Functional Analysis of the Murine Cytomegalovirus Chemokine Receptor Homologue M33: Ablation of Constitutive Signaling Is Associated with an Attenuated Phenotype In Vivo^▿

PCR and cloning.

Primers used for mutagenesis and cloning are shown in Table 1, and the locations of restriction sites are shown in Fig. 1. The M33 and UL33 sequences were derived from the HCMV K181 (Perth) and AD169 strains, respectively. Mutagenesis was achieved by either the megaprimer or overlap extension PCR methods (34, 54). All constructs were verified by sequencing prior to use.

Plasmids for signaling assays.

The M33 cDNA sequence (i.e., lacking the 5′ intron) in pTEJ (a kind gift of M. Waldhoer, The Panum Institute, Copenhagen, Denmark) was reamplified and cloned into the EcoRI site of pcDNA3 (Invitrogen) to generate pcDNA3/M33. The cDNA of UL33 was cloned into the EcoRI site of pcDNA3 to generate pcDNA3/UL33. The various modified forms of M33 were generated by PCR-mutagenesis and cloning of either the full M33 cassette into pcDNA3 or mutated fragments of M33 into pcDNA3/M33. For N-terminal truncation, pcDNA3/M33ΔN10 was generated by amplification of M33 lacking the initial ATG, such that translation would initiate at the next available ATG a further nine amino acids downstream. For C-terminal truncations, pcDNA3/M33ΔC57, pcDNA3/M33ΔC38, pcDNA3/M33ΔC24, and pcDNA3/M33ΔC8 were generated by amplification of M33 incorporating a stop codon prior to the final 57, 38, 24, and 8 C-terminal amino acids, respectively.

Plasmids for fluorescence imaging studies.

Hemagglutinin (HA) and c-myc tagging of the N terminus of M33 was achieved by PCR amplification of M33 lacking the initiator ATG (from pcDNA3/M33) which was cloned into the EcoRI site of pCMV-HA or pCMV-myc, respectively (BD Biosciences, Oxford, United Kingdom). A pTEJ-based plasmid containing M33 fused in frame with enhanced green fluorescent protein (EGFP) at the C terminus was kindly provided by M. Waldhoer, The Panum Institute, Copenhagen, Denmark. The M33-EGFP cassette from this plasmid was PCR amplified and cloned into pcDNA3 to generate pcDNA3/M33-GFP. All mutant M33 constructs (described above in “Plasmids for signaling assays”) were fused with EGFP at the C terminus to facilitate protein localization studies in transfected cells. Mutations upstream of the PfiMI site (position 890 within the M33 coding sequence (Fig. 1) were subcloned from the mutant pcDNA3/M33 plasmids as PfiMI/HindIII fragments and introduced into PfiMI/HindIII-digested pcDNA3/M33-GFP. The C-terminal deletion mutants were PCR amplified as full-length cassettes, using primers that incorporated an EcoRI upstream of the start codon and an ApaI site at the position of the 5′ truncation, and cloned in frame into EcoRI/ApaI-digested pEGFP-N1 (BD Biosciences, Oxford, United Kingdom). For the generation of UL33-GFP, a three-way ligation was performed between HindIII/NotI-digested pcDNA3, a UL33 fragment that had been amplified from pcDNA/UL33 with 5′ HindIII and 3′ HincII restriction sites (the latter blunt site replacing the stop codon), and an EGFP fragment amplified from p-EGPF-N1 (Clontech) with a 5′ PmlI restriction site (concomitantly removing the initiating methionine) and a 3′ NotI site.

Plasmids for preparation of MCMV recombinants.

The recombination plasmid pING/M33, comprising an MCMV BglII fragment [between 39.4 and 44.4 kb of the MCMV K181 (Perth) genomic sequence] that contains the M33 sequence and approximately 5 kb of flanking sequence, has been described previously (16). A second recombination plasmid, designated pGEM/M33, was generated by cloning the above BglII fragment into the PvuII site of pGEM 4(Z). To facilitate the introduction of N-terminal mutations of M33, the pGEM/M33 plasmid was subsequently modified by the introduction of a unique HindIII site upstream of the M33 start codon to produce pGEM/hM33. The generation of 24- and 38-residue C-terminal truncations of M33 was achieved by the introduction of in-frame stop codons at appropriate sites within M33 while retaining M33 sequences downstream of the stop codon. The modified fragments were subsequently cloned into pING/M33 as BstEII/NotI fragments to produce pING/M33ΔC24 and pING/M33ΔC38, respectively. M33 TM III point mutations N130D and R131Q were introduced into pGEM/hM33 by subcloning BstEII/NsiI fragments from their respective pcDNA3 signaling plasmids (described above) into pGEM/hM33. The M33 TM II F79D mutation was cloned from pcDNA3/M33(F79D) as a HindIII/BstEII fragment into pGEM/hM33. GFP-tagged wild-type (wt) and mutated M33 were introduced into pING/M33 via shuttling of a BstEII/NotI fragment from the relevant GFP-tagged expression plasmid (described above). Replacement of M33 with untagged UL33 was achieved by amplification, with 5′ HindIII and 3′ NotI sites, from plasmids derived from either genomic DNA (intron plus) or cDNA (intron minus). The UL33 cassettes were then inserted between the HindIII/NotI sites of pGEM/hM33 to produce plasmids pGEM/UL33 and pGEM/UL33Δi, respectively. Replacement of the wt M33 sequence (intron plus) with M33 lacking the intron was achieved by amplification of M33 (with 5′ HindIII and 3′ NotI sites) from pcDNA3/M33 and cloning into the HindIII/NotI sites of pGEM/hM33 to produce plasmid pGEM/M33Δi.

Purification of expression plasmids.

Plasmids used in transfection experiments were purified using Qiagen EndoFree plasmid kits to an optical density ratio (260 nm:280 nm) of greater than 1.8 prior to use.

Luciferase reporter assays.

HEK-293 cells were grown in Dulbecco's modified Eagle's medium (DMEM)-GlutaMAX (Invitrogen) supplemented with 10% heat inactivated fetal calf serum, 180 U/ml penicillin, and 45 μg/ml streptomycin at 37°C and 5% CO₂. Cells were seeded (5 × 10⁴ cells/well) in 96-well black and white isoplates (PerkinElmer Life and Analytical Sciences) using medium without antibiotics. One day after seeding, cells were cotransfected with various concentrations of pcDNA3-based receptor plasmid and either 50 ng of pNFAT-Luc (Pathdetect cis-reporting system) or 50 ng of pFR-Luc and 6 ng of pFA2-CREB (Pathdetect trans-reporting system; both from Stratagene). Transfections were carried out using OptiMEM and Lipofectamine 2000 reagent according to the manufacturer's guidelines (Invitrogen). The medium was replaced with DMEM-GlutaMAX containing 0.5% heat-inactivated fetal calf serum without antibiotics at 5 h posttransfection. After a 24-h incubation, the cells were washed with phosphate-buffered saline (PBS), and an equal volume of LucLite was added (PerkinElmer). Luminescence was measured for 5 s at 22°C, following 10 min of dark adaptation on a TopCount microplate scintillation and luminescence counter.

Phosphatidylinositol turnover assay.

HEK-293 cells were seeded (3.5 × 10⁴ cells/well) in poly-l-lysine-coated 96-well plates and cultured overnight in growth medium (DMEM containing 10% FBS, 180 U/ml penicillin, and 45 U/ml streptomycin) at 37°C and 10% CO₂. The cells were cotransfected the following day with receptor constructs at 3, 6, or 10 ng/well, and the promiscuous Gα subunit GαΔ6qi4myr ((kindly provided by Evi Konstenis, 7TM-Pharma, Denmark) at 10 ng/well using the serum-free medium OptiMEM and Lipofectamine 2000 at 10 μl/ml. At 5 h posttransfection the cells were transferred to normal growth medium. Twenty-four hours later the cells were labeled with 5 μCi of [myo-³H]inositol per well and incubated overnight at 37°C. The following day, the cells were washed once in assay buffer (20 mM HEPES, 140 mM NaCl, 5 mM KCl, 1 mM MgSO₄, and CaCl₂) and incubated for 90 min at 37°C in assay buffer containing 10 mM LiCl. Subsequently, the cells were incubated on ice in cold formic acid (10 mM) for 30 min. A total of 20 μl of the lysis solution was subsequently transferred to a new white 96-well plate and mixed with 80 μl of freshly diluted scintillation proximity assay YSi beads at 12.5 mg/ml. The plates were vigorously shaken at room temperature for 30 min, centrifuged at 1,500 rpm for 5 min, left overnight at room temperature, and counted the following day using a TopCounter (Packard) after a final round of centrifugation.

Quantitation of expression in transfected cells.

HEK-293 cells were transfected with various amounts (6.25 to 200 ng/well) of expression plasmids for GFP-tagged (C terminus) wt and mutated M33, using the methods described above (luciferase reporter assays) with the following modifications. Cells were seeded into 96-well optical bottom black-walled plates (Fluorcarbon/Black; Nunc). At 24 h posttransfection, cells were washed three times in Hanks balanced salt solution supplemented with Ca²⁺ and Mg²⁺ salts (HBSS plus CaCl₂ and MgCl₂; Invitrogen). Fluorescence was quantified using a POLARstar fluorimeter (BMG Labtech), with excitation (485 nm) and measurement of emission (520 nm) via the base of the plate.

Quantitation of expression in infected cells.

Mouse embryonic fibroblasts (MEFs), cultured in minimal essential medium-GlutaMAX supplemented with 10% fetal calf serum, 180 U/ml penicillin, and 45 U/ml streptomycin (Invitrogen), were seeded at 4 × 10⁴ cells per well in 96-well optical bottom black-walled plates (Fluorcarbon/Black; Nunc). After overnight growth, cultures were infected at a high multiplicity of infection (3 PFU/cell) with either K181 or recombinant MCMV expressing GFP-tagged M33 (wt or mutated). At 15 h postinfection, the culture medium was removed, and cells were washed (two times) using PBS and fixed in 2% paraformaldehyde-PBS for 15 min at room temperature. Cells were then washed (three times) with PBS and permeabilized using 0.2% Triton X-100-PBS for 15 min at room temperature. Cells were processed for immunofluorescence as follows (room temperature throughout): wash (twice in PBS), block (5% normal goat serum [NGS]-PBS for 30 min), wash (one time in PBS), primary antibody (1/1,000 rabbit anti-GFP [Abcam AB290] in 1% NGS-PBS for 1 h), wash (three times in PBS), secondary antibody (1/1,000 goat anti-rabbit Alexa Fluor 488 conjugate [Molecular Probes A-11034] in 1% NGS-PBS for 1 h), and wash (three times in PBS). Fluorescence was quantified by fluorimeter as described above.

Fluorescent imaging.

HEK-293 cells (10⁶ cells/well in six-well trays; Corning) were grown in antibiotic-free DMEM-GlutaMAX supplemented with 8% fetal calf serum for 24 h prior to transfection with 2 μg of plasmid DNA using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. At 6 h after transfection, cells were trypsinized and distributed evenly across three wells of glass coverslip cultures in 24-well trays (Corning). Following a further 18-h incubation at 37°C, coverslips were processed for confocal microscopy, using HBSS plus CaCl₂ and MgCl₂ (Invitrogen) for all wash and incubation steps, as follows. Cells were washed (two times) and then incubated (for 10 min at 37°C) with Image-iT LIVE plasma membrane and a nuclear labeling kit (Invitrogen), using 200 μl of labeling solution (5 μg/ml Alexa Fluor 594 wheat germ agglutinin) per coverslip. Cells were then washed (two times) and fixed using 2% paraformaldehyde-HBSS for 15 min at 37°C. Fixed cells were washed (three times), and coverslips were mounted onto glass slides using ProLong Gold Antifade reagent (Invitrogen). Slides were examined using a Leica TCS SP2 confocal microscope fitted with a 63× objective, and images were acquired with the Leica confocal software using a zoom factor of ×2.5. Images were imported into Adobe Photoshop CS2, version 9, and overlays of the GFP and Alexa Fluor 594 images were produced.

Generation of recombinant virus.

Purified genomic DNA was obtained from MEFs infected with KΔ33B_T2 recombinant virus, which contains the E. coli lacZ cassette within the second exon of M33 (15). Accordingly, this parent virus exhibits blue plaques upon staining with 5-bromo-4-chloro-3-inolyl-β-d-galactopyranoside. Transfection of MEFs was performed using the calcium phosphate method as previously described (16). Recombinant virus progeny were selected by the presence of “white” plaques. Recombination frequencies were of the order of 0.5 to 4%. Recombinant viruses were plaque purified at least twice and verified by sequencing across the M33 region. As a control, five independent transfections were performed between KΔ33B_T2 genomic DNA and the wt recombination plasmid pING/M33 to establish whether “repair” of the M33 region conferred wt properties of the resulting recombinant progeny in each case.

Growth of MCMV recombinants in vitro.

Following several rounds of plaque purification, working stocks of MCMV recombinants were prepared, and titers were determined by plaque assay on MEFs. Multistep growth of each of the recombinants in NIH-3T3 cells was compared with that of wt MCMV over a 5-day period. Subconfluent cultures of NIH-3T3 cells were prepared in six-well dishes and infected with each of the MCMV recombinants or wt MCMV at a multiplicity of infection (MOI) of 0.01. Infected culture medium (cell free) was harvested daily for 5 days and stored at −80°C until quantification by plaque assay on MEFs. Values shown are the average from two independent cultures for each virus.

Growth of MCMV recombinants in vivo.

In the first set of experiments (“low dose”), groups of five weanling BALB/c female mice were inoculated intraperitoneally with 5 × 10³ PFU of either wt MCMV or recombinants. Seventeen days postinoculation, the mice were euthanized, and the virus titer from the pooled salivary glands was determined by plaque assay on MEFs. A second experiment (“high dose”) was performed in the same manner, except a higher virus dose was used as the inoculum (4 × 10⁴ PFU), and the titers from the salivary glands were assessed from individual mice. Results are expressed as the total number of PFU recovered per salivary gland.

Western blotting.

COS-7 cells (2 × 10⁵ cells/well in six-well trays) were seeded at 37°C and 10% CO₂ in minimal essential medium-GlutaMAX supplemented with 10% fetal calf serum (antibiotic free) prior to transfection with 1 μg/well of plasmid DNA using Lipofectamine 2000 (Invitrogen) according to the manufacturer's guidelines. Untransfected cells were used as controls. At 24 h posttransfection, cells (harvested from two wells for each sample) were washed twice with PBS and lysed with 200 μl of ice-cold nonionic solubilization buffer (20 mM Tris-HCl, 150 mM NaCl, 1% sodium deoxycholate, 1% NP-40, 2 mM EDTA, and 10 mM iodoacetamide, supplemented with 1 protease inhibitor cocktail tablet/10 ml [Roche] prior to use). Samples were centrifuged at 17,900 × g for 15 min at 4°C, and the supernatants were discarded. Pellets were lysed in 50 μl of the above solubilization buffer that had been supplemented with 0.1% sodium dodecyl sulfate (SDS). Samples were centrifuged at 17,900 × g for 15 min at 4°C to remove insoluble material. Soluble lysates were mixed 1:1 with 2× reducing sample buffer (125 mM Tris-HCl, 20% glycerol, 4% SDS, 0.2% bromophenol blue, and 200 mM dithiothreitol) and loaded directly (without heating) onto 10% polyacrylamide gels, using 16 μl of diluted lysate (derived from 6.4 × 10⁴ cells) per lane. Duplicate gels were stained with Coomassie for total protein detection. Following SDS-polyacrylamide gel electrophoresis, protein was transferred to nitrocellulose, blocked with PBS-0.05% Tween 20 (PBS-T) containing 5% bovine serum albumin (BSA), and probed with a mouse anti-GFP monoclonal antibody (monoclonal antibody JL8; BD Living Colors). Following extensive washing with PBS-T, membranes were probed with a horseradish peroxidase-conjugated secondary antibody (Dako) in PBS-T containing 5% bovine serum albumin and again washed thoroughly in PBS-T, and bound antibody was visualized with an enhanced chemiluminescence kit (Amersham Biosciences). Protein separation was visualized with rainbow prestained markers (Bio-Rad); biotinylated protein markers (Bio-Rad) were additionally used for marker visualization on ECL Hyperfilm.

Statistics.

For CREB and NFAT signaling assays, wt and mutant M33 samples were tested alongside the pcDNA3 empty vector control in quadruplicate in each assay. Assays were repeated two to four times, giving 8 to 16 sets of data in each case. For phosphatidylinositol (PI) turnover assays, both groups were tested in triplicate, with assays repeated three times, for a total of nine sets of data. Within each assay, raw data were normalized as a percentage of the negative control (100 × S/C, where S is the value of the sample and C is the mean value of the vector control samples). The transformed values were then used to calculate the overall means and standard errors for each sample from the combined values of the repeated assays, expressed as a percentage of pcDNA3. For determination of expression levels in transfected cells by fluorimetry, samples were tested in quadruplicate and analyzed as above (n = 4). For determination of expression levels in infected cells by fluorimetry, six replicates were used (n = 6), and results were normalized to K181-infected (GFP negative) control wells. Graphing and statistical tests (analysis of variance with Bonferroni posttests for comparison of wt M33 with mutants) were performed using GraphPad Prism, version 5.00, for Windows (www.graphpad.com).

Cell surface expression and signaling properties of wt M33 and UL33.

It has been demonstrated previously that C-terminally tagged M33 and the HCMV homologue, UL33, are expressed at the cell surface and are constitutively active (24, 68). In the absence of available antiserum to M33, both N-terminal (c-myc and HA) and C-terminal (EGFP) tagging strategies were assessed to determine the optimal method for the detection of wt M33 at the cell surface, to which subsequent M33 mutants could be compared. Cellular expression of wt M33 in HEK-293 cells was examined by confocal microscopy, using wheat germ agglutinin as a marker for cell surface expression; wt UL33 was included as a comparison. Both wt M33-GFP and wt UL33-GFP were clearly detected at the cell surface as well as in the perinuclear Golgi region and small vesicular structures (see Fig. 3), consistent with previous studies. In contrast, the HA and c-myc N-terminal tags of wt M33 each failed to be detected at the cell surface, with concomitant loss of constitutive signaling (data not shown), suggesting that the N-terminal region of M33 is important for translocation of the protein to the cell surface. Consequently, C-terminal EGFP tagging was used to compare the distribution of M33 mutants with that of wt M33.

Cell surface expression and signaling properties of M33 mutants.

To identify regions of critical importance to M33 expression and signaling, N- and C-terminal truncations, as well as TM II and TM III point mutations were generated as summarized in Fig. 2, and the mutants were examined for cellular expression and constitutive CRE- and NFAT-mediated signaling in HEK-293 cells.

N- and C-terminal truncation mutants of M33.

While the N termini of seven-transmembrane receptors (7TMR) are not generally implicated in the regulation of constitutive 7TMR signaling per se, signaling could potentially be modulated via their interaction with cognate ligands. Thus, for M33, it was predicted that little or no effect in constitutive signaling in vitro would result following mutation of the N terminus. Nevertheless, if M33 signaling in vivo was regulated by ligand interaction, then such a mutation may exhibit an in vivo phenotype. A deletion of 10 amino acids from the N terminus (M33ΔN10) prevented detection of the receptor at the cell surface (Fig. 3B) and ablated activation of CRE- and NFAT-mediated transcription (Fig. 4A and B). Taken together with the observations of disrupted cellular distribution and signaling by the N-terminally tagged (HA or c-myc) M33 mutants, these data suggest that the N terminus is required for correct subcellular trafficking of M33.

Regulation of G protein coupling and cell surface expression have been ascribed to the C termini of numerous 7TMR. Thus, we examined how progressive C-terminal deletions of M33 affected signaling, expression, and in vivo function. Truncations of 8, 24, 38, and 57 amino acids (designated M33ΔC8, M33ΔC24, M33ΔC38, and M33ΔC57, respectively) were generated. Of these truncation mutants, M33ΔC8, M33ΔC24 and M33ΔC38 (lacking amino acids 370 to 377, 354 to 377, and 340 to 377, respectively) were all localized at the cell surface, but there was no detectable cell surface expression for the mutant with the largest deletion (amino acids 321 to 377), M33ΔC57 (Fig. 3B). M33ΔC8 and M33ΔC24 were constitutively active in CREB and NFAT signal transduction pathways (Fig. 4A and B). Notably, M33ΔC8 stimulated CRE-mediated transcription that was between 70 to 130% higher than wt M33 at all doses examined (the differences were statistically significant, with a P value of <0.001 between 1 to 5 ng/well). However, CRE- and NFAT-mediated transcription was undetectable for M33ΔC38 despite its localization at the cell surface. As expected, there was no constitutive activation for M33ΔC57. These data indicate that while the final 38 amino acids (340 to 377) of the M33 C terminus are dispensable for cell surface expression, the region between amino acids 340 to 353 of M33 (present in M33ΔC24 but deleted in M33ΔC38) is important for constitutive CRE- and NFAT-mediated signaling.

M33 TM II point mutations.

Within the second predicted TM domain of the 7TMR family is a highly conserved, positively charged aspartic acid located at II:10, according to the numbering system introduced by Baldwin (1) and modified by Schwartz (58). In many 7TMR this residue has been shown to be important for either ligand binding or signal transduction or both. In addition, it has a major role in the sensitivity of some 7TMR to sodium ion regulation of ligand binding (11, 20). ORF74 from HHV-8 lacks this conserved acidic residue, and it was recently demonstrated that the introduction of aspartic acid at this locus resulted in loss of constitutive activity but retention of its ability to be stimulated by agonist (50). M33 also lacks an acidic amino acid within TM II, and therefore several residues in this region were altered to aspartic acid to assess the effect on receptor function. These included N⁸¹ [M33(N81D); II:10], which is the residue most likely to correspond to the conserved Asp determined by alignment with other 7TMR, and the surrounding residues, F⁷⁹ [M33(F79D); II:08], A⁸⁰ [M33(A80D); II:09], and L⁸² [M33(L82D); II:11]. In addition, residue Y⁷⁸ (II:07) was mutated to alanine M33(Y78A) as alanine is much more commonly found in 7TMR at this position. Of the TM II mutants, only M33(Y78A) and M33(F79D) could be detected at the cell periphery (Fig. 3C). The three other mutants that were not detected at the cell surface were negative for constitutive signaling (Fig. 4C and D). M33(Y78A) retained constitutive activation of both CRE- and NFAT-mediated transcription. In contrast, despite its detection at the cell surface, M33(F79D) did not stimulate either signaling pathway (Fig. 4C and D).

Mutations of the M33 NRY motif.

An amino acid triplet motif, (D/E)RY, is found at the cytoplasmic end of TM III of most 7TMR. This sequence plays an important role in 7TMR activation. Of these residues, the arginine is the most highly conserved, and mutations of this residue generally confer a loss of G protein-mediated signaling (2, 44, 57, 71). In contrast, mutational analysis of the D/E residue, whereby the acidic residue is replaced by a neutral residue, has revealed a gain of constitutive activity for several receptors (42, 47, 57). UL33 has the prototypic DRY motif, whereas the equivalent motif within R33 and M33 is NRY. To assess the effects of conversion of this motif to the conventional DRY or modification of the arginine in M33, both N130D and R131Q mutations were made. Expression of both mutant receptors was detected at the cell surface similar to wt M33 (Fig. 3D); however, only M33(N130D) was constitutively active (Fig. 4E and F). This mutant activated CRE-mediated signaling at a level 60 to 200% greater than wt M33 levels (P < 0.05, 1 to 5 ng), whereas a much higher increase from 200 to 440% was observed through NFAT signaling (P < 0.05, 0.25 to 5 ng/well).

PLC activity.

The mutant receptors that retained cell surface distribution were further assessed for membrane-associated PLC activation by measurement of PI turnover across a plasmid dose range. In agreement with previous reports (60, 68), constitutive activity was observed for wt M33 (Fig. 5). All mutants which were positive for CREB and NFAT signaling[M33ΔC8, M33ΔC24, M33(Y78A), and M33(N130D)] were positive for PI turnover (Fig. 5), although in the case of (Y78A)M33the level of signaling was somewhat reduced (30 to 60% lower than wt M33; P < 0.01, 3 to 10 ng/well). However, while no CREB- or NFAT-mediated signaling was detected for the M33ΔC38 mutant, there was activation of PLC (Fig. 5A), albeit at 30 to 50% reduced levels compared with wt M33 (the difference was statistically significant only at 6 ng; P < 0.001). The other two mutants that were negative for CREB- and NFAT-mediated signaling [M33(F79D) and M33(R131Q)] were also negative for IP turnover (Fig. 5B and C). Therefore, the effect most mutants had on PI turnover was similar to its effect on the downstream CREB and NFAT signal transduction pathways.

Detection of wt M33 and mutated derivatives by Western blotting.

To determine whether the M33 mutants were aberrantly expressed compared with wt M33, lysates from transfected COS-7 cells were examined by Western blotting, using an anti-GFP monoclonal antibody (JL8; Clontech). Due to the number of constructs examined, the transfections and Western blotting were carried out in batches on separate occasions, with wt M33 included each time as a control. Typical results are shown in Fig. 6A to D. The GFP-tagged wt M33 protein was detected as a major band running slightly lower than the predicted mass of 69 kDa. Slight anomalies in migration from the predicted size may have been due to the fact that samples were not heated prior to loading to prevent M33 aggregation. Point mutants of M33 displayed the same mobility as wt M33, while truncations conferred minor changes in mobility, consistent with the sizes of deletion. Higher forms of each of the proteins were also detected, which is consistent with oligomer formation. These studies confirmed that expression of wt M33 and the mutated derivatives was at approximately the expected mass and suggest that the mutations did not affect oligomerization of the receptors. For wt M33 and the various mutants, a minor, lower-molecular-weight band was also detected, which may be a breakdown product or otherwise truncated form of the full-length proteins.

Quantitation of expression of wt M33 and mutated derivatives by fluorimetry.

In order to determine the relative levels of protein expression for the various mutants, relevant to the signaling assays, HEK-293 cells were transfected with GFP-tagged mutant and wt M33, and total protein levels were determined via GFP fluorescence (Fig. 6E and F) All of the mutants which were negative for cell surface expression were found to have reduced levels of GFP fluorescence (generally 30 to 60% of wt M33 levels; P < 0.05 for all cell surface negative mutants from 12.5 to 200 ng/well). The remaining mutants were expressed at levels similar to wt M33 across the range of DNA doses tested, with the exception of M33ΔC38, which was consistently lower (approximately 60% wt M33 levels; P < 0.05, 25 to 200 ng/well).

In vitro and in vivo growth properties of MCMVs bearing M33 mutations.

A number of the M33 mutants were selected for the generation of MCMV recombinants. Prior to investigations of in vivo replication, each of the recombinants was assessed for replication in vitro. Multistep growth analysis revealed that there was no significant difference between the recombinants and wt in their ability to replicate in NIH-3T3 cells (Fig. 7). At various occasions during the course of construction and in vivo assay of mutant viruses, the ability of wt M33 to restore the wt M33 phenotype from the lacZ-disrupted M33 parental virus was assessed. Thus, recombinant progeny from five independent transfections that resulted in repair of the disrupted M33 region were evaluated for their ability to disseminate to the salivary glands in weanling BALB/c mice. Titers of M33 revertants were compared with those of M33-disrupted parent virus (KΔ33B_T2) and wt MCMV [K181 (Perth) strain] (Table 2). While the K181 virus strain replicated to high levels in the salivary glands, virus was not detected in mice infected with the KΔ33B_T2 parent recombinant, consistent with previous studies (16). In contrast, all MCMV “wt-reconstituted” revertants grew to levels similar to wt K181, demonstrating that restoration of the wt phenotype upon repair of M33 was reproducible. The large difference in titer of MCMV between wt and the M33 deletion mutant enabled us to determine whether selected M33 recombinants resulted in full or partial recovery of the wt phenotype for salivary gland replication.

The mutations incorporated into MCMV recombinants were as follows: (i) C-terminal truncations of 24 and 38 amino acids; (ii) point substitutions of TM II F79D, TM III N130D, and TM III R131Q; and (iii) replacement of M33 with UL33. For the F79D mutant, the intron near the 5′ end of the M33 coding sequence was also lacking. In order to determine whether loss of the intron had a major effect upon virus replication, a recombinant with the intron removed but no other mutations of M33 was also examined. All recombinant viruses were sequenced across the entire M33 coding region to confirm the introduction of the mutation prior to conducting in vivo infections (data not shown). In addition, virus recovered from the salivary glands (where detected) was also sequenced to confirm the presence of the relevant mutation (data not shown).

The titers of viruses recovered from the salivary glands following a series of intraperitoneal infection experiments, using either a high- or low-dose inoculum, are shown in Table 2. Following infection at the lower dose (5 × 10³ PFU/mouse), the intronless, control recombinant (M33Δi) rescued salivary gland replication but yielded a salivary gland titer approximately 10-fold lower than that of the revertants that retained the intron. This indicates that retention of the M33 intron sequence is required for full salivary gland virulence, but loss of the intron results in a relatively minor attenuation. Whether this effect is mediated via transcriptional regulation of M33 expression or some other mechanism is not known. The C-terminal truncation mutants had displayed similar cell surface distribution but markedly different signaling properties. Whereas M33ΔC24 (signaling similar to wt M33) grew to levels in the salivary glands that were up to 10-fold lower than wt MCMV, the replication of the ΔC38 truncation mutant (reduced signaling compared with wt) was reduced approximately 1,000-fold. Similar to the truncation mutants, the point mutants had retained cell surface expression but showed markedly different signaling properties. Whereas the M33(N130D) mutant (signaling increased compared to wt M33) replicated to wt MCMV levels in the salivary glands, the M33(F79D) mutation (negative for signaling) was negative for salivary gland replication. Following infection with an eightfold higher dose (4 × 10⁴ PFU/mouse), the level of M33ΔC38 virus recovered from the salivary glands increased to a level almost 100-fold higher than that recovered following the lower-dose inoculation. Even at this increased dose, there was a highly statistically significant difference between the M33ΔC38 titers and those of wt MCMV (P < 0.001; unpaired two-tailed Student's t test). In contrast, both signaling-inactive mutants, M33(F79D) and M33(R131Q), failed to yield detectable virus in the salivary glands at the higher dose. wt MCMV, M33ΔC24, and the M33(N130D) viruses gave modest increases (<10-fold) in salivary gland titer, compared with the lower-dose infection, suggesting that a maximum obtainable level of virus replication had been reached. There were no statistically significant differences between the titers of M33ΔC24 or M33(N130D) and wt MCMV (P > 0.1, Student's t test). In order to determine whether the three mutations resulting in in vivo attenuation were associated with reduced expression levels of M33 in the context of virus-infected cells, recombinant MCMV expressing GFP-tagged derivatives of each of the mutants (F79D, R131Q and ΔC38) were generated. M33 is expressed at a relatively low level in MCMV-infected cells, so M33-GFP expression was quantitated by fluorimetry following immunofluorescent staining with a polyclonal anti-GFP rabbit antibody. MEFs were infected at high multiplicity (MOI of 3 PFU/cell), and expression of M33-GFP in replicated cultures (n = 6) was determined at 15 h postinfection (Fig. 8). Expression levels of each of the mutated M33-GFP proteins were similar to that of wt M33-GFP, with no statistically significant differences between the mean levels (P > 0.05). Taken together, these data suggest that disruption of constitutive signaling is associated with an attenuated phenotype in vivo.

The UL33 substitution mutant enabled assessment of the ability of UL33 to functionally substitute for M33 in vivo. Full replacement of M33 with UL33 (including the endogenous intron of UL33) resulted in low, but detectable virus replication in the salivary glands, suggesting that UL33 can provide partial compensation for the absence of M33. This compensation was ablated in the UL33 virus recombinant with the intron removed.