Lymphotoxin α-deficient (LTα−/−) mice show dramatically reduced IgG responses after either primary or secondary immunizations with sheep red blood cells (SRBC). When splenocytes from SRBC-primed wild-type donor mice were infused into irradiated naive wild-type recipient mice, they generated a robust memory IgG response, but not when infused into LTα−/− recipients, indicating that the microenvironment that develops in LTα−/− mice is incompetent to support the activation of this memory response. When irradiated wild-type mice were reconstituted with splenocytes from primed LTα−/− donors and then challenged with the same immunizing Ag, no memory response was observed, indicating further that memory cells could not be generated in the LTα−/− environment. To address which lymphocyte subsets were impaired in the LTα−/− mice, we performed reconstitution experiments using a hapten/carrier system and T cells and B cells from different primed donors. There was no detectable defect in either the generation or expression of memory T cells from LTα−/− donors. In contrast, B cells were not primed for memory in the microenvironment of LTα−/− mice. Additionally, primed wild-type memory B cells could not express a memory IgG response in the LTα−/− microenvironment. Thus, splenic white pulp structure, which depends on the expression of LTα for its development and maintenance, is needed to support the generation of memory B cells and to permit existing memory B cells to express an isotype switched memory Ig response following antigenic challenge.

Recent studies have shown that signals induced by membrane lymphotoxin (LT)4 are essential for the organogenesis of peripheral lymph nodes and Peyer’s patches and the formation of both the primary and secondary B cell follicles in the spleen white pulp (1, 2, 3, 4). Membrane LT is a heterotrimer of LTα- and LTβ-chains with predominant stoichiometry LTα1β2 (5). Signaling by the LTα1β2 heterotrimer is mediated through the LTβ receptor (LTβR). Thus, membrane LT signals are independent of the type I or type II TNF receptors that mediate signals from the soluble LTα3 homotrimer (5, 6). Mice rendered deficient in LTα, LTβ, or the LTβR are born with a dramatic impairment of lymph node and Peyer’s patch biogenesis. In addition, LTα−/−, LTβ−/−, and LTβR−/− mice each fail to form distinct splenic T cell and B cell zones, follicular dendritic cell (FDC) clusters, or germinal centers (GC). These structural disturbances are associated with impaired high affinity isotype-switched Ig responses following primary or secondary immunization with T cell-dependent Ags (such as sheep red blood cells (SRBC) or keyhole limpet hemocyanin (KLH)) when administered without adjuvants (7, 8, 9, 10).

Long-term reconstitution of lethally irradiated LTα−/− mice with bone marrow cells from wild-type (wt) mice leads to restoration of the ability to form FDC clusters, GC, and adjuvant-independent strong IgG responses (8). We have recently shown that B cells alone are required to deliver the LTα-dependent signals that restore the formation of FDC clusters, GC, and recovery of IgG responses in LTα−/− mice (11). Expression of LTα by T cells is not required. In contrast, when sublethally irradiated LTα−/− mice are reconstituted with spleen cells from wt donors rather than with bone marrow cells, no FDC clusters, GC, or T cell-dependent high affinity IgG responses are detected 2 wk after cell transfer and immunization with SRBC (8). These findings suggest that the microenvironment that develops in LTα−/− mice cannot support a high affinity isotype-switched Ig response, but that at least some aspects of this microenvironment are plastic and can be restored under the influence of sustained LT-dependent signaling through the LTβR. The ability to mount a high affinity isotype-switched Ig response appears to be correlated with the presence of clusters of FDC, which can be detected between 2 and 3 wk following transfer of LTα-expressing B cells into LTα−/− recipients. The present study was undertaken to determine whether the lymphoid tissue microenvironment that forms under the influence of LT is required to support either the induction or expression of memory responses.

Several studies suggest that memory B cells form primarily in GC and that their formation requires the presence of FDC clusters (12, 13, 14, 15). A prominent contribution of GC to the formation of memory B cells is suggested by the observation that the B cell Ag receptors expressed by memory B cells carry a significant number of somatic mutations and generally show high affinity for the eliciting Ag. Thus, memory B cells appear to have passed through GC during their development. Like maturing B cells, Ag-specific T cells also undergo phenotypic changes within GC (16, 17, 18); however, it remains unclear whether these GC-dependent changes in T cells also represent an integral part of the program leading to the generation of memory T cells. It also remains unclear whether GC or FDC clusters play an important role in the expression of memory responses by established memory T and B cells.

Until recently, the lack of animal models in which the clustering of FDC and the formation of GC could be regulated has limited our understanding of the role of these structures in the generation of memory cells and in the maintenance of memory responses. LTα−/− mice fail to form clusters of FDC and are unable to generate GC following immunization with T-dependent Ags (7). These mice provide a model in which to study whether these structures contribute to the generation of memory T or B cells and to the functional expression of the memory response. We demonstrate here that the microenvironment of LTα−/− mice supports the generation of memory T cells but not of memory B cells. Furthermore, the microenvironment in LTα−/− mice cannot support the expression of functional memory IgG responses, even when sensitized memory cells from wt mice are provided by adoptive transfer. Of interest, when sublethally irradiated LTα−/− mice are reconstituted with splenocytes from wt mice, immunization elicits the formation of clusters of peanut agglutinin+ (PNA+) cells, apparently in the absence of associated FDC, suggesting that normal B cells can be activated to form GC-like structures without FDC; however, these clusters of PNA+ cells fail to differentiate into functional memory cells in the splenic microenvironment of LTα−/− mice. The dysfunctional nature of these GC-like structures is underscored by their dissociation from the apoptotic process that normally appears to be a consequence of B cell selection in these structures (15, 19).

C57BL/6J, 129Sv, and B cell receptor (BCR)−/− (C57BL/6J-Igh-6tmlCgn) mice were obtained from The Jackson Laboratory (Bar Harbor, ME). LTα−/− mice (1) were maintained on a mixed 129Sv × C57BL/6 background and were bred under specific pathogen-free conditions.

Four milliliters of 10% SRBC (Colorado Serum, Denver, CO) in PBS were incubated with 4 ml of NP-Osu (1 mg/ml, Biosearch Technologies, San Rafael, CA) in 0.15 M NaHCO3 for 2 h at room temperature. The NP-conjugated SRBC were washed with PBS and resuspended in 8 ml of PBS. Immunization was with 0.1 ml of this suspension injected i.p. NP13-KLH was also purchased from Biosearch Technologies.

Specific Abs were measured and analyzed as previously described (8). For measurement of anti-SRBC Abs, 96-well Falcon plates (Becton Dickinson, Lincoln Park, NJ) were coated with SRBC (150 μl of 0.1% SRBC in PBS per well). For anti-NP Abs, 96-well Immulon 4 plates (Dynatech Laboratories, Chantilly, VA) were coated with NP2.5-BSA (10 μg/ml, Biosearch Technologies) for 1 h. Unbound Ags were removed by washing with PBS. Diluted mouse sera were then added and incubated at 4°C for 1 h. Bound Abs were detected using 100 μl of 1:2000 diluted alkaline phosphatase-conjugated goat anti-mouse IgG-specific antiserum (Southern Biotechnology Associates, Birmingham, AL), followed by addition of the alkaline phosphatase substrate p-nitrophenyl phosphate (Sigma, St. Louis, MO) at 1 mg/ml. The mean OD405 from triplicate wells was compared to a various dilutions of a standard anti-NP immune serum to calculate the relative units (RU) using linear regression analysis. The results are reported as means ± SEM.

Whole spleen cell suspensions were prepared from single donor mice and were injected i.v. into recipients that had been irradiated with 750 rads 3 h earlier. When indicated, SRBC were injected i.v. together with the spleen cell suspensions. Each recipient received all of the cells derived from a single donor spleen.

Nylon wool columns (Polysciences, Warrington, PA) were used to enrich spleen cell suspensions for T or B cells. Splenocytes were incubated on nylon wool columns with 10% FCS in DMEM for 1 h at 37°C. Nonadherent cells were eluted with 10% FCS in DMEM at 37°C. Enriched T cells in the fraction of nonadherent cells were further purified by panning on tissue culture dishes coated with goat anti-mouse Ig H/L chain (Southern Biotechnology Associates), yielding 75–85% T cells. The contamination by B cells was <5%. Cells that adhered to nylon wool were eluted using cold PBS. Eluted cells were 80–90% B cells. After panning on tissue culture dishes coated with anti-Thy1.2 Ab, contamination by T cells was reduced to <5%.

Spleens were harvested, embedded in OCT compound (Miles, Elkhart, IN), and frozen in liquid nitrogen. Frozen sections (6–10 μm thick) were fixed, quenched, and stained as previously described (8) using 0.2% H2O2 in methanol. After washing, the sections were stained by first incubating with FITC-conjugated B220 (PharMingen, San Diego, CA), and biotinylated PNA (Vector, Burlingame, CA), all at 1:100 dilution. HRP-conjugated rabbit anti-FITC (Dako, Glostrup, Denmark; diluted 1:10) was added 1 h later. Sections were then incubated for 1 h with one drop of alkaline phosphatase (AP)-conjugated streptavidin (Zymed, South San Francisco, CA), and color development for bound AP and HRP was with an AP reaction kit (Vector) and with diaminobenzidine.

Cells undergoing apoptosis were detected using a modified TUNEL method (20). Tissue sections were incubated with 2 mM digoxigenin-conjugated dUTP (Boehringer Mannheim, Indianapolis, IN) and 5 U of TdT in 0.5 M cacodylate (pH 6.8), 1 mM CoCl2, 0.5 mM DTT, 0.05% BSA, and 0.15 M NaCl. After washing in Tris-buffered saline, sections were incubated with sheep anti-digoxigenin Ab (Boehringer Mannheim) in Tris-buffered saline, washed, and further incubated with HRP-conjugated anti-sheep Ig Ab (Jackson ImmunoResearch, West Grove, PA). Color development for bound HRP was with 100 μg/ml 3-amino-9-ethylcarbazole in 0.17 M sodium acetate (Sigma). Sections were then counterstained with 1% methyl green.

We previously reported that LTα−/− mice developed a dramatically reduced Ag-specific IgG response following immunization with the T cell-dependent Ags SRBC or KLH without adjuvant (8). We now extend these experiments to test the character of the IgG response in LTα−/− mice that received repeated immunizations intended to elicit immunological memory. Three months after initial priming, LTα−/− and wt mice were challenged by i.p. immunization with 108 SRBC without adjuvant. Anti-SRBC IgG was measured 5 days after the booster immunization (Fig. 1,A). Measurements were made 5 days after challenge because at this time memory responses in wt mice show strong IgG production, whereas primary IgG responses remain undetectable or low. Wt mice mounted an anti-SRBC memory response that was more than 200-fold higher than that of LTα−/− mice, suggesting that there was a severe impairment of the memory response in the LTα−/− mice. We had previously shown that LTα−/− mice were not absolutely unable to generate an IgG response, but could do so following immunization with either NP-OVA (21) or SRBC (8) when the Ag was administered with an adjuvant. Therefore, we tested the memory IgG response following priming with IFA (Fig. 1 B). Although LTα−/− mice showed a robust IgG response when immunized with SRBC in IFA, when challenged 3 mo later, they showed only a weak anti-SRBC memory response. These experiments suggest that LTα−/− mice either are unable to express a memory response or fail to generate long-lived memory lymphocytes.

FIGURE 1.

Impaired IgG anti-SRBC memory response in LTα−/− mice. Groups of three to five 8-wk-old mice (wt, □; LTα−/−, ▪) were immunized i.p. on day 0 with 108 SRBC without adjuvant (A) or with IFA (B) and given a boost with the same dose without adjuvant 3 mo later. Serum was collected 10 days after the initial immunization (Primary) and 5 days after the booster immunization (Memory). SRBC-specific IgG was measured by ELISA as described in Materials and Methods. Data shown represent the means ± SEM of triplicate determinations from three to five mice. Similar results were obtained in a replicate experiment.

FIGURE 1.

Impaired IgG anti-SRBC memory response in LTα−/− mice. Groups of three to five 8-wk-old mice (wt, □; LTα−/−, ▪) were immunized i.p. on day 0 with 108 SRBC without adjuvant (A) or with IFA (B) and given a boost with the same dose without adjuvant 3 mo later. Serum was collected 10 days after the initial immunization (Primary) and 5 days after the booster immunization (Memory). SRBC-specific IgG was measured by ELISA as described in Materials and Methods. Data shown represent the means ± SEM of triplicate determinations from three to five mice. Similar results were obtained in a replicate experiment.

Close modal

To test whether the lymphoid tissue microenvironment that forms in congenitally LTα-deficient mice can support the function of memory lymphocytes, we transferred splenic lymphocytes from either naive or SRBC-immunized wt mice into irradiated naive wt or LTα−/− recipients and then challenged these chimeric recipients with SRBC. In previously immunized wt animals, a strong memory IgG response can typically be detected as early as day 3–5 after i.p. challenge. When naive cells were transferred, development of a significant primary IgG response was not detected 5 days after immunization of the recipients, although it was by 10 days (Fig. 2). In contrast, when primed lymphocytes from SRBC-immunized wt mice (expected to contain memory B and T cells) were transferred into naive wt recipients, challenge with SRBC elicited a high titer IgG response on day 5 with a further increase by day 10. Strikingly, when lymphocytes were transferred from primed wt mice to LTα−/− recipients, rechallenge with SRBC failed to produce a high titer memory IgG response (even by day 10 after rechallenge). Thus, preformed memory cells elicited in wt mice were unable to express a mature memory response in the disturbed lymphoid tissue environment that exists in LTα−/− mice. The disturbed microenvironment in LTα−/− mice can support neither primary (8) nor memory responses, even when LTα-expressing naive or memory cells are provided.

FIGURE 2.

Primed memory cells do not generate memory IgG responses following transfer to LTα−/− mice. Naive wt or LTα−/− mice were sublethally irradiated with 750 rads, and then were treated with an i.v. infusion of 108 SRBC and spleen cells from wt mice that were either unprimed or that had been primed i.p. with 108 SRBC in PBS 2 mo prior to transfer. Three mice were pooled from each group as donors. Five and 10 days later, serum samples were collected and anti-SRBC IgG was measured by ELISA. □, Splenocytes from naive donors transferred to a wt recipient; ▥, splenocytes from naive donors transferred to a LTα−/− recipient; ▪, splenocytes from primed wt donors transferred to a wt recipient; and ▤, splenocytes from primed wt donors transferred to a LTα−/− recipient. The results represent means ± SEM of three to five recipient mice per group. This experiment was repeated three times with similar results.

FIGURE 2.

Primed memory cells do not generate memory IgG responses following transfer to LTα−/− mice. Naive wt or LTα−/− mice were sublethally irradiated with 750 rads, and then were treated with an i.v. infusion of 108 SRBC and spleen cells from wt mice that were either unprimed or that had been primed i.p. with 108 SRBC in PBS 2 mo prior to transfer. Three mice were pooled from each group as donors. Five and 10 days later, serum samples were collected and anti-SRBC IgG was measured by ELISA. □, Splenocytes from naive donors transferred to a wt recipient; ▥, splenocytes from naive donors transferred to a LTα−/− recipient; ▪, splenocytes from primed wt donors transferred to a wt recipient; and ▤, splenocytes from primed wt donors transferred to a LTα−/− recipient. The results represent means ± SEM of three to five recipient mice per group. This experiment was repeated three times with similar results.

Close modal

The studies described above showed that primed wt memory cells were unable to express a memory IgG response following transfer into an LTα−/− recipient mouse. To compare the ability of the lymphoid microenvironments of wt and LTα−/− mice to support the formation of memory lymphocytes, we immunized wt and LTα−/− mice with the T-dependent Ag SRBC to provide a stimulus for memory cell formation. Splenocytes from the primed mice were then transferred to sublethally irradiated wt recipients. These reconstituted animals were challenged with SRBC to stimulate memory cells that might have been transferred (Fig. 3). Five days after challenge, mice that had received cells from primed wt donors showed a brisk IgG anti-SRBC response characteristic of established B cell memory. In contrast, cells transferred from primed LTα−/− mice supported no IgG response, similar to cells transferred from naive LTα−/− donors. This clearly indicates that functional memory cells had not been induced in the LTα−/− environment.

FIGURE 3.

Failure to form memory cells in the splenic microenvironment that develops in LTα−/− mice. Naive wt mice were sublethally irradiated with 750 rads, and then were reconstituted with spleen cells from naive wt or LTα−/− donors or from wt or LTα−/− donors that had been primed i.p. with 108 SRBC in PBS 2 mo previously. The reconstituted mice also received at the same time 108 SRBC i.v. Five days later, serum samples were collected and anti-SRBC IgG was measured by ELISA. □, Transfer of splenocytes from an LTα−/− donor; ▪, transfer of splenocytes from a wt donor. The results represent means ± SEM of three to five mice per group. One representative experiment of two is shown.

FIGURE 3.

Failure to form memory cells in the splenic microenvironment that develops in LTα−/− mice. Naive wt mice were sublethally irradiated with 750 rads, and then were reconstituted with spleen cells from naive wt or LTα−/− donors or from wt or LTα−/− donors that had been primed i.p. with 108 SRBC in PBS 2 mo previously. The reconstituted mice also received at the same time 108 SRBC i.v. Five days later, serum samples were collected and anti-SRBC IgG was measured by ELISA. □, Transfer of splenocytes from an LTα−/− donor; ▪, transfer of splenocytes from a wt donor. The results represent means ± SEM of three to five mice per group. One representative experiment of two is shown.

Close modal

Memory cells of both the B and T cell lineages are required to coordinate memory responses in vivo (14, 22). To address which memory cell compartment(s) failed to form in the microenvironment of LTα−/− mice, a hapten-carrier system was applied. To investigate whether LTα−/− mice could generate memory B cells, the following strategy was used. BCR−/− mice were primed i.p. with SRBC in PBS to generate anti-SRBC memory T cells that were free of contamination by naive or memory B cells. Wt mice and LTα−/− mice were primed i.p. with NP-KLH in PBS to provide potential sources of anti-NP memory B cells. Then, 107 SRBC-primed T cells from the spleens of BCR−/− mice were mixed with an equal number of partially purified splenic B cells from NP-KLH-immunized wt or LTα−/− mice and transferred to sublethally irradiated (750 rads) naive wt recipients. Following challenge with NP-SRBC, mice that received B cells from the NP-KLH-immunized wt mice showed a robust memory anti-NP IgG response, whereas mice that received B cells from immunized LTα−/− mice showed at least a 50-fold lower anti-NP IgG response (Fig. 4). These data demonstrate a substantial impairment in the formation of memory B cells in LTα−/− mice.

FIGURE 4.

Lack of memory B cells in the spleen of primed LTα−/− mice. To evaluate the ability of LTα−/− mice to generate memory B cells, BCR−/− mice were primed with 108 SRBC and 2 mo later the splenocytes were used as a common source of anti-SRBC memory T cells. Wt mice and LTα−/− mice were primed with NP-KLH (50 μg) to provide the source of primed anti-NP B cells. Equal numbers of T and B cells (107 each) were transferred into sublethally irradiated (750 rads) wt recipients, and then these mice were challenged with 108 NP-SRBC. Five days (□) or 10 days (▪) later, sera were collected and the anti-NP IgG response was measured by ELISA. The results represent means ± SEM of three mice per group. One representative experiment of two is shown.

FIGURE 4.

Lack of memory B cells in the spleen of primed LTα−/− mice. To evaluate the ability of LTα−/− mice to generate memory B cells, BCR−/− mice were primed with 108 SRBC and 2 mo later the splenocytes were used as a common source of anti-SRBC memory T cells. Wt mice and LTα−/− mice were primed with NP-KLH (50 μg) to provide the source of primed anti-NP B cells. Equal numbers of T and B cells (107 each) were transferred into sublethally irradiated (750 rads) wt recipients, and then these mice were challenged with 108 NP-SRBC. Five days (□) or 10 days (▪) later, sera were collected and the anti-NP IgG response was measured by ELISA. The results represent means ± SEM of three mice per group. One representative experiment of two is shown.

Close modal

To address whether LTα−/− mice could generate memory T cells, wt mice and LTα−/− mice were primed i.p. with SRBC in PBS to provide a source of potential anti-SRBC memory T cells. Wt and LTα−/− mice were primed i.p. with NP-KLH in PBS to provide a source of potential anti-NP memory B cells. Primed T and B cells (107 each) purified from the SRBC-immunized and the NP-KLH-immunized mice were then transferred to irradiated wt recipients. Five days after i.p. challenge with NP-SRBC in PBS, a robust IgG anti-NP response was detected in mice that received primed B cells from wt mice, but not from LTα−/− mice (Fig. 5), confirming that the LTα−/− environment does not support the formation of memory B cells. Of interest, functional memory T cells were recovered from both wt and LTα−/− donors. Thus, the altered microenvironment of LTα−/− mice retains the ability to support maturation of T cells to effective memory function. The methods we have employed may not detect modest or partial impairments of memory T cell formation in LTα−/− mice. Such potential memory T cell defects might be revealed by adoptive transfer of smaller numbers of T cells, or by immunization with more limiting doses of Ag. Nevertheless, our data do show that the tissue requirements for the formation of memory B cells and memory T cells are substantially different.

FIGURE 5.

Robust formation of memory T cells in LTα−/− mice. Wt (+/+) and LTα−/− (−/−) mice were immunized i.p. with 108 SRBC in PBS, and 60 days later T cells were purified from their spleens. B cells were purified from wt mice or LTα−/− mice that had been primed i.p. with NP-KLH (50 μg) in PBS 60 days earlier. Equal numbers of primed T and B cells (107 each) were transferred to irradiated wt recipients, and at the same time the recipients were challenged by i.v. immunization with 108 NP-SRBC. Five days later, sera were collected and anti-NP IgG was measured by ELISA. The results represent means ± SEM of three mice per group. One representative experiment of two is shown.

FIGURE 5.

Robust formation of memory T cells in LTα−/− mice. Wt (+/+) and LTα−/− (−/−) mice were immunized i.p. with 108 SRBC in PBS, and 60 days later T cells were purified from their spleens. B cells were purified from wt mice or LTα−/− mice that had been primed i.p. with NP-KLH (50 μg) in PBS 60 days earlier. Equal numbers of primed T and B cells (107 each) were transferred to irradiated wt recipients, and at the same time the recipients were challenged by i.v. immunization with 108 NP-SRBC. Five days later, sera were collected and anti-NP IgG was measured by ELISA. The results represent means ± SEM of three mice per group. One representative experiment of two is shown.

Close modal

GC, with their prominent clusters of FDC and scattered Ag-specific T cells, are thought to provide primary venues for the formation of memory B cells (13, 14, 19, 22). The GC represent a dynamic microenvironment in which B cells differentiate first into rapidly proliferating PNA+ cells that then move toward the FDC clusters where further maturation takes place (13, 14, 23). The FDC are thought to play a central role in GC function, with Ag deposited on their surfaces primarily in the form of complement-coated immune complexes serving to select B cells with high affinity Ag receptors. In the course of this selection process, PNA+ GC cells are thought to be converted to the precursors of both IgG-producing B cells and long-lived memory B cells. To investigate further the nature of the disturbed B cell memory function in the LTα−/− mice, we examined the roles of LTα and the lymphoid tissue microenvironment in the formation and function of GC-like clusters of PNA+ cells. Wt or LTα−/− mice were sublethally irradiated, reconstituted with splenocytes from wt donors, and immediately immunized with an i.v. infusion of 108 SRBC. Ten days after transfer and immunization, sections of spleen were stained with PNA (Fig. 6). Both wt and LTα−/− recipients could support the development of clusters of PNA+ cells in response to Ag, although the sizes and number of the clusters generally appeared to be reduced in the LTα−/− recipients. Importantly, no clusters of PNA+ cells were detected in LTα−/− recipients that received splenocytes from LTα−/− donors (data not shown and 8). Thus, the development of the PNA+ cells was dependent on transferred LTα-expressing splenocytes. The PNA+ clusters that were produced by wt splenocytes 10 days after transfer into the disturbed microenvironment of LTα−/− recipients were not associated with detectable clusters of FDC (data not shown and 8) and were unable to support the development of a detectable Ag-specific serum IgG response (Fig. 2 and 8). In addition, the clusters of PNA+ cells that were found in LTα−/− recipients of wt splenocytes showed no association with apoptotic selection (Fig. 6). Wt SRBC-immunized mice that received wt splenocytes showed prevalent clusters of TUNEL+ cells associated with GC. No clusters of TUNEL+ cells were detected in the white pulp of SRBC-immunized LTα−/− mice that had been reconstituted with wt donor splenocytes. Apoptosis of GC cells is thought to represent deletion of B cells that express low affinity Abs. This deletion is thought to correlate with the selection of B cells with high affinity and their conversion into memory cells (14, 15, 19). Our data indicate that the LTα−/− microenvironment characterized by lack of FDC clusters can support the differentiation of Ag-stimulated B cells into PNA+ GC-like cells, but fails to support either affinity-based selection of these cells or their further differentiation into either IgG producing cells or into memory B cells.

FIGURE 6.

Lack of apoptotic activity in PNA+ clusters in LTα−/− mice reconstituted with wt splenocytes. Irradiated wt mice (left panels) and LTα−/− mice (right panels) were reconstituted with splenocytes from naive wt mice together with 108 SRBC. The recipients’ spleens were collected 10 days after transfer and immunization, and sections were stained with PNA (blue) and anti-IgD (brown) in the top panels and by the TUNEL procedure in the lower panels. TUNEL+ cells are brown on the background of methyl green counterstain.

FIGURE 6.

Lack of apoptotic activity in PNA+ clusters in LTα−/− mice reconstituted with wt splenocytes. Irradiated wt mice (left panels) and LTα−/− mice (right panels) were reconstituted with splenocytes from naive wt mice together with 108 SRBC. The recipients’ spleens were collected 10 days after transfer and immunization, and sections were stained with PNA (blue) and anti-IgD (brown) in the top panels and by the TUNEL procedure in the lower panels. TUNEL+ cells are brown on the background of methyl green counterstain.

Close modal

The data presented in this study demonstrate that the disturbed lymphoid tissue microenvironment that forms in LTα−/− mice is unable to support the generation of memory B cells from naive progenitors. This LTα−/− microenvironment is also unable to support the conversion of preformed wt memory B cells (produced in a wt environment) into Ag-producing cells that express a mature memory response. At least three characteristics of the normal spleen white pulp fail to form properly in LTα−/− mice (1, 2, 7, 21): 1) the segregated T cell-predominant periarteriolar lymphoid sheath (PALS), 2) the segregated B cell-predominant marginal sinus structure with MAdCAM-1+ (mucosal addressin cell adhesion molecule-1+) vascular endothelium and MOMA-1+ metallophilic macrophage components, and 3) primary B cell follicles with clusters of FDC. Cooperative interactions between T and B lymphocytes are required for the formation of functional GC and high affinity isotype-switched Ig responses. Cooperative B/T interactions have also been suggested to be required for the generation of memory B cells (13, 22, 24). It is possible that segregation of T and B cells into separate zones within the splenic white pulp facilitates properly regulated interactions of Ag-specific T and B cells during both primary and memory responses. LTα−/− mice lack normal T and B cell segregation (1, 8). This might be expected to impair effective collaboration between the two cell populations and explain the observed inability to form memory B cells during a primary response or to productively activate memory B cells in a memory response; however, several observations argue against a major role of disorganization of T and B cell zones in the impaired memory responses. First, 6 wk after lethally irradiated LTα−/− mice had been reconstituted with wt bone marrow, segregation of T and B cell compartments remained incomplete but FDC clusters were restored (8). Mice reconstituted in this fashion can generate GC and secondary IgG responses similar in magnitude to wt mice. Second, in preliminary studies, we detect successful generation of memory B cells in LTα−/− mice that have been reconstituted with wt bone marrow in this fashion (data not shown). Thus, normal T cell/B cell segregation appears not to be required for the memory response. Therefore, we speculate, that LTα-dependent disturbances in the primary B cell follicle structure are more likely to underlie the disturbances of the memory response.

Consistent with this hypothesis, the generation of memory T cells appears to require a microenvironment distinct from that required for the formation of memory B cells. Memory T cells can be formed effectively in LTα−/− mice, with GC and FDC appearing not to be required for the generation of memory T cells. FDC clusters, on the other hand, are thought to play critical roles in the formation of B cell follicles and in the presentation of Ag to activated B cells (13, 14). We suggest that the lack of FDC clusters in LTα−/− mice primarily underlies the failure to generate memory B cells and also to support the response of passively transferred memory B cells to rechallenge. Consistent with this hypothesis, our preliminary data suggest that LTα−/− mice fail to manifest affinity maturation following repeated challenge with hapten in the absence of adjuvant (data not shown). Generally, a lack of somatic mutation correlates tightly with failure to elaborate memory B cells (14, 25).

The generation of the B cell arm of the memory response requires the induction of memory B cells as well as their maintenance. The experiments described here cannot distinguish between failure of the initial differentiation of activated B cells into B cells of the memory phenotype and failure to sustain the survival of these cells after Ag is catabolized and cleared. Additional experiments in which exposure to Ag is sustained continuously will be required to discriminate between these potential mechanisms of the failure to generate transferrable memory B cells in the LTα−/− mice.

In addition, the failure of memory B cells induced in a wt environment to express a memory IgG response after adoptive transfer into sublethally irradiated LTα−/− recipients could be based on several potential mechanisms. Expression of a memory response by adoptive transferred cells is contingent first on their survival in the recipient mice. Although we know that similar numbers of donor B cells are present in the spleen of recipient wt and LTα−/− mice after adoptive transfer (data not shown), the methods used here cannot measure specifically the survival of memory cells. Assuming survival of memory B cells after adoptive transfer, the expression of the memory IgG response requires activation of these cells by Ag, and then their conversion into IgG secreting cells. Our observation that primed wt cells can support the Ag-dependent formation of clusters of PNA+ cells after adoptive transfer into LT−/− recipients suggests that Ag recognition does occur in this setting, and that at least one key defect in the LTα-deficient microenvironment may be in conversion from activated, proliferating B cells to Ab-secreting cells. Thus, the failure to express B cell memory in this environment may represent a general failure of maturation of proliferating, isotype-switched cells to Ab-secreting B cells and plasma cells.

GC contain significant numbers of Ag-specific T cells, and these cells appear to be essential for GC physiology, supporting somatic mutation and affinity maturation of GC B cells (12, 13, 22, 26). However, it has not been previously determined whether GC are required for the generation of T cell memory. Our data indicate that generation and subsequent activation of memory T cells and memory B cells require different environmental elements. In contrast to memory B cells, memory T cells can be generated in easily detectable numbers without the formation of GC and FDC (Fig. 5).

Our data support the emerging concept that members of the TNF ligand/receptor family play critical roles not only in the establishment of normal lymphoid tissue structure, but also in the cellular interactions that occur characteristically within these tissues. In addition to the LTβ receptor, several other members of the TNF receptor family, including OX40, the nerve growth factor receptor (NGFR), CD40, and type I TNF receptor (5, 27) are required for different steps of the memory IgG response. OX40 and CD40 on B cells interact with their ligands on the surfaces of T cells, with these interactions prominent during the generation of B cell memory responses and crucial for the development of high affinity IgG responses (5, 24). Signaling via the NGFR delivers autocrine signals supporting the survival of memory B cells (28). In these interactions, the B cell expresses the TNF receptor family member and receives signals primarily from T cells to support the Ab response. In contrast, B cells do not express the LTβR. Instead, B cells express LT ligands (LTα3 and LTα1β2). Via these ligands, B cells signal LTβR-bearing cells to support the development and maintenance of the lymphoid tissue structure that is required for mature B cell responses (11). The data presented here indicate that this LT-dependent microenvironment is required both for the generation of memory B cells and for their proper activation by recall Ag. These results underscore the central role of the NGFR/TNF receptor family in the generation of the memory response.

In summary, this study has demonstrated that LTα−/− mice show defective generation of memory B cells in response to T-dependent Ags. Furthermore, they cannot support LTα-expressing memory B cells to express memory responses to recall Ags. LTα-expressing B cells, 10 days after they have been transferred into LTα−/− mice, can be activated by Ag to form clusters of PNA+ cells; however, these cells do not form IgG-producing cells. Unlike T cells that can develop memory responses in LTα−/− mice, B cell memory requires LTα-dependent structures, with either clusters of FDC or structures closely linked to the expression of FDC being centrally required for the development of this response. Our studies clearly separate the requirements for the formation of memory T and memory B cells and establish lymphotoxin as a member of the growing group of TNF family ligands that are required for the maturation of the B cell response and the development of B cell memory.

1

This work was supported by Grants AI 01431 (Y.-X.F.) and AI 34580 (D.D.C.) from the National Institutes of Health. D.D.C. is an investigator of the Howard Hughes Medical Institute.

3

Address correspondence and reprint requests to Dr. David D. Chaplin, Division of Allergy and Immunology, Department of Internal Medicine, Washington University School of Medicine, 4566 Scott Avenue, Box 8122, St. Louis, MO 63110. E-mail address: chaplin@im.wustl.edu

4

Abbreviations used in this paper: LT, lymphotoxin; LTβR, LT-β receptor; FDC, follicular dendritic cell; GC, germinal center; KLH, keyhold limpet hemocyanin; wt, wild type; NP, nitrophenyl; PNA, peanut agglutinin; BCR, B cell Ag receptor, RU, relative units; AP, alkaline phosphatase.

1
De Togni, P., J. Goellner, N. H. Ruddle, P. R. Streeter, A. Fick, S. Mariathasan, S. C. Smith, R. Carlson, L. P. Shornick, J. Strauss-Schoenberger, et al
1994
. Abnormal development of peripheral lymphoid organs in mice deficient in lymphotoxin.
Science
264
:
703
2
Banks, T. A., B. T. Rouse, M. K. Kerley, P. J. Blair, V. L. Godfrey, N. A. Kuklin, D. M. Bouley, J. Thomas, S. Kanangat, M. L. Mucenski.
1995
. Lymphotoxin-α-deficient mice: effects on secondary lymphoid organ development and humoral immune responsiveness.
J. Immunol.
155
:
1685
3
Rennert, P. D., J. L. Browning, R. Mebius, F. Mackay, P. S. Hochman.
1996
. Surface lymphotoxin α/β complex is required for the development of peripheral lymphoid organs.
J. Exp. Med.
184
:
1999
4
Rennert, P. D., D. James, F. Mackay, J. L. Browning, P. S. Hochman.
1998
. Lymph node genesis is induced by signaling through the lymphotoxin β receptor.
Immunity
9
:
71
5
Ware, C. F., T. L. VanArsdale, P. D. Crowe, J. L. Browning.
1995
. The ligands and receptors of the lymphotoxin system.
Curr. Top. Microbiol. Immunol.
198
:
175
6
Beutler, B..
1995
. TNF, immunity and inflammatory disease: Lesson of the past decade.
J. Invest. Med.
43
:
227
7
Matsumoto, M., S. Mariathasan, M. H. Nahm, F. Baranyay, J. J. Peschon, D. D. Chaplin.
1996
. Role of lymphotoxin and type I TNF receptor in the formation of germinal centers.
Science
271
:
1289
8
Fu, Y.-X., H. Molina, M. Matsumoto, G. Huang, J. Min, D. D. Chaplin.
1997
. Lymphotoxin-α supports development of splenic follicular structure that is required for IgG responses.
J. Exp. Med.
185
:
2111
9
Koni, P. A., R. Sacca, P. Lawton, J. L. Browning, N. H. Ruddle, R. A. Flavell.
1997
. Distinct roles in lymphoid organogenesis for lymphotoxin α and β revealed in lymphotoxin β-deficient mice.
Immunity
6
:
491
10
Futterer, A., K. Mink, A. Luz, M. H. Kosco-Vilbois, K. Pfeffer.
1998
. The lymphotoxin β receptor controls organogenesis and affinity maturation in peripheral lymphoid tissues.
Immunity
9
:
59
11
Fu, Y.-X., G. Huang, Y. Wang, D. D. Chaplin.
1998
. B lymphocytes induce the formation of follicular dendritic cell clusters in a lymphotoxin-α-dependent fashion.
J. Exp. Med.
187
:
1009
12
Liu, Y. J..
1997
. Sites of B lymphocyte selection, activation, and tolerance in spleen.
J. Exp. Med.
186
:
625
13
MacLennan, I. C..
1994
. Germinal centers.
Annu. Rev. Immunol.
12
:
117
14
Rajewsky, K..
1996
. Clonal selection and learning in the antibody system.
Nature
381
:
751
15
Kelsoe, G..
1996
. Life and death in germinal centers (redux).
Immunity
4
:
107
16
Zheng, B., S. Han, G. Kelsoe.
1996
. T helper cells in murine germinal centers are antigen-specific emigrants that downregulate Thy-1.
J. Exp. Med.
184
:
1083
17
Zheng, B., W. Xue, G. Kelsoe.
1994
. Locus-specific somatic hypermutation in germinal centre T cells.
Nature
372
:
556
18
Zheng, B., S. Han, Q. Zhu, R. Goldsby, G. Kelsoe.
1996
. Alternative pathways for the selection of antigen-specific peripheral T cells.
Nature
384
:
263
19
Nossal, G. J..
1997
. B lymphocyte physiology: the beginning and the end.
Ciba Found. Symp.
204
:
220
20
Gavrieli, Y., Y. Sherman, S. A. Ben-Sasson.
1992
. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation.
J. Cell Biol.
119
:
493
21
Matsumoto, M., S. F. Lo, C. J. Carruthers, J. Min, S. Mariathasan, G. Huang, D. R. Plas, S. M. Martin, R. S. Geha, M. H. Nahm, D. D. Chaplin.
1996
. Affinity maturation without germinal centers in lymphotoxin-α (LTα) deficient mice.
Nature
382
:
462
22
Zinkernagel, R. M., M. F. Bachmann, T. M. Kundig, S. Oehen, H. Pirchet, H. Hengartner.
1996
. On immunological memory.
Annu. Rev. Immunol.
14
:
333
23
Tew, J. G., J. H. Wu, D. H. Qin, S. Helm, G. F. Burton, A. K. Szakal.
1997
. Follicular dendritic cells and presentation of antigen and costimulatory signals to B cells.
Immunol. Rev.
156
:
39
24
van Essen, D., H. Kikutani, D. Gray.
1995
. CD40 ligand-transduced co-stimulation of T cells in the development of helper function.
Nature
378
:
620
25
Weiss, U., K. Rajewsky.
1990
. The repertoire of somatic antibody mutants accumulating in the memory compartment after primary immunization is restricted through affinity maturation and mirrors that expressed in the secondary response.
J. Exp. Med.
172
:
1681
26
Ahmed, R., D. Gray.
1996
. Immunological memory and protective immunity: understanding their relation.
Science
272
:
54
27
Matsumoto, M., Y.-X. Fu, H. Molina, D. D. Chaplin.
1997
. Lymphotoxin-α-deficient and TNF receptor-I-deficient mice define developmental and functional characteristics of germinal centers.
Immunol. Rev.
156
:
137
28
Torcia, M., L. Bracci-Laudiero, M. Lucibello, L. Nencioni, D. Labardi, A. Rubartelli, F. Cozzolino, L. Aloe, E. Garaci.
1996
. Nerve growth factor is an autocrine survival factor for memory B lymphocytes.
Cell
85
:
345