Exogenous administration of protease-resistant, non-matrix-binding IGFBP-2 inhibits tumour growth in a murine model of breast cancer.

BACKGROUND: Insulin-like growth factors (IGF-I and IGF-II) signal via the type 1 IGF receptor (IGF-1R) and IGF-II also activates the insulin receptor isoform A (IR-A). Signalling via both receptors promotes tumour growth, survival and metastasis. In some instances IGF-II action via the IR-A also promotes resistance to anti-IGF-1R inhibitors. This study assessed the efficacy of two novel modified IGF-binding protein-2 (IGFBP-2) proteins that were designed to sequester both IGFs. The two modified IGFBP-2 proteins were either protease resistant alone or also lacked the ability to bind extracellular matrix (ECM). METHODS: The modified IGFBP-2 proteins were tested in vitro for their abilities to inhibit cancer cell proliferation and in vivo to inhibit MCF-7 breast tumour xenograft growth. RESULTS: Both mutants retained low nanomolar affinity for IGF-I and IGF-II (0.8-2.1-fold lower than IGFBP-2) and inhibited cancer cell proliferation in vitro. However, the combined protease resistant, non-matrix-binding mutant was more effective in inhibiting MCF-7 tumour xenograft growth and led to inhibition of angiogenesis. CONCLUSIONS: By removing protease cleavage and matrix-binding sites, modified IGFBP-2 was effective in inhibiting tumour growth and reducing tumour angiogenesis.

Suckling defect in mice lacking the soluble haemopoietin receptor NR6.

Cytokines control a variety of cellular responses including proliferation, differentiation, survival and functional activation, via binding to specific receptors expressed on the surface of target cells [1]. The cytokine receptors of the haemopoietin family are defined by the presence of a conserved 200 amino acid extracellular domain known as the haemopoietin domain [2]. We report here the isolation of NR6, a haemopoietin receptor that, like the p40 subunit of interleukin-12 (IL-12) [3] and the EBI3 gene induced by Epstein-Barr virus infection in lymphocytes [4], contains a typical haemopoietin domain but lacks transmembrane and cytoplasmic domains. Although in situ hybridisation revealed NR6 expression at multiple sites in the developing embryo, mice lacking NR6 did not display obvious abnormalities and were born in the expected numbers. Neonatal NR6(-/-) mice failed to suckle, however, and died within 24 hours of birth, suggesting that NR6 is necessary for the recognition or processing of pheromonal signals or for the mechanics of suckling itself. In addition, NR6(-/-) mice had reduced numbers of haemopoietic progenitor cells, suggesting a potential role in the regulation of primitive haemopoiesis.

DAN is a secreted glycoprotein related to Xenopus cerberus.

We report that DAN, a potential cell cycle regulator and tumour suppressor, is a secreted glycoprotein related to Xenopus cerberus. DAN, cerberus, its mouse relative Cer-1/cer-l/Cerberus-like/Cerr1, and the recently described factor DRM/Gremlin, appear to be members of the cystine knot superfamily, which includes TGFbetas and BMPs. Like cerberus and mCer-1, DAN-induced cement glands as well as markers of anterior neural tissue and endoderm in Xenopus animal cap assays, features of BMP signalling blockade. During mouse embryogenesis, Dan was expressed from E8.5 in cranial mesenchyme and somites, then later in limb and facial mesenchyme. The pattern in somites was highly dynamic, with transcripts initially localized to the caudal half of the nascent epithelial somite, then, after maturation, to sclerotomal cells adjacent to the neural tube. Dan was also expressed in the developing myotome. The expression domains include sites in which BMP inhibition is known to be important for development. Thus, DAN appears to be a secreted factor belonging to the cystine knot superfamily, and one of a growing number of antagonists acting to modulate BMP signalling during development.

Purification and refolding of vascular endothelial growth factor-B.

Vascular endothelial growth factor (VEGF)-A interacts with the receptor tyrosine kinases VEGF-R1 and R2, and the importance of this interaction in endothelial cell (EC) function and blood vessel development has been well documented. Other ligands that interact differentially with these receptors and that are structurally related to VEGF-A include VEGF-B, VEGF-C, VEGF-D, and placenta growth factor (PLGF). Compared with VEGF-A, relatively little is known about the biological role of the VEGF-R1 specific ligand, VEGF-B. Two splice variant isoforms that differ at the COOH-terminus and which retain unique solubility characteristics are widely expressed throughout embryonic and postnatal development. Recent analysis of mice with a targeted deletion of the VEGF-B gene has revealed a defect in heart development and function consistent with an important role in vascularization of the myocardium (Bellomo D et al., 2000, Circ Res 86:E29-E35). To facilitate further characterization of VEGF-B, we have developed a protocol for expression and purification of refolded recombinant protein from Escherichia coli inclusion bodies (IBs). The approach developed resolves a number of significant issues associated with VEGF-B, including the ability to heterodimerize with endogenous VEGF-A when co-expressed in mammalian cells, a complex secondary structure incorporating inter- and intrachain disulfide bonds and hydrophobic characteristics that preclude the use of standard chromatographic resins. The resulting purified disulfide-linked homodimer was demonstrated to bind to VEGF-R1 and to compete with VEGF-A for binding to this receptor.

Primary structure of ovine pituitary basic fibroblast growth factor.

The complete amino acid sequence of basic FGF (146 residues) from ovine pituitary glands has been established. This has been achieved by the sequence analysis of subnanomole amounts of the intact molecule and of peptides derived by enzymatic digestions with clostripain, chymotrypsin, pepsin and Staphylococcus aureus V8 protease. Microbore HPLC, employing 1-2 mm i.d. columns, was used to purify, concentrate and buffer-exchange the FGF peptides. A novel application of ion-pairing chromatography was employed to isolate peptides which were not retained on conventional reversed-phase systems. There is only one positional difference between the ovine and bovine basic FGFs, but there are 3 positional differences between ovine and human basic FGFS.

Synergies between micropreparative high-performance liquid chromatography and an instrumental optical biosensor.

The recent development of an automated surface plasmon resonance technology for the measurement of biomolecular interactions (Pharmacia BIAcore) has provided new opportunities for the detection and analysis of protein-protein interactions. In the BIAcore, detection is based on changes in surface plasmon resonance which are monitored optically. Changes in surface plasmon resonance correspond to changes in surface concentration of macromolecules and can be monitored in real time. We have found that the detection sensitivity obtainable with this technology (ng/ml concentrations of specific ligands are readily detectable for many applications) is complementary "in a bidirectional manner" to micropreparative HPLC. Thus micropreparative HPLC may be used to purify and characterise reagents for the biosensor, whilst the biosensor may be used to define chromatographic parameters such as elution conditions for affinity chromatography or serve as an affinity detector for fractions obtained during chromatographic purification. Examples of such applications, including the potential of the biosensor to search for and monitor the purification of unknown ligands for which the target molecule has been identified, are shown. In particular, the use of the biosensor to monitor the purification of soluble epidermal growth factor receptor from A431 cell conditioned media is demonstrated.

Targeting of acute myeloid leukemia in vitro and in vivo with an anti-CD123 mAb engineered for optimal ADCC.

Acute myeloid leukemia (AML) is a biologically heterogeneous group of related diseases in urgent need of better therapeutic options. Despite this heterogeneity, overexpression of the interleukin (IL)-3 receptor α-chain (IL-3 Rα/CD123) on both the blast and leukemic stem cell (LSC) populations is a common occurrence, a finding that has generated wide interest in devising new therapeutic approaches that target CD123 in AML patients. We report here the development of CSL362, a monoclonal antibody to CD123 that has been humanized, affinity-matured and Fc-engineered for increased affinity for human CD16 (FcγRIIIa). In vitro studies demonstrated that CSL362 potently induces antibody-dependent cell-mediated cytotoxicity of both AML blasts and CD34(+)CD38(-)CD123(+) LSC by NK cells. Importantly, CSL362 was highly effective in vivo reducing leukemic cell growth in AML xenograft mouse models and potently depleting plasmacytoid dendritic cells and basophils in cynomolgus monkeys. Significantly, we demonstrated CSL362-dependent autologous depletion of AML blasts ex vivo, indicating that CSL362 enables the efficient killing of AML cells by the patient's own NK cells. These studies offer a new therapeutic option for AML patients with adequate NK-cell function and warrant the clinical development of CSL362 for the treatment of AML.

Protein engineering and preclinical development of a GM-CSF receptor antibody for the treatment of rheumatoid arthritis.

BACKGROUND AND PURPOSE: For antibody therapies against receptor targets, in vivo outcomes can be difficult to predict because of target-mediated clearance or antigen 'sink' effects. The purpose of this work was to engineer an antibody to the GM-CSF receptor α (GM-CSFRα) with pharmacological properties optimized for chronic, s.c. treatment of rheumatoid arthritis (RA) patients. EXPERIMENTAL APPROACH: We used an in silico model of receptor occupancy to guide the target affinity and a combinatorial phage display approach for affinity maturation. Mechanism of action and internalization assays were performed on the optimized antibody in vitro before refining the modelling predictions of the eventual dosing in man. Finally, in vivo pharmacology studies in cynomolgus monkeys were carried out to inform the predictions and support future clinical development. KEY RESULTS: Antibody potency was improved 8600-fold, and the target affinity was reached. The refined model predicted pharmacodynamic effects at doses as low as 1 mg kg(-1) and a study in cynomolgus monkeys confirmed in vivo efficacy at 1 mg kg(-1) dosing. CONCLUSIONS AND IMPLICATIONS: This rational approach to antibody drug discovery enabled the isolation of a potent molecule compatible with chronic, s.c. self-administration by RA patients. We believe this general approach enables the development of optimal biopharmaceuticals.

Comparison of iodinated [Nle15]- and [Met15]-gastrin17 prepared by reversed-phase HPLC.

125I-[Nle15]-gastrin17 prepared by the iodogen method can be separated by reversed-phase high performance liquid chromatography into two peaks, both of which elute after [Nle15]-gastrin17. Direct determination of the specific activities of the two derivatives by microbore reversed-phase HPLC indicated that they were the mono- and di-iodinated species. In contrast the two peaks obtained with [Met15]-gastrin17 iodinated under the same conditions eluted earlier, relative to the appropriate gastrin17 standard, than the [Nle15]-gastrin derivatives. Treatment of either peak with 0.75 M dithiothreitol at 56 degrees C or 95 degrees C resulted in progressive conversion to compounds migrating in relative positions similar to the 125I-[Nle15]-gastrin17 derivatives. Direct determination of the specific activity of the earlier eluting [Met15]-gastrin17 derivative before reduction indicated that it was the mono-iodinated species. It thus appears likely that iodination of [Met15]-gastrin17 by the iodogen method results predominantly in the formation of mono- and di-125I-[Met sulphoxide15]-gastrin17. To avoid problems arising from oxidation of the methionine residue of gastrin during iodination, the use of 125I-[Nle15]-gastrin17 in binding studies is therefore recommended.

Platelet-derived growth factor stimulates the release of protein kinase A from the cell membrane.

The mitogenic action of growth factors involves the stimulation of intracellular protein kinases. In this report we have characterized the major protein kinase released from Balb/c 3T3 and normal rat kidney plasma membranes by the action of platelet-derived growth factor (PDGF). PDGF appears to stimulate the release of approximately 10 proteins, at least one of which is a kinase capable of phosphorylating proteins on Ser or Thr (as determined by the lability of the phosphate to alkali treatment). More than 90% of the Ser/Thr kinase activity was inhibited by PKI5-22, a specific peptide inhibitor of the cAMP-dependent protein kinase (PKA). We used immunoblotting to confirm that the kinase released in response to PDGF was PKA. cAMP also stimulated the release of PKA, and the set of protein substrates phosphorylated was similar following PDGF or cAMP stimulation. Interestingly, in the presence of a cAMP analogue ((Rp)-cAMPS), cAMP could not induce dissociation of PKA from the membranes, whereas stimulation by PDGF increased the level of PKA activation. Furthermore, unlike Swiss 3T3 cells, neither Balb/c 3T3 fibroblasts nor normal rat kidney cells accumulate cAMP in response to PDGF, yet the level of PKA in the cytosol of these intact cells increases in response to PDGF. Thus, it appears as though PDGF activation of the membrane-associated form of the PKA holoenzyme occurs by a mechanism independent of an elevation in cAMP levels.

Micropreparative ion exchange high performance liquid chromatography: applications to microsequence analysis.

Data on the characterization of anion and cation micropreparative (50 x 1.6 mm i.d.) HPLC columns is presented. It is shown how subnanomole quantities of protein can be efficiently recovered from such columns, rendering them compatible for use in multidimensional chromatographic strategies for the purification of trace biological samples. By selection of appropriate solvent systems (e.g., buffer-free sodium chloride solutions), the small eluant peak volumes can be loaded directly onto the gas phase sequencer, and N-terminal sequence data obtained. The potential of the technique is illustrated for the purification of a GTPase activating protein (GAP-3).

Characterization of bovine heparin-binding neurotrophic factor (HBNF): assignment of disulfide bonds.

The topology of the disulfides in native heparin-binding neurotrophic factor (HBNF) isolated from bovine brain was studied by proteolytic digestion using trypsin, Asp-N endoproteinase and chymotrypsin and peptide mapping. Disulfide-linked peptides were identified by automated Edman degradation. It has been shown that there are disulfide bonds between Cys15-Cys44, Cys23-Cys53, Cys30-Cys57, Cys67-Cys99 and Cys77-Cys109.

Mapping of the antibody- and receptor-binding domains of granulocyte colony-stimulating factor using an optical biosensor. Comparison with enzyme-linked immunosorbent assay competition studies.

An automated optical biosensor instrument for measuring molecular interactions (Pharmacia BIAcore) has been used to characterise the epitopes recognised by 15 monoclonal antibodies raised against recombinant human granulocyte colony-stimulating factor (G-CSF). The BIAcore combines an autosampler and integrated microfluidic cartridge for the introduction and transportation of samples to the sensor chip surface, with surface plasmon resonance to detect binding events. A rabbit anti-mouse Fc antibody, coupled to the sensor surface in situ using conventional protein chemistry techniques, was used to capture an anti-G-CSF monoclonal antibody. G-CSF was bound to this antibody by injection over the sensor surface. Multi-site binding experiments were then performed in which other anti-G-CSF monoclonal antibodies were injected sequentially over the surface, and their ability to bind to the G-CSF in a multimolecular complex monitored in real time. Results obtained using the biosensor have been compared with data obtained by cross competition studies using biotinylated antibodies or antibody binding studies using chemically or enzymatically derived G-CSF peptide fragments or synthetic peptides. The results of these studies are in excellent agreement with the data from the BIAcore, although modification of the antibody or G-CSF occasionally altered the epitope affinity.

The major colonic cell mitogen extractable from colonic mucosa is an N terminally extended form of basic fibroblast growth factor.

Colonic growth factors (CGFs) were extracted from porcine intestinal epithelium and mucosa. Under acidic conditions, very little mitogenic activity (as assayed using murine 3T3 fibroblasts and a human colonic cell line) was extractable. However, by extracting at neutral or slightly alkaline pH, significant mitogenic activity for both the murine fibroblasts and human colonic carcinoma cell line could be detected. CGFs are present throughout the intestine and cecum. The epithelial mucosa of the distal colorectal region appeared to contain mitogens which were more potent for the colonic cells than the 3T3 fibroblasts. Purification of CGFs from the colonic mucosa required removal of associated mucin by pH precipitation prior to chromatographic fractionation. It was then possible to develop a complete purification (390,000-fold) scheme for the major CGF, an 18-kDa protein which bound to heparin-Sepharose. N-terminal sequence analysis yielded a single sequence (Q)SPGGAMAAGSITTLPALP, i.e. an N-terminally extended form of basic fibroblast growth factor. Apart from the substitution of Gly in bovine basic fibroblast growth factor by a Ser in porcine CGF, the proteins are identical. A similar extraction procedure using purified human colonic crypt epithelial cells yielded a mitogen for the human colonic cell line with similar chromatographic properties.

Rsr1 and Rap1 GTPases are activated by the same GTPase-activating protein and require threonine 65 for their activation.

The Rsr1 protein of Saccharomyces cerevisiae has been shown to be essential for bud site selection (Bender, A., and Pringle, J. (1989) Proc. Natl. Acad. Sci. U.S.A. 86, 9976-9980). This protein of 272 amino acids shares approximately 50% sequence identity with both Ras and Rap GTPases. However, neither GTP binding nor GTPase activity of the Rsr1 protein has been reported. The Rsr1 protein shares with human Rap1 GTPases the four specific motifs, i.e. Gly-12, residues 32-40, Ala-59, and residues 64-70, that are required for GAP3-dependent activation of the Rap1 GTPases. In this paper we demonstrate that the intrinsic GTPase activity of the Rsr1 protein is stimulated by GAP3 purified from bovine brain cytosol. The Rsr1 GTPase is not activated by either GAP1 or GAP2 which are specific for the Ras and Rho GTPases, respectively. Thus, it appears that the Rsr1 GTPase is a new member of the Rap1 GTPase family. Replacement of Gly-12 by Val in the Rsr1 GTPase completely abolishes the GAP3-dependent activation. The chimeric GTPases, Ras(1-60)/Rsr1(61-168) and Rsr1(1-65)/Ras(66-189), are activated by GAP3 but not by GAP1. Replacement of Thr-65 by Ser in the latter chimeric GTPase completely abolishes the GAP3-dependent activation, indicating that Thr-65 is required for distinguishing GAP3 from GAP1. We have previously shown that Gln-61 and Ser-65 are sufficient to determine the GAP1 specificity. Replacement of Thr-35 by Ala in the common effector domain (residues 32-40) of the chimeric Ras/Rsr1 GTPases completely abolishes GAP3-dependent activation.

Structural characterisation of native and recombinant forms of the neurotrophic cytokine MK.

The retinoic acid (RA)-inducible midkine (MK) gene encodes a heparin-binding protein which can induce neurite outgrowth in cultured mammalian embryonic brain cells. This cytokine shares 65% amino acid sequence identity with another RA-inducible cytokine, pleiotropin (PTN). Both proteins contain 10 conserved cysteine residues, all of which appear to be disulphide linked. MK and PTN are also rich in lysine and arginine residues rendering them susceptible to proteolysis during purification, and making large-scale preparation of these molecules inherently difficult. Recombinant MK has been expressed as a fusion protein using a pGEX vector transfected into E. coli. To enable refolding of MK, the fusion protein was stored in solution at 4 degrees C for 14 days in the presence of dithiothreitol (DTT). Thrombin cleavage of the fusion protein, post storage, typically generated 5 mg of MK per litre of bacterial pellet. To establish the structural integrity of the recombinant product, we have analysed the refolding kinetics and compared the disulphide bond assignment of recombinant MK with that of native MK and native PTN. The synergistic use of micropreparative HPLC, to separate and recover in small eluant volumes enzymatically derived peptide fragments, with matrix assisted laser desorption mass spectrometry (MALD-MS) and N-terminal sequence analysis has allowed the unambiguous identification of the disulphide bonded fragments of native and recombinant MK. The disulphide bond assignment of MK is C12-C36, C20-C45, C27-C49, C59-C91 and C69-C101, and is equivalent to that of PTN.

The purification of a Rap1 GTPase-activating protein from bovine brain cytosol.

Two GTPase-activating proteins (GAPs) have been detected in extracts from bovine brain: GAP-1, which is specific for the activation of ras GTPases, and GAP-3, which is specific for the activation of the rap1 GTPases. We present a strategy for the purification to homogeneity of a cytosolic form of GAP-3 from bovine brain. The 100,000 x g supernatant from homogenized brains was chromatographed sequentially on DEAE Fast Flow, green H-E4BD Sepharose, Bio-Gel A1.5, hydroxyapatite, and phenyl-Sepharose prior to high resolution separation on Mono Q HR 5/5, phenyl-Superose HR 5/5, Mono Q PC 1.6/5, and Superose 12 PC 3.2/30. This procedure resulted in an approximately 18,000-fold purification, yielding 50 micrograms of GAP-3 from 1.6 kg of tissue. Purified cytosolic GAP-3 migrated as a single band of apparent Mr 55,000 on sodium dodecyl sulfate-polyacrylamide gel electrophoresis. However, on gel filtration cytosolic GAP-3 chromatographed as a dimer with an apparent Mr 92,000. Purified GAP-3 does not activate ras or rho GTPases and possesses no intrinsic GTPase activity. Amino acid sequence data indicated a proline-rich N terminus. The amino acid sequences of peptides generated by Staphylococcus aureus V8 digestion of reduced and pyridine-ethylated GAP-3 showed no similarity to the predicted primary structure of GAP-1 or any other proteins in the nucleic acid or protein data bases. By comparison with the data of Rubinfeld et al. (Rubinfeld, B., Munemitsu, S., Clark, R., Conroy, L., Watt, K., Crosier, W.J., McCormick, F., and Polakis, P. (1991) Cell 65, 1033-1042), it appears that the membrane-associated (Mr 85,000-95,000) and cytosolic forms of GAP-3 are derived from equivalent, or closely related, genes.

Human granulocyte-macrophage colony-stimulating factor (hGM-CSF): identification of a binding site for a neutralizing antibody.

One approach to the localization of functionally active regions of human granulocyte-macrophage colony-stimulating factor (hGM-CSF) is to map the epitopes recognized by neutralizing anti-hGM-CSF monoclonal antibodies. We have defined the epitope recognized by one neutralizing antibody (LMM102) using proteolytic fragments obtained by enzymic digestion of bacterially synthesized hGM-CSF. RP-HPLC fractionation of a tryptic digest resulted in the identification of an immunoreactive "tryptic core" peptide containing 66 amino acids (52% of the protein). Further digestion of this "tryptic core" with S. aureus V8 protease produced a unique immunoreactive hGM-CSF product comprising two peptides, residues 86-93 and 112-127, linked by a disulfide bond between residues 88 and 121. The individual peptides, generated by reduction with dithiothreitol, were not recognized by the antibody. An analog of this peptide has been synthesized chemically and shown to have similar immunoreactivity to the epitope obtained by enzymic digestion. A series of modified peptides has also been synthesized to identify further the region required for antibody recognition.

Murine epidermal growth factor: structure and function.

Murine epidermal growth factor (EGF), a 53 amino acid protein, has been modified by enzymic digestion, site-specific chemical reactions, and recombinant DNA technology. After trypsin digestion the EGF derivatives EGF1-48 (called EGF-T) and EGF1-45 (called EGF-T2) were separated from the residual EGF and the C-terminal pentapeptide by reversed-phase high-performance liquid chromatography. EGF-T competes for binding to EGF receptors with the same efficiency as EGF. The EGF-T2 derivative had no detectable receptor binding activity even at 100 nM. The in vitro mitogenic potencies of EGF and EGF-T for Balb/c 3T3 cells were indistinguishable. Treatment of EGF-T with carboxypeptidase Y yielded two derivatives, EGF-T-(des-Arg48) and EGF-T-des(Leu47-Arg48). There was only a 3-7-fold diminution in the binding efficiency and mitogenic potency for EGF-T-(des-Arg48). However, there was more than a 100-fold decrease in the binding efficiency and mitogenic activity of EGF-T-des (Leu47-Arg48). These results indicated that Leu47 is intimately involved in the formation of the ligand-receptor complex. Studies with a number of proteases indicated that the C-terminus of EGF was susceptible to enzymic digestion; however, the N-terminus appears to be folded into a conformation which prevents access to proteolytic digestion. Consequently, the N-terminus was modified by preparing an analogue with recombinant DNA technology. Oligonucleotides corresponding to EGF(3-48). Met3 Lys21 residues were ligated in frame to a beta-galactosidase expression vector. The beta-Gal-EGF fusion protein was cleaved with cyanogen bromide and EGF(4-48).Lys21 purified.(ABSTRACT TRUNCATED AT 250 WORDS)