Biochimica et Biophysica Acta 1833 (2013) 1787–1797
Contents lists available at SciVerse ScienceDirect
Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbamcr
Genetically encoded Ca 2 + indicators: Properties and evaluation☆ Vadim Pérez Koldenkova a, Takeharu Nagai a, b,⁎ a b
Department of Biomolecular Science and Engineering, The Institute of Scientific and Industrial Research, Osaka University, Mihogaoka 8-1, Ibaraki, Osaka 567-0047, Japan PRESTO, Japan Science and Technology Agency, Sanbancho, Chiyoda-ku, Tokyo 102-0075, Japan
a r t i c l e
i n f o
Article history: Received 31 October 2012 Received in revised form 2 January 2013 Accepted 10 January 2013 Available online 22 January 2013 Keywords: Genetically encoded calcium ion indicators Ca2 + imaging Fluorescent proteins FRET Ratiometric Intensiometric
a b s t r a c t Genetically encoded calcium ion (Ca 2+) indicators have become very useful and widely used tools for Ca2+ imaging, not only in cellular models, but also in living organisms. However, the in vivo and in situ characterization of these indicators is tedious and time consuming, and it does not provide information regarding the suitability of an indicator for particular experimental environments. Thus, initial in vitro evaluation of these tools is typically performed to determine their properties. In this review, we examined the properties of dynamic range, affinity, selectivity, and kinetics for Ca2+ indicators. Commonly used strategies for evaluating these properties are presented. This article is part of a Special Issue entitled: 12th European Symposium on Calcium. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Calcium ion (Ca 2+) is a common second messenger in cellular signal transduction by which many biological phenomena such as muscle contraction, neuronal transmission, fertilization and hormonal secretion are regulated [1]. Ca2+ imaging is the powerful technique to investigate how Ca 2+ exerts its action on these phenomena. To monitor Ca2+ dynamics in living cells, optical Ca2+ indicators are indispensable. Thus, developing methods for Ca2+ measurement is still the subject of a growing number of studies. The first experiments visualizing intracellular Ca2+ changes [2] were conducted using the Ca2+-sensitive bioluminescent protein aequorin [3]. These and subsequent experiments were performed by directly injecting purified protein from the jellyfish Aequorea into the cells
Abbreviations: [Ca2 +], free calcium ion concentration; BRET, bioluminescence resonance energy transfer; cDNA, complementary DNA; cp, circularly permuted; D, dynamic range; EDTA, 2-({2-[Bis (carboxymethyl) amino] ethyl} (carboxymethyl) amino) acetic acid; EGTA, ethylene glycol-bis(2-aminoethylether)-N,N,N′,N′-tetraacetic acid; F, fluorescence intensity or FRET ratio; F0, fluorescence intensity or FRET ratio in resting conditions (in vivo); FP, fluorescent protein; FRET, Förster resonance energy transfer; GECI(s), genetically encoded Ca2 + indicator(s); GFP, green fluorescent protein; I, fluorescence intensity; Kd, dissociation constant; kon, koff, rate constants of Ca2 + binding and unbinding reactions, respectively; n, Hill constant; N, number of photons; R, FRET ratio; SNR, signal-to-noise ratio; YC, yellow cameleons; ΔF, fluorescence intensity or FRET ratio change; τ, time constant of dissociation of the Ca2 +-indicator complex; Θ, fractional saturation ☆ This article is part of a Special Issue entitled: 12th European Symposium on Calcium. ⁎ Corresponding author at: Department of Biomolecular Science and Engineering, The Institute of Scientific and Industrial Research, Osaka University, Mihogaoka 8-1, Ibaraki, Osaka 567-0047, Japan. Tel.: +81 6 6879 8480; fax: +81 6 6875 5724. E-mail address:
[email protected] (T. Nagai). 0167-4889/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbamcr.2013.01.011
examined since the method of introduction of exogenous genes into eukaryotic cells was established more than a decade later [4]. The development of Ca2+-sensitive dyes such as Quin2 (characterized by a dim fluorescence) [5] and fura-2 [6] as well as the use of their acetoxymethylated ester forms for non-invasive cell loading [7] were later breakthroughs that allowed much broader use of Ca2+-imaging. The next leap in Ca2+-imaging technology, the introduction of genetically encoded Ca2+ indicators, GECIs, was preceded by three important steps: the discovery of green fluorescent protein (GFP, reported in the same study in which aequorin was presented [3]), the creation of GFP color variants, which are suitable for Förster resonance energy transfer (FRET) experiments [8], and the biochemical study of Ca 2 + binding through a fusion of calmodulin with its binding peptide, M13 derived from the myosin light chain kinase [9]. These achievements lead to the studies regarding development of FRET-based Ca2+ indicators, FIP-CBSM [10] and cameleon [11]. FIP-CBSM was composed of the M13 peptide sandwiched between two fluorescent proteins, BGFP and RGFP. The FRET signal of FIP-CBSM was reduced under interaction with endogenous Ca2+-calmodulin in the cell. In the case of cameleon, two fluorescent proteins were fused directly to calmodulin and M13 peptide; in this indicator FRET signal was increased upon Ca2+ binding. These two pioneering works established the basis for further development of genetically encoded indicators. Next, other improvements in GECIs included development of single-fluorescent protein Ca 2 +-sensors (camgaroos [12], pericams [13], GCaMP [14]), implementation of circularly permuted fluorescent proteins (FPs) to improve the dynamic range of FRET-based indicators (YC 2.60 and YC3.60 [35]), replacement of the Ca2+-sensing moiety calmodulin by troponin C (TN-indicators family [15]), or structure-guided remodeling of the calmodulin-M13 interface (DcpV-indicators families
1788
V. Pérez Koldenkova, T. Nagai / Biochimica et Biophysica Acta 1833 (2013) 1787–1797
[16]) to allow reduction of non-specific interactions in the cytoplasm. More recent improvements involving large-scale screening after random (GECOs [17]) or semi-rational mutation (GCaMP5-indicator family [18]), have led to color and high dynamic range variants of single fluorescent protein-based indicators. Other important modifications included the complete redesigning of the Ca 2+-sensing moiety by duplicating the troponin C C-terminal EF hands in the sensor TN-XXL [19] and the significant increase in Ca 2 +-affinity in the YC-Nano indicators [20]. Interestingly, a completely new type of fluorescent Ca 2 +-indicator, CatchER has been recently developed by replacement of 5 residues to create a low-affinity Ca 2 +-sensing motif in EGFP, becoming the smallest GECI to date [21]. This quickly increasing diversity of fluorescent GECIs has been described in many recent reviews [22–25]. Ectopic expression of aequorin in eukaryotic cells by gene transfection was employed in the 1990s for Ca 2 +-imaging experiments, although a dim chemiluminescence prevented its wide usage for monitoring intracellular Ca 2 + dynamics. It was not until aequorin and fluorescent proteins were fused that the use of aequorin expanded [26], but by then, it had to compete with fluorescent Ca2+ indicators. Recently, other types of bioluminescent indicators such as BRAC [27] and Nano-lantern (Ca 2+) [28] were also developed by using Renilla luciferase derivatives. These indicators showed excellent performance under conditions in which external light excitation was undesirable, such as in highly autofluorescent plant tissues [27], or when coexpressed with the light-sensitive optogenetic tool channelrhodopsin-2 [28]. Despite increases in the number of indicators, little is known regarding the details underlying their functions, and schematic representations are still primarily used to explain observed properties. Thus, to further understand the mechanisms by which the indicators can work, we reviewed the composition and working principles of available fluorescent and bioluminescent GECIs (Section 2) and have briefly described the selection strategies used for fast preliminary evaluation of indicator variants (Section 3). Properties of interest for any GECI, including dynamic range, affinity, selectivity, and kinetics, are described in Section 4, together with other approaches aimed to obtain a detailed picture regarding the functioning of these indicators. 2. Reporting Ca 2+ with GECIs 2.1. Genetically encoded fluorescent Ca 2+ indicators Genetically encoded fluorescent Ca 2+ indicators can be categorized in two classes according to the number of fluorescent proteins present in the indicator: some GECIs contain a single fluorescent protein (non-ratiometric with some exceptions, Fig. 1) and some contain two fluorescent proteins (ratiometric probes, Fig. 2). Single fluorescent protein-based GECIs typically share a common principle of action involving a change in fluorescence intensity upon Ca2+ binding. Ca2+-chelating properties in most available indicators are provided by the calmodulin moiety, which is fused with the fluorescent protein and the calmodulin-binding peptide M13. An exception, which contains only a calmodulin fragment sandwiched between the two halves of a split GFP molecule, is the Camgaroo indicator family [12] (Fig. 1). Binding of Ca2+ promotes an intramolecular rearrangement that alters the chromophore protonation state and its associated fluorescent properties [29,30]. A completely different mechanism underlies the low-affinity Ca2+ sensor CatchER, in which the Ca 2+-binding motif formed by the introduction of five negatively charged residues in front of the tyrosine-derived hydroxyl group of the chromophore affects the chromophore electron density distribution, reducing the background fluorescence of this protein [21]. Bound Ca2+ shields this negative charge, allowing the recovery of fluorescence intensity. In all of these cases, the chromophore chemical structure is not modified during conformational rearrangement, and therefore, the change in fluorescence is associated with an increase or decrease in the absorption
spectra (or change in fluorescence quantum yield, in the case of inverse-pericam [13]), leading to a change in the resulting emitted fluorescence intensity. Since only a single wavelength is measured, these types of indicators are known as intensiometric or non-ratiometric (as opposed to the ratiometric indicators, described below). Monitoring Ca2+ dynamics using GECIs containing two fluorescent proteins takes advantage of the concept of FRET. During FRET, one of the fluorescent proteins acts as a donor that transfers absorbed energy to the second fluorescent protein, the acceptor, when excited (Fig. 2). The efficiency of energy transfer depends on several factors, including the overlap of donor emission and acceptor absorption spectra, donor–acceptor distance, donor lifetime, and relative orientation of the donor and acceptor transition dipole moments [31]. FRET efficiency is highly dependent on the distance between the donor and the acceptor moieties [31], and therefore, it can only be used at distances of 1–10 nm. In our case, GECIs are designed in such a way that after Ca2+ binding the FRET efficiency increases, either due to approaching the two fluorescent protein moieties each other or due to changing in their relative orientations. FRET-based techniques can be used to assess different aspects of donor–acceptor interactions [32,33], but for detecting changes in Ca 2 + levels, a simple donor–acceptor emission ratio is typically used: R ¼ IA =I D ;
ð2:1Þ
where R is the FRET ratio, IA and ID are the peak emission intensities of the donor and the acceptor, respectively. An oppositely directed intensity change at two emission wavelengths allows reductions in intensity fluctuations associated with varying sensor concentrations (to which single-wavelength indicators are more sensitive). Because the FRET ratio value provides information regarding Ca 2+ binding, these indicators are known as ratiometric indicators. The first intensiometric indicators were developed on the basis of Aequorea GFP-derivatives and showed spectral characteristics similar to those of GFP, but more recently developed non-ratiometric probes employ other GFP color variants or are based on other fluorescent proteins [17], thus allowing simultaneous multicolor Ca 2+ imaging in different intracellular compartments, the combined use of Ca2+-imaging with fluorescent-tag labeling, or simultaneous Ca2+ imaging with optogenetic manipulation. Several ratiometric indicators composed of different FRET pairs, such as blue-green [10,11], or green-orange [34], have been examined; however, the cyan-yellow pairs of fluorescent proteins provide the highest ratio change upon Ca 2+ binding. Thus, most currently available ratiometric GECIs use a combination of these proteins, primarily an enhanced cyan fluorescent protein such as ECFP [8] as a donor, and a yellow fluorescent protein such as Venus [35] or Citrine [36,15] as an acceptor. Brighter variants of cyan fluorescent proteins, such as Cerulean [37] and mTurquoise2 [38], were also tested as FRET donors; however, these sensors showed a lower dynamic range than those based on ECFP [19 and personal observations]. 2.2. Genetically encoded bioluminescent Ca 2+ indicators As described in the Introduction, aequorin was the first chemiluminescent protein employed to detect Ca2+ changes in vivo. Targeting of recombinant aequorin allowed monitoring of Ca2+ changes in several intracellular organelles, including the mitochondria, nucleus, sarcoplasmic reticulum, endoplasmic reticulum, Golgi, secretory granules, and gap junctions [39,40]. However, the low turnover rate of luminescent reaction [41] and low quantum yield with its associated difficulties in signal detection prevented widespread use of aequorin for Ca2+ imaging. The low signal level of aequorin was overcome through fusion with its natural counterpart, GFP (resulting in the “GA” sensor [26], Fig. 3, Table 3), which exploits the same bioluminescent resonance
V. Pérez Koldenkova, T. Nagai / Biochimica et Biophysica Acta 1833 (2013) 1787–1797
A
1789
B 2+
+Ca 2+
+Ca
2+
–Ca 2+
–Ca
GCaMPs
C
GECOs
D
2+
+Ca
2+
–Ca
2+
+Ca
Camgaroos
2+
–Ca
2+
+Ca
E
2+
+Ca
2+
–Ca
2+
–Ca
Pericams
CatchER
Fig. 1. Single-fluorescent protein Ca2+ indicator families. A. GCaMPs [14,18,69], B. GECOs [17], C. Pericams [13], D. Camgaroos [12,36], and E. CatchER [21]. Based on fluorescent proteins, these indicators require external excitation with light of different frequencies (wavy arrows). Binding of Ca2+ promotes intramolecular conformational rearrangement to changes the fluorescence intensity upon irradiation. Colors denote the fluorescence emission range characteristic of each family. Ca2+ affinity is conferred by Ca2+-binding moieties derived from calmodulin or by an artificially created low-affinity Ca2+-binding site in the case of CatchER (see details in Section 2.1 and indicator properties in Table 1).
energy transfer (BRET) principle underlying Aequorea jellyfish bioluminescence. Interestingly, Ca2+ appears to play an important role in increasing BRET efficiency in the GFP-aequorin fusion by promoting a shift in the GFP absorption spectra [42]. Currently, all bioluminescent GECIs are based on the resonance energy transfer mechanism, which is similar to FRET but uses a donor a chemiluminescent protein such as aequorin or a luciferase, rather than a fluorescent moiety that requires external light excitation. In addition to GFP, aequorin fusions with other fluorescent proteins have been reported, such as with the yellow fluorescent protein Venus [43], mRFP1 [43], and other red emitting fluorescent proteins [44], resulting in chemiluminescent Ca 2+ indicators with significantly longer emission wavelengths than those of aequorin alone. Luciferase-based bioluminescent GECIs were developed using design strategies previously utilized to create fluorescent GECIs. For the indicator BRAC [27], the RLuc8 moiety was fused to the C-terminus of calmodulin-M13, while the acceptor fluorescent protein Venus was fused to the N-terminal end (Fig. 3). Later, for the improved variant Nano-lantern (Ca2+) [28], the RLuc8 moiety was split in two parts by introducing the calmodulin-M13 fragment, allowing reconstitution of luciferase activity upon Ca2+ binding. As was done previously, Venus was fused to the N-terminal end of RLuc8 (Fig. 3) to increase the emission signal from the indicator through efficient BRET. The Ca2+ dependence of aequorin luminescence has been extensively analyzed [41]. The light flash emitted upon Ca2+ binding was shown to contain both slow and fast components, the proportion of which depends on the interaction between the three EF hands of aequorin. This
determines the amplitude of the emitted chemiluminescent signal [45]. Luminescence intensity, as well as its spectral characteristics and time course, also depend on the coelenterazine analog used as the light-emitting substrate. Therefore, aequorin complexes with different coelenterazine types can display drastically different apparent sensitivities for Ca2+ [46,47]. The Ca 2+-sensing range is also determined by the chosen aequorin variant [48]. In aequorin fusion constructs, as in aequorin alone, the emitted fluorescence signal is also determined by aequorin luminescence intensity; therefore, Ca2+ changes are reported as changes in signal intensity (in a non-ratiometric manner), despite the use of a resonance energy transfer mechanism. A similar mechanism is also characteristic of split RLuc8-based Nano-lantern (Ca2+) indicators [28], in which chemiluminescence depends on Ca2+ binding. In contrast, the indicator BRAC [27], which is based on the complete RLuc-8 moiety, can be used as a ratiometric probe because donor emission does not depend on Ca2+ concentration and only modulates the BRET ratio by changing the distance between the donor and the fluorescent acceptor protein. This uncouples the light-emitting and Ca2+-sensing mechanisms of this indicator. 3. Selection of GECI variants Useful GECIs can be obtained through a selection process that can vary depending on the goal. Typically, primary constructions or construction libraries are expressed in Escherichia coli strains, so that screening can be performed by analyzing the colonies formed, a cleared lysate of
1790
V. Pérez Koldenkova, T. Nagai / Biochimica et Biophysica Acta 1833 (2013) 1787–1797
A 2+
+Ca
2+
–Ca
2+
+Ca
2+
–Ca
Cameleons
If the number of samples to be screened is small, such as when circular permutation or site-specific random mutagenesis is used, selection of the most promising indicator variant is performed using the lysates of bacterial cultures, which are obtained by selecting individual colonies. Lysis is performed either mechanically by pressure shock or sonication or by enzymatic treatment in lysis buffer. After centrifugation to remove the cell debris, the cleared supernatant is added to solutions containing Ca 2+ or EGTA, and the fluorescence intensity change is measured. This simplified screening method also allows evaluation of spectral changes in the indicator under different levels of Ca2+ when it is technically difficult to conduct these measurements in colonies. To study in detail the properties of indicators selected after rapid initial screening, the indicators are subjected to purification after cell lysis. Purification is carried out by expressing a protein with a purification tag, which is typically included in the cloning vector. The polyhistidine tag is the most common tag used for purifying GECIs by using the affinity of this tag to the nickel or cobalt ions supplied in specific isolation media. 4. Evaluation of GECIs
B 2+
+Ca
2+
–Ca
TN Fig. 2. Two-fluorescent protein (FRET) Ca 2 + indicator families. A. Cameleons [11,16,20,34,35,72], B. TN indicators [19]. These indicators also require external excitation with light of different frequencies (wavy arrows). Binding of Ca2+ promotes intramolecular conformational changes that lead to the convergence of the donor and acceptor fluorescent proteins, allowing FRET to occur (see in Section 2.1). Colors denote the fluorescence emission ranges characteristic of each family. Ca2+ affinity is conferred by Ca2+-binding moieties derived from calmodulin (Cameleon family) or from troponin C (TN Ca2+-indicators) (see details in Section 2.1 and indicator properties in Table 2).
bacterial culture, or purified protein. Protein purification is a time-consuming process, and various techniques have been developed to perform fast initial recognition screening of the variants with better performance. Large-scale (~2 × 105 colonies) screening after random mutagenesis with error-prone PCR was implemented for the development of GECO indicators, by using GCaMP3 and pericam as a template [17]. A large number of variants were analyzed for Ca 2+-dependent fluorescence intensity changes that could not be determined by direct addition of Ca2+/EGTA solutions to bacterial cells expressing the proteins in the cytoplasm. However, individual culturing and analysis of such a large number of samples is time consuming. To allow screening of colonies, proteins were expressed using the TorA periplasmic export tag that targets the protein to the periplasmic space between the bacterial outer membrane, which is largely permeable to solutes, and the inner membrane [49]. This strategy allowed selection of colonies showing the highest change in fluorescence intensity produced by the spray application of 2 mM EGTA directly onto the agar plate containing the colonies. A similar approach can also be used to screen bright variants of bioluminescent GECIs, particularly those in which a circularly permuted or mutated variant of luciferase are used. Colonies can be sprayed with phosphate-buffered saline (PBS) containing coelenterazine at a concentration of 5 μg/mL [28]. Coelenterazine can easily pass through the plasma membrane, and thus, it can be directly applied to assess the fluorescence intensity of proteins expressed only in the bacterial cytoplasm.
In vitro evaluation of GECIs aims to not only determine their physicochemical properties and assess the applicability of the indicators for particular tasks, but also to compare the properties of different indicators. However, the biological environment in which GECIs are used for Ca2+ imaging can differ significantly from the standardized solutions used for in vitro evaluation, and therefore, analyzing performance in vivo under stimulation is carried out to determine the reproducibility of the Ca 2+ changes generated for the indicator. For fluorescent GECIs, absorption spectra are measured to determine the excitation wavelengths. For single-fluorescent protein-based indicators, the excitation wavelength corresponds to one of the absorption maximum peaks of the indicator, while for two-fluorescent protein-based (ratiometric) indicators, the excitation wavelength corresponds to the donor protein absorption maximum. Based on the emission spectra of the indicators, the specific wavelength (or two wavelengths in case of ratiometric indicators), at which fluorescence intensity is the highest, is determined. This emission wavelength(s) are used to record Ca2+-evoked fluorescence signals. For bioluminescent indicators, spectral measurements are complicated by the consumption of a light-emitting substrate, which potentially introduces fluctuations to the resulting intensity values due to changes in luminescence intensity related to substrate consumption. To avoid this, indicator properties are determined under conditions of a quasi-steady-state reaction in the presence of excess substrate. As an initial condition, coelenterazine concentrations ranging from 1 to 10 μM (95 μL [27,28]) can be tested with 5 μL of diluted solution (100 nM) of the bioluminescent indicator, providing a quasi-stable bioluminescent signal over sufficient amount of time to perform spectral measurements. These concentration and volume values, however, are dependent on the indicator turnover rate. Some properties of the bioluminescent GECIs are presented in Table 3. 4.1. Dynamic range and signal-to-noise ratio Dynamic range is one of the most important properties of GECIs, which is defined for non-ratiometric indicators as the maximal fluorescence intensity (typically in a Ca 2+-bound state) divided by minimal fluorescence intensity (determined in the presence of EGTA): Dnonratiometric ¼ I max =Imin ;
ð4:1Þ
where D is the dynamic range, and Imax and Imin are, respectively, are the maximal and minimal fluorescence intensities obtained under Ca 2 +-saturated or Ca 2 +-depleted conditions.
V. Pérez Koldenkova, T. Nagai / Biochimica et Biophysica Acta 1833 (2013) 1787–1797
1791
B
A
2+
2+
+Ca
2+
–Ca
+Ca
2+
–Ca
GFP-aequorin
Aequorin
D
C
2+
2+
+Ca
+Ca
2+
2+
–Ca
–Ca
Nano-lantern(Ca2+)
BRAC
Fig. 3. Bioluminescent Ca2+ indicators. A. Chemiluminescent protein aequorin [3], B. GFP-aequorin (GA) indicator [26], C. RLuc8-based indicator BRAC [27], and D. Split RLuc8-based Ca2+-indicator Nano-lantern (Ca2+) [28]. These indicators do not require external excitation with light, but instead require the light-emitting cofactor coelenterazine. The luminescent signal of aequorin (A), aequorin-based indicators (B), and Nano-lantern (Ca2+) (D) directly depends on Ca2+ binding, while Ca2+ binding in the case of BRAC (C) only modulates the energy transfer efficiency from the luciferase to the acceptor fluorescent protein (see details in Section 2.2). The colors denote the emission ranges of each indicator. Ca2+ affinity is conferred by intrinsic Ca2+-binding properties of aequorin or by a calmodulin moiety (BRAC and Nano-lantern (Ca2+) indicators) (see details in Section 2.2 and indicator properties in Table 3).
For ratiometric probes, dynamic range is calculated in a similar manner, typically as the maximum FRET ratio (observed in the Ca2+-bound state) divided by the minimal FRET ratio (in the virtual absence of Ca 2+): Dratiometric ¼ Rmax =Rmin ¼ IAðþCaÞ =IDðþCaÞ = I Að−CaÞ =I Dð−CaÞ ;
ð4:2Þ
where Rmax and Rmin are, respectively, the maximal and minimal FRET ratios calculated from the donor (ID(Ca)) and acceptor (IA(Ca)) intensities obtained under conditions of either Ca2+ saturation (IA(+Ca), ID(+Ca)) or Ca2+ depletion (IA(−Ca), ID(−Ca)). Thus, the dynamic range is a dimensionless measure that states how many times brighter the Ca 2+-bound state is than the Ca2+-free state or how many times the maximal FRET ratio is larger than the minimal observed FRET ratio. Dynamic range can be also expressed as a percent and can be determined as follows: D ¼ ðF max =F min −1Þ 100%;
ð4:3Þ
where F is either fluorescence intensity or FRET ratio. As can be observed in the equation, a two-fold increase of either fluorescence intensity or FRET ratio corresponds to a 100% increase in dynamic range. In living cells, resting Ca2+ concentrations are sometimes within the detectable range of the indicator (see detailed in 4.2 Affinity section), and therefore, the reported intracellular baseline fluorescence intensity (or FRET ratio) under resting conditions does not typically correspond to the minimal fluorescence (or FRET ratio) determined in vitro. We call here a measure of the signal reported by an indicator during transient Ca2+ change the signal-to-baseline ratio, or simply ΔF/F0 (measured ratio of the transient fluorescence response to baseline fluorescence, [50]): SBR ¼ ΔF=F 0 ¼ ðF–F 0 Þ=F 0 ;
ð4:4Þ
where the signal-to-baseline ratio (SBR) or the (ΔF/F0) parameter for an intensiometric indicator is determined as the fluorescence intensity
change registered during Ca2+ change (ΔF) divided by average baseline fluorescence (F0) under resting conditions [50]. A similar formula can be used to determine SBR for ratiometric indicators. Generally, the ability of an indicator to report Ca 2+ transient over background fluorescence fluctuations is determined on the basis of the signal-to-noise ratio (SNR) characteristic of the indicator, which is determined as the ratio between the fluorescence signal (ΔF) and the shot noise on the baseline fluorescence (~ F0/N 1/2 [50]) associated with the random character of the fluorescent molecule diffusion and variations in the detected number of photons within defined intervals of time even at a constant fluorescence intensity (see details in [51]): SNR ¼ ΔF=F 0 N
1=2
; also determined as ΔF=SDF0
ð4:5Þ
where N is the number of photons counted [52] and SDF0 is the standard deviation of the baseline fluorescence (previous to the Ca2+ transient). The inverse equation allows determination of the minimal number of photons that should be registered in order to obtain a SNR higher than 1 (i.e., to obtain a more than 100% increase over the background signal) [50]: −2
N > ðΔF=F 0 Þ
:
ð4:6Þ
In the experiment, SNR can be improved by background subtraction [50], as this component of the signal can be significant because of autofluorescence. In this case, the FRET ratio R can be calculated as follows: R ¼ I A −IAbg = ID −IDbg ;
ð4:7Þ
where (IA − IAbg) is the acceptor fluorescence intensity with the subtracted background intensity and (ID − IDbg) is the same calculation for the donor channel. SNR is particularly important when performing Ca 2+ imaging in very small compartments such as neuronal spines or, more generally,
1792
V. Pérez Koldenkova, T. Nagai / Biochimica et Biophysica Acta 1833 (2013) 1787–1797
when the indicator concentration is low and the photon emission process becomes stochastic. GECIs are characterized by a poorer SNR than synthetic Ca2+-sensitive dyes due to their approximately 10-times higher resting baseline fluorescence [52]. Unfortunately, a low signal-to-noise ratio is typically observed in these experiments, and therefore, various strategies are used to increase SNR. One strategy involves increasing the expression of the indicator by using strong promoters, but under some conditions, this strategy can lead to adverse effects in the expression system [53]. Another strategy includes statistical analysis of the obtained FRET data to isolate a meaningful signal [54,55]. An important consideration is the indicator fractional saturation reflecting the fraction of a GECI population bound to Ca 2+ that does not participate in the detection of a Ca 2+ transient. This property depends both on Ca2+ concentration present in the study environment and on indicator Ca2+-binding properties (see Eq. (4.8)). If fractional saturation is high (due to a high Ca2+ concentration of an indicator with a particular Ca2+ affinity), a further increase in Ca2+ will produce a very small change in SNR. To evaluate fractional saturation of the indicator in vivo after Ca 2+ a transient measurement, the effective range of the indicator is evaluated by Ca2+ depletion and saturation (addition of EGTA and a high Ca2+ concentration in the presence of an agent increasing the membrane permeability, such as ionomycin). Fractional saturation can be determined as follows: Θ ¼ ðF–F min Þ=ðF max –F min Þ;
ð4:8Þ
where Θ is fractional saturation, F is fluorescence intensity or FRET ratio signal, Fmin and Fmax are the fluorescence intensities or FRET ratio values obtained under Ca2+-depleted and Ca2+-saturated conditions, respectively [52]. The relationship between the dependence of fluorescence intensities (or FRET ratios) on the Ca2+ concentration is determined using the following equation: h i h i 2þ 2þ þ Kd ; ðF–F min Þ=ðF max –F min Þ ¼ Ca = Ca
ð4:9Þ
where [Ca2+] is the Ca2+ concentration measured in the sample and Kd is the dissociation constant of the Ca2+-indicator complex, which is used to describe the GECI Ca2+-binding properties; this is further analyzed in Section 4.2. 4.2. Affinity Reactions contributing to the changes in the optical properties of a GECI upon Ca 2+ binding can be expressed using a model describing binding of a ligand to a protein. In the case of a single Ca 2+ binding, this equation will look as follows: 2þ kon
G þ Ca
2þ
⇄ G−Ca :
koff
ð4:10Þ
Under equilibrium conditions, the system can be described using the Ca2+ indicator dissociation constant Kd characterized by the on- (kon) and off-rates (koff) of the forward and reverse reactions, respectively: K d ¼ koff =kon :
ð4:11Þ
This macroscopic “average” constant reflects Ca2+ binding to the EF hands, conformational changes induced by such binding, and donoracceptor interactions leading to resonance energy transfer and reflecting an apparent Ca2+ concentration at which the indicator reports its halfmaximal fluorescence intensity or FRET ratio. The Kd value allows comparison of different indicators in a qualitative manner, and provides an idea of the approximate range of Ca2+ concentration in which a GECI can function. However, it does not specify
the range of Ca2+ concentrations that the indicator can sense (its capacity) and does not provide information regarding the kinetic properties of the indicator. The relationship between the indicator fractional saturation, Ca2+ concentration, and indicator apparent Kd value is expressed using the Hill equation: h i h i 2þ n 2þ n = Ca þ Kd ; Θ ¼ Ca
ð4:12Þ
where n is the Hill constant, and the other variables have the same meaning as described above. The Hill constant provides information regarding the steepness of the indicator fluorescence change at a Ca2+ concentration equal to the indicator Kd constant, and thus indicates the degree of cooperativity of binding of each subsequent Ca2+. Theoretically, the Hill constant can have values higher, lower, or equal to 1, corresponding to cases in which binding of the first Ca2+ promotes, inhibits, or does not affect the binding of the next Ca2+. In practice, however, the Hill constant for most GECIs has a value larger than 1 and is determined on the basis of cooperativity in Ca 2+ binding to the N- and C-terminal EF hand pairs of calmodulin- or troponin C-derived moieties [56]. This is likely why a Hill constant value of approximately 2 is characteristic for Ca2+ binding for many two-fluorescent protein GECIs (see in [22,35,57] and Tables 1, 2). However, similar to the “averaged” Kd constant, the value of n also includes components not directly related to Ca2+ binding. Such an “averaged” Hill constant can provide information regarding the range at which the Ca2+ indicator can report Ca2+ concentration around the Kd value; as noted above, a higher value of n (characteristic for some single-fluorescent protein GECIs, Table 1 and [18]) indicates a higher increase in fluorescence intensity or FRET ratio per Ca 2+ concentration unit change, but at the same time, it implies a narrowed Ca2+ range when compared with an indicator of a similar dynamic range that has a lower Hill constant (see in [58]). Interestingly, binding of a single Ca 2+ in the low-affinity Ca2+-sensitive GFP is reflected as a Hill constant almost equal to 1 [21], but a similar n value is also characteristic of some cameleon variants with a redesigned calmodulin-M13 interface (Table 2 and [16]). This may reflect the influence of intramolecular conformational changes on the otherwise cooperative Ca 2+ binding to calmodulin. Because the different Ca2+-binding sites on GECIs can have very different affinities, the fluorescence intensity or the FRET ratio of a particular indicator is not always well described by the “1 Ca2+ ion per indicator” or the single-binding site Hill equation described above. A two-component Hill equation is used in this case to fit the data: F
h
2þ
Ca
i
¼
α1 α2 n þ n ; 1 þ K d1 = Ca2þ 1 þ K d2 = Ca2þ
ð4:13Þ
where α1/2 refer to the fractions of total fluorescence of the FRET ratio response corresponding to different apparent dissociation constants Kd1 and Kd2. The difference between these individual Kd values can be significant (see yellow cameleon 4.60, Table 2 and [35], see also D1cpV [16]), and while such indicators are advantageous because of the increased Ca2+ range sensed (up to hundreds of micromoles in the cases of YC 4.60 and D1cpV), this increase comes at a cost of a reduced dynamic range at each Ca2+ concentration range, implying a reduced SNR. Kd value(s) for Ca 2+ can be determined by measuring indicator fluorescence intensity or the FRET ratio in a series of free Ca2+ concentration which can be prepared by reciprocal mixing of EGTA and Ca2+-EGTA buffers [59]. The data are normalized to the maximal and minimal fluorescence intensities or FRET ratio value (defined as 1 under Ca2+ saturation and 0 under Ca2+ depletion conditions, respectively) and fitted using any of the Hill equations presented. Typically, values obtained in vitro provide an idea of the Kd magnitude in vivo, but differences in these environments appear to be result in discrepancies. It has been well established that ionic strength, ionic
V. Pérez Koldenkova, T. Nagai / Biochimica et Biophysica Acta 1833 (2013) 1787–1797 Table 1 In vitro properties of single-FP genetically encoded Ca2+ indicators, by family.
Em — emission wavelength(c), D — dynamic range, KD — dissociation constant, n — Hill constant, τrt — time constant of the dissociation reaction determined at room temperature (20–25 °C), pKa — acidity constant determined at room temperature (20–25 °C) [70,71]. When the koff value is specified in the reference instead of τ, the last is calculated from the relationship τ = Θ /koff, assuming Θ = 1. The table includes data obtained from [22]. a Values respectively in Ca2+ depleted/saturated conditions, b Values from different references are separated by a vertical bar |. c Same as above.
composition, and pH can drastically affect the affinity of GECIs for Ca2+ when expressed in vivo ([52] and references therein). After determining the apparent Kd value in vitro and the maximum FRET ratio in situ, Ca2+ concentration can be calculated on the basis of the signal reported by a GECI, as proposed by Grynkiewicz et al. [6]: h
Ca
2þ
i
¼ K d ðSf2 =Sb2 ÞðR–Rmin Þ=ðRmax −RÞ;
ð4:14Þ
where Sf2/Sb2 represents the ratio of the acceptor fluorescence intensities at its emission wavelength in Ca2+-depleted and Ca 2+-bound states, respectively. This original equation was developed for a 1:1 Ca2+:indicator binding ratio, and thus, a slightly different equation was proposed by Adams et al. [60] for measuring Ca 2+ by using yellow cameleon variants [11,16]: h
Ca
2þ
i
1=n
¼ K d ððR–Rmin Þ=ðRmax –RÞÞ
:
1793
Ca 2+/Mg 2+ selectivity of the Ca 2+-binding motifs used in GECIs, derived from calmodulin and troponin C, constitutes a wide research field (for example, see [62]). Nevertheless, the selectivity of GECIs has been analyzed in very few cases, and primarily for troponin-based variants since this Ca 2+-binding protein is known to bind Mg 2+ in an non-selective manner ([19] and references therein). Mg2+ sensitivity is conferred by specific residues within the EF hand motif [22] that were replaced to develop the family of Mg2+-insensitive troponin-based Ca2+ indicators [15,22]. Selectivity should be determined under conditions similar to environmental conditions in which the GECI will be used. For Ca2+ indicators residing in the cytoplasm, selectivity can first be assayed in solutions containing either Ca 2+ or Mg2+ (concentrations around 1 mM in a background of 100–150 mM KCl in a pH buffer) and compared with values obtained in solutions containing the Ca2+ and Mg2+ chelators EGTA and EDTA. If Mg 2 + can provoke a change in the fluorescence intensity or FRET ratio similar to the changes evoked in the media containing Ca 2 +, then selectivity should be assayed in solutions containing physiologically relevant concentrations of both cations. A combination of 50–100 nM to 20 μM of Ca 2 + (respectively, for resting and Ca 2 +-elevated conditions) containing 0.5 mM to 1–5 mM Mg 2 + will likely fit well to the proportion of Ca 2 + and Mg 2 + present in the cytoplasm of a typical cell. 4.4. Kinetic properties As can be observed in formula (Eq. (4.10)), the Kd value is determined by dividing the rate constant for the reverse reaction (koff, s −1) with that for the forward reaction (kon, M−1 s−1). These constants can be used to characterize not only the rate of Ca2+ binding/unbinding to the EF hands, but also the kinetic properties of the induced intramolecular conformational change, the kinetic of the calmodulin–M13 interaction (when present), and other allosteric effects associated with fluorescent protein interaction in ratiometric GECIs. It is not surprising, therefore, that rate constants for both forward and reverse reactions induced by Ca2+ binding/unbinding in GECIs reflect significantly slower response kinetics for these indicators than in the case of synthetic Ca 2+-sensitive dyes. Table 2 In vitro properties of two-FP genetically encoded Ca2+ indicators.
ð4:15Þ
For high-affinity indicators (or indicators located in intracellular compartments), measurements of Rmin in situ can be problematic; so, to calculate Ca2+ concentration, the use of dynamic range rather than FRET ratio value was proposed (adapted from [61] for multiple binding sites): h
Ca
2þ
i
1=n
¼ K d ððR=Rmax –1=DÞ=ð1–R=Rmax ÞÞ
;
ð4:16Þ
where D is the dynamic range (Rmax/Rmin). Based on this equation, changes in [Ca2+] from baseline values can be estimated with regard to the obtained SNR (see details in [61]). 4.3. Selectivity Free Ca 2+ concentration in the cytoplasm is in the order of tensto-hundreds of nanomoles. Meanwhile, another common divalent cation, Mg 2+, is present at a concentration of hundreds of micromoles to millimoles, competing with Ca 2+ for its binding sites.
Em — emission wavelengths, D — dynamic range, KD — dissociation constant, n — Hill constant, τrt — time constant of the dissociation reaction determined at room temperature (20–25 °C) [70,71]. When the koff value is specified in the reference instead of τ, the last is calculated from the relationship τ = Θ /koff, assuming Θ = 1. The table includes data obtained from [22]. a Emission wavelengths of, respectively, donor (acceptor) fluorescent proteins, b Values from different references are separated by a vertical bar |. c Values obtained from two-component fittings are separated by a forward slash /, d Same as above. e [Value] obtained at 30 °C.
1794
V. Pérez Koldenkova, T. Nagai / Biochimica et Biophysica Acta 1833 (2013) 1787–1797
Table 3 Properties of bioluminescent genetically encoded Ca2+ indicators.
Em — emission wavelengths, D — dynamic range, KD — dissociation constant, n — Hill constant, τrt — time constant of the dissociation reaction determined at room temperature (20–25 °C), half-life — time during which the protein concentration in the cytosol is reduced by 50% [73,74]. When the koff value is specified in the reference instead of τ, the last is calculated from the relationship τ = Θ /koff, assuming Θ = 1. a Emission wavelength, or emission wavelengths of donor(acceptor) moieties. b Values from different references are separated by a vertical bar |. c Same as above. d N-l — nano-lantern (Ca2+). e Measurements in 150 mM KCl+ 1 mM Mg2+.
The kinetics of Ca 2 + binding to EF hands is no more than 10 times slower than binding to Ca 2 +-sensitive dyes (see [22]). More detailed analyses have shown that binding of Ca 2 + to troponin occurs with a kon of 10 8 M − 1 s − 1, and koff values of 500–800 s − 1 for dissociation from N-terminal lower-affinity EF hands, while the koff for highaffinity C-terminal EF hands is of a magnitude about 1–2 orders lower, corresponding to a slower off-reaction. For calmodulin, much faster Ca 2 +-binding and unbinding from the EF hands was observed. The dissociation constant koff is more important for determining the shape of the Ca2+ transient, while kon acts as a scaling factor for the indicator response [63,64]. The apparent dissociation of Ca 2+ in GECIs is much slower than for synthetic dyes; for the latter, koff values of 100–370 s−1 are characteristic (τoff b 10 ms [65,22], while for GECIs, τoff values range from 142 ms (the fastest two-fluorescent protein TN-XXL [19]) to 300–1000 ms (koff ~ 1–3 s−1) for single fluorescent GECIs (Table 1) and higher values for two-fluorescent indicators (koff ~ 0.3 s−1; Table 2 and [20,22]). It seems plausible to associate the fast off reaction of TN-XXL with the introduction of a second bending site at the C-terminal part of each repeated Ca2+-binding motif between the donor and acceptor fluorescent proteins, while in conventional calmodulin and troponin, there is only one bending site within the indicator core, presumably located in the central part of the calmodulin moiety. Moreover, it is interesting that the low-affinity Ca 2 +-sensitive fluorescent protein CatchER, in which the bound Ca 2 + ion can directly affect chromophore fluorescence, shows a time-constant and off-rate very similar to those of Ca 2 +-sensitive dyes; 70% of the fluorescence changes occurs within 2.2 ms with a koff of 700 s − 1 [21]. The kinetic properties of an indicator should be considered when an accurate representation of the Ca2+ transient is required, particularly in rapid processes. This likely drives improvement of indicators aimed to monitor neural activity. Various strategies can be employed to improve the kinetic properties of GECIs, including 1) directly linking the chromophore response to Ca2+ binding, thus eliminating the waiting time required for intramolecular conformational changes to occur, or 2) make this conformational change Ca2+-sensitive at most of the steps to increase the cooperativity and to reduce the time between Ca 2+ binding and the indicator response associated with such binding (i.e., start: Ca2+ binding→ bending of the indicator → binding of peptide→ approaching fluorescent proteins or conformational change within the protein→ result: fluorescent property change).
4.5. Other biochemical and biophysical properties Large intramolecular changes that induce changes in GECI optical properties are reasonably well understood. However, clarifying the details requires structural knowledge, as well as information regarding the dynamics of this structure. The 3D structure of several single-fluorescent protein GECIs has been resolved in several studies [18,30,66] to obtain improved GCaMP variants by using a structure-guided design strategy [18]. Nevertheless, no structures are available regarding FRET-based indicators and the folding mechanisms of the fluorescent moieties on which the sensors are based remain unknown. To identify the molecular-scale processes associated with Ca2+ reporting, various techniques can be used. An example of a very detailed examination is that conducted for the TN-XXL sensor [67]. In this study, Ca2+ binding to each of the EF hands was analyzed by using steady-state and stopped flow spectroscopy as well as by nuclear magnetic resonance. For these experiments, only the troponin-derived Ca2+-binding moiety was used. To determine the effect of the provoked conformational change on donor and acceptor fluorescence, either the donor or acceptor was replaced by a similar fluorescent protein with an “Amber”-like mutation, resulting in a non-fluorescent moiety to allow the analysis of the signal from an individual fluorescent protein. An analysis of fluorescence lifetime revealed the absence of a non-functional fraction of sensors. However, more interestingly, hydrodynamic properties, studied by using size-exclusion chromatography and analytical ultracentrifugation, provided insight regarding the indicator shape change after Ca2+ binding, while experiments involving small-angle X-ray scattering (SAXS) allowed modeling of the shape of TN-XXL in the presence and absence of Ca2+. Other properties of GECIs for determining whether a particular indicator can be used under specific conditions include pH sensitivity and temperature sensitivity. Cytosolic pH is typically strictly controlled, but during processes such as programmed cell death, pH may be reduced. At the same time, some of the fluorescent proteins within a specific indicator can be more sensitive to such environmental changes, introducing artifacts for reported Ca2+ values. Thus, when designing a new or choosing an existing GECI variant, the effect of the measurement conditions on the properties of the fluorescent protein moiety or moieties employed should first be evaluated. Preliminary evaluation is particularly important when a GECI is used for monitoring Ca 2+ dynamics within intracellular organelles, which can be characterized on the basis of the levels of pH or Ca 2+ that differ significantly from those observed in the cytoplasm. 5. Conclusions Currently, GECI properties are not comparable to those of Ca 2+sensitive dyes. However, GECIs offer some advantages not achievable with synthetic Ca2+ probes, including precise targeting, prolonged imaging over days and months, and simultaneous multicolor Ca2+ imaging. GECIs are becoming a valuable tool for monitoring the activity of in situ neurons [25], and are now also used for feedback of lightevoked excitation using optogenetic tools [28,68]. Signals reported by GECIs will always be affected by both the used indicator properties and the environment in which it is used. Distinguishing signals requires an understanding not only of indicator properties, but also of the environment in which Ca 2+ imaging is performed. In vivo evaluation maybe the most important test for GECIs, but good performance in one cell type does not guarantee similar behavior in other cell types. The extent to which a particular GECI can fulfill the needs of an experimenter depends only on the goal of the study. In Table 4, as a general reference source, we present common applications for which GECIs are used, including Ca2+ monitoring and measurement. However, GECI use is not restricted only to these tasks; their use as Ca 2+-buffering agents can allow determination of the role of the Ca 2+ signal in specific signal cascades.
V. Pérez Koldenkova, T. Nagai / Biochimica et Biophysica Acta 1833 (2013) 1787–1797
1795
Table 4 Selection of an appropriate GECI indicator for specific applications.
Refer to Tables 1–3 and the text for detailed properties of each indicator. Indicators in (parenthesis) can potentially be used in the specified conditions due to their properties, but their performance in those conditions was not evaluated. Deep-in-tissue Ca2+ monitoring requires two-photon excitation of fluorescent Ca2+ indicators [75–87]. Monitoring Ca2+ with aequorin-based indicators requires an accurate selection of the appropriate aequorin mutant and light-emitting substrate (see in [46,47]). a Can potentially be saturated at higher Ca2+ concentrations. b D4 (D4cpv with Citrine instead of circular permuted Venus as an acceptor fluorescent protein [16]) performance will depend on the pH value in the monitored environment. Lower pH values can drastically reduce the fluorescence, impeding to perform the experiment. Indicators containing Citrine instead of cpVenus have a lower dynamic range. c CatchER will not work in an acidic pH environment (e.g., plant apoplast) due to pH-sensitivity. d High-affinity GECIs tend to slow the reported kinetics of fast Ca2+ transients. e Long-time experiments can result in incorrect estimations, due to changes in the indicator concentration (as a result of expression/degradation, or bleaching). f Consumption of light-emitting substrate should be taken into account. Long-time experiments (over several minutes) can require constant substrate supply. The indicator calibration inside organelles can be challenging.
Acknowledgements We thank the members of the laboratory of T.N. for their valuable comments. This work was partly supported by MEXT Grant-in-Aid for Scientific Research on Innovative Areas (23115003) and PRESTO, JST to T.N. References [1] E. Carafoli, The calcium-signaling saga: tap water and protein crystals, Nat. Rev. Mol. Cell Biol. 4 (2003) 326–332. [2] E.B. Ridgway, C.C. Ashley, Calcium transients in single muscle fibres, Biochem. Biophys. Res. Commun. 29 (1967) 229–234. [3] O. Shimomura, F.H. Johnson, Y. Saiga, Extraction, purification and properties of aequorin, a bioluminescent protein from the luminous hydromedusan, Aequorea, J. Cell. Comp. Physiol. 59 (1962) 223–239.
[4] N. Hsiung, H. Warrick, J.K. de Riel, D. Tuan, B.G. Forget, A. Skoultchi, R. Kucherlapati, Cotransfer of circular and linear prokaryotic and eukaryotic DNA sequences into mouse cells, Proc. Natl. Acad. Sci. U. S. A. 77 (8) (1980) 4852–4856. [5] R.Y. Tsien, T. Pozzan, T.J. Rink, Calcium homeostasis in intact lymphocytes: cytoplasmic free calcium monitored with a new, intracellularly trapped fluorescent indicator, J. Cell Biol. 94 (2) (1982) 325–334. [6] G. Grynkiewicz, M. Poenie, R.Y. Tsien, A new generation of Ca2+ indicators with greatly improved fluorescence properties, J. Biol. Chem. 260 (6) (1985) 3440–3450. [7] R.Y. Tsien, A non-disruptive technique for loading calcium buffers and indicators into cells, Nature 290 (1981) 527–528. [8] R. Heim, R.Y. Tsien, Engineering green fluorescent protein for improved brightness, longer wavelengths and fluorescence resonance energy transfer, Curr. Biol. 6 (2) (1996) 178–182. [9] T. Porumb, P. Yau, T.S. Harvey, M. Ikura, A calmodulin-target peptide hybrid molecule with unique calcium-binding properties, Protein Eng. 7 (1996) 109–115. [10] V.A. Romoser, P.M. Hinkle, A. Persechini, Detection in living cells of Ca2+-dependent changes in the fluorescent emission of an indicator composed of two green
1796
[11]
[12]
[13]
[14] [15]
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23] [24] [25] [26]
[27]
[28]
[29] [30]
[31] [32] [33] [34]
[35]
[36]
[37] [38]
V. Pérez Koldenkova, T. Nagai / Biochimica et Biophysica Acta 1833 (2013) 1787–1797 fluorescent protein variants linked by a calmodulin-binding sequence, J. Biol. Chem. 272 (1997) 13270–13274. A. Miyawaki, J. Llopis, R. Heim, J.M. McCaffery, J.A. Adams, M. Ikura, R. Tsien, Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin, Nature 388 (1997) 882–887. G.S. Baird, D.A. Zacharias, R.Y. Tsien, Circular permutation and receptor insertion within green fluorescent proteins, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 11241–11246. T. Nagai, A. Sawano, E.S. Park, A. Miyawaki, Circularly permuted green fluorescent proteins engineered to sense Ca2+, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 3197–3202. J. Nakai, M. Ohkura, K. Imoto, A high signal-to noise Ca2+ probe composed of a single green fluorescent protein, Nat. Biotechnol. 19 (2001) 137–141. N. Heim, O. Griesbeck, Genetically encoded indicators of cellular calcium dynamics based on troponin C and green fluorescent protein, J. Biol. Chem. 279 (2004) 14280–14286. A.E. Palmer, M. Giacomello, T. Kortemme, S.A. Hires, V. Lev-Ram, D. Baker, R.Y. Tsien, Ca2+ indicators based on computationally redesigned calmodulin-peptide pairs, Chem. Biol. 13 (5) (2006) 521–530. Y. Zhao, A. Satoko, J. Wu, T. Teramoto, Y.-F. Chang, M. Nakano, A.S. Abdelfattah, M. Fujiwara, T. Ishihara, T. Nagai, R.E. Campbell, An expanded palette of genetically encoded Ca indicators, Science 333 (6051) (2011) 1888–1891. J. Akerboom, T.-W. Chen, T.J. Wardill, L. Tian, J.S. Marvin, S. Mutlu, N. Carreras Calderón, F. Esposti, B.G. Borghuis, X.R. Sun, A. Gordus, M.B. Orger, R. Portugues, F. Engert, J.J. Macklin, A. Filosa, A. Aggarwal, R.A. Kerr, R. Takagi, S. Kracun, E. Shigetomi, B.S. Khakh, H. Baier, L. Lagnado, S.S.-H. Wang, C.I. Bargmann, B.E. Kimmel, V. Jayaraman, K. Svoboda, D.S. Kim, E.R. Schreiter, L.L. Looger, Optimization of a GCaMP calcium indicator for neural activity imaging, J. Neurosci. 32 (40) (2012) 13819–13840. M. Mank, D.F. Reiff, N. Heim, M.W. Friedrich, A. Borst, O. Griesbeck, A FRET-based calcium biosensor with fast signal kinetics and high fluorescence change, Biophys. J. 90 (2006) 1790–1796. K. Horikawa, Y. Yamada, T. Matsuda, K. Kobayashi, M. Hashimoto, T. Matsu-ura, A. Miyawaki, K. Mikoshiba, T. Nagai, Spontaneous network activity visualized by ultra-sensitive Ca 2+ indicators, yellow Cameleon nano, Nat. Methods 7 (2010) 729–732. S. Tang, H.-C.h. Wong, Zh.-M. Wang, Y. Huang, J. Zou, Y. Zhuo, A. Pennati, G. Gadda, O. Delbono, J.J. Yang, Design and application of a class of sensors to monitor Ca2+ dynamics in high Ca2+ concentration cellular compartments, Proc. Natl. Acad. Sci. U. S. A. 108 (39) (2011) 16265–16270. M. Mank, O. Griesbeck, Genetically encoded calcium indicators, Chem. Rev. 108 (5) (2008) 1550–1564, http://dx.doi.org/10.1021/cr078213v. M. Whitaker, Genetically-encoded probes for measurement of intracellular calcium, Methods Cell Biol. 99 (2010) 153–182. A.E. Palmer, Y. Qin, J.G. Park, J.E. McCombs, Design and application of genetically encoded biosensors, Trends Biotechnol. 29 (3) (2011) 144–152. C. Grienberger, A. Konnerth, Imaging calcium in neurons, Neuron 73 (5) (2012) 862–885. V. Baubet, H. Le Mouellic, A.K. Campbell, E. Lucas-Meunier, P. Fossier, P. Brûlet, Chimeric green fluorescent protein-aequorin as bioluminescent Ca 2+ reporters at the single-cell level, Proc. Natl. Acad. Sci. U. S. A. 97 (13) (2000) 7260–7265. K. Saito, N. Hatsugai, K. Horikawa, K. Kobayashi, T. Matsu-Ura, K. Mikoshiba, T. Nagai, Auto-luminescent genetically-encoded ratiometric indicator for real-time Ca2+ imaging at the single cell level, PLoS One 5 (4) (2010) e9935. K. Saito, Y.-F. Chang, K. Horikawa, N. Hatsugai, Y. Higuchi, M. Hashida, Y. Yoshida, T. Matsuda, Y. Arai, T. Nagai, Luminescent proteins for high-speed singlecell and whole body imaging, Nat. Commun. 3 (2012), http://dx.doi.org/10.1038/ ncomms2248. Q. Wang, B. Shui, M.I. Kotlikoff, H. Sondermann, Structural basis for calcium sensing by GCaMP2, Structure 16 (12) (2008) 1817–1827. J. Akerboom, J.D. Rivera, M.M. Guilbe, E.C. Malavé, H.H. Hernandez, L. Tian, S.A. Hires, J.S. Marvin, L.L. Looger, E.R. Schreiter, Crystal structures of the GCaMP calcium sensor reveal the mechanism of fluorescence signal change and aid rational design, J. Biol. Chem. 284 (10) (2009) 6455–6464. J.R. Lackowicz, Principles of Fluorescence Spectroscopy, third ed. Springer, Singapore, 2006. E.A. Jares-Erijman, T.M. Jovin, FRET imaging, Nat. Biotechnol. 21 (2003) 1387–1395. C. Berney, FRET or no FRET: a quantitative comparison, Biophys. J. 84 (6) (2003) 3992–4010. M. Waldeck-Weiermair, M.R. Alam, M.J. Khan, A.T. Deak, N. Vishnu, F. Karsten, H. Imamura, W. Graier, R. Malli, Spatiotemporal correlations between cytosolic and mitochondrial Ca2+ signals using a novel red-shifted mitochondrial targeted cameleon, PLoS One 7 (9) (2012) e45917. T. Nagai, S. Yamada, T. Tominaga, M. Ichikawa, A. Miyawaki, Expanded dynamic range of fluorescent indicators for Ca2+ by circularly permuted yellow fluorescent proteins, Proc. Natl. Acad. Sci. U. S. A. 101 (29) (2004) 10554–10559. O. Griesbeck, G.S. Baird, R.E. Campbell, D.A. Zacharias, R.Y. Tsien, Reducing the environmental sensitivity of yellow fluorescent protein, J. Biol. Chem. 276 (2001) 29188–29194. M.A. Rizzo, G.H. Springer, B. Granada, D.W. Piston, An improved cyan fluorescent protein variant useful for FRET, Nat. Biotechnol. 22 (4) (2004) 445–449. J. Goedhart, D. von Stetten, M. Noirclers-Savoye, M. Lelimousin, L. Joosen, M.A. Hink, L. van Weeren, T.W.J. Gadella, A. Royant, Structure-guided evolution of cyan fluorescent proteins toward a quantum yield of 93%, Nat. Commun. 3 (2012) 751, http: //dx.doi.org/10.1038/ncomms1738.
[39] R. Rizzuto, M. Brini, T. Pozzan, Intracellular targeting of the photoprotein aequorin: a new approach for measuring, in living cells, Ca2+ concentrations in defined cellular compartments, Cytotechnology 11 (Suppl. 1) (1993) S44–S46. [40] M. Brini, L. Pasti, C. Bastianutto, M. Murgia, T. Pozzan, R. Rizzuto, Targeting of aequorin for calcium monitoring in intracellular compartments, J. Biolumin. Chemilumin. 9 (3) (1994) 177–184. [41] O. Shimomura, Bioluminescence, Chemical Principles and Methods, Mainland Press Pte Ltd., Singapore, 2008. [42] A.Yu. Gorokhovatsky, V.V. Marchenkov, N.V. Rudenko, T.V. Ivashina, V.N. Ksezenko, N. Burkhardt, G. Semisotnov, L.M. Vinokurov, Yu.B. Alakhov, Fusion of Aequorea victoria GFP and aequorin provides their Ca2+-induced interaction that results in red shift of GFP absorption and efficient bioluminescence energy transfer, Biochem. Biophys. Res. Commun. 320 (2004) 703–711. [43] T. Curie, K.L. Rogers, C. Colasante, P. Brûlet, Red-shifted aequorin-based bioluminiscent reporters for in vivo imaging of Ca 2+ signaling, Mol. Imaging 6 (1) (2007) 30–42. [44] A. Bacayan, C.F. Vaquero, F. Picazo, J. Llopis, Red fluorescent protein-aequorin fusions as improved bioluminescent Ca2+ reporters in single cells and mice, PLoS One 6 (5) (2011) e19520. [45] L. Tricoire, K. Tsuzuki, O. Courjean, N. Gibelin, G. Bourout, J. Rossier, B. Lambolez, Calcium dependence of aequorin bioluminescence dissected by random mutagenesis, Proc. Natl. Acad. Sci. U. S. A. 103 (25) (2006) 9500–9505. [46] O. Shimomura, B. Musicki, Y. Kishi, Semi-synthetic aequorins with improved sensitivity to Ca2+ ions, Biochem. J. 261 (1989) 913–920. [47] A.L. Miller, E. Karplus, L.F. Jaffe, Imaging [Ca2+]i with aequorin using a photon imaging detector, R. Nuccitelli (Ed.), A practical guide to the study of calcium in living cellsMethods Cell Biol. 40 (1994) 305–338. [48] M. Montero, M.T. Alonso, E. Carnicero, I. Cuchillo-Ibañez, A.G. García, J. García-Sancho, J. Alvarez, Chromaffin-cell stimulation triggers fast millimolar mitocondrial Ca2+ transients that modulate secretion, Nat. Cell Biol. 2 (2000) 57–61. [49] C.M. Barret, N. Ray, J.D. Thomas, C. Robinson, A. Bolhuis, Quantitative export of a reporter protein, GFP, by the twin-arginine translocation pathway in Escherichia coli, Biochem. Biophys. Res. Commun. 304 (2) (2003) 279–284. [50] R. Yasuda, E.A. Nimchinsky, V. Scheuss, T.A. Pologruto, T. Oertner, B.L. Sabatini, K. Svoboda, Imaging calcium concentration dynamics in small neuronal compartments, Sci. STKE 219 (2004) l5, http://dx.doi.org/10.1126/stke.2192004pl5. [51] H. Qian, On the statistics of fluorescence correlation spectroscopy, Biophys. Chem. 38 (1990) 49–57. [52] S.A. Hires, L. Tian, L.L. Looger, Reporting neural activity with genetically encoded calcium indicators, Brain Cell Biol. 36 (1–4) (2008) 69–86. [53] L. Tian, S.A. Hires, T. Mao, D. Huber, M.E. Chiappe, S.H. Chalasani, L. Petreanu, J. Akerboom, S.A. McKinney, E.R. Schreiter, C.I. Bargmann, V. Jayaraman, K. Svoboda, L.L. Looger, Imaging neural activity in worms, flies and mice with improved GCaMP indicators, Nat. Methods 6 (2009) 875–881. [54] J. Broder, A. Majumder, E. Porter, G. Srinivasamoorthy, C. Keith, J. Lauderdale, A. Sornborger, Estimating weak ratiometric signals in imaging data. I. Dual-channel data, J. Opt. Soc. Am. 24 (9) (2007) 2921–2931. [55] X. Fan, A. Majumder, S.S. Reagin, E.L. Porter, A.T. Sornborger, C.H. Keith, J.D. Lauderdale, New statistical methods enhance imaging of cameleon fluorescence resonance energy transfer in cultured zebrafish spinal neurons, J. Biomed. Opt. 12 (3) (2007) 034017. [56] N.V. Valeyev, D.G. Bates, P. Heslop-Harrison, I. Postlenthwaite, N.V. Kotov, Elucidating the mechanisms of cooperative calcium–calmodulin interactions: a structural systems biology approach, BMC Syst. Biol. 2 (2008) 48. [57] T. Hendel, M. Mank, B. Schnell, O. Griesbeck, A. Borst, D.F. Reiff, Fluorescence changes of genetic calcium indicators and OGB-1 correlated with neural activity and calcium in vivo and in vitro, J. Neurosci. 28 (29) (2008) 7399–7411. [58] J. Weiss, The Hill equation revisited: uses and misuses, FASEB J. 11 (1997) 835–841. [59] R. Tsien, T. Pozzan, Measurements of cytosolic free Ca2+ with Quin2, Methods Enzymol. 172 (1989) 230–262. [60] S.R. Adams, B.J. Bacskai, S.S. Taylor, R.Y. Tsien, Optical probes for cyclic AMP, in: W.T. Mason (Ed.), Fluorescent Probes for Biological Activity of Living Cells—A Practical Guide, Academic Press, 1993, pp. 133–149. [61] M. Maravall, Z.F. Mainen, B.L. Sabatini, K. Svoboda, Estimating intracellular calcium concentrations and buffering without wavelength ratioing, Biophys. J. 78 (2000) 2655–2667. [62] J.L. Gifford, M.P. Walsh, H.J. Vogel, Structures and metal-ion-binding properties of the Ca2+-binding helix-loop-helix EF-hand motifs, Biochemistry 405 (2) (2007) 199–221. [63] A. Struk, G. Szücs, H. Kemmer, W. Meizer, Fura-2 calcium signals in skeletal muscle fibres loaded with high concentrations of EGTA, Cell Calcium 23 (1) (1998) 23–32. [64] M.G. Klein, B.J. Simon, G. Szücs, M.F. Schneider, Simultaneous recording of calcium transients in skeletal muscle using high and low-affinity calcium indicators, Biophys. J. 53 (1988) 971–988. [65] A.L. Escobar, P. Velez, A.M. Kim, F. Cifuentes, M. Fill, J.L. Vergara, Kinetic properties of DM-nitrophen and calcium indicators: rapid transient response to flash photolysis, Pflügers Arch.–Eur, J. Physiol. 434 (1997) 615–631. [66] L. Leder, W. Stark, F. Freuler, M. Marsh, M. Meyerhofer, T. Stettler, L.M. Mayr, O.V. Britanova, L.A. Strukova, D.M. Chudakov, E.A. Souslova, The structure of Ca2+ sensor Case16 reveals the mechanism of reaction to low Ca2+ concentrations, Sensors 10 (9) (2010) 8143–8160. [67] A. Geiger, L. Russo, T. Gensch, T. Thestrup, S. Becker, K.-P. Hopfner, C. Griesinger, G. Witte, O. Griesbeck, Correlating calcium binding, Förster resonance energy transfer, and conformational change in the biosensor TN-XXL, Biophys. J. 102 (2012) 2401–2410.
V. Pérez Koldenkova, T. Nagai / Biochimica et Biophysica Acta 1833 (2013) 1787–1797 [68] M. Ohkura, T. Sasaki, C. Kobayashi, Y. Ikegaya, J. Nakai, An improved genetically encoded red fluorescent Ca2+ indicator for detecting optically evoked action potentials, PLoS One 7 (7) (2012) e39933. [69] Y.N. Tallini, M. Ohkura, B.R. Choi, G. Ji, K. Imoto, R. Doran, J. Lee, P. Plan, J. Wilson, H.B. Xin, A. Sanbe, J. Gulick, J. Mathai, J. Robbins, G. Salama, J. Nakai, M.I. Kotlikoff, Imaging cellular signals in the heart in vivo: cardiac expression of the high-signal Ca 2+ indicator GCaMP2, Proc. Natl. Acad. Sci. U. S. A. 103 (12) (2006) 4753–4758. [70] D.F. Reiff, A. Ihring, G. Guerrero, E.Y. Isacoff, M. Joesch, J. Nakai, A. Borst, In vivo performance of genetically encoded indicators of neural activity in flies, J. Neurosci. 25 (2005) 4766–4768. [71] M. Ohkura, M. Matsuzaki, H. Kasai, K. Imoto, J. Nakai, Genetically encoded bright Ca2+ probe applicable for dynamic Ca2+ imaging of dendritic spines, Anal. Chem. 77 (18) (2005) 5861–5869. [72] A. Miyawaki, O. Griesbeck, R. Heim, R.Y. Tsien, Dynamic and quantitative Ca2+ measurements using improved cameleons, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 2135–2140. [73] J.M. Kendall, G. Sala-Newby, V. Ghalaut, R.L. Dormer, A.K. Cambell, Engineering the Ca2+-activated photoprotein aequorin with reduced affinity for calcium, Biochem. Biophys. Res. Commun. 187 (1992) 1091–1097. [74] S. de la Fuente, R.I. Fonteriz, P.J. de la Cruz, M. Montero, J. Alvarez, Mitochodrial free [Ca2+] dynamics with a novel low-Ca2+ affinity aequorin probe, Biochem. J. 445 (2012) 371–376. [75] G. Haspel, M.J. O'Donovan, A. Hart, Motoneurons dedicated to either forward or backward locomotion in the nematode C. elegans, J. Neurosci. 30 (33) (2010) 11151–11156. [76] H. Lütcke, M. Murayama, T. Hahn, D.J. Margolis, S. Astori, S. Meyer zum Alten Borgloh, W. Göbel, Y. Yang, W. Tang, S. Kügler, R. Sprengler, T. Nagai, A. Miyawaki, M.E. Larkum, F. Helmchen, M.T. Hasan, Optical recording of neuronal activity with a genetically-encoded calcium indicator in anesthesized and freely moving mice, Front. Neural Circuits 4 (9) (2010), http://dx.doi.org/10.3389/fncir.2010.00009. [77] M. Iwano, T. Entani, H. Shiba, M. Kakita, T. Nagai, H. Mizuno, A. Miyawaki, T. Shoji, K. Kubo, A. Isogai, S. Takayama, Fine-tuning of the cytoplasmic Ca 2 +
[78]
[79]
[80] [81]
[82] [83]
[84]
[85]
[86]
[87]
1797
concentration is essential for pollen tube growth, Plant Physiol. 150 (3) (2009) 1322–1334. L.H. Tay, I.E. Dick, W. Yang, M. Mank, O. Griesbeck, D.T. Yue, Nanodomain Ca2+ of Ca2+ channels detected by a tethered genetically encoded Ca2+ sensor, Nat. Commun. 3 (2012) 778, http://dx.doi.org/10.1038/ncomms1777. M. Mank, A.F. Santos, S. Direnberger, T.D. Mrsic-Flogel, S.B. Hofer, V. Stein, T. Hendel, D.F. Reiff, C. Levelt, A. Borst, T. Bonhoeffer, M. Hübener, O. Griesbeck, A genetically encoded calcium indicator for chronic in vivo two-photon imaging, Nat. Methods 5 (9) (2008) 805–811. T. Pozzan, R. Rudolf, Measurements of mitochondrial calcium in vivo, Biochim. Biophys. Acta Bioenerg. 1787 (11) (2009) 1317–1323. M. Giacomello, I. Drago, M. Bortolozzi, M. Scorzeto, A. Gianelle, P. Pizzo, T. Pozzan, Ca2+ hot spots on the mitochondrial surface are generated by Ca2+ mobilization from stores, but not by activation of store-operated Ca2+ channels, Mol. Cell 38 (2010) 280–290. A. Mithöfer, C. Mazars, Aequorin-based measurements of intracellular Ca2+signatures in plant cells, Biol. Proced. Online 4 (2002) 105–118. N. Mehlmer, N. Parvin, C.H. Hurst, M.R. Knight, M. Teige, U.C. Vothknecht, A toolset of aequorin expression vectors for in planta studies of subcellular calcium concentrations in Arabidopsis thaliana, J. Exp. Bot. 63 (4) (2012) 1751–1761. K.L. Rogers, S. Picaud, E. Roncali, R. Boisgard, C. Colasante, J. Stinnakre, B. Tavitian, P. Brûlet, Non-invasive in vivo imaging of calcium signaling in mice, PLoS One 2 (10) (2007) e974, http://dx.doi.org/10.1371/journal.pone.0000974. Y. Yamada, T. Michikawa, M. Hashimoto, K. Horikawa, T. Nagai, A. Miyawaki, M. Häusser, K. Mikoshiba, Quantitative comparison of genetically encoded Ca2+ indicators in cortical pyramidal cells and cerebellar Purkinje cells, Front. Cell. Neurosci. 5 (2011), (article 18). R. Rizzuto, A.W.M. Simpson, M. Brini, T. Pozzan, Rapid changes of mitochondrial Ca2+ revealed by specifically targeted recombinant aequorin, Nature 358 (1992) 325–327, http://dx.doi.org/10.1038/358325a0. Z.M. Wang, S. Tang, M.L. Messi, J.J. Yang, O. Delbono, Residual sarcoplasmic reticulum Ca2+ concentration after Ca2+ release in skeletal myofibers from young, adult and old mice, Pflugers Arch. 463 (4) (2012) 615–624.