Functional consequences of neuropeptide Y Y₂ receptor knockout and Y₂ antagonism in mouse and human colonic tissues

Abstract

1. Neuropeptide Y (NPY), peptide YY (PYY) and pancreatic polypeptide (PP) differentially activate three Y receptors (Y₁, Y₂ and Y₄) in mouse and human isolated colon.

2. The aim of this study was to characterise Y₂ receptor-mediated responses in colon mucosa and longitudinal smooth muscle preparations from wild type (Y₂+/+) and knockout (Y₂−/−) mice and to compare the former with human mucosal Y agonist responses. Inhibition of mucosal short-circuit current and increases in muscle tone were monitored in colonic tissues from Y₂+/+ and Y₂−/− mice±Y₁ ((R)-N-[[4-(aminocarbonylaminomethyl)phenyl)methyl]-N²-(diphenylacetyl)-argininamide-trifluoroacetate (BIBO3304) or Y₂ (S)-N²-[[1-[2-[4-[(R,S)-5,11-dihydro-6(6H)-oxodibenz[b,e]azepin-11-yl]-1-piperazinyl]-2-oxoethyl]cyclopentyl]acetyl]-N-[2-[1,2-dihydro-3,5(4H)-dioxo-1,2-diphenyl-3H-1,2,4-triazol-4-yl]ethyl]-argininamide (BIIE0246) antagonists.

3. Predictably, Y₂−/− tissues were insensitive to Y₂-preferred agonist PYY(3-36) (⩽100 nM), but unexpectedly Y₄-preferred PP responses were right-shifted probably as a consequence of elevated circulating PP levels, particularly in male Y₂−/− mice (Sainsbury et al., 2002).

4. BIBO3304 and BIIE0246 elevated mucosal ion transport, indicating blockade of inhibitory mucosal tone in Y₂+/+ tissue. While BIBO3304 effects were unchanged, those to BIIE0246 were absent in Y₂−/− mucosae. Neither antagonist altered muscle tone; however, BIIE0246 blocked NPY and PYY(3-36) increases in Y₂+/+ basal tone. BIBO3304 abolished residual Y₁-mediated NPY responses in Y₂−/− smooth muscle.

5. Tetrodotoxin significantly reduced BIIE0246 and PYY(3-36) effects in Y₂+/+ mouse and human mucosae, but had no effect upon Y-agonist contractile responses, indicating that Y₂ receptors are located on submucosal, but not myenteric neurones.

6. Tonic activation of submucosal Y₂ receptors by endogenous NPY, PYY or PYY(3-36) could indirectly reduce mucosal ion transport in murine and human colon, while direct activation of Y₂ receptors on longitudinal muscle results in contraction.

Introduction

Neuropeptide Y (NPY) suppresses synaptic excitation in the central and peripheral nervous systems (King et al., 2000; Weiser et al., 2000; Silva et al., 2001) by activating distinct Y receptors, most commonly postsynaptic or postjunctional Y₁ receptors and presynaptic Y₂ receptors (Dumont et al., 1998; Weiser et al., 2000; Kaga et al., 2001; Bahn et al., 2002). In the hypothalamus, activation of Y₂ receptors inhibits NPY-mediated tonic inhibition of adjacent pro-opiomelanocortin (POMC) neurones leading to satiety in mouse and man (Batterham et al., 2002). In the peripheral nervous system, Y₂ receptors also provide significant presynaptic (or prejunctional) inhibition that can either be autoinhibitory (Malmström et al., 2002), inhibitory upon noradrenergic (Cunningham et al., 1994) or cholinergic neurotransmission (Smith-White et al., 2002), but little is known about Y₂ receptor modulation of nonadrenergic, noncholinergic (NANC) neurotransmission.

Y₂ receptors are not, however, always presynaptic, nor are Y₁ receptors exclusively postsynaptic, or postjunctional (Cox & Cuthbert, 1990; McAuley & Westfall, 1992; Mannon et al., 1999; for reviews see, Cox, 1998; Michel et al., 1998). For example, epithelial Y₂ receptors alone mediate NPY-, PYY- and PYY(3-36)-induced inhibition of Cl⁻ secretion across rat jejunum mucosa (Cox et al., 1988; Cox & Cuthbert, 1990; Cox & Tough, 2000). In the rat colon mucosa, Y₂ receptors are coactivated with Y₁ receptors to inhibit anion secretion and in this tissue each receptor type exhibits pre- and postjunctional responses (Tough & Cox, 1996). The selective nonpeptide antagonists for Y₁ (BIBO3304, Wieland et al., 1998) and Y₂ receptors (BIIE0246, Doods et al., 1999; Dumont et al., 2000) have been crucial for the pharmacological characterisation of NPY, PYY and PYY(3-36) effects, especially in tissues coexpressing multiple Y receptor types. The two antagonists have enabled determination of the functional significance of Y₁ and Y₂ receptors in many isolated tissues and in vivo systems. For example in human isolated colon where NPY, and its hormone analogues, peptide YY (PYY), PYY(3-36) and pancreatic polypeptide (PP) are all antisecretory (and each agonist was equi-effective, Cox & Tough, 2002), only PP effects (i.e. Y₄ receptor-mediated effects) were insensitive to treatment with a combination of a Y₁ and Y₂ receptor antagonist. The same was also true of PP responses in the 129Sv mouse colon mucosa where all four peptides were inhibitory (but Y₁ responses predominated, Cox et al., 2001). Thus, in isolated human and murine colon mucosae, Y₁, Y₂ and Y₄ receptors have been shown to differentially mediate Y agonist inhibition of ion transport, and Y₅ receptors play no functional role in either tissue (Cox et al., 2001; Cox & Tough, 2002).

In the intestine, NPY released locally from enteric submucous secretomotor neurones innervating the mucosa (Ekblad et al., 1987;1988; Sang & Young, 1996; see also review by Furness, 2000) can be antisecretory by activating either Y₂ or Y₁ receptors, or both. In contrast, the endocrine peptides PYY, its fragment PYY(3-36) and PP, which are expressed and released from intestinal or pancreatic type F endocrine cells, respectively (Sundler et al., 1993; Arantes & Nogueira, 1997), will differentially activate Y₁, Y₂ or Y₄ receptors in murine and human large bowel to provide further antisecretory, or proabsorptive, influences following ingestion of a meal. How these effects are coordinated with changes in the patterns of smooth muscle contraction and the modulatory roles that each Y receptor plays in the final integrated intestinal response, remains to be elucidated. Few studies of Y-agonist effects upon intestinal smooth muscle have actually included selective Y antagonists. However, from the agonist orders of potency, we may predict that tonic contractions stimulated by NPY and PYY (Pheng et al., 1999; Ferrier et al., 2000) are Y₂-mediated, while those to PP are most likely Y₄-mediated (in rat, Pheng et al., 1999; Ferrier et al., 2000; and rabbit intestine, Feletou et al., 1999). Neither the Y₁ receptor (Feletou et al., 1999; Ferrier et al., 2000) nor the Y₅ receptor (Feletou et al., 1999) appear to have a significant direct role on normal rodent intestine smooth muscle.

Thus, our initial aim was to elucidate the functional role(s) of Y₂ receptors in isolated smooth muscle and mucosal preparations using the Y₂ antagonist, BIIE0246 (Doods et al., 1999; Dumont et al., 2000) in wild-type (Y₂+/+) murine tissue. The functional phenotype exhibited by these preparations were compared with those from germline Y₂ receptor knockout (Y₂−/−) mice and preliminary investigations have already been presented to the British Pharmacological Society (Hyland et al., 2002). Our second aim was to establish whether Y₂ receptors were pre- or postjunctional and these studies were performed with murine and human colon mucosae. Thirdly, recent studies by Sainsbury et al. (2002) using the same germline Y₂−/− mouse, described an unexpected gender-specific increase in circulating PP in male Y₂−/− mice (together with complex changes in several parameters of energy homeostasis). How elevated circulating PP levels alter peripheral tissue sensitivity to Y agonists or to blockade of such responses by Y receptor antagonists, was of particular interest to us. We therefore set out to establish whether gender-specific differences in Y receptor responses were apparent in isolated intestinal tissue from germline Y₂−/− in comparison with Y₂+/+ mice.

Methods

Tissue preparation

Germline Y₂−/− and Y₂+/+ mice were generated and maintained on a mixed C57BL/6–129/SvJ background as described previously (Sainsbury et al., 2002). Age-matched adult mice (both sexes, 12 weeks or more) were asphyxiated with CO₂, weighed, the ascending and descending colon taken and placed immediately in fresh Krebs–Henseleit (KH) solution with the following composition (in mM): NaCl 118, KCl 4.7, NaHCO₃ 25, KH₂PO₄ 1.2, MgSO₄ 1.2, CaCl₂ 2.5, glucose 11.1 (pH 7.4). Luminal contents were removed by washing gently with KH.

Murine and human descending colon mucosal preparations

Human colon was obtained with the consent of patients undergoing elective bowel resection for primary intestinal carcinoma (the study was approved by the Guy's and St Thomas' Hospitals Research Ethics Committee; Cox & Tough, 2002). Mucosae were prepared by removing overlying circular and longitudinal smooth muscle layers by dissection under a microscope. Mouse and human colon routinely provided six or eight adjacent mucosal pieces, respectively. Either were placed between two halves of Perspex Ussing chambers (exposed area 0.2 cm² for mouse or 0.6 cm² for human tissues), bathed both sides with 5 ml of oxygenated (95% O₂/5% CO₂) KH and maintained at 37°C. Mucosae were voltage-clamped at 0 mV (using a DVC 1000 automatic voltage clamp, World Precision Instruments, Stevenage, U.K.) as previously described (Cox et al., 1988;2001). Once a stable short-circuit current (I_sc) was obtained, agonists or antagonists were added to the basolateral reservoir only and the resultant changes in I_sc were recorded continuously. Murine descending colon mucosae were pretreated with vasoactive intestinal polypeptide (VIP) (30 nM) and 15 min later with either a single concentration of a Y agonist or an appropriate Y receptor antagonist was added (for a further 10–15 min) prior to agonist addition. Tetrodotoxin (TTX) pretreatment was for 15 min prior to VIP addition in mouse mucosa and for 30 min once a stable basal I_sc had been achieved with human mucosae. Maximal agonist-induced changes in I_sc were pooled (and quoted as μA cm⁻²) and means±1 s.e.m. were calculated, where each mucosal or smooth muscle preparation provided a single n value.

Ascending colon longitudinal smooth muscle preparation

Each section of ascending colon provided two adjacent segments of longitudinal smooth muscle (each 1 cm long), which were cut distal to the caecal junction. Segments were washed with KH, attached with thread and suspended in an organ bath (10 ml) in oxygenated (95% O₂/5% CO₂) KH, maintained at 37°C. Tissues were stretched to a basal tension of 1 g and were allowed to equilibrate (for 45 min) with three intermittent KH washes. Isometric changes in basal tension were recorded in response to Y agonists in the absence or presence of specific Y antagonists (added 15 min prior to the agonist). Agonist-induced maximum increases in basal tone (within 5 min of agonist addition) were pooled and are quoted as increases in g tension throughout (mean±1 s.e.m.). Carbachol (CCh, 10 μM) was added as an internal contractile control at the end of each assay.

Statistical analysis

Each mucosal and smooth muscle preparation provided single observations, which were not paired but were pooled to provide means±1 s.e.m. For agonist–response curves, EC₅₀ values (with 95% confidence limits) were calculated from pooled single agonist additions using GraphPad Prism (v. 3.0, GraphPad Software Inc., CA, U.S.A.). Student's unpaired t-test was used to compare responses±antagonist. Multiple comparisons between data groups were evaluated using one-way ANOVA with either Bonferroni's or Dunnett's post-test, where appropriate. In each case, a P-value of less than 0.05 was considered statistically significant.

Materials

Peptides were purchased from Bachem U.K. Ltd (Merseyside, U.K.) unless otherwise stated and aliquots were frozen and stored at −20°C, only undergoing a single freeze–thaw cycle. The porcine (p) sequences of PYY, NPY and PP (the latter from Eli Lilly Inc., IN, U.S.A.), plus the human (h) sequences of PP, PYY(3-36) and Pro³⁴PYY were used as indicated. [(S)-N²-diphenylacetyl)-N-[(4-hydroxyphenyl)methyl]-argininamide] (acetate salt) (BIBP3435), ((R)-N-[[4-(aminocarbonylaminomethyl)-phenyl)methyl]-N²-(diphenyl acetyl)-argininamide-trifluoroacetate (BIBO3304) and (S)-N²-[[1-[2-[4-(R,S)-5, 11-dihydro-6(6H)-oxidobenz[b,e]azepin-11-y1]-1-piperazinyl]-2-oxoethyl]cyclopentyl]acetyl]-N-[2-[1,2-dihydro-3,5(4H)dioxo-1,2-diphenyl-3H-1,2,4-triazol-4-yl]ethyl]-argininamide (BIIE0246) were all obtained from Boehringer Ingelheim Pharma KG (Biberach an der Riss, Germany). [Ac-4NO₂-Phe-c(D-Cys-Tyr-D-Trp-Lys-Thr-Cys)-D-Tyr-NH₂ (CYN-154806) was a gift from Glaxo Institute of Applied Pharmacology (Cambridge, U.K.). 5-bromo-N-(4,5-dihydro-1H-imidazol-2-yl)-6-quinoxalinamine (UK14,304) was purchased from Research Biochemical International (Natick, MA, U.S.A.) and CCh and TTX were from Sigma (Poole, U.K.).

Results

General features of Y₂+/+ and Y₂−/− mice and their descending colon mucosa

Age-matched female Y₂+/+ and Y₂−/− mice weighed significantly less than their male counterparts of similar age, and Y₂−/− mice were leaner than Y₂+/+ mice of a similar age (P-values ranging from 0.05 to 0.001, Table 1) as observed previously by Baldock et al. (2002). Since Sainsbury et al. (2002) described significantly elevated circulating PP levels in male germline Y₂−/− (∼15 nM) compared with male Y₂+/+ mice (∼5 nM), and lower PP levels in female mice of either genotype (∼2 nM in Y₂−/− and 1 nM in Y₂+/+ mice), we segregated our data accordingly.

Table 1.

A comparison of basal current, basal resistance, in I_sc following addition of 30 nM VIP and weight between Y₂+/+ and Y₂−/− mice

	Y2+/+		Y2−/−
	Male	Female	Male	Female
Body weight (g)	31.9±0.4 (25)	25.6±0.4 (22)***	28.6±0.6 (37)+++	22.6±0.5 (32)***+
Age (weeks)	17.4±1.4 (16)	18.8±1.0 (20)	18.7±1.0 (31)	19.4±1.6 (26)
Resistance (Ω cm2)	31.9±1.7 (85)	37.6±2.9 (65)	34.3±1.9 (72)	34.7±1.4 (88)
Basal current (μA cm−2)	52.7±4.3 (138)	41.0±2.3 (103)	44.6±2.8 (167)	36.6±3.0 (152)
VIP (30 nM) (μA cm−2)	66.9±5.1 (105)	67.8±4.4 (109)	65.9±4.1 (173)	70.4±4.2 (155)
EC50 PYY(3–36) (nM)	10.1 (6.7–15.4)	10.2 (4.3–23.9)	No response	No response
EC50 Pro 34PYY (nM)	17.6 (9.2–33.8)	6.2 (1.9–20.1)	39.2 (25.0–61.4)	18.3 (7.3–46.2)
EC50 hPP (nM)	3.7 (0.3–41.7)	9.9 (3.2–30.1)	58.3 (43.0–79.2)	16.3 (5.7–46.8)

Values are ±1 s.e.m. with n values in parenthesis. All EC₅₀ values (with 95% confidence limits) are calculated from the pooled agonist concentration–response curves. No responses were recorded with PYY(3–36) ⩽100 nM.

⁺P⩽0.05

⁺⁺⁺P⩽0.001 comparisons between Y₂+/+ and Y₂−/− mice and

^***P⩽0.001 for comparisons between males and females within each group.

Mucosal resistance and basal I_sc levels were not significantly different in preparations from male and female Y₂+/+ and Y₂−/− mice (Table 1). Electrogenic responses to basolateral VIP (30 nM, which stimulates cAMP-mediated epithelial Cl⁻ secretion and consequently elevates I_sc) were unchanged. Neither the maxima (Table 1) nor the time courses (data not shown) of VIP responses were different between mucosae from Y₂+/+ and Y₂−/− tissue of either gender.

Mucosal I_sc responses to the Y₂ preferred agonist PYY(3-36) and to PYY± the Y₁ antagonist, BIBO3304 or the Y₂ antagonist, BIIE0246

In Y₂+/+ colon mucosae, PYY(3-36) attenuated VIP-elevated I_sc in a concentration-dependent manner with similar potency in wild-type male and female tissue (Figure 1a, b; Table 1). No PYY(3-36) responses were observed (up to 100 nM, Figure 1b) in Y₂−/− tissue from either gender and the small decrease in I_sc observed with 300 nM PYY(3-36) (Figure 1b, male data only) was abolished by BIBO3304 (300 nM, data not shown, Hyland et al., 2002). PYY (10 nM) responses following PYY(3-36) (10 nM) addition were not significantly different in Y₂−/− tissues (−25.8±6.6 μA cm⁻², n=8) from those in Y₂+/+ mucosa (−27.8±10.3 μA cm⁻², n=4) indicating the predominance of Y₁-mediated responses to the full-length peptide. Subsequent addition of the α₂-adrenoceptor agonist, UK14,304 also elicited similar-sized reductions in I_sc in wild-type and knockout groups (Figure 1a for representative trace, pooled data not shown). In Y₂+/+ mucosa, the competitive Y₂ receptor antagonist BIIE0246 (1 μM) raised I_sc (by 2.5±0.4 μA cm⁻², n=26 and 2.1± 0.3 μA cm⁻², n=21 in male and female tissue respectively, compared with vehicle controls containing BIBP3435 (1 μM) of 0.3±0.2 μA cm⁻², n=6 in male wild-type tissues) and this effect was lost from Y₂−/− tissue (0.6±0.6 μA cm⁻², n=6 and 0.3 μA cm⁻², n=2, male and female tissues respectively). Thus, the inhibitory Y₂ tone revealed by BIIE0246, as well as Y₂-stimulated agonist responses were predictably absent from Y₂−/− male and female colon mucosae, with no changes in non-Y receptor-stimulated effects. The combination of Y₁ and Y₂ antagonists to Y₂+/+ mucosa resulted in a significant sustained elevation in I_sc, indicating a combined Y₁ and Y₂ receptor-mediated inhibitory tone in wild-type colon (Figure 2a, b).

Figure 1.

The presence or absence of PYY(3-36) responses in Y₂+/+ and Y₂−/− mucosae (a, b) and longitudinal smooth muscle (c, d). (a) Representative responses to VIP (30 nM), PYY(3-36) (100 nM), PYY (10 nM) and UK14,304 (α2 adrenoceptor agonist) (1 μM) in Y₂+/+ (upper trace) and Y₂−/− (lower trace). Values to the left of each trace in (a) are the basal I_sc prior to VIP addition. (b) Pooled PYY(3-36) data from male tissue in Y₂+/+ and Y₂−/− colon. Values are the mean±1 s.e.m., n=3 throughout (where n represents the number of preparations). (c) Contractile effects of PYY(3-36) (100 nM) on smooth muscle in Y₂+/+ (upper trace) and Y₂−/− (lower trace). (d) Pooled data showing PYY(3-36) or NPY (both at 100 nM) induced increases in tone in Y₂+/+ and Y₂−/− colon, respectively. Each bar is the mean+1 s.e.m. for between five and seven observations. Significant differences between NPY responses in the presence of both antagonists and vehicle control (BIBP3435, ^*P⩽0.05) and PYY(3-36) responses in Y₂−/− compared with Y₂+/+ tissue, ^***P<0.001.

Figure 2.

The effects of BIBO3304 (grey bars) or a combination of BIBO3304 and BIIE0246 (hatched bars) upon inhibitory hPP responses (open bars, in the presence of antagonist(s)) compared with control responses (black bars, in the absence of antagonist(s)) in Y₂+/+, male (a) and female (b), and Y₂−/−, male (c) and female (d) colon. All values are the mean±1 s.e.m. with n numbers in parentheses. Each 300 nM hPP response, in male and female Y₂+/+ tissue, was significantly larger (^*P⩽0.05, ^**P⩽0.01, ^***P⩽0.001; one-way ANOVA with Bonferonni's post-test) than the responses to 30 and 100 nM hPP in those groups. Unpaired Student's t-test was used to compare control hPP responses in the presence or absence of antagonist(s), where+P⩽0.05.

Smooth muscle responses to the Y₂ preferred agonist PYY(3-36) and NPY± antagonists, BIBO3304 and BIIE0246

Longitudinal muscle contractile responses to PYY(3-36) (100 nM) were characterised (in Y₂+/+ ascending colon) by an initial increase in basal tone and a concomitant increase in frequency and amplitude of spontaneous activity (Figure 1c). These effects were present for 3–5 min in Y₂+/+, but were absent from Y₂−/− colon (Figure 1c, d). PYY(3-36)-stimulated increases in Y₂+/+ basal tone (0.12±0.02 g, n=5) were abolished by BIIE0246 (1 μM, 0.0±0.0 g, n=3, P<0.01) the antagonist doing nothing per se. NPY (100 nM) also stimulated Y₂+/+ basal tone (Figure 1d) with associated increases in spontaneous activity (data not shown) and both components were unaffected by prior treatment with either the inactive enantiomer of Y₁ antagonist BIBP3226, BIBP3435 (300 nM), or the more selective Y₁ antagonist BIBO3304 (300 nM). NPY-stimulated Y₂+/+ basal tone was abolished by BIIE0246 (1 μM added in combination with BIBO3304, 300 nM, Figure 1d), indicating that Y₂ receptors predominantly mediate NPY-induced contraction in wild-type tissue. In Y₂−/− colon, however, residual NPY-induced increases in basal tone (in the presence of BIBP3435) were abolished by BIBO3304 (Figure 1d), indicating that a Y₁ receptor-mediated component is present in Y₂ knockout longitudinal smooth muscle. CCh-stimulated contractions were not altered by any of the treatments above and there were no observable differences between muscarinic responses in male and female tissue (Figure 3a–d).

Figure 3.

The effect of pPP (30 nM) upon basal tone of ascending colon preparations obtained from male (a) Y₂+/+ and female (b) Y₂+/+ mice and Y₂−/− tissue from male (c) and female (d) mice. Responses in the presence of either BIBP3435 (300 nM, an inactive enantiomer of the Y₁ receptor antagonist, BIBP3226), or BIBO3304 (300 nM, a selective Y₁ receptor antagonist) are shown. Each bar is the mean+1 s.e.m. with n numbers as shown in parenthesis. There are no significant differences between BIBP3435- and BIBO3304-pretreated pPP responses or between responses from male and female tissue.

Mucosal responses to Y₄-preferring human pancreatic polypeptide (hPP) and Y₁-preferred Pro³⁴PYY±Y antagonists

At concentrations between 1 and 100 nM, human pancreatic polypeptide (hPP) was antisecretory with an EC₅₀ of 3.7 nM in male and 9.9 nM in female Y₂+/+ colon (Table 1). Within this concentration range, the sensitivity to exogenous hPP was approximately halved in female Y₂−/− colon (where notably plasma PP levels doubled, Sainsbury et al., 2002) and in male Y₂−/− mucosa the response curve was shifted an order of magnitude to the right compared with male Y₂+/+ tissue (Table 1, and the former exhibited three-fold elevated PP levels compared with male Y₂+/+ controls). In Figure 2, we show the effects of competitive antagonists (in combination, hatched bars or BIBO3304 alone in grey bars) followed by inhibitory responses to three hPP concentrations only (30, 100 and 300 nM) either in the absence or presence of Y₁ and Y₂ antagonists. In Y₂+/+ colon treated with BIBO3304 (300 nM) and BIIE0246 (1 μM), Y₄ receptors remain unaffected and can be stimulated by a Y₄-preferring agonist hPP. Thus, the antagonist combination raised I_sc levels (Figure 2a, b) indicating endogenous Y₁ and Y₂ receptor-mediated inhibitory tone in wild-type colon mucosae. Secondly, in male and female Y₂+/+ tissues, control 300 nM hPP responses were significantly larger than those at 100 nM (and were excluded from EC₅₀ calculations). Antagonist treatment significantly reduced these high concentration responses to the levels observed with 30 or 100 nM hPP (Figure 2a, b) showing that at 300 nM hPP is able to stimulate Y₁ (and probably Y₂) as well as Y₄ receptors in this tissue.

In male and female Y₂−/− colon, BIBO3304 alone caused a small but significant elevation in basal I_sc (Figure 2c, d) indicating a similar degree of Y₁ receptor-mediated endogenous tone as described for male 129Sv colon mucosa (Cox et al., 2001). In Y₂−/− mucosae from either gender, BIBO3304 pretreatment did not significantly inhibit hPP effects (Figure 2c, d). Notably in male Y₂−/− mucosa, the hPP concentration–response curve was shifted to the right (Table 1).

The potency of exogenous Y₁-preferred agonist, Pro³⁴PYY (0.3–300 nM) was not significantly reduced when comparing Y₂−/− responses with those in respective wild types (Table 1). The Pro³⁴PYY-response curve maxima (which were eight to 10 times greater than those for either hPP or PYY(3-36)) were unchanged in the four groups and BIBO3304 (300 nM) abolished Pro³⁴PYY antisecretory effects (data not shown). Additionally, the inactive enantiomer of BIBP3226, BIBP3435 (1 μM) had no significant effect upon either Pro³⁴PYY responses or other Y agonist effects in colon mucosal sheets (data not shown).

Smooth muscle responses to pPP in Y₂+/+ and Y₂−/− ± either BIBO3304 or BIBP3435

In smooth muscle preparations pPP (30 nM) increased basal tone in both Y₂+/+ and Y₂−/− tissue and these responses were not significantly altered by pretreatment with BIBO3304 or BIBP3435 (Figure 3). Neither compound altered basal tone per se (data not shown). Responses to pPP after either BIBO3304 or BIBP3435 were no different from those recorded in untreated male tissues (0.26±0.09 g, n=5 in Y₂+/+ colon; 0.08±0.03 g, n=3, Y₂−/− tissue). Subsequent CCh responses were unaffected either by BIBO3304 or BIBP3435 treatment, and were no different between male, female, wild-type or knockout groups (Figure 3).

Effects of the Y₂ antagonist BIIE0246 and TTX upon Y₂-preferred PYY(3-36) responses in murine and human colon mucosae

The competitive Y₂ antagonist, BIIE0246 (1 μM) increased I_sc in both murine and human colon mucosae and these effects were significantly attenuated by TTX (100 nM, Figure 4a, b) indicating that Y₂ receptors are present on inhibitory submucous neurones and that they are tonically active in both tissues. While TTX alone did not reduce mouse colon basal I_sc (from 26.5±6.9 μA cm⁻², n=9 to 24.7±6.7 μA cm⁻², n=9), it did significantly inhibit subsequent antisecretory PYY(3-36) responses (by 87.2%, P<0.001, Figure 4a). This agonist activation of Y₂ receptors was predominantly neurogenic although a residual ∼10% of the PYY(3-36) response (30 nM) was not TTX sensitive, indicating a minor postjunctional component, probably epithelial Y₂ receptor expression in this tissue. The identity of the final effector inhibiting I_sc across the wild-type mouse colon epithelium could be one of a number of inhibitory enteric neuropeptides, such as somatostatin (sst). A selective sst₂ receptor antagonist, CYN-154806 (1 μM, Feniuk et al., 2000; Tough & Cox, 2002) was tested. However, it had no effect upon PYY(3-36) responses (controls −5.7± 1.1 μA cm⁻², n=7; plus CYN-154806, −6.6±0.9 μA cm⁻², n=7), but it significantly reduced sst (30 nM) responses (from −50.0±8.5 μA cm⁻², n=7 to −10.2±3.4 μA cm⁻², n=7, P<0.001).

Figure 4.

The effects of BIIE0246 (1 μM) and TTX (100 nM) per se and upon subsequent PYY(3-36) (30 nM in (a) and 100 nM in (b)) responses in (a); wild-type mouse descending colon mucosa and in (b), human colon mucosae. Each bar is the mean±1 s.e.m. from between three and six observations and ^*P⩽0.05, ^**P⩽0.01, ^***P⩽0.001.

In human colon mucosa, TTX significantly reduced basal I_sc (from 66.2±18.5 μA cm⁻², n=6 to 17.8±8.9 μA cm⁻², n=5; P<0.05, Figure 4b). Subsequent PYY(3-36) (100 nM) responses were abolished by BIIE0246 (as shown previously, Cox & Tough, 2002) and by TTX (P<0.01, Figure 4b). Thus, Y₂ receptors in human colon mucosa are also predominantly prejunctional and reduce electrogenic epithelial ion transport, a mechanism similar to that observed in murine colon mucosa (Figure 4a).

Effects of TTX upon PYY(3-36)-induced contractile responses in mouse colon smooth muscle

In wild-type ascending colon, TTX (200 nM) caused an increase in basal tone (0.34±0.08 g, n=7) and spontaneous activity (data not shown), neither of which were significantly altered in Y₂−/− tissue (basal tone; 0.21±0.03 g, n=7). PYY(3-36) effects following TTX were no different from those in naïve tissues (0.13±0.04 g, n=6; compared with controls of 0.12±0.02 g, n=5) and are therefore direct excitatory effects upon longitudinal smooth muscle.

Conclusion

General features

The reduced body weight observed in male and female Y₂−/− mice is consistent with the lean phenotype observed previously in these germline knockout mice, where the long-term loss of hypothalamic Y₂ receptors has been associated with elevations in orexigenic NPY and Agouti-related peptide (AgRP) and with coincident decreases in anorexic POMC and cocaine-and amphetamine-regulated transcript (CART) mRNA in the arcuate nucleus (Sainsbury et al., 2002). Y₂ receptors in this nucleus are thought to mediate the inhibitory food intake response observed in humans and mice, that occurs for up to 12 h following elevated postprandial PYY(3-36) (Batterham et al., 2002). Male Y₂−/− mice, in addition to being lean, also exhibit significantly increased circulating PP levels (Sainsbury et al., 2002) compared with Y₂+/+ and a transgenic PP overexpressing mouse with elevated circulating levels of PP also exhibits a lean phenotype (Ueno et al., 1999). Conversely, human obesity syndromes and the genetically obese ob/ob mouse demonstrate reduced PP plasma levels. How raised PP occurs as a consequence of germ line Y₂ receptor knockout remains unclear, although sexual dimorphism in the functioning of the hypothalamo-pituitary-adrenal axis is evidently partially responsible (Sainsbury et al., 2002). Such changes in circulating PP will not only alter hypothalamic mechanisms, but also peripheral tissue sensitivities to the hormone and potentially to other Y agonists with overlapping pharmacology (for example, hPP can activate murine Y₁ as well as Y₄ receptors).

Predicted losses in sensitivity to the Y₂-preferred agonist, PYY(3-36) in Y₂−/− compared with Y₂+/+ colon mucosa and smooth muscle

Y₂ receptors predominantly mediate PYY(3-36) responses (up to 100 nM) in colon mucosa and smooth muscle. This agonist's concentration–response curve in Y₂+/+ mucosa was comparable with that from 129Sv mouse colon mucosa (Cox et al., 2001). No sensitivity to this fragment was observed in either female or male Y₂−/− mucosae, up to concentrations of 100 nM. The small decreases in I_sc seen to 300 nM PYY(3-36) in Y₂−/− were abolished by BIBO3304 pretreatment, showing that the fragment can stimulate Y₁ as well as Y₂ receptors, albeit at high nM concentrations.

Endogenous PP is predicted to preferentially stimulate Y₄ receptors, although costimulation of Y₁ receptors may also occur (Cox et al., 2001). The consequence of either or both of these events would attenuate electrogenic anion secretion across mucosal preparations thereby lowering basal I_sc levels. Such a pattern was observed with Y₂−/− mucosae (Table 1) and correlates with the robust elevations in circulating PP levels established in Y₂−/− mice of both genders compared to Y₂+/+ mice (Sainsbury et al., 2002). The absence of differences in VIP-stimulated I_sc responses and basal mucosal resistances, between the four tissues, argues against nonspecific mucosal changes in knockout tissues.

In Y₂+/+ ascending colon longitudinal smooth muscle, Y₂ receptors exclusively mediate PYY(3-36) and NPY contractile effects (Figure 1d). The effects of Y agonists that we observe in mouse ascending colon are similar to the tonic contractions recorded in rat ascending colon longitudinal smooth muscle by Ferrier et al. (2000) and Pheng et al. (1999). In the latter, NPY, PYY and PP responses were abolished by TTX and partially inhibited by atropine, indicating that activation of Y receptors (probably Y₂ and Y₄) resulted in ACh and NANC transmitter release to cause contraction indirectly. RT–PCR analysis of rat colon showed expression of Y₁ and Y₄ receptor mRNA (Ferrier et al., 2000) and there were no Y₂-mediated NPY(13-36) effects. Y₂ mRNA was also lacking from ‘nonepithelial' preparations of rat colon, where PCR-based detection identified Y₁, Y₄ and Y₅ receptor mRNA (Goumain et al., 1998). It is important also to note that the Y₆ receptor is not expressed in adult mouse tissue (Gregor et al., 1996). Thus, our data clearly demonstrate a functional role for Y₂ receptors in mouse colon and indicate disparities between species, not only in terms of the Y receptor type(s) involved, but also the mechanism by which Y agonist-mediated excitation occurs. The TTX insensitivity of PYY(3-36) and also PP contractile responses (the latter were also atropine- and neurokinin-insensitive, Singh et al., 2002; Tough et al., 2002) indicate that both peptides act predominantly directly upon mouse longitudinal smooth muscle to cause contraction. In Y₂−/− tissue, we also observed a Y₁ receptor-mediated excitatory component, which was abolished with BIBO3304. Whether this represents a compensatory upregulation of the Y₁ receptor type specifically in smooth muscle remains to be determined.

Wild-type colon longitudinal smooth muscle therefore contracts following stimulation of Y₂ and Y₄ receptors and there was no evidence of inhibitory tone in this preparation, in contrast with colonic mucosae from both human and mouse. It is clear, however, that in murine colon mucosa Y₂ receptors are located predominantly on submucosal neurones (Figure 4) and that a significant level of Y₂-mediated inhibitory tone is present in this tissue. Human colon mucosa also exhibits a marked Y₂ tone which was also TTX-sensitive. Subsequent PYY(3-36) responses were abolished by either TTX or the Y₂ antagonist, BIIE0246, indicating a solely prejunctional Y₂-mediated mechanism.

Unpredicted functional changes in Y₂−/− compared with Y₂+/+ tissues

The increases in PP EC₅₀ values in male Y₂−/− mucosae could be a consequence of sustained Y₄ receptor stimulation (this reducing basal epithelial I_sc slightly in Y₂−/− mucosa) by elevated endogenous PP (Sainsbury et al., 2002). At concentrations of 30 nM (and higher, Cox et al., 2001) hPP coactivates murine Y₁ as well as Y₄ receptors (Figure 2a, b, Tough et al., 2002). Thus, prolonged elevations of endogenous PP levels in male Y₂−/− could also be manifested by reductions in Y₁-agonist potency, which our observations with Pro³⁴PYY do suggest. The EC₅₀ values for Pro³⁴PYY were greatest in male Y₂−/− tissue (from mice where PP levels of ∼15 nM have been recorded), next highest in male Y₂+/+ mucosa (where PP levels were ∼5 nM), followed by female Y₂−/− and Y₂+/+ tissue (where PP levels were ∼2 and ∼1 nM respectively, Sainsbury et al., 2002). In female Y₂−/− tissues, hPP potency was reduced by half compared with that in female Y₂+/+. The observation that both female groups exhibited reduced hPP sensitivity compared with those from males (Table 1) implicates additional gender-specific differences that have altered female tissue sensitivity to exogenous peptide.

In smooth muscle, PP responses (30 nM) were not sensitive to BIBO3304 in any group and there were no significant differences in the size of these peptides or CCh responses between Y₂+/+ and Y₂−/− preparations (Figure 3). The cellular mechanisms underpinning PP-mediated contractile effects in intestinal smooth muscle do involve voltage-gated Ca²⁺ channels since responses are abolished by nifedipine (Singh et al., 2002). All the Y receptor types are coupled to G_i/o, pertussis toxin (PT)-sensitive inhibition of adenylate cyclase (for review, see Michel et al., 1998), but in neuroblastoma cells (Lynch et al., 1994) a PT-insensitive, direct coupling to receptor-operated Ca²⁺ channels provides for depolarisation and resultant contraction. We have yet to determine the exact cellular mechanism(s) involved in these Y receptor-mediated smooth muscle responses.

From these functional studies, we conclude that Y₂ receptors are differentially expressed, being present postjunctionally on murine smooth muscle (together with Y₄ receptors), while in mucosal preparations Y₂ receptors are predominantly prejunctional. Activation of the latter by the endogenous Y₂ agonist PYY(3-36) results in intrinsic inhibitory Y₂ tone and therefore attenuates epithelial ion transport indirectly in human and wild-type mouse colon. Germline knockout of Y₂ receptors predictably led to loss of sensitivity to the preferred Y₂ agonist, PYY(3-36) in isolated preparations, while the associated elevation in circulating PP levels in Y₂−/− mice resulted in functional blunting, not just of exogenous PP (Y₄-mediated) responses, but also of Pro³⁴PYY (Y₁-mediated) antisecretory effects.

Abbreviations

BIBO3304

((R)-N-[[4-(aminocarbonylaminomethyl)phenyl)methyl]-N²-(diphenylacetyl)-argininamide-trifluoroacetate

BIBP3226

N²-(diphenylacetyl)-N-[(4-hydroxy-phenyl)methyl]-D-arginine amide

BIBP3435

[(S)-N²-diphenylacetyl)-N-[(4-hydroxyphenyl)methyl]-argininamide] (acetate salt)

BIIE0246

(S)-N²-[[1-[2-[4-[(R,S)-5, 11-dihydro-6(6H)-oxodibenz[b,e]azepin-11-yl]-1-piperazinyl]-2-oxoethyl]cyclopentyl]acetyl]-N-[2-[1,2-dihydro-3,5(4H)- dioxo-1,2-diphenyl-3H-1,2,4-triazol-4-yl]ethyl]-argininamide

CCh

carbachol

CYN-154806

Ac-4NO₂-Phe-c(D-Cys-Tyr-D-Trp-Lys-Thr-Cys)-D-Tyr-NH₂

hPP

human pancreatic polypeptide

KH

Krebs Henseleit

NPY

neuropeptide Y

Pro³⁴PYY

human [Leu³¹, Pro³⁴]PYY

PYY

peptide YY

PYY(3-36)

peptide YY(3-36)

sst

somatostatin 14-28

TTX

tetrodotoxin

UK14,304

5-bromo-N-(4,5-dihydro-1H-imidazol-2-yl)-6-quinoxalinamine

VIP

vasoactive intestinal polypeptide