Inhibition of degranulation and interleukin-6 production in mast cells derived from mice deficient in protein kinase Cβ

The production of mice lacking PKCβ was previously described.15 The mice had a mixed 129/Sv/129/Ola background, with controls derived from litter mates of heterozygous matings. In the current study, PKCβ-deficient mice and control litter mates were killed on the same day. Mouse BMMC were obtained as previously described.18 Briefly, bone marrow cells were obtained from the femurs and tibias of control and PKCβ-deficient 2-month-old mice. We used bone marrow from 1 or 2 mice per flask. Cells from control and PKCβ-deficient mast cells were grown in individual flasks according to the date the mice were killed. The cells were cultured at a starting density of 0.1 × 10⁶ cells/mL at 37°C in a humidified atmosphere containing 5% carbon dioxide (CO₂). The culture medium was RPMI 1640 supplemented with 2 mmol/L of L-glutamine, 2 mmol/L of nonessential amino acids, 100 U/mL of penicillin, 100 μg/mL of streptomycin, 50 μmol/L of β-2-mercaptoethanol, 10% fetal-calf serum, and 100 U/mL of yeast-derived interleukin 3 (IL-3; Genzyme Diagnostics, Cambridge, MA). The nonadherent cells from the bone marrow cultures were transferred every 7 days into fresh IL-3–containing medium after verification that there were similar cell densities for the control and PKCβ-deficient cells. The cultures were maintained for 3 to 5 weeks. Mast cells were identified by toluidine blue staining.

Mice were injected with 10 mL of normal saline intraperitoneally. After gentle abdominal massage, the peritoneal fluid was aspirated, 250 μL of the fluid was centrifuged, and the pellet was resuspended in 1 mL of normal saline. Then, 200 μL of this suspension was spun for 5 minutes in a cytospin centrifuge. The cells were stained by using a modification of the original Enerbac protocol19 that uses 0.5% Alcian blue and 0.3% acetic acid. They were then rinsed with water and counterstained with 0.1% safranin and 0.1% acetic acid.

IgE sensitization was carried out by incubating 2 × 10⁶ cells with 30 μg of monoclonal IgE against dinitrophenol (DNP) (provided by F.-T. Liu, La Jolla, CA) for 2 hours at 37°C in RPMI medium. The cells were washed once with 1 mL of Tyrode buffer and resuspended in the same buffer. Fifty-microliter aliquots of this cell suspension were put into microfuge tubes containing increasing amounts of human serum albumin (HSA) conjugated to DNP (DNP-HSA) or ionomycin (Sigma, St Louis, MO) for 45 minutes. The release of β-hexosaminidase was then determined in triplicate in a 96-well plate. The results were expressed as the net percentage release of hexosaminidase, as previously described.20 

For experiments investigating cytokine mRNA production and release, the cells were preincubated for 4 days with IgE at the same concentration described above and then incubated with 100 ng/mL of HSA-DNP in tissue-culture medium for 8 to 10 hours. This incubation time was based on the work by Zhu et al21 and Marshall et al,22 in which the production of IL-6 in activated mast cells was determined on the level of mRNA and of protein after overnight incubation. The cells were then centrifuged and the supernatants were collected and frozen at −70°C for later cytokine determination. The pellets containing the activated mast cells were used for RNA preparation. The IL-6 secreted into the medium was measured by an enzyme-linked immunosorbent assay kit (Endogen, Woburn, MA).

The proliferation assay used was described previously.17Cells were cultured in 200 μL of medium supplemented with doses of mouse IL-3 defined according to the requirements of each experiment. Cells from each treatment were divided into 4 replicates and were incubated for 24 hours at 37°C in 5% CO₂ in round-bottomed 96-well microtiter plates (Nunc, Denmark) at a density of 2 × 105 cells per well. The cells were labeled with 0.037 MBq of tritium-thymidine for the last 4 hours at 37°C and transferred onto glass-fiber filter paper (Titerek, United Kingdom). Thymidine incorporation was measured as described previously.18 

BMMC were analyzed for apoptosis by using 2-color flow cytometry. The cells were stained with annexin V antibody (Boehringer Mannheim, Germany)23 and then with biotin-labeled antimouse antibody, followed by avidin-phycoerythrin (BD-Pharmingen, San Diego, CA). The cells were also stained with the fluorescent DNA binding agent 7-amino actinomycin D (7AAD), which was used as a substitute for propidium iodide to mark the cells in the last phase of apoptosis.24The percentage of cell survival was calculated by subtracting the number of cells stained with 7AAD, which stains DNA in dead cells, from the total cell count.

Cell pellets were lysed directly with 1 mL of Tri-Reagen, and total RNA was isolated and purified according to the manufacturer's instructions. The following primer pairs were designed from the mouse complementary DNA sequences available in GenBank: for IL-6, 5′-CTGGTGACAACCACGGCCTCCCCT-3′ and the 3′ primer 5′-TGGATCACGCAATACGGATTCGTA-3′, which amplify a 620 base pair (bp); for interleukin 10 (IL-10), 5′-CCAGTTTTACCTGGTAGAAGTGATG-3′ and the 3′ primer 5′-TGTCTAGGTCCTGGAGTCCAGCAGACTCAA-3′, which amplify a 440 bp; for β-actin, 5′-TGGAATCCTGTGGCATCCATGAAAC-3′ and the 3′ primer 5′-GCCTGACAATGACTCGACGCAAA-3′, which amplify a 348 bp; and for JunD, 5′-ATGGAAACGCCCTTCTATGGC-3′ and the 3′ primer 5′-CGTTCTTGCGTGTCCATGTCG-3′, which amplify a 783 bp.

Reverse transcription and polymerase chain reaction (PCR) were done by using the Titan one-tube RT-PCR System (Boehringer Mannheim). The linear range of amplification was determined by measuring gene expression after 25 to 35 cycles (nonsaturated PCR product). After the PCR reaction, loading buffer was added to each 5-μL reaction and aliquots were resolved on 2% agarose gel containing 0.008% ethidium bromide. DNA molecular-weight standards were run in parallel.

Peritoneal mast cells are different in many respects from cultured BMMC. For instance, they contain heparin in their granules that is stained strongly by safranin, whereas BMMC contain chondroitin sulfate E in their granules that is stained by Alcian blue.19 In vitro, BMMC can differentiate into heparin-containing mast cells if coincubated with fibroblasts or c-kit ligand.25 26Thus, primary mast cell cultures like BMMC are now considered precursors of heparin-containing mast cells. If heparin-containing mast cells are dependent on PKCβ for their differentiation, one would not expect to see peritoneal mast cells from PKCβ-deficient mice stained with safranin. However, we found that the peritoneal mast cells in cytospin preparations from mouse peritoneum from both control and PKCβ-deficient mice were stained with safranin (Figure1A and Figure 1B). Hence, PKCβ-deficient mice, in contrast to mouse strains such as c-kit-deficient mice, contain large numbers of mast cells that can differentiate into heparin-containing mast cells.

Our initial assessment of BMMC cell growth was by cell count after 2, 3, and 4 weeks of cell culture. At this stage, most of the cells were mast cells (as determined by toluidine blue staining). BMMC cell density increased in similar proportions in control and PKCβ-deficient mast cells (data not shown). We then measured cell proliferation in 3 individual experiments. BMMC derived from wild-type and from PKCβ-deficient mice were incubated for 48 hours in the presence of increasing concentrations of IL-3 and were assessed for their rate of proliferation by tritium-thymidine incorporation into DNA and cell number (Figure 2). A similar pattern of IL-3 dose-dependent proliferation rate was observed in these 2 types of BMMC. Thus, the deficiency of PKCβ in these cells did not affect their rate of proliferation.

The effect of IL-3 deprivation on apoptosis of BMMC derived from the PKCβ-deficient mice was also assessed in 3 separate experiments. These experiments were carried out with 2 different staining reagents: annexin V, which stains phosphatidylserine and thus determines an early event in apoptosis23; and 7AAD, which stains DNA and so reveals apoptotic cells at a later stage.20PKCβ-deficient and wild-type mast cells were incubated with or without IL-3 for 18 hours and then stained with annexin V and 7AAD. As shown in Figure 3A and Figure 3B, the rates of apoptosis after IL-3 deprivation were similar in PKCβ-deficient and control BMMC.

Mitomycin C is a chemotherapeutic agent that has been shown to induce apoptosis in a variety of cells. Therefore, the effect of mitomycin C on both wild-type and PKCβ-deficient BMMC was determined. Cells were grown in medium containing a suboptimal concentration of IL-3 (20 U/mL) that was found to prevent BMMC apoptosis and 30 μmol/L of mitomycin C. No significant difference in the apoptotic rate was observed between wild-type and PKCβ-deficient BMMC (Figure 3C). Thus, deficiency of PKCβ in mast cells did not affect their rate of proliferation or apoptosis.

Wild-type and PKCβ-deficient BMMC were sensitized with monoclonal mouse IgE against HSA-DNP and challenged with increasing doses of HSA-DNP–released β-hexosaminidase (Figure4). In 4 consecutive experiments, degranulation was 52% ± 6% (mean ± SE) lower in BMMC derived from the PKCβ-deficient mice than in wild-type BMMC.

In 3 separate experiments, cells were incubated for 20 minutes at 37°C with increasing concentrations of ionomycin. Degranulation was 75% ± 7% (mean ± SE) lower in the PKCβ-deficient BMMC than in the wild-type BMMC (Figure 5).

Our initial experiments to assess cytokine release from BMMC failed to detect important amounts of cytokines in the medium, probably because the preincubation period with IgE was only 2 hours. Because several articles reported up-regulation of IgE receptors by IgE,27-29 we preincubated the cells for 4 days with IgE before stimulating them with the antigen for 8 to 10 hours. The culture medium was then collected and the mRNA was extracted from the cells.

Because it was demonstrated that IL-10 is induced in human mast cells by IgE-Ag stimulation,30 we investigated IL-10 mRNA accumulation in PKCβ-deficient and wild-type BMMC with and without IgE-Ag activation, with β-actin as the control. The level of induction of IL-10 was similar in PKCβ-deficient and wild-type cells (Figure 6A), showing that there is probably no general defect in IgE-Ag induction in PKCβ-deficient BMMC.

Overexpression of PKCβ in RBL cells induces the accumulation of IL-6 mRNA.13 Therefore, we measured the accumulation of IL-6 mRNA in IgE-Ag–activated PKCβ-deficient BMMC. Figure 6B shows that the mRNA level of IL-6 was significantly reduced in the PKCβ-deficient BMMC compared with the wild-type cells.

The homodimer of JunD/JunD has been suggested to be a regulator of IL-6 gene expression.14 We showed that, in mast cells, the expression of JunD on the level of mRNA is dependent on PKC.5 We therefore decided to determine whether the level of JunD mRNA accumulation is different in BMMC derived from PKCβ-deficient mice compared with wild-type BMMC. As shown in Figure6C, compared with the findings in wild-type BMMC, there was indication of a reduction in the level of JunD mRNA in PKCβ-deficient BMMC after activation.

To determine whether the PKCβ deficiency in BMMC also inhibited IL-6 production, we measured the FcεRI-dependent synthesis of IL-6 in these BMMC (Figure 7). No IL-6 protein was detected in the resting BMMC. However, the average results from 3 separate experiments showed that IL-6 was significantly reduced in the PKCβ-deficient BMMC (381 ± 66 pg/mL per 10⁶cells) compared with the wild-type BMMC (769 ± 154 pg/mL per 10⁶ cells). Thus, the PKCβ deficiency in these cells significantly affected their capacity to produce IL-6.

The finding of peritoneal mast cells stained with safranin in peritoneal exudates from PKCβ-deficient mast cells shows that PKCβ-deficient mice contain mast cells that can differentiate into heparin-containing cells. Because no previous studies indicated that PKCβ is essential for mast cell proliferation or differentiation, this result was not unexpected, and we therefore focused on the analysis of the specific functions of mast cells by using BMMC.

Our hypotheses regarding specific differences between control and PKCβ-deficient mast cells were based on 2 kinds of evidence—that obtained by using mice lacking in Btk, which display an immunophenotype very similar to PKCβ-deficient mice15; and that obtained with in vitro cellular studies. The similarity between deficiency of Btk and deficiency of PKCβ has been shown in studies in knockout mice. One explanation for the similarity in phenotype could be an association between Btk and PKCβ. However, since signal-transduction pathways can vary substantially between different cell types, finding a difference in the phenotype of these 2 mutant cell types was not surprising. Our observation that PKCβ-deficient mast cells undergo apoptosis at the same rate as control cells in response to growth-factor deprivation is in contrast to the situation in Btk-deficient mast cells. Btk-deficient mast cells undergo apoptosis at a lower rate than control mast cells in response to growth-factor deprivation.16 Several explanations for the difference in phenotype between Btk-deficient and PKCβ-deficient mast cells could be postulated, such as the use of a different kind of PKC for pathways involving Btk for the induction of apoptosis in PKCβ-deficient mast cells. Additional work is needed to explain the differences in the phenotypes.

Most chemotherapeutic agents cause cell death by inducing apoptosis. Interestingly, safingol, a PKC inhibitor, was shown to greatly enhance mitomycin C-induced apoptosis of gastric tumor cells.31Others reported that PKCβ inhibition reduced the apoptotic rate of B lymphocytes.32 We therefore tried to assess the chemotherapy-induced apoptotic response of PKCβ-deficient mast cells. However, even in this context, we observed no significant difference between PKCβ-deficient BMMC and wild-type BMMC in the mitomycin C-induced death rate.

Our observations that the total number of mast cells grown with IL-3 and the proliferation rates were not different in control and PKCβ-deficient mast cells (Figure 2) indicate that PKCβ has a nonessential role (if any) in mast cell growth. This is in disagreement with the evidence from studies in a rat mast cell line that overexpressed PKCβ and had an increase in cellular proliferation rate.13 Unless these results are due to dissimilarities of the mast cell subtypes assessed in the different studies, it seems that although an abnormal signal from PKCβ can increase cellular proliferation rate, this enzyme does not have an essential role in this process. In contrast to this discrepancy, our finding of a substantial decrease in mast cell degranulation in PKCβ-deficient mast cells compared with control cells (Figure 4) is in agreement with results from reconstitution and overexpression experiments. Although PKCδ could also allow degranulation in reconstitution experiments, it could not totally replace the absence of PKCβ in cells lacking this enzyme. Hence, PKCβ has an essential role in degranulation.

Because ionomycin was unable to rescue the degranulation defect in the PKCβ-deficient mice, it appears that PKCβ might function further downstream of the FcεRI receptor itself. It has been shown that calcium ionophore-induced secretion from mast cells correlates with myosin light-chain phosphorylation by PKC33; thus, myosin light-chain might present a target for PKCβ. Alternatively, it has been reported that PKC is involved in the regulation of coat protein assembly in mast cells and therefore in the modulation of secretion.34 Additional experiments are needed to elucidate the physiologic target of PKCβ in mast cells.

IgE-Ag induction of 2 cytokine mRNAs, IL-6 and IL-2, was increased in mast cells that overexpressed PKCβ.13 We detected a decrease in IL-6 released by activated PKCβ-deficient mast cells compared with control cells. An even greater decrease in the amount of IL-6 mRNA was observed in activated PKCβ-deficient mast cells compared with control cells. It is almost impossible to determine whether the decrease in the level of IL-6 protein in mast cells activated overnight was due to a decay in the level of IL-6 mRNA. Additional work must be performed to find out whether the control of IL-6 synthesis is on the transcriptional, translation, or mRNA accumulation level. Experiments to investigate calcium flux in these cells would be of great interest because degranulation is dependent on calcium, and IL-6 production is probably dependent on calcium by means of the activation of an NF-AT or NF-AT–like factor, although a classical NF-AT site is not present in the IL-6 promoter.35Interestingly, there was no significant change in the amount of mRNA for IL-10 between the different cell types, which suggests that there is no general defect in IgE-receptor signal transduction in these cells. In contrast, mRNA levels of JunD, which has been shown to be a regulator of the IL-6 gene,14 were lower in activated PKCβ-deficient mast cells than in wild-type cells. This suggests that JunD induction by PKCβ is important for IgE-Ag induction of the IL-6 gene. In previous work, we demonstrated that JunD mRNA accumulation is dependent on PKC.5 We also showed that the PKCβ isozyme is the predominant PKC isozyme in BMMC.8 Therefore, it is not surprising that in this study we found indication that PKCβ-deficient BMMC had lower levels of IgE-induced JunD mRNA.

Thus, our study not only supports the main conclusions from reconstitution and overexpression experiments regarding the role of PKCβ in signal transduction leading to degranulation and cytokine induction, but it also shows that under normal conditions the role of this enzyme is not redundant and cannot be totally replaced by other PKC isozymes. Specific PKC isozyme inhibitors are currently being formulated. Studies using different PKC knockout mice should prove to be invaluable not only for assessing the possible effects of such inhibitors but also for increasing our understanding of the complex role of different PKC isozymes in signal transduction in different cell types.

We thank the members of the A. Tarakhovsky's laboratory, especially I. Mecklenbrauker.