Biomedical subjects
Publications and source records attributed to K Matsubara.
A 3'-directed partial cDNA clone that matches exactly a genomic sequence in GenBank was isolated while collecting transcribed sequences from adult lung by a random approach. This is the first report of active gene identification on genomic sequence without the aid of Northern hybridization.
Alterations of genomic DNAs in primary hepatocellular carcinomas (HCCs) were examined by restriction landmark genomic scanning (I. Hatada et al., Proc. Natl. Acad. Sci. USA, 88: 9523-9527, 1991) which is a 2-dimensional gel analysis that allows detection of deletion, amplification, or other rearrangements of genomic DNA. Sixteen HCC samples together with their normal counterparts were tested in this manner. Each HCC sample was micromanipulated to minimize possible carryover from non-malignant cells. DNAs from HCCs and their normal counterparts were cleaved with the restriction enzyme NotI, end labeled with 32P, and size fractionated by 2-dimensional electrophoresis using HinfI as the second cleavage enzyme. The resulting spots (about 2000) in HCC samples were compared with their normal counterparts. Five spots were more intense in 10-14 of the 16 HCCs (63-88%). The intensity of several spots was reduced to about half, suggesting the loss of one of two alleles. Some of these decreases were observed frequently in different HCC samples, whereas others were sporadic. Sixty of these spots reproducibly decreased in > 2 cases, with 27 showing a decrease in > 50% of the informative cases. The highest incidence was observed in 14 of 16 samples (88%). No significant correlations were observed between these changes in spots and hepatitis B virus or hepatitis B virus infection. The use of landmarks that show a reproducible increase or decrease in intensity is discussed in conjunction with future studies of genomic alterations inherent in HCC.
Thymidine kinase (TK)-deficient cells were established from six human leukemia cell lines to evaluate the role of TK in maintaining intracellular TTP pools. The residual TK activities in mutant cells were less than 3% of those of wild-type strains, except for a B-lymphoid cell line, Ball-1 (8.7%). In a promyelocytic leukemia cell line (HL-60), a splenic B cell line (WI-L2) and Ball-1, a mutational loss of TK resulted in a decrease of TTP pools by 80%, 33% and 54%, respectively. On the other hand, in the T cell lines, Molt-3, Molt-4 and CEM, TTP did not show any significant differences between parent and TK-deficient cells. TK-deficient HL-60 cells had, however, comparable levels of dATP, dGTP and dCTP with wild-type cells. An analysis of growth characteristics showed that the decrease of TTP was not due to the change of the cell cycle distribution. These results indicate that TK plays a different role in maintaining TTP pools among human leukemia cell lines.
The mechanism of cell death induced by calcium ionophore, A23187, was investigated in six human leukemia cell lines. Following exposure to 1 microM A23187, the myelogenous cell lines (HL-60, U-937, KG-1) underwent apoptosis within 3 h as determined by their morphology and DNA fragmentation assay. In contrast, T-lymphoblastic leukemia cell lines (Molt-4, Molt-3, CEM) revealed necrotic cell death after 24 h of incubation. However, an initial rise of intracellular free calcium concentrations and growth inhibition after treatment with A23187 were similar in the two cell types. We further showed that an endonuclease capable of mediating internucleosomal DNA fragmentation was constitutively expressed in the cytosol but not in the nuclei of the myelogenous cell lines, although this endonuclease was not detected in either the nuclei or the cytosol of the T-lymphoblastic cell lines. The activation of the endonuclease in myelogenous cells is calcium-independent and has an optimal pH of 7.5-9. It is inhibited by 1 mM zinc ion or 300 microM aurintricarboxylic acid. We propose that this constitutive endonuclease may be related to the susceptibility of myelogenous leukemia cell lines to apoptotic cell death.
We recently reported that treatment with calcium ionophore, A23187, induces apoptosis in human myelogenous leukemia cells but causes necrotic cell death in T-lymphoblastic leukemia cells. To better understand the underlying mechanisms of such different modes of cell death, we established hybridomas between HL-60 promyelocytic and CEM T-lymphoblastic leukemia cells. The resulting hybridomas were divided into three groups in terms of their susceptibility to apoptosis following exposure to A23187: (1) hybridomas highly sensitive to apoptosis, (2) hybridomas with intermediate sensitivity to apoptosis which occurs later and to a lesser extent, and (3) hybridomas resistant to apoptosis. However, growth inhibition after 72 h of incubation and an initial rise in intracellular free calcium concentrations induced by A23187 were similar in the three groups. Expression of Ca(2+)-independent/Mg(2+)-dependent endonuclease, which had an optimal pH of 7.5-8.5 and was inhibited by Zn2+, was correlated with the susceptibility of the hybridomas to A23187-induced apoptosis. Thus, this endonuclease may play, at least in part, an important role in the induction of apoptosis in leukemia cell lines. Analysis of hybridomas between apoptosis-sensitive and apoptosis-resistant cells is useful in the elucidation of genetic factors which regulate cell death.
The effect of cadmium chloride (CdCl2) on cell proliferation and EDRF (endothelium-derived relaxing factor) production by cultured human umbilical arterial endothelial cells (HUAECs) was investigated. The viability of HUAECs decreased dose-dependently after the addition of Cd (cadmium chloride). Morphologic examination by phase contrast microscopy revealed severe damaging effects of Cd at higher concentrations. The cytotoxic effect of Cd on DNA synthesis was also concentration-dependent. The effect of Cd on EDRF production by indomethacin-treated HUAECs was assessed by its anti-platelet aggregatory effect. Platelet aggregation studies were carried out in cuvettes lined with HUAECs using an aggregometer. The anti-platelet aggregatory effect was decreased dose-dependently by pretreatment with Cd. These findings suggest that HUAECs are susceptible to concentration-dependent Cd cytotoxicity, and that Cd can inhibit the production of EDRF by HUAECs.
When the hepatitis B virus (HBV) replicates, a full-size transcript of the viral genome, called pregenome RNA, is made and is selectively packaged into virus core particles. This selective encapsidation is dependent upon the presence of a specific cis-acting sequence called the packaging signal that is found at the 5' end of the pregenome RNA. Deletion analysis revealed that this packaging signal was located in a 70 bp stretch of nucleotides. Packaging defective virus genome containing the 70 bp packaging signal sequence at several locations, showed that the packaging occurs effectively only when the signal is located near the 5' end of the pregenome RNA. Packaging also occurred with a 2.4 kb viral RNA whose 5' region sequence was different from that of the pregenome RNA, but which has the packaging signal sequence inserted at the 5' region.
Frequencies of somatic mutations in pediatric patients with leukemia were evaluated following intensive treatment at three different loci: the hypoxanthine-guanine phosphoribosyl transferase (HPRT), T-cell antigen receptor (TCR), and glycophorin A (GPA) gene. Thirty-two children with acute lymphoblastic leukemia (ALL), nine children with acute myelogenous leukemia (AML), and 20 age-matched healthy controls were included in the study of mutant frequencies (Mfs) at the HPRT and TCR loci. Among these patients and controls, individuals with heterozygous MN blood type, i.e., 14 children with ALL, three children with AML, and nine healthy controls, served for the further assessment of variant frequency (Vf) at the GPA locus. In ALL patients, geometric mean Mfs and Vfs at these loci were significantly higher than in healthy controls. The high Mf value at the HPRT locus persisted for up to 8 years after the end of chemotherapy. On the other hand, the Mf values at the TCR locus and Vf values at the GPA locus declined gradually with time. In AML patients, on the other hand, the geometric mean Mf only at the TCR locus was significantly higher than in the controls, albeit to a lesser degree than in ALL patients. These data suggest that anti-cancer therapy induces somatic mutations at various loci and that ALL patients are more susceptible to mutagenic intervention than are AML patients.
Pancreatic secretory trypsin inhibitor, an acute phase reactant protein, is expressed in the liver in response to inflammatory cytokines, especially in hepatocellular carcinoma. Northern blots of 25 dissected liver tissues from non-hepatitis, chronic hepatitis and cirrhosis patients revealed that 10 (40%) expressed pancreatic secretory trypsin inhibitor. The expression seemed to be closely associated with hepatitis B viral infection, since among the 11 hepatitis B virus-infected samples, nine (81%) were pancreatic secretory trypsin inhibitor-positive. In contrast, this augmented expression was absent in non-infected livers (0/5; 0%), and rare in those infected with hepatitis C virus (1/9; 11%). There was no significant correlation between the pancreatic secretory trypsin inhibitor expression in the liver and the serum level of glutamic pyruvic transaminase, the hepaplastin test, the 15-min retention rate of indocyanine green, or the histological findings of the liver tissues such as lymphocyte infiltration and pseudolobular formation. Furthermore, we identified an almost three-fold increase in the transactivation of pancreatic secretory trypsin inhibitor gene expression in HepG2 cells after transient transfection with HBV-DNA or the X gene in an expression vector. These results suggest that the induction of pancreatic secretory trypsin inhibitor gene expression in livers with chronic hepatitis and cirrhosis is directly affected by hepatitis B virus.
An expression profile of genes active in the human colonic mucosa was obtained by collecting 959 partial sequences from a 3'-directed cDNA library. Seven genes were found to produce mRNA each of which comprised more than 1% of total mRNA. Four of these genes are novel, and are likely to be uniquely expressed in the colonic mucosa, and the other three have been identified as genes for fatty acid binding protein, immunoglobulin lambda chain, and carcinoma-associated antigen GA733-2. In the remaining 952 clones, 310 were composed of 118 species occurred recurrently but less than 1%, and 533 clones appeared only once. Because the 3'-directed cDNA library faithfully represents the mRNA population in the source tissue, these numbers represent the relative activities of the gene expression. Altogether 156 gene species were identified in GenBank, and a significant portion of these genes encode proteins found in Golgi apparatus and lysosomes, chromosome-encoded mitochondrial proteins, cell surface proteins, and components in the protein synthesis machinery. The types and proportions of genes identified is consistent with the known major activities of the colonic mucosa such as mucous protein production, energy-dependent water absorption, and rapid cell proliferation and turnover.
By collecting 3'-directed cDNA sequences called gene signatures (GSs) on a large scale, it is possible to make an expression profile of genes in a particular tissue, as well as discovering a number of novel genes. A total of 305 novel GSs collected from granulocytoid cells derived from HL60, a human promyelocytic leukemia cell line, by exposure to dimethyl sulfoxide were radiolabeled and used for Southern blot analyses to determine the copy number of the corresponding genes. Of these, 198 GSs identified as representing single-copy genes were then used as probes for hybridization analyses using a monochromosomal hybrid cell DNA panel. Sixty-nine of them were assigned to individual chromosomes. These results demonstrate that the chromosomal distribution of the GSs seems not to be proportional to the cytogenetic length of each chromosome.
Overlapping primers and tailed short primers are effective agents for mapping very short cDNA sequences. By using such primers, human cDNAs as short as 32 nucleotides in length can produce PCR bands. Using these and other primers of ordinary size, 44 cDNAs were assigned to chromosomes, of which 24 were assigned to single chromosomes, and 2 were assigned to two chromosomes and two were assigned to three chromosomes, respectively. Among the 24 cDNAs, all of which matched GenBank entries, 6 cDNAs were observed to map to the same chromosomes as reported previously.
An expression profile of genes active in the human lung was obtained by collecting 797 partial sequences from a 3'-directed cDNA library. Three genes were found to produce mRNA each of which comprised more than 1% of total mRNA. These three have been identified as genes for pulmonary surfactant apoprotein (PSP-A), Clara cells 10-kDa secretory protein, and HLA-E heavy chain. In the remaining 745 clones, 221 were composed of 89 species that occurred recurrently, and 524 clones appeared only once. Because the 3'-directed cDNA library faithfully represents the mRNA population in the source tissue, these numbers represent the relative activities of the gene expression. Altogether 437 gene species were novel, and 179 gene species were identified in GenBank. A significant portion of these genes encode proteins found in secretory proteins, cell surface proteins, and components in the protein synthesis machinery, representing the function of the lung.
We report a case of malignant melanoma of unknown primary origin which presented with a giant metastatic tumor in his right inguinal region. A 94-year-old man noticed a small subcutaneous tumor in the right inguinal region 3 years earlier, which eventually became as larger as 9 cm in diameter without treatment. Although a histological examination of the lesion showed malignant melanoma, extensive examination did not reveal its primary lesion or any metastasis other than that to the right inguinal area. Our case took an interesting course in that this well-growing metastatic tumor was localized in only one region and supported a previous report indicating that malignant melanoma with unknown primary origin has a low tendency to metastasize and a relatively good prognosis.
We recently showed that recombinant human granulocyte-colony stimulating factor (rhG-CSF) maintained the viability of human neutrophils in incubation for up to 72 hours. However, it is not known whether rhG-CSF can enhance neutrophil survival in in vivo situations. To clarify this issue, we investigated neutrophil survival in vitro following in vivo injection of rhG-CSF. Neutrophils were obtained from 4 pediatric patients with malignancies and healthy adult volunteers before and after rhG-CSF administration. Neutrophils obtained before rhG-CSF treatment started to undergo apoptosis after 24 h of incubation. In contrast, the survival of neutrophils drawn after rhG-CSF administration increased by approximately 24 h. Concomitantly, the appearance of typical ladder-like DNA fragmentation was delayed. Such an increase in neutrophil survival was inhibited by co-incubation with either H 7 (10 mumol/l) or H 8 (20 mumol/l), which worked as protein kinase C inhibitors. Although our study did not measure neutrophil survival in vivo directly, it provides us with further evidence that rhG-CSF may function to prolong neutrophil life expectancy in vivo.
This study is designed to investigate whether apoptosis occurs in vivo in pediatric patients with acute leukemia during induction therapy. When patients with common acute lymphoblastic leukemia (cALL) and acute myeloblastic leukemia (AML) were treated with prednisolone (60 mg/m2/day, p.o. or i.v.) and etoposide (150 mg/m2/day, i.v.), respectively, the blast cell counts fell to below 30% and 5%, respectively, in 1 week. However, during this cytoreduction phase, neither morphologically apoptotic cells nor fragmentation of DNA derived from peripheral blast cells were detected at any preparations. On the other hand, cALL but not AML cells spontaneously undergo apoptosis following their culture in vitro. The addition of autologous serum instead of fetal calf serum substantially prevented apoptosis from occurring spontaneously in cALL cells. When cALL and AML cells freshly obtained from patients before therapy were treated in vitro with 10 mumol/l prednisolone and 20 micrograms/ml etoposide, respectively, these cells underwent apoptosis within 6 hours, as determined by a morphological and DNA fragmentation assay. These in vivo and in vitro findings suggest that, although anticancer drugs may induce apoptosis in vivo, these apoptotic cells cannot be detected due to their rapid removal from the circulation.
The novel nucleoside oxetanocin G, 9-(2-deoxy-2-hydroxymethyl-beta-D-erythro-oxetanosyl)guanine (OXT-G), that is a derivative of oxetanocin A, was studied in relation to its action on the synthesis of hepatitis B virus (HBV) DNA and cellular DNA in an HBV-producing cell line, HB611 (T. Tsurimoto, A. Fujiyama, and K. Matsubara, Proc. Natl. Acad. Sci. USA 84:444-448, 1987). The median effective concentration of OXT-G against HBV replication was 1.5 microM, and the median cytotoxic concentration was more than 1,000 microM. At the same concentration, OXT-G did not inhibit cellular DNA synthesis or viral RNA synthesis. Chemically synthesized OXT-GTP inhibited the HBV endogenous DNA polymerase reaction and was incorporated into HBV DNA strands at a low efficiency compared with the incorporation of dGTP. A synthetic primer-template study revealed that OXT-GTP was incorporated into DNA strands at a low efficiency and that further extension of the DNA strand by using the 2' position of the incorporated OXT-G could take place.
Hemodynamic parameters during supine leg exercise-induced myocardial ischemia were investigated in ten patients without left ventricular hypertrophy (group I), and in ten patients with left ventricular hypertrophy (group II). Both groups of patients had previously been diagnosed as having totally asymptomatic silent myocardial ischemia. Both groups were comparable in age, sex, left ventricular ejection fraction at rest, and extent of coronary artery disease. All of the patients exercised to the same extent without angina or its usual equivalent. Left ventricular filling pressure increased from 4 +/- 2 mmHg at rest to 11 +/- 3 mmHg at peak exercise in group I, and from 7 +/- 4 mmHg at rest to 19 +/- 7 mmHg at peak exercise in group II. Left ventricular filling pressure at peak exercise was significantly greater in group II than in group I (p < 0.01). Totally asymptomatic silent myocardial ischemia can be seen with either normal or mildly abnormal left ventricular filling dynamics during exercise. In addition patients with coexistent left ventricular hypertrophy can have severely abnormal left ventricular filling dynamics during exercise without ischemic cardiac pains. Therefore, severe ischemia and/or stretching of the ventricular wall may be necessary, but not sufficient, for the occurrence of chest pain.