QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: dmd.106.012880v1 35/2/306    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bouligand, J. Articles by Vassal, G. Search for Related Content PubMed PubMed Citation Articles by Bouligand, J. Articles by Vassal, G.

Induction of Glutathione Synthesis Explains Pharmacodynamics of High-Dose Busulfan in Mice and Highlights Putative Mechanisms of Drug Interaction

Unité Propre de Recherche et de l'Enseignement Supérieur, Equipe d'Accueil 3535, Pharmacology and New Treatments of Cancers, IFR54, University Paris XI and Institut Gustave Roussy, Villejuif, France (J.B., N.S., J.M., E.D., M.R., A.P., G.V.); Mass Spectrometry Platform, IFR54, Institut Gustave Roussy, Villejuif, France (A.D., G.V.); and Unité Mixte de Recherche 8121, Vectorology and Gene Transfer, IFR54, Centre National de la Recherche Scientifique and Institut Gustave Roussy, Villejuif, France (P.O., E.C.)

(Received September 11, 2006; accepted November 17, 2006)

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Busulfan is an example of a drug eliminated through glutathione S-transferase (GST)-catalyzed conjugation with reduced glutathione (GSH). We studied the pharmacokinetics and toxicity of busulfan in C57BL6 mice in correlation with liver GST activity and GSH synthesis by accurate determination of precursors, namely, -glutamyl-cysteine and cysteine. A significantly lower incidence of acute toxicity was observed in mice receiving busulfan 16.5 mg/kg twice a day compared with animals receiving 33 mg/kg once a day. In both cases, a total dose of 132 mg/kg was administered over 4 days. The difference in toxicity was explained by pharmacokinetics since a strong induction of clearance was observed only in animals treated twice daily. Induction of metabolism was correlated with an increase in liver cysteine content and enhanced glutathione synthesis rate, whereas GST activity was unchanged. To our knowledge, this is the first time that in vivo flux of GSH synthesis has been shown to be closely related to a drug plasma clearance and toxicity. These results allow hypothesizing that GSH liver synthesis may directly influence busulfan clearance in humans with possible implications in the occurrence of hepatic veno-occlusive disease.

View larger version (17K): [in this window] [in a new window]   FIG. 1. Glutathione metabolic pathways and busulfan metabolism.

The aim of this work was thus to study the regulation of GSH synthesis in vivo in mice during high-dose busulfan. This information was correlated to the clearance and toxicity to finally highlight putative mechanisms for metabolic drug interaction with busulfan, through the GSH/GST pathway.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Chemicals. Busulfan (1,4-butanediol-dimethylsulfonate) was purchased from Fluka (Steinheim, Germany) and busulfan-d4 was synthesized from [2,2,3,3-²H]-1,4-butanediol as previously reported (Vassal et al., 1988). Cyclophosphamide (Endoxan) was purchased from Baxter Oncology (Maurepas, France). Tetrahydrothiophene, 2-ethylethiophene, tetraethoxypropane, sodium iodide, and 4-nitrobenzylpyridine were provided by Aldrich (Steinheim, Germany); thiobarbituric acid was provided by Sigma (St. Louis, MO). Polyethylene glycol 400, dimethyl sulfoxide (Sigma), and N,N-dimethylacetamide (Merck, Hohenbrunn, Germany) were analytical grade. Dextrose (5%) was purchased from MacoPharma (Tourcoing, France). High-purity reduced glutathione (L--glutamyl-L-cysteinyl-glycine), oxidized glutathione, -glutamylcysteine tri-fluoroacetate salt, cysteine hydrochloride salt, cystine, homocysteine, homocystine, cysteinyl-glycine, and glutathione ethyl ester were supplied by Sigma (St Quentin Fallavier, France). High-performance liquid chromatography-grade acetonitrile, methylene chloride, heptane, and hexane were provided by Carlo Erba (Rodano, Italy), formic acid by Merck (Darmstadt, Germany), potassium chloride by Prolabo (Paris, France), and ethylenediaminetetraacetic acid disodium salt dihydrate by Labosi (Elancourt, France). Iodoacetic acid, ammonium bicarbonate, and bathophenanthrolinedi-sulfonic acid disodium salt were purchased from Aldrich, and bovine albumin was obtained from Euromedex (Mundosheim, France). Sodium hydroxide was provided by Prolabo. Sterile water was provided by Fresenius (Sèvres, France) and deionized water was prepared using a Milli-Q system (Millipore, Saint Quentin-en-Yvelines, France). Dulbecco's minimal essential medium was purchased from Invitrogen (Paisley, Scotland, UK).

Animal Experiments. Animal experiments were carried out in compliance with the conditions established by the European Union (Directives n°86/609/CEE and n°2003/65/CE). Male mice, C57BL/6 inbred strains, 2 months old, with an average weight of 25 g, were used for these experiments. Because fasting may lead to a significant decrease of hepatic glutathione contents, mice were not fasted before and during experiments.

Busulfan Treatment and Syngeneic Bone Marrow Transplantation. Mice were treated with intraperitoneally injected busulfan during 4 days at a total dose of 132 mg/kg. The intraperitoneal route was chosen based on previous experiments that have shown a bioavailability close to 100% for crystal-free busulfan solutions (Boland et al., 1999). This route of administration was preferred to the intravenous route since repeated reliable injections were required. The dose was chosen according to the literature and preliminary experiments with evaluation of a single-injection dose ranging from 16.5 mg/kg up to 40 mg/kg and total dose ranging from 107 mg/kg up to 133 mg/kg. Two schedules were compared: injection of busulfan twice daily and once daily. Busulfan was dissolved in N,N-dimethylacetamide to obtain a 60 mg/ml solution. The solution was then diluted in polyethylene glycol 400 to prepare a stable 6.4 mg/ml solution. This anhydrous solution of busulfan is then diluted extemporaneously in dextrose 5% before each injection, to a final concentration of between 1.6 and 3.2 mg/ml. Mice were transplanted with 5 x 10⁶ nucleated cells of fresh syngeneic bone marrow 24 h (twice daily schedule) after the last injection of busulfan. Syngeneic bone-marrow from sacrificed inbred animals was obtained by flushing femurs and tibias with Dulbecco's minimal essential medium. Survival was recorded during 1 year after BMT.

Busulfan Pharmacokinetics. Busulfan assay. Plasma busulfan concentrations were determined using a gas chromatography and mass spectrometry (GC-MS) assay with deuterated internal standard (busulfan-d4), as reported previously (Vassal et al., 1988). The GC-MS system consisted of a GC HP6890 II and MSD 5973 detector (Agilent Technologies, Massy, France). Accuracy and stability of busulfan concentration in injectable solutions were checked using a high-performance thin-layer chromatography assay (CA-MAG; Chromacim, Lyon, France) with derivatization by 4-nitrobenzylpyridine, as reported previously (Bouligand et al., 2004).

Pharmacokinetics. The treatment was started at 8:00 A.M. For the pharmacokinetics determination, mice were sacrificed at predetermined times along the treatment, i.e., before drug injection and 1, 4, and 8 h after drug injection. Four mice were sacrificed by sampling time point. A blood sample (1 ml) was immediately collected by cardiac puncture in a heparinized vial and plasma was separated by centrifugation (10 min at 2500g). Liver and other organs were collected and frozen immediately in liquid nitrogen and stored at –80°C until analysis. The descending segment of the concentration-time curve was monoexponential, as reported previously (Boland et al., 1999). Elimination half-lives (t_1/2) were determined from a log-linear curve with the three time points, 1, 4, and 8 h. The areas under the curve from 0 to 8 h (AUC_0–8h) were calculated using the linear trapezoidal rule from 0 to 8 h, and extrapolated to infinity according to the elimination half-life (AUC_0- = AUC_0–8h + C_8h/k_e, with k_e = ln 2/t_1/2). The plasma clearance uncorrected to bioavailability (CL/F) was calculated by dividing the dose by the AUC_0-.

In Vitro Metabolism: Conjugation of Busulfan to Glutathione. Preparation of cytosol. Mouse liver cytosol was prepared according to the previously published method (Abernathy et al., 1971). All steps were performed at +4°C. Frozen liver (150 mg) was thawed, grossly minced, washed, and suspended in approximately 1.5 ml of buffer (0.25 M sucrose, 1 mM EDTA, 1 mM Tris, pH 7.5). The suspension was homogenized using a Potter homogenizer at 750 to 3000 rpm, during 1 min. Crude homogenate was centrifuged at 10,000g for 10 min. The supernatant was centrifuged at 100,000g during 1 h. The cytosol (100,000g supernatant) used in the experiment was stored at –80°C until analysis. The protein content was determined using the micro-BCA protein reagent assay (Pierce, Rockford, IL).

Conjugation metabolism. These experiments were performed according to the previously published methods (Gibbs et al., 1997; Ritter et al., 1999). Busulfan solubility in water (1.2 mM) is far below the K_m value. The V_max and K_m therefore cannot be assessed separately. Intrinsic clearance (V_max/K_m) was calculated according to the method of Gibbs et al. (1997). In brief, incubation mixtures contained 0.25 mg of cytosolic protein and 0.7 µmol of reduced glutathione in 1.0 ml of assay buffer (100 mM sodium phosphate, pH 7.4 and 1 mM EDTA). Preincubation was carried out for 5 min at 37°C followed by addition of busulfan solution (100 mM) in dimethyl sulfoxide and incubation for 20 min at 37°C. Two different final concentrations of busulfan (250 µM and 500 µM) were prepared. The formation of the conjugate was linear for protein content from 0.2 to 1 mg/ml. The enzymatic reaction was stopped by adding 5 ml of methylene chloride. After vortexing for 20 s, the mixture was centrifuged for 15 min at 2500g. For alkaline hydrolysis and THT extraction, 200 µl of hexane, 100 µl of NaOH (1 M), and 40 µl of 2-ethylthiophene (5 µg/ml in isopropanol) were added to 700 µl of supernatant. After vortexing and centrifugation for 10 min at 10,000g, 100 µl of hexane phase was removed and a 2-µl aliquot was analyzed by GC-MS.

Tetrahydrothiophene assay. Sulfonium ion of glutathione was determined after alkaline conversion into tetrahydrothiophene using a GC-MS assay with a slight modification of the method (Ritter et al., 1999). In brief, the GC-MS system consisted of a GC HP5890 II and MSD 5971A (Agilent Technologies). GC was performed in the splitless mode on a 30 m x 0.25 mm HP-5 capillary column (Agilent Technologies). Helium (99.99%) was used as carrier gas using the constant flow mode with an inlet pressure of 75 kPa. An aliquot (2 µl) of hexane was injected at 280°C in splitless mode. The initial oven temperature was set at 45°C and held for 1.5 min, increased by 20°C/min to 165°C, and maintained for 0.5 min. In these conditions, retention times were 3.5 and 4.1 min for THT and 2-ethylthiophene, respectively. MS detection was performed in the electron impact mode with the following conditions of temperature and energy: transfer line at 280°C, detector at 170°C, and an electron-energy of 70 eV. THT calibrations were performed from 5 to 500 ng/ml using 2-ethylthiophene as an internal standard at a 270 ng/ml concentration. Initial THT stock solutions were prepared in isopropanol and diluted into cytosol or plasma.

In Vivo Synthesis of Glutathione in Mouse Liver. Assay of glutathione and precursors. Reduced and oxidized glutathione and precursors (-glutamylcysteine, cysteinyl-glycine, cysteine, cystine, homocysteine, and homocystine) were quantified in mouse liver using a liquid chromatography-tandem mass spectrometry (LC-MS/MS) assay after iodoacetic acid derivatization, as previously reported (Bouligand et al., 2006). The LC-MS/MS system consisting of an 1100 series high-performance liquid chromatography system (Agilent Technologies) and a Quattro-LCZ triple quadrupole mass spectrometer equipped with the orthogonal electrospray source (Micromass, Manchester, UK).

Kinetics of glutathione and precursors. The AUC_0–8h was calculated using the linear trapezoidal rule. The mean concentration was then calculated.

Various Assays. -Glutamyltranspeptidase was analyzed in plasma samples or in liver homogenates using the Synchron LX assay reagent (Beckman-Coulter, Fullerton, CA). Thiobarbituric acid reactive species was used to determine lipoperoxidation according to the previously published method (Wallin et al., 1993).

Western Blot. Frozen liver (50 mg) was thawed at +4°C, grossly minced, suspended in 100 µl of lysis buffer (containing 0.15 M NaCl, 1 mM KH₂PO₄, 5 mM MgCl₂, 1 mM EDTA, pH 6.4, 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, 1 mM benzamide, 1 µg/ml aprotinine, 10 µg/ml soybean trypsin inhibitor), and homogenized with a homogenizer (PowerGen125; Fisher Scientific, Pittsburgh, PA). A volume of 200 µl of NaCl (0.55 M) was added and incubated for 1 h on ice for whole-cell extract. After centrifugation at 13,000g for 30 min, the protein content was determined by the micro-BCA method (Pierce). Proteins (30 µg) were separated electrophoretically in 7.5% to 15% SDS-polyacrylamide gels and then transferred to a polyvinylidene difluoride membrane (Hybond P; GE Healthcare, Little Chalfont, Buckinghamshire, UK). Blots were incubated with rabbit polyclonal antibodies anti-C terminus human Nrf2 at 1:1000 dilution [anti-Nrf2 (C-20); Santa Cruz Biotechnology, Santa Cruz, CA] during 1 h, and followed by the anti-rabbit Ig horseradish peroxidase-linked whole antibody from donkey (GE Healthcare) at 1:2000 dilution. Detection was performed using a chemiluminescence (ECL) enzyme immunoassay (GE Healthcare).

Histology. Liver, lung, intestine, and femur were fixed in Finefix (Milestone Medical, Bergamo, Italy) and embedded in paraffin. For morphology analysis, 4-µm-thick paraffin sections were stained with hematoxylin-eosin-safranin. Liver was frozen with liquid nitrogen, embedded in Cryomatrix (Thermo Electron Corporation, Waltham, MA). For neutral lipid revelation, 4-µm frozen sections were stained with Sudan dye, a specific staining for fatty acids.

Statistical Analysis. Groups of values were always reported as mean ± S.D. Proportions were compared using the ² test and values were compared using the Mann-Whitney and Wilcoxon tests; for more than two groups, the Kruskal-Wallis nonparametric and Dunn's multiple comparison tests were used.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Toxicology. Survival. Survival of mice treated with busulfan at a total dose of 132 mg/kg given once or twice daily is shown in Fig. 2. There is a strong effect of administration schedule on mortality since all mice receiving busulfan once daily died before those receiving busulfan twice daily (p < 0.0001). Acute toxicity occurred during the 4 days of once daily busulfan treatment. More than 50% of animals were dead before day 0. The median survival was 3.5 days after the first day of busulfan treatment. All the deaths occurred more than 10 h after the third busulfan injection and this toxicity did not bias the metabolic study. However, BMT was not practicable in this group.

View larger version (7K): [in this window] [in a new window]   FIG. 2. Influence of busulfan dosing schedule on mouse survival. Mice were treated with busulfan injected intraperitoneally during 4 days (day –4 to day –1). There were eight mice by experimental group. Two schedules were compared: injection of busulfan 16.5 mg/kg twice daily versus injection of 33 mg/kg once daily. The twice daily schedule (median survival of 13 days without BMT) was significantly better tolerated (p < 0.0001) than the once daily schedule (median survival 3.5 days). Only the twice daily schedule without immediate toxicity allowed performance of BMT at day 0 and rescue of all animals from the bone marrow failure.

Body weight. Body weight is a significant marker of toxicity. Animals treated with busulfan 33 mg/kg once daily during 4 days presented a severe body weight loss with a mean ± S.D. of 16 ± 5% after 2 days of treatment, and continued to decrease thereafter. This effect was reduced in animals treated according to the 16.5 mg/kg twice daily schedule, with a body weight loss of 9 ± 3% after 2 days of treatment. The difference was statistically significant with p < 0.01. Moreover, in the twice daily group, the animals presented a progressive adaptation to the treatment with a recovery of the weight, up to –4 ± 1% of baseline after 4 days of treatment.

Histology. Toxicity was studied by systematic histological analysis of femur, lung, intestine, and liver. At onset of toxicity, animals treated twice daily presented a dramatic drop of the bone marrow cellularity. This confirmed that animals died of bone marrow failure. During this twice daily treatment, and after bone marrow transplantation, no significant morphologic changes were identified in lung, intestine, and liver. On the other hand, animals treated once daily presented an acute liver toxicity during busulfan treatment, whereas no significant morphologic changes were identified in lung and intestine. After 2 days of treatment, 50% of animals presented characteristic liver histology features with a diffuse microvesicular steatosis. This liver toxicity was dose-dependent since the incidence of mice with steatosis increased with the dose: 50% at 33 mg/kg/day and 80% at 40 mg/kg/day. This toxicity was associated with lethargy and severe body weight loss.

Long-time follow-up post bone marrow transplantation. BMTs were performed only on animals treated with busulfan administered twice daily. All transplanted animals survived more than 1 year after a syngeneic BMT. The experiment was censored after 1 year. No significant morphological changes were identified in liver, lung, and intestine. However, all busulfan-treated mice presented a hair-graying and a significant growth delay with an average body weight of 28 ± 1 g (–28%) 1 year post-BMT in the busulfan-treated group compared with 39 ± 4 g in the control group (p < 0.05).

Busulfan Plasma Pharmacokinetics. We studied busulfan plasma pharmacokinetics to explain the difference in toxicity between the two schedules of administration (once and twice daily) with the same total dose of 132 mg/kg. All the animals were followed over the same period (first day of busulfan treatment, 8:00 A.M., to third day, 4:00 P.M.). No deaths occurred during the study period. The plasma concentration-time curves of busulfan after intraperitoneal administration of busulfan are shown in Fig. 3. For each point, there was a low interindividual variability (about 20%) due to the use of inbred animals and standardized conditions of experimentation. It allowed estimating the AUC_0- of busulfan after administration of busulfan at 16.5 mg/kg (first and fifth dose) and at 33 mg/kg (first and third dose).

View larger version (10K): [in this window] [in a new window]   FIG. 3. Busulfan pharmacokinetics in mice. Plasma busulfan pharmacokinetics was determined after injection of busulfan 16.5 mg/kg twice daily (a) versus injection of 33 mg/kg once daily (b). These determinations were performed on untreated mice (1st dose) and in "induced" mice after 2 days of treatment (3rd or 5th dose). For the pharmacokinetics, mice were sacrificed over the treatment period at predetermined times, i.e., before drug injection and 1, 4, and 8 h after drug injection. Four mice were sacrificed by sampling time. Busulfan concentration was determined in blood plasma by GC/MS. Time (hours) is plotted on the x-axis and busulfan concentration (nmol/ml) on the y-axis.

0-

Induction of plasma clearance after several busulfan injections. The influence of repeated busulfan injections on plasma clearance was studied through the determination of pharmacokinetics after the first dose and after 2 days of treatment, i.e., at the third or the fifth dose (Fig. 3, a and b). We observed a strong increase of plasma clearance when animals were treated according to the schedule 16.5 mg/kg twice daily. The exposure determined after the fifth dose was 91 ± 27 h · nmol · ml^–1 compared with 220 ± 34 h · nmol · ml^–1 after the first injection. This induction of clearance (mean ± S.D.) from 7.4 ± 2.2 to 17.9 ± 2.8 ml/h was statistically significant (CL/F x 2.4, p < 0.0001). On the other hand, in animals treated according to the schedule 33 mg/kg once daily, the exposure determined after the third dose was 490 ± 114 h · nmol · ml^–1 compared with 604 h · nmol · ml^–1 after the first dose. The induction of the clearance from 5.4 ± 0.8 to 6.7 ± 1.6 ml/h was less intensive (CL/F x 1.2) compared with the twice daily schedule (p < 0.0001). We used a model of continuous monoexponential decay of the exposure fit to the experimental data to simulate the induction. According to this model, total exposure of mice receiving busulfan once daily (4 injections, total AUC = 2081 h · nmol · ml^–1) was 2.4-fold greater than the exposure of mice receiving the twice daily scheme (eight injections, total AUC = 917 h · nmol · ml^–1). This strong and significant difference in plasma exposure (p < 0.0001) may explain the difference in toxicity.

Study of the Mechanisms of Induction of Clearance. Busulfan rate of conjugation to glutathione in vivo. Plasma concentrations of sulfonium ion (THT⁺) were determined and the molar ratios of THT⁺/busulfan were calculated to estimate the rate of conjugation of GSH to busulfan. The mean molar ratio was very low (5.6 ± 3.8%) whatever the experimental conditions. Moreover, plasma concentration-time curves of busulfan and THT⁺ seem to be parallel at dose 1 (Fig. 4). In this mouse strain, the elimination rate of THT⁺ was thus quite similar to the elimination rate of busulfan. For this reason, the molar ratio THT⁺/busulfan is not an accurate parameter to estimate the induction of conjugation-rate. An in vitro approach to measure the specific busulfan GST activity was thus proposed.

View larger version (20K): [in this window] [in a new window]   FIG. 4. Plasma concentrations of busulfan and THT+. Plasma busulfan and THT+ concentrations were determined after injection of busulfan 16.5 mg/kg twice daily (a and b) versus injection of 33 mg/kg once daily (c and d). These determinations were performed on untreated mice at first dose (a and c) and in "induced" mice after 2 days of treatment at the fifth (b) or third (d) dose. For the study, mice were sacrificed over the treatment period at predetermined times, i.e., before drug injection and 1, 4, and 8 h after drug injection. Four mice were sacrificed by sampling time. Busulfan concentration () and THT+ concentration () were determined in blood plasma by two different GC/MS assay methods. Time (hours) is plotted on the x-axis and concentration (nmol/ml) on a log-scale y-axis.

Busulfan rate of conjugation to glutathione in vitro. The in vitro rate of conjugation to GSH was determined in the different experimental groups. The rate of conjugation was unchanged after 2 days of treatment. The intrinsic clearance (V_max/K_m) was 0.61 ± 0.02 µl · min^–1 · mg^–1 in untreated mice, 0.62 ± 0.01 µl · min^–1 · mg^–1 in adjuvant-treated mice, 0.58 ± 0.03 µl · min^–1 · mg^–1 in once daily-treated mice, and 0.62 ± 0.04 µl · min^–1 · mg^–1 in twice daily-treated mice, respectively. Furthermore, the expression of Nrf2, the percentage of oxidized glutathione (5.1 ± 2.7%), and lipoperoxidation determined through thiobarbituric acid reactive species levels (112 ± 57 pmol/mg) were not enhanced in the liver under the various conditions. In our experiment, induction of clearance was thus not due to an increase of busulfan-specific glutathione S-transferase activity through oxidative stress and Nrf2 activation.

In vivo glutathione synthesis. GSH and its precursors were simultaneously quantified in the mouse liver samples to have an overview of the flux of precursors for GSH synthesis (Fig. 5). In mice treated with busulfan 16.5 mg/kg twice daily, the level of GSH was sustained throughout the activation of the metabolic pathway. The mean concentration of GSH (83.6 ± 18.4 nmol/mg) during 8 h after the fifth busulfan injection (Fig. 5a) was not different from the mean concentration (74.8 ± 21.1 nmol/mg) after the first dose. However, GSH concentrations measured just before the fifth injection of busulfan (93 ± 26 nmol/mg) were significantly higher (p < 0.05) than the concentrations measured just before the first injection (53 ± 14 nmol/mg). Activation of the metabolic pathway was observed through a significant enhancement of the concentrations of GSH precursors, namely, -glutamyl-cysteine and cysteine. The mean concentration of -glutamyl-cysteine (Fig. 5b) after the fifth busulfan injection was 2.43 ± 0.62 nmol/mg compared with the mean concentration of 1.26 ± 0.39 nmol/mg after the first dose (p < 0.001). The mean concentration of cysteine (Fig. 5c) after the fifth busulfan injection was 6.06 ± 1.30 nmol/mg compared with the mean concentration of 3.93 ± 0.86 nmol/mg after the first dose (p < 0.001). In mice treated with busulfan twice daily, activation of glutathione synthesis, with increased liver content of both -glutamyl-cysteine and cysteine, might explain induction of clearance. To explain the enhanced level of cysteine, we determined the concentration of homocysteine and cysteinyl-glycine before and after the first and fifth busulfan injections. The mean concentration of homocysteine after the fifth busulfan injection was 0.21 ± 0.04 nmol/mg compared with a mean concentration of 0.27 ± 0.04 nmol/mg after the first dose. The mean concentration of cysteinyl-glycine after the fifth busulfan injection was 1.12 ± 0.24 nmol/mg compared with a mean concentration of 0.92 ± 0.27 nmol/mg after the first dose. No statistically significant increase in mean cysteinyl-glycine levels was observed with a significant increase of liver -glutamyltranspeptidase activity from 6.5 ± 1.0 mU/mg to 13.0 ± 3.3 mU/mg (p < 0.01).

View larger version (25K): [in this window] [in a new window]   FIG. 5. Kinetics of liver glutathione and precursors after busulfan administration. GSH and precursors were determined in mouse liver after injection of busulfan 16.5 mg/kg twice daily (a, b, and c) versus injection of 33 mg/kg once daily (d, e, and f). These determinations were performed on untreated mice (1st dose) and in "induced" mice after 2 days of treatment (3rd or 5th dose). For the kinetics study, mice were sacrificed over the treatment period at predetermined times, i.e., before drug injection and 1, 4, and 8 h after drug injection. Four mice were sacrificed by sampling time. GSH (a, d), -glutamyl-cysteine (b, e), and cysteine (c, f) were determined in liver mice by LC-MS/MS. Time (hours) is plotted on the x-axis and content (nmol/mg protein) on the y-axis. GLU-CYS, -glutamyl-cysteine.

In mice treated with busulfan 33 mg/kg once daily, the glutathione system was saturated, which may explain the nonlinearity of the pharmacokinetics. Activation of the glutathione metabolic pathway was observed with increased concentrations of glutathione, -glutamyl-cysteine, and cysteine, 1 h after the first busulfan injection (Fig. 5). However, the concentration of GSH (78 ± 19 nmol/mg) measured just before the third injection of busulfan was not significantly higher than the concentration (53 ± 14 nmol/mg) measured just before the first injection. Moreover, the mean concentration of glutathione (55.9 ± 8.8 nmol/mg) during 8 h after the third busulfan injection (Fig. 5d) was lower than the mean concentration (76.3 ± 17.1 nmol/mg) after the first dose. This statistically significant (p < 0.01) decrease of glutathione mean concentration (–27%) was explained by the observation that the metabolic system was unable to sustain the mean concentration of glutathione precursors, -glutamyl-cysteine and cysteine, after the third injection. The mean concentration of -glutamyl-cysteine (Fig. 5e) was 1.22 ± 0.35 nmol/mg after the third busulfan injection compared with a mean concentration of 1.59 ± 0.50 nmol/mg after the first dose. This difference was not statistically significant. Furthermore, the mean concentration of cysteine (Fig. 5f) after the third busulfan injection with 4.13 ± 1.19 nmol/mg was lower than the mean concentration of 5.59 ± 1.61 nmol/mg after the first dose (p < 0.05). Mice treated with busulfan 33 mg/kg once daily were unable to sustain glutathione level and to adapt their metabolism through -glutamyl-cysteine and cysteine accumulation. The mean concentration of homocysteine after the fifth busulfan injection was 0.29 ± 0.04 nmol/mg compared with a mean concentration after the first dose of 0.35 ± 0.05 nmol/mg. The mean concentration of cysteinyl-glycine after the fifth busulfan injection was 1.05 ± 0.21 nmol/mg compared with a mean concentration of 0.97 ± 0.36 nmol/mg after the first dose and a stable -glutamyltranspeptidase activity of 8.1 ± 2.0 mU/mg at the third dose.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
Mice were treated with busulfan during 4 days at a total dose of 132 mg/kg according to two different schedules of administration: 33 mg/kg once daily versus 16.5 mg/kg twice daily. Under these conditions, busulfan 132 mg/kg given according to the once daily schedule leads to an excessive toxicity that is not compatible with the BMT procedure. No significant histological alterations in the lung and the intestine were observed at the onset of immediate toxicity. Microvesicular steatosis was observed in the liver, which may be associated in some conditions with altered GSH synthesis at the mitochondrial level (Fernandez-Checa and Kaplowitz, 2005). Under our conditions, these histological features may be qualified as nonspecific since they were associated with severe weight loss. The animals were probably not eating adequately because of lethargy. However, there was no major alteration of the metabolic capacity since GSH was not decreased at steady state.

On the other hand, animals treated twice daily properly tolerated the treatment, with adaptation phenomena assessable on the weight variations over the 4 days of treatment. The animals treated twice daily were efficiently rescued with a syngeneic BMT. Late tissue-specific toxicity after BMT, with growth retardation and hair-graying, was similar to previous observations (Down et al., 1989).

The difference in terms of immediate toxicity, related to the frequency of busulfan injections, may be explained through a difference of plasma clearance. A significant induction of the clearance was observed in mice treated twice daily. It is the first time that an adaptation of busulfan clearance during repeated injections is reported. Because busulfan was injected intraperitoneally, we specify that the ratio of drug-amount (A) and area under plasma concentration-time curve (AUC) estimate the clearance "uncorrected for bioavailability" (CL/F = A/AUC). However, previous experiments performed in our laboratory (Boland et al., 1999) have shown that busulfan bioavailability in mice was close to 100% when busulfan was given intraperitoneally (CL/F CL).

Busulfan is a unique example of a drug almost exclusively eliminated through liver metabolism with conjugation by glutathione S-transferases (Czerwinski et al., 1996; Gibbs et al., 1998). Induction of clearance may thus be due to an adaptation of busulfan metabolism through the GSH/GST pathway. It may concern either the conjugation rate (GST) or the substrate availability (GSH). This animal experiment allowed the study of the GSH/GST pathway regulation during a strong electrophile stress.

We showed that the induction of clearance is not due to an induction of the specific busulfan GST activity. GST activity was determined in vitro since the plasma ratio THT⁺/busulfan determined in vivo did not allow accurate estimation of the conjugation rate. This was probably due to a high rate of elimination of THT⁺ in mice, contrary to the observations done in humans with THT⁺ accumulation (Gibbs et al., 1997). On the other hand, the induction of busulfan clearance was correlated with enhanced GSH synthesis. In our conditions, GSH synthesis rate through availability of precursors is the limiting parameter for conjugation of busulfan to GSH. The induction of the clearance was very rapid, and a saturation of this biochemical pathway was observed for a dose of busulfan between 16.5 and 33 mg/kg/injection. It is a relevant observation since the synthesis of GSH influenced directly the clearance and the toxicity of busulfan in C57BL6 male mice. Previous studies have shown that GSH stressors such as acetaminophen frequently lead to a rebound effect of GSH synthesis after a prior depletion (Buttar et al., 1977).

Our study is the first report of the in vivo flux of liver GSH synthesis being so closely related to the plasma clearance and toxicity of a drug. GCS activity has been presented for a long time as the rate-limiting step of GSH synthesis, because this enzymatic activity is inhibited in vitro by the high level of GSH (Richman and Meister, 1975). Our study in mice supports the recently published results observed in yeasts (Lafaye et al., 2005). It was proposed that the enzymes of the metabolic pathway are not saturated under standard conditions and that increasing the pool of cysteine is sufficient to increase -glutamyl-cysteine level and GSH synthesis. Enhanced cysteine flux through modification of the intracellular metabolism with enhanced anabolism (Lafaye et al., 2005) or decreased catabolism (Lee et al., 2004) may thus be a determinant for glutathione synthesis. Furthermore, this observation is very interesting for therapeutic purposes. In children receiving high-dose busulfan every 6 h over 4 consecutive days, busulfan plasma levels exhibited a significant circadian rhythm with a higher mean level at 6:00 A.M. compared with that at 12:00 P.M., 6:00 P.M., and midnight (Vassal et al., 1993). It was proposed that this circadian rhythm was due to hepatic GSH content (Vassal et al., 1993) since hepatic GSH, and not GST, exhibits circadian variations (Belanger et al., 1991). These observations are strong arguments to hypothesize that GSH synthesis in the liver may directly influence busulfan clearance, not only in mice, but also in humans.

It was proposed a long time ago that busulfan administered first in a BMT conditioning regimen might deplete the hepatic GSH content and sensitize the liver to the toxicity of the second alkylating agent, melphalan (Meresse et al., 1992; Bouligand et al., 2003) or cyclophosphamide (Hassan et al., 2000). This hypothesis was supported by clinical observations and laboratory experiments. Shulman et al. (1987) showed that melphalan injected after a strong alteration of GSH synthesis can produce HVOD in dogs. DeLeve (1996) demonstrated with various in vitro or ex vivo experiments that SECs isolated from mouse liver are significantly more susceptible than hepatocytes to the toxicity of various toxins implicated in HVOD. During these in vitro experiments, the profound depletion of GSH preceded the onset of toxicity (DeLeve, 1996). This observation explains why injury is initiated on SECs and not on hepatocytes during HVOD. On the other hand, the differences observed between SECs and hepatocytes could be a consequence of cell isolation procedure or cell culture conditions. Furthermore, only high concentrations of busulfan (250 µg/ml), not achievable in vivo, were able to profoundly deplete GSH in murine hepatic cells in vitro (DeLeve and Wang, 2000).

Our study is original because it elucidates the in vivo role of GSH synthesis during high-dose busulfan treatment. We showed that mice treated with high-dose busulfan sustain efficiently their liver GSH content through activation of synthesis and cysteine accumulation. In clinical practice, patients receive busulfan at a total dose of 16 mg/kg in adults and up to 30 mg/kg in children. Busulfan is generally given every 6 h over 4 consecutive days. According to whether it is given by either oral or intravenous administration, mean busulfan concentrations are far below the concentrations determined in mice receiving 132 mg/kg (Bouligand et al., 2003). On the other hand, the GSH content in human liver with 50 nmol/mg (Srivastava et al., 2004) is quite similar to the content determined in mice. It is thus unlikely that high-dose busulfan profoundly depletes GSH content in human liver in therapeutic conditions. Another mechanism of drug interaction may be proposed, such as the activation of different metabolic pathways enabling increase of GSH synthesis in response to the electrophile stress. The conception of GSH/GST pathway may be reviewed considering that this metabolic pathway is also a sensor of the chemical stress (Griffith, 1999). It has been demonstrated through the Nrf2/keap-1 pathway (Motohashi and Yamamoto, 2004) that several phase 2 genes are activated in response to chemical stress in parallel with the activation of GSH synthesis. Moreover, the role of many other pathways is still unknown (Tsuji, 2005). We think that the electrophile stress during busulfan treatment, detected through intensive GSH conjugation, may enhance the expression of several proteins, procoagulant, proinflammatory, or profibrotic, participating in the biological events characterizing HVOD (DeLeve et al., 2002). The second alkylating agent administered in the conditioning regimen would initiate the systemic disease through alteration of the endothelium and cytokine release. This hypothesis is supported by recent observations showing that endothelial cells enhanced their production of thromboplastin (Ritter et al., 2002) or activin-A (Dressel et al., 2003) during exposure to busulfan through an efficient GSH conjugation. The activation of the GSH pathway could be related to the elevation of homocysteine levels during HVOD (Gerecitano et al., 2003). It is noteworthy that cysteine and homocysteine are potentially toxic for the endothelium (Adinolfi et al., 2005). Moreover, cysteine was shown to be a strong reducer of iron, and elevation of cysteine could thus maintain free iron in the toxic ferrous form responsible for the "Fenton reaction" (Park and Imlay, 2003). That is very interesting because it has been proposed that an iron-generated pro-oxidant state may partially explain the occurrence of HVOD during BMT (Evens et al., 2004).

In summary, we showed that activation of GSH synthesis through the increase of liver cysteine content is essential for clearance and detoxification of busulfan in mice. To our knowledge, it is the first time that the in vivo flux of liver GSH synthesis has been shown to be so closely related to plasma clearance and toxicity of a drug.

	Acknowledgments

 
We thank Patrice Ardouin and the staff of the animal unit at the Institute Gustave Roussy for their care of the animals.

	Footnotes

 
This work was supported by Association de la Recherche sur le Cancer, Académie Nationale de Médecine, Institute Gustave Roussy, and Centre National de la Recherche Scientifique, Ligue contre le Cancer, and Paris XI University.

Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.

doi:10.1124/dmd.106.012880.

ABBREVIATIONS: GSH, reduced glutathione; AUC, area under concentration-time curve; BMT, bone marrow transplantation; CL/F, clearance uncorrected for bioavailability; GCS, -glutamyl-cysteine synthetase; F, bioavailability; GC, gas chromatography; GST, glutathione S-transferase; k_e, terminal phase elimination rate constant; LC, liquid chromatography; MS, mass spectrometry; MS/MS, tandem mass spectrometry; THT, tetrahydrothiophene; THT⁺, sulfonium ion of glutathione; t_1/2, half-life.

Address correspondence to: Professor Gilles Vassal, UPRES EA3535, Pharmacology and New Treatments of Cancers, Institut Gustave Roussy, 39 rue Camille Desmoulins, Villejuif 94800, France. E-mail: gilles.vassal{at}igr.fr

	References

Top Abstract Materials and Methods Results Discussion References

 


Abernathy CO, Hodgson E, and Guthrie FE (1971) Structure-activity relationships on the induction of hepatic microsomal enzymes in the mouse by 1,1,1-trichloro-2,2-bis(p-chlorophenyl) ethane(DDT) analogs. Biochem Pharmacol 20: 2385–2393.[CrossRef][Medline]

Adinolfi LE, Ingrosso D, Cesaro G, Cimmino A, D'Anto M, Capasso R, Zappia V, and Ruggiero G (2005) Hyperhomocysteinemia and the MTHFR C677T polymorphism promote steatosis and fibrosis in chronic hepatitis C patients. Hepatology 41: 995–1003.[CrossRef][Medline]

Belanger PM, Desgagne M, and Bruguerolle B (1991) Temporal variations in microsomal lipid peroxidation and in glutathione concentration of rat liver. Drug Metab Dispos 19: 241–244.[Abstract]

Boland I, Vassal G, Morizet J, Terrier-Lacombe MJ, Valteau-Couanet D, Kalifa C, Hartmann O, and Gouyette A (1999) Busulphan is active against neuroblastoma and medulloblastoma xenografts in athymic mice at clinically achievable plasma drug concentrations. Br J Cancer 79: 787–792.[CrossRef][Medline]

Bouligand J, Boland I, Valteau-Couanet D, Deroussent A, Kalifa C, Hartmann O, and Vassal G (2003) In children and adolescents, the pharmacodynamics of high-dose busulfan is dependent on the second alkylating agent used in the combined regimen (melphalan or thiotepa). Bone Marrow Transplant 32: 979–986.[CrossRef][Medline]

Bouligand J, Deroussent A, Morizet J, Paci A, and Vassal G (2006) Liquid chromatography-tandem mass spectrometry assay of reduced and oxidized glutathione and main precursors in mice liver. J Chromatogr B Anal Technol Biomed Life Sci 832: 67–74.[Medline]

Bouligand J, Paci A, Mercier L, Vassal G, and Bourget P (2004) High-performance thin-layer chromatography with a derivatization procedure, a suitable method for the identification and the quantitation of busulfan in various pharmaceutical products. J Pharm Biomed Anal 34: 525–530.[CrossRef][Medline]

Buttar HS, Chow AY, and Downie RH (1977) Glutathione alterations in rat liver after acute and subacute oral administration of paracetamol. Clin Exp Pharmacol Physiol 4: 1–6.[Medline]

Czerwinski M, Gibbs JP, and Slattery JT (1996) Busulfan conjugation by glutathione S-transferases alpha, mu, and pi. Drug Metab Dispos 24: 1015–1019.[Abstract]

DeLeve LD (1996) Cellular target of cyclophosphamide toxicity in the murine liver: role of glutathione and site of metabolic activation. Hepatology 24: 830–837.[CrossRef][Medline]

DeLeve LD, Shulman HM, and McDonald GB (2002) Toxic injury to hepatic sinusoids: sinusoidal obstruction syndrome (veno-occlusive disease). Semin Liver Dis 22: 27–42.[CrossRef][Medline]

DeLeve LD and Wang X (2000) Role of oxidative stress and glutathione in busulfan toxicity in cultured murine hepatocytes. Pharmacology 60: 143–154.[CrossRef][Medline]

Down JD, Berman AJ, Warhol M, Van Dijken PJ, Ferrara JL, Yeap B, Hellman S, and Mauch PM (1989) Late tissue-specific toxicity of total body irradiation and busulfan in a murine bone marrow transplant model. Int J Radiat Oncol Biol Phys 17: 109–116.[Medline]

Dressel D, Ritter CA, Sperker B, Grube M, Maier T, Klingebiel T, Siegmund W, Beck JF, and Kroemer HK (2003) Busulfan induces activin A expression in vitro and in vivo: a possible link to venous occlusive disease. Clin Pharmacol Ther 74: 264–274.[CrossRef][Medline]

Evens AM, Mehta J, and Gordon LI (2004) Rust and corrosion in hematopoietic stem cell transplantation: the problem of iron and oxidative stress. Bone Marrow Transplant 34: 561–571.[CrossRef][Medline]

Fernandez-Checa JC and Kaplowitz N (2005) Hepatic mitochondrial glutathione: transport and role in disease and toxicity. Toxicol Appl Pharmacol 204: 263–273.[CrossRef][Medline]

Gerecitano J, Mathias C, Mick R, Duffy KM, Luger S, Stadtmauer EA, Schuster SJ, Tsai D, Nasta S, Berlin J, et al. (2003) Homocysteine and prothrombin fragment 1+2 levels in patients with veno-occlusive disease after stem cell transplantation. J Hematother Stem Cell Res 12: 215–223.[CrossRef][Medline]

Gibbs JP, Murray G, Risler L, Chien JY, Dev R, and Slattery JT (1997) Age-dependent tetrahydrothiophenium ion formation in young children and adults receiving high-dose busulfan. Cancer Res 57: 5509–5516.[Abstract/Free Full Text]

Gibbs JP, Yang JS, and Slattery JT (1998) Comparison of human liver and small intestinal glutathione S-transferase-catalyzed busulfan conjugation in vitro. Drug Metab Dispos 26: 52–55.[Abstract/Free Full Text]

Griffith OW (1999) Biologic and pharmacologic regulation of mammalian glutathione synthesis. Free Radic Biol Med 27: 922–935.[CrossRef][Medline]

Hartmann O, Benhamou E, Beaujean F, Pico JL, Kalifa C, Patte C, Flamant F, and Lemerle J (1986) High-dose busulfan and cyclophosphamide with autologous bone marrow transplantation support in advanced malignancies in children: a phase II study. J Clin Oncol 4: 1804–1810.[Abstract]

Hassan M and Ehrsson H (1987) Urinary metabolites of busulfan in the rat. Drug Metab Dispos 15: 399–402.[Abstract]

Hassan M, Ljungman P, Ringden O, Hassan Z, Oberg G, Nilsson C, Bekassy A, Bielenstein M, Abdel-Rehim M, Georen S, et al. (2000) The effect of busulphan on the pharmacokinetics of cyclophosphamide and its 4-hydroxy metabolite: time interval influence on therapeutic efficacy and therapy-related toxicity. Bone Marrow Transplant 25: 915–924.[CrossRef][Medline]

Lafaye A, Junot C, Pereira Y, Lagniel G, Tabet JC, Ezan E, and Labarre J (2005) Combined proteome and metabolite-profiling analyses reveal surprising insights into yeast sulfur metabolism. J Biol Chem 280: 24723–24730.[Abstract/Free Full Text]

Lee JI, Londono M, Hirschberger LL, and Stipanuk MH (2004) Regulation of cysteine dioxygenase and gamma-glutamylcysteine synthetase is associated with hepatic cysteine level. J Nutr Biochem 15: 112–122.[CrossRef][Medline]

McDonald GB, Hinds MS, Fisher LD, Schoch HG, Wolford JL, Banaji M, Hardin BJ, Shulman HM, and Clift RA (1993) Veno-occlusive disease of the liver and multiorgan failure after bone marrow transplantation: a cohort study of 355 patients. Ann Intern Med 118: 255–267.[Abstract/Free Full Text]

Meresse V, Hartmann O, Vassal G, Benhamou E, Valteau-Couanet D, Brugieres L, and Lemerle J (1992) Risk factors for hepatic veno-occlusive disease after high-dose busulfan-containing regimens followed by autologous bone marrow transplantation: a study in 136 children. Bone Marrow Transplant 10: 135–141.[Medline]

Motohashi H and Yamamoto M (2004) Nrf2-Keap1 defines a physiologically important stress response mechanism. Trends Mol Med 10: 549–557.[CrossRef][Medline]

Park S and Imlay JA (2003) High levels of intracellular cysteine promote oxidative DNA damage by driving the fenton reaction. J Bacteriol 185: 1942–1950.[Abstract/Free Full Text]

Peters WP, Henner WD, Grochow LB, Olsen G, Edwards S, Stanbuck H, Stuart A, Gockerman J, Moore J, and Bast RCJ (1987) Clinical and pharmacologic effects of high dose single agent busulfan with autologous bone marrow support in the treatment of solid tumors. Cancer Res 47: 6402–6406.[Abstract/Free Full Text]

Richman PG and Meister A (1975) Regulation of gamma-glutamyl-cysteine synthetase by nonallosteric feedback inhibition by glutathione. J Biol Chem 250: 1422–1426.[Abstract/Free Full Text]

Ritter CA, Bohnenstengel F, Hofmann U, Kroemer HK, and Sperker B (1999) Determination of tetrahydrothiophene formation as a probe of in vitro busulfan metabolism by human glutathione S-transferase A1-1: use of a highly sensitive gas chromatographic-mass spectrometric method. J Chromatogr B Biomed Sci Appl 730: 25–31.[CrossRef][Medline]

Ritter CA, Sperker B, Grube M, Dressel D, Kunert-Keil C, and Kroemer HK (2002) Overexpression of glutathione S-transferase A1-1 in ECV 304 cells protects against busulfan mediated G2-arrest and induces tissue factor expression. Br J Pharmacol 137: 1100–1106.[CrossRef]

Santos GW, Tutschka PJ, Brookmeyer R, Saral R, Beschorner WE, Bias WB, Braine HG, Burns WH, Elfenbein GJ, and Kaizer H (1983) Marrow transplantation for acute nonlymphocytic leukemia after treatment with busulfan and cyclophosphamide. N Engl J Med 309: 1347–1353.[Abstract]

Shulman HM, Luk K, Deeg HJ, Shuman WB, and Storb R (1987) Induction of hepatic veno-occlusive disease in dogs. Am J Pathol 126: 114–125.[Abstract]

Srivastava A, Poonkuzhali B, Shaji RV, George B, Mathews V, Chandy M, and Krishnamoorthy R (2004) Glutathione S-transferase M1 polymorphism: a risk factor for hepatic venoocclusive disease in bone marrow transplantation. Blood 104: 1574–1577.[Abstract/Free Full Text]

Tsuji Y (2005) JunD activates transcription of the human ferritin H gene through an antioxidant response element during oxidative stress. Oncogene 24: 7567–7578.[CrossRef][Medline]

Vassal G, Challine D, Koscielny S, Hartmann O, Deroussent A, Boland I, Valteau-Couanet D, Lemerle J, Levi F, and Gouyette A (1993) Chronopharmacology of high-dose busulfan in children. Cancer Res 53: 1534–1537.[Abstract/Free Full Text]

Vassal G, Re M, and Gouyette A (1988) Gas chromatographic-mass spectrometric assay for busulfan in biological fluids using a deuterated internal standard. J Chromatogr 428: 357–361.[Medline]

Wallin B, Rosengren B, Shertzer HG, and Camejo G (1993) Lipoprotein oxidation and measurement of thiobarbituric acid reacting substances formation in a single microtiter plate: its use for evaluation of antioxidants. Anal Biochem 208: 10–15.[CrossRef][Medline]

This article has been cited by other articles:

				I. R. Younis, M. Elliott, C. J. Peer, A. J. L. Cooper, J. T. Pinto, G. W. Konat, M. Kraszpulski, W. P. Petros, and P. S. Callery Dehydroalanine Analog of Glutathione: An Electrophilic Busulfan Metabolite That Binds to Human Glutathione S-Transferase A1-1 J. Pharmacol. Exp. Ther., December 1, 2008; 327(3): 770 - 776. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: dmd.106.012880v1 35/2/306    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bouligand, J. Articles by Vassal, G. Search for Related Content PubMed PubMed Citation Articles by Bouligand, J. Articles by Vassal, G.