| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Physiology, Pharmacology and Toxicology, Marshall University School of Medicine, Huntington, West Virginia
Submitted 8 September 2005; accepted in final form 5 February 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Noguchi 1996; Scheepens et al. 2000) and decreased GH levels are associated with a variety of cognitive impairments including memory disruption (reviewed in van Dam et al. 2000). Although the primary source of GH is the anterior pituitary, there are other sites for GH production, including the hippocampus (Gossard et al. 1987; Hojvat et al. 1982). In addition to producing GH, the hippocampus also contains GH receptors (Burton et al. 1992; Lobie et al. 1993; Mustafa et al. 1994a).
Beyond its role in development, GH also plays a role in adult memory processing. GH modulates long-term memory (Schneider-Rivas et al. 1995) and may attenuate effects of aging on memory (Ramsey et al. 2004; Thornton et al. 2000). The effects of GH on memory may be mediated by the hippocampus (Nyberg 2000). Interestingly, hippocampal GH expression is upregulated during learning (Donahue et al. 2002), suggesting a possible autocrine or paracrine function for GH in the hippocampus. Chronic GH treatment alters N-methyl-D-aspartate (NMDA) receptor gene expression in the hippocampus (Le Greves et al. 2002) and this could at least partially explain the effects of GH on memory function. Although previous studies have examined effects of long-term GH treatment and GH deficiency, acute effects of GH on hippocampal function have not been investigated.
The GH receptor (GHR) belongs to the cytokine receptor superfamily. GH stimulation causes receptor dimerization, activation of the Janus kinase 2 (JAK2) tyrosine kinase, and phosphorylation of signal transducer and activator of transcription 5 (STAT5; reviewed in Herrington and Carter-Su 2001; Moutoussamy et al. 1998). Other downstream signaling events include stimulation of mitogen-activated protein kinase (MAP kinase) and phosphatidylinositol-3 (PI3) kinase (Anderson 1992; Argetsinger et al. 1995; Jeay et al. 2001; Shoba et al. 2001; Sotiropoulos et al. 1994). GH is well known for its ability to rapidly stimulate protein synthesis (Dreskin and Kostyo 1980; Jeffereson et al. 1975; Kostyo and Nutting 1973; Mowbray et al. 1975). Our main objective in this study was to determine whether somatotropic signaling exerts short-term effects on hippocampal function and, if so, which signaling pathway or pathways are required.
We found that GH enhanced both 
-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)- and NMDA-receptor–mediated excitatory postsynaptic potentials (EPSPs), but did not alter 
-aminobutyric acid type A (GABAA)–receptor-mediated inhibitory synaptic transmission. The effects of GH on excitatory synaptic transmission required JAK2, MAP kinase, PI3 kinase, and protein synthesis. MAP kinase and PI3 kinase were required for initial enhancement of EPSPs, but were not required for maintained enhancement of EPSPs.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
All experimental procedures were approved by the Institutional Animal Care and Use Committee at Marshall University. Hippocampal slices were prepared from 1.5- to 3-mo-old male Sprague–Dawley rats (Hilltop Laboratory Animals). Animals were sedated by inhalation of a CO2/air mixture and decapitated. The skull was opened and the brain was removed and submerged in chilled, oxygenated (95% O2-5% CO2), low Ca2+/high Mg2+ artificial cerebrospinal fluid (ACSF) composed of: 124 mM NaCl, 26 mM NaHCO3, 3 mM KCl, 0.5 mM CaCl2, 5.0 mM MgSO4, and 10 mM glucose. While submerged in chilled low Ca2+/high Mg2+ ACSF the brain was trimmed to a block containing both hippocampi. The block was glued to the stage of a vibrating microtome (Campden Instruments), immersed in a bath of chilled, oxygenated, low Ca2+/high Mg2+ ACSF, and 400-µm coronal sections were cut. Sections containing the hippocampus in transverse profile were selected and transferred to a small petri dish, where they were further dissected to free the hippocampus from surrounding tissue. The CA3 region was removed from slices which were later treated with the GABAA receptor antagonist bicuculline. After dissection, hippocampal slices were transferred to a holding chamber where they were stored for later use.
Slices were maintained in the holding chamber at room temperature (20–22°C) at the ACSF/atmosphere (95% O2-5% CO2) interface. The holding chamber was filled with standard ACSF composed of 124 mM NaCl, 26 mM NaHCO3, 3.4 mM KCl, 1.2 mM NaH2PO4, 2.0 mM CaCl2, 2.0 mM MgSO4, and 10 mM glucose. Slices were incubated in the holding chamber for a minimum of 1 h before use.
Slices were withdrawn from the holding chamber as needed and placed in a low-volume (about 200 µL) interface recording chamber, where they were continuously perfused at a rate of 1–1.5 ml/min with standard ACSF. The recording chamber was kept at a temperature of 25 ± 0.5°C. A minimum 30-min period was allowed for recovery after transferring slices from the holding chamber to the recording chamber.
Field potential recording
Extracellular potentials were recorded through low-impedance (3–4 M
) glass micropipettes filled with ACSF and placed into the stratum radiatum of area CA1. Signals were amplified (gain 1,000) and filtered (0.1–3,000 Hz) using a WPI DAM50 amplifier, then digitized (10 kHz; National Instruments) and stored on a personal computer.
Whole cell patch-clamp recording
Whole cell recordings were obtained from the somata of CA1 pyramidal neurons by the method of Blanton et al. (1989). Patch electrodes (3–4 M) were filled with a solution of 140 mM cesium or potassium gluconate, 10 mM sodium HEPES (N-[2-hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid]), and 3 mM MgCl2. Positive pressure was applied to the back of the patch electrodes as they were lowered into the somatic layer of area CA1, and the electrode resistance was continuously monitored. When electrode resistance increased, positive pressure was released, and gentle negative pressure was applied to form a high-resistance seal (>1 G, typically 2–5 G) with the cell membrane. The membrane patch was then ruptured to obtain the whole cell recording configuration. Most recordings were done in current-clamp mode, although some recordings were made in whole cell voltage-clamp mode with membrane potential clamped to the initial level resting level throughout the recording.
Membrane potentials were measured with an Axoclamp 2B (Axon Instruments) operating in continuous current-clamp mode. Access resistance was measured and compensated using the Axoclamp bridge balance circuitry. Cell input resistance was monitored throughout experiments by passing small hyperpolarizing and depolarizing currents into the cell (±50–100 pA). Cells were discarded if access or input resistances showed large, abrupt, irreversible changes. Reported membrane potentials were compensated for a liquid junction potential of 10 mV. Series resistance compensation was not applied during whole cell voltage-clamp recordings.
Synaptic stimulation
Postsynaptic potentials were evoked by delivery of constant-voltage stimuli through a bipolar stimulating electrode placed into the stratum radiatum. Stimuli were delivered at a 15-s interval. In some field potential recordings, paired stimuli (50-ms interstimulus interval) were delivered to measure paired-pulse facilitation (PPF), which was quantified as the ratio of the second response divided by first response.
Postsynaptic potentials evoked in standard ACSF were quantified by measuring the slope of the linear portion of the initial response. In some recordings we isolated synaptic responses mediated by NMDA receptors or GABA receptors. NMDA-receptor–mediated EPSPs were isolated by perfusing slices with the AMPA receptor antagonist DNQX (30 µM) and the GABAA receptor antagonist bicuculline (bicuculline methiodide, 10 µM). GABA-receptor–mediated inhibitory postsynaptic potentials (IPSPs) were isolated by perfusing slices with the AMPA-receptor antagonist DNQX (6,7-dinitroquinoxaline-2,3-dione, 30 µM) and the NMDA-receptor antagonist D-AP5 (D-2-amino-5-phosphonopentanoic acid, 50 µM). Isolated synaptic responses were quantified by measuring peak amplitude.
Occlusion test
We tested for mutual occlusion between effects of GH treatment and high-frequency (tetanic) stimulation, by applying two trains of 100-Hz, 1-s stimuli at an intertrain interval of 30 s. One group of slices was first pretreated with growth hormone (60–120 min), washed in standard ACSF for 30–60 min, and then stimulated with two trains of 100-Hz stimuli as described above. A second group of slices received repeated rounds of tetanic stimulation until potentiation reached a stable, maximal amplitude, followed 20–60 min later with application of GH. Results from these two groups of slices were compared with slices receiving tetanic stimulation alone or GH alone.
Reagents
Reagents used in this study were: recombinant human GH (Bachem); recombinant rat GH (Cell Sciences); bicuculline methiodide, D-AP5, DNQX (Tocris); 1,4-diamino-2,3-dicyano-1,4-bis(2-aminophenylthio)-butadiene (U-0126), tyrphostin AG 490 (LC Labs); wortmanin, cycloheximide (3-[2-(3,5-dimethyl-2-oxocyclohexyl)-2-hydroxyethyl] glutarimide) (Sigma). Tyrphostin AG 490, wortmanin, and cycloheximide were dissolved in dimethyl sulfoxide (DMSO) and stored as a concentrated stock solution; all other reagents were dissolved into water. Control recordings revealed that DMSO, at the concentrations used in this study (
0.01%), had no observable effects on baseline synaptic activity or effects of GH. Tyrphostin AG 490, U-0126, wortmanin, and cycloheximide also had no effects on baseline synaptic activity. Stock solutions were dissolved into ACSF at the final working concentration for application to slices. Salts and other compounds were from Sigma or Fisher.
Data analysis
Changes in synaptic response caused by GH treatment or tetanization were expressed as percentage of baseline before treatment. Synaptic responses and membrane potentials were recorded and initially analyzed using the WinWCP program (Strathclyde Electrophysiology Software, John Dempster, University of Strathclyde). Further data analysis was performed with Origin (Microcal Software) and Excel (Microsoft). Statistical significance was assessed by paired or unpaired t-tests, as appropriate, with P < 0.05 (two-tailed) considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
We used field potential recordings to determine whether recombinant human GH (rhGH) application affects excitatory synaptic transmission. Application of rhGH (22 ng/ml) produced a slow increase in EPSP slope (Fig. 1). EPSPs began to increase within a few minutes of rhGH application, but required 
60 min to reach a stable level. EPSPs remained enhanced throughout continued rhGH application. In slices where a presynaptic fiber volley could be clearly resolved, the increase in EPSP slope occurred without any change in the fiber volley (Fig. 1).
|
; Gomez-Orellana et al. 1998; Li 1962; Pfund et al. 1996). In a second control experiment we compared rhGH to recombinant rat GH (rrGH). Primate GHs, including rhGH, are effective ligands at both GH and prolactin receptors, whereas nonprimate GHs, which retain activity at GH receptors, are ineffective ligands at prolactin receptors (reviewed in Goffin et al. 1996). To verify that the EPSP enhancement we observed in response to rhGH was caused by GH receptor stimulation, and not prolactin receptor stimulation, we compared the effects of rhGH with rrGH. The results are summarized in Fig. 2. Both GHs enhanced EPSPs, with similar time course and magnitude. EPSPs from slices treated with rrGH (n = 7) increased to 224.6 ± 25.3% of baseline (–0.26 ± 0.07 V/s) after 60 min of treatment. This value was not significantly different from that obtained with rhGH (P > 0.99). Because rhGH and rrGH produced comparable results, subsequent recordings were done using rhGH.
|
4 h. After 4 h of GH application, EPSPs averaged 217.2 ± 30.4% of the initial baseline (–0.12 ± 0.2 V/s, n = 8), a value significantly greater than that from slices perfused with ACSF alone for an equivalent period of time (71.4 ± 8.7% of the initial baseline level of –0.15 ± 0.03 V/s, n = 4, P < 0.002).
|
; Hess et al. 1987; Otmakhov et al. 1993; Zucker 1989). To determine whether increased probability of glutamate release might underlie the GH enhancement, we examined PPF during GH application. Although 1 h of GH application caused a substantial and significant increase in EPSP slope to 218.1 ± 18.6% of the initial baseline (–0.21 ± 0.03 V/s, P < 0.0001), there was no significant change in PPF ratio (n = 15, Fig. 4A). Before GH application, the PPF ratio averaged 1.38 ± 0.09, and after 60 min of GH treatment, the PPF ratio was 1.30 ± 0.08 (P > 0.27). Our failure to observe a change in PPF during GH application is not the result of a lack of sensitivity in our methods because changes in PPF were easily observed during application of high-Ca2+ ACSF (Fig. 4B) and during the posttetanic facilitation that follows high-frequency tetanic stimulation (data not shown; see also Grover 1998). Because GH caused a significant increase in EPSPs without causing significant alteration in PPF, it seems unlikely that increased probability of glutamate release underlies the GH enhancement of excitatory synaptic transmission. Our PPF data do not, however, argue against other changes in presynaptic function that do not involve altered probability of release.
|
; Collingridge et al. 1983; Davies and Collingridge 1989; Koerner and Cotman 1982), the changes in field potentials that we observed during GH treatment most likely reflect alterations in AMPA-receptor–mediated synaptic transmission. To determine whether this is, in fact, the case, we conducted an additional set of recordings where we blocked NMDA receptors with the antagonist D-AP5 (50 µM) before applying GH to slices. As shown in Fig. 5, we observed a large increase in AMPA-receptor EPSPs during GH application. After 60 min of GH treatment, EPSPs averaged 389.6 ± 101.5% of the initial baseline (–0.19 ± 0.03 V/s, n = 5). This increase was significant (P < 0.05).
|
We used whole cell recordings to examine the effects of GH on EPSPs, pharmacologically isolated NMDA-receptor–mediated EPSPs (NMDA-EPSPs), and pharmacologically isolated IPSPs. We also assessed possible effects of GH on resting membrane potential and input resistance (RN).
GH ENHANCED EPSPS AND HYPERPOLARIZED PYRAMIDAL NEURONS. During GH application, EPSPs slowly increased (Fig. 6A1), as we saw during field potential recordings. After 30 min of GH application, whole cell EPSPs were significantly increased to 181 ± 22% of baseline (from an initial slope of 1.19 V/s, n = 7, P < 0.01; Fig. 9A). The slow increase in whole cell EPSP was paralleled by a slow hyperpolarization of about 3 mV and a slow decrease in RN of about 30% (P values of <0.05 and <0.01, respectively; n values of 10 and 11; Figs. 6B and 9B).
|
|
; Collingridge et al. 1983; Davies and Collingridge 1989; Koerner and Cotman 1982), and the enhancement of these EPSPs by GH indicates an effect on AMPA-receptor–dependent synaptic transmission, a conclusion supported by our earlier field potential recordings (Fig. 5). To determine whether NMDA-receptor–dependent synaptic function is also affected by GH, we treated slices with the AMPA-receptor antagonist DNQX and the GABAA receptor antagonist bicuculline. Afferent stimulation under these conditions results in a PSP consisting of an NMDA-receptor–mediated EPSP (NMDA-EPSP) followed by a GABAB IPSP. In preliminary recordings (not shown), we found that the GABAB IPSPs ran down substantially during the first 30 min of whole cell recording. Because the NMDA-EPSP and GABAB IPSP partially overlap in time, rundown of the GABAB IPSP allows the NMDA-EPSP to increase. Therefore to avoid possible confounding effects of GABAB rundown and GH, we either 1) increased the duration of the baseline recording before application of GH to 30 min to allow GABAB IPSP rundown to complete, or 2) substituted Cs+ for K+ in our whole cell pipette solution to block the GABAB conductance (Ling and Benardo 1994; Otis et al. 1993). Under these conditions, NMDA-EPSPs were stable before GH application, and it was possible to determine whether GH altered the NMDA-EPSP. Because the NMDA-EPSP is voltage-dependent, in these recordings we injected DC holding current to maintain the original resting membrane potential (–65 to –75 mV) and compensate for any GH-induced hyperpolarization. As shown in Fig. 7A, GH application caused enhancement of isolated NMDA-EPSPs similar to its effect on EPSPs evoked in standard ACSF (compare Figs. 6 and 7). In the presence of GH, NMDA-EPSPs slowly increased over time, reaching a mean level of 171 ± 10% of baseline (initial amplitude 1.17 ± 0.21 mV, n = 3) after 30 min of application (P < 0.01, Fig. 9A). Similar results were obtained during whole cell voltage-clamp recordings (Fig. 7B). Isolated NMDA-receptor–mediated EPSCs increased from a mean of –64.3 ± 36.3 to –109.7 ± 47.1 pA after 30 min of GH treatment (n = 3).
|
|
|
|
|
|
Our whole cell and field potential recordings revealed pronounced effects of GH on both AMPA- and NMDA-receptor–mediated EPSPs, membrane potential, and input resistance, but no effect on GABAA-receptor–mediated IPSPs. To begin characterizing the signaling pathway responsible for GH enhancement of EPSPs, we applied specific inhibitors of JAK2, PI3 kinase, or MAP kinase kinase (MEK) along with GH, and examined effects on field EPSPs. The JAK inhibitor tyrphostin AG 490 (10 µM), applied beginning 30 min before and during GH, caused a significant decrease in the normal EPSP enhancement seen during GH (Fig. 11); after 60 min of GH + tyrphostin AG 490, EPSPs averaged 153 ± 22% of baseline (–0.17 ± -0.02 V/s, n = 6), compared with 255 ± 27% when GH was applied alone (P < 0.05, n = 12). As shown in Fig. 11, complete inhibition was seen when GH was applied with 20 µM of the MEK inhibitor U0126 (96 ± 7% of baseline, –0.26 ± 0.05 V/s, n = 7, P < 0.001 vs. GH alone) or 50 nM of the PI3 kinase inhibitor wortmanin (105 ± 8% of baseline, –0.21 ± 0.02 V/s, n = 6, P < 0.01 vs. GH alone). GH signaling in area CA1 of the hippocampus therefore involves both PI3-kinase and MAP-kinase signaling pathways, as shown previously in other tissues (Anderson 1992; Argetsinger et al. 1995; Jeay et al. 2001; Shoba et al. 2001; Sotiropoulos et al. 1994).
|
In support of a temporally limited requirement for both PI3- and MAP-kinase signaling pathways, the addition of either wortmanin (50 nM) or U0126 (10 µM) during the final 30 min of a 60-min period of GH treatment failed to affect the EPSP enhancement (Fig. 12). EPSPs were increased to 244 ± 65% of baseline (initially –0.16 ± 0.02 V/s, n = 5) in slices treated with GH followed by wortmanin (P > 0.30 compared with GH alone) and were increased to 212 ± 16% of baseline (initially –0.14 ± 0.01 V/s, n = 5) in slices treated with GH followed by U0126 (P > 0.50 compared with GH alone, n = 12).
GH is capable of rapidly stimulating protein synthesis (Dreskin and Kostyo 1980; Jeffereson et al. 1975; Kostyo and Nutting 1973; Mowbray et al. 1975). To determine whether synthesis of new proteins contributes to GH enhancement of EPSPs, we pretreated slices with the protein synthesis inhibitor cycloheximide (60 µM) for 30 min before GH treatment, and continued to apply cycloheximide with GH for 60 min. To control for possible nonspecific effects of cycloheximide on synaptic responses, we treated a second group of slices with cycloheximide alone. Results from these two groups of slices were compared with results from slices treated with GH alone for 60 min. The control group treated with cycloheximide alone showed no change in field EPSP (at 60-min averaging 94 ± 14% of the initial baseline of –0.18 ± 0.02 V/s, n = 5, P > 0.60; see Fig. 13). Slices treated with GH + cycloheximide failed to show the EPSP enhancement normally seen during GH application. At the end of the 60-min GH + cycloheximide application, EPSPs were unchanged from baseline (99 ± 22% of the initial baseline of –0.18 ± 0.04 V/s, n = 6, P > 0.90) and were indistinguishable from the cycloheximide-alone control slices. Slices treated with GH alone (no cycloheximide, n = 12) showed significant EPSP enhancement in comparison with both the GH + cycloheximide and cycloheximide-alone groups (P values all <0.001).
|
; Sweatt 2001). To test whether common signaling mechanisms are involved in GH enhancement of EPSPs and LTP, we conducted a final set of experiments. In these experiments we looked for mutual occlusion between GH and LTP induced by high-frequency (100-Hz) stimulation. In the first experiment we pretreated slices with GH, then washed GH from the tissue and attempted to induce LTP by 100-Hz tetanization; results were compared with slices that received 100-Hz tetanization without prior GH treatment. Sample results from two slices, one receiving 100-Hz tetanization without prior GH treatment and one receiving 100-Hz tetanization after GH treatment, are shown in Fig. 14, A and B. Slices receiving 100-Hz tetanization showed robust LTP, as expected, averaging 160 ± 17% of baseline (–0.15 ± 0.02 V/s, n = 9; Fig. 14C). In contrast, slices pretreated with GH showed posttetanic potentiation (PTP) only, and no LTP (at 50 min posttetanus, EPSPs averaged 97 ± 4% of the initial baseline, –0.21 ± 0.04 V/s, n = 6; P < 0.01; Fig. 15C). The difference between these two groups, at the 50-min posttetanus time point, was significant (P < 0.01). In a second experiment, we reversed the order of treatment, first inducing maximal LTP by repeated delivery of 100-Hz stimulation, then applying GH; results were compared with slices receiving GH with no prior tetanization. Sample results from two slices are shown in Fig. 15, A and B to illustrate the protocols used. Prior induction of LTP, by repeated 100-Hz stimulation, completely abolished the normal response to GH; after 60 min of treatment (Fig. 15C); EPSPs averaged 109 ± 5% of the initial baseline (–0.30 ± 0.07 V/s, n = 5), a value significantly less (P < 0.0002) than that obtained in slices treated with GH without prior tetanization (255 ± 27% of baseline, n = 12). |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Lobie et al. 1993; Mustafa et al. 1994a) and hippocampal function is affected by chronic GH administration or deficiency (Le Greves et al. 2002; Ramsey et al. 2004; van Dam et al. 2000). The effects of GH on memory may be mediated by the hippocampus (Nyberg 2000). Our results demonstrate that hippocampal function is affected by GH over a much shorter timescale—minutes to hours—than previously appreciated. GH administration caused a selective enhancement of excitatory synaptic transmission in area CA1 (Figs. 1–7, 9, and 10) without affecting fast GABAA-receptor–mediated synaptic inhibition (Figs. 8 and 9). GH enhanced both AMPA- and NMDA-receptor–mediated excitatory synaptic transmission (Figs. 5, 7, and 9) without altering paired-pulse facilitation, an index of presynaptic function that changes when the probability of transmitter release is altered (Fig. 4). This increase in excitatory synaptic function was correlated with significant changes in postsynaptic neuron membrane function: hyperpolarization and decreased input resistance (Figs. 6, 9, and 10). However, these changes in postsynaptic neuron membrane function do not explain the increase in EPSPs. Decreased input resistance might lead to smaller EPSPs, but would not cause enhancement of EPSPs, and EPSP enhancement, at least for NMDA-receptor–mediated EPSP/Cs was observed when changes in membrane potential were prevented (Figs. 7 and 9). Although our findings indicate that GH does affect postsynaptic membrane function, decreasing input resistance and hyperpolarizing CA1 neurons, and we found no change in PPF during GH application, suggesting that the probability of glutamate release from presynaptic terminals is not altered, we cannot exclude the possibility that GH has presynaptic effects. For example, GH could increase the number of presynaptic release sites without changing PPF so long as the new release sites have the same overall probability of release as the original release sites.
GH enhancement of EPSPs requires JAK2, PI3 kinase, and MEK (Fig. 11) as well as synthesis of new proteins (Fig. 13). Although PI3-kinase, MEK, and protein synthesis are required for GH effects on excitatory synaptic transmission, the requirement is temporally limited because application of inhibitors beginning 30 min after the start of GH application failed to reverse the EPSP enhancement (Figs. 11 and 13). On the contrary, EPSPs continued to increase during inhibitor application (Fig. 13). That EPSPs continued to increase during maintained GH application for 60–70 min, either with or without inhibitors present (Figs. 1–5 and 13), may indicate the existence of additional downstream signaling targets whose activity outlasts initial signaling events. Alternatively, the continuing increase in EPSPs beyond the first 30 min of GH application may simply reflect the final consequences of signaling events that occur during the first 30 min. For example, the final delivery of new proteins that are synthesized within the first 30 min of GH stimulation could require >30 min.
GH signaling has been studied extensively in several nonneuronal tissues (reviewed in Herrington and Carter-Su 2001; Kelly et al. 2001; Piwien-Pilipuk et al. 2002). GH binding to GH receptors induces receptor dimerization and activation of JAK2. In response to GH receptor-JAK2 activation, several distinct signaling pathways are stimulated, including the STAT (primarily STAT5, but also STAT1 and STAT3), PI3-kinase, and MAP-kinase pathways. GH signaling in hippocampus requires at least JAK2, PI3 kinase, and MEK, indicating substantial similarity with other tissues. We have not yet investigated the possible role of STATs in GH enhancement of excitatory synaptic transmission. In response to GH, STATs are phosphorylated, dimerize, and stimulate transcription of specific genes (Herrington and Carter-Su 2001; Kelly et al. 2001; Piwien-Pilipuk et al. 2002), leading eventually to synthesis of new proteins. Considering the time required for formation of new proteins in response to STAT activation and the transport of these new proteins from somatic to synaptic regions of the neuron, it seems unlikely that STATs could participate during the initial response to GH, which begins within a few minutes of the start of GH application (see Figs. 1–7 and 10). However, it is possible that the sustained response to GH, which persists for 4 h, might require activation of one or more of the STATs. The possible contributions of one or more of the STATs will need to be addressed in future work. Future experiments will also need to determine which of the GH-stimulated pathways is required for GH effects on postsynaptic membrane function (membrane potential and input resistance).
Future experiments should also address the possible contribution of insulin-like growth factor I (IGF-I) to the acute GH effects we observed. IGF-I expression is induced by GH and IGF-I in turn is responsible for mediating many but by no means all of the effects of GH (reviewed in Jones and Clemmons 1995; LeRoith et al. 2001). Previous studies in the hippocampus suggest that IGF-I can act as a modulator of synaptic transmission (Huang et al. 2004; Ramsey et al. 2005). Although IGF-I might contribute to GH effects on hippocampal function, it seems unlikely that IGF-I acts as a mediator during the initial response to GH, for the same reason that the STATs seem unlikely to contribute during the early phase of the response. IGF-I might, however, contribute to the maintained effects of GH several hours after the initial application.
Primate, but not nonprimate, GH can bind to and activate prolactin receptors (Goffin et al. 1996; Kossiakoff et al. 1994). The hippocampus contains prolactin receptors (Fechner and Buntin 1989; Lai et al. 1992; Mustafa et al. 1994a,b) in addition to GH receptors, and they might contribute to effects of rhGH. We therefore examined rrGH, which does not bind the prolactin receptor (Amit et al. 1987; Møldrop et al. 1990). We found identical effects for both rrGH and rhGH, indicating that prolactin receptors are not required for GH effects in the hippocampus.
Our results indicate a number of similarities between GH enhancement of EPSPs and other forms of long-lasting synaptic enhancement in hippocampal area CA1, including high-frequency stimulation-induced (tetanus-induced) long-term potentiation (LTP) and brain-derived neurotrophin (BDNF)–induced potentiation. We found that GH enhancement of EPSPs required activation of both the PI3-kinase and MAP-kinase pathways, and also required synthesis of new proteins. Similarly, tetanus- and neurotrophin-induced LTP also require PI3-kinase and MAP-kinase signaling, and synthesis of new proteins (reviewed in Kelleher et al. 2004; Lynch 2004). It appears that GH, like BDNF, shares at least some of the signaling mechanisms involved in tetanus-induced LTP. In agreement with this conclusion, we found mutual occlusion between the effects of GH treatment and high-frequency stimulation-induced LTP (Figs. 14 and 15).
Previous studies have shown that GH is required for normal development of the brain (reviewed in Harvey and Hull 2003; Noguchi 1996; Scheepens et al. 2000), and memory deficiency including both short- and long-term memory is a well known syndrome in GH-deficient patients (Aleman et al. 2000; Deijen et al. 1996). Chronic (3-mo) GH supplementation improved attentional performance in adult patients suffering from hypopituitarism with GH deficiency (Oertel et al. 2004). An age-related decrease of GH is associated with defects in spatial memory in animals (van Dam et al. 2000) and may contribute to age-related decline in memory function during normal human aging (Creyghton et al. 2004; van Dam et al. 2000). The hippocampus contains receptors for GH (Burton et al. 1992; Lobie et al. 1993; Mustafa et al. 1994a) and chronic, systemic treatment with GH alters the pattern of expression of NDMA-receptor subunits in the hippocampus (Le Greves et al. 2002), alters GABAA 1 subunit expression, short-term synaptic plasticity (paired-pulse inhibition), and improves hippocampal-dependent spatial learning (Ramsey et al. 2004). The beneficial effects of GH on memory function may result from the alteration of glutamate and GABA-receptor expression during long-term treatment with GH. In addition, our results suggest more immediate effects of GH on hippocampal synaptic function, which might also contribute to the ability of GH to modulate hippocampal-dependent memory function. In addition to GH receptors, GH itself is produced in the hippocampus (Gossard et al. 1987; Hojvat et al. 1982). In a recent study, Donahue et al. (2002) showed upregulation of hippocampal GH mRNA after a hippocampal-dependent learning task (trace eye-blink conditioning). This finding together with our results suggests that endogenous hippocampal GH might act in an autocrine or paracrine manner to enhance or help maintain altered synaptic function during memory formation. GH may therefore influence hippocampal function in multiple ways, over both short and long timescales.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: L. M. Grover, Department of Physiology, Pharmacology and Toxicology, Marshall University School of Medicine, 1542 Spring Valley Drive, Huntington, WV 25704 (E-mail: grover{at}marshall.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Amit T, Barkey RJ, Guy J, and Youdim MB. Specific binding sites for prolactin and growth hormone in the adult rabbit lung. Mol Cell Endocrinol 49: 17–24, 1987.[CrossRef][ISI][Medline]
Anderson NG. Growth hormone activated mitogen-activated protein kinase and S6 kinase and promotes intracellular tyrosine phosphorylation in 3T3-F442A preadipocytes. Biochem J 284: 649–652, 1992.
Andreasen M, Lambert JDC, and Jensen MS. Effects of new non-N-methyl-D-aspartate antagonists on synaptic transmission in the in vitro rat hippocampus. J Physiol 414: 317–336, 1989.
Argetsinger LS, Campbell GS, Yang XJ, Witthuhn BA, Silvennoline O, Ihle JN, and Carter-Su C. Identification of JAK2 as a GHR-associated tyrosine kinase. Cell 74: 237–244, 1993.[CrossRef][ISI][Medline]
Argetsinger LS, Hsu GW, Myers MG Jr, Billestrup N, White MF, and Carter-Su C. Growth hormone, interferon-gamma, and leukemia inhibitory factor promoted tyrosyl phosphorylation of insulin receptor substrate-1. J Biol Chem 270: 14685–14692, 1995.
Bewley TA. Thermal denaturation of human somatotropin, human chorioammotropin and ovine prolactin. Int J Peptide Protein Res 19: 63–66, 1982.[ISI][Medline]
Bjork S, Jonsson B, Westphal O, and Levin JE. Quality of life of adults with growth hormone deficiency: a controlled study. Acta Paediatr Scand Suppl 356: 55–59, 1989.[Medline]
Blanton MG, Lo Turco J, and Kriegstein AR. Whole cell recording from neurons in slices of reptilian and mammalian cerebral cortex. J Neurosci Methods 30: 203–210, 1989.[CrossRef][ISI][Medline]
Burton KA, Kabigting EB, Clifton DK, and Steiner RA. Growth hormone receptor messenger ribonucleic acid distribution in the adult male rat brain and its colocalization in hypothalamic somatostatin neurons. Endocrinology 131: 958–963, 1992.[Abstract]
Collingridge GL, Kehl SJ, and McLennan H. Excitatory amino acids in synaptic transmission in the Schaffer collateral-commissural pathway of the rat hippocampus. J Physiol 334: 33–46, 1983.
Creyghton WM, van Dam PS, and Koppeschaar HP. The role of the somatotropic system in cognition and other cerebral functions. Semin Vasc Med 4: 167–172, 2004.[CrossRef][Medline]
Davies SN and Collingridge GL. Role of excitatory amino acid receptors in synaptic transmission in area CA1 of rat hippocampus. Proc R Soc Lond B Biol Sci 236: 373–385, 1989.[Medline]
Deijen JB, de Boer H, Blok GJ, and van der Veen EA. Cognitive impairments and mood disturbances in growth hormone deficient men. Psychoneuroendocrinology 21: 313–322, 1996.[CrossRef][ISI][Medline]
Donahue CP, Jensen RV, Ochiishi T, Eisenstein I, Zhao M, Shors T, and Kosik KS. Transcriptional profiling reveals regulated genes in the hippocampus during memory formation. Hippocampus 12: 821–833, 2002.[CrossRef][ISI][Medline]
Dreskin SC and Kostyo JL. Acute effects of growth hormone on the function of ribosomes of rat skeletal muscle. Horm Metab Res 12: 60–66, 1980.[ISI][Medline]
Dunwiddie TV and Haas HL. Adenosine increases synaptic facilitation in the in vitro r
