Introduction

Viruses require host cell factors for their multiplication. The roles of such host factors (HFs) for virus multiplication have been investigated thoroughly in mammals, insects, and plants. Identification of HFs is important not only for understanding basic mechanisms underlying virus multiplication but also for therapeutics, in which HFs are attractive targets for antiviral agents [1,2,3]. Over the past decade, an increasing number of HFs for fish viruses have been identified. One trend in fish HF research is the discovery of defense-related factors, many of which are involved in the interferon system. These are categorized into positive [e. g., 4,5,6,7,8,9,10] and negative [e. g., 11,12,13,14,15] regulators of fish virus multiplication. Another research trend is the identification of heat shock proteins (HSPs) as fish HFs. Similar to the defense-related factors mentioned above, HSP HFs can be classified into two groups: facilitators [e. g., 16,17,18,19,20] and inhibitors [21, 22] of virus multiplication. The identification of the several HSPs as fish HFs is not surprising, because in mammals, plants, and insects, many HSPs are known to be HFs for viruses [23, 24]. HSPs comprise a large family of proteins and are known to function as molecular chaperones that are required for correct protein folding and intracellular transport in both prokaryotic and eukaryotic cells [25]. HSPs are classified into families (e.g., HSP60, HSP70, HSP90) according to their molecular weight, and each family has multiple isoforms [26]. However, many of the studies investigating HSPs as fish HFs were conducted without specifying which isoform was being studied [16,17,18, 20, 27]. In addition, although there have been several studies of HFs targeting specific isoforms of fish HSPs [22, 28,29,30,31], no comprehensive studies have been performed to clarify the biological importance of more than one isoform in a given HSP family. In this study, we focused on the HSP70 and HSP90 families in fish and examined the importance of many isoforms in these families as HFs. In order to improve the efficiency of the study, we used a betanodavirus, a model fish virus, and medaka (Oryzias latipes), a model fish, for the study.

Betanodaviruses are the causative agents of viral nervous necrosis of marine fish [32] and belong to the family Nodaviridae. Betanodaviruses are small in size [33] and possess a bipartite positive-sense RNA genome that is the smallest among fish viruses thus far identified [34, 35]. The larger genomic RNA (RNA1) is 3.1 kb in length and encodes an RNA-dependent RNA polymerase (protein A), while the smaller one (RNA2) is 1.4 kb in length and encodes a coat protein. The subgenomic RNA3, transcribed from RNA1, is 0.4 kb in length and encodes a suppressor for post-transcriptional gene silencing (protein B2) [36, 37] and an anti-necrotic death factor (protein B1) [38]. The virus can be classified into four genotypes, designated redspotted grouper nervous necrosis virus (RGNNV), striped jack nervous necrosis virus (SJNNV), barfin flounder nervous necrosis virus, and tiger puffer nervous necrosis virus, based on similarities in RNA2 sequences [39, 40]. The biological importance of the viral RNAs and encoded proteins have been studied extensively [37, 41,42,43,44,45] using reverse genetic technology based on a cDNA-mediated infectious RNA transcription strategy [46]. Recently, we established an experimental infection system of betanodaviruses using the model fish medaka and its cell lines [47, 48].

Medaka is an excellent vertebrate model that is used in various areas of study, including development, genetics, environmental toxicology, and human disease [49,50,51]. In addition to preferable characteristics as a model animal, such as small body size, high fecundity, tolerance to a wide range of environmental conditions, and easiness to rear, the availability of several inbred lines and enormous genetic diversity between those lines [52] are distinct features of medaka not found in other model fish species. Moreover, whole-genomic sequencing [53] and the establishment of transgenesis and mutagenesis systems have been done with medaka [54,55,56].

In this study, we performed overexpression and knockdown experiments on many genes of HSP70 and HSP90 isoforms in medaka to evaluate their importance in betanodavirus multiplication. We also examined the inducibility of these isoforms by heat shock or virus infection stress. The determination of which isoforms are involved in virus multiplication should not only help us to understand the mechanism of virus infection but also provide important information for the development of specific antiviral drugs targeting specific HSP isoforms.

Materials and methods

Cells and virus

The medaka cell line OLHNI-2 [57] was provided by H. Mitani and cultured at 30°C in Leibovitz's L-15 medium (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 15% fetal bovine serum (FBS) (Nichirei Biosciences, Tokyo, Japan). E-11 cells [58] were cultured at 25°C in L-15 medium supplemented with 5% FBS. The betanodavirus used in this study was RGNNV (strain SGWak97) [59]. The virus was prepared from OLHNI-2 cells transfected with in vitro-transcribed viral RNAs according to a procedure described previously [43]. The RNA sequences of the resulting progeny virus were verified as described previously [43]. Viral titers were determined based on the TCID50 method [60] using E-11 cells.

Plasmids

The mRNA sequences of the five HSP70 isoforms (HSP70-1, HSP70-2, HSP70-5, HSP70-8, and HSP70-9) and the five HSP90 isoforms (HSP90-α1, HSP90-α2, HSP90-β, GRP94, and TRAP1) of medaka were retrieved from Ensemble Genome Browser (http://www.ensembl.org/index.html) and NBRP medaka (http://www.shigen.nig.ac.jp/medaka/). The nomenclature of HSPs in the present study was according to the guidelines of Kampinga et al. [26] and Tavaria et al. [61]. Based on the mRNA sequences obtained, specific primers (Table 1) were designed to amplify DNA that encodes each of the HSP isoforms fused with an HA tag at the N- or C-terminus. PCR was performed using the primers and cDNA prepared from mRNA of the medaka strain Cab. Amplified fragments were digested with XbaI and cloned into the XbaI site of the pCI-Neo expression vector (Promega, Madison, WI, USA). The sequence integrity of all the constructs was verified by DNA sequencing using a BigDye Terminator v1.1 Cycle Sequencing Kit and an Applied Byosystems 3130xl Genetic Analyzer (Thermo Fisher Scientific).

Table 1 Oligonucleotides used in the present study
Full size table

Northern blot and Western blot analysis

Total RNA and proteins were prepared from cells inoculated with RGNNV using ISOGEN reagent (Nippon Gene, Tokyo, Japan) according to the manufacturer's instructions. Total RNA was subjected to Northern blot analysis as described previously [46]. Briefly, DIG-labelled RNA probes were hybridized with blots in the hybridization buffer containing 50% formamide, 750 mM NaCl, 75 mM sodium citrate, 2% blocking reagent (Roche Diagnostics, Basel, Switzerland), 0.1% N-lauroylsarcosine, and 0.02% sodium dodecyl sulfate (SDS) at 68°C for 6 h. The blots then were washed with a washing buffer containing 15 mM NaCl, 1.5 mM sodium citrate, and 0.1% SDS at 68°C. Hybridization signals were detected using anti-DIG-AP antibody (Roche Diagnostics) and CDP-Star (Roche Diagnostics) according to the supplier's instructions. The coat protein in the protein samples extracted from an equal number of cells was detected by Western blot analysis as described previously [46], except that a rabbit antiserum, raised against the recombinant RGNNV coat protein, was used as the primary antibody. The alkaline-phosphatase-conjugated goat anti-rabbit secondary antibody (Dako, Santa Clara, CA, USA) bound to the primary antibody on the blots was detected using CDP-Star. All chemiluminescent images of the blots were photographed using a ChemiDoc XR system (Bio-Rad, Hercules, CA, USA), and quantitative data were obtained by analysis using Quantity One software (Bio-Rad).

Treatment of cells with HSP inhibitors

The HSP90 inhibitors 17-(allylamino)-17-demethoxygeldanamycin (17-AAG) and radicicol were purchased from Wako Pure Chemical Industries (Osaka, Japan). Quercetin, which inhibits both HSP70 and HSP90, was purchased from the same company. All of the chemicals were dissolved in dimethyl sulfoxide (DMSO) as stock solutions (5 mM for 17-AAG and radiciol, 20 mM for quercetin) and stored at –20°C until use.

OLHNI-2 cells seeded in a 24-well culture plate (Sumitomo Bakelite, Tokyo, Japan) at a density of 0.5 × 106 cells per well were cultured for 16 h at 30°C. The cells were then inoculated with RGNNV at a multiplicity of infection (MOI) of 1.0 for 30 min and further cultured with medium containing HSP inhibitors. Control cells were incubated with medium containing the same amount of DMSO. At 12 h after virus inoculation, the culture supernatants were collected, and virus particles were purified using the PEG precipitation method [62] to remove the inhibitors from the samples. The viral titers of the samples thus obtained were determined as described above. Total RNA and proteins were also isolated from cells in replicate wells and subjected to Northern blot and Western blot analysis as described above. In order to assess the cytotoxic effects of the inhibitors, OLHNI-2 cells were seeded in a 96-well culture plate (Iwaki, Shizuoka, Japan) at a density of 0.5 × 105 cells per well. The cells were then cultured with medium containing each of the inhibitors, and cell viability was determined periodically using a WST-1 Cell Proliferation Assay Kit (Roche Diagnostics) according to the manufacturer’s recommendations.

Thermal stress on the cells

OLHNI-2 cells were seeded in a 6-well culture plate (Sumitomo Bakelite) at a density of 1.0 × 106 cells per well and cultured for 16 h at 30°C. The cells were then thermally stressed by submerging the culture plate in water at 42°C for 1 h, followed by recovery at 30°C for 1 h. The time point when the recovery step was finished was designated as 0 h after thermal stress. The cytotoxic effect of thermal stress was assessed using a WST-1 Cell Proliferation Assay Kit according to the manufacturer’s recommendations.

Overexpression of HSP genes

OLHNI-2 cells were transfected with plasmids encoding each of the HSP70 and HSP90 isoforms, using Lullaby Transfection Reagent (OZ Biosciences, Marseille, France) according to the procedure recommended by the manufacturer, with some modifications. Briefly, 500 ng of plasmid DNA and 1 μl of Lullaby Transfection Reagent were diluted separately in 50 μl Opti-MEM-I (Thermo Fisher Scientific). The diluted plasmid DNA and the transfection reagent then were combined and incubated for 20 min at room temperature. Finally, the transfection complex was combined with 400-μl cell suspensions (5 × 105 cells/ml), plated in a 24-well cell culture plate (Sumitomo Bakelite), and cultured for 48 h. These transfected cells were then used for virus inoculation.

Knockdown of HSP genes

siRNAs targeting each of the medaka HSP70 and HSP90 isoforms were designed using BLOCK-iT RNAi Designer with Stealth modification and purchased from Thermo Fisher Scientific. The Stealth modification is a chemical modification of the siRNA that enhances its potency and stability and reduces the occurrence of off-target effects. The target sequences of the siRNAs are listed in Table 2. Stealth RNAi GFP reporter control (Thermo Fisher Scientific), which targets the enhanced green florescent protein (EGFP) gene, was used as a control siRNA (siREGFP). OLHNI-2 cells were transfected with each of the siRNAs using X-tremeGENE siRNA Transfection Reagent (Roche Diagnostics) according to the manufacturer’s instructions. Total RNA was isolated from cells in replicate wells at 48 h post-transfection, and target mRNAs in the samples were quantified by real-time RT-PCR as described below. At 48 h post-transfection, the cells were used for virus inoculation.

Table 2 siRNAs used in the present study
Full size table

Quantitative real-time RT-PCR

Total RNA was isolated from cells using the acid guanidinium thiocyanate-phenol-chloroform method [63] and treated with RQ1 RNase-free DNase (Promega), followed by purification using an RNeasy Mini Kit (QIAGEN, Venlo, The Netherlands). To obtain cDNA, 0.5 µg of purified total RNA was incubated with 0.5 μl of the oligo dT primer (20 μM; Eurofins Genomics, Val Fleuri, Luxembourg) at 70°C for 10 min, and, after addition of 2 μl of 5× RT reaction buffer (Takara, Otsu, Japan), 2 μl of dNTP mix (2.5 mM each dNTP, Takara), 100 units of M-MLV reverse transcriptase (RNase H-, Takara), and 10 units of RNase inhibitor (Toyobo, Osaka, Japan) in a final reaction volume of 10 μl, incubated further at 42°C for 60 min. The synthesized cDNA samples were diluted tenfold with double-distilled water and stored at –20°C until use.

Quantitative real-time RT-PCR was performed using Thunderbird SYBR qPCR Mix (Toyobo) and a Chromo4 Real-Time PCR Detection System (Bio-Rad). The primer sets used to detect mRNAs of the HSP isoforms and β-actin are listed in Table 1. The reaction mixture for PCR contained 10 μl of Thunderbird SYBR qPCR Mix, 0.5 μl each of the primers (10 μM), 3 μl of the cDNA sample, and 6 μl of double-distilled water. The amplification procedures included one cycle of 3 min at 95°C, followed by 40 cycles of 30 s at 95°C and 30 s at 60°C. The specificity of the PCR reaction was confirmed by melting curve analysis. Relative mRNA expression was calculated by the standard curve method using β-actin mRNA expression as the control for normalization.

Immunocytochemistry

An indirect immunofluorescence assay to detect HA-tagged HSP70 and HSP90 was performed as described previously with some modifications [47]. Briefly, cells were fixed with 4% paraformaldehyde and permeabilized by treatment with 0.1% NP-40 in PBS. The cells then were treated with a 1:500 dilution of anti-HA monoclonal antibody (Wako Pure Chemical Industries), followed by treatment with a 1:2000 dilution of Alexa Fluor 555 goat anti-mouse IgG (Thermo Fisher Scientific). Cells were stained simultaneously with 4',6-diamidino-2-phenylindole (DAPI) (Dojindo, Kumamoto, Japan) and were observed under a fluorescence microscope (ORCA-1394 and AQUA-Lite version 1.10 systems; Hamamatsu Photonics, Hamamatsu, Japan).

Statistical analysis

Statistical analysis was performed using a two-tailed Student t-test to compare the means of two groups, and a one-way ANOVA followed by Tukey’s HSD test for comparing the means of more than two groups, as appropriate. A p-value less than 0.05 was considered statistically significant.

Results

Effects of the HSP inhibitors on virus multiplication in medaka cells

To determine whether the medaka HSP70 and HSP90 families include any HFs for RGNNV, we tested the effects of three well-known chemical inhibitors of HSPs on virus multiplication in medaka cells. Although the concentrations of inhibitors used in this study (5 μM for 17-AAG and radicicol, 25 μM for quercetin) have been suggested to be the most effective and the least toxic to insect cells [64], a modest cytotoxic effect was observed in medaka cells 24 h after treatment with these inhibitors (Fig. 1A). Accordingly, hereafter, effects of these inhibitors were investigated within 12 h post-treatment.

Fig. 1
figure 1

Effect of HSP inhibitors on virus multiplication. OLHNI-2 cells were incubated with 5 μM radiciocl, 5 μM 17-AAG, or 20 μM quercetin after inoculation with RGNNV (MOI = 1) for 30 min. Cells incubated with DMSO only are indicated as “vehicle”. Cells incubated with normal culture medium were used as a control. (A) Cytotoxic effects of the inhibitors, assessed using a WST-1 Cell Proliferation Assay Kit. The relative cell viability compared to the control is shown. (B) Production of progeny virus in culture supernatants, determined by the TCID50 method 12 h after virus inoculation. The numbers above the bars indicate the relative viral titers compared to the control cells. (C) Northern blot analysis of the viral RNAs and Western blot analysis of the coat protein, performed 12 h after inoculation with RGNNV. Representative data are shown. Positions of RNA1, RNA2, and RNA3 are indicated to the right. Expression of β-actin was measured as a loading control. (D) Quantitative analysis of the Northern blot and Western blot images in panel C. The chemiluminescent data of the blots were quantified using Quantity One software. The relative expression was calculated compared to control cells after normalization to β-actin expression. All data, except in panel C, are the mean values of three independent experiments with standard deviation. Statistical analysis was performed by one-way ANOVA followed by Tukey’s HSD test (*, p < 0.05; **, p < 0.01, compared with vehicle).

Full size image

First, the effects of these inhibitors on the production of progeny virus were examined. Viral titers in the supernatants 12 h after virus inoculation were reduced by a ratio of 0.28, 0.03, and 0.07 in cells treated with 17-AAG, radicicol, and quercetin, respectively, when compared to cells without any treatment (Fig. 1B). A slight reduction in viral titer was observed in control cells treated with vehicle only. To further elucidate the steps of virus multiplication affected by the HSP inhibitors, viral RNA and protein were analyzed quantitatively by Northern blotting and Western blotting, respectively. Production of viral RNA and protein was suppressed in cells treated with each of the inhibitors, the effect of which was the most significant in cells treated with radicicol (Fig 1C and D). No obvious difference was found in replication competence between the positive- and negative-strand RNAs. Interestingly, production of the coat protein was decreased by treatment with the inhibitors, the level of which was lower than that observed for RNA2 replication. These results suggest that inhibition of HSPs affected viral protein production as well as viral RNA replication. Alternatively, HSP inhibition might exclusively affect the production of viral proteins, including protein A, thereby reducing viral RNA replication.

Effect of the thermal stress on virus multiplication

Since the inhibition of HSP activity suppressed virus multiplication, we next examined the effect of artificial induction of HSPs on virus multiplication by subjecting cells to non-lethal thermal stress. We found that treatment of cells at 42°C for 1 h was the minimum requirement to upregulate HSPs (data not shown), although a slight cytotoxic effect on the cells was also observed at 12 h after treatment (Fig. 2A). When cells were inoculated with RGNNV immediately after the thermal stress, from 4.64 to 6.12 times more progeny virus was produced compared with control cells (Fig. 2B). Similarly, production of viral RNA and protein was increased by heat treatment (Fig. 2D and E). On the other hand, when cells were inoculated with RGNNV 12 h after the thermal stress, from 1.45 to 2.82 times more virus was produced (Fig. 2C).

Fig. 2
figure 2

Effect of thermal stress on virus multiplication. OLHNI-2 cells were thermally stressed at 42°C for 1 h and allowed to recover at 30°C for 1 h (designated as 0 h post-thermal-stress). Cells continuously incubated 30°C were used as a control. (A) Cytotoxic effect of thermal stress, assessed using a WST-1 Cell Proliferation Assay Kit. The relative cell viability against the control is shown. Thermally stressed cells were then inoculated with RGNNV (MOI = 1) for 1 h at 0 (B), or 12 h (C) post-thermal-stress. The amount of progeny virus in the culture supernatants was determined by the TCID50 method 48 h after virus inoculation. The numbers above the bars indicate the relative viral titers compared to the control. (D) Northern blot analysis of viral RNAs and Western blot analysis of the coat protein, performed 12 h after inoculation with RGNNV. Representative data are shown. Positions of RNA1, RNA2, and RNA3 are indicated at the left. Expression of β-actin was measured as a loading control. (E) Quantitative analysis of the Northern blot and Western blot images shown in panel D. The chemiluminescent data of the blots were quantified using Quantity One software. The relative expression was calculated compared to the control cells after normalization to β-actin expression. Statistical analysis was performed using a two-tailed Student's t-test (*, p < 0.05; **, p < 0.01, compared to the control).

Full size image

We examined the effect of thermal stress on the expression of the major HSP70 isoforms (HSP70-1, HSP70-2, HSP70-5, HSP70-8, and HSP70-9) and all of the HSP90 isoforms (HSP90-α1, HSP90-α2, HSP90-β, GRP94, and TRAP1) of medaka. Among the 10 HSPs tested, HSP70-1, HSP70-2, HSP70-5, HSP90-α1, and HSP90-α2 were strongly upregulated immediately after the thermal stress (Fig 3). The upregulation of HSP70-2 and HSP90-α2 was sustained for at least 48 h after thermal stress, while the expression of HSP70-1 and HSP90-α1 returned to the control level 48 h after the thermal stress. Upregulation of HSP70-5 was observed only immediately after the thermal stress, and a return to the basal level was observed 12 h after the thermal stress.

Fig. 3
figure 3

Effect of thermal stress on the expression of HSP70 and HSP90 isoforms. OLHNI-2 cells were thermally stressed at 42°C for 1 h and allowed to recover at 30°C for 1 h (designated as 0 h post-thermal-stress). Cells continuously incubated at 30°C were used as a control. Total RNA was isolated from cells at 0, 12, and 48 h post-thermal-stress and subjected to quantitative real-time RT-PCR analysis. The relative mRNA expression was calculated against the control after normalization to β-actin expression. Statistical analysis was performed using a two-tailed Student's t-test (*, p < 0.05; **, p < 0.01, compared to the control).

Full size image

Effect of virus infection on the expression of the HSP70 and HSP90 isoforms

We next examined whether virus infection itself induces the expression of HSPs. Among the 10 HSP isoforms tested, HSP70-1, HSP70-2, HSP90-α1, and HSP90-α2 were strongly upregulated by virus infection, and the levels increased with time (Fig. 4). In contrast, no upregulation was observed for HSP70-5, HSP70-9, GRP94, or TRAP1. HSP70-8 and HSP90-β were slightly upregulated at 6 h postinfection (Fig. 4). However, the levels of upregulation induced by virus infection were relatively low compared with those induced by thermal stress (Fig. 3).

Fig. 4
figure 4

Effect of virus infection on the expression of the HSP70 and HSP90 isoforms. OLHNI-2 cells were inoculated with RGNNV (MOI = 1) for 1 h. Cells without virus inoculation were used as a control. Total RNA was isolated from cells at 6, 12, 18, and 24 h post-inoculation and subjected to quantitative real-time RT-PCR analysis. The relative mRNA expression was calculated compared to the control after normalization to β-actin expression. Statistical analysis was performed using a two-tailed Student's t-test (*, p < 0.05; **, p < 0.01, compared to the control).

Full size image

Effects of overexpression of the HSP isoform genes on virus multiplication

In order to identify the HSP isoforms that support virus multiplication, we overexpressed each of the 10 HSP isoforms in medaka cells prior to virus inoculation. HA-tagged HSPs were successfully expressed in cells transfected with each of the expression constructs (Fig. 5A). When compared with the control, the levels of virus multiplication were elevated in cells transfected with the vector expressing HSP70-1, HSP70-2, HSP70-5, HSP90-α1, HSP90-α2, or HSP90-β (Fig. 5B), while the levels of virus multiplication were reduced to less than those of the control when cells were transfected with the vector expressing HSP70-8, HSP70-9, GRP94, or TRAP1. These results suggest that HSP70-1, HSP70-2, HSP70-5, HSP90-α1, HSP90-α2, and HSP90-β serve as positive factors for betanodavirus multiplication. Conversely, HSP70-8, HSP70-9, GRP94, and TRAP1 might function as negative factors.

Fig. 5
figure 5

Effects of overexpression of the HSP isoforms on virus multiplication. OLHNI-2 cells were transfected with expression plasmids encoding each of the HSP70 and HSP90 isoforms. Cells transfected with an expression construct encoding EGFP was used as a control. Cells were then inoculated with RGNNV (MOI = 10) at 48 h post-transfection. (A) Cells expressing each of the HA-tagged HSP70 and HSP90 isoforms, detected by immunocytochemistry at 48 h post-transfection using an anti-HA monoclonal antibody and Alexa Fluor 555 goat anti-mouse IgG. The red fluorescence images of HSP-expressing cells were merged with the blue fluorescence images of DAPI-stained nuclei. Representative data from three independent experiments are shown. The green fluorescence micrograph indicates the expression of EGFP in the control cells. (B) Viral titers in culture supernatants, determined by the TCID50 method 48 h after virus inoculation. The numbers above the bars indicate the relative viral titers compared to the control. Data are shown as mean values of three independent experiments with standard deviation. Statistical analysis was performed with a one-way ANOVA, followed by Tukey’s HSD test (*, p < 0.05; **, p < 0.01, compared to the control).

Full size image

Effects of knockdown of the HSP isoform genes on virus multiplication

Based on the results of the overexpression experiments, we further examined the positive roles of HSP70-1, HSP70-2, HSP70-5, HSP90-α1, HSP90-α2, and HSP90-β in virus multiplication via knockdown experiments. Treatment with siRNAs, targeting HSP70-1, HSP70-2, HSP70-5, HSP90-α1, HSP90-α2, and HSP90-β reduced the expression levels of the corresponding mRNA to 42.1, 46.2, 49.7, 48.5, 42.7, and 38.9% of the control level, respectively (Fig. 6A). When cells transfected with each of the six siRNAs were inoculated with RGNNV, the amount of progeny virus in all of the treatments decreased compared to the control (Fig. 6B).

Fig. 6
figure 6

Effect of knockdown of HSP isoform genes on virus multiplication. OLHNI-2 cells were transfected with siRNAs targeting each of the HSP70 and HSP90 isoforms. Cells transfected with an siRNA targeting EGFP (siREGFP) were used as a control. Cells were then inoculated with RGNNV (MOI = 1) at 48 h post-transfection. (A) Knockdown efficiency of the siRNAs, assessed by the relative mRNA expression of the isoforms. Total RNA was isolated from cells at 48 h post-transfection and subjected to quantitative real-time RT-PCR analysis. The relative mRNA expression was calculated compared to the control after normalization to β-actin expression. (B) Titers of progeny virus in culture supernatants, determined by the TCID50 method 48 h after virus inoculation. The numbers above the bars indicate the relative viral titers compared to the control. Statistical analysis was performed using a two-tailed Student's t-test in panel A, and using a one-way ANOVA followed by Tukey’s HSD test in panel B (*, p < 0.05; **, p < 0.01, compared to the control).

Full size image

Discussion

Non-lethal stressors, including thermal stress, activate the synthesis of HSPs and protect fish from bacterial infections [65, 66]. For example, treatment with the chemical compound TEX-OE and its variant induces HSPs non-lethally in salmon and gilthead sea bream (Sparus aurata), which protects these fish from Vibrio anguillarum infection [66]. Bacterial infections upregulate HSP70 in rainbow trout (Oncorhynchus mykiss) and sea bream (Sparus sarba), and this has been suggested to be a defense response against bacterial pathogens [67,68,69,70,71]. However, these inhibitory functions of HSPs in bacterial infections do not necessarily apply to viral infections. Relatively more fish HSPs have thus far been described as positive HFs for viruses [e.g., 16,17,18,19,20] than as negative factors [21, 22]. As for HSPs involved in betanodavirus infection, HSP70-5 supports RGNNV multiplication in grouper (Epinephelus coioides) GF-1 cells [29], and HSP90-β also serves as a positive HF for the virus in GF-1 cells [72] and in marine medaka (Oryzias melastigma) hMMES1 cells [31], which is consistent with our overexpression and knockdown experiments using OLHNI-2 cells (Figs. 5 and 6). In contrast, HSP27 in sea perch (Lateolabrax japonicus) exerts an anti-RGNNV effect by regulating the apoptosis signaling pathway [21]. In the present study, HSP70-1, HSP70-2, HSP90-α1, and HSP90-α2 as well as HSP70-5 and HSP90-β facilitated RGNNV multiplication in medaka cells (Figs. 5 and 6). In contrast, HSP70-8, HSP70-9, GRP94, and TRAP1 apparently suppressed virus multiplication, although further studies, including knockdown of each of these four genes, are required to confirm this. One curious observation is that overexpression of HSP70-8 suppressed RGNNV multiplication in medaka cells, which apparently contradicts a previous report by Chang and Chi [28]. A possible explanation for the discrepancy is that our data were obtained using OLHNI-2 cells derived from medaka, the host specificity of which is different from that of grouper, which was used by Chang and Chi [28]. Medaka is susceptible to SJNNV as well as RGNNV infection, although the levels of susceptibility to RGNNV are higher than those to SJNNV [48]. Conversely, groupers are susceptible to RGNNV infection but entirely resistant to SJNNV [42, 43]. Chang and Chi [28] have suggested that HSP70-8 functions as an RGNNV receptor or coreceptor, which probably distinguishes the structural difference between SJNNV and RGNNV. Therefore, some protein other than HSP70-8 might function as an RGNNV receptor in medaka. In any event, more studies are required to obtain a complete picture of the role of HSP isoforms in fish virus multiplication. Furthermore, in vivo investigations of the importance of HSP isoforms in fish virus multiplication are required to develop our hypotheses based on this in vitro study.

Among the three inhibitors used in this study, radicicol and 17-AAG inhibit HSP90 activity specifically through the inhibition of ATP binding to HSP90 [73]. Quercetin inhibits the activity of the transcription factor heat shock factor 1, resulting in broad downregulation of HSPs, including HSP70 and HSP90 [74, 75]. In this study, radicicol, 17-AAG, and quercetin had inhibitory effects on the RNA replication and progeny virus production of RGNNV (Fig. 1). These results are consistent with previous studies showing that treatment of marine medaka hMMES1 cells with ganetespib or NVP-AUY922, which are inhibitors of HSP90ab1, significantly decreased RGNNV entry into the cells [31]. Furthermore, geldanamycin (GA, a prototype of 17-AAG) and radicicol negatively affected the replicative and multiplicative competence of Flock House virus (FHV), which, like RGNNV, is a member of the family Nodaviridae [64]. Similar negative effects of HSP inhibitors have been reported for other viruses, including hepatitis viruses [76,77,78,79], influenza virus [80], Ebola virus [81], and herpes simplex virus [51]. However, in contrast, treatment of GF-1 cells with GA has been shown to facilitate the multiplication of RGNNV [72]. One possible reason for these contradictory findings is the difference in the dose of inhibitors used. While we treated cells with 5 µM 17-AAG, Chen et al. [72] used a relatively low concentration of GA (1.5 µM). Since inhibition of HSP90 by GA is known to trigger upregulation of HSP90 [82], the concentration of GA used by Chen et al. [72] might be sufficient to upregulate HSP90 but not to inhibit HSP90 activity. Alternatively, treatment of GF-1 cells with GA could have unexpected effects on virus multiplication. Although the cytotoxic effect of GA on GF-1 cells was not tested by Chen et al. [72], we found that GA had a harmful effect on medaka cells within 12 h at concentrations as low as 1 µM, which prompted us to use 17-AAG instead of GA in this study.

In this study, overexpression and knockdown experiments on the HSP70 and HSP90 isoforms revealed that HSP70-1, HSP70-2, HSP70-5, Hso90-α1, Hso90-α2, and HSP90-β facilitated virus multiplication. Overexpression of these six isoforms increased the levels of viral multiplication from 1.69 to 3.63 times more than that of the control (Fig. 5B). However, this level of increase was less than that observed in a previous study [29]. One possible explanation for this discrepancy is that the expression levels of the exogenous HSP genes were not high enough to allow a strong increase in viral titer in this study because of low transfection efficiency (Fig. 5A). Similarly, knockdown of the endogenous genes encoding the six HSP isoforms decreased the levels of viral multiplication to 0.26 to 0.55 times that of the control (Fig. 6B), which is less efficient than what has been reported previously by Su et al. [29]. Since the knockdown levels of the isoform mRNAs obtained in this study were around 40% of the control (Fig. 6A), it is to be expected that more-efficient knockdown of the HSPs should impair virus multiplication more efficiently.

We found that HSP70-1, HSP70-2, HSP70-5, HSP90-α1, HSP90-α2, and HSP90-β play positive roles in betanodavirus multiplication. Surprisingly, all of these but HSP70-5 were upregulated by virus infection, whereas the other HSP isoforms tested were not, even though all 10 HSP isoforms are known to be upregulated by various physiological or pathological stimuli in humans [26, 61, 82]. Su et al. [29] reported that upregulation of HSP70-5 occurred in GF-1 cells 48 and 72 h after inoculation with RGNNV (MOI = 5). We could not examine the expression of HSP70-5 in medaka cells at 48 h post-inoculation because almost all of the cells were disrupted by a cytopathic effect at this time point (data not shown). HSP90-β in GF-1 cells was upregulated by inoculation with RGNNV (MOI = 0.1) as early as 6 h post-inoculation [72], which is consistent with our data showing that HSP90-β was upregulated in medaka cells 6 h after RGNNV inoculation (Fig. 4). Collectively, these results imply that RGNNV actively induces those six HSP isoforms to facilitate its multiplication. Viral manipulation of host cells via alteration of gene expression has been reported for various viruses [83,84,85,86]. Nevertheless, we still cannot rule out the possibility that induction of some or all of the virologically important HSP isoforms results from the general host stress response to virus infection and is incidentally utilized for virus multiplication.