High titer virus stocks of bacterial artificial chromosome (BAC)-derived mCMV-egfp (Angulo et al., 2000), mCMV-flox-egfp and mCMV-rec-egfp (Sacher et al., 2008) were prepared from infected murine embryonic fibroblasts (MEF) according to standard protocol (Podlech et al., 2002). C57BL/6 mice were purchased from Janvier Labs. Mcpt5-cre transgenic mice (Scholten et al., 2008) were bred in-house. All mice were housed under specified pathogen-free conditions in the translational animal research center (TARC) of the University Medical Center of the Johannes Gutenberg-University Mainz, Germany. MEF were prepared from C57BL/6 mice by a standard protocol (Podlech et al., 2002).

Bone marrow (BM)-derived mast cells (BMMC) from C57BL/6 and Mcpt5-cre mice were generated by using standard procedures (Stassen et al., 2006). In brief, BM cells were isolated and subsequently cultivated in Iscove's modified Dulbecco's medium (IMDM), supplemented with 10% FCS, 2 mM L-glutamine, 1 mM sodium pyruvate, 100 U/ml penicillin, 100 μg/ml streptomycin, 20 U/ml murine IL-3, 200 ng/ml stem cell factor (SCF; c-kit ligand), and 50 U/ml murine IL-4. Non-adherent cells were transferred to fresh culture plates every week to remove adherent macrophages and fibroblasts. After 4–5 weeks of culture, the resulting BMMC population had reached a purity of more than 95%.

mCMV-Δm38.5-egfp was generated by partial deletion of ORF m38.5 between nucleotides 51,958 and 52,368 (Rawlinson et al., 1996, GenBank accession number NC_004065) through en-passant mutagenesis of the BAC plasmid pSM3fr-rev (Angulo et al., 2000) as described (Tischer et al., 2010). Briefly, a kanR cassette flanked by homologous viral sequences was amplified from plasmid pori6K-RIT (Hammer et al., 2018) using the oligonucleotides Δm38.5_for GCG AAC ATC CTC TCG GTG TTC GGC ACG ATT GTC GTT GTC GAC ACC GCC ATC GCT GTC CCC TCC ACA ACT ACA ACC GAC GCA TCG TGG CCG GAT CTC and Δm38.5_ rev TAA AAG TTG CGC GGA CGG TGC CGC GGG TTG TAG TTG TGG AGG GGA CAG CGC GAT TGT CGT TGT CGA CAC CGC CAT GTG ACC ACG TCG TGG AAT GC. The resulting product was transformed into GS1783 bacteria carrying pSM3fr-rev. After Red recombination, arabinose-induced I-SceI expression, and a second round of Red recombination, BAC pΔm38.5-egfp was purified, and successful mutagenesis was confirmed by sequencing (Eurofins Genomics). After BAC DNA purification, the recombinant mCMV was reconstituted by transfection of the DNA into MEF and propagated for six passages until residual BAC sequences were lost as verified by PCR (Lemmermann et al., 2010).

(i) Infection of BMMC. BMMC were centrifugally infected with 0.2 plaque-forming-units (PFU) of mCMV per cell, corresponding to a multiplicity of infection (MOI) of 4 (Podlech et al., 2002, and references therein), stimulated with 1 μM ionomycin in supplemented IMDM (see above), and inspected under the Axiovert 200 M fluorescence microscope (Carl Zeiss). (ii) Infection of peritoneal exudate cells (PEC). PEC were isolated from the peritoneal cavity of C57BL/6 mice by standard protocol (Ray and Dittel, 2010). 1 × 10⁶ PEC were seeded in IMDM medium supplemented with 5% FCS, 100 U/ml penicillin, and 100 μg/ml streptomycin and centrifugally infected with 0.2 PFU per cell (MOI 4) of mCMV directly after seeding. (iii) Infection of mice. Intraperitoneal infection of C57BL/6 mice was performed with 5 × 10⁵ PFU of mCMV diluted in 300 μl PBS. Infected PEC were isolated 18 h later for analysis.

C57BL/6 and Mcpt5-cre BMMC were infected with mCMV-flox-egfp or mCMV-rec-egfp. Five days later, cell culture supernatants were harvested and released infectious virus was quantitated on permissive C57BL/6 MEF by standard plaque assay under conditions of centrifugal enhancement. Total plaque formation and eGFP⁺ (green) plaques were counted 3 days later under the Axiovert 200M fluorescence microscope.

To generate m38.5 expression plasmid, PCR was performed using mCMV MW97.01 (Wagner et al., 1999) genomic DNA with primers pIRES-m38.5-XhoI-for ATA TCT CGA GAT GGA GAG TGT GCG CC and pIRES-m38.5-EcoRI-rev GCG CGG AAT TCC TAG AAT GTG TAA TCT C, amplifying the full length ORF m38.5. The PCR product was subcloned within the XhoI and EcoRI restriction sites into vector pIRES2-egfp (Takara Bio). Successful generation of pIRES2-m38.5-egfp was confirmed by sequencing (Eurofins Genomics).

BMMC (1 × 10⁶ cells in 0.2 ml serum-free IMDM) were transfected with 7 μg plasmid DNA by electroporation in 0.2-cm cuvettes at room temperature using a Bio-Rad Gene Pulser (Richmond, CA) set at 290 V, 200 Ω, and 600 μF. Cells were allowed to recover for 4 h in IMDM supplemented with 5% FCS and antibiotics, harvested, washed with PBS containing 1% BSA and 2 mM EDTA, and analyzed by flow cytometry as outlined below.

BMMC were harvested 4 h after transfection. Cell surface staining was performed with the following antibodies for cytofluorometric analysis: BV421-conjugated anti-CD117 (clone ACK2, eBioscience), PE-Cy7-conjugated anti-FcεRI (clone MAR-1; BioLegend, San Diego, CA, USA), and eFluor660-conjugated anti-CD107a (clone 1D4B; eBioscience). PEC were isolated and stained accordingly ex vivo after either in corpore or ex corpore infection. Optionally, to facilitate gating for peritoneal exudate-derived MC (PEMC), PEC were additionally stained with APC-conjugated anti-Ly-6A/E (Sca-1) (clone D7; BioLegend). Derivation of PEC and PEMC from hematopoietic lineages was verified by staining with BV510-conjugated anti-panCD45 (clone 30-F11, BioLegend). PEC preparations contain almost no cells other than leukocytes. Doublets were excluded by gating on singlets in the sideward scatter height (SSC-H) vs. area (SSC-A) plot prior to setting a “live cell gate” based on scatter properties SSC-A and forward scatter area (FSC-A). Gating for cells expressing eGFP and MC cell surface markers was done as established in our own previous work, in which an isotype control for degranulation marker CD107a was used to exclude unspecific antibody binding to infected MC (Becker et al., 2015). Cytofluorometric analyses of cell surface staining and intracellular eGFP fluorescence were performed with BD FACSymphony (BD Biosciences) and BD FACSDiva V6.1.3 using FlowJo V.10.6.2 software (BD Biosciences).

To evaluate statistical significance of differences between two independent sets of data, Student's t-test (two-sided, unpaired) was used with Welch's correction to account for unequal variance. Differences were considered as being statistically significant at levels of P < 0.05 (*), < 0.01 (**), and < 0.001 (***). Calculations were performed with Graph Pad Prism 8.4.1 (Graph Pad Software).

Previous work with eGFP-tagged virus has shown that mCMV can infect peritoneal exudate-derived CD117⁺FcεRI⁺ MC (PEMC) in corpore and can trigger their degranulation (Ebert et al., 2014). Recombination of mCMV-flox-egfp (Sacher et al., 2008) in cre-transgenic B6-Mcpt5-cre mice, which express Cre recombinase selectively in MC, generated infectious mCMV-rec-egfp (Becker et al., 2015) that also spread to other cell types such as hepatocytes and endothelial cells in the liver (Podlech et al., 2015). In contrast, BMMC were found to be not permissive for infection, not even expressing the immediate-early (IE) phase protein IE1 (Ebert et al., 2014). This discrepancy between PEMC and BMMC was explained by different stages of MC maturation (Matsushima et al., 2004). So, a cell-culture system for infection of MC was missing so far.

Interestingly, pre-treatment of BMMC with the Ca²⁺ ionophore ionomycin, which rapidly raises the intracellular level of Ca²⁺, renders MC permissive to mCMV infection (Figure 1A). This gives a first hint to a critical role of Ca²⁺ levels in the infection of MC and suggests that the previously observed difference between PEMC and BMMC (see above) was based on differences in Ca²⁺ mobilization. Although BMMC cultures reach high MC purity, some contamination by macrophages, which are permissive to mCMV infection, cannot be excluded. To verify that productively infected cells in BMMC cultures are MC, we again used the strategy of MC-specific Cre recombination to remove the stop cassette in virus mCMV-flox-egfp for eGFP expression selectively in MC (Figure 1B). Indeed, whereas pre-recombined mCMV-rec-egfp infected BMMC derived from both C57BL/6 (B6) and cre-transgenic B6-Mcpt5-cre mice produced infectious eGFP⁺ progeny, mCMV-flox-egfp generated infectious eGFP⁺ progeny only from B6-Mcpt5-cre BMMC (Figures 1C,D). This gives final evidence for productive infection of BMMC.

As elevated levels of cytosolic Ca²⁺ are known to be decisive for MC degranulation (for reviews, see Ma and Beaven, 2011; Wernersson and Pejler, 2014), and given the experience with the Ca²⁺ ionophore ionomycin in the infection of BMMC (this report), we considered the possibility that a viral protein involved in Ca²⁺ mobilization might be required for the degranulation of mCMV-infected MC. As hCMV vMIA-UL37x1 has been reported to mobilize Ca²⁺ from endoplasmic reticulum stores into the cytosol (Sharon-Friling et al., 2006), we considered mCMV vMIA-m38.5 as a candidate, even though this was thinking laterally because a functional analogy between mCMV and hCMV vMIAs in terms of Ca²⁺ regulation has not yet been established. To test a putative role for mCMV vMIA-m38.5 in MC degranulation, we transfected BMMC with gene m38.5 expression plasmid pIRES2-m38.5-egfp and indeed found elevated numbers of cells expressing the degranulation marker CD107a among transfected eGFP⁺CD117⁺FcεRI⁺ BMMC compared to the pIRES2-egfp empty vector control expressing only eGFP (Figure 2; for gating controls using pIRES2-egfp, see Figure S1A). This is the first positive evidence for a role of mCMV vMIA-m38.5 in MC degranulation.

As a reciprocal approach for identifying a role of m38.5 expression in MC degranulation, we isolated CD45⁺ PEC (Figure S2) and infected them ex corpore with the m38.5 gene deletion mutant mCMV-Δm38.5-egfp or with virus mCMV-egfp encoding m38.5. The deletion significantly reduced the numbers of degranulated CD107a⁺ eGFP⁺CD117⁺FcεRI⁺ PEMC (Figure 3; for gating in the mCMV-egfp control, see Figure S1B). Note that CD117⁺FcεRI^low cells were included in the analysis in order not to miss infected PEMC that have already downregulated FcεRI in response to activation (Rivera and Gilfillan, 2006; Yoshioka et al., 2007). Finally, to account for putative modulating influences of an in vivo cytokine/chemokine milieu in the peritoneal cavity, infection was performed in corpore and confirmed the importance of m38.5 expression for triggering MC degranulation (Figure 4; for gating in the mCMV-Δm38.5-egfp group, see Figure S1C).