High-protein diet differently modifies intestinal goblet cell characteristics and mucosal cytokine expression in ileum and colon

Available online at www.sciencedirect.com

ScienceDirect Journal of Nutritional Biochemistry xx (2014) xxx – xxx

High-protein diet differently modifies intestinal goblet cell characteristics and mucosal cytokine expression in ileum and colon ☆ Annaïg Lan a,⁎, Mireille Andriamihaja a , Jean-Marc Blouin a , Xinxin Liu a , Véronique Descatoire b, Caroline Desclée de Maredsous a , Anne-Marie Davila a , Francine Walker b , Daniel Tomé a , François Blachier a a

AgroParisTech, Institut National de la Recherche Agronomique (INRA), Centre de Recherche en Nutrition Humaine-Ile de France, UMR 914 Physiologie de la Nutrition et du Comportement Alimentaire, Paris, France b Service d'Anatomie et Cytologie Pathologique, Hôpital Xavier Bichat, Paris, France

Received 8 January 2014; received in revised form 2 September 2014; accepted 11 September 2014

Abstract We have previously shown that high-protein (HP) diet ingestion causes marked changes in the luminal environment of the colonic epithelium. This study aimed to evaluate the impact of such modifications on small intestinal and colonic mucosa, two segments with different transit time and physiological functions. Rats were fed with either normal protein (NP; 14% protein) or HP (53% protein) isocaloric diet for 2 weeks, and parameters related to intestinal mucoussecreting cells and to several innate/adaptive immune characteristics (myeloperoxidase activity, cytokine and epithelial TLR expression, proportion of immune cells in gut-associated lymphoid tissues) were measured in the ileum and colon. In ileum from HP animals, we observed hyperplasia of mucus-producing cells concomitant with an increased expression of Muc2 at both gene and protein levels, reduction of mucosal myeloperoxidase activity, down-regulation of Tlr4 gene expression in enterocytes and down-regulation of mucosal Th cytokines associated with CD4+ lymphocyte reduction in mesenteric lymph nodes. These changes coincided with an increased amount of acetate in the ileal luminal content. In colon, HP diet ingestion resulted in a lower number of goblet cells at the epithelial surface but increased goblet cell number in colonic crypts together with an increased Muc3 and a slight reduction of Il-6 gene expression. Our data suggest that HP diet modifies the goblet cell distribution in colon and, in ileum, increases goblet cell activity and decreases parameters related to basal gut inflammatory status. The impact of HP diet on intestinal mucosa in terms of beneficial or deleterious effects is discussed. © 2014 Elsevier Inc. All rights reserved. Keywords: High-protein diet; Intestine; Goblet cells; Mucus; Basal gut inflammatory status

1. Introduction An increase in dietary protein intake increases the quantity of unabsorbed proteins, peptides and amino acids reaching the large intestine through the ileocecal junction [1]. In the large intestine, undigested and partially digested dietary proteins and peptides from dietary and endogenous origins [2,3] are degraded by endogenous and microbiota proteases/peptidases in peptides and free amino acids, these latter being precursors of numerous bacterial metabolites including ammonia, hydrogen sulfide, amines, short-chain and branched-chain fatty acids, indoles, phenols and organic acids [4–7]. These changes in the small and large intestinal luminal composition may affect mucosal physiology and immunity [8]. In the rat model, consumption of a high amount of protein markedly modifies the colonic luminal content including increased water, ammonia and short-chain fatty acids (SCFAs) content, higher protease activities in both the small and large intestines [9] and modification of ☆

Grants: This research was supported by AgroParisTech and INRA. Corresponding author. AgroParisTech-INRA-CRNH-IdF, UMR 914 Nutrition Physiology and Ingestive Behavior, 16 rue Claude Bernard, F-75005 Paris, France. ⁎

http://dx.doi.org/10.1016/j.jnutbio.2014.09.007 0955-2863/© 2014 Elsevier Inc. All rights reserved.

the microbiota composition and diversity [10]. Numerous metabolites, depending on their luminal concentrations, are known to exert either beneficial or deleterious effects on the intestinal colonic mucosa [11]. For instance, some bacterial metabolites whose production is increased after high-protein (HP) diet ingestion, such as ammonia or sulfide, inhibit oxygen consumption in colonocytes [12,13]. Conversely, other metabolites like SCFAs which are also increased in the colonic luminal content after HP diet ingestion are fuels for the colonocytes. Owing to the role of intestinal epithelial cells (IECs) as luminal signal integrator, and then as central regulators of barrier function and mucosal immune homeostasis [14], we hypothesized that changes in the luminal environment of the IEC induced by HP ingestion might lead to modifications of epithelial renewal and/or functions. Numerous evidences have underscored the central role of IECs in relation with immune-cell functions (for review, see Ref. [15]). We have previously shown that HP diet ingestion alters the distal colonocyte morphology by markedly reducing the height of brush-border membranes without affecting mechanical barrier as judged by the tight junction structure [9]. However, the consequences of luminal composition changes together with the higher dietary antigenic load after HP diet consumption remain to be determined on parameters related to basal gut mucosal immune status. To that aim, rats were fed for 15 days with a normal protein

2

A. Lan et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

2.6. Analysis of mucin expression by immunohistochemistry

(NP; 14%) or HP (53%) isocaloric diet, and then several components involve in mucosal protection and immune homeostasis (mucoussecreting cell activity, luminal IgA concentrations, expression of antimicrobial peptides, TLRs and cytokines; proportion of lymphocyte populations and basal inflammatory status) were assessed in both the ileum and colon mucosa in order to compare the effects of such a dietary change on two intestinal segments with different transit time and physiological functions.

Mucin-2 expression was evaluated with immunohistochemistry experiments carried out on 6-μm transversal ileal sections using the Leica Novolink Polymer Detection Systems kit (Leica Biosystems, Newcastle, UK). The antigens were unmasked by heating the sections for 20 min in sodium citrate buffer 10 mM (pH 6.0) supplemented with 0.05% Tween 20 (vol/vol) at 95°C–100°C. Expression of MUC2 was detected using anti-MUC2 antibody (ab134119, 1:250; Abcam, Cambridge, UK) in Trisbuffered saline 50 mM containing 1% bovine serum albumin. Negative control was performed in the same conditions with omission of the primary antibody anti-MUC2.

2. Methods and materials

2.7. Quantitation of intestinal gene expression by real-time polymerase chain reaction

2.1. Animals and diets

Intestinal tissues were lysed in Trizol (Invitrogen, Life Technologies, Illkirch, France), and total RNA was extracted and cleaned up with the RNeasy kit and RNasefree DNase I (Qiagen, Courtaboeuf, France) digestion based on the manufacturer's protocol. For isolated IECs, total RNA was directly extracted with the RNeasy kit and RNase-free DNase I (Qiagen) digestion. Quantitative real-time polymerase chain reaction (qRT-PCR) was performed with rat-specific primers (Eurogentec, Angers, France) designed based on published sequences of the above-mentioned target genes (sequences available on demand) using Oligo Explorer 1.1.0 software. Marker genes of chemical barrier function (Muc2, Muc3, Muc4, Defb1, Reg3g, Defa5, Defcr4, Spla2), of mechanical barrier function (Cldn2, Ocln, Tjp1), of immune receptors (Tlr2, Tlr4, Tlr5) and intestinal cytokines (Il-1β, Il-6, Il-10, Il-13, Il-17f, Il-22, Ifnγ, Tgfβ, Tnfα) were analyzed. After cDNA synthesis from mRNA using High Capacity cDNA Reverse Transcription Kit (Applied Biosystems), real-time PCR was performed subsequently on cDNA using the power SYBR Green PCR master mix and StepOne Real-Time PCR system (Applied Biosystems, Life Technologies). cDNA samples were assayed in triplicate, and gene expression levels for each sample were normalized relative to Rlp13a with 2−ΔΔCt calculation.

Male Wistar Lewis rats (Harlan, Gannat, France) weighing 150 g (5–6 weeks) were fed for 1 week with a standard rodent diet containing 16% protein by weight. The animals then received for 15 days either a normoproteic (140 g/kg whole milk protein) or a hyperproteic/hypoglucidic (530 g/kg whole milk protein) isocellulose isocaloric diet (14.6 kJ/g in both diets) [9] and water ad libitum. After 15 days of NP or HP diet, the animals were anesthetized with pentobarbital sodium (40 mg/kg body wt) for biological sample collection. All aspects of the present protocol are in accordance with the guidelines of the French Committee for Animal Care and the European convention of vertebrate animals used for experimentation, under European council directive, and received written agreement from the local animal ethical committee (COMETHEA at Jouy-en-Josas, France, No. 11/042 and No. 12/090). 2.2. Tissue collection The colon and small intestine were recovered and the luminal contents were removed by expulsion and stored at −80°C until analysis. After rinsing with PBS, small intestine was resected and divided into six segments: the first, third and sixth segments corresponded to duodenum, jejunum and ileum, respectively. After rinsing, the colon was divided into three segments, namely, proximal, mid and distal segments. Samples in the middle of each intestinal segment were immediately frozen in liquid nitrogen and stored at −80°C for myeloperoxidase (MPO) assay, RNA or protein analysis. The latest 1-cm intestinal segments (ileum and colon) were fixed in 4% buffered formaldehyde for histological analysis. In some experiments, isolation of colon and small intestine epithelial cells was prepared from the whole colon or from the whole small intestine as previously described [16]. 2.3. SCFA analysis in intestinal contents SCFAs (acetate, propionate, butyrate, valerate and caproate) and branched-chain fatty acids (isobutyrate and isovalerate) were measured by a modification of the method of [17] using gas chromatography and a capillary column (30m, 0.32 mm ID, RestekRtx 502.2; Restek, Lisses, France). The amount of SCFAs was determined with reference to internal standards. 2.4. Total IgA quantification in luminal intestinal samples by ELISA Frozen luminal intestinal samples diluted 10-fold (wt/vol) in PBS containing protease inhibitor cocktail (Roche, Boulogne-Billancourt, France) were homogenized and centrifuged 10 min at 16,000×g to collect supernatants. Flat bottom 96-well plates coated with 100 μl/well of purified anti-rat IgA (2 μg/ml; BD Biosciences, Le Pont de Claix, France) were incubated after blocking, with 100 μl of either intestinal supernatant or standard rat purified IgA (BD Biosciences) for 1 h 30 min at 37°C. After washing, fixed antibodies were detected with 100 μl of biotinylated anti-rat IgA (2 μg/ml; BD Biosciences) for 1 h 30 min at 37°C. After a 30-min incubation at 37°C with extravidin–peroxidase (Sigma), the reaction was revealed with 100 μl/well of extravidin–peroxidase substrate (O-phenylenediamine; Sigma) and H2O2 30%, and stopped by the addition of H2SO4. Absorbance was read at 490 nm. 2.5. Histological preparation and examination of ileal and colonic sections After an overnight fixation, 0.5-cm section of ileum and colon was embedded in paraffin wax. Serial transversal 6-μm-thick sections were stained with hematoxylin and eosin or with periodic acid-Schiff (PAS) or Alcian blue counterstained with hematoxylin in order to visualize both acidic and neutral or acidic mucins, respectively. Histology was performed on the Cochin HistIM Facility, and histological analysis of ileal and colonic sections was performed using the viewer and the image analysis software Calopix TRIBVN (Chatillon, France). Colonic crypt length and number of PAS-positive cells per crypt were determined by analysis of 30 to 40 well-oriented crypts per animal. Number of PAS positive cells on colonic epithelial plateau was determined by the analysis of a mean epithelial surface of 36,000±3774 μm2 per animal (n=9/group). Number of PAS positive cells in ileum was performed by automating counting of total and PAS-positive cells in selected area of ileal mucosa (mean surface analyzed per animal: 3.394±0.423mm2 per animal, n=9–10/group). Villus–crypt length was determined by measuring 54.6±4.7 ileal crypt–villus unit per animal (n=5/group).

2.8. Cytokine concentration analysis in small intestine tissue lysates Concentration of TGFβ1, IL-10 and IL-13 were measured in total protein lysate from an ileum segment by Luminex technology using Bio-Plex kits (Bio-Rad, Marnes-LaCoquette, France) and performed according to the manufacturer instructions. Total protein lysate was diluted to 900 μg/ml in Bio-Plex Cell Lysis buffer and run in duplicate after a 1:2 dilution in sample diluent. Median fluorescence intensities were collected on a Luminex-200 instrument using Bio-Plex Manager software version 4.1 (Bio-Rad). 2.9. Lymphocyte population analysis in gut-associated lymphoid tissue by flow cytometry analysis 2.9.1. Cell isolation Peyer patches (PPs) from the whole small intestine were carefully excised from the serosal side of the intestine. The tissues were gently dilacerated in RPMI and incubated with collagenase type I (Gibco, Life Technologies) for 50 min at 37°C. Digested PP and mesenteric lymph nodes (MLNs) were passed through a 70-μm cell strainer filter (BD Biosciences), centrifuged at 400×g for 5 min at 4°C, and the pellets were washed thrice with RPMI containing heat-inactivated calf serum and antibiotics. After trypan blue exclusion test (Sigma-Aldrich), cell counts were adjusted to 1×106cells/ml using PBS BSA1%, and 1 ml of this cell suspension was used for flow cytometry analysis. 2.9.2. Flow cytometry analysis To analyze the percentage of T and B and NK cells in PP and MLN, three-color immunofluorescence staining was performed and the intensity of stained cells was analyzed by flow cytometry (FACScalibur, BD). To determine T lymphocytes (CD3+ cells), T lymphocytes CD4+ (CD3+ CD4+ cells) and T lymphocytes CD8+ (CD3+ CD8a+), 1×106 isolated cells were incubated with antirat lymphocyte T cocktail (anti-CD8a-FITC, anti-CD4-PE and anti-CD3-AP; BD Biosciences); 1×106 isolated cells were also incubated with CD45RA-FITC, CD161a-PE and CD3-APC (BD Biosciences) to measure proportion of lymphocyte populations (NK cells: CD161a+CD3− cells; B lymphocytes: CD45RA+CD3− cells and T lymphocytes: CD3+ cells). The incubation (30 min in the dark) was followed by a wash in PBS BSA1%. Then, the cells were resuspended in PBS BSA 1% prior to analysis by flow cytometry. Lymphocytes were gated on the basis of their characteristic light scatter. Fluorescence intensity was depicted on a three-decade logarithmic scale and in single-parameter analysis as histograms. Absolute numbers of lymphocytes were calculated according to lymphocyte subset. 2.10. Measurement of MPO activity MPO activity was determined as an indicator of neutrophil infiltration using an odianasidinedihydrochloride assay. Small intestine (duodenum, jejunum and ileum) and colon (proximal, mid, distal) samples were homogenized over ice using an ultrathurax (3×10 s) in 0.5% hexadecyl-trimethylammonium bromide (Sigma-Aldrich, Saint-Quentin Fallavier, France) in potassium phosphate buffer (pH 6.0). Then samples were centrifuged (10,000×g, 30 min, 4°C) and the supernatant was kept on ice. The pellet was frozen, thawed and homogenized for 10 s before centrifugation (10,000×g, 30 min, 4°C); this cycle was repeated two times. Final supernatant (50 μl) was mixed

A. Lan et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

Table 1 Effect of HP diet on luminal amount of SCFAs SCFA (μmol/total luminal content) Acetate Propionate Butyrate

Small intestine NP

11.12±2.25 ⁎

5.65±0.55 ND 0.06±0.06

polyclonal antibody to Occludin (Abcam ab31721, 1/250) or with mouse monoclonal to E-Cadherin (Abcam ab76055, 1/1000) diluted in blocking solution. After washes, blots were incubated for 2 h at room temperature with an antirabbit or antimouse HRPlinked secondary antibody (1/5000; Jackson Immuno Research Laboratories, West Grove, PA, USA) or goat anti-actin-HRP (1/1000, C-11; Santa Cruz Biotechnologies, Heidelberg, Germany) in blocking solution. After additional washes, the specific bands were detected by chemiluminescence using ECL system according to the manufacturer instructions (Pierce Biotechnology, Courtabœuf, France). Acquisition was performed using the FluorChem FC2 device and the AlphaView software (Cell Biosciences, Santa Clara, CA, USA), and specific band intensity was measured using β-actin intensity to normalize data on three independent experiments.

Colon HP ND 0.60±0.36

NP

HP

6.22±1.33 1.33±0.28 0.57±0.10

11.06±1.71 ⁎ 3.29±0.41 ⁎⁎ 1.38±0.27 ⁎

3

SCFAs were measured in luminal contents of small intestine (n=13–15 animals per group) and colon (n=8 animals per group) after 15 days of normoproteic (NP) and hyperproteic (HP) diet ingestion. Results are expressed as mean±S.E.M. ND: not detectable. ⁎ Pb.05. ⁎⁎ Pb.01.

2.12. Statistical analysis Statistical analysis was performed using Prism software (Prism 6.0; GraphPad Software, La Jolla, CA, USA). Unpaired t test was used to compare data with Welch's correction when variances were unequal. Differences were considered significant if Pb.05. Data are presented as means±S.E.M. together with the number of animals per experiments.

with 250 μl of potassium phosphate buffer containing 0.48 mg/ml of o-dianasidine dihydrochloride (Sigma-Aldrich) and 0.48 mM of hydrogen peroxide. Absorbance was read at 450 nm for 10 min. In parallel, protein assay on 10 μl of supernatant was performed using Bio-Rad DC Protein Assay (Bio-Rad). Activity was expressed as the difference after a 10-min reaction time between the absorbance and the baseline absorbance per milliliter of supernatant and per milligram of proteins.

3. Results

2.11. Expression analysis of tight junction and adherent junction proteins by Western blotting

3.1. Luminal acetate content increases in the small intestine after HP diet consumption but not other SCFAs nor organic acids

Total proteins (40 μg) from scrapped ileal mucosa were loaded onto 4%–12% Criterion XT (Bio-Rad) and run using 1× MOPS buffer. After transfer onto nitrocellulose membrane and incubation in blocking solution (TBS pH 7.5, 0.05% Tween 20 and 5% [wt/vol] nonfat dry milk), membranes were incubated overnight (4°C) with rabbit

HP diet feeding induced a twofold higher acetate content in the small intestine of HP animals (P=.031) when compared with NP animals (Table 1). The amount of butyrate in the small intestine

Colon NP

Ileum HP

a

b

e

NP

HP

c

f

B 300

d

g

Colonic crypt length (µm)

A

h

200

100

j

k

D 200

*

150 100 50 0 NP

HP

Percent of PAS-positive cells on colonic epithelial surface (% of total epithelium surface)

Number of PAS-positive cells per mm of colonic crypt depth

C

l

NP

HP

NP

HP

15000

10000

5000

0

E 30

20

*

10

0 NP

HP

Percent of ileal PAS-positive area (% of total ileal mucosa surface)

i

Number of nuclei per mm2 of ileal mucosa

0

25

*

20 15 10 5 0 NP

HP

Fig. 1. Effect of HP diet on intestinal morphology and mucus-producing cells. (A) Histological examination of colonic and ileal sections after a 15-day normoproteic (NP) or HP diet stained with hematoxylin–eosin (a–d) and PAS (e-j) or Blue Alcian (k–l), these latter dyes being used to visualize mucus producing cells (k and l: enlargement of ileal sections g and h, magnification: ×200). (B) Colonic crypt length and total number of nuclei per mm2 of ileal mucosa after a 15-day NP (white) or HP (black) diet. (C) Number of PAS-positive cells per length of colonic crypt. (D) Percentage of goblet cells on colonic epithelial surface. (E) Percentage of PAS-positive area normalized to total ileal mucosa surface. Images are representative of 10 rats per group. Results are expressed as mean±S.E.M. (n=10 animals per group). *Pb.05.

4

A. Lan et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

luminal content recovered from both HP and NP rats was close to the limit of detection, not exceeding 0.60±0.36 μmol/total content. The weight of the ileal luminal content was significantly higher in HP animals compared to NP group averaging 1.16±0.01 and 0.69±0.07 g, respectively (n=13–15 animals in each group, P=.0058). Thus, the acetate concentration in the ileal luminal content of NP and HP animals averaged 8.2 and 9.6 mM, respectively. Other SCFAs (propionate, valerate and caproate), branched-chain fatty acids (isobutyrate, isovalerate) or organic acids (lactate, succinate and ethanol) were below the limit of detection in the small intestinal luminal contents of both groups (data not shown). In the whole colonic content, the amounts of acetate, propionate and butyrate were significantly higher being approximately doubled in the HP group when compared with the NP group (Pb.05).

markedly lower in HP rats (8.74%±0.64% of total surface epithelial cells) than in NP rats (20.69%±4.39% of total surface epithelial cells, P=.026). In addition, mucus-producing cell hyperplasia was observed in ileum after HP diet consumption when compared with NP rats (Fig. 1A g–h and E). Indeed, the PAS-positive area averaged 18.47%± 1.88% of the ileal mucosa surface in HP group, while representing 13.62%±0.91% in the NP group (P=.037). Alcian blue staining of ileal sections (and also staining of colon sections; data not shown) confirmed higher mucin mucosal content in HP-group (Fig. 1A k–l) and revealed that both acidic and neutral mucins were more abundant in response to HP diet than to NP diet feeding, owing to higher number of goblet cells (4017±368.3 mucus-positive cells per mm2 of ileal mucosa in HP group) compared to 2920±208.8 in the NP group (P=.018). Gene expression corresponding to Muc2, the main gel-forming secretory mucin in intestine, increased in the ileal mucosa of HP rats when compared to NP rats (P=.015) but not in the colonic mucosa (Fig. 2A). This increased expression of MUC2 in the ileum was also observed at the protein level by immunohistochemistry (Fig. 2B). Expression of Muc3 gene was 3-fold higher in the colonic mucosa from HP rats (P=.032) and 3.5-fold more expressed in the ileal mucosa from HP rats (P=.003) than in NP rats (Fig. 2A). In contrast, Muc4, another membrane-associated mucin, remained unchanged in HP rats when compared to NP rats.

3.2. HP diet consumption induces goblet cell hyperplasia and higher intestinal mucus content in ileum, and modifies goblet cell distribution in colonic epithelium HP diet has no effect on the overall morphology of the colonic and ileal mucosa (Fig. 1A). Colonic crypt length was not different between NP and HP rat groups (234.1±9.7 μm for NP rats compared to 241.4± 17.1 μm for HP rats; Fig. 1B). In the ileum, no difference of villus–crypt length was observed between the two groups (NP group: 376.9± 12.0 μm; HP group: 395.4±32.4 μm, n=10 rats per group; data not shown). Moreover, the total number of nuclei per mm2 of ileal mucosa was similar between the two groups of rats (12,629±273 vs. 11,859±243 cells/mm2 of ileal surface in the NP and HP groups, respectively; Fig. 1B). When rats were fed with the HP diet, hyperplasia of intestinal goblet cells was observed. Indeed, the number of PAS positive cells, i.e., goblet cells along the crypt depth (Fig. 1A e–f and C) was increased in HP-colon crypts compared to NPcolon crypts (155.3±4.8 goblet cells/mm of crypt in HP rats vs. 135.3 ±5.9 in NP rats, P=.018). In contrast, the relative number of goblet cells in the colonic surface epithelium (Fig. 1A i–j and D) was

A

To further assess the effect of HP diet on chemical barrier function, IgA concentrations were measured in the luminal contents of the small intestine, cecum, colon and in feces. As indicated in Table 2, no significant difference was found for the luminal IgA concentrations in any intestinal anatomical part and in feces recovered from either NP or HP rats. Furthermore, no change in gene expression of antimicrobial peptides such as the β-defensin 1 (Defb1), the regenerating islet-

Muc2 2.0

Fold change relative to normalised control

3.3. HP diet consumption has no effect on IgA luminal concentrations nor antimicrobial peptide gene expression

Colon

Muc3 Ileum

*

Colon

8

Muc4 Ileum

*

2.0

1.5

6

1.0

4

1.0

0.5

2

0.5

0.0

**

0 NP

B

HP

NP

Ileum

1.5

0.0

HP

NP

NP rat ileum

a

Colon

HP

NP

HP

NP

HP rat ileum

b

HP

NP

HP

Negative control

c

Fig. 2. Effect of HP diet on mucin expression. (A) Gene expression of Muc2, Muc3 and Muc4 was analyzed by qRT-PCR after a 15-day normoproteic (NP; white) or HP (black) diet in colonic and ileal homogenates. RNA expression relative to NP-fed animals is expressed as mean ΔΔCt±S.E.M. (n=8 animals per group). *Pb.05; **Pb.01. (B) Staining of MUC-2 by immunohistochemistry on rat ileal section after a 15-day normoproteic (NP; a) or HP (b) diet (magnification: ×100). Images are representative of 6 rats per group (a, b); c: negative control.

A. Lan et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

Table 2 Effect of HP diet on luminal IgA concentrations

Table 4 Effect of the HP diet on intestinal cytokine gene expression

IgA (μg/ml luminal content)

Small intestine Caecum Colon Feces

Ileum

NP

HP

4.71±1.19 32.20±7.61 99.20±35.30 138.69±62.37

2.93±0.96 20.88±4.67 76.90±19.35 163.7±51.30

Secretory IgA concentrations were measured in intestinal contents after 15 days of normoproteic (NP) and hyperproteic (HP) diet ingestion. Results are expressed as mean±S.E.M. (n=10 animals per group).

derived3 gamma (Reg3g) or Paneth cell-specific peptides (such as the α-defensin 5, Defa5; cryptdin-4, Defcr4 and the secretory phospholipase 2, Spla2) was measured in mucosa of HP rats (data not shown; n=10 rats for both groups). 3.4. Gene expression of ileal Tlr4 and cytokines, and Th subpopulation proportion are modified after HP diet consumption To address whether expression of the main cell-surface toll-like receptors (TLRs) in intestine were altered by the HP diet, we measured Tlr2, Tlr4 and Tlr5 expression in epithelial cells isolated from the whole small intestine (enterocytes) or from the entire colon (colonocytes) obtained from NP- and HP-fed rats (Table 3). In enterocytes recovered from HP rats, Tlr4 mRNA level was significantly lower (P=.043) than in NP rats, while expression of other TLRs remained similar in both experimental groups. Regarding TLR gene expression in colonocytes recovered from NP and HP rats, no significant difference was measured when comparing these 2 groups of animals. Furthermore, expression of cytokines by isolated enterocytes and colonocytes was similar in both groups of rats. Indeed, no change of Il-10, Tgfβ, Il-1β and Tnfα gene expression measured by qRT-PCR was recorded between NP and HP rats for both types of IECs (data not shown). In whole colonic mucosa homogenate, no change of cytokine gene expression was observed except for a slight but significant reduction of Il-6 mRNA expression (Table 4). Conversely in ileum mucosa, HP diet markedly down-regulated gene expression of the major cytokines, i.e., Il-1β, Il-4, Il-10, Il-17f, Ifnγ and Tnfα (twofold to fivefold decrease, Pb.05) and up-regulated expression of the regulatory cytokine Tgf-β in the total ileal mucosa homogenate of HP rats compared to NP rats (Table 4). Using the total protein lysate from ileal tissue homogenate, the concentration of Tgf-β1 originating from HP animals was significantly increased compared to NP animals (206.4± 9.0 pg/ml of total protein lysate vs. 178.7±7.6 in HP and NP groups, respectively; P=.041; Fig. 3). Conversely, ileal concentration of IL-10 was slightly lower in HP animals than in NP (168.7±23.6 pg/ml of total protein lysate vs. 197.5±37.6 pg/ml in the HP and NP groups, respectively). In addition, Il-13 gene was threefold more expressed in

Table 3 Effect of the HP diet on TLR gene expression of IECs. Enterocytes

Tlr2 Tlr4 Tlr5

5

Colonocytes

NP

HP

NP

HP

0.999±0.070 1.000±0.052 0.999±0.065

0.794±0.143 0.770±0.084 ⁎ 0.993±0.208

1.018±0.084 1.030±0.070 1.033±0.080

1.167±0.073 1.330±0.143 0.958±0.160

Gene expression of Tlr-2, Tlr-4 and Tlr-5 were analyzed by qRT-PCR after a 15-day normoproteic (NP) or HP diet in isolated small intestine and colonic epithelial cells. RNA expression relative to NP-fed animals is expressed as mean ΔΔCt±S.E.M. (n=6 animals per group). ⁎ P=.043.

Colon

NP

HP

Mean S.E.M. Mean Tgf-β Il-1β Il-4 Il-6 Il-10 Il-13 Il-17f Il-22 Ifnγ Tnfα

1.076 0.947 1.477 1.120 1.104 1.100 0.898 1.688 1.094 0.998

0.123 0.113 0.451 0.248 0.198 0.162 0.218 0.617 0.203 0.25

P NP HP value S.E.M. Mean S.E.M. Mean

1.408 ⁎ 0.088 0.573 ⁎ 0.088 0.455 ⁎ 0.172 0.982 0.227 0.247 ⁎ 0.048 3.061 ⁎ 0.662 0.302 ⁎ 0.111 1.004 0.339 0.218 ⁎ 0.069 0.476 ⁎ 0.177

.046 .020 .048 NS .010 .013 .043 NS .025 .033

1.155 1.074 1.062 1.016 1.021 1.487 ND 1.092 0.813 1.1

0.161 0.19 0.088 0.092 0.132 0.677 0.375 0.09 0.232

0.958 1.077 1.224 0.726 ⁎ 0.989 1.291 ND 0.561 0.654 1.068

S.E.M.

P value

0.072 0.161 0.403 0.086 0.213 0.748

NS NS NS .049 NS NS

0.128 0.187 0.194

NS NS NS

Cytokine mRNA expression in ileal and colonic homogenates was analyzed by qRT-PCR after 15 days of normoproteic (NP) and hyperproteic (HP) diet ingestion. RNA expression relative to NP-fed animals is expressed as mean ΔΔCt±S.E.M. (n=6 animals per group). ND: not detectable. NS: PN.05. ⁎ Pb.05.

ileal mucosa of HP rats than of NP rats (P=.013). These latter finding could not been confirmed at the protein level since the IL-13 concentration in ileal total protein lysate was closed to the limit of detection (data not shown). Flow cytometry analysis of lymphocyte populations (T, B and NK cells) and T-cell subpopulations in gutassociated lymphoid tissue revealed a reduction of T-cell trafficking in MLNs of HP rats (Table 5). Indeed, the percentage of Th subpopulation (CD3+CD4+) was significantly lower in MLN of HP rats (9.2% less cells of CD3+CD4+ total population of the NP group, P=.016), while no other significant change of the percentages of lymphocytes was observed between groups.

3.5. HP diet consumption reduces basal MPO activity in ileum To assess the effect of HP diet on the basal intestinal inflammatory status, MPO activity was measured in duodenum, jejunum and ileum, and in proximal, mid and distal colon (Fig. 4). MPO activity was reduced in the ileum of HP rats (2.68±0.31 AU MPO/mg protein) when compared with the activity measured in NP rats (4.85±0.82 AU MPO/mg proteins, P=.037), whereas MPO activity was similar in all other intestinal anatomical parts when comparing NP and HP animals. This lower MPO activity in the ileum of HP rats was not associated with any measurable change in the gene expression of tight junction proteins involved in the mechanical intestinal barrier function. Indeed, gene expression corresponding to the tight-junction proteins Claudin-2 (Cldn2), Occludin (Ocln) and ZO-1 (Tjp1) was similar between HP and NP rats when measured in isolated ileal and colonic epithelial cells (data not shown). Lastly, the Western blot analysis of Occludin and the adherent junction protein E-Cadherin found no measurable difference in the specific band intensity average of each protein measured in the scrapped ileal lysates from NP and HP rats, with important variability in each group of animals (data not shown, n=6 rats for both groups).

4. Discussion The present study indicates that the ingestion of an HP diet during 15 days modifies parameters related to intestinal barrier function and mucosal immunity, both in the distal part of the small intestine and large intestine, but with different features, when compared with rats receiving an NP diet.

6

A. Lan et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

IL-10

pg/mL of total protein lysate

TGFβ1 250

*

200

250

200

150

150

100

100

50

50

0

0 NP

HP

NP

HP

Fig. 3. Effect of the HP diet on regulatory cytokines in ileal tissue homogenates. TGFβ1 and IL-10 concentrations were quantified by Bio-Plex assay after a 15-day normoproteic (NP; white) or HP (black) diet in total ileal protein lysate. Cytokine concentrations are expressed as mean pg/ml of total protein lysate±S.E.M. (n=6 animals per group). *Pb.05.

The present results indicate that increasing the dietary supply of protein can affect goblet cell distribution and possibly their function mainly in the ileum and, to a lesser extent, in the colonic epithelium. The intestinal mucus layer forms a compact mesh-like network of viscous and permeable gel covering the gastrointestinal mucosa that lubricates the epithelial surface and protects it from potential luminal insults such as endogenous and exogenous deleterious compounds, but also participates in the transport of nutrients [18]. In the present study, HP diet when compared to NP diet induced both hyperplasia of goblet mucin-producing cells and up-regulation of transcriptional and translational levels of mucin expression in the ileum, suggesting an increase of mucus synthesis in ileum that would reinforce mucous barrier. Indeed, in the ileal mucosa, more goblet cells were enumerated in HP rats as evidenced by a higher number of PASpositive cells and increased PAS-positive area. Accordingly, Muc2 expression at both gene and protein levels was increased, as well as Muc3 gene expression. Conversely, in the colonic surface epithelium the relative number of goblet cells was decreased more than twofold after HP diet but a slight increase of the number of goblet cells in the colonic crypts and a marked increase of the expression of the Muc3 gene in the colonic mucosa were observed. Although the mechanisms involved in such modification of goblet cell distribution in the colonic epithelium remained to be determined, this modification may be related to alteration of the differentiation process of colonic epithelial cells when moving from the crypt to the epithelial surface. This is coherent with previous observation made in our laboratory which showed that the absorptive colonocytes in the surface epithelium in HP animals are characterized by reduced height of brush-border membrane when compared with colonocytes in NP animals [9].

Table 5 Effect of the HP diet on intestinal lymphocyte populations Group

Number of PP LB (%) NK (%) LT (%) CD3+CD4+ (%) CD3+ CD8+ (%)

PP

MLN

NP

HP

NP

HP

20.25±2.06 48.24±3.32 1.98±0.11 16.68±0.88 12.64±0.81 3.60±0.46

21.25±3.07 44.96±2.38 2.04±0.18 16.12±0.75 12.62±0.56 3.20±0.38

– 19.62±2.41 1.46±0.16 66.24±3.50 50.32±2.35 16.30±1.85

– 21.26±1.70 1.99±0.21 60.62±2.48 45.88±1.81 ⁎ 15.00±1.31

Lymphocyte B (LB), lymphocyte T (LT) and natural killer (NK) populations (CD3/ CD45RA/CD161a staining) and T-cell subpopulations (CD3/CD4/CD8staining) were analyzed by flow cytometry in PPs and MLNs after 15 days of normoproteic (NP) and hyperproteic (HP) diet ingestion and were calculated as percentages of total cells. Results are expressed as mean±S.E.M. (n=5 animals in both experimental groups). ⁎ P=.016.

Regulation of mucin expression is triggered by many factors [19], among which a larger amount of amino acids provided by the HP diet, as precursors for mucin synthesis is likely to play a major role. Mucins are large glycoproteins with protein backbone structures particularly rich in threonine, serine and proline, which are linked to a wide variety of O-linked oligosaccharide side chains. Regarding threonine, this essential amino acid is known to be critical for mucin production [20] and gut function [21,22] and is extensively used by the intestine for mucosal protein synthesis [23]. Interestingly, the results showed a threefold increase of Il-13 gene expression in the ileum of HP rats when compared with NP rats, this Th2 cytokine having been reported to promote the differentiation and hyperplasia of goblet cells [24]. It is tempting to propose (but would require further experiments with more sensitive method of IL-13 assay) that Il-13 would be responsible for Muc2 gene expression increase and goblet cell hyperplasia in the terminal part of the small intestine. This result raises the question of which intestinal mucosa cells produce this cytokine and then requires additional flow cytometry analysis to precisely identify Il-13 producing cells. HP diet feeding compared to NP diet down-regulated a series of parameters related to mucosal immunity in the ileum with a specific decreased Tlr4 gene expression among other TLRs in isolated enterocytes; a down-regulation of the expression of genes related to mucosal cytokines involved in innate and adaptive T-cell immune responses (Il-1β, Il-4, Il-10, Il-17f, Ifnγ and Tnfα), as well as a reduction of T-helper cell proportion in MLN. Lastly, we observed an upregulation of Tgfβ expression at both gene and protein levels and a reduction of the basal inflammatory status. Although these results would require further development regarding the functional consequences on intestinal physiology, increased mucus content may decrease luminal antigenic stimulation of the intestinal immune cells, which would be reflected by a lower expression of intestinal cytokines. In addition, the modifications observed may be related to previous results showing that TLR signaling in the intestine is involved in epithelial cell proliferation, IgA production, and maintenance of tight junctions [25,26]. Regarding these latter parameters, our data indicate, however, that HP diet consumption did not modify IgA luminal concentration in the intestine, and, according to the data obtained in this study and previous studies [9], there is no evidence indicating that HP diet would reinforce the mechanical intestinal barrier function when compared with the control NP diet situation. However, since the HP diet increases the weight of the ileal luminal content (this study) and the weight of the colonic luminal content [9], it can be deduced that the amount of luminal IgA in the intestinal lumen is increased by the HP diet both in the distal small intestine and in the colon.

A. Lan et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

B 15

10

5

*

0

Duodenum

Jejunum

Ileum

AU MPO activity/mg of protein

AU MPO activity/mg of protein

A

7

2.5 2.0 1.5 1.0 0.5 0.0

Proximal

Mid

Distal

Fig. 4. Effect of the HP diet on basal intestinal inflammatory status. MPO activity was measured in small intestine (A) and colon (B) segments after a 15-day normoproteic (NP, white bar) or HP (black bar) diet. Results are expressed as mean±S.E.M. (n=8 animals per group). *Pb.05.

The present study suggests an overall down-regulation of T-cell responses in ileal mucosa of HP rats evidenced by the reduced gene expression of the most major cytokines involved in Th1, Th2, Th17 and Treg responses, which might explain the reduction of CD3+CD4+ cells in MLN and absence of secretory IgA increase [27], although specific characterization of immune cells involved in intestinal mucosal immune responses and susceptible to be affected by HP ingestion remained to be done in further studies. These observations can be related to the increased gene and protein expression of the cytokine TGF-β1 in ileal mucosa, which apparently had no epithelial origin as its gene expression remained unchanged when comparing enterocytes isolated from both rat groups. This regulatory cytokine is known to promote the development of a tolerogenic environment that characterizes intestinal immune responses [28]. Furthermore, reduction of gene expression related to proinflammatory cytokines might be directly related to MUC2 expression since interestingly it has been recently shown that this mucin induces attenuation of proinflammatory cytokine production by dendritic cells [29]. In the colon, the present results showed after HP diet consumption a slight reduction of the proinflammatory Il-6 gene expression in colonic mucosa that was concomitant with an increased amount of SCFAs in the luminal colonic content. In the ileum, we found a marked increase of the amount of acetate in the luminal content recovered from HP animals. However, and as previously found in the rat colon [10], the increased of the luminal bulk content in the small intestine of HP rats led to a similar acetate concentration in the intestinal lumen of NP and HP rats, thus allowing homeostasis of this parameter. Incidentally, the twofold increase of the amount of acetate in the small intestine luminal content after HP diet ingestion was likely due, in part at least, to an increased conversion by the small intestine microbiota of the casein-derived amino acids into this SCFA. Indeed, several essential (threonine and lysine) and nonessential amino acids (alanine, aspartate, glutamate and glycine) are precursors for acetate synthesis by the microbiota [11]. This is compatible with the idea that a part of luminal amino acids in the small intestine is lost for intestinal utilization, absorption and utilization in the liver and peripheral organs since they are converted to acetate by the intestinal microbiota [30]. In ileum of HP rats, the question of a relationship between the acetate luminal content, the higher mucin production and the modifications of the ileal microbiota composition and/or activities remains open and will require further development out of the scope of the present study. Regarding this latter aspect, we have recently shown that HP diet consumption causes major changes in colonic microbiota composition, such as a significant decrease concentration

of the acetate consumer Faecalibacterium prausnitzii [10], which has been shown to diminish the effects of an acetate producer bacterium on goblet cells and mucins in gnobiotic rats [31]. Although in our previous work, HP diet consumption was shown to induce major changes in the luminal environment and ultrastructural modifications of the absorptive colonocytes in the distal colon [9], it did not lead in the present study to any improper immune response on the basis of the parameters measured, but led notably to modification of the distribution of goblet cells between the crypts and the epithelial surface. Whether the decreased relative amount of goblet cells at the surface epithelium changes the protection by mucus of the differentiated absorptive colonic epithelial cells remains to be explored. Collectively, the present study demonstrates that HP diet consumption led to an increase of mucus-producing cell activity and a reduction of several parameters related to the intestinal mucosal immunity mainly in the distal part of the small intestine. It can be proposed that the impact of HP diet is presumably either direct (e.g., via providing more amino acids for mucus synthesis) and/or indirect (e.g., via intestinal adaptation to changes in the luminal environment). These data raise the question on the one hand of the effects of these modifications in physiological conditions and, on the other hand, of the impact of HP diet intake on moderate, acute or severe intestinal inflammation. Induction and maintenance of intestinal immune homeostasis in a healthy gastrointestinal tract relies on complex and highly regulated immune responses by different mucosal cell types. Gastrointestinal immunity is indeed finely balanced between inflammatory and tolerogenic responses by complex regulatory pathways in order to not overreact to inoffensive luminal antigens and to induce a protective immune response against pathogens when necessary. It is well known that in susceptible individuals, defects in the delicate equilibrium in favor of inflammatory responses can disrupt the homeostatic mechanisms and lead to intestinal inflammation. However, little is known about the consequences of a diet-induced down-regulation of intestinal cytokine expression and of basal inflammatory status on long-term gastrointestinal health. Because low-grade inflammation results from an appropriate crosstalk between luminal antigens and intestinal mucosal cells, it can be hypothesized that an intestinal immune response lessening could be detrimental for intestinal homeostasis, as any disturbance of intestinal immune response balance. Finally, the present work raises new questions and asks for further development on the consequences of the diet-induced down-regulation of mucosal immune responses and modification of mucus-secreting goblet cell characteristics in physiological and physiopathological contexts.

8

A. Lan et al. / Journal of Nutritional Biochemistry xx (2014) xxx–xxx

Disclosure/Conflict of interest The authors declared no conflict of interest.

[14] [15]

Acknowledgments The authors greatly acknowledge M. Favier and C. Saffre of the Cochin HistIM Facility, and INRA and AgroParisTech for financial support.

[16]

[17]

[18]

References [1] Gibson JA, Sladen GE, Dawson AM. Protein absorption and ammonia production: the effects of dietary protein and removal of the colon. Br J Nutr 1976;35(1):61–5 [Epub 1976/01/01]. [2] Chacko A, Cummings JH. Nitrogen losses from the human small bowel: obligatory losses and the effect of physical form of food. Gut 1988;29(6):809–15 [Epub 1988/06/01]. [3] Gaudichon C, Bos C, Morens C, Petzke KJ, Mariotti F, Everwand J, et al. Ileal losses of nitrogen and amino acids in humans and their importance to the assessment of amino acid requirements. Gastroenterology 2002;123(1):50–9 [Epub 2002/07/10]. [4] Cummings JH, Macfarlane GT. The control and consequences of bacterial fermentation in the human colon. J Appl Bacteriol 1991;70(6):443–59 [Epub 1991/06/01]. [5] Windey K, De Preter V, Verbeke K. Relevance of protein fermentation to gut health. Mol Nutr Food Res 2012;56(1):184–96 [Epub 2011/11/29]. [6] Hughes R, Magee EA, Bingham S. Protein degradation in the large intestine: relevance to colorectal cancer. Curr Issues Intest Microbiol 2000;1(2):51–8 [Epub 2001/11/17]. [7] Macfarlane GT, Macfarlane S. Fermentation in the human large intestine: its physiologic consequences and the potential contribution of prebiotics. J Clin Gastroenterol 2011;45:S120–7 [Suppl., Epub 2011/10/14]. [8] Ruemmele FM. Bacterial mucosa cross-talk and pathophysiology of inflammation. J Pediatr Gastroenterol Nutr 2009;48(Suppl 2):S49–51 [Epub 2009/03/28]. [9] Andriamihaja M, Davila AM, Eklou-Lawson M, Petit N, Delpal S, Allek F, et al. Colon luminal content and epithelial cell morphology are markedly modified in rats fed with a high-protein diet. Am J Physiol Gastrointest Liver Physiol 2010;299(5): G1030–7 [Epub 2010/08/07]. [10] Liu X, Blouin JM, Santacruz A, Lan A, Andriamihaja M, Wilkanowicz S, et al. High protein diet modifies colonic microbiota and luminal environment but not colonocyte metabolism in the rat model: the increased luminal bulk connection. Am J Physiol Gastrointest Liver Physiol 2014;307(4):G459–70. [11] Blachier F, Mariotti F, Huneau JF, Tome D. Effects of amino acid-derived luminal metabolites on the colonic epithelium and physiopathological consequences. Amino Acids 2007;33(4):547–62. [12] Leschelle X, Goubern M, Andriamihaja M, Blottiere HM, Couplan E, GonzalezBarroso MD, et al. Adaptative metabolic response of human colonic epithelial cells to the adverse effects of the luminal compound sulfide. Biochim Biophys Acta 2005;1725(2):201–12 [Epub 2005/07/06]. [13] Mouille B, Robert V, Blachier F. Adaptative increase of ornithine production and decrease of ammonia metabolism in rat colonocytes after hyperproteic diet

[19]

[20] [21]

[22]

[23]

[24]

[25]

[26] [27]

[28]

[29]

[30]

[31]

ingestion. Am J Physiol Gastrointest Liver Physiol 2004;287(2):G344–51 [Epub 2004/04/06]. Peterson LW, Artis D. Intestinal epithelial cells: regulators of barrier function and immune homeostasis. Nat Rev Immunol 2014;14(3):141–53 [Epub 2014/02/26]. Artis D. Epithelial-cell recognition of commensal bacteria and maintenance of immune homeostasis in the gut. Nat Rev Immunol 2008;8(6):411–20 [Epub 2008/05/13]. Guihot G, Blachier F, Colomb V, Morel MT, Raynal P, Corriol O, et al. Effect of an elemental vs a complex diet on L-citrulline production from L-arginine in rat isolated enterocytes. JPEN J Parenter Enteral Nutr 1997;21(6):316–23 [Epub 1997/12/24]. Kristensen NB, Pierzynowski SG, Danfaer A. Net portal appearance of volatile fatty acids in sheep intraruminally infused with mixtures of acetate, propionate, isobutyrate, butyrate, and valerate. J Anim Sci 2000;78(5):1372–9 [Epub 2000/06/02]. Kim YS, Ho SB. Intestinal goblet cells and mucins in health and disease: recent insights and progress. Curr Gastroenterol Rep 2010;12(5):319–30 [Epub 2010/08/13]. Andrianifahanana M, Moniaux N, Batra SK. Regulation of mucin expression: mechanistic aspects and implications for cancer and inflammatory diseases. Biochim Biophys Acta 2006;1765(2):189–222 [Epub 2006/02/21]. Forstner G, Forstner J, Fahim R. Small intestinal mucin: polymerization and the “link glycopeptide”. Symp Soc Exp Biol 1989;43:259–71 [Epub 1989/01/01]. Law GK, Bertolo RF, Adjiri-Awere A, Pencharz PB, Ball RO. Adequate oral threonine is critical for mucin production and gut function in neonatal piglets. Am J Physiol Gastrointest Liver Physiol 2007;292(5):G1293–301 [Epub 2007/01/20]. Hamard A, Mazurais D, Boudry G, Le Huerou-Luron I, Seve B, Le Floc'h N. A moderate threonine deficiency affects gene expression profile, paracellular permeability and glucose absorption capacity in the ileum of piglets. J Nutr Biochem 2010;21(10):914–21 [Epub 2009/12/04]. Schaart MW, Schierbeek H, van der Schoor SR, Stoll B, Burrin DG, Reeds PJ, et al. Threonine utilization is high in the intestine of piglets. J Nutr 2005;135(4):765–70 [Epub 2005/03/30]. Dohi T, Fujihashi K, Koga T, Shirai Y, Kawamura YI, Ejima C, et al. T helper type-2 cells induce ileal villus atrophy, goblet cell metaplasia, and wasting disease in T cell-deficient mice. Gastroenterology 2003;124(3):672–82 [Epub 2003/03/04]. Rakoff-Nahoum S, Paglino J, Eslami-Varzaneh F, Edberg S, Medzhitov R. Recognition of commensal microflora by toll-like receptors is required for intestinal homeostasis. Cell 2004;118(2):229–41 [Epub 2004/07/21]. Santaolalla R, Fukata M, Abreu MT. Innate immunity in the small intestine. Curr Opin Gastroenterol 2011;27(2):125–31 [Epub 2011/01/21]. Wu Y, Kudsk KA, DeWitt RC, Tolley EA, Li J. Route and type of nutrition influence IgA-mediating intestinal cytokines. Ann Surg 1999;229(5):662–7 [discussion 7–8. Epub 1999/05/11]. Coombes JL, Siddiqui KR, Arancibia-Carcamo CV, Hall J, Sun CM, Belkaid Y, et al. A functionally specialized population of mucosal CD103+ DCs induces Foxp3+ regulatory T cells via a TGF-beta and retinoic acid-dependent mechanism. J Exp Med 2007;204(8):1757–64 [Epub 2007/07/11]. Shan M, Gentile M, Yeiser JR, Walland AC, Bornstein VU, Chen K, et al. Mucus enhances gut homeostasis and oral tolerance by delivering immunoregulatory signals. Science 2013;342(6157):447–53 [Epub 2013/09/28]. Davila AM, Blachier F, Gotteland M, Andriamihaja M, Benetti PH, Sanz Y, et al. Intestinal luminal nitrogen metabolism: role of the gut microbiota and consequences for the host (reprinted from Pharmacological Research vol 68, pg 95–107, 2012). Pharmacol Res 2013;69(1):114–26. Wrzosek L, Miquel S, Noordine ML, Bouet S, Chevalier-Curt MJ, Robert V, et al. Bacteroides thetaiotaomicron and Faecalibacterium prausnitzii influence the production of mucus glycans and the development of goblet cells in the colonic epithelium of a gnotobiotic model rodent. BMC Biol 2013;11:61 [Epub 2013/05/23].