The rostral hypothalamus, especially the preoptic-anterior hypothalamus (POAH), contains temperature-sensitive and -insensitive neurons that form synaptic networks to control thermoregulatory responses. Previous studies suggest that the cyclic nucleotide cGMP is an important mediator in this neuronal network, since hypothalamic microinjections of cGMP analogs produce hypothermia in several species. In the present study, immunohistochemisty showed that rostral hypothalamic neurons contain cGMP, guanylate cyclase (necessary for cGMP synthesis), and CNG A2 (an important cyclic nucleotide-gated channel). Extracellular electrophysiological activity was recorded from different types of neurons in rat hypothalamic tissue slices. Each recorded neuron was classified according to its thermosensitivity as well as its firing rate response to 2–100 μM 8-bromo-cGMP (a membrane-permeable cGMP analog). cGMP has specific effects on different neurons in the rostral hypothalamus. In the POAH, the cGMP analog decreased the spontaneous firing rate in 45% of temperature-sensitive and -insensitive neurons, an effect that is likely due to cGMP-enhanced hyperpolarizing K+ currents. This decreased POAH activity could attenuate thermoregulatory responses and produce hypothermia during exposures to cool or neutral ambient temperatures. Although 8-bromo-cGMP did not affect the thermosensitivity of most POAH neurons, it did increase the warm sensitivity of neurons in other hypothalamic regions located dorsal, lateral, and posterior to the POAH. This increased thermosensitivity may be due to pacemaker currents that are facilitated by cyclic nucleotides. If some of these non-POAH thermosensitive neurons promote heat loss or inhibit heat production, then their increased thermosensitivity could contribute to cGMP-induced decreases in body temperature.
- guanosine monophosphate
- neuronal activity
- soluble guanylate cyclase
- cyclic nucleotide-gated channel
neurons in the rostral hypothalamus, especially in the preoptic-anterior hypothalamus (POAH), are important in the regulation of body temperature. Thermosensitive POAH neurons sense changes in the central core temperature, and these same neurons receive synaptic afferent inputs from skin thermoreceptors (1, 2). This integration of central and peripheral thermal information produces appropriate physiological and behavioral responses that regulate body temperature. Both in vivo and in vitro electrophysiological studies have recorded the activity of POAH neurons during changes in hypothalamic temperature (17, 19, 24, 27). Although the majority of these neurons are considered to be temperature insensitive, at least 20% are classified as warm sensitive because their firing rates significantly increase during hypothalamic warming and decrease during hypothalamic cooling. Thermosensitivity is an intrinsic property of these warm-sensitive neurons, and it is due to ionic channels that produce thermosensitive pacemaker potentials (17, 38). In addition, a small proportion of neurons (i.e., <5%) have the opposite firing rate responses to temperature, and these are classified as cold sensitive (1, 2).
Cyclic guanosine monophosphate (cGMP) appears to be an important mediator in the central control of body temperature. When cGMP analogs are microinjected in the POAH or administered intracerebroventricularly (icv), there are consistent decreases in body temperature in rats (34), guinea pigs (21), rabbits (6, 20), and cats (7). These previous studies used either dibutyryl cGMP or 8-bromo-cGMP (8-Br-cGMP), i.e., analogs that are more membrane permeable and more resistant to phosphodiesterase degradation compared with native cGMP.
The present in vitro electrophysiological study was conducted in rat hypothalamic tissue slices to determine the effect of 8-Br-cGMP on the firing rate and thermosensitivity of different types of neurons in the POAH and other nearby hypothalamic regions. In addition, this study immunohistochemically examined the hypothalamic distribution of cGMP and soluble guanylate cyclase (sGC), which is necessary for cGMP synthesis. Also, since cyclic nucleotide-gated (CNG) channels are highly dependent on cGMP (15, 22, 40), the present study examined the distribution of CNG A2, an important CNG channel.
Preparation of tissue slices for electrophysiological recording.
As in previous studies (8, 9, 17, 39), horizontal hypothalamic tissue slices were prepared from male Sprague-Dawley rats. Each rat was decapitated by a guillotine according to procedures approved by the National Institutes of Health and the Ohio State University Laboratory Animal Care and Use Committee. Following removal of the brain, a tissue block of the hypothalamus was cut, mounted on a vibratome, and sectioned to a thickness of 350 μm. One or two slices were transferred to an interface-recording chamber (23) and allowed to incubate for 1.5–2 h before any recordings were attempted. The chamber was aerated with a humidified 95% O2-5% CO2 gas mixture. The tissue slices were continuously perfused with artificial cerebrospinal fluid (aCSF) consisting of (in mM) 124 NaCl, 26 NaHCO3, 10 glucose, 5 KCl, 2.4 CaCl2, 1.3 MgSO4, 1.24 KH2PO4, and gas saturated with 95% O2-5% CO2. The aCSF was heated to 36–37°C using a thermoelectric Peltier assembly. This thermoelectric assembly also allowed the aCSF and tissue slice to be periodically warmed and cooled to characterize the thermosensitivity of each recorded neuron. The aCSF flowed hydrostatically (1–2 ml/min) into the recording chamber. In addition to normal aCSF, tissue slices were periodically exposed to perfusions of aCSF containing 2- to 100-μM concentrations of 8-Br-cGMP. This concentration range activates CNG ion channels as well as cGMP-dependent protein kinase (15, 32). Valves were used to control the switching between the different perfusion media. Tissue slice temperature was monitored continuously by a thermocouple in the medium below the slice. The air temperature within the recording chamber was also monitored by a thermocouple located above the slices and was maintained at 36–38°C.
Spontaneous action potentials were recorded extracellularly with 1- to 2-μm tip glass microelectrodes filled with 3 M NaCl. Amplified extracellular activity was displayed on a storage oscilloscope, and the oscilloscope trigger level was adjusted to isolate the action potentials from background electrical noise. The criterion for acceptable recordings consisted of spontaneous extracellular action potentials having signal-to-noise ratios greater than 3:1. A ratemeter processed this triggered spike activity into an integrated firing rate of impulses per second (imp/s). Integrated firing rate, tissue slice temperature, and air temperature were recorded on a polygraph and computer using a Digidata 1320A Data Acquisition System and Axoscope software (Axon Instruments).
After the spontaneous firing rate was recorded at 36–37°C (thermoneutrality), the tissue slice temperature was varied 3–5°C above and below thermoneutrality to determine neuronal thermosensitivity. Most neurons were tested over a 32- to 40°C range. Neuronal firing rate was plotted as a function of tissue temperature, and the thermal coefficient or slope (m) of this plot was defined as each neuron's thermosensitivity (imp·s−1·°C−1). Neuronal thermosensitivity was determined over a minimal 3°C temperature range in which the neuron was most temperature sensitive. Figure 1 shows examples of different types of temperature-sensitive and temperature-insensitive neurons recorded from hypothalamic tissue slices. As in previous studies (4, 18, 39, 41), neurons were classified as warm sensitive if they had a positive thermal coefficient of ≥0.8 imp·s−1·°C−1 (see Fig. 1A) or cold sensitive if they had a negative thermal coefficient of at least −0.6 imp·s−1·°C−1 (see Fig. 1B). All other neurons were considered temperature insensitive, and these were subdivided into two classes: moderate-slope temperature insensitive and low-slope temperature insensitive. Moderate-slope temperature-insensitive neurons exhibited small thermally induced changes in firing rate, and their thermosensitivities were <0.8 imp·s−1·°C−1 but >0.2 imp·s−1·°C−1 (see Fig. 1C). Low-slope temperature-insensitive neurons exhibited little or no change in firing rate during a temperature change, and their thermosensitivities ranged between +0.2 and −0.2 imp·s−1·°C−1 (see Fig. 1D). Using similar criteria, previous studies have demonstrated physiological and morphological distinctions between these different neuronal types (1, 18).
Once control firing rate and thermosensitivity were determined, the perfusion medium was switched from control aCSF to an experimental medium to determine the effect of 8-Br-cGMP on the spontaneous firing rate. In some cases, the neuron's thermosensitivity was reevaluated during the 8-Br-cGMP treatment. The experimental medium was then switched back to the control aCSF to demonstrate that the effects of the experimental medium could be reversed. For each recorded neuron, the location of the microelectrode tip was determined with a microscope mounted above the tissue chamber. These locations were plotted on standardized tissue slice maps (8), and the corresponding hypothalamic structure was identified.
Data analysis for electrophysiological studies.
Values are expressed as means ± SE. Mean firing rates (imp/s) were determined over a 1- to 2-min period at 36–37°C and were compared before, during, and after 8-Br-cGMP treatment. As in a previous study (39), the criteria for a change in firing rate were 1) a 10% change in firing rate that was ≥1 imp/s and 2) a firing rate response showing at least partial recovery upon return to control conditions. In all comparisons of firing rate and thermosensitivity, statistical differences were determined by Student's t-tests or ANOVA. When appropriate, post hoc Fisher's least significant difference test determined which comparisons were statistically different. Statistical significance was defined as P < 0.05. All statistical tests were performed using JMP IN software (SAS Institute).
Male Sprague-Dawley rats were anesthetized with pentobarbital sodium (100 mg/kg ip). Once anesthetized, the animals were perfused through the heart with phosphate-buffered saline (PBS) followed by ice-cold 4% paraformaldehyde/0.1 M PBS (fixative). The brains were removed and placed in fixative overnight at 4°C and then transferred to PBS (containing 30% sucrose) until the brains sank. A tissue block containing the hypothalamus was cut and sectioned at 50 μm in either coronal or horizontal planes. Sections containing the POAH were pretreated with 5% normal sheep serum (Jackson ImmunoResearch Laboratories) in PBS containing 0.3% Triton X (PBT) for 2 h at room temperature and then incubated for 48 h at 4°C with constant agitation in various antibody solutions: 1) rabbit polyclonal antibody for sGC (1:2,500; United States Biological, Swampscott, MA), 2) rabbit polyclonal antibodies for cGMP (1:2,500; US Biological), or 3) rabbit polyclonal antibodies for CNG A2 (1:200; Alomone Laboratories, Jerusalem, Israel). All antibodies were diluted in PBT. The sections were then rinsed in PBS and sequentially placed in sheep anti-rabbit IgG (1:500 in PBT; ICN Pharmaceuticals) and rabbit peroxidase antiperoxidase (1:500 in PBT; ICN Pharmaceuticals) for 1 h each at room temperature with constant agitation. Following a final rinse in PBS, the sections were processed using the glucose oxidase procedure (33) to visualize the distribution of sGC, cGMP, and CNG A2. At least three animals for each antibody were used in this study. Tissue sections were examined using a Zeiss bright-field microscope, and the hypothalamic distribution of sGC, cGMP, and CNG A2 was recorded with a Zeiss Axiocam black and white digital camera attached to the microscope.
Two types of controls were used to verify the specificity of the antibodies, and no immunolabeling was observed with either type of control. First, tissue sections were processed in parallel in solutions that did not contain the primary antibodies. Second, absorption controls were conducted where blocking peptides for either sGC (supplied by United States Biological) or CNG A2 (supplied by Alomone Laboratories) were added to their respective primary antibodies at a concentration of 0.02 μg/μl and stored overnight before being applied to tissue sections. In addition, the specificity of the cGMP antibody was established by ELISA techniques (United States Biological), which demonstrated that the antibody recognized the cGMP antigen.
Figure 2 indicates that many neurons in the rostral hypothalamus (including the POAH) contain detectable levels of cGMP, and the synthesis of this cGMP can be mediated by sGC. In addition, some of these neurons express CNG ion channels (i.e., channels containing the CNG A2 subunit) that can be activated by increased levels of intracellular cGMP. Figure 2, A–D, shows cGMP immunostaining in a horizontal section through the rostral hypothalamus containing the POAH. cGMP immunoreactivity is present in cell bodies as well as some processes. Fig. 2, E–H, shows sGC immunostaining in a coronal section of rat hypothalamus containing the medial preoptic area. Figure 2, I–L, shows CNG A2 immunostaining on neuronal cell bodies and processes in a coronal hypothalamic section containing the preoptic area. This immunohistochemical analysis supports electrophysiological evidence that cGMP is a putative mediator capable of influencing neuronal activity in and near the POAH.
Forty-five neurons were recorded in hypothalamic tissue slices and classified according to their thermosensitivity as well as their firing rate response before, during, and after exposure to 8-Br-cGMP. This included 12 warm-sensitive neurons (27%), 17 moderate-slope temperature-insensitive neurons (38%), 14 low-slope temperature-insensitive neurons (31%), and two cold-sensitive neurons (4%). These proportions are similar to those reported previously (1, 9, 18, 39, 41). The mean neuronal firing rate was 10.0 ± 1.4 imp/s, and warm-sensitive neurons had significantly higher mean firing rates (17.2 ± 2.5 imp/s, P < 0.05) than either moderate-slope temperature-insensitive (7.7 ± 2.1 imp/s) or low-slope temperature-insensitive neurons (7.5 ± 2.4 imp/s).
Effects of 8-Br-cGMP on firing rate in hypothalamic neurons.
Neuronal firing rate responses were recorded before, during, and after experimental perfusion with various 8-Br-cGMP concentrations. This included low concentrations (2–5 μM, n = 10), medium concentrations (10–20 μM, n = 25), and high concentrations (40–100 μM, n = 10). There were no statistical differences between these different concentrations in terms of the reported effects on neuronal firing rate. Table 1 summarizes the firing rate effects of 8-Br-cGMP, and Figs. 3–7 provide examples of responsive and unresponsive neurons.
Figure 3, top, shows a record of firing rate and tissue temperature for an anterior hypothalamic neuron whose activity was unaffected by exposure to 20 μM 8-Br-cGMP. When the neuron's firing rate was plotted as a function of temperature (Fig. 3, bottom, 1), the slope of the regression line was −0.1 imp·s−1·°C−1 and the cell was classified as a low-slope temperature-insensitive neuron. This neuronal thermosensitivity did not change appreciably during 8-Br-cGMP (Fig. 3, bottom, 2) nor during washout with control aCSF (Fig. 3, bottom, 3).
Figure 4 shows examples of three different types of POAH neurons that decreased their firing rates during 20- or 50-μM 8-Br-cGMP perfusions. These include a low-slope temperature-insensitive neuron (Fig. 4A), a moderate-slope temperature-insensitive neuron (Fig. 4B), and a warm-sensitive neuron (Fig. 4C). For all three examples, following the cGMP-induced decreased firing rates, each neuron's activity eventually returned toward its original baseline level during the final washout period. The example in Fig. 5 is a different response to cGMP. In this case, Fig. 5 shows a moderate-slope temperature-insensitive neuron that increased its firing during 8-Br-cGMP perfusion. Once again, during the final washout period, this neuron's activity returned toward its original baseline level.
Only two cold-sensitive neurons were recorded and treated with 8-Br-cGMP. One of these neurons was unaffected by 50 μM 8-Br-cGMP (not shown), but the other cold-sensitive neuron significantly decreased its firing rate in response to 5 μM 8-Br-cGMP (Fig. 6, top). This neuron's thermosensitivity was determined to be −0.8 imp·s−1°·C−1 (Fig. 6, bottom, 1). When treated with 8-Br-cGMP the firing rate decreased, but there was no change in the neuron's thermosensitivity (Fig. 6, bottom, 2). During the final perfusion with control aCSF, the neuron's firing rate increased and its thermosensitivity remained unchanged (Fig. 6, bottom, 3). Given the small number of cold-sensitive neurons recorded, no analysis compared cold-sensitive neurons with other neuronal types.
Table 1 summarizes the effect of 8-Br-cGMP on the firing rate in temperature-insensitive and warm-sensitive neurons. In this study, 49% of the recorded neurons exhibited firing rate changes during 8-Br-cGMP exposure, i.e., 35% decreased firing rate (i.e., cGMP-inhibited) and 14% increased firing rate (i.e., cGMP excited). The proportions of cGMP-inhibited neurons were similar for low-slope temperature-insensitive (36%), moderate-slope temperature-insensitive (35%), and warm-sensitive (33%) neurons. The proportions of cGMP-excited neurons were smaller but also similar for low-slope temperature-insensitive (7%), moderate-slope temperature-insensitive (18%), or warm-sensitive neurons (17%). There was no significant difference in the duration of 8-Br-cGMP exposure in those neurons whose firing rates changed during 8-Br-cGMP (15 ± 1 min, n = 21) compared with those neurons whose firing rates did not change (14 ± 1 min, n = 22).
Cyclic GMP increases thermosensitivity in warm-sensitive neurons.
The effect of 8-Br-cGMP on neuronal thermosensitivity was examined in 24 neurons. As an example of cGMP-enhanced thermosensitivity, Fig. 7, top, shows a warm-sensitive neuron recorded in the perifornical region of the lateral hypothalamus. During treatment with 20 μM 8-Br-cGMP, this neuron showed little change in its firing rate, but there was a dramatic increase in its thermosensitivity. During the original control aCSF perfusion, the neuron's thermosensitivity was 0.9 imp·s−1·°C−1 (Fig. 7, bottom, 1); however, this increased to 1.9 imp·s−1·°C−1 during 8-Br-cGMP treatment (Fig. 7, bottom, 2). During the washout period when the perfusate was switched back to control aCSF, the thermosensitivity returned to 0.9 imp·s−1·°C−1 (Fig. 7, bottom, 3).
In Fig. 8A, each neuron's thermosensitivity during 8-Br-cGMP is plotted as a function of its control thermosensitivity before 8-Br-cGMP exposure. Data points that lie along the diagonal dashed line represent neurons that did not change thermosensitivity during perfusion with 8-Br-cGMP. Data points that lie above or below this dashed line represent neurons that increased or decreased, respectively, their thermosensitivity during 8-Br-cGMP. The small square outlined by dotted lines in Fig. 8A, bottom left, contains the entire population of temperature-insensitive neurons (i.e., thermosensitivity <0.8 imp·s−1·°C−1). All data points residing outside this square are warm-sensitive neurons (i.e., thermosensitivity ≥0.8 imp·s−1·°C−1). As a population, temperature-insensitive neurons (n = 15) exhibited very little change in thermosensitivity during treatment with 8-Br-cGMP, whereas warm-sensitive neurons (n = 9) tended to increase thermosensitivity with 8-Br-cGMP. If there had been no net change in the thermosensitivity of the neuronal population, linear regression analysis would show a slope (m) of 1.0. Instead, Fig. 8A shows that the m increased to 1.45, and this was due to an increased thermosensitivity primarily in those neurons displaying the greatest initial thermosensitivity. This is confirmed in Fig. 8B, where the thermosensitivity increased significantly for warm-sensitive neurons during 8-Br-cGMP when compared with temperature-insensitive neurons.
Regional differences in responses to 8-Br-cGMP.
Differences in neuronal responses to 8-Br-cGMP were found when POAH neurons were compared with other hypothalamic neurons located outside the POAH. For this comparison, the total neuronal population of warm-sensitive and temperature-insensitive neurons was subdivided into two separate regions, i.e., those neurons located within the POAH (n = 31) and those hypothalamic neurons located dorsal, posterior, and lateral to the POAH (i.e., non-POAH, n = 12). These non-POAH neurons were recorded in the lateral hypothalamus (n = 7), posterior hypothalamus (n = 3), and dorsomedial or dorsoanterior hypothalamus (n = 2).
In terms of regional sensitivity to the 8-Br-cGMP, POAH neurons were more likely to exhibit firing rate decreases compared with non-POAH neurons. Figure 9A shows that 55% (17 of 31 cells) of POAH neurons changed firing rate during 8-Br-cGMP treatment; specifically, 45% (14 of 31 cells) decreased their firing rate and 10% (3 of 31 cells) increased their firing rate. In contrast, Figure 9B shows that 25% (3 of 12) of non-POAH neurons increased firing rate during 8-Br-cGMP treatment, and only one of the non-POAH neurons (1 of 12) decreased firing rate. Furthermore, Figure 9C indicates that the relative change in firing rate during 8-Br-cGMP was −19 ± 5% for POAH neurons (n = 31) compared with +4 ± 8% for the non-POAH neurons (n = 12), and these changes were significantly different (P < 0.03).
Figure 10 shows the relative change in firing rate for POAH and non-POAH neurons when these populations are further divided into low-slope temperature-insensitive, moderate-slope temperature-insensitive, and warm-sensitive subpopulations. For POAH neurons (Fig. 10A), 8-Br-cGMP tended to decrease the firing rate in all subpopulations, and this effect was most evident in the low-slope temperature-insensitive neurons (−34 ± 9%) and warm-sensitive neurons (−25 ± 10%). On the other hand, non-POAH neurons (Fig. 10B) showed only minor changes in firing rate during 8-Br-cGMP. When POAH and non-POAH subpopulations were compared with each other, the POAH low-slope temperature-insensitive neurons and POAH warm-sensitive neurons exhibited significantly decreased firing rates during 8-Br-cGMP. For the moderate-slope temperature-insensitive neurons, however, there were no differences in the firing rate changes between the POAH (−7 ± 7%) and non-POAH populations (−8 ± 16%).
8-Br-cGMP increases thermosensitivity in non-POAH warm-sensitive neurons.
Although 8-Br-cGMP often decreased firing rates in POAH neurons, it also tended to increase the thermosensitivity of warm-sensitive neurons, particularly neurons located outside the POAH. For the entire neuronal population, Fig. 8 indicates that 8-Br-cGMP increased the thermosensitivity of warm-sensitive neurons but had little effect on the thermosensitivity of temperature-insensitive neurons. Figure 11 indicates that this is true predominantly for non-POAH neurons, such as the example in Fig. 7 that shows a non-POAH neuron in the perifornical region near the lateral hypothalamus. Figure 11 plots the change in thermosensitivity during 8-Br-cGMP exposure for POAH (n = 14) and non-POAH neurons (n = 10). Figure 11A shows that most POAH neurons did not change their thermosensitivity (m = 1.2 for all POAH neurons) during 8-Br-cGMP, whereas most warm-sensitive non-POAH neurons increased their thermosensitivity (m = 1.8 for all non-POAH neurons) during 8-Br-cGMP (Fig. 11B). Figure 12 shows that the cGMP-induced increase in thermosensitivity was only 0.1 ± 0.1 imp·s−1·°C−1 for POAH warm-sensitive neurons (n = 5); however, it was 0.8 ± 0.2 imp·s−1·°C−1 for non-POAH warm-sensitive neurons (n = 4), and these changes were significantly different (P < 0.05).
cGMP modulates body temperature.
The rostral hypothalamus, especially the POAH, is an important neural region controlling body temperature by heat loss, heat retention, and heat production responses (1). Whole animal studies suggest that cGMP plays a role in the neural control of these responses since changes in hypothalamic cGMP consistently produce changes in body temperature (6, 7, 20, 21, 34). When hypothalamic cGMP levels increase, body temperature decreases, and when hypothalamic cGMP levels decrease (via inhibition of sGC), body temperature increases. In those previous studies, direct hypothalamic microinjections implicate the POAH as a particularly sensitive region for cGMP signaling.
Supporting the thermoregulatory role of hypothalamic cGMP, several studies have examined the central effects of nitric oxide (NO) on body temperature. NO binds to and activates sGC and stimulates cGMP production (10). As might be expected, POAH microinjections of sodium nitroprusside (SNP; an NO donor) produce decreases in body temperature (34, 36). Conversely, other studies examined the thermoregulatory effects of inhibitors of either NO synthase (NOS) or sGC. NOS is the enzyme that synthesizes NO, which in turn activates sGC to stimulate cGMP production. Inhibition of NOS or sGC should decrease cGMP production and reduce intracellular cGMP concentrations. When NOS was inhibited by icv NG-nitro-l-arginine methyl ester (L-NAME) (3, 11) or when sGC was inhibited by icv 1H-[1,2,4]oxadiazolo[4,3]-quinoxalin-1-one (35), body temperature increased. Thus, body temperature is inversely related to hypothalamic intracellular cGMP.
Cyclic GMP production and utilization.
The present immunohistochemistry study indicates that many rostral hypothalamic neurons possess the cellular mechanisms to produce and utilize cGMP. The coronal and horizontal sections in Fig. 2 show cell bodies and neuronal processes that contain both cGMP, as well as sGC, which is necessary for cGMP production. Other immunohistochemical studies have identified cGMP-positive neurons in the preoptic area of rats (12, 36). Previous in situ hybridization studies in rats detected guanylate cyclase mRNA and sGC in the medial preoptic, supraoptic, and paraventricular nuclei (5, 16). The sGC immunoreactivity in preoptic cell bodies confirms that these neurons are capable of synthesizing cGMP in response to NO.
The observed changes in hypothalamic neuronal activity may be due to cGMP-dependent protein kinases and protein phosphorylation (13, 14, 28, 30, 37). In addition, Fig. 2, I–L, shows rostral hypothalamic neurons that contain CNG A2 channels, and these channels are also likely candidates for the observed neuronal responses to 8-Br-cGMP. Using in situ hybridization, a previous study detected CNG A2 mRNA in rat supraoptic, paraventricular, and preoptic hypothalamic nuclei (25). In the present study, CNG A2 immunoreactivity was detected in POAH neuronal cell bodies and processes. This confirms that preoptic neurons express CNG channels capable of responding to cGMP. CNG channels and cGMP serve important physiological roles that could explain the observed changes in neuronal activity (40). First, CNG channels can depolarize some presynaptic neurons via inward Ca2+ and Na+ currents in response to cyclic nucleotides (40). This depolarization can trigger action potentials and release neurotransmitters that either inhibit or excite postsynaptic neurons. Second, CNG channels allow intracellular Ca2+ entry and the activation of Ca2+-dependent pathways. This includes Ca2+-activated K+ currents causing transient hyperpolarizations that prolong interspike intervals and decrease spontaneous firing rates (15, 22). Third, in some cells, cGMP can enhance outward K+ leak currents (26). If this occurs in POAH neurons, it would cause hyperpolarization and reduced firing rates, particularly in temperature-insensitive neurons that have strong K+ leak currents (38). Therefore, all three of the above possibilities offer explanations for the reduced firing rates in many neurons in the present study.
Finally, the present study also found that cGMP increased the thermosensitivity of some neurons, particularly non-POAH neurons. A recent study indicates that hyperpolarization-activated CNG channels (HCN) or hyperpolarization-activated cation currents contribute to hypothalamic neuronal warm sensitivity by controlling the depolarizing prepotentials (DPPs) or pacemaker potentials that produce the action potentials (38). Since cyclic nucleotides facilitate HCN channels (29), this may be a key mechanism by which cGMP enhances the thermosensitivity of warm-sensitive neurons, particularly in the non-POAH neurons reported in the present study.
Regional differences in neuronal responses to 8-Br-cGMP.
Although Fig. 2 suggests that cGMP, sGC, and CNG A2 are present in many neurons throughout the rostral hypothalamus, a whole animal study in rats indicates that there are regional differences in response to 8-Br-cGMP hypothalamic microinjections (34). Specifically, microinjections in the anteroventral preoptic area produced significant decreases in body temperature; however, nearby hypothalamic microinjections outside this region had little effect on body temperature. The present electrophysiological study also found regional differences in neuronal responses to 8-Br-cGMP. The cGMP analog affected the firing rates of 55% of the POAH neurons but only 33% of non-POAH neurons. Moreover, 8-Br-cGMP inhibited 45% of the POAH neurons, but in contrast, it inhibited only 8% of the non-POAH neurons. Thus, cGMP appears to have different effects not only on different neuronal types, but also in different hypothalamic regions.
In support of the present findings, an extracellular electrophysiological study by Schmid et al. (31) used rat hypothalamic tissue slices to record neuronal responses to 10–100 μM SNP, an NO donor that increases intracellular cGMP. In that previous study, 17 spontaneously firing POAH neurons were characterized by their thermosensitivity and their response to SNP. This included seven warm-sensitive neurons and 10 temperature-insensitive neurons. SNP inhibited the firing rates of all 17 POAH neurons. Moreover, in two of these neurons, the inhibition could be mimicked by subsequent exposures to 100 μM 8-Br-cGMP or the NO donor 3-morpholinosydnonimine (SIN-1). It should be noted that this previous study also recorded five non-POAH neurons nearby in the diagonal band of Broca, and none of these neurons responded to SNP. Therefore, both the previous study and the present study suggest that cGMP preferentially decreases firing rates in POAH neurons. If some of these neurons control thermoregulatory responses, the cGMP-induced attenuation of POAH neuronal activity may explain the observed decreases in body temperature, at least during exposures to cool and neutral ambient temperatures. As mentioned above, cGMP-induced inhibition could be due to either synaptic inhibition or hyperpolarizing K+ currents. To determine whether synaptic or cellular mechanisms underlie this response, future intracellular recording studies should compare the effects of cGMP analogs on the synaptic and intrinsic activity of POAH and non-POAH neurons.
In addition to regional differences in firing rate responses, there are also regional differences in the effects of 8-Br-cGMP on neuronal thermosensitivity. In general, warm-sensitive hypothalamic neurons increased thermosensitivity during 8-Br-cGMP treatments, whereas temperature-insensitive neurons did not change their thermosensitivity. When these neurons were further divided into POAH and non-POAH populations, it was the non-POAH warm-sensitive neurons that exhibited a significant increase in thermosensitivity with 8-Br-cGMP. As indicated above, it is possible that this increased thermosensitivity is due to CNG channels that determine pacemaker potentials or DPPs in warm-sensitive neurons. Our previous studies have identified DPPs as the primary determinants of neuronal warm sensitivity (4, 17). Major components of these thermosensitive prepotentials are HCN or hyperpolarization-activated cationic channels that provide the inward Na+ currents of pacemaker potentials (38). Since HCN channels are gated by cyclic nucleotides, it might be predicted that cGMP enhancement of HCN currents would lead to an increased thermosensitivity of pacemaker potentials and, subsequently, an increased firing rate thermosensitivity. It remains for future studies to determine whether there are regional differences in CNG channels and HCN currents that explain the responses of different populations of hypothalamic neurons.
Perspectives and significance.
Since central microinjections of cGMP analogs produce hypothermia in several species, the present electrophysiological study examined 8-bromo-cGMP's effect on different types of neurons in rat hypothalamic tissue slices. Cyclic GMP-induced changes in hypothalamic neuronal firing rates and thermosensitivies may underlie the central mechanisms for thermal adaptation and changes in thermal set points. H. T. Hammel's popular neuronal model (2) suggests that set point temperature for all thermoregulatory responses is determined by the balance of excitatory and inhibitory inputs from warm-sensitive and temperature-insensitive neurons synapsing upon heat loss and heat production effector neurons. In this model, regulated body temperature would be lowered either by decreasing the firing rates of temperature-insensitive neurons or by increasing the firing rate thermosensitivities of warm-sensitive neurons. In the present study, cGMP decreased the firing rates of many POAH neurons. cGMP can enhance outward K+ currents such as calcium-activated K+ currents and TREK potassium leak currents (26, 40), and this would hyperpolarize neurons, especially temperature-insensitive neurons having strong K+ leak currents (38). Moreover, the cGMP analog increased the thermosensitivity of hypothalamic warm-sensitive neurons but not temperature-insensitive neurons. Neuronal warm sensitivity is due to thermally dependent pacemaker potentials that are partially determined by CNG HCN channels. Therefore, enhancement of HCN currents could lead to increased thermosensitivity of pacemaker potentials and, subsequently, increased firing rate thermosensitivity.
This research was supported by National Institute of Neurological Disorders and Stroke Grants NS-14644 and NS-045758. C. L. Wright is a Presidential Dissertation Fellow and a Medical Scientist Fellow at the Ohio State University. Presently, Dr. Wright is in the Department of Radiology at the Ohio State University. J. A. Boulant is Hitchcock Professor of Environmental Physiology.
We express our sincere gratitude to M. L. Kaple for helpful assistance in this work.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society