Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Inflammatory bowel disease is often associated with increased painful sensations in response to colorectal distension (Ritchie, 1973; Whitehead et al., 1990; Naliboff et al., 1997; Holzer, 1998; Evangelista, 2001). This hypersensitivity to mechanical stimuli has been detected in patients with ulcerative colitis as well as Crohn's disease, the two most common forms of inflammatory bowel disease (Thompson et al., 1999). In experimental models of colonic inflammation in rodents, hyperalgesic responses to colonic distension have been observed (Williams et al., 1988; Toulouse et al., 2000). Several lines of evidence suggest that hyperalgesia after irritation of the colon in both humans and animals may involve the release of inflammatory agents such as tachykinins, which have been implicated in both acute and chronic phases of intestinal inflammation (Julia et al., 1994; Maggi, 1997; Toulouse et al., 2000; Evangelista, 2001).

Tachykinins are a family of neuropeptides that includes substance P, neurokinin A (NKA) and NKB (Maggi et al., 1993; Patacchini and Maggi, 2001). Release of tachykinins can elicit a wide range of actions in a number of cell types via interaction with three types of tachykinin receptors: NK1, NK2, and NK3 (Maggi et al., 1993; Holzer and Holzer-Petsche, 2001; Patacchini and Maggi, 2001). In the gastrointestinal system, the effects of tachykinins released from extrinsic or intrinsic neurons as well as by non-neuronal cells may be mediated by stimulation of tachykinin receptors in intestinal smooth muscle, epithelial cells, as well as neurons (Holzer and Holzer-Petsche, 2001; Southwell and Furness, 2001).

There are a number of studies that support the involvement of both NK1 and NK2 tachykinin receptors in the regulation of gastrointestinal motility as well as pain sensitivity. NK1 receptors seem to be primarily involved in the physiological control of motility, whereas NK2 receptors seem to exert a prominent role in inflammation and control of motility in pathological conditions (Lecci et al., 1994; Holzer and Holzer-Petsche, 1997; Lecci et al., 1997; Maggi, 1997). For example, NK2 receptor antagonists can exert prominent effects on abdominal cramping or the number of intestinal contractions evoked by rectal distension (Julia et al., 1994; Santicioli et al., 1997; Toulouse et al., 2000; Lordal et al., 2001; Patacchini and Maggi, 2001).

To explore the involvement of NK2 receptors in visceral nociception, the present study evaluated the effect of a selective NK2 tachykinin antagonist, nepadutant (MEN 11420), in a rat model of acute colonic irritation induced by trinitrobenzenesulfonic acid (TNBS) (Traub et al., 1992; Santicioli et al., 1997; Catalioto et al., 1998). We and others have previously reported that visceral stimulation increases proto-oncogene (Fos, Jun) expression in second order spinal cord neurons as well as dorsal root ganglion (DRG) cells (Birder and de Groat, 1992, 1993; Traub et al., 1992). We used this method to evaluate proto-oncogene expression in both spinal cord (Fos) and DRG (Jun) neurons in response to a non-noxious (colorectal distension, CRD) as well as noxious (TNBS-induced irritation) stimulation of the colon. Nepadutant was evaluated for its effect on Fos or Jun induced by irritation or by low-pressure CRD, a physiological stimulus. The results obtained indicate that activation of NK2 receptors in afferent pathways may be involved in the response to colonic irritation and also that NK2 antagonists may be useful for the treatment of motility dysfunction. Preliminary reports of these findings exist in abstract form (Birder et al., 2000).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

All procedures were approved by the Institutional Animal Care and Use Committee of the University of Pittsburgh School of Medicine.

Experiments were performed on a total of 92 adult anesthetized (halothane, 2%) female Sprague-Dawley rats (200-250 g). In 24 animals, colitis was induced using a procedure described previously (Miampamba et al., 1998). Briefly, in halothane (2%)-anesthetized rats, TNBS (30 mg in 50% ethanol; Sigma-Aldrich, St. Louis, MO) was instilled via a cannula into the colon 6 cm proximal to the anus. After 30 min, the remaining solution was withdrawn. A separate group of rats (n = 24) received repetitive (5 min on/5 min off) constant pressure inflation (10 mm Hg) of a balloon catheter (size 6 cm; Bio-Tek Instruments, Inc., Winooski, VT) for 2 h to produce CRD.

The following five control groups were used: 1) normal animals not subjected to any experimental procedure; 2) animals in which saline was instilled via a cannula into the colon; 3) animals in which ethanol vehicle was instilled via a cannula into the colon; and 4 and 5) intravenous administration of saline in place of the NK2 antagonist in two experimental preparations, TNBS-induced colonic irritation or non-noxious CRD.

The effect of nepadutant on non-noxious CRD (10 mm Hg) and TNBS-induced inflammation was evaluated in rats (n = 32) by pretreating with either vehicle or nepadutant (200 nmol/kg i.v.) at least 20 min before colonic stimulation. In preliminary experiments, we also evaluated the effects of lower doses of nepadutant (50-150 nmol/kg i.v.). Two hours after instillation of saline, vehicle, or TNBS into the colon, all animals were sacrificed by intracardiac perfusion with cold Krebs' buffer followed by paraformaldehyde fixative. This time period was previously demonstrated to be the peak for irritation induced proto-oncogene expression. The tissue (DRG or spinal cord) was removed, postfixed overnight, and cryoprotected in 25% sucrose. Serial frozen sections of spinal cord (28 µm) were processed for immunoreactivity (IR) to Fos protein, and DRG sections (14 µm) were processed for immunoreactivity to Jun protein using specific (rabbit polyclonal) antisera. Control tests conducted without primary antisera or antisera preabsorbed with these proteins eliminated the staining.

To identify colon DRG neurons, a dye tracer (fast blue, 4% in saline) was injected using aseptic procedures into the colon wall in a subset of halothane-anesthetized rats. After 1 week, the rats were subjected to various experimental paradigms (detailed above) and perfused with fixative. Transverse serial frozen sections of DRG neurons were processed for Jun as described above. Epifluorescence or transmitted light microscopy was used to examine the processed sections and to capture images using a high-resolution color camera.

A separate group of rats (n = 10) was pretreated with either capsaicin (100 mg/kg s.c.) or capsaicin vehicle (10% Tween 80, 20% ethanol, 70% saline, n = 3) 4 to 5 days before induction of colitis using TNBS. All capsaicin-pretreated animals were unresponsive to application of a dilute solution of capsaicin (20 µg/ml) to the surface of the eye (indicative of desensitization).

In a separate group (n = 10) of rats, 2 h after instillation of saline, vehicle, or TNBS, the animals were sacrificed and the colon removed and postfixed. Sections of the colon were stained for hematoxylin-eosin and evaluated microscopically for signs of inflammation.

Counts of Fos-positive spinal cord cells in four spinal cord regions: medial dorsal horn (MDH), lateral dorsal horn (LDH), dorsal commissure (DCM), and the autonomic region near the sacral parasympathetic nucleus (SPN) (Birder and de Groat, 1993) on both sides of the spinal cord are presented as number of cells per spinal cord section or as the percentage of change compared with control. Sections used for the counts were separated by at least 100 µm to avoid double counting. For counts of Jun-positive DRG neurons, the number of labeled neurons in each DRG section was estimated from counts of positively stained profiles containing nuclei in at least 10 sequential sections separated by 100 µm.

Statistics. All counts are presented as the mean number of positive neurons per section or percentage of increase or decrease (for each group of experiments) ± S.E.M. Student's t test was used to estimate the probability that differences between means of the experimental groups could have occurred by chance.

Materials. Chemicals were obtained as follows: normal goat serum, biotinylated goat anti-rabbit IgG, and ABC Elite kit with VIP substrate were from Vector Laboratories (Burlingame, CA). FOS and JUN primary antisera were purchased from Calbiochem (San Diego, CA). Other chemical compounds were obtained from Sigma-Aldrich. Slides (SuperFrost) used to mount the tissue sections were obtained from Fisher Scientific (Pittsburgh, PA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Control Experiments

Control animals not subjected to experimental procedures exhibited low basal levels of Fos in L₁-S₂ segments of the spinal cord (4 cells/section, n = 5) (Fig. 1A). Increased numbers of Fos-positive cells in the L₆-S₁ spinal cord (range 20-35 cells/section) were also detected after either saline (range 11-29 cells/section) or ethanol (17-35 cells/section) instillation into the colon (Fig. 1A).

View larger version (34K): [in this window] [in a new window]   Fig. 1.   Number of Fos-positive neurons after non-noxious CRD or TNBS-induced colitis. A, histogram depicts the mean number of Fos-positive neurons per section in the L6 rat spinal cord. Control: untreated rats (n = 8); saline: intracolonic saline instillation (n = 8); ETOH: intracolonic ethanol (50%) instillation (n = 8); CRD: repetitive (5 min on/10 min off) constant pressure (10 mm Hg) non-noxious distension (n = 8); TNBS: intracolonic TNBS instillation (n = 8); CapRx: pretreatment with capsaicin (100 mg/kg s.c.) 4 days before TNBS (n = 4); and nepadutant + TNBS: treatment with nepadutant (200 nmol/kg i.v.) before TNBS (n = 8). B, segmental distribution of Fos-IR after colon irritation. Histogram showing the segmental distribution of Fos-positive cells (number/section) in the rat spinal cord (L1-S2) after TNBS-induced irritation of the colon. C, histogram depicting the percentage of the total number of Fos-positive spinal cord neurons located in four L6 spinal cord regions (MDH, LDH, DCM, and SPN) after either non-noxious CRD (solid columns) or TNBS treatment (hatched columns). D, histogram depicting the percent decrease after nepadutant treatment (200 nmol/kg i.v.) in the number of Fos-positive spinal cord neurons in four L6 spinal cord regions (MDH, LDH, DCM, and SPN) induced by either non-noxious CRD (solid black columns, n = 8) or intracolonic instillation of TNBS (hatched columns, n = 8). Values represent mean ± S.E.M., *, P < 0.05.

Proto-Oncogene (c-fos, c-jun) Expression Induced by Stimulation of Colon

TNBS-Induced Experimental Colitis. Instillation of TNBS into the colon of rats produces colitis evidenced by inflammation and ulceration of the colon as noted by Miampamba et al. (1998) and Khan (2002). Histological evaluation of colonic tissue taken from animals instilled with TNBS revealed regions of inflammation and tissue damage, including disruption of the epithelial lining and edema in both mucosa and submucosa, that were not evident in tissue from normal control rats (data not shown). TNBS significantly increased (8-10-fold increase above control) the number of Fos-positive cells in all four regions of the L₆-S₁ spinal cord: MDH, LDH, DCM, and SPN (Figs. 1, A-C, and 2A). Relatively small increases (10-50 cells/section) in c-fos expression were detected in adjacent segments (L₁-L₄, S₂) (Fig. 1B). TNBS significantly (>50%) increased the number of JUN-IR neurons as well as those dye-labeled as colorectal L₆-S₁ DRG neurons (Figs. 3B and 4C) with mainly small- to medium-diameter soma (18-40 µm).

View larger version (56K): [in this window] [in a new window]   Fig. 2.   NK2 antagonist nepadutant decreases Fos-IR after TNBS-induced colitis. A, bright-field photomicrograph of L6 rat spinal cord section depicting Fos-IR 2 h after TNBS-induced colitis. B, photomicrograph from L6 rat spinal cord depicting Fos-IR in an animal pretreated with nepadutant 20 min before TNBS instillation. Scale bar, 55 µm.

View larger version (81K): [in this window] [in a new window]   Fig. 3.   Effect of colorectal distension or TNBS-induced irritation on c-jun expression in dorsal root ganglia. Bright-field photomicrographs depicting Jun-IR in L6 DRG neurons (shown at arrow) in an untreated rat (A) or 2 h after TNBS-induced colitis (B). C, histogram depicts the number of Jun-IR cells/L6 DRG section in control animals or after either CRD or TNBS. Values represent mean ± S.E.M. Scale bar, 20 µm.

Non-Noxious CRD. CRD (repetitive distension of the colon at 10 mm Hg pressure), below the threshold for production of visceromotor responses and considered to be non-noxious (Traub et al., 1992), induced small numbers of Fos-positive cells (range 16-31) in the lower L₆-S₁ spinal cord (Fig. 1A). Fos-positive cells were located in the dorsal horn (medial and lateral), DCM, and SPN regions (Fig. 1C). The distribution of Fos-positive cells was similar to that occurring after TNBS treatment. The numbers of Jun-IR cells in DRG were similar to controls (Fig. 3C).

Effect of Capsaicin Pretreatment on Proto-Oncogene Expression Induced by TNBS

Capsaicin has been shown to affect a subset of small-diameter polymodal nociceptive afferents first by stimulating then desensitizing the response to noxious stimuli (Maggi, 1993). To determine whether capsaicin-sensitive primary afferent neurons contribute to the increase in Fos expression in spinal cord neurons after exposure to TNBS, a group of rats (n = 4) was pretreated with capsaicin before exposure to the irritant. Capsaicin-treated animals did not respond to application of a dilute solution of capsaicin (20 µg/ml) to the surface of the eye (a negative eye-wipe test indicates desensitization of corneal afferents). In the absence of TNBS induced colitis, Fos expression was low in animals pretreated with capsaicin vehicle (10% Tween 80, 20% ethanol, 70% saline, n = 3) or in animals pretreated with capsaicin 1 week before the experiment (n = 3). However, in capsaicin-pretreated animals, the Fos response to TNBS-induced irritation was significantly reduced (85% decrease) in all spinal cord regions compared with the response to TNBS in untreated animals (Fig. 1A).

Effect of Nepadutant on Proto-Oncogene Expression after either Non-Noxious CRD or Colon Irritation (TNBS)

The L₆-S₁ level of the cord was used to evaluate the effect of the neurokinin antagonist. Pretreatment (20 min) with nepadutant before non-noxious CRD did not have a significant effect on Fos-expression (Fig. 1D). However, pretreatment with nepadutant (200 nmol/kg i.v., 20 min) before exposure of the colon to TNBS significantly decreased the number of Fos-positive neurons (70% of control) in all four spinal cord regions (MDH, LDH, DCM, and SPN) (Figs. 1, A and D, 2B). Lower concentrations of the antagonist (50-150 nmol/kg i.v.) did not significantly alter proto-oncogene expression after TNBS-induced colitis (data not shown).

Effect of CRD or TNBS on JUN-IR in DRG Neurons

Injury or inflammation is known to increase the expression of the c-jun proto-oncogene in DRG neurons. As noted by other investigators, in the absence of nociceptive stimuli DRG from control animals exhibited low levels of c-jun expression (Fig. 3, A and C). In contrast, 2 h after exposure of the colon to TNBS, the number of DRG neurons expressing JUN-IR significantly increased (2-3-fold increase; Fig. 3C) compared with either control or vehicle-treated animals (data not shown). This increase was localized to both small- (presumably nociceptors) and medium-diameter DRG neurons.

Effect of Nepadutant on TNBS-Induced JUN-IR in Colon DRG Neurons

JUN-IR was also detected in only small numbers (<5 cells/section) of dye-labeled colon DRG neurons from control animals (Fig. 4) or in vehicle-treated animals (data not shown). Pretreatment with nepadutant before induction of TNBS-induced colitis significantly decreased (*P < 0.05) Jun-IR in both small- and medium-diameter colon DRG cells but had no effect on the number of Jun-IR-positive colon DRG cells after non-noxious CRD (Fig. 4C).

View larger version (47K): [in this window] [in a new window]   Fig. 4.   Effect of nepadutant on Jun-IR in identified colon L6 DRG neurons after either colorectal distension or irritation. A, photomicrograph depicting dye-labeled colon DRG neurons after injection of a fluorescent dye into the colon (left, shown at arrow) and in an adjacent section, the same DRG neuron that also expressed Jun-IR. B, histogram depicting the number of Jun-IR cells/L6 colon DRG neurons after either non-noxious CRD or TNBS alone (solid columns) and the effect of nepadutant (hatched columns) after either non-noxious CRD (n = 8) or TNBS treatment (n = 8). Values represent mean ± S.E.M.. N.S., not significant. *, P < 0.05.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our studies revealed that the peptide NK2 antagonist nepadutant significantly suppresses the increased expression of c-fos and c-jun in spinal cord neurons and colon DRG neurons, respectively, after TNBS-induced colitis in the rat but did not alter the increased proto-oncogene expression in spinal cord and in colon DRG neurons induced by non-noxious CRD (10 mm Hg). These results indicate that nepadutant is more effective at suppressing afferent activity evoked by inflammation of the colon than the activity induced by non-noxious mechanical distension in the absence of inflammation. This selective action of nepadutant may reflect, in part, the involvement of different primary afferent neurons and/or different peripheral mechanochemical mechanisms underlying the afferent activation by these two types of stimuli. These data suggest that nepadutant possesses specific visceral analgesic- and antihyperalgesic actions. The effect of the NK2 antagonist nepadutant to reduce the proto-oncogene expression after irritation of the colon is consistent with both electrophysiological and behavioral studies in the rat that demonstrate this compound is effective in decreasing locomotive activity or hypersensitivity (as measured by abdominal striated muscle contractions) after colorectal inflammation in rodents (Toulouse et al., 2000).

Fos, the protein product of c-fos, can be expressed in specific populations of second order spinal cord neurons after non-noxious as well as nociceptive stimuli (Birder and de Groat, 1992, 1993; Traub et al., 1992). Our observation that c-fos expression in spinal cord neurons after TNBS-induced colitis is significantly elevated is consistent with the demonstration that TNBS-induced colon irritation activates a population of polymodal afferents and second order spinal cord neurons (Traub et al., 1992). Systemic pretreatment with the C-fiber neurotoxin capsaicin significantly reduced the TNBS-induced expression of Fos-IR in spinal cord neurons as noted after noxious chemical stimulation of the urinary bladder (Birder and de Groat, 1998), suggesting that capsaicin-sensitive, peptidergic afferents transmit noxious input from the colon to the cord. The similarity in the effects of capsaicin and nepadutant suggests that the NK2 antagonist alters the activity of capsaicin-sensitive C-fiber afferent neurons.

Tachykinin NK2 receptor antagonists have been demonstrated to be effective in a number of animal models of visceral inflammation, suggesting that NK2 receptors play an important role in the initiation of visceral hyperalgesia. The effect of nepadutant is attributable to an action on peripheral NK2 receptors to decrease the responses to nociceptive stimuli because the drug does not penetrate the blood-brain barrier (Catalioto et al., 1998; Laird et al., 2001). This is consistent with the effect of the drug to decrease Jun in DRG neurons. In the absence of irritation, our data indicate that colon distension pressures of 10 mm Hg, which are assumed to be non-noxious, produced smaller increases in Fos-IR in the spinal cord compared with TNBS-induced colitis. This is consistent with previous observations demonstrating that physiological distension of the urinary bladder could activate a significantly smaller number of Fos cells in the cord compared with noxious stimulation (Birder and de Groat, 1993). Our findings that nepadutant was not effective in reducing the Fos response to distension alone is consistent with the observation that nepadutant reduces the firing of spinal neurons to CRD after inflammation of the colon (Laird et al., 2001), whereas in the absence of colonic inflammation, treatment with nepadutant was ineffective in reducing the neuronal response to CRD.

Although the site of action of nepadutant in the peripheral nervous system was not established in the present experiments, it seems likely that the drug could act directly on NK2 receptors on afferent nerves that express these receptors or on non-neuronal cells (mast cells) (Maggi, 1993, 1997; Sculptoreanu and de Groat, 2000). The receptors could in turn be activated by tachykinins released from intrinsic neurons, primary afferent nerves, or non-neuronal cells in the gut. NK2-positive nerve fibers are in part localized in the submucosa where they may be in functional communication with epithelial or other non-neuronal cells (Grady et al., 1996). Release of tachykinins from afferent nerves could potentially activate autoreceptors on afferent nerves as well as non-neuronal cells (mast cells, immune cells, and epithelial cells) that are in proximity to the nerves. Certain types of non-neuronal cells such as immune cells not only express NK2 receptors but also may synthesize/release tachykinins (Maggi, 1997; Brunelleschi et al., 1998). Thus, non-neuronal as well as neuronal release of tachykinins could alter the responsiveness of nearby non-neuronal cells, including that of smooth muscle as well as the activity of underlying nerves. This type of complex interaction would not be unique to the gastrointestinal system, as a similar relationship has been suggested to exist between afferent fibers, epithelial, and other non-neuronal cells in the airway, the urinary bladder, as well as in the inner ear (Hudspeth and Corey, 1977; Folkerts and Nijkamp, 1998; Birder et al., 2001). Although further studies are necessary to further explore the mechanism by which nepadutant alters the afferent mechanisms during colitis, taken together, these findings support the claim that NK2 receptors are involved in the colorectal hypersensitivity associated with inflammation.

In conclusion, the present study demonstrates that both non-noxious and noxious irritation of the colon activates colon afferents and results in increased proto-oncogene expression in spinal cord and DRG neurons. The increased gene expression in neurons induced by irritation but not by non-noxious distension is significantly reduced in all spinal cord laminae after pretreatment with the NK2 receptor antagonist nepadutant. Taken together, these results provide evidence that irritation of the colon activates NK2 receptors and increases activity of nociceptive afferents. Thus, a therapy that includes NK2 receptor antagonists may be beneficial for the treatment of colitis and chronic visceral dysfunction.