Margaret M. McCarthy

Infusion of antisense oligodeoxynucleotides to the oxytocin receptor in the ventromedial hypothalamus reduces estrogen-induced sexual receptivity and oxytocin receptor binding in the female rat

Exogenous administration of the neuropeptide oxytocin reliably facilitates sexual behavior in the female rat and exposure to estrogen increases oxytocin receptor (OTR) binding in the ventromedial nucleus (VMN) of the hypothalamus. We have used a novel approach to investigate the role of hypothalamic OTR in controlling behavior by infusing antisense oligodeoxynucleotides (oligo) to the 5-region of the human OTR mRNA into the VMN of hormonally primed rats. Control infusions consisted of a scrambled-sequence oligo that had little or no homology to known mRNAs. OTR antisense oligo infusion significantly reduced lordosis frequency and intensity in females primed with estrogen. There was also a significantly greater number of rejection behaviors exhibited by antisense-oligo-infused estrogen-treated females versus controls and no evidence of decreased locomotion by either treatment. In contrast to the effects in estrogen-primed-females, when females were primed to be sexually receptive with estrogen plus progesterone, OTR antisense-oligo infusion had no effect on sexual behavior. The lack of effectiveness of OTR antisense oligo in females primed with progesterone may be the result of the action of this steroid on other neurotransmitter systems that also facilitate lordosis and thereby override a deficit in oxytocin binding. Alternatively, via previously described mechanisms, progesterone may enhance the effectiveness of oxytocin binding at its receptor. In vitro receptor autoradiography in estrogen-primed females indicated a 31% reduction in VMN OTR binding in the vicinity of the cannula tip in antisense-oligo-infused females compared to controls. There was no significant difference in the level of OTR binding in the central nucleus of the amygdala.(ABSTRACT TRUNCATED AT 250 WORDS)

Intracerebral administration of antisense oligodeoxynucleotides to GAD65 and GAD67 mRNAs modulate reproductive behavior in the female rat

Increased GABA activity in the medial hypothalamus (HYP) and midbrain central gray (MCG), but not the preoptic area (POA), facilitates sexual receptivity in the female rat [40]. In the current experiments, ovariectomized females were chronically treated with estrogen (via silastic capsules) to maintain a continuously high level of lordosis response. Administration of crystalline antisense oligodeoxynucleotide to the GABA synthetic enzyme, GAD67, into the HYP and MCG significantly and reversibly reduced lordosis response for 1-2 days, but did not inhibit lordosis when administered into the POA. Administration of a control oligonucleotide, consisting of the same nucleotide bases but in a scrambled sequence, did not significantly modulate behavior when infused into any brain areas. When oligodeoxynucleotide antisense to GAD67 was suspended in oil and then infused into the HYP or MCG it was more effective and resulted in less inter-animal variability. Subsequent experiments involving infusions into the MCG compared the effectiveness of antisense oligonucleotides to the two different forms of GAD, known as GAD65 and GAD67. Oligodeoxynucleotides antisense to the mRNA for either gene were effective at reducing lordosis behavior but with a different time course. Oligonucleotide antisense to GAD67 significantly reduced behavior within 24 h of infusion and there was full recovery by 4 days post-infusion. GAD65 antisense oligonucleotide did not significantly reduce behavior until 48 h post infusion and animals did not fully recover to pretest levels of lordosis until 5 days post-infusion. When antisense oligonucleotide for the two genes was administered simultaneously, the inhibition of lordosis was maximal at 24 h and stayed depressed for 4 days. There did not appear to be an additive effect of the two different antisense oligonucleotides when administered together. Tissue GABA levels in HYP and MCG of individual rats assayed by HPLC were no longer correlated with lordosis score after antisense oligonucleotide infusion but were after infusions of scrambled control oligos. Immunoblotting for the two forms of GAD revealed that GAD67 antisense oligonucleotide infusion led to significant decreases in both GAD67 and GAD65 protein levels as compared to infusions of scrambled control oligo. In addition, the levels of a neuronal marker, neuron-specific enolase, also decreased (although nonsignificantly) suggesting either a temporary shutdown of protein synthesis or a degeneration of GABAergic neurons after GAD67 antisense oligonucleotide infusion.

Midbrain central gray GABAA receptor activation enhances, and blockade reduces, sexual behavior in the female rat

SummaryThe inhibitory neurotransmitter, GABA, has been implicated in the control of lordosis behavior. Previous studies indicate that modulation of GABAA transmission can have dual effects on lordosis, being facilitative in the ventromedial hypothalamus and inhibitory in the preoptic area. The midbrain central gray (MCG) is also known to be an important neural site for regulating lordosis as well as defensive and escape behaviors, and plays an integral role in the control of nociception. Because of the multitude of behaviors regulated at the level of the MCG, we utilized a two-chamber testing apparatus that allowed simultaneous measurement of the females proceptive (hopping and darting), receptive and rejection behaviors, as well as an index of nociception and general motor activity. We found that microinfusion of the GABAA antagonist, bicuculline, into the MCG of steroid-primed female rats resulted in a significant decrease in lordosis and proceptive behaviors at 5 min post-infusion. There was full recovery to pretest levels by 60 min. Furthermore, microinfusion of the GABAA agonist, muscimol, to estrogen-treated females that displayed low levels of receptivity and high levels of rejection behavior during a pretest, resulted in a significant increase in lordosis responding and a decrease in rejection behaviors. Neither drug significantly affected time spent in the vicinity of the male, motor activity or vocalizations. It is concluded that the decrease in lordosis resulting from blockade of GABA transmission is not solely due to the induction of antagonistic behaviors since there was no increase in rejections after bicuculline administration. The current findings are consistent with the interpretation that GABA facilitates lordosis in the MCG via disinhibition. When the retrograde tracer, Fluoro-gold, was infused into the same cannula sites in the MCG as the GABAA drugs it demonstrated the presence of strong projections from the ventromedial nucleus, zona incerta, medullary reticular formation and spinal cord. These projections to the MCG may be important for the integration of the diverse behaviors regulated at the level of the MCG and GABAergic transmission may play a role in this integration.

Steroid Regulation and Sex Differences in [3H]Muscimol Binding in Hippocampus, Hypothalamus and Midbrain in Rats

The gonadal steroids estradiol and progesterone have previously been shown to modulate the specific binding of the GABAA agonist, [3H]muscimol, in the CA1 region of the hippocampus, the ventromedial nucleus of the hypothalamus and the midbrain central gray of ovariectomized female rats. In this report we show a sex difference in the level of binding in the very caudal ventromedial nucleus of the hypothalamus. In contrast to females, there is no steroid modulation of [3H]muscimol binding in the ventromedial nucleus of the hypothalamus and midbrain central gray of males. These effects may be functionally related to GABAergic control of female sexual behavior. In contrast, steroid modulation of [3H]muscimol binding in the CA1 region of the hippocampus occurred to the same degree in males and females, and there was no difference in the level of binding in any region of the hippocampus between gonadectomized males and females.

Speculations Concerning the Physiological Significance of Central Oxytocin in Maternal Behavior

Central administration of the neuropeptide oxytocin was first reported to induce short-latency maternal behavior in the rat by Pederson and Prange, a finding that has been replicated and extended* but apparently requires specific testing condition^.^.^ The narrow definition of the environmental and hormonal circumstances that allow the effect may have in fact contributed to furthering our understanding of the mechanisms of action of oxytocin on maternal behavior. In particular, the role of variables such as testing environment, degree of anosmia, and genetic strain may offer useful insights into how oxytocin is acting in the brain to modify behavior. J Furthermore, oxytocin has since been found to have wide-ranging effects on other reproductive behaviors, such as arousal, satiety, mating behavior, and social contact, in a variety of species including man (see other chapters in this volume). Oxytocin is also an important neurohormone involved in the response to stress, and we have proposed that many of the behavioral effects exerted by this neuropeptide may be the result of the underlying influence of stress.* The purpose of the current discussion will be to review in some detail our findings on oxytocin effects on maternal behavior in mice and rats and then to further our argument concerning the mechanisms of action of oxytocin.

Midbrain PAG Control of Female Reproductive Behavior: In Vitro Electrophysiological Characterization of Actions of Lordosis-Relevant Substances

Among the various innate behavioral repertoires of mammals, lordosis, as a major component of female reproductive behaviors is one of the best characterized. It is a stereotyped posture consisting of dorsiflexion of the vertebral column, which causes the elevation of the head and rump, accompanied by hind limb extension (Pfaff and Lewis, 1974). Somatosensory input given by a male partner is essential for every component of naturally occurring lordosis; it can be mimicked by an experimenter’s manual stimulation to the flanks, posterior rump, and perineum. The most important feature of lordosis is its steroid hormone dependency; an elevated level of estrogen is a prerequisite for the induction of lordosis both in naturally cycling and experimentally manipulated females. The nature and mechanisms of estrogen and progesterone control of lordosis have been extensively investigated and are well reviewed elsewhere (Pfaff and Schwartz-Giblin, 1988).

Ovarian steroid modulation of [3H]muscimol binding in the spinal cord of the rat

[3H]Muscimol binding was measured in the lumbar spinal cord of female rats by in vitro quantitative autoradiography. Ovariectomized rats were treated subcutaneously with either oil, estradiol benzoate (EB) or EB plus progesterone (P) in a regime known to reliably induce sexual receptivity. The level of [3H]muscimol binding was highest in laminae I-III and in the region around the central canal. Binding was lower in laminae IV-VI and was frequently undetectable in the ventral horn. There was a significant increase in the level of binding in laminae I-III after EB treatment. There was also a significant increase after treatment with EB+P in comparison to both the ovariectomized and EB-treated groups in this same region of the spinal cord.

Modulation by estrogen and progesterone of the effect of muscimol on nociception in the spinal cord.

The GABAA agonist, muscimol, administered intrathecally (IT) to the spinal cord at a dose (1 microgram) that was subthreshold for affecting pain thresholds (vocalization-threshold-to-tail-shock: VTTS, and tail-flick latency: TFL) in ovariectomized, hormonally untreated rats, showed a significant increase in VTTS up to 30 min postinjection in intact females only in proestrus or estrus. This treatment produced no significant effect on TFL at any stage of the estrous cycle. IT muscimol produced a significant increase in VTTS (but not TFL) in ovariectomized rats primed with estradiol benzoate (EB) for 2 days and tested 40 hr after the second injection but had no effect in females primed with a single EB injection and tested 15 min later. By contrast, ovariectomized females primed with progesterone (P) for 15 min exhibited a significant increase in pain thresholds after IT muscimol in both the VTTS and TFL tests. When EB-primed females (2 days) received P 4 hr prior to muscimol there was no analgesia produced by IT muscimol, in contrast to EB-primed females receiving P 15 min prior to IT muscimol in which there was significant analgesia. These results suggest a mechanism for antagonistic effects of estrogen and progesterone.

The medullary reticular formation is a site of muscle relaxant action of diazepam on deep back and neck muscles in the female rat.

We tested the hypothesis that the effect of systemic injections of diazepam (DZ, 125 mg/kg) to reduce the quality of the reproductive behavior, lordosis, and to reduce the EMG of lumbar back muscles involved in lordosis (Schwartz-Giblin et al., 1984) is exerted through a reticulospinal pathway with cells of origin in the nucleus gigantocellularis that excites lumbar motoneurons indirectly (Robbins et al., 1990, Robbins et al., 1992). In contrast, DZ facilitates lordosis behavior when infused into the midbrain central gray (McCarthy et al., 1995). Direct deposits of crystalline mixtures of DZ (20-80 ng) in dextrose were delivered to the medullary reticular formation (MRF) by diffusion from a cannula inserted through a guide to which a bipolar stimulating electrode was attached. The multiunit EMG response evoked by 20 (300 ms long) stimulus trains was recorded in back and neck muscles, lateral longissimus and splenius before and 5, 15, 30 and 60 min after local DZ deposits. There was a significant reduction in EMG response over this time period when stimulus intensities were within the range of 1.2-1.5 times threshold (Friedman two-way non-parametric test, P < 0.002). Large amplitude motor units that provide large tensions were the most sensitive to DZ-induced inhibition. Control deposits of dextrose had no significant effect. Systemic injections of progesterone (1 mg, i.p.) 60 min after DZ deposits, but not after dextrose deposits, further reduced the MRF-evoked EMG responses over the course of 1 h. As predicted, DZ infusions into the midbrain central gray did not reduce the reticulospinal-evoked axial muscle response, consistent with the facilitatory effect of midbrain central gray infusions of DZ on the lordosis quotient. The results suggest that benzodiazepine agonists (if endogenous) acting at sites in the MRF would be effective muscle relaxants during pregnancy, prior to the fall in progesterone that precedes labor.

Nerve growth factor affects defense-related behaviors, but not lordosis, in ovariectomized, estrogen-treated rats

Effects of NGF and anti-NGF on estrogen-sensitive behaviors were examined in ovariectomized, estrogen-treated rats. Intracerebroventricular (i.c.v.) administration of NGF resulted in a significant decrease in body weight. Daily treatment with low levels of estradiol resulted in a steady increase in lordosis behavior as reflected by average lordosis quotient and lordosis score. No effects of NGF or anti-NGF on lordosis behavior were detected. Estrogen treatment also resulted in a significant increase in the number of vocalizations elicited from female controls by male contact during sex behavior. NGF-treatment enhanced this effect, resulting in significantly more vocalizations elicited earlier in the course of estrogen treatment than were elicited from non-NGF-treated controls. These effects were blocked by progesterone. An increase in the number of rejections elicited by male contact during sex behavior was also observed in NGF-treated animals relative to controls. In addition, i.c.v. infusions of anti-NGF prevented the estrogen-mediated increase in elicited vocalizations, suggesting that NGF may have a physiological role in regulating this behavior. These data implicate NGF in the regulation of specific defense-related behaviors in estrogen-treated rats. Effects of NGF and anti-NGF on immunocytochemical staining for p75NGFR-, and ChAT-like immunoreactivity were also analyzed and are discussed.