|
|
||||||||
Department of Microbiology and Immunology and Sanders-Brown Research Center on Aging, University of Kentucky, Lexington, KY 40536
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
TI-2 Ags stimulate B cells without the requirement for MHC class
II-restricted presentation to T cells and subsequent cognate T cell-B
cell interactions (2, 10, 11). Nevertheless, the responses of B cells
to this class of Ags is dependent on cytokines secreted by T cells and
accessory cells (2, 12, 13, 14, 15, 16). When TI-2 Ags are added to cultures of
adult splenic cells, the Ag-specific B cells readily respond by
secreting Abs to the immunizing Ag. However, when purified adult B
cells are used instead of total splenocytes, the B cells respond to
TI-2 Ags only when cultures are provided with certain cytokines. In
agreement with the multiple cellular sources of cytokines, purified
splenic B cells from adult mice overcome their unresponsiveness
in vitro to TI-2 Ags when stimulated with cytokines such as IL-1, -2,
-4, -5, and -6 and IFN-
(13, 14, 15, 16). These cytokines are required for a
number of steps in the B cell response to Ags, including clonal
expansion, differentiation into Ab-secreting cells, the level of Ab
secretion, and isotype switching (2, 3, 17, 18, 19, 20, 21, 22).
Recent studies have emphasized the predominance of the Th2 type of cytokine profile in neonatal immune responses. Thus, both in the neonatal transplantation tolerance model and in the immune responses to Cas-Br-M murine leukemia virus, Th2 cytokines are primarily produced in neonates (23). Despite this, neonates are deficient in TI-2 Ag responses, which have been shown to be enhanced by Th2 cytokines such as IL-4, -5, and -6 (13, 14, 15, 24, 25, 26), suggesting that another cell type or cytokine critical for activation of such Th2 cells or TI-2 Ag-specific B cells is deficient in neonates. Since numerous previous studies have shown TI-2 Ag responses to be accessory cell dependent (26, 27, 28, 29), it is conceivable that neonates have a deficiency in such a cell type or in cytokines produced by these cells. In this context, similar to neonatal mice, peripheral lymph node cells from normal healthy adult mice do not respond either to the haptenated TI-2 Ag, TNP-Ficoll, or to the pneumococcal polysaccharide-containing Pnu-Imune vaccine (26, 29). Moreover, aged mice and humans are hyporesponsive to TI-2 Ags, increasing their susceptibility to infections with capsulated bacteria. Extensive analyses of these models in our laboratory showed that the unresponsiveness of both young adult murine lymph node cells and aged mouse spleen cells to TI-2 Ags is due to a defect in an accessory cell population, which can be compensated by the addition of young mouse splenic accessory cells or IL-1, an accessory cell-derived cytokine (22, 25, 26, 29). In vivo, this defect is overcome by the injection of accessory cell-activating agents, such as monophosphoryl lipid A (30). Even Xid B cells can be stimulated to respond to TNP-Ficoll when cultures are provided with synovial fluid from arthritis patients (31), which is an enriched source of a number of cytokines, including the accessory cell-derived IL-1 and IL-6 (32).
Despite the immune-compromised status of neonates with regard to capsular bacterial infections, neonatal B cells are not completely unresponsive to polysaccharides. Mosier (33) established that neonatal B cells do indeed proliferate in vivo in response to TNP-Ficoll and could subsequently be induced to secrete TNP-specific Abs upon stimulation in vitro with LPS. Furthermore, certain polysaccharide Ags upon conjugation to a protein carrier, become immunogenic in neonates, a strategy that has been used very effectively in designing vaccines against certain diseases caused by encapsulated bacteria, such as H. influenzae (5, 34, 35). Both these studies imply that TI-2 Ag-specific B cells are present in neonates but are unable to respond to these Ags. Since adult B cells require cytokines to respond to polysaccharide Ags, the lack of responsiveness in neonates to these Ags could be due to a cytokine deficit.
In this study we have measured the Ab-secreting cell responses to TNP-Ficoll, a prototype polysaccharide TI-2 Ag, after supplementing neonatal splenocyte cultures with a panel of accessory and T cell-derived cytokines shown previously to modulate B cell function. Our study demonstrates that neonatal B cells, when provided with the right mixture of cytokines, such as the accessory cell-derived IL-1 and IL-6, respond to polysaccharide Ags in vitro by secreting Abs of appropriate avidity against the immunizing Ag. We also show that although adult splenocyte cultures respond to TNP-Ficoll without the addition of exogenous cytokines, once purified, the adult B cell Ab response to TNP-Ficoll is modulated by cytokines in a remarkably similar fashion to that of purified neonatal B cells.
| Materials and Methods |
|---|
|
|
|---|
Pathogen-free, 2- to 3-mo-old BALB/c mice were purchased from the National Institute of Aging (Bethesda, MD), while 6-wk-old BALB/c C57BL/6 and mice were purchased from the National Cancer Institute (Frederick, MD). All mice were housed in microisolator cages in our American Association for Laboratory Animal Accreditation and Certification-approved rodent facility. Neonatal BALB/c (1012 days old, unless otherwise stated) were bred in our rodent facility.
Cytokines and Ags
TNP-Ficoll was prepared from the aminoethyl carbamylmethyl derivative of Ficoll as described previously (36) and was a gift from John Inman, National Institutes of Health (Bethesda, MD). The F(ab')2 goat anti-IgM heavy chain was purchased from Organon Technika (Aurora, OH). To obtain D10 T cell supernatant, 10 x 106 D10.G4.1 cells (C. A. Janeway, Yale University, New Haven, CT) were cultured with irradiated splenocytes (2000 rad) from one C57BL/6 mouse in 10 ml of RPMI 1640 nutrient medium (Life Technologies, Grand Island, NY) supplemented with 10% FBS (HyClone, Logan, UT) at 37°C in 5% CO2. On the fourth day, D10 supernatant was collected and stored at -20°C until needed. For the IL-1 bioassay, the activated D10 cells were depleted of dead cells and then cultured in RPMI in the presence of supernatant (10%, v/v) derived from Con A-stimulated rat splenocytes. The cells were ready for use 7 days after antigenic stimulation. Human rIL-1ß, which was functionally interchangeable with murine IL-1 (37), was obtained through the Biologic Response Modifiers Program of the National Cancer Institute (Bethesda, MD). Mouse rIL-4 and -6 were purchased from Genzyme (Cambridge, MA). All three cytokines were certified to be free of endotoxin, and the tissue culture components were routinely tested for endotoxin using the Limulus amebocyte lysate kit (BioWhittaker, Walkersville, MD). IL-10 was a gift from Dr. Jay Fine (Schering-Plough Research Institute, Kenilworth, NJ).
Isolation of splenocytes and B cells
Lymphocytes were isolated from spleens of unimmunized mice and were processed according to protocols described previously (26, 29, 30). Briefly, adhering fat and connective tissue were removed from spleens, and lymphocytes were dispersed by pressing the spleens against the bottom of a tissue culture dish containing HBSS (Life Technologies) using the flat surface of a syringe plunger. Cells were left undisturbed for 1 min to let the splenic debris settle and then were collected and washed by centrifuging for 5 min at 400 x g. RBCs were depleted with Geys solution according to a previously described protocol (38).
B cells were enriched by panning with goat anti-IgM immobilized on tissue culture-treated dishes using a modification of the protocol reported by Mage (39). Briefly, plastic adherent cells were removed by incubating 120 x 106 splenocytes in 13 ml of Iscoves modified Dulbeccos medium supplemented with 5% newborn bovine serum on a 150 x 25-mm polystyrene tissue culture dish (Corning Glass Works, Corning, NY) for 1.5 h at 37°C. The nonadherent cells were then transferred to a dish that had previously been precoated with a mixture of 10 µg/ml anti-IgM and 90 µg/ml goat IgG (Sigma, St. Louis, MO) and were incubated at 4°C. After 70 min, the unbound cells were removed, and the dish was gently rinsed with HBSS. The bound cells were dislodged by vigorous vortexing of the dish, and the cells were recovered after repeated flushing with HBSS. The purity of the cells was determined by staining aliquots with anti-IgM-FITC (Southern Biotechnology Associates, Birmingham, AL), anti-B220-phycoerythrin (clone RA3 6B2; Sigma), anti-Thy1.2-FITC, and anti-Mac-1-phycoerythrin (PharMingen, San Diego, CA). The cells were also occasionally stained with biotinylated Abs to CD5 (Becton Dickinson, San Jose, CA) and heat-stable Ag (HSA; CD24) (PharMingen). The cells were then fixed with 1% paraformaldehyde (Sigma) and analyzed on a FACSCalibur flow cytometer (Becton Dickinson).
Cell culture
Lymphocytes were cultured in IF-12 tissue culture nutrient medium, that had previously been shown to support in vitro responses from B cells (29, 40). It consisted of a 1/1 mixture of Iscoves modified Dulbeccos medium and Hams F-12 nutrient mixture (Life Technologies) supplemented with 10% FBS (Atlanta Biologics, Norcross, GA), insulin (Life Technologies), L-glutamine, 50 µM 2-ME (Sigma), transferrin (Sigma), progesterone (Pfaltz & Bauer), and gentamicin (Fujisawa, Deerfield, IL). For in vitro immunization, cultures contained 2.0 x 106 total splenocytes or 0.8 x 106 purified B cells/0.5 ml of IF-12 in 48-well tissue culture plates (Costar, Cambridge, MA). For proliferation assays, cultures contained 2.0 x 105 cells/0.2 ml in 96-well tissue culture plates (Costar). Cultures were set up in triplicate and placed in a 5% CO2 humidified incubator (Forma Scientific, Marietta, OH) maintained at 37°C.
The radioactive based assay to determine cell proliferation was performed as described previously (41, 42). Briefly, cultures were pulsed with 1 µCi of [3H]thymidine (New England Nuclear, Boston, MA) 2 days or, in the case of bioassays, 3 days after initiation of cultures, and 4 h later the cells were harvested onto filter mats (Skatron, Sterling, VA) using a cell harvester (Packard, Meriden, CT). The levels of tritium trapped in the mats was measured with a Matrix 96 beta particle counter (Packard).
Hemolytic plaque-forming cell (PFC) assay
The number of IgM anti-TNP-secreting cells was determined on the fourth day of culture using a glass slide version of the technique of localized hemolysis in a gel (43). Briefly, a 1-ml packed cell volume of SRBC (Colorado Serum, Denver, CO) was coupled with 2,4,6-trinitrobenzene sulfonic acid (Eastman Kodak, Rochester, NY) following published protocols (44). The cultured splenocytes were washed with HBSS and then mixed with 50 µl of 13.5% (v/v) TNP-coupled SRBC, 200 µl of 2x basal Eagles medium (Life Technologies), and 200 µl of 1.6% agarose (FMC Bioproducts, Rockland, ME) and poured onto a glass microscope slide (Goldseal, Portsmouth, NH). The slides were incubated for 1 h at 37°C, and the plaques were developed during an additional 1-h incubation at 37°C with guinea pig complement (Pel-Freez Biologics, Rogers, AR). The plaques were viewed under a low power microscope. Ag-specific Ab-forming cells (AFC) were calculated by taking the number of plaques for any given treatment and subtracting the number of plaques obtained in the corresponding cultures incubated without Ag. Results are provided as the arithmetic mean ± SE, and the statistical significance of different treatments was evaluated by Students t test (two tailed, unequal variance).
A comparative Ab avidity assay was employed to determine the avidity of Abs produced in neonates and adult culture total splenocyte cultures. Briefly, the PFC assay was performed in the presence of increasing concentrations of the monovalent TNP-lysine (Research Organics, Cleveland, OH) or the multivalent TNP-Ficoll, which resulted in reduced plaque formation. The results were plotted as the concentration of free hapten on the abscissa vs the percentage of maximum response on the ordinate axis. A measure of the relative avidity of Abs secreted due to the different treatments was obtained by comparing the amounts of free hapten needed to obtain 50% inhibition, which was termed average avidity (45).
Apoptosis assay
The level of apoptosis was determined using the protocol of Reid et al. (46). Briefly, cells were stained with the DNA-binding viable dye, Hoechst 33342 (Molecular Probes, Eugene, OR), for 30 min at 37°C. The resulting level of fluorescence was an indicator of the amount of DNA per cell, thus determining the position of each cell in the cell cycle at the time of harvest (47). The cells were washed and stained with the naturally fluorescent lipophilic merocyanine 540 (Molecular Probes) for 20 min at room temperature. This dye selectively stained disordered or loosely packed membrane bilayers, and its level of fluorescence upon binding was a measure of the degree of packing of membrane lipids (48), which was an early characteristic of apoptosis (49). The cells were finally stained with anti-B220-FITC (Sigma), and cells positive for B220 and Hoechst were analyzed using a FACStar (Becton Dickinson) as described previously (50).
Determination of cytokine levels
Splenocytes or B cells were cultured for 1 or 2 days, and then the supernatants from replicate cultures were pooled and immediately frozen at -20°C. The bioassays were performed according to recently published methods (51, 52). Briefly, a twofold serial dilution of a culture supernatant or cytokine standard was conducted in a 96-well tissue culture plate in triplicate. For the IL-1 assay, each dilution was incubated with 4 x 104 activated D10 cells (see Cytokines and Ags) and 1.25 µg/ml Con A. For the IL-6-dependent assay, each dilution was incubated with 2 x 103 B9 cells, and proliferation was determined 3 days later. The amount of cytokine in the supernatant was calculated by comparing the dilution of the culture supernatant required to achieve 50% of the maximum proliferation with that required for the cytokine standard (53). IL-1 was also estimated by ELISA at the Cytokine Core Laboratory (Baltimore, MD). The minimum detection limits were 1 U/ml (IL-6 bioassay), 3 U/ml (IL-1 bioassay), and 3 pg/ml (ELISA).
| Results |
|---|
|
|
|---|
The inability of neonates to respond to polysaccharide Ags was
best exemplified when splenocytes from both adult (2- to 3-mo-old) and
neonatal (10- to 12-day-old) mice were cultured with TNP-Ficoll, a
haptenated prototype TI-2 Ag, for 4 days. Using the PFC assay, a good
anti-TNP response of 208 ± 64 AFC/culture was detected in
adult cultures, which was about 20-fold higher than the neonatal
response of 11 ± 9 AFC/culture in eight experiments
(p < 0.001), one of which is represented in
Figure 1
A.
|
To determine which cytokine or cytokines present in D10 supernatant
could be responsible for stimulating neonatal B cells to respond to
TNP-Ficoll, we substituted defined cytokines in place of D10
supernatant. Since D10 supernatant was collected immediately after
stimulating the D10 cells by coculturing them with irradiated C57BL/6
splenocytes, this supernatant served as a good source of both Th2
cytokines secreted by the D10 cells (56) as well as accessory
cell-derived cytokines from the irradiated C57BL/6 splenocytes. Thus,
we initially screened a number of Th2 and accessory cell-derived
cytokines and finally some Th1 cytokines that were known to modulate B
cell function, both alone and in combination. Individually,
granulocyte-macrophage CSF and IL-2, -3, -4, and -10 did not stimulate
neonatal spleen cells to respond to TNP-Ficoll (data not shown),
whereas IL-1 alone consistently induced a robust response that was
>50-fold higher than that of neonatal splenocyte cultures in the
absence of IL-1 (hatched bars in Fig. 1
A). In this
experiment, IL-6 did not induce any response from neonatal splenocytes,
but in one of three experiments, it had a modest effect (data not
shown). IL-5, like IL-1, was also able to consistently overcome the
neonatal unresponsiveness, but this is the subject of another
manuscript.
Purified B cells only responded with a mixture of IL-1 and IL-6
Since the in vitro immunization with TNP-Ficoll was performed using total splenocytes, IL-1 might have overcome the unresponsiveness of neonatal B cells to TNP-Ficoll either by directly stimulating neonatal B cells, by activating other splenic cells to secrete cytokines that, in turn, activated the appropriate B cells, or both. To elucidate whether B cells were indeed directly responding to the exogenous IL-1, purified neonatal B cells would have to be cultured instead of total splenocytes.
First we characterized the lymphocyte population in neonates and adults
by flow cytometry. As expected, 3-day-old mice had about 18% B cells,
which increased with age (Table I
). They
were further characterized by HSA expression (57), which showed that
the number of immature B cells (IgM+HSAhigh)
was greater in neonates than in adults. The frequency of the
HSAhigh population was about the same in 3- and 10-day-old
neonates. In general, the 10- to 12-day-old neonates had fewer B and T
cells, but more Mac-1+ cells (Table I
) and pre-B cells
(15% of spleen cells as opposed to 2% in adults), with about 25% of
the spleen cells not in any of these categories. Thus, any procedure
employing negative selection would yield neonatal B cells contaminated
with the uncharacterized IgM- population, implying
that IgM+ B cells would have to be enriched by positive
selection. Consequently, we purified B cells by positively selecting
IgM+ cells by panning with the F(ab')2
fragments of goat anti-IgM immobilized on tissue culture-treated
dishes, yielding B cells of 85 to 95% purity. Positive selection using
anti-IgM has been previously employed to purify neonatal B cells
without affecting B cell function (58, 59, 60).
|
Since IL-1 and -6 together stimulated a response to TNP-Ficoll from panned neonatal and adult B cells, this suggested that these two cytokines were directly acting on B cells to overcome their unresponsiveness. To show conclusively that B cells were indeed the target of both cytokines, this experiment was repeated twice with 98% pure IgM+ adult B cells obtained by flow cytometry. In both instances, IL-1 and IL-6 together induced B cells to produce an Ab response against TNP-Ficoll (431 ± 124 AFC/culture with cytokines and 10 ± 3 AFC/culture without cytokines), implying that these cytokines were indeed acting directly on B cells, enabling them to overcome their unresponsiveness.
IL-1 and -6 together were effective even in 3-day-old neonates
Having established that IL-1 and -6 together could drive neonatal
B cells to produce a good Ab response to TNP-Ficoll, we were curious to
see how early in the ontogeny this combination of cytokines would work.
Total splenocytes from 3-day-old murine neonates responded to
TNP-Ficoll only when cultures were provided with Ag plus IL-1 (Fig. 2
). The response was small and not
significantly above background. Interestingly, the adult response
increased over twofold due to IL-1 treatment (p
< 0.05). However, if 3-day-old splenocytes were provided with both
IL-1 and -6, an excellent Ab response to TNP-Ficoll was detected on the
fourth day of culture (p < 0.001), and IL-6
also improved the adult IL-1-stimulated response
(p < 0.02). The response of 3-day-old
splenocytes was less than that of adults or even that of 10- to
12-day-old neonates, presumably because 3-day-old mice contain fewer B
cells (Table I
).
|
Since the concentrations of IL-1 and IL-6 used up to this point
were determined from titrations performed with only neonatal splenocyte
cultures, such concentrations could possibly mask a lower threshold for
one or both cytokines in purified adult B cell cultures. To determine
whether there was a reduced threshold in adult B cells, panned B cells
from both age groups were cultured with increasing concentrations of
either IL-1 or IL-6. Since purified B cells would only respond in the
presence of both cytokines, the second cytokine was maintained at its
optimal concentration. A reduced threshold for either cytokine in adult
B cells was not observed (Fig. 3
), and
surprisingly, for any given concentration of cytokine, the neonatal B
cells tended to produce a higher anti-TNP response.
|
The avidity of Abs secreted against some Ags, including
TNP-Ficoll, was reduced in aged mice (61, 62). To determine whether the
cytokine-induced TNP Abs made by neonates were immunologically
compromised, the avidities of the IL-1- and IL-6-induced TNP Abs
secreted by neonatal and adult splenocyte cultures were compared. We
could not use the Abs secreted by purified adult B cells as the
benchmark for avidity, because these cultures also needed to be
stimulated with the same pair of cytokines as the purified neonatal B
cells. Hence, we used Abs secreted by adult total splenocyte cultures
as the standard, to which we also added IL-1 and IL-6, to determine
whether these cytokines modified the avidities of the secreted Abs. The
relative avidities of the anti-TNP Abs produced by the PFC assay
were determined by inhibiting plaque formation in the presence of
varying concentrations of free hapten. Free hapten competes with the
hapten coupled to the target SRBC for the plaque-producing Abs, thus
reducing the incidence of plaque formation (63). The profiles of the
inhibition curves of the Abs secreted by both adult and neonatal
splenocyte cultures were very similar, suggesting that the avidities to
the TNP determinant were comparable (Fig. 4
) and indicating that this cytokine
treatment enabled neonatal B cells to secrete immunologically effective
Abs against a TI-2 Ag.
|
Since purified murine B cells needed both IL-1 and IL-6 to respond
to TNP-Ficoll, this raised the interesting possibility that one
cytokine may be required for the second to exert its function, such as
up-regulation of the second cytokines receptor. To address this
issue, a number of B cell cultures containing TNP-Ficoll were
established on day 0, and IL-1 and/or IL-6 were added either
immediately or after 1 or 2 days. The best anti-TNP response was
observed when both cytokines were added at the initiation of the 4-day
culture (Fig. 5
). If the addition of both
cytokines was delayed by 1 day, there was a small, but insignificant,
reduction in the PFC response, but a further 1-day delay completely
abrogated the response. If either cytokine was added at the start of
the culture, and the second cytokine was added a day later, the
response was comparable to that to adding both cytokines on the first
day of culture. However, when the addition of the second cytokine was
delayed to day 2, the response was significantly reduced, but still
higher than that when adding both cytokines on day 2. In summary, the
best Ab response to TNP-Ficoll occurred if both cytokines were added
within 1 day after initiation of culture, and the order of addition of
cytokines was immaterial.
|
Since IL-1 was also known to amplify the proliferation of mature adult B cells in response to anti-IgM (37, 64), we were interested to determine whether IL-1 and IL-6 were acting by overcoming the inability of neonatal B cells to proliferate in response to TNP-Ficoll. Since TNP-Ficoll-induced B cell proliferation cannot be measured reliably due to the small number of TNP-specific B cell clones, we used anti-IgM instead, which had previously been established to be a polyclonal TI-2 Ag (65, 66). Immature B cells, including neonatal splenic B cells, unlike mature B cells, do not proliferate in response to anti-IgM unless cytokines such as IL-4 are provided (59).
Although IL-1 on its own did not induce B cell proliferation, when
added together with anti-IgM, neonatal B cells responded at a rate
comparable to that of adult B cells stimulated with anti-IgM alone
(Fig. 6
). While IL-6 alone did not
stimulate neonatal B cells to proliferate in response to anti-IgM,
it did augment adult B cell proliferation. When both cytokines were
added together, this resulted in an additive effect in adult B cell
proliferation and a good synergy in neonatal B cell proliferation in
response to anti-IgM, bringing the neonatal B cell response close
to that of adult B cells. The combination of IL-1 and IL-6 was more
effective than IL-4 alone or IL-4 with IL-1 or IL-6 (Fig. 6
).
Interestingly, there was no cooperative effect between IL-6 and IL-4, a
cytokine whose effects on both neonatal and adult B cell proliferation
were very similar to those of IL-1, nor did IL-4 influence the
IL-1-induced or IL-1- plus IL-6-induced proliferation in response to
anti-IgM, suggesting a special relationship between IL-1 and IL-6.
Preliminary results from whole splenocyte cultures from adult Xid mice,
which did not respond to anti-IgM (40, 65), indicated that just as
with neonatal BALB/c B cells, IL-1 induced proliferation of Xid B cells
in response to anti-IgM, and IL-6 enhanced the effect of IL-1
(R. L. Chelvarajan, H.-J. Wu, and S. Bondada, manuscript in
preparation).
|
Neonatal lymphocytes have been reported to have a higher rate of spontaneous apoptosis in vitro (67). Friedlander et al. (68) demonstrated that if IL-1 was provided exogenously to HeLa cells in culture before an apoptotic stimulus, it had the ability to stave off apoptosis. It was thus possible that IL-1 improved the survival of neonatal B cells in culture, thus increasing the numbers of B cells that could respond to anti-IgM or TNP-Ficoll. To test this hypothesis, splenocytes were cultured in the presence or the absence of IL-1, and an apoptosis assay was performed immediately after harvesting the cells. After about 2 days in culture, the proportion of B220+ cells undergoing spontaneous apoptosis in the unstimulated neonatal cultures was slightly higher than that in adult splenocyte cultures (52 ± 5 vs 39 ± 10%; data not shown). If cultures were provided with IL-1 at the initiation of culture, this reduced the level of apoptosis in both neonatal and adult B220+ cells to 39 ± 4% (p = 0.057) and 29 ± 7% (p = 0.335), respectively (data not shown). Addition of the IL-1 and IL-6 together did not improve the level of survival afforded by either cytokine on its own (data not shown). Thus, the reduction of spontaneous apoptosis might play some role in the ability of IL-1 to overcome neonatal B cell unresponsiveness in vitro.
Neonatal splenocyte cultures do secrete IL-6, but not IL-1
Since neonatal splenocyte cultures responded to TNP-Ficoll when
supplied with IL-1, while purified B cells needed IL-1 and IL-6, could
a deficiency of IL-1 and/or IL-6 account for the unresponsiveness of
neonatal B cells to polysaccharides Ags? Both neonatal and adult
splenic cells spontaneously secreted IL-6 (Fig. 7
A). The level of IL-6 in
neonatal cultures was about 4 times higher than that in adult cultures.
Since macrophages were known to readily take up TNP-Ficoll among other
carbohydrates (69, 70), cultures were also incubated with TNP-Ficoll to
determine whether this polysaccharide would activate macrophages,
leading to a boost in IL-6 production. However, this treatment did not
increase the level of IL-6 secreted (Fig. 7
A). Even though B
cells have been shown previously to secrete IL-6 (71), in our hands,
supernatants from cultures of purified B cells (with or without
TNP-Ficoll) from either age group contained undetectable levels of IL-6
(data not shown).
|
| Discussion |
|---|
|
|
|---|
The additional requirement for IL-6 by purified B cells from either age group is in agreement with previous observations of synergy between these two cytokines in inducing the adult B cell response to dextran sulfate (74). Our ability to detect IL-6 production in cultures of whole spleen cells but not of purified B cells explains the need for addition of IL-6 for purified B cells to respond to TNP-Ficoll. Our observation that neonatal splenocytes secrete more IL-6 than those from adults contrasts with the report by Schibler et al. (75) that human neonatal monocytes are defective in IL-6 secretion. The authors considered neonatal monocytes to be defective, not because they failed to make any IL-6 at all but because they made about half as much as adults. The authors, in fact, showed that upon stimulation, human neonatal monocytes are quite competent in up-regulating the secretion of IL-6. Although many transformed B cells have been shown to produce IL-6 (76) and IL-6 mRNA is detected in anti-Ig-stimulated B cells (77), its production by TNP-Ficoll-activated B cells must be insufficient to induce a PFC response. The non-B cell populations in the neonatal spleen are at least as efficient as those in the adult (if not better) in producing sufficient IL-6 in whole spleen cultures, such that additional supplementation is unnecessary.
Unlike IL-6, IL-1 production is deficient in neonatal splenocytes.
Previous studies with B cells from young mouse adult lymph nodes and
aged mouse spleens suggest that the source of IL-1 is a non-B cell,
most likely an accessory cell, such as macrophages and/or dendritic
cells (26, 29, 78). These B cells also fail to respond to the TI-2
stimuli unless supplemented with splenic accessory cells or IL-1 (26, 29). The level of IL-1 is low despite the presence of a significantly
larger macrophage population (Mac-1+ and
F4/80+) in neonatal spleens vs adult spleens (Table I
and
data not shown), which suggests that these accessory cells may be
deficient in IL-1 secretion, and such a deficiency has been reported in
accessory cells of aged mice (79). Dendritic cells can also produce
IL-1, but have been shown to be functionally immature in neonates (78, 80), which could contribute to the defective IL-1 production in the
neonate. In support of this possibility, we were unable to detect IL-1
in neonatal splenocyte cultures stimulated with TNP-Ficoll (data not
shown) or anti-IgM, which can activate most B cells. Before this
report, only one other publication reported IL-1 production by normal B
cells (72), although that study was performed with human B cells.
However, the levels of IL-1 produced in an Ag-specific response do not
appear sufficient to drive the differentiation of purified B cells from
adults.
Although the causes of the lack of a neonatal response to TI-2 Ags have not been characterized in detail, one early study showed that neonatal injection of TNP-Ficoll induces clonal expansion of TNP-specific B cells, which could be subsequently induced to differentiate into Ab-secreting cells with mitogens such as LPS (33). Similarly, we have shown that lymph node B cell unresponsiveness may be due to a defect in differentiation, since in vivo priming with TNP-Ficoll expands TNP-specific B cell precursors without Ig secretion (22). The ability of accessory cell-derived cytokines to induce neonatal B cell responses to TNP-Ficoll is consistent with these observations. It also explains the finding that another macrophage activator, monophosphoryl lipid A, can overcome neonatal unresponsiveness to type III pneumococcal polysaccharide (81).
In contrast to these studies that point to a defect in the differentiation step of B cells, experimental models that use anti-Ig have found neonatal B cells to be defective at the clonal expansion step (82, 83). Although neonatal B cells enter into early G1 phase, they fail to progress into late G1 and S phases upon B cell receptor cross-linking. In these systems, neonatal B cells are defective in B cell receptor-mediated signal transduction compared with adult B cells. Thus, neonatal B cells are defective in the production of inositol 1,4,5-triphosphate; expression of the protein tyrosine kinases, Fgr and Fyn; induction of the immediate early gene c-myc; as well as induction of cdk2, cdk4, and cyclin E proteins (84, 85, 86).
IL-1 has been implicated in the initial activation of B cells (87) and, later, in promoting Ag-induced clonal expansion (64, 88). This is in agreement with our finding that IL-1 was required early, during the first 1 to 2 days of culture. What is surprising is that IL-6, a cytokine that acts directly on preactivated B cells (89) to induce the final maturation of B cells into Ab-producing cells (90, 91), is required during the same time frame as the early acting IL-1. Maybe IL-1 hastens the activation and clonal expansion of the TNP-Ficoll-stimulated B cells, such that the differentiation occurs 1 day earlier. If these clonally expanded cells are not stimulated with IL-6 immediately, they probably become unresponsive and undergo apoptosis. Alternatively, IL-6 could be another B cell growth factor. There is ample evidence that IL-6 supports the growth of transformed murine and human B cells (92, 93). Moreover, Vink et al. (74) noticed that IL-6 enhances the IL-1- plus anti-IgM-induced proliferation in adult murine splenic B cells, while Mizutani et al. (94) showed that IL-6 enhanced the IL-1- plus Con A-induced proliferation of the D10.G4 T cell line. Since neither these studies nor ours noticed significant proliferation induced by IL-6 in the absence of IL-1, it appears that IL-6 only amplifies the growth response delivered by IL-1. Further evidence, albeit circumstantial, for the early requirement for IL-6 is provided by Freeman et al., who reported that, upon anti-IgM stimulation, adult splenic B cells rapidly expressed the IL-6 gene, with the peak of expression occurring about 4 h after stimulation (77).
Neonatal splenocytes have a greater tendency to undergo spontaneous apoptosis in vitro (data not shown) (67), and this might play a major role in inhibiting the response to TI-2 Ags. In both cultured primary cortical neurons (95) and in a murine fibrosarcoma cell line (96), pretreatment with IL-1 significantly improved cell viability upon stimulation with an apoptotic signal. The expression of egr-1 is also found to be deficient in immature B cells from the bone marrow (97). Induction of the egr-1 gene may be a key event, since our recent studies with an immature B cell lymphoma cell line showed that egr-1 expression is causally related to the apoptotic vs the growth response induced by anti-IgM (98). In agreement with these models we are also unable to demonstrate the proliferation of neonatal B cells upon treatment with anti-IgM Abs unless they are supplemented with IL-1. Thus, the accessory cell-derived cytokines can completely overcome the proliferation defect in neonatal B cells. It is conceivable that IL-1 promotes neonatal B cell resistance to apoptosis by facilitating egr-1 expression.
In our study, IL-1 reduces the proportion of B220+ cells undergoing apoptosis, but this reduction is rather small compared with the level of Ab response to TNP-Ficoll encountered in the presence of IL-1. Therefore, the antiapoptotic activity of IL-1 might play a minor role in overcoming neonatal unresponsiveness to TI-2 Ags. The major role may be in overcoming the block in the proliferation step, as outlined above, although our IL-4 data show that this alone is not sufficient.
The Th2 cytokine, IL-4, also induces anti-IgM-stimulated neonatal B cells to proliferate to a level comparable to that induced by IL-1, but, unlike IL-1, IL-4 does not induce neonatal splenocytes to differentiate into anti-TNP-secreting cells. When IL-4 was added in combination with IL-1 (neonatal splenocytes) or with IL-1 and IL-6 (purified neonatal B cells), it frequently inhibits the differentiation response to TNP-Ficoll (data not shown). In fact, Alderson et al. (99), using a single cell study, showed that although IL-4 induces TI-2 Ag-dependent proliferation of single B cells, it is a poor agent for differentiation. Interestingly, they too noticed that IL-4 inhibited IL-1-induced Ag-dependent differentiation of B cells into Ig-secreting cells induced by IL-1. In our hands, the level of inhibition caused by IL-4 varies from experiment to experiment, and this phenomenon is being further investigated.
Recently, Snapper et al. (60) were able to induce adult-like levels of IgM secretion upon stimulation of purified neonatal murine B cells with anti-IgD-dextran, a prototype TI-2 Ag, in the presence of IL-4, IL-5, and CD40L (i.e., CD154). The ability of neonatal B cells to respond to anti-IgD-dextran in the presence of a plethora of T cell-derived stimuli is not surprising, since it is known that neonatal B cells can respond to protein-polysaccharide conjugates that can activate T cell help (34, 35). Our system is unique in that unseparated spleen cells need only IL-1 to respond to TNP-Ficoll, while just one other cytokine, IL-6, is required to activate purified B cells, and no T cell signal, such as CD40L, is required. Since many agents can augment IL-1 or IL-6 production from a variety of cell types in the body (100), it may be much easier to devise a means to induce neonatal responses to polysaccharide Ags with nontoxic IL-1- and/or IL-6-elevating adjuvants. This has a very important clinical significance, since it had been difficult to extend the success of H. influenzae-diphtheria toxin conjugate to the pneumococcal polysaccharide system due to the large number of serotypes of pneumococcal bacteria that vary in their capsular polysaccharide structures (2).
In summary, we have shown that immature neonatal B cells, long considered to be intrinsically incapable of mounting an immune response to TI-2 Ags, differentiated into Ig-secreting cells at a level comparable to that of adult B cells, and secreted TNP Abs of appropriate avidity when stimulated with TNP-Ficoll in the presence of IL-1 and IL-6. The defect in neonates might be due to the inability of accessory cells in the neonatal spleen to secrete sufficient quantities of IL-1.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Current address: School of Biological Sciences, University of Kentucky, Lexington, KY 40506. ![]()
3 Address correspondence and reprint requests to Dr. Subbarao Bondada, 329 Sanders-Brown Center on Aging, 800 S. Limestone, University of Kentucky, Lexington, KY 40536-0230. E-mail address: ![]()
4 Abbreviations used in this paper: Xid, X-linked immunodeficiency; TI-2, thymus-independent type 2; TNP, trinitrophenyl; HSA, heat-stable Ag; AFC, antibody-forming cell; PFC, plaque-forming cell. ![]()
Received for publication March 5, 1998. Accepted for publication June 1, 1998.
| References |
|---|
|
|
|---|
antibody as a probe of pathogenesis and therapy of rheumatoid disease. Immunol. Rev. 144:195.[Medline]
and secretion of IL-1ß. J. Immunol. 143:864.[Abstract]
This article has been cited by other articles:
![]() |
M. Gururajan, A. Simmons, T. Dasu, B. T. Spear, C. Calulot, D. A. Robertson, D. L. Wiest, J. G. Monroe, and S. Bondada Early Growth Response Genes Regulate B Cell Development, Proliferation, and Immune Response J. Immunol., October 1, 2008; 181(7): 4590 - 4602. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Kanswal, N. Katsenelson, A. Selvapandiyan, R. J. Bram, and M. Akkoyunlu Deficient TACI Expression on B Lymphocytes of Newborn Mice Leads to Defective Ig Secretion in Response to BAFF or APRIL J. Immunol., July 15, 2008; 181(2): 976 - 990. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Chelvarajan, D. Popa, Y. Liu, T. V. Getchell, A. J. Stromberg, and S. Bondada Molecular mechanisms underlying anti-inflammatory phenotype of neonatal splenic macrophages J. Leukoc. Biol., August 1, 2007; 82(2): 403 - 416. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. R. Alugupalli, S. Akira, E. Lien, and J. M. Leong MyD88- and Bruton's Tyrosine Kinase-Mediated Signals Are Essential for T Cell-Independent Pathogen-Specific IgM Responses J. Immunol., March 15, 2007; 178(6): 3740 - 3749. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Ke, M. Gururajan, A. Kumar, A. Simmons, L. Turcios, R. L. Chelvarajan, D. M. Cohen, D. L. Wiest, J. G. Monroe, and S. Bondada The Role of MAPKs in B Cell Receptor-induced Down-regulation of Egr-1 in Immature B Lymphoma Cells J. Biol. Chem., December 29, 2006; 281(52): 39806 - 39818. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. P. Wiertsema, S.-K. Khoo, G. Baynam, R. H. Veenhoven, I. A. Laing, G. A. Zielhuis, G. T. Rijkers, J. Goldblatt, P. N. LeSouef, and E. A. M. Sanders Association of CD14 Promoter Polymorphism with Otitis Media and Pneumococcal Vaccine Responses. Clin. Vaccine Immunol., August 1, 2006; 13(8): 892 - 897. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. E. Butler, D. H. Francis, J. Freeling, P. Weber, and A. M. Krieg Antibody Repertoire Development in Fetal and Neonatal Piglets. IX. Three Pathogen-Associated Molecular Patterns Act Synergistically to Allow Germfree Piglets to Respond to Type 2 Thymus-Independent and Thymus-Dependent Antigens J. Immunol., November 15, 2005; 175(10): 6772 - 6785. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. L. Chelvarajan, S. M. Collins, J. M. Van Willigen, and S. Bondada The unresponsiveness of aged mice to polysaccharide antigens is a result of a defect in macrophage function J. Leukoc. Biol., April 1, 2005; 77(4): 503 - 512. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. L. Chelvarajan, S. M. Collins, I. E. Doubinskaia, S. Goes, J. Van Willigen, D. Flanagan, W. J. S. de Villiers, J. S. Bryson, and S. Bondada Defective macrophage function in neonates and its impact on unresponsiveness of neonates to polysaccharide antigens J. Leukoc. Biol., June 1, 2004; 75(6): 982 - 994. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. E. Butler, P. Weber, M. Sinkora, D. Baker, A. Schoenherr, B. Mayer, and D. Francis Antibody Repertoire Development in Fetal and Neonatal Piglets. VIII. Colonization Is Required for Newborn Piglets to Make Serum Antibodies to T-Dependent and Type 2 T-Independent Antigens J. Immunol., December 15, 2002; 169(12): 6822 - 6830. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Muthukkumar, J. Goldstein, and K. E. Stein The Ability of B Cells and Dendritic Cells to Present Antigen Increases During Ontogeny J. Immunol., November 1, 2000; 165(9): 4803 - 4813. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. P. Arulanandam, J. N. Mittler, W. T. Lee, M. O'Toole, and D. W. Metzger Neonatal Administration of IL-12 Enhances the Protective Efficacy of Antiviral Vaccines J. Immunol., April 1, 2000; 164(7): 3698 - 3704. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |