MATERIALS and METHODS

Mice

Adult mice, aged 6–12wk, of the BALB/C, C57BL/6, C3H/HeN, C3H/HeJ, WBB6F1-W/W^v, WBB6F1-+/+, WCB6F1-Sl/Sl^d, WCB6F1-+/+ strains were used. These mice were either produced and maintained in our domestic colony or purchased from Jackson Laboratories (Bar Harbor, ME). Experimental procedures were carried out with the animals under general anesthesia, achieved by intraperitoneal injection of ketamine and xylazine. Each control or experimental group consisted of five mice. Each Experimental protocol was repeated at least twice with similar results.

Reagents

2,4-dinitro-1fluorobenzene (DNFB), oxazolone, dexametha- sone (DX), and ethylenediamine tetraacetic acid were purchased from Sigma (St. Louis, MO). IgE anti-dinitrophenyl (DNP) and DNP-human serum albumin (HSA) were purchased from Sigma. IgE anti-dansyl was purchased from Pharmingen (San Diego, CA). TNF- and anti-TNF- antibodies were purchased from Genzyme (Cambridge, MA).

UVR and TNF- treatment

Shaved abdominal skin was exposed to UVB from a bank of four FS-20 fluorescent lamps with a tube to a target distance of 46cm as previously described (^{Yoshikawa & Streilein 1990}). These bulbs have a broad emission spectrum (250–400nm) and high output primarily in the UVB range (290–320nm). As measured by an IL700 radiometer with a SEE 240 UVB photodetector, these lamps delivered an average flux of 1.4 per m² per s. Mice were exposed to a single dose or four consecutive daily doses of UVR (400J per m²) on the shaved abdominal cutaneous surface prior to sensitization. In some cases, 200l of TNF- (3 10⁴U) or PBS was injected intradermally into shaved abdominal skin prior to sensitization as reported previously (^{Yoshikawa et al. 1992}).

Sensitization and elicitation of contact sensitivity

After mice were anesthetized, 25l of 0.004% DNFB or 25l of 2% oxazolone in acetone was carefully applied to shaved abdominal skin on day 0. Five days later, in order to elicit CH, the dorsal surface of the right ears was challenged with 20l of 0.05% DNFB or 0.5% oxazolone. In some cases anti-TNF- or rabbit anti-bovine serum albumin antibodies (2 10⁴U) were injected i.p. 6h prior to sensitization. The extent of ear swelling was used as a measure of CH. Unsensitized animals were also challenged and the ear swelling measurement gives the global background irritant value. Twenty-four hours after antigen challenge, the degree of ear thickness of both left and right ears was measured using an engineer's micrometer (Mitutoyo, Japan) and the right ear value (the experimental value) was compared with ear thickness prior to challenge (0h) and to left ear thickness (used as background for each individual animal). Final value (expressed inm) = (right ear value 24h -right ear value 0h)– (left ear value 24h -left ear value 0h).

IgE-triggered mast cell degranulation

Two hundred microliters of IgE anti-DNP or IgE anti-dansyl at 1g perml was injected i.d. at day 0 as described (^{Gordon & Galli 1991}). Then at day 1, mice received an i.v. injection of DNP_30-40–HSA 100g in 100l of HEPES-Hanks' balanced salt solution 1% Evan's blue dye, which permitted visual localization of the cutaneous site with increased vascular permeability. This phenomenon was observed as soon as 10min following the injection of DNP–HSA.

DX treatment

Mice received a daily i.p. injection of DX (30mg per kg) for a total of 3d as described previously (^{Wershil et al. 1995}). One hour after the last DX injection the mice were exposed to UVR.

Isolation of mast cells and activation in vitro

Primary cultures of interleukin (IL)-3-dependent, bone marrow-derived cultured mast cells were prepared from BALB/C and B6 mice as described (^{Nakano et al. 1985}). After 3–4wk more than 90% of nonadherent cells stained positively with neutral red. The number of mast cells obtained from both BALB/C and C57BL/6 mice were the same in all experiments. For FcRI-dependent activation, mast cells were sensitized using a monoclonal IgE anti-DNP antibody at 5ng perl for 3h. After two washes, the cells were stimulated with DNP_30-40–HSA at 100ng perml. Supernatants were harvested and frozen at -20°C for further assay of TNF- content.

TNF- and IL-4 enzyme-linked immunosorbent assays

Ninety-six well plates were coated with purified coating anti-TNF- or anti-IL-4 antibodies (Pharmingen) at 2g perml in 50l of carbonate buffer. After a blocking step with bovine serum albumin, samples were added in duplicate and incubated for 9h at 4°C. Following four washes, detecting biotinylated anti-TNF- or anti-IL-4 antibodies (Pharmingen), different from coating antibodies, were then added at 2g perml for 45min at room temperature, washed and avidin–alkaline phosphatase (Sigma) at 1/400 was added in each well for 30min at room temperature. Finally, Substrate (Sigma) was distributed in each well and optical density was measured at 405nm. The limit of detection of both assay was about 10pg perml using as standard TNF- and IL-4 provided by Pharmingen.

Preparation of fresh Langerhans cells and allogeneic mixed epidermal lymphocyte reaction

The epidermis was separated from the dermis after incubation of ventral and dorsal skin sheets in PBS containing 0.25% trypsin for 1h at 37°C. The epidermis was disaggregated after an additional 10min incubation in 0.25% trypsin using a 5ml syringe and filtered through a 75mm nylon cell strainer. Enrichment for Ia⁺ cells was performed by running the cells on 3ml of Accu-prep Lymphocytes (Accurate Chemical and Scientific, Westbury, NJ). The cells localized at the interphase layer were recovered and referred to as fresh Langerhans cells. The percentage of recovered Ia⁺ cells was between 10 and 15% for all strains of mice used. T cells were prepared and purified from lymph nodes. Red blood cells were lysed using Gey's solution and accessory cells were depleted by passing the cell suspension through a mouse T cell enrichment column (R&D Systems, Minneapolis, MN). The percentage of enriched T cells was between 93 and 96% for all strains used. Allogeneic T cells (3 10⁵ per well) were cultured with irradiated (2000RAD) fresh Langerhans cells (5 10⁴ per well) in 96 round bottom microtiter plates at 37°C, 5% CO₂ for 5d. In some experiments, 0.1g of anti-TNF- was added. One microcurie [³H]thymidine was added to each well 16–18h before termination of culture. The cells were harvested with Tomtec Harvester (Wallac, LKB, NJ) and assayed for [³H]thymidine incorporation.

Statistical analysis

The statistical significance of differences between means of each experimental group was determined by analysis of variance. Mean differences were considered to be significant when p <0.05. Analyses were performed using StatView 512^+. Each experiment was repeated at least twice.

Results

IgE-triggered mast cell degranulation impairs CH induction

TNF- is an important mediator of the deleterious effects of UVR on CH induction. Using two different UVB-S strains (C3H/HeN and C57BL/6), immunohistochemistry was performed on skin sections snap frozen 30min after exposure to one dose of UVR. It seems that mast cells became degranulated and that their granules were positive for TNF- (data not shown). The staining was similar to that observed upon treatment of the skin with CGRP (^{Niizeki et al. 1997}). These data suggested that mast cells may be the source of TNF- released in response to UVR. The following experiments were then designed to determine whether TNF- released by mast cells can mediate impairment of CH in a manner similar to that found with UVR treatment. It is known that mast cells express receptors for the Fc region of IgE molecules, Fc3RI, and that ligation of IgE molecules to these receptors can trigger degranulation (^{Gordon & Galli 1991}). We used this approach in the next series of experiments. Anti-DNP IgE antibodies were injected intradermally into shaved abdominal skin of naive C3H/HeN (UVB-S) and C3H/HeJ (UVB-R) mice. Control mice received anti-dansyl IgE antibodies instead. One day later, DNP–HSA diluted in 1% Evan's blue dye was injected intravenously. Within 10min, the injected sites turned intensely blue due to the dilatation of the capillary system, indicating that local release of histamine from mast cell granules had been achieved (data not shown). Within 30min, oxazolone (2%) was painted epicutaneously on the injected site or irradiated site. The ears of these mice were challenged 5d later with dilute oxazolone. As revealed in Figure 1a, CH responses of IgE-DNP treated C3H/HeN mice (UVB-S) were significantly reduced compared with positive controls (anti-dansyl IgE). In contrast, CH of comparable intensities was elicited in the ears of C3H/HeJ mice (UVB-R) whether the injected IgE antibodies were anti-DNP or anti-dansyl (Figure 1b). To confirm that the relevant factor being released from mast cells was TNF-, neutralizing anti-TNF- antibodies were injected intraperitoneally 6h before the i.v. injection of DNP–HSA. Oxazolone was applied epicutaneously to the injected site 30min after the i.v. injection of DNP–HSA. The ears of these mice were challenged with dilute hapten 5d later. The results of these experiments are included in Figure 1a. Anti-TNF- antibodies were shown to restore the CH responses of UVB-S mice that had received anti-DNP IgE antibodies, followed by DNP–HSA prior to hapten sensitization.

Figure 1.

IgE triggered degranulation of mast cells mimics UVR-induced impairment of CH. Anti-DNP IgE antibodies were injected intradermally into shaved abdominal skin of C3H/HeN (A) or C3H/HeJ mice (b). Control mice received anti-dansyl IgE antibodies instead. One day later, DNP–HSA diluted in 1% phenol blue was injected intravenously. Within 30min, oxazolone (2%) was painted epicutaneously on the injected site or on the irradiated site, for mice exposed to one dose of UVR (400J per m²). In some cases, anti-TNF- or anti-bovine serum albumin antibodies (2 10⁴U) were injected i.p. 6h prior to sensitization. The ears of these mice were challenged 5d later with dilute oxazolone and the ear swelling response was measured 24h later. Bars indicate mean ear swelling responses SEM. An asterisk indicates values significantly less than positive control (anti-dansyl IgE-treated group) (p <0.04).

Full figure and legend (24K)

These findings indicate that release of TNF- from mast cells, triggered by an IgE-dependent mechanism, leads to impaired CH induction in C3H/HeN mice. As similar changes are observed in C3H/HeN mice exposed to UVR, and as these changes are also reversed by anti-TNF- antibodies, these data suggest that UVR may impair CH induction by triggering the release of TNF- from dermal mast cells.

UVR-impaired CH induction is dependent upon mast cells

To investigate the possible participation of mast cells in impaired CH induction after UVR, we used mast cell-deficient mice which should belong to the UVR susceptible category due to their background (WBxB6F1) (^{Yoshikawa & Streilein 1990}). Groups of mast cell-deficient mice (W/W^v) and their congenic normal littermates received a single UVR exposure to shaved abdominal skin. Immediately thereafter, an optimal dose of DNFB (0.004%, 25l) was painted on the exposed site. When the ears of these mice were painted with dilute DNFB 5d later, the results displayed in Figure 2 were obtained. DNFB readily induced CH in nonirradiated W/W^v mice, as well as in their congenic normal littermates. When DNFB was painted on UVR-exposed skin of the wild-type mice, CH induction was impaired, indicating that these mice were UVB susceptible as expected. DNFB applied to UVR-exposed skin of mast-cell deficient mice, however, induced CH of comparable intensity with that of the nonirradiated controls. Similar results were observed using a different regimen of UVB and oxazolone at a conventional sensitizing dose (2%, 25l) in a different strain of mast cell deficient mice (Sl/Sl^d). These results are displayed in Table 1. These findings indicate that in wild-type littermates of Sl/Sl^d or W/W^v mice, susceptibility to the deleterious effects of UVR on CH induction is dependent upon the presence of mast cells. In the absence of mast cells, UVR failed to prevent CH induction.

Figure 2.

Participation of mast cells in impaired CH induction after UVR. Mast cell-deficient mice (W/W^v) and their congenic normal littermates received a single UVR exposure to shaved abdominal skin. Immediately thereafter, DNFB (0.004%, 25l) was painted on the exposed site. Five days later, the ears of these mice were painted with dilute DNFB (0.05%) and the ear swelling was measured 24h following challenge. Bars indicate mean ear swelling responses SEM. An asterisk indicates values significantly less than positive control (p <0.001).

Full figure and legend (25K)

Table 1 - UVR-impaired CH induction is dependent upon mast cells.

Full table

To confirm that the deficit of mast cells was directly responsible for the lack of a UVR effect in mast cell deficient mice, skin from these mice was injected intradermally with anti-DNP IgE, followed by systemic injection of DNP–HSA. No impairment of CH was observed in mast cell deficient mice as was expected (Figure 3a), whereas the wild-type control exhibited impairment of CH (Figure 3b). Intradermal injection of TNF- in Sl/Sl^d, however, was able to induce impairment of CH (Figure 4) in a manner similar to that observed in other UVB-R and UVB-S strains of mice (^{Dai & Streilein 1997}), suggesting that Langerhans cells from mast cell deficient mice are susceptible to the deleterious effect of TNF-. These results strongly suggest that the release of TNF- by mast cells is required in order for CH induction to be impaired by UVR.

Figure 3.

IgE-triggered degranulation of mast cells impairs CH induction. Anti-DNP IgE antibodies were injected intradermally into shaved abdominal skin of wild-type (A) or Sl/Sl^d mice (b). Control mice received anti-dansyl IgE antibodies instead. One day later, DNP–HSA diluted in 1% Evan's blue was injected intravenously. Within 30min, oxazolone (2%) was painted epicutaneously on the injected site. The ears of these mice were challenged 5d later with dilute oxazolone and the ear swelling response was measured 24h later. Bars indicate mean ear swelling responses SEM. An asterisk indicates values significantly less than positive control (anti-dansyl IgE-treated group) (p <0.01).

Full figure and legend (21K)

Figure 4.

Intradermal injection of TNF- in Sl/Sl^d mice induces impairment of CH. Two hundred microliters of TNF- (3 10⁴U) or PBS was injected intradermally into shaved abdominal skin of Sl/Sl^d mice. Within 30min, oxazolone (2%) was painted epicutaneously on the injected site. The ears of these mice were challenged 5d later with dilute oxazolone and the ear swelling response was measured 24h later. Bars indicate mean ear swelling responses SEM. An asterisk indicates values significantly less than positive control (PBS treated group) (p <0.02).

Full figure and legend (12K)

Source of UVR-triggered release of TNF- in mast cells

Release of TNF- from mast cells may occur by direct secretion of preformed TNF-, or by the induction of newly synthesized TNF-. It has been shown by others that IgE antibodies can trigger mast cells via their Fc receptors to release preformed TNF-. It has also been reported, however, that within 30–45min of IgE-dependent activation of mast cells, newly synthesized TNF- can be detected (^{Gordon & Galli 1991}). To discriminate between stored or newly synthesized TNF-, we used the steroid DX, which has been shown to inhibit mast cell synthesis of TNF- by a post-transcriptional mechanism (^{Schmidt et al. 1996}). C3H/HeN mice received i.p. injections of DX (30mg per kg) for three consecutive days, without any consequences on CH induction (Figure 5). One hour after the last injection, the abdominal skin was exposed to a single dose of UVR (400J per m²). Thirty minutes later, the surface was painted with oxazolone (2%). Positive control mice received DX, but no UVR. Five days later the ears of these mice were challenged with dilute oxazolone. The results presented in Figure 5 indicate that DX restored CH responsiveness to mice exposed to UVR. These findings support the view that the mast cell-derived TNF- induced by UVR exposure, and which is involved in impairment of CH, is newly synthesized rather than preformed.

Figure 5.

Dexamethasone restores CH responsiveness induced by UVR. C3H/HeN mice received i.p. injections of DX (30mg per kg) for three consecutive days. One hour after the last injection, the abdominal skin was exposed to a single dose of UVR (400J per m²). Thirty minutes thereafter, the surface was painted with oxazolone (2%). Positive control mice received DX, but no UVR. Five days later the ears of these mice were challenged with dilute oxazolone and the ear swelling response measured at 24h. Bars indicate mean ear swelling responses SEM. An asterisk indicates values significantly less than positive control (oxazolone sensitized group without DX treatment) (p <0.001).

Full figure and legend (11K)

Quantity of TNF- released by mast cells from UVB-S and UVB-R mice

Next, we investigated whether there were differences in the amount of TNF- released from mast cells from UVB-S and UVB-R mice, as this could explain the phenotypes of resistance and sensitivity to UVR (^{Kurimoto & Streilein 1994}). We generated mast cells from bone marrow suspensions prepared from BALB/C (UVB-R) and C57BL/6 (UVB-S) mice. The cells were cultured for 3–4wk in the presence of WEHI-3 supernatant, which is known to contain the mast cell mitogen, IL-3. At the completion of these cultures, the cells were harvested, loaded with anti-DNP IgE antibodies, and exposed to DNP–HSA in vitro. Supernatants were harvested 1 or 2h later and assayed for TNF- and IL-4 content by enzyme-linked immunosorbent assay. As indicated in Table 2, mast cells prepared from C57BL/6 mice released significantly more TNF- than BALB/C mice, whereas both of them released similar amounts of IL-4, ruling out the possibility of a differential mast cell growth rate. To determine whether the TNF- measured in these supernatants was newly synthesized, we have also measured TNF- production by anti-DNP triggered mast cells treated with actinomycin D at a concentration that inhibits gene transcription. In the presence of actinomycin D, mast cells prepared from BALB/C mice released significantly less TNF- in the medium (15–38%) compared with untreated mast cells (100%), suggesting that most of the TNF- released from mast cells during the first hour is newly synthesized. Taken together, these data suggest that the phenotypes of resistance and sensitivity to UVR may result from differences in the quantity of TNF- synthesized and released from mast cells after UVR exposure.

Table 2 - Differential release of TNF by mast cells prepared from UVB-R versus UBV-S mice.

Full table

Analysis of active TNF- produced by mast cells from UVB-S versus UVB-R mice

There is a possibility that different ratios of active versus inactive TNF- are released from mast cells of UVB-S versus UVB-R mice. Using a functional assay to test this hypothesis, we examined whether TNF- released by mast cells from UVB-S versus UVB-R mice could inhibit the ability of Langerhans cells to induce T cell proliferation to the same extent.^{Dai & Streilein (1995)} have previously shown that the effects of UVR on Langerhans cells and CH induction can be reproduced in vitro using hapten-derived Langerhans cells treated with TNF- in a T cell proliferation assay. Single cell suspensions enriched for Ia⁺ Langerhans cells were prepared from normal BALB/C or C57BL/6 skin. These cells were then incubated in vitro for 2h with supernatants collected from C57BL/6 or BALB/C mast cells, respectively, that had been stimulated for 4h to release their granular contents by anti-DNP IgE antibodies and DNP–HSA. These supernatants were tested for their content of TNF- and the concentration adjusted to 50pg of TNF- perml. The inhibition of allogeneic T cell proliferation using C57BL/6 Langerhans cells incubated with supernatants from mast cells generated from C57BL/6 mice was greater than 70% (Figure 6a). Moreover, when Langerhans cells from C57BL/6 mice were incubated simultaneously with mast cell supernatants and anti-TNF- (0.1g of anti-TNF- which neutralizes 1000U of TNF-), allogeneic T cell proliferation was retained, indicating that the suppression observed above was mediated by TNF- (Figure 6a). Interestingly, although C57BL/6 Langerhans cell functions were inhibited by supernatants from mast cells generated from either C57BL/6 or BALB/C mice, BALB/C Langerhans cells retained most of their ability to induce allogeneic T cell proliferation (Figure 6b). These results indicate that mast cells from UVB-S versus UVB-R produce comparable amounts of active TNF-.

Figure 6.

Mast cell supernatants inhibit the ability of Langerhans cells to activate allogeneic T cells. Single cell suspensions enriched for Ia⁺ Langerhans cells were prepared from normal BALB/C and C57BL/6 skin. C57BL/6 Langerhans cells were treated for 2h or not with anti-TNF- and mast cell supernatants from C57BL/6 mast cells that had been triggered to release their granular contents by anti-DNP IgE antibodies and DNP–HSA (adjusted at 50pg perml of TNF-). These Langerhans cells were then washed and incubated for 5d with allogeneic BALB/C T cells (A). Similarly, BALB/C and C57BL/6 Langerhans cells after treatment with supernatants from C57BL/6 or BALB/C mast cells were incubated with allogeneic T cells (isolated from C57BL/6 and BALB/C mice, respectively) for 5d (b). Background cpm were less than 1000cpm. The cultures were pulsed with [³H]thymidine during the last 18h of the culture.

Full figure and legend (18K)

Evaluation of the susceptibility of Langerhans cells from UVB-S versus UVB-R mice to TNF-

As there appeared to be a difference in the response of Langerhans cells from BALB/C versus C57BL/6 to TNF--containing supernatants produced by mast cells, the next experiments were designed to determine whether susceptibility to UVR could involve susceptibility of Langerhans cells to TNF-. Langerhans cells were isolated from normal C57BL/6 and BALB/C skin and incubated in vitro for 2h with various concentrations of TNF-. Langerhans cells from C57BL/6 incubated with low levels of TNF- were prevented from inducing allogeneic T cell proliferation (Figure 7). In contrast, the dose of TNF- required to inhibit the ability of BALB/C Langerhans cells to stimulate allogeneic T cells was 10–100 times higher (Figure 7); this indicated a resistance of UVB-R Langerhans cells to the deleterious effects of TNF-. These results suggest that the difference in UVB susceptibility between C57BL/6 and BALB/C mice may be due to a difference in sensitivity to the downregulatory effects of TNF-.

Figure 7.

Differential doses of TNF- inhibit the ability of Langerhans cells to activate allogeneic T cells. Langerhans cells were enriched from normal BALB/C and C57BL/6 skin and incubated in vitro for 2h with various concentrations of TNF-. Langerhans cells were then washed and incubated with allogeneic T cells (isolated from C57BL/6 and BALB/C mice, respectively) for 5d, that were pulsed with [³H]thymidine during the last 18h of the culture. Background counts per minute were less than 1000cpm. Asterisks indicate values significantly less than positive control (group without TNF- treatment) (*p <0.001, **p <0.01, ***p <0.03).

Full figure and legend (7K)

Discussion

Exposure of the skin to UVR leads to impairment of CH when the hapten is applied shortly thereafter to the irradiated site. Several UVR-dependent factors that may participate in failed CH induction have been identified, such as TNF- (^{Yoshikawa et al. 1992}), cis-urocanic acid (^{Kurimoto & Streilein 1992}), -melanocyte-stimulating hormone (^{Shimizu & Streilein 1994}), IL-10 (^{Enk et al. 1994}), reactive oxygen intermediates (^{Nakamura et al. 1997}), and DNA photoproducts (^{Applegate et al. 1989}). We turned our attention to TNF- as our laboratory has previously demonstrated that neutralizing anti-TNF- antibodies can restore the capacity of UVB-treated skin to support the induction of CH (^{Yoshikawa et al. 1992}). The cellular source of TNF- that is responsible for the impairment of CH is still unknown. Two types of cells in the skin appear to be candidates. It has been reported that human keratinocytes are induced to produce TNF- 12h after UVR (^{Kock et al. 1990}), whereas, the ''sunburn'' response is accompanied by release of TNF- from human dermal mast cells (^{Walsh 1995}). Furthermore, mast cells are known to store preformed TNF- and to secrete newly synthetized TNF- within 30min following certain stimuli (^{Gordon & Galli 1990}). As hapten is painted on the exposed skin 30min after irradiation, we were looking for a rapid source of TNF- that could act quickly (within a few hours) on cutaneous antigen-presenting cells. Therefore, in this study we tested the hypothesis that mast cells are responsible for the deleterious effects of UVB on CH induction.

Our immunohistochemistry data suggested that mast cells may be the source of TNF- released in response to UVR. Furthermore, activation of mast cells through their Fc receptor resulted in a significant reduction in Ia⁺ epidermal cells in C3H/HeN mice, as well as a change in morphology of the remaining cells such as loss of dendrites and presence of swollen and rounded soma (data not shown). Moreover, CH induction was inhibited when hapten was applied to the same site. The release of TNF- from mast cells appeared to mimic the effects of UVR, as both induced impairment of CH and loss of Ia⁺ cells which were reversed with anti-TNF- antibodies. Our data are supported by those of^{Wille et al. (1995)} who have shown that topical application of degranulating agents (e.g., chloroquine, and more importantly cis-urocanic acid, a mediator of UVR effects) to the skin induces mast cells to degranulate as well as impairing CH induction. More recently, we have shown that UVR-impaired induction of CH is mediated by CGRP, which triggers the release of TNF- from mast cells (^{Niizeki et al. 1997}). Further, we found that UVR does not impair CH induction when hapten is painted on irradiated skin of mast cell deficient mice (Sl/Sl^d), even though CH induction is impaired by UVR in their wild-type littermates. Finally, Sl/Sl^d mice treated locally with subcutaneous injections of c-kit ligand showed impairment of CH after exposure of UVB (data not shown). These results suggest that mast cells play a key part in the aborted ability of hapten-painted Langerhans cells to induce CH induction after exposure to UVR. Moreover, experiments using DX suggest that newly synthesized TNF- may be necessary for UVR-mediated impairment of CH. It is also possible, however, that DX may have prevented the production of a factor or cytokine that acts as a cofactor with TNF- or as a component of a pathway acting downstream from the activity of TNF-. As UVR impairs CH induction through a TNF--dependent mechanism, our data strongly suggest that UVR triggers the release of TNF- from dermal mast cells, and that mast cell-derived TNF- interferes with generation of the hapten-specific signal that is required for CH induction.

Recently,^{Hart et al. (1998a)} have reported that the CH unresponsiveness that follows a single large exposure to UVR depends upon the presence of mast cells at the exposed site. Specifically, these investigators exposed shaved dorsal skin of mice to 12kJ per m² UVR, and then applied a sensitizing dose of hapten to ventral (unexposed) skin 5d later. These treatments resulted in immunosuppression and failure to develop CH, that was dependent upon the existence of mast cells. The experiments of^{Hart et al. (1998a)} differ substantially from ours in that we induce tolerance by painting hapten on the skin immediately after the exposure to UVR. Substantial evidence indicates that failed CH of the type we report differs mechanistically from the immunosuppression induced systemically by high-dose UVR that was described by^{Hart et al. (1998a)} Nonetheless, both effects appear to be dependent upon mediators released from dermal mast cells. In our case, however, the mediator is TNF-, whereas in the system described by^{Hart et al. (1998a)} histamine appears to be involved. In support of our data, the most recent publication by^{Hart et al. (1998b)} shows that TNF- is critical to the mechanism of immunosuppression following hapten application to the UV radiated site.

Our laboratory has identified a polymorphism within the 5'-untranslated region of the TNF- gene, which appears to be responsible for the two different mouse phenotypes, UVB-R and UVB-sensitive (^{Vincek et al. 1993,1994}). As the polymorphism resides in the 5'-untranslated region of the TNF- gene, it has been hypothesized that the quantity of TNF- available after exposure to UVB in UVB-R mice is lower than in UVB-S mice (^{Kurimoto & Streilein 1994}). To test this hypothesis, we compared the level of TNF- released from cultured mast cells isolated from the bone marrow of UVB-R versus UVB-S mice. We used a well-established in vitro system to induce release of TNF- from mast cells by triggering their Fc receptor with IgE anti-DNP. Under these conditions, mast cells from UVB-S mice appeared to produce more TNF- than those from UVB-R mice, suggesting that the phenotype of UVB-S and UVB-R may result from differences in the amount of TNF- released from mast cells after UVR exposure. Recent data published by^{Hart et al. (1998b)} have shown that the dermis of BALB/C mice contains 50% fewer mast cells than the dermis of C57BL/6 mice, supporting the idea that quantitative differences in TNF- released in the skin after UVB treatment exist in these two strains. Furthermore, the fact that mast cells are located deeper in the dermis in the case of BALB/C versus C57BL/6 mice (60% of BALB/C mast cells versus only 20% for C57BL/6 mice were located in the subcutaneous fatty tissues) may affect their ability to degranulate in response to UVR.

Experiments were also performed to determine whether mast cells from UVB-S versus UVB-R mice produce different amounts of active TNF-. An in vitro system was used to examine whether TNF- released by mast cells from UVB-S versus UVB-R could inhibit, to the same extent, the ability of Langerhans cells to induce T cell proliferation. The data indicate that supernatants released by mast cells generated from UVB-S versus UVB-R are identical in their capacity to inhibit Langerhans cell induction of T cell proliferation. Langerhans cells from UVB-R mice, however, appeared to be less sensitive to inhibition, suggesting that the sensitivity of Langerhans cells to TNF- could also contribute to the susceptibility/resistance phenotype. Similar results were found using an in vivo system in our laboratory;^{Vermeer & Streilein (1990)} showed that the dose of injected TNF- necessary to reduce the density and alter the morphology of Langerhans cells was 10–100 times higher in UVB-R than UVB-S mice (^{Vermeer & Streilein 1990}). A similar observation was made concerning the impairment of CH induction after intracutaneous injection of TNF- (^{Yoshikawa & Streilein 1990}).

In conclusion, we have demonstrated that exposure of the skin to UVR triggers dermal mast cells to release TNF-, which in turn inhibits cutaneous antigen-presenting cells from inducing CH (^{Yoshikawa et al. 1992}). Our data further indicate that it is both the quantity of TNF- released by mast cells and the sensitivity of Langerhans cells themselves to TNF- that define the phenotypes of resistance versus sensitivity to UVB.

References

1. Applegate, LA, Ley, RD, Alcalay, J, Kripke, ML Identification of the molecular target for the suppression of contact hypersensitivity by ultraviolet radiation. J Exp Med, 1989 170, 1117–1131,  | Article | PubMed | ISI | ChemPort |
2. Chabot, B, Stephenson, DA, Chapman, VM, Besmer, P, Bernstein, A The proto-oncogene c-kit encoding a transmembrane tyrosine kinase receptor maps to the mouse W locus. Nature, 1988 335, 88–89,  | Article | PubMed | ISI | ChemPort |
3. Dai, R & Streilein, JW Sensitizing capacity of Langerhans' cells obtained from ultraviolet-B-exposed murine skin. Immunology, 1995 86, 661–667,  | PubMed | ISI | ChemPort |
4. Dai, R & Streilein, JW Ultraviolet B-exposed and soluble factor-pre-incubated epidermal Langerhans cells fail to induce contact hypersensitivity and promote DNP-specific tolerance. J Invest Dermatol, 1997 108, 721–726,  | Article | PubMed | ISI | ChemPort |
5. De Kossodo, S, Cruz, PD, Dougherty, I, Thompson, P MS-V, BB: Expression of the tumor necrosis factor gene by dermal fibroblasts in response to ultraviolet irradiation or lipopolysaccharide. J Invest Dermatol, 1995 104, 318–322,  | Article | PubMed | ChemPort |
6. Enk, AH, Saloga, J, Becker, D, Mohamadzadeh, M, Knop, J Induction of hapten-specific tolerance by interleukin 10 in vivo. J Exp Med, 1994 179, 1397–1402,  | Article | PubMed | ISI | ChemPort |
7. Galli, SJ & Hammel, I Unequivocal delayed hypersensitivity in mast cell-deficient and beige mice. Science, 1984 226, 710–713,  | PubMed | ISI | ChemPort |
8. Galli, SJ, Tsai, M, Gordon, JR, Geissler, EN, Wershil, BK Analyzing mast cell development and function using mice carrying mutations at W/c-kit or Sl/MGF (SCF) loci. [Review]. Ann N Y Acad Sci, 1992 664, 69–88,  | PubMed | ISI | ChemPort |
9. Geissler, EN, Ryan, MA, Housman, DE The dominant-white spotting (W) locus of the mouse encodes the c-kit proto-oncogene. Cell, 1988 55, 185–192,  | Article | PubMed | ISI | ChemPort |
10. Gordon, JR & Galli, SJ Mast cells as a source of both preformed and immunologically inducible TNF-alpha/cachectin. Nature, 1990 346, 274–276,  | Article | PubMed | ISI | ChemPort |
11. Gordon, JR & Galli, SJ Release of both preformed and newly synthesized tumor necrosis factor alpha (TNF-alpha) /cachectin by mouse mast cells stimulated via the Fc epsilon RI. A mechanism for the sustained action of mast cell-derived TNF-alpha during IgE-dependent biological responses. J Exp Med, 1991 174, 103–107,  | Article | PubMed | ISI | ChemPort |
12. Hart, PH, Grimbaldeston, MA, Swift, GJ, Jaksic, A, Noonan, FP, Finlay-Jones, JJ Dermal mast cells determine susceptibility to ultraviolet B-induced systemic suppression of contact hypersensitivity responses in mice. J Exp Med, 1998a 187, 2045–2053,  | Article | PubMed | ISI | ChemPort |
13. Hart, PH, Grimbaldeston, MA, Swift, GJ, Sedgwick, D, Korner, H, Finlay-Jones, JJ TNF modulates susceptibility to UVB-induced systemic immunomodulation in mice by effects on dermal mast cell prevalence. Eur J Immunol, 1998b 28, 2893–2901,  | Article | ISI | ChemPort |
14. Kitamura, Y & Go, S Decreased production of mast cells in Sl/Sld mice. Blood, 1979 53, 492–497,  | PubMed | ISI | ChemPort |
15. Kitamura, Y, Go, S, Hatanaka, K Decrease of mast cell in W/Wv mice and their increase by bone marrow transplantation. Blood, 1978 52, 447–452,  | PubMed | ISI | ChemPort |
16. Kock, A, Schwarz, T, Kirnbauer, R, Urbanski, A, Perry, P, Ansel, JC, Luger, TA Human keratinocytes are a source for tumor necrosis factor alpha: evidence for synthesis and release upon stimulation with endotoxin or ultraviolet light. J Exp Med, 1990 172, 1609–1614,  | Article | PubMed | ISI | ChemPort |
17. Kurimoto, I & Streilein, JW cis-urocanic acid suppression of contact hypersensitivity induction is mediated via tumor necrosis factor-. J Immunol, 1992 148, 3072–3078,  | PubMed | ISI | ChemPort |
18. Kurimoto, I & Streilein, JW Studies of contact hypersensitivity induction in mice with optimal sensitizing doses of hapten. J Invest Dermatol, 1993 101, 132–136,  | Article | PubMed | ISI | ChemPort |
19. Kurimoto, I & Streilein, JW Characterization of the immunogenetic basis of ultraviolet-B light effects on contact hypersensitivity induction. Immunology, 1994 81, 352–358,  | PubMed | ISI | ChemPort |
20. Mekori, YA & Galli, SJ Undiminished immunologic tolerance to contact sensitivity in mast cell-deficient W/Wv and Sl/Sld mice. J Immunol, 1985 135, 879–885,  | PubMed | ISI | ChemPort |
21. Nakamura, T, Pinnel, SR, Streilein, JW Vitamicin C abrogates the deleterious effects of UVB radiation on cutaneous immunity by a TNF- independent mechanism. J Invest Dermatol, 1997 109, 20–24,  | Article | PubMed | ISI | ChemPort |
22. Nakano, T, Sonoda, T, Hayashi, C et al. Fate of bone marrow-derived cultured mast cells after intracutaneous, intraperitoneal, and intravenous transfer into genetically mast cell-deficient W/Wv mice. Evidence that cultured mast cells can give rise to both connective tissue type and mucosal mast cells. J Exp Med, 1985 162, 1025–1043,  | Article | PubMed | ISI | ChemPort |
23. Niizeki, H, Alard, P, Streilein, JW Calcitonin gene-related peptide is necessary for ultraviolet B-impaired induction of contact hypersensitivity. J Immunol, 1997 159, 5183–5186,  | PubMed | ISI | ChemPort |
24. Schmidt, CA, Furuta, GT, Lavigne, JA, Galli, SJ, Wershil, BK The regulation of tumor necrosis factor-alpha production in murine mast cells: pentoxifylline or dexamethasone inhibits IgE-dependent production of TNF-alpha by distinct mechanisms. Cell Immunol, 1996 171, 140–146,  | Article | PubMed | ChemPort |
25. Shimizu, T & Streilein, JW Influence of alpha-melanocyte stimulating hormone on induction of contact hypersensitivity and tolerance. J Dermatol Sci, 1994 8, 187–193,  | Article | PubMed | ChemPort |
26. Tsai, M, Shih, LS, Newlands, GF et al. The rat c-kit ligand, stem cell factor, induces the development of connective tissue-type and mucosal mast cells in vivo. analysis by anatomical distribution, histochemistry, and protease phenotype. J Exp Med, 1991 174, 125–131,  | Article | PubMed | ISI | ChemPort |
27. Vermeer, M & Streilein, JW Ultraviolet B light-induced alterations in epidermal Langerhans cells are mediated in part by tumor necrosis factor-alpha. Photodermatol, Photoimmunol Photomed, 1990 7, 258–265,  | PubMed | ISI | ChemPort |
28. Vincek, V, Kurimoto, I, Medema, JP, Prieto, E, Streilein, JW Tumor necrosis factor alpha polymorphism correlates with deleterious effects of ultraviolet B light on cutaneous immunity. Cancer Res, 1993 53, 728–732,  | PubMed | ISI | ChemPort |
29. Vincek, V, Kurimoto, I, Streilein, JW Polymorphism at the tumor necrosis factor-alpha locus and susceptibility to toxoplasmic encephalitis, ultraviolet B-induced immune impairment, and mouse AIDS [letter]. J Acquired Immune Deficiency Syndromes, 1994 7, 734–735,  | ISI | ChemPort |
30. Walsh, L Ultraviolet B irradiation of skin induces mast cell degranulation and release of tumour necrosis factor-. Immunol Cell Biol, 1995 73, 226–233,  | PubMed | ISI | ChemPort |
31. Wershil, BK, Furuta, GT, Lavigne, JA, Choudhury, AR, Wang, ZS, Galli, SJ Dexamethasone or cyclosporin A suppress mast cell-leukocyte cytokine cascades. Multiple mechanisms of inhibition of IgE- and mast cell-dependent cutaneous inflammation in the mouse. J Immunol, 1995 154, 1391–1398,  | PubMed | ISI | ChemPort |
32. Wille, J, Njiehs, F, Amin, P, Kydonieus, A Topical delivery of mast cell degranulating agents for treatment of transdermal drug-induced hypersensitivity. Proc Intern Symp Controll Rel Bioact Mater, 1995 22, 119–120,
33. Yoshikawa, T & Streilein, JW Genetic basis of the effects of ultraviolet light B on cutaneous immunity. Evidence that polymorphism at the TNF- and Lps loci governs susceptibility. Immunogenetics, 1990 32, 398–405,  | PubMed | ISI | ChemPort |
34. Yoshikawa, T, Kurimoto, I, Streilein, JW Tumour necrosis factor-alpha mediates ultraviolet light B-enhanced expression of contact hypersensitivity. Immunology, 1992 76, 264–271,  | PubMed | ISI | ChemPort |
35. Zsebo, KM, Williams, DA, Geissler, EN et al. Stem cell factor is encoded at the sl locus of the mouse and is the ligand for the c-kit tyrosine kinase receptor. Cell, 1990 63, 213–224,  | Article | PubMed | ISI | ChemPort |

Acknowledgments

This work was supported by National Institutes of Health Grants AR44130–11. We wish to thank Dr. Naili Ma for her advice concerning i.v. injection and immunostaining procedures, and Dr. Michele M. Kosiewicz and Dr. Jacqueline M. Doherty for their comments on the manuscript.