Engineering a soluble extracellular erythropoietin receptor (EPObp) in Pichia pastoris to eliminate microheterogeneity, and its complex with erythropoietin.

The extracellular ligand-binding domain (EPObp) of the human EPO receptor (EPOR) was expressed both in CHO (Chinese Hamster Ovary) cells and in Pichia pastoris. The CHO and yeast expressed receptors showed identical affinity for EPO binding. Expression levels in P. pastoris were significantly higher, favoring its use as an expression and scale-up production system. Incubation of EPO with a fourfold molar excess of receptor at high protein concentrations yielded stable EPO-EPObp complexes. Quantification of EPO and EPObp in the complex yielded a molar ratio of one EPO molecule to two receptor molecules. Residues that are responsible for EPOR glycosylation and isomerization in Pichia were identified and eliminated by site-specific mutagenesis. A thiol modification was identified and a method was developed to remove the modified species from EPObp. EPObp was complexed with erythropoietin (EPO) and purified. The complex crystallized in two crystal forms that diffracted to 2.8 and 1.9 A respectively. (Form 1 and form 2 crystals were independently obtained at AxyS Pharmaceuticals, Inc. and Amgen, Inc. respectively.) Both contained one complex per asymmetric unit with a stoichiometry of two EPObps to one EPO.

Homodimerization of erythropoietin receptor by a bivalent monoclonal antibody triggers cell proliferation and differentiation of erythroid precursors.

Erythropoietin (EPO) stimulates proliferation and differentiation of erythroid progenitor cells. Several lines of evidence indicate that the most likely mechanism of EPO receptor (EPO-R) activation by EPO is homodimerization of the receptor on the surface of erythrocyte precursors. Therefore, we argued that it should be possible to raise EPO-R monoclonal antibodies (MoAbs) that would activate the receptor by dimerization and thus mimic EPO action. We have identified such an agonist MoAb (MoAb34) directed against the extracellular EPO binding domain of the EPO-R. This bivalent IgG antibody triggers the proliferation of EPO-dependent cell lines and induces differentiation of erythroid precursors in vitro. In contrast, the monovalent Fab fragment, which cannot dimerize the receptor, is completely inactive. The mechanism of receptor activation by homodimerization implies that at high ligand concentrations the formation of 1:1 receptor/ligand complexes is favored over 2:1 complexes, thereby turning the ligand agonist into an antagonist. Thus, EPO and MoAb34 should self-antagonize at high concentrations in both cell proliferation and differentiation assays. Our data indeed demonstrate that EPO and MoAb34 antagonize ligand-dependent cell proliferation with IC50 values of approximately 20 and 2 mumol/L, respectively. Erythroid colony formation (BFUe) is inhibited at MoAb34 concentrations above 1 mumol/L. Furthermore, we analyzed the MoAb34:EPO-R interaction using a mathematic model describing antibody-mediated receptor dimerization. The data for proliferation and differentiation activity were consistent with the receptor dimer formation on the cell surface predicted by the model.

Carboxyl-terminal truncation of leucine76 converts heparin-binding EGF-like growth factor from a heparin-enhancible to a heparin-suppressible growth factor.

Previous studies have indicated that heparin differentially regulates heparin-binding EGF-like growth factor (HB-EGF) and amphiregulin (AR) mitogenic activity. To further explore this phenomenon, these mitogens were compared under identical cell culture conditions in two different assays. The results of our present investigation demonstrated that AR-mediated mitogenic activity in the murine AKR-2B fibroblast-like cell line was inhibited by heparin, while HB-EGF activity was enhanced. However, the absolute effect of heparin appeared to be cell type specific since HB-EGF mitogenic activity was not dramatically affected by coincubation with heparin when tested on human dermal fibroblasts. Several studies have indicated that mutation of a conserved leucine in the carboxyl-terminal region of both EGF and transforming growth factor-alpha results in decreased affinity for EGF receptors. Since this leucine is present in the analogous position of HB-EGF, but absent in AR, we examined the effect of deleting this residue by carboxyl-terminal truncation of HB-EGF. Analysis of recombinant forms of HB-EGF demonstrated that HB-EGF can be converted to a heparin-inhibited growth factor if the putative mature form of the protein is truncated by two residues (leucine76 and proline77) at the carboxyl terminus. Further analysis demonstrated that only leucine76 appears to be required for heparin-dependent enhancement of HB-EGF-mediated mitogenic activity, indicating that this amino acid may play a pivotal role in controlling the response of HB-EGF to heparin or related glycosaminoglycan sulfates. Our results also suggest that expression of different HB-EGF forms in vivo could result in the production of HB-EGFs with divergent responses to sulfated glycosaminoglycans and proteoglycans.

Differential effects of a heparin antagonist (hexadimethrine) or chlorate on amphiregulin, basic fibroblast growth factor, and heparin-binding EGF-like growth factor activity.

Amphiregulin (AR) and heparin-binding EGF-like growth factor (HB-EGF) are two recently identified members of the EGF family. Both AR and HB-EGF share with EGF the ability to interact with the type-1 EGF receptor; however, AR and HB-EGF differ from EGF in that both of these mitogens bind to heparin while EGF does not. To determine whether interactions with heparin-like molecules on the cell surface influence binding of AR and HB-EGF with EGF receptors and the subsequent mitogenic activity exerted by these growth factors, murine AKR-2B and Balb/MK-2 cells were treated with either an inhibitor of proteoglycan sulfation (chlorate) or a heparin antagonist (hexadimethrine). As expected, neither treatment significantly altered the specific binding of 125I-EGF on AKR-2B cells. Interestingly, treatment with either chlorate or hexadimethrine inhibited the ability of AR to compete with 125I-EGF for cell surface binding and also attenuated AR-mediated DNA synthesis. Thus, as has been suggested for other heparin-binding growth factors such as basic fibroblast growth factor (bFGF), the interaction of AR with an EGF-binding receptor appears to be facilitated by interaction with cell-associated sulfated glycosaminoglycans or proteoglycans. Unexpectedly, however, neither chlorate nor hexadimethrine treatment caused an inhibition of HB-EGF-induced mitogenic activity. Chlorate treatment did not significantly alter the ability of HB-EGF to compete with 125I-EGF for cell surface binding sites, however, heparin and hexadimethrine reduced the ability of HB-EGF to compete for 125I-EGF binding. These results suggest that, in AKR-2B cells, HB-EGF may mediate its mitogenic response at least in part through a receptor which appears to be selective for HB-EGF and permits HB-EGF-mediated mitogenic responses in the presence of hexadimethrine or heparin. Finally, hexadimethrine inhibited the specific binding and mitogenic activity of bFGF, suggesting that this cationic polymer can function as an antagonist of heparin-binding mitogens other than AR.

Trinitrophenyl-ATP binding to the ArsA protein: the catalytic subunit of an anion pump.

The ars operon of the conjugative R-factor R773 confers resistance to arsenicals by coding for an anion pump for extrusion of arsenicals from cells of Escherichia coli. Extrusion of arsenite requires only two polypeptides, the ArsA and ArsB proteins. Purified ArsA protein exhibits oxyanion-stimulated ATPase activity and has been shown to bind ATP by photoaffinity labeling with [alpha-32P]ATP. From sequence analysis the ArsA protein is predicted to have two nucleotide binding folds, one in the N-terminal half and one in the C-terminal half of the protein. Purified ArsA protein bound a fluorescent ATP analogue, 2',3'-O-(2,4,6-trinitrophenylcyclohexadienylidene)adenosine- 5'-triphosphate, with an apparent stoichiometry of 2 mol of nucleotide per mole of ArsA. Strains expressing plasmids with mutations in the N-terminal consensus nucleotide sequence bound only 1 mol of nucleotide per mole of protein.

Ligand interactions in the ArsA protein, the catalytic component of an anion-translocating adenosinetriphosphatase.

The ars operon of the conjugative R-factor R773 produces resistance to arsenicals in cells of Escherichia coli. The operon encodes an oxyanion pump which is composed of a membrane subunit, the 45.5-kDa ArsB protein, and a catalytic subunit, the 63-kDa ArsA protein. Purified ArsA protein is an arsenite(antimonite)-stimulated ATPase. From its amino acid sequence, as deduced from the nucleotide sequence, the ArsA protein has four tryptophanyl residues which could serve as intrinsic fluorescent probes for the study of substrate-induced conformational changes. Both static and dynamic measurements of tryptophan fluorescence were performed with the ArsA protein. Results from static anisotropy measurements indicated differences in molecular motion with addition of ATP, SbO2-, or Mg2+. These results were supported by time decay measurements of fluorescence anisotropy. The results of time decay measurements indicated a shorter correlation time, reflecting localized motion in the vicinity of the probe, and a longer correlation time, which could have arisen from rotation of the major portion of the molecule. The longer correlation time changed with addition of the various effectors, especially MgCl2, suggesting that binding of Mg2+ decreases probe mobility.

Substrate-induced dimerization of the ArsA protein, the catalytic component of an anion-translocating ATPase.

The ArsA protein, the catalytic component of the plasmid-encoded resistance system for removal of the toxic oxyanions arsenite, antimonite, and arsenate from bacterial cells, catalyzes oxyanion-stimulated ATP hydrolysis. Three lines of evidence suggest that the ArsA protein functions as a homodimer. First, the ArsA protein was modified with 5'-p-fluorosulfonyl-benzoyladenosine (FSBA). Antimonite potentiated FSBA inhibition, while ATP or ADP afforded partial protection. ATP and antimonite together provided complete protection, indicating interaction of the anion- and nucleotide-binding sites. The estimated Ki values for FSBA were 0.4 mM in the absence of antimonite and 0.1 mM in the presence of antimonite, suggesting that the binding of antimonite increased the affinity of ArsA protein for FSBA. Incorporation of [14C]FSBA was examined. Extrapolation of the amount of FSBA required to inactivate the protein indicated that 1 mol of FSBA was sufficient to inhibit the activity of 1 mol of ArsA protein in the absence of substrates, while only 0.5 mol was required in the presence of the anionic substrate antimonite. Second, chemical cross-linking of the 63-kDa ArsA protein with N-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinoline resulted in formation of a species approximately twice the size of the monomer in the presence of antimonite but not ATP. Third, determination of the average mass of the ArsA protein in solution by light scattering demonstrated that the average species was 66 kDa in the absence of substrates. In the presence of antimonite the weight average molecular mass increased to a mass in excess of 100 kDa. These results are consistent with the ArsA protein existing in an equilibrium between monomer and dimer, with the equilibrium favoring dimerization upon binding of the anionic substrate. Moreover, total loss of ATPase activity in the half-modified enzyme suggests that the catalytic sites on each monomer must interact.

A plasmid-encoded anion-translocating ATPase.

An anion-translocating ATPase has been identified as the product of the arsenical resistance operon of resistance plasmid R773. When expressed in Escherichia coli this ATP-driven oxyanion pump catalyzes extrusion of the oxyanions arsenite, antimonite and arsenate. Maintenance of a low intracellular concentration of oxyanion produces resistance to the toxic agents. The pump is composed of two polypeptides, the products of the arsA and arsB genes. This two-subunit enzyme produces resistance to arsenite and antimonite. A third gene, arsC, expands the substrate specificity to allow for arsenate pumping and resistance.

Mutagenesis of a nucleotide-binding site of an anion-translocating ATPase.

The ars operon of the conjugative R-factor R773 confers resistance to arsenicals by coding for an anion pump for extrusion of arsenicals from cells of Escherichia coli. The operon encodes three structural genes arsA, arsB, and arsC. The anion pump requires only two polypeptides, the ArsA and ArsB proteins. Purified ArsA protein exhibits oxyanion-stimulated ATPase activity and was demonstrated to bind ATP by photoaffinity labeling with [alpha-32P]ATP. Analysis of the amino acid sequence deduced from the nucleotide sequence of the arsA gene suggests that the ArsA protein contains two potential nucleotide binding folds, one in the N-terminal half and one in the C-terminal half of the protein. A combination of site-directed and bisulfite mutagenesis was used to alter the glycine-rich region of the N-terminal putative nucleotide-binding sequence G15KGGVGKTS23. Four mutant proteins (G18----D, G18----R, G20----S, and T22----I) were analyzed. Strains bearing the mutated plasmids were all arsenite sensitive and were unable to extrude arsenite. Each purified mutant protein lacked oxyanion-stimulated ATPase activity and ATP binding. These results suggest that the N-terminal sequence is part of a nucleotide-binding domain required for catalysis.

Molecular analysis of an ATP-dependent anion pump.

The plasmid-borne arsenical resistance operon encodes an ATP-driven oxyanion pump for the extrusion of the oxyanions arsenite, antimonite and arsenate from bacterial cells. The catalytic component of the pump, the 63 kDa ArsA protein, hydrolyses ATP in the presence of its anionic substrate antimonite (SbO2-). The ATP analogue 5'-p-fluorosulphonylbenzoyladenosine was used to modify the ATP binding site(s) of the ArsA protein. From sequence analysis there are two potential nucleotide binding sites. Mutations were introduced into the N-terminal site. Purified mutant proteins were catalytically inactive and incapable of binding nucleotides. Conformational changes produced upon binding of substrates to the ArsA protein were investigated by measuring the effects of substrates on trypsin inactivation. The hydrophobic 45.5 kDa ArsB protein forms the membrane anchor for the ArsA protein. The presence of the ArsA protein on purified inner membrane can be detected immunologically. In the absence of the arsB gene no ArsA is found on the membrane. Synthesis of the ArsB protein is limiting for formation of the pump. Analysis of mRNA structure suggests a potential translational block to synthesis of the ArsB protein. Northern analysis of the ars message demonstrates rapid degradation of the mRNA in the arsB region.

Molecular characterization of an anion pump. The arsA gene product is an arsenite(antimonate)-stimulated ATPase.

The products of the arsenical resistance operon of resistance plasmid R733 form an efflux system for arsenicals. Detoxification results from active efflux of the oxyanions, preventing their concentration from reaching toxic levels. The largest polypeptide encoded by the ars operon was purified. From N-terminal sequencing the purified protein, termed the ArsA protein, was shown to correspond to the product of the arsA gene. The purified protein was demonstrated to bind ATP by two methods. First, a photoadduct of the protein with [alpha-32P]ATP was formed by irradiation at 254 nm. Second, the purified protein bound a fluorescent ATP analogue, 2',3'-o-(2,4,6)trinitrophenyl ATP, with a half-maximal affinity of 2 microM. By both assays competition was observed with ATP or ADP, but not with AMP, GTP, CTP, or UTP. In both nucleotide binding assays, Mg2+ was required, but neither arsenite nor antimonate had any affect. In contrast, the ArsA protein exhibited an ATPase activity which was dependent on the presence of arsenite or antimonate. The results suggest that the ArsA protein is the catalytic subunit of an oxyanion-translocating ATPase.