Back to Application Notes
 

Calcium Imaging Systems: How fast is fast enough?

The Journal of NIH Research
November 1995 VOL 7

 

Authors: Eric Gruenstein and Jesus Luna
Reviewed: Mitch Villereal, Ph.D. University of Chicago

                                The Biology: How Fast Does Calcium Signaling Occur?ur?

    In this application note we discuss the speeds at which intracellular calcium signaling processes occur and what sort of equipment is required to image them. We have divided these calcium signals into three categories: (1) slow changes are defined as those that take many seconds or tens of seconds to reach peak values; (2) moderate changes are those that reach peak values in just a few seconds; and (3) rapid changes are peak values that are complete in one second or less. In the slow category are found many of the calcium signaling events that occur in response to such stimuli as growth factors and cytokines. These responses typically take many seconds or tens of seconds to reach peak cytoplasmic calcium levels and even longer to return to basal values. For these calcium signals, imaging at one ratio/sec or even slower is adequate.

    Moderate rates of cytoplasmic calcium change typically occur in cells that are undergoing calcium oscillations. An example of this class of calcium kinetics is shown in figure 1 where two neurons are shown undergoing synchronized calcium spiking in response to activation of NMDA receptors. Increasing the rate of data acquisition (compare panels A & B) results in the resolution of a greater number of spikes per minute. There is also an increase in peak calcium amplitude due to the increased likelihood of capturing data near the peak. Consequently, for signaling events in the moderate category, imaging at rates of several calcium change associated with processes such as the generation of an action potential or with the front of a calcium wave as it moves through the cytoplasm will usually require data acquisition at rates between 10 and 100 ratios/sec.

                                        The Technology: Key Factors For Speed and Cost

    Most current calcium imaging systems, with the possible exception of some confocal systems, can collect data faster than one ratio/sec., making them suitable for all experiments involving slow calcium changes as well as many types of moderate kinetics. The price for one of these systems, complete with computer, video camera, filter changer, and fluorescence microscope, can vary from $30,000 to $70,000. An important factor affecting price (but not necessarily performance) is the type of video camera employed. Most calcium imaging systems today use intensified CCD (ICCD) video cameras which are very light sensitive, can acquire fluorescence images in 33 msec, and typically cost about $15,000. Although the rate of image acquisition by ICCD cameras suggests that only 66 msec should be needed to acquire the two images to be ratioed, in actual practice ICCD cameras are quite noisy and almost always require the averaging of at least 8 video frame/image. This extends the time for acquiring a pair of images at least 1/2 sec and thus limits the practical speed of data capture to 2 ratios/sec or slower.

    Recently, integrating video cameras which substitute variable exposure times for image intensifiers have become available. With the proper choice of light source and filter sets, these integrating CCD cameras can often acquire low noise images from fura-2 loaded cells in only 100-200 msec/image. Thus, under real experimental conditions, integrating cameras may actually be faster than ICCDs and they cost less than 1/10 as much.

    For imaging fast calcium kinetics, cooled CCD cameras capable of capturing small sub-regions at image rates faster than 33 msec are required. These cameras will typically cost $30,000 or about twice the price of an ICCD and 30 times that of an integrating CCD.
A second important factor in determining the price and performance of a calcium imaging system is the mechanism for switching excitation wavelengths. Essentially two options are available: filter changers and choppers. Filter changers consist of rotating wheels or reciprocating solenoids which can operate at speeds of 25-100 msec per change and thus add about 50-200 msec/ratio. Choppers use a rotating mirror which can operate at speeds up to 1000 Hz, more than adequate to keep up with even the fastest cameras, but this speed comes at a considerable increase in cost.

    In summary, for most applications involving the imaging of slow to moderate calcium kinetics, ratiometric imaging systems employing integrating cameras should be acceptable and are available for as little as $30,000. For cells which take up fura-2 poorly or which are extremely sensitive to UV light, imaging systems with intensified CCD cameras may be preferable but will cost about $60,000. And for imaging fast calcium kinetics requiring more than 10 ratioed images/sec, figure on a system with a cooled camera and light chopper, and expect to pay about $90,000.