Sulfolobus islandicus Employs Orc1-2-Mediated DNA Damage Response in Defense against Infection by SSV2

ABSTRACT

All living organisms have evolved DNA damage response (DDR) strategies in coping with threats to the integrity of their genome. In response to DNA damage, Sulfolobus islandicus activates its DDR network in which Orc1-2, an ortholog of the archaeal Orc1/Cdc6 superfamily proteins, plays a central regulatory role. Here, we show that pretreatment with UV irradiation reduced virus genome replication in S. islandicus infected with the fusellovirus SSV2. Like treatment with UV or the DNA-damaging agent 4-nitroquinoline-1-oxide (NQO), infection with SSV2 facilitated the expression of orc1-2 and significantly raised the cellular level of Orc1-2. The inhibitory effect of UV irradiation on the virus DNA level was no longer apparent in the infected culture of an S. islandicus orc1-2 deletion mutant strain. On the other hand, the overexpression of orc1-2 decreased virus genomic DNA by ~10²-fold compared to that in the parent strain. Furthermore, as part of the Orc1-2-mediated DDR response genes for homologous recombination repair (HRR), cell aggregation and intercellular DNA transfer were upregulated, whereas genes for cell division were downregulated. However, the HRR pathway remained functional in host inhibition of SSV2 genome replication in the absence of UpsA, a subunit of pili essential for intercellular DNA transfer. In agreement with this finding, lack of the general transcriptional activator TFB3, which controls the expression of the ups genes, only moderately affected SSV2 genome replication. Our results demonstrate that infection of S. islandicus by SSV2 triggers the host DDR pathway that, in return, suppresses virus genome replication.

IMPORTANCE Extremophiles thrive in harsh habitats and thus often face a daunting challenge to the integrity of their genome. How these organisms respond to virus infection when their genome is damaged remains unclear. We found that the thermophilic archaeon Sulfolobus islandicus became more inhibitory to genome replication of the virus SSV2 after preinfection UV irradiation than without the pretreatment. On the other hand, like treatment with UV or other DNA-damaging agents, infection of S. islandicus by SSV2 triggers the activation of Orc1-2-mediated DNA damage response, including the activation of homologous recombination repair, cell aggregation and DNA import, and the repression of cell division. The inhibitory effect of pretreatment with UV irradiation on SSV2 genome replication was no longer observed in an S. islandicus mutant lacking Orc1-2. Our results suggest that DNA damage response is employed by S. islandicus as a strategy to defend against virus infection.

INTRODUCTION

All organisms have evolved DNA damage response (DDR) strategies to cope with constant endogenous and exogenous insults to their genome. Eukaryotic DDR and bacterial DDR, known as the SOS response, have been extensively investigated, and they both entail damage detection, control of transcriptional activities, activation of DNA repair enzymes, and repression of DNA replication and cell division (1, 2). Although archaeal DDR is less well understood, recent studies have provided convincing evidence indicating that this regulatory pathway functions in archaea (3, 4). Virus infection has been shown to activate DDR, which either inhibits or facilitates virus replication in eukaryotes (5). The bacterial SOS response is known to inactivate the repressor of a temperate bacteriophage as a result of DNA damage, but the potential impact of the lytic infection of a bacteriophage on the bacterial SOS response has seldomly been examined (6). The relationship between virus infection and DDR in archaea is far less clear.

Archaea are often found to flourish in extreme habitats of high temperature, salinity, or high or low pH and, therefore, under considerable stress. Among them, thermoacidophilic archaea of the family Sulfolobaceae grow in acidic thermal habitats under aerobic conditions and are exposed to sunlight (7). Conceivably, the genomic DNA of these organisms would experience increased attacks from spontaneous hydrolysis, reactive oxygen species (ROS), UV irradiation, and other DNA-damaging agents in their native habitats (8–10). The transcriptional response of Saccharolobus or Sulfolobus species, members of Sulfolobaceae, to UV irradiation has been investigated, with differences between species or strains of the same species observed in the change of the expression of DNA repair genes following UV treatment (10–13). In a series of studies focused on Saccharolobus solfataricus, the UV response of the organism was found to involve the upregulation of up to 55 genes (11), including orc1-2, which encodes the archaeal-origin recognition complex 1/cell division control protein 6 (Orc1/Cdc6) superfamily (3), mre11 and rad50, both of which code for functions in homologous recombination repair (HRR) (14). In addition, a strong transcriptional induction of genes responsible for pilus formation, and thus, cell aggregation was also observed, suggesting an increased exchange of DNA through conjugation (11). It was subsequently confirmed that the pili biogenesis operon, termed the UV-inducible pili formation (ups) operon, mediated UV-inducible cell aggregation, which allowed chromosomal marker exchange with high frequency in S. solfataricus (15, 16). The crenarchaeal system for exchange of DNA (Ced) was involved in the intercellular DNA transfer (17). Furthermore, the pilus knockout mutants of Sulfolobus acidocaldarius showed decreased survival upon UV treatment (18). Therefore, the UV-inducible DNA transfer process and subsequent homologous recombination appear to represent an important DDR mechanism employed by these hyperthermophilic organisms to maintain their genome integrity.

More recently, Orc1-2 was found to function as a master regulator in Sulfolobus islandicus DDR (3). It activates genes encoding HRR (e.g., radA, mre11, and rad50) (19), UV-inducible pili formation (ups) (15, 16), the ced system (17), the general transcription factor TFB3, etc. (4, 20). Expression of the ups operon and ced system, under the dual control of Orc1-2 and TFB3, permits cell aggregation and subsequent intercellular DNA transfer for HRR (3, 21). Orc1-2 also represses genes for cell division (e.g., cdvA, cdvB or escort-III, and cdvC), DNA replication initiation (e.g., oriC1 and oriC3), and genome segregation (e.g., parA and parB) (3). However, there are very few studies on DDR regulation following virus infection in archaea. Transcriptomic analyses revealed no significant changes in the expression of orc1-2 and the other DDR-related genes in S. solfataricus following infection by the fusellovirus SSV2 (22). The DDR genes, except for the cdv operon, were not differentially expressed either in S. islandicus LAL14/1 upon infection by S. islandicus rod-shaped virus 2 (SIRV2) (23). Therefore, it remains to be determined if and how DDR pathways would respond to virus infection in S. islandicus.

In this study, we investigated the interaction between virus infection and host DDR, using SSV2 and S. islandicus strain E233S as the virus-host system. We found that preinfection, UV irradiation decreased the host’s susceptibility to SSV2 infection. Virus infection activated the host DDR, which, in turn, reduced virus production.

RESULTS

SSV2 genome replication was reduced in S. islandicus treated with preinfection UV irradiation.

We started by testing the susceptibility of UV-irradiated S. islandicus cells to infection by SSV2. A culture of S. islandicus strain E233S in the early log phase was irradiated with UV light (254 nm) at a dose of 30 J/m², and incubation was continued for 1 h at 75°C with shaking in SCVYU medium (Fig. 1A). The culture was then infected with SSV2 at a multiplicity of infection (MOI) of ~5. Moderate growth retardation was observed in the infected culture compared to the uninfected control (Fig. 1B). The virus genome concentrations (VGCs) in the infected culture were determined at intervals by quantitative PCR (qPCR) (Fig. 1C). As shown in Fig. 1C, the virus yield was significantly lower in the UV-irradiated culture than in the unirradiated control, and the two cultures differed in VGC by 18.2 (±0.5)-fold 24 h postinfection (hpi). This observation was also supported by virion counting under an electron microscope (Fig. 1D). The drastic decrease in VGC in the UV-irradiated culture cannot be traced back to UV-induced killing of the cells, as a survival rate of 60% (±1.2%) was recorded for the organism at the dose of UV light used in this study (see Fig. S1 in the supplemental material). In addition, the virus infected both the UV-irradiated and unirradiated host cells at similar rates (Fig. 1E). Therefore, the SSV2 genome appeared to be replicated less efficiently in UV-irradiated S. islandicus cells than in unirradiated control cells.

FIG 1.

Infection of UV-irradiated S. islandicus E233S by SSV2. (A) Sketch showing the experimental procedure of the infection of UV-irradiated and unirradiated S. islandicus cells with SSV2. (B) Effect of UV irradiation, SSV2 infection, and sequential treatment with UV irradiation and SSV2 infection on the growth of S. islandicus E233S. S. islandicus E233S was grown to the early exponential phase (OD₆₀₀, ~0.2) in SCVYU medium. The culture was irradiated with UV light, infected with SSV2 at an MOI of ~5, or first irradiated with UV and subsequently infected with SSV2 at an MOI of ~5. Incubation was continued, and OD₆₀₀ measurements were made in triplicate at the indicated times. (C) Effect of preinfection UV irradiation on the genome copy number of SSV2 in infected S. islandicus E233S cells. S. islandicus E233S was grown to the early exponential phase in SCVYU medium and irradiated with UV or left untreated. The UV-treated and untreated cells were infected with SSV2 at an MOI of ~5. Samples were collected at the indicated times. The measurements were made in triplicate. Significance test was performed by Student's t test. NS, P > 0.05; *, P < 0.05; **, P < 0.01; ***, P < 0.001. (D) Quantification of SSV2 particles in UV-irradiated and untreated S. islandicus E233S cultures by TEM. The S. islandicus E233S culture was grown to the early exponential phase in SCVYU medium, irradiated with UV or left untreated, and incubated for 1 h. Both the UV-irradiated and untreated cultures were then infected with SSV2 at an MOI of ~5. The infected cultures were further incubated for 12 h at 75°C. Samples of the cultures were negatively stained and viewed at ×8,000 magnification (TEM), and virus particles were counted in >10 fields from multiple grids. The significance test was performed using Student's t test. (E) Infection rate assays. S. islandicus E233S was grown to the early exponential phase in SCVYU medium and treated with UV or left untreated. After incubation with shaking for 1 h at 75°C, the cultures were infected with SSV2 at an MOI of ~5. Samples were taken at 0, 0.5, 1, 2, and 3 hpi. Each sample was diluted 10³- and 10⁴-fold. Dilutions were spread on SCVYU solid medium and incubated at 75°C for 7 days. At least 30 colonies were picked from one plate (either the 10³- or 10⁴-fold dilution) for each time point for PCR analysis using primers targeting the SSV2 vp1 gene. The reaction mixtures were subjected to agarose gel electrophoresis. Amplification of a sequence of the expected size was taken to indicate successful infection. (F) Cellular levels of Orc1-2 in S. islandicus upon UV treatment. S. islandicus E233S was grown to the early exponential phase in SCVYU medium and treated with UV or left untreated. Samples were taken at 0, 1, and 2 h after further incubation and subjected to analysis by immunoblotting with anti-Orc1-2 antibody. PCNA was used as a loading control. The ratios of the blot intensity of Orc1-2 in the UV-irradiated cells to those in the untreated cells at various time points, corrected for the loading control, are shown at the bottom.

DDR-related genes were upregulated in S. islandicus infected by SSV2.

UV irradiation triggers DNA damage response (DDR) in cells. Orc1-2 was shown to be a key upstream regulator in DNA damage response in S. islandicus (3). Indeed, we found that the cellular level of Orc1-2 increased significantly in the S. islandicus E233S culture 2 h following UV treatment, as revealed by immunoblotting (Fig. 1F). We then sought to determine if SSV2 infection would induce the DDR pathway in the organism. As shown in Fig. 2A, S. islandicus E233S infected with SSV2 at the MOI of ~5 grew more slowly than the uninfected control, and the VGC increased by over 10⁴-fold in the infected culture by 36 hpi (Fig. 2B). We then quantified the expression of genes encoding proteins involved in DDR by reverse transcription-quantitative PCR (qRT-PCR). Among the DDR-related genes tested were those encoding functions in (i) HRR (mre11, rad50, radA, and herA) (24, 25), (ii) intercellular DNA transfer (cedA) (17), (iii) UV-inducible pili formation (upsA) (15, 16), (iv) general transcription activation (tfb3) (4, 21), and (v) cell division (cdvA and cdvB) (26–28). We found that orc1-2, mre11, rad50, radA, tfb3, cedA, upsA, and herA were upregulated by 3.6 (±0.4)-, 1.8 (±0.3)-, 2.0 (±0.2)-, 2.8 (±0.4)-, 1.9 (±0.5)-, 15.4 (±0.6)-, 3.5 (±0.1)-, and 4.6 (±0.5)-fold, respectively, whereas cdvA and cdvB were downregulated by 3.5 (±0.1)- and 4.4 (±0.2)-fold, respectively, in the infected cells at 12 hpi compared to those in the uninfected cells (Fig. 2C). We also measured the protein level of Orc1-2 in the infected cells at various times. The protein was synthesized to higher cellular levels (e.g., a 2.8-fold increase at 48 hpi) for up to 48 h (the duration of the experiment) in the infected cells than in the uninfected cells (Fig. 2D). Interestingly, as observed with S. islandicus cells treated with DNA-damaging agents (3), S. islandicus E233S cells tended to aggregate upon infection by SSV2, and more than 95% of the cells formed aggregates 12 hpi, as revealed by confocal laser scanning microscopy (Fig. 2E and F). In comparison, fewer than 5% of the cells of an S. islandicus mutant strain defective in UpsA (ΔupsA), which is essential for cell aggregation (16), aggregated following SSV2 infection (Fig. 2E and F). Therefore, our results suggest that SSV2 infection was similar to treatment with DNA-damaging agents (e.g., NQO) in activating the DDR pathway in S. islandicus.

FIG 2.

Effect of SSV2 infection on S. islandicus E233S. (A) Growth curves of S. islandicus E233S with or without SSV2 infection. S. islandicus E233S was grown to the early exponential phase in SCVYU medium and infected with SSV2 at an MOI of ~5. All OD measurements were made in triplicate. (B) Time course of SSV2 genome replication in infected S. islandicus E233S. S. islandicus E233S was grown to the early exponential phase in SCVYU medium and infected with SSV2 at an MOI of ~5. Samples were taken at the indicated time points, and the VGCs were determined by qPCR. All data points were determined in triplicate. (C) Transcription levels of DDR-related genes in S. islandicus E233S infected with SSV2. S. islandicus E233S was grown to the early exponential phase in SCVYU medium and infected with SSV2 at an MOI of ~5. Samples were taken at 12 hpi, and RNAs were extracted. The transcription products of the specified genes were quantified by qRT-PCR. The 16S rRNA gene from S. islandicus E233S was used as an internal control. (D) Cellular levels of Orc1-2 in S. islandicus E233S at various time points following SSV2 infection. Cells were grown in the early exponential phase in SCVYU medium and infected with SSV2 at an MOI of ~5. Samples were collected from the SSV2-infected culture and an uninfected control culture at the indicated time points and subjected to immunoblotting with anti-Orc1-2 antibody. PCNA was used as a loading control. The ratios of the blot intensity of Orc1-2 in the SSV2-infected cells to those in the uninfected cells at various time points, corrected for the loading control, are shown at the bottom. (E) Aggregation of S. islandicus E233S cells following SSV2 infection. S. islandicus E233S was grown to the early exponential phase in SCVYU medium and infected with SSV2 at an MOI of ~5. Samples collected at 12 hpi were observed under a confocal laser scanning microscope. (F) Quantification of cells in aggregates after infection. At least 300 cells from ≥10 fields from the grids obtained in panel E were counted for each sample. A cluster of >3 cells was considered a cell aggregate. Significance test was performed by Student’s t test. ***, P < 0.001.

Orc1-2 played a vital role in the host defense of S. islandicus against infection by SSV2.

The activation of DDR by SSV2 infection suggests that the host might use the DNA damage response system in its defense against the virus infection. To test this possibility, we infected an orc1-2 deletion mutant of S. islandicus E233S (Δorc1-2), which was either preirradiated with UV or unirradiated, with SSV2. Δorc1-2 displayed a survival rate of 33% following UV irradiation at the dose used in the experiment (Fig. S1). Interestingly, the VGCs in the infected culture were similar whether it was pretreated with UV irradiation or untreated (Fig. 3A). When a plasmid overproducing Orc1-2 under the control of an arabinose-inducible promoter (pSeSD-orc1-2) was introduced into Δorc1-2, the resulting complementary strain Δorc1-2/pSeSD-orc1-2 supported less SSV2 genome replication than Δorc1-2 following preinfection UV irradiation. A 10.5 (±0.6)-fold difference between the two strains in VGC was observed 24 h after UV treatment. As expected, Orc1-2 increased in both S. islandicus E233S and Δorc1-2/pSeSD-orc1-2 in response to UV treatment but was not detectable in Δorc1-2 (Fig. 3C). To further understand the role of Orc1-2 in host defense against virus infection, we constructed E233S/pSeSD-orc1-2, an S. islandicus strain overproducing Orc1-2. The Orc1-2 content in E233S/pSeSD-orc1-2 was considerably higher than that in E233S/pSeSD under the inducing conditions (Fig. S2). When the two strains were infected with SSV2, the VGC in E233S/pSeSD-orc1-2 was ~10²-fold lower than that in E233S/pSeSD 24 hpi (Fig. 3B). Our results indicate that Orc1-2 served a critical role in the host defense of S. islandicus against infection by SSV2.

FIG 3.

Effect of UV irradiation on SSV2 genome replication in the Orc1-2-deficient strain Δorc1-2. (A) Effect of UV irradiation on SSV2 genome replication in S. islandicus E233S, Δorc1-2, and Δorc1-2/pSeSD-orc1-2. S. islandicus E233S and Δorc1-2 were grown in SCVYU medium, whereas Δorc1-2/pSeSD-orc1-2 was grown in ACV medium for the induction of the expression of orc1-2. All strains were grown to the early exponential phase and irradiated with UV. Following further incubation for 1 h, the UV-treated and untreated cultures were infected with SSV2 at an MOI of ~5. Samples were collected at 24 hpi, and the VGCs were determined by qPCR. All data points were determined in triplicate. (B) Time course of the virus genome replication in the Orc1-2-overproducing strain E233S/pSeSD-orc1-2. Cells were grown to the early exponential phase in ACV medium and infected with SSV2 at an MOI of ~5. Samples were taken at the indicated times, and the VGC in each sample was determined by qPCR in triplicate. (C) Cellular levels of Orc1-2 in S. islandicus E233S and Δorc1-2/pSeSD-orc1-2. Cultures were grown to the early exponential phase in SCVYU medium and irradiated with UV. After further incubation for 2 h, when Orc1-2 was most significantly overproduced at this time compared to that at other tested time points, samples were taken and subjected to immunoblotting with anti-Orc1-2 antibody. Δorc1-2 was used as a control (left). The Δorc1-2/pSeSD-orc1-2 sample was loaded in two different amounts. The ratios of the blot intensity of Orc1-2 in the UV-irradiated cells to that in the untreated cells, corrected for the loading control, are shown at the bottom. ***, P < 0.001.

The S. islandicus HRR pathway functioned in host defense against infection by SSV2.

To learn if synthesis of any of the DDR genes controlled by Orc1-2 would restore the ability of Δorc1-2 to defend against SSV2 infection, we introduced plasmid pSeSD encoding His-tagged Mre11, Rad50, RadA, CedA, UpsA, HerA, CdvA, or CdvB into the deletion mutant, yielding strains denoted Δorc1-2/pSeSD-mre11, Δorc1-2/pSeSD-rad50, Δorc1-2/pSeSd-radA, Δorc1-2/pSeSD-cedA, Δorc1-2/pSeSD-upsA, Δorc1-2/pSeSD-herA, Δorc1-2/pSeSD-cdvA, and Δorc1-2/pSeSD-cdvB, respectively. All of the strains were able to produce the desired proteins, as revealed by immunoblotting (Fig. S3). We then infected these strains, as well as Δorc1-2/pSeSD, with SSV2. As shown in Fig. 4, the VGCs in Δorc1-2/pSeSD-mre11, Δorc1-2/pSeSD-rad50, Δorc1-2/pSeSD-radA, Δorc1-2/pSeSD-cedA, and Δorc1-2/pSeSD-upsA were ~10- to 10²-fold lower than those in Δorc1-2/pSeSD 24 hpi, indicating that overproduction of any of the above-described proteins compensated for the defect of Δorc1-2 in defense against SSV2 infection. In comparison, Δorc1-2/pSeSD-herA, which overproduced HerA, a helicase involved in DNA end resection in DNA repair (24, 25), appeared to be an exception, as it showed no measurable differences from E233S/pSeSD in VGC. On the other hand, the VGCs in Δorc1-2/pSeSD-cdvA and Δorc1-2/pSeSD-cdvB were 14.5 (±0.8)- and 13.7 (±0.7)-fold higher than those in Δorc1-2/pSeSD, suggesting that overproduction of the proteins involved in host cell division facilitated virus production (Fig. 4). Since some of the tested proteins, such as Mre11, Rad50, CedA, UpsA, and CdvB, function in complex or association with other proteins in DDR (16, 17, 19, 26–29), it was somewhat unexpected that overproduction of these proteins individually was at least partially effective in restoring the ability of Δorc1-2 to defend against SSV2. We then explored if it was possible that overproduction of one of these proteins led to the overproduction of other proteins with which the protein forms a complex. However, this did not appear to be the case since the cellular level of neither Mre11 nor RadA increased in E233S/pSeSD-rad50, as measured by immunoblotting (data not shown). While the compensation of overproducing Mre11, Rad50, CedA, UpsA, or CdvB for the defect of Δorc1-2 in inhibiting SSV2 genome replication remains to be understood mechanistically, a possibility exists that none of these proteins were present at cellular levels required for the optimal function of the processes in which they are involved. Based on our results, we conclude that the HRR pathway plays a role in the defense of S. islandicus against infection by SSV2.

FIG 4.

Effect of the overexpression of the DDR-related genes on SSV2 genome replication in the Orc1-2-deficient S. islandicus strain Δorc1-2. Strains were grown to the early exponential phase in SCVYU medium and infected with SSV2 at an MOI of ~5. Samples were taken at 24 hpi, and the VGC in each sample was determined by qRT-PCR in triplicate.

S. islandicus was able to use HRR to inhibit SSV2 genome replication in the absence of UpsA.

Among the tested DDR-related genes, upsA is part of the ups operon which encodes Ups pili required for the initiation of cell aggregate formation and subsequent exchange of genomic DNA for the repair of DNA double-stranded breaks (15, 16). We would therefore like to learn the influence of the ups system on SSV2 genome replication in S. islandicus. We infected the S. islandicus upsA deletion strain ΔupsA and parent strain E233S with SSV2 and found that the VGC was 10.2 (±0.2)-fold higher in the mutant strain than in the parent strain 24 hpi, in agreement with the observation above that UpsA was involved in host cell defense against infection by the virus (Fig. 5A). In comparison, the VGCs in pre-UV-irradiated ΔupsA and parent strains were similar and ~10- to 20-fold lower than that in the unirradiated parent strain (Fig. 5A). The cellular levels of Orc1-2 increased substantially in both the mutant and parent strains following UV irradiation (Fig. 5B). Therefore, it appears that the Ups pili were involved in the defense of S. islandicus against SSV2 infection. Once irradiated with UV, the organism mobilized greater defense against the virus even in the absence of the intercellular transfer of DNA.

FIG 5.

Effect of the overexpression of the DDR-related genes on SSV2 genome replication in the UpsA-deficient S. islandicus strain ΔupsA. (A) SSV2 genome replication in ΔupsA. S. islandicus E233S and ΔupsA, grown to the early exponential phase in SCVYU medium, were irradiated with UV and incubated for 1 h. Both the UV-treated and untreated cells were then infected with SSV2 at an MOI of ~5 and grown at 75°C. Samples were collected at 24 hpi, and the VGC in each sample was determined by qPCR in triplicate. (B) Detection of Orc1-2 in S. islandicus E233S and ΔupsA. Cells were grown to the early exponential phase in SCVYU medium and irradiated with UV. Samples were taken 2 h after UV irradiation and subjected to immunoblotting with anti-Orc1-2 antibody. PCNA was used as a loading control. The ratios of the blot intensity of Orc1-2 in the UV-irradiated cells to those in the untreated cells, corrected for the loading control, are shown at the bottom. (C) SSV2 genome replication in ΔupsA overproducing DDR-related proteins. Strains were grown to the early exponential growth phase in ACV medium and then infected with SSV2 at an MOI of ~5. Samples were taken at 24 hpi, and the VGC in each sample was determined by qPCR in triplicate. ***, P < 0.001.

To shed more light on the inhibition of SSV2 genome replication by host DNA repair processes in the absence of UpsA, we prepared ΔupsA strains overproducing each of the following proteins, UpsA, Mre11, Rad50, RadA, and CedB, and infected the resulting strains, i.e., ΔupsA/pSeSD-upsA, ΔupsA/pSeSD-mre11, ΔupsA/pSeSD-rad50, ΔupsA/pSeSD-radA, and ΔupsA/pSeSD-cedB, as well as ΔupsA/pSeSD, with SSV2. We found that the VGCs were 9.3 (±0.5)-, 17.1 (±0.8)-, 28.6 (±1.9)-, and 17.4 (±0.7)-fold lower in strains overproducing UpsA, Mre11, Rad50, and RadA, respectively, than that in ΔupsA/pSeSD (Fig. 5C). Therefore, inhibition of SSV2 production by HRR occurred in the absence of UpsA. As predicted from the dependence of the intercellular transfer of DNA on Ups pili, the VGC in ΔupsA/pSeSD-cedB was similar to that in ΔupsA/pSeSD (Fig. S4).

SSV2 genome replication was moderately reduced in S. islandicus lacking TFB3.

It was recently shown that the general transcriptional activator TFB3 served to activate the expression of some of the DDR-related genes, including the ced and ups genes (4, 21). To determine if TFB3 played a role in the defense of S. islandicus against SSV2 infection, we constructed two S. islandicus derivative strains, i.e., a tfb3 deletion mutant strain (Δtfb3) and a TFB3-overproducing strain (E233S/pSeSD-tfb3) (Fig. S5), infected them with SSV2, and monitored the VGC at intervals during postinfection incubation. We found that virus yields were somewhat lower in Δtfb3 than in S. islandicus E233S and higher in E233S/pSeSD-tfb3 than in E233S/pSeSD (Fig. 6). For example, the VGC was 3.5 (±0.3)-fold lower in the mutant than in the parent strain and 3.8 (±0.3)-fold higher in the TFB3 overproducer than in the control strain at 24 hpi. Therefore, we conclude that TFB3 did not contribute to the inhibition of virus infection by the host and instead moderately increased the virus yield in the infected culture.

FIG 6.

Effect of TFB3 on SSV2 genome replication in S. islandicus E233S. (A) SSV2 genome replication in the TFB3-deficient S. islandicus strain Δtfb3. Cells were grown at 75°C in SCVYU medium and infected with SSV2 at an MOI of ~5. Samples were taken at the indicated time points, and the VGC in each sample was determined by qPCR in triplicate. The significance test was performed by Student’s t test. **, P < 0.01. (B) SSV2 genome replication in the TFB3-overproducing S. islandicus strain E233S/pSeSD-tfb3. Cells were grown at 75°C in ACV medium and infected with SSV2 at an MOI of ~5. Samples were taken at the indicated time points, and the VGC in each sample was determined by qPCR in triplicate.

DISCUSSION

S. islandicus thrives in harsh habitats and presumably relies on DDR to maintain the integrity of its genome (10). It is of interest to understand how the DNA-damaging agents would affect host susceptibility to virus infection. In this report, we show that UV-irradiated S. islandicus became significantly more inhibitory to the replication of the SSV2 genome than the unirradiated control strain. The inhibition appears to have resulted from the activation of Orc1-2, the master regulator of the DDR regulation pathways in S. islandicus (3), because the yields of the progeny virions from an infected orc1-2 deletion strain were similar whether it was irradiated with UV or not. To our knowledge, this represents the first report of the inhibition of virus production by DDR in archaea.

It is worth noting that Saccharolobus infected with SSV1, a fusellovirus closely related to SSV2 (30), releases a large number of virus particles following irradiation with UV light or treatment with mitomycin C (31, 32). As the only member of the family Fuselloviridae with a UV-inducible life cycle, SSV1 encodes F55, a transcriptional factor, which regulates the expression of the UV-inducible, as well as the early, genes and serves a key role in the control of the transition from the lysogenic to the induced state of SSV1 (33, 34). Therefore, it is possible that the production of SSV1 particles following UV treatment of the infected cells is the combined result of the opposing effects of the virus-encoded and host-encoded UV responses.

We showed that, like treatment with DNA-damaging agents (e.g., NQO), infection with SSV2 elicited Orc1-2-mediated DDR, including activation of genes for HRR (mre11, rad50, radA), cell aggregation (upsA) and DNA import (cedA), and repression of genes for cell division (cdvA and cdvB) (3). Transcriptional analyses of the response of Sulfolobus and Saccharolobus species to infection by several viruses have been reported (22, 23, 35–37). Checkpoint-like control was found in host response to virus infection in several studies. Following infection by SIRV2, the complete cdv operon was downregulated 10-fold, and the three cdvB paralogs were downregulated 3- to 10-fold in infected S. islandicus LAL14/1 cells (23). The decrease in the expression of the cdv genes, which was likely part of the checkpoint-like control activated by DNA degradation induced by virus infection, presumably resulted in the inhibition of cell division (27). Similarly, cell cycle-related genes (e.g., cdvA, cdvB, and cdvC) were significantly downregulated by 4.85-, 3.18-, and 3.05-fold, respectively, in S. islandicus upon infection with the archaeal enveloped virus STSV2 (37). Reduced expression of the cell cycle-related genes allows more time for DNA repair before cell division. In addition, the endosomal sorting complex required for transport (ESCRT) system is known to function in the budding of a variety of eukaryotic enveloped viruses, such as HIV-1 and Ebola virus (38, 39). CdvB3, an ESCRT-III homolog, was also reported to be involved in the budding of STSV2 (28). SSV1 was shown to exit the host cells through budding (40). Therefore, it will be of interest to determine if the Cdv proteins are also involved in the budding of SSV2.

Intriguingly, however, changes in the expression of orc1-2 or HRR genes were rarely detected in archaeal host cells following virus infection (22, 23, 35–37, 41, 42). Among the many reported transcriptional analyses, infection of S. islandicus with STSV2 was a rare example in which an increase in the expression of DNA repair genes (e.g., mre11, rad50, radA, herA, and NurA) by 2.5- to 6.3-fold at the maximum during the infection period was detected (37). We speculate that the extent of DNA damage inflicted during virus infection determines how the host would respond. When the damage is beyond repair by the host and would lead to the programmed cell death or otherwise be readily repaired without the need for the mobilization of excess repair capacity, the Orc1-2-mediated repair network might not be measurably activated. In addition, the failure to detect the Orc1-2-mediated response might also be attributed to experimental design. For example, in our previous transcriptomic analysis of the infection of S. solfataricus P2 by SSV2, little overexpression of orc1-2 or other downstream DDR-related genes was observed (22). The MOI for the infection experiments in the previous study was considerably lower than that in this study (i.e., 0.5 versus 5). This is presumably consistent with the observation that growth retardation of the host as a result of SSV2 infection was not apparent in the previous study compared to that in the present study. Therefore, the experimental design in the current study may allow for better detection of the expression of the DDR-related genes in response to virus infection.

Orc1-2 regulates the expression of a large number of DDR-related genes, as demonstrated previously for S. islandicus treated with the DNA-damaging agent NQO (3). To shed light on processes in DDR critical to the susceptibility of S. islandicus to SSV2 infection, we tested the roles of individual DDR proteins in compensating for the deletion of orc1-2 in host response to infection by the virus. Our results suggest that HRR was a key DDR process involved in host resistance to SSV2 infection since overproduction of Mre11, RadA, and Rad50 markedly decreased the virus yield in the orc1-2 deletion mutant strain. Rad50 and Mre11 are known to function in DNA damage signal transduction in eukaryotes (43, 44). Archaeal RadA, Rad50, and Mre11 are likely involved in a similar fashion in signal transduction in archaea (45, 46), which could also explain why the archaeal HerA, and probably NurA as well, did not have a role in SSV2 inhibition. The Ups and Ced systems, both of which are required for intercellular DNA transfer, have been proposed to be involved in DNA repair in Sulfolobales (15–17). It was observed in this study that, upon SSV2 infection, S. islandicus cells formed aggregates in which intercellular DNA transfer would presumably occur. Deletion of upsA, which encodes UV-inducible pili required for cell aggregation (15, 16), resulted in an increase in virus genome replication in the infected host. Notably, however, the HRR pathway appeared to be functional in SSV2 inhibition in the absence of intercellular DNA transfer since the significant inhibition of the SSV2 genome replication by HRR remained in host cells defective in the ups systems. This is also consistent with the moderate effect of the deletion of tfb3, which activates the expression of the ced and ups genes (4, 21), on the host defense against SSV2.

The mechanistic aspect of Orc1-2-mediated DDR, especially HRR, in inhibiting virus genome replication in S. islandicus and, possibly, in other members of Sulfolobales as well remains to be understood. A possibility exists that the DDR network serves to maintain the genome integrity and, thus, the cellular viability of the infected host, allowing it to defend against the infecting virus using a plethora of its antivirus mechanisms, such as CRISPR-Cas, restriction-modification, and toxin-antitoxin systems.

MATERIALS AND METHODS

Strains and growth conditions.

Strains and plasmids used in this study are listed in Table S1 in the supplemental material. S. islandicus strains were grown with shaking at 75°C in SCV medium (3) or SCVYU medium (34). S. islandicus strains containing plasmid pSeSD for the overexpression of an indicated gene under the control of the arabinose promoter were grown in ACV medium (3).

Construction of overproducer strains.

To overproduce a target gene in S. islandicus E233S (47), a DNA sequence containing the gene was amplified by PCR using the genomic DNA from S. islandicus E233S as the template and a pair of the corresponding primers (Table S2). The PCR product was digested with SalI and NotI and purified with the V-ELUTE Gel Mini Purification Kit (Zomanbio). The purified DNA fragment was inserted into plasmid pSeSD at the SalI and NotI sites, yielding an overexpression plasmid. The plasmid was transformed by electroporation into S. islandicus E233S cells (48).

UV irradiation.

S. islandicus E233S and its derivative strains were grown with shaking in SCVYU medium to the early exponential phase (optical density at 600 nm [OD₆₀₀], ~0.2 to 0.3), and a sample (10 mL) of the culture was transferred to a sterile petri dish (15 cm in diameter). The petri dish was placed under a UV illuminator (254 nm; GS gene linker UV chamber; Bio-Rad) and irradiated with UV light at 30 J/m². The illumination time is about 3 s. The treated and untreated cultures were then incubated with shaking at 75°C in SCVYU medium.

Survival assays.

S. islandicus E233S was grown to the early exponential phase in SCVYU medium and irradiated with UV or left untreated. After incubation with shaking for 12 h at 75°C, samples were taken and serially diluted. Properly diluted samples were both spotted and spread on SCVYU solid medium. After incubation at 75°C for 7 days, the amounts of cell growth on the spots were compared, and colonies on the spread plates were counted. The rate of survival of the UV-treated sample was calculated by comparing the numbers of CFU in the UV-treated and untreated samples.

Infection rate assays.

S. islandicus E233S was grown to the early exponential phase in SCVYU medium and irradiated with UV or left untreated. After incubation with shaking for 1 h at 75°C, the cultures were infected with SSV2 at an MOI of ~5. Samples were taken at 0, 0.5, 1, 2, and 3 hpi. Each sample was diluted by 10³- and 10⁴-fold. The dilutions were spread on SCVYU solid medium and incubated at 75°C for 7 days. At least 30 colonies were picked from one plate (either the 10³- or the 10⁴-fold dilution) for each time point for PCR analysis using primers targeting the SSV2 vp1 gene (Table S2). The PCR products were subjected to agarose gel electrophoresis. Amplification of a sequence of the expected size was taken to indicate successful infection.

Virus preparation.

S. islandicus REY15/4 carrying SSV2 (49) was grown with shaking in SCVYU medium to an OD₆₀₀ of ~1.0. Cells were centrifuged at 5,000 × g for 20 min at 4°C, and the supernatant was filtered through a 0.45-μm membrane filter (MerckMillipore). The filtrate was centrifuged at 120,000 × g for 1 h at 4°C, and the pelleted virus particles were resuspended in Zillig’s basal salts (50).

Virus infection.

S. islandicus E233S and its derivative strains were grown to the early exponential phase (OD₆₀₀, ~0.2 to 0.3) with shaking at 75°C in SCV or SCVYU medium. SSV2 was added to the cultures at an MOI of ~5. The MOI was determined as follows. By our measurement, an S. islandicus culture at an OD₆₀₀ of 1 contains ~5 × 10⁸ cells/mL. The concentrations of our SSV2 preparations were determined by qPCR as shown below. The SSV2 preparation and the culture were mixed to obtain the desired MOI. Incubation was continued at 75°C.

Quantification of the SSV2 genomic DNA by qPCR.

Total genomic DNA was extracted from the S. islandicus E233S culture infected with SSV2, and the viral DNA was quantified by qPCR with the primer pair vp1 qPCR-F/vp1 qPCR-R (Table S2) as described previously (51). Briefly, a 140-bp fragment in the vp1 gene, which codes for the major capsid protein of SSV2, was amplified with LA-Taq polymerase (TaKaRa). The PCR product was purified by using the V-Elute gel minipurification kit (Zomanbio) and cloned into the T1 vector by using the pEASY-T1 cloning kit (TransGen). The plasmid was then serially diluted and amplified as the known control to generate a standard curve relating the copy number of viral DNA with threshold cycle (C_T) values. Data were analyzed with LightCycler 480 software, version 1.5.1.62, using the Fit Points method to calculate the C_T values according to the software instructions.

qRT-PCR.

A sample (50 mL) of the SSV2-infected culture was centrifuged at 5,000 × g for 20 min at 4°C. Cells were resuspended in 1 mL TRIzol reagent (Invitrogen) and processed according to the manufacturer’s instructions. Genomic DNA in the RNA preparations was removed by treatment with DNase I (1.5 U; Roche) for 30 min at 37°C. The quality and quantity of the resulting total RNAs were determined by measuring absorbance at 260 nm and 280 nm using the NanoDrop ND-1000 spectrophotometer (Labtech). cDNAs were prepared with the random hexamer primers using reverse transcriptase (Thermo Scientific) according to the manufacturer’s instructions. Real-Time PCR was performed with fast, ΔΔC_T, and SYBR green settings on the Applied Biosystems 7500 Fast machine as described (52). The level of transcription for each sample was normalized to an endogenous control obtained with the 16S rRNA gene of S. islandicus E233S. Normalization was done by subtracting the C_T value obtained with primers targeting the 16S rRNA gene from that obtained with primers for an indicated DDR-related gene using the same cDNA template. The ratio of the target to reference RNA was calculated according to the manufacturer’s instructions, and the target and reference sequences are shown in Table S2.

Immunoblotting.

S. islandicus cells were harvested, sonicated, and subjected to immunoblotting as described previously (3). Orc1-2, PCNA, and TFB3 were detected, respectively, with rabbit anti-Orc1-2, rabbit anti-PCNA, and goat anti-TFB3 antibodies, all of which were prepared in our laboratories. His-tagged proteins were detected with Mouse anti-His antibodies (Abbkine). The blots were subsequently developed with corresponding secondary antibodies, i.e., goat anti-rabbit IgG (Promega), goat anti-mouse IgG (Promega), donkey anti-goat IgG (Promega), and goat anti-mouse IgG (Promega). The proteins were visualized by chemiluminescent detection with the Western ECL substrate (Bio-Rad). Immunoblots were quantified by ImageJ software. The relative intensity for each sample was obtained by dividing the signal intensity of the sample by that of PCNA, a loading control. The signal intensity ratio of a sample treated as indicated represents the ratio of the relative intensity of the treated sample to that of an untreated control.

Confocal laser scanning microscopy.

A sample (10 μL) of an exponentially grown S. islandicus culture was spotted on a poly-l-lysine-coated slide (Sigma-Aldrich) and air-dried. The slide was observed under a confocal laser scanning microscope (Leica).

Transmission electron microscopy (TEM).

An aliquot (8 μL) of an S. islandicus culture infected with SSV2 was negatively stained for 1 min with 2% (wt/vol) uranyl acetate and observed under a transmission electron microscope (JEM-1400; Jeol).