VDE_273.

fm Page 339 Thursday, November 15, 2001 7:20 PM

Veterinary Dermatology 2001, 12, 339 346

Skin mast cell histamine release following stem cell factor and
high-affinity immunoglobulin E receptor cross-linking
in dogs with atopic dermatitis

Blackwell Science Ltd

BRUCE HAMMERBERG,* THIERRY OLIVRY and SUSAN M. ORTON*

*Department of Microbiology, Pathology and Parasitology and
Department of Clinical Science, College of Veterinary Medicine, North Carolina State University,
Raleigh, NC 27606, USA
(Received 21 March 2001; accepted 8 June 2000)

Abstract Stem cell factor (SCF) influences mast cell activation and inflammatory mediator release, and is
elevated in tissues undergoing allergic inflammation. Wheal formation in response to the injection of SCF or antiimmunoglobulin (Ig )E antibody injection was compared between normal (n = 10) and nonlesional atopic
(n = 10) canine skin. In situ SCF secretion was compared between lesional and nonlesional skin using immunohistochemistry. Histamine release by skin cell suspensions after stimulation with SCF, concanavalin A (ConA)
or rabbit anticanine IgE antibodies was compared between normal and atopic dogs. All dogs exhibited strong
responses to intradermal SCF injection at 10 and 50 ng mL1. Atopic dogs had significantly (P = 0.002) larger
wheal responses to anti-IgE than normal dogs; but there was no difference in numbers of skin mast cells bearing
IgE as detected by immunohistochemistry. Only atopic dogs exhibited interstitial deposition of SCF in both
lesional and nonlesional skin specimens. Median histamine release stimulated by SCF in the absence of IgE from
lesional skin cells was higher in atopic than normal dogs (P = 0.04). These experiments suggest that dermal SCF
secretion could potentiate histamine release following IgE receptor cross-linking and thus, could be one of the
explanations for the inherent mast cell hyperexcitability observed in canine atopic dermatitis.

Keywords: atopic dermatitis, histamine, mast cell, stem cell factor.

INTRODUCTION
Naturally occurring allergic diseases in dogs, as manifested in skin and intestine, are remarkably similar to
human allergic diseases of these tissues.13 In addition,
these diseases have been very well characterized as
inherited traits in different lines of dogs.4 Atopic dermatitis in dogs is immunoglobulin (Ig)E dependent,4
but total serum IgE does not correlate with disease
manifestation.5,6 We have shown that grossly nonlesional skin of atopic individuals differs markedly from
the skin of nonatopic dogs regarding immune and
inflammatory cell populations.7,8 Factors important
to the presence of these cell populations may be
dependent upon inherited differences in the functional
properties of mast cells and/or differences in skin cell
production of mast cell-activating cytokines. These
inherited differences would be present throughout the
skin, including grossly nonlesional sites. Stem cell
factor (SCF) is a cytokine produced by fibroblasts and
keratinocytes that has been shown to enhance human,9
as well as canine,10 mast cell degranulation response
to cross-linking of high-affinity receptors for IgE
(Fc RI).

Correspondence: Bruce Hammerberg. Tel.: +1-919-513-6226, Fax:

+1-919-513-6455, E-mail: bruce_hammerberg@ncsu.edu
2001 Blackwell Science Ltd

Naturally occurring hyperexcitability by blood

leukocytes releasing a greater percentage of their
histamine in response to anti-IgE cross-linking of
Fc RI has been shown to distinguish atopic dogs from
nonatopic dogs.11 Mast cells from the skin of atopic
dogs released a greater percentage of histamine than
cells from the skin of normal dogs in response to concanavalin A (ConA).12 In addition, it has been reported
that human asthma patients have primed basophils,
which release significantly more histamine than normal basophils when incubated with thapsigargin.13
This study was conducted to determine whether
mast cells from either lesional or nonlesional sites of
atopic dogs were more responsive to SCF than mast
cells from the skin of normal dogs, and to determine if
Fc RI-mediated histamine release at nonlesional sites
was different between atopic and nonatopic dogs.

METHODS
Dogs
Dogs that presented at the Dermatology Service at the
North Carolina State University College of Veterinary
Medicine were determined to be affected with atopic
dermatitis based on standard criteria.14 In addition,
an important feature of diagnosis was the exclusion of
other causes of pruritic dermatitis. Dogs diagnosed as
339

VDE_273.fm Page 340 Thursday, November 15, 2001 7:20 PM

340

B. Hammerberg et al.

atopic and included in this study were evaluated for

Sarcoptes scabei and helminth parasite infections, primary flea bite hypersensitivity, food hypersensitivity
and insect hypersensitivity based on standard diagnostic methods. The identification of allergens involved
was made using a 70-allergen intradermal test and a
32-allergen-specific IgE serologic test. The 10 dogs
included in this study exhibited positive responses to
more than three allergens on intradermal testing and
more than one allergen on serologic testing, and were
determined to not be infected with S. scabei, fleas or
intestinal parasites. All dogs received an elimination
diet to rule out food-induced dermatitis (intolerance,
allergies).
Ten normal dogs were of mixed breed, maintained
indoors under NIH guidelines by the Laboratory Animal Resources Department at North Carolina State
University. These dogs were judged by the criteria
stated above to not be affected with atopic dermatitis.
Intradermal testing for responses to SCF and anti-IgE
The standard protocol for intradermal testing for allergen hypersensitivity was conducted routinely at the
College of Veterinary Medicine Dermatology Service.
The site for testing was the side of the dorsal thoracic
region opposite that used for skin biopsy. Recombinant canine stem cell factor (rc-SCF; Amgen Inc.,
Thousand Oaks, CA, USA) and rabbit IgG antidog
IgE were injected intradermally (0.1 mL) at 50 ng mL1
and 10 ng mL1 for rc-SCF, and at 20 g mL1 for
rabbit antidog IgE. The wheal-and-flare reaction at the
injection site (average of the diameters of the wheal
measured in two perpendicular directions) was scored
by a test-blinded individual 20 min after intradermal
injections.
Rabbit IgG anticanine IgE
Rabbit IgG anticanine IgE antibodies used for histamine release studies and for intradermal injection were
produced in New Zealand white rabbits immunized
with 100 g of the canine monoclonal IgE in Titermax
adjuvant (Sigma). The rabbits were boosted after
4 weeks with 50 g of IgE in Titermax adjuvant
(Sigma) and sera were harvested 1014 days post
boost. IgG antibodies were purified by elution from
a protein G-agarose column (Zymed, San Francisco,
CA, USA) after removing cross-reactive anticanine
IgG antibodies by passing sera over an affinity column
of canine IgG linked to Affi-gel beads (Bio-Rad,
Melville, NY, USA). Optimal amounts of rabbit IgG
anti-IgE used for intradermal injections and for in vitro
histamine release were determined by serial dilution
assays.
Immunohistochemistry for SCF and IgE
Skin biopsies were placed in OCT compound (Sakura
Finetek USA, Inc., Torrance, CA, USA), flash frozen
in liquid nitrogen and kept at 70 C until 5-m sections were cut for immunohistochemical staining. Sections were fixed in cold acetone for 10 min and washed
2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 339 346

in phosphate-buffered saline (PBS) for 3 min. To detect

SCF, slides were incubated in 10% normal goat serum
in PBS with 5% bovine serum albumin (BSA) for
20 min, followed by incubation in 0.3% H2O2 in methanol for 5 min and washed in PBS. The slides were
incubated with a canine SCF cross-reactive rabbit antimouse SCF (R&D Systems, Minneapolis, MN, USA)
diluted 1:2000 in PBS containing 1 g mL1 dog IgG
for 1 h in a moist chamber at room temperature. After
washing, the slides were incubated with biotinylated
goat antirabbit IgG (Vector Labs, Bulingame, CA,
USA) diluted 1:400 in PBS for 30 min in a moist chamber at room temperature. The slides were washed again
and incubated with streptavidin labelled with horseradish peroxidase (HRP; Zymed) diluted 1:400 in PBS
and incubated for 30 min. After a final wash, the
AEC chromagen (AEC Substrate kit, Biogenex, San
Ramon, CA, USA) was applied for 1 min. Slides were
counterstained with Harris haematoxylin (Sigma). To
detect IgE, slides prepared as above were incubated
with mouse monoclonal antidog IgE, 5.91, specific for
a linear, heat stable, epitope unique to epsilon chains,6
at 1 g mL1 for 1 h at room temperature in a moist
chamber. After washing in PBS, the slides were incubated for 30 min at room temperature with biotinylated horse antimouse IgG (Vector Labs) diluted
1:400 in PBS. The remainder of the procedure was as
described above for staining for SCF.
Enzyme-linked immunosorbant assay for serum IgE
Capture enzyme-linked immunosorbant assay (ELISA)
for IgE was performed using capture and detection
with mouse monoclonal antibody, 5.91. Monoclonal
antibody 5.91 at 10 g mL1 in 0.1 carbonate bicarbonate buffer, pH 9.6, was used to coat polystyrene
microtitre plates (Dynex Technologies, Inc., Chantilly,
VA, USA) at 4 C overnight. Coated plates were
blocked for 2 h at room temperature or overnight at
4 C with PBS plus 0.5% Tween 20 (Sigma; PBST)
containing 1% nonfat milk. A standard curve of serial
dilutions of monoclonal canine IgE15 (Bethyl Labs,
Montgomery, TX, USA), starting at 1 g mL1, was
run on each plate. Canine sera were serially diluted
from 1:10 to 1:160 and all samples were run in triplicate. After 2 h incubation, the plates were washed with
PBST and biotinylated monoclonal 5.91 at 1 g mL1
in PBST containing 10% normal mouse serum was
used to detect quantities of captured IgE by incubation on the plates for 2 h followed by washing and a
30-min incubation with HRP-labelled streptavidin
(Zymed), diluted 1:1000. Following a final series of
five washes with PBST, ABST substrate (Kirkegaard
and Perry Labs, KPL, Gaithersburg, MD, USA) was
added and colour development was read after 60 min
at 405 nm on a MicroTek microtiter plate reader
(Fisher Scientific, Atlanta, GA, USA).
Skin biopsy and cell culture
Multiple skin biopsy samples were taken from one side
of the dorsal thoracic region of normal and atopic

VDE_273.fm Page 341 Thursday, November 15, 2001 7:20 PM

Stem cell factor and histamine release

dogs. Atopic dogs were also biopsied at sites showing
characteristic skin lesions or sites typically associated
with lesional skin (abdominal or inguinal regions).
The skin samples were placed in minimal essential
medium (MEM; MediaTek, Raleigh, NC, USA), subcutaneous fatty tissue was removed and the skin was
finely chopped into fragments measuring 1 mm or less.
The fragments were washed in MEM and digested
(1 g tissue in 35 mL) in a mixture consisting of
MEM, 100 unit mL1 penicillin/streptomycin, 0.1% BSA,
2 mg mL1 collagenase (Type I, Sigma, St. Louis,
MO, USA), 1.2 mg mL1 hyaluronidase (Sigma) and
1 mg mL1 protease (Pronase E, Type XIV, Sigma).
The digest was incubated for 1 h at 37 C on a rotator
plate at 120 r.p.m. After 1 h the digest was passed
through a 150-m mesh nylon filter, a fresh digest
cocktail was added to the remaining fragments and
digested for an additional 1 h at 37 C with shaking.
The second digest was also passed through a 150-m
filter and pooled with the first. The digest was washed
twice with Hanks balanced salts solution (HBSS,
MediaTek) without Ca2+ and Mg2+ containing 0.1%
BSA. After the last wash, the digest was resuspended in
histamine release buffer that was HBSS containing
100 units mL1 penicillin/streptomycin, 0.1% BSA,
25 m HEPES, 1.3 m Ca2+ and 1.1 m Mg2+.
To correct for the loss of cell-bound IgE during
skin digestion, half of the cells were pre-incubated
for 30 min with 2 g mL1 monoclonal canine IgE15
(Bethyl Labs) followed by washing prior to incubation
with the various secretagogues. Histamine release
assays were also performed on cells treated exactly the
same without the addition of IgE. Cell counts and
viability were assessed using Trypan Blue. Cytospins
were prepared from each sample and stained with 0.1%
toluidine blue to determine the percentage of mast
cells in each digest.
Histamine release
Skin digest cells (5 105) were incubated with the
following treatments: histamine release buffer (spontaneous release), ConA (Sigma) at 100 g mL1, rc-SCF
(Amgen Inc.) at 1 g mL1 and rabbit IgG anti-IgE
polyclonal antibodies at 100 g mL1. After 30 min
at 37 C, the samples were centrifuged for 30 s at
12 000 g and supernatants harvested from the cell

341

pellets. Cell pellets were lysed by resuspension in water,

followed by three rounds of freeze/thaw cycles. Supernatants and pellet lysates were frozen at 70 C until
analysis. Histamine was measured using a commercially available ELISA (Coulter Immunotech, Hialeh,
FL, USA). Values for histamine release were expressed
as a percentage of total histamine.
Statistical analysis
The parametric Students t-test was used to compare
median values of serum SCF and IgE. These values
showed a Gaussian distribution. The nonparametric
Wilcoxon signed-rank test was used to determine statistical differences of medians and reported as P-values
of histamine release, because these values did not show
Gaussian distribution. Anticipated changes were compared by one-tailed tests as indicated. Two-tailed tests
were used for all other reported differences. Median
values were reported with 95% confidence intervals
(95% CI). Values were calculated using the
software (GraphPad Software, Inc., San Diego,
CA, USA).

RESULTS
Response to intradermal injections
Nonlesional sites on the lateral thorax area of atopic
dogs and corresponding sites on normal dogs were
injected with SCF, rabbit anticanine IgE, saline or histamine. The diameter of wheals formed 20 min after
injection was markedly greater in atopic dogs in
response to anti-IgE antibodies (Fig. 1). In contrast,
atopic and normal dogs responded equally strongly to
SCF injection at both 50 and 10 ng mL1 when compared with saline controls (Fig. 1). There was no difference between the two groups in response to histamine
injection (data not shown).
Distribution of SCF and IgE in skin
SCF was detected by immunohistochemistry in mast
cells, identified by Alcian Blue staining (not shown) in
serial sections, in lesional and nonlesional skin of
atopic dogs and of normal dogs (Fig. 2a). However,
only atopic dogs exhibited clear evidence of SCF distributed between dermal collagen fibres, external to

Figure 1. Swelling response 20 min after

intradermal injection of stem cell factor
(SCF), rabbit anti-IgE and saline. Two
concentrations of SCF were tested: 50 and
10 ng mL1. N designates normal dogs
(black squares). A designates atopic dogs
(open squares). Horizontal lines mark
median values. P-Values were calculated by
two-tailed Wilcoxon test.
2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 339346

VDE_273.fm Page 342 Thursday, November 15, 2001 7:20 PM

342

B. Hammerberg et al.

Figure 2. In normal canine skin (a), stem cell factor (SCF) was found in a granular pattern in or around dermal cells with round nuclei; there
was no interstitial SCF deposition. In contrast, dermal SCF was discovered primarily between collagen fibres of both nonlesional (b) and lesional
(c) atopic canine skin. The intensity of interstitial dermal SCF immunoreactivity was higher in lesional (c) than nonlesional (b) atopic skin.
Immunoperoxidase technique, SCF-specific monoclonal antibody, amino-ethyl-carbazol chromogen; bar = 30 m.

Figure 3. Serum immunoglobulin (Ig)E amounts in normal and

atopic dogs. Total IgE in sera taken at the time of skin biopsy from
normal dogs (n = 10) and atopic dogs (n = 10) was measured by enzymelinked immunosorbant assay (ELISA). There was no statistical
difference between the groups. Horizontal lines mark median values.

cells in both lesional and nonlesional skin (Fig. 2b,c).

All mast cells, identified by Alcian Blue staining, in
either lesional or nonlesional skin of atopic dogs, or
skin from normal dogs showed strong staining for
surface IgE, but no interstitial staining for IgE was
detected (data not shown). Total serum IgE values
were not statistically different between the normal and
atopic groups, medians of 1.5 (0.25.7) and 2.0 (1.4
4.4) g mL1, respectively (Fig. 3).
In vitro histamine release response to SCF,
anti-IgE and ConA
The percentage of mast cells obtained from enzymatic
digests of biopsies taken at lesional and nonlesional sites
of atopic dogs and sites corresponding to nonlesional
sites on normal dogs were not statistically different,
containing 1.22 (0.612.49), 1.14 (0.761.93) and 1.82
(1.362.62)% mast cells, respectively. Similarly, lesional,
2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 339 346

nonlesional and normal dog total histamine content in

5 105 skin cells from digests were not statistically
different, containing median values of 46.0 (38.894.7),
59.6 (44.685.2) and 61.7 (28.585.0) n histamine,
respectively.
As reported by others for humans,9 skin mast cells
assayed immediately following isolation show limited
responses to stimulation by anti-IgE, SCF or ConA.
Even though these responses result in the release of a
modest amount of histamine compared with cells subjected to prolonged post isolation incubation, we considered long-term in vitro maintenance to be potentially
confounding to detection of differences in mast cell
responses relatable to skin site and atopy. In addition,
because the process of cell preparation removes
innately bound IgE, the loss of bound IgE in this study
was either corrected by pre-incubation with canine monoclonal IgE for one set of cells (Fig. 4) or not (Fig. 5) to
reveal the effect of saturating the high-affinity receptors for IgE on responses to IgE-independent stimuli.
As expected, histamine release response immediately
following skin cell isolation was low, but ConA-stimulated cells from lesional and nonlesional sites on atopic
dogs and from normal dogs released significantly more
histamine than nonstimulated cells from the same sites
(Fig. 4). In contrast, SCF stimulation did not show
increased histamine release by cells from either atopic
or normal dogs pre-incubated with IgE (Fig. 4). Surprisingly, anti-IgE failed to stimulate histamine release
from IgE-pre-incubated cells from lesional sites even
though nonlesional site cells from atopic dogs responded
as expected (Fig. 4).
There were clear differences in histamine release
responses between cells stimulated in the presence

VDE_273.fm Page 343 Thursday, November 15, 2001 7:20 PM

Stem cell factor and histamine release

343

Figure 4. Percentage of histamine release by skin cells from atopic dogs at lesional (LA) and nonlesional (NLA) sites and from normal dogs
(NC) at sites comparable with nonlesional sites on atopic dogs. Cells were pre-incubated with immunoglobulin (Ig)E at 2 g mL1 for 30 min
prior to stimulation. Histamine released into cell culture supernatants 30 min after concanavalin A (ConA), stem cell factor (SCF) and rabbit
anti-IgE stimulation was calculated as a percentage of total histamine in supernatants and cell pellets. Horizontal lines mark median values.

Figure 5. Percentage of histamine release by skin cells from atopic dogs at lesional (LA) and nonlesional (NLA) sites and from normal dogs
(NC) sites comparable with nonlesional sites on atopic dogs. Cells were not pre-incubated with immunoglobulin (Ig)E prior to stimulation.
Histamine released into cell culture supernatants 30 min after concanavalin A (ConA), stem cell factor (SCF) and rabbit anti-IgE stimulation
was calculated as a percentage of total histamine in supernatants and cell pellets. Horizontal lines mark median values.

(Fig. 4) and absence (Fig. 5) of IgE. In the absence of

IgE the responses by all cell types to SCF tended to be
higher and more broadly distributed than in the presence of IgE.
Cells from normal dogs had higher spontaneous
histamine release in the absence of added IgE than in
its presence, P = 0.05 (Fig. 6). Withholding IgE supplementation from cells from lesional and nonlesional
sites on atopic dogs did not show a significant effect on
spontaneous release (Fig. 6).
Table 1 shows the statistical analyses of results represented graphically in Figs 4 and 5. Possibly because
of the tendency for raised spontaneous release values
in the absence of IgE (Fig. 6), no cells, regardless of the
type of stimulus, showed histamine release values
significantly higher than their nonstimulated controls
(Table 1 and Fig. 5). Spontaneous histamine release by
cells from atopic dogs tended to be higher than normal
dogs and this was statistically significant when comparing lesional (A) site cells with normal (C) skin cells
in the presence of IgE (Table 1 and Fig. 4). In the
absence of added IgE, lesional site cells stimulated
with ConA (D) or SCF (G) released significantly more

Figure 6. The effect of pre-incubating skin cells with immunoglobulin

(Ig)E at 2 g mL1 on spontaneous release of histamine. The same values
for spontaneous histamine release with or without pre-incubation with
IgE shown in Figs 4 and 5 are compared here. Horizontal lines mark
median values. P-Values were calculated by two-tailed Wilcoxon test.

histamine than ConA (F) or SCF (I) stimulated cells

from normal dogs, respectively (Table 1 and Fig. 5). The
lesional site cells stimulated in the absence of IgE (D,
G) likewise released significantly more histamine than
nonstimulated nonlesional site cells from atopic dogs
(B) (Table 1 and Fig. 5).
2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 339346

VDE_273.fm Page 344 Thursday, November 15, 2001 7:20 PM

344

B. Hammerberg et al.

Table 1. Comparison of median per cent histamine release values within treatment groups as shown in Figs 4, 5

A Lesional spont.
B Nonlesional spont.
C Normal dog spont.
D Lesional ConA
E Nonlesional ConA
F Normal dog ConA
G Lesional SCF
H Nonlesional SCF
I Normal dog SCF
J Lesional anti-IgE
K Nonlesional anti-IgE
L Normal dog anti-IgE

IgE added

Significant* differences

No IgE added

10.4%
6.4
3.8
14.1
11.0
7.5
6.6
4.9
5.4
7.8
10.8
5.6

C P = 0.04

10.2%
9.6
7.4
18.6
16.0
9.8
15.8
8.4
11.0
14.8
11.0
12.1

A P = 0.04
A P = 0.04
B P = 0.01; F P = 0.03
C P = 0.006; E P = 0.03

K P = 0.01
B P = 0.03; J P = 0.01
C P = 0.04

Significant differences

B P = 0.04; F P = 0.04
D P = 0.04
B P = 0.04; I P = 0.04
G P = 0.04

*Wilcoxon test of P-values of like treatments and untreated cells pre-incubated with IgE.
Wilcoxon test of P-values of like treatments and untreated cells not pre-incubated with IgE.

DISCUSSION
This is the first report of comparisons of in vivo and in
vitro responses to SCF by groups of individual animals
that are manifesting lesions of naturally occurring
allergic disease or not. As in humans, it is apparent that
there is an inherited basis for allergic diseases in dogs
but the mechanisms remain unknown.4 The observation that peripheral blood leukocytes from atopic dogs
release a higher percentage of histamine than normal
dogs when stimulated with antibodies against IgE,
even though there was no difference between these
groups in total serum IgE, suggests that basophilic leukocytes from atopic dogs are hyperreactive.11 ConA
stimulation of skin mast cells in vitro showed a greater
percentage release of histamine by atopic skin mast
cells than normal skin cells.12 A similar hyperreactive
state is indicated by the response of atopic dogs in this
study to an intradermal injection of rabbit antibodies
against IgE.
Toward elucidating the mechanism for occurrence
of this proposed hyperexcitable mast cell/basophil
population we considered one of the cytokines most
often cited for its action on mast cells. The recently
documented profound affects of SCF, as soluble and
membrane-bound forms, on mast cell differentiation
and function16,17 suggest the importance of knowing
whether there are differences between nonlesional skin
sites on atopic dogs compared with the same sites on
normal dogs regarding the presence of SCF or histamine release response to SCF.
Immunohistochemistry revealed that there was an
obvious difference in the detectable presence of SCF
between areas of skin sites from atopic dogs not showing gross signs of inflammation and same-site skin
from normal dogs. The presence of similar sites in
dogs genetically predisposed to atopic dermatitis
but not yet manifesting disease will be investigated
in future studies. As expected, lesional sites showed
strong staining for intercellular SCF as reported previously for urticaria pigmentosa in humans18 and
nasal biopsies from humans presenting with allergic
rhinitis.16
2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 339 346

These nonlesional sites formed much larger wheals

in response to Fc RI cross-linking by rabbit anti-IgE
IgG compared with the same sites on normal dogs. The
heightened response may well be due to increased local
SCF16, but the possibility that mast cells from atopic
skin could be more responsive to either rabbit antiIgE or SCF alone needed to be tested in vitro. This
led to the discovery that skin cell preparations from
normal dogs had a statistically higher spontaneous
release of histamine in the absence of IgE replenishment following isolation by enzymatic digestion that
was not present in comparisons of spontaneous
release by lesional sites or by nonlesional sites from
atopic dogs. It was surprising to find that the absence
of IgE permitted elevated histamine release responses
by lesional skin cells to SCF and ConA that were not
apparent in the presence of IgE, suggesting that IgE
occupation of FcRI lowers spontaneous release.19
The higher response by lesional skin cells to ConA
compared with normal dog skin cells in the absence of
IgE is in agreement with a previous report by DeMora
et al.12 However, because of the small population
sizes and wide ranges, it is possible that this statistical
difference observed for SCF was due to a type 1 error.
Confirmatory studies are being planned using genetically high-risk dogs sampled during a preclinical
state.
Also in agreement with this previous report,12 we did
not find statistical differences between any of the skin
cell preparations with regard to the numbers of mast
cells. However, we did not confirm their observation
that mast cells from lesional sites contained more histamine. The reason for this discrepancy is not clear because
we did recover more histamine per mast cell and more
mast cells from skin tissues than reported by these authors,
thus reducing the possibility that our technique led to
a selective loss of histamine by lesional skin mast cells
during tissue digestion and cell preparation.
The results of this study suggest that higher levels
of SCF production in the skin of atopic dogs could
be a cause of hyperexcitability in skin mast cells. If
skin mast cells of atopic dogs are hyperexcitable
then it is reasonable to suspect that the consequent

VDE_273.fm Page 345 Thursday, November 15, 2001 7:20 PM

Stem cell factor and histamine release

release of histamine, tumour necrosis factor (TNF) and several other pro-inflammatory mediators could
initiate and help sustain the lesions associated with
atopic dermatitis. The question remains as to whether
there is an inherited increased sensitivity to SCF by
mast cells from atopic dogs and/or increased production
of SCF. This question can be answered in preclinical
dogs with a high risk of developing atopic dermatitis
by repeating this study at a time prior to the development of macroscopic inflammatory lesions.

8.

9.

10.

ACKNOWLEDGEMENTS
11.

The authors thank Dr Shelly Vaden for allowing

biopsies from colony dogs. The authors thank Stanley
Dunston for performing immunohistochemical staining
and Christopher A. Manuel for all other technical
analyses. This study was funded by a grant from the
Canine Health Foundation of the American Kennel
Club.
REFERENCES
1. Hogan, M.B., Harris, K.E., Patterson, R. A naturally
occurring model of immunoglobulinE antibodymediated hypersensitivity in laboratory animals. Journal
of Laboratory and Clinical Medicine 1994; 123: 899905.
2. Ermel, R.W., Kock, M., Griffey, S.M. et al. The atopic
dog: a model for food allergy. Laboratory Animal Science
1997; 47: 40 9.
3. Frick, O.L. Food allergy in atopic dogs. Advances in
Experimental Medicine and Biology 1996; 409: 17.
4. de Weck, A.L., Mayer, P., Stumper, B. et al. Dog allergy,
a model for allergy genetics. International Archives of
Allergy and Immunology 1997; 113: 557.
5. Hill, P.B., Moriello, K.A., DeBoer, D.J. Concentrations
of total serum IgE, IgA, and IgG in atopic and parasitized dogs. Veterinary Immunology and Immunopathology
1995; 44: 105 13.
6. Hammerberg, B., Bevier, D., DeBoer, D.J. et al. Auto
IgG anti-IgE and IgG IgE immune complex presence
and effects on ELISA-based quantitation of IgE in
canine atopic dermatitis, demodectic acariasis and helminthiasis. Veterinary Immunology an Immunopathology
1997; 60: 33 46.
7. Olivry, T., Moore, P.F., Affolter, V.K. et al. Langerhans
cell hyperplasia and IgE expression in canine atopic

12.

13.

14.

15.

16.

17.

18.

19.

345

dermatitis. Archives of Dermatological Research 1996;

288: 57985.
Olivry, T., Naydan, D.K., Moore, P.F. Characterization
of the cutaneous inflammatory infiltrate in canine atopic
dermatitis. American Journal of Dermatopathology 1997;
19: 47786.
Frenz, A.M., Gibbs, B.F., Pearce, F.L. The effect of
recombinant stem cell factor on human skin and lung
mast cells and basophil leukocytes. Inflammation
Research 1997; 46: 359.
Brazis, P., Queralt, M., de Mora, F. et al. Stem cell factor
enhances IgE-mediated histamine and TNF- release
from dispersed canine cutaneous mast cells. Veterinary
Immunology and Immunopathology 2000; 75: 97108.
Jackson, H.A., Miller, H.A., Halliwell, R.E. Canine
leukocyte histamine release: response to antigen and
anti-IgE. Veterinary Immunology and Immunopathology
1996; 53: 195206.
DeMora, F., Garcia, G., Puigdemont, A. et al. Skin mast
cell releasability in dogs with atopic dermatitis. Inflammation Research 1996; 45: 4247.
Lie, W.J., Knol, E.F., Mul, F.P.J. et al. Basophils from
patients with allergic asthma show a primed phenotype.
Journal of Allergy and Clinical Immunology 1999; 104:
10007.
Willemse, T.A. Atopic dermatitis. In: Nesbitt, G.H., ed.
Dematology. New York: Churchill Livingstone, 1987:
pp. 5777.
Gebhard, D., Orton, S.M., Edmiston, D. et al. Canine
IgE monoclonal antibody specific for a filarial antigen:
production by a canine mouse heterohybridoma using
Bcells from a clinically-affected lymph node. Immunology 1995; 85: 42934.
Otsuka, H., Kusumi, T., Kanai, S. et al. Stem cell factor
mRNA expression and production in human nasal epithelial cells: contribution to the accumulation of mast
cells in the nasal epithelium of allergy. Journal of Allergy
and Clinical Immunology 1998; 102: 75764.
Schmidt-Choudhury, A., Meiner, J., Seebeck, J. et al.
Stem cell factor influences neuro-immune interactions:
the response of mast cells to pituitary adenylate cyclase
activating polypeptide is altered by stem cell factor. Regulatory Peptides 1999; 83: 7380.
Onuma, H., Matsui, C., Morohashi, M. Enhanced
expression of SCF in the dermis is a prognostic factor for
the regression of urticaria pigmentosa. European Journal
of Dermatology 1999; 9: 62932.
Bischoff, S.C., Zwahlen, R., Stuki, M. et al. Basophil
histamine release and leukotriene production in response
to anti-IgE and anti-IgE receptor antibodies. International
Archives of Allergy and Immunology 1996; 110: 26171.

Rsum Le Stem cell factor (SCF) influence lactivation des mastocytes et la libration des mdiateurs de
linflammation. Il est lev dans les tissus prsentant une inflammation dorigine allergique. Cette tude a compar la formation de plaques orties la suite dune injection de SCF ou dun anticorps anti-immunoglobuline
(Ig)E, dans la peau de chien sains (n = 10), ou en zone non lsionnelle chez des chiens atopiques (n = 10). La
scrtion de SCF au niveau de zones saines ou de zones lses a galement t compare par immunohistochimie.
La libration dhistamine par des suspensions de cellules cutanes stimules par le SCF, la concanavaline A
(ConA) ou par des anticorps de lapin anti-IgE de chien a t compare pour les chiens sains et les chiens atopiques.
Tous les chiens ont prsent de fortes rponses linjection intradermique de SCF la dose de 10 et de 50 ng mL1.
Les chiens atopiques ont prsent des plaques significativement (P = 0.002) plus importantes que les chiens sains;
en revanche, limmunohistochimie na pas permis de montrer de diffrence quant au nombre de mastocytes
cutans portant des IgE. Seuls les chiens atopiques prsentaient un dpt interstitiel de SCF au niveau des prlvements en peau lse et en peau saine. La libration moyenne dhistamine aprs stimulation par le SCF en labsence
2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 339346

VDE_273.fm Page 346 Thursday, November 15, 2001 7:20 PM

346

B. Hammerberg et al.
dIgE par les cellules cutanes de peau lse tait plus importante chez les chiens atopiques que chez les chiens
sains (P = 0.04). Ces expriences suggrent que la scrtion dermique de SCF pourrait potentialiser la libration
dhistamine aprs pontage des rcepteurs IgE, et tre une des causes dhyperexcitabilit mastocytaire observe
dans la dermatite atopique du chien.
Resumen El factor de clulas madre (SCF) influye la activacin de mastocitos y la liberacin de mediador de
la inflamacin, y se encuentra elevado en tejidos con una inflamacin alrgica. La formacin de habones como
respuesta a la inyeccin de SCF o anticuerpo anti-inmunoglobulina (Ig)E fue comparada entre piel canina normal
(n = 10) y atpica sin lesiones (n = 10). La secrecin de SCF in situ fue comparada entre piel lesionada y nolesionada utilizando inmunohistoqumica. La liberacin de histamina en suspensiones de clulas cutneas
despus de la estimulacin con SCF, concanavalina A (ConA) o anticuerpos de conejo anti- IgE fue comparada
entre perros normales y atpicos. Todos los perros mostraron fuertes respuestas a la inoculacin intradrmica
de SCF a 10 y 50 ng mL1. Los perros atpicos mostraron habones significativamente mayores (P = 0.002) a los
anti-IgE que los perros normales; pero no hubo diferencia en el nmero de mastocitos cutneos portadores de
IgE, detectados mediante inmunohistoqumica. Slo los perros atpicos mostraron una deposicin intersticial
de SCF tanto en muestras cutneas lesionadas como en no lesionadas. La liberacin media de histamina estimulada con SCF en ausencia de IgE, de clulas de piel lesionada fue superior en perros atpicos que en perros normales (P = 0.04). Estos experimentos sugieren que la secrecin de SCF drmica podra potenciar la liberacin
de histamina despus de un enlace cruzado con el receptor de IgE y, as, podra ser una de las explicaciones de
la hiperexcitabilidad inherente de los mastocitos observada en la dermatitis atpica canina.
Zusammenfassung Stammzellenfaktor (SZF) beeinflusst Mastzellenaktivierung und Freisetzung von Entzndungsmediatoren, die Stammzellenfaktorkonzentration in Geweben mit allergischer Entzndung ist erhht.
Quaddelbildung nach Injektion von SZF oder Anti-Immunglobulin E-Antikrpern in der Haut von normalen
(n = 10) und in der lsionsfreien Haut atopischer Hunde (n = 10) wurde verglichen. In situ SZF Sekretion von
lsionsfreier Haut und Haut mit Lsionen wurde mittels Immunhistochemie verglichen. Histaminfreisetzung von
Hautzellsuspensionen nach Stimulation mit SCF, Concanavalin A (ConA) oder Kaninchenantikrper gegen
Hunde-IgE-Antikrper von normalen und atopischen Hunden wurde verglichen. Alle Hunde zeigten starke
Reaktionen nach intradermaler SZF Injektion in einer Dosierung von 10 und 50 ng mL1. Atopische Hunde
hatten signifikant grssere Quaddelbildung gegen IgE als normale Hunde (P = 0.002), zwischen der Anzahl der
durch Immunhistochemie entdeckten Hautmastzellen mit gebundenem IgE war kein Unterschied festzustellen.
Nur atopische Hunde zeigten interstitielle Ablagerung von SZF in lsionsfreier Haut und Haut mit Lsionen.
Durch SZF stimulierte durchschnittliche Histaminfreisetzung in Abwesenheit von IgE von Zellen der mit
Lsionen befallenen Haut war hher bei atopischen als bei normalen Hunden (P = 0.04). Diese Experimente
deuten daraufhin, dass dermale SZF-Sekretion mglicherweise Histaminfreisetzung nach Querverbindung von
IgE-Rezeptoren erhht und knnte eine der Grnde fr die bei atopischer Dermatitis des Hundes gesehene den
Mastzellen eigene erhhte Erregbarkeit sein.

2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 339 346