Sustained stimulation of vasopressin and oxytocin release by ATP and phenylephrine requires recruitment of desensitization-resistant P2X purinergic receptors

Abstract

Coexposure of hypothalamo-neurohypophyseal system explants to ATP and phenylephrine [PE; an alpha1-adrenergic receptor (alpha1-AR) agonist] induces an extended elevation in vasopressin and oxytocin (VP/OT) release. New evidence is presented that this extended response is mediated by recruitment of desensitization-resistant ionotropic purinergic receptor subtypes (P2X-Rs): 1) Antagonists of the P2X2/3 and P2X7-Rs truncated the sustained VP/OT release induced by ATP+PE but did not alter the transient response to ATP alone. 2) The P2X2/3 and P2X7-R antagonists did not alter either ATP or ATP+PE-induced increases in [Ca(2+)](i). 3) P2X2/3 and P2X7-R agonists failed to elevate [Ca(2+)](i), while ATP-gamma-S, an agonist for P2X2-Rs increased [Ca(2+)](i) and induced a transient increase in VP/OT release. 4) A P2Y1-R antagonist did not prevent initiation of the synergistic, sustained stimulation of VP/OT release by ATP+PE but did reduce its duration. Thus, the desensitization-resistant P2X2/3 and P2X7-R subtypes are required for the sustained, synergistic hormone response to ATP+PE, while P2X2-Rs are responsible for the initial activation of Ca(2+)-influx by ATP and ATP stimulation of VP/OT release. Immunohistochemistry, coimmunoprecipitation, and Western blot analysis confirmed the presence of P2X2 and P2X3, P2X2/3, and P2X7-R protein, respectively in SON. These findings support the hypothesis that concurrent activation of P2X2-R and alpha1-AR induces calcium-driven recruitment of P2X2/3 and 7-Rs, allowing sustained activation of a homeostatic circuit. Recruitment of these receptors may provide sustained release of VP during dehydration and may be important for preventing hemorrhagic and septic shock.

Fig. 1.

Effect of ATP-γ-S, an agonist at P2X1, 2, 3, and 4-Rs, on [Ca²⁺]_i in supraoptic nuclei (SON) neurons filled with fura-2 AM. A: time course of the response to ATP-γ-S (100 μM) followed after a 30-min washout by ATP (100 μM) in SON neurons simultaneously recorded in a single hypothalamo neurohypophyseal system (HNS) explant preparation (means ± SE, n = 20). Both ATP-γ-S and ATP induced rapid increases in [Ca²⁺]_i, as monitored by the change in 340:380 emission ratio, which returned toward baseline prior to washout of the ligand. B: comparison of the peak 340/380 ratios induced by ATP-γ-S and ATP in 36 neurons combined from 2 HNS preparations (means ± SE). The peak response to ATP was significantly greater than that elicited by ATP-γ-S (*P < 0.001) This is consistent with ATP activation of both P2X and P2Y-Rs (39). The y-axis in B is the same as in A. C: immunohistochemistry for P2X2-R in rat SON (C) and PVN (C′). Note the intense labeling of the cytoplasm throughout the dorsal/ventral extent of SON and in both magnocellular (mc) and parvocellular (pc) neurons in PVN. The P2X2-R immunoreactivity was eliminated by preincubation of the antibody with the peptide antigen (not shown). Scale bars: 100 μm. OC, optic chiasm; VGL, ventral glia lamina.

Fig. 2.

Effect of a P2X3 and 2/3-R selective agonist, (α,βMeATP; 100 μM) and antagonist (A317491; 300 nM) on [Ca²⁺]_i, and ATP-induced vasopressin/oxytocin (VP/OT) release. A: time course of the [Ca²⁺]_i response to α,βMeATP followed by ATP (100 μM). B: mean peak response to ATP was significantly greater than that to α,βMeATP (n = 38 neurons in 2 explants, H = 56.26, P < 0.001). C: time course of the change in [Ca²⁺]_i in response to ATP (blue trace) or ATP+PE (red trace; each at 100 μM) alone or in the presence of A317491 (300 nM). D: analysis of the peak and sustained elevations in [Ca²⁺]_i induced by exposure to ATP or ATP+PE in the presence or absence of A317491. Data from two explants for each treatment are combined (n = 36 neurons for ATP, 22 for ATP+PE). All neurons analyzed responded. As reported previously, the peak response was comparable to ATP or ATP+PE (40), and it was not altered by A317491 (F = 1.8, P = 0.15). The sustained elevation in [Ca²⁺]_i induced by ATP+PE was greater than that to ATP alone [H = 43.398, P < 0.001] (40), but it was not altered by A317491 (P > 0.05 b vs. c, end ratio). Values are expressed as means ± SE; bars labeled with the same letter indicate no significant difference from each other. The y-axis in D is the same as in C. ATP induced a transient increase in VP and OT release (E and F: *P < 0.05 vs. other times between 5 and 7 h) that was not altered by the prior addition of A317491 (300 nM, A31.ATP; E: VP: F_time = 6.29, P < 0.001, F_group = 0.8, P = 0.4; F: OT: F_time =3.4, P = 0.005; F_group = 0.001, P = 0.9; n = 6/group). Arrows indicate the time that agonists/antagonists reached the explants. The drugs were present during the remainder of the experiment. Note that the small decrease in VP/OT release observed in response to A317491 alone was not statistically different from that in unexposed explants (ATP group from hour 4.5 to 5.5).

Fig. 3.

Effect of A317491 (300 nM; a P2X2/3 antagonist) and Brilliant Blue G (BBG; 1 μM; a selective P2X7-R antagonist) on ATP+PE stimulated VP/OT release from HNS explants. A and A′: Following equilibration, all explants were exposed to ATP+PE for the rest of experiment. Two hours later, A317491 (A317491.AP, n = 6) or BBG (BBG.AP, n = 5) was added to the perifusion medium. Both VP and OT release increased during exposure to ATP+PE (P = 0.028, VP; P < 0.001, OT). Both A317491 and BBG reduced VP/OT release relative to the group maintained with ATP+PE alone (n = 5; F = 1.9, P = 0.01 for VP; F = 3.5, P < 0.001 for OT; individual mean comparisons: *P ≤ 0.04 both drug groups vs. ATP+PE alone; #P = 0.04 BBG.AP vs. ATP+PE only). B and B′: A317491 (n = 6) and BBG (n = 5) were added 1 h before ATP+PE and remained in the perifusion medium for the rest of the experiment. Control explants (n = 5) were perifused with basal medium throughout the experiment. In the absence of either antagonist, ATP+PE (n = 6) induced a significant increase in VP/OT release compared with the control (F = 14.9, P < 0.001 for VP; F = 6.8, P = 0.018 for OT; *Individual mean comparisons P ≤ 0.05), but neither VP nor OT release was significantly different from controls in the A317491.AP or BBG.AP groups following the addition of ATP+PE. Neither drug significantly altered basal VP or OT release (an hour prior to ATP+PE addition).

Fig. 4.

Effect of BBG on [Ca²⁺]_i and ATP-induced VP/OT release. A: time course of the [Ca²⁺]_i response to ATP (100 μM; blue trace) or ATP+PE (100 μM each; red trace) alone or in the presence of BBG (1 μM). B: analysis of the peak and sustained elevations in [Ca²⁺]_i induced by exposure to ATP or ATP+PE in the presence or absence of BBG. Data combined from two explants/treatment (n = 24 neurons for ATP, 36 for ATP+PE). All neurons that were analyzed responded. The peak response was not altered by BBG. Again, ATP+PE induced a sustained elevation in [Ca²⁺]_i compared with ATP alone (P < 0.05 C vs. D, end ratio) that was not altered by BBG. Values are expressed as means ± SE; labels with the same letter are not significantly different from each other. The y-axis label in B is the same as in A. C and D: ATP induced a transient increase in VP and OT release (*P < 0.05 vs. hours 5 and 6) that was not altered by the prior addition of BBG (VP: F_time = 5.09, P < 0.001, F_group = 0.12, P = 0.7; F: OT: F_time = 4.43, P < 0.001; F_group = 0.22, P = 0.6; n = 5/group). Basal VP and OT release was not significantly different in the presence of BBG during the hour before ATP addition.

Fig. 5.

BzATP effect on [Ca²⁺]_i. A: time course of the [Ca²⁺]_i response to BzATP (20 μM followed by 100 μM) and subsequently ATP (100 μM). Values are means ± SE from 18 neurons in one hypothalamo-neurohypophyseal system (HNS) explant. B: peak change in 340:380 ratio was significantly greater in response to 100 μM compared with 20 μM BzATP (*P < 0.05). The response to ATP was significantly greater than the response to 100 μM BzATP (#P < 0.05). C: peak response to 200 μM BzATP (BzATP 200) was comparable to the response to ATP-γ-S and significantly less than the response to 100 μM ATP (*P < 0.001).

Fig. 6.

Molecular evidence for the presence of P2X2/3-R in SON. A: Coimmunoprecipitation of P2X2 and P2X3. The Western blot, reacted with anti-P2X2-R, reveals immunoreactive protein bands of the appropriate size for P2X2-R in microdissected rat SON homogenized in co-IP buffer (SON lane) and in the elution fraction of SON tissue immunoprecipitated with anti-P2X3-R (X3-IP lane). No immunoreactive band was present in the elution fraction from anti-P2X3 Sepharose beads when SON protein was omitted (IP-cont. lane). Size markers are indicated in left lane. B: immunohistochemistry for P2X3-R in rat SON. Scale bar: 100 μm. B′: digital magnification of the area denoted by the dashed box in B. Note the primarily cytoplasmic location of the immunoreactivity throughout the dorsal/ventral extent of SON. Immunoreactivity was eliminated by omission of the primary antibody (not shown).

Fig. 7.

P2X7-immunoreactive protein is present in SON from rat (rSON) and wild-type mouse (WTson), but not P2X7 knockout mouse (X7KOson). The 75-kDa protein band reported by others is present in all lanes, but the ∼70 kDa P2X7-ir protein that is present in rat and wild-type mouse SON is absent in the P2X7-KO mouse. Preincubation of the P2X7 antibody with antigen peptide eliminated all bands.

Fig. 8.

Effect of calmidazolium on the sustained stimulation of VP and OT release by ATP+PE. Calmidazolium (Cal, 10 or 100 μM), an antagonist that blocks P2X7-R ion conductance, but not P2X7-R pore formation, was added 2 h after the addition of ATP+PE (100 μM each) to perifused HNS explants. ATP+PE increased VP (A) and OT (B) release relative to control explants (F = 5.1, P = 0.009 for VP; F = 4.5, P = 0.035 for OT). Both VP and OT release remained elevated in the presence of calmidazolium [VP: *P < 0.05, ATP+PE (n = 9) and calmidazolium 10 μM.AP (n = 3) vs. control (n = 6); #P < 0.05, calmidazolium 100 μM.AP (n = 4) vs. control; OT: *P < 0.05 calmidazolium100.AP (n = 5) vs. control (n = 5); #P < 0.05 ATP+PE (n = 5) vs. control].

Fig. 9.

Effect of MRS2179 (100 μM), a P2Y1-R antagonist, on VP and OT release induced by combined exposure to ATP+PE (100 μM each). A: VP release was significantly elevated by ATP+PE (P < 0.001). The presence of MRS2179 (MRS+AP) did not alter this response during the first 2 h (*P < 0.05 for ATP+PE and MRS2179+AP vs. control), but subsequently, the response decayed in the MRS2179+AP group, such that MRS2179+AP differed statistically from the both ATP+PE and control ($) and then was not different from control, while ATP+PE remained significantly greater than control (#P < 0.01). B: OT release initially was significantly elevated in both the ATP+PE and MRS2179+AP groups (*P < 0.001 for ATP+PE and MRS+AP vs. control) but then decreased precipitously leaving only the ATP+PE group greater than control (#P = 0.01, 0.055, 0.045, and 0.051 at the next 4 respective time points).