A lot of things that are shown to happen in animals do not necessarily transfer to humans

Professor Tonmoy Sharma

They discovered that mice lacking the protein in their brain circuits during a critical early period of their lives fail to develop normal reactions to anxiety-producing situations.

A team from Columbia University in New York used gene technology to create mice that lacked the protein.

Anxiety symptoms

They found that the mice, as adults, mimicked signs of human anxiety.

They moved around less than normal animals in open spaces, balked at entering elevated mazes, and were slower to begin eating in new environments.

When the serotonin receptors were activated, they were far more confident about being placed in the strange environment.

The key protein is usually found in two areas of the brain - one in the forebrain and another deep in an area called the brainstem

By using sophisticated genetic technology techniques, the researchers found the anxious behaviour was linked to a lack of the protein in the forebrain.

Further tests revealed the lack of protein only seemed to damage behaviour in the early stages of development.

Mice who were given drugs to shut down the function of the protein later in life did not show signs of anxiety.

The crucial period appears to be between five and 21 days after birth.

Human scans

Writing in the journal Nature, the researchers said that serotonin stimulation of the forebrain receptor during this period appears to trigger "long lasting changes in brain chemistry or structure that are essential for normal emotional behaviour throughout life".

Researchers are now carrying out brain imaging tests to try to determine whether the same protein malfunction is responsible for anxiety disorder in humans.

Professor Tonmoy Sharma, a consultant psychiatrist at Stonehouse Hospital, Dartford, told BBC News Online the theory was certainly very plausible.

He said anxiety disorders were relatively common, and anything that pointed to a root cause was welcome.

However, he said many people did not develop symptoms until their late 20s and 30s, and it was likely that they were caused by a complex interaction of biology and life experience.

He said: "A lot of things that are shown to happen in animals do not necessarily transfer to humans."

The research is published in the journal Nature.