SEARCH

SEARCH BY CITATION

Keywords:

  • enteric glia;
  • enteric nervous system;
  • neuro-transmitters;
  • Pediococcus acidilactici;
  • pig;
  • probiotic

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments and Disclosures
  8. References

Background  The enteric nervous system (ENS) contains chemically coded populations of neurons that serve specific functions for the control of the gastrointestinal tract. The ability of neurons to modify their chemical code in response to luminal changes has recently been discovered. It is possible that enteric neuronal plasticity may sustain the adaptability of the gut to changes in intestinal activity or injury, and that gut neurons may respond to an altered intestinal environment by changing their neuropeptide expression.

Methods  We used immunohistochemical methods to investigate the presence and localization of several neuronal populations and enteric glia in both the small (ileum) and large (cecum) intestine of piglets. We assessed their abundance in submucosal and myenteric plexus from animals treated with the probiotic Pediococcus acidilactici compared with untreated controls.

Key Results  The treated piglets had a larger number of galanin- and calcitonin gene-related peptide (CGRP)-immunoreactive neurons than controls, but this was limited to the submucosal plexus ganglia of the ileum. Moreover, immunohistochemistry revealed that glial fibrillary acidic protein-positive enteric glial cells were significantly higher in the inner and outer submucosal plexuses of treated animals.

Conclusions & Inferences  The neuronal and glial changes described here illustrate plasticity of the ENS in response to an altered luminal environment in the gastrointestinal tract.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments and Disclosures
  8. References

Probiotics are defined as living microorganisms or components of bacteria that, when administered in the diet, have been shown to have beneficial effects on the health of the host.1 A putative probiotic bacterium should be non-pathogenic, possess gastric acid and bile stability, persist for some time in the gut, and produce antimicrobial substances, be antagonistic against pathogenic bacteria, and able to modulate the immune response.2,3 Once ingested and temporarily present in sufficient numbers in the gut lumen, probiotics are potentially able to influence gastrointestinal functions due to their interactions with the different components of the intestinal wall. It is recognized that the enteric nervous system (ENS) has a unique ability to mediate local reflex activities independently of inputs from the central nervous system (CNS) and the peripheral nervous system.4,5 The very complex regulatory activities of the ENS are made possible by the presence of different types of neurons within the wall of the gastrointestinal tract, including sensory, motor, secreto-motor and interneurons, accompanied by enteric glial cells. The ENS in mammals contains an estimated 108 intramural neurons: this number is approximately the same as the number of neurons found in the mammalian spinal cord.6 The events that are controlled by the ENS are multiple and interconnected, and include motor activity, secretion, absorption, and mucosal blood flow changes.7 In addition, gut intramural neurons are involved in regulating the mucosal activities related to the detection, acquisition and processing of antigens and other molecules originating from microorganisms and nutrients.8

We are interested in the question of whether microorganisms that reside in the gut able to influence the phenotypic characteristics of enteric neurons. More specifically, are changes in the gut microbial community, like those that are linked to the presence of dietary probiotics, able to communicate with enteric neurons, in such a way as to modify gut ‘behavior’? To date, only a very limited number of studies have investigated the influence of probiotics on the chemical code of neurons in the ENS. Kamm et al.9 showed that dietary probiotics (Saccharomyces cerevisiae sp. boulardii) affect the chemical coding of swine myenteric neurons, and Bar et al.10 demonstrated that Escherichia coli Nissle 1917 affects the contractile activity of human isolated smooth muscle strips.

The aim of the present study was to investigate the neural and glial populations in both the submucosal and myenteric plexuses of the pig ileum and cecum in animals fed a standard diet compared to those that had also received the probiotic Pediococcus acidilactici. Such dietary supplementation with P. acidilactici has previously been studied by our group, revealing an increase in villus height and crypt depth, and a larger number of proliferating enterocytes in treated animals compared to controls.11 In fact, there has been increasing interest in the morphology, neurochemistry and function of the ENS in the pig, owing to its structural and physiological similarities to the human gut.12

Material and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments and Disclosures
  8. References

Animals and diet

Sixteen cross breed (Large white × Landrace) female pigs weighing 7.0 ± 0.2 kg and aged 25 ± 2 days, were purchased from a conventional herd. On arrival the piglets were clinically healthy. They were divided into two treatment groups (eight animals per group) to compare the effects of either a control diet (Ctr, = 8) or the same diet supplemented with Pediococcus acidilactici (Pa, = 8). Animals were housed in a standard laboratory facility in individual pens (approximately 3.5 m2) with straw bedding, a light regime of 8 h light/16 h dark, and within sight and sound of one another. The piglets were given ad libitum access to water. All piglets received a starter liquid diet that was either the control (Ctr) or contained lactobacilli (Pa) at a rate of 1 g kg−1 dry feed (2 × 109 CFU g−1 of dry feed). Pediococcus acidilactici was a single live strain (MA18/5M) (Bactocell, Lallemand, France). Feed was automatically dispensed to the piglets three times per day by a liquid feed delivery system. Diets were fortified to meet or exceed nutrient requirements (NRC 1998) for all nutrients. All animals were treated in accordance with both the policies and the principles of laboratory animal care consistent with the European Union guidelines (86/609/EEC), which were approved by the Italian Ministry of Health (Law 116/92).

Micro-anatomical analyses of the gut

At the end of the trial (42 days), the 16 piglets were slaughtered by approved procedures (Italian Ministry of Health; DL n.333/1998), and small segments of both the ileum and cecum (apex) were collected from each animal immediately after killing. The samples (total = 32) were immediately fixed in 4% para-formaldehyde in 0.01 M phosphate-buffered saline (PBS) pH 7.4 for 24 h at 4 °C, dehydrated in a graded series of ethanol, cleared with xylene and embedded in paraffin. For each ileum and cecum sample, three separate serial sections (500 μm apart) were used, as described by Slavin et al.,13 so that three different levels of depth were analyzed. Serial microtome sections (4-μm thick) were obtained from each 500 μm thick section, and were stained with Hematoxylin–Eosin (HE) sequential staining to ascertain structural details.

Other sections from both the ileum and cecum were utilized as follows.

Immunohistochemical analyses (IHC) and cell counts (histometry)

Dewaxed and re-hydrated sections were treated with 0.6% H2O2 in absolute methanol for 15 min, and with normal goat (Dakocytomation, Milan, Italy) diluted at 1 : 20 for 30 min at room temperature to inhibit non-specific reactivity. Sections were then incubated overnight at room temperature in a humid chamber with the primary antisera (Table 1). The primary antisera were diluted with a 0.05 M pH 7.4 Tris–HCl saline buffer (TBS: 0.05 M, pH 7.4, 0.55 M NaCl). After rinsing in TBS, the sections were incubated with biotinylated rabbit IgG for 60 min, rinsed, and then incubated with StreptAvidin–Biotin Complex conjugated to Horseradish Peroxidase (StreptABC/HRP; Vector Laboratories Inc., Burlingame, CA, USA) for 30 min. The peroxidase reaction was developed in a solution of 3,3′-diaminobenzidine tetrahydrochloride (Sigma, Milan, Italy) (0.04% w/v in Tris–HCl 0.05 M, pH 7.4) and H2O2 (0.005%). Developed sections were counterstained with Mayer’s Hematoxylin solution.

Table 1.   Primary antisera used in the study
 CodeSourceDilutionIncubation
  1. 5HT, 5-hydroxytryptamine; CGRP, calcitonin gene-related peptide; ChAT, choline acetyltransferase; GFAP, glial fibrillary acidic protein; nNOS, neuronal nitric oxide synthase; NPY, neuropeptide Y; SP, substance P; TH, tyrosine hydroxylase; VIP, vasoactive intestinal polypeptide.

PGP 9.5AB1761Chemicon1 : 1000Overnight at RT
ChATAb68779Abcam1 : 100Overnight at RT
GalaninT7153Peninsula Lab1 : 1000Overnight at RT
CGRPT4032Peninsula Lab1 : 1000Overnight at RT
SPT4107Peninsula Lab1 : 500Overnight at RT
VIPT4246Peninsula Lab1 : 1000Overnight at RT
nNOSSC648Santa Cruz1 : 100Overnight at RT
NPYSC14728Santa Cruz1 : 100Overnight at RT
5HTAB125Chemicon1 : 100Overnight at RT
GFAPN1506Dakocytomation1 : 1000Overnight at RT
THAB151Chemicon1 : 200Overnight at RT

The specificity of the immunostaining was verified by incubating sections with: (i) PBS instead of the specific primary antibodies (see Table 1); (ii) preimmune sera instead of the primary antisera; (iii) PBS instead of the secondary antibodies; (iv) absorption of some antisera with excess of respective immunogens (3 μg μl−1, see Table 2) before incubation. The results of these controls were negative (i.e. staining was abolished).

Table 2.   List of immunogens used in the study
ImmunogensCodeSource
  1. 5HT, 5-hydroxytryptamine; CGRP, calcitonin gene-related peptide; nNOS, neuronal nitric oxide synthase; NPY, neuropeptide Y; SP, substance P; VIP, vasoactive intestinal polypeptide.

GalaninH 1365Bachem AG, Bubendorf, Switzerland
CGRPH 4924Bachem AG, Bubendorf, Switzerland
SPH 1890Bachem AG, Bubendorf, Switzerland
VIPV 3628Sigma Chemicals, St. Louis, MO, USA
nNOSsc-648 PSanta Cruz Biotechnologies, Inc., Santa Cruz, CA, USA
NPYH 6375Bachem AG, Bubendorf, Switzerland
5HTH 9523Sigma Chemicals, St. Louis, MO, USA

Immunoreactive neurons and enteric glial were counted in the ENS of both ileum and cecum in: (i) inner submucosal plexus (Meissner plexus), (ii) outer submucosal plexus (Schabadasch’s plexus), and (iii) myenteric plexus (Auerbach plexus). Mucosal endocrine cells were also calculated in both ileum and cecum.

Quantifications of mucosal endocrine cells, glial and neuronal cells were referred to each intestinal section area and extrapolated to mm−2 to allow comparison of the data, thus reflecting mucosal endocrine cell density, neuronal cell density and glial cell density.14,15

The observations were made using an Olympus BX51 light microscope (Olympus, Milan, Italy), equipped with a digital camera. The observer was not aware of the origin of the sections.

Double immunofluorescence

The dewaxed and re-hydrated sections were incubated with the first-step primary antiserum, 1 : 10 goat anti-rabbit choline acetyltransferase (ChAT), for 24 h at 18–20 °C, then washed in TBS, and subsequently treated with the Avidin-Biotin blocking kit solution (Vector Laboratories Inc.). The sections were then washed in TBS for 10 min and incubated with a solution of goat biotinylated anti-rabbit IgG (Vector Laboratories Inc.) 10 μg ml−1 in TBS for 1 h at 18–20 °C. After rinsing twice in TBS, the sections were treated with Fluorescein-avidin D (Vector Laboratories Inc.), 10 μg ml−1 in NaHCO3, 0.1 M, pH 8.5, 0.15 M NaCl for 1 h at 18–20 °C. The sections were then washed in TBS and incubated with rabbit IgG (Vector Laboratories Inc.) for 2 h to inhibit binding of the second primary antiserum to the goat anti-rabbit IgG used in the first sequence.16 For the second step of the double immunofluorescence procedure, the slides were subsequently treated with galanin or calcitonin gene-related peptide (CGRP) antiserum (Table 1). Sections were then rinsed in TBS for 10 min and incubated with 10 μg mL−1 goat biotinylated anti-rabbit IgG (Vector Labs) for 1 h at 18–20 °C. The sections were then washed twice in TBS, and treated with Rhodamine-Avidin D (Vector Laboratories Inc.), 10 μg mL−1 in NaHCO3, 0.1 M, pH 8.5, with 0.15 M NaCl for 1 h at 18–20 °C. Finally, slides with tissue sections were embedded in Vectashield Mounting Medium (Vector Laboratories Inc.) and observed using a Confocal Laser Scanning Microscope (FluoView FV300; Olympus). The immunofluororeactive structures were excited using Argon/Helio–Neon–Green lasers with excitation and barrier filters set for fluorescein and rhodamine. Images containing superimposition of fluorescence were obtained by sequentially acquiring the image slice of each laser excitation or channel.

Statistical analyses

Statistical analysis of the quantitative data was performed using the general linear model of the SAS (version 8.1, Cary Inc., NC, USA). Histometrical analyses (cells counts) were conducted by anova using the proc mixed of the SAS package. The mixed model included the fixed effect of treatment and the random effect of the piglet. The individual piglet values were considered to be the experimental unit of all response variables. The data were presented as least squared mean ± SEM. Differences between means were considered significant at < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments and Disclosures
  8. References

To assess if the probiotic used in the study possibly colonized the gut, at the end of the trial some fecal bacterial populations were studied. Lactobacillus ssp population, collected from ileal digesta, was increased by P. acidilactici addition when compared with control diet (Ctr: 10 × 108 CFU g−1; Pa: 27 × 108 CFU g−1; SE: 9.41 × 106; < 0.05, A. Di Giancamillo, G. Savoini, C. Domeneghini, unpublished data), while no difference was observed in the E. coli population. No influence after P. acidilactici oral treatment was observed in the cecal digesta.

When stained with H&E, the sections from both the ileum and cecum revealed the typical structural details for this species. With both H&E and PGP 9.5 (Fig. 1A) -IHC (see below), the ganglia in the submucosal and myenteric plexuses were always evident, containing variable numbers of enteric neurons. The submucosal plexus was distinct with an inner and an outer compartment, in both the ileum and cecum.

image

Figure 1.  Immunopositive staining in ileum samples of treated pigs. Immunofluorescence for PGP 9.5 (A) is clearly visible both in the nerve cell bodies and fibers of the myenteric plexus. Immunoreactivity for choline acetyltransferase (B) is evident in the myenteric plexus (arrow). Galanin (C) immunoreactive (IR) neurons are present in a ganglion belonging to the outer submucosal plexus (arrow); vasoactive intestinal peptide (VIP) (D) IR neurons are present in a ganglion belonging to the inner submucosal plexus (arrows); neuronal nitric oxide synthase (nNOS) (E) IR neurons are present in a ganglion belonging to the outer submucosal plexus (arrows). 5-Hydroxytryptamine (5HT) (F) immunopositive endocrine cells are present in the deep portion of some intestinal glands (arrows). Scale bar = 10 μm (A), 50 μm (B–F).

Download figure to PowerPoint

We used immunoreactivity towards the PGP 9.5 antibody to precisely identify and count neuronal cell bodies. The above-mentioned pan-neuronal marker revealed no changes in response to the P. acidilactici treatment in terms of the density of neuronal cell bodies either in the ileum or in the cecum.

In addition, single neuronal populations were identified by revealing their chemical code using immunohistochemistry. The different neuronal populations were: ChAT-immunoreactive (IR) (Fig. 1B), galanin-IR (Fig. 1C), CGRP-IR, substance P (SP)-IR, vasoactive intestinal polypeptide (VIP)-IR (Fig. 1D), neuronal nitric oxide synthase (nNOS)-IR (Fig. 1E), neuropeptide Y (NPY)-IR, and 5-hydroxytryptamine (5HT)-IR. Mucosal endocrine cells that were IR 5-HT antibody were also detected (Fig. 1F). Moreover, enteric glial cells IR to glial fibrillary acidic protein (GFAP) were detected.

The density of IR neurons and their relative counts are summarized in Table 3 for the ileum and Table 4 for the cecum. In the ileum and cecum of both treated and control animals the ChAT-IR neurons formed the largest population, which out-numbered the other ones that were IR to the different studied neuromodulators and neurotransmitters; ChAT-IR neurons were detected in both the submucosal (both inner and outer) and the myenteric plexuses.

Table 3.   Effect of Pediococcus acidilactici on histometrical analyses related to immuno-histochemistry of ileum
 CtrPaP
ILEUMMeasureNeurons  or endocrine cells mm−2Neurons  or endocrine cells mm−2
  1. Ctr, control animals; Pa, Pediococcus acidilactici-treated animals; 5HT, 5-hydroxytryptamine; CGRP, calcitonin gene-related peptide; ChAT, choline acetyltransferase; GFAP, glial fibrillary acidic protein; nNOS, neuronal nitric oxide synthase; NPY, neuropeptide Y; SP, substance P; VIP, vasoactive intestinal polypeptide.

  2. Values are mean ± SEM; n/ treatment = 8.

PGP 9.5
 Inner submucosal plexus80.68 ± 9.994.00 ± 11.60.488
 Outer submucosal plexus58.66 ± 12.868.03 ± 10.80.866
 Myenteric plexus126.44 ± 16.7122.98 ± 10.70.815
ChAT
 Inner submucosal plexus72.93 ± 16.773.10 ± 14.20.923
 Outer submucosal plexus30.62 ± 11.829.15 ± 9.60.981
 Myenteric plexus70.94 ± 17.378.00 ± 15.80.411
TH
 Inner submucosal plexus
 Outer submucosal plexus
 Myenteric plexus
Galanin
 Inner submucosal plexus19.97 ± 8.863.26 ± 9.90.035
 Outer submucosal plexus15.54 ± 6.443.01 ± 7.30.040
 Myenteric plexus
SP
 Inner submucosal plexus18.55 ± 9.324.11 ± 10.10.802
 Outer submucosal plexus11.47 ± 3.214.00 ± 2.80.713
 Myenteric plexus
CGRP
 Inner submucosal plexus9.65 ± 2.521.75 ± 2.10.041
 Outer submucosal plexus5.94 ± 2.0310.22 ± 0.90.038
 Myenteric plexus
VIP
 Inner submucosal plexus31.35 ± 7.236.08 ± 6.90.469
 Outer submucosal plexus8.15 ± 2.212.13 ± 2.150.599
 Myenteric plexus
nNOS
 Inner submucosal plexus8.65 ± 2.39.93 ± 2.80.673
 Outer submucosal plexus6.73 ± 2.09.73 ± 2.90.555
 Myenteric plexus19.32 ± 4.117.42 ± 5.30.94
NPY
 Inner submucosal plexus13.70 ± 3.411.80 ± 2.90.815
 Outer submucosal plexus9.78 ± 4.112.04 ± 3.30.733
 Myenteric plexus7.83 ± 1.88.24 ± 2.30.602
5HT
 Inner submucosal plexus2.55 ± 0.54.56 ± 1.90.301
 Outer submucosal plexus
 Myenteric plexus
 Endocrine cells286.44 ± 30.6281.13 ± 24.50.873
GFAP
 Inner submucosal plexus136.81 ± 19.3177.42 ± 15.30.033
 Outer submucosal plexus172.89 ± 17.9215.01 ± 20.80.048
 Myenteric plexus107.28 ± 10.8108.21 ± 11.10.958
Table 4.   Effect of Pediococcus acidilactici on histometrical analyses related to immuno-histochemistry of cecum
 CtrPaP
CECUMMeasureNeurons or endocrine cells mm−2Neurons or endocrine cells mm−2
  1. Ctr, control animals; Pa, Pediococcus acidilactici-treated animals; 5HT, 5-hydroxytryptamine; CGRP, calcitonin gene-related peptide; ChAT, choline acetyltransferase; GFAP, glial fibrillary acidic protein; nNOS, neuronal nitric oxide synthase; NPY, neuropeptide Y; SP, substance P; VIP, vasoactive intestinal polypeptide.

  2. Values are mean ± SEM; n/ treatment = 8.

PGP 9.5
 Inner submucosal plexus85.36 ± 5.361.81 ± 6.10.205
 Outer submucosal plexus286.44 ± 63.0296.16 ± 78.60.871
 Myenteric plexus158.11 ± 15.6162.31 ± 11.30.902
ChAT
 Inner submucosal plexus61.81 ± 9.884.30 ± 14.10.445
 Outer submucosal plexus126.85 ± 25.2135.42 ± 19.00.717
 Myenteric plexus102.22 ± 15.8110.08 ± 17.90.811
TH
 Inner submucosal plexus
 Outer submucosal plexus
 Myenteric plexus
Galanin
 Inner submucosal plexus
 Outer submucosal plexus78.70 ± 11.070.23 ± 11.90.752
 Myenteric plexus
SP
 Inner submucosal plexus
 Outer submucosal plexus73.74 ± 16.667.59 ± 9.90.831
 Myenteric plexus
CGRP
 Inner submucosal plexus
 Outer submucosal plexus68.33 ± 13.665.04 ± 11.00.673
 Myenteric plexus
VIP
 Inner submucosal plexus
 Outer submucosal plexus75.66 ± 9.366.20 ± 11.20.301
 Myenteric plexus
nNOS
 Inner submucosal plexus
 Outer submucosal plexus45.08 ± 4.744.16 ± 5.30.958
 Myenteric plexus67.07 ± 5.968.81 ± 3.20.806
NPY
 Inner submucosal plexus
 Outer submucosal plexus65.23 ± 8.572.33 ± 9.60.822
 Myenteric plexus28.19 ± 4.325.15 ± 7.20.799
5HT
 Inner submucosal plexus30.67 ± 3.826.89 ± 4.00.602
 Outer submucosal plexus8.11 ± 2.25.37 ± 1.80.669
 Myenteric plexus
 Endocrine cells113.58 ± 15.291.63 ± 13.80.563
GFAP
 Inner submucosal plexus71.4 ± 9.391.8 ± 14.50.232
 Outer submucosal plexus65.0 ± 12.190.2 ± 16.20.114
 Myenteric plexus95.36 ± 9.4109.9 ± 12.60.388

Ileum

Galanin-, SP-, CGRP-, and VIP-IR neurons were detected in either the inner or the outer submucosal plexus in the ileum of all the studied animals, whereas nNOS- and NPY-IR neurons were also observed in both the submucosal and the myenteric plexus. 5-Hydroxytryptamine-immunoreactivity was observed in both neurons and mucosal endocrine cells (Fig. 1F): 5-HT-IR neurons were only present in the inner submucosal plexus. Pediococcus acidilactici-treated animals showed a higher number of galanin-IR neurons in the inner (= 0.025) and outer (= 0.045) submucosal plexus of the ileum than the controls. In addition, P. acidilactici dietary supplementation was accompanied by a higher number of CGRP-IR neurons in both the inner (= 0.050) and outer (= 0.044) submucosal plexus of the ileum in comparison with control animals. Furthermore, GFAP-positivity was present in small cell bodies and in their cytoplasmic processes in ganglia belonging to the inner and outer, and in the myenteric plexuses. Anti-GFAP-IR cells were significantly more numerous in the inner and outer submucosal plexuses of the ileum of treated animals in comparison with controls (= 0.033 and = 0.048, respectively).

Cecum

Galanin-, SP-, CGRP-, and VIP-IR neurons were detected in the outer submucosal plexus in the cecum of all the studied animals, whereas nNOS- and NPY-IR neurons were observed both the outer submucosal and in the myenteric plexus. 5-Hydroxytryptamine-immunoreactivity was observed in both neurons and mucosal endocrine cells: 5-HT-IR neurons were present in both the inner and outer submucosal plexus, whereas they were absent in the myenteric plexus, as in the ileum. The P. acidilactici dietary treatment was not accompanied by significant quantitative changes in the different neuronal populations and glial cells in the cecum.

Double fluorescence labeling revealed quite similar results in both the ileum and cecum. Double immunofluorescence aimed at identifying the nature of the two types of neurons that were found to be quantitatively affected in the ileum by the dietary intervention (see above), revealed that galanin- (Fig. 2) and CGRP-IR neurons displayed a 100% co-localization with ChAT-IR neurons in both the inner and the outer submucosal plexus of the ileum. The co-localization is referred only to galanin- and CGRP-IR neurons vs ChAT-IR neurons, but not vice versa.

image

Figure 2.  Immunofluorescent stained ileum samples of treated pigs. Choline acetyltransferase-immunoreactive (IR) (A) and galanin-IR neurons (B) are evident in the same inner submucosal plexus ganglion. Scale bar = 20 μm.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments and Disclosures
  8. References

The ENS contains chemically coded populations of neurons that serve specific functions for the control of the gastrointestinal tract and as such it has been described as the ‘second brain’ in the mammalian body.17 This term is applied not only with respect to neuron numbers, but also with respect to their extremely complex interactions, as revealed by the presence of neurotransmitters and a multiplicity of both inhibitory and excitatory neuromodulators.4,5

Neuronal plasticity can be defined as the potential of the elements of the nervous system to react with adaptive changes to intrinsic or extrinsic inputs. Neuronal plasticity, therefore, is basically a flexible property of the neurons, or neuronal networks, to change, temporarily or permanently their biochemical and physiological characteristics. Similarly to neurons of the CNS, aspects relating to enteric neuronal plasticity have been have recently been described. First of all, increasing age,18 and some pathological gut conditions,19–21 affect the number and types of enteric neurons and nerve fibers. Other authors have described qualitative and/or quantitative changes in enteric neuronal populations following experimental treatments. Mucosal acid challenge has been shown to affect the intramural nitrergic neurons of the rat stomach that are localized in the myenteric plexus;22,23 dietary yeast has been shown to affect swine myenteric neurons;9 and bacterial lipopolysaccharide has been reported to be able to quantitatively influence cultured porcine myenteric neurons.24 Cowen et al.25 demonstrated that a restricted diet reduced the number of enteric neurons that underwent extensive age-related death in the rat, whereas Oste et al.26 showed that the passage from total parenteral to enteral nutrition in preterm pigs enhanced the number of nitrergic neurons. Quite recently, Liu et al.27 showed that the ENS is highly susceptible to injuries caused by the gut luminal contents, and for this reason the plasticity of its neurons may be expressed, in adult mice, by a possible local neurogenesis.

This study provided a quantitative evaluation of neuronal populations in the submucosal and myenteric plexuses of the pig ileum and cecum, and described specific changes in the numbers of some enteric neuronal populations in animals treated with dietary P. acidilactici, suggesting that changes in the intestinal microorganism community, like that linked to dietary probiotic administration, are able to interact with intramural neurons.

The density of intramural neurons was not affected by the studied dietary intervention in either the submucosal or the myenteric plexuses, at least of those neurons visualized by PGP 9.5-immunohistochemistry. In addition, the density of intramural cholinergic neurons did not show any changes in response to the dietary treatment. In the present study, the cholinergic neuronal population largely out-numbered the other neuron types. In humans,28 and in large mammalian species,9,29 the enteric cholinergic neurons account for 50–70% of enteric neurons, whereas in laboratory animals (guinea-pig) their number is larger (around 80% in the small intestine).4 The fact that the cholinergic neurons in both the ileum and cecum out-numbered the intramural neurons shown by the PGP 9.5 neuronal marker is possibly due to technical aspects, because the effective possibility of revealing neurons by the so-called pan neuronal markers may vary in different species and localizations.30 We did not observe catecholaminergic neurons, unlike Li et al.,31 who described the presence of dopaminergic neurons in the mouse intestine.

On the other hand, in comparison with the controls, the P. acidilactici-treated animals had a larger density of galaninergic and CGRP-positive neurons in both the inner and the outer submucosal plexus of the ileum, but not of the cecum. A previous study by our group,11 on dietary intervention with P. acidilactici showed that it was effective on some other morpho-functional aspects of the pig ileum but not of the cecum, possibly due to different mechanisms of action and the presence of different target structures of the probiotics. The other studied neuronal populations, the one linked to the synthesis and release of nitric oxide, and those linked to the synthesis and release of neuropeptides (SP, VIP, NPY) and also serotonin, were not affected by the dietary intervention studied here. Moreover, the endocrine cell types linked to the synthesis and release of serotonin were not quantitatively influenced by the treatment in this study either.

Galanin and CGRP are widely distributed in the mammalian ENS, where they can act as neuromodulators and neurotransmitters. Galanin displays multiple roles, being involved in regulating peristalsis, secretion, blood flow, and feeding behavior. Calcitonin gene-related peptide is also involved in many gut physiological functions, including the modulation of sensory functions and the regulation of smooth muscle activity. The fact that the larger numbers of galaninergic and CGRP-positive neurons in treated than in control pigs were detected in the two compartments of the ileum submucosal plexus lead us to hypothesize that the regulating functions of these two peptides are predominantly exerted towards the ileum tunica mucosa. Why only these two neuropeptides appeared to be influenced by the dietary treatment in this study is at present a matter of debate (also on the basis of the paucity of similar data in the body of literature on swine) and deserves further investigation.

The neuroglial component was also affected by the studied dietary intervention, even though the effect was limited to the inner and outer submucosal plexuses of the ileum. The functions of glial cells in the gut (which resemble CNS astrocytes32) are very complex and not particularly well known,33 but, briefly, they may support enteric neurons and contribute to mucosal defensive functions, and helping the ENS in its adaptation to intestinal injuries or changed conditions.34,35 We have found that the inner and outer submucosal plexuses of the ileum are characterized by an enhanced density of glial cells in treated animals. Accordingly, Van Haver et al.35 found that increasing age and the introduction of enteral food (with its bacterial counterparts) in preterm pigs is linked with an increased number of glial cells in the inner submucosal plexus, possibly aimed at ensuring the structural and functional integrity of the intestinal mucosa. Interestingly, glial cells are also reputed to be involved in the regulation of enteric neuron plasticity by the secretion of neurotrophic factors.33,35

We suggest that the neuronal and enteric glial cell changes we have described as consequences of a probiotic dietary intervention in swine may conceptually support aspects related to enteric neuronal plasticity. Enteric neuroplasticity consists of the ability of the nervous system to adapt its structural organization to new situations. Consequently, gut neurons may respond to an altered intestinal environment by changing their neuropeptide expression, and consequently their chemical coding. This putative mechanism may be referred as ‘neurotransmitter/neuromodulator plasticity’, and may play a crucial role in the progressive adjustment of neuronal functions to altered gut conditions, both in health and disease.21,36,37

Acknowledgments and Disclosures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments and Disclosures
  8. References

The authors have no financial interest in the drugs used for this study. The authors wish to thank Dr. E. Chevaux (Lallemand, France) for the gift of the probiotic used in this study.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgments and Disclosures
  8. References