The storage globulin phaseolin represents about half of the total protein content of the common bean (Phaseolus vulgaris)(Reference Ma and Bliss1). The nutritive value of phaseolin is limited by its low content in sulphur amino acids and tryptophan and its high resistance to enzymatic hydrolysis(Reference Marquez and Lajolo2, Reference Jivotovskaya, Senyuk and Rotari3). However, phaseolin digestion is markedly improved with heat treatment due to alterations in phaseolin structure(Reference Deshpande and Nielsen4, Reference Deshpande and Damodaran5).
Unheated phaseolin and/or its digestion fragments exert a secretagogue activity on the gut of rats fed a single test meal since intestinal endogenous protein losses (e.g. shed cells, digestive enzymes, gastrointestinal mucus and blood serum proteins) increase when phaseolin intake increases(Reference Santoro, Grant and Pusztai6). These authors postulated that mucin could be a possible significant contributor to these losses(Reference Santoro, Grant and Pusztai6). By contrast, in a chronic feeding trial, the dietary level of unheated phaseolin had little effects on rat small intestine architecture and enzymatic activity(Reference Montoya, Lallès and Beebe7).
The composition, thickness and protective effect of the mucus layer are determined by the dynamic balance between two processes: synthesis and secretion by goblet cells v. degradation by physical abrasion and proteolysis(Reference Montagne, Piel and Lallès8). An intact mucus layer is required at the gut epithelial surface for optimal protection(Reference Engel, Guth and Nishizaki9). Mucin could represent 11 % of total endogenous N losses at the ileum of pigs(Reference Lien, Sauer and Fenton10). Previous studies in single-stomached animals showed that food components, including fibre and protein source and their level of incorporation into the diet could stimulate mucin secretion in the small intestine(Reference Montagne, Toullec and Formal11, Reference Toden, Bird and Topping12). However, no information is available on the effect of prolonged intake of diets differing in phaseolin level or with heated phaseolin on mucin flow in the gut lumen and on mucin gene expression in gut tissues of rats.
The aim of the present study was to test the hypothesis that unheated phaseolin modulates mucin flow and tissue gene expression in the intestines of rats and that heat treatment of phaseolin abolishes these effects.
Experimental methods
Phaseolin purification
The bean cultivar used in the present study contained phaseolin of the T type. It was provided by the International Centre of Tropical Agriculture (Cali, Colombia). Phaseolin was isolated by successive protein solubilisation and precipitation steps as previously reported(Reference Montoya, Lallès and Beebe7, Reference Hall, McLeester and Bliss13). The final phaseolin precipitate was dialysed against distilled water, frozen and freeze-dried before being incorporated into the experimental diets. Phaseolin isolated with this protocol was found to be pure as revealed by SDS-PAGE(Reference Montoya, Lallès and Beebe7).
Animals and diets
The experiment was conducted in agreement with the guidelines of the National University of Colombia for care and use of laboratory animals(Reference Mrad de Osorio and Cardozo de Martinez14). Twenty young adult Wistar male rats with an initial body weight of 110 (sd 5) g were randomly allocated to one of the four treatments described below, and placed in individual metabolism cages (Tecniplast 150-300, Buguggiate, Italy) for the whole experimental period. The control diet (P0) contained casein as the sole protein source. In the other three diets, protein was provided at 33 % (P33) and 67 % (unheated form; P67) and 67 % (heated form; P67-H) by purified phaseolin, respectively. No attempt was made to incorporate 100 % of protein as phaseolin because it was previously shown to drastically reduce food intake in our previous studies with rats(Reference Montoya, Lallès and Beebe7). The complement of protein in the diets was casein (Table 1). Heat treatment of purified phaseolin was carried out under pressure at 121°C for 15 min, as described previously(Reference Montoya, Lallès and Beebe7). Chromium oxide was added to the diets as an indigestible marker for determining the flow and the apparent digestibility of food components along the gut. The rats were fed the experimental diets for 6 d only because the amounts of purified phaseolin were limited. Food intake was fixed at 10 % of body weight in order to limit food refusals(Reference Montoya, Lallès and Beebe7, Reference Montoya, Leterme and Beebe15). The rats had free access to water.
Ingredient and analytical composition of the experimental diets
* Diets: P0, casein control; P33 and P67, diets with phaseolin T contributing to 330 and 670 g/kg of total dietary protein. Casein was supplemented with 30 g dl-methionine/kg DM casein.
† P67-H, heat-treated (121°C for 15 min) phaseolin.
‡ Purified phaseolin of the T type.
§ Mineral and vitamin mixture supplied per kg of diet: 7·5 mg vitamin A; 0·2 mg vitamin D3; 15 mg vitamin E; 6 mg vitamin K; 10 mg vitamin B2; 35 mg calcium pantothenate; 75 mg niacin; 2·5 mg vitamin B6; 0·05 mg vitamin B12; 0·05 mg biotin; 200 mg choline; 150 mg Mn; 500 mg Zn; 40 mg Cu, 200 mg Fe; 2 mg iodine; 0·5 mg Se, 1 mg Co.
Collection and preparation of faeces, digesta and gut tissue samples
Faeces were collected during the last day of the feeding experiment and mixed (350 g/l) with cold saline (9 g NaCl/l) (4°C). At the end of the trial, rats had access to the experimental diets for 4 h and then were euthanised with an injection of Ketamine and Rompun® (1:1, v/v). Gut digesta and tissues were sampled as described previously(Reference Montoya, Lallès and Beebe7, Reference Montoya, Leterme and Beebe15). All digesta present in the distal 20 cm of the small intestine (referred to as ileum) and in the whole colon were collected. Ileal and colonic digesta were gently flushed from the segments with 10 ml of cold saline (9 g NaCl/l) (4°C) using a syringe. Gut digesta and faeces were immediately mixed with 40 g NaN3/l with a final concentration of 2 g/l in order to minimise protein degradation by bacterial enzymes. Digesta and faeces were fractionated into two aliquots that were frozen and stored at − 20°C. An aliquot was kept frozen until mucin analysis, while the other aliquot was freeze-dried and ground (1 mm mesh screen) for digestion studies.
Whole tissue samples (1·5 cm in length) were collected in the middle of the ileum and colon, opened longitudinally and washed three times in cold saline (9 g NaCl/l) (4°C). They were stored immediately in cold TRIzol reagent (4°C), then frozen in liquid N2 and finally stored in a deep freezer at − 80°C.
Enzyme-linked lectin assay for high molecular weight mucin along the gut
Ileal and colonic digesta and faeces were assayed for high molecular weight glycosylated mucin by enzyme-linked lectin assay using wheat germ agglutinin as the lectin, as described by Trompette et al.(Reference Trompette, Blanchard and Zoghbi16). Casein, phaseolin and the experimental diets were also checked by enzyme-linked lectin assay test for possible lectin binding. Porcine gastric mucin (reference M-1778; Sigma-Aldrich, Saiat-Quentin Fallavier, France) was used as the standard. Briefly, dilutions of standards and samples were prepared in carbonate buffer (0·5 m-Na2CO3, pH 9·6) before being coated on ninety-six-well microtitre plates (reference 469914; NUNC microplates, Roskilde, Denmark; 100 μl/well). After overnight incubation at 4°C, the plates were washed four times with PBS Tween (1 g/l) (pH 7·2). Microplate saturation was made with 250 μl/well of a PBS Tween solution of bovine serum albumin (20 g/l PBS Tween) incubated for 1 h at 37°C. Plates were washed again, and 100 μl of biotinylated wheat germ agglutinin (reference B-1025; Abcys, Paris, France; 5 mg/l) in PBS Tween–bovine serum albumin was added and incubated for 1 h at 37°C. Plates were washed, and 100 μl/well of avidin-peroxidase conjugate (reference A-7419; Sigma) was added for 1 h at room temperature. After washing, 100 μl/well of o-phenylenediamine (OPD) solution (reference P-9187; Fast OPD, Sigma) was incubated in the dark at room temperature for 5–10 min. The reaction was stopped by adding 25 μl/well of 3 m-H2SO4. Absorbance was read at 492 nm using an ELISA reader (Multiskan Spectrum reference 5118550; Thermo Electron Corporation, Vantaa, Finland). Mucin concentration in samples was calculated from porcine gastric mucin standard curve. Data were expressed as mg or μg mucin per g digesta or faeces, depending on its concentration along the gut.
RNA extraction from gut tissues
Gut tissue samples were homogenised in TRIzol reagent (1 ml/100 mg) with tissue lyser (Qiagen Inc., Valencia, CA, USA) at room temperature. Then, 200 μl of chloroform was added, and the sample was mixed and centrifuged at 11 300 g for 15 min at 4°C. The chloroform upper phase was recovered, mixed with 500 μl of isopropanol and centrifuged at 11 300 g for 15 min at 4°C. The precipitated RNA was rinsed with 75 % ethanol, centrifuged at 7500 g for 5 min at 4°C, and dissolved into 100 μl of RNase-free water and stored at − 80°C until further analysis. RNA concentration and purity were determined by measuring the absorbance at 260 and 280 nm using an Agilent 2100 bioanalyser (Agilent Technologies, Palo Alto, CA, USA). Finally, samples were treated with DNase (DNA-free kit, Applied Biosystems, Foster City, CA, USA) following the manufacturer's recommendations.
Quantitative real-time RT-PCR
Quantitative real-time RT-PCR was conducted as previously reported(Reference Trompette, Blanchard and Zoghbi16) with slight modifications. Briefly, mucin complementary DNA rat Muc1, Muc2, Muc3, Muc4, trefoil factor 3 (Tff3) and 18S mRNA were amplified by PCR using the primer sequences shown in Table 2. For retrotranscription, total RNA (2 μg) was added with RNase-free water to a final volume of 20 μl. The reaction mixture (High Capacity complementary DNA Reverse Transcription Kit; Applied Biosystems) had a final volume of 20 μl, and contained 4 μl of 10 × RT buffer, 1·6 μl of 25 × deoxy nucleotide triphosphate, 4 μl of 10 × random primers, 2 μl of RT (50 Iu/μl) and 8·4 μl of sterile water. The reaction mixture and the sample were mixed together (40 μl, final volume), then incubated at 25°C for 10 min, at 37°C for 2 h, and finally cooled on ice. Afterwards complementary DNA was ready for use in real-time PCR.
Nucleotide sequences of the PCR primers used to measure the effect of phaseolin in rats
Real-time PCR was performed in duplicate for each sample using ABI PRISM Sequence Detection System 7000 (Applied Biosystems). A reaction mixture containing the following components was prepared: 5·8 μl ultrapure water, 0·75 μl of forward and 0·75 μl of reverse primer (5 μmol/l), 12·5 μl SYBR Green PCR Master Mix kit and 0·2 μl uracil-DNA glycosylase (1 Iu/μl) (Applied Biosystems). The reaction mixture (20 μl) was mixed with sample complementary DNA (5 μl). The cycling conditions were as follows: 50°C for 2 min for uracil-DNA glycosylase action. Then, initial denaturation was conducted at 95°C for 10 min and then followed by forty amplification cycles of 95°C for 15 s and 60°C for 1 min. The Muc genes and Tff3 were expressed relatively to 18S RNA as reported previously(Reference Giulietti, Overbergh and Valckx17).
Chemical analysis
Diets were analysed for ash (AOAC 942.05), diethyl ether extract (AOAC 920.39) and neutral-detergent fibre (AOAC 2002.04). Gross energy was determined in diets using a Parr 1342 calorimeter (Parr Instruments, Moline, IL, USA). Diets, faeces and digesta were also analysed for DM (AOAC 930.15) and N (Kjeldahl method). Chromium concentration in diets, ileal and colonic digesta and faeces was determined colorimetrically after nitro-perchloric acid hydrolysis(Reference Furukawa and Tsukahara18).
Digesta flow and digestibility calculations
The flows of DM, N and enzyme-linked lectin assay mucin at the ileum, in the colon and in the faeces were calculated from marker concentrations in diets and digesta or faeces as reported previously(Reference Montagne, Toullec and Formal11). Apparent ileal and faecal digestibilities of DM and N were calculated using the following equations(Reference Montoya, Gomez and Lallès19):
where DMi, Ni and Cri are the DM, N and chromium contents of ileal digesta, and DMf, Nf and Crf are those in the faeces; DMd, Nd and Crd are the DM, N and chromium contents of the diet.
Statistical analysis
Two separate ANOVA of data were conducted using the Mixed Model procedure of Statistical Analysis System software package version 8.0 (SAS Institute Inc., Cary, NC, USA). In the first analysis, the effect of unheated phaseolin level (0, 33 and 67 % of total N) in the diet was tested for linear and quadratic variations using polynomial orthogonal contrasts(Reference Steel and Torrie20). The second ANOVA was conducted in order to test the effect of heat treatment of phaseolin, and diets with untreated phaseolin, heated phaseolin and casein were compared. When the P value of treatment effects was ≤ 0·10, the diets were compared two by two using appropriate contrasts (P0 v. P67, P0 v. P67-H and P67 v. P67-H).
Results
The voluntary DM and N intakes were on average across treatments 9·6 (sem 0·14) g/d and 151 (sem 4) mg/d, respectively, and did not differ significantly between treatments (P>0·05) (data not shown).
Flow and apparent digestibility of DM and nitrogen
The flow of DM in the faeces, but not at the ileum and in the colon, increased linearly (P < 0·001) with increasing the level of raw phaseolin in the diet (Table 3). As a result, the apparent faecal digestibility of DM decreased quadratically with increasing the level of phaseolin (P < 0·001). The ileal digestibility of DM was not influenced by this factor. The flow of N at all digestive sites increased linearly with increasing the level of unheated phaseolin in the diet (P = 0·011 to < 0·001). As a consequence, both ileal and faecal digestibilities of N decreased linearly with increasing dietary phaseolin level (P < 0·001). Heat treatment of phaseolin reduced DM and N flows along the intestines and increased DM and N digestibilities (P < 0·001) to values close to those observed with the casein-based control diet.
DM, nitrogen and mucin flows along the gastrointestinal tract, and apparent ileal and faecal digestibilities of DM and nitrogen in rats fed graded levels of unheated phaseolin and of heat-treated phaseolin (n 4–5)
a,b Values within a row with unlike superscript letters between P0, P67 and P67-H treatments were significantly different (P < 0·05).
* P0, casein control; P33 and P67, diets with unheated phaseolin contributing to 330 and 670 g/kg of the total dietary protein in the diet; P67-H, diet with heated phaseolin contributing to 670 g/kg of total protein in the diet.
† Comparisons between diets P0, P33 and P67.
‡ Comparisons between diets P0, P67 and P67-H.
Flow of mucin along the gastrointestinal tract and tissue expression of mucin family genes
The enzyme-linked lectin assay test revealed no binding with casein or phaseolin or with the experimental diets, indicating the lack of carbohydrate moieties recognised by wheat germ agglutinin in these ingredients and diets. The flow of mucin in the colon decreased linearly with increasing the dietary phaseolin level (P < 0·001) (Table 3). This factor did not influence the flow of mucin at the ileum or in the faeces (P>0·05). The colonic flow of mucin was lower with P67-H than with the casein control (P < 0·05), but it was not different from that with unheated phaseolin (P67) (probabilities of differences for P0 v. P67, P0 v. P67-H and P67 v. P67-H: P = 0·001, 0·003 and 0·288, respectively).
The expression of Muc3 and Muc4 mRNA tended to decrease quadratically (P = 0·060 and 0·090, respectively) in the ileal tissue and increased or tended to increase linearly (P = 0·093 and 0·023, respectively) in the colonic tissue as the dietary level of unheated phaseolin increased in the diet (Table 4). By contrast, the expressions of Muc1, Muc2 and Tff3 mRNA in the ileal and colonic tissue were not influenced by the level of unheated phaseolin in the diet (P>0·1).
Mucin (Muc) gene expression in ileal and colonic tissues of rats fed graded levels of unheated phaseolin (n 4–5)
Tff3, trefoil factor 3.
* P0, casein control; P33 and P67, diets with unheated phaseolin contributing to 330 and 670 g/kg of the total dietary protein in the diet.
The expression of Muc genes and Tff3 tended to be influenced by heat treatment of phaseolin in colonic tissues, except for the Muc1 gene (Fig. 1). mRNA expressions of Muc2, Muc3, Muc4 and Tff3 were lower with heat-treated phaseolin than unheated phaseolin (P < 0·001 to < 0·05). mRNA expression of the Muc family genes and Tff3 in ileal tissues was not influenced by heat treatment of phaseolin (P>0·05). Muc2 mRNA levels were lower in ileal and colonic tissues of rats fed the heat-treated phaseolin diet than in those fed the casein-based control diet (P = 0·003 and 0·001, respectively). Finally, Muc2 gene expression in the ileal tissue was lower (P = 0·004), and those of Muc3 and Muc4 in the colonic tissue of rats fed unheated phaseolin tended to be higher (P = 0·058 and 0·005) when compared with the corresponding tissues of rats fed the casein-based control diet.
Influence of heat treatment of phaseolin on mucin (Muc) and trefoil factor 3 (Tff3) gene expression in ileal and colonic tissues of rats: Muc2 in the ileum (a); Muc2 in the colon (b); Muc3 in the ileum (c); Muc3 in the colon (d); Muc4 in the ileum (e); Muc4 in the colon (f); Tff3 in the ileum (g); Tff3 in the colon (h). Values are means (n 4–5), with their standard errors represented by vertical bars. P0, casein control diet; P67, diet containing phaseolin at 670 g/kg total protein; P67-H, diet containing heat-treated phaseolin at 670 g/kg total protein. The overall probabilities for treatment effects were for Muc2 (P = 0·004 and 0·001), Muc3 (P = 0·789 and 0·027), Muc4 (P = 0·732 and 0·003) and Tff3 (P = 0·432 and 0·062) in the ileum and in the colon, respectively.
Discussion
The present investigation shows for the first time that phaseolin intake and heat treatment can modulate mucin flow and mRNA levels of various mucin genes in the gut tissues of rats.
Influence of phaseolin on mucin flow and gut tissue expression of Muc family genes
Santoro et al.(Reference Santoro, Grant and Pusztai6) based on acute feeding experiments assumed that the poor nutritional value of unheated phaseolin was due to increased intestinal losses of endogenous N, suggestively mucin, when the level of phaseolin increased in the diet. The present results do not support this view because raw phaseolin intake did not increase mucin flow at the end of the small intestine and did not alter intestinal mRNA levels of Muc2, the main component of intestinal secreted mucin(Reference Corfield, Myerscough and Longman21). The discrepancies between the present investigation and studies by Santoro et al.(Reference Santoro, Grant and Pusztai6, Reference Santoro, Grant and Pusztai22) may come from different experimental approaches (acute v. repeated feeding experiments) or from methodologies for evaluating endogenous N losses. The present data are consistent with our previous observations showing limited effects of phaseolin on intestinal anatomy and enzyme activities(Reference Montoya, Lallès and Beebe7, Reference Montoya, Leterme and Beebe15).
We observed that dietary phaseolin level and heat treatment influenced the expression of Muc2, Muc3 and Muc4 and Tff3 genes in different ways along the intestines. The actual reasons for these effects and the possible consequences in terms of gut protection are not fully understood yet. Mucin 2 is the major component of mucin secreted along the gastrointestinal tract(Reference Corfield, Myerscough and Longman21). Muc2 knock out mice develop colitis spontaneously, indicating the important role of this mucin in colonic protection(Reference Van der Sluis, De Koning and De Bruijn23). In the present study, both unheated and heat-treated phaseolin reduced Muc2 gene expression in the ileum, and heat treatment of phaseolin reduced Muc2 gene expression in the colon as compared to the casein control (Fig. 1). These observations may suggest a potentially weaker gut protection following phaseolin intake, with differential effects in the ileum and the colon depending on phaseolin cooking.
Muc3 contributes to the protection of the intestinal epithelium. A higher intestinal expression of Muc3 mRNA following hypoxia suggests a protective mechanism during episodes of diminished oxygen delivery(Reference Louis, Hamilton and Canny24). Conversely, a reduction in Muc3 mRNA levels has been reported in Crohn's disease patients(Reference Corfield, Myerscough and Longman25). In the present study, the level of unheated phaseolin in diets affected Muc3 mRNA levels in opposite ways depending on the intestinal segment: it decreased in ileal tissue and increased in colonic tissue. These results suggest that ileal Muc3 gene expression may be negatively regulated by ileal N flow, while that of colonic Muc3 may be positively regulated by colonic N flow. Although the correlation between ileal Muc3 mRNA level and ileal N flow did not reach significance, colonic Muc3 mRNA level was found to be positively correlated to colonic N flow (r 0·70, P < 0·05). Heat treatment of phaseolin decreased the expression of Muc3 mRNA in the colon, an observation that supports the latter assumption. Further work is needed to demonstrate causal links between mucin gene expression and digesta flow or fermentation in order to contribute to explain regional variations in the expression of these genes along the gut.
The Muc4 gene is expressed in cells at the basement of crypts in the small intestine, but its expression is higher in the colon where it is located in goblet cells(Reference Rong, Rossi and Zhang26). Muc4 appears to play important roles in epithelial growth, cell differentiation(Reference Rossi, McNeer and Price-Schiavi27), mucosal defence(Reference Hoebler, Gaudier and De Coppet28) and intestinal lubrication(Reference Rong, Rossi and Zhang26). In the present study, the increased expression of Muc4 mRNA levels in colonic tissues and the trend in ileal tissues, in response to increased levels of dietary phaseolin, could not be explained by changes in the flow of DM (or fresh digesta) in the ileum and the colon that were NS. Heat-treated phaseolin also reduced the expression of Muc4 mRNA.
The trefoil family 3 (Tff3) gene is also expressed in the mucin secretory cells. It helps to protect and stabilise the mucus layer and heal the epithelium(Reference Kindon, Pothoulakis and Thim29). As for the Muc family genes (except Muc1), a reduction in the expression of Tff3 mRNA in the colonic tissue after thermal treatment of phaseolin was observed. A positive association between intestinal trefoil factor and Muc3 expressions has been linked with mucosal hypoxia and altered epithelial barrier function(Reference Louis, Hamilton and Canny24).
The link between Muc gene tissue expressions and mucin flow along the intestines does not seem to be straightforward, because this flow reflects the balance between mucin production and luminal degradation by the microbiota(Reference Montagne, Piel and Lallès8) and also because Muc gene expression varies regionally along gut compartments for a given diet(Reference Hedemann, Theil and Bach Knudsen30). The most consistent point in the present study was the lower colonic mucin flow with heated phaseolin, which could reflect at least partly the lower colonic Muc2 mRNA levels with this diet. By contrast, Muc gene levels and mucin flow varied in different ways along the gut according to the dietary treatments (e.g. in the ileum, the casein-based control diet had similar mucin flow and higher Muc2 gene expression than phaseolin-containing diets).
In previous studies with longer periods of feeding (2 weeks) similar phaseolin-based diets, we did not observe health problems or important alterations in gut anatomy and enzyme activities(Reference Montoya, Leterme and Beebe15). Therefore, it can be suggested that the changes in the expression of Muc genes in gut tissues (mainly in colon) as noted here might have had little effect on gut function, at least in our experimental conditions.
Nutritional factors modulating Muc gene expression and mucin synthesis
A number of studies suggest that casein may influence mucin secretion and gene expression in the intestine(Reference Montagne, Toullec and Formal11, Reference Moughan, Butts and Rowan31–Reference Zoghbi, Trompette and Claustre34). Varying the level and the origin (animal v. plant) of protein in milk formulas modulated mucin gut flow in baby calves(Reference Montagne, Toullec and Formal11). Hydrolysed casein increased ileal endogenous amino acid losses in human subjects(Reference Moughan, Butts and Rowan31) and gut tissue Muc3 and Muc4 mRNA levels in rats(Reference Han, Deglaire and Sengupta32). Hydrolysed casein and related peptides (e.g. β-casomorphin-7) also induced mucin secretion in isolated and perfused rat jejunum(Reference Claustre, Toumi and Trompette33) and increased the expression of Muc2 and Muc3 genes in intestinal mucin-producing cells(Reference Zoghbi, Trompette and Claustre34). Here, casein may have been responsible for the higher Muc2 mRNA levels observed in the ileal tissues. However, in the colon, Muc gene expression was similar to or higher than those observed with casein when the level of unheated phaseolin increased in the diet. Possible effects of casein peptides on intestinal mucin gene expression and flow make the interpretation of data more difficult in studies where casein is used as the control and is substituted with other protein sources (e.g. phaseolin), as in the present study. However, it must be borne in mind that luminal concentrations of casein peptides, including that of β-casomorphin-7 after milk or casein intake, have never been determined in vivo (Reference Claustre, Toumi and Trompette33). Also, two independent studies revealed that the bioactive peptide β-casomorphin-7 is not detected in digests after casein digestion in vitro (Reference Schmelzer, Schöps and Reynell35, Reference Picariello, Ferranti and Fierro36). Additionally, the demonstration of β-casomorphin-7 secretory properties on jejunal mucin with isolated and perfused rat jejunum(Reference Claustre, Toumi and Trompette33) does not provide evidence for such effects on the ileum or the colon, the two sites under study in the present study. Finally, results from investigations with hydrolysed casein(Reference Moughan, Butts and Rowan31, Reference Claustre, Toumi and Trompette33) do not mean that entire casein that is subsequently hydrolysed by endogenous proteases and peptidases may have resulted in the same effects as exogenously hydrolysed casein on mucin flow and gene expression. Collectively, there is no evidence to date showing that casein peptides (like β-casomorphin-7) are released during casein digestion and are bioactive in vivo. Therefore, the most reasonable interpretation of the present data is that phaseolin was responsible for the observed changes in intestinal mucin flow and gene expression.
Dietary threonine is also important to consider because it is highly represented in mucin, its dietary restriction reduces intestinal mucin synthesis specifically(Reference Faure, Moënnoz and Montigon37, Reference Nichols and Bertolo38) and it may become a limiting amino acid for mucin synthesis in intestinal inflammation in rats and pigs(Reference Faure, Mettraux and Moennoz39, Reference Rémond, Buffière and Godin40). Threonine restriction reduced Muc gene expression in the small and large intestines of rats(Reference Faure, Mettraux and Moennoz39). According to the amino acid composition of casein and phaseolin(Reference Montoya, Leterme and Victoria41) and the ileal digestibility of N in the present diets, the theoretical availability of amino acids decreased as the level of unheated phaseolin increased in the diet. The casein-based control diet provided 25 and 68 % more threonine than phaseolin P33 and P67 diets, respectively. These diets with unheated phaseolin did not lead to a reduction in ileal mucin flow. The limited threonine supply with the unheated phaseolin-containing diets may have been responsible at least partly for the changes recorded in the ileal Muc gene expression in the present study. In the colon, the mucin flow decreased in proportion to the unheated phaseolin included in the diet. This could have been caused by bacterial fermentation being stimulated by indigested phaseolin components, thus leading to enhanced mucin degradation in the colon. Dietary protein type and amino acid composition influence the colonic microbiota and resulting profiles of SCFA, which, in turn, modulate the mucus layer(Reference Montagne, Piel and Lallès8, Reference Toden, Bird and Topping12) and the expression of secreted (Muc2) and membrane-bound (Muc1, Muc3 and Muc4) mucins(Reference Hedemann, Theil and Bach Knudsen30, Reference Burger-van Paassen, Vincent and Puiman42, Reference Gaudier, Rival and Buisine43). It can partially explain the increased proportion of Muc3 and Muc4 gene expressions as dietary unheated phaseolin was included in the diet.
Although the casein control diet and the heat-treated phaseolin diet had similar and high digestibilities, they displayed different Muc2 mRNA levels in ileal and colonic tissues and different colonic mucin flow. Both of these observations could be explained by the β-casomorphin-7 peptides of casein(Reference Claustre, Toumi and Trompette33) and/or changes in microbiota(Reference Hedemann, Theil and Bach Knudsen30, Reference Burger-van Paassen, Vincent and Puiman42, Reference Gaudier, Rival and Buisine43) as explained above. New investigations are needed to determine the influence of phaseolin on gut fermentation profiles and possible links with gut Muc gene expression.
Effect of phaseolin on digesta flow along the gastrointestinal tract and on digestibility
The present results revealed increased flows of DM and N that led to a reduction in apparent N digestibility in rats fed with the highest level of unheated phaseolin in the diet, in agreement with our previous studies(Reference Montoya, Leterme and Beebe15). The low nutritional value of unheated phaseolin consumed chronically may be due mainly to its high resistance to enzymatic hydrolysis, as evidenced by increased ileal and faecal output of undigested phaseolin polypeptides(Reference Montoya, Lallès and Beebe44). Heat treatment of phaseolin reduced ileal and faecal N output, thus increasing the apparent digestibility of N. These observations are in agreement with previous investigations(Reference Marquez and Lajolo2, Reference Montoya, Lallès and Beebe7, Reference Liener and Thompson45, Reference Marquez and Lajolo46). Recently, we showed that these improvements were related to the disappearance of undigested phaseolin polypeptides in ileal digesta following heat treatment of phaseolin(Reference Montoya, Leterme and Beebe15, Reference Montoya, Lallès and Beebe44).
In conclusion, the present study provides evidence that the level and the source of protein influence the flow of mucin and the expression of various Muc family genes in the ileal and colonic tissues. Additionally, we showed that different sources of protein (casein v. phaseolin) with similar digestibility could influence Muc gene expression in the intestines differentially. Further work is required to elucidate the actual mechanisms of Muc gene modulation by phaseolin and to evaluate the possible functional outcomes of phaseolin intake in terms of gut protection.
Acknowledgements
The authors thank the Volkswagen Foundation (Hannover, Germany), COLCIENCIAS (Bogotá, Colombia), ECOS-Nord (Université de Paris 5, France) and Conseil Régional de Bretagne (Rennes, France) for financial support. The study was not subject to conflicts of interest of any kind. The research is public. The bean varieties containing the different phaseolin types are available to research scientists at the International Centre of Tropical Agriculture, as long as they are not commercialised afterwards. P. L., J. C. and J.-P. L. designed the study and revised the manuscript; C. A. M. did the laboratory work and prepared the manuscript under the supervision of J.-P. L.; S. B. produced the beans and the phaseolins; and V. R. did the gene expression analysis.
