Search

Lymphomatoid gastropathy, which was first reported in 2010, is a rare NK-cell proliferation of cyCD3+, CD4−, CD5−, CD8−, CD56+phenotypes with unknown etiology. The diagnosis is challenging, as there is histopathologic similarity to malignant lymphoma. In the 2010 report on 10 cases, all lesions were located in the stomach, and all regressed without any therapy. In the present study, we analyzed 6 cases of lymphomatoid gastropathy by investigating the clinicopathologic, immunohistochemical, and molecular findings. Endoscopic and morphologic appearances of all cases were consistent with previous reports, but 2 cases showed previously unreported unique immunophenotypes of CD4−CD8+. Three of 6 patients underwent lower gastrointestinal examination (1 case underwent double-balloon endoscopic examination), but no patient had lesions in the lower gastrointestinal tract. No obvious difference of histology was found between the cases of CD4-CD8-typical phenotype and ones of CD4−CD8+phenotype. Both cases had similar clinical behavior as the other 4 cases, implying that the spectrum of the disease is broader than initially thought. Careful clinical and endoscopic follow-up is required for the diagnosis of lymphomatoid gastropathy, and additional case studies and molecular studies are warranted to further investigate the pathophysiology of this peculiar benign mimic of lymphoma.

We have reported previously that duodenal follicular lymphoma (FL) is distinct from nodal FL and showed more resemblance to mucosa-associated lymphoid tissue lymphoma, and that FL frequently involved the duodenal second portion. In the present study, we examined duodenal FLs and gastric/colonic FLs to clarify the clinicopathological and immunological differences between the tumor types. We analyzed 8 samples of gastric FL, 17 of duodenal ones, and 5 of colonic/rectal ones, and characterized them by immunohistochemistry, immunogenotyping, and histology. Gastric and colonic FLs presented in submucosal to subserosal areas, whereas duodenal ones presented in the mucosal to submucosal layers. Immunohistochemical analysis revealed that duodenal FLs exhibited the following phenotypes: CD10 (+), B-cell lymphoma 2 (BCL-2) (+), BCL-6 (+), activation-induced cytidine deaminase (AID) (−), BACH2 (+), CD27 (+), MUM-1 (−), Blimp-1 (−), and loose CD21 network (duodenal pattern). Gastric/colonic FLs exhibited the following phenotypes: CD10 (+), BCL-2 (+), BCL-6 (+), AID (+), BACH2 (+), CD27 (−), MUM-1 (−), Blimp-1 (−), and a dense CD21 network (nodal pattern). Expression of AID and CD27 in lymphoma cells and the CD21 network pattern were considerably different between duodenal FLs and gastric/colonic ones. Moreover, in situ hybridization revealed that, in the duodenal FLs, BACH2 was expressed at the periphery of the tumor follicle and tumor villi. The number of immunoglobulin heavy-chain variable domains VH4 and VH5 were higher in duodenal follicular lymphomoas than in gastric FLs. The lymphoma cells of duodenal FLs are different from those of gastric/colonic FLs, and duodenal FL is distinct even within the gastrointestinal tract. Somatic hypermutation in immunoglobulin genes and CD27 expression are hallmarks of memory B cells. We suggest that duodenal FL cells are in the memory B-cell stage, and require BACH2 instead of AID for ongoing mutation.

We have reported previously that duodenal follicular lymphoma (FL) is distinct from nodal FL and showed more resemblance to mucosa-associated lymphoid tissue lymphoma, and that FL frequently involved the duodenal second portion. In the present study, we examined duodenal FLs and gastric/colonic FLs to clarify the clinicopathological and immunological differences between the tumor types. We analyzed 8 samples of gastric FL, 17 of duodenal ones, and 5 of colonic/rectal ones, and characterized them by immunohistochemistry, immunogenotyping, and histology. Gastric and colonic FLs presented in submucosal to subserosal areas, whereas duodenal ones presented in the mucosal to submucosal layers. Immunohistochemical analysis revealed that duodenal FLs exhibited the following phenotypes: CD10 (+), B-cell lymphoma 2 (BCL-2) (+), BCL-6 (+), activation-induced cytidine deaminase (AID) (−), BACH2 (+), CD27 (+), MUM-1 (−), Blimp-1 (−), and loose CD21 network (duodenal pattern). Gastric/colonic FLs exhibited the following phenotypes: CD10 (+), BCL-2 (+), BCL-6 (+), AID (+), BACH2 (+), CD27 (−), MUM-1 (−), Blimp-1 (−), and a dense CD21 network (nodal pattern). Expression of AID and CD27 in lymphoma cells and the CD21 network pattern were considerably different between duodenal FLs and gastric/colonic ones. Moreover, in situhybridization revealed that, in the duodenal FLs, BACH2 was expressed at the periphery of the tumor follicle and tumor villi. The number of immunoglobulin heavy-chain variable domains VH4 and VH5 were higher in duodenal follicular lymphomoas than in gastric FLs. The lymphoma cells of duodenal FLs are different from those of gastric/colonic FLs, and duodenal FL is distinct even within the gastrointestinal tract. Somatic hypermutation in immunoglobulin genes and CD27 expression are hallmarks of memory B cells. We suggest that duodenal FL cells are in the memory B-cell stage, and require BACH2 instead of AID for ongoing mutation.

Although most follicular lymphomas are believed to be of nodal origin, they sometimes originate from the duodenum. We have reported that the latter differ from nodal follicular lymphomas in having lower clinical stages and uniformly low histological grades, along with variable region of immunoglobulin heavy chain gene (VH) usage that is more similar to mucosa-associated lymphoid tissue (MALT) lymphomas. Little is known, however, about whether they possess other characteristics of nodal follicular lymphomas, particularly ongoing mutations with follicular dendritic cells. We examined 17 cases for which PCR identified the monoclonal bands of the immunoglobulin gene. The duodenal cases showed ongoing mutations, but they lacked activation-induced cytidine deaminase (AID) expression, a statistically significant difference from the nodal cases (P<0.001), and their follicular dendritic cell networks were disrupted. Moreover, not only were VH deviations observed but also they used very restricted VH genes. Although the mechanisms of ongoing mutation without AID and follicular dendritic cell were not clarified, restricted VH usage strongly suggested that antigen stimulation was involved, and that was similar to MALT lymphomas. In conclusion, duodenal follicular lymphomas were shown to be unique, in that they had ongoing hypermutations such as nodal cases, but the mechanisms involved in the hypermutation were quite different; furthermore, restricted VH usage suggested a strong similarity to the antigen-dependent origin of MALT lymphomas.Modern Pathology (2009) 22, 940–949; doi:10.1038/modpathol.2009.51; published online 24 April 2009

Although most follicular lymphomas are believed to be of nodal origin, they sometimes originate from the duodenum. We have reported that the latter differ from nodal follicular lymphomas in having lower clinical stages and uniformly low histological grades, along with variable region of immunoglobulin heavy chain gene (VH) usage that is more similar to mucosa-associated lymphoid tissue (MALT) lymphomas. Little is known, however, about whether they possess other characteristics of nodal follicular lymphomas, particularly ongoing mutations with follicular dendritic cells. We examined 17 cases for which PCR identified the monoclonal bands of the immunoglobulin gene. The duodenal cases showed ongoing mutations, but they lacked activation-induced cytidine deaminase (AID) expression, a statistically significant difference from the nodal cases (P<0.001), and their follicular dendritic cell networks were disrupted. Moreover, not only were VHdeviations observed but also they used very restricted VH genes. Although the mechanisms of ongoing mutation without AID and follicular dendritic cell were not clarified, restricted VHusage strongly suggested that antigen stimulation was involved, and that was similar to MALT lymphomas. In conclusion, duodenal follicular lymphomas were shown to be unique, in that they had ongoing hypermutations such as nodal cases, but the mechanisms involved in the hypermutation were quite different; furthermore, restricted VHusage suggested a strong similarity to the antigen-dependent origin of MALT lymphomas.

Established risk factors for gastric cancer include a diet high in nitrate or nitrite and low in vitamin C and the presence of achlorhydria or hypochlorhydria. The aim of this study was to investigate the relationship between intragastric nitrite concentration and atrophic change of the stomach or gastric carcinogenesis in Japanese Helicobacter pylori-infected patients. Gastric juice pH, nitrite, and total vitamin C concentrations in gastric juice, serum pepsinogen I and II concentrations, and specific Helicobacter pyloriantibody were analyzed. Intragastric total vitamin C concentration was decreased by Helicobacter pyloriinfection of the gastric mucosa and with progression of the atrophic grade. There was a significant positive correlation between atrophic grade and intragastric nitrite concentration. In conclusion, the levels of nitrite in gastric juice play a causal role in the development of cancer in Helicobacter pylori-associated atrophic gastric mucosa.

This study investigated whether the serum nitrite concentration reflects Helicobacter pylori-induced inflammation and atrophic changes of gastric mucosa. Ninety-seven patients underwent biopsy of both antrum and fundus. Samples were analyzed by the rapid urease test and histopathological examination according to the updated Sydney system. Fasting serum samples from each subject were analyzed for specific IgG Helicobacter pyloriantibodies, pepsinogen I and II concentrations, and NO2−/NO3−content. Eleven patients had H. pylorieradicated with proton pump-based triple therapy. There was a strong positive correlation between the Helicobacter pyloridensity in the gastric mucosa and the serum nitrite concentration, but a negative correlation existed between the atrophic grade of the gastric mucosa and both serum nitrite concentration and Helicobacter pyloridensity in the gastric mucosa. Serum nitrite concentrations decreased significantly after successful eradication of Helicobacter pylori. Therefore, serum nitrite concentration may be a useful marker for oxidative DNA damage and apoptosis associated with Helicobacter pyloriinfection.

This study was conducted to investigate the efficacy of rebamipide against experimental colitis induced by dextran sulfate sodium (DSS) in a rat model of inflammatory bowel disease. Experimental colitis was induced in male Wistar rats by oral administration of 3% DSS solution for one week. The rats were provided with standard diet containing 0.105% rebamipide (160 mg/kg/day) for 1 week. In rats treated with rebamipide, clinical (body weight loss, bloody diarrhea, reduced physical activity, severe anemia, shortened colonic length, and perianal injury) and histopathological (pathological lesion score) findings of DSS colitis were significantly less than in rats with DSS colitis not treated with rebamipide. Rebamipide thus inhibited the induction of colitis. Rebamipide significantly reduced concentrations of both interleukin-1α and GRO/CINC-1 (IL-8-like substance) and cell infiltrates in colonic wall, in parallel with decreased activity of myeloperoxidase. It also reduced expression of IL-1 mRNA but did not influence expression of GRO/CINC-1 mRNA. The attenuation of colonic indices of colitis by rebamipide in this rat model suggests that this drug might have beneficial effects in the treatment of human ulcerative colitis. These effects of rebamipide are attributable to its inhibition of inflammatory cytokine-mediated granulocyte (neutrophil) infiltration into the colon.

Antral gastrin and somatostatin gene expression during starvation and after refeeding with liquid meals of varying composition were studied. Northern and slot-blot hybridization analyses showed that starvation caused a marked decrease in antral gastrin messenger RNA (mRNA) level by 12 hours associated with an increase in somatostatin mRNA. After 48 hours of fasting, antral gastrin mRNA was 26% and somatostatin mRNA was 136% of their prefasting levels. Refeeding caused increased 2-hour integrated gastrin mRNA levels after liquid peptone (+45%), phenylalanine (+31%), and olive oil (+13%), but no changes were observed with glucose or saline solutions. Integrated 2-hour immunoreactive antral gastrin content was increased after peptone (+106%), phenylalanine (+68%), and olive oil (+32%) meals but was not increased after glucose (−11%) or saline (−10%). In some cases, both gastrin mRNA and peptide responses could be measured as early as 15 minutes. The same nutrients that increased gastrin mRNA levels caused decreased 2-hour integrated somatostatin mRNA levels; peptone (−30%), phenylalanine (−28%), and olive oil (−21%), but neither glucose nor saline, altered somatostatin mRNA levels. These results suggest that antral gastrin and somatostatin genes were regulated in opposite directions, in a coordinate manner, by specific gastric nutrients that stimulate gastrin release.

Somatostatin in gastric juice was determined in normal subjects and patients with duodenal ulcer. Gel exclusion chromatography of gastric juice revealed that the main immunoreactivity existed at the position of somatostatin-14. A large amount of somatostatin was present in gastric juice, and the quantity increased following tetragastrin stimulation. Furthermore, there was a good inverse correlation between somatostatin concentration and acidity of gastric juice; however, there was no difference between normal subjects and patients with duodenal ulcer in the amount of somatostatin released into gastric juice.

ß-Endorphin-like immunoreactivity was detected in the mucosa and muscle layer of normal colon, adenocarcinomas derived from the colon mucosa, and colon polyps which were histologically confirmed to be adenoma without a focus of carcinoma or with in situ carcinoma. The contents of ß-endorphin-like immunoreactivity in adenocarcinomatous tissue (11.94± 1.77 pmollg wet wt) and colon polyps without focus of carcinoma (10.71 ±1.50 pmollg wet wt) were found to be significantly higher than those in the mucosal layer (6.86± 0.64 pmollg wet wt) and muscle layer (8.30 ±0.68 pmollg wet wt) of normal colon. These data suggest that the production of ß-endorphin-like immunoreactivity is specifically increased in some adenocarcinomas and adenomatous polyps and may be related to the alteration of bowel habits. Gel exclusion chromatography of ß- endorphinlike immunoreactivity revealed three peaks corresponding to ß-endorphin, ß-lipotropin, and an immunoreactive form between the two. In the mucosal layer and muscle layer of the colon, a broad major peak was eluted at the position of ß-endorphin, and minor peaks were eluted at the position of ß-lipotropin and between ß-endorphin and ß-lipotropin. In adenocarcinoma and polyp, the peak size corresponding to authentic ß-lipotropin was greater than that of ß-endorphin. This study demonstrated that ß-endorphin-like immunoreactivity existed at a high concentration in some colon adenocarcinomas and polyps whose elution patterns were different from those of normal colon tissue.

β-Endorphinlike immunoreactivity and somatostatinlike immunoreactivity were detected in the mucosa and muscle layer of normal gastric antrum and corpus and in moderately differentiated adenocarcinoma derived from the antral mucosa. The concentration of β-endorphinlike immunoreactivity in the normal gastric tissues was 4–15 pmol/g wet wt tissue; this value varied from 9.64 to 241.39 pmol/g wet wt tissue (81.38 ± 37.82 pmol/g wet wt tissue) in adenocarcinomas. The concentration of somatostatinlike immunoreactivity was 18–25 pmol/g wet wt tissue in normal gastric mucosa, whereas it was 1–2 pmol/g wet wt tissue in adenocarcinomas. Gel exclusion chromatography of β-endorphinlike immunoreactivity revealed two peaks corresponding to β-endorphin and β-lipotropin. In normal mucosa and in the whole layer of antrum, the major peak was eluted near the position of β-lipotropin, and the minor broad peak was eluted near the position of β-endorphin. In contrast, in adenocarcinoma, β-endorphinlike immunoreactivity was eluted broadly at the position of β-endorphin and the other smaller peak was at the position of β-lipotropin. Gel exclusion chromatography of somatostatinlike immunoreactivity also showed different patterns between antral mucosa and adenocarcinoma. This study revealed the presence of the opioid peptide, β-endorphinlike immunoreactivity, not only in normal gastric tissue but also in adenocarcinomas with highly increased concentration and different elution patterns in combination with decreased concentration of somatostatinlike immunoreactivity.

Pirenzepine has inhibitory effects on gastrin secretion bothin vivo andin vitro. The aim of this study was to determine the mechanism responsible for the suppression of omeprazole-induced hypergastrinemia that occurs with pirenzepine treatment. The effects were measured in rats treated with oral omeprazole plus intraperitoneal pirenzepine or saline once daily for seven days in the antrum. The serum gastrin level increased significantly by more than sixfold with omeprazole treatment; additional treatment with pirenzepine suppressed this increase by 48%. Pirenzepine treatment did not change the level of gastrin mRNA but significantly increased the level of somatostatin mRNA. Combination treatment with omeprazole plus pirenzepine significantly decreased the gastrin mRNA level to half and significantly increased the somatostatin mRNA level up to 1.4-fold of the levels achieved with omeprazole treatment alone. These results suggest that the stimulatory effect of omeprazole on gastrin synthesis is partially blocked by pirenzepine via mediation of somatostatin synthesis in the antrum.

We examined the effects of gastrin and histamine on rat gastric H+/K+-ATPase, the enzyme responsible for H+ secretion, gene expressionin vivo. Gastrin 17 (G 17) or histamine dihydrochloride (histamine) was continuously infused through the femoral vein of anesthetized rats. Gastric H+/K+-ATPase mRNA levels were measured using northern blot analysis. Infusion of G 17 and histamine increased the H+/K+-ATPase mRNA level significantly compared with basal control level or vehicle control level (P<0.01). However, pretreatment with famotidine, a potent histamine-2 (H2)-receptor antagonist, inhibited the increase of rat gastric H+/K+-ATPase mRNA following G 17 and histamine infusion. These findings indicate that both histamine and G 17 increase expression of H+/K+-ATPase mRNA by activating H2 receptor on the parietal cell.

The aim of this study was to investigate whethergastrin regulates morphological changes andalpha-subunit gene expression in parietal cells throughthe gastrin/CCK-B receptor on enterochromaffin-likecells by histamine release. Treatment with 100 mg/kgof YM022, a potent and selective gastrin/CCK-B receptorantagonist, for one week in rats did not alter mRNAlevels of histidine decarboxylase or H+,K-ATPase. However, parietal cell morphology predominantlychanged to the resting form, although the serum gastrinconcentration was significantly increased. Additionaltreatment with YM022 and oral omeprazole, 100 mg/kg, for one week markedly suppressed theincreases of mRNA levels of histidine decarboxylase andH+, K-ATPase and completely blocked themorphological transformation of the parietal cells to astimulated form induced by treatment with omeprazolealone. This indicates that the morphologicaltransformation of parietal cells to an activated formwith a subsequent increase in H+, K-ATPasesynthesis caused by hypergastrinemia is mediated by increasedhistidine decarboxylase gene expression inenterochromaffin-like cells via gastrin/CCK-Breceptors.

This study was performed to examine the effectsof additional enprostil administration onhypergastrinemia and gastric acid suppression induced byomeprazole. Serum gastrin concentrations were measured in 10 peptic ulcer patients (six Helicobacterpylori-positive and four Helicobacter pylori-negativepatients) before treatment, after two weeks ofomeprazole (20 mg/day), and after two weeks ofomeprazole and enprostil (50 μg/day). The additionalacid inhibitory effect of enprostil was evaluated by24-hr intragastric pH measurements in five healthyHelicobacter pylori-negative volunteers. Afteromeprazole treatment, the serum gastrin level ofHelicobacter pylori-positive patients (3.5-fold ofcontrol) was markedly higher than that of Helicobacterpylori-negative patients (1.7-fold of control).Additional treatment with enprostil suppressed serumgastrin levels to 0.4-fold and 0.7-fold of omeprazoletreatment levels in Helicobacter pylori-positive andHelicobacter pylori-negative patients, respectively. In healthy volunteers, median pH recordedduring the nonmeal daytime interval increasedsignificantly with additional enprostil. Thus, enprostilreduces undesirable omeprazole-induced hypergastrinemia, especially in Helicobacter pylori-positivepatients, and effectively suppresses acidsecretion.

The specific binding of125I-labeled gastrin-17 was studied in samples of human gastric mucosa, duodenal mucosa, colonic mucosa or pancreatic tissue obtained surgically. With regard to gastric fundic mucosa, the criteria for receptor binding were studied, and saturability, high affinity, tissue specificity and hormone specificity were demonstrated. The dissociation constant for gastric fundic mucosa was 1.6 x 10-9M, and the binding capacity was 15 fmol/mg protein. Tissue specificity studies revealed a high degree of gastrin receptor binding in human gastric fundic mucosa, duodenal mucosa and pancreas, whereas antral mucosa and colonic mucosa demonstrated a low degree of binding.

Abstract: Omeprazole effectively suppresses acid secretion, resulting in the long-term elevation of intragastric pH and serum gastrin level. Pirenzepine has been reported to inhibit gastrin secretion. This study was carried out to examine the effects of additional pirenzepine treatment on the hypergastrinemia and gastric acid suppression induced by omeprazole. Concentrations of serum gastrin and plasma somatostatin were measured in 28 peptic ulcer patients before treatment, after omeprazole treatment (20 mg/day) for 2 weeks, and after omeprazole and pirenzepine (100 mg/day) treatment for 2 weeks. The acid inhibitory effect of pirenzepine treatment in addition to omeprazole was evaluated by 24-h intragastric pH measurement in six healthy volunteers. Serum gastrin level was increased significantly, to 2.4-fold the pretreatment level, by omeprazole treatment. Additional treatment with pirenzepine suppressed serum gastrin level to 0.6-fold the omeprazole-treatment level. The serum somatostatin level was not altered significantly either by omeprazole treatment or by omeprazole and pirenzepine treatment. In healthy volunteers whose pH 3 holding time on 24-h intragastric pH monitoring was 70% by omeprazole treatment, omeprazole and pirenzepine treatment markedly increased the pH 3 holding time, to 89%. These findings suggest that pirenzepine is useful in reducing the undesirable effects of omeprazole-induced hypergastrinemia, i.e., the excessive trophic effect of omeprazole on the acid-secreting part of the stomach and the overstimulation of acid secretion. The additional pirenzepine treatment is also effective in suppressing acid secretion.

Correction to: Modern Pathology (2013) 26, 22–31; doi:10.1038/modpathol.2012.127; published online 17 August 2012. In this article, the name of the 16th author in the byline is incorrect. The correct name is Lamia Abd Al-Kader.