Senescence of endothelial cells increases susceptibility to Kaposi's sarcoma-associated herpesvirus infection via CD109-mediated viral entry

Abstract

The aging process is characterized by cellular functional decline and increased susceptibility to infections. Understanding the association between virus infection and aging is crucial for developing effective strategies against viral infections in older individuals. However, the relationship between Kaposi's sarcoma-associated herpesvirus (KSHV) infection, a cause of increased Kaposi's sarcoma prevalence among the elderly without HIV infection, and cellular senescence remains enigmatic. This study uncovered a link between cellular senescence and enhanced KSHV infectivity in human endothelial cells. Through a comprehensive proteomic analysis, we identified caveolin-1 and CD109 as host factors significantly upregulated in senescent cells that promote KSHV infection. Remarkably, CRISPR/Cas9-mediated KO of these factors reduced KSHV binding and entry, leading to decreased viral infectivity. Furthermore, surface plasmon resonance analysis and confocal microscopy revealed a direct interaction between KSHV virions and CD109 on the cell surface during entry, with recombinant CD109 protein exhibiting inhibitory activity of KSHV infection by blocking virion binding. These findings uncover a previously unrecognized role of cellular senescence in enhancing KSHV infection through upregulation of specific host factors and provide insights into the complex interplay between aging and viral pathogenesis.

Keywords: Aging; Cellular senescence; Endothelial cells; Molecular biology; Virology.

Figure 1. Increased KSHV infection in senescent human endothelial cells.

(A) Senescence was induced in primary human endothelial cells (HUVECs) by repeated subculture over 35 passages or treatment with doxorubicin. In doxycycline-inducible immortalized human endothelial cells (HuARLT cells), senescence was induced by culturing without doxycycline. SA-β-gal staining was used for validation of senescence through microscopy or flow cytometry. p, cell culture passages. dc, doxycycline. KSHV infectivity was measured by GFP expression in cells infected with recombinant KSHV BAC16. (B) Analysis of KSHV-infected cells in nonsenescent and senescent human endothelial cells by flow cytometry at 24 hpi. Data are representative of 3 independent experiments. Data are shown as mean ± SD. n = 3. ^***P < 0.001, unpaired 2-tailed Student’s t test. (C) Quantification of the KSHV genome in KSHV-infected nonsenescent and senescent human endothelial cells by quantitative PCR at 24 hpi. Data are representative of 3 independent experiments. Data are shown as mean ± SD. n = 3. ^**P < 0.01, ^***P < 0.001, unpaired 2-tailed Student’s t test. (D) Assessment of the relative expression of KSHV ORF71 mRNA in KSHV-infected nonsenescent and senescent human endothelial cells using quantitative reverse transcription PCR at 24 hpi. Data are representative of 3 independent experiments. Data are shown as mean ± SD. n = 3. ^**P < 0.01, unpaired 2-tailed Student’s t test.

Figure 2. Enhanced entry of KSHV in senescent human endothelial cells.

(A) Immunofluorescence assay of the entry of KSHV into nonsenescent and senescent cells. KSHV was visualized at 4 hpi using KSHV ORF65 antibody. Phalloidin (F-actin) and DAPI were used to visualize the shapes of the cells and nuclei, respectively. Scale bars: 50 μm. (B) Analysis of the number of internalized KSHV particles per cell in the images in A and Supplemental Figure 5. Data are representative of 10 independent experiments. Data are shown as mean ± SD. n = 10. ^***P < 0.001 using unpaired 2-tailed Student’s t test. (C) Quantification of the internalized KSHV genome in the KSHV-infected nonsenescent and senescent human endothelial cells by quantitative PCR. Genomic DNA was extracted from KSHV-infected cells at 4 hpi. The KSHV genome was quantified in the extracted DNA by quantitative PCR using primers for KSHV ORF26. Data are representative of 3 independent experiments. Data are shown as mean ± SD. n = 3. ^**P < 0.01; ^***P < 0.001, unpaired 2-tailed Student’s t test.

Figure 3. Enhanced binding of KSHV in senescent human endothelial cells.

(A) Confocal microscopy images of KSHV binding to the cell surface. After infection, cells were immediately fixed and stained without permeabilization. Scale bars: 10 μm. (B) Analysis of the number of KSHV particles binding to the cell surface per cell in the images in A and Supplemental Figure 6. Data are representative of 10 independent experiments. Data are shown as mean ± SD. n = 10. ^***P < 0.001, unpaired 2-tailed Student’s t test. (C) Quantification of the cell-surface–bound KSHV genome in KSHV-infected cells by quantitative PCR. After 1 hour of KSHV infection, the cells were immediately scraped and genomic DNA was extracted. The KSHV genome was quantified in the extracted DNA by quantitative PCR using primers for KSHV ORF26. Data are representative of 3 independent experiments. Data are shown as mean ± SD. n = 3. ^**P < 0.01; ^***P < 0.001, unpaired 2-tailed Student’s t test.

Figure 4. Identification of candidate proteins associated with increased KSHV infectivity in senescent endothelial cells.

(A) Heatmap of the differential expression of proteins from nonsenescent (+dc, culture with doxycycline) and senescent (–dc, culture without doxycycline) HuARLT cells with KSHV (K) or without KSHV (M). The cell lysate and cell pellet from each conditioned cell were analyzed by LC-MS/MS. (B) Western blot analysis of the selected candidate proteins. Cav-1, caveolin-1; ITG, integrin; dc, doxycycline. GAPDH was used as a housekeeping protein for normalization. (C) Flow cytometric analysis of candidate proteins in nonsenescent (+dc HuARLT and –dox HUVEC) and senescent (–dc HuARLT and +dox HUVEC) cells. Data are representative of 3 independent experiments. Data are shown as mean ± SD. n = 3. ^*P < 0.05; ^**P < 0.01; ^***P < 0.001, unpaired 2-tailed Student’s t test.

Figure 5. Establishment and characterization of KO clones for the candidate proteins in HuARLT cells.

(A) Schematic diagram of the gRNA sequence for caveolin-1 (CAV1), integrin α2 (ITGA2), F11R, and CD109. The gRNA-recognizing site is indicated as the CRISPR target sequence. (B) Western blot analysis of each KO protein. WT, WT HuARLT cell; KO, KO HuARLT cell. (C) Target sequence analysis in WT and KO HuARLT cells. The PCR products containing the gRNA targeting region from the genomic DNA of WT and KO cells were cloned into a T-vector. The sequences of 10 colonies were analyzed by Sanger sequencing. The mutated region is indicated by a box.

Figure 6. Analysis of KSHV infectivity in KO clones of CAV1, ITGA2, F11R, and CD109.

An equivalent quantity of KSHV was used to infect the same number of WT and KO HuARLT cells, with or without induction of senescence. KSHV was prepared as GFP infectious units of 1 to infect approximately 90% of nonsenescent WT cells, followed by the infection of a 2-fold serially diluted virus into each group of conditioned cells. KSHV infectivity was measured using GFP expression. (A and B) Fluorescence microscopic visualization of KSHV-infected cells for each KO clone in nonsenescent (+dc, A) and senescent (–dc, B) HuARLT cells. CAV1, caveolin-1; ITGA2, integrin-α₂. Scale bar: 100 μm. (C and D) Flow cytometric analysis of KSHV-infected cells for each KO clone in nonsenescent (+dc, C) and senescent (–dc, D) HuARLT cells. At the indicated GFP infectious units (GIU), the percentage of KSHV-infected cells was compared among each group. Data are representative of 3 independent experiments. Data are shown as mean ± SD. n = 3. ^***P < 0.001, Dunnett’s test for multiple comparisons.

Figure 7. KSHV binding to the cell surface of CAV1-KO and CD109-KO HuARLT cells.

(A) Representative confocal images of KSHV virus particles in CAV1-KO and CD109-KO HuARLT cells. Scale bars: 10 μm. (B) Analysis of the number of KSHV particles that bind to the cell surface per cell in the images in A and Supplemental Figure 13. Data are representative of 10 independent experiments. Data are shown as mean ± SD. n = 10. ^***P < 0.001, Dunnett’s test for multiple comparisons. (C) Quantification of the cell-surface–bound KSHV genome in the KSHV-infected cells by quantitative PCR. After 1 hour of KSHV infection, the cells were immediately scraped, and genomic DNA extracted after washing. The KSHV genome was quantified in the extracted DNA by quantitative PCR using primers for KSHV ORF26. Data are representative of 3 independent experiments. Data are shown as mean ± SD. n = 3. ^**P < 0.01; ^***P < 0.001, Dunnett’s test for multiple comparisons.

Figure 8. Analysis of KSHV interaction with caveolin-1 and CD109.

(A and B) Three-dimensional confocal microscopy images of the colocalization of KSHV with CD109 (A) or caveolin-1 (Cav-1) (B). Senescent HuARLT cells infected with KSHV for 1 hour were stained with KSHV K8.1 antibody and target proteins, and Z-axis scanning was performed at 4 μm intervals, generating more than 10 scans. The bar-shaped images at the edges represent cross-sections, guided by white solid lines in the stacked image. (C) Quantitative analysis of the colocalization of KSHV with CD109 or caveolin-1 by Manders’ colocalization coefficient in KSHV-infected senescent HuARLT cells. n = 10. ^***P < 0.001 (D) SPR analysis of KSHV binding to immobilized recombinant CD109. Data are representative of 3 independent experiments. (E) Neutralization of KSHV infectivity using a recombinant CD109 protein. Data are representative of 3 independent experiments. Data are presented as mean ± SD. n = 3. ^***P < 0.001, Dunnett’s test for multiple comparisons.

Figure 9. KSHV gH/gL interacts with CD109.

Co-IP analysis of CD109 with KSHV glycoproteins gB (A), gH/gL (B), and K8.1 (C). HEK-293T cells were cotransfected with the indicated plasmids for 24 hours, after which cell lysates were subjected to IP using anti-FLAG magnetic agarose beads. The resulting complexes were then analyzed by immunoblotting (IB) using the indicated antibodies. Data are representative of 3 independent experiments.