Journal list menu

Volume 588, Issue 10 p. 1899-1905
Short communication
Free Access

The Schizosaccharomyces pombe Hikeshi/Opi10 protein has similar biochemical functions to its human homolog but acts in different physiological contexts

Yuumi Oda

Yuumi Oda

Cellular Dynamics Laboratory, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan

Graduate School of Nanobioscience, Yokohama City University, 22-2 Seto, Kanazawa-ku, Yokohama, Kanagawa 236-0027, Japan

Search for more papers by this author
Makoto Kimura

Makoto Kimura

Cellular Dynamics Laboratory, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan

Search for more papers by this author
Shingo Kose

Shingo Kose

Cellular Dynamics Laboratory, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan

Search for more papers by this author
Milo B. Fasken

Milo B. Fasken

Department of Biochemistry, Emory University School of Medicine, Atlanta, GA 30322, USA

Search for more papers by this author
Anita H. Corbett

Anita H. Corbett

Department of Biochemistry, Emory University School of Medicine, Atlanta, GA 30322, USA

Search for more papers by this author
Naoko Imamoto

Corresponding Author

Naoko Imamoto

Cellular Dynamics Laboratory, RIKEN, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan

Corresponding author. Fax: +81 48 462 4716.Search for more papers by this author
First published: 24 April 2014
Citations: 5

Abstract

Human Hikeshi (HsHikeshi) is a nuclear import carrier for Hsp70s and is required for cell survival after heat shock. The Hikeshi homolog in Schizosaccharomyces pombe (SpHikeshi/Opi10) localizes to the nuclear rim, interacts with the Hsp70 homolog Ssa2, and mediates its nuclear import in a reconstituted mammalian nuclear transport system. However, SpHikeshi/Opi10 is not required for heat stress response and survival after heat stress. Instead, SpHikeshi/Opi10 is required for the normal expression of stress response genes under optimal conditions and for cell growth during glucose deprivation. Here, the functions of SpHikeshi/Opi10 are discussed and compared to the functions of HsHikeshi.

1 Introduction

Most nuclear proteins are transported between the cytoplasm and the nucleus by nucleocytoplasmic transport receptors (NTRs) of importin-β family (Imp-βs) through nuclear pores embedded in the nuclear envelope [1]. The transport by Imp-βs depends on the gradient of the GTP- and GDP-bound forms of the small GTPase Ran across the nuclear envelope [2]. Importin-α family proteins (Imp-αs) are adaptors for importin-β, the most well studied member of the Imp-β family [3]. In mammalian cells under heat stress conditions, these predominant importin-mediated transport pathways are down-regulated because of the collapse of the RanGTP/RanGDP gradient and the nuclear retention of Imp-αs [4, 5]. In mammalian and Drosophila cells, however, Hsp70s relocate from the cytoplasm to the nucleus during heat stress [6-9]. HsHikeshi is a nuclear transport carrier of Hsp70s and facilitates the nuclear import of Hsp70s under heat stress conditions. The nuclear import of Hsp70s by HsHikeshi is crucial for the attenuation and reversal of the heat-shock response and, thus, for cell viability after heat shock [10]. Although the gene coding for HsHikeshi (C11orf73) is conserved among eukaryotes, including Schizosaccharomyces pombe (SPBC21H7.06c) and Saccharomyces cerevisiae (YOL032W), no thermo sensitive mutants of the homologs have been described in yeasts. The only reported homolog is the S. cerevisiae OPI10 gene, and the opi10 mutant is associated with the Opi phenotype: overproduction and excretion of inositol in the absence of inositol and choline [11]. However, the biochemical functions of both S. cerevisiae Opi10p (ScOpi10p) and SpHikeshi/Opi10 have yet to be characterized. Thus, the S. pombe gene is tentatively annotated as an “OPI10 homolog” due to sequence identity. Here, we characterized SpHikeshi/Opi10 to explore the functional conservation between the human and S. pombe homologs.

2 Materials and methods

2.1 Strains

The S. pombe strains used are listed in Table 1 . Yeast extract (YE) medium and Edinburgh minimal medium (EMM) [12] were supplemented with adenine, uracil, and leucine as needed. The disruption of the opi10 gene and the absence of the SpHikeshi/Opi10 protein in the strain ED666 (opi10Δ) (Bioneer Inc.) were confirmed by PCR and Western blotting, respectively (Supplementary Fig. S1).

Table Table 1. S. pombe strains.
Strain Genotype Reference or source
ED666 h+ ade6-M210 ura4-D18 leu1-32 Bioneer Inc.
ED666 (opi10Δ) h+ opi10::kanMX6 ade6-M210 ura4-D18 leu1-32 Bioneer Inc.
AM2 (opi10-YFP) a h90 leu1-32::opi10YFH-leu1+ [17]
AM2 (ssa2-YFP) a h90 leu1-32::ssa2YFH-leu1+ [17]
AM2 (ssa2-YFP opi10Δ) a h+ opi10::kanMX6 leu1-32::ssa2YFH-leu1+ This work b
  • a YFP is contained in a C-terminal YFP-FLAG-His6 (YFH) tag.
  • b Constructed by genetic crosses as described [12].

2.2 Plasmids and proteins

The cDNAs coding for SpHikeshi/Opi10 and Ssa2 (SPCC1739.13) were provided by the National BioResource Project of Japan (clone names: spa73g12 and spa101a06, respectively). To express SpHikeshi/Opi10 in S. pombe, the cDNA was inserted into the pREP81 vector [13] with and without an N-terminal FLAG tag sequence. The SpHikeshi/Opi10-His6 protein was expressed in Escherichia coli using the pET21b vector (Novagen), purified on Ni-NTA agarose (Qiagen), and passed through HiTrap Q and Mono S columns (GE healthcare) for further purification. The His6 tagged green fluorescent protein (GFP)-Ssa2 fusion protein was expressed using the pQE80L vector (Qiagen) in E. coli and purified on TALON resin (Clontech).

2.3 FLAG-affinity purification and mass spectrometry

ED666 (opi10Δ)(pREP81-opi10) and ED666 (opi10Δ)(pREP81-FLAG-opi10) were grown in EMM at 30 °C to mid-log phase (3 × 107 cells/mL) and harvested. The cells were frozen in liquid N2, disrupted by Cryopress (Microtech Nichion), and suspended in ×1 cell volume of ×2 concentration of Nonidet buffer (NB) (20 mM HEPES-KOH (pH 7.3), 110 mM KOAc, 2 mM MgOAc, 5 mM NaOAc, 0.5 mM EGTA, 2 mM dithiothreitol, 0.2% Nonidet P40 and 1 μg/mL of each aprotinin, pepstatin A, and leupeptin). The lysates were centrifuged at 200,000×g for 15 min, and ×1/50 bed volume of anti FLAG M2 agarose (Sigma) equilibrated with NB was added to the supernatants. The mixtures were rotated for 2.5 h, and the resin was washed three times with NB. Then, the proteins were eluted with 100 μg/mL of FLAG peptide in NB and separated by SDS–PAGE. The protein band was excised and in-gel-digested using trypsin. The peptides were analyzed by LC–MS/MS (LTQ, Thermo Fisher Scientific) and identified using Mascot (Matrix Science) against the NCBI database. The parameters and the results of the Mascot search are shown in Supplementary Table S1.

2.4 Transport assay

Digitonin-permeabilized HeLa S3 cells were prepared [14] and a transport reaction was reconstituted using a heat-shocked HeLa S3 cytosolic extract depleted of Imp-βs and HsHikeshi as described previously [10].

2.5 Live-cell imaging of the S. pombe cells

S. pombe cells were cultured in a glass-bottom dish coated with concanavalin A in liquid EMM, and fluorescent images were captured with a DeltaVision RT microscopy system (Sekitechnotron). The temperature was quickly shifted up (to 43 °C) or down (to 30 °C) by medium exchange, and controlled by a heat chamber (MI-IBC Olympus). The medium temperature was directly measured by a probe thermometer (HD1100K Anritsu).

2.6 DNA microarray

ED666 and ED666 (opi10Δ) were cultured in liquid EMM at 30 °C. When the cell density was 3 × 107 cells/mL, the temperature was shifted to 43 °C for 1 h and then shifted back to 30 °C. Cells were harvested at four time points: at 30 °C before the initial shift up of temperature (t 1), after 1 h at 43 °C (t 2), and 1 (t 3) and 3 h (t 4) after the shift back down to 30 °C. The cells were disrupted with glass beads and RNA was purified as described [12, 15]. cDNA was synthesized and labeled with biotin using a GeneChip 3'IVT Express kit (Affymetrix) and hybridized to GeneChip Yeast Genome 2.0 arrays (Affymetrix). After incubation with streptavidin phycoerythrin, the arrays were scanned using a GeneChip Scanner 3000 7G (Affymetrix). The data were analyzed with GeneSpring GX ver. 12.5 (Agilent Technologies) with 75th percentile normalization. The experimental condition and data were deposited to the Gene Expression Omnibus database with the accession number GSE52259.

3 Results

3.1 SpHikeshi/Opi10 interacts with the Hsp70 homolog Ssa2 in S. pombe and mediates its nuclear import in a mammalian transport system

The gene coding for HsHikeshi is conserved in S. pombe and S. cerevisiae. Fig. 1 shows the deduced amino-acid sequences aligned by Clustal W2 [16]. In this alignment, 83, 38, and 53 amino acids are identical between SpHikeshi/Opi10 and HsHikeshi, ScOpi10p and HsHikeshi, and SpHikeshi/Opi10 and ScOpi10p, respectively (the protein lengths are shown in Fig. 1). SpHikeshi/Opi10 is more similar to HsHikeshi (42% identity) than ScOpi10p (19% identity to human), suggesting that the function of SpHikeshi/Opi10 may be more similar to that of HsHikeshi.

figure image
Amino acid sequences of SpHikeshi/Opi10, HsHikeshi and ScOp10p. The sequences of the Hikeshi homologs in S. pombe (O60175), H. sapiens (Q53FT3), and S. cerevisiae (Q08202) are aligned by ClustalW2. Highlighted in black, identical amino acids; shaded in gray, similar amino acids. In the shading, similarity between S. pombe protein and H. sapiens or S. cerevisiae protein is prioritized over that between H. sapiens and S. cerevisiae proteins.

When an SpHikeshi/Opi10 protein fused to yellow fluorescent protein (YFP) (SpHikeshi/Opi10-YFP) was expressed in S. pombe, it localized mainly to the nuclear envelope in a punctate manner reminiscent of nuclear pore complexes (NPCs) (as in the S. pombe Postgenome database [17]) (Fig. 2 A). Thus, SpHikeshi/Opi10 possibly interacts with the NPCs like HsHikeshi. To identify S. pombe soluble proteins that directly interact with SpHikeshi/Opi10, we expressed FLAG tagged or untagged SpHikeshi/Opi10 in the S. pombe opi10Δ strain and isolated protein complexes from the cell extracts using an anti-FLAG antibody. One protein with the molecular mass of 70 kDa specifically associated with the SpHikeshi/Opi10 (Fig. 2B). LC–MS/MS analysis revealed that the protein was Ssa2, an Hsp70 homolog (Supplementary Table S1). Ssa1, another Hsp70 homolog, was also identified, though it had lower peptide coverage. In the control (untagged) lane in Fig. 2B, a faint band with the same mobility as Ssa2 is seen, but the level is negligible. To determine whether SpHikeshi/Opi10 can import Ssa2 into cell nuclei, purified SpHikeshi/Opi10 and GFP-Ssa2 were analyzed in a reconstituted transport system using digitonin-permeabilized HeLa cells. In this system, hydrophobic NTRs like Imp-βs and Hikeshi in a HeLa cytosolic extract are depleted with phenyl Sepharose [10]; thus, the added substrate (GFP-Ssa2 in this case) is imported into the nuclei by the added NTR (SpHikeshi/Opi10 in this case) with the assistance of factors in the HeLa cytosolic extract and ATP. GFP-Ssa2 was imported into nuclei in an SpHikeshi/Opi10-dependent manner (Fig. 2C) when both the cytosolic extract and ATP were present (Fig. 2D). SpHikeshi/Opi10 mediated the nuclear import of human Hsc70 as well (Supplementary Fig. S2). These results indicate that SpHikeshi/Opi10 has similar biochemical properties, such as binding to Hsp70s and the NPCs, with HsHikeshi and mediates the nuclear import of Hsp70s.

figure image
SpHikeshi/Opi10 is the nuclear import carrier of Ssa2. (A) SpHikeshi/Opi10 localizes to the nuclear envelope. S. pombe cells expressing SpHikeshi/Opi10-YFP were observed by fluorescence microscopy. (B) SpHikeshi/Opi10 interacts with the Hsp70 homolog Ssa2. Untagged or FLAG tagged SpHikeshi/Opi10 was expressed from a plasmid in the S. pombe opi10Δ strain, and proteins associated with the FLAG-SpHikeshi/Opi10 were isolated from the cell extracts with an anti-FLAG antibody. The proteins were separated by SDS–PAGE, and the gel was silver stained. A major protein (arrowhead) specifically associated with the FLAG-SpHikeshi/Opi10 was identified as Ssa2 by mass spectrometry (Supplementary Table S1). FLAG-SpHikeshi/Opi10 and immunoglobulin light chain (IgL) from the antibody showed the same mobility in SDS–PAGE. (C) SpHikeshi/Opi10 imports Ssa2 into the cell nuclei in a reconstituted HeLa transport system. Purified GFP-Ssa2 (2 μM) and the indicated amount of SpHikeshi/Opi10 were added into the reconstituted transport system consisting of permeabilized HeLa cells, NTRs-depleted HeLa cytosolic extract (CE), and an ATP regeneration system. The nuclear import of GFP-Ssa2 was observed by fluorescence microscopy. (D) Nuclear import of Ssa2 by SpHikeshi/Opi10 depends on CE and ATP. CE or ATP was omitted as indicated from the transport system used in (C). All reactions were performed in the presence of 2 μM SpHikeshi/Opi10. Apyrase was added to the ATP minus reaction mixture to hydrolyze residual ATP.

3.2 SpHikeshi/Opi10 is dispensable for the nuclear localization of Ssa2 in living cells and cell survival after heat stress

The S. pombe opi10 gene is a non essential gene. The disruption of the opi10 gene did not affect cell growth at the optimal temperature (Fig. 3A, 30 °C). Because HsHikeshi plays a crucial role in cell survival after heat shock [10], we examined the viability of the S. pombe opi10Δ strain at a high temperature. At 37 °C, the opi10Δ strain grew almost equivalently to the opi10+ (wt) strain (Fig. 3A, 37 °C). Both the strains do not form single colonies at higher temperatures (>38 °C), and a short time exposure to higher temperatures decreases cell viability. We examined the survival rates of the wt and opi10Δ strains after a severe heat treatment. Liquid cultures of the strains grown at 30 °C were incubated at 47 °C for 20 min. Then, a dilution series of cells were spotted on a plate and incubated at 30 °C. A significant population of cells died as a consequence of this heat treatment (40–60% colony survival), but the survival rates were almost the same between the wt and opi10Δ strains (Fig. 3 B). Cell growth after the heat treatment was also almost the same (Supplementary Fig. S3). Slow growth of S. cerevisiae opi10Δ strain was observed at 42 °C (Supplementary Fig. S4A), but the S. cerevisiae opi10Δ strain, like the S. pombe strain, survived equivalently as the wt strain after severe heat treatments (Supplementary Fig. S4B). Thus, SpHikeshi/Opi10 and ScOpi10p are dispensable for cell survival after heat stress.

figure image
SpHikeshi/Opi10 is dispensable for cell survival in response to heat stress and the nuclear localization of Ssa2 in living cells. (A) The S. pombe opi10+ (wt) and opi10Δ strains grow equivalently at a higher temperature. The wt and opi10Δ strains, ED666 and ED666 (opi10Δ), respectively, were streaked on YE plates and incubated at 30 or 37 °C for the indicated durations. (B) The S. pombe opi10+ (wt) and opi10Δ strains survive heat stress equivalently. Liquid cultures of the wt and opi10Δ strains grown to the mid-log phase at 30 °C were treated at 47 °C for 20 min or kept at 30 °C. The twofold dilution series of the cultures were spotted on a YE plate and incubated at 30 °C for 3 days. (C) Ssa2 is localized in the nuclei in the absence of SpHikeshi/Opi10. The S. pombe opi10+ (wt) and opi10Δ strains expressing Ssa2-YFP, AM2 (ssa2-YFP) and AM2 (ssa2-YFP opi10Δ), respectively, were cultured in liquid medium at 30 °C, and the temperature was shifted up to 43 °C. Fluorescent images of the live cells were captured at the indicated time points. Insets show magnifications of a typical cell.

Next, we analyzed the nuclear localization of Ssa2 during heat stress by live-cell imaging of the wt and opi10Δ strains expressing Ssa2-YFP (Fig. 3C). A heat-shock response was induced in S. pombe cells by applying a heat stress (Supplementary Fig. S5). At 30 °C, the Ssa2-YFP distributed to the nucleus and cytoplasm in both the wt and opi10Δ strains. At 10–20 min after the temperature shift to 43 °C, Ssa2-YFP accumulated in a restricted area in the nucleus around the nucleolus in both strains. After 30 min at 43 °C, Ssa2-YFP accumulated in a wider area in the nucleus. Again, no significant difference between the wt and opi10Δ strains was observed for the Ssa2-YFP localization. These results indicate that the depletion of SpHikeshi/Opi10 does not inhibit the nuclear localization of Ssa2 in living yeast cells (see Section 4).

3.3 SpHikeshi/Opi10 is involved in the regulation of cellular gene expression at the optimal temperature but not during heat stress

To investigate the effect of opi10 disruption on the cellular heat response, we performed a DNA microarray assay of the wt and opi10Δ strains. Gene expression profiles for the strains cultured at 30 °C (t 1), shifted to 43 °C and cultured for 1 h (t 2), and then returned to 30 °C and cultured for 1 h (t 3) and 3 h (t 4) were analyzed (Fig. 4 A). In cells shifted to 43 °C, the heat shock transcription factor Hsf1p translocated from the cytoplasm into the nucleus, and, after temperature shift back to 30 °C, Hsf1p relocalized to the cytoplasm (Supplementary Fig. S5). Thus, these temperature shifts induced heat-shock response and recovery. In the wt strain, 982 in 4814 total genes were induced more than twofold after the 1 h heat treatment (wt[t 2]/wt[t 1] > 2), and most of the 982 genes were also induced in the opi10Δ strain (Fig. 4B, lanes 2 and 6). At t 2 in the opi10Δ strain, only 2 or 8 of the 982 genes were expressed at levels higher (opi10Δ[t 2]/wt[t 2] > 2) or lower (opi10Δ[t 2]/wt[t 2] < 0.5), respectively, than twofold intensities of the wt strain. Similarly, 796 genes were repressed more than twofold in the wt strain after the 1 h heat treatment (wt[t 2]/wt[t 1] < 0.5), and at t 2 in the opi10Δ strain, only 2 genes among them were expressed at more (opi10Δ[t 2]/wt[t 2] > 2) or less (opi10Δ[t 2]/wt[t 2] < 0.5) than twofold intensities of the wt strain. Thus, during the heat treatment, the cellular gene expression profiles of both the strains were almost alike, arguing that SpHikeshi/Opi10 does not affect the heat stress response. The heat-induced or -repressed gene expression at t 2 was restored after the temperature shift back to 30 °C at t 3 and t 4 (Fig. 4B and C, lanes 3, 4, 7, and 8), and this gene expression recovery was almost equivalent between the wt and opi10Δ strains. Thus, the opi10 deletion does not affect largely on the recovery from heat shock response. Although certain differences in the gene expression between the strains were observed during the recovery (Fig. 4B and C, lanes 3, 4, 7, and 8), we could not find common features among the differentially expressed genes. Note that even after 3 h culture at 30 °C (t 4), the genes did not recover to the original expression levels shown at t 1, probably because complete recovery takes much longer time. This is supported by the fact that the cell growth that was arrested by the heat treatment restarted after 10 h culture at 30 °C (Supplementary Fig. S3).

figure image
SpHikeshi/Opi10 is involved in the regulation of cellular gene expression at the optimal temperature but not during heat stress. (A) Scheme of the heat treatment and recovery. The S. pombe opi10+ (wt) and opi10Δ strains, ED666 and ED666 (opi10Δ), respectively, cultured at 30 °C were heat-treated for 1 h at 43 °C, and then returned to 30 °C. At four time points (green dot, t 1t 4), the cells were harvested and the RNA was extracted for DNA microarray assays. (B) Gene expression profile of heat-induced genes. Normalized expression levels of the 982 genes that were induced by the heat treatment more than twofold (wt[t 2]/wt[t 1] > 2) are presented by a heat map. The expression levels are presented by colors as indicated by the range scale. Black represents unreliable values. On the right, clusters of genes differentially expressed in the wt and opi10Δ strains at t 1 are indicated (see text). (C) Gene expression profile of heat-repressed genes. Normalized expression levels of the 796 genes that were repressed by the heat treatment more than twofold (wt[t 2]/wt[t 1] < 0.5), are presented by a heat map.

At the optimal condition (before the heat treatment: t 1), however, a group of genes were down-regulated in the opi10Δ strain compared with the wt strain (Fig. 4B, compare lanes 1 and 5, regions indicated by “a” at the right). In the 982 heat-induced genes, 42 (4.3%) were down-regulated in the opi10Δ strain more than twofold compared to the wt strain at t 1 (opi10Δ[t 1]/wt[t 1] < 0.5, Supplementary Fig. S6), whereas in the 4814 total genes, only 62 (1.3%) were down-regulated in the opi10Δ strain more than twofold compared to the wt strain at t 1. Thus, the genes down-regulated at 30 °C (t 1) in the opi10Δ strain were concentrated in the heat-induced genes (binomial distribution P = 4 × 10−11). Importantly, Gene Ontology analysis revealed that 26 of the 42 genes (62%) are annotated as “response to stress”. Although 292 of the 982 heat-induced genes (30%) in the wt strain are also annotated as “response to stress”, the “response to stress” genes were concentrated in the 42 genes down-regulated in the opi10Δ strain (binomial distribution P = 1.6 × 10−5). Although some heat-repressed genes were also differentially expressed at the optimal condition (t 1: Fig. 4C, lanes 1 and 5), we could not find common features among those genes.

3.4 SpHikeshi/Opi10 is required for cell growth during glucose deprivation

S. pombe does not feed on glycerol because it lacks the glyoxylate cycle [18, 19]; however, the addition of 0.02–0.04% glucose into the glycerol-based medium supports slow growth [20]. In S. pombe, glucose deprivation induces endogenous oxidative stress [21], and at the same time, the expression of Pyp2, which is the phosphatase for Sty1 and Pmk1 mitogen-activated protein kinases (MAPKs) [22], is induced through the Sty1 mediated stress activated protein kinase (SAPK) pathway [23]. In our DNA microarray assay, 42 of the 982 heat-induced genes were down-regulated in the opi10Δ strain at 30 °C (opi10Δ[t 1]/wt[t 1] < 0.5), and 26 of these 42 genes are related to stress response, as mentioned above. Of them, only 2 are related to heat stress but 7 are related to oxidative stress, and the pyp2 gene was included in the 42 repressed genes (Supplemental Fig. S6). We compared the growth of the wt and opi10Δ strains on glycerol-based media. The opi10Δ strain grew much slower than the wt strain on both EMM (synthetic medium) and YE (complex medium) plates containing glycerol as the main carbon source (Fig. 5 ). Because the opi10Δ strain is sensitive to glucose deprivation, and because in the opi10Δ strain the expression of stress response genes was repressed in the DNA microarray assay, SpHikeshi/Opi10 must be involved in stress response. Although we examined sensitivity of the opi10Δ strain for several other stresses, such as exogenous oxidative stress, osmotic stress, low temperature, nitrogen deprivation, and heavy metal ions, we did not find any phenotypes.

figure image
The S. pombe opi10 disruptant strain is sensitive to glucose deprivation. The S. pombe opi10+ (wt) and opi10Δ strains, ED666 and ED666 (opi10Δ), respectively, were streaked on the plates and incubated for the indicated durations at 30 °C. (A) Minimal medium (EMM) containing 2% glucose (left) or 0.03% glucose and 3% glycerol (right). (B) Complex medium (YE) containing 3% glucose (left) or 3% glycerol (right).

4 Discussion

The nuclear accumulation of Hsp70s in mammalian and Drosophila cells during heat shock has been described for three decades without the elucidation of the localization mechanism or the nuclear function of Hsp70s [6-9]. Hikeshi was identified as the NTR for Hsp70s in human cells [10]. Although HsHikeshi is crucial for cell survival after heat shock and the gene is evolutionarily conserved, no reports in yeasts have connected this protein to heat stress or nuclear transport. In this study, we analyzed the functions of SpHikeshi/Opi10 in relation to the functions of HsHikeshi. The biochemical functions of HsHikeshi and SpHikeshi/Opi10 are conserved because SpHikeshi/Opi10 interacts with the Hsp70 family protein Ssa2 in S. pombe and imports Ssa2 into nuclei in a HeLa cell reconstituted transport system. The nuclear import of Ssa2 by SpHikeshi/Opi10 depends on cytosolic extract and ATP, indicating that the import activity of SpHikeshi/Opi10 involves the ATP-ADP exchange of Ssa2 mediated by an Hsp110 family protein, as in the case of HsHikeshi [10].

Unlike human Hsp70s, S. pombe Ssa2 localizes to the nucleus as well as in the cytoplasm at the optimal temperature. Further accumulation of Ssa2 in the nucleus during heat stress was scarcely observed. Therefore, human and S. pombe Hsp70s may respond differently to heat stress. Notably, Ssa2 localized to the nucleus under optimal and heat stress conditions, and this distribution was not perturbed by the opi10 gene disruption. This suggests that S. pombe has another NTR for Ssa2. S. cerevisiae might also have another NTR for Hsp70 family proteins. When Hikeshi is knocked-down in HeLa cells, the heat-induced nuclear import of Hsp70s is inhibited, and, consequently, cell survival after heat shock is reduced because the attenuation and reversal of the nuclear heat response is significantly delayed [10]. Thus, the viability of the S. pombe opi10Δ strain at 37 °C and after the heat treatment (47 °C for 20 min) may be attributable to the nuclear localization of Ssa2.

In HeLa cells depleted of Hikeshi, the heat-shock response is not inhibited [10]. In the S. pombe opi10Δ strain, the gene expression profile responding to the heat treatment (43 °C for 1 h) was similar to that in the wt strain. Thus, SpHikeshi/Opi10 may be dispensable for heat response, similar to HsHikeshi. Nevertheless, SpHikeshi/Opi10 is most likely involved in stress response because many stress-related genes were repressed in the opi10Δ strain at the optimal temperature and because the opi10Δ strain was sensitive to glucose deprivation, in which the SAPK and Pmk1 MAPK pathways are supposed to be activated [21, 23]. As glucose is not only an energy source but also the precursor of many molecules, including inositol, the participation of yeast Hikeshi homologs in other cellular processes is also possible. Thus, SpHikeshi/Opi10 and HsHikeshi function in different physiological contexts. In light of the non-descript growth phenotype of the S. pombe opi10Δ strain, a screen for gene mutations that exhibit synthetic phenotypes with the opi10Δ mutation may be helpful for further analysis of SpHikeshi/Opi10.

Acknowledgements

We thank Yoko Yashiroda, Akihisa Matsuyama, and Minoru Yoshida for providing yeasts and Yuriko Morinaka for technical assistance. We are grateful to the Support Unit for Bio-Material Analysis at the RIKEN BSI Research Resources Center for technical help with mass spectrometry and DNA microarray analysis. This work was supported by the (JSPS) through the “Funding Program for Next Generation World-Leading Researchers (NEXT Program)” initiated by the Council for Science and Technology Policy (CSTP) to NI.

    Appendix A A

    Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.febslet.2014.04.018.