Immunotoxins containing Pseudomonas exotoxin that target LeY damage human endothelial cells in an antibody-specific mode: relevance to vascular leak syndrome

Vascular leak syndrome (VLS) was originally found to be a major dose-limiting toxicity in humans with cancer treated with several immunotoxins (ITs) containing ricin A chain or blocked ricin. Recently, VLS has also been observed in patients treated with an IT containing the murine monoclonal antibody (MAb) B3 coupled to LysPE38, a recombinant truncated form of Pseudomonas exotoxin (PE) A. Antibody B3 (IgG1k) recognizes LewisY and related carbohydrate epitopes present on many human solid tumors, and B3-LysPE38 showed excellent antitumor activity in nude mice bearing tumors that express the B3 antigen. In the clinical trial, the development of VLS has prevented the administration of the amount of IT necessary to achieve blood levels required for good therapeutic responses. We have now investigated the effects of several PE-based ITs on different human endothelial cell lines to elucidate the mechanism of VLS induced by ITs containing PE. To assess the cytotoxic effect of IT on endothelial cells, various ITs were incubated with cells for 2 or 20 h, and the incorporation of [3H]leucine into protein was measured. The endothelial cells studied were human umbilical vein endothelial cells, human lung-derived microvascular endothelial cells (HUVECs), human adult dermal microvascular endothelial cells, human pulmonary artery endothelial cells, and human aortic endothelial cells. We found that both B3-LysPE38 (LMB-1), a chemical conjugate of MAb B3 with PE38, as well as B3(Fv)-PE38 (LMB-7), a recombinant single chain immunotoxin, inhibited protein synthesis, with 50% inhibitory concentrations between 600 and 1000 ng/ml for 20-h incubation in HUVECs, human lung-derived microvascular endothelial cells, and human adult dermal microvascular endothelial cells but not on human pulmonary artery endothelial cells. The cytotoxic effect was specific since PE38 itself or PE coupled to several other antibodies did not inhibit protein synthesis in these cells even at 10,000 ng/ml. Further evidence that the cytotoxicity of B3-containing ITs is due to specific B3 binding to endothelial cells comes from the fact that the cytotoxicity can be blocked by excess free MAb B3. HUVECs undergo overt morphological changes after treatment with B3-LysPE38 or B3(Fv)PE38. Gaps between the cells are formed after a 20-h exposure but not after 2 h. These studies suggest that VLS in patients is due to capillary damage caused by prolonged exposure to high concentrations of LMB-1.     

A dual role for endothelial cells in cytomegalovirus infection? A study of cytomegalovirus infection in a series of rat endothelial cell lines

Several clinical findings point to the involvement of microvascular endothelial cells in cytomegalovirus-related pathology. In this study the interactions of cytomegalovirus (CMV) with microvascular endothelial cells was investigated in an in vitro rat model. A series of rat endothelial cell lines, considered representative for the heterogeneity of heart microvascular endothelium in vivo, were infected with rat CMV (RCMV). The course of infection and production of infectious virus were examined using immunofluorescence staining and plaque titration assays, and was compared with infection of fully permissive rat fibroblasts. These endothelial cell lines displayed differences in susceptibility to CMV infection. Two endothelial cell lines (RHEC 50 and 191) were practically non-permissive, while four endothelial cell lines (RHEC 3, 10, 11 and 116) were partly permissive for CMV infection. In contrast to CMV infection in fibroblasts, only limited infection of the permissive endothelial cell lines was observed without spreading of CMV infection through the monolayer, although infectious virus was produced. Detachment of infected endothelial cells and recovery of the monolayer with time was observed. The detached endothelial cells were able to transmit CMV infection to fibroblast monolayers, but not to endothelial monolayers. Our in vitro results demonstrate differences in permissiveness for RCMV between the series of rat endothelial cell lines, which is suggestive for endothelial heterogeneity to CMV infection in vivo. Our findings indicate that endothelial cells are relatively resistant to CMV infection and that, upon infection, the endothelial monolayer may dispose of the virus via detachment of the infected cells. This points to a dual role for the endothelium in CMV infection in vivo: a barrier for CMV infection (by the endothelial monolayer) on the one hand and spreading of CMV infection (by detached infected cells) on the other hand.     

Evidence for the involvement of Fas ligand and perforin in the induction of vascular leak syndrome

Endothelial cell injury resulting in vascular leak syndrome (VLS) is one of the most widely noted phenomenons in a variety of clinical diseases. In the current study we used IL-2-induced VLS as a model to investigate the role of cytolytic lymphocytes in the cytotoxicity of endothelial cells. Administration of IL-2 (75,000 U/mouse, three times a day for 3 days) into BL/6 wild-type mice triggered significant VLS in the lungs, liver, and spleen. Interestingly, perforin-knockout (KO) mice exhibited a marked decrease in IL-2-induced VLS in all three organs tested. Also, Fas ligand-defective (gld) mice and Fas-deficient (lpr) mice exhibited decreased VLS in the liver and spleen, but not in the lungs. The decreased VLS seen in perforin-KO, gld, and lpr mice was not due to any defect in lymphocyte migration or homing to various organs because histopathologic studies in these mice demonstrated significant and often greater perivascular infiltration of lymphocytes compared with the IL-2-treated wild-type mice. Ultrastructural studies of the lungs demonstrated significant damage to the endothelial cells in IL-2-treated wild-type mice and decreased damage in perforin-KO mice. IL-2 administration caused up-regulation of CD44 in all strains of mice tested and triggered increased LAK activity against an endothelial cell line in wild-type and gld mice, but not in perforin-KO mice. The current study demonstrates for the first time that perforin and Fas ligand may actively participate in endothelial cell injury and induction of VLS in a variety of organs.     

Inhibitory effects of malotilate on invasion and metastasis of rat mammary carcinoma cells by modifying the functions of vascular endothelial cells

Malotilate (diisopropyl,1,3-dithiol-2-ylidenemalonate, MT) is clinically used as a hepatoprotective agent. Because we noticed that MT induced the differentiation of cultured vascular endothelial cells, we have examined its effects on lung metastasis of the highly metastatic rat mammary carcinoma c-SST-2. MT was orally administered to syngeneic SHR rats from 7 days before or after s.c. inoculation of c-SST-2 cells to the end of the experiments. In the MT-treated rats, pulmonary metastasis was markedly suppressed compared with the non-treated rats. In the rats treated with MT for 19 days after i.v. inoculation of c-SST-2 cells, lung metastasis was also significantly suppressed. An in vitro invasion assay using a rat lung endothelial (RLE) cell monolayer revealed that pretreatment of the RLE cells with MT, but not c-SST-2 cells, significantly reduced the invasion of the RLE monolayer by c-SST-2 cells. An in vitro vascular permeability assay demonstrated that MT prevented the increase in permeability of the RLE monolayer by serum starvation. On the other hand, in vivo and in vitro growth, gelatinase production and adhesion to the RLE cell monolayer of c-SST-2 cells were not affected by MT treatment. These findings suggest that MT suppressed tumour metastasis by intensifying the cell-to-cell contact of endothelial cells, thus preventing tumour cells from invading vascular endothelium.     

Cell shape changes and cytoskeleton reorganization during transendothelial migration of human melanoma cells

An in vitro system has been established to study the migration of human melanoma cells through a monolayer of endothelial cells. Endothelial cells were cultured to confluence on Matrigel before the seeding of melanoma cells. Laser scanning confocal microscopy showed that, prior to migration, melanoma cells appeared round and showed cortical F-actin staining. The initial stage of transmigration was characterized by numerous membrane blebs protruding from basolateral surfaces of the melanoma cells, and contact regions showed an abundance of filaments arising in the underlying endothelial cells. Later, pseudopods from the melanoma cells inserted into contact regions between endothelial cells. Eventually, the melanoma cells intercalated with the endothelial cells. At this stage, many endothelial filament bundles terminated at contacts between the endothelial cells and the transmigrating melanoma cell, suggesting active interactions between the two cell types. Upon contact with the Matrigel, melanoma cells began to spread beneath the endothelium, displaying a fibroblastic morphology with prominent stress fibers. To reestablish the monolayer, adjacent endothelial cells extended processes over the melanoma cell. Tumor necrosis factor alpha did not affect the transmigration of melanoma cells from cell lines isolated from several stages of metastasis. However, tumor necrosis factor did promote the transmigration of melanoma cells derived from a non-metastatic lesion. These results thus define cell attachment and cell penetration of the monolayer as two distinct steps in transmigration and suggest that tumor necrosis factor may enhance the metastatic potential of tumor cells.     

Morphological and functional characterization of an in vitro blood-brain barrier model

Cell culture models have been extensively used for studies of blood-brain barrier (BBB) function. However, several in vitro models fail to reproduce some, if not most, of the physiological and morphological properties of in situ brain microvascular endothelial cells. We have recently developed a dynamic, tridimensional BBB model where endothelial cells exposed to intraluminal flow form a barrier to ions and proteins following prolonged co-culturing with glia. We have further characterized this cell culture model to determine whether these barrier properties were due to expression of a BBB phenotype. Endothelial cells of human, bovine or rodent origin were used. When co-cultured with glia, intraluminally grown endothelial cells developed features similar to in vivo endothelial cells, including tight junctional contacts at interdigitating processes and a high transendothelial resistance. This in vitro BBB was characterized by the expression of an abluminal, ouabain-sensitive Na/K pump, and thus favored passage of potassium ions towards the lumen while preventing K+ extravasation. Similarly, the in vitro BBB prevented the passage of blood-brain barrier-impermeant drugs (such as morphine, sucrose and mannitol) while allowing extraluminal accumulation of lipophylic substances such as theophylline. Finally, expression of stereo-selective transporters for Aspartate was revealed by tracer studies. We conclude that the in vitro dynamic BBB model may become an useful tool for the studies of BBB-function and for the testing of drug passage across the brain endothelial monolayer.     

Astrocytic contributions to blood-brain barrier (BBB) formation by endothelial cells: a possible use of aortic endothelial cell for in vitro BBB model

Astrocytic contribution of endothelial cell monolayer permeability was examined in two blood-brain barrier (BBB) models, using the coculture in a double chamber system: rat astrocytes and bovine aortic endothelial cells (BAECs) or bovine brain endothelial cells (BBECs). In system 1, where astrocytes were separated from endothelial cells, a 40% reduction in L-glucose permeability of the BBEC monolayer, but not the BAEC monolayer, was observed by cocultivation with astrocytes. Although several passages of BBEC in culture elicited morphological transformation from spindle-shapes to cobblestone-like features, the passaged BBECs remained responsive to astrocytes in coculture in system 1 (37% reduction of the L-glucose permeability). By contrast, in system 2, where respective endothelial cells and astrocytes layered on the upper and lower surfaces of a membrane, the permeability of both BAEC and BBEC monolayers was reduced by cocultivation with astrocytes (75% reduction for BAEC and 40% reduction for BBEC). BAECs in this contiguous coculture (system 2) with astrocytes showed numerous tight junction-like structures characteristic of the BBB in vivo. These results suggest that primary cultured BBECs, which had been primed by astrocytes in vivo, retain a higher sensitivity to astrocytes possibly through an astrocytic soluble factor (s) to exhibit BBB-specific phenotypes, and that even BAEC from extra-neural tissues, when cultured with astrocytes in close proximity in vitro, may acquire the similar phenotypes and serve for an extensive use of BBB model in vitro.     

Monocyte diapedesis through an in vitro vessel wall construct: inhibition with monoclonal antibodies to thrombospondin

Human peripheral blood monocytes were examined for migration across an endothelial cell monolayer in an in vitro vessel wall construct. Few monocytes invaded in the absence of a chemotactic gradient, despite significant adhesion to the endothelial monolayer. However, the addition of zymosan-activated human plasma to the lower compartment, to create a chemotactic gradient across the vessel wall, resulted in significantly enhanced monocyte migration. Pretreatment of the monocytes with monoclonal antibodies to thrombospondin (TSP) dramatically inhibited monocyte diapedesis into the vessel wall. The same treatment inhibited monocyte adhesion to endothelial cells in two-dimensional monolayer cultures as well as in vessel wall constructs (no chemotactic gradient). Of interest, however, the monoclonal antibodies had no inhibitory effect on monocyte migration into collagen gels devoid of endothelial cells in response to the same chemotactic gradient, suggesting the importance of TSP in monocyte-endothelial cell interactions. Monoclonal antibodies to fibronectin and normal mouse immunoglobulin G did not inhibit migration in this model of a vessel wall. Furthermore, monoclonal antibodies to TSP showed no inhibition of human peripheral blood neutrophil migration. Previous studies have shown that monocytes synthesize TSP and express this moiety on their surface. The present data suggest that monocytes may utilize TSP to interact with endothelial cells lining the vessel wall during diapedesis.     

Chronic in vitro shear stress stimulates endothelial cell retention on prosthetic vascular grafts and reduces subsequent in vivo neointimal thickness

OBJECTIVE: The absence of endothelial cells at the luminal surface of a prosthetic vascular graft potentiates thrombosis and neointimal hyperplasia, which are common causes of graft failure in humans. This study tested the hypothesis that pretreatment with chronic in vitro shear stress enhances subsequent endothelial cell retention on vascular grafts implanted in vivo. METHODS: Cultured endothelial cells derived from Fischer 344 rat aorta were seeded onto the luminal surface of 1.5-mm internal diameter polyurethane vascular grafts. The seeded grafts were treated for 3 days with 1 dyne/cm2 shear stress and then for an additional 3 days with 1 or 25 dyne/cm2 shear stress in vitro. The grafts then were implanted as aortic interposition grafts into syngeneic rats in vivo. Grafts that were similarly seeded with endothelial cells but not treated with shear stress and grafts that were not seeded with endothelial cells served as controls. The surgical hemostasis time was monitored. Endothelial cell identity, density, and graft patency rate were evaluated 24 hours after implantation. Endothelial cell identity in vivo was confirmed with cells transduced in vitro with beta-galactosidase complementary DNA in a replication-deficient adenoviral vector. Histologic, scanning electron microscopic, and immunohistochemical analyses were performed 1 week and 3 months after implantation to establish cell identity and to measure neointimal thickness. RESULTS: The pretreatment with 25 dyne/cm2 but not with 0 or 1 dyne/cm2 shear stress resulted in the retention of fully confluent endothelial cell monolayers on the grafts 24 hours after implantation in vivo. Retention of seeded endothelial cells was confirmed by the observation that beta-galactosidase transduced cells were retained as a monolayer 24 hours after implantation in vivo. In the grafts with adherent endothelial cells that were pretreated with shear stress, immediate graft thrombosis was inhibited and surgical hemostasis time was significantly prolonged. Confluent intimal endothelial cell monolayers also were present 1 week and 3 months after implantation. However, 1 week after implantation, macrophage infiltration was observed beneath the luminal cell monolayer. Three months after the implantation in vivo, subendothelial neointimal cells that contained alpha-smooth muscle actin were present. The thickness of this neointima averaged 41 +/- 12 micrometer and 60 +/- 23 micrometer in endothelial cell-seeded grafts that were pretreated with 25 dyne/cm2 shear stress and 1 dyne/cm2 shear stress, respectively, and 158 +/- 46 micrometer in grafts that were not seeded with endothelial cells. CONCLUSION: The effect of chronic shear stress on the enhancement of endothelial cell retention in vitro can be exploited to fully endothelialize synthetic vascular grafts, which reduces immediate in vivo graft thrombosis and subsequent neointimal thickness.     

Apposition-dependent induction of plasminogen activator inhibitor type 1 expression: a mechanism for balancing pericellular proteolysis during angiogenesis

Plasminogen-activator inhibitor type I (PAI-1), the primary inhibitor of urinary-type plasminogen activator, is thought to play an important role in the control of stroma invasion by both endothelial and tumor cells. Using an in vitro angiogenesis model of capillary extension through a preformed monolayer, in conjunction with in situ hybridization analysis, we showed that PAI-1 mRNA is specifically induced in cells juxtaposed next to elongating sprouts. To further establish that PAI-1 expression is induced as a consequence of a direct contact with endothelial cells, coculture experiments were performed. PAI-1 mRNA was induced exclusively in fibroblasts (L-cells) contacting endothelial cell (LE-II) colonies. Reporter gene constructs driven by a PAI-1 promoter and stably transfected into L-cells were used to establish that both mouse and rat PAI-1 promoters mediate apposition-dependent regulation. This mode of PAI-1 regulation is not mediated by plasmin, as an identical spatial pattern of expression was detected in cocultures treated with plasmin inhibitors. Because endothelial cells may establish direct contacts with fibroblasts only during angiogenesis, we propose that focal induction of PAI-1 at the site of heterotypic cell contacts provides a mechanism to negate excessive pericellular proteolysis associated with endothelial cell invasion.     

Two distinct patterns of peritoneal involvement shown by in vitro and in vivo ovarian cancer dissemination models

We established an in vitro peritoneal dissemination model using six ovarian cancer cell lines and cultured mesothelial cells. Ovarian cancer cells were classified into two types, invasive or adhesive, on the basis of their interaction with the mesothelial cell monolayer. The ovarian cancer cell lines derived from mucinous cystadenocarcinoma, poorly differentiated adenocarcinoma and undifferentiated carcinoma, which belonged to the invasive type, began to invade beneath the mesothelial monolayer from several hours after seeding in vitro, expelling the mesothelial cells at the periphery and forming colonies directly on the dish surface. On the other hand, cancer cell lines of clear cell carcinoma, which belonged to the adhesive type, showed colony formation with adhesion on the mesothelial monolayer even 18 h after seeding. Invasive-type cell lines invaded into the mesothelial monolayer at various rates in vitro, and the degree of invasiveness showed good correlation with the degree of peritoneal dissemination in vivo after intraperitoneal injection of cancer cells into nude mice. Adhesive-type cells showed rather higher dissemination rates in vivo. Microscopic observation of in vivo peritoneal dissemination at one day after inoculation also revealed two patterns of peritoneal involvement similar to those in vitro. In the in vitro model, anti-integrin alpha 2- and beta 1-antibodies inhibited the infiltration of invasive-type cells into the mesothelial monolayer, but did not affect colony formation by adhesive-type cells on the monolayer, indicating that invasion by both cell types was mediated by different molecules. This in vitro model is thought to be useful for analysis of the molecular mechanisms of peritoneal dissemination.     

Modulation of retinal endothelial barrier in an in vitro model of the retinal microvasculature

The prediabetic/diabetic condition functionally alters the microvascular bed of the eye and the breakdown in the transvascular barrier may be produced by changes in the retinal endothelial barrier. To better understand how retinal microvessel barrier is maintained and is altered in vivo this study applies and extends our previously described in vitro permeability technique to study retinal endothelial monolayers. The model of the retinal microvasculature consists of retinal capillary endothelial cells cultured on porous microcarrier beads and perfused in chromatographic 'cell-columns'. This model design relies on indicator-dilution techniques to measure the permeability of the retinal endothelial monolayer and detects small changes in retinal endothelial permeability produced by treatments. Bovine retinal capillary endothelial cells (RCE) were obtained using an endothelial selective media. RCE were seeded at 3 x 10(4) cells cm-2 of fibronectin-coated gelatin microcarriers. After 7 days of microcarrier culture, microcarriers were poured to form columns 0.66 cm in diameter and 1.6 cm in length. The cell-column elution patterns of coinjected optically absorbing tracers (blue dextran 2 x 10(6) Da; cyanocobalamin 1355 Da; sodium fluorescein 376 Da) were analysed to estimate the permeability of the RCE monolayers covering the microcarriers. Scanning electron microscopic examination showed complete monolayer formation on the surface of the microcarriers. We found that baseline monolayer permeability averaged 7.57 +/- 0.57 x 10(-5) cm sec-1 for cyanocobalamin and 9.29 +/- 0.78 x 10(-5) cm sec-1 for sodium fluorescein (mean +/- S.E.M., n = 39). Permeability did not increase over 2 hr of cell-column perfusion. Permeability was decreased by 1 micron isoproterenol (n = 3) and increased by 1 microgram ml-1 cytochalasin D (n = 5). This is one of the first reports of in vitro permeability values for the transport barrier formed by retinal microvascular endothelial cells. Furthermore, the endothelial component of the retinal barrier is dynamic, and is enhanced by isoproterenol and diminished by cytochalasin D.     

Vascular endothelial growth factor (VEGF)-C synergizes with basic fibroblast growth factor and VEGF in the induction of angiogenesis in vitro and alters endothelial cell extracellular proteolytic activity

Vascular endothelial growth factor-C (VEGF-C) is a recently characterized member of the VEGF family of angiogenic polypeptides. We demonstrate here that VEGF-C is angiogenic in vitro when added to bovine aortic or lymphatic endothelial (BAE and BLE) cells but has little or no effect on bovine microvascular endothelial (BME) cells. As reported previously for VEGF, VEGF-C and basic fibroblast growth factor (bFGF) induced a synergistic in vitro angiogenic response in all three cells lines. Unexpectedly, VEGF and VEGF-C also synergized in the in vitro angiogenic response when assessed on BAE cells. Characterization of VEGF receptor (VEGFR) expression revealed that BME, BAE, and BLE cell lines express VEGFR-1 and -2, whereas of the three cell lines assessed, only BAE cells express VEGFR-3. We also demonstrate that VEGF-C increases plasminogen activator (PA) activity in the three bovine endothelial cell lines and that this is accompanied by a concomitant increase in PA inhibitor-1. Addition of alpha2-antiplasmin to BME cells co-treated with bFGF and VEGF-C partially inhibited collagen gel invasion. These results demonstrate, first, that by acting in concert with bFGF or VEGF, VEGF-C has a potent synergistic effect on the induction of angiogenesis in vitro and, second, that like VEGF and bFGF, VEGF-C is capable of altering endothelial cell extracellular proteolytic activity. These observations also highlight the notion of context, i.e., that the activity of an angiogenesis-regulating cytokine depends on the presence and concentration of other cytokines in the pericellular environment of the responding endothelial cell.     

Fluorescence activated cell sorting: a reliable method in tissue engineering of a bioprosthetic heart valve

BACKGROUND: Techniques of tissue engineering are used to seed human autologous cells in vitro on degradable mesh to create new functional tissue like a bioprosthetic heart valve. A precondition is subsequent seeding of native-valve-analogous pure endothelial and myofibroblast cell lines. The aim of this study is to find a safe method of isolating viable cell lines out of tissues from the operating room. METHODS: Mixed cells from ascending aorta obtained from the operating room were incubated with an endothelial-specific fluorescent marker. The labeled cells were activated and sorted by flow cytometry. Isolated cell lines were cultured and thereafter square sheets of polymeric scaffold were seeded with myofibroblasts, followed by endothelial cells. The created tissue was stained with hematoxylin and eosin, van Gieson stain, and stains for factor VIII and CD34. RESULTS: Control culture samples (n = 25) revealed vital uncontaminated endothelial and myofibroblast cell lines. Microscopy of the seeded meshes (n = 16) demonstrated a tissue-like structure. Van Gieson stain showed production of collagen. Endothelial cells formed a superficial monolayer, demonstrated by factor VIII and CD34; no invasive formation of capillaries was detectable. CONCLUSIONS: These results demonstrate that fluorescence activated cell sorting is a reliable and safe method to gain pure vital autologous cell lines out of human mixed cells for subsequent seeding on degradable mesh and that those cells are active to form new tissue.     

Vascular leak syndrome: a side effect of immunotherapy

The major dose-limiting toxicity of interleukin-2 (IL-2) and of immunotoxin (IT) therapies is vascular leak syndrome (VLS). VLS is characterized by an increase in vascular permeability accompanied by extravasation of fluids and proteins resulting in interstitial edema and organ failure. Manifestations of VLS include fluid retention, increase in body weight, peripheral edema, pleural and pericardial effusions, ascites, anasarca and, in severe form, signs of pulmonary and cardiovascular failure. Symptoms are highly variable among patients and the causes are poorly understood. The pathogenesis of endothelial cell (EC) damage is complex and can involve activation or damage to ECs and leukocytes, release of cytokines and of inflammatory mediators, alteration in cell-cell and cell-matrix adhesion and in cytoskeleton function. VLS restricts the doses of IL-2 and of ITs which can be administered to humans and, in some cases, necessitates the cessation of therapy. This review discusses the diversity of clinical manifestation, possible mechanisms and therapeutic modalities for VLS induced by IL-2 and ITs.     

Bradycardia induced by intravascular versus direct stimulation of the vagus nerve

Ricin is a heterodimeric cytotoxin composed of RTB, a galactose binding lectin, and RTA, an enzymatic N-glycosidase. The toxin is endocytosed, and after intracellular routing, RTA is translocated to the cytoplasm where it inactivates ribosomes resulting in a loss of host cell protein synthesis and cell death. We show for the first time that the cytotoxicity against cultured T cells by several RTA mutants is directly proportional to the enzyme activity of RTA, suggesting this is a reliable system to measure translocation effects. Large discrepancies between cytotoxicity and enzyme action for a given pair of toxins are therefore attributable to differences in cell binding, uptake, or membrane translocation. Fluid phase uptake and cytotoxicity of isolated RTA are essentially identical to that of the single chain toxin PAP. This important finding suggests that RTA, and the A chain of class 2 RIPs in general, has not evolved special translocation signals to complement the increased target cell binding facilitated by RTB. Experiments with the lectin RCA and with ebulin suggest those toxins have diminished cytotoxicity probably mediated by comparative deficiencies in B chain binding. Addition of a KDEL sequence to RTA increases fluid phase uptake, consistent with the notion that transport to the ER is important for cytotoxicity. Fusion of MBP or GST to the amino terminus of RTA has little effect on enzyme action or cytotoxicity. This result is not altered by protease inhibitors, suggesting the fusion proteins are probably not cleaved prior to translocation of the toxic A chain and implying that the toxins can carry large passenger proteins into the cytoplasm, an observation with interesting potential for analytical and therapeutic chemistry.     

Cell surface density of p185(c-erbB-2) determines susceptibility to anti-p185(c-erbB-2)-ricin A chain (RTA) immunotoxin therapy alone and in combination with anti-p170(EGFR)-RTA in ovarian cancer cells

Approximately 30% of ovarian and breast cancers overexpress p185(c-erbB-2) with as many as 10(6) receptors/cell. Normal cells have as few as 10(4) receptors/cell. We have examined the susceptibility of SKOv3 human ovarian cancer cells to anti-c-erbB2 antibodies and immunotoxins as a function of c-erbB-2 density on the cell surface. A panel of SKOv3 clones that expressed different densities of p185(c-erbB-2) receptor were generated through transfection with the c-erbB-2 gene. A significant correlation was found between p185(c-erbB-2) density and susceptibility to killing by anti-p185(c-erbB-2)-ricin A chain (anti-p185(c-erbB-2)-RTA) immunotoxins. With 10(5) copies/cell of p185(c-erbB-2), <10% of clonogenic ovarian cancer cells could be eliminated, whereas in clones that expressed 10(6) copies/cell of p185(c-erbB-2), 99.9% of clonogenic tumor cells were killed. In cell lines that overexpressed p185(c-erbB-2) and also expressed p170(EGFR), anti-p185(cerbB-2)-RTA and anti-p170(EGFR)-RTA immunotoxins exerted synergistic cytotoxicity. Treatment with the two immunotoxins could eliminate 99.99% of clonogenic cells. Importantly, tumor cells that had survived first treatment with anti-p185(c-erbB2)-RTA alone still retained sensitivity to repeat treatment with the same immunotoxin and also proved susceptible to the synergistic cytotoxicity of anti-p185(cerbB-2)-RTA in combination with anti-p170(EGFR)-RTA. Growth characteristics of the clones expressing various levels of p185(c-erbB-2) were also studied. No correlation was found between p185(c-erbB-2) expression levels and the rate of anchorage-dependent growth, anchorage-independent growth, or in vivo growth in nude mice.     

Bacterial lipopolysaccharide disrupts endothelial monolayer integrity and survival signaling events through caspase cleavage of adherens junction proteins

Bacterial lipopolysaccharide or endotoxin induces actin reorganization, increased paracellular permeability, and endothelial cell detachment from the underlying extracellular matrix in vitro. We studied the effect of endotoxin on transendothelial albumin flux and detachment of endothelial cells cultured on gelatin-impregnated filters. The endotoxin-induced changes in endothelial barrier function and detachment occurred at doses and times that were compatible with endotoxin-induced apoptosis. Since the actin cytoskeleton and cell-cell and cell-matrix adhesion all participate in the regulation of the paracellular pathway and cell-matrix interactions, we studied whether protein components of the actin-linked adherens junctions were modified in response to endotoxin. Components of cell-cell (beta- and gamma-catenin) and cell-matrix (focal adhesion kinase and p130(Cas)) adherens junctions were cleaved by caspases activated during apoptosis with dose and time requirements that paralleled those seen for barrier dysfunction and detachment. Cleavage of focal adhesion kinase led to its dissociation from the focal adhesion-associated signaling protein, paxillin, resulting in reduced paxillin tyrosine phosphorylation. Inhibition of caspase-mediated cleavage of these proteins protected against detachment but not opening of the paracellular pathway. Therefore, endotoxin-induced disruption of endothelial monolayer integrity and survival signaling events is mediated, in part, through caspase cleavage of adherens junction proteins.     

Expression of human aldehyde dehydrogenase-3 associated with hepatocellular carcinoma: promoter regions and nuclear protein factors related to the expression

Studies in our laboratory have shown that as early as day 8.5 of development, mouse yolk sac cells can generate T cells when placed in a thymic microenvironment. At this stage, yolk sac cells can also differentiate into myeloid cells in vitro. B cell differentiation in vitro was achieved with day 9 yolk sac by providing a bone marrow stromal feeder layer. We have now established endothelial cell lines and clones from yolk sacs of day 8-12 mouse embryos. These vary in their ability to support stem cell maintenance and differentiation. Our principal work has been carried out with day 12 cloned endothelial cell lines. One clone supported the > 100 fold expansion of yolk sac hematopoietic stem cells that subsequently could generate B cells, T cells and myeloid cells both in vitro and in vivo. Preliminary experiments with endothelial cells from younger embryos are also described.     

Expression and subcellular distribution of basic fibroblast growth factor are regulated during migration of endothelial cells

Migration of endothelial cells is involved in normal and pathological angiogenesis and in re-endothelialization after vascular injury or rupture of atherosclerotic plaques. Several types of endothelial cells are known to synthesize basic fibroblast growth factor (bFGF); in some of these, migration is increased by exogenous bFGF and inhibited by anti-bFGF antibodies. Using immunocytochemical techniques and RNase protection analysis, we studied endothelial cells from bovine coronary arteries and veins as well as from adrenal microvessels. We found that bFGF mRNA and peptide were present in confluent endothelial cells and were upregulated during migration stimulated by removal of some cells from the monolayer. During migration, extracellular matrix stores of bFGF were depleted, and bFGF immunoreactivity began to accumulate in the cytoplasm of endothelial cells between 2 and 6 hours. After migration had begun, but before the initiation of DNA synthesis, bFGF immunoreactivity increased in the nuclei and nucleoli. Exogenous bFGF stimulated endothelial migration, and antibodies to bFGF markedly inhibited migration, suggesting that an intracrine function of nuclear bFGF is not sufficient for cell migration. In all three types of endothelial cells studied, bFGF was identified as an endogenous regulator, but not as the sole regulator, or migration. Moreover, bFGF expression and subcellular localization were found to be regulated during endothelial cell migration.