Transient receptor potential vanilloid 4 (TRPV4) is one member of the TRP superfamily of nonselective cation channels. Recently, the possibility has been raised that TRPV4 is an osmoreceptor, because it is found in the circumventricular organs where osmoreceptors are supposed to be distributed and because it is sensitive to osmotic pressure in in vitro experiments. In addition, TRPV4 knockout mice have abnormal osmosensitivity. In this study, effects of 4α-phorbol 12,13-didecanoate (4α-PDD), a TRPV4 agonist, on drinking behavior were examined to investigate roles for TRPV4 as an osmoreceptor in vivo in wild-type animals. Intracerebroventricular injections of 4α-PDD inhibited water intake under normal conditions in both light and dark periods of the day, after food deprivation, and after administration of angiotensin II. However, this drug did not influence increased water intake after administration of a hypertonic solution or after water deprivation that significantly increased plasma osmolality. Locomotor activity of the 4α-PDD-injected group decreased slightly compared with that of the vehicle-injected group; however, sweet taste, food intake, and body temperature were not different between the two groups. The antidipsogenic effects of 4α-PDD were blocked by preinjection into the ventricle of TRPV4 antagonists such as ruthenium red or gadolinium. These findings suggest that TRPV4 regulates drinking behavior under certain conditions, and the regulation interacts with the angiotensin II pathway.
- water intake
- transient receptor potential vanilloid 4
homeostasis of osmotic pressures of body fluids is maintained on the basis of a balance between water intake and excretion. Osmoreceptors in the central nervous system (CNS) perform a key step in osmoregulation, a step that detects changes in osmotic pressures of body fluids. Osmoreceptors are suggested to be distributed in the circumventricular organs, especially the anterior wall adjacent to the third ventricle, the hypothalamic paraventricular and supraoptic nuclei where the cell bodies of vasopressin-containing neurons localize, and the subfornical organ (3). However, molecular identification of the osmoreceptor is not clarified.
The transient receptor potential (TRP) superfamily of nonselective cation channels are known to be present in many species from worms to man (5). Of six subfamilies, some members of the TRPV and TRPM subfamilies are temperature sensitive, although each has a different range of temperature sensitivity. Therefore, it is suggested that TRP channels of the CNS are involved in the regulation of body temperature, whereas those of the skin are involved in the regulation of thermosensation (17). Recently, Liedtke et al. (12) found that TRPV4, one member of the TRPV subfamily, is sensitive to osmotic pressure, as well as to temperature. TRPV4 is activated by hypotonic stimuli, resulting in an increase in intracellular Ca2+ concentration ([Ca2+]i). In addition, this channel is found in the circumventricular organ of the CNS (7, 12). Therefore, it has been suggested that TRPV4 functions as an osmoreceptor in the CNS. Since then, many studies have been published concerning TRPV4, and most of them were carried out using an electrophysiological technique or [Ca2+]i measurement method in TRPV4-transfected cells.
In 2003, two laboratories succeeded in generating TRPV4 knockout (KO) mice and showed that they have abnormal osmosensitivity (13, 14). Mizuno et al. (14) demonstrated that the urinary volume and ion excretion, water intake, and vasopressin release of the KO mice are normal under basal conditions, although hypertonic stimuli cause an increased release of vasopressin in the KO mice compared with wild-type (WT) mice. However, the KO mice are supposed to show some abnormality even under basal conditions if, as they point out, TRPV4 is constitutively activated by body temperature. In addition, it remains unexplained why KO and WT mice show similar responses to hypotonic stimuli. On the other hand, KO mice from the laboratory of Liedtke and Friedman (13) have shown abnormalities in plasma osmolality, water intake, and/or vasopressin release compared with WT mice under basal conditions and after both hyper- and hyposmotic stimuli. Concerning vasopressin release, the two kinds of KO mice show opposite responses to hypertonic stimuli. These discrepancies make it difficult to understand a role of TRPV4 as an osmoreceptor in the KO mouse phenotype. Therefore, it is necessary to investigate the role of TRPV4 as an osmoreceptor in WT animals. In this study, we examined the effect of the TRPV4 agonist 4α-phorbol 12,13-didecanoate (4α-PDD) (25), microinjected into the lateral ventricle, on the water intake of WT animals.
MATERIALS AND METHODS
Animals and surgical procedure.
Experiments were carried out according to the Guidelines for Animal Care and Use of Nagoya City University. The experimental protocol was approved by the Institutional Animal Care and Use Committee of Nagoya City University Graduate School of Medical Sciences. The detailed procedure is described elsewhere (21–23). We used male Wistar rats (9 wk old; SLC, Hamamatsu, Japan), which were housed in the animal center of Nagoya City University under controlled conditions (temperature, 23 ± 1°C; humidity, 65%; and a 12:12-h light-dark cycle with lights on at 0800 and off at 2000). After the rats were anesthetized with pentobarbiturate (40 mg/kg body wt ip), a guide cannula (AG-8; Eicom, Tokyo, Japan) was implanted into the right lateral ventricle and fixed to the skull with dental cement and small screws according to coordinates provided by Konig and Klippel's atlas (11). During the 1-wk postoperative recovery period, rats were acclimated to handling and to the experimental cage used for drug administration. We cannulated a total of 140 rats, 18 of which were removed from the study because of weight loss in 6 rats after the operation and because of cannula misplacement in 12 rats.
Measurements of water and food intakes.
Measurements were carried out under one of the experimental conditions shown in Table 1: 1) food deprivation (food removed on the previous day; no access for 16 h); 2) normal conditions during light period without treatment; 3) normal conditions during dark period without treatment; 4) water deprivation (drinking water removed on the previous day; no access for 16 h); 5) hypertonic NaCl administration (intragastric administration of 4 ml of 0.5 M NaCl carried out just before the rat was returned to its home cage; free access to food and drinking water); and 6) sweet drinking (8% sucrose given to rats for 3 h from 1200 to 1500 for 4 days, and intakes of sucrose solution were measured at 1500). Drug was administered at 1150 on the fourth day. On the day of the experiment, drugs were microinjected into the lateral ventricle by using a microinjection cannula inserted into the guide cannula. The microinjection cannula was connected via a polyethylene tube to a microsyringe containing two drug solutions. These solutions were separated by a small air bubble. In addition, 1 μl of sterilized saline was loaded between the solutions with air bubbles. The total volume of the drug solutions was 10–13 μl. The interval between the first and second drug microinjections was 30 min. Five minutes after the second injection, the microinjection cannula was switched to a dummy cannula, and the rats were returned to their home cages. Water and food intakes were measured 0.5, 1.0, 1.5, 2.0, 3.0, and 4.0 h after administration of the second drug. In one animal, two measurements were carried out with an interval of at least 1 wk in one experimental series. The drug combinations were randomly selected. However, when effects of 4α-PDD on ANG II-induced drinking behavior was examined under normal conditions (Table 1, group 2b), three drug combinations were administered in one animal with an interval of 1 wk between them and in the following order: ANG II + vehicle, ANG II + 4α-PDD, and ANG II + vehicle.
Measurement of body temperature.
When the guide cannula was implanted into the ventricle, a telemetry transmitter (Minimitter; Eicom) was embedded in the abdomen. Body temperature was measured 3 h after drug administration under the condition of group 2a in Table 1. In one animal, vehicle or 4α-PDD was randomly microinjected into the ventricle with an interval of at least 1 wk between injections.
Measurement of locomotor activity.
Locomotor activity was measured as the number of sensor beam interruptions using an activity monitor (Automex II; Columbus Instruments, Columbus, OH) under the condition of group 2a in Table 1. In one animal, vehicle and 4α-PDD were randomly microinjected into the ventricle with an interval of at least 1 wk.
Measurement of plasma osmolality.
Plasma osmolality was measured just after vehicle injection under the conditions of groups 1 and 4 in Table 1, 5–10 min after vehicle injection and free access to food under the conditions of group 1 (after food intake without water intake), 0.5 h after vehicle injection under the conditions of groups 2 and 5, and 2 h after vehicle injection under the condition of group 3 in Table 1. After decapitation of the rat, blood was collected into an ice-cold tube containing heparin and centrifuged at 3,000 rpm and 4°C. Plasma osmolality was measured using the freezing-point depression method (Osmostat OM-6020; Kyoto Daiichi Kagaku, Kyoto, Japan).
All of the data are expressed as means ± SE. Statistical analyses were carried out using one-way or two-way ANOVA followed by Fisher's protected least significant difference (PLSD) test (post hoc test) or Student's t-test when appropriate (StatView 4.0; Macintosh). Differences were considered significant if the P value was <0.05.
4α-PDD was obtained from Alexis Biochemicals (San Diego, CA); ruthenium red (RR) was obtained from Latoxan (Valence, France); and gadolinium (Gad) and ANG II were obtained from Sigma (St. Louis, MO). The chemicals used were the highest grade. 4α-PDD was dissolved in DMSO and diluted with sterile physiological saline (Otsuka Chemicals, Tokyo, Japan) as used. The final concentration of DMSO was 8%. The other drugs were dissolved in sterile physiological saline.
Effects of 4α-PDD on water intake.
Food-deprived rats drink more water than normally fed rats once the food supply is resumed after food deprivation. Intracerebroventricular (ICV) injections of 4α-PDD (20 μg) significantly inhibited this increased water intake, although food intake was not influenced (Fig. 1, A–C). Water and food intakes under food deprivation conditions in Fig. 1 were simultaneously measured in the same animals. The 4α-PDD-induced antidipsogenic effect was dose dependent [water intakes for 0–0.5 h after injection: vehicle, 4.0 ± 0.4 g, n = 9; 20 μg 4α-PDD, 1.1 ± 0.4 g, n = 8; 10 μg 4α-PDD, 2.6 ± 0.3 g, n = 4; 2 μg 4α-PDD, 4.4 ± 1.4 g, n = 5; P < 0.05, 20 μg 4α-PDD vs. vehicle or 2 μg 4α-PDD groups by Fisher's PLSD test after one-way ANOVA: F(3, 22) = 4.12, P < 0.05]. Pretreatments with RR (1 μg) and Gad (1 μg), two TRPV4 antagonists (8), injected into the ventricle completely blocked 4α-PDD-induced effects (Fig. 1, A and B). ICV injection of 1 μg of either RR or Gad into the ventricle did not influence food and water intakes (data not shown). However, at ∼5 min after administration of RR, the bodies of 2 of 20 rats shook a few times, and locomotor activity seemed to decrease. These rats recovered after 30 min and took in the same volume of food and water as the vehicle-injected rats.
The antidipsogenic effects induced by 4α-PDD also were observed under the normal light-dark condition without any treatment (Fig. 1, D–F). The 4α-PDD-induced effect during the dark period was blocked by RR (1 μg) (Fig. 1F). Food intake during the dark period after 4α-PDD administration did not change compared with that after vehicle administration (4α-PDD: 14.8 ± 0.9 g, n = 6; vehicle: 16.0 ± 0.8 g, n = 6). RR itself did not influence water or food intakes during the dark period.
ANG II is well known to be one of the most powerful dipsogenic agents. Increased water intake after ICV injection of ANG II (100 ng) was completely inhibited by 4α-PDD in five of nine rats (Fig. 2). After 1 wk, when ANG II alone was injected into the same rats once more, this induced the dipsogenic effects to the degree similar to the first effects.
After a 16-h deprivation of water or gastric administration of hypertonic NaCl solution, animals increased their ingestion of water. However, water intakes for 4 h after injection were not different between the 4α-PDD- and the vehicle-injected groups (Fig. 3).
Effects of 4α-PDD on other behavior.
Because drinking behavior also is influenced by various factors such as locomotor activity, taste aversion, and body temperature, these factors were examined after 4α-PDD administration (20 μg). Locomotor activity was slightly, but significantly, decreased after administration of 4α-PDD compared with after administration of vehicle (Table 2). To examine changes in taste aversion or illness/malaise after 4α-PDD administration, we measured how much sucrose solution rats drank. As shown in Fig. 4A, ICV injections of 4α-PDD did not influence the intake of sucrose solution. Neither body temperature after 4α-PDD administration was different from those after vehicle administration (Fig. 4B).
Table 3 shows plasma osmolality under experimental conditions for water intake measurement. Osmolality increased after a 16-h water deprivation period and after gastric administration of hypertonic solution, when water intake was not changed by 4α-PDD (Fig. 3). On the other hand, under a 16-h food deprivation, the normal conditions, and an ANG II-pretreated condition, the antidipsogenic effect of 4α-PDD was shown and osmotic pressure did not change. Plasma osmolality under 16-h food deprivation condition tended to increase after some food intake (1.4 ± 0.3 g, n = 8) without water intake (5–10 min after the rat was given free access to food); however, this was not statistically significant compared with osmolality during light and dark periods.
We found that 4α-PDD microinjected into the lateral ventricle inhibits water intake under certain conditions such as normal conditions during light and dark periods, after food deprivation, and during an ANG II-induced thirst condition. Sweet taste, food intake, and body temperature did not differ between the 4α-PDD- and vehicle-injected groups. Locomotor activity slightly decreased after 4α-PDD administration; however, this decrease is probably not the cause of the antidipsogenic effect. Because food and sucrose intakes were not influenced by 4α-PDD administration despite a decrease in locomotor activity, the 4α-PDD-administered animals did not show taste aversion or illness/malaise. Therefore, it is presumed that the 4α-PDD-induced antidipsogenic effect is not a secondary effect. In addition, the effects were abolished by pretreatment with RR and Gad. Because 4α-PDD is an agonist of TRPV4 and RR and Gad are antagonists of TRPV4, these findings show that the antidipsogenic effect involves TRPV4.
TRPV4 was shown to be sensitive to hyposmotic stimuli in an in vitro experimental system and to be present in the osmoreceptor-containing circumventricular region by immunohistochemical studies (12). Antidipsogenic effects of 4α-PDD were not observed under any in vivo condition. Indeed, 4α-PDD did not inhibit increased water intake following water deprivation or intragastric administration of hypertonic solution. These conditions significantly increased plasma osmotic pressure. We hypothesized that neurons expressing TRPV4 detect decreased osmotic pressure and regulate water balance under physiological conditions in which animals detect as little as a 1–2% change from normal osmolality; this results in thirst and water intake (1). TRPV4 in the CNS may play an important role as an osmoreceptor to detect hyposmolality and to stop water intake under physiological conditions.
Large changes in plasma osmolality influence a variety of factors, including 1) ion/substrate concentration, 2) cell volume, 3) membrane stretch strength, and 4) membrane structure/composition (10, 24, 27), which are followed by activation of intracellular signaling, and ion channels, among others. The Nax channel was found to be a Na+ sensor in the CNS and influences drinking behavior (9). Bourque and colleagues (24, 27) demonstrated that osmotic changes affect the length of the cell membranes, cell volume, and activity of ion channels and then promote vasopressin release, resulting in the regulation of osmolality of body fluids. Activation of mitogen-activated protein kinases is well known under hypertonic conditions (2, 28). Therefore, multiple mechanisms may operate powerfully during large osmotic pressure changes. We suppose that the effect of TRPV4 is masked or that TRPV4 is not involved in the regulation under these conditions.
Liedtke et al. (13) demonstrated that KO mice decrease drinking water volume after hypertonic NaCl stimuli, compared with WT mice. This finding suggests that WT mice drink water as mediated through activation of TRPV4 under hypertonic conditions, although under hypotonic conditions, TRPV4 is activated and WT mice do not drink water. They proposed a hypothesis that there is another molecule presented in vivo that responds to hypertonic stimuli, as well as to hypotonic stimuli. This phenomenon is inconsistent with our result that the TRPV4 agonist did not affect water intake after hypertonic NaCl stimuli. The presumed reasons are as follows. 1) Because TRPV4 in WT animals is fully activated by hypertonic stimuli, agonist administration will not be able to produce an additional effect any longer. 2) KO mice are not only WT mice without TRPV4; the regulatory mechanisms in KO mice are considerably disturbed directly and indirectly by the absence of TRPV4 because they become adapted to this condition since embryos. Two kinds of KO mouse phenotypes reported from the laboratories of Lietdke et al. (13) and Mizuno et al. (14) are different from each other. 3) From the whole body of KO mice, all of the target molecules are deficient; therefore, there may be a discrepancy between the KO mouse phenotype and an agonist/antagonist-induced effect in WT mice after the drug affects limited sites.
Vasopressin release is also regulated by plasma osmolality. Vasopressin controls urine volume, which is the largest component of water excretion from the body, and plays an important role in the regulation of body fluid osmolality. The presence of osmoreceptors on the cell bodies of vasopressin-containing neurons is well known (3, 24). Electrophysiological studies demonstrated that they are nonselective cation channels and are activated by hypertonic, but not hypotonic, stimulation (3, 18). Therefore, the osmoreceptors on the cell bodies of vasopressin-containing neurons may be a different molecule from TRPV4. However, in TRPV4 KO mice, the regulatory mechanisms of vasopressin secretion are disturbed, although effects of TRPV4 on vasopressin secretion are controversial (13, 14). TRPV4 may possibly indirectly regulate vasopressin secretion through neurons expressing TRPV4. To elucidate roles for TRPV4 in the regulatory mechanisms of osmolality homeostasis, further in vivo experiments on vasopressin release are needed in WT animals.
TRPV4 in the choroid plexus (CP) is the first candidate for the action site of 4α-PDD in this study. TRPV4 in the CP is the nearest target for agonist molecules injected into the ventricle rather than other sites, such as the vascular organ of the lamina terminalis (OVLT) and the subfornical organ, where immunohistochemical studies (12–14) show the localization of TRPV4 in the CNS. The CP is known to generate the cerebrospinal fluid (CSF). Probably, TRPV4 in the CP detects osmotic pressure and regulates ion components in the CSF, resulting in control of water balance in vivo. In fact, there is some evidence that changes in the osmotic pressure/sodium concentration of the CSF regulate water intake (1, 16, 19). The organs around the anteroventral third ventricle are also important, because we found an interaction of TRPV4 and ANG II in the regulation of drinking water. All of the organs are needed for ANG II-induced dipsogenic effects after injection into the ventricle (4). Liedtke et al. (13) suggested that the OVLT is a key region for TRPV4 in osmotic regulation.
TRPV4 channels are also sensitive to heat from 30 to 42°C (7, 17). Therefore, at normal rat body temperature, the TRPV4 channel is activated, but not fully. Electrophysiological studies show that additional stimuli, for example, an agonist or osmotic stimuli, probably open TRPV4 channels synergistically, not additively (6). In in vitro studies of TRPV4-expressed cells, TRPV4 is activated by protein kinase C and tyrosine kinase (6, 15, 26). Regulation by [Ca2+]i is complex; ordinary increases activate TRPV4, but excessive increases inhibit TRPV4 (20).
In addition, our study showed an interaction between ANG II and TRPV4 pathways. It is suggested that TRPV4 is present downstream from ANG II signaling, because the TRPV4 agonist inhibited the dipsogenic effects of ANG II. Activating ANG II signaling may close TRPV4 directly or indirectly. In vitro experiments using TRPV4-transfected cells have shown that TRPV4 is closed by temperatures <30°C (7, 17), hyperosmolality (12), and excessive increases in [Ca2+]i (20). So far, there is no evidence that these mechanisms are linked to ANG II signaling. One possibility is that ANG II produces excessive increases in [Ca2+]i and then promotes protein/peptide phosphorylation, resulting in TRPV4 inactivation. In general, phosphorylation is involved in gating mechanisms of various ion channels.
Our study showed that only one-half of the ANG II-induced dipsogenic effects were inhibited by a TRPV4 agonist. We suppose that TRPV4 is not expressed on all the neurons downstream of ANG II signaling. Distribution of TRPV4 in the hypothalamus, especially along with ANG II, needs to be studied in detail. ANG II is well known to be one of the strongest dipsogenic drugs. In addition, ANG II also detects sodium concentrations in body fluids and regulates vasopressin secretion (1). Therefore, it is suggested that TRPV4 is present downstream of ANG II signaling and plays a very important role in the regulation of osmolality homeostasis.
We conclude that TRPV4 is involved in physiological regulation, via drinking behavior, of the osmotic pressure of body fluids. ANG II signaling is suggested to be present upstream of this regulation. Hyperosmolality-induced thirst is not influenced by activation of TRPV4.
This study was partly supported by grant-in-aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (no. 16659060) and from Salt Science Foundation (no. 0336; Tokyo, Japan).
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 © 2006 the American Physiological Society