FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2131356098> ?p ?o ?g. }

• W2131356098 endingPage "34671" @default.
• W2131356098 startingPage "34665" @default.
• W2131356098 abstract "Expression of heat shock genes is regulated mainly at the level of transcription by heat shock transcription factors (HSFs). In avian, HSF1 and HSF3 are maintained in a cryptic monomer and dimer form, respectively, in the cytoplasm in the absence of stress. Upon heat stress, they undergo conformational change associated with the formation of a trimer and nuclear translocation. In this study, we identified regions that are necessary for the dimer-to-trimer transition of chicken HSF3 (cHSF3) upon stress conditions. One of these regions coincides with a functional nuclear localization signal (NLS) of cHSF3 that was recognized by a nuclear transport receptor importin α. Point mutations of basic amino acids in the NLS inhibit both nuclear translocation and trimer formation of cHSF3. These results demonstrate that the NLS acts positively on the trimer formation of cHSF3 upon stress conditions. Expression of heat shock genes is regulated mainly at the level of transcription by heat shock transcription factors (HSFs). In avian, HSF1 and HSF3 are maintained in a cryptic monomer and dimer form, respectively, in the cytoplasm in the absence of stress. Upon heat stress, they undergo conformational change associated with the formation of a trimer and nuclear translocation. In this study, we identified regions that are necessary for the dimer-to-trimer transition of chicken HSF3 (cHSF3) upon stress conditions. One of these regions coincides with a functional nuclear localization signal (NLS) of cHSF3 that was recognized by a nuclear transport receptor importin α. Point mutations of basic amino acids in the NLS inhibit both nuclear translocation and trimer formation of cHSF3. These results demonstrate that the NLS acts positively on the trimer formation of cHSF3 upon stress conditions. heat shock transcription factor chicken HSF nuclear localization signal hydrophobic heptad repeat green fluorescence protein glutathione S-transferase phosphate-buffered saline polyacrylamide gel electrophoresis Cells respond to various kinds of physiological signals and environmental stresses through the expression of programmed sets of genes. These inducible systems of gene expression are regulated by transcription factors that bind to specific DNA sequences. Many transcription factors that govern inducible gene expression remain in the cytoplasm in a cryptic form, and in response to signaling they are transported to the nucleus where they bind to and activate specific sets of genes. One of the examples of inducible gene expression is heat shock gene expression induced by heat stress (1Wu C. Annu. Rev. Cell Biol. 1995; 11: 441-469Crossref Scopus (967) Google Scholar). The expression of eukaryotic heat shock genes is controlled by heat shock transcription factors (HSFs)1 that bind to heat shock element located in the promoter regions of all heat shock genes. In mammals, HSF1 is activated by heat stress and induces expression of heat shock genes (2Baler R. Dahl G. Voellmy R. Mol. Cell. Biol. 1993; 13: 2486-2496Crossref PubMed Scopus (392) Google Scholar, 3Sarge K.D. Murphy S.P. Morimoto R.I. Mol. Cell. Biol. 1993; 13: 1392-1407Crossref PubMed Scopus (744) Google Scholar, 4McMillan D.R. Xiao X. Shao L. Graves K. Benjamin I.J. J. Biol. Chem. 1998; 273: 7523-7528Abstract Full Text Full Text PDF PubMed Scopus (431) Google Scholar). HSF1 not only protects cells subjected to stresses by inducing a set of heat shock genes but also is involved in normal development (5Jedlicka P. Mortin M.A. Wu C. EMBO J. 1997; 16: 2452-2462Crossref PubMed Scopus (233) Google Scholar, 6Xiao X. Zuo X. Davis A.A. McMillan D.R. Curry B.B. Richardson J.A. Benjamin I.J. EMBO J. 1999; 18: 5943-5952Crossref PubMed Scopus (446) Google Scholar). Furthermore, HSF1 induces apoptosis of spermatocytes subjected to thermal stress (7Nakai A. Suzuki M. Tanabe M. EMBO J. 2000; 19: 1545-1554Crossref PubMed Scopus (151) Google Scholar). In the absence of stress, HSF1 is mostly located in the cytoplasm in an inactive monomeric form. HSF1 is activated in roughly two steps (1Wu C. Annu. Rev. Cell Biol. 1995; 11: 441-469Crossref Scopus (967) Google Scholar, 8Larson J.S. Schuetz T.J. Kingston R.E. Nature. 1988; 335: 372-375Crossref PubMed Scopus (188) Google Scholar,9Morimoto R.I. Genes Dev. 1998; 12: 3788-3796Crossref PubMed Scopus (1523) Google Scholar). First, HSF1 is translocated to the nucleus, a process that is associated with a conversion to a trimer form that can bind to heat shock element with high affinity. Second, it acquires transcriptional activity probably by exposing a carboxyl-terminal activation domain (10Green M. Schuetz T. Sullivan K. Kingston R.E. Mol. Cell. Biol. 1995; 15: 3354-3362Crossref PubMed Scopus (136) Google Scholar, 11Shi Y. Kroeger P. Morimoto R.I. Mol. Cell. Biol. 1995; 15: 4309-4318Crossref PubMed Scopus (101) Google Scholar, 12Zuo J. Rungger D. Voellmy R. Mol. Cell. Biol. 1995; 15: 4319-4330Crossref PubMed Scopus (203) Google Scholar). All members of the HSF gene family share common structural features including a conserved DNA-binding domain that exhibits a winged helix-turn-helix motif (13Harrison C.J. Bohm A.A. Nelson H.C.M. Science. 1994; 264: 224-227Crossref PubMed Scopus (219) Google Scholar, 14Littlefield O. Nelson H.C. Nat. Struct. Biol. 1999; 6: 464-470Crossref PubMed Scopus (135) Google Scholar), an extended heptad repeat of hydrophobic amino acids (HR-A/B) involved in trimerization (15Sorger P.K. Nelson H.C.M. Cell. 1989; 59: 807-813Abstract Full Text PDF PubMed Scopus (286) Google Scholar, 16Peteranderl R. Nelson H.M.C. Biochemistry. 1992; 31: 12272-12276Crossref PubMed Scopus (121) Google Scholar). With the exception of HSFs in budding yeasts and mammalian HSF4, HSFs also have another carboxyl-terminal hydrophobic repeat (HR-C) (17Clos J. Westwood J.T. Becker P.B. Wilson S. Wu C. Cell. 1990; 63: 1085-1097Abstract Full Text PDF PubMed Scopus (266) Google Scholar, 18Nakai A. Tanabe M. Kawazoe Y. Inazawa J. Morimoto R.I. Nagata K. Mol. Cell. Biol. 1997; 17: 469-481Crossref PubMed Scopus (284) Google Scholar). Although a consensus has yet to be reached (19Zandi E. Tran T.-N.T. Chamberlain W. Parker C.S. Genes Dev. 1997; 11: 1299-1314Crossref PubMed Scopus (44) Google Scholar, 20Saltsman K.A. Prentice H.L. Kingston R.E. Yeast. 1998; 15: 733-746Crossref Scopus (2) Google Scholar), HR-C is suggested to maintain HSF in an inactive state by interacting intramolecularly with the HR-A/B (21Rabindran S.K. Haroun R.I. Clos J. Wisniewski J. Wu C. Science. 1993; 259: 230-234Crossref PubMed Scopus (384) Google Scholar, 22Zuo J. Balar R. Dahl G. Voellmy R. Mol. Cell. Biol. 1994; 14: 7557-7568Crossref PubMed Scopus (164) Google Scholar, 23Farkas T. Kutskova Y.A. Zimarino V. Mol. Cell. Biol. 1998; 18: 906-918Crossref PubMed Scopus (61) Google Scholar). HSF1 is also negatively regulated in trans by molecular chaperone complexes (24Ali A. Bharadwaj S. O'Carroll R. Ovsenek N. Mol. Cell. Biol. 1998; 18: 4949-4960Crossref PubMed Scopus (241) Google Scholar, 25Zuo J. Guo Y. Guettouche T. Smith D.F. Voellmy R. Cell. 1998; 94: 471-480Abstract Full Text Full Text PDF PubMed Scopus (933) Google Scholar, 26Shi Y. Mosser D. Morimoto R.I. Genes Dev. 1998; 12: 654-666Crossref PubMed Scopus (486) Google Scholar, 27Bharadwaj S. Ali A. Ovsenek N. Mol. Cell. Biol. 1999; 19: 8033-8041Crossref PubMed Scopus (146) Google Scholar). Sequestration of molecular chaperones to denatured substrates by heat stress is speculated to lead to the activation of HSF1. The existence of a bipartite type of NLS was suggested in HSF2, and these sequences are conserved among HSFs (28Sheldon L.A. Kingston R.E. Genes Dev. 1993; 7: 1549-1558Crossref PubMed Scopus (75) Google Scholar). Avian cells respond to heat stress by activating HSF3 as well as HSF1. Chicken HSF3 (cHSF3) stays as an inactive dimer exclusively in the cytoplasm in normally growing cells and is translocated into the nucleus and converted to an active trimer upon heat stress (29Nakai A. Kawazoe Y. Tanabe M. Nagata K. Morimoto R.I. Mol. Cell. Biol. 1995; 15: 5168-5178Google Scholar). Activation of cHSF3 exhibits a delayed response as compared with the relatively rapid activation of HSF1 and occurs predominantly upon exposure to extremely high temperatures (30Tanabe M. Nakai A. Kawazoe Y. Nagata K. J. Biol. Chem. 1997; 272: 15389-15395Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar, 31Kawazoe Y. Tanabe T. Sasai N. Nagata K. Nakai A. Eur. J. Biochem. 1999; 265: 688-697Crossref PubMed Scopus (27) Google Scholar). Fundamentally, however, the activation mechanism of cHSF3 may be similar to that of HSF1 (32Nakai A. Cell Stress Chaperone. 1999; 4: 86-93Crossref PubMed Scopus (80) Google Scholar). Analysis of cHSF3-deficient B lymphocytes showed that cHSF3 is essential for the burst activation of heat shock genes even in the presence of cHSF1 (33Tanabe M. Kawazoe Y. Takeda S. Morimoto R.I. Nagata K. Nakai A. EMBO J. 1998; 17: 1750-1758Crossref PubMed Scopus (85) Google Scholar). cHSF3 is also specifically activated by the c-Myb proto-oncogene product through a direct interaction and may involve in cell growth and development (34Kanei-Ishii C. Tanikawa J. Nakai A. Morimoto R.I. Ishii S. Science. 1997; 277: 246-248Crossref PubMed Scopus (55) Google Scholar). Thus, cHSF3 has significant roles in activating heat shock genes in response to heat stress and physiological stress. Studies using various experimental systems have attempted to identify cis-acting domains that regulate the monomer-to-trimer transition of HSF1. In addition to the core hydrophobic heptad repeat domains (HR-A/B and HR-C domains), several other domains play roles in maintaining HSF1 in an inactive form (35Orosz A. Wisniewski J. Wu K. Mol. Cell. Biol. 1996; 16: 7018-7030Crossref PubMed Scopus (66) Google Scholar, 36Liu P.C.C. Thiele D.J. J. Biol. Chem. 1999; 274: 17219-17225Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). However, no domain has been identified to participate positively in activating HSFs. In this study, we analyzed the dimer-to-trimer transition of a series of deletion mutants of chicken HSF3 and found domains that are necessary for trimer formation. One of them coincides with the NLS of cHSF3. We suggest a novel regulatory mechanism of HSF regulation. Chicken HSF3 expression vector pCMV-cHSF3 was constructed by ligating a 2665-base pairEcoRI fragment of pCHSF3–14a into the vector pCMV/Blue (Pharmingen). All of the expression vectors of cHSF3 with deletions or point mutations were created by using QuickChange site-directed mutagenesis kit (Stratagene) and pCMV-cHSF3 plasmid as a template. To confirm mutations, sequencing reactions were performed by using ALFexpress AutoRead sequencing kit (Amersham Pharmacia Biotech), ALFexpress dATP labeling mix, and each synthetic oligonucleotide. Sequences were analyzed by using an ALFexpress sequencer (Amersham Pharmacia Biotech). To construct expression vectors pGFP-aa (), pGFP-aa (), and pGFP-aa (), each cDNA fragment was modified by polymerase chain reaction to introduce an EcoRI site and ligated into pEGFP2-C1 (which is an expression vector of a tandem repeat GFP (a gift from Y. Eguchi, Osaka University). To examine sequences, sequence reactions were performed by using a pEGFP-C1 primer (5′-CATGGTCCTGCTGGAGTTCGTG-3′). To construct pGEX2T-NLS vector, a cDNA fragment corresponding to amino acids 202–224 of cHSF3 was modified by polymerase chain reaction to introduce a SmaI site and ligated into pGEX2T plasmid vector (Amersham Pharmacia Biotech). pGEX2T-NLS6, pGEX2T-NLS7, and pGEX2T-NLS8 were created by using QuickChange site-directed mutagenesis kit (Stratagene) and pGEX2T-NLS vector as a template. To confirm mutations, sequence reactions were performed by using a pGEX2T primer (5′-GTATTGAAGCTATCCCAC-3′). DT40 cells were maintained as described previously (33Tanabe M. Kawazoe Y. Takeda S. Morimoto R.I. Nagata K. Nakai A. EMBO J. 1998; 17: 1750-1758Crossref PubMed Scopus (85) Google Scholar). For generating cells expressing a mutant cHSF3, 1 × 107 of cHSF3-null cells (33Tanabe M. Kawazoe Y. Takeda S. Morimoto R.I. Nagata K. Nakai A. EMBO J. 1998; 17: 1750-1758Crossref PubMed Scopus (85) Google Scholar) suspended in 0.7 ml of PBS were mixed with 0.1 ml of DNA solution that contained a linear pCMV-cHSF3 mutant plasmid vector (25 μg) digested with XhoI and a linear pZeoSV2(+) (3 μg) (Invitrogen) digested with XhoI. After keeping on ice for 10 min, cells were electroporated with a Gene Pulser apparatus (Bio-Rad) at 550 V and 25 microfarad and then incubated on ice for 10 min. After suspension in 20 ml of fresh medium, cells were divided into two 96-well plates and incubated for 20 h. Medium containing zeocine (Invitrogen) was then added to a final concentration of 300 μg/ml. After 10–14 days, drug-resistant clones were expanded, and whole cell extracts were prepared as described previously (30Tanabe M. Nakai A. Kawazoe Y. Nagata K. J. Biol. Chem. 1997; 272: 15389-15395Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar). Aliquots containing equal amount of protein were subjected to SDS-PAGE, transferred to a nitrocellulose membrane, and then blotted with αHSF3γ antiserum as described previously (29Nakai A. Kawazoe Y. Tanabe M. Nagata K. Morimoto R.I. Mol. Cell. Biol. 1995; 15: 5168-5178Google Scholar). The level of cHSF3 protein in each clone was compared with that of endogenous cHSF3 in wild type DT40 cells, and clones expressing less cHSF3 protein than wild type cells were analyzed further. To examine the oligomer form of cHSF3 mutant-2 (amino acids 392–467 were deleted), antiserum αHSF3β that recognizes the whole cHSF3 molecule was used (data not shown). Size fractionation of cell extracts was performed as described previously (29Nakai A. Kawazoe Y. Tanabe M. Nagata K. Morimoto R.I. Mol. Cell. Biol. 1995; 15: 5168-5178Google Scholar). Briefly, whole cell extracts (200 μl containing 500–1000 μg of protein) from each cell were applied to a Superdex 200 HR column for fast protein liquid chromatography (Amersham Pharmacia Biotech). The samples were eluted at 0.3 ml/min with a buffer containing 1% glycerol, 20 mmTris-HCl (pH 7.9), 200 mm KCl, and 1.5 mmMgCl2. The fractions (0.5 ml) were precipitated with trichloroacetic acid, washed with acetone, dried, suspended in gel sample buffer, and analyzed by SDS-PAGE and Western blot. Cells were washed with PBS and stuck to a slide glass by using a centrifugal cell collector (Tomy, Tokyo, Japan) according to the manufacturer's instructions. They were then fixed with 4% paraformaldehyde. After blocking with 1% bovine serum albumin (fraction V) in PBS at room temperature for 1 h, cells were incubated with a 1: 300 dilution of antiserum (αHSF3γ) in 1% bovine serum albumin/PBS/10% normal goat serum in PBS, and then with fluorescein isothiocyanate-conjugated goat anti-rabbit IgG antibody (1: 200 dilution) (Cappel). The cells were washed with PBS and mounted in 80% glycerol in 10 mm Tris-HCl (pH 7.5). Immunolabeled cells were visualized by using fluorescence microscopy (Axiophot, Zeiss). QT6 quail fibroblasts were maintained at 37 °C in 5% CO2 in Dulbecco's modified Eagle's medium containing 5% fetal calf serum, 2% chicken serum, and 2% tryptose phosphate broth. Cells were cultured on glass coverslips for 24 h and were transfected with each expression vector by the CaPO4 method. After 14 h in transfection salts, the plates were washed three times with PBS, fed with fresh medium, and incubated further for 24 h. Then the cells were fixed with 4% paraformaldehyde for 30 min at 4 °C. The coverslips were washed and mounted in 80% glycerol in 10 mm Tris-HCl (pH 7.5) on coverglasses. Photograph was taken by using by fluorescence microscopy (Axiophot, Zeiss). To examine the level of GFP fusion proteins, cells were lysed in Nonidet P-40 lysis buffer consisting of 150 mm NaCl, 1% Nonidet P-40, 50 mm Tris (pH 8.0), 1 mmphenylmethylsulfonyl fluoride, 1 μg/ml leupeptin, and 1 μg/ml pepstatin. Western blot analysis was then performed using anti-GFP polyclonal antibody (CLONTECH). HeLa cells were maintained in Dulbecco's modified Eagle's medium containing 10% fetal calf serum and these extracts were prepared in Nonidet P-40 lysis buffer. Cultures of Esherichia coli (strain JM109) transformed with pGEX2T, pGEX2T-NLS, pGEX2T-NLS6, pGEX2T-NLS7, and pGEX2T-NLS8 were treated with 0.4 mm isopropyl-1-thio-β-d-galactopyranoside for 3 h. The cells were washed and suspended in PBS containing 1 mm phenylmethylsulfonyl fluoride, 1 μg/ml leupeptin, and 1 μg/ml pepstatin. After sonication, soluble fractions were mixed with glutathione-Sepharose 4B beads (Amersham Pharmacia Biotech) on ice for 1 h. After being washed with PBS, beads (bed volume of 20 μl) were mixed with HeLa cell extracts on ice for 2 h. After being washed with PBS, the beads were added to SDS sample buffer (20 μl) and boiled at 95 °C for 3 min. Proteins bound to the beads were analyzed by Western blot analysis by using mouse monoclonal antibody against importin α (anti-Rch-1, Transduction Laboratories) and peroxidase-conjugated goat anti-mouse IgG antibody (Cappel) as a second antibody. The signals were visualized using an enhanced chemiluminescence system (Amersham Pharmacia Biotech). We transfected cHSF3 expression vector into DT40 cells in which HSF3 gene was disrupted (33Tanabe M. Kawazoe Y. Takeda S. Morimoto R.I. Nagata K. Nakai A. EMBO J. 1998; 17: 1750-1758Crossref PubMed Scopus (85) Google Scholar) and isolated lines stably expressing cHSF3 under the control of cytomegalovirus promoter. The heat shock response was restored in these cells (33Tanabe M. Kawazoe Y. Takeda S. Morimoto R.I. Nagata K. Nakai A. EMBO J. 1998; 17: 1750-1758Crossref PubMed Scopus (85) Google Scholar). Unexpectedly, cHSF3 was observed as two bands with molecular sizes of 65 and 85 kDa. The mobility of the lower band is identical to that of a translated product from cHSF3 transcript in rabbit reticulocyte lysate, whereas the mobility of the upper band is identical to that of endogenous cHSF3 in cells (Fig.1 A). These results suggest that cHSF3 polypeptide is modified post-translationally (see “Discussion”). We examined the oligomer form of the reintroduced cHSF3 by gel filtration analysis. The 85-kDa form of the reintroduced cHSF3 was eluted at the dimer position, fractions 21–24, and after heat shock at 45 °C for 30 min, it was eluted at the trimer position, fractions 18–20 (Fig. 1 B, panels d–f,upper arrows), just like endogenous cHSF3 (Fig.1 B, panels a–c) (29Nakai A. Kawazoe Y. Tanabe M. Nagata K. Morimoto R.I. Mol. Cell. Biol. 1995; 15: 5168-5178Google Scholar). In contrast, the 65-kDa form was eluted at a monomer position (Fig. 1 B, panel d, lower arrow). After heat shock, its elution peak was observed a little at a trimer position and disappeared from the soluble fraction at higher temperatures such as 45 °C (Fig. 1 B,panel f). These results suggest that a dimer-to-trimer transition of the 85-kDa form of reintroduced cHSF3 is correctly regulated just like endogenous cHSF3 irrespective of the presence of the 65-kDa form. In this study, we examined the oligomer form of only the 85-kDa form. We generated cells stably expressing a series of deletion mutants of cHSF3 (Fig.2). Because negative regulation of HSF1 activation was shown to be limited (3Sarge K.D. Murphy S.P. Morimoto R.I. Mol. Cell. Biol. 1993; 13: 1392-1407Crossref PubMed Scopus (744) Google Scholar, 19Zandi E. Tran T.-N.T. Chamberlain W. Parker C.S. Genes Dev. 1997; 11: 1299-1314Crossref PubMed Scopus (44) Google Scholar, 21Rabindran S.K. Haroun R.I. Clos J. Wisniewski J. Wu C. Science. 1993; 259: 230-234Crossref PubMed Scopus (384) Google Scholar) and was shared with negative regulation of cHSF3 activation (32Nakai A. Cell Stress Chaperone. 1999; 4: 86-93Crossref PubMed Scopus (80) Google Scholar, 33Tanabe M. Kawazoe Y. Takeda S. Morimoto R.I. Nagata K. Nakai A. EMBO J. 1998; 17: 1750-1758Crossref PubMed Scopus (85) Google Scholar), we analyzed cells that expressed less cHSF3 mutant protein than endogenous cHSF3 protein in wild type DT40 cells (Fig. 2 B). cHSF3 mutants lacking an N terminus (see lines Z10, Z11, and Z12) or carboxyl terminus (line Z2; data not shown and see “Materials and Methods”) were not modified. The central region including HR-B was also necessary for the modification (see line Z8). We analyzed the oligomer forms of all of the cHSF3 mutants in control and heat shocked cells by gel filtration analysis (29Nakai A. Kawazoe Y. Tanabe M. Nagata K. Morimoto R.I. Mol. Cell. Biol. 1995; 15: 5168-5178Google Scholar).Figure 2Structure and expression of cHSF3 mutants. A, schematic representation of cHSF3 deletions containing a wild type (number 1) and deletion mutants (numbers 2–15). Positions of the deleted amino acids are indicated in each mutants. B, expression of mutant cHSF3 proteins in HSF3-deficient DT40 cells. Each expression vector was stably transfected into cells, and whole cell extracts from control and heat shocked (43 °C for 30 min) cells were prepared. Aliquots containing same amounts of protein (40 μg) were subjected to SDS-PAGE followed by Western blotting using an antiserum αHSF3γ. The protein levels of cHSF3 mutants were below that of endogenous cHSF3 in wild type (WT) DT40 cells. The cell lines expressing cHSF3 mutants are denoted as Z1–Z15 according to the serial number of cHSF3 mutants shown above.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Previous works analyzing the monomer-to-trimer transition of mammalian and Drosophila HSF1 identified two hydrophobic heptad repeats that maintain HSF1 in a monomer state (21Rabindran S.K. Haroun R.I. Clos J. Wisniewski J. Wu C. Science. 1993; 259: 230-234Crossref PubMed Scopus (384) Google Scholar, 22Zuo J. Balar R. Dahl G. Voellmy R. Mol. Cell. Biol. 1994; 14: 7557-7568Crossref PubMed Scopus (164) Google Scholar, 23Farkas T. Kutskova Y.A. Zimarino V. Mol. Cell. Biol. 1998; 18: 906-918Crossref PubMed Scopus (61) Google Scholar). We reported a similar observation that a cHSF3 mutant lacking carboxyl-terminal region containing the hydrophobic heptad repeat (HR-C) exhibited a DNA-binding form in vitro in rabbit reticulocyte lysate (37Nakai A. Morimoto R.I. Mol. Cell. Biol. 1993; 13: 1983-1997Crossref PubMed Scopus (208) Google Scholar). We extended these works and showed that a mutant lacking a hydrophobic heptad repeat at the carboxyl terminus formed a trimer in the absence of heat shock (Fig.3 A, line Z3). Furthermore, substitution of isoleucine at amino acid 371 or leucine at amino acid 375 in the HR-C with glutamic acid made cHSF3 constitutively a trimer (data not shown). Although there is controversy about the role of HR-C (19Zandi E. Tran T.-N.T. Chamberlain W. Parker C.S. Genes Dev. 1997; 11: 1299-1314Crossref PubMed Scopus (44) Google Scholar, 20Saltsman K.A. Prentice H.L. Kingston R.E. Yeast. 1998; 15: 733-746Crossref Scopus (2) Google Scholar), our data conclusively showed that this structure plays a role in maintaining cHSF3 in an inactive oligomeric state. Previous studies have identified some regions of HSF1 other than the HR-C that contribute to the suppression of HSF1 trimerization (21Rabindran S.K. Haroun R.I. Clos J. Wisniewski J. Wu C. Science. 1993; 259: 230-234Crossref PubMed Scopus (384) Google Scholar, 35Orosz A. Wisniewski J. Wu K. Mol. Cell. Biol. 1996; 16: 7018-7030Crossref PubMed Scopus (66) Google Scholar,36Liu P.C.C. Thiele D.J. J. Biol. Chem. 1999; 274: 17219-17225Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). However, there is no evidence that suggests the existence of domain acting positively on the HSF1 trimerization. Surprisingly, we found that two deletion mutants of cHSF3, cHSF3 mutant-6 and mutant-14, remained as dimer even after cells were heat shocked at 45 °C for 30 min. Regions necessary for cHSF3 trimerization include one containing a putative NLS downstream of the trimerization domain (formerly denoted as NLS2 in human HSF2; see Ref. 28Sheldon L.A. Kingston R.E. Genes Dev. 1993; 7: 1549-1558Crossref PubMed Scopus (75) Google Scholar) (Fig. 3 B, line Z6) and a carboxyl-terminal region of the DNA-binding domain (which corresponds to a linker region of hHSF1; see Ref. 36Liu P.C.C. Thiele D.J. J. Biol. Chem. 1999; 274: 17219-17225Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar) (Fig.3 C, line Z14). The significance of the latter region will be discussed elsewhere. The lack of trimer formation of cHSF3 mutant-6 and mutant-14 was observed not only when cells were heat shocked but also when cells were treated with sodium arsenite, hydrogen peroxide, high concentration of NaCl (osmotic shock), proline analogue azetidine, prostaglandin A1, and proteasome inhibitor lactacystin (data not shown). These results suggest that stress signals are not transmitted to cHSF3 without the putative NLS. To examine whether the putative NLS, which is necessary for the trimer transition of cHSF3, is functional, we analyzed first the subcellular localization of cHSF3 mutants. Endogenous and reintroduced wild type cHSF3 changed their localization from the cytoplasm to the nucleus after heat shock (Fig.4, lines wt andZ1). In contrast, the deletion mutant of the putative NLS did not move to the nucleus after heat shock (line Z6). The cHSF3 mutant-14 did not translocate to the nucleus after heat shock (Fig. 4,line Z14), nor did a mutant-15, which lacks another putative NLS between the DNA-binding domain and oligomerization domain (formerly denoted as NLS1 in human HSF2) (28Sheldon L.A. Kingston R.E. Genes Dev. 1993; 7: 1549-1558Crossref PubMed Scopus (75) Google Scholar), move to the nucleus (line Z15). Other cHSF3 mutants did translocate to the nucleus after heat shock (Fig. 4). Staining patterns of cHSF3 and cHSF3 mutants in the nucleus of heat shocked cells were same. We next examined whether each of the three regions necessary for nuclear translocation of cHSF3 could transport GFP when fused to one of these polypeptides. We expressed GFP-NLS fusion protein as well as GFP fused to another putative nuclear localization sequence (amino acids 111–139) or GFP fused to another region (amino acids 79–110), which is also necessary for trimer formation (Fig.5, A–C). GFP-NLS was accumulated in the nucleus (Fig. 5 B, lane c), whereas GFP, GFP-aa (), and GFP-aa () located mostly in the cytoplasm (Fig. 5 B, panels a, b, and d). We further examined whether the NLS of cHSF3 was recognized by a nuclear import receptor. The NLS of cHSF3 (amino acids 202–224) was fused to carboxyl terminus of GST (GST-NLS), and a GST-NLS column was generated. After mixing with HeLa cell extract, complexes bound to GST-NLS were examined (Fig. 5 D). Because importin α is known to bind to the typical bipartite NLS among various kinds of nuclear import receptors, we performed immunoblotting with antibody against importin α. It was revealed that importin α efficiently bound to the NLS of cHSF3 (Fig. 5 D, lane 3). These results indicate that the NLS (amino acids 202–224) in cHSF3 is a classical NLS. To further clarify which amino acids are responsible for both trimer formation and nuclear translation, we created a series of point mutations and examined their oligomer formation and subcellular localization before and after heat shock. The NLS in cHSF3 could be a bipartite nuclear localization sequence consisting of four charged amino acids (204KRKR207) and one charged amino acid (Lys219) spaced by 11 amino acids (Fig.6 A). Clusters of charged amino acids are known to be necessary and sufficient for nuclear localization. Therefore, we substituted one or two of charged amino acids into alanine in the NLS sequences (Fig. 6 A). Substitution of a single amino acid did not affect the heat shock-induced nuclear translocation (Fig. 6 A, mutants NLS1–NLS5; data not shown). cHSF3 could still move to the nucleus after heat shock, although less efficiently, by the substitution of two amino acids including Lys (amino acid 204) in a cluster of charged amino acids and R (amino acid 196 or 222) (Fig. 6, A andB, mutants NLS7 and NLS9). The nuclear translocation was clearly blocked when two Lys (amino acids 204 and 219) were substituted (Fig. 6, A and B,mutant NLS8). Deletion of the KRKR sequence (amino acid 204–207) also prevented nuclear translocation (Fig. 6, Aand B, mutant NLS6). These results suggested that KRKR (amino acids 204–207) and Lys (amino acid 219) are essential amino acids for nuclear translocation of cHSF3. These amino acids necessary for nuclear translocation in the NLS of cHSF3 were consistent with those of one of the putative NLS of human HSF2 (NLS2; see Ref.28Sheldon L.A. Kingston R.E. Genes Dev. 1993; 7: 1549-1558Crossref PubMed Scopus (75) Google Scholar). We next examined the ability of mutated NLS to bind to importin α. Although the amount of importin α coprecipitated with GST-NLS7 was greatly reduced compared with that coprecipitated with GST-NLS, it was above the background level (Fig. 5 D,lanes 2 and 4). In contrast, neither GST-NLS8 nor GST-NLS-6 bound to importin α at all (Fig. 5 D, lanes 5 and 6). These results showed that the abilities of NLS mutants to bind to importin α are well correlated with nuclear localization (Fig. 6). We then examined the oligomeric form of these cHSF3 mutants. cHSF3 mutant NLS6, which lacked a cluster of charged amino acids KRKR, remained a dimer after heat shock, and more than half of the mutant NLS8, which did not move to the nucleus, eluted at dimer fractions (fractions 21–23) after heat shock at 45 °C for 30 min (fractions 18–20) (Fig. 6 C). In contrast, NLS7 was eluted mostly at the trimer position (Fig. 6 C), and effects of other point mutations in the NLS sequences on the trimer formation were undetectable (Fig. 6 A and data not shown). These results indicated that mutations that block nuclear translocation also prevent trimer formation. The basic amino acids in the NLS are conserved among all of the vertebrate HSFs (Fig. 6 D). Our mutational analysis of cHSF3 provided unexpected results. We identified two regions that are essential for trimer formation of cHSF3 upon heat stress. One of these regions is coincident with a functional NLS that is a classical bipartite type and is recognized by a nuclear import receptor importin α. Point mutations of the basic amino acids in the NLS inhibit both stress-induced trimer formation and nuclear translocation of cHSF3. Among vertebrate HSFs, HSF1 and HSF3 respond to environmental stresses such as heat stress and change subcellular localization. In the absence of stress, HSF1 and HSF2 are located both in the cytoplasm and in the nucleus (2Baler R. Dahl G. Voellmy R. Mol. Cell. Biol. 1993; 13: 2486-2496Crossref PubMed Scopus (392) Google Scholar, 3Sarge K.D. Murphy S.P. Morimoto R.I. Mol. Cell. Biol. 1993; 13: 1392-1407Crossref PubMed Scopus (744) Google Scholar, 28Sheldon L.A. Kingston R.E. Genes Dev. 1993; 7: 1549-1558Crossref PubMed Scopus (75) Google Scholar, 35Orosz A. Wisniewski J. Wu K. Mol. Cell. Biol. 1996; 16: 7018-7030Crossref PubMed Scopus (66) Google Scholar), HSF3 is mostly in the cytoplasm (29Nakai A. Kawazoe Y. Tanabe M. Nagata K. Morimoto R.I. Mol. Cell. Biol. 1995; 15: 5168-5178Google Scholar), and HSF4 is in the nucleus (18Nakai A. Tanabe M. Kawazoe Y. Inazawa J. Morimoto R.I. Nagata K. Mol. Cell. Biol. 1997; 17: 469-481Crossref PubMed Scopus (284) Google Scholar, 38Tanabe M. Sasai N. Nagata K. Liu X.D. Liu P.C. Thiele D.J. Nakai A. J. Biol. Chem. 1999; 274: 22756-27845Google Scholar). Upon heat stress, HSF1 and HSF3 accumulate in the nucleus, where they bind to DNA and activate heat shock gene expression. Thus, HSF1 and HSF3 are strictly inactivated by unique mechanisms in the absence of stress. The first step of HSF1 and HSF3 activation is the conversion to a trimer that is associated with nuclear translocation. The trimer formation of HSF1 is shown to be partly under the negative control of chaperone complexes (24Ali A. Bharadwaj S. O'Carroll R. Ovsenek N. Mol. Cell. Biol. 1998; 18: 4949-4960Crossref PubMed Scopus (241) Google Scholar, 25Zuo J. Guo Y. Guettouche T. Smith D.F. Voellmy R. Cell. 1998; 94: 471-480Abstract Full Text Full Text PDF PubMed Scopus (933) Google Scholar, 27Bharadwaj S. Ali A. Ovsenek N. Mol. Cell. Biol. 1999; 19: 8033-8041Crossref PubMed Scopus (146) Google Scholar), and the sequestration of heat shock proteins under condition of stress may alleviate the negative regulation and lead to the activation of HSF1. Negative regulation of cHSF3 should share this mechanism because depletion of cHSF3 resulted in increased negative regulation to cHSF1 activation (32Nakai A. Cell Stress Chaperone. 1999; 4: 86-93Crossref PubMed Scopus (80) Google Scholar, 33Tanabe M. Kawazoe Y. Takeda S. Morimoto R.I. Nagata K. Nakai A. EMBO J. 1998; 17: 1750-1758Crossref PubMed Scopus (85) Google Scholar). In this study, we suggest another mechanism that controls the first step of cHSF3 activation under stress conditions. Our results indicate that release from the negative regulator is not sufficient for the cHSF3 activation. How does the NLS act positively on trimer formation of cHSF3? One possibility is that the NLS may affect the structure of the trimerization domain (HR-A/B), which is located just upstream of the NLS. The HR-A/B is thought to interact with another hydrophobic heptad repeat HR-C on unstressed condition, and the interaction should be dissociated on stress conditions. Mutation of basic amino acids in the NLS may alter the structure of the HR-A/B and inhibit the stress-induced dissociation of the two hydrophobic heptad repeats. Another possibility is that positively acting factors may be necessary for trimer formation of cHSF3. These factors can interact to cHSF3 through the NLS. Nuclear import receptor is a candidate for a positively acting factor because they bind to the NLS of cHSF3 and could induce conformational change of cHSF3. Conformational change induced by binding of nuclear transport receptor is shown in the yeast transcriptional activator GAL4 (39Chan C.-K. Jans D.A. FEBS Lett. 1999; 462: 221-224Crossref PubMed Scopus (27) Google Scholar). It may be possible that nuclear transport receptor affects an oligomeric form of cHSF3 in addition to transport it to the nucleus. We are aware of some discrepancies in the roles of the NLS and HR-C. The HR-C of human and Drosophila HSF1 were first shown to maintain HSF1 as an inactive monomer in the absence of stress, as examined by overexpression of mutant proteins in cells (21Rabindran S.K. Haroun R.I. Clos J. Wisniewski J. Wu C. Science. 1993; 259: 230-234Crossref PubMed Scopus (384) Google Scholar). This result was confirmed by experiments conducted in Xenopusoocytes (22Zuo J. Balar R. Dahl G. Voellmy R. Mol. Cell. Biol. 1994; 14: 7557-7568Crossref PubMed Scopus (164) Google Scholar) and in vitro in rabbit reticulocyte lysate (23Farkas T. Kutskova Y.A. Zimarino V. Mol. Cell. Biol. 1998; 18: 906-918Crossref PubMed Scopus (61) Google Scholar). However, analysis of cells in which the expression of mutant HSF1 was from integrated promoter showed that HR-C is not necessary for maintaining Drosophila HSF1 in an inactive state (19Zandi E. Tran T.-N.T. Chamberlain W. Parker C.S. Genes Dev. 1997; 11: 1299-1314Crossref PubMed Scopus (44) Google Scholar). It is hard to compare these results because the experiments were not performed under physiological conditions. Our experimental system has the advantage that cHSF3 gene-disrupted cells stably expressing cHSF3 mutants are used. Regulation of cHSF3 mutants was monitored in nearly physiological conditions. Our results conclusively indicated that HR-C is necessary for maintaining cHSF3 in an inactive state (Fig. 3 A; data of point mutations were not shown), suggesting that the HR-C of human and Drosophila HSF1 play a similar role. HR-C is expected to prevent trimer formation through HR-A/B by binding intramolecularly to HR-A/B (21Rabindran S.K. Haroun R.I. Clos J. Wisniewski J. Wu C. Science. 1993; 259: 230-234Crossref PubMed Scopus (384) Google Scholar, 22Zuo J. Balar R. Dahl G. Voellmy R. Mol. Cell. Biol. 1994; 14: 7557-7568Crossref PubMed Scopus (164) Google Scholar, 23Farkas T. Kutskova Y.A. Zimarino V. Mol. Cell. Biol. 1998; 18: 906-918Crossref PubMed Scopus (61) Google Scholar). Heat stress may cause dissociation of the interaction and induce homo-trimerization of HR-A/B. cHSF3 exists as a dimer in normally growing cells. This dimer state is necessary for cHSF3 to be correctly converted to a trimer after heat shock. The unmodified form of cHSF3 (65 kDa) stayed a monomer and was converted to incorrectly aggregated after heat shock (Fig. 1). To understand the molecular mechanism of cHSF3 regulation, it is necessary to identify the nature of post-translational modification. We have tried to reveal the modification by examining phosphorylation, glycosylation, and covalent modifications in other polypeptides. Unfortunately, we could not identify the cHSF3 modification. The mobility of 85-kDa cHSF3 was not affected by phosphatase treatment of cell lysates (data not shown). cHSF3 did not bind to a concanavalin A column nor wheat germ agglutinin column (data not shown). cHSF3 was not conjugated by ubiquitin-like proteins such as ubiquitin, NEDD-8, sentrin-1 (Sumo-1), sentrin-2, or sentrin-3. 2T. Kamitani, personal communication. Because the 65-kDa cHSF3 is not degradation product (Fig. 1 A), 85-kDa cHSF3 must be post-translationally modified by unknown mechanism. Trimer formation of cHSF3 is shown to be specifically regulated by factors that are involved in cell proliferation, differentiation, and death. A proto-oncogene product c-Myb induces trimer formation of cHSF3 through binding to a DNA-binding domain of cHSF3 (34Kanei-Ishii C. Tanikawa J. Nakai A. Morimoto R.I. Ishii S. Science. 1997; 277: 246-248Crossref PubMed Scopus (55) Google Scholar). Tumor suppressor protein p53 also binds to cHSF3 and inhibits c-Myb-induced trimer formation of cHSF3 by competing c-Myb binding to cHSF3 (40Tanikawa J. Ichikawa-Iwata E. Kanei-Ishii C. Nakai A. Matsuzawa S. Reed J.C. Ishii S. J. Biol. Chem. 2000; 275: 15578-15585Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar). It is possible that positively acting factors such as c-Myb may regulate trimer formation of cHSF3 more generally on stress conditions. We are grateful to N. Imamoto and R. I. Morimoto for critical discussion, Y. Eguchi for plasmid, T. Kamitani for analysis of HSF3 modification, and Y. Kawazoe and M. Tanabe for technical assistance." @default.
• W2131356098 created "2016-06-24" @default.
• W2131356098 creator A5004693769 @default.
• W2131356098 creator A5044261042 @default.
• W2131356098 date "2000-11-01" @default.
• W2131356098 modified "2023-10-16" @default.
• W2131356098 title "A Nuclear Localization Signal Is Essential for Stress-induced Dimer-to-Trimer Transition of Heat Shock Transcription Factor 3" @default.
• W2131356098 cites W1479962123 @default.
• W2131356098 cites W1967138582 @default.
• W2131356098 cites W1973510010 @default.
• W2131356098 cites W1987871229 @default.
• W2131356098 cites W1987938397 @default.
• W2131356098 cites W1989996595 @default.
• W2131356098 cites W1990000982 @default.
• W2131356098 cites W1994131234 @default.
• W2131356098 cites W2014914786 @default.
• W2131356098 cites W2031793125 @default.
• W2131356098 cites W2042368077 @default.
• W2131356098 cites W2059524588 @default.
• W2131356098 cites W2068073076 @default.
• W2131356098 cites W2068903746 @default.
• W2131356098 cites W2070641117 @default.
• W2131356098 cites W2080082509 @default.
• W2131356098 cites W2080834659 @default.
• W2131356098 cites W2082157468 @default.
• W2131356098 cites W2083761143 @default.
• W2131356098 cites W2087423119 @default.
• W2131356098 cites W2089176336 @default.
• W2131356098 cites W2091477850 @default.
• W2131356098 cites W2101911420 @default.
• W2131356098 cites W2104699504 @default.
• W2131356098 cites W2107038109 @default.
• W2131356098 cites W2108201412 @default.
• W2131356098 cites W2118509212 @default.
• W2131356098 cites W2127283650 @default.
• W2131356098 cites W2129091231 @default.
• W2131356098 cites W2131405590 @default.
• W2131356098 cites W2145760315 @default.
• W2131356098 cites W2148890339 @default.
• W2131356098 cites W2152622809 @default.
• W2131356098 cites W2167257333 @default.
• W2131356098 doi "https://doi.org/10.1074/jbc.m005302200" @default.
• W2131356098 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/10942768" @default.
• W2131356098 hasPublicationYear "2000" @default.
• W2131356098 type Work @default.
• W2131356098 sameAs 2131356098 @default.
• W2131356098 citedByCount "26" @default.
• W2131356098 countsByYear W21313560982014 @default.
• W2131356098 countsByYear W21313560982015 @default.
• W2131356098 countsByYear W21313560982016 @default.
• W2131356098 countsByYear W21313560982017 @default.
• W2131356098 countsByYear W21313560982018 @default.
• W2131356098 countsByYear W21313560982019 @default.
• W2131356098 countsByYear W21313560982020 @default.
• W2131356098 crossrefType "journal-article" @default.
• W2131356098 hasAuthorship W2131356098A5004693769 @default.
• W2131356098 hasAuthorship W2131356098A5044261042 @default.
• W2131356098 hasBestOaLocation W21313560981 @default.
• W2131356098 hasConcept C104317684 @default.
• W2131356098 hasConcept C12554922 @default.
• W2131356098 hasConcept C137984847 @default.
• W2131356098 hasConcept C138885662 @default.
• W2131356098 hasConcept C178790620 @default.
• W2131356098 hasConcept C179926584 @default.
• W2131356098 hasConcept C185592680 @default.
• W2131356098 hasConcept C194232998 @default.
• W2131356098 hasConcept C205260736 @default.
• W2131356098 hasConcept C2776403692 @default.
• W2131356098 hasConcept C2776667301 @default.
• W2131356098 hasConcept C2779546866 @default.
• W2131356098 hasConcept C2780114586 @default.
• W2131356098 hasConcept C2780723820 @default.
• W2131356098 hasConcept C41895202 @default.
• W2131356098 hasConcept C49805395 @default.
• W2131356098 hasConcept C52981337 @default.
• W2131356098 hasConcept C55493867 @default.
• W2131356098 hasConcept C59157529 @default.
• W2131356098 hasConcept C68991219 @default.
• W2131356098 hasConcept C86339819 @default.
• W2131356098 hasConcept C86803240 @default.
• W2131356098 hasConcept C95444343 @default.
• W2131356098 hasConceptScore W2131356098C104317684 @default.
• W2131356098 hasConceptScore W2131356098C12554922 @default.
• W2131356098 hasConceptScore W2131356098C137984847 @default.
• W2131356098 hasConceptScore W2131356098C138885662 @default.
• W2131356098 hasConceptScore W2131356098C178790620 @default.
• W2131356098 hasConceptScore W2131356098C179926584 @default.
• W2131356098 hasConceptScore W2131356098C185592680 @default.
• W2131356098 hasConceptScore W2131356098C194232998 @default.
• W2131356098 hasConceptScore W2131356098C205260736 @default.
• W2131356098 hasConceptScore W2131356098C2776403692 @default.
• W2131356098 hasConceptScore W2131356098C2776667301 @default.
• W2131356098 hasConceptScore W2131356098C2779546866 @default.
• W2131356098 hasConceptScore W2131356098C2780114586 @default.
• W2131356098 hasConceptScore W2131356098C2780723820 @default.
• W2131356098 hasConceptScore W2131356098C41895202 @default.
• W2131356098 hasConceptScore W2131356098C49805395 @default.
• W2131356098 hasConceptScore W2131356098C52981337 @default.

• next