Notch signaling: cell fate control and signal integration in development.

Notch signaling defines an evolutionarily ancient cell interaction mechanism, which plays a fundamental role in metazoan development. Signals exchanged between neighboring cells through the Notch receptor can amplify and consolidate molecular differences, which eventually dictate cell fates. Thus, Notch signals control how cells respond to intrinsic or extrinsic developmental cues that are necessary to unfold specific developmental programs. Notch activity affects the implementation of differentiation, proliferation, and apoptotic programs, providing a general developmental tool to influence organ formation and morphogenesis.

Processing of the notch ligand delta by the metalloprotease Kuzbanian.

Signaling by the Notch surface receptor controls cell fate determination in a broad spectrum of tissues. This signaling is triggered by the interaction of the Notch protein with what, so far, have been thought to be transmembrane ligands expressed on adjacent cells. Here biochemical and genetic analyses show that the ligand Delta is cleaved on the surface, releasing an extracellular fragment capable of binding to Notch and acting as an agonist of Notch activity. The ADAM disintegrin metalloprotease Kuzbanian is required for this processing event. These observations raise the possibility that Notch signaling in vivo is modulated by soluble forms of the Notch ligands.

Calcium depletion dissociates and activates heterodimeric notch receptors.

Notch receptors participate in a highly conserved signaling pathway that regulates morphogenesis in multicellular animals. Maturation of Notch receptors requires the proteolytic cleavage of a single precursor polypeptide to produce a heterodimer composed of a ligand-binding extracellular domain (N(EC)) and a single-pass transmembrane signaling domain (N(TM)). Notch signaling has been correlated with additional ligand-induced proteolytic cleavages, as well as with nuclear translocation of the intracellular portion of N(TM) (N(ICD)). In the current work, we show that the N(EC) and N(TM) subunits of Drosophila Notch and human Notch1 (hN1) interact noncovalently. N(EC)-N(TM) interaction was disrupted by 0.1% sodium dodecyl sulfate or divalent cation chelators such as EDTA, and stabilized by millimolar Ca(2+). Deletion of the Ca(2+)-binding Lin12-Notch (LN) repeats from the N(EC) subunit resulted in spontaneous shedding of N(EC) into conditioned medium, implying that the LN repeats are important in maintaining the interaction of N(EC) and N(TM). The functional consequences of EDTA-induced N(EC) dissociation were studied by using hN1-expressing NIH 3T3 cells. Treatment of these cells for 10 to 15 min with 0.5 to 10 mM EDTA resulted in the rapid shedding of N(EC), the transient appearance of a polypeptide of the expected size of N(ICD), increased intranuclear anti-Notch1 staining, and the transient activation of an Notch-sensitive reporter gene. EDTA treatment of HeLa cells expressing endogenous Notch1 also stimulated reporter gene activity to a degree equivalent to that resulting from exposure of the cells to the ligand Delta1. These findings indicate that receptor activation can occur as a consequence of N(EC) dissociation, which relieves inhibition of the intrinsically active N(TM) subunit.

Calcium binding to tandem repeats of EGF-like modules. Expression and characterization of the EGF-like modules of human Notch-1 implicated in receptor-ligand interactions.

The Ca(2+)-binding epidermal growth factor (cbEGF)-like module is a structural component of numerous diverse proteins and occurs almost exclusively within repeated motifs. Notch-1, a fundamental receptor for cell fate decisions, contains 36 extracellular EGF modules in tandem, of which 21 are potentially Ca(2+)-binding. We report the Ca(2+)-binding properties of EGF11-12 and EGF10-13 from human Notch-1 (hNEGF11-12 and hNEGF10-13), modules previously shown to support Ca(2+)-dependent interactions with the ligands Delta and Serrate. Ca2+ titrations in the presence of chromophoric chelators, 5,5'-Br2BAPTA and 5-NBAPTA, gave two binding constants for hNEGF11-12, Kd1 = 3.4 x 10(-5) M and Kd2 > 2.5 x 10(-4) M. The high-affinity site was found to be localized to hNEGF12. Titration of hNEGF10-13 gave three binding constants, Kd1 = 3.1 x 10(-6) M, Kd2 = 1.6 x 10(-4) M, and Kd3 > 2.5 x 10(-4) M, demonstrating that assembly of EGF modules in tandem can increase Ca2+ affinity. The highest affinity sites in hNEGF11-12 and hNEGF10-13 had 10 to 100-fold higher affinity than reported for EGF32-33 and EGF25-31, respectively, from fibrillin-1, a connective tissue protein with 43 cbEGF modules. A model of hNEGF11-12 based on fibrillin-1 EGF32-33 demonstrates electronegative potential that could contribute to the higher affinity of the Ca(2+)-binding site in hNEGF12. These data demonstrate that the Ca2+ affinity of cbEGF repeats can be highly variable among different classes of cbEGF containing proteins.

The effects of methylmercury on Notch signaling during embryonic neural development in Drosophila melanogaster.

Methylmercury (MeHg) is a ubiquitous toxicant that targets the developing fetal nervous system. MeHg interacts with the Notch signaling pathway, a highly-conserved intercellular signaling mechanism required for normal development. Notch signaling is conveyed by activation of the genes in the enhancer of split (E(spl)) locus in Drosophila. We have previously shown that acute high doses of MeHg upregulate several E(spl) genes in Drosophila neural-derived C6 cells. Furthermore, MeHg induction of E(spl) can occur independent of the Notch receptor itself. We now show that MeHg, unlike inorganic mercury (HgCl2), preferentially upregulates E(spl)mδ and E(spl)mγ in Drosophila C6 cells. This is distinct from Delta ligand-induced Notch signaling in which no induction of E(spl)mδ is seen. MeHg is also seen to specifically upregulate E(spl)mδ in Drosophila embryos where HgCl2 showed no such effect. Additionally, treatment of embryos with MeHg caused a consistent failure in axonal outgrowth of the intersegmental nerve (ISN). This ISN phenotype was partially replicated by genetic activation of the Notch pathway, but was not replicated by increasing expression of E(spl)mδ. These data suggest a role for Notch signaling and the E(spl)mδ target gene in MeHg toxicity, however, the site of action for E(spl)mδ in this system remains to be elucidated.

Kuz and TACE can activate Notch independent of ligand.

A central mechanism in activation of the Notch signaling pathway is cleavage of the Notch receptor by ADAM metalloproteases. ADAMs also cleave Delta, the ligand for Notch, thereby downregulating Notch signals. Two ADAMs, Kuzbanian (Kuz) and TNF-alpha converting enzyme (TACE), are known to process both Delta and Notch, yet the role of these cleavages in signal propagation has remained controversial. Using an in vitro model, we show that Kuz regulates Notch signaling primarily by activating the receptor and has little overall effect on signaling via disabling Delta. We confirm that Kuz-dependent activation of Notch requires stimulation of Notch by Delta. However, over-expression of Kuz gives ligand-independent Notch activation. In contrast, TACE, which is elevated in expression in the developing Drosophila nervous system, can efficiently activate Notch in a ligand-independent manner. Altogether, these data demonstrate the potential for Kuz and TACE to participate in context- and mechanism-specific modes of Notch activation.

The mechanism of inactivation of human factor V and human factor Va by activated protein C.

The cleavage of human factor V and human factor Va by human activated protein C (APC) was analyzed in the absence and presence of phospholipid vesicles containing 75% phosphatidylcholine (PC) and 25% phosphatidylserine (PS). Membrane-bound human factor V (250 nM) is cleaved by APC (2.5 nM) to give M(r) = 200,000, 70,000, 45,000, and 30,000 fragments and an M(r) = 22/20,000 doublet. These fragments are released after four sequential cleavages of the membrane-bound procofactor at Arg306, Arg506, Arg679, and Lys994. No cofactor activity is observed following thrombin treatment of the membrane-bound APC-cleaved procofactor. In the absence of a membrane surface, no cleavage of factor V by APC is observed, and following thrombin activation factor Va retains full cofactor activity. Membrane-bound human factor Va (600 nM) loses more than 90% of its initial cofactor activity after 10 min of incubation with APC (10.9 nM), and virtually no cofactor activity is observed after 1 h of incubation. Under similar conditions but in the absence of PCPS vesicles, factor Va is cleaved but retains approximately 80% of its initial cofactor activity after 2 h of incubation with APC. In the presence of PCPS vesicles, the APC related loss of activity is correlated with cleavage of the heavy chain and appearance of fragments of M(r) = 45,000, 30,000, and of 28/26,000, and 22/20,000 doublets. These products correspond to three cleavages of the heavy chain (at Arg306, Arg506, and Arg679). Cleavage at Arg506 of factor Va precedes and appears to be required for cleavage at Arg306 and Arg679. In the absence of membrane, proteolysis at Arg506 produces an M(r) = 75,000 fragment which corresponds to the NH2-terminal portion of the human factor Va heavy chain (residues 1-506), and a carboxyl-terminal doublet of M(r) = 28/26,000 (residues 507-709) which is cleaved by APC at Arg679 to generate an M(r) = 22/20,000 doublet and an M(r) = 6,000 peptide. No cleavage of the light chain of the human cofactor is observed in the presence or absence of PCPS vesicles following 2 h of incubation with APC. Our data demonstrate that inactivation of human factor V and human factor Va only occurs in the presence of a membrane surface after cleavage at Arg306. However, while this cleavage site is exposed on membrane-bound human factor V, cleavage at Arg506 on the heavy chain of factor Va appears necessary for complete exposure of the cleavage site at Arg306.

Platelet coagulation factor Va: the major secretory platelet phosphoprotein.

Platelet-derived coagulation factor Va is the primary secreted substrate for a thrombin-stimulation-dependent platelet kinase. Human platelet factor Va, consisting of a molecular weight (M(r)) 105,000 heavy chain and an M(r) 74,000 light chain, incorporates phosphate in at least two sites on the light chain. Phosphorylated factor Va represents 50% of the secreted protein-associated phosphate. This modification occurs exclusively at serine residues and is inhibited by H-7 and staurosporine, which suggests a protein kinase C (PKC)-mediated event. Purified plasma factor V and Va are phosphorylated in the light chain region by rat brain PKC. The activity of platelet factor Va in prothrombinase on platelets is not altered when phosphorylation is inhibited by staurosporine. Plasma-derived factor Va in the presence of thrombin stimulated platelets is phosphorylated on both the heavy chain and the light chain. Plasma factor V and factor Va heavy chain phosphorylation occurs without light chain phosphorylation in the presence of added 32P gamma-ATP and non-stimulated or collagen-stimulated platelets or casein kinase II. This differential phosphorylation of factor Va heavy and light chain shows two independent platelet kinase activities that act on factor Va. The heavy chain factor V/Va kinase activity is similar to casein kinase II, which we have demonstrated previously to act on factor Va and accelerate activated protein C inactivation of the cofactor. Our data show platelet-dependent phosphorylation of platelet and plasma factor V and Va resulting in significant covalent modifications of the cofactor. These modifications may play a role in directing the extracellular distribution of factor V and factor Va.

Factor Va-membrane interaction is mediated by two regions located on the light chain of the cofactor.

Factor Va was incubated with 1-azidopyrene, a fluorescent lipophilic probe, in the presence of phospholipid vesicles composed of various proportions of phosphatidylcholine (PC) and phosphatidylserine (PS). The majority of the label was associated with the light chain of factor Va. The light chain was found to be labeled in the presence of phospholipid vesicles containing either 100% PC or 100% PS. After cleavage by factor Xa and incubation with PC/PS vesicles composed of 75% PC and 25% PS, label was found both on the M(r) = 30,000 fragment, derived from the NH2-terminal portion of the bovine factor Va light chain (residues 1537-1752), and on the M(r) = 46,000/48,000 carboxyl-terminal fragment of the factor Va light chain (residues 1753-2183). The M(r) = 46,000/48,000 fragment incorporated 1-azidopyrene independent of the phospholipid composition, while label incorporation into the M(r) = 30,000 fragment required phospholipid vesicles containing PC. No labeling of the M(r) = 30,000 fragment was observed with phospholipid vesicles composed of 100% PS. The label incorporation into the two portions of the molecule was found to be independent of the ionic strength in the presence of phospholipid vesicles containing 75% PC and 25% PS. In contrast, the labeling of the M(r) = 46,000/48,000 fragment with phospholipid vesicles composed of 100% PS was ionic strength dependent.(ABSTRACT TRUNCATED AT 250 WORDS)

Factor V turnover in a primate model.

The isolation and characterization of baboon plasma factor V (FV) were performed for the development of an in vivo model for studying factor V/Va physiology in nonhuman primates. Baboon FV was purified by immunoaffinity chromatography with an antihuman FV monoclonal antibody and exhibits a specific activity of 1,940 U/mg. Baboon FV activation by thrombin proceeds through two proteolytic pathways similar to those observed with human and bovine FV. Limited amino acid sequencing of FV and its thrombin activation fragments shows 95% identity with human and 79% identity with bovine FV. 125I-Factor V and a mixture of thrombin cleaved 125I-FV activation products were infused into normal male baboons and evaluated by blood sample radioactivity measurements and by autoradiography of plasma samples following resolution by gel electrophoresis. Factor V disappeared with a half-life (t1/2) of 12.98 +/- 1.85 hours and was cleared without obvious degradation of the molecule during circulation. The radioactivity associated with the thrombin activated FV mixture, which consisted of the Mr = 220,000 activation intermediate, the Mr = 150,000 activation peptide, the heavy chain (HC) and the light chain (LC) of FVa, was cleared in a nonlinear manner. The HC and LC were removed with t1/2 < 20 minutes. The apparent molecular weight (Mr) = 220,000 and Mr = 150,000 fragments were cleared with t1/2 > 6 hours and t1/2 > 30 hours, respectively.

Characterization of the molecular defect in factor VR506Q.

A poor anticoagulant response of plasma to activated protein C is correlated with a single mutation in the factor V molecule (Arg506-->Gln). Factor V was purified to homogeneity from plasma of two unrelated patients (patient I, factor VI, and patient II, factor VII), who are homozygous for this mutation. The factor V molecule from both patients has normal procoagulant activity when compared with factor V isolated from normal plasma in both a clotting time-based assay and in an assay measuring alpha-thrombin formation. The cleavage and subsequent inactivation by activated protein C (APC) of the alpha-thrombin-activated membrane-bound cofactor (factor Va) from both patients were analyzed and compared with the cleavage and inactivation of normal human factor Va. In normal factor Va, cleavage at Arg506 generates a M(r) = 75,000 fragment and a M(r) = 28,000/26,000 doublet and is necessary for the optimum exposure of the sites for subsequent cleavage at Arg306 and Arg679. Proteolysis at these sites leads to the appearance of M(r) - 45,000 and 30,000 fragments and a M(r) = 22,000/20,000 doublet. Cleavage at Arg306 is membrane-dependent and is required for complete inactivation. Following 5 min of incubation with APC (5.4 nM) membrane-bound normal factor Va (280 nM) has virtually no cofactor activity whereas under similar experimental conditions factor VaI and factor VaII retain approximately 50% of their initial activity. After 1 h of incubation with APC, factor VaI retains 20% of its initial cofactor activity whereas factor VaII has 10% remaining cofactor activity. The initial loss in cofactor activity (approximately 70%) of membrane-bound factor VaI and factor VaII during the first 10 min of the inactivation reaction is correlated with cleavage at Arg306 and appearance of a M(r) = 45,000 fragment and a M(r) = 62,000/60,000 doublet. Subsequently, the M(r) = 62,000/60,000 doublet is cleaved at Arg679 to generate a M(r) = 56,000/54,000 doublet resulting in complete loss of cofactor activity. Both procofactors, factor VI and factor VII, were inactivated following cleavage at Arg306 and Arg679, with APC inactivation rates equivalent to those observed for normal factor V. Our data demonstrate that: 1) cleavage at Arg506 is required for optimum exposure of the cleavage sites at Arg306 and Arg679 and rapid inactivation of membrane-bound factor Va; and 2) cleavage at Arg306 by APC on membrane-bound factor V occurs at the same rate in both normal and APC-resistant individuals. Thus cleavage at Arg306 and Arg679 and subsequent inactivation of the membrane-bound procofactor, factor V, does not require prior cleavage at Arg506 for optimum exposure.

Phosphorylation of factor Va and factor VIIIa by activated platelets.

Platelet activation leads to the incorporation of 32[PO4(2-)] into bovine coagulation factor Va and recombinant human factor VIII. In the presence of the soluble fraction from thrombin-activated platelets and (gamma-32P) adenosine triphosphate, radioactivity is incorporated exclusively into the M(r) = 94,000 heavy chain (H94) of factor Va and into the M(r) = 210,000 to 90,000 heavy chains as well into the M(r) = 80,000 light chain of factor VIII. Proteolysis of the purified phosphorylated M(r) = 94,000 factor Va heavy chain by activated protein C (APC) gave products of M(r) = 70,000, 24,000, and 20,000. Only the intermediate M(r) = 24,000 fragment contained radioactivity. Because the difference between the M(r) = 24,000 and M(r) = 20,000 fragments is located on the COOH-terminal end of the bovine heavy chain, phosphorylation of H94 must occur within the M(r) = 4,000 peptide derived from the carboxyl-terminal end of H94 (residues 663 through 713). Exposure of the radioactive factor VIII molecule to thrombin ultimately resulted in a nonradioactive light chain and an M(r) = 24,000 radioactive fragment that corresponds to the carboxyl-terminal segment of the A1 domain of factor VIII. Based on the known sequence of human factor VIII, phosphorylation of factor VIII by the platelet kinase probably occurs within the acidic regions 337 through 372 and 1649 through 1689 of the procofactor. These acidic regions are highly homologous to sequences known to be phosphorylated by casein kinase II. Results obtained using purified casein kinase II gave a maximum observed stoichiometry of 0.6 mol of 32[PO4(2-)]/mol of factor Va heavy chain and 0.35 mol of 32[PO4(2-)]/mol of factor VIII. Phosphoamino acid analysis of phosphorylated factor Va by casein kinase II or by the platelet kinase showed only the presence of phosphoserine while phosphoamino acid analysis of phosphorylated factor VIII by casein kinase II showed the presence of phosphothreonine as well as small amounts of phosphoserine. The platelet kinase responsible for the phosphorylation of the two cofactors was found to be inhibited by several synthetic protein kinase inhibitors. Finally, partially phosphorylated factor Va was found to be more sensitive to APC inactivation than its native counterpart. Our findings suggest that phosphorylation of factors Va and VIIIa by a platelet casein kinase II-like kinase may downregulate the activity of the two cofactors.

Blood clotting in minimally altered whole blood.

The sequences of events regulating thrombin generation during tissue factor-initiated clotting in whole blood at 37 degrees C in which the contact pathway was suppressed with corn trypsin inhibitor are studied using quantitative Western blotting of factor V, prothrombin, platelet factor 4, antithrombin III, and fibrinogen. In addition, fibrinopeptide A (FPA), thrombin-antithrombin III (TAT) complex formation, and prothrombin fragment 1.2 (F1.2) were measured via commercially available enzyme-linked immunosorbent assays (ELISAs). In a typical experiment initiated with 40 pmol/L recombinant tissue factor, visual clot time (4.5 minutes), was preceded by significant cleavage of factor V resulting in 65% factor Va heavy-chain generation but only 10% light-chain formation. At this point, 50% of the platelet factor 4 is released, suggesting that half (approximately 700 pmol/L) of the platelet prothrombinase sites available have been generated. At clot time, approximately 15 nmol/L thrombin B-chain is present; however, analyses of FPA release demonstrate that only 15% of the thrombin is acting on fibrinogen. This thrombin is produced by the action of 7 pmol/L prothrombinase. The maximum rate of thrombin production is reached well after clot time and is consistent with the presence of approximately 150 pmol/L prothrombinase (at about 7 minutes). These results suggest that factor Xa is the limiting factor for thrombin generation. After 60 minutes, 75% of the initial prothrombin (1.24 mumol/L) is consumed yielding 400 nmol/ L prethrombin 2 and 360 nmol/l thrombin (B-chain) products. The sum of these values (800 nmol/L) is similar to the (corrected) F1.2 concentration determined by ELISA. The incomplete cleavage of prothrombin indicates both the prothrombinase complex and the formation of prothrombinase are inhibited in the reaction. TAT complex measured by ELISA is almost equivalent to B-chain concentration, but sodium dodecyl sulfate stable thrombin-antithrombin III complexes are not observed until well after clot formation and are never equivalent to ELISA-TAT values. At the point of clot formation, 80% of the fibrinogen is depleted from the fluid phase, whereas only 35% to 45% of the FPA is released, suggesting a significant incorporation of uncleaved fibrinogen into the initial clot formed.

Factor VNew Brunswick: Ala221-to-Val substitution results in reduced cofactor activity.

We have characterized the factor V protein and cDNA of a patient displaying factor V deficiency (parahemophilia) and correlated the reduced activity with a missense mutation of Ala221-to-Val. Plasma from the subject individual (C1) presented reduced factor V antigen (39% of normal) that displayed reduced activity (approximately 26% of normal). Factor V purified from this individual by standard techniques shows normal migration on sodium dodecyl sulfate gels and a normal pattern of activation by thrombin. Purified antigen from sibling C2 gives a much reduced specific activity of 263 U/mg (17% of normal). Sibling C3, the mother, and the father have antigen within the normal range (57% to 200%) that has approximately normal specific activity. The cDNA encoding the factor Va heavy and light chains of the subject individual was polymerase chain reaction-amplified and sequenced and revealed an A-to-G substitution at position 3 of codon 51 (silent mutation), a C-to-T substitution in position 2 of codon 221 (Ala221-Val), a T-to-C substitution at position 3 of codon 708 (silent mutation), and a G-to-A substitution at position 1 of codon 2185 (Thr2185-Ala). The latter mutation was also observed in control individuals and is proposed to be a possible polymorphism. Restriction analyses demonstrated the presence of one mutant and one normal allele in the father. The subject individual (C1) and sibling C2 carry only the mutant allele. The mother and sibling C3 carry only the normal allele. The inheritance pattern suggests the presence of a missing or nonexpressed allele in the mother that is passed on to all the siblings. Expression of only the mutant allele by the subject individual (C1) and sibling C2 is consistent with reduced factor V antigen and activity in these patients. We have designated this mutant as Factor VNew Brunswick.