The effects of Ca2+ buffers on cytosolic Ca2+ signalling.

Ca2+ is one of the most important and universal intracellular signalling agents, controlling a multitude of vitally important cellular processes (Petersen & Verkhratsky 2016). The regulating cytosolic Ca2+ signals vary enormously with regard to timing, localization and spatial extent, depending on the specific cell type and the function to be controlled (Petersen & Verkhratsky, 2016). Because of the many different types of Ca2+ sensors (Ca2+‐binding proteins) found at different locations inside cells, there is often a requirement for Ca2+ signals to be strictly localized. A well‐known example of this is the control of neurotransmitter release by Ca2+ entry through voltage‐gated Ca2+ channels, generating short‐lived local nano‐domains of high Ca2+ concentration ([Ca2+]i) (Meinrenken et al. 2003). A more recent, and very interesting, example is the physiological activation of two Ca2+‐dependent transcription isoforms, NFAT1 and NFAT4. This occurs through two coincident, but spatially segregated, intracellular Ca2+ signals, namely in the immediate vicinity of Ca2+ release‐activated Ca2+ channels in the plasma membrane and close to inositol trisphosphate receptors in the inner nuclear membrane (Kar et al. 2016).

Different spatio‐temporal Ca2+ signal patterns can occur in the same cell depending on the type, intensity and duration of stimulation (Petersen 1992). In pancreatic acinar cells, both local Ca2+ spiking in the region of secretory control and global [Ca2+]i elevations can be generated, which is important because Ca2+ signals control not only secretion but also many other functions, including growth (cell division) (Petersen, 1992). In nerve terminals, functions other than neurotransmitter secretion may depend on the global [Ca2+]i. There has therefore, for a long time, been much interest in measuring the volume‐averaged [Ca2+]i (Meinrenken et al. 2003), which can also be particularly helpful for a precise quantitative evaluation of Ca2+ handling.

The nature and concentration of intracellular Ca2+ buffers play an important role in determining the timing and spreading of Ca2+ released into the cytosol by opening of Ca2+ channels either in the plasma membrane or in organelle membranes. An early example of experiments showing this phenomenon was the demonstration, in whole‐cell patch clamp current recording studies on pancreatic acinar cells, that intracellular addition of a highly mobile low‐affinity Ca2+ buffer, for example citrate, transformed ACh‐evoked local and short‐lasting Ca2+ spikes into global and much more prolonged Ca2+ transients (Petersen, 1992). The timing and spatial extension of physiological Ca2+ signals therefore depends not only on the strength and type of stimulation but also on the affinities, mobilities and concentrations of the various intracellular Ca2+ buffers, which can vary significantly between different cell types. There is also a practical issue relating to the ability of Ca2+ buffers to influence intracellular Ca2+ signals. Because all Ca2+‐sensitive fluorescent probes are Ca2+ buffers, they can distort the [Ca2+]i signals they are designed to monitor.

It has been challenging to obtain reliable estimates for the parameters that define the dynamics of physiological [Ca2+]i changes. In this issue of The Journal of Physiology, Erwin Neher and his colleagues (Lin et al. 2017) now describe the currently most precise quantitative approach to solving this problem by once more taking advantage of the calyx of Held, a giant mammalian glutamatergic nerve terminal which – following the pioneering work of Ian Forsythe (Forsythe, 1994) – has been extensively studied by several groups (Meinrenken et al. 2003). Lin et al. (2017) describe the dynamic changes of global [Ca2+]i during single and repetitive voltage‐clamp depolarizations and provide quantitative data on Ca2+ inflow, Ca2+ buffering and Ca2+ clearance.

Using low concentrations of the low‐affinity Ca2+ indicator Fura‐6F, in order not to overwhelm the endogenous Ca2+ buffers, Lin et al. (2017) studied the voltage‐clamped nerve terminals with patch pipettes containing solutions with minimal Ca2+ buffer concentrations. This allowed them to determine the Ca2+ binding properties of the endogenous fixed buffers and also the Ca2+ clearance mechanism. With regard to the latter, a comparison was made between the results obtained with Cs+‐ or K+‐based pipette solutions. The data from these experiments confirmed the importance of K+‐dependent Na+–Ca2+ exchange for Ca2+ extrusion (Schnetkamp, 2004). In other experiments, Lin et al. (2017) used pipette solutions with 500 μm of the widely used Ca2+ chelator EGTA, determining its Ca2+ binding characteristics under realistic intracellular conditions. It turned out that the Ca2+ dissociation constant of EGTA is more than 3 times higher than the value previously obtained in vitro. This result is of great practical importance as EGTA has been, and no doubt will continue to be, a useful tool as a slow Ca2+ chelator buffering global rather than local [Ca2+]i.

Overall, the major importance of the work reported by Lin et al. (2017) is that, based on very sensitive experimental protocols, they have been able to generate a consistent set of parameters for modelling [Ca2+]i transients in a mammalian presynaptic nerve terminal. Estimates of some of the parameters determining [Ca2+]i dynamics have been reported previously, but they were based on experiments carried out under a variety of different conditions, whereas the new study by Lin et al. (2017) has resulted in a comprehensive set of parameters valid for recording conditions generally used for studies of the calyx of Held. The results of the work of Lin et al. (2017) that ‘one set of parameters accurately describes [Ca2+]i measurements covering a wide range of amplitudes and obtained using quite different stimulation protocols and ionic conditions’ is remarkable and promises that this set will turn out to be of real help as a firm quantitative basis for further studies in this field.

Additional information

Competing interests

No competing interests declared.

Funding

The author is a Medical Research Council Professor (G19/22/2) and is supported by a Medical Research Council Programme grant (MR/J002771/1).