|Home | About | Journals | Submit | Contact Us | Français|
The Autographa californica M nucleopolyhedrovirus (AcMNPV) orf79 (ac79) gene is a conserved gene in baculoviruses and shares homology with genes in ascoviruses, iridoviruses, and several bacteria. Ac79 has a conserved motif and structural similarities to UvrC and intron-encoded endonucleases. Ac79 is produced at early times during infection and concentrates in the nucleus of infected cells at late times, suggesting a cellular compartment-specific function. To investigate its function, an ac79-knockout bacmid was generated through homologous recombination in Escherichia coli. Titration assays showed that budded virus (BV) production was reduced in the ac79-knockout virus compared to control viruses, following either virus infection or the transfection of bacmid DNA. The ac79-knockout virus-infected cells produced plaques smaller than those infected with control ac79-carrying viruses. No obvious differences were observed in viral DNA synthesis, viral protein accumulation, or the formation of occlusion bodies in ac79-knockout and control viral DNA-transfected cells, indicating progression into the late and very late phases of viral infection. However, comparative analyses of the amounts of BV genomic DNA and structural proteins in a given quantity of infectious virions suggested that the ac79-knockout virus produced more noninfectious BV in infected cells than the control virus. The structure of the ac79-knockout BV determined by transmission electron microscopy appeared to be similar to that of the control virus, although aberrant capsid protein-containing tubular structures were observed in the nuclei of ac79-knockout virus-infected cells. Tubular structures were not observed for ac79 viruses with mutations in conserved endonuclease residues. These results indicate that Ac79 is required for efficient BV production.
Baculoviruses have circular double-stranded DNA genomes of between 80 and 180 kbp (12) and infect arthropods. Baculoviruses include four genera: Alphabaculovirus, Betabaculovirus, Gammabaculovirus, and Deltabaculovirus (13). Alphabaculoviruses and betabaculoviruses infect larvae of Lepidoptera; while gammabaculoviruses and deltabaculoviruses, infect larvae of Hymenoptera and Diptera, respectively (13).
During alphabaculovirus replication, two types of virions are produced, the budded virus (BV) and the occlusion-derived virus (ODV). The BV is produced as the nucleocapsid buds out from the plasma membrane of the infected cell. BVs are essential for establishing systemic infection in an infected insect by transmitting infection from cell to cell. The ODV is formed in the nucleus of the infected cell prior to being embedded within a crystalline matrix composed mainly of the protein polyhedrin. ODV spreads infection from larvae to larvae when a dead host releases ODV-containing polyhedra into the environment and polyhedra are ingested by another host.
The Autographa californica M nucleopolyhedrovirus (AcMNPV) open reading frame (ORF) orf79 (ac79) has not been studied in detail, and our knowledge of the function of ac79 stems from genomic or proteomic analyses. The genomic sequence of ac79 predicts a gene product of 104 amino acids with a putative molecular mass of 12.2 kDa (2). Homologs of ac79 are found in baculoviruses of the genera Alphabaculovirus and Betabaculovirus, ascoviruses, iridoviruses, and bacteria (23). A previous proteomic study determining the protein composition of ODVs of AcMNPV concluded that Ac79 was associated with ODVs (5). It was suggested that Ac79 is a member of the DNA repair UvrC endonuclease superfamily with similarities to intron-encoded endonucleases, since it is predicted to contain two tyrosines spaced by about 10 amino acids, the hallmark sequence RX3[YH], and a key glutamate downstream of this sequence (1).
In this study, we characterized ac79 by mapping the transcription start sites and determining protein synthesis and the localization of its product during virus replication. In order to investigate the role of Ac79 during AcMNPV replication, we generated an ac79-knockout virus, Ac79KO-PG, through homologous recombination in Escherichia coli. The deletion of ac79 resulted in a decrease in BV production but did not affect viral DNA replication or late and very late protein accumulation. Ac79KO-PG was able to produce BV, but more virions than the control virus were not infectious. Electron micrographs did not reveal virion structure defects in the absence of ac79. Elongated tubular structures containing the major capsid protein VP39 were produced in ac79-knockout virus-infected cells. These tubular structures were not observed for viruses carrying ac79 mutations in the UvrC/intron-encoded endonuclease conserved residues.
Bacmid bMON14272 (Invitrogen), here referred to as AcBAC, containing an AcMNPV genome, was propagated in E. coli BJ5183 cells as described previously (3). The virus AcWT-PG was constructed by introducing the polyhedrin gene and the enhanced green fluorescent protein gene (egfp) at the polyhedrin locus of AcBAC (32). DH10B cells with helper plasmid pMON7124, encoding a transposase, were purchased from Invitrogen. The Sf9 insect cell line, clonal isolate 9 from IPLB-Sf21-AE cells, derived from the fall armyworm Spodoptera frugiperda (26), was purchased from the ATCC and cultured at 27°C in TC-100 medium (Invitrogen) supplemented with 10% fetal bovine serum, penicillin G (60 μg/ml), streptomycin sulfate (200 μg/ml), and amphotericin B (0.5 μg/ml).
To map the transcription start sites for ac79, RNA was extracted from AcWT-PG-infected Sf9 cells and collected at 6, 12, and 24 h postinfection (p.i.) by use of TRIzol reagent (Sigma) according to the manufacturer's protocol. RNA was quantified by measuring the optical density at 260 nm. Rapid amplification of 5′ cDNA ends (5′ RACE) was performed by using a 5′ RACE kit, version 2.0, according to the handbook provided by the manufacturer (Invitrogen). ac79-specific primer Ac79SP1 (5′-GGTTTGGCTTTGATGAGACGC-3′) was used to synthesize first-strand cDNA. A nested ac79-specific primer, Ac79SP2 (5′-GCTGTGATACACGAGCCGTA-3′), and the provided Abridged Anchor Primer were used for PCR amplification using the first-strand cDNA as a template. PCR products were gel purified with the Gel Extraction kit (Qiagen) and cloned into pCRII (Invitrogen) prior to deriving the nucleotide sequence. A total of five clones per time point were sequenced in each of two independent experiments.
A monolayer of Sf9 cells (1 × 106 cells) was infected at a multiplicity of infection (MOI) of 5 with Ac79HARep-PG, an ac79-knockout bacmid expressing Ac79 at a different locus (see below). The protein synthesis inhibitor cycloheximide and the DNA synthesis inhibitor aphidicolin were added, and cell samples were collected as previously described (32). Proteins were analyzed by sodium dodecyl sulfate (SDS)–15% polyacrylamide gel electrophoresis (PAGE) followed by immunoblotting or stored at −20°C until further use.
Sf9 cells (1 × 106 cells) were infected with Ac79HARep-PG at an MOI of 5 and collected at 6, 12, 24, 48, and 72 h p.i. To fractionate the cells into cytoplasmic and nuclear fractions, the cell pellet was resuspended in 50 μl of ice-cold buffer A (10 mM Tris [pH 7.9], 10 mM NaCl, 5 mM MgCl2, 1 mM dithiothreitol, 0.5% [vol/vol] NP-40) and kept on ice for 5 min. The cells were disrupted and nuclei were released by using a prechilled Dounce homogenizer (10 strokes with a tight pestle). Cells were centrifuged at 1,000 × g for 5 min, and the supernatant was retained as the cytoplasmic fraction. Pelleted nuclei were washed five times with buffer A and resuspended in 50 μl of buffer A. Fractions were stored at −20°C until further use.
Immunoblotting was performed as previously described (32). The primary antibodies used in these study were (i) mouse monoclonal anti- hemagglutinin (HA) antibody (Covance), to detect proteins tagged with an HA tag; (ii) rabbit polyclonal VP39 antiserum (16) (a gift from Kai Yang), to detect the major capsid protein; (iii) mouse monoclonal anti-GP64 antibody (eBioscience), to detect the viral fusion envelope protein; (iv) mouse monoclonal anti-IE-1 antibody (a gift from Linda Guarino), to detect the transactivator IE-1; and (v) rabbit polyclonal Orgyia pseudotsugata MNPV (OpMNPV) polyhedrin antiserum (a gift from George Rohrmann), to detect the cross-reactive AcMNPV polyhedrin.
The ac79-knockout bacmid was generated through homologous recombination in E. coli as previously described (32), in which the ac79 gene of AcBAC was replaced with the chloramphenicol resistance gene (Cm). A 1,178-bp PCR fragment, containing a 1,038-bp Cm cassette and 70 bp of ac79 flanking regions at each end, was amplified by using pCMR (29) as the template and primers Ac79D1 (5′-CTGCGGCAAGACAATGGAAAATTGTACACGGGCATCACGAGCAATCTTAACAGACGCATAAAACAGCATTCCCTTTCGTCTTCGAATAAA-3′) and Ac79D2 (5′-TTATGCAACAAAAGTGGTTTGGCTTTGATGAGACGCAATTTGAAATACTTGCTGCATTTACGCTTAAGATTAAACCAGCAATAGACAAA-3′). Purified PCR fragments (1 μg) were electroporated into electrocompetent BJ5183 cells, and the recombinant ac79-knockout bacmid (Ac79KO) was selected, as previously described (32), and confirmed by PCR amplification of inserted and flanking DNA.
Four primers were used to confirm the replacement of ac79 by Cm in the ac79 locus of AcBAC. Primers Cm5 (5′-CTTCGAATAAATACCTGTGA-3′) and Cm3 (5′-AACCAGCAATAGACATAAGC-3′) were used to detect the correct insertion of the Cm cassette. Primers Ac7951 and Ac7931, which bind to the sequences outside the region of recombination, were used to confirm the expected deleted region and the insertion of Cm at the ac79 locus. Primer pairs Ac7951/Cm3 and Cm5/Ac7931 were used to examine the recombination junctions.
To facilitate the examination of virus-infected cells and to determine if the deletion of ac79 had any effect on occlusion morphogenesis, the egfp and polyhedrin genes were inserted into the polyhedrin locus of AcBAC by site-specific transposition, as previously described (32). To this end, several donor plasmids were constructed. Two primers, Ac7952 (5′-GAGCTCATCTGCGTCTGCCAACATAT-3′ [the SacI restriction site is underlined]) and Ac7933 (5′-TCTAGATAATTCCCGCCGACACGTTG-3′ [the XbaI restriction site is underlined]), were designed to amplify an 816-bp fragment containing the native ac79 promoter (300 bp upstream of the ac79 ATG) and the ac79 ORF (Ac79POA), using AcBAC DNA as a template. This PCR product was ligated into pCRII, and the resulting plasmid, pCRII-Ac79POA, was confirmed by nucleotide sequencing. pCRII-Ac79POA was digested with SacI and XbaI, and the Ac79POA fragment was inserted into pFB1-PH-GFP (31) to generate a donor plasmid named pFB-PG-Ac79POA. AcBAC and primers Ac7952 and Ac7932 (5′-TCTAGATTAGGCGTAATCTGGGACGTCGTATGGGTACAACTTATTTGCTAACAGGA-3′ [the XbaI restriction site is underlined]) were used to amplify a 654-bp fragment that contained the ac79 native promoter and the ac79 ORF with an HA tag prior to the stop codon (Ac79POHA). The PCR product was ligated into pCRII and confirmed by nucleotide sequencing. The resulting plasmid, pCRII-Ac79POHA, was digested with SacI and XbaI to obtain the Ac79POHA fragment and inserted into pFB-PG-pA (32), generating the donor plasmid pFB-PG-Ac79POHA. Electrocompetent DH10B cells containing the helper plasmid pMON7124 and Ac79KO were transformed with donor plasmid pFB1-PH-GFP, pFB-PG-Ac79POA, or pFB-PG-Ac79POHA to generate the ac79-knockout virus Ac79KO-PG and two repair viruses, Ac79Rep-PG with ac79 and Ac79HARep-PG with an HA-tagged ac79. Control bacmid AcWT-PG, carrying the polyhedrin and egfp genes, was described previously (32). All bacmids were cured of helper plasmids as described previously (32). Bacmid DNA was extracted and purified with a Qiagen Large-Construct kit and quantified by optical density.
To determine whether conserved amino acids in Ac79 were important for function, we changed the nucleotide sequences of tyrosine and glycine at positions 24 and 26, respectively, to make Ac79Y24AG26A-PG; arginine at position 34 to encode lysine and construct Ac79R34K-PG; or glutamic acid at position 72 to encode aspartic acid and construct Ac79E72D-PG. Three primer pairs were used to construct donor plasmids pFB-PG-Ac79Y24AG26A, pFB-PG-Ac79R34K, and pFB-PG-Ac79E72D for transposition by using a QuikChange XL site-directed mutagenesis kit (Agilent Technologies) and pFB-PG-Ac79POHA as the template. Primers Ac79 GIYmut f (5′-CAAGACAATGGAAAATTGGCCACGGCCATCACGAGCAATCTTAAC-3′) and Ac79 GIYmut r (5′-GTTAAGATTGCTCGTGATGGCCGTGGCCAATTTTCCATTGTCTTG-3′) were designed to introduce mutations into Ac79Y24AG26A-PG, primers Ac79 RtoKmut f (5′-CACGGGCATCACGAGCAATCTTAACAGAAAGATAAAACAGCATTCGAAC-3′) and Ac79 RtoKmut r (5′-GTTCGAATGCTGTTTTATCTTTCTGTTAAGATTGCTCGTGATGCCCGTG-3′) were designed to introduce a mutation into Ac79R34K-PG, and primers Ac79 EtoDmut f (5′-GCCGCCCGCATGGATTACAATCTTAAGCGTAAATGC-3′) and Ac79 EtoDmut r (5′-GCATTTACGCTTAAGATTGTAATCCATGCGGGCGGC-3′) were designed to introduce a mutation into Ac79E72D-PG. Electrocompetent DH10B cells containing helper plasmid pMON7124 and Ac79KO were transformed with donor plasmid pFB-PG-Ac79Y24AG26A, pFB-PG-Ac79R34K, or pFB-PG-Ac79E72D to generate Ac79Y24AG26A-PG, Ac79R34K-PG, or Ac79E72D-PG, as described above.
Sf9 cells (1.0 × 106 cells/35-mm-diameter dish) were transfected with 1 μg of bacmid DNA using Grace's unsupplemented medium (Invitrogen) and Lipofectin (6), as previously described (32), or infected with virus at an MOI of 0.01 or 5. The supernatants of transfected or infected cells were collected at various time points to determine titers by 50% tissue culture infective dose (TCID50) endpoint dilution assays (22) using Sf9 cells. Time zero was defined as the time when the DNA-Lipofectin mixture or virus was replaced with fresh medium.
Sf9 cells were infected with AcWT-PG or Ac79KO-PG at an MOI of 0.01, 0.005, or 0.001, and plaque assays were performed as previously described (31). Virus plaque diameters were measured at 72 h p.i. under a fluorescence microscope using an eyepiece micrometer. The results were analyzed by using GraphPad Prism, version 5.01 (GraphPad Software, Inc.).
To compare the numbers of BV genome copies of Ac79KO-PG and AcWT-PG, the same number of infectious BV units (based on TCID50, 5 × 107 PFU) of each virus was concentrated by centrifugation, and genomic DNA was purified as previously described (22). Purified genomic DNA was resuspended in 50 μl distilled water, and 5 μl of DNA was used as the template for quantitative PCR (Q-PCR) as previously described (32).
The same number of infectious Ac79KO-PG or AcWT-PG units (5 × 107 PFU) was concentrated, and viral proteins were immunodetected with anti-GP64 antibody or anti-VP39 antiserum.
Sf9 cells (1.0 × 106 cells/35-mm-diameter dish) were infected in triplicate with Ac79KO-PG or AcWT-PG at an MOI of 5 and collected at different time points. Total DNA was prepared, and Q-PCR was used to determine viral DNA replication in virus-infected cells as previously described (32). The results were analyzed by using GraphPad Prism, version 5.01 (GraphPad Software, Inc.).
To evaluate the effect of ac79 on viral protein expression, Sf9 cells were infected with A79KO-PG or AcWT-PG at an MOI of 5. At different time points, cells were collected, and samples were prepared for immunoblotting as described above. VP39 antiserum and OpMNPV polyhedrin antiserum, which cross-reacts with AcMNPV polyhedrin, were used as primary antibodies to immunodetect representative late and very late proteins.
Sf9 cells were infected with Ac79HARep-PG at an MOI of 0.01. At 96 h p.i., the pellet of infected cells and the supernatant were separated by centrifugation. BV was concentrated and purified as described previously (22). To collect polyhedra, 200 third-instar Trichoplusia ni larvae were infected by the contamination of an artificial diet with Ac79HARep-PG polyhedra, which were prepared from Ac79HARep-PG-infected Sf9 cells. ODVs were purified as previously described (30).
Sf9 cells were infected with Ac79KO-PG or AcWT-PG at an MOI of 0.01. The supernatant of infected cells containing BV was collected at 120 h p.i. The purification of BV was carried out as previously described (22). For negative staining, an electron microscope copper grid was placed over 10 μl of purified BV to absorb virions. Grid-bound virions were stained with 10 μl of a 2% uranyl acetate solution in water. Stained virions were observed by using a Philips CM 100 transmission electron microscope at an accelerating voltage of 100 kV.
Sf9 cells (1.0 × 106 cells/35-mm-diameter dish) were infected with the indicated virus at an MOI of 5. Infected cells were collected at 72 h p.i., fixed, dehydrated, embedded, sectioned, and stained as described previously (17). Samples were viewed as described above.
For immunoelectron microscopy, cells were embedded in LR White resin (Ted Pella, Inc.), according to the technical notes provided by the manufacturer. VP39 in Ac79KO-PG-infected cells was detected with rabbit polyclonal VP39 antisera, and HA-tagged Ac79 was detected with anti-HA antibody in Ac79HARep-PG-infected cells. These primary antibodies and goat anti-rabbit immunoglobulin G 10-nm gold-conjugated secondary antibody (Sigma) were used at a dilution of 1:50. Samples lacking the primary antibody were used as background labeling controls. Images were obtained with a Philips CM 100 transmission electron microscope at an accelerating voltage of 100 kV.
The ac79 transcription initiation sites were determined by 5′ RACE analyses using total RNAs isolated at 6, 12, and 24 h p.i. from AcWT-PG-infected cells. 5′ RACE products (Fig. 1A) obtained from RNAs harvested at each time point were cloned, sequenced, and mapped to four transcription start sites (Fig. 1B, asterisks). At 6 and 12 h p.i., we detected 5′ RACE products that mapped to 55, 166, 167, and 168 nucleotides (nt) upstream of the predicted Ac79 translational start site (Fig. 1B). These sites were not reminiscent of early or late start sites (i.e., TATA box, initiator sequence, or TAAG). The larger 5′ RACE product observed at 24 h p.i. (Fig. 1A, triangle) maps to 18 nt upstream of the gp41 translation initiation codon and corresponds to the gp41 transcription start site (Fig. 1B). gp41 is present upstream of ac79 and is transcribed in the same direction.
Sf9 cells were infected with Ac79HARep-PG, a bacmid carrying an HA-tagged ac79 gene at the polyhedrin locus, and the production of Ac79 was analyzed by immunoblotting using an anti-HA antibody. An HA-immunoreactive band of approximately 16 kDa was detected at 6 h p.i., and detection continued through 96 h p.i. (Fig. 2A). The DNA synthesis inhibitor aphidicolin and the protein synthesis inhibitor cycloheximide were individually added to cells to assess if Ac79 synthesis required prior DNA replication. The synthesis of Ac79 was reduced but was detected in aphidicolin-treated cells; however, as expected, it was nearly abolished in the presence of cycloheximide (Fig. 2B), suggesting that late genes or DNA replication was not necessary for Ac79 synthesis. As a positive control, the same membrane was stripped from antibodies and reprobed with anti-IE-1 antibody to detect an early viral protein (Fig. 2B).
To determine the subcellular localization of Ac79 during infection, Sf9 cells were infected with Ac79HARep-PG and collected at various time points. Cells were biochemically fractionated into cytoplasmic and nuclear fractions, and proteins were detected following immunoblotting. Ac79 fractionated in both the cytoplasmic and nuclear fractions; however, it was more abundant in the nuclear fraction at late and very late phases (Fig. 2C). To assess the fractionation efficiency, the same membrane was reprobed with a GP64 or IE-1 antibody to detect a mainly cytoplasmic or nuclear protein, respectively (Fig. 2C).
We also examined whether Ac79 was associated with the BV or ODV. BV or ODV was prepared from the supernatant of Ac79HARep-PG-infected cells or whole larvae, respectively, and HA-tagged Ac79 was detected by immunoblotting with anti-HA antibody. As a positive control, Ac92HARep-PG (32) BV was used to detect an HA-tagged protein. Ac79 was associated with neither BV nor ODV (Fig. 2D). To verify the purification and loading of BV and ODV, IE-1 was detected in all samples by immunoblotting (Fig. 2D). This result indicates that Ac79 is not a virion-associated protein. In addition, Ac79 did not specifically localize within Ac79HARep-PG-infected cells (Fig. 2E).
Ac79KO was constructed by deleting the central portion of ac79 and retaining 115 nt from the 5′ end and 95 nt from the 3′ end of the ac79 coding region to ensure the expression of the neighboring genes ac78 and gp41 (ac80) (Fig. 3A). The remaining 105-nt coding sequence (nt 65385 to 65498 of AcMNPV ) was replaced with the Cm cassette (Fig. 3A).
The deletion of ac79 from the ac79 locus of AcBAC and the insertion of the Cm cassette were confirmed by PCR (Fig. 3B). Primers Ac7951 and Ac7931 (Fig. 3A) were used to confirm the deletion of 105 bp within the ac79 coding region and its replacement with the 1,038-bp Cm cassette. Primers pairs Ac7951/Cm3 and Cm5/Ac7931 (Fig. 3A) were used to confirm the recombination junctions upstream and downstream of ac79, respectively. Primer pair Cm5/Cm3 was also used to confirm the insertion of the Cm cassette. The sizes of the PCR-amplified products obtained were as predicted following successful recombination (Fig. 3B).
To facilitate an examination of the effects of the deletion of ac79 on virus infection, an ac79-knockout mutant, Ac79KO-PG, containing the polyhedrin gene under polyhedrin promoter control and egfp under ie-1 promoter control, was constructed by the transposition of the polyhedrin and egfp genes into the polyhedrin locus of Ac79KO (Fig. 3Di). Two repair bacmids were constructed to rescue and confirm the phenotype resulting from the deletion of ac79, Ac79Rep-PG, carrying the ac79, polyhedrin, and egfp genes (Fig. 3Dii), and Ac79HARep-PG, carrying the polyhedrin gene, egfp, and ac79 with an HA tag at the C terminus (Fig. 3Diii). AcWT-PG, AcBAC carrying the polyhedrin and egfp genes, was described previously (32) and used as a positive control (Fig. 3C). All constructs were confirmed by PCR, the expression of egfp, and the formation of occlusions (data not shown).
Sf9 cells were transfected with DNA from Ac79KO-PG, Ac79Rep-PG, Ac79HARep-PG, or AcWT-PG, and the expression of egfp was monitored by fluorescence microscopy. No obvious differences in egfp expression were observed from 24 to 72 h posttransfection (p.t.) among the viruses (Fig. 4A). An inspection of cells by light microscopy showed no differences in the approximate number of occlusion bodies formed and the timing of occlusion body formation at 48 h p.t. (data not shown) and 72 h p.t. (Fig. 4A).
Virus growth curve analyses were performed to assess the effect of the deletion of ac79 on virus replication. Cells were transfected with bacmid DNA, and at selected time points, BV titers were determined by TCID50 endpoint dilution assays. A steady increase in virus production was observed with each virus. However, the virus titer of Ac79KO-PG was lower than the titers of the other viruses, with obvious differences starting at 48 h p.i. (Fig. 4B). At 72 h p.t., the titer of Ac79KO-PG was 25-fold lower than that of Ac79HARep-PG. To confirm this defect, cells were infected with each virus at a low MOI of 0.01 or at a high MOI of 5, and BV titers were determined by TCID50 endpoint dilution assays at selected time points. The titer of Ac79KO-PG was again lower than those of any of the other viruses from about 48 to 96 h p.t. at both multiplicities of infection (Fig. 4C and D), and the onset of BV production was not drastically affected (Fig. 4B and D, insets). For example, Ac79KO-PG-infected cells at two different MOIs produced 10.5-fold-lower titers at 72 h p.i. (Fig. 4C) or 9.15-fold-lower titers at 48 h p.i. (Fig. 4D) than Ac79HARep-PG. Together, these data show a reduction rather than a delay in BV production in the absence of ac79.
To confirm the defect of BV production in the ac79-knockout virus and, thus, virus spread, the diameters of randomly selected plaques were measured at 72 h p.i. Plaques from AcWT-PG-infected cells were significantly larger than those from Ac79KO-PG-infected cells (P < 0.001) (Fig. 4E), supporting the defect of Ac79KO-PG in BV production.
We considered that the reduction in virus titers from cells infected with Ac79KO-PG compared to those from AcWT-PG-infected cells could be attributed to a reduction in the amount of infectious BV. To determine if Ac79KO-PG produced less infectious BV than AcWT-PG, we first used Q-PCR to determine the relative number of viral DNA copies produced during infection. To this end, viral DNA was purified from 5 × 107 PFU and used as a template for Q-PCR. Using the same units of infectious BV, more viral genomic DNA was amplified in Ac79KO-PG (Fig. 5A), suggesting that more virions were not infectious. Immunoblotting was performed using the same amount of infectious Ac79KO-PG- or AcWT-PG-purified BVs, and blots were probed with VP39 antiserum or GP64 antibody (Fig. 5B). The immunosignal was much stronger with proteins from Ac79KO-PG than that with proteins from AcWT-PG. This suggested that more noninfectious BV was produced in Ac79KO-PG- than in AcWT-PG-infected cells.
To evaluate if the lack of ac79 affected the structure of BV, BV was purified and prepared for visualization by transmission electron microscopy. The rod-shaped structures and sizes of Ac79KO-PG and AcWT-PG virions were comparable (Fig. 5C).
To assess whether viral DNA replication was affected by the deletion of ac79, Q-PCR was performed. Cells were infected with AcWT-PG or Ac79KO-PG at an MOI of 5 and collected at selected time points. Total DNA was purified and used as a template, amplifying gp41 by Q-PCR. DNA replication dynamics between the viruses showed no significant differences (Fig. 6), confirming that ac79 did not affect viral DNA replication.
To explore the effects of ac79 on late and very late viral protein synthesis, the levels of production of VP39 and polyhedrin from a virus lacking ac79 were compared to those from a virus carrying ac79. VP39 and polyhedrin were detected from 18 and 24 h p.i., respectively, in either Ac79KO-PG- or AcWT-PG-infected cells (Fig. 7). The results indicate that the deletion of ac79 did not affect the relative accumulation of VP39 and polyhedrin.
Thin sections from virus-infected cells collected at 72 h p.i. were visualized by electron microscopy to compare virion morphogenesis among the viruses. Cells infected with AcWT-PG showed typical characteristics of baculovirus replication, that is, the appearance of rod-shaped nucleocapsids in the electron-dense virogenic stroma and in bundle formations in the ring zone, microvesicles, and polyhedra containing embedded ODVs in the ring zone (Fig. 8A). Cells infected with Ac79Rep-PG or Ac79HARep-PG showed characteristics similar to those of AcWT-PG-infected cells (data not shown). In general, typical infection characteristics were also observed for Ac79KO-PG-infected cells (Fig. 8B); however, in contrast to cells infected with AcWT-PG, Ac79KO-PG-infected cells contained clusters of elongated tubular structures in the ring zone (Fig. 8B, arrows). These tubular structures appeared translucent, suggesting that they lacked DNA. To confirm whether these structures were capsid assemblages, VP39 antiserum and gold-conjugated antibodies were applied prior to transmission electron microscopy visualization. The tubular structures in Ac79KO-PG-infected cells reacted specifically with VP39 antiserum (Fig. 8C and D), suggesting that these structures were capsid-like structures. From this experiment, it is not clear if the presence of noninfectious BV was related to these tubular structures.
Iterative database searches revealed significant similarities among motifs present in UvrC, intron-encoded endonucleases, and other proteins, including Ac79 (1). This motif consists of RX3[YH], tyrosines spaced by 10 amino acids, and a glutamic acid and is referred to as the Uri motif (UvrC and intron-encoded endonucleases). The tyrosines are present in the GIY-YIG superfamily of intron-encoded endonucleases at the N terminus and are part of the catalytic domain (25). The arginine and glutamic acid corresponding to amino acids 34 and 72, respectively, in Ac79 are important for intron-encoded endonuclease catalysis (9, 15). Intron-encoded endonucleases and UvrC endonucleases were suggested previously to have the same lineage (10).
The alignment of baculovirus Ac79 orthologs shows that the key amino acids within the Uri motif are conserved (Fig. 9A). The GIY-YIG sequence present in intron-encoded endonucleases is not completely conserved in baculovirus Ac79 orthologs, which lack the first glycine (Fig. 9A). In Ac79KO-PG, the conserved glutamic acid was deleted, but the rest of the Uri motif was not (Fig. 9A, top line following the alignment). We constructed three recombinant bacmids, Ac79Y24AG26A-PG, Ac79R34K-PG, and Ac79E72D-PG, with mutations in the Uri motif of Ac79 (Fig. 9A, bottom). These mutations include two residues in the GIY-YIG-corresponding sequence, arginine 34 and a conserved glutamic acid at amino acid 72. These viruses expressed the altered Ac79 proteins following the transfection of bacmid DNA into Sf9 cells (Fig. 9B). To assess whether the mutations affected virus replication, virus growth curve analyses were performed. Cells were transfected with bacmid DNA, and BV titers were determined by TCID50 endpoint dilution assays at 24, 48, and 72 h p.t. A steady increase in virus production was observed with each virus. However, the virus titer of Ac79E72D-PG was lower throughout the time course than that of other viruses (Fig. 9C), similar to the reduction observed for Ac79KO-PG-infected cells. Ac79Y24AG26A-PG and Ac79R34K-PG were comparable to the control virus, Ac79HARep-PG (Fig. 9C).
Transmission electron microscopy analysis was also performed to determine whether the mutations affected the virus structure. Ac79Y24AG26A-PG, Ac79R34K-PG, or Ac79E72D-PG DNA-transfected cells showed similar ultrastructural characteristics of baculovirus replication (Fig. 9D). No elongated tubular structures were found in the ring zone of the transfected cells.
In recent years, the number of studies of the function of conserved baculovirus genes has increased with the development of bacmid technology, facilitating the generation of viruses null for genes of interest and evaluations of their phenotypes in infected or bacmid DNA-transfected cells. We sought to characterize ac79 since it is conserved in many baculoviruses, and orthologs are found in other viruses and bacteria. ac79 is not an essential gene for virus replication, and viruses lacking ac79 were able to produce infectious BVs (Fig. 4) and ODVs that were infectious in insects (data not shown). However, it is required for efficient levels of infectious BV production, since the deletion of ac79 resulted in about a 10-fold decrease in levels of infectious BV. Since ac79 is not conserved in all baculovirus genomes, its function may be partially redundant with that of other viral or host genes, or it may provide an advantage to virus replication in some hosts.
ac79 is an early gene based on its expression being independent of DNA replication, its timing of expression, and its transcription initiation being at a site other than a late or very late consensus start site. ac79 does not appear to affect late or very late protein accumulation, although we did not test if the onset of gene expression at any phase was affected. Ac79 was detected from 6 to 96 h p.i. (Fig. 2A). It localized to the nucleus and cytoplasm (Fig. 2C) but accumulated in the nucleus as the time course of infection progressed, suggesting a nuclear-specific role during virus replication. The appearance of tubular structures containing capsid proteins in the nucleus of cells infected with Ac79KO-PG supports a nuclear role.
Viruses lacking ac79 produced reduced levels (about 10-fold) of infectious BV; however, viral DNA replication and late and very late gene expression appeared to be unaffected, suggesting the completion of the replication cycle. In addition, the timing of the onset of infectious BV production was similar for all viruses. Together, these observations suggest that Ac79 may have a role in facilitating events after viral DNA replication, such as proper capsid assembly or virion maturation, the efficient transport of nucleocapsids from the nucleus to the cytoplasm, or virus budding from the cell.
Previous studies showed that the deletion of the AcMNPV gp64, ac17, exon0, ac66, me53, or pp31 gene results in a reduction in the level of BV production but that viral DNA replication remains unaffected (7, 8, 11, 14, 19, 21, 33). Although these mutants share similar phenotypes, the specific functions of each gene in the process of BV production may differ. Defects in BV production could be due to defective virus egress; impaired nucleocapsid structure integrity; or alterations in the transport of nucleocapsids within the cell, packaging, or virus assembly. The lack of ac79 did not appear to affect the egress of BV from the cell or the transport of nucleocapsids from the nucleus to the cytoplasm, since infectious and noninfectious virions were released (Fig. 4 and and5).5). In addition, consistent with the finding that Ac79 was not a structural component of the BV or ODV (Fig. 2D and E), we found that the lack of ac79 did not affect the gross morphology of budded virions released from the cells (Fig. 5C). This is supported by previous studies that did not identify Ac79 as a component of BV by proteomic methods (28).
Ac79KO-PG-infected cells produced tubular VP39-containing structures along the inner nuclear membrane. Interestingly, previous studies that described mutations in the ac53, 38K, vlf-1, and alkaline nuclease genes also described similar tubular sheaths (18, 20, 24, 31). This phenotype has been associated with nucleocapsid transport, nucleic acid resolution, or packaging defects. In addition, capsid protein-containing tubular structures have been observed following treatment with cytochalasin D, a microfilament elongation inhibitor, even though viral DNA synthesis was not affected (27). The interference of cytochalasin D with capsid assembly indicated that microfilaments were involved in this nuclear process (27). Given that defects in different genes result in similar phenotypes, it makes it difficult to determine if ac79 functions in any of these processes or has another function.
A previous study suggested that Ac79 may be related to bacterial DNA repair UvrC excision endonucleases and intron-encoded endonucleases, based on the presence of the Uri motif (1). To explore this further, we compared the Ac79 peptide sequence to sequences in protein structure databases using HHpred v 2.0 (4). The results showed significant predicted structural similarities between Ac79 and UvrC, bacteriophage T4 endonuclease II, and the I-TevI intron-encoded endonucleases with the GIY-YIG domain (data not shown), indicating structural parallels between these endonucleases and Ac79. It is not known if Ac79 binds DNA, a function that is localized at the C terminus of GIY-YIG endonucleases (25), or if it has endonuclease function; however, structural similarities, along with the presence of key residues important for nuclease function, suggest that Ac79 is related to these endonucleases.
To test the importance of residues conserved between Ac79 and GIY-YIG-containing endonucleases, we constructed viruses with mutations in the conserved GIY-YIG-corresponding motif (Ac79 amino acids Y24 and G26) or in residues predicted to participate in endonucleolytic catalysis (Ac79 amino acids R34 and E72). None of the mutants showed tubular capsid-like structures similar to those observed in Ac79KO-PG-infected cells. It is possible that the elongated structures were caused by the presence of the undeleted N-terminal fragment of Ac79 (amino acids 1 to 38) in Ac79KO-PG, which contained the conserved tyrosines and the RX3H sequence, which may have hindered the activity of cellular or viral proteins necessary for proper nucleocapsid formation. However, this N-terminal peptide does not have a dominant negative function, since it is also present in the amino acid point mutants. Among the viruses with point mutations in Ac79, only the virus with the conservative E72D change showed reduced BV production. This glutamic acid was also deleted in Ac79KO-PG, suggesting a role for this amino acid in productive virus infection. Curiously, we did not observe reduced BV production when Y24 and G26 or R34 were mutated. It is possible that these mutations were repaired by recombination events, with the N-terminal 38 amino acids remaining at the ac79 locus, even though our experiments were carried out with transfected bacmid DNA to minimize homologous recombination events during reiterative virus replication cycles. Further work is needed to determine the requirement of Y24, G26, and R34 in infectious BV production. It appears that the tubular capsid protein-containing structures observed in Ac79KO-PG-infected cells are not required for the defect in infectious BV production. Although the E72D mutation suggests that endonucleolytic activity is important for infectious BV production, additional experiments will be required to further define this role.
Mutations in baculovirus DNA replication and processing genes result in altered capsid protein-containing structures, suggesting that viral DNA affects the nucleocapsid architecture (23). We tested whether there is an interaction between Ac79 and VLF-1, which is involved in viral genome processing, hypothesizing that Ac79 may provide the endonuclease activity needed during genome processing and packaging, but we could not detect an interaction by coimmunoprecipitation (our unpublished data). Although Ac79KO-PG produces elongated capsid protein-containing structures, it also produces normal nucleocapsids in both occluded and budded virions. It is premature to speculate whether Ac79 functions similarly to UvrC, cleaving phosphodiester bonds during DNA repair, or to T4 endonuclease II, degrading host DNA to reutilize nucleotides for its DNA synthesis, or whether it functions in creating double-strand breaks characteristic of a homing endonuclease like I-TevI endonuclease. Further work is needed to determine if Ac79 has endonucleolytic activity and to further define its role during BV production.
This research was supported by U.S. Department of Agriculture award 2008-35302-18849.
We thank Rollie Clem for valuable discussions.
Published ahead of print 14 March 2012
This article is contribution 11-401-J from the Kansas Agricultural Experiment Station.