Uridine adenosine tetraphosphate is a novel neurogenic P2Y1 receptor activator in the gut

Abstract

Enteric purinergic motor neurotransmission, acting through P2Y1 receptors (P2Y1R), mediates inhibitory neural control of the intestines. Recent studies have shown that NAD(+) and ADP ribose better meet criteria for enteric inhibitory neurotransmitters in colon than ATP or ADP. Here we report that human and murine colon muscles also release uridine adenosine tetraphosphate (Up4A) spontaneously and upon stimulation of enteric neurons. Release of Up4A was reduced by tetrodotoxin, suggesting that at least a portion of Up4A is of neural origin. Up4A caused relaxation (human and murine colons) and hyperpolarization (murine colon) that was blocked by the P2Y1R antagonist, MRS 2500, and by apamin, an inhibitor of Ca(2+)-activated small-conductance K(+) (SK) channels. Up4A responses were greatly reduced or absent in colons of P2ry1(-/-) mice. Up4A induced P2Y1R-SK-channel-mediated hyperpolarization in isolated PDGFRα(+) cells, which are postjunctional targets for purinergic neurotransmission. Up4A caused MRS 2500-sensitive Ca(2+) transients in human 1321N1 astrocytoma cells expressing human P2Y1R. Up4A was more potent than ATP, ADP, NAD(+), or ADP ribose in colonic muscles. In murine distal colon Up4A elicited transient P2Y1R-mediated relaxation followed by a suramin-sensitive contraction. HPLC analysis of Up4A degradation suggests that exogenous Up4A first forms UMP and ATP in the human colon and UDP and ADP in the murine colon. Adenosine then is generated by extracellular catabolism of ATP and ADP. However, the relaxation and hyperpolarization responses to Up4A are not mediated by its metabolites. This study shows that Up4A is a potent native agonist for P2Y1R and SK-channel activation in human and mouse colon.

Keywords: P2Y1 receptor; Up4A; enteric nervous system; intestine; purinergic signaling.

Fig. 1.

Up4A is released in human colon tunica muscularis. (A) HPLC-DAD analysis at 200- to 270-nm absorbance wavelengths identified Up4A with a lower detection limit of 25 pmol per injection. Each absorbance wavelength tested is represented by a different color chromatogram; optimum Up4A detection is at 260 nm absorbance. mAU, milli-absorbance units. (B) Up4A, etheno-derivatized with 2-chloroacetaldehyde (80 °C, 40 min), formed eUp4A that was identified with HPLC-FLD excitation (Ex) at 200–270 nm and emission (Em) at 400–470 nm. Each wavelength tested is represented by a different color chromatogram; optimum: Ex, 230 nm, and Em, 420 nm, with a lower detection limit of 0.05 pmol per injection. LU, luminescence units. (C) HPLC-FLD analysis demonstrated that the detection sensitivity of eUp4A (optimum: Ex 230 nm, Em 420 nm) was more than six orders of magnitude higher than that of Up4A (optimum: Ex 260 nm, Em 400 nm). (D) The 8.9-min fraction (presumably containing Up4A) of tissue superfusates (60 chambers combined) collected before (prestimulation, PS) and during EFS (ST, 16 Hz, 0.3–0.5 ms, for 30 s) and etheno-derivatized verified basal and EFS-evoked release of Up4A in human colon. The segment of the peak at the left of the dotted line was identified as eUp4A, because this segment coaligned precisely with the eUp4A standards. (E, Top) Segments of original chromatograms showing eUp4A in tissue superfusates. (Middle and Bottom) Basal (PS, Middle) and EFS-evoked (ST, 16 Hz, 0.3 ms, for 30 s; Bottom) overflow of Up4A (Control) was reduced by 0.5 μM TTX. (F) Averaged data (means ± SEM) for basal overflow (fmol/mg tissue) (PS) in the absence (control) and presence of 0.5 μM TTX. Stimulus-induced release of Up4A is shown as stimulated release minus basal release, ST − PS. *P < 0.05 vs. no-TTX controls; the number of experiments is shown in parentheses. Note that TTX inhibited ∼50% of the basal Up4A overflow and about 90% of the stimulation-evoked overflow of Up4A.

Fig. 2.

Up4A causes a P2Y1R- and SK-channel–mediated relaxation in human colon muscularis that is more potent than the relaxation in response to numerous adenine nucleotides and adenosine. (A) Inhibition of spontaneous colonic contractions by 2-min exposure (black lines) to Up4A in the absence and presence of atropine and l-NNA; w, washout. (B and C) Inhibition of spontaneous colonic contractions by Up4A (black lines) was abolished by the selective and specific antagonist of P2Y1R MRS 2500 (B) and by apamin (C). (D) Summary (means ± SEM) of concentration-dependent inhibition of spontaneous colonic contractions by Up4A (1–10 μM for 2 min) and responses to 10 μM Up4A in the presence of 1 μM atropine, atropine plus 100 μM l-NNA, 1 μM MRS 2500, and 0.3 μM apamin. Open circles denote significant differences from effects of 10 μM Up4A: ^oP < 0.05, ^ooP < 0.01. Asterisks denote a significant difference from the effect of Up4A in the presence of atropine plus l-NNA: ***P < 0.001. The number of experiments is given in parentheses. (E) Concentration-dependent inhibition of spontaneous colonic contractions by Up4A (closed squares) and the selective and specific P2Y1R agonist MRS 2365 (open circles), each at 1–30 μM. Data are presented as means ± SEM. Asterisks denote a significant difference between the effects of the two substances, each at 1 μM: **P < 0.01. n = 3–6. Note that at 3–30 μM the two substances are equally effective. (F) Summary (means ± SEM) of the inhibition of spontaneous colonic contractility by 10 μM Up4A, 10 μM MRS 2365, and 10–1,000 μM of ATP, ADP, AMP, ADO, NAD⁺, and ADPR. Asterisks denote significant differences from 10 μM Up4A: **P < 0.01, ***P < 0.001. The number of experiments is given in parentheses.

Fig. 3.

Electrophysiological responses to Up4A in murine circular muscle and isolated PDGFRα⁺ cells. (A and B) Responses to pressure ejection (spritzes) of Up4A at impalement sites in circular muscle sheets of murine proximal (A) and distal (B) colon before (black traces) and after (red traces) MRS 2500. (C and D) Summarized data in which the spritz pulses were varied from 10–200 ms (n = 6 for proximal and distal colon). (E and F) Hyperpolarization responses to Up4A (black traces) were inhibited by 0.3 μM apamin (red trace in E; 71 ± 5.8% inhibition, n = 3, P = 0.012) and 1 μM UCL 1684 (red trace in F; 68 ± 4.1% inhibition, n = 4, P = 0.00003). (G) Hyperpolarization responses under current-clamp conditions elicited by Up4A in an isolated PDGFRα⁺ cell (n = 5). Responses reached E_K, suggesting activation of a K⁺ conductance, and were blocked by MRS 2500, suggesting hyperpolarization was mediated by P2Y1R. (H) Responses to ramp potentials (−80 to +80 mV) applied under voltage-clamp conditions before (black trace, a) and after (red trace, b) Up4A. Reversal potential of current activated by Up4A shifted toward E_K. Outward current was reduced by 1 μM UCL 1684 (blue trace, c).

Fig. 4.

Up4A forms UMP and ATP in the human colon (A–G) and UDP and ADP in the murine colon (H–N). (A and H) Excerpts of chromatograms obtained with HPLC-FLD of eUp4A (50 nM) in the absence of tissue [(−) tissue] and after 30-s contact with tissue [(+) tissue] in human (A) or murine (H) colonic muscle preparations, demonstrating a decrease in Up4A concentration after 30-s contact with tissue. LU, luminescence units. (B and I) HPLC-FLD chromatograms of etheno-purines in tissue superfusates in the absence of tissue [(−) tissue] and after 30-s contact with tissue [(+) tissue] of human (B) and murine (I) colons. eADO is observed only after the tissue was exposed to Up4A. (C and J) HPLC-DAD chromatograms showing the formation of UMP, but not UDP, in human colon (C) and the formation of UDP, but not UMP, in murine colon (J) exposed to eUp4A substrate (50 nM). mAU, milli-absorbance units. (D and K) Summary (means ± SEM) of Up4A (50 nM) degradation after contact with human colon (D) for 30 s (black bars) or 120 s (gray bars) or after 30-s contact with murine proximal (black bars) or distal (gray bars) colon (K). Asterisks denote significant differences from control [(−) tissue]: *P < 0.05, **P < 0.01, n = 8 (30 s); n = 4 (120 s); n = 4 (mouse proximal and distal colon). (E and L) Summary (means ± SEM) of detected eATP, eADP, eAMP, and eADO in the eUp4A substrate solution in the absence of tissue (white bars) and after 30-s (black bars) or 120-s (gray bars) exposure of human colon to Up4A (E) or after 30-s exposure of murine proximal (black bars) and distal (gray bars) colon to Up4A (L). Asterisks denote significant differences from control [(−) tissue]: *P < 0.05, **P < 0.01, ***P < 0.001; n = 8 (30 s); n = 4 (120 s); n = 4 (murine proximal and distal colon). ^oooP < 0.001 vs. 30-s exposure to eUp4A in human colon. (F and M) Summary (means ± SEM) of detection of UTP, UDP, and UMP in the eUp4A substrate solution in the absence of tissue (white bars) and after 30-s (black bars) or 120-s (gray bars) exposure of human colon to Up4A (F) or after 30-s exposure of murine proximal (black bars) and distal (gray bars) colon to Up4A (M). Asterisks denote significant differences from control [(−) tissue]: *P < 0.05, ***P < 0.001. Open circles denote significant differences from 30-s contact of tissue with eUp4A. ^oooP < 0.001; n = 8 (30 s); n = 4 (120 s); n = 4 (mouse proximal and distal colon). (G and N) Chemical structure of Up4A and proposed sites of cleavage (scissors sign) and primary products in human (G) and murine (N) colonic muscles. Based on data demonstrated in A–F, we propose that in human colonic muscles Up4A forms ATP and UMP. ATP is rapidly and sequentially degraded to ADP, AMP, and ADO, so that the end detectable products are ADO and UMP. From data demonstrated in H–M we propose that in the murine colon Up4A first forms ADP and UDP. ADP is rapidly and sequentially degraded to AMP and ADO. The detectable end products are ADP, AMP, ADO, and UDP.