The role of distinct p185neu extracellular subdomains for dimerization with the epidermal growth factor (EGF) receptor and EGF-mediated signaling

The extracellular domain of p185c-neu can be viewed as a complex structure of four subdomains, two of which are cysteine-rich subdomains. We have investigated the contribution of these distinct p185c-neu extracellular subdomains to p185/epidermal growth factor receptor (EGFR) heteromer formation and EGF-induced heteromeric signaling. Our studies indicate that at least two separate p185 subdomains, a region spanning subdomains I and II and subdomain IV are involved in association of p185 with the EGFR. We also demonstrated that subdomain IV reduced the heteromeric signaling and transforming activities induced by EGF after associating with EGFR. When 126 aa were deleted from subdomain IV, this small subdomain IV-derived fragment could still lead to heterodimers with EGFR and suppress EGF-induced mitogen-activated protein kinase activation and subsequent transformation abilities. These data provide information about trans-inhibitory mechanisms of mutant p185 species and also indicate that both the entire and a part of subdomain IV may represent a therapeutic target for erbB-overexpressing tumors. Finally, these studies define a basic feature of receptor-receptor associations that are determined by cystine-knot containing subdomains.

Platelets modulate gastric ulcer healing: Role of endostatin and vascular endothelial growth factor release

Bleeding and delayed healing of ulcers are well recognized clinical problems associated with the use of aspirin and other nonsteroidal antiinflammatory drugs, which have been attributed to their antiaggregatory effects on platelets. We hypothesized that antiplatelet drugs might interfere with gastric ulcer healing by suppressing the release of growth factors, such as vascular endothelial growth factor (VEGF), from platelets. Gastric ulcers were induced in rats by serosal application of acetic acid. Daily oral treatment with vehicle, aspirin, or ticlopidine (an ADP receptor antagonist) was started 3 days later and continued for 1 week. Ulcer induction resulted in a significant increase in serum levels of VEGF and a significant decrease in serum levels of endostatin (an antiangiogenic factor). Although both aspirin and ticlopidine markedly suppressed platelet aggregation, only ticlopidine impaired gastric ulcer healing and angiogenesis as well as reversing the ulcer-associated changes in serum levels of VEGF and endostatin. The effects of ticlopidine on ulcer healing and angiogenesis were mimicked by immunodepletion of circulating platelets, and ticlopidine did not influence ulcer healing when given to thrombocytopenic rats. Incubation of human umbilical vein endothelial cells with serum from ticlopidine-treated rats significantly reduced proliferation and increased apoptosis, effects reversed by an antibody directed against endostatin. Ticlopidine treatment resulted in increased platelet endostatin content and release. These results demonstrate a previously unrecognized contribution of platelets to the regulation of gastric ulcer healing. Such effects likely are mediated through the release from platelets of endostatin and possibly VEGF. As shown with ticlopidine, drugs that influence gastric ulcer healing may do so in part through altering the ability of platelets to release growth factors.

Targeted mutation of the murine arylhydrocarbon receptor nuclear translocator 2 (Arnt2) gene reveals partial redundancy with Arnt

The ubiquitously expressed basic helix–loop–helix (bHLH)-PAS protein ARNT (arylhydrocarbon receptor nuclear transporter) forms transcriptionally active heterodimers with a variety of other bHLH-PAS proteins, including HIF-1α (hypoxia-inducible factor-1α) and AHR (arylhydrocarbon receptor). These complexes regulate gene expression in response to hypoxia and xenobiotics, respectively, and mutation of the murine Arnt locus results in embryonic death by day 10.5 associated with placental, vascular, and hematopoietic defects. The closely related protein ARNT2 is highly expressed in the central nervous system and kidney and also forms complexes with HIF-1α and AHR. To assess unique roles for ARNT2 in development, and reveal potential functional overlap with ARNT, we generated a targeted null mutation of the murine Arnt2 locus. Arnt2−/− embryos die perinatally and exhibit impaired hypothalamic development, phenotypes previously observed for a targeted mutation in the murine bHLH-PAS gene Sim1 (Single-minded 1), and consistent with the recent proposal that ARNT2 and SIM1 form an essential heterodimer in vivo [Michaud, J. L., DeRossi, C., May, N. R., Holdener, B. C. & Fan, C. (2000) Mech. Dev. 90, 253–261]. In addition, cultured Arnt2−/− neurons display decreased hypoxic induction of HIF-1 target genes, demonstrating formally that ARNT2/HIF-1α complexes regulate oxygen-responsive genes. Finally, a strong genetic interaction between Arnt and Arnt2 mutations was observed, indicating that either gene can fulfill essential functions in a dose-dependent manner before embryonic day 8.5. These results demonstrate that Arnt and Arnt2 have both unique and overlapping essential functions in embryonic development.

Structural basis for fibroblast growth factor receptor 2 activation in Apert syndrome

Apert syndrome (AS) is characterized by craniosynostosis (premature fusion of cranial sutures) and severe syndactyly of the hands and feet. Two activating mutations, Ser-252 → Trp and Pro-253 → Arg, in fibroblast growth factor receptor 2 (FGFR2) account for nearly all known cases of AS. To elucidate the mechanism by which these substitutions cause AS, we determined the crystal structures of these two FGFR2 mutants in complex with fibroblast growth factor 2 (FGF2) . These structures demonstrate that both mutations introduce additional interactions between FGFR2 and FGF2, thereby augmenting FGFR2–FGF2 affinity. Moreover, based on these structures and sequence alignment of the FGF family, we propose that the Pro-253 → Arg mutation will indiscriminately increase the affinity of FGFR2 toward any FGF. In contrast, the Ser-252 → Trp mutation will selectively enhance the affinity of FGFR2 toward a limited subset of FGFs. These predictions are consistent with previous biochemical data describing the effects of AS mutations on FGF binding. Alterations in FGFR2 ligand affinity and specificity may allow inappropriate autocrine or paracrine activation of FGFR2. Furthermore, the distinct gain-of-function interactions observed in each crystal structure provide a model to explain the phenotypic variability among AS patients.

Evidence for a circulating islet cell growth factor in insulin-resistant states

Insulin resistance is a feature of many common disorders including obesity and type 2 diabetes mellitus. In these disorders, the β-cells compensate for the insulin resistance for long periods of time with an increase in secretory capacity, an increase in β-cell mass, or both. To determine whether the β-cell response might relate to a circulating growth factor, we have transplanted normal islets under the kidney capsule of normoglycemic insulin-resistant mice with two different models of insulin resistance: lean mice that have a double heterozygous deletion of the insulin receptor and insulin receptor substrate-1 (DH) or the obese, hyperglycemic ob/ob mice. In the grafts transplanted into both hosts, there was a marked increase in β-cell mitotic activity and islet mass that was comparable with that observed in the endogenous pancreas. By contrast, islets of the DH mouse transplanted into normal mice showed reduced mitotic index. These data suggest the insulin resistance is associated with a circulating islet cell growth factor that is independent of glucose and obesity.

Identification of urocortin III, an additional member of the corticotropin-releasing factor (CRF) family with high affinity for the CRF2 receptor

The corticotropin-releasing factor (CRF) family of neuropeptides includes the mammalian peptides CRF, urocortin, and urocortin II, as well as piscine urotensin I and frog sauvagine. The mammalian peptides signal through two G protein-coupled receptor types to modulate endocrine, autonomic, and behavioral responses to stress, as well as a range of peripheral (cardiovascular, gastrointestinal, and immune) activities. The three previously known ligands are differentially distributed anatomically and have distinct specificities for the two major receptor types. Here we describe the characterization of an additional CRF-related peptide, urocortin III, in the human and mouse. In searching the public human genome databases we found a partial expressed sequence tagged (EST) clone with significant sequence identity to mammalian and fish urocortin-related peptides. By using primers based on the human EST sequence, a full-length human clone was isolated from genomic DNA that encodes a protein that includes a predicted putative 38-aa peptide structurally related to other known family members. With a human probe, we then cloned the mouse ortholog from a genomic library. Human and mouse urocortin III share 90% identity in the 38-aa putative mature peptide. In the peptide coding region, both human and mouse urocortin III are 76% identical to pufferfish urocortin-related peptide and more distantly related to urocortin II, CRF, and urocortin from other mammalian species. Mouse urocortin III mRNA expression is found in areas of the brain including the hypothalamus, amygdala, and brainstem, but is not evident in the cerebellum, pituitary, or cerebral cortex; it is also expressed peripherally in small intestine and skin. Urocortin III is selective for type 2 CRF receptors and thus represents another potential endogenous ligand for these receptors.

Control of Meristem Development by CLAVATA1 Receptor Kinase and Kinase-Associated Protein Phosphatase Interactions1

The CLAVATA1 (CLV1) gene encodes a putative receptor kinase required for the proper balance between cell proliferation and differentiation in Arabidopsis shoot and flower meristems. Impaired CLV1 signaling results in masses of undifferentiated cells at the shoot and floral meristems. Although many putative receptor kinases have been identified in plants, the mechanism of signal transduction mediated by plant receptor-like kinases is largely unknown. One potential effector of receptor kinase signaling is kinase-associated protein phosphatase (KAPP), a protein that binds to multiple plant receptor-like kinases in a phosphorylation-dependent manner. To examine a possible role for KAPP in CLV1-dependent plant development, the interaction of CLV1 and KAPP was investigated in vitro and in vivo. KAPP binds directly to autophosphorylated CLV1 in vitro and co-immunoprecipitates with CLV1 in plant extracts derived from meristematic tissue. Reduction of KAPP transcript accumulation in an intermediate clv1 mutant suppresses the mutant phenotype, and the degree of suppression is inversely correlated with KAPP mRNA levels. These data suggest that KAPP functions as a negative regulator of CLV1 signaling in plant development. This may represent a general model for the interaction of KAPP with receptor kinases.

CrkII signals from epidermal growth factor receptor to Ras.

A rat fibroblast mutant defective in oncogenic transformation and signaling from epidermal growth factor receptor to Ras has been isolated. The mutant contains dominant negative-type point mutations in the C-terminal SH3 domain of one crkII gene. Among the adapters tested, the mutant is complemented only by crkII cDNA. Expression of the mutated crkII in parent cells generates the phenotype indistinguishable from the mutant cell. Yet overexpression or reduced expression of Grb2 in the mutant before and after complementation with crkII have little effect on its phenotype. We conclude that adapter molecules are highly specific and that the oncogenic growth signal from epidermal growth factor receptor to Ras is predominantly mediated by CrkII in rat fibroblast.

Physiological coupling of growth factor and steroid receptor signaling pathways: estrogen receptor knockout mice lack estrogen-like response to epidermal growth factor.

Past studies have shown that epidermal growth factor (EGF) is able to mimic the uterotropic effects of estrogen in the rodent. These studies have suggested a "cross-talk" model in which EGF receptor (EGF-R) signaling results in activation of nuclear estrogen receptor (ER) and its target genes in an estrogen-independent manner. Furthermore, in vitro studies have indicated the requirement for ER in this mechanism. To verify the requirement for ER in an in vivo system, EGF effects were studied in the uteri of ER knockout (ERKO) mice, which lack functional ER. The EGF-R levels, autophosphorylation, and c-fos induction were observed at equivalent levels in both genotypes indicating that removal of ER did not disrupt the EGF responses. Induction of DNA synthesis and the progesterone receptor gene in the uterus were measured after EGF treatment of both ERKO and wild-type animals. Wild-type mice showed increases of 4.3-fold in DNA synthesis, as well as an increase in PR mRNA after EGF treatment. However, these responses were absent in ERKO mice, confirming that the estrogen-like effects of EGF in the mouse uterus do indeed require the ER. These data conclusively demonstrate the coupling of EGF and ER signaling pathways in the rodent reproductive tract.

Synthesis and characterization of a photoactivatable analog of corticotropin-releasing factor for specific receptor labeling.

A novel photoactivatable analog of ovine corticotropin-releasing factor (ovine photoCRF) has been synthesized and characterized. A diazirine group, the 4-(1-azi-2,2,2-trifluoroethyl)benzoyl residue, was covalently bound to the amino terminus of ovine CRF (oCRF), which was N-terminally extended by a tyrosyl residue for radioactive labeling with 125I. Under mild conditions, photolysis yielded highly reactive carbenes, responsible for the formation of covalent bonds to the CRF receptor. Ovine photoCRF was shown to bind to the high-affinity site of the CRF receptor with a similar Kd value as oCRF. When radioactively iodinated ovine photoCRF (ovine 125I-photoCRF) was covalently linked to rat CRF receptor, type 1 (rCRFR1), permanently transfected into human embryonic kidney (HEK) 293 cells, a highly glycosylated 75-kDa protein was identified with SDS/PAGE. The specificity of ovine 125I-photoCRF was demonstrated by the finding that this analog could be displaced from the receptor by oCRF, but not other unrelated peptides such as vasoactive intestinal peptide. The observed size of the 75-kDa cross-link was in agreement with the molecular weight reported earlier for native CRFR1 from rat brain. Deglycosylation of the 75-kDa cross-link with peptide:N-glycosidase (PNGase) yielded a 46-kDa protein, in agreement with the molecular weight estimated from cDNA coding for rat CRFR1. The developed CRF analog, photoCRF, is expected to facilitate future biochemical and physiological analysis of CRF receptors and--by analogous strategies--of other peptide receptors.

Glial cell line-derived neurotrophic factor activates the receptor tyrosine kinase RET and promotes kidney morphogenesis.

The receptor tyrosine kinase RET functions during the development of the kidney and the enteric nervous system, yet no ligand has been identified to date. This report demonstrates that the glial cell line-derived neurotrophic factor (GDNF) activates RET, as measured by tyrosine phosphorylation of the intracellular catalytic domain. GDNF also binds RET with a dissociation constant of 8 nM, and 125I-labeled GDNF can be coimmunoprecipitated with anti-RET antibodies. In addition, exogenous GDNF stimulates both branching and proliferation of embryonic kidneys in organ culture, whereas neutralizing antibodies against GDNF inhibit branching morphogenesis. These data indicate that RET and GDNF are components of a common signaling pathway and point to a role for GDNF in kidney development.

FRAG1, a gene that potently activates fibroblast growth factor receptor by C-terminal fusion through chromosomal rearrangement.

A constitutively active form of fibroblast growth factor 2 (FGFR2) was identified in rat osteosarcoma (ROS) cells by an expression cloning strategy. Unlike other tyrosine kinase receptors activated by N-terminal truncation in tumors, this receptor, FGFR2-ROS, contains an altered C terminus generated from chromosomal rearrangement with a novel gene, designated FGFR activating gene 1 (FRAG1). While the removal of the C terminus slightly activates FGFR2, the presence of the FRAG1 sequence drastically stimulates the transforming activity and autophosphorylation of the receptor. FGFR2-ROS is expressed as a unusually large protein and is highly phosphorylated in NIH 3T3 transfectants. FRAG1 is ubiquitously expressed and encodes a predicted protein of 28 kDa lacking significant structural similarity to known proteins. Epitope-tagged FRAG1 protein showed a perinuclear localization by immunofluorescence staining. The highly activated state of FGFR2-ROS appears to be attributed to constitutive dimer formation and higher phosphorylation level as well as possibly altered subcellular localization. These results indicate a unique mechanism of receptor activation by a C terminus alteration through a chromosomal fusion with FRAG1.

Identification of ciliary neurotrophic factor (CNTF) residues essential for leukemia inhibitory factor receptor binding and generation of CNTF receptor antagonists.

Ciliary neurotrophic factor (CNTF) drives the sequential assembly of a receptor complex containing the ligand-specific alpha-receptor subunit (CNTFR alpha) and the signal transducers gp130 and leukemia inhibitory factor receptor-beta (LIFR). The D1 structural motif, located at the beginning of the D-helix of human CNTF, contains two amino acid residues, F152 and K155, which are conserved among all cytokines that signal through LIFR. The functional importance of these residues was assessed by alanine mutagenesis. Substitution of either F152 or K155 with alanine was found to specifically inhibit cytokine interaction with LIFR without affecting binding to CNTFR alpha or gp130. The resulting variants behaved as partial agonists with varying degrees of residual bioactivity in different cell-based assays. Simultaneous alanine substitution of both F152 and K155 totally abolished biological activity. Combining these mutations with amino acid substitutions in the D-helix, which enhance binding affinity for the CNTFR alpha, gave rise to a potent competitive CNTF receptor antagonist. This protein constitutes a new tool for studies of CNTF function in normal physiology and disease.

Identification of the zinc-dependent endothelial cell binding protein for high molecular weight kininogen and factor XII: identity with the receptor that binds to the globular "heads" of C1q (gC1q-R).

High molecular weight kininogen (HK) and factor XII are known to bind to human umbilical vein endothelial cells (HUVEC) in a zinc-dependent and saturable manner indicating that HUVEC express specific binding site(s) for those proteins. However, identification and immunochemical characterization of the putative receptor site(s) has not been previously accomplished. In this report, we have identified a cell surface glycoprotein that is a likely candidate for the HK binding site on HUVECs. When solubilized HUVEC membranes were subjected to an HK-affinity column in the presence or absence of 50 microM ZnCl2 and the bound membrane proteins eluted, a single major protein peak was obtained only in the presence of zinc. SDS/PAGE analysis and silver staining of the protein peak revealed this protein to be 33 kDa and partial sequence analysis matched the NH2 terminus of gC1q-R, a membrane glycoprotein that binds to the globular "heads" of C1q. Two other minor proteins of approximately 70 kDa and 45 kDa were also obtained. Upon analysis by Western blotting, the 33-kDa band was found to react with several monoclonal antibodies (mAbs) recognizing different epitopes on gC1q-R. Ligand and dot blot analyses revealed zinc-dependent binding of biotinylated HK as well as biotinylated factor XII to the isolated 33-kDa HUVEC molecule as well as recombinant gC1q-R. In addition, binding of 125I-HK to HUVEC cells was inhibited by selected monoclonal anti-gC1q-R antibodies. C1q, however, did not inhibit 125I-HK binding to HUVEC nor did those monoclonals known to inhibit C1q binding to gC1q-R. Taken together, the data suggest that HK (and factor XII) bind to HUVECs via a 33-kDa cell surface glycoprotein that appears to be identical to gC1q-R but interact with a site on gC1q-R distinct from that which binds C1q.