Báo cáo khoa học: "Mucosal mast cell-derived chondroitin sulphate levels in and worm expulsion from FcRγ-knockout mice following oral challenge with Strongyloides venezuelensis" doc

-2851$/ 2)
9H W H U L Q D U \ 
6FLHQFH
J. Vet. Sci.
(2004),
/
5
(3), 221–226
Mucosal mast cell-derived chondroitin sulphate levels in and worm
expulsion from FcR
γ
-knockout mice following oral challenge with
Strongyloides venezuelensis
Denis Nnabuike Onah
1,
*, Yukifumi Nawa
2
1
Department of Veterinary Parasitology & Entomology, University of Nigeria, Nsukka, Nigeria
2
Department of Parasitology, Miyazaki Medical College, Kiyotake, Miyazaki 889-1692, Japan
Mucosal mast cell-derived chondroitin sulphates
(sulphated proteoglycans) were assayed in gut washings
and homogenate of FcR
γ
-knockout (KO) and wild-type
(WT) C57BL/6 mice challenged with
Strongyloides
venezuelensis
in order to assess their possible role in
secondary immunity against enteric nematodes. Groups

of immune KO and WT mice were challenged by oral
gavage with 300 infective larvae (L
3
). Establishment of
infection was assessed by daily faecal analysis to
determine the number of eggs per gram of faeces (EPG)
and by adult worm recovery on days 5 and 13 post
challenge. Mucosal mast cell (MMC) counts were done on
days 5 and 13 post challenge while MMC-derived
chondroitin sulphates in gut washings (days 1 and 5) and
homogenate (day 8) were assayed by high performance
liquid chromatography (HPLC). Results showed that
patent infection occurred in challenged KO but not WT
mice despite significantly higher mastocytosis in jejunal
sections of KO than WT mice (
p
<0.001). Similarly but
against prediction, significantly higher concentration of
MMC-derived chondroitin sulphates was observed in gut
homogenate of KO than WT mice (
p
< 0.05). In contrast,
significantly higher concentration of chondroitin
sulphates was observed in gut washings of WT than KO
mice (
p
< 0.05). These results suggest that MMC in KO
mice failed to release sufficient amount of sulphated
proteoglycans into the gut lumen as did the WT mice,
which may have been part of the hostile environment that

prevented the establishment in and eventual expulsion of
adult
S. venezuelensis
from the gut of WT mice following
challenge.
Key words:

Strongyloides venezuelensis
, mast cells, mucosal
immunity, chondroitin sulphates,

gut washings,

mice
Introduction
The immune expulsion of adult
Strongyloides
species
from the small intestine of mice and rats is associated with
intraepithelial mast cell hyperplasia [1,2,7,14]. Mast cells
are thought to mediate this expulsion by creating an
environment hostile to the establishment and survival of
gastrointestinal nematodes through inflammation-associated
changes and release of inflammatory mediators [19]. The
release of these mediators is induced by mast cell
degranulation, which in turn is induced by mast cell
activation triggered by cross-linking of the high affinity
immunoglobulin Fc receptor (FcR) for IgE (Fc
ε
RI) with an

antigen-IgE immune complex [3,8,11,13]. This type of
acute inflammation, also known as type I hypersensitivity is
therefore thought to depend on mast cells, its Fc
ε
RI and on
IgE. Perturbation in any one of these three components
should result in the loss of type I hypersensitivity responses
[18]. This, in fact, is the case with Fc
ε
RI. To be expressed on
the surface of cells and for signal transduction into the
interior of the cell, Fc
ε
RI requires the homodimeric
γ
subunit of FcRs (FcR
γ
) [5,10,17]. Ablation of the
γ
subunit
by targeted gene deletion results in the loss of Fc
γ
RI
expression on mast cells coupled with loss of mast cell
functions including degranulation and mediator release in
the mutant mouse [23]. Using these mice we showed that
while the wild type (WT) counterparts were able to expel a
primary
Strongyloides venezuelensis
infection, FcR

γ
-
knockout (KO) mice failed to do so [16]. However, the
confounding aspect of our study was that both intestinal
mastocytosis and serum mouse mast cell protease-I
(mMCP-I), the levels of which have negative correlation
with nematode burden in sheep [22] were similar. Since
FcR
γ
-KO mice fail to assemble Fc
γ
RI on their mast cells
and therefore are unable to express mast cell degranulation
and mediator release [23], we concluded that mMCP-1 may
be released spontaneously without requiring the mast cell
Fc
γ
RI cross-linking with the immune complex formed by
the parasite antigen/IgE and that mMCP-1 is not involved in
*Corresponding author
Tel: +234-42-770106
E-mail:
222 Denis Nnabuike Onah, Yukifumi Nawa
adult
S. venezuelensis
expulsion [16]. However, because
mastocytosis has a strong association and often corresponds
with the time of worm expulsion [14], we stated that mast
cells must be involved in the process in a manner yet to be
elucidated. We then speculated that failure of FcR

γ
-KO mice
to expel
S. venezuelensis
might be related to inability of their
mast cells to degranulate and release inflammatory
mediators other than mMCP-1 [16]. Sulphated goblet cell
mucins [7] and experimental introduction of glycosaminoglycans
of the type produced by mucosal mast cells [12] have been
shown to prevent the establishment of
S. venezuelensis
infection and mediate its expulsion from infected mice.
Chondroitin sulphates are the major proteoglycans
contained by mucosal mast cells in mouse [25]. In this study
therefore, we assayed the amounts of chondroitin sulphates
in gut homogenate and washings of FcR
γ
-KO mice and their
WT counterparts following oral challenge with
S.
venezuelensis
. This was in order to ascertain (i) if
differences exit in their concentrations in the gut of KO and
WT mice and (ii) whether differences in worm
establishment and expulsion following challenge might be
related to any differences in their concentrations within the
gut niche occupied by the parasite in the two mouse types.
Materials and Methods
Animals
FcR

γ
-knockout C57BL/6 male mice of 8 weeks old were
generous gifts from Professor Toshiyuki Takai and Dr.
Masao Ono (Tohoku University, Sendai, Japan). Age-
matched specific pathogen free wild type C57BL/6 mice of
the same sex were purchased from Japan SLC (Shizuoka,
Japan). All animals were kept in our laboratory animal unit
for 2 weeks to acclimatize before use at 10 weeks old.
Before and during the period of the experiment they were
supplied with feed and water ad libitum.
Parasite
The strain of
S. venezuelensis
originally isolated from a
brown rat in Okinawa Prefecture, Japan [6] and established
as a laboratory strain [20] was used in this study. Infective
third stage larvae (L
3
) used for infecting experimental
animals orally (oral gavage) were obtained from the lungs of
mice given primary infection and sacrificed on day 3 post
infection. Briefly, 10 C57BL/6 male mice were each
infected with 5000 L
3
of
S. venezuelensis
obtained by the
filter-paper faecal culture method and sacrificed on day 3
post-infection by anaesthetic overdose using diethyl ether.
The lungs were removed by dissection, shredded with fine-

tipped thumb forceps in fine-meshed coffee strainers placed
in a beaker of warm PBS and incubated at 37
o
C for 3 h.
Emerged lung L
3
were recovered by centrifugation, re-
suspended and washed three time in fresh warm (37
o
C) PBS,
counted and adjusted to 1500 lung L
3
/ml.
Infection of experimental animals
All animal groups for challenge were primed with 2000 L
3
subcutaneously 30 days before challenge. For uniform
treatment, every mouse in both the challenge and challenge
control (primary infection) groups was treated orally with
20-mg/kg mebendazole (Sigma, St. Louis, MO) on day 26
post priming to get rid of any residual adult worms from the
priming dose. On day 30 post priming, each animal in the
challenge and control groups was infected with 300 lung-
recovered L
3
contained in 200
µ
l PBS and introduced
directly into the stomach using a stomach needle with blunt
oval tip.

Experimental protocol
Thirty knockout (KO) and thirty wild-type (WT) mice
were used for the experiment. Twenty each of the KO and
WT mice were primed as stated above. For challenge,
primed KO and WT mice with 10 each of their naïve
controls were infected as described above. The challenge
groups were sacrificed 5 mice each on day 1, 5, 8 and 13
post challenge for sample collection while 5 each of their
controls were sacrificed on day 5 and 13 post infection.
Daily faecal egg counts expressed as eggs per gram of faeces
(EPG) and adult worm recovery to establish whether patent
infection occurred were carried out until day 13, and on day
5 and 13 post infection respectively in both challenge and
control animals. In the challenge animals only, histology for
MMC numbers was carried out on day 5 and 13, while
various chondroitin sulphates and mouse mast cell protease-
1 (mMCP-1) in gut washings were assayed on day 1 and 5
post challenge. Chondroitin sulphates in gut homogenate
were assayed on day 8 post challenge.
EPG and adult worm recovery
As stated above, faecal samples for daily EPG was
collected only from individual animal of the groups
sacrificed on day 13 post challenge. To ascertain worm
establishment, adult worms were recovered from sacrificed
challenge and control animals on day 5 and 13 post
infection. The entire small intestine of each sacrificed
animal was isolated and processed for adult worm recovery.
The methods used for EPG and adult worm recovery, were
as previously described [9,20].
Histology and serology

Jejunal pieces were taken and histological sections
prepared, stained and mucosal mast cells were enumerated
as previously described [16] on day 5 and 13 post challenge.
Also samples of gut washings as described below were
collected on days 1 and 5 post challenge and analysed for
mMCP-1 concentration by ELISA as described [16].
Proteoglycan assay in gut washings and homogenate
The entire small intestine isolated from each of the
Chondroitin sulphates and worm expulsion in FcR
γ
-knockout mice 223
sacrificed animal on the days specified above was washed
out twice with a total volume of 10 ml sterile distilled water.
Each sample was then centrifuged at 350 g for 10 min to
remove debris. The supernatant was frozen until used. For
gut homogenate, approximately 15 cm of the upper small
intestine was cut off, washed out as above and minced using
a homogeniser (Polytron homogeniser, Kinematika AG,
Littau, Switzerland). The minced samples were suspended
in 5 ml sterile PBS, centrifuged as above and the supernatant
was frozen until used. For the assay of the various
chondroitin sulphates, each sample was removed from the
freezer, thawed under room temperature and was then
processed and analysed by high-performance liquid
chromatography according to the methods of Yoshida
et al.
[24] and Shinmei
et al.
[21].
Statistical analysis

Data were compiled and subjected to descriptive statistics
while differences between group-means were obtained by
the Students
t
-test using Microsoft Excel Statistical
Toolpack. Differences at
p
= 0.05 were considered significant.
Results
Eggs per gram of faeces (EPG)
The mean daily EPG of animals sacrificed on day 13 post
challenge are presented in Fig. 1. All control WT and KO
animals as well as all challenged KO mice persistently shed
eggs in their faeces until the day of sacrifice. In contrast,
there were no eggs in the faeces of all challenged WT mice.
One significant observation however, is that the oral route of
infection with lung-recovered L
3
does not seem to be an
efficient means of establishing patent infection judging from
the level of EPG of the animals (less than 1000 at peak EPG)
when compared with the EPG of mice infected by the
subcutaneous route in which peak EPG usually runs in tens
of thousands [15,16].
Adult worm recovery
Consistent with the EPG result, control WT and KO as
well as challenged KO mice developed persistent patent
S.
venezuelensis
infection. The mean number of adult worms

recovered from animals sacrificed on days 5 and 13 are
shown in Fig. 2. As expected, no adult worms were
recovered from any of the challenged WT mice on both
days, which agreed with the zero EPG recorded for this
group. Again, the very few number of adult worms
recovered from these mice indicate that oral implantation of
third stage larvae is not very efficient in establishing patent
infection.
Intestinal mast cell numbers and serum mMCP-1
concentration
Mast cells were enumerated in jejunal sections prepared
on days 5 and 13 post oral challenge. Results (Table 1a)
show that significantly higher numbers of mast cells were
counted in KO than WT mice on day 5 (
p
< 0.01) and day
13 (
p
< 0.001). On the other hand, there was no significant
difference (
p
> 0.05) in the amount of mMCP-1 concentration
in gut washings of challenged KO and WT mice on days 1
and 5 post challenge (Table 1b).
Sulphated proteoglycan concentration in gut washings
and homogenate
Results of the assay for various sulphated proteoglycans
(Chondroitin sulphates A, C, D, E, and total chondroitin
sulphate) in gut washings of challenged KO and WT mice
sacrificed on days 1 and 5 are presented in Figs. 3a and 3b

respectively. The results show that significantly higher
concentrations of the chondroitin sulphates occurred in the
gut washings of WT than KO mice on these days (
p
< 0.05).
In contrast, significantly higher concentrations of these
Table 1.
(a) Mucosal mast cell number/10 villus crypt unit (Mean ± SD) in FcR
γ
-KO and WT mice challenged orally with 300 lung-
recovered
S. venezuelensis
L
3
Day PC Animal MMC/10VCU ± SD
p
Value
5 KO 320 ± 51
WT 156 ± 22 < 0.01
13 KO 456 ± 91
WT 057 ± 10 < 0.001
(b) Mean ± SD of mMCP-1 level (ng/ml) in gut washings of FcR
γ
-KO and WT mice challenged orally with 300 L
3
of
S.
venezuelensis

mMCP-1 Concentration (ng/ml)

Day 1 Day 5
KO 91 ± 36 205 ± 82
WT 85 ± 28 191 ± 56
p
Value > 0.05 > 0.05
224 Denis Nnabuike Onah, Yukifumi Nawa
sulphated sugars were obtained in gut homogenate
preparations of KO than WT mice on day 8 post-challenge
(
p
< 0.05, Fig. 3c).
Discussion
Primary infection of mice with
S. ratti
and
S.
venezuelensis
results in the development of a strong
immunity, which completely aborts a patent infection
following secondary challenge of the animals with the
parasites [4,20]. This strong immunity depends on CD4
+
T
cells which make Th
2
cytokines and induce mast cell-
dependent gut inflammatory responses and changes in gut
physiology, all of which act in concert to create an
environment hostile to worm establishment in and their
eventual expulsion from the intestinal niche [15]. Our

prediction was that an important part of this inflammatory
response and change in gut physiology resulting in worm
expulsion is the presence of sulphated proteoglycans
released by degranulating mast cells. Therefore, if FcR
γ
-KO
mice are unable to undergo mast cell degranulation and
mediator release [23], the enabling hostile environment for
the ablation of secondary infections will be lacking in
primed KO animals. Consequently, if sulphated proteoglycans
play a role in worm expulsion, then the introduction of
migrating L
3
into a hostile intestinal environment in primed
KO and WT mice would most likely result in a patent
infection in the former but not in the latter. Results of this
study largely support our predictions. First, significantly
higher concentrations of chondroitin sulphates were present
in gut washings of WT, in which the challenge infection was
sterile than in KO, in which persistent patent infection was
established. Secondly, there was no significant difference in
the concentrations of mMCP-1 in gut washings taken from
both mouse types irrespective of the fact that significantly
higher numbers of intraepithelial mast cells were counted in
KO than in WT animals.
Taken together, it can be inferred that (a) mMCP-1 may be
released by mast cells in a manner not dependent on
degranulation i.e., not requiring the antigen-IgE-Fc
ε
R cross-

linking and triggering, (b) mMCP-1 may not be important in
the mucosal immune mechanisms resulting in the expulsion
of adult
S. venezuelensis
from the gut and (c) mucosal mast
cell-derived glycosaminoglycans (chondroitin sulphates) are
apparently involved in and therefore play a role in the
prevention of the establishment of adult
S. venezuelensis
in
and their eventual immune expulsion from the gut. These
agree with the suggestions of Onah
et al.
[16] regarding
mMCP-1 and its possible role in worm expulsion as well as
with their speculation that FcR
γ
-KO mice are probably
unable to expel primary
S. venezuelensis
infection as a result
of failure of their mast cells to degranulate and release
mediators other than mMCP-1. In fact, that significantly
higher concentrations of chondroitin sulphates were
obtained from the homogenized gut tissue of KO (which
contained more mast cells) than WT is an added support that
the intraepithelial mast cells in KO contain them but are not
releasing them in enough quantities into the gut lumen to
effect worm expulsion. In addition, we have similar
evidence that when immune KO and WT mice are

challenged by subcutaneous introduction of 3000
S.
venezuelensis
L
3
, fewer larvae are recovered from the lungs
of KO than in WT animals 3 days later yet patent infection
F
ig. 1. Daily numbers (mean ± SD) of eggs per gram of faec
es
(
EPG) in immune and control FcRγ-KO and WT mi
ce
c
hallenged orally with 300 lung-recovered L
3
of
S. venezuelens
is
a
nd sacrificed on day 13 post challenge.
F
ig. 2. Number of adult worms (mean ± SD) recovered from t
he
s
mall intestines of immune and control FcRγ-KO and WT mi
ce
c
hallenged orally with 300 lung-recovered L
3

of
S. venezuelens
is
a
nd sacrificed on days 5 and 13 post challenge.
Chondroitin sulphates and worm expulsion in FcR
γ
-knockout mice 225
occurs only in the latter animals. In this infection protocol
serum mMCP-1 was also similar but sulphated
proteoglycans were found to be significantly higher in the
WT than KO animals. Moreover, our results in this study
agree with and support those of Maruyama and his
associates [12] who showed that glycosaminoglycans of the
type produced by mucosal mast cells significantly inhibited
the invasion and establishment of adult
S. venezuelensis
implanted into the duodenum of naïve mice.
In conclusion, defective secondary immunity against
S.
venezuelensis
in FcR
γ
-KO mice is associated with
significant decrease in the amount of mast cell-derived
chondroitin sulphates released into the gut lumen,
suggesting that the sugars are intimately associated with
worm expulsion. It is our opinion that extensive further
studies into the role of sulphated sugars of the type produced
by mucosal mast cells in parasitic gastrointestinal nematode

expulsion is essential and worthy of support as they seem
attractive candidates for anthelmintic drug investigation and
development.
Acknowledgments
We thank Mrs. Eri Ono for excellent technical assistance
F
ig. 3. (A) Chondroitin sulphate (ChS) concentration (
µ
g/ml) (mean ± SD) in gut washings of FcR
γ
-KO and WT mice challenged oral
ly
w
ith 300 lung-recovered L
3
of
S. venezuelensis
and sacrificed on day 1 post challenge. ChS-A:E = Chondroitin sulphate A-E; U
s-
C
h = Unsulphated chondroitin. (B) Chondroitin sulphate (ChS) concentration (
µ
g/ml) (mean ± SD) in gut washings of FcR
γ
-KO a
nd
W
T mice challenged orally with 300 lung-recovered L
3
of

S. venezuelensis
and sacrificed on day 5 post challenge. Ch
S-
A
:E = Chondroitin sulphate A-E; Us-Ch = Unsulphated chondroitin. (C) Chondroitin sulphate (ChS) concentrations (
µ
g/m
l)
(
mean ± SD) in gut homogenates of FcR
γ
-KO and WT mice challenged orally with 300 lung-recovered L
3
of
S. venezuelensis
a
nd
s
acrificed on day 8 post challenge. ChS-A:E = Chondroitin sulphate A-E; Us-Ch = Unsulphated chondroitin.
226 Denis Nnabuike Onah, Yukifumi Nawa
and Professor Toshiyuki Takai and Dr. Masao Ono (Tohoku
University, Sendai, Japan) for generously providing the
mutant mice used in this study. DNO was a JSPS
postdoctoral research fellow and funding for this work was
provided by the Grant-in-Aid for scientific research from the
Ministry of Education, Culture, Sports, Science and
Technology, Japan. We also thank the University of Nigeria,
Nsukka for granting DNO the study leave which enabled
him undertake the fellowship in Japan.
References

1. Abe T, Nawa Y. Reconstitution of mucosal mast cells in W/
W
v
mice by adoptive transfer of bone marrow-derived
cultured mast cells and its effects on the protective capacity
to
Strongyloides ratti
infection. Parasite Immunol 1987, 9,
31-38.
2. Abe T, Nawa Y. Localisation of mucosal mast cells in W/W
v
mice after reconstitution with bone marrow cells or cultured
mast cells, and its relation to the protective capacity to
Strongyloides ratti
infection. Parasite Immunol 1987, 9, 477-
485.
3. Beaven MA, Metzger H. Signal transduction by Fc
receptors: the Fc
ε
RI case. Immunol Today 1993, 14, 222-
226.
4. Dawkins HJS, Muir GM, Grove DI. Histopathological
appearances in primary and secondary infections with
Strongyloides ratti
in mice. Int J Parasitol 1981, 11, 97-103.
5. Ernst LK, Duchemin AM, Anderson CL. Association of
high-affinity receptor for IgG (Fc
γ
RI) with the
γ

subunit of
the IgE receptor. Proc Natl Acad Sci USA 1993, 90, 6023-
6027.
6. Hasegawa H, Orido Y, Sato Y, Otsuru M.
Strongyloides
venezuelensis
Brumpt, 1934 (Nematoda: Strongyloididae)
collected from
Rattus norvegicus
in Naha, Okinawa, Japan.
Jap J Parasitol 1988, 37, 429-434.
7. Ishiwata K, Uchiyama F, Maruyama H, Kobayashi T,
Kurokawa M, Nawa Y. Glycoconjugates and host-parasite
relationship in the mucosal defense against intestinal
nematodes. In: Ogra PL, Mestecky J, Lamm ME, Strober

W,
Bienenstock J, McGhee JR (eds.). Mucosal Immunology. pp.
925-933. Academic Press, San Diego, 1999.
8. Ishizaki T, Ishizaki K. Biology of immunoglobulin E. Prog
Allergy 1975, 19, 61-121.
9. Korenaga M, Nawa Y, Mimori T, Tada I. Effects of
preintestinal larval antigenic stimuli on the generation of
intestinal immunity in
Strongyloides ratti
infection. J
Parasitol 1983, 69, 78-82.
10. Kurosaki T, Ravetch JV. A single amino acid in the glycosyl
phosphatidylinositol attachment domain determines the
membrane topology of Fc

γ
RIII. Nature 1989, 342, 805-807.
11. Lorentz A, Schwengberg S, Sellge G, Manns MP, Bischoff
SC. Human intestinal mast cells are capable of producing
different cytokine profiles: role of IgE receptor cross-linking
and IL-4. J Immunol 2000, 164, 43-48.
12. Maruyama H, Yabu Y, Yoshida A, Nawa Y, Ohta N. A role
of mast cell glycosaminoglycans for the immunological
expulsion of intestinal nematode,
Strongyloides venezuelensis.
J Immunol 2000, 164, 3749-3754.
13. Metzger H. The receptor with high affinity for IgE. Immunol
Rev 1992, 125, 37-48.
14. Nawa Y, Ishikawa N, Tsuchiya K, Horii Y, Abe T, Khan
AI, Shi B, Itoh H, Ide H, Uchiyama F. Selective effector
mechanisms for the expulsion of intestinal helminths.
Parasite Immunol 1994, 16, 333-338.
15. Onah DN, Nawa Y. Mucosal Immunity against parasitic
gastrointestinal nematodes. Korean J Parasitol 2000, 38, 209-
236.
16. Onah DN, Uchiyama F, Nagakui Y, Ono M, Toshiyuki T,
Nawa Y. Mucosal defense against gastrointestinal
nematodes: responses of mucosal mast cells and mouse mast
cell protease 1 during primary
Strongyloides venezuelensis
infection in FcR
γ
-knockout mice. Infec Immun 2000, 68,
4968-4971.
17. Ra C, Jouvin MH, Blank U, Kinet JP. A macrophage Fcã

receptor and the mast cell receptor for IgE share an identical
subunit. Nature 1989, 431, 752-754.
18. Ravetch JV. Fc receptors: rubor redox. Cell 1994, 78, 553-
560.
19. Rothwell TLW. Immune expulsion of parasitic nematodes
from the alimentary tract. Int. J. Parasitol. 1989, 19, 139-168.
20. Sato Y, Toma H.
Strongyloides venezuelensis
infection in
mice. Int J Parasitol 1990, 20, 57-62.
21. Shinmei M, Miyauchi S, Machida A, Miyazaki K.
Quantitation of chondroitin 4-sulfate and chondroitin 6-
sulfate in pathologic joint fluid. Arthritis Rheumat 1992,
35,1304-1308.
22. Stevenson LM, Huntley JF, Smith WD, Jones DG. Local
eosinophil- and mast cell-related responses in abomasal
nematode infections of lambs. FEMS Immunol Med
Micrbiol 1994, 8, 167-173.
23. Takai T, Li M, Sylvestre D, Clynes R, Ravetch JV. FcR
γ
chain deletion results in pleiotropic effector cell defects. Cell
1994, 76, 519-529.
24. Yoshida K, Miyauchi S, Kikuchi H, Tawada A, Tokuyasu
K. Analysis of unsaturated disaccharides from glycosamino-
glycuronan by high-performance liquid chromatography.
Analyt Biochem 1989, 177, 327-332.
25. Yoshikawa T, Gon Y, Matsui M, Yodoi J. IgE-mediated
allergic responses in the mucosal immune system. In: Ogra
PL, Mestecky J, Lamm ME, Strober

W, Bienenstock J,
McGhee JR (eds.). Mucosal Immunology. pp. 925-933.
Academic Press, San Diego, 1999.