The neutralizing function of the anti-HTLV-1 antibody is essential in preventing in vivo transmission of HTLV-1 to human T cells in NOD-SCID/γcnull (NOG) mice

Abstract

Background: Human T-cell leukemia virus type 1 (HTLV-1) causes both neoplastic and inflammatory diseases, including adult T-cell leukemia and HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP). Because these life-threatening and disabling diseases are not yet curable, it is important to prevent new HTLV-1 infections.

Findings: In this study, we have established a simple humanized mouse model of HTLV-1 infection for evaluating prophylactic and therapeutic interventions. In this model, HTLV-1-negative normal human peripheral blood mononuclear cells (PBMCs) are transplanted directly into the spleens of severely immunodeficient NOD-SCID/γcnull (NOG) mice, together with mitomycin-treated HTLV-1-producing T cells. Using this model, we tested the efficacy of monoclonal antibodies (mAbs) specific to HTLV-1 as well as human IgG isolated from HAM/TSP patients (HAM-IgG) in preventing HTLV-1-infection. One hour before and 24 h after transplantation of the human cells, each antibody sample was inoculated intraperitoneally. On day 14, human PBMCs isolated from the mouse spleens were tested for HTLV-1 infection. Whereas fresh CD4-positive and CD8-positive T cells isolated from untreated mice or mice treated with isotype control mAb, HTLV-1 non-neutralizing mAbs to envelope gp46, gag p19, and normal human IgG were all infected with HTLV-1; the mice treated with either HTLV-1 neutralizing anti-gp46 mAb or HAM-IgG did not become infected.

Conclusions: Our data indicate that the neutralizing function of the antibody, but not the antigen specificity, is essential for preventing the in vivo transmission of HTLV-1. The present animal model will also be useful for the in vivo evaluation of the efficacy of candidate molecules to be used as prophylactic and therapeutic intervention against HTLV-1 infection.

Figure 1

In vivo infection of HTLV-1 in engrafted human PBMCs in hu-PBMC-NOG-spl mice. A. Live cells were gated on their forward and side light scatter characteristics, and then cell surface markers within the HLA-class I-positive population were analyzed. B. There tended to be higher frequencies of CD4-positive cells than CD8-positive cells. The numbers represent the percentage of the cell population within the HLA-class I-positive gate. C. Genomic PCR to confirm HTLV-1 infection. Genomic DNA was extracted from human CD4 and CD8-positive T cells recovered from the spleens of hu-PBMC-NOG-spl mice sacrificed 14 days post infection, and then a fragment of the HTLV-1 pX region was amplified. β-actin was used as a control. The lower limit of detection was one copy of HTLV-1 tax per 10⁴ PBMCs. D. RT-PCR to confirm HTLV-1 infection. RNA was extracted from human CD4 and CD8-positive T cells recovered from the spleens of hu-PBMC-NOG-spl mice sacrificed 14 days post infection. cDNA was synthesized and amplified from HTLV-1 tax and the HBZ region as described previously [15]. GAPDH was used as a control.

Figure 2

Characteristics of HTLV-1-infected human T cells recovered from hu-PBMC-NOG-spl mice. Tax protein expression in human lymphocytes recovered from the spleens of hu-PBMC-NOG-spl mice by flow cytometry. A. Human lymphocytes recovered from mouse spleens express very low levels of Tax protein (upper panel). After a short-term (16 h) cultivation ex vivo, Tax expression was rapidly induced (lower panel). The numbers represent the percentage of the Tax-positive cell population within the HLA-class I-positive gate. B. Tax-expressing cells are more frequently positive for CCR4 than Tax-negative cells.

Figure 3

HTLV- 1 infection in hu- PBMC- NOG- spl mice was completely inhibited by neutralizing, but not non- neutralizing, antibodies. In vivo transmission of HTLV-1 and protective efficacy of various monoclonal antibodies was evaluated using quantitative real-time PCR analysis of HTLV-1 proviral DNA. Genomic DNA was extracted from the human lymphocytes recovered from hu-PBMC-NOG-spl mice. A. All of the mice immunized with neutralizing mAbs against Env (clone LAT-27) were completely protected against HTLV-1 infection, whereas non-neutralizing mAbs against Env (clone LAT-25), anti-Gag (clone GIN-7), anti-HCV (clone MO-8), or anti-OX40 mAb (clone B-7B5) did not protect against infection. The mice immunized with human immunoglobulin isolated from HAM/TSP patients (HAM-IgG) were also protected against HTLV-1 infection, whereas human IgG isolated from normal uninfected controls (NC-IgG) did not protect against infection. Results are shown as mean ± SE. To test for significant differences among the different groups, one-way analysis of variance was performed, followed by Scheffe’s multiple comparisons test. The lower limit of detection was one copy of HTLV-1 tax per 10⁴ PBMCs. B. Flow cytometric studies indicated that the human lymphocytes recovered from mouse spleens immunized with anti-Env neutralizing mAbs or HAM-IgG express only a trace amount of Tax protein after short-term (16 h) cultivation ex vivo, which may be a background false-positive staining artifact. In contrast, a significant amount of Tax protein was expressed in human lymphocytes recovered from non-immunized mouse spleens (PBS-injected) or mouse spleens immunized with NC-IgG. The numbers represent the percentage of the cell population within the HLA-class I-positive/CD4-positive gate.