Main

IBS is a disorder characterized by chronic abdominal pain and discomfort associated with alterations in bowel habits in absence of any detectable organic disease (1). Although IBS is a heterogeneous disorder in terms of etiology and pathophysiology, alterations in bowel habits are likely related to alterations in autonomic regulation of gut motility, whereas symptoms of abdominal pain and discomfort are thought to involve changes in perception of visceral events resulting from visceral hyperalgesia or allodynia (2,3).

Progress in the development of effective therapies has been hampered due to the lack of relevant animal models mimicking the key features of IBS. Many preclinical investigations have been performed with models of acute inflammatory insults to the gut (glycerol, mustard oil, acetic acid, zymosan), thereby mimicking more IBD than IBS, which is characterized by the lack of apparent inflammatory changes within the gut mucosa. Recently, several potential IBS models have been proposed, all of which mimicking some aspects of the human syndrome (4–6). They include an early life colon irritation (4), an adult stress-induced hypersensitivity to colonic distension (7), and an adult postinfective state in rodents (8,9). However, all of these models only reproduced some aspects of the pathophysiological features of the human syndrome.

Despite the incomplete knowledge about the pathophysiology of IBS, different types of stressors appear to play key roles in the development of IBS as well as in the modulation and maintenance of the disease throughout life. Indeed, early traumatic experiences such as childhood neglect, abuse, loss of a parent, and life-threatening situations during childhood have been shown to increase the risk of IBS development (10–12). Consequently, many investigators have attempted to develop animal models of stressful experience in childhood and, postnatal handling, early separation and daily periodic NMD have been proposed. The pioneering work of Levine et al. (13–15) has shown that even quite subtle alterations in the experience of rats during the early postnatal period can have long lasting consequences in terms of behavior, emotionality and stress responsiveness. Multiple variations of the NMD procedures exist with regard to duration (1–24 h) and number (1–14 d during the first 3 postnatal weeks) of the separation episodes. A primary feature of the NMD phenotype has been an enduring dysregulation of HPA axis reactivity to stress (16). Indeed, in the developing rodent, a period around postnatal d 2–14 is characterized by a limited, or even absent, adrenal response to stress (17), and has been defined as the stress hyporesponsive period (SHRP). Numerous experiments have demonstrated that maternal factors are critical for the regulation of the pup's HPA axis and the maintenance of SHRP (15). NMD abolishes this physiologic inhibitory state and induces a long-lasting hyperresponsiveness of the HPA axis response (18). In addition, rats periodically deprived during the early postnatal period show weak learning ability, memory capacity, associated with lower levels of acetylcholinesterase activity within the CNS (19), and a decreased immediate exploratory locomotor response to novelty (20). In agreement with these neuroendocrine changes, deprived rats also exhibit an increased anxiety-like behavior, anhedonia, an enhanced preference for ethanol, and impairment of male sexual behavior (16,21–24). However, it is well known that the effects of maternal separation depend upon such variables as duration of each episode of separation, number of episodes experienced, and timing of episodes during development. Nevertheless, among the different variables of NMD models, one procedure where newborn rats are daily removed from the mother for several hours during the first 2 wk of life (22,24–27) has been largely used to describe the repercussions of NMD on the gastrointestinal tract, and to develop an experimental model to investigate the etiology and pathophysiology of IBS (28–31).

NEONATAL NOXIOUS STIMULI

The early neonatal period is a time of great plasticity for both somatic and visceral sensory systems. This neuronal plasticity often contributes to adaptive or maladaptive function of the mammalian nervous system, and possibly to the development of chronic pain. The nociceptive neuronal circuits are both formed during embryonic and postnatal times when painful stimuli are absent or limited. During this critical period, particularly before the maturation of the descending inhibitory systems (32,33), pain can lead to prolonged structural and functional alterations in nociceptive pathways that can be maintained in adult life (4,34–36).

The first model of neonatal noxious somatic stimuli has been proposed by Anand and co-workers (37,38). In this model, the noxious stimulus was applied daily from day of birth to postnatal d 7. The paws of rat pups were stimulated once, twice, or four times a day at hourly intervals with a needle inserted rapidly through the paw. This experimental model may lead to an altered development of the nociceptive system characterized by decreased pain thresholds and hypersensitivity (37,38). Another model of early pain experience (laparotomy) on the day of birth has also highlighted the occurrence of long-lasting changes in pain behavior, probably mediated by alterations in the HPA-axis and antinociceptive circuitry (39).

Considering the gut, the first model of neonatal noxious stimuli has been developed by Al-Chaer et al. (4,40). In this model, colon injury was applied daily using colorectal distension during postnatal development between d 8 and 21. The distension was applied using an angioplasty balloon, inserted rectally into the colon in the awake neonate. Colon injury in neonates results in chronic visceral hypersensitivity, with characteristics of allodynia and hyperalgesia in adult rats, associated with peripheral sensitization, as well as central sensitization in the absence of identifiable peripheral pathology (4,40). Finally, the importance of neonatal noxious stimuli in the development of human functional intestinal disorders in later life has been recently pointed out by Anand et al. Indeed, noxious stimulation caused by gastric suction at birth may promote the development of long-term visceral hypersensitivity and cognitive hypervigilance, leading to an increased prevalence of functional intestinal disorders in later life (41). Although each neonatal noxious model mimics one feature of human syndrome and also evidence the importance of these traumatic experiences during neonatal period, all of these experimental models fail individually to reproduce all of the IBS features (Table 1).

Table 1 Neonatal stress models and irritable bowel syndrome features
Full size table

NEONATAL INFLAMMATORY STIMULI

Developmental plasticity in physiologic systems is an important mechanism through which organisms can adapt their physiologic responses to better meet environmental demands. Although such alterations may prove some benefit for immediate survival, they may also result in a permanent alteration of physiologic responses to environmental challenges, and alter predisposition to pathology later in life (42–44). Thus, some groups have investigated the impacts of different inflammatory stimuli [lipopolysaccharide (LPS) administration, colonic inflammation] applied during early stages of postnatal development, on adult physiology.

Exposure of neonatal rats to a low dose of Salmonella enteritidis endotoxin (i.p. route) during the first week of life resulted in long-term changes in HPA axis activity (45). Moreover, this neonatal endotoxin exposure induced long-lasting effects on immune regulation, including increased sensitivity of lymphocytes to stress-induced suppression of proliferation and a protection from adjuvant-induced arthritis (45). Neonatal (d 3 and 5) LPS exposure alters behavioral and clinical course of periodontal disease in adult rats (46). In addition, it has been reported that a single LPS exposure during development (d 14) impacts upon adult pain and sensory processing, leading to a decrease in nociceptive thresholds and increases in responses to painful and even innocuous stimuli in adulthood (47).

Neonatal colonic inflammation, induced by mustard oil between postnatal d 8 and 21 results in chronic visceral hypersensitivity associated with central neural sensitization in the absence of identifiable peripheral pathology in adult rats (4). Taken together, these models highlight the emphasis of the neonatal inflammatory stimuli in the development of some aspects of the human IBS in later life. However, none of these models mimics all of the IBS features (Table 1).

NEONATAL MATERNAL DEPRIVATION

There is a growing body of evidence that indicates that all different aspects of mother–infant interactions play an important role in the development of the newborn. Studies, mainly performed in developing rodents, have demonstrated that the mother behavior influences various physiologic parameters in the infant as heart rate, sleep/wake cycles, and growth hormone production (48). Specific physiologic changes in the infant occurring slowly over a relatively protracted period of separation can be tightly linked to specific features of mother-infant interaction. Thus, 24 h of NMD decreases mean cardiac frequency (49). This decrease may result from the long duration of mother's milk deprivation, rather than a lack of other aspects of maternal care such as growth hormone secretion induced by tactile stimulation from the mother.

Although less investigated than emotional behavior or stress reactivity, several studies have established that early traumatic experience alters gastrointestinal homeostasis in the adult. The first studies have reported that early maternal deprivation increases gastric ulcer risk, and that early weaning predisposes rats to exacerbated stress-induced ulcer formation (50–52). More recently, some laboratories have developed NMD models, using different rat strains, which have highlighted the consequences of NMD on intestinal physiology in adults (28–31,53,54). In all these models, rat pups were daily exposed to a 180-min period of NMD during the stress-hyporesponsive phase (Table 2).

Table 2 Long-term consequences of different NMD models on gut functions
Full size table

Gut mucosa immune status.

NMD is described to modify the gut immune status in young and adult rats (31,55). Macroscopic observation of the colon of adult deprived rats reveals the presence of both focal hyperemia and mesenteric adherences (31). These macroscopic alterations are associated with several immune modifications. In both 14-d-old and 12-wk-old rats, colonic mucosa of deprived rats presents a greater number of mast cells than in controls (55). An increase of polymorphonuclear neutrophils number is also observed in both colonic and jejunal mucosa of adult maternally deprived rats (31,53). Moreover, expression of mRNA encoding for cytokines as interferon (IFN)-γ, IL-1β, IL-2, IL-4, and IL-10 in the colon is higher in deprived than in control rats. Thus, NMD induces an increase of both Th-1 and Th-2 cytokine profiles associated with a low-grade systemic inflammatory state in colon (31). In addition, it has been shown that NMD in rats facilitates primary infection by Nippostrongylus brasiliensis and enhances the inflammatory response of the jejunum, but does not induce severe breakdown of immunity to N. brasiliensis (53). Deprived rats exhibited an increase number of jejunal worms and fecal eggs and larvae on d 7 after primary infection associated with an increased MPO activity in jejunal mucosa. These data support that NMD promotes long-term alterations of the intestine immune status and reduces intestinal defense toward a nematode infection.

Visceral sensitivity and motility.

Coutinho et al. (30) have shown that deprived rats exhibit visceral hyperalgesia and cutaneous hypoalgesia under baseline conditions, as well as visceral hyperalgesia and enhanced colonic motility in response to an acute psychological stressor in later life. Thus, exposure to a novel environment or acute stress resulted in a significantly enhanced fecal pellet output, used as an index of colonic motility, in deprived rats compared with nondeprived rats. More recently, using the same neonatal stress protocol it has been reported that deprived rats also develop acute and delayed stress-induced visceral hyperalgesia, associated with an enhanced stress-induced increase in colonic motility and increased anxiety-like behavior (56).

NMD provokes a greater visceral hypersensitivity to rectal distension in female than in male adult (28). This visceral hyperalgesia is also greater when all pups are removed from home cage than when only half of littermates is removed (28). This indicates that stress of the mother plays a role in the long-term effects of NMD. However, acute restraint stress induces hyperalgesia in control females only, and this effect was similar when either half of littermates or all pups were removed. These results indicate that long-term visceral hyperalgesia depends upon the type of NMD and that females are more sensitive than males (28).

Taken together, these data make evident that early traumatic experiences induce long-term changes of gut sensitivity under basal and stressful situations.

Intestinal epithelial permeability.

A first series of experiments performed by Perdue et al. (54) has demonstrated that NMD provokes an immediate increase in passage of macromolecule through the colonic mucosa associated with higher amounts of bacteria adhering to and penetrating into the colonic epithelium. However, these colonic alterations rapidly return to normal values (54). In addition, a mild stress increase short-circuit current, conductance, and transepithelial transport of macromolecules in maternally deprived rats, while having minimal effects in controls (29), evidencing that neonatally deprived rats develop persistent mucosal barrier dysfunction characterized by an impaired host defense to luminal bacteria and enhance gut mucosa vulnerability to stress.

Other studies have reported that repeated NMD induce long-term alterations of gut paracellular permeability in adult rats (31,53,55). Male Wistar pups separated during postnatal d 2–14 exhibit at 12 wk of age, an increase in both jejunal and colonic paracellular permeability, associated with bacterial translocation (31). Similarly, weaning caused marked disturbances in intestinal barrier function, as demonstrated by significant reductions in transepithelial electrical resistance and increases in intestinal permeability to mannitol compared with unweaned control pigs (57).

Pathways involved in the genesis of visceral and permeability alterations.

Among previous studies, some of mediators and cell type have been characterized in the genesis of visceral sensitivity and gut permeability alterations induced by NMD. Thus, NGF and CRF (29,54,55) and mast cells (55) participate in triggering and maintaining gut dysfunction provoked by NMD. However, other cell types such as polymorphonuclear neutrophils and neurons and other mediators such as cytokines (58,59) and proteases (60) might be also involved in the gut alterations induced by NMD. Polymorphonuclear neutrophils and nerve fibers are described in some models of pain to act in concert with mast cells to generate noxious stimuli (61,62). In addition, mediators released from mast cells such as cytokines [IFN-γ, tumor necrosis factor (TNF)-α and IL-4] or proteases (rat mast cell protease II) promote visceral pain and/or increase in gut paracellular permeability.

CRF is largely known to be involved in stress-induced gut disturbances in adult animals, and its role in gut dysfunctions induced by NMD has been described recently. Indeed, treating (i.p. route) animals with either nonselective CRF-R1/2 (α-helical CRF 9-41) or selective CRF-R1 (NBI-35965, CP-154, 526 and SSR-125543) receptor antagonists abolishes the effects of NMD on gut permeability, bacterial translocation, and visceral hypersensitivity (29,54,56,63,64). It is established that CRF-R1 mediates stress-induced gut disturbances in adult animals (64–66); these results confirm that CRF-R1 is likewise involved in NMD-induced gut dysfunctions in newborn rat. However, since the literature is controversial regarding the ability of CRF antagonists to cross the blood-brain barrier, it is unclear whether CRF mediated NMD-induced gut alterations through pathways involving central and/or peripheral mechanisms. In addition, the origin of CRF is still under debate. Although CRF secretion is mainly localized in the CNS, a colonic CRF production has been detected in the mucosal cells in the neighborhood of the base of the colonic crypts (67,68). In agreement, we have shown that both preproCRF and mature CRF are present within the colonic mucosa, and that NMD increases preproCRF expression in colonic mucosal, while it does not modify colonic mucosal CRF content in adult rats (63). These results also reinforce the likely peripheral involvement of CRF in NMD-induced gut disturbances.

NGF is able to trigger noxious stimuli (62) and NGF levels within the CNS have been found increased in rats submitted to NMD (69). Recently, it has also been shown that NGF participates in the genesis of gut dysfunction triggered by NMD (55). Indeed, NGF neutralization by specific antibodies (i.p. route), given before each neonatal stress session, suppressed the effects of NMD on visceral hypersensitivity to rectal distension and gut permeability (55,63). However, little is known about central and/or peripheral mechanisms involved in the NGF induced gut dysfunction in deprived rat. Nevertheless, since NGF depletion (autoimmunization) produces thermal hypoalgesia, associated with high serum titers of anti-NGF IgG, and an absence of anti-NGF IgG in the cerebrospinal fluid (70), one can hypothesize that antibodies directed against NGF do not cross the blood-brain barrier and that thermal hypoalgesia is the consequence of peripheral NGF neutralization. Accordingly, gut dysfunctions induced by NMD are likely mainly triggered by a peripheral mechanism. In agreement with this hypothesis, we have observed an enhanced colonic mRNA and protein expression of NGF in deprived rats that can be related to these gut alterations (55).

These data raise the question of the interplay between CRF and NGF, and several studies have brought investigators to focus on mast cells as a pivotal effecter of NMD. During the last decades, studies from various groups have highlighted the importance of mast cells in stress-related changes in intestinal barrier function (65), i.e. mucin or ion secretion, water secretion, para- and transcellular permeability, and visceral pain. Thus, in adult rats, repeated stress has been shown to increase the number of colonic mucosal mast cells, which play an important role in stress-induced increase in gut paracellular permeability and visceral hyperalgesia, as demonstrated by using mast cell deficient Ws/Ws rats (71), or mast cells stabilizers (7,72). It is noteworthy that the use of a mast cell membrane stabilizer abolishes both visceral hypersensitivity and increased permeability induced by NMD in adult rats (55). Moreover, CRF is a well-known mast cell secretagogue, and mast cells synthesize, store, and release NGF (73). Thus, the close link between CRF, mast cells, and NGF has been recently established in animals submitted to NMD (63), since it has been reported that CRF, through CRF-R1 receptor, promotes NGF release from mast cells, inducing a downstream elevation of gut permeability (Fig. 1) (63).

Figure 1
figure 1

Putative mechanisms involved in NMD-induced visceral hyperalgesia and increase in gut paracellular permeability. The interaction between CRF and mast cell CRF-R1 receptors provokes NGF and other mediators release. The resulting activation and release of mediators such as NGF, proteases or cytokines may (1) directly target the intestinal epithelial cells or (2) sensitize nerve ending to promote visceral hyperalgesia and activate enterocyte to increase the paracellular permeability.

Full size image

Neonatal maternal deprivation, a pathophysiological model for IBS.

Recently, clinical studies on biopsies from IBS patients have brought an important advance in the understanding of the mechanisms involved in visceral pain, and strengthened the observations in animal models (74,75). In IBS biopsies, mast cell density is increased in the intestinal mucosa, and mast cells appear closer to nerve fibers than in controls (74,75). Indeed, in the ileum and rectosigmoid mucosa of IBS patients, the density of neuron specific enolase, substance P, and serotonin positively stained nerve fibers was increased and appeared in clusters, surrounding an increased number of mast cells (74). Moreover, colonic mast cell infiltration and mediator release in proximity to mucosal innervation may contribute to abdominal pain perception in IBS patients (75). The severity and frequency of perceived abdominal painful sensations are also correlated with the presence of activated mast cells (tryptase release) in proximity of nerve endings in the gut wall (75). In addition, clinical studies using sodium cromoglycate as a mast cell stabilizer have reported the primary role of mast cells in IBS. Thus, mast cell–mediated food intolerance has been evidenced as a major factor in the pathogenesis of IBS, and sodium cromoglycate treatment for one month resulted in a significant improvement of intestinal IBS symptoms (76–78).

In agreement with these clinical data, we have reported that NMD promotes long-term changes of colonic mast cell phenotype and nerve terminal distribution, characterized by an increase of nerve endings density in close proximity of mast cells within the colonic wall (unpublished data). In addition, the amount of mediators (RMCP II) released is also increased within the colonic mucosa of deprived rats (unpublished data). Since mast cells mediators (tryptase, cytokines) are known to alter enteric nervous system physiology and to promote visceral hypersensitivity (60,79), the higher release of such mediators concomitant with the closer proximity of colonic innervation also supported that these anatomical modifications may contribute to the disturbed sensory-motor function (28,30).

In conclusion, the studies reviewed herein indicate that various types of early traumatic events have long-term consequences on gastrointestinal functions. These studies have brought important advances in the understanding of mechanisms probably involved in the development of IBS. Among the different early traumatic experiences applied during the neonatal period, NMD results in permanent visceromotor and somatic alterations associated with neurochemical changes, altered HPA responsiveness to stressors, and an increased risk of developing depression-like behaviors, thereby mimicking all the main features of IBS in humans (Table 1). Consequently, NMD may constitute a valuable experimental model to investigate the pathophysiology of IBS and to identify novel pharmacological targets.