Sympathetic vasoconstrictor regulation of mouse colonic submucosal arterioles is altered in experimental colitis

Abstract

Recent studies suggest that altered neural regulation of the gastrointestinal microvasculature contributes to the pathogenesis of inflammatory bowel disease. Therefore, we employed video microscopy techniques to monitor nerve-evoked vasoconstrictor responses in mouse colonic submucosal arterioles in vitro and examined the effect of 2,4,6-trinitrobenzene sulphonic acid (TNBS) colitis. Nerve stimulation (2-20 Hz) caused frequency-dependent vasoconstrictor responses that were abolished by tetrodotoxin (300 nm) and guanethidine (10 microm). The P2 receptor antagonist suramin (100 microm) or the alpha(1)-adrenoceptor antagonist prazosin (100 nm) reduced the vasoconstriction and the combination of suramin and prazosin completely abolished responses. Nerve-evoked constrictions of submucosal arterioles from mice with TNBS colitis were inhibited by prazosin but not suramin. Superfusion of ATP (10 microm) resulted in large vasoconstrictions in control mice but had no effect in mice with colitis whereas constrictions to phenylephrine (3 microm) were unaffected. P2X(1) receptor immunohistochemistry did not suggest any alteration in receptor expression following colitis. However, Western blotting revealed that submucosal P2X(1) receptor expression was increased during colitis. In contrast to ATP, alphabeta-methylene-ATP (1 microm), which is resistant to catabolism by nucleotidases, constricted control and TNBS arterioles. This indicates that reduced purinergic transmission to submucosal arterioles may be due to increased degradation of ATP during colitis. These data comprise the first description of the neural regulation of mouse submucosal arterioles and identify a defect in sympathetic regulation of the GI vasculature during colitis due to reduced purinergic neurotransmission.

Figure 1. Frequency dependence of mouse colon submucosal arteriole vasoconstriction to nerve stimulation

A, raw data traces of arteriolar diameter responses to nerve stimulation at various frequencies. Nerve stimulation decreased arteriolar diameter (downward deflections). The amplitude of evoked constrictions increased with increasing stimulus frequencies. B, mean ±s.e.m. frequency–response data for nine arterioles from control mouse colon submucosal arterioles.

Figure 2. Pharmacology of sympathetic vasoconstrictions

A and B, raw data illustrating the inhibitory effect of the P2 receptor antagonist suramin and the α₁-adrenoceptor antagonist prazosin on vasoconstrictions to 20 Hz electrical stimuli. C, summary data showing that suramin (n = 38) and prazosin (n = 8) individually reduced evoked vasoconstrictions and that the combination of suramin plus prazosin (n = 16) abolished neurally mediated vasoconstrictions. *P < 0.001, one-way ANOVA plus Tukey's multiple comparisons test.

Figure 3. Pharmacology of sympathetic vasoconstrictions in arterioles from TNBS colitis mice

A, mean ±s.e.m. data which demonstrate that suramin (n = 10) had no inhibitory effect on vasoconstrictions evoked by 20 Hz stimuli in these preparations whereas prazosin (n = 5) inhibited a large component of evoked constrictions. The combination of suramin plus prazosin (n = 9) abolished the constriction response to nerve stimulation but there is no statistical difference between the effect of prazosin alone and prazosin plus suramin. B, comparison of the percentage inhibition by suramin (100 μm) of sympathetic vasoconstrictor responses at 5 and 20 Hz in control and TNBS arterioles. Suramin had significantly less effect on constrictions in TNBS colon compared to controls. C, prazosin had significantly more effect at 20 Hz on constrictions in TNBS colon versus controls. This indicates that the adrenergic component of sympathetic vasoconstrictions compensated for the colitis-induced loss of purinergic signalling. *P < 0.05, one way ANOVA with Tukey's multiple comparisons test for A and unpaired t test with Welch's correction for B and C (comparisons of control versus TNBS).

Figure 4. Immunohistochemical analysis of sympathetic neuroanatomy in normal (A) and TNBS colitis (B) colons

Submucosal preparations were stained with antiserum that detects tyrosine hydroxylase (TH), a vital enzyme for catecholamine synthesis. Immunoreactivity was present in a perivascular plexus along submucosal arterioles and was also detected in varicosities that surround neurons in the submucosal plexus. Although there was a decrease in the amount of staining in the submucosal plexus, the innervation of submucosal arterioles by TH-immunoreactive fibres was not reduced during colitis (C).

Figure 5. Responsiveness of submucosal arterioles to sympathetic vasoconstrictors in normal and TNBS-inflamed colons

A, superfusion of ATP caused large vasoconstrictions that rapidly desensitize in arterioles from control colons. TNBS colitis abolished the responsiveness of arterioles to ATP. B, summary data on constriction amplitude to ATP (control: n = 17; TNBS: n = 7) and phenylephrine (control: n = 9; TNBS: n = 12), an α-adrenoceptor agonist in control and TNBS colons. Colitis abolished constrictions to ATP whereas constrictions to phenylephrine were unaffected. *P < 0.01.

Figure 6. Signalling pathways downstream of receptor activation differ between purinergic and adrenergic constrictions

A, purinergic constrictions rely on Ca²⁺ influx and were abolished by superfusion with zero-Ca²⁺ Krebs solution (n = 3 each for ATP and phenylephrine). B, purinergic constrictions were unaltered by 2-APB (n = 5), an IP₃ receptor antagonist, whereas constrictions to phenylephrine (n = 10) were abolished following blockade of IP₃ receptor-mediated release of sarcoplasmic reticulum stores of Ca²⁺. *P < 0.05.

Figure 7. Immunohistochemical and immunoblot detection of P2X₁ receptor in submucosal arterioles

A, in control colon submucosa, receptor immunoreactivity was localized to smooth muscle cells within arterioles. B, TNBS colitis did not affect P2X₁ receptor immunoreactivity in arterioles but increased immunoreactivity in non-vascular cell types, probably reflecting influx of immune cells during colitis. C, Western blotting of submucosal protein with a P2X₁ receptor antiserum detected a band at the predicted molecular weight of ∼59 kDa; β-actin antiserum revealed a band at ∼40 kDa. D, analysis of the ratio of densitometry measurements of P2X₁ receptor and β-actin bands in submucosal preparations from 6 control and 6 TNBS colons. TNBS colitis significantly increased P2X₁ receptor expression. ***P < 0.001, unpaired t test with Welch's correction.

Figure 8. The purinergic agonistαβ-methylene-ATP, but not ATP, constricts arterioles in TNBS colon

Responses to ATP and αβ-methylene-ATP were compared in arterioles from 12 control mice and 6 mice with TNBS colitis. ATP and αβ-methylene ATP cause equal constrictions in control mice (P = 0.98; paired t test), but in TNBS arterioles there is a marked difference in the responsiveness of arterioles (###P < 0.0001, paired t test). A and C, vasoconstrictions of arterioles to ATP is abolished in TNBS colitis (***P < 0.0001, unpaired t test with Welch's correction) whereas arteriolar responsiveness to αβ-methylene-ATP is increased in colitis (B and C; ***P < 0.0001, unpaired t test with Welch's correction).