Enhancement of Human Intestinal
Mast Cell Mediator Release in Active
Ulcerative Colitis
CHARITY C. FOX, AUDREY J. LAZENBY, WENDY
JOHN H. YARDLEY, THEODORE M. BAYLESS,
and LAWRENCE M. LICHTENSTEIN
The Johns Hopkins Asthma and Allergy Center, Division of Gastroenterology of the Department
of Medicine, and Department of Pathology, Johns Hopkins University School of Medicine,
Baltimore, Maryland
To further define the role of mast cells in the
idiopathic inflammatory bowel diseases, mediator
release from intestinal mast cells derived from ac-
tively inflamed and relatively quiescent areas of
ulcerative colitis was studied. It was hypothesized
that mast cells in the actively diseased segments
would indicate involvement in the disease process by
releasing a different profile of mediators than cells in
uninflamed tissue. Mast cell-containing suspensions
derived from matched segments of 12 ulcerative
colitis specimens were compared for responsiveness
to the mast cell stimulus goat anti-human immuno-
globulin E. Supernatants from challenged cells were
analyzed for levels of three mast cell mediators,
histamine, prostaglandin D,, and the sulfidopeptide
leukotriene C. Mast cells from the actively involved
areas released significantly greater amounts of hista-
mine, prostaglandin D,, and sulfidopeptide leuko-
triene. The difference in histamine release was not a
result of greater stores of histamine in the active
tissue cells, because the total histamine content of
the mast cells from the active areas was not signifi-
cantly greater. The enhanced release of both pre-
formed and newly generated mediators indicates
activation of those cells in the course of the disease
and points to the mast cell contribution to the
inflammatory process in these disorders.
It
has long been recognized that the intestinal mast
cell performs a crucial role in gastrointestinal al-
lergy and host defense against parasitic infection (1,Z).
Consequently, most research in mast cell function in
the gastrointestinal tract has focused on these immuno-
globulin E (IgE)-mediated responses of the mast cell.
However, recent research in mast cell function in
C. MOORE,
other organ systems has shown that mast cells may
respond to many other stimuli, such as substance P (31,
opiates (a), bradykinin analogs (51, and the newly
described histamine-releasing factors (6).
It has also been noted that there is a significant
increase of mucosal and submucosal mast cells in the
idiopathic inflammatory bowel diseases (IBD), ulcer-
ative colitis (UC), and Crohn’s disease (CD) (7,8,9). On
the basis of these observations, previous investigators
have suggested that the mast cell plays a critical role in
IBD, especially since degranulating mast cells have
been observed in the active areas of disease (10). Not
surprisingly, this has led to the obvious hypothesis of
food allergy as an etiology of IBD; however, our recent
understanding of other mast cell stimuli suggests that
other mechanisms are more likely.
The mast cell produces an impressive array of
inflammatory mediators, including histamine, the
arachidonic acid metabolites, and neutral proteases,
which are known to increase vascular permeability
and smooth muscle contraction (11). The effect of these
mediators could contribute to the symptoms of diar-
rhea, cramping, and mucosal edema seen in IBD.
Until recently there was no in vitro system in which
to study the mediator production of the intestinal
mast cell from inflamed tissue. We have developed
Abbreviations used in this paper: CD, Crohn’s disease; EDTA,
edetic add; IBD, idiopathic inflammatory bowel disease; IgE,
immunoglobulln Is; LTC, leukotrlene C; CMF, caldum and magne-
sium free: HRSS, Hank’s balanced salt solution; PAG, PIPE!%
Albumln-Glucoee buffer; PAGCM, PAG buffer with calcium and
magnesium; PIPES, plperazlne-N,N-bls (Zsthanesulfonlc add)
PGD, prostaglandin D,; RIA, radioimmunoassay; UC, ulcerative
colitis.
0 1990 by the American Gastroenterological Association
OOM-5085/90/$3.00
120 FOX ET AL.
a method of producing mast cell-containing suspen-
sions of human intestinal mucosa from surgical
specimens (12). With these preparations, we examined
the production of histamine, prostaglandin D, (PGD,),
and leukotriene C (LTC) by mast cells derived from
the grossly abnormal and relatively quiescent areas of
surgical UC specimens.
Materials and Methods
Patient Population
Twelve patients were admitted for surgical resection
for UC refractory to medical treatment. They included 5
women and 7 men ranging in age from 25-45 years. Tissue
was obtained through a waste tissue protocol approved by
the Joint Committee on Clinical Investigation.
Pathological Correlation
To correlate histopathology with the isolated mast
cell findings, representative histological slides were selected
for review. The histological slides were chosen from sites
adjacent to areas from which the mast cells were harvested.
If the disease process was uniform, one slide was chosen
from each mucosal area. However, if the disease process
was variable in intensity, two slides were chosen from the
“diseased” areas and then averaged to better reflect the total
histological picture. A total of 28 slides from the 12 cases
were thus chosen. This histological material was part of the
usual surgical pathology specimen processing: it had been
fixed in formalin, embedded in paraffin wax, and stained
with H&E. Because the tissue was fixed in formalin, which
does not preserve intestinal mucosal mast cell granules; it
was not possible to stain and count the mast cells in the
histological material.
The selected slides were then coded and examined in a
blinded fashion for six active and chronic disease parame-
ters. Each slide was evaluated for percent of total surface
ulcerated or eroded, crypt distortion (none or rare, mild,
moderate, and severe), and amount of mononuclear cells in
the lamina propria (normal or mild, moderate, and severe
increase]. Morphometric counts of crypt neutrophils, as an
indicator of acute inflammation, were performed by count-
ing numbers of crypts with neutrophils (cryptitis or crypt
abcesses] per multiple 10x fields; they were then averaged.
Morphometric counts of eosinophils were done by counting
those cells per multiple 40xfields and then averaging.
Isolation of Human Intestinal Mucosal Mast
Cells
Human colonic mucosa was obtained from surgical
resections for complications of active and chronic UC. From
each of 12 biopsy-proven cases of UC, 10 cm2 of grossly
abnormal diseased mucosa and relatively inactive-appear-
ing mucosa were obtained. The specimens were processed
in parallel. The mast cells were obtained from the mucosa
by a technique of mechanical and enzymatic dispersion, as
previously described (12).
GASTROENTEROLOGY Vol. 99. No. 1
Briefly, after dissection from the underlying muscle lay-
ers, the mucosa was cut into 1 x l-cm pieces and washed to
remove adherent mucus with calcium- and magnesium-free
Hank’s balanced salt solution (CMF-HBSS) with 25 mmol/L
HEPES (CMF-HBSS/HEPES] and 1 mmol/L dithiothreitol.
The pieces were incubated for 60 minutes at 37’C in
CMF-HBSWHEPES with 0.75 mmol/L edetic acid (EDTA)
to remove the epithelial cell layer. The remaining tissue was
further cut into 2 x 2-mm pieces and incubated for 60
minutes at 37’C with stirring in 100 mL HBSS/HEPES with
30 units collagenase (Gibco, Grand Island, NY] and 20%
heat-inactivated fetal calf serum. At the end of incubation,
the suspension was filtered through lOO+m mesh Nitex
cloth (Tetko, Elmsford, NY] and saved. The tissue was
subjected to a repeat collagenase digestion, and the filtrates
from both collagenase steps were combined.
The remaining tissue was further dispersed by incubation
with pronase (1.5 mg/g wet weight tissue] (Calbiochem, San
Diego, CA] and chymopapain (0.5 mg/g wet weight tissue]
(Sigma, St. Louis, MO] for 40 minutes at 22% The tissue was
rinsed with CMF-HBSWHEPES, the filtered cell suspen-
sion was saved, and the tissue was incubated again with
pronase and chymopapain for 30 minutes. The tissue was
then disrupted further by “syringing” it lightly with a 30-mL
syringe and then washed again over Nitex. The cell filtrates
from the enzyme incubations were combined with those
from the final disruption, centrifuged at 200xg for 8 minutes.
These cell pellets were resuspended in CMF-HBSWHEPES
and pelleted again. Mast cells were counted using a low- pH
alcian blue stain (13). There was sufficient tissue for matched
studies of histamine release in all cases, but there were
sufficient cells for matched studies of prostaglandin release
in only seven cases. Only four cases had enough cells for
matched sulfidopeptide leukotriene assays, since leuko-
triene determinations require large quantities of cells (2-
3 x lo6 mast cells from each active and inactive segment].
Short-Term Culture
The cells were suspended in RPM1 1640 with 25
mmol/L HEPES buffer, 2 mmol/L L-glutamine, 100 pgg/mL
penicillin, and 100 rg/mL gentamicin at a concentration
of 2 x 10’ cells/ml. The cell suspension was then aliquoted
to multiwell tissue culture plates and incubated at 37°C in
humidified 95% air and 5% CO,. The cells were harvested
the next day after 12-18 hours, washed several times in PAG
buffer [PIPES, 7.6 g/L; NaCl, 6.4 g/L; KCl, 0.37 g/L; dextrose,
1.0 g/L; human serum albumin, 3.0 mg/L], pelleted, and
counted.
Histamine Release Assay
The mast cells in the single cell suspensions were
selectively challenged with the mast cell stimulus, goat anti-
human IgE. Duplicate samples containing 2 x lo4 mast cells
were challenged in PAGCM buffer (PAG buffer with
CaCl, - 2 H,O, 0.14 g/L (1 mmol/L), and MgCl, - 6 H,O, 0.20
g/L (1 mmol/L)) in a volume of 0.5 mL for 30 minutes at
37°C. Supernatants were assayed for histamine by auto-
July1990 ENHANCED MAST CELL MEDIATOR RELEASE IN UC 121

mated fluorometric analysis (14). Results were expressed as Results

percent of total histamine content as determined by lysis of
cells in 2% perchloric acid.
Radioimmunoassays
Prostaglandin D, and LTC, the predominant mast
cell cyclooxygenase and lipoxygenase pathway products,
were measured by radioimmunoassay (RIA), as previously
described (15,16), with dextran-coated charcoal used to
separate bound from free ligand. The PGD antiserum shows
a 4% cross-reactivity with PGF, and less than 1% with all
other heterologous ligands (15). The sensitivity limit was 20
pg/O.l mL sample. The rabbit LTC antiserum had 80%
cross-reactivity with leukotriene D and 10% cross-reactivity
with leukotriene E on a molar basis. The sensitivity limit was
also 20 pg/O.l mL sample (IS]. Quantities of sulfidopeptide
leukotriene C are expressed as picograms of immunoreac-
tive LTC per lo6 mast cells.
Statistical Tests
Data for the histamine, prostaglandin, and leuko-
triene release were compared by paired two-sample Stu-
dent’s t test and the Wilcoxon matched-pairs signed-rank
test.
Pathological Correlation
Tissue was obtained from 12 cases of UC
resected because of complications of active disease,
treatment failure, or long-standing chronic disease. As
shown on Table 1, histological review of the 12 active
UC segments showed that all had very significant
disease with severe distortion of the crypt architec-
ture, and all but two with severe mononuclear cell
infiltration. Five of these segments had marked ulcer-
ation, and all but one had significant neutrophil
infiltration. The corresponding grossly normal appear-
ing segments had mild disease with few or no neutro-
phils and little crypt distortion, with the exception of
one case in which the crypt distortion was severe but
other indices of inflammation were mild, indicating
quiescent disease. The differences between the inac-
tive and active tissue parameters were found to be
statistically significant on analysis by Student’s t test
and the Wilcoxon matched-pairs signed-rank test
(P 5 0.05 and P 5 0.004, respectively). Statistical anal-
ysis showed that the active and inactive disease groups
were significantly different for all parameters (Table

Table 1. Review of Histological Sections

Crypts with
Tissue Ulcers Erosions neutrophils Eosinophils Mononuclear CrYpt
site (70surface) (5%surface] (mean/lpf ] (mean/hpf ) cell& distortion

Inactive
IUC Rcolon 0.00 0.12 Mild Severe
2uc Rcolon 0.40 4.80 Moderate Mild
3uc Rcolon 0.00 1.20 Mild Rare
4uc R colon 0.00 13.50 Normal Rare
5uc R colon 0.00 6.70 Mild Mild
6UC R colon 0.00 0.85 Mild Mild
7uc R colon 0.00 1.10 Mild Mild
6UC R colon 0.70 8.70 Moderate Mild
9uc R colon 0.00 12.00 Moderate Mild
IOUC R colon 0.50 0.15 Mild Mild
1lUC R colon 0.00 3.10 Normal None
12uc R colon 0.00 21.60 Mild Mild
Active
1uc Trans colon 0 0.00 2.30 Mild Severe
2uc Trans colon 60 5.80 11.00 Severe Severe
3uc L colon 0 1.45 6.69 Moderate Severe
4uc Trans colon 40 0.63 42.70 Severe Severe
5uc L colon 0 0.40 5.30 Severe Severe
6UC L colon 0 0.12 1.50 Severe Severe
7uc L colon 0 1.40 2.00 Severe Severe
8UC Trans colon 0 0.75 34.00 Severe Severe
9uc L colon 0 1.40 16.30 Severe Severe
IOUC Trans colon 60 1.60 1.30 Severe Severe
IIUC L colon 55 0.88 1.55 Severe Severe
12uc L colon 4 1.40 17.85 Severe Severe

R, right; L. left; lpf. low-power field; hpf, high-power field; Trans. transverse.
“Inactive vs. active difference significant at P c 0.05.
bInactive vs. active difference significant at P c 0.004.
122 FOX ET AL.
1). There was no correlation between any of the
individual pathological parameters and percent total
histamine release by mast cells in either actively
involved or quiescent tissue.
Characteristics of Cell Suspensions
There were no statistically significant differ-
ences between active and inactive segments in yield
of mast cells per gram wet weight tissue, purity of mast
cells in those suspensions, or histamine content per
cell. Suspensions from the UC tissue yielded a mean
of 2.6 x lo5 (+1.7 SD) mast cells per gram wet weight of
inactive tissue and 4.1 x lo5 (+ 3.4 SD] mast cells per
gram wet weight of actively involved tissue. Suspen-
sions from inactive UC tissue contained, on average,
3.1% (+1.7% SD) mast cells while those from the
active segments contained 1.8% (+1.8’S SD] mast
cells. Mast cells from inactive tissue contained a mean
of 2 pg histamine per cell (kl.1 pg SD] and cells from
active tissue contained a mean of 2.3 (k1.1 SD) pg
histamine per cell.
Mediator Release
The intestinal cell suspensions were stimulated
with goat anti-human IgE to assess the level of mast
cell responsiveness. Mast cells derived from the active
areas of the 12 specimens released a greater propor-
tion of their preformed cellular stores of histamine
than did those from the uninvolved segments (Figure
1). This difference was highly significant at maximal
mast cell histamine release at 3 pg/mL anti-IgE
[P I 0.001). The difference was also significant
*Q) 40 1
6i J
2 metabolites showed that mast cells derived from the
$ 30- * * * **
g newly generated mediators prostaglandin D, and imu-
._
$J 20-
‘;
z cells was significant at 0.3 pg/mL (P 5 0.02),0.1 pg/mL
7 lo- _ >*
P
5 sulfidopeptide leukotriene release by mast cells from
I-
8 om,001 .Ol .l 1 10 100
Anti-IgE @g/ml)
Figure 1. Comparison of histamine release by human intestinal
mucosal mast cells from active (m and inactive (0) segments of UC
specimens on challenge with goat anti-human IgE. Each point
represents the mean * SJIM of duplicate determinations from 12
preparations. Differences are statistically signi5cant at **P 5
0.001, *P I 0.01, ttP 5 0.02 and tP < 0.05.
GASTROENTEROLOGY
Vol. 99. No. 1
60 -
50 -
40 -
30 -
20 -
10 -
Anti-IgE @g/ml)
Figure 2. Comparison of prostaglandin II, release by human
intestinal mucosal mast cells from active (w) and inactive (0)
segments of UC specimens on challenge with goat anti-human IgE.
Each point represents the mean * SEM of duplicate determina-
tions from seven preparations. Differences are statistically signi5-
cant at ***P I 0.02, **P I 0.03, and *P < 0.04.
(P 5 0.01) at 10, 1, 0.3, and 0.1 pg/mL anti-IgE. The
maximal histamine release from the cells of inactive
tissue, 11.7% + 1.9 SEM, was not statistically different
from the maximal release from the last 12 preparations
from the grossly normal margins of adenocarcinoma
specimens we have processed, 15% + 2.9 SEM. There
is normally large variation in the histamine release
from the mast cells from carcinoma margins, and the
maximal histamine release in these 12 preparations
ranged from l%-41% of total cellular histamine. The
maximal histamine release from cells derived from
the active segments of UC bowel, 24.4% + 4 SEM, was
statistically different from that of the “normal” cells
from carcinoma cases (P I 0.05).
Analysis of the supernatants for arachidonic acid
more active areas also released greater amounts of the
noreactive leukotriene C [Figures 2 and 3). The differ-
ence in PGD generated by “abnormal” and “normal”
(P 5 0.031, and 0.03 pg/mL (P 5 0.04) anti-IgE. The
active areas appeared substantially greater than that
from cells from inactive segments; however, the num-
bers were too small to achieve statistical significance.
Since the variability in amount of arachidonic acid
metabolites released is usually greater than that seen
with histamine, the PGD results are especially signifi-
cant (17). For example, maximal PGD release varied
from 11 ng/106 cells to 101 ng/lO’ mast cells and
maximal immunoreactive LTC release varied from
5-101 ng/106 mast cells.
July 1990
60 1 T .
50 -
40 -
30 - whereas only two of the controls were atopic. Further-
20 - disease at the time of lavage, it was difficult to assess
10 -
,001 .Ol .l 1 IO 100
Anti-IgE (pg/ml)
Figure 3. Comparison of immunoreactive LTC release by human
intestinal mucosal mast cells from active (H) and inactive (II)
segments of UC specimens on challenge with goat anti-human IgE.
Each point represents the mean + SEM of duplicate determina-
tions from four preparations.
Discussion
We have shown that mast cells derived from
actively inflamed UC tissue release greater amounts of
histamine, prostaglandin D,, and sulfidopeptide leuko-
triene than mast cells from relatively quiescent seg-
ments. This is not because cells from the segments
with active disease contain significantly more hista-
mine, but because they release a greater proportion of
their stored histamine. Because the arachidonic acid
products are freshly generated on stimulation of the
cell, the greater release of these may reflect either
larger stores of precursor or amplification of the
arachidonic acid metabolism pathways.
Although we have not definitively proven that other
cells in the suspension, such as eosinophils or mono-
cytes, did not contribute to the levels of arachidonic
acid metabolites measured in this study, it is unlikely
that cells other than mast cells were the source for
PGD and LTC released on stimulation. The stimulus
for mediator release was goat anti-human IgE, which
induces release by cross-linking IgE bound to the
high-affinity IgE receptor on the mast cell. Eosino-
phils, lymphocytes, and monocytes are known to have
low-affinity receptors for IgE (18,19), but the condi-
tions of the cell isolation procedure, multiple washes,
and overnight culture in the absence of exogenous IgE
are expected to remove IgE from a low-affinity recep-
tor.
Our study is the first report of in vitro comparison of
human mast cells from inflamed and noninflamed
intestinal tissue. The activation we found may be
unique to idiopathic IBD or could be common to other
inflammatory processes. In fact, we anticipated that
mast cells would respond to the inflammatory milieu
with the sort of “activation” or hyperresponsiveness
ENHANCED MAST CELL MEDIATOR RELEASE IN UC 123
we observed because, in the bronchial lavage study of
Flint et al., the mast cells from allergic asthmatics
were found to release more histamine than those from
normal controls (20). However, in that study the
asthmatics were all known to be allergic by skin test,
more, because the asthmatics did not have active
the contribution of inflammation. In our study we used
each patient as his or her own control, thereby elimi-
nating some of the wide interpersonal differences in
histamine release that are normally seen; in addition,
we compared the means to those of theoretically
“normal” control mast cells from the grossly normal
margins of adenocarcinoma resections.
It is especially interesting that the differences in
mast cell responsiveness are seen even after multiple
washes and 12 hours in culture away from the inflam-
matory environment. This would not be the case if the
mast cell response were caused merely by an agent
present at the time of stimulation; rather, it suggests
that a lymphokine or other cytokine has caused a
substantial change in the signal transduction appara-
tus. The increased levels of histamine could be ex-
plained by changes in the cytoskeleton that allow
greater exocytosis of granules or changes in the gran-
ules themselves, further allowing greater liberation of
histamine from the proteoglycan matrix. However,
that would not account for the increased levels of
newly generated mediators such as the arachidonic
acid metabolites. In that case, the increase is perhaps
better explained either by a modulation of the signal
transduction pathways, such as increased levels of the
enzymes involved in cell activation [protein kinases,
phospholipases, etc.], or by an increase in the levels of
precursors. Conversely, there is also the possibility
that the hyperresponsiveness seen in the active areas
is caused by removal of “suppression” rather than
activation and that a factor that usually downregulates
the mast cell is not present.
Analysis of the histological sections did not show
any single pathological feature that correlated with the
difference in mast cell histamine release. However,
statistical analysis of the pathological data showed that
the “normal” and “abnormal” segments were signifi-
cantly different in the histological features usually
associated with disease activity in UC. This confirms
that the “abnormal” segments did indeed have more
active disease, which was reflected in the mast cell
hyperresponsiveness, and that the gross appearance
was a good indicator of disease activity.
While our data support the concept of a role for mast
cells in the inflammation of IBD, this study does not
establish the reason for the local mast cell activity. The
stimulus for mast cell secretion in IBD is still unidenti-
124 FOX ET AL.
fied. The hypothesis of food antigens as inciting agents
for IBD has been fashionable (2l), but sensitivity to
food, especially cow’s milk proteins, is not a consistent
finding: it is now thought by some to represent an
epiphenomenon (22,23). The intense inflammation of
UC probably allows many protein antigens to escape
unprocessed through the mucosa, which might explain
the findings of serum antibodies to colon constituents
such as epithelial cell proteins (24). In fact, we have
preliminary data to suggest that mast cells from some
patients may respond to these epithelial proteins with
mediator release (25). It is also possible that in IBD the
mast cell is part of an amplification process rather than
an initiator of events as it is in allergic disorders. The
mast cell could be extending or perpetuating the
inflammation in response to signals from other partici-
pants, and terminating such an amplification cascade
could be an effective means of reducing the overall
inflammation. An obvious next step is to investigate
the drugs that are effective in IBD to assess their action
on human intestinal mast cells.
The finding of enhanced release in active ulcerative
colitis further implicates the mast cell as a contributor
to the inflammation of IBD. This in vitro system gives
us a unique opportunity to investigate the responses of
mast cells in inflamed human tissue: also, further
study will yield important insights into the mast cell’s
role in IBD and other gastrointestinal disorders.
References
1. Barrett KE. Metcalfe DD. The mucosal mast cell and its role in
gastrointestinal allergic diseases. Clin Rev Allergy 1984;2:39-53.
2. Woodbury RG, Miller HR, Huntley JF, Newlands GF, Palliser
AC, Wakelin D. Mucosal mast cells are functionally active
during spontaneous expulsion of intestinal nematode infections
in the rat. Nature 1984;312:450-452.
3. Lawrence ID, Warner JA, Cohan VL. Hubbard WC, Kagey-
Sobotka A, Lichtenstein LM. Purification and characterization
of human skin mast cells: evidence for human mast cell
heterogeneity. J Immunoll987;139:3062-3069.
4. Tharp MD, Kagey-Sobotka A. Fox CC, Marone G, Lichtenstein
LM, Sullivan TJ. Functional heterogeneity of human mast cells
from different anatomical sites: In vitro responses to morphine
sulfate. J Allergy Clin Immunoll987;79:646-652.
5. Lawrence ID, Warner JA, Cohan VL, Lichtenstein LM. Kagey-
Sobotka A. Vavrek RJ, Stewart JM, Proud D. Induction of
histamine release from human skin mast cells by bradykinin
analogs. Biochem Pharmacoll989;38:227-233.
6. Schulman ES, Proud D, Liu MC, MacGlashan DW Jr, Lichten-
stein LM, Plaut M. Human lung macrophages induce histamine
release from basophils and mast cells. Am Rev Respir Dis
1985;131:230-235.
7. Dvorak AM, Monahan RA, Osage JE, Dickersin GR. Crohn’s
disease: transmission electron microscopic studies II. Immuno-
logic inflammatory response. Alterations of mast cells, Baso-
phils, eosinophils, and the microvasculature. Hum Path01 1980;
11:606-619.
8. Sommers SC. Mast cells and paneth cells in ulcerative colitis.
Gastroenterology 1966;51:841-848.
9. Hiatt RB, Katz L. Mast cells in inflammatory conditions of the
gastrointestinal tract. Am J Gastroenteroll962;37:541-545.
GASTROENTEROLOGY Vol. 99, No. 1
10.Dvorak AM, Monahan RA, Osage JE, Dickersin GR. Mast-cell
degranulation in Crohn’s disease. Lancet 1978;1:498.
11.Wasserman SI. Mediators of immediate hypersensitivity. J
Allergy Clin 1mmuno11983;72:101-115.
12.Fox CC, Dvorak AM, Peters SP, Kagey-Sobotka A, Lichtenstein
LM. Isolation and characterization of human intestinal mucosal
mast cells. J Immunol1985;135:483-491.
13.Gilbert HS, Ornstein L. Basophil counting with a new staining
method using alcian blue. Blood 1975;46:279-286.
14.Siraganian RP. An automated continuous flow system for the
extraction and fluorometric analysis of histamine. Anal Bio-
them 1974;57:383-394.
15.Schulman ES, Newball HH, Demers LM, Fitzpatrick FA,
Adkinson NF Jr. Anaphylactic release of thromboxane A,,
prostaglandin Dz, and prostacyclin from human lung paren-
chyma. Am Rev Respir Dis 1981;124:402-406.
16.Hayes EC, Lombard0 DL, Girard Y, Maycock AL, Rokach J,
Rosenthal AS, Young RN, Egan RW, Zweerink HJ. Measuring
leukotrienes of slow-reacting substance of anaphylaxis: develop-
ment of a specific radioimmunoassay. J Immunol1983;131:429-
433.
17.MacGlashan DW Jr, Peters SP, Warner J, Lichtenstein LM.
Characteristics of human basophil sulfidopeptide leukotriene
release: releasability defined as the ability of the basophil to
respond to dimeric cross-links. J 1mmuno11986;136:2231-2239.
18.Capron M, Jouault T, Prin L, Joseph M, Arneisen JC. Butter-
worth AE, Papin J-P, Kusnierz JP, Capron A. From parasites to
allergy: a second receptor for IgE. Immunol Today 1986;7:15-18.
19.Spiegelberg HL. Structure and function of Fc receptors for IgE
on lymphocytes, monocytes, and macrophages. Adv Immunol
1984;35:61-88.
20.Flint KC, Leung KBP, Hudspith BN, Brostoff J, Pearce FL,
Johnson NM. Bronchoalveolar mast cells in extrinsic asthma: a
mechanism for the initiation of antigen specific bronchoconstric-
tion.BrMed J 1985;291:923-926.
21.Taylor KB, Truelove SC, Wright R. Serologic reactions to gluten
and cow’s milk proteins in gastrointestinal disease. Gastroenter-
ology 1964;46:99-108.
22.Kirsner JB, Shorter RG. Recent developments in nonspecific
inflammatory bowel disease. N Engl J Med 1982;306:837-848.
23.McCaffery TD Jr, Kraft SC, Rothberg RM. The influence of
different techniques in characterizing human antibodies to
cow’s milk proteins. Clin Exp Immunol1972;11:225-234.
24.Aronson RA, Cook SL, Roche JK. Sensitization to epithelial
antigens in chronic mucosal inflammatory disease I. Purifica-
tion, characterization and immune reactivity of epithelial cell-
associated components (ECAC). J Immunoll983;131:2796-2804.
25.Fox CC, Roche JK, Lichtenstein LM. Immediate hypersensitiv-
ity responses in idiopathic inflammatory bowel disease (IBD]:
histamine release from human intestinal mast cells in response
to gut epithelial antigens (abstr). Gastroenterology 1987;92:1394.
Received January 20,1989. Accepted January 19,199O.
Address requests for reprints to: Charity C. Fox, M.D., The Johns
Hopkins Asthma and Allergy Center, 301 Bayview Boulevard,
Baltimore, Maryland 21224.
Publication #008 from the Johns Hopkins Asthma and Allergy
Center.
This work was supported by the Myerhoff Digestive Disease/
Inflammatory Bowel Disease Center, the National Foundation for
Ileitis and Colitis, and the National Institute of Allergy and Infec-
tious Disease Grants AI07290 and AI08270. The authors wish to
acknowledge the assistance of Drs. Edward J. Wolf and Mary L.
Harris in obtaining surgical specimens for this study.