Rapid-Onset, Short-Duration Induction of Colorectal Contractions in Anesthetized, Adult, Male Rats

Abstract

Substantial clinical and preclinical evidence indicates that transient receptor potential vanilloid 1 (TRPV1) receptors are expressed on terminals of colorectal chemoreceptors and mechanoreceptors and are involved in various rectal hypersensitivity disorders with common features of colorectal overactivity. These stimulatory properties of TRPV1 receptors on colorectal function suggested that brief stimulation of TRPV1 might provide a means of pharmacologically activating the colorectum to induce defecation in patients with an “unresponsive” colorectum. The current studies explored the basic features of TRPV1 receptor-induced contractions of the colorectum in anesthetized rats with and without acute spinal cord injury (aSCI). Cumulative concentration–response curves to intrarectal (IR) capsaicin (CAP) solutions (0.003%–3.0%) were performed in anesthetized aSCI and spinal intact rats. CAP produced an “inverted U,” cumulative concentration–response curve with a threshold for inducing colorectal contractions at 0.01% and a peak response at 0.1% and slight decreases in responses up to 3%. Decreases in responses with concentrations >0.1% are due to a rapid desensitization (i.e., ≤30 minutes) of TRPV1 receptors to each successive dose. Desensitization appeared fully recovered within 24 hours in spinal intact rats. Colorectal contractions were completely blocked by atropine, indicating a reflexogenic activation of parasympathetic neurons, and responses were completely unaffected by a neurokinin 2 receptor antagonist, indicating that release of neurokinin A from afferent terminals and subsequent direct contractions of the smooth muscle was not involved. IR administration of three other TRPV1 receptor agonists produced similar results as CAP.

SIGNIFICANCE STATEMENT

Individuals with spinal cord injury often lose control of defecation. Time-consuming bowel programs using digital stimulation of the rectum are used to empty the bowel. This study shows that intrarectal administration of the transient receptor potential vanilloid 1 (TRPV1) receptor agonist, capsaicin, can induce rapid-onset, short-duration colorectal contractions capable of inducing defecation in spinal cord injured and intact rats. Therefore, TRPV1 agonists show promise as potential therapeutics to induce defecation in individuals with neurogenic bowel.

Introduction

The inability to initiate defecation is a distressing condition that can result from benign conditions such as idiopathic constipation, as well as severe conditions such as spinal cord injury or multiple sclerosis (Simpson et al., 2012; Callaghan et al., 2018; Bourbeau et al., 2020). Benign conditions rely on enemas and laxatives to facilitate bowel movements, whereas severe conditions can require digital stimulation of the rectum and manual extraction of stool, in addition to enemas and laxatives (Cotterill et al., 2018). Enemas and laxatives are contraindicated for continuous use, and the timing of their onset of action and the duration of action are difficult to predict, which can cause issues with fecal incontinence, while manual bowel programs are very time consuming (hours), uncomfortable, and stigmatizing for the subject and their caregivers (Wheeler et al., 2018). Obviously, a drug that produces safe and efficient defecation with a rapid onset of action (i.e., <5 minutes) without lingering effects (i.e., <15 minutes) could provide a useful therapy for many, especially the spinal cord injured, who cannot voluntarily initiate defecation under most circumstances (Callaghan et al., 2018; Cotterill et al., 2018).

Transient receptor potential vanilloid 1 (TRPV1) receptors are a class of ion channel receptors that respond to temperature, acidic environments, various endogenous substances, and ingredients contained in hot peppers, such as capsaicin (CAP) (Caterina et al., 1997; Yang and Zheng, 2017). TRPV1 receptors are located in the colorectum and trigger the urge to defecate when stimulated (Spencer et al., 2008; Matsumoto et al., 2009). These effects led us to propose that TRPV1 receptor agonists might be useful to induce defecation. This proposal was first tested by measuring colorectal contractions in anesthetized, male rats, with and without acute spinal cord injury (aSCI). Because TRPV1 receptors exhibit substantial desensitization to repeated application of CAP in other tissues, rectal TRPV1 receptor desensitization was measured 1 hour after a conditioning dose of CAP was administered.

Methods

Animals

Male, Sprague–Dawley rats (2–4 months old, Charles River Laboratories, Raleigh, NC) were housed in standard rodent cages with wood chip bedding (2–3 animals per cage) in a temperature and humidity-controlled vivarium on a 12 hour/12 hour light/dark cycle. Environmental enrichment was provided in the home cage (plastic red translucent tube and crinkle nesting paper) and food and water were available ad libitum. All experiments conformed to National Institutes of Health Guidelines for the Care and Use of Laboratory Animals and were approved by the Integrated Laboratory Systems Animal Care and Use Committee. At the end of the experiment, rats were euthanized with an overdose of urethane followed by thoracotomy.

Surgical Procedures.

Rats (total of N = 96) were anesthetized with subcutaneous urethane (1.2–1.4 g/kg), which was supplemented by isoflurane (1%–2% in O₂) during the surgical procedures. Blood pressure was monitored via a cannula inserted into the carotid artery, and intravenous drugs were delivered through a cannula inserted into the jugular vein. After the surgical procedures, fecal pellets were manually removed from the rectum by palpation over the skin and by manual rectal stimulation with a saline-soaked cotton swab to remove fecal pellets from the rectum. Colorectal pressure (CRP) was monitored via a balloon catheter inserted ∼3 cm into the rectum through the anus and connected to a pressure transducer (Utah Medical Products; DelTran II) and bridge amplifier (Transbridge 4M, World Precision Instruments). The colorectal catheter was slowly filled with saline in 0.05-ml increments until a stable baseline of 10–15 mm Hg was attained.

In 51 rats, the spinal cord was transected at T3/4 as previously described (Rupniak et al., 2023) under full anesthesia using aseptic procedures. A complete spinal transection at T3 was chosen because individuals with high thoracic lesions (above T6) have a high incidence of autonomic dysreflexia, and bowel stimulation can induce autonomic dysreflexia (Inskip et al., 2018). Briefly, the skin and muscle over the upper-middle thoracic vertebrae were incised; the spinal cord was exposed by a laminectomy, and transected at T3/4 followed by placement of gel-foam between the transected cord. The muscle and skin overlying the injury site was sealed with wound clips. An additional nine rats, used in the neurokinin 2 receptor and muscarinic receptor antagonist studies, were transected at T9 as previously described (Marson et al., 2018).

Compounds and/or Drugs.

CAP was administered intrarectally 2–4 cm from the anal opening as a 0.1–0.2 ml of solution (0.003%–3% w/v in a mixture of 0.03%–28% ethanol in water) via a PE10 cannula or FN20-30 feeding tube. The TRPV1 agonists, nonivamide (J&K Scientific), N-oleoyldopamine, (OLDA, Alomone Laboratories), and oleoylethanolamide (OEA, Tocris) were solubilized in Tween-80:ethanol:saline (1:1:8 ratio). Because intrarectal (IR) administration of these TRPV1 receptor agonists have not been performed to our knowledge, we chose concentrations of nonivamide, OLDA, and OEA, based on their relative potencies at TRPV1 receptors compared with CAP. The neurokinin 2 receptor (NK2R) agonist [Lys5, MeLeu9, Nle10] NKA(4–10)] (LMN-NKA, Genscript) was administered in saline via an intravenous bolus injection through a cannula placed in the jugular vein. The NK2R antagonist, GR159897 (Tocris) was prepared at 50 mM (22.65 mg/ml) in DMSO and diluted to 1 mg/ml with saline the day of the experiment and administered via a subcutaneous injection. DMSO 4.8% in saline was used for the vehicle. GR159897 was given at least 10 minutes before the agonists. Doses of LMN-NKA and GR159897 were determined based on our previous studies in rats (Marson et al., 2018; Cook et al., 2022). The muscarinic agonist bethanechol (Abcam) was diluted in saline to 0.1 mg/ml and given intravenously at 0.1 mg/kg. The nonselective muscarinic antagonist, atropine (Sigma-Aldrich) was diluted in saline as a 0.5 mg/ml solution (2.5 mg in 5 ml) and given intravenously at a dose of 0.5 mg/kg. Atropine was given at least 5 minutes before the agonists. Doses of bethanechol and atropine used were based on doses previously used in rats (Takeuchi et al., 1990; Kullmann et al., 2008). All drug administration took place at least 45 minutes after the completion of surgical procedures.

Data Analysis.

Arterial blood and colorectal pressures were recorded continuously throughout the experiment via LabChart software through a PowerLab/8SP data acquisition system (version 7 and 8, ADInstruments). Mean arterial pressure (MAP) and heart rate were calculated offline from the blood pressure signal. Baseline pre-dose and post-dose values were used to calculate the percentage change in MAP and heart rate. Colorectal activity was calculated as a change from pre-dose baseline values. Blinding was not performed during the dose administration. The data analysis was done offline by two people who independently analyzed and reviewed the data. Animals were removed from analysis if they were outliers based on values that were 2 S.D. above or below the mean of the total population. This resulted in removing one rat from each dose group as a statistical outlier. Increasing concentrations from 0.003% to 3% were administered every 5 minutes until the concentration–response curve was completed. The baseline value in each experiment was measured prior to administering any CAP and was used to calculate change in baseline for each concentration. The peak CRP was measured and area under the CRP-time curve (AUC, 0–5 minute post-dose) was calculated. Because higher concentrations of ethanol were required to solubilize higher CAP concentrations, we examined the CRP responses to three vehicle concentrations of ethanol/water (0.03%, 2.85%, and 28.5%). No significant increases in CRP were observed to any concentration of ethanol, and there was no difference between the CRP responses of spinal intact compared with aSCI animals to the vehicle solutions (one-way ANOVA; intact rats: CRP P = 0.3869 and CRP AUC P = 0.3116; aSCI rats: CRP P = 0.8500 and CRP AUC P = 0.3805). Therefore, the data from all the vehicle trials were grouped for comparison with CAP (Fig. 2). To examine if desensitization occurred with the high concentrations, in seven animals the highest concentration of CAP (either 1 or 3%,) was readministered an hour after the first high dose. Statistical analysis was performed using computer software (GraphPad Prism version 10) with either a one-way or two-way ANOVA, followed by Dunnett’s or Tukey’s multiple comparisons tests, as appropriate. A P value <0.05 was considered statistically significant. Data are presented as mean and S.D.

Results

IR administration of CAP solutions initiated colorectal contractions within 1 minute. Contractions peaked at 2 minutes and lasted for approximately 4 minutes (Fig. 1). Administration of increasing, cumulative concentrations of CAP solution to individual rats showed concentration-dependency of colorectal contractions and responder rates from 0.003% to 0.1% (Figs. 1 and 2, A and B). At a concentration of 0.1% CAP, 92% of the spinal intact and 83% of the aSCI rats responded with a colorectal contraction >10 mm Hg (Table 1), which on occasion expelled the recording balloon through the anus. (The balloon was reinserted before the subsequent dose was administered.) Colorectal contractions and responder rates plateaued or decreased at higher concentrations (0.3%–3%) (Fig. 2, A and B). MAP rose ∼10 mm Hg at higher concentrations (0.3% and 1%) of CAP in aSCI rats, but not spinal intact rats, with no changes in heart rate in either group (Fig. 3). Responses to direct colorectal smooth muscle stimulation with LMN-NKA, prior to and after cumulative dosing with CAP (n = 96), indicated that colorectal smooth muscle strength was the same at the end of the experiment as at the beginning. The ability of TRPV1 receptor agonists to initiate colorectal contractions was examined for three other TRPV1 receptor agonists: nonivamide, OLDA, and OEA. Each of these agonists initiated colorectal contractions that were similar in magnitude, time to onset, and duration of the contraction at similar concentrations as CAP (Fig. 4). The ethanol/water vehicles did not induce a change in colorectal pressure at any concentration (for example for 28% ethanol in water the CRP mean ± S.D. was 5.09 ± 4.63, N = 13 and 7.9 ± 7.8, N = 11 for spinal intact and aSCI rats, respectively).

TABLE 1.

Number of animals responding to IR dosing of CAP in intact and aSCI rats

CAP (%)	0	0.003	0.01	0.03	0.1	0.3	1	3
Intact	7/31 (22)	2/10 (10)	4/10 (40)	11/12 (92)*	11/12 (92)*	7/11 (64)*	6/9 (66)*	7/10 (70)*
aSCI	3/36 (8)	1/12 (8)	6/12 (50)*	10/12 (83)*	10/12 (83)*	10/13 (77)*	6/10 (60)*	6/11 (55)*

Rats received vehicle (0) and cumulative doses of CAP. A change in colorectal pressure of ≥10 mm Hg from the pre-CAP baseline value was used to identify responders. Values indicate the number of animals responding/total number of animals tested × 100 (responder rate). *Indicates responder rates significantly different to vehicle (P < 0.05 Fisher’s exact test).

Fig. 1.

Physiograph tracings of blood pressure and CRP in response to cumulative doses of CAP administered intrarectally at 5-minute intervals. The concentration of CAP is shown below the arrows and is expressed as a % concentration in ethanol/water vehicle. (A) In a spinal intact rat, CAP (0.01%–0.3%) evoked a rapid increase in CRP. At 1% CAP, the response was delayed and smaller and no response was seen at 3% CAP. (B) In the acute spinal cord injured rat, CAP (0.01%–0.03%) evoked a rapid increase in CRP. At higher doses of 0.1%–3%, the response was either markedly reduced or absent. No obvious drug-induced changes in blood pressure were observed.

Fig. 2.

CRP responses to cumulative doses of IR CAP. Complete desensitization to repeated IR administration of high dose CAP in aSCI rats. Cumulative doses of CAP were administered at 5-minute intervals. The % concentration of is shown on the horizontal axis. CAP increased (A) CRP and (B) CRP AUC in both intact and aSCI groups (pressure: [F(7,214) = 11.44, P < 0.0001; AUC: F(7,208) = 9.697, P < 0.0001]. *Indicates a statistical difference from vehicle (0), Tukey’s multiple comparison test. No group difference was observed in CRP [F(1,214) = 0.2400, P = 0.6247], but a difference was observed in the colorectal AUC [F(1,208) = 4.064, P = 0.0451] with no dose matched difference between intact and aSCI. The vehicle (0) groups include three doses of ethanol (0.03%, 2.85%, and 28.5% ethanol in water). Intact: N = 31, aSCI N = 36 for 0; intact N = 9–12, and aSCI N = 10–13 for each concentration of CAP. Complete desensitization was observed to a high dose (1% or 3%) of CAP (second dose) administered 1 hour after the first high dose administration (first dose). Both the peak (C) CRP and (D) CRP AUC responses were nearly abolished. *Indicates P < 0.05 compared with the previous response to CAP, Student’s paired t test, n = 7.

Fig. 3.

Effects of IR CAP on MAP and heart rate in anesthetized intact and aSCI rats. CAP (0.3%–1%) transiently increased MAP by less than 15% in aSCI [F(7,206) = 3.616, P = 0.0011], but not in intact rats. *Indicates a difference from vehicle (0) within group. A significant interaction [F(7,206) = 2.688, P = 0.0110] and difference between intact and aSCI groups [F(1,206) =14.60, P = 0.0002] was observed. No difference in heart rate was observed with dose [F(7,212) = 1515, P = 0.1635] or between groups [F(1,212) = 2.390, P = 0.1236]. Vehicle, (0) N = 26; CAP, N = 11–14/group.

Fig. 4.

Physiograph tracings of CRP in response to TRPV1 agonists, OLDA, OEA, and nonivamide. All three agonists evoked a rapid increase in CRP. The wide black arrow indicates when a cotton plug (0.1–0.15 g weight saturated in ∼0.1–0.2 ml solution) was placed into the rectum. The concentration of each dose is shown near the arrow at the horizontal axis. The black downward arrows indicate when the cotton plug was removed. *The colorectal balloon catheter was expelled.

To determine if the colorectal response to CAP was due to reflex activation of parasympathetic efferent neurons or due to release of neurokinin A from the colorectal afferent terminals, rats were administered the muscarinic cholinergic receptor antagonist, atropine, or the NK2R antagonist, GR159897. To confirm that the atropine dose was adequate; bethanechol, the muscarinic cholinergic receptor agonist was administered before and after atropine. To confirm that the dose of GR159897 was adequate; LMN-NKA, was administered before and after GR159897. As shown in Fig. 5A, a dose of atropine that blocked the bethanechol-induced colorectal contraction also completely blocked the response to CAP (but had no effect on the LMN-NKA–induced contraction). As shown in Fig. 5B, a dose of GR159897 that completely blocked the LMN-NKA–induced colorectal contraction had no effect on the CAP-induced contraction.

Fig. 5.

CAP-induced increase in CRP is mediated via muscarinic receptor activation in an aSCI rat. (A) Physiographs are shown from a single rat that was administered the drugs in the sequence shown. Atropine (0.5 mg/kg) blocks the CRP increase induced by bethanechol and blocks the CAP (0.1%–1%)-induced CRP activity. In the presence of atropine, the NK2 agonist, LMN-NKA, still induces a smooth muscle-mediated increase in CRP. (B) CAP-induced increase in CRP is not mediated via NK2R activation in an aSCI rat. Physiographs are shown from a single rat that was administered the drugs in the sequence shown. The NK2R agonist, LMN-NKA, induced a robust CRP response that was blocked by the NK2R antagonist, GR159897 (1 mg/kg). Administration of GR159897 did not block the CAP-induced increase in CRP. The concentrations of CAP are expressed as a % concentration.

The decrease in colorectal responses and responder rates (Fig. 1, A and B; Table 1) with cumulative dosing of CAP suggested tachyphylaxis due to TRPV1 receptor desensitization produced by the previously administered doses. Figure 6 shows that colorectal responses to administration of the same concentration of CAP, administered three times, produced decreased responses at 10 and 30 minutes after the first administration, with the highest concentration (1%) producing no response after 30 minutes. Responses to CAP applied 1 hour after a previous high dose application consistently showed complete desensitization (Fig. 2, C and D). No significant decrease in NK2R-induced colorectal contractions was observed in any rats when LMN-NKA was administered immediately prior to or immediately after complete desensitization to CAP (n = 7). Rats that were tested 24 hours after complete desensitization to CAP the day before showed complete recovery of colorectal responses (n = 7).

Fig. 6.

Representative colorectal pressure traces showing desensitization during repeated IR administration of CAP in three aSCI rats. CAP was administered over 2 minutes via a cotton plug soaked with 0.2 ml of CAP solution at time = 0, 10 minutes, and 30 minutes. Upward pointing arrows indicate when the CAP-saturated plug was administered, and downward arrows indicate removal of the plug. (A) CRP following repeated administration of 0.01% CAP. (B) CRP following repeated administration of 0.1% CAP. (C) CRP following repeated administration of 1% CAP.

Discussion

These studies demonstrate that IR administration of TRPV1 receptor agonists, including CAP, can rapidly (in 1 minute) induce colorectal contractions. The magnitude of the colorectal contractions produced by CAP is similar to NK2R agonist-induced colorectal contractions that in our previous studies induced functional defecation (Marson et al., 2018, 2020). Although fecal pellets were manually removed from the rectum during surgery, mid to high dose CAP administration sometimes resulted in expulsion of the colorectal catheter followed by fecal pellet excretion. Furthermore, our preliminary behavioral studies in conscious rats have shown that IR CAP produces rapid defecation within 5 minutes at concentrations that are well-tolerated (unpublished data).

Based on blockade of CAP-induced colorectal contractions by atropine, we propose that CAP-activated TRPV1 receptors on intrarectal primary afferent terminals that reflexively activated parasympathetic neurons leading to colorectal contractions (Matsumoto et al., 2009, 2011). Whether the afferent terminals mediating the reflex are on intrinsic (i.e., myenteric plexus and submucosal plexus sensory neurons) or extrinsic (i.e., sacral dorsal root ganglion neurons) requires additional studies, such as repeating our study after sacral rhizotomy. Similarly, whether the efferent pathway mediating the reflex involves sacral parasympathetic preganglionic neurons or only parasympathetic enteric nervous system neurons cannot be concluded. However, it is tempting to speculate that the reflex is mediated by the peripheral, enteric nervous system, because colorectal responses to CAP were the same in animals with aSCI (that produces spinal shock and reduces spinal parasympathetic activity) or intact spinal cord. Colorectal afferent terminals contain transmitters, such as substance P, neurokinin A, and calcitonin gene-related peptide. These peptides provide “afferent” terminals the ability to behave like “efferent” fibers, because the peptides are released when the afferent fibers fire an action potential (Mózsik et al., 2007; Spencer et al., 2020a,b). Because of the NKA content of colorectal terminals and the important role of NK2R agonists on colorectal function, we tested whether an NK2R antagonist, GR159897, would inhibit the colorectal response to CAP. Because no effect was detected, it is reasonable to exclude an efferent function of the CAP-sensitive colorectal terminals, at least in regard to NKA and NK2Rs.

The absence of effects of IR CAP on arterial pressure and heart rate in spinal intact, anesthetized rats suggest that IR CAP may be free of cardiovascular effects at efficacious doses. However, the elevation in arterial pressure in aSCI rats at doses only slightly higher than the efficacious doses suggests that cardiovascular effects should be monitored closely in people with spinal cord injuries rostral to T6 (Kurnick, 1956). These higher spinal levels of injury remove supraspinal, homeostatic control of vascular sympathetic preganglionic neurons in the thoracic cord, allowing latent connections from colorectal afferent fibers to elevate sympathetic activity. Spinal injuries often present symptoms of autonomic dysreflexia (hypertensive crisis) through the same mechanisms, and TRPV1 receptors have been implicated in producing these blood pressure responses (Ramer et al., 2012). Although this situation requires careful monitoring, it is possible that CAP-induced defecation in spinal-injured people, which lasts for only a few minutes, will produce smaller elevations in blood pressure than the manual bowel programs that are required for two-thirds of the people with spinal cord injuries. The manual bowel program, with digital stimulation of rectal fibers to induce defecation, which can take hours to sufficiently evacuate the colorectum, can be painful. In addition, manual bowel programs are inconvenient and stigmatizing for the individual, especially if a caregiver’s assistance is needed for digital stimulation (e.g., quadriplegia). For these reasons, one might expect that a mild, warm/burning sensation that accompanies IR CAP therapy will be preferred over a manual bowel program. Furthermore, many individuals with spinal cord injuries do not have colorectal sensation so a mild burning sensation may be irrelevant in those cases.

As expected, repeated application of CAP produced TRPV1 receptor desensitization, which was prevalent at 30 minutes and lasted up to 2 hours. The colorectal contraction response appeared to be fully recovered by the next day, when IR CAP (0.1%) was given to isoflurane anesthetized rats. However, this observation requires verification with a larger sample size and longer durations of treatment.

Limitations and Future Directions

One of the limitations of the current study is that female rats were not included. We focused on male rats because ∼80% of spinal cord injuries occurs in men (Shackelford et al., 1998). Future studies are required to determine the efficacy of IR CAP in females and whether there is any influence of the estrus cycle on colorectal function (Werner et al., 2023). Bowel dysfunction after spinal cord injury is correlated to the level and severity of the injury and although changes occur during the acute period, bowel function and management is generally thought to stabilize around 1 year post spinal cord injury, with reports of increased constipation 10 years post injury (Faaborg et al., 2008; Pavese et al., 2019; Round et al., 2021). In addition to the loss of autonomic inputs (parasympathetic and sympathetic nervous systems) mediating colorectal function following spinal cord injury, changes in the enteric nervous system occur, including plasticity in the number and arborization of interstitial cells of Cajal in the myenteric plexus, and an increase in smooth muscle thickness (Holmes and Blanke, 2019). Published reports of colorectal function in spinal cord injured rodents have shown that functional and anatomic changes in the colorectum change from the acute phase (1–3 days post injury) to the chronic phase (3–6 weeks post spinal cord injury) (White and Holmes, 2018; Werner et al., 2023). Therefore, studies examining the efficacy of IR CAP in chronic spinal cord injured rats have been initiated.

In summary, the current findings support the concept of using IR CAP therapy as a rapid-onset, short-duration, drug-induced defecation therapy. Future studies are needed to determine if the therapy can be used on a daily basis or must be reserved for ad hoc use with a frequency of usage to be determined.

Abbreviations

aSCI

acute spinal cord injury

AUC

area under the curve

CAP

capsaicin

CRP

colorectal pressure

IR

Intrarectal

LMN-NKA

[Lys5, MeLeu9, Nle10] NKA(4-10)]

MAP

mean arterial pressure

OLDA

N-oleoyldopamine

NK2R

neurokinin 2 receptor

OEA

oleoylethanolamide

TRPV1

transient receptor potential vanilloid 1

Authorship Contributions

Participated in research design: Cook, Thor, Marson.

Conducted experiments: Cook, Piatt, Marson.

Performed data analysis: Cook, Piatt, Marson.

Wrote or contributed to the writing of the manuscript: Cook, Piatt, Burgard, Thor, Marson.