Download Histidine Decarboxylase, a Pyridoxal Phosphate

0013-7227/03/$15.00/0
Printed in U.S.A.
The Journal of Clinical Endocrinology & Metabolism 88(4):1445–1452
Copyright © 2003 by The Endocrine Society
doi: 10.1210/jc.2002-021761
Histidine Decarboxylase, a Pyridoxal PhosphateDependent Enzyme, Is an Autoantigen of Gastric
Enterochromaffin-Like Cells
FILIP SKÖLDBERG, GUIDA M. PORTELA-GOMES, LARS GRIMELIUS, GUNNAR NILSSON,
JAAKKO PERHEENTUPA, CORRADO BETTERLE, EYSTEIN S. HUSEBYE, JAN GUSTAFSSON,
ANDERS RÖNNBLOM, FREDRIK RORSMAN, AND OLLE KÄMPE
Departments of Medical Sciences (F.S., A.R., F.R., O.K.), Women’s and Children’s Health (J.G.), and Genetics and Pathology
(G.M.P.-G., L.G., G.N.), Uppsala University, University Hospital, 751 85 Uppsala, Sweden; Hospital for Children and
Adolescents (J.P.), University of Helsinki, 00029 Helsinki, Finland; Department of Medical and Surgical Sciences (C.B.),
University of Padova, 35128 Padova, Italy; and Division of Endocrinology (E.S.H.), Institute of Medicine, Haukeland
Hospital, 5021 Bergen, Norway
Patients with autoimmune polyendocrine syndrome type 1
often have autoantibodies against neurotransmitter synthesizing enzymes, including the pyridoxal phosphate-dependent enzymes glutamic acid decarboxylase and aromatic Lamino acid decarboxylase. Using a candidate approach, we
have identified the histamine-synthesizing enzyme histidine
decarboxylase, also pyridoxal phosphate dependent, as an autoantigen in this disorder. Anti-histidine decarboxylase antibodies reacting with in vitro translated antigen were found in
36/97 (37%) of autoimmune polyendocrine syndrome type 1
patients studied. The antibodies also reacted with the native
enzyme in HMC-1 cell lysates and did not cross-react with the
A
UTOIMMUNE POLYENDOCRINE syndrome type 1
(APS1), also known as autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy (OMIM 240300),
is an autosomal recessive disorder caused by mutations of
the AIRE (autoimmune regulator) gene encoded on chromosome 21 (1, 2). The classical features are chronic mucocutaneous candidiasis, hypoparathyroidism, and adrenal
failure, two of which should be present for the clinical diagnosis of APS1 (3). Other manifestations include gonadal
failure, alopecia, vitiligo, insulin-dependent diabetes mellitus (IDDM), pernicious anemia, intestinal dysfunction, and
chronic active hepatitis (4). APS1 can be regarded as a model
disorder for organ-specific autoimmunity and recently, targeted inactivation of the AIRE gene in mice has provided an
animal model, which shares some features of the human
syndrome (5, 6). Several manifestations have been linked to
the presence of specific autoantibodies, often directed against
cytochrome P450 enzymes or neurotransmitter synthesizing enzymes, with restricted tissue distribution. These
autoantigens include the pyridoxal phosphate-dependent
enzymes glutamic acid decarboxylase (GAD) and aromatic
Abbreviations: AADC, Aromatic L-amino acid decarboxylase; AIRE,
AIRE autoimmune regulator; APS1, autoimmune polyendocrine syndrome type 1; CgA, chromogranin A; EC, enterochromaffin; ECL, enterochromaffin-like; GAD, glutamic acid decarboxylase; HDC, histidine
decarboxylase; IDDM, insulin-dependent diabetes mellitus; MBP, maltose binding protein; TPH, tryptophan hydroxylase; VMAT-2, vesicular
monoamine transporter-2.
highly homologous aromatic L-amino acid decarboxylase.
Anti-histidine decarboxylase antibodies were associated with
a history of intestinal dysfunction (P ⴝ 0.017). Gastric and
duodenal biopsies from a patient with anti-histidine
decarboxylase antibodies were studied by immunohistochemistry. The oxyntic mucosa was found to lack the histamine
producing enterochromaffin-like cells, suggestive of an autoimmune destruction. To our knowledge, this is the first report
of autoantibodies against histidine decarboxylase and absence of gastric enterochromaffin-like cells. (J Clin Endocrinol Metab 88: 1445–1452, 2003)
L-amino acid decarboxylase (AADC; Refs. 7 and 8). Intestinal dysfunction, including steatorrhea, diarrhea, and
constipation is reported to be present in 15–25% of APS1
patients (3, 4, 9, 10) and is associated with an autoimmune
reaction against tryptophan hydroxylase (TPH) and an
autoimmune destruction of the serotonin producing enterochromaffin (EC) cells of the intestine (11, 12, 18).
There have been no reports on histamine producing enterochromaffin-like (ECL) cells being affected in APS1. As opposed
to EC cells, which are primarily located in the antrum and
duodenum, ECL cells are located in the oxyntic mucosa of the
stomach. They express histidine decarboxylase (HDC), the histamine-synthesizing enzyme, and are the main site of histamine
synthesis in the gastric mucosa (13). HDC is a pyridoxal phosphate-dependent enzyme (14) expressed in the brain, mast cells,
gastric mucosa, and fetal liver (discussed in Ref. 15), structurally
related to AADC and GAD (16), which are 52% homologous
over 476 amino acids and 27% homologous over 316 amino
acids by BLASTP alignment, respectively. The aim of the
present study was to investigate whether HDC can be targeted
by autoantibodies in APS1, and the possible relation to alterations of ECL cells.
Patients and Methods
Patients
Serum samples were analyzed from 10 Swedish, 16 Norwegian, 57
Finnish, and 14 Italian patients with APS1. As controls, we analyzed 44
patients with Addison’s disease, 54 patients with IDDM, 30 patients with
1445
1446
J Clin Endocrinol Metab, April 2003, 88(4):1445–1452
Sköldberg et al. • HDC Is an ECL Cell Autoantigen
FIG. 1. Scattergram showing the anti-HDC antibody reactivity of serum samples from patients with APS1 (n ⫽
97), Addison’s disease (n ⫽ 44), IDDM (n ⫽ 54), Hashimoto’s thyroiditis (n ⫽ 30), vitiligo (n ⫽ 30), and blood
donors (n ⫽ 108). Dashed line indicates the cut-off value
of positive results (mean value of blood donors ⫹ 4 SD).
Hashimoto’s disease, 30 patients with vitiligo, and 108 healthy blood
donors. The clinical characteristics of the APS1 patients have been described elsewhere (17–19). We have used the definition of intestinal
dysfunction as periodic steatorrhea, diarrhea, or severe constipation
(18). Data on antibody reactivity against AADC and TPH in the Nordic
patients were available from previous studies (17, 18). The methods used
were approved by the local ethics committee.
for human AADC in pBluescript was generously donated by Dr. Hiroshi
Ichinose (21). The pMALc2 expression vector was kindly provided by
Dr. Mark Peakman. Bases 83–1930 of the AADC cDNA (GenBank accession no. NM_000790) were subcloned into the EcoRI site of pMALc2,
for the production of recombinant AADC as described below. Normal
rabbit serum was purchased from DAKO Corp. (Glostrup, Denmark).
In vitro transcription/translation
Plasmids and antisera
The plasmid pVL-HDC74, containing a cDNA encoding full-length
human HDC and a rabbit antiserum against human HDC, were kindly
provided by Dr. Kimio Yatsunami (20). The cDNA was subcloned into
the BamHI site of pGEM3zf⫹ (Promega Corp., Madison, WI) and the
integrity verified by sequencing of the entire insert. A full-length cDNA
Human full-length HDC (amino acids 1– 662) and AADC were expressed by coupled in vitro transcription and translation of 35S-labeled
antigen in the TnT system (Promega Corp.) as previously described.
Translation products were analyzed by SDS-PAGE and measurement of
35
S-methionine incorporation by trichloroacetic acid precipitation, followed by scintillation counting.
Sköldberg et al. • HDC Is an ECL Cell Autoantigen
J Clin Endocrinol Metab, April 2003, 88(4):1445–1452 1447
FIG. 2. Immunoprecipitates of 35S-methionine labeled
HMC-1 cell lysates subjected to SDS-PAGE and transfer to
nitrocellulose, followed by analysis by phosphorimaging
(A), and immunoblotting with a specific anti-HDC rabbit
antiserum (B), respectively. Lane 1, Anti-HDC rabbit antiserum; lane 2, normal rabbit serum; lanes 3–5, anti-HDC
positive APS1 patient sera; lane 6, anti-HDC negative
APS1 patient serum; lanes 7 and 8, healthy blood donor
sera; lane 9, unlabeled HMC-1 cell lysate; lane 10,
prestained molecular weight standard (not visible); and
lane 11, in vitro translated HDC. B, Composite of two exposure times (lanes 1 and 2 and 3–11, respectively) due to
high background in the first two lanes.
Immunoprecipitation of in vitro translated antigen
Immunoprecipitation of in vitro translated HDC was carried out in
96-well plates, and all serum samples were analyzed in duplicate. An
anti-HDC reactivity index of each serum was calculated as follows: (cpm
of unknown sample ⫺ cpm of negative control)/(cpm of positive control ⫺ cpm of negative control) as described (22). Approximately 20,000
cpm of radiolabeled HDC and 2.5 ␮l serum in a final volume of 50 ␮l
was used for each reaction. For some experiments, immunoprecipitations of HDC and AADC were carried out in microcentrifuge tubes.
Bound immune complexes were then washed six times with 1 ml of 20
mm Tris-HCl (pH 8), 150 mm NaCl, 0.02% sodium azide, 1% Tween 20,
and bound proteins analyzed by SDS-PAGE, followed by PhosphorImager analysis (Molecular Dynamics, Inc., Sunnyvale, CA).
Immunoprecipitation and immunoblotting of HMC-1
cell lysates
The human mast cell line HMC-1, which expresses HDC (23) was
maintained as described (24). Cells were labeled for 5 h with 35Smethionine (125 ␮Ci/ml) in DMEM with 10% dialyzed FCS, washed in
PBS, and stored at –70 C. Cell pellets (⬃2 ⫻ 107 cells) were either lysed
on ice for 1 h in 1 ml nondenaturing lysis buffer (20 mm Tris-HCl, pH
8; 150 mm NaCl; 0.02% sodium azide; 1% Triton X-100) or resuspended
in 50 ␮l PBS, followed by addition of 50 ␮l 2⫻ denaturing lysis buffer
(20 mm Tris-HCl, pH 8; 150 mm NaCl; 0.02% sodium azide; and 2%
sodium dodecyl sulfate) and denatured at 80 C for 10 min. Denatured
lysates were diluted with 10 vol nondenaturing lysis buffer and incubated on ice for 30 min. Insoluble material was removed by centrifugation at 16,000 ⫻ g for 30 min. In immunoprecipitations, nondenatured
lysates were used for human sera, whereas denatured lysates were used
for rabbit sera. Immunoprecipitation, SDS-PAGE, and transfer to nitrocellulose membranes were performed essentially as described (25), except that antibodies were bound to Fast Flow Protein A-Sepharose
(Amersham Pharmacia Biotech, Uppsala, Sweden) before adding the cell
lysates. Bound proteins on the same nitrocellulose membrane were
analyzed both by phosphorimaging and by probing with a specific
anti-HDC antiserum (dilution 1:5000; Ref. 20). Antibody binding was
detected using a horseradish peroxidase conjugated rat antirabbit antibody (Amersham Pharmacia Biotech) diluted 1:5000 and Western Blotting Luminol Reagent chemiluminescent substrate (Santa Cruz Biotechnology, Inc., Santa Cruz, CA).
FIG. 3. Immunoprecipitation of 35S-methionine labeled in vitro
translated HDC (first and second rows), and AADC (third and fourth
rows), respectively, followed by SDS-PAGE and phosphorimaging.
Serum samples were preincubated with extracts of bacteria expressing MBP alone (first and third rows) or MBP-AADC fusion protein
(second and fourth rows) before adding the respective radiolabeled
antigen. Lanes 1 and 2, Two different APS1 patient sera positive for
anti-HDC and anti-AADC antibodies; lane 3, APS1 patient serum
positive for anti-HDC and negative for anti-AADC antibodies; lane 4,
APS1 patient serum negative for anti-HDC and positive for antiAADC antibodies; and lane 5, serum from a healthy blood donor.
Competition experiments
To investigate possible cross-reactivity between anti-HDC antibodies
and AADC, competition experiments were performed essentially as
described (26). The pMALc2 bacterial expression vector was used to
express amino acids 4 – 480 of human AADC as a fusion protein with
maltose binding protein (MBP) in the XL1-Blue MRF’ Escherichia coli
strain grown in Terrific Broth. Protein expression was induced with 0.5
mm isopropyl-␤-d-thiogalactopyranoside at 18 C. Cells were harvested
after 22 h and lysed with lysozyme and Triton X-100 as described (27).
Competition was conducted by preincubating the sera with E. coli extract
containing approximately 10 ␮g MBP-AADC fusion protein, or control
E. coli extract, for 1 h at 4 C before adding 35S-labeled in vitro translated
HDC or AADC and conducting an immunoprecipitation as described
above. Immune complexes were analyzed by SDS-PAGE, followed by
phosphorimaging.
1448
J Clin Endocrinol Metab, April 2003, 88(4):1445–1452
Sköldberg et al. • HDC Is an ECL Cell Autoantigen
TABLE 1. Clinical disorders and anti-HDC antibodies in 97 patients with APS1
a
Clinical disorder
Number with disorder/total
Intestinal dysfunction
Chronic active hepatitis
Hypoparathyroidism
Adrenal insufficiency
Gonadal failure
Diabetes mellitus
Alopecia
Vitiligo
Pernicious anemia
26/97 (27%)
16/97 (16%)
82/97 (85%)
78/97 (80%)
29/97 (30%)
13/97 (13%)
30/97 (31%)
20/97 (21%)
14/97 (14%)
Number with anti-HDC antibodies/total
With disorder
Without disorder
15/26 (58%)
11/16 (69%)
30/82 (37%)
32/78 (41%)
12/29 (41%)
8/13 (62%)
11/30 (37%)
8/20 (40%)
6/14 (43%)
21/71 (30%)
25/81 (31%)
6/15 (40%)
4/19 (21%)
24/68 (35%)
28/84 (33%)
25/67 (37%)
28/77 (36%)
30/83 (36%)
Pa
0.017
0.0089
0.78
0.12
0.65
0.066
⬎0.99
0.80
0.77
Calculated by use of Fisher’s exact test.
TABLE 2. Summary of histological findings in the gastric and
duodenal mucosa of the APS1 patient studied
Corpus
Antrum
Duodenum
Mucosal atrophy
CgA positive cells 2
Epithelial VMAT-2 positive cells 2 (mast cells present)
Somatostatin positive cells 2
Slight foveolar hyperplasia
Normal
Decreased number of cells denoted by 2.
Immunohistochemistry
Archival, paraffin-embedded biopsies were available from one single
APS1 patient who had undergone gastroscopy because of epigastrial
pains. This was a 26-yr-old woman with a history of mucocutaneous
candidiasis, hypoparathyroidism, adrenal failure, hypogonadism, alopecia, and vitiligo. At gastroscopy, no macroscopic abnormalities were
found. As controls, paraffin sections of histologically normal mucosa
from patients who underwent surgery for gastric tumors were immunostained.
Biopsies from gastric corpus, antrum, and distal duodenum of the
present patient (n ⫽ 3 each) and controls were fixed in 10% buffered
neutral formalin and routinely processed to paraffin. The sections, 4 ␮m
thick, were stained with hematoxylin-eosin, according to van Gieson or
immunostained using the Vectastain ABC kit (Vector Laboratories, Burlingame, CA) with mouse monoclonal antibodies to human chromogranin A (CgA, clone LK2H10, Roche Molecular Biochemicals, Mannheim, Germany; dilution 1:1000), human serotonin (clone 5HT-H209,
code no. M0758, DAKO Corp., Glostrup, Denmark; dilution 1:50) human
N-terminal gastrin (clone 4C7A1, code no. 1537, Immunotech, Marseille,
France; dilution 1:500), or human mast cell tryptase (code no 444905,
Calbiochem-Novabiochem, San Diego, CA; dilution 1:3000) and a rabbit
polyclonal antiserum to human C-terminal vesicular monoamine transporter-2 (VMAT-2; code no. H-V003, Phoenix Pharmaceuticals, Inc.,
Belmont, CA; dilution 1:2000), or human somatostatin (code no. A 568,
DAKO Corp.; dilution 1:500).
All immunostainings were performed without microwave pretreatment unless indicated. Furthermore, immunostaining for CgA was performed on paraffin sections pretreated for enhanced sensitivity in a
microwave oven (Whirlpool Nordic AB, Stockholm, Sweden) for 2 ⫻ 5
min at 750 W, using a citrate buffer (pH 6.0) as retrieval solution.
Statistical analyses
Frequencies of intestinal dysfunction and other manifestations in
APS1 patients with and without anti-HDC antibodies were compared by
the use of Fisher’s exact test. P value less than 0.05 was considered
significant.
Results
Immunoprecipitation of in vitro translated HDC with
patient sera
Coupled in vitro transcription and translation of the fulllength HDC and AADC cDNA yielded major products of
approximately 64 kDa and 50 kDa, respectively, as estimated
by SDS-PAGE. The in vitro translated HDC was recognized
specifically by the anti-HDC antiserum in immunoblotting
(data not shown). In vitro translated HDC was used in a
96-well immunoprecipitation assay, and the antibody reactivity was expressed as an anti-HDC reactivity index. The
results from the 96-well immunoprecipitation assay are
shown in Fig. 1. Using a cut-off value at 0.188, at 4 sd above
the mean index value of the blood donors, 36 of 97 APS1
patients (37%) were positive, whereas all of the control sera
tested were negative. We have not been able to detect any
antibody reactivity directed against the C-terminal approximately 20-kDa fragment (amino acids 478 – 680) expressed in
vitro (data not shown).
Immunoprecipitation and immunoblotting of HMC-1
cell lysates
To determine whether the antibodies reacting with in
vitro translated HDC were also able to immunoprecipitate
the native protein from lysates of cells expressing endogenous HDC, immunoprecipitation experiments with 35Smethionine-labeled HMC-1 cell lysates were performed.
Patient sera previously found to be positive in the antiHDC assay reacted with a protein migrating at approxi-
FIG. 4. Immunohistochemical stainings of biopsies from the gastric corpus (A–H), antrum (I–L), and duodenum (M and N) obtained from an
anti-HDC-positive APS1 patient (A, C, E, G, I, K, and M) and control patients (B, D, F, H, J, L, and N), respectively. A, Corpus mucosa of the
APS1 patient stained for CgA. The mucosa displays slight atrophy and marked reduction of CgA positive cells (microwave pretreated section).
Two CgA-positive cells can be seen (indicated by the square). Inset, Higher magnification of the immunoreactive cells. Bar, 100 ␮m; inset bar,
35 ␮m. B, Normal corpus mucosa stained for CgA showing numerous immunoreactive cells (no microwave pretreatment). Bar, 100 ␮m. C, Corpus
mucosa of the APS1 patient stained for VMAT-2. A few immunoreactive cells, presumably mast cells, are present in the stroma (indicated by
arrows). Bar, 42 ␮m. D, Normal corpus mucosa stained for VMAT-2. Numerous positive cells are present, both in the glandular epithelium (ECL
cells) and in the stroma. Bar, 100 ␮m. E, Corpus mucosa of the APS1 patient stained for somatostatin, displaying only a single immunoreactive
cell (indicated by the arrow). Bar, 42 ␮m. F, Normal corpus mucosa stained for somatostatin. Bar, 42 ␮m. G, Corpus mucosa of the APS1 patient
Sköldberg et al. • HDC Is an ECL Cell Autoantigen
J Clin Endocrinol Metab, April 2003, 88(4):1445–1452 1449
stained for mast cell tryptase. Immunoreactive mast cells are present in the stroma. Bar, 42 ␮m. H, Normal corpus mucosa stained for mast
cell tryptase. Bar, 50 ␮m. I, Antral mucosa of the APS1 patient stained for CgA, displaying a normal frequency of immunoreactive cells. Bar,
63 ␮m. J, Normal antral mucosa stained for CgA. Bar, 63 ␮m. K, Antral mucosa of the APS1 patient stained for gastrin, displaying a normal
frequency of immunoreactive cells. Bar, 63 ␮m. L, Normal antral mucosa stained for gastrin. Bar, 63 ␮m. M, Duodenal mucosa of the APS1
patient stained for CgA, displaying a normal frequency of immunoreactive cells. Bar, 63 ␮m. N, Normal duodenal mucosa stained for CgA. Bar,
63 ␮m.
1450
J Clin Endocrinol Metab, April 2003, 88(4):1445–1452
mately 64 kDa, which comigrated with the major band
immunoprecipitated with a specific rabbit anti-HDC serum (Fig. 2A). Sera of anti-HDC-negative patients and
healthy blood donors did not react with this protein. Immunoblotting of the same nitrocellulose membrane with
the rabbit anti-HDC serum showed that HDC was present
in these lanes and had an apparently identical molecular
mass, strongly suggesting that the radioactive 64-kDa
band represents HDC (Fig. 2B).
Competition experiments
Considering the sequence similarity between AADC, a
well known autoantigen in APS1, and HDC we investigated
whether anti-HDC reactivity could be reduced by preincubation with excess AADC. As shown in Fig. 3, it was found
that the amount of HDC immunoprecipitated by different
sera was only marginally decreased by preincubation with
AADC. This was true for sera, which had both anti-HDC and
anti-AADC reactivity, as well as a serum, which was only
anti-HDC positive. On the other hand, binding of radiolabeled AADC was almost abolished when preincubating sera
with unlabeled AADC.
Clinical characteristics of anti-HDC positive APS1 patients
Anti-TPH antibodies and the specific loss of TPH expressing EC cells of the gastrointestinal tract have been found in
APS1 patients with intestinal dysfunction (18, 28). As HDC
is present in a distinct endocrine cell type of the gastrointestinal tract, namely the ECL cells, we investigated whether
anti-HDC antibodies may also be linked to intestinal dysfunction. APS1 patients with and without anti-HDC antibodies were compared regarding the frequency of intestinal
dysfunction and other supposedly autoimmune manifestations of APS1. Fifteen of twenty-six (58%) of patients with a
history of intestinal dysfunction were anti-HDC positive,
compared with 20 of 71 (28%) patients without intestinal
dysfunction (P ⬍ 0.02). Anti-HDC antibodies were also
found more frequently among patients with a history of signs
of chronic active hepatitis (11 of 16, 69%) than those without
(25 of 81, 31%; P ⬍ 0.01). No other associations between
anti-HDC antibodies and different manifestations of APS1
were found (Table 1). Data on AADC and TPH immunoreactivity was available in 83 patients, 21 of which had a history
of intestinal dysfunction (17, 18). No patients with intestinal
dysfunction were identified which were anti-HDC positive
and anti-TPH negative, suggesting that anti-HDC antibodies
do not add to the sensitivity in identifying patients with
intestinal dysfunction. Notably, 13 of 21 (62%) patients with
intestinal dysfunction were positive for all three antibodies,
whereas this was the case in only 12 of 62 (19%) patients
without intestinal dysfunction (P ⫽ 0.0006).
Immunohistochemical findings
To investigate whether anti-HDC antibodies could be
linked to loss of HDC-expressing cells (ECL cells and mast
cells), sections of biopsies from the corpus, antrum, and
duodenum of an HDC-positive APS1 patient were studied.
The biopsies from the corpus showed some atrophy of the
Sköldberg et al. • HDC Is an ECL Cell Autoantigen
mucosa and those from the antrum a slight foveolar hyperplasia. The duodenal mucosa showed a normal histology.
None of the biopsies displayed any signs of inflammation.
Parietal cells and chief cells could be identified by hematoxylin-eosin staining (data not shown). The histopathological
findings are summarized in Table 2.
Sections of the mucosal biopsies were immunostained for
CgA, which is expressed in several endocrine cells (29). In the
corpus, only very few CgA-positive cells could be observed
(Fig. 4A) compared with the normal control section (Fig. 4B).
In the APS1 patient biopsies, these cells could only be detected after microwave pretreatment of the sections to enhance sensitivity (Fig. 4A), whereas this was not required for
the control sections (Fig. 4B).
The histamine producing ECL cells are the major CgA
positive cell type in the oxyntic mucosa (30), and to confirm
that ECL cells were lacking, we did immunostainings of the
corpus biopsies with a VMAT-2 antiserum (31). The corpus
mucosa completely lacked ECL cells detected by VMAT-2
staining (Fig. 4C). Some cells were present in the stroma,
presumably mast cells, expressing VMAT-2 (Fig. 4C, indicated by arrows). The normal control section contained numerous ECL cells in the middle portion of the crypts (Fig.
4D). Somatostatin cells, which do not express CgA in the
corpus (29), were almost completely absent from the corpus
mucosa. In three sections from separate biopsies from the
APS1 patient, one single somatostatin-positive cell could be
found (Fig. 4E). In the control section, a small number of
somatostatin-positive cells were seen scattered in the mucosa
(Fig. 4F). The presence of mast cells could be detected in both
the APS1 patient biopsies and the control sections by staining
for mast cell tryptase (Fig. 4, G and H). In the antrum, the
number of CgA-positive and gastrin-positive cells was comparable to that in normal biopsies (Fig. 4, I–L). The abundance of CgA-positive cells in the duodenal biopsies from the
APS1 patient and control patient was also similar (Fig. 4, M
and N). Serotonin-positive EC cells could be detected in the
duodenal biopsies of both the APS1 patient and the control
patient (data not shown).
Discussion
Using a candidate approach based on structural similarity,
we have identified HDC as a novel pyridoxal-phosphate
dependent autoantigen in APS1. The anti-HDC antibodies
were found to react with in vitro translated HDC as well as
the native enzyme present in HMC-1 cells and showed little
cross-reactivity with recombinant AADC, indicating that
unique epitopes are targeted. Purified HDC has been found
to have a lower molecular weight than predicted from the
cDNA sequence (32–34). This difference is believed to be due
to posttranslational cleavage of the C-terminus, which does
not have a homolog in the AADC amino acid sequence. In rat
gastric mucosa, this has been proposed to represent proteolytic activation of the HDC (35). We have not been able to
detect antibody reactivity directed against the C-terminal
approximately 20-kDa fragment, indicating that the antibodies are directed against the part of the protein that is similar
to AADC. Knowledge of the existence of specific non-crossreactive autoantibodies against these distinct proteins with
Sköldberg et al. • HDC Is an ECL Cell Autoantigen
structural similarities may be useful for future mapping of
conformational epitopes using chimeric hybrid molecules
and also for studies of possible intermolecular epitope
spreading (36, 37).
In the APS1 patient material studied, anti-HDC antibodies
were associated with intestinal dysfunction and chronic active hepatitis. HDC is expressed in a subset of endocrine cells
of the stomach, the ECL cells, and an immune reaction to
these cells may possibly be linked to gastrointestinal symptoms. In adult liver on the other hand, HDC is not normally
detectable at significant levels, suggesting that the association with chronic active hepatitis may rather be of indirect
nature.
Gastric and duodenal biopsies were available from one
patient with anti-HDC antibodies for immunohistochemical
studies. It was found that the oxyntic mucosa almost completely lacked endocrine cells, including ECL cells, suggestive of an autoimmune destruction. Högenauer et al. (28)
reported an APS1 patient with gastrointestinal symptoms, in
which a subset of duodenal endocrine cells were absent, and
subsequently reappeared in conjunction with clinical improvement. This suggests that enteroendocrine cells are not
lacking, e.g. due to a congenital aplasia caused by the defect
AIRE gene, and supports the hypothesis of an autoimmune
pathogenesis. To our knowledge, this is the first report of
absence of ECL cells in human and provides a novel basis for
studies of ECL cell function in the human gastric mucosa. In
contrast, mucosal mast cells, which also express HDC, could
be detected at all levels. The patient studied also had antibodies against AADC and TPH (8, 18). AADC activity has
been demonstrated in ECL cells (38) and both AADC and
TPH has been detected in serotonin-containing EC cells (18,
39). This highlights that the connection between autoantibody reactivity, antigen distribution, and histopathological
findings in autoimmune disorders is yet to be fully understood. Similarly, the widely expressed E2 component of the
pyruvate dehydrogenase enzyme complex is the main autoantigen in primary biliary cirrhosis, where biliary epithelial cells are specifically targeted by the immune system
(discussed in Ref. 40). Possibly, these apparent discrepancies
could be related to differences in proteasome mediated processing and presentation of autoantigenic epitopes (41).
Autoantibodies against intracellular proteins are generally
not believed to be pathogenic but may reflect activation of
autoreactive T cells, which could mediate target cell destruction. The identification of HDC as an autoantigen illustrates
the intriguing propensity of intracellular enzymes to be the
targets of B cell responses in autoimmune disorders (42).
These enzymes may have conserved structural features that
can provoke immune responses or may not be presented to
T cells in the thymus or periphery to a sufficient extent to
achieve immunological tolerance. The gene mutated in APS1,
AIRE, and its murine ortholog, have been identified (1, 2, 43)
and targeted inactivation of this gene in mice may provide
a tool for addressing this issue (5, 6).
In conclusion, we have found that HDC, the histaminesynthesizing enzyme, is a B cell autoantigen in APS1 and
report an APS1 patient with anti-HDC antibodies who was
found to lack ECL cells in the gastric oxyntic mucosa. These
findings extend our knowledge of autoantibody specificity
J Clin Endocrinol Metab, April 2003, 88(4):1445–1452 1451
and alterations of endocrine cells in this disorder. They
also provide a basis for studies of ECL cell physiology in
human and a candidate antigen for the development of immunotherapy of ECL-derived tumors. Notably, the vitiligoassociated APS1 autoantigen SOX10 (19) was recently found
to be targeted by tumor infiltrating lymphocytes in a patient with malignant melanoma who responded to immunotherapy (44).
Acknowledgments
We thank Dr. Mona Landin-Olsson for sera from diabetes patients,
Dr. Anthony P. Weetman for sera from vitiligo patients, Dr. Kimio
Yatsunami for the pVL-HDC74 plasmid and the anti-HDC antiserum,
Dr. Hiroshi Ichinose for the human AADC cDNA, and Dr. Mark Peakman for the pMALc2 plasmid. We are also indebted to Dr. Rolf Håkanson and Niclas Olsson for valuable discussions and technical advice,
and to Lars Berglund (Uppsala Clinical Research Center) for expert
statistical advice.
Received November 12, 2002. Accepted January 2, 2003.
Address all correspondence and requests for reprints to: Dr. Filip
Sköldberg, Department of Medical Sciences, Uppsala University, University Hospital, 751 85 Uppsala, Sweden. E-mail: Filip.Skoldberg@
medsci.uu.se.
This work was supported in part by the Medical Research Council,
the Torsten and Ragnar Söderberg Foundation, the Petrus and Augusta
Hedlund Foundation, the Swedish Medical Society, the Claes Groschinsky Memorial Foundation, the Agnes and Mac Rudberg Foundation,
the Tore and Wera Cornell Foundation, and the Professor Nanna Svartz’
Foundation.
References
1. Consortium TF-GA 1997 An autoimmune disease, APECED, caused by mutations in a novel gene featuring two PHD-type zinc-finger domains. The
Finnish-German APECED Consortium. Autoimmune PolyendocrinopathyCandidiasis-Ectodermal Dystrophy. Nat Genet 17:399 – 403
2. Nagamine K, Peterson P, Scott HS, Kudoh J, Minoshima S, Heino M, Krohn
KJ, Lalioti MD, Mullis PE, Antonarakis SE, Kawasaki K, Asakawa S, Ito F,
Shimizu N 1997 Positional cloning of the APECED gene. Nat Genet 17:393–398
3. Neufeld M, Maclaren N, Blizzard R 1980 Autoimmune polyglandular syndromes. Pediatr Ann 9:154 –162
4. Betterle C, Greggio NA, Volpato M 1998 Clinical review 93: autoimmune
polyglandular syndrome type 1. J Clin Endocrinol Metab 83:1049 –1055
5. Ramsey C, Winqvist O, Puhakka L, Halonen M, Moro A, Kämpe O, Eskelin
P, Pelto-Huikko M, Peltonen L 2002 Aire deficient mice develop multiple
features of APECED phenotype and show altered immune response. Hum Mol
Genet 11:397– 409
6. Anderson MS, Venanzi ES, Klein L, Chen Z, Berzins S, Turley SJ, Von
Boehmer H, Bronson R, Dierich A, Benoist C, Mathis D 2002 Projection of
an immunological self-shadow within the thymus by the Aire protein. Science
298:1395–1401
7. Björk E, Velloso LA, Kämpe O, Karlsson FA 1994 GAD autoantibodies in
IDDM, stiff-man syndrome, and autoimmune polyendocrine syndrome type
I recognize different epitopes. Diabetes 43:161–165
8. Rorsman F, Husebye ES, Winqvist O, Björk E, Karlsson FA, Kämpe O 1995
Aromatic-l-amino-acid decarboxylase, a pyridoxal phosphate-dependent enzyme, is a ␤-cell autoantigen. Proc Natl Acad Sci USA 92:8626 – 8629
9. Brun JM 1982 Juvenile autoimmune polyendocrinopathy. Horm Res 16:308 –316
10. Ahonen P, Myllarniemi S, Sipila I, Perheentupa J 1990 Clinical variation of
autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy (APECED)
in a series of 68 patients. N Engl J Med 322:1829 –1836
11. Ekwall O, Sjöberg K, Mirakian R, Rorsman F, Kämpe O 1999 Tryptophan
hydroxylase autoantibodies and intestinal disease in autoimmune polyendocrine syndrome type 1. Lancet 354:568
12. Ward L, Paquette J, Seidman E, Huot C, Alvarez F, Crock P, Delvin E, Kampe
O, Deal C 1999 Severe autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy in an adolescent girl with a novel AIRE mutation: response to
immunosuppressive therapy. J Clin Endocrinol Metab 84:844 – 852
13. Rubin W, Schwartz B 1979 Electron microscopic radioautographic identification of the ECL cell as the histamine-synthesizing endocrine cell in the rat
stomach. Gastroenterology 77:458 – 467
14. Ono S H 1959 Pyridoxal phosphate: a coenzyme for histidine decarboxylase.
Nature 184:1143–1144
1452
J Clin Endocrinol Metab, April 2003, 88(4):1445–1452
15. Grzanna R 1984 Histidine decarboxylase: isolation and molecular characteristics. Neurochem Res 9:993–1009
16. Sandmeier E, Hale TI, Christen P 1994 Multiple evolutionary origin of pyridoxal-5⬘-phosphate-dependent amino acid decarboxylases. Eur J Biochem
221:997–1002
17. Husebye ES, Gebre-Medhin G, Tuomi T, Perheentupa J, Landin-Olsson M,
Gustafsson J, Rorsman F, Kämpe O 1997 Autoantibodies against aromatic
l-amino acid decarboxylase in autoimmune polyendocrine syndrome type I.
J Clin Endocrinol Metab 82:147–150
18. Ekwall O, Hedstrand H, Grimelius L, Haavik J, Perheentupa J, Gustafsson
J, Husebye E, Kämpe O, Rorsman F 1998 Identification of tryptophan hydroxylase as an intestinal autoantigen. Lancet 352:279 –283
19. Hedstrand H, Ekwall O, Olsson MJ, Landgren E, Kemp EH, Weetman AP,
Perheentupa J, Husebye E, Gustafsson J, Betterle C, Kampe O, Rorsman F
2001 The transcription factors SOX9 and SOX10 are vitiligo autoantigens in
autoimmune polyendocrine syndrome type I. J Biol Chem 276:35390 –35395
20. Yatsunami K, Tsuchikawa M, Kamada M, Hori K, Higuchi T 1995 Comparative studies of human recombinant 74- and 54-kDa L-histidine decarboxylases. J Biol Chem 270:30813–30817
21. Ichinose H, Kurosawa Y, Titani K, Fujita K, Nagatsu T 1989 Isolation and
characterization of a cDNA clone encoding human aromatic L-amino acid
decarboxylase. Biochem Biophys Res Commun 164:1024 –1030
22. Petersen JS, Hejnaes KR, Moody A, Karlsen AE, Marshall MO, HoierMadsen M, Boel E, Michelsen BK, Dyrberg T 1994 Detection of GAD65
antibodies in diabetes and other autoimmune diseases using a simple radioligand assay. Diabetes 43:459 – 467
23. Maeda K, Taniguchi H, Ohno I, Ohtsu H, Yamauchi K, Sakurai E, Tanno Y,
Butterfield JH, Watanabe T, Shirato K 1998 Induction of L-histidine decarboxylase in a human mast cell line, HMC-1. Exp Hematol 26:325–331
24. Olsson N, Piek E, Sundström M, ten Dijke P, Nilsson G 2001 Transforming
growth factor-␤-mediated mast cell migration depends on mitogen-activated
protein kinase activity. Cell Signal 13:483– 490
25. Winqvist O, Karlsson FA, Kämpe O 1992 21-Hydroxylase, a major autoantigen in idiopathic Addison’s disease. Lancet 339:1559 –1562
26. Bonifacio E, Lampasona V, Bingley PJ 1998 IA-2 (islet cell antigen 512) is the
primary target of humoral autoimmunity against type 1 diabetes-associated
tyrosine phosphatase autoantigens. J Immunol 161:2648 –2654
27. Sambrook J, Russel D 2001 Molecular cloning: a laboratory manual. Cold
Spring Harbor, NY: Cold Spring Harbor Laboratory Press
28. Högenauer C, Meyer RL, Netto GJ, Bell D, Little KH, Ferries L, Santa Ana
CA, Porter JL, Fordtran JS 2001 Malabsorption due to cholecystokinin deficiency in a patient with autoimmune polyglandular syndrome type I. N Engl
J Med 344:270 –274
29. Portela-Gomes GM, Stridsberg M, Johansson H, Grimelius L 1997 Complex
co-localization of chromogranins and neurohormones in the human gastrointestinal tract. J Histochem Cytochem 45:815– 822
30. Simonsson M, Eriksson S, Håkanson R, Lind T, Lonroth H, Lundell L,
O’Connor DT, Sundler F 1988 Endocrine cells in the human oxyntic mucosa.
A histochemical study. Scand J Gastroenterol 23:1089 –99
Sköldberg et al. • HDC Is an ECL Cell Autoantigen
31. Erickson JD, Schafer MK, Bonner TI, Eiden LE, Weihe E 1996 Distinct pharmacological properties and distribution in neurons and endocrine cells of two
isoforms of the human vesicular monoamine transporter. Proc Natl Acad Sci
USA 93:5166 –5171
32. Tran VT, Snyder SH 1981 Histidine decarboxylase. Purification from fetal rat
liver, immunologic properties, and histochemical localization in brain and
stomach. J Biol Chem 256:680 – 686
33. Watabe A, Fukui T, Ohmori E, Ichikawa A 1992 Purification and properties
of l-histidine decarboxylase from mouse stomach. Biochem Pharmacol 43:
587–593
34. Yamauchi K, Sato R, Tanno Y, Ohkawara Y, Maeyama K, Watanabe T, Satoh
K, Yoshizawa M, Shibahara S, Takishima T 1990 Nucleotide sequence of the
cDNA encoding L-histidine decarboxylase derived from human basophilic
leukemia cell line, KU-812-F. Nucleic Acids Res 18:5891
35. Dartsch C, Chen D, Håkanson R, Persson L 1999 Histidine decarboxylase in
rat stomach ECL cells: relationship between enzyme activity and different
molecular forms. Regul Pept 81:41– 48
36. Schwartz HL, Chandonia JM, Kash SF, Kanaani J, Tunnell E, Domingo A,
Cohen FE, Banga JP, Madec AM, Richter W, Baekkeskov S 1999 Highresolution autoreactive epitope mapping and structural modeling of the 65
kDa form of human glutamic acid decarboxylase. J Mol Biol 287:983–999
37. Kawasaki E, Yu L, Rewers MJ, Hutton JC, Eisenbarth GS 1998 Definition of
multiple ICA512/phogrin autoantibody epitopes and detection of intramolecular epitope spreading in relatives of patients with type 1 diabetes. Diabetes
47:733–742
38. Håkanson R, Owman C 1967 Concomitant histochemical demonstration of
histamine and catecholamines in enterochromaffin-like cells of gastric mucosa.
Life Sci 6:759 –766
39. Hunyady B, Zolyomi A, Hoffman BJ, Mezey E 1998 Gastrin-producing endocrine cells: a novel source of histamine in the rat stomach. Endocrinology
139:4404 – 4415
40. Gershwin ME, Ansari AA, Mackay IR, Nakanuma Y, Nishio A, Rowley MJ,
Coppel RL 2000 Primary biliary cirrhosis: an orchestrated immune response
against epithelial cells. Immunol Rev 174:210 –225
41. Kuckelkorn U, Ruppert T, Strehl B, Jungblut PR, Zimny-Arndt U, Lamer S,
Prinz I, Drung I, Kloetzel PM, Kaufmann SH, Steinhoff U 2002 Link between
organ-specific antigen processing by 20S proteasomes and CD8(⫹) T cellmediated autoimmunity. J Exp Med 195:983–990
42. Song YH, Li Y, Maclaren NK 1996 The nature of autoantigens targeted in
autoimmune endocrine diseases. Immunol Today 17:232–238
43. Blechschmidt K, Schweiger M, Wertz K, Poulson R, Christensen HM,
Rosenthal A, Lehrach H, Yaspo ML 1999 The mouse Aire gene: comparative
genomic sequencing, gene organization, and expression. Genome Res 9:
158 –166
44. Khong HT, Rosenberg SA 2002 The Waardenburg syndrome type 4 gene,
SOX10, is a novel tumor-associated antigen identified in a patient with a
dramatic response to immunotherapy. Cancer Res 62:3020 –3023