Locating Inositol 1,4,5-Trisphosphate in the Nucleus and Neuronal

Science, The University of Tokyo, Shirokane-dai, Minato-ku, Tokyo 108-8639, Japan .... The conserved Arg241-Glu439 salt bridge determines flexibil...
0 downloads 0 Views 475KB Size
Download PDF
Anal. Chem. 2005, 77, 4751-4758

Locating Inositol 1,4,5-Trisphosphate in the Nucleus and Neuronal Dendrites with Genetically Encoded Fluorescent Indicators Moritoshi Sato,†,‡ Yoshibumi Ueda,† Masabumi Shibuya,§ and Yoshio Umezawa*,†

Department of Chemistry, School of Science, The University of Tokyo, Japan Science and Technology Agency, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan, PRESTO, Japan Science and Technology Agency, 4-1-8 Honcho Kawaguchi, Saitama, Japan, and Division of Genetics, Institute of Medical Science, The University of Tokyo, Shirokane-dai, Minato-ku, Tokyo 108-8639, Japan

Inositol 1,4,5-trisphosphate (InsP3) is a key second messenger in many cell types and also in distinct subcellular regions of single living cells; however, little is examined about the subcellular dynamics of InsP3 in a variety of cell types. We have developed fluorescent indicators to locate InsP3 dynamics in single living cells based on an intramolecular fluorescence resonance energy transfer. Our indicator has visualized InsP3 dynamics in the cytoplasm of cultured cells and even in single thin dendrites of hippocampal neurons, which has been unseen previously. We have further localized the present indicator in the nucleus and pinpointed nuclear InsP3 dynamics. The observation with our nuclear InsP3 indicator has solved a question on nuclear propagation of InsP3 from the cytoplasm and has drawn a conclusion that the nuclear InsP3 dynamics synchronously occurs with cytosolic InsP3 dynamics evoked by agonist stimulations. The present approach contributes to the understanding of when, where, and how InsP3 is generated and removed in a variety of living cells. Stimulation with diverse biological stimuli, such as neurotransmitters and hormones, activates phospholipase Cs (PLCs), which then induces hydrolysis of phosphatidylinositol 4,5-bisphosphate (PtdIns(4,5)P2) to generate InsP3.1,2 To measure the generation of InsP3 in living cells, Hirose et al. previously developed a fusion protein (GFP-PHD) of green fluorescent protein (GFP) and pleckstrin homology domain (PHD) from PLCδ.3 They showed that this GFP-PHD, which is bound to PtdIns(4,5)P2 at the plasma membrane, translocates from the plasma membrane to the cytoplasm in response to an increased concentration of InsP3. Simultaneous detection of fluorescence intensities of both the plasma membrane and cytoplasm was required for the analysis * To whom correspondence should be addressed: Phone: +81-3-5841-4351. Fax: +81-3-5841-8349. E-mail: [email protected]. † School of Science, The University of Tokyo, and Japan Science and Technology Agency. ‡ PRESTO. § Institute of Medical Science, The University of Tokyo. (1) Berridge, M. J.; Irvine, R. F. Nature 1984, 312, 315-321. (2) Irvine, R. F. Nat. Rev. Mol. Cell Biol. 2003, 4, 586-590. (3) Hirose, K.; Kadowaki, S.; Tanabe, M.; Takeshima, H.; Iino, M. Science 1999, 284, 1527-1530. 10.1021/ac040195j CCC: $30.25 Published on Web 06/15/2005

© 2005 American Chemical Society

of InsP3 generation with the GFP-PHD. van der Wal et al. have previously developed two fusion proteins, each having cyan fluorescent protein (CFP) and yellow fluorescent protein (YFP), named CFP-PHD and YFP-PHD.4 Both the CFP-PHD and YFP-PHD translocate from the plasma membrane to the cytoplasm in response to InsP3 concentrations; likewise, the GFPPHD, decreasing intermolecular fluorescence resonance energy transfer (FRET) from the CFP-PHD protein to YFP-PHD protein. However, several shortcomings remain to be overcome for these previous methods. The GFP-PHD and the pair of CFPPHD and YFP-PHD respond not only to the InsP3 generation but also to the decrease in PtdIns(4,5)P2 concentration, and the difference in PtdIns(4,5)P2 concentration between cells does not allow the comparison of InsP3 concentration between cells.5 In addition, the previous methods do not allow the detection of InsP3 concentrations in subcellular compartments, such as the nucleus and gap junctions. In the present study, to overcome these limitations of the previous methods, we have developed fluorescent indicators for quantitatively visualizing InsP3 dynamics in living cells on the basis of an intramolecular FRET approach6-8 (Figure 1). RESULTS AND DISCUSSION Design of the Present Fluorescent Indicator for InsP3. The most important target of generated InsP3 is probably InsP3 receptors (InsP3Rs) that predominantly localize at membranes of the endoplasmic reticulum.9-11 So far, their structures and functions have been extensively studied.12 Type 1 InsP3R (InsP3R1) is a polypeptide with three major functionally distinct regions: the amino-terminal InsP3-binding region, the central modulatory region, and the carboxy-terminal channel region. A crystal (4) van der Wal, J.; Habets, R.; Varnai, P.; Balla, T.; Jalink, K. J. Biol. Chem. 2001, 276, 15337-15344. (5) Xu, C.; Watras, J.; Loew, L. M. J. Cell Biol. 2003, 161, 779-791. (6) Miyawaki, A.; Tsien, R. Y. Methods Enzymol. 2000, 327, 472-500. (7) Sato, M.; Ueda, Y.; Takagi, T.; Umezawa, Y. Nat. Cell Biol. 2003, 5, 10161022. (8) Sato, M.; Umezawa, Y. Methods 2004, 32, 451-455. (9) Mikoshiba, K. Curr. Opin. Neurobiol. 1997, 7, 339-345. (10) Mikoshiba, K.; Hattori, M. Sci. STKE 2000, pe1 2000. (11) Vermassen, E.; Parys, J. B.; Mauger, J.-P. Biol. Cell 2004, 96, 3-17. (12) Taylor, C. W.; da Fonseca, P. C. A.; Morris, E. P. Trends Biochem. Sci. 2004, 29, 210-219.

Analytical Chemistry, Vol. 77, No. 15, August 1, 2005 4751

Figure 1. Fluorescent indicators for InsP3 in single living cells. (a) Principle of fretino for visualizing InsP3. CFP and YFP are differently colored mutants of green fluorescent protein from Aequorea victoria with mammalian codons and additional mutation. Upon binding of InsP3 with the hIP3R1224-579 within fretino, a conformational change of fretino takes place, which changes the efficiency of FRET from CFP to YFP. (b) Schematic representations of domain structures of the present fretinos. Shown at the top of each bar are the restriction sites. hIP3R1224-579 and hIP3R11-604 are derived from a InsP3-binding domain of human IP3R1 and selectively bind with InsP3. R504Q shows a mutation in which the arginine 504 is replaced with a glutamine to partly reduce the affinity of the mutant hIP3R1224-579 with InsP3. K508A represents a mutation in which the lysine 508 is replaced with an alanine to inhibit the binding of InsP3 with the mutant hIP3R1224-579. NLS stands for a nuclear localization sequence, the amino acid sequence of which is shown at the bottom.

structure of the amino-terminal InsP3-binding region of InsP3R1 shows that the region has the asymmetric and boomerang-like structure consisting of an N-terminal β-trefoil domain and a C-terminal R-helical domain containing an armadillo repeat-like fold.13 We sandwiched this InsP3-binding region of InsP3R1 (InsP3R1224-579) with CFP and YFP, which are, respectively, the donor and acceptor for FRET (Figure 1a). Upon binding with InsP3, this fusion protein is expected to change its conformation, which accompanies a change in the distance or orientation (13) Bosanac, I.; et al. Nature 2002, 420, 696-700.

4752 Analytical Chemistry, Vol. 77, No. 15, August 1, 2005

between CFP and YFP. This InsP3-dependent conformational change of the fusion protein results in the change in the efficiency of intramolecular FRET from CFP to YFP (Figure 1a). Monitoring the fluorescence of both CFP and YFP should allow the quantitative detection of InsP3 concentrations in living cells. We named this FRET-based indicator for inositol trisphosphate, fretino (Figure 1b). Response of Fretino for InsP3. At first, we examined the FRET response of fretino for InsP3. cDNA encoding fretino was constructed (Figure 1b). We also constructed cDNA encoding

fretino-2, in which the arginine 504 was substituted by a glutamine14 to partly reduce the affinity of the indicator for InsP3 (Figure 1b). The cDNA encoding fretino and that encoding fretino-2 were separately transfected in Madin-Darby canine kidney (MDCK) cells. Cells expressed with fretino and those with fretino-2 were permeabilized by treating with a solution containing a detergent, saponin, to introduce InsP3 into the cells. When various concentrations of InsP3 were added one after the other to the extracellular solution of the fretino-2-expressed cells, a stepwise increase in an emission ratio of CFP to YFP (CFP/YFP) was observed (Figure 2a). This increase in the CFP/YFP emission ratio indicates that FRET from CFP to YFP is decreased upon binding of InsP3 with fretino-2. The change in the CFP/YFP emission ratio was dependent upon the InsP3 concentration, with an apparent dissociation constant of 1.9 × 10-7 M (Figure 2b). Fretino showed a response to InsP3 with an apparent dissociation constant of 7.6 × 10-9 M (Figure 2b), indicating that fretino has an ∼25-fold higher affinity for InsP3 than that of fretino-2, as expected. Permeabilization of MDCK cells yielded nearly the same increase in the emission ratio, as compared to microinjection of InsP3 in the cells (Figures 2a and 4h). We next examined the responses of fretino and fretino-2 for InsP3 generated by physiological stimulations. Fretino and fretino-2 were separately expressed in NIH-KDR cells that stably express receptors for vascular endothelial growth factor (VEGF).15 In the NIH-KDR cells, phospholipase Cγ becomes activated upon VEGF stimulation, and InsP3 is then generated.15 Figure 2c shows pseudocolor images of a cell expressed with fretino-2, illustrating a change in the CFP/YFP emission ratio of fretino-2. After stimulation of the cell with 50 ng/mL VEGF, a distinct red-shift in the pseudocolor was observed, showing that InsP3 was generated in the cell upon VEGF stimulation (Figure 2c). A time course of the CFP/YFP emission ratio in the cytoplasm shows an immediate InsP3 generation upon 50 ng/mL VEGF stimulation, and then the emission ratio reached a plateau in 1200 s (Figure 2d). At 5 ng/mL of VEGF, InsP3 was transiently generated, and a gradual removal of the generated InsP3 was observed (Figure 2d). At 0.5 ng/mL of VEGF, a slight but significant increase in InsP3 concentration was observed (Figure 2d). The FRET response of fretino-2 for the VEGF-dependent generation of InsP3 indicates that fretino-2 monitors the change in InsP3 concentration upon physiological stimulations. Fretino having the higher affinity for InsP3 than fretino-2 exhibited a higher emission ratio than fretino-2 in the absence of VEGF (Figure 2e). However, upon stimulation with VEGF, the CFP/YFP emission ratio of fretino reached nearly the same plateau level to that of fretino-2 (Figure 2e). We constructed and transfected cDNAs encoding fretino-3, in which the lysine 50814 was mutated to an alanine to disrupt the binding with InsP3 (Figure 1b). Fretino-3 did not show any significant response upon VEGF stimulation because it does not bind with InsP3 (Figure 2e). Importantly, fretino-3 exhibited nearly the same low emission ratio to fretino-2 in the absence of VEGF (Figure 2e). Furthermore, when the NIH-KDR cell expressed with fretino was treated with an inhibitor of phospholipases, 5 µM of U-73122, a gradual decrease in the CFP/YFP emission ration was observed (Figure

2f). The emission ratio of fretino then reached nearly the same ratio as that of fretino-3 (Figure 2f). These results indicate that fretino is already bound with a basal InsP3, even in the absence of VEGF, due to its high affinity. In contrast, fretino-2 having moderate affinity for InsP3 responds to the InsP3 generated upon the agonist stimulation but not to the basal InsP3 concentration in the absence of the agonist. Many amino acids in the InsP3binding region of InsP3R1 have been revealed to affect the affinity for InsP3 to each different extent when mutated to other amino acids.14 These mutant InsP3-binding regions can also be a good way of tuning the affinity for InsP3 of the present fluorescent indicators, as exemplified with fretino-2. To confirm that the response of the present indicators is actually reflecting a FRET change, we carried out photoinactivation of YFP within fretino-2. Excitation at 540 ( 12.5 nm of the cell expressed with fretino-2 resulted in photobleaching of the FRET acceptor, YFP, and an increase in the donor emission from CFP accompanied it due to breakdown of the energy transfer (Figure 2g). This photobleached cell did not respond to VEGF stimulation (Figure 2h). Fretino did not also respond to VEGF when photobleached (data not shown). This demonstrates that fretinos exhibit fluorescence responses based on FRET changes upon binding with InsP3. Imaging InsP3 in Single Dendrites of Neurons with Fretino. InsP3 is considered to be a key molecule in many cell types and also in distinct subcellular regions of single living cells; however, little is examined about the subcellular dynamics of InsP3 in a variety of cells. In particular, neurons have a great number of dendrites branched from a cell body, in which InsP3 plays a critical role in neuronal transmission.16-18 The dendrites of neurons have very thin structures so that it appears to be quite difficult to quantitatively analyze the InsP3 dynamics in single dendrites with the previously reported methods. In contrast, the present fretino-2 is based on the intramolecular FRET that is easily detectable, even in such thin dendrites; thus, we next show that fretino-2 is applicable for visualization of the InsP3 dynamics in single dendrites. cDNA encoding fretino-2 was introduced into hippocampal neurons prepared from the brains of rat embryos. Expression of fretino-2 was observed under a fluorescence microscope, showing that it was distributed throughout the branched dendrites and cell body of a neuron (Figure 3a and c). This allowed observation of InsP3 dynamics in each dendrite and the cell body. When the neuron expressed with fretino-2 was stimulated with 0.1 µM glutamate, the CFP/YFP emission ratio immediately increased in both the dendrites and cell body and then decreased to the basal emission ratio in 100 s (Figure 3b and d). This change in the CFP/YFP emission ratio represents a transient generation of InsP3 and its removal in the dendrites and cell body. The transient generation of InsP3 was augmented in a glutamatedependent fashion when the cell was stimulated with 1 µM glutamate (Figure 3b and d). Interestingly, the InsP3 dynamics provoked by 10 µM glutamate was not always monotonic but, rather, was often oscillating (Figure 3b and d). However, at 10 µM glutamate, as well as at 0.1 and 1 µM glutamate, the InsP3 dynamics was sometimes transient, and the increased InsP3 concentration was sometimes sustained for more than 500 s (data

(14) Yoshikawa, F.; et al. J. Biol. Chem. 1996, 271, 18277-18284. (15) Takahashi, T.; Shibuya, M. Oncogene 1997, 14, 2079-2089.

(16) Finch, E. A.; Augustine, G. J. Nature 1998, 396, 753-756. (17) Berridge, M. J. Neuron 1998, 21, 13-26. (18) Rose, C. R.; Konnerth, A. Neuron 2001, 31, 519-522.

Analytical Chemistry, Vol. 77, No. 15, August 1, 2005

4753

Figure 2. Fluorescence imaging with fretinos. (a) FRET responses of fretino-2 expressed in MDCK cells upon addition of various concentrations of InsP3 to the extracellular solution. Each concentration of InsP3 was added at the time shown with a broken line. The cells expressed with fretino-2 were permeabilized by treating with a detergent-containing solution to introduce InsP3 into the cells. The vertical axis represents the emission ratio of CFP (480 ( 15 nm) to YFP (535 ( 12.5 nm) when excited at 440 ( 10 nm at 25 °C. (b) InsP3 titration plots of fretino and fretino-2. The results are of the means with standard deviations of three independent experiments (mean ( S. D.). (c) Pseudocolor images of the CFP/YFP emission ratio before (time 0 s) and at 200, 400, 1200 s after the addition of 50 ng/mL VEGF. The results were obtained with NIH-KDR cells expressed with fretino-2. (d) Time courses of the CFP/YFP emission ratio of fretino-2 upon stimulation with 0.5 (2), 5 (O), and 50 ng/mL VEGF (b). (e) Time courses of the CFP/YFP emission ratio of fretino (O), fretino-2 (b), and fretino-3 (4) upon stimulation with 50 ng/mL VEGF. (f) Time courses of the CFP/YFP emission ratio of fretino-2 (b) and fretino-3 (O) upon stimulation with a PLC inhibitor, 5 µM U-73122. (g) Changes in the fluorescence intensity of CFP and YFP after the cell expressed with fretino-2 was excited at 540 ( 12.5 nm, resulting in photobleaching of the acceptor fluorophore, YFP. (h) Pseudocolor images of the CFP/YFP emission ratio of the cell expressed with fretino-2 before and after the photobleaching of YFP in fretino-2 and an image of the cell stimulated with 50 ng/mL VEGF after the photobleaching. 4754 Analytical Chemistry, Vol. 77, No. 15, August 1, 2005

Figure 3. Imaging InsP3 dynamics in neuronal dendrites with fretino-2. (a) Pseudocolor images of dendrites of a neuron expressed with fretino-2 before and at 150 s after stimulation with 10 µM glutamate. (b) Time courses of the CFP/YFP emission ratio in dendrites of neurons expressed with fretino-2 upon stimulation with 0.1, 1, and 10 µM glutamate. Glutamate was added at the time shown with a broken line. (c) Pseudocolor images of the cell body of a neuron expressed with fretino-2 before and at 150 s after stimulation with 10 µM glutamate. (d) Time courses of the CFP/YFP emission ratio in the cell body of neurons expressed with fretino-2 upon stimulation with 0.1, 1, and 10 µM glutamate. Glutamate was added at the time shown with a broken line.

Analytical Chemistry, Vol. 77, No. 15, August 1, 2005

4755

not shown). We observed no significant difference in the InsP3 dynamics between the dendrites and the cell body (Figure 3b and d). Fretino-2 allowed visualization of the complex InsP3 dynamics not only in the cell body but also in single, thin dendrites of living neurons. We expect that the present fretino-2 makes a substantial contribution to uncovering functional roles of InsP3 in neuronal transmission. Imaging Nuclear InsP3 Dynamics with a Genetically Targeted Fretino Variant. We next show that the present fretino-2 is genetically targetable to various subcellular regions, such as the nucleus and gap junctions. These subcellularly localized indicators allow us to visualize the respective local dynamics of InsP3 in living cells. Here, we illustrate an indicator to visualize InsP3 dynamics in the nucleus, named fretino-2nuc. This fretino-2nuc is based on fretino-2, but it is fused with a nuclear localization sequence to its C terminus for directing fretino-2nuc to the nucleus (Figure 1b). The nuclear InsP3 is considered necessary for the generation of other secondary messengers, Ca2+ released from InsP3Rs on the nucleoplasmic reticulum19 and inositol polyphosphates20-22 that control gene expression through chromatin remodeling and mRNA export from the nucleus. However, it remains controversial whether the nuclear membrane with nuclear pore complexes may act as a diffusion barrier for nuclear propagation of InsP3 generated in the cytoplasm.23,24 We expected that the present fretino-2nuc might solve this key question on the nuclear InsP3 dynamics that is crucial for the cellular functions derived from the nucleus. Fretino-2nuc was expressed in the MDCK cell. As shown in Figure 4a, bright fluorescence was observed exclusively from the nucleus. The nuclear localization of fretino-2nuc was confirmed by staining the nucleus with SYTOX-Orange after the cell was fixed with paraformaldehyde (Figure 4a). In contrast, fretino-2 was expressed in the cytoplasm and excluded from the nucleus due to the lack of NLS (Figure 4b). Each strict localization of the two kinds of indicators guarantees that fretino-2nuc and fretino-2 faithfully report InsP3 dynamics in the nucleus and cytoplasm, respectively. MDCK cells that endogenously express G protein-coupled P2Y receptors25 were stimulated with adenosine triphosphate (ATP). ATP-bound P2Y receptors provoke the generation of InsP3 through the activation of PLC. When MDCK cells expressed with fretino-2nuc were sequentially stimulated with 1 and 10 µM ATP, ATP-dependent transient responses were observed in the nucleus (Figure 4c). Upon further stimulation with 100 µM ATP, the MDCK cells expressed with fretino-2nuc exhibited a larger response than that upon stimulation with 1 and 10 µM ATP (Figure 4c). In addition, this response to 100 µM ATP was not monotonic but, rather, often oscillating (Figure 4c). The ATP-dependent response of fretino2nuc that represents InsP3 dynamics in the nucleus was quite (19) Echevarria, W.; Leite, M. F.; Guerra, M. T.; Zipfel, W. R.; Nathanson, M. H. Nat. Cell Biol. 2003, 5, 440-446. (20) York, J. D.; Odom, A. R.; Murphy, R.; Ives, E. B.; Wente, S. R. Science 1999, 285, 96-100. (21) Odom, A. R.; Stahlberg, A.; Wente, S. R.; York, J. D. Science 2000, 287, 2026-2029. (22) Shen, X.; Xiao, H.; Ranallo, R.; Wu, W.-H.; Wu, C. Science 2003, 299, 112116. (23) Hennager, D. J.; Welsh, M. J.; DeLisle, S. J. Biol. Chem. 1995, 270, 49594962. (24) Bootman, M. D.; Thomas, D.; Tovey, S. C.; Berridge, M. J.; Lipp, P. Cell. Mol. Life Sci. 2000, 57, 371-378. (25) Insel, P. A.; et al. Clin. Exp. Pharmacol. Physiol. 2001, 28, 351-354.

4756

Analytical Chemistry, Vol. 77, No. 15, August 1, 2005

similar to the response of fretino-2 that represents InsP3 dynamics in the cytoplasm when stimulated with 1, 10, and 100 µM ATP (Figure 4c and d). Furthermore, whenever InsP3 was injected into the cytoplasm of the cell (cell A) expressed with fretino-2nuc, quick responses of fretino-2nuc were observed in the nucleus of the injected cell A, while the cell (cell B) adjacent to the injected cell A remained silent (Figure 4e and f). Injection of a vehicle solution did not elicit a significant response of fretino-2nuc (Figure 4f, inset). In addition to fretino-2nuc in the nucleus, fretino-2 exhibited a quick response to InsP3 injected in the cytoplasm of cell C but not in the cytoplasm of the adjacent cell D (Figure 4g and h). Taking these results together, we conclude that the nuclear membrane does not hinder the nuclear propagation of InsP3 that is generated in the cytoplasm. Thus, nuclear InsP3 dynamics occurs in an almost synchronized manner with agonistevoked InsP3 dynamics in the cytoplasm. The nuclear InsP3 delivered in this way appears to control a wide range of cell functions through the generation of nuclear secondary messengers, Ca2+ 19and inositol polyphosphates.20-22 While this manuscript was in preparation, Tanimura et al. published an independent report of a FRET-based InsP3 indicator, LIBRA,26 analogous to the present fretino, except that they used a whole N-terminal domain of InsP3R (InsP3R31-604) rather than the present truncated N-terminal domain of InsP3R (InsP3R1224-579). We constructed fretino-4, which mimics LIBRA and possesses the whole N-terminal domain of InsP3R (InsP3R11-604), and compared the FRET response of fretino-4 with that of fretino-2 (Figure 1b). Fretino-4 was expressed in NIH-KDR cells, and the cells were stimulated with 50 ng/mL of VEGF. The FRET response of fretino-4 for VEGF was much smaller than that of fretino-2 (Figure 5). We also performed microinjection of InsP3 into MDCK cells expressed with fretino-4; however, fretino-4 showed only a subtle response again (data not shown). Compared to fretino-4 and LIBRA, fretino-2 lacks N-terminal amino acids 1-223 and Cterminal amino acids 580-604 of the InsP3 binding domain. The N- and C-terminal parts of the InsP3 binding domain appears to have hindered the substantial conformational change in the InsP3 binding domain of fretino-4. The deletion of the N- and C-terminal parts of the InsP3 binding domain has escaped their unwanted effect on inducing larger FRET responses for InsP3, as exemplified with the present fretino-1, -2, and -2nuc. In addition, LIBRA has a membrane-targeting signal and localizes to the plasma membrane and Golgi area. Tanimura et al. did not show whether LIBRA works in the cytoplasm and nucleus. Except for fretino-4, the present fretinos report subcellular dynamics of InsP3 in single living cells with a large FRET response. In conclusion, we have developed fluorescent indicators, named fretinos, to locate InsP3 dynamics in single living cells on the basis of an intramolecular FRET approach. We have demonstrated that the fretino makes it possible to visualize InsP3 dynamics, even in single, thin dendrites of neurons, which has been unseen previously. We have further localized the fretino in the nucleus, and the nuclear InsP3 dynamics has been pinpointed. As a result, it has been found that the nuclear membrane allows nuclear propagation of InsP3 from the cytoplasm so that nuclear InsP3 dynamics synchronously occurs with cytosolic InsP3 dynamics (26) Tanimura, A.; Nezu, A.; Morita, T.; Turner, R. J.; Tojyo, Y. J. Biol. Chem. 2004, 279, 38095-38098.

Figure 4. Imaging nuclear InsP3 dynamics with fretino-2nuc. (a, b) Fluorescence images of MDCK cells expressed with fretino-2nuc (a) and fretino-2 (b), taken with an emission filter for YFP (540 ( 12.5 nm). (c, d) Time courses of the CFP/YFP emission ratio of fretino-2nuc (c) and fretino-2 (d) in MDCK cells upon sequential stimulation with 1, 10, and 100 µM ATP. These concentrations of ATP were achieved at the time shown with broken lines. (e) A merged image of a fluorescence image and a phase contrast image, showing microinjection of InsP3 in the cytoplasm of a cell (cell A) expressed with fretino-2nuc. Two fluorescent nuclei from cells A and B are observed to express fretino-2nuc in this image. (f) Time courses of the CFP/YFP emission ratio of fretino-2nuc in cell A and the neighbor cell B upon injection of InsP3 into cell A. InsP3 was injected at the time shown with broken lines. Injection of a vehicle solution had no significant effect on the CFP/YFP emission ratio of fretino-2nuc in MDCK cells (inset). (g) A merged image of a fluorescence image and a phase contrast image, showing microinjection of InsP3 in the cytoplasm of a cell (cell C) expressed with fretino-2. Two fluorescent cells, cells C and D, are observed to express fretino-2 in this image. (h) Time courses of the CFP/YFP emission ratio of fretino-2 in cell C and the adjacent cell D upon injection of InsP3 into cell C. InsP3 was injected at the time shown with broken lines. Injection of a vehicle solution had no significant effect on the CFP/YFP emission ratio of fretino-2 in MDCK cells (inset).

Analytical Chemistry, Vol. 77, No. 15, August 1, 2005

4757

Figure 5. Comparison of the FRET responses between fretino-2 (O) and fretino-4 (b) in NIH-KDR cells upon stimulation with 50 ng/ mL of VEGF. The FRET response of fretino-4 is also shown with a narrow scale (inset).

evoked by agonist stimulations. The present approach should provide a unique, powerful tool to reveal local InsP3 dynamics not only in the nucleus but also in postsynaptic densities of neurons,17,18 gap junctions,27 etc., where InsP3 is believed to play a critical role in cellular functions. In addition, fretinos should contribute to our understanding of when, where, and how InsP3 is generated and removed in a variety of cell types, such as muscle cells, neurons, eggs on fertilization,28 and also in tissues and individuals. METHODS Plasmid Construction. To construct cDNAs of fretinos, fragment cDNAs of CFP, hIP3R1224-579, hIP3R1224-579 with the R504Q mutation, hIP3R1224-579 with the K508A mutation, YFP, and YFP with NLS were generated by standard PCR, and each was subcloned into pBluescript SK(+). All cloning enzymes were from Takara Biomedical (Tokyo, Japan) and were used according to the manufacturer’s instructions. All PCR fragments were sequenced with an ABI310 genetic analyzer (Applied Biosystems, Foster City, CA). Each cDNA encoding fretino was subcloned at the HindIII and XhoI sites of a mammalian expression vector, pcDNA3.1(+) (Invitrogen Co., Carlsbad, CA). (27) Niessen, H.; Harz, H.; Bedner, P.; Kramer, K.; Willecke, K. J. Cell Sci. 2000, 113, 1365-1372. (28) Kurokawa, M.; Sato, K.; Fissore, R. A. Biol. Cell 2004, 96, 37-45.

4758

Analytical Chemistry, Vol. 77, No. 15, August 1, 2005

Cell Culture and Transfection. MDCK cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal calf serum at 37 °C in 5% CO2. NIH-KDR cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% calf serum and 200 µg/mL Geneticin. Primary hippocampal neurons were prepared from Wister rat embryos (embryonic day 17) and cultured in Neurobasal medium supplemented with 2% B27 and 0.5 mM glutamine at 37 °C in 5% CO2. Cells were plated onto glass-bottom dishes, transfected with LipofectAMINE 2000 reagent, and left for 24 h at 37 °C in 5% CO2. Fluorescence Imaging of Cells. Culture medium was replaced with a Hank’s balanced salt solution (HBSS) for fluorescence imaging experiments. As described previously,7,8 the cells were imaged at room temperature on a Carl Zeiss Axiovert 135 microscope with a cooled CCD camera, MicroMAX (Roper Scientific Inc, Tucson, AZ), controlled by MetaFluor (Universal Imaging, West Chester, PA). The fluorescence images were obtained through 480 ( 15-nm and 535 ( 12.5-nm filters with a 40× oil immersion objective (Carl Zeiss, Jena, Germany). MDCK cells expressed with fretino and fretino-2 were permeabilized by treating with 10 mM HEPES-KOH buffer (pH7.4) containing 20 mM NaCl; 120 mM KCl; 1 mM EGTA; 330 µM CaCl2; and a detergent, saponin (Wako Pure Chemical Industries, Osaka, Japan), to introduce InsP3 into the cells. We observed and compared cells that express to each other as close to equal amounts as possible of the present InsP3 indicators. In addition, to minimize an unavoidable increase in InsP3 buffering as a result of overexpression of the InsP3 indicators, cells that express high concentration of the indicators were not used for imaging. Nuclear Staining with SYTOX-Orange. MDCK cells separately expressed with fretino-2 and fretino-2nuc were fixed with 4% paraformaldehyde for 10 min. The cells were treated with 2.5 µM SYTOX-Orange (Molecular Probes, Inc., Eugene, OR) for 10 min to stain the nucleus and were washed with HBSS. Coverslips with the stained cells were mounted onto the slide and observed under a confocal laser scanning microscope (LSM 510, Carl Zeiss). ACKNOWLEDGMENT This work was supported by grants from Japan Science and Technology Agency (JST) and Japan Society for the Promotion of Science (JSPS). We thank Prof. N. Matsuki and Dr. M. Yamada for their guidance on preparing the dissociation culture of the hippocampal neurons.

Received for review December 29, 2004. Accepted May 8, 2005. AC040195J