cis-Urocanic acid induces mast cell degranulation and release of preformed TNF-α: A possible mechanism linking UVB and cis-urocanic acid to immunosuppression of contact hypersensitivity

Article abstract of DOI:10.1159/000029842

The search for effective inhibitors of transdermal drug-induced contact sensitization was directed to dermal mast-cell-degranulating agents (MCDA). Human skin organ cultures were employed to test whether cis-urocanic acid (C-UA) and other potential MCDAs cause mast cell degranulation. These were then tested for their ability to inhibit the induction phase of the contact hypersensitivity reaction (CHR). C-UA at 1 μg/ml significantly depleted mast cell chymase, whereas trans-urocanic acid (T-UA) was relatively ineffective. C-UA, but not T-UA, induced local effects of liberated mast cell TNF-α, as detected by E-selectin expression on the micro-vascular dermal endothelium. C-UA significantly reduced (> 70%) the ear swelling response in Balb/c mice, when applied 24 h prior to application of a sensitizing amount of dinitrochlorobenzene (DNCB), and induced a prolonged (> 3 weeks) state of immune tolerance (> 40%). Similar effects on local immunosuppression of CHR were observed with topical chloroquine and capsaicin, while cromolyn, a mast cell membrane stabilizer, was unable to inhibit DNCB-induced CHR. It is suggested that MCDAs may interfere with downstream events associated with accessory cell function.

Full text of DOI:10.1159/000029842

Original Research Article
Received: Sept. 22, 1998
Accepted: Dec. 18, 1998
Skin Pharmacol Appl Skin Physiol 1999;12:18–27
cis-UrocanicAcidInducesMastCell
DegranulationandReleaseofPreformed
TNF-·: A Possible Mechanism Linking UVB
and cis-Urocanic Acid to Immunosuppression
ofContactHypersensitivity
John J. Willea Agis F. Kydonieusa George F. Murphyb
aConvaTec, Division of Bristol-Myers Squibb Company, Princeton, N.J., and
bDepartment of Pathology, Anatomy and Cell Biology, Jefferson University School of Medicine,
Philadelphia, Pa., USA
KeyWords
vascular dermal endothelium. C-UA signifi-
cis-Urocanic acid W Contact dermatitis W Mast
cell degranulation W Skin sensitization W
Transdermal drugs
cantly reduced (170%) the ear swelling re-
sponse in Balb/c mice, when applied 24 h
prior to application of a sensitizing amount of
dinitrochlorobenzene (DNCB), and induced a
prolonged (13 weeks) state of immune toler-
ance (140%). Similar effects on local immu-
Abstract
The search for effective inhibitors of trans- nosuppression of CHR were observed with
dermal drug-induced contact sensitization topical chloroquine and capsaicin, while cro-
was directed to dermal mast-cell-degranulat- molyn, a mast cell membrane stabilizer, was
ing agents (MCDA). Human skin organ cul- unable to inhibit DNCB-induced CHR. It is
tures were employed to test whether cis-uro- suggested that MCDAs may interfere with
canic acid (C-UA) and other potential MCDAs downstream events associated with accesso-
cause mast cell degranulation. These were ry cell function.
then tested for their ability to inhibit the
induction phase of the contact hypersensitiv-
ity reaction (CHR). C-UA at 1 Ìg/ml signifi-
cantly depleted mast cell chymase, whereas
trans-urocanic acid (T-UA) was relatively
ineffective. C-UA, but not T-UA, induced local
effects of liberated mast cell TNF-·, as de-
tected by E-selectin expression on the micro-
Introduction
A major obstacle to the transdermal deliv-
ery of drugs is the development of allergic
contact dermatitis. Delayed contact hypersen-
© 1999 S. Karger AG, Basel
ABC
Dr. John J. Wille
Bioderm Technologies Inc., 36 South Broad Street
Trenton, NJ 08608 (USA)
1422–2868/99/0122–0018$17.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch
www.karger.com
Accessible online at:
http://BioMedNet.com/karger
Tel. +1 609 656 0784, Fax +1 609 396 8603
E-Mail JJWille@aol.com

sitivity responses are reportedto occur for tiation of such a photoreaction is urocanic
many topical medicaments surveyed by patch acid, a skin chromophore that absorbs sun-
testing, including benzocaine, chlorphenir- light in the UVB range. trans-Urocanic acid
amine, diphenhydramine, neomycin sulfate, (T-UA) is formed and stored naturally in the
penicillin, promazine and sulfanilamide [1]. uppermost layers of the epidermis [9] from
For example, transdermal delivery of cloni- the deamination of histidine by histidase.
dine causes contact dermatitis reactions in up Upon exposure to UVB, T-UA is photoisom-
to 20% of treated patients [2], as do most ß- erized to cis-urocanic acid (C-UA).
blockers, antiasthmatics and antihistamines
There is an extensive literature demon-
that have been tested in humans [3]. The strating an immunosuppressive effect of C-
search for transdermal therapies that effec- UA on CHR [15–18]. Recently, it has been
tively prevent contact dermatitis has identi- reported that C-UA has differential effects on
fied several agents including lanthanum ci- the immunosuppression of delayed-type hy-
trate and diltiazem [4], corticosteroids [5], persensitivity versus CHR [19]. Antibodies to
verapamil hydrochloride [6] and amiloride C-UA relieve UVB-induced suppression of
and its analogues [7]. Recently, we reported delayed-type hypersensitivity [17] and reverse
that ethacrynic acid prevents sensitization to UV-induced changes in Langerhans cells but
a number of topically applied skin-sensitizing fail tu prevent CHR or accumulation of den-
drugs [8].
dritic cells in the local draining lymph nodes
It is well-known that low-dose ultraviolet [20]. In the experiments presented here, we
light of the B wavelength (UVB) is also immu- sought to investigate the linkage between C-
nosuppressive [9] and often results in an anti- UA immunosuppression and mast cell de-
gen-specific state of immune tolerance [10]. granulation [21].
This effect is mimicked by intradermal injec-
tion of TNF-· [11]. UVB also degranulates
mast cells [12], and thus this cell is a potential
Materials and Methods
source for endogenous TNF-· release. We
Chemicals
previously published that normal human
mast cells synthesize, store and secrete TNF-·
in amounts that significantly exceed those of
other dermal cell types [13]. We have also
learned that when mast cells degranulate and
release TNF-· in situ, a sensitive and relative-
ly specific direct bioassay for the TNF-· effect
is induction of E-selectin synthesis and ex-
pression in adjacent microvascular endothe-
lial cells [14], an event also reproduced by
exposing human skin to UVB [12]. In aggre-
gate, these findings have led to the following
hypothesis: UVB triggers a photoreaction
which induces mast cell secretion leading to
local TNF-· release and ultimately to inhibi-
tion of the contact hypersensitivity reaction
All chemicals except for C-UA were purchased
from Sigma Chemical Company (St. Louis, Mo.,
USA).
Preparation of Purified C-UA
C-UA was prepared following a 30-hour exposure
of an aqueous saturated solution (3.5 mg/ml) of T-UA
to 313 nm of UV light using a spectrolinker (Spectron-
ics) UV cabinet. An equilibrium ratio of 70:30 of
cis:trans isomers was achieved with typical yields of
1.0 mg/100 J/cm2/day. C-UA was purified (195%) by
Dowex-1 X8 (Biorad) ion exchange chromatography
with a linear gradient of ammonium acetate. The
pooled eluates containing the C-UA fractions were
characterized by HPLC as depicted in figure 1. The
purified materials were lyophilized and stored under
vacuum at room temperature until further formulation
and use. Nuclear magnetic resonance studies con-
firmed the molecular structure to be 199% the cis iso-
(CHR). A logical candidate molecule for ini- mer. The solubility of purified C-UA was determined
Skin Pharmacol Appl Skin Physiol
1999;12:18–27
Mast Cell Degranulation Abrogates
CHR
19

sults: 2.5 g of hydroxypropyl methylcellulose (HPMC),
75 ml of ethanol and 0.1–20 g (w/v) of active agents
added as solids and brought to 100 ml with deionized
sterile water to make the final water-in-ethanol gel.
Placebo HPMC gels contained added weight in water
(w/v) in place of active agents.
Skin Permeation
Skin permeation studies were performed as pre-
viously described [22]. Excised full-thickness dorsal
mouse skin or human epidermis was prepared from
human banked skin (New York Firefighters Skin
Bank, New York, N.Y., USA) and quadruplicate 0.79-
cm2 skin samples were positioned epidermis side up in
upright Franz diffusion cells, with the dermal side in
contact with 7.5 ml of 1 ! phosphate-buffered saline
on the receptor phase medium. Approximately 0.5 ml
of purified C-UA sample was placed in the donor com-
partment in contract with the epidermis, and the open-
ing was sealed with parafilm. The cells were incubated
in a heating module (Reacti-Therm; Pierce Co., Rock-
°
ford, Ill., USA) at 37 C and continuously stirred. Sam-
ples were removed from the receptor phase medium at
regular time intervals, and the amount of C-UA per-
meated was assessed by HPLC analytical methods.
Fluxes were calculated by linear-regression analysis of
a plot of total accumulated amount of C-UA per-
meated at each time point over the linear portion of the
plot.
Fig. 1. HPLC analysis of solutions containing: T-UA
in water ( ), a 70:30 mixture of C-UA:T-UA obtained
after equilibrium (30 h) of the photoisomerization
reaction ( ) and chromatographically purified C-UA
). Ordinate: relative absorbance units; abscissa: elu-
a
b
(c
tion time (min). Chemical structures of T-UA and
C-UA are depicted in the inserts.
Mast Cell Degranulation and E-Selectin Induction
Neonatal foreskin organ culture was prepared as
previously described [14] and exposed to various con-
centrations of test agents in culture medium. The tis-
sue was cryostatically preserved, sectioned and stained
with a monoclonal antibody to human mast-cell-spe-
cific chymase.
Mast cell profiles were identified by the immuno-
peroxidase staining procedure [23] and were counted
per square millimeter (at least 3 mm3 evaluated per
section). The positive control was 5 mM of morphine
sulfate and the negative control was culture medium.
E-selectin expression on dermal microvascular vessels
was monitored by exposing human skin organ cultures
to varying concentrations of C-UA, T-UA, control
TFN-· and culture medium alone. The tissue was cryo-
preserved, sectioned and stained with primary antihu-
man E-selectin antibodies. Endothelial vessel profiles
were identified by an immunocytological staining pro-
cedure and by semiquantitative image analysis as pre-
viously described [14]. A total of 30 random micro-
scopic sections were evaluated, and the presence of
immunoperoxide-positive stain on the endothelial cell
and found to be 150.7 mg/ml in ethanol and 25.2 mg/
ml in water. By contrast, the solubility of the trans iso-
mer was 3.7 mg/ml in ethanol and 1.1 mg/ml in water.
High-Pressure Liquid Chromatography
HPLC was employed to characterize mixtures or
chromatographically purified solutions of C-UA. Sam-
ples were dissolved in 0.02 M monobasic phosphate
buffer containing 3 ml/l acetic acid, 1 g/l heptanesul-
fonic acid sodium salt and 6% acetonitrile. The uro-
canic acid isomers were resolved on a C18 Bondapak
(10-Ìm) column (Waters). C-UA eluates at 6–7 min
and T-UA is well separated from C-UA at 7.5–8.5 min
of elution time.
Topical Countersensitizer Formulation
The following topical gel formulation was used as a
vehicle in all of the experiments reported under Re-
Skin Pharmacol Appl Skin Physiol
1999;12:18–27
20
Wille/Kydonieus/Murphy

membrane was taken as evidence of a ‘positive micro-
vessel profile’.
Mouse skin
150
Sensitization and Immune Tolerance of Mice
Balb/c mice were housed in the Bristol-Myers
Squibb Company animal facility at Lawrenceville,
N.J., and protocols were approved by the local AALAC
committee. Four groups of 8 mice each were employed
in both the sensitization and countersensitization pro-
tocols. Group A, which served as the positive control,
was sensitized by a single application of 0.1 ml of 1%
dinitrochlorobenzene (DNCB) in acetone on shaved
abdominal skin. Five days later, 20 Ìl of 1% DNCB
was applied to one ear to elicit a sensitization response.
Group B, which served as the negative control, was
sensitized as group A but was challenged with acetone
on one ear. Group C animals had Finn chambers
placed on shaved abdominal skin which contained
0.2 ml of HPMC gel (vehicle control). Group D ani-
mals had Finn chambers placed on shaved abdominal
skin, which contained either C-UA, chloroquine or
capsaicin in vehicle. Finn chambers were applied 24 h
prior to sensitizing the mice as described for the con-
trol groups. Ear thickness measurements were con-
ducted at 0, 24, 48 and 72 h after ear challenge with 1%
DNCB, using an engineer’s micrometer caliper. The
DNCB in acetone was applied within the circumfer-
ence outlined by the Finn chamber, and the C-UA,
chloroquine and capsaicin applications extended well
beyond the circumference covered by the DNCB skin
application. To test for the induction of immune toler-
ance, two separate groups of mice were employed.
Group A mice were chosen from DNCB-treated ani-
mals, whose sensitization response was significantly
suppressed earlier by a single 24-hour exposure to C-
UA. These mice were then reexposed to DNCB at a
different skin site (shaved backs) 3 weeks after DNCB
treatment at the primary skin site. Group B control
mice were completely naive and age-matched to group
A mice. They were shaved on the abdomens and
received acetone at the time when group A mice were
sensitized on the abdomen. As in the sensitization pro-
tocol, all animals were painted 5 days later on one ear
with 1% DNCB in acetone to elicit an immune
response. Ear thickness measurements were made 0,
24 and 48 h later as described above.
100
50
Human skin
0
0
5
10
15
20
Concentration in gel (%)
Fig. 2. Permeation of C-UA through human (P) and
mouse skin ($) in vitro. Ordinate: observed fluxes;
abscissa: concentration of C-UA in HPMC gel.
Results
C-UA in Gel Formulation Permeates
Mouse and Human Skin
Figure 2 summarizes results on the per-
meation of C-UA through mouse and human
skin. A roughly parabolic relationship is ob-
served between flux (Ìg/cm2/h) and the con-
centration of the topically applied C-UA in gel
vehicle for both mouse and human skin. As
expected, mouse skin with a thinner stratum
corneum than human skin was 2–3 times
more permeable than human skin over the
concentration range tested. Percutaneous ab-
sorption of C-UA formulated in the ethanolic
gel vehicle (75% ethanol in 2.5% HPMC) is
enhanced over an aqueous vehicle by more
than 5-fold (data not shown). Of further rele-
vance for the use of C-UA as a topical agent
for transdermal delivery is the fact that signif-
icant fluxes were achieved in the range of
C-UA concentrations needed to induce sup-
pression of CHR, i.e. 10–50 Ìg/cm2/h.
Statistical Methods
The Student t test was applied to the means of all
animal group data to determine whether the differ-
ences in means were significant at the ^0.05 level of
significance.
Skin Pharmacol Appl Skin Physiol
1999;12:18–27
Mast Cell Degranulation Abrogates
CHR
21

120
100
80
60
40
20
0
MSO4 vs. SCM p = 0.0037
SCM vs. 1 mg p = 0.003
SCM vs. 10 mg p = 0.001
Fig. 3. Depletion of mast-cell-spe-
cific chymase by C-UA in human
skin organ culture. Positive con-
trol, 5 mM morphine sulfate
(MSO4); negative control, skin cul-
ture medium (SCM); C-UA treat-
ment, dose = 1 Ìg/ml; C-UA treat-
ment, dose = 10 Ìg/ml; ordinate:
chymase-positive mast cells per
microscopic field. The box encloses
the
p values calculated from
MSO4
SCM
C-UA 1 mg
C-UA 10 mg
ANOVA two-way comparisons as
shown.
TNF-a,
positive
control
CIS data
TRANS data
100
80
60
40
20
0
Fig. 4. Induction of E-selectin ex-
pression by C-UA versus T-UA.
Controls (left panel): positive con-
trol, 100 ng/ml of TNF-·; negative
control, skin culture medium
(SCM). Dose (right panel): C-UA
and T-UA were added at 10, 75
and 200 Ìg/ml to the culture me-
dium of the human skin organ cul-
tures. Ordinate: number of positive
microvessel profiles per square
millimeter. Oval boxes enclose the
probability values p for two-way
ANOVA.
CIS vs.
TRANS
p = 0.0001
SCM,
negative
control
CIS vs.
TRANS
p = 0.0017
CIS vs.
TRANS
p = 0.0019
10 mg
75 mg
Dose
200 mg
Controls
C-UA Depletes Mast Cell Chymase
with 5 mM morphine sulfate (37 B 15 posi-
Figure 3 shows that as little as 1 Ìg/ml of tive mast cells/mm2), a known mast cell secre-
C-UA significantly degranulated mast cells, as tagogue, and which results in TNF-· libera-
shown by a 75% reduction in number of chy- tion [13]. In addition, in skin organ culture
mase-positive mast cells versus untreated experiments, we found that T-UA was less
controls (88 B 12 positive mast cells/mm2) in effective in depleting mast cell chymase than
preparations stained immunocytochemically C-UA by a factor of 5 at the optimal mast-
by specific antihuman chymase antibody. The cell-degranulating concentration of C-UA
effect of C-UA was as potent as treatment (data not shown).
Skin Pharmacol Appl Skin Physiol
1999;12:18–27
22
Wille/Kydonieus/Murphy

Table 1. C-UA inhibits contact sensitivity
C-UA-Stimulated Mast Cell
Degranulation Induces TNF-·-Dependent
E-Selectin
The induction of E-selectin expression was
monitored in human skin organ cultures ex-
Treatment
Ear
Ear
swelling
Suppression
%
thickness
mm ! 103 mm ! 103
posed to varying concentrations of C-UA or None (HPMC gel)
24 h
72 h
244B8
–
–
–
–
T-UA. Previous work in our laboratory had
established that the peak of E-selectin expres-
sion on the dermal microvessels occurs at 6 h
after mast cell secretagogue, substance P or
UVB exposure of human skin organ cultures,
and results from the release of mast cell
TNF-· [12–14]. Figure 4 presents data show-
ing that C-UA results in significantly greater
levels of E-selectin expression on dermal mi-
crovessels than equivalent concentrations of
T-UA. C-UA at 10 Ìg/ml was equivalent to
the positive control (5 ng/ml of TNF-·).
241B4
DNCB (100 Ìg)
24 h
72 h
332B21
313B32
89
72
–
–
C-UA (1%) + DNCB (100 Ìg)
24 h
72 h
270B16
288B23
26
48
71
33
C-UA (5%) + DNCB (100 Ìg)
24 h
72 h
261B15
270B15
18
30
80
59
C-UA Inhibits Contact Sensitivity
Table 2. Purified C-UA inhibits contact hypersensi-
Our previous work with ethacrynic acid [8]
showed that inhibitory effects were indepen-
dent of whether DNCB or sensitizing drugs
were used to induce CHR. We chose DNCB
here because the protocol is less complex.
Table 1 shows that purified C-UA incorpo-
rated at either 1% (w/v) or 5% (w/v) in a 2.5%
HPMC ethanolic gel and applied 24 h prior to
DNCB sensitization is effective in preventing
the murine ear swelling response induced by
DNCB. With 1% concentration of C-UA in
vehicle, immune suppression at 24 h after
challenge was 170%, and with 5% C-UA in
vehicle, there was an 80% inhibition of CHR.
Inhibition of CHR was observed up to 72 h
after challenge. This experiment was repeated
tivity: dose response effect
Treatment
dose, %
Number of
experiments (24 h), %
Inhibition
5.0
7.5
10
4
3
4
60B6
69B5
76B10
The percent inhibition shown is significantly dif-
ferent from both the negative control (placebo HPMC
gel) and the positive control (1% DNCB) at the 0.05%
level of confidence by the Student t test.
more than a dozen times with similar results. of purified C-UA (table 2). The extent of im-
No sign of local irritation or inflammation mune suppression increased with concentra-
was ever seen as a result of prolonged skin tions of C-UA up to 10%. An additional ques-
contact (124 h) with C-UA in vehicle at any tion we addressed experimentally was wheth-
of the tested concentrations below 20% in er there was an advantage to pretreatment
HPMC gel.
versus codelivery of C-UA with the sensitizer.
In a second series of experiments, mouse Table 3 demonstrates our finding that code-
skin was exposed for 24 h to increasing doses livery is less effective than a 24-hour pretreat-
Skin Pharmacol Appl Skin Physiol
1999;12:18–27
Mast Cell Degranulation Abrogates
CHR
23

Table 3. Comparison of the codelivery and 24-hour
pretreatment on immune suppression by C-UA
develop formulations with maximum pene-
tration. A number of nonspecific detergents
including sodium dodecyl sulfate (10%) and
the dimeric acid of linoleic acid (15%) were
ineffective in suppressing DNCB-induced
CHR.
Treatment
Ear
Ear
swelling
Suppression
%
thickness
mm ! 103 mm ! 103
None (HPMC gel), n = 5
24 h
48 h
238B9
230B9
–
–
–
–
Induction of Immune Tolerance
Figure 5 shows that purified C-UA in-
duced a state of immune unresponsiveness or
immune tolerance. Mice displaying immune
tolerance had received only a single applica-
tion of C-UA prior to DNCB sensitization (on
the abdomen) and displayed a marked sup-
pression of ear swelling after DNCB chal-
lenge. There was a 70 and 82% suppression at
24 h after elicitation for 1% C-UA gel and 5%
C-UA gel, respectively. These same mice were
then resensitized with DNCB at a second skin
site (shaved back) and rechallenged with
DNCB on a different ear than the first DNCB
challenge without any intervening C-UA
treatment. Control mice were age-matched
litter mates that had been sensitized on one
ear only prior to their second-site DNCB sen-
DNCB (100 Ìg), n = 7
24 h
48 h
292B13
291B20
55
60
–
Codelivery (C-UA 5% + DNCB 100 Ìg), n = 8
24 h
48 h
277B27
273B23
39
43
28
29
None (HPMC gel), n = 5
24 h
48 h
230B4
251B9
–
–
–
Pretreatment (C-UA 5% + DNCB 100 Ìg), n = 6
24 h
48 h
260B19
259B12
29
20
52
69
n = Number of mice per group.
ment protocol at both 24 and 48 h after chal- sitization. The data show that one-time pre-
lenge. It is presumed that DNCB penetrates treatment with 1% C-UA gel or with 5%
mouse skin too quickly and acts before C-UA C-UA gel induced greater than 50 and 43%
can bring about mast cell degranulation.
suppression relative to control litter mates,
We tested two other mast-cell-degranulat- respectively. As expected, the extent of inhibi-
ing agents (MCDAs), chloroquine [24] and tion exhibited following immune tolerance in-
capsaicin [25], for their ability to inhibit duction was less than when these mice had
DNCB sensitization. The same protocols only a single round of DNCB sensitization
were employed as for suppression by C-UA. and challenge. It can be inferred from these
When 2% chloroquine in vehicle was applied results that C-UA does not suppress CHR
24 h prior to DNCB, it significantly inhibited through an immunological null event.
the ear swelling response (140%). Similar-
ly, 1% capsaicin in vehicle applied 24 h
prior to DNCB inhibited DNCB sensitization
(130%). These values were significant at the
Cromolyn Does Not Inhibit Contact
Sensitization
We hypothesized that agents that stabilize
0.05 level by Student t test. These experi- mast cell membranes should prove to be inef-
ments were performed to determine if fective inducers of immune suppression. For
MCDAs as a class would inhibit sensitization; this experiment, we chose sodium cromolyn.
no optimization studies were performed to Because of its poor skin permeability, it was
Skin Pharmacol Appl Skin Physiol
1999;12:18–27
24
Wille/Kydonieus/Murphy

injected intradermally a few hours prior to
initiation of sensitization. Preliminary studies
showed that injection of 20 Ìg of cromolyn
alone had no effect on ear swelling of nonsen-
sitized mice. Also mice were sensitized with
trinitrobenzene sulfonate, a water-soluble an-
tigen that cross-reacts with DNCB [26]. Ta-
ble 4 presents results confirming that cromo-
lyn is ineffective in preventing sensitization,
and, although not rising to the level of signifi-
cance, there was a trend toward cromolyn
stimulation of CHR. It has previously been
shown that a dose of 20 Ìg of cromolyn pre-
vents UV-induced mast cell degranulation
[pers. commun. from G.F.M.].
24 h
48 h
3 weeks
100
80
60
40
20
0
2
10
Concentration (mg)
Fig. 5. Induction of immune suppression and immune
tolerance by C-UA. Balb/c mice (8 per group) received
a single application of C-UA prior to primary-site sen-
sitization with DNCB and displayed immune suppres-
sion at 24 h (i) and 48 h (n) after ear challenge; these
same mice were then resensitized 2–3 weeks later with
1% DNCB at a second skin site and subsequently
rechallenged 5 days later. The resensitization data (g)
show that there is a significant reduction in ear swell-
ing relative to control mice that did not receive C–UA
treatment during primary-site sensitization. In all
cases, the bar values had a standard error of the mean
of less than 10%.
Discussion
This is the first study to present the novel
findings that C-UA, but not T-UA, promotes
mast cell degranulation in an ex vivo human
skin model. This is likely an underestimate of
the superiority of C-UA as an MCDA in vivo,
as T-UA is sequestered in the upper layers of
the epidermis where, as we have shown here,
C-UA is at least 20 times more soluble than
T-UA and is likely to rapidly diffuse into the
dermis upon UVB-induced isomerization.
We also report that C-UA, but not T-UA, is
more effective in inducing E-selectin expres-
sion on dermal microvessels, and, thus, C-UA
produces local effects consonant with release
of mast cell stores of TNF-·. Finally, we show
that a single application of highly purified
C-UA (195%) during the primary sensitiza-
tion phase is an effective inhibitor of contact
hypersensitivity and also confers a state of
immune tolerance.
Table 4. Effect of cromolyn on contact sensitivity
Treatment
Ear
Ear
swelling
Suppression
%
thickness
mm ! 103 mm ! 103
None (saline)
24 h
17B3
17B1
–
–
–
–
48 h
TNBS (2.4 Ìg)
24 h
101B16
161B28
84
144
–
–
48 h
TNBS (2.4 Ìg) + cromolyn (20 Ìg)
24 h
48 h
112B12
169B16
96
169
+14
+17
It remains an open question as to whether
all MCDAs possess the ability to induce im-
mune suppression and immune tolerance,
when administered according to the protocol
described herein. We examined two known
In this case, no suppression was observed; the cro-
molyn group was more positive than the group treated
with trinitrobenzene sulfonate (TNBS).
Skin Pharmacol Appl Skin Physiol
1999;12:18–27
Mast Cell Degranulation Abrogates
CHR
25

skin irritants and potential MCDAs, chloro- induction of sensitization may interrupt the
quine and capsaicin, and each inhibited CHR normal sequence of steps required for Langer-
in this mouse model. We firmly established hans cell maturation.
that C-UA is both a potent MCDA and inhibi-
In aggregate, these results indicate that
tor of CHR. In this regard, cromolyn, a mast stimulation of mast cell degranulation at criti-
cell membrane stabilizer, failed to induce im- cal time points in the induction phase of CHR
mune suppression and actually increased the interferes with the acquisition of sensitiza-
extent of the DNCB-induced CHR.
tion. The chromophore model predicts that
Perhaps, the most compelling evidence other MCDAs applied to skin prior to the
that C-UA mediates UVB-induced immuno- induction phase of CHR could produce a local
suppression is evidence that dermal injection cytokine environment that is inhibitory to
of an antibody to TNF-· reverses the immu- induction of sensitization, e.g. by releasing
nosuppressive effects of both UVB [11] and cytokines that alter Langerhans cell expres-
C-UA [19], suggesting that TNF-· released sion of B7 costimulatory signals or through
locally in the area of Langerhans cells is induction of IL-10 and other immunosup-
responsible for the cascade of events mediat- pressive cytokines [32].
ing immunosuppression. Recently, Yamazaki
et al. [27] reported that TNF-·, RANTES and
Acknowledgments
MCP-1 are potent chemotactic attractants for
Langerhans cells and may influence their mi-
gration from the epidermis to regional lymph
nodes during contact sensitization. Further,
Jakob and Udey [28] reported that both IL-1
and TNF-· aid in the emigration of Langer-
hans cells by reducing E-cadherin interaction
of Langerhans cells with epidermal keratino-
cytes. In experiments using anti-IL-1ß anti-
body administered systemically prior to skin
sensitization, Langerhans cell accumulation
in the draining lymph nodes was abolished, as
was TNF-·-induced migration and dendritic
cell accumulation [29]. Interestingly, Price
et al. [30] have reported that antibody to
·6-integrin (laminin) but not an anti-·4-inte-
grin (fibronectin) blocks Langerhans emigra-
tion and dendritic cell accumulation during
the induction phase of contact sensitization.
These findings support the early discovery
that ·6-integrin is expressed on epidermal
Langerhans cells exposed to TNF-· [31] and
supports the hypothesis that Langerhans cell
migration is orchestrated by a cascade of cyto-
kines. Inappropriate overexposure of Langer-
hans cells to mast-cell-derived TNF-· prior to
We would like to thank Dr. J. Wayne Streilein for
sharing the results of his experiments on sodium cro-
molyn as part of a ConvaTec sponsored project
awarded to the University of Miami. We also acknowl-
edge the help of Dr. François Njeha in preparing puri-
fied C-UA, Ms. Prafulla Amin for her excellent HPLC
analytical work and Ms. Lisy Wang for her technical
help with the animal studies.
Skin Pharmacol Appl Skin Physiol
1999;12:18–27
26
Wille/Kydonieus/Murphy

References
1 Fischer AA: Contact dermatitis
from topical medicaments. Semin
Dermatol 1982;1:49–57.
2 Maibach HI: Clonidine: Irritant and
contact dermatitis assays. Contact
Dermatitis 1985;12:192–196.
13 Walsh LJ, Trinchieri G, Waldorf
HA, Whitaker D, Murphy GF: Hu-
man dermal mast cells contain and
release tumor necrosis factor-·
which induces endothelial leukocyte
adhesion molecule 1. Proc Natl
Acad Sci USA 1991;88:4220–4224.
14 Klein LM, Lavker RM, Matis WL,
Murphy GF: Degranulation of hu-
man mast cells induces an endothe-
lial antigen central to leukocyte ad-
hesion. Proc Natl Acad Sci USA
1989;86:8972–8976.
15 Laerma AI, Maibach HI: Topical
cis-urocanic acid suppresses both in-
duction and elicitation of contact
hypersensitivity in Balb/c mice.
Acta Derm Venereol (Stockh) 1995;
77:272–275.
16 Yarosh DB, Alas L, Kibitel AL, Ull-
rich SE: Urocanic acid, immunosup-
pressive cytokines, and the induc-
tion of human immunodeficiency
virus. Photodermatol Photoimmu-
nol Photomed 1992;9:127–135.
17 Moodycliffe AM, Kimber I, Norval
M: The effect of ultraviolet B irria-
diation and urocanic acid isomer on
dendritic cell migration. Immunolo-
gy 1992;77:394–399.
18 Kurimoto I, Streilein JW: Deleteri-
ous effects of cis-urocanic acid and
UVB radiation on Langerhans cells
and on the induction of contact hy-
persensitivity are mediated by tu-
mor necrosis factor-·. J Invest Der-
matol 1992;99:69s–70s.
19 Moodycliffe AM, Bucana CD,
Kripke ML, Norval M, Ullrich SE:
Differential effects of a monoclonal
antibody to cis-urocanic acid on the
suppression of delayed and contact
hypersensitivity following ultravio-
let irradiation. J Immunol 1996;
157:2891–2899.
20 El-Ghorr AA, Norval M: A mono-
clonal antibody to cis-urocanic acid
prevents the UV-induced changes in
Langerhans cells and DTH re-
sponses in mice, although not pre-
venting dendritic cell accumulation
in lymph nodes draining the site of
irradiation and contact hypersensi-
tivity responses. J Invest Dermatol
1994;105:264–269.
ment of transdermal drug-induced
hypersensitivity. Proc Int Symp Co-
trol Rel Bioact Mater 1995;22:119–
120.
22 Shaw JE, Prevo ME, Amkraut AA:
Testing of controlled release trans-
dermal dosage forms. Arch Derma-
tol 1987;123:1548–1556.
23 Kaminer MS, Lavker RM, Walsh
LJ, Whitaker D, Ziverman B, Mur-
phy GF: Extracellular localization of
human connective tissue mast cell
granule contents. J Invest Dermatol
1991;96:1–8.
24 Green KB, Lim HW: Effects of
chloroquine on release of mediators
from mast cells. Skin Pharmacol
1989;2:77–85.
25 Maurer M, Paus R, Galli SJ: Capsai-
cin-induced cutaneous inflamma-
tion in mast cell-deficient mice. J
Invest Dermatol 1993;101:722.
26 Shimizu M, Streilein JW: Transepi-
dermal induction of contract hyper-
sensitivity in mice with a water-solu-
ble antigen. J Invest Dermatol 1993;
101:749–753.
27 Yamazaki S, Yokezeki H, Sahoh T,
Katayama I, Nishioka K: TNF-·,
RANTES, and MCP 1 are major che-
moattractants of murine Langerhans
cells to the regional lymph nodes.
Exp Dermatol 1998;7:35–41.
28 Jakob T, Udey MC: Regulation of
E-cadherin-mediated adhesion in
Langerhans cell-like dendritic cells
by inflammatory mediators that mo-
bilize Langerhans cell in vivo. J Im-
munol 1998;160:4067–4073.
29 Cumberbatch M, Dearman RJ,
Kimber I: Langerhans cells require
signals from both tumor necrosis
factor-alpha and interleukin-1 beta
for migration. Immunology 1997;
92:388–395.
30 Price AA, Cumberbatch M, Kimber
I, Ager A: Alpha 6 integrins are re-
quired for Langerhans cell migra-
tion from the epidermis. J Exp Med
1997;186:1725–1735.
3 Rietschel RL, Fowler JE Jr (eds):
Fisher’s Contact Dermatitis, ed 4.
Baltimore, Williams
1995.
&
Wilkins,
4 Diezel W, Gruner S, Diaz LA, An-
halt GJ: Inhibition of cutaneous
contact hypersensitivity by calcium
transport inhibitors lanthanum and
diltiazem. J Invest Dermatol 1989;
93:322–326.
5 Amkraut A, Jordan W, Taskovich L:
Effect of co-administration of corti-
costeroids on the development of
contact sensitization. J Am Acad
Dermatol 1996;35:27–31.
6 McFadden J, Bacon K, Camp R:
Topically applied verapamil hydro-
chloride inhibits tuberculin-induced
delayed-type hypersensitivity reac-
tions in human skin. J Invest Der-
matol 1992;99:784–786.
7 Gallo RL, Granstein RD: Inhibition
of allergic contact dermatitis and ul-
traviolet irradiation-induced tissue
swelling in the mouse by amiloride.
Arch Dermatol 1989;125:502–506.
8 Kalish RS, Wood JA, Kydonieus A,
Wille JJ: Prevention of contact hy-
persensitivity to topically applied
drugs by ethacrynic acid: Potential
application to transdermal drug de-
livery. J Control Rel 1997;48:79–
87.
9 DeFabo E, Noonan F: Mechanism
of immune suppression by ultravio-
let light in vivo. J Exp Med 1983;
157:84–98.
10 Streilein JW: cis-Urocanic acid in-
duces immune tolerance. J Invest
Dermatol 1993;105:76.
11 Kurimoto I, Streilein JW: cis-Uro-
canic acid suppression of contact hy-
persensitivity induction is mediated
via tumor necrosis factor-·. J Invest
Dermatol 1992;148:3072–3078.
12 Lavker RM, Kaminer MS, Murphy
GF: Mast cell degranulation results
in endothelial activation after acute
exposure of human skin to UV irra-
diation; in Marks R, Plewig G (eds):
Environmental Threat to the Skin.
London, Dunitz, 1992, pp 125–129.
31 Ioffreda MD, Whitaker D, Murphy
GF: Mast cell degranulation upregu-
lates ·6 integrins on epidermal Lan-
gerhans cells.
J Invest Dermatol
1993;101:150–154.
21 Wille JJ, Njieha F, Amin P, Kydo-
nieus A: Topical delivery of mast
cell degranulating agents for treat-
32 Lambert RW, Granstein RD: Neu-
ropeptides and Langerhans cells.
Exp Dermatol 1998;7:73–80.
Skin Pharmacol Appl Skin Physiol
1999;12:18–27
Mast Cell Degranulation Abrogates
CHR
27