Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Monoamine oxidases (EC 1.4.3.4.; MAO) are flavoproteins integrated to outer mitochondrial membranes. They oxidatively deaminate neurotransmitters and xenobiotic amines. The enzyme exists in two isoforms, MAO-A and MAO-B, which differ in primary structure, substrate specificity, and inhibitor sensitivity. In the brain, MAO-A oxidizes mainly serotonin, norepinephrine, and dopamine and is found primarily in catecholaminergic neurones, whereas MAO-B oxidizes phenethylamine and dopamine and is mainly localized in serotoninergic neurones and glial cells. Both forms are important for regulation of monoaminergic transmission. Fluctuation in functional MAO activity may be associated with human diseases such as Parkinson's and Alzheimer's diseases, depression, and certain psychiatric disorders (Brunner et al., 1993). Moreover, it was shown that heavy smokers possess lower brain MAO-A and MAO-B levels (Fowler et al., 1996a,b) and that their inhibition facilitates smoking cessation (Berlin et al., 1995).

Positron emission tomography (PET) is a noninvasive imaging technique allowing the study of metabolism, receptors, and in the present case, enzymes in living human brain (Fowler et al., 1987; Pappata et al., 1996). Few radiotracers have been developed for MAO-A PET studies such as [¹¹C]clorgyline and [¹¹C]harmine (Fowler et al., 1987, 2000; Bergström et al., 1997a,b,c). However, both tracers present some drawbacks. A recent study show that [¹¹C]clorgyline present an unexplained species difference in that [¹¹C]clorgyline was not retained in baboon brain in contrast to results in human (Fowler et al., 2001). [¹¹C]Harmine is extensively metabolized in plasma; therefore, the input function determination (unchanged compound) is subject to large errors (Bergström et al., 1997a). Brofaromine was the first reversible MAO-A inhibitor labeled for PET (with ¹¹C), but the brain uptake could not be significantly reduced by a pretreatment with either irreversible (clorgyline) or reversible (moclobemide) MAO-A inhibitors (Ametamey et al., 1996).

Befloxatone, an oxazolidinone derivative, is a selective MAO-A reversible inhibitor with potential antidepressant properties (Curet et al., 1994; Rovei et al., 1994; Wouters et al., 1999). The biochemical and pharmacological profiles of befloxatone were extensively studied in rats and in human tissues (Caille et al., 1996; Curet et al., 1996). These studies revealed that befloxatone is a selective, competitive, specific, and reversible MAO-A inhibitor in rat as well in human tissues. Befloxatone is very potent at inhibiting in vitro MAO-A activity (K_i = 2 versus 150 nM for MAO-B), more potent than the other reversible MAO-A inhibitors (including moclobemide, brofaromine, and harmaline). The specificity of befloxatone for MAO-A was confirmed against other neurotransmitter and transporter systems. In vivo, befloxatone increases brain levels of norepinephrine, dopamine, and serotonin and decreases the levels of corresponding deaminated metabolites (dihydroxyphenylacetic acid and 5-hydroxyindolacetic acid (Curet et al., 1996). Befloxatone is much more potent (10- to 500-fold) than other reversible or irreversible MAO-A inhibitors in classical antidepressant tests in rodents (Caille et al., 1996). Befloxatone displays a higher activity in test involving MAO-A (with and ED₅₀ value of about 0.1 to 0.2 mg/kg p.o. or 0.07 mg/kg i.v.) than in test involving MAO-B (an ED₅₀ value of about 58 mg/kg p.o.) (Caille et al., 1996).

Due to its selective and reversible MAO-A inhibition, befloxatone possesses a good safety profile, particularly regarding the observation of tyramine induced pressor effect and the drug interactions (Caille et al., 1996; Rosenzweig et al., 1998).

All the biochemical and pharmacological characteristics of the drug suggest that befloxatone is an excellent candidate for the PET study of the MAO-A in human. Moreover, the structure of the molecule allows labeling with carbon-11, a positron emitter isotope, using phosgene.

This work aimed to characterize the properties of the drug in living brain and to determine whether [¹¹C]befloxatone is a suitable radioligand to study the MAO-A by PET in vivo. This preclinical study was performed in nonhuman primate using PET.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. All animal use procedures were in strict accordance with the recommendations of the European Community (86/609/CEE) and the French National Committee (décret 87/848) for the care and use of laboratory animals. Twelve PET experiments were carried out on four male Papio anubis baboons (mean weight 14.4 ± 5 kg). Each animal was allowed to rest for a period of at least 2 weeks between PET studies.

Drugs. Befloxatone [(5R)-5-(methoxymethyl)-3-[4-[(3R)-4,4,4-trifluoro-3-hydroxybutoxy]phenyl]-2-oxazolidinone)], moclobemide, and lazabemide were provided by Synthélabo (Bagneux, France). All drugs were dissolved in a mixture of physiological saline/ethanol/propanediol (98:2:2, v/v/v) and were injected intravenously.

Magnetic Resonance Imaging. A magnetic resonance imaging (MRI) examination was performed for each baboon to provide detailed anatomical images corresponding to the PET slices. The examinations were performed with a 1.5-Tesla SIGNA system (General Electric, Milwaukee, WI), and a custom-made receive-only coil was used in proximity to the baboon's head to provide a high sensitivity. The animal was anesthetized by i.m. injection of a mixture of ketamine (15 mg/kg, Ketalar; PanPharma, Fougere, France) and xylazine (1.5 mg/kg, Rompun; Bayer, Puteaux, France) (Banknieder et al., 1978) and positioned using a stereotaxic head-holder so that the stereotaxic reference plane was aligned with a laser beam figuring the axial reference plane of the MR scanner. The imaging protocol used a T₁-weighted inversion-recovery sequence in three-dimensional mode and a 256 × 192 matrix over 124 slices of 1.5 mm in thickness.

Positron Emission Tomography Data Acquisition. PET studies were performed with a high-resolution tomograph (ECAT 953B/31; CTI PET system; CTI-Siemens, Knoxville, TN), which allowed reconstruction of 31 slices every 3.3 mm with spatial and axial resolution of 5.7 and 5.0 mm, respectively (Bendriem, 1991). Animals were anesthetized using isoflurane/nitrous-oxide mixture (1:66%), controlled by an Ohmeda ventilator (Ohmeda OAV 7710, Madison, WI) with 33% oxygen. The tidal volume was adjusted to achieve stable end-tidal carbon dioxide tension between 38 to 40 mm Hg. The baboon's head was fixed in a head-holder and positioned in the scanner gantry for axial plane acquisition. A transmission scan (⁶⁸Ge rods, 15 min) was recorded to correct for -ray attenuation.

Input Function and Metabolite Studies. During each PET acquisition, arterial blood samples (n = 30) were withdrawn from a femoral artery at designated times. Blood and plasma radioactivity were measured in a gamma-counter, and the blood-plasma time-activity curves were corrected for [¹¹C] decay from the beginning of the PET acquisition (T_{0 min}).

PET Data Analysis. Regions of interest (ROI) for anatomical structures such as cortex, cerebellum, pons, thalamus, and striata were delineated on MRI images of each baboon and then transferred to the PET studies of the corresponding animal. The concentration of radioactivity in each ROI was determined on each sequential scan and expressed as a percentage of the injected dose per 100 ml of tissue. Distribution volumes for individual ROI (DV) were calculated using the graphic analysis developed by Logan et al. (1990) for reversible tracers. This computation required the uptake PET data from a ROI versus time and input function, which is the unchanged tracer concentration in arterial plasma. As the plasma input function depends on the amount of befloxatone administered (see Results section), the ratio [brain TACs] to [plasma AUC_{0120 min}] were calculated to allow comparison between PET experiments.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Blood and Plasma Time Activity Curves. After [¹¹C]befloxatone causes a rapid phase of distribution in blood (about 5 min), the decline in radiotracer plasma concentration was fitted by a monoexponential function (intercept = 0. 14% ID/100 ml of plasma, t_1/2 = 56.8 min in control experiment and intercept = 0.83% ID/100 ml of plasma, t_1/2 = 86.6 min in saturation experiments). As shown in Fig. 1, the [¹¹C]befloxatone plasma TACs showed a marked difference between control experiments and saturation experiments (preinjection of high amount of unlabeled befloxatone). After the distribution phase, the plasma radioactive concentration was 8-fold higher in the saturation experiments than in the control experiments (0.6 ± 0.2 versus 0.07 ± 0.03% ID/100 ml, respectively, at T_{30 min}). Similarly, in displacement experiments, the administration of unlabeled befloxatone at T_{30 min} induced an increase of radiotracer plasma concentration. In these experiments, the plasma radioactive level, measured at 90 min (1 h after cold drug administration), was correlated with the amount of injected cold drug (r = 0.994; p < 0.001). In all experiments (n = 12), the blood to plasma concentration ratio was constant during the 2 h of the experimentation (0.9 ± 0.06) consistent with a similar kinetic of [¹¹C]befloxatone in both blood and plasma.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   [11C]Befloxatone plasma TACs in control (; n = 2) and saturation experiments with befloxatone (; n = 2). Values are expressed a percentage of ID per 100 ml of plasma. The inset represent the percentage of nonmetabolized [11C]befloxatone in plasma during the control experiments.

Metabolite Analysis. [¹¹C]Befloxatone was extracted from plasma with an extraction efficiency of 96%. After tracer injection, [¹¹C]befloxatone was relatively stable in vivo since 78% of the radioactivity in plasma at T_{30 min} represented unchanged befloxatone (inset in Fig. 1). The detectable metabolite has a shorter retention time (2.1 min), indicating a more hydrophilic compound. No peak was detectable at the retention time of the desmethylbefloxatone (5.3 min).

Cerebral Distribution of [¹¹C]Befloxatone. Tomographic images obtained after i.v. [¹¹C]befloxatone display a heterogeneous distribution of radioactivity (Fig. 2). A high uptake was found in basal ganglia, thalamus, pons, and cortical structures, whereas cerebellum and white matter present lower uptake (Table 1). Similar results were found using the calculated DV in the different structures; high DV values (19-23 ml/g) in he basal ganglia, thalamus, and cortex and low values (10-11 ml/g) in cerebellum and white matter (Table 1). The DV values in the different brain structures are strongly correlated with the uptake values obtained directly from the PET image (30 min after tracer injection) (r = 0.83; p < 0.001).

View larger version (96K): [in this window] [in a new window]   Fig. 2.   PET axial brain slices obtained 30 min after i.v. [11C]befloxatone and superimposed with the MRI image. Radioactive concentration is reported in color. A high uptake was found in basal ganglia, thalamus, pons, and cortical structures, whereas cerebellum and white matter present lower uptake.

Cerebral Time Activity Curves. After i.v. [¹¹C]befloxatone, radioactivity was detected early in the brain (in the first 30-s PET image) and increased rapidly. In brain structures with a high uptake, the radioactivity reached maximal values at 30 min (Table 1) and then decreased slowly until the end of the experiments (Fig. 3). The tracer TACs was slightly different in structures with lower uptake such as cerebellum and white matter. In these structures, the maximal uptake was reached in less than 5 min and the washout was faster (Fig. 3).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   [11C]Befloxatone TACs in several brain structures in a control experiment (radiotracer is injected i.v. at T0 min). Radioactivity in plasma is also displayed. Values are expressed in a percentage of the ID per 100 ml of tissue.

Saturability of [¹¹C]Befloxatone Brain Uptake. Preinjection of unlabeled befloxatone (0.4 mg/kg) before the radiotracer prevents the tracer from accumulating in the brain (Fig. 4). The radioactive concentrations were very low and identical in all structures (0.25 ± 0.035% ID/100 ml). After correction of the plasma input function, the nonspecific uptake represents less than 5% (from T_{45 min}) of the total uptake.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   [11C]Befloxatone TACs in the frontal cortex obtained in the control (; n = 2) and saturation with befloxatone (; n = 2) experiments and compared with TACs obtained after pretreatment with the MAO-A specific inhibitor moclobemide (10 mg/kg; ) and the MAO-B specific inhibitor lazabemide (0.5 mg/kg; ). Values are PET data corrected by the plasma AUC0120.

Selectivity of [¹¹C]Befloxatone Brain Uptake. Brain uptake of [¹¹C]befloxatone was blocked by pretreatment with the MAO-A specific inhibitor moclobemide (10 mg/kg; Fig. 4). Preinjection of the MAO-B specific inhibitor lazabemide (0.5 mg/kg) did not induce significant change of the [¹¹C]befloxatone brain uptake compared with that observed in control experiments (Fig. 4).

Reversibility of [¹¹C]Befloxatone Brain Uptake. Increasing doses of unlabeled befloxatone were administered 30 min after the tracer injection (T_{30 min}) in separate experiments. Figure 5 shows that, compared with control cerebral TACs of [¹¹C]befloxatone, administration of befloxatone produced a washout of the radioactivity. The new equilibrium between brain and plasma was reached at T_{95 min}, i.e., 60 min after the unlabeled befloxatone administration.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   [11C]Befloxatone TACs in the frontal cortex obtained in four separates displacement experiments () compared with the control () and saturation with befloxatone () experiments. To make the figure more readable, data are normalized to the value at 30 min.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Dose-displacement relationship in four brain structures. Experimental data () are fitted using the logistic model (line).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present study shows that befloxatone, labeled with a positron emitter (carbon 11), is an excellent tool for PET assessment of MAO-A binding sites as demonstrated by displacement and saturation experiments. In vivo, [¹¹C]befloxatone penetrates and accumulates rapidly in the brain and displays a high specific uptake that can be rapidly and dose dependently reversed. [¹¹C]Befloxatone presents a high selectivity for the isoform A of MAO. These results, obtained in vivo, confirm the biochemical and pharmacological profile of befloxatone found in rodent and in human tissues (Caille et al., 1996; Curet et al., 1996).

[¹¹C]Befloxatone was rapidly distributed in the body after i.v. administration, as shown by the distribution phase of the radioactivity in plasma. During the elimination phase, the [¹¹C]befloxatone plasma concentration was very low when a tracer dose is administered. This concentration was dramatically increased (Fig. 1) when a high amount of MAO-A inhibitor (befloxatone or moclobemide) was administered before the tracer. This is due to the presence of high level of MAO-A sites in several peripheral organs such as the liver, kidney, myocardium, and duodenum (Saura et al., 1996). Indeed, the saturation of theses sites by the pretreatment dose of MAO-A inhibitors reduces the distribution volume of the [¹¹C]befloxatone injected and subsequently increase its plasma concentration. This feature was also found in humans using an irreversible MAO-A specific ligand [¹¹C]harmine after a 7-day treatment with a MAO-A inhibitors (Bergström et al., 1997c). It is therefore obvious that brain TACs of [¹¹C]befloxatone, which are highly dependent of the plasma input function, have to be corrected by this parameter (see Materials and Methods).

[¹¹C]Befloxatone at a tracer dose, was relatively stable in vivo. The only radiolabeled metabolite detected at T_{30 min} in the plasma accounted for about 20% of the total plasma radioactivity. The short retention time of this compound indicates a low lipophilicity, suggesting a poor brain penetration of this compound.

The distribution of radioactivity measured in vivo by PET in different brain regions (Fig. 2) paralleled the areas of MAO-A concentration determined in vitro in primate brain either by histochemistry (Westlund et al., 1985) or autoradiography (Saura et al., 1992). Similar results were found in humans using the irreversible MAO-A PET ligand [¹¹C]clorgyline (Fowler et al., 1987). As in in vitro studies, we did not find any region that was devoid of MAO-A and significant radioactivity was even detectable in white matter.

The noradrenergic system is thought to be the main compartment of MAO-A in most species examined, including monkey and human brain. Autoradiographical and histochemical studies and transcript expression of MAO-A has been clearly imaged most abundant in the human locus coeruleus (3 to 5 times higher than in cortex) and, to a lesser extent, in the interpeduncular nuclei (Westlund et al., 1988; Willoughby et al., 1988; Saura et al., 1992; Luque et al., 1996). In PET studies, due to the small size of these structures and the limited resolution of the technique, it is not possible to clearly identify these nuclei within the pons. Our so-called "pons" region that contains these nuclei displayed a high radioactivity. Compared with the values we obtained in the cortex, however, the activity in the pons was not as high as expected from in vitro studies. This can be explained by the fact that, due to their small size, compared with the resolution of the PET camera, the nuclei are more sensitive than larger structures to partial volume effect. Thus, PET-measured activity in these nuclei is underestimated when compared with structures such as cortical regions and even basal ganglia.

The brain uptake of [¹¹C]befloxatone showed that the tracer freely crosses the blood-brain barrier and binds rapidly to its target sites with a maximum at 30 min. This is consistent with studies in rodents showing that maximal inhibition of MAO-A brain activity is obtained at 30 to 60 min after befloxatone administered p.o.(Curet et al., 1996). The very low residual [¹¹C]befloxatone uptake (less than 5% of the total uptake) obtained after injection of a high dose of unlabeled befloxatone indicates that the most part of the brain radioactivity is specifically bound to MAO-A. Moreover, pretreatment with another selective MAO-A inhibitor, moclobemide (Da Prada et al., 1989), has the same effect. Only a low residual radioactivity was observed, confirming that [¹¹C]befloxatone binds in vivo with a high specificity to the MAO-A.

To test in vivo the MAO isoform selectivity of [¹¹C]befloxatone, a high dose of lazabemide, a selective MAO-B inhibitor (Haefely et al., 1990), was injected before [¹¹C]befloxatone. The lazabemide dose (0.5 mg/kg) was chosen high enough to saturate MAO-B isoform (Bench et al., 1991). In our study, in spite of this high dose of lazabemide, the brain uptake of [¹¹C]befloxatone was not decreased (Fig. 4). This results demonstrates that in vivo [¹¹C]befloxatone binds with a high specificity and selectivity to MAO-A. This confirms the high selectivity of befloxatone in vivo for MAO-A as found in the in vitro and ex vivo enzyme inhibition studies of befloxatone (Curet et al., 1996), as well as in the in vivo pharmacological profile (Caille et al., 1996).

Biochemical studies have shown that befloxatone binds to MAO-A in a reversible and competitive manner in rat and human tissues (Curet et al., 1996). To verify these characteristics in vivo, competition experiments were performed with increasing doses of unlabeled befloxatone administered after the tracer injection (at T_{30 min}). We showed that befloxatone displaces the specifically bound [¹¹C]befloxatone in all brain structures. The radioactivity washout was very rapid in that a new equilibrium between brain and plasma was achieved after 30 min. This confirms that, in vivo, the MAO-A befloxatone binding is rapidly reversible.

[¹¹C]Befloxatone was displaced in a dose-dependent manner, confirming the reversible interaction of befloxatone with MAO-A demonstrated in vitro (Curet et al., 1996; Wouters et al., 1999). In our study, we found that the estimated dose of befloxatone needed to displace half of the specifically bound radioactivity in brain is very low (about 0.02 mg/kg). The high affinity of befloxatone for MAO-A (K_i = 2.5 nM), associated with an extensive brain penetration may explain these results. Befloxatone was shown to be very potent in inhibiting the rat brain MAO-A activity (increase brain concentration of monoamines ED₅₀ = 0.06 mg/kg, p.o.; decrease monoamine metabolites ED₅₀ = 0.03 mg/kg, p.o.) (Curet et al., 1996).

In conclusion, the in vivo properties of befloxatone may bring new insights for the therapeutic use of befloxatone. The knowledge of the brain pharmacokinetics of befloxatone may be of great interest for a better control of the intensity and the duration of the enzyme inhibition. This, together with the determination of the dose of befloxatone that inhibits its target, may lead to a better definition of the therapeutical doses. Moreover, the competitivity of befloxatone binding to its target, demonstrated in vivo, tend to minimize the interaction with tyramine absorbed with the food and thus to avoid the cheese effect in patients, a major side effect of irreversible MAO inhibitors. All these features may provide a significant improvement in the therapeutic use of befloxatone.

The present article shows that in vivo [¹¹C]befloxatone enters readily the brain and binds to MAO-A in a potent, selective, and reversible manner. Moreover, [¹¹C]befloxatone presents a very low, nonsaturable uptake and is metabolized rather slowly. All these biochemical characteristics suggest that [¹¹C]befloxatone is a unique tool for the in vivo study of MAO-A brain.