Passively transferred IgG enhances humoral immunity to a red blood cell alloantigen in mice

Abstract

Antibodies are typically thought of as the endpoint of humoral immunity that occur as the result of an adaptive immune response. However, affinity-matured antibodies can be present at the initiation of a new immune response, most commonly because of passive administration as a medical therapy. The current paradigm is that immunoglobulin M (IgM), IgA, and IgE enhance subsequent humoral immunity. In contrast, IgG has a "dual effect" in which it enhances responses to soluble antigens but suppresses responses to antigens on red blood cells (RBCs) (eg, immunoprophylaxis with anti-RhD). Here, we report a system in which passive antibody to an RBC antigen promotes a robust cellular immune response leading to endogenous CD4+ T-cell activation, germinal center formation, antibody secretion, and immunological memory. The mechanism requires ligation of Fcγ receptors on a specific subset of dendritic cells that results in CD4+ T-cell activation and expansion. Moreover, antibodies cross-enhance responses to a third-party antigen, but only if it is expressed on the same RBC as the antigen recognized by the antibody. Importantly, these observations were IgG subtype specific. Thus, these findings demonstrate that antibodies to RBC alloantigens can enhance humoral immunity in an IgG subtype-specific fashion and provide mechanistic elucidation of the enhancing effects.

Graphical abstract

Figure 1.

Experimental design. Recipient mice received a passive immunization with 1 μg of anti-HOD monoclonal antibody (anti-HOD mAb), nonspecific mAb, or PBS control followed by an RBC transfusion. Each transfusion consisted of 100 μL of packed leukoreduced RBCs at 20% hematocrit, with 50 μL of allogeneic HOD RBCs, and 50 μL syngeneic B6 RBCs, labeled with DiO and DiI lipophilic dyes, respectively. Before transfusion, labeled HOD-DiO and B6-DiI RBCs were mixed at a 1:1 ratio and analyzed on a flow cytometer to determine the pretransfusion ratio. After transfusion, RBCs and sera were collected from experimental mice and used to determine RBC survival and alloantibody production.

Figure 2.

Passive immunization with anti-HOD mAb IgG2a or IgG2c enhances alloantibody production upon HOD RBC transfusion. Recipient mice received a passive immunization of individual anti-HOD mAb subtypes followed by an RBC transfusion. RBCs and sera were collected at multiple time points over 21 days. (A) The survival of allogeneic HOD RBCs, as a function of control B6 RBCs, was calculated. (B) The percentage of circulating HOD RBCs was determined. The percentage of HOD RBCs in PBS-treated animals were normalized to 100% across all time points. Sera were analyzed by flow crossmatch against HOD and control B6 RBCs for IgM (C), Igs (D), IgG1 (E), IgG2a (F), IgG2b (G), IgG2c (H), and IgG3 (I) anti-HOD alloantibodies. (J) Representative flow plots of flow crossmatch data are provided. Each experiment was performed 3 times with 3 to 5 mice per group. Representative data are shown. Data were analyzed with a repeated measures 2-way ANOVA with Dunnett multiple comparisons to PBS-treated animals. *P ≤ .05, **P ≤ .01, ***P ≤ .001, ****P ≤ .0001.

Figure 3.

Generation of a new FcRγ^fl/fl mouse. A new mouse was designed to have conditional expression of activating Fc receptors. (A) Fcer1g gene locus showing insertion of LoxP recognition sequences flanking exons 2 and 3. Splenocytes were harvested from B6, FcRγ^fl/fl, FcRγ^fl/fl × CMV^Cre+, and Fcer1g^−/− mice and (B) evaluated for expression of FcγRI, FcγRIII, FcγRIV, and FcγRIIb on multiple leukocyte subsets; MFIs of FcγRs are displayed in a heat map. To evaluate whether insertion of LoxP sites interfered with clearance and alloantibody production, B6 and FcRγ^fl/fl mice were passively immunized with anti-HOD mAb followed by an RBC transfusion. HOD RBC survival (C) and anti-HOD alloantibodies (D) were evaluated over 21 days. In analogous experiments, immune responses in FcRγ^fl/fl × CMV^Cre+, and Fcer1g^−/− were directly compared and HOD RBC survival (E) and anti-HOD alloantibody production (F) were evaluated. Each experiment was performed 3 times with 3 to 5 mice per group. Representative data are shown. Data were analyzed with a repeated measures 2-way ANOVA with multiple comparisons test to control B6 mice. ****P ≤ .0001.

Figure 4.

Absence of activating FcγR expression in cells driven by CD11c or Zbtb promoters abrogates RBC alloimmunization. (A) FcRγ^fl/fl mice were bred with cre-expressing transgenic mice and spleens were harvested and stained with antibodies to delineate cell subsets and FcγR expression and relative MFIs of individual FcγRs are displayed as a heat map. To assess the functional consequence of modulating expression of activating FcγRs, recipients were passively immunized with anti-HOD mAb IgG2c, nonspecific anti-K mAb IgG2c, or control PBS followed by an RBC transfusion (as described in Figure 1). Survival of HOD RBCs was calculated at 24 hours posttransfusion (B) and over a 21-day time course (C). IgM (D) and total Igs (E) anti-HOD alloantibodies were determined for days 7, 14, and 21 by flow crossmatch. Each experiment was performed 3 times with 3 to 5 mice per group. Representative data are shown. Data were analyzed with a 1-way ANOVA with Dunnett multiple comparisons test to control B6 mice. *P ≤ .05, **P ≤ .01, ***P ≤ .001, ****P ≤ .0001.

Figure 5.

Anti-HOD mAb IgG2c leads to enhanced T-cell proliferation and increases RBC consumption by DCs. To evaluate RBC consumption, B6 mice were passively immunized followed by a DiO⁺ HOD RBC transfusion. Spleens were harvested 18 to 24 hours posttransfusion and leukocytes were stained to delineate cell subsets. The percent erythrophagocytosis of total Thy1.2⁻TER119⁻ leukocytes was determined (A) and the DiO MFI of individual APC subsets was calculated (B). To assess whether passive immunization modulated T-cell responses, B6 mice were adoptively transferred with 1 × 10⁵ purified CD4⁺ OTII T cells labeled with CellTrace-far-red (FR). The next day, recipient mice were passively immunized with PBS, nonspecific anti-K mAb IgG2c, or anti-HOD mAb IgG2c followed by a HOD RBC transfusion. CellTrace-FR dilution (C) and absolute number (D) was determined for CD4⁺Va2⁺Vb5⁺Thy1.1⁺ OTII T cells 3 days posttransfusion. Each experiment was performed 3 times with 3 to 5 mice per group. Representative data are shown. Data were analyzed with a 1-way ANOVA with Dunnett’s multiple comparisons test to the control PBS group. *P ≤ .05, **P ≤ .01, ****P ≤ .0001.

Figure 6.

Passive immunization with anti-HOD mAb promotes memory lymphocyte formation. (A) General experimental design to test whether passive immunization can lead to RBC-specific lymphocyte memory formation. (B) One week posttransfusion sera from B6.IgH^a recipients were collected assayed for IgG1^b alloantibodies by flow crossmatch against HOD and control B6 RBCs. The percent of IgG1^b+ HOD RBCs for a representative experiment is shown (C) and the number of mice with detectable memory responses across all experiments is provided (D). In parallel, splenocytes were harvested and stained with antibodies against CD19, I-A^b, GL7, and PNA to identify germinal center B cells. Representative flow plots (E) and percentage of germinal center B cells (F) are shown. Flow plots are gated on CD19⁺I-A^b+Thy1.1⁻Thy1.2⁻ leukocytes. Each experiment was performed 3 times with 3 to 5 mice per group. Data were analyzed with a 1-way ANOVA with Dunnett multiple comparisons test to the control PBS group. *P ≤ .05; ****P ≤ .0001.

Figure 7.

Anti-RBC mAbs cross-enhance to third-party alloantigens only if they are expressed on the same RBCs. Recipient B6 mice were pretreated with either PBS, anti-HOD mAb, or anti-K mAb. For each pretreatment, 4 different groups were set up, each receiving a different RBC transfusion: HOD RBCs, K RBCs, HOD RBCs mixed with K RBCs (HOD+K), or F1 RBCs expressing both alloantigens. Serum was obtained 21 days after transfusion and was assayed for polyclonal anti-HOD antibodies (A) and polyclonal anti-K antibodies (B) by indirect immunofluorescence using HOD or K RBC targets, respectively. Data were analyzed with a 1-way ANOVA using a Dunnett multiple comparison test to the control PBS group. *P ≤ .05; ****P ≤ .0001.