Angiotensin IV elevates oxytocin levels in the rat amygdala and produces anxiolytic-like activity through subsequent oxytocin receptor activation
- First Online:
- Cite this article as:
- Beyer, C.E., Dwyer, J.M., Platt, B.J. et al. Psychopharmacology (2010) 209: 303. doi:10.1007/s00213-010-1791-1
- 228 Views
The effects of angiotensin (Ang) IV result from binding to a constitutively active metallopeptidase known as the AT4 receptor (or oxytocinase/insulin-regulated membrane aminopeptidase). While in vitro evidence indicates that Ang IV inhibits the peptidase activity of AT4 receptors, leading to increases in the concentrations of several peptides, including oxytocin, the consequence of inhibiting AT4 peptidase activity in vivo remains unresolved.
Microdialysis coupled to immunoassay techniques revealed that systemic and intra-amygdala injection of Nle-Ang IV, a metabolically stable derivative of Ang IV, significantly elevated extracellular levels of oxytocin in the rat amygdala. Based on earlier reports describing the anxiolytic-like effects of oxytocin, we investigated whether disrupting AT4 peptidase activity would yield similar responses. In the mouse four-plate test, acute treatment with either Nle-Ang IV or LVV-hemorphin-7, a related AT4 receptor ligand, elicited significant increases in the number of punished crossings. These behavioral responses were comparable to the anxiolytic-like effects of oxytocin and to the standard anxiolytic agent, chlordiazepoxide. Cotreatment with either the AT4 receptor antagonist, divalinal, or the selective oxytocin receptor antagonist, WAY-162720, reversed the anxiolytic-like effects of Nle-Ang IV, while combining ineffective doses of Nle-Ang IV and oxytocin increased the number of punished crossings in this assay. Conversely, Nle-Ang IV and LVV-hemorphin-7 were inactive in the mouse tail suspension test of antidepressant activity. These findings represent the first in vivo demonstration of the peptidase activity of AT4 receptors, confirm the anxiolytic-like properties of Ang IV, and reveal a unique and previously uncharacterized relationship between AT4 and oxytocin receptor systems.
KeywordsAT4 receptorsNle-angiotensin IVLVV-hemorphin-7DivalinalOxytocinAnxiety
Angiotensin (Ang) IV (angiotensin 3–8; VTIHPP) is an endogenous hexapeptide purported to be involved in a number of diverse functions, including regulating heart and renal vasculature, cell growth, and biological processes associated with learning and memory (reviewed in Chai et al. 2004; von Bohlen und Halbach 2003). The central nervous system (CNS) effects of Ang IV result from binding to a type II, membrane-spanning protein that is a constitutively active metallopeptidase called the AT4 receptor—also known as oxytocinase and insulin-regulated membrane aminopeptidase (Keller et al. 1995; Swanson et al. 1992). Within the brain, AT4 receptors are located in areas distinct from other related Ang receptors (namely AT1 and AT2) and are highly expressed in the hippocampus, entorhinal, prefrontal, and insular cortices and moderately to highly expressed in the substantia nigra, hypothalamus, and amygdala (Chai et al. 2004; von Bohlen und Halbach 2003).
While the CNS functions of AT4 receptors are not well understood, the constitutively active peptidase activity of this receptor has been described. Evidence suggests that ligands with high affinity for AT4 receptors sufficiently inhibit the catalytic activity of these receptors resulting in an increase in the concentrations of various peptide substrates—including but not limited to—oxytocin (Herbst et al. 1997; Matsumoto et al. 2000). In fact, it is postulated that interactions between the AT4 and oxytocin systems are responsible for the behavioral observations produced by administration of Ang IV and its synthetic analogs in animal models (Albiston et al. 2003; Kovacs and De Wied 1994). For example, central administration of Ang IV and LVV-hemorphin-7, a related endogenous peptide known to inhibit the peptidase activity of AT4 and produce behavioral effects similar to Ang IV (Chai et al. 2000), improves performance in various preclinical memory paradigms (Lee et al. 2004), and these behavioral responses are mimicked by studies reporting that administration of oxytocin at low doses enhances social recognition memory (Popik et al. 1996) and restores memory deficits in oxytocin knockout mice (Ferguson et al. 2001). These nootropic findings provide the foundation for a popular hypothesis that the biological effects of Ang IV result from blocking the constitutively active peptidase activity of AT4 receptors, which subsequently results in elevated levels of oxytocin in the CNS (see Discussion in Gard et al. 2007).
Although these potential neurochemical interactions are intriguing, there are currently no published reports demonstrating that Ang IV results in elevations of these neuropeptides in the CNS of freely moving animals. Accordingly, we sought to investigate whether application of Ang IV—by virtue of disrupting the inherent peptidase activity of AT4 receptors—would result in robust increases in oxytocin. As recent in vitro work from Gard et al. (2007) reveal an interaction between these two peptidergic systems, in particular, that Ang IV influences the ability of oxytocin to contract smooth muscles isolated from rat uterus, we propose that Ang IV would elicit increases in extracellular oxytocin levels as measured by in vivo microdialysis techniques. Furthermore, since exogenous administration of oxytocin is reported to possess anxiolytic-like activity in a variety of preclinical behavioral models (McCarthy et al. 1996; Bale et al. 2001; Ring et al. 2006), we investigated whether Nle-Ang IV and LVV-hemorphin-7 would elicit similar anxiolytic responses in the mouse four-plate test (FPT). Finally, by employing the selective oxytocin receptor antagonist, WAY-162720 (Ring et al. 2006), we aimed to explore the possible involvement of the oxytocin receptor system in mediating anxiolytic-like effects of Nle-Ang IV in the FPT.
All experiments were performed in accordance with the specifications of both the National Institutes of Health Guide for the Care and Use of Laboratory Animals (Pub. 85-23, rev 1996) and Wyeth’s Internal Animal Care and Use Committee. For microdialysis studies, adult, male Sprague–Dawley rats (200–225 g) were group (three per cage) housed prior to stereotaxic surgery. For behavioral tests, adult, male Swiss Webster mice (17–22 g) were group housed (five per cage). All animals were obtained from Charles River (Shrewsbury, MA, USA) and housed in an Association for Assessment and Accreditation of Laboratory Animal Care-accredited facility, maintained on a 12-h light/dark cycle (lights on at 0600 hours) with food and water provided ad libitum. All neurochemical and behavioral experiments were performed during the light period.
Nle-Ang IV, LVV-hemorphin-7, divalinal, and oxytocin (all from AnaSpec, San Jose, CA, USA) were dissolved in artificial cerebrospinal fluid (aCSF; 125 mM NaCl, 3 mM KCl, 0.75 mM MgSO4, and 1.2 mM CaCl2, pH 7.4) for intracerebroventricular (i.c.v.) injections or 0.9% saline for either subcutaneous (s.c.) or intraperitoneal (i.p.) injections. WAY-162720 was synthesized by Wyeth’s Chemical and Screening Sciences group (Collegeville, PA, USA) and dissolved in 2% Tween 80/0.5% methylcellulose. All chemicals were prepared fresh on the day of experimentation.
In vivo microdialysis
Stereotaxic surgery was performed to implant a microdialysis guide cannula as recently described (Schechter et al. 2008). Briefly, rats were anesthetized with isoflurane while a microdialysis guide cannula (CMA/12; CMA Microdialysis, Sweden) was positioned above the central nucleus of the amygdala using the following coordinates: AP—2.7 mm from bregma; ML—4.6 mm from midline; DV—7.2 mm from dura (Paxinos and Watson 1986). Microdialysis probes (CMA/12; 20 kD cutoff; CMA Microdialysis, Sweden) with a 2-mm active membrane were perfused with aCSF at 3.0 μL/min according to the manufacturer’s suggestions. Animals were given a postoperative recovery period of 24 h—a time frame shown to be sufficient to not alter neurotransmitter levels as compared to a 7-day recovery period (unpublished observation). Microdialysis probes were inserted into the amygdala and allowed to sit for 3 h to equilibrate. After this time, three baseline samples were collected every 30 min into plastic tubes (CMA microdialysis) containing 10 μl 0.1 N HCl using a flow rate of 3.0 μL/min. Immediately following baseline sample collection, rats received one of the following treatments: 0.9% saline (1 ml/kg, s.c.; vehicle for s.c. studies), aCSF (vehicle for intra-amygdala infusion studies), Nle-Ang IV (0.5 mg/kg, s.c.), or Nle-Ang IV (1 or 10 μM) infused (60 min) into the amygdala by reverse dialysis. Dialysis samples (100 μl total volume) were collected for at least 3 h postinjection/infusion. At the end of the experiment, animals were euthanized and probe placement was verified histologically.
An oxytocin immunoassay (R&D Systems, DE1900) was performed according to manufacturer’s instruction. Briefly, 100 μl microdialysis samples were added to 96-well plates precoated with goat antirabbit antibody. Next, 50 μl of oxytocin conjugate and 50 μl of oxytocin antibody from the immunoassay kit were added into the wells, followed by an incubation of 18–24 h at 4°C. On the next day, plates were washed three times with supplied washing buffer, followed by incubation with para-nitrophenylphosphate substrate for 1 h at room temperature. The plates were then read at 570 nm immediately after stop solution was added. Background signal, nonspecific binding and maximum binding were included in the analysis for quality control. Consistent with a competitive binding assay, the intensity of the color is inversely proportional to the concentration of oxytocin in the sample. Oxytocin concentration was deducted from a standard curve which is generated using four-parameter logistic curve-fit.
These i.c.v. injections have been previously published by Ring et al. (2006). Briefly, only mice in the i.c.v. studies were lightly anesthetized with isoflurane (3% in oxygen). A 10-μL Hamilton syringe (26 gauge, 3 mm tip) was used to pierce the skull and administer an injection (2 μL volume) directly into the left or right lateral ventricle by visual location. All i.c.v. injections were performed by an experimenter that has demonstrated a ≥95% success rate while using a blue dye to verify injection success histologically. Only once an experiment consistently achieved above 95% accuracy in performing these i.c.v. injections did they conduct the behavioral experiments.
The four-plate apparatus, as originally described (Aron et al. 1971), is a Plexiglas chamber (18 × 25 × 16 cm) with a floor consisting of four metal plates (8 × 11 cm), wired to a shock generator (Med Associates, St Albans, VT, USA). Mice (n = 10 per group) were acclimated to the experimental chamber for 18 s immediately prior to the 1-min test session. During the test, the animal’s innate exploration of the novel environment was suppressed by the delivery of a mild foot-shock (0.8 mA; 0.5 s) when the animal moved from one plate to another (“punished crossing”). A “punished crossing” was defined as a mouse placing two paws onto an adjacent plate while moving in the direction of the adjacent plate. The shock was administered while the mouse was roughly halfway between two plates. Following each punished crossing, there was a 3-s timeout, during which crossings were not punished. The total number of punished crossings during the testing period was recorded. Data are represented as total number of punished crossings. Test compounds were administered 20 min (i.c.v.) or 30 min (i.p.) prior to all studies, including combination (i.e., antagonist) studies.
To examine the putative antidepressant effects of Ang IV, mice were tested in the tail suspension test (TST) as originally described (Steru et al. 1985). In this test, mice (n = 12 per group) were suspended upside down by taping their tails with adhesive laboratory tape (VWR International) to a flat metal bar connected to a force transducer within a sound-attenuating chamber (Med Associates Inc., St. Albans, VT, USA) for a 6-min test session. The bar is positioned such that the mouse is unable to reach the top or sides of the chamber. In this model, mice received a single injection of either vehicle, Nle-Ang IV (0.3–30 μg, i.c.v.), or LVV-hemorphin-7 (1–10 μg, i.c.v.) 20 min prior to the test session. Data are represented as total immobility time (seconds)
Oxytocin levels during the baseline samples were averaged, and this value was denoted as 100%. Subsequent sample values were expressed as a percentage of this pre-injection baseline value (% of baseline). Neurochemical data for oxytocin, excluding pre-injection values, were analyzed by a two-way ANOVA with repeated measures (time). Post hoc analyses were made using the Bonferroni–Dunn adjustment for multiple comparisons. All behavioral data for the FPT and TST were analyzed with an ANOVA and least significant difference post hoc test to determine the effect of compounds.
Angiotensin IV elevates extracellular levels of oxytocin in the central nucleus of the amygdala
Angiotensin IV and LLV-hemorphin-7 are anxiolytic in the mouse FPT
Anxiolytic properties of angiotensin IV, but not oxytocin, are blocked by divalinal
Acute treatment with either Nle-Ang IV or oxytocin increased the number of punished crossings in the mouse FPT (Fig. 3a, b). Nle-Ang IV (3 mg/kg, i.p.) when administered alone produced an increase (44%) in punished crossings relative to vehicle-treated animals [F(3, 39) = 3.368, p < 0.05]. This behavioral response was consistent to our earlier findings with this ligand (see Fig. 2a). Acute administration of oxytocin, similar to previous findings (Ring et al. 2006), produced an increase (27%) in punished crossings relative to vehicle-treated animals [F(3, 38) = 2.733, p = 0.0584]. Interestingly, divalinal, a putative AT4 receptor antagonist (Krebs et al. 1996), at a dose (3.4 μg, i.c.v.) that did not elicit a behavioral response on its own, reversed the anxiolytic-like activity of Nle-Ang IV (Fig. 3a; p < 0.05) but not the responses evoked by oxytocin (Fig. 3b).
Interactions between the AT4 and oxytocin receptor systems: the selective oxytocin receptor antagonist blocks the anxiolytic-like effects of angiotensin IV
Angiotensin IV lacks antidepressant-like activity in the mouse TST
In the present series of experiments, we demonstrate that inhibiting the constitutively active peptidase activity of AT4 receptors results in marked increases in extracellular levels of oxytocin in the central nucleus of the amygdala. These neurochemical findings prompted us to investigate the behavioral consequences of inhibiting the peptidase activity of AT4 receptors in vivo. Our characterization of various AT4 ligands (Nle-Ang IV and LVV-hemorphin-7) reveals putative therapeutic importance in a mouse behavioral model of anxiety (FPT). These findings were similar to the effects elicited by exogenous oxytocin treatment and the clinically effective anxiolytic, chlordiazepoxide. Moreover, the anxiolytic-like effects of Ang IV were reversed by pretreatment with either an AT4 antagonist (divalinal) or a selective oxytocin receptor antagonist (WAY-162720). These results shed new light on the interactions between the AT4 and oxytocin receptor systems and confirm, for the first time, that the anxiolytic-like properties of Ang IV are mediated through the oxytocinergic system.
Ang IV binds with high affinity to the AT4 receptor (Swanson et al. 1992) and inhibits its constitutively active peptidase activity responsible for the degradation of various peptides, including, but not limited to, oxytocin (Matsumoto et al. 2000). Using in vivo microdialysis techniques coupled to an immunoassay detection system for oxytocin, we demonstrate that both systemic and intra-amygdala administration of Nle-Ang IV—a metabolically stable derivative of Ang IV—produces an increase in local concentrations of oxytocin in the amygdala. More specifically, we demonstrated that a 60-min infusion of Nle-Ang IV resulted in dose-dependent and significant increases in oxytocin levels, reaching a peak 128 ± 31% above baseline levels at 2 h after the start of infusion. We also observed a similar elevation in extracellular levels of oxytocin in the rat amygdala following systemic administration of Nle-Ang IV. These neurochemical results are consistent with the hypothesis that Ang IV inhibits the constitutively active peptidase activity of AT4 receptors in vitro (Herbst et al. 1997; Matsumoto et al. 2000), which is a primary degradation pathway for several peptides, leading to elevations in levels of peptides. Our work is the first in vivo demonstration that in the rat amygdala, a brain region with overlap of oxytocin and both oxytocin and AT4 receptors (Sofroniew 1983; Chai et al. 2004; von Bohlen und Halbach 2003), inhibiting AT4 receptors, leads to elevations in oxytocin, a peptide with a well-characterized role in modulating the body’s responses to various emotional and physical stressors (Neumann et al. 2000).
A variety of rodent (McCarthy et al. 1996; Bale et al. 2001; Ring et al. 2006; Winslow and Insel 2002) and human (Kirsch et al. 2005; Kosfeld et al. 2005) studies reveal that the oxytocinergic system may represent a therapeutically important target for ameliorating symptoms of anxiety, likely thru direct effects of oxytocin within the amygdala and hypothalamic brain systems (Neumann et al. 2000). Moreover, exogenous administration of oxytocin elicits anxiolytic-like activity (McCarthy et al. 1996; Ring et al. 2006; Windle et al. 1997), while direct infusion of oxytocin into the amygdala reverses prenatal stress-induced behaviors (Lee et al. 2007) and systemic application of WAY-267464, the only small molecule, nonpeptide oxytocin receptor agonist currently available, was recently demonstrated to possess anti-anxiety- and antipsychotic-like activity in various preclinical behavioral paradigms (Ring et al. 2010). As we found that Ang IV elevated oxytocin levels and did so in the central nucleus of the amygdala—a brain region intimately involved in mediating responses of fear and anxiety—we investigated whether inhibiting the peptidase activity of AT4 receptors would also yield an anxiolytic-like profile in a preclinical mouse model. Consistent with this hypothesis, we discovered that both systemic and central administration of Nle-Ang IV produced a significant increase in the number of punished crosses in the mouse FPT. Additionally, central treatment with another AT4 ligand, LVV-hemorphin-7, produced comparable anxiolytic-like effects in the same model. Both of these responses were similar to the behavioral effects of oxytocin and the clinically effective anxiolytic agent, chlordiazepoxide. It is worth point out that the effects of Nle-Ang IV and LVV-hemorphin-7 are not likely due to an impact on sensory function or analgesia since morphine, at doses that are analgesic, does not affect the number of punished crossings in the FPT (Ripoll et al. 2006). Additionally, the behavioral effects of Ang IV were found to be initially mediated by the AT4 receptor since the anxiolytic-like behaviors of this peptide—but not those responses induced by oxytocin—were blocked by the selective AT4 receptor inhibitor, divalinal (Krebs et al. 1996).
While our behavioral data suggest that AT4 receptors are necessary for the anxiolytic-like effects of Ang IV, we also demonstrated that these behavioral responses could be reversed by pretreatment with WAY-162720, a selective, nonpeptide oxytocin receptor antagonist shown to block the anxiolytic-like effects of oxytocin (Ring et al. 2006). Findings from these pharmacological studies provide compelling evidence for a unique relationship and interaction between the AT4 and oxytocin receptor systems that has yet to be fully explored. Although effects in different species has yet to be studies, our research results suggest that inhibition of AT4 receptors in vivo elicits marked increases in the release of oxytocin (microdialysis data) that subsequently engages the oxytocin receptor to exhibit anxiolytic-like activity. This interaction between these two peptide systems is supported by recent work from Gard et al. (2007) showing that Ang IV potentiates the ability of oxytocin to contract uterine smooth muscle in vitro and abolished the antidepressant-like effects of oxytocin in the mouse forced swim test.
Given that anxiety is often comorbid with depression (Rush et al. 2005), we evaluated the behavioral profile of Nle-Ang IV in the mouse TST. Interestingly, unlike the anxiolytic-like findings in the four-plate anxiety model, we found no evidence of antidepressant-like effects following a variety of doses and routes of administration of Nle-Ang IV and LVV-hemorphin-7. These are findings consistent with data showing that Ang IV lacks antidepressant-like activity in the mouse forced swim test (Gard et al. 2007). However, given the interactions between the AT4 and oxytocin systems, as well as previous reports demonstrating that exogenous oxytocin administration does indeed produce antidepressant-like effects (Arletti and Bertolini 1987; Gard et al. 2007), the lack of effect in the TST is intriguing. At the moment, the possible reasons for the lack of antidepressant responses remains speculative; however, it is conceivable that the differences between the pathways involved in the anxiolytic- and antidepressant-like effects are indeed functionally distinct. Moreover, it is also possible that given the promiscuity of AT4 receptor to degrade other peptides—and not just oxytocin—results in the elevation in additional neuropeptides include somatostatin, metenkaphalin, substance P, and vasopressin (Herbst et al. 1997; Matsumoto et al. 2000) that could prohibit an antidepressant-like response in this assay. This hypothesis is supported, in part, by data showing that vasopressin, which is a peptide substrate for AT4 receptors, evokes depression-like effects (Mlynarik et al. 2007).
In summary, we demonstrated that inhibiting the constitutively active peptidase activity of AT4 receptors results in marked increases in the extracellular levels of oxytocin in the rat amygdala. Moreover, acute administration of ligands with high affinity for AT4 receptors produced an anxiolytic-like profile in the mouse FPT. Pharmacological antagonism studies revealed that these anxiolytic-like effects were completely blocked by cotreatment with either an AT4 or oxytocin receptor antagonist. While additional empirical studies are needed to explore whether disruption of AT4 receptor peptidase activity results in similar elevations of other neuropeptides within the CNS, our collective findings shed new light on the interactions between the AT4 and oxytocin receptor systems and confirm, for the first time, that the anxiolytic-like properties of Ang IV are mediated through the oxytocinergic system.