Disruption of maternal behaviour by acute conspecific interaction induces selective activation of the lateral periaqueductal grey

Maternal behaviour is sensitive to stress and opioidergic activation. The periaqueductal grey (PAG) is involved in coping strategies to stress, whereas morphine inhibition of maternal behaviour depends on the activation of the PAG. The aim of this study was to investigate whether the PAG is activated by disrupting maternal behaviour. Lactating Wistar rats were assigned to four groups: C (control); E1 (acute exposure to a male rat); E2 (daily 2-h exposure to another lactating female and a male rat from Day 3 to 6 of lactation); and E1 + 2 (treated first as E2 and, on Day 9, as E1). Maternal behaviour was recorded on Day 9 of lactation and analysed for 1 h. The E1 group spent more time retrieving their pups, took longer to initiate nursing, had shorter nursing bouts and spent more time in non-maternal activities compared with control. Rats submitted to E2 or E1 + 2 did not differ from the control. In another experiment, lactating rats were treated as above, except that 90 min after the end of the observation period the rats were killed and their brains were processed for immunohistochemical detection of Fos protein in the PAG. Fos increased in the lateral PAG only in the E1 group. We also observed that neurons activated by acute conspecific interaction in the PAG could be responsible for an opioid-dependent decrease in maternal behaviour as this effect was reversed by a microinjection of naltrexone, nor-binaltorphimine or naloxonazine into the lateral PAG. Chronic conspecific interaction alters the way this circuitry responds to acute conspecific interaction.     

Systemic morphine reduces GABA release in the lateral but not the medial portion of the midbrain periaqueductal gray of the rat

Neuroanatomical, electrophysiological and pharmacological studies have provided indirect evidence indicating that GABAergic neurons play a key role in opiate analgesia mediated by the midbrain periaqueductal gray (PAG) and ventromedial medulla. Although these studies suggest that systemic administration of opiates inhibits GABA release in the PAG, there have been no investigations to date that have directly examined this issue. The present study was thus designed to determine whether systemic morphine injection inhibits GABA release in the PAG of awake, freely moving rats using in vivo microdialysis and subsequent HPLC analysis. Extracellular levels of GABA, glutamate, aspartate, glycine, homocysteic acid and taurine were monitored with the microdialysis technique in either the lateral or medial portion of the ventrocaudal PAG in unanesthetized, unrestrained rats. Amino acid release was induced by infusing veratridine (75 microM, a sodium channel activator) directly through the dialysis probe. The effect of veratridine alone and the effect of veratridine in the presence of systemic morphine on the concentrations of amino acids in the PAG dialysate were determined. There were no significant differences in the basal concentrations of GABA, taurine, aspartate, glutamate, homocysteic acid and glycine between dialysates collected from the medial versus the lateral ventrocaudal PAG. Glycine, taurine and glutamate were present in the highest concentrations in dialysis samples both before and after treatment with veratridine, whereas GABA, homocysteic acid and aspartate were present in the lowest concentrations. Perfusion of veratridine into the ventrocaudal PAG resulted in significant elevation of all amino acids investigated. Except for taurine, no significant difference in veratridine-induced release between the lateral and medial PAG was observed. Tetrodotoxin (TTX) significantly blocked veratridine-induced release of GABA, aspartate, glutamate, glycine and taurine but not homocysteic acid. When rats were injected with morphine (10 mg/kg i.p.), veratridine-induced release of GABA was selectively and significantly decreased in the lateral but not the medial PAG as compared to control rats injected with saline followed by veratridine perfusion. Systemic injection of morphine or saline caused no significant change in the basal concentration of amino acids in PAG dialysate samples. These findings are consistent with the proposed mechanism of action of morphine in the lateral ventrocaudal PAG and offer the first direct evidence that systemic opiates decrease GABA release in this midbrain region.     

A re-evaluation of the role of superior colliculus in turning behaviour

There is much debate on the role of the superior colliculus (SC) in turning behaviour. In order to clarify this issue, unilateral kainate lesions were made by infusing 0.25 μg of kainate at two different anterior planes (0.8 mm apart), in the lateral or in the medial aspects of the deep collicular layers (DLSC), in the dorsal mesencephalic reticular formation (MRF), or in the lateral periaqueductal grey (PAG), both in normal rats and in rats made unilaterally supersensitive to DA-receptor agonists by unilateral infusion of 6-OHDA in the rostral substantia nigra. The effect of kainate lesions on spontaneous and apomorphine-induced motor behaviour was studied. In normal rats, unilateral kainate lesions of lateral DLSC or dorsal MRF resulted in short-lasting, spontaneous ipsiversive turning and persistant ipsiversive circling after peripheral administration of apomorphine. In 6-OHDA rats, kainate lesions of lateral DLSC or of dorsal MRF ipsilateral to 6-OHDA denervation reduced or even reversed the contralateral circling normally elicited in these rats by peripheral administration of apomorphine. Lesions of dorsal MRF, when compared with lesions of lateral DLSC, were more effective in producing these changes. Kainate lesions restricted to medial DLSC or to the PAG failed to elicit motor asymmetries in normal rats or to significantly modify the intensity of contralateral turning in 6-OHDA rats. These results clearly indicate that the SC plays an important role in turning behaviour. Failure of previous studies to reach this conclusion probably derives from inadequate localization of collicular lesions and from the use of bilateral lesions.     

Periaqueductal gray matter projections to midline and intralaminar thalamic nuclei of the rat

The periaqueductal gray matter (PAG) projections to the intralaminar and midline thalamic nuclei were examined in rats. Phaseolus vulgaris-leucoagglutinin (PHA-L) was injected in discrete regions of the PAG, and axonal labeling was examined in the thalamus. PHA-L was also placed into the dorsal raphe nuclei or nucleus of Darkschewitsch and interstitial nucleus of Cajal as controls. In a separate group of rats, the retrograde tracer cholera toxin beta-subunit (CTb) was injected into one of the intralaminar thalamic nuclei-lateral parafascicular, medial parafascicular, central lateral (CL), paracentral (PC), or central medial nucleus-or one of the midline thalamic nuclei-paraventricular (PVT), intermediodorsal (IMD), mediodorsal, paratenial, rhomboid (Rh), reuniens (Re), or caudal ventral medial (VMc) nucleus. The distribution of CTb labeled neurons in the PAG was then mapped. All PAG regions (the four columns of the caudal two-thirds of the PAG plus rostral PAG) and the precommissural nucleus projected to the rostral PVT, IMD, and CL. The ventrolateral, lateral, and rostral PAG provided additional inputs to most of the other intralaminar and midline thalamic nuclei. PAG inputs to the VMc originated from the rostral and ventrolateral PAG areas. In addition, the lateral and rostral PAG projected to the zona incerta. No evidence was found for a PAG input to the ventroposterior lateral parvicellular, ventroposterior medial parvicellular, caudal PC, oval paracentral, and reticular thalamic nuclei. PAG --> thalamic circuits may modulate autonomic-, nociceptive-, and behavior-related forebrain circuits associated with defense and emotional responses.     

Neurohistological and behavioral evidence for lordosis-inhibiting tract from lateral septum to periaqueductal gray in male rats

To verify the anatomical and functional connection of the lateral septum (LS) and periaqueductal gray (PAG) in inhibiting female sexual behavior, lordosis, in male rats, retrograde (Fluoro-Gold, FG) or anterograde (Phaseolus vulgaris-leucoagglutinin, PHA-L) tracer was injected into the PAG or LS on the right side, respectively, and FG-labeled cells or PHA-L-labeled axons in the forebrain and mesencephalon were determined in estrogen-treated castrated male rats. A ventral cut (VC) of the septum and a behavioral test were also conducted in some FG-injected rats. Furthermore, lordosis behavior was observed after chemical destruction of the septum by ibotenate. As a result, the lordosis quotient (LQ) in VC males was higher than that in control males without VC. FG-labeled neuronal cell bodies were found in the ipsilateral intermediate part of the LS in the control males but not in this area of the VC males. When neuronal cells in the intermediate part of the bilateral LS were completely destroyed by ibotenate, the LQ was higher than that in sham-lesioned male rats. These results suggest that a direct neural connection of the intermediate LS to the PAG has an inhibitory role in regulating lordosis in male rats. In addition, neuronal cell bodies in the intermediate LS exert an inhibitory influence. In the PHA-L experiment, labeled axons were seen in the ventral part of the LS, the medial forebrain bundle at the chiasmatic level, the lateral hypothalamus, the median region of the mesencephalon, and the rostral PAG in the side ipsilateral to the tracer injection site of the LS. Thus, these areas are thought to be involved in the pathway for lordosis-inhibition from the intermediate part of the LS to the PAG in male rats.     

Increased c-Fos expression in select lateral parabrachial subnuclei following chemical versus electrical stimulation of the dorsal periaqueductal gray in rats

The parabrachial nucleus (PBN) is located in the rostral dorsolateral pons and has been identified as a critical relay for cardiovascular responses (sympathoexcitation and baroreflex attenuation) evoked by the dorsal periaqueductal gray (PAG). We examined the pattern of c-Fos protein immunoreactivity throughout the rostral-caudal extent of the PBN in four groups of anesthetized male Sprague-Dawley rats to identify the specific PBN regions activated by dorsal PAG stimulation. Both electrical stimulation and chemical (0.3 mM bicuculline methobromide) activation of the dorsal PAG elicited a selective increase in Fos-like immunoreactivity (FLI) in the superior lateral and central lateral subnuclei of the rostral lateral PBN (LPBN) relative to surgery and blood pressure control groups. In the middle LPBN chemical stimulation of the dorsal PAG selectively increased FLI in the central lateral subnucleus while electrical stimulation increased FLI in the Kolliker-Fuse area only. Finally, in the caudal LPBN only electrical stimulation of the dorsal PAG induced significant changes in FLI above control. Significant changes in FLI in the medial PBN were not observed under any experimental conditions. These results confirm neuroanatomical data demonstrating that neurons in superior lateral and central lateral subnuclei of the rostral and middle LPBN are the primary targets of the dorsal PAG. Our results also demonstrate that this descending projection to the central lateral and superior lateral subnuclei of the LPBN is in part excitatory. Finally, our results raise the possibility that neurons in the central lateral subnucleus of the middle and rostral LPBN are integrally involved in descending modulation of sympathetic drive associated with dorsal PAG activation.     

Disruption of maternal behaviour in virgin and postparturient rats following sagittal plane knife cuts in the preoptic area-hypothalamus

Parasagittal knife cuts along the medial preoptic area-medial anterior hypothalamus were placed at either the medial (near lateral or NL cuts) or lateral (far lateral or FL cuts) border of the medial forebrain bundle in separate groups of virgin (Experiment 1) and postparturient (Experiment 2) rats. FL cuts were placed so as to spare preoptic-hypothalamic efferent connections with the medial forebrain bundle. NL and FL knife cuts were equally effective in preventing the induction of maternal behaviour produced by repeated exposure of virgin rats to foster pups. Both types of cuts also reduced nest building in virgins. In postparturient rats, NL and FL cuts both abolished pup retrieval and reduced nursing behaviour. However, only NL cuts disrupted lactation and nest building. NL cuts also produced chronic hyperthermia. The deficits in maternal behaviour and nest building may be independent of this hyperthermia, as shaving the fur in animals with NL cuts reduced body temperatures to control levels but did not restore these behaviours. These findings indicate that while preoptic/hypothalamic connections through the medial forebrain bundle are important for nest building and possibly lactation, other lateral connections must also be important for pup retrieval and nursing behaviour. The identity of these connections remains to be determined.     

Tolerance to non-opioid analgesics in PAG involves unresponsiveness of medullary pain-modulating neurons in male rats

Opiate analgesia can be hampered by a reduction in pharmacological effectiveness (tolerance), and this crucially depends on the periaqueductal gray matter (PAG). Non-opioids like metamizol (dipyrone) or aspirin also induce PAG-dependent analgesia and tolerance, but the neuronal bases of this tolerance are unknown. Metamizol is a pyrazolon derivative and cyclooxygenase inhibitor with widespread use as an analgesic in Europe and Latin America. Metamizol was microinjected into the PAG of awake male rats, and antinociception was assessed by the tail flick (TF) and hot plate (HP) tests. Microinjection twice daily for 2.5 days caused tolerance to metamizol. The rats were then anesthetized and recordings from pain-facilitating on-cells and pain-inhibiting off-cells of the rostral ventromedial medulla (RVM) were performed. PAG microinjection of morphine or metamizol depresses on-cells, activates off-cells and thus inhibits nociception, including TF and HP. In metamizol-tolerant rats, however, PAG microinjection of metamizol failed to affect on- or off-cells, and this is interpreted as the reason for tolerance. In metamizol-tolerant rats morphine microinjection into PAG also failed to affect RVM neurons or nociception (cross-tolerance). In naïve, non-tolerant rats the antinociceptive effect of PAG-microinjected metamizol or morphine was blocked when CTOP, a μ-opioid antagonist, was previously microinjected into the same PAG site. These results emphasize a close relationship between opioid and non-opioid analgesic mechanisms in the PAG and show that, like morphine, tolerance to metamizol involves a failure of on- and off-cells to, respectively, disfacilitate and inhibit nociception. Cross-tolerance between non-opioid and opioid analgesics should be important in the clinical setting.     

Reciprocal connections between the medial preoptic area and the midbrain periaqueductal gray in rat: a WGA-HRP and PHA-L study.

The midbrain periaqueductal gray (PAG) participates in diverse functions such as analgesia, autonomic regulation, sexual behavior, and defense/escape responses. Anatomical studies of the circuits involved in such functions have largely focused on the connections of PAG with the medulla. Projections to PAG from forebrain structures are extensive, but their organization has received little attention. Previous anatomic studies indicate that the medial preoptic area (MPO), involved in a variety of physiological and behavioral functions, is a major source of afferent input to the periaqueductal gray. Here, we have examined the topography of reciprocal connections between these two structures in the rat by using wheat germ agglutinin conjugated horseradish peroxidase (WGA-HRP) and Phaseolus vulgaris leucoagglutinin (PHA-L). Multiple WGA-HRP injections at several rostrocaudal levels of PAG retrogradely labeled large numbers of neurons in the medial preoptic area; labeled cells were primarily located in the medial preoptic nucleus, the median preoptic nucleus, and the region lateral to the medial preoptic nucleus. The distribution of labeled cells shifted medially to laterally along the rostral to caudal axis of the medial preoptic area. Rostrally, there was selective retrograde labeling in the central and lateral divisions of medial preoptic nucleus, whereas caudally, labeled cells were primarily located only in the lateral subdivision of medial preoptic nucleus. Tracer injections in PAG also produced strong anterograde labeling in MPO. WGA-HRP and PHA-L injections in the medial preoptic area resulted in dense anterograde labeling along the entire rostrocaudal axis of PAG. The terminal labeling in PAG from the medial preoptic area was not uniformly distributed throughout PAG, however. Instead, this projection formed one or two rostrocaudally oriented longitudinal columns that terminated in different subregions of PAG along the entire rostrocaudal axis of this structure. Rostrally, inputs from the medial preoptic area project heavily to dorsomedial PAG, and at mid-PAG levels, the projection becomes distinctly bipartite with two discrete longitudinal terminal columns in dorsomedial and lateral PAG; caudally, the heaviest labeling is in ventrolateral PAG. The projection also exhibited a central to peripheral (radial) gradient; labelled fibers and terminals were heaviest near the aqueduct and much lower in the peripheral parts of PAG. WGA-HRP injections in MPO also produced retrograde labeling of neurons at all rostrocaudal levels of PAG; more neurons were labeled in the rostral than the caudal half of PAG. The majority of labeled cells were located in dorsomedial and ventral/ventrolateral parts of PAG; only a few neurons in the dorsal raphe region appear to project to MPO.(ABSTRACT TRUNCATED AT 400 WORDS)     

Differential effects of NK1 receptors in the midbrain periaqueductal gray upon defensive rage and predatory attack in the cat

This study utilized anatomical and behavioral-pharmacological methods to determine the role of NK(1)-Substance P receptors in the midbrain periaqueductal gray (PAG) in defensive rage behavior in cats. For behavioral pharmacological experiments, monopolar stimulating electrodes were implanted in the medial hypothalamus for elicitation of defensive rage behavior and cannula-electrodes were implanted in the PAG for microinjections of receptor compounds. Microinjections of the NMDA antagonist, AP-7 (2 nmol), into the dorsal PAG blocked defensive rage elicited by medial hypothalamic stimulation, thus establishing the PAG as a synaptic region that receives hypothalamic inputs linked to defensive rage behavior. Microinjections of the NK(1) agonist, GR73632, into the same injection sites facilitated defensive rage in a dose-dependent manner, and also induced spontaneous hissing in five cats. The effects of GR73632 were reduced by pretreatment of the PAG with the NK(1) antagonist, GR82334 (16 nmol), microinjected into the same sites. Microinjections of GR73632 (8 nmol) into the PAG also suppressed predatory attack elicited by stimulation of the lateral hypothalamus. Immunohistochemical methods utilized to detect Substance P and Fos immunoreactivity revealed that neurons in the PAG activated after defensive rage-inducing medial hypothalamic stimulation lie in the same region as Substance-P-immunoreactive processes. Fos immunoreactivity was highest in the dorsomedial aspect of the rostral PAG after medial hypothalamic stimulation. Cats that were unstimulated or that exhibited predatory attack after lateral hypothalamic stimulation had low c-fos expression levels in the PAG. Substance P immunoreactivity was high throughout the dorsal PAG. The results indicate that NK(1) receptors in the PAG potentiate defensive rage and suppress predatory aggression in the cat.     

Nuclei within the rostral ventromedial medulla mediating morphine antinociception from the periaqueductal gray

The relative contributions of nuclei within the rostral ventromedial medulla (RVM) involved in mediating morphine induced antinociception from the periaqueductal gray (PAG) were examined. Lidocaine injections (4%) at the time of morphine's maximal response were used to provide a localized neural block and were administered in the nucleus raphe magnus/reticularis gigantocellularis pars alpha (RMg/GiA; commonly referred to as RMg), reticularis gigantocellularis (Gi) and reticularis paragigantocellularis lateralis (LPGi). Microinjection of morphine (6 nmol; 0.5 μl) into the PAG of awake rats produced an inhibition of the tail-flick reflex that was maximal after 30 min. This response was unaffected by a single medial lidocaine injection (0.5 μl) into the RMg/GiA or Gi, bilateral injections into the Gi or LPGi or triple injections that included both the RMg/GiA and LPGi. A partial, non-significant block of morphine's response was observed either by bilateral injections (0.5 μl) into both the Gi and LPGi (%inhibition= 16.4±24.8) or by bilateral injections in the LPGi and a single medial injections into the Gi (%inhibition= 41.5±29.8). However, injection of a greater volume of lidocaine (1 μl) into the RMg/GiA or bilaterally into the LPGi affected adjacent medial and lateral tissue, and completely inhibited morphine's response. Furthermore, triple injections of lidocaine (0.5 μl) into the Gi or bilateral injections (0.5 μl) into the Gi and a single medial injection into the RMg/GiA completely blocked morphine's antinociceptive response. These results indicate that morphine antinociception from the PAG is mediated by a large volume of tissue in the RVM containing nuclei located both medially and laterally. Additionally, the principal nuclei involved in this response appear to be the Gi and RMg/GiA.     

Lesions of the periaqueductal gray disrupt input to the rostral ventromedial medulla following microinjections of morphine into the medial or basolateral nuclei of the amygdala

Microinjections of morphine into the basolateral (BLa) and medial (MEa) nuclei of the amygdala differentially affect rostral ventromedial medulla (RVM) neuronal activity and nocifensive behaviors. PAG lesions attenuated or blocked the effects of both BLa and MEa morphine on RVM cell activity, and interfered with the behavioral antinociception produced by BLa infusions. These results demonstrate that the influences from both the BLa and MEa to the RVM are relayed via the PAG.     

Pontine control of the urinary bladder and external urethral sphincter in the rat

Neurons in the rostral pontine tegmentum are known to have an important role in controlling micturition. The present experiments used urethane anesthetized rats to examine the effects of electrical stimulation at various sites in the pons on bladder and external urethral sphincter activity and on the volume threshold for inducing micturition. Stimulation with short trains of pulses (50 Hz, 1-3 s trains, 1-15 V) in the laterodorsal tegmental nucleus (LDT), the periaqueductal grey (PAG) or the lateral parabrachial nucleus (L-PBN) elicited contractions of a partially filled, quiescent bladder. However stimulation during a bladder contraction aborted the contraction indicating that these areas have inhibitory as well as excitatory effects. Continuous stimulation (50 Hz) in the PAG or L-PBN during a cystometrogram decreased bladder capacity (mean decrease 36%). Conversely, continuous stimulation in the pontine reticular formation (in or near the dorsal subcoeruleus nucleus and medial parabrachial nucleus) increased bladder capacity (mean increase 50%). Stimulation at pontine sites (LDT, PAG and L-PBN) which elicited bladder contractions also elicited an increase in external urethral sphincter activity. A similar increase in urethral sphincter activity occurred during reflex micturition induced by bladder distension. These data suggest that bladder capacity and the coordination of bladder and external urethral functions are controlled by various neuronal populations in the rostral pons of the rat.     

Connections of the precommissural nucleus.

The connections of the precomissural nucleus (PRC) have been examined with anterograde and retrograde axonal tracing methods in the rat. Experiments with cholera toxin B subunit (CTb) indicate that the PRC shares a number of common afferent sources with the dorsolateral periaqueductal gray (PAG). Thus, we have shown that the nucleus receives substantial inputs from the prefrontal cortex, specific domains of the rostral part of the lateral septal nucleus, rostral zona incerta, perifornical region, anterior hypothalamic nucleus, ventromedial hypothalamic nucleus, dorsal premammillary nucleus, medial regions of the intermediate and deep layers of the superior colliculus, and cuneiform nucleus. Moreover, the PRC also receives inputs from several PAG regions and from neural sites involved in the control of attentive or motivational state, including the laterodorsal tegemental nucleus and the ventral tegmental area. The efferent projections of the PRC were analyzed by using the Phaseolus vulgaris-leucoagglutinin (PHA-L) method. Notably, the PRC presents a projection pattern that resembles in many ways the pattern described previously for the rostral dorsolateral PAG in addition to projections to a number of targets that also are innervated by neighboring pretectal nuclei, including the rostrodorsomedial part of the lateral dorsal thalamic nucleus, the ventral part of the lateral geniculate complex, the medial pretectal nucleus, the nucleus of the posterior commissure, and the ventrolateral part of the subcuneiform reticular nucleus. Overall, the results suggest that the PRC might be viewed as a rostral component of the PAG, and the possible functional significance of the nucleus is discussed in terms of its connections.     

Comparison of the antinociceptive effect of morphine and glutamate at coincidental sites in the periaqueductal gray and medial medulla in rats

In rats, microinjection of the opioid agonist morphine (15-45 nmol/0.5 microliter) and the excitatory amino acid monosodium glutamate (30-60 nmol/0.5 microliter) into identical brainstem sites within the mesencephalic periaqueductal gray matter (PAG; 23 sites) and the rostral ventromedial medulla (RVM; 22 sites) produced an increase of tail flick and hot plate response latencies. At both PAG and RVM sites, there was a statistically significant relationship between the effect obtained with morphine and with glutamate on the two nociceptive responses. While morphine and glutamate produced indistinguishable inhibition of tail flick and hot plate response latencies in the PAG, the tail flick inhibition following RVM morphine, but not RVM glutamate, displayed a clear plateau. One parsimonious interpretation of these data is that (1) glutamate directly increases the activity in the bulbospinal pathway, and (2) morphine inhibits an evoked or tonic suppression of a bulbospinal projection.     

Morphine pretreatment increases opioid inhibitory effects on maternal behavior

Ongoing maternal behavior in rats is under the inhibitory influence of opiates. Exposure to drugs of abuse may result in a progressive and enduring enhancement of their reinforcing effects. Little attention has been paid to the possibility that puerperal treatment with morphine may lead to sensitization to this drug, ultimately influencing the effects of opiates on maternal behavior. The aim of the present study was to investigate if the abrupt withdrawal of repeated treatment with morphine chlorhydrate (MC) during late pregnancy and early lactation may influence maternal behavior in lactating rats. The premise that a possible change in sensitivity to the inhibitory effect of MC on maternal behavior would last at least until day 17 of lactation without any reinforcement was tested. In addition, the hypothesis that the MC-induced inhibition would be reversed by the opioid antagonist naloxone was also tested. In all experiments female Wistar rats were treated with MC (5.0 mg/kg/day, subcutaneous [s.c.]) or saline for 7 days starting on the 17th day of pregnancy. After the abrupt discontinuation of long-term treatment, animals were acutely challenged with MC (5.0 mg/kg, s.c.) or saline and tested for maternal behavior in three different experimental situations: first, on days 5, 10, and 17 postpartum (Experiment 1); second, on day 17 postpartum (Experiment 2); third, on day 6 postpartum following naloxone pretreatment (1.0 mg/kg; Experiment 3). In Experiment 1, animals were treated for 7 days with morphine and acutely challenged with MC (group MM). Experimental MM animals showed significantly longer latencies for all maternal behavior parameters than all other groups during all observation days. The other groups (treated with MC for 7 days and acutely challenged with saline, group MS; treated with saline for 7 days and acutely challenged with MC, group SM; and treated with saline for 7 days and acutely challenged with saline, group SS) did not differ significantly from one another. In Experiment 2, in which rats were submitted to a single test on day 17 of lactation, the MM group showed significantly longer latencies for all behavioral parameters as compared to group SM. Previous acute naloxone treatment (Experiment 3) reversed the inhibitory effects of MC on maternal behavior in lactating rats. These data suggest that repeated administration of MC to female rats during late pregnancy sensitizes the animals to the inhibitory effects of opioids on rat ongoing maternal behavior.     

Subregions of the periaqueductal gray topographically innervate the rostral ventral medulla in the rat.

Previous anatomical and physiological studies have revealed a substantial projection from the periaqueductal gray (PAG) to the nucleus paragigantocellularis (PGi). In addition, physiological studies have indicated that the PAG is composed of functionally distinct subregions. However, projections from PAG subregions to PGi have not been comprehensively examined. In the present study, we sought to examine possible topographic specificity for projections from subregions of the PAG to PGi. Pressure or iontophoretic injections of wheat germ agglutinin-conjugated horseradish peroxidase, or of Fluoro-Gold, placed into the PGi of the rat retrogradely labeled a substantial number of neurons in the PAG from the level of the Edinger-Westphal nucleus to the caudal midbrain. Retrogradely labeled neurons were preferentially aggregated in distinct subregions of the PAG. Rostrally, at the level of the oculomotor nucleus, labeled neurons were i) compactly aggregated in the ventromedial portion of the PAG corresponding closely to the supraoculomotor nucleus of the central gray, ii) in the lateral and ventrolateral PAG, and iii) in medial dorsal PAG. More caudally, retrogradely labeled neurons became less numerous in the dorsomedial PAG but were more widely scattered throughout the lateral and ventrolateral parts of the PAG. Only few retrogradely labeled neurons were found in the ventromedial part of the PAG at caudal levels. Injections of retrograde tracers restricted to subregions of the PGi suggested topography for afferents from the PAG. Injections into the lateral portion of the PGi yielded the greatest number of labeled neurons within the rostral ventromedial PAG. Medially placed injections yielded numerous retrogradely labeled neurons in the lateral and ventrolateral PAG. Injections placed in the rostral pole of the PGi (medial to the facial nucleus) produced the greatest number of retrogradely labeled neurons in the dorsal PAG. To examine the pathways taken by fibers projecting from PAG neurons to the medulla, and to further specify the topography for the terminations of these afferents in the PGi, the anterograde tracer Phaseolus vulgaris-leucoagglutinin was iontophoretically deposited into subregions of the PAG that contained retrogradely labeled neurons in the above experiments. These results revealed distinct fiber pathways to the rostral medulla that arise from the dorsal, lateral/ventrolateral, and ventromedial parts of the PAG. These injections also showed that there are differential but overlapping innervation patterns within the PGi. Consistent with the retrograde tracing results, injections into the rostral ventromedial PAG near the supraoculomotor nucleus yielded anterograde labeling immediately ventral to the nucleus ambiguus in the ventrolateral medulla, within the retrofacial portion of the PGi.(ABSTRACT TRUNCATED AT 400 WORDS)     

Oxytocin Messenger Ribonucleic Acid Levels in the Medial Preoptic Area are Increased During Lactation

The peptide hormone oxytocin has an important role in parturition, lactation and maternal behavior. The present study employed in situ hybridization histochemistry to determine whether oxytocin messenger ribonucleic acid (mRNA) levels in cells in the medial preoptic area, a brain area known to control maternal behavior, change during pregnancy and lactation in the rat. Female rats were perfused on either Day 18 or 22 of pregnancy or Day 5 of lactation. Ovariectomized female rats were included as an additional control group. Cells expressing oxytocin mRNA were detected by in situ hybridization using an [¹²⁵ l]-labeled 38 base synthetic oligodeoxynucleotide probe complementary to the C-terminal coding region of the preprooxytocin. Relative differences in oxytocin mRNA levels were determined by silver grain counting of labeled cells. A group of oxytocin neurons in the dorsal medial preoptic area, called the lateral subcommissural nucleus, showed elevated oxytocin mRNA levels in lactating animals relative to ail other groups. Oxytocin mRNA levels in the neurons of the periventricular nucleus of the preoptic area did not change across pregnancy and lactation. This result extends the findings of others showing elevated oxytocin mRNA levels in magnocellular nuclei of lactating animals. The results are discussed in terms of the possible role of oxytocin cells in the medial preoptic area in the expression of maternal behavior.     

Efferent projections of ProTRH neurons in the ventrolateral periaqueductal gray.

Our previous study has shown that prothyrotropin-releasing hormone (proTRH) gene expression is increased in the ventrolateral periaqueductal gray (PAG) neurons following precipitated morphine withdrawal and continues to be activated even 24 h after withdrawal. We have hypothesized that peptide products of proTRH may participate in the recovery from morphine withdrawal. To identify neuroanatomical substrates of the proposed action of proTRH-derived peptides originating from the ventrolateral PAG proTRH neurons, projections of these neurons were investigated by a series of anterograde and retrograde tract-tracing experiments. First, Phaseolus vulgaris-leucoagglutinin (PHA-L) was injected in the ventrolateral PAG in Sprague-Dawley rats. Following transport of the tracer, simultaneous immunolabeling for PHA-L and proTRH peptides was performed and mapped in discrete brain regions. PHA-L-immunoreactive (IR) fibers showing preterminal and terminal-like arborization that contained proTRH were identified in the dorsolateral and lateral PAG, deep layer of superior colliculus (CS), parafascicular nucleus (PF), ventromedial zona incerta (ZI) and at the border of the locus coeruleus (LC) and Barrington's nucleus. Scattered double-labeled fibers were present in the lateral septal nucleus, ventromedial preoptic nucleus, lateral hypothalamus, perifornical area and in the periventricular region at the diencephalon/midbrain junction. The retrogradely transported marker, cholera toxin beta-subunit (CTb) was then injected in the dorsolateral PAG, CS, PF, ZI and medial to the LC. Double-labeled perikarya for both CTb and proTRH in the ventrolateral PAG were found for each region injected with CTb, corroborating the findings by the anterograde tracing experiment. These studies demonstrate that proTRH neurons in the ventrolateral PAG project to several regions of the brain that are involved in autonomic and behavioral regulation and thereby, may function as an integrating center to coordinate responses to opiate withdrawal.