230 DISC

Veterinary Dermatology 2001, 12, 4958

Characterization of the inammatory inltrate during

IgE-mediated late phase reactions in the skin of normal
and atopic dogs
THIERRY OLIVRY,* STANLEY M. DUNSTON,* K. MARCY MURPHY* and PETER F. MOORE{
*Department of Clinical Sciences, North Carolina State University College of Veterinary Medicine, Raleigh,
NC 27606, USA and {Department of Pathology, Microbiology and Immunology, University of California
School of Veterinary Medicine, Davis, CA 95616, USA
(Received 30 December 1999; accepted 4 May 2000)

Abstract In canine and human atopic patients, the intracutaneous injection of oending allergens is followed
by the development of both immediate and late-phase reactions. The present study was performed to expand
on the characterization and dynamics of inammatory cell subsets during IgE-mediated late-phase reactions
in canine skin. Three normal dogs and three Dermatophagoides farinae-allergic dogs were selected for this
experiment. All dogs were challenged intradermally with mite allergen, puried anticanine IgE antibodies
(positive control) or phosphate-buered saline (negative control). Skin biopsies were obtained before and 6, 12
and 24 h post-injection. Sections were stained with metachromatic and eosinophil-specic histological stains.
Additionally, we used an immunohistochemical method with antibodies specic for canine leukocyte antigens.
This study conrmed the occurrence of a late-phase reaction in atopic skin following allergen challenge, and in
normal and atopic canine skin after intradermal injection of IgE-specic antibodies. Whereas early emigrating
dermal cells were composed chiey of neutrophil and activated eosinophil granulocytes, there was an inux of
ab T-lymphocytes and dermal dendritic cells in later stages of the late-phase reactions. Because IgE-mediated
late-phase reactions resemble spontaneous atopic canine skin lesions, both at macroscopic and microscopic
levels, we propose the use of similar challenges to study the anti-inammatory eects of anti-allergic drugs in a
pre-clinical setting.
Keywords: allergy, atopic dermatitis, dog, late-phase reaction, skin.

INTRODUCTION
Intradermal injection of allergens in the skin of
humans with atopic dermatitis (AD) results in a
biphasic response. Cutaneous wheal-and-are follows
the immediate degranulation of dermal mast cells, as
seen in urticarial reactions characteristic of type-I
hypersensitivity. IgE-mediated mast cell activation 1,2
also triggers the sequential chemoattraction of
inammatory cells, a phenomenon known as the
late-phase reaction (LPR).3 During allergen-induced
LPR, there is an inux of neutrophil and eosinophil
granulocytes that occurs in less than 6 h.47 Basophils
also can be seen in biopsies of early stages of LPR.8,9
Polymorphonuclear inltration is followed by the
emigration of mononuclear cells including memory
T-lymphocytes.57,1013 Since the cutaneous inltrate
of allergen-induced LPR resembles that present in
mucosae or skin of humans with atopic rhinitis,
asthma or dermatitis,14 the concept that LPR are
relevant to the pathogenesis of IgE-mediated allergic
Correspondence and requests for reprints: Dr Thierry Olivry,
Department of Clinical Sciences, North Carolina State University,
College of Veterinary Medicine, 4700 Hillsborough Street, Raleigh,
NC 27606, USA, Tel.: (919) 5136276, Fax: (919) 513 6336
# 2001 Blackwell Science Ltd

reactions has gained momentum.14,15 This hypothesis

is supported by the demonstration of attenuated
bronchial LPR following allergen-specic immunotherapy in mite sensitized asthmatic children.16
In dogs sensitized experimentally to ragweed,
intradermal injection of allergen results in macroand microscopic LPR.18 These cutaneous LPR are
similar to those observed in atopic humans, i.e.
characterized by sequential granulocytic and mononuclear dermal inltration.18 Similar LPR have been
observed in the skin of dogs with spontaneously
arising AD after intradermal injection of house dust
mite extracts.19 The relevance of LPR to the
pathogenesis of canine AD is supported indirectly
by the observation of a reduction of allergic pruritus
and skin lesions in an open clinical trial with
misoprostol.20 Indeed, misoprostol is a prostaglandin
E1 analogue which selectively inhibits late-phase but
not immediate cutaneous allergic reactions.21
The present study was undertaken to expand the
previous knowledge on the cellular nature and
dynamics that occur during IgE-mediated LPR in
the skin of dogs with AD. We will conrm that LPR
develop, in mite-sensitized atopic dogs, following
intradermal injection of puried Dermatophagoides
farinae allergens. Similar inammatory responses also
49

230 DISC
50

T. Olivry et al.

will be provoked by intradermal challenge with antiIgE antibodies in normal or atopic canine skin.
Because the dermal inltrate of full-blown LPR
resembles that which is seen in the dermis of dogs
with AD,22,23 we propose that such reactions could be
used as a model to study IgE-mediated allergic
inammation in canine skin.
METHODS
Selection of canine subjects
Three adult dogs, diagnosed with AD, were entered
in the study after an informed consent was obtained
from their owner. In these dogs the diagnosis of AD
was carried out by fullment of the clinical criteria
proposed by Willemse.24 Resembling pruritic diseases
(e.g. adverse food reactions, scabies and insect bite
hypersensitivities) were ruled-out by standard diagnostic or therapeutic methods (elimination diet,
acaricidal therapy and ea control, respectively). All
three dogs exhibited positive immediate reactions
after intradermal injection of a commercially available Dermatophagoides farinae (Df) solution (1: 1000
1: 50000 w/v; Greer laboratories, Lenoir, NC, USA).
Anti-inammatory drugs had been discontinued for
variable lengths of time, depending upon the drug
category, so that they would not interfere with
immediate reaction generation (e.g. 1 week for
antihistamines, 3 weeks for oral corticosteroids and
6 weeks for repository glucocorticoid formulations).
It was hoped that these withdrawal times would be
sucient to not interfere with LPR generation, as
there are no published withdrawal guidelines established for the interference of these drugs on cutaneous
LPR in the canine species.
Three adult research beagle dogs were chosen, as
normal controls, because of lack of breed and family
history of pruritus, skin lesions and cutaneous bacterial
infections. None of these dogs exhibited any immediate
reactions after intradermal challenge with the Df
commercial extract at 1: 10001: 50000 dilutions.
Generation of IgE-mediated LPR
Each of the three atopic and three normal dogs were
challenged, on the lateral thorax, with triplicate
intradermal injections of Df house dust mite allergens
and suitable controls.
Allergen challenge was performed with a puried
fraction of Df house dust mite extract known to
contain the high molecular weight major allergens for
dogs (Df-AFT) (Esch, R. E., Plaster, J., Girone, C.,
Grier, T. J., Olivry, T. Isolation and characterization
of a major dust mite (Dermatophagoides farinae)
allergenic fraction in dogs. Proceedings of the Annual
Meeting of the AAVD/ACVD. Nashville, TN, 1997:
878.). To avoid individual variability in the antigen
provocation, the three atopic dogs were tested
initially with varying dilutions of the Df-AFT mite
fraction. The minimum challenge dose (MCD) of Df# 2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 4958

AFT, found to elicit a 10-mm wheal in each dog after

15 min, was determined to be 0.05 mL of a 3-mg
mL71 dilution (i.e. an approximate dose of 150 ng of
Df-AFT per injection). Since the generation of LPR
in human allergic individuals is amplied with
increased concentrations of allergens,10 atopic and
normal dogs were injected intradermally with 10
times the MCD, i.e. *1.5 mg of Df-AFT in a nal
volume of 0.05 mL.
To serve as positive control, we used protein G
anity-puried rabbit polyclonal IgG specic for
canine IgE (courtesy of Dr Bruce Hammerberg, NC
State University, Raleigh, NC, USA).25
Anticanine IgE antibodies induce degranulation of
canine mast cells in vitro, at least in part via an IgEdependent mechanism26 and such antibodies have been
used to induce immediate-type hypersensitivity reactions in vivo in canine skin.27 Preliminary experiments
for the present study demonstrated that intradermal
injection of pre-immune rabbit IgG in normal canine
skin neither resulted in immediate nor late-phase cutaneous reactions (B. Hammerberg, NC State University,
Raleigh, NC, USA, 1997, unpublished data).
The MCD necessary to produce a 10-mm wheal in
each atopic dog was estimated to be *700 ng of
anticanine IgE antibodies, as determined by previous
intradermal challenges using serial dilutions of this
reagent. To ensure optimal generation of cutaneous
LPR, all dogs were injected intradermally with 10
times the MCD (approximately 7 mg of anti-IgE
antibodies in a volume of 0.05 mL). As the negative
control, 0.05 mL of phosphate-buered saline (PBS)
was injected in the dermis of normal and atopic dogs.
Specimen collection and processing
One punch biopsy specimen (8 mm) was collected
from normal appearing skin on the lateral thorax
before injection. Similar samples were collected 6, 12
and 24 h post-injection from skin sites challenged
with Df, anti-IgE and PBS. Immediately after
sampling, skin samples were bisected. One half of
each specimen was placed in neutral buered
formalin for routine embedding in paran. The
second half was deposited in Optimal Cutting
Temperature medium (OCT Tissue Tek, Baxter
Diagnostics Inc., McGaw Park, IL, USA), immersed
in isopentane cooled to its freezing point in liquid
nitrogen, and then stored at 7708C until processed
for immunohistochemical staining.
Histological staining of granulocytes and mast cells
Five micrometer sections were obtained from paran
blocks and stained with routine haematoxylin-eosin
for visualization of the inammatory reaction pattern.
For determination of dermal neutrophilia and eosinophilia, paran sections were coloured with Luna's
stain for eosinophils.28 Preliminary studies were
performed to select a technique that permitted the
easy dierentiation of basophils from mast cells in the
dermis. Among the various methods tried (Giemsa

230 DISC
Canine cutaneous late-phase reactions
and toluidine blue dyes, esterase histochemical stain
or tryptase/chymase immunohistochemistry), we selected a histological staining with alcian blue and
nuclear fast red counterstain. This method permitted
dierentiation of round from segmented nuclei
unmasked from the abundant cytoplasmic granules.
Immunophenotyping of cutaneous mononuclear cells
To determine the phenotype and numbers of dermal
mononuclear cells, we used a three-step labelled
streptavidin method23 and the reagents listed in Table
1 (all diluted at 1 : 10). All primary monoclonal
antibodies were specic for canine leukocyte antigens
and have been generated at the University of
California Davis by one of the authors (PFM).2931
Grading of macroscopic reactions
Injected skin sites were examined 20 min, and 6, 12
and 24 h after intradermal challenge for any
macroscopic signs of inammation. The diameter of
visible erythematous patches was measured in two
perpendicular directions and averaged. Similarly, at
all time points and sites, skin thickness was measured
with callipers. Measurements were averaged among
the three atopic or the three normal dogs. Statistical
analyses were not performed because of the low
number of dogs in each group.
Enumeration of dermal cells
Granulocytes, mast cells and mononuclear cells were
counted in the dermis of each biopsy using a
microscope at 100 6 total magnication and a
rectangular eld covering 0.06 mm2. Sixteen consecutive elds, i.e. a total surface of 1 mm2, were
examined in the supercial dermis and the various
cell types were enumerated. Dermal cell counts were
averaged between biopsies from the three dogs in
each group. Tallies were expressed as number of cells
per mm2 of dermis.
RESULTS
LPR can be provoked in normal and atopic canine skin
In normal dogs, urticarial reactions, characteristic of
type I hypersensitivity, were seen 20 min following
the injection of anti-IgE polyclonal antibodies.
Erythematous patches, averaging 11 mm in diameter
and 1 mm of increased skin thickness, were disTable 1. Panel of antibodies specic for canine leukocyte antigens
Cell type

CD

Clone

Dendritic cells

CD1c

CA13.9H11

CD3
CD4
CD8a
TCR(C)ab
TCR(C)gd
TCR(C)gd

CA17.2A12
CA13.1E4
CA9.JD3
CA15.8G7
CA20.6A3
CA20.8H1

CD21

CA2.1D6

T-lymphocytes

b-lymphocytes

51

covered 6 h after the injection of this positive control

(Fig. 1a,b). Such LPR decreased progressively in
diameters but persisted for the entire duration of the
study (i.e. 24 h). Immediate or late-phase reactions
were not observed following PBS or Df-AFT
intradermal injection in normal canine skin.
In atopic dogs, wheals developed within 20 min of
intradermal challenge with Df-AFT fraction and antiIgE antibodies. Erythematous patches, indicative of
LPR, arose at both allergen and positive control sites.
The acme of macroscopic inammation was observed
between 6 and 12 h post-challenge (Figs 1a,b and 2).
There were no immediate or late-phase reactions
following intradermal injections of PBS.
These results conrmed that LPR develop following Df allergen intradermal challenge in mite-allergic
dogs. Similar macroscopic reactions could be provoked, in normal or atopic dogs, after intracutaneous
injection of anti-IgE antibodies. Even though the
intensity of these reactions was lower in normal than
atopic dogs, this study establishes clearly that IgEmediated LPR can be induced in normal canine skin.
Mast cells are activated during cutaneous LPR in
canine skin
Alcian blue staining with nuclear fast red counterstain permitted the visualization, in a perivascular
location, of large cells with round nuclei and a low
nuclear: cytoplasmic ratio. These cells were identied
as mast cells. Basophil granulocytes, with a smaller
cell diameter and a segmented nucleus, were not
recognized in any of the specimens examined.
In normal canine biopsies, the number of dermal
mast cells varied, over time, from an average of 12
15, 1216 and 1210 cells/mm2 after challenge with
PBS, Df-AFT and anti-IgE, respectively.
Similarly, in specimens collected from atopic dogs,
dermal mast cell counts changed from 19 to 25, 1922
and 1910 cells/mm2 following intradermal injection
with PBS, Df-AFT and anti-IgE, respectively. In
atopic skin challenged with Df-AFT, or in normal
and atopic specimens injected with anti-IgE antibodies, dermal mast cells often exhibited a fainter
cytoplasmic colour than in samples previously
challenged with PBS. This nding was interpreted
as mast cell degranulation following activation by
allergens or anti-IgE antibodies.
Neutrophil and eosinophil granulocytes emigrate early
during LPR in canine skin
In normal dogs, intradermal injection of anti-IgE
antibodies was followed by the early emigration of
neutrophil and eosinophil granulocytes in the supercial dermis, as visualized with haematoxylin-eosin or
Luna's stain (Fig. 3a,b). The pattern of inammation
was supercial perivascular to interstitial. All biopsy
specimens, challenged with the positive control,
contained activated eosinophils characterized by
aggregated intra-or paracellular eosinophilic granules. Dermal eosinophil counts were usually twice as
# 2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 4958

230 DISC
52

T. Olivry et al.

Figure 1. Intradermal injection of allergen or anti-IgE antibodies is followed by the development of a late-phase reaction characterized by
erythematous patches (1a) and increased skin thickness (1b).

(Fig. 4a-d). The peak of neutrophil inux occurred

between 6 and 12 h, whereas that of eosinophils arose
at later times (Fig. 3a,b). In all biopsies, cytological
markers of activation, evidenced by granule conuence, were seen in supercial dermal eosinophils
(Fig. 5). The magnitude of dermal eosinophilia in
atopic canine skin following anti-IgE antibody injection was almost twice that occurring in normal skin
specimens similarly challenged (Fig. 3b). In atopic
biopsies, the number of eosinophils was two to 10-fold
that of neutrophils in the same specimens (Fig. 3a,b).

Figure 2. An erythematous patch occurred 6 h after intradermal

injection of Dermatophagoides farinae antigen in a mite-allergic dog.

high as neutrophil tallies. Eosinophils were not seen

at sites of previous PBS or Df injections.
Identical granulocytic inux was seen in atopic
specimens following intracutaneous challenges with DfAFT house dust mite fraction and anti-IgE antibodies
(Fig. 3a,b). The inammatory pattern again was
characterized as supercial perivascular to interstitial
# 2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 4958

Late-emigrating lymphocytes consist mostly of alphabeta T-lymphocytes that express either CD4 or CD8
In normal and atopic skin biopsies injected with antiIgE antibodies, T-lymphocytes, expressing CD3 and
the ab T-cell receptor, began emigrating into the
dermis 6 h post-challenge and continued to rise in
numbers during the course of the study (Fig. 3c).
Similar T-cell inux, albeit present at a higher degree,
occurred in atopic skin following Df-AFT allergen
provocation (Figs 3c and 6). Remarkably, a doubling
of ab T-cells (maximum of 50 cells/mm2) was seen also
in normal canine skin challenged with mite allergens.
T-cells expressing the gd T-cell receptor were not
discovered in the dermis of any normal canine skin
samples. In atopic specimens, they were uncovered in
very low numbers (i.e. less than 10 cells/mm2) in 9/30
biopsies. B-lymphocytes, expressing CD21, were not
seen in any of the stimulated skin samples.
In allergen and anti-IgE stimulated samples, Tlymphocytes expressed either CD4 or CD8 (data not

230 DISC
Canine cutaneous late-phase reactions

53

Figure 3. Early stages of IgE-mediate late phase reactions are characterized by the emigration of neutrophil (a) and eosinophil (b)
granulocytes. Later stages are associated with an inux of CD3+T-lymphocytes (c) and dermal dendritic cells expressing CD1c (d).

shown). Precise analysis of CD4:CD8 cellular ratios

proved meaningless as, in most biopsies, the number
of cells expressing CD4 or CD8 was higher than that
of CD3+ lymphocytes. These results suggest that
other cells, such as activated dermal dendritic cells or
neutrophils29 expressed CD4.
Dermal dendritic cells compose the main subset of
dermal mononuclear cells in late canine cutaneous LPR
In atopic skin samples challenged with mite allergen,
and in normal or atopic biopsies provoked with
anti-IgE antibodies, the number of CD1c+ dendritic
cells arose progressively during the course of the
study (Fig. 3d). In the supercial dermis of most
biopsies, dermal dendritic cells were disposed in
dense clusters, a sign of cellular activation (Fig. 7).
These clusters often occurred in the vicinity of
mononuclear cells presumed to be T-lymphocytes.
The number of dermal dendritic cells was, in general,
two times higher than that of CD3+ T-lymphocytes
in each of the specimens and reached 230 (in Dfinjected sites) to 360 cells/mm2 (in anti-IgE-challenged specimens).

DISCUSSION
The present study conrms that intradermal injection
of Df house dust mite allergens in canine atopic
patients triggers not only immediate urtication, but
also long lasting visible and microscopic inammation with phenotypic features of IgE-mediated LPR
in human skin. Remarkably, similar reactions can be
induced, in normal or atopic canine skin, following
intracutaneous challenge with puried anticanine IgE
polyclonal rabbit IgG antibodies. We expand on the
previous knowledge18,19 that, microscopically, canine
cutaneous LPR are associated with the early emigration of neutrophil and eosinophil granulocytes.
Furthermore, we establish that ab T-lymphocytes
and dermal dendritic cells migrate to the dermis in
later stages of inammation.
In atopic dogs challenged with Df or in normal and
atopic dogs provoked with anti-IgE antibodies,
clinical signs of LPR consist of erythema and
increased skin thickening. These LPR lesions, but
not the urticarial wheals seen in immediate allergic
reactions, resemble those seen in dogs with AD. The
# 2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 4958

230 DISC
54

T. Olivry et al.

Figure 4. There is no inammation in normal canine skin (a). Six (b), 12 (c) and 24 (d) hours after Dermatophagoides farinae antigen challenge
in atopic canine skin, dermal inammation occurs in a perivascular to interstitial pattern. Luna's stain for eosinophils, bar = 300 mm.

Figure 6. Twenty-four hours after Dermatophagoides farinae

antigen injection in atopic canine skin, T-lymphocytes cluster in
the supercial dermis. Immunohistochemical method, ab T-cell
receptor-specic monoclonal antibody CA15.8G7, amino-ethylcarbazole chromogen and haematoxylin counterstain, bar = 75 mm
(inset, 40 mm).

Figure 5. Activated eosinophils, characterized by aggregated intraand paracellular eosinophil granules, are seen 12 h after
Dermatophagoides farinae antigen injection in an atopic dog.
Luna's stain for eosinophils, bar = 20 mm.
# 2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 4958

clinical acme of allergen and anti-IgE-induced LPR

occurs around 6 h after challenge, similarly to a
previous observation.18
Because LPR could be relevant to the pathogenesis
of canine AD, we recommend that evaluation of
intradermal allergy tests should encompass not only a
reading of immediate reactions after 1530 min, but
also an assessment of LPR 6 h after challenge.

230 DISC
Canine cutaneous late-phase reactions

Figure 7. Dendritic cells aggregate around mononuclear cells in the

supercial dermis of biopsies collected 24 h after Dermatophagoides
farinae antigen injection in the skin of a mite-allergic dog.
Immunohistochemical
method,
CD1c-specic
monoclonal
antibody CA13.9H11, amino-ethyl-carbazole chromogen and
haematoxylin counterstain, bar = 75 mm (inset, 40 mm).

The alcian blue/nuclear fast red histological stain

selected for metachromatic cell identication permitted the enumeration of dermal mast cells in every
sample. Throughout the course of the study, the
number of mast cells did not vary markedly within
each group of samples. However, there was usually a
decrease in the intensity of the cytoplasmic staining.
This observation was attributed to the release of mast
cell granules, suggesting that these cells had been
activated following IgE receptor cross-linking after
either allergen or anti-IgE antibody challenge.
Remarkably, even though our colouration allowed
easy distinction of nuclear contours, we did not notice
any basophil granulocytes in any specimen. In a
previous study of LPR in dogs, basophils similarly
were not seen.18 These observations are in marked
contrast with reports of LPR induced in human
skin.9,10 In these experiments, basophils were detected
within 68 h following allergen challenge. These
diverging results could be explained by a dierence in
the pathogenesis of LPR between species, or by a lack
of appropriate detection of basophils by our method.
In support of the latter hypothesis, Irani et al. reported
that LPR-recruited basophils were undetectable after
toluidine blue staining, but were visible after immunohistochemical colouring using a specic monoclonal
antibody.9 Unfortunately, antibodies developed
against canine basophils are currently not available.
Our study corroborated that granulocytes, both
neutrophils and eosinophils, were the rst cells
migrating into the sites of IgE-mediated LPR in
canine skin. Indeed, neutrophil inux was seen early
(30 min to 1 h) after antigen challenge in two previous
investigations on canine cutaneous LPR.18,19 In the
present study, the magnitude of neutrophil inux
after allergen challenge was similar to that previously
reported in both canine18,19 and human skin.32,33 In
dogs, neutrophil migration has been explained, in
part, by the preceding release of leukotriene B4 into
sites of provocation.19 In humans, however, the early

55

release of chemokines, e.g. interleukin (IL)-8, is

hypothesized to trigger neutrophil inux.34,35
This study conrmed earlier ndings that eosinophil granulocytes were the cells most commonly
emigrating into the dermis within 12 h post-challenge
with allergen18 or anti-IgE antibodies. Because most
dermal eosinophils exhibited microscopic signs of
degranulation, it is likely that their cytotoxic proteins
could be responsible for some of the tissue damage
and resulting erythema and oedema of early LPR in
canine skin. In humans, the peak of eosinophil inux
also takes place approximately 6 h after allergen LPR
induction.4,5,10,11,33 The number of migrating eosinophils after allergen challenge, however, appears much
lower in humans (75150 cells/mm2 at 6 h
peak)4,5,11,33 than in canine atopic skin (500 cells/
mm2). The reason for this dierence is not known at
present times, but it could be due to the magnitude or
nature of the secretion of chemokines involved in
eosinophil chemoattraction (e.g. eotaxins, RANTES
and MCP-3 and -4).36
During late phases of LPR in canine skin, mononuclear cells have been shown to represent the
predominant cell category.18 Our study conrmed
this observation and expanded the previous knowledge with the demonstration that emigrating mononuclear cells consisted of T-lymphocytes and CD1+
dermal dendritic cells. Lymphocytes most commonly
expressed the ab T-cell receptor. T-cells expressing
the gd antigen receptor were observed only rarely in
atopic specimens after intradermal allergen challenge.
These results are in contrast to those of natural
canine atopic skin lesions in which these cells are seen
frequently in the epidermis and dermis.23 Because gd
T-lymphocytes are involved in epithelial defence,37
we hypothesize that the observed dierence between
lesional atopic skin samples and specimens in this
study is best explained by the route of allergen
challenge. Whereas epicutaneous provocation with
allergens, as suspected for canine AD skin, would
result in the hyperplasia of epitheliotropic gd Tlymphocytes, intradermal antigen challenge would
lead to ab T-cell emigration in the present experiment. Alternatively, gd T-cell hyperplasia could be
the result of prolonged or continuous allergen
challenge (as seen in atopic patients), a provocation
which would be dierent from the single intradermal
injection of this study.
In canine allergen and anti-IgE-induced skin LPR,
CD1c+ dendritic cells emigrated in high numbers to
the sites of intradermal challenge. Clustering of these
dendritic cells at the vicinity of lymphocytes suggested
that active foci of antigen capture or presentation
occurred in the dermis. Conversely, the number of
dermal CD1a+ dendritic cells did not increase over
time in a study of LPR-induction in human atopic
skin.13 In the skin of dogs with spontaneous AD,
however, dendritic cells are hyperplastic.22 Whereas
Langerhans' cell hyperplasia and clustering occur in
the epidermis of lesional atopic canine skin, dermal
# 2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 4958

230 DISC
56

T. Olivry et al.

dendritic cells are in high number in the dermis of

both lesional and nonlesional atopic specimens.22 The
results of the present study therefore are similar to the
observations in spontaneous canine AD skin, albeit
reecting the dierence in the location (intradermal
vs. epidermal) allergen challenges.
In conclusion, our study of LPR generation in both
atopic and normal canine skin highlights the similarity
between the late stages of induced reactions and the
dermal inammation of spontaneous canine atopic
skin. Even though our study was limited to 24 h postchallenge, we presume that later biopsies would have
yielded dermal inammation richer in mononuclear
cells and poorer in granulocytes, hence, even more
similar to that of natural AD in dogs. Whether or not
cutaneous LPR could serve as an experimental model
for AD skin lesions remains the subject of debate in
humans.7 Indeed, early phases of intradermally
provoked LPR are richer in granulocytes than `atopy
patch tests' and atopic skin lesions.7 This dierence in
cellular dynamics, however, could be explained by the
dierence in chemokine release following dermal
instead of epidermal allergen provocation.
Nevertheless, we documented that intradermal
injection of anti-IgE antibodies into the dermis of
normal dogs is followed by an LPR. This nding is not
surprising, in light of the presence of IgE on skin mast
cells of normal dogs.22,38 Cutaneous reactions after
intradermal challenges with IgE-specic antibodies in
normal dogs: (1) most likely result from mast cell
activation following IgE-receptor cross-linking, (2) are
similar to those occurring after allergen provocation in
atopic dogs, and (3) resemble atopic skin lesion both
macro and microscopically. We propose, therefore,
that IgE-mediated LPR in normal canine skin could be
used to study inhibitory eects of `anti-atopic'
pharmacological agents in a pre-clinical setting.
Indeed, such an experimental approach has been
validated recently by dierent investigators.27

2.

3.

4.

5.

6.

7.

8.
9.

10.

11.

ACKNOWLEDGEMENTS
This study was funded by a competitive research
grant from the American Academy of Veterinary
Dermatology. The authors thank Drs Robert Esch
(Greer Laboratories, Lenoir, NC, USA) and Bruce
Hammerberg (NC State University, College of
Veterinary Medicine, NC, USA) for providing the
mite allergenic fraction and anticanine IgE antibodies, respectively. We are grateful to Wendy Savage
(Biomedical Communication, NC State University,
College of Veterinary Medicine, NC, USA) for
helping with this paper's colour gures.
REFERENCES
1. Solley, G.O., Gleich, G.J., Jordon, R.E., Schroeter,
A.L. Late phase of the immediate wheal and are skin
# 2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 4958

12.

13.

14.
15.

reactions: its dependence on IgE antibodies. Journal of

Clinical Investigation 1976; 58: 40820.
Sawada, K., Nagai, H., Basaki, Y. et al. The expression
of murine cutaneous late phase reaction requires both
IgE antibodies and CD4 T cells. Clinical and
Experimental Allergy 1997; 27: 22531.
Charlesworth, E.N. The skin as a model to study the
pathogenesis of IgE-mediated acute and late-phase
responses. Journal of Allergy and Clinical Immunology
1994; 94: 124050.
Frew, A.J., Kay, A.B. The relationship between
inltrating CD4+ lymphocytes, activated eosinophils
and the magnitude of the allergen-induced late-phase
cutaneous reactions in Man. Journal of Immunology
1988; 141: 415864.
Tsicopoulos, A., Hamid, Q., Haczku, A. et al. Kinetics
of cell inltration and cytokine messenger RNA
expression after intradermal challenge with allergen
and tuberculin in the same atopic individuals. Journal
of Allergy and Clinical Immunology 1994; 94: 76472.
Werfel, S., Massey, W., Lichtenstein, L.M., Bochner,
B.S. Preferential recruitment of activated, memory T
lymphocytes into skin chamber uids during human
cutaneous late-phase allergic reactions. Journal of
Allergy and Clinical Immunology 1995; 96: 5765.
Langeveld-Wildschut, E.G., Thepen, T., Bihari, I.C. et
al. Evaluation of the atopy patch test and the
cutaneous late-phase reaction as relevant models for
the study of allergic inammation in patients with
atopic eczema. Journal of Allergy and Clinical
Immunology 1996; 98: 101927.
Charlesworth, E.N. The role of basophils and mast
cells in acute and late reactions in the skin. Allergy
1997; 52: 3143.
Irani, A.M.A., Huang, C., Xia, H.Z. et al.
Immunohistochemical detection of human basophils
in late-phase skin reactions. Journal of Allergy and
Clinical Immunology 1998; 101: 35462.
Charlesworth, E.N., Hood, A.F., Soter, N.A., KageySobotka, A., Norman, P.S., Lichtenstein, L.M.
Cutaneous late-phase response to allergenmediator
release and inammatory cell inltration. Journal of
Clinical Investigation 1989; 83: 151926.
Gaga, M., Frew, A.J., Varney, V.A., Kay, A.B.
Eosinophil activation and T lymphocyte inltration in
allergen-induced late phase skin reactions and classical
delayed-type hypersensitivity. Journal of Immunology
1991; 147: 81622.
Lessin, S.R., Benoit, B.M., Li, G.Q., Moskovitz, A.,
Zweiman, B. Quantitative analysis of T-cell receptor
beta variable-gene usage in cutaneous late-phase
reactions: Implications for T-lymphocyte recruitment
in cutaneous inammation. Clinical and Diagnostic
Laboratory Immunology 1999; 6: 858.
Ying, S., Barata, L.T., Meng, Q. et al. Highanity immunoglobulin E receptor (Fc epsilon RI) bearing eosinophils, mast cells, macrophages and
Langerhans' cells in allergen-induced late-phase
cutaneous reactions in atopic subjects. Immunology
1998; 93: 2818.
Zweiman, B. The late-phase reaction: role of IgE, its
receptor and cytokines. Current Opinion in Immunology
1993; 5: 9505.
Kaliner, M.K. Hypotheses on the contribution of latephase allergic responses to the understanding and

230 DISC
Canine cutaneous late-phase reactions

16.

17.

18.

19.

20.

21.

22.

23.

24.
25.

26.

treatment of allergic diseases [Editorial]. Journal of

Allergy and Clinical Immunology 1984; 73: 311.
Warner, J.O., Soothill, J.F., Price, J.F., Hey, E.N.
Controlled
trial
of
hyposensitization
to
Dermatophagoides pteronyssinus in children with
asthma. Lancet 1978; 2: 9127.
Cockcroft, D.W., Run, R.E., Hargreave, F.E.
Appeareance of allergen-induced increases in airway
responsiveness only after repeated allergen inhalations
in two subjects. Clinical and Experimental Allergy 1989;
19: 2257.
Becker, A.B., Chung, K.F., McDonald, D.M., Frick,
O.L., Gold, W.M. Cutaneous allergic response in
atopic dogs: relationship of cellular and histamine
response. Journal of Allergy and Clinical Immunology
1988; 81: 4418.
Thomsen, M.K., Kristensen, F., Elling, F. Species
specicity in the generation of eicosanoids: emphasis on
leukocyte-activating factors in the skin of allergic dogs
and humans. In: Ihrke, P.J., Mason, I.S., White, S.D.
eds. Advances in Veterinary Dermatology, Vol 2
Oxford: Pergamon Press; 1993: 6378.
Olivry, T., Guaguere, E., Heripret, D. Treatment of
canine atopic dermatitis with the prostaglandin E1
analog misoprostol: an open study. Journal of
Dermatological Treatment 1997; 8: 2437.
Alam, R., DeJarnatt, A., Staord, S., Forsythe, P.A.,
Kumar, D., Grant, J.A. Selective inhibition of the
cutaneous late but not immediate allergic response to
antigens by misoprostol, a PGE analogResults of a
double-blind, placebo-controlled randomized study.
American Review of Respiratory Diseases 1993; 148:
106670.
Olivry, T., Moore, P.F., Aolter, V.K., Naydan, D.K.
Langerhans cell hyperplasia and IgE expression in
canine atopic dermatitis. Archives of Dermatological
Research 1996; 288: 57985.
Olivry,
T.,
Naydan,
D.K.,
Moore,
P.F.
Characterization of the cutaneous inammatory
inltrate in canine atopic dermatitis. American
Journal of Dermatopathology 1997; 19: 47786.
Willemse, T. Atopic skin disease: a review and a
reconsideration of diagnostic criteria. Journal of Small
Animal Practice 1986; 27: 7718.
Gebhard, D., Orton, S., Edmiston, D., Nakagaki, K.,
DeBoer, D., Hammerberg, B. Canine IgE monoclonal
antibody specic for a larial antigen: production by a
canine x murine heterohybridoma using B cells from a
clinically aected lymph node. Immunology 1995; 85:
42934.
Brazis, P., Queralt, M., deMora, F., Ferrer, L.,
Puigdemont, A. Comparative study of histamine
release from skin mast cells dispersed from atopic,

27.

28.
29.

30.

31.

32.
33.

34.

35.

36.
37.

38.

57

ascaris-sensitive and healthy dogs. Veterinary

Immunology and Immunopathology 1998; 66: 4351.
Deboer, D.J., Cooley, A.J. Use of induced cutaneous
immediate-type hypersensitivity reactions to evaluate
anti-inammatory eects of triamcinolone topical
solution in three dogs. Veterinary Dermatology 2000;
11: 2533.
Luna, L.G. Manual of histologic staining methods of the
Armed Forces Institute of Pathology. 3rd edn. New
York: McGraw-Hill Book Company; 1968: 1112.
Moore, P.F., Rossitto, P.V., Danilenko, D.M.,
Wielenga, J.J., Ra, R.F., Severns, E. Monoclonal
antibodies specic for canine CD4 and CD8 dene
functional T-lymphocyte subsets and high density
expression of CD4 by canine neutrophils. Tissue
Antigens 1992; 40: 7585.
Moore, P.F., Rossitto, P.V. Development of
monoclonal antibodies to the canine T-cell receptor
complex (TCR/CD3) and their utilization in the
diagnosis of T-cell neoplasia. Veterinary Pathology
1993; 30: 457.
Moore, P.F., Rossitto, P.V., Olivry, T. Development of
monoclonal antibodies to canine T cell receptor-1
(TCR-gd) and their utilization in the diagnosis of
epidermotropic cutaneous T-cell lymphoma. Veterinary
Pathology 1994; 31: 597.
Frew, A.J., Varney, V.A., Gaga, M., Kay, A.B.
Immunohistology of human allergic late-phase skin
reactions. Skin Pharmacology 1991; 4(Suppl. 1): 718.
Ying, S., Meng, Q., Taborda-Barata, L., Kay, A.B.
Association of apoptosis of neutrophils and eosinophils
and their ingestion by macrophages with resolution of
the allergen-induced cutaneous late-phase response in
atopic human subjects. Proceedings of the Association
of American Physicians 1997; 109: 4250.
Zweiman, B., Kaplan, A.P., Tong, L.J., Moskovitz, A.R.
Cytokine levels and inammatory responses in developing late-phase allergic reactions in the skin. Journal of
Allergy and Clinical Immunology 1997; 100: 1049.
Zweiman, B., Moskovitz, A.R., von Allmen, C.
Comparison of inammatory events during developing immunoglobulin E-mediated late-phase reactions
and delayed-hypersensitivity reactions. Clinical and
Diagnostic Laboratory Immunology 1998; 5: 5747.
Kaplan, A.P., Kuna, P. Chemokines and the late-phase
reaction. Allergy 1998; 53: S27S32.
Bluestone, J.A., Khattri, R., Sciammas, R., Sperling,
A.I. TCR gamma delta cells: a specialized T-cell subset
in the immune system. Annual Review of Cell and
Developmental Biology 1995; 11: 30753.
Halliwell, R.E.W. The localization of IgE in canine
skin: an immunouorescent study. Journal of
Immunology 1973; 110: 42230.

Resume Chez les chiens et les humains atopiques, l'injection intracutanee des allergenes responsables est
suivie par l'apparition de reactions immediates et retardees. Cette etude a ete realisee pour determiner la
nature et la variation de l'inltrat inammatoire observe pendant les reactions retardees mediees par les IgE
dans la peau des chiens. Trois chiens normaux et trois chiens allergiques a Dermatophagoides farinae ont ete
inclus. Tous les chiens ont ete testes par voie intradermique avec un extrait d'acariens, un anticorps purie
anti-IgE canine (controle positif) et du solute sale (controle negatif). Des biopsies cutanees ont ete realisees
avant et 6, 12 et 24h apres injection. Les sections ont ete colorees par des colorations histologiques metachromatiques et speciques des eosinophiles. En outre, nous avons utilise une technique immunohistochimique avec des anticorps speciques des antigenes leucocytaires canins. L'etude a conrme l'apparition
d'une reaction retardee chez les chiens atopiques apres la stimulation antigenique, et chez les chiens sains et les
# 2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 4958

230 DISC
58

T. Olivry et al.
chiens atopiques apres l'injection intradermique d'anticorps speciques d'IgE. Bien que les premieres cellules
observees dans le derme aient ete des neutrophiles et des granulocytes eosinophiles actives, un aux de
lymphocytes T ab et de cellules dendritiques a ete observe dans les derniers stades des reactions retardees.
Comme les reactions retardees mediees par les IgE ressemblent aux lesions spontanees de dermatite atopique,
a la fois macroscopiquement et microscopiquement, nous proposons l'utilisation de tests similaires pour
etudier les eets anti-inammatoires des molecules anti-allergiques. [Olivry, T., Dunston, S. M., Murphy, K.
M. et Moore, P. F. Characterization of the inammatory inltrate during IgE-mediated late phase reactions in
the skin of normal and atopic dogs. (Caracterisation de l'inltrat inammatoire des reactions retardees a IgE
dans la peau de chiens normaux et de chiens atopiques.) Veterinary Dermatology 2001; 12: 4958.]
Resumen En pacientes atopicos caninos y humanos, la inoculacion intracutanea de alergenos ofensivos es
seguida por una reaccion inmediata y de fase tard a. Este estudio se realizo para ampliar la caracterizacion y la
dinamica de diferentes subgrupos de celulas inamatorias durante las reacciones de fase tard a mediadas por
IgE en la piel canina. Tres perros normales y tres perros alergicos a Dermatophagoides farinae fueron
seleccionados para este experimento. Todos los perros fueron inoculados intradermicamente con alergeno del
acaro, anticuerpos anti-IgE caninos puricados (control positivo) o suero salino tamponado con fosfato
(control negativo). Se obtuvieron biopsias cutaneas antes y 6, 12 y 24 h post-inoculacion. Las secciones fueron
tenidas con tinciones metacromaticas y espec cas de eosinolos. Ademas, utizamos un metodo
inmunohistoqu mico con anticuerpos espec cos para leucocitos caninos. Este estudio conrmo la
existencia de una reaccion de fase tard a en piel atopica despues de entrar en contacto con el alergeno, y
en piel canina normal y atopica despues de la inoculacion intradermica de anticuerpos IgE-espec cos.
Mientras las celulas emigrantes dermicas en fase temprana se compon an principalmente por neutrolos y
granulocitos eosinolos activados, hab a un inujo de linfocitos ab T y celulas dendr ticas dermicas en
reacciones de estad os posteriores y tard as. Puesto que las reacciones de fase tard a mediadas por IgE se
asemejan a las lesiones cutaneas espontaneas de la atopia canina, tanto a nivel microscopico como
macroscopico, proponemos utilizar prebas similares para estudiar los efectos anti-inamatorios de farmacos
anti-alergicos en un estudio pre-cl nico. [Olivry, T., Dunston, S. M., Murphy, K. M. y Moore, P. F.
Characterization of the inammatory inltrate during IgE-mediated late phase reactions in the skin of normal
and atopic dogs. (Caracterizacion del inltrado inamatorio durante las reacciones de fase tard a mediadas
por IgE en la piel de perros normales y atopicos.) Veterinary Dermatology 2001; 12: 4958.]
Zusammenfassung Bei atopischen Hunden und Menschen entwickeln sich nach intrakutaner Injektion von
relevanten Allergenen Reaktionen vom Sofort-wie auch vom Spattyp. Diese Studie wurde durchgefuhrt, um
die Subtypen der Entzundungszellen wahrend IgE-vermittelter Spatreaktionen in der Hundehaut weiter zu
charakterisieren. Drei normale Hunde und drei Hunde mit einer Allergie gegen Dermatophagoides farinae
wurden fur dieses Experiment ausgewahlt. Alle Hunde wurden mit Milbenallergen, gereinigten kaninen AntiIgE Antikorpern (positive Kontrolle) oder mit mit Phosphat gepuerter Salzlosung (negative Kontrolle)
behandelt. Hautbiopsien wurden vor und 6, 12 und 24 Stunden nach der Injektion genommen. Die Schnitte
wurden mit metachromatischen und eosinophil-spezischen histologischen Schnitten gefarbt. Zusatzlich
verwendeten wir eine immunhistochemische Methode mit fur kanine Leukozyten spezischen Antikorpern.
Diese Studie bestatigt das Vorhandensein einer Spatreaktion in atopischer Haut nach Antigenexponierung
und in normaler und atopischer Haut nach der Injektion von IgE-spezischen Antikorpern. Wahrend es sich
bei den fruh in die Haut emigrierenden Zellen um neutrophile und aktivierte eosinophile Granulozyten
handelte, war in den nachfolgenden Phasen der Spatreaktion ein Inux von ab T-Lymphozyten und dermalen
dendritischen Zellen festzustellen. Weil beim Hund IgE-vermittelte Spatreaktionen mikroskopisch und
makroskopisch spontanen atopischen Lasionen ahneln, schlagen wir diese Methode vor, um
entzundungshemmende Wirkungen von anti-allergischen Medikamenten in einem vorklinischen Stadium zu
testen. [Olivry, T., Dunston, S. M., Murphy, K. M. und Moore, P. F. Characterization of the inammatory
inltrate during IgE-mediated late phase reactions in the skin of normal and atopic dogs. (Charakterisierung
des entzundlichen Inltrats wahrend der IgE-vermittelten Spatreaktion in der Haut von normalen und
atopischen Hunden.) Veterinary Dermatology 2001; 12: 4958.]

# 2001 Blackwell Science Ltd, Veterinary Dermatology, 12, 4958