A simple method for quantitative calcium imaging in unperturbed developing neurons

https://doi.org/10.1016/j.jneumeth.2009.08.004Get rights and content

Abstract

Calcium imaging has been widely used to address questions of neuronal function and development. To gain deeper insights into the actions of calcium as a second messenger, but also to measure synaptic function, it is necessary to quantify the level of calcium at rest and during calcium transients. While quantification of calcium levels is straightforward when using ratiometric calcium indicators, these dyes have several draw-backs due to their short wavelength excitation spectra, such as light scattering and cytotoxicity. In contrast, many single-wavelength indicators exhibit superior photostability, low phototoxicity, extended dynamic ranges and very high signal to noise ratios. However, quantifying calcium levels in unperturbed neurons has not been performed with these indicators. Here, we explore a new approach for determining the calcium concentration at rest as well as calcium rises during evoked and spontaneous neuronal activity in unperturbed developing neurons using a single-wavelength calcium indicator. We show that measuring the maximal fluorescence at the end of an imaging experiment allows determining calcium levels with high resolution. Specifically, we assessed the limits of calcium measurements with a CCD camera in small neuronal processes and found that even in small diameter dendrites and spines the intracellular calcium concentration and its changes can be estimated accurately. This approach may not only allow mapping patterns of neuronal activity quantitatively with the resolution of single synapses and a few tens of milliseconds, but also facilitate investigating the role of calcium as a second messenger.

Introduction

Calcium is an important second messenger in many cell types and in particular in neurons. Calcium signaling underlies various aspects of structural and functional plasticity in mature and developing nerve cells (Gomez and Spitzer, 2000, Redmond and Ghosh, 2005, Lohmann and Wong, 2005, Konur and Ghosh, 2005). On the other hand, calcium dynamics reflect neuronal activity, on the level of neuronal networks as well as single synapses (Feller, 1999, Ikegaya et al., 2004, Goldberg and Yuste, 2005). Therefore, calcium imaging has been widely used to address questions of neuronal function and development. To gain further insight into the actions of calcium as a second messenger, but also to measure, e.g. synaptic strength, it is necessary to quantify the level of calcium at rest and during activation of calcium transients.

Classically, ratiometric dyes that can easily be calibrated have been used for quantitative calcium imaging (Tsien, 1989a, Tsien, 1989b). However, ratiometric imaging is difficult to perform in many experimental situations because of a number of limitations. First, ratiometric imaging, for example with Fura-2, requires sophisticated and typically slow experimental set-ups for switching excitation wavelengths. Second, all currently applied chemical ratiometric indicators have short wavelength excitation and emission maxima and are therefore associated with light scattering in neuronal tissue and are potentially cytotoxic. Furthermore, there is a large selection of non-ratiometric dyes offering different dissociation constants and fluorescence spectra and some of them, such as Oregon Green BAPTA-1 exceed ratiometric dyes in dynamic range and intensity. Therefore, current efforts have been directed at developing methods for quantifying calcium dynamics in recordings with non-ratiometric dyes.

Recently, Maravall et al. (2000) demonstrated the feasibility of quantitative calcium recordings with single wavelength dyes. They showed that absolute [Ca2+]i levels as well as relative changes from rest can be determined with subcellular resolution. Essential for this procedure is to establish the maximal fluorescence of the dye at each location for which measurements are being taken. One possibility to determine the maximum fluorescence is to charge the cell with action potential bursts at increasing frequencies, which in turn will raise the intracellular calcium concentration ([Ca2+]i). The maximal fluorescence of the dye can then be deduced by extrapolation of fluorescence intensities to saturation. For this approach the imaged neuron has to be loaded with the calcium indicator and stimulated in whole-cell patch clamp configuration.

Patching a cell, however, will inevitably lead to an exchange of the cytoplasm with the pipette solution. As a consequence, this approach is not practical in experiments where the composition of the cytoplasm must be unperturbed, e.g. to determine the native resting calcium concentration or specific forms of calcium signaling that require the presence of second messengers. In addition, patch-clamp experiments are not feasible for long term experiments. An alternative technique for loading single neurons with calcium indicators is single cell electroporation which allows calcium recordings over hours without interfering with intracellular signaling (Haas et al., 2001, Lang et al., 2006, Nevian and Helmchen, 2007). However, at present there is no straightforward method for quantifying absolute calcium levels or changes in [Ca2+]i in experiments where the composition of the neuronal cytoplasm is maintained in its native state.

We have explored a new approach for determining the maximum fluorescence in situations where recordings in unperturbed cells over longer periods of time are necessary. The maximum intensity is being determined by generating a saturating calcium influx into the recorded neurons with ionomycin at the end of the experiment, which leads to a saturation of the intracellular dye without adding extracellular background fluorescence (Hesketh et al., 1983, Kao et al., 1989). Furthermore, we determined the limits of calcium measurements with a CCD camera in small neuronal processes and found that even in small diameter dendrites (1 μm) [Ca2+]i can be estimated accurately.

Section snippets

Slice preparation

Hippocampal slice cultures were prepared from postnatal day (P) 0–2 Wistar rats. After decapitation, the hippocampi were dissected in ice-cold GBSS medium. Transversal slices (400 μm) were cut with a tissue chopper (McIlwain) and plated onto Millicell membrane inserts (CM; Millipore, Billerica, MA), serum containing culture medium was added, and slices were incubated for two to ten days at 37 °C with 5% CO2. All animal procedures followed national and international ethics guidelines.

Single cell electroporation and patch-clamping of CA3 pyramidal cells

The recording

Results

An elegant method to obtain absolute levels or changes of [Ca2+]i using single-wavelength calcium indicators is to compare changes in fluorescence with the maximum fluorescence at each recorded site (Maravall et al., 2000). We wished to explore strategies for inducing maximal fluorescence without patch-clamping to quantify calcium levels in unperturbed developing hippocampal neurons. As a reference we chose the approach of obtaining the maximal fluorescence by measuring [Ca2+]i increases during

Discussion

We have investigated new approaches for quantifying calcium levels at rest and calcium dynamics with non-ratiometric calcium indicators. We believe that such approaches will facilitate the quantitative investigation of, e.g. synaptic function in developing, and possibly mature, neurons. We show that imaging neurons filled with calcium indicators by single cell electroporation and subsequent analysis using the here proposed approach allows determining [Ca2+]rest and Δ[Ca2+]i in cells whose

Acknowledgements

We would like to thank Nicole Stöhr for preparation and maintenance of slice cultures and Thomas Kleindienst and Volker Staiger for technical advice. We also thank Corette Wierenga and Volker Scheuss for critically reading an earlier version of the manuscript and Tobias Bonhoeffer for generous support. This work was supported by MPG, DFG, and KNAW.

Cited by (12)

  • Transport of Glucose by the Plasma Membrane Affects the Removal and Concentration of Ca<sup>2+</sup> at Rest in Neurons – Implications of a Condition Prior to Alzheimer's Disease?

    2020, Neuroscience
    Citation Excerpt :

    In order to evaluate Ca2+ ion concentration in the neurons, the influx was normalized by addition of ionomycin for further quantification of the steady-state Ca2+ clearance activity. Consequently, the fluorescence steady-state plateau period translate the activity of Ca2+ clearance mechanisms, since the Ca2+ influx is the same in both situations (for details, Maravall et al., 2000; Albantakis and Lohmann, 2009; Oh et al., 2013; Helmchen and Tank, 2015). The data demonstrated considerable intracellular increases of Ca2+ in the neurons submitted to a reduction of GLUT3; however, neurons under hypoglycemia for 4 days increased the glucose transport and controversially lead to an outsized Ca2+ plateau transient, which might indicate problems in intracellular glucose metabolization or a disbalance between SERCA/PMCA activity in Ca2+ clearance; therefore, more experiments are needed to rule out these hypotheses.

  • Probing synaptic function in dendrites with calcium imaging

    2013, Experimental Neurology
    Citation Excerpt :

    Similar reports provided insights into the distribution and functioning of NMDARs and VSCCs in individual hippocampal spines (Nimchinsky et al., 2004; Sabatini and Svoboda, 2000). Calcium imaging can also be employed to determine intracellular calcium concentrations during rest, synaptic stimulation and spontaneous activity (Albantakis and Lohmann, 2009; Cornelisse et al., 2007; Maravall et al., 2000). These reports show that the comparison of fluorescence intensities at rest and during synaptic activity with the intensity at saturation of the calcium indicator can be used to quantify calcium concentrations on the subcellular level during baseline conditions as well as during activation and synaptic plasticity.

  • Wave digital model of calcium-imaging-based neuronal activity of mice

    2023, International Journal of Numerical Modelling: Electronic Networks, Devices and Fields
View all citing articles on Scopus
1

Current address: Universitat Pompeu Fabra, Department of Technology, Computational Neuroscience, Roc Boronat 138, 08018 Barcelona, Spain.

View full text