Safety of intraocular anti-VEGF antibody treatment under in vitro HTLV-1 infection

Abstract

Introduction: HTLV-1 (human T-cell lymphotropic virus type 1) is a retrovirus that infects approximately 20 million people worldwide. Many diseases are caused by this virus, including HTLV-1-associated myelopathy, adult T-cell leukemia, and HTLV-1 uveitis. Intraocular anti-vascular endothelial growth factor (VEGF) antibody injection has been widely used in ophthalmology, and it is reportedly effective against age-related macular degeneration, complications of diabetic retinopathy, and retinal vein occlusions. HTLV-1 mimics VEGF₁₆₅, the predominant isoform of VEGF, to recruit neuropilin-1 and heparan sulfate proteoglycans. VEGF₁₆₅ is also a selective competitor of HTLV-1 entry. Here, we investigated the effects of an anti-VEGF antibody on ocular status under conditions of HTLV-1 infection in vitro.

Methods: We used MT2 and TL-Om1 cells as HTLV-1-infected cells and Jurkat cells as controls. Primary human retinal pigment epithelial cells (HRPEpiCs) and ARPE19 HRPEpiCs were used as ocular cells; MT2/TL-Om1/Jurkat cells and HRPEpiCs/ARPE19 cells were co-cultured to simulate the intraocular environment of HTLV-1-infected patients. Aflibercept was administered as an anti-VEGF antibody. To avoid possible T-cell adhesion, we lethally irradiated MT2/TL-Om1/Jurkat cells prior to the experiments.

Results: Anti-VEGF antibody treatment had no effect on activated NF-κB production, inflammatory cytokines, chemokines, HTLV-1 proviral load (PVL), or cell counts in the retinal pigment epithelium (RPE) under MT2 co-culture conditions. Under TL-Om1 co-culture conditions, anti-VEGF antibody treatment did not affect the production of activated NF-κB, chemokines, PVL, or cell counts, but production of the inflammatory cytokine IL-6 was increased. In addition, anti-VEGF treatment did not affect PVL in HTLV-1-infected T cells.

Conclusion: This preliminary in vitro assessment indicates that intraocular anti-VEGF antibody treatment for HTLV-1 infection does not exacerbate HTLV-1-related inflammation and thus may be safe for use.

Keywords: HTLV-1 uveitis; VEGF; aflibercept; human T-cell leukemia virus type 1; ocular inflammation; uveitis.

Figure 1

(A) Multi-receptor model of the initial phase of HTLV-1 entry into target cells. HTLV-1 SU interacts with HSPG, resulting in the initial attachment and concentration of HTLV-1 particles on the cell surface. The interaction of HSPG with SU and NRP-1, and the direct binding of SU to NRP-1 then result in the recruitment of NRP-1, enabling stable binding of SU to the HSPG/NRP-1 complex. (B) Schematic representation of interactions between VEGF₁₆₅, HSPG, and NRP-1. The sequence encoded by exon 7 of the VEGF₁₆₅ gene binds to HSPG, and the sequence encoded by exon 8 binds directly to NRP-1. The NRP-1 dimer is then formed, resulting in enhanced stability. HTLV-1: human T-cell lymphotropic virus type 1; SU: surface subunit; HSPG: heparan sulfate proteoglycan; NRP-1: neuropilin-1.

Figure 2

ELISA for phospho p65 NF-κB. Effect of anti-VEGF treatment on activation of nuclear factor-κB (NF-κB) in ARPE19 cells shown as the ratio of phosphor to total. No significant change in NF-κB activation of ARPE19 cells in corresponding groups was seen following aflibercept addition. Data are taken from three independent biological experiments and presented as the mean ± SEM (n.s., not significant).

Figure 3

Levels of inflammatory cytokines were measured in the culture supernatants of HRPEpiCs and HRPEpiCs co-cultured for 48 h with MT2, TL-Om1, or Jurkat cells with/without 0.5 mg/mL aflibercept. In co-culture of HRPEpiCs with MT2 cells, aflibercept did not significantly affect the levels of IL-6, IL-8, and IFN-γ. In co-culture of HRPEpiCs with TL-Om1 cells, production of IL-6 increased significantly following aflibercept addition, but levels of IL-8 and IFN-γ did not change significantly. Data are taken from three independent biological experiments and presented as the mean ± SEM (units: pg/μL) (*P < 0.05; n.s., not significant).

Figure 4

Levels of chemokines measured in the culture supernatants of HRPEpiCs and HRPEpiCs co-cultured for 48 h with MT2, TL-Om1, or Jurkat cells with/without 0.5 mg/mL aflibercept. Secretion of CXCL10, CCL2, CXCL9, and CCL5 was monitored, but no significant differences were detected in all corresponding groups, with or without aflibercept. Data are taken from three independent biological experiments and presented as the mean ± SEM (units: pg/μL) (n.s., not significant).

Figure 5

Proviral load (PVL) in MT2 cells (A) or TL-Om1 cells (B) treated with/without 0.5 mg/mL aflibercept for 48 (h) The number of each type of cells was 5 × 10⁵. Aflibercept had no effect on the PVL of either cell type. Data are taken from three independent biological experiments and presented as the mean ± SEM (n.s., not significant).

Figure 6

HTLV-1 proviral DNA (PVL) was monitored in HRPEpiCs transferred three times after culture alone or co-culture with MT2, TL-Om1, or Jurkat cells. Aflibercept had no effect on HTLV-1 PVL of HRPEpiCs in the corresponding groups. The number of each type of cells was 1 × 10⁵. Data are taken from three independent biological experiments. Error bars represent standard deviation (n.s., not significant).

Figure 7

Enumeration of HRPEpiCs (A) and ARPE19 cells (B) co-cultured with irradiated T-cell lines or cultured alone; data are number of cells per well with/without 0.5 mg/mL aflibercept after three transfers. The number of HRPEpiCs and ARPE19 cells at the beginning of each culture was 1.5 × 10⁵. No significant changes in the number of HRPEpiCs or ARPE19 cells were observed in the corresponding groups with aflibercept addition. Data are taken from three independent biological experiments and presented as the mean ± SEM (n.s., not significant).