Endocannabinoid signaling, glucocorticoid-mediated negative feedback, and regulation of the hypothalamic-pituitary-adrenal axis

The hypothalamic-pituitary-adrenal (HPA) axis regulates the outflow of glucocorticoid hormones under basal conditions and in response to stress. Within the last decade, a large body of evidence has mounted indicating that the endocannabinoid system is involved in the central regulation of the stress response; however, the specific role endocannabinoid signaling plays in phases of HPA axis regulation, and the neural sites of action mediating this regulation, were not mapped out until recently. This review aims to collapse the current state of knowledge regarding the role of the endocannabinoid system in the regulation of the HPA axis to put together a working model of how and where endocannabinoids act within the brain to regulate outflow of the HPA axis. Specifically, we discuss the role of the endocannabinoid system in the regulation of the HPA axis under basal conditions, activation in response to acute stress, and glucocorticoid-mediated negative feedback. Interestingly, there appears to be some anatomical specificity to the role of the endocannabinoid system in each phase of HPA axis regulation, as well as distinct roles of both anandamide and 2-arachidonoylglycerol in these phases. Overall, the current level of information indicates that endocannabinoid signaling acts to suppress HPA axis activity through concerted actions within the prefrontal cortex, amygdala, and hypothalamus.

Presynaptic noradrenergic regulation of glutamate inputs to hypothalamic magnocellular neurones

Glutamate and norepinephrine transmitter systems play critical roles in the synaptic control of hypothalamic magnocellular neurones. We recently reported on a norepinephrine-sensitive glutamate circuit within the paraventricular nucleus (PVN) that projects to magnocellular neurones. Here, we present evidence for norepinephrine regulation of glutamate release in the PVN and supraoptic nucleus (SON) via actions on presynaptic terminals. Whole-cell synaptic currents were recorded in magnocellular neurones of the SON and PVN in an acute slice preparation. Bath application of norepinephrine (100 microm) caused a robust, reversible increase in the frequency of spontaneous glutamatergic excitatory postsynaptic currents in 100% of SON neurones (246%) and in 88% of PVN magnocellular neurones (259%). The norepinephrine-induced increase in glutamate release was mediated by activation of both presynaptic alpha1 receptors and alpha2 receptors, but the alpha1-receptor component was the predominant component of the response. The presynaptic actions of norepinephrine were predominantly, although not completely, resistant to blockade of Na-dependent spikes, implicating a presynaptic terminal locus of action. Interestingly, the spike-dependent component of the response was greater in PVN than in SON magnocellular neurones. This robust presynaptic facilitation of glutamate release by norepinephrine, combined with the known excitatory postsynaptic actions of norepinephrine, activational effects on local glutamate circuits, and inhibitory effects on gamma-aminobutyric acid release, indicate a strong excitatory role of norepinephrine in the regulation of oxytocin and vasopressin release during physiological stimulation.

Neurosecretory and non-neurosecretory parvocellular neurones of the hypothalamic paraventricular nucleus express distinct electrophysiological properties

Parvocellular neurones of the hypothalamic paraventricular nucleus (PVN) comprise neurosecretory and non-neurosecretory subpopulations. We labelled neurosecretory neurones with intravenous injection of the retrograde tracer, fluoro-gold, and recorded from fluoro-gold-positive and negative PVN parvocellular neurones in hypothalamic slices. Non-neurosecretory parvocellular neurones generated a low-threshold spike (LTS) and robust T-type Ca2+ current, whereas neurosecretory neurones showed no LTS and a small T-current. LTS neurones were located in non-neurosecretory regions of the PVN, and non-LTS neurones were located in neurosecretory regions of the PVN. These findings indicate that neurosecretory and non-neurosecretory subtypes of parvocellular PVN neurones express distinct membrane electrical properties.

A slow transient potassium current expressed in a subset of neurosecretory neurons of the hypothalamic paraventricular nucleus

Type I putative magnocellular neurosecretory cells of the hypothalamic paraventricular nucleus (PVN) express a prominent transient outward rectification generated by an A-type potassium current. Described here is a slow transient outward current that alters cell excitability and firing frequency in a subset of type I PVN neurons (38%). Unlike most of the type I neurons (62%), the transient outward current in these cells was composed of two kinetically separable current components, a fast activating, fast inactivating component, resembling an A-type potassium current, and a slowly activating [10-90% rise time: 20.4 +/- 12.8 (SE) ms], slowly inactivating component (time constant of inactivation: tau = 239.0 +/- 66.1 ms). The voltage dependence of activation and inactivation and the sensitivity to block by 4-aminopyridine (5 mM) and tetraethylammonium chloride (10 mM) of the fast and slow components were similar. Compared to the other type I neurons, the neurons that expressed the slow transient outward current were less excitable when hyperpolarized, requiring larger current injections to elicit an action potential (58.5 +/- 13.2 vs. 15.4 +/- 2.4 pA; 250-ms duration; P < 0.01), displaying a longer delay to the first spike (184.9 +/- 15.7 vs. 89.7 +/- 8.8 ms with 250- to 1,000-ms, 50-pA current pulses; P < 0.01), and firing at a lower frequency (18. 7 +/- 4.6 vs. 37.0 +/- 5.5 Hz with 100-pA current injections; P < 0. 05). These data suggest that a distinct subset of type I PVN neurons express a novel slow transient outward current that leads to a lower excitability. Based on double labeling following retrograde transport of systemically administered fluoro-gold and intracellular injection of biocytin, these cells are neurosecretory and are similar morphologically to magnocellular neurosecretory cells, although it remains to be determined whether they are magnocellular neurons.

Noradrenergic regulation of parvocellular neurons in the rat hypothalamic paraventricular nucleus

Noradrenergic projections to the hypothalamic paraventricular nucleus have been implicated in the secretory regulation of several anterior pituitary hormones, including adrenocorticotropin, thyroid-stimulating hormone, growth hormone and prolactin. In an attempt to elucidate the effects of norepinephrine on the central control of pituitary hormone secretion, we looked at the actions of norepinephrine on the electrical properties of putative parvocellular neurons of the paraventricular nucleus using whole-cell current-clamp recordings in hypothalamic slices. About half (51%) of the putative parvocellular neurons recorded responded to norepinephrine with either a synaptic excitation or a direct inhibition. Norepinephrine (30-300microM) caused a marked increase in the frequency of excitatory postsynaptic potentials in about 36% of the parvocellular neurons recorded. The increase in excitatory postsynaptic potentials was blocked by prazosin (10microM), but not by propranolol (10microM) or timolol (20microM), indicating that it was mediated by alpha(1)-adrenoreceptor activation. It was also blocked by ionotropic glutamate receptor antagonists, suggesting that the excitatory postsynaptic potentials were caused by glutamate release. The increase in excitatory postsynaptic potentials was completely abolished by tetrodotoxin, indicating the spike dependence of the norepinephrine-induced glutamate release. In a separate group comprising 14% of the parvocellular neurons recorded, norepinephrine elicited a hyperpolarization (6.2+/-0.69mV) that was blocked by the beta-adrenoreceptor antagonists, propranolol (10microM) and timolol (20microM), but not by the alpha(1)-receptor antagonist, prazosin (10microM). This response was not blocked by tetrodotoxin (1.5-3microM), suggesting that it was caused by a direct postsynaptic action of norepinephrine. The topographic distribution within the paraventricular nucleus of the norepinephrine-responsive and non-responsive parvocellular neurons was mapped based on intracellular biocytin labeling and neurophysin immunohistochemistry. These data indicate that one parvocellular subpopulation, consisting of about 36% of the paraventricular parvocellular neurons, receives an excitatory input from norepinephrine-sensitive local glutamatergic interneurons, while a second, separate subpopulation, representing about 14% of the parvocellular neurons in the paraventricular nucleus, responds directly to norepinephrine with a beta-adrenoreceptor-mediated inhibition. This suggests that excitatory inputs to parvocellular neurons of the paraventricular nucleus are mediated mainly by an intrahypothalamic glutamatergic relay, and that only a relatively small subset of paraventricular parvocellular neurons receives direct noradrenergic inputs, which are primarily inhibitory.

Neurosteroid modulation of GABA IPSCs is phosphorylation dependent

The neurosteroid 3alpha-hydroxy-5alpha-pregnan-20-one (allopregnanolone) facilitates GABA(A) receptor-mediated ionic currents via allosteric modulation of the GABA(A) receptor. Accordingly, allopregnanolone caused an increase in the slow decay time constant of spontaneous GABA-mediated IPSCs in magnocellular neurons recorded in hypothalamic slices. The allopregnanolone effect on IPSCs was inhibited by a G-protein antagonist as well as by blocking protein kinase C and, to a lesser extent, cAMP-dependent protein kinase activities. G-protein and protein kinase C activation in the absence of the neurosteroid had no effect on spontaneous IPSCs but enhanced the effect of subsequent allopregnanolone application. These findings together suggest that the neurosteroid modulation of GABA-mediated IPSCs requires G-protein and protein kinase activation, although not via a separate G-protein-coupled steroid receptor.

Voltage-gated currents distinguish parvocellular from magnocellular neurones in the rat hypothalamic paraventricular nucleus

1. Magnocellular and parvocellular neurones of the hypothalamic paraventricular nucleus (PVN) differentially regulate pituitary hormone secretion and autonomic output. Previous experiments have suggested that magnocellular, or type I neurones, and parvocellular, or type II neurones, of the PVN express different electrophysiological properties. Whole-cell patch-clamp recordings were performed in hypothalamic slices to identify the voltage-gated currents responsible for the electrophysiological differences between type I and type II PVN neurones. 2. Type I neurones, which display transient outward rectification and lack a low-threshold spike (LTS), generated a large A-type K+ current (IA) (mean +/- s.e. m.: 1127.5 +/- 126.4 pA; range: 250-3600 pA; voltage steps to -25 mV) but expressed little or no T-type Ca2+ current (IT). Type II neurones, which lack transient outward rectification but often display an LTS, expressed a smaller IA (360.1 +/- 56.3 pA; range: 40-1100 pA; voltage steps to -25 mV), and 75 % of the type II neurones generated an IT (-402.5 +/- 166.9 pA; range: -90 to -2200 pA; at peak). 3. The voltage dependence of IA was shifted to more negative values in type I neurones compared to type II neurones. Thus, the activation threshold (-53.5 +/- 0.9 and -46.1 +/- 2.6 mV), the half-activation potential (-25 +/- 1.9 and -17.9 +/- 2.0 mV), the half-inactivation potential (-80.4 +/- 9.3 and -67.2 +/- 3.0 mV), and the potential at which the current became fully inactivated (-57.4 +/- 2.1 and -49.8 +/- 1.5 mV) were more negative in type I neurones than in type II neurones, respectively. 4. IT in type II neurones activated at a threshold of -59.2 +/- 1.2 mV, peaked at -32. 6 +/- 1.7 mV, was half-inactivated at -66.9 +/- 2.2 mV, and was fully inactivated at -52.2 +/- 2.2 mV. 5. Both cell types expressed a delayed rectifier current with similar voltage dependence, although it was smaller in type I neurones (389.7 +/- 39.3 pA) than in type II neurones (586.4 +/- 76.0 pA). 6. In type I neurones IA was reduced by 41.1 +/- 7.0 % and the action potential delay caused by the transient outward rectification was reduced by 46.2 +/- 10.3 % in 5 mM 4-aminopyridine. In type II neurones IT was reduced by 66.8 +/- 10.9 % and the LTS was reduced by 76.7 +/- 7.8 % in 100 microM nickel chloride, but neither IT nor LTS was sensitive to 50 microM cadmium chloride. 7. Thus, differences in the electrophysiological properties between type I, putative magnocellular neurones and type II, putative parvocellular neurones of the PVN can be attributed to the differential expression of voltage-gated K+ and Ca2+ currents. This diversity of ion channel expression is likely to have profound effects on the response properties of these neurosecretory and non-neurosecretory neurones.

Noradrenergic excitation of magnocellular neurons in the rat hypothalamic paraventricular nucleus via intranuclear glutamatergic circuits

Noradrenergic projections to the hypothalamus play a critical role in the afferent control of oxytocin and vasopressin release. Recent evidence for intrahypothalamic glutamatergic circuits prompted us to test the hypothesis that the excitatory effect of noradrenergic inputs on oxytocin and vasopressin release is mediated in part by local glutamatergic interneurons. The voltage response to norepinephrine (30-300 microM) was tested with whole-cell recordings in putative magnocellular neurons of the paraventricular nucleus (PVN) in hypothalamic slices (400 micrometers). Norepinephrine elicited an alpha1 receptor-mediated direct depolarization in 23% of the magnocellular neurons tested; however, the most prominent response, seen in 42% of the magnocellular neurons, was an alpha1 receptor-mediated increase in the frequency of EPSPs. The norepinephrine-induced increase in EPSPs was blocked by tetrodotoxin and by ionotropic glutamate receptor antagonists, suggesting that norepinephrine excited presynaptic glutamate neurons to cause an increase in spike-mediated transmitter release. The increase in EPSPs also was observed in a surgically isolated PVN preparation (64% of cells) and with microdrop applications of norepinephrine (1 mM, 33% of cells) and glutamate (0.5-1 mM, 28%) in the PVN, indicating that the norepinephrine-sensitive presynaptic glutamate neurons are located within the PVN. Biocytin injection and subsequent immunohistochemical labeling revealed that both oxytocin and vasopressin neurons responded to norepinephrine. Our data indicate that magnocellular neurons of the PVN receive excitatory inputs from intranuclear glutamatergic neurons that express alpha1-adrenoreceptors. These glutamatergic interneurons may serve as an excitatory relay in the afferent noradrenergic control of oxytocin and vasopressin release under certain physiological conditions.

Modulation of multiple potassium currents by metabotropic glutamate receptors in neurons of the hypothalamic supraoptic nucleus

We studied the effects of activation of the metabotropic glutamate receptors on intrinsic currents of magnocellular n urons of the supraoptic nucleus (SON) with whole cell patch-clamp and conventional intracellular recordings in coronal slices (400 micron) of the rat hypothalamus. Trans-(+/-)-1-amino-1,3-cyclopentane dicarboxylic acid (trans-ACPD, 10-100 microM), a broad-spectrum metabotropic glutamate receptor agonist, evoked an inward current (18.7 +/- 3.45 pA) or a slow depolarization (7.35 +/- 4.73 mV) and a 10-30% decrease in whole cell conductance in approximately 50% of the magnocellular neurons recorded at resting membrane potential. The decrease in conductance and the inward current were caused largely by the attenuation of a resting potassium conductance because they were reduced by the replacement of intracellular potassium with an equimolar concentration of cesium or by the addition of potassium channel blockers to the extracellular medium. In some cells, trans-ACPD still elicited a small inward current after blockade of potassium currents, which was abolished by the calcium channel blocker, CdCl2. Trans-ACPD also reduced voltage-gated and Ca2+-activated K+ currents in these cells. Trans-ACPD reduced the transient outward current (IA) by 20-70% and/or the IA-mediated delay to spike generation in approximately 60% of magnocellular neurons tested. The cells that showed a reduction of IA generally also showed a 20-60% reduction in a voltage-gated, sustained outward current. Finally, trans-ACPD attenuated the Ca2+-dependent outward current responsible for the afterhyperpolarization (IAHP) in approximately 60% of cells tested. This often revealed an underlying inward current thought to be responsible for the depolarizing afterpotential seen in some magnocellular neurons. (RS)-3,5-dihydroxyphenylglycine, a group I receptor-selective agonist, mimicked the effects of trans-ACPD on the resting and voltage-gated K+ currents. (RS)-alpha-methyl-4-carboxyphenylglycine, a group I/II metabotropic glutamate receptor antagonist, blocked these effects. A group II receptor agonist, 2S,1'S,2'S-2carboxycyclopropylglycine and a group III receptor agonist, (+)-2-amino-4-phosphonobutyric acid, had no effect on the resting or voltage-gated K+ currents, indicating that the reduction of K+ currents was mediated by group I receptors. About 80% of the SON cells that were labeled immunohistochemically for vasopressin responded to metabotropic glutamate receptor activation, whereas only 33% of labeled oxytocin cells responded, suggesting that metabotropic receptors are expressed preferentially in vasopressinergic neurons. These data indicate that activation of the group I metabotropic glutamate receptors leads to an increase in the postsynaptic excitability of magnocellular neurons by blocking resting K+ currents as well as by reducing voltage-gated and Ca2+-activated K+ currents.

Physiological evidence for local excitatory synaptic circuits in the rat hypothalamus

We conducted whole cell voltage-clamp and current-clamp recordings in slices of rat hypothalamus to test for local excitatory synaptic circuits. Local excitatory inputs to neurons of the paraventricular nucleus (PVN) and supraoptic nucleus (SON) were studied with the use of electrical and chemical stimulation. Extracellular electrical stimulation provided indirect evidence of local excitatory circuits. Single stimuli evoked multiple excitatory postsynaptic potentials (EPSPs) or excitatory postsynaptic currents (EPSCs) in some PVN and SON cells, invoking polysynaptic excitatory inputs. Repetitive stimulation (10-20 Hz, 2-10 s) elicited long afterdischarges of EPSPs/EPSCs, suggesting a potentiation of upstream synapses in a polysynaptic circuit. Bath application of metabotropic glutamate receptor agonists provided more conclusive evidence for local excitatory circuits. Metabotropic receptor activation caused an increase in the frequency of EPSPs/EPSCs that was blocked by tetrodotoxin, suggesting that it was mediated by activation of local presynaptic excitatory neurons. The local excitatory inputs to SON and PVN neurons were mediated by glutamate release, because the EPSPs/EPSCs elicited with electrical stimulation and metabotropic receptor activation were blocked by ionotropic glutamate receptor antagonists. Finally, glutamate microstimulation furnished the most direct demonstration of local excitatory synaptic circuits. Glutamate microstimulation of perinuclear sites elicited an increase in the frequency of EPSPs/EPSCs in 13% of the PVN and SON neurons tested. Two sites provided most of the local excitatory synaptic inputs to PVN neurons, the dorsomedial hypothalamus and the perifornical region. These experiments provide converging physiological evidence for local excitatory synaptic inputs to hypothalamic neurons, inputs that may play a role in pulsatile hormone release.

Presynaptic modulation by metabotropic glutamate receptors of excitatory and inhibitory synaptic inputs to hypothalamic magnocellular neurons

The effects of activation of metabotropic glutamate receptors (mGluRs) on synaptic inputs to magnocellular neurons of the hypothalamic supraoptic nucleus (SON) were studied with the use of whole cell patch-clamp and microelectrode recordings in acute hypothalamic slices. Application of the mGluR agonist trans-(+/-)-1-amino-1,3-cyclopentane dicarboxylic acid (trans-ACPD, 100 microM) elicited an increase in the frequency of spontaneous excitatory postsynaptic potentials (EPSPs) and excitatory postsynaptic currents (EPSCs) in 20% of the cells, and of spontaneous inhibitory postsynaptic potentials (IPSPs) and inhibitory postsynaptic currents (IPSCs) in 50% of the cells tested in normal medium. The increased frequency of spontaneous EPSPs/EPSCs and IPSPs/IPSCs was blocked by tetrodotoxin (TTX), indicating that mGluRs act to excite the somata/dendrites of presynaptic glutamatergic and GABAergic neurons. (RS)-3,5-dihydroxyphenylglycine (50 microM), a selective group I receptor agonist, mimicked the presynaptic somatic/dendritic effects of trans-ACPD, suggesting that the presynaptic somatic/dendritic receptors responsible for increased spike-dependent glutamate and gamma-aminobutyric acid (GABA) release belong to the group I mGluRs. In the presence of TTX, trans-ACPD caused a decrease in the frequency of miniature EPSCs (up to 90%) in 13 of 16 cells, and a decrease in the frequency of miniature IPSCs (up to 80%) in 10 of 16 cells tested. Miniature EPSC and IPSC amplitudes usually did not change in trans-ACPD, suggesting that activation of metabotropic receptors located at presynaptic glutamatergic and GABAergic terminals led to a reduction in transmitter release onto SON magnocellular neurons. L(+)-2-amino-4-phosphonobutyric acid (100-250 microM), a selective group III receptor agonist, mimicked the effects of trans-ACPD at presynaptic terminals, decreasing the frequency of miniature EPSCs and IPSCs by up to 85% without affecting their amplitude. Thus the metabotropic receptors at presynaptic glutamate and GABA terminals in the SON belong to group III mGluRs. EPSCs evoked by electrical stimulation were enhanced by the group III receptor antagonist (S)-2-amino-2-methyl-4-phosphonobutanoic acid, suggesting that presynaptic metabotropic receptors are activated by the release of endogenous glutamate. These data indicate that mGluRs in the hypothalamus have opposing actions at presynaptic somata/dendrites and at presynaptic terminals. Activation of group I receptors (mGluR1 and/or mGluR5) on presynaptic somata/dendrites led to an increase in spike-dependent transmitter release, whereas activation of the group III receptors (mGluR4, 7, and/or 8) on presynaptic terminals suppressed glutamate and GABA release onto SON neurons. No differences were seen in the effects of mGluR activation between immunohistochemically identified oxytocin and vasopressin neurons of the SON.

Physiological mapping of local inhibitory inputs to the hypothalamic paraventricular nucleus

Local inhibitory synaptic inputs to neurons of the rat hypothalamic paraventricular nucleus (PVN) were studied by using glutamate microstimulation and conventional intracellular and whole-cell patch-clamp recording in coronal, horizontal, and parasagittal slices of rat hypothalamus. PVN cells were classified as magnocellular or parvocellular neurons on the basis of electrophysiological and post hoc immunohistochemical analyses; GABA-producing neurons were localized with in situ hybridization. Glutamate microstimulation of different sites around the PVN evoked volleys of postsynaptic potentials in 43% of the PVN neurons tested. Some responses to stimulation at each site were blocked by bicuculline, suggesting that they were mediated by the activation of presynaptic GABA neurons. In the coronal plane, presynaptic inhibitory sites were located lateral to the PVN and ventral to the fornix, corresponding to the lateral hypothalamic area and the posterior bed nucleus of the stria terminalis (BNST). In the horizontal plane, presynaptic inhibitory sites were found rostral, lateral, and caudal to the nucleus, corresponding to parts of the anterior hypothalamic area, the posterior BNST, the medial preoptic area, and the dorsomedial hypothalamus. In the parasagittal plane, presynaptic inhibitory neurons were revealed at sites rostral and caudal to the nucleus, corresponding to the medial preoptic area and the dorsomedial hypothalamus, and in a site dorsal to the optic chiasm that included the suprachiasmatic nucleus. These presynaptic sites each contained GABA-producing neurons based on in situ hybridization with a glutamic acid decarboxylase riboprobe and together formed a three-dimensional ring around the PVN. Unexpectedly, both magnocellular and parvocellular neurons received inhibitory synaptic inputs from common sites.

Electrical properties of neocortical neurons in slices from children with intractable epilepsy

1. The intrinsic electrical properties of human neocortical neurons were studied with current-clamp and single-electrode voltage-clamp techniques in slices obtained from children, aged 3 mo to 15 yr, undergoing surgical treatment of intractable epilepsy. Neocortical samples were classified as most or least abnormal based on clinical data. Recorded neurons were labeled with biocytin for correlation of electrical properties with morphological characteristics and laminar position. All recorded neurons were divided into three cell types--fast-spiking, low-threshold spiking (LTS) and non-LTS cells--on the basis of their electrical characteristics. 2. Fast-spiking cells generated brief, rapidly repolarizing action potentials. Most of these cells showed only weak spike-frequency adaptation. Fast-spiking cells labeled with biocytin were aspiny or sparsely spiny nonpyramidal neurons located in cortical layers 2-4. 3. LTS cells generated Ca(2+)-dependent low-threshold potentials and were the most numerous of the three cell types. Their Na(+)-dependent action potentials were broader than those of fast-spiking cells and showed marked spike-frequency adaptation. The size of low-threshold Ca2+ potentials and currents varied across cells, but they never supported more than two or, occasionally, three fast action potentials. LTS cells were pyramidal neurons located throughout cortical layers 2-6. Unlike the bursting neocortical cells described in lower mammals, LTS neurons in neocortex from children failed to generate bursts of inactivating Na+ action potentials. 4. Non-LTS cells also had relatively broad Na(+)-dependent action potentials and showed spike-frequency adaptation, but they did not generate detectable low-threshold potentials or currents. Non-LTS cells were also pyramidal neurons located throughout layers 2-6. 5. The electrical properties of cells from different age groups (< or = 1, 2-8, and 9-15 yr) and from most-abnormal and least-abnormal tissue samples were compared. A statistically significant trend toward a lower input resistance, a faster membrane time constant, and a decreased spike duration was observed with increasing age. There were no significant differences between the electrical properties of cells from the most-abnormal tissue and cells from the least-abnormal tissue. 6. These data indicate that the intrinsic electrical properties of neocortical neurons from children vary according to cell morphology and change with increasing age, as has been observed in rodent and feline neocortical neurons. No obvious evidence of epileptogenicity was detected in the intrinsic electrical properties of any of the neurons studied.

Neurophysiology of neocortical slices resected from children undergoing surgical treatment for epilepsy

The recent emergence of surgical treatment of childhood epilepsy has led to the accessibility of young human cerebral tissue for electrophysiological studies of the mechanisms involved in epileptogenesis. Intracellular recordings were obtained from neurons in slices prepared from neocortical tissue resected from children (3 months to 15 years) with catastrophic epilepsy. Data from 'least abnormal' versus 'most abnormal' tissue were compared; the evaluation of the degree of abnormality was based on several clinical criteria. Hypotheses concerning NMDA receptors, local synaptic circuits, and epileptiform bursts were tested. The NMDA receptor-mediated component of synaptic responses, which was isolated pharmacologically, had a voltage dependence that was functionally mature by 8-10 months of age and did not appear to be altered even in the most abnormal tissue. Local inhibitory and excitatory synaptic circuits were present as early as 11 months and 8 months, respectively. Local excitatory circuits were sufficiently extensive in young children to initiate and sustain epileptiform activity when synaptic inhibition was suppressed. Bicuculline-induced epileptiform bursts were similar to those in adult human or animal neocortical slices. Burst duration and the presence of after-discharges were unrelated to patient age or tissue abnormality. These data demonstrate that (1) the electrophysiological properties of human neocortical neurons are very similar to those observed in animal experiments, (2) the mechanisms of neuronal communication are qualitatively mature within the first year of life, and (3) synaptic transmission and local neuronal circuits appear qualitatively normal, even in the most abnormal tissue from children with catastrophic epilepsy.

Connections from the subfornical organ to the oxytocin and vasopressin systems in the lactating rat. A study using electrical stimulations, lesions and electrophysiological recordings

The medial septal area has been implicated in the control of the magnocellular neurosecretory cells of the hypothalamus, and in particular, in the regulation of neurons secreting oxytocin. The present study investigated the hypothesis that this medial septal pathway originates in the subfornical organ. Brief electrical stimulation of the subfornical organ or of the medial septum both evoked a transient rise in intramammary pressure equivalent to that caused by an i.v. injection of 1 mU oxytocin. The optimal frequency was 5-20 Hz for 5-10 s. Prolonged stimulation also elicited at its onset a single transient response, similar to that evoked by brief stimulation. Extracellular recordings were made from neurosecretory cells of the supraoptic nucleus identified by antidromic stimulation of the neural stalk and further classified as vasopressinergic and oxytocinergic by their reaction at the time of reflex milk ejection induced by suckling. Single-pulse stimulation of the subfornical organ rarely produced excitation, but short trains of stimuli evoked a large excitation in most oxytocinergic and vasopressinergic neurons. To delineate further the pathway from the subfornical organ to the magnocellular neurons, stimulations were combined with various lesions of the medial forebrain. The effects of stimulation of the subfornical organ were abolished after a section immediately rostral to the organ, and in most cases after lesion of the medial septum. Stimulation of the medial septum no longer had an effect after the subfornical organ had been lesioned a week prior to experiments, a period sufficient to allow degeneration of subfornical efferents. This study shows that the excitatory afferent input to the oxytocin and vasopressin-secreting neurons of the hypothalamus from the medial septal area originates in the subfornical organ. This input is not involved in the main afferent control of the milk ejection reflex since lesions of the subfornical organ and of the medial septum had no effect on the reflex. It is suggested, therefore, that the subfornical input to both oxytocin and vasopressin cells intervenes to facilitate synergistic action of both hormones in non-reproductive functions.

Local inhibitory synaptic inputs to neurones of the paraventricular nucleus in slices of rat hypothalamus

1. Intracellular recordings were obtained from neurones in the region of the paraventricular nucleus in slices of rat hypothalamus. Glutamate microdrops were applied to the surface of the slices at sites dorsal, lateral and ventral to the paraventricular nucleus to selectively activate local presynaptic neurones. The gamma-aminobutyric acidA (GABAA)-receptor antagonists picrotoxin or bicuculline were bath-applied to block synaptic inhibition. 2. Glutamate microapplication caused a tonic depolarization and often repetitive action potentials in twenty of forty-seven recorded cells. This was probably caused by the direct exposure of the dendrites of the recorded cells to the glutamate microdrops. 3. Glutamate microstimulation elicited inhibitory synaptic responses in nine of forty-seven neurones tested. Glutamate microdrops caused discrete, hyperpolarizing postsynaptic potentials (PSPs) in four cells recorded with microelectrodes containing potassium acetate and evoked depolarizing PSPs in four cells recorded with KCl-filled microelectrodes. Glutamate microapplication inhibited spontaneous spike firing in another cell recorded with a potassium acetate microelectrode. 4. Bath application of GABAA-receptor antagonists completely blocked the hyperpolarizing PSPs elicited by glutamate microstimulation in three of three cells recorded with potassium acetate electrodes and the depolarizing PSPs in two of two cells recorded with KCl electrodes, indicating they were inhibitory PSPs caused by the release of GABA. Suppression of GABAA-mediated synaptic inhibition did not reveal any glutamate-evoked excitatory PSPs. 5. Recorded cells were identified as magnocellular, parvocellular or non-paraventricular bursting neurones on the basis of their electrophysiological properties. Direct depolarization and local inhibitory synaptic responses were observed in all three cell types. 6. Several conclusions can be drawn from these data: (1) functional glutamate receptors are distributed throughout neuronal populations in the paraventricular region of the hypothalamus, confirming and extending previous observations; (2) local synaptic inputs to neurones in the paraventricular nucleus are primarily inhibitory, supplied by perinuclear GABAergic neurones; (3) both magnocellular and parvocellular subpopulations receive local inhibitory synaptic inputs. The possibility that these local GABAergic circuits mediate inhibitory inputs to paraventricular neurones from limbic structures is discussed.

Membrane properties of identified guinea-pig paraventricular neurons and their response to an opioid mu-receptor agonist: evidence for an increase in K+ conductance

Intracellular recordings were obtained from neurons in the paraventricular nucleus (PVN) of guinea-pig hypothalamic slices. Passive and active properties of the neurons were determined, and when possible, recorded neurons were injected with biocytin. The effects of a mu-receptor opioid agonist [D-Ala2, Nme-Phe4, Gly5-ol]enkephalin (DAGO) were studied in order to determine which types of cells have mu receptors and to test the hypothesis that an increase in K+ conductance causes mu-receptor-mediated inhibition in the PVN. The recorded cells inside the PVN were divided into two groups, primarily on the basis of the presence or absence of a low threshold Ca2+ spike (LTS). In one group of neurons, LTS potentials could not be evoked (non-LTS cells, n = 42). In another group of neurons (n = 35), LTS potentials with one or two Na+ spikes could be initiated with depolarizing pulses superimposed on steady hyperpolarizing currents; however, these neurons did not fire robust bursts (i.e. non-bursting LTS cells). The mean time constant of non-LTS cells (19.9 +/- 1.6 ms; mean +/- SEM) was significantly shorter than that of non-bursting LTS cells (26.7 +/- 2.1 ms). There were no differences in the mean resting membrane potential, spike amplitude, spike duration, input resistance, spike threshold and pattern of synaptic inputs between the two groups. Intracellular labeling with biocytin combined with cresyl violet counter-staining demonstrated that the two types of cells were located in the PVN.(ABSTRACT TRUNCATED AT 250 WORDS)

Local synaptic circuits and epileptiform activity in slices of neocortex from children with intractable epilepsy

1. Single and dual intracellular recordings were performed in neocortical slices obtained from tissue samples surgically removed from children (8 mo to 15 yr) for the treatment of intractable epilepsy. Electrical stimulation and glutamate microapplication were used to study local synaptic inputs to pyramidal cells. 2. In recordings with potassium-acetate electrodes, activation of presynaptic neocortical neurons with glutamate microdrops did not elicit a clear increase in postsynaptic potentials (PSPs) but did suppress current-evoked repetitive spike firing in recorded neurons. Bicuculline (10 microM) blocked this effect, suggesting it was caused by the activation of presynaptic gamma-aminobutyric acid (GABA) cells. In recordings with KCl electrodes, glutamate microdrops elicited an increase in the frequency and amplitude of depolarizing PSPs. Bicuculline (5-10 microM) blocked the glutamate-evoked PSPs, suggesting they were reversed GABAA-receptor-mediated inhibitory postsynaptic potentials (IPSPs). In one cell recorded with a KCl electrode (total n = 8), current-evoked spike trains elicited afterdischarges of reversed IPSPs, thus revealing a recurrent inhibitory circuit. Therefore local inhibitory synaptic circuits were robust and could be observed in tissue from patients as young as 11 mo. 3. In addition to short-latency (10-25 ms), monosynaptic excitatory postsynaptic potentials (EPSPs), electrical stimulation at low intensities sometimes elicited delayed EPSPs (20-60 ms). When GABAA-receptor-mediated synaptic inhibition was partially reduced in bicuculline (5-10 microM), electrical stimulation evoked large EPSPs at long and variable latencies (100-300 ms). Glutamate microapplication caused an increase in the frequency and amplitude of EPSPs; preliminary results suggest that glutamate microdrops were less likely to evoke EPSPs in tissue from younger patients (8-12 mo) than in slices from patients greater than 4 yr. Evidence for local excitatory synaptic circuits was thus found when synaptic inhibition was partially reduced. 4. After further reduction of inhibition in bicuculline (5-50 microM), electrical stimulation elicited epileptiform bursts. In pairs of simultaneously recorded neurons, bursts were generated synchronously from long-latency EPSPs (100-300 ms) in slices from patients as young as 8 mo. Reflected EPSPs at very long and variable latencies (500-1,100 ms) and repetitive epileptiform bursts could be evoked synchronously in pairs of cells. Glutamate activation of local presynaptic neurons elicited robust epileptiform events in recorded cells. This was seen in slices from patients as young as 16 mo. 5. These data provide physiological evidence for the presence of local inhibitory and excitatory synaptic circuits in human neocortex at least as early as 11 and 8 mo, respectively.(ABSTRACT TRUNCATED AT 400 WORDS)

In vivo Intracellular Recording of Neurons in the Supraoptic Nucleus of the Rat Hypothalamus

Abstract Intracellular recordings were made from cells in the hypothalamic supraoptic nucleus in the urethane-anaesthetized male rat using the ventral surgical approach. Impalements lasted from 5 min to 1 h and recorded cells had an input resistance of 55 to 170 megohms. Spikes of over 50 mV were recorded from 14 cells which could be antidromically activated by stimulation of the neural stalk. The spikes showed a hyperpolarizing afterpotential and the broadening characteristic of rapidly firing magnocellular neurons, which recovered rapidly (<200 ms). When depolarized, the cells showed evidence of a transient potassium current. Recurrent synaptic coupling between the recorded cell and adjacent cells would be expected to alter the hyperpolarizing afterpotential of an antidromic spike as compared with a spontaneous spike; no perceptible difference in the waveforms of the different types of spike could be detected in 11 spontaneously active cells. Application of just subthreshold stimuli to the neural stalk did not evoke depolarizing or hyperpolarizing potentials. Suprathreshold shocks to the neural stalk, when the antidromic spike was prevented by collision, also had no discernible effect on membrane potential. Thus intracellular recordings from magnocellular neurons in vivo revealed electrophysiological properties similar to those seen in vitro. No evidence for synaptic interconnection between magnocellular neurons was found in male rats.

Comparison of three intracellular markers for combined electrophysiological, morphological and immunohistochemical analyses

Hypothalamic paraventricular and supraoptic neurons were recorded intracellularly in coronal slices and injected with Lucifer yellow, ethidium bromide or biocytin. Electrical properties, morphological staining and neurophysin immunohistochemistry were compared among the 3 markers. Lucifer yellow electrodes had a high resistance and frequently blocked during experiments. Neurons recorded with Lucifer yellow electrodes had low input resistances and low-amplitude, broad spikes. Lucifer yellow labeling in whole mount was highly fluorescent, revealing distal dendrites and axons. Of cells injected with Lucifer yellow, 64% were recovered but were faint after immunohistochemical processing. Recordings with ethidium bromide electrodes were similar to controls, although electrode blockage sometimes occurred. Only somata and proximal dendrites of ethidium bromide-filled neurons were visible in whole-mount. Forty percent of cells injected with ethidium bromide were recovered after immunohistochemical processing; these were invariably faint. Recordings with biocytin-filled electrodes were similar to control recordings. Biocytin-filled, HRP-labeled cells showed distal dendrites and often dendritic spines and axons in 50-75-microns sections. Seventy percent of biocytin-injected cells labeled with fluorescent markers were recovered and remained strongly labeled after immunohistochemical processing. Biocytin had the best electrical and staining properties for combined electrophysiological and anatomical studies.

Immunohistochemical differentiation of electrophysiologically defined neuronal populations in the region of the rat hypothalamic paraventricular nucleus

Intracellular recording and labeling were combined with neurophysin immunohistochemistry to study neurons in the paraventricular nucleus region of the rat hypothalamus. Neuronal membrane properties were examined in hypothalamic slices, and cells were labeled by injecting biocytin or Lucifer yellow. Slices were then embedded, sectioned, and immunohistochemically processed for neurophysin. Immunoreactivity patterns, and in some cases counterstaining, enabled determinations of the cytoarchitectonic positions of recorded cells to be made. Recorded cells were divided into three types according to their electrophysiological characteristics. The first type lacked low-threshold Ca2+ spikes and displayed linear current-voltage relations, a short time constant, and evidence for an A current. These were relatively large cells that were typically immunoreactive for neurophysin and were situated near other neurophysin-positive neurons. The second type had relatively small low-threshold potentials that did not generate bursts of Na+ spikes. These cells had heterogeneous current-voltage relations and intermediate time constants. They did not label for neurophysin, and most were located in the parvicellular subregion of the paraventricular nucleus. The third type had large low-threshold Ca2- spikes that generated bursts of Na+ spikes, and these cells had nonlinear current-voltage relations and long time constants. These neurons were dorsal or dorsolateral to the paraventricular nucleus and were not immunoreactive for neurophysin. These results indicate that paraventricular magnocellular neurons lack low-threshold potentials, whereas paraventricular parvicellular neurons display low-threshold potentials that generate one or two action potentials. Neurons that fire spike bursts from low-threshold potentials are adjacent to the paraventricular nucleus, confirming earlier reports.

Electrophysiological properties of neurones in the region of the paraventricular nucleus in slices of rat hypothalamus

1. Neurones in the region of the hypothalamic paraventricular nucleus (PVN) of the rat were studied with intracellular recording in the coronal slice preparation. Three types of hypothalamic neurones were distinguished according to their membrane properties and anatomical positions. Lucifer Yellow or ethidium bromide was injected intracellularly to determine the morphology of some recorded cells. 2. The most distinctive electrophysiological characteristic was the low-threshold depolarizing potentials which were totally absent in type I neurones, present but variable in type II neurones and very conspicuous in type III neurones. Type II neurones generally showed relatively small low-threshold depolarizations (26.5 +/- 2.2 mV) which generated at most one to two action potentials. Type III neurones, on the other hand, generated large low-threshold potentials (40.3 +/- 2.8 mV) which gave rise to bursts of three to six fast action potentials. Deinactivation of the low-threshold conductance in both type II and type III neurones was voltage- and time-dependent. Low-threshold potentials persisted in TTX (1-3 microM) but were blocked by solutions containing low Ca2+ (0.2 mM) and Cd2+ (0.5 mM), suggesting they were Ca(2+)-dependent. 3. Type I neurones had a significantly shorter membrane time constant (14.5 +/- 1.7 ms) than those of type II (21.6 +/- 1.7 ms) and type III neurones (33.8 +/- 4.4 ms). Input resistance and resting membrane potential did not differ significantly among the cell groups. 4. Current-voltage (I-V) relations were significantly different among the three cell types. Type I neurones had linear I-V relations to -120 mV, while type III neurones all showed marked anomalous rectification. I-V relations among type II neurones were more heterogeneous, although most (75%) had linear I-V curves to about -90 to -100 mV, inward rectification appearing at more negative potentials. 5. Type I neurones generated fast action potentials of relatively large amplitude (64.2 +/- 1.1 mV, threshold to peak) and long duration (1.1 +/- 0.1 ms, measured at half-amplitude); the longer duration was due to a shoulder on the falling phase of the spike. Type II neurones had large spikes (66.5 +/- 1.6 mV) of shorter duration (0.9 +/- 0.1 ms) with no shoulder. Type III neurones had relatively small spikes (56.1 +/- 2.2 mV) of short duration (0.8 +/- 0.1 ms) with no shoulder. 6. The three cell populations showed different patterns of repetitive firing in response to depolarizing current pulses. Type I neurones often generated spike trains with a delayed onset and little spike-frequency adaptation.(ABSTRACT TRUNCATED AT 400 WORDS)

Electrophysiology of GABA-mediated synaptic transmission and possible roles in epilepsy

Epileptogenic conditions come about from a disequilibrium between excitatory and inhibitory mechanisms, creating a state of neuronal hypersynchrony. From experimental studies in animal models of epilepsy it appears that several mechanisms, alone or in combination, could be responsible for this imbalance. An alteration of GABA-mediated inhibition has long been considered to be one of the most likely candidates. We review recent data on the synaptic physiology of GABA-mediated inhibition, with emphasis on GABAA and GABAB receptors and their conductances. We describe the integrative role of GABAergic local-circuit neurons in the normal control of recurrent excitation. We then discuss possible alterations in GABAA-mediated inhibition in two chronic animal models of epilepsy, the kindled rat and the kainate-treated rat. Finally, we review studies on GABA inhibition in human epileptic cortex resected for the treatment of intractable epilepsy.

Osmolality-induced changes in extracellular volume alter epileptiform bursts independent of chemical synapses in the rat: importance of non-synaptic mechanisms in hippocampal epileptogenesis

The contribution of non-synaptic mechanisms to the seizure susceptibility of rat CA1 hippocampal pyramidal cells was examined in vitro by testing the effects of osmolality on synchronous neuronal activity, using solutions which blocked chemical synaptic transmission both pre- and post-synaptically. Decreases in osmolality, which shrink the extracellular volume, caused or enhanced epileptiform bursting. Increases in osmolality with membrane-impermeant solutes, which expand the extracellular volume, blocked or greatly reduced epileptiform discharges. Reductions in the extracellular volume, therefore, can enhance synchronization among CA1 hippocampal neurons through non-synaptic mechanisms. Since similar osmotic treatments are known to modify epileptiform discharges in several models of epilepsy, non-synaptic mechanisms are probably more important in hippocampal epileptogenesis than previously realized and may contribute to the high susceptibility of this brain region to epileptic seizures in animals and humans. These data also provide a possible explanation for the observation in humans that decreased plasma osmolality, which can be associated with a wide range of clinical syndromes, leads to seizures.

Synaptic transmission in human neocortex removed for treatment of intractable epilepsy in children

Synaptic transmission to pyramidal cells was studied in slices of neocortex resected from infants and children (n = 10, age 8 months to 13 years) undergoing surgical treatment for intractable epilepsy. Most specimens were from the least abnormal area of the resection. Stable intracellular recordings could be obtained for up to 8 hours. Most of the recorded neurons had electrophysiological characteristics similar to those of regular-firing pyramidal cells and were in layers III to V, which was confirmed by intracellular staining with Lucifer yellow. Local extracellular stimulation evoked a sequence of excitatory and inhibitory postsynaptic potentials. After application of the gamma-aminobutyric acid antagonist, bicuculline (10-30 microM), extracellular stimulation induced large excitatory postsynaptic potentials and epileptiform bursts. Spontaneous bursts occasionally occurred in bicuculline. This effect of bicuculline was observed in all the tissue samples, even those from infant patients (n = 4, age 8-16 months). Kynurenic acid depressed or abolished both spontaneous and stimulation-induced bursts. The competitive antagonist for N-methyl-D-aspartate receptors, DL-2-amino-5-phosphonopentanoic acid decreased the duration of bicuculline-induced bursts. These data provide evidence that, similar to rat and cat neocortex, excitatory and inhibitory amino acids are important transmitters to pyramidal cells in immature human neocortex.

Intrinsic and synaptic mechanisms of hypothalamic neurons studied with slice and explant preparations

The use of slice and explant preparations has allowed major advances in our understanding of the membrane physiology of mammalian hypothalamic neurons. This article will review intracellular electrophysiological studies of neurons in or immediately surrounding the supraoptic and paraventricular nuclei. Considerable information is now available on the intrinsic membrane mechanisms that control action potential generation and burst firing in magnocellular neuroendocrine cells (MNCs) within these nuclei. Neurons surrounding the paraventricular nucleus have different electrical properties than the MNCs, including low-threshold Ca2+ spikes and pronounced anomalous rectification. Bicuculline and kynurenic acid strongly depress fast IPSPs and EPSPs in MNCs, thus suggesting that inhibitory and excitatory amino acids mediate fast synaptic transmission in the hypothalamus. The effects of neuromodulators, such as noradrenaline and opioid peptides, have also been examined. Noradrenaline excites supraoptic neurons and leads to phasic firing through an alpha-1 mechanism and decreased K+-conductance. Opioid peptides act directly on mu-receptors to hyperpolarize about half of the neurons through an increased K+-conductance. In conclusion, using the magnocellular neuroendocrine system as a model, in vitro slice and explant preparations have allowed the characterization of electrophysiological properties, the identification of neurotransmitters for synaptic events, and studies on the mechanism of action of neuromodulators.

The effects of neonatal capsaicin treatment on the sensory innervation of the nipple and on the milk ejection reflex in the rat

Much of the sensory innervation of the nipple is provided by fibers of small calibre (A delta and C). In order to determine the contribution of unmyelinated C-fibers to this innervation and to the physiology of lactation, mammary afferents were studied in rats that had succeeded in lactating after neonatal treatment with capsaicin, a neurotoxin which selectively destroys C-fibers. After subcutaneous injection of horseradish peroxidase-wheat germ agglutinin (HRP-WGA) into the nipple of capsaicin-treated lactating rats, cell counts and surface area estimates of peroxidase-labelled and unlabelled cells were made in the corresponding dorsal root ganglia (DRG) and compared to values obtained in untreated lactating females that had received similar tracer injections. The segmental distribution of HRP-labelled primary sensory neurons in the capsaicin-treated rats was similar to that in untreated controls, but the number of labelled cells was significantly reduced at each segmental level. This reduction reflected a marked decrease in C-fibers, since there was a striking reduction in the number of small HRP-labelled and unlabelled cell bodies in the DRG and unmyelinated fibers in the dorsal roots. Peroxidase labelling within the dorsal horn of capsaicin-treated rats was also substantially diminished. About 40% of the females that had been treated neonatally with capsaicin gave birth and lactated. Although the average weight gain of their litters was retarded with respect to that of litters of untreated controls, the milk ejection reflex appeared to function normally.(ABSTRACT TRUNCATED AT 250 WORDS)

Afferent projections from the mammary glands to the spinal cord in the lactating rat--I. A neuroanatomical study using the transganglionic transport of horseradish peroxidase-wheatgerm agglutinin

Horseradish peroxidase-wheatgerm agglutinin was injected subcutaneously into one or more nipples of lactating rats to determine the spinal organization of sensory afferents emanating from the mammary glands. After survival periods of 45-96 h, dorsal root ganglia and segments of the spinal cord and/or medulla oblongata were sectioned and reacted histochemically with tetramethylbenzidine to reveal the transganglionically transported tracer. For each nipple injected, the peroxidase reaction product was found in somata, ranging in diameter from 15 to 60 microns, and fibres in 5-11 contiguous dorsal root ganglia. The number of labelled profiles was highest in the 2-4 central-most ganglia of the series and generally decreased progressively rostrally and caudally. After separate injections into each of the six ipsilateral nipples, labelling occurred in all ipsilateral dorsal root ganglia between the 5th cervical and 6th lumbar spinal segments. Substantial overlap of the spinal projections from adjacent mammary glands was seen, a given dorsal root ganglion innervating 2-3 different glands. Label in the spinal cord was restricted to the medial portion of the superficial dorsal horn. It occurred in what appeared to be terminal fields and fibres essentially in the substantia gelatinosa, but was also seen to extend into the marginal zone and sometimes into deeper regions of the dorsal horn. Label was found in both the gracile and cuneate nuclei of the medulla oblongata, though only occasionally and then only very sparsely. The substantial spread and segmental overlap of labelled mammary afferents, and the fact that most labelled afferents terminated in the dorsal horn, suggest that this spinal region may be an important site for the integration of sensory input from the mammary glands that may play a role in the sensory induction of reflex milk ejection.

Recurrent mammary gland contractions induced by a low tonic release of oxytocin in rats

In urethane-anaesthetized lactating rats, intramammary pressure occasionally displayed recurrent variations or oscillations having a slow rise time, low amplitude, long duration and a periodicity of 1-4 min. These oscillations differed from changes in intramammary pressure characteristic of reflex milk ejections induced by suckling, and were also observed in unsuckled rats. They were suppressed by lesions of the pituitary stalk or by stimulating the septum, a structure that inhibits the activity of the magnocellular system. They could be induced by long-term low frequency stimulation of the pituitary stalk, lesions of the septum or long-term infusions of oxytocin at a low rate of 0.05-0.3 mu./min. We suggest that the recurrent oscillations in intramammary pressure constitute a particular mode of response of the mammary gland to a tonic release of oxytocin resulting from a moderate but sustained increase in the basal level of electrical activity of the oxytocin-secreting neurones.