Inhibition of DNA2 nuclease as a therapeutic strategy targeting replication stress in cancer cells

Replication stress is a characteristic feature of cancer cells, which is resulted from sustained proliferative signaling induced by activation of oncogenes or loss of tumor suppressors. In cancer cells, oncogene-induced replication stress manifests as replication-associated lesions, predominantly double-strand DNA breaks (DSBs). An essential mechanism utilized by cells to repair replication-associated DSBs is homologous recombination (HR). In order to overcome replication stress and survive, cancer cells often require enhanced HR repair capacity. Therefore, the key link between HR repair and cellular tolerance to replication-associated DSBs provides us with a mechanistic rationale for exploiting synthetic lethality between HR repair inhibition and replication stress. DNA2 nuclease is an evolutionarily conserved essential enzyme in replication and HR repair. Here we demonstrate that DNA2 is overexpressed in pancreatic cancers, one of the deadliest and more aggressive forms of human cancers, where mutations in the KRAS are present in 90–95% of cases. In addition, depletion of DNA2 significantly reduces pancreatic cancer cell survival and xenograft tumor growth, suggesting the therapeutic potential of DNA2 inhibition. Finally, we develop a robust high-throughput biochemistry assay to screen for inhibitors of the DNA2 nuclease activity. The top inhibitors were shown to be efficacious against both yeast Dna2 and human DNA2. Treatment of cancer cells with DNA2 inhibitors recapitulates phenotypes observed upon DNA2 depletion, including decreased DNA double strand break end resection and attenuation of HR repair. Similar to genetic ablation of DNA2, chemical inhibition of DNA2 selectively attenuates the growth of various cancer cells with oncogene-induced replication stress. Taken together, our findings open a new avenue to develop a new class of anticancer drugs by targeting druggable nuclease DNA2. We propose DNA2 inhibition as new strategy in cancer therapy by targeting replication stress, a molecular property of cancer cells that is acquired as a result of oncogene activation instead of targeting currently undruggable oncoprotein itself such as KRAS.

Role for nitric oxide in the hyperpermeability and hemodynamic changes induced by intravenous VEGF

PURPOSE

To explore the effects of brief intravenous (IV) infusion of vascular endothelial growth factor (VEGF) on vascular albumin permeability, blood flow, and vascular conductance (blood flow normalized to arterial blood pressure) in ocular tissues and brain and to assess the role of nitric oxide in mediating these changes.

METHODS

A quantitative, double-tracer, radiolabeled albumin permeation method was combined with radiolabeled microspheres for assessment of changes in vascular permeability and blood flow, respectively, induced in ocular tissues by IV infusion of recombinant human VEGF165 for 20 minutes (80-450 picomoles/kg body weight). An inhibitor of nitric oxide synthase (NOS), NG-monomethyl-L-arginine (L-NMMA; 50 micromoles/kg body weight infused simultaneously with VEGF), was used to explore the role of nitric oxide in mediating the vascular changes induced by VEGF.

RESULTS

Infusion of VEGF165 in thiopental-anesthetized rats dose-dependently increased 125I albumin permeation in the retina, anterior uvea, and choroid/sclera and in brain, aorta, lung, kidney, small intestine, and peripheral nerve. Mean arterial blood pressure, cardiac output, and stroke volume were decreased only at the highest dose of VEGF, whereas heart rate remained unchanged. Blood flow was increased in the anterior uvea, and vascular conductance was increased in retina, anterior uvea, choroid/sclera, and brain at the highest dose of VEGF. The NOS inhibitor, L-NMMA, blocked VEGF-induced vascular hyperpermeability in all ocular and nonocular tissues, prevented the increase in vascular conductance in all ocular tissues, and blocked the decrease in mean arterial blood pressure, cardiac output, and stroke volume. Infusion of L-NMMA alone decreased vascular conductance in choroid/sclera and kidney, slightly increased mean arterial blood pressure, and in general, did not affect 125I-albumin permeation. (L-NMMA slightly decreased albumin permeation in the retina and increased it in the brain.)

CONCLUSIONS

Intravenous infusion of VEGF can acutely impair endothelial cell barrier functional integrity and relax resistance arterioles in ocular tissues and brain through a mechanism involving activation of NOS.

Development of novel monoclonal antibodies for the analysis of functional sites in FGF-2

Fibroblast growth factor 2 (FGF-2) can function as a potent mitogen, as well as a survival factor for a variety of mammalian cell types. The biological effects of FGF-2 are mediated by its interaction with two types of cellular binding sites: (1) high affinity tyrosine kinase receptors; and (2) low affinity heparan sulfate proteoglycans (HSPGs) on the cell surface. Although numerous FGF-2 antibodies have been used previously to analyze its biological actions, few studies have utilized antibodies to analyze domains within FGF-2 involved in its interactions with the two binding sites. In this report, we describe the generation and use of two monoclonal antibodies against human recombinant FGF-2 (254F1 and 256A12) that inhibit FGF-2 function. However, these antibodies appear to target preferentially different domains within the FGF-2 molecule, and therefore differentially influence the interactions of FGF-2 with its low and high affinity receptors. 254F1 is a more effective inhibitor of the high affinity, receptor tyrosine kinase binding site, whereas 256A12 appears to be a better inhibitor of the low affinity, HSPG interactions. We also demonstrate that the two antibodies are potent inhibitors of FGF-2 stimulated vascular cell proliferation, and as such have potential use in the treatment of vascular hyperproliferative diseases.

Role for heparin-binding growth factors in glucose-induced vascular dysfunction

Vascular hyperpermeability and excessive neovascularization are hallmarks of early and late vascular endothelial cell dysfunction induced by diabetes. Vascular endothelial growth factor (VEGF) appears to be an important mediator for these early and late vascular changes. We reported previously, using skin chambers mounted on backs of SD rats, that neutralizing antibodies directed against VEGF blocked vascular permeability and blood flow changes induced by elevated tissue glucose and sorbitol levels in a dosage-dependent manner. We report in this study, using the same skin chamber model and neutralizing antibodies directed against basic fibroblast growth factor (FGF-2), that another member of the heparin-binding growth factor family also mediates glucose- and sorbitol-induced vascular permeability and blood flow increases. In addition, we show that 1) TBC1635, a novel heparin-binding growth factor antagonist, blocks the vascular hyperpermeability and blood flow increases induced by elevated tissue levels of glucose and sorbitol and by topical application of VEGF and FGF-2 to granulation tissue in skin chambers, and 2) suramin, a commercially available growth factor antagonist, blocks glucose-induced vascular dysfunction. These results suggest an early role for heparin-binding growth factors in the vascular dysfunction caused by excessive glucose metabolism, possibly via the sorbitol pathway.

Vascular dysfunction induced by elevated glucose levels in rats is mediated by vascular endothelial growth factor

The purpose of these experiments was to investigate a potential role for vascular endothelial growth factor (VEGF) in mediating vascular dysfunction induced by increased glucose flux via the sorbitol pathway. Skin chambers were mounted on the backs of Sprague-Dawley rats and 1 wk later, granulation tissue in the chamber was exposed twice daily for 7 d to 5 mM glucose, 30 mM glucose, or 1 mM sorbitol in the presence and absence of neutralizing VEGF antibodies. Albumin permeation and blood flow were increased two- to three-fold by 30 mM glucose and 1 mM sorbitol; these increases were prevented by coadministration of neutralizing VEGF antibodies. Blood flow and albumin permeation were increased approximately 2.5-fold 1 h after topical application of recombinant human VEGF and these effects were prevented by nitric oxide synthase (NOS) inhibitors (aminoguanidine and N(G)-monomethyl L-arginine). Topical application of a superoxide generating system increased albumin permeation and blood flow and these changes were markedly attenuated by VEGF antibody and NOS inhibitors. Application of sodium nitroprusside for 7 d or the single application of a calcium ionophore, A23187, mimicked effects of glucose, sorbitol, and VEGF on vascular dysfunction and the ionophore effect was prevented by coadministration of aminoguanidine. These observations suggest a potentially important role for VEGF in mediating vascular dysfunction induced by "hypoxia-like" cytosolic metabolic imbalances (reductive stress, increased superoxide, and nitric oxide production) linked to increased flux of glucose via the sorbitol pathway.

Identification of the extracellular domains of Flt-1 that mediate ligand interactions

Vascular Endothelial Growth Factor (VEGF) mediates its actions through the Flt-1 and KDR(Flk-1) receptor tyrosine kinases. To localize the extracellular region of Flt-1 that is involved in ligand interactions, we prepared secreted fusion proteins between various combinations of its seven extracellular IgG-like folds. Ligand binding studies show that in combination, domains one and two (amino acids 1-234) are sufficient to achieve VEGF165 interactions. Either domain alone is insufficient to achieve this effect. However, Scatchard analysis reveals that despite the binding capabilities of this construct, the Kd is five fold lower than ligand binding to the full extracellular domain. We find that addition of domain three to this minimal site restores high affinity receptor binding. Further, we show that domains one and two are sufficient to achieve interactions of Flt-1 with Placental Growth Factor (PIGF-1).

Vascular endothelial growth factor, a multifunctional polypeptide

Angiogenesis, the sprouting of new blood vessels from pre-existing vessels, is a complex, multicellular phenomenon involving capillary endothelial cell (EC) proliferation, migration, and tissue infiltration. The elucidation of the biochemical and molecular factors which control angiogenesis is fundamental to our understanding of normal blood vessel development, as well as of the pathogenesis of abnormal blood vessel formation. Angiogenesis is associated with numerous physiological processes, including embryogenesis, wound healing, organ regeneration, and the female reproductive cycle. However, abnormal angiogenesis also plays a major role in the pathogenesis of tumor growth, rheumatoid arthritis, atherosclerosis and various retinopathies. The cellular and molecular mechanisms underlying both physiological and pathophysiological angiogenesis are only now beginning to be understood. Vascular endothelial growth factor was initially discovered as an unidentified tumor-derived factor which increased microvascular permeability (vascular permeability factor, VPF). Subsequently, it was determined that the protein exhibited mitogenic effects on endothelial cells, but not other cell types. Multiple receptor subtypes have been described which may in part explain the multiplicity of biological actions that have been ascribed to VEGF/VPF in the literature. In this overview, we briefly summarize what is currently known about VEGF and VEGF receptor biology, as well as VEGF receptor signal transduction mechanisms in endothelial cells.

Characterization of the subtype of muscarinic receptor coupled to the stimulation of phosphoinositide hydrolysis in 132-1N1 human astrocytoma cells

Stimulation of muscarinic receptors increases phosphoinositide (PI) hydrolysis in 132-1N1 human astrocytoma cells. To evaluate the subtype of receptors which mediate PI hydrolysis in 132-1N1 cells, the effects of: a) the nonselective M1 agonist, carbachol; b) the selective M1 agonist, 4-hydroxy-2-butynyl-trimethylammonium chloride-m-chlorocarbinilate (McN-343); c) the nonselective antagonists, atropine and scopolamine; d) the relatively selective M1 antagonist, pirenzepine; e) the relatively selective M2 antagonists, AF-DX 116 (11-2-diethylaminomethyl-1-piperidinylacetyl-5, 11-dihydro-6H-pyrido-2,3-b-1,4-benzodiazepine-6-one) and methoctramine and f) the relatively selective M3 antagonist, hexahydrosila-difenidol (HHSiD) on PI hydrolysis in 132-1N1 cells were studied. The cell pools of inositol-phospholipids were prelabelled by incubating 132-1N1 cells in a low inositol containing medium (CMRL-1066) supplemented with [3H]inositol (2 microCi/ml) for 20-24 hours at 37 degrees C. The cells were washed and resuspended in a physiological salt solution, and PI hydrolysis was measured by accumulation of [3H]inositol-1-phosphate (IP) in the presence of 10 mM LiCl. Carbachol produced time and concentration dependent PI hydrolysis (EC50, 37 microM). McN-A343 did not cause significant hydrolysis of PI in 132-1N1 cells indicating that the receptor was not of M1 type. All the above muscarinic antagonists caused a concentration dependent decrease in the level of IP in response to carbachol (100 microM). The rank order of their affinities (pA2 values) was: atropine (8.8) > HHSiD (7.6) > pirenzepine (6.8) > methoctramine (6.0) > AF-DX 116 (5.8). This rank order supports the concept that M3 (other names, M2 beta, glandular M2) receptors are linked to PI hydrolysis in 132-1N1 cells. HHSiD, which is selective for M3 receptors of the smooth muscle has higher affinity for muscarinic receptors in 132-1N1 cells than AF-DX 116 which is selective for M2 receptors in cardiac tissue. If the receptor in 132-1N1 cells had been M2, part of the rank order for affinities would have been methoctramine > AF-DX 116 > HHSiD > pirenzepine. From all of these observations, the muscarinic receptor for PI hydrolysis in 132-1N1 cells is tentatively characterized as of M3 type.

Subtype of muscarinic receptor coupled to the attenuation of hormone-stimulated cAMP accumulation in NG108-15 neuroblastoma x glioma hybrid cells

The subtype of muscarinic receptor which mediates cAMP attenuation is not established. Therefore, several selective muscarinic antagonists were used to characterize the subtype of muscarinic receptor coupled to the inhibition of hormone-stimulated cAMP accumulation using NG108-15 neuroblastoma x glioma hybrid cells. These cells were prelabeled with [2-3H]-adenine, washed, and resuspended in a culture medium containing the phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine (0.5 mM). The labeled cells were preincubated with the different antagonists 12-15 min. before they were challenged with agonists. The formation of [3H]-cAMP was activated by PGE1 (1 microM) or forskolin (1 microM). In all cases, [3H]-cAMP formed was separated and measured. Carbachol (100 microM) and McN-A343 (10 mM) were used as standard muscarinic agonists. These studies gave the following results: a) McN-A343 (10 mM), an M1 receptor agonist, was only a partial agonist causing 40% inhibition of cAMP accumulation indicating that this effect was not mediated by an M1 receptor; b) The M1-selective antagonist, pirenzepine, exhibited low affinity (pA2 6.2) further suggesting that an M1 receptor was not coupled to the attenuation of cAMP accumulation; c) Two selective M2 antagonists (AF-DX 116 and methoctramine) and M3 antagonist (HHSiD) were used to further characterize these muscarinic receptors. The order of all antagonists based on their affinities (pA2 values) could be arranged in the following order: atropine (9.0) > methoctramine (7.6) > HHSiD (6.9) > AF-DX 116 (6.6) > pirenzepine (6.2). HHSiD exhibits the same degree of affinity to M2 receptors of other tissues as it does to those of NG cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Isolation of two distinct type I angiotensin II receptor genes

A rat genomic Southern blot, probed with a type I angiotensin II receptor probe, demonstrated that two highly homologous type I angiotensin II receptors were present. A rat genomic library was subsequently screened and four clones were isolated. From restriction mapping, differential hybridization, polymerase chain reaction amplification and sequence analyses we have determined that there are two unique type I angiotensin II receptor genes. The first of these genes corresponds to the published rat vascular complementary DNA sequence; the second, corresponds to a novel receptor not previously described.