FOCUS - September 17, 2004 - RADIOLOGY: Imaging Method Reveals Which Mice Develop Type 1 Diabetes

September 17, 2004

Cell Biology:
Live-cell Studies Pick Up Pattern in Vesicle Traffic

Radiology:
Imaging Method Reveals Which Mice Develop Type 1 Diabetes

Endocrinology:
Fat Hormone Revives Reproductive Systems of Lean Women

Systems Biology
Systems Biology, the New Physiology, Marks First Year at HMS

Chemical Staples Turn Flimsy Peptide into Cancer Killer

Female Flies Join Food Fight

Time Zone Controls Limb Size

Images of Rotavirus Entry Show Bug the Exit as Childhood Killer

Contract Supports Research in Biodefense Proteomics

HSPH and Cyprus Establish International Initiative

HMS Welcomes Incoming Students

Stearns Appointed Associate Master of Castle

The Myrto Lefkopoulou Lectureship

Applications Wanted for Health Care Research

Longwood Symphony Season Opens

Honors and Advances

In Memoriam:
Leroy Vandam
Leonard Safon
Robert Moylan

Students Aid Families with Special-needs Children

Students Orient Themselves Toward Medicine

Front Page

RADIOLOGY

Imaging Method Reveals Which Mice Develop Type 1 Diabetes

A diagnosis of type 1 diabetes is like the worst of visitors--never welcome and always late. By the time the disease is detected, the immune system has already begun its attack on the insulin-producing beta cells of the pancreas and may even have laid waste to the lion's share of them. "You are not diagnosed with having diabetes until most of these beta cells are either lost or rendered inactive," said Diane Mathis, HMS professor of medicine at Joslin Diabetes Center. For the more than 13,000 Americans who are diagnosed each year, most of whom are children and adolescents, the only recourse is to begin a life-long course of insulin injections or to replace the lost cells with transplanted tissue, which presents its own hazards.

Nanoparticles light up the capillaries surrounding a healthy pancreas (click on left video) but leave a fuzzier trace as immune cells attack the pancreas of a mouse with type 1 diabetes (right video). (Images courtesy of Ralph Weissleder)

A far better strategy would be to stop the immune system from killing off the beta cells in the first place. But that would require detecting the disease very early, and so far no one has been able to pull off such a feat. Now a team of HMS researchers reports that it has literally shown a light on the earliest stages of type 1 diabetes--just when the renegade immune troops are amassing at the borders of the pancreas. Mathis, along with Maria Denis, Umar Mahmood, Christophe Benoist, and Ralph Weissleder, deployed a fleet of tiny iron particles into the bloodstream of diabetes-prone mice. The particles were picked up by immune system scouts, macrophages, inside the pancreas, essentially leaving a trace of their incursion that was detected by microscopic and magnetic resonance imaging (MRI) techniques. The findings appear in the Aug. 24 Proceedings of the National Academy of Sciences.

The signals sent by the beacon-bearing macrophages inside the pancreas of the living mice were so vivid that Mathis, Weissleder, HMS professor of radiology at Massachusetts General Hospital, and their colleagues have just gotten approval to try the MRI approach in a clinical trial with a small number of patients.

Maria Denis (left) worked with (left to right) Diane Mathis, Ralph Weissleder, and Christophe Benoist and with Umar Mahmood (not pictured) to develop a new method for imaging the early stages of type 1 diabetes. (Denis photo courtesy of Maria Denis; Mathis photo by Steve Gilbert; Weissleder and Benoist photos by Graham Ramsay)

If the approach is successful--and clinical trials using the nanoparticles for other medical purposes suggest that they can enter the pancreas--it could provide a first step toward a personalized approach to the diagnosis and treatment of type 1 diabetes, something that is sorely needed. "Currently, you can tell people who have had an attack on the pancreas because they develop auto-antibodies against pancreatic antigens," said Benoist. "But then it can be immediately, 20 years, or never that the person manifests diabetes as elevated blood glucose levels. You need something like this so you can follow people and see how their disease is progressing."

Probes for Therapy

Of course, catching the immune system as it is about to mount its attack will be of little use if there is no way to stop the invasion. Though some researchers have tried to develop drugs to correct the aberrant immune cells, the variable and often slow process by which type 1 diabetes develops has made it an unappealing target for pharmaceutical companies. "It can take five to seven years to know if a drug has had any influence," Mathis said. "If there were some way that it could be predicted very early on that this drug is or is not going to have an effect, it would be incredibly useful. Looking at changes in vasculature using these probes would in theory be a great way to do that."

This is not the first time that a fleet of nanoparticles has been deployed to the site of an immune cell attack. Weissleder and colleagues have used magnetic and other probes to detect otherwise invisible early steps in inflammation associated with atherosclerosis and rheumatoid arthritis. So effective was their approach that the National Institutes of Health awarded the MGH and Joslin researchers a collaborative program project grant to explore ways to apply their imaging methods to the study of autoimmune disease, including type 1 diabetes. (For an interactive introduction to Weissleder's probes, see Lab Works).

Pinpointing Disease

Denis, who was an HMS research fellow and is now in her native Greece, began by launching magnetic probes into normal mice and into mice genetically engineered to develop diabetes according to a fairly predictable time course. Initially, the magnetic nanoparticles were equipped with green-light emitting bulbs, or fluorochromes, so they could be detected visually by a confocal microscope. The point was to make sure that the magnetic signals the researchers would later detect with MRI were accurate. To make sure the nanoparticles were being taken into the bloodstream, Denis and colleagues imaged the mice immediately. Sure enough, the entire vasculature of both groups of mice, normal and diabetic, appeared green.

"If there were some way that it could be predicted very early on that this drug is or is not going to have an effect, it would be incredibly useful. Looking at changes in vasculature using these probes would in theory be a great way to do that."

The researchers went back 24 hours later and exteriorized the pancreas of each mouse, essentially pulling the organ out of the abdomen, and viewed it by confocal microscopy. While the blood vessels surrounding the pancreas of normal mice exhibited clear-cut contours, those surrounding the pancreas of the diabetic mice were fuzzy, suggesting that the walls of the capillaries had become porous and leaky. And the pancreas itself appeared to be turning green, presumably because it was filling with nanoparticle-carrying macrophages. What is more, the extent of the green correlated with the time course of the disease. "It is most intense two to three weeks after birth and then tapers off because these lesions become less aggressive," said Weissleder.

The same striking results--leaky microvasculature around the pancreases of diabetic mice but not of normal ones--played out when the experiments were repeated using MRI. "This is the tip of the iceberg. It opens the door to therapeutic interventions," said Weissleder. In fact, the researchers have used the magnetic probes to monitor various therapies in mice, and preliminary results suggest the approach works. In humans, such monitoring could mean that "for the first time you can actually titrate treatments in patients," said Weissleder. "Before we did not know how much drug to treat them with. So this approach could move us toward personalized medicine."

--Misia Landau