Celia D. Sladek* and Zhilin Song
Department of Physiology and Biophysics, University of Colorado at Denver and Health Sciences Center,
Aurora, CO, USA
Abstract: Arginine vasopressin (AVP) neurons of the hypothalamo-neurohypophseal system (HNS) are innervated by numerous afferent pathways carrying information about two physiologically important parameters: blood volume/pressure and osmolality. These pathways use a variety of neurotransmitters/ neuropeptides. In order to understand normal and pathological regulation of VP secretion, the mechanisms underlying integration of these complex afferent signals by the AVP neurons must be understood. The importance of neurotransmitter interactions in determining hormone release is highlighted by the finding that simultaneous exposure to adenosine triphosphate (ATP, a neurotransmitter acting on purinergic receptors) and phenylephrine (PE; to mimic norepinephrine activation of a1-adrenergic receptors) results in potentiation of AVP release that is characterized by an increase in the peak response and conversion of a transient response to a response that is sustained for hours. Evaluation of the mechanisms responsible for this response indicated that (1) activation of P2X purinergic receptors (P2X-R) is required, (2) protein kinase C (PKC) activation is required, (3) the sustained component requires new gene transcription, (4) the synergism does not involve presynaptic mechanisms nor does it occur directly in the neural lobe and (5) live-cell Ca + + imaging techniques demonstrated a sustained increase in [Ca + + ]; and that ATP activates P2Y-Rs as well as P2X-Rs in supraoptic neurons. Since the subtypes of P2X-Rs differ in their rate of desensitization, identification of the subtype of P2X-Rs participating in the initial and sustained responses to ATP+PE may elucidate mechanisms underlying the abrupt and transient responses to orthostatic hypotension versus sustained responses to chronic hypovolemia or vasodilation.
Keywords: vasopressin; oxytocin; ATP; purinergic; norepinephrine; adrenergic; haemorrhage; shock
The hormones secreted from the hypothalamo-neurohypophyseal system (HNS) regulate important
Corresponding author. Tel.: +303 7244526; Fax: +303 7244501; E-mail: [email protected]
homeostatic functions. Arginine vasopressin (AVP) is critical for maintenance of water balance and cardiovascular function via its antidiuretic and vasoconstrictor actions. Oxytocin (OTX), in addition to its roles in reproduction (lactation and parturition), also participates in maintenance of fluid and electrolyte balance via its natriuretic effects (Verbalis et al., 1991). Sustained elevations in AVP and OTX
secretion persist for days in response to dehydration (Dunn et al., 1973; Windle et al., 1993) which, as a combined osmotic and hypovolemic stimulus, activates CNS pathways carrying information about plasma osmolality as well as the pathways carrying information about blood pressure and volume. Prolonged decreases in blood pressure and blood volume independent of hyperosmolarity also induce elevations in plasma AVP and OTX that can be sustained for hours to days. Hypotension induced by intravenous injection of hydralazine, a potent vasodilator, induces large increases in both plasma OTX and AVP for at least 90 min (Schiltz et al., 1997). Both non-hypotensive and moderate (20% of blood volume) haemorrhages induce prompt and prominent increases in plasma AVP (Grimes et al., 1987; Block et al., 1989; Blair et al., 1991). Without intervention, blood volume restitution in response to a moderate haemorrhage can require more than 24 h (Grimes et al., 1987; Blair and Mickelsen, 2006). Cardiovascular collapse occurring in haemorrhagic and septic shock and in patients suffering from cardiac arrest or following extended cardiopulmo-nary bypass is associated with a failure of AVP secretion to maintain plasma AVP at levels sufficient to prevent vasodilation and can be averted by administration of exogenous AVP (Morales et al., 1999; Landry and Oliver, 2001; Sharshar et al., 2003). Thus, sustained elevation in AVP secretion in response to hypotension and hypovolemia is critical for maintaining cardiovascular homeostasis.
Pathways transmitting hypotension and hypovolemia signals to AVP neurons
As previously reviewed (Sladek, 2000), AVP secretion in response to cardiovascular information is transmitted to the AVP neurons by multiple pathways carrying information about decreases in blood pressure and blood volume and a separate inhibitory pathway carrying information about increases in blood pressure. Baroreceptors in the carotid sinus and stretch receptors in the right atria of the heart monitor blood pressure and volume, respectively. This information is carried to the nucleus tractus solitarius and dorsal vagal complex in the brainstem by the IXth and Xth cranial nerves, and the information is relayed to the A1 and C1 catecholamine neurons in the ventrolateral medulla which in turn innervate the AVP neurons of the HNS. Destruction of either the A1 (noradrenergic) projection to the supraop-tic nucleus (SON) or the C1 (epinephrinergic) neurons of the rostral ventrolateral medulla compromises the response of AVP neurons to hypotension or simulated hypovolemia (Day and Sibbald, 1990; Smith et al., 1995; Madden et al., 2006).
Neurotransmitters responsible for stimulation of AVP neurons in response to hypovolemia/ hypotension
Although the A1 pathway was first recognized for its ability to secrete norepinephrine (NE) and convincing evidence supported the importance of the A1 pathway for stimulation of AVP release in response to moderate decreases in blood pressure (Raby and Renaud, 1989; Smith et al., 1995), the studies that intended to demonstrate that NE is the transmitter responsible for this response were unsuccessful (Day et al., 1990). Adrenoceptor antagonists did not block A1 activation of AVP cells, prompting the suggestion that these neurons use neurotransmitters in addition to NE as their principal transmitter (Day et al., 1990). Injections of the broad spectrum excitatory amino acid (EAA) receptor antagonist, kynurenic acid, were ineffective in blocking excitation induced by stimulation of the A1 region, ruling out the possibility that glutamate is responsible for AVP responses to A1 input (Day et al., 1990). Neuropeptides such as neuropeptide Y and substance P that are co-localized in subsets of A1 neurons (Everitt et al., 1984; Sawchenko et al., 1985; Blessing et al., 1986; Bittencourt et al., 1991) were considered candidates for transmitting cardiovascular information, but were found to play modu-latory roles (Willoughby and Blessing, 1987; Sibbald et al., 1989; Kapoor and Sladek, 2001). A role for ATP, the nucleotide which is commonly co-localized in catecholamine vesicles (Fried, 1980; Whittaker, 1982), in mediating responses to activation of the A1 pathway was supported by the finding that application of the P2 receptor blocker, suramin (10 mM), in the SON reversibly blocked the excitation of AVP cells by A1
stimulation without preventing the excitatory effect of locally applied NE (Day et al., 1993). Histological and electrophysiological evidence that SON neurons express multiple subtypes of puri-nergic receptors, the receptor family activated by ATP, further supports a role for purinergic transmission (Hiruma and Bourque, 1995; Shibuya et al., 1999) and supports a role for ATP as a neurotransmitter involved in cardiovascular regulation of AVP secretion.
Response to ATP and activation of a1-adrenergic receptors
The above evidence as well as data demonstrating that ATP and NE are co-released in hypothalamic slices (Sperlagh et al., 1998) prompted us to evaluate the impact of co-exposure to ATP and NE on AVP and OTX release. Since our early experiments had demonstrated that the effect of NE on AVP release was situation dependent due to offsetting effects on a- and ß-adrenergic receptors (Sladek and Yagil, 1990), we examined whether the response to ATP and phenylephrine (PE), an a1-adrenergic receptor (a1-R) agonist that excites SON neurons and stimulates AVP release (Armstrong et al., 1986), was different when the agents were applied separately or together. As shown in Fig. 1, using acutely prepared and perifused explants of the HNS that include the SON neurons with their axons projecting through the median eminence and terminating in the neural lobe, we observed that the combined exposure to ATP and PE (ATP + PE) resulted in a significantly larger increase in AVP release than was observed with either agent alone and even more impressively, the response was converted from a transient increase in AVP release to one which was sustained for several hours (Kapoor and Sladek, 2000). There was a similar synergistic effect of ATP + PE on OTX release. As is more evident in Fig. 2B, the synergistic response to ATP + PE is frequently characterized by an initial, rather small transient increase in hormone release followed by a delayed, larger increase in hormone release. We have invested considerable effort in understanding the cellular and molecular mechanisms underlying the sustained increase in hormone release, because this
may be critical for maintenance of cardiovascular function during haemorrhage or chronic hypoten-sive conditions induced by cardiac failure or arrest, vasodilation or cardiopulmonary bypass surgery.
Mechanism of ATP + PE synergism
Presynaptic vs. postsynaptic site of action of ATP and PE
Since SON neurons express both purinergic and adrenergic receptors, it is tempting to assume that ATP and PE act postsynaptically to alter AVP secretion. However, it is also possible that one of these agents acts presynaptically to increase excitatory afferent input (i.e. glutamate release). ATP has been shown to stimulate the release of glutamate and GABA in SON (Ponzio and Hatton, 2004), and NE increases excitatory postsynaptic
potentials (epsp) on magnocellular neurons via an a1-R-mediated effect (Daftary et al., 1998; Gordon and Bains, 2005). This suggests noradrenergic regulation of local glutamatergic neurons that are known to exist in the perinuclear zone of SON. The possibility of presynaptic actions of ATP and/or NE was addressed using a cocktail of EAA receptor antagonists. Combined exposure to CNQX and AP5 (antagonists at AMPA and NMDA receptors, respectively) decreased basal AVP release but did not alter the response to combined exposure to ATP and PE (Fig. 2A; Song and Sladek, 2006). The decrease in basal release indicates that glutamatergic tone contributes to basal AVP release in HNS explants, but the synergistic action of ATP + PE is not dependent on activation of ionotropic EAA receptors. A non-selective meta-botropic glutamate receptor (mGluR) antagonist [MCPG, (RS)-a-methyl-4-carboxyphenylglycine], also did not alter the response to ATP + PE (Fig. 2B; Song and Sladek, 2006). This does not eliminate the possibility that either ATP or PE acts presynaptically to alter the release of some other excitatory input; however, glutamate is the predominant local excitatory input to SON (Van den Pol et al., 1990) and, therefore, the prime candidate for presynaptic modulation in HNS explants.
Do ATP and/or PE act on the neurohypophyseal nerve terminals?
Purinergic and a1-R mediated effects have also been reported in the neural lobe. ATP has been reported to increase AVP release from isolated neurohypophyseal terminals (Troadec et al., 1998), and to alter K+ efflux from pituicytes (Troadec et al., 2000). a1-Rs are present in the neural lobe (DeSouza and Kuyatt, 1987), and it receives noradrenergic innervation from the A2 neurons in the medulla as well as sympathetic innervation from the superior cervical ganglion (Alper et al., 1980; Saavedra, 1985; Garten et al., 1989). The possibility that the nerve terminals in posterior pituitary are the site of synergism was addressed by perifusing isolated neural lobes with ATP and PE alone or together. Synergistic stimulation of AVP release by ATP + PE was not observed in isolated, perifused neural lobes (Fig. 3; Song and Sladek, 2006). Furthermore, as shown in Fig. 3, blocking action potential propagation with tetro-dotoxin (TTX, 3 mM) eliminated the response to ATP and PE indicating that action potentials are required for the response to ATP + PE (Song and Sladek, 2006). Note that TTX also decreased basal release indicating that a large portion of the hormone release from HNS explants reflects action potential initiated exocytosis. Thus, the synergistic
effect is not due to synergistic actions of P2 and a1-Rs occurring at the level of the neural lobe, and it requires action potentials initiated in either the hypothalamus or the neural lobe.
Role of adenosine to the ATP+PE synergistic stimulation of AVP/OTX release
ATP is metabolized to adenosine in the extracellular space by ecto-5'-nucleotidase (Dunwiddie et al., 1997). Adenosine inhibits SON neurons (Ponzio and Hatton, 2005), and adenosine receptors (AR) are expressed in SON neurons (Noguchi and Yamashita, 2000). Since conversion of a transient ATP-induced response to sustained stimulation of AVP and OTX release is an important feature of the ATP + PE synergism, removing an inhibitory influence might underlie the development of synergism. Since adenosine is a metabolite of ATP and is inhibitory to SON neurons, we postulated that metabolism of ATP to adenosine could be responsible for the transient response to ATP alone.
This possibility was tested using a combination of a potent ecto-5'-nucleotidase inhibitor, a,ß-methylene adenosine 5'-diphosphate (AMP-CP), and a competitive substrate for ecto-5'-nucleotidase (guano-sine monophosphate, GMP). Enzymatic inhibition did not affect basal AVP release. It did slightly prolong the response to ATP, but not for the duration of exposure to ATP; the response to ATP was not greater than that observed with ATP alone (Fig. 4; Song and Sladek, 2005). Therefore, although conversion of exogenously applied ATP to adenosine contributes to termination of ATP-induced stimulation of AVP release, production of adenosine alone cannot account for the sustained responses observed with ATP + PE.
Purinergic receptor subtypes activated by ATP in SON neurons
Extracellular ATP acts as a neurotransmitter by activation of two classes of purinergic receptors: P2X receptors (P2X-R) and P2Y receptors (P2Y-R).
Was this article helpful?