Public Release: 

Mice with human protein, COX-2, exhibit age-related memory loss similar to Alzheimer's disease

Johns Hopkins Medicine

A new player is emerging in the complex world of the brain. This player, a protein called COX-2, appears to be critically important in the brain's normal functioning, as evidenced by its ability to wreak havoc in mice that have too much of it.

These mice, engineered to make the human COX-2 protein, develop memory problems as they age, mimicking problems seen in Alzheimer's disease, report scientists from Johns Hopkins School of Medicine and the biomedical research company Pharmacia. The results are scheduled for presentation Nov. 14 at the annual meeting of the Society for Neuroscience in San Diego.

Additional studies seem to link COX-2 to loss of brain cells in animal models of other neurologic diseases, as well, including stroke, Parkinson's disease and Lou Gehrig's disease.

"The really exciting thing is that this protein is turning out to be involved in so many things, and we already have the means to target it, to block it," says Katrin Andreasson, M.D., assistant professor of neurology and neuroscience at Johns Hopkins. "It makes it very conceivable that in the years to come we could prevent these diseases."

Drugs that reduce swelling, like ibuprofen and newer anti-inflammatories like Celebrex and Vioxx, work by blocking COX-2. However, Andreasson cautions that a lot of work remains to learn whether these drugs might prevent neurological disease or damage in people.

Building on their report in the Oct. 15 issue of the Journal of Neuroscience, Andreasson's colleagues Alena Savonenko and Alicja Markowska are to present new data at the neuroscience meeting that shows the more COX-2 protein the mice make in the brain, the more pronounced their memory problems and the faster those problems develop.

Andreasson's colleagues measured the animals' abilities to learn and remember using a battery of tests, including mazes and swimming tests. The animals' behaviors in the test situations reflected how brain levels of the human COX-2 protein affected their learning and memory at different ages.

"The effect of having lots of the human COX-2 protein is remarkable in these mice," says Andreasson. "At age seven months -- roughly similar to humans in their 20s and 30s -- they are fine, but as these mice get older they exhibit progressively greater memory deficits. These mice have real age-dependent memory loss, and higher age-dependent loss of brain cells that parallels the behavioral changes."

Because so much is already known about the human COX-2 protein, learning more about what it does in the brain, even the mouse brain, is a little easier, says Andreasson. "We already have tools to see where COX-2 is located in the brain and in individual nerve cells, and we can also test whether it's working by measuring levels of the molecules it makes."

COX-2 helps make a group of five molecules, called prostaglandins, that send signals to the nerve cell. Andreasson says one of the prostaglandins may be the real culprit behind COX-2's effects on memory and it's role in disease.

"If it's a prostaglandin, we can find out which is good and which is bad, and potentially target the bad one specifically to prevent memory loss or damage from aging or stroke or disease," she says. "The possibility would exist to prevent neurological problems and to do so with fewer side effects than blocking COX-2 itself."

While targeting prostaglandins is still hypothetical, current research is beginning to evaluate COX-2 inhibitors to prevent or delay neurological problems. For example, an ongoing study at Johns Hopkins and three other centers will determine whether using these drugs can affect development of Alzheimer's disease in older patients at high risk for it.

COX-2's role in exacerbating damage from strokes is another hot topic of pursuit among neurologists and neuroscientists. Another Hopkins presentation at the Society for Neuroscience meeting shows that the area of the brain affected by a stroke is much larger in mice with extra COX-2, but importantly, the study also shows it might be possible to save a large part of that affected area.

"In a stroke, a core part of the affected brain is destroyed, but there's a region around the core that isn't immediately destroyed, but is at high risk for damage," explains Andreasson. "If you can save that area, called the penumbra, or save a large part of it, that would have huge potential to improve recovery after stroke."

While studies looking at COX-2 inhibitors in patients with stroke haven't begun, scientists at Johns Hopkins have started a study of people at high risk of developing Alzheimer's disease to see if using COX-2 inhibitors can prevent or delay that disease. Since participants in the study don't have the disease, it will take many years for the researchers to know for sure if there is an effect.


Authors on the Society for Neuroscience presentation on age-related memory loss in COX-2 mice are Savonenko, Andreasson and Paul Worley of Johns Hopkins, Peter Isakson of Pharmacia Research and Development (Peapack, NJ), and Markowska, formerly of Johns Hopkins University, now at the National Institute on Aging. (Wednesday, Nov. 14, poster session 4-5 pm)

Authors on the Oct. 15 Journal of Neuroscience paper on age-related memory loss in COX-2 mice are Andreasson, Savonenko, Worley, Isakson and Markowska, as well as Sveta Vidensky and Walter Kaufmann of Johns Hopkins, and Joseph Goellner, Yan Shang and Alex Shaffer of Pharmacia Research (St. Louis). (J. Neurosci., 2001; 21(20): 8198-8209.) This study was funded by the National Institutes of Health, the Alzheimer's Association and Pharmacia Research and Development.

Authors on the analysis of stroke in the COX-2 mice are Sylvain Dore, Nobuo Sugo, Richard Traytsman, Raymond Koehler and Patricia Hurn of anaesthesiology and critical care medicine at Johns Hopkins; Worley, Andreasson and Isakson. (Monday, Nov. 12, 11 a.m. - noon poster session)

Related Web sites: (search for meeting abstracts at

Johns Hopkins Medical Institutions' news releases are available on an EMBARGOED basis on EurekAlert at and from the Office of Communications and Public Affairs' direct e-mail news release service. To enroll, call 410-955-4288 or send e-mail to

On a POST-EMBARGOED basis find them at and Quadnet at

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.