Christian Friis

Putative biomarkers for evaluating antibiotic treatment: an experimental model of porcine Actinobacillus pleuropneumoniae infection

Biomarkers of infection were screened for their possible role as evaluators of antibiotic treatment in an aerosol infection model of porcine pneumonia caused by Actinobacillus pleuropneumoniae (Ap). Following infection of 12 pigs, clinical signs of pneumonia developed within 20 h, whereafter the animals received a single dose of either danofloxacin (2.5mg/kg) or tiamulin (10 mg/kg). To test the discriminative properties of the biomarkers, the dosage regimens were designed with an expected difference in therapeutic efficacy in favour of danofloxacin. Accordingly, the danofloxacin-treated pigs recovered clinically within 24h after treatment, whereas tiamulin-treated animals remained clinically ill until the end of the study, 48 h after treatment. A similar picture was seen for the biomarkers of infection. During the infection period, plasma C-reactive protein (CRP), interleukin-6 and haptoglobin increased, whereas plasma zinc, ascorbic acid and alpha-tocopherol decreased. In the danofloxacin-treated animals, CRP, interleukin-6, zinc, ascorbic acid and alpha-tocopherol reverted significantly towards normalisation within 24h of treatment. In contrast, signs of normalisation were absent (CRP, zinc and ascorbic acid) or less marked (interleukin-6 and alpha-tocopherol) in the tiamulin-treated animals. Plasma haptoglobin remained elevated throughout the study in both groups. This indicates that CRP, zinc, ascorbic acid and to a lesser extent interleukin-6 and alpha-tocopherol might be used to evaluate antibiotic treatment of acute Ap-infection in pigs. The present model provides a valuable tool in the evaluation of antibiotic treatments, offering the advantage of clinical and pathological examinations combined with the use of biochemical infection markers.

Pig hepatocytes as an in vitro model to study the regulation of human CYP3A4: prediction of drug-drug interactions with 17α-ethynylestradiol

The objective of this study was to provide evidence of the validity of pig hepatocytes as a model to study the regulation of human CYP3A4 with special emphasis on drug-drug interactions. Thirteen different drugs were incubated with primary monolayer cultures of pig hepatocytes (n = 4). The study included both drugs reported to cause drug interactions in the clinic with 17 alpha-ethynylestradiol (EE2), other drugs metabolized by CYP3A4, and drugs not reported to cause any problems. Effect of the drug exposure to pig hepatocytes was determined by immunodetection using a monoclonal human CYP3A4 antibody and measurement of 6 beta-hydroxylation of testosterone and 2-hydroxylation of 17 alpha-ethynylestradiol (EE2), both reactions known to be catalyzed by CYP3A4 in humans. Data were compared to data from human hepatocytes and to reported observations of drug-drug interactions in the clinic. The drugs known to be inducers of CYP3A4 in humans significantly increased a CYP isoform in pigs catalyzing 6 beta-hydroxylation of testosterone and 2-hydroxylation of EE2, whereas drugs not reported to have clinical interactions with EE2 had no or only marginal effect. Induction by the drugs known to be inducers of CYP3A4 increased with drug exposure time and the CYP3A4 activity, represented by testosterone 6 beta-hydroxylation, was highest at 72 h for the investigated induction periods (24, 48 and 72 h), except for dexamethasone where the effect peaked after 24 h. Induction of the 2-hydroxylation of EE2 correlated well with the increase in 6 beta-hydroxylation of testosterone (except for sulphinpyranzone) and the increase in the protein level of CYP3A detected by a monoclonal human CYP3A4 antibody, thus confirming the 2-hydroxylation of EE2 in pigs as being biotransformed by a CYP isoform presumably belonging to the CYP3A subfamily as in humans. In conclusion, these results indicate that pig hepatocytes may be a valuable model to mimic the regulation of human CYP3A4.

Penetration of amoxycillin to the respiratory tract tissues and secretions in Actinobacillus pleuropneumoniae infected pigs

The pharmokinetic properties of amoxycillin, and its penetration into respiratory tract tissue, were determined in 18 Actinobacillus pleuropneumoniae infected pigs, after a single i.v. dose of 8.6 mg amoxycillin kg(-1) bodyweight. Pleuropneumoniae was produced experimentally in pigs by an aerosol infection model. The infection created a homogeneous response, characterised by depression of breathing and increased body temperature. The clinical symptoms were accompanied by increased haptoglobin levels and circulating white blood cell counts. At necropsy the findings were characterised by a bilateral fibrinous pleuropneumonia. Twenty hours after infection, the pigs were administered amoxycillin i.v. The plasma concentration-time curve was described by a three compartment open model. The mean residence time and the elimination half-life were 1.5 and 3.4 hours, respectively. The steady-state volume of distribution was 0.67 litres kg(-1), and the clearance was 0.46 litres kg(-1) hour(-1). There were no significant differences between these values and those reported previously for healthy pigs. The concentration of amoxycillin in bronchial secretions, lung tissue and diseased lung tissue peaked two hours after intravenous drug administration, while amoxycillin concentration in pleural fluid, lymph nodes and tonsil tissue peaked at the first sampling point one hour after drug administration. The concentration of amoxycillin in secretions and tissue decreased by a slower rate than amoxycillin concentration in plasma, resulting in an increasing tissue-to-plasma concentration ratio. The distribution ratios (AUCtissue/AUCplasma) was 0.53 for bronchial secretions, 0.44 for pneumonic lung tissue, 0.42 for lung tissue, 1.04 for pleural fluid, 0.58 for lymph nodes and 0.37 for tonsil tissue. The distribution of amoxycillin to secretions was increased compared with that previously reported for healthy pigs, while only minor changes were observed in lung tissue.

Antimicrobial disposition in pulmonary epithelial lining fluid of horses, Part III. Cefquinome

Cefquinome concentrations, following intravenous and aerosol administration to horses, in pulmonary epithelial lining fluid (PELF) were examined and compared to plasma concentrations. Single dose of cefquinome sulphate (1 mg/kg) was administered intravenously to six horses followed by a single aerosol administration (225 mg) with a wash-out period of 14 days between treatments. After each drug administration, cefquinome concentrations in plasma and PELF, obtained by intrabronchial cotton swabs, were determined. After intravenous administration, cefquinome concentrations in plasma declined fast and were not detectable after 12 h. After aerosol administration, plasma concentrations were low or below limit of quantification (LOQ) during the entire sampling period. The degree of penetration of cefquinome into PELF after intravenous administration as described by the AUC(PELF) /AUC(plasma) ratio was 0.33. Following aerosol administration, cefquinome concentrations in PELF were high, but only detectable for 4 h. Based on AUC values, total cefquinome concentrations in PELF were one-third of total plasma concentrations after intravenous administration together with shorter time above Minimum Inhibitory Concentrations (T > MIC) in PELF, thus twice daily dosing may be required when treating lower airway infections in horses. Lower doses of cefquinome can be administered as aerosols providing high local drug concentrations in lung, but additional optimization of formulation is needed to improve distribution and persistence in lung.

Evaluation of a single dose versus a divided dose regimen of danofloxacin in treatment of Actinobacillus pleuropneumoniae infection in pigs.

A single versus a divided dose regimen of danofloxacin was evaluated in treatment of porcine Actinobacillus pleuropneumoniae infection using clinical observations combined with biochemical infection markers: C-reactive protein, zinc and ascorbic acid. Twenty hours after experimental infection, the 18 pigs received danofloxacin intravenously as a single dose of 2.5mg/kg or four doses of 0.6 mg/kg administered at 24h intervals. These dosage regimens resulted in similar AUCs of the plasma danofloxacin vs time curve. The maximum concentration was 3.5-fold higher using the single dose regimen, while the time with concentrations above the MIC was 2.5-fold longer using the fractionated regimen. Using the single dose regimen, temperature was normalised 32 h post-infection. In contrast, normalisation was delayed until 44 h post-infection using four low doses and a relapse with elevated temperatures at 52 and 68 h was observed. No other significant differences between the treatments were found, neither regarding clinical, haematological nor biochemical observations. The use of the more convenient single dose regimen was appropriate, as it was at least equivalent to the fractionated regimen.

Penetration of amoxycillin into the respiratory tract tissues and secretions in pigs.

The pharmacokinetic properties of amoxycillin, and its penetration into respiratory tract tissues (alveolar macrophages, bronchial secretions, bronchial mucosa, lung tissue and lymph nodes), were determined in 20 healthy female pigs weighing 29 to 55 kg, after a single intravenous dose of 8.6 mg kg(-1) bodyweight. Following intravenous administration the plasma concentration-time curves were best described by a three-compartment open model. The elimination half-life and the mean residence time were 2.5 and 1.4 hours, respectively. The volume of distribution at steady state was 0.52 litres kg(-1), and the body clearance was 0.40 litres hour(-1) kg(-1). In all structures (except alveolar macrophages) amoxycillin concentration peaked at the first sampling point, one hour after drug administration. The tissue to plasma ratio (based on AUC values) were 0.33 for bronchial secretions, 0.37 for bronchial mucosa, 0.39 for lung tissue and 0.68 for lymph nodes. Traces of amoxycillin were found in alveolar macrophages, but the concentrations were below the limit of quantification. The concentration of amoxycillin in secretions and tissue decreased by a slower rate than the concentration in plasma, resulting in increasing secretion- and tissue-to-plasma concentration ratios.

Expression changes of CYP2A and CYP3A in microsomes from pig liver and cultured hepatocytes

The P450 enzymes of the liver are responsible for the metabolism of a wide range of chemical compounds, and hepatocytes are used in pharmacological and toxicological in vitro tests. Thus, it is important to know how stable these enzymes are in culture. We measured the activity of CYP2A and CYP3A in microsomes isolated from both pig liver and primary pig hepatocyte cultures, together with the apoprotein concentration using Western blotting. The CYP2A activity and apoprotein concentration decreased rapidly; only about 5 percent remained after 48 hr incubation, whereas the CYP3A activity and apoprotein concentration was constant. CYP3A was induced 3 times after exposure to rifampicin, whereas neither rifampicin nor pyrazole could induce CYP2A. The hepatocytes were also incubated with varying concentration of FCS and autologous serum, however without effect on the stability of CYP2A, nor did different concentrations of growth hormone and testosterone added to the cultures have any effect.

Analyses of CYP2C in Porcine Microsomes

The cytochrome P450 2C (CYP2C) subfamily in human beings includes four different isoenzymes that metabolize different substrates although with some cross reactivity. Some of these substrates (e.g. diclofenac and tolbutamide), have been investigated in porcine microsomes, but without identifying the specific CYP catalysing the reactions. The objective of this study was therefore to test some CYP2C substrates and identify the porcine CYP2C responsible for the reaction. Three substrates, paclitaxel, tolbutamide and omeprazole, were chosen, as they are metabolized by three different CYP2C isoenzymes in human beings. Microsomes, isolated from 20 different pigs, 12 conventional, and 8 minipigs, were incubated with these compounds, and correlations between the metabolism rates of these three substrates were found indicating that the reactions are catalysed by the same enzyme. Male minipigs tend to have higher average activity than female minipigs, which is in contrast to the gender-dependent expression seen for other CYP isoenzymes. The metabolic activities correlated with the protein level determined in Western blotting, using anti-human CYP2C9, indicating that this enzyme is responsible for the reaction. The expression of the CYP2C enzymes was analysed by real-time polymerase chain reaction, using a primer set that could amplify CYP2C8, CYP2C9 and CYP2C19. The melting curves (peaks) revealed that all three genes were present, showing very different expression levels in the various types of pigs. The area of one of the peaks, however, correlated with the CYP2C9-like enzyme concentration, suggesting that this peak represents CYP2C9. Among paclitaxel, tolbutamide and omeprazole, omeprazole is the best probe of CYP2C9-like enzyme in the pig.

Determination of ptaquiloside and pterosin B derived from bracken (Pteridium aquilinum) in cattle plasma, urine and milk.

Ptaquiloside (PTA) is a toxin from bracken fern (Pteridium sp.) with genotoxic effects. Hydrolysis of PTA leads to the non-toxic and aromatised indanone, pterosin B (PTB). Here we present a sensitive, fast, simple and direct method, using SPE cartridges to clean and pre-concentrate PTA and PTB in plasma, urine and milk followed by LC-MS quantification. The average recovery of PTA in plasma, urine, and milk was 71, 88 and 77%, respectively, whereas recovery of PTB was 75, 82 and 63%. The method LOQ for PTA and PTB in plasma was 1.2 and 3.7ngmL(-1), 52 and 33ngmL(-1) for undiluted urine and 5.8 and 5.3ngmL(-1) for milk. The method is repeatable within and between days, with RSD values lower than 15% (PTA) and 20% (PTB). When PTA and PTB spiked samples were stored at -18°C for 14 days both compounds remained stable. In contrast, the PTA concentration was reduced by 15% when PTA spiked plasma was left for 5h at room temperature before SPE clean-up, whereas PTB remained stable. The method is the first to allow simultaneous quantification of PTA and PTB in biological fluids in a relevant concentration range. After intravenous administration of 0.092mg PTA per kgbw in a heifer, the plasma concentration was more than 300ngmL(-1) PTA and declined to 9.8ngmL(-1) after 6h, PTB was determined after 10min at 50ngmL(-1.)

Antimicrobial disposition in pulmonary epithelial lining fluid of horses. Part I. Sulfadiazine and trimethoprim

Sulfadiazine (SDZ) and trimethoprim (TMP) concentrations were examined in plasma and pulmonary epithelial lining fluid (PELF), following intravenous and oral administration and compared to minimum inhibitory concentrations (MICs) of common bacterial isolates from equine lower airway infections. SDZ/TMP (25/5 mg/kg) was administered intravenously, intragastric or per os to fed horses, and blood samples were collected before and 11 times, over 24 h, after administration. PELF samples were collected via a tampon device four times after drug administration and analysed for drug concentrations. Additionally, MICs of SDZ and TMP alone and in combination were determined in a selection of clinical respiratory isolates. Bioavailability was 74% for SDZ and 46% for TMP after paste administration in fed horses. The degree of penetration of SDZ and TMP into PELF, as described by AUC(PELF) /AUC(plasma) ratios, was 0.68 and 0.72, respectively, after intravenous administration. After oral administration, the degree of penetration for SDZ and TMP was 0.92 and 0.46, respectively. MIC measurements using SDZ/TMP ratios of 5:1 and 10:1 did not affect the interpretation of the results. The results indicate that clinically relevant drug concentrations of mainly TMP are difficult to maintain in PELF, especially after oral administration of SDZ/TMP.