Interactions of serine proteinases with pNiXa, a serpin of Xenopus oocytes and embryos.

In a previous study, kinetic assays showed that pNiXa, an Ni(II)-binding serpin of Xenopus oocytes and embryos, strongly inhibits bovine chymotrypsin, weakly inhibits porcine elastase, and does not inhibit bovine trypsin. In this study, analyses by SDS-PAGE and gelatin zymography showed that an SDS-resistant complex is formed upon the interaction of pNiXa with bovine chymotrypsin. No such pNiXa-enzyme complex was detected after pNiXa interactions with porcine elastase, bovine trypsin, or human cathepsin G. The major products of pNiXa cleavage by the four proteinases were partially sequenced by Edman degradation. The cleavage products were also tested by immunoblotting with an antibody to the His-cluster of pNiXa, and by radio-blotting with 63Ni(II). These assays showed that chymotrypsin and elastase cleave pNiXa at the P1-P1 (Thr-Lys) peptide bond near the C-terminus, while trypsin and cathepsin G cleave pNiXa at specific peptide bonds near the N-terminus, within an interesting 26-residue segment, rich in Lys and Gln, that separates the His-cluster of pNiXa from the rest of the molecule. The segment lacks homology to other serpins, but resembles a domain of Xenopus POU3 transcription factor. This study identifies the specific sites for interactions of four serine proteinases with pNiXa, indicates that pNiXa inhibition of chymotrypsin involves a serpin-like mechanism, and shows that 63Ni(II)-binds to the His-cluster of pNiXa.

Characterization of pNiXa, a serpin of Xenopus laevis oocytes and embryos, and its histidine-rich, Ni(II)-binding domain.

A Ni(II)-binding serpin, pNiXa, is abundant in Xenopus oocytes and embryos. Kinetic assays show that purified pNiXa strongly inhibits bovine alpha-chymotrypsin (Ki = 3 mM), weakly inhibits porcine elastase (K1 = 0.5 microM), and does not inhibit bovine trypsin. The reversible, slow-binding inhibition of alpha-chymotrypsin by pNiXa is unaffected by Ni(II). Ovochymase in egg exudates is inhibited by pNiXa, but to a limited extent, even at high pNiXa concentrations. An octadecapeptide that models the His-rich domain (-HRHRHEQQGHHDSAKHGH-) of pNiXa forms six-coordinate, octahedral Ni(II)-complexes when the N-terminus is acetylated, and a square-planar Ni(II)-complex when the N-terminus is unblocked. Spectroscopy reveals two distinct types of octahedral Ni(II)-coordination to the N-acetylated octadecapeptide, involving, respectively, 3-4 and 5-6 imidazole nitrogens; the octadecapeptide undergoes partial, reversible precipitation in pH- and Ni(II)-dependent fashion, suggesting an insoluble, Ni(II)-coupled (Hx)n-dimer. Such (Hx)n-peptide interaction is confirmed by an enzyme-linked biotinavidin assay with N-biotin-KHRHRHE-amide and N-acetyl-KHRHRHE-resin beads, which become coupled after adding Ni(II) or Zn(II). H2O2 oxidation of 2'-deoxyguanosine to mutagenic 8-hydroxy-2'-deoxyguanosine is enhanced by the octahedral Ni(II)-octadecapeptide complex, although the effect is more intense with the square-planar Ni(II)-octadecapeptide complex. Immunoperoxidase staining of whole mounts with pNiXa antibody shows that pNiXa is distributed throughout gastrula-stage embryos and is localized during organogenesis in the brain, eye, spinal cord, myotomes, craniofacial tissues, and other sites of Ni(II)-induced anomalies. Patterns of pNiXa staining are similar in controls and Ni(II)-exposed embryos. Binding of Ni(II) to pNiXa may cause embryotoxicity by enhancing oxidative reactions that produce tissue injury and genotoxicity. Although the natural target proteinases for pNiXa inhibition have not been established, pNiXa may be an important regulator of proteolysis during embryonic development.

Cyclic AMP is a requisite messenger in the action of big PTTH in the prothoracic glands of pupal Manduca sexta.

Prothoracicotropic hormone (PTTH), a peptide produced by the insect brain, stimulates the prothoracic glands to secrete ecdysteroids. The big form of this peptide (25.5 kDa) has been postulated to act through cyclic AMP in larval Manduca sexta, but the role of the cyclic nucleotide in the action of PTTH in pupal glands has been less clear. Results of the present study indicate that PTTH-stimulated ecdysteroid secretion and protein phosphorylation by glands removed from pupal Manduca sexta are blocked by two inhibitors of cAMP-dependent protein kinase: Rp-cAMPS, an antagonist of cAMP binding to the regulatory subunit of the kinase, and H-89, an inhibitor of the catalytic subunit of the kinase. Further, PTTH stimulates significant accumulation of cAMP in pupal glands, although less than that previously seen in PTTH-stimulated larval glands. Cyclic AMP-dependent protein kinase is found in cytoplasmic and membrane-associated glandular subfractions, as measured by incorporation of [32P]8-N3cAMP into the regulatory subunit of the kinase. PTTH enhances cytoplasmic cAMP content and appears to increase the amount of cAMP bound to a cytoplasmic type II regulatory subunit of cAMP-dependent protein kinase. The results indicate that cAMP plays a requisite role in PTTH action in pupal glands, thus arguing in favor of a uniform mechanism of action for the peptide during Manduca development.

Xenopus lipovitellin 1 is a Zn(2+)- and Cd(2+)-binding protein.

This report discusses the identification of a Zn(2+)- and Cd(2+)-binding protein of Xenopus laevis that is abundant in vitellogenic oocytes and in embryos from fertilization to stage 46. Oocyte or embryo homogenates were fractionated by SDS-PAGE, blotted onto nitrocellulose, and probed with 65Zn2+ or 109Cd2+. The resulting autoradiograms showed binding of both radionuclides to a protein, designated pCdZn. Freon extraction of oocyte and embryo homogenates showed pCdZn to be a yolk protein. When pCdZn was isolated from oocyte homogenates by ammonium sulfate precipitation, delipidation, and chromatography, it co-purified with lipovitellin 1. The amino acid composition of pCdZn closely resembled the reported composition of lipovitellin 1 and the molecular weight of purified pCdZn (approximately 115 kD) corresponded to reported values for lipovitellin 1 (111-121 kD). Amino acid sequence analyses of five peptides derived from pCdZn yielded 94% identity to the reported sequence of lipovitellin 1, deduced from the DNA sequence of the Xenopus vitellogenin A2 precursor gene. Based on these findings, pCdZn was identified as lipovitellin 1. This study suggests that lipovitellin 1 is the major storage protein for zinc in mature oocytes and developing embryos of Xenopus laevis.

The 40 kDa 63Ni(2+)-binding protein (pNiXc) on western blots of Xenopus laevis oocytes and embryos is the monomer of fructose-1,6-bisphosphate aldolase A.

A Ni(2+)-binding protein (pNiXc, 40 kDa), present in Xenopus laevis oocytes and embryos, was isolated from mature oocytes by chromatography on DEAE-cellulose and cellulose phosphate, followed by FPLC on Ni-iminodiacetate-Agarose, or reverse-phase HPLC on a C-4 column. Size-exclusion HPLC showed that intact pNiXc is approximately 155 kDa, consistent with tetrameric structure. After cleavage with Lys-C proteinase or cyanogen bromide, six peptides were separated by HPLC and sequenced by Edman degradation, providing sequence data for 83 residues. Data-base search showed similarity of pNiXc to eukaryotic aldolases, with 96% identity to human aldolase A. pNiXc demonstrated aldolase activity with fructose 1,6-bisphosphate as substrate (Km, 30 microM Vmax 26 mumol min-1 mg-1); the aldolase activity was inhibited non-competitively by Cu2+, Cd2+, Co2+, or Ni2+. Equilibrium dialysis showed high affinity binding (Kd, 7 microM) of 1 mole of Ni per mole of 40 kDa subunit. Based on metal-blot competition assays, the abilities of metals to compete with 63Ni2+ for binding to pNiXc were ranked: Cu2+ >> Zn2+ > Cd2+ > Co2+. This study identifies pNiXc as the monomer of fructose-1,6-bisphosphate aldolase A, and raises the possibility that aldolase A is a target enzyme for metal toxicity.

Lipovitellin 2 beta is the 31 kD Ni(2+)-binding protein (pNiXb) in Xenopus oocytes and embryos.

An Ni(2+)-binding protein (pNiXb, 31 kD) present in mature Xenopus laevis oocytes and in embryos from fertilization in N/F stage 42, was isolated and characterized. After oocytes or embryos were fractionated by PAGE, electroblotted onto nitrocellulose, and probed with 63Ni2+, pNiXb was detected by autoradiography. pNiXb, a yolk protein located in the embryonic gut, was purified from yolk platelets by ammonium sulfate precipitation, delipidation, gel filtration chromatography, and HPLC analysis. During these steps, pNiXb copurified with lipovitellin 2. The N-terminal sequence of purified pNiXb exactly matched that of Xenopus lipovitellin 2 beta, deduced from the DNA sequence of the Xenopus vitellogenin A2 precursor gene. Since pNiXb and lipovitellin 2 beta agree in N-terminal sequence, amino acid composition, and apparent molecular weight, they appear to be identical. Based on a metal-blot competition assay, the abilities of metal ions to compete with 63Ni2+ for binding to pNiXb were ranked: Zn2+ approximately Cu2+ approximately Co2+ > Cd2+ approximately Mn2+ > Sn2+. This study shows that Xenopus lipovitellin 2 beta is a metal-binding protein in vitro, and raises the possibility that it may function similarly in vivo.

Developmental changes in cyclic AMP-dependent protein kinase associated with increased secretory capacity of Manduca sexta prothoracic glands.

In Manduca sexta, basal and PTTH-stimulated secretion of ecdysteroids by prothoracic glands in vitro increases from days 1 to 4 of the fifth larval stage. Glandular content of cAMP-dependent protein kinase was analyzed to determine if the enzyme changes in concert with increased secretory response. Photoaffinity labeling with [32P]8-N3 cAMP revealed a 55-kDa cAMP-binding protein characteristic of the regulatory subunit of type-II cAMP-dependent protein kinase (RII). It appears that RII is one of a limited number of cellular proteins that is phosphorylated in the presence of [gamma-35S]ATP; the thiophosphorylated protein and the photoaffinity-labeled regulatory subunit possess the same M(r) and pI, and thiophosphorylation is blocked by mammalian cAMP-dependent protein kinase inhibitor. From days 1 to 4 of the fifth instar, glandular content of RII increases in conjunction with increased ecdysteroid secretory capacity. Application of JH analog on day 1 significantly inhibits the observed increase in RII. Catalytic subunit activity does not change from days 1 to 4 of the fifth instar, nor does cellular content of a 34-kDa protein previously shown to be phosphorylated in response to PTTH. While it is unlikely that increased content of RII is solely responsible for enhanced ecdysteroid secretion by the prothoracic glands, it may serve as a convenient marker for investigating the mechanism by which steroidogenic capacity is regulated.