September 2004, Issue 3

Diabetes at the Subcellular Level

Orian Shirihai, MD, PhD, and his group focus on diseases in which mitochondria play a key role, including diabetes and anemia. One innovative technology they use is the Optical LiveCell Array, a transparent micron-sized array of wells. The array is manufactured by Molecular Cytomics, a US/Israeli company with which Shirihai has a long-standing collaboration. "The cell array is made of glass, so it is transparent under the microscope," Shirihai says. "Every cell has its own room with windows going to the adjoining rooms, so each cell can interact with six other cells, like a honeycomb."

Shirihai's enthusiasm for the technology is well founded. Using an imaging apparatus and analysis programs, an investigator can acquire and quantify fluorescent signals from single cells while they are living and communicating within a complex of 10,000. "My goal — what I think will have the strongest impact on diabetes technology — is to make a new tool for diabetes and obesity research that will enable us to look at the large population of beta cells and encompass the heterogeneity of this population," Shirihai says. Beta cells are the insulin-producing cells of the pancreatic islets of Langerhans.

Shirihai joined the Tufts school of medicine's Department of Pharmacology and Experimental Therapeutics in December 2003. He comes to Tufts from the Marine Biological Laboratory in Woods Hole, Massachusetts, where he founded the Laboratory for Molecular Physiology of Mitochondria in 2001. Prior to Woods Hole, Shirihai did postdoctoral work in Stuart Orkin's laboratory at Harvard. He earned his MD and PhD degrees from the faculty of medicine at the Technion—Israel Institute of Technology, where his research uncovered the roles of membrane potential in blood cell differentiation.

Each islet of Langerhans contains from 1,000 to 10,000 beta cells, depending on the size of the islet, and each beta cell is a complex processor of energy input and output. Beta cells analyze levels of energy molecules derived from glucose, fatty acids, and amino acids, as well as messages from other pancreatic cells and hormonal inputs such as glucagon-like peptide (GLP1), in order to calculate one output: insulin secretion. Insulin is the signal to cells to take up glucose from the blood, and a disruption of this system usually results in diabetes. Shirihai is currently using primary cultures of beta cells and the Optical LiveCell Array technology to generate "a pancreatic islet on a chip." He follows individual beta cells as they respond to varying levels of energy molecules.

Shirihai's group found that beta cells do not respond uniformly to stimuli for insulin secretion. While all beta cells increase and decrease insulin secretion in an oscillatory manner, some cells have stronger insulin responses while others have weaker responses. These metabolically heterogeneous beta cells can be divided into subpopulations based on insulin responsiveness to nutrients and other messages. Shirihai's group is currently studying similarities and differences in various subpopulations of beta cells.

The 200 or so mitochondria within each beta cell are critical to the cell's energy analysis and insulin secretion. The mitochondria produce adenosine triphosphate (ATP) in response to the availability of energy-yielding molecules. ATP levels in the cytoplasm then signal the beta cell to increase or decrease insulin secretion. Because ATP production generates a membrane potential that can be measured by the cell's electrical charge (voltage), Shirihai's group can use voltage-sensitive fluorescent dyes and time-lapse confocal/2-photon microscopy to study mitochondrial activity in multiple individual cells of an intact islet.

Insulin secretion in an individual beta cell varies through oscillations in its mitochondrial activity (i.e., ATP production), and subpopulations of cells have distinctive frequencies of ATP production that result in different levels of insulin output. "Using a laboratory model for type 2 diabetes, we showed that the coordination and regularity of oscillations in mitochondrial activity are altered in diabetes," Shirihai says. "Current research is aimed at elucidating the mechanisms that control and synchronize these oscillations. We are studying the role of fatty acids as modulators of metabolic oscillations, and whether this pathway contributes to the onset of type 2 diabetes." Shirihai's group is collaborating on this project with the group of Barbara Corkey, PhD, who heads the Obesity Center at Boston University, and with Marc Prentki, MSc, PhD, at the University of Montreal.

Moving from the level of an individual cell to that of an individual mitochondrion, Shirihai's group found that, at any given moment, some mitochondria are more active than others in their production of ATP. Then the question became whether a single mitochondrion maintains a constant level of activity or whether the level varies from day to day. "When you have thousands of mitochondria in one cell, following one is very hard because they move like crazy," Shirihai says. The challenge was to find a label for mitochondria that fluoresced in a different wavelength than that of the voltage-sensitive dyes. The group introduced a transgene expressing a photoactivatable variant of green fluorescent protein (GFP) into the mitochondria. They could then target a single mitochondrion to activate the GFP. Because voltage-sensitive dye fluoresces red, they can use software to seek out the green mitochondrion and follow the membrane potential in real time. The Shirihai laboratory has already used this innovative technique to produce about 300 movies of single mitochondria.

The group found that different populations of mitochondria seem to maintain their level of activity until one interacts with another, which happens every 30 minutes or so. Shirihai was surprised to see that when a mitochondrion containing a cytoplasmic photoactivated GFP interacts with another mitochondrion, sometimes the fluorescence is passed between the mitochondria, indicating they fuse part of their membranes and share cytoplasm. "That was very exciting," Shirihai says. "We could actually see that a protein was passed from one to another. So then we asked, 'How is this influencing membrane potential?' We found that after the mitochondria share their fluids and split again, usually the membrane potential of one daughter mitochondrion is very high and one is very low." The group has now embedded photoactivatable GFP in the mitochondrion outer membrane to see if and when fluorescence is passed from one to the other. "Now we'll be able to tell if one mitochondrion is taking a bite out of the other," Shirihai says with a smile.

Shirihai collaborates actively with both academic and industry investigators. He hasn't yet started collaborations with colleagues at Tufts, but he's seeking out prospects. He would like to find a collaborator in neuronal biology, in order to follow the activity of mitochondria as they move up and down the neuron. Shirihai believes that the Optical LiveCell Array technology has a myriad of potential applications, and has offered to provide arrays and share his expertise with anyone at Tufts.

For more information, please contact Dr. Shirihai at


Tufts University, Office of the Vice Provost
Health Sciences Campus: (617) 636-6550
Medford Campus: (617) 627-3417
Copyright 2005 Tufts University. All Rights Reserved.

Please send questions/comments about this site to Webmaster.