Solutions and Standards
MMC (Medac, Hamburg, Germany) was dissolved in Glucose 5% to a stock concentration of 3 mM (1 mg/ml), aliquoted and stored in liquid nitrogen. Calibrators at 100, 10, 1 and 0.1 μM were made up in Ringer's-Acetate (R-A) (Fresenius Kabi, Halden, Norway) and stored at -70°C for up to two years without notable degradation. Aliquots of these calibrators were thawed, diluted, and included in each analytical series for calibration and quality control. MMC in microdialysates was quantified against calibrators in R-A. The same calibrators were diluted in rat plasma at the day of use for quantification of MMC in plasma samples. The internal standard solution was 5 μM tinidazole (T3021, Sigma-Aldrich Company Ltd, St. Louis, MO, USA) dissolved in R-A. Ethyl acetate, hexane and HPLC grade methanol were from Merck (Darmstadt, Germany). All aqueous reagents were made up in water purified through a Milli-Q UF-PLUS system (MillPore Corp., Bedford, MA, USA).
Determination of mitomycin C concentrations in plasma and microdialysates
MMC concentrations in plasma were determined by reversed phase high performance liquid chromatography (HPLC) after solid-phase extraction (SPE). The analytical method and routines for sample handling was based on previously published studies [23–25]. Some modifications to the methods were made, e.g. the internal standard used was no longer available and adaptation to differences in equipment and instruments used. Chromatographic analysis was performed on an Agilent 1100 system (Agilent Technologies, Santa Clara, California, USA) with a G1311A Quaternary pump, G1379A degasser, G1313A autosampler, G1316A Thermostated Column compartment and a G1365B Multi-Wavelength Detector. System control and data acquisition and integration were performed by Chemstation software Rev. B. 02.01. The mobile phase buffer was a 20 mM potassium phosphate buffer pH 7.0. The chromatography system was set up with buffer/methanol: 90/10 in reservoir A and buffer/methanol: 60/40 in reservoir B delivered in equal portions with a flow rate of 0.5 ml/min at 50°C. The column used for separation was a Supelcosil LC-18, 150 × 3 mm with 3 μm particles protected by a 20 × 2.1 mm guard column (Supelco, Bellefonte, PA, USA). Tinidazole was eluted at 2.8 minutes and MMC at 4.2 minutes, followed by a 4-minute washout with 100% B to eliminate late eluting peaks. One sample was injected every 15 minutes. When analyzing microdialysates no sample preparation was necessary. Ten μl of mobile phase A was added to 10 μl of dialysate and 15 μl of the mixture was injected. For these samples the washout with 100% B was not required, and samples could be injected every 5 minutes. For extraction of plasma samples, 1-ml Isolute SPE columns containing 25 mg of endcapped C18 sorbent were used (International Sorbent Technology LTD, Hengoed Mid Glamorgan, UK). To each 25-μl plasma sample, 150 μl of internal standard solutions and 25 μl R-A (or calibrator solution) were added. All plasma samples were split and extracted in duplicate by loading approximately half of the volume onto each of two SPE-columns. All steps up to the second wash (5% methanol) were performed without vacuum, the remaining steps were performed using gentle vacuum to pull the liquids through the columns. The columns were preconditioned with 1 ml methanol and 1 ml water. After loading, columns were washed with 0.5 ml R-A, 0.2 ml 5% methanol in water, and 0.2 ml hexane. Columns were dried by flushing with air for 2-3 minutes prior to elution using two times 100 μl of a 50/50 mixture of methanol and ethyl acetate. The eluates were evaporated to dryness for 30 minutes in a vacuum centrifuge. The residues were dissolved in 25 μl of mobile phase A, centrifuged and 15 μl of sample was injected onto the HPLC system.
The HPLC assays were run with the acceptance/rejection criteria suggested by Shah et al [26, 27]. The lower limit of quantization for the determination of total plasma concentrations of MMC was 0.1 μM. The upper limit was 4 μM. Within this range the method had a median accuracy of 100% (range 92%-111%), with a precision of 0.1%-6%. The lower detection limit for the determination of MMC in microdialysates was 0.02 μM. The tested upper limit was 100 μM, but the assay was set up with 10 μM as upper limit for the in-vivo studies. The method had a median accuracy of 99% (range 97%-103%), with a precision from 2% at MMC concentration 100 μM to 13% at MMC concentration 0.02 μM.
MD was performed using CMA/20 Elite microdialysis probes with membrane length 10 mm, diameter 0.5 mm and molecular cutoff 20 kDa. A CMA/100 syringe pump, 1-ml glass micro syringes and a CMA/140 fraction collector, all from CMA/Microdialysis AB (Solna, Sweden) were used to administer perfusion solution and handle collected fractions. MD fractions were collected in glass vials, diluted 1:1 with mobile phase A and stored at -70°C until analysis.
In vitro microdialysis in Ringer's - Acetate
The general mathematical expression of RR is:
in which Cin
is the concentration in perfusate, dialysate and the medium surrounding the probe membrane, respectively [28
]. Two experimental setups were applied in order to compare in vitro RR by gain (RRgain
) and RR by loss (RRloss
). To determine RRgain
, the probes were immersed in R-A spiked with MMC to concentrations 1, 5 and 10 μM and perfused with blank R-A. In this case Cin
= 0, and RRgain
was calculated by the equation:
The same probes were used to determine RRloss
by the retrodialysis method, in which identical concentrations of MMC in R-A were used as perfusate and the probes were immersed in blank R-A. In this case Cm
= 0, and RRloss
was calculated by the equation:
The experiments were conducted for four hours at 37°C with flow rate 1 μl/min and sample volume 20 μl, i.e. 12 replicates were collected from each probe.
In vitro microdialysis in plasma
To assess the influence on RR by changing the concentration, perfusion flow rate and temperature, three sets of experiments were performed with MD probes immersed in male rat plasma spiked with MMC at concentrations 1.0, 2.5 and 5.0 μM. Unless otherwise stated, the temperature in the medium was 37°C and flow rate was 1 μl/min. All experiments were performed in triplicate and the mean of three 10-μl samples was calculated for each data point. The experiments with variable flow rate (flow rate experiments) were performed with flow 6, 4, 2, 1 and 0.5 μl/min, and the temperature experiments were performed at 37, 40, 42 and 44°C. In these experimental setups RR was calculated using equation (2), in which Cm was the concentration of MMC in plasma. The zero-net flux (ZNF) method can be used both to calculate the solute concentration and for probe calibration, provided a stable concentration of the solute in the medium. In ZNF the perfusate contains the solute of interest and by varying Cin in a stepwise manner and measuring the resulting Cout, Cm and RR can be calculated using equation (1) . ΔC (=Cout-Cin) is plotted against Cin and a linear regression line is drawn between these points. The point of interception of the regression line on the x-axis (ΔC = 0) is where Cin is equal to the free fraction of Cm, while the slope of the regression line represents RR for the probe . The ZNF-experiments were performed with Cin 0.5, 4 and 10 μM.
Fifteen locally bred Rowett nude male rats with mean weight 346 g (range 302-396) were used, divided into three groups. Four animals were used to assess the stability of RR by the retrodialysis method in four different compartments over seven hours. For pharmacokinetic studies, six animals were given i.v. and five were given i.p. infusions of MMC. The animals were maintained under specific pathogen-free conditions, and food and water were supplied ad libitum. Housing and all procedures involving animals were performed according to protocols approved by the animal care and use committee.
Anesthesia was initialized with isoflurane (Sevorane, Abbott Scandinavia AB, Solna, Sweden) and maintained with subcutaneous (s.c.) injections of a mixture of tiletamine, 5.9 mg/ml and zolazepam, 5.9 mg/ml, (Zoletil vet, Virbac Laboratories, Carros, France), xylazine, 9.5 mg/ml, (Narcoxyl vet, Roche, Basel, Switzerland) and butorphanol, 0.23 mg/ml, (Torbugesic, Fort Dodge Laboratories, Fort Dodge, Iowa, USA). The initial dose of this mixture was 0.17 ml/100 g, and maintenance doses were 0.043 ml/100 g administered hourly. On completion of the experiments the animals were sacrificed using intracardial injections of penthobarbital (Haukeland Hospital Pharmacy, Bergen, Norway). During the experiments the body temperature was maintained at 37°C using a CMA/150 temperature controller (CMA/Microdialysis AB, Solna, Sweden). Infusion catheters for administering i.v. infusions and probes for performing MD were inserted through the right and left pectoral muscles into the external jugular veins through small skin incisions. Through a small midline laparotomy, placement of the MD probe in the extraperitoneal (XP) position in the flank was facilitated by tunneling outside the peritoneum with a blunt surgical probe under visual control. The MD probe was gently pushed in position beside the surgical probe which was then removed. MD probe or infusion catheter placement in the peritoneal cavity was accomplished through the abdominal wall and the muscle probe was placed in the left biceps femoris muscle through a separate skin incision. Except for the XP location, the probes and catheters were placed using an introducer consisting of a needle and split-tube.
Stability of relative recovery during in vivo microdialysis
To evaluate the stability of RR in the compartments named above (vein, peritoneum, XP and muscle), three experiments were performed for each compartment. Because the equipment permitted only three MD probes per experiment, all compartments could not be examined simultaneously. After surgery the probes were flushed with R-A for 15 min at flow rate 10 μl/min, for the rest of the experiment the flow rate was 1 μl/min. After another 15 min with blank R-A, perfusion with 2 μM MMC in R-A was initiated and the probes were perfused for 90 min for equilibration of the system before start of sampling. MD was performed for seven hours with continuous sampling of 20-μl fractions, i.e. 21 replicates were collected from each probe. To avoid clotting on the vein probe, dalteparin 25 IE/ml (Fragmin, Pfizer, Limoges, France) was added to the perfusate in this probe. The concentration in the tissue being zero, RR of each sample collected was calculated using equation (3).
Tissue distribution of mitomycin C during intravenous and intraperitoneal infusion
In this set of experiments, MD was performed to evaluate tissue distribution of MMC in rats during and after i.v. and i.p. bolus infusion. In the i.v. group the MD probes were placed in the vein, XP and peritoneum compartments and the infusion catheter in the left external jugular vein, while the i.p. group had probes in the vein, XP and muscle and an i.p. infusion catheter. For equilibration after surgery, the probes were perfused with R-A for 30 min and with 2 μM MMC in R-A for 90 min as described above. Individual calibration of the probes by the retrodialysis method was then performed in each experiment prior to the infusion of MMC. Four 10-μl samples were collected from each probe, and RR was calculated using equation (3) with Cout defined as the mean MMC-concentration in the dialysates. After calibration, the remainder of the experiments were performed using blank R-A as perfusate, starting with a 30 min washout of the drug. MMC 2.5 mg/kg (750 μM in saline) was then infused over 30 min using an Asana syringe pump (Alaris medical systems, Basingstoke, UK). Dialysate sampling was started 6 min after infusion start, allowing time for emptying of the dead-volume of the probe. Continuous sampling was performed for four hours at flow rate 1 μl/min with sample volumes of 10 μl. The tissue concentration of MMC was calculated by reformulating equation (2), dividing the dialysate concentrations by RR determined by the probe calibration. Dalteparin was added to the perfusate in the vein probe throughout the experiments as described above. Blood samples for plasma analyses were taken from the lesser saphenous veins 35, 65 and 95 min after start of infusion, immediately centrifuged and plasma was stored at -70°C until analysis. To calculate the fraction of MMC absorbed from the abdominal cavity during the i.p. infusion experiments, the abdomen was opened at the end of sampling and the remaining fluid was collected for MMC content analysis.
Calculations and statistical analysis
Pharmacokinetic parameters were determined by non-compartmental analysis using Kinetica ver 4.4.1 (Thermo Fisher Scientific Inc, Philadelphia, PA, USA). Statistical calculations were performed using SPSS software version 16.0 (SPSS GmBH, Chicago, Ill). Comparisons of area under the curve (AUC) in vein, peritoneum, XP, and muscle after i.v. and i.p. infusion were performed using the Mann-Whitney test, and p-values < 0.05 were considered significant.