Aberrations in the fragile histidine triad (FHIT) gene in idiopathic pulmonary fibrosis.

Idiopathic pulmonary fibrosis (IPF) seems to be closely associated with lung carcinogenesis. To identify the genetic characteristics of precancerous IPF lesions in the peripheral lung, we performed PCR-based microsatellite analysis with DNA extracted from microdissected tissues; fluorescent in situ hybridization (FISH) analysis of the fragile histidine triad (FHIT) gene and immunohistochemical analysis of Fhit protein expression in samples of metaplasias and bronchiolar epithelia obtained from patients with IPF. We used four microsatellite markers of the FHIT gene within or flanking the FHIT gene on chromosome 3p for loss of heterozygosity (LOH) analysis. LOH of the FHIT locus was frequently found among the lesions of metaplasias and bronchiolar epithelia in the patients with IPF [62 (52%) of 119 informative lesions]. Fifty-four (73%) of the 74 lesions of metaplasias and bronchiolar epithelia obtained from the IPF patients with lung carcinoma and 8 (17%) of the 46 samples obtained from the IPF patients without lung carcinoma showed LOH at the FHIT gene (P < 0.0001). We confirmed allelic loss in the metaplasias and bronchiolar epithelia of IPF by FISH analysis of the FHIT gene. Additionally, the level of Fhit protein expression in the metaplastic cells of IPF was frequently reduced. Our findings suggest that allelic loss of the FHIT gene may be involved in carcinogenesis in the peripheral lung of patients with IPF.

Plasma concentrations of tumor necrosis factor-alpha (TNF-alpha) and soluble TNF receptors in patients with anorexia nervosa.

Tumor necrosis factor-alpha (TNF-alpha) is a cytokine with numerous immunological and metabolic activities. To study the role of TNF-alpha on the pathophysiology of anorexia nervosa and its complications, plasma concentrations of TNF-alpha, 2 soluble TNF receptors (sTNF-RI and sTNF-RII), and leptin were measured in 20 female patients with anorexia nervosa (AN) and 20 age-matched normal women (N). Plasma TNF-alpha concentrations in AN were significantly higher than those in N (4.1 +/- 0.6 pg/mL vs. 1.6 +/- 0.1 pg/mL; P < 0.01). Although no significant difference was observed in plasma sTNF-RI concentrations between the two groups, plasma sTNF-RII concentrations in AN were significantly higher than those in N (2094.0 +/- 138.5 pg/mL vs. 1569.5 +/- 84.0 pg/mL; P < 0.01). Plasma concentrations of TNF-alpha and sTNF-RII after treatment of 8 anorectic patients were not different from those before treatment, although body fat mass and plasma leptin concentrations significantly increased after treatment. Plasma TNF-alpha concentrations were not related to body fat mass in anorectic patients. These results suggest that the adipose tissue may not be the immediate source of TNF-alpha in anorectic patients and that TNF-alpha may contribute to the pathophysiology of immunological and metabolic abnormalities in anorexia nervosa.

Up-regulation of high-affinity dehydroepiandrosterone binding activity by dehydroepiandrosterone in activated human T lymphocytes.

Although evidence indicates that dehydroepiandrosterone (DHEA) exerts direct physiological effects, its mechanism of action remains unknown. DHEA binding sites were examined using a whole-cell binding assay in a human T lymphoid cell line, PEER, revealing that a single class of high-affinity binding sites for DHEA (dissociation constant = 7.4 +/- 0.53 nmol/L, mean +/- SE, n = 4) was greatly increased when treated with DHEA, phorbol-12-myristate-13-acetate, and the Ca2+ ionophore A23187. Bound [3H]DHEA was displaced sensitively by DHEA and secondarily by dihydrotestosterone, but not effectively by other steroids, including DHEA sulfate. These results not only indicate the existence of a DHEA receptor, but also suggest that T cells become susceptible to regulation by DHEA during the process of signal-induced activation.

Does plasma immunoreactive ouabain originate from the adrenal gland?

It was reported recently that the endogenous digitalis-like factor ouabain may mainly originate from the adrenal gland. To ascertain the pathophysiological significance of endogenous ouabain and to examine if it originates in the adrenal gland, we determined plasma immunoreactive ouabain levels in patients with various cardiovascular and endocrine diseases. Plasma immunoreactive ouabain levels were also determined in the adrenal venous blood by adrenal venous sampling. Plasma immunoreactive ouabain levels were significantly increased in patients with essential hypertension, primary aldosteronism, Cushing's syndrome, pheochromocytoma, acromegaly, and chronic renal failure. Plasma immunoreactive ouabain levels were decreased in patients with primary aldosteronism after unilateral adrenalectomy, acromegaly after pituitary adenomectomy, and chronic renal failure after hemodialysis. Plasma immunoreactive ouabain levels in patients after bilateral adrenalectomy were similar to those in healthy subjects. There was no significant step-up of immunoreactive ouabain levels in the adrenal vein from the peripheral vein in three patients, whereas one patient with hypertension and right adrenal tumor but without any known adrenal hormone excess showed higher plasma immunoreactive ouabain levels in the right adrenal vein than those in the peripheral vein. These results suggest an important pathophysiological significance of endogenous ouabain in various cardiovascular and endocrine diseases. It is unlikely that the adrenal gland is a major source of plasma ouabain, although a possible excess production of ouabain by the adrenal tumor remains to be elucidated.

Evidence for the existence of des-Asp1-angiotensin II in human uterine and adrenal tissues.

Renin is present in various tissues outside the kidney. In contrast, the levels of angiotensins (ANG), the active products of the renin-angiotensin system, have not been thoroughly evaluated in tissues. In this study, we demonstrated the presence of immunoreactive (ir) ANG I and ANG II in various human tissues by RIA. Of the tissues examined, uterine tissue contained the most ir-ANG II. Since the anti-ANG II antibody used had significant cross-reactivity with ANG III, high performance liquid chromatography was performed to separate ANG II from ANG III. The major portion of the ir-ANG II in the plasma was ANG II. In contrast, the major portion of the ir-ANG II in uterine tissue was determined to be ANG III, a known biologically active peptide. The adrenal gland and testis also contained ANG III. From these results, it can be postulated that ANG III may contribute to the biological activity of ANG in some tissues.

Effects of glutaraldehyde and other drugs on concanavalin A-mediated red blood cell adsorption to nonsenescent, senescent and transformed human fibroblasts.

Concanavalin A (Con A)-mediated red blood cell (RBC) adsorption with the RBC coating method (in which Con A-coated RBC's are adsorbed to fibroblasts) was greatly increased by glutaraldehyde fixation of RBCs before Con A-coating and decreased by the fixation of fibroblasts. On the other hand, RBC adsorption with the fibroblast coating method (in which RBCs are adsorbed to Con A-coated fibroblasts) was decreased by the fixation of RBCs and increased by the fixation of fibroblasts before Con A coating. The fixation of RBCs or fibroblasts after Con A coating did not have these effects. In addition, the fixation of both RBCs and fibroblasts nearly completely abolished RBC adsorption with either method. However, the amount of [3H] Con A binding was not affected by the fixation. RBC adsorption with the fibroblast coating method was also affected by cytochalasin B, colchicine, NaN3 and dibucane treatments of fibroblasts. These drug treatments of fibroblasts, however, did not affect RBC adsorption with the RBC coating method, except cytochalasin B. In addition, the effects of drug treatments of fibroblasts examined occurred nearly to the same extent for nonsenescent, senescent, and transformed cells. Our results suggest that secondary processes after Con A binding, receptor mobility and receptor association with cytoskeletals, play important roles in RBC adsorption, but that the roles do not change with aging or transformation.

Age-related cell surface changes in human diploid fibroblasts revealed by lectin-mediated red blood cell adsorption assay: a lectin survey.

Two types of age-related cell surface changes could be demonstrated in human diploid fibroblasts with the two methods of the lectin-mediated red blood cell (RBC) adsorption assay: the fibroblast coating method (in which RBCs are adsorbed to lectin-coated fibroblasts) and the RBC coating method (in which lectin-coated RBCs are adsorbed to fibroblasts). With the fibroblast coating method, concanavalin A and agglutinin L from Phaseolus vulgaris gave a change in RBC adsorption which did not occur throughout the phase II period, but increased with the advance of the phase III period (type I). With the RBC coating method, these lectins gave another type of change in RBC adsorption which increased continuously from early phases of cell passage up through cell senescence (type II). Ricinus communis agglutinin 120 also gave the type I change in RBC adsorption with the fibroblast coating method. On the other hand, even with the fibroblast coating method, Phaseolus vulgaris agglutinin E and wheat-germ agglutinin gave the type II change in RBC adsorption. Soy bean agglutinin and Bauhinia purpurea agglutinin gave only a restricted amount of RBC adsorption. Lens culinaris agglutin, pokeweed mitogen, Dolichos biflorus agglutinin, Lotus tetragonolobus agglutinin, Limulus polyhemus agglutinin and divalent succinylated concanavalin A did not give any RBC adsorption throughout the life span of human diploid fibroblasts.

Cell-surface changes accompanying aging in human diploid fibroblasts: effects of tissue, donor age and genotype.

The generality of age-related changes in concanavalin A (Con A)-mediated red blood cell (RBC) adsorption to human diploid fibroblasts was investigated on fibroblast-like cells from fetal lung, heart, liver, skin and muscle tissues. All the cells examined showed the continuous increase from early passages in RBC adsorption with the RBC coating method (in which Con A-coated RBCs are adsorbed to fibroblasts) and the incraease only at phase III with the fibroblast coating method (in which RBCs are adsorbed to Con A-coated fibroblasts). All of the four strains of lung fibroblasts gave nearly the same extent of the age-related change in RBC adsorption, when expressed as a function of percentage life span consumed, indicating that the change in RBC adsorption is independent of genetic heterogeneity and conditions of primary culture. Liver and heart fibroblasts also gave results similar to those of lung fibroblasts. However, skin and muscle fibroblasts were lower in their RBC adsorption capacity throughout the life span. The continuous age-related increase in RBC adsorption to these cells could be sensitized by using glutaraldehyde-prefixed RBCs, trypsinized RBCs or phytohemagglutinin P in place of Con A. The relevance of the phenotype of in vitro aging revealed by RBC adsorption to in vivo aging was also demonstrated on skin fibroblasts from different ages of donors using glutaraldehyde-prefixed RBCs. In addition, fibroblasts from patients with Werner's syndrome, an hereditary disease manifested by early and widespread degenerative changes, showed senescent phenotype in RBC adsorption even at early passages.

Altered expression of several genes in highly metastatic subpopulations of a human pulmonary adenocarcinoma cell line.

Non-small cell lung cancer is associated with approximately 85% mortality due to its high metastatic potential. Therapeutic efforts have failed to produce a significant improvement in prognosis. In this situation, a better understanding of the key factors of metastasis may be useful for designing new molecular targets of therapy. In order to identify these factors, we compared the expression profiles of two subpopulations of an adenocarcinoma cell line with a high metastatic potential, PC9/f9 and PC9/f14, with the parent cell line, PC9, using a cDNA array. The expression of 15 genes was found to be significantly enhanced or reduced in the highly metastatic subpopulations. The expression of matrix metalloproteinase-2 (MMP-2), plasminogen activator inhibitor-1 (PAI-1) and interleukin-1 (IL-1 alpha) were upregulated in the highly metastatic subpopulations, while the expression of carcinoembryonic antigen (CEA), caspase-5, Fas ligand, Prk/FNK, cyclin E, cyclin B1, Ki-67, proliferating cell nuclear antigen (PCNA), Smad4, macrophage proinflammatory human chemokine-3 alpha (MIP-3 alpha)/LARC, Met and CD44 were downregulated. Data from the literature suggest that the altered expression of MMP-2, PAI-1, IL-1 alpha, CEA, caspase-5, Fas ligand, Prk/FNK and Smad4 promotes the highly metastatic phenotype. The differential expression of these genes was confirmed by Northern blot analysis, standard reverse transcription-polymerase chain reaction (RT-PCR) and real-time quantitative RT-PCR. This analysis in subpopulations of a lung cancer cell line indicated that the highly metastatic potential of lung cancer may be induced not by an alteration in the expression of a single gene, but by the accumulation of alterations in the expression of several genes involved in extracellular matrix (ECM) adhesion disruption, ECM degradation, escape from apoptosis, and resistance to transforming growth factor-beta(1) (TGF-beta(1)). Strategies for inhibiting metastasis of pulmonary adenocarcinoma should be designed accordingly.

Effect of natural menopause on serum levels of IGF-I and IGF-binding proteins: relationship with bone mineral density and lipid metabolism in perimenopausal women.

The present study was performed to examine the effect of natural menopause on serum levels of IGF-I, IGF-binding protein (IGFBP)-2 and -3 as well as on bone mass and lipid metabolism in perimenopausal women. One hundred and twenty-one healthy Japanese women, who were 45-55 years old, were studied (71 premenopausal and 50 postmenopausal women 1-9 years after menopause). Bone mineral density (BMD) was measured at the middle third of the radius by using dual energy X-ray absorptiometry. Serum levels of IGF-I, but not those of IGFBP-2 or -3, were significantly reduced in the postmenopausal group compared with the premenopausal group. One year after menopause, serum IGF-I levels were significantly lower, and the biochemical markers of bone turnover, such as serum total alkaline phosphatase level and urinary calcium to creatinine ratio, were significantly higher than the premenopausal levels. Serum levels of IGF-I, but not those of IGFBP-2 or -3, were positively correlated with BMD. Serum levels of IGFBP-2, but not those of IGF-I or IGFBP-3, were negatively correlated with body mass index and body weight. Finally, serum levels of IGFBP-3, but not those of IGF-I, were positively correlated with serum levels of total cholesterol and triglyceride. The present findings suggest that a rapid decrease in serum IGF-I levels after menopause might be partly involved in bone loss following gonadal failure and that IGFBP-2 and -3 might be related to the regulation of body mass and lipid metabolism during perimenopause respectively, although the mechanisms remain unknown.

Mutation analysis of the gene encoding the human mannose 6-phosphate/insulin-like growth factor 2 receptor (M6P/IGF2R) in human cell lines resistant to growth inhibition by transforming growth factor beta(1) (TGF-beta(1)).

The mannose 6-phosphate/insulin-like growth factor 2 receptor (M6P/IGF2R) is involved in activating the transforming growth factor beta(1) (TGF-beta(1)), an inhibitor of the cell proliferation, and limiting the insulin-like growth factor 2 mediated-growth stimulation. The M6P/IGF2R gene has been reported to be mutated and deleted in various cancers, and is a candidate tumor suppressor gene. We studied the genomic structure of the M6P/IGF2R gene and designed the intron primers to detect mutations in the M6P/IGF2R gene of genomic DNA samples. The M6P/IGF2R gene consists of 48 exons. The previously reported 23 mutations of the M6P/IGF2R gene in human cancers, liver, breast, and gastrointestinal tumors, are located in five exons, exon 27, 28, 31, 40, 48. Using the intron primers designed in this study, polymerase chain reaction-single strand conformation polymorphism (PCR-SSCP) analysis, and direct sequencing, we performed an initial analysis of the complete coding sequences of the M6P/IGF2R gene in 21 human cell lines resistant to growth inhibition by TGF-beta(1). An adenine-to-guanine transition, resulting in an asparagine-to-serine amino acid substitution, was found in one lung adenocarcinoma cell line at exon 40 where the mutation has been previously reported in human cancers. This is the first report of a mutation of the M6P/IGF2R gene in lung tumor. These results indicated that the mutation in M6P/IGF2R may be involved in human lung cancinogenesis.

Alteration of the PTEN/MMAC1 gene locus in primary lung cancer with distant metastasis.

The PTEN/MMAC1 gene located at 10q23, has been proposed to be a tumor suppressor gene. To determine the involvement of alteration of the PTEN/MMAC1 gene in carcinogenesis and the progression of primary lung cancers, we analyzed tumor samples of primary and distant metastatic sites and normal lung tissue samples of 30 patients with advanced lung cancer with distant metastasis. The tissues were analyzed for allelic deletion and mutational inactivation of PTEN/MMAC1 by loss of heterozygosity (LOH) analysis, polymerase chain reaction-single strand conformation polymorphism (PCR-SSCP), and direct sequence analysis. LOH of the PTEN/MMAC1 locus was common in each histologic type of primary lung cancer. In this study, the overall allelic deletion rate was 33.3% (7/21). Allelic loss at the primary site and that at the metastatic site of each patient, were identical; in most cases, it seemed that the allelic loss had occurred before metastasis. Sequence analysis of the PTEN/MMAC1 gene revealed a G to C substitution located 8 bp upstream of the coding region of exon 1 and which seems to be a polymorphism, in 4 of the 30 cases. Somatic mutations of the PTEN/MMAC1 gene were not identified in any of the tumors at the primary and metastatic sites. These data indicate that point mutations in the PTEN/MMAC1 gene are probably not an important factor in tumorigenesis and the progression of a major subset of lung cancers. Due to frequent allelic loss at the PTEN/MMAC1 locus occurring at a stage earlier than the metastatic process, alternative mechanisms in which the remaining allele is inactivated such as methylation or homozygous deletion of a small region of the gene that can not be detected by the usual analysis, or alteration of other important tumor suppressor genes lying close to the PTEN/MMAC1 gene on 10q23, may be involved in the tumorigenesis of lung cancers of all histologic subtypes.

Genomic structure of the human MAD2 gene and mutation analysis in human lung and breast cancers.

Some of the many human cancers that exhibit chromosomal instability also carry mutations in mitotic checkpoint genes and/or reveal reduced expression of some of those genes, such as hMAD2. To facilitate investigation of alterations of hMAD2, we determined its genomic structure and intronic primers designed to amplify the entire coding region. Since general impairment of the mitotic checkpoint is frequently reported in lung cancers, and reduced expression of hMAD2 has been reported in breast cancers as well, we searched for mutations throughout the coding sequence of this gene in the genomic DNA of 30 primary lung tumors, 30 lung-cancer cell lines and 48 primary breast cancers. Our approach, which involved polymerase chain reaction-single strand conformation polymorphism (PCR-SSCP) analysis and direct sequencing, revealed nucleotide variants in only two of the 108 specimens. One was a cytosine-to-adenine substitution 3 bp upstream of exon 4 that occurred in one lung cancer cell line and one primary breast tumor, a change that did not alter transcriptional sequence. The other was an adenine-to-guanine substitution within exon 4, of the same lung cell line; this change already had been reported as a polymorphism. The results suggested that the hMAD2 gene is not commonly mutated in either lung nor breast cancers. Further studies should focus on other mechanisms that might account for reduced expression of the hMAD2 gene, and/or pursue analyses of other mitotic checkpoint genes for mutations in human cancer. Nevertheless, the genomic structure, the intronic primer sequences, and polymorphisms of the hMAD2 gene presented here will facilitate future studies to determine the full spectrum and frequency of the genetic events that can affect expression of the hMAD2 gene in human tumors.

Leptin in women with eating disorders.

The aim of the present study was to determine the factors controlling leptin secretion and to clarify the role of leptin in eating disorders. The subjects were 152 eating-disordered women with different fat mass, eating behavior, and endocrine abnormalities and 24 age-matched control subjects. The body fat mass, eating behavior score, and plasma leptin, luteinizing hormone (LH), follicle-stimulating hormone (FSH), triiodothyronine (T3), free thyroxine (T4), insulin, and cortisol levels were evaluated for each subject. In patients with eating disorder, logarithmic values for leptin were significantly correlated with the body fat mass (r = .828, P < .001), eating behavior score (r = .777, P < .001), and LH (r = .465, P < .001), FSH (r = .440, P < .001), T3 (r = .572, P < .001), insulin (r = .410, P < .001), and cortisol (r = -.389, P < .001) levels. After adjusting for fat mass, the partial correlations of log leptin with LH, FSH, insulin, and cortisol were not statistically significant, but log leptin remained correlated with T3 (r = .390, P < .01). Stepwise regression analysis showed that the body fat mass and eating behavior score were significant determinants of leptin levels. These results suggest that eating behavior, as well as the body fat mass, is the control factor for leptin secretion in eating disorders.

Immunoreactive alpha-human atrial natriuretic polypeptide in human plasma.

A radioimmunoassay (RIA) has been developed for the determination of alpha-human atrial natriuretic polypeptide (alpha-hANP) in human plasma. Antibodies generated in rabbits recognized alpha-hANP-related peptides containing the subsequence flanked by two cysteine residues at position 7 and 23 equally. Radiolabelled tracer prepared by iodination with chloramine-T method was purified by high performance liquid chromatography. Immunoreactive (ir-) alpha-hANP was extracted from human plasma by Sep-Pak C18 column. The plasma ir-alpha-hANP concentrations in normal, healthy adults were 178 +/- 16 pg/ml in male and 182 +/- 18 pg/ml in female, respectively. Plasma ir-alpha-hANP increased significantly after acute intravenous administration of isotonic saline. Plasma levels were elevated in patients with various disease states accompanying increased body fluid volume, whereas those in patients with idiopathic edema were decreased despite excessive salt and water retention. These results suggest that alpha-hANP plays an important role in the regulation of body fluids and may have primary or secondary pathophysiological significance in various disease states.

Plasma level of basic fibroblast growth factor increases with progression of chronic liver disease.

Basic fibroblast growth factor (FGF) is thought to be involved in carcinogenesis and, to clarify its clinical significance, the study of its blood level in cancer patients is important. Plasma levels of basic FGF are reported to be elevated in some cancers. However, little is known of basic FGF levels in plasma in hepatocellular carcinoma (HCC). In this study, we measured basic FGF plasma levels in patients with chronic liver disease and compared the levels in chronic hepatitis (CH), liver cirrhosis (LC), and HCC. We also examined whether these levels were related to serum levels of asparate aminotransferase, alanine aminotransferase, gamma-glutamyl transpeptidase, alkaline phosphatase, leucine aminopeptidase, total bilirubin, total protein, and albumin, and to the indocyanine green test (i.e., liver function tests) and to type III procollagen. 7S domain of IV type collagen, and hyaluronic acid (i.e., markers of liver fibrosis). Levels of basic FGF, determined by a quantitative "sandwich" enzyme immunoassay, were significantly elevated with the progression of liver disease; being 3.67 +/- 2.37 (mean +/- SD). 7.78 +/- 6.61, and 12.37 +/- 7.67 pg/ml in the CH, LC, and HCC groups, respectively. FGF levels were elevated to a greater extent in the HCC patients than in the CH (P < 0.0001) and LC patients (P = 0.0117). Levels were higher in LC than in CH (P = 0.0204). None of the liver function test findings or levels of markers of liver fibrosis were correlated with levels of basic FGF. These results suggest that circulating basic FGF could serve as a new indicator of the progression of chronic liver disease. The extremely elevated plasma of level basic FGF in the HCC group suggests that basic FGF may be related to the development of HCC.

Unchanged frequency of loss of heterozygosity and size of the deleted region at 8p21-23 during metastasis of lung cancer.

The genetic mechanisms involved in lung cancer development and progression are beginning to be understood. Many studies have documented frequent loss of heterozygosity (LOH) at specific chromosomal regions in cancer cells; this implies that tumor suppressor genes (TSG) are usually present in those regions. Recently, it has been reported that LOH or chromosomal deletions at chromosome 8p21-23 represent early events frequently occurring in lung cancer. In addition, the size of these chromosome 8 deletions, as well as their frequency, was also reported to increase during lung cancer progression. To determine the spectrum and frequency of alterations of chromosome 8p21-23 in human lung cancer and whether these increase with progression of the tumors, we performed LOH analysis of chromosome 8p and 3p in the genomic DNA from cells from primary and metastatic sites of lung cancer, as well as from normal lung. We studied 35 subjects with primary lung cancer including 30 tumors with distant metastasis. Detection of allelic deletion utilized a PCR-based approach of microsatellite polymorphism analysis, which was performed using the microsatellite markers D8S1130, D8S1106, D8S511, D8S1827, D8S549, D8S261, LPL, D8S258, D8S136, NEFL, D3S1295, D3S1313, D3S1234, D3S1300, D3S1351, D3S1339, and D3S1340. The overall allelic deletion rates were 10 of 28 (35.7%) at 8p and 13 of 33 (39.4%) at 3p. The allelic deletions in the primary cancer and its metastatic sites were in each case identical in both frequency and size of the deleted regions. In our analysis, 8p21-23 deletions were not always associated with 3p deletions in primary lung cancer. These results therefore suggest that allelic deletion at chromosome 8p21-23 is an early and frequent event in the carcinogenesis and development of lung cancer, independent of chromosome 3p deletion. However, a continuing increase in the frequency of LOH at 8p21-23 and in the size of the deleted region rarely occurs during the process of metastasis.

Serum anti-Lewis X antibody is associated with VacA seropositivity but not atrophic gastritis in patients with Helicobacter pylori infection.

OBJECTIVES: Lipopolysaccharides of Helicobacter pylori have an antigenic structure that mimics Lewis X occurring in gastric mucosa. The pathogenic role of antigenic mimicry in H. pylori-induced gastritis has been of recent interest. The aim of this study was to examine the relevance of anti-Lewis X antibody in the development of atrophic gastritis in H. pylori infection. METHODS: A total of 72 patients were studied. Serum samples were collected to measure IgG antibodies to H. pylori, CagA, VacA and Lewis X. Biopsy specimens were obtained from the antrum and the corpus to examine the grade and the type of atrophic gastritis. RESULTS: Mean anti-Lewis X antibody titres were higher in 38 VacA-seropositive patients than in 13 seronegative patients (P < 0.05). The difference was not significant between patients with diffuse-type atrophic gastritis and those with multi-focal type. No significant correlation was observed between the titre of anti-Lewis X antibody and the grade of glandular atrophy, whereas CagA seropositivity was associated with glandular atrophy. CONCLUSIONS: Anti-Lewis X antibody may play a role in persistent gastric inflammation, particularly in VacA-seropositive H. pylori infection. However, anti-Lewis X antibody does not seem itself to be associated with atrophic gastritis in patients with H. pylori infection.

Elevated plasma atrial natriuretic peptide level in the early phase of microalbuminuria in patients with non-insulin-dependent diabetes mellitus.

Plasma atrial natriuretic peptide (ANP) levels were examined in 66 patients with non-insulin-dependent diabetes mellitus (NIDDM), and in 9 age-matched normal controls and 18 hypertensive controls. The diabetic patients were classified into three groups according to random urine albumin/creatinine ratio (ACR, mg/g or mg/88.4 nM); group 1 (normo-albuminuria, ACR less than 20, n = 34), group 2 (borderline microcalbuminuria, 20 less than or equal to ACR less than 100, n = 17) and group 3 (manifest microalbuminuria and macroalbuminuria, 100 less than or equal to ACR less than 2000, n = 15). Plasma ANP levels (pg per ml) were significantly elevated in group 2 (46.0 +/- 19.0 SD) when compared with either normal controls (23.8 +/- 14.2), group 1 (28.9 +/- 15.6) or group 3 (26.0 +/- 12.9). This increase in plasma ANP levels was not related to hypertension. Furthermore, plasma ANP levels correlated positively with log(ACR) among the patients with ACR under 100 (groups 1 and 2 combined, r = 0.4701, p less than 0.01). These results suggest that an elevated plasma ANP level in the early phase of microalbuminuria possibly plays a pathophysiological role in the development of nephropathy in NIDDM patients.

Effects of interferon-gamma on cell differentiation and cytokine production of a human monoblast cell line, U937.

The U937 cell, a human monoblast cell line, has been used as a model to study the function of human monocytes. We investigated the effects of interferon-gamma (IFN-gamma) on superoxide anion (O2-) production, cell surface antigens, and cytokine production of U937 cells. IFN-gamma treatment enhanced O2- production of fMLP or PMA-stimulated U937 cells. IFN-gamma increased the ratio of CD23 and CD11b positive cells. The fluorescence intensity of CD14 and CD25 was enhanced by IFN-gamma treatment. U937 cells produced IL-1 alpha, IL-1 beta, IL-6, and TNF-alpha by lipopolysaccharide (LPS) stimulation. IFN-gamma treatment enhanced TNF-alpha production, but decreased IL-6 production. These results suggest that IFN-gamma differentiates U937 cells to monocyte-like cells and it regulates the production systems of IL-6 and TNF-alpha separately in U937 cells.