Main

Human neonates show greater susceptibility to infections than do adults and when infected are able to mount only weak immune responses, consisting mainly of IgM antibody(1). They exhibit low levels of endogenous serum IgM, little if any IgA or IgE, and their IgG is of maternal origin(1). Although serum IgM and IgG levels rise significantly by 6-12 mo of age, IgG antibody responses to TI polysaccharide antigens remain deficient through age 2(2). Consequently, children of this age group are susceptible to infections with bacteria having capsular polysaccharides including Streptococcus pneumoniae, Neisseria meningitides, and Hemophilus influenzae(3–5). Immunization efforts against some of these organisms in children under 2 y of age recently have become successful as a result of the use of conjugate vaccines, which consist of poorly immunogenic polysaccharide molecules linked to immunogenic protein carrier molecules(6, 7).

In vitro experiments demonstrate that this immunologic dysfunction reflects immaturity of both the B and T cell compartments(8–10). Neonatal B cells secrete only small amounts of IgM and almost no IgG or IgA upon stimulation with T cell-dependent stimuli, such as pokeweed mitogen, and are somewhat more responsive to the TI stimuli SAC and EBV(11–14). Studies of the newborn's antibody repertoire reveal normal precursor frequency for autoantibodies of the IgM class, but a decreased number of precursors for IgG antibodies against foreign antigens(15–17). A distinct B cell subpopulation expressing the CD5 marker (B1 cells) has been found to predominate among human cord blood B cells and is known to secrete most of the human autoantibodies(18, 19). Differences have also been shown to exist between the phenotype of neonatal and adult peripheral blood B cells. A greater percentage of unstimulated neonatal B cells express activation antigens compared with unstimulated adult B cells(20). It is unclear whether these B cells have undergone some form of activation or if they represent a distinct phenotype.

When neonatal B cells are cultured with suboptimal concentrations of T cell activators, they demonstrate enhanced responses, whereas adult B cells do not(17). At optimal concentrations, the emergence of T cell suppressive effects occurs in the neonate but not in the adult(21). T cells from cord blood also exhibit a deficient helper function for Ig synthesis and dysregulated production of certain cytokines(22–26). Cytokine receptors for IL-2, IL-4, IL-6, IL-7, tumor necrosis factor, and interferon-γ all have been found to be expressed at lower levels in cord blood lymphocytes when compared with adult lymphocytes(27). These observations suggest that the combination of an intrinsic B cell and/or T cell immaturity, a lack of immunologic memory, an incomplete cytokine response, and/or a helper/suppressor T cell imbalance might lead to a deficient humoral immune response in neonates, which consists of an antibody pattern with low IgM and almost no IgG, IgA, or IgE.

The exact mechanism underlying B cell immaturity in the neonate is unclear. Previous studies that demonstrated defects in neonatal B cell responses used non-antigen-specific B cell polyclonal immunologic activators or immobilized, CD3-activated T cells(22). Because the most profound defect of neonates is their unresponsiveness to TI antigens, utilization of other stimuli may not provide an appropriate model with which to study the basis for neonatal unresponsiveness to polysaccharide antigens. In a previous study we have shown that dex-conjugated anti-Ig provides a model to study TI B cell activation events in the adult(28). To date no study has directly compared the ability of neonatal human B cells to become activated after ligation of their sIg by bivalent or multivalent cross-linking stimuli with that of adult B cells. To address this issue, we have used unconjugated anti-Ig, dex-conjugated anti-Ig antibody, and SAC to study the responses of adult and neonatal cord blood B cells. We examined the magnitude of cellular proliferation, changes in [Ca2+]i, alterations in MHC class II expression, and changes in cell size. We also studied the effect of direct B cell activation by stimulation with PKC activators and calcium ionophores. The results reveal that cord blood B cells are able to respond to sIg-stimulated activation in a manner similar to that observed in control adult peripheral blood B lymphocytes, and suggest that the defect in the neonatal response is distal to these early activation events.

METHODS

Antibodies. IgD MAb (δIA6.2) was a gift from Dr. John Kearny (University of Alabama, Birmingham, AL). Anti-δ MAb, cloneδIA6.2αδ (IgG2a subclass), and a control MAb of the IgG2a subclass (UPC-10) were grown as ascites fluid in Pristane-primed BALB/c mice and were a generous gift from Dr. Fred Finkelman (University of Cincinnati College of Medicine, Cincinnati, OH). Anti-δ MAb and control MAb UPC-10 were purified from ascites by ammonium sulfate precipitation and DE-52 anion exchange chromatography, as previously described(28). Affinity-purified F(ab′)2 fragments of goat anti-human IgM(anti-μ) were obtained from Jackson Immunoresearch Laboratories (Avondale, PA). Anti-δ and anti-μ MAbs, and the control MAb UPC-10, were conjugated to high molecular weight dex as previously described(30). Briefly, they were covalently linked through stable thioether bonds to molecules of aminoethyl-carbamylmethyl dex using a heteroligation technique.

An antibody mixture for human non-B cell surface epitopes (rB#16) was a kind gift of Dr. Carl June (Naval Medical Research Institute, Bethesda, MD) and was made by combining murine MAb 9.6 (anti-CD2), G10-1 (anti-CD8), and FC-2 (anti-CD16), which were a generous gift from Dr. Jeffrey Ledbetter(Oncogene Corp., Seattle, WA); MAb 20.3 (anti-CD14) and 38.1 (anti-CD3), a kind gift from Dr. John Hansen (Fred Hutchinson Cancer Research Center, Seattle, WA); OKM1 (anti-CD11) (American Type Culture Collection, Rockville, MD); and murine anti-human NKH-1 MAb (Coulter Clone, Hialeah, FL). This mixture was used to negatively select for B cells in the preparation described below.

FITC-conjugated anti-CD19, PE-conjugated anti-CD19 and HLA-DR, FITC-IgG1 + PE-IgG2 (Simultest Control), and rat anti-mouseκ light chain MAb were obtained from Becton Dickinson (San Jose, CA). FITC-conjugated anti-human B1 (anti-CD20)-specific MAb was obtained from Coulter Clone (Hialeah, FL). FITC-conjugated anti-δ MAb was a generous gift from Dr. Fred Finkelman (University of Cincinnati College of Medicine, Cincinnati, OH).

Reagents. SAC was purchased from Calbiochem-Behring (La Jolla, CA). PDBU, phorbol myristate acetate, and ionomycin were obtained from Sigma Chemical Co. (St. Louis, MO).

B cell preparations. Cord blood samples were obtained as heparinized blood from newborn placentas of healthy term infants (38-42 wk) with no maternal complications or evidence of sepsis at the time of birth. MNCs from cord blood were isolated using Ficoll-Paque (Pharmacia, Uppsala, Sweden) density gradient centrifugation. Monocyte-depleted cell fractions were obtained by density gradient centrifugation with Sepracell (Sepratech Corp., Oklahoma City, OK). Sepracell centrifugation separated the cells into three layers. The top layer contained most of the monocytes(31). The middle and bottom layers contained most of the B cells, and only a few percent were monocytes.

Cells were also isolated from the buffy coat obtained from adult peripheral blood of normal volunteer subjects from the National Institutes of Health Blood Bank and were either handled as above or were separated by counterflow centrifugal elutriation, as previously described(32). The latter technique yielded a monocyte-depleted, MNC-enriched population of small, dense lymphocytes in fractions 2 and 3, and was kindly provided by Dr. Larry Wahl (National Institute of Dental Research, Bethesda, MD).

In some experiments the cord blood and adult peripheral blood MNC were further enriched for B cells. Cells were incubated at 4°C for 50 min with saturating amounts of antibody mixture rB#16 containing antibodies specific for T cell, natural killer cells, and monocyte surface antigens as listed above. This was followed by incubation at 4°C for 1 h on a rotator with goat anti-mouse IgG-coated magnetic beads (Dynabeads M450, Dynal Inc., Oslo, Norway). The cell suspension was placed in a samarium cobalt magnetic field(Edmund Scientific, Barrington, NJ), and the nonadherent cells were harvested. These cells contained >95% (adult) or 62 ± 8% (neonatal) CD19+ cells by FACS analysis.

Cell culture and proliferation assays. Cells were cultured in RPMI 1640 (GIBCO Laboratories, Gaithersburg, MD) medium containing L-glutamine, gentamicin (50 μg/mL), and 10% lipopolysaccharide-free FCS(Hyclone, Logan, UT) in flat-bottom microtiter plates (Nunc, Copenhagen, Denmark). Unless otherwise specified, cells were cultured at a density of 1× 105 cells/0.2 mL of medium per well. Cultures were incubated for 72 h at 37°C in a humidified, 5% CO2 environment. Proliferation was assayed by pulsing each well with 1.0 μCi of [3H]thymidine 18 h before harvesting. Each culture was performed in triplicate with results expressed as the arithmetic mean ± the SD of counts/min per well. All experiments were carried out several times, and a representative experiment is given in the “Results.”

Analysis of cell size. B cell-enriched, monocyte-depleted cord blood MNC were incubated for 24 h at 37°C with anti-δ-dex at a concentration of 0.1 μg/mL or with media in 24-well tissue culture plates(Costar, Cambridge, MA). Cell size was measured with a Coulter Counter and Coulter Channelizer (Coulter Electronics, Hialeah, FL). The data are displayed as population size distribution curves.

Fluorescence labeling of cell surface antigens and analysis by FMF. Cell suspensions (50 μL) of cord blood or adult cells were incubated at 4°C for 30 min with saturating concentrations of the appropriate fluorochrome-labeled antibodies for specific cell surface antigens. Human AB serum (10%) was always added to the incubation medium to prevent nonspecific binding of antibody to Fc receptors. The cells were washed twice, resuspended in 2% paraformaldehyde, and maintained at 4°C until analysis by single or dual parameter FMF. Data from single parameter (5× 105 cells) or dual parameter (1 × 106 cells) FMF were collected with a FACScan (FACS Systems, Becton Dickinson, Sunnyvale, CA) as previously described(33).

Indo-1 assay of intracellular calcium. Measurement of[Ca2+]i was performed using the ratio of Ca2+-dependent Indo-1 fluorescence as described elsewhere(34). Briefly, a B cell-enriched fraction was loaded with Indo-1 medium [Hanks' balanced salt solution with Ca2+ and Mg2+ (Quality Biologicals, Inc., Gaithersburg, MD) with 1.0% FCS]. Pluronic/Indo working solution was made by reconstituting a 50-μg ampule of Indo 1-AM (Molecular Probes, Eugene, OR) with 50 μL of DMSO. To this was added 25 μL of Pluronic F-127 stock (Molecular Probes, Eugene, OR) and 62.5μL of filtered FCS (Hyclone Laboratories, Inc., Logan, UT). The cells were incubated at 37°C for 30 min with the Pluronic/Indo solution, washed, stained with PE-labeled anti-CD19, and resuspended in Indo-1 medium before FACS analysis. The peak mean [Ca2+]i increase for B cells, the percentage of B cells responding, and the integrated area under the curve of the mean [Ca2+]i above baseline versus time plot was calculated, as previously described(35).

RESULTS

Neonatal cord blood B lymphocytes proliferate in response to sIg cross-linking. To identify the nature of the defect(s) underlying the reduced response of neonates to TI antigens, we used anti-Ig antibodies and SAC to stimulate B cells via their surface antigen receptors. We compared the proliferative response of cord blood to that of adult B cell preparations. Monocyte-depleted MNC populations were prepared from adult peripheral blood and neonatal cord blood, and were then cultured with soluble anti-μ antibodies or SAC. The percentage of CD19+ B lymphocytes was similar in the adult and cord blood sample (12.2 and 10.6%, respectively). Cells from both adult and neonatal samples proliferated in response to anti-μ and SAC(Table 1). The magnitude of the proliferative response was greater in the adult samples overall, but was variable from experiment to experiment. This differential response might be due to differences in the B cell percentages and/or in the composition of the non-B cell portion of the neonatal and adult cell preparations, or it may be secondary to an intrinsic disparity in the neonatal B cell response.

Table 1 Monocyte-depleted neonatal MNCs proliferate in response to soluble anti-μ or SAC
Full size table

Stimulation of cord blood B cell proliferation using nanogram concentrations of anti -δ- dex antibody. We have shown previously that anti-Ig antibodies coupled to dex provide a useful model to study TI type 2-mediated B cell activation and are effective at concentrations that are 1000-10 000-fold lower than those required to stimulate murine or human adult B cells with unconjugated anti-Ig(28). We wished to determine whether multivalent sIg cross-linking stimuli would induce a tolerogenic response in cord blood B cells or whether neonatal B cells would be as equally responsive as adult B cells to such low concentrations of anti-Ig-dex. Monocyte-depleted MNC populations from adult peripheral blood and neonatal cord blood (same cell populations as listed in Table 1) were incubated with varying concentrations of anti-δ-dex. Mitogenic responses of both adult and neonatal cells peaked at 0.01 μg/mL anti-δ-dex, and significant proliferative responses were seen at concentrations as low as 1 ng/mL. Anti-δ-dex stimulated a proliferative response of neonatal B cells that was greater than that stimulated by a 10 000-fold higher concentration of the unconjugated anti-μ antibody (compare Fig. 1 with Table 1). Results obtained in other experiments using anti-μ-dex antibodies were similar to those using anti-δ-dex (data not shown). A control MAb that was coupled to dex (UPC-10-dex) did not lead to proliferation higher than that seen with medium alone (data not shown). Thus, neonatal B cells, similar to adult B cells, possess the capability to proliferate after stimulation with low concentrations of anti-Ig-dex that may more closely resemble physiologic concentrations of multivalent antigens seen in vivo.

Figure 1
figure 1

Neonatal B lymphocytes proliferate upon stimulation with anti-Ig-dex. MNC (1 × 105) from neonatal cord blood(A) or adult peripheral blood (B) were cultured in microtiter plates in the presence of varying concentrations of anti-δ-dex antibody for 72 h. [3H]Thymidine (1.0 μCi/well) was added 18 h before harvesting. Results are expressed as the arithmetic mean counts/min per triplicate culture.

Full size image

To demonstrate that anti-Ig-dex-induced proliferation reflected B cell-dependent events and was directly dependent on the numbers of cultured B cells, a sample of cord blood monocyte-depleted MNC was enriched for B cells and then cultured with varying concentrations of anti-δ-dex. The proliferative response of unenriched (14% CD19+ cells) and B cell-enriched (61% CD19+ cells) fractions were both maximal using 0.01μg/mL of anti-δ-dex, but the response of the B cell-enriched sample was 3-4-fold higher at this concentration (data not shown). To rule out the possibility that non-B cells were responsible for the anti-δ-dex-induced proliferative response, a cord blood lymphocyte-depleted MNC sample that contained <1% CD19+ B cells was stimulated with anti-δ-dex. No proliferation of these cells was observed in response to anti-δ-dex or SAC (data not shown). These data support the idea that proliferation of cord blood cells in response to anti-Ig-dex reflects the ability of these stimuli to ligate sIg on B lymphocytes and initiate intracellular events ultimately leading to proliferation.

Anti -δ- dex induces a dose-dependent increase in [Ca2+]i in cord blood B lymphocytes. Increases in[Ca2+]i have been identified as an early activation event in anti-Ig-stimulated human and murine B cells(28, 35), but not in response to other B cell activators. To investigate whether the early sIg-mediated events in cord blood are associated with increases in [Ca2+]i and whether the magnitude is comparable in adult and cord blood B cells, we evaluated levels of [Ca2+]i induced in enriched cord blood B cell populations after addition of varying concentrations of anti-δ-dex. Proliferative responses, as assessed by [3H]thymidine incorporation, were measured for the same samples. The changes in[Ca2+]i are expressed as the total integrated area under the curve, which represents the magnitude of calcium mobilization, the duration of the response, and the number of cells responding. Because it is unclear from the proliferation data whether the majority of B cells or just a subset are responding, the determination of [Ca2+]i by this method allows us to assess what portion of the cells are responding. Increases in [Ca2+]i correlated with increasing concentrations of anti-δ-dex, and 10 μg/mL anti-δ-dex, which caused little if any [3H]thymidine incorporation in a simultaneous proliferation experiment, led to the largest increase in[Ca2+]i (Fig. 2). Interestingly, although few if any resting adult peripheral blood B cells exhibit[Ca2+]i above 200 nM(36), we found that, in this experiment, close to 40% of unstimulated cord blood B cells had a [Ca2+]i above 200 nM (data not shown). These data indicate that some cord blood B cells have high levels of[Ca2+]i without stimulation and that signal transduction via sIg leads to additional increases in[Ca2+]i in the majority of the cord blood B cells; however, there is no apparent relationship between the magnitude of the proliferative response and the extent of calcium mobilization that was stimulated at the higher concentrations of anti-δ-dex.

Figure 2
figure 2

Comparison of increases in total[Ca2+]i and [3H]thymidine incorporation of neonatal B cells stimulated by anti-δ-dex. B cell-enriched cord blood MNC were first incubated with PE-labeled Leu-12 (CD19) and loaded with Indo-1 as described in “Methods.” Cells were then stimulated with different concentrations of anti-δ-dex. Indo-1 fluorescence was measured in the PE-positive B lymphocytes by FMF. The net[Ca2+]i mobilization above baseline is shown as the integrated area under the curve (nM/s) (♦). Also shown is the proliferation data for the same cells after 72 h incubation with similar antibody concentrations. [3H]Thymidine (1.0 μCi/well) was added 18 h before harvesting. Results are expressed as the arithmetic mean counts/min of triplicate cultures (⊡). Data for the net [Ca2+]i mobilization and the cellular proliferation were compared with the maximum values obtained, where 100% reflects the proliferation induced by 0.01μg/mL anti-δ-dex, and the net [Ca2+]i increase induced by 10 μg/mL of the same antibody.

Full size image

Anti -δ- dex induces cell size increases in the majority of neonatal B cells. The finding that anti-δ-dex stimulated increased levels of [Ca2+]i in a large percentage of cord blood B cells might reflect the fact that the majority of B cells are stimulated to exit the G0 phase of the cell cycle but does not indicate whether they proceed to G1. To study this question, B cell-enriched, monocyte-depleted neonatal cord blood MNC were incubated with 0.1 μg/mL anti-δ-dex or medium for 24 h and then analyzed for size distribution using a Coulter Channelizer (Fig. 3). The major shift in size distribution observed suggests that the majority of the neonatal B cells were stimulated by anti-Ig-dex to enter G1.

Figure 3
figure 3

Anti-δ-dex induces large increases in neonatal B cell size. B cell-enriched, monocyte-depleted cord blood MNC fractions (58% CD20+ cells) were incubated for 24 h with 0.1 μg/mL anti-δ-dex or medium (control). Cell size was measured by a Coulter Counter and Channelizer.

Full size image

Anti -δ- dex induces an up-regulation in HLA-DR expression. B cell MHC class II antigens play a crucial role in direct activation of B cells, antigen presentation, and T cell-B cell collaboration. It has been reported previously that cross-linking of sIg by anti-Ig antibodies induces an up-regulation in MHC class II antigen expression on mature murine and human B cells(29, 30). It is possible that a defect in neonatal B cell HLA-DR up-regulation might result in impaired B cell differentiation or impaired T cell-B cell collaboration, with decreased antibody production. To study the effect of sIg cross-linking on HLA-DR antigen expression by cord blood B lymphocytes, B cell-enriched, monocyte-depleted cord blood MNC were incubated for 24 h with medium or anti-δ-dex, and subsequently labeled with PE-conjugated HLA-DR and FITC-conjugated B1 (CD20) antibodies or Simultest control. Dual fluorescence analysis revealed that, similar to adult B cells (data not shown), all B1+ unstimulated neonatal B cells expressed HLA-DR antigen, and that there was a marked increase in HLA-DR antigen expression after 1 d of incubation with anti-δ-dex (Fig. 4). Maximal changes in B cell HLA-DR expression were seen around 72 h in both cord and adult B cells, with cord blood B cells showing up to a 7.3-fold increase compared with an increase up to 8.4-fold in comparably cultured adult B cells. As with the cell size increase, the increase in expression of MHC class II molecules reflected a response of the majority of cells and not merely a minority subset.

Figure 4
figure 4

Anti-δ-dex enhances HLA-DR antigen expression on neonatal B cells. B cell-enriched, monocyte-depleted cord blood MNC fractions were incubated for 24 h with 0.1 μg/mL anti-δ-dex or medium(control). Cells were then harvested, stained with saturating concentrations of PE-labeled HLA-DR and FITC-labeled B1 (CD20), and analyzed by FMF. Data are presented as dual fluorescence contour plots.

Full size image

Direct activators of PKC enhance anti -δ- dex induced cellular proliferation. Activators that induce plasma membrane translocation of PKC, such as phorbol myristate acetate, PDBU, or indolactam, induce proliferation of human adult B cells and are able to synergize with anti-Ig in mediating proliferative responses(36). To examine the effect of PKC activation in cord blood B lymphocytes, B cell-enriched, monocyte-depleted cord blood MNC preparations were cultured with anti-δ-dex, PDBU, or anti-δ-dex plus PDBU. PDBU alone at the optimal concentration of 1 μM (dose response data not shown) induced modest proliferation of cord blood B cells (Fig. 5). When PDBU and anti-δ-dex were combined, however, the effect of PDBU was additive at lower concentrations of anti-δ-dex and synergistic at higher concentrations of the same antibody (Fig. 5). In both cord and adult samples, maximal responses were observed when 0.01 μg/mL of anti-δ-dex was combined with 1 μM PDBU, with cord blood B cells showing a 34-fold increase over baseline proliferation and adult B cells showing a 36-fold increase in the same experiment. These data suggest that activation of PKC in cord blood B cells leads to both direct activation of these cells and enhancement of anti-Ig-mediated stimulation. To compare the response of adult and cord blood B cells to stimuli that mimic intracellular messengers triggered by sIg cross-linking, adult and cord blood B cells were stimulated with the calcium ionophore ionomycin alone or in combination with the phorbol ester, PDBU (Fig. 6). Ionomycin alone caused little if any proliferation, and PDBU alone led to a low level of proliferation in both neonatal and adult B cells. The combination of the stimulators, however, resulted in proliferation of neonatal B cells that was comparable to that observed in adult B cells treated with both stimulators. These results suggest that the PKC-dependent mechanism of B cell proliferation, which is stimulated by phorbol esters, is intact in neonatal B cells.

Figure 5
figure 5

The phorbol ester, PDBU, enhances anti-δ-dex-induced neonatal cord blood B cell proliferation. B cell-enriched, monocyte-depleted cord blood MNC were cultured with medium alone (○), 1 μM PDBU alone (•), various concentrations of anti-δ-dex alone (▵), or anti-δ-dex with 1 μM PDBU(▴). Cells were cultured for 72 h. [3H]Thymidine (1.0μCi/well) was added 18 h before harvesting. Results are expressed as the arithmetic mean counts/min per triplicate culture.

Full size image
Figure 6
figure 6

The calcium ionophore, ionomycin, and the phorbol ester, PDBU, synergize in inducing human B cell proliferation. B cell-enriched, monocyte-depleted MNCs from neonatal cord blood or adult peripheral blood were cultured for 72 h with 1 μM ionomycin, 10 μM PDBU, 1 μM ionomycin, and 10 μM PDBU, or medium (control).[3H]Thymidine (1.0 μCi/well) was added 18 h before harvesting. Results are expressed as the arithmetic mean counts/min per triplicate culture± SEM.

Full size image

DISCUSSION

The data presented in this report demonstrate that activation of cord blood B cells by cross-linking of sIg results in cellular proliferation. The response of neonatal cord blood B cells was similar to that seen in adult human B cells, suggesting that cross-linking of sIg by antigens induces signals in cord blood B cells that culminate in B cell clonal expansion. Although other groups have reported that neonatal B cells can be induced to proliferate when cultured with activated human T cells or with antigen-nonspecific B cell polyclonal activators(11–14, 22), this is the first study demonstrating that cross-linking membrane Ig by both anti-Ig and by multivalent anti-Ig-dex leads to cellular proliferation. These results suggest that the inability of neonatal B cells to respond to multivalent sIg cross-linking antigens, such as bacterial polysaccharides, does not simply reflect an inability of these antigens to mediate effective signal transduction via sIg in a TI-type manner. However, antibody production is a later event, which does not always occur after the induction of cellular proliferation and therefore would need to be assessed separately.

Although Chang et al.(37) reported deficient proliferative responses of purified neonatal murine splenic B cells to anti-Ig-dex, their data are not necessarily in conflict with our findings. They demonstrated that dex-conjugated anti-Ig stimulates substantial responses(≈50 000 cpm) in neonatal murine B cells (versus background in unstimulated cells), although lower than that stimulated in adult murine B cells (≈175 000 cpm). The striking inability of neonatal murine B cells to respond to sIg cross-linking stimuli when not presented as a multivalent ligand contrasts with the rather “adult-like” response of human cord blood cells. This difference may reflect the fact that neonatal cord blood has a greater percentage of CD5+ B cells(18) than is found in neonatal murine spleen(38). It may also reflect a different responsiveness of circulating blood B cells compared with splenic B cells. Additionally, the neonatal mice studied by Chang et al. were 5-8 d old and are not immunologically equivalent to the term human newborn, who is known to be capable of mounting an immune response from early in the second trimester of gestation to a number of infectious agents(1). The murine response to anti-Ig is known to be age-dependent(39), and up-regulation of MHC class II expression recently has been shown to be deficient in the neonatal mouse(40), in contrast to our finding of up-regulation of HLA-DR expression in the human newborn. Finally, we cannot exclude the presence of T cells from our B cell-enriched cord blood samples, even though vigorous attempts were made to eliminate T cells with anti-CD2, anti-CD3, and anti-CD8 MAbs. These T cells may have a role in enhancing proliferative responses to sIg-activated B cells. We did confirm, however, that the magnitude of proliferation correlated not with the presence of T cells but with the percentage of B cells in culture, although we did not achieve the degree of B cell purity (98%) reported by Chang et al.(37).

The data presented demonstrate the similarity between the elevation in intracellular ionized calcium induced in neonatal cord blood B cells and that previously described in adult B cells after sIg cross-linking(36). There is a dose-dependent increase in[Ca2+]i influx in neonatal B cells that correlates with the concentration of anti-Ig-dex, but does not always correlate with the magnitude of cellular proliferation. The highest concentration of anti-Ig-dex tested (10 μg/mL) led to the maximum increase in[Ca2+]i, but did not lead to cellular proliferation above background levels.

The observation that cord blood B cells proliferated upon activation with the appropriate concentration of anti-Ig-dex indicates that human cord blood contains B cells capable of responding to sIg cross-linking stimuli, but did not distinguish whether the responding cells represented a minority or a majority of the population. The dual fluorescence FMF experiments, however, revealed increases in [Ca2+]i in >90% of B cells after stimulation with anti-Ig-dex. Furthermore, the majority of cord blood B cells were stimulated to show increased expression of HLA-DR antigen after activation with anti-Ig-dex. Finally, increases in cell size were documented in the majority of the B cell-enriched cell population, indicating that most of the B cells are capable of responding to anti-Ig-dex and entering into at least the G1 phase of the cell cycle. These findings suggest that the proliferation induced by anti-Ig-dex reflects the response of a large fraction of the cord blood B cell population.

Activation of mature human B cells with PKC activators, either alone or in conjunction with calcium ionophores serving as pharmacologic mediators, stimulates the same pathways of activation induced by sIg cross-linking and in our study, was capable of stimulating proliferation in cord blood B cells. Although neonatal B cells displayed a modest level of proliferation when stimulated with the phorbol ester PDBU alone, at optimal concentrations of anti-Ig-dex a synergistic response was observed between these two activators, as has been previously reported with adult B cells(36). Similarly, the combination of ionomycin and PDBU stimulated comparable levels of proliferation in neonatal and adult B cells, indicating that the targets for these mediators are intact in neonatal B cells.

The neonatal response to polysaccharide antigens has been compared with that of splenectomized adults(41), patients with Wiskott Aldrich syndrome(42), and patients with common variable immunodeficiency(43). Neonates lack adult-type splenic marginal zone B cells, which are known to play a role in the response to TI antigens(44). In the murine system, splenic marginal zone B cells proliferate in response to sIg cross-linking signals but are defective in Ig secretion when stimulated with anti-Ig-dex(44). Responsiveness to cytokines derived from non-T cells may be a critical regulator of responses to TI antigens(45), and low level expression of cytokine receptors in the murine spleen as well as in the human neonate may explain their lack of response. The human neonate is known to have deficient cord blood expression of receptors for IL-2, IL-4, IL-6, IL-7, tumor necrosis factor, and interferon-γ(27). A recent proposal for the steps involved in B cell response to polysaccharide antigens emphasizes the importance of costimulatory signals, such as interferon-γ, granulocyte/macrophage-colony-stimulating factor, IL-2, and certain bacterial constituents, in Ig secretion and class switching(46). Further clarification of the later events involved in the mature B cell response to TI type 2 antigens will allow for additional research into the neonatal response.

In summary, our data strongly suggest that there is no apparent deficiency in early activation events leading to entry into G1, or the later events leading to DNA synthesis, in neonatal cord blood B cells in response to bivalent or multivalent sIg cross-linking ligands. These results suggest that the origin of the deficiency in the human neonatal response to TI antigens lies at a point distal to those studied.