Rapid quantification of digitoxin and its metabolites using differential ion mobility spectrometry-tandem mass spectrometry.

This study focuses on the quantitative analysis of the cardiac glycoside drug digitoxin and its three main metabolites digitoxigenin-bisdigitoxose, digitoxigenin-monodigitoxose, and digitoxigenin using electrospray ionization-differential ion mobility spectrometry-tandem mass spectrometry (ESI-DMS-MS/MS). Despite large molecular weight differences, gas-phase separation of the four compounds in the DMS drift cell was not possible, even by utilizing additional volatile chemical modifiers. Baseline separation was achieved after adduct formation with alkali metal ions, however, and efficiency was shown to improve with increasing size of the alkali ion, reaching optimum conditions for the largest cesium ion. Subsequently, an assay was developed for quantification of digitoxin and its metabolites from human serum samples and its analytical performance assessed in a series of proof-of-concept experiments. The method was applied to spiked human serum pools with concentration levels between 2 and 80 ng/mL. After a short reversed-phase chromatographic step for desalting the sample, rapid DMS separation of the analytes was carried out, resulting in a total run time of less than 1.5 min. The instrumental method showed good repeatability; the calculated coefficients of variation ranged from 2% to 13%.

Rapid, presumptive identification of seed-based toxins using direct analysis in real time mass spectrometry (DART-MS) and its variants.

Detection of seed-based toxins is a need for forensic chemists when suspected poisonings occur. The evidence that is found is often physically unidentifiable, as the seeds are mashed to extract the toxin. This work investigates potential strategies for rapid detection of seed-based toxins and seed mashes containing these toxins using chemical signatures obtained by direct analysis in real time mass spectrometry (DART-MS). Seven toxins (digoxin, digitoxin, hypaconitine, hyoscyamine, lanatoside, oleandrin, and scopolamine) and six seeds containing these toxins were studied. While detection of four of the toxins was readily attainable, detection of digoxin, digitoxin, and lanatoside was hindered by the inability to thermally desorb these larger compounds under normal operating conditions. The use of DART-MS variants capable of higher desorption temperatures (thermal desorption (TD)-DART-MS and infrared thermal desorption (IRTD)-DART-MS) enabled detection of these compounds. Detection of toxins from direct analysis of seed mashes and methanolic seed mash extracts was found to be compound and technique dependent. Principal component analysis (PCA) of generated mass spectra enabled differentiation of seed species, even in cases where the toxins were undetectable.

A novel LC-IDMS/MS method for the determination of the cardiac glycosides digoxin and digitoxin using caesium adducts.

This article describes an essential improvement of the published candidate reference measurement procedure for digoxin and digitoxin and compares it with the original method. The novelty of the method lies in the measurement of the caesium (Cs+) ion as product ion in the multiple reaction monitoring mode (MRM) with potentially improved analytical specificity whilst retaining a comparable accuracy and precision at therapeutic levels. The original measurement procedure used the single-ion mode (SIM). The dissociation of the Cs+ adducts in MRM leads to the formation of Cs+ ions as main charged product in high yield. The present method results in a product ion signal intensity in MRM for digoxin and digitoxin of up to 80% of the precursor ion signal intensity in SIM. The precision, expressed as the coefficient of variation of the new method for digoxin was 3.18% (SIM) and 2.28% (MRM) at a concentration of 0.66 microg/l and 1.26% (SIM) or 1.65% (MRM) at 2.0 microg/l. The corresponding data for digitoxin were 1.21% (SIM) and 1.62% (MRM) at 24 microg/l and 1.46% (SIM) and 1.13% (MRM) at 42 microg/l.

Disposition of digitoxin in renal failure.

The disposition of digitoxin was studied for a period of 8 days in 6 uremic patients given a single oral dose of 1 mg 3H-digitoxin. In plasma, the time-course of radioactivity indicated a diminished absorption velocity of tritium compared to that of control subjects already reported and, after reaching of a pseudostate-equilibrium at 24 hr, an exponential decline with a mean half-life of 8.0 days. In urine, smaller amounts of tritiated compounds were eliminated in uremic patients (8.7% of the dose) than in controls (22.5%). The average fecal excretion of digitoxin and its metabolites was not significantly increased. Chloroform extraction and thin-layer chromatography in plasma, urine and feces suggested no qualitative alteration in the metabolism of digitoxin. Calculations of the total body tritium content (body stores) after each 24-hr interval and its pharmacokinetic behavior showed that the elimination of digitoxin is determined by the transfer constant from tissue to plasma. The differences in elimination kinetics of digitoxin and its metabolites of uremic patients and healthy subjects were not significant.

Studies on digitalis. XII. Kinetic pattern of digitoxin metabolism in patients with biliary fistulas.

The metabolic pattern of cardioactive and inactive, conjugated metabolites (a maximum of 24 substances) was studied after a single intravenous dose of 0.6 mg digitoxin in two female patients (aged 72 and 62 yr) with biliary fistulas. Bile and urine were collected every twenty-fourth hour and the 1-, 2-, 4-, 6-, and 8-day samples were analyzed. With the methods used, enzymatic cleavage of conjugation bonds, TLC (thin-layer chromatography), and a modified 86Rb method, the products of hydroxylation, hydrolysis, and conjugation could be separated. All cardioactive metabolites were present in bile and all were conjugated. Unchanged digitoxin was the main substance excreted. Hydrolyzed and conjugated metabolites formed a greater part of the substances excreted in bile than hydroxylated metabolites. The metabolic pattern in bile did not change much with time. The metabolic pattern in urine showed no close resemblance to that in bile. Hydroxylated, hydrolyzed, and conjugated metabolites were equally predominant in urine. Interruption of the enterohepatic circulation by T tube drainage not only changed the elimination kinetics of digitoxin but also changed the pattern of digitoxin metabolites in urine.

Studies on digitalis. X. Digitoxin metabolites in human myocardium and relationship between myocardial and serum concentrations of digitoxin in patients on maintenance treatment.

The levels of digitoxin and cardioactive metabolites were measured in 42 atrial biopsies with a 86Rb method modified for analysis of myocardial samples. The mean value was 91.0 ng/gm wet weight (SD 54.4). Myocardial and serum concentrations were compared in 23 patients; there was no significant correlation. The ratio of total drug concentration in myocardium and serum ranged from 1 to 38 with a mean value of 5.4. Calculated from the free drug concentrations, the mean myocardial serum ratio was 200, which reflects the high affinity of digitoxin and cardioactive metabolites to the myocardium. The metabolic pattern of cardioactive and inactive metabolites (conjugates with glucuronic and sulfuric acid) was studied in autopsy samples from left ventricular myocardium from 7 patients. Significant differences between the myocardial and serum patterns of cardioactive and inactive metabolites were demonstrated. The myocardium contained less unchanged digitoxin (25.7%) and more hydrolyzed (55.4%) and conjugated (54.1%) metabolites than serum (57.6%, 31.0%, and 33.1%, respectively). Hydroxylated metabolites in myocardium (15.8%) were not significantly changed compared to serum (10.0%).

Effect of heparin on digitoxin protein binding.

Reduction of digitoxin binding to plasma proteins after heparin has been reported. Our aim was to determine whether this reduction is an in vivo effect or occurs only after blood collection as a result of heparin-induced lipolysis that increases levels of nonesterified fatty acids in vitro. The effect of heparin on digitoxin protein binding was studied in 10 patients undergoing hemodialysis receiving digitoxin maintenance therapy. Digitoxin free fraction increased after heparin, from 2.5% +/- 0.7% to 4.4% +/- 1.1%, but after inhibition of in vitro lipolysis with diethyl p-nitrophenyl phosphate (2mM), a potent lipase inhibitor, there was no increase in the free fraction (2.3% +/- 0.4% before heparin and 2.4% +/- 0.5% after heparin). Digitoxin salivary levels were also unchanged (0.41 +/- 0.08 ng/ml before heparin and 0.41 +/- 0.08 ng/ml after heparin [n = 8]). These data indicate that the binding of digitoxin to plasma proteins in vivo is not altered by heparin. The reduced binding reported elsewhere was a result of heparin-induced in vitro lipolysis.

Chemiluminescent enzyme immunoassay using beta-D-galactosidase as the label and the bis(2,4,6-trichlorophenyl)oxalate-fluorescent dye system.

A chemiluminescent enzyme immunoassay (EIA) using beta-D-galactosidase as the enzyme label is described in this report. The activity of beta-D-galactosidase, after separation of bound and free fractions, was measured using a coupled enzyme procedure: lactose and glucose oxidase were used as the substrate and coupling enzyme, respectively. Hydrogen peroxide generated from glucose was determined by the chemiluminescence reaction using the bis(2,4,6-trichlorophenyl)oxalate-fluorescent dye system. This assay system was applied to the EIA of phenytoin using beta-D-galactosidase as the label and assay sensitivity was increased about 10 times compared with the original method. This method may also be applied to other EIAs of digitoxin, alpha-fetoprotein and thyroid-stimulating hormone.

The uptake of cardiac glycosides by intestinal smooth muscle of the guinea-pig in relation to digitalis receptors.

1. The accumulation and release of (3)H-digitoxin, (3)H-digoxin and (3)H-ouabain by isolated guinea-pig intestinal smooth muscle has been studied and compared with a pharmacological action due to inhibition of the sodium pump.2. The uptake of labelled cardiac glycosides can be described by means of an exponential function. The t of uptake was similar for the three compounds and did not depend on the concentration.3. Analysis of the curve relating the uptake of cardiac glycosides at equilibrium to the bath concentration enabled a non-saturable and a saturable binding site to be distinguished.4. In contrast to the uptake observations, the onset of the pharmacological effect was dependent on the concentration, and furthermore the t((1/2)) for this effect was shorter.5. The release of cardiac glycosides proceeded more slowly than the uptake.6. The uptake of a labelled glycoside was reduced in the presence of another glycoside. The amount of displaceable glycoside was nearly equivalent to the capacity of the saturable binding site.7. The significance of these results is discussed.

Cat assay for the emetic action of digitalis and related glycosides (digitoxin, digoxin, lanatoside C, ouabain and calactin).

1. A titration assay with two end points is described for comparison of the emetic and lethal potencies of digitalis-like drugs.2. A drug was infused at constant rate to a conscious, unrestrained cat, through an indwelling venous cannula. At the moment of vomiting the cat was rapidly anaesthetized and infusion continued at the same rate until the moment of cardiac arrest.3. With very slow and very fast infusions, the emetic and lethal doses tended to rise. In the range between these extremes (which varied from drug to drug) they were independent of time.4. The observations could be accounted for by analogue computation, assuming that the drugs entered an initial pool and were distributed at finite rates to receptors in the CNS (vomiting centre) and heart.5. Half times of metabolic loss derived from this computation for digitoxin, digoxin and ouabain (17, 9.9 and 1.8 h, respectively) were in the same ratio as the threefold longer half times reported for these drugs in man.6. When measured with infusion rates in the time independent range, the ratio of lethal to emetic doses did not vary between the drugs studied. All caused vomiting at 40% of the lethal dose.7. From a review of the literature, the emetic and cardiotoxic actions of digitalis-like drugs appear inseparable and probably share a common biochemical mechanism.8. It is concluded that foreseeable improvements in digitalis-like drugs are small and would depend on the elimination of any local emetic effect on gut receptors which they may have.

Plasma concentration, uptake by liver, and biliary excretion of tritiated cardiac glycosides in the isolated perfused guinea-pig liver.

1. Investigations were carried out on isolated perfused guinea-pig livers. Different doses of tritiated ouabain, digoxin, and digitoxin were added to the perfusion medium and the subsequent plasma elimination, hepatic uptake, and biliary excretion quantitatively measured. After the perfusion, extracts of liver, bile and plasma were subjected to thin layer chromatography in order to detect the radioactively labelled glycosides and their metabolites.2. The ouabain concentration in the plasma approached the equilibrium stage within 45 minutes. At this time 40% of the administered dose had been taken up by the liver, and no further elimination occurred. The elimination curve for ouabain followed a simple exponential function. After 1 h the tissue medium (T/M) ratio was approximately 3. In bile hardly any radioactivity could be detected. Ouabain was therefore not excreted by the liver.3. Up to 80% of the digitoxin was eliminated from the plasma within 4 hours. The elimination of radioactive material for the dose range studied could be described by a hyperbolic function. The T/M ratio in the liver varied with time. At the beginning it was as high as 10 and after 4 h reduced to approximately 3. After 45-60 min the concentration of radioactive material in the bile was 500 times as high as that in the plasma. Almost 70% of the administered radioactivity was excreted with the bile within 4 hours. At the end of the perfusion almost all the identifiable substances in plasma and bile were polar metabolites, as shown by thin layer radiochromatography.4. Digoxin behaved similarly to digitoxin.5. The findings led to the following hypothesis: uptake of cardiac glycosides into the liver cells occurs by a passive diffusion process and is related to their lipid solubility. On the other hand excretion in the bile occurs in general if polar metabolites are formed in the liver cells.

Concentration of digoxin, methyldigoxin, digitoxin and ouabain in the myocardium of the dog following coronary occulsion.

26 mongrel dogs were given a single dose of 0.03mg/kg tritium-labelled digoxin, beta-methyldigoxin, digitoxin or ouabain 2 hrs or 95 hrs following experimental coronary occlusion. Examination of the epicardial ECG was performed by moving from intact to ischemic or necrotic zones. 60 min after glycoside administration the animals were sacrificed and tissue samples from the marked heart muscles areas and from the skeletal muscle were analysed for glycoside content. The early glycoside uptake in acute ischemic or necrotic myocardium was diminished independently of the physicochemical properties of the glycoside. Significantly higher glycoside concentrations (ng/g wet weight) were measured in the injured myocardium 3 hrs after coronary occlusion than 96 hrs afterward (p less than 0.005). The values in acute ischemic myocardium varied considerably. This nonhomogeneity of glycoside uptake in the acute ischemic heart muscle may partly explain the increased sensitivity to glycosides in myocardial infarction. The decline of glycoside concentration correlates with the alterations in the epicardial ECG. The cardiac effects of cardenolides 60 min after intravenous administration was caused by the unchanged glycoside. In contrast to the myocardium, glycoside accumulation could not be found in the skeletal muscle. The concentrations of digoxin, beta-methyldigoxin and digitoxin in the skeletal muscle were significantly higher than the concentration of ouabain, which was rapidly eliminated via the urine.

The Fab fragment of anti-digoxin antibody (digibind) binds digitoxin-like immunoreactive components of Chinese medicine Chan Su: monitoring the effect by measuring free digitoxin.

Chan Su, a Chinese medicine prepared from the skin glands of Chinese toads, is used in the treatment of cardiovascular diseases. Severe toxicity and even death has been reported from overdose with Chan Su. The cardiotonic effect of Chan Su is attributed to bufadienolides, which also have apparent digitoxin activity. We demonstrated that these components of Chan Su could be neutralized by digibind, both in vitro and in vivo. For in vitro experiments, we supplemented drug-free serum pools with aqueous extract of Chan Su. Then, to aliquots of serum pool containing Chan Su, various amounts of digibind (10, 25 or 50 microg/ml of serum) were added. After incubation, total and free digitoxin concentrations (in the protein-free ultrafiltrate) were measured using the fluorescence polarization immunoassay (FPIA) and a FLX/TDx analyzer. For in vivo experiments, mice were fed with Chan Su by gavage. After 45 min, 200 microg of digibind was administered by injection. Fifteen minutes after injection, blood was collected for analysis of total and free apparent digitoxin activities. We observed complete removal of apparent digitoxin activity from protein-free ultrafiltrate both in vitro and in vivo by digibind, indicating that digibind successfully binds Chan Su. We conclude that digibind neutralizes Chan Su, and measuring the free digitoxin concentrations can monitor such an effect.

Neutralization of cardiac toxins oleandrin, oleandrigenin, bufalin, and cinobufotalin by digibind: monitoring the effect by measuring free digitoxin concentrations.

Oleandrin plant poisoning is common in children and the plant extract is used in Chinese medicines. The toxicity is due to oleandrin and the deglycosylated metabolite oleandrigenin. Bufalin and cinobufotalin (toad cardiac toxins) are also widely used in Chinese medicines like Chan SU, and Lu-Shen -WU. Severe toxicity from bufalin after consumption of toad soup has been reported. Taking advantage of structural similarities of these toxins with digitoxin, we demonstrated that these compounds can be rapidly detected in blood by the fluorescence polarization immunoassay for digitoxin. The cross reactivities of these compounds with digoxin assay were much lower. For example, when a drug free serum was supplemented with 10 microg/ml of oleandrin, we observed 127.7 ng/ml of digitoxin equivalent but only 2.4 ng/ml of digoxin equivalent concentration. Digibind neutralized all cardiac toxins studied as evidenced by significant fall of free concentrations. When aliquots of serum pool containing 50.0 microg/ml of oleandrin were supplemented with 0, 10.0, 25.0, 50.0, 100, and 200 microg/ml of digibind, the mean free concentrations were 30.6, 23.3, 16.0, 10.7, 7.8 and 5.5 microg/ml respectively. Similarly, with 50.0 microg/ml of oleandrigenin (total concentration: 36.2 ng/ml), the free concentration was 14.5 ng/ml digitoxin equivalent in the absence of digibind and 5.4 ng/ml in the presence of 200 microg/ml of digibind. In another specimen containing 500 ng/ml bufalin (total concentration: 156.9 ng/ml), the free concentration was 8.6 ng/ml in the absence of digibind and none detected in the presence of 100.0 microg/ml digibind. Because such neutralization may also occur in vivo, digibind may be useful in treating patients exposed to these toxins.

7 beta -hydroxycardenolide glycosides.

The microbial transformation of digitoxin (I) by Streptomyces sp. yielded digoxin (III) as main product along with the by-products 7beta -hydroxydigitoxin (II) and 7beta -hydroxydigoxin (IV). The present paper is concerned with the structure elucidation of 7beta-hydroxycardenolides as well as with the formation of some of their derivatives.

High-performance liquid chromatographic analysis of digitoxin formulations.

A rapid, selective, and simple high-performance liquid chromatographic assay for digitoxin formulations is described. The method utilizes a conventional octadecyl-bonded phase column with detection at 220 nm. The isocratic solvent system resolves digitoxin from its potential degradation products and provides an accurate assay for tablet and injectable formulations with a relative standard deviation of 1.4 and 3.3%, respectively. The method is sufficiently sensitive to monitor content uniformity of tablets and the minimum quantifiable amount of digitoxin was determined to be 20 ng. The total chromatograph time was approximately 15 min.

Determination of cardiac glycosides in digitoxin tablets and deslanoside injections by micro-HPLC.

A micro high-performance liquid chromatographic (micro-HPLC) procedure for the assay of digitoxin tablets and deslanoside injections has been developed. Micro-HPLC is performed on an ODS micro column, with acetonitrile-methanol-water (10:20:17) for digitoxin tablets and acetonitrile-water (21:70) for deslanoside injections. The effluent is monitored by UV absorption at 220 nm. Quantitation of cardiac glycosides in tablets and injections is carried out by the internal standard method. The composite assay results for digitoxin tablets and deslanoside injections provide average values of 101.2 and 99.9% with standard deviations of 1.2 and 0.93%, respectively. This micro-HPLC method is sensitive, quantitative, and reproducible. It is suitable for use in examining the content uniformity of pharmaceutical preparations.

HPLC method validation for Digitalis and its analogue by pulsed amperometric detection.

We developed a highly sensitive and selective reversed-phase HPLC-pulsed amperometric detection (RP-HPLC-PAD) method for cardiac glycoside detection. Eight cardiac glycosides were completely separated within 45 min on a reversed-phase column using a water-acetonitrile gradient, and were detected using a PAD under NaOH alkaline conditions. The detection (S/N=3) and quantification (S/N=10) limits for the cardiac glycosides were 0.1-0.3 and 0.3-0.8 ng, respectively. The linear regression coefficient was 0.9962-0.9998 for concentrations of 1-25 µg/mL. Cardiac glycosides in the Digitalis purpurea leaf displayed intra- and inter-day precisions (RSDs) of <9.30% and average recoveries of 98.63-99.94%. The contents of gitoxin, digitonin, and digitoxin in the D. purpurea were 0.197, 0.11, and 0.379 mg/g for leaf dried at 60 °C, 0.058, 0.11, and 0.090 mg/g for leaf dried at ambient temperature, and N.D. (not detected), and 18.379 mg/g, N.D. for seed, respectively. We conclude that our method shows good precision and accuracy.