Clin Exp Immunol 2000; 119:264±269

Development of dextran sulphate sodium-induced experimental colitis is

suppressed in genetically mast cell-de®cient Ws/Ws rats

Y. ARAKI, A. ANDOH, Y. FUJIYAMA & T. BAMBA Department of Internal Medicine, Shiga University of Medical
Science, Otsu, Japan

(Accepted for publication 15 September 1999)

SUMMARY
Ws/Ws rats have a small deletion of the c-kit gene, and are de®cient in both mucosal-type mast cells
(MMC) and connective tissue-type mast cells (CTMC). In the present study we investigated the role of
intestinal MMC in the development of dextran sulphate sodium (DSS)-induced experimental colitis
using Ws/Ws rats. Ws/Ws and control (‡/‡) rats were given a 3% DSS aqueous solution orally for
10 days, and the subsequent mucosal damage was evaluated macroscopically and histologically. The
mucosal myeloperoxidase (MPO) activities and histamine levels were also measured. (i) DSS induced
severe oedema and hyperaemia with sporadic erosions in the control (‡/‡) rats, but these changes were
signi®cantly attenuated in the Ws/Ws rats (P < 0´01). (ii) The microscopic mucosal damage score was
lower in the Ws/Ws rats than in the control (‡/‡) rats (P ˆ 0´06). (iii) There were no signi®cant
differences in mucosal MPO activity between the Ws/Ws and control (‡/‡) rats (P ˆ 0´46). (iv) The
mucosal histamine levels in the colon were signi®cantly reduced in the Ws/Ws rats compared with the
control (‡/‡) rats (P < 0´05). (v) Signi®cant positive correlations were observed between mucosal
histamine levels and the degree of mucosal oedema (calculated as colonic wet weight/protein content)
(r ˆ 0´778, P < 0´01), and between histamine levels and the macroscopic damage (r ˆ 0´623, P < 0´05),
respectively. (vi) DSS induced a local recruitment of MMC in the colonic mucosa of Ws/Ws rats, and
mucosal damage gradually increased in accordance with this MMC recruitment. These results indicate
that MMC play an important role in the development of DSS colitis.

Keywords in¯ammatory bowel disease c-kit gene mucosal mast cell

INTRODUCTION induces experimental colitis in animal models, and these colonic

lesions are similar to those observed in UC patients [10].
Mast cell activation leads to the release of large quantities of
Therefore, the purpose of the present study was to elucidate the
chemical mediators such as histamine [1,2], leukotrienes [3] and
role of mast cell activation in the development of DSS-induced
platelet-activating factor (PAF) [4], which then play an important
experimental colitis in rats. The role of mast cell activation was
role in acute and chronic tissue injury. The pathogenesis of
evaluated in mast cell-de®cient Ws/Ws (white spotting in the skin)
in¯ammatory bowel disease (IBD), including ulcerative colitis
rats. Ws/Ws rats have a 12-base deletion in the tyrosine kinase
(UC) and Crohn's disease (CD), remains unknown. However,
domain of the c-kit cDNA, and are genetically de®cient in both
recent studies have suggested that mast cells may play a role in
connective tissue-type (CTMC) and MMC [11]. These rats are
the pathogenesis of IBD [5,6]. For example, it has been reported
useful for studying the role of mast cells in the pathogenesis of
that mast cell in®ltration increases in the colonic mucosa of active
IBD.
IBD patients [7], and that mucosal histamine levels are signi®-
cantly elevated in UC patients [8]. Furthermore, the therapeutic
agents used for the treatment of IBD patients can also inhibit the MATERIALS AND METHODS
release of chemical mediators from intestinal mucosal-type mast
Animals
cells (MMC) [9].
The origin of Ws/Ws rats has been described in detail elsewhere
The oral administration of dextran sulphate sodium (DSS)
[12,13]. Ws/Ws rats with a large white spot and coat colour dilution
Correspondence: Yoshio Araki MD, Second Department of Internal were maintained by the mating of heterozygous Ws/‡ rats at SCL,
Medicine, Shiga University of Medical Science, Seta Tukinowa, Otsu 520- Inc. (Shizuoka, Japan). These matings generate Ws/Ws, Ws/‡ and
2192, Japan. (‡/‡) rats. We used speci®c pathogen-free male Ws/Ws and (‡/‡)

264 q 2000 Blackwell Science

 DSS colitis in Ws/Ws rats 265
rats (n ˆ 6), weighing 200±250 g, in this experiment. The rats were (14 000 rev/min, for 20 min), and the supernatants were used for
maintained under standard laboratory conditions with controlled the measurement of MPO activity with 0´0005% hydrogen per-
temperature (20±228C), humidity (50±60%) and a light±dark oxide as the substrate. The protein content of the supernatant was
cycle (12 h:12 h). The experimental protocol was approved by also measured according to the method of Lowry [19].
the Animal Care and Use Committee of Shiga University of
Medical Science. Measurement of intestinal mucosal histamine
The concentration of mucosal histamine was measured by high
Development of DSS colitis performance liquid chromatography (HPLC) using a minor modi-
The rats were fed standard rat chow CE-2 (Nippon Clea Inc., ®cation of the previously described method [20]. Brie¯y, 800 ml
Tokyo, Japan), and were provided with distilled water containing of 10% aqueous TCA solution was added to 400 ml of the super-
3% (w/v) DSS (mol. wt 5000 D, total sulphur 15´0±20´0%; Wako natant prepared as described above. After mechanical shaking and
Pure Chemical Industries Ltd, Osaka, Japan) ad libitum for 10 days. centrifugation, the supernatant was ®ltered through a membrane
The DSS intake was recorded daily, and rat body weight was also ®lter (pore size 0´45 mm) and injected into the HPLC system. The
monitored daily during the experimental period. histamine in these samples was separated with three reverse phase
columns in series (Cosmosil 5C18-AR; Nacalai Tesque Inc.,
Measurement of wet colonic weight and macroscopic evaluation Kyoto, Japan). The temperature was maintained at 408C through-
of colitis out this experiment. The histamine was then detected using a RF-
After 10 days of DSS administration, the rats were fasted overnight 535 ¯uorescence detector (Shimadzu, Kyoto, Japan) at excitation
and anaesthetized with an i.p. injection of pentobarbital sodium and emission wavelengths of 330 nm and 430 nm, respectively.
(40 mg/kg). After the rats were killed by cervical dislocation, the The HPLC mobile phase was a mixture of 50 mM sodium borate
body weight was measured and then a laparotomy was performed. buffer (pH 11) and acetonitrile (75:25, v/v), containing 2 mM
The large intestine from the anus to the caecocolonic junction was o-phthalaldehyde (OPA) and 2 mM N-acetyl-L-cysteine (NAC),
removed, irrigated with chilled saline, cut along the anti-mesen- which was shielded from exposure to sunlight and was delivered
teric border and then the wet weight was measured. To evaluate isocratically at a ¯ow rate of 0´5 ml/min.
any macroscopic changes, the mucosa was photographed and the
damaged mucosal area (colour change to purple and red, structural Statistical analysis
change with irregular folds, oedema and erosion) per total colonic The data were expressed as means 6 s.e.m. The variance was
surface was determined using an image analysis apparatus (NIH analysed by the F-test. Subsequently, Student's t-test for unpaired
image version 4.0/Macintosh). values was performed to compare the means of the normally
distributed data. Mann±Whitney U-test was also performed to
Histological examination compare the means of non-parametric or abnormally distributed
A specimen (10 ´ 10 mm) was removed at 1 cm distance from the data. P < 0´05 was considered statistically signi®cant.
anal margin, ®xed in Carnoy's ®xative for 3 h, and then embedded
in paraf®n. The histological samples were cut into 5 mm thick
RESULTS
sections, and were stained with haematoxylin and eosin (H±E)
after paraf®n removal. The mucosal damage score was determined Wet colonic weight and damaged area
according to a previously described method [14]. The following In the control (‡/‡) rats, diarrhoea occurred on days 2±3, and
three parameters were used: surface epithelial loss, crypt destruc- macroscopic bloody stools appeared on days 6±7. However, in the
tion and in¯ammatory cell in®ltration into the mucosa. A score Ws/Ws rats, the diarrhoea and bloody stools did not occur during
of 0±4 was then assigned to each parameter: 0 ˆ no change; the experimental period. None of the rats in either group died
1 ˆ localized and mild; 2 ˆ localized and moderate; 3 ˆ extensive during the experimental period. The DSS intake in the Ws/Ws rats
and moderate; and 4 ˆ extensive and severe. The sum of the scores was comparable to that of the control (‡/‡) rats (0´14 6 0´02 ml/day
from all three parameters was de®ned as the mucosal damage score per kg versus 0´12 6 0´004 ml/day per kg, respectively). Signi®cant
for each animal. In addition, toluidine blue staining (pH 2´5) was differences in body weight gains between Ws/Ws rats and control
also performed. For the immunohistochemical detection of MMC, (‡/‡) rats were observed during the experimental period (Table 1).
the sections were incubated with a polyclonal rabbit anti-rat mast The Ws/Ws rats remained healthy, and their body weight increased
cell protease II (RMCP II; Moredun, Edinburgh, UK) antibody, slightly. However, the body weight in the control (‡/‡) rats
followed by reaction with a biotin-conjugated goat anti-rabbit IgG decreased gradually. The wet colonic weight per unit body weight
and a peroxidase±streptavidin complex system (Dako, Glostrup, was signi®cantly lower in the Ws/Ws rats compared with control
Denmark), as described previously [15,16]. RMCP II is regarded as (‡/‡) rats, indicating that oedematous changes were prevented in
a speci®c marker of rat MMC [17]. the Ws/Ws rats (Table 1). Macroscopic examination of the colon
revealed that hyperaemia, erosions and occasional tiny blood
Mucosal myeloperoxidase activity and mucosal protein content coagula occurred mainly in the rectum in both control (‡/‡) and
Intestinal myeloperoxidase (MPO) activity, a biochemical maker Ws/Ws rats. However, this mucosal damage extended signi®cantly
of MPO‡ granulocytes, was measured using a standard assay as further towards the oral side in the control (‡/‡) rats than in the
described previously [18]. Brie¯y, a mucosal scraping from the Ws/Ws rats (Fig. 1).
remnant colon was placed in 1 ml of a hexadecyltrimethylammo-
nium bromide solution (0´5%, w/w). The solution was then homo- Histological ®ndings
genized and sonicated, and the resulting homogenate was subjected H±E staining revealed a marked in®ltration of in¯ammatory
to three rapid cycles of freezing and thawing. The samples were cells into the mucosa and submucosa in both control (‡/‡) and
centrifuged in a microfuge to remove any insoluble material Ws/Ws rats. Crypt loss and surface epithelial loss were also evident
q 2000 Blackwell Science Ltd, Clinical and Experimental Immunology, 119:264±269
266 Y. Araki et al.
Table 1. The body weight gain, colonic wet weight per unit body weight, macroscopic colonic damaged area per
entire colonic area and mucosal damage score

Group

Ws/Ws ‡/‡

Body weight gains (g) 16´7 6 4´9* 14´2 6 11´7

Colon wet weight per unit body weight (%) 0´52 6 0´03² 0´63 6 0´06
Macroscopic colonic damaged areas per entire colonic area (%) 55´3 6 1´31** 90´8 6 2´16
Mucosal damage score³ 5´00 6 1´31 8´17 6 0´70

Results are presented as means 6 s.e.m. from six rats in each group.
Signi®cantly different from the values for the ‡/‡ rats: *P < 0´05; **P < 0´0001 by Student's t-test; and
²P < 0´05 by Mann±Whitney U-test.
³ The mucosal damage score was determined according to the method of Oda [14].

(Fig. 2). The mucosal damage, as quanti®ed by the scoring system, some mast cells and many macrophages containing DSS particles
is summarized in Table 1. Mucosal damage scores in the Ws/Ws with metachromasia in the mucosa and submucosa (Fig. 3).
rats were lower than in control (‡/‡) rats, but these differences RMCP II-immunopositive cells were scattered in the mucosa of
were not statistically signi®cant (P ˆ 0´06, Table 1). The oedema- the control (‡/‡) rats. In the Ws/Ws rats, RMCP II-immunoposi-
tous changes in the mucosa and submucosa of control (‡/‡) rats tive cells were also noted, but the number was lower (Fig. 4). In
were more severe than those observed in the Ws/Ws rats (Fig. 1). In addition, sporadic globule leucocytes, which have been character-
the control (‡/‡) and Ws/Ws rats, toluidine blue staining revealed ized as mast cells in®ltrating into the epithelial layer, were also
observed in both group of rats [21].

Mucosal MPO activity, mucosal histamine level and mucosal

protein content
The intestinal mucosal MPO activities in the Ws/Ws rats were
comparable to those of control (‡/‡) rats (Table 2). On the other
hand, the mucosal histamine levels in the Ws/Ws rats were sig-
ni®cantly lower than those in control (‡/‡) rats (Table 2). The ratio
of the wet weight to the protein content was used to evaluate the
oedematous changes in the colonic tissue samples. A relative
decrease in mucosal oedema was observed in Ws/Ws rats compared
with control (‡/‡) rats, but there were no signi®cant differences
(P ˆ 0´09, Table 2). A positive correlation was observed between
histamine levels and degree of mucosal oedema in Ws/Ws rats (Fig.
5). A positive correlation was also found between mucosal histamine
levels and the area of macroscopic damage in the colon (Fig. 5).

Fig. 1. Dextran sulphate sodium (DSS)-induced colitis in Ws/Ws and

control (‡/‡) rats. The rats were treated with a 3% DSS solution for
DISCUSSION
10 days. (A) Ws/Ws rats. (B) Control rats. The mucosal damage in the Animal models of UC can be induced by the oral administration of
control rats extended further towards the oral side than the Ws/Ws rats. sulphated polysaccharides such as DSS. It has been reported that

Fig. 2. Microscopic ®ndings of dextran sulphate sodium (DSS)-induced colitis in the Ws/Ws and control (‡/‡) rat. Rectal specimens taken at
1 cm from the anal margin were preserved in Carnoy's ®xative, sectioned, and stained with H±E. (A) Ws/Ws rat, and (B) control (‡/‡) rat
(´ 100).
q 2000 Blackwell Science Ltd, Clinical and Experimental Immunology, 119:264±269
 DSS colitis in Ws/Ws rats 267

Fig. 3. Microscopic ®ndings of dextran sulphate sodium (DSS)-induced colitis in the Ws/Ws and control (‡/‡) rat. Toluidine blue staining
was performed as described in Materials and Methods. (A) Ws/Ws rat. (B) Control (‡/‡) rat. Some mast cells (open triangle) and
macrophages (arrow) containing ingested DSS particles, with metachromasia, were detected in the mucosa and submucosa (´ 400).

Fig. 4. Rat mast cell protease II (RMCP II)-immunopositive cells in the mucosa. (A) Ws/Ws rat. (B) Control (‡/‡) rat. The RMCP II-
immunopositive cells were scattered in the mucosa of Ws/Ws and (‡/‡) rats. Sporadic globule leucocytes, which are regarded as mast cells,
in®ltrating the epithelial layer, were also observed (arrow) (´ 200).

Table 2. Colonic mucosal myeloperoxidase (MPO) activity, histamine level and degree
of oedema

Group

Ws/Ws ‡/‡

MPO activity (U/mg protein) 0´276 6 0´033 0´322 6 0´050

Histamine level (mol/mg protein, ´ 10 8) 0´184 6 0´054* 0´343 6 0´028
Degree of oedema² 41´56 6 4´07 65´08 6 10´26

Results are presented as means 6 s.e.m. from six rats in each group.
² The ratio of wet weight to protein content was used to evaluate the degree of oedema
in the colonic tissue sample.
Signi®cantly different from the values for the ‡/‡ rats: *P < 0´05 by Student's t-test.

models of DSS-induced colitis exhibit some of the clinical and remain unclear. With respect to the pathogenic factors in the
histological features of UC patients. For example, several studies development of DSS-induced colitis, previous reports have postu-
have reported that similar to UC patients, DSS initially induced lated the importance of various factors such as local immunologi-
colonic lesions in the distal colon, which spread to the whole colon cal disturbance [25], the activation of mucosal macrophages [26],
[10,22±24]. In this model, the development of colitis is dependent effects related to the strong negative charge of DSS [27], the
on the molecular weight and sulphation of the DSS, in addition obliteration of the crypt lumina [28], and changes in the intestinal
to the dosage and the duration of administration [22]. However, micro¯ora [29]. However, the role of intestinal MMC in the
the precise mechanisms responsible for DSS-induced colitis development of DSS colitis has not been fully evaluated.
q 2000 Blackwell Science Ltd, Clinical and Experimental Immunology, 119:264±269
268 Y. Araki et al.
to Ws/Ws and control (‡/‡) rats for 10 days in accordance with the
procedures described in several previous reports [14,18,22,24]. In
addition, because DSS-induced colitis occurs at the distal colon in
both Ws/Ws and control (‡/‡) rats, we evaluated and compared the
most severe lesions at the rectum. Mast cells have been reported to
be closely associated with the development of tissue oedema and
hyperaemia. In this process, the histamine released by mast cells
plays an important role. In the present study, the development of
colitis was signi®cantly attenuated in the Ws/Ws rats. Moreover, a
positive correlation between the development of colitis and muco-
sal histamine levels was observed. This indicates that the MMC
contributed to the development of DSS-induced colitis at least via
histamine release, although mast cell activation is accompanied
by the release of other chemical mediators such as leukotrienes
and/or PAF [31]. These results are also compatible with the pre-
vious report that ketotifen, an inhibitor of the release of chemical
mediators from mast cells, prevented the development of DSS
colitis [14].
We previously reported that there were few mast cells in the
intact intestinal mucosa of Ws/Ws rats [32]. Interestingly, DSS
administration over 10 days induced the recruitment of MMC in
the colonic mucosa of Ws/Ws rats. This ®nding is consistent with
the previous study from Arizono and co-workers, who reported that
Nippostrongylus brasiliensis infection induces the appearance of
intestinal MMC in Ws/Ws rats [33]. Alizadeh and co-workers also
reported that a small number of MMC appear in the small intestine
of W/WV mice after infection with Trichinella spiralis [34]. Since
we detected a statistical correlation between the mucosal histamine
levels and the macroscopic damaged area, it is feasible that the
recruited MMC might have contributed to the process of mucosal
damage in the Ws/Ws rats. In another report, signi®cant increases
in IL-3 levels were reported in the mesenteric lymph nodes of mice
with DSS-induced colitis [35]. In general, IL-3 as well as the stem
cell factor c-kit ligand have been established as potent growth
factors for mast cells. These observations suggest that DSS may be
able to induce intestinal MMC recruitment through the stimulation
of IL-3 generation. On the other hand, in previous studies of mast
cell-de®cient animals, there have been no descriptions of MMC
recruitment in the colonic mucosa. Therefore, it is possible that the
discrepancy between the present study and previous studies of mast
cell-de®cient animals is due to differences in MMC recruitment
during the experimental period. It may be important to investigate
the participation of recruited MMC in the development of experi-
Fig. 5. Relationship between mucosal histamine levels and mucosal mental colitis in mast cell-de®cient animals.
damage markers. (a) Correlation between mucosal histamine levels and In conclusion, the development of DSS-induced colitis was
the degree of mucosal oedema (the ratio of the wet weight to the protein
attenuated in Ws/Ws rats. These results strongly suggest that mast
content in the colonic mucosa). The line represents the regression line
for data from both Ws/Ws and control rats (Y ˆ 0´038X 0´011, R ˆ 0´778,
cells play an important role in the development of DSS-induced
P < 0´01). (b) Correlation between mucosal histamine levels and the ratio colitis. Moreover, the sequential induction of MMC may affect
of the macroscopically damaged area to the entire colonic surface tissue in¯ammation in Ws/Ws rats.
(Y ˆ 0´9074X ‡ 0´4916, R ˆ 0´623, P < 0´05).

To de®ne the role of mast cells in the development of experi- REFERENCES

mental colitis, two studies have reported on mast cell-de®cient
mice. One study reported that mast cells were not involved in the 1 Boros M, Kaszaki J, Nagy S. Histamine release during intestinal
colitis induced by intrarectal administration of trinitrobenzene ischemia±reperfusion: role of irons and hydrogen peroxide. Circ Shock
1991; 35:174±80.
sulphonic acid (TNBS) in mast cell-de®cient W/WV mice [18].
2 Boros M, Kaszaki J, Nagy S. Oxygen free radical induced histamine
The other experiment revealed that oral DSS administration induced release during intestinal ischemia and reperfusion. Eur Surg Res 1989;
colitis in both mast cell-de®cient W/WV and control (‡/‡) mice in 21:297±304.
a similar fashion [30]. In this study, we attempted to elucidate the 3 Lehr HA, Guhlmann A, Nolte D et al. Leukotrienes as mediators in
potential role of mast cell activation in the pathogenesis of DSS- ischemia±reperfusion injury in a microcirculation model in the ham-
induced colitis in rats. To induce the colitis, we administered DSS ster. J Clin Invest 1991; 87:2036±41.
q 2000 Blackwell Science Ltd, Clinical and Experimental Immunology, 119:264±269
 DSS colitis in Ws/Ws rats 269
4 Kubes P, Ibbotson G, Russell JM et al. Role of platelet-activating factor 21 Murray M, Miller HRP, Jarrett WFH. The globule leukocyte and its
in ischemia/reperfusion induced leukocyte adherence. Am J Physiol derivation from the subepithelial mast cell. Lab Invest 1968; 19:222±
1990; 259:G300±5. 34.
5 McAuley RL, Somers SC. Mast cells in nonspeci®c ulcerative colitis. 22 Okayasu I, Hatakeyama S, Yamada M, Ohkusa T, Inagaki Y, Nakaya R.
Am J Dig Dis 1961; 6:233±6. A novel method in the induction of reliable experimental acute and
6 Dvorak AM, Monahan RA, Osage JE, Dickersin GR. Crohn's disease: chronic ulcerative colitis in mice. Gastroenterology 1990; 98:694±702.
transmission electron microscopic studies. Immunologic in¯ammatory 23 Axelsson LG, Midtvedt T, Bylund-Fellenius AC. The role of intestinal
response. Alteration of mast cells, basophils, eosinophils, and the bacteria, bacterial translocation and endotoxin in dextran sodium
microvasculature. Hum Pathol 1980; 11:606±19. sulphate-induced colitis in the mouse. Microb Ecol Health Dis 1996;
7 Hiatt RB, Katz L. Mast cells in in¯ammatory conditions of the 9:225±37.
gastrointestinal tract. Am J Gastroenterol 1962; 37:541±5. 24 Kanauchi O, Nakamura T, Agata K, Mitsuyama K, Iwanaga T. Effects
8 Baenkler HW, Lux GG, GuÈnthner R, KohlhaÈu¯ M, Matek W. Biopsy of germinated barley foodstuff on dextran sulfate sodium-induced
histamine in ulcerative colitis and Crohn's disease. Hepatogastroenter- colitis in rats. J Gastroenterol 1998; 33:179±88.
ology 1987; 34:289±90. 25 Kobayashi K, Asakura H, Hamada Y et al. T lymphocyte subpopula-
9 Fox CC, Moore WC, Lichtenstein LM. Modulation of mediator release tions and immunoglobulin-containing cells in the colonic mucosa of
from human intestinal mast cells by sulfasalazine and 5-amino-salicylic ulcerative colitis; a morphometric and immunohistochemical study.
acid. Dig Dis Sci 1991; 36:179±84. J Clin Lab Immunol 1988; 25:63±8.
10 Elson CO, Sartor RB, Tennyson GS, Riddell RH. Experimental models 26 Abraham R, Fabian RJ, Golberg L et al. Role of lysosomes in
of in¯ammatory bowel disease. Gastroenterology 1995; 109:1344±67. carrageenan-induced cecal ulceratation. Gastroenterology 1976; 67:
11 Niwa Y, Kasugai T, Ohno K et al. Anemia and mast cell depletion in 1169±81.
mutant rats that are homozygous atÃWhite Spotting (Ws)¨ locus. Blood 27 Watt J, Marcus R. Experimental ulcerative disease of the colon. Meth
1991; 78:1936±41. Achiev Exp Pathol 1975; 7:56±71.
12 Tsujimura T, Hirota T, Nomura S et al. Characterization of Ws mutant 28 Iwanaga T, Hoshi O, Han H et al. Morphological analysis of acute
allele of rats: a 12-base deletion in tyrosine kinase domain of c-kit gene. ulcerative colitis experimentally induced by dextran sulfate sodium
Blood 1991; 78:1942±6. in the guinea pig: some possible mechanisms of cecal ulceration.
13 Tei H, Kasugai T, Tsujimura T et al. Characterization of cultured mast J Gastroenterol 1994; 29:430±8.
cells derived from Ws/Ws mast cell-de®cient rats with a small deletion 29 Onderdonk AB, Hermos JA, Bartlett JG. The role of the intestinal
at tyrosine kinase domain of c-kit. Blood 1994; 83:916±25. micro¯ora in experimental colitis. Am J Clin Nutr 1977; 30:1819±25.
14 Oda T. Role of mast cells in dextran sulfate sodium-induced experi- 30 Minocha A, Thomas C, Omar R. Lack of crucial role of mast cells in
mental colitis in rats. J Kyoto Pref Univ Med 1995; 104:1069±82. pathogenesis of experimental colitis in mice. Dig Dis Sci 1995;
15 Kimura T, Andoh A, Fujiyama Y et al. A blockade of complement 40:1757±62.
activation prevents rapid intestinal ischaemia±reperfusion injury by 31 Wasserman SI. Mediators of immediate hypersensitivity. J Allergy Clin
modulating mucosal mast cell degranulation in rats. Clin Exp Immunol Immunol 1983; 72:101±15.
1998; 111:484±90. 32 Andoh A, Kimura T, Fukuda M, Araki Y, Fujiyama Y, Bamba T. Rapid
16 Kanwer S, Kubes P. Mast cells contribute to ischemia±reperfusion- intestinal ischaemia±reperfusion injury is suppressed in genetically
induced granulocyte in®ltration and intestinal dysfunction. Am J mast cell-de®cient Ws/Ws rats. Clin Exp Immunol 1999; 116:90±3.
Physiol 1994; 267:G316±21. 33 Arizono N, Kasugai T, Yamada M et al. Infection of Nippostrongylus
17 Kido H, Fukusen N, Katsunuma N. Chymotrypsin and trypsin type brasiliensis induces development of mucosal-type but not connective
serine proteases in rat mast cells: properties and functions. Arch tissue-type mast cells in genetically mast cell-de®cient Ws/Ws rats.
Biochem Biophys 1985; 239:436±43. Blood 1993; 81:2572±8.
18 Chin KW, Barrett KE. Mast cells are not essential to in¯ammation in 34 Alizadeh H, Murrell KD. The intestinal mast cell response to Trichi-
murine model of colitis. Dig Dis Sci 1994; 39:513±25. nella spiralis infection in mast cell-de®cient W/Wv mice. J Parasitol
19 Lowry OH, Rosenbrough NJ, Farr AL, Udenfriend S. Protain measure- 1984; 70:767±73.
ment with the Folin phenol reagent. J Biol Chem 1951; 193:265±76. 35 Dieleman LA, Ridwan BU, Tennyson GS et al. Dextran sulfate sodium-
20 Saito K, Horie M, Nose N et al. Determination of polyamines in foods induced colitis occurs in severe combined immunode®cient mice.
by liquid chromatography with on-column ¯uorescence derivatization. Gastroenterology 1994; 107:1643±52.
Anal Sci 1992; 8:675±80.

q 2000 Blackwell Science Ltd, Clinical and Experimental Immunology, 119:264±269