|Home | About | Journals | Submit | Contact Us | Français|
Viral factories are compartmentalized centers for viral replication and assembly in infected eukaryotic cells. Here, we report the formation of a replication focus by prototypical archaeal Sulfolobus islandicus rod-shaped virus 2 (SIRV2) in the model archaeon Sulfolobus. This rod-shaped virus belongs to the viral family Rudiviridae, carrying linear double-stranded DNA (dsDNA) genomes, which are very common in geothermal environments. We demonstrate that SIRV2 DNA synthesis is confined to a focus near the periphery of infected cells. Moreover, viral and cellular replication proteins are recruited to, and concentrated in, the viral replication focus. Furthermore, we show that of the four host DNA polymerases (DNA polymerase I [Dpo1] to Dpo4), only Dpo1 participates in viral DNA synthesis. This constitutes the first report of the formation of a viral replication focus in archaeal cells, suggesting that organization of viral replication in foci is a widespread strategy employed by viruses of the three domains of life.
IMPORTANCE The organization of viral replication in foci or viral factories has been mostly described for different eukaryotic viruses and for several bacteriophages. This work constitutes the first report of the formation of a viral replication center by a virus infecting members of the Archaea domain.
The compartmentalization of viral genome replication is a well-known feature of many eukaryotic viruses, where replication is confined to specific subcellular microenvironments termed viral factories, viroplasms, viral replication compartments, or viral replication centers. In general, viral factories function as a scaffold containing viral genomes and proteins involved in viral replication and assembly, and this is hypothesized to increase the efficiency of viral replication and confer protection from the cellular antiviral immune responses (1, 2).
The formation of viral factories has been observed for DNA, double-stranded RNA (dsRNA), positive-sense single-stranded RNA (ssRNA), and negative-sense ssRNA viruses, and it has been extensively studied for nucleocytoplasmic large DNA viruses (e.g., poxviruses and asfarviruses) and positive-sense ssRNA viruses, including flaviviruses, coronaviruses, picornaviruses, and togaviruses (1). Although these viruses infect a wide range of hosts and their corresponding viral factories differ in their morphologies, components, and intracellular locations, their modes of formation share some similarities. First, formation of viral factories usually involves extensive organizational changes in the host cell membranes and/or cytoskeleton. Second, viral factories constitute a scaffold for the concentration of viral genomes and of viral and cellular proteins required for viral replication and assembly (1, 2).
To date, the formation of viral factories has been studied mainly for eukaryotic viruses, although the organization of viral replication has been reported for bacteriophages ϕ29, PRD1, SSP1, and 201ϕ2-1 (3,–5). Viruses that infect the Archaea domain exhibit a wide variety of morphologies and taxonomic diversity comparable to those of eukaryotic viruses, but relatively little is currently known about their replicative cycles or their interactions with hosts (6, 7). The Rudiviridae constitute one of the most common archaeal viral morphotypes found in hyperthermophilic environmental samples, and the prototype rudivirus, Sulfolobus islandicus rod-shaped virus 2 (SIRV2), is one of the most extensively studied (8). Rudiviruses infect members of the acidothermophilic crenarchaeal order of the Sulfolobales and include four isolated viruses, SIRV1, SIRV2, Sulfolobus islandicus rod-shaped virus (SRV), and Acidianus rod-shaped virus (ARV) (9,–11), and a further two members, Sulfolobales Mexican rudivirus 1 (SMR1) and ARV2, assembled from metagenomes (12, 13).
In this work, we provide evidence for the early formation of a replication focus in cells infected with SIRV2 where viral DNA synthesis takes place and where viral single-stranded DNA (ssDNA) binding protein gp17 and host proteins host proliferating cell nuclear antigen (PCNA) and DNA polymerase I (Dpo1) are localized. In addition, it is shown that the host replicative Dpo1, rather than the other DNA polymerases (Dpo2 to Dpo4), is primarily responsible for viral DNA synthesis.
Recently, we showed that multiple copies of ssDNA intermediates are produced during SIRV2 replication (14). This allows us to determine the intracellular location of SIRV2 DNA synthesis by fluorescence microscopy using an anti-bromodeoxyuridine (BrdU) monoclonal antibody which exclusively targets BrdU in ssDNA but not in double-stranded DNA (dsDNA) (15). Since Sulfolobus species do not carry a native thymidine kinase (tk) gene, a plasmid-borne tk gene was introduced into Sulfolobus solfataricus Sens1 which enabled incorporation of BrdU, a thymidine analog, into DNA in the transformant strain S. solfataricus SsoTK.
The expression of the tk gene in SsoTK cells was induced by arabinose in ACVY medium prior to infection with SIRV2 (multiplicity of infection [MOI] of 5) and addition of 1 mM BrdU to the culture. In the infected cells, the BrdU signal was clearly observed at 2 h postinfection (hpi), consistent with the presence of numerous ssDNA replication intermediates from the virus (Fig. 1A). In contrast, the uninfected SsoTK cells contained only dsDNA and DNA denaturation with HCl was required in order to detect BrdU-containing DNA (Fig. 1A). Notably, the BrdU signal in the infected cells was located at a single focus at the periphery of the cell (Fig. 1A). This strongly suggested that the SIRV2 replication was organized and confined to a single intracellular site.
To investigate whether viral replication proteins were recruited to the site of viral DNA synthesis, we localized viral protein gp17 in the infected SsoTK cells. This ssDNA-binding protein is considered to be essential for viral replication and is highly expressed throughout the SIRV2 replication cycle (16, 17). First, the SsoTK cells carrying the tk gene were infected with SIRV2 and supplemented with BrdU as described above. gp17 was then visualized by epifluorescence microscopy using a guinea pig anti-gp17 polyclonal antibody, and BrdU-containing ssDNA was visualized as described above. Figure 1B shows the result from the immunostaining of both gp17 (red) and BrdU-containing ssDNA (green). In a given cell, the gp17 protein was also located at a single peripheral focus that overlapped completely with the site of DNA synthesis, strongly suggesting that the focus constitutes a viral replication center.
Next, the formation of the SIRV2 replication focus was examined throughout the viral life cycle, which lasts about 12 h (17, 18). To circumvent any deleterious effects caused by the incorporation of BrdU into the viral DNA, we excluded BrdU from the SIRV2-infected S. solfataricus 5E6 cultures (MOI of 5) and monitored the formation of the viral replication focus by examining the distribution of gp17. The gp17 foci were detected already at 1 hpi and increased in size and intensity as infection progressed (Fig. 1C). At 10 hpi, when virus-induced cell lysis initiates (17, 18), cells devoid of both gp17 and DNA were observed (Fig. 1C, white asterisks), consistent with the occurrence of cell lysis and virus release (18).
To test whether host replisomal proteins are also recruited to the viral replication focus, we investigated the location of PCNA in S. solfaraticus 5E6 cells infected with SIRV2 (MOI of 10). PCNA is an essential host replisomal protein that functions as the processivity factor for the replicative DNA polymerase and as a scaffold to recruit proteins involved in DNA replication and repair (19). In addition, several SIRV2 proteins have been reported to interact with the host PCNA in vitro (20), suggesting that PCNA participates in viral DNA replication. In uninfected cells, PCNA showed a peripheral multifocus distribution (Fig. 2, 0 hpi), as described earlier for Sulfolobus acidocaldarius (21). In contrast, PCNA is redistributed to a single focus overlapping with that of gp17 in SIRV2-infected cells (Fig. 2, 2 to 3 hpi).
Next, we examined the intracellular distribution of host DNA polymerases which are implicated in SIRV2 replication because the viral genome does not encode a DNA polymerase (6). Four DNA polymerases are encoded in the genome of Sulfolobus: Dpo1 (sso0552), Dpo2 (sso1459), and Dpo3 (sso0081) of the B family of DNA polymerases and Dpo4 (sso2448) of the Y family (22). In uninfected S. solfataricus 5E6 cells, each DNA polymerase was distributed in several foci throughout the cell (Fig. 3A). Upon SIRV2 infection, Dpo1 was redistributed to a single site that overlapped with the gp17 focus. In contrast, distribution of the other three DNA polymerases remained unaltered in the infected cells and did not overlap the viral replication focus significantly (Fig. 3B to toE).E). Moreover, the Dpo1 signal intensified during the progression of SIRV2 replication, in a manner similar to that of the SIRV2 gp17 protein, whereas Dpo2 to Dpo4 showed an unaltered distribution pattern throughout the viral life cycle (Fig. 3C to toE).E). This preferential concentration of Dpo1, but not of Dpo2 to Dpo4, with respect to the gp17 focus suggests a primary role for Dpo1 in SIRV2 genome replication (see below). In addition, these results provide further support for the idea of the occurrence of a SIRV2 replication focus where essential host replication proteins concentrate.
The SIRV2 genes are likely to be transcribed by the host RNA polymerase; therefore, we examined the intracellular distribution of RNA polymerase subunits Rpo4/7, which form a stalk in the archaeal RNA polymerase. Archaeon heterotrimeric translation initiation factor 2 gamma (aIF2γ) was also examined in SIRV2-infected cells to assess whether the location of the translation machinery was affected by SIRV2 infection (Fig. 4). The Rpo4/7 stalk was spread throughout the cytoplasm in both infected and uninfected cells (data not shown), and there was no evidence for it being concentrated in the viral replication focus. In contrast, aIF2γ changed its distribution dramatically early during infection and was concentrated at a single cellular site throughout the first half of the infection cycle (Fig. 4). However, the aIF2γ focus did not overlap the gp17 focus, indicating that replication and translation were highly organized but physically separated processes in the infected cells.
The main role of Dpo1 in SIRV2 genome replication was further investigated in the individual knockout (KO) mutants of Dpo2, Dpo3, and Dpo4 and also in the knockdown mutant of Dpo1. First, single-deletion mutants of Dpo2 to Dpo4, namely, Dpo2KO, Dpo3KO, and Dpo4KO, were constructed in Sulfolobus islandicus LAL14/1 using type I-A clustered regularly interspaced short palindromic repeat (CRISPR)-based genome editing (23). Each deletion mutant was infected with SIRV2 (MOI of 1) and incubated for 24 h, and then the amount of virus present in the supernatant was quantified. Infection of the wild-type LAL14/1 strain was used as a reference. The results showed that production of viral progeny was not negatively affected by the deletion of Dpo2, Dpo3, or Dpo4 (Fig. 5A), demonstrating that they are not essential for SIRV2 replication.
Given the key role of Dpo1 in Sulfolobus genome replication and the possible essentiality of its gene in Sulfolobus, dpo1 was knocked down using a type III CRISPR-dependent RNA silencing approach (Fig. 5B) (24). A spacer targeting dpo1 mRNA but not the DNA gene was cloned into a plasmid-borne artificial mini-CRISPR locus to produce pAC-dpo1. After transforming the plasmid into S. islandicus LAL14/1, the mRNA level of dpo1 was, on average, 50% lower than that in the wild-type strain as determined by quantitative reverse transcription-PCR (qRT-PCR) (Fig. 5C). Moreover, the transformant harboring pAC-dpo1 exhibited growth retardation relative to the wild-type strain (data not shown). To determine whether silencing of dpo1 affects SIRV2 DNA replication, the wild-type LAL14/1 strain, and the transformant harboring plasmid pAC-dpo1, were infected with SIRV2 (MOI of 1) and the level of viral progeny produced was quantified after 24 h. The average amount of virus produced in the transformant was found to be 24% of the amount of virus produced in the wild-type strain (Fig. 5D). Since both LAL14/1 and pAC-dpo1 do not grow after SIRV2 infection, which is lytic to the hosts, the lower viral yield in the latter is thus unlikely have resulted from the slower growth observed before viral infection. Taken together (Fig. 5A and andC),C), these results confirm that Dpo1 is the primary polymerase involved in SIRV2 genome replication.
For eukaryal viruses, the compartmentalization of viral replication in the structures named viroplasms or viral factories has been long known and extensively documented. These structures can be nuclear or cytoplasmic, and they compartmentalize viral and cellular factors required for DNA and RNA synthesis (1, 2). With regard to bacterial viruses, a few examples of the organization of viral replication have been reported (3, 4, 25), with one recent study showing that the phage 201ϕ2-1 replication compartment can display the complexity of the viral factories formed by eukaryal viruses (5). We demonstrate the formation of a viral replication focus at the periphery of Sulfolobus cells, after infection with SIRV2, where viral DNA synthesis takes place and where replication-related proteins of the virus (gp17) and host (PCNA and Dpo1) are concentrated.
The majority of the eukaryal viruses studied to date require the reorganization of cellular membranes and the cytoskeleton for the formation of their viral replication compartments. Eukaryotic RNA and DNA viruses replicating in the cytoplasm require the relocalization of organelles and reorganization of cellular membranes (1). In contrast, eukaryotic DNA viruses replicating in the nucleus do not seem to require membranous structures for the formation of viral factories; instead, they make use of nuclear components such as the promyelocytic leukemia (PML) nuclear bodies (2). As for bacterial viruses, while phage SPP1 replicates at a single focus close to the cell poles, suggesting attachment to the cell membrane, phages ϕ29 and PRD1 show a nucleoid organization of replication that is dependent on the presence of the membrane-associated MreB actin-like cytoskeleton (4). Phage 201ϕ2-1 also forms a replication compartment that is delimited by the gp105 viral protein and positioned at the middle of the cell by a bipolar viral tubulin-based spindle, PhuZ (5).
Whether formation of the SIRV2 replication focus in Sulfolobus involves any membranous structures remains to be investigated; the peripheral location of the SIRV2 replication focus suggests that it may be attached to the cellular membrane, which could serve as a scaffold for organizing SIRV2 replication. Interestingly, PCNA distribution in Sulfolobus cells, in the absence of viral infection, is also peripheral (21). Therefore, attachment to the membrane could be a cellular mechanism for organizing replisomes that the virus exploits to its own advantage. PCNA was recently reported to interact with several SIRV2 proteins in vitro (20), and these interactions may underlie the mechanism by which the virus captures the host replication machinery for replicating its own genome.
Apart from the involvement of the cellular membrane, the replication mechanism of SIRV2 may also contribute to the formation of the viral replication focus. The 35-kbp linear dsDNA genome of SIRV2 carries hairpin ends and is replicated via strand displacement combined with multiple reinitiation events on a single parental template. This generates large replication intermediates (greater than 1.2 Mbp in size) such that all viral progeny genomes remain linked until processing and maturation start at 4 hpi (14). This mechanism necessarily facilitates occurrence of both replication and virion assembly at a localized site. Indeed, SIRV2 particles were found to be well aligned and tightly packed into 1 to 3 bundles before virus release (18), underlining that SIRV2 virion assembly is well organized spatially. While the cytoskeleton appears to play a crucial function in the compartmentalization of viral replication for both eukaryal and bacterial viruses, the role of the host cytoskeleton in the different stages of SIRV2 replication remains to be addressed.
We propose a model for the localized SIRV2 replication. Following infection, the viral genome is localized to a specific site in the periphery of the host cell. Then, viral proteins (e.g., gp16, gp17) and cellular proteins (e.g., PCNA, Dpo1) involved in viral DNA replication are recruited and DNA synthesis begins. The SIRV2 gp16 Rep protein is the putative replication initiator (26), while the highly abundant ssDNA gp17 binding protein is hypothesized to bind to and protect the large ssDNA regions generated during replication (16). Resolution into monomeric genomes is referred to as being affected by the viral Holliday junction resolvase (Hjr) gp35 (27), which, owing to its interaction with the major capsid protein of SIRV2 (27), provides a link between viral replication and virion assembly.
Whereas viroplasms are formed by members of a broad range of eukaryotic viral families, the extent to which archaeal viruses form viral replication centers remains unknown. Limited by the small cellular sizes, prokaryotes have rarely been studied for spatial organization of viral replication. The detection of a SIRV2 replication focus in the archaeon Sulfolobus suggests that the formation of specialized viral replication microenvironments is a mechanism commonly employed by viruses of the three domains of life.
All strains used in the study are listed in Table 1. Sulfolobus solfataricus 5E6 (17) cells were grown in small-colony variant Y (SCVY) or ACVY media. Sulfolobus islandicus LAL14/1 (31) cells were grown in SCVYU medium. These media contain the salt base solution (32), one of the three types of carbon source (0.2% sucrose, 0.2% d-arabinose, or 0.2% glucose for SCVY/SCVYU, ACVY, or GCVY media, respectively), 0.2% Casamino Acids, Wolin's vitamin mixture, and 0.005% yeast extract (20 mg/liter uracil in SCVYU medium) (33). Cultures were incubated at 78°C with agitation.
Competent cells were prepared and transformed by electroporation as described previously (33). Transformants were selected on SCVY or GCVY plates, as indicated for each case. Subsequently, the obtained colonies were grown in liquid medium.
S. islandicus LAL14/1 cells were grown to an optical density at 600 nm (OD600) of approximately 0.2 and were then infected with SIRV2 and incubated for 2 to 3 days until the OD600 started to decrease. Cells were then removed by centrifugation at 4,025 × g for 10 min, and virions in the supernatant were precipitated with 10% polyethylene glycol 6000 and 1 M NaCl at 4°C overnight. Virions were recovered by centrifugation at 11,180 × g for 30 min and resuspended in 10 mM Tris-acetate (pH 6). Cell debris was removed by centrifugation at 4,025 × g for 5 min. SIRV2 stocks were titrated by focus-forming assay as described below.
For infection of cell cultures, the corresponding strain was grown in liquid medium to an OD600 of 0.2 and then infected with SIRV2 with a multiplicity of infection (MOI) of 5 to 10 and incubated at 78°C for the indicated times in each experiment.
SIRV2 stocks and viral supernatants were titrated by the focus-forming assay as follows. Aliquots of a Sulfolobus solfataricus 5E6 culture with an OD600 of 0.2 were infected with 10-fold serial dilutions of the SIRV2 sample and incubated for 4 h at 78°C. Cells were then harvested, fixed, and processed for immunofluorescence microscopy using the anti-gp17 antibody as described below (“Immunofluorescence staining”).
The percentage of infected cells for each dilution was estimated by counting the number of cells positive for gp17 signal. The amount of virus particles, in terms of focus-forming units (FFUs) per milliliter (FFU/mL), was derived from the Poisson distribution equation as follows, where Po is the fraction of uninfected cells:
Ideal counts derived from virus dilutions that resulted in 20 to 60% of infected cells.
The thymidine kinase gene of Pyrobaculum aerophilum was amplified from plasmid pTK3, kindly provided by Stephen D. Bell (21), using primers TKndeIF and TKstuIR (Table 2). The resulting PCR product was cloned into the NdeI and StuI sites of plasmid pEXA2 (30) to generate pEXA2-TK, where the tk gene is under the control of an inducible arabinose promoter. pEXA2-TK was electroporated into S. solfataricus Sens1 (28), and transformants were selected on SCVY plates and screened by PCR. One colony was grown in liquid medium and labeled strain SsoTK. The ability of SsoTK to incorporate BrdU was tested by a dot blot assay (data not shown).
SsoTK cells were grown overnight in ACVY medium to an OD600 of 0.2. BrdU was then added to a final concentration of 1 mM, and, if applicable, cultures were infected with SIRV2 with an MOI of 5. Cells were further incubated for 2 h at 78°C and then fixed with 4% paraformaldehyde and processed for epifluorescence microscopy as described below. To denature DNA, the corresponding samples were treated with 2 N HCl–0.5% Triton X-100 for 15 min at room temperature and neutralized with 0.2 M sodium tetraborate for 2 min before immunostaining was performed.
Plasmids for generation of the deletion mutants were constructed according to the CRISPR-based genome editing method (23). Plasmid pEXA-arrayIA carrying an artificial mini-CRISPR array was constructed by inserting a DNA fragment of two tandem copies of the CRISPR repeat separated by two oppositely oriented SapI restriction sites. Genes SiL_0653, SiL_1893, and SiL_0229, encoding DNA polymerases Dpo2, Dpo3, and Dpo4, respectively, were identified in the genome of S. islandicus LAL14/1. A spacer targeting each of the genes was designed and generated by annealing two complementary oligonucleotides (Table 2). Briefly, each pair of oligonucleotides was heated at 95°C for 10 min and then cooled gradually to room temperature. The annealed spacer was inserted into the pEXA-arrayIA vector at the SapI sites. The genomic regions flanking each gene were retrieved, and a single composite PCR product containing both flanking regions was amplified by PCR using the corresponding oligonucleotides (Table 2). The resulting composite PCR products were digested with SphI and XhoI and inserted into the pEXA-arrayIA vector at the equivalent restriction sites. The resulting vectors were electroporated into S. islandicus LAL14/1, and transformants were selected on SCV plates, yielding strains Dpo2KO, Dpo3KO, and Dpo4KO. Selected colonies were screened by PCR followed by sequencing of the PCR products. Effective gene deletion and purity of the transformants were tested by RT-PCR (data not shown).
A spacer fragment targeting the dpo1 gene of S. islandicus LAL14/1 (SiL_1453) was prepared by annealing the corresponding oligonucleotides (Table 2). The annealed spacer was cloned into plasmid pSe-Rp (24) at the BspMI sites, yielding plasmid pAC-dpo1, which contains an artificial mini-CRISPR locus. pAC-dpo1 was then transformed into S. islandicus LAL14/1, and the resulting transformants were selected on plates containing GCVY medium, followed by PCR screening.
Cell cultures of the corresponding strain were grown to an OD600 of 0.2 and infected with SIRV2 with an MOI of 1. Infected cultures were incubated at 78°C for 24 h. Afterward, cells were collected by centrifugation at 6,200 × g for 3 min and the supernatants were titrated by focus-forming assay. The relative virus titers were calculated using SIRV2-infected S. islandicus LAL14/1 cultures as a reference.
Total RNA was extracted using TRI reagent (Sigma-Aldrich), and RNA quality was assessed by gel electrophoresis. RNA samples were treated with DNase I (Thermo Fisher Scientific) before reverse transcription was performed using a RevertAid First Strand cDNA synthesis kit (Thermo Fisher Scientific) with random hexamers as primers. Reaction mixtures without reverse transcriptase were included to test for the presence of residual DNA after DNase I treatment. Amplification of the cDNA samples by quantitative PCR (qPCR) was performed using Maxima SYBR green/Rox qPCR master mix (Thermo Fisher Scientific) with a CFX96 real-time detection system (Bio-Rad). Oligonucleotides for amplification of dpo1 (SiL_1453) and tfbI (SiL_1547) are described in Table 2. The following cycling parameters were used: 95°C for 5 min and 39 cycles of 95°C for 10 s, 51°C for 10 s, and 72°C for 25 s. Reactions were carried out in triplicate, and average values were used for quantification. Data were analyzed with Bio-Rad CFX managing software, the amounts of the dpo1 target gene and the tfbI reference gene were determined, and their relative levels of expression with respect to the wild-type S. islandicus LAL14/1 strain were calculated using the threshold cycle (ΔCT) method (34).
The guinea pig anti-gp17 polyclonal antibody was produced by Innovagen AB (Lund, Sweden) using a His-tagged gp17 open reading frame (ORF), cloned from the SIRV2 genome, that was expressed in Escherichia coli and purified as described previously (16). For detection of BrdU, the commercial rat monoclonal anti-BrdU [BU1/75 (ICR1)] antibody (Abcam; catalog no. ab6326) was used.
Visualization of each of the Sulfolobus DNA polymerases was accomplished using the rabbit polyclonal sera obtained from P. Guengerich (22). The rabbit polyclonal antibodies anti-PCNA1, anti-Rpo4/7, and anti-aIF2γ (35) were kindly provided by Changyi Zhang, Finn Werner, and Udo Bläsi, respectively. The specificity of the polyclonal antibodies was tested on cellular lysates by Western blotting.
The commercial secondary antibodies anti-guinea pig Alexa Fluor 594, anti-rabbit Alexa Fluor 488, and anti-rat Alexa Fluor 488 (Invitrogen; catalog no. A11076, A11008, and A21210, respectively) were used for detection of their corresponding primary antibodies during fluorescence microscopy experiments.
Cells were recovered by centrifugation at 6,200 × g for 3 min and fixed by suspending the cell pellet in 4% paraformaldehyde for 10 min at room temperature. Cells were washed for three times with phosphate-buffered saline (PBS) buffer (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4) by pelleting the cells at 6,200 × g for 3 min and resuspending them in PBS. Then the samples were permeabilized with PBS–0.1% Tween20 (PBST) for 15 min at room temperature and afterward washed for three times with PBS. Approximately 3.6 × 106 cells contained in 10 μl of cell suspension were spotted on poly-l-lysine-coated slides (Sigma-Aldrich) and air dried. The slides were rinsed with PBS and incubated with PBST–1% bovine serum albumin (BSA) for 15 min at room temperature. Afterward, the slides were incubated with the corresponding primary antibodies diluted in PBST–1% BSA for 1 h at room temperature. The anti-gp17 antibody was used at a dilution of 1:5,000, the anti-BrdU antibody was used at a dilution of 1:100, the anti-Dpo1 and anti-Dpo-3 antibodies were used at a dilution of 1:1,000, and the anti-Dpo2 and anti-Dpo4 antibodies were used at a dilution of 1:200. After three washes of 2 min each with PBS, the slides were incubated with the corresponding secondary antibodies diluted in PBST–1% BSA at room temperature for 1 h; all the secondary antibodies were used at a dilution of 1:1,000. The slides were washed for three times with PBS and incubated with 10 μM 4′,6-diamino-2-phenylindole (DAPI) (Sigma-Aldrich) for 5 min at room temperature. Then, the slides were again washed for three times with PBS and mounted with PVP mounting medium (1× PBS, 78% glycerol, 0.2% polyvinyl pyrrolidone).
Slides were analyzed by using a Zeiss Axio Imager.Z1 ApoTome microscope coupled to a charge-coupled-device (CCD) Hamamatsu ORCA-ER camera (C4742-80) or using a Nikon Eclipse Ti-E inverted microscope coupled to an Andor Zyla 5.5 sCMOS (scientific complementary metal-oxide semiconductor) camera. Images were processed using Volocity 6.3 software (PerkinElmer) and Adobe Photoshop CS6.
We thank Stephen D. Bell for providing the thymidine kinase gene and Yang Guo for purification of gp17. We also thank P. Guenguerich, C. Zhang, F. Werner, and U. Bläsi for kindly providing some of the antibodies used in this work.
This work was supported by EU FP7 project HotZyme (265933) and by The Danish Council for Independent Research/Natural Sciences (grant no. DFF-4181-00274B). L.M.-A. was supported by a doctoral scholarship from Consejo Nacional de Ciencia y Tecnología (312536). The funding institutions had no role in any decision taken about the design, development, or publication of the present work.