Pneumococcal polysaccharide abrogates conjugate-induced germinal center reaction and depletes antibody secreting cell pool, causing hyporesponsiveness

Abstract

Background: Plain pneumococcal polysaccharide (PPS) booster administered during second year of life has been shown to cause hyporesponsiveness. We assessed the effects of PPS booster on splenic memory B cell responses and persistence of PPS-specific long-lived plasma cells in the bone marrow (BM).

Methods: Neonatal mice were primed subcutanously (s.c.) or intranasally (i.n.) with pneumococcal conjugate (Pnc1-TT) and the adjuvant LT-K63, and boosted with PPS+LT-K63 or saline 1, 2 or 3 times with 16 day intervals. Seven days after each booster, spleens were removed, germinal centers (GC), IgM(+), IgG(+) follicles and PPS-specific antibody secreting cells (AbSC) in spleen and BM enumerated.

Results: PPS booster s.c., but not i.n., compromised the Pnc1-TT-induced PPS-specific Abs by abrogating the Pnc1-TT-induced GC reaction and depleting PPS-specific AbSCs in spleen and limiting their homing to the BM. There was no difference in the frequency of PPS-specific AbSCs in spleen and BM between mice that received 1, 2 or 3 PPS boosters s.c.. Repeated PPS+LT-K63 booster i.n. reduced the frequency of PPS-specific IgG(+) AbSCs in BM.

Conclusions: PPS booster-induced hyporesponsiveness is caused by abrogation of conjugate-induced GC reaction and depletion of PPS-specific IgG(+) AbSCs resulting in no homing of new PPS-specific long-lived plasma cells to the BM or survival. These results should be taken into account in design of vaccination schedules where polysaccharides are being considered.

Figure 1. Subcutaneous administration of PPS-1 booster depletes Pnc1-TT-induced PPS-1-specific AbSC pool in the spleen.

PPS-1 (upper panels) and TT (lower panels) specific IgG⁺ AbSC measured by ELISPOT in spleen (A and B) and bone marrow (C and D), shown as number of spots (mean±SD) per 10⁶ spleen cells and PPS-1- and TT-specific IgG levels (mean EU/ml±SD) in serum measured by ELISA (E and F). Results are shown for mice boosted by either i.n. or s.c. route with saline (open columns), PPS-1+LT-K63 (black columns), Pnc1-TT+LT-K63 (grey columns) and unvaccinated controls (stribe columns), as indicated. Statistical difference between vaccinated groups is indicated in the graphs. All vaccinated mice had significantly higher frequency of PPS-1 -specific IgG⁺ AbSCs in spleen and BM and also higher serum IgG anti-PPS-1 levels compared to unvaccinated controls (B–F), except mice that received PPS-1 booster s.c. which had comparable frequency of PPS-1-specific IgG⁺ AbSCs in spleen (A). The results shown in A–F are from one of two independent experiments showing comparable results (8 mice per group in each experiment).

Figure 2. Plain PPS-1 booster abrogates the Pnc1-TT-induced GC reaction in mice primed as neonates.

(A–C) Mean number of follicles and (D) ratio of GC/follicle per section in consecutive sections from spleen of mice that received booster by either s.c. (left columns) or i.n. (right columns) route with saline (open columns), PPS-1+LT-K63 (black columns), Pnc1-TT+LT-K63 (grey columns) and unvaccinated controls (stribe columns), as indicated. Spleens were removed 7 days after booster in mice primed as neonates with Pnc1-TT+LT-K63 by the same route as the booster. Half of the spleen was snap frozen, serial cryosection prepared, cutting into 7 µm sections at four levels, starting 700 µm into the tissue and the levels separated by 210 µm. Section from all 4 levels were stained with (A) anti-IgM, (B) anti-IgG, (C) PNA, and results (mean±SD) shown for each group. (D) Mean ratio of GC/follicle was calculated for every spleen at all 4 levels for individual mice and results (mean±SD) shown for each group. Statistical difference between vaccinated groups is stated in the graphs. Results in A–D are from one representative of two independent experiments (8 mice per group) showing comparable results.

Figure 3. Plain PPS-1 booster s.c. abrogates the Pnc1-TT-induced GC reaction in mice primed as neonates.

Active germinal centers in spleen sections were enumerated with PNA staining (upper panels). Double fluorescent staining was performed with PNA and MOMA-1 (metallophilic marginal macrophages) to show the follicular structure (top panel). IgM⁺ and IgG⁺ follicles were identified with anti-IgM (middle panel) and anti-IgG (lower panel) staining, 7 days after booster with 5.0 µg PPS-1 and 5.0 µg LT-K63 s.c. (left) or i.n. (right). Spleen sections, 7 µm, were prepared from four different levels in the spleen, starting 700 µm into the tissue and each level separated by 210 µm. One representative section per group is shown. Results are from one representative of two independent experiments (8 mice per group) showing comparable results.

Figure 4. PPS-1 booster depletes Pnc1-TT-induced memory cells in spleen preventing homing of PPS-1-specific AbSCs to BM.

PPS-1-specific (A–B) and TT-specific (D–E) IgG⁺ AbSCs, shown as number of spots (mean±SD) per 10⁶ cells, in spleen (A and D) and bone marrow (B and E) measured by ELISPOT, and PPS-1- (C) and TT-specific (F) IgG Abs (mean EU/ml±SD) in serum measured by ELISA, at day 7, 23 and 39 after booster with saline (open squares), PPS-1+LT-K63 (filled squares) or unvaccinated controls (open circles). The results shown are from one of two independent experiments (eight mice/group for each time point) showing comparable results.

Figure 5. PPS-1 booster abrogation of Pnc1-TT-induced GCs reduces levels, avidity and protective efficacy of PPS-1-specific IgG.

PPS-1-specific IgG levels (mean EU/ml±SD) in serum (A) and their avidity index (mean AI±SD) (B) measured by ELISA −2 days before and 7, 23 and 39 days after s.c. (left panels) or i.n. (right panels) booster with saline (open squares), PPS-1+LT-K63 (filled squares) in mice primed with Pnc1-TT as neonates or unvaccinated controls (open circles). Six weeks after the booster the mice were challenged with live S. pneumoniae serotype 1 to assess protection against bactermia and lung lung infection . Pneumococcal colony forming units (CFU/ml) in blood (C, left graph) and lungs (C, right graph) 24 h after challenge shown for each mouse (N = 8/group), the median for each group is indicated by a line. Statistical difference in bacteremia or lung infection between vaccinated groups is indicated in the graphs. The results shown are from one of two independent experiments (eight mice/group for each time point) showing comparable results.

Figure 6. A single PPS-1 booster s.c. is sufficient to completely deplete PPS-1-specific-memory established by neonatal Pnc1-TT-priming.

PPS-1-specific IgG⁺ AbSCs, shown as number of spots (mean±SD) per 10⁶ cells, in spleen (A) and bone marrow (B) measured by ELISPOT, and PPS-1-specific IgG Abs (mean EU/ml±SD) in serum (C) measured by ELISA, at day 7, 23 and 39 after mice receiving s.c. booster with saline, PPS-1+LT-K63 or unvaccinated controls. Statistical difference between test groups and unvaccinated controls are indicated as; * P<0.05; ** P≤0.001, and the difference between vaccinated groups is stated in the graphs. The results are shown for one of two independent experiments (eight mice/group for each time point), showing comparable results.

Figure 7. Repeated plain PPS-1 boosters i.n. reduce PPS-1-specific long-lived plasma cell pool in the bone marrow.

PPS-1-specific IgG⁺ AbSCs, shown as number of spots (mean±SD) per 10⁶ cells, in spleen (A) and bone marrow (B) measured by ELISPOT, and PPS-1-IgG Abs (mean EU/ml±SD) in serum (C) measured by ELISA, at day 7, 23 and 39 after mice receiving i.n. booster with saline, PPS-1+LT-K63 or unvaccinated controls. Statistical difference between test groups and unvaccinated controls are indicated as; * P<0.05; ** P≤0.001, and the difference between vaccinated groups is stated in the graphs. The results are shown for one of two independent experiments (eight mice/group for each time point), showing comparable results.