| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
CLINICAL CASE SEMINAR |
Departments of Pediatrics (M.O.-H.), Medicine (G.V.B.), Dermatology (G.J.A.), and Pathology (J.H.Y.), Johns Hopkins University School of Medicine and Hospital, Baltimore, Maryland 21287
Address all correspondence and requests for reprints to: John H. Yardley, M.D., Department of Pathology, Johns Hopkins Hospital, Baltimore, Maryland 21287. E-mail: jyardley{at}jhmi.edu.
| Abstract |
|---|
|
|
|---|
Objective: The objective of the study was to study the histopathological and immunological findings in the GI mucosa of a patient with typical features of APS I, malabsorption, and pernicious anemia.
Design and Patient: Biopsies from the GI tract of a patient with APS I were immunostained with chromogranin for GI endocrine cells (GIECs). Blinded slides were graded for numbers of endocrine cells. Normal gastric mucosa was exposed to the patients serum to test for circulating anti-GIEC and antiparietal cell antibodies using indirect immunofluorescence.
Setting: The study was conducted at the Departments of Pediatrics and Medical Gastroenterology in an academic medical center.
Results: The patients GI mucosa demonstrated absence of GIECs throughout, including gastric gastrin-secreting cells, and her laboratory tests for serum gastrin levels were low normal. Both GIECs and parietal cells were absent in her gastric corpus. The patients serum contained anti-GIEC antibody but no antiparietal cell antibody.
Conclusions: These observations suggest that GIECs in APS I are subject to an autoimmune destruction that can cause widespread GIEC loss. This could explain the GI dysfunctions that are often noted in the syndrome including malabsorption and atrophic gastric changes with pernicious anemia. We also hypothesize that absence of gastric parietal cells may result mainly from hypogastrinemia that is mainly the loss of gastrin-secreting cells rather than from immune-mediated destruction of parietal cells like that seen in the atrophic gastritis associated with adult-onset pernicious anemia.
| Introduction |
|---|
|
|
|---|
Nonendocrine manifestations can include dental enamel hypoplasia, nail dystrophy, vitiligo, keratoconjunctivitis (25), chronic mucocutaneous candidiasis, and dystrophy of ectodermal tissues (1, 2, 3, 4, 5). APS I is inherited in an autosomal recessive pattern and is now known to be caused by mutations in the autoimmune regulator gene (AIRE) (6, 7). The product of the AIRE gene is expressed in the thymus and appears to be involved in T cell self-tolerance to endocrine tissues. The presence of autoantibodies to tissue-specific antigens has been reported in patients who lack the AIRE gene product (8, 9, 10, 11, 12).
Nearly 25% of patients with APS I have significant gastrointestinal (GI) dysfunction manifested primarily by steatorrhea or watery diarrhea, which is very difficult to control (1, 2, 14, 15, 16). The pathophysiological mechanisms leading to this malabsorptive diarrhea are unclear; previous explanations in some cases have included a direct effect of hypoparathyroidism and hypocalcemia, intestinal candidiasis, mucosal atrophy, pancreatic insufficiency, and bile salt deficiency (17, 18, 19). Approximately 15% of patients may develop chronic atrophic gastritis and pernicious anemia (13).
We previously reported a patient with a striking paucity of GI endocrine cells (GIECs) and hypothesized that this finding could be an important factor in the steatorrhea associated with APS I (19). Studies by others (20, 21, 22) have suggested the importance of enteroendocrine cells in the pathophysiology of APS I-associated diarrhea. In this report we expand our preliminary case report on GIEC findings in APS I by demonstrating the absence of GIECs throughout the patients GI tract, as well as highlighting the gastric mucosal and hematological aspects and possible mechanism for the pernicious anemia.
| Patient and Methods |
|---|
|
|
|---|
A 12-yr-old female was referred to pediatric gastroenterologists at the Johns Hopkins Childrens Center for evaluation of chronic diarrhea, which was present since 21 months of age. She was diagnosed with APS I at 15 months when she developed the first of multiple manifestations of this syndrome beginning with hypocalcemic tetany secondary to hypoparathyroidism. Subsequently she developed hypothyroidism, mucocutaneous candidiasis, lipoid nephrosis, ovarian failure, type 1 diabetes, vitiligo, nail dystrophy, dental enamel hypoplasia, hypercalcuria with nephrolithiasis, and corneal keratopathy secondary to vitamin A deficiency. She is said, by unconfirmed report from the patients family, to have the major Finnish mutation R257X, the most common AIRE mutation.
Medications on initial presentation to Johns Hopkins included calcitriol, calcium carbonate, levothyroxine, vitamin B12, pancrelipase, fluoride, fat-soluble vitamin preparations, and parathormone.
GI findings were significant for the passage of two to 10 nonbloody, foul-smelling, fatty-appearing loose stools per day. After onset of the chronic diarrhea, she never had periods of formed, normal-appearing stools. Abnormal findings from previous medical evaluations included at least three elevated 72-h fecal fat tests and one abnormally low bentiromide test with less than 6% p-aminobenzoic acid excretion at 8 h. Pernicious anemia was diagnosed when she was 4 yr old. Findings included positive serum antibodies to intrinsic factor on one occasion, endoscopic and histological evidence of gastric atrophy, chronic mild anemia with high mean corpuscular volume, low mean corpuscular hemoglobin and mean corpuscular hemoglobin concentration, and low serum vitamin B12. After her first admission to Johns Hopkins, there were multiple admissions elsewhere, primarily for hypocalcemia, chronic diarrhea, and dehydration for which there were additional GI endoscopic procedures.
At age 20 yr, she was admitted on the adult medicine service at Johns Hopkins because of increased watery diarrhea for 2 months accompanied by malaise, weight loss, and fatigue but without fever or chills. She was a chronically ill-appearing woman who looked younger than her stated age. Her vital signs were normal, and her abdomen was nontender and revealed normal bowel sounds. Work-up for GI infection (bacterial overgrowth, Clostridium difficile, Giardia, other parasites and ova) was negative. Upper endoscopy and colonoscopy were again performed (for findings, see Results).
Histopathology
Multiple endoscopic biopsies taken at ages 3, 12, 17, and/or 20 yr were available from the esophagus, gastric body and antrum, duodenum, colon, and/or rectum. After fixation in buffered formalin, the specimens were paraffin embedded, sectioned in the standard manner, stained with hematoxylin and eosin (H&E) for routine study and, for GIECs, by an immunostain for chromogranin (Dako, Carpinteria, CA). In selected specimens the Grimelius silver stain was also used to demonstrate GIECs.
Gastric and duodenal biopsies were also stained by the combined periodic acid Schiff and Alcian blue methods for mucosubstances and fungal infection, and the gastric antrum was immunostained with antihuman gastrin for G cells. A modified Giemsa stain (Diff-Quik; Dade-Behring, Deerfield, IL) was used to detect Helicobacter pylori. An immunostain for H. pylori (Chemicon, Temecula, CA) was also used ad libitum to specifically rule out H. pylori infection.
As control specimens, the above stains were applied to histologically unremarkable biopsies from the upper and lower GI tracts of 3- to 18-yr-old male and female patients without APS I. In addition, gastric biopsies from a 54-yr-old male with adult-onset pernicious anemia, gastric atrophy, and intact GI ECs were used for comparison with the present case.
The densities of GIECs were semiquantified in blinded chromogranin-stained slides using the following grading scale: 0, no stained GIECs observed; ±, rare GIECs or indefinite staining; 1+, few GIECs in only scattered glands; 2+, reduced or absent GIECs in some glands; 3+, normal GIEC numbers present in most glands; and 4+, increased GIECs present in all glands and/or in the lamina propria.
Immunological studies
To test for the presence of anti-GIEC antibodies in the patients serum, frozen sections were prepared from unfixed segments of surgically resected normal stomach and duodenum that contained abundant endocrine cells. These were exposed to the patients serum followed by testing for uptake of antiendocrine cell antibodies by a standard technique (23) for indirect immunofluorescent study using fluorescein-conjugated polyclonal goat antihuman IgG (ICN/Cappel, Aurora, OH) and then evaluated by UV microscopy. The preparations were also used to examine parietal cells in the normal gastric test specimens for evidence of antiparietal cell antibody uptake from the patients serum.
| Results |
|---|
|
|
|---|
At age 12 yr, upper GI endoscopy showed mild gastritis with focal erythema and edema in the gastric antrum. No endoscopic changes were described in the gastric body, although histological evidence of atrophy was noted (see below). However, at age 20 yr, marked mucosal atrophy was seen to involve the entire stomach. The mucosa appeared flat and smooth with striking loss of rugae in the gastric body region. The esophagus and duodenum looked normal. Flexible sigmoidoscopy at age 12 yr revealed normal colonic and rectal mucosa, as did colonoscopy at age 20 yr except for mild widespread edema.
Histopathology
General. H&E-stained esophageal biopsies taken from the patient at ages 12, 17, and 20 yr were similar, revealing mild reactive epithelial changes as seen in gastroesophageal reflux. No further studies were done on these specimens.
The patients gastric body mucosa revealed a striking autoimmune-type atrophic gastritis with mild focal acute inflammation and moderate to marked chronic inflammation with focal intraepithelial lymphocytes (Fig. 1A
) that were shown by immunostaining (Fig. 1B
) to be mainly CD3 positive (i.e. T cells). In addition, parietal (oxyntic) cells were completely absent, and there was no intestinal metaplasia (Fig. 1A
), unlike in the gastritis of adult-onset pernicious anemia, in which intestinal metaplasia is almost universal (24).
|
Biopsies taken at age 3 yr from the duodenal bulb and third duodenum showed normal villous architecture, intact columnar epithelium, and no excess inflammatory cells in the lamina propria or epithelium. There was no evidence of candidal or other fungal infection in the duodenum by H&E, PAS/AB, or Diff-Quik stains. The ileal, colonic, and rectal mucosae were within normal limits by H&E staining at all ages studied.
GIECs
All biopsies of the patients GI tract available for study (gastric body and antrum, duodenum, ileum, colon, and rectum) showed total or near total absence of GIECs in chromogranin-immunostained biopsies at all studied ages (Figs. 2
, B and D, and 3). These observations were also confirmed by Grimelius silver staining in selected specimens. In contrast, control specimens from normal individuals stained with chromogranin immunostain always showed scattered GIECs that were found inserted between secretory and/or absorptive cells (Fig. 2
, A and C).
|
|
Immunohistology
The patients serum was shown to contain antibodies to GIECs when tested against normal endocrine cells in small intestine (duodenum) and against GIEC in normal gastric body mucosae (Fig. 4
, AC). Although a serological test for antiintrinsic factor was positive on one occasion, there was no evidence of serum antibody uptake by parietal cells or other gastric cell types in the current indirect immunofluorescence test (Fig. 4C
).
|
| Discussion |
|---|
|
|
|---|
The pernicious anemia-associated gastric findings in this patient are similar to those seen in adult-onset autoimmune gastritis insofar as both show atrophic gastritis with loss of acid-secreting parietal cells. But as seen in this patient, the two conditions can differ markedly in their pathogenesis. Absence of parietal cells in adult-onset pernicious anemia stems from autoimmune destruction of the parietal cells that leads in turn to hyperstimulation of gastrin secretion and resultant high serum gastrin levels. The elevated gastrin then causes secondary hyperplasia of gastric GIECs. Both features are lacking in this patient, who had serum gastrin levels only in the low normal range and absence of gastrin-secreting GIECs. This leads one to ask: if not immune mediated, what might cause the absence of parietal cells in APS I? One possibility is that the syndrome might include autoimmune destruction of parietal cells as another nonendocrine feature of APS I. Support for that possibility in this patient is limited to demonstration of some T cell infiltration in glands of the gastric body mucosa (Fig. 1B
). However, we saw no evidence of antibody uptake from the patients serum by intact parietal cells. A possible alternate explanation for absence of parietal cells is that they disappeared due to the low normal serum gastrin levels because gastrin, in addition to stimulating acid secretion, has a trophic function that stimulates replication of parietal cells. Thus, insufficient trophic stimulation by gastrin-secreting GIECs in our patient may have led to marked reduction of the parietal cell population.
Indirect support for a possible gastrin-related trophic deficiency in APS I is provided by the studies of Nagata et al. (28) in which cholecystokinin/gastrin receptor-deficient mice were generated by gene targeting. The resulting homozygous animals showed a remarkable atrophy of the gastric mucosa due to a decrease in parietal cells, suggesting that absence of gastrin stimulation could lead to loss of parietal cells such as that noted in our patient with APS I. These points clearly deserve further study.
Two groups of investigators have described serum autoantibodies to enzymes found in GIECs: tryptophan hydroxylase (21, 22) and histidine decarboxylase (28). Both groups demonstrated that autoantibodies to the enzymes were detected in the sera of the majority of patients with both APS I and GI dysfunction and that smaller proportions of serological positivity to autoantibody had occurred in their APS I patients without GI dysfunction. Those investigators also noted the absence or reduction of stainable GIECs in biopsy specimens from patients with the autoantibodies. The temporary reappearance of GIECs in the colorectum of our patient is consistent with the spontaneous but temporary histological and clinical recovery seen by Hogenauer et al. (20) in their patient with APS I and provides encouraging evidence that such restoration is possible.
Our observations give strong additional support for the conclusion that GIECs are included in the autoimmune-mediated destruction of endocrine organs in patients with APS I and that the widespread loss of these cells may play a significant role in the various GI dysfunctions that are often associated with this syndrome, including malabsorption and pernicious anemia.
| Acknowledgments |
|---|
| Footnotes |
|---|
First Published Online May 9, 2006
Abbreviations: AIRE, Autoimmune regulator gene; APS I, autoimmune polyglandular syndrome type I; GI, gastrointestinal; GIEC, GI endocrine cell; H&E, hematoxylin and eosin.
Received November 21, 2005.
Accepted May 2, 2006.
| References |
|---|
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |