Original Paper

Int Arch Allergy Immunol 2010;151:80–88 Received: January 13, 2009

Accepted after revision: April 23, 2009
DOI: 10.1159/000232573
Published online: August 7, 2009

Pathophysiological Role of Skin Mast Cells in

Wound Healing after Scald Injury: Study with
Mast Cell-Deficient W/WV Mice
Naotaka Shiota Yoriko Nishikori Eiichi Kakizoe Keiko Shimoura
Tomomi Niibayashi Chiko Shimbori Tetsuya Tanaka Hideki Okunishi
Department of Pharmacology, Shimane University School of Medicine, Izumo, Japan

Key Words jured tissues of W/W V mice throughout the experiment.

Angiogenesis ⴢ Chymase ⴢ Fibrosis ⴢ Remodeling ⴢ Conclusions: Wound healing after skin scald injury was par-
Transforming growth factor- ␤1 tially impaired in mast cell-deficient mice. Mast cells may
contribute to the wound healing process, especially in the
proliferative and remodeling phases after scald injury.
Abstract Copyright © 2009 S. Karger AG, Basel
Background: The major role of mast cells in wound healing
process has not been identified. In this study, we used mast
cell-deficient W/W V mice and their congenic control (+/+) Introduction
mice to examine the role of mast cells in scald wound heal-
ing. Methods: The size of the scald wound, thickness of the Injury to tissues stimulates rapid wound healing. The
dermis, collagen deposition, vascularization, number of process of cutaneous wound healing consists of 3 sequen-
mast cells and chymase activity were measured before and tial phases: acute inflammation, proliferation and re-
at 3, 7, 14 and 21 days after inducing scald injury. Results: modeling. First, neutrophils infiltrate into damaged tis-
Although the process of wound closure and re-epithelializa- sues to remove microbes and debris in the acute inflam-
tion was not markedly different between W/W V mice and matory phase, and then macrophages arrive to clear
+/+ mice, the degree of fibrous proliferation at the wound damaged tissue and matrix. Macrophages also produce a
edge and wound vascularization in the proliferative phase variety of cytokines and growth factors that activate tis-
was significantly lower in W/W V mice than in +/+ mice, and sue proliferation and regeneration. Thereafter, tissue re-
no vascular regression in the late remodeling phase was ob- modeling proceeds and results in wound closure [1, 2].
served in W/W V mice. Mast cells producing chymase, FGF2, Mast cells are multifunctional effector cells of the im-
TGF-␤1 and VEGF were highly accumulated at the edge of mune system, and skin is one of the major sites of resting
scald wound in +/+ mice during the proliferative and remod- mast cell accumulation. In the acute inflammatory phase
eling phases at days 14 and 21. Chymase activity in the in-
jured tissues of +/+ mice decreased in the acute phase, but
recovered to no-injury level at days 14 and 21. The number
of mast cells and chymase activity were very low in the in- N.S. and Y.N. contributed equally to this study.

© 2009 S. Karger AG, Basel Correspondence to: Dr. Naotaka Shiota

1018–2438/10/1511–0080$26.00/0 Department of Pharmacology
Fax +41 61 306 12 34 Shimane University School of Medicine
E-Mail karger@karger.ch Accessible online at: 89-1 Enya-cho, Izumo, Shimane 693-8501 (Japan)
www.karger.com www.karger.com/iaa Tel. +81 853 20 2132, Fax +81 853 20 2130, E-Mail shiotana@med.shimane-u.ac.jp
after incisional injury, mast cells are activated and de- Materials and Methods
granulated within several hours [3, 4]. Recent studies
Animal Preparation
with mast cell-deficient W/W V mice showed that plasma Male W/W V mice genetically deficient for mast cells and age-
extravasation and neutrophil recruitment after excision- matched congenic normal +/+ mice (9 weeks of age) were obtained
al injury are impaired in the absence of mast cells [5, 6]. from SLC Japan (Shizuoka, Japan) and housed under convention-
These results suggest that mast cells contribute to the in- al animal laboratory conditions with access to standard chow and
flammatory response during the acute phase of wound tap water ad libitum. Deep-dermal second-degree scald injury
was induced as described previously [13]. Briefly, hairs on the dor-
healing. sal skin of W/W V and +/+ mice were removed using a depilatory
Mast cells also produce various growth factors and cream (Kanebo Co., Tokyo, Japan) 1 day before induction of scald
proteases. Our previous study showed that the number of injury. Under ketamine anesthesia, scald injury was induced on 1
mast cells increased markedly during fibrous prolifera- side of the dorsal skin by applying a heated (100 ° C) cylindrical
tion in the skin of tight-skin mice, and mast cell chymase lead block (10 mm diameter) for 5 s. At 3, 7, 14 and 21 days after
induction of scald injury, scald-injured W/W V mice (n = 6 for
promoted activation of latent TGF-␤1 in fibrous tissue, each time-point) and +/+ mice (n = 6 for each time-point) were
thereby accelerating the development of skin fibrosis in sacrificed by administration of a lethal dose of ketamine, and dor-
tight-skin mice [7]. However, the importance of mast cells sal skins with panniculus carnosus muscle layers were collected
in the proliferative and remodeling phases of the wound for histological analysis and measurement of chymase activity.
healing process is still unclear and controversial. Com- Age-matched male W/W V (n = 6) and +/+ mice (n = 6), which un-
derwent hair removal but had no scald injury, served as uninjured
pound 48/80 (a mast cell degranulator) and exogenous normal controls. This study was conducted in accordance with
histamine increase tensile strength and hydroxyproline the ‘Guide for the Care and Use of Laboratory Animals’ prepared
content in incised wounds [8, 9]. Furthermore, a study by Shimane University School of Medicine.
with mast cell-deficient W/W V mice showed that wound
closure after excisional injury is impaired in the absence Histological Analysis
Skin tissues were fixed with neutral buffered formalin, paraf-
of mast cells [6]. These results suggest that mast cell-de- fin-embedded, and cut into 5-␮m-thick serial sections. Cross sec-
rived factors may promote a fibroproliferative response tions through the center of the scald wound were used for histo-
and accelerate wound healing. However, other studies us- logical analysis. Using a light microscope, morphologic charac-
ing W/W V mice showed that the absence of mast cells has teristics of skin sections from scald-injured W/W V and +/+ mice,
no effect on the proliferative aspects of wound healing and from uninjured W/W V and +/+ mice, were analyzed. The size
of the scald wound was indicated by the volume of an elliptical
after incisional injury [5], but does raise the hydroxypro- cone calculated from the macroscopic surface area and depth
line content of granulation tissue after excisional injury (both measured in hematoxylin-eosin-stained sections) [13].
[10]. Wound edges were defined as the border separating the damaged
Several points should be noted regarding rodent mod- from the undamaged areas, and the distance between the wound
els of cutaneous injury. First of all, the major mechanism edges was measured. The length of regenerated epithelium cover-
ing the wound was also measured. Then percentage of re-epithe-
of wound closure after surgical injury in rodents is con- lialization was calculated by dividing the total length of regener-
traction, which leads to rapid closure of defects before ated epithelium by the distance between wound edges. Dermal
granular tissue can form [11]. Secondly, the number of thickness (from the surface of the regenerated epidermis to the
skin mast cells does not increase above the control level deep connective tissue layer just above the panniculus carnosus
in uninjured skin following surgical injury [4, 12]. On muscle layer) was measured at the center and at both edges of the
scald wound. We could not measure dermal thickness at the cen-
the contrary, potent granular tissue formation and ter of the scald wound on days 3 and 7 because the center region
marked increase in the number of mast cells were ob- was not yet covered with regenerated epidermis, and it was im-
served in our previous study after deep-dermal scald in- possible to identify the border between necrotic and viable tissue.
jury [13]. Therefore, scald injury may be more suitable Therefore, we could not calculate the fibrotic area and the number
than uncomplicated surgical injury in rodents for ana- of vessels and mast cells in the center of the scald wound on days
3 and 7. The extent of fibrosis on Azan-stained sections was mea-
lyzing the role of mast cells, especially in the late phase sured using an image analyzer (Image Pro; Media Cybernetics,
of wound healing. To further explore the role of mast Silver Spring, Md., USA) and indicated by the blue-stained area
cells in wound healing, we investigated the process of re- expressed as a percentage per square millimeter of the area mea-
epithelialization, vascularization, and fibrous prolifera- sured. For analyzing wound vascularization, sections were stained
tion after scald injury in mast cell-deficient mice with primary antibody to factor VIII (1:200 dilution; Santa Cruz
Biotechnology, Santa Cruz, Calif., USA) overnight at 4 ° C, and
(WBB6F1-W/W V; W/W V) and their congenic normal then the antibody was detected using a Dako Envision kit (Dako-
(WBB6F1-+/+; +/+) littermates. cytomation, Tokyo, Japan) according to the manufacturer’s direc-
tions, and visualized with diaminobenzidine substrate (Dakocy-

Role of Mast Cells in Wound Healing Int Arch Allergy Immunol 2010;151:80–88 81
 20 NS +/+
W/WV
NS
100 NS
15

Scald wound size (mm3)

NS

Re-epithelialization (%)
Fig. 1. Closure and re-epithelialization of 10
the scald wounds of +/+ mice and W/W V
mice. Wound size (a) and the percentage 50
of wound re-epithelialization (b) of scald
NS
wounds in +/+ mice and W/W V mice were 5
calculated at 3, 7, 14 and 21 days after scald NS
injury. There were no statistically signifi- NS
cant between-group differences in the ex- NS
tent of wound closure and re-epithelializa- 0 0
tion during the experiment. Results are 3 7 14 21 3 7 14 21
shown as the mean 8 SEM (n = 6 for each a Days b Days
time point). NS = Not significant.

tomation). The number of vessels (including capillaries and large Results

vessels) was counted and the density (vessels per square millime-
ter) was calculated. Toluidine blue staining was used to identify Closure and Re-Epithelialization of Scald Wounds
mast cells on tissue sections [7, 13]. The number of mast cells was
counted at the center and at both edges on toluidine blue-stained The wound was swollen immediately after scald in-
sections, and the counts were expressed as n mast cells/mm2 of jury. Thereafter, the upper layer of skin became necrotic
the area measured. by day 3 and then covered by crust by day 7. The wound
size in +/+ mice and W/W V mice decreased rapidly by day
Measurement of Chymase Activity 14, and almost all wounds were closed by day 21 (fig. 1a).
Chymase was extracted from the dorsal skin, and its activity
was measured as described previously [13]. Since the size of skin In both groups, re-epithelialization started from the
tissue samples was too small to measure chymase activity sepa- wound edge by day 3, and was about 80% complete by day
rately at the center and edge, chymase activity was measured in 14 and complete by day 21 (fig. 1b). There were no statis-
the entire sample. tically significant between-group differences in scald
wound size and percentage of re-epithelialization at each
Immunohistochemical Analysis of Chymase, FGF2, TGF-␤1
and VEGF time point (fig. 1). Wound closure and re-epithelializa-
For detection of chymase, FGF2, TGF-␤1 and VEGF in paraf- tion occurred at same speed in +/+ mice and W/W V mice.
fin sections of skin tissues, polyclonal rabbit antibodies against Hence, the absence of mast cells appeared to have no ef-
chymase (kindly provided by Dr. G.H. Caughey [14]), FGF2, TGF- fect on wound closure and re-epithelialization.
␤1 and VEGF (Santa Cruz Biotechnology) were used. Paraffin
sections were incubated overnight at 4 ° C with anti-chymase
(1:10,000), anti-FGF2 (1:2,000), anti-TGF-␤1 (1:500), anti-VEGF Dermal Thickness and Fibrous Proliferation
(1:200), or equal amounts of nonimmune serum (control for the Dermal thickness at the wound edge was significantly
specificity of antibodies). The antibodies were then detected us- increased after scald injury in both +/+ mice and W/W V
ing a Dako Envision kit (Dakocytomation) and visualized with mice. There was no statistically significant between-
diaminobenzidine substrate. group difference in dermal thickness at the wound edge
Statistical Analysis at days 3 and 7. At day 7, this thickness had peaked in
All data are expressed as the means 8 SEM. Comparisons be- W/W V mice, continued to increase at day 14 in +/+ mice,
tween 2 groups were made by Student’s t test for unpaired data. and was thinner in W/W V mice than in +/+ mice in the
For multiple comparisons, data were examined by analysis of late phase of wound healing at days 14 and 21 (fig. 2, 3a).
variance (ANOVA) followed by the Bonferroni/Dunn’s test. Dif- Fibrotic area at the wound edge was significantly in-
ferences were considered statistically significant when p ! 0.05.
creased by day 7 in both groups (fig. 3b), but by days 14
and 21, it was smaller in W/WV mice than in +/+ mice

82 Int Arch Allergy Immunol 2010;151:80–88 Shiota/Nishikori/Kakizoe/Shimoura/

Niibayashi/Shimbori/Tanaka/Okunishi
 Color version available online
(fig. 2, 3b). At the wound center, dermal thickness was not a
increased after scald injury, and there were no between-
group differences in dermal thickness and fibrotic area at
each time point (data not shown).

Angiogenesis
Vessel regeneration was observed at the wound edge
in both groups by day 7. The number of vessels at the
wound edge peaked in +/+ mice at day 14 and then de-
creased significantly at day 21, while it increased con-
tinuously in W/W V mice until the end of experiment
at day 21. This number was higher in +/+ mice than in
W/W V mice at day 7, but lower in +/+ mice than in
W/W V mice at day 21 (fig. 4a). At the wound center, the b
number peaked in +/+ mice at day 14 and then de-
creased significantly at day 21, but in W/W V mice, the
number continued to increase up to day 21. The num-
ber of vessels at the wound center was higher at day 14
but lower at day 21 in +/+ mice than in W/W V mice
(fig. 4b). In both groups, angiogenesis took place during
the proliferative phase, but to a lesser extent in W/W V
mice than in +/+ mice at days 7 and 14. Furthermore,
vascular regression took place in the late remodeling
phase at day 21 in normal +/+ mice but was not ob-
served in W/W V mice.

Number of Mast Cells and Chymase Activity

At the wound edge in +/+ mice, the number of mast Fig. 2. Impaired fibrous proliferation at the wound edge in the late
cells was significantly decreased at day 3 and began to phase of healing in W/W V mice. Representative photographs of
increase in the deep dermis at day 7. Then mast cells at Azan-stained cross sections of skin from a +/+ mouse and W/W V
mouse at 21 days after scald injury. The dermis was thinner and
the wound edge in +/+ mice increased markedly around fibrous proliferation was less extensive at the wound edge at 21
granulation tissues at days 14 and 21 (fig. 5a, 6a). At the days after inducing scald injury in W/W V mice than in +/+ mice.
wound center in +/+ mice, the number of mast cells was Scale bar = 200 ␮m.
decreased at day 3, and recovered to pre-injury control
level by days 14 and 21 (data not shown). Few mast cells
were observed in the injured tissues of W/W V mice (fig. Discussion
5b, 6a). In +/+ mice, the chymase activity of whole in-
jured tissues was decreased at days 3 and 7, recovered The presence of mast cells just beneath the epidermis
to pre-injured control level at day 14, and continued to suggests that they may have some role in regulating ho-
increase at day 21 (fig. 6b). On the other hand, in W/WV meostasis of epidermis. Previous studies showed that
it remained very low throughout the experiment mast cell-derived factors, such as histamine, heparin and
(fig. 6b). TNF-␣, inhibited keratinocyte growth in vitro [15–17].
Furthermore, increased numbers of mast cells have been
Immunohistochemical Analysis of Chymase, FGF2, reported near the epithelial margin of chronic nonheal-
TGF-␤1 and VEGF ing leg ulcers, and chymase derived from mast cells im-
In +/+ mice, mast cells in the skin tissues (particular- paired keratinocyte migration and adherence [18]. Thus,
ly around the granulation tissues in the late phase of mast cells may inhibit re-epithelialization at least in vitro
wound healing) were immunopositive for chymase and in pathological conditions. Therefore, we expected
(fig. 7a), FGF2 (fig. 7c), TGF-␤1 (fig. 7e) and VEGF that re-epithelialization might be promoted in mast cell-
(fig. 7g) at day 14 (data not shown) and day 21 (shown). deficient W/W V mice compared with normal +/+ mice

Role of Mast Cells in Wound Healing Int Arch Allergy Immunol 2010;151:80–88 83
 1.0 +/+ 80
W/WV b b
Fig. 3. Dermal thickness and fibrotic area b
a

Fibrotic area at edge of scald wound (%)

a

Thickness at edge of scald wound (mm)

at the edge of scald wounds in +/+ mice a b
and W/W V mice. Dermal thickness (a) and 0.8 NS
a 60 NS NS
fibrotic area (b) at the edge of scald wounds NS a a
a a a a
in +/+ mice and W/W V mice were mea- a a
0.6 a
sured before (no injury) and after scald in- a
NS
jury. The dermis at the wound edge in the 40
NS
late phase of healing (at 14 and 21 days) was
thinner in W/W V mice than in +/+ mice 0.4
(a). Fibrotic area at the wound edge in the
late phase of healing (at 14 and 21 days) was 20
smaller in W/W V mice than in +/+ mice 0.2
(b). Results are shown as the mean 8 SEM
(n = 6 for each time point). a p ! 0.05, in-
jured vs. uninjured normal (no injury) 0 0
skin tissues within the same mouse strain. No injury 3 7 14 21 No injury 3 7 14 21
b
p ! 0.05, +/+ mice vs. W/W V mice. NS = a Days b Days
Not significant.

b
300 +/+ 300
b
W/WV
a

Vessels at center of scald wound (mm–2)

Vessels at edge of scald wound (mm–2)

b b
b
Fig. 4. The number of vessels at the edge a
a
200 NS 200
and center of scald wounds in +/+ mice b
and W/W V mice. The number of vessels at a
a
the edge (a) and center (b) of scald wounds a
in +/+ mice and W/W V mice was mea- a a
sured before (no injury) and after scald in- NS a
jury. Angiogenesis was retarded and vas- 100 100
cular regression was not observed in the NS NS
wound tissues of W/W V mice. Results are
shown as the mean 8 SEM (n = 6 for each
time point). a p ! 0.05, injured vs. unin-
jured normal (no injury) skin tissues with- 0 0
in the same mouse strain. b p ! 0.05, +/+ No injury 3 7 14 21 No injury 3 7 14 21
mice vs. W/W V mice. NS = Not signifi- a Days b Days
cant.

after scald injury. However, there was no difference in re- and ␣2-macroglobulin in interstitial fluid strongly inhib-
epithelialization between normal +/+ mice and W/W V it chymase [19, 20]. Chymase released from degranulated
mice. Another study using W/WV mice also found mast mast cells may be suppressed by natural protease inhibi-
cells played no role in re-epithelialization after excisional tors. Because chymase is an important inhibitory factor
skin injury [5]. The precise reason is unclear why mast for re-epithelialization [18], inhibition of chymase activ-
cells may not have major role on the physiological re-ep- ity early in the wound healing process may favor initia-
ithelialization process, but the substantial suppression of tion of re-epithelialization. Actually, re-epithelialization
chymase activity within the wound of +/+ mice at days 3 in +/+ mice after scald injury started within 3 days, and
and 7 may partly explain the reason. The wound region was not delayed compared with that in W/W V mice. Oth-
became edematous immediately after scald injury and er mediators may also be released into the wound regions
filled with interstitial fluid containing a variety of natu- immediately after scald wounding; however, their role on
ral protease inhibitors. Especially, ␣1-antichymotrypsin re-epithelialization was not determined.

84 Int Arch Allergy Immunol 2010;151:80–88 Shiota/Nishikori/Kakizoe/Shimoura/

Niibayashi/Shimbori/Tanaka/Okunishi
 Color version available online
a b

Fig. 5. Mast cell accumulation at the edge

of scald wounds of +/+ mice. Representa-
tive photographs of toluidine blue-stained
cross sections of skin from a +/+ mouse (a)
and W/W V mouse (b) at 21 days after scald
injury. Reacting with toluidine blue, mast
cells are metachromatic (i.e. stained red-
purple). Many mast cells were present at
the wound edge in +/+ mice, but few were
found at the wound edge in W/W V mice.
Scale bar = 50 ␮m.

150 +/+ 40
Fig. 6. The number of mast cells at the edge W/WV b
and chymase activity of skin tissues in +/+
Mast cells at edge of scald wound (mm–2)

a
mice and W/W V mice. The number of Chymase activity (mU/mg protein)
mast cells at the wound edge (a) and chy- b
30 b
mase activity (b) in the skin tissues of +/+ a
100
mice and W/W V were measured before (no b
injury) and after scald injury. The number b
of mast cells increased dramatically at the b 20
wound edge in +/+ mice (a). Chymase ac- a
tivity in the skin tissues of +/+ mice was b
decreased significantly 3 and 7 days after 50
b a
injury, and then recovered to the pre- b 10
injury level at 14 and continued to in- b a
crease at 21 days (b). Results are shown as a
the mean 8 SEM (n = 6 for each time
point). a p ! 0.05, injured vs. uninjured 0 0
normal (no-injury) skin tissues within the No injury 3 7 14 21 No injury 3 7 14 21
same strain of mice. b p ! 0.05, +/+ mice vs. a Days b Days
W/W V mice.

In our study, although wound closure and re-epitheli- suggest that mast cells may contribute to wound healing
alization were not significantly different between W/WV during the proliferative phase. However, previous reports
mice and +/+ mice, stronger fibrous proliferation occurred suggested that mast cells were superfluous during prolif-
at the wound edge in +/+ mice than in W/WV mice. Impor- eration after excisional skin injury [5, 10]. This discrep-
tantly, the number of mast cells increased markedly at the ancy may be due to the difference in the methods used for
wound edge in +/+ mice during the proliferative phase. The inducing skin injury. Another study showed that re-epithe-
proliferative response and increase of mast cell number oc- lialization speed was slower and fibroblastic proliferation
curred in the same area at the same time. These results amount was greater for scald wounds than excisional

Role of Mast Cells in Wound Healing Int Arch Allergy Immunol 2010;151:80–88 85
 Color version available online
a b

c d

e f

Fig. 7. Immunohistochemical localization

of chymase, FGF2, TGF-␤1 and VEGF in
the scald-injured skin of +/+ mice. Cross
sections of skin from +/+ mice at 21 days g h
after scald injury were incubated with an-
tibody against chymase, FGF2, TGF-␤1 or
VEGF. The brown stain indicates the pres-
ence of chymase (a), FGF2 (c), TGF-␤1 (e)
or VEGF (g). To identify mast cells, adja-
cent sections were stained with toluidine
blue. Mast cells displayed toluidine blue
metachromasia (b, d, f, h) and were immu-
nopositive for chymase, FGF2, TGF-␤1
and VEGF. Arrowheads point to mast
cells. Scale bar = 50 ␮m.

wounds [21]. Healing after scald injury is a more compli- at day 7, and continued to increase markedly at days 14 and
cated and prolonged process than that after excisional or 21. Although the precise mechanism is unclear, recruit-
incisional injury. Importantly, the number of mast cells in ment of mast cell progenitors into the injured regions and/
surgically injured skin has been found to decrease in the or local proliferation of mast cells may be more strongly
early phase and then recover without increasing further in activated in the late phase of wound healing after scald in-
the late phase of wound healing [4, 12]. In our present study, jury than that after uncomplicated surgical injury.
the number of mast cells at wound edge in normal +/+ mice Mast cells produce various profibrotic factors includ-
was also decreased at day 3, but exceeded the control level ing TGF-␤1. TGF-␤1 is a multifunctional growth factor

86 Int Arch Allergy Immunol 2010;151:80–88 Shiota/Nishikori/Kakizoe/Shimoura/

Niibayashi/Shimbori/Tanaka/Okunishi
known to stimulate fibroblast proliferation and matrix stimulates endothelial cell function and neovasculariza-
formation and to have wound healing effects [22, 23]. tion at low concentration and inhibits angiogenesis at rel-
TGF-␤ applied to incisional wounds in rats causes atively high concentration [32]. Interestingly, our study
marked deposition of collagen, increases mechanical found vascular regression in the wound tissues of normal
strength of regenerated tissue, and accelerates the rate of +/+ mice in the remodeling phase at day 21, but not in the
healing [24]. On the other hand, exogenous addition of wound tissues of mast cell-deficient W/WV mice. In the
neutralizing antibody against TGF-␤1 and TGF-␤2 to in- latter, the number of vessels continued to increase through-
cisional wounds in rats reduces extracellular matrix de- out the experiment. The number of mast cells and chy-
position [25]. Furthermore, granulation tissue formation, mase activity peaked in the wound tissues of normal +/+
vascularity, and collagen deposition were all decreased in mice at 21 days after scald injury, suggesting that active
full-thickness excisional wounds of TGF-␤1-deficient TGF-␤1 in the scald wound of normal +/+ mice may reach
mice [26]. These results suggest that TGF-␤1 is crucial for a high enough concentration to inhibit angiogenesis dur-
normal wound healing. Our present study showed that ing late remodeling. Furthermore, mast cells are potent
mast cells also produce TGF-␤1. In addition, a mast cell- regulator of matrix remodeling [33]. Composition and mi-
specific protease, chymase, plays a key role in the regula- croenvironment of extracellular matrix are also crucial
tion of extracellular matrix formation through activation angiogenesis-regulatory factors. Recent study showed that
of latent TGF-␤1 [7, 27]. Interestingly, mast cells also pro- collagen reorganization in the regenerated skin tissues of
duce a latent form of TGF-␤1, which is immediately con- mast cell-deficient W/WV mice was suppressed in the late
verted by co-secreted chymase into active TGF-␤1 [28]. phase of wound healing after excisional injury [10]. Taken
Our present study showed that the number of mast cells together, these findings suggest that mast cells may play an
expressing both chymase and TGF-␤1 was markedly in- important role in the regulation of matrix reorganization
creased at the wound edge in the late phase of wound and affect the delicate balance between pro-angiogenic
healing. Chymase activity in the total wound area was factors and anti-angiogenic factors in wound tissues.
also significantly higher in +/+ mice than in W/W V mice. The limitation of the present study is that, for techni-
These results suggest that the concentration of active cal reasons, we could not provide data for mast cell-
TGF-␤1 in wound tissues may be higher in +/+ mice than reconstituted W/W V mice. Although W/W V mice ex-
W/W V mice, especially late in the wound healing process. press profound deficiency of mast cells, W/W V mice also
Thus, mast cells may affect the proliferative phase of show such characteristics as macrocytic anemia, neutro-
wound healing after scald injury. penia, sterility, impaired melanogenesis, impaired prolif-
New blood vessels are needed at the wound site to in- eration of keratinocytes in hair follicles, lack of intestinal
crease the blood supply and promote tissue repair. During cells of Cajal and decrease in the number of intestinal
the proliferative phase of wound healing, numerous new lymphocytes [34–36]. With regard to dermal wound heal-
capillaries penetrate the granulation tissue to create a mi- ing process, in addition to mast cell deficiency, neutrope-
crovascular network. As granulation tissue is transformed nia and impaired proliferation of keratinocytes may also
to collagen-rich scar tissue during the late matrix remodel- affect the wound healing process. Neutrophils may not
ing phase of normal wound healing, the density of blood have a major role in the proliferative and remodeling
vessels diminishes [29]. Since mast cells accumulate near phase of the late wound healing process, and re-epitheli-
blood vessels and produce several potent angiogenic fac- alization process was not different between W/W V and
tors such as FGF2, VEGF and TGF-␤1, mast cells have +/+ mice in our present study, but we could not exclude
been implicated in angiogenesis [30, 31]. However, a previ- these alternative possibilities without doing the experi-
ous study using W/WV mice could not determine whether ment with mast cell-reconstituted W/W V mice.
mast cells affect wound angiogenesis after excisional in- In conclusion, our present study showed that wound
jury [5], and thus the precise role of mast cells has been healing after skin scald injury was partially impaired in
unclear. Our present study showed that the angiogenic re- mast cell-deficient mice. Although no obvious effect of
sponse to scald injury during the proliferative phase of mast cells on wound closure and re-epithelialization was
healing was weaker in mast cell-deficient W/WV mice observed, mast cells affected proliferative response and
than in normal +/+ mice. Although the pro-angiogenic vascular homeostasis in wound tissues. Since mast cells
role of FGF2 and VEGF has been well defined, that of are increased markedly during the proliferative and re-
TGF-␤1 has not, but its effects on angiogenesis are like- modeling phases after scald injury, studying the role of
ly to be biphasic and concentration-dependent. TGF-␤1 mast cells in this model of scald injury has advantages.

Role of Mast Cells in Wound Healing Int Arch Allergy Immunol 2010;151:80–88 87
Interestingly, wound closure was completed within 3 Acknowledgment
weeks, whereas the number of mast cells in injured skin
The authors would like to thank Dr. G.H. Caughey, University
was continuously increased in +/+ mice. Further studies of California, San Francisco for kindly providing the antibody
in the late remodeling phase after wound closure are war- against chymase.
ranted to clarify whether the observed impairment in
mast cell-deficient mice is transient or affects the general
quality of the healed wounds.

References

1 Eming SA, Krieg T, Davidson JM: Inflam- promotion of matrix remodeling in burn by transforming growth factor-beta. Science
mation in wound repair: molecular and cel- wound healing in mice: relevance of chymase. 1987;237:1333–1336.
lular mechanisms. J Invest Dermatol 2007; Arch Dermatol Res 1998;290:553–560. 25 Shah M, Foreman DM, Ferguson MW: Neu-
127:514–525. 14 Fang KC, Wolters PJ, Steinhoff M, Bidgol A, tralising antibody to TGF-beta 1,2 reduces
2 Martin P, Leibovich SJ: Inflammatory cells Blount JL, Caughey GH: Mast cell expression cutaneous scarring in adult rodents. J Cell
during wound repair: the good, the bad and of gelatinases A and B is regulated by kit li- Sci 1994;107:1137–1157.
the ugly. Trends Cell Biol 2005; 15:599–607. gand and TGF-beta. J Immunol 1999; 162: 26 Brown RL, Ormsby I, Doetschman TC,
3 El Sayed SO, Dyson M: Responses of dermal 5528–5535. Greenhalgh DG: Wound healing in the
mast cells to injury. J Anat 1993; 182: 369– 15 Symington FW: Lymphotoxin, tumor necro- transforming growth factor-beta-deficient
376. sis factor, and gamma interferon are cyto- mouse. Wound Repair Regen 1995;3:25–36.
4 Hebda PA, Collins MA, Tharp MD: Mast cell static for normal human keratinocytes. J In- 27 Taipale J, Lohi J, Saarinen J, Kovanen PT,
and myofibroblast in wound healing. Der- vest Dermatol 1989; 92:798–805. Keski-Oja J: Human mast cell chymase and
matol Clin 1993;11:685–696. 16 Pillai S, Gilliam L, Conrad HE, Holleran leukocyte elastase release latent transform-
5 Egozi EI, Ferreira AM, Burns AL, Gamelli WM: Heparin and its non-anticoagulant an- ing growth factor-ß1 from the extracellular
RL, Dipietro LA: Mast cells modulate the alogues inhibit human keratinocyte growth matrix of cultured human epithelial and en-
inflammatory but not the proliferative re- without inducing differentiation. J Invest dothelial cells. J Biol Chem 1995; 270: 4689–
sponse in healing wounds. Wound Repair Dermatol 1994;103:647–650. 4696.
Regen 2003;11:46–54. 17 Huttunen M, Hyttinen M, Nilsson G, But- 28 Lindstedt KA, Wang Y, Shiota N, Saarinen J,
6 Weller K, Foitzik K, Paus R, Syska W, Mau- terfield JH, Horsmanheimo M, Harvima IT: Hyytiainen M, Kokkonen JO, Keski-Oja J,
rer M: Mast cells are required for normal Inhibition of keratinocyte growth in cell Kovanen PT: Activation of paracrine TGF-
healing of skin wounds in mice. FASEB J culture and whole skin culture by mast cell beta1 signaling upon stimulation and de-
2006;20:2366–2368. mediators. Exp Dermatol 2001;10:184–192. granulation of rat serosal mast cells: a novel
7 Shiota N, Kakizoe E, Shimoura K, Tanaka T, 18 Huttunen M, Harvima IT: Mast cell tryptase function for chymase. FASEB J 2001; 15:
Okunishi H: Effect of mast cell chymase in- and chymase in chronic leg ulcers: chymase 1377–1388.
hibitor on the development of scleroderma in is potentially destructive to epithelium and 29 Moulin V: Growth factors in skin wound
tight-skin mice. Br J Pharmacol 2005; 145: is controlled by proteinase inhibitors. Br J healing. Eur J Cell Biol 1995; 68:1–7.
424–431. Dermatol 2005;152:1149–1160. 30 Hiromatsu Y, Toda S: Mast cells and angio-
8 Yoshida S, Flancbaum L, Fitzpatrick JC, Berg 19 Schechter NM, Sprows JL, Schoenberger OL, genesis. Microsc Res Tech 2003;60:64–69.
RA, Fisher H: Effects of compound 48/80 Lazarus GS, Cooperman BS, Rubin H: Reac- 31 Noli C, Miolo A: The mast cell in wound
and exogenous histamine on wound healing tion of human skin chymotrypsin-like pro- healing. Vet Dermatol 2001;12:303–313.
in mice. Proc Soc Exp Biol Med 1989;191:90– teinase chymase with plasma proteinase 32 Bertolino P, Deckers M, Lebrin F, ten Dijke
92. inhibitors. J Biol Chem 1989; 264: 21308– P: Transforming growth factor-beta signal
9 Bairy KL, Rao CM, Ramesh KV, Kulkarni 21315. transduction in angiogenesis and vascular
DR: Effect of histamine on wound healing. 20 Walter M, Sutton RM, Schechter NM: High- disorders. Chest 2005; 128(6 suppl):585S–
Indian J Physiol Pharmacol 1991; 35: 180– ly efficient inhibition of human chymase by 590S.
182. alpha(2)-macroglobulin. Arch Biochem Bio- 33 Galli SJ, Tsai M: Mast cells: versatile regula-
10 Iba Y, Shibata A, Kato M, Masukawa T: Pos- phys 1999;368:276–284. tors of inflammation, tissue remodeling,
sible involvement of mast cells in collagen 21 Schaffer CJ, Reinisch L, Polis SL, Stricklin host defense and homeostasis. J Dermatol
remodeling in the late phase of cutaneous GP: Comparisons of wound healing among Sci 2008;49:7–19.
wound healing in mice. Int Immunophar- excisional, laser-created, and standard ther- 34 Peters EM, Maurer M, Botchkarev VA, Jen-
macol 2004;4:1873–1880. mal burns in porcine wounds of equal depth. sen K, Welker P, Scott GA, Paus R: Kit is ex-
11 Galiano RD, Michaels J, Dobryansky M, Wound Repair Regen 1997;5:52–61. pressed by epithelial cells in vivo. J Invest
Levine JP, Gurtner GC: Quantitative and 22 Werner S, Grose R: Regulation of wound Dermatol 2003;121:976–984.
reproducible murine model of excisional healing by growth factors and cytokines. 35 Grimbaldeston MA, Chen CC, Piliponsky
wound healing. Wound Repair Regen 2004; Physiol Rev 2003;83:835–870. AM, Tsai M, Tam SY, Galli SJ: Mast cell-defi-
12:485–492. 23 Beanes SR, Dang C, Soo C, Ting K: Skin re- cient W-sash c-kit mutant Kit W-sh/W-sh
12 Persinger MA, Lepage P, Simard JP, Parker pair and scar formation: the central role of mice as a model for investigating mast cell bi-
GH: Mast cell numbers in incisional wounds TGF-beta. Expert Rev Mol Med 2003; 5: 1– ology in vivo. Am J Pathol 2005;167:835–848.
in rat skin as a function of distance, time and 22. 36 Zhou JS, Xing W, Friend DS, Austen KF, Katz
treatment. Br J Dermatol 1983; 108:179–187. 24 Mustoe TA, Pierce GF, Thomason A, Gra- HR: Mast cell deficiency in Kit(W-sh) mice
13 Nishikori Y, Kakizoe E, Kobayashi Y, Shi- mates P, Sporn MB, Deuel TF: Accelerated does not impair antibody-mediated arthri-
moura K, Okunishi H, Dekio S: Skin mast cell healing of incisional wounds in rats induced tis. J Exp Med 2007;204:2797–2802.

88 Int Arch Allergy Immunol 2010;151:80–88 Shiota/Nishikori/Kakizoe/Shimoura/

Niibayashi/Shimbori/Tanaka/Okunishi