Gene expression relevant to osteoclastogenesis in the synovium and bone marrow of mature rats with collagen-induced arthritis.

OBJECTIVES: To investigate gene expression relevant to osteoclastogenesis in the synovium and bone marrow during the development of collagen-induced arthritis (CIA) in mature rats. METHODS: Total messenger RNAs (mRNAs) were obtained from CIA synovium and bone marrow after immunization. First, reverse transcriptase-polymerase chain reactions (RT-PCR) were carried out to detect the mRNA encoding receptor activator of NF-kappaB (RANK), RANK ligand (RANKL), osteoprotegerin (OPG), tumour necrosis factor-alpha (TNF-alpha), interleukin (IL)-1beta, IL-6 and the osteoclast markers tartrate-resistant acid phosphatase (TRAP) and cathepsin K. Secondly, the genes detected clearly by RT-PCR were quantified using real-time PCR. RESULTS: In the synovium, expression of all genes was confirmed by specific single bands in RT-PCR. In real-time PCR, the expression levels of TNF-alpha, IL-1beta, IL-6, RANKL, TRAP and cathepsin K mRNA increased, whereas the expression levels of RANK and OPG were unchanged and decreased respectively. RANKL expression was highly correlated with the two osteoclast markers. In the bone marrow, RT-PCR did not clearly detect the expression of IL-6, RANKL or OPG mRNA. Quantitative real-time PCR showed that TNF-alpha, RANK and TRAP mRNA expression did not change significantly with time, and that IL-1beta and cathepsin K changed slightly compared with those in the synovium. CONCLUSIONS: In the early stages of arthritis, synovial RANKL is closely involved in osteoclastogenesis, and various changes in synovial cytokines, including down-regulation of OPG, probably accelerate osteoclast formation. In contrast, cytokine mRNA in the bone marrow showed little fluctuation. We suggest that synovial cytokines affect osteoclastogenesis not only in the synovium but in the bone marrow.

Immunohistochemical study of p53, c-erbB-2, and PCNA in barrett's esophagus with dysplasia and adenocarcinoma arising from experimental acid or alkaline reflux model.

PURPOSE: An immunohistochemical study of p53, c-erbB-2, and proliferating cell nuclear antigen (PCNA) in Barrett's esophagus with dysplasia and adenocarcinoma, arising from experimental acid or alkaline reflux, was performed in dogs. METHODS: Cardiectomy was performed in group A (n = 26) as an acid reflux model, and total gastrectomy was performed in group B (n = 24) as an alkaline reflux model. After surgery, the esophageal mucosa was observed and biopsied endoscopically every 3 months over a period of 6 years. Immunohistochemical staining of p53. c-erbB-2, and PCNA was performed, using biopsied specimens. RESULTS: In group A, Barrett's esophagus developed in 14 of the 26 dogs. Low-grade dysplasia occurred in 5 of the 26 dogs, and in 1 of these 5 dogs, it developed into high-grade dysplasia. In this animal, adenocarcinoma arose 63 months after the operation. In group B, Barrett's esophagus developed in 10 of the 24 dogs. Low-grade dysplasia was observed in 4 of the 24 dogs. In 1 of these 4 dogs, the dysplasia became high-grade and adenocarcinoma occurred 66 months after the operation. In group A, PCNA was positive in adenocarcinoma; the PCNA labeling index (LI) was 58. c-erbB-2 and p53 were negative in all animals in group A. In group B, PCNA was positive in Barrett's esophagus with high-grade dysplasia and adenocarcinoma; the PCNA LI was 77. p53 was positive in adenocarcinoma. c-erbB-2 was negative in adenocarcinoma. CONCLUSIONS; The results of this study provided evidence of the dysplasia-carcinoma sequence arising from alkaline reflux, as well as from acid reflux. To the best of our knowledge, this is the first report of the use of an alkaline reflux model and a 6-year study using dogs to observe the course of Barrett's esophagus.

Polymorphisms of two fucosyltransferase genes (Lewis and Secretor genes) involving type I Lewis antigens are associated with the presence of anti-Helicobacter pylori IgG antibody.

Helicobacter pylori attach to the gastric mucosa with adhesin, which binds to Lewis b (Le(b)) or H type I carbohydrate structures. The Secretor (Se) gene and Lewis (Le) gene are involved in type I Le antigen synthesis. The present study was performed to investigate the possibility that Se and Le gene polymorphisms alter the risk of H. pylori infection. Two hundred thirty-nine participants were genotyped for Se and Le and tested for the presence of anti-H. pylori IgG antibodies. Using the normal gastric mucosa from 60 gastric cancer patients, we assessed immunohistochemically whether type I Le antigen expression depended on the Se and Le genotypes. The H. pylori infection rate was positively associated with the number of Se alleles (se/se group, 45.1%; Se/se group, 64.6%; and Se/Se group, 73.3%) and negatively associated with the number of Le alleles (le/le group, 76.4%; Le/le group, 68.3%; and Le/Le group, 55.6%). When the subjects were classified into three groups [low risk, (se/se, Le/Le) genotype; high risk, (Se/Se, le/le), (Se/Se, Le/le), and (Se/se, le/le) genotypes; moderate risk, other than low- or high-risk group], the odds ratio relative to the low-risk group was 3.30 (95% confidence interval, 1.40-7.78) for the moderate-risk group and 10.33 (95% confidence interval, 3.16-33.8) for the high-risk group. Immunohistochemical analysis supported the finding that Se and Le genotypes affected the expression of H. pylori adhesin ligands. We conclude that Se and Le genotypes affect susceptibility to H. pylori infection.

A remodeling system of the 3'-sulfo-Lewis a and 3'-sulfo-Lewis x epitopes.

It has been reported that the chemically synthesized 3'-sulfo-Le(a) and 3'-sulfo-Le(x) epitopes have a high potential as a ligand for selectins. To elucidate the physiological functions of 3'-sulfated Lewis epitopes, a remodeling system was developed using a combination of a betaGal-3-O-sulfotransferase GP3ST, hitherto known alpha1,3/1,4-fucosyltransferases (FucT-III, IV, V, VI, VII, and IX) and arylsulfatase A. The pyridylaminated (PA) lacto-N-tetraose (Galbeta1-3GlcNAcbeta1-3Galbeta1-4Glc) was first converted to 3'-sulfolacto-N-fucopentaose II (sulfo-3Galbeta1-3(Fucalpha1-4)GlcNAcbeta1-3Galbeta1-4Glc)-PA by sequential reactions with GP3ST and FucT-III. The 3'-sulfolacto-N-fucopentaose III (sulfo-3Galbeta1-4(Fucalpha1-3)GlcNAcbeta1-3Galbeta1-4Glc)-PA was then synthesized from lacto-N-neotetraose (Galbeta1-4GlcNAcbeta1-3Galbeta1-4Glc)-PA by GP3ST and FucT-III, -IV, -V, -VI, -VII, or -IX in a similar manner. The substrate specificity for the 3'-sulfated acceptor of the alpha1,3-fucosyltransferases was considerably different from that for the non-substituted and 3'-sialylated varieties. When the GP3ST gene was introduced into A549 and Chinese hamster ovary cells expressing FucT-III, they began to express 3'-sulfo-Le(a) and 3'-sulfo-Le(x) epitopes, respectively, suggesting that GP3ST is responsible for their biosynthesis in vivo. The expression of the 3'-sialyl-Le(x) epitope on Chinese hamster ovary cells was attenuated by the introduction of GP3ST gene, indicating that GP3ST and alpha2,3-sialyltransferase compete for the common Galbeta1-4GlcNAc-R oligosaccharides. Last, arylsulfatase A, which is a lysosomal hydrolase that catalyzes the desulfation of 3-O-sulfogalactosyl residues in glycolipids, was found to hydrolyze the sulfate ester bond on the 3'-sulfo-Le(x) (type 2 chain) but not that on the 3'-sulfo-Le(a) (type 1 chain). The present remodeling system might be of potential use as a tool for the study of the physiological roles of 3'-sulfated Lewis epitopes, including interaction with selectins.

Epigenetic silencing of PEG3 gene expression in human glioma cell lines.

Genomic imprinting, the phenomenon in which alleles of genes are expressed differentially depending on their parental origins, has important consequences for mammalian development, and disturbance of normal imprinting leads to abnormal embryogenesis and some inherited diseases and is also associated with various cancers. In the context of screening for novel imprinted genes on human chromosome 19q13.4 with mouse A9 hybrids, we identified a maternal allele-specific methylated CpG island in exon 1 of paternally expressed imprinted gene 3 (PEG3), a gene that exhibits paternal allele-specific expression. Because PEG3 expression is downregulated in some gliomas and glioma cell lines, despite high-level expression in normal brain tissues, we investigated whether the loss of PEG3 expression is related to epigenetic modifications involving DNA methylation. We found monoallelic expression of PEG3 in all normal brain tissues examined and five of nine glioma cell lines that had both unmethylated and methylated alleles; the remaining four glioma cell lines exhibited gain of imprinting with hypermethylated alleles. In addition, treatment of glioma cell lines with the DNA demethylating agent 5-aza-2'-deoxycytidine reversed the silencing of PEG3 biallelically. In this article, we report that the epigenetic silencing of PEG3 expression in glioma cell lines depends on aberrant DNA methylation of an exonic CpG island, suggesting that PEG3 contributes to glioma carcinogenesis in certain cases.

Molecular cloning and characterization of UDP-GlcNAc:lactosylceramide beta 1,3-N-acetylglucosaminyltransferase (beta 3Gn-T5), an essential enzyme for the expression of HNK-1 and Lewis X epitopes on glycolipids.

A new member of the UDP-N-acetylglucosamine:beta-galactose beta1,3-N-acetylglucosaminyltransferase (beta3Gn-T) family having the beta3Gn-T motifs was cloned from rat and human cDNA libraries and named beta3Gn-T5 based on its position in a phylogenetic tree. We concluded that beta3Gn-T5 is the most feasible candidate for lactotriaosylceramide (Lc(3)Cer) synthase, an important enzyme which plays a key role in the synthesis of lacto- or neolacto-series carbohydrate chains on glycolipids. beta3Gn-T5 exhibited strong activity to transfer GlcNAc to glycolipid substrates, such as lactosylceramide (LacCer) and neolactotetraosylceramide (nLc(4)Cer; paragloboside), resulting in the synthesis of Lc(3)Cer and neolactopentaosylceramide (nLc(5)Cer), respectively. A marked decrease in LacCer and increase in nLc(4)Cer was detected in Namalwa cells stably expressing beta3Gn-T5. This indicated that beta3Gn-T5 exerted activity to synthesize Lc(3)Cer and decrease LacCer, followed by conversion to nLc(4)Cer via endogenous galactosylation. The following four findings further supported that beta3Gn-T5 is Lc(3)Cer synthase. 1) The beta3Gn-T5 transcript levels in various cells were consistent with the activity levels of Lc(3)Cer synthase in those cells. 2) The beta3Gn-T5 transcript was presented in various tissues and cultured cells. 3) The beta3Gn-T5 expression was up-regulated by stimulation with retinoic acid and down-regulated with 12-O-tetradecanoylphorbol-13-acetate in HL-60 cells. 4) The changes in beta3Gn-T5 transcript levels during the rat brain development were determined. Points 2, 3, and 4 were consistent with the Lc(3)Cer synthase activity reported previously.

Construction of 700 human/mouse A9 monochromosomal hybrids and analysis of imprinted genes on human chromosome 6.

As an in vitro assay system for the identification of human imprinted genes, a library of human/mouse A9 monochromosomal hybrids containing a single, intact bsr-tagged human chromosome of known parental origin, derived from normal human fibroblasts, has been previously generated by microcell-mediated chromosome transfer (MMCT). To supplement this assay system, we constructed additional 700 A9 monochromosomal hybrids, using a pSTneo or pPGKneo selection marker. To validate the A9 hybrids, we screened them with chromosome-specific polymorphic markers, and identified the hybrids containing either human chromosome 6, 7, 14, 18, or 21 of known parental origin. Matching paternal and maternal chromosome pairs of A9 hybrids were identified for chromosomes 6, 7, 14, and 18. The paternal-specific expression of ZAC (zinc finger protein, which regulates apoptosis and cell cycle arrest) and HYMAI (hydatidiform mole-associated and imprinted transcript), and the maternal-specific methylation of a CpG island within an imprinted domain on human chromosome 6q24, were maintained in A9 hybrids. For an example, we profiled the expression of expressed sequence tags (ESTs) and the methylation of CpG islands in the 300-kb imprinted domain around 6q24, which may be associated with cancers and transient neonatal diabetes mellitus (TNDM). Thus, the 700 A9 hybrids should be useful for various aspects of imprinting studies.

CD15 expression in mature granulocytes is determined by alpha 1,3-fucosyltransferase IX, but in promyelocytes and monocytes by alpha 1,3-fucosyltransferase IV.

The CD15 carbohydrate epitope is expressed in mature human neutrophils, monocytes, and promyelocytes. We aimed to determine the alpha1,3-fucosyltransferase responsible for the expression of CD15 in each subpopulation of leukocytes. Three alpha1,3-fucosyltransferases, FUT4, FUT7, and FUT9, are expressed in human leukocytes. We demonstrated that FUT9 exhibits 20-fold stronger activity for CD15 synthesis than FUT4, whereas FUT4 exhibits 4.5-fold stronger activity for CDw65 synthesis than FUT9. By competitive reverse transcriptase-polymerase chain reaction, FUT9 was found to be strongly expressed in mature granulocytes and peripheral blood mononuclear cell, but not in monocytes. CD34(+) and CD15(+) cells in cord blood and myeloid cell lines (HL-60 and U937) did not express FUT9 at all. FUT4 transcripts were ubiquitously expressed in all blood cells and all cultured cell lines, with HL-60 and U937 cells in particular expressing a number of FUT4 transcripts. Transfection of the FUT9 gene into Jurkat and U937 cells demonstrated that FUT9 has the potential to express CD15 in myeloid and lymphoid cells. These findings suggest that the expression of CD15 in mature granulocytes is directed by FUT9, whereas it is determined in promyelocytes and monocytes by FUT4. Measurement of CD15 synthesizing activity in cell homogenates of each cell population using the polylactosamine acceptor further supported these conclusions.

Comparison of fit and fill between anatomic stem and straight tapered stem using virtual implantation on the ORTHODOC workstation.

The objective of this article was to determine the influence of stem design on fit and fill using the preoperative planning workstation of the ROBODOC system. Anatomic ABG and straight Osteolock femoral components were virtually implanted into 50 femora (25 from patients with developmental dysplasia of the hip (DDH), 25 morphologically normal) on the workstation display. Fit and fill, and length of the proximal posterolateral femoral cortex removed by milling (LPFCR), were measured on the cross-sectional images. Lateral curvature (alpha angle) and anteversion of the femur were evaluated. The ABG components showed significantly better fit than the Osteolock components at the levels proximal to the lesser trochanter. The Osteolock components showed significantly greater LPFCR than the ABG components, especially in the patients with DDH. The patients with DDH showed significantly greater alpha angle and femoral anteversion than those with morphologically normal femora. With the Osteolock components, the alpha angle correlated significantly with femoral anteversion and LPFCR. Use of an anatomic proximal body of the stem helped to improve the proximal canal fit. Greater LPFCR was required when a straight stem was implanted in patients with a relatively high alpha angle.

Expression of cutaneous lymphocyte-associated antigen regulated by a set of glycosyltransferases in human T cells: involvement of alpha1, 3-fucosyltransferase VII and beta1,4-galactosyltransferase I.

Cutaneous lymphocyte-associated antigen (CLA), which plays a key part in skin homing of human CD4+ memory T cells via CLA/E-selectin binding, is upregulated by IL-12 and downregulated by IL-4. Although alpha1,3-fucosyltransferase VII is essential for synthesis of the CLA carbohydrate epitope, little is known about how the CLA expression is regulated by a number of glycosyltransferases. A 6 wk long-term culture for the in vitro differentiation of naïve Th cells to memory Th1 cells was employed. By repeated activation in the presence of IL-12, naïve T cells differentiated into memory Th1 cells, resulting in the upregulation of CLA expression. The switching of cytokine from IL-12 to IL-4 at three cycles resulted in a marked downregulation of CLA. The transcript levels of 16 glycosyltransferases and P-selectin glycoprotein ligand-1, all considered to be potentially involved in CLA synthesis, were determined after each cycle. The level of CLA expression was well correlated with the amounts of alpha1,3-fucosyltransferase VII and beta1,4-galactosyltransferase I. Both were upregulated by IL-12 and downregulated by IL-4. In particular, alpha1,3-fucosyltransferase VII levels decreased markedly in the presence of IL-4. P-selectin glycoprotein ligand-1 and Core 2 beta1, 6-N-acetylglucosaminyltransferase were progressively up-regulated by repeated IL-12 stimulation, but they were not downregulated by IL-4. The transcript levels of some genes examined were constitutive without any correlation to CLA expression. These results suggest that the level of CLA expression is determined by alpha1, 3-fucosyltransferase VII and beta1,4-galactosyltransferase I, the other enzymes merely participating in the synthesis of CLA. In peripheral blood mononuclear cells, IL-12 and IL-4 profoundly upregulated and downregulated the alpha1,3-fucosyltransferase VII transcripts, respectively, but not the beta1,4-galactosyltransferase I ones, within only 2 h of in vitro culture. This suggested that alpha1,3-fucosyltransferase VII is transcriptionally regulated directly by IL-12 and IL-4.

Multipoint imprinting analysis in sporadic colorectal cancers with and without microsatellite instability.

Disrupted imprinting is implicated in certain tumorigenesis. Since aberrant methylation has been described for a majority of microsatellite instability (MSI)-positive sporadic colorectal cancers, we have investigated alteration to the imprinting in 55 sporadic colorectal cancers with or without MSI. Loss of imprinting (LOI) of IGF2 and PEG1/MEST was observed in 42% and 35% of informative cancers, respectively. H19 expression was not detected in 24% of informative cancers. SNRPN and NDN retained monoallelic expression in all the cancers examined. These findings indicate no simultaneous disruption of the imprinted genes. LOI of IGF2 and PEG1/MEST was also observed in colorectal mucosa from almost all the patients with LOI in tumor tissue. Moreover, MSI-positive colorectal cancers exhibit LOI of IGF2 with a high frequency compared to MSI-negative cancers (P=0.013). These observations, consistent with a previous report, establish an association between LOI of IGF2 and MSI in colorectal cancers and provide insight into susceptibility of tumor development.

Molecular mechanism involved in increased expression of sialyl Lewis antigens in ductal carcinoma of the pancreas.

Increased expression of sialyl Lewis antigens, sLe(a) and sLe(x), is frequent during malignant transformation and tumor progression of gastrointestinal cancers and it is assumed to be correlated with the increased metastatic potential of tumor cells and, consequently, with poor patient survival. To determine the influence of the increased expression of these antigens in disease progression of ductal carcinoma of the pancreas, immunohistochemical expressions of sLe(x) and sLe(a) in 51 ductal carcinomas of the pancreas were examined. We also examined the expression of glycosyltransferase genes, which are involved in the synthetic pathways of these antigens to understand the molecular mechanism involved in the increased expression of these antigens. Of the 51 primary ductal carcinomas of the pancreas, 40 (78.4%) were sLe(a)-positive and 11 (21.6%) were sLe(a)-negative; 16 (31.4%) were sLe(x)-positive and 35 (68.6%) were sLe(x)-negative. Although there were no significant differences in any examined clinicopathological factors such as age, sex, histological type, tumor size, presence of lymph node metastasis, or presence of vessel invasion between the positive and negative groups with both the sLe(a) and sLe(x) antigens, patient survival tended to be worse in the antigens-positive group than in the antigens-negative group. Increased expression, however, was not dependent on the increased expression of a single glycosyltransferase gene examined among five such genes, which are postulated to be responsible for the synthesis of the sLe(a) and sLe(x) epitopes in the glycosylation pathway. Furthermore, the increased expression of these antigens was not closely associated with mutations status of the K-ras or p53 genes. These findings suggested that increased expression of sialyl Lewis antigens are involved in pancreatic tumorigenesis and that the accumulation of genetic alterations or epigenetic changes is responsible for the molecular mechanisms of increased expression of the sLe(a) and sLe(x) antigens in ductal carcinomas of pancreas.

A novel glycosyltransferase with a polyglutamine repeat; a new candidate for GD1alpha synthase (ST6GalNAc V)(1).

The fifth type GalNAcalpha2,6-sialyltransferase (mST6GalNAc V) was cloned from a mouse brain cDNA library. mST6GalNAc V exhibited type II transmembrane topology containing a polyglutamine repeat, which showed 42.6% and 44.8% identity to mouse ST6GalNAc III and IV, respectively. Northern blot analysis revealed that the mST6GalNAc V gene was specifically expressed in forebrain and cerebellum. mST6GalNAc V exhibited GD1alpha synthetic activity from GM1b the same as mST6GalNAc III and IV. The activity ratio of GM1b toward fetuin and the expression pattern were completely different among the three ST6GalNAcs. Interestingly, the polyglutamine repeat number was different from that of inbred mice. We report the first glycosyltransferase with a polymorphic polyglutamine repeat.

Cloning and expression of a human gene encoding an N-acetylgalactosamine-alpha2,6-sialyltransferase (ST6GalNAc I): a candidate for synthesis of cancer-associated sialyl-Tn antigens.

The sialyl-Tn (sTn) antigen is a well known cancer-associated antigen, the expression of which is related to the prognosis of cancer patients. We aimed to isolate a human gene encoding an N -acetylgalactosamine alpha2,6-sialyltransferase which synthesizes sTn antigen, and to characterize the enzyme. Degenerate primers encoding sialyl motifs were used for the polymerase chain reaction to amplify complementary DNAs prepared from RNAs of human pyloric mucosae with intestinal metaplasia, which abundantly expressed sTn antigen, followed by screening of full-length cDNAs using the amplified DNA fragment as a probe. We isolated two human cDNA clones, long-form (2.46 kb) and short-form (2.23 kb) cDNAs. The former encodes an active enzyme with a predicted 600 amino acid sequence. The latter, a splice-variant of the long-form, encodes an inactive enzyme. HCT15 human colorectal cancer cells stably expressing the long-form cDNA expressed sTn epitopes on O -glycans. The long form cDNA was considered to encode a human homologue of chick ST6GalNAc I for the following reasons: (1) the putative amino acid sequence showed greater homology to that of chick ST6GalNAc I (55%) compared to other sialyltransferases, (2) it encodes the extraordinarily long stem region that is a typical feature of chick ST6GalNAc I, and (3) the substrate specificity was very similar to that of chick ST6GalNAc I. In situ hybridization demonstrated that the localization of transcripts correlated well with that of sTn antigen in gastric cancer cells and Goblet cells in intestinal metaplastic glands. Thus, we determined that the long-form cDNA of the human ST6GalNAc I gene encodes the probable candidate for the human sTn synthase(s).

Up-regulation of Lewis enzyme (Fuc-TIII) and plasma-type alpha1,3fucosyltransferase (Fuc-TVI) expression determines the augmented expression of sialyl Lewis x antigen in non-small cell lung cancer.

Sialyl Lewis a and x antigens are well-known tumor-associated antigens expressed in many cancer tissues. The expression of the genes encoding 5 alpha1,3fucosyltransferases, which are able to synthesize the sialyl Lewis antigens, was examined in normal and cancerous lung tissues of patients with non-small cell lung carcinoma. In all 20 cases examined, the transcripts only for the Lewis gene, encoding the Lewis enzyme (alpha1,3/4fucosyltransferase, Fuc-TIII), were abundantly expressed in lung tissue, and interestingly they were markedly up-regulated in the lung cancer tissues of all 20 cases in comparison with normal lung tissues. Myeloid-type alpha1,3fucosyltransferase (Fuc-TIV) was expressed at an intermediate level but was not up-regulated in lung cancer tissues. The transcripts for plasma-type alpha1,3fucosyltransferase (Fuc-TVI) gene were detected at a very low level but were apparently up-regulated in cancer tissues. Fuc-TVI was found to exhibit stronger relative activity for sialyl Lewis x synthesis (almost 6. 4-fold that of Fuc-TIII). The amount of sialyl Lewis x antigen on mucins in the lung cancer tissues was found to be determined by both enzymes, the Lewis enzyme (Fuc-TIII) and Fuc-TVI. However, the amount of the sialyl Lewis a antigens was not determined by any of the alpha1,3-fucosyltransferases, although the expression of sialyl Lewis a antigens definitely required the Lewis enzyme.

Molecular mechanisms of expression of Lewis b antigen and other type I Lewis antigens in human colorectal cancer.

Lewis b (Leb) antigens are gradiently expressed from the proximal to the distal colon, i.e., they are abundantly expressed in the proximal colon, but only faintly in the distal colon. In the distal colon, they begin to increase at the adenoma stage of cancer development and then increase with cancer progression. We aimed to clarify the molecular basis of Leb antigen expression in correlation with the expression of other type I Lewis antigens, such as Lewis a (Lea) and sialylated Lewis a (sLea), in colon cancer cells. Considering the Se genotype and the relative activities of the H and Se enzymes, the amounts of Leb antigens were proved to be determined by both the H and Se enzymes in noncancerous and cancerous colon tissues. But the Se enzyme made a much greater contribution to determining the Lebamounts than the H enzyme. In noncancerous colons, the Se enzyme were gradiently expressed in good correlation with the Leb expression, while the H enzyme was constantly expressed throughout the whole colon. In distal colon cancers, the H and Se enzymes were both significantly upregulated in comparison with in adjacent noncancerous tissues. In proximal colon cancers, expression of the H enzyme alone was highly augmented. The augmented expression of Leb antigens in distal colon cancers is caused mainly by upregulation of the Se enzyme and partly by the H enzymes, while it is caused by upregulation of the H enzyme alone in proximal colon cancers. The Se gene dosage profoundly influences the amounts of the Leb, Lea, and sLea antigens in whole colon tissues, regardless of whether they are noncancerous or cancerous tissues. It suggests that the Se enzyme competes with alpha2,3 sialyltransferase(s) and the Le enzyme for the type I acceptor substrates.

Cloning, expression, and characterization of a novel UDP-galactose:beta-N-acetylglucosamine beta1,3-galactosyltransferase (beta3Gal-T5) responsible for synthesis of type 1 chain in colorectal and pancreatic epithelia and tumor cells derived therefrom.

The sialyl Lewis a antigen is a well known tumor marker, CA19-9, which is frequently elevated in the serum in gastrointestinal and pancreatic cancers. UDP-galactose:N-acetylglucosamine beta1, 3-galactosyltransferase(s) (beta3Gal-Ts) are required for the synthesis of the sialyl Lewis a epitope. In the present study, a novel beta3Gal-T, named beta3Gal-T5, was isolated from a Colo205 cDNA library using a degenerate primer strategy based on the amino acid sequences of the four human beta3Gal-T genes cloned to date. Transfection experiments demonstrated that HCT-15 cells transfected with the beta3Gal-T5 gene expressed all the type 1 Lewis antigens. In gastrointestinal and pancreatic cancer cell lines, the amounts of beta3Gal-T5 transcripts were quite well correlated with the amounts of the sialyl Lewis a antigens. The beta1,3Gal-T activity toward agalacto-lacto-N-neotetraose was also well correlated with the amounts of beta3Gal-T5 transcripts in a series of cultured cancer cells, and in Namalwa and HCT-15 cells transfected with the beta3Gal-T5 gene. Thus, the beta3Gal-T5 gene is the most probable candidate responsible for the synthesis of the type 1 Lewis antigens in gastrointestinal and pancreatic epithelia and tumor cells derived therefrom. In addition, beta3Gal-T5 is a key enzyme that determines the amounts of the type 1 Lewis antigens including the sialyl Lewis a antigen.

Molecular behavior of mutant Lewis enzymes in vivo.

The expression of type-1 Lewis antigens on erythrocytes and in digestive organs is determined by a Lewis type alpha(1,3/1, 4)-fucosyltransferase (Lewis enzyme) encoded by the Fuc-TIII gene ( FUT3 gene; Lewis gene). We have classified the Lewis alleles in the Japanese population into four types, the wild-type allele ( Le ) and three mutated alleles, i.e., le1, which has missense mutations T59G and G508A, le2, which has T59G and T1067A, and le3, which has only T59G. Here we carried out an extensive study on the biological properties of the three mutant Lewis enzymes, the le1, le2, and le3 enzymes, using native tissues and obtained the following results. (1) In in vivo and in vitro experiments, the le1 and le2 enzymes were found to be susceptible to protease digestion probably because the one missense mutation in the catalytic domains, i.e., Gly170 to Ser in the le1 enzyme and Ile356 to Lys in the le2 enzyme, makes the three-dimensional structures of the enzymesunstable, while the le3 and wild-type Lewis enzymes wereresistant to protease digestion. (2) The le1 and le2 enzymes cannot synthesize type 1 Lewis antigens on either glycolipids or mucins. The le3 enzyme cannot synthesize Lewis-active glycolipids, which result in the Lewis antigen-negative phenotype of erythrocytes, while it can synthesize Lewis antigens on mucins in normal and cancerous colon tissues. The missense mutation, Leu20 to Arg, in the transmembrane domain reduces retention of the le3 enzyme in the Golgi membrane resulting in an apparent reduction of enzyme activity as revealed by the lack of Lewis antigen synthesis. (3) The Lewis gene dosage actually has effects in vivo on the amount of the Lewis enzyme, its activity, and finally the amounts of Lewis carbohydrate antigens. This is the first article that clearly demonstrates the gene dosage effects on the amount of the glycosyltransferase protein, its activity, and the amounts of carbohydrate products in vivo.

The aberrant expression of Lewis a antigen in intestinal metaplastic cells of gastric mucosa is caused by augmentation of Lewis enzyme expression.

Immunohistochemical staining showed an aberrant expression of Le(a) antigen in the intestinal metaplastic glands of the gastric mucosa of secretors, as reported by others. In this study, we have demonstrated for the first time that the Lewis enzyme is well colocalized with Le(a) antigen, indicating that the Lewis enzyme is responsible for Le(a) antigen synthesis in the gastric mucosa. The staining intensity of the Lewis enzyme was much stronger in the cells with intestinal metaplasia than the cells without metaplasia, regardless of the secretor status. The amount of transcript of the Lewis gene was related to the degree of metaplasia; i.e., the more severe the metaplastic change was, the more abundantly the transcripts of the Lewis gene were expressed. This augmentation of the Lewis enzyme in metaplastic tissues was also confirmed by Western blotting analysis using a specific antibody against the Lewis enzyme. We conclude that intestinal metaplastic change of gastric mucosa is usually accompanied by a marked augmentation of the Lewis enzyme expression, which results in the enhanced expression of Le(a) antigens, particularly in secretors.

Telomerase activity and microsatellite instability in colorectal cancer and adenoma.

In order to examine their roles in carcinogenesis or in progression of colorectal carcinoma, we investigated telomerase activity and microsatellite instability in 67 non-familial colorectal cancers and in 18 adenomas. The incidence of detectable telomerase activity increased from 22% of normal colorectal mucosas adjacent to carcinoma, and 33% of adenomas, to 75% of carcinomas. On the other hand, the incidence of detectable microsatellite instability in carcinomas (30%) was almost the same as in adenomas (22%). No significant correlation was detected in the incidence of telomerase activity and microsatellite instability in carcinomas or in adenomas. Moreover, the incidence of telomerase activity and microsatellite instability did not increase during the progression of carcinomas. These results indicate that telomerase activity and microsatellite instability are independent events in colorectal carcinogenesis, and that telomerase activity and microsatellite instability are not correlated with the progression of colorectal carcinoma. However, in 13 multiple cancers, the incidence of telomerase activity (92%) and the incidence of microsatellite instability (54%) was higher than that of telomerase activity (70%) and that of microsatellite instability (24%) in 54 sporadic cancers. Moreover, the incidence of telomerase activity and that of microsatellite instability in adenomas with carcinomas (45% and 36% respectively) was higher than that of telomerase activity and microsatellite instability in adenomas without carcinomas (14% and 0% respectively). These results indicate that telomerase activity and microsatellite instability may play an important role in multicentric carcinogenesis in colorectal carcinoma.