Quantitative Determination of Protein Nuclear Transport Induced by

For monitoring the nuclear transport of proteins induced by a ligand binding, we ... Labor-effective manipulation of marine and beetle luciferases for...
0 downloads 0 Views 542KB Size
Download PDF
Anal. Chem. 2005, 77, 6928-6934

Quantitative Determination of Protein Nuclear Transport Induced by Phosphorylation or by Proteolysis Sung Bae Kim,† Ryohei Takao,† Takeaki Ozawa,‡ and Yoshio Umezawa*,†

Department of Chemistry, School of Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-Ku, Tokyo 113-0033, Japan, and Department of Molecular Structure, Institute for Molecular Science, Myodaiji, Okazaki, Aichi 444-8585, and PREST, Japan Science and Technology Agency, 4-1-8 Honcho Kawaguchi, Saitama, Japan

Nucleocytoplasmic transport of proteins in eukaryotic cells is a fundamental process for gene expression. The transport is regulated by posttranslational modifications of the proteins, such as ligand-binding, phosphorylation, and proteolysis. For monitoring the nuclear transport of proteins induced by a ligand binding, we have recently developed a genetically encoded bioluminescent indicator based on reconstitution of split fragments of Renilla reniformis (RLuc) by protein splicing with DnaE inteins. We herein describe that the technique is used for detecting phosphorylation- or proteolysis-induced nuclear transports of a target protein. Two model proteins, signal transducer and activator of transcription 3 (STAT3) and sterol-regulatory element binding protein-2 (SREBP-2), were exemplified as phosphorylation- and proteolysisinduced nuclear transport, respectively. Each STAT3 or SREBP-2 is connected with C-terminal halves of RLuc and DnaE. If the protein translocates into the nucleus, the C-terminal fragment of RLuc meets the N-terminal fragment of RLuc, and full-length RLuc is reconstituted by protein splicing in the nucleus. The indicator with SREBP-2 enabled us to quantify the intracellular concentrations of cholesterol. The indicator with STAT3 quantified the extent of the nuclear transport induced by representative cytokines. This simple assay based on protein nuclear transports allows the selection of suitable drugs among candidates and has significant potential for risk assessments, such as carcinogenic chemical screening in vitro and in vivo. Eukaryotic cells are characterized by distinct nuclear and cytoplasmic compartments. In response to extra- or intracellular stimuli, certain proteins enter the nucleus to control magnitude and specificity of gene expression. The gene expression is regulated by posttranslational protein modifications, such as (1) ligand-binding, (2) phosphorylation, and (3) proteolysis.1-4 Defects in the protein modifications or nuclear transports are known to * To whom correspondence should be addressed. Phone: +81-3-5841-4351. Fax: +81-3-5841-8349. E-mail: [email protected]. † The University of Tokyo. ‡ Institute for Molecular Science and PREST. (1) Gorlich, D.; Kutay, U. Annu. Rev. Cell. Dev. Biol. 1999, 15, 607-660.

6928 Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

cause malfunctions and cancers in different types of cells.2-6 Simple detection methods for the nuclear transports are, therefore, of importance for drug screening or risk assessment in pharmacological and toxicological research. A technique to detect the protein movements inside the cells is to use the immunocytochemistry or automated fluorescence microscopy with a green fluorescent protein (GFP) and its variants. Although these approaches allow analyses of protein dynamics within cells, image acquisition can be slow and tedious. Immunocytochemical analysis requires a complex sample-preparation procedure, which prevents high-throughput drug screening or risk assessment. The obtained results are qualitative rather than quantitative because the number of analyzed cells is limited. In the case of automated fluorescence microscopy, the algorithm to determine the subcellular localization of a GFP-tagged protein is imprecise. To address such limitations, we have recently developed a new indicator based on reconstitution of split fragments of Renilla reniformis (RLuc) by protein-splicing with a DnaE intein (a catalytic subunit of DNA polymerase III).7 RLuc provides distinct merits, such as background-fluorescence-free and highly sensitive bioluminescence detection, because it does not require an excitation light, which is needed, for example, for GFP. Protein splicing is an autocatalytic process in which an intervening sequence, termed intein, excises itself out of a precursor protein with concomitant ligation of the flanking sequences, termed exteins. The N-terminal fragment of split RLuc connected with the N-terminal part of DnaE are localized in the nucleus due to a cofused nuclear localization signal (NLS) (Figure 1). The Cterminal fragment of split RLuc is fused with the rest of DnaE and a target cytosolic protein, which is localized in the cytosol. Translocation of the target protein into the nucleus results in protein splicing with DnaE, thereby producing a full-length RLuc and recovering its bioluminescence activity. In this primary experiment, a typical ligand-binding protein, androgen receptor, (2) Lee, S. J.; Sekimoto, T.; Yamashita, E.; Nagoshi, E.; Nakagawa, A.; Imamoto, N.; Yoshimura, M.; Sakai, H.; Chong, K. T.; Tsukihara, T.; Yoneda, Y. Science 2003, 302, 1571-1575. (3) Kau, T. R.; Way, J. C.; Silver, P. A. Nat. Rev. Cancer 2004, 4, 106-117. (4) Gasiorowski, J. Z.; Dean, D. A. Adv. Drug Delivery Rev. 2003, 55, 703716. (5) Weigel, N. L.; Zhang, Y. X. J. Mol. Med. 1998, 76, 469-479. (6) Mori, K. Traffic 2003, 4, 519-528. (7) Kim, S. B.; Ozawa, T.; Watanabe, S.; Umezawa, Y. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 11542-11547. 10.1021/ac050966z CCC: $30.25

© 2005 American Chemical Society Published on Web 10/06/2005

Figure 1. (A) Schematic illustration of protein transport into the nucleus. Signaling pathways 1, 2, and 3 show the nuclear transport of a target protein induced by ligand-receptor binding, proteolysis, and phosphorylation, respectively. The C-terminal RLuc linked with the target protein is initially localized in the cytosol. The N-terminal RLuc is localized in the nucleus due to a cofused nuclear localization signal (NLS). Upon stimulation of a ligand, the C-terminal RLuc linked with the target protein is translocated into the nucleus through either of the above signaling pathways 1, 2, and 3, resulting in protein splicing with the N-terminal RLuc in the nucleus. The full-length RLuc is thereby formed and recovers its bioluminescence activity. (B) Schematic structures of cDNA constructs. cDNA sequences encoding CFNLSH, CFNLSH, and KFAEY are inserted between rLuc and dnaE in pcDRc-SREBP-2, pcDRc-STAT3, and pcRDn-NLS, respectively, for efficient protein splicing.

was used to demonstrate usefulness and applicability of the splitRLuc indicator.7 In the present work, we have explored the potential of the splitRLuc indicator for detecting phosphorylation- and proteolysisinduced nuclear transport of a target protein. Of many phosphorylated proteins, signal transducer and activator of transcription 3 (STAT3) was selected, because STAT3 is known to contribute to human diseases, making the study of STAT3 an important topic of current research. We also demonstrated high-throughput sensing of proteolysis-induced nuclear import by using a representative protein, sterol-regulatory element binding protein-2 (SREBP-2), which senses concentrations of intracellular cholesterol. This technology is expected to increase the diversity

available for the various kinds of proteins that target to the nucleus. EXPERIMENTAL SECTION Construction of Plasmids. The plasmid pcRDn-NLS encoding the N-terminal domains of RLuc (RLuc-N; 1 to 229 aa) and DnaE (DnaE-N; 1 to 123 aa) was described in our previous report.7 At the splicing junction between dnaE and rLuc, cDNA sequences encoding an additional five amino acids were inserted for efficient splicing (Figure 1B).8 A cDNA sequence encoding a GC linker, GGGGSG, was inserted next to rLuc-c for providing the expressed (8) Evans, T. C.; Martin, D.; Kolly, R.; Panne, D.; Sun, L.; Ghosh, I.; Chen, L. X.; Benner, J.; Liu, X. Q.; Xu, M. Q. J. Biol. Chem. 2000, 275, 9091.

Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

6929

probes with its flexibility. The plasmids, pcDRc-SREBP2 and pcDRc-STAT3, were prepared from pcDRc-AR by replacing the cDNA of AR with SREBP-2 and STAT3 as follows: The cDNA encoding each full length of SREBP-2 or STAT3 was modified by PCR to add unique enzyme sites, NotI and XhoI, at the N- and C-terminal ends, respectively. Each cDNA fragment was subcloned into the expression vector pcDRc-AR at NotI/XhoI sites, named pcDRc-SREBP2 and pcDRc-STAT3. The PCR products were sequenced to ensure fidelity with a BigDye Terminator Cycle Sequencing kit and a genetic analyzer ABI Prism310 (PE Biosystems). Cell Culture and Transfection. COS-7 or HEK 293 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Sigma) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (P/S) at 37°C in 5% CO2. The cells were transfected with the constructed plasmids using a lipofection reagent, LipofectAMINE 2000 (Invitrogen). Western Blot. For the study on SREBP-2, COS-7 cells were transiently cotransfected with the plasmids, pcRDn-NLS and pcDRc-SREBP2, and incubated for 4 h. The medium was then replaced with DMEM supplemented with a 5% cholesterol-free newborn calf serum (NCS), 1% P/S, 50 µM compactin, 50 µM mevalonate, and 0.2% ethanol. Soon after the medium change, the cells were supplemented with a mixture of 25 µM cholesterol and 2.5 µM 25-hydroxycholesterol for 16 h. A calpain inhibitor, N-acetyl-leucine-leucine-norleucinal (ALLN), was then added to the medium for 4 h, and then the cells were lysed in 200 µL of a lysis buffer (1% SDS/10% glycerol/10% 2-mercaptoethanol/0.001% bromophenol blue/50 mM Tris-HCl, pH 6.8). Each aliquot of the samples was electrophoresed in 8% acrylamide gel and transferred to a nitrocellulose membrane and blotted with mouse anti-FLAG epitope antibody (Sigma) and mouse anti-β-actin antibody (Sigma). The blots were incubated with alkaline phosphatase-conjugated secondary antibodies (Jackson) and visualized by a chemiluminescence reagent (New England Biolabs) and a luminescence image analyzer (LAS-1000, Fuji Film). For the study on STAT3, HEK 293 cells were transiently cotransfected with the plasmids, pcRDn-NLS and pcDRc-STAT3. The cells were incubated for 24 h and stimulated with 100 ng/ mL human oncostatin M (hOSM, Sigma) for 4 h. Western blot was performed with anti-c-Myc antibody (Roche), anti-β-actin antibody (Sigma), and anti-pY-STAT3 antibody (Santa Cruz) under otherwise identical conditions. Immunocytochemistry. COS-7 and HEK 293 cells were cultured on thin cover glasses to 90% confluence and transfected with the respective plasmids. The medium was replaced with a cholesterol-free DMEM for making a cholesterol starvation condition or with a DMEM containing 25 µM cholesterol and 2.5 µM 25-hydroxycholesterol for a reference. On the other hand, the HEK 293 cells were stimulated with or without 100 ng/mL hOSM. The cells were washed with PBS, and fixed with a 3% paraformaldehyde (PFA) solution on the cover glasses. The cells on cover glasses were blocked with 0.2% fish skin gelatin (FSG) and then incubated with mouse anti-SREBP-2 antibody (Santa Cruz) or mouse anti-FLAG epitope antibody (Sigma) for 1 h. The cells were then incubated with Cy-5-conjugated mouse secondary antibody (Jackson) for 30 min, and the nuclei were stained with a specific marker, Sytox Green (Molecular Probes). The fluorescence 6930

Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

images were recorded using a confocal laser-scanning microscope (LSM510, Zeiss) fitted with a band-pass filter (514-535 nm) for Sytox Green and a long-pass filter (665 nm) for Cy-5. Cell Fractionation Study. HEK 293 cells transiently cotransfected with pcRDn-NLS and pcDRc-STAT3 were stimulated with 100 ng/mL hOSM for 4 h. The cells were harvested and then suspended in a buffer solution (0.25 M sucrose, 5 mM EDTA, 20 mM Tris-HCl, pH 7.4). The plasma membranes of cells were crushed with a tip sonicator (U200Scontrol, IKA labortechnik), and the homogenate was centrifuged at 600g for 15 min to separate the nuclei from the other cell components. The luminescence intensities of each fraction were determined with a luminometer (Minilumat LB9506, Berthold), and the results displayed on the digital readout were shown in relative light units (RLU) per 1.0 µg of protein. Cell-Based in Vitro Assay. COS-7 cells cultured in 12-well microplates were transiently cotransfected with pcRDn-NLS and pcDRc-SREBP2 and incubated for 4 h. The medium was replaced with DMEM supplemented with a 5% steroid-free newborn calf serum (NCS), 1% P/S, 50 µM compactin, 50 µM mevalonate, 0.2% ethanol, and differing concentrations of cholesterol or 25-hydroxycholesterol. The cells were incubated for 16 h. Four hours after the injection of ALLN, the cells were harvested, and their luciferase activities were evaluated with a luminometer. For the study on STAT3, HEK 293 cells raised in 12-well microplates were transiently cotransfected with pcRDn-NLS and pcDRc-STAT3 and incubated for 24 h. The cells on each well were stimulated with differing concentrations of human oncostatin M (hOSM), human interleukin-6 (hIL-6), human leukemia inhibitory factor (hLIF), and 1:2 mixtures of hIL-6 and human interleukin-6 soluble receptor (hIL-6sR), respectively, and extensively incubated for 4 h. The cells were harvested, and their luciferase activities were measured with a luminometer. RESULTS An Indicator for Proteolysis-Induced Nuclear Transport. Intracellular Localization of the Indicators. SREBP-2 is synthesized as an inactive precursor form bound to the endoplasmic reticulum (ER) membrane. Upon decrease in the intracellular cholesterol, the precursor undergoes a cleavage by proteases, and thereby, its matured N-terminal domain transports into the nucleus. We examined by the immunocytochemical method whether the SREBP-2 connected with C-terminal halves of DnaE and RLuc was correctly localized in the ER membrane (Figure 2A and B). COS-7 cells were transiently transfected with pcDRc-SREBP2 or pcRDnNLS carrying respective constructs of fusion proteins, and localization of each expressed fusion protein was imaged with its specific primary and Cy-5-conjugated secondary antibodies. The results showed that with 25 µM cholesterol, SREBP-2 connected with C-terminal halves of DnaE and RLuc was localized in the cytosol (Figure 2A-a). The localization of SREBP-2 overlapped exactly the localization of the ER, which was stained with ER-specific fluorescent marker, BODIPY-Brefeldin A (Figure 2B). Without 25 µM cholesterol, the nuclei were stained faintly with a Cy-5labeled antibody, indicating that a large amount of SREBP-2 on the ER membrane was translocated into the nucleus by proteolysis. In contrast, the N-terminal fusion of RLuc and DnaE connected with a nuclear localization signal (NLS; (DPKKKRKV)3) was predominantly localized in the nucleus, regardless whether 25 µM

Figure 2. Characterization of the indicators. (A) Immunocytochemical analyses of the localization of the indicator proteins. The SREBP-2- and STAT3-fused C-terminal proteins and the counterpart N-terminal protein were expressed from pcDRc-SREBP2, pcDRc-STAT3, and pcRDnNLS, respectively. The respective proteins were recognized by anti-SREBP-2 (a), anti-c-Myc antibody (c), and anti-Flag antibodies (b and d), respectively, and stained with Cy-5-labeled secondary antibody (left column of each block). The nuclei were stained with Sytox Green (middle of each block), and their merged images were shown with the transmission (right column of each block). +Chol: 25 µM cholesterol added. -Chol: the vehicle (DMSO) added. +OSM: 100 ng/mL oncostatin M added. -OSM: the vehicle (PBS) added. (B) Immunocytochemical analyses of the localization of the SREBP-2 fusion protein. Endoplasmic reticulum (ER) was stained with its specific fluorescent probe, BODIPY-Brefeldin A (Green). SREBP-2 fusion protein was recognized with anti-SREBP-2 antibody and stained with Cy-5-conjugated secondary antibody. The two fluorescence images were superimposed on the transmission images. (C) Western blotting analysis. (a) Western blot of protein extracts from COS-7 cells (lane 1) and from the cells cotransfected with pcRDn-NLS and pcDRc-SREBP-2 in the presence (lane 2) and absence (lane 3) of a mixture of 25 µM cholesterol and 2.5 µM 25-hydroxycholesterol. As a reference for the amounts of proteins electrophoresed, β-actin was stained with its specific antibody. (b) Western blot of protein extracts from HEK 293 cells (lane 1) and from the cells carrying pcRDn-NLS and pcDRc-STAT3 after being incubated in the absence (lane 2) and presence (lane 3) of 100 ng/mL oncostatin M. Phosphotyrosine in the STAT3 was blotted with anti-pY-STAT3 antibody. The STAT3-linked fusion protein was stained with anti-c-Myc antibody. (c) Western blot of protein extracts from HEK 293 cells (lane 1) and from the cells transfected with pcRDn-NLS and pcDRc-STAT3 in the absence (lane 2) and presence (lane 3) of 100 ng/mL oncostatin M. As a reference for the amounts of the electrophoresed proteins, β-actin was stained with its specific antibody.

cholesterol was added (Figure 2A-b). These results demonstrate that N- and C-terminal halves of RLuc were colocalized in the nucleus only when the cholesterol was depleted in the cytoplasm. To ensure that protein splicing occurs upon N- and C-terminal halves of RLuc colocalized in the nucleus, the spliced products were analyzed by Western blot. The COS-7 cells cotransfected

with pcDRc-SREBP2 and pcRDn-NLS were stimulated with a mixture of 25 µM cholesterol and 2.5 µM 25-hydroxycholesterol (25-HC), and their crude extracts were electrophoresed. In the presence of the cholesterol mixture, anti-FLAG epitope antibody recognized a specific band at 45 kDa, the size of which was the same as the unspliced precursor protein, SREBP-2 connected with Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

6931

Figure 3. Cholesterol concentration dependence of the luminescence intensity from the cells carrying pcDRc-SREBP2 and pcRDn-NLS. The luminescence intensities were induced by varying concentrations of cholesterol (Chol), 25-hydroxycholesterol (25-HC), or both. The mean luminescence intensities (n ) 3) were determined at each cholesterol concentration.

C-terminal halves of DnaE and RLuc. In the privation of cholesterol, anti-FLAG epitope antibody recognized a specific band at 96 kDa in addition to an unspliced precursor protein at 45 kDa. Electrophoretic mobility of the 96-kDa polypeptide was consistent with the predicted size of the spliced product, SREBP-2 connected with a full-length RLuc (Figure 2C-a). The results demonstrate that the proteolysis-induced translocation of SREBP-2 into the nucleus triggered protein splicing, and thereby, a full-length RLuc was reconstituted in the nucleus. Cell-Based Assay of the Nuclear Transport of SREBP-2. To develop a cell-based assay for detecting the proteolysis-activated nuclear transport, COS-7 cells were transiently cotransfected with the plasmids of pcDRc-SREBP2 and pcRDn-NLS. Before harvest, the cells on the culture plate were stimulated with differing concentrations of cholesterol. The luminescence intensities are expressed as RLU ratio, RLU1/RLU0, where RLU1 is the luminescence intensity of the cell lysate with each cholesterol stimulation, and RLU0 is that of the lysate stimulated with a reference cholesterol concentration (10-7 M cholesterol or 10-7 M 25hydroxycholesterol), respectively. When cholesterol concentrations were below 10-11 M, the luminescence intensities showed a remarkable increase (Figure 3). Over 10-11 M concentrations of cholesterol, the luminescence intensity was completely suppressed. The dramatic change in luminescence from 10-12 to 10-11 M cholesterol concentrations corresponds to the change in level of the 96-kDa protein shown in Figure 2C-a. The suppressed extent of the luminescence intensities upon changing from cholesterol-privation to cholesterolrich conditions was highly influenced by the kinds of added cholesterols. The suppressive effects of cholesterol analogues on the luminescence intensities decreased in the following order: a mixture of 25 µM cholesterol and 2.5 µM 25-hydroxycholesterol > 25 µM cholesterol > 25 µM 25-hydroxycholesterol. These results conclude that high concentrations of cholesterol inhibit nuclear transport of SREBP-2, and the mixture of cholesterol and 25-hydroxycholesterol is most effective in inhibiting the nuclear transport of SREBP-2. An Indicator for Phosphorylation-Induced Protein Nuclear Transport. Intracellular Localization of the Indicators. It is known 6932 Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

that phosphorylation of a 705-tyrosine residue in the STAT3 leads to translocation of STAT3 from cytosol into the nucleus, in which the STAT3 binds to the promoters of genes and activates transcription. We examined the localization of STAT3 with antic-Myc antibody, which recognizes a fusion protein of STAT3 with C-terminal halves of RLuc and DnaE. Without hOSM, the fusion protein was predominantly in the cytoplasm, whereas addition of 100 ng/mL hOSM resulted in the nuclear localization of the fusion protein (Figure 2A-c and -d). To confirm that the translocation of STAT3 connected with the C-terminal halves of RLuc and DnaE was, indeed, triggered by phosphorylation of the 705-tyrosine residue, Western blot analysis with anti-phosphotyrosine antibody against STAT3 (anti-pY-STAT3 antibody) was performed. Lysates of HEK 293 cells cotransfected with pcDRc-STAT3 and pcRDnNLS were prepared with or without 100 ng/mL hOSM for 15 min. Western blot analysis revealed that anti-pY-STAT3 antibody recognized a specific peptide band only in the lane with a hOSM stimulation, whereas anti-c-Myc antibody recognized specific bands at 103 kD with or without 100 ng/mL hOSM (Figure 2C-b). The results demonstrate that the fusion protein including STAT3 was phosphorylated by 100 ng/mL hOSM and thereby translocated into the nucleus. For ensuring that the protein splicing occurs in the nucleus after nuclear transport of the phosphorylated STAT3, the HEK 293 cells were transiently cotransfected with pcDRc-STAT3 and pcRDn-NLS. Crude extracts were obtained from the cells after the stimulation with 100 ng/mL hOSM. The anti-c-Myc antibody recognized a specific band of an unspliced protein at 103 kDa whether 100 ng/mL hOSM is present (Figure 2C-c). In addition, the stimulation of hOSM resulted in a specific new polypeptide at 130 kDa. The electrophoretic mobility corresponded to the predicted size of the spliced product, that is, STAT3 connected with full-length RLuc. Possible reasons for the low level of STAT3linked full-length RLuc at 130 kD may be that the expression of the C-terminal fragment of STAT3-linked split RLuc is low, or its spliced product was less stable, as compared to the SREBP-2linked full-length RLuc. Next, to investigate whether RLuc activity recovers in the nucleus, we prepared nucleus and cytosol fractions and measured

Figure 4. Analysis of luminescence intensities of nuclear and cytoplasmic fraction. The cellular fractions were obtained from the cells cotransfected with pcDRc-STAT3 and pcRDn-NLS with (+) or without (-) 100 ng/mL oncostatin M.

their luminescence intensities (Figure 4). The luminescence intensity from the nuclei stimulated with 100 ng/mL hOSM was found to be strong, as compared with those from the cytosolic fractions and the nuclei without stimulation with 100 ng/mL hOSM, demonstrating that RLuc refolded correctly and recovered its luminescence activity in the nucleus only when the cells were stimulated with 100 ng/mL hOSM. Cell-Based Assay of the Nuclear Transport of STAT3. We examined changes in the luminescence intensities for various concentrations of hOSM by using the HEK 293 cells carrying pcDRc-STAT3 and pcRDn-NLS (Figure 5). The luminescence intensities are expressed as an RLU ratio, RLU1/RLU0, where RLU1 is the luminescence intensity of the cell lysate with a cytokine stimulation, and RLU0 is that of a control lysate. Murine and human OSMs have 65% similarity in their amino acids. Their functions are also quite different in embryonic stem cell lines and murine hematopoietic cell lines.9,10 Here, we examined differences in the activity between the murine and human OSMs for the nuclear transport of STAT3 with HEK 293 cells. When the cells were stimulated with hOSM, the luminescence intensities increased efficiently in hOSM concentrations from 1 ng/mL to 100 ng/mL. In contrast, mOSM did not induce any increase in luminescence intensities in the tested concentration ranges. Next, we characterized this phosphorylation-triggered nuclear transport of STAT3 using three different kinds of cytokines, human leukemia inhibitory factor (hLIF), human interleukin 6 (hIL-6), and hOSM, all of which belong to the interleukin-6 (IL-6) subfamily of cytokines (Figure 6). We measured the activities of three different types of cytokines with HEK 293 cells carrying pcDRcSTAT3 and pcRDn-NLS. Addition of hLIF showed no increase in (9) Ichihara, M.; Hara, T.; Kim, H.; Murate, T.; Miyajima, A. Blood 1997, 90, 165-173. (10) Yoshimura, A.; Ichihara, M.; Kinjyo, I.; Moriyama, M.; Copeland, N. G.; Gilbert, D. J.; Jenkins, N. A.; Hara, T.; Miyajima, A. EMBO J. 1996, 15, 1055-1063.

Figure 5. hOSM concentration dependence of the luminescence intensity from the cells carrying pcDRc-STAT3 and pcRDn-NLS. The luminescence intensities were induced by varying concentrations of human OSM (hOSM) or by murine OSM (mOSM). The mean luminescence intensities (n ) 3) were determined at each OSM concentration.

Figure 6. Cytokine concentration dependence of the luminescence intensity from the cells carrying pcDRc-STAT3 and pcRDn-NLS. HEK 293 cells cotransfected with pcDRc-STAT and pcRDn-NLS were stimulated with each concentration of human leukemia inhibitory factor (hLIF), human interleukin-6 (hIL-6), and human oncostatin M (hOSM). The mean luminescence intensities (n ) 3) were determined at each cytokine concentration.

the luminescence intensities in the tested concentration ranges, whereas hOSM rapidly increased the luminescence intensities in the range from 1 to 100 ng/mL. hIL-6 did not induce any detectable increase in the luminescence intensity. The 1:2 mixture of hIL-6 and its soluble receptor (hIL-6sR) was found to induce small luminescence intensities in the range from 10 to 100 ng/ mL, which was twice that of the background. The results indicate that hIL-6 complexed with hIL-6sR induced phosphorylation of STAT3 through a cytokine signaling cascade and subsequent Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

6933

nuclear transport of STAT3. The extent of luminescence intensities induced by each 100 ng/mL cytokine decreased in the following order: hOSM > 1:2 mixture of hIL-6 and hIL-6sR > hIL-6 > hLIF. The results demonstrate that hOSM is the most potent cytokine for a STAT3 phosphorylation. DISCUSSIONS We constructed a pair of genetically encoded indicators for imaging proteolysis-triggered nuclear transport of SREBP-2. When cells are deprived of cholesterol, SREBP-2 is escorted to the Golgi apparatus, where it is proteolyzed. The N-terminal fragment enters the nucleus and activates the transcription of genes.2,11,12 Thus, the nuclear transport of N-terminal fragment of SREBP-2 is a measure of the cholesterol privation in cell lines. With this technology, we evaluated the inhibitory effects of cholesterols on the proteolysis-activated nuclear trafficking. In cholesterol-rich conditions, the luminescence intensities were suppressed because a proteolysis-induced nuclear transport of SREBP-2 was repressed, whereas in cholesterol-deprived conditions, the luminescence intensities sharply increased due to a quick proteolysis and nuclear transport of SREBP-2 (Figure 3). The dose-response curve by differing cholesterol concentrations reflects the content of the proteolyzed N-terminal SREBP-2 into the cellular nucleus. We found that the nuclear transport of SREBP-2 is initiated at around 10-11 M of cholesterol and reached to a plateau at around 10-13 M. The result suggests that their cholesterol-regulatory system works when their cellular cholesterol levels decrease below 10-11 M. For detecting the proteolysis-stimulated nuclear transport, SREBP-2 was connected with C-terminal halves of DnaE and RLuc. Generally, such a protein tagging often causes its mislocalization and an unexpected repressive influence on its activity. Therefore, we confirmed the localization of the SREBP-2 with immunocytochemistry (Figure 2B). As shown in Figure 2B, the localization data demonstrated that modification of both N- and C-terminal ends of SREBP-2 did not influence the SREBP-2 localization in the ER. The result implies that SREBP-2 may have its intrinsic ER localization signal in the middle of its amino acid sequence, not in the N- or C-terminal ends. (11) Miserez, A. R.; Muller, P. Y.; Barella, L.; Barella, S.; Staehelin, H. B.; Leitersdorf, E.; Kark, J. D.; Friedlander, Y. Atherosclerosis 2002, 164, 1526. (12) Shimano, H. Prog. Lipid Res. 2001, 40, 439-452. (13) Yang, J.; Chatterjee-Kishore, M.; Staugaitis, S. M.; Nguyen, H.; Schlessinger, K.; Levy, D. E.; Stark, G. R. Cancer Res. 2005, 65, 939-947. (14) Bromberg, J. F.; Wrzeszczynska, M. H.; Devgan, G.; Zhao, Y. X.; Pestell, R. G.; Albanese, C.; Darnell, J. E. Cell 1999, 98, 295-303. (15) Yu, H. E.; Jove, R. Clin. Cancer Res. 2003, 9, 6275S-6276S.

6934

Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

It has been reported that the sustained activation of STAT3 is frequently observed in many human tumors, including various carcinomas, lymphomas, leukemias, and melanomas.13-15 From such previous experimental results, we took a hint that nuclear translocation of STAT3 may be a novel target for anticancer drug screenings. Herein, we constructed a new genetically encoded indicator to quantitatively detect the phosphorylation-stimulated nuclear transport of STAT3. The amount of nuclear transport of STAT3 is a quantitative index of the phosphorylated STAT3 by cytokines. The selective phosphorylation of STAT3 by the stimulation of hOSM suggests that the HEK 293 cells express a common receptor for cytokines, gp130, which is known to be required for the phosphorylation of STAT3. In the cell fractionation study, we observed that the background intensity of luminescence was high even in the absence of a hOSM stimulation. This may be due to the stimulationindependent nuclear transport of STAT3. Recent reports indicated that unphosphorylated STAT3 is in part transported into the nucleus and drives gene expressions even without an extracellular stimulation.13 However, such stimulation-independent nuclear transport gives limited and minor effects in this technology, and therefore, it is possible to estimate the extent of STAT3 translocation if the background luminescence is correctly measured and subtracted. To conclude, we developed a nontranscriptional assay system using the new genetically encoded indicators, with which the magnitude of protein nuclear transport was quantified. The nuclear transport was evaluated with SREBP-2 and STAT3, induced by proteolysis and phosphorylation, respectively. With this technology, the translocational activity of the target proteins triggered by cytokines and starvation of cholesterols was quantitatively estimated. This simple and nontranscriptional assay allows selection of suitable drugs among candidates and also has significant potential for application for risk chemical screening in vitro and in vivo. This method is generally useful for imaging the cellular signal transduction of a target protein on the basis of the trafficking of the protein by stimulations of bioactive small molecules. ACKNOWLEDGMENT This work was supported by grants from Japan Science and Technology Agency (JST) and Japan Society for the Promotion of Science (JSPS).

Received for review June 2, 2005. Accepted August 31, 2005. AC050966Z