Developmental expression of a unique carbohydrate antigen, Tn antigen, in mouse central nervous tissues.

Using an anti-Tn monoclonal antibody, the Tn antigen was detected immunohistochemically in prenatal and early postnatal central nervous tissues. On embryonic day 9 (E9), the antigen was distributed throughout the single neuroepithelial layer in the neocortex and then became more prominent in the preplate than in the ventricular zone along with formation of the preplate. Following division of the preplate and concomitant formation of the cortical plate, distinct labeling of the neocortex occurred in the marginal, subplate and intermediate zones, whereas in the cortical plate and ventricular zone were virtually not immunostained. It is notable that thalamocortical afferent fibers were also immunostained specifically on E14. After birth, the localization of the antigen became less noticeable and by 3 weeks after birth, the antigen had substantially disappeared. In the developing cerebellum, prominent labeling was also observed in the molecular layer and outskirts of the cerebellar nuclei on early postnatal days. To characterize the glycoprotein bearing the Tn antigen biochemically, immunoblot analysis was performed. The glycoprotein, most of which was extracted with a salt solution, migrated as a broad smeared band corresponding to a molecular weight of about 250 kDa on SDS-PAGE. Among the various tissues examined, this glycoprotein was only detected in the brain and its amount increased until an early postnatal stage with a peak on postnatal day 3 (P3), and then decreased gradually with age. This spatially and developmentally regulated expression of the Tn antigen suggests that this antigen plays a significant role in brain development.

High density O-glycosylation of the MUC2 tandem repeat unit by N-acetylgalactosaminyltransferase-3 in colonic adenocarcinoma extracts.

A synthetic peptide corresponding to the human MUC2 tandem repeat unit was glycosylated in vitro using UDP-GalNAc and extracts of colonic adenocarcinoma and paired normal mucosa, followed by fractionation of the products by reverse phase high-performance liquid chromatography. Several peaks of glycopeptides with different numbers of GalNAc residues attached were detected. It is notable that the adenocarcinoma extract was capable of glycosylating peptides to a much greater extent than was normal mucosa. The levels of mRNA for N-acetylgalactosaminyltransferases-1, -2, and -3 were determined by reverse transcription-PCR. Only N-acetylgalactosaminyltransferase-3 mRNA was expressed at a higher level in the adenocarcinoma than in the normal tissue. When the MUC2 tandem repeat peptide was glycosylated with a mixture of the normal mucosa extract and recombinant N-acetylgalactosaminyltransferase-3, larger amounts of glycopeptides with higher contents of GalNAc residues were produced. The MUC2 tandem repeat peptides glycosylated extensively by recombinant N-acetylgalactosaminyltransferase-1, -2, or -3 were prepared and characterized. Substitution at each Thr residue, as revealed by Edman degradation sequencing, in conjunction with evidence obtained on mass spectrometry indicated a heterogeneous pattern of site-specific glycosylation within the MUC2 tandem repeat. It was found that maximum numbers of 6, 8, and 11 GalNAc residues were incorporated by N-acetylgalactosaminyltransferases-1, -2, and -3, respectively, and that only N-acetylgalactosaminyltransferase-3 could completely glycosylate both consecutive sequences composed of three and five Thr residues in the MUC2 tandem repeat unit. These results suggest that O-glycosylation of the clustered Thr residues is a selective process controlled by N-acetylgalactosaminyltransferase-3 in the synthesis of clustered carbohydrate antigens.

Identification of the core protein carrying the Tn antigen in mouse brain: specific expression on syndecan-3.

We isolated glycoproteins carrying the Tn antigen, which was expressed spatiotemporally in the developing mouse brain. The Tn antigen was expressed on two molecular species with a molecular weight from 200 to 350 kDa and 110 to 160 kDa, as judged on SDS-PAGE. Although the two glycoproteins showed different susceptibilities to heparitinase I and solubilities in a salt solution, after treatment with V8 protease they showed the same mobility corresponding to a molecular weight of 90 kDa on SDS-PAGE, suggesting that these two molecules shared a common core protein. Partial N-terminal sequences of the glycoproteins were determined, i.e. AQRXRNENFERPV and ALAAPXAPAMLP, which were identified as the sequences of the N-terminal and central portions of syndecan-3, respectively. Both glycoproteins were reactive to anti-mouse syndecan-3 antibody. These results suggest that one is a soluble syndecan-3 cleaved between mucin-like domain and transmembrane domain, and the other is a membrane-bound syndecan-3 lacking N-terminal glycosaminoglycan attachment sites, and that both glycoproteins have a mucin-like domain characteristic of syndecan-3, in which the Tn antigen may be expressed.

Stimulation of macrophages by mucins through a macrophage scavenger receptor.

We found that phorbol ester-primed THP-1 cells (a human monocyte cell line), which express a scavenger receptor, were stimulated by mucins through the macrophage scavenger receptor, resulting in enhanced secretion of IL-1beta. The activity was abolished by treatment of the mucins with sialidase, indicating that sialic acid is involved in binding. (125)I-Labeled ovine submaxillary mucin could bind to COS 7 cells transfected with cDNA encoding the scavenger receptor. Binding was inhibited by mucins, fucoidan, and polyinosinic acid but not by polycytidylic acid, this being consistent with the characteristics of the scavenger receptor. When phorbol ester-primed THP-1 cells were cocultured with colon cancer cells producing mucins, IL-1beta secreted from the THP-1 cells increased significantly. Adhesion between colon cancer cells and a scavenger receptor transfectant was observed, and binding was inhibited partly by mucins and ligands for the scavenger receptor.

Binding characteristics of an anti-Siaalpha2-6GalNAcalpha-Ser/Thr (sialyl Tn) monoclonal antibody (MLS 132).

To determine the epitopic structure for an anti-Siaalpha2-6GalNAcalpha-Ser/Thr (anti-sialyl Tn) monoclonal antibody, MLS 132, ovine submaxillary mucin (OSM) was digested with the combination of trypsin and thermolysin and the digest fractionated by immunoaffinity column chromatography and HPLC. From tryptic digest, a major glycopeptide designated as T3 was obtained as an immunoaffinity column-bound fraction. On solid-phase radioimmunoassay, it was found that T3 exhibited strong immunoreactivity with MLS 132. On treatment with thermolysin, T3 was converted into about 50 fragments, as found on fractionation by HPLC. Several of them were strongly immunoreactive and had the same amino acid sequence, i.e. Phe-Ser*-Gly-Glu-Thr*-Ser*-Thr*-Thr*-Val-Ile-Ser*-Gly-Thr*-Asn-Val, where asterisks denote the sites of attachment of carbohydrate. Of these, one was fully sialylated, the others having one Ser or Thr with unsialylated GalNAc attached. Results of analyses of the carbohydrate attached in these glycopeptides led us to postulate that a cluster composed of four sialyl Tn antigens is the essential epitopic structure for MLS 132.

Glycosylation of the tandem repeat unit of the MUC2 polypeptide leading to the synthesis of the Tn antigen.

A synthetic peptide corresponding to the human MUC2 tandem repeat domain containing 14 Thr residues was glycosylated in vitro using UDP-GalNAc and microsomal membranes of the colorectal cancer cell line, LS180. The products were fractionated by reverse phase HPLC, which gave seven glycopeptide fractions. Their molecular weights were estimated by matrix-assisted laser desorption/ionization mass spectrometry, the values obtained corresponding to glycopeptides containing from one to ten GalNAc residues. On solid phase radioimmunoassaying involving a monoclonal anti-Tn antibody (MLS128), it was found that the glycopeptides containing nine or ten GalNAc residues were strongly immunoreactive, whereas the glycopeptides containing less than six GalNAc residues were inactive, indicating that a cluster of GalNAc-Thr is essential for the Tn antigenicity.

Three-step tumor imaging with biotinylated monoclonal antibody, streptavidin and 111In-DTPA-biotin.

The purpose of this study was to test the three-step targeting of tumors in mice using biotinylated antibody, streptavidin and radiolabeled biotin. Nude mice bearing subcutaneous LS180 human colon cancer xenografts were intravenously administered with 200 microg of the biotinylated anti-Tn monoclonal antibody MLS128, and 2 days later they got intravenous injection of 50 microg of streptavidin. They were intravenously injected 1, 4 or 7 days later with 0.5 microg of 111In-diethylenetriamine pentaacetic acid (DTPA)-biotin. The tumor uptake, determined 2 h later, was 1.4, 0.5 and 0.6% injected dose/gram of tissue (ID/g), respectively, and the blood radioactivity was 1.0, 0.2 and 0.2% ID/g, respectively. When the interval between the streptavidin and radiolabeled biotin injections was prolonged from 1 day to 7 days, the tumor-to-blood ratio 2 h after injection of 111In-labeled biotin increased from 1.5 to 4.0. Clear tumor images were obtained as early as 2 h after injection of radiolabeled biotin. In conclusion, these preliminary data suggested that the three-step method using the streptavidin-biotin system would be applicable in an experimental mouse tumor model and provides images of tumors rapidly and clearly after injection of radiolabeled biotin.

Effect of administration route and dose of streptavidin or biotin on the tumor uptake of radioactivity in intraperitoneal tumor with multistep targeting.

The effect of the administration route and dose of streptavidin or biotin on the biodistribution of radioactivity in multistep targeting was studied in nude mice bearing intraperitoneal (IP) colon cancer xenograft. The multistep targeting included a two-step method using biotinylated antibody and radiolabeled streptavidin and a three-step method with radiolabeled biotin based on the two-step method. A monoclonal antibody, MLS128, which recognizes Tn antigen on mucin, was biotinylated and injected intravenously (i.v.) or i.p. in nude mice bearing human colon cancer LS180 IP xenografts for pretargeting. In the two-step method, i.p.-injected streptavidin showed a higher tumor uptake and tumor-to-nontumor ratios than i.v.-injected streptavidin regardless of administration route of pretargeting. The tumor uptake of radiolabeled streptavidin was increased with a high dose of biotinylated antibody pretargeting, but decreased with an increasing dose of streptavidin. In the three-step targeting, i.p. injection also gave a higher tumor uptake of radiolabeled biotin than i.v. injection. In conclusion, i.p. administration of radiolabeled streptavidin or biotin resulted in more efficient IP tumor targeting with the multistep methods.

Improved intratumoral penetration of radiolabeled streptavidin in intraperitoneal tumors pretargeted with biotinylated antibody.

UNLABELLED: Inefficient intratumoral penetration of pharmaceuticals is one of the major limiting factors against effective tumor-targeting therapy. This study investigated the effect of the distribution pattern of the binding site in tumors on the penetration of target material. METHODS: In the first experiment, radiolabeled biotinylated monoclonal antibody, MLS128, was injected intraperitoneally or intravenously into nude mice bearing intraperitoneal human colon cancer xenografts. In the second experiment, radiolabeled streptavidin was injected intraperitoneally in the tumor-bearing mice after the pretargeting with the unlabeled biotinylated antibody. Intratumoral distribution of radioactivity was examined with quantitative autoradiography. RESULTS: There was no difference in the biodistribution of biotinylated antibody between intraperitoneal and intravenous administrations, but autoradiography showed a higher uptake in the margin and a lower uptake in the center of radioactivity in tumor nodules with intraperitoneal injection and a more uniform intratumoral radioactivity distribution with intravenous injection. In the two-step method, radioactivity in a low dose of streptavidin with intraperitoneal pretargeting primarily localized at the tumor margin. By increasing the dose, streptavidin penetrated more deeply. In tumors with intravenous pretargeting, a more uniform intratumoral distribution of streptavidin was obtained. The biodistribution of radiolabeled streptavidin was the same between different pretargeting routes. CONCLUSION: The better intratumoral penetration of radiolabeled streptavidin after intravenous pretargeting than intraperitoneal pretargeting with biotinylated antibody may be the result of different intratumoral distribution of the binding site for the radiolabel.

Anti-murine antibody response to mouse monoclonal antibodies in cancer patients.

Although development of human anti-murine immunoglobulin antibody (HAMA) is often seen in patients receiving murine antibodies, the variety of methods used for detecting HAMA makes it difficult to compare directly the HAMA responses measured by different assays. In the present study, several parameters of the HAMA response to two murine monoclonal antibodies were evaluated. The anti-sialosyl Tn antibody MLS102 and anti-CA125 antibody 145-9, which were labeled with 111In, were injected intravenously into 17 colorectal cancer patients and 11 ovarian cancer patients for immunoscintigraphy, respectively. HAMA was measured by enzyme-linked immunosorbent assay. There was no difference in baseline HAMA levels before antibody injection between the two groups. HAMA developed more frequently in ovarian cancer patients receiving the 145-9 antibody than in colorectal cancer patients receiving the MLS102 antibody (9/11 vs. 6/17, P < 0.05). No significant difference was observed in maximal HAMA levels between the two groups of patients. However, time to reach the maximal levels was delayed and the duration of the response seemed longer in ovarian cancer patients. Among 11 patients receiving the 145-9 antibody three patients became positive for HAMA more than 2 months after antibody injection and the other two had HAMA activity in their sera for more than 17 months. HAMA response was different between the two antibodies, and late onset or long duration of HAMA response against the 145-9 antibody suggests the importance of HAMA measurement in patients who receive a second injection of murine antibodies even after a long interval.

Metabolic changes of E-selectin caused by the binding of a mucin carrying sialylLewis A antigens.

Mucin-type glycoproteins carrying sialylLeA antigens (SL-GP) were isolated from the ascites fluid of a patient with colorectal cancer. SL-GP bound to E-selectin on endothelial cells in Ca2+- and sialylLeA antigen-dependent manners. To examine the metabolic change in E-selectin caused by ligation, endothelial cells were labeled with 32P-phosphate or 35S-Met and 35S-Cys. Phosphorylation at one or more serine residues of E-selectin was elevated by ligation with SL-GP but not with sialylLeA hexasaccharide. Pulse-labeling of E-selectin with 35S-Met and 35S-Cys in the presence of SL-GP indicated that the degradation of E-selectin was accelerated by SL-GP ligation, but labeling after pre-ligation with SL-GP revealed an increase in the synthesis of E-selectin. The synthesis may reflect compensation for the E-selectin degraded on pre-ligation. These results indicate that the overall metabolism of E-selectin was enhanced by the ligation of SL-GP, with degradation and synthesis being apparently balanced.

Intravenous avidin chase improved localization of radiolabeled streptavidin in intraperitoneal xenograft pretargeted with biotinylated antibody.

In the present study, we examined the effect of avidin administered intravenously (i.v.) on the biodistribution of radiolabeled streptavidin in mice bearing intraperitoneal (IP) xenografts pretargeted with biotinylated antibody. Tumors were established in nude mice by IP inoculation of LS180 human colon cancer cells. Monoclonal antibody MLS128, which recognizes Tn antigen on mucin, was biotinylated and injected IP into the IP tumor-bearing mice. Radioiodinated streptavidin was administered IP or i.v. 48 h after pretargeting of biotinylated antibody. Avidin was administered i.v. 30 min prior to streptavidin injection. The localization of radioiodinated streptavidin in the tumor pretargeted with biotinylated antibody was significantly higher than that without pretargeting and that of radioiodinated MLS128 by the one-step method. Avidin administration significantly accelerated the clearance of radioiodinated streptavidin in blood and other normal tissues and increased the tumor-to-blood radioactivity ratio regardless of administration route of streptavidin. The i.v. avidin chase improved tumor localization of radiolabeled streptavidin in the IP xenografts pretargeted with biotinylated antibody.

Novel glycoproteins inhibiting the binding of colorectal cancer cells to E-selectin.

Novel glycoproteins carrying sialyl-LeA (SLeA) antigens (SL-GP) were isolated from ascites fluid from a patient with colorectal cancer by immunoaffinity chromatography. Their characteristics, including binding capacity to E-selectin, were investigated. SL-GP showed a typical mucin type amino acid composition in which Ser, Thr and Pro together accounted for greater than 50% of the total amino acid residues. A large amount of carbohydrate (about 80%) was present in SL-GP. The number of O-glycans carrying SLeA antigens comprised about 9% of the total number of O-glycosidic chains. SL-GP could bind to IL-1 beta treated HUVEC, and the binding was inhibited by anti-E-selectin and anti-SLeA monoclonal antibodies. The binding of colorectal cancer cells, LS 180, to HUVEC was assayed in the presence of SL-GP, oligosaccharides prepared from SL-GP and human milk SLeA hexasaccharide. SL-GP inhibited the binding most effectively, whereas equivalent amounts of the SL-GP oligosaccharides and milk SLeA hexasaccharide inhibited it only slightly. These results constitute direct evidence that a unique arrangement of SLeA antigens on the polypeptide chain, probably a cluster, is essential for the binding to E-selectin.

Accelerated evolution of crotalinae snake venom gland serine proteases.

Eight cDNAs encoding serine proteases isolated from Trimeresurus flavoviridis (habu snake) and T. gramineus (green habu snake) venom gland cDNA libraries showed that nonsynonymous nucleotide substitutions have accumulated in the mature protein-coding regions to cause amino acid changes. Southern blot analysis of T. flavoviridis genomic DNAs using two proper probes indicated that venom gland serine protease genes form a multigene family in the genome. These observations suggest that venom gland serine proteases have diversified their amino acid sequences in an accelerating manner. Since a similar feature has been previously discovered in crotalinae snake venom gland phospholipase A2 (PLA2) isozyme genes, accelerated evolution appears to be universal in plural isozyme families of crotalinae snake venom gland.

Impaired plasticity of neurons in aging. Biochemical, biophysical, and behavioral studies.

Age-related correlation of impaired plasticity of neurons (biochemical and biophysical aspects) and behavioral alterations were investigated in young (3.5 months) and extremely aged (approximately 40 months) female Wistar rats. Age-dependent significant differences in second messenger (cAMP and Ins (1,4,5)P3) concentration and signal transduction via muscarinic and dopaminergic receptors were found. The results point to the specifically impaired coupling between dopamine D1 receptor and GS protein, which underlies normal brain aging. However, cholinergic neurotransmission may be modulated at another level in extremely aged rats. Thus, it appears that the site of affection in coupling of receptor and G protein and/or G protein-dependent signal transduction in aging cannot be generalized. This indicates that alterations in the coupling of signal transduction depend on diverse neurotransmitter receptors with advanced age. The age-dependent alterations in the cAMP and PI signal pathways could be due to changes in the physical properties of the membranes. To support this hypothesis, age-dependent changes in the physical state and the biochemical composition of synaptosomal membranes from the cortex, cerebellum, and striatum were examined by measuring the steady-state fluorescence amisotropy of the membrane probes 1,6-diphenyl-1,3,5-hexatriene (DPH), trimethylammonium-DPH (TMA-DPH), and trimethylammoniumpropyl-DPH (TMAP-DPH). Significant differences in the physical properties of the synaptosomal membranes existed between young and very aged rats, expressed by a higher anisotropy in the 40-month-old rat brain tissue. The changes in the physical properties of the membranes were in line with the determined age-dependent alterations in the chemical composition, e.g., the increase in cholesterol content of the aged membranes.

Expression of the T antigen on a T-lymphoid cell line, supT1.

We have measured glycosyltransferase activities of SupT1 cells, a T-lymphoid cell line shown to react with autoantibodies in the sera of many HIV patients. Since considerable alpha-N-acetylgalactosaminyl-transferase and beta 1, 3 galactosyltransferase activities were found in SupT1 cells, at least the O-glycan core 1 structure can probably be synthesized. FACS analysis using an anti-T monoclonal antibody showed expression of the T antigen (Gal beta 1-3 GalNAc). Glycoproteins with the T antigen were isolated by immunoprecipitation with the anti-T antibody from a SupT1 cell lysate labelled metabolically with 3H-glucosamine and then analysed by SDS-PAGE. It was revealed that the precipitate contained a glycoprotein with a molecular weight corresponding to that of leukosialin. O-glycans were prepared from the immunoprecipitate by alkaline-borohydride treatment and then fractionated on Bio-Gel P-2, GalNAcOH and Gal-GalNAcOH being identified inter alia. These results suggest that an anti-T antibody may be included in the autoantibodies found in HIV-1 infected individuals.

Improved targeting of radiolabeled streptavidin in tumors pretargeted with biotinylated monoclonal antibodies through an avidin chase.

UNLABELLED: Radiolabeled streptavidin can be accumulated in tumors pretargeted with biotinylated anti-tumor antibodies. However, circulating biotinylated antibody and endogenous biotin may interfere with the tumor targeting of streptavidin. To reduce biotinylated antibody concentration in the blood, we injected avidin before streptavidin administration. The effects of avidin administration on the biodistribution and tumor targeting of radiolabeled streptavidin were examined. METHODS: Biotinylated anti-human colon cancer monoclonal antibody (MAb) MLS128 was injected intravenously into nude mice bearing human colon cancer xenografts for pretargeting. After intraperitoneal injection of avidin, radioiodinated streptavidin was administered and its biodistribution and tumor accumulation was investigated. RESULTS: Radioiodinated streptavidin specifically localized in the tumor pretargeted with biotinylated antibody. Avidin preadministration accelerated the tumor uptake and blood clearance of radioiodinated streptavidin. The tumor-to-blood radioactivity ratio at 6 and 24 hr after radiolabeled streptavidin injection were 1.23 +/- 0.29 and 3.04 +/- 0.86, respectively, in mice with avidin chase (mean +/- s.d., n = 7), and 0.82 +/- 0.17 and 2.29 +/- 0.29, respectively, in those without chase (mean +/- s.d., n = 7). CONCLUSION: Localization of radiolabeled streptavidin in tumors pretargeted with biotinylated MAb could be improved by avidin chase. This approach may be useful for tumor radioimmunoimaging and radioimmunotherapy.

Radioimmunoimaging of colon cancer xenografts with anti-Tn monoclonal antibody.

Tn antigen is a glycosylated tumor associated antigen and a murine monoclonal antibody, MLS128, has been identified to react with it. The potential of MLS128 for the radioimmunoimaging of colorectal cancer was studied. MLS128 was labeled with radioiodine by the chloramine-T method or indium-111 (111In) by using isothiocyanatobenzyl EDTA, and was injected into nude mice bearing human colon cancer xenografts. Radiolabeled MLS128 showed a high and specific localization in xenografted tumor. At 48 h after injection, the %ID/g of 125I-labeled MLS128 in the tumor was 34.69, whereas that of isotype matched control antibody, FLOPC21, was 5.58 and the tumor-to-nontumor radioactivity ratios of 125I-labeled MLS128 reached to 4.56, 17.84 and 23.62 for the blood, liver and bone, respectively. 111In-labeled MLS128 showed similar results. High accumulation of MLS128 in xenografted tumors suggested that the monoclonal antibody MLS128 is promising for radioimmunoimaging of colorectal cancer.

Distribution of Tn antigen recognized by an anti-Tn monoclonal antibody (MLS128) in normal and malignant tissues of the digestive tract.

Alterations in the normal glycosylation process are often associated with oncogenic transformation. Using an anti-Tn monoclonal antibody, MLS128, we have investigated the immunohistochemical localization of Tn antigen in normal and malignant tissues of the digestive tract. In normal tissues, MLS128 was immunoreactive with the squamous epithelium of the esophagus and was weakly reactive with the columnar epithelia of the stomach, duodenum, colon, bile duct and pancreatic duct. In malignant tissues, positive immunostaining was detected with high frequency (75%-100%) in carcinomas of the esophagus, stomach colon, biliary tract and pancreas, whereas 2 of 11 (18%) hepatocellular carcinomas were positive. Tn antigen was detected in the upper two-thirds of the normal squamous epithelium, and was often detected in squamous cell carcinomas with cancer pearls (keratinization). These results suggest that the expression of Tn antigen is related to the differentiation of squamous epithelium, or to keratinization. In normal columnar epithelial cells. Tn antigen was localized mainly to the Golgi area. This intracellular localization was preserved in well-differentiated papillary adenocarcinomas of the colon, but was lost in most cases of tubular adenocarcinomas.