Conditional Cytomegalovirus Replication In Vitro a
http://www.100md.com
病菌学杂志 2005年第1期
Max von Pettenkofer Institut für Virologie, Ludwig-Maximilians-Universit?t München, Munich, Germany
ABSTRACT
We have established a conditional gene expression system for cytomegalovirus which allows regulation of genes independently from the viral replication program. Due to the combination of all elements required for regulated expression in the same viral genome, conditional viruses can be studied in different cell lines in vitro and in the natural host in vivo. The combination of a self-sufficient tetracycline-regulated expression cassette and Flp recombinase-mediated insertion into the viral genome allowed fast construction of recombinant murine cytomegaloviruses carrying different conditional genes. The regulation of two reporter genes, the essential viral M50 gene and a dominant-negative mutant gene (m48.2) encoding the small capsid protein, was analyzed in more detail. In vitro, viral growth was regulated by the conditional expression of M50 by 3 orders of magnitude and up to a millionfold when the dominant-negative small capsid protein mutant was used. In vivo, viral growth of the dominant-negative mutant was reduced to detection limits in response to the presence of doxycycline in the organs of mice. We believe that this conditional expression system is applicable to genetic studies of large DNA viruses in general.
INTRODUCTION
Temperature-sensitive (ts) mutants, the first generation of conditional mutants in virology, have been helpful and often superior to null mutants in mapping and identifying viral functions (26). Random procedures for the generation of ts mutants have been described for alphaherpesviruses (29). However, this method did not lead to major findings in betaherpesviruses, which have as their most prominent member human cytomegalovirus (HCMV), an important human pathogen. Propagation of conditional betaherpesvirus mutants by selection of ts alleles is technically difficult, and a specific gene has not been assigned for any of the reported ts mutants (1). However, one ts allele was generated recently for the UL122 gene of HCMV by rational mutagenesis (11).
Conditional gene expression is desirable for the study of gene functions as well as for the development of vaccines and gene therapy vectors both for safety reasons and for the opportunity to adjust the amount of the delivered gene product. As a conditional principle, cell lines expressing tet regulators have been used to control tetracycline-responsive promoters in the herpesvirus genome in order to modulate reporter gene expression (14, 18) or viral growth (33, 34). Recently, all elements required for tet-regulated growth were placed into the genomes of adenovirus (25) and human immunodeficiency virus (30). In these examples, the targeted viral promoters control major regulators of the viral genetic program in order to achieve general inhibition of virus replication. In the present study, we intended to regulate individual genes during morphogenesis and maturation. This goal requires that the gene of interest is regulated as independently as possible from the viral genetic program.
Here, we have established a conditional system for cytomegalovirus in which both the regulator and the regulated transcription unit are constructed in one cassette, which is integrated into the viral genome. The cassette was inserted at a predefined intergenic locus, which appears to be neutral in vitro and in vivo (5). Thus, regulation of the target gene should not necessarily affect the viral program of gene expression. We show that this conditional system has the capacity to regulate the activity of the luciferase reporter about 80-fold in the context of viral replication. In order to study the regulation of viral genes, we subjected two genes to conditional expression: (i) the M50 gene, an essential gene of murine cytomegalovirus (MCMV) (4); and (ii) a dominant-negative (DN) mutant gene (m48.2) encoding the small capsid protein (SCP) (3). The conditional M50 recombinant virus was regulated in vitro at least 1,000-fold; the DN recombinant virus was regulated in vitro 100- to 1,000,000-fold depending on the cell line. In vivo, the productive growth of the DN virus was reduced to the detection limit upon doxycycline (DOX) administration.
MATERIALS AND METHODS
Cells and viruses. BALB/c mouse embryonic fibroblasts (MEF), M2-10B4 cells (ATCC CRL-1972) and SVEC 4-10 cells (ATCC CRL-2181) were prepared and treated as described previously (19). Mouse mammary epithelial C127 cells (ATCC CRL-1616) were maintained in Dulbecco modified Eagle medium supplemented with 10% fetal calf serum. The MCMV wild-type (wt) and wt-FRT viruses used in this study were derived from the bacterial artificial chromosomes (BACs) pSM3fr (32) and wt-FRT-MCMV (5), respectively. MCMV BACs were reconstituted to viruses by transfection of BAC DNA into MEF by use of SuperFect transfection reagent (QIAGEN) according to the manufacturer's instruction. To induce expression of the conditional M50 gene, 1 μg of DOX/ml was added every third day for the reconstitution of the M50-1 virus. Virus stocks were propagated in M2-10B4 cells and quantified on MEF by plaque assay as described previously (17, 24). After the removal of cell debris, the wt, GFP-1, GFP-2, SCP-6FP-1, SCP-GFP2, and M50-1 virus stocks were purified by pelleting them (23,450 x g for 3 h), followed by an ultracentrifugation step over a 15% sucrose cushion. In the case of M50-1, virus stocks were purified additionally by using an OptiPrep gradient (Sigma) according to the manufacturer's instruction.
Growth analysis. For determination of in vitro growth of MCMV mutants, MEF, SVEC 4-10, or C127 cells were infected at a multiplicity of infection (MOI) of 0.1 or 1 PFU per cell. The inoculum was removed after 1 h, and normal medium or medium with 1 μg of DOX/ml was added. Supernatants of infected cells were harvested daily from day 0 (input virus) until day 5 postinfection (p.i.). On day 3, DOX was added a second time. Supernatants of samples were taken in duplicate and titrated on MEF.
Plasmid construction. (i) Construction of plasmids with the inducible promoter PCMV-tetO2. PCMV-tetO2 is composed of the HCMV immediate-early promoter-enhancer (PCMV) and two tet operators (tetO2) downstream of the TATA element (35). In order to place the respective open reading frame (ORF) under control of the inducible promoter PCMV-tetO2, the luciferase ORF was excised from pGL3-Control (Promega) by using HindIII/XbaI and inserted into pcDNA4/TO (Invitrogen), giving rise to pIET-Luc. The conditional expression unit of pIET-Luc was excised by PvuII/MluI, filled in by Klenow polymerase, and inserted into the vector pOriR6K-zeo (GenBank accession no. AY700021) (4), which was linearized by EcoRV, giving rise to pO6-IET-Luc. For the generation of the enhanced green fluorescent protein (EGFP) expression vector, the EGFP ORF was amplified by PCR with primers gfpCN and gfpCE from pEGFP-C1 (Clontech), cut with AflII/ApaI, and inserted into pcDNA4/TO linearized by AflII/ApaI, giving rise to pIET-gfp. For cloning of the DN acting SCP-GFP fusion gene, the EGFP ORF was amplified by PCR from pEGFP-C1 with primers gfpCN and gfpCF and cut with AflII/SpeI. The SCP ORF corresponding to nucleotide (nt) positions 73445 to 73864 in the MCMV genome (23) was amplified from pSM3fr with primers SCPforF and SCPrev and cut with SpeI/ApaI. Both fragments were inserted in one step into pcDNA4/TO linearized with AflII/ApaI, giving rise to pIET-gfpSCP. The inserts from pIET-gfp and pIET-gfpSCP were excised and transferred into pOriR6K-zeo-ie (GenBank accession no. AY700022) (4) by use of NdeI/ApaI, giving rise to pO6-IET-gfp and pO6-IET-gfpSCP, respectively.
In order to insert the tetR gene into the pO6-IET constructs, the tetR expression unit was excised from pcDNA6/TR (Invitrogen) by using MluI and treated with T4 DNA polymerase. This fragment was inserted into pO6-IET-Luc, which was cut with BamHI and filled in, giving rise to pO6-IET-Luc-TR. The same fragment was inserted into pO6-IET-gfp and pO6-IET-gfpSCP, which were linearized by SphI and treated with T4 DNA polymerase, giving rise to pO6-IET-gfp-TR and pO6-IET-gfpSCP-TR, respectively. For the construction of the insertion vector pO6-IET-M50-TR, the M50 ORF was excised from pOriR6K-zeo-ie-M50 (4) with KpnI/XhoI, treated with T4 DNA polymerase, and inserted into pO6-IET-Luc-TR instead of the luciferase ORF, which was removed by a partial digestion with PmeI.
(ii) Construction of plasmids with the inducible promoter PSV40-tetO2. The simian virus 40 (SV40) early enhancer was amplified from pcDNA4/TO by PCR with primers SV40for and SV40rev, and the product was cut with SpeI/NcoI. The minimal CMV promoter including tetO2 was amplified by PCR from pcDNA4/TO with primers miniPfor and miniPrev, and the product was cut with NcoI/AflII. These fragments were inserted in one step into the vector LIT-MUS28 (New England BioLabs) cut with SpeI/AflII, giving rise to pL-SVT.
The vector pOriR6K-zeo was modified by insertion of the hybridized oligonucleotides MCS3 and MCS4 into the KpnI/BamHI sites, giving rise to pOriR6K-zeo-Asc. The tetR gene was excised from pcDNA6/TR by MluI and inserted into pOriR6K-zeo-Asc linearized by AscI, giving rise to pO6-TR.
The luciferase ORF and the bovine growth hormone poly(A) was excised from pIET-Luc and inserted into pL-SVT by AflII/PvuII, giving rise to pSVT-Luc. The conditional SV40-derived expression unit was excised from pSVT-Luc by PvuII/SnaBI and inserted into pO6-TR linearized by EcoRV, giving rise to pO6-SVT-Luc-TR. For construction of pO6-SVT-gfp-TR and pO6-SVT-gfpSCP-TR, the EGFP and SCP-GFP ORFs were excised from pO6-IET-gfp and pO6-IET-gfpSCP, respectively, by PmeI/ApaI, treated with T4 DNA polymerase, and inserted into pO6-SVT-Luc-TR instead of the luciferase ORF, which was removed by partial digestion with PmeI. The sequential steps for the construction of the plasmids used in this study are illustrated in a figure (see Fig. S1 in the supplemental material). The primers used are listed in Table 1.
Generation of recombinant viral BACs. Escherichia coli strain DH10B (Invitrogen) containing the wt-FRT-MCMV BAC and the temperature-sensitive Flp recombinase expression plasmid pCP20 (7) was transformed with the insertion vectors pO6-IET-Luc, pO6-IET-Luc-TR, pO6-SVT-Luc-TR, pO6-IET-gfp-TR, pO6-SVT-gfp-TR, pO6-IET-gfpSCP-TR, and pO6-SVT-gfpSCP-TR and treated as described previously (4). For insertion of pO6-IET-M50-TR, the M50-FRT-MCMV BAC was used, which was derived from the wt-FRT-MCMV BAC by replacement of most of the M50 ORF (nt 75504 to 76451 according to Rawlinson et al. [23]) with a kanamycin resistance cassette by use of a linear fragment amplified with primers 5'-M50 and 3'-M50 (4) as described previously (31). The generated BACs were used for reconstitution to viruses Luc-0, Luc-1, Luc-2, GFP-1, GFP-2, SCP-GFP-1, SCP-GFP-2, and M50-1, respectively.
Luciferase assay. M2-10B4 cells were infected with Luc-0, Luc-1, and Luc-2 at an MOI of 0.1 in the absence or presence of a 1-μg/ml concentration or a serial dilution of DOX, and the cells were harvested after 1 to 4 days. The cells were washed in phosphate-buffered saline and lysed in 100 μl of 1x reporter lysis buffer (Promega). Cell debris was removed by centrifugation. The activity of luciferase was determined with the luciferase assay reagent (Promega) according to the manufacturer's instructions by using dilutions of the lysates (1/100 to 1/20,000 of the 100-μl volume) in order to be in the linear range of detection. The concentration of proteins in the cell lysates was determined with the BCA protein assay kit (Pierce) according to the manufacturer's instructions. Induction levels were calculated for each pair by dividing the normalized luciferase activity in the presence of DOX by that in the absence of DOX.
Western blot analysis. M2-10B4 cells were infected in the absence or presence of 1 μg of DOX/ml with Luc-2 at an MOI of 0.1 for 1 to 4 days, or for 1 day with M50-1 and GFP-1 at an MOI of 0.1 with centrifugal enhancement (700 x g for 30 min), which results in about a 20-fold enhancement of infectivity, i.e., an effective MOI of ca. 2. Cells were lysed in reporter lysis buffer (Promega) for detection of the TetR and in lysis buffer (20 mM Tris, 150 mM NaCl, 1% Triton X-100, freshly supplemented with 1% protease inhibitor cocktail [Roche]) for detection of M50/p35, major capsid protein (MCP), immediate-early protein 1 (IE1), and cellular actin. Lysed samples were separated in a sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel containing 7.5% (IE1 and MCP), 12.5% (TetR), or 15% (M50 and actin) polyacrylamide and transferred to Hybond-P membranes (Amersham). The membranes were treated with polyclonal rabbit antisera specific to TetR (MoBiTec), actin (aa20-33) (Sigma), or M50 (20), a monoclonal mouse antibody to viral IE1 protein pp89 (CROMA101) (provided by S. Jonjic, University of Rijeka, Rijeka, Croatia), and a polyclonal rat antiserum against MCP, which was raised against a synthetic peptide with the amino acid sequence RIQQSSQKDLPESQF (Eurogentec). For immunodetection, horseradish peroxidase-coupled anti-mouse, anti-rat, and anti-rabbit immunoglobulin-specific antisera (Dianova) were used, followed by ECL-Plus (Amersham).
Fluorescence microscopy. MEF, SVEC 4-10, or C127 cells were infected with SCP-GFP-2 at an MOI of 0.1 for 5 days without or with the addition of 1 μg of DOX/ml on day 0. DOX was added again on day 3 (1 μg/ml in the case of MEF or 0.5 μg/ml in the case of SVEC 4-10 or C127 cells). The green fluorescence and the bright-field or phase-contrast images were collected with an Axiovert 200 M system (Zeiss).
Animal experiments. Female BALB/c mice (6 to 10 weeks old) were kept under specific-pathogen-free conditions during the experiments. The animal experiments were approved by the responsible state office (approval no. 211-2531-38/99). Half of the animals received DOX (2 mg/ml in a 5% sucrose solution) in drinking water starting 7 days before infection. DOX was replaced every second day until the end of the experiment. Mice were infected with 5 x 105 PFU of virus intraperitoneally and sacrificed 5 days after infection. The harvested organs were homogenized, and the viral load was determined by a plaque assay using MEF with centrifugal enhancement, which results in a 20-fold enhancement of infectivity.
RESULTS
Design of the regulation cassettes and construction of MCMV recombinants. To regulate gene expression in MCMV, we used the elements of the Tn10-encoded Tet repressor (TetR) and its binding sequence, the tet operator (tetO). TetR proteins bind to operator sequences positioned within the promoter of the gene of interest (Fig. 1A) (35), thereby interfering with the transcription initiation complex. TetR binding is controlled by tetracycline or its analogue, DOX, which causes a reduction in the affinity of the TetR for the operator sequence, thus leading to the release of the transcriptional block (Fig. 1A). The regulation cassettes, which were designed to carry all elements required for regulation in order to be independent of special cell lines, were each composed of two expression units: (i) the constitutively expressed regulator, i.e., the tetR under control of the HCMV major immediate-early enhancer-promoter (PCMV), and (ii) the inducible transcription unit containing two minimal tet operators (tetO2). First, the PCMV-tetO2 (35) was used as an inducible promoter in the regulation cassette. This promoter carries two minimal tet operators (tetO2) inserted 10 bp downstream of the TATA element of the PCMV. In addition, to reduce the basal activity of the PCMV-tetO2 construct, we replaced the HCMV major immediate-early enhancer with the SV40 early enhancer, resulting in PSV40-tetO2. These inducible promoters were analyzed by expression of either luciferase, M50, EGFP, or SCP-GFP, a fusion protein of EGFP and SCP (the m48.2 gene product) of MCMV, which is traceable by its green fluorescence and has been claimed to inhibit viral growth in a DN manner (3). In order to generate conditional MCMV mutants, the plasmids carrying the regulation cassettes for luciferase, EGFP, or SCP-GFP were inserted into the MCMV BAC containing an Flp recombination target (FRT) site in the region between ORFs m16 and m17 (5) (Fig. 1B). The plasmid containing the M50 regulation cassette was inserted into the M50-FRT-MCMV BAC, in which the authentic M50 ORF had been deleted (Fig. 1C). The recombinant MCMV BACs containing the respective regulation cassettes were used for the reconstitution of MCMV mutants.
Regulated expression of the luciferase reporter gene by MCMV. The regulatory capacities of the different constructs within the viral genome were tested by luciferase expression in infected M2-10B4 cells. In cells infected at an MOI of 0.1 with Luc-1 (PCMV-tetO2-controlled luciferase) and Luc-2 (PSV40-tetO2-controlled luciferase) (Fig. 1B), the reporter gene expression was induced in the presence of 1 μg of DOX/ml 11- and 20-fold on day 1 p.i., and induction levels increased on day 2 p.i. 29- and 51-fold, respectively (Fig. 2A). The luciferase expression did not respond to DOX in cells infected with the control virus Luc-0, which lacks the tetR gene. The induction levels were similar for Luc-1 and Luc-2 after infection at a 10-fold-higher MOI (MOI of 1), namely, 9- and 24-fold on day 1 p.i. and 29- and 68-fold on day 2 p.i., respectively (data not shown). Luc-2 with its superior inducibility was used for further analysis. Luc-2 expressed about one-third of the luciferase activity expressed by Luc-0 after induction by DOX (3,325 ± 468 versus 10,036 ± 5,493 relative light units [RLU] per μl [mean ± standard deviation] for a 1:200 dilution of each sample) at an MOI of 0.1 on day 1 p.i., while the basal activity of Luc-2 in the noninduced stage was very low (171 ± 49 and 933 ± 299 RLU/μl for a 1:200 dilution of each sample on day 1 and day 2 p.i., respectively). In Luc-2-infected cells, the maximal induction levels (60- to 100-fold) were observed on days 3 and 4 p.i. (Fig. 2B). In the same cell lysates, which were used for measurement of the luciferase activity, TetR protein was detectable starting from day 3 p.i. (Fig. 2C). Luciferase activity above the background level was seen at a DOX concentration of 0.5 to 1 ng/ml with a half-effective dose of 2 ng/ml on days 1 and 2 p.i. (Fig. 2D). Thus, in M2-10B4 cells, a conditional reporter activity was detectable for several days, reflecting repeated MCMV replication cycles.
Since in this conditional system all elements required for regulation are combined in one viral genome, the conditional gene expression could be analyzed in different cell lines. The regulation of the luciferase expression was studied in the endothelial SVEC 4-10 cell line and the epithelial C127 cell line (Fig. 2E). The induction levels in Luc-2-infected SVEC 4-10 cells were similar to those in M2-10B4 cells: 13- and 46-fold on days 1 and 2 p.i., respectively. Yet, in C127 cells, the induction levels were lower and reached only 3-fold and 11-fold on days 1 and 2 p.i., respectively. The low induction levels in these cells was not due to an altered susceptibility to DOX since the maximal induction levels were reached in both cell lines at a DOX concentration of about 20 ng/ml on day 2 p.i. (Fig. 2F).
Regulated expression of the essential M50 gene. In order to apply the conditional system to a gene essential for virus replication, we used M50 as an example (4). M50 homologues are conserved among the Herpesviridae (6, 23). Inactivation of the homologous genes in certain -herpesviruses severely affects viral growth in cell culture, most probably because of their crucial role in the nuclear egress of viral capsids (15, 21, 28). For the M50 gene product, a similar role has been proposed (20). The M50 regulation cassette was tested in the context of a BAC-derived MCMV genome lacking the authentic M50 ORF (Fig. 1C). Since the essential M50 gene is required for viral propagation, the reconstitution of M50 mutants had to be carried out in the presence of DOX. The standard purification protocol was extended by an additional gradient ultracentrifugation step in order to remove residual trace amounts of DOX from M50-1 virus stocks. M50-1 and control viruses were tested in primary fibroblasts under multistep growth conditions. In the absence of DOX, the growth of M50-1 was reduced at least 3 orders of magnitude, whereas the addition of 1 μg of DOX/ml rescued the growth phenotype (Fig. 3). The control viruses grew similarly irrespective of the presence of DOX.
The regulation of the M50 gene at the ectopic position should affect virus replication by an inducible block of the capsid export from the nucleus (20) but should not affect the normal viral gene expression cascade. Therefore, we tested M50-1 and GFP-1-infected M2-10B4 cells for viral protein expression. The cells were infected at an effective MOI of 2 (MOI of 0.1 with centrifugal enhancement) in the absence and presence of 1 μg of DOX/ml in order to infect all cells simultaneously. Cell lysates were analyzed by Western blotting on day 1 p.i. (Fig. 4). Notably, M50/p35 protein was not detectable in the cells infected with M50-1 in the absence of DOX, indicating a tight regulation (Fig. 4, lane 3). In M50-1-infected cells in the presence of DOX (Fig. 4, lane 4) and in GFP-1-infected cells irrespective of the presence of DOX (lane 1 and 2), M50/p35 was present. We then analyzed the expression of the MCP, which is produced in large amounts in the late phase of viral replication. The degree of MCP expression did not vary in response to DOX, neither in M50-1- nor in GFP-1-infected cells. Comparable protein amounts were also detected for the viral immediate-early protein 1 (IE1), which is found during all phases of viral gene expression, both as full-length pp89 and smaller posttranslational modification products (13).
Thus, the conditional system could be applied to an essential viral gene in order to establish a DOX-dependent growth phenotype. In presence of DOX, the M50-1 virus grew in a manner that was indistinguishable from that of wt and GFP control viruses. There was no detectable effect on the expression of other viral genes.
Regulated expression of DN SCP-GFP. The DN principle is a general tool in cell biology, and we applied it in order to control the growth of MCMV. The SCP encoded by UL48/49 of HCMV (2, 10) and m48.2 of MCMV is essential for viral replication (3). The homologous protein of HSV-1, VP26, is not essential, and fusion to the GFP resulted in viable and infectious virions (8, 9). Recently, it was shown that the constitutive expression of the fusion protein of SCP with EGFP (SCP-GFP) prevents reconstitution of CMV even in the presence of the authentic SCP gene (3). Therefore, we tested the DN function by conditional expression of the fusion gene in the context of a BAC-derived MCMV containing all wt MCMV genes (Fig. 1B). The reconstitution of SCP-GFP mutants was carried out in the absence of DOX in order to prevent the expression of DN SCP-GFP protein. Interestingly, both conditional constructs, SCP-GFP-1 and SCP-GFP-2, were reconstituted like the wt MCMV BAC.
First, we studied the conditional replication of the SCP-GFP mutants in vitro. In the absence of DOX, the titers of BAC-derived wt virus, GFP-1, and SCP-GFP-2 revealed no difference in growth on MEF under multistep growth conditions (MOI of 0.1) (Fig. 5A). In contrast, the presence of 1 μg of DOX/ml completely inhibited growth of the SCP-GFP-2 virus (Fig. 5A) but not of control viruses. The SCP-GFP-1 virus was regulated similarly, yet there were infectious particles released in the supernatant during the observed period (Fig. 5B). Under single-step growth conditions (MOI of 1) in the absence of DOX, the controls and SCP-GFP-2 grew likewise indistinguishably (Fig. 5C). Under this condition, the growth of SCP-GFP-2 was reduced by a factor of 1,000 in response to DOX. The inhibition of virus replication (MOI of 0.1) was DOX concentration dependent (Fig. 5D): concentrations of at least 10 ng of DOX/ml resulted in an almost complete inhibition of SCP-GFP-2 growth, while 1 ng of DOX/ml did not affect the growth of the mutant. Thus, the insertion of a regulation cassette containing the DN SCP-GFP into MCMV resulted in regulation of viral growth of over 6 orders of magnitude. The dose response of SCP-GFP-2 to DOX was comparable to that of Luc-2 (Fig. 2D).
The green fluorescence of SCP-GFP revealed both inhibitor expression and virus spread. The SCP-GFP fusion proteins accumulated in the nucleus, leading to a punctate staining as described previously (3). In infected MEF (Fig. 6a to d), a weak fluorescent signal was detected for SCP-GFP in the absence of DOX (Fig. 6a), indicating a low basal expression. The signal was clearly increased in the presence of 1 μg of DOX/ml (Fig. 6b). Importantly, induction of SCP-GFP was associated with a lack of plaque formation (Fig. 6d).
Epithelial and endothelial cells are major targets of MCMV (16, 22). Therefore, we studied the conditional SCP-GFP mutants in the endothelial cell line SVEC 4-10 (Fig. 6e to h) and the epithelial cell line C127 (Fig. 6i to l). The SCP-GFP expression was induced by DOX in both cell lines (Fig. 6f and j). Despite basal expression of SCP-GFP in the absence of DOX (Fig. 6e and i), viral plaques were formed (Fig. 6g and k). Comparably to the situation observed with MEF (Fig. 6d), no plaques were detected in infected SVEC 4-10 cells in the presence of DOX (Fig. 6 h). Yet, in infected C127 cells, viral plaques were seen even in the presence of DOX (Fig. 6l), although they were smaller than in the absence of DOX. In order to correlate the different expression levels of SCP-GFP with the release of infectious viral particles from cells, the supernatants of infected SVEC 4-10 and C127 cells were titrated. No infectious particles were detected in the supernatants of SVEC 4-10 cells infected with SCP-GFP-1 or SCP-GFP-2 in the presence of DOX (Fig. 6m). In contrast, both recombinants replicated in the epithelial cell line C127 in the absence as well as in the presence of DOX (Fig. 6n). However, there was a DOX-mediated reduction in viral titers of about 100-fold. Notably, the DOX-mediated induction of the luciferase activity in C127 cells was also not as high as it was in M2-10B4 or SVEC 4-10 cells (Fig. 2E), which was not due to a different DOX susceptibility of C127 cells (Fig. 2F). Additionally, the expression of SCP-GFP in C127 cells was inducible by DOX, as detected by the increase in the punctate GFP fluorescence in the nuclei of infected cells (Fig. 6j).
For an inhibitory effect, a certain level of DN proteins is required in relation to their wt counterparts. Since there is no antibody available against SCP of MCMV yet, we analyzed the expression ratio of SCP and SCP-GFP genes in infected SVEC 4-10 and C127 cells by semiquantitative reverse transcription-PCR. In repeated experiments, the normalized SCP-GFP mRNA was found to be induced by DOX 10- to 15-fold in SVEC 4-10 cells and 4- to 6-fold in C127 cells, and the ratio between the SCP-GFP- and SCP-specific signals was higher in SVEC 4-10 cells than in C127 cells (data not shown).
Regulated virus replication in vivo. The differences in induction levels of the luciferase activity and the growth of the SCP-GFP-2 virus reflect that the efficiency of the tet regulation is dependent on the cell line. However, to study the biological relevance of viral functions, efficient regulation in the natural host is also required. Therefore, we analyzed the growth of the conditional lethal recombinant MCMV in mice. BALB/c mice were infected with 5 x 105 PFU of BAC-derived wt, GFP-2, and SCP-GFP-2 MCMV with or without DOX administration. Virus titers were determined in lungs and spleens 5 days p.i. (Fig. 7). The DOX administration did not affect the growth of GFP-2 and wt virus (data not shown). The observed difference between the in vivo fitness of the wt and the recombinant viruses is probably due to the increase in genome length, which is in line with previous findings (32). In mice receiving SCP-GFP-2 and DOX, virus titers were found to be strongly reduced in lungs and spleens. In lungs, the regulation reached a range of 3 orders of magnitude (Fig. 7A). Thus, conditional virus replication could also be achieved in vivo, and the presence of DOX reduced viral replication to the detection limit.
DISCUSSION
Here we report on a conditional viral gene expression system in which the regulator and the regulated gene are introduced as one cassette into a herpesvirus genome in order to achieve the regulation of a target gene without necessarily interfering with the viral replication program. Since all elements are integrated into the viral genome, the resulting virus can be studied in any permissive cell line as well as in vivo.
Conditional expression of viral genes is superior to the use of null mutants to study essential functions. One advantage is that the "off" and "on" situations can be studied with the very same construct or virus preparation, thus providing controlled conditions. Additionally, there is no need to construct revertants to ensure that the observed phenotype is linked to the regulated gene.
Conditional gene expression requires the presence of two elements in the same cell: (i) the regulated transcription unit, and (ii) the regulator. Based on expression kinetics, herpesvirus genes of the lytic phase fall into three groups: immediate-early, early, and late (27). In the late phase, the number of viral genomes and transcripts increases enormously. Many herpesviruses modulate host gene expression, which could also affect the regulator, when it is expressed by the infected cell. In addition, by using regulator-expressing cell lines, the risk that amplified late genes could escape regulation is high, if the expression level of the regulator remains static. Thus, the regulator should be amplified as well. Tight control is crucial, particularly in the case of DN allele regulation, since DN proteins, on the one hand, are detrimental to the virus if not repressed sufficiently and, on the other hand, have to be amplified abundantly to fulfill their inhibitory function. The most convenient way to achieve an equilibrium in all target cells at any stage of viral replication is to combine all elements required for regulated expression on the same viral genome. In contrast to other examples, which blocked viral replication right from the start (25, 30), in the constructs presented here the viral gene expression cascade and synthesis of other viral proteins should not be affected. Conditional expression of M50 did not affect the synthesis of the immediate-early protein pp89 or MCP, a late protein. With the DN SCP-GFP mutant, we intended to block virus assembly, a late step in virus replication. The transcription level of authentic SCP was not affected. The inducible block of virus assembly resulted in a DOX-dependent growth reduction of recombinant SCP-GFP-2 MCMV in different cell lines in vitro. We also showed for the first time that DOX-regulated growth of a recombinant virus is feasible in vivo in its native host.
Two viral promoters-enhancers rendered inducible by insertion of two tetO sequences were used. The new fusion promoter PSV40-tetO2 was constructed; it turned out to be more useful due to its reduced background activity in the "off" setting than the published PCMV-tetO2 (35). Using PSV40-tetO2-controlled luciferase in the viral context, the regulation of the luciferase reporter remained robust until completion of cell lysis. Neither the insertion of the regulation cassette into an intergenic region nor the constitutive expression of TetR had detectable effects on virus replication in vitro. For instance, the mutant with regulated expression of EGFP showed wt growth characteristics both in the "off" and "on" settings.
The incorporation of both regulator and the regulated gene into the same viral genome allowed the analysis of tet-regulated viral growth in different cell lines. There was a difference in the degree of regulation in various cell lines, as reported for other tet systems (12). Based on this fact, we believe that further improvement of the conditional system should be possible: while keeping the core promoter, which is not induced by CMV infection, the enhancer might be exchanged with regards to different cell types. Our data underline that conditional phenotypes observed in one specific cell line cannot simply be extrapolated to other cell lines or tissues.
In vivo, the growth of the conditional SCP-GFP mutant was regulated by DOX over several orders of magnitude. However, the recombinants had a loss of fitness compared to that of wt MCMV even without induction of EGFP or DN SCP-GFP. Effects on neighboring genes, basal expression of the immunogenic EGFP, and increase in genome length may contribute to this. Although there may be other suitable insertion sites for the regulation cassette, the far-reaching effects of the introduced enhancers are probably difficult to avoid. EGFP is not an essential element of the conditional expression principle and may be omitted in further constructs. Despite technical limitations, regulation of genes with strong biological effects can now be carried out using the very same constructs both in vitro and in vivo. So far, we have subjected only genes involved in lytic replication to conditional expression. It might be of interest to study the regulation of genes involved in chronic and latent infection in vivo. An attractive option is the possibility of analyzing biological phenotypes that operate mainly in vivo and which, therefore, cannot be studied using the ts approach. For instance, the functions of immunomodulatory genes can now be analyzed in more detail. Finally, we believe that this conditional expression system is applicable to genetic studies of large DNA viruses in general.
ACKNOWLEDGMENTS
This work was supported by grants from the DFG (SFB455) and the Friedrich-Baur-Stiftung (0020/2003).
Supplemental material for this article may be found at http://jvi.asm.org/.
REFERENCES
Akel, H. M., and C. Sweet. 1993. Isolation and preliminary characterisation of twenty-five temperature-sensitive mutants of mouse cytomegalovirus. FEMS Microbiol. Lett. 113:253-260.
Baldick, C. J., Jr., and T. Shenk. 1996. Proteins associated with purified human cytomegalovirus particles. J. Virol. 70:6097-6105.
Borst, E. M., S. Mathys, M. Wagner, W. Muranyi, and M. Messerle. 2001. Genetic evidence of an essential role for cytomegalovirus small capsid protein in viral growth. J. Virol. 75:1450-1458.
Bubeck, A., M. Wagner, Z. Ruzsics, M. Lotzerich, M. Iglesias, I. R. Singh, and U. H. Koszinowski. 2004. Comprehensive mutational analysis of a herpesvirus gene in the viral genome context reveals a region essential for virus replication. J. Virol. 78:8026-8035.
Bubi, I., M. Wagner, A. Krmpoti, T. Saulig, S. Kim, W. M. Yokoyama, S. Jonji, and U. H. Koszinowski. 2004. Gain of virulence caused by loss of a gene in murine cytomegalovirus. J. Virol. 78:7536-7544.
Chee, M. S., A. T. Bankier, S. Beck, R. Bohni, C. M. Brown, R. Cerny, T. Horsnell, C. A. Hutchison III, T. Kouzarides, J. A. Martignetti, et al. 1990. Analysis of the protein-coding content of the sequence of human cytomegalovirus strain AD169. Curr. Top. Microbiol. Immunol. 154:125-169.
Cherepanov, P. P., and W. Wackernagel. 1995. Gene disruption in Escherichia coli: TcR and KmR cassettes with the option of Flp-catalyzed excision of the antibiotic-resistance determinant. Gene 158:9-14.
Desai, P., N. A. DeLuca, and S. Person. 1998. Herpes simplex virus type 1 VP26 is not essential for replication in cell culture but influences production of infectious virus in the nervous system of infected mice. Virology 247:115-124.
Desai, P., and S. Person. 1998. Incorporation of the green fluorescent protein into the herpes simplex virus type 1 capsid. J. Virol. 72:7563-7568.
Gibson, W., K. S. Clopper, W. J. Britt, and M. K. Baxter. 1996. Human cytomegalovirus (HCMV) smallest capsid protein identified as product of short open reading frame located between HCMV UL48 and UL49. J. Virol. 70:5680-5683.
Heider, J. A., W. A. Bresnahan, and T. E. Shenk. 2002. Construction of a rationally designed human cytomegalovirus variant encoding a temperature-sensitive immediate-early 2 protein. Proc. Natl. Acad. Sci. USA 99:3141-3146.
Imhof, M. O., P. Chatellard, and N. Mermod. 2002. Comparative study and identification of potent eukaryotic transcriptional repressors in gene switch systems. J. Biotechnol. 97:275-285.
Keil, G. M., M. R. Fibi, and U. H. Koszinowski. 1985. Characterization of the major immediate-early polypeptides encoded by murine cytomegalovirus. J. Virol. 54:422-428.
Kim, H. J., C. Gatz, W. Hillen, and T. R. Jones. 1995. Tetracycline repressor-regulated gene repression in recombinant human cytomegalovirus. J. Virol. 69:2565-2573.
Klupp, B. G., H. Granzow, and T. C. Mettenleiter. 2000. Primary envelopment of pseudorabies virus at the nuclear membrane requires the UL34 gene product. J. Virol. 74:10063-10073.
Koffron, A. J., M. Hummel, B. K. Patterson, S. Yan, D. B. Kaufman, J. P. Fryer, F. P. Stuart, and M. I. Abecassis. 1998. Cellular localization of latent murine cytomegalovirus. J. Virol. 72:95-103.
Lutarewych, M. A., M. R. Quirk, B. A. Kringstad, W. Li, C. M. Verfaillie, and M. C. Jordan. 1997. Propagation and titration of murine cytomegalovirus in a continuous bone marrow-derived stromal cell line (M2-10B4). J. Virol. Methods 68:193-198.
McVoy, M. A., and E. S. Mocarski. 1999. Tetracycline-mediated regulation of gene expression within the human cytomegalovirus genome. Virology 258:295-303.
Menard, C., M. Wagner, Z. Ruzsics, K. Holak, W. Brune, A. E. Campbell, and U. H. Koszinowski. 2003. Role of murine cytomegalovirus US22 gene family members in replication in macrophages. J. Virol. 77:5557-5570.
Muranyi, W., J. Haas, M. Wagner, G. Krohne, and U. H. Koszinowski. 2002. Cytomegalovirus recruitment of cellular kinases to dissolve the nuclear lamina. Science 297:854-857.
Neubauer, A., J. Rudolph, C. Brandmuller, F. T. Just, and N. Osterrieder. 2002. The equine herpesvirus 1 UL34 gene product is involved in an early step in virus egress and can be efficiently replaced by a UL34-GFP fusion protein. Virology 300:189-204.
Podlech, J., R. Holtappels, N. Wirtz, H. P. Steffens, and M. J. Reddehase. 1998. Reconstitution of CD8 T cells is essential for the prevention of multiple-organ cytomegalovirus histopathology after bone marrow transplantation. J. Gen. Virol. 79:2099-2104.
Rawlinson, W. D., H. E. Farrell, and B. G. Barrell. 1996. Analysis of the complete DNA sequence of murine cytomegalovirus. J. Virol. 70:8833-8849.
Reddehase, M. J., F. Weiland, K. Munch, S. Jonjic, A. Luske, and U. H. Koszinowski. 1985. Interstitial murine cytomegalovirus pneumonia after irradiation: characterization of cells that limit viral replication during established infection of the lungs. J. Virol. 55:264-273.
Rittner, K., H. Schultz, A. Pavirani, and M. Mehtali. 1997. Conditional repression of the E2 transcription unit in E1-E3-deleted adenovirus vectors is correlated with a strong reduction in viral DNA replication and late gene expression in vitro. J. Virol. 71:3307-3311.
Roizman, B., and E. A. Sears. 1996. Herpes simplex viruses and their replication, p. 2248. In B. N. Fields, D. M. Knipe, and P. M. Howley (ed.), Fields virology. Lippincott-Raven Publishers, Philadelphia, Pa.
Roizman, B., E. A. Sears, E. Kieff, E. S. Mocarski, J. I. Cohen, and S. E. Straus. 1996. Herpesviridae, p. 2258-2537. In B. N. Fields, D. M. Knipe, and P. M. Howley (ed.), Fields virology. Lippincott-Raven Publishers, Philadelphia, Pa.
Roller, R. J., Y. Zhou, R. Schnetzer, J. Ferguson, and D. DeSalvo. 2000. Herpes simplex virus type 1 U(L)34 gene product is required for viral envelopment. J. Virol. 74:117-129.
Schaffer, P., V. Vonka, R. Lewis, and M. Benyesh-Melnick. 1970. Temperature-sensitive mutants of herpes simplex virus. Virology 42:1144-1146.
Smith, S. M., M. Khoroshev, P. A. Marx, J. Orenstein, and K. T. Jeang. 2001. Constitutively dead, conditionally live HIV-1 genomes. Ex vivo implications for a live virus vaccine. J. Biol. Chem. 276:32184-32190.
Wagner, M., A. Gutermann, J. Podlech, M. J. Reddehase, and U. H. Koszinowski. 2002. Major histocompatibility complex class I allele-specific cooperative and competitive interactions between immune evasion proteins of cytomegalovirus. J. Exp. Med. 196:805-816.
Wagner, M., S. Jonjic, U. H. Koszinowski, and M. Messerle. 1999. Systematic excision of vector sequences from the BAC-cloned herpesvirus genome during virus reconstitution. J. Virol. 73:7056-7060.
Yao, F., and E. Eriksson. 1999. A novel anti-herpes simplex virus type 1-specific herpes simplex virus type 1 recombinant. Hum. Gene Ther. 10:1811-1818.
Yao, F., and E. Eriksson. 2002. Inhibition of herpes simplex virus type 2 (HSV-2) viral replication by the dominant negative mutant polypeptide of HSV-1 origin binding protein. Antivir. Res. 53:127-133.
Yao, F., T. Svensjo, T. Winkler, M. Lu, C. Eriksson, and E. Eriksson. 1998. Tetracycline repressor, tetR, rather than the tetR-mammalian cell transcription factor fusion derivatives, regulates inducible gene expression in mammalian cells. Hum. Gene Ther. 9:1939-1950.(Brigitte Rupp, Zsolt Ruzs)
ABSTRACT
We have established a conditional gene expression system for cytomegalovirus which allows regulation of genes independently from the viral replication program. Due to the combination of all elements required for regulated expression in the same viral genome, conditional viruses can be studied in different cell lines in vitro and in the natural host in vivo. The combination of a self-sufficient tetracycline-regulated expression cassette and Flp recombinase-mediated insertion into the viral genome allowed fast construction of recombinant murine cytomegaloviruses carrying different conditional genes. The regulation of two reporter genes, the essential viral M50 gene and a dominant-negative mutant gene (m48.2) encoding the small capsid protein, was analyzed in more detail. In vitro, viral growth was regulated by the conditional expression of M50 by 3 orders of magnitude and up to a millionfold when the dominant-negative small capsid protein mutant was used. In vivo, viral growth of the dominant-negative mutant was reduced to detection limits in response to the presence of doxycycline in the organs of mice. We believe that this conditional expression system is applicable to genetic studies of large DNA viruses in general.
INTRODUCTION
Temperature-sensitive (ts) mutants, the first generation of conditional mutants in virology, have been helpful and often superior to null mutants in mapping and identifying viral functions (26). Random procedures for the generation of ts mutants have been described for alphaherpesviruses (29). However, this method did not lead to major findings in betaherpesviruses, which have as their most prominent member human cytomegalovirus (HCMV), an important human pathogen. Propagation of conditional betaherpesvirus mutants by selection of ts alleles is technically difficult, and a specific gene has not been assigned for any of the reported ts mutants (1). However, one ts allele was generated recently for the UL122 gene of HCMV by rational mutagenesis (11).
Conditional gene expression is desirable for the study of gene functions as well as for the development of vaccines and gene therapy vectors both for safety reasons and for the opportunity to adjust the amount of the delivered gene product. As a conditional principle, cell lines expressing tet regulators have been used to control tetracycline-responsive promoters in the herpesvirus genome in order to modulate reporter gene expression (14, 18) or viral growth (33, 34). Recently, all elements required for tet-regulated growth were placed into the genomes of adenovirus (25) and human immunodeficiency virus (30). In these examples, the targeted viral promoters control major regulators of the viral genetic program in order to achieve general inhibition of virus replication. In the present study, we intended to regulate individual genes during morphogenesis and maturation. This goal requires that the gene of interest is regulated as independently as possible from the viral genetic program.
Here, we have established a conditional system for cytomegalovirus in which both the regulator and the regulated transcription unit are constructed in one cassette, which is integrated into the viral genome. The cassette was inserted at a predefined intergenic locus, which appears to be neutral in vitro and in vivo (5). Thus, regulation of the target gene should not necessarily affect the viral program of gene expression. We show that this conditional system has the capacity to regulate the activity of the luciferase reporter about 80-fold in the context of viral replication. In order to study the regulation of viral genes, we subjected two genes to conditional expression: (i) the M50 gene, an essential gene of murine cytomegalovirus (MCMV) (4); and (ii) a dominant-negative (DN) mutant gene (m48.2) encoding the small capsid protein (SCP) (3). The conditional M50 recombinant virus was regulated in vitro at least 1,000-fold; the DN recombinant virus was regulated in vitro 100- to 1,000,000-fold depending on the cell line. In vivo, the productive growth of the DN virus was reduced to the detection limit upon doxycycline (DOX) administration.
MATERIALS AND METHODS
Cells and viruses. BALB/c mouse embryonic fibroblasts (MEF), M2-10B4 cells (ATCC CRL-1972) and SVEC 4-10 cells (ATCC CRL-2181) were prepared and treated as described previously (19). Mouse mammary epithelial C127 cells (ATCC CRL-1616) were maintained in Dulbecco modified Eagle medium supplemented with 10% fetal calf serum. The MCMV wild-type (wt) and wt-FRT viruses used in this study were derived from the bacterial artificial chromosomes (BACs) pSM3fr (32) and wt-FRT-MCMV (5), respectively. MCMV BACs were reconstituted to viruses by transfection of BAC DNA into MEF by use of SuperFect transfection reagent (QIAGEN) according to the manufacturer's instruction. To induce expression of the conditional M50 gene, 1 μg of DOX/ml was added every third day for the reconstitution of the M50-1 virus. Virus stocks were propagated in M2-10B4 cells and quantified on MEF by plaque assay as described previously (17, 24). After the removal of cell debris, the wt, GFP-1, GFP-2, SCP-6FP-1, SCP-GFP2, and M50-1 virus stocks were purified by pelleting them (23,450 x g for 3 h), followed by an ultracentrifugation step over a 15% sucrose cushion. In the case of M50-1, virus stocks were purified additionally by using an OptiPrep gradient (Sigma) according to the manufacturer's instruction.
Growth analysis. For determination of in vitro growth of MCMV mutants, MEF, SVEC 4-10, or C127 cells were infected at a multiplicity of infection (MOI) of 0.1 or 1 PFU per cell. The inoculum was removed after 1 h, and normal medium or medium with 1 μg of DOX/ml was added. Supernatants of infected cells were harvested daily from day 0 (input virus) until day 5 postinfection (p.i.). On day 3, DOX was added a second time. Supernatants of samples were taken in duplicate and titrated on MEF.
Plasmid construction. (i) Construction of plasmids with the inducible promoter PCMV-tetO2. PCMV-tetO2 is composed of the HCMV immediate-early promoter-enhancer (PCMV) and two tet operators (tetO2) downstream of the TATA element (35). In order to place the respective open reading frame (ORF) under control of the inducible promoter PCMV-tetO2, the luciferase ORF was excised from pGL3-Control (Promega) by using HindIII/XbaI and inserted into pcDNA4/TO (Invitrogen), giving rise to pIET-Luc. The conditional expression unit of pIET-Luc was excised by PvuII/MluI, filled in by Klenow polymerase, and inserted into the vector pOriR6K-zeo (GenBank accession no. AY700021) (4), which was linearized by EcoRV, giving rise to pO6-IET-Luc. For the generation of the enhanced green fluorescent protein (EGFP) expression vector, the EGFP ORF was amplified by PCR with primers gfpCN and gfpCE from pEGFP-C1 (Clontech), cut with AflII/ApaI, and inserted into pcDNA4/TO linearized by AflII/ApaI, giving rise to pIET-gfp. For cloning of the DN acting SCP-GFP fusion gene, the EGFP ORF was amplified by PCR from pEGFP-C1 with primers gfpCN and gfpCF and cut with AflII/SpeI. The SCP ORF corresponding to nucleotide (nt) positions 73445 to 73864 in the MCMV genome (23) was amplified from pSM3fr with primers SCPforF and SCPrev and cut with SpeI/ApaI. Both fragments were inserted in one step into pcDNA4/TO linearized with AflII/ApaI, giving rise to pIET-gfpSCP. The inserts from pIET-gfp and pIET-gfpSCP were excised and transferred into pOriR6K-zeo-ie (GenBank accession no. AY700022) (4) by use of NdeI/ApaI, giving rise to pO6-IET-gfp and pO6-IET-gfpSCP, respectively.
In order to insert the tetR gene into the pO6-IET constructs, the tetR expression unit was excised from pcDNA6/TR (Invitrogen) by using MluI and treated with T4 DNA polymerase. This fragment was inserted into pO6-IET-Luc, which was cut with BamHI and filled in, giving rise to pO6-IET-Luc-TR. The same fragment was inserted into pO6-IET-gfp and pO6-IET-gfpSCP, which were linearized by SphI and treated with T4 DNA polymerase, giving rise to pO6-IET-gfp-TR and pO6-IET-gfpSCP-TR, respectively. For the construction of the insertion vector pO6-IET-M50-TR, the M50 ORF was excised from pOriR6K-zeo-ie-M50 (4) with KpnI/XhoI, treated with T4 DNA polymerase, and inserted into pO6-IET-Luc-TR instead of the luciferase ORF, which was removed by a partial digestion with PmeI.
(ii) Construction of plasmids with the inducible promoter PSV40-tetO2. The simian virus 40 (SV40) early enhancer was amplified from pcDNA4/TO by PCR with primers SV40for and SV40rev, and the product was cut with SpeI/NcoI. The minimal CMV promoter including tetO2 was amplified by PCR from pcDNA4/TO with primers miniPfor and miniPrev, and the product was cut with NcoI/AflII. These fragments were inserted in one step into the vector LIT-MUS28 (New England BioLabs) cut with SpeI/AflII, giving rise to pL-SVT.
The vector pOriR6K-zeo was modified by insertion of the hybridized oligonucleotides MCS3 and MCS4 into the KpnI/BamHI sites, giving rise to pOriR6K-zeo-Asc. The tetR gene was excised from pcDNA6/TR by MluI and inserted into pOriR6K-zeo-Asc linearized by AscI, giving rise to pO6-TR.
The luciferase ORF and the bovine growth hormone poly(A) was excised from pIET-Luc and inserted into pL-SVT by AflII/PvuII, giving rise to pSVT-Luc. The conditional SV40-derived expression unit was excised from pSVT-Luc by PvuII/SnaBI and inserted into pO6-TR linearized by EcoRV, giving rise to pO6-SVT-Luc-TR. For construction of pO6-SVT-gfp-TR and pO6-SVT-gfpSCP-TR, the EGFP and SCP-GFP ORFs were excised from pO6-IET-gfp and pO6-IET-gfpSCP, respectively, by PmeI/ApaI, treated with T4 DNA polymerase, and inserted into pO6-SVT-Luc-TR instead of the luciferase ORF, which was removed by partial digestion with PmeI. The sequential steps for the construction of the plasmids used in this study are illustrated in a figure (see Fig. S1 in the supplemental material). The primers used are listed in Table 1.
Generation of recombinant viral BACs. Escherichia coli strain DH10B (Invitrogen) containing the wt-FRT-MCMV BAC and the temperature-sensitive Flp recombinase expression plasmid pCP20 (7) was transformed with the insertion vectors pO6-IET-Luc, pO6-IET-Luc-TR, pO6-SVT-Luc-TR, pO6-IET-gfp-TR, pO6-SVT-gfp-TR, pO6-IET-gfpSCP-TR, and pO6-SVT-gfpSCP-TR and treated as described previously (4). For insertion of pO6-IET-M50-TR, the M50-FRT-MCMV BAC was used, which was derived from the wt-FRT-MCMV BAC by replacement of most of the M50 ORF (nt 75504 to 76451 according to Rawlinson et al. [23]) with a kanamycin resistance cassette by use of a linear fragment amplified with primers 5'-M50 and 3'-M50 (4) as described previously (31). The generated BACs were used for reconstitution to viruses Luc-0, Luc-1, Luc-2, GFP-1, GFP-2, SCP-GFP-1, SCP-GFP-2, and M50-1, respectively.
Luciferase assay. M2-10B4 cells were infected with Luc-0, Luc-1, and Luc-2 at an MOI of 0.1 in the absence or presence of a 1-μg/ml concentration or a serial dilution of DOX, and the cells were harvested after 1 to 4 days. The cells were washed in phosphate-buffered saline and lysed in 100 μl of 1x reporter lysis buffer (Promega). Cell debris was removed by centrifugation. The activity of luciferase was determined with the luciferase assay reagent (Promega) according to the manufacturer's instructions by using dilutions of the lysates (1/100 to 1/20,000 of the 100-μl volume) in order to be in the linear range of detection. The concentration of proteins in the cell lysates was determined with the BCA protein assay kit (Pierce) according to the manufacturer's instructions. Induction levels were calculated for each pair by dividing the normalized luciferase activity in the presence of DOX by that in the absence of DOX.
Western blot analysis. M2-10B4 cells were infected in the absence or presence of 1 μg of DOX/ml with Luc-2 at an MOI of 0.1 for 1 to 4 days, or for 1 day with M50-1 and GFP-1 at an MOI of 0.1 with centrifugal enhancement (700 x g for 30 min), which results in about a 20-fold enhancement of infectivity, i.e., an effective MOI of ca. 2. Cells were lysed in reporter lysis buffer (Promega) for detection of the TetR and in lysis buffer (20 mM Tris, 150 mM NaCl, 1% Triton X-100, freshly supplemented with 1% protease inhibitor cocktail [Roche]) for detection of M50/p35, major capsid protein (MCP), immediate-early protein 1 (IE1), and cellular actin. Lysed samples were separated in a sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel containing 7.5% (IE1 and MCP), 12.5% (TetR), or 15% (M50 and actin) polyacrylamide and transferred to Hybond-P membranes (Amersham). The membranes were treated with polyclonal rabbit antisera specific to TetR (MoBiTec), actin (aa20-33) (Sigma), or M50 (20), a monoclonal mouse antibody to viral IE1 protein pp89 (CROMA101) (provided by S. Jonjic, University of Rijeka, Rijeka, Croatia), and a polyclonal rat antiserum against MCP, which was raised against a synthetic peptide with the amino acid sequence RIQQSSQKDLPESQF (Eurogentec). For immunodetection, horseradish peroxidase-coupled anti-mouse, anti-rat, and anti-rabbit immunoglobulin-specific antisera (Dianova) were used, followed by ECL-Plus (Amersham).
Fluorescence microscopy. MEF, SVEC 4-10, or C127 cells were infected with SCP-GFP-2 at an MOI of 0.1 for 5 days without or with the addition of 1 μg of DOX/ml on day 0. DOX was added again on day 3 (1 μg/ml in the case of MEF or 0.5 μg/ml in the case of SVEC 4-10 or C127 cells). The green fluorescence and the bright-field or phase-contrast images were collected with an Axiovert 200 M system (Zeiss).
Animal experiments. Female BALB/c mice (6 to 10 weeks old) were kept under specific-pathogen-free conditions during the experiments. The animal experiments were approved by the responsible state office (approval no. 211-2531-38/99). Half of the animals received DOX (2 mg/ml in a 5% sucrose solution) in drinking water starting 7 days before infection. DOX was replaced every second day until the end of the experiment. Mice were infected with 5 x 105 PFU of virus intraperitoneally and sacrificed 5 days after infection. The harvested organs were homogenized, and the viral load was determined by a plaque assay using MEF with centrifugal enhancement, which results in a 20-fold enhancement of infectivity.
RESULTS
Design of the regulation cassettes and construction of MCMV recombinants. To regulate gene expression in MCMV, we used the elements of the Tn10-encoded Tet repressor (TetR) and its binding sequence, the tet operator (tetO). TetR proteins bind to operator sequences positioned within the promoter of the gene of interest (Fig. 1A) (35), thereby interfering with the transcription initiation complex. TetR binding is controlled by tetracycline or its analogue, DOX, which causes a reduction in the affinity of the TetR for the operator sequence, thus leading to the release of the transcriptional block (Fig. 1A). The regulation cassettes, which were designed to carry all elements required for regulation in order to be independent of special cell lines, were each composed of two expression units: (i) the constitutively expressed regulator, i.e., the tetR under control of the HCMV major immediate-early enhancer-promoter (PCMV), and (ii) the inducible transcription unit containing two minimal tet operators (tetO2). First, the PCMV-tetO2 (35) was used as an inducible promoter in the regulation cassette. This promoter carries two minimal tet operators (tetO2) inserted 10 bp downstream of the TATA element of the PCMV. In addition, to reduce the basal activity of the PCMV-tetO2 construct, we replaced the HCMV major immediate-early enhancer with the SV40 early enhancer, resulting in PSV40-tetO2. These inducible promoters were analyzed by expression of either luciferase, M50, EGFP, or SCP-GFP, a fusion protein of EGFP and SCP (the m48.2 gene product) of MCMV, which is traceable by its green fluorescence and has been claimed to inhibit viral growth in a DN manner (3). In order to generate conditional MCMV mutants, the plasmids carrying the regulation cassettes for luciferase, EGFP, or SCP-GFP were inserted into the MCMV BAC containing an Flp recombination target (FRT) site in the region between ORFs m16 and m17 (5) (Fig. 1B). The plasmid containing the M50 regulation cassette was inserted into the M50-FRT-MCMV BAC, in which the authentic M50 ORF had been deleted (Fig. 1C). The recombinant MCMV BACs containing the respective regulation cassettes were used for the reconstitution of MCMV mutants.
Regulated expression of the luciferase reporter gene by MCMV. The regulatory capacities of the different constructs within the viral genome were tested by luciferase expression in infected M2-10B4 cells. In cells infected at an MOI of 0.1 with Luc-1 (PCMV-tetO2-controlled luciferase) and Luc-2 (PSV40-tetO2-controlled luciferase) (Fig. 1B), the reporter gene expression was induced in the presence of 1 μg of DOX/ml 11- and 20-fold on day 1 p.i., and induction levels increased on day 2 p.i. 29- and 51-fold, respectively (Fig. 2A). The luciferase expression did not respond to DOX in cells infected with the control virus Luc-0, which lacks the tetR gene. The induction levels were similar for Luc-1 and Luc-2 after infection at a 10-fold-higher MOI (MOI of 1), namely, 9- and 24-fold on day 1 p.i. and 29- and 68-fold on day 2 p.i., respectively (data not shown). Luc-2 with its superior inducibility was used for further analysis. Luc-2 expressed about one-third of the luciferase activity expressed by Luc-0 after induction by DOX (3,325 ± 468 versus 10,036 ± 5,493 relative light units [RLU] per μl [mean ± standard deviation] for a 1:200 dilution of each sample) at an MOI of 0.1 on day 1 p.i., while the basal activity of Luc-2 in the noninduced stage was very low (171 ± 49 and 933 ± 299 RLU/μl for a 1:200 dilution of each sample on day 1 and day 2 p.i., respectively). In Luc-2-infected cells, the maximal induction levels (60- to 100-fold) were observed on days 3 and 4 p.i. (Fig. 2B). In the same cell lysates, which were used for measurement of the luciferase activity, TetR protein was detectable starting from day 3 p.i. (Fig. 2C). Luciferase activity above the background level was seen at a DOX concentration of 0.5 to 1 ng/ml with a half-effective dose of 2 ng/ml on days 1 and 2 p.i. (Fig. 2D). Thus, in M2-10B4 cells, a conditional reporter activity was detectable for several days, reflecting repeated MCMV replication cycles.
Since in this conditional system all elements required for regulation are combined in one viral genome, the conditional gene expression could be analyzed in different cell lines. The regulation of the luciferase expression was studied in the endothelial SVEC 4-10 cell line and the epithelial C127 cell line (Fig. 2E). The induction levels in Luc-2-infected SVEC 4-10 cells were similar to those in M2-10B4 cells: 13- and 46-fold on days 1 and 2 p.i., respectively. Yet, in C127 cells, the induction levels were lower and reached only 3-fold and 11-fold on days 1 and 2 p.i., respectively. The low induction levels in these cells was not due to an altered susceptibility to DOX since the maximal induction levels were reached in both cell lines at a DOX concentration of about 20 ng/ml on day 2 p.i. (Fig. 2F).
Regulated expression of the essential M50 gene. In order to apply the conditional system to a gene essential for virus replication, we used M50 as an example (4). M50 homologues are conserved among the Herpesviridae (6, 23). Inactivation of the homologous genes in certain -herpesviruses severely affects viral growth in cell culture, most probably because of their crucial role in the nuclear egress of viral capsids (15, 21, 28). For the M50 gene product, a similar role has been proposed (20). The M50 regulation cassette was tested in the context of a BAC-derived MCMV genome lacking the authentic M50 ORF (Fig. 1C). Since the essential M50 gene is required for viral propagation, the reconstitution of M50 mutants had to be carried out in the presence of DOX. The standard purification protocol was extended by an additional gradient ultracentrifugation step in order to remove residual trace amounts of DOX from M50-1 virus stocks. M50-1 and control viruses were tested in primary fibroblasts under multistep growth conditions. In the absence of DOX, the growth of M50-1 was reduced at least 3 orders of magnitude, whereas the addition of 1 μg of DOX/ml rescued the growth phenotype (Fig. 3). The control viruses grew similarly irrespective of the presence of DOX.
The regulation of the M50 gene at the ectopic position should affect virus replication by an inducible block of the capsid export from the nucleus (20) but should not affect the normal viral gene expression cascade. Therefore, we tested M50-1 and GFP-1-infected M2-10B4 cells for viral protein expression. The cells were infected at an effective MOI of 2 (MOI of 0.1 with centrifugal enhancement) in the absence and presence of 1 μg of DOX/ml in order to infect all cells simultaneously. Cell lysates were analyzed by Western blotting on day 1 p.i. (Fig. 4). Notably, M50/p35 protein was not detectable in the cells infected with M50-1 in the absence of DOX, indicating a tight regulation (Fig. 4, lane 3). In M50-1-infected cells in the presence of DOX (Fig. 4, lane 4) and in GFP-1-infected cells irrespective of the presence of DOX (lane 1 and 2), M50/p35 was present. We then analyzed the expression of the MCP, which is produced in large amounts in the late phase of viral replication. The degree of MCP expression did not vary in response to DOX, neither in M50-1- nor in GFP-1-infected cells. Comparable protein amounts were also detected for the viral immediate-early protein 1 (IE1), which is found during all phases of viral gene expression, both as full-length pp89 and smaller posttranslational modification products (13).
Thus, the conditional system could be applied to an essential viral gene in order to establish a DOX-dependent growth phenotype. In presence of DOX, the M50-1 virus grew in a manner that was indistinguishable from that of wt and GFP control viruses. There was no detectable effect on the expression of other viral genes.
Regulated expression of DN SCP-GFP. The DN principle is a general tool in cell biology, and we applied it in order to control the growth of MCMV. The SCP encoded by UL48/49 of HCMV (2, 10) and m48.2 of MCMV is essential for viral replication (3). The homologous protein of HSV-1, VP26, is not essential, and fusion to the GFP resulted in viable and infectious virions (8, 9). Recently, it was shown that the constitutive expression of the fusion protein of SCP with EGFP (SCP-GFP) prevents reconstitution of CMV even in the presence of the authentic SCP gene (3). Therefore, we tested the DN function by conditional expression of the fusion gene in the context of a BAC-derived MCMV containing all wt MCMV genes (Fig. 1B). The reconstitution of SCP-GFP mutants was carried out in the absence of DOX in order to prevent the expression of DN SCP-GFP protein. Interestingly, both conditional constructs, SCP-GFP-1 and SCP-GFP-2, were reconstituted like the wt MCMV BAC.
First, we studied the conditional replication of the SCP-GFP mutants in vitro. In the absence of DOX, the titers of BAC-derived wt virus, GFP-1, and SCP-GFP-2 revealed no difference in growth on MEF under multistep growth conditions (MOI of 0.1) (Fig. 5A). In contrast, the presence of 1 μg of DOX/ml completely inhibited growth of the SCP-GFP-2 virus (Fig. 5A) but not of control viruses. The SCP-GFP-1 virus was regulated similarly, yet there were infectious particles released in the supernatant during the observed period (Fig. 5B). Under single-step growth conditions (MOI of 1) in the absence of DOX, the controls and SCP-GFP-2 grew likewise indistinguishably (Fig. 5C). Under this condition, the growth of SCP-GFP-2 was reduced by a factor of 1,000 in response to DOX. The inhibition of virus replication (MOI of 0.1) was DOX concentration dependent (Fig. 5D): concentrations of at least 10 ng of DOX/ml resulted in an almost complete inhibition of SCP-GFP-2 growth, while 1 ng of DOX/ml did not affect the growth of the mutant. Thus, the insertion of a regulation cassette containing the DN SCP-GFP into MCMV resulted in regulation of viral growth of over 6 orders of magnitude. The dose response of SCP-GFP-2 to DOX was comparable to that of Luc-2 (Fig. 2D).
The green fluorescence of SCP-GFP revealed both inhibitor expression and virus spread. The SCP-GFP fusion proteins accumulated in the nucleus, leading to a punctate staining as described previously (3). In infected MEF (Fig. 6a to d), a weak fluorescent signal was detected for SCP-GFP in the absence of DOX (Fig. 6a), indicating a low basal expression. The signal was clearly increased in the presence of 1 μg of DOX/ml (Fig. 6b). Importantly, induction of SCP-GFP was associated with a lack of plaque formation (Fig. 6d).
Epithelial and endothelial cells are major targets of MCMV (16, 22). Therefore, we studied the conditional SCP-GFP mutants in the endothelial cell line SVEC 4-10 (Fig. 6e to h) and the epithelial cell line C127 (Fig. 6i to l). The SCP-GFP expression was induced by DOX in both cell lines (Fig. 6f and j). Despite basal expression of SCP-GFP in the absence of DOX (Fig. 6e and i), viral plaques were formed (Fig. 6g and k). Comparably to the situation observed with MEF (Fig. 6d), no plaques were detected in infected SVEC 4-10 cells in the presence of DOX (Fig. 6 h). Yet, in infected C127 cells, viral plaques were seen even in the presence of DOX (Fig. 6l), although they were smaller than in the absence of DOX. In order to correlate the different expression levels of SCP-GFP with the release of infectious viral particles from cells, the supernatants of infected SVEC 4-10 and C127 cells were titrated. No infectious particles were detected in the supernatants of SVEC 4-10 cells infected with SCP-GFP-1 or SCP-GFP-2 in the presence of DOX (Fig. 6m). In contrast, both recombinants replicated in the epithelial cell line C127 in the absence as well as in the presence of DOX (Fig. 6n). However, there was a DOX-mediated reduction in viral titers of about 100-fold. Notably, the DOX-mediated induction of the luciferase activity in C127 cells was also not as high as it was in M2-10B4 or SVEC 4-10 cells (Fig. 2E), which was not due to a different DOX susceptibility of C127 cells (Fig. 2F). Additionally, the expression of SCP-GFP in C127 cells was inducible by DOX, as detected by the increase in the punctate GFP fluorescence in the nuclei of infected cells (Fig. 6j).
For an inhibitory effect, a certain level of DN proteins is required in relation to their wt counterparts. Since there is no antibody available against SCP of MCMV yet, we analyzed the expression ratio of SCP and SCP-GFP genes in infected SVEC 4-10 and C127 cells by semiquantitative reverse transcription-PCR. In repeated experiments, the normalized SCP-GFP mRNA was found to be induced by DOX 10- to 15-fold in SVEC 4-10 cells and 4- to 6-fold in C127 cells, and the ratio between the SCP-GFP- and SCP-specific signals was higher in SVEC 4-10 cells than in C127 cells (data not shown).
Regulated virus replication in vivo. The differences in induction levels of the luciferase activity and the growth of the SCP-GFP-2 virus reflect that the efficiency of the tet regulation is dependent on the cell line. However, to study the biological relevance of viral functions, efficient regulation in the natural host is also required. Therefore, we analyzed the growth of the conditional lethal recombinant MCMV in mice. BALB/c mice were infected with 5 x 105 PFU of BAC-derived wt, GFP-2, and SCP-GFP-2 MCMV with or without DOX administration. Virus titers were determined in lungs and spleens 5 days p.i. (Fig. 7). The DOX administration did not affect the growth of GFP-2 and wt virus (data not shown). The observed difference between the in vivo fitness of the wt and the recombinant viruses is probably due to the increase in genome length, which is in line with previous findings (32). In mice receiving SCP-GFP-2 and DOX, virus titers were found to be strongly reduced in lungs and spleens. In lungs, the regulation reached a range of 3 orders of magnitude (Fig. 7A). Thus, conditional virus replication could also be achieved in vivo, and the presence of DOX reduced viral replication to the detection limit.
DISCUSSION
Here we report on a conditional viral gene expression system in which the regulator and the regulated gene are introduced as one cassette into a herpesvirus genome in order to achieve the regulation of a target gene without necessarily interfering with the viral replication program. Since all elements are integrated into the viral genome, the resulting virus can be studied in any permissive cell line as well as in vivo.
Conditional expression of viral genes is superior to the use of null mutants to study essential functions. One advantage is that the "off" and "on" situations can be studied with the very same construct or virus preparation, thus providing controlled conditions. Additionally, there is no need to construct revertants to ensure that the observed phenotype is linked to the regulated gene.
Conditional gene expression requires the presence of two elements in the same cell: (i) the regulated transcription unit, and (ii) the regulator. Based on expression kinetics, herpesvirus genes of the lytic phase fall into three groups: immediate-early, early, and late (27). In the late phase, the number of viral genomes and transcripts increases enormously. Many herpesviruses modulate host gene expression, which could also affect the regulator, when it is expressed by the infected cell. In addition, by using regulator-expressing cell lines, the risk that amplified late genes could escape regulation is high, if the expression level of the regulator remains static. Thus, the regulator should be amplified as well. Tight control is crucial, particularly in the case of DN allele regulation, since DN proteins, on the one hand, are detrimental to the virus if not repressed sufficiently and, on the other hand, have to be amplified abundantly to fulfill their inhibitory function. The most convenient way to achieve an equilibrium in all target cells at any stage of viral replication is to combine all elements required for regulated expression on the same viral genome. In contrast to other examples, which blocked viral replication right from the start (25, 30), in the constructs presented here the viral gene expression cascade and synthesis of other viral proteins should not be affected. Conditional expression of M50 did not affect the synthesis of the immediate-early protein pp89 or MCP, a late protein. With the DN SCP-GFP mutant, we intended to block virus assembly, a late step in virus replication. The transcription level of authentic SCP was not affected. The inducible block of virus assembly resulted in a DOX-dependent growth reduction of recombinant SCP-GFP-2 MCMV in different cell lines in vitro. We also showed for the first time that DOX-regulated growth of a recombinant virus is feasible in vivo in its native host.
Two viral promoters-enhancers rendered inducible by insertion of two tetO sequences were used. The new fusion promoter PSV40-tetO2 was constructed; it turned out to be more useful due to its reduced background activity in the "off" setting than the published PCMV-tetO2 (35). Using PSV40-tetO2-controlled luciferase in the viral context, the regulation of the luciferase reporter remained robust until completion of cell lysis. Neither the insertion of the regulation cassette into an intergenic region nor the constitutive expression of TetR had detectable effects on virus replication in vitro. For instance, the mutant with regulated expression of EGFP showed wt growth characteristics both in the "off" and "on" settings.
The incorporation of both regulator and the regulated gene into the same viral genome allowed the analysis of tet-regulated viral growth in different cell lines. There was a difference in the degree of regulation in various cell lines, as reported for other tet systems (12). Based on this fact, we believe that further improvement of the conditional system should be possible: while keeping the core promoter, which is not induced by CMV infection, the enhancer might be exchanged with regards to different cell types. Our data underline that conditional phenotypes observed in one specific cell line cannot simply be extrapolated to other cell lines or tissues.
In vivo, the growth of the conditional SCP-GFP mutant was regulated by DOX over several orders of magnitude. However, the recombinants had a loss of fitness compared to that of wt MCMV even without induction of EGFP or DN SCP-GFP. Effects on neighboring genes, basal expression of the immunogenic EGFP, and increase in genome length may contribute to this. Although there may be other suitable insertion sites for the regulation cassette, the far-reaching effects of the introduced enhancers are probably difficult to avoid. EGFP is not an essential element of the conditional expression principle and may be omitted in further constructs. Despite technical limitations, regulation of genes with strong biological effects can now be carried out using the very same constructs both in vitro and in vivo. So far, we have subjected only genes involved in lytic replication to conditional expression. It might be of interest to study the regulation of genes involved in chronic and latent infection in vivo. An attractive option is the possibility of analyzing biological phenotypes that operate mainly in vivo and which, therefore, cannot be studied using the ts approach. For instance, the functions of immunomodulatory genes can now be analyzed in more detail. Finally, we believe that this conditional expression system is applicable to genetic studies of large DNA viruses in general.
ACKNOWLEDGMENTS
This work was supported by grants from the DFG (SFB455) and the Friedrich-Baur-Stiftung (0020/2003).
Supplemental material for this article may be found at http://jvi.asm.org/.
REFERENCES
Akel, H. M., and C. Sweet. 1993. Isolation and preliminary characterisation of twenty-five temperature-sensitive mutants of mouse cytomegalovirus. FEMS Microbiol. Lett. 113:253-260.
Baldick, C. J., Jr., and T. Shenk. 1996. Proteins associated with purified human cytomegalovirus particles. J. Virol. 70:6097-6105.
Borst, E. M., S. Mathys, M. Wagner, W. Muranyi, and M. Messerle. 2001. Genetic evidence of an essential role for cytomegalovirus small capsid protein in viral growth. J. Virol. 75:1450-1458.
Bubeck, A., M. Wagner, Z. Ruzsics, M. Lotzerich, M. Iglesias, I. R. Singh, and U. H. Koszinowski. 2004. Comprehensive mutational analysis of a herpesvirus gene in the viral genome context reveals a region essential for virus replication. J. Virol. 78:8026-8035.
Bubi, I., M. Wagner, A. Krmpoti, T. Saulig, S. Kim, W. M. Yokoyama, S. Jonji, and U. H. Koszinowski. 2004. Gain of virulence caused by loss of a gene in murine cytomegalovirus. J. Virol. 78:7536-7544.
Chee, M. S., A. T. Bankier, S. Beck, R. Bohni, C. M. Brown, R. Cerny, T. Horsnell, C. A. Hutchison III, T. Kouzarides, J. A. Martignetti, et al. 1990. Analysis of the protein-coding content of the sequence of human cytomegalovirus strain AD169. Curr. Top. Microbiol. Immunol. 154:125-169.
Cherepanov, P. P., and W. Wackernagel. 1995. Gene disruption in Escherichia coli: TcR and KmR cassettes with the option of Flp-catalyzed excision of the antibiotic-resistance determinant. Gene 158:9-14.
Desai, P., N. A. DeLuca, and S. Person. 1998. Herpes simplex virus type 1 VP26 is not essential for replication in cell culture but influences production of infectious virus in the nervous system of infected mice. Virology 247:115-124.
Desai, P., and S. Person. 1998. Incorporation of the green fluorescent protein into the herpes simplex virus type 1 capsid. J. Virol. 72:7563-7568.
Gibson, W., K. S. Clopper, W. J. Britt, and M. K. Baxter. 1996. Human cytomegalovirus (HCMV) smallest capsid protein identified as product of short open reading frame located between HCMV UL48 and UL49. J. Virol. 70:5680-5683.
Heider, J. A., W. A. Bresnahan, and T. E. Shenk. 2002. Construction of a rationally designed human cytomegalovirus variant encoding a temperature-sensitive immediate-early 2 protein. Proc. Natl. Acad. Sci. USA 99:3141-3146.
Imhof, M. O., P. Chatellard, and N. Mermod. 2002. Comparative study and identification of potent eukaryotic transcriptional repressors in gene switch systems. J. Biotechnol. 97:275-285.
Keil, G. M., M. R. Fibi, and U. H. Koszinowski. 1985. Characterization of the major immediate-early polypeptides encoded by murine cytomegalovirus. J. Virol. 54:422-428.
Kim, H. J., C. Gatz, W. Hillen, and T. R. Jones. 1995. Tetracycline repressor-regulated gene repression in recombinant human cytomegalovirus. J. Virol. 69:2565-2573.
Klupp, B. G., H. Granzow, and T. C. Mettenleiter. 2000. Primary envelopment of pseudorabies virus at the nuclear membrane requires the UL34 gene product. J. Virol. 74:10063-10073.
Koffron, A. J., M. Hummel, B. K. Patterson, S. Yan, D. B. Kaufman, J. P. Fryer, F. P. Stuart, and M. I. Abecassis. 1998. Cellular localization of latent murine cytomegalovirus. J. Virol. 72:95-103.
Lutarewych, M. A., M. R. Quirk, B. A. Kringstad, W. Li, C. M. Verfaillie, and M. C. Jordan. 1997. Propagation and titration of murine cytomegalovirus in a continuous bone marrow-derived stromal cell line (M2-10B4). J. Virol. Methods 68:193-198.
McVoy, M. A., and E. S. Mocarski. 1999. Tetracycline-mediated regulation of gene expression within the human cytomegalovirus genome. Virology 258:295-303.
Menard, C., M. Wagner, Z. Ruzsics, K. Holak, W. Brune, A. E. Campbell, and U. H. Koszinowski. 2003. Role of murine cytomegalovirus US22 gene family members in replication in macrophages. J. Virol. 77:5557-5570.
Muranyi, W., J. Haas, M. Wagner, G. Krohne, and U. H. Koszinowski. 2002. Cytomegalovirus recruitment of cellular kinases to dissolve the nuclear lamina. Science 297:854-857.
Neubauer, A., J. Rudolph, C. Brandmuller, F. T. Just, and N. Osterrieder. 2002. The equine herpesvirus 1 UL34 gene product is involved in an early step in virus egress and can be efficiently replaced by a UL34-GFP fusion protein. Virology 300:189-204.
Podlech, J., R. Holtappels, N. Wirtz, H. P. Steffens, and M. J. Reddehase. 1998. Reconstitution of CD8 T cells is essential for the prevention of multiple-organ cytomegalovirus histopathology after bone marrow transplantation. J. Gen. Virol. 79:2099-2104.
Rawlinson, W. D., H. E. Farrell, and B. G. Barrell. 1996. Analysis of the complete DNA sequence of murine cytomegalovirus. J. Virol. 70:8833-8849.
Reddehase, M. J., F. Weiland, K. Munch, S. Jonjic, A. Luske, and U. H. Koszinowski. 1985. Interstitial murine cytomegalovirus pneumonia after irradiation: characterization of cells that limit viral replication during established infection of the lungs. J. Virol. 55:264-273.
Rittner, K., H. Schultz, A. Pavirani, and M. Mehtali. 1997. Conditional repression of the E2 transcription unit in E1-E3-deleted adenovirus vectors is correlated with a strong reduction in viral DNA replication and late gene expression in vitro. J. Virol. 71:3307-3311.
Roizman, B., and E. A. Sears. 1996. Herpes simplex viruses and their replication, p. 2248. In B. N. Fields, D. M. Knipe, and P. M. Howley (ed.), Fields virology. Lippincott-Raven Publishers, Philadelphia, Pa.
Roizman, B., E. A. Sears, E. Kieff, E. S. Mocarski, J. I. Cohen, and S. E. Straus. 1996. Herpesviridae, p. 2258-2537. In B. N. Fields, D. M. Knipe, and P. M. Howley (ed.), Fields virology. Lippincott-Raven Publishers, Philadelphia, Pa.
Roller, R. J., Y. Zhou, R. Schnetzer, J. Ferguson, and D. DeSalvo. 2000. Herpes simplex virus type 1 U(L)34 gene product is required for viral envelopment. J. Virol. 74:117-129.
Schaffer, P., V. Vonka, R. Lewis, and M. Benyesh-Melnick. 1970. Temperature-sensitive mutants of herpes simplex virus. Virology 42:1144-1146.
Smith, S. M., M. Khoroshev, P. A. Marx, J. Orenstein, and K. T. Jeang. 2001. Constitutively dead, conditionally live HIV-1 genomes. Ex vivo implications for a live virus vaccine. J. Biol. Chem. 276:32184-32190.
Wagner, M., A. Gutermann, J. Podlech, M. J. Reddehase, and U. H. Koszinowski. 2002. Major histocompatibility complex class I allele-specific cooperative and competitive interactions between immune evasion proteins of cytomegalovirus. J. Exp. Med. 196:805-816.
Wagner, M., S. Jonjic, U. H. Koszinowski, and M. Messerle. 1999. Systematic excision of vector sequences from the BAC-cloned herpesvirus genome during virus reconstitution. J. Virol. 73:7056-7060.
Yao, F., and E. Eriksson. 1999. A novel anti-herpes simplex virus type 1-specific herpes simplex virus type 1 recombinant. Hum. Gene Ther. 10:1811-1818.
Yao, F., and E. Eriksson. 2002. Inhibition of herpes simplex virus type 2 (HSV-2) viral replication by the dominant negative mutant polypeptide of HSV-1 origin binding protein. Antivir. Res. 53:127-133.
Yao, F., T. Svensjo, T. Winkler, M. Lu, C. Eriksson, and E. Eriksson. 1998. Tetracycline repressor, tetR, rather than the tetR-mammalian cell transcription factor fusion derivatives, regulates inducible gene expression in mammalian cells. Hum. Gene Ther. 9:1939-1950.(Brigitte Rupp, Zsolt Ruzs)