Digoxin-specific Fab fragments impair renal function in the rat.

The effect on renal function, and the plasma and urinary disposition, of digoxin-specific antibody fragments (DSFab), were studied using the rat as an experimental model. After 24h, DSFab (2 mg kg-1, i.v.) caused decreases in urine volume and creatinine clearance of 34 and 33%, respectively, when measured in the same rats. However, only the creatinine clearance was significantly changed when compared with a separate saline-treated control group. Plasma and urinary creatinine concentrations were unaffected by DSFab treatment. Since creatinine clearance approximates to glomerular filtration rate (GFR), it appears that a dose of DSFab equivalent to about one-fifth of the usual clinical dose, causes a reduction in GFR of about one-third. In patients undergoing digitalis therapy, a degree of renal impairment is common and it is possible that this may be exacerbated by treatment with DSFab. DSFab had an elimination half-life of 178 min, an apparent volume of distribution (Vd) of 106 mL kg-1 and a plasma clearance of 0.42 mL kg-1 min-1. If it is assumed that the plasma volume of a rat is approximately 35 mL kg-1, the measured Vd suggests appreciable penetration of DSFab into the extracellular fluid at this dose. Seventy-two hours after injection, only 7.6% of the administered dose of DSFab was found in the urine.

The plasma disposition and renal elimination of digoxin-specific Fab fragments and digoxin in the rabbit.

Administering digoxin-specific antibody fragments (DSFab, 1.9 mg kg-1, i.v.) to rabbits 1 h after digoxin (15 micrograms kg-1 or 12.5 microCi kg-1, i.v.) produced a redistribution of digoxin associated with a 5-fold elevation in total plasma concentration and 36-86% reductions in elimination half-life, apparent volume of distribution at steady-state and total body clearance (CLT). Renal clearance (CLR) was also reduced (54%), but urinary digoxin excretion was increased by one-third (35% vs 25%). This apparent anomaly is due to the large rise in total plasma digoxin concentration with a consequent increase in the area under the plasma concentration curve (AUC). The AUC, which is the denominator term in calculating CLR (and CLT), was increased to a greater extent than urinary digoxin excretion (numerator term in calculating CLR) so that an overall reduction in CLR occurred. The initial presence of digoxin appeared to alter the distribution of DSFab, since their plasma concentrations were markedly higher when the antibody was given after the hapten. The digoxin also reduced (from 3 to 1%) the amount of detectable DSFab in the urine.

Digoxin-specific Fab fragments impair renal function in the rabbit.

A 2 mg kg-1 intravenous bolus dose of digoxin-specific Fab fragments produced a 28% reduction in creatinine clearance in rabbits after 24 h. Urine output was reduced, while plasma and urinary creatinine concentrations were unaffected and increased, respectively. By 5 days the creatinine clearance had returned to normal. The fractional excretion of Na+ was nearly halved, indicating that the tubular reabsorption of Na+ increased to compensate for the reduced glomerular filtration rate, suggesting that tubular (as opposed to glomerular) function was not impaired.

The plasma kinetics of digoxin-specific Fab fragments and digoxin in the rabbit.

The plasma kinetics of total and free digoxin, and digoxin-specific antibody fragments (DSFab) in rabbits which had been given [3H]digoxin one hour before DSFab has been studied over a 5 day period. Injection of DSFab caused a 4- to 5-fold rise in total digoxin and reduced elimination half-life (t1/2 beta), apparent volume of distribution at steady-state (Vdss) and systemic clearance (CL) by 40, 90 and 75% respectively. Early in the experimental period, DSFab reduced free digoxin concentration (measured by ultrafiltration) from 4.1 ng mL-1 to a minimum of 1.3 ng mL-1 at 15 min. However, the concentration had rebound to 2.5 ng mL-1 by 60 min. Subsequently, free digoxin fell to 0.63 ng mL-1 and remained relatively constant over a 7 to 90 h period. The distribution half-life, t1/2 beta, Vdss and CL for DSFab (concentrations measured by enzyme-linked immunosorbent assay) were 0.3 h, 3.2 h, 185 mL kg-1 and 57 mL kg-1 h-1, respectively. A considerable molar excess (about 5) of DSFab in the plasma was necessary to maintain minimum free digoxin concentrations. When the DSFab:digoxin molar ratio was less than 4 during the initial treatment period, free (toxicologically active) concentrations increased. With the elevation in total digoxin, however, an opposite situation appeared to apply. By 24 h the relatively short DSFab t1/2 beta meant that the plasma DSFab concentration was less than 0.05 micrograms mL-1 giving a DSFab:digoxin molar ratio of below 0.06, yet the antibody-induced rise in total digoxin concentration was still detectable at 100 h.

The use of enzyme-linked immunosorbent assays to study the plasma disposition of sheep polyclonal and rat monoclonal digoxin-specific Fab fragments in the rabbit.

The plasma disposition of sheep polyclonal digoxin-specific Fab (fragment antigen-binding) fragments has been studied in rabbits after their intravenous injection (1 mg kg-1) using enzyme-linked immunosorbent assays exploiting both the species-specificity (ELISA1) and the digoxin-specificity (ELISA2) of digoxin-specific Fab fragments. The log concentration versus time profiles were best described by a biexponential plasma disposition when either assay was used. Although the plasma concentrations determined by ELISA1 and ELISA2 at each sampling time were not significantly different, there was a tendency for certain ELISA2 values to be higher. This resulted in the ELISA2-derived data giving a significantly longer distribution half-life (t1/2 alpha), but similar values for elimination half-life (t1/2 beta), apparent volume of distribution at steady state (Vdss), and clearance. Using ELISA2, which was generally the more sensitive assay, to compare the plasma disposition of the sheep polyclonal digoxin-specific Fab fragments with rat monoclonal digoxin-specific Fab fragments, it was shown that the rat product had a shorter t1/2 alpha (11 vs 22 min), a t1/2 beta which was not significantly different (253 vs 168 min), but a faster clearance (1.2 vs 0.7 mL kg-1 min-1), associated with a much larger Vdss (321 vs 108 mL kg-1). The extracellular fluid volume, using thiocyanate as a marker, was about 216 mL kg-1 for the nine rabbits used. This suggests that the rat preparation penetrates more extensively into the extracellular space and may indicate that some degree of extracellular binding or cell penetration is occurring.

The plasma disposition of sheep antibody (Fab) fragments in the guinea-pig and rabbit.

The plasma elimination of sheep digoxin-specific Fab (fragment antigen-binding) antibody fragments has been studied after intravenous injection (1 mg kg-1) in guinea-pigs and rabbits using an enzyme-linked immunosorbent assay. The log concentration versus time profiles were best described by biexponential and triexponential functions for the response in the guinea-pig and rabbit, respectively. However, the elimination half-lives and apparent volumes of distribution were similar in both species (about 140 min and 120 mL kg-1, respectively). The value for the Fab distribution volume suggests that the antibody fragments distribute out of the vascular compartment but do not fully occupy the extracellular space. Our estimates of the latter, using thiocyanate as a marker, ranged from 220 to 327 mL kg-1 (rabbits and guinea-pigs, respectively). The distribution of Fab fragments in these two species differs significantly from that in the rat, where our earlier studies have shown that these antibody fragments are confined to the intravascular compartment with a distribution volume approximately equivalent to that of plasma (about 40 mL kg-1).

Kinetics of digoxin and anti-digoxin antibody fragments during treatment of digoxin toxicity.

Anti-digoxin antibody fragments (ADAF, 80 mg) were infused intravenously to successfully treat severe digoxin toxicity in an 82 year old woman. During treatment, total and free digoxin were determined using an Abbot TDX analyser and an ultrafiltration technique. ADAF were measured by an enzyme-linked immunosorbent assay. By 1 h after ADAF, total serum digoxin concentrations had risen 12-fold from a pretreatment level of 15.4 nmol l-1 but free digoxin fell from 10 to 0.1 nmol l-1, indicating greater than 99.9% digoxin binding to ADAF. However, the low free levels had rebounded to 7.7 nmol l-1 by 12 h, but despite this rise the patient's condition had improved. A serum ADAF/digoxin molar ratio of around five was associated with the low concentration of free digoxin at 1 h, while at later times with ratios roughly between 3 and 4, the free digoxin concentrations ranged between 2.0 and 7.7 nmol l-1. ADAF were mainly confined to the plasma during the first hour, but subsequently distributed into an apparent volume of 193 ml kg-1. The elimination half-lives of ADAF and total digoxin were 96 and 55 h, respectively. More than 50% of the estimated digoxin load had been excreted in the urine by 5 days; for ADAF the equivalent figure was only about 3%. Renal and/or bacterial degradation may have contributed to the low detection of urinary ADAF.

The use of an enzyme-linked immunosorbent assay to study the disposition of sheep digoxin-specific immunoglobulin G and Fab fragments in the rat.

An enzyme-linked immunosorbent assay was developed for the measurement of sheep digoxin-specific immunoglobulin G and Fab fragments. With the latter, two preparations were examined, one available commercially (Digibind, Wellcome) and one prepared by ourselves (DSFab). The assay exhibited a greater sensitivity towards immunoglobulin G compared with Digibind and DSFab, presumably because the Fab preparations lacked some of the sheep-specific antigens present on the whole antibody molecule. The assay was used subsequently to examine the disposition of the antibody preparations after injection of 1 mg/kg i.v. into anaesthetised bile duct-cannulated rats. The plasma distribution half-lives (1.8-3.3 min) were similar for all three preparations, but while plasma elimination half-life values for Digibind and DSFab were much the same (110-115 min), that for immunoglobulin G was longer (425 min). The shorter half-life values for Fab fragments were linked to a rate of urinary elimination 10-20 fold faster. No antibody excretion in the bile was detected. The apparent volume of distribution of immunoglobulin G was 35 ml/kg, indicating that the whole antibody was largely confined to the plasma space. The volume of distribution for Digibind or DSFab (about 46 ml/kg) was not significantly larger than that for immunoglobulin G and much smaller than the extracellular fluid volume, which was measured as 305 ml/kg. Thus the distribution of sheep Fab fragments in the rat markedly differs from that in the baboon (Smith et al., 1979) where the apparent volume of distribution approximates to the extracellular fluid volume.

The effect of drug-specific active immunization on digoxin and benzylpenicillin disposition in the bile duct-cannulated rat.

Rats were immunized with a digoxin-human serum albumin conjugate i.m. This resulted in a several hundred-fold increase in plasma radioactivity and a 90% reduction in biliary drug elimination when [3H]digoxin (10 micrograms kg-1, i.v.) was subsequently injected into anaesthetized bile duct-cannulated rats. It was calculated that about 90% of the drug dose remained antibody-bound within the plasma compartment, with essentially no drug distributing into organs such as the heart and liver. Digoxin-specific antibody levels, determined by equilibrium dialysis, were high in the plasma but at least an order of magnitude lower in the bile. Immunization via Peyer's patches did not increase antibody levels in the bile. Immunization (i.m.) with a benzylpenicillin-human serum albumin conjugate gave specific antibody plasma titres with values less than 10% of those obtained after immunization with a digoxin-protein conjugate. However, although subsequent injection of the hapten (40 micrograms kg-1, [14C]benzylpenicillin, i.v.) was associated with much lower increases and decreases in plasma and biliary radioactivity, respectively, they were still statistically significant. It appears that endogenously-formed drug-specific antibodies, when present in the blood, will inhibit drug distribution and elimination. It is unlikely that their secretion in the bile plays a significant role in mediating biliary drug hapten elimination.

The effect of hydroxocobalamin on the nitroprusside-induced relaxation of rat aortic preparations.

Hydroxocobalamin (HOCb) when mixed with sodium nitroprusside (SNP) in a 10:1 or 1:1 molar ratio had no significant effect on the relaxation of noradrenaline-precontracted rat aortic rings. However, the addition of HOCb did prolong the time taken for the relaxant response to occur. The time taken for a 0.34 microM SNP dose to exert 50% of its relaxant effect was increased by about 120 and 35% with 10:1 and 1:1 HOCb/SNP molar ratio mixtures, respectively. A 0.5:1 HOCb/SNP molar ratio mixture caused no prolongation. In incubations with [14C]SNP, it was found that mixing with equimolar, or greater, amounts of HOCb, reduced aortic tissue uptake of radioactivity. The results support the hypothesis that both SNP alone and SNP/HOCb complexes breakdown extracellularly to release the same amount of active species, but that in the case of the complexes the active agent is released at a slower rate.

The influence of digoxin-specific antibody fragments on digoxin disposition in the rat.

Pentobarbitone-anaesthetized bile duct-cannulated female rats were injected intravenously with an equimolar dose of digoxin-specific sheep antibody fragments (DS-Fab) at 2 or 60 min after a dose of [3H]digoxin. The plasma drug levels were promptly elevated by 7-fold or 12-30-fold when the DS-Fab were given at 2 or 60 min respectively. When tissue drug concentrations were measured 2 min after a dose of DS-Fab (given 60 min after digoxin) which caused a 30-fold increase in plasma concentration, reductions could be detected if corrections were made for the presence in the tissues of high plasma concentrations of DS-Fab-bound drug. For instance, reductions in the heart, liver and small intestine were 63, 58 and 48% respectively. However, by 120 min after digoxin injection the only detectable effects on tissue drug concentration were in the kidney, where concentrations had increased 14-fold or 7-fold when the DS-Fab were given at 2 or 60 min respectively. Over the 120 min period the urinary excretion of digoxin-derived radioactivity was enhanced, and in the case where DS-Fab were given at 2 min, a 3-fold increase in urinary excretion was seen, which resulted in a net increase in the overall drug elimination. This greater urinary elimination was accompanied by a marked increase in the amount of bound drug in the urine (control and experimental values were 4 and 36% respectively). The cumulative biliary excretion of radioactivity seemed to be slightly reduced by DS-Fab administration at 2 or 60 min, although this was not statistically significant. A lack of significant drug-specific binding in the bile suggested that the liver is not involved in the elimination of hapten-DS-Fab complexes. There was little effect on the intestinal secretion of the drug.

Sodium nitroprusside: pharmacological aspects of its interaction with hydroxocobalamin and thiosulphate.

Hydroxocobalamin (HOCb), when mixed with sodium nitroprusside (SNP) in a 10:1 or 1:1 molar ratio and injected (i.v.) into the anaesthetized rat, prolonged the depressor response to SNP by 25-50%, but did not affect the degree of blood pressure lowering. Both the 'onset' and 'offset' components of the response were prolonged. Injecting [14C]SNP along with a 10-fold molar excess of HOCb resulted in a 2- to 3-fold elevation of plasma radioactivity which was maintained during the first 10 min of a 40 min experimental period. These effects of HOCb on the pharmacodynamics and pharmacokinetics of SNP are probably due to complex formation between the two compounds. Sodium thiosulphate (ST) added to SNP (12:1 molar ratio) had no effect on the depressor response to SNP. This mixing of ST and SNP had a less-marked influence on the plasma [14C] SNP-derived radioactivity than occurred with HOCb. There was no initial elevation of radioactivity, but the levels were raised by 50-60% at 4, 6 and 10 min. Since the depressor response to SNP was unaffected by ST, it is presumed that the higher concentrations of radioactivity were due to inactive degradation products rather than the active species itself.

The effect of digoxin dosage on the digoxin-quinidine interaction in the bile duct-cannulated rat.

Pretreating anaesthetized bile duct-cannulated rats with 9 mg kg-1 quinidine significantly decreased the cumulative biliary excretion of digoxin and its metabolites after 10 or 100 micrograms kg-1 [3H]digoxin, although the effect was more marked in animals receiving the high dose of digoxin. In contrast, however, although quinidine pretreatment raised plasma radioactivity levels by 50-80% in animals given the higher dose of digoxin, no significant effect on circulating plasma levels was observed in rats receiving 10 micrograms kg-1 digoxin. Generally, quinidine had no statistically significant effect on other aspects of digoxin disposition, although with both digoxin doses there were trends towards a reduction in the direct intestinal secretion and urinary excretion of digoxin-derived radioactivity with an increase in tissue levels of radioactivity (apart from the small intestine wall where concentrations were reduced). The radioactivity in the bile after 100 or 10 micrograms kg-1 digoxin comprised about 25 and 33% of digoxin and digoxigenin bis-digitoxoside, respectively, as well as appreciable amounts of the monodigitoxoside and a highly polar component. This metabolite profile was unaffected by quinidine. The influence of cardiac glycoside dosage shown by the present work indicates that the digoxin-quinidine interaction and possibly analogous interactions involving other cardiac glycosides, may not always be readily detectable from plasma concentration data.

The effect of digoxin-specific active immunization on digoxin toxicity and distribution in the guinea-pig.

In guinea-pigs intravenously infused with digoxin, prior immunization using a digoxin-human serum albumin conjugate increased by 3- and 2.4-fold, respectively, the digoxin doses causing the first signs of cardiotoxicity and death. At death, serum digoxin concentration was four times higher in immunized than in control animals. In the immunized guinea-pigs 50% of the serum digoxin was protein bound, presumably mainly to digoxin-specific antibodies, since in the controls the bound fraction was only 1-2%. Generally, tissue digoxin concentrations were not increased to the same extent as the lethal dose, and in the heart and lungs the increase was not significant. With cardiac (ventricle) subcellular fractions, there was no difference between control and immunized animals in the digoxin concentration of the 'microsomal' pellet. This subfraction contains the plasma membrane and the associated sodium pumps which are considered to be the sites at which the pharmacologically active digoxin binds. It seems likely, therefore, that the greater digoxin resistance in the immunized animals can be explained on the basis of reduced drug access to the site of action within the heart.

A strain dependent sex difference in ouabain-induced cardiotoxicity in rats.

Male rats of the BDIX strain were about twice as sensitive to the lethal effects of intravenously infused ouabain than were their female counterparts. This sex difference in ouabain sensitivity was not exhibited by rats of the Wistar strain. In bile duct-cannulated animals, ouabain biliary excretion was much faster in female BDIX rats. This finding could partly explain the greater ouabain resistance, although certain aspects of the cardiac/plasma ouabain concentration data did suggest that there may have been a component of the sex-related resistance that was not explicable on a pharmacokinetic basis. The faster ouabain biliary excretion in female rats could also partly explain the fact that for similar rates of infusion, the plasma ouabain concentration during infusion was similar in both sexes, despite the body size of the female rats being about one-third smaller.

The biliary elimination of amaranth, indocyanine green and nitrazepam in germ-free rats.

In anaesthetised bile duct-cannulated rats the overall rate of bile flow was 15-50% lower in both male and female germ-free (GF) rats. Except in the case of amaranth in female GF rats, this was reflected in a lower rate of biliary excretion of the three test xenobiotics, of which two (amaranth and indocyanine green) are excreted unchanged and one (nitrazepam) is excreted solely as metabolites. It was also noted that compared with conventional (CV) rats the relative liver weights (g/kg body weight) were about 20% lower in GF animals. After the intravenous injection of 14C-nitrazepam, thin-layer chromatographic separation of biliary nitrazepam-derived radioactivity revealed three loci (A, B and C in decreasing order of polarity). The relative proportions of A, B and C were similar in GF and CV rats, with C and B being the major and minor "metabolites" respectively. When 14C-nitrazepam was given intragastrically to non-anaesthetised rats, by 9 days about 20% and 70% of the dose had been recovered in the urine and faeces respectively of both GF and CV rats. The rate of elimination of urinary radioactivity was similar in GF and CV rats. However, faecal elimination was much slower in GF animals, for example after 24 hr the respective amounts of radioactivity excreted in GF and CV rats corresponded to 13% and 52% of the dose. These findings indicate that the indigenous bacterial population of an animal may indirectly affect the disposition of a xenobiotic whether or not it is metabolised by the bacteria.

The influence of digoxin antibodies on digoxin disposition and effect: studies in guinea-pigs and HeLa cells.

Pretreatment of guinea-pigs with digoxin-specific Fab (fragment antigen binding) fragments reduced the cardiotoxicity of intravenously infused digoxin (the lethal doses in Fab-treated and control animals were 1.0 and 0.6 mgkg-1, respectively). At death the serum digoxin concentration was elevated 2 fold in the Fab-treated animals, while the tissue concentrations were generally lower. The 30-40% lower cardiac digoxin concentration (seen in whole homogenate and throughout the subcellular fractions examined) was surprising; presumably this reflects a difference from the controls in the proportion of pharmacologically active/inactive digoxin in this organ. Adding digoxin-specific immunoglobulin G or the Fab fragments to HeLa cells before incubation with digoxin, reduced specific digoxin binding (Na pump-bound) slightly more than the non-specific binding. Adding specific antibody after digoxin, however, did not reduce digoxin binding or effect a recovery in Na pump activity. It seems that the protective effect of digoxin-specific antibodies seen in the guinea-pig can to some extent be simulated using HeLa cells. However, this is apparently not so regarding the widely-reported ability of these antibodies to reverse the action of digoxin.

The serum pharmacokinetics of digoxin as an immunogen and hapten in the rabbit.

3H-digoxin was given intradermally 4 weeks before, and at 6 and 44 weeks after immunisation with a 3H-digoxin-human serum albumin conjugate. Before immunisation, the serum digoxin distribution and elimination half-life (t1/2) values were 4.2 h and 2.1 days respectively. The immunogen-associated radioactivity showed characteristic fluctuations during a prolonged distribution phase of about 5 days, but the t1/2 of 2.7 days was similar to that of digoxin, indicating that appreciable cleavage of digoxin from the albumin may have occurred during the distribution period. At 6 weeks after immunisation, following hapten injection there was again a prolonged distribution phase of about 5 days during which concentration of digoxin were some five-fold higher than corresponding pre-immunisation values. The serum elimination t1/2 was 4.1 days. At 44 weeks the differences were even more marked; the distribution phase was some 7 days, during which serum hapten concentrations were approximately ten-fold higher than pre-immunisation values. The serum elimination t1/2 was in this case about 25 days. Surprisingly digoxin-specific antibody titres at 6 and 44 weeks were not significantly different, indicating that measurement of titre, which is a function of both antibody concentration and affinity, is not in itself reliable in predicting changes in hapten pharmacokinetics.

The effect of immunization with digoxin-specific antibodies on digoxin disposition in the mouse.

After intravenous dosing, digoxin was rapidly distributed to tissues, with a distribution half-life of 3.0 min. The highest digoxin concentrations at 1 hr post dosing were found in lymph nodes, adrenals, gallbladder (including contents), liver and kidney respectively. Digoxin concentrations in the heart, spleen, brain, lung, skeletal muscle and fat were similar to, or lower than, those in the plasma. The apparent volume of distribution (AVd) was 1 l/kg, and the plasma elimination half-life and clearance (Cl) 2.8 hr and 0.25 l/kg per hr respectively. When digoxin was given one day after passive immunization with digoxin-specific immunoglobulin G (IgG) or Fab fragments the respective plasma digoxin concentrations were elevated some 26- and 5-fold respectively compared with control values. Consequently there were reductions in AVd (93 and 32%) and Cl (94 and 50%). The effect of IgG treatment on clearance was still apparent when the hapten was given up to 14 days after immunization, while the weaker effect of Fab-treatment was less persistent. Although tissue digoxin concentrations were slightly lower in the immunised mice, it was only in the lymph nodes at 10 and 14 days after IgG treatment that the reduction in hapten concentration was statistically significant.