Less Is More: a Mutation in the Chemical Defense Pathway of Erysimum cheiranthoides (Brassicaceae) Reduces Total Cardenolide Abundance but Increases Resistance to Insect Herbivores.

Erysimum cheiranthoides L (Brassicaceae; wormseed wallflower) accumulates not only glucosinolates, which are characteristic of the Brassicaceae, but also abundant and diverse cardenolides. These steroid toxins, primarily glycosylated forms of digitoxigenin, cannogenol, and strophanthidin, inhibit the function of essential Na+/K+-ATPases in animal cells. We screened a population of 659 ethylmethanesulfonate-mutagenized E. cheiranthoides plants to identify isolates with altered cardenolide profiles. One mutant line exhibited 66% lower cardenolide content, resulting from greatly decreased cannogenol and strophanthidin glycosides, partially compensated for by increases in digitoxigenin glycosides. This phenotype was likely caused by a single-locus recessive mutation, as evidenced by a wildtype phenotype of F1 plants from a backcross, a 3:1 wildtype:mutant segregation in the F2 generation, and genetic mapping of the altered cardenolide phenotype to one position in the genome. The mutation created a more even cardenolide distribution, decreased the average cardenolide polarity, but did not impact most glucosinolates. Growth of generalist herbivores from two feeding guilds, Myzus persicae Sulzer (Hemiptera: Aphididae; green peach aphid) and Trichoplusia ni Hübner (Lepidoptera: Noctuidae; cabbage looper), was decreased on the mutant line compared to wildtype. Both herbivores accumulated cardenolides in proportion to the plant content, with T. ni accumulating higher total concentrations than M. persicae. Helveticoside, a relatively abundant cardenolide in E. cheiranthoides, was not detected in M. persicae feeding on these plants. Our results support the hypothesis that increased digitoxigenin glycosides provide improved protection against M. persicae and T. ni, despite an overall decrease in cardenolide content of the mutant line.

Cardenolide genin pattern in Isoplexis plants and shoot cultures.

An HPLC method for the quantitative analysis of cardenolides was developed and applied. The procedure was optimised for analysing small samples (40 mg dw) of plant and tissue culture material. ISOPLEXIS CANARIENSIS plants obtained from seeds accumulated cardenolides to about 20 - 40 micromol g (-1) dw as calculated from the levels of cardenolide genins released after acidic hydrolysis of methanolic extracts. The relative contents of xysmalogenin, digitoxigenin, uzarigenin and canarigenin were 5 - 15 %, 0 - 5 %, 10 - 15 % and 70 - 90 %, respectively. Shoot cultures were initiated from seeds, established as permanent cultures and cultivated on agar-solidified or in liquid medium. Shoot cultures maintained on solid medium contained an average of about 6 micromol cardenolides g (-1) dw. A relatively high proportion of digitoxigenin was found in two-thirds of the shoot cultures examined. The cardenolide content of amphibian shoot cultures averaged to about 1 micromol g (-1) dw. Plants regenerated from shoot cultures and maintained under hydroponic conditions accumulated the same amount of cardenolides as plants collected in the field. No cardenolides could be detected in callus cultures.

Structural mechanism of specific ligand recognition by a lipocalin tailored for the complexation of digoxigenin.

DigA16 is an artificial digoxigenin-binding protein, which was derived from the bilin-binding protein, a lipocalin of Pieris brassicae, via reshaping of its natural ligand pocket. Here we report the crystal structures of DigA16 in the presence of either digoxigenin or digitoxigenin and for the apo-protein at resolutions below 1.9A. As a consequence of the altogether 17 amino acid substitutions within the binding site significant structural changes have occurred in the four loops that form the entrance to the ligand pocket on top of the structurally conserved beta-barrel framework. For example, one loop adopts a new alpha-helical backbone structure, which seems to be induced by few critical side-chain contacts. Digoxigenin becomes almost fully buried (by 95%) upon complexation, whereby specificity for the hydrophilic steroid is maintained through hydrogen-bonding networks and shape complementarity. The differential binding of the related steroid digitoxigenin is mainly governed by an internal histidine residue, whose side-chain undergoes significant induced fit. Among those amino acids that line the ligand pocket two tyrosine and one tryptophan residue provide the largest contacts. Interestingly, corresponding three side-chains are found with the same mutual orientation in the anti-digoxigenin antibody 26-10, even though the hapten orientation is quite different there and only 66% of the steroid surface is buried in the combining site. Hence, in the case of the engineered lipocalin DigA16 an example of convergent in vitro evolution is observed. Generally, the remarkable structural plasticity of the loop region and the role of polar residues in the binding site illustrate the potential of the lipocalin scaffold for the generation of specific receptor proteins towards a variety of ligands.

Synthesis and biological evaluation of 2-hydroxy derivatives of digitoxigenin and 3-epidigitoxigenin.

The four stereoisomers of the 2-hydroxy derivatives of digitoxigenin and 3-epidigitoxigenin have been synthesized, their structures established by NMR, and their binding affinity for the digitalis receptor on Na+, K(+)-ATPase evaluated. These derivatives showed lower affinities than the parent compounds. The hydrophilic hydroxy groups in the alpha position are more detrimental to the affinity than hydroxy groups in the beta position.

Biotransformation of digitoxigenin by cultured Strophanthus hybrid cells.

Hybrid cells between Strophanthus gratus and S. amboensis were obtained by electrofusion and confirmed to be hybrids through isozyme and RFLP analyses. Because a new and hybrid compound, 17 beta H-periplogenin beta-D-glucoside, was isolated as a biotransformation product of digitoxigenin by the hybrid cells, isomerization of 17 beta-lactone ring on S. gratus and glucosylation on S. amboensis were demonstrated simultaneously as the biotransformation abilities in the hybrid cells. Moreover, the productivity of the hybrid compound was increased by raising the sucrose concentration.

Förster energy transfer in ultrathin polymer layers as a basis for biosensors.

A method of detecting the binding of analyte molecules to biospecific receptors, like antibodies, is described. Förster energy transfer is used in connection with monomolecular organic films. The films are built up from pre-polymerized materials using the Langmuir-Blodgett or self-assembly techniques. Fluorescent dyes (as energy transfer donors) as well as reactive groups for covalent immobilization of protein receptors are integrated into the polymers. Several different methods for immobilizing biomolecules are described, including the use of protein A and the biotin/streptavidin couple. The studies suggest that the combination of Förster transfer and ultrathin organic films can be used for the construction of biosensors working either by displacement or by competitive assays. The mannose/concanavalin A, digoxin/antibody, and mouse immunoglobulin G/antibody systems are investigated. Further, a simplified meter optimized for measuring the fluorescence ratio at two wavelengths is described.

Biotransformation of digitoxigenin by cultured ginseng cells.

Nine compounds, including a new compound (digitoxigenin beta-D-glucoside malonyl ester), were isolated as biotransformation products of digitoxigenin by cell suspension cultures of Panax ginseng (Pg-3 cell line). At the same time, two known products were identified by TLC and HPLC.

Formation of a beta-glucuronidase-resistant glucuronide conjugate of digitoxin by dog liver microsomes.

The aim of these studies was to characterize the glucuronide conjugates of digitoxin and digitoxigenin monodigitoxoside (DMD) produced by liver microsomes from the dog with respect to hydrolysis by beta-glucuronidase and to behavior on HPLC. These results have been compared with studies of conjugates produced by liver microsomes from the rat. Glucuronidation was similar with both substrates with dog microsomes, whereas rat microsomes formed the glucuronide with DMD but not with digitoxin. The DMD glucuronide from both species was completely hydrolyzed by beta-glucuronidase, but no hydrolysis of digitoxin glucuronide was detected. The digitoxin glucuronide was hydrolyzed by a buffer at pH 1.5 but not at pH 10. After acid hydrolysis, the major products appear to be digitoxigenin and DMD glucuronide. These results suggest that glucuronidation of certain drugs by the dog is quite different from that of other species and that the dog may be the only species that possesses a glucuronosyltransferase capable of forming a glucuronide conjugate with digitoxin. The dog also has a glucuronosyltransferase, similar to that in the rat, which is responsible for glucuronidation of DMD. Whether this represents a single glucuronosyltransferase or two different enzymes remains to be elucidated.

UDP-glucuronosyltransferase activity toward digitoxigenin monodigitoxoside in human liver microsomes.

The properties of UDP-glucuronosyltransferase (UDPGT) toward digitoxigenin-monodigitoxoside (DT1) have been studied in human liver microsomes. The enzyme activity determined in nonactivated microsomes was very low (20 pmol/min/mg protein) compared with previously published values in rats (104 pmol/min/mg protein) or mice (379 pmol/min/mg protein) DT1-UDPGT activity was increased (180 to 220% of control activity) by Lubrol PX, Triton X-100, or (3-[3-cholamidopropyl]dimethylammonio)-1-propane sulfonic acid. The rate of DT1 glucuronidation determined for 29 different human liver microsomes was variable (18 to 87 pmol/min/mg protein). The KM found was approximately 4.5 microM. DT1-UDPGT activity was tentatively correlated with other known UDPGT activities. No significant correlations were found between DT1 and p-nitrophenol or 4-hydroxybiphenyl UDPGT activities. On the other hand, a strong correlation (r = 0.64, p < 0.05) was observed between DT1-UDPGT activity and digoxigenin-monodigitoxoside (DG1, another cardiac glycoside) glucuronidation rate.

Hydroxysteroid sulfotransferase and a specific UDP-glucuronosyltransferase are involved in the metabolism of digitoxin in man.

In vitro experiments were performed with cytosolic and microsomal fractions of human liver specimens in order to investigate which enzyme forms of sulfotransferase (ST) and UDP-glucurosyltransferase (GT) are involved in the metabolism of digitoxin (dt-3) and/or its cleavage products. It was found that the cytosolic STs preferentially react with digitoxigenin (dt-0) whereas microsomal GTs conjugate digitoxigenin-monodigitoxoside (dt-1) and in traces the bisdigitoxoside (dt-2). Dt-3 and dt-0 cannot be glucuronidated. By separation of different sulfotransferases it was found that the hydroxysteroid-ST is responsible for dt-0 and 3-epidigitoxigenin (epi-dt-0) sulfation. The hydroxysteroid-ST could be purified and characterized (apparent Km and Vmax for dt-0 sulfation: approx. 17 mumol/l and 2.7 nmol/min mg protein, respectively). Of various model substrates and endogenous compounds (steroids, bilirubin) none caused a competitive inhibition of the microsomal dt-1 glucuronidation except dt-2 and dt-3. Therefore it can be supposed that a new GT form catalyses this reaction. It is characterized by an extraordinarily high affinity towards dt-1 with Km values ranging between 0.7 and 27 mumol/l.

Metabolism of digoxin, digoxigenin digitoxosides and digoxigenin in human hepatocytes and liver microsomes.

In vitro metabolism of digoxin and its cleavage-related compounds was investigated using hepatocytes in primary culture and microsomal fractions both isolated from human livers. On these models, digoxin (DG3) and digoxigenin bisdigitoxoside (DG2) were not shown to be significantly metabolized in vitro in man. Therefore, it appeared that the stepwise cleavage of DG3 and DG2 sugars was not cytochrome P450 dependent. This enzymatic system probably plays a minor role in humans for this particular reaction. However, digoxigenin monodigitoxoside (DG1) and digoxigenin (DG0) which are known to be formed after intra-gastric hydrolysis of DG3, were extensively converted to polar compounds (mainly glucuronides). In addition, using human liver microsomes, a wide variability in UDP-glucuronyl transferase (UDPGT) activities responsible for DG1 glucuronidation was demonstrated. These results suggest that two main factors may contribute to the overall interindividual variability of digoxin biotransformation: 1), the individual intra-gastric pH which influences the sugar cleavage leading to DG1 and DG0; ii), a variability in the level of the hepatic UDPGT specific for digitalis compounds conjugation.

Purification of a rat liver cytosolic sulfotransferase responsible for the conjugation of digitoxigenin.

Previous investigations on the digitoxin metabolism hardly considered the role of the sulfate ester conjugation. Therefore, this study examined whether digitoxin (dt-3) or one of its cleavage products might be sulfated in vitro. It was proven that digitoxigenin (dt-0) is by far the best substrate for the cytosolic sulfotransferases (ST). Digitoxigenin-monodigitoxoside (dt-1) and digitoxigenin-bisdigitoxoside (dt-2) are sulfated in trace amounts whereas dt-3 is not sulfated at all. The purification of the responsible enzyme was performed by liquid chromatography on Q-Sepharose and hydroxyapatite. During the purification procedure this enzymatic activity corresponded exactly to that towards dehydroepiandrosterone (DHEA). The 134-fold purified and gel electrophoretically homogeneous enzyme protein (Mr 33,000) showed a Vmax of 12.5 nmoles dt-0 sulfate/min mg protein and a KM of 37 mumol/L. The purified enzyme conjugated dt-1 and dt-2 in trace amounts only and was inhibited competitively by DHEA. It can be concluded that in the rat a 3 beta-hydroxy-steroid sulfotransferase is responsible for the sulfation of dt-0. The purified enzyme reacts with dt-1, dt-2 and digoxigenin (dg-0) in traces only, a sulfation of dt-3 is not detectable.

UDP-glucuronosyltransferase activity toward digitoxigenin-monodigitoxoside. Differences in activation and induction properties in rat and mouse liver.

Glucuronidation of digitoxigenin-monodigitoxoside (DT1), a metabolite of the cardiac glycoside digitoxin, is mediated by the microsomal isozymes, UDP-glucuronosyltransferase(s) (UDP-GT). The present studies examined the activation and induction properties of UDP-GT activity toward DT1 in hepatic microsomes of rats and mice. When compared to enzyme activity present in native (latent) microsomes of the rat (0.104 +/- 0.010 nmol/min/mg of protein), the activity toward digitoxigenin-monodigitoxoside in mouse native microsomes was 3.5-fold higher (0.379 + 0.44 mumol/min/mg of protein). After treatment with ionic (sodium cholate), zwitterionic [3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid (CHAPS)], or nonionic (Emulgen 911, Triton X-100) detergents, or with UDP-N-acetylglucosamine, enzyme activity in rat microsomes remained unchanged. In contrast, UDP-GT activity (DT1) in mouse liver microsomes treated with detergents or with the nucleotide was increased 2-3-fold above native enzyme activity. Pretreatment of rats with the microsomal enzyme inducers, 3-methylcholanthrene and phenobarbital, had no effect on this enzyme activity, whereas pretreatment with pregnenolone-16 alpha-carbonitrile (PCN) and dexamethasone (DEX) increased enzyme activity toward DT1 800 and 380%, respectively. These findings support the hypothesis that PCN and DEX induce a unique form of UDP-GT in the rat that selectively glucuronidates DT1. In marked contrast, the activity of this enzyme in mouse liver was not affected by pretreatment with any of the microsomal inducers, including PCN and DEX. In both rat and mouse, the P-450p-dependent N-ethylmorphine demethylase activity was increased 10-15-fold in PCN-pretreated animals.(ABSTRACT TRUNCATED AT 250 WORDS)

Evidence for a digitoxin conjugating UDP-glucuronosyltransferase in the dog.

Liver microsomes of male Beagle dogs contain a form of UDP-glucuronyltransferase which is capable of conjugating digitoxin and its cleavage products digitoxigenin-bisdigitoxoside and digitoxigenin-monodigitoxoside. The highest reaction rates (Vmax 236 pmoles/mg microsomal protein min) were found for digitoxin and digitoxigenin-monodigitoxoside whereas the lowest Km was obtained for digitoxigenin-bisdigitoxoside (29 microM). Digoxin cannot be glucuronidated and digitoxigenin is glucuronidated only in traces. The result may explain the fast digitoxin elimination in dogs. Mutual induction experiments utilizing cardenolides and model substrates of UDP-glucuronyltransferase result in the conclusion that a specific form of UDP-glucuronyltransferase is responsible for glucuronidating digitoxigenin glycosides.

Simultaneous analysis of digitoxin and its clinically relevant metabolites using high-performance liquid chromatography and radioimmunoassay.

A specific assay for determining the urinary excretion of unchanged digitoxin and its metabolites is described. The procedure includes solvent extraction of urine at pH 8.5, reversed-phase high-performance liquid chromatography (HPLC) and radioimmunoassay (RIA) of equivalent fractions. Confidence limits showed good linearity and precision of recovery and high sensitivity, accuracy and specificity. Cross-reactivities were high for digitoxigenin (DGTN) and digitoxigenin bisdigitoxoside (Bis-DGTN), they were low for digitoxigenin monodigitoxoside (Mono-DGTN) or digoxin when [125I]digitoxin RIA was used. The interference of endogenous compounds in urine in the RIA was overcome by using HPLC. Compared with results reported in the literature, the urinary recovery of unchanged digitoxin was lower, being only 8.11 +/- 1.51% of the dose administered. Amounts of 6.52 +/- 1.31% were excreted hydrolysed as Bis-DGTN, Mono-DGTN, DGTN or C12-hydroxylated as digoxin.

Phosphorylated intermediates of Na,K-ATPase proteoliposomes controlled by bilayer cholesterol. Interaction with cardiac steroid.

Fragmental Na,K-ATPase from the electric eel forms three phosphorylated intermediates (EP) with MgATP and Na+: ADP-sensitive K+-insensitive EP (E1P), ADP- and K+-sensitive EP (E*P), and K+-sensitive ADP-insensitive EP (E2P). The EP composition varied with the Na+ concentration. In the reconstituted Na,K-ATPase proteoliposomes (PL), the EP composition of the inside-out form was controlled not only by the intravesicular (extracellular) Na+ concentration, but also by the temperature and the cholesterol content of the lipid bilayer. When the lipid bilayer of PL contained less than 30 mol % cholesterol, the E*P content did not change significantly while the E2P content increased with an elevation in temperature (3-20 degrees C). In contrast, when the lipid bilayer contains more than 35 mol % cholesterol, the E*P content increased while the E2P content stayed less than 10% under the same temperature change. These observations suggest that a high cholesterol content in the lipid bilayer interferes with the E*P to E2P conversion. This cholesterol effect was reversed by ionophores (monensin, nigericin, and A23187). Therefore, E1P-rich EP, E*P-rich EP, or E2P-rich EP could be obtained in the PL under a constant Na+ concentration by using various concentrations of cholesterol and ionophores. The reaction between the proteoliposomal EPs and digitoxigenin (lipid-soluble cardiac steroid) occurred in a single turnover, thereby avoiding unphysiologically high Na+ concentrations. The increase in the ADP- and K+-insensitive EP, which indicated formation of the digitoxigenin-Na,K-ATPase complex, was equivalent to the decrease in the E*P under six different sets of conditions, without any significant change in the E1P and E2P contents. This result indicated that E*P is the active intermediate of the Na,K-ATPase for cardiac steroid binding. Although the E2P has been thought to be the active form for binding, it cannot bind with the cardiac steroid in the presence of Na+ and in the absence of free Mg2+.

Induction of digitoxigenin monodigitoxoside UDP-glucuronosyltransferase activity by glucocorticoids and other inducers of cytochrome P-450p in primary monolayer cultures of adult rat hepatocytes and in human liver.

We have recently proposed that glucocorticoids induce cytochrome P-450p, a liver microsomal hemoprotein originally isolated from rats treated with the antiglucocorticoid pregnenolone 16 alpha-carbonitrile (PCN), through a mechanism that involves a stereospecific recognition system clearly distinguishable from the classic glucocorticoid receptor (Schuetz, E. G., Wrighton, S. A., Barwick, J. L., and Guzelian, P. S. (1984) J. Biol. Chem. 259, 1999-2012). We now report that digitoxigenin monodigitoxoside UDP-glucuronosyltransferase (DIG UDP-glucuronosyltransferase), a liver microsomal enzyme activity induced by PCN in rats, is also inducible, as is P-450p, in primary monolayer cultures of adult rat hepatocytes. DIG UDP-glucuronosyltransferase activity closely resembled reported characteristics of induction of P-450p in its time course of induction, concentration-response relationships, exclusivity of induction by steroids with glucocorticoid properties, unusual rank order of potency of glucocorticoid agonists, unusually high ED50 for induction by glucocorticoids, enhanced induction rather than inhibition by anti-glucocorticoids in the presence of glucocorticoids, and finally, induction by nonsteroidal inducers of P-450p. DIG UDP-glucuronosyltransferase activity was also readily detected in human liver microsomes and was elevated in two patients who had received inducers of P-450p. We conclude that the liver enzymes controlled by the postulated PCN recognition system include not only P-450p but also one or more UDP-glucuronosyltransferases.

In vitro digitoxin metabolism. Rate-limiting step and alteration following spironolactone pretreatment.

The biotransformation of digitoxin was considered as a linear transformation comprised of dual deoxy sugar cleavages culminating with product conjugation to glucuronic acid. Digitoxin and two metabolites, digitoxigenin bisdigitoxoside and digitoxigenin monodigitoxoside, were each incubated for 1 hr with rat liver slices. Measurements of amounts of various metabolites allowed determination of rate constants, showing the cleavage of the terminal sugar from digitoxigenin bisdigitoxoside to be the rate-limiting step. The inductive action of spironolactone predominately affected the rate-limiting step.

Separation, purification, and characterization of digitoxigenin-monodigitoxoside UDP-glucuronosyltransferase activity.

Glucuronidation of digitoxigenin-monodigitoxoside was investigated in liver microsomes from spironolactone-induced male Wistar rats. Isolation of a specific digitoxigenin-monodigitoxoside UDP-glucuronosyltransferase was possible utilizing chromatofocusing chromatography with a gradient from pH 10.1 to 8.0 after solubilizing the microsomal protein with the nonionic detergent Emulgen 911. The digitoxigenin-monodigitoxoside UDP-glucuronosyltransferase was further purified using UDP-hexanolamine Sepharose 4B affinity chromatography. The highly purified (75-fold) enzyme showed activity toward digitoxigenin-monodigitoxoside and slight activity toward digitoxigenin-bisdigitoxoside, whereas digitoxin and substrates for p-nitrophenol, 17 beta-OH steroid, and 3 alpha-OH steroid UDP-glucuronosyltransferases were not glucuronidated. In addition, bilirubin, morphine, estrone, 4-hydroxybiphenyl, and aromatic amines were not glucuronidated by this protein. These results strongly confirm the presence of a form of UDP-glucuronosyltransferase, which is highly specific for the glucuronidation of digitoxigenin-monodigitoxoside.

Altered distribution and toxicity of digitoxigenin in fasted mice.

Intravenous administration of digitoxigenin (DTXGN) evokes seizure episodes in mice which may be dependent on brain biogenic amines such as serotonin (5-HT). Fasting is known to have effects on both drug toxicity and brain 5-HT synthesis. The purpose of this study was to assess the effects of overnight fasting on DTXGN toxicity. The i.v. LD-50 of DTXGN was increased by 61% in fasted mice. Adjustment of DTXGN dose for the decrease in body weight of fasted mice did not alter the fasting induced protection. A loading dose of 1-tryptophan (25 mg/kg, i.p.) did not alter mortality rates in either fed or fasted mice. Cortical levels of 3H-DTXGN were decreased significantly by 25% in fasted mice. Liver and blood levels were elevated significantly. These data suggest that decreased DTXGN toxicity is associated with a decrease in its distribution to the cerebral cortex and emphasize the importance of acute dietary status in the expression of drug toxicity.