UL45 of Human Cytomegalovirus Promotes Macrophage Survival and Modulates RIPK3-Associated Cell Death

Figure 1. UL45st mutagenesis. A stop codon and a single amino acid deletion were introduced at the start codon of the UL45 gene in TB40E-BAC4, resulting in a stop-frameshift mutation. (A) An overview of the two-step Red recombinase mutagenesis method used to create the UL45st mutant. (B) A summary of the specific mutation made to generate the UL45st mutant. (C) Restriction fragment-length analysis of the intermediate UL45st insert compared to original TB40E BAC DNA. The UL45st insert contained an additional ~4.5 kb DNA fragment containing the kanamycin resistance cassette compared to the original BAC DNA when digested with an EcoRI restriction enzyme (indicated by red box). (D) After generating infective viral particles from the BAC DNA, viral DNA was extracted from stock virus and sequenced by Illumina next-generation sequencing. Genome reads were then compared to a reference genome (TB40E-BAC4 isolate), and a variant allele frequency (VAF) plot was generated. Wild-type virus and UL45st virus share similar levels of low-frequency mutations (VAF < 0.25). An expected INDEL mutation (VAF = 1) occurs at 93,128 kb in the UL45st mutant with no other mutations in the genome. A previously…

Figure 2. Growth properties of wild-type TB40E virus and UL45st virus in HFF cells at high and low MOI. HFF cells were infected at an MOI of either 3 (panel (A)) or 0.01 plaque-forming units (PFUs) per cell (panel (B)). Titers of the resulting progeny from both whole-cell lysates and supernatants were determined at the indicated times post infection. The titer at 0 hpi represents the input virus (circles, wild-type TB40E; triangles, UL45st mutant) Each data point is the mean of 3 replicate cultures, with standard deviation bars displayed. Some data sets showed very little variability.

Figure 3. Cell viability of wild-type and UL45st-infected M1 or M2 macrophages. PBMC-differentiated M1 and M2 macrophages from 12 individuals were infected with wild-type TB40E virus and UL45st virus for 24 h. (A,B) 18 hpi, half the cells were treated with TNF-α, BV6, and Z-VAD-FMK. (C) Macrophages from half the donors were treated with GSK840 at the time of infection. Note that cells were differentiated into M1 and M2 macrophages as described previously, but data were pooled for analysis. Cell viability was measured using a Cell Titer Glo ATP assay. All values are normalized to the mock-infected and mock-treated macrophages and show standard deviation and mean for the individuals. For graphs (A–C), n = 12. p-Values (p) of <0.05 were considered significant and are indicated with ns (not significant), * p < 0.05, ** p < 0.01, **** p < 0.0001.

Figure 4. Necrotic-like cell death in UL45st-infected macrophages. Brightfield images of WT and UL45st-infected macrophages at 24 hpi. Nuclei are indicated by arrows. Asterisks (*) indicate large vacuoles in cells (if present). (A) Representative TB40E-infected cell. (B–D) Brightfield images of selected UL45st-infected cells with significant vacuolization. The additional fluorescent panels correspond to panels (A,D), respectively. (E,F) Fluorescence imaging of TB40E (E) or UL45st (F) stained with DAPI (to indicate location of nuclei), and indirect immunofluorescence using antibodies recognizing HCMV IE1 (an infection marker) and host RIPK1 detected using species-specific fluorescently labelled secondary antibodies, along with a brightfield image of the same cell. Note that panels (E,F) are the same cells as panels (A,D), respectively, and are shown at the same magnification.

Figure 5. Caspase-dependent and -independent cell death in multiple cell types. (A) Human colon carcinoma cell line (HT29), primary PBMC differentiated into M1-like macrophages (Mφ), primary human fibroblasts (HFFs), a and a fetal lung fibroblast cell line (MRC5) were treated for 24 h with a combination of TNF-α: tumor necrosis factor-alpha (T), BV6: a Smac-mimetic (S), Z-VAD-FMK: a pan-caspase inhibitor (V), and/or GSK 840: a RIPK3 inhibitor (GSK). Twenty-four hours post treatment, cell viability was measured by Cell Titer Glo ATP luciferase assay. (B) HFF, HT29, M1-like macrophages and MRC5 cells were harvested and tested for endogenous RIPK3 and MLKL levels by Western blot. (C) Densitometric analysis of endogenous RIPK3 and MLKL levels to β-actin.

Figure 6. pUL45 partially rescues caspase-independent and caspase-dependent cell death in HT29 cells. HT29 cells were transfected with an empty vector (EV), a plasmid expressing UL45 (UL45myc) or a plasmid expressing UL45 lacking an IPAM domain (ΔIPAM). Twenty-four hours post transfection, media was replaced with cell death media for 24 h that contained the following drugs: TNF-α and BV6 (TS); TNF-α, BV6, and GSK840 (TS+GSK840); TNF-α, BV6, and Z-VAD-FMK (TSV); or TNF-α, BV6, Z-VAD-FMK, and GSK840 (TSV+GSK840). Twenty-four hours post treatment, relative numbers of viable, metabolically active cells were measured with a Cell Titer Glo ATP luciferase assay. p-Values (p) of <0.05 were considered significant and are indicated with ns (not significant), ** p < 0.01, *** p < 0.001.