Functional and structural aspects of the epithelial lymphocyte, with implications for coeliac disease and tropical sprue

Tropical sprue

Tropical sprue (TS) is a clinical entity of unknown etiology characterized by an acquired chronic diarrheal illness and malabsorption that affects indigenous inhabitants and expatriates, either long-term residents or short-term visitors, in the tropical countries. The exact pathogenetic sequence of TS remains incompletely characterized. Bacterial overgrowth, disturbed gut motility, and hormonal and histopathologic abnormalities contribute to the development of TS in a susceptible host. Treatment with tetracycline and folate is effective in some patients, although relapses after treatment are common. Research in the areas of microbial factors, pathogenesis, immunogenetics, and hormonal and immune regulation, using modern diagnostic techniques, may be able to settle some of the unanswered issues and open new venues for diagnosis, prevention, and treatment of tropical sprue.

Changes in jejunal mucosa after long-term feeding of germfree rats with gluten

BACKGROUND

Coeliac disease (CD) or gluten-sensitive enteropathy is a chronic gastrointestinal disease of children and adults. An experimental model using inbred germfree rats has been developed to study the effects of intragastric gliadin application on intestinal mucosa.

METHODS

AVN strain Wistar rats (inbred F 87)-germfree were used. Gliadin was applied by intragastric probe from birth until day 63 (0.5-5 mg of gliadin per immunization). Intraepithelial lymphocytes (IEL) were separated from the jejunum, and surface marker characterization was performed using flow cytometry. Isolated IEL were labelled with fluorescein isothiocyanate and injected into control jejunal loops. After 1 h and 6 h the abdominal cavity was reopened. The samples of jejunum were fixed.

RESULTS

Prolonged application of gliadin led to the shortening of jejunal villi, crypt hyperplasia, increased number of mitoses in the crypt epithelium, and increased number of IEL-characteristic CD8+, RGL-1+, and TcR alpha/beta +. Transfer of IEL separated from rats fed with gliadin into the intestinal loops of untreated rats led to tight junctions in the enterocytes of the intestinal loops. The IEL isolated from controls (albumin-treated) induced no mucosal changes in intestinal loops.

CONCLUSION

These data suggest that IEL isolated from gliadin-treated rats transfer mucosal damage and that gluten-induced enteropathy has an autoimmune component.

Reconstitution of cultured intestinal epithelial monolayers with a mucosal-derived T lymphocyte cell line. Modulation of epithelial phenotype dependent on lymphocyte-basolateral membrane apposition

In vivo, epithelial cells which line the intestine are intimately associated with lymphocytes, termed intraepithelial lymphocytes. Previous studies have demonstrated that intraepithelial lymphocytes are present in the uninflamed mucosa, and become especially prominent in various human enteropathies including coeliac disease, tropical sprue, dermatitis herpetiformis, and giardiasis. Using the intestinal crypt cell line T84, and a previously well-defined human mucosa-derived lymphocyte (MDL) line with phenotypic features similar to (but not specific for) intraepithelial lymphocytes, we describe a co-culture model to study the functional sequellae of MDL-T84 cell interactions in vitro. A co-culture method was defined which permitted reconstitution of the paracellular spaces of physiologically confluent epithelial monolayers with MDL. Such co-cultures thus mimicked the correct geometry of intraepithelial lymphocytes-epithelial cell interactions. The presence of physiologically positioned MDL brought about specific and dramatic effects on intestinal epithelial monolayer function. In a dose-dependent fashion, the presence of MDL significantly attenuated barrier function (expressed as a decrease in monolayer resistance), decreased epithelial electrogenic Cl- secretion, and modulated epithelial-neutrophil interactions. Such effects were not reproduced in monolayers similarly reconstituted with inert polystyrene beads equivalent in size to MDL. These MDL-elicited effects on epithelial function specifically required direct MDL apposition to the epithelial basolateral membrane. Furthermore, this specific form of MDL-epithelial basolateral contact released soluble factors which were able to confer the MDL-reconstituted phenotype on virgin epithelial monolayers in the absence of MDL. We have previously shown that many aspects of the MDL converted epithelial phenotype described here can be induced by IFN-gamma. While IFN-gamma, a cytokine produced by many lymphocytes including intraepithelial lymphocytes, was detectable in conditioned supernatants from co-cultures, it existed at concentrations insufficient to fully explain the physiologic effects observed here.

Studies of intestinal lymphoid tissue. XIV--HLA status, mucosal morphology, permeability and epithelial lymphocyte populations in first degree relatives of patients with coeliac disease

Fifty two first degree relatives of patients with coeliac disease were investigated for HLA status, small intestinal permeability, and mucosal morphology together with the size of the intraepithelial lymphocyte pool and indices of lymphocyte activation, in an attempt to identify genetically determined markers of the disease. Thirty eight per cent of these subjects had increased intraepithelial lymphocyte populations and a highly significant association with HLA-DR3 compared with controls. Their intestinal permeability to 51chromium-labelled ethylenediamine tetraacetate was invariably normal and there was no evidence of abnormal mucosal architecture, increased crypt cell mitotic activity or lymphocyte 'activation'. Although increased intraepithelial lymphocyte counts clearly do not cause alterations in intestinal structure or function, it is likely that this parameter together with the HLA-DR3 status identifies a genetically determined predisposition to the disease which may only become clinically evident with larger doses of ingested gluten.

Grains of truth: evolutionary changes in small intestinal mucosa in response to environmental antigen challenge

Images

Studies of intestinal lymphoid tissue. XII. Epithelial lymphocyte and mucosal responses to rectal gluten challenge in celiac sprue

The immunopathologic, structural, and functional changes within rectal mucosa of known celiac sprue subjects were quantitated during local challenge with a peptic-tryptic digest of gluten. In the celiac sprue patients challenged with 2 g of digest, major effects occurred in lamina propria, submucosa, and local microvasculature. The lamina propria swelling was biphasic, starting 1-2 h after challenge with widespread extravascular deposition of fibrinogen, indicative of increased microvascular permeability, receding by 24 h postchallenge. A rapid fall in mast cells together with granule discharge suggested their involvement in this response. The late-phase swelling (48-72 h) was preceded by a rapid influx of neutrophils and basophils, the latter showing evidence of degranulation beyond 72 h. Reestablishment of vessel lumina, a rise in mast cells, and loss of neutrophils indicated tapering of the inflammatory cellular cascade by 96 h. Lymphocytes, first seen to enter the lamina by 2 h postchallenge, increased progressively, thereby resulting in substantial infiltration between 36 and 96 h. A marked rise in epithelial lymphocytes, maximal at 6-8 h, waned by 24 h. Volumes of surface and crypt epithelium remained constant throughout. In another challenge series with 4 g of gluten digest, electrical potential difference across rectal mucosa decreased significantly 12 h postchallenge, but the associated decreases in net sodium and chloride absorptive fluxes were insignificant. It is concluded that rectal mucosa is sensitized to gluten in celiac sprue disease and thus offers a promising and convenient in vivo substrate for investigative and diagnostic purposes.

Jejunal enteropathy associated with human immunodeficiency virus infection: quantitative histology

Jejunal biopsy specimens from 20 human immunodeficiency virus (HIV) positive male homosexual patients were analysed and compared with those of a control group to determine whether the abnormalities were caused by the virus or by opportunistic infection. The degree of villous atrophy was estimated with a Weibel eyepiece graticule, and this correlated strongly with the degree of crypt hyperplasia, which was assessed by deriving the mean number of enterocytes in the crypts. The density of villous intraepithelial lymphocytes fell largely within the normal range, either when expressed in relation to the number of villous enterocytes or in relation to the length of muscularis mucosae. Villous enterocytes showed mild non-specific abnormalities. Pathogens were sought in biopsy sections and in faeces. Crypt hyperplastic villous atrophy occurred at all clinical stages of HIV disease and in the absence of detectable enteropathogens. An analogy was drawn between HIV enteropathy and the small bowel changes seen in experimental graft-versus-host disease. It is suggested that the pathogenesis of villous atrophy is similar in the two states, the damage to the jejunal mucosa in HIV enteropathy being inflicted by an immune reaction mounted in the lamina propria against cells infected with HIV.

The immunopathology of the small intestinal reaction in gluten-sensitivity

Computerised image-analysis was used to define the spectrum of immunopathological changes in small intestinal mucosa in established celiac sprue disease; dermatitis herpetiformis; 1 degree relatives of celiac sprue patients, and treated celiac sprue patients challenged with varying doses of a peptic-tryptic digest of gluten. Typically, in flat ('Type 2') lesion there was a reduced number of large, mitotically active lymphocytes in surface epithelium, but an increased lymphocyte population in crypts. In approximately 50% untreated DH patients and in 20% 1 degree celiac sprue relatives, mucosal architecture was well-preserved although surface (villous) epithelium contained an expanded population of small, non-mitotic lymphocytes ('Type 1' lesion), with or without crypt hyperplasia. Similar changes were also induced by small dose gluten challenge. Larger dose challenges caused a progression from a Type 1 to a Type 2 lesion during a 5 day period of observation. In addition, observations on a few patients over 2-4 years showed a similar sequence of mucosal changes. A major feature of this sequence was the early appearance of crypt hypertrophy, before significant villous flattening had occurred. These changes parallel T lymphocyte-mediated graft- versus-host reactions in animals, suggesting that the specific immunopathologic features seen in gluten sensitivity are fundamentally cell-mediated in type, the degree of change probably dependent on host genetic factors. Finally, these data show that in becoming flat the mucosa must initially pass through the earlier Type 1 lesion in which crypt hypertrophy is a prominent response.

Reduction in uterine intra-epithelial lymphocytes during early gestation in pigs

The numbers of intra-epithelial lymphocytes per 5 mm of uterine epithelium were counted in cyclic (n = 18) and pregnant (n = 18) gilts on days 10, 13, 16 and 19 postestrus. In cyclic animals these cells increased (P less than 0.01) from 26.4 +/- 1.1 per 5 mm at the mid-luteal phase (day 10) to 47.7 +/- 1.5 per 5 mm at proestrus (day 19). Intra-epithelial lymphocytes were significantly (P less than 0.01) fewer at all stages in pregnant females and were reduced from 17.7 +/- 1.5 on day 10 to 6 +/- 1.4 per 5 mm on day 19. The significant reduction in intra-epithelial lymphocytes during the second and third weeks of gestation indicates that the developing pig conceptus can reduce the numbers of potentially damaging cells at the placental interface.

Studies of the rectal mucosa in coeliac sprue: the intraepithelial lymphocyte

The dynamics of the rectal surface epithelial lymphocyte and leucocyte response to wheat, gluten, and gliadin enema challenges in control individuals and in patients with coeliac sprue in remission is shown. There is a clear increase in intraepithelial lymphocytes and polymorphonuclear (PMN) leucocytes in response to these enemas in coeliac sprue, but not in controls. The peak response was at eight hours and cleared within 24 hours. There was no change in the crypt epithelium. These data add further support to the role of wheat, gluten, and gliadin in the pathogenesis of coeliac sprue, at least in the rectal mucosa.

Gastrointestinal food allergy and its role in large domestic animals

The significance of food allergy as a primary cause for gastrointestinal disturbances in domestic animals, especially calves and piglets, is discussed. The immunological backgrounds and pathogenesis are described in some detail. The clinical and pathological manifestations in animals are related to those in man. Diagnostic possibilities, therapy and prevention, as far as known in animals, are mentioned and, based on human experiences, further extensions are proposed.

Studies of intestinal lymphoid tissue. X-observations on granular epithelial lymphocytes (gEL) in normal and diseased human jejunum

A proportion of epithelial lymphocytes in various mammalian species is characterised by cells containing cytoplasmic granules. We have studied the total number of granular lymphocytes within surface and crypt epithelium of jejunal mucosae (per 10(4) micron2 muscularis mucosae) from six groups of subjects, comprising (i) young healthy volunteers (ii) family relatives of known coeliac patients, patients with gastrointestinal disorders associated with either (iii) normal or (iv) "flat" mucosae, and groups of (v) untreated and (vi) treated patients with coeliac disease. There was no difference in the absolute number of gEL between the three control groups with normal mucosal architecture, the proportion of granular to total EL per unit of tissue varying between 30-40%. In untreated coeliac mucosae, there was a significantly increased population of gEL, compared with the same control groups (p less than 0.001): the ratio of granular to total EL approximated 65%, and did not differ from flat-control mucosae in which the proportion of gEL was 55%. On withdrawal of gluten, the absolute number of gEL fell significantly in comparison with the untreated coeliac group (p less than 0.05). To further evaluate the effect of gluten challenge, granular lymphocytes were monitored during a five-day period in groups of treated coeliac patients orally challenged with increasing doses (500-3000 mg) of a peptic-tryptic digest of gluten. A significant rise in the absolute number of granular lymphocytes occurred at 12 h, but without any deterioration in mucosal architecture.

Expression of major histocompatibility complex class II subregion products by jejunal epithelium in patients with coeliac disease

The MHC class II subregion products (HLA-DR), HLA-DP, and HLA-DQ) were located by immunofluorescence in serial sections of ethanol-fixed, paraffin-embedded jejunal mucosa from control subjects and patients with coeliac disease (CD). DR staining was seen in a granular luminal distribution and basolaterally on surface epithelial cells in both untreated and treated CD patients and in controls. In untreated CD the crypt epithelium was positive for DR almost to the bottom of the glands. This contrasted with virtually absent glandular DR staining in controls and weak staining including only the upper part of the crypts in 5 out of 11 treated patients. HLA-DP was present apically in the surface epithelium in all untreated patients, in 5 out of 11 treated patients, and in 4 out of 11 controls. HLA-DQ appeared only in three untreated patients and was restricted to patches of surface epithelium. The number of intraepithelial T lymphocytes per millimetre of surface epithelium was significantly higher in untreated than in treated CD patients or controls; it was also significantly higher in specimens with epithelial DP expression than in those without. This suggested that intraepithelial lymphocytes modulate epithelial class II expression.

Histological changes associated with wheat protein antibodies in the absence of villous atrophy

A retrospective study was conducted to assess the association of alpha-gliadin antibodies with intraepithelial lymphocyte counts. Twelve subjects with apparently normal small intestinal histology and raised alpha-gliadin antibody titres had significantly increased intraepithelial lymphocyte counts (42 (SEM) 5.9) when compared with 16 subjects with normal alpha-gliadin antibody titres (17 (3.2); p less than 0.001). These findings show that in the absence of gross pathology raised alpha-gliadin antibody titres are associated with increased numbers of intraepithelial lymphocytes and may reflect continuous immunological processes in the small intestine.

Morphometric analysis of small intestinal mucosa. III. The quantitation of crypt epithelial volumes and lymphoid cell infiltrates, with reference to celiac sprue mucosae

With the aid of computerised image-analysis, morphometric techniques were used to measure the volumes of crypt epithelium, with reference to a constant test area (10(4) microns2) of muscularis mucosae, in untreated and treated celiac sprue mucosae in comparison with four other groups of control jejunal specimens. Crypt epithelial lymphocyte populations were also analyzed in terms of absolute numbers (N), mean nuclear (DN) and cytoplasmic (DCYT) diameters, and mean nuclear (VN) and cell (VCELL) volumes. Untreated celiac sprue crypts, despite a 3-5 fold increase in volume over control mucosae, contained a markedly expanded population of lymphocytes which was localised predominantly to the upper crypt regions and comprised approximately 6% large lymphocytes (DN greater than 6 microns: DCYT greater than 9 micron). These changes were entirely reversed by dietary treatment and hence were considered to be gluten-driven. The infiltrate might reflect the expression of gluten receptors on maturing upper crypt enterocytes: but why these lymphocytes do not cause any apparent injury to crypt epithelium (unlike surface epithelium) or influence its well-known compensatory response, remains unclear.