Excitotoxic amino acids as neuroendocrine research tools

Standards of evidence in chronobiology: A response

A number of recent studies have debated the existence and nature of clocks outside the suprachiasmatic nucleus that may underlie circadian rhythms in conditions of food entrainment or methamphetamine administration. These papers claim that either the canonical clock genes, or the circuitry in the dorsomedial nucleus of the hypothalamus, may not be necessary for these forms of entrainment. In this paper, we review the evidence necessary to make these claims. In particular, we point out that it is necessary to remove classical conditioning stimuli and interval timer (homeostatic) effects to insure that the remaining entrainment is due to a circadian oscillator. None of these studies appears to meet these criteria for demonstrating circadian entrainment under these conditions. Our own studies, which were discussed in detail by a recent Review in these pages by Mistlberger and colleagues, came to an opposite conclusion. However, our studies were designed to meet these criteria, and we believe that these methodological differences explain why we find that canonical clock gene Bmal1 and the integrity of the dorsomedial nucleus are both required to produce true circadian entrainment under conditions of restricted feeding.

A non-excitatory paradigm of glutamate toxicity

Retinal ganglion cells are driven by glutamatergic synapses, but they are also very susceptible to glutamate toxicity. Whereas the conventional excitotoxicity model of glutamate-induced cell death requires membrane depolarization, we have found that glutamate toxicity need not be linked with excitation. A large subset of ganglion cells possesses high-affinity kainate receptors that are calcium permeable. At 1-5 microM, kainate produced elevation of internal calcium but did not significantly depolarize ganglion cells. This low concentration of kainate caused ganglion cell death, which could be inhibited by specific kainate receptor antagonists. The toxic effect of kainate may be associated with calcium influx, because toxicity was reduced by polyamines that suppress calcium influx and by an inhibitor of calcium phosphatase. Thus activation of ionotropic glutamate receptors can produce neurotoxicity uncoupled from neuroexcitation.

Prolactin response to anesthetic stress and beta-endorphin is altered in female rats treated neonatally with monosodium glutamate

MSG (4 mg/g, sc) or saline was administered to neonatal female rats on days 1, 3, 5, 7 and 9. Study 1) Prl levels were assessed at 30, 60 and 75 days after birth to monitor possible development of hyperprolactinemia. No hyperprolactinemia was observed at any time studied. Study 2) MSG and control rats were administered pentobarbital anesthesia at 2 months of age. At 20, 60 and 90 min following anesthesia, plasma was collected for assay of Prl. 20 min prior to the 90 min bleeding, BE (5 micrograms/5 ul) was stereotaxically administered, into the third ventricle. MSG-treated rats had an attenuated Prl response to the stress of anesthesia (bleeding 1). Prl levels in control and MSG-treated rats were similar at 60 min post-anesthesia (bleeding 2) which represented a return of Prl levels to baseline after stress-induced elevation of Prl. Control and MSG-treated rats exhibited an increase in plasma Prl following intracerebroventricular BE; however, the amplitude of this response was markedly attenuated in the MSG-treated animals (bleeding 3). Thus, an observed loss of TH-positive neurons in the arcuate nucleus of the hypothalamus following MSG treatment and the attenuated response of Prl to anesthesia-stress and BE administration suggests that Prl secretion in response to these agents is operative through inhibition of the TIDA system. Furthermore, these studies show that the Prl response to these agents (anesthetic and BE) is intact but sub-operational in MSG-treated rats.

Excitotoxin lesions of the zona incerta/lateral tegmentum continuum: effects on male sexual behavior in rats

In male rats, ibotenic acid and N-methyl-D-aspartic acid were used to destroy neuronal perikarya intrinsic to an anterior-posterior continuum including the caudal zona incerta and lateral tegmentum. Some lesions virtually eliminated male sexual behavior - an effect most closely associated with damage to the lateral hypothalamus and zona incerta. Many lesioned males who copulated to ejaculation with normally active females showed little or no mating with receptive, but relatively inactive, females. Although it is possible to identify a critical region within the subthalamus whose destruction eliminates male sexual behaviour, sexually-relevant neuronal cell bodies appear to be distributed throughout the lateral hypothalamic/incertal/tegmental continuum.

Computer-assisted analysis of behavior-brain damage relationships

Experimental destruction of subcortical brain tissue by electrolytic, radio frequency, or excitotoxin lesions is commonly used in the study of the neural control of behavior. Relating variability in the locus and extent of brain damage to lesion-produced variability in behavior is essential to a complete analysis of the behavioral effects of brain damage. We describe the use of commercially available computer accessories and software for measuring the area of irregular polygons to accurately and rapidly estimate the amount of damage to any of a number of brain structures included within an area of experimental destruction. We then describe the use of statistical programs for correlation and multiple regression in calculating relationships between damage to particular structures and behavior. As an example, we show that excitotoxin (and presumably axon-sparing) lesions of lateral hypothalamic tissue adjacent to the caudal zona incerta virtually eliminate male sexual behavior in rats. The use of multiple regression analysis, in conjunction with procedures for calculating destruction to different brain regions, provides a way of correlating brain damage and behavior that has widespread application in studies of the behavioral effects of brain lesions.

Excitotoxic lesions of the paraventricular hypothalamus: metabolic and cardiac effects

The excitotoxin, N-methyl-D-aspartic acid (NMDA), was used to lesion cell bodies, but not fibers-of-passage, in the paraventricular hypothalamus. Bilateral injections of NMDA (12.6 nmol/100 nl) were made into the paraventricular hypothalamus in halothane-anesthetized male Sprague-Dawley rats. Water intake, food intake, urine output and body weight were measured daily for 26 days after lesioning. Lesioned rats exhibited a modest, but significant, reduction in the rate of gain of body weight, which was most closely correlated with decreases in food intake. Water intake and urine output were not significantly different among the groups. Resting blood pressure, heart rate and baroreflex sensitivity (using the infusion of phenylephrine method) were similar in conscious animals of both groups, 4-5 weeks after lesioning. Neuronal loss, primarily of parvocellular elements, was evident in the paraventricular hypothalamus and neuronal loss frequently extended into the ventro-medial thalamus adjacent to the paraventricular hypothalamus in NMDA-lesioned rats. In a second experiment, injections of NMDA were given acutely into the paraventricular hypothalamus of halothane-anesthetized rats. Upon recovery from anesthesia, behavioral excitation and increases in blood pressure and heart rate were evident for 1-2 hr. Histological examination of hearts taken 48 hr after injection of NMDA revealed a largely mononuclear inflammatory infiltration, hyperemia and myocardial hemorrhage and focal myocardial necrosis. Inflammatory and degenerative changes were most prominent in the left ventricular subendocardium. The cardiomyopathy possessed similarities with catecholamine-induced myocardial necrosis. The results indicated that NMDA-induced lesions of parvocellular elements of the paraventricular hypothalamus did not cause hyperphagia or obesity or alter the resting systemic circulatory function. However, an inflammatory cardiomyopathy, termed "excitotoxin-induced myocardial necrosis", was associated with injections of NMDA into the hypothalamus. Excitotoxin-induced myocardial necrosis may complicate any hemodynamic studies performed in rats in which lesions of the CNS have been produced by means of application of excitotoxins.

Transneuronal degeneration in the midbrain central gray following chemical lesions in the ventromedial nucleus: a qualitative and quantitative analysis

In the preceding experiments with electrolytic lesions of the ventromedial nucleus of the hypothalamus, we showed pre- and postsynaptic degeneration in the midbrain central gray of the rat. The postsynaptic degeneration seen may indicate a transneuronal effect of the ventromedial nucleus on the midbrain central gray. Electrolytic lesions, however, destroy afferent endings and fibers in passage, so that the postsynaptic degeneration seen in the midbrain central gray may be due to retrograde degeneration of midbrain central gray afferents to the ventromedial nucleus or due to degeneration of fibers in passage. In order to distinguish among these possibilities, chemical, i.e. kainic acid and N-methyl aspartate, lesions were made in the ventromedial nucleus and the ultrastructure of the midbrain central gray and cerebral cortex was examined at various intervals following the lesions. Both of these excitotoxins have been shown to destroy neurons, sparing afferent terminals and fibers in passage. Animals receiving kainic acid lesions in the right ventromedial nucleus were allowed to survive for one week, and animals receiving N-methyl aspartate lesions in the right ventromedial nucleus were permitted to survive for four, eight, and 20 days. Midbrain central gray tissue of unlesioned animals served as a control for both kainic acid and N-methyl aspartate lesions. In addition, other control animals received injections of the same amount of N-methyl aspartate in the right parietal cortex and were permitted to survive for four and eight days. For each of the above injection and survival conditions, the left cortex and subdivisions of the midbrain central gray were removed and processed for electron microscopy. Animals receiving ventromedial hypothalamic lesions with both kainic acid and N-methyl aspartate showed signs of pre- and postsynaptic degeneration. A quantitative analysis (General Linear Model Procedure) of degeneration was performed on the cortex and midbrain central gray of animals receiving N-methyl aspartate lesions in the ventromedial nucleus and cortex, and several parameters were measured. Animals receiving ventromedial hypothalamic lesions and surviving for eight and 20 days show significantly higher ratios of degenerating presynaptic elements to total presynaptic elements, degenerating postsynaptic elements to total postsynaptic elements, and degenerating total elements to total elements, in the midbrain central gray than in the cortex. Furthermore, the ratio of degenerating postsynaptic elements to total postsynaptic elements is larger than the other ratios.(ABSTRACT TRUNCATED AT 400 WORDS)

Projections of ventromedial hypothalamic neurons to the midbrain central gray: an ultrastructural study

Ventromedial hypothalamic projections to the midbrain central gray may be involved in the mediation of female reproductive behavior. In order to demonstrate and examine projections of the ventromedial nucleus in the midbrain central gray, in the rat, electrolytic lesions were placed in the ventromedial nucleus and the midbrain central gray was examined for ultrastructural signs of degeneration at various intervals, i.e. 27.5 h and two, four, six and eight days following the lesions. The fine structure of the midbrain central gray of unlesioned animals was also examined to characterize its normal morphology and to establish a baseline with which to compare the effects of the lesion. In unlesioned animals, the neuropil of midbrain central gray contained several synaptic types, with axodendritic synapses appearing to be the most predominant. Dendrites contained well-preserved microtubules. Synaptic endings contained many clear, round vesicles and some contained dense-cored vesicles as well. Neuropil synapses were both asymmetric and symmetric. Cell bodies were characterized by light cytoplasm and had asymmetric and symmetric synapses on their surface. Following electrolytic lesions in the ventromedial nucleus, various types of degenerative patterns were seen in the midbrain central gray, including electron-dense, flocculent, watery, and pinocytotic degeneration. Specific characteristics of degeneration included shrunken, dense axons and endings, clumped synaptic vesicles, abnormally large, dark mitochondria, membranous sacs of various sizes, swollen endings with reduced numbers of synaptic vesicles, and endings and processes containing large numbers of coated vesicles. Some of these signs were already evident at 27.5 h following the lesion. In addition, degenerating postsynaptic processes and cell bodies were seen in the midbrain central gray. At 27.5 h survival time, degenerating dendritic processes often appeared swollen, devoid of microtubules, and contained enlarged mitochondria. At longer survival times neuronal degeneration was observed in the midbrain central gray, characterized by electron-dense cell bodies and pycnotic nuclei. Both degenerating pre- and postsynaptic elements appeared to be engulfed by glial processes. Control lesions in non-hypothalamic regions which project to the midbrain central gray, i.e. nucleus gigantocellularis and pontine reticular formation and in a non-projecting region, i.e. parietal cortex, were performed.(ABSTRACT TRUNCATED AT 400 WORDS)

Bidirectional transfer between kindling induced either by L-glutamate or L-aspartate and electrical stimulation in rats

Repeated spaced injections into the amygdala (AM) of rats of a subconvulsive dose of either L-glutamate (GLU) or L-aspartate (ASP) produced progressive seizure development culminating in generalized convulsion strikingly similar to electrically kindled seizure. Both GLU- and ASP-kindled AM sites responded to electrical stimulation with very rapid development of kindled seizure. When electrically kindled AM sites received identical injections of GLU or ASP, a similar positive transfer effect was obtained. These results suggest that (1) either GLU or ASP alone can kindle the AM and (2) GLU, ASP or GLU/ASP kindling share common neurobiological mechanisms with electrical kindling.

Bidirectional transhemispheric interaction between amygdaloid kindling induced by excitatory amino acids and electrical stimulation

Using transhemispheric transfer and post-transfer interference as specific measures, the relationship between chemical and electrical kindling at the amygdala (AM) in rats was examined. Chemical kindling was accomplished by repeated injections of a subconvulsive dose of combined L-glutamate (GLU) and L-aspartate (ASP) in a molar ratio of 1:3 (GLU 0.375 microM:ASP 1.125 microM). Strong bidirectional transhemispheric transfer between electrical and chemical kindling occurred at the secondary-site AM. In addition, at the primary-site retest, bidirectional posttransfer interference between electrical and chemical kindling was observed. These findings strongly suggest that endogenous GLU/ASP participates in the transsynaptic changes which are presumed to underlie the kindling phenomenon.

Bidirectional transfer between kindling induced by excitatory amino acids and electrical stimulation

Chemical (C) kindling by means of repeated spaced injections into the amygdala (AM) of a subconvulsive dose of L-glutamate and L-aspartate combined in a molar ratio of 1:3. (Glu/Asp) produced progressive seizure development culminating in generalized convulsion strikingly similar to electrically kindled Stage 5 seizure in rats. These C-kindled AM sites responded readily to electrical stimulation with very rapid development of kindled seizure. When a separate group of rats electrically kindled at the AM were subjected to identical C-kindling at the kindled AM site, a similar positive transfer effect was observed. In addition, 30 days following the last injection, kindled Stage 5 seizure was triggered with a single injection of Glu/Asp, one-half of the dose used for C-kindling. These results suggest that the glutamate and/or aspartate systems participate in the development and persistence of increased seizure susceptibility induced by AM kindling.

Aging and responses to toxins in female reproductive functions

Reproductive senescence in laboratory rodents has been well characterized using vaginal smear analysis, a relatively simple and inexpensive technique to monitor changes in estrous cyclicity. Although some differences exist, laboratory animal models are available to study most aspects of age-related changes in human reproductive physiology (4,13). Thus presumptive reproductive toxins can be tested on laboratory rodents using the qualitative and quantitative parameters for estrous cyclicity described here to assess damage. Before implementing such tests, however, certain limitations must be considered. Quantitative changes in estrous cycles, for example, cycle length distribution, which indicate more subtle impairments in reproductive function, require more refined data analysis. In particular, baseline data for at least one month is required before exposure to the presumptive toxin. In addition, the period of maximum cycling regularity is fairly short (three to four months duration) so long-term exposures to presumptive toxins would not be applicable. Longer term exposure (greater than or equal to six months) would be permissible if qualitative changes in cyclicity (i.e., acyclic vs. cyclic) were the dependent variable used to assess toxicity. Finally, if other parameters of reproductive function (e.g., fertility, litter size) are used to assess reproductive toxicity, we urge that cycling characteristics of the offspring be carefully monitored to assess possible cryptic damage to later neuroendocrine functions that occurred to the fetuses in utero.

Area postrema lesions block feeding induced by systemic injections of monosodium glutamate

Glutamate is an amino acid neurotransmitter capable of producing widespread receptor-mediated neuronal excitation. Recently we reported that high doses of monosodium glutamate (MSG) given systemically stimulate food intake in a dose-related fashion. Since glutamate does not cross the blood-brain barrier, it seems possible that feeding was stimulated by an action of glutamate on neurons within circumventricular organs (CVOs), areas of the brain in which the blood-brain barrier is deficient. In this experiment, we tested the hypothesis that systemic MSG stimulates feeding by an action on the area postrema (AP), a CVO in the caudal hindbrain. AP-lesioned rats (APLs) and sham-operated controls (shams) were injected with saline or MSG (2 and 6 g/kg, SC, one dose per week). Food intake was measured for 3 hr immediately following the injection. Shams increased their food intake significantly in response to both doses of MSG, but APLs did not. This result suggests that systemic glutamate may stimulate feeding by an action on the AP.

Dose-related stimulation of feeding by systemic injections of monosodium glutamate

Monosodium glutamate (MSG) is an excitotoxin capable of both stimulating and lesioning neurons in circumventricular organs (CVOs) after systemic administration. In this study, MSG and equiosmotic concentrations of NaCl were administered subcutaneously to adult rats in order to observe the effects on food and water intake. MSG (0.5, 1, 2 and 6 g/kg), but not NaCl, stimulated feeding. The magnitude of the feeding was dose-related. After the highest dose, rats consumed 4.4 g of pelleted food. Since MSG does not cross the blood-brain barrier, we conclude that feeding was stimulated by an action of glutamate on CVOs. Doses of MSG that stimulated feeding did not alter blood glucose concentration. Neonatal MSG treatment, which is known to be more damaging to circumventricular neurons than adult treatment, greatly reduced or abolished subsequent MSG-induced stimulation of feeding in adults. Both MSG and NaCl stimulated drinking. Since the magnitude of the drinking response was similar for both solutes and was directly related to the osmotic strength of the solutions, we conclude that the drinking response after MSG was mediated by cellular dehydration.