Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Neuromedin U (NmU) is a bioactive peptide originally isolated from porcine spinal cord and initially characterized by its potent contractile activity in rat uterine smooth muscle (Minamino et al., 1985a,b). Since the initial identification of porcine NmU-25 and porcine NmU-8 (which shares the common C-terminal sequence of NmU-25), NmU peptides from a variety of species have been identified and characterized, including rat NmU-23 (Conlon et al., 1988; Minamino et al., 1988), canine NmU-25 (O'Harte et al., 1991), and human NmU-25 (Austin et al., 1995), among others. Although the NmU peptides are structurally unlike any other group of peptides, the amino acid sequence homology between the NmU peptides is strikingly similar.

NmU-like immunoreactivity has been detected in mammalian central and peripheral nervous systems as well as in the gastrointestinal and genitourinary tracts (Domin et al., 1986, 1987; Augood et al., 1988; Ballesta et al., 1988; Furness et al., 1989). Since the original observation of the potent contractile activity of NmU in rat uterus (Minamino et al., 1985a,b), further smooth muscle-stimulating activity has been demonstrated in human isolated ileum and urinary bladder (Maggi et al., 1990). The studies described in this report were undertaken in an effort to gain a better understanding of the role of NmU in mammalian smooth muscle physiology with the ultimate goal of determining whether NmU and its recently identified receptors (see below) are implicated in disease states of the genitourinary or gastrointestinal systems. Although sequence homology and activity are well conserved between the identified NmU peptides from various species, the function of NmU seems to be species-specific. Although NmU causes potent contraction of human urinary bladder, preliminary experiments in our laboratory could not identify the same contractile effect on urinary bladder smooth muscle from a number of other mammals, including rat, guinea pig, mouse, rabbit, and ferret. However, binding of radiolabeled NmU to canine urinary bladder membranes was detected and urinary bladder smooth muscle-stimulating activity in the canine confirmed by both in vitro and in vivo measurements of contractile activity. The effect of NmU on a number of other canine smooth muscle-containing tissues was investigated in an effort to further characterize the peripheral actions of NmU in this species and to assess the potential of this model in the future study of NmU and its receptors.

The recent cloning and subsequent identification of the ligand-receptor pairing of two NmU-activated receptors (FM-3 and FM-4 or NmU-1 and NmU-2), simultaneously by a number groups, has led to an investigation of NmU receptor tissue distribution (Tan et al., 1998; Fujii et al., 2000; Howard et al., 2000; Hosoya et al., 2000; Raddatz et al., 2000; Szekeres et al., 2000). However, a comparison of expression profiles for NmU-1 and NmU-2 in smooth muscle from the digestive and reproductive tracts where NmU has peripheral actions is lacking. Therefore, in an attempt to elucidate the receptor(s) mediating the contractile effects of NmU in these smooth muscles, quantitative PCR was used to assess expression levels of NmU-1 and NmU-2 in both human and canine tissues.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Contraction Studies. Human bladder was obtained from the National Disease Research Institute (Philadelphia, PA) and kept in iced UW transplant media until being transferred to Krebs' solution in preparation for experiments. Experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals (NIH publication 85-23). The Institutional Animal Care and Use Committee of GlaxoSmithKline approved procedures using laboratory animals. Female rats (200-300 g) were anesthetized with 5% isoflurane in O₂ and exsanguinated. Both beagle and mongrel dogs (female and male) were humanely killed with 100 mg/kg i.v. sodium pentobarbital. The uterus and urinary bladder (from rats and dogs) as well as the ileum, stomach, and colon (from dogs) were removed, cleaned of connective tissue, and strips (approximately 2-3 mm in width by 2-3 cm in length) were mounted in 10- or 20-ml vertical tissue baths. Urinary bladder strips from rat, guinea pig, mouse, rabbit, and ferret were prepared as described above. The smooth muscle strips (aside from uterine strips) were allowed to equilibrate under a resting tension of 1 to 2 g at 37°C for 1 h in a physiological salt solution of the following composition: 118 mM NaCl, 25 mM NaHCO₃, 4.7 mM KCl, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 2.5 mM CaCl₂, and 11 mM glucose; and bubbled with 95% O₂, 5% CO₂. To reduce spontaneous activity, uterine strips were bathed at 25°C in DeJalon's solution of the following composition: 154 mM NaCl, 6 mM NaHCO₃, 45.6 mM KCl, 0.4 mM CaCl₂, and 2.7 mM glucose. Tension was recorded using two systems: 1) TSD125C 50-g transducers connected to DA 100B general purpose transducer amplifiers connected to a PC-based MP100 system, including AcqKnowledge 3.5.7 software (Biopac, La Jolla, CA); and 2) Grass FT03 transducers connected to a Grass D7 polygraph (Grass-Telefactor, West Warwick, RI).

In Vivo Studies. Either mongrel or beagle female dogs (12-15 kg) were anesthetized with (6 mg/kg i.v.) propofol, followed by tracheal intubation. Anesthesia was maintained by mechanical ventilation using O₂ at 100 ml/kg/min and 1.5% isoflurane. An intravenous line was inserted into the front leg for i.v. injections and the femoral artery was cannulated to monitor blood pressure via a P23XL transducer (Gould, Cleveland, OH) connected to a Grass 7D polygraph. A 10-French catheter was inserted into the urinary bladder via the urethra. The bladder was drained of urine and then filled with 60 ml of saline. When the bladder pressure stabilized, 30 nmol/kg canine NmU-8 was injected i.v. bolus and bladder pressure was monitored.

Binding Studies. Urinary bladder was obtained as described above, cleaned of connective tissue, and weighed. Approximately 3 g of tissue was homogenized on ice using a TR-10 tissue homogenizer (Tekmar, Cincinnati, OH) for five 15-s time periods in buffer containing 20 mM Tris, pH 7.5, 0.25 M sucrose, 5 mM EDTA, 0.02% soybean trypsin inhibitor, 4 µg/ml leupeptin, 0.25 mg/ml bacitracin, 1 µM phosphoramidon, 0.2 mM phenylmethylsulfonyl fluoride, and 0.1 mg/ml aprotinin (buffer A). The homogenate was centrifuged for 10 min at 1000g (4°C) and filtered through one layer of cheesecloth before being centrifuged again at 40,000g (4°C) for 30 min. The pellet was resuspended in a solution containing 50 mM Tris, pH 7.5, and 10 mM MgCl₂ with the protease inhibitors mentioned above. HEK-293 cells expressing the canine recombinant NmU-2 were detached from 150-cm² flasks with 1 mM EDTA in Ca²⁺/Mg²⁺-free Dulbecco's phosphate-buffered saline, washed by centrifugation at 300g, and stored as a frozen pellet. The membranes were then prepared essentially as described above.

Tissue Localization Using TaqMan. Human cDNA prepared from urinary bladder, small intestine, colon, uterus, kidney, and testis was obtained from CLONTECH (Palo Alto, CA). Canine cDNA was prepared in-house from total RNA (using the oligo-dT-primed reverse transcription system; Promega, Madison, WI) isolated from 0.5- to 1-g sections of urinary bladder, uterus, stomach, and kidney using TriReagent (Sigma-Aldrich, St. Louis, MO) methodology.

Statistics. Values in the text refer to mean ± S.E.M. Statistical significance of the results was tested by Student's paired t test. Differences were considered significant when P < 0.05.

Drugs. Porcine NmU-8, porcine NmU-25, rat NmU-23, neuromedin B, and neuromedin C (American Peptide, Sunnyvale, CA; Bachem, King of Prussia, PA) as well as human NmU-25 (synthesized in-house by GlaxoSmithKline) were dissolved in distilled water and kept as 1 mM stock solutions. Bacitracin and phosphoramidon (Sigma-Aldrich) were dissolved in distilled water and kept as 2.5-mg/ml and 1 mM stock solutions, respectively. Porcine ¹²⁵I-NmU-25 and human ¹²⁵I-NmU-25 were made in-house. Atropine sulfate, carbamylcholine chloride (carbachol), and tetrodotoxin (all from Sigma-Aldrich) were also dissolved in distilled water and kept as 1 mM stock solutions.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Peripheral Effects of NmU Peptides. The effect of NmU on urinary bladder smooth muscle from a variety of species was investigated (Fig. 1). Interestingly, although confirming literature reports of the potent contractile effect of NmU on smooth muscle from rat uterus and human urinary bladder, activity in isolated urinary bladder strips from a number of small mammals, including rat, guinea pig, rabbit, mouse, and ferret, was absent (Fig. 1).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Response of urinary bladder strips from human, dog, rat, mouse, guinea pig, rabbit, ferret, and uterine strips from rat to porcine NmU-8 (10 µM) and carbachol (10 µM). n = 3 to 4 for each value. Vertical bars indicate S.E.M.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Contractile response of canine isolated urinary bladder strips to human NmU-25, porcine NmU-8, rat NmU-23, or porcine NmU-25. Each value is the mean of four to six experiments. Vertical bars indicate S.E.M.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Effect of bacitracin and TTX on responses of canine isolated urinary bladder strips to human NmU-25 (n = 3 to 4 for each point). Vertical bars indicate S.E.M.

View larger version (6K): [in this window] [in a new window]   Fig. 4.   Effect of an i.v. bolus injection of NmU-8 (30 nmol/kg) on urinary bladder pressure (a) and mean arterial blood pressure (b) in the dog.

Binding Studies. Binding studies using porcine ¹²⁵I-NmU-25 revealed saturable and high-affinity binding sites in canine urinary bladder (Fig. 5a). However, although high-affinity binding sites (K_d = 0.06 nM) were detectable, the receptor density was quite low (B_max = 24 fmol/mg). In competition binding studies the four NmU peptides displaced ¹²⁵I-NmU-25 binding to canine urinary bladder membranes with similar potencies (Fig. 5a). The K_i values were 0.11 ± 0.03, 0.08 ± 0.04, 0.15 ± 0.05, and 0.24 ± 0.04 nM for human NmU-25, rat NmU-23, porcine NmU-25, and porcine NmU-8, respectively. Structurally unrelated peptides such as neuromedin B and neuromedin C were not able to displace ¹²⁵I-NmU-25 binding at concentrations up to 10 µM.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Displacement of 125I-NmU-25 binding from canine urinary bladder membranes (a) and HEK-293 cells transiently transfected with the canine NmU-2 receptor by increasing concentrations of human NmU-25, porcine NmU-8, rat NmU-23, or porcine NmU-25 (b). Results are expressed as mean ± S.E.M. of three experiments.

Quantitative PCR. Recently, a number of groups have identified and cloned the first two NmU receptors, FM-3 and FM-4 or NmU-1 and NmU-2. Using TaqMan quantitative reverse transcription-PCR analysis, significant expression levels of NmU-1 were found in human tissues activated by NmU, including urinary bladder, small intestine, colon, and uterus (Fig. 6). In contrast, expression levels of NmU-2 were much lower or absent in these tissues (Fig. 6). Expression levels of NmU-1 and NmU-2 in the other two tissues tested (kidney and testis) were very similar.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Expression levels of NmU-1 and NmU-2 (using TaqMan PCR) in human tissues.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of this study demonstrate that the NmU peptides cause a potent contraction of canine urinary bladder both in vitro and in vivo. Because the action of NmU on isolated urinary bladder was atropine- and TTX-resistant, it would seem that NmU acts directly on the smooth muscle and intramural nerves are not involved. When contractions of canine urinary bladder strips from the dome and body were compared, no difference in responses was seen, suggesting that at least in these parts of the urinary bladder, the action of NmU is fairly uniform.

Peptidase inhibitors such as phosphoramidon and bacitracin have been shown to inhibit the degradation of various contractile peptides, including substance P and neurokinin A (Rouissi et al., 1990; Honda et al., 1991; Russell et al., 1996). These two compounds, however, do not seem to affect the potency of NmU in the canine urinary bladder. Taken together with the nature of the contraction to NmU (which is fairly well maintained for a number of minutes), it seems that endogenous peptidases do not significantly affect responses to exogenous NmU in this tissue.

Although EC₅₀ concentrations for the various NmU peptides could not be determined, it is clear from the contraction data that (at least at concentrations up to 10 µM) the various NmU peptides have similar concentration-effect curves in canine urinary bladder. This is supported by membrane binding data, which show that the four NmU peptides used in this study displace ¹²⁵I-NmU-25 binding with similar IC₅₀ values.

To further characterize the actions of NmU in the dog, the effect of NmU on a number of other intestinal smooth muscles was assessed. Similar to findings in human tissue by Maggi et al. (1990), NmU caused contraction of isolated ileum longitudinal muscle. In addition, we found that NmU caused contraction of canine stomach and longitudinal muscle from the colon.

We were able to replicate findings by two groups that the various NmU peptides cause a potent contraction of human urinary bladder smooth muscle (Maggi et al., 1990; Salmon et al., 2000). However, we were not able to do further characterization because additional tissue was unavailable. These findings prompted us to investigate whether NmU caused contraction of urinary bladder smooth muscle from other species. Surprisingly, no activity was found in any other mammalian species we examined including rat, rabbit, guinea pig, mouse, and ferret. NmU has been isolated from a number of these species, including rat, rabbit, and guinea pig, suggesting NmU plays some role in their physiology (Conlon et al., 1988; Minamino et al., 1988; Murphy et al., 1990; Kage et al., 1991). Indeed, porcine NmU-8 alters ion transport in porcine jejunum (Brown and Quito, 1988) and in contrast to our findings in the canine, has a significant hypertensive effect in rats (15-20 mm Hg at 3 nmol/kg i.v.; Minamino e al., 1985a; Gardiner et al., 1990). However, although no contractile activity was seen in guinea pig ileum (Minamino et al., 1985a) or porcine jejunum (Brown and Quito, 1988), NmU-8 did cause contraction of human as well as canine isolated ileum (Maggi et al., 1990; vide supra). Therefore, although the numerous isolated NmU peptides have high sequence homology and similar activity in responsive tissues, the function of NmU seems to be species-specific. Contractile activity in human and canine urinary bladder could suggest a physiological role for NmU in these tissues. In contrast, the lack of activity and binding of NmU in the urinary bladder of the rat (as well as the guinea pig, rabbit, mouse, and ferret) would imply that NmU is not involved in the peripheral regulation of the urinary bladder in these species.

Although the successful cloning of canine NmU-2 was achieved by our colleagues in Molecular Biology at GlaxoSmithKline, the cloning of canine NmU-1 has proved to be problematic (personal communication). However, if the binding profile for the NmU peptides in HEK-293 cells transfected with the canine NmU-2 receptor is compared with the binding profile for the NmU peptides in canine urinary bladder membranes, striking differences are apparent. Although the four peptides were approximately equipotent in the canine urinary bladder, porcine NmU-8 had much lower affinity than porcine NmU-25, rat NmU-23, and human NmU-25 at the cloned canine NmU-2 receptor, indicating the contractile response of canine urinary bladder to NmU is not mediated by this receptor.

Although the NmU peptides were first identified in 1985 and specific receptor binding in rat uterus was demonstrated in 1993 (Nandha et al., 1993; indicating the existence of at least one endogenous receptor), interest has waned with the lack of any sequenced receptors. However, the recent cloning of NmU-1 and NmU-2 has led to a renewed interest in these peptides and their biological function and allowed investigators to get a clearer and more detailed view of where NmU acts in the body (Tan et al., 1998; Fujii et al., 2000; Howard et al., 2000; Hosoya et al., 2000; Raddatz et al., 2000; Szekeres et al., 2000). Elucidating what receptors are involved in each physiological response is still quite difficult to determine from the available data. Using TaqMan quantitative PCR, Szekeres et al. (2000) found high expression levels of human NmU-1 in adipose tissue, intestine, lymphocytes, stomach, pancreas, bone marrow, and spleen. The same technology was used to detect expression of NmU-1 in human testis, uterus, prostate, kidney, and small intestine (Raddatz et al., 2000). Using Northern blot analysis, Howard et al. (2000) found message for human NmU-1 in testis, small intestine, and stomach as well as in pancreas, adrenal cortex, and liver. Message for human NmU-2, on the other hand, was virtually absent in the periphery aside from in testis, but was widespread throughout the brain (Howard et al., 2000; Raddatz et al., 2000). In the rat (like in the human), high expression levels of NmU-1 were found in small intestine, however very low levels of expression were seen in rat uterus (Fujii et al., 2000; Hosoya et al., 2000). In contrast to other peripheral tissues, high expression levels for NmU-2 were found in rat uterus (Hosoya et al., 2000).

Without specific agonists and antagonists it is quite difficult to determine which NmU receptor(s) mediates its contractile effects in smooth muscle. However, in the present study significant expression levels of NmU-1 were detected in human tissues that are stimulated by NmU, including urinary bladder, uterus, small intestine, and colon. On the other hand, little or no expression of NmU-2 was detected in these human tissues or in canine tissues that are stimulated by NmU. Based on these data from our laboratory and published expression data from other groups, it would seem likely that the NmU-1 receptor mediates the contractile effects of NmU in most peripheral tissues, including both the human and canine urinary bladder as well as portions of the intestinal tract. The main exception seems to be in the rat uterus where expression of NmU-1 is low and NmU-2 is high (Hosoya et al., 2000).

The results of this study show that NmU causes a potent contraction of smooth muscle from the urinary bladder and digestive tract in both humans and dogs, which seems to be mediated by the recently cloned NmU-1 receptor.