Download Lymphocyte activation as measured by interleukin-2

Immunol. Cell Biol. (1990)68. 155-160
Lymphocyte activation as measured by interleukin-2 receptor
expression to gluten fraction 111 in coeliac disease
I. A. Penttila, C. E. Gibson, B. D. Forrest, A. G. Cummins and J. T. LaBrooy
Department of Medicine. Royal Adelaide Hospital and University of Adelaide, Adelaide,
SA 5000. Australia
(Submitted 8 September 1989. Accepted for publication 9 March 1990.)
Summary Lymphocyte activation was examined by interIeukin-2 (IL-2) receptor expression on
peripheral blood mononuciear cells from coeliac and control subjects. Purified T cells were incubated
with gluten fraction 111 (a known toxic peptidc for coeliac subjects), soyabean hydrolysate (an
unrelated hydrolysed food antigen), and Concanavalin-A (Con-A. a non-specific mitogen). After 1-5
days incubation, expression of IL-2 receptors was assessed using a cellular enzyme-linked
immunosorbent assay (ELISA). Gluten fraction 111 induced expression of IL-2 receptors on T
lymphocytes from coeliac but not ftom normal subjects ( P ^ 00005). whereas soyabean hydrolysate
did not induce IL-2 receptor expression. Lymphocytes from both coetiac and normal subjects had
similar increased IL-2 receptor expression after incubation with Con-A. Flowcytometry was also used
to contirm specific expression of IL-2 receptor expression of lymphocytes from coeliac subjects.
Interlcukin-2 receptor expression increased from 0 to 5-4% of cultured mononuclear cells after 7 days
incubation with gluten fraction III. These cells were CD3-positiveand CD4-positivc. We conclude that
peripheral blood lymphocytes from coeliac subjects are sensitized specifically to gluten fraction III.
INTRODUCnON
Coeliac disease is characterized by a hypersensitivity reaction of the small intestine to cereals
containitig gluten which results in intestinal
damage (I). A soluble peptic/tryptic digest of
gluten, gluten fraction 111. is toxic in coeliac
patients (2). The mechanism causing damage to
the small intestine in coeliac disease has not
been fully elucidated, but is thought to involve
an abnormal immune response to the gluten
component of cereals (1.3).
Evidence implicating a cell-mediated immune
response to gluten in coeliac disease is accumulating. Coeliac disease has many similarities to a
cell-mediated immune response in the gut. as
exemplified by mucosal graft-versus-hos! reaction (4), with intestinal crypt hyperplasia,
increased intra-epithelial lymphocytes and
mucosal mast cells, and increased expression of
Correspondence: Dr I. A. Penttila, Department of
Medicine. Royal Adelaide Hospital, North Terrace,
Adelaide, SA 5000. Australia.
Abbreviations u.sed in this paper: Con-A, Concanavalin-A; DTH, deiayed-type hypersensitivity reaction;
ELiSA. enzyme-linked immunosorbent assay; IL-2,
interleukin-2; IL-2R. interleukin-2 receptor; LMIF,
leucocyte migration inhibition factor; PBS, phosphate-buffered saline.
class II antigens (1.4-7). Furthermore, mesenteric node lymphocytes from coeliac patients
proliferate in vitro to gluten fraction 1! 1 but do
not respond to other food antigens (8.9). In contrast, evidence for systemic lymphocyte activation in coeliac subjects is limited. Peripheral
blood T lymphocytes from coeliac subjects are
hyporesponsive to gluten as measured by either
proliferation or skin testing for a delayed-type
hypersensitivity (DTH) reaction (9.10). Despite
the hyporesponsiveness of systemic T lymphocytes to gluten, these cells may nonetheless be
sensitized to gluten even in the absence of in
v(7ro proliferation. Zeitzc/fl/. (11) have recently
reported that mucosal lymphocytes provided
antigen-specific help but did not proliferate during Chlamydia infection of primates. Similarly,
evidence for activation of lymphocytes in coeliac disease is suggested by the observation that
lymphocytes from the peripheral blood produce
leucocyte migration inhibition factor (LMIF), a
cytokine whose activity is correlated with cellmediated immunity in man, af\er stimulation
with gluten or its derivatives (3.12,13).
Activation antigens define the state of activation of human mature lymphocytes (14).
These antigens are not expressed on resting
lymphocytes and can be detected using monoclonal antibodies. We have therefore investigated systemic lymphocyte activation using
136
I. A.
interleukin-2 receptor expression, as measured
by a cellular enzyme-linked immunosorbent
assay (ELISA) technique on lymphocytes from
coeliac and normal subjects after iticubation
with gluten fraction 111. Flow cytometry was
also used to characterize the phenotype of cultured lymphocytes and to confirm interleukin-2
(IL-2) receptor expression.
SUBJECTS AND METHODS
Subjects
Blood was collected from six adult coeliacs and a
similar number of normal subjects. Gluten-sensitive
enteropathy was diagnosed in coeliac subjects by the
presence of sub-total or partial villous atrophy on
small bowel biopsy, which improved on a gluten-free
diet at the time of a second biopsy. All coeliac patients
were on a gluten-free diet.
Gluten fraction I I I
Gluten fraction 111 was prepared as outlined in the
method of Frazer et al. (2). Gluten fraction III was
dissolved in RPMI-1640 medium (Flow Laboratories.
Sydney) and sterilized by filtration. Soyabean hydrolysate was purchased from Sigma (St Louis. USA).
In vitro culture
Peripheral blood mononuclear cells were obtained
from venous hcparinized blood which was layered on
Ficoll-Paquc (Pharmacia Fine Chemicals. Uppsala.
Sweden) density gradient, and cenirifuged at 400 g for
25 min. Cells were washed twice m phosphate-buffered saline (PBS), and rcsuspended in complete
medium (RPMI-1640. 10% fetal calf serum. Flow
Laboratories, 25 mmol/L Hepes buffer. L-glutamine.
lOOng/mL streptomycin. 100 ng/mL penicillin).
Mononuclear cell suspensions were enriched for T
lymphocytes by passage over nylon wool and the
concentration adjusted to 1 X lOVmL in complete
medium. 01 mLaiiquots of the cells from either coeliac or control subjects were cultured in flat-bottomed
96-well tissue culture plates (Linbro. Flow Laboratories). Experiment cultures were stimulated with 100
jig/mL of either gluten fraction III. soyabean hydrolysateor lOiig/mLCon-A. All wells were adjusted to 0-2
mL. Control wells contained cells withoul antigen.
Plates were incubated at 37°C in 5% CO: for N5 days
before assay for interleukin-2 receptor (IL-2R)
expression.
Interleukin-2 receptor cell ELISA
An ELISA technique was used to measure IL-2
receptors on human lymphocytes and was modified
from that described by Igietseme and Herscowitz {1 5).
After incubation with antigen, plates were centrifuged
at 500 g for 5 min. The supernatant was gently
removed from each well. Freshly prepared glutaraldehyde (100 nL/well of a 0-125% v/v solution) was
added, ihe plates were centrifuged again at 500^ for 5
min. and ineubated at room temperature for 30 min.
The fixed cells were washed three times with PBS and
drained. An aliquot of 100 mmol/L glycine (100 nL)
was added into each well, the plates were incubated at
room temperature for 30 min. the fluid removed and
drained; 100 nL of anti-IL-2 receptor antibody (1/500
dilution. Dakopatts. Glostrup, Denmark) was added
in PBS containing 0 05% bovine serum albumin.
The plates were incubated at 4''C overnight, washed
five times and drained. The second antibody (100 ^L,
sheep anti-mouse antibody conjugated to alkaline
phosphatase. Dakopatts) was added, incubated al
37''C for 4 h and washed five times in PBS/0-05% BSA.
Freshly prepared y>nitrophenyl phosphate (Sigma)
was added, and the plates incubated for 3 h at 37°C.
Absorbance was read at 405 nm.
Flow cytometry
Purified peripheral blood lymphocytes from three
coeliac subjects were cultured with no antigen, gluten
fraction 111 (100 ng/mL) or Con-A (10 ng/mL) for 7
clays. Cells were purified on a Ficoll-Paque discontinuous gradient and washed with PBS/5% BSA. Aliquots
of resuspended ceils (I X lO'/niL. 50 fiL) were incubated on ice for 30 min with mouse IgG monoclonal
antibodies against CD3. CD4. CDS. HML-1
(Immunotech, Marseilles, 16), IL-2R, as well as mixtures of certain combinations of these antibodies.
Double mixing was used as a means of determining
dual labelling. Antibody labelling was detected using
FITC-anti-mouse IgG F(ab')2 goat antibody (1/20
dilution in PBS/5% BSA, 30 min. Cappel). Cells were
washed and fixed in 0-4% paraformaldehyde/2%
sucrose/0-02% sodium azide/PBS. Flow cytometry
was performed using a Coulter EPICS V with recording of fluorescence histograms of relative number
of cells versus fluorescence intensity on a four decade
logarithmic scale of 256 channels.
Proliferation assay
Purified peripheral blood lymphocytes (5 X
lO^/mL. 100 jxL) were incubated for 3 and 5-6 days
with no antigen, gluten fraction 111 (100 jjg/mL), soyabean hydrolysate (100 ng/mL) or Con-A (10 ng/mL).
[3H]-Thymidine (1 ^Ci) was added to each aliquot of
cells during the last 24 h of incubation. Cells were harvested and radioactivity measured usmg a liquid scintillation counter. The stimulation index was detined as
the ratio of counts per min of stimulated cells divided
by counts per minute of unstimulated cells.
Statistics
Readings of optical density for IL-2R expression by
lymphocytes from coeliac and control subjects were
compared using Analysis of Variance.
RESULTS
interleukin-2 receptor ELISA
Interleukin-2 receptor expression of lymphocytes isolated from coeliac subjects peaked after
137
IL.-2R ON COELIAC T CELLS
0.0-" ,
Tlnw(»i)
Time (h)
Fig. 1. Intedeukin-2 receptor expression by T
lymphocyte-enriched peripheral blood cells from coeliac (•) and normal subjects (o) after stimulation with
1 (X) i^ig/mL gluten fraction 111. Results are expressed
as the mean and s.d. of six subjects per group.
5-6 of incubation with gluten fraction 111
(/'<0.0001). whereas lymphocytes isolated
from control subjects showed no response (Fig.
1). There was a significant difference of coeliacderived lymphocytes compared to those derived
from normal subjects {/*= 0.0005). This indicated sensitization of coeliac-derived lymphocytes. Soyabean hydrolysate did not induce
IL-2R expression on lymphocytes isolated from
either coeliac or normal subjects and was similar
to incubation without any antigen (Figs 2,3)This demonstrated that sensitization was
specific to the gluten fraction III. The nonspecific mitogen. Con-A. however, induced IL-2
expression on lymphocytes from both coeliac
Fig. 3. Interleukin-2 receptor expression by T
lymphocyte-enriched peripheral blood cells from coetiac (•) and normal subjects (•) after culture in media
alone. Results are expressed as the mean and s.d. of six
subjects per group.
fer between the two groups. This showed that
lymphocytes from normal subjects are capable
of expressing IL-2 receptors and therefore indicated that they were not sensitized to gluten
fraction III.
Flow cytometry
Representative results offlowcytometry for one
of the three coeliac subjects whose peripheral
blood lymphocytes were cultured is presented in
Table 1. The majority (76-80%) of cells were T
lymphocytes as shown by CD3 expression. The
proportion of the CD4 subset remained the
same, but CD8 and HML-1 expressing cells
increased after incubation with gluten fraction 11
and Con-A. Interleukin-2 receptor expression
increased after gluten fraction III and Con-A
{P=0.0004) and control (P<0.000\) subjects
(Fig. 4). However, IL-2R expression did not dif-
Tlme (h)
0.0 •"
Tinia(h)
Fig. 2. Interleukin-2 receptor expression by T
lymphocyte-enriched peripheral blood eells from coeliac (•) and normal subjects (•) after stimulation with
100 |ig/mL of soyabean hydrolysate. Results are
expressed as the mean and s.d. of six subjects per
group.
Fig. 4. Interleukin-2 receptor expression by T
lymphocyte-enriched peripheral blood cells from coeliac (•) and normal subjects (a) after stimulation with
10 lig/mL Con-A. Results are expressed as tbe mean
and s.d. of six subjects per group.
158
I. A. PENTTILA £7-.4/...
Table I. Phenotype and activation of peripheral blood
nionotiuclear cells from a coeliac subject after
incubation for 7 days with no antigen, gluten fraction
III or Con-A.
-ve
niagc of positive ceils
Determinant
CD3
CD4
CDS
HML-I
IL-2R
CD3/1L-2R
CD4/IL-2R
CD8/IL-2R
No
Gluten rraciion
antigen
III
76
56
19
0
0
75
53
16
76
50
18
0-2
0 1
75
48
26
80
52
25
7
5
79
54
34
Con-A
V*
80
56
29
14
13
80
53
34
incubation (Fig. 5). Double mixing of CD3/IL2R and CD4/IL-2R antibodies did not increase
the percentage reading for CD3 or CD4 alone.
This indicated co-expression on the same cell.
Thus CD3 and CD4 positive cells were activated
to express IL-2 receptors. The reading for
CD8/IL-2R was higher than for CD8 alone. This
was best seen after incubation with gluten fraction III, and to a lesser extent with Con-A. In the
case of gluten fraction III, the data indicated that
CD8 cells did not express lL-2 receptors, as there
was a simple summation of readings for CDS
and IL-2R that was approximately equal to the
value for the mixture of CD8/IL-2R. After ConA incubation, however, some CD8 cells may
have expressed IL-2 receptors, as there was
incomplete summation of readings.
Proliferation assay
Purified peripheral blood lymphocytes from
four coeliac subjects showed a wide range of stimulation indices after incubation with gluten
fraction III, soyabean hydrolysate of Con-A
(Table 2). However, the median values of
lymphocyte proliferation after gluten fraction
III or soyabean hydrolysate Incubation were
much lower than after Con-A.
DISCUSSION
We have used an ELISA technique, similar to
that reported by Igietseme and Herscowitz (1 5)
for mouse lytnphocytes, to measure the activation of gluten-stimulated lymphocytes from
coeliac patients. The ELISA assay detects 1L-2R
on the surface of cells and is simple, reproducible and convenient to perform in the laboratory. These IL-2R are induced during T cell
activation after incubation with antigens or
mitogens (14.17).
\^ 1
Gluten fraction III
o
.a
E
o:
Con A
/
4k
i rm
r
13.1%
1111 f n 1 1
111 rr.*,i.i I
Fluorescence (log)
Fig. 5. lnterIeukin-2 receptor expression of purified
peripheral blood mononuclear cells after incubation
for 7 days with no antigen, gluten fraction III or ConA. Histograms show relative number of cells versus
logarithm ofthe fluorescence intensity. The percentage reading to the right ofthe arrow is given.
in the past, activation of lymphocytes by
gluten fraction 111 has given variable results, for
example, peripheral blood lymphocytes proliferate poorly in response to gluten fraction III,
whereas mesenteric lymph node cells proliferate
strongly after incubation with this antigen (8).
Gluten fraction 111 is a peptic/tryptic digest of
gluten, which is known to be toxic to the coeliac
epithelium, and has been demonstrated to
induce migration inhibition of leucocytes from
coeliac subjects, but not from normal subjects
(2,3.18). Until the present time, this LMIF assay
1L-2R ON COELIAC T CELLS
Table 2. Proliferation of purified peripheral blood
lymphocytes after incubation for 3 and 5-6 days with
alulcn fraction III. soyabean hydrolysate or Con-A.
pHl-Thymidine incorporation was expressed as the
stimulation index, which is ihe ratio of eounis per
minute of stimulated lo unstimulated cells.
Median
Con-A
Gluicn fraction Ml
Soyabean hydrolysaie
Days 5-6
Con-A
Gluten fraction HI
Soyabiran hytlrolysaif
109
41
2-9
26
46
2-3
Range
n
4-810 6
I 7-740 7
1-6-7-3 3
5-6-57
0-5-128
1'2-I52
4
4
.1
has been the principal in vitro technique for
detnonstrating a cell-mediated imtnune reaction
in coeliac disease, because of the difficulty of
inducing proliferation of peripheral blood
lymphocytes to gluten-derived antigen. We also
found relatively poor 4-5 fold increased proliferation to gluten fraction III. compared with
26-109 fold increased proliferation to Con-A.
The growth of T lymphocytes after activation
by antigen is regulated by binding of IL-2. Intertcukin-2 binds to high-affinity membrane receptors to cause lymphocyte proliferation (14.17),
In tinstimulated or resting T lymphocytes, the
number of IL-2R is low and the receptors are of
low affinity. These receptors increase in number
and affinity after antigenic stimulation but later
decline with prolonged stimulation (15). The
decay in expression of IL-2R correlates with a
loss of proliferative capacity, indicating that a
threshold number of IL-2R need to be occupied
for proliferation to occur, even though the cell is
in an activated state (19). In fact, lymphocyte
activation, as measured by IL-2R expression,
still occurs in the presence of inhibitors of DNA
synthesis, but not to inhibitorsof prolein synthesis (15). Thus, absence of proliferation of peripheral blood lymphocytes does not indicate an
absence of activation. This is further highlighted
by the findings of Zeitz et ai (II) who showed
that mucosal lymphocytes provide antigen specific help for immunoglobulin production after
antigen stimulation, but do not proliferate in
response to the same antigen.
Quantitation of IL-2R on lymphocytes from
coeliac subjects after gluten stimulation is a sensitive measure of lymphocyte activation in the
peripheral blood. In this study, we have shown
that these cells from coeiiac subjects are specifically activated ater exposure to gluten fraetion
139
111 to express IL-2R. lt is likely that these
lymphocytes circulating systemically had been
sensitized in vivo while in the gut mucosa,
although they do not express IL-2R prior to
incubation. Lymphocyte activation was not
demonstrated to soyabean hydrolysate. which is
an unrelated food antigen. This indicated specific activation. Mitogen stimulation (Con-A)
resulted in activation of lymphocytes and
increased IL-2R expression on lymphocytes
from both coeliac and normal subjects, indicating that lymphocytes from normal individuals
were capable of expressing IL-2R.
Crabtree et al. (20) have recently demonstrated that soluble IL-2R concentration is
increased in untreated coeliac disease and
decreases on gluten-free diet. Our study supports this work and indicates that the likely
origin of soluble IL-2R are activated T
lymphocytes. As we have shown that these sensitized lymphocytes do not express 1L-2R while
circulating in peripheral blood, presumably activation and release of IL-2R occurs mucosally
where some gluten antigen may be present.
The phenotype of lymphocytes expressing IL2R was CD3-positive and CD4-positive. CD8
lymphocytes did not express IL-2R after incubation with gluten fraction III. at least as
assessed by dual mixing of relevant antibodies.
However, the same technique suggested that
some CD8-positive lymphocytes expressed IL2R after Con-A stimulation. This was because
there was an incomplete summation of percentage readings for CD8 or 1L-2R alone compared
with the mixture of CD8/IL-2R.
Usually ingestion of dietary proteins induces
tolerance of both cell-mediated and humoral
responses to a given antigen (7,21). Gluten is a
dietary antigen which is similar to other food
antigens in inducing tolerance, at least in mice
(22). However, in coeliac subjects. lymphocytes
in the peripheral blood are sensitized, and can be
activated in vitro if stimulated with gluten or its
derivatives to express IL-2R, or to produce
LMIF. In coeliac disease, specific tolerance may
be impaired, as is suggested by impaired suppressor cell function to gluten (23).
.Acknowledgements
We aeknowlcdge the National
Health & Medical Research Council and the Royal
Adelaide Hospital Research Fund for financial
support. We are grateful l o M r A. Bishop (Department
of Human Immunology, IMUS, Adelaide) for
performing the flow cytometry.
160
I. A. PENTTILA £r^Z..
REFERENCES
1. Marsh, M.N. 1988. Studies of intestinal lymphoid
tissue. XI-The immunopathology of cell mediated
reactiotis in gluten sensitivity and other enteropathies. Scand. Microsc. 2: 1663-1684.
2. Fraser. A. C. Fletcher, R. F.. Ross. C. A. C . Shaw,
B.. Sammons. H. G. and Schneider. R. 1959.
Gluten induced enteropathy. The effect of partially digested gluten. Lancet ii: 252-255.
3. Guan, R., Rawclifie, P. M.. Priddle, J. D. and
Jewell, D. P. 1987. Cellular hypersensitivity to
gluten derived peptides in coeliac disease. Gut 27:
426-434.
4. Ferguson, A. 1987. Models of immunologieally
driven small intestinal damage. In Immunopathology of the Small Intestine. M. N. Marsh (ed.),
John Wiley. Chichester. pp. 225-252.
5. Seiby. W. S.. Janossy, G.. Bofill. M. and Jewell, D.
P. 1983. Lymphocyte subpopuiations in the
human small intestine. The findings in normal
mucosa and in the mucosa of patients with adult
coeliac disease. Clin. Exp. Immunol. 52: 219228.
6. Ciclitira, P. J.. Nelufer. J. M.. Ellis. H. J. and
Evans. D. J. 1986. The effect of gluten on HLADR in the small intestinal epithelium of patients
with coeliac disease. Clin. Exp. Immunol. 63:
101-104.
7. Mowat, A. M. and Ferguson. A. 1981. Hypersensitivity in the small intestinal mucosa. V. Induction of cell mediated immunity to dietary gluten.
Clin. E.xp. Immunol. 43: 574-582.
8. Housley. J., Asquith. P. and Cooke. W. T. 1969.
Immune response to gluten in adult coeliac disease. Brit. Med J. 2: 159-161.
9. Sikora, K.. Anand, B. S.. Truelove, S. C, Ciclitara,
P. J. and Offord. R. E. 1976. Stimulation of
lymphocytes from patients with coeliac disease by
a sub-fraction of gluten. Lancet ii: 389-391.
10. Cunningham-Rundles. S., Pollack, M. S.. Good.
R. A. and Dupont. B. Response to wheat antigens
in in vitro lymphocyte transformation among
HLA-B8 positive normal donors. Transplant.
Proc. 10: 977-979.
11. Zeitz, M., Quinn, C. T., Graeff, A. S. and James.
S. P. 1988. Mucosal T cells provide helper function but do not proliferate when simulated by specific antigen in Lymphogranuloma venerium proctitis in nonhuman primates. Gastroenterology 94:
353-366.
12. Howdle, P. D., Bullen, A. W. and Losowsky, M. S.
1982. Cell-mediated immunity to gluten within
the small intestinal mucosa in coeliac disease. Gut
23: 115-122.
13. Lyd ford-Da vis, H.. Karagiannis, J. A.. Priddle,
L. D. and Jewell. D. P. (1987). Preliminary
characterisation of leucocyte migration inhibition
factor (LMIF) produced by lymphocytes from coeliac patients stimulated with gluten peptides. Clin.
Sci. 72: 89P.
14. Robb, R. J..Munck, A. andSmith, K. A. 1981. T
cell growth factor: quantification, specificity and
biological relevance. / Exp. Med. 154: 14551474.
15. Igietseme. J. U. and Herscowitz H. B. 1987. Quantitative measurement of T lymphocyte activation
by an enzyme-linked immunosorbent assay
(ELISA) detecting interleukin 2 receptor
expression. J. Immunol. Meth. 97: 123-131.
16. Cerf-Bensussan. N., Jarry, A., Brousse. N.,
Lisowska-Grospierre. B.. Guy-Grand. D. and
Griscelti. C. 1987. A monoclonal antibody (HML1) defining a noval membrane molecule present
on human intestinal lymphocytes. Eur. J. Immunol. 17: 1279-1285.
17. Robb, R. J.. Greene. W. C. and Rusk. C M. 1984.
Low and high affinity cellular receptors for interleukin 2. Implications for the level of Tac antigen.
J. Exp. Med. 160: 1126-1146.
18. Bullen. A. W. and Losowsky. M. S. 1978. Cell
mediated immunity to gluten fraction 111 in adult
coeliac disease. Gtii. 19: 126-131,
19. Churilla.A.andBraciale. V. L. 1987. Lack of IL-2
dependent proliferation despite significant
expression of high-affinity IL-2 receptors on
murine cytolytic lymphocytes clones late after
antigenic stimulation. J. Immunol. 138: 13381342.
20. Crabtree, J. E., Heatley. R. V., Juby, L. D.,
Howdle. P. D. & Losowsky, M. S. 1989. Serum
interleukin-2 receptor in coeliac disease: response
to treatment and gluten challenge. Clin. Exp.
Immunol. 11: 345-348.
21. Strobel, S. and Ferguson. A. 1984. Immune
responses to fed protein antigens in mice. 3.
Systemic tolerance or priming is related to the age
which antigen is first encountered. Pediatr. Res.
18: 588-594.
22. Troncone. R. and Ferguson, A. 1988. Gliadin
presented via the gut induces oral tolerance in
mice. Clin. Exp. Immunol. 72: 284-287.
23. Pignata.C. Troncone, R.. Monaco, G. cffl/. 1985.
Impaired suppressor activity in children affected
by coeliac disease. Gut 26: 285-290.