Reversible, short-lasting, and dose-dependent effect of (+)-fenfluramine on neocortical serotonergic axons

Plasma levels of parent compound and metabolites after doses of either d-fenfluramine or d-3,4-methylenedioxymethamphetamine (MDMA) that produce long-term serotonergic alterations

Plasma levels of parent compounds and metabolites were determined in adult rhesus monkeys after doses of either 5mg/kg d-fenfluramine (FEN) or 10mg/kg d-3, 4-methylenedioxymethamphetamine (MDMA) i.m. twice daily for four consecutive days. These treatment regimens have been previously shown to produce long-term serotonin (5-HT) depletions. Peak plasma levels of 2.0+/-0.4 microM FEN were reached within 40min after the first dose of FEN, and then declined rapidly, while peak plasma levels (0.4+/-0.1 microM) of the metabolite norfenfluramine (NFEN) were not reached until 6h after dosing. After the seventh (next to last) dose of FEN, peak plasma levels of FEN were 35% greater than after the first dose while peak NFEN-levels were 500% greater. The t(1/2) for FEN was 2.6+/-0.3h after the first dose and 3.2+/-0.2h after the seventh. The estimated t(1/2) for NFEN was more than 37.6+/-20.5h. Peak plasma levels of 9.5+/-2.5 microM MDMA were reached within 20min after the first dose of MDMA, and then declined rapidly, while peak plasma levels (0.9+/-0.2 microM) of the metabolite 3,4-methylenedioxyamphetamine (MDA) were not reached until 3-6h after dosing. After the seventh (next to last) dose of MDMA, peak plasma levels of MDMA were 30% greater than the first dose while peak MDA levels were elevated over 200%. The t(1/2) for MDMA was 2.8+/-0.4h after the first and 3.9+/-1.1h after the seventh dose. The estimated t(1/2) for MDA was about 8.3+/-1.0h. Variability in plasma levels of MDMA and MDA between subjects was much greater than that for FEN and NFEN. This variability in MDMA and MDA exposure levels may have lead to variability in the subsequent disruption of some behaviors seen in these same subjects. There were 80% reductions in the plasma membrane-associated 5-HT transporters 6 months after either the FEN or MDMA dosing regimen indicating that both treatments produced long-term serotonergic effects.

Neuroendocrine correlates of antisocial personality disorder in abstinent heroin-dependent subjects

The function of the central alpha-adrenergic, serotoninergic and dopaminergic systems was investigated in 30 heroin-dependent subjects, 6 - 8 weeks after detoxification and in 22 psychophysically healthy controls (group C). Twelve heroin-dependent subjects with antisocial personality disorder (ASPD) (group A), 18 heroin-dependent subjects without other Axis I and II pathologies (group B) were included among abstinent substance abusers. The norepinephrine (NE) function was evaluated by the GH responses to acute stimulation with clonidine (clon); the serotonin (5-HT) function by the PRL and cortisol (CORT) responses to acute stimulation with d-fenfluramine (d-fen) and the dopamine (DA) function was investigated by growth hormone (GH) and prolactin (PRL) responses to acute administration of bromocriptine (brom). Alpha-adrenergic sensitivity, as measured by the GH-clon test, was found significantly reduced in A subjects (ASPD), in comparison with B subjects and controls. PRL and CORT responses to d-fen were significantly blunted both in A and B subjects, in comparison with control subjects. DA receptors sensitivity seems to be reduced significantly in ASPD (A subjects); in contrast, heroin addicts without open psychiatric co-morbidity showed unimpaired responses to brom challenge; a significantly lower GH response to brom and a lack of PRL suppression in ASPD subjects could express D2 postsynaptic receptor hyposensitivity possibly related to DA gene variants associated to co-morbid disorder. In sum, the study of central monoamine function revealed an alteration of the 5-HT system in all detoxified heroin-dependent subjects. A significant reduction of alpha-adrenergic receptors sensitivity and the hyposensitivity of postsynaptic DA receptors in ASPD subjects suggest once again that specific biological correlates of psychiatric co-morbidity may characterize substance abusers subtypes.

Long-term impairment of anterograde axonal transport along fiber projections originating in the rostral raphe nuclei after treatment with fenfluramine or methylenedioxymethamphetamine

To further evaluate the serotonin (5-HT) neurotoxic potential of substituted amphetamines, we used tritiated proline to examine anterograde transport along ascending axonal projections originating in the rostral raphe nuclei of animals treated 3 weeks previously with (+/-)fenfluramine (FEN, 10 mg/kg, every 2 h x 4 injections; i.p.) or (+/-)3,4-methylenedioxymethamphetamine (MDMA, 20 mg/kg, twice daily for 4 days; s.c.). The documented 5-HT neurotoxin, 5,7-dihydroxytryptamine (5,7-DHT, 75 microg; ICV; 30 min after pretreatment with pargyline, 50 mg/kg; i.p., and desipramine 25 mg/kg; i.p.), served as a positive control. Along with anterograde axonal transport, we measured two 5-HT axonal markers, 5-HT and 5-hydroxyindoleacetic acid (5-HIAA). Prior treatment with FEN or MDMA led to marked reductions in anterograde transport of labeled material to various forebrain regions known to receive 5-HT innervation. These reductions were associated with lasting decrements in 5-HT axonal markers. In general, decreases in axonal transport were less pronounced than those in 5-HT and 5-HIAA. However, identical changes were observed after 5,7-DHT. These results further indicate that FEN and MDMA, like 5,7-DHT, are 5-HT neurotoxins.

Hyperthermia-enhanced serotonin (5-HT) depletion resulting from d-fenfluramine exposure is preventable

Recent findings indicate that elevations in body temperature during acute d-fenfluramine (Fen) exposure enhance long-term 5-HT depletion. Therefore, we hypothesized that when repeated exposure to d-Fen produced repeated elevations in body temperature, 5-HT reductions would be greater in comparison to a single d-Fen exposure. Groups of animals were exposed to d-Fen for 1 or 4 days in a 28 degrees C environment. Exposure to d-Fen in the 28 degrees C environment induced an increase in body temperature and resulted in a long-term decrease in brain 5-HT. However, brain 5-HT was not different between the two groups. An additional experiment revealed that if the initial exposure to d-Fen does not induce elevations in body temperature, then long-term 5-HT depletion can be prevented. We conclude that the central nervous system rapidly adapts to the 5-HT depleting action of d-Fen thereby preventing further decreases in 5-HT concentrations from d-Fen exposure. In addition, this rapid adaptation circumvented the hyperthermia-enhanced 5-HT depletion that results from d-Fen exposure in a warm environment.

Effects of methamphetamine-induced neurotoxicity on the development of neural circuitry: a hypothesis

Exposure of the developing brain to methamphetamine has well-studied biochemical and behavioral consequences. We review: (1) the effects of methamphetamine on mature serotonergic and dopaminergic pathways; (2) the mechanisms of methamphetamine neurotoxicity and (3) the role of serotonergic and dopaminergic signaling in sculpting developing neural circuitry. Consideration of these data suggest the types of neural circuit alterations that may result from exposure of the developing brain to methamphetamine and that may underlie functional defects.

Neuroendocrine correlates of depression in abstinent heroin-dependent subjects

The functions of the central alpha-adrenergic, serotoninergic and dopaminergic systems were investigated in 28 heroin-dependent subjects 6-8 weeks after detoxification, and in 22 healthy control subjects (group C). Fourteen heroin-dependent subjects with depressive comorbidity (group A), and 14 heroin-dependent subjects without other Axis I and II pathologies (group B) were included among abstinent substance abusers. Norepinephrine (NE) function was evaluated by growth hormone (GH) responses to acute stimulation with clonidine (clon); serotonin (5-HT) function by prolactin (PRL) and cortisol (CORT) responses to acute stimulation with D-fenfluramine (D-fen) and dopamine (DA) function by GH and PRL responses to acute administration of bromocriptine (brom). Central NE activity, as measured by the GH-clon test, seems to be well preserved both in A and B subjects. PRL and CORT responses to D-fen were significantly blunted both in A subjects and in B subjects, in comparison with control subjects (C); the PRL response in A subjects was significantly lower than in B subjects. The DA system of B subjects was found unimpaired; in contrast, a significantly higher GH response to brom in A subjects (depressed) could express D2 post-synaptic receptor hypersensitivity and, therefore, decreased pre-synaptic DA release. In sum, the study of central monoamine function revealed an alteration only of the 5-HT system in detoxified heroin-dependent subjects without psychiatric comorbidity, which might be a trait character of these subjects, possibly involved in the pathogenesis of the disorder. A more significant impairment of 5-HT function and the hypersensitivity of post-synaptic DA receptors in A subjects suggests that specific biological correlates of psychiatric comorbidity may characterize substance abuser subtypes.

Serotonin reuptake inhibition does not enhance short term modulation of the exercise ventilatory response

Increased respiratory dead space causes a serotonin (5-HT) dependent augmentation of the exercise ventilatory response known as short term modulation (STM). Contrary to predictions, 5-HT reuptake inhibition with fluoxetine failed to enhance, and even impaired STM with large dead space volumes (0.4-0.6 L). In this study, we tested the hypotheses that: (1) fluoxetine similarly impairs STM with smaller dead space volumes (0.2 L); whereas (2) acute 5-HT release and reuptake inhibition with fenfluramine would enhance STM. Ventilatory and blood gas measurements were made on five goats (37-58 kg) during rest and exercise, with the mask alone or with increased dead space (0.2 L). STM protocols were performed following chronic fluoxetine (>/=21 days, 1 mg/kg, SQ, SID) and acute fenfluramine administration (1 mg/kg, IV). Following fluoxetine, STM was partially impaired. Fenfluramine had no detectable effects on STM. The data suggest that: (1) chronic fluoxetine diminishes STM, possibly via down-regulation of relevant 5-HT receptors, and (2) drugs that release 5-HT acutely do not enhance STM.

Comparative study of fluoxetine, sibutramine, sertraline and dexfenfluramine on the morphology of serotonergic nerve terminals using serotonin immunohistochemistry

We compared the effects of treatment with high doses of fluoxetine, sibutramine, sertraline, and dexfenfluramine for 4 days on brain serotonergic nerve terminals in rats. Methylenedioxymethamphetamine (MDMA) and 5,7-dihydroxytryptamine (5,7-DHT) were used as positive controls because both compounds deplete brain serotonin. Food intake and body weight changes were also monitored and yoked, pair-fed animals were used to control for possible changes in morphology due to nutritional deficits. Fluoxetine, sibutramine, sertraline and dexfenfluramine all produced a significant reduction in body weight. Fluoxetine, sibutramine and sertraline treatment resulted in no depletion of brain serotonin but produced morphological abnormalities in the serotonergic immunoreactive nerve network. In contrast, dexfenfluramine and MDMA depleted brain serotonin and produced morphological changes in the serotonin nerve network. These results indicate that even though fluoxetine, sibutramine and sertraline do not deplete brain serotonin, they do produce morphological changes in several brain regions (as identified by serotonin immunohistochemistry). Dexfenfluramine and MDMA, on the other hand, markedly deplete brain serotonin and also produce morphological changes. Collectively, these results lend support to the concept that all compounds acting on brain serotonin systems, whether capable of producing serotonin depletion or not, could produce similar effects on the morphology of cerebral serotonin systems.

Neurotoxic effects of +/-fenfluramine and phenteramine, alone and in combination, on monoamine neurons in the mouse brain

Until recently, (+/-)fenfluramine (FEN) was widely prescribed as an appetite suppressant. In animals, FEN is a potent and selective brain serotonin neurotoxin. The present studies assessed the effects of phentermine (PHEN), an appetite suppressant frequently used clinically in combination with FEN, on FEN-induced serotonin neurotoxicity. Groups (n = 6/group) of mice were treated with FEN (10 mg/kg), PHEN (20 mg/kg or 40 mg/kg), FEN (10 mg/kg) plus PHEN (20 mg/kg or 40 mg/kg), or vehicle twice daily for four days. Food intake and body weight were measured during and after drug treatment. Brains were evaluated for regional brain serotonin and dopamine axonal markers two weeks after drug treatment. PHEN enhanced the anorectic and weight-reducing effects of FEN. PHEN also significantly enhanced FEN's long-term toxic effects on 5-HT axons. This effect was evident in some (hypothalamus, striatum) but not all (hippocampus, cortex) brain regions examined. PHEN alone produced no long-term effects on 5-HT axonal markers. However, whether given alone or in combination with FEN, PHEN produced significant, dose-related decreases in striatal DA axonal markers. These results, coupled with those from previous studies, suggest that PHEN has the potential to exacerbate FEN-induced serotonin neurotoxicity, if utilized in certain doses. Further, the present results indicate that PHEN possesses dopamine (DA) neurotoxic potential. The relevance of these data to humans previously treated with FEN/PHEN is discussed.

The effect of fenfluramine dosage regimen and reduced food intake on levels of 5-HT in rat brain

Male Wistar rats received fenfluramine in subacute (5 mg/kg b.i.d. i.p. for 4 days) or escalating (0.5, 1, 1.5, 2, 3, 4 and 5 mg/kg b.i.d. i.p., each dose given for 4 days) doses. Saline-treated controls received food ad libitum, or were pair-fed with the fenfluramine-treated animals. Rats were killed 1, 15 and 30 days after drug withdrawal. On day 1, plasma and brain fenfluramine levels were higher, and hypothalamus norfenfluramine levels were lower following escalating compared to subacute dosing, although total active drug levels were unaltered. Both treatment regimes, and pair-feeding reduced food intake and body weight. Subacute fenfluramine reduced brain 5-hydroxytryptamine (5-HT) and 5-hydroxyindoleacetic acid (5-HIAA) levels for up to 30 days. Brain 5-HT and 5-HIAA levels were unaltered following escalating-doses of fenfluramine. Additionally, pair-feeding transiently decreased hippocampal 5-HT levels. These data suggest that escalating-doses of fenfluramine prevent the 5-HT-depleting effect of a sub-cute challenge without altering the anorexic action of the drug.

Neuropharmacological effects of low and high doses of repeated oral dexfenfluramine in rats: a comparison with fluoxetine

The neuropharmacological effects of repeated oral doses of dexfenfluramine (DF; 1.25-10 mg/kg, twice daily for 21 days) were examined in rats and related to the drug brain levels. Results were compared with fluoxetine (FL) given at similar doses relative to its anorectic ED50. Both drugs dose-dependently slowed body weight gain and reduced brain serotonin (5-HT). However, at 1.25 mg/kg DF caused only a slight and transient decrease in cortical 5-HT. Comparable doses of FL (6.25-12.5 mg/kg) lowered 5-HT more than DF, besides slightly reducing striatal dopamine. At higher doses DF markedly reduced 5-HT in all regions, and to a lesser extent noradrenaline in hippocampus. There was a negative relationship between 5-HT and log total active drug levels and the indole was approximately halved at drug levels about 50 times lower with DF than FL. However, the ratio between drug levels causing marked 5-HT reductions and those considered anorectic was similar for DF and FL because brain levels at the anorectic ED50 were higher with FL than DF. Long-lasting reductions of 5-HT were also observed but recovery was only consistently slow beginning from 5 mg/ kg DF. Comparable doses of FL could not be used because its general toxicity leads to the death of rats after only 2-4 multiples of its anorectic ED50.

Effects of repeated oral doses of dexnorfenfluramine on 5-HT levels and 5-HT uptake sites in rat brain

The effects of oral dexnorfenfluramine (DNF; 1-4 mg/kg, twice daily for 4 days), the active metabolite of dexfenfluramine, were examined on rat regional brain indole contents and [3H]citalopram binding. Two hours after the last dose, serotonin (5-HT) and 5-hydroxyindoleacetic acid (5-HIAA) were dose-dependently lowered at doses above 1.5 mg/kg, with slight regional differences. Cortical 5-HT uptake sites were reduced only at the highest dose. Above 2 mg/kg DNF also caused a more lasting reduction (4 weeks) of regional indoles and cortical 5-HT uptake sites. At this longer time while the decrease in hippocampal 5-HT levels and cortical 5-HT uptake sites remained essentially constant, cortical and striatal 5-HT levels were lowered less than at 2 h, suggesting a return toward control values.

Effects of dexfenfluramine or 5,7-dihydroxytryptamine on tryptophan hydroxylase and serotonin transporter mRNAS in rat dorsal raphe

Dexfenfluramine (DF), given in high doses, can produce long-lasting decreases in brain levels of serotonin (5-HT) and 5-HT transporter (5-HTT) protein. The purpose of this study was to determine if DF-induced decreases in 5-HT and 5-HTT in rat forebrain are correlated with compensatory changes in the expression of the genes for tryptophan hydroxylase (TPH) and 5-HTT in the dorsal raphe nucleus. Gene transcripts were measured by quantitative reverse transcription-polymerase chain reaction (RT-PCR). Rats were treated with either one or eight injections of DF at either high (10 mg/kg) or low (2 mg/kg) doses. A positive control group for 5-HT cell loss received a single cerebroventricular injection of 5,7-dihydroxytryptamine (DHT). Rats were killed either 5, 15 or 30 days after their last treatment. Paroxetine binding to the 5-HTT protein in frontal cortex was, as expected, reduced in all of the treated groups relative to vehicle controls. TPH mRNA levels in the dorsal raphe of animals that received DHT were significantly higher than those measured in all other treatment groups 15 days following treatment. By 30 days, the amount of TPH mRNA in DHT-treated rats had fallen to well below control levels. None of the DF regimens significantly affected TPH mRNA levels. Unlike the TPH mRNA changes in DHT-treated rats, the 5-HTT mRNA levels in the dorsal raphe declined progressively throughout the 30 day survival period. None of the DF regimens significantly affected 5-HTT mRNA levels. The significance of these data are discussed in terms of whether loss of forebrain markers for 5-HT reflects either the loss of fine caliber 5-HT axon terminals or a decrease in the expression of these markers in the somata of these cells which are located in the dorsal raphe.

Down-regulation of rat brain 5-HT uptake carriers after treatment with high doses of D-fenfluramine

Male rats were treated with 10 mg/kg D-fenfluramine (DF) i.p., twice a day for 4 days. Five days later there was a strong reduction (70-100%) in the Bmax of [3H]citalopram binding and the Vmax of [3H]5-HT uptake in cortical and hippocampal synaptosomes; 2 months after the treatment these parameters were reduced by 40-70%. The effect of treatment was also evaluated in synaptosomes preloaded with [3H]5-HT, superfused and exposed for 3 min to a releasing stimulus (15 mM K+ or 0.5 microM DF). In our experimental conditions, the stimulated [3H]5-HT release is Ca(2+)-dependent and takes place only from 5-HT nerve endings. The K(+)-stimulated release was not consistently altered by the DF treatment whereas DF-stimulated [3H]5-HT release was markedly reduced, either 5 days and 2 months after the treatment. The effect of chronic DF was different from the effect of i.c.v. 5,7-DHT, a specific 5-HT neurotoxin which completely abolished the K(+)-induced release. Since the decrease of synaptosomal [3H]5-HT uptake induced by 5,7-DHT (82%) was similar to that found after chronic DF (70-80%), these data suggest that the decrease of 5-HT uptake sites induced by chronic DF is not (only) due to neurodegeneration. That chronic DF could induce a functional down-regulation of 5-HT uptake sites (i.e. decreased density per intact nerve ending) was suggested by the decrease of DF-induced release, since the releasing activity of DF is dependent on functional 5-HT uptake sites. However, due to the characteristics of our model, our results are compatible with either the absence or the presence of a concomitant, partial neurodegeneration of 5-HT nerve endings in DF-treated rats. In summary, our data indicate that after treatment with high doses of DF, the 5-HT uptake carriers undergo a long-lasting down-regulation, thus totally or partly explaining the lower [3H]citalopram binding and the lower synaptosomal [3H]5-HT uptake.

Effect of chronic dexfenfluramine on Fos in rat brain

The acute appetite suppressant effect of dexfenfluramine (DF) in rats, which may depend upon its action to release serotonin (5-HT) in the brain, often declines with repeated dosing (tolerance). The mechanisms of this tolerance remain unclear. Previously, we used Fos-like immunoreactivity (Fos-IR) to map potential brain sites activated by single injections of DF in rats. A dose of 5 mg DF/kg activated the central amygdala (CeA), bed nucleus of the stria terminalis (BST), caudate-putamen (CPu), lateral parabrachial nucleus (LPB), nucleus tractus solitarius (NST), frontal cerebral cortex and the parvocellular paraventricular hypothalamic nucleus (PVN). We now report studies using Fos-IR in an attempt to understand which regions might underlie tolerance to the action of DF. Pretreatment of rats with an escalating dosage regimen of DF (0.5-4 mg/kg, i.p.) was associated with complete loss of Fos-IR to a probe dose (5 mg DF/kg) in the cortex, CPu, PVN and NTS, and partial loss of Fos-IR in the BST, CeA and LPB. Second, repeated treatment with DF (2 mg/kg), which has been shown to produce tolerance the anorexia caused by DF but not cholecystokinin (CCK), likewise reduced Fos-IR induced in the above brain regions, but had no effect on Fos-IR induced by either CCK or the 5-HT agonist, 5-carboxamidotryptamine. Third, repeated treatment with 5-HT (2 mg/kg, s.c.) had no effect on Fos-IR induced by a probe dose of DF. These data show that regionally heterogeneous hyporesponsiveness to the induction of Fos by DF develops after repeated low doses of DF; however, the Fos response to other putative anorectics or weight reducing agents is not affected. This may be related to behavioral tolerance.

Fenfluramine depletes serotonin from the developing cortex and alters thalamocortical organization

A previous experiment from our laboratory showed that neonatal destruction of cortical serotoninergic (5-HT) axons with 5,7-dihydroxytryptamine (5,7-DHT) reduced the size of the clusters of vibrissae-related thalamocortical axons. This result suggested an important role for 5-HT in thalamocortical development, but could be questioned because of potentially direct toxic effects of 5,7-DHT on thalamocortical axons. In the present study, 5-HT was depleted from the cortex using a different method, neonatal administration of +fenfluramine, and vibrissae-related patches of thalamocortical afferents were measured when animals reached 6 days of age. Fenfluramine reduced cortical 5-HT levels to 93.9 +/- 6.0% of normal (P < 0.01) and decreased the average area of vibrissae-related lamina IV patches by 23.8 +/- 4.4% (P < 0.05). Depletion of 5-HT with +fenfluramine did not significantly reduce body, brain, or cortical weight, or the overall dimensions of the somatosensory cortex. Thus, these results extend our previous studies by showing that thalamocortical organization can be altered when 5-HT is depleted without the potential for direct toxic effects on thalamic axons.

Depletion and time-course of recovery of brain serotonin after repeated subcutaneous dexfenfluramine in the mouse. A comparison with the rat

The indole-depleting effects of repeated subcutaneous doses of dexfenfluramine (D-F) (2.5, 5, 10, 20 and 40 mg/kg/day, for four days) in mice were examined with regard to the initial response and time-course of recovery and related to the pharmacokinetics of D-F and its active metabolite dexnorfenfluramine (D-NF). Steady-state plasma and brain concentrations of D-F rose dose-dependently with a metabolite-to-drug ratio averaging 0.4 in brain. This confirmed that in mice D-NF contributes less than in other species to the effects of D-F. Regional serotonin (5-HT) and 5-hydroxyindoleacetic acid (5-HIAA) contents were decreased dose-dependently 4 hr after the last injection of D-F. However, two weeks after D-F (2.5-10 mg/kg/day) brain indoles had almost totally recovered, and the long-term effects of the 20 mg/kg/day dose were completely reversed by six weeks, when significant effects are still observable in rats. Although substantial recovery was evident even at 40 mg/kg/day, 5-HT but not 5-HIAA was still slightly reduced nine weeks later. Comparative studies in rats given 2.5-20 mg/kg/day D-F indicated much more severe initial indole depletions than in mice. Brain levels of D-F and D-NF were much higher in rats than in mice. The total active drug brain concentration (D-F + D-NF) was significantly correlated with 5-HT content in both species, with approx 20 nmol/g of total drug causing 50% reduction. These findings point to species differences in D-F kinetics as a main reason for differences in the neurochemical response, supporting the view that the recovery of indoles over time is related to the extent of initial depletion, which in turn depends on critical drug brain concentrations. In view of the qualitative and quantitative species differences in the pharmacodynamics and pharmacokinetics of D-F neither of these rodent species is a suitable model for predicting potential drug toxicity in humans.

Brain serotonin, carbohydrate-craving, obesity and depression

Serotonin-releasing brain neurons are unique in that the amount of neurotransmitter they release is normally controlled by food intake: Carbohydrate consumption--acting via insulin secretion and the "plasma tryptophan ratio"--increases serotonin release; protein intake lacks this effect. This ability of neurons to couple neuronal signaling properties to food consumption is a link in the feedback mechanism that normally keeps carbohydrate and protein intakes more or less constant. However, serotonin release is also involved in such functions as sleep onset, pain sensitivity, blood pressure regulation, and control of the mood. Hence many patients learn to overeat carbohydrates (particularly snack foods, like potato chips or pastries, which are rich in carbohydrates and fats) to make themselves feel better. This tendency to use certain foods as though they were drugs is a frequent cause of weight gain, and can also be seen in patients who become fat when exposed to stress, or in women with premenstrual syndrome, or in patients with "winter depression," or in people who are attempting to give up smoking. (Nicotine, like dietary carbohydrates, increases brain serotonin secretion; nicotine withdrawal has the opposite effect.) It also occurs in patients with normal-weight bulimia. Dexfenfluramine constitutes a highly effective treatment for such patients. In addition to producing its general satiety-promoting effect, it specifically reduces their overconsumption of carbohydrate-rich (or carbohydrate-and fat-rich) foods.

The nature of d,l-fenfluramine-induced 5-HT reuptake transporter loss in rats

The administration of the anorexigenic drug d,l-fenfluramine (Ponderax) to laboratory animals results in a dose-dependent reduction in presynaptically located serotonergic reuptake transporter protein. This long-term effect may represent an altered mechanism of synthesis of the transporter (downregulation). Alternatively, fenfluramine may destroy the serotonergic terminals on which 5-HT transporters are located. To distinguish between these two alternatives, we applied an assay of neurotransmitter-specific nerve endings (alpha) to brain tissue from two animal models of reduced 5-HT transporter density. In Model 1, serotonergic nerve terminals were destroyed (rats received 5,7-dihydroxytryptamine [5,7-DHT] intracisternally); in Model 2, there was a loss of 5-HT transporter per se on otherwise intact serotonergic nerve terminals. The manner in which alpha declined as transporter density was decreased (reducing Vmax values) in animal Models 1 and 2 was found to be significantly different. In rats treated with fenfluramine, the association between 5-HT transporter density and alpha was the same as in the neurotoxic model.

Effect of d-fenfluramine on the indole contents of the rat brain after treatment with different inducers of cytochrome P450 isoenzymes

The effects of pretreatment with inducers of hepatic cytochrome P450 isoenzymes (phenobarbital, dexamethasone and beta-naphthoflavone) on the metabolism of d-fenfluramine (d-F) and its acute and long-lasting indole-depleting effects were studied in rats, in an effort to obtain further information on the importance of hepatic drug metabolism in relation to its neurochemical actions. Twenty-four hours after the last dose of each inducer, rats were injected with d-F hydrochloride (5 mg/kg, IP) and killed at various times thereafter for parallel determination of indoles and drug concentrations in plasma and brain. Additional rats were treated as above and killed 1 week after d-F hydrochloride (5 and 10 mg/kg) to study the recovery of indole in the cortex, a particularly sensitive brain area. Phenobarbital and beta-naphthoflavone and, to a lesser degree, dexamethasone, stimulated the metabolism of d-F, as evidenced by a decrease in plasma and brain areas under the curve (AUC) compared to vehicle-treated rats. This indicated that multiple isoenzymes are capable of mediating the drug's metabolism, primarily by N-dealkylation to d-norfenfluramine (d-NF). None of the inducers raised plasma and brain AUC of the nor-derivative, and in fact phenobarbital and particularly beta-naphthoflavone reduced it. These different effects were even apparent in rats given d-NF (2.5 mg/kg), indicating that both phenobarbital and beta-naphthoflavone also stimulate the sequential metabolism of the nor-metabolite (by N-deamintaion) which, however, is apparently enhanced most actively by beta-naphthoflavone-inducible forms of P-450.(ABSTRACT TRUNCATED AT 250 WORDS)

Quantitative aspects of drug and toxicant-induced astrogliosis

A universal cellular reaction to damage of the CNS is hypertrophy of astrocytes. The hallmark of this response, often termed 'reactive gliosis', is the enhanced expression of the major intermediate filament protein of astrocytes, glial fibrillary acidic protein (GFAP). This latter observation suggests that increased synthesis of GFAP would occur in response to diverse neurotoxic insults. To investigate this possibility, prototype neurotoxicants were administered to experimental animals and the effects of these agents on the tissue content of GFAP was determined by immunoassay. Assays of GFAP were found to reveal dose-, time- and region-dependent patterns of neurotoxicity at toxicant dosages below those that cause light microscopic evidence of cell loss or damage. Moreover, the temporal and regional increments in GFAP correspond to the temporal and regional patterns of argyrophilia, as revealed by the cupric silver degeneration stain of de Olmos. Our findings indicate that assays of GFAP represent a sensitive, simple and quantitative approach for evaluation of nervous system damage. Combining this indirect yet quantitative indicator of neurotoxicity with more traditional neuroanatomical endpoints, should augment the armamentarium of techniques useful for detection and characterization of neurotoxicity.

Quantitative immunocytochemical evaluation of serotonergic innervation in alcoholic rat brain

Neurotoxicity can be divided into three levels: depletion, degeneration and denervation. The first level is determined by the transmitter content, and the second and third levels, which require anatomical evaluation, can be analyzed by quantitative immunocytochemistry on a specific neurotransmitter system. An antibody specific to serotonin (5-HT) can reveal detailed normal as well as degenerative morphology of 5-HT neurons. Quantitative alterations of 5-HT fibers in a particular brain region indicate degenerative or plastic changes. This study demonstrates quantitative immunocytochemistry by using image analysis of immunostained 5-HT fibers in selectively-bred, alcohol-preferring and nonpreferring rats, which are known to have divergent drinking behaviors, and 5-HT contents in specific brain regions. The method of the image analysis is described in detail and the advantages and disadvantages of using this method to detect the degeneration of a particular fiber system are discussed. The 'area density' traditionally measured in image analysis was converted (with Zhou-Tam formula) into 'volume density' to correct the mismeasurement of fibers through the optical depth. Quantitative immunostaining shows that the difference in 5-HT fiber density in particular brain regions between the two rat lines is consistent with changes in content of the 5-HT/5-HIAA and hypersensitivity of 5-HT1 a receptor. This result indicates either (a) the 5-HT content is too low to be detected in the nerve terminals; (b) degeneration of 5-HT fibers occurs in P rats sometime during development; or (c) a smaller number of 5-HT fibers was preprogrammed in the brain regions of P than NP rats.(ABSTRACT TRUNCATED AT 250 WORDS)

Chronic D-fenfluramine decreases serotonin transporter messenger RNA expression in dorsal raphe nucleus

In situ hybridization was used to measure the effects of chronic fenfluramine administration on serotonin transporter messenger RNA expression in cells of the dorsal raphe nucleus complex. Fenfluramine produced a significant, but transient, down-regulation of serotonin transporter mRNA in cells which lie in the ventral portion of the dorsal raphe nucleus, but not in the dorsal part of the dorsal raphe nucleus. Our findings suggest that cells which lie in the ventral part of the dorsal raphe nucleus are more sensitive to the effects of chronic fenfluramine administration, but that fenfluramine does not cause long-term changes in gene expression in serotonin cell bodies.

Loss of serotonin uptake sites and immunoreactivity in rat cortex after dexfenfluramine occur without parallel glial cell reactions

The frontal cortices of rats which received either D,L- or D-fenfluramine (DFEN) for 4 days were examined 18 h to 2 weeks following treatment for changes in synaptosomal uptake of serotonin (5HT), paroxetine binding, 5HT-immunoreactivity (5HT-IR), and both astrocytic (GFAP) and microglial markers. Additional rats received intracerebroventricular injections of the neurotoxin 5,7-dihydroxytryptamine (DHT). Consistent with previous reports, D,L- and DFEN produced dose-dependent losses of both 5HT uptake and paroxetine binding, and loss of 5HT-IR which coincided with an abnormal or 'swollen' appearance of immunoreactive axon processes. Recovery of these serotonergic indices was greatest following the lowest doses of DFEN, but was absent after 5,7-DHT treatment. No evidence for an increase in GFAP synthesis or microglial activation was observed in frontal cortices of rats treated with either DFEN or 5,7-DHT. We conclude that the presence of swollen 5HT-IR axons in the cortices of both the 5,7-DHT and DFEN groups is insufficient to trigger the glial responses often associated with neuronal degeneration. Thus, it remains to be determined if swollen 5HT-IR axons are a prelude to neurodegeneration, or whether they represent reversible changes in axonal immunochemistry associated with decreases in 5HT levels. The implications of the data for the clinical safety of DFEN are briefly discussed.

Effect of d-fenfluramine and 5,7-dihydroxytryptamine on the levels of tryptophan hydroxylase and its mRNA in rat brain

Repeated high doses of d-fenfluramine (dF; 10 mg/kg, i.p. twice daily for 4 days) markedly reduced serotonin (5-HT) concentrations in the hippocampus and striatum of rat brain up to 1 month after treatment, while tryptophan hydroxylase (TPH) levels were reduced only in the hippocampus 5 days after injection. Unlike dF, an intracerebroventricular (i.c.v.) injection of 5,7-dihydroxytryptamine (5,7-DHT 150 micrograms/20 microliters) induced a marked and long-lasting reduction of 5-HT and TPH in both brain regions. Thirty days after injection, 5,7-DHT, but not dF, markedly reduced the number of labelled neurons in the dorsal and ventral regions of the nucleus raphe dorsalis (NRD) and raised the levels of TPH mRNA in the spared neurons at all times examined. TPH mRNA levels were raised 5 and 15 days after dF treatment in the NDR suggesting that changes in the TPH gene expression or transcript stability result following 5-HT depletion. These data are in agreement with the suggestion that 5,7-DHT damages 5-HT nerve terminals and perikarya, but leave unanswered the question of the mechanism of the long-lasting reduction of 5-HT levels caused by high, repeated doses of dF.

The regeneration of d,l-fenfluramine-destroyed serotonergic nerve terminals

The regeneration of serotonergic nerve terminals subsequent to their destruction by high-dose fenfluramine administration was examined. Treating rats with fenfluramine (80 mg/kg over 2 days) destroyed 80% of serotonergic nerve terminals, indicated by reduced maximal [3H]paroxetine binding to 5-hydroxytryptamine (5-HT) uptake sites on synaptic membranes (Bmax) and maximal [14C]5-HT uptake rate into synaptosomes (Vmax). 25 weeks later, these indices of serotonergic nerve terminals had returned to 72% of control. Maximal synaptosomal loading (alpha) with [14C]5-HT also recovered (to 79% of control), reflecting an increased number of serotonergic synaptosomes. This suggests that the rebound in 5-HT uptake site density found after fenfluramine illustrates the regeneration of 5-HT-containing nerve endings.

New evidence for a loss of serotonergic nerve terminals in rats treated with d,l-fenfluramine

Fenfluramine has been classified as a neurotoxin because animals treated with this anorectic lose 5-HT uptake sites located on serotonergic nerve terminals. However, there are two possible bases for this finding: either uptake sites are lost because the terminals themselves have been destroyed (neurotoxicity); or uptake sites are lost from otherwise intact terminals. To distinguish between these possibilities, we established an animal model in which male Wistar rats were injected (intraperitoneally) with an irreversible 5-HT uptake site antagonist (EEDQ). Since their 5-HT sites were inhibited (blocked) non-competitively, by this agent, such animals had effectively lost 5-HT uptake sites from intact serotonergic terminals. Synaptosomes prepared from such animals showed the predicted reduction in the Bmax of [3H]paroxetine binding to the 5-HT uptake site, and a reduction in the Vmax of [14C]5-HT uptake. However, they showed no significant reduction in maximal [14C]5-HT loading (alpha) compared with synaptosome from sham-injected controls. In contrast, fenfluramine-treated animals showed reduced [3H]paroxetine binding, reduced maximal [14C]5-HT uptake and significantly (P < 0.02) reduced synaptosomal [14C]5-HT loading. Therefore, the results suggest that fenfluramine does indeed cause the destruction of serotonergic nerve terminals.

Administration of dexfenfluramine in pregnant rats: effect on brain serotonin parameters in offspring

Dexfenfluramine (DFEN) was infused SC at doses of either 6 or 12 mg/kg/day during the last week of pregnancy in rats. Compared with untreated controls, weight gain of dams was attenuated by DFEN without effect on the number or birth weight of offspring. Brain serotonin (5-HT) concentration and/or paroxetine binding to the 5-HT uptake carrier was reduced by 20% on the day after birth in one study but not in two other studies. No decreases in brain 5-HT parameters were observed on or after the sixth postnatal day. In contrast, mothers sustained large depletions of brain 5-HT when measured at least 3 weeks after giving birth. These data indicate that fetal brains either are protected from or recover rapidly from the 5-HT-depleting actions of high-dosage regimens of DFEN in rats.

Single- and multiple-dose kinetics of d-fenfluramine in rats given anorectic and toxic doses

1. High parenteral doses of a twice-daily schedule of d,l-fenfluramine (d,l-F) may cause long-lasting decrease of functional indices of brain serotoninergic neurones in rats. The single- and multiple-dose (b.i.d. x 4 days) kinetics of low (1.25 mg/kg) and high (12.5 mg/kg) subcutaneous (s.c.) doses of d-F, which accounts of the anorectic effects of the racemate, and its deethylated metabolite d-norfenfluramine (d-NF), were therefore examined and compared with those of pharmacologically effective oral doses (0.3-1.25 mg/kg) in rats. 2. There were dose-dependent alterations of kinetic parameters after s.c. and oral dosing, indicating that hepatic clearance of d-F in the rat can be saturated either by increasing the size of the single dose or during repeated dosing. Nonlinearity was also observed for d-NF. Consequently at high doses exposure of rat to the drug, as measured by the sum of area under the plasma concentration-time curve (AUC) of d-F and d-NF considerably exceeded that expected from simple dosage considerations, particularly with repeated administration of d-F. 3. Total exposure at the high doses considerably exceeded that at pharmacological doses, however, indicating an ample margin in favour of anorectic activity. The possibility that the long-term depletion of brain 5-HT by d-F and/or its metabolite d-NF may have relevance at the usual therapeutic dose, is discussed.