Induction of B-cell immune tolerance by antigen-modified cytotoxic T lymphocytes

Abstract

Background: Third-party-specific cytotoxic T lymphocytes (CTL), or veto CTL, are being assessed as a cellular therapeutic for the induction of T-cell tolerance during transplantation. Conceptually, veto cell-expressed antigens (Ags) may induce B-cell immune responses, and this may have deleterious consequences. Whether veto cells induce immunity, tolerance, or are ignored by B lymphocytes has, however, not been addressed.

Methods: CTL were retrovirally transduced with a model cell surface Ag to generate veto CTL. The impact of CTL-specific Ag expression on the activation and tolerization of Ag-specific B cells was assessed in vitro and, using adoptive transfer models, in vivo.

Results: In vitro, CTL-expressed Ag induced an abortive proliferative response in specific B lymphocytes, whereby an initial proliferative burst was followed by cell death. In vivo, the administration of veto CTL also induced B-cell tolerance. Specific immunoglobulin was not detected after subsequent immunization with a veto cell-expressed Ag. Modeling of this effect with Ag-specific B-cell receptor transgenic B lymphocytes demonstrated that Ag-specific B cells were eliminated by the veto CTL; the cell division was accompanied by the exhaustion and depletion of responding cells. Veto-induced B-cell tolerance could be wholly abrogated by treatment with the toll-like receptor ligand lipopolysaccharide, implying that this tolerance resulted from the absence of adequate supplemental signals during antigenic stimulation.

Conclusions: Veto CTL are effective promoters of B-cell tolerance. Further assessment of their therapeutic potential in this regard is warranted.

Figure 1. HEL expression on CTL

(A) Diagram of the retroviral vector used to express HEL on CTL. Nucleotide sequence shows junctional segments between the chimeric gene's components. Amino acid sequence is shown above this and restriction sites used for cloning and segment identification below. Additional sequence can be found under the following GenBank accessions: G gallus lysozyme, NM_205281; CD8, XM_132621. (B) Retrovirus containing the MSCV-I-GFP vector or the HEL construct shown was used to transduce activated, purified CD8 T cells. Cells were surface stained with a HEL-specific antibody and flow cytometrically analyzed 5 days after transduction.

Figure 2. Cross stimulation by HEL-specific MD-4 B cells and HEL-modified CTL

(A) 5×10⁴ resting HEL- or vector-modified CTL transduced and flow cytometrically purified for the expression of CD8 and GFP were re-stimulated with the indicated number of irradiated MD-4 BCR Tg or control non-Tg C57BL/6 splenocytes 9 days after initial transduction. After 48 h, cultures were pulsed with ³H-thymidine and ³H incorporation determined 16 h later by scintillation counting. No proliferative response was observed. Viability of the modified CTL was confirmed by stimulation with control splenocytes in the presence of the mitogen conA. (B) 5×10⁴ purified vector- of HEL-modified CTL, 9 days after transduction, were cultured alone, or co-cultured with 2.5×10⁵ irradiated MD-4 Tg or non-Tg C57BL/6 splenocytes. 48 h later cell free supernatant was isolated and IFN-γ production measured by ELISA. Cells were additionally stimulated with non-Tg splenocytes in the presence of conA as a positive control. ND, none detected. (C) To determine if MD-4 B cell engagement of HEL on HEL-modified CTL induced B cell lysis, MD-4 splenocytes (∼50% B lymphocytes) were co-cultured in triplicate with HEL-CTL or vector-modified CTL for 6 h at a 1:1 ratio. B lymphocytes were then enumerated by staining with B220 and MD-4 allotype specific IgM^a reactive antibodies and quantitative flow cytometry. A significant increase in survival (2-sided t-test, p<0.05) is apparent among B cells co-cultured with HEL-CTL in the experiment shown, but was inconsistently seen in additional analyses. (D, E) To determine if MD-4 B lymphocytes are stimulated to proliferate and expand in response to the HEL-CTL, CFSE-labeled MD-4 splenocytes were co-cultured with HEL- or vector-transduced CTL, or cultured untreated or in the presence of LPS for up to 3 days. Cells were analyzed at daily intervals by quantitative flow cytometry. Total number of viable cells detected are plotted (D). (E) Plots showing CFSE dilution with proliferation. Representative data is shown for each subfigure from 2 or more experimental repeats.

Figure 3. HEL-CTL induce B cell tolerance in vivo

HEL-CTL or control vector-transduced CTL (10⁷) were adoptively transferred i.v. into C57BL/6 mice 9, 6, and 3 days prior to immunization with HEL/CFA. Serum was isolated prior to initial immunization with HEL (day 0) and on day 28 and antibody titers tested by staining T cells transduced with the HEL construct or with control vector with serum from the treated animals. (A) Serum collected from mice on day 0 prior to immunization. Plot shows mean fluorescence intensity (MFI) of staining against HEL-transduced cells at various dilutions or control MSCV vector-transduced cells at the highest titer only. (B) Serum collected from mice on day 28 after immunization. (C) Serum collected on day 0 from mice that also received 50 μg LPS i.p. with each infusion of the HEL- or vector-modified CTL. (D) Serum collected on day 28 from mice treated as in (C). Mean ± S.E.M. is shown for each value. 6-8 mice were studied per treatment group in this representative experiment.

Figure 4. Co-transfer of MD-4 B cells and modified CTL into Rag1^-/- mice

HEL- or control vector-transduced CTL (10⁷) were adoptively transferred i.v. (retro-orbital) into B and T cell deficient Rag1^-/- mice. MD-4 splenocytes (2×10⁷) were transferred through the alternate retro-orbital plexus. On day 2, the mice were sacrificed and numbers of residual MD-4 B cells determined by staining with anti-B220 and MD-4 allotype-specific anti-IgM^a antibodies. Percent B220⁺IgM^a+ cells among splenocytes or mixed LN cells from control or HEL-CTL treated mice is shown. Representative data is shown from one of 3 mice per experimental group with essentially identical results.

Figure 5. Co-transfer of MD-4 B cells and modified CTL into wild type mice

CFSE-labeled MD-4 splenocytes were adoptively transferred into wild-type C57BL/6 mice i.v. retro-orbitally and HEL- or control vector-modified CTL then transferred through the alternate retro-orbital plexus. 4 days later, the indicated organs were isolated and lymphocytes purified. Cells were stained for IgM^a and analyzed by flow cytometry. (A) Representative dot plots of LN cells, gated for lymphocytes based on scatter properties, showing IgM^a versus CFSE fluorescence. (B) Histogram plots of CFSE-staining among IgMa⁺ lymphocytes, gated based on the box shown in (A), is shown. Plots are normalized so that an equal number of scatter-gated lymphocytes were analyzed for the paired HEL-CTL and vector control plots to allow visual comparability. Percent of total lymphocytes that are IgM^a+ MD-4 B cells is indicated. Representative plots are shown.

Figure 6. LPS promotes the survival of proliferating, HEL-CTL-treated MD-4 B cells

MD-4 mice were bred onto the CD45.1 background, and CFSE-labeled splenocytes transferred into congenic C57BL/6 (CD45.1^-CD45.2⁺) mice. HEL-CTL or vector-control CTL were transferred as in figure 5. LPS or saline (No LPS) was administered i.p. on the day of transfer. Lymphocytes in the spleen and LN were isolated 6 d later, and gated on the adoptively transferred CD45.1⁺ cells. (A) IgM^a and CFSE staining among splenocytes. Numbers of dots on the different plots were normalized to show approximately equivalent numbers of IgM^a- CFSE⁺ cells, which comprise primarily transferred T cells. Percent of IgM^a- CFSE⁺ and IgM^a+ cells are indicated in the respective boxes. (B) Analyses were performed as in (A), but for LN cells. Representative plots are shown.