Crystal structure of the tumor-promoter okadaic acid bound to protein phosphatase-1.

Protein phosphatase-1 (PP1) plays a key role in dephosphorylation in numerous biological processes such as glycogen metabolism, cell cycle regulation, smooth muscle contraction, and protein synthesis. Microorganisms produce a variety of inhibitors of PP1, which include the microcystin class of inhibitors and okadaic acid, the latter being the major cause of diarrhetic shellfish poisoning and a powerful tumor promoter. We have determined the crystal structure of the molecular complex of okadaic acid bound to PP1 to a resolution of 1.9 A. This structure reveals that the acid binds in a hydrophobic groove adjacent to the active site of the protein and interacts with basic residues within the active site. Okadaic acid exhibits a cyclic structure, which is maintained via an intramolecular hydrogen bond. This is reminiscent of other macrocyclic protein phosphatase inhibitors. The inhibitor-bound enzyme shows very little conformational change when compared with two other PP1 structures, except in the inhibitor-sensitive beta12-beta13 loop region. The selectivity of okadaic acid for protein phosphatases-1 and -2A but not PP-2B (calcineurin) may be reassessed in light of this study.

The adsorption of microcystin-LR by natural clay particles.

The microcystin cyanobacterial hepatotoxins represent an increasingly severe global health hazard. Since microcystins are found world wide in drinking water reservoirs concern about the impact on human health has prompted investigations into remedial water treatment methods. This preliminary study investigates the scavenging from water of microcystin-LR by fine-grained particles known to have a high concentration of the clay minerals kaolinite and montmorillonite. The results show that more than 81% of microcystin-LR can be removed from water by clay material. Thus, microcystin-LR is indeed scavenged from water bodies by fine-grained particles and that this property may offer an effective method of stripping these toxins from drinking water supplies.

Inhibition of protein phosphatase-1 by clavosines A and B. Novel members of the calyculin family of toxins.

Site-directed mutagenesis was used to investigate the mechanism of interaction between the catalytic subunit of human protein phosphatase-1 (PP-1cgamma) and members of the calyculin family of toxins. Clavosines A and B are related to calyculins but are glycosylated with a trimethoxy rhamnose group. We provide experimental evidence implicating Tyr-134 as an important residue in PP-1cgamma that mediates interactions with the calyculins. Mutation of Tyr-134 to Phe, to prevent hydrogen bond formation, resulted in a slight increase in sensitivity of PP-1cgamma to clavosines A and B and calyculin A. In contrast, a Y134A mutant was 10-fold less sensitive to inhibition by all three inhibitors. The greatest effect on inhibition was found by substituting an Asp for Tyr-134 in the phosphatase. Clavosine B inhibited PP-1cgamma Y134D with a 310-fold decrease in potency. Clavosine A and calyculin A were also markedly poorer inhibitors of this mutant. These results suggest that a hydrogen bond between Tyr-134 and the calyculins is unlikely to be essential for inhibitor binding to the phosphatase. The clavosines and calyculin A were tested for their ability to inhibit other mutants of PP-1cgamma (including Ile-133, Val-223, and Cys-291). Our mutagenesis studies provide an experimental basis for assessing models of calyculin binding found in the literature (Lindvall, M. K., Pihko, P. M., and Koskinen, A. M. (1997) J. Biol. Chem. 272, 23312-23316; Gupta, V., Ogawa, A. K., Du, X., Houk, K. N., and Armstrong, R. W. (1997) J. Med. Chem. 40, 3199-3206; Gauss, C. M., Sheppeck, I. J., Nairn, A. C., and Chamberlain, R. (1997) Bioorg. Med. Chem. 5, 1751-1773). A new model for clavosine and calyculin A binding to PP-1c is presented that is consistent with previous structure-function experiments and which accommodates key structural features of the clavosines, including the novel rhamnose moiety.

Mutation of the toxin binding site of PP-1c: comparison with PP-2B.

The catalytic cores of PP-1c and PP-2B (calcineurin) are structurally conserved. However, PP-2B is resistant to inhibition by toxins of the okadaic acid and cyclic peptide classes, while PP-1c is potently inhibited. Molecular docking of the structure of microcystin-LR onto the catalytic core of PP-2B identified residues that may be responsible for blocking access of toxins to the catalytic site. Amino acids in PP-1c were substituted with these PP-2B residues to investigate their contribution to PP-2B toxin resistance. Mutants of PP-1c were also produced to test the importance of hydrophobic interactions to toxin binding. Our results suggest that different classes of toxin inhibitors interact with the same hydrophobic side chains of PP-1c through different mechanisms. Substitution of amino acids in PP-1c with PP-2B residues demonstrated no highly significant changes in toxin inhibition. We hypothesize that an interaction outside the catalytic core causing the L7 loop of PP-2B to block the catalytic site may be responsible for PP-2B resistance to toxins.

Molecular mechanisms underlying he interaction of motuporin and microcystins with type-1 and type-2A protein phosphatases.

Heptapeptide microcystin and pentapeptide motuporin (nodularin-V) are equipotent inhibitors of type-1 and type-2A protein phosphatase catalytic subunits (PP-1c and PP-2Ac). Herein we describe elucidation of the molecular mechanisms involved in the interaction of these structurally similar hepatotoxins with PP-1c/PP-2Ac and identification of an important functional difference between their mode of interaction with these enzymes. Microcystin-LR, microcystin-LA, and microcystin-LL were found to interact with PP-2Ac and PP-1c by a two-step mechanism involving rapid binding and inactivation of the protein phosphatase (PPase) catalytic subunit, followed by a slower covalent interaction (within hours). Covalent adducts comprising PPase-toxin complexes were separated from free PPase by C-18 reverse-phase liquid chromatography, thus allowing the time course of covalent adduct formation to be quantitated. In contrast to microcystins, motuporin (nodularin-V) and nodularin-R were unable to form covalent complexes with either PP-1c or PP-2Ac even after 96 h incubation. Specific reduction of microcystin-LA to dihydromicrocystin-LA abolished the ability of the toxin to form a covalent adduct with PP-2Ac. Specific methyl esterification of the single Glu residue in microcystin-LR rendered this toxin inactive as a PPase inhibitor and abolished subsequent formation of a covalent adduct. Our data indicate that inactivation of PP-2Ac/PP-1c by microcystins precedes covalent modification of the PPases via a Michael addition reaction between a nucleophilic phosphatase residue and Mdha in the heptapeptide toxin. In contrast, following rapid inactivation of PP-2Ac/PP-1c by motuporin, the equivalent N-methyldehydrobutyrine residue in this toxin is unreactive and does not form a covalent bond with the PPases. These results are consistent with structural data for (i) the NMR solution structures of microcystin-LR and motuporin, which indicate a striking difference in the relative positions of their corresponding dehydroamino acids in the toxin peptide backbone, and (ii) X-ray crystallographic data on an inactive complex between PP-1c and microcystin-LR, which show a covalent bond between Cys-273 and the bound toxin.

Identification and characterization of hydrophobic microcystins in Canadian freshwater cyanobacteria.

Hepatotoxic microcystins produced by cyanobacteria in freshwater lakes represent a significant health hazard to humans and agricultural livestock. Liquid chromatography (LC)-linked protein phosphatase (PPase) bioassay analysis of blooms of Microcystis aeruginosa produced in a Canadian drinking water lake identified several PPase inhibitors with significantly greater hydrophobicity than microcystin-LR, based on their retention time on C18 reverse phase LC columns. Seven PPase inhibitors were purified to homogeneity by bioassay-guided fractionation involving Sephadex LH-20 chromatography and two-step reverse phase at pH 6.5 and 2.0. One of the PPase inhibitors, isolated in a final yield of 1.5 micrograms/g lyophilized cyanobacteria, was identified as microcystin-LL by amino acid analysis and mass spectrometry. A further PPase inhibitor (20 ng/g cyanobacteria) was identified as microcystin-LL but with D-Ala replaced by an unknown amino acid. Four PPase inhibitors (< 20 ng/g cyanobacteria) were characterized by amino acid analysis and identified as microcystin-LV, -LM, -LF and -LZ (where Z represents an unknown hydrophobic amino acid). A further microcystin was also identified (< 10 ng/g cyanobacteria) in which arginine was apparently absent. The biological activity of the seven microcystins as inhibitors of the catalytic subunit of protein phosphatase-1 (PP-1c) was compared with microcystin-LR and motuporin (a hydrophobic analogue of nodularin). All of the compounds inhibited PP-1c with IC50 values of 0.06-0.4 nM, consistent with their identification as microcystins. These findings further demonstrate the applicability of a sensitive PPase bioassay for the identification of variant microcystins in the natural environment.

Chemical and biological evidence links microcystins to salmon 'netpen liver disease'.

Evidence is presented that links microcystins to a severe liver disease that occurs in Atlantic salmon that are netpen-reared in coastal British Columbia. Liquid chromatography-linked protein phosphatase bioassay analysis of extracts of liver tissue taken from Atlantic salmon afflicted with netpen liver disease showed the presence of an inhibitor of protein phosphatase that was chromatographically indistinguishable from microcystin-LR. Analysis of liver tissue from healthy control fish showed a complete absence of microcystin-LR. Intraperitoneal injection of microcystin-LR into healthy Atlantic salmon re-created the pathologic changes of netpen liver disease, including diffuse necrosis and hepatic megalocytosis.

Quantification of diarrhetic shellfish toxins and identification of novel protein phosphatase inhibitors in marine phytoplankton and mussels.

Liquid chromatography (LC)-linked protein phosphatase 1/2A (PP-1/PP2A) bioassay was used to quantitatively identify diarrhetic shellfish toxins in marine phytoplankton (cultured and natural assemblages) and commercially available mussels. Using this approach, multiple protein phosphatase inhibitor profiles of varying composition were found in diarrhetic mussels from Holland and Canada. Based on LC elution positions and relative activity versus PP-1 and PP-2A, at least six inhibitors distinct from known diarrhetic shellfish toxins were identified and termed mussel phosphatase inhibitor (MPI) 19,22,23,25,33 and 42. The levels of these inhibitors, in okadaic acid equivalent units, varied from 100 pg to 3350 ng per g shellfish tissue. The combined levels of PP-1/2A inhibitors in all instances superseded that of okadaic acid/dinophysistoxin-1 and may contribute to the diarrhetic shellfish toxin profile of the contaminated mussels. The efficacy of LC-protein phosphatase bioassay was established for cultured phytoplankton where picogram levels of okadaic acid could be detected from microgram extracts of Prorocentrum lima. Analyses of plankton net tows from estuarine mussel culture sites in Eastern Canada revealed a heterogeneous population of protein phosphatase inhibitors, with dinophysistoxin-1 being most prevalent. This toxin was predominant for at least 2 months in mussel populations in the immediate vicinity of plankton sampling sites. The results are consistent with a hypothetical model in which marine bacteria, cyanobacteria and dinoflagellates combine to produce a variety of protein phosphatase inhibitors effective against signal transduction pathways in higher eukaryotes.

Ubiquitin gene expression: response to environmental changes.

It has previously been shown that the yeast ubiquitin genes UBI1, 2 and 3 are strongly expressed during the log-phase of batch culture growth, whereas the UBI4 gene is weakly expressed. We found that heat shock, treatment with DNA-damaging agents, starvation, and the feeding of starved cells all transiently induced UBI4. These results suggest that UBI4 is induced whenever a change in culture conditions dictates a dramatic shift in cellular metabolism, and that UBI4 expression returns to lower levels once cellular metabolism has adapted to the new conditions. In contrast, all of the treatments tested, except starvation, transiently repressed the UBI1, 2 and 3 genes. Although starvation also repressed UBI1, 2 and 3 its effect was not transient, and expression only recovered upon the addition of fresh media. These results, together with others presented here, suggest that high levels of UBI1, 2 and 3 expression are dependant upon ongoing cell growth, and that treatments which slow or stop growth repress their expression.

Inhibition of protein phosphatases-1 and -2A with acanthifolicin. Comparison with diarrhetic shellfish toxins and identification of a region on okadaic acid important for phosphatase inhibition.

Acanthifolicin (9,10-epithio-okadaic acid from Pandoras acanthifolium) inhibited protein phosphatase-1 (PP1) similarly to okadaic acid (IC50 = 20 nM and 19 nM, respectively) but was slightly less active against protein phosphatase-2A (PP2A) (IC50 = 1 nM and 0.2 nM, respectively). Methyl esterification of acanthifolicin sharply reduced its activity. PP2A was inhibited with an IC50 = 5.0 microM, whilst PP1 was inhibited less than 10% at 250 microM toxin. Okadaic acid methyl ester was similarly inactive whereas dinophysistoxin-1 (35-methyl okadaic acid) inhibited PP1/2A almost as potently as okadaic acid. Pure acanthifolicin/okadaic acid methyl ester may be useful as specific inhibitors of PP2A at 1-10 microM concentrations in vitro and perhaps in vivo. The data also indicate that a region on these toxins important for PP1/2A inhibition comprises the single carboxyl group.

The interactions of cytochrome c and porphyrin cytochrome c with cytochrome c oxidase. The resting, reduced and pulsed enzymes.

Cytochrome c oxidase forms tight binding complexes with the cytochrome c analog, porphyrin cytochrome c. The behaviour of the reduced and pulsed forms of the oxidase with porphyrin cytochrome c have been followed as functions of ionic strength; this behaviour has been compared with that of the resting oxidase [Kornblatt, Hui Bon Hoa and English (1984) Biochemistry 23, 5906-5911]. All forms of the cytochrome oxidase studied bind one porphyrin cytochrome c per 'functional' cytochrome oxidase (two heme a); it appears as though porphyrin cytochrome c and cytochrome c compete for the same site on the oxidase. The resting enzyme binds cytochrome c 8 times more strongly than porphyrin cytochrome c; the reduced enzyme, in contrast, binds the two with almost equal affinity. In all three cases, resting, pulsed and reduced, the heme-to-porphyrin distance is estimated to be about 3 nm. The tight-binding complexes formed between cytochrome oxidase and porphyrin cytochrome c can be dissociated by salt. Debye-Hückel analysis of salt titrations indicate that the resting enzyme and the reduced enzyme are similar in that the product of the interaction charges on the two proteins is about -14. The product of the charges for the pulsed enzyme is -25, indicating that on average another positive and negative charge take part in the interaction of the two proteins. While there is one tight binding site for cytochrome c per two heme a, cytochrome c is able to 'communicate' with four heme a. In the absence of cytochrome c, electron transfer from tetramethylphenylenediamine to the oxidase to oxygen results in the conversion of the resting form to the 'oxygenated'; in the presence of cytochrome c, the same electron transfer results in the appearance of the 'pulsed' form. Cytochrome c titrations of the enzyme show that a ratio of only one cytochrome c to four heme a is sufficient to convert all the oxidase to the 'pulsed' form. Porphyrin cytochrome c, like cytochrome c, catalyzes the same conversion with the same stoichiometry. The binding data and salt effects indicate that major structural alterations occur in the oxidase as it is converted from the resting to the partially reduced and subsequently to the pulsed form.