1H NMR studies of the mercuric ion binding protein MerP: Sequential assignment,secondary structure and global fold of oxidized MerP

Summary The oxidized form of the mercuric ion binding protein MerP has been studied by two-dimensional NMR. MerP, which is a periplasmic water-soluble protein with 72 amino acids, is involved in the detoxification of mercuric ions in bacteria with resistance against mercury. The mercuric ions in the periplasmic space are first scavenged by the MerP protein, then transported into the cytoplasm by the membrane-bound transport protein MerT, and finally reduced to elementary (nontoxic) mercury by the enzyme mercuric reductase. In this work, the ¹H NMR spectrum of oxidized MerP (closed disulfide bridge) has been assigned by using homonuclear 2D NMR techniques. The secondary structure and global fold have been inferred from the nuclear Overhauser effect (NOE) data. The secondary structure comprises four -strands and two -helices, in the order ₁₁₂₃₂₄. The protein folds into an antiparallel -sheet, ₂₃₁₄, with the two antiparallel helices on one side of the sheet. The folding topology is similar to that of acylphosphatase, the activation domain of porcine pancreatic procarboxypeptidase B, the DNA-binding domain of bovine papillomavirus-1 E2 and the RNA-binding domains of the U1 snRNP A and hnRNP C proteins. However, there is no structural similarity between MerP and other bacterial periplasmic binding proteins.     

Amylase activity and growth in internodes of deepwater rice

Isoelectrofocusing, product analysis, thermal denaturation studies and affinity chromatography on cycloheptaamylose-Sephadex were used to identify the amylolytic enzymes in internodes of deepwater rice (Oryza sativa L.). Amylolytic activity in internodes of deepwater rice consists of -amylase (sometimes separated into two isoforms) and of -amylase. During submergence of whole plants, -amylase activity increases in young, growing internodes, but -amylase activity declines. Although non-growing, mature internodes contain higher levels of -amylase than do the elongating younger internodes, the effect of submergence on amylase activities in both tissues follows the same trend. Submergence, gibberellic acid (GA₃) and ethylene all promote -amylase activity in growing and non-growing internodes of excised deepwater-rice stem sections. Inhibitor studies showed that submergence and ethylene promote -amylase activity in the absence of endogenous gibberellin (GA), and GA₃ enhances -amylase activity when ethylene action is inhibited. Therefore, ethylene and GA appear to increase -amylase activity independently of each other. Enhanced -amylase activities are probably responsible for the mobilization of carbohydrates which are needed to support internode elongation during submergence of deepwater rice.Abbreviations CHA cycloheptaamylose - GA₃ gibberellic acid - NBD 2,5-norbornadiene - TCY tetcyclacis     

Improving Feedback Merging for Source-Adaptive Layered Multicast Schemes

Current source-adaptive layered multicast schemes exploit merging capabilities at special nodes in the network to combine feedback from the multicast group. The merging procedures reduce network load and avoid feedback implosion at the source. In this paper, we demonstrate how to increase the efficiency of a merger node. We show how to tune two mechanisms: feedback transmission, on the timer settings used by receivers, and temporal merging, on the timer settings used by intermediate nodes. Our approach is generic, relating purely to timer settings, and does not touch on content merging, or the decision regarding which packets to forward. We investigate the effects of temporal merging on packet suppression. We show that periodic feedback transmission coupled with a suitable timer setting at the merger node increases efficiency.     

Do neurons have a voltage or a current threshold for action potential initiation?

The majority of neural network models consider the output of single neurons to be a continuous, positive, and saturating firing ratef(t), while a minority treat neuronal output as a series of delta pulses (t — t _i ). We here argue that the issue of the proper output representation relates to the biophysics of the cells in question and, in particular, to whether initiation of somatic action potentials occurs when a certain thresholdvoltage or a thresholdcurrent is exceeded. We approach this issue using numerical simulations of the electrical behavior of a layer 5 pyramidal cell from cat visual cortex. The dendritic tree is passive while the cell body includes eight voltage- and calcium-dependent membrane conductances.We compute both the steady-state (I ^static (V _m )) and the instantaneous (I _o (V_m)) I–V relationships and argue that the amplitude of the local maximum inI ^static (V _m ) corresponds to the current thresholdI _th for sustained inputs, while the location of the middle zero-crossing ofI _o corresponds to a fixed voltage thresholdV _th for rapid inputs. We confirm this using numerical simulations: for rapid synaptic inputs, spikes are initiated if the somatic potential exceedsV _th, while for slowly varying inputI _th must be exceeded. Due to the presence of the large dendritic tree, no charge thresholdQ _th exists for physiological input.Introducing the temporal average of the somatic membrane potential (V _m) while the cell is spiking repetitively, allows us to define a dynamic I-V relationship ^dynamic ((V _m)). We find an exponential relationship between (V _m) and the net current sunk by the somatic membrane during spiking (diode-like behavior). The slope ofI/dynamic((V _m)) allows us to define a dynamic input conductance and a time constant that characterizes how rapidly the cell changes its output firing frequency in response to a change in its input.     

Selective staining of Y chromosomal loops in Drosophila hydei,D. neohydei,and D. eohydei

Modified Giemsa procedures have been developed which elicit differential and highly selective staining of individual Y chromosomal lamp-brush loops in spermatocyte nuclei of Drosophila hydei, D. neohydei, and D. eohydei. In all three species the Y loop pair known as the clubs stains a brilliant dark red with Giemsa at pH 10. With the same treatment other loop pairs either remain unstained, e.g. the threads, or show a differentiation between light blue and pink staining matrical material, e.g. pseudonucleolus and cones in D. hydei and D. eohydei. With eosin at pH 2.8 the threads in D. hydei can be stained intensely, as well as one matrical component of the pseudonucleolus. Pretreatment with RNase or TCA removes all stainability from the Y loops with Giemsa at pH 10. TCA treatment enhances eosin staining at pH 2.8. These and other variations of Giemsa may be utilized to establish homologies between Y loops in different species. The molecular basis of the staining reactions remains to be elucidated.     

Novel Type of Signaling Molecules: Protein Kinases Covalently Linked with Ion Channels

Recently we identified a new class of protein kinases with a novel type of catalytic domain structurally and evolutionarily unrelated to the conventional eukaryotic protein kinases. This new class, which we named alpha-kinases, is represented by eukaryotic elongation factor-2 kinase and the Dictyosteliummyosin heavy chain kinases. Here we cloned, sequenced, and analyzed the tissue distribution of five new putative mammalian -kinases: melanoma -kinase, kidney -kinase, heart -kinase, skeletal muscle -kinase, and lymphocyte -kinase. All five are large proteins of more than 1000 amino acids with an -kinase catalytic domain located in the carboxyterminal part. We expressed the catalytic domain of melanoma -kinase in Escherichia coli, and found that it autophosphorylates at threonine residues, demonstrating that it is a genuine protein kinase. Unexpectedly, we found that long aminoterminal portions of melanoma and kidney -kinases represent new members of the TRP ion channel family, which are thought to mediate the capacitative Ca²⁺entry in nonexcitable mammalian cells. This suggests that melanoma and kidney -kinases, which represent a novel type of signaling molecule, are involved in the regulation of Ca²⁺influx in mammalian cells.     

Trace metal contamination initiates the apparent auto-aggregation,amyloidosis, and oligomerization of Alzheimer’s Aβ peptides

Nucleation-dependent protein aggregation (seeding) and amyloid fibril-free formation of soluble SDS-resistant oligomers (oligomerization) by hydrophobic interaction is an in vitro model thought to propagate -amyloid (A) deposition, accumulation, and incur neurotoxicity and synaptotoxicity in Alzheimers disease (AD), and other amyloid-associated neurodegenerative diseases. However, A is a high-affinity metalloprotein that aggregates in the presence of biometals (zinc, copper, and iron), and neocortical A deposition is abolished by genetic ablation of synaptic zinc in transgenic mice. We now present in vitro evidence that trace (0.8 µM) levels of zinc, copper, and iron, present as common contaminants of laboratory buffers and culture media, are the actual initiators of the classic A1–42-mediated seeding process and A oligomerization. Replicating the experimental conditions of earlier workers, we found that the in vitro precipitation and amyloidosis of A1–40 (20 µM) initiated by A1–42 (2 µM) were abolished by chelation of trace metal contaminants. Further, metal chelation attenuated formation of soluble A oligomers from a cell-free culture medium. These data suggest that protein self-assembly and oligomerization are not spontaneous in this system as previously thought, and that there may be an obligatory role for metal ions in initiating A amyloidosis and oligomerization.     

Excretion of ureide and other nitrogenous compounds by the root system of soybean at different growth stages

Direct excretion of nitrogenous compounds into a N-free nutrient solution, which was allowed to drip onto the root system of soybean (Glycine max (L) Merr. cv. Kurosengoku) was examined at different growth stages; vegetative, flowering and pod-filling. Considerable amounts of nitrogenous compounds were excreted at all the growth stages, with the highest amount recorded at the pod-filling stage.The root was found to be the major site of N compound excretion and its dry weight was linearly correlated with N amount excreted. Maximum nitrogen excretion rate during vegetative and flowering stages was recorded during the day, however at the pod-filling stage, the highest was recorded during the night. Ureide was excreted at all growth stages, but the highest amount was recorded at the pod-filing stage. The root was found to be the active site of ureide excretion. Composition of the total nitrogen examined i. e. ureide, soluble proteins, ammonia and amino acids, was found to be changing during the growth stages, suggesting possible different major pathways of excretion at different plant age. Among the N compounds monitored, were soluble proteins, ammonia and amino acids. Only a few of the several amino acids found in the root tissues were observed in the excreted solution, notably phosphoserine and phosphoethanolamine, at all the growth stages, whilst -amino-butyric-acid and serine were observed in trace amounts during vegetative and flowering stages. Quantitatively the ammonia found in the excreted solution was far greater than in the tissues.     

Cross-reactivity between human and fish pituitary hormones as demonstrated by immunocytochemistry

Summary In this communication we describe the immunocytochemical cross-reactivity between antisera to various human pituitary hormones and specific hormone producing cell types in the pituitary gland of sexually mature male platyfish (Xiphophorus maculatus). Antisera to human pituitary hormones cross-reacted either with cells known to produce corresponding hormones (or hormone subunits) in the platyfish (e.g., ACTH, prolactin, TSH , LH , FSH , TSH ) or with no pituitary cells at all (e.g., LH , FSH ). The one exception was antiserum to human growth hormone which cross-reacted with MSH and ACTH producing cells. The platyfish pituitary is proposed as a test system for immunocytochemically screening antisera for purity and specificity in order to determine their applicability in particular studies.     

Synthesis of FimH receptor-active manno-oligosaccharides by reverse hydrolysis using α-mannosidases from Penicillium citrinum,Aspergillus phoenicis and almond

Recombinant Penicillium citrinum -1,2-mannosidase, expressed in Aspergillus oryzae, was employed to carry out regioselective synthesis of -d-mannopyranosyl-(12)-d-mannose. Yields (w/w) of 16.68% disaccharide, 3.07% trisaccharide and 0.48% tetrasaccharide were obtained, with 12 linkages present at 98.5% of the total linkages formed. Non-specific -mannosidase from almond was highly efficient in reverse hydrolysis and oligosaccharide yields of 45–50% were achieved. The products of the almond mannosidase were a mixture of disaccharides (30.75%, w/w), trisaccharides (12.26%, w/w) and tetrasaccharides (1.89%, w/w) with 12, 13 and 16 isomers. -1,2-linkage specific mannosidase from P. citrinum and -1,6-linkage-specific mannosidase from Aspergillus phoenicis were used in combination to hydrolyse the respective linkages from the mixture of isomers, resulting in -d-mannopyranosyl-(13)-d-mannose in 86.4% purity. The synthesised oligosaccharides can potentially inhibit the adhesion of pathogens by acting as "decoys" of receptors of type-1 fimbriae carried by enterobacteria.     

Über die Teilungsfolge in den Fäden von zwei Grünalgen

Zusammenfassung In den Fäden der GrünalgenMicrospora amoena undSpirogyra crassa wechseln geschlossene Abschnitte von eigentlich ruhenden Zellen mit Abschnitten von präprophasischen, das sind Mitose-bereiten, mitotisch aktiven und posttelophasischen, also unmittelbar nach der Mitose stehenden Zellen ab. Nur ausnahmsweise schieben sich Zellen des einen Typus in Abschnitte von Zellen des anderen Typus ein.Dieses Verhalten wird hypothetisch mit dem Auftreten und der Weiterleitung teilungsinduzierender Hormone erkärt. Daß die Teilungsstadien einschließlich der vor- und nachmitotischen Stadien nicht entsprechend einem Gefälle streng geordnet aufeinanderfolgen, beruht wahrscheinlich auf der Verschiedenartigkeit der äußeren und inneren Faktoren, die auf die einzelnen Zellen einwirken. Genealogische Faktoren spielen offenbar eine untergeordnete Rolle.     

DNA fragmentation and detachment of enterocytes induced by anti-CD3 mAb-activated intraepithelial lymphocytes

To elucidate the role of intraepithelial lymphocytes (IEL) and enterocytes in the defense mechanism of the small intestine, we designed experiments to stimulate the IEL by anti-CD3, anti-TCR, or anti-TCR monoclonal antibodies (mAbs), and to examine the subsequent changes to the enterocytes. The enterocytes of the duodenum and jejunum, but not of the ileum, showed massive DNA fragmentation 30 min after intraperitoneal injection of anti-CD3 mAb. These responses were also induced by anti-TCR mAb, but not by anti-TCR mAb, and were completely inhibited by cyclosporin A. Nearly half of the enterocytes of the villi in the duodenum and jejunum were exfoliated into the lumen 4 h after the injection of the mAb. Administration of anti-CD3 mAb also induced DNA fragmentation in Fas-deficient MRL/lpr mice, indicating that the Fas-Fas ligand system was not involved in these events. The anti-CD3 mAb treatment also induced massive DNA fragmentation in the intestinal epithelium of the duodenum and jejunum in TNF-receptor-1-deficient mice, whereas TNF- induced only the detachment of intestinal epithelium of wild-type mice, implying the dissociation of two independent factors and/or mechanisms for DNA fragmentation and the subsequent epithelial cell detachment in the murine duodenum and jejunum. The mAb failed to exfoliate the epithelium in TNF-R1-deficient mice. Thus, TCR⁺ IEL, when treated with anti-CD3 or anti-TCR mAbs, induced rapid DNA fragmentation and subsequent detachment of the duodenal and jejunal epithelia, but not in the ileum (the silent ileum), partly because of the paucity of TCR⁺ IELs in the ileum.K. Yaguchi and S. Kayaba contributed equally to this workThis work was in part supported by a Grant-in-aid for Scientific Research from the Ministry of Education, Science and Culture, Japan (07407066, 10470002, and 13670002 to T.I., and 10770001 to H.S.), and by The Funds for Comprehensive Research on Long Term Chronic Diseases from the Ministry of Health and Welfare of Japan (to T.I.)     

Protein kinase C isoforms in pituitary cells displaying differential sensitivity to phorbol ester

Investigations with protein kinase C (PKC) isoform-specific antisera, revealed distinct profiles of PKC isoform content amongst pituitary tissues. Western analysis revealed the and isoforms of PKC are present in rat anterior and posterior pituitary tissue as well as in the GH₃ somatomammotrophic cell line. AtT-20/D16-V corticotrophic and T3-1 gonadotrophic murine cell lines contained no PKC-. The or isoforms were undetected in any pituitary tissue. PKC activity measurements revealed Ca²⁺-independent PKCs in T3-1 and GH₃ cells which were more sensitive to activation by phorbol-dibutyrate (PDBu) than the corresponding PKC activity found in COS cells. However, Ca²⁺-dependent PKC activities were of similar sensitivity to PDBu in GH₃, T3-1 and COS cells, indicating that functional differences observed in PDBu-sensitivity in these cells may be due to differential activation of Ca²⁺-independent PKC isoforms. Moreover, substrate-specificity of these PKCs were also compared indicating that the amount of Ca²⁺-dependency of the observed PKC activity from the same pituitary tissue is dependent upon the substrate utilized by the PKC isotypes present. These findings explain differential sensitivities of PKC-mediated actions that have previously been observed in a range of pituitary cells. (Mol Cell Biochem 000-000, 1999)     

A morphometric analysis of the subcellular distribution of LHβ and FSHβ in secretory granules in the pituitary gonadotrophs of the frog (Rana japonica)

Immunohistochemical localization of lutropin (LH) and follitropin (FSH) in the pituitary gland of the frog Rana japonica was studied by the peroxidase-anti-peroxidase method and the two-face, double-labeling method with different-sized gold particles at the light-and electron-microscopic levels, respectively, using monoclonal antibodies against bullfrog LH and FSH. Light-microscopic immunohistochemistry indicated that approximately 66.0% of all the gonadotrophs in the pituitary contained both LH and FSH, whereas 33.4% of gonadotrophs contained only LH, and 0.6% contained only FSH. The staining intensity of LH and FSH varied from cell to cell. The gonadotrophs were classified into four types (Types I–IV) in terms of their ultrastructural and immunolabeling characteristics. Moreover, several secretory granule types were recognized according to differences in their shape and electron density. In all the cell types, both LH and FSH were often seen in the same secretory granules, but the proportion of granules bearing both hormones ranged from 5.5% in Type I to 32.7% in Type IV. Most secretory granules in Types I and II were immunolabeled with LH alone, whereas a small number of granules were immunolabeled with FSH alone. More immunolabeled FSH granules were present in Types III and IV than in Types I and II.     

Histochemistry and ultrastructure of adrenergic and acetylcholinesterase-containing nerves supplying follicles and endocrine cells in the guinea-pig ovary

Summary With the use of an anti-human S-100 protein antibody, it was possible to reveal a characteristic cell type in the anterior lobe of the normal human pituitary. These cells, so-called folliculo-stellate cells, were present in all pituitaries studied but their number varied from one gland to another. Immunoreactive cells, isolated or grouped, were arranged close to various secretory granulated cells. Especially by use of double immunoenzymatic labeling, it was evident that these cells are spatially related either to somatotropes, prolactin cells and corticotropes, or to glycoprotein-containing cells. Such immunoreactive cells were rare or absent in pseudo-follicular arrangements of secretory granulated cells. Since it is now possible to identify this cell type by light microscopy and since no reliable functional significance is known, it seems more advisable to term this cell type stellate cell instead of folliculostellate cell.     

Differential Expression and Association of Calcium Channel Subunits in Development and Disease

Voltage-gated calcium channels (VDCC) are essential to neuronal maturation and differentiation. It is believed that important signaling information is encoded by VDCC-mediated calcium influx that has both spatial and temporal components. VDCC are multimeric complexes comprised of a pore-forming ₁ subunit and auxiliary and ₂/ subunits. Changes in the fractional contribution of distinct calcium conductances to the total calcium current have been noted in developing and differentiating neurons. These changes are anticipated to reflect the differential expression and localization of the pore-forming ₁ subunits. However, as in vitro studies have established that regulates the channel properties and targeting of ₁, attention has been directed toward the developmental expression and assembly of isoforms. Recently, changes in the component of the omega-conotoxin GVIA (CTX)-sensitive N-type VDCC have indicated differential assembly of _1B with in postnatal rat brain. In addition, unique properties of 4 have been noted with respect to its temporal pattern of expression and incorporation into N-type VDCC complexes. Therefore, the expression and assembly of specific ₁/ complexes may reflect an elaborate cellular strategy for regulating VDCC diversity. The importance of these developmental findings is bolstered by a recent study which identified mutations in the 4 as the molecular defect in the mutant epileptic mouse (lethargic; lh/lh). As 4 is normally expressed in both forebrain and cerebellum, one may consider the impact of the loss of 4 upon VDCC assembly and activity. The importance of the lb and 4 isoforms to calcium channel maturation and assembly is discussed.     

Degradation stages of the oak forests in the area of Algiers

Long lasting human impact on the natural Mediterranean vegetation is the reason for its degradation. Uncontrolled felling of trees, fire and overgrazing are the most important ecological causes of this process.Geobotanical investigations made in the area of Algiers facilitated the characterization of degradation changes in the structure of vegetation and in the floristic composition in all the degradation stages.Under the influence of the degradation factors, mentioned above, the oak forests, growing on different, more or less calcareous substrates, are transformed into various shrub formations, mainly into many types of maquis and garrigue. The floristic composition of these communities is given in Table 2.Further degradation leads to widespread Cistus-formations and then to palmitto, a formation created by the dwarf palm (Chamaerops humilis).When the anthropopressure becomes stronger even this dwarf palm retreats. Loose swards, replacing the palmitto formation, often cannot stop the subsequent degradation. Bare rock is the extreme, relatively rare stage of forest degradation in this area.The most common form of natural regeneration is the invasion of Pinus halepensis, observed in all degradation stages identified. This pine is also one of trees, most commonly planted on mountain slopes to prevent their erosion.     

Lectin cytochemistry of cell types in human and canine pituitary

Summary Labeled lectins specific for different sugars were employed to identify different cell types in pituitaries from six human autopsies and seven dogs. To determine the lectins bound by each cell type, fixed-paraffin embedded sections serial to those stained with lectins were immunostained for specific hormones and the serial pairs were examined in a comparison microscope. In human pituitaries corticotrophs stained selectively with lectins having affinity for -l-fucose and the core region of complex type N-glycosyl-proteins. Some corticotrophs also stained for the presence of terminal -galactose. Thyrotrophs stained selectively with a periodate oxidation-borohydride reduction-concanavalin A sequence. Some mammotrophs evidenced content of glycoconjugate with terminal -galactose. Dendritic cells stained selectively for abundant glycogen with the periodate-reduction-concanavalin A sequence and a lectin from Griffonia simplicifolia. Adenohypophyseal cells of dog pituitary differed in showing absence of terminal -galactose in corticotrophs, presence of terminal -galactose in thyrotrophs, presence of glycoconjugate with N-glycosidically bound oligosaccharide in thyrotrophs and gonadotrophs and presence of terminal -galactose with a different lectin affinity in mammotrophs. The main contributions of lectin histochemistry applied to the pituitary include providing an additional histologic method for identification of some cell types, and localizing glycosylated prohormone or other biochemically unrecognized non-hormone glycoconjugates whose function in pituitary cells remains to be explained.This research was supported by NIH Grants AM-10956 and HL-29775 and United Health and Medical Research Foundation of South Carolina, Inc. Grant #79     

Binding of the galactose-specificPseudomonas aeruginosa lectin,PA-I,to glycosphingolipids and other glycoconjugates

The carbohydrate-binding specificity ofPseudomonas aeruginosa lectin I (PA-I) in iodinated or biotinylated form was studied. A large number of glycosphingolipids, as well as some glycoproteins and neoglycoproteins were used as ligands. Also, inhibition by free saccharides of PA-I binding to glycosphingolipids was tested. It was found that the lectin binds most strongly to terminal and nonsubstituted Gal3Gal- or Gal4Gal-structures.Abbreviations PA-I Pseudomonas aeruginosa lectin I - Cer ceramide - lactosylceramide Gal4GlcCer - iso globotriaosylcerami Gal3Gal4GlcCer - globotriaosylceramide Gal4Gal4GlcCer - globoside or globotetraosylceramide GalNAc3Gal4Gal4GlcCer - Forssman glycolipid GalNAc3GalNAc3Gal4Gal4GlcCer - P1 glycolipid Gal4Gal4GlcNAc3Gal4GlcCer - lactoneotetraosylceramide Gal4GlcNAc3Gal4GlcCer - B5 glycolipid Gal3Gal4GlcNAc3Gal4GlcCer - gangliotetraosylceramide Gal3GalNAc4Gal4GlcCer - GM1 Gal3GalNAc4(NeuAc3)Gal4GlcCer - RBC red blood cells - BSA bovine serum albumin - PBS phosphate-buffered saline - SDS sodium dodecyl sulfate - TLC thin-layer chromatography - HPLC high pressure liquid chromatography - MS mass spectrometry - FAB fast-atom bombardment - EI electron impact     

Over expression of integrin α 5 β 1 in human hepatocellular carcinoma cell line suppresses cell proliferation in vitro and tumorigenicity in nude mice

Integrin 5 1 and 2 1 are the major integrin receptors in human hepatocytes. However, in human hepatocellular carcinoma cells it was found that the expression of integrin 5 1 was decreased and another integrin 6 1 increased. In this study, the SMMC7721 human hepatocellular carcinoma cells cotransfected or singlely transfected with integrin 5 and/or 1 cDNAs were established, and designated 5 1.6-7721, 5.3-7721, and 1.6-7721 cell lines, respectively. Transfection with cDNAs of integrin 5 and 1 subunits resulted in the overexpression of each integrin and modified biological properties, including a slowed growth rate, changes in the cell cycle from 15.5% of control cells in the G2/M phase to 12.1%, 9.6% and 9.4% in 5.3-7721, 1.6-7721, 5 1.6-7721, respectively, and a decrease in the Cell Mitosis Index from 1.6 in controls to 0.96, 0.95, and 0.72, and 34%, 28% and 52% derived from colony forming ability, respectively. Tumorigenicity was also tested in nude mice with inoculation of cells subcutaneously. Tumor masses growing in nude mice following inoculation with 1.6-7721,and 5 1.6-7721 cells weighed only 52% or 31% those of control cells. These results indicated that deletion or low expression of integrin 5 1 may play an important role in the development of hepatocellular carcinoma. Therefore, induction of expression of the integrin 5 1 in malignant cells could be a potential means of treating hepatocellular carcinoma.