ORIGINAL ARTICLE: GASTROENTEROLOGY
Effects of Amoxicillin and Clavulanic Acid on the Spontaneous Mechanical Activity of Juvenile Rat Duodenum Steven L. Ciciora, Kent C. Williams, and Cheryl E. Gariepy ABSTRACT Objectives: There are a limited number of medications for the treatment of foregut dysmotility. Enteral amoxicillin/clavulanic acid induces phase III duodenal contractions in a fasting pediatric patient. The mechanism by which this occurs is unknown. We examined the individual contributions of amoxicillin and clavulanic acid on the spontaneous mechanical activity of juvenile rat duodenum to better understand this phenomenon. Methods: Duodenal segments from juvenile rats were longitudinally attached to force transducers in organ baths. Samples were cumulatively exposed to amoxicillin or clavulanic acid. Separate samples were exposed to carbachol alone to assess response in both the presence and absence of amoxicillin or clavulanic acid. Basal tone, frequency, and amplitude of contractions were digitized and recorded. Results: The amplitude of the spontaneous contractions increased with amoxicillin. Inhibition of neuronal activity prevented this effect. Clavulanic acid did not affect the spontaneous contractions. Basal tone and the rate of contractions did not differ with either drug. Stimulation with carbachol in the presence of amoxicillin caused a statistically significant increase in the contractility compared with carbachol alone. Conclusions: Amoxicillin alters the spontaneous longitudinal mechanical activity of juvenile rat duodenum. Our results suggest that amoxicillin modulates the spontaneous pattern of cyclic mechanical activity of duodenal smooth muscle through noncholinergic, neurally mediated mechanisms. Our work provides an initial physiologic basis for the therapeutic use of amoxicillin in patients with gastrointestinal dysmotility. Key Words: amoxicillin, clavulanic acid, duodenum, motility, prokinetic drugs
(JPGN 2015;61: 340–345)
M
otility of the gastrointestinal (GI) tract involves the coordinated propagation of food and secretions to the distal GI tract. In dysmotility, the inherent, coordinated neuromuscular activity that directs peristalsis and movement of materials through the GI tract is abnormal. Dysmotility occurs because of a variety of reasons including inflammatory conditions, postinfectious state,
Received July 31, 2014; accepted March 25, 2015. From the Division of Pediatric Gastroenterology, Nationwide Children’s Hospital, Columbus, Ohio. Address correspondence and reprint requests to Steven L. Ciciora, MD, Division of Pediatric Gastroenterology, Nationwide Children’s Hospital, 700 Children’s Dr, Columbus, OH 43205-2696 (e-mail: steven. [email protected]). The authors report no conflicts of interest. Copyright # 2015 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition DOI: 10.1097/MPG.0000000000000804
What Is Known
Therapies for pediatric dysmotility are limited and oftentimes have adverse effects. Amoxicillin/clavulanic acid leads to increased intestinal motor activity in pediatric manometric studies. Amoxicillin/clavulanic acid is used clinically in patients with dysmotility.
What Is New
Amoxicillin alters activity in our system, whereas clavulanic acid does not. The action of amoxicillin appears to be neurally mediated. There is a need to study amoxicillin clinically as a promotility agent.
prior surgery, or genetic disorders, or it may be idiopathic. The most severe form is chronic intestinal pseudo-obstruction (CIPO), which is characterized by a great deal of morbidity. CIPO may be neuropathic, myopathic, or of a mixed type. The neuropathic pattern is more common than a myopathic one (1). Patients with GI dysmotility experience symptoms ranging from gastroesophageal reflux and bloating to massive abdominal distension with severe intestinal dysfunction (2,3). Medical treatments for GI dysmotility are limited. Therapies include the use of a limited number of prokinetic drugs, parenteral nutrition, along with placement of gastrostomy and enterostomies. Gastric pacemakers and intestinal transplantations are used in severe cases (4). The small number of medications used to improve motility in the stomach and small bowel include dopamine-receptor antagonists, serotonergic agents, and antibiotics such as erythromycin, a motilin agonist. Unfortunately, these drugs also have adverse effects, which limit their safe use (5–9). The oftentimes overwhelming symptoms of motility disorders along with the lack of effective therapies are problematic in adults and children alike with clear effects on quality of life (10). Past work with amoxicillin/clavulanic acid has shown promising results for its use as an enterokinetic agent. An adult manometric study revealed that administration of amoxicillin/clavulanic acid results in increased fasting motor activity in the small intestine (11). According to a study by Gomez et al (12), amoxicillin/ clavulanic acid induces phase III contractions in fasting pediatric patients. Its potential use as treatment for GI dysmotility is bolstered by the wide availability, low cost, and a history of safe use of amoxicillin/clavulanic acid in the pediatric population.
340 JPGN Volume 61, Number 3, September 2015 Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
JPGN
Volume 61, Number 3, September 2015
Amoxicillin is an oral b-lactam antibiotic that inhibits bacterial wall synthesis and is used to treat a wide range of infections. Clavulanic acid is a b-lactamase inhibitor that helps overcome antibiotic resistance seen in b-lactamase-producing bacteria. On its own, clavulanic acid has minimal antibiotic activity and is not commercially available for isolated use. The 2 are combined in a ratio of 4–7:1 and together provide a broader spectrum of antibacterial activity. The combination also has a role in the treatment of small bowel bacterial overgrowth (13). Presently, it is unclear whether either amoxicillin or clavulanic acid alone can change neuromuscular activity in the small intestine. Notably, rates of nausea, cramping, emesis, and diarrhea are higher with amoxicillin/clavulanic acid versus amoxicillin alone suggesting clavulanic acid may play a role in increasing motility (14,15). There is a risk of drug-induced hepatitis, however, with the use of amoxicillin/clavulanic acid that is greater than that of amoxicillin alone (9,16). Should amoxicillin on its own alter GI neuromuscular function in vitro, a case could be made to further study it as a prokinetic on its own. We set out to determine whether amoxicillin and clavulanic acid individually influence the spontaneous mechanical activity of juvenile rat duodenum in vitro and to characterize such influence too. With this knowledge, we will have a better initial mechanistic understanding of the use of amoxicillin and/or clavulanic acid in children with dysmotility.
Amoxicillin/Clavulanic Acid Duodenum Effect equilibration. Only strips with rhythmic spontaneous contractions were used in the analysis.
Myography
Wistar Kyoto rats, ages 4 to 6 weeks, were used. All animal procedures were approved by the Committee on Use and Care of Animals at Nationwide Children’s Research Institute (AR1100055). Rats were housed as specific pathogen free in temperatureand humidity-controlled environments, with free access to water and a standard rat chow with a 12:12-hour light:dark cycle.
Samples were cumulatively exposed to increasing doses of amoxicillin (0.1–100 mmol/L) or clavulanic acid (0.1–100 mmol/ L) at 15-minute intervals while the motor activity was recorded. The volume of drug added to an organ bath was never >1% the total volume to achieve a specific concentration. Control sample data were captured at the same 15-minute time points as experimental conditions to reflect changes that occur over time in the motor activity with the addition of vehicle (distilled water). The same trials were carried out on different tissue samples with serial additions of amoxicillin (3–100 mmol/L) in the presence of hexamethonium (100 mmol/L), atropine (1 mmol/L), or tetrodotoxin (1 mmol/L), which had been added to the organ bath system 30 to 60 minutes earlier. All samples were exposed to carbachol (10 mmol/L) to ensure tissue viability at the end of the trials with the exception of those previously exposed to atropine. We next aimed to determine whether the contractile response evoked by carbachol could be further augmented by amoxicillin (10–100 mmol/L), clavulanic acid (10–100 mmol/L), or the 2 in combination (amoxicillin 30–100 mmol/L and clavulanic acid 3– 10 mmol/L, a ratio similar to that which is commercially available). Duodenal segments were first exposed to carbachol (10 mmol/L) alone to determine cholinergic contractile response. After serial washouts with Krebs solution, tissues were incubated with the test agents for 30 minutes and then exposed again to carbachol. Contractile response was again recorded. Samples were finally exposed again to carbachol (10 mmol/L) to ensure tissue viability. During all of the trials, tension, frequency, and the amplitude of contractions were amplified, digitized, and recorded.
Drugs
Data and Statistical Analyses
Amoxicillin, potassium clavulanate, atropine (muscarinic antagonist), tetrodotoxin (sodium channel antagonist), carbachol (cholinergic agonist), and hexamethonium (nicotinic antagonist) were purchased from Sigma-Aldrich (St Louis, MO). Drugs were solubilized with distilled water into stock solutions. From the stock solutions, working drug solutions were made fresh daily in a range of concentrations such that the volume of solution added to an organ bath to reach a certain concentration was never >1% of the total volume.
The effect of each agent on mean tension, rate of spontaneous contractions, and amplitude of spontaneous contractions was calculated after addition of the agent from baseline. Contractile response was calculated as the lowest tension recorded just before the addition of carbachol subtracted from the peak tension recorded following the addition of carbachol. The effect of each agent was calculated as the contractile response with carbachol stimulation in the presence of the agent compared with contractile response with carbachol alone in which a result of 1 would represent an equivalent contractile response. Data were analyzed with 1-way analysis of variance with Bonferroni method and paired t tests using GraphPad Prism 5 (GraphPad Software, La Jolla, CA). Results were considered significant if P < 0.05.
METHODS Animals
Duodenum Isolation Rats were sacrificed in a carbon dioxide inhalation chamber. The proximal 10 to 15 mm of the duodenum was harvested from each rat, debrided of associated mesentery, and then gently flushed with Krebs solution (NaCl 119 mmol/L, KCl 4.5 mmol/L, MgSO4 2.5 mmol/L, NaHCO3 25 mmol/L, KH2PO4 1.2 mmol/L, CaCl2 2.5 mmol/L, and glucose 11.1 mmol/L) to cleanse the luminal contents. Each segment was in turn placed in ice-cold oxygenated (95% O2 and 5% CO2) Krebs solution. These whole-mount duodenal segments were individually placed in a 15-mL organ bath with oxygenated 378C Krebs solution and attached longitudinally to an isotonic force transducer. The activity was digitized and recorded in a laptop computer (PowerLab 16/35 and LabChart 7; ADInstruments, Colorado Springs, CO). Whole-mount strips were allowed to equilibrate for 30 to 60 minutes before the experiments, and an initial tension of approximately 1 gm was achieved. The organ bath solution was periodically replaced every 10 to 20 minutes during this time of www.jpgn.org
RESULTS Cumulative Exposure With Amoxicillin or Clavulanic Acid Isolated duodenal segments exhibited spontaneous longitudinal contractions at an average rate of 35 to 38 contractions per minute with an average amplitude of 300 to 500 mg tension (Fig. 1). After addition of amoxicillin in serially increasing doses, the average amplitude of spontaneous contractions increased by 26% (Figs. 1 and 2). The amplitude of spontaneous contractions did not change with the addition of clavulanic acid in serially increasing doses (Figs. 1 and 2). There was no significant change in the amplitude under control conditions over the same time course as
341
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
Ciciora et al
JPGN
Volume 61, Number 3, September 2015
1 gm
15 min 0.5 gm
100 µM
30 µM
10 µM
3 µM
1 µM
0.3 µM
Clavulanic acid
0.1 µM
Baseline
Amoxicillin
15 sec.
FIGURE 1. Representative tracings showing concentration-dependent effect of amoxicillin and clavulanic acid on spontaneous mechanical activity of rat duodenum. Concentrations shown represent addition of drug at each 15-minute interval. Inset shows detail of tracing activity.
Tetrodotoxin (1 mmol/L) induced a small increase in cyclic amplitude at baseline, which has been previously seen in other studies (17). Serial addition of amoxicillin in the presence of tetrodotoxin did not have an excitatory effect and paralleled control conditions (Fig. 3C).
experimental conditions (Fig. 2). Neither drug had an effect on tension or average rates of contraction (data not shown).
Cumulative Exposure With Amoxicillin in the Presence of Inhibitors
Contractile Response
To characterize the nature of amoxicillin’s effect on amplitude, we conducted further trials in the presence of neuronal inhibitors. The effects of amoxicillin on the amplitude persisted in the presence of hexamethonium (100 mmol/L), an acetylcholine nicotinic antagonist (Fig. 3A). Hexamethonium alone had no effect on the amplitude. Atropine (1 mmol/L), an acetylcholine muscarinic antagonist, significantly reduced the spontaneous amplitude of all of the segments on its own before the addition of amoxicillin (data not shown). When this baseline change was accounted for, the effects of amoxicillin on amplitude with increasing doses appeared to persist (Fig. 3B). This observation, however, did not reach statistical significance because of wide variation between samples.
To further investigate amoxicillin’s potential effect on the response of duodenal segments to cholinergic activity, response trials with carbachol were completed. The higher doses of amoxicillin tested (30–100 mmol/L) increased the contractile response (Fig. 4A). There was no effect on contractile response seen in the presence of clavulanic acid alone (Fig. 4B). Amoxicillin (100 mmol/L)/clavulanic acid (10 mmol/L) (mimicking a therapeutic ratio) increased the contractile response to carbachol (Fig. 4C). Serial addition of carbachol (10 mmol/L) did not have an additive effect on contraction in separate trials (data not shown).
Amplitude change (%)
40 * 20
*
*
* Amoxicillin Clavulanic acid
0
Vehicle −20
–7
–6
–5
Log [Drug] mol
–4
L–1
FIGURE 2. Concentration response curves showing effect of increasing dosing of amoxicillin and clavulanic acid on the mechanical activity. Amplitude change (%) versus baseline activity. Amplitudes increase at higher doses of amoxicillin (3–100 mmol/L) but do not change with serial additions of clavulanic acid (0.1–100 mmol/L). Control condition amplitudes did not change significantly over time. P < 0.05 versus baseline amplitude. Each point is mean standard error of the mean. n ¼ 13 duodenal segments for each drug.
342
www.jpgn.org
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
JPGN
Volume 61, Number 3, September 2015
Amoxicillin/Clavulanic Acid Duodenum Effect
Amplitude change (%)
A 30
* *
20 10
Amoxicillin Control (Hexamethonium)
0 −10
Amplitude change (%)
−20
50 40 30 20 10 0 −10 −20 −30
Amplitude change (%)
B
40
40
Amoxicillin & hexamethonium −6 −5 −4 Amoxicillin [log mol L−1]
Amoxicillin Control (Atropine) Amoxicillin & atropine −6 −5 −4 Amoxicillin [log mol L−1]
C
20
Amoxicillin Control (Tetrodotoxin)
0
Amoxicillin & tetrodotoxin
−20 −40
−6 −5 −4 Amoxicillin [log mol L−1]
FIGURE 3. Effect of increasing dosing of amoxicillin on the mechanical activity in the presence of hexamethonium, atropine, and tetrodotoxin. Amplitude change (%) versus baseline activity. A, Amplitudes increase at higher doses of amoxicillin (10–100 mmol/L) in the presence of hexamethonium (100 mmol/L). Hexamethonium (100 mmol/L) alone has no effect on mechanical activity over time. B, Data presented after correcting for decrease in amplitude caused by atropine across all samples. Amplitudes increase serially in the presence of atropine (1 mmol/L) but not to a statistically significant effect because of large variance between samples. C, Amplitudes do not increase in the presence of tetrodotoxin (1 mmol/L) and follow the same pattern as control conditions with tetrodotoxin alone over time. Amoxicillin data in (A–C) is the same in each figure. P < 0.05 versus baseline. Each point is meanstandard error of the mean. n ¼ 6 to 8 duodenal segments for each inhibitor used.
DISCUSSION This study shows for the first time that amoxicillin has a significant effect on the spontaneous mechanical activity of juvenile rat duodenum. We also show that inhibition of neuronal activity prevents this effect, but it persists in the presence of cholinergic antagonism. We were unable to detect any effect on spontaneous mechanical activity because of clavulanic acid. Neither drug had an effect on the frequency of contractions. We also show an amplification of cholinergic contractile response with amoxicillin. As our experiments did not demonstrate any effect because of clavulanic acid, our work demonstrates an impelling and exciting rationale for evaluating amoxicillin alone as a treatment for pediatric foregut motility disorders. Furthermore, given clavulanic acid’s lack of efficacy in our experiments and the associated risks and costs above those seen with amoxicillin alone, clavulanic acid’s use in combination with amoxicillin as an enterokinetic agent is called into question (9,14,15). In addition, the use of a narrower spectrum antibiotic to treat a noninfectious problem ought to lessen the concerns about antibiotic stewardship, inducing antibiotic resistance and increasing www.jpgn.org
the risk of Clostridium difficile colitis (but certainly not eliminate them) (18–21). It is notable that amoxicillin at a concentration of 30 mmol/L increased contractile activity (Fig. 4A), whereas that same concentration of amoxicillin in combination with clavulanic acid did not have an effect on contractile activity (Fig. 4C). This may be because of an inhibitory effect of clavulanic acid, or it reflects the small number of trials we did in this portion of our work. The concentrations of amoxicillin in the organ baths in which a statistically significant effect was seen in our experiments can be reached in the serum with oral dosing of amoxicillin giving further physiologic credence to our findings (22). The prokinetic effect seen by Gomez et al (12), however, is notable, wherein amoxicillin/ clavulanic acid–induced phase III contractions in fasting pediatric patients was seen minutes after small intestine infusion of the drug via an antroduodenal motility catheter. This observation of immediate efficacy suggests a direct, luminal mechanism of action for amoxicillin rather than one because of systemic absorption and remote action. It was for this reason that we chose to keep the mucosa intact in our experimental system rather than conducting these trials with segments without mucosa.
343
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
Ciciora et al
JPGN
A Drug + Carbachol Carbachol alone
1.3 *
1.2
*
1.1 1.0 0.9
−5
−4.5
−4
Amoxicillin [log mol L−1]
Drug + Carbachol Carbachol alone
B 1.5
1.0 −5 0.5
−4.5
−4
Clavulanic acid [log mol L−1]
Drug + Carbachol Carbachol alone
C 1.3
*
1.0
0.7 Amoxicillin −5 Clav. acid −5.5
Amoxicillin −4.5 Clav. acid −5.5 [log mol L−1]
Amoxicillin −4 Clav. acid −5
FIGURE 4. Effect of drug on the maximal contractility of duodenal segments when stimulated with carbachol as compared with maximal contractility with carbachol alone. Horizontal line at 1.0 represents activity with carbachol (10 mmol/L) alone. A, Amoxicillin (30–100 mmol/L) increases the maximal contractility compared with carbachol alone. B, Clavulanic acid (10–100 mmol/L) has no effect on the maximal contractility. C, The highest dose tested (amoxicillin 100 mmol/L and clavulanic acid 1 mmol/L) increases the maximal contractility compared with carbachol (10 mmol/L) alone. Horizontal line in the box is mean, whiskers are the range of results. Contractile response is calculated as the lowest tension recorded just before the addition of carbachol subtracted from the peak tension recorded following the addition of carbachol. P < 0.05 versus carbachol alone. n ¼ 6 duodenal segments for each trial at each concentration.
Interestingly, our findings suggesting a promotility effect of amoxicillin in the duodenum contradict a prior clinical study where amoxicillin was found to have little promotility effect on its own. Bortolotti et al (23) performed gastroduodenal recording in adults with functional dyspepsia and Helicobacter pylori gastritis. They found that clarithromycin induced manometric activity, whereas amoxicillin did not and suggested clarithromycin was responsible for the upper GI symptoms oftentimes experienced by those taking oral H pylori eradication regimens. This discrepancy may be because of the age of the subjects. Specifically, Gomez et al (12) studied children, and we purposefully studied young, not fully grown, rats because it has previously been shown that neurotransmitter effects can be age dependent in the GI tract (24).
344
Volume 61, Number 3, September 2015
In our trials, given the amplitude increases with serial addition of amoxicillin in the presence of an acetylcholine nicotinic antagonist (hexamethonium), a noncholinergic mechanism of action is involved (Fig. 3A). Furthermore, the finding that amoxicillin increases maximum contractility beyond that which is achieved with carbachol stimulation alone also supports this conclusion (Fig. 4). In addition, the effects of amoxicillin on amplitude with increasing doses appeared to persist in the presence of atropine (muscarinic antagonist) but did not reach significance because of wide variation. Given the observation that the atropine control curve continues to decay at our last recorded observation point, it may be that atropine does not have its full effect during the time periods we observed and may thus serve as an explanation for the variability which precluded this observation from reaching statistical significance (Fig. 3B). The absence of an effect with amoxicillin in the presence of tetrodotoxin suggests a neurally mediated process. Prior studies suggest that nervous elements are responsible for the variability of mechanical activity of the rat duodenum, whereas the inherent mechanical activity itself is myogenic (17,25). As the segments we used in our trials, however, were whole-mount duodenal segments and not isolated muscle strips, the specific influence each muscle layer (longitudinal and circular) has on the mechanical activity relative to amoxicillin cannot be established. Furthermore, as our segments had intact mucosa, the influence of mucosal enterochromaffin cells remained intact whereas other studies often remove the mucosa to solely assess the role of the myenteric plexuses. Therefore, our initial findings that demonstrate amoxicillin’s neurally mediated effect represents an area that could be further studied with isolated muscle strips to allow for further clarification of the specific mechanism. Given that the mucosal layer was kept intact, a tachykinergic or serotonergic mechanism may be contributing as well. Tachykinins (substance P, neurokinin A, and neurokinin B) have similar distribution in the human and rodent intestine and can both stimulate and inhibit intestinal tract motor activity depending on the site and the levels of expression. Some GI disorders associated with dysmotility feature reduced tachykinin sensitivity, which, when antagonized, gives resolution of these motor disturbances (26–29). Serotonin also plays a major role in intestinal tract motor and secretory function and has been implicated in disease states (30,31). Its varied role in GI tract physiology (including its role in the migrating motor complex) is mediated by a number receptor subtypes (32,33). Furthermore, experiments in the presence of tachykinergic or serotonergic inhibitors could be performed to further delineate the mechanism of action of amoxicillin too. Although our results are encouraging, we recognize our work has limitations. Specifically, the relationship between the motility of rat duodenum and human duodenum is unknown. This limits the immediate clinical applicability of our findings. It has been shown that some drugs may have different effects at different sites of the GI tract in animal models compared with their site of action in humans (34). In addition, the methods in this study are unable to measure the complex motor patterns of peristalsis in humans as is done in antroduodenal manometric studies (35). Coupling our findings with the clinical data of Gomez et al, however, there is a clear indication for a clinical study examining the use of amoxicillin alone to improve duodenal motility in children. Given the potential benefit in light of the low cost and wide availability, amoxicillin as an enterokinetic deserves further investigation. The dearth of therapies available to those with motility disorders makes these findings all the more important. In conclusion, our results are the first to demonstrate that amoxicillin modulates the spontaneous pattern of mechanical activity of rat duodenum through noncholinergic, neurally mediated www.jpgn.org
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
JPGN
Volume 61, Number 3, September 2015
mechanisms. Clavulanic acid does not impact the pattern of activity in our system. Our novel findings suggest a distinct role for the use of amoxicillin in the treatment of pediatric dysmotility. This work begins to provide a better understanding of amoxicillin’s function for those patients whose present pharmacologic therapeutic options are extremely limited. Acknowledgments: The authors thank Naoko Murakami for animal husbandry, Drs Jackie Wood, Yusen Liu, and Fievos Christofi for technical advice, and Drs Pam Lucchesi and Aaron Trask for the use of equipment.
REFERENCES 1. Connor FL, Di Lorenzo C. Chronic intestinal pseudo-obstruction: assessment and management. Gastroenterology 2006;130 (2 suppl 1): S29–36. 2. Milla PJ. Motility disorders in childhood. Baillieres Clin Gastroenterol 1998;12:775–97. 3. Yeung AK, Di Lorenzo C. Primary gastrointestinal motility disorders in childhood. Minerva Pediatr 2012;64:567–84. 4. Gariepy CE, Mousa H. Clinical management of motility disorders in children. Semin Pediatr Surg 2009;18:224–38. 5. Quigley EM. Cisapride: what can we learn from the rise and fall of a prokinetic? J Dig Dis 2011;12:147–56. 6. Di Lorenzo C, Youssef NN. Diagnosis and management of intestinal motility disorders. Semin Pediatr Surg 2010;19:50–8. 7. Langer JC, Bramlett G. Effect of prokinetic agents on ileal contractility in a rabbit model of gastroschisis. J Pediatr Surg 1997;32:605–8. 8. Dicken BJ, Sergi C, Rescorla FJ, et al. Medical management of motility disorders in patients with intestinal failure: a focus on necrotizing enterocolitis, gastroschisis, and intestinal atresia. J Pediatr Surg 2011;46:1618–30. 9. Hita EO, Ruiz-Extremera A, Garcia JA, et al. Amoxicillin-clavulanic acid hepatotoxicity in children. J Pediatr Gastroenterol Nutr 2012; 55:663–7. 10. Schwankovsky L, Mousa H, Rowhani A, et al. Quality of life outcomes in congenital chronic intestinal pseudo-obstruction. Dig Dis Sci 2002;47:1965–8. 11. Caron F, Ducrotte P, Lerebours E, et al. Effects of amoxicillin-clavulanate combination on the motility of the small intestine in human beings. Antimicrob Agents Chemother 1991;35:1085–8. 12. Gomez R, Fernandez S, Aspirot A, et al. Effect of amoxicillin/clavulanate on gastrointestinal motility in children. J Pediatr Gastroenterol Nutr 2012;54:780–4. 13. Lykova EA, Bondarenko VM, Parfenov AI, et al. [Bacterial overgrowth syndrome in the small intestine: pathogenesis, clinical significance and therapy tactics]. Eksp Klin Gastroenterol 200551–7, 113.. 14. Salvo F, De Sarro A, Caputi AP, et al. Amoxicillin and amoxicillin plus clavulanate: a safety review. Expert Opin Drug Saf 2009;8:111–8. 15. Turck D, Bernet JP, Marx J, et al. Incidence and risk factors of oral antibiotic-associated diarrhea in an outpatient pediatric population. J Pediatr Gastroenterol Nutr 2003;37:22–6. 16. Robles M, Andrade RJ. [Hepatotoxicity by antibiotics: update in 2008]. Revista espanola de quimioterapia: publicacion oficial de la Sociedad Espanola de Quimioterapia. Rev Esp Quimioter 2008;21:224–33.
www.jpgn.org
Amoxicillin/Clavulanic Acid Duodenum Effect
17. Postorino A, Mancinelli R, Racanicchi C, et al. Spontaneous electromechanical activity in the rat duodenum in vitro. Arch Int Physiol Biochim 1990;98:35–40. 18. Rahnama’i MS, Wagenvoort JH, van der Linden CJ, et al. (Augmentin) resistant Escherichia coli in bacterial peritonitis after abdominal surgery—clinical outcome in ICU patients. Neth J Med 2009;67: 173–6. 19. Savini V, Catavitello C, Balbinot A, et al. Discrepancy between Vitek 2 and the agar disk method in detecting the amoxicillin-clavulanate resistance in a Klebsiella pneumoniae isolate. Neth J Med 2010; 18:53–4. 20. Ferjani A, Marzouk M, Ben Moussa F, et al. [Resistance of Escherichia coli isolated from the urinary tract to amoxicillin-clavulanic acid and other antibiotics]. Med Mal Infect 2010;40:161–4. 21. McGregor JA. Ticarcillin/clavulanate for the treatment of female genital tract infections. Efficacy, safety and comparative microbiology. J Reprod Med 1990;35 (3 suppl):333–8. 22. Barr WH, Zola EM, Candler EL, et al. Differential absorption of amoxicillin from the human small and large intestine. Clin Pharmacol Ther 1994;56:279–85. 23. Bortolotti M, Brunelli F, Sarti P, et al. Effects of oral clarithromycin and amoxycillin on interdigestive gastrointestinal motility of patients with functional dyspepsia and Helicobacter pylori gastritis. Aliment Pharmacol Ther 1998;12:1021–5. 24. Wittmeyer V, Merrot T, Mazet B. Tonic inhibition of human small intestinal motility by nitric oxide in children but not in adults. Neurogastroenterol Motil 2010;22:1078–282. 25. Wood JD. Excitation of intestinal muscle by atropine, tetrodotoxin, and xylocaine. Am J Physiol 1972;222:118–25. 26. Holzer P, Holzer-Petsche U. Tachykinin receptors in the gut: physiological and pathological implications. Curr Opin Pharmacol 2001;1: 583–90. 27. Hellstrom PM, Murthy KS, Grider JR, et al. Coexistence of three tachykinin receptors coupled to Caþþ signaling pathways in intestinal muscle cells. J Pharmacol Exp Ther 1994;270:236–43. 28. Improta G, Broccardo M. Tachykinins: role in human gastrointestinal tract physiology and pathology. Curr Drug Targets 2006;7:1021–9. 29. Valero MS, Fagundes DS, Grasa L, et al. Contractile effect of tachykinins on rabbit small intestine. Acta Pharmacol Sin 2011;32:487–94. 30. Deloose E, Janssen P, Depoortere I, et al. The migrating motor complex: control mechanisms and its role in health and disease. Nat Rev Gastroenterol Hepatol 2012;9:271–85. 31. Bottner M, Barrenschee M, Hellwig I, et al. The enteric serotonergic system is altered in patients with diverticular disease. Gut 2013;62: 1753–62. 32. Hasler WL. Serotonin and the GI tract. Curr Gastroenterol Rep 2009;11:383–91. 33. Cirillo C, Vanden Berghe P, Tack J. Role of serotonin in gastrointestinal physiology and pathology. Minerva Endocrinol 2011;36:311–24. 34. Cinci L, Faussone-Pellegrini MS, Rotondo A, et al. GLP-2 receptor expression in excitatory and inhibitory enteric neurons and its role in mouse duodenum contractility. Neurogastroenterol Motil 2011;23:e383– 92. 35. Lentle RG, De Loubens C, Hulls C, et al. A comparison of the organization of longitudinal and circular contractions during pendular and segmental activity in the duodenum of the rat and guinea pig. Neurogastroenterol Motil 2012;24:686–95.
345
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
Effects of Amoxicillin and Clavulanic Acid on the Spontaneous Mechanical Activity of Juvenile Rat Duodenum Steven L. Ciciora, Kent C. Williams, and Cheryl E. Gariepy ABSTRACT Objectives: There are a limited number of medications for the treatment of foregut dysmotility. Enteral amoxicillin/clavulanic acid induces phase III duodenal contractions in a fasting pediatric patient. The mechanism by which this occurs is unknown. We examined the individual contributions of amoxicillin and clavulanic acid on the spontaneous mechanical activity of juvenile rat duodenum to better understand this phenomenon. Methods: Duodenal segments from juvenile rats were longitudinally attached to force transducers in organ baths. Samples were cumulatively exposed to amoxicillin or clavulanic acid. Separate samples were exposed to carbachol alone to assess response in both the presence and absence of amoxicillin or clavulanic acid. Basal tone, frequency, and amplitude of contractions were digitized and recorded. Results: The amplitude of the spontaneous contractions increased with amoxicillin. Inhibition of neuronal activity prevented this effect. Clavulanic acid did not affect the spontaneous contractions. Basal tone and the rate of contractions did not differ with either drug. Stimulation with carbachol in the presence of amoxicillin caused a statistically significant increase in the contractility compared with carbachol alone. Conclusions: Amoxicillin alters the spontaneous longitudinal mechanical activity of juvenile rat duodenum. Our results suggest that amoxicillin modulates the spontaneous pattern of cyclic mechanical activity of duodenal smooth muscle through noncholinergic, neurally mediated mechanisms. Our work provides an initial physiologic basis for the therapeutic use of amoxicillin in patients with gastrointestinal dysmotility. Key Words: amoxicillin, clavulanic acid, duodenum, motility, prokinetic drugs
(JPGN 2015;61: 340–345)
M
otility of the gastrointestinal (GI) tract involves the coordinated propagation of food and secretions to the distal GI tract. In dysmotility, the inherent, coordinated neuromuscular activity that directs peristalsis and movement of materials through the GI tract is abnormal. Dysmotility occurs because of a variety of reasons including inflammatory conditions, postinfectious state,
Received July 31, 2014; accepted March 25, 2015. From the Division of Pediatric Gastroenterology, Nationwide Children’s Hospital, Columbus, Ohio. Address correspondence and reprint requests to Steven L. Ciciora, MD, Division of Pediatric Gastroenterology, Nationwide Children’s Hospital, 700 Children’s Dr, Columbus, OH 43205-2696 (e-mail: steven. [email protected]). The authors report no conflicts of interest. Copyright # 2015 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition DOI: 10.1097/MPG.0000000000000804
What Is Known
Therapies for pediatric dysmotility are limited and oftentimes have adverse effects. Amoxicillin/clavulanic acid leads to increased intestinal motor activity in pediatric manometric studies. Amoxicillin/clavulanic acid is used clinically in patients with dysmotility.
What Is New
Amoxicillin alters activity in our system, whereas clavulanic acid does not. The action of amoxicillin appears to be neurally mediated. There is a need to study amoxicillin clinically as a promotility agent.
prior surgery, or genetic disorders, or it may be idiopathic. The most severe form is chronic intestinal pseudo-obstruction (CIPO), which is characterized by a great deal of morbidity. CIPO may be neuropathic, myopathic, or of a mixed type. The neuropathic pattern is more common than a myopathic one (1). Patients with GI dysmotility experience symptoms ranging from gastroesophageal reflux and bloating to massive abdominal distension with severe intestinal dysfunction (2,3). Medical treatments for GI dysmotility are limited. Therapies include the use of a limited number of prokinetic drugs, parenteral nutrition, along with placement of gastrostomy and enterostomies. Gastric pacemakers and intestinal transplantations are used in severe cases (4). The small number of medications used to improve motility in the stomach and small bowel include dopamine-receptor antagonists, serotonergic agents, and antibiotics such as erythromycin, a motilin agonist. Unfortunately, these drugs also have adverse effects, which limit their safe use (5–9). The oftentimes overwhelming symptoms of motility disorders along with the lack of effective therapies are problematic in adults and children alike with clear effects on quality of life (10). Past work with amoxicillin/clavulanic acid has shown promising results for its use as an enterokinetic agent. An adult manometric study revealed that administration of amoxicillin/clavulanic acid results in increased fasting motor activity in the small intestine (11). According to a study by Gomez et al (12), amoxicillin/ clavulanic acid induces phase III contractions in fasting pediatric patients. Its potential use as treatment for GI dysmotility is bolstered by the wide availability, low cost, and a history of safe use of amoxicillin/clavulanic acid in the pediatric population.
340 JPGN Volume 61, Number 3, September 2015 Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
JPGN
Volume 61, Number 3, September 2015
Amoxicillin is an oral b-lactam antibiotic that inhibits bacterial wall synthesis and is used to treat a wide range of infections. Clavulanic acid is a b-lactamase inhibitor that helps overcome antibiotic resistance seen in b-lactamase-producing bacteria. On its own, clavulanic acid has minimal antibiotic activity and is not commercially available for isolated use. The 2 are combined in a ratio of 4–7:1 and together provide a broader spectrum of antibacterial activity. The combination also has a role in the treatment of small bowel bacterial overgrowth (13). Presently, it is unclear whether either amoxicillin or clavulanic acid alone can change neuromuscular activity in the small intestine. Notably, rates of nausea, cramping, emesis, and diarrhea are higher with amoxicillin/clavulanic acid versus amoxicillin alone suggesting clavulanic acid may play a role in increasing motility (14,15). There is a risk of drug-induced hepatitis, however, with the use of amoxicillin/clavulanic acid that is greater than that of amoxicillin alone (9,16). Should amoxicillin on its own alter GI neuromuscular function in vitro, a case could be made to further study it as a prokinetic on its own. We set out to determine whether amoxicillin and clavulanic acid individually influence the spontaneous mechanical activity of juvenile rat duodenum in vitro and to characterize such influence too. With this knowledge, we will have a better initial mechanistic understanding of the use of amoxicillin and/or clavulanic acid in children with dysmotility.
Amoxicillin/Clavulanic Acid Duodenum Effect equilibration. Only strips with rhythmic spontaneous contractions were used in the analysis.
Myography
Wistar Kyoto rats, ages 4 to 6 weeks, were used. All animal procedures were approved by the Committee on Use and Care of Animals at Nationwide Children’s Research Institute (AR1100055). Rats were housed as specific pathogen free in temperatureand humidity-controlled environments, with free access to water and a standard rat chow with a 12:12-hour light:dark cycle.
Samples were cumulatively exposed to increasing doses of amoxicillin (0.1–100 mmol/L) or clavulanic acid (0.1–100 mmol/ L) at 15-minute intervals while the motor activity was recorded. The volume of drug added to an organ bath was never >1% the total volume to achieve a specific concentration. Control sample data were captured at the same 15-minute time points as experimental conditions to reflect changes that occur over time in the motor activity with the addition of vehicle (distilled water). The same trials were carried out on different tissue samples with serial additions of amoxicillin (3–100 mmol/L) in the presence of hexamethonium (100 mmol/L), atropine (1 mmol/L), or tetrodotoxin (1 mmol/L), which had been added to the organ bath system 30 to 60 minutes earlier. All samples were exposed to carbachol (10 mmol/L) to ensure tissue viability at the end of the trials with the exception of those previously exposed to atropine. We next aimed to determine whether the contractile response evoked by carbachol could be further augmented by amoxicillin (10–100 mmol/L), clavulanic acid (10–100 mmol/L), or the 2 in combination (amoxicillin 30–100 mmol/L and clavulanic acid 3– 10 mmol/L, a ratio similar to that which is commercially available). Duodenal segments were first exposed to carbachol (10 mmol/L) alone to determine cholinergic contractile response. After serial washouts with Krebs solution, tissues were incubated with the test agents for 30 minutes and then exposed again to carbachol. Contractile response was again recorded. Samples were finally exposed again to carbachol (10 mmol/L) to ensure tissue viability. During all of the trials, tension, frequency, and the amplitude of contractions were amplified, digitized, and recorded.
Drugs
Data and Statistical Analyses
Amoxicillin, potassium clavulanate, atropine (muscarinic antagonist), tetrodotoxin (sodium channel antagonist), carbachol (cholinergic agonist), and hexamethonium (nicotinic antagonist) were purchased from Sigma-Aldrich (St Louis, MO). Drugs were solubilized with distilled water into stock solutions. From the stock solutions, working drug solutions were made fresh daily in a range of concentrations such that the volume of solution added to an organ bath to reach a certain concentration was never >1% of the total volume.
The effect of each agent on mean tension, rate of spontaneous contractions, and amplitude of spontaneous contractions was calculated after addition of the agent from baseline. Contractile response was calculated as the lowest tension recorded just before the addition of carbachol subtracted from the peak tension recorded following the addition of carbachol. The effect of each agent was calculated as the contractile response with carbachol stimulation in the presence of the agent compared with contractile response with carbachol alone in which a result of 1 would represent an equivalent contractile response. Data were analyzed with 1-way analysis of variance with Bonferroni method and paired t tests using GraphPad Prism 5 (GraphPad Software, La Jolla, CA). Results were considered significant if P < 0.05.
METHODS Animals
Duodenum Isolation Rats were sacrificed in a carbon dioxide inhalation chamber. The proximal 10 to 15 mm of the duodenum was harvested from each rat, debrided of associated mesentery, and then gently flushed with Krebs solution (NaCl 119 mmol/L, KCl 4.5 mmol/L, MgSO4 2.5 mmol/L, NaHCO3 25 mmol/L, KH2PO4 1.2 mmol/L, CaCl2 2.5 mmol/L, and glucose 11.1 mmol/L) to cleanse the luminal contents. Each segment was in turn placed in ice-cold oxygenated (95% O2 and 5% CO2) Krebs solution. These whole-mount duodenal segments were individually placed in a 15-mL organ bath with oxygenated 378C Krebs solution and attached longitudinally to an isotonic force transducer. The activity was digitized and recorded in a laptop computer (PowerLab 16/35 and LabChart 7; ADInstruments, Colorado Springs, CO). Whole-mount strips were allowed to equilibrate for 30 to 60 minutes before the experiments, and an initial tension of approximately 1 gm was achieved. The organ bath solution was periodically replaced every 10 to 20 minutes during this time of www.jpgn.org
RESULTS Cumulative Exposure With Amoxicillin or Clavulanic Acid Isolated duodenal segments exhibited spontaneous longitudinal contractions at an average rate of 35 to 38 contractions per minute with an average amplitude of 300 to 500 mg tension (Fig. 1). After addition of amoxicillin in serially increasing doses, the average amplitude of spontaneous contractions increased by 26% (Figs. 1 and 2). The amplitude of spontaneous contractions did not change with the addition of clavulanic acid in serially increasing doses (Figs. 1 and 2). There was no significant change in the amplitude under control conditions over the same time course as
341
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
Ciciora et al
JPGN
Volume 61, Number 3, September 2015
1 gm
15 min 0.5 gm
100 µM
30 µM
10 µM
3 µM
1 µM
0.3 µM
Clavulanic acid
0.1 µM
Baseline
Amoxicillin
15 sec.
FIGURE 1. Representative tracings showing concentration-dependent effect of amoxicillin and clavulanic acid on spontaneous mechanical activity of rat duodenum. Concentrations shown represent addition of drug at each 15-minute interval. Inset shows detail of tracing activity.
Tetrodotoxin (1 mmol/L) induced a small increase in cyclic amplitude at baseline, which has been previously seen in other studies (17). Serial addition of amoxicillin in the presence of tetrodotoxin did not have an excitatory effect and paralleled control conditions (Fig. 3C).
experimental conditions (Fig. 2). Neither drug had an effect on tension or average rates of contraction (data not shown).
Cumulative Exposure With Amoxicillin in the Presence of Inhibitors
Contractile Response
To characterize the nature of amoxicillin’s effect on amplitude, we conducted further trials in the presence of neuronal inhibitors. The effects of amoxicillin on the amplitude persisted in the presence of hexamethonium (100 mmol/L), an acetylcholine nicotinic antagonist (Fig. 3A). Hexamethonium alone had no effect on the amplitude. Atropine (1 mmol/L), an acetylcholine muscarinic antagonist, significantly reduced the spontaneous amplitude of all of the segments on its own before the addition of amoxicillin (data not shown). When this baseline change was accounted for, the effects of amoxicillin on amplitude with increasing doses appeared to persist (Fig. 3B). This observation, however, did not reach statistical significance because of wide variation between samples.
To further investigate amoxicillin’s potential effect on the response of duodenal segments to cholinergic activity, response trials with carbachol were completed. The higher doses of amoxicillin tested (30–100 mmol/L) increased the contractile response (Fig. 4A). There was no effect on contractile response seen in the presence of clavulanic acid alone (Fig. 4B). Amoxicillin (100 mmol/L)/clavulanic acid (10 mmol/L) (mimicking a therapeutic ratio) increased the contractile response to carbachol (Fig. 4C). Serial addition of carbachol (10 mmol/L) did not have an additive effect on contraction in separate trials (data not shown).
Amplitude change (%)
40 * 20
*
*
* Amoxicillin Clavulanic acid
0
Vehicle −20
–7
–6
–5
Log [Drug] mol
–4
L–1
FIGURE 2. Concentration response curves showing effect of increasing dosing of amoxicillin and clavulanic acid on the mechanical activity. Amplitude change (%) versus baseline activity. Amplitudes increase at higher doses of amoxicillin (3–100 mmol/L) but do not change with serial additions of clavulanic acid (0.1–100 mmol/L). Control condition amplitudes did not change significantly over time. P < 0.05 versus baseline amplitude. Each point is mean standard error of the mean. n ¼ 13 duodenal segments for each drug.
342
www.jpgn.org
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
JPGN
Volume 61, Number 3, September 2015
Amoxicillin/Clavulanic Acid Duodenum Effect
Amplitude change (%)
A 30
* *
20 10
Amoxicillin Control (Hexamethonium)
0 −10
Amplitude change (%)
−20
50 40 30 20 10 0 −10 −20 −30
Amplitude change (%)
B
40
40
Amoxicillin & hexamethonium −6 −5 −4 Amoxicillin [log mol L−1]
Amoxicillin Control (Atropine) Amoxicillin & atropine −6 −5 −4 Amoxicillin [log mol L−1]
C
20
Amoxicillin Control (Tetrodotoxin)
0
Amoxicillin & tetrodotoxin
−20 −40
−6 −5 −4 Amoxicillin [log mol L−1]
FIGURE 3. Effect of increasing dosing of amoxicillin on the mechanical activity in the presence of hexamethonium, atropine, and tetrodotoxin. Amplitude change (%) versus baseline activity. A, Amplitudes increase at higher doses of amoxicillin (10–100 mmol/L) in the presence of hexamethonium (100 mmol/L). Hexamethonium (100 mmol/L) alone has no effect on mechanical activity over time. B, Data presented after correcting for decrease in amplitude caused by atropine across all samples. Amplitudes increase serially in the presence of atropine (1 mmol/L) but not to a statistically significant effect because of large variance between samples. C, Amplitudes do not increase in the presence of tetrodotoxin (1 mmol/L) and follow the same pattern as control conditions with tetrodotoxin alone over time. Amoxicillin data in (A–C) is the same in each figure. P < 0.05 versus baseline. Each point is meanstandard error of the mean. n ¼ 6 to 8 duodenal segments for each inhibitor used.
DISCUSSION This study shows for the first time that amoxicillin has a significant effect on the spontaneous mechanical activity of juvenile rat duodenum. We also show that inhibition of neuronal activity prevents this effect, but it persists in the presence of cholinergic antagonism. We were unable to detect any effect on spontaneous mechanical activity because of clavulanic acid. Neither drug had an effect on the frequency of contractions. We also show an amplification of cholinergic contractile response with amoxicillin. As our experiments did not demonstrate any effect because of clavulanic acid, our work demonstrates an impelling and exciting rationale for evaluating amoxicillin alone as a treatment for pediatric foregut motility disorders. Furthermore, given clavulanic acid’s lack of efficacy in our experiments and the associated risks and costs above those seen with amoxicillin alone, clavulanic acid’s use in combination with amoxicillin as an enterokinetic agent is called into question (9,14,15). In addition, the use of a narrower spectrum antibiotic to treat a noninfectious problem ought to lessen the concerns about antibiotic stewardship, inducing antibiotic resistance and increasing www.jpgn.org
the risk of Clostridium difficile colitis (but certainly not eliminate them) (18–21). It is notable that amoxicillin at a concentration of 30 mmol/L increased contractile activity (Fig. 4A), whereas that same concentration of amoxicillin in combination with clavulanic acid did not have an effect on contractile activity (Fig. 4C). This may be because of an inhibitory effect of clavulanic acid, or it reflects the small number of trials we did in this portion of our work. The concentrations of amoxicillin in the organ baths in which a statistically significant effect was seen in our experiments can be reached in the serum with oral dosing of amoxicillin giving further physiologic credence to our findings (22). The prokinetic effect seen by Gomez et al (12), however, is notable, wherein amoxicillin/ clavulanic acid–induced phase III contractions in fasting pediatric patients was seen minutes after small intestine infusion of the drug via an antroduodenal motility catheter. This observation of immediate efficacy suggests a direct, luminal mechanism of action for amoxicillin rather than one because of systemic absorption and remote action. It was for this reason that we chose to keep the mucosa intact in our experimental system rather than conducting these trials with segments without mucosa.
343
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
Ciciora et al
JPGN
A Drug + Carbachol Carbachol alone
1.3 *
1.2
*
1.1 1.0 0.9
−5
−4.5
−4
Amoxicillin [log mol L−1]
Drug + Carbachol Carbachol alone
B 1.5
1.0 −5 0.5
−4.5
−4
Clavulanic acid [log mol L−1]
Drug + Carbachol Carbachol alone
C 1.3
*
1.0
0.7 Amoxicillin −5 Clav. acid −5.5
Amoxicillin −4.5 Clav. acid −5.5 [log mol L−1]
Amoxicillin −4 Clav. acid −5
FIGURE 4. Effect of drug on the maximal contractility of duodenal segments when stimulated with carbachol as compared with maximal contractility with carbachol alone. Horizontal line at 1.0 represents activity with carbachol (10 mmol/L) alone. A, Amoxicillin (30–100 mmol/L) increases the maximal contractility compared with carbachol alone. B, Clavulanic acid (10–100 mmol/L) has no effect on the maximal contractility. C, The highest dose tested (amoxicillin 100 mmol/L and clavulanic acid 1 mmol/L) increases the maximal contractility compared with carbachol (10 mmol/L) alone. Horizontal line in the box is mean, whiskers are the range of results. Contractile response is calculated as the lowest tension recorded just before the addition of carbachol subtracted from the peak tension recorded following the addition of carbachol. P < 0.05 versus carbachol alone. n ¼ 6 duodenal segments for each trial at each concentration.
Interestingly, our findings suggesting a promotility effect of amoxicillin in the duodenum contradict a prior clinical study where amoxicillin was found to have little promotility effect on its own. Bortolotti et al (23) performed gastroduodenal recording in adults with functional dyspepsia and Helicobacter pylori gastritis. They found that clarithromycin induced manometric activity, whereas amoxicillin did not and suggested clarithromycin was responsible for the upper GI symptoms oftentimes experienced by those taking oral H pylori eradication regimens. This discrepancy may be because of the age of the subjects. Specifically, Gomez et al (12) studied children, and we purposefully studied young, not fully grown, rats because it has previously been shown that neurotransmitter effects can be age dependent in the GI tract (24).
344
Volume 61, Number 3, September 2015
In our trials, given the amplitude increases with serial addition of amoxicillin in the presence of an acetylcholine nicotinic antagonist (hexamethonium), a noncholinergic mechanism of action is involved (Fig. 3A). Furthermore, the finding that amoxicillin increases maximum contractility beyond that which is achieved with carbachol stimulation alone also supports this conclusion (Fig. 4). In addition, the effects of amoxicillin on amplitude with increasing doses appeared to persist in the presence of atropine (muscarinic antagonist) but did not reach significance because of wide variation. Given the observation that the atropine control curve continues to decay at our last recorded observation point, it may be that atropine does not have its full effect during the time periods we observed and may thus serve as an explanation for the variability which precluded this observation from reaching statistical significance (Fig. 3B). The absence of an effect with amoxicillin in the presence of tetrodotoxin suggests a neurally mediated process. Prior studies suggest that nervous elements are responsible for the variability of mechanical activity of the rat duodenum, whereas the inherent mechanical activity itself is myogenic (17,25). As the segments we used in our trials, however, were whole-mount duodenal segments and not isolated muscle strips, the specific influence each muscle layer (longitudinal and circular) has on the mechanical activity relative to amoxicillin cannot be established. Furthermore, as our segments had intact mucosa, the influence of mucosal enterochromaffin cells remained intact whereas other studies often remove the mucosa to solely assess the role of the myenteric plexuses. Therefore, our initial findings that demonstrate amoxicillin’s neurally mediated effect represents an area that could be further studied with isolated muscle strips to allow for further clarification of the specific mechanism. Given that the mucosal layer was kept intact, a tachykinergic or serotonergic mechanism may be contributing as well. Tachykinins (substance P, neurokinin A, and neurokinin B) have similar distribution in the human and rodent intestine and can both stimulate and inhibit intestinal tract motor activity depending on the site and the levels of expression. Some GI disorders associated with dysmotility feature reduced tachykinin sensitivity, which, when antagonized, gives resolution of these motor disturbances (26–29). Serotonin also plays a major role in intestinal tract motor and secretory function and has been implicated in disease states (30,31). Its varied role in GI tract physiology (including its role in the migrating motor complex) is mediated by a number receptor subtypes (32,33). Furthermore, experiments in the presence of tachykinergic or serotonergic inhibitors could be performed to further delineate the mechanism of action of amoxicillin too. Although our results are encouraging, we recognize our work has limitations. Specifically, the relationship between the motility of rat duodenum and human duodenum is unknown. This limits the immediate clinical applicability of our findings. It has been shown that some drugs may have different effects at different sites of the GI tract in animal models compared with their site of action in humans (34). In addition, the methods in this study are unable to measure the complex motor patterns of peristalsis in humans as is done in antroduodenal manometric studies (35). Coupling our findings with the clinical data of Gomez et al, however, there is a clear indication for a clinical study examining the use of amoxicillin alone to improve duodenal motility in children. Given the potential benefit in light of the low cost and wide availability, amoxicillin as an enterokinetic deserves further investigation. The dearth of therapies available to those with motility disorders makes these findings all the more important. In conclusion, our results are the first to demonstrate that amoxicillin modulates the spontaneous pattern of mechanical activity of rat duodenum through noncholinergic, neurally mediated www.jpgn.org
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
JPGN
Volume 61, Number 3, September 2015
mechanisms. Clavulanic acid does not impact the pattern of activity in our system. Our novel findings suggest a distinct role for the use of amoxicillin in the treatment of pediatric dysmotility. This work begins to provide a better understanding of amoxicillin’s function for those patients whose present pharmacologic therapeutic options are extremely limited. Acknowledgments: The authors thank Naoko Murakami for animal husbandry, Drs Jackie Wood, Yusen Liu, and Fievos Christofi for technical advice, and Drs Pam Lucchesi and Aaron Trask for the use of equipment.
REFERENCES 1. Connor FL, Di Lorenzo C. Chronic intestinal pseudo-obstruction: assessment and management. Gastroenterology 2006;130 (2 suppl 1): S29–36. 2. Milla PJ. Motility disorders in childhood. Baillieres Clin Gastroenterol 1998;12:775–97. 3. Yeung AK, Di Lorenzo C. Primary gastrointestinal motility disorders in childhood. Minerva Pediatr 2012;64:567–84. 4. Gariepy CE, Mousa H. Clinical management of motility disorders in children. Semin Pediatr Surg 2009;18:224–38. 5. Quigley EM. Cisapride: what can we learn from the rise and fall of a prokinetic? J Dig Dis 2011;12:147–56. 6. Di Lorenzo C, Youssef NN. Diagnosis and management of intestinal motility disorders. Semin Pediatr Surg 2010;19:50–8. 7. Langer JC, Bramlett G. Effect of prokinetic agents on ileal contractility in a rabbit model of gastroschisis. J Pediatr Surg 1997;32:605–8. 8. Dicken BJ, Sergi C, Rescorla FJ, et al. Medical management of motility disorders in patients with intestinal failure: a focus on necrotizing enterocolitis, gastroschisis, and intestinal atresia. J Pediatr Surg 2011;46:1618–30. 9. Hita EO, Ruiz-Extremera A, Garcia JA, et al. Amoxicillin-clavulanic acid hepatotoxicity in children. J Pediatr Gastroenterol Nutr 2012; 55:663–7. 10. Schwankovsky L, Mousa H, Rowhani A, et al. Quality of life outcomes in congenital chronic intestinal pseudo-obstruction. Dig Dis Sci 2002;47:1965–8. 11. Caron F, Ducrotte P, Lerebours E, et al. Effects of amoxicillin-clavulanate combination on the motility of the small intestine in human beings. Antimicrob Agents Chemother 1991;35:1085–8. 12. Gomez R, Fernandez S, Aspirot A, et al. Effect of amoxicillin/clavulanate on gastrointestinal motility in children. J Pediatr Gastroenterol Nutr 2012;54:780–4. 13. Lykova EA, Bondarenko VM, Parfenov AI, et al. [Bacterial overgrowth syndrome in the small intestine: pathogenesis, clinical significance and therapy tactics]. Eksp Klin Gastroenterol 200551–7, 113.. 14. Salvo F, De Sarro A, Caputi AP, et al. Amoxicillin and amoxicillin plus clavulanate: a safety review. Expert Opin Drug Saf 2009;8:111–8. 15. Turck D, Bernet JP, Marx J, et al. Incidence and risk factors of oral antibiotic-associated diarrhea in an outpatient pediatric population. J Pediatr Gastroenterol Nutr 2003;37:22–6. 16. Robles M, Andrade RJ. [Hepatotoxicity by antibiotics: update in 2008]. Revista espanola de quimioterapia: publicacion oficial de la Sociedad Espanola de Quimioterapia. Rev Esp Quimioter 2008;21:224–33.
www.jpgn.org
Amoxicillin/Clavulanic Acid Duodenum Effect
17. Postorino A, Mancinelli R, Racanicchi C, et al. Spontaneous electromechanical activity in the rat duodenum in vitro. Arch Int Physiol Biochim 1990;98:35–40. 18. Rahnama’i MS, Wagenvoort JH, van der Linden CJ, et al. (Augmentin) resistant Escherichia coli in bacterial peritonitis after abdominal surgery—clinical outcome in ICU patients. Neth J Med 2009;67: 173–6. 19. Savini V, Catavitello C, Balbinot A, et al. Discrepancy between Vitek 2 and the agar disk method in detecting the amoxicillin-clavulanate resistance in a Klebsiella pneumoniae isolate. Neth J Med 2010; 18:53–4. 20. Ferjani A, Marzouk M, Ben Moussa F, et al. [Resistance of Escherichia coli isolated from the urinary tract to amoxicillin-clavulanic acid and other antibiotics]. Med Mal Infect 2010;40:161–4. 21. McGregor JA. Ticarcillin/clavulanate for the treatment of female genital tract infections. Efficacy, safety and comparative microbiology. J Reprod Med 1990;35 (3 suppl):333–8. 22. Barr WH, Zola EM, Candler EL, et al. Differential absorption of amoxicillin from the human small and large intestine. Clin Pharmacol Ther 1994;56:279–85. 23. Bortolotti M, Brunelli F, Sarti P, et al. Effects of oral clarithromycin and amoxycillin on interdigestive gastrointestinal motility of patients with functional dyspepsia and Helicobacter pylori gastritis. Aliment Pharmacol Ther 1998;12:1021–5. 24. Wittmeyer V, Merrot T, Mazet B. Tonic inhibition of human small intestinal motility by nitric oxide in children but not in adults. Neurogastroenterol Motil 2010;22:1078–282. 25. Wood JD. Excitation of intestinal muscle by atropine, tetrodotoxin, and xylocaine. Am J Physiol 1972;222:118–25. 26. Holzer P, Holzer-Petsche U. Tachykinin receptors in the gut: physiological and pathological implications. Curr Opin Pharmacol 2001;1: 583–90. 27. Hellstrom PM, Murthy KS, Grider JR, et al. Coexistence of three tachykinin receptors coupled to Caþþ signaling pathways in intestinal muscle cells. J Pharmacol Exp Ther 1994;270:236–43. 28. Improta G, Broccardo M. Tachykinins: role in human gastrointestinal tract physiology and pathology. Curr Drug Targets 2006;7:1021–9. 29. Valero MS, Fagundes DS, Grasa L, et al. Contractile effect of tachykinins on rabbit small intestine. Acta Pharmacol Sin 2011;32:487–94. 30. Deloose E, Janssen P, Depoortere I, et al. The migrating motor complex: control mechanisms and its role in health and disease. Nat Rev Gastroenterol Hepatol 2012;9:271–85. 31. Bottner M, Barrenschee M, Hellwig I, et al. The enteric serotonergic system is altered in patients with diverticular disease. Gut 2013;62: 1753–62. 32. Hasler WL. Serotonin and the GI tract. Curr Gastroenterol Rep 2009;11:383–91. 33. Cirillo C, Vanden Berghe P, Tack J. Role of serotonin in gastrointestinal physiology and pathology. Minerva Endocrinol 2011;36:311–24. 34. Cinci L, Faussone-Pellegrini MS, Rotondo A, et al. GLP-2 receptor expression in excitatory and inhibitory enteric neurons and its role in mouse duodenum contractility. Neurogastroenterol Motil 2011;23:e383– 92. 35. Lentle RG, De Loubens C, Hulls C, et al. A comparison of the organization of longitudinal and circular contractions during pendular and segmental activity in the duodenum of the rat and guinea pig. Neurogastroenterol Motil 2012;24:686–95.
345
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
