Fig 1.
The Eph RTK interaction maps to the N-terminal region of KSHV gH.
A Co-Immunoprecipitation of depicted V5-tagged RRV/KSHV gH chimeras in complex with KSHV gL-Flag identifies KSHV gH domains necessary for binding of EphA2. Monoclonal antibody to the V5-tag was used for precipitation of complexes. Equal amounts of HA epitope-tagged EphA2 ectodomain was added to each reaction. EphA2, gH, gL were detected using their respective tag. RRV-derived parts of the chimeric constructs are displayed in light grey; lane 1: KSHV gH, lane 9: RRV gH. Four gH chimeras (lanes 3 and 5–7) were not expressed to detectable levels. All chimeras harboring domain I of KSHV gH (lanes 2, 4, 8) were still able to interact with EphA2. B Co-Immunoprecipitation of V5-tagged KSHV gH in complex with Flag-tagged RRV/KSHV gL chimeras identifies KSHV gL domains necessary for binding of EphA2. gH-V5/gL-Flag complexes were immunoprecipitated in the presence of full-length EphA2-myc using monoclonal antibody to the V5-tag and precipitates were analyzed by Western blot as in A. RRV-derived parts of the chimeric constructs are displayed in light grey; first lane: KSHV gL, fourth lane: RRV gL. C Co-Immunoprecipitation of V5-tagged KSHV gH in complex with a Flag-tagged C-terminally truncated KSHV gL mutant (gLΔ135–164). gH-V5/gL-Flag complexes were immunoprecipitated in the presence of EphA2-HA (ectodomain) using monoclonal antibody to the V5-tag and precipitates were analyzed by Western blot as in A. Full-length KSHV gH/gL serves as a positive control, KSHV gH alone serves as a negative control. Asterisks indicate non-specific bands. Abbreviations: D: domain, TM: transmembrane domain, SP: signal peptide, IVD: intravirion domain, IP: immunoprecipitation, IB: immunoblotting.
Fig 2.
Two amino acids of a conserved E-L-E-F-N motif in the N-terminal region of KSHV and RRV gH are essential for Eph interaction.
A Domain structure of KSHV and RRV gH. Multiple sequence alignment of domain I of gH of KSHV and the two RRV isolates 26–95 and 17577 (enlarged inset, numbers corresponding to KSHV gH). The EBV gH sequence is included as a reference. B Mutational scan of the N-terminal region of KSHV gH identifies EphA2-interacting residues. V5-tagged KSHV gH mutants were co-expressed with Flag-tagged KSHV gL. gH-V5/gL-Flag complexes were immunoprecipitated in the presence of full-length EphA2-myc using monoclonal antibody to the V5-tag and precipitates were analyzed by Western blot. KSHV gH alone serves as negative control. C Mutational scan of the N-terminal region of RRV gH identifies EphB3-interacting residues. V5-tagged gH mutants were co-expressed with Flag-tagged RRV gL. gH-V5/gL-Flag complexes were immunoprecipitated in the presence of full-length EphB3-myc using monoclonal antibody to the V5-tag and precipitates were analyzed by Western blot. RRV gH alone serves as negative control. Residues in the conserved E-L-E-F-N motif that are critical for Eph interaction are indicated by black lines. Asterisks indicate non-specific bands. Abbreviations: D: domain, TM: transmembrane domain, SP: signal peptide, IVD: intravirion domain, IP: immunoprecipitation, IB: immunoblotting.
Fig 3.
The E-L-E-F-N motif is located in a putative beta-hairpin at the KSHV and RRV gH/gL interaction site.
Homology-based structure prediction of the KSHV gH/gL and RRV gH/gL complexes based on the crystal structure of the EBV gH/gL complex (PDB number 3PHF) using the Iterative Threading ASSembly Refinement (I-TASSER) server and the CO-THreader algorithms for protein-protein complex structure and multi-chain protein threading. A KSHV gH/gL complex. Domain I is colored in blue. gL is colored in green. B Enlarged view of the inset indicated in A by dotted lines, showing the E-L-E-F-N motif in a putative beta-hairpin. Amino acids Glu52 and Phe53 that are critical for Eph binding are highlighted by asterisks. C RRV gH/gL complex. Domain I is colored in blue. gL is colored in green. D Enlarged view of the inset indicated in C by dotted lines, showing the E-L-E-F-N motif in a putative beta-hairpin. Amino acids Glu54 and Phe55 that are critical for Eph binding are highlighted by asterisks.
Fig 4.
A List of bacmid-derived recombinant viruses generated for this study. B Schematic representation of mutations introduced into gH/gL. C SLK cells infected with wt or mutant KSHV or RRV.
Fig 5.
Mutation of the E-L-E-F-N motif is sufficient for Eph receptor detargeting.
A Dose-dependent inhibition of KSHV infection by soluble EphA2-Fc on SLK cells. KSHV wt or gH-ELAAN were pre-incubated with EphA2-Fc. Fc alone and PBS were used as controls. GFP expression as indicator of infection was measured by flow cytometry. Infection without protein (PBS control) was set to 1 (duplicates, error bars represent range). B Dose-dependent inhibition of RRV infection by soluble EphB3-Fc on SLK cells. RRV 26–95 wt, ΔgL or gH-AELAAN were pre-incubated with EphB3-Fc. Fc alone and PBS were used as controls. YFP expression as indicator of infection was measured by flow cytometry. Infection with Fc 0.8nM was set to 1 (duplicates, error bars represent range). C Target cells were pre-incubated with a soluble ephrinA4-Fc fusion protein at 2μg/ml for 30min prior to infection with KSHV wt or gH-ELAAN. Infection was measured as in A. Infection without protein (PBS control) was set to 100% (triplicates, error bars represent sd). Non-normalized infection (%GFP+ cells) ±sd is listed below the respective bars. D Target cells were pre-incubated with a soluble ephrin-Fc fusion protein mix (ephrinA1, ephrinA2, ephrinA3, ephrinA4, ephrinA5, ephrinB1, ephrinB2, ephrinB3) at 2μg/ml each for 30min prior to infection with RRV 26–95 wt, ΔgL or gH-AELAAN. Infection was measured as in B. Infection without protein (PBS control) was set to 100% (triplicates, error bars represent sd). Non-normalized infection (%YFP+ cells) ±sd is listed below the respective bars. ns: not significant, *: p-value < 0.05, **: p-value < 0.01, ***: p-value < 0.001.
Fig 6.
Eph-binding-negative RRV and KSHV mutants exhibit normal attachment and reduced specific infectivity.
A Attachment of KSHV on LECs is not affected by mutational ablation of the Eph interaction. Cells were incubated with cold virus at the indicated concentrations at 4°C for 30min followed by genomic DNA isolation. The ratio of viral to cellular DNA as a measurement for attached virus was calculated based on ΔCt values of a genomic (CCR5) and a viral locus (ORF59, KSHV or ORF73, RRV) as determined by qPCR and plotted against input viral genome number. B Attachment of RRV on LECs is not affected by mutational ablation of the Eph interaction. Attachment was determined as in A. C-D Eph-binding-negative RRV and KSHV mutants exhibit a reduced specific infection. Target cells were infected with KSHV wt and gH-ELAAN (C) or RRV wt, gH-AELAAN and ΔgL (D) at the indicated virus concentrations. GFP (KSHV) or YFP (RRV) expression as indicator of infection was measured by flow cytometry. Solid lines represent one representive experiment (triplicates, error bars indicate sd). Dotted lines represent non-linear fitting of combined representative experiments of three independent pairs (KSHV) or two independent triplets (RRV) of virus stocks. The ratio Kwt/Kmutant of the rate constant K of fitted curves for wt (Kwt) and mutant viruses (KELAAN, KAELAAN, KΔgL) represents the effect of introduced mutations on specific infectivity. Bar graphs represent infections achieved by two specific input virus concentrations normalized to genome copies for one representative experiment per cell type. *: p-value < 0.05, **: p-value < 0.01, ***: p-value < 0.001.
Fig 7.
Contribution of the gH/gL-Eph interaction to RRV infection is cell type-specific.
A-B Comparative infection on LEC (A) or HUVEC (B) and RF by RRV wt, RRV gH-AELAAN, and RRV ΔgL. RF and LEC or HUVEC were infected with the same inocula of the respective virus stock, and the percentage of reporter gene-positive cells as determined by flow cytometry for each dilution was plotted. C Micrograph of RF and LEC infected with the same respective inocula of wt and Eph-binding-negative RRV gH-AELAAN. D-E Comparative infection on LEC (D) or HUVEC (E) and RF by RRV wt, RRV gH-AELAAN, and RRV ΔgL as in A based on mean fluorescence intensity (MFI) as determined by flow cytometry. ***: p-value < 0.001.