Sex hormones and the hippocampus

Dorn Spinal Therapy

Spine Healing Therapy

Get Instant Access

Sex Differences Estrogen and androgen receptors are expressed in the hippocampus of the developing and adult rat. In the developing rat brain, both estrogen receptors and aromatizing enzymes are expressed perina-tally at the time of perinatal sexual differentiation and then decline to much lower levels that persist into adulthood (MacLusky et al., 1994; MacLusky et al., 1987; O'Keefe et al., 1995; O'Keefe and Handa, 1990). The cellular pattern of estrogen receptor expression is very similar in developing and adult hippocampus (S. Ha-yashi and C. Orikasa, personal communication).

Administration of estrogens to newborn female rats mimics the masculine sexual differentiation seen in males as far as the use of global spatial strategies in learning is concerned (Williams and Meck, 1991). Indeed, male and female rats use different strategies in spatial learning and memory, with males preferring global spatial cues, and females, local contextual information (Williams and Meck, 1991).

The hippocampus displays a number of anatomical sex differences. Notably, males have a larger dentate gy-

rus and more granule neurons, and dentate gyrus volume is increased in developing female rats by neonatal androgen treatment (Roof and Havens, 1992). In addition, sex differences in the pattern of dendritic branching and the density of dendritic spines on CA3 pyramidal neurons have been reported (Gould, Westlind-Daniels-son, et al., 1990). Finally, environmental enrichment has different effects on hippocampal morphology in male and female rats (Juraska, Fitch, and Washburne, 1989; Juraska et al., 1985).

Ovarian Hormones and Synaptogenesis Adult female rats express low levels of estrogen receptors, and these receptors are expressed in interneurons scattered throughout the hippocampus and cerebral cortex (Weiland et al., 1997). These recent studies confirm and extend earlier findings using steroid autoradiography (Loy, Gerlach, and McEwen, 1988) and immunocy-tochemistry (DonCarlos, Monroy, and Morrell, 1991). The newly discovered beta estrogen receptor has been reported to be expressed in CA1 and CA2 pyramidal neurons (Li, Schwartz, and Rissman, 1997), but this finding is disputed (Weiland, Alves, and Lopez, personal communication).

During the estrous cycle of the female rat, new excitatory spine synapses are produced and broken down (Woolley and McEwen, 1992; Woolley, Gould, and McEwen, 1990). This finding may help explain the variations in seizure susceptibility of the dorsal hippocampus during the estrous cycle reported in 1968 (Terasawa and Timiras, 1968), with a marked decrease in seizure threshold on the afternoon of proestrus, at the peak of estrogen and progesterone levels. stimulated by that paper, a morphological study that demonstrated estrogen induction of dendritic spines and new synapses in the ventromedial hypothalamus of the female rat led to the discovery that the density of dendritic spines on CA1 pyramidal neurons was also increased by estradiol (Gould, Woolley, Frankfurt, et al., 1990). These effects were specific for CA1, and dendritic spine density changed cyclically during the estrous cycle of the female rat (Woolley, Gould, and McEwen, 1990). There were also parallel changes in synapse density on dendritic spines, strongly supporting the notion of synapse turnover during the natural reproductive cycle of the rat (Woolley and McEwen, 1992).

There are two particularly important features of synapse formation and breakdown. First, progesterone administration rapidly potentiated estrogen-induced spine formation but then triggered the down-regulation of spines on CA1 neurons; the progesterone antagonist RU38486 blocked the natural down-regulation of spines during the estrous cycle (Woolley and McEwen, 1993).

Second, estrogen induction of new spine synapses on CA1 pyramidal neurons is blocked by concurrent administration of NMDA receptor antagonists (Woolley and McEwen, 1994). A recent study by Woolley showed, by electrophysiological recording, dye filling, and morphometry, that estrogen treatment increases not only spine density on CA1 neurons but also currents mediated by NMDA receptors when AMPA/kainate receptors are blocked (Woolley et al., 1997).

Moreover, a prominent effect of estrogen treatment is to induce NMDA receptor binding sites in the CA1 region of the hippocampus (Gazzaley et al., 1996; Weiland, 1992a). It is therefore possible that activation of NMDA receptors themselves could lead to induction of new synapses, in which case estrogen induction of NMDA receptors would then become a primary event leading to synapse formation.

The problem with all of this plasticity in CA1 pyramidal neurons is that, as noted earlier, the only detectable intracellular estrogen receptors are found in interneu-rons, not in the CA1 pyramidal neurons themselves. The possible exception is the presence of the beta ER in CA1 and CA2 neurons, discussed previously. There are also no detectable estrogen-inducible intracellular progestin receptors in hippocampus, based upon immunocyto-chemistry (N. G. Weiland and S. Alves, unpublished); however, there is a binding study showing low levels of such receptors in hippocampal tissue (Parsons et al., 1982). Nevertheless, recent studies with an estrogen antagonist argue strongly for a role of alpha estrogen receptors in interneurons in synapse induction. The estrogen antagonist CI-628 blocks estrogen induction of spines on CA1 neurons and does not produce any agonist effect (McEwen, Tanapat, and Weiland, 1999). This situation is reminiscent of blockade of progestin receptor induction in hypothalamus by CI-628, which is undoubtedly an alpha estrogen receptor-mediated effect (Roy, MacLusky, and McEwen, 1979). This experiment argues against a nongenomic effect of E and is consistent with the presence of alpha, but not beta, estrogen receptors in the inhibitory interneurons (Weiland et al., 1997; McEwen, Tanapat, and Weiland, 1999).

Recent studies on embryonic hippocampal neurons in culture have revealed that estrogens induce spines on dendrites of dissociated hippocampal neurons in culture by a process that is blocked by an NMDA receptor blocker and not by an AMPA/kainate receptor blocker (Murphy and Segal, 1996). In a subsequent study, estrogen treatment was found to increase expression of phos-phorylated CREB and CREB-binding protein, and a specific antisense to CREB prevented both the formation of dendritic spines and the elevation in phospho-CREB (Murphy and Segal, 1997a). In agreement with the in vivo data (already discussed), estrogen receptors were located in the cultures on GAD-immunoreactive cells that constitute around 20% of neurons in the culture; estrogen treatment caused GAD content and the number of neurons expressing GAD to decrease, and mimicking this decrease with an inhibitor of GABA synthesis, mercaptopropionic acid, caused an up-regulation of dendritic spine density, mimicking the effects of estra-diol (Murphy and Segal, 1997b). However, the situation in the in vivo hippocampus may be somewhat more complicated.

Estrogen induction of the mRNA for glutamic acid decarboxylase (GAD), a GABA synthesizing enzyme, was detected in interneurons residing within the CA1 pyramidal cell layer, a finding consistent with the known localization of estrogen receptors (Weiland, 1992b). Moreover, estrogen treatment induces GABAa receptor binding in the CA1 region of hippocampus, an effect that may be secondary to the action on the interneurons (McCarthy et al., 1992; Schumacher, Coirini, and McEwen, 1989). However, these effects were detected a number of days after estrogen treatment and are not inconsistent with a transient repression of GAD expression, as noted earlier in discussing the cell culture studies.

Concentrating for the moment on the interneurons containing the alpha ER, it is possible that the estrogen effects on formation of synapses on CA1 pyramidal neurons are indirect and mediated through these GABA in-terneurons. Wong and Moss (1992) reported that two-day estradiol treatment prolongs the EPSP and increases the probability of repetitive firing in some CA1 neurons in response to Schaffer collateral (glutamatergic) stimulation. One explanation is that estrogen actions on inhibitory interneurons might in some manner disinhibit the pyramidal neurons, allowing for removal of the magnesium blockade and activation of NMDA receptors (Or-chinik and McEwen, 1995). Two alternative pathways are outlined in figure 12.2. Estrogens may decrease the activity of the GABA inhibitory input to CA1 pyramidal neurons and thus disinhibit the CA1 neurons to allow for up-regulation of NMDA receptors and excitatory synapses; alternatively, estrogens may increase the activity of GABA inhibitory input to other inhibitory interneurons that synapse on CA1 pyramidal neurons, resulting in a disinhibition. As noted previously, the recent evidence from the cell culture model suggests that estrogens transiently suppress expression of glutamic acid decarboxy-lase in inhibitory interneurons (Murphy and Segal, 1997b), thus supporting the first model shown in figure 12.2. Neuroanatomical data on the type of GABA inter-neuron that expresses estrogen receptors are consistent with this notion (N. G. Weiland, unpublished).

Figure 12.2 Alternative schemes for the role of GABA inter-neurons in synaptogenesis on CA1 pyramidal neurons. Estrogens may act to repress GABA activity in interneurons that synapse on CA1 neurons and produce a state of disinhibition that allows synapse formation to proceed. Alternatively, estrogens may act on GABA interneurons that synapse on other in-terneurons, in which case the estrogen effect might be to up-regulate GABA activity and produce a disinhibition on the CA1 pyramidal neurons.

Figure 12.2 Alternative schemes for the role of GABA inter-neurons in synaptogenesis on CA1 pyramidal neurons. Estrogens may act to repress GABA activity in interneurons that synapse on CA1 neurons and produce a state of disinhibition that allows synapse formation to proceed. Alternatively, estrogens may act on GABA interneurons that synapse on other in-terneurons, in which case the estrogen effect might be to up-regulate GABA activity and produce a disinhibition on the CA1 pyramidal neurons.

Androgens and the Hippocampus The hippocampus expresses androgen receptors that are prominently localized in CA1 pyramidal neurons (J. E. Kerr et al., 1995). Androgens positively regulate androgen receptor mRNA levels in CA1, although androgen receptor binding is less sensitive to castration and hormone replacement than mRNA levels (J. E. Kerr et al., 1995).

In turn, androgen treatment induces changes in NMDA receptor expression and alters neuronal activity mediated by NMDA receptors in CA1 pyramidal neurons. Electrophysiological^, chronic treatment of castrated rats with the potent androgen dihydrotestosterone (DHT) increased the action potential duration of CA1 pyramidal neurons and decreased the amplitude of the fast hyperpolarization; moreover, DHT treatment promoted recovery of the membrane potential after depolarization with NMDA (Pouliot, Handa, and Beck,

1996). Binding of the antagonist MK801 to NMDA receptors in the CA1 region was elevated by castration, and this increase was prevented by DHT treatment, indicating that the antagonist-binding form of the NMDA receptor is affected by androgen treatment (Kus et al., 1995).

Castration increased levels of the mRNA for GnRH receptor not only on CA1 pyramidal neurons but also in the CA3 region and the dentate gyrus; curiously, however, castration decreased the second-messenger response to GnRH in hippocampal slices (Jennes et al., 1995). Except for electrophysiological studies, there is nothing to suggest a function for GnRH receptors in the hippocampus.

Androgens also have an influence on the adrenocorti-cal response system of the hippocampus. Androgen treatment suppressed expression of Type II (glucocorti-coid) receptor mRNA but had no effect on Type I (min-eralocorticoid) receptor mRNA levels ( J. E. Kerr, Beck, and Handa, 1996a). In this connection, castration has been reported to increase reactivity of the hypo-thalamo-pituitary-adrenal (HPA) axis (Handa et al., 1993). One possible reason for this effect is that castration increases reactivity of neural circuits involved in regulating HPA function. Consistent with this notion is the finding that castration increased, and androgen replacement suppressed, the immediate early gene responses to novelty in the CA1 region of the hippocampus but not in CA3 or dentate gyrus ( J. E. Kerr, Beck, and Handa, 1996b).

Was this article helpful?

0 0

Post a comment