Crucial Role of Phospholipase Cε in Induction of Local Skin Inflammatory Reactions in the Elicitation Stage of Allergic Contact Hypersensitivity

Contact hypersensitivity (CHS) is a form of T cell-mediated cutaneous immune and inflammatory reactions induced by repeated exposure to the chemical substance, hapten. The development of CHS is composed of two stages: induction/sensitization and elicitation (1). In the elicitation/sensitization stage, a hapten penetrates the skin, where it combines with protein present in the epidermis to form a haptenated peptide. The haptenated peptide can then function as a complete Ag and is subsequently transported to the local draining lymph nodes, where an immune response is stimulated. The elicitation stage can be initiated by a subsequent challenge of the sensitized animal with the same hapten. The haptenated peptide is presented by APC, such as epidermal Langerhans cells, and hapten-specific T cells are selectively recruited to the extravascular space at the challenged sites within 2 h (2). The recruited T cells start to produce cytokines, such as IL-17 and IFN-γ, which stimulate skin cells, including keratinocytes and fibroblasts, to express cytokines and cell adhesion molecules that facilitate leukocyte recruitment (3, 4). Through these actions, T cells elicit cutaneous inflammatory reaction phase that peaks after ∼24 h (2). The dermatitis in this stage is characterized by edema, itching, redness, pain, rash, etc. Because it can be easily reproduced in animals, CHS serves as a good experimental model for studying allergy and autoimmunity.

Phosphoinositide-specific phospholipase C (PLC) plays fundamental roles in intracellular signaling by hydrolyzing phosphatidylinositol 4,5-bisphosphate to 1,2-diacylglycerol and inositol 1,4,5-trisphosphate. 1,2-diacylglycerol binds to a variety of the target proteins, including protein kinase C (PKC) isozymes and Ras guanyl nucleotide-releasing proteins (RasGRPs), and regulates their activities. Inositol 1,4,5-trisphosphate induces the release of calcium ion from the intracellular stores, leading to an increase in the cytosolic-free calcium concentration. At least 13 mammalian PLC isoforms have been identified and organized into six classes (β, γ, δ, ε, ζ, and η) (5, 6). PLCε was identified as an effector of Ras-family small GTPases: Ras, Rap1, and Rap2 (7–10). Similar to other effectors, such as cRaf-1 and B-Raf, PLCε is activated by direct association with the GTP-bound form of the small GTPases through its Ras/Rap-associating domain (7–10). Subsequent studies have demonstrated that PLCε is also activated by another small GTPase RhoA and heterotrimeric G proteins α₁₂ and β₁γ₂ subunits (10).

We showed that PLCε^−/− mice, in which PLCε was catalytically inactivated by gene targeting, exhibited marked resistance to tumor formation in the two-stage skin chemical carcinogenesis protocol using 7,12-dimethylbenz(a)anthracene as an initiator and a phorbol ester PMA as a promoter (11). Subsequent studies showed that PMA-induced skin inflammation, which is intimately involved in tumor promotion, was attenuated in PLCε^−/− mice (12). Using primary-cultured dermal fibroblasts prepared from PLCε^−/− mice, we showed that PMA treatment gives rise to the activation of PLCε, which is mediated by Rap1 activation induced by the direct activation of two PMA targets, RasGRP3 and PKC, thereby inducing the expression of proinflammatory molecules, such as IL-1α (12). Recently, we have demonstrated that PLCε^−/− mice also exhibit marked resistance to Apc^Min-dependent intestinal tumorigenesis, which appears to be closely associated with the attenuation of tumor-associated inflammation (13). These results suggest a crucial role of PLCε in a wide variety of inflammations. However, little is known about how PLCε functions in inflammatory reactions.

In this study, we use a mouse CHS model in which allergic dermatitis is induced by a hapten 2,4-dinitro-1-fluorobenzene (DNFB). Substantial attenuation of the inflammatory reactions in the elicitation stage of CHS is observed in PLCε^−/− mice. Comparison of the infiltration of CD4⁺ T cells and the expression of various proinflammatory cytokines between PLCε^+/+ and PLCε^−/− mice, combined with the adoptive transfer of CD4⁺ T cells, indicate that PLCε functions in efficient production of proinflammatory cytokines from dermal fibroblast and epidermal keratinocytes in response to T cell-derived cytokines, such as IL-17, IFN-γ, TNF-α, and IL-4.

Mice homozygous for the inactivated PLCε allele (PLCε⁻), created by in-frame deletion of an exon coding for the catalytic X domain, were generated as described previously (11, 14) and backcrossed to C57BL/6JJcl mice (CLEA Japan, Tokyo, Japan) for at least eight generations. Genotyping of the mutant animals was performed by PCR using allele-specific primers (11, 14). The use and care of the animals were reviewed and approved by the Institutional Animal Committee of Kobe University Graduate School of Medicine.

Primary Abs used were as follows: anti-CD4 (550280, BD Pharmingen, San Diego, CA), anti-Gr-1 (MAB1037, R&D Systems, Minneapolis, MN), anti-CXCL-1 (AF-453-NA, R&D Systems), anti-CXCL-2 (AF-452-NA, R&D Systems), anti-CCL-20 (AF760, R&D Systems), anti-IL-1α (AF-400-NA, R&D Systems), anti-IL-1β (AF-401-NA, R&D Systems), and anti-PLCγ1 (sc-81, Santa Cruz Biotechnology, Santa Cruz, CA). Anti-PLCε Ab raised against the C terminus of mouse PLCε was described (15). Secondary Abs labeled with Alexa dye were purchased from Invitrogen (Carlsbad, CA). HRP-conjugated secondary Abs were purchased from GE Healthcare (Buckinghamshire, U.K.). Ligands and chemicals used were TNF-α (315-01, PeproTech, Rocky Hill, NJ), IFN-γ (315-05, PeproTech), IL-4 (112-40-134S, AMS Biotechnology, Abingdon, Oxford, U.K.), IL-17 (421-ML/CF, R&D Systems), DNFB (D1529, Sigma-Aldrich, St. Louis, MO), and PMA (P-8139, Sigma-Aldrich).

A mouse was sensitized with DNFB by painting its shaved dorsal skin with 25 μl 0.4% (w/v) DNFB dissolved in acetone. Five days later, 20 μl 0.2% (w/v) DNFB was applied to the backside of the right ear. The same volume of acetone was applied to the left ear as an unchallenged control. For noninvasive evaluation of the inflammation, ear swelling was calculated by subtracting the thickness of the left ear from that of the right ear after measurement with a pair of calipers (MDC-25MJ, Mitutoyo, Kawasaki, Japan) as described (12).

Blood samples (30 μl) were collected from the retro-orbital venous plexus of 8-wk-old male mice (n = 5 for each PLCε genotype) and diluted in buffer (155 mM NH₄Cl, 10 mM KHCO₃, 0.1 mM EDTA) for hemolysis of erythrocytes. Subsequently, cells were collected by centrifugation, resuspended in 300 μl PBS, loaded on a hematocytometer, and counted as WBCs under a microscope.

CD4⁺ T cells were purified from the spleen of the sensitized animal at 5 d after DNFB sensitization by using mouse CD4⁺ T isolation kit (130-090-860, Miltenyi Biotec, Bergisch Gladbach, Germany) and AutoMACS (Miltenyi Biotec). The purity of the recovered CD4⁺ T cells was estimated >95% by the immunostaining with anti-CD4 Ab (data not shown). The purified CD4⁺ T cells (5 × 10⁶ cells) were suspended in 0.2 ml sterilized PBS and injected i.v. into recipient mice. One hour later, the recipients’ ears were challenged with DNFB or acetone and subjected to the evaluation of the response as described previously.

Tissues were collected from sacrificed animals, fixed in 4% paraformaldehyde, embedded in paraffin or OCT compound, and sectioned as described previously (12). The sections were subjected to H&E staining or immunostaining. For immunostaining, the sections were fixed again in 4% paraformaldehyde (for anti-CD4, anti-PLCε, anti-PLCγ1, and anti-Gr-1 immunostaining) or Zamboni’s Fixative (Muto Pure Chemicals, Tokyo, Japan) for 30 min, washed, and treated with blocking solutions containing IgG. Probing with primary Abs and fluorescent active Alexa-labeled secondary Abs (Invitrogen), nuclear staining with DAPI, and observation of the fluorescence were performed as described (12–15).

Leukocytes and neutrophils were identified on H&E-stained skin sections (Supplemental Fig. 1). Particularly, pink-stained leukocytes with nuclei definitely divided into two to five lobes were identified as neutrophils. The identification of neutrophils was performed also with immunostaining using anti-Gr-1 Ab (Supplemental Fig. 1). CD4⁺ T cells were identified by immunostaining with anti-CD4 Ab. Cells present in the dermal layer of the ear auricles were counted on three randomly-chosen areas (at least 1 mm² in size) of each section. Three animals of each group were examined.

Total cellular RNAs were isolated from cultured cells with Trizol (Invitrogen) and from tissues with RNeasy Mini kit (74134, Qiagen). cDNA synthesis and RT-PCR were performed as described previously (12). Real-time quantitative RT-PCR (QRT-PCR) was carried out using SYBR Premix Ex Taq II kit (Takara Bio, Kyoto, Japan) with Thermal Cycler Dice Real Time System (Takara Bio) as described (13). Relative mRNA levels were determined by the comparative Ct method, followed by normalization with the β-actin mRNA level in each cDNA sample. The sequences of the primers used for RT-PCR and QRT-PCR are listed in Table I (12, 13, 16, 17).

Supernatant was collected from dermal fibroblast culture after stimulation with cytokines for 12 h and subjected to quantification of various cytokines by using ELISA kits purchased from R&D Systems (MLA00 for IL-1α, MLB00B for IL-1β, MM200 for CXCL-2, and MCC200 for CCL-20).

Primary culture was performed as described previously (12) with minor modifications. Briefly, epidermal keratinocytes were isolated from the dorsal skin of newborn mice at postnatal day 1 (P1) and seeded with Defined Keratinocyte-SFM supplemented with growth factors (10744-019, Invitrogen) onto the plastic plates coated with type I-A collagen (Nitta gelatin, Osaka, Japan). Keratinocyte differentiation was induced by exposing cells to the media containing 0.5 mM Ca²⁺ for 2 d. The differentiation was monitored by assessment of the expression of differentiation-specific markers as well as the morphology. Dermal fibroblasts were isolated from the dorsal skin of newborn mice at P1–3 and cultured on ordinary culture plates in DMEM supplemented with 10% FBS.

Preparation of cell lysates, determination of protein concentrations, SDS-PAGE, and immunoblotting were performed as described previously (12, 13).

Cytosolic-free calcium concentration in cultured dermal fibroblasts was monitored in cells loaded with Fura-2 AM (Nacalai Tesque, Kyoto, Japan) as described previously (12). Using F-4500 fluorometer (Hitachi, Tokyo, Japan), changes in the fluorescence intensities of the emissions were measured at 510 nm after excitation at 340 and 380 nm. Results are expressed as relative concentrations of cytosolic-free calcium calculated as described (18).

The conditioned medium from primary-cultured PLCε^+/+ dermal fibroblasts stimulated with 10 ng/ml IL-17 for 12 h was added to fibroblasts. At 0, 3, and 6 h after the addition of the conditioned medium, mRNA levels of various cytokine genes were assessed by QRT-PCR.

Values are expressed as the means ± SD. Unpaired Student t test was performed for determination of p-values using Graphpad Prism software (Graphpad Software). In case the p-values were <0.05, differences were considered to be statistically significant.

To study the role of PLCε in CHS, PLCε^+/+ and PLCε^−/− mice were sensitized by painting their dorsal skins with DNFB. Five days later, they were challenged on the ears with DNFB, and the inflammatory reactions were assessed on their ear sections at 24 h after the DNFB challenge (Fig. 1A and Supplemental Fig. 2). H&E staining of the sections indicated that the ears of PLCε^+/+ mice showed development of strong inflammatory reactions at the challenged sites, which were characterized by marked spongiosis and extensive granulocyte infiltration in the swollen dermis. In contrast, such reactions were substantially attenuated in the ears of PLCε^−/− mice. Examination of the time course of ear swelling indicated that the ear thickness sharply increased until 12 h after the challenge without noticeable difference depending on the PLCε genetic background (Fig. 1B). Thereafter, substantial difference developed depending on the PLCε background: the ear thickness continuously increased until 48 h after the challenge in PLCε^+/+ mice, whereas it exhibited substantial reduction and reached a plateau around 24–36 h in PLCε^−/− mice (Fig. 1B). As the patterns of the ear swelling were found gender-independent (Fig. 1B), additional studies were carried out using male animals only. Next, the time course of the leukocyte infiltration was examined on the H&E-stained ear sections (Fig. 1C). After the DNFB challenge, the number of overall leukocytes present in the challenged site showed a robust increase in a PLCε-independent manner until 6 h. Thereafter, it increased continuously up to 48 h. By contrast, in PLCε^−/− mice, the increase was substantially blunted starting from 12 h and thereafter. We paid particular attention on neutrophils because they play a crucial role in DNFB-induced CHS (19). As examined on the H&E-stained ear sections, neutrophil infiltration was substantially attenuated on the PLCε^−/− background (Fig. 1C), which was confirmed by immunostaining for Gr-1 (Supplemental Fig. 1). These results suggested the involvement of PLCε in the middle to late phases of the elicitation stage of CHS, starting around 12 h after DNFB challenge.

The circulating WBC counts were not affected by the PLCε background [(32.4 ± 4.0) × 10² cells/mm³ in PLCε^+/+ mice versus (34.0 ± 5.3) × 10² cells/mm³ in PLCε^−/− mice; p = 0.382 (n = 5 per group)]. In addition, the PLCε background did not seem to intrinsically influence the function of Gr-1-positive granulocytes because granulocytes do not express a detectable amount of PLCε (12). Thus, the suppression of neutrophil infiltration in PLCε^−/− mice is likely to be accounted for by attenuated expression of proinflammatory cytokines in the elicitation stage. To test this assumption, the levels of cytokine mRNAs at the DNFB-challenged sites were quantified by QRT-PCR (Table I) at the different phases of the elicitation stage, namely, 3 h and 24 h after the DNFB challenge representing the early phase and the late phase, respectively (Fig. 2). The level of IL-1α, whose PLCε-dependent expression had been observed in PMA-induced skin inflammation (12), exhibited significant increase at 3 h, followed by marked increase at 24 h in a PLCε-dependent manner. IL-1β showed a similar expression pattern. Chemokines, CXCL-2, CXCL-1, and CCL-20, showed significant elevation at 3 h, though, in a PLCε-independent manner, followed by marked elevation at 24 h in a PLCε-dependent manner. PLCε-dependent expression of these five cytokines in situ was demonstrated by immunostaining of the DNFB-challenged ears; they showed robust expression at 24 h after the challenge in both epidermal keratinocytes and dermal fibroblasts of PLCε^+/+ mice, which was substantially attenuated in those of PLCε^−/− mice (Fig. 3, Supplemental Fig. 3). These results suggested that the suppression of neutrophil infiltration in PLCε^−/− mice could be accounted for by reduced production of these proinflammatory cytokines. Because production of these cytokines after T cell infiltration is triggered by T cell-derived cytokines, we next examined the expression of such cytokines, IL-17A, IFN-γ, IL-4, and TNF-α, which had been implicated in the development of CHS (20–22) (Fig. 2). The levels of IL-17A, IFN-γ, and IL-4 exhibited a small but significant increase at 3 h, followed by a marked increase at 24 h. The increase of these cytokine levels was found independent of the PLCε background at the both time points, which apparently contradicts with the previously reported data that PLCε was involved in Ras-dependent upregulation of cytokine genes in CD4⁺ T cells (16). Our results were further supported by an analysis of PLCε expression in CD4⁺ T cells, which demonstrated the absence of PLCε at both the protein and mRNA levels in contrast to the abundant presence of PLCγ1 (Fig. 4, Supplemental Figs. 4 and 5). In contrast, TNF-α showed no increase at 3 h, followed by a marked elevation in a PLCε-dependent manner at 24 h. This PLCε-dependent increase may be accounted for by the attenuated infiltration of neutrophils (Fig. 1), another species of TNF-α-producing cells (23), in PLCε^−/− mice. IL-22, another T cell-derived cytokine implicated in the pathogenesis of certain types of skin inflammation (24), was not induced at all within 24 h after the DNFB challenge (Supplemental Fig. 6), indicating that this cytokine does not play a role in the elicitation. These results hinted possible involvement of PLCε in downstream signaling from IL-17A, IFN-γ, TNF-α, or IL-4, which is produced by T cells infiltrated into the challenged site.

During the development of CHS, IL-17 plays a key role in induction of local tissue inflammation (3, 25). The major targets of this cytokine include fibroblasts and keratinocytes (3, 24–26), where PLCε is abundantly present (12) (Fig. 4A). Recent studies demonstrated that one of the major producers of IL-17 is a subclass of CD4⁺ Th cells, called Th17 (3). We therefore asked whether the PLCε background indeed affected the infiltration of CD4⁺ T cells to the DNFB-challenged site (Fig. 5, A–C, Supplemental Fig. 7). The time course of the CD4⁺ T cell infiltration exhibited no difference depending on the PLCε background throughout the experimental periods ranging from 3 to 72 h after challenge as shown by immunostaining (Fig. 5A, 5B, Supplemental Fig. 6). Also, quantification of the CD4 mRNA levels supported this notion (Fig. 5C). These results indicated that CD4⁺ T cell infiltration was not affected by the PLCε background, suggesting that PLCε may mainly function in the inflammatory reactions after the T cell infiltration. To test this possibility, we carried out adoptive transfer experiments of CD4⁺ T cells between PLCε^+/+ and PLCε^−/− mice. At first, CD4⁺ T cells purified from the DNFB-sensitized mice carrying the PLCε^+/+ or PLCε^−/− background were adoptively transferred into naive PLCε^+/+ mice, followed by DNFB challenge of the recipients’ ears. As expected, the PLCε background of the donor mice did not affect the development of the inflammatory reactions (Fig. 5D). In contrast, when CD4⁺ T cells from the DNFB-sensitized PLCε^+/+ mice were transferred into naive mice carrying the PLCε^+/+ or PLCε^−/− background, the PLCε^−/− recipient mice developed substantially attenuated inflammatory reactions compared with the PLCε^+/+ recipients (Fig. 5E). These results implied that PLCε exerts its function in cells other than CD4⁺ T cells, presumably, fibroblasts or keratinocytes of the skin, to augment inflammatory reactions during the elicitation stage, but not the sensitization stage, of CHS.

The results so far indicated the importance of fibroblasts or keratinocytes of the skin as the cellular site of the PLCε action, which presumably involves expression of proinflammatory cytokines in response to T cell-derived signals. Thus, we examined whether T cell-derived cytokines implicated in CHS, such as IL-17, IFN-γ, TNF-α, and IL-4 (20–22), could induce proinflammatory cytokine expression from fibroblasts or keratinocytes in a PLCε-dependent manner. To this end, primary cultures of fibroblasts (Fig. 6) and keratinocytes (Fig. 7) established from PLCε^+/+ and PLCε^−/− mice skins were stimulated by each of the four cytokines for 12 h and analyzed for the production of a variety of proinflammatory cytokines by QRT-PCR (Figs. 6A, 7B) and ELISA (Figs. 6B, 7C). In fibroblasts, IL-1α was found to be efficiently induced by IL-17 stimulation, which showed clear dependence on the PLCε background. Likewise, CXCL-2 was induced by IFN-γ stimulation, whereas CCL-20 was induced by TNF-α stimulation, both of which exhibited significant PLCε-dependence. Further, certain combinations of these cytokines yielded more efficient induction of IL-1α and CXCL-2 in a PLCε-dependent manner. Under our experimental conditions, induction of IL-1β was not detected in fibroblasts (data not shown). In contrast, keratinocytes cultured in the low-calcium media, which were almost negative for keratin 1 (Fig. 7A), failed to show PLCε-dependent induction of any of the cytokines tested on stimulation with IL-17, IFN-γ, TNF-α, and IL-4 (data not shown). However, when keratinocytes were induced to differentiate in the high-calcium media containing 0.5 mM Ca²⁺ and became highly positive for keratin 10 and keratin 1 (Fig. 7A), marked production of IL-1β was observed on stimulation with the combination of IL-17 and TNF-α or of IL-17 and IL-4 in a PLCε-dependent manner (Fig. 7B, 7C), suggesting that the differentiation status of keratinocytes may affect this induction. Stimulation with the combination of IL-17 and IFN-γ or with IL-17, IFN-γ, TNF-α, or IL-4 alone failed to cause PLCε-dependent induction of cytokines such as IL-1α in differentiated keratinocytes (Supplemental Fig. 8). These results indicated that PLCε plays a crucial role in induction of proinflammatory cytokines at the challenged sites, where infiltrated CD4⁺ T cells produce their intrinsic cytokines, thereby augmenting local inflammatory reactions.

To gain mechanistic insights into the role of PLCε in cytokine induction, we examined the kinetics of the cytokine expression from cultured skin cells by measuring mRNA levels at 3 and 12 h after stimulation with T cell-derived cytokines, representing the early and middle phases, respectively, of the elicitation stage (Fig. 8A, 8B). As a control, we first tested the induction of IL-1α and CXCL-2 from dermal fibroblasts on stimulation with PMA, which had been shown to occur through activation of PLCε in a manner dependent on both PKC and RasGRP3 (12). The IL-1α level showed robust elevation in a PLCε-dependent manner at 3 h and subsequently returned to almost the basal levels by 12 h. CXCL-2 showed a similar expression pattern although the level and PLCε-dependence of the induction were less apparent. In contrast, CCL-20 failed to show any induction. These data were consistent with those reported previously (12). In contrast, stimulation with IL-17 resulted in modest elevation of IL-1α at 3 h, which was independent of the PLCε background. After that, the IL-1α level exhibited a marked increase at 12 h with clear dependence on the PLCε background. Likewise, CXCL-2 and CCL-20 also showed a PLCε-dependent increase at 12 h after IL-17 administration. Similar time courses were observed with both the TNF-α–stimulated induction of CCL-20 and the IFN-γ–stimulated induction of CXCL-2. In the case of the IL-4–stimulated CCL-20 induction, the CCL-20 level showed increase only at 12 h, which was dependent on the PLCε background. We next carried out similar experiments with differentiated keratinocytes (Fig. 8B). We chose IL-1β as an output because IL-1β, but not IL-1α, was found to be induced in a PLCε-dependent manner (Supplemental Fig. 8). Stimulation with the combination of IL-17 and TNF-α or of IL-17 and IL-4 resulted in robust elevation of IL-1β at 12 h, which showed clear dependence on the PLCε background. No elevation was observed at 3 h. These results demonstrated delayed kinetics of PLCε-dependent cytokine induction by T cell-derived cytokines compared with that by PMA, hinting that the T cell-derived cytokines may not correspond to the upstream ligand directly coupled to PLCε.

To test this possibility, we examined the activation of PLCε in cultured dermal fibroblasts upon stimulation with PMA or the T cell-derived cytokines by monitoring elevation of the cytosolic-free calcium concentration (Fig. 8C). As reported (12), PMA stimulation induced rapid and sharp elevation of the calcium concentration. In contrast, stimulation with IL-17 failed to induce even slight elevation of the calcium concentration. Essentially, similar results were obtained with stimulation by TNF-α, IL-4, or IFN-γ (data not shown).

The results described previously hinted the existence of an identified factor (factors), which is (are) secreted from skin cells on stimulation by the T cell-derived cytokines and induces (induce) expression of proinflammatory cytokines from skin cells through PLCε activation. To test this possibility, we examined the induction of the IL-1α, CXCL-2, and CCL-20 mRNAs in cultured dermal fibroblasts carrying the different PLCε backgrounds by treatment with the conditioned medium from IL-17–stimulated dermal fibroblasts (Fig. 9). After the addition of the conditioned medium, PLCε-dependent induction of CXCL-2 and CCL-20 was detected at 3 h (Fig. 9), which was much earlier than the time triggered by the direct IL-17 stimulation (Fig. 8A). In contrast, the induction of IL-1α failed to show significant PLCε-dependency at 3 h after the addition of the conditioned medium (Fig. 9). Such difference in the kinetics of cytokine induction suggests that more than one soluble factor secreted from IL-17–activated cells is involved in the cytokine gene activation via receptor engagement (Fig. 10).

In the current study, we have shown that PLCε plays a crucial role in induction of the inflammatory reactions in the elicitation stage of CHS. Our observation that the extents and kinetics of CD4⁺ T cell infiltration were not affected by the PLCε background (Fig. 5, A–C) and that PLCε is not expressed in CD4⁺ T cells (Fig. 4) indicated that PLCε exerts its function in cells other than CD4⁺ T cells, presumably, fibroblasts or keratinocytes of the skin, to augment inflammatory reactions. This was supported by the results of the adoptive transfer experiments of CD4⁺ T cells from DNFB-sensitized animals (Fig. 5D, 5E). Expression of proinflammatory cytokines, such as IL-1α, IL-1β, CXCL-1, CXCL-2, and CCL-20, in both epidermal keratinocytes and dermal fibroblasts was markedly attenuated in PLCε^−/− mice as assessed by QRT-PCR and immunostaining (Figs. 2, 3). Given that these cytokines have pleiotropic functions such as upregulation of the expression of cell adhesion molecules (27) and chemoattracting activity toward granulocytes and immature dendritic cells (28–30), the PLCε’s function in augmenting inflammatory reactions is possibly mediated by the induction of their expression. In addition, experiments with primary cultures of epidermal keratinocytes and dermal fibroblasts showed that T cell-derived cytokines, such as IL-17A, IFN-γ, TNF-α, and IL-4, alone or in combination, were capable of inducing expression of proinflammatory cytokines, such as IL-1α, IL-1β, CXCL-2, and CCL-20, in a PLCε-dependent manner (Figs. 6, 7). These results taken together imply that PLCε plays a crucial role in induction of proinflammatory cytokines from keratinocytes and fibroblasts at the challenged sites, where infiltrated CD4⁺ T cells produce their intrinsic cytokines, thereby augmenting local inflammatory reactions. Such a function of PLCε in inflammation is unique among the PLC isozymes because PLC isozymes of other classes were reported to contribute to inflammation through distinct ways; PLCγ2, which is most highly expressed in immune cells, regulates activation of immune cells, thereby inducing immune inflammatory reactions (31, 32) and PLCδ1 negatively regulates expression of proinflammatory cytokines such as IL-1β in keratinocytes (31, 33).

As for the mechanism underlying the PLCε’s action, the delayed kinetics of the proinflammatory cytokine induction by the T cell-derived cytokines (Fig. 8A, 8B), the failure of the T cell-derived cytokines in inducing PLCε activation in dermal fibroblasts (Fig. 8C), and difference in the kinetics of the cytokine induction in cells treated with the conditioned medium from IL-17–stimulated cells (Fig. 9) suggested that the cytokine-inducing function of the T cell-derived cytokines is mediated by soluble factors, which are produced by fibroblasts or/and keratinocytes in a PLCε-independent manner on stimulation with the T cell-derived cytokines (Fig. 10). An example of delayed activation of an intracellular signaling pathway probably mediated by the production of secondary factors was reported in the case of IL-17–stimulated ERK activation in mouse embryonic fibroblasts mediated by Ras and/or Rap signaling, where ERK was markedly activated by 24 h IL-17 treatment but not by short-term (∼10–30 min) treatment (34). Further studies including the identification of such factor(s) will be needed to clarify the whole picture of the signaling pathway(s) mediated by PLCε leading to cytokine gene activation.

It should be noted that some discrepancies exist between the expression patterns of the proinflammatory cytokines observed in vivo and in vitro. Although fibroblasts and keratinocytes in culture showed PLCε-dependent expression of different sets of the cytokines, IL-1α, CXCL-2, and CCL-20, versus IL-1β, on stimulation with the T cell-derived cytokines (Figs. 6–8), these cells in vivo almost equally produced all the four cytokines in a PLCε-dependent manner (Fig. 3). Also, the differentiation status-dependent expression of IL-1β in cultured keratinocytes (Fig. 7) was apparently not observed in vivo (Fig. 3). These discrepancies could be explained by the involvement of other intercellular molecules not examined in the current study, which are produced from skin constituents excluding T cells and stimulate fibroblasts or keratinocytes to induce PLCε-dependent expression of certain cytokines missing in the in vitro study.

It is of particular interest that IL-17 is capable of inducing the cytokine expression in a PLCε-dependent manner because IL-17 is a pleiotropic proinflammatory cytokine having a wide array of target cells (3, 35) and implicated in the development of a variety of inflammatory diseases, including not only CHS, but also rheumatoid arthritis (36), psoriasis (37, 38), inflammatory bowel disease (39), and asthma (40). Also, the observation that TNF-α induced the proinflammatory cytokine expression in a PLCε-dependent manner is reminiscent of the crucial roles of PLCε in skin chemical carcinogenesis (11, 41–43) and Apc^Min/+-dependent intestinal tumorigenesis (13, 42, 43). The roles of PLCε in various inflammatory diseases are under investigation in our laboratory. The results of these studies as well as the current one may lay the solid foundation in introducing PLCε as a promising molecular target for the development of novel anti-inflammatory drugs.

We thank Dr. Toshio Imai (KAN Research Institute, Kobe, Japan) and our laboratory members, particularly Dr. Mingzhen Li and Kemmei Ikuta, for helpful discussions and suggestions.

Disclosures The authors have no financial conflicts of interest.