Introduction

Top Abstract Introduction Methods Results Discussion References

Changes in the density of DAT sites have been described in several disorders of the central nervous system, including Parkinson's disease (Janowsky et al., 1987; Kaufman and Madras, 1991), Alzheimer's disease (Allard et al., 1990) and Tourette's syndrome (Singer et al., 1991). Because the DAT plays a major role in regulating DA transmission, radiolabeled tracers suitable for visualizing and measuring changes in these sites in vivo may provide valuable tools for the diagnosis and follow-up of treatment of neurodegenerative diseases associated with dopaminergic neurons.

A number of radiotracers that are inhibitors of DA uptake have been developed to visualize DA uptake sites with the use of PET, such as [¹⁸F]GBR 13119 (Kilbourn et al., 1989), [¹¹C]nomifensine (Aquilonius et al., 1987) and [¹⁸F]N-1-[1-(2-benzothienyl)-cyclo-hexyl]-piperidine (Ponchant et al., 1993). More recently, many analogs of cocaine have been developed as potent radioligands to explore this transporter with the use of PET or SPECT, such as [¹¹C]-CIT (Müller et al., 1993), [¹¹C]-CIT-fluoroethyl (Halldin et al., 1996), [¹²³I]-CIT (Innis et al., 1991), [¹²³I]-CIT-fluoropropyl (Abi-Dargham et al., 1996; Ishikawa et al., 1996; Kuikka et al., 1995b; Neumeyer et al., 1994) and [¹²³I]-CIT-FE (Kuikka et al., 1995a). Among these compounds, the best known iodinated derivative of cocaine -CIT permits exploration of the DAT in human diseases (Innis et al., 1993; Kuikka et al., 1995c; Seibyl et al., 1995). The feasibility and choice of quantification method depend on the specificity and in vivo kinetics of PET and SPECT radioligands. The equilibrium method requires similar association and dissociation rates, whereas the peak uptake equilibrium method is used for radioligands displaying fast uptake and washout kinetics, which provide peak activity soon after injection (Abi-Dargham et al., 1994, 1996). -CIT has great affinity in vitro for the DAT but also for both 5-HT and NE transporters (Boja et al., 1992; Neumeyer et al., 1991). Moreover, quantification with this tracer can be performed only 20 hr after injection (Brücke et al., 1993; Laruelle et al., 1993). To develop SPECT applications in human diseases, it is better to use tracers allowing quantification on the same day as injection, regardless of the quantification method used, thus limiting the use of -CIT.

Several other analogs have been described to improve selectivity and kinetic properties of -CIT. RTI-121, a 2-carboisopropyloxy analog of -CIT (Carroll et al., 1992; Scheffel et al., 1992), has better selectivity in vitro for the DAT compared with -CIT. However, its relatively high nonspecific uptake may be a serious disadvantage for the use of this tracer in SPECT imaging (Al Tikriti et al., 1995). Structure-affinity relationship studies on tropane derivatives have demonstrated that the nature and position of the substituent on the 3-phenyl ring are very important for ligand/transporter interactions (Abraham et al., 1992; Carroll et al., 1992) and that N-substitutions do not affect DAT binding properties (Neumeyer et al., 1994). Therefore, a new analog of cocaine, IPT, was developed and tested (Goodman et al., 1994; Kung et al., 1995). When labeled with ¹²³I, IPT is an interesting imaging agent for use in in vivo exploration of DA uptake sites. However, it has good in vitro affinity for the 5-HT transporter, resulting in discrepancies between in vivo and in vitro selectivities (Kung et al., 1995). We therefore hypothesized that a new substituent on the 3-phenyl ring could improve in vitro specificity for the DAT, and we synthesized a new analog of cocaine, -CDIT. We previously demonstrated that this compound has high specific in vivo binding on the DAT in the rat (Emond et al., 1997). Moreover, preliminary SPECT exploration in monkeys showed fast uptake and excellent images in the striatum (Emond et al., 1996). It was therefore necessary to further characterize -CDIT to quantify its affinity in relation to the three monoamine transporters and visualize its cerebral biodistribution in the rat. The present study describes the in vitro binding of [¹²⁵I]-CDIT in rat striatal membranes and its pharmacological characterization. The specificity of -CDIT was studied in vitro in rat cerebral preparations and compared with other DA uptake inhibitors, and the biodistribution of this compound was visualized in vivo in rat brains using autoradiographic studies. Moreover, the effects of this new iodinated ligand on locomotor activity were evaluated in mice and compared with those of -CIT. The results suggest that -CDIT will be a valuable ligand for in vitro characterization of DA transporter sites and for the exploration of these sites in vivo by SPECT.

	Methods

Top Abstract Introduction Methods Results Discussion References

Chemicals and Radiolabeling

The stannyl precursor for preparation of -CDIT was synthesized from ()-cocaine by hydrostannylation of the N-propargyl derivative (Goodman et al., 1994; Jung and Light, 1982). Unlabeled -CDIT was achieved with 70% yield by metal halogen exchange of the stannyl precursor using iodine in chloroform. Stable compounds were characterized by mass and NMR spectrometry (fig. 1).

View larger version (K): [in this window] [in a new window]   Fig. 1.   Preparation of stable -CDIT and [125I]-CDIT.

No-carrier-added radioiodinated -CDIT was prepared by iododestannylation of the stannyl derivative using radiolabeled sodium iodine and H₂O₂ as oxidizing agent. The specific activity was 2000 Ci/mmol, and radiochemical purity was >95%.

In Vitro Binding Studies

Tissue preparation. Male Wistar rats weighing 200 to 250 g (Centre d'Elevage Dépré, Saint-Doulchard, France) were rapidly decapitated, and striata were dissected on ice. Tissue was prepared according to Bonnet et al. (1986). The tissue was homogenized in 10 vol of 0.32 M sucrose using an Ultraturrax T25. After 1000 × g centrifugation for 10 min at 2°C, the supernatant and pellet were collected separately. The pellet was homogenized, washed and centrifuged as described above. Supernatants were pooled and centrifuged at 17,500 × g for 30 min at 2°C. Pellets were homogenized in 20 vol of the assay buffer and then centrifuged at 50,000 × g for 10 min at 2°C. The pellets were suspended in a minimum volume of assay buffer, and the protein concentration was determined according to Bradford (1976) using bovine serum albumin as standard.

[¹²⁵I]-CDIT binding assays. The binding studies of [¹²⁵I]-CDIT were conducted on rat striatal membranes by saturation assays, and its pharmacological characterization was determined by competition with drugs known to bind to the DA, 5-HT and NE transporters.

K_i Determinations

Transporter binding affinity of -CDIT was evaluated by competitive in vitro radioaffinity assays for DA, 5-HT and NE transporter sites in rat brain tissue using [³H]GBR 12935, [³H]paroxetine and [³H]nisoxetine, respectively.

[³H]GBR 12935 assay. Each sample contained 2.4 ml of hydrogencarbonate buffer (9 mM NaHCO₃, 5 mM NaH₂PO₄, 5 mM EDTA, 120 mM NaCl and 0.01% bovine serum albumin, pH 7.5), 0.4 ml of [³H]GBR 12935 (specific activity, 45.7 Ci/mmol; New England Nuclear Research Prodcuts, Boston, MA) at a constant concentration of 1 nM, 0.2 ml of the unlabeled -CDIT or other DA uptake inhibitors at various concentrations and 1 ml of 100 µg of membrane protein preparation. Nonspecific binding was determined with 10⁶ M mazindol (a gift from Sandoz, Rueil-Malmaism, France). Samples were incubated at 37°C for 1 hr and rapidly filtered through Whatman GF/B filters. The filters were washed twice with 3 ml of ice-cold buffer, and the residual radioactivity was measured with a -counter (LKB Rack Beta 1215) after the addition of 5 ml of scintillator (LKB Optiphase Highsafe II). The binding assays were run in duplicate.

[³H]Paroxetine assay. Each sample contained 1.2 ml of incubation buffer (50 mM Tris·HCl, 120 mM NaCl, 5 mM KCl, pH 7.4), 0.2 ml of [³H]paroxetine (specific activity, 18.1 Ci/mmol; New England Nuclear) at a constant concentration of 0.5 nM, 0.2 ml of unlabeled -CDIT or other 5-HT uptake competitors at various concentrations and 0.5 ml of 140 µg of protein preparation. Nonspecific binding was determined with 10⁶ M fluvoxamine (a gift from Duphar). Samples were run in duplicate and incubated for 90 min at 22°C. Samples were then treated as previously described.

[³H]Nisoxetine assay. Each sample contained 0.2 ml of incubation buffer (50 mM Tris·HCl, 300 mM NaCl, 5 mM KCl, pH 7.4), 0.1 ml [³H]nisoxetine (specific activity, 80 Ci/mmol, New England Nuclear) at a constant concentration of 0.5 nM, 0.1 ml of unlabeled -CDIT or other NE uptake competitors at various concentrations and 0.2 ml of 125 µg of protein preparation. Nonspecific binding was determined with 10⁶ M desipramine (RBI Bioblock, Illkirch, France). Samples were run in duplicate and incubated for 5 hr at 4°C. Samples were then treated as previously described.

Locomotor Activity

Nine-week-old male Swiss mice (Centre d'Elevage Janvier, France) were used. Before the experimental testing, they were housed five to a standard cage that contained a constant supply of food pellets and water. All animals were kept on a 12-hr reversed light/dark cycle from 8:00 p.m. to observe animals in their high activity period (dark period). Each mouse was tested only once.

Open-Field Test

The test apparatus was a gray polyvinyl chloride circular open field that was 40 cm in diameter and 30 cm high. The floor was divided into six peripheral and one circular central sectors, all of the same area and covered by a white sheet of paper that was changed after each mouse. The device was lit by a 100-W bulb placed 80 cm above the floor of the open field and provided the only room illumination. Each mouse was introduced in the center of the open field. The number of sector crossings (locomotion) was recorded during 5-min sessions. -CIT/D-()-tartrate and -CDIT/D-()-tartrate were dissolved in saline and administered 30 min before testing. Treatments were administered intraperitoneally in concentrations with an injection volume of 10 ml/kg b.wt. Mice (n = 9-16 animals/group) were randomly allocated to the following groups: vehicle control (saline), -CIT/D-()-tartrate (0, 0.078, 0.156, 0.312, 0.625, 1.25, 2.5, 5 and 10 mg/kg, respectively) and -CDIT/D-()-tartrate (0, 0.078, 0.156, 0.312, 0.625, 1.25, 2.5, 5, 10, 15 and 20 mg/kg, respectively).

Actograph Test

The actographs were eight toggle-floor boxes, each of which was divided into two 20 × 10-cm sections connected by a 3 × 3-cm opening. For each mouse, the number of transitions between the two parts was automatically recorded with a microswitch that was connected to the floor of the box. Mice were individually placed into the apparatus for 240 min, and the number of transitions was recorded experimenter every 5 min. In each experimental session, all doses of the drug administered were tested in an equal number of animals. Because it was not practically possible to run all animals in a given experiment simultaneously, a small number of mice received each treatment. This was repeated until a total of 8 to 16 animals had received all treatments.

Statistical Analysis

In the open-field and actograph experiments, data underwent analysis of variance. A posteriori comparisons were made with Tukey's HSD test.

Ex Vivo Autoradiograhic Studies

Ex vivo autoradiographic studies were performed in male Wistar rats (200-250 g). Rats were injected intravenously with 0.3 ml of 120 to 130 µCi of [¹²⁵I]-CDIT. Three groups of rats received GBR 12909 (5 mg/kg), paroxetine (5 mg/kg) or saline (300 µl), respectively, 30 min before the injection of [¹²⁵I]-CDIT. Rats were killed at 2 hr after the injection by rapid decapitation. The brains were removed, rapidly frozen on dry ice and maintained at 70°C until use. Sections 20 µm thick were cut at 18°C and thaw-mounted on glass slides. Sections were placed in X-ray cassettes and exposed to max Hyperfilms (Amersham, Buckinghamshire, UK) for 4 weeks at room temperature. Autoradiograms were developed (Kodak L X24), fixed (Kodak AL4) and quantified using a computer imaging system (Biocom).

	Results

Top Abstract Introduction Methods Results Discussion References

Scatchard analysis of specific [¹²⁵I]-CDIT binding and pharmacological characterization. The affinity and density of specific [¹²⁵I]-CDIT binding sites were determined using increasing concentrations of [¹²⁵I]-CDIT (0.02-2 nM). Binding was saturable and had high affinity. Scatchard transformation of the data resulted in a linear curve suggesting a one-site model (n_H = 0.96) with a K_d value of 0.18 ± 0.07 nM (mean ± S.E.) and a B_max value of 500 fmol/mg of protein (fig. 2). The specific binding of [¹²⁵ I]-CDIT was studied in rat striatal membranes in competition with other ligands of DA, 5-HT and NE uptake sites (table 1). In rat striatal membranes, inhibitors of DA uptake such as GBR 12909, mazindol and cocaine are competitors for [¹²⁵I]-CDIT binding, with K_i values of 5 ± 2, 50 ± 15 600 ± 30 nM, respectively. Paroxetine, a specific 5-HT uptake blocker, does not inhibit binding of [¹²⁵I]-CDIT in rat striatal membranes (K_i > 1000 nM). Similarly, nisoxetine, a selective norepinephrine uptake blocker with high affinity for this site, does not appear to inhibit binding at all (K_i > 1000 nM).

View larger version (K): [in this window] [in a new window]   Fig. 2.   Saturation of specific [125I]-CDIT binding in the rat striatum. Striatal membranes were incubated with increasing concentrations of [125I]-CDIT (0.02-2 nM). Nonspecific binding was determined by the addition of 30 µM cocaine. Scatchard transformation of the resulting data was determined using EBDA program (Kd = 0.18 ± 0.07 nM, Bmax = 500 fmol/mg of protein).

View this table: [in this window] [in a new window]   TABLE 1 Relative potencies of competitors for [125I]-CDIT binding in rat striatum Experiments were performed at 0.2 nM concentration in homogenized rat striatum samples. Results are mean ± S.D. of three independent experiments.

K_i determinations. The specificity of -CDIT binding was also determined in relation to the three monoamine transporters by competition with [³H]GBR 12935 in rat striatal membranes or with [³H]nisoxetine or [³H]paroxetine in rat cortical membranes.

Locomotor activity. Figure 3 shows the dose-response function for the increase in locomotor activity induced by -CIT and -CDIT in mice placed in an open field for 5 min. Both -CIT and -CDIT produced significant hyperactivity [F(100,8) = 4.42, P < .0001 and F(110,10) = 32.30, P < .0001, respectively]. -CIT increased locomotion at doses ranging from 0.312 to 10 mg/kg. -CDIT was less potent insofar as it increased locomotion at doses ranging from 5 to 20 mg/kg.

View larger version (K): [in this window] [in a new window]   Fig. 3.   Open field test, in which locomotor activity as a function of the dose was determined 30 min after intraperitoneal injection of -CIT or -CDIT. Values are presented as the mean number of sector crossings recorded during 5-min sessions ± S.E.M. *, P < .0001 compared with control value (intraperitoneal injection of saline).

View larger version (K): [in this window] [in a new window]   Fig. 4.   Time course of the locomotor stimulation produced by the lowest effective doses of -CIT and -CDIT in the actographs. Values are presented as the mean number of transitions per 5 min during the 4-hr test session ± S.E.M. *, P < .05, **, P < .01, ***, P < .001: differences between controls and -CIT-treated mice, , P < .05, , P < .01, , P < .001: differences between controls and -CDIT-treated mice.

Ex Vivo Autoradiographic Studies. Absorbance values were evaluated after establishment of regions of interest, using a computer imaging system (Biocom). Autoradiograms (fig. 5) showed a high uptake of [¹²⁵I]-CDIT in areas rich in DA transporter: striatum (Fig. 5b), accumbens nucleus (Fig. 5c) and olfactory tubercles (Fig. 5d). In contrast, a low level was found in the cortical area (Fig. 5a), providing a striatum to frontal cortex ratio of 6 in control rats. This ratio was reduced to 1 in rats pretreated with a saturating dose (5 mg/kg) of GBR 12909 but remained unchanged in rats pretreated with the same dose of paroxetine.

View larger version (K): [in this window] [in a new window]   Fig. 5.   Ex vivo autoradiographic studies in rats. Autoradiograms of coronal brain sections were obtained at 2 hr after injection of [125I]-CDIT. Animals received an intravenous injection of saline (1), GBR 12909 at the dose of 5 mg/kg (2) or paroxetine at the dose of 5 mg/kg (3) 30 min before [125I]-CDIT. In 1 and 3, high levels of binding were observed in the striatum (b), accumbens nucleus (c) and olfactory tubercle (d), whereas a low level was observed in the frontal cortex (a). In 2, low levels were observed in all regions.

	Discussion

Top Abstract Introduction Methods Results Discussion References

Exploration of the human DA transporter by scintigraphy is of great value, in particular in neurodegenerative diseases such as Parkinson's or Alzheimer's disease. Many radiotracers that are inhibitors of DA uptake have been developed for in vivo exploration of the DAT by SPECT or PET, but most of them do not possess the necessary properties to be used as specific radioligands in vivo. Indeed, -CIT appears to be a promising agent for the exploration of DA uptake sites, but it has two major disadvantages: it binds to the 5-HT transporter as well as to the DAT, and it has slow kinetics in vivo. Imaging can thus be performed only ~20 hr after injection (Brücke et al., 1993; Laruelle et al., 1993). To obtain a new ligand with better in vitro and in vivo specificity for the DAT and faster kinetics in vivo, we synthesized a new cocaine analog with aromatic and N-methyl substitutions: -CDIT (Emond et al., 1997). In the present study, we characterized -CDIT in vitro by saturation and competition studies and ex vivo in rats by autoradiographic studies. In addition, its pharmacological effects on locomotor activity in mice were compared with those of -CIT.

[¹²⁵I]-CDIT in vitro binding was evaluated in the striatum, a region of high DAT abundance. These experiments demonstrated that [¹²⁵I]-CDIT had high affinity for this transporter, with a K_d value obtained in Tris·NaCl buffer of 0.18 ± 0.07 nM and a B_max value of 500 ± 80 fmol/mg of protein. The affinity is comparable to the affinities of -CIT (K_d = 0.12 nM) (Carroll et al., 1992), IPT (K_d = 0.25 nM) (Kung et al., 1995) or RTI-121 (K_d = 0.15 nM) (Carroll et al., 1995). Under the experimental conditions reported in the present study, one-site binding of [¹²⁵I]-CDIT was observed to the DAT in the striatum. The Hill coefficient was close to unity, which is consistent with a one-site binding model. In similar binding conditions, Goodman et al. (1994) also demonstrated a one-site binding model for two N-substituted cocaine analogs. Similarly, Kung et al. (1995) reported one-site binding for IPT.

The specific binding of [¹²⁵I]-CDIT was studied in rat homogenates using a variety of drugs known to bind to DA, NE and 5-HT transporters. The results showed that GBR 12909, mazindol and cocaine are high competitors for [¹²⁵I]-CDIT binding. In contrast, paroxetine and nisoxetine, inhibitors of 5-HT and NE transporters, do not compete at all with [¹²⁵I]-CDIT (K_i = 1000 nM for paroxetine and 1000 nM for nisoxetine). The rank order in competing [¹²⁵I]-CDIT binding (GBR 12909 > mazindol > cocaine > paroxetine and nisoxetine) was similar to that found in the rat striatum with [¹²⁵I]IPT (Kung et al., 1995) and for [¹²⁵I]RTI-55 (Boja et al., 1992). These results are in agreement with specific binding to DA uptake sites.

The specificity of -CDIT was also evaluated on rat striatal membranes using [³H]GBR 12935 and on rat cortex membranes using [³H]paroxetine and [³H]nisoxetine as monoamine transporter ligands in comparison with -CIT. The results showed that -CDIT inhibits specific binding of [³H]GBR 12935 with a K_i value of 29 nM. This is similar to the value obtained for -CIT under the same conditions (27.5 nM). In contrast, -CDIT has a lower affinity for the 5-HT transporter (K_i = 50 nM) than -CIT (K_i = 3.1 nM). Similarly, the affinity of -CDIT for the NE transporter (K_i = 500 nM) is 6 times lower than the affinity of -CIT (K_i = 80 nM) measured under the same conditions.

We showed in behavioral experiments that both -CDIT and -CIT induced stimulation of the locomotor activity. This type of response is known to be mediated through dopaminergic systems and is consistent with blocking of DA uptake (Ross, 1979). The maximal effect obtained with -CIT occurred with a dose of ~1 mg/kg, which is in agreement with previously reported data (Cline et al., 1992). We observed that the maximal effect was obtained for -CDIT with a 10-fold higher dose. However, we showed that in vitro affinity for the DAT was identical for -CIT and -CDIT. Relationships between the level of occupancy of the striatal DAT by an inhibitory compound and the intensity of its stimulant locomotor effect have not been well established (Cline et al., 1992; Vaugeois et al., 1993). The differences observed in vivo on locomotor activity could reflect differences in bioavailability of either compound. Indeed, we previously observed that passage through the blood-brain barrier was lower for -CDIT than for -CIT (Emond et al., 1997). This was probably related to higher lipophilicity for -CDIT than for -CIT, leading to strong binding to plasma proteins and cell membranes for the former. It is unlikely that the lower activity of -CDIT compared with -CIT results from instability or fast metabolism, because we have previously shown that -CDIT is stable in the blood and striatum (Emond et al., 1997). In addition, the time course of the locomotor activity induced by the lowest active dose of either compound showed that the stimulatory effect of -CDIT appeared to be faster than that of -CIT. This observation is consistent with faster in vivo kinetics for -CDIT than for -CIT.

Ex vivo autoradiographic studies showed a high uptake of [¹²⁵I]-CDIT in the striatum, a brain region known to have a high density of DA nerve terminals. Other brain regions with low DAT site density revealed very low binding of [¹²⁵I]-CDIT. No significant binding of [¹²⁵I]-CDIT was revealed in the cortex, an area known to contain a high level of 5-HT transporter sites. In rats pretreated with GBR 12909, a potent DAT blocker, accumulation of [¹²⁵I]-CDIT in the striatum was reduced to the low level found in the cerebral cortex. In contrast, it remained unchanged in rats pretreated with paroxetine. The results of this study clearly demonstrate the selective in vivo binding of [¹²⁵I]-CDIT to DA transporter sites. Moreover, we have previously shown with ex vivo saturation experiments that a preinjection of GBR 12909 prevented 80% striatal fixation of [¹²⁵I]-CDIT, whereas under the same experimental conditions, striatal fixation of [¹²⁵I]-CIT was prevented by only 30% (Emond et al., 1997). These experiments also revealed that no fixation of -CDIT was observed in the frontal cortex, with a frontal cortex/cerebellum ratio of 1 at 2 hr after injection. By contrast, a ratio of 5 was obtained under the same conditions for [¹²⁵I]-CIT. Preinjection of paroxetine prevented 55% cortical fixation. All these findings argue in favor of specific in vivo binding of -CDIT to the DAT and are consistent with in vitro experiments.

From all these results, it appears that disubstitution on the phenyl ring conserved a high affinity for the DAT. Similar results have been reported for analogs of -CIT (Carroll et al., 1992). Analog results have also been previously described for other N-substituted derivatives such as IPT (Goodman et al., 1994; Kung et al., 1995) and N-fluoroalkyl analogs of -CIT (Neumeyer et al., 1994). In addition, we found in our experiments that [¹²⁵I]-CDIT has an affinity 16 times lower than that of -CIT for the 5-HT transporter and 6 times lower for the NE transporter. It therefore appears that the combination of substitution on nitrogen and the 3-phenyl ring improved in vitro selectivity for the DAT. Moreover, disubstitution leads to better selectivity for -CDIT compared with the monosubstituted compound IPT, which also has in vitro affinity for the 5-HT transporter (Kung et al., 1995). In addition, ex vivo autoradiographic studies clearly demonstrated specific binding to DAT in the rat striatum. Therefore, the combination of phenyl ring disubstitution and N-substitution increases the specificity of -CDIT binding without modification of the affinity to the DA uptake site.

In conclusion, this new iodinated ligand, -CDIT, displays high affinity for DA transporter sites in the rat brain and better selectivity than -CIT. In vitro and in vivo characterization of this compound confirms that it is a selective DA uptake site ligand that could be used as a SPECT imaging agent.