Human cytomegalovirus entry into epithelial and endothelial cells depends on genes UL128 to UL150 and occurs by endocytosis and low-pH fusion

Abstract

Human cytomegalovirus (HCMV) replication in epithelial and endothelial cells appears to be important in virus spread, disease, and persistence. It has been difficult to study infection of these cell types because HCMV laboratory strains (e.g., AD169 and Towne) have lost their ability to infect cultured epithelial and endothelial cells during extensive propagation in fibroblasts. Clinical strains of HCMV (e.g., TR and FIX) possess a cluster of genes (UL128 to UL150) that are largely mutated in laboratory strains, and recent studies have indicated that these genes facilitate replication in epithelial and endothelial cells. The mechanisms by which these genes promote infection of these two cell types are unclear. We derived an HCMV UL128-to-UL150 deletion mutant from strain TR, TRdelta4, and studied early events in HCMV infection of epithelial and endothelial cells, and the role of genes UL128 to UL150. Analysis of wild-type TR indicated that HCMV enters epithelial and endothelial cells by endocytosis followed by low-pH-dependent fusion, which is different from the pH-independent fusion with the plasma membrane observed with human fibroblasts. TRdelta4 displayed a number of defects in early infection processes. Adsorption and entry of TRdelta4 on epithelial cells were poor compared with those of TR, but these defects could be overcome with higher doses of virus and the use of polyethylene glycol (PEG) to promote fusion between virion and cellular membranes. High multiplicity and PEG treatment did not promote infection of endothelial cells by TRdelta4, yet virus particles were internalized. Together, these data indicate that genes UL128 to UL150 are required for HCMV adsorption and penetration of epithelial cells and to promote some early stage of virus replication, subsequent to virus entry, in endothelial cells.

FIG. 1.

HCMV genome and structure of UL128-to-UL150 mutant TRΔ4. (A, lines 1 and 2) Schematic diagrams of the HCMV AD169 and TR genomes. Indicated are the unique long (U_L) and unique short (U_S) segments and inverted terminal and internal repeats (gray and black rectangles). The open rectangle represents the UL128-to-UL150 genes. AD169 lacks most of the UL128-to-UL150 genes and has mutations in others (35). (Line 3) The UL128-to-UL150 region of the TR genome. Hatched rectangles represent 60-bp sequences used to mediate homologous recombination with TR-BAC sequences. The positions of EcoRI restriction sites (Ec) and the predicted fragment sizes (10 kb and 22.3 kb) are indicated. (Line 4) Schematic diagram of UL128-to-UL150 genes in TRΔ4 replaced with LacZ/Amp^r sequences (wide horizontal black rectangle). (B) Southern blot analysis of viral DNA. Eight ampicillin-resistant BAC clones were screened for the replacement of UL128-to-UL150 genes with LacZ/Amp^r sequences. DNA was digested with EcoRI and subjected to electrophoresis and Southern blot analysis with a probe directed against Amp^r gene sequences. W.T., parental (wild type) TR. A 15-kb fragment was expected for recombinants in which UL128 to UL150 were replaced with LacZ/Amp^r sequences.

FIG. 2.

Effects of polyethylene glycol and centrifugation on infection of ARPE-19 epithelial cells by HCMV. Replicate cultures of ARPE-19 cells were inoculated with 10 PFU/cell of HCMV strain TR, TRΔ4, or AD169 and were either incubated at room temperature for 2 h or centrifuged at 800 × g at room temperature for 2 h. All cultures were then shifted to 37°C for 2 h and subsequently treated with either PBS or 44% PEG for 30 s. After 48 h, cells were fixed and analyzed by immunofluorescence to detect HCMV IE86. Representative fields are shown.

FIG. 3.

Effects of polyethylene glycol and centrifugation on infection of t-HUVECs by HCMV. Replicate cultures of t-HUVECs were inoculated with 10 PFU/cell of HCMV strain TR, TRΔ4, or AD169 and subsequently centrifuged as in Fig. 2 and treated with 40% PEG. After 48 h, IE86 was detected by immunofluorescence. Representative fields are shown.

FIG. 4.

Internalization of radiolabeled HCMV particles. (A) Purified [³H]thymidine-labeled HCMV TR particles were centrifuged onto fibroblasts or epithelial cells at 4°C for 2 h. Cells were washed extensively with PBS and shifted to 37°C for 0, 30, or 120 min. Radiolabeled particles that remained on the cell surface were removed by placing the cells on ice and treating them with PK at 4°C. The internalization of virus was calculated as follows: % of PK-resistant virus after 30 or 120 min (virus remaining with cells after PK treatment/virus remaining with cells not PK treated) − % PK-resistant virus at 0 min (virus remaining with cells after PK treatment at 0 min/virus remaining with cells not PK treated at 0 min). (B) Radiolabeled TR, AD169, or TRΔ4 particles were centrifuged onto fibroblasts, epithelial cells, or endothelial cells at 4°C, and the cells were washed extensively with PBS. Cells were treated with proteinase K immediately or after 3 h at 37°C. Shown is the internalization of virus as described for panel A. Values are the mean of three experiments, and error bars represent standard deviations.

FIG. 5.

Effects of lysosomotropic agents on HCMV infection of fibroblasts, epithelial cells, and endothelial cells. NHDF, ARPE-19 or t-HUVEC cultures were either not treated (open bars) or treated (black bars) for 1 h with (A) 50 mM NH₄Cl, (B) 35 nM bafilomycin, or (C) 100 μM chloroquine and then infected with HCMV TR (5 PFU/cell). hpi, hours postinfection. Alternatively, drugs were added 4 h after the addition of virus (gray bars). After 20 h (fibroblasts and epithelial) or 8 h (endothelial), cells were fixed and analyzed by immunofluorescence for the HCMV IE86 protein and DAPI-stained nuclei. For each condition, at least 350 cells were counted in random fields and the ratio of IE86-positive nuclei to DAPI-positive nuclei was determined. Each experiment was performed at least three times, and representative results are shown.

FIG. 6.

Plaque formation by HCMV on fibroblasts and epithelial cells. Fibroblasts or ARPE-19 epithelial cells were infected with HCMV strain TR, AD169, or TRΔ4 at approximately 20 PFU/well, and the cells and virus were centrifuged at 800 × g for 2 h at room temperature. The epithelial cells were then treated briefly with 44% PEG. All cultures were incubated at 37°C in the presence of neutralizing antibody for 21 days, fixed, and analyzed by immunofluorescence for HCMV IE86 protein (red) and gB (green). Representative fields are shown. The magnification of the panels showing epithelial cells was three times greater than that for fibroblasts.