Neuroscience Institute, National Research Council, Padua Section, 35121 Padua, ItalyVenetian Institute of Molecular Medicine, 35129 Padua, ItalyDepartment of Biomedical Sciences, University of Padua, 35121 Padua, Italy
Fluorescent sensors classes.Class I: Ligand-binding sensors. This first class accounts for probes in which the binding of the ligand results in changes of the physicochemical properties of the sensor such as: (A) FPs that intrinsically respond to a parameter of interest, e.g., pH probes; (B) FPs, in which the binding domain is inserted in the sequence of an FP; (C) intramolecular reaction, as in chemiluminescence, where binding of the ligand causes the oxidation of the cofactor and photon emission. Class II: FRET-based sensors. Includes probes where ligand binding changes the FRET efficiency between the two chromophores. They are subdivided into (A) intermolecular FRET sensors; (B) intramolecular FRET sensors; and (C) BRET, the donor chromophore is a chemiluminescent protein and the acceptor is an FP. Class III: Translocation- or accumulation-based biosensors. (A) Translocation probe: the GFP-PKC example. The GFP-tagged protein kinase C accumulates at the PM upon increases in DG levels at the PM. (B) Accumulation probe: the TMRM example. This dye accumulates within mitochondria depending on their membrane potential. For other examples such as PM potential dyes that reorient in the plane of the membrane after changes or nonfluorescent probes, see Targeting organic sensors to subcellular compartments (PM section).
Targeted synthetic sensors. (A) Transmission of [Ca2+]c changes into the mitochondrial matrix of single hepatocytes challenged with a maximal dose of vasopressin (VP). [Ca2+]m was monitored with dihydro-Rhod 2-AM, and [Ca2+]c was measured with Fura 2-AM. The inset shows cell population mean responses for [Ca2+]c and [Ca2+]m. The figure is modified from Hajnóczky et al. (1995) with permission from Elsevier. (B) PM potential measured using the FRET-based sensor. (left) Scheme of the voltage-sensitive FRET mechanism. At resting negative membrane potential (top), the permeable oxonols have a high concentration at the extracellular surface of the PM, and energy transfer from the extracellularly bound FL-WGA (acceptor molecule) is favored. FRET is symbolized as the straight arrow from lectin to oxonol. Upon membrane depolarization (bottom), the anions diffuse inside the cell and energy transfer is greatly reduced. (right) Confocal image of a voltage-clamped astrocytoma cell at −70 (A) and 50 mV (B), stained with FLOX6 (from González and Tsien  with permission from Elsevier). (C) Mitochondrial membrane potential measured using TMRM. Cerebellar granule neurons loaded with TMRM (left) were exposed to the uncoupler FCCP and monitored over 60 min using TMRM in both the quenched (100 nM, middle) and nonquenched (30 nM, right) mode (from Ward et al.  with permission from the Society for Neuroscience). (D) [Ca2+]ER measured with Mag-Fura2. (left) Pseudo-color ratio image of permeabilized BHK-21 cells loaded with mag-Fura2. (middle) Same cell of left panel after stimulation with InsP3, causing release of Ca2+ from the ER. (right) Representative kinetic of mag-Fura2 ratio collected from selected areas of the cell (from Hofer et al.  with permission from the Federation of American Societies for Experimental Biology). (E) Exocytosis from large dense core vesicles (LDCVs) measured using FFN511. (left) Chemical structure of FFN511. (middle) Multiphoton image of a chromaffin cell shows distribution of FFN511 in LDCVs. Bar, 5 µm. (right) FFN511 exocytosis from an LDCV observed with total internal reflection fluorescence microscopy (TIRFM) images. The upper row shows consecutive images of a single vesicle. Orthogonal section through this vesicle and its integrated intensity are in the middle and bottom panels. The dotted line indicates stimulation by high potassium (from Gubernator et al.  with permission from The American Association for the Advancement of Science).
Schematic representation of targeting strategies. Targeting strategies used to achieve selective mitochondria-targeted biosensors (A), ER-targeted probes (B), Golgi-targeted sensors (C), and nuclear-targeted sensors (D) and localization of the available sensors.
Schematic representation of targeting strategies. Targeting strategies used to generate probes selectively targeted to PM (A), endosomes (B), lysosomes (C), secretory granules/synaptic vesicles (D), and peroxisomes (E) and localization of the available probes.
Ca2+dynamics in different cellular compartments. (A) Cytosolic and nuclear Ca2+ dynamics evaluated using cytosolic green aequorin (CytGA) and nuclear red aequorin (NucRA). (A, inset) CytGA and NucRA localization in HEK293T cells. Bar, 10 µm. (A) Representative kinetics of [Ca2+]C and [Ca2+]N (from Manjarrés et al.  with permission from Springer). (B) Microscopic SR calcium release, the so-called Ca2+ sparks, evaluated using Fluo-3 (from Cannell and Kong  with permission from Elsevier). (C) Measurements of nuclear and ER Ca2+dynamics using the D3 and D4 Cameleons variants, respectively. (inset) D4ER fluorescence (green). Bar, 10 µm. Representative kinetics of nuclear (gray) and ER (black) fluorescent signals in a single BHK cell coexpressing H2B-D3cpv and D4ER and stimulated with bradykinin (BK) and the SERCA inhibitor, cyclopiazonic acid (CPA), in a Ca2+-free medium (from Greotti et al.  with permission from MDPI AG). (D) SR Ca2+ release and cytosolic Ca2+ increase in flexor digitorum brevis (FDB) monitored using the FRET-based D4cpv-Casq1 SR sensor and the cytosolic dye X-rhod-1. FDB was voltage-clamped and subjected to depolarization as indicated. (left, top) Normalized line scan of fluorescence of X-rhod-1 (cytosolic signal). (bottom) FRET ratio (R(x,t)) of D4cpv-Casq1 (SR signal). (right) Plot of the line averages (from Manno et al. [2013b] with permission from The Physiological Society).
The heterogeneity of GA in Ca2+handling. (A) [Ca2+] and molecular toolkit along the secretory pathway. The GA can be divided in three distinct subcompartments: the cis-Golgi, with a luminal [Ca2+] around 250 µM, the medial Golgi, with a luminal [Ca2+] lower compared with that of the cis-Golgi (i.e., ∼150–200 µM), and trans-Golgi, with a luminal [Ca2+] around 130 µM. The efflux and influx Ca2+ toolkit is also shown. TGN, trans-Golgi network; SV, secretory vesicles (from Pizzo et al.  with permission from Elsevier). (B–E) Ca2+ handling by medial Golgi in intact cells monitored with a targeted Cameleon probe: (B) the fluorescence microscope image of a medialGo-D1cpv–expressing SH-SY5Y cell. Bar, 10 µm. (C–E) SHSY-5Y cells were incubated in medium supplemented or not with 1 mM CaCl2 or 300 µM EGTA and challenged with the indicated stimuli: (C) bradykinin (BK), demonstrating the presence of an IP3 sensitive pool; (D) cyclopiazonic acid (CPA), demonstrating the presence of the SERCA pump; (E) ionomycin (Iono), a ionophore demonstrating the presence of another molecular component besides IP3Rs and SERCA, such as SPCA1 (from Wong et al.  with permission from Oxford University Press). (F–H) Ca2+ handling by trans-Golgi in single intact cells monitored with a targeted Cameleon probe: (F) confocal microscopy image of a cardiomyocyte cell expressing transGo-D1cpv and the mRFP-Zasp construct (red). Bar, 10 µm. (G and H) HeLa cells (G) and cardiac myocytes (H) were exposed to different stimuli demonstrating that this compartment is enriched of SPCA1 (ionomycin-sensitive pool) and RyRs (caffeine-sensitive pool), but neither SERCA (CPA-sensitive pool) nor IP3Rs (histamine-sensitive pool) are present (from Lissandron et al.  with permission from the National Academy of Sciences).
Peroxisomal Ca2+dynamics. Peroxisomal Ca2+ dynamics explored using a targeted Cameleon. (A) Colocalization of transiently expressed D3cpv-SKL and the peroxisome marker catalase in HeLa cells. Bar, 10 µm. (B and C) Increases in [Ca2+]c are followed by a slow rise in intraperoxisomal [Ca2+]. Fluorescence changes of GH3 cells transiently expressing D3cpv-KVK-SKL selectively within peroxisomes (two cells, dashed and dotted traces), mistargeted to the cytosol (continuous trace; B), or loaded with fura-2 (C). Where indicated, 30 mM KCl and 2 mM EGTA were added. (D) Cells permeabilized with digitonin. The experiment shows that no driving force supplied by ATP is needed for Ca2+ to enter peroxisomes (from Drago et al.  with permission from The American Society for Biochemistry and Molecular Biology).
Mitochondrial [Ca2+] hotspots. Pixel by pixel correlation showing the presence of Ca2+ microdomains on mitochondrial outer membrane (OMM) upon stimulation. HeLa cells cotransfected with nuclear (H2BD1cpv)- and OMM (N33D1cpv)-targeted Cameleons were treated with 100 µM histamine where indicated by the arrow in A. (A) The average [Ca2+] rises, expressed as ΔR/R0, of nucleus (green trace) and OMM (blue trace) are very similar. (B) The number and intensity of single pixels, expressed as ΔRmax/ΔR0 of single pixels, of the nucleus (green) or OMM (blue) reached during the first 4 s after histamine stimulation were plotted. The Gaussian fit of the nuclear (green) and OMM (red) pixel distribution highlights the right tail caused by hot-spot formation on OMM upon cellular stimulation. (C) Yellow to red color representation of ΔRmax/R0 spatial distribution of pixels in 2D (left) and 3D (right), superimposed to YFP fluorescence image. Only pixels that during the 4 s after the histamine challenge have a ΔRmax/R0 exceeding by 125% the ΔR/R0 of the whole compartment were color coded. Bar, 5 µm (from Giacomello et al.  with permission from Elsevier).
Intracellular resting [Ca2+] and GEI-targeted mechanism. In resting conditions, cells maintain a [Ca2+] gradient between the cytosol ([Ca2+] ∼100 nM) and the extracellular medium ([Ca2+] ∼1.5–2 mM) and some organelles. The [Ca2+] of each compartment is color coded (bar on the right). The figure also includes a schematic representation illustrating the targeting mechanisms used to target GEIs to the different cell compartments.
cAMP signaling. (A–C) Localized increases of cAMP. (A) Scanning ion conductance microscopy image and corresponding cAMP level obtained in cardiomyocytes expressing Epac1-cAMP sensor upon local β2-adrenergic receptors (β2AR) stimulation in the cell crest (B) and in the T-tubule (C). Cells were stimulated with a β2AR agonist (ISO and CGP) in the presence of a β1AR antagonist, showing β2-cAMP signals only in T-tubules, whereas no cAMP signal was detected in cell crest (from Nikolaev et al.  with permission from The American Association for the Advancement of Science). (D and E) cAMP diffusion is low in the cytoplasm of adult rat ventricular myocytes. GEI for cAMP EpacSH187 (from Klarenbeek et al.  with permission from PLOS) was expressed in myocytes stimulated with β-adrenoceptor agonist and antagonist to investigate cAMP diffusion. (D) Time course of FRET ratio in proximal and distal ROIs (top) and pseudocolor representative image (bottom); (E) longitudinal profile of FRET ratio, relative to boundary position, measured under resting conditions (light gray squares), during the last 10 s of β-agonist and β-antagonist additions (black circles) to one end of the cell, and during the last 10 s of uniform exposure to β-agonist (dark gray triangles). The experiment shows the low cAMP diffusion in the cytoplasm of adult ventricular myocyte (from Richards et al.  with permission from Oxford University Press). (F–H) Kinetics of cAMP levels and cAMP-dependent phosphorylation in the cytosol and on the OMM. (F) cAMP and mitochondria. Representative kinetics of intramitochondrial cAMP levels (expressed as ΔR/R0 changes) in control cells or in cells overexpressing MCU (mitochondrial Ca2+ uniporter). Where indicated, the cells were stimulated with an IP3-generating agonist (ATP) and a SERCA pump inhibitor (tert-Butylhydroquinone [TBHQ]) to induce a massive Ca2+ accumulation within the mitochondrial matrix. The data demonstrate that the amplitude of the cAMP increase in the mitochondria depends on the amplitude of the Ca2+ increase within the matrix (from Di Benedetto et al.  with permission from Elsevier). (G) Kinetics of cAMP-dependent phosphorylation of the sensor AKAR4 localized in the cytosol or on the OMM; (H) kinetics of the cAMP increase with sensors localized in the cytosol or on the OMM. The data demonstrate that the kinetics of the sensor phosphorylation is very different in the cytosol and OMM (G), whereas the amplitude and kinetics of cAMP levels in the two compartments are very similar (H; from Lefkimmiatis et al.  with permission from The Rockefeller University Press).
Table 2.Summary of some of the most used organelle-targeted indicators
Pellegatti et al., 2005
Imamura et al., 2009
Vishnu et al., 2014
Gajewski et al., 2003; Imamura et al., 2009; Nakano et al., 2011; Lefkimmiatis et al., 2013
Marsault et al., 1997; Mao et al., 2008
Brini et al., 1993; Bengtson et al., 2010; Giacomello et al., 2010; Simonetti et al., 2013
Kendall et al., 1992; Palmer et al., 2004; Tang et al., 2011; Kipanyula et al., 2012; de la Fuente et al., 2013; Rodriguez-Garcia et al., 2014; Suzuki et al., 2014; Henderson et al., 2015; Rodríguez-Prados et al., 2015; Waldeck-Weiermair et al., 2015; Navas-Navarro et al., 2016
Pinton et al., 1998; Lissandron et al., 2010; Wong et al., 2013; Aulestia et al., 2015; Rodríguez-Prados et al., 2015
Rizzuto et al., 1992, 1994; Arnaudeau et al., 2001; Griesbeck et al., 2001; Nagai et al., 2001; Rapizzi et al., 2002; Filippin et al., 2003, 2005; Palmer et al., 2006; Giacomello et al., 2010; Park et al., 2010; Alam et al., 2012; Jean-Quartier et al., 2012; Loro et al., 2012; Akerboom et al., 2013
Mitchell et al., 2001; Shen et al., 2012; McCue et al., 2013; Albrecht et al., 2015; Ronco et al., 2015
Drago et al., 2008; Lasorsa et al., 2008
DiPilato et al., 2004; Dyachok et al., 2006; Terrin et al., 2006
DiPilato et al., 2004
DiPilato et al., 2004; Terrin et al., 2006; Di Benedetto et al., 2013; Lefkimmiatis et al., 2013
Sato et al., 2006
Sato et al., 2006
Sato et al., 2006
Sato et al., 2006
Harvey et al., 2008
Fehr et al., 2004
Fehr et al., 2005
Hydrogen Peroxide (H2O2)
Malinouski et al., 2011
Merksamer et al., 2008
Belousov et al., 2006; Malinouski et al., 2011; Ermakova et al., 2014
Malinouski et al., 2011
Phosphoinositides (PI3P2, PI3P3)
Ananthanarayanan et al., 2005
Sato et al., 2005; Kapoor et al., 2014
Barros et al., 2013
Lindenburg et al., 2013b
Zhao et al., 2011
Hung et al., 2011; Zhao et al., 2011; Bilan et al., 2014
Mukhtarov et al., 2013
Llopis et al., 1998; Miesenböck et al., 1998; Wu et al., 2000
Kneen et al., 1998; Llopis et al., 1998; Abad et al., 2004; Porcelli et al., 2005; Azarias and Chatton, 2011; Poburko et al., 2011; Tantama et al., 2011
Miesenböck et al., 1998; Ohara-Imaizumi et al., 2002; Zhu et al., 2009; Li and Tsien, 2012
Jankowski et al., 2001; Lasorsa et al., 2008
Allen and Zhang, 2006; Depry et al., 2011
Allen and Zhang, 2006
Liu et al., 2011
Allen and Zhang, 2006; Lefkimmiatis et al., 2013
Gallegos et al., 2006
Gallegos et al., 2006
Gallegos et al., 2006
Gallegos et al., 2006
van Lith et al., 2011
Hanson et al., 2004; Gutscher et al., 2008; Liu et al., 2012a; Albrecht et al., 2014
Yano et al., 2010
Storace et al., 2016
Qin et al., 2011
Qin et al., 2011
Dittmer et al., 2009; Miranda et al., 2012
Vinkenborg et al., 2009
References are provided for original works or reviews. For most of these sensors, the cytosolic forms are available and are not included in this table. Sensors for glutamate (Marvin et al., 2013), pyruvate (San Martín et al., 2014), G protein activation (van Unen et al., 2016), cGMP (Russwurm et al., 2007), and phosphate (Gu et al., 2006) are currently available only in the form localized in the cytosol.