Tinigible Body Macrophages in Regulation of Germinal Center Reactions

Tingible body macrophages (TBM), long thought simply as scavengers of apoptotic lymphocytes, are located in the unique microenvironment of germinal centers in close proximity to antigen-retaining follicular dendritic cells (FDC). Observations that TBM endocytose FDC-iccosomal (immune-complex coated bodies) antigen suggested that TBM might present this antigen and help regulate the germinal center reaction. To test for antigen presentation, the ovalbumin (OVA)-specific TH hybridoma, 3DO-54.8, which produces IL-2 on receiving effective presentation of OVA, were used as responders to OVA-bearing TBM. Results showed that OVA-bearing TBM failed to induce IL-2 production. Furthermore, addition of TBM to IL- 2-inducing positive controls (B cells) not only failed to augment IL-2 production, but rather TBM significantly (55-90%) reduced B-cell induction of IL-2. We found that TBM were rich in prostaglandin by comparison with other nongerminal center lymph node macrophages and that addition of indomethacin to the cultures reversed the inhibitory effect of TBM. Depletion of TBM from enriched preparations, prior to addition to positive control cultures, also abrogated the inhibitory effect on IL-2 production. These data support the concept that TBM, within the unique microenvironment of germinal centers, may be specialized to downregulate the germinal center reaction.


de in germina
centers of secondary *Corresponding author.lymphoid tissues (Flemming, 1885).TBM contain many phagocytized, apoptotic cells (referred to as tingible bodies) (Flemming, 1885) in various states of degradation (Swartzendruber and Congdon, 1963).Unique to this subset of macrophages is the germinal center microenvironment, characterized functionally by long-term antigen retention on follicular dendritic cells, antigen presentation by B cells to T-helper cells, Present address: Department of Anatomy and Cell Biology, ECU School of Medicine, Brody Building, 7N-94, Greenville, NC 27858. 105TBM-enriched cells were cultured for 24 hr with 3 10 3DO-54.8Tl cells.Results shown were always greater than control cultures of 3DO cells or TBM alone.bl.0 )< 10 TBM-enriched cells and 50/xg OVA were cultured for 24 hr with 3 10 3DO-54.8Tni cells.c3 10 CTL cells were cultured with either 50/xl IL-2 containing Con A supernatant or 50/xM human rIL-2.aData represent 3H-thymidine incorporation by 3 104 IL-2-dependent CTLL cells and are expressed as mean cpm +_ standard errors of triplicate cultures.a high rate of somatic mutations, aff nity maturation, induction of antibody-forming cells and memory Bcell development (Coico et al., 1983; Szakal et al.,  1989; Tew et al., 1990).Although the scavenging function of TBM in germinal centers is obvious, it has also been suggested that TBM may be important in initiating the germinal center reaction (Kamperdijk et  al., 1978, 1982).

Regarding the germinal center reaction, it has been shown that germinal center B cells are extremely effective antigen-presenting cells.Three to eight days after a booster immunization, germinal center B cells obtain antigen from FDCs in the form of immune complex coated bodibs (iccosomes) and are capable o processing and presenting the iccosomal antigen to T cells in an MHC-restricted manner (Kosco et al.,  1988; Szakal et al., 1988).Typically, macrophages phagocytose antigens, process them, express them on class II molecules, and present the processed antigen to T-helper cells.Observations that TBM, like germinal center B cells, also endocytosed iccosomes (Szakal et al., 1988) and that many TBM are class II positive (Smith et al., 1988), suggest that TBM might also be able to present the FDC-derived antigen to T cells (Smith et al., 1988).

The objectives of the present study was to deter- mine if TBM can indeed present antigen and if antigen presentation by TBM may be implicated in the regulation of the germinal center reaction.The results showed that TBM-enriched preparations were not good antigen pres nters but were highly effective inhibitors of T-cell lymphokine secretion induced by B cells.Selective depletion of TBM from the enriched preparations and/or addition of the prostaglandin synthesis inhibitor indomethacin to the cultures restored lymphokine activity.Histochemical observa- tions confirmed that TBM are a rich source of prostaglandins.Thus, these data are consistent with the concept that the microenvironment of germinal centers favors an inhibitory role for TBM in the regulation of the germinal center reaction via a prostaglandin-mediated mechanism.


RESULTS


Antigen Presentation

Germinal center B cells obtain antigen from FDC for processing and presentation to T cells during a period between 3 and 8 days after booster immunization.Typically, numerous germinal center B cells (20%) were present in the TBM prepar

th TBM and germinal c
nter B cells take up iccosomes (Kosco  et al., 1988; Szakal et al., 1988).Therefore, we   reasoned that the addition of TBM-enriched cells from an OVA-immune animal to the OVA-specific Tcell hybridoma 3DO-54.8should result in antigen presentation and IL-2 production.As shown in Table I, there was no evidence of presentation of in vivo obtained antigen by the TBM preparation.When antigen was added in vitro (TBM + OVA), to be certain that antigen was not limiting, a significant (p < 0.01) but very modest level of IL-2 production was detected in two experiments (Table I, Experiments 1 and 4).In both cases, the production of IL-2 was less 55.0 60,000 3DO-54.8Tr cells (3 104) were cultured for 24 hr with:

aGerminal center B cells (106) or TA3 B-cell hybridoma'/(104) and 50 #g OVA (positive control cultures).bFootnote a and 105 TBM-enriched cells.c3 10 CTLL cells were cultured with either 50/xl IL-2 containing Con A supernatant or 50/zM human rIL-2.

dData represent 3H-thymidine incorporation by 3 10 IL-2-dependent CTLL cells and are expressed as mean cpm _+ standard errors of triplicate cultures.In Experiments to 3, the values for 3DO cells or TBM cultured alone were always less than 500.In Experiment 4, the values for cultures of 3DO alone and TBM alone were 1532 67 nd 1087 53, respectively.96,400 +_ 5,700 42,700 6,350   40  66,300 10,300 69 3DO-54.8T1 cells (3 104) were cultured for 24 hr with:

aGerminal center B cell

(106) or TA3 B-cell hybridoma'/(104) and 50/zg OVA.bFootnote a and
105 TBM-enriched cells.Values for 3DO cells and TBM alone were always less than 500 cpm.CFootnote b and 100/zM indomethacin.dData represent 3H-thymidine incorporation by 3 10 IL-2-dependent CTLL cells and are expressed as mean cpm _+ standard errors of triplicate cultures.than 10% of the response attained in the control when a near optimal level of IL-2 was added.

To determine if TBM might suppress antigen presentation, TBM-enriched cells were cocultured with 3DO cells and B cells in the presence of OVA.

The results from four experiments are shown in Table II.A 55-90% suppression of the IL-2 response was observed when TBM-enriched populations were added.


Effect of Indomethacin on TBM Suppression of Antigen Presentation

Macrophages are a major source of prostaglandins and generally the larger most phagocytically active macrophages are the most prolific prostaglandin producers (Lee and Berry, 1977; L

and Wong,   1982; Guid
s et al., 1987).The striking phagocytic activity of TBM and their large size prompted us to reason that TBM might suppress through a prosta- glandin-dependent mechanism.To begin testing this, the prostaglandin synthesis inhibitor indomethacin was added to cultures containing TBM-enriched cells, B cells, and 3DO cells in the presence of OVA.As

shown in Table III, the addition of 100 #M indom- ethacin relieved most of the suppression of IL-2 production in TBM-containing cultures.Controls showed that indomethacin did not inherently augment the IL-2 response in positive control cultures.Further- more, indomethacin did not inherently increase the proliferation of CTLL over background levels (data not shown).


TBM and Prostaglandins

PGE2 is made by a membrane-bound prostaglandin synthetase.We reasoned that antibody specific for Depending on the stage of maturation, state of activation, and source, macrophages have been reported to either enhance or suppress immune responses (Stenson and Parker, 1980).Classically, macrophages are believed to be important accessory cells for antigen presentation (Rosenthal and Shevach,  1973a, 1973b; Unanue, 1981).A requirement for macrophages to function as accessory cells is the expression of MHC products.In contra t, in many cases, it has been shown that macrophages do not function as accessory cells but instead suppress immune reactions (Parkhouse and Dutton, 1966).In a series of publications by Lee and colleagues, it was s ggested that the functional heterogeneity expressed by macrophages in the modulation of immune responses can be differentiated according to macro- phage size as measured by velocity sedimentation rates (Lee and Berry, 1977; Lee and Wong, 1982;   Guido et al., 1987).Antigen-presenting activity was found to be characteristic of small macrophages, whereas large macrophages tended to be immuno- suppressive.Furthermore, large macrophages that were induced to express class II molecules also exhibited suppressive effects on immune responses.Thus, the large size of TBM, even though they express class II (Smith et al., 1988) is compatible with the hypothesis that TBM may have a suppressive function to downregulate or limit the germinal center reaction.

Prostaglandins appear to be responsible for much of the immunosuppression mediated by macrophages.The suppressive effects of prostaglandins on immune responses are diverse and have been reviewed (Stenson and Parker, 1980; Goodwin and Ceuppens, 1983).

The binding of immune complexes by macrophages stimulates phagocytosis, which in turn stimulates prostaglandin synthesis with the major product being PGE2 with smaller amounts of PGF1 (Bonney et al.,   1979).This immune-complex stimulation may help explain the high level of prostaglandin in TBM since they may encounter immune complexes directly on the processes of FDC and in the form of iccosomes (Szakal et al., 1988).The data in this study show that indomethacin, an inhibitor of prostaglandin synthesis, abrogates the suppressive effect of the TBM prepara- tion on B-cell antigen presentation.Controls indicated that indomethacin did not inherently augment the IL-2 response in the assay system (data not shown).A known suppressive effect of prostaglandins on immune responses is the inhibition of lymphokine synthesis by TH cells.Whereas most of the lympho- cytes in germinal centers are B cells, a small number of T cells are present and most of these T cells are of the helper phenotype (Rouse et al., 1982a, 1982b).Interestingly, a close association between TBM and germinal center T cells has been observed (Nieu- wenhuis and Opstelten, 1984, and unpublished).

Germinal centers are sites of memory B-cell production (Coi o et al., 1983) and production of memory B cells involves somatic mutation and maintenance of B cells that express high-affinity receptors for the stimulating antigen (Wysocki et al.,   1986).It has been suggested that B cells with low- affinity antigen receptors cannot compete for antigen persisting on FDC and are clonally deleted (Wysocki  et al., 1986).The tingible bodies in TBM may represent phagocytized low-affinity B cells that have been condemned to undergo apoptosis (Manser et al.,  1987; Schad and Phipps, 1988).Consistent with this hypothesis, Odartchenko et al. (1966) and Fliedner  (1967) have reported that the majority of the tingible bodies located in TBM represent cells that have died either during late G2

mitotic
vent.

It has also been suggested that TBM play an active role in initiating the germinal center reaction (Kamperdijk et al., 1978, 1982).This suggestion was based on the observations that TBM initially appeared at the onset of germinal center developm

t and that peak num
ers of TBM were seen at peak germinal center development.However, in recent studies, it has been shown that TBM are not absolutely essential for germinal center development.Germinal centers can develop in old mice in the absence of TBM (Smith et  al., 1990).The data shown in this report are consistent with a regulatory role for TBM and it appears that TBM are more likely to down regulate than to stimulate the germinal center reaction.


MATERIALS AND METHODS


Animals

Ten-to sixteen-week-old female Balb/C (H-2d) mice were purchased from Charles River Laboratories (Wilmington, MA).The animals were housed under pathogen-free conditions in standard cages equipped with filter tops and allowed food and water ad libitum.


Cell Lines and mAb

The T 1 cell hybridoma 3DO-54.8 was a gift from J.

Kappler and P. Marrack (National Jewish Hospital, Denver).Cultures of 3DO-54.8 cells (3DO) produce IL-2 when stimulated by the antigen ovalbumin (OVA) recognized in the context of the MHC molecule I-Ad.The IL-2-dependent cytotoxic lym- phocyte line CTLL used to assay IL-2 was provided by P. Allen and E. Unanue (Washington University, St. Louis).The B-cell hybridoma TA3 (I-ATM) was a generous gift from Laurie Glimscher.Supernatants from the hybridoma M3/38 (obtained from Hybritech, San Diego) was used as the source of anti-Mac-2 antibody.Anti-Thy-1 antibody was obtained from Pharming n (San Diego).Recombin nt human IL-2 was obtained from Southern Biotech.


Immunizations Active immunization.

Germinal center B cells were obtained from ovalbu- min (OVA; #A5503, Sigma Chemical, St. Louis)- immune mice.Mice were injected with 0.5 mg of OVA emulsified with complete Freunds adjuvant (CFA, Difco Laboratories, Detroit) behind the neck followed by a second 0.5-mg injection of OVA in CFA 14 days later.Seven days prior to germinal center B-cell isolation, mice received challenge injections of 0.5/g OVA in all footpads.Anti-OVA sera was prepared in rabbits immunized subcuta- neously with 1.0 mg OVA emulsified in CFA followed by a booster immunization without CFA.Rabbits were bled via ear veins 7 days after booster immunization.


Passive immunization.

Pha

cytosis of Freunds adjuvant by TBM and other
macrophages can affect their densities and can cause erroneous isolation and the identification of non-TBM macrophages as TBM.Therefore, to obtain a popula- tion of TBM not contaminated by Freunds-adjuvant- containing macrophages, mice were passively immunized by intraperitoneal injections of 0.5 ml of rabbit anti-OVA.Eighteen to twenty-four hours later, challenge injections of 8-10 /g of OVA diluted in 0.9% phosphate-buffered saline (PBS), pH 7.4, were injected into all four footpads.By using this immun- ization protocol, germinal center development and TBM were first evident 3 days after the antigenic challenge.Optimal numbers of TBM were present 8-10 days after challenge (Smith et al., 1990; Tew et  al., 1990).

Isolation and Enrichment of TBM and B Cells Techniques for obtaining enriched populations o TBM (Smith et al., 1988) and germinal center B cells (Kosco et al., 1988) were previously described in detail.Briefly, 8 to 10 days after antigen challenge of OVA immune mice, lymph nodes were collected, enzymatically dissociated, and enriched on a discon- tinuous Percoll (Pharmacia, Piscataway, NJ) gradient (densities 1.042 and 1.064).Cells at the 1.064 interface were collected, washed, and incubated in tissue-culture dishes (Costar #3100, Cambridge, MA) at 37C for 45 rain to remove strongly adherent cells, which were primarily typical macrophages.Weakly adherent cells, including TBM, were then flushed off the culture dishes using a Pasteur pipette and counted.This population consisted of 7-10% TBM, 50-60% B cells (including abo

20% PNA hi GC B cells) and 30-35
T cells.

Enriched populations of germinal center B cells were obtained as previously described (Kosco et al.,  1988).Briefly, lymph nodes from actively immunized mice were collected at 3 and 5 days after challenge with OVA, enzymatically dissociated, and enriched on a continuous Percoll gradient.Further enrichment was achieved by panning for germinal center B cells on peanut agglutinin (PNA; Sigma)-coated tissue- culture dishes (Costar #3100).Unbound cells were removed by gentle swirling and PNA-positive germi- nal center B cells were then eluted from the dishes by adding 0.2 M d-galactose (Sigma) for 1 hr before vigorous flushing with a pipette.By using this technique, 70-80% enriched populations of highly PNA germinal center B cells were commonly attained.


Depletion of TBM Using Dynabeads

It is known that TBM are the only Mac-2-bearing cells in lymph nodes (Flotte et al., 1983; Smith et al.,  1990), and, in addition, TBM also express Thy-1 (Smith et al., 1988).Consequently, TBM can be selectively removed from TBM-enriched cell populat

ns using anti-Mac-2 and anti-Thy-1 o
immuno- magnetic beads (Dynal, Great Neck, NY).The beads were purchased precoated with Fc-specific sheep anti- rat IgG.The beads were washed five times, incubated with a cocktail of anti-Mac-2 (1:4 dilution) and mouse anti-rat Thy-1 (1:20) for 1 hr at 4C with intermittent agitation, and gently washed three times with HBSS.The beads were then incubated with the Mac-2/Thy-1-treated TBM-enricheff preparation at a bead-to-cell ratio of 40:1 for 30 min on ice with gentle intermittent agitation.Following incubation, the Dynabead roset- ted TBM were removed using the Dynal magnetic chamber.The remaining cells were collected with a Pasteur pipette, washed, and adjusted to 2 105 cells/ 50/zl in complete media.This preparation was used as the TBM-depleted population.

IL-2 Assay for Antigen Presentation Antigen presentation was assessed using an assay system similar to the one originally described by Shimonkevitz et al. (1983).In this system, the OVA- specific Try-cell hybridoma 3DO-54.8(3DO) will produce the lymphokine IL-2 when cocultured with OVA and a source of antigen-presenting cells.The production of IL-2 can then be quantitated by transferring aliquots of media from the cocultured cells to the IL-2-dependent T-cell line CTLL.The presence of IL-2 in the media stimulates the prolifera- tion of CTLL cells, which can be measured by 3Hthymidi e incorporation.Ultimately, then, the proliferation of CTLL serves as an index for antigen presentation.From 5 10 4 to 10 6 antigenpresenting cells (germinal center B cells or TA3 cells) were cocultured with 3.0 10 4 3DO cells in 96-well flat-bottom plates (Falcon #3596) at 37C, in 5% CO2 atmosphere for 24 hr in the presence of 50 /zg of OVA.RPMI 1640 (GIBCO, Grand Island, NY) supplemented with 10% pyrogen-free fetal calf serum, 100 /zg/ml gentamicin (GIBCO), 0.2 mM glutamine (GIBCO), and 10 mM Hepes buffer (GIBCO) was used as the culture media.Positive controls for antigen presentation were either (1) cocultures of 5 105 to 1.0 10 6 germinal center B cells with 5.0 10 4 3DO cells in the presence of 50 /zg of OVA or (2) 5.0 104 TA3 cells with 3.0 104 3DO cells in the presence of 50/zg OVA.

To assess the production of IL-2 by 3DO cells,   150-/zl aliquots of coculture media were transferred to 3 10 4 CTLL cells in 96-well U-bottom plates (Costar #3799) and cultured for 24 hr at 37C, at 5% CO2 atmosphere.At 16-20 hr, each well was pulsed with 1.0 /xCi 3H-thymidine (New England Nuclear, Boston) and harvested (PHD Harvester, Cambridge Technical) at 24 hr onto glass filter str

s (Costar, #240-1).Proliferation of CTLL cells induced by IL-2  in t
e coculture media was then measured as thymidine incorporation using a Packard 2200CA scintilla- tion counter.All culturing were performed in tripli- cate and the data were expressed as the mean 3H-thymidine uptake.Cultures of CTLL cells alone and supernatant from 3DO cells alone in the presence or absence of OVA were used as controls and were consistently negative (mean cpm _+ standard error 76).As a control for T-cell contamination in the germinal center B-cell preparation, germinal center B cells were cultured alone in the presence of antigen.The germinal center B-cell preparation appeared to contain some T cells, as indicated by slight but inconsequential IL-2 production when cultured in the presence of antigen.


Influence of TBM on Antigen Presentation by Germinal Center B Cells

To determine if TBM have an affect on the capacity of B cells to present antigen, 1-2 105 BMenriched cells (containing 14,000-16,000 TBM) were cocultured with B cells (either 5.0 105 to 1.0 106 germinal center B cells or 5.0 10 4 TA3 cells) and 3.0 X 10 4 3DO cells in the presence of 50/xg OVA.As in the antigen-presenting assay, in some cultures, TBM were removed from the enriched preparations using Mac-2-coated Dynabeads prior to culturing.In some experiments, 50 /xl of 100 /xM indomethacin (Sigma, St. Louis), a prostaglandin inhibitor, was added.Prostaglandins are known to inhibit IL-2 synthesis.As a control, to ensure that indomethacin did not inherently affect the assay, indomethacin was also added to cocultures of germinal center B cells, antigen, and 3DO cells in the absence of TBMenriched cells.lmmunocytochemistry Reactivity for pros