The arsonomethyl analogue of 3-phosphoglycerate

Chemical synthesis of the organoarsenical antibiotic arsinothricin

We report two routes of chemical synthesis of arsinothricin (AST), the novel organoarsenical antibiotic. One is by condensation of the 2-chloroethyl(methyl)arsinic acid with acetamidomalonate, and the second involves reduction of the N-acetyl protected derivative of hydroxyarsinothricin (AST-OH) and subsequent methylation of a trivalent arsenic intermediate with methyl iodide. The enzyme AST N-acetyltransferase (ArsN1) was utilized to purify l-AST from racemic AST. This chemical synthesis provides a source of this novel antibiotic for future drug development.

Arsinothricin is prepared from 2-chloroethyl(methyl)arsinic acid or by reduction of N-acetyl protected derivative of hydroxyarsinothricin and methylation with methyl iodide.

Semisynthesis of the Organoarsenical Antibiotic Arsinothricin

Arsinothricin [AST (1)], a new broad-spectrum organoarsenical antibiotic, is a nonproteinogenic analogue of glutamate that effectively inhibits glutamine synthetase. We report the chemical synthesis of an intermediate in the pathway to 1, hydroxyarsinothricin [AST-OH (2)], which can be converted to 1 by enzymatic methylation catalyzed by the ArsM As(III) S-adenosylmethionine methyltransferase. This is the first report of semisynthesis of 1, providing a source of this novel antibiotic that will be required for future clinical trials.

Proteomic analysis of Pteris vittata fronds: two arbuscular mycorrhizal fungi differentially modulate protein expression under arsenic contamination

Arbuscular mycorrhizae (AM) are the most widespread mutualistic symbioses between the roots of most land plants and a phylum of soil fungi. AM are known to influence plant performance by improving mineral nutrition, protecting against pathogens and enhancing resistance or tolerance to biotic and abiotic stresses. The aim of this study was to investigate the frond proteome of the arsenic hyperaccumulator fern Pteris vittata in plants that had been inoculated with one of the two AM fungi (Glomus mosseae or Gigaspora margarita) with and without arsenic treatment. A protective role for AM fungi colonisation in the absence of arsenic was indicated by the down-regulation of oxidative damage-related proteins. Arsenic treatment of mycorrhizal ferns induced the differential expression of 130 leaf proteins with specific responses in G. mosseae- and Gi. margarita-colonised plants. Up-regulation of multiple forms of glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate kinase, and enolase, primarily in G. mosseae-inoculated plants, suggests a central role for glycolytic enzymes in arsenic metabolism. Moreover, a putative arsenic transporter, PgPOR29, has been identified as an up-regulated protein by arsenic treatment.

Anion activation of 3-phosphoglycerate kinase requires domain closure

3-Phosphoglycerate kinase is a typical two-domain "hinge-bending" enzyme, which is known to be regulated by multivalent anions. Here a relationship between this regulation and the hinge-bending domain closure is proposed on the basis of enzyme kinetic analysis and molecular modeling. Activation of the pig muscle enzyme at low concentrations and inhibition at high concentrations of various anionic analogues of the substrate 3-phosphoglycerate or of the nonsubstrate metal-free ATP are described by a two-site model assuming separate sites for activation and inhibition, respectively. Kinetic experiments with various pairs of analogues suggest the presence of a common site for activation by all effectors, separate from the catalytic site for 3-phosphoglycerate; and a common site for inhibition, except for metal-free ATP, identical with the catalytic site of 3-phosphoglycerate. An additional inhibiting site for all of the anions investigated, including metal-free ATP, is also proposed. A similar two-site model can describe activation of the enzyme by a large excess of each substrate; here the ligand binds to the catalytic site as a substrate and to the regulatory site as an activator. Activation is exerted not only by the physiological substrate, 3-phophoglycerate, but also by a synthetic weak substrate. The activity in the reaction with 3-phosphoglycerate and MgATP is greatly enhanced by the simultaneous presence of the weak substrate. This finding clearly proves the existence of a regulatory site, separate from the catalytic site. This regulatory site, however, may only exist in the catalytically competent closed conformation of the enzyme, as indicated by molecular modeling. Docking of the regulator anions into the known X-ray structures of the enzyme revealed the appearance of an anion binding site between the two domains, including the invariant residues of Lys-215 (C-domain) and of Arg-65 among other residues of the basic cluster (N-domain), as a consequence of the large-scale substrate-induced conformational change that leads to domain closure.

The transport of acidic amino acids and their analogues across monolayers of human intestinal absorptive (Caco-2) cells in vitro

The X-AG system, a sodium-dependent, acidic amino-acid transport system has been implicated in the transport of L-aspartate and L-glutamate across monolayers of human Caco-2 cells, an in vitro model of intestinal absorption. This system, which shares many properties with the L-glutamate carrier present in the human jejunum, is highly saturable (> 95% at 50 microM), vectorial (apical-to-basolateral >> basolateral-to-apical) and sodium-, pH- and temperature-dependent. L-Aspartate was also transported against a 10-fold reverse concentration gradient. These data are consistent with a major (saturable) carrier-mediated pathway superimposed onto a minor non-saturable (diffusional) pathway. The carrier has an absolute sodium-dependence and the Michaelis constants for the sodium-dependent transport component (Km) for L-aspartate and L-glutamate were 56 +/- 3 microM and 65 +/- 6 microM, respectively. Cross-inhibition studies showed that strong interaction with the carrier was limited to close analogues of the natural substrates. Potent inhibitors included L-aspartate, D-aspartate (Ki, 70 microM), L-glutamate (Ki 180 microM) and threo-beta-hydroxy-DL-aspartate (Ki, 55 microM), while partial inhibitors included alpha-methyl-DL-aspartate, D-glutamate, L-asparagine, L-proline and L-alanine. Replacement of the side-chain -COO- group (aspartate) with -SO-3 (L-cysteate, Ki, 65 microM) or -(H)P(O)O- (DL-3-(hydroxyphosphoryl)alanine, Ki, 60 microM) maintained strong interaction with the carrier while -As(O)(OH)O- (DL-3-arsonoalanine, Ki, 1100 microM) and -P(O)(OH)O- (DL-3-phosphonoalanine, Ki, 3270 microM) were much more weakly bound, with the larger, but probably less ionised, arsono analogue being more tightly bound than the phosphono compound. The corresponding analogues of glutamate (homologous extension of the methylene chain) showed negligible interaction. We conclude that Caco-2 monolayers are a relevant experimental model for the study of the transport of acidic amino acids and their analogues in man.

Arsenite release on enzymic transformation of arsonomethyl substrate analogues: a potentially lethal synthesis by glycerol-3-phosphate dehydrogenase

The isosteric arsenical analogue of glycerol 3-phosphate, 3,4-dihydroxybutylarsonic acid, is a good substrate for rabbit muscle glycerol-3-phosphate dehydrogenase. Its oxidation is accompanied by release of arsenite. This release seems to be due to a spontaneous elimination of arsenite by 3-oxoalkylarsonic acids, as it is also observed in (1) the oxidation of 3-hydroxypropylarsonic acid by yeast alcohol dehydrogenase, (2) treatment of 3,4-dihydroxybutylarsonic acid with periodate and (3) nonenzymic transamination of the glutamate analogue 2-amino-4-arsonobutyric acid. Enzymic formation of 3-oxoalkylarsonic acids in cells can therefore be lethal, as arsenite is poisonous to most organisms because of its high affinity for dithiols such as dihydrolipoyl groups.

Enolase and the arsonomethyl analogue of 2-phosphoglycerate

(RS)-3-Arsono-2-(hydroxymethyl)propionic acid was synthesized by the action of alkaline arsenite on 3-bromo-2-(bromomethyl)propionic acid. It is a substrate for yeast enolase (EC 4.2.1.11) with a Km of 6.5 mM (for 2-phospho-D-glycerate Km = 0.08 mM). The catalytic constant of the enzyme with the arsonomethyl analogue is 230 times lower than with 2-phosphoglycerate.

Modelling of substrate binding to 3-phosphoglycerate kinase with analogues of 3-phosphoglycerate

Two short analogues of 3-phosphoglycerate, -OOC-CHOH-CH2-O-PO32-, phosphonolactate, (-OOC-CHOH-CH2-PO32-) and arsonolactate (-OOC-CHOH-CH2-AsO32-) have been tested with 3-phosphoglycerate kinase. None of these served as substrate for the kinase reaction, unlike the previously studied [Orr, G. A. & Knowles, J. R. (1974) Biochem. J. 141, 721-723] analogues -OOC-CHOH-CH2-CH2-PO32- and -OOC-CHOH-CH2-CH2-AsO32-, which are isosteric with 3-phosphoglycerate. Thus, a decrease in the substrate size and the accompanying stereochemical changes cannot be tolerated by the catalytic mechanism. Instead, both analogues acted as relatively poor competitive inhibitors with respect to both 3-phosphoglycerate and MgATP. AT pH 8.5 and 20 degrees C, the inhibitory constants (Ki) of phosphonolactate and arsnolactate against both substrates are 17 +/- 5 mM and 30 +/- 7 mM, respectively. Surprisingly, however, both analogues proved to be more effective than either 3-phosphoglycerate or its isosteric analogues in protecting the enzyme against modification of its fast-reacting thiols. This comparison suggests that the shorter analogues bind differently, and that the catalytic mechanism demands a precise fitting of the -CH2-O-PO32- segment of the substrate.

Synthesis of 2-aminoethylarsonic acid. A new synthesis of primary

2-Aminoethylarsonic acid was prepared from 2-choloethylarsonic acid. The route constitutes a new procedure for making primary amines from haloalkanes; chloride was displaced by treatment with 2-aminoethanol at 70 degrees C, and the product was converted into the required primary amine by treatment with periodate.

The phosphate group of 3-phosphoglycerate accounts for conformational changes occurring on binding to 3-phosphoglycerate kinase. Enzyme inhibition and thiol reactivity studies

Steady-state kinetic study of the inhibition of 3-phosphoglycerate kinase reaction by the substrate analogues D-glycerol 3-phosphate, 2-phosphoglycolate, tartronate and malonate revealed competition with respect to 3-phosphoglycerate. D-Glycerate had no detectable inhibitory effect. The data indicate that (a) the phosphate of 3-phosphoglycerate plays an essential role in the formation of its complex with the enzyme and, taking into account the relatively strong binding of 3-phosphoglycerate, (b) the two charged groups of the substrate might cause a synergic interaction with the protein. The carboxyl-lacking D-glycerol 3-phosphate is a non-competitive inhibitor with respect to MgATP, while all the investigated carboxyl-containing inhibitors compete for MgATP binding. The inhibitory analogues of 3-phosphoglycerate reduce the reactivity of both the two fast-reacting and the five slow-reacting thiol groups of the enzyme molecule. In the case of the fast-reacting thiols the effect is specifically associated with the presence of a ligand's phosphate group. Similarly mainly the phosphate-containing nucleotides and analogues slow down significantly the reaction rate of the fast-reacting thiols, while adenosine is less effective and the competitive inhibitor adenine has no effect at all. MgADP has an especially dramatic effect as compared to MgATP, in line with the known X-ray structural data. The fast-reacting thiols are of particular interest, since their reactivity is possibly controlled by ligand-induced conformational changes. This is shown by the similar ligand protection against alkylation irrespective of the reagent's electrostatic charge (iodoacetamide or iodoacetate) and also by the similar substrate-binding properties of carboxamidomethylated and the unmodified enzyme.

The phosphonomethyl analogue of 3-phosphoglycerate is a potent competitive inhibitor of phosphoglycerate mutases

The phosphonomethyl analogue of 3-phosphoglycerate (2-hydroxy-4-phosphonobutanoate) is a potent competitive inhibitor of cofactor-dependent phosphoglycerate mutase from yeast and of cofactor-independent phosphoglycerate mutase from wheat germ. For the yeast enzyme Ki is 1.3 mM (Km for substrate is 0.71 mM); for the wheatgerm enzyme Ki is 18 mM (Km for substrate is 0.86 mM). This analogue should be a useful tool for n.m.r. spectroscopic studies on the mechanism of action of the two mutases. The arsonomethyl analogue of 3-phosphoglycerate (4-arsono-2-hydroxybutanoate) was a relatively poor inhibitor.

Inhibitors of angiotensin-converting enzyme containing a tetrahedral arsenic atom

A series of tetrahedral oxo acids of Group VA and VIA elements and of silicon and boron were examined as inhibitors of angiotensin-converting enzyme. Arsenate is a competitive inhibitor with a Ki of 27 +/- 1 mM, at least 10-fold more potent than phosphate. Dimethylarsinate is a competitive inhibitor with a Ki of 70 +/- 9 mM, 2-fold more potent than dimethylphosphinate. Oxo acids of boron, silicon, antimony, sulphur and selenium are not inhibitors. On the basis of these results and the strong inhibition of this zinc metallopeptidase by substrate analogues containing a tetrahedral phosphorus atom, two substrate analogues containing a tetrahedral arsenic atom were prepared. 2-Arsonoacetyl-L-proline is a competitive inhibitor with a Ki of 18 +/- 7 mM, more than 2000-fold weaker than that of its phosphorus analogue 2-phosphonoacetyl-L-proline. 4-Arsono-2-benzylbutanoic acid is a mixed inhibitor with a Ki of 0.5 +/- 0.2 mM, indistinguishable in potency from its phosphorus analogue 2-benzyl-4-phosphonobutanoic acid.

The mechanism of pyrophosphorolysis of RNA by RNA polymerase. Endowment of RNA polymerase with artificial exonuclease activity

DNA-directed RNA polymerase from Escherichia coli can break down RNA by catalysing the reverse of the reaction: NTP + (RNA)n = (RNA)n+1 + PPi where n indicates the number of nucleotide residues in the RNA molecule, to yield nucleoside triphosphates. This reaction requires the ternary complex of the polymerase with template DNA and the RNA that it has synthesized. It is now shown that methylenebis(arsonic acid) [CH2(AsO3H2)2], arsonomethylphosphonic acid (H2O3As-CH2-PO3H2) and arsonoacetic acid (H2O3As-CH2-CO2H) can replace pyrophosphate in this reaction. When they do so, the low-Mr products of the reaction prove to be nucleoside 5'-phosphates, so that the arsenical compounds endow the polymerase with an artificial exonuclease activity, an effect previously found by Rozovskaya, Chenchik, Tarusova, Bibilashvili & Khomutov [(1981) Mol. Biol. (Moscow) 15, 636-652] for phosphonoacetic acid (H2O3P-CH2-CO2H). This is explained by instability of the analogues of nucleoside triphosphates believed to be the initial products. Specificity of recognition of pyrophosphate is discussed in terms of the sites, beta and gamma, for the -PO3H2 groups of pyrophosphate that will yield P-beta and P-gamma of the nascent nucleoside triphosphate. Site gamma can accept -AsO3H2 in place of -PO3H2, but less well; site beta can accept both, and also -CO2H. We suggest that partial transfer of an Mg2+ ion from the attacking pyrophosphate to the phosphate of the internucleotide bond of the RNA may increase the nucleophilic reactivity of the pyrophosphate and the electrophilicity of the diester, so that the reaction is assisted.

The arsonomethyl analogue of adenosine 5'-phosphate. An uncoupler of adenylate kinase

Adenosine was converted into the arsonomethyl analogue of AMP. The reactions used provide a general route for converting an alcohol, R-CH2-OH, into the arsonomethyl analogue, R-CH2-CH2-AsO3H2, of its phosphate, R-CH2-O-PO3H2. The analogue of AMP proves to be a substrate for rabbit adenylate kinase, which shows a limiting velocity with it of 1/17 that with AMP, a Michaelis constant raised 70-fold to about 10 mM, and hence a specificity constant lowered about 1200-fold. The product of transfer of a phospho group from ATP to the analogue is, like all anhydrides of arsonic acids, unstable to hydrolysis, and so breaks down to yield orthophosphate and regenerate the analogue. Hence adenylate kinase is converted into an ATPase by the presence of the analogue.

The arsonomethyl group as an analogue of phosphate. An X-ray investigation

The X-ray structure analysis of three compounds of interest as enzyme substrates is reported. They are the hydrated forms of (I) DL-2-amino-4-arsonobutanoic acid [HO-AsO2--CH2-CH2-CH(NH3+)-CO2H], (II) DL-2-amino-4-phosphonobutanoic acid [HO-PO2--CH2-CH2-CH(NH3+)-CO2H] and the hydrated barium salt of (III) D-3-phosphoglycerate [HO-PO2--O-CH2-CH(OH)-CO2-]. The structures were fully refined to R factors of 0.033, 0.053 and 0.046. For the compounds (I) and (II) the charge distribution was directly determined by locating all H atoms. The co-ordination around As and P is approximately tetrahedral, with the valency angle between the two charged O atoms enlarged to 112 degrees in compound (I), 166 degrees in compound (II) and 122 degrees in compound (III). The As-X bond distances are increased relative to P-X to accommodate the increased atomic radius. The analysis establishes that the compounds are structural analogues. Tables of co-ordinates for H atoms, anisotropic thermal parameters, bond lengths and bond angles for the three compounds have been deposited as Supplementary Publication SUP 50122 (5 pages) with the British Library Lending Division, Boston Spa, Wetherby, West Yorkshire LS23 7BQ, U.K., from whom copies can be obtained directly [see Biochem J. (1983) 209, 5].