Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The tachykinin neuropeptide family consists of a large number of mammalian and nonmammalian peptides that share a similar carboxyl-terminal sequence. Substance P (SP), neurokinin (NK) A and neurokinin B are the principal members of this family and exert a broad spectrum of actions in both the central nervous system and peripheral tissues via NK-1, NK-2, and NK-3 receptors, respectively (Harrison and Geppetti, 2001). SP, NKA, and an elongated form of NKA, neuropeptide  (NP), are abundant in the gastrointestinal tract of various mammalian species, including human. SP and NKA occur primarily in intrinsic enteric neurons, where they are colocalized with Ach, although SP-like immunoreactivity also occurs in extrinsic sensory neurons, whose cell bodies are located in the dorsal root ganglia (Holzer and Holzer-Petsche, 1997a). Recently, neurokinin B was identified in a subset of SP-immunoreactive neurons in the rat ileum (Yunker et al., 1999). In the gut, tachykinins are involved in contraction of smooth muscle, neuroneuronal communication between enteric neurons, regulation of water/ion secretion, and blood flow as well as proinflammatory responses (Holzer and Holzer-Petsche, 1997a,b; Quartara and Maggi, 1998).

Much of our knowledge about tachykinins in the gastrointestinal system is derived from extensive investigation in the guinea pig, where NK-1 receptors are prominent in contractile mechanisms (Holzer and Holzer-Petsche, 1997a,b) and have recently been identified on interstitial cells of Cajal (ICC), which have a pacemaker function (Lavin et al., 1998; Southwell and Furness, 2001). In contrast, in human colon, NK-2 receptors dominate the contraction of smooth muscle by tachykinins (Giuliani et al., 1991; Kölbel et al., 1994; Croci et al., 1998; Cao et al., 2000). These receptors have been localized autoradiographically on the circular muscle and muscularis mucosae (Gates et al., 1989; Warner et al., 2000). The actions of NK-1 receptors in human colon, on the other hand, are more diverse because immunohistochemical and autoradiographic studies have shown that NK-1 receptors are expressed on a variety of cell types, including smooth muscle, neurons, immune cells, glands, and vascular endothelium (Mantyh et al., 1988; Goode et al., 2000a; Renzi et al., 2000), implying that NK-1 receptors may be involved in the control of smooth muscle excitation, immune responses, secretion, and blood flow.

There is growing evidence that imbalanced function of the tachykinin system may influence or accompany the pathophysiology of some intestinal disorders. For instance, some radioimmunoassay and autoradiographic studies demonstrated that SP and NK-1 receptors are up-regulated in the colon of patients with ulcerative colitis and Crohn's disease (Koch et al., 1987; Mantyh et al., 1988; Goldin et al., 1989, 1995), although other studies are contradictory (cited in Lee et al., 2002). Studies using in situ hybridization and immunohistochemistry revealed that ulcerative colitis and Crohn's disease are associated with up-regulation of NK-1 and NK-2 receptors (Goode et al., 2000a; Renzi et al., 2000).

We have previously characterized NK-2 receptors in the circular muscle of human colon using the technique of radioligand membrane binding (Warner et al., 1999). To date, however, no study has attempted to illustrate the nature of NK-1 receptors in the human colon. This is rather surprising because NK-1 receptors obviously play an important role in both physiological and pathophysiological states. In this study, we have reported the localization and characterization of NK-1 receptor in normal human colon from different regions, using the selective NK-1 receptor radioligand [¹²⁵I]Bolton-Hunter-[Sar⁹,Met(O₂)¹¹]SP ([¹²⁵I]BH-SarSP) (Lew et al., 1990).

Recently, in vitro alterations in NK-2 receptor-mediated circular muscle contraction in ulcerative colitis, Crohn's disease, and idiopathic chronic constipation have been reported (Shang et al., 2000; Menzies et al., 2001; Mitolo-Chieppa et al., 2001). The second aim of this study, therefore, was to investigate the contractile responses of human colon circular muscle strips to SP, carried out in normal colon compared with colon from patients with ulcerative colitis and diverticular disease.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Specimen Collection. Specimens of ascending, transverse, descending, and sigmoid colon were collected from male and female patients, aged 28 to 77, undergoing colectomy at the St. George Hospital (Sydney, Australia). Forty-five specimens were obtained from patients with carcinoma, who had not received radiotherapy or chemotherapy treatment before surgery. Whole ring segments (4 cm in length) of macroscopically normal regions, taken 10 to 20 cm away from the tumor, were immediately placed in cold, well carbogenated (95% O₂, 5% CO₂) Krebs-Henseleit solution (composition 118 mM NaCl, 4.7 mM KCl, 25 mM NaHCO₃, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄, 2.5 mM CaCl₂, and 11.7 mM D-glucose). In addition, specimens of sigmoid colon from patients undergoing operation for ulcerative colitis (n = 12) and acute diverticular disease (n = 11) were taken, from areas away from the most grossly inflamed lesions. Specimens were transported to the laboratory in ice-cold Krebs' solution, where they were dissected on the day of arrival or the next day. The serosa, mucosa, and submucosal layers were first removed, and the circular muscle was then separated from the taenia coli. The band of circular muscle (containing the myenteric plexus and a thin layer of longitudinal muscle) was used for functional studies or sectioned into 0.5-g portions, frozen in liquid nitrogen, and stored at 70°C for later homogenate binding. For autoradiographic studies, segments of colon with all layers intact were used. The procedure and experimental protocols were reviewed and approved by the University of New South Wales Human Ethics Committee (ethics 97139). Any specimens from carcinoma patients ("normal colon") appearing inflamed or showing abnormal histological features were discarded.

Homogenate Binding Studies. The radioiodination of [Sar⁹, Met(O₂)¹¹]SP followed the method described previously (Lew et al., 1990). [¹²⁵I]Bolton-Hunter ([¹²⁵I]BH) reagent (1 mCi, specific activity 2200 Ci/mmol) was reacted with [Sar⁹,Met(O₂)¹¹]SP (50 µg, dissolved in 50 µl of borate buffer, pH 8.5) at 4°C for 60 min. The product, [¹²⁵I]BH-SarSP, was then purified with reverse phase high-performance liquid chromatography using a linear gradient of 18 to 48% acetonitrile (containing 0.01% trifluoroacetic acid) at a flow rate of 1.5 ml/min over 80 min.

Autoradiographic Studies. Full-thickness pieces of colon from eight patients were embedded in octane compound (Tissue-Tek; Sakura Finetek, Torrance, CA) and immediately frozen in liquid nitrogen and stored at 70°C. Transverse sections (10 µm) were cut on a cryostat and thaw-mounted onto gelatin-coated glass microscope slides and stored desiccated at 20°C until required. The procedure for autoradiographic experiments was similar to that described previously (Lew et al., 1990). After preincubation (3 × 5 min) in 50 mM Tris-HCl buffer (pH 7.4, 25°C) containing 0.02% BSA, sections were incubated with [¹²⁵I]BH-SarSP (100 pM) in 5 ml of incubation buffer comprising 50 mM Tris-HCl buffer (pH 7.4, 25°C), 3 mM MnCl₂, and 0.02% BSA for 60 min at 25°C. Nonspecific binding was determined in the presence of 1 µM [Sar⁹,Met(O₂)¹¹]SP. The reaction was terminated by washing sections (3 × 5 min) with ice-cold 50 mM Tris-HCl buffer (pH 7.4, 4°C) containing 3 mM MnCl₂ and 0.02% BSA, followed by two brief rinses in ice-cold distilled water and rapid drying. Labeled sections were then fixed in paraformaldehyde vapor at 70°C for 30 min, dipped in molten photographic emulsion LM-1 (Amersham Biosciences, Sydney, Australia), and exposed in the dark for 10 days at 4°C before developing and staining with pyronin Y. Adjacent unlabeled sections were fixed with 4% formaldehyde and stained with Masson's trichrome for histological examination.

Functional Studies. Circular muscle strips (3 × 10 mm) were cut along the circular axis, mounted in 2-ml organ baths containing Krebs-Henseleit solution at 37°C, and aerated with carbogen. The preparations was set at an initial tension of 1 g and allowed to equilibrate for 60 min. Muscle activity was recorded isometrically using FTO3C force transducers (Grass Instruments, Quincy, MA) and recorded using Polygraph (University of New South Wales). After equilibration, 10 mM Ach was added to each organ bath to obtain the maximal response of the muscle strip, followed by thorough washing and a further 60-min equilibration. Discrete concentration-response curves for SP and NKA were then constructed using a 30- to 60-min concentration cycle, with peptide contact time 2 to 3 min. Tachyphylaxis was not observed under these conditions.

Data Analysis. Results are presented as mean ± S.E.M. For binding studies, data were fitted to a one- or two-site model using the nonlinear regression analysis program of GraphPad Prism, version 3 (GraphPad Software, San Diego, CA). The affinities of competitors for [¹²⁵I]BH-SarSP binding sites were expressed as IC₅₀ values. In functional studies, contractile responses of tachykinins were measured in grams tension. Unless otherwise indicated, these data were then expressed as a percentage of the 10 mM Ach response. The concentration-effect curves were fitted using the nonlinear regression analysis program of GraphPad Prism. Agonist potencies were expressed as pD₂ (negative log of the EC₅₀) except in preparations from disease patients, where some concentration-response curves had clearly not reached their maxima, and here the potencies were expressed as pEC₃₀Ach. For the insurmountable antagonist SR48968, the apparent affinity (pK_B) in inhibiting NKA-induced contraction was determined according to Kenakin (1993): a double-reciprocal plot of equieffective concentrations of agonist (A) in the absence (1/A) and presence (1/A') of the antagonist (B) was constructed and pK_B was derived from the equation pK_B = log{(slope  1)/[B]}.

Materials. SP, NKA, NP, [Pro⁹]SP, [Sar⁹,Met(O₂)¹¹]SP, [Lys⁵, MeLeu⁹,Nle¹⁰]NKA(4-10), and senktide were purchased from Auspep (Melbourne, Australia). CP99994 was obtained from Dr M. Snider (Pfizer, Groton, CT) and SR48968 from Dr. X. Emonds-Alt (Sanofi-Synthélabo Recherche, Montpellier, France). The peptidase inhibitors bestatin, captopril, chymostatin, pepstatin, and phosphoramidon were purchased from Sigma (Sydney, Australia). Bacitracin (zinc salt) was from ICN (Sydney, Australia) and diprotin A from Bachem (Bubendorf, Switzerland). The [¹²⁵I]BH-reagent (1 mCi/mmol) was the product of PerkinElmer Life Sciences and purchased from Geneworks (Adelaide, Australia). Stock solutions of peptides were made in 0.01 M acetic acid containing 1% -mercaptoethanol and stored in aliquots at 20°C.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Radioligand Binding

Optimization of Binding Conditions for [¹²⁵I]BH-SarSP Binding. At an incubation temperature of 25°C, specific binding of [¹²⁵I]BH-SarSP to human colon circular muscle membranes reached equilibrium by 30 min and was stable up to 120 min. Specific binding was not significantly enhanced by the presence of peptidase inhibitors bestatin (10 µM), phosphoramidon (10 µM), captopril (1 µM), chymostatin (4 µg/ml), diprotin A (10 µM), and bacitracin (40 µg/ml). Binding assays were subsequently carried out for 60 min of incubation, without any peptidase inhibitors. Under these conditions, specific binding of [¹²⁵I]BH-SarSP was 50 to 75%.

Saturation Studies. Saturation studies were carried out in circular muscle homogenates of ascending, descending, and transverse colon, but mainly focused on the sigmoid colon. In all regions, specific binding of [¹²⁵I]BH-SarSP was saturable and to a homogenous population of binding sites (Fig. 1). Figure 2 shows the B_max and K_D values for individual specimens from different regions, illustrating an absence of regional differences. Overall, [¹²⁵I]BH-SarSP binding was of high affinity (K_D = 68 ± 8.9 pM; n = 12) and low capacity (B_max, = 0.31 ± 0.07 fmol/mg of wet weight tissue; n = 12).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Representative "hot" saturation data obtained using increasing concentrations of [125I]BH-SarSP incubated with human sigmoid colonic circular muscle membranes. A, saturation data illustrating total (), specific (), and nonspecific () binding. B, Scatchard plot showing a single binding site, of KD = 0.06 nM and Bmax = 0.304 fmol/mg of wet weight tissue. Eight other experiments gave similar data.

View larger version (11K): [in this window] [in a new window]   Fig. 2.   KD values (A) and Bmax values (B) from different regions of the human colon, showing a lack of regional differences in binding affinity and capacity.

Competition Studies. Naturally occurring tachykinins, selective agonists, and antagonists were used in competition studies to characterize the [¹²⁵I]BH-SarSP binding site (Table 1). Data analysis showed that for the majority of agonists, the slope factors were shallow and the curves were best fitted to two sites, of high and low affinity (Table 1). On the other hand, the slopes for all antagonists and for the selective NK-2 receptor agonist [Lys⁵,MeLeu⁹,Nle¹⁰]-NKA(4-10) were close to unity, indicating binding to a single site. The rank order of affinity for ligands to inhibit [¹²⁵I]BH-SarSP binding was SP  [Pro⁹]SP  CP99994 NKA  NP > [Lys⁵,MeLeu⁹,Nle¹⁰]-NKA(4-10)  SR48968 senktide (selective NK-3 receptor agonist).

Autoradiographic Studies. In transverse sections of normal human colon, moderate-to-dense specific binding of [¹²⁵I]BH-SarSP was seen over the circular muscle and blood vessels (Figs. 3 and 4). In the circular muscle, intensity of binding was greatest near the longitudinal muscle, declining toward the submucosa, although a broken border of denser binding often occurred at the submucosal edge (Fig. 3B). Negligible specific binding was visible on the myenteric ganglia (Fig. 4B) or muscularis of the longitudinal muscle (Figs. 3B and 4B).

View larger version (152K): [in this window] [in a new window]   Fig. 3.   Low-power photomicrographs of adjacent transverse sections of two different specimens of sigmoid colon, illustrating histological staining (A and D), total binding of [125I]BH-SarSP (B and E), and nonspecific binding (C and F). A to C, longitudinal muscle (lm) containing blood vessels (bv), circular muscle (cm), myenteric plexus (mp), and part of the submucosa (s). D to F, submucosa (s) with blood vessels (arrowed), and mucosa with muscularis mucosae (mm) glands (g) and lamina propria (lp). Specific binding occurred on blood vessels of the longitudinal muscle and submucosa, and on the circular muscle. Denser binding was seen on the submucosal edge of the circular muscle. Weak specific binding occurred in the lamina propria. A number of macrophages exhibiting nonspecific binding are seen in the mucosa and submucosa of E and F, and several are circled (refer to *, left side of D and F).

View larger version (178K): [in this window] [in a new window]   Fig. 4.   High-power photomicrographs of adjacent transverse sections of sigmoid colon, illustrating histological staining (A and D), total binding of [125I]BH-SarSP (B and E), and nonspecific binding (C and F). A to C, longitudinal muscle (lm), circular muscle (cm), and ganglia of the myenteric plexus (mp). Specific binding occurred on the circular muscle but not the longitudinal muscle or myenteric ganglia. D to F, part of the submucosa (s). Structures showing specific binding are on circular muscle, two arterioles (a) with specific binding on the endothelium, small blood vessels (bv), and a number of microvessels (*).

Functional Studies

Normal Colon. The contractile effects of SP and NKA were examined in circular muscle strips of sigmoid, ascending, and descending colon. SP and NKA contracted the strips in a concentration-dependent manner, although NKA was approximately 2 orders of magnitude more potent than SP (Fig. 5; Table 2). There was no significant difference in maximal contractile responses (equivalent to about 60% of the 10 mM Ach) or potency (pD₂) values between different regions (Table 2). In subsequent studies, data from descending and sigmoid colon strips were combined and are described as "sigmoid" below.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Concentration-response curves to SP in circular muscle strips from normal sigmoid colon. Data represent the mean ± S.E.M. A, responses to SP were significantly reduced by 100 nM SR48968 (P < 0.0001, two-way ANOVA) but not by 100 nM CP99994 (n = 7-26). B, responses were significantly reduced by 1 µM atropine (P = 0.012, two-way ANOVA) (n = 11-25). C, responses were unaffected by 1 µM indomethacin (n = 6-26).

Functional Studies in Colon with Gastrointestinal Diseases. Contractile responses were also examined in specimens of sigmoid colon from patients with ulcerative colitis and diverticular disease. The mean weights of the circular muscle strips were similar: control, 74 mg; colitis, 75 mg; and diverticular, 77 mg. There was a trend for a potentiation of responses to Ach in strips from colitis patients (maximum increase in tension 5.6 ± 1.1 g; n = 8), whereas responses to Ach were unaltered in diverticular disease (3.9 ± 0.75 g; n = 11) compared with control (4.2 ± 0.56 g; n = 23).

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Concentration-response curves to SP in sigmoid circular muscle strips from normal, ulcerative colitis (UC) and diverticular disease (DD) colonic specimens. Data represent the mean ± S.E.M. and represent absolute increase in tension (A and D) or percentage of maximum Ach response (B, C, E, and F). A, responses to SP were reduced (P = 0.049, two-way ANOVA) in colitis compared with control (n = 12-26). B, when data from A are expressed as percentage of maximum Ach, responses were not significantly changed. C, atropine (1 µM) had no significant effect in specimens with ulcerative colitis (n = 5-12). D and E, responses to SP were very markedly reduced in diverticular disease (n = 11-26), whether expressed as increase in gram tension (D) (P = 0.0028, two-way ANOVA) or as percentage of maximum Ach (E) (P = 0.0001, two-way ANOVA). F, indomethacin (1 µM) did not influence responses in specimens with diverticular disease (n = 9-11).

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Tachykinins, especially SP, play an important role in the various functions of the gastrointestinal tract. In humans, evidence suggests that the tachykinin NK-2 receptor is the predominant tachykinin receptor in the colon circular muscle (Giuliani et al., 1991; Kölbel et al., 1994; Croci et al., 1998). Our previous study investigating NK-2 receptors in normal sigmoid colon circular muscle showed that binding of ¹²⁵I-NKA was of high capacity (B_max of 2.1 fmol/mg of wet weight tissue) (Warner et al., 1999). In this study, we have focused on SP and the NK-1 receptor, using the selective radioligand [¹²⁵I]BH-SarSP. Binding of [¹²⁵I]BH-SarSP was of high affinity (K_D = 68 pM) and low capacity (B_max = 0.3 fmol/mg of wet weight tissue) compared with the NK-2 receptor. There were no regional differences observed in receptor affinity and number between normal ascending, descending, transverse, and sigmoid colon. NK-1 receptor radioligand binding data from homogenates of human tissue do not seem to have been reported previously.

Our autoradiographic studies showed the presence of [¹²⁵I]BH-SarSP binding sites on circular muscle, submucosal blood vessels, with a high density of silver grains on longitudinal muscle blood vessels and a low density in the mucosa. This matches the immunohistochemical localization of SP around blood vessels, circular muscle, and in the mucosa, as well as in myenteric ganglia (Wattchow et al., 1988). A similar distribution of NK-1 binding sites (Mantyh et al., 1988, 1995) and mRNA (Goode et al., 2000a; Renzi et al., 2000) is seen in other studies in human colon, with expression up-regulated on blood vessels in specimens from ulcerative colitis and Crohn's disease patients (Mantyh et al., 1988, 1995; Goode et al., 2000a). The broken border of denser binding occurring near the submucosal edge of the circular muscle (Fig. 3B) may correspond to ICC, which are prominent in this location in human colon (Rumessen et al., 1993; Vanderwinden et al., 1996). ICC are also found in throughout the circular muscle layer and around the enteric ganglia in the distal human colon (Vanderwinden et al., 1996).

The competition binding profile of [¹²⁵I]BH-SarSP using tachykinins, selective tachykinin agonists, and antagonists demonstrated binding to the NK-1 receptor. The affinity of CP99994 for [¹²⁵I]BH-SarSP binding sites reported herein (pIC₅₀ = 10.48) is comparable with the pK_B value of 9.9 in the human pulmonary artery (Corboz et al., 1998). Although in saturation studies this radioligand bound to a single class of sites, most tachykinin agonists seemed to bind to more than one component. This ability to resolve the competition data into two sites was a notable feature of our study. It was also of interest that the affinity of these peptides was higher than reported in binding studies with cell lines using a similar radioligand (Fong et al., 1992). Such high-affinity, multiple-site binding was also reported in homogenates from guinea pig lung (Geraghty et al., 1992). As shown in our autoradiographic studies, there were at least two populations of [¹²⁵I]BH-SarSP binding sites in the preparation used in our homogenate binding studies, one on the circular muscle and the other on blood vessels located within the external longitudinal muscle. Whether these two populations represent two distinct molecular forms is unclear from this study. Alternatively, the two binding sites may represent different activation states of the same receptor protein, or represent binding to different G proteins (Maggi and Schwartz, 1997). Contraction of human colon sigmoid circular muscle in response to NK-2 receptor agonists is mediated via G_q (Cao et al., 2000).

A striking result of this study is the absence of functional evidence for the participation of NK-1 receptors in circular muscle contraction. The highly selective agonist [Pro⁹]SP was virtually ineffective, and SP was 100-fold less potent than NKA in contracting isolated circular muscle strips. Furthermore, responses to SP were almost completely inhibited by the NK-2 but not by the NK-1 receptor antagonist. Thus, SP seems to contract the circular muscle via NK-2 receptors, but not via NK-1 receptors. This is at variance with the clear demonstration of NK-1 binding sites on circular muscle, shown in binding and autoradiographic studies. The question is, what is the physiological function of NK-1 receptors (strictly speaking, binding sites) on human colon circular muscle? Do these sites represent the full-length NK-1 receptor or are they a short, nonfunctional form? Both short as well as full-length versions of the human NK-1 receptor have been described, in a glioblastoma cell line (Fong et al., 1992) and in human colonic mucosa (Goode et al., 2000b). Studies using in situ hybridization and quantitative polymerase chain reaction indicate that the full-length form of the NK-1 receptor is present in the circular muscle, but whether it coexists with the short isoform is unknown (Goode et al., 2000a; Renzi et al., 2000). It should also be pointed out that not only smooth muscle cells occur in "circular muscle" but also neurons, ICC, and other cell types.

Tachykinins have both direct and indirect actions to contract smooth muscle. In several tissues, including rat duodenum, there is evidence for tachykinin-prostaglandin interactions (Hallgren et al., 1998), but this did not seem to be the case in human sigmoid colon circular muscle. We found that a minor component of the contractile response was apparently due to Ach release, as in guinea pig ileum (Holzer and Lembeck, 1980), because responses to SP (but not to NKA) were slightly but significantly inhibited by atropine. A similar result was reported by Kölbel et al. (1994). Thus, although a direct interaction of SP with NK-2 receptors of the smooth muscle predominates, there is also a presence of facilitatory NK-1 receptors on cholinergic neurons. Although we and others (Mantyh et al., 1995) did not find autoradiographic evidence for NK-1 receptors on myenteric ganglia of normal colon, other techniques have demonstrated NK-1 receptor mRNA on enteric neurons of normal human ileum and colon (Goode et al., 2000a; Renzi et al., 2000).

Therefore, one plausible explanation for our data with atropine is that, first, in the normal colon, a limited number of neuronal NK-1 receptors are expressed on ganglia, autoradiographically undetectable using a low concentration of radioligand. Second, there may be some different tachykinin receptor subtype on the ganglia. A third hypothesis is that SP coreleased from cholinergic motoneurons to the circular muscle may act locally to modulate Ach release. This might occur if NK-1 receptors were localized to terminals of cholinergic motor neurons innervating circular muscle. Alternatively, because our autoradiographic studies showed denser binding at the submucosal edge of circular muscle, corresponding to ICC, it is likely that NK-1 receptors are present on human ICC, as shown in the guinea pig intestine (Lavin et al., 1998; Southwell and Furness, 2001). ICC form a link between neurons and smooth muscle cells. We do not have any direct evidence for these hypotheses, although the binding sites seen throughout the circular muscle layer may be located on motoneurons (as well as or instead of muscle fibers and ICC).

Altered colonic motility is a feature of most if not all gastrointestinal disorders. A novel finding of our study was that contractile responses to SP were significantly decreased in both potency and efficacy in colon from diverticular disease patients. Because cholinergic responses were unchanged in diverticular disease, as reported previously (Snape et al., 1991), the altered contractility to SP is rather due to a disease-related alteration in SP receptor/signaling mechanisms than to a nonspecific damage to the tissue by inflammation. Although prostaglandins are involved in inflammatory cascades, there was no evidence for their participation in responses to SP in circular muscle specimens with diverticular disease.

The attenuated contractility to SP found in diverticular disease is unlikely to be related to the change of NK-2 receptor density and affinity because our binding studies with ¹²⁵I-NKA showed no change in K_D or B_max values in colon circular muscle from patients with diverticular disease (Shang et al., 2000). The hypocontractility in diverticular disease was, however, not seen in response to NKA (data not shown). Thus, the reduced responsiveness to SP in this disease may well be due to deficiency of the NK-1 receptor system. To date, there is little information available on the tachykinin system in diverticular disease.

Attenuations in responses to SP were also seen in specimens from patients with ulcerative colitis, but of lesser magnitude than diverticular disease. A similar result was reported recently (Menzies et al., 2001). This finding is supported by results from our binding study, where a reduced affinity of ¹²⁵I-NKA was observed in the colonic circular muscle from ulcerative colitis patients (Shang et al., 2000). No changes in ganglionic or circular muscle NK-1 receptors occurred in specimens with ulcerative colitis, although there was marked up-regulation of NK-1 receptors associated with other cell types (Mantyh et al., 1995; Goode et al., 2000a). Thus, the reduced responsiveness to SP is probably related to decreased affinity of the NK-2 receptor in ulcerative colitis. The nature of NK-1 and NK-2 receptor and signaling systems in human colon from patients with diverticular disease and ulcerative colitis is certainly worthy of further investigation.