CD36 and lipid metabolism in the evolution of atherosclerosis.

Background: CD36 is a multi-functional class B scavenger receptor, which acts as an important modulator of lipid homeostasis and immune responses. Sources of data: This review uses academic articles. Areas of agreement: CD36 is closely related to the development and progression of atherosclerosis. Areas of controversy: Both persistent up-regulation of CD36 and deficiency of CD36 increase the risk for atherosclerosis. Abnormally up-regulated CD36 promotes inflammation, foam cell formation, endothelial apoptosis, macrophage trapping and thrombosis. However, CD36 deficiency also causes dyslipidemia, subclinical inflammation and metabolic disorders, which are established risk factors for atherosclerosis. Growing points: There may be an 'optimal protective window' of CD36 expression. Areas timely for developing research: In addition to traditionally modulating protein functions using gene overexpression or deficiency, the modulation of CD36 function at post-translational levels has recently been suggested to be a potential therapeutic strategy.

A clinicopathological study of serotonin of sigmoid colon mucosa in association with chronic symptoms in uncomplicated diverticulosis.

INTRODUCTION: Neurotransmitter imbalance is hypothesised as a pathogenetic mechanism in several bowel conditions. We previously reported increased 5-HT in the sigmoid mucosa of colon resected for complicated diverticular disease (DD). We aimed to identify if abnormal 5-HT expression is associated with symptoms of uncomplicated DD. METHODS: This was a prospective, comparative study and follow-up survey of symptoms. We examined the differences in 5-HT between DD patients and controls, as well as the presence of bowel symptoms at time of endoscopy and also 2 years later. Sigmoid biopsies were collected at colonoscopy. Immunohistochemical staining for 5-HT cells was performed. RESULTS: Eighty-seven patients were recruited, 37 (42.5 %) DD and 50 (57.5 %) controls. No patients underwent surgery. There was no significant difference in total mean number of 5-HT-positive cells in DD compared to controls or between patients and controls with abdominal symptoms. Forty-one patients (47.1 %) responded to questionnaires at median 57.8 months from biopsy. Eighteen (43.9 %) were DD and 23(56.1 %) controls. 5-HT counts showed no significant association to symptom persistence. DISCUSSION: Although 5-HT expression has previously been found to be increased in complicated DD in whole bowel-resected specimens, the same is not confirmed on colonic mucosal biopsies. This raises the suggestion that 5-HT may be involved in the development of acute complications but may not be involved in the pathogenesis of chronic symptoms.

Histopathological features for coexistent invasive cancer in large colorectal adenomatous polyps.

BACKGROUND: Histopathological features associated with coexistent invasive adenocarcinoma in large colorectal adenomas have not been described. This study aimed to determine the association of histopathological features in areas of low-grade dysplasia with coexistent invasive adenocarcinoma. METHODS: High-grade lesions (containing high-grade dysplasia or adenocarcinoma) from a cohort of large (at least 20 mm) colorectal adenomas removed by endoscopic resection were subjected to detailed histopathological analysis. The histopathological features in low-grade areas with coexistent adenocarcinoma were reviewed and their diagnostic performance was evaluated. RESULTS: Seventy-four high-grade lesions from 401 endoscopic resections of large adenomas were included. In the low-grade dysplastic areas, a coexistent invasive adenocarcinoma was associated significantly with a cribriform or trabecular growth pattern (P < 0.001), high nuclear grade (P < 0.001), multifocal intraluminal necrosis (P < 0.001), atypical mitotic figures (P = 0.006), infiltrative lesion edges (P < 0.001), a broad fibrous band (P = 0.001), ulceration (P < 0.001), expansile nodules (P < 0.001) and an extensive tumour-infiltrating lymphocyte pattern (P = 0.04). Lesions with coexistent invasive adenocarcinoma harboured at least one of these features. The area under the receiver operating characteristic curve (AUROC) for coexistent invasive adenocarcinoma, using frequencies of adverse histopathological factors in low-grade areas, was 0.92. The presence of two or more of these adverse histopathological features in low-grade areas had a sensitivity of 86 per cent and a specificity of 84 per cent for coexistent invasive adenocarcinoma. CONCLUSION: Several histopathological features in low-grade dysplastic areas of adenomas could be predictive of coexistent adenocarcinoma.

Soluble factors in tolerance and contact sensitivity to 2,4-dinitrofluorobenzene in mice. III. Histocompatibility antigens associated with the hapten dinitrophenol serve as target molecules on 2,4-dinitrofluorobenzene-immune T cells for soluble suppressor factor.

Previous studies have shown that suppression of 2,4-dinitrofluorobenzene (DNFB) contact sensitivity by soluble suppressor factor (SSF) requires that the donor of immune lymph node (LN) cells and of SSF share either the H-2K and/or H-2D region of the major histocompatibility complex. Thus, target or acceptor molecules for SSF appear to be coded for by genes within the H-2K and H-2D loci. Experiments were done to investigate the nature of these target molecules and to determine what cell types expressed them. It was found that purified lymph node T cells are suppressed by SSF indicating that T cells express the acceptor molecules. Adsorption experiments showed that the only cells capable of adsorbing the suppressor factor are DNFB-immune T cells from donors which share with the factor-producing strain either the H-2K or H-2D locus. This adsorption can be specifically blocked by pretreating the immune LN cells with antibodies directed against H-2K and/or H-2D determinants or against the hapten DNP but not by antibodies against Ia or theta-antigens. Collectively, these results indicate that the target molecules are expressed only by DNFB-immune T cells and are comprised of histocompatibility antigens associated with DNP.

Antigen receptors on murine T lymphocytes in contact sensitivity. III. Mechanism of negative feedback regulation by auto-anti-idiotypic antibody.

Contact sensitivity (CS) to 2,4-dinitrofluorobenzene (DNFB) is maximal 6 d after sensitization but declines rapidly. Previous studies have shown that this rapid decline is due to auto-anti-idiotypic (anti-Id) antibodies produced by the host. The present study was done to investigate the mechanism(s) involved in his down-regulation of the effector phase of the CS reaction. Using transfer of CS to mimic the natural effector phase, we found that the inhibition of transfer by treating DNFB-sensitized lymph node (LN) cells with either auto-anti-Id or syngeneic anti-Id serum is complement (C) independent. This inhibition requires Ia+ T cells in the immune population. Depleting immune LN cells of Ia+ T cells rendered them insensitive to inhibition by anti-Id alone, although the same population is inhibited by anti-Id plus C. This cell population is rendered sensitive to inhibited by anti-Id alone by addition of untreated DNFB-sensitized LN cells, but not by addition of normal LN cells. Further studies showed that the suppression of immunity by anti-Id-activated Ia+ T cells is not systemic, but rather occurs locally at the skin test site and is antigen nonspecific. We interpret these data to indicate that the natural regulation of CS to DNFB by auto-anti-Id antibodies is an active process that involves a negative feedback regulatory loop.

Antigen receptors on murine T lymphocytes in contact sensitivity. I. Functional inhibition of effector T cells by monovalent 2,4-dinitrophenol: implication for a two-receptor model.

Experiments were done to investigate the nature of the antigen receptor on lymph node(LN) T cells from mice sensitized to 2,4-dinitrofluorobenzene (DNFB). LN cells or purified T cells were treated in vitro with monovalent or different multivalent 2,4-dinitrophenol (DNP) ligands. The effect of this treatment was measured by testing the ability of the cells to transfer contact sensitivity (CS) to DNFB into naive recipients. We found that treatment of the T cells in vitro with either epsilon-DNP-L-lysine or DNP-protein conjugates inhibits the transfer of CS in a dose-dependent way. The inhibition is hapten specific and is not mediated by activation of suppressor cells. Inhibition of the T cells by hapten in vitro is rapid (15-30 min) and temperature independent but requires divalent cations in the treatment medium. In addition, it was found that hapten-treated T cells are unable to adsorb specific anti-idiotype antibody, and this inhibition of adsorption is hapten specific. Collectively, these data indicate that DNFB-immune T cells express a receptor specific for the hapten DNP and provide evidence that supports a two-receptor model for T cell recognition of antigen.

Functional anatomy of the lymphocyte in immunological reactions in vitro.

The motile lymphocyte in vitro has a prominent "tail" that becomes a means of "attachment" to other cells and debris during interaction. The term "uropod" is proposed to designate this specialized cytoplasmic projection which appears totally different, anatomically and functionally, from the pseudopods. Observations of lymphoblasts during mitosis indicate that the uropod is formed immediately following mitosis at the point of final cytoplasmic connection between daughter cells, a fact that may prove significant as lymphocyte function is better understood. In the mixed leukocyte reaction the lymphocyte interacts with macrophages, cell debris, and lymphoblasts via the uropod, suggesting that stimulatory material may be acquired through this specialized appendage. Lymphoblast-lymphocyte interaction is noteworthy and implies that immunologically committed cells may be mustered through horizontal as well as vertical processes: horizontally by lymphoblast-lymphocyte interaction and vertically by mitosis of transformed lymphoblasts. The possible relevance of these in vitro observations to lymphocyte functions in vivo is discussed.

Immunologic reactions to haptens on autologous carriers. I. Participation of both thymus-derived and bone marrow-derived cells in the secondary in vitro response.

Both thymus-derived (T) and bone marrow-derived (B) lymphocytes participate in the response to a hapten 4-hydroxy-3-iodo-5-nitrophenylacetic acid (NIP), coupled to a nonimmunogenic isologous carrier, mouse gamma globulin (MGG). Spleen cells from mice immunized with NIP-MGG show increased DNA synthesis in vitro when cultured with NIP-MGG. The participation of and requirement for T cells in the response was demonstrated by treating the spleen cells with anti-theta serum. This treatment resulted in a 77% inhibition of the antigen response. Furthermore, adoptively transferred normal thymus cells could be specifically "activated" by NIP-MGG in vivo and they responded secondarily to the antigen in vitro. The active participation of B cells in the secondary response was demonstrated by passing the immune spleen cells through a column coated with polyvalent anti-MGG serum. Column filtration reduced the number of NIP-specific plaque-forming cells and NIP-specific rosette-forming cells (both functions of B cells) and produced a 47% inhibition of the NIP-MGG response. The ability of the cells to respond to phytohemagglutinin (PHA) was not affected by column filtration showing that T cells were not being selectively removed. The participation of B cells in the in vitro NIP-MGG response was also shown by treatment of the spleen cells with antiserum specific for MGG and MGG determinants. B cells were removed by treatment with anti-IgM or polyvalent anti-MGG serum plus complement, resulting in a respective 46 and 49% inhibition of the response to NIP-MGG. (Treatment with anti-IgM serum had no effect on T cells.) The contribution of the hapten NIP to stimulation of T cells was investigated using NIP-MGG-activated thymus cells. These activated T cells responded in vitro very well to the NIP-MGG complex but not to the MGG carrier alone demonstrating the requirement of the hapten for T cell stimulation. The response was also partially inhibited (41%) by incubating the activated cells with NIP coupled to a single amino acid (epsilon-aminocaproic acid) before addition of NIP-MGG. These results demonstrated that T cells recognize the hapten NIP when it is coupled to the isologous carrier MGG.

Immunity tolerance to a hapten (NIP) coupled to an isologous carrier (mouse gamma globulin).

A hapten, 4-hydroxy-3-iodo-5-nitrophenylacetic acid (NIP) when coupled to isologous mouse gamma globulin (MgammaG) elicits a hapten-specific immune response in mice if administered in Freund's complete adjuvant. This response is measurable by the capacity of the sera to bind N(125)IP, by detection of NIP-specific plaque-forming cells (B cells), and by in vitro secondary type antigen-driven DNA synthesis (T cells and probably B cells). The in vitro response requires both the hapten and carrier since neither by itself is capable of stimulating the spleen cells. This same antigen gives rise to hapten-specific tolerance when given in the soluble form. Mice pretreated with soluble NIP-MgammaG and challenged with NIP coupled to a heterologous carrier give a normal antibody response to the carrier but have barely detectable levels of antibody to NIP. Spleen cells from mice made tolerant to NIP-MgammaG do not respond in vitro with increased DNA synthesis. This implies that thymus-derived cells as well as bone marrow-derived cells are involved in hapten-specific tolerance.

Active suppression of 1-fluoro-2,4-dinitrobenzene-immune T cells. Requirement of an auxiliary T cell induced by antigen.

We investigated T-T cell interactions in the suppression of contact sensitivity. Suppressor cells that block the efferent limb of sensitivity (Ts-eff) can inhibit the passive transfer of contact sensitivity mediated by 1-fluoro-2,4-dinitrobenzene immune cells (T DH). But, Ts-eff cannot block the passive transfer of TDH which comes from cyclophosphamide (Cy) pretreated sensitized mice. We interpret these results to indicate that lymph node cells from sensitized mice contain not only TDH but also another intermediate cell which is required for the suppression of TDH by Ts-eff. This intermediate cell is sensitive to cyclophosphamide and requires antigen activation for its development. It is sensitive to adult thymectomy and anti-brain associated theta serum and is therefore designated as an auxiliary T-suppressor cell (Ts-aux). It is not sensitive to splenectomy and it carries I-J determinants. Ts-aux are required for the activity of suppressors of the efferent limb (Ts-eff) but not of suppressors of the afferent limb (Ts-aff). Thus, in the feedback loops in contact sensitivity, the generation of Tdh is coordinated with the development of auxiliary Ts which are essential for the suppression of those TDH.

Induction of tolerance to 1-fluoro-2,4-dinitrobenzene contact sensitivity with hapten-modified lymphoid cells. II. Selective tolerance in F1 mice of T cell subsets recognizing 1-fluoro-2,4-dinitrobenzene associated with parental major histocompatibility complex antigens.

F1 animals were tolerized to 1-fluoro-2,4-dinitrobenzene (DNFB) contact sensitivity with parentally derived, in vitro hapten-modified spleen cells. This tolerant state was found, upon adoptive transfer to naive parental strain recipients, to affect only that T cell subpopulation that recognized the parental haplotype of the cell used as the tolerogen, and did not inhibit the ability of the remaining T cell subset to confer immunity. This demonstrates that this tolerant state involves the inactivation of a cell required for the expression of contact sensitivity by recognizing DNFB in association with self major histocompatibility complex gene products.

A simple and sensitive ELISA to detect immune (gamma) interferon induced I-A on a macrophage line.

A convenient and sensitive cell surface ELISA is described to detect immune (gamma) interferon induced I-A determinants on a murine macrophage line, P388D1. The ELISA is performed on monolayers of P388D1 cells grown in 96-well culture plates for 2 days in the presence of recombinant gamma IFN or supernatants from cultures of antigen-activated T cells. After glutaraldehyde fixation, the cultured cells are overlayed with fetal calf serum to block nonspecific antibody binding during the assay. A double sandwich technique employing a monoclonal anti-I-Ad (MKD6) antibody followed by peroxidase-conjugated goat anti-mouse gamma chain antibody is then used to detect the presence of I-A on the cell monolayers. Detection of I-A induced by both r gamma IFN and the supernatant from an antigen-stimulated T cell line was highly reproducible and sensitive using this method. Furthermore, the assay is easily performed and many sample supernatants can be rapidly screened for this activity. The assay is used by this laboratory to detect the antigen-stimulated production of gamma IFN by T cell clones and hybridomas.

Glomerular structures and lipids in progressive renal disease.

In the last few years, remarkable advances have been made in the understanding of lipoprotein metabolism in the pathogenesis of renal disease in animal models and in vitro cell culture. Central to this work is the problem of the progression of renal disease in humans. This review recapitulates the theory (Lancet 1982; II: 1309-1312) that the progression of disease depends in part on the damage inflicted on the glomerulus by lipoproteins. The glomerular environment of high or low pressure, basement membrane damage, and destruction or damage of the mesangial and epithelial cells permits the filtration of protein, the consequence of which is hyperlipidemia. Whatever the therapeutic measures employed, if proteinuria persists, hyperlipidemia will follow. This suggest that lipoprotein toxicity may contribute to the final common path of renal damage in progressive renal disease. "Lipoprotein toxicity" in arteries is called atherosclerosis, but this term ignores the complexity of the glomerulus and the possible tubular damage that might be caused by filtered lipoprotein. It is clear there is insufficient knowledge of the metabolism of the damaged kidney to confidently attribute the pathology of progression of disease to any single process.

The enhancing and sensitizing effects of donor blood components, including dendritic cells, in a rat renal allograft model.

In the DA-to-Lewis renal allograft model, donor whole blood enhanced renal allograft survival (14.5 +/- 7.6 days versus 6.9 +/- 0.6 days in controls [P less than 0.01]). The effect of individual cell components given in numbers equivalent to those present in the enhancing volumes of donor whole blood was studied. Immunization with donor red cells alone produced greater enhancement than that produced by whole blood (36.14 +/- 19.5 days [P less than 0.01]). B lymphocytes also enhanced allograft survival (16.0 +/- 3.9 days [P less than 0.01]). Although slight enhancement was observed with platelets (8.5 +/- 0.6 days) and 10(5) dendritic cells (8.4 +/- 0.5 days), in terms of allograft function dendritic cell immunization produced evidence of dose-dependent accelerated rejection. A similar finding was obtained with donor T cell immunization. Donor plasma had no effect. We conclude that, although donor blood has an overall enhancing effect on renal allograft survival in this model, the sensitizing and enhancing effects can be ascribed to individual types of cells.

In vitro cyclosporine toxicity. The effect of verapamil.

The epithelial cell line LLC-PK1, which expresses many proximal tubular characteristics, was used to investigate the relationship between calcium, the calcium channel blocker verapamil, and cyclosporine toxicity. The LLC-PK1 cells took up cyclosporine when this was added in a concentration of 2 micrograms/ml, and this uptake was maximal at 30 min (112 +/- 3 ng cyclosporine/mg cell protein). At 12 micrograms/ml it inhibited the sodium glucose cotransporter, as assessed by phlorizin-inhibitable 14C-alpha-methyl glucopyranoside (alpha-MG) uptake (control 37.2 +/- 6.3, 12 micrograms/ml 21.2 +/- 1.1 mumol/hr/mg protein). Cyclosporine at 2 micrograms/ml did not affect cell growth after 5 days (control 945 +/- 60 micrograms cell protein per 25 cm2 flask, 2 micrograms/ml cyclosporine/ml 1046 +/- 32 micrograms protein/flask), even in the presence of 7.6 mM ionized calcium (862 +/- 37 micrograms protein/flask). Cyclosporine at 12 micrograms/ml inhibited cell growth (286 +/- 27 micrograms protein/flask), and raising the ambient ionized calcium concentration to 7.6 mM reduced cell growth further (91 +/- 6 micrograms protein/flask). Cyclosporine at concentrations of 2 and 12 micrograms/ml produced increasing cell vacuolation, as seen in vivo. Short-term uptake of 2 micrograms/ml cyclosporine could be inhibited by 1.0 mM and 0.5 mM verapamil (49 +/- 9.5 and 71 +/- 6.4 ng cyclosporine/mg cell protein, respectively, at 30 min). However, in the presence of 2 micrograms/ml cyclosporine 0.1 mM verapamil was toxic to the cells grown over five days (44 +/- 5 micrograms protein/flask). At 0.01 mM verapamil was not toxic to cell growth (921 +/- 29 micrograms protein/flask), but raising the medium calcium to 7.6 mM reduced cell growth (652 +/- 96 micrograms/ml). Inhibition of cyclosporine uptake did not occur with 0.01 mm verapamil (control 145.6 +/- 12.3 vs. 0.01 mM verapamil 150.4 +/- 3.8 ng cyclosporine/mg cell protein). The LLC-PK1 cell line represents a good in vitro model for cyclosporine renal tubular toxicity, as the in vivo observation of glycosuria and proximal tubular cell vacuolation in cyclosporine nephrotoxicity can be reproduced. In vitro this is shown to be associated with inhibition of sodium-dependent glucose cotransport. Verapamil inhibited cyclosporine uptake, but only at concentrations that were toxic to the cells. Verapamil potentiated rather than reduced the increased cyclosporine toxicity produced by increasing the medium calcium concentration. The suggested protective effect of verapamil against cyclosporine nephrotoxicity is therefore unlikely to be due to inhibition of cyclosporine uptake or of calcium entry into proximal tubular cells.

Lipiduria in renal disease.

This article reviews the published data on lipiduria in both health and disease. Small amounts of lipid appear in the urine under normal circumstances but, in the nephrotic syndrome in humans, there is also a considerable amount of high-density lipoprotein in the urine as well as smaller amounts of other lipoproteins. Potential tubular re-uptake mechanisms for lipoproteins have been demonstrated in both animal and cell-culture models. In humans, there is no direct evidence for these specific re-uptake mechanisms--it is only through specific staining of renal biopsies for apolipoproteins that the presence of such mechanisms in intracellular vesicular structures is suggested. It is possible that lipoprotein filtration and re-uptake by the tubule are important mechanisms in tubular injury.

Intraplatelet serotonin, beta-thromboglobulin, and histamine concentrations and thromboxane A2 synthesis in renal disease.

Intraplatelet serotonin (5-HT), beta-thromboglobulin (beta-TG), and histamine content as well as platelet total thromboxane A2 (TXA2) synthesizing capacity were measured in 53 patients with chronic renal disease: nephrotic syndrome (n = 18); end-stage renal failure (ESRF; n = 13); continuous ambulatory peritoneal dialysis (CAPD; n = 9); hemodialysis (HD; n = 13). These indices of platelet function were correlated with plasma total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglyceride (TG) concentrations. When compared with controls, intraplatelet 5-HT was significantly reduced in all patient groups studied and beta-TG was diminished in all patient groups except CAPD. Total platelet TXA2 synthesizing capacity was increased in ESRF and HD groups. Intraplatelet histamine content was not altered in any of the patient groups studied. There was a significant inverse correlation between intraplatelet 5-HT content on the one hand and plasma TC, LDL-C, and TG on the other. The depletion of intraplatelet 5-HT and beta-TG and the increase in total TXA2 synthesizing capacity are consistent with platelet activation in chronic renal disease. The correlation between these indices of platelet activation and TC, LDL-C, HDL-C, and TG suggests that changes in the concentrations of these lipids may contribute to the activation of platelets in these conditions.

Glycosylated haemoglobin in chronic renal failure and after renal transplantation.

Haemoglobin A1 (HbA1) was determined by ion-exchange chromatography in 37 normoglycaemic patients with chronic renal failure (CRF) and 26 with successful renal transplants. Blood glucose concentrations in patients with CRF were similar to those in controls, and there was a significant correlation between fasting blood glucose concentration and HbA1 in these groups. HbA1 in patients with CRF was, however, significantly lower than that in control subjects. Concentrations of HbA1 in patients on haemodialysis, peritoneal dialysis, and those with steady state CRF prior to dialysis were not significantly different from each other. Whereas patients with successful renal transplants of greater than 3 months' duration had HbA1 concentrations indistinguishable from controls, HbA1 in patients with transplants of shorter duration were significantly lower. These observations are suggestive of a shortened erythrocyte survival in CRF per se. Furthermore, these results indicate: (a) the inadequacy of HbA1 in monitoring the quality of diabetic control in patients with CRF, and (b) the absence of a specific effect of dialysis on HbA1, and the restoration to normal HbA1 after successful renal transplantation.

Plasma histamine in patients with chronic renal failure and nephrotic syndrome.

Plasma histamine concentrations were measured using a commercially available monoclonal antibody radioimmunoassay in 38 patients with nephrotic syndrome, end stage renal failure, those receiving haemodialysis, and those receiving continuous ambulatory peritoneal dialysis to determine whether histamine may mediate damage to glomerular capillaries and arterial endothelium. Plasma histamine concentrations were significantly increased in all four patient groups when compared with those of controls and were the highest in two patients with pruritus. Raised plasma histamine concentrations in such patients are consistent with the hypothesis that histamine may contribute to the damage to glomerular capillaries and to arterial endothelium. These effects may be relevant to the pathogenesis of glomerular disease and atherosclerosis. Histamine may also contribute to the pathogenesis of pruritus in patients with chronic renal failure.

Gastric function and histology in chronic renal failure.

Gastric function and histology were investigated in 24 patients with untreated chronic renal failure. At endoscopy nine patients had oesophagitis, 12 patients were considered to have gastritis, and the duodenum appeared inflamed in 20 patients. Endoscopic biopsies were taken at standard sites in the stomach and duodenum; gastritis was found in all patients, and 17 patients had duodenitis. Stimulated acid secretion was impaired in seven out of 20 patients and acid hypersecretion was found in a further two patients. Pepsin output correlated well with acid output in these patients. Fasting serum gastrin levels were elevated in 12 of the 19 patients tested. Patients with atrophic gastritis had low acid outputs and hypergastrinaemia, and when extensive gastritis was present, the patients tended to have more severe renal failure and hyposecretion of acid. Three patients were studied again after regular haemodialysis or renal transplantation and were found to show marked endoscopic and histological improvement.