Targeting human langerin promotes HIV-1 specific humoral immune responses

Abstract

The main avenue for the development of an HIV-1 vaccine remains the induction of protective antibodies. A rationale approach is to target antigen to specific receptors on dendritic cells (DC) via fused monoclonal antibodies (mAb). In mouse and non-human primate models, targeting of skin Langerhans cells (LC) with anti-Langerin mAbs fused with HIV-1 Gag antigen drives antigen-specific humoral responses. The development of these immunization strategies in humans requires a better understanding of early immune events driven by human LC. We therefore produced anti-Langerin mAbs fused with the HIV-1 gp140z Envelope (αLC.Env). First, we show that primary skin human LC and in vitro differentiated LC induce differentiation and expansion of naïve CD4+ T cells into T follicular helper (Tfh) cells. Second, when human LC are pre-treated with αLC.Env, differentiated Tfh cells significantly promote the production of specific IgG by B cells. Strikingly, HIV-Env-specific Ig are secreted by HIV-specific memory B cells. Consistently, we found that receptors and cytokines involved in Tfh differentiation and B cell functions are upregulated by LC during their maturation and after targeting Langerin. Finally, we show that subcutaneous immunization of mice by αLC.Env induces germinal center (GC) reaction in draining lymph nodes with higher numbers of Tfh cells, Env-specific B cells, as well as specific IgG serum levels compared to mice immunized with the non-targeting Env antigen. Altogether, we provide evidence that human LC properly targeted may be licensed to efficiently induce Tfh cell and B cell responses in GC.

Fig 1. Targeting human skin LC promotes IgG production by naïve B cells.

(A) Schematics of the procedure of the epidermal skin LC treated with αLC.Env (10nM) or control mAbs, cultured with naïve CD4+ T cells for 4 days and then with autologous naïve B cells for 6 days. (B) Human LC isolated from skin explants were cocultured with allogenic CD4⁺ naïve T cells as in A. Four days post culture, Tfh-differentiated cells were assessed by FACS, using CD45RO, CXCR5, PD-1, Bcl-6 and ICOS markers. DC were generated in vitro from CD14+ purified monocytes and cultured with skin LC as comparison. MFI: mean of fluorescence intensity. (C) IgG concentrations in LC/T/B co-culture supernatants were measured by ELISA at day 10 after LC treatment. The 10–1074 mAb with a known concentration served as reference. Statistics were obtained using the Holm-Sidak’s multiple comparisons test (*, P < 0.05; **, P < 0.01; ***, P < 0.001; ns; non-significant).

Fig 2. Targeting of CD34-LC induces the activation of Tfh cells and Ig class-switch by naïve B cells, in a complete autologous system, without any cytokine.

(A) As in Fig 1A, schematics of the procedure of the CD34-LC treated with αLC.Env (10nM) or control mAbs, cultured with autologous Tfh cells and autologous naïve B cells for 6 days. (B) B cell phenotype was performed before and after coculture and was based on the expression of CD21 and CD27 among CD19⁺ population. Representative dot plots of one out of four donors are depicted on the left (AM, activated memory, RM, resting memory; TLM; tissue-like memory; N, naïve). Two-tailed paired T-test, ** P < 0.01 (C) As in Fig 1C, IgG concentrations in LC/T/B co-culture supernatants measured by ELISA at day 10 after LC treatment. Statistics were obtained using the Holm-Sidak’s multiple comparisons test (*, P < 0.05; **, P < 0.01; ***, P < 0.001; ns; non-significant).

Fig 3. Targeting CD34-LC induces the secretion of Env-specific Ig by memory B cells from HIV⁺ donors.

(A) Schematics of the procedure, as in Fig 2 but adding total B cells isolated from PBMCs of HIV⁺ donors at day 4. (B) As in Fig 2C, total IgG were measured in culture supernatants of LC treated or not with αLC.Env, and cultured (left) or not (right) with naïve CD4⁺ T cells. (C) ELISA-based concentration analysis of Env gp140z-specific Ig in culture supernatants at day 17. Statistics were obtained using the Holm-Sidak’s multiple comparisons test (*, P < 0.05; **, P < 0.01; ***, P < 0.001; ns; non-significant).

Fig 4. Targeting skin LC promotes a specific cytokine environment.

(A) Multiplex beads analysis of 20 cytokines and comparison of culture supernatants at day 4 or 10 (as in Fig 1A) with skin LC treated by αLC.Env or IgG4.Env, cultured alone 4 days (left), with allogenic naïve CD4+ T cells (middle) or with total B cells from same donors (right). Color scale indicates median values of each condition (n = 4 to 9) and is expressed in pg/mL. Values of IL-21 and APRIL are detailed in (B), and IL-18 and CXCL13 in (C), respectively. Dotted line indicates threshold of detection.

Fig 5. The in vivo targeting of HIV-1 Env to the skin LC promotes Tfh responses, GC B cell responses and antigen-specific humoral responses.

(A) Serum antibody responses to the Env antigen. Xcr1^DTA mice were immunized sc with 1 μg of αLC.Env (plain circles) or 1 μg of mIgG2b.Env (open circles). Non-immunized Xcr1^DTA mice were used as controls (squares). Antibody titers were measured by Luminex assay, using trimeric gp140z Env-coated beads. Means values of 6 mice per group (± SEM) are indicated. Dotted line: background of fluorescence evaluated in naïve mice. MFI, mean of fluorescence intensity. (B) Absolute number of Bcl6⁺ Tfh cells in dLN (left) and spleen (right), one week post boost (day 28 post prime), in Xcr1^DTA versus Xcr1^Cre-mTFP1 mice. B6 mice were used as controls (namely PBS). (C) Same as in (B), measuring absolute number of GC B cells in dLN (left) and spleen (right). (D-F) Same as in (B) measuring absolute number of memory B cells, plasma cells and antibody-secreting plasma cells in spleen. (G) From IgM^- GC B cells in the spleen, absolute numbers of those specific to the Env gp140z antigen are shown. Dotted line indicates threshold of positivity (mean of B6 values + 3 SD). Non-parametric Kruskal-Wallis tests were performed (*; P < 0.05; **, P < 0.01; ***, P < 0.001; ns, non-significant).