How do you make a memory? One important ingredient researchers have been studying is calcium. Daniel Johnston, a professor of neuroscience at Baylor College of Medicine (Houston, TX), and his colleagues use charge-coupled-device (CCD) cameras capable of detecting very low levels of light to watch the mechanisms they think trigger memories and learning. It's a demanding application, and scientists are counting on improvements in imaging technology to help them see more.
The researchers start with slices of a rat's brain and stimulate them electrically to mimic neural activity that goes on during learning. They treat the brain with a fluorescent dye and illuminate it with 380-nm light from a tungsten bulb. The dye emits around 500 nm when it binds to a calcium ion. The CCD camera allows the researchers to record changes in calcium concentration throughout a neuron, including both the central cell body and the branching dendrites—the jagged twigs that transmit information to neighboring neurons.
"The rise in calcium triggers some biochemical events that lead to more permanent changes in the neuron," Johnston said. Ultimately those changes lead to the synthesis of proteins, which form the chemical basis of memory. Scientists hope to discover how the brain learns, and such research could give them insight into memory disorders such as Alzheimer's disease.
"It's a very active field in neuroscience today," Johnston said. "We're still a long way from totally understanding the mechanism."
Technical challenges
Imaging the changes in calcium concentration presents several technical challenges. For one thing, the fluorescent dye emits extremely low levels of light. And the change in fluorescence the researchers measure can be as little as 1%. Neurons at rest contain approximately 50 nanomolars (nM) of calcium but can reach up to 100 µM. These researchers, however, are interested in watching concentrations go from 50 nM to only 100 to 200 nM.
In addition, the dendrites can extend 500 or 600 µm and can fan out to half a millimeter wide. Changes in concentration happen rapidly, as well, so researchers not only want to be able to capture low light levels but also need high spatial and temporal resolution. They want to know the magnitude of the changes and the progression and to see where in the whole neuron the changes take place. "It actually puts quite severe constraints on the type of imaging system we use," Johnston said.
His group has recently upgraded its imaging system with a new cooled CCD camera (Roper Scientific; Tucson, AZ) that has two to four times the sensitivity of previous instruments the group has used. Because it is back-illuminated, the camera has a quantum efficiency of around 80%, compared to 20% to 40% for other cameras. It is cooled to -30°C, which Johnston said greatly improves the signal-to-noise ratio. At 3 MHz, the camera is six times as fast as the 500-kHz camera he had used before, and it has a resolution of 512 x 512 pixels of 13 x 13 µm, compared to 384 x 288 pixels for the previous device.
"The cameras keep getting better, so the better the camera, the more detailed and accurate measurements can be made," Johnston said.