Search

The objective of this study was to develop an in vitro cartilage degradation model that emulates the damage seen in early-stage osteoarthritis. To this end, cartilage explants were collagenase-treated to induce enzymatic degradation of collagen fibers and proteoglycans at the articular surface. To assess changes in mechanical properties, intact and degraded cartilage explants were subjected to a series of confined compression creep tests. Changes in extracellular matrix structure and composition were determined using biochemical and histological approaches. Our results show that collagenase-induced degradation increased the amount of deformation experienced by the cartilage explants under compression. An increase in apparent permeability as well as a decrease in instantaneous and aggregate moduli was measured following collagenase treatment. Histological analysis of degraded explants revealed the presence of surface fibrillation, proteoglycan depletion in the superficial and intermediate zones and loss of the lamina splendens. Collagen cleavage was confirmed by the Col II-3/4Cshort antibody. Degraded specimens experienced a significant decrease in proteoglycan content but maintained total collagen content. Repetitive testing of degraded samples resulted in the gradual collapse of the articular surface and the compaction of the superficial zone. Taken together, our data demonstrates that enzymatic degradation with collagenase can be used to emulate changes seen in early-stage osteoarthritis. Further, our in vitro model provides information on cartilage mechanics and insights on how matrix changes can affect cartilage's functional properties. More importantly, our model can be applied to develop and test treatment options for tissue repair., (© 2013 Elsevier Ltd. All rights reserved.)

The role of cell surface integrins in cell migration, proliferation, and attachment to matrix molecules is well known. Integrin-matrix interactions have been implicated in mechanotransduction and load transmission from the outside to the inside of the cell. In this study, the effect of cyclic strain on the cell proliferation, attachment, and expression of integrin subunits beta1, beta3, and alpha5 was determined in anterior cruciate ligament (ACL) and medial collateral ligament (MCL) fibroblasts grown on polystyrene, Type I collagen, laminin, elastin, and fibronectin. ACL fibroblast proliferation was not affected by growth substrate whereas MCL cells reached confluence more rapidly on fibronectin compared with collagen or polystyrene. Exposure to 5% cyclic strain resulted in a significant decrease in ACL and MCL fibroblast proliferation on fibronectin and Type I collagen. MCL cells showed a greater strain-dependent inhibition of cells grown on a fibronectin substrate than those grown on collagen. This matrix-dependent effect of strain on cell proliferation was not seen with ACL cells. Attachment of ACL and MCL fibroblasts was stronger to fibronectin compared with Type I collagen, laminin, and polystyrene. In the absence of applied load, the expression of beta1, beta3, and alpha5 subunits was not substrate dependent and the expression of beta1 and alpha5 integrin subunits was higher in MCL cells than ACL cells on all substrates. In contrast, the expression of beta3 integrin subunit was higher in ACL cells than MCL cells. In response to 5% strain, beta1, and alpha5 expression increased in all fibroblasts with MCL cells having a higher magnitude of expression. beta3 expression showed a 90% increase in response to load when grown on laminin for both MCL and ACL fibroblasts and demonstrated no change in expression on Type I collagen or fibronectin. The duration of applied strain from 2 versus 22 h had no effect on cell proliferation or integrin expression., ((c) 2005 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res.)

The present study was undertaken to define the nature of key transport processes for sodium, glucose, proline, and sulfate in primary culture of canine anterior cruciate ligament (ACL) and medial collateral ligament (MCL) cells. Uptake studies using radiolabeled isotopes were performed and Na,K-ATPase activity was determined in cell lysates. At 25 degrees C both ACL and MCL cells showed a significant uptake of 86Rb. Ouabain inhibited Rb uptake by 55% in ACL cells and by 60% in MCL cells. The transport activity of Na,K-ATPase in intact cells was calculated to be 57 and 71 nmol.(mg protein)-1.(15 min)-1, respectively. The enzymatic activity of Na,K-ATPase in cell lysates was observed to be 104 for ACL cells and 121 nmol.(mg protein)-1.(15 min)-1 for MCL cells. Cytochalasin B, a known inhibitor of sodium-independent D-glucose transport, completely inhibited D-glucose uptake in ACL and MCL cells. Removal of Na+ or addition of 10-5 mol/L phlorizin, a potent inhibitor of the sodium-D-glucose cotransporter, did not alter D-glucose uptake, suggesting that glucose entered the cells using a sodium-independent pathway. Both ACL and MCL cells exhibited high sulfate uptake that was not altered by replacement of Na+ by N-methyl-D-glucamine, whereas DIDS, an inhibitor of sulfate/anion exchange abolished sulfate uptake in both cell types. Thus, neither cell type seems to possess a sodium-sulfate cotransport system. Rather, sulfate uptake appeared to be mediated by sulfate/anion exchange. Proline was rapidly taken up by ACL and MCL cells and its uptake was reduced by 85% when Na+ was replaced by N-methyl-D-glucamine, indicating that proline entered the cells via sodium-dependent cotransport systems. The data demonstrate that both ACL and MCL cells possess a highly active sodium pump, a secondary active sodium-proline cotransport system, and sodium-independent transport systems for D-glucose and sulfate.

Injuries to the avascular region of the meniscus occur frequently and may be difficult to repair. This study was designed to determine whether growth factors could diffuse from a collagen sponge or a collagen gel into meniscal tissue and stimulate healing of defects using an in vitro model. The diffusion of platelet-derived growth factor (PDGF) from the collagen carriers into the medium was rapid with approximately 50% being released from the collagen sponge within the first hour. After 5 d of incubation, 8% of the PDGF was present in the meniscus, 11% in the collagen sponge, and 62% had been released into the medium. Similar results were obtained when a collagen gel was used as a carrier. Histological evaluation of the meniscal explants after 2 wk in culture revealed extensive proteoglycan staining in the areas surrounding defects treated with either hepatocyte growth factor (HGF) or PDGF compared with controls without growth factor. The HGF-PDGF treatment resulted in alignment and migration of meniscal cells toward the defect, which was not observed in untreated controls. At 3-7 d, increased number of cells were observed in defects treated with collagen gels (but not the sponge) with PDGF-HGF. At 4 wk, combined HGF-PDGF treatment resulted in the formation of tissue with birefringence by polarized microscopy, suggestive of organized collagen. The data suggest that use of specific PDGF-HGF may enhance the repair of meniscal injuries.

When liver cytosol prepared from rats administered [(14)C]-3'-Methyl-N,N-dimethyl-4-aminoazobenzene was subjected to Sephadex gel chromatography, four peaks of radioactivity containing proteins (Peak-I-IV) and one peak devoid of protein (Peak-V) were obtained. Translocation of azo dye metabolites from these various cytosolic fractions into nucleus was studied in an in vitro system and a maximum of about 10% of the radioactivity associated with a particular cytosolic fraction (Peak-II) could translocate into the nuclei. Radioactivity (%) translocated did not increase upon addition of excess nuclei. Passage of this protein fraction through an immobilized protease column reduced the azo dye metabolite translocation by 65%, concomitant with the degradation of proteins. Translocation was not observed with protein-free metabolites extracted from this cytosolic fraction; addition of proteins corresponding to peak-II from normal rat liver cytosol significantly restored the metabolite translocation. This observation suggests that specific cytosolic proteins are involved in the translocation of azo dye carcinogen metabolites from liver cytoplasm into the nucleus. When the liver cytosolic proteins corresponding to this fraction (Peak-II) were iodinated with (125)I-iodine and incubated with purified nuclei, translocation of three specific proteins into nucleus was observed as seen by SDS-PAGE and fluorography of nuclear proteins. Covalent binding of azo dye metabolites to DNA was not observed when cytosolic peak-II fraction containing azo dye metabolites was incubated with isolated liver DNA instead of liver nuclei. This suggests that the interaction of azo dye metabolites with nuclear macromolecules necessitate further prior processing which actually may occur in the nucleus.

Articular cartilage is subjected to cyclic compressive stresses during joint loading. There is increasing experimental evidence that this loading is essential for the chondrocytes to maintain the functionality of the cartilage extracellular matrix (ECM) and that members of the integrin family of transmembrane receptors may play an important role in signal mechanotransduction between the ECM and chondrocytes. Of particular interest are the integrin subunits alpha5 and beta1, which are known to form the receptor for fibronectin, an important ECM protein, and to be involved in mechanotransduction as well as in the regulation of cytokine production. In this study, we measured the amounts of the integrin subunits alpha5 and beta1 in chondrocytes from young (immature) and adult (mature) bovine articular cartilage explants which were subjected to a continuously applied cyclic compressive stress of 1 MPa for 6 and 24 h. The integrin content per chondrocyte was measured immediately after load cessation by flow cytometry following matrix digestion to release the cells. We found that a mechanical stress induced an increase in the number of integrin subunit alpha5 in immature and mature cartilage but not in the integrin subunit beta1 content. The integrin contents were greatest after 6 h of loading and returned to control levels after 24 h of unloading. The results of this study supply further experimental evidence that chondrocytes respond to changes in their mechanical environment and that the integrin alpha5beta1 may act as a mechanical signal transducer between the chondrocyte and the ECM for the modulation of cellular physiology.

This study examined the effect of several cytokines on the chemotactic migration of fibroblasts derived from 3 different parts of the canine shoulder: the upper part of the medial glenohumeral ligament (equivalent to the anterior part of the inferior glenohumeral ligament of the human shoulder); the inferior part of the medial glenohumeral ligament (equivalent to the axillary pouch of the human shoulder); and the posterior capsule (equivalent to the thin posterior capsule in the human shoulder). Platelet-derived growth factor-AB stimulated the migration of all 3 cell types in a dose-dependent manner, with increases from 150% to 300% at 1 ng/mL to 500% to 700% at 10 ng/mL. Hepatocyte growth factor also stimulated the migration of all 3 cell types in a dose-dependent manner (130% to 310%). Insulinlike growth factor-1 increased the migration of all 3 types of fibroblasts by 160% to 250%. Bone morphogenic protein-2, interleukin-1, and transforming growth factor-b had no significant effect on migration of shoulder capsular fibroblasts. These data demonstrate that capsular fibroblasts are responsive to specific growth factors and suggest the potential for use of growth factors to augment healing and/or remodeling of the shoulder capsule.

Cells in normal tendon are in a resting G0 state, performing maintenance functions. However, traumatic injury introduces growth factors such as platelet-derived growth factor and insulin-like growth factor from blood as well as activates endogenous growth factors. These factors stimulate migration and proliferation of tendon cells at the wound area. Tendon cells require growth-promoting factors to transit the cell cycle. To evaluate the contribution of endogenous growth factors in tendon, extracts of the epitenon and internal compartment of avian flexor tendon as well as medium of cultured cells from the epitenon (tendon surface cells) and internal tendon (tendon internal fibroblasts) were collected to assess their ability to stimulate DNA synthesis. Acid-ethanol extracts of tissues and medium were chromatographed on a P-30 molecular sieve column and assayed for mitogenic activity by quantitating [3H]thymidine incorporation into tendon cell DNA. The extract from the internal tendon compartment was more stimulatory for DNA synthesis than that from the epitenon, particularly when tested on tendon internal fibroblasts. However, conditioned medium fractions from surface epitenon cells stimulated DNA synthesis to a high degree on both tendon surface cells and tendon internal fibroblasts. Conditioned medium from tendon internal fibroblasts was also stimulatory. An anti-insulin-like growth factor-I antibody ablated most of the mitogenic activity present in both tissues and conditioned medium. The levels of acid-extractable insulin-like growth factor-I in tendon were determined by competitive radioimmunoassay as 1.48+/-0.05 ng/g tissue for the epitenon and 3.83+/-0.03 ng/g tissue for the internal compartment. Results of Western immunoblots of conditioned medium revealed insulin-like growth factor-I at the 7.5 kDa position. Cultured tendon surface cells and tendon internal fibroblasts as well as cells in intact flexor tendon expressed insulin-like growth factor-I mRNA detected by reverse transcriptase-polymerase chain reaction. In situ hybridization histochemistry positively identified insulin-like growth factor-I mRNA in tendons from 52-day-old chickens. Platelet-derived growth factor was not detected at the protein or message levels. Furthermore, tendon surface cells and tendon internal fibroblasts both expressed receptors for insulin-like growth factor-I detected by flow cytometry. These data suggest that tendon cells express insulin-like growth factor-I mRNA and synthesize insulin-like growth factor-I in both the epitenon and the internal compartment of tendon, which is present in an inactive form, most likely bound to insulin-like growth factor-binding proteins.

A method for measuring the expression of integrin subunits on the cell surface of knee ligament fibroblasts was developed with use of flow cytometry and immunofluorescence. The ligament cells exhibited uniform size and density, as shown by forward and side-scatter properties, and showed minimal nonspecific binding of isotype control antibodies compared with unstained cells. All cells expressed the alpha5 integrin subunit; lateral collateral ligament cells stained with antibody to alpha5 showed a mean fluorescence intensity 2-fold higher than that of medial collateral ligament cells, 1.5-fold higher than that of posterior cruciate ligament cells, and 3-fold higher than that of anterior cruciate ligament cells, indicating a greater expression of the alpha5 subunit by lateral collateral ligament cells than by medial collateral, posterior cruciate, and anterior cruciate ligament cells. All cells expressed the beta1 integrin subunit; the expression by posterior cruciate ligament cells was 3-fold higher than that by medial collateral ligament or lateral collateral ligament cells and 5-fold higher than that by anterior cruciate ligament cells. All cells expressed the beta3 integrin subunit; the expression by posterior cruciate ligament cells was 1.5, 3, and 4.5-fold greater than that by lateral collateral, anterior cruciate, and medial collateral ligament cells, respectively. Our data suggest there is a differential expression of integrin subunits in knee ligament fibroblasts, and this in part may explain differences in their attachment and adherence to extracellular matrix molecules.

We determined the effect of cytokines on the proliferation and migration of cells isolated from the inner-third (white-white), middle-third (red-white), and outer-third (red-red) regions of bovine meniscus. Cells from the outer, or peripheral, region of the meniscus exhibited higher DNA synthesis in the presence of 10% serum compared with cells from the inner or central regions. Recombinant human platelet-derived growth factor-AB, hepatocyte growth factor/scatter factor, and bone morphogenic protein-2 stimulated DNA synthesis of all meniscal cells in a dose-dependent manner, with a two- to threefold maximal stimulation at 10 ng/ml. Cell migration was also stimulated by addition of cytokines. Platelet-derived growth factor and hepatocyte growth factor caused an increase in the migration of cells derived from all three zones, while interleukin-1 selectively stimulated the migration of outer-zone meniscal cells. Epidermal growth factor was much less effective and stimulated the migration of cells in the inner and outer zones by 40% to 50%, while bone morphogenic protein-2 and insulin-like growth factor-1 stimulated the migration of meniscal cells from the middle zone by 40% to 50%. The identification of cytokines that stimulate both the growth and migration of meniscal cells may provide new tools for modulation of meniscal healing.

Migration and proliferation of ligament fibroblasts are essential for the healing of ligament injuries. This study was designed to evaluate the migration of intraarticular (anterior and posterior cruciate) and extraarticular (medial and lateral collateral) ligament fibroblasts in response to cytokines and to determine the effect of cell passage on cell proliferation. Recombinant human platelet-derived growth factor, hepatocyte growth factor/scatter factor, and bone morphogenic protein-2 stimulated the migration of all ligament cells in a dose-dependent manner, with optimal migration at 10 ng/ml. Recombinant human epithelial growth factor preferentially stimulated the migration of intraarticular ligament fibroblasts, whereas recombinant human interleukin-1 was more effective with extraarticular ligament fibroblasts. Recombinant human insulin-like growth factor-1, insulin-like growth factor-2, transforming growth factor-beta, and fibroblast growth factor had no significant effect on the migration of ligament-derived fibroblasts. These data suggest that specific cytokines stimulate the migration of knee ligament fibroblasts and provide a rationale for possible therapeutic approaches to optimize ligament healing. Fibroblasts derived from the anterior cruciate ligament have been shown to proliferate at a slower rate than those derived from the medial collateral ligament. We have extended these observations and have demonstrated that fibroblasts from both the posterior and anterior cruciate ligaments proliferate at a slower rate than lateral and medial collateral ligament-derived fibroblasts. The differences between the growth rates of intraarticular and extraarticular fibroblasts become insignificant with serial passaging of the cells.

Based upon findings that the scatter factor/hepatocyte growth factor (SF/HGF) has strong mitogenic and motogenic properties, and that the sperm cell acquires its fertilizing capacity and motility in the distal parts of mammalian epididymis, the present study was conducted to investigate the role of SF/HGF in initiation of sperm cell motility. This was investigated by determining the expression of SF/HGF in various regions of the murine male genital tract by scatter and cell tracking assays using MDCK epithelial cells, Western blot procedure, and the immunohistochemical procedure using paraffin sections of various regions of the male genital tract. The findings from all these assays indicate that SF/HGF is differentially expressed in various parts of the male genital tract with slight or no expression in the testes, caput epididymis, and vas deferens, and with the highest expression in cauda and corpus (distal) epididymis followed by expression in the corpus (proximal) epididymis. This region-specific SF/HGF expression pattern coincides with the pattern of acquiring the fertilizing capacity and motility by the sperm cell during its transit through the male genital tract. However, wherever SF/HGF was expressed in the male genital tract, its molecular weight was slightly higher (Mr, 82 kD), compared to the SF/HGF expressed in various other somatic tissues (Mr, 78 kD), indicating that the genital tract SF/HGF may be a different molecular species that shares some immunoreactive epitopes with the somatic cell SF/HGF. Incubation of immotile sperm from caput epididymis with the purified human placental SF/HGF of 78 kD initiated motility in 5-15% of sperm population. These results strongly suggest that the SF/HGF-like activity is expressed in the male genital tract in a region-specific manner, and this activity may have a role in initiation of sperm motility acquired during its transit through the epididymis in mammals.

Using an automated cell analyzer system, the effect of hepatocyte growth factor/scatter factor (HGF/SF), epidermal growth factor (EGF), basic fibroblast growth factor (bFGF), endothelial acidic fibroblast growth factor (a-FGF), platelet derived growth factor (PDGF), and recombinant human insulinlike growth factor (IGF) on the motility and morphology of Madin-Darby canine kidney (MDCK), rat hepatomas, C2, and H5-6 and murine mammary carcinoma (EMT-6) cells was investigated. Treatment of MDCK cells with HGF/SF, bFGF, EGF, and a-FGF resulted in an increase in average cell velocity and in the fraction of moving cells. Cells treated with the PDGF and IGF did not show significant alterations in velocity. MDCK cells treated with each growth factor were classified into groups of "fast" and "slow" moving cells based on their average velocities, and the average morphologic features of the two groups were quantitated. Fast-moving cells had larger average area, circularity, and flatness as compared to slow-moving cells. Factors that stimulated cell movement also induced alterations in cell morphologic parameters including spreading, flatness, area, and circularity. HGF/SF also scattered and stimulated motility of C2 and H5-6 hepatoma cells. In contrast to MDCK cells, there was no significant difference between the morphology of the fast moving and slow moving C2 and H5-6 cells. These studies suggest that growth factor cytokines have specific effects on motility of normal and tumor cells.

In view of significant species-specific sequence differences between human and rat placental hepatocyte growth factor (HGF)/scatter factor (SF), the rat placental HGF/SF (rpSF) was purified, and its properties compared with human placental HGF/SF (hpSF). Like hpSF, rpSF scattered Madin-Darby canine kidney cells at 1-2 ng/ml and is composed of two subunits of 60 kDa and 30 kDa. Higher amounts (> 50%) of uncleaved 90-kDa form was present in the HGF/SF preparations from both human and rat placentas. Rat placental SF reacts with antibodies raised against hpSF in rabbits and chickens. The SF activity when expressed per gram rat placental tissue rises rapidly up to 9 days and then levels off. When expressed per milligram tissue protein it also increases rapidly up to 9 days and then declines. The expression of HGF/SF mRNA during development parallels that of HGF/SF activity. The specific activity of HGF/SF receptor (c-met) mRNA also appears to peak at 6 days. These findings suggest that (i) in spite of significant (> 10%) sequence differences between rpSF and hpSF, they exhibit similar structural, biologic, and immunologic characteristics and (ii) HGF/SF and its receptor are expressed in high amounts on Day 6 and then decline in developing placenta.

Scatter factor (also known as hepatocyte growth factor) is a glycoprotein secreted by stromal cells that stimulates cell motility and proliferation. In vitro, scatter factor stimulates vascular endothelial cell migration, proliferation, and organization into capillary-like tubes. Using two different in vivo assays, we showed that physiologic quantities of purified native mouse scatter factor and recombinant human hepatocyte growth factor induce angiogenesis (the formation of new blood vessels). The angiogenic activity was blocked by specific anti-scatter factor antibodies. Scatter factor induced cultured microvascular endothelial cells to accumulate and secrete significantly increased quantities of urokinase, an enzyme associated with development of an invasive endothelial phenotype during angiogenesis. We further showed that immunoreactive scatter factor is present surrounding sites of blood vessel formation in psoriatic skin. These findings suggest that scatter factor may act as a paracrine mediator in pathologic angiogenesis associated with human inflammatory disease.

Scatter factor (SF) is a glycoprotein which is secreted by mesenchymal cells and which causes cohesive epithelial cell colonies to spread out, separate into individual cells, and assume a fibroblastic morphology (i.e., to "scatter"). SF is now known to be identical or nearly identical to hepatocyte growth factor, a serum-derived mitogen for various normal cell types. SF, tumor necrosis factor-alpha (TNFa), and interleukin-1 (IL1) share the ability to stimulate scattering, motility, and protease production in a variety of human tumor cell types. SF and TNFa stimulate vascular endothelial cell motility in vitro and induce angiogenesis, the formation of new blood vessels, in vivo. These factors may participate in a cytokine network which regulates tumor invasion and metastasis directly by enhancing the malignant epithelial phenotype and indirectly by inducing tumor neovascularization.

HGF-SF is a recently discovered cytokine which has both mitogenic and motogenic effects on a wide variety of cells. We used a computerized digital imaging system to measure motility and morphology of isolated cells. In this chapter we will describe the effect of HGF-SF and of other growth factors on the velocity, area, circularity, and flatness of normal and tumor cells. We will then discuss the possible mechanism of HGF-SF induced motility and the possible role of this factor in biological and pathological processes.

Scatter factor (SF), a fibroblast-derived cytokine characterized by its ability to convert non-motile epithelial cells to a motile fibroblast-like phenotype, is identical to hepatocyte growth factor (HGF), a broad-spectrum mitogen. SF is a heterodimeric glycoprotein that is homologous to plasminogen and other blood coagulation proteases but lacks proteolytic activity. Its receptor is the c-met proto-oncogene product, a growth factor receptor-like transmembrane tyrosine kinase. This unique cytokine is also synthesized and secreted by vascular smooth muscle cells and acts on endothelial cells to stimulate migration, protease production, invasion, proliferation, and differentiation into capillary-like tubes in vitro. SF-containing implants in mouse subcutaneous tissue and rat cornea induce directed ingrowth of new blood vessels from surrounding tissue, with maximal angiogenic responses at doses of 100-200 ng of SF. Immunoreactive SF is expressed at sites of neovascularization within human psoriatic plaques. These findings suggest that SF may play a significant role in the formation and repair of blood vessels under physiologic and pathologic conditions.

We investigated the effects of human placental scatter factor (hSF), mouse scatter factor (mSF) and recombinant human hepatocyte growth factor (HGF) on motility and morphology of individual Madin-Darby canine kidney cells using a computerized cell tracking system. All three factors increased the velocity of individual cells and the ratio of moving to stationary cells. Similarly, all three factors caused changes in morphologic features of cells, leading to increased area, flatness, and polarity. Increases in area and flatness but not polarity were slightly greater with HGF than with hSF or mSF. These results suggest that SFs and HGF have similar effects on motility and morphology of isolated epithelial cells.

Scatter factor (SF) is a protein produced by fibroblasts, smooth muscle cells, and human placenta which scatter cohesive epithelial cell colonies and increases cellular motility. SF bound to concanavalin A and other lectins with high affinity. SF could also be stained with a glycoprotein specific stain. Incubation of producer cells (N-ras-transformed 3T3), with tunicamycin homolog A1 did not have any significant effect on the secretory activity of SF. The treatment of SF with N- and O-glycanases as well as endoglycosidase H had no effect on its activity. However, treatment of target (Madin Darby canine kidney) cells with tunicamycin A1, abolished the scattering response. These studies suggest that scatter factor is a glycoprotein, but glycosylation is not required for its secretion or activity by the producer cells; however, glycosylation of proteins in the target cells is required for SF action.

1. The hepatic content, biliary excretion, cytosolic protein binding and nuclear translocation of metabolites of i.v. administered 14C-3'-methyl-N,N-dimethyl-4-aminoazobenzene (3'-methyl-DAB) were investigated in rats at various stages of 2-acetamidofluorene (AAF)-induced hepatocarcinogenesis. 2. At nodular and post-nodular stages biliary excretion of radioactive metabolites was decreased, although hepatic content of radioactivity was similar to controls not dosed with AAF. The secretion in bile of a major azo dye binding protein was also decreased at these stages. 3. Binding of dye metabolites to cytosolic proteins was decreased by 40% at nodular and post-nodular stages compared to controls. 4. Translocation in vitro of dye metabolites from cytosol to nucleus at nodular and post-nodular stages was 40% less than that of controls. Since specific soluble proteins control translocation from cytosol into the nucleus (and bile), this decreased binding of metabolites may explain the diminished translocation of carcinogen metabolites into the nucleus.

Scatter factor (SF) causes contiguous sheets of epithelium to spread and cells to separate from each other. SF also increases the velocity, area, and reduces the circularity of individual cells. These changes are mediated in part by alterations in protein synthesis, protein phosphorylation, cytoskeletal reorganization, and cell surface components. SF has been purified from the conditioned medium of ras transformed 3T3 cells and human placenta. Sequence information suggests that SF from 3T3 cells is closely related to hepatocyte growth factor. SF is a glycoprotein, but glycosylation is not necessary for its activity. Glycosylation of target cell proteins, however, is required for SF action.

Two radiolabeled hepatocarcinogens, N,N-dimethyl-4-aminoazobenzene (DAB) and 3'-methyl-N,N-dimethyl-4-aminoazobenzene (3'-Me-DAB), were rapidly cleared from the blood of rats after i.v. administration, with half-lives of 40 and 70 sec, respectively. Rates of hepatic uptake and biliary secretion of [14C]-3'-Me-DAB were double that of [14C]DAB within 30 min of administration. Two hr after azo dye injection, the hepatic output into bile of [14C]-3'-Me-DAB-derived radioactivity was three times that of [14C]DAB. Fifty and 75% of the total 3'-Me-DAB-derived radioactivity was recovered in blood, liver, and bile 30 and 120 min after injection while only 30 to 40% of the administered [14C]DAB-derived radioactivity was recovered at these times. We postulate the existence of an extrahepatic azo dye accumulation site which may compete with the ability of the liver to clear azo dye from the circulation and which releases 3'-Me-DAB-derived radioactivity more readily than that of DAB. Azo dye metabolites were isolated from liver, bile, and blood. The chromatographic pattern of liver metabolites generated in vivo by rats which received either hepatocarcinogen was obtained and compared with that of biliary metabolites. With either azo dye, some metabolites were located exclusively in the liver, some were secreted immediately into bile, while others were present in both liver and bile, indicating selectivity in biliary excretion.

Eight-week-old rats had twofold higher hepatic ligandin concentration than 10-day-old animals as determined immunologically and by steroid isomerase and glutathione S-transferase assays. Increased ligandin content was accompanied by parallel increase in subunit synthesis as determined by [3H]leucine incorporation into each subunit relative to incorporation into total cytosolic proteins. The mRNA content for each ligandin subunit was twofold higher in older animals as determined by cell-free in vitro translation followed by immunoprecipitation and dot hybridization using a ligandin cDNA probe. When poly A mRNA from the postmitochondrial fraction of liver from young or old rats was subjected to agarose gel electrophoresis under denaturing conditions and hybridized to ligandin cDNA probe, a single 11 S band was obtained. With RNA from total liver, an additional 13 S band was obtained, suggesting the existence of a precursor form of ligandin mRNA. Since precursor polypeptides were not observed with RNA from total liver in cell-free in vitro translation systems, the precursor form requires processing to the 11 S form before the mRNA becomes functional.

Following administration of phenobarbital to rats, liver ligandin content, bilirubin binding, glutathione-S-transferase and steroid isomerase activities increased by 150% and the 22000-dalton subunit was selectively increased. Following administration of 3'-methyl-N,N-dimethyl-4-aminoazobenzene, rat liver ligandin content and steroid isomerase decreased by 65%, glutathione-S-transferase increased by 100%, bilirubin binding was abolished, and the relative proportion of the 22000- and 25000-dalton subunits remained unchanged.

Small amounts of metabolite-binding protein (MBP) originally characterized from the bile were detected in rat serum and cytosol by an indirect enzyme-linked immunoabsorbant assay. The site of MBP synthesis was shown to be the liver based upon results of (1) the in vitro translation of liver poly(A)+ mRNA, followed by immunoprecipitation with anti-MBP sera and sodium dodecyl sulfate-polyacrylamide gel electrophoresis and fluorography of the immunoprecipitate, and (2) immunoprecipitation of bile collected from [3H]leucine perfused liver in situ and SDS-polyacrylamide gel electrophoresis and fluorography of the immunoprecipitate. To determine whether part of the MBP in bile is derived from the circulation, [125I]MBP was injected intravenously and bile was collected and subjected to SDS-polyacrylamide gel electrophoresis and fluorography. Intact [125]MBP was not detected in bile even though several other iodinated bile proteins were taken up by the liver from the circulation and secreted intact into the bile under similar experimental conditions. These data indicate that MBP is synthesized in the liver and secreted into the bile and circulation independently. In addition, MBP was not found in brain, spleen or kidneys.

Several lines of evidence indicate that ligandin consists of two different subunits. The protein dissociates into two components that are detected by electrophoresis in a discontinuous sodium dodecyl sulfate system, or in acid-urea gels, and by isoelectric focusing in the presence of urea. The apparent molecular weights of the two polypeptides are 25,000 and 22,000. Alkylated or succinylated ligandins also exhibit subunit heterogeneity and resolved into two bands in these electrophoretic systems. Cross-linked ligandin showed only one band in sodium dodecyl sulfate-gel electrophoresis indicating that the two subunits are part of a heterodimeric protein rather than monomers of two different proteins. No dansylated terminal amino acids were detected suggesting that the NH2-terminal residues of both chains are blocked. One mole of arginine or phenylalanine was released per mole of ligandin after digestion with carboxypeptidase B or A, respectively. Tryptic maps of succinylated ligandin were consistent with identical disposition of arginine residues in both chains, but several additional tryptic peptides were obtained with native ligandin as compared to the predicted number if both subunits were identical. These observations are consistent with the possibility that both subunits contain common sequences and that a small peptide of about 25 to 30 amino acid residues is cleaved from the COOH-terminal of the larger subunit to produce the smaller subunit.

Using a monospecific antibody to the major cytosolic glutathione-S-transferase of human liver, we have isolated a cDNA clone from a human liver cDNA expression vector library in lambda gt11. The clone cross-hybridizes with a rat liver ligandin (glutathione-S-transferase 1-2) cDNA probe. The clone has an insert of 1.25 kb, a size sufficient to code for the 23 kilodalton subunit of human GST. Digestion of the insert with Hinf I produced three fragments (0.8 kb, 0.4 kb and 0.1 kb). A similar pattern of multiple bands was observed when rat liver GST1-2 cDNA probe was used for Southern blot analysis of Pst digests of rat and human genomic DNAs. These data suggest that these two functionally similar proteins exhibit sequence homology between their respective cDNAs and at ligandin loci, in spite of the lack of immuno-crossreactivity between them.

Physical methods and chemical modifications were used to discriminate between the bilirubin-binding capacity and glutathione-S-transferase activity of ligandin which was purified from rat liver. Binding of bilirubin occurs at a primary high affinity site (KA = 5 X 10(7) M-1) and at a secondary, lesser affinity site (KA = 3 X 10(5) M-1). Circular dichroism and fluorescence-quenching methods were used to distinguish between these sites. Cross-linked as well as reduced and alkylated ligandin lost high affinity bilirubin-binding capacity, but retained glutathione-S-transferase activity, bilirubin binding at a secondary site, and immunological reactivity. Succinylation of ligandin abolished catalytic activity and bilirubin binding at high and low affinity sites, but not immunological reactivity. Catalytic activity was unaffected by concentrations of bilirubin which saturated the primary binding site. These results suggest that the high affinity site at which bilirubin is bound to ligandin is independent from the site at which catalytically reactive substrates bind. The latter substrates probably interact at the secondary bilirubin binding site where bilirubin competitively inhibits glutathione-S-transferase activity.

Human lung acidic glutathione S-transferase is irreversibly inhibited by 1-chloro-2,4-dinitrobenzene (CDNB) in the absence of the co-substrate glutathione (GSH). The time-dependent inactivation is pseudo-first-order and demonstrates saturation kinetics, suggesting that inactivation occurs from an EI complex. The Ki was 0.14 mM; and kobs was 0.32 min-1 at 0.6 mM CDNB. The enzyme was protected against CDNB inactivation by GSH. The other two classes of glutathione S-transferase, the basic and near-neutral, are not significantly inactivated by CDNB. Incubation with [14C]CDNB indicated covalent binding to all three classes of transferase. One peptide fraction was found to be radiolabelled in both the basic and acidic transferases when these were incubated with [14C]CDNB and GSH, cleaved with cyanogen bromide, and chromatographed by HPLC. Incubation in the absence of GSH yielded one and two additional labelled peptide fractions for the basic and acidic transferases, respectively. Our results suggest that while CDNB arylates all three classes of human transferases, only the acidic transferase possesses a specific GSH-sensitive CDNB binding site, binding to which leads to time-dependent inactivation.

The pulse-chase technique was employed to determine the synthesis of the subunits of ligandin (glutathione S-transferase 1-2) by isolated hepatocytes. Ligandin comprised 2.5-3% of the total proteins synthesized. A slightly higher incorporation of [35S]methionine into the 22 k than the 25 k subunit was observed. However, the ratio of [35S]methionine incorporation into the subunits remained constant throughout the chase period, suggesting that, in spite of the considerable sequence homology, the conversion of 25 k to 22 k subunit does not occur in vivo.

Photoaffinity techniques were employed to affect the covalent binding of [35S]sulfobromophthalein to proteins of rat and human liver cytosol. In rat liver cytosol at low concentrations, sulfobromophthalein bound to the 22 kDa subunit of ligandin. In human liver cytosol, binding to a 23.5 kDa subunit was observed. At higher concentrations, sulfobromophthalein also bound to 12, 23.5, 37, and 42 kDa peptides. When the peptides resulting from CNBr cleavage of [35S]sulfobromophthalein-ligandin complex were resolved by high-performance liquid chromatography, radioactivity was associated with two peptides. The peptide containing 80% of the radioactivity was isolated and characterized. Its molecular weight is 3.4 kDa, it contains the single tryptophan residue of ligandin and has a glutamate (glutamine) as the N-terminal amino acid.

Ligandin was decreased by 75% as determined immunologically and by glutathione-S-transferase or steroid isomerase activities in rat hepatocellular carcinomas induced by exposure to N-2-fluorenylacetamide. Minor variable differences in ligandin levels were noted between the putative, premalignant nodules induced by this regimen and normal liver.

When liver cytosol prepared from rats administered [14C]3'-methyl-N,N-dimethyl-4-aminoazobenzene was subjected to Sephadex gel chromatography, four peaks (I-IV) of radioactivity containing proteins and one peak (V) of radioactivity devoid of protein were obtained. Forty to fifty-five per cent of the radioactivity in the protein peaks was butanol-extractable. When the protein peaks were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, over 90% of the radioactivity was separated from the proteins, indicating lack of covalent binding. Several differences in the metabolite patterns were seen when the butanol-extractable metabolites from the five chromatographic peaks were analyzed by TLC. When pooled fractions of the peaks were incubated with isolated rat liver nuclei, only radioactivity associated with peak II was translocated into the nucleus. Translocation was time- and temperature-dependent and was maximal at 40 min at 37 degrees C. Only 10 to 12% of the radioactivity associated with peak II could be translocated even in the presence of an excess of nuclei, indicating that specific protein metabolite adduct(s) present in this fraction is/are translocated. Five per cent of translocated radioactivity was irreversibly bound to DNA, 3% to RNA, 67% to non-histone proteins, and 7.5% to histones; the remaining was not associated with any of these macromolecules.

Synthesis of both subunits (Ya and Yb) of ligandin in equal amounts was observed when poly(A)+ mRNA isolated from the post-mitochondrial fraction was translated in an in vitro wheat-germ system and the products were immunoprecipitated by monospecific antibody to ligandin and analyzed by SDS-polyacrylamide gel electrophoresis and fluorography. When the Mg2+ or K+ concentrations were increased in the in vitro wheat-germ system the ratio of synthesis of Yb/Ya subunits was 3. With a mRNA-dependent reticulocyte lysate, the synthesis of Ya subunits was 20-30% higher than Yb subunits. At a fixed K+ and Mg2+ concentration, the ratio of incorporation of [35S]methionine into Yb/Ya subunits remained 1 and 0.7 in wheat-germ and reticulocyte lysate systems, respectively, up to 60 min. When poly(A)+ mRNA was fractionated on a 5-20% sucrose gradient, ligandin mRNA was present in fractions having a peak sedimentation value of 11 S. When poly(A)+ mRNA was fractionated by gel electrophoresis, fractions enriched in mRNA for each subunit were obtained. By administration of [3H]leucine followed by determination of radioactivity in ligandin and total proteins by immunoprecipitation and trichloroacetic acid precipitation, respectively, synthesis of the Ya subunits was selectively stimulated by phenobarbital administration. When poly(A)+ mRNA from liver of rats administered phenobarbital was translated in vitro a selective increase in the mRNA content of Ya subunits was observed. When poly(A)+ RNA from testes was translated in the wheat-germ system and products analyzed, Yb subunits were the predominant subunit (greater than 90%) synthesized, reflecting the subunit composition of testicular ligandin. These results suggest that in spite of the close sequence homology between the two subunits of ligandin, there are separate mRNA's for each subunit which are independently regulated.

Uptake, metabolism, and biliary secretion of 3'-methyl-N,N-dimethyl-4-aminoazobenzene (3'-Me-DAB) were studied in isolated rat liver which was perfused with protein-free fluorocarbon medium. [14C]3'-Me-DAB (5-10 nmol) was injected into the portal vein and allowed to recirculate. The recovery of radioactivity in bile was 7.5, 14, and 20% at 15, 30, and 45 min of injection, respectively. At 45 min, the liver contained an additional 17% of injected radioactivity. Azo dye metabolites in perfused liver differed from those in vivo; metabolites co-migrating with 3'-CHO-DAB and 3'-methyl-n,n-methyl aminoazobenzene (Me-MAB) (and the parent compound 3'-Me-DAB) were present, while metabolites co-migrating with 3'-Me-4'-OH-AB and 3'-CH2OH-MAB were increased and metabolites co-migrating with 3'-CH2OH-DAB were decreased. In bile from perfused liver, metabolites co-migrating with 3'-CHO-DAB and 3'-Me-MAB were undetectable. When proteins from normal rat bile were injected into portal vein 15 min after the administration of 3'-Me-DAB, the compounds co-migrating with 3'-Me-MAB, 3'-CHO-DAB, 3'-Me-4'-OH-AB, and 3'-CH2OH-MAB decreased, and compounds co-migrating with 3'-CH2OH-DAB increased in the liver; in bile, there was increase in 3'-Me-MAB, 3'-CHO-DAB, and 3'-Me-4'-OH-MAB, which there was a decrease in N-Ac-3'-Me-4'-OH-AB and 3'-COOH-DAB. Appearance of protein-metabolite adducts in bile was also observed after addition of normal bile proteins to the perfusate.(ABSTRACT TRUNCATED AT 250 WORDS)

Rat testicular and liver cystols contain ligandin as determined immunologically, and have high glutathione-S-transferase activity. Unlike liver cytosol, testicular cytosol does not contain protein components that bind bilirubin or sulfobromophthalein with high affinity. To investigate these effects, ligandin was purified to homogeneity from rat testis. Whereas rat liver ligandin consists of equal amounts of two subunits with molecular weights of 22,000 (Ya) and 25,000 (Yb), more than 90% of testicular ligandin consists of Yb. Rat testicular ligandin is immunologically similar to liver ligandin, and has identical glutathione-S-transferase activity, but lacks the capacity for high affinity binding of bilirubin and sulfobromophthalein. The amino acid composition and other properties of testicular ligandin are similar to those of the Yb subunit of liver ligandin. Sulfobromophthalein and bilirubin biphasically inhibit the glutathione-S-transferase activity of liver ligandin: initial high affinity inhibition is followed by reduced inhibition. Testicular ligandin has only low affinity inhibition kinetics. These results suggest that Ya is required for high affinity binding, and that reduced organic anion binding by testicular ligandin results from the lower amounts of Ya in testis.

Following intravenous administration to rats of the azo dye hepatocarcinogen 3'-methyl-N,N-dimethyl-4-aminoazo-[14C]benzene, 60-70% of the injected dose was recovered in bile in 2 h. Approximately 10% of bile radioactivity was trichloroacetic acid-precipitable, not extracted by n-butanol and non-dialyzable. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of bile followed by fluorography revealed two major and several minor proteins to which radiolabelled azo dye metabolites were bound; one of these major proteins (50 kDa) was purified from bile and shown to be homogeneous by SDS-polyacrylamide gel electrophoresis, isoelectric focusing (pI 7) under denaturing conditions and N-terminal analysis. The protein (MBP) comprises 15% of the total bile protein. Amino acid analysis revealed a preponderance of acidic and hydrophobic amino acids. The absorption spectrum of the native protein had a major peak at 280 nm and minor peaks at 345, 400, 600 and 650 nm. The fluorescence spectrum showed a major excitation maxima at 285 and 350 nm and corresponding emission maxima at 345 and 440 nm. Atomic absorption spectroscopy revealed 5 atoms of Cu per mol protein. Approximately 90% of the Cu was EPR silent. MBP did not react with antisera directed against rat serum IgA, albumin or ceruloplasmin; nor did these proteins react against antisera to MBP. Seven distinct peptide bands ranging from 5 to 18 kDa were obtained when MBP was subjected to CNBr cleavage and the digests were analyzed by SDS-polyacrylamide gel electrophoresis. The [14C]-Me-DAB derived radioactivity was present in only two of the peptides, indicating specific binding sites for azodye metabolites.

Bilirubin glucuronoside glucuronosyltransferase (EC 2.4.1.95) converts bilirubin monoglucuronide to bilirubin diglucuronide and is concentrated in plasma membrane-enriched fractions of rat liver homogenates. The enzyme was purified 2,000-fold to homogeneity from rat liver. The pI of the enzyme is 7.9 +/- 0.2. The enzyme has a molecular weight of 160,000 and is an oligomer of 28,000 dalton subunits. Km for purified enzyme was 35 microM and Vmax was 2.2 mumol of bilirubin diglucuronide formed/min/mg of protein. Freshly biosynthesized bilirubin monoglucuronide was injected intravenously into homozygous Gunn rats which had bile duct cannulation. Gunn rats lack UDP-glucuronate glucuronyltransferase activity (EC 2.4.1.17), have normal bilirubin glucuronoside glucuronosyltransferase activity, cannot form bilirubin monoglucuronide in vitro or in vivo, and do not excrete bilirubin glucuronides after intravenous injection of unconjugated bilirubin. Within 1 h, approximately 75% of the injected conjugated bilirubin was recovered in bile, of which 20% consisted of bilirubin diglucuronide. These results indicate that bilirubin glucuronide glucuronosyltransferase catalyzes conversion of bilirubin monoglucuronide to diglucuronide in vivo.

1. Butan-1-ol solubilizes that portion of rat liver mitochondrial aspartate aminotransferase (EC 2.6.1.1) that cannot be solubilized by ultrasonics and other treatments. 2. A difference in electrophoretic mobilities, chromatographic behaviour and solubility characteristics between the enzymes solubilized by ultrasonic treatment and by butan-1-ol was observed, suggesting the occurrence of two forms of this enzyme in rat liver mitochondria. 3. Half the aspartate aminotransferase activity of rat kidney homogenate was present in a high-speed supernatant fraction, the remainder being in the mitochondria. 4. A considerable increase in aspartate aminotransferase activity was observed when kidney mitochondrial suspensions were treated with ultrasonics or detergents. 5. All the activity after maximum activation was recoverable in the supernatant after centrifugation at 105000g for 1hr. 6. The electrophoretic mobility of the kidney mitochondrial enzyme was cathodic and that of the supernatant enzyme anodic. 7. Cortisone administration increased the activities of both mitochondrial and supernatant aspartate aminotransferases of liver, but only that of the supernatant enzyme of kidney.