KSHV/HHV-8 infection of human hematopoietic progenitor (CD34+) cells: persistence of infection during hematopoiesis in vitro and in vivo

Abstract

The cellular reservoir for Kaposi sarcoma-associated herpesvirus (KSHV) infection in the hematopoietic compartment and mechanisms governing latent infection and reactivation remain undefined. To determine susceptibility of human CD34+ hematopoietic progenitor cells (HPCs) to infection with KSHV, purified HPCs were exposed to KSHV, and cells were differentiated in vitro and in vivo. Clonogenic colony-forming activity was significantly suppressed in KSHV-infected CD34+ cells, and viral DNA was predominantly localized to granulocyte-macrophage colonies differentiated in vitro. rKSHV.219 is a recombinant KSHV construct that expresses green fluorescent protein from a cellular promoter active during latency and red fluorescent protein from a viral lytic promoter. Infection of CD34+ HPCs with rKSHV.219 showed similar patterns of infection, persistence, and hematopoietic suppression in vitro in comparison with KSHV. rKSHV.219 infection was detected in human CD14+ and CD19+ cells recovered from NOD/SCID mouse bone marrow and spleen following reconstitution with rKSHV.219-infected CD34+ HPCs. These results suggest that rKSHV.219 establishes persistent infection in NOD/SCID mice and that virus may be disseminated following differentiation of infected HPCs into the B-cell and monocyte lineages. CD34+ HPCs may be a reservoir for KSHV infection and may provide a continuous source of virally infected cells in vivo.

Figure 1.

Quantitation of KSHV and rKSHV.219 infection of CD34⁺ cells cultured in vitro. (A) CD34⁺ HPCs were infected with wild-type KSHV or UV-irradiated KSHV (UV, 10 μW/cm for 1 hour) (100 viral genome equivalents per cell) or were mock-infected. Infected CD34⁺ HPCs (5 × 10 cells) were cultivated in IMDM supplemented with 10% FBS at 37°C, 5% CO₂ for up to 25 days. Cell aliquots were collected daily, and DNA was extracted and analyzed by RQ-PCR using KSHV KS330₂₃₃ (ORF 26 minor capsid protein, 233 bp) and human β-globin (240 bp) primers. KSHV genomic copy number was normalized to 0.5 × β-globin copy number, representing one human cell. Bars represent mean KSHV genomic copies per human cell at indicated periods after infection, and error bars represent standard error of the mean (SEM). (B) CD34⁺ HPCs were infected with rKSHV.219, UV-irradiated rKSHV.219 (UV, 10 μW/cm for 1 hour), heat-inactivated rKSHV.219 (HEAT, 65°C for 1 hour) (600 viral genome equivalents per cell), or mock-infected. Infected CD34⁺ HPCs were cultivated, cell aliquots collected, and DNA was extracted and analyzed by RQ-PCR as in panel A. Infected CD34⁺ HPCs were cultivated for up to 16 days. Bars represent mean rKSHV.219 genomic copies per human cell at indicated time periods after infection, and error bars represent SEM.

Figure 2.

Latent and lytic viral gene expression in KSHV- and rKSHV.219-infected CD34⁺ HPCs. (A) CD34⁺ HPCs infected with KSHV or rKSHV.219 and cells were permeabilized and incubated with primary rat monoclonal antibody (mAb) directed against KSHV ORF 73 (LANA-1) followed by incubation with either an FITC-labeled goat anti-rat IgG secondary antibody (KSHV-infected cells) or AMCA-labeled rabbit anti-rat IgG secondary antibody (rKSHV.219-infected cells). Samples were analyzed by flow cytometry for the presence of LANA-1, and cells were gated according to patterns demonstrated by staining with secondary antibody alone. (B-D) CD34⁺ HPCs infected with KSHV and rKSHV.219 were incubated with primary rabbit polyclonal antibodies directed against (B) ORF K2 (vIL-6), primary mouse mAbs directed against (C) ORF 59 (PF-8) and (D) ORF K8.1A of KSHV, followed by incubation with Cy5-conjugated goat anti-rabbit or goat anti-mouse IgG secondary antibody. Samples were analyzed by flow cytometry for the presence of vIL-6, ORF 59, and ORF K8.1A, and gated as described in panel A. (E) CD34⁺ HPCs infected with KSHV were fixed onto siliconized glass slides, permeabilized, and incubated with primary rat mAb directed against KSHV LANA-1, primary rabbit polyclonal antibodies directed against KSHV vIL-6, and primary mouse mAbs directed against KSHV ORF59 and ORF K8.1A. Cells were washed with PBS, followed by incubation with TRITC-conjugated goat anti-rat immunoglobulin (IgG), FITC-conjugated goat anti-rabbit IgG, or FITC-conjugated goat anti-mouse IgG. Cells were washed and counterstained with TO-PRO-3 iodide nuclear stain. KSHV LANA-1 staining is represented by red. KSHV vIL-6, ORF 59, and ORF K8.1A staining are represented by green. TO-PRO-3 nuclear staining is represented by blue. KSHV-infected CD34⁺ HPCs were analyzed at 3 and 15 days after infection (dpi) and are from the cultures presented in panels A-D. TPA-treated BCBL-1 cells and mock-infected CD34⁺ HPCs were similarly analyzed. Cells were visualized through a 100 ×1.40 NA oil-immersion Plan-Apochromat objective lens (Nikon, Tokyo, Japan); total magnification was × 1000.

Figure 3.

KSHV infection of CD34⁺ HPCs in the absence of polybrene. CD34⁺ HPCs were infected with KSHV in the presence or absence of polybrene. CD34⁺ HPCs infected with KSHV were permeabilized and incubated with primary rat monoclonal antibody (mAb) directed against KSHV LANA-1 followed by incubation with AMCA-labeled rabbit anti-rat IgG secondary antibody. Samples were analyzed by flow cytometry for the presence of LANA-1, and cells were gated according to patterns demonstrated by staining with secondary antibody alone.

Figure 4.

KSHV and rKSHV.219 infection suppresses clonogenic colony-forming potential of CD34⁺ HPCs in vitro. (A) CD34⁺ HPCs were infected with KSHV or with UV-irradiated KSHV (10 μW/cm for 1 hour). Cells were washed 3 times with PBS and plated in 2 mL MethoCult H4433 medium (3000 cells per 35mm × 10mm plate) in triplicate at 4 hours after infection. Colonies were visually scored and enumerated at 12 to 15 days after plating. Experiments were repeated 3 times using purified CD34⁺ HPCs from different donor tissues. The mean CFU-GM, BFU-E, HPP, and total colony numbers are presented. Statistically significant differences in the mean total colony numbers between KSHV and UV-KSHV (P = .021) groups were determined by single-tail ANOVA analysis. (B) CD34⁺ HPCs were infected with rKSHV.219, UV-irradiated rKSHV.219 (10 μW/cm for 1 hour), or heat-inactivated rKSHV.219 (65°C for 1 hour). Infected HPCs were washed and plated in MethoCult H4433 as described in panel A. Colonies were visually scored at 12 to 15 days after plating and used to calculate mean CFU-GM, BFU-E, HPP, and total colony numbers. Statistically significant differences in the mean total colony numbers between the rKSHV.219, UV-rKSHV.219, and HEAT-rKSHV.219 (P = 2.31 × 10^-6) groups were determined by single-tail ANOVA analysis. Experiments were repeated twice. Error bars represent standard error of the mean (SEM).

Figure 5.

Viral gene expression in spleen and bone marrow cells recovered from NOD/SCID-hu mice reconstituted with rKSHV.219-infected CD34⁺ HPCs. Bone marrow cells recovered from rKSHV.219 mouse D6 were cultured ex vivo for 33 days in complete IMDM. Bone marrow cells were analyzed by IFA using a rat anti-LANA-1 primary antibody and an AMCA antirat secondary antibody. Bone marrow cells recovered from a NOD/SCID mouse inoculated with heat-inactivated rKSHV.219-infected HPCs also were analyzed. BCBL-1, a KSHV-positive primary effusion lymphoma cell line, was used as a positive control for LANA-1 IFA. U937, a myeloid leukemia cell line negative for KSHV infection, was used as a negative control. Cells were visualized under a 20 ×.0.45 Plan-Fluor extra-long working distance (ELWD) dark medium (DM) objective lens (Nikon, Tokyo, Japan); total magnification, ×/200.