Cutting Edge: Enhancement of Antibody Responses Through Direct Stimulation of B and T Cells by Type I IFN

Mice

C57BL/6 mice were purchased from Charles River or from the specific pathogen-free unit at the Institute for Animal Health (Compton, U.K.). 129S6 (129), 129-Ifnar1^tm1Agt (Ifnar1^−/−), B6.129S2-Igh-6^tm1Cgn/J (μMT), and B6;129P-Tcrb^tm1Mom Tcrd^tm1Mom/J (Tcrb^−/−Tcrd^−/−) mice were purchased from the Institute for Animal Health. Bone marrow (BM) chimeras were produced by injecting 5 × 10⁶ total BM cells into irradiated (900 cGy) recipients as indicated in the text.

To generate mice with a conditional IFN-αβR α-chain, IB10 embryonic stem cells were gene targeted so that exon 10 of the Ifnar1 gene was loxP flanked (Ifnar1^F/F). Upon Cre-mediated deletion of exon 10, a frameshift results in an open reading frame devoid of the transmembrane region and the cytoplasmic signaling domain. Control experiments revealed complete IFN-αβR inactivation upon exon 10 deletion (E. Kamphuis and U. Kalinke, manuscript in preparation). 129/Sv-Ifnar1^tm1Uka (Ifnar1^F/F) mice generated from targeted embryonic stem cells and C.129P2-Cd19^tm1(cre)Cgn/J (CD19-Cre mice) (20) were backcrossed 10 times with C57BL/6 mice before both strains were intercrossed. Mice homozygous for Ifnar1^F/F and carrying one CD19-Cre allele showed B cell-specific Ifnar1 deletion that was >97% efficient, as indicated by genetic and functional analysis (data not shown). Mice with a conditional Ifnar1 gene were bred under specific pathogen-free conditions.

All animal experimentation was done with the approval of the Home Office and the Ethical Review Committee of the Institute for Animal Health.

Immunizations

One hundred micrograms of chicken gamma globulin (CGG) (Stratech Scientific) was injected s.c. either in PBS alone or in PBS containing 10⁵ U IFN-α. Recombinant mouse IFN-α4 was produced by NSo mouse myeloma cells in serum-free medium as described previously (5). In IFN-α-treated mice, IFN-α was also injected 1 and 2 days after administration of Ag at the site of the primary injection (4).

Assay of serum Ab by ELISA

CGG-specific Abs were detected by ELISA as described previously (4). For measurement of IgM and IgG subclasses, rat anti-mouse-IgM (R6-60.2), IgG1 (A85-1), -IgG2a (R19-15), -IgG2b (12-3), and -IgG3 (R40-82) Abs (all from obtained from BD Biosciences Pharmingen) were used. To distinguish IgG2a^a and IgG2a^b Abs, allotype-specific mouse anti-mouse IgG2a^a (8.3) and IgG2a^b (5.7) Abs (both from BD Biosciences Pharmingen) were used for detection.

Visualization of germinal centers (GC)

Draining lymph nodes (LN) were obtained from mice 12 days after immunization. Sections (5–6 μm) were cut from frozen tissues using a CM1900 cryostat (Leica Microsystems). Acetone-fixed sections were labeled with biotinylated peanut agglutinin (Vector Laboratories) at 50 μg/ml for 30 min, and, after washes, with HRP-avidin-biotin-complex (Vector Laboratories). Sections were then treated for 4 min with a peroxidase diaminobenzidine solution (Vector Laboratories), after which the reaction was stopped with water. Sections were counterstained with Meyer’s hematoxylin. The average number of GC per section was determined by counting 3–9 different sections per LN, analyzing 9 LN each for mice receiving injections with CGG and CGG + IFN-α. No GC were detected in LN sections from control unimmunized mice. Sections were visualized and photographed using a DMLS microscope (Leica Microsystems) and a Polaroid DMC.

Impaired ability of IFN-α to enhance Ab response when B cells lack expression of the IFN-αβR

Previously, we reported that injection of IFN-αβ greatly enhances the Ab response against soluble protein Ags (4). The augmenting effects of IFN-αβ included stimulating production of IgM and all subclasses of IgG and were found to increase when IFN-αβ was injected not only at the same time as the Ag but also 1 and 2 days after administering the Ag. To investigate further the mechanisms of IFN-αβ adjuvant activity, we used the same immunization protocol, i.e., s.c. injection of soluble Ag (CGG) plus IFN-α on day 0 followed by injections of IFN-α alone on days 1 and 2 (given at the same site as the day 0 injection). As shown in Fig. 1⇓, injection of IFN-α caused a significant increase in the number of GC generated after immunization with CGG.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Stimulation of GC formation by IFN-αβ. A, Representative sections of LN from mice receiving injections with CGG (top) or CGG + IFN-α (bottom), showing GC (indicated by arrows) at 12 days after immunization. B, Average number of GC per section for LN from CGG- or CGG + IFN-α-injected mice. Data are mean counts for 3–9 sections per LN, 9 LN per treatment (±SD). Differences are statistically significant (p < 0.005 by two-sample t test).

As an initial approach to determine whether B cells serve as direct targets of IFN-α adjuvant activity, we generated mixed BM chimeras in which B cells were deficient in expression of the IFN-αβ receptor. B cell-deficient μMT mice (21) were irradiated and reconstituted with a 1:1 mixture of BM from syngeneic (μMT) mice and Ifnar1^−/− mice. In the resulting chimeras (termed IFN-αβR^−/− B cell chimeras), all B cells were derived from Ifnar1^−/− progenitors, whereas other BM-derived cells arose from both Ifnar1^−/− and Ifnar1^+/+ (μMT-derived) precursors. Control chimeras were produced by reconstituting irradiated μMT mice with μMT plus 129 (WT) BM; in these mice, all cells were IFN-αβR^+/+.

Groups of both types of chimeras were immunized by injection of CGG alone or CGG + IFN-α, and CGG-specific serum Ab titers were measured 10 days later. IFN-α markedly enhanced the anti-CGG response, including production of IgM and all subclasses of IgG, in control chimeras (Fig. 2⇓). In striking contrast, injection of IFN-α engendered little or no increase in the anti-CGG Ab response in IFN-αβR^−/− B cell chimeras. Because B cells represented the only uniformly IFN-αβR^−/− population in these mice, the results suggested that direct stimulation of B cells was required for optimal enhancement of Ab responses by IFN-α.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Defective stimulation of Ab response by IFN-α in chimeras in which B cells lack IFN-αβR expression. IFN-αβR^−/− B cell chimeras (▪) or control chimeras (□) mice were immunized by injection of CGG alone or CGG + IFN-α, and serum Ab levels were measured 10 days later. Data show mean endpoint titers ± SD for IgM and the indicated IgG subclasses (2–3 mice/group), and are representative of two separate experiments.

In IFN-αβR^−/− B cell chimeras, non-B hemopoietic cells included a mixture of IFN-αβR^−/− and IFN-αβR^+/+ cells. To exclude the possible influence of IFN-αβR^−/− non-B cells on the adjuvant activity of IFN-α, a second approach was taken to target IFN-αβR-deficiency selectively to B cells, using the Cre-loxP system (22). Mice were generated in which the Ifnar1 gene, which encodes one of the two chains of the IFN-αβR, was flanked by loxP sites (Ifnar1^F/F). These mice were then crossed to mice expressing the Cre recombinase under control of the CD19 promoter (CD19-Cre⁺), leading to deletion of the IFN-αβR selectively in B cells (20). The number of B cells in these mice was equivalent to that in control mice, indicating that there is no requirement for IFN-αβR signaling in B cell development, consistent with normal lymphocyte production in IFN-αβR^−/− mice (data not shown).

CD19-Cre⁺ Ifnar1^F/F mice and Cre-negative Ifnar1^F/F mice were immunized with CGG or CGG + IFN-α, and serum Abs were measured 10 or 35 days later (Fig. 3⇓). Consistent with the results from the mixed BM chimeras, the ability of IFN-α to augment the anti-CGG Ab response was greatly reduced when B cells were unable to respond directly to IFN-αβ. Thus, injection of IFN-α strongly enhanced the production of anti-CGG Abs in Cre⁻ Ifnar1^F/F mice but stimulated minimal augmentation of the response in CD19-Cre⁺ Ifnar1^F/F mice. Of note, IFN-α showed equivalent adjuvant activity in control CD19-Cre⁺ and Cre-negative mice, indicating that expression of Cre alone in B cells did not alter Ab response (data not shown). Therefore, these data provide strong evidence in an independent model that direct stimulation of B cells by IFN-α is important in IFN-α-mediated enhancement of the Ab response.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Defective stimulation of Ab response by IFN-α in mice selectively lacking the IFN-αβR on B cells. Cre⁻ Ifnar1^F/F (□) and CD19-Cre⁺ Ifnar1^F/F mice (▪) were immunized by injection of CGG alone or CGG + IFN-α, and serum Ab levels were measured 10 (upper graphs) or 35 days (lower graphs) later. Data show mean endpoint titers ± SD for IgM and the indicated IgG subclasses (3 mice/group), and are representative of two separate experiments. IgM titers were very low in all groups on day 35 (data not shown).

Impaired ability of IFN-α to enhance Ab response when T cells lack expression of the IFN-αβR

Because the Ab response to CGG is completely dependent on T cell help, it was of interest to determine whether T cells also represent direct targets of IFN-α in the enhancement of Ab responses. For this purpose, mixed BM chimeras were generated to produce mice in which T cells were IFN-αβR-deficient. T cell-deficient Tcrb^−/−Tcrd^−/− mice were irradiated and reconstituted with a mixture of BM from Tcrb^−/−Tcrd^−/− and Ifnar1^−/− mice, so that all T cells were derived from Ifnar1^−/− progenitors. In these mice (termed IFN-αβR^−/− T cell chimeras), B cells (and other BM-derived cells) were both IFN-αβR^−/− and IFN-αβR^+/+. Usefully, however, Abs produced by the two types of B cells could be distinguished, because Ifnar1^−/−- and Tcrb^−/−Tcrd^−/−-derived B cells expressed h chains of a and b allotypes, respectively.

IFN-αβR^−/− T cell chimeras and control chimeras (Tcrb^−/−Tcrd^−/− recipients reconstituted with 129 (WT) and Tcrb^−/−Tcrd^−/− BM) were immunized as described above and Ab responses measured 10 days later (Fig. 4⇓). As expected, when IgH^a Abs (produced by Ifnar1^−/− B cells in IFN-αβR^−/− T cell chimeras, and by 129 (WT)-derived B cells in control chimeras) were measured, it was apparent that IFN-α strongly enhanced the response in control chimeras but not in IFN-αβR^−/− T cell chimeras (Fig. 4⇓A). This result was in accordance with the requirement for direct stimulation of B cells by IFN-α demonstrated above. Significantly, a similarly reduced response was apparent in IFN-αβR^−/− T cell chimeras when IgH^b Abs (produced by Tcrb^−/−Tcrd^−/−-derived B cells in both types of chimeras) were measured (Fig. 4⇓B). Thus, when T cells were uniformly IFN-αβR^−/−, expression of the IFN-αβR on B cells was insufficient to allow for IFN-α-mediated enhancement of the Ab response. These results indicated that IFN-αβR-mediated stimulation of T cells also contributes to the enhancement of Ab responses by IFN-α.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Defective stimulation of Ab response by IFN-α in chimeras in which T cells lack IFN-αβR expression. IFN-αβR^−/− T cell chimeras (▪) and control chimeras (□) were immunized by injection of CGG alone or CGG + IFN-α, and serum Ab levels were measured 10 days later. A, Endpoint titers (±SD) of IgG2a^a Abs, which were produced by 129-derived B cells in control chimeras and by Ifnar1^−/−-derived B cells in IFN-αβR^−/− T cell chimeras. B, Endpoint titers (±SD) of IgG2a^b Abs, which were produced by Tcrb^−/− Tcrd^−/−-derived B cells in both types of chimeras. Data are means from 3 mice/group.

Overall, these results demonstrate a clear stimulatory effect of IFN-αβR-mediated signaling in B and T cells relevant to the generation of Ab responses in vivo. In considering the potential mechanism(s) of action, it should be noted that the present data do not establish that IFN-αβ acts directly on those cells that are responding specifically to Ag. For example, it is conceivable that IFN-α triggers B or T cells globally to release cytokines that act indirectly to stimulate Ag-specific cells. Moreover, although it seems most likely that the requirement for T cell responsiveness to IFN-αβ relates to the priming and/or function of CD4⁺ Th cells, we cannot rule out the possibility that triggering of other T cells (including CD8⁺ T cells) might lead to signals that indirectly stimulate the Ab response.

Notwithstanding these caveats, in vitro studies have indicated a number of stimulatory effects of IFN-αβ on B and T cells that could be relevant to our in vivo observations. For B cells, these include protection from apoptosis, augmentation of Ag receptor-triggered proliferation, and promotion of differentiation into Ab-forming cells (15, 16, 17, 19). Similarly, protection of activated T cells from apoptosis could lead to increased and prolonged availability of T cell help (18). However, given the sometimes conflicting results stemming from in vitro investigation of IFN-αβ function, elucidation of how IFN-αβ enhances Ab responses will require further examination of the effects of IFN-αβR signaling in T and B cells in vivo. It will be of particular interest to investigate whether IFN-αβ can enhance one or more stimulus essential for T cell-B cell collaboration and the GC reaction, such as those mediated by CD28-B7, CD40-CD40L, OX40-OX40L, or BAFF-BAFF-R interactions (23, 24, 25), perhaps through up-regulation of the relevant receptor.

The results of this study provide further insight into the way that infection-associated signals can impact on the adaptive immune response. The demonstration that exposure of B and T cells to the innate cytokine IFN-αβ enhances the in vivo Ag-specific Ab response adds to previous observations that these cells express certain Toll-like receptors and can respond to pathogen components (26, 27). Hence, lymphocytes are sensitive to direct and indirect signs of infection and can modify their function in response to these signals. Therefore, optimal generation of immune responses is likely to rely upon innate triggering of multiple cell types, including DCs, B cells, and T cells.