IL-4-dependent IgE switch in membrane IgA-positive human B cells

IgE responses by human B cells, separated according to membrane Ig classes, were analyzed in a clonal assay using EL-4 thymoma cells as helper cells, T cell supernatant, and rIL-4. In cultures seeded by means of the autoclone apparatus of the FACS, IgE responses were generated frequently by either IgM (mu+/gamma-alpha-) or IgA (alpha +/mu-)-positive B cells (16 and 14% of the Ig producing wells, respectively), but rarely by IgG (gamma +/mu-)-positive B cells (1.3% of Ig producing wells). The total amounts of Ig secreted by IgM-, IgG-, or IgA-positive cells and the total proportions of responding autoclone wells (23-27%) were comparable. All IgE secretion was IL-4 dependent. When the Ig secretion patterns from alpha +/mu- vs alpha +/mu-epsilon- B cells were compared, most autoclone wells from both types of cells produced IgA only, and similar proportions of IgA producing wells (6.2 and 6.0%) also secreted IgE. In addition, IgE restricted responses occurred 6 times more frequently with alpha +/mu- than with alpha +/mu-epsilon- cells, which suggests that membrane IgA+E double-positive, IgE committed B cells occur in vivo. The isotype pattern generated by alpha +/mu-epsilon- B cells cannot be explained by a chance assortment of separate IgA and IgE precursors or by cytophilic antibody. Thus, IL-4 dependent switch to IgE occurred frequently in IgM- or IgA-positive, but rarely among total IgG-positive, B cells. This could be relevant to IgE production in mucosal tissues rich in IgA expressing B cells.     

Suppression by IL-2 of IgE production by B cells stimulated by IL-4.

IgE production was obtained from B cells of BALB/c or nude mice when these cells were cultured with IL-4 plus LPS. IL-2 added to these cultures at the start (day 0), 1 or 2 days later completely suppressed the production of IgE. The production of IgG1 was also inhibited, but only if IL-2 was added on day 0. The production of other isotypes (IgM, IgG2a, IgG2b) was only slightly decreased by addition of IL-2. No suppression of IgE or IgG1 production was observed if monoclonal anti-IL-2 was added, whereas anti-IFN-gamma had no effect on the suppression of the production of these isotypes. The expression of CD23 on the third day of culture on B cells stimulated with LPS and IL-4 was markedly decreased when IL-2 was added to the cultures on day 0. Addition of monoclonal anti-IL-2 suppressed all effects produced by IL-2, whereas addition of anti-IFN-gamma had no effect. These results show that the suppression by IL-2, at least for the first signaling processes, are different from the suppression produced by IFN-gamma.     

Human interdigitating dendritic cells induce isotype switching and IL-13-dependent IgM production in CD40-activated naive B cells

Interdigitating dendritic cells (IDC) represent a mature progeny of dendritic cells (DC) in vivo and are exhibiting a strong lymphocyte stimulatory potential. Because of the restricted localization to secondary lymphoid organs where decisive cellular interactions take place in the initial events of immunity, IDC regulatory function was addressed in relation to naive B cells. In this study, we demonstrate that human tonsillar IDC induce a dual response from CD40-activated IgD+/CD38- naive B lymphocytes. IDC direct naive B cells toward either isotype switching or an IL-13-dependent IgM secretion. IDC-dependent proliferation, isotype switching, and Ig production are all strictly mediated by soluble factors, suggesting that such skewing in B cell activation is the result of differential cytokine expression. Moreover, IDC-expressed IL-13 represents a novel source of a cytokine with recently established effects in Th2 induction as well as in immunological disorders resulting in allergic reactions.     

The role of IL-4 in proliferation and differentiation of human natural killer cells. Study of an IL-4-dependent versus an IL-2-dependent natural killer cell clone

The role of IL-4 in proliferation and differentiation of human NK cells was studied using newly established sublines of an IL-4-dependent NK cell clone (IL4d-NK cells) and an IL-2-dependent NK cell clone (IL2d-NK cells) derived from a parental conditioned medium-dependent NK cell clone (CM-NK cells). IL-4 induced the higher proliferation of CM-NK cells, but abolished their NK activity and decreased CD16 and CD56 Ag expression. In contrast, IL-2 induced the higher NK activity and increased CD16 and CD56 Ag expression. Addition of anti-IL-4 antibody to the culture of CM-NK cells with CM inhibited the proliferation, but slightly increased NK activity, and largely increased CD56 Ag expression. Addition of anti-IL-2 antibody to the culture of CM-NK cells with CM inhibited both proliferation and cytotoxicity. Proliferation of IL4d-NK cells, which is totally dependent on rIL-4, is greater than that of IL2d-NK cells, which was greater than parental CM-NK cells. Morphologically, IL4d-NK cells are small and round, whereas IL2d-NK cells are large and elongated. Anti-IL-4 antibody inhibited proliferation of IL4d-NK but not IL2d-NK cells, whereas anti-IL-2 antibody inhibited that of IL2d-NK but not IL4d-NK cells. IL-2 was not detected in the supernatant from IL4d-NK cells, nor was IL-2-mRNA expressed in IL4d-NK cells. In contrast, IFN-gamma production and protein expression in IL4d- and IL2d-NK cells were detected. NK cell activation markers (CD16 and CD56) were expressed on IL2d-NK cells but not IL4d-NK cells. IL4d-NK cells were not cytotoxic to any tumor cells tested, whereas IL2d-NK cells displayed potent NK activity and lymphokine-activated killer activity. IL4d-NK cells failed to bind K562 tumor cells, whereas one-third of the IL2d-NK cells did. IL4d-NK cells responded to rIL-2, proliferated, and differentiated into cytotoxic NK cells, whereas IL2d-NK cells failed to respond to rIL-4 and died. These results raise a possibility that IL4d-NK cells or IL2d-NK cells primarily represent the immunologic properties of immature or activated types of human NK cells, respectively. Our results provide the first evidence of the capability of IL-4 to support continuous proliferation of a lymphocyte clone with immature NK cell characteristics and to stimulate IFN-gamma production in the clone. IL-4 is suggested as a potential growth factor for certain types of human NK cell progenitors.     

IgE-enhancing activity directly and selectively affects activated B cells: evidence for a human IgE differentiation factor

T cells from highly atopic individuals spontaneously secrete in vitro a factor that specifically induces IgE synthesis from normal human B cells. We investigated the effects of such T cell supernatants derived from atopic individuals (TCSN-A) on functionally distinct B cell subsets to determine at what developmental stage B cells become responsive to this IgE-enhancing activity. B cells from normal and allergic donors were separated into subsets of small resting and large activated cells by density centrifugation or unit gravity sedimentation. When stimulated by TCSN-A, large activated B cells made more IgE than small resting B cells. The difference was as much as 3300% in comparing these subsets from allergic donors. Similarly, resting B cells stimulated by Staphylococcus aureus Cowan I (SAC) made 52 to 125% more IgE in response to TCSN-A than unstimulated small resting B cells. However, IgE production from large B cells, already activated in vivo, was not enhanced by the addition of SAC. Notably, the IgE level synthesized by in vivo large activated B cells from allergic persons was markedly greater than that seen with similar cells from normal donors, whereas resting B cells purified from allergic and normal donors produced comparable levels of IgE in response to TCSN-A. These results suggest that this enhancing activity functions as an IgE differentiation factor for activated B cells. This was further confirmed by the effects of TCSN-A on the IgM- and IgE-secreting EBV-transformed human B cell line K1D5. TCSN-A specifically enhanced IgE synthesis from these cells; TCSN from normal donors, IL 2, IFN-gamma, and BCGF did not. These results confirm that this activity functions as an IgE-specific differentiation factor, directly influencing activated B cells to synthesize IgE.     

Noncognate contact-dependent B cell activation can promote IL-4-dependent in vitro human IgE synthesis

We have previously shown that IL-4 is an essential mediator for the synthesis of human IgE in vitro. In this study we demonstrate that prior physical contact with T cells is required by B cells to synthesize IgE in response to IL-4. Both autologous and allogeneic freshly prepared T cells were consistently able to support IL-4-dependent IgE synthesis, provided that they were added to B cells together with, or before, the addition of IL-4. In addition, most CD4+, as well as a proportion of CD8+, PHA-induced T cell clones (TCC) established from two HLA-DR incompatible donors, supported, in the presence of exogenous IL-4, the synthesis of IgE in B cells from the majority of individuals tested including both donors of cloned T cells. An alloreactive TCC able to produce IL-4 in response to HLA-DR4+ B cells and to induce HLA-DR4+ B cells to synthesize IgE, acquired the ability to support IgE synthesis by B cells lacking the appropriate alloantigen provided that exogenous IL-4 was added. Although the ability of freshly prepared T cells to support IgE synthesis was consistently abrogated by fixation with paraformaldehyde (PF), such a treatment variably affected the IgE-inducing ability of TCC. Preactivation with anti-CD3 before treatment with PF maintained or even enhanced the ability of TCC to support IL-4-dependent IgE synthesis. More importantly, preactivation with anti-CD3, followed by fixation with PF, enabled TCC, apparently devoid of IgE-inducing activity in unfixed condition, to support IL-4-dependent IgE synthesis. Taken together these data suggest that at least two signals are involved in the triggering of human B cells to IgE production: the first is delivered by a T-B cell contact and the second by IL-4. The physical signal delivered by T cells does not necessarily consist of cognate interaction. Non-cognate contact-dependent induction of B cells to IgE synthesis in response to IL-4 appears to be related to molecule(s) distinct from the TCR/CD3 complex, but fully expressed on the membrane of TCR/CD3-activated T cells.     

IL-4 regulates differentiation and proliferation of human precursor B cells

The mechanism by which precursor and pre-B cells undergo differentiation is unclear; however, it is known that growth factors play an important role in this maturation process. The lymphokine, IL-4 has been shown to increase expression of class II Ag on B cells and induce B cell proliferation. In the murine system, IL-4 induced differentiation of precursor B cells into pre-B cells. In order to analyze growth factors on B cell development we have established an in vitro culture system for human bone marrow cells. We found that in the presence of IL-4, normal human precursor and pre-B cells can be induced to differentiate in the absence of cell proliferation with four days of culture. Furthermore, IL-4 depressed proliferation induced by supernatant from a T cell line. The differentiation was measured by an increase in both the number of cytoplasmic mu and surface IgM-positive cells. The effect of IL-4 on precursor and pre-B cell differentiation was detected as soon as 14 h of exposure to the lymphokine in the absence of an adherent feeder layer. These data suggest that IL-4 directly affects the differentiation process of normal human precursor and pre-B cells, and may antagonistically affect cell proliferation.     

IL-21 effects on human IgE production in response to IL-4 or IL-13

In human atopic disease, IgE sensitizes the allergic response, while IgG4 is protective. Because IL-4 and IL-13 trigger switch recombination to both IgE and IgG4, additional agents must regulate the balance between these isotypes to influence susceptibility or tolerance to atopy. In this report, we define in vitro conditions leading to activation or inhibition of human IgE and IgG4 production by IL-21. IL-21 reduced IL-4-driven IgE synthesis by mitogen-stimulated human PBMC. IL-21 inhibition of human IgE production was not a direct effect on B cells, was not seen following B cell activation with IL-13, and was overcome by CD40 ligation. Neither IFN-gamma, IL-10, IL-12, CD40L expression, nor apoptosis was responsible for the inhibitory effect. In contrast, IL-21-stimulated secretion of IgG4 from PBMC. Our findings indicate that IL-21 may influence the production of both human IgE and IgG4, and thus contribute to the regulation of atopic reactions.     

Differential inhibition of T and B cell function in IL-4-dependent IgE production by cyclosporin A and methylprednisolone

The present study examines the role of the immunosuppressive agents methylprednisolone (MPN) and cyclosporin (Cs)A on IL-4-dependent IgE and IgG production. Addition of optimal amounts of IL-4 (100 U/ml) to cultures of tonsil mononuclear cells resulted in a mean increase in IgG production of 175% and in IgE production of 2460%. Frequency analysis of IgE- and IgG-producing B cells, using an ELISA spot assay, showed parallel increases in both Ig production and numbers of Ig-secreting B cells. IgE production was also enhanced by addition of IL-2 (10 U/ml) and maximal IgE production was obtained with a combination of IL-4 and IL-2. MPN (10(-7) M) and CsA (1 microgram/ml) markedly reduced IL4-induced IgE and IgG production as well as numbers of Ig-secreting cells in a dose-dependent fashion. The suppression of Ig production by the cyclosporins was restricted to the immunosuppressive compounds CsA, CsG, and dihydro-CsD, but not the nonimmunosuppressive drug CsH. Delayed addition of CsA revealed that inhibition was maximal when the drug was added during the first 48 h after addition of IL-4 to the culture. Addition of IL-2 (10 U/ml) partially overcame the inhibition induced by CsA. In coculture experiments, in which separated T or B cells were precultured with the drugs and the cells were then combined and further incubated in the presence of IL-4, the suppressive effects of CsA on IgE production were related to pretreatment of the T but not B cells. The maximum inhibiting effects of MPN were similarly observed when the drug was present in the cultures from the beginning, and addition of IL-2 also partially reversed this inhibition. In contrast to the results with CsA, pretreatment of the B but not T cells with MPN-reduced IgE production. These studies demonstrate that IL-4 increases both numbers of IgE-secreting cells as well as IgE production and CsA and MPN differentially affect the responding T and B cells, resulting in inhibition of Ig production.     

IL-12p70-dependent Th1 induction by human B cells requires combined activation with CD40 ligand and CpG DNA

The detection of microbial molecules via Toll-like receptors (TLR) in B cells is not well characterized. In this study, we found that both naive and memory B cells lack TLR4 (receptor for LPS) but express TLR9 (receptor for CpG motifs) and produce IL-6, TNF-alpha, and IL-10 upon stimulation with CpG oligonucleotides (ODN), synthetic mimics of microbial DNA. Consistent with the lack of TLR4, purified B cells failed to respond to LPS. Similar to CpG ODN, CD40 ligand (CD40L) alone induced IL-6, TNF-alpha, and IL-10. Production of these cytokines as well as IgM synthesis was synergistically increased when both CpG ODN and CD40L were combined. Unlike IL-6, TNF-alpha, and IL-10, the Th1 cytokine IL-12p70 was detected only when both CpG ODN and CD40L were present, and its induction was independent of B cell receptor cross-linking. CpG ODN did not increase the capacity of CD40L-activated B cells to induce proliferation of naive T cells. However, B cells activated with CpG ODN and CD40L strongly enhanced IFN-gamma production in developing CD4 T cells via IL-12. Together, these results demonstrate that IL-12p70 production in human B cells is under the dual control of microbial stimulation and T cell help. Our findings provide a molecular basis for the potent adjuvant activity of CpG ODN to support humoral immune responses observed in vivo, and for the limited value of LPS.     

Testosterone acts directly on CD4+ T lymphocytes to increase IL-10 production

Males are less susceptible than females to experimental autoimmune encephalomyelitis and many other autoimmune diseases. Gender differences in cytokine production have been observed in splenocytes of experimental autoimmune encephalomyelitis mice stimulated with myelin proteins and may underlie gender differences in susceptibility. As these differences should not be limited to responses specific for myelin proteins, gender differences in cytokine production upon stimulation with Ab to CD3 were examined, and the mechanisms were delineated. Splenocytes from male mice stimulated with Ab to CD3 produced more IL-10 and IL-4 and less IL-12 than those from female mice. Furthermore, splenocytes from dihydrotestosterone (DHT)-treated female mice produced more IL-10 and less IL-12 than those from placebo-treated female mice, whereas there was no difference in IL-4. IL-12 knockout mice were then used to determine whether changes in IL-10 production were mediated directly by testosterone vs indirectly by changes in IL-12. The results of these experiments favored the first hypothesis, because DHT treatment of female IL-12 knockout mice increased IL-10 production. To begin to delineate the mechanism by which DHT may be acting, the cellular source of IL-10 was determined. At both the RNA and protein levels, IL-10 was produced primarily by CD4+ T lymphocytes. CD4+ T lymphocytes were then shown to express the androgen receptor, raising the possibility that testosterone acts directly on CD4+ T lymphocytes to increase IL-10 production. In vitro experiments demonstrated increased IL-10 production following treatment of CD4+ T lymphocytes with DHT. Thus, testosterone can act directly via androgen receptors on CD4+ T lymphocytes to increase IL-10 gene expression.     

Type 1 fimbriae deliver an LPS- and TLR4-dependent activation signal to CD14-negative cells

Fimbriae target bacteria to different mucosal surfaces and enhance the inflammatory response at these sites. Inflammation may be triggered by the fimbriae themselves or by fimbriae-dependent delivery of other host activating molecules such as lipopolysaccharide (LPS). Although LPS activates systemic inflammation through the CD14 and Toll-like receptor 4 (TLR4) pathways, mechanisms of epithelial cell activation by LPS are not well understood. These cells lack CD14 receptors and are unresponsive to pure LPS, but fimbriated Escherichia coli overcome this refractoriness and trigger epithelial cytokine responses. We now show that type 1 fimbriae can present an LPS- and TLR4-dependent signal to the CD14-negative epithelial cells. Human uroepithelial cells were shown to express TLR4, and type 1 fimbriated E. coli strains triggered an LPS-dependent response in those cells. A similar LPS- and fimbriae-dependent response was observed in the urinary tract of TLR4-proficient mice, but not in TLR4-defective mice. The moderate inflammatory response in the TLR4-defective mice was fimbriae dependent but LPS independent. The results demonstrate that type 1 fimbriae present LPS to CD14-negative cells and that the TLR4 genotype determines this response despite the absence of CD14 on the target cells. The results illustrate how the host "sees" LPS and other microbial products not as purified molecules but as complexes, and that fimbriae determine the molecular context in which LPS is presented to host cells.     

IL-12p70 production by Leishmania major-harboring human dendritic cells is a CD40/CD40 ligand-dependent process

Leishmaniasis, a vector-borne parasitic disease, is transmitted during a sandfly blood meal as the parasite is delivered into the dermis. The parasite displays a unique immune evasion mechanism: prevention of IL-12 production within its host cell, the macrophage (i.e., where it differentiates and multiplies). Given the close proximity of skin dendritic cells (DC) to the site of parasite delivery, their critical role in initiating immune responses and the self-healing nature of Leishmania major (Lm) infection, we examined the interaction between myeloid-derived human DC and Lm metacyclic promastigotes (infectious-stage parasites) to model the early "natural" events of infection. We found that DC can take up Lm and, after this internalization, undergo changes in surface phenotype suggesting "maturation". Despite the intracellular location of the parasite and resultant up-regulation of costimulatory and class II molecules, there was no detectable cytokine release by these Lm-harboring DC. However, using intracellular staining and flow cytometry to analyze cytokine production at the single-cell level, we found that Lm-harboring DC, but not monocytes, produce large amounts of IL-12p70 in a CD40 ligand (CD40L)-dependent manner. Finally, DC generated from mononuclear cells from patients with cutaneous leishmaniasis (Lm), once loaded with live metacyclic promastigotes, were found to reactivate autologous primed T lymphocytes and induce a CD40L-dependent IFN-gamma response. Our results link the required CD40/CD40L interactions for healing with DC-derived IL-12p70 production and provide a mechanism to explain the genesis of a protective T cell-mediated response in the face of local immune evasion within the macrophage at the site of Leishmania delivery.     

IL-7 up-regulates IL-4 production by splenic NK1.1+ and NK1.1- MHC class I-like/CD1-dependent CD4+ T cells.

NK T cells are an unusual subset of T lymphocytes. They express NK1. 1 Ag, are CD1 restricted, and highly skewed toward Vbeta8 for their TCR usage. They express the unique potential to produce large amounts of IL-4 and IFN-gamma immediately upon TCR cross-linking. We previously showed in the thymus that the NK T subset requires IL-7 for its functional maturation. In this study, we analyzed whether IL-7 was capable of regulating the production of IL-4 and IFN-gamma by the discrete NK T subset of CD4+ cells in the periphery. Two hours after injection of IL-7 into mice, or after a 4-h exposure to IL-7 in vitro, IL-4 production by CD4+ cells in response to anti-TCR-alphabeta is markedly increased. In contrast, IFN-gamma production remains essentially unchanged. In beta2-microglobulin- and CD1-deficient mice, which lack NK T cells, IL-7 treatment does not reestablish normal levels of IL-4 by CD4+ T cells. Moreover, we observe that in wild-type mice, the memory phenotype (CD62L-CD44+) CD4+ T cells responsible for IL-4 production are not only NK1.1+ cells, but also NK1.1- cells. This NK1.1-IL-4-producing subset shares three important characteristics with NK T cells: 1) Vbeta8 skewing; 2) CD1 restriction as demonstrated by their absence in CD1-deficient mice and relative overexpression in MHC II null mice; 3) sensitivity to IL-7 in terms of IL-4 production. In conclusion, the present study provides evidence that CD4+MHC class I-like-dependent T cell populations include not only NK1.1+ cells, but also NK1.1- cells, and that these two subsets are biased toward IL-4 production by IL-7.     

CD28 delivers a costimulatory signal involved in antigen-specific IL-2 production by human T cells.

CD4+ T cells require two signals to produce maximal amounts of IL-2, i.e., TCR occupancy and an unidentified APC-derived costimulus. Here we show that this costimulatory signal can be delivered by the T cell molecule CD28. An agonistic anti-CD28 mAb, but not IL-1 and/or IL-6, stimulated T cell proliferation by tetanus toxoid-specific T cells cultured with Ag-pulsed, costimulation-deficient APC. Furthermore, the ability of B cell tumor lines to provide costimulatory signals to purified T cells correlated well with expression of the CD28 ligand B7/BB-1. Finally, like anti-CD28 mAb, autologous human APC appeared to stimulate a cyclosporine A-resistant pathway of T cell activation. Together, these results suggest that the two signals required for IL-2 production by CD4+ T cells can be transduced by the TCR and CD28.     

IL-4-dependent up-regulation of IL-4 receptor expression in murine T and B cells

This study examined mRNA levels and cell surface expression of IL-4 receptor (IL-4R) in murine T and B cells after incubation with IL-4. Northern blot analysis of mRNA levels of T cells isolated from mesenteric lymph nodes and spleen revealed that IL-4 induced a transient augmentation of IL-4R mRNA in a dose-dependent manner. Maximal levels of mRNA were detected as early as 5 h after initiation of culture. These data were complemented by studies examining the cell surface expression of IL-4R using an anti-IL-4R mAb. Resting T and B lymphocytes express IL-4R (T greater than B) and incubation of these cells with exogenous IL-4 increased IL-4R expression to a maximum after 24 h. This effect was abolished after addition of anti-IL-4 antibody. Continuous incubation of T cells in the presence of high concentrations of IL-4 resulted in a down-regulation of IL-4R expression. Addition of the protein synthesis inhibitor cycloheximide blocked the induced increases in IL-4R expression, indicating the requirement for de novo protein synthesis. Both the levels of mRNA and cell surface expression of IL-4R were not affected by addition of exogenous IL-2, and IL-4 regulation of IL-4R expression was not influenced by the immunosuppressive drug cyclosporin A. These data demonstrate that in T and B cells, IL-4 induces a transient up-regulation of IL-4 mRNA levels that is subsequently reflected in increased numbers of IL-4R displayed on the cell surface. This regulation of IL-4R expression by IL-4 provides an important mechanism for amplification of IL-4-dependent activation pathways.     

B and T lymphocyte attenuator-mediated signal transduction provides a potent inhibitory signal to primary human CD4 T cells that can be initiated by multiple phosphotyrosine motifs

The B and T lymphocyte attenuator (BTLA) is a recently identified member of the CD28 family of cell receptors. Initial reports demonstrated that mice deficient in BTLA expression were more susceptible to experimental autoimmune encephalomyelitis, indicating that BTLA was likely to function as a negative regulator of T cell activation. However, cross-linking of BTLA only resulted in a 2-fold reduction of IL-2 production, questioning the potency with which BTLA engagement blocks T cell activation. We established a model in which BTLA signaling could be studied in primary human CD4 T cells. We observed that cross-linking of a chimeric receptor consisting of the murine CD28 extracellular domain and human BTLA cytoplasmic tail potently inhibits IL-2 production and completely suppresses T cell expansion. Mutation of any BTLA tyrosine motifs had no effect on the ability of BTLA to block T cell activation. Only mutation of all four tyrosines rendered the BTLA cytoplasmic tail nonfunctional. We performed structure-function studies to determine which factors recruited to the BTLA cytoplasmic tail correlated with BTLA function. Using pervanadate as a means to phosphorylate the BTLA cytoplasmic tail, we observed both Src homology protein (SHP)-1 and SHP-2 recruitment. However, upon receptor engagement, we observed only SHP-1 recruitment, and mutations that abrogated SHP-1 recruitment did not impair BTLA function. These studies question whether SHP-1 or SHP-2 have any role in BTLA function and caution against the use of pervanadate as means to initiate signal transduction cascades in primary cells.