| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
NEUROREGULATION AND MOTILITY
1Department of Physiology, University of Melbourne, Parkville; and 2Howard Florey Institute, Parkville, Victoria, Australia
Submitted 22 April 2008 ; accepted in final form 30 June 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
fed-state/postprandial motor pattern; computational modeling; enteric nervous system; myenteric plexus; enteric reflex circuits; feedforward networks
With the recent development of an in vitro preparation and associated video recording approach (1, 9, 13), segmentation can be observed directly. Segmentation consists of several types of motor patterns that occur as distinct episodes of activity separated by quiescence (12). The most prominent pattern is stationary contractions that are repeated at regular intervals. There are also short-length propagating (SL) contractions that occur at regular intervals and slowly migrate either orally or anally along the intestine for a few centimeters. Finally, there are whole-length propagating (WL) contractions that are stronger and propagate rapidly along the whole length of the segment being studied. WL contractions travel in either the oral or anal directions and may correspond to retrograde and forward peristalsis, respectively, whereas stationary and SL contractions are unique to segmentation. Importantly, the stationary and SL contractions originate from specific fixed locations, which are consistent between episodes of activity. This is consistent with in vivo observations. Furthermore, intestinal transit and nutrient absorption each depend on the relative numbers of stationary (including SL) and propagating (WL) contractions (18–20).
Several lines of evidence indicate that the enteric nervous system (ENS) generates segmentation. Pharmacological agents acting on the ENS, including tetrodotoxin, hexamethonium, and hyoscine, abolish segmentation (12). Furthermore, stationary contractions occur concurrently with excitatory junction potentials and associated muscle action potentials that differ in frequency from slow waves in the same intestinal regions (11, 12). Indeed, stationary contractions are time locked with inhibitory junction potentials in the circular muscle immediately oral and anal to the contraction, indicating the action of a complex neural pacemaker (11).
This raises the question of how the ENS generates stationary motor patterns. Well-studied reflexes such as ascending and descending excitation may account for propagating contractions (4, 27). The ascending excitatory and descending inhibitory pathways form feedforward networks (4, 25) that simplistically should produce only propagating contractions, and not stationary contractions, as activity flows along them. Furthermore, the excitatory and inhibitory motor neurons are polarized (6), so even monosynaptic reflexes would produce oral contractions and anal dilations, not stationary contractions with inhibitory junction potentials in the circular muscle on the oral side. Another mystery is why stationary and SL contractions originate at specific fixed locations along the length of the intestine.
Using a conceptually simple mathematical model simulated on a computer, we investigated general mechanisms by which the ENS could produce segmentation-like patterns. We modeled populations of interneurons and associated motor neurons in the ascending excitatory and descending inhibitory pathways and responses in the muscle. We did not explicitly model the rhythm generator, but we tested locating it in the afferent pathway or in one of the feedforward pathways. Predictions of the model were tested by video imaging of guinea pig jejunum infused with decanoic acid, an established in vitro model of segmentation (12).
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
We developed a simple mathematical model that describes the local mean firing rate of a population of neurons (26). Similar models have been used in the central nervous system (16, 24) and the ENS (E. A. Thomas and J. C. Bornstein, unpublished observations). The intestine was divided into segments along the longitudinal axis (Fig. 1) and the mean firing rate of neurons in functional populations was described by two equations. The functional populations used in this investigation were the interneurons and associated motor neurons in the ascending excitation and descending inhibition. These are the two major reflex pathways in the small intestine and the main pathways causing circular muscle contractions and circular muscle dilations, both of which are important in segmentation (11). The combined output from both pathways was integrated by a simple equation to predict the response of the circular muscle. Input into the interneuron pathways was from an afferent circuit that includes Dogiel type II/AH neurons (3, 4, 10) and possibly other interneurons but was not explicitly modeled in this study. Since segmentation contractions are rhythmic, there must be a rhythm generator in the circuit (11). We examined the implications of having the rhythm generator located in the afferent input or in the ascending or descending pathway (Fig. 2). The ascending and descending pathways were each coupled to form feedforward networks (4, 25). This was implemented by having neuron populations in the ascending pathway projecting orally to two adjacent populations and populations in the descending pathway projected to four adjacent populations in the anal direction (Fig. 1) [descending interneurons project at least twice as far as ascending interneurons (6)]. A range of coupling constants within each pathway and other parameters was tested for all simulation protocols.
|
|
 | (1) |
 | (2) |
E is the time constant for excitation, D is the time constant for synaptic depression, I is exogenous input, and β is a constant to prevent spontaneous activity. Exogenous synaptic input and recurrent activity within the population drives further activity according to a sigmoid activation function, 
E, defined as
 | (3) |
Similarly, activity in the neuron population drives synaptic depression, such as receptor desensitization or negative feedback, according to a sigmoid activation function, D, defined as
 | (4) |
E, kE, D, and kD are parameters that determine the dynamics of the excitation and depression of activity in the population of neurons. The numerical values of the constants can be set to produce two types of dynamic behavior in a local population: accommodation or oscillation (Fig. 2). An oscillating dynamical response was chosen because segmentation is driven by a neural rhythm generator. A second adaptive response was also chosen to describe activity in populations of neurons that do not intrinsically oscillate during a constant input stimulus.
These equations describe activity in a single population of neurons in a short segment of intestine, where the activity will be homogeneous. To extend this model to cover neurons along a length of intestine, we use a variable, Ei, to describe activity at the ith location. Populations at different locations are then coupled to form the feedforward networks of the ascending and descending pathways (25). Therefore, Eq. 1 becomes
 | (5) |
The output of each pathway was played into a simple model that compared the relative strength of input from the ascending and descending pathways to predict the contractile response of the circular muscle. The equation describing the response was
 | (6) |
The constants in Eqs. 1, 2, 3, 4, and 5 (E = 1, D = 2, E = 0.2, kE = 0.05, D = 0.5, kD = 0.05, β = 0.1 or 0.25) were set to produce two dynamical responses (Fig. 2). All numerical values in the model are in arbitrary units and have no direct physiological meaning, other than the dynamical response they produce. Synaptic coupling constants in Eq. 5 were also arbitrary values and a wide range of values were tested in all simulations to cover all physiological possibilities. Output was graphed, unless otherwise stated, as a spatiotemporal map (12) with intestine length on the x-axis, time on the y-axis, and muscle state as a grayscale with white indicating a contraction and black representing a dilation.
In Vitro Experiments
The in vitro segmentation experiments were conducted in accordance with the guidelines of the National Health and Medical Research Council of Australia and with approval from the University of Melbourne Animal Experimentation Ethics Committee. For control experiments, guinea pig jejunum was prepared according to our previously published procedure (12). In lesioned preparations, tissue was first placed in a dish with physiological saline then pinned at the ends (through the tissue) and in the middle (through the attached mesentery). An incision was made following a blood vessel from the mesentery into the muscle, cutting through the longitudinal muscle, myenteric plexus, and circular muscle layers, right around the circumference of the segment (a myotomy). Care was taken to avoid cutting the submucous plexus, which may have altered the structural integrity of the intestinal tube. Lesioned preparations were then placed in an organ bath in the same way as control preparations. After experiments, lesioned tissue was fixed in formaldehyde solution (4% formaldehyde in 0.1 M phosphate buffer) and stained for NADPH-diaphorase (with 10 mg β-NADPH, 2.5 mg nitroblue tetrazolium, 10 ml 0.1 M Tris·HCl, and 20 µl Triton X). Stained tissue was mounted in glycerol and viewed under a microscope to verify that the myenteric plexus was completely interrupted.
Once the tissue was in the organ bath, physiological saline was flushed through the lumen and the preparation was left to equilibrate for 30–60 min. During this period, a 20-min video recording of spontaneous activity was made and the pressure threshold for initiation of peristaltic contractions was determined by raising the inflow pressure in steps of 1 cmH2O at intervals of 
30 s. The lumen was then flushed with saline solution containing decanoic acid (Sigma-Aldrich, Castle Hill, NSW, Australia) at a concentration of 1 mM. Over the following 120–180 min, video recordings were made in 20-min blocks. To allow for varying onset and duration of segmentation between preparations (12), the most active 20-min recording was used for comparisons between preparations.
Video recordings were processed to produce spatiotemporal maps (12): two-dimensional plots with intestinal length on the x-axis, time on the y-axis, and diameters represented by a grayscale.
Contractions were classified within the maps by eye as WL contractions, SL contractions, or stationary contractions, according to published criteria (12). For each contraction, its speed, length, and position along the longitudinal axis were measured. To compensate for longitudinal movements, the position of each frame was adjusted to align the lesion, so that contractions were measured relative to the lesion.
Periods of quiescence and isolated bursting activity were also determined. A period of quiescence was defined as 20 s or longer without a contraction. We measured frequency and duration of quiescent periods. An isolated burst of activity was defined as three or more contractions of a similar length (within 50% of each contraction length), occurring at similar intervals (within 50%), at similar longitudinal positions (the center of each contraction was within the most oral and most anal edges of the largest contraction within the burst) and no other contractions could pass through the same region during the isolated burst. Identifying bursts of activity by these criteria enabled an accurate measurement of the output from the neural rhythm generator by determining the frequency at which contractions occurred within each burst. We also calculated the rate at which isolated bursts were observed and the number of contractions in each burst.
Data in the text are reported as means ± SE. Statistical comparisons were made by unpaired t-tests (P values <0.05 were taken as significance) or by ensuring that mean values were outside the 95% confidence intervals for comparable means.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Local stimuli produce an oral contraction and anal dilation. To test the hypothesis that stationary contractions are due to local nutrient concentration, we modeled the response to a stimulus localized to a small region of intestine. Activity in the ascending pathway could remain in the stimulated region only, propagate a small distance in the oral direction, or propagate the entire length of the segment in the oral direction, when the synaptic coupling strength in the ascending pathway was weak, intermediate, or strong, respectively. Activity in the descending pathway had the same dependence on synaptic coupling strength. Therefore, an orally propagating contraction and an anally propagating dilation were observed when the synaptic coupling strength in both pathways were intermediate or high (Fig. 3A), similar to the response to local distension. This was observed regardless of whether the rhythm generator was located in the afferent pathway or one of the feedforward pathways.
|
Local variations in synaptic coupling create stationary oscillating contractions. We tested the hypothesis that localized contractions result from variations in synaptic coupling, even when the stimulus is uniform along the whole of the intestine. Intermediate or high synaptic coupling strength was required in both the ascending and descending pathways for variations in synaptic coupling to have an effect on the patterns observed. We first tested locating the rhythm generator in the afferent pathway (Fig. 2A). We cut transmission into a single population of descending interneurons from the orally located populations (Fig. 1B), mimicking a local decrease in postsynaptic responses in this feedforward network. Immediately anal to the location of the disruption, descending interneurons received only afferent input and so the activity in this population was reduced. Therefore, inhibitory input into the muscle was reduced, resulting in an oscillating contraction in the region of the pathway with reduced descending input. This pattern mimics stationary contractions seen in vitro.
Similarly, we cut transmission into a single population of ascending interneurons from the anally located populations, mimicking a local decrease in synaptic transmission in this feedforward network. At the location of disruption, ascending interneurons received only afferent input and so the activity in this population was reduced. Therefore, excitatory input into the muscle was reduced, resulting in oscillating dilations at the location of the disruption. This would also produce a pattern of activity similar to segmentation if there were basal tone in the smooth muscle.
We increased the strength of transmission into a single population of ascending interneurons by a factor of 5 (Fig. 1C), mimicking a local increase in synaptic transmission in this pathway. The increased activity in this population drove an oscillating contraction in the muscle at that location, mimicking stationary contractions seen in vitro.
We also increased the strength of transmission into a single population of descending interneurons by a factor of 5, mimicking a local increase in synaptic transmission in this pathway. The increased activity in this population resulted in dilations in the muscle at that location. Local contractions at that point were not produced.
Finally, synaptic transmission from the afferent pathway to both the ascending and descending pathways was abolished in a small region. Directly within the altered region, input into the ascending pathway only came from anally located ascending interneurons, which propagated in small waves into the region of reduced sensory input. Similarly, activity in the descending pathway propagated anally in waves into the region of reduced sensory input. In the middle of the altered region, there were alternating contractions and dilations as the propagating activity in both pathways crossed over each other. Although this produced localized activity, it was qualitatively different from the motor patterns seen in vitro because the contractions and dilations were not coordinated since they were not receiving afferent input.
The rhythm generator must be in the afferent pathway. In the previous experiments, the rhythm generator was in the afferent input into the feedforward networks (Fig. 2A). We also located the rhythm generator in the ascending pathway with both feedforward pathways receiving constant afferent input (Fig. 2B). Under these conditions, the only local variation in synaptic transmission that produced localized activity was a simultaneous reduction in transmission from the afferent pathway to both feedforward pathways, but when the rhythm generator was in the afferent input, this localized activity was qualitatively different to contractions seen in vitro because there was coordination between the contractions and dilations. A decrease in coupling in the descending pathway or an increase in coupling in the ascending pathway only produced a single contraction after removal of the stimulus. Similar results were obtained when the rhythm generator was located in the descending pathway.
Connections between the ascending and descending pathways. The circuitry of the ENS is more complex than two separate reflex pathways. We tested the possibility that synaptic connections between the ascending and descending pathways (15) may influence these motor patterns. When this additional coupling between the ascending and descending pathways was weak, it had no effect on responses to a local or distributed stimuli. When there was strong coupling in only one direction (that is, the ascending pathway received strong input from the descending pathway or vice versa) there were only minor changes on the patterns observed in response to distributed stimuli. Under these conditions segmentation-like patterns were still observed with the local variations in synaptic coupling within the pathways described above. However, in response to a local stimulus, intermediate or strong coupling in only one direction had a major effect on the patterns observed. When the ascending pathway received intermediate input from the descending pathway, in addition to an oral contraction and anal dilation, an anally propagating contraction arose immediately behind the anally propagating dilation (Fig. 3C). This appeared as an anally directed contraction preceded by dilation, very much like descending excitation as seen in vitro. It was only seen in a limited range of coupling strengths between elements of the ascending feedforward pathway. Similarly, when the descending pathway received intermediate or strong input from the ascending pathway, an orally propagating dilation was observed immediately behind the orally propagating contraction triggered by a local stimulus. Strong reciprocal coupling between the ascending and descending pathways resulted in uncontrolled positive feedback and caused contractions and/or dilations to continue indefinitely after either a local or distributed stimulus was removed.
A testable prediction. Our model reproduces localized rhythmic activity via local variations in synaptic transmission. One such variation is an incision through the longitudinal muscle, myenteric plexus, and circular muscle around the entire circumference of the gut wall (a myotomy). We modeled this by disrupting transmission through a location in the simulated intestine for all pathways (Fig. 1D). Note that this differs from the manipulations performed previously, where transmission into a specific location was altered, but longer synaptic connections were able to pass through, in only one pathway at a time (Fig. 1). Similar to previous variations in synaptic coupling, modeling this lesion required intermediate or higher synaptic coupling strength in both the ascending and descending pathways for the lesion to have an effect on the patterns observed. Results for the experiment when the rhythm generator was in the afferent input (Fig. 2A) are illustrated in Fig. 3D. Ascending interneurons in the region immediately oral to the lesion have reduced input, resulting in a reduced excitatory input into the muscle. This in turn results in an oscillating dilation. Further oral activity in the ascending interneurons approaches that of interneurons well away from the lesion. However, because the ascending pathway accommodates, the baseline of activity shifted, allowing a more rapid rise in activity and causing an imbalance relative to the constant inhibitory activity, resulting in weak oscillating contractions. Descending interneurons immediately anal to the lesion received input only from the afferent pathway, resulting in reduced activity in these neurons and hence reduced inhibitory input into the muscle, which in turn produced an oscillating contraction. Thus a lesion would be expected to induce stationary contractions on both sides.
When the rhythm generator was in the ascending pathway with constant afferent input, the lesion only produced a single contraction on the anal side of the lesion and only when the stimulus was removed. Thus our model predicts that if the rhythm generator is in the afferent input, then a lesion induces stationary contractions immediately oral and anal to the lesion.
Segmentation-Like Contractions Are Induced by Lesion In Vitro
We induced segmentation in isolated guinea pig jejunum using decanoic acid and analyzed video recordings to determine the effects of an acute lesion on intestinal movements. Before the addition of decanoic acid, spontaneous activity was recorded for 20 min and the pressure threshold for the initiation of peristaltic contractions was measured (Fig. 4, A and B). The lesion did not affect spontaneous activity (1.3 ± 0.6 contractions/min, n = 10, compared with control 1.6 ± 0.6, n = 10). Similarly, the lesion did not affect the pressure threshold for the initiation of peristaltic contractions (2.0 ± 0.2 cmH2O control, n = 9; 1.9 ± 0.2 cmH2O lesion, n = 10). However, peristaltic contractions did not propagate through the lesion; rather, peristaltic contractions occurred on either side of the lesion (Fig. 4B). These were usually synchronized, but the peristaltic contraction on the anal side of the lesion sometimes occurred prior to the one on the oral side.
|
Contraction rates were calculated in 0.25-mm bins and plotted against longitudinal position (Fig. 5). In lesion preparations, stationary and SL contraction rates were significantly higher up to 2 mm oral and 3.25 mm anal of the lesion (as indicated by the 95% confidence intervals). The lesions did not significantly change either the length of stationary contractions (lesion 3.5 ± 0.3 mm, n = 10; control 4.5 ± 0.4 mm, n = 10) or the length (lesion 14.5 ± 0.8 mm, n = 10; control 17.1 ± 1.5 mm, n = 10) or speed of propagation (lesion 9.1 ± 0.4 mm/s, n = 10; control 9.9 ± 1.0 mm/s, n = 10) of SL contractions. The lesions abolished whole-length propagating contractions.
|
= 0.043), but did not effect their duration (lesion 35.1 ± 3.9 s, n = 10; control 53.5 ± 11.6 s, n = 10; > 0.1). Lesions did not alter contraction frequencies within isolated bursts (lesion 0.198 ± 0.027 Hz, n = 10; control 0.210 ± 0.021 Hz, n = 10) or the number of contractions within an isolated burst (lesion 3.7 ± 0.2, n = 10; control 3.9 ± 0.6, n = 10). However, lesions significantly increased incidence of isolated contraction bursts (lesion 1.3 ± 0.2 per minute, n = 10; control 0.3 ± 0.1 per minute, n = 10). |
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Local Stimuli
In response to local stimuli, our model produced an orally propagating contraction and anally propagating dilation, which is consistent with reflex responses to local distension (3). An anally propagating contraction was only observed when the ascending pathway received intermediate or strong input from the descending pathway. This gave the appearance of an anally propagating dilation closely followed by a contraction, which is similar to descending excitation. Some ascending interneurons do receive input from descending interneurons and descending interneurons receive input from ascending interneurons (15). Further simulations incorporating these additional connections and a separate descending pathway to excitatory motor neurons (14) will be needed to address issues related to peristalsis (27).
Distributed Stimuli With Local Variations in Synaptic Coupling
Local stationary contractions could not be produced from local stimuli because of the heavily polarized feedforward networks. Surprisingly, localized activity in our model required a distributed stimulus. Distributed stimuli allowed the output of the feedforward networks to cancel at the level of the circular muscle, leaving the circular muscle in its basal state.
Under these conditions, local variations in synaptic coupling within either the ascending or descending pathway could induce localized activity in or adjacent to that region provided the synaptic coupling strength in both pathways was intermediate or high (allowing activity to propagate down these feedforward pathways without diminishing). Physiologically, activity in these feedforward pathways can propagate without diminishing because orally propagating contractions and anally propagating dilations have been reported to travel the length of the intestine (2, 7) and the explanation of our in vitro results also requires propagating activity (see below). Therefore, drugs that interfere with synaptic transmission in the ascending reflex pathway should prevent segmentation and it has been previously reported hexamethonium abolishes segmentation (12). Furthermore, drugs that prevent synaptic transmission in the descending reflex pathway should induce contractions. However, if the drug is allowed to act over the entire length of intestine, it is likely to produce large propagating contractions over the entire length of segment instead of localized rhythmic contractions. It has been reported that the P2X antagonist PPADS does not significantly alter segmentation (12); and this may be due to different subgroups of descending interneurons been active during segmentation or because the drug is acting over the entire length of intestine instead of locally reducing synaptic transmission.
Our model predicts that severing both interneuron pathways will induce stationary contractions on either side of the lesion. We tested this by lesioning neural pathways in an intestinal segment in vitro; the lesion induced stationary contractions on both its oral and anal edges. The lesion interrupted whole-length propagating contractions, but it did not affect the number, speed, and length of short-length propagating contractions. Nor did the lesion affect the length of the stationary contractions. Thus the lesion induced localized stationary contractions without affecting properties of stationary or short-length propagating contractions.
Our results indicate that both the ascending and descending pathways must be active during the fed state. Contractions occurring on the oral side of the lesion imply that reducing activity levels in the ascending pathway, while activity levels in the descending pathway remain unchanged, can produce contractions when there is a rhythmic input. Similarly, contractions occurring on the anal side of the lesion imply that decreased activity levels in the descending pathway, while activity levels in the ascending pathway remain unchanged, can also produce contractions when there is a rhythmic input.
Our results indicate that one switch between peristalsis and segmentation could be a change from a local stimulus to a distributed stimulus. In our model, anally propagating contractions required strong synaptic coupling from the descending to the ascending pathway. Our model still produced stationary localized contractions with this coupling present, indicating the same circuit can generate two different motor responses. However, adding synaptic connections from ascending interneurons to descending interneurons when there was strong synaptic coupling from the descending to the ascending pathway had no effect if the synaptic coupling was very weak or led to uncontrolled positive feedback when the coupling was strong. No value of this coupling led to a stable intermediate level of activity. Thus it is unlikely that synaptic coupling from ascending to descending interneurons is strong, unless there are additional mechanisms to control activity. Further investigations into the electrophysiological properties of these synaptic connections are needed to test our proposal that a change from a local to a distributed stimulus switches the motor pattern from peristalsis to segmentation.
In our model, localized contractions occur at specific sites due to mismatches in opposing outputs from the ascending and descending reflex pathways arising from disruptions to activity propagating down these feedforward networks. Recently it has been suggested that a neural program activates inhibitory motor neurons at some locations and inactivates inhibitory motor neurons at adjacent locations (27), which could correspond to a disruption to the descending inhibitory pathway in our model, but currently there is no evidence proving that such a neural program exists. Since the lesion was able to induce stationary contractions, structural variations in the feedforward pathways could also produce localized activity. Under normal physiological conditions, these variations could be structural, for example numbers of synaptic connections or numbers of neurons, or transient, such as receptor desensitization, synaptic rundown, or action of locally released neuromodulators. Currently there is no anatomical evidence supporting or refuting structural variations along the length of the gut, but since sites at which stationary contractions occur can be maintained for hours (12), there must be something special about them. Transient variations have been demonstrated in the guinea pig myenteric plexus, including receptor desensitization (8, 28) and other forms of synaptic rundown (17). These transient variations could cause localized activity anywhere along the length of the intestine.
The Segmentation Rhythm Generator
We tested locating the rhythm generator in the afferent pathway or in one of the feedforward pathways. The model produced stationary rhythmic contractions at a lesion site only when the rhythm generator was in the afferent input. When the lesion induced stationary contractions in vitro, the frequency at which these contractions occurred during isolated bursts did not change. This also suggests that severing the feedforward pathways did not affect the part of the neural circuit responsible for generating rhythmic contractions.
Thus it appears that the rhythmic contractions observed during segmentation are driven by an oscillating afferent input. The afferent input represents activity in AH/Dogiel type II neurons and possibly other interneurons. Previous modeling has predicted that the AH/Dogiel type II neurons can oscillate as a whole population because they form a recurrent network (E. A. Thomas and J. C. Bornstein, unpublished observations). The present model requires all AH/Dogiel type II neurons along a length of intestine to oscillate in a relatively synchronized fashion. Synchronization of activity along long lengths of the myenteric plexus has been demonstrated in the guinea pig colon in response to a stretch stimulus (21–23). The oscillating stationary contractions observed on either side of the experimental lesion were often, but not always, synchronized. Therefore, it is likely there is some form of communication between neuron populations of either side of the lesion. This could be through the submucous plexus or perhaps mechanical coupling through the intestinal wall.
Conclusion
This is the first study to investigate how the known enteric neural circuits could be responsible for segmentation. Our results suggest that spatially localized rhythmic contractions arise from a local imbalance between ascending excitatory and descending inhibitory muscle inputs due to local variations in ascending and descending reflex pathways. The afferent input into the ascending and descending reflex pathways must be distributed along the length of the intestine and must contain the rhythm generator responsible for oscillatory nature of contractions.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
