Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The peripheral benzodiazepine receptor (PBR) is localized in the outer mitochondrial membrane (Anholt et al., 1986). It is composed of a 18-kDa protein, which contains isoquinoline and carboxamide binding sites, associated to a voltage-dependent anion channel and an adenine nucleotide carrier (McEnery et al., 1992). The PBR has a widespread distribution in the nervous system and peripheral tissues, but the highest levels are found in steroid-producing tissues. In the central and peripheral nervous systems, PBR localization was mainly described on glial cells (Krueger and Papadopoulos, 1990; Lacor et al., 1996), but a neuronal localization has also been reported (Anholt et al., 1984). An increase in PBR density has been observed following experimental injuries of the nervous system and in other pathological situations, including sciatic nerve lesion, cerebral infarction, multiple sclerosis, Alzheimer's disease, and astrocytomas (Benavides et al., 1988; Vowinckel et al., 1997; Lacor et al., 1999). In most lesions, the increase of PBR density was associated with the inflammatory reaction of the nervous tissue (Benavides et al., 1988).

Although the exact function of the PBR is not yet fully established, this receptor has been involved in the control of several mitochondrial functions including the respiratory chain and ion channel activities (Szabo et al., 1993), as well as in the modulation of immune functions (Carayon et al., 1996). Activation of steroid biosynthesis is a well characterized activity for the PBR in endocrine and nervous tissues, both in vitro and in vivo (Papadopoulos et al., 1992; Lacor et al., 1999). The PBR is required for the transport of cholesterol from the outer to the inner mitochondrial membrane and its availability for the cytochrome P450_scc, which catalyzes the synthesis of pregnenolone, a precursor of all steroids (Krueger and Papadopoulos, 1990). The newly formed pregnenolone then leaves the mitochondria for the endoplasmic reticulum, where it's converted to progesterone by the 3-hydroxysteroid dehydrogenase. The presence and activity of this enzyme has been described in both glial cells and neurons. By measuring the cerebral accumulation of pregnenolone after injection of the 3-hydroxysteroid dehydrogenase inhibitor trilostane, it has been demonstrated that the conversion of pregnenolone is a very active metabolic pathway in the brain (Schumacher et al., 2000). Neurosteroids are synthesized in the nervous system where they can influence central nervous system functions by modulating the activity of different ionotropic neurotransmitter receptors, such as -aminobutyric acid A, N-methyl-D-aspartate, or -receptors (Baulieu et al., 1999). Some neurosteroids have also been shown to have neurotrophic and neuroprotective activities (Koenig et al., 1995; Melcangi et al., 1999; Schumacher et al., 2000).

Recently, the involvement of the PBR in the control of apoptotic processes has also been suggested in light of data showing that PBR ligands protect Jurkat cells, transfected with human PBR cDNA, against apoptosis induced by H₂O₂ (Carayon et al., 1996). Furthermore, Bono et al. (1999) described the protective effect of the PBR ligand Ro5-4864 against apoptosis induced by TNF in U937 lymphoblastoid cells, suggesting that pharmacological modulation of the PBR may influence cell survival. These data, together with the steroidogenic potential of PBR ligands, led us to hypothesize that compounds that modulate the PBR function could play neurotrophic and neuroprotective roles during neuronal damage. To investigate this possibility, we have evaluated the potential neuroprotective effects of SSR180575 (Fig. 1), a pyridazinoindole derivative that possesses a high affinity for the PBR, in animal models of progressive degeneration of the central and peripheral nervous systems. SSR180575 was evaluated in an experimental model of motoneuron degeneration induced by axotomy of facial motoneurons in immature rats and in two models of peripheral neuropathy, acrylamide-induced axonopathy and traumatic nerve injury in the young rat. We have also measured the effect of the compound on steroid concentrations in plasma and nervous tissue in the rat.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   The chemical structure of SSR180575.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Adult male Sprague-Dawley rats (Iffa-Credo, L'Arbresle, France) weighing 100 to 300 g were used for binding studies, acrylamide-induced axonopathy, or in in vivo steroidogenesis studies. Young rats with a body weight of 45 to 50 g or rat pups of 4 days of age were used for facial nerve regeneration or motoneuron survival experiments, respectively. Animals were housed at 22°C under a 12-h light/dark cycle in animal care facilities and were fed rat chow and water ad libitum. Experimental protocols have been approved by the Animal Care and Use Committee of Sanofi-Synthélabo Recherche.

Drugs

SSR180575 (Fig. 1), Ro5-4864, and trilostane were synthesized by Sanofi-Synthélabo Recherche (Bagneux, France). [³H]PK11195, [³H]Ro5-4864, and [³H]flumazenil were purchased from PerkinElmer Life Sciences (Courtaboeuf, France) and [³H]alpidem from Amersham Bioscience (Orsay, France). For in vitro studies, drugs were dissolved in dimethyl sulfoxide (10³ M) and diluted in the appropriate buffer. For in vivo studies, SSR180575, Ro5-4864, or trilostane was prepared as a suspension in Tween-80 (0.1%). For all studies, doses were chosen by their ability to inhibit 50 to 70% of [³H]alpidem binding in vivo. Doses were expressed as free doses.

In Vitro Binding Studies

The affinity of SSR180575 for the rat and human PBR was evaluated by measuring its ability to compete with [³H]Ro5-4864 or [³H]PK11195 binding to membrane preparations from rat kidney (Schoemaker et al., 1983) and human keratinocytes (Canat et al., 1993), respectively. Selectivity versus the central benzodiazepine receptor was determined by measuring the capacity of SSR180575 to displace [³H]flumazenil binding to membrane preparations from the rat cerebellum. Its potency was compared with that of Ro5-4864, a selective PBR ligand.

Assay of [³H]Ro5-4864 Binding to Rat Kidney

Animals were killed by decapitation; the kidneys were rapidly removed, weighed, and homogenized in ice-cold 50 mM phosphate-buffered saline (PBS), pH 7.4, at 0-4°C. The kidney membrane preparation was incubated in PBS containing 0.5 nM [³H]Ro5-4864 (80 Ci/mmol) at 4°C for 3 h. The reaction was terminated by vacuum filtration through Whatman GF/B filters (Merck Eurolab, Fontenay sous bois, France). After washing, the filters were dried, and the radioactivity was determined by liquid scintillation spectrometry. Nonspecific binding was determined by the addition of 1 µM of unlabeled Ro5-4864. Specific binding was defined as the difference between the total and nonspecific binding. Data are presented as the compound concentration required to inhibit 50% of the specific radioligand binding. IC₅₀ values were determined by using the nonlinear regression analysis program ALLFIT (De Lean et al., 1978).

Assay of [³H]Flumazenil Binding

The binding of [³H]flumazenil to the central benzodiazepine receptors was measured in the rat cerebellum, as described by Schoemaker et al. (1997). Briefly, the cerebellum was homogenized in 50 mM Tris-HCl, pH 7.4, containing 120 mM NaCl and 5 mM KCl. After incubation with [³H]flumazenil (1 nM; specific activity, 70-87 Ci/mmol) for 45 min at 0-4°C; membranes were recovered by vacuum filtration using Whatman GF/B filters and washed, and filter-bound radioactivity was quantified by liquid scintillation spectrometry. Nonspecific binding was determined as binding in the presence of 1 µM diazepam. Data are presented as the percentage of inhibition of the specific radioligand binding at 1 µM.

[³H]PK11195 to Human Keratinocytes

The human keratinocyte cell line A431 was purchased from ECACC (European Collection of Cell cultures, Centre for Applied Microbiology & Research, Salisbury, Wiltshire, UK). Cells were routinely grown in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal calf serum (Invitrogen), 2 mM Glutamax (Invitrogen), 1 mM sodium pyruvate (Invitrogen), 100 µg/ml gentamicin (Invitrogen), and 1% tylocine (Tylosin) (Invitrogen).

PK11195 binding to human keratinocytes was performed according to Canat et al. (1993). Human keratinocytes (0.2 ml) were incubated for 30 min in the presence of [³H]PK11195 (specific radioactivity, 85.5 Ci/mmol) at 5 nM and SSR180575 at concentrations ranging from 10¹⁰ to 10⁵ M. Incubations were ended by filtration through GF/B Whatman glass fiber filters. Filters were then washed and dried, and the bound radioactivity was measured by liquid scintillation spectrometry. Nonspecific binding was defined by the addition of 1 µM of unlabeled PK11195. Values were expressed as specific binding obtained by the difference between the total and nonspecific binding. Data are presented as the compound concentration required to inhibit 50% of the specific radioligand binding.

In Vivo Binding Studies

The ability of SSR180575 to interact with the rat PBR in vivo was evaluated by measuring its ability to displace the in vivo binding of [³H]alpidem, a selective PBR ligand (Arbilla et al., 1993) in the rat brain and spleen. The kinetic of the displacement by SSR180575 of in vivo [³H]alpidem binding in rat brain and spleen was also studied to determine the duration of PBR occupancy.

[³H]Alpidem (specific activity, 49 Ci/mmol) was injected i.v. (0.2 µCi/0.2 ml of a 0.9% NaCl solution) into rats via the tail vein. SSR180575 or vehicle was administered i.p. or p.o. 30 min or 1 h, respectively, before [³H]alpidem injection. Animals were sacrificed 90 min later by decapitation, and their brain and spleen were rapidly removed. Tissue samples were weighed and dissolved in 500 µl of Solvable (Packard Instrument Co., Meriden, CT) before adding 8 ml of scintillation cocktail for the determination of radioactivity by liquid scintillation spectrometry. Data were expressed as the dose required to inhibit 50% of [³H]alpidem binding (ID₅₀), determined using Fig. P curve-fitting software (Windows; Microsoft, Redmond, WA). Time-related displacement of in vivo [³H]alpidem binding in rat brain and spleen by SSR180575 at 1 mg/kg p.o. was studied using the same method at 0.25, 1, 3, 6, 15, and 24 h after SSR180575 administration.

Motoneuron Survival after Axotomy of Facial Nerve in the Immature Rat

Surgery. Rat pups 4 days old were anesthetized using pentobarbitone (3 mg/kg i.p.). The right facial nerve was transected distally to its exit from the stylomastoid foramen. Care was taken to remove a 2- to 3-mm segment of the nerve to prevent reinnervation of the distal nerve stump.

Drug Administration. In a first experiment, SSR180575 was given orally, twice a day at 3, 6, or 10 mg/kg with a 6-h interval between the two administrations. The first treatment was administered 5 min before nerve transection. In the second experiment, SSR180575 or Ro-54864 was given intraperitoneally at 3 and 5 mg/kg twice a day, respectively. In this case, the first injection was given 10 min after axotomy. In these two experiments, animals were killed 8 days after injury for the determination of neuronal survival.

Histology. Brainstems were removed and frozen at 40°C in isopentane. Serial 10-µm sections were cut on a cryostat (CM 3000; Leica, France) throughout the entire length of the facial nucleus, and every ninth section was stained with Nissl substance or peripherin. The number of motoneurons was counted serially at 90-µm intervals after excluding the slices corresponding to the facial nucleus extremities, using a Histo-2000 station (Biocom, France).

Immunohistochemistry. Frozen sections were fixed with 4% paraformaldehyde for 15 min and permeabilized with 4% Triton X-100 in 0.01 M PBS, rinsed, and then incubated with peripherin antibody (Chemicon, Euromedex-France) at 1:4000 dilution in PBS, pH 7.4, containing 1% bovine serum albumin for 30 min at 4°C using a Cadenza immunostainer (Shandon, UK). After washing, the sections were incubated with the biotin-conjugated secondary antibody (1:100) for 30 min and then with an avidin-biotin enzyme system (Vector, Biovalley, France) for an additional 30 min. The visualization of immunopositive cells was performed in a 0.05% solution of diaminobenzidine in 0.01% H₂O₂. After washing, the sections were counterstained with hematoxylin for 5 min and mounted in a DAKO fluorescent mounting medium (DAKO, Trappes, France).

Facial Nerve Regeneration in the Young Rat

Young male Sprague-Dawley (Harlan, Indianapolis, IN) rats (45-50 g) were anesthetized, and the right facial nerve was injured unilaterally by freezing (Koenig et al., 1995) at its exit from the stylomastoid foramen, distal to the posterior auricular branch. Freezing lesions induce nerve fibers destruction without interrupting nerve sheaths continuity, thus allowing optimal alignment of the proximal and distal nerve segments and favoring nerve regeneration compared with mechanical injuries (nerve crush or section).

Facial nerve regeneration was assessed by recovery of the blink reflex, measured by applying a stream of air on the eye. Neurological tests were performed using a score of 0 to 4 points for gradual signs of facial nerve regeneration: 0, complete paralysis (open eye); 1, blink reflex running the eyelid to less than halfway; 2, semi blink reflex; 3, blink reflex running the eyelid to more than halfway (without complete closure); 4, full blink reflex (identical to the contralateral eye). Functional scores were performed by comparing the lesioned ipsilateral side to the intact contralateral side. They were monitored twice daily from day 5 after nerve lesion until complete recovery.

To evaluate the effect of SSR180575, animals were administered intraperitoneally with the compound at 3 mg/kg or vehicle (n = 30), starting 30 min before the nerve lesion followed by a second injection, 6 h later, then twice a day for 11 days. Functional evaluation was performed 1 h before treatment. Statistical analysis was performed by two-way ANOVA followed by Dunnett's test for the time course of blink reflex recovery. One-way ANOVA and Dunnett's t-test were applied for the area under the curve (AUC).

Acrylamide-Induced Neuropathy in the Rat

Acrylamide induces a distal neuropathy characterized by alterations of slow anterograde and retrograde axonal transport in several classes of neurons, leading to a dying-back axonopathy with reproducible alterations of sensory and motor functions (Sickles et al., 1996). To test the effect of SSR180575 on motor deficits induced by acrylamide, male Sprague-Dawley rats (250-300g) were administered i.p. with a water solution of acrylamide (Sigma Chemical Co., St. Louis, MO) at 50 mg/kg three times a week for 3 weeks (a total dosage of 450 mg/kg b.wt.) (Fournier et al., 1993). SSR180575 was given orally at 2.5, 5, or 10 mg/kg/day. In the first experiment, SSR180575 or vehicle was administered 2 h after the first acrylamide injection and then once daily for 16 days. In the second experiment, SSR180575 or vehicle were first administered on day 13, after the beginning of the intoxication with acrylamide and then once a day for 4 days. Sensorimotor function was assessed by measuring the maximal angle at which the animals retained their position on an inclined plan twice a week from 1 to 13 days and then every day until the day 17. In this study, 10 animals/group were used. Data were expressed as the mean angle ± S.E.M. and were analyzed by one-way ANOVA. Subsequent comparisons between treatment groups and control were carried out using Dunnett's t test.

Measurement of Pregnenolone Concentration in Rat Plasma and Nervous System

To measure the formation and accumulation of newly synthesized pregnenolone in the brain and sciatic nerve after PBR ligand stimulation, rats were pretreated with trilostane, a competitive inhibitor of the 3--hydroxysteroid dehydrogenase, an enzyme that converts pregnenolone to progesterone. In this experiment, three groups of rats (200-220 g; n = 10/group) were used: vehicle, trilostane, and trilostane followed by SSR180575. Trilostane was injected i.p. at the dose of 10 mg/kg 15 min before i.p. administration of SSR180575 (3 mg/kg).

Fifteen minutes after SSR180575 treatment, animals were killed by decapitation. Blood samples were collected in heparinized tubes and centrifuged at 2000g at 4°C. Brains and sciatic nerves were rapidly removed, frozen, and immediately stored at 30°C until the extraction of steroids.

Pregnenolone concentrations were measured using high-performance liquid chromatography (HPLC) followed by gas chromatography-mass spectrometry, as described by Liere et al. (2000). Briefly, steroids were extracted from rat plasma (200 µl), brain, or sciatic nerve (50 mg) in 10 volumes of methanol overnight at room temperature. After sonication and centrifugation, the supernatant was collected, and 1 ng of the internal standard 3,5 -tetrahydro-androstanedione was added for quantification. Free steroids were eluted from C₁₈ columns with a mixture of methanol/water (85:15, v/v) after a preliminary washing with methanol/water (40:60, v/v). After filtration on a 0.45-µm Millipore polytetrafluoroethylene membrane (Millipore Corporation, Bedford, MA), a HPLC purification step was carried out.

The HPLC fraction containing pregnenolone was submitted to derivatization by adding 20 µl of heptafluorobutyric anhydride and 200 µl anhydrous acetone for 30 min at 20°C. After complete evaporation of the reaction mixture under a stream of nitrogen, the residue was dissolved in 12 µl of hexane.

The derivatized steroids were analyzed by gas chromatography-mass spectrometry using an Automass mass spectrometer (model 150; Thermo Finnigan, San Jose, CA) coupled to a GC 8000 Top gas chromatograph with an AS 800 autosampler (Carlo Erba, Italy). The gas chromatograph was equipped with a BPX5 capillary column from SGE (Courtaboeuf, France). Helium was used as the carrier gas, with a constant flow rate of 0.9 ml/min. Two microliters of the derivatized sample was injected in the injector in the splitless-mode during 1 min. Electron impact occurred in the ionization chamber with a 70-eV ionization energy. Detection was done in single ion-monitoring mode. The ionic species m/z 283 Da and 298 Da for pregnenolone heptafluorobutyrate and m/z 486 Da for 3,5-tetrahydro-androstanedione-heptafluorobutyrate were analyzed. Pregnenolone was expressed as nanograms per gram of tissue for brain and sciatic nerve and in nanograms per milliliter of plasma (± S.E.M.). For each tissue, the concentration of pregnenolone was analyzed by one-way ANOVA followed by Newman-Keuls' test for multiple comparisons.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Affinity of SSR180575 for PBR in Vitro and in Vivo. As shown in Table 1, SSR180575 and Ro5-4864 displaced the binding of [³H]Ro5-4864 in the rat kidney (with an IC₅₀ in the nanomolar range), SSR180575 being 4 times more potent than Ro5-4864. In contrast, SSR180575 and Ro5-4864 did not show any noticeable affinity for the central type benzodiazepine receptors labeled by [³H]flumazenil (IC₅₀, >1 µM; Table 1). Likewise, SSR180575 lacks activity at 1 µM in a large variety of binding assays for receptors, ion channels, and enzymes (data not shown). SSR180575 also recognizes human PBR with high affinity since it displaces [³H]PK11195 from human keratinocytes with an IC₅₀ of 3.5 nM.

Inhibition of in Vivo Binding of [³H]alpidem to PBR in the Rat. After intraperitoneal administration, SSR180575 inhibited the binding of [³H]alpidem, a PBR ligand (Arbilla et al., 1993) to rat brain and spleen in a dose-related manner, with an ID₅₀ of 0.1 and 0.3 mg/kg, respectively (Fig. 2A). When given orally, SSR180575, also potently inhibited [³H]alpidem binding in rat brain or spleen with almost similar ID₅₀ values (Fig. 2B). With both routes of administration, the ID₅₀ values of SSR180575 in the brain and spleen were comparable. The occupancy of brain and spleen PBR at different doses of SSR180575 was calculated from the inhibition curves of the in vivo [³H]alpidem binding and is listed in Table 2. A very substantial occupancy of PBR (51-88%) was observed in both structures following i.p. or p.o. administration of SSR180575 at doses that were used in subsequent studies. The time-related displacement by SSR180575 of in vivo [³H]alpidem binding in rat brain and spleen at 1 mg/kg p.o. showed that about 50% PBR occupancy was still observed in both organs at 12 h after injection, indicating a long duration of action of the compound (Fig. 3).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Inhibition by SSR180575 of in vivo binding of [3H]alpidem to PBR in the rat brain and spleen. Each value ± S.E.M. represents the mean of six animals. The drug was injected i.p. 30 min (A) or p.o. 1 h (B) before [3H]alpidem injection. Animals were sacrificed 90 min later by decapitation. ID50 values ± S.E.M. were determined using Fig. P curve-fitting software. ID50 values were calculated from the equation of the curve fitting: y = (100  a) · (1  x/(x + ID50)) + a, where a represents nonspecific binding.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Time-related displacement of in vivo [3H]alpidem binding by SSR180575 at 1 mg/kg p.o. in brain and spleen. Data are means determined using Fig. P curve-fitting software on plots from means obtained on six animals per group.

Motoneuron Survival in the Immature Rat. Facial nerve axotomy induced a degeneration of the motoneurons in the facial nucleus of 4-day-old rats at 8 days after injury. The cell bodies of the surviving motoneurons showed a shrunken aspect and an increased glial cell density around them (Fig. 4). The number of axotomized motoneurons was reduced by 50% on the ipsilateral side in control animals (Fig. 5, A and B).

View larger version (101K): [in this window] [in a new window]   Fig. 4.   Photomicrographs of Nissl-stained transverse sections of facial nucleus of 4-day-old rats at 8 days after axotomy and treatment with SSR180575 or vehicle. A and B, contralateral side from a representative control animal; many motoneurons (arrow) are found. C and D, ipsilateral injured side from a representative control animal; the number of motoneurons is reduced with an increase of perineuronal glial cells (asteric). E and F, ipsilateral facial nucleus from an SSR180575-treated (10 mg/kg p.o. b.i.d.) animal showing an increase in the number of surviving motoneurons. Scale bars are 100 µm.

View larger version (41K): [in this window] [in a new window]   Fig. 5.   Effect of SSR180575 on motoneuron survival 8 days after facial nerve axotomy in 4-day-old rats. Data are expressed as the number of motoneurons/section of facial nucleus (means ± S.E.M. from six animals per group) on the contralateral or ipsilateral side. A, oral administration of SSR180575 at 3, 6, and 10 mg/kg b.i.d. B, intraperitoneal treatment with SSR180575 (3 mg/kg b.i.d.) or Ro5-4864 (5 mg/kg b.i.d.) ***, p < 0.001 (Dunnett's test; treated versus ipsilateral control group).

Peripherin-Immunoreactivity in Axotomized Facial Motoneurons. Peripherin is a neuronal-specific intermediate filament protein whose expression is selectively up-regulated during axonal regeneration (Chadan et al., 1994). In normal facial nuclei from 4-day-old rat pups, the perikaryon of the motoneurons showed a very weak immunoreactivity for peripherin, whereas their axons were highly immunoreactive (data not shown). Axotomy of facial nerve induced peripherin immunoreactivity in motoneurons on the ipsilateral side (Fig. 6, C, D, I, and J), whereas a faint staining was seen on the contralateral side (Fig. 6, A, B, G, and H).

View larger version (137K): [in this window] [in a new window]   Fig. 6.   Photomicrographs of peripherin-stained transverse sections from facial nucleus of 4-day-old rats at 2 days (A-F) and 8 days (G-L) after axotomy and treatment with vehicle or SSR180575 (10 mg/kg p.o. b.i.d.). A, B, G, and H, contralateral side from a representative control rat; C, D, I, and J, ipsilateral injured side from a control animal showing motoneurons immunoreactive to peripherin (arrow); E, F, K, and L, ipsilateral injured side from a SSR180575-treated rat. Scale bars are 200 µm.

View larger version (20K): [in this window] [in a new window]   Fig. 7.   Effect of SSR180575 (10 mg/kg p.o. b.i.d.) on the number of peripherin-immunoreactive motoneurons of the facial nucleus at different times after axotomy in the immature rat. Data are expressed as the percentage of motoneurons immunoreactive to peripherin per Nissl-stained motoneurons in adjacent sections on the ipsilateral side. ***, p < 0.001; **, p < 0.01 (Dunnett's test; treated versus ipsilateral control).

Facial Nerve Regeneration in the Young Rat. Facial nerve freeze-injury in the rat resulted in a complete ipsilateral facial paralysis 24 h later, evidenced by a loss of blink reflex and vibrissae movements and abnormal vibrissae orientation.

View larger version (15K): [in this window] [in a new window]   Fig. 8.   Effect of SSR180575 (3 mg/kg i.p. b.i.d.) on functional recovery after facial nerve freeze-injury in the rat. A, changes in blink reflex score at days 1 to 11 following nerve injury in vehicle-treated (- - - -) or SSR180575-treated () rats. Results are the mean (± S.E.M.) scores obtained with 30 rats per group. **, p < 0.01; *, p < 0.05 compared with vehicle-treated rats (Dunnett's test). B, calculated AUC between day 5 and day 9.5 days. **, p < 0.05 (Dunnett's test).

Acrylamide-Induced Neuropathy in the Rat. In the vehicle-treated group, acrylamide-intoxicated animals showedwide gait and low stance in the hind limbs. These symptoms appeared during the second week after the initiation of acrylamide intoxication and progressed to a complete paralysis of the posterior limbs in the third week.

View larger version (40K): [in this window] [in a new window]   Fig. 9.   Effect of SSR180575 on acrylamide-induced neuropathy in the rat. Motor function was assessed on an inclined plan. All groups received acrylamide (50 mg/kg i.p.) three times a week. Animals were treated daily with SSR180575 (2.5 to 10 mg/kg p.o.) or vehicle for different periods. A, B, and C, treatment from day 1 to 17. D, E, and F, treatment from day 13 to 17. Results are the mean (± S.E.M.) angle values obtained with 10 rats per group. **, p < 0.01; *, p < 0.05 compared with vehicle-treated rats (Dunnett's t test).

Pregnenolone Concentration in Plasma and Nervous System of the Rat. The effects of SSR180575 on the accumulation of pregnenolone in brain, sciatic nerve, and plasma have been determined in animals pretreated with trilostane, a 3--hydroxysteroid dehydrogenase inhibitor that prevents the conversion of pregnenolone to progesterone.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The aim of the present study was to evaluate the neuroprotective potential of a PBR ligand in the rat. Our results clearly indicate that SSR180575, a potent and selective PBR ligand, promotes neuronal survival and repair in a number of models of nerve degeneration, increasing facial motoneuron survival following axotomy and enhancing functional recovery after local facial nerve injury and acrylamide-induced peripheral neuropathy.

SSR180575 is a novel PBR ligand that belongs to an original pyridazinoindole chemical series. This compound shows nanomolar affinity for the rat and human PBR and high specificity versus a large number of receptors including the central-type benzodiazepine receptor. SSR180575 potently inhibited the in vivo binding of [³H]alpidem to PBR in the rat brain and spleen after p.o. or i.p. administration. The ID₅₀ compound was similar after i.p. and p.o. administration in both structures, indicating a good bioavailability and brain penetration. In vivo binding studies also revealed a long duration of action (>12 h) for SSR180575. Thus, SSR180575 seems to be ideally suited to investigate the neuroprotective potential of PBR ligands.

The neuroprotective activities observed in the various experimental models with SSR180575 were obtained at doses that occupied about 50 to 70% of PBR, as determined by in vivo [³H]alpidem binding, indicating that these neuroprotective effects require only partial occupancy of PBRs.

The fact that SSR180575 displayed significant neuroprotective effects on either neuronal cell body or axons suggests that this compound may act via multiple mechanisms. In the immature rat, the transection of the facial nucleus causes retrograde degeneration of the affected motoneurons, which die essentially by apoptosis (Rossiter et al., 1996). It has been suggested that apoptosis occurs through opening of the permeability transition pore, outer mitochondrial membrane rupture, and cytochrome c release (Kroemer et al., 1998). In the cytosol, cytochrome c forms a complex with Apaf-1 and catalyzes the activation of caspase cascades, which finally can initiate cleavage of other cellular substrates that results in the morphological features of apoptosis (Kroemer et al., 1998). The voltage-dependent anion channel (VDAC), which is part of the PBR complex (McEnery et al., 1992), has also been proposed to be part of the permeability transition pore (Szabo and Zoratti, 1993). Thus, following transection of the facial nerve in the immature rat, the decrease in motoneuron death that we observed after treatment with the PBR ligands, SSR180575, or Ro5-4864 may occur through a PBR-mediated antiapoptotic effect. This assumption is supported by recent in vitro findings, showing that Ro5-4864 and SSR180575 can rescue neuroblastoma SH-SYSY and lymphoblastoid U937 cells from TNF-induced apoptosis (Bono et al., 1999; F. Bono, I. Lamarche, V. Prabbnnaud, G. Le Fur, and J. M. Herbert, manuscript in preparation). The view that PBR stimulation may protect cells from apoptotic death is also supported by data showing the ability of Jurkat cells, not expressing the PBR, to become resistant to apoptosis induced by H₂O₂ after transfection with human PBR cDNA (Carayon et al., 1996). Whether SSR180575 or Ro5-4864 may increase neuronal survival by blocking the apoptotic process directly at the neuronal level is presently unknown. In fact, autoradiographic studies in neonatal rats have shown that only low levels of [³H]Ro5-4864 are found in the brain, PBR expression being restricted to nerve terminals in the olfactory bulb (Anholt et al., 1984). These studies performed in normal animals do not preclude the possibility that PBR expression might increase in injured neurons, and such studies remain to be performed.

Alternatively, SSR180575 or Ro5-4864 might promote neuronal survival by acting via glial cells. The neuronal degeneration induced by transection of the facial nerve is accompanied by massive local microglial and astrocytic reaction (Graeber et al., 1998). As demonstrated by several authors, microglial cells and astrocytes express high levels of PBR in many types of brain injury (Stephenson et al., 1995). The increase in PBR density has been attributed to activated microglia invasion and astrocytic reaction. The neuroprotective effects of SSR180575 may thus be due to increased survival of glial cells and to an enhanced production by these cells of mediators, such as neurosteroids, cytokines, or other neurotrophic factors that support motoneuron survival (Raivich et al., 1996).

We have shown that SSR180575 is able to enhance pregnenolone concentration in the brain and sciatic nerve, suggesting that the neuroprotective effects of SSR180575 might also be mediated via PBR stimulation of neurosteroid synthesis. Indeed, stimulation of neurosteroid biosynthesis by PBR ligands has been documented in the central and peripheral nervous systems (Lacor et al., 1999).

The conversion of cholesterol to pregnenolone is a rate-limiting step in steroid synthesis, and its stimulation may lead to increased formation of steroids with neuroprotective or neurotrophic effects. However, the steroid metabolic pathways that are activated in response to PBR ligands remain to be established, as in the present study we measured the accumulation of pregnenolone after blocking its further conversion by trilostane. Pregnenolone itself has been proposed to play a role in neuronal plasticity (Schumacher et al., 2000). Other neurosteroids, in particular progesterone, can be synthesized by glial cells and have been shown to protect facial motoneurons from death in the adult rat after axotomy (Yu, 1989).

In our experiments, the protective effect induced by SSR180575 on motoneurons was associated with an increase in the number of peripherin-positive motoneurons that was already maximal 2 days after the lesion, reflecting an early activation of peripherin synthesis; this increase was maintained at the same level at 8 days after lesion. Peripherin is a neuronal type III intermediate filament protein in which expression is selectively up-regulated during neuronal regeneration (Chadan et al., 1994). Neurotrophic factors, such as nerve growth factor, or cytokines from the CNTF family, such as CNTF, LIF, or interleukin 6, have been shown to induce peripherin expression and to promote neurite outgrowth in several types of neurons (Djabali et al., 1993; Sterneck et al., 1996). Interestingly, the mouse peripherin gene was identified as a target gene for LIF (Lecompte, 1998). Furthermore, the ability of LIF and other neuropoïetic cytokines to influence the survival and differentiation of central and peripheral neuronal cells is well established (Dale et al., 1994, Shellard et al., 1996). However, cytokine production is also modulated by steroids (Brattsand and Linden, 1996), indicating a complex interplay between the PBR, steroids, and cytokines, leading to neuroprotective effects. Thus, PBR ligands, by interacting with their receptor on glial cells, may activate cholesterol transport and steroid biosynthesis, which in turn could induce neuroprotective effects by a direct action mediated by steroid receptors or by modulating the production of cytokines, such as those from the CNTF family (Fig. 10). Feedback regulations between cytokines and PBR may also be considered. Indeed, data from in vivo experiments have shown that cytokines, like interleukin-1 and TNF, induce an up-regulation of PBR sites in the rat brain (Bourdiol et al., 1991). Identification of these relationships may help to understand the mechanisms responsible for the neuroprotective effects of PBR ligands.

View larger version (30K): [in this window] [in a new window]   Fig. 10.   Potential mechanisms involved in the neuroprotective effects of PBR ligands. PBR ligands interact with their receptor on mitochondria and activate cholesterol transport and neurosteroid biosynthesis. Neurosteroids could induce neuroprotection by direct action on glial or neuronal cells or indirectly by modulating cytokine production in glial cells, such as cytokines from the CNTF family. An alternative mechanism may involve VDAC; PBR, by interacting with VDAC, may inhibit apoptosis in glial/neuronal cells.

In models of peripheral neuropathies, local facial nerve injury leads to distal nerve degeneration, which is followed by rapid spontaneous regeneration in the rat. In humans, the rate of regeneration is much slower than in the rat, and recovery is usually incomplete and dysfunctional (Fu and Gordon, 1997). SSR180575 induced a 20% increase in facial nerve regeneration, an effect comparable to that observed under the same experimental conditions with the neuroimmunophilin compound FK506 (B. Ferzaz, unpublished data).

This facilitation of nerve repair was confirmed in a more global type of peripheral nerve damage induced by acrylamide, which leads to a dying-back axonopathy characterized by alteration of axonal transport in several classes of neurons (Sickles et al., 1996). In this model, SSR180575 was able to completely antagonize the sensorimotor deficits following concomitant treatment with acrylamide and to markedly reduce the evolution of the sensorimotor deficits when administered therapeutically. Several data have shown that steroids, such as progesterone, promote sciatic nerve regeneration after local application and stimulate the activity of the myelin gene promoters P0 and PMP22 (Melcangi et al., 1999; Schumacher et al., 2000). In addition, the therapeutic potential of androgens and estradiol has also been reported in facial nerve regeneration after crush injury in the adult hamster (Tanzer and Jones, 1997). In peripheral nerves, increased production of neurosteroids at the level of Schwann cells may facilitate functional recovery after toxic or traumatic injury. Although the mechanism responsible for this nerve repair facilitation or neuroprotective effect is unknown, a relationship between SSR180575 neuroprotective activities and its steroidogenic potential may be suggested.

In conclusion, we have shown for the first time that a PBR ligand, SSR180575, promotes neuronal survival and regeneration in experimental models of axotomy and neuropathy. These effects may be mediated by a local synthesis of neurosteroids and/or cytokines, mainly from glial cells, which lead to an increased neuronal survival and an improvement of the repair phenomenon. Additionally, inhibition of apoptosis either at the neuronal and/or glial level could also play a significant role in the neuroprotective effects of SSR180575. These results suggest that PBR ligands like SSR180575 may have potential for the treatment of neurodegenerative disorders (e.g., traumatic, diabetic, or iatrogenic peripheral neuropathies and amyotrophic lateral sclerosis).