Design, development, and evaluation of gene therapeutics specific to KSHV-associated diseases

Abstract

Kaposi sarcoma associated herpesvirus (KSHV) is the causative agent of Kaposi sarcoma (KS) and two human B cell lymphoproliferative diseases. KSHV-encoded latency-associated nuclear antigen (LANA) is expressed in KSHV-infected cancer cells. Thus, LANA is an attractive target for therapeutic intervention for KSHV-associated diseases. Here, we devised a cancer gene therapy vector using the adeno-associated virus (AAV), which capitalizes on the LANA's function to maintain terminal repeat (TR)-containing circular genomes and the TR's enhancer function for KSHV-inducible gene promoters. By including two TR copies with a lytic inducible gene promoter (TR2-OriP), we prepared an AAV vector, which expresses an engineered thymidine kinase (TK) selectively in KSHV-infected cells. Ganciclovir (GCV), an anti-herpesvirus drug, effectively eradicated multiple KSHV-infected cells that include induced pluripotent stem cell-derived epithelial colony-forming cells but not non-KSHV-infected counterparts in the presence of AAV8-TR2-OriP-TK. In addition, AAV8-TR2-OriP-TK prevents KSHV from producing virions from reactivated cells. Anti-cancer drugs, known to reactivate KSHV, stimulated TK expression from the vector and, therefore, synergized with AAV8-TR2-OriP-TK/GCV. Finally, the AAV8-TR2-OriP-TK/GCV effectively suppressed the growth of KSHV-infected cancer cells in the xenograft tumor model, whereas systemic intravenous AAV8-TR2-OriP-TK injection/GCV showed no detectable side effects. Our proof-of-concept studies highlight a promising strategy for targeting cancers driven by herpesviruses.

Keywords: Kaposi's sarcoma; Kaposi’s sarcoma-associated herpesvirus; MT: Regular Issue; adeno-associated virus; cancer gene therapy; oncogenic herpesvirus; oncolytic therapy; promoter-enhancer interaction; terminal repeats.

Graphical abstract

Figure 1

Construction of KSHV-infection-specific gene expression cassette (A) Schematic diagram of cloning strategy to generate specific gene therapeutic vector. Plasmid names are depicted on the left, and restriction enzyme sites used for cloning are also shown. TR, terminal repeat; ITR, inverted terminal repeat. (B) Schematic representation of the regulatory mechanism of the gene cassette. The construct consists of terminal repeats (TR) and Ori RNA promoter (OriP) and the mCardinal fluorescent reporter gene. (C) Fluorescent and bright-field images of the 293/KSHV and the parental 293 cells. The pAAV-TR2-OriP-mCardinal vector was transfected into 293/KSHV and the parental 293 cells. Images were taken 48 h post-transfection. Scale bars, 200 μm. (D) Mean fluorescence intensity (MFI). The mCardinal signal in 293/KSHV and the parental 293 cells after transfection of the pAAV-TR2-OriP-mCardinal vector was measured with a flow cytometer at 2 and 5 days post-transfection. (E) The proportion of the mCardinal-positive cells. The number of mCardinal-expressing cells was determined by the Cy5 channel with flow cytometry at indicated days after transfection. The relative proportion of mCardinal-positive cells was depicted taking the number of mCardinal cells at day 2 as 100%. (F) Enhanced exogenous gene expression with TR fragments. The pAAV-TR2-OriP-mCardinal vector with or without two copies of TR sequences was transfected to the 293/KSHV cells. Fluorescent and bright-field images were taken 2 days after the transfection. Scale bars, 300 μm.

Figure 2

Activation of TR2-OriP promoter in KSHV-infected cells (A) Schematic diagram of iodixanol gradient ultracentrifugation for AAV isolation. The fraction with 40% iodixanol contains AAVs. (B) SDS-PAGE gels. The 40% iodixanol layer was fractionated from bottom to top 1 to 6. Coomassie staining shows the AAV capsid proteins VP1, VP2, and VP3. Fractions 1–4 for AAV8-TR2 OriP mCardinal were pooled and concentrated, DNA copies were measured, and the fractions were used for the following studies. The molecular size of the marker is indicated on the left side of the gel. (C) Transmission electron microscopy (TEM). The representative TEM images for the AAV8-TR2-OriP-mCardinal are shown. Putative empty capsids are marked with an arrow. (D) Fluorescent and bright-field images. The 293/KSHV and the parental 293 cells were infected with AAV8-TR2-OriP-mCardinal, and images were taken 72 h post-transduction. Scale bars, 100 μm. (E) Mean fluorescence intensity (MFI) of the mCardinal signal and AAV genome copies. MFI was measured with a flow cytometer, and AAV DNA copies in transduced cells were determined by qPCR. Relative AAV DNA levels were measured at the indicated days after infection. GAPDH coding sequence was used for internal control. Data were analyzed using a two-sided unpaired Student’s t test and shown as mean ± SD. (F) Relative abundance of AAV DNA copies. AAV DNA copies were measured by qPCR and compared between the 293/KSHV and the parental 293 cells. Twenty-four hours after AAV-mCardinal transduction in 293/KSHV cells was designated as 1. The GAPDH coding sequence was used as an internal control. Data were analyzed using a two-sided unpaired Student’s t test and shown as mean ± SD. ∗∗ and ∗ indicates a stastically significant difference between groups (p < 0.01 and p < 0.05).

Figure 3

Inhibition of cell growth by AAV8-TR2-OriP-TK in KSHV-infection-specific manner (A) Growth of 293 cells; 293 cells or KSHV-infected 293 cells were seeded in 12-well plates and transduced with AAV8-TR2-OriP-TK. Cells were treated with mock, GCV (5 μM), or a combination of GCV and OTX015 (200 nM). Live 293 cells were counted every day for 3 days. The total number of cells in each well was counted in triplicate, and a bar graph was generated with mean ± SD. (B) Bar chart with or without AAV8-TR2-OriP-TK transduction in 293 cells. Live 293 and 293/KSHV cells at day 3 with or without AAV8-TR2-OriP-TK are shown in bar charts. Data were analyzed using a two-sided unpaired Student’s t test and shown as mean ± SD. (C) iSLK cell growth. iSLK cells or KSHV-infected iSLK cells were seeded in 12 well plates and transduced with AAV8-TR2-OriP-TK. Cells were treated with mock, GCV (5 μM), or GCV and OTX015 (200 nM). The total number of cells in each well was counted in triplicate, and a bar graph was generated with mean ± SD. (D) Bar chart with or without AAV8-TR2-OriP-TK transduction in iSLK cells. Live iSLK and iSLK/KSHV cells at day 3 with or without AAV8-TR2-OriP-TK are shown in bar charts. Data were analyzed using a two-sided unpaired Student’s t test and shown as mean ± SD. (E) Fluorescent and bright-field images. iSLK cells with or without KSHV infection were transduced with AAV8-TR2-OriP-TK. Cells were treated with the indicated drug combination. Images were taken 3 days after treatment of OTX015 (200 nM) or SAHA (1 μm); Scale bars, 200 μm.

Figure 4

AAV8-TR2-OriP-TK selectively inhibits KSHV-infected ECFC growth (A) Schematic diagram of ECFC differentiation from iPSCs. iPSC differentiation was induced by bFGF, BMP4, and VEGF₁₆₅ in the Stemline II media, followed by EGF-II culture. (B) Relative gene expression of iPSCs and ECFCs (day 26). The CD34, Prox-1, Flt-4, and LYVE-1 genes were utilized as differentiation markers for ECFCs, whereas the Oct3/4, Nanog, and Sox2 genes were employed as reprogramming markers; 18S was used for internal control. Data were analyzed using a two-sided unpaired Student’s t test and shown as mean ± SD. (C) Fluorescent and bright-field cell images. r.219 KSHV (MOI = 1) was infected to differentiating cells at 12 days post-induction of iPSC differentiation or parental iPSCs. Images were taken 14 days after infection. Scale bars, 200 μm. (D) Fluorescent and bright-field cell images. ECFCs or KSHV-infected ECFCs transduced with AAV8-TR2-OriP-TK were treated with GCV (5 μM), and images were taken 3 days after GCV treatment. Scale bars, 100 μm. (E) Cell growth. ECFCs, or KSHV-infected ECFCs were seeded in 6-well plates and transduced with AAV8-TR2-OriP-TK. The following day, cells were treated with DMSO or GCV (5 μM). Live ECFCs were counted daily for three consecutive days. The total number of cells in each well was counted in triplicate, and a bar graph was generated with mean ± SD.

Figure 5

AAV8-TR2-OriP-TK prevents KSHV reactivation and reinfections (A) Fluorescence and bright-field cell images. iSLK/KSHV cells were treated as indicated, and the cells were imaged with fluorescence microscopy at 48 h post-stimulation. AAV8-TR2-OriP-TK transduction with GCV treatment resulted in a decrease in the RFP signal, indicating reduced PAN-RNA promoter activation. Scale bars, 100 μm. (B) Viral gene expression. PAN-RNA or K8.1 transcripts were measured by RT-qPCR with specific primer pairs. Transcripts were normalized with 18S rRNA. The relative transcripts in AAV8-TR2-OriP-TK transduced without GCV were normalized as 1. Data were analyzed using a two-sided unpaired Student’s t test and shown as mean ± SD. (C) Capsidated viral DNA copies. The KSHV virion copy number with or without reactivation in the presence or absence of GCV (5 μM) was measured with qPCR. KSHV virions were collected from the culture supernatant 4 days post-reactivation. Data were analyzed using a two-sided unpaired Student’s t test and presented as mean ± SD. (D) KSHV infection in freshly prepared iSLK cells. The KSHV infection with reactivated virus was evaluated by infecting freshly prepared iSLK cells. 1 mL culture supernatant 4 days post-reactivation in the presence or absence of GCV (5 μM) was mixed with freshly prepared iSLK cells. Images were taken 2 days after infection. (E) Flow cytometry. Frequencies of KSHV infection were measured with flow cytometry. Data were analyzed using a two-sided unpaired Student’s t test and shown as mean ± SD.

Figure 6

Bystander effects and transcription profiles (A) Schematic diagram of study design for the bystander effect. Two types of KSHV-infected iSLK cells were prepared (green and red), and AAV8-TR2-OriP-TK was transduced only to EGFP-positive cells. Green and red (mCherry-positive) cells were mixed at a 1:1 ratio (10⁵ cells each) and cultured for 4 days in the presence of GCV (5 μM). Bystander effects were assessed with the viability of red cells. (B) Fluorescent cell images. Arrows indicate dead red cells determined by cell morphology and fractured nucleus. Images were taken 4 days after the mixture of iSLK/KSHV (green) and iSLK/mCherry-KSHV (red) cells. Scale bars, 20 μm. (C) Principal-component analyses. Total RNA-sequencing was performed with iSLK/KSHV and 293/KSHV cells. Cells were treated as indicated in the panel. (D) Pathway analyses. Cellular signaling pathways that are significantly altered in GCV with AAV8-TR2-OriP-TK transduction are shown. The results indicated a strong induction of DNA damage responses with subsequent induction of cell apoptosis. (E) Immunofluorescence assays. AAV8-TR2-OriP-TK-transduced iSLK/KSHV cells were stained with antibodies specific to cleaved caspase 3 and rabbit-647 secondary antibody. Cleaved caspase 3 expression in iSLK/KSHV cells under GCV (5 μM) treatment was confirmed with immune staining. Scale bars, 100 μm.

Figure 7

AAV8-TR2-OriP-TK inhibits tumor growth in a xenograft model (A) Schematic diagram of treatment schedules. KSHV-infected iSLK cells were transduced with AAV8-TR2-OriP-TK and subcutaneous xenograft. AAV8-TR2-OriP-TK transduced iSLK cells were implanted into the right hind leg of the mice, whereas non-transduced iSLK cells were injected into the left hind leg. GCV was administered as indicated. (B) Measurement of tumor sizes. Tumor volumes (mm³) were measured every 4 days. (C) Mouse images at day 38. (D) Images of tumor mass extracted from subcutaneous location. (E) Tumor volumes. The weight of tumor mass (mg) was plotted, and treatments were indicated at the bottom of the panel. (F) H&E staining and immunohistochemistry (IHC) staining of Ki-67 in mouse tissue sections. Representative images of IHC staining and EGFP signals found in individual tumors. Scale bars, 20 μm (top and middle) and 200 μm (bottom).

Figure 8

AAV8-TR2-OriP-TK inhibits tumor growth in a KSHV-infection-specific manner in a xenograft model (A) Schematic diagram of treatment schedules. KSHV-infected Capi-1 (SLK cells) and parental SLK cells were transduced with AAV8-TR2-OriP-TK and xenograft subcutaneously to the right and left hind legs, respectively. GCV was administered at 50 mg/kg twice a day for 5 days. (B) Measurement of tumor sizes. Tumor volumes (mm³) were measured every 3 to 4 days. (C) Mouse images at day 40. (D) Images of tumor mass extracted from subcutaneous. (E) Tumor volumes. The weight of tumor mass (mg) was plotted, and treatments were indicated at the bottom of the panel. ∗∗ indicates a statistically significant difference between groups (p < 0.05).