Dietary ceramide 2-aminoethylphosphonate, a marine sphingophosphonolipid, improves skin barrier function in hairless mice.

Sphingolipids are one of the major components of cell membranes and are ubiquitous in eukaryotic organisms. Ceramide 2-aminoethylphosphonate (CAEP) of marine origin is a unique and abundant sphingophosphonolipid with a C-P bond. Although molluscs such as squids and bivalves, containing CAEP, are consumed globally, the dietary efficacy of CAEP is not understood. We investigated the efficacy of marine sphingophosphonolipids by studying the effect of dietary CAEP on the improvement of the skin barrier function in hairless mice fed a diet that induces severely dry-skin condition. The disrupted skin barrier functions such as an increase in the transepidermal water loss (TEWL), a decrease in the skin hydration index, and epidermal hyperplasia were restored by CEAP dietary supplementation. Correspondingly, dietary CAEP significantly increased the content of covalently bound ω-hydroxyceramide, and the expression of its biosynthesis-related genes in the skin. These effects of dietary CAEP mimic those of dietary plant glucosylceramide. The novel observations from this study show an enhancement in the skin barrier function by dietary CAEP and the effects could be contributed by the upregulation of covalently bound ω-hydroxyceramide synthesis in the skin.

Chirality-Driven Mode of Binding of α-Aminophosphonic Acid-Based Allosteric Inhibitors of the Human Farnesyl Pyrophosphate Synthase (hFPPS).

Thienopyrimidine-based allosteric inhibitors of the human farnesyl pyrophosphate synthase (hFPPS), characterized by a chiral α-aminophosphonic acid moiety, were synthesized as enantiomerically enriched pairs, and their binding mode was investigated by X-ray crystallography. A general consensus in the binding orientation of all (R)- and (S)-enantiomers was revealed. This finding is a prerequisite for establishing a reliable structure-activity relationship (SAR) model.

Sphingoid bases of dietary ceramide 2-aminoethylphosphonate, a marine sphingolipid, absorb into lymph in rats.

Various functions of dietary sphingolipids have been reported; however, little is known about marine sphingolipids. Ceramide 2-aminoethylphosphonate (CAEP), an abundant sphingolipid in marine mollusks, frequently has a unique triene type of sphingoid base [2-amino-9-methyl-4,8,10-octadecatriene-1,3-diol (d19:3)]. We previously reported that dietary CAEP prepared from the skin of squid was digested in the intestinal mucosa of mice via ceramides to yield free sphingoid bases. How dietary CAEP is then used in the body remains unclear. Here, we investigated the absorption of dietary CAEP using a lipid absorption assay on the lymph collected from rats with thoracic duct cannulation. Our results reveal that sphingoid bases derived from CAEP, including d16:1, d18:1, and d19:3, were detected in the lymph after administration of CAEP. Lymphatic recovery of d19:3 was lower than that of other sphingoid bases. A large fraction of the absorbed sphingoid bases was present as complex sphingolipids, whereas a smaller portion was present in the free form. Fatty acids in ceramide moieties found in the lymph were partially different from dietary CAEP, which indicates that sphingoid bases derived from CAEP could be, at least in part, resynthesized into complex sphingolipids. Future studies should elucidate the metabolism of sphingoid bases derived from CAEP.

Digestion of Ceramide 2-Aminoethylphosphonate, a Sphingolipid from the Jumbo Flying Squid Dosidicus gigas, in Mice.

Ceramide 2-aminoethylphosphonate (CAEP), a sphingophosphonolipid containing a carbon-phosphorus bond, is frequently found in marine organisms and has a unique triene type of sphingoid base in its structure. CAEP has not been evaluated as a food ingredient, although it is generally contained in Mollusca organisms such as squids and shellfish, which are consumed worldwide. In this study, we aimed to elucidate the effects of CAEP as a food component by evaluating the digestion of CAEP extracted from the skin of the jumbo flying squid Dosidicus gigas. Our results revealed that dietary CAEP was digested to free sphingoid bases via ceramides by the mouse small intestinal mucosa. At pH 7.2, CAEP was hydrolyzed more rapidly than the major mammalian sphingolipid sphingomyelin; however, the hydrolysis of CAEP was similar to that of sphingomyelin at pH 9.0. Thus, the digestion of CAEP may be catalyzed by alkaline spingomyelinase and other enzymes. Our findings provide important insights into the digestion of the dietary sphingophosphonolipid CAEP in marine foods.

Synthesis and biological analysis of novel glycoside derivatives of l-AEP, as targeted antibacterial agents.

To develop targeted methods for treating bacterial infections, the feasibility of using glycoside derivatives of the antibacterial compound l-R-aminoethylphosphonic acid (l-AEP) has been investigated. These derivatives are hypothesized to be taken up by bacterial cells via carbohydrate uptake mechanisms, and then hydrolyzed in situ by bacterial borne glycosidase enzymes, to selectively afford l-AEP. Therefore the synthesis and analysis of ten glycoside derivatives of l-AEP, for selective targeting of specific bacteria, is reported. The ability of these derivatives to inhibit the growth of a panel of Gram-negative bacteria in two different media is discussed. ß-Glycosides (12a) and (12b) that contained l-AEP linked to glucose or galactose via a carbamate linkage inhibited growth of a range of organisms with the best MICs being <0.75mg/ml; for most species the inhibition was closely related to the hydrolysis of the equivalent chromogenic glycosides. This suggests that for (12a) and (12b), release of l-AEP was indeed dependent upon the presence of the respective glycosidase enzyme.

In vitro and in vivo anti-angiogenic activities and inhibition of hormone-dependent and -independent breast cancer cells by ceramide methylaminoethylphosphonate.

Ceramide methylaminoethylphosphonate (CMAEPn) was isolated from eastern oyster ( Crassostrea virginica ) and screened against in vitro and in vivo angiogenesis and against MCF-7 and MDA-MB-435s breast cancer cell lines. In vitro angiogenesis was evaluated by the vascular endothelial growth factor (VEGF)-induced human umbilical vein endothelial cell (HUVEC) tube formation assay. MCF-7 and MDA-MB-435s cell viability was evaluated by the CellTiter 96 AQ(ueous) One Solution Cell Proliferation assay. Apoptosis was evaluated by the caspase-9 assay, autophagy by acridine orange staining and beclin-1 level. Our study indicates that CMAEPn at 50 microM inhibited VEGF-induced tube formation by HUVEC. The viability of MCF-7 and MDA-MB-435s breast cancer cells exposed to 125 microM CMAEPn for 48 h was reduced to 76 and 85%, respectively. The viability of MCF-7 and MDA-MB-435s cells exposed to 250 microM CMAEPn for 48 h under the same conditions was reduced to 38 and 45%, respectively. CMAEPn at 125 microM inhibited VEGF-induced MDA-MB-435s cell migration and invasion. CMAEPn at 125 microM also decreased VEGF, EGF levels in the conditioned media, PI3K, IkappaB phosphorylation and degradation in the cytoplasmic extracts, and NFkappaB nuclear translocation. Both acridine orange staining and beclin-1 indicated autophagic cell death in MCF-7 and MDA-MB-435s cells, respectively. In vivo, CMAEPn at 30 mg/kg body weight inhibited bFGF-induced angiogenesis and caused a 57% reduction in hemoglobin levels in the matrigel plug assay within 7 days. This is the first report on CMAEPn-inhibited angiogenesis both in vitro and in vivo.

Structures of an alanine racemase from Bacillus anthracis (BA0252) in the presence and absence of (R)-1-aminoethylphosphonic acid (L-Ala-P).

Bacillus anthracis, the causative agent of anthrax, has been targeted by the Oxford Protein Production Facility to validate high-throughput protocols within the Structural Proteomics in Europe project. As part of this work, the structures of an alanine racemase (BA0252) in the presence and absence of the inhibitor (R)-1-aminoethylphosphonic acid (L-Ala-P) have determined by X-ray crystallography to resolutions of 2.1 and 1.47 A, respectively. Difficulties in crystallizing this protein were overcome by the use of reductive methylation. Alanine racemase has attracted much interest as a possible target for anti-anthrax drugs: not only is D-alanine a vital component of the bacterial cell wall, but recent studies also indicate that alanine racemase, which is accessible in the exosporium, plays a key role in inhibition of germination in B. anthracis. These structures confirm the binding mode of L-Ala-P but suggest an unexpected mechanism of inhibition of alanine racemase by this compound and could provide a basis for the design of improved alanine racemase inhibitors with potential as anti-anthrax therapies.

Inhibition of rabbit brain 4-aminobutyrate transaminase by some taurine analogues: a kinetic analysis.

The use of the antiepileptic drug, 4-aminobutyrate transaminase (GABA-T) inhibitor vigabatrin (VIGA), has been recently cautioned because it is associated to irreversible field defects from damage of the retina. Since novel GABA-T inhibitors might prove useful in epilepsy or other CNS pathologies as VIGA substitutes, the aim of the present investigation was to characterize the biochemical properties of some taurine analogues (TA) previously shown to act as GABA-T inhibitors. These include (+/-)piperidine-3-sulfonic acid (PSA), 2-aminoethylphosphonic acid (AEP), (+/-)2-acetylaminocyclohexane sulfonic acid (ATAHS) and 2-aminobenzenesulfonate (ANSA). Kinetic analysis of the activity of partially purified rabbit brain GABA-T in the presence of VIGA and TA showed that PSA and AEP caused a linear, mixed-type inhibition (Ki values 364 and 1010 microM, respectively), whereas VIGA, ANSA and ATAHS behaved like competitive inhibitors (Ki values 320, 434 and 598 microM, respectively). Among the compounds studied, only VIGA exerted a time-dependent, irreversible inhibition of the enzyme, with Ki and k(inact) values of 773 microM and 0.14 min(-1), respectively. Furthermore, the ability of VIGA and TA to enhance GABA-ergic transmission was assessed in rabbit brain cortical slices by NMR quantitative analysis. The results demonstrate that VIGA as well as all TA promoted a significant increase of GABA content. In conclusion, PSA, ANSA and ATAHS, reversible GABA-T inhibitors with Ki values close to that of VIGA, represent a new class of compounds, susceptible of therapeutic exploitation in many disorders associated with low levels of GABA in brain tissues.

Protection of intrinsic nerves of guinea-pig detrusor strips against anoxia/glucopenia and reperfusion injury by taurine.

There is ample evidence that ischaemia is associated with partial denervation of the detrusor muscle and that this is responsible for much of its abnormal contractile behaviour, resulting in bladder dysfunction (instability). In guinea-pig nerves are very susceptible to the ischaemic damage as compared to the muscle cells. The purpose of this study was to assess the neuroprotection afforded by taurine on guinea-pig detrusor under ischaemic-like conditions. Guinea-pig detrusor strips were subjected for 60 min to ischaemic-like conditions, followed by 150 min reperfusion. Intrinsic nerves underwent every 30 min electrical field stimulation (EFS) by 5-s trains of square voltage pulses of 0.05 ms duration (15 Hz, 50 V). Detrusor strips were perfused with 0.1, 1, 3 or 10 mM taurine during the ischaemia-like exposure and the first 30 min of reperfusion. Taurine (1 and 3 mM) significantly improved the response of the strips to EFS both at the end of ischaemia and reperfusion. On the contrary, neither 0.1 nor 10 mM taurine had significant effects. It is concluded that taurine can partially counteract the ischaemia-reperfusion injury in the guinea-pig urinary bladder.

Further observations on the uptake and effects of phosphonates in perfused rat liver studied by (31)P-NMR.

We examined the route of uptake of 2-aminoethylphosphonate (NEthPo) and of phenylphosphonate (PhePo; 10 mM each) in perfused liver by (31)P-NMR. Uptake of NEthPo was concentrative. The rate of uptake was reduced to 21 +/- 2% (n = 3; all percentages refer to control rates) by substituting choline for Na(+), and to 21 +/- 4% (n = 3), 32 +/- 6% (n = 5) and 70 +/- 5% (n = 3) by replacing Cl(-) by gluconate, SO(4)(2-) or NO(3)(-), respectively. Taurine (20 mM) reduced NEthPo uptake to 38 +/- 6% (n = 3). The data are consistent with uptake of NEthPo by the Na(+)-coupled Cl(-)-dependent beta-amino acid transporter. A small fraction of NEthPo was incorporated into phospholipid. PhePo uptake evolved over 1 h towards levels of the membrane-permeant volume marker dimethyl methylphosphonate. Uptake depended on H(+), and was inhibited by 4, 4'-diisothiocyanato-stilbene-2,2'-disulphonic acid (100 microM), bumetanide and furosemide (1 mM each) and alpha-cyano-4-OH-cinnamic acid (5 mM) to 31 +/- 4% (n = 4), 28 +/- 4% (n = 4), 27 +/- 5% (n = 6) and 40 +/- 7% (n = 4), respectively. These characteristics of PhePo uptake are reminiscent of H(+)-coupled monocarboxylate transport. The monocarboxylates, lactate and acetate (20 mM), and the substrate analogue, phenylalanine (20 mM), were not inhibitory, while benzoic acid (20 mM) slightly inhibited (to 82 +/- 5%; n = 4) PhePo uptake. The tested phosphonates (10 mM) did not significantly affect hepatic extraction of [(3)H]-cholate or [(3)H]-taurocholate (25 microM each; 1:3 bile salt:albumin). The monocarboxylate analogue, PhePo (10 mM), did not significantly interfere with disposal of lactate (0.3-5 mM).

Slow-binding and competitive inhibition of 8-amino-7-oxopelargonate synthase, a pyridoxal-5'-phosphate-dependent enzyme involved in biotin biosynthesis, by substrate and intermediate analogs. Kinetic and binding studies.

8-Amino-7-oxopelargonate synthase catalyzes the first committed step of biotin biosynthesis in micro-organisms and plants. Because inhibitors of this pathway might lead to antibacterials or herbicides, we have undertaken an inhibition study on 8-amino-7-oxopelargonate synthase using six different compounds. d-Alanine, the enantiomer of the substrate of this pyridoxal-5'-phosphate-dependent enzyme was found to be a competitive inhibitor with respect to l-alanine with a Ki of 0.59 mm. The fact that this inhibition constant was four times lower than the Km for l-alanine was interpreted as the consequence of the inversion-retention stereochemistry of the catalyzed reaction. Schiff base formation between l or d-alanine and pyridoxal-5'-phosphate, in the active site of the enzyme, was studied using ultraviolet/visible spectroscopy. It was found that l and d-alanine form an external aldimine with equilibrium constants K = 4.1 mm and K = 37.8 mm, respectively. However, the equilibrium constant for d-alanine aldimine formation dramatically decreased to 1.3 mm in the presence of saturating concentration of pimeloyl-CoA, the second substrate. This result strongly suggests that the binding of pimeloyl-CoA induces a conformational change in the active site, and we propose that this new topology is complementary to d-alanine and to the putative reaction intermediate since they both have the same configuration. (+/-)-8-Amino-7-oxo-8-phosphonononaoic acid (1), the phosphonate derivative of the intermediate formed during the reaction, was our most potent inhibitor with a Ki of 7 microm. This compound behaved as a reversible slow-binding inhibitor, competitive with respect to l-alanine. Kinetic investigation showed that this slow process was best described by a one-step mechanism (mechanism A) with the following rate constants: k1 = 0.27 x 103 m-1.s-1, k2 = 1.8 s-1 and half-life for dissociation t1/2 = 6.3 min. The binding of compound 1 to the enzyme was also studied using ultraviolet/visible spectroscopy, and the data were consistent with the kinetic data (K = 4.2 microm). Among the other compounds tested, two potential transition state analogs, 4-carboxybutyl(1-amino-1-carboxyethyl)phosphonate (4) and 2-amino-3-hydroxy-2-methylnonadioic acid (5) were found to be competitive inhibitors with respect to l-alanine with Ki of 68 microm and 80 microm, respectively.

Differential modification of activities of the high-affinity and low-affinity insulin receptors of 3T3-L1 fibroblasts by phosphonolipids in vivo.

The low-affinity and high-affinity forms of the insulin receptor respond differently to modifications of cellular phospholipid content in mouse 3T3-L1 fibroblasts in vivo. When cells are cultured with 2-aminoethylphosphonate the resulting phosphonolipid, which has previously been demonstrated to prevent the insulin-induced differentiation of the fibroblasts into adipocytes [J. D. Smith et al., Biochem. Arch. 8, 339-344 (1992)] results in alterations in both the affinity for insulin and receptor number of the low-affinity receptor while leaving the high-affinity receptor unaffected. That this phospholipid modification induces a specific change in the cellular insulin effect is demonstrated by the lack of alteration in the mobilization of GLUT-4 and glucose transport in the lipid modified cells. The results suggest that this specific cellular phospholipid modification will be useful in dissecting the specific functions of the two forms of the mammalian insulin receptor.

Kinetics of endosomal acidification in Dictyostelium discoideum amoebae. 31P-NMR evidence for a very acidic early endosomal compartment.

We have examined the pH of the various endosomal compartments in the amoebae of the cellular slime mould Dictyostelium discoideum. This was accomplished both by fluorescence and by in vivo 31P-NMR methods. The fluid-phase marker, fluorescein-labeled dextran, was fed to the amoebae to report the average pH of their endocytic vesicles. During the progressive loading of successive endosomal compartments, we observed an early acidification down to a minimum value of pH < or = 5.3 after 30 min at 20 degrees C followed by an increase to an average pH of 5.8 when all the endosomal compartments were loaded by the fluid-phase marker. The weak fluorescence intensity of FITC-dextran at acidic pH precluded a more detailed investigation and we checked various phosphonate compounds as potential 31P-NMR pH probes for the endosomal compartments. Two molecules, aminomethylphosphonate and 2-aminoethylphosphonate, were selected for this study because of the large amplitudes of their chemical shift variation with pH (2 and 2.5 ppm, respectively) and their acidic pKs of 5.5 and 6.3, respectively. They were only moderately toxic (IC50% approximately 10 mM) towards both the axenic growth and the differentiation program of Dictyostelium amoebae. Internalization of the two aminophosphonates occurred only through the fluid-phase pinocytosis pathway as revealed by the full inhibition of their entry with 1 mM vanadate or 7.5 mM caffeine, two previously characterized inhibitors of endocytosis in Dictyostelium. We found that in vivo 31P-NMR of amoebae suspensions incubated with the aminophosphonates allowed the detection of three distinct intracellular compartments at pH 4.3, 5.8-6.0 and 7.3. Kinetics of aminophosphonate entry were analyzed and the results allowed us to reconstruct the time course for the acidification sequence during endocytosis. The data are consistent with the hypothesis that in Dictyostelium amoebae phosphonates occupy a highly acidic early endosomal compartment (t1/2 = 18 min; pH 4.3) before reaching a less acidic late endosomal/prelysosomal compartment (pH 5.8-6.0) from where they are immediately transported to, and trapped in, the cytoplasm (pH 7.3).

Inhibition of alanine racemase by alanine phosphonate: detection of an imine linkage to pyridoxal 5'-phosphate in the enzyme-inhibitor complex by solid-state 15N nuclear magnetic resonance.

Inhibition of alanine racemase from the Gram-positive bacterium Bacillus stearothermophilus by (1-aminoethyl) phosphonic acid (Ala-P) proceeds via a two-step reaction pathway in which reactivation occurs very slowly. In order to determine the mechanism of inhibition, we have recorded low-temperature, solid-state 15N NMR spectra from microcrystals of the [15N]Ala-P-enzyme complex, together with spectra of a series of model compounds that provide the requisite database for the interpretation of the 15N chemical shifts. Proton-decoupled spectra of the microcrystals exhibit a line at approximately 150 ppm, which conclusively demonstrates the presence of a protonated Ala-P-PLP aldimine and thus clarifies the structure of the enzyme-inhibitor complex. We also report the pH dependence of Ala-P binding to alanine racemase.

Time-dependent inhibition of Bacillus stearothermophilus alanine racemase by (1-aminoethyl)phosphonate isomers by isomerization to noncovalent slowly dissociating enzyme-(1-aminoethyl)phosphonate complexes.

An alanine racemase encoded by a gene from the thermophilic Gram-positive bacterium Bacillus stearothermophilus is overproduced to 0.3% of the soluble protein when carried on plasmid pICR4 in Escherichia coli [Inagaki, K., Tanizawa, K., Badet, B., Walsh, C. T., Tanaka, H., & Soda, K. (1986) Biochemistry (third paper of four in this issue)]. Purification of large quantities (50 mg) of racemase permits study of time-dependent inactivation by D and L isomers of the antibacterial (1-aminoethyl)phosphonate (Ala-P), the phosphonate analogue of alanine. The time-dependent activity loss by this compound now appears general to Gram-positive but not to Gram-negative racemases [Badet, B., & Walsh, C. (1985) Biochemistry 24, 1333] and is shown to occur by extremely slow dissociation of a noncovalent E X Ala-P complex. Ala-P binds initially in a weak, reversible (KI = 1 mM) competitive manner but is slowly isomerized (kinact = 6-9 min-1) to a stoichiometric enzyme complex, which in turn dissociates extremely slowly, with a half-time about 25 days. Thus, Ala-P is a slow but not a tight-binding inhibitor. The E X Ala-P complex is not reducible by borohydride but does perturb the fluorescence of bound pyridoxal 5'-phosphate coenzyme. Determination of the sequence of an active site octapeptide of the B. stearothermophilus alanine racemase shows homology with the sequence of a Gram-negative Salmonella typhimurium alanine racemase that is not susceptible to time-dependent inhibition by Ala-P. Studies with Ala-P analogues suggest the phosphonate dianion is crucial for stable formation of an isomerized long-lived E X Ala-P-inhibited complex.

Phosphatidylcholine homeostasis in phosphatidylethanolamine-depleted Tetrahymena.

The relative contributions of the two pathways of phosphatidylcholine biosynthesis, phosphatidylethanolamine N-methyltransferase (EC 2.1.1.17) and diacylglycerol: CDP-choline cholinephosphotransferase (EC 2.7.8.1), are altered in the ciliate protozoan Tetrahymena thermophila whose phospholipid composition has been modified by culturing the organism in the presence of one of several aminophosphonic acids, as determined by measuring the incorporation of [methyl-3H]choline and [methyl-14C]methionine into phosphatidylcholine in vivo. In control cells the phosphotransferase pathway provides about 40% of the phosphatidylcholine, while in cells grown with 2-aminoethylphosphonate (AEP), 3-aminopropylphosphonate (APP), and N,N,N-trimethylaminoethyl-phosphonate (TMAEP) the contribution of the phosphotransferase pathway to phosphatidylcholine formation is 75, 90, and 26%, respectively. In AEP- and APP-grown cells, in which 80% of the phosphatidylethanolamine has been replaced by the corresponding phosphonolipid, the methyltransferase is less active since the level of the substrate phosphatidylethanolamine is reduced and neither of the phosphonolipids is a substrate for the enzyme. In TMAEP-grown cells, TMAEP competes with and reduces the incorporation of phosphocholine by the phosphotransferase pathway, leading to a smaller contribution of the pathway to phosphatidylcholine biosynthesis. The relative amounts of the two different radioactive labels incorporated into diacylphosphatidylcholine vs alkylacylphosphatidylcholine are also altered in the phosphonate-grown cells. The exogenous AEP induces a change in the glyceryl ether content of the 2-aminoethylphosphonolipid--33% in the AEP-grown cells compared to 70% in the control cells--indicating that the exogenous AEP is entering the phospholipids by the ethanolamine-phosphotransferase pathway rather than by the route of the endogenous AEP.

(beta-Chloro-alpha-aminoethyl)phosphonic acids as inhibitors of alanine racemase and D-alanine:D-alanine ligase.

The (beta-chloro-, (beta, beta-dichloro-, and (beta, beta, beta-trichloro-alpha-aminoethyl)phosphonic acids have been synthesized and their inhibitory properties on the alanine racemases [EC 5.1.1.1] and the D-Ala:D-Ala ligases [EC 6.3.2.4] from Pseudomonas aeruginosa and Streptococcus faecalis have been evaluated. The monochloro and the dichloro derivatives of Ala-P exhibit a strong inhibition on the racemases of the two species tested but do not behave as suicide substrates. Only the D-Ala:D-Ala ligase of S. faecalis is inhibited by these compounds. The poor antibacterial activity observed with beta-chloro- and beta, beta-dichloro-Ala-P might be enhanced by the peptide-transport strategy.

Purification of an alanine racemase from Streptococcus faecalis and analysis of its inactivation by (1-aminoethyl)phosphonic acid enantiomers.

An alanine racemase has been purified some 30 000-fold almost to homogeneity from Gram-positive Streptococcus faecalis NCIB 6459; the enzyme has been purified to the same extent (4000-fold) from an O-carbamyl-D-serine-resistant mutant with a 7-fold higher enzyme level in crude extract. The racemase has one pyridoxal phosphate molecule per 42-kDa subunit, has a Vmax of 3570 units/mg and a Km of 7.8 mM in the L to D direction, and has a Vmax of 1210 units/mg and a Km of 2.2 mM in the D to L direction. The Keq is 0.8 and kcat/Km values are ca. 3 X 10(5) M-1 s-1. The purified enzyme is inhibited in a time-dependent manner by both L- and D-(l-aminoethyl)phosphonates (Ala-P), confirming observations of Atherton et al. in crude extracts of this organism [Atherton, F. R., Hall, M. J., Hassal, C. H., Holmes, S. W., Lambert, R. W., Lloyd, W. J., & Ringrose, P. S. (1980) Antimicrob. Agents Chemother. 18, 897]. Studies with [1-2H]-, [1-3H]-, and [1,2-14C]Ala-P rule out enzymic activation and processing as the basis for irreversible inhibition. Thus, enzyme after exposure to [14C]Ala-P or [alpha-3H]Ala-P and gel filtration contains stoichiometric amounts of radioactive label, but denaturation quantitatively releases intact Ala-P into solution as revealed by high-performance liquid chromatography and cocrystallization with authentic material. The Ala-P isomers are slow binding inhibitors of this racemase as is the alpha,alpha'-dimethyl analogue but not the D or L isomers of the corresponding phosphinate.(ABSTRACT TRUNCATED AT 250 WORDS)