to depression or mood disorders, and the loss of acetylcholine in the cerebral cortex is characteristic of Alzheimer’s disease.

Brain scanning

Within individual neurons, signals are formed by electrochemical pulses. Collectively, this electrical activity can be detected outside the scalp by an electroencephalogram (EEG).
These signals have wave-like patterns, which scientists classify from alpha (common while we are relaxing or sleeping), through to gamma (active thought). When this activity goes awry, it is called a seizure. Some researchers think that synchronizing the activity in different brain regions is important in perception.

Other ways of imaging brain activity are indirect. Functional magnetic resonance imaging (fMRI) or positron emission tomography (PET) monitor blood flow. MRI scans, computed tomography (CT) scans and diffusion tensor images (DTI) use the magnetic signatures of different tissues, X-ray absorption, or the movement of water molecules in those tissues, to image the brain.

These scanning techniques have revealed which parts of the brain are associated with which functions. Examples include activity related to sensations, movement, libido, choices, regrets, motivations and even racism. However, some experts argue that we put too much trust in these results and that they raise privacy issues.

Before scanning techniques were common, researchers relied on patients with brain damage caused by strokes, head injuries or illnesses, to determine which brain areas are required for certain functions. This approach exposed the regions connected to emotions, dreams, memory, language and perception and to even more enigmatic events, such as religious or "paranormal" experiences.

One famous example was the case of Phineas Gage, a 19th century railroad worker who lost part of the front of his brain when a 1-metre-long iron pole was blasted through his head during an explosion. He recovered physically, but was left with permanent changes to his personality, showing for the first time that specific brain regions are linked to different processes.

Structure in mind

The most obvious anatomical feature of our brains is the undulating surfac of the cerebrum - the deep clefts are known as sulci and its folds are gyri. The cerebrum is the largest part of our brain and is largely made up of the two cerebral hemispheres. It is the most evolutionarily recent brain structure, dealing with more complex cognitive brain activities.

It is often said that the right hemisphere is more creative and emotional and the left deals with logic, but the reality is more complex. Nonetheless, the sides do have some specialisations, with the left dealing with speech and language, the right with spatial and body awareness.
See our Interactive Graphic for more on brain structure

Further anatomical divisions of the cerebral hemispheres are the occipital lobe at the back, devoted to vision, and the parietal lobe above that, dealing with movement, position, orientation and calculation.

Behind the ears and temples lie the temporal lobes, dealing with sound and speech comprehension and some aspects of memory. And to the fore are the frontal and prefrontal lobes, often considered the most highly developed and most "human" of regions, dealing with the most complex thought, decision making, planning, conceptualising, attention control and working memory.

They also deal with complex social emotions such as regret, morality and empathy.
Another way to classify the regions is as sensory cortex and motor cortex, controlling incoming information, and outgoing behaviour respectively.

Below the cerebral hemispheres, but still referred to as part of the forebrain, is the cingulate cortex, which deals with directing behaviour and pain. And beneath this lies the corpus callosum, which connects the two sides of the brain. Other important areas of the forebrain are the basal ganglia, responsible for movement, motivation and reward.

Urges and appetites

Beneath the forebrain lie more primitive brain regions. The limbic system, common to all mammals, deals with urges and appetites. Emotions are most closely linked with structures called the amygdala, caudate nucleus and putamen. Also in the limbic brain are the hippocampus - vital for forming new memories; the thalamus - a kind of sensory relay station; and the hypothalamus, which regulates bodily functions via hormone release from the pituitary gland.

The back of the brain has a highly convoluted and folded swelling called the cerebellum, which stores patterns of movement, habits and repeated tasks - things we can do without thinking about them.

The most primitive parts, the midbrain and brain stem, control the bodily functions we have no conscious control of, such as breathing, heart rate, blood pressure, sleep patterns, and so on. They also control signals that pass between the brain and the rest of the body, through the spinal cord.

Though we have discovered an enormous amount about the brain, huge and crucial mysteries remain. One of the most important is how does the brain produces our conscious experiences?
The vast majority of the brain’s activity is subconscious. But our conscious thoughts, sensations and perceptions - what define us as humans - cannot yet be explained in terms of brain activity.

A new study points to rare gene duplications and deletions that are believed to play a significant role in the psychological disorder

By Nikhil Swaminathan

A new study indicates that the genetic culprits behind schizophrenia may be much less common than previously believed. Researchers report this week in Science that a rare but devastating change in one of several different genes may dramatically increase the risk of developing the debilitating brain disorder affecting 1 percent of the world's population and marked by psychotic behavior, hallucinations and delusions. Until now, most scientists believed that it was likely
that a cluster of relatively common genetic mutations was to blame.

"We're not saying that our results themselves are irrefutable proof that schizophrenia is all rare mutations," says study co-author Jonathan Sebat, a geneticist at Cold Spring Harbor Laboratory in Long Island, N.Y. "But, rare mutations are an important part of the story, and it's this type of risk factors that tend to have the strongest effects."

Until recently, researchers trying to unravel the mysterious disease searched the genomes of schizophrenia patients for flaws not present in the genes of healthy people. Their probes turned up a few possible genetic suspects, but the findings were contradicted by those from other studies. In addition, candidate mutations typically only showed up in no more than 10 percent of schizophrenia sufferers sampled. The new study identifies more than 20 genes that may trigger the disease. If researchers can positively link any or all of these genes to the disease, it would set the stage for development of new therapies.

In this study, researchers combed the genomes of 150 schizophrenia sufferers and 268 healthy individuals for never-before-seen copy number variations (CNVs)—mutations that result in large swaths of DNA encompassing multiple genes either being deleted or duplicated. Some
such mutations have been found to be benign, but others have been implicated in ailments such as autism and cancer. The team of scientists, from research facilities across the U.S., found novel gene alterations in 5 percent of the healthy volunteers and 15 percent of the schizophrenia patients; new CNVs showed up in 20 percent of those subjects who developed symptoms at or before the age of 18.

"What we prove is that, collectively, there is a threefold effect in schizophrenia and a fourfold effect in the childhood-onset disorder," says Mary-Claire King, a geneticist at the University of Washington School of Medicine and another co-author, referring to the incidence of these rare variants between the groups.

The researchers determined which affected genes were likely to cause damage—and where in the body that damage that might occur. The flawed genes in the schizophrenia patients were overwhelmingly linked to changes in pathways responsible for communication between and within nerve cells. In fact, one gene, ERBB4, is known to code for a receptor that interacts with neuregulin 1, a protein that's been associated with the illness for a decade. "The silver lining is that these new findings show us … that we were already on the right track," Sebat says.

Daniel R. Weinberger, director of the Genes, Cognition and Psychosis Program at the National Institute of Mental Health in Bethesda, Md., agrees that CNVs likely play a role. But he doesn't believe the new study demonstrates that they operate alone.

King says the next step is to screen the 20 suspect genes to pinpoint specific defects common among large groups of schizophrenia patients. Sebat acknowledges that the data presented is merely consistent with the rare mutation gene model and not direct proof of it. "It's premature to say that these findings have diagnostic value [right now]," he says. "But, that's exactly where we're headed."

Brain map project set to revolutionize neuroscience

March 2008
NewScientist News Service
By: Peter Aldhous

Take the most complex organ in the human body, superimpose the legacy of biology’s biggest research project, and what have you got? An unprecedented brain map that is set to transform studies of neuroscience and brain disease.

The Allen Institute for Brain Science in Seattle, Washington, US, is today launching a four-year, $55-million effort to build a three-dimensional map documenting the levels of activity of some 20,000 different genes across the human brain.

“The Human Genome Project was the ‘what’, and our project is the ‘where’,” says Allan Jones, the institute’s chief scientific officer.

Established in 2003 with a $100-million gift from Microsoft co-founder Paul Allen, the Allen institute has already created a similar atlas of the mouse brain, unveiled in December 2006.

By revealing patterns of gene activity, the mouse atlas has allowed neuroscientists to identify functionally important regions that were invisible simply by looking at the brain’s anatomy.
Evolution insights

“That’s why the brain is such a unique structure,” says Greg Foltz, a neurosurgeon at the Swedish Neuroscience Institute, also in Seattle. “Its function is very much embedded in its anatomy.”

Foltz’s team has also used the mouse atlas to help home in on two genes, known as BEX1 and BEX2, which seem to be silenced in a form of brain cancer called glioma. An atlas of the human brain should be an even more powerful tool in identifying what goes wrong at the gene level in cancer and other diseases, he says.

For instance, some neuroscientists suspect that autism may be linked to abnormalities in a paired structure called the amygdala, involved in processing emotional information. The genes likely to be involved could be narrowed down by looking in the atlas for those that are active in the amygdala. Researchers could then compare the brains of autism patients with those from people without the condition, matched for age and gender, studying the activity of these candidate genes.

Comparisons between the mouse and human brain atlases should also yield insights into the evolution of our advanced cognitive abilities, suggests David Anderson, a neuroscientist at the California Institute of Technology in Pasadena, and one of the Allen institute’s scientific advisers. “Are there fundamental differences in the organisation of the brain?” he asks.
Huge task

The sheer size of the human brain will make the new project a much bigger challenge.

The mouse atlas was produced using a method called “in situ hybridisation”, in which thin slices of brain tissue are bathed in a solution containing molecular probes that bind to messenger RNA sequences produced by each gene. This gives a very detailed map of gene activity, down to the level of individual cells.

Trying to repeat this effort for all 20,000 genes across an organ about 2000 times larger than the mouse brain is impractical, for now. So Allen institute scientists will instead divide the human brain into between 500 and 2000 anatomical regions, and study gene activity in each by washing extracts from tissues in these regions across “gene chips” that can record which messenger RNA is present.

Once results from this initial phase of the project are in, which will take about two years, the institute’s scientists will perform in situ hybridisation across the whole brain for up to 500 genes with the most interesting patterns of activity.

As well as launching the human brain atlas, the Allen institute is starting two further projects. The first, costing around $15 million over the next two years, will look at the activity of around 4000 mouse genes at different stages in embryological and juvenile development.

A second project, taking about a year to complete at a cost of $2 million, will make an atlas of gene activity in the mouse spinal cord.

The Human Brain - With one hundred billion nerve cells, the complexity is mind-boggling. Learn more in our cutting edge special report.

Perspectives: The flip side to multiple personalities

15 March 2008
From New Scientist
Rita Carter

In 1791, a German doctor called Eberhardt Gmelin reported a bizarre case: one of his patients regularly transformed from a middle-class German woman into a French aristocrat. She would suddenly "exchange" her personality for the manners and ways of a French-born lady, speaking French perfectly but German as a Frenchwoman might. As her French self, she could remember everything she had said or done during her previous French "episodes". As a German woman, she knew nothing of her French personality.

Gmelin's French aristo was followed by other similarly odd cases. Felida X, for example, had three different personalities, each with their own illnesses. One of them even had her own pregnancy, unknown, at first, to the others. These cases introduced the contentious multiple personality disorder (MPD).

For much of the 20th century MPD was eclipsed by Freud's notions of hysteria and repression, but in the 1980s, for no obvious reason, it returned explosively. Between 1985 and 1995, an estimated 40,000 cases were diagnosed - twice as many as in the entire preceding century. But this time, MPD was considered more than just a psychiatric oddity. Under the label dissociative identity disorder (DID), doctors believed it was closely linked with childhood trauma.

The professionals were split. One camp argued that personality-switching was elaborate play-acting, encouraged by naive or needy therapists and fueled by an emerging "victim" culture. The other camp argued it indicated the surfacing of "alters" - selves created to carry the burden of traumatic events, which had then "split off" and become buried. We now know the two theories are not necessarily in conflict: the brain is able both to create false memories and recover memories that seemed to be lost.

The International Society for the Study of Trauma and Dissociation was at the heart of it all because dissociation, the "slicing-up" of experiences (and thus memories) into different streams of consciousness, was identified as the mechanism behind MPD. Once researchers were involved, dissociation became psychopathology and the notion of "normal" dissociation more or less disappeared.

But the truth is we all dissociate to some extent: if we didn't, we would be overwhelmed by the barrage of stimuli that continually assaults our senses. In Multiplicity, I argue one effect of this normal dissociation is to create a normal multiplicity. So the gap between "crazy" people with MPD/DID and the rest of us is a matter of degree.

With MPD/DID, the personalities are discrete, but those "multiples" in the rest of us share perceptions, thoughts and emotions. Crucially, this means they also share many memories. While people with MPD have amnesic gaps for the periods when they assume another personality, the rest of us just have hazy patches in our recall and, tellingly, the odd moment when we look back on a deviation from our characteristic behavior and wonder: "What got into me?"

To understand how multiplicity comes about, we need to look at the dissociation spectrum. At one end is the everyday focusing of attention we call "concentration". Moving along there is the intense absorption in a subset of stimuli, or "flow". Then there is day-dreaming, mild trance-like states, and such oddities as out-of-body experiences and lucid dreaming. Further still is detachment, including depersonalisation (the feeling of being dislocated from one's body or mind) and derealisation (feeling the world to be distant, or crushingly close, distorted or unreal).

Detachment is a defence mechanism, evolving from the "play dead" or "freeze" reaction to danger. Its adaptive function is obvious: in a car accident, an injured victim who can detach from their pain may be better able to escape. A doctor coming to help who detaches from their emotions may act more effectively. This normal response to an abnormal situation is only pathological if it continues beyond the threat or challenge.

Although we are not conscious of dissociated experiences at the time, they are nevertheless registered by the brain and may form memories. Later, such memories may be consciously recalled, resulting in a sort of "delayed" experience. In the 1970s, Ernest Hilgard, a psychologist at Stanford University, California, tested this hypnosis on patients who were, for various medical reasons, facing surgery without anaesthesia. Before their operations Hilgard put them into a trance and told them a "hidden observer" would feel their pain for them. After surgery, the patients recalled having the operation but reported no pain. But when Hilgard hypnotised them again and asked the "hidden observer" to speak, they reported the agony of the knife. The pain seemed to be stored in a "compartment" the patient was unable to access ordinarily.

The compartmentalization of MPD differs from other dissociative states in that entire, protracted experiences are stored separately, rather than just one component of an experience, such as pain. Prolonged experiences consist of sensations (the record of outside events) together with the emotions and thoughts we have at the time. Part of these internal events is a sort of background "hum" of self-recognition (the knowledge of who you are, your past and so on) plus characteristic ways of seeing and responding to the world. These are the habits of mind and behavior we recognize as a "personality".

Entire compartmentalized experiences will therefore hold a sense of identity and personality as well as sensations and emotions. So when one is recalled, the self that experienced it is recalled too. If it happened in childhood, the recaller will return to childhood, and while the recollection lasts will have no awareness of being adult or of subsequent experiences.

So personal identity and personality are functions of experience, not separate from it. In theory, we become a new personality in each situation. In practice, most experiences and the personalities they incorporate are similar enough to be partly integrated.

But as life becomes more heterogeneous, our personalities will be less closely conjoined. Compare the wide, fragmented experiences of children now with the narrow, continuous experiences of their grandparents. A child transported across cultures may still remain connected to her old life through her family. At school she speaks one language, with one set of opinions, habits and behaviors, while at home she behaves entirely differently. To avoid inner conflict, she flips between two personalities. The personalities are not "acts": she feels and thinks totally differently in each case. Neither is faked, but neither is more "real". Increasingly, personalities will become compartmentalized as modern life drives us along the dissociative spectrum.

Those who associate dissociation with pathology may find this alarming, and there is evidence that more people are suffering from chronic detachment. But there is already evidence that normal dissociation can protect people from pain, infection and depression. People who report they are more "multiple" suffer less from stress-related conditions. Say "Judy" has a sporty personality, A, and an academic personality, B. If A loses a tennis match, A is annoyed, which results in tensed muscles and a backache. If A was the only personality Judy had she would be tense all day. But if she goes off to college, switching to Judy B, her muscles relax because B doesn't care about the tennis match. So Judy suffers less than if she was only personality A.

It looks as if normal multiplicity could prove useful in helping people function in an increasingly complex world. But first we will need to recognize it as quite different from its pathological, 18th-century origins.

The Human Brain - With one hundred billion nerve cells, the complexity is mind-boggling.

Study Finds Traces of Drugs in Drinking Water in 24 Major U.S. Regions

A potential risk to mental health?
Monday , March 10, 2008
AP

A vast array of pharmaceuticals — including antibiotics, anti-convulsants, mood stabilizers and sex hormones — have been found in the drinking water supplies of at least 41 million Americans, an Associated Press investigation shows.

To be sure, the concentrations of these pharmaceuticals are tiny, measured in quantities of parts per billion or trillion, far below the levels of a medical dose. Also, utilities insist their water is safe.

But the presence of so many prescription drugs — and over-the-counter medicines like acetaminophen and ibuprofen — in so much of our drinking water is heightening worries among scientists of long-term consequences to human health.

In the course of a five-month inquiry, the AP discovered that drugs have been detected in the drinking water supplies of 24 major metropolitan areas — from Southern California to Northern New Jersey, from Detroit to Louisville, Ky.

Water providers rarely disclose results of pharmaceutical screenings, unless pressed, the AP found. For example, the head of a group representing major California suppliers said the public "doesn't know how to interpret the information" and might be unduly alarmed.

How do the drugs get into the water?

People take pills. Their bodies absorb some of the medication, but the rest of it passes through and is flushed down the toilet. The wastewater is treated before it is discharged into reservoirs, rivers or lakes. Then, some of the water is cleansed again at drinking water treatment plants and piped to consumers. But most treatments do not remove all drug residue.

And while researchers do not yet understand the exact risks from decades of persistent exposure to random combinations of low levels of pharmaceuticals, recent studies — which have gone virtually unnoticed by the general public — have found alarming effects on human cells and wildlife.

"We recognize it is a growing concern and we're taking it very seriously," said Benjamin H. Grumbles, assistant administrator for water at the U.S. Environmental Protection Agency.

Members of the AP National Investigative Team reviewed hundreds of scientific reports, analyzed federal drinking water databases, visited environmental study sites and treatment plants and interviewed more than 230 officials, academics and scientists. They also surveyed the nation's 50 largest cities and a dozen other major water providers, as well as smaller community water providers in all 50 states.

Here are some of the key test results obtained by the AP:

_Officials in Philadelphia said testing there discovered 56 pharmaceuticals or byproducts in treated drinking water, including medicines for pain, infection, high cholesterol, asthma, epilepsy, mental illness and heart problems. Sixty-three pharmaceuticals or byproducts were found in the city's watersheds.

_Anti-epileptic and anti-anxiety medications were detected in a portion of the treated drinking water for 18.5 million people in Southern California.

_Researchers at the U.S. Geological Survey analyzed a Passaic Valley Water Commission drinking water treatment plant, which serves 850,000 people in Northern New Jersey, and found a metabolized angina medicine and the mood-stabilizing carbamazepine in drinking water.

_A sex hormone was detected in San Francisco's drinking water.

_The drinking water for Washington, D.C., and surrounding areas tested positive for six pharmaceuticals.

_Three medications, including an antibiotic, were found in drinking water supplied to Tucson, Ariz.

The situation is undoubtedly worse than suggested by the positive test results in the major population centers documented by the AP.

The federal government doesn't require any testing and hasn't set safety limits for drugs in water. Of the 62 major water providers contacted, the drinking water for only 28 was tested. Among the 34 that haven't: Houston, Chicago, Miami, Baltimore, Phoenix, Boston and New York City's Department of Environmental Protection, which delivers water to 9 million people.

Some providers screen only for one or two pharmaceuticals, leaving open the possibility that others are present.

The AP's investigation also indicates that watersheds, the natural sources of most of the nation's water supply, also are contaminated. Tests were conducted in the watersheds of 35 of the 62 major providers surveyed by the AP, and pharmaceuticals were detected in 28.

Yet officials in six of those 28 metropolitan areas said they did not go on to test their drinking water — Fairfax, Va.; Montgomery County in Maryland; Omaha, Neb.; Oklahoma City; Santa Clara, Calif., and New York City.

The New York state health department and the USGS tested the source of the city's water, upstate. They found trace concentrations of heart medicine, infection fighters, estrogen, anti-convulsants, a mood stabilizer and a tranquilizer.

City water officials declined repeated requests for an interview. In a statement, they insisted that "New York City's drinking water continues to meet all federal and state regulations regarding drinking water quality in the watershed and the distribution system" — regulations that do not address trace pharmaceuticals.

In several cases, officials at municipal or regional water providers told the AP that pharmaceuticals had not been detected, but the AP obtained the results of tests conducted by independent researchers that showed otherwise. For example, water department officials in New Orleans said their water had not been tested for pharmaceuticals, but a Tulane University researcher and his students have published a study that found the pain reliever naproxen, the sex hormone estrone and the anti-cholesterol drug byproduct clofibric acid in treated drinking water.

Of the 28 major metropolitan areas where tests were performed on drinking water supplies, only Albuquerque; Austin, Texas; and Virginia Beach, Va.; said tests were negative. The drinking water in Dallas has been tested, but officials are awaiting results. Arlington, Texas, acknowledged that traces of a pharmaceutical were detected in its drinking water but cited post-9/11 security concerns in refusing to identify the drug.

The AP also contacted 52 small water providers — one in each state, and two each in Missouri and Texas — that serve communities with populations around 25,000. All but one said their drinking water had not been screened for pharmaceuticals; officials in Emporia, Kan., refused to answer AP's questions, also citing post-9/11 issues.

Rural consumers who draw water from their own wells aren't in the clear either, experts say.

The Stroud Water Research Center, in Avondale, Pa., has measured water samples from New York City's upstate watershed for caffeine, a common contaminant that scientists often look for as a possible signal for the presence of other pharmaceuticals. Though more caffeine was detected at suburban sites, researcher Anthony Aufdenkampe was struck by the relatively high levels even in less populated areas.

He suspects it escapes from failed septic tanks, maybe with other drugs. "Septic systems are essentially small treatment plants that are essentially unmanaged and therefore tend to fail," Aufdenkampe said.

Even users of bottled water and home filtration systems don't necessarily avoid exposure. Bottlers, some of which simply repackage tap water, do not typically treat or test for pharmaceuticals, according to the industry's main trade group. The same goes for the makers of home filtration systems.

Contamination is not confined to the United States. More than 100 different pharmaceuticals have been detected in lakes, rivers, reservoirs and streams throughout the world. Studies have detected pharmaceuticals in waters throughout Asia, Australia, Canada and Europe — even in Swiss lakes and the North Sea.

For example, in Canada, a study of 20 Ontario drinking water treatment plants by a national research institute found nine different drugs in water samples. Japanese health officials in December called for human health impact studies after detecting prescription drugs in drinking water at seven different sites.

In the United States, the problem isn't confined to surface waters. Pharmaceuticals also permeate aquifers deep underground, source of 40 percent of the nation's water supply. Federal scientists who drew water in 24 states from aquifers near contaminant sources such as landfills and animal feed lots found minuscule levels of hormones, antibiotics and other drugs.

Perhaps it's because Americans have been taking drugs — and flushing them unmetabolized or unused — in growing amounts. Over the past five years, the number of U.S. prescriptions rose 12 percent to a record 3.7 billion, while nonprescription drug purchases held steady around 3.3 billion, according to IMS Health and The Nielsen Co.

"People think that if they take a medication, their body absorbs it and it disappears, but of course that's not the case," said EPA scientist Christian Daughton, one of the first to draw attention to the issue of pharmaceuticals in water in the United States.

Some drugs, including widely used cholesterol fighters, tranquilizers and anti-epileptic medications, resist modern drinking water and wastewater treatment processes. Plus, the EPA says there are no sewage treatment systems specifically engineered to remove pharmaceuticals.

One technology, reverse osmosis, removes virtually all pharmaceutical contaminants but is very expensive for large-scale use and leaves several gallons of polluted water for every one that is made drinkable.

Another issue: There's evidence that adding chlorine, a common process in conventional drinking water treatment plants, makes some pharmaceuticals more toxic.

Human waste isn't the only source of contamination. Cattle, for example, are given ear implants that provide a slow release of trenbolone, an anabolic steroid used by some bodybuilders, which causes cattle to bulk up. But not all the trenbolone circulating in a steer is metabolized. A German study showed 10 percent of the steroid passed right through the animals.

Water sampled downstream of a Nebraska feedlot had steroid levels four times as high as the water taken upstream. Male fathead minnows living in that downstream area had low testosterone levels and small heads.

Other veterinary drugs also play a role. Pets are now treated for arthritis, cancer, heart disease, diabetes, allergies, dementia, and even obesity — sometimes with the same drugs as humans. The inflation-adjusted value of veterinary drugs rose by 8 percent, to $5.2 billion, over the past five years, according to an analysis of data from the Animal Health Institute.

Ask the pharmaceutical industry whether the contamination of water supplies is a problem, and officials will tell you no. "Based on what we now know, I would say we find there's little or no risk from pharmaceuticals in the environment to human health," said microbiologist Thomas White, a consultant for the Pharmaceutical Research and Manufacturers of America.

But at a conference last summer, Mary Buzby — director of environmental technology for drug maker Merck & Co. Inc. — said: "There's no doubt about it, pharmaceuticals are being detected in the environment and there is genuine concern that these compounds, in the small concentrations that they're at, could be causing impacts to human health or to aquatic organisms."

Recent laboratory research has found that small amounts of medication have affected human embryonic kidney cells, human blood cells and human breast cancer cells. The cancer cells proliferated too quickly; the kidney cells grew too slowly; and the blood cells showed biological activity associated with inflammation.

Also, pharmaceuticals in waterways are damaging wildlife across the nation and around the globe, research shows. Notably, male fish are being feminized, creating egg yolk proteins, a process usually restricted to females. Pharmaceuticals also are affecting sentinel species at the foundation of the pyramid of life — such as earth worms in the wild and zooplankton in the laboratory, studies show.

Some scientists stress that the research is extremely limited, and there are too many unknowns. They say, though, that the documented health problems in wildlife are disconcerting.

"It brings a question to people's minds that if the fish were affected ... might there be a potential problem for humans?" EPA research biologist Vickie Wilson told the AP. "It could be that the fish are just exquisitely sensitive because of their physiology or something. We haven't gotten far enough along."

With limited research funds, said Shane Snyder, research and development project manager at the Southern Nevada Water Authority, a greater emphasis should be put on studying the effects of drugs in water.

"I think it's a shame that so much money is going into monitoring to figure out if these things are out there, and so little is being spent on human health," said Snyder. "They need to just accept that these things are everywhere — every chemical and pharmaceutical could be there. It's time for the EPA to step up to the plate and make a statement about the need to study effects, both human and environmental."

To the degree that the EPA is focused on the issue, it appears to be looking at detection. Grumbles acknowledged that just late last year the agency developed three new methods to "detect and quantify pharmaceuticals" in wastewater. "We realize that we have a limited amount of data on the concentrations," he said. "We're going to be able to learn a lot more."

While Grumbles said the EPA had analyzed 287 pharmaceuticals for possible inclusion on a draft list of candidates for regulation under the Safe Drinking Water Act, he said only one, nitroglycerin, was on the list. Nitroglycerin can be used as a drug for heart problems, but the key reason it's being considered is its widespread use in making explosives.

So much is unknown. Many independent scientists are skeptical that trace concentrations will ultimately prove to be harmful to humans. Confidence about human safety is based largely on studies that poison lab animals with much higher amounts.

There's growing concern in the scientific community, meanwhile, that certain drugs — or combinations of drugs — may harm humans over decades because water, unlike most specific foods, is consumed in sizable amounts every day.

Our bodies may shrug off a relatively big one-time dose, yet suffer from a smaller amount delivered continuously over a half century, perhaps subtly stirring allergies or nerve damage. Pregnant women, the elderly and the very ill might be more sensitive.

Many concerns about chronic low-level exposure focus on certain drug classes: chemotherapy that can act as a powerful poison; hormones that can hamper reproduction or development; medicines for depression and epilepsy that can damage the brain or change behavior; antibiotics that can allow human germs to mutate into more dangerous forms; pain relievers and blood-pressure diuretics.

For several decades, federal environmental officials and nonprofit watchdog environmental groups have focused on regulated contaminants — pesticides, lead, PCBs — which are present in higher concentrations and clearly pose a health risk.

However, some experts say medications may pose a unique danger because, unlike most pollutants, they were crafted to act on the human body.

"These are chemicals that are designed to have very specific effects at very low concentrations. That's what pharmaceuticals do. So when they get out to the environment, it should not be a shock to people that they have effects," says zoologist John Sumpter at Brunel University in London, who has studied trace hormones, heart medicine and other drugs.

And while drugs are tested to be safe for humans, the timeframe is usually over a matter of months, not a lifetime. Pharmaceuticals also can produce side effects and interact with other drugs at normal medical doses. That's why — aside from therapeutic doses of fluoride injected into potable water supplies — pharmaceuticals are prescribed to people who need them, not delivered to everyone in their drinking water.

"We know we are being exposed to other people's drugs through our drinking water, and that can't be good," says Dr. David Carpenter, who directs the Institute for Health and the Environment of the State University of New York at Albany.