The Journal of Nutrition
Nutrient Requirements and Optimal Nutrition
Intestinal Inflammation Increases Gastrointestinal
Threonine Uptake and Mucin Synthesis in
Enterally Fed Minipigs
1–3
Didier Re´mond,
4* Caroline Buffie`re,
4Jean-Philippe Godin,
5Philippe Patureau Mirand,
4Christiane Obled,
4Isabelle Papet,
5Dominique Dardevet,
4Gary Williamson,
5Denis Breuille´,
5and Magali Faure
54INRA, UMR1019 Human Nutrition Unit, F-63122 Clermont-Fd/Theix, France and5Nestle´ Research Center, CH-1000 Lausanne, Switzerland
Abstract
The high requirement of the gut for threonine has often been ascribed to the synthesis of mucins, secreted threonine-rich glycoproteins protecting the intestinal epithelium from injury. This requirement could be even greater during intestinal inflammation, when mucin synthesis is enhanced. In this study, we used an animal model to investigate the effects of an acute ileitis on threonine splanchnic fluxes. Eight adult multi-catheterized minipigs were fed with an enteral solution. Four of them were subjected to experimental ileitis involving direct administration of trinitrobenzene sulfonic acid (TNBS) into the ileum (TNBS-treated group) and the other 4 were not treated (control group). Threonine fluxes across the portal-drained viscera (PDV) were quantified with the use of simultaneous i.g.L-[15N]threonine and i.v.L-[U-13C]threonine infusions. Ileal mucosa was sampled for mucin fractional synthesis rate measurement, which was greater in the TNBS-treated group (114 6 15%/d) than in the control group (61 6 8%/d) (P¼ 0.021). The first-pass extraction of dietary threonine by the PDV and liver did not differ between groups and accounted for ;27 and 10% of the intragastric delivery, respectively. PDV uptake of arterial threonine increased from 25 6 14 mmolkg21.
h21in the control group to 171 6 35 mmolkg21.
h21in the TNBS-treated group (P , 0.001). In conclusion, ileitis increased intestinal mucin synthesis and PDV utilization of threonine from arterial but not luminal supply. This leads to the mobilization of endogenous proteins to meet the increased threonine demand associated with acute intestinal inflammation. J. Nutr. 139: 720–726, 2009.
Introduction
The gastrointestinal tract is a highly secretory and proliferative
tissue and, although it contributes only 3–6% of the body weight, it
accounts for 20–35% of the whole-body protein turnover (1). The
composition of secreted proteins largely affects its requirement for
individual amino acids. For instance, the core proteins of intestinal
mucins secreted in the lumen by the epithelium contain high
amounts of threonine (2) and their production could explain the
high rate of gastrointestinal threonine utilization (3,4). Whether
mucins secreted in the upper part of the digestive tract can be
digested, and their amino acids reabsorbed in the small intestine,
remains conceivable. However, threonine recycling appears to be
low (5) and threonine ileal loss particularly high, at least with
respect to the whole-body threonine requirement (6). In
inflamma-tory bowel diseases such as ulcerative colitis and Crohn’s disease,
the mucus layer and mucin production are often qualitatively and
quantitatively impaired (7). Using models of chronic colitis in rats,
we observed that although mucin synthesis was unchanged, the
mucin composition was altered all along the gut, showing a decrease
in threonine content (8). These data suggested that in inflammatory
diseases, because of its involvement in the synthesis of acute phase
proteins and proteins of the immune system (9), and possibly
because of an increase in its ileal loss, threonine may become a
limiting amino acid for mucin synthesis. Indeed, an increased supply
of a mixture of amino acids, including threonine, serine, proline,
and cysteine, increased mucin synthesis during colitis (10). Thus, an
adequate threonine supply seems essential to restore gut barrier
integrity and function during intestinal inflammation and therefore
to improve recovery. It has been suggested that besides
corticoste-roid therapy, the use of enteral nutrition could be considered as a
primary treatment in inflammatory bowel diseases (11). In this
context, the definition of amino acid requirement, and particularly
threonine requirement, by the gastrointestinal tract in the course of
intestinal inflammation appeared crucial.
In this study, we investigated threonine fluxes in the
portal-drained viscera (PDV)
6using minipigs as an animal model. We
1
Supported by the Institut National de la Recherche Agronomique (France) and the Nestle´ Research Center (Switzerland).
2
Author disclosures: D. Re´mond, C. Buffie`re, J.-P. Godin, P. P. Mirand, C. Obled, I. Papet, D. Dardevet, G. Williamson, D. Breuille´, and M. Faure, no conflicts of interest. 3
Supplemental Appendix 1, Supplemental Figure 1, and Supplemental Tables 1–3 are available with the online posting of this paper at jn.nutrition.org. * To whom correspondence should be addressed. E-mail: dremond@clermont.inra.fr.
6
Abbreviations used: FSR, fractional synthesis rate; MPE, molar percent excess; MPO, myeloperoxidase; PDV, portal-drained viscera; TNBS, trinitrobenzene sulfonic acid.
0022-3166/08 $8.00ª 2009 American Society for Nutrition.
720 Manuscript received October 29, 2008. Initial review completed December 2, 2008. Revision accepted January 13, 2009. First published online February 4, 2009; doi:10.3945/jn.108.101675.
hypothesized that an intestinal inflammation (ileitis), through an
increase in mucin synthesis, would increase specifically
threo-nine utilization by the digestive tract and possibly impair the
reabsorption of endogenous protein secretions in the ileum.
Using stable isotopically labeled threonine by both i.v. and
intraduodenal routes, we investigated the effect of an acute ileitis
on the partitioning of threonine uptake by the gastrointestinal
tract, dietary vs. arterial supply, in minipigs undergoing enteral
feeding with an elemental solution.
Materials and Methods
All procedures were in accordance with the guidelines formulated by the European Community for the use of experimental animals (L358–86/ 609/EEC) and the study was approved by the Regional Committee for Ethics in Animal Experimentation (CREEA d’Auvergne, Aubie`re, France).
Animals.The study involved 8 adult Pitmann-Moore minipigs (CEGAV) (8–10 mo old; 26–30 kg body weight). They were housed in individual pens (1 3 1.5 m) separated by Plexiglass walls in a ventilated room with controlled temperature (20–23C).
Surgical procedures.Before surgery, the minipigs were deprived of food for 24 h. One-half hour after premedication by i.m. injection of 7 mg/kg ketamine (Imalge`ne 1000; Rhoˆne Me´rieux), 1.5 g/kg azaperone (Sresnil, Janssen Pharmaceutica), and atropine (Atropine 0.1%, Labo-ratoire Aguettant), anesthesia was induced via a snout mask with 10% isoflurane (Forene, Abbot) carried by oxygen for 10 min. After oral endotracheal intubation, the isoflurane concentration was reduced to 2% to maintain general anesthesia during the total surgery time. After midline laparotomy, minipigs were fitted with catheters (polyvinyl chloride; 1.1-mm i.d., 1.9-mm o.d.) in the portal vein, the aorta, and the caval vein. The portal vein catheter was placed via the splenic vein with the tip positioned in the liver hilus. This catheter was used for Ringer’s lactate (500–1000 mL) delivery during the remainder of the surgery. A transit time ultrasonic blood flow probe (Transonic Systems) was implanted around the portal vein (14-mm probe, A-series). Finally, a T-shaped cannula (silicone rubber; 12-mm i.d., 17-mm o.d.) was inserted into the ileum 15 cm upstream of the ileocecal valvule and a catheter (silicone rubber; 1.57-mm i.d., 3.18-mm o.d.) was inserted into the stomach. Vascular catheters and probe cables were exteriorized through the skin on the right flank of the minipig, whereas ileal cannula and gastric catheter were exteriorized on the left flank. The first 4 d after surgery, minipigs were treated with an antibiotic (Clamoxyl, Pfizer Sante´ Animal) as a prophylactic measure and Flunixin (Finadyne, Shering-Plough) was administered i.v. for postoperative analgesia. Feed allot-ments were progressively increased over 7 d up to the 500-g daily ration. A minimum of 2 wk was allowed for recovery from surgery before initiating the experiment.
Experimental ileitis.Ileitis was induced by intraluminal injection of trinitrobenzene sulfonic acid (TNBS) using a similar procedure used by Shibata et al. (12) in dogs. Minipigs were food deprived for 12 h. They were anesthetized by i.v. injection of 8 mg/kg ketamine, which was thereafter added if necessary to maintain anesthesia adequately during the full procedure of ileitis induction. The ileal cannula was removed, offering direct access to intestinal lumen. A fiberscope was introduced into the intestine toward the jejunum over ;70 cm. A tube fitted at its tip with 2 inflatable latex balloons at 20-cm intervals was slid into the intestine until its extremity passed beyond the fiberscope one. The first balloon was then inflated (20 cm3) and the fiberscope moved back to the
2nd balloon, which was inflated in turn before removing the fiberscope. The 20-cm segment closed between the 2 inflated balloons was rinsed out with 50% ethanol. A 30-mL solution of 120 mg/kg TNBS (Sigma-Aldrich chimie) dissolved in 50% ethanol was introduced into the isolated segment and kept in place for 30 min. The 2 balloons were then deflated, the tube removed, and the cannula was replaced. On the day of treatment, minipigs were not fed. One day before and for 3 d following
TNBS challenge, arterial serum was collected for C-reactive protein determination as a marker of inflammation (acute phase response). Experimental protocol.Throughout the experimental period, a liquid diet was enterally infused into the gastric catheter via a swivel system connected to a pump. The daily diet was prepared by mixing Nestle´ f.a.a (40 mL/kg of body weight), which is an elemental nutrition support containing free amino acids, with Nutriose FB (Roquette; 6 g/L of Nestle´ f.a.a), which provided soluble dietary fiber. The mixture was made up to 1.2 L with water. The diet was infused at a rate of 200 mL/h for 6 h (between 0900 and 1500). The enteral feeding provided 2 g protein equivalentkg21d21, 1674 kJkg21d21, and 140 mg of
threoninekg21d21.
The effect of intestinal inflammation was evaluated by measuring threonine fluxes 4 d before (d 24) and 3 d after (d 3) ileitis induction in 4 TNBS-treated minipigs (TNBS-treated group) and 4 control minipigs (control group) that did not undergo the procedure described above for ileitis induction but were submitted only to the period of starvation. On the days of sampling, baseline arterial and portal blood samples were drawn before starting the enteral nutrition. Once the infusion of the enteral solution was started, after an initial bolus equivalent to 1 h of infusion,L-[U-13C]threonine andL-[15N]threonine (Cambridge Isotope Laboratories) in sterile saline were infused (6 mL/h) at a rate of 4 mmolkg21h21in the caval vein and the stomach, respectively. Blood samples were then simultaneously taken from the artery and the portal vein at 60, 120, 210, 270, 300, 330, 360, and 390 min. Portal blood flow was continuously recorded during the sampling session.
At the end of the last sampling day, immediately after the last blood samples were taken, the minipigs were killed by i.v. injection of 125 mg/kg body weight of sodium pentobarbital (Dole´thal, Vetoquinol). The abdomen was opened and the ileum was then rapidly excised, flushed with saline, and opened. Two 20-cm segments were isolated: one corresponding to the inflamed area, the other one being collected about 1 m upstream. For both segments, examination of the mucosa and scoring of macroscopically visible damage was performed according to the scale proposed by Rodriguez-Cabezas et al. (13). A small piece of the whole intestinal wall was fixed in Bouin’s (Pioneer Research Chemicals) for later histological examination and the mucosa of the remaining intestine was removed by scraping with a microscope slide, immediately frozen in liquid nitrogen, and stored at 280C.
Sample processing. Blood samples were collected in prechilled syringes containing either EDTA-K (for whole blood analysis) or heparin lithium (for blood gas analysis and plasma) as anticoagulant. Hemoglo-bin concentration and packed cell volume were immediately determined using an automatic blood-gas analyzer (ABL510, Radiometer). Plasma was separated by centrifugation at 1500 3 g; 10 min at 4C. Whole blood and plasma spiked with norvaline and norleucine, respectively, as internal standards were deproteinized with sulfosalicylic acid (190 mmol/L, final concentration) and stored at 280C.
Histological examination and scoring of inflammation in ileum specimens.Fixed tissues, stored in 70% ethanol, were dehydrated, embedded in paraffin blocks, sectioned (5 mm thick), and successively stained with Harris hematoxylin, eosin-erythrosin, and saffron for evaluation of epithelial damage. Slices were scored by a histopathologist unaware of the treatments on a scale from 0 to 20 according to criteria proposed by Millar et al. (14). Slices were equally stained with iron diamine/Alcian blue for evaluation of the number of mucin-containing goblet cells.
Analytical methods.Biochemical analysis of ileum myeloperoxidase (MPO) activity was performed with a dianisidine-H2O2assay (15). The
C-reactive protein concentration in serum was determined by solid phase sandwich immunoassay (Tridelta Development).
Plasma amino acid concentrations were measured by ion-exchange chromatography (Bio-Tek Instruments A.R.L). The threonine concen-tration in blood was measured by reversed-phase HPLC using precolumn derivatization with O-phthaldehyde reagent (16). For isotopic analysis, deproteinized blood samples (400 mL) were applied to a 3-mL bed
volume column of AG50-X8 resin (100–200 mesh) in the H1
form. After a water wash, the amino acids were eluted with NH4OH and dried under
vacuum. The isotopic measurements ofL-[U-13C] threonine andL-[15N] threonine in plasma samples were carried out by GC-MS using N(O,S)-ethoxycarbonyl ethyl ester derivative. Briefly, the analyses were per-formed on a gas chromatograph (6890 series; Hewlett-Packard) using helium as carrier gas, coupled to a quadrupole mass spectrometer (MSD 5975; Agilent) operating in electron impact-ionization mode. Ions at mass:charge ratios of 175, 176, and 177 for the M, M11 (15
N-threonine), and M12 (13C-threonine) isotopomers, respectively, were
monitored to determine the isotopic enrichment of [15N]threonine and
[13C]threonine. To check and correct the accuracy of the13C-isotopic
measurement carried out in this condition, mixtures of a known amount of unlabeled threonine and a known amount of labeled [U-13
C]Thre-onine were used and analyzed by GC-MS. Moreover, because the iso-topic enrichment is in the low range for the GC-MS device, a few samples were split in 2 and analyzed by GC-MS and GC-combustion-isotope ratio mass spectrometer and did not differ. The isotopic enrichments were calculated using the tracer/tracee approach and then subsequently transformed to a molar percent excess (MPE) value according to Rosenblatt et al. (17). This approach takes into account the isotopic contributions of labeled atoms and the natural abundance of the elements present in the derivatized molecules.
Mucins were purified from the ileal mucosa of minipigs and their amino acid composition was analyzed as previously described by Faure et al. (8). The 13C-isotopic enrichment of threonine in mucins was
performed using N(O,S)-ethoxycarbonyl ethyl ester derivatives (18). Calculations.Measurements were performed under steady-state condi-tions. Amino acid fluxes in fed minipigs were calculated using only the last 4 sampling times, 300–390 min after the onset of enteral feeding and tracer perfusions, when amino acid concentrations and [13C]threonine
and [15N]threonine enrichment reached a plateau (Supplemental Fig. 1).
We estimated the arterial blood flow to the portal-drained vein viscera (BFa) as follows:
BFa¼ BFv3Hbv=Hba;
where BFvis measured blood flow (Lkg21h21) in the portal vein and
Hbaand Hbvare hemoglobin concentrations (g/L) in arterial and venous
whole blood, respectively. We used the hemoglobin ratio to take in account water movement across the PDV.
Plasma amino acid net fluxes across the PDV (Fnet AA,
mmolkg21h21) were calculated as
FnetAA ¼ ð½AAv3PFvÞ 2 ð½AAa3PFaÞ;
where [AA] is the blood amino acid concentration (mmol/L), PF is plasma flow, and subscripts a and v refer to arterial and venous data, respectively. PF was calculated from blood flow with the relation PF ¼ BF 3 (1 2 PCV).
Whole-body threonine flux (Ra, mmolkg21h21) was derived from the equation
Ra ¼ Inf 3 ½ðEi Thrinf=Ei ThrartÞ 2 1;
where Inf is the rate of infusion of tracer (mmolkg21h21) and Ei Thrinf
and Ei Thrartare isotopic enrichments (MPE) in infused solution and
arterial blood, respectively. Ra was calculated with i.v. tracer (Raiv) or
i.g. tracer (Raig).
The difference between the whole-body fluxes calculated from the i.v. and the i.g. tracers was assumed to represent the first-pass splanchnic extraction of the i.g. tracer. Thus, the first-pass splanchnic extraction rate (Extsplanchnic, %) was calculated as
Extsplanchnic¼ ðRaig2RaivÞ=Raig:
I.v. and i.g. infusion of labeled threonine allowed us to determine the unidirectional fluxes that underlie net fluxes across tissues. At the PDV level, it offers the possibility to distinguish the origin (luminal vs. arterial) of the threonine used by the mucosa. The unidirectional fluxes of threonine across the PDV were calculated from arterial and portal threonine isotopic enrichments and whole blood concentrations, and blood flow. A schematic model of the different fluxes is depicted (Fig. 1).
The equations used to calculate these fluxes are detailed in Supplemental Appendix 1.
The fractional synthesis rate (FSR; %d21) of mucins was calculated from threonine13C isotopic enrichment in mucins (Ei MUCmucosa, MPE)
and the isotopic enrichment of free threonine in the mucosa (Ei Thrmucosa, MPE). Ei Thrmucosais not constant during the entire infusion
period and the plateau value is not reached immediately. The time course of Ei Thrmucosacan be described by a simple exponential: Ei Thrmucosa¼
Ei Thrmucosamax (1 – e2lt), where Ei Thrmucosamax is the isotopic
enrichment during the steady state and l is an empirically-fitted rate constant. The evolution of free threonine 13C enrichment in arterial
whole blood can also be described by a single exponential analogous to the equation presented above for tissues. We assumed that l in mucosa is equivalent to the value determined in plasma, and FSR was calculated as
FSR ¼ Ei MUCmucosa=Ei Thrmucosa3 l=½tl ð1 2 e2 ltÞ:
Statistics.Data were analyzed using the repeated option of the PROC MIXED procedure of SAS (SAS/STAT Users Guide, release 8.1; SAS Institute), with subjects as random effect and day, group, day 3 group as factors. When a significant day 3 group interaction was found, we used the LSMEANS procedure to test differences at specific days, between and within groups. Unpaired Student’s t tests were used to compare between the 2 groups data obtained on intestinal mucosa (d 3). Results are presented as means 6 SEM. Means are considered different at P , 0.05.
Results
Inflammatory consequences of TNBS treatment.
In the
TNBS-treated group, 3 d after treatment, annular or
irregular-shaped lesions covered by yellowish exudates were observed
macroscopically in the ileum at the site of TNBS administration
(20 cm), but no apparent lesions were observed in the small
intestine upstream of the treated area. The intestinal wall was
generally thickened and edema was evident. No apparent
lesions were observed in the small intestine upstream of the
treated area. The average macroscopic score of the
TNBS-treated group was greater than that of the control group (Table
1). A more extensive analysis of ileitis was performed
histolog-ically by scoring inflammation (Table 1). Ileum specimens of
TNBS-treated minipigs were characterized by focal necrosis of
mucosal epithelium, edema, and submucosal neutrophilic
infil-tration. In agreement, ileum MPO activity tended to be greater
(t test, P ¼ 0.057) in the TNBS-treated group than in the control
group (Table 1). Histological examination and MPO activity
revealed no inflammation in the ileum segment upstream of the
treated area.
FIGURE 1 Schematic representation of the metabolic fate of enteral and systemic threonine in the PDV of adult minipigs receiving i.g. infusion of an elemental solution (3 d after the enteral feeding onset). (A) I.g. infusion of threonine. (B) First-pass uptake of dietary threonine. (C) Release of dietary threonine in the portal vein. (D) Uptake of arterial threonine. (E) Release of nondietary threonine in the portal vein. Values are means (expressed as mmolkg21
h21), n
¼ 8.
722 Re´mond et al.
The C-reactive protein concentration in peripheral serum of
TNBS-treated minipigs increased up to 200 mg/L within the first
24 h following TNBS challenge (P , 0.01). However, it returned
to prechallenge values on d 3 (data not shown), indicating a
transient acute phase protein response occurring in the liver
following the local intestinal inflammation induction.
Plasma amino acid arterial concentrations and net fluxes
across the PDV.
Except for glutamine, glutamate, and
histi-dine, the arterial concentrations of the different amino acids
were not affected by group, day of sampling, or their interaction
(Supplemental Table 1). The arterial glutamate concentration in
the control group was greater (P ¼ 0.049) at d 3 (167 6 7 mmol/
L) than at d 24 (126 6 9 mmol/L), whereas it was unchanged in
the TNBS-treated group (P ¼ 0.202). Conversely, for glutamine,
the arterial concentration was lower (P ¼ 0.044) at d 3 (235 6
10 mmol/L) than at d 24 (357 6 25 mmol/L) in the TNBS-treated
group, but did not change in the control group (P ¼ 0.189).
The arterial threonine concentration was 728 6 89 mmol/L. The
portal blood flow tended to decrease between d 24 and d 3 in
the control group (P ¼ 0.089) but increased in the TNBS-treated
group (P ¼ 0.024) (Table 2).
The pattern of amino acid net fluxes across the PDV was
characterized by a decrease in threonine (P ¼ 0.050), serine (P ¼
0.008), glutamic acid (P ¼ 0.032), leucine (P ¼ 0.001), and
proline (P ¼ 0.003) net release, and glutamine net uptake (P ¼
0.008) in the TNBS-treated group between d 24 and d 3
(Supplemental Table 2). However, because the net release of
serine, glutamic acid, and leucine also decreased slightly in the
control group between d 24 and d 3, the difference between the
2 groups at d 3 was significant only for threonine (P ¼ 0.003)
and proline (P ¼ 0.033). When expressed as a proportion of the
i.g. infusion rate, the PDV net flux of amino acids at d 24
showed a classical pattern, with a net production of glycine and
alanine and a low portal apparent recovery of aspartate and
glutamate (Table 2). The apparent recovery of indispensable
amino acids ranged from 75 to 95%, except for methionine,
which was only 36%. The TNBS challenge in the control group
induced a decrease in the portal apparent recovery of dietary
proline (P ¼ 0.020) and threonine (P ¼ 0.045).
Threonine fluxes.
Whatever the tracer used in calculation,
the whole-body flux of threonine was not affected by day of
sampling or the treatments (Table 3). The first-pass splanchnic
TABLE 1 Macroscopic and microscopic damage scores and MPO activity in minipig ileum 3 d after
induction of ileitis with TNBS1
Macroscopic score (1–10) Microscopic score (1–20) MPO activity, units/mg tissue
Control minipigs 0 0 0.37 6 0.13
TNBS-treated minipigs 6 (3–8) 8 (2–12) 0.89 6 0.18
1Values are mean scores (ranges) or means 6 SEM, n
¼ 4.
TABLE 2 Portal blood flow and amino acid net recovery in enterally fed minipigs 4 d before and 3 d
after induction of ileitis with TNBS1
I.g. infusion d 24
d 3 ANOVA2
Control TNBS Pd Pt Pd3t
n mmolkg21h21 8 4 4
Blood flow, mLmin21.kg21 30 6 2 24 6 2 39 6 2 0.509 0.014 0.012
Dispensable amino acids % of i.g. infusion
Tyrosine 27 6 1 84 6 5 63 6 8 51 6 9 0.002 0.691 0.148 Cystine 33 6 1 46 6 4 42 6 3 38 6 7 0.373 0.566 0.630 Glycine 88 6 2 146 6 15 125 6 13 138 6 26 0.293 0.539 0.049 Aspartate 99 6 2 16 6 4 17 6 4 14 6 4 0.950 0.206 0.194 Alanine 111 6 2 302 6 30 291 6 17 259 6 41 0.187 0.310 0.284 Arginine 113 6 3 77 6 8 63 6 6 64 6 7 0.022 0.983 0.900 Proline 143 6 3 95 6 4 88 6 6 60 6 9 0.016 0.245 0.020 Serine 156 6 3 80 6 8 73 6 3 53 6 5 0.047 0.682 0.075 Glutamate 380 6 8 17 6 3 15 6 3 6 6 1 0.018 0.407 0.030
Indispensable amino acids
Histidine 42 6 1 84 6 8 84 6 8 53 6 4 0.174 0.129 0.174 Methionine 66 6 2 36 6 5 31 6 3 30 6 6 0.127 0.881 0.458 Phenylalanine 139 6 3 82 6 6 75 6 7 70 6 6 0.178 0.990 0.446 Lysine 157 6 4 93 6 10 83 6 7 68 6 13 0.069 0.710 0.313 Isoleucine 163 6 4 81 6 6 71 6 12 58 6 11 0.023 0.974 0.045 Valine 182 6 4 78 6 9 76 6 13 72 6 10 0.336 0.774 0.083 Threonine 193 6 4 75 6 5 78 6 8 51 6 6 0.078 0.307 0.016 Leucine 326 6 7 79 6 5 78 6 10 56 6 7 0.004 0.404 0.006
1Values are means 6 SEM, n¼ 4 (d 3) or 8 (d 24).
2Data were analyzed by a mixed-model ANOVA with subjects as random effect and day, group, day 3 group as factors. P
d, P for a day
effect; Pt, P for a group effect; PdxtP for an interaction between days and groups.
extraction of dietary threonine was ;37% and was not
significantly affected by the TNBS challenge. Similarly, the
first-pass extractions of threonine by the PDV and by the liver
were not affected and accounted for 27 and 10% of the dietary
threonine, respectively. The TNBS challenge induced a 5-fold
increase in PDV uptake of arterial threonine, which increased
from 2 to 12% of the arterial influx. In the TNBS-treated group,
the contribution of the PDV to the whole-body flux of threonine
increased from 13 to 58% between d 24 and d 3 (P ¼ 0.001),
whereas it was not affected in the control group (P ¼ 0.354). The
PDV release of nondietary threonine on d 3 was 2-fold greater in
the TNBS-treated group than in the control group.
Ileal mucin synthesis.
Finally, the number of mucin-containing
goblet cells was greater (t test, P , 0.05) in the treated area of
the TNBS-treated group (220 6 14 goblet cells/mm intestine)
than in the control group (168 6 28 goblet cells/mm intestine).
The FSR of ileal mucins in the TNBS-treated group (114 6
15%/d) was nearly twice that of the control group (61 6 8%/d;
P ¼ 0.021). The amino acid composition of collected mucins
was not affected by ileitis (P . 0.10). Threonine accounted
for 10.4 6 2.3% of the total amino acid content of mucins
(Supplemental Table 3).
Discussion
Threonine, a nutritionally indispensable amino acid, seems to be
of critical importance for gut function given its high occurrence
in the mucins secreted along the length of the digestive tract (1).
In the present study, we demonstrated that increased mucin
synthesis in the case of intestinal inflammation substantially and
specifically affects the gastrointestinal tract uptake of threonine.
Determination of the quantitative utilization of threonine by
the gastrointestinal tract was not possible in humans, so we used
the minipig as an animal model for adult humans. In this model,
as previously observed for calves (19) and pigs (20), threonine
accounted for 10% of the total amino acids in intestinal mucins.
This contribution was much lower than that reported in humans
(2). Nevertheless it is still high compared with other body
proteins such as muscle proteins (5–6% of threonine). The
experimental approach combined the arterio-venous difference
technique with the infusion of labeled threonine. Simultaneous
infusion of different tracers in the systemic circulation and
stomach allowed differentiation of the unidirectional fluxes
(uptake, release) that constitute the net flux at the level of the
PDV. To determine the effect of a local inflammation of the
intestine, we developed an experimental model of ileitis. This
model, involving luminal administration of TNBS dissolved in
ethanol, was initially set up to induce recto-colic inflammation
in rats (21) and was thereafter adapted to produce ileitis in dogs
and pigs (12,22). The approach used in the present study was
similar to the one reported in dogs (12), but the tube was inserted
via a permanent external fistula in the distal ileum instead of
using the rectal route. Macroscopic and histological examination
demonstrated the effectiveness of the procedure in ileitis induction.
The PDV first-pass extraction of dietary threonine in the
present study with adult minipigs was low (30%) compared with
the one in growing pigs on solid food (50%) and in enterally
fed piglets (70%) (4,5). Besides the type of feeding, the
phy-siological development stage could be a major determinant of
this pass extraction. Indeed, in humans, splanchnic
first-pass extraction of leucine, another indispensable amino acid,
decreases from 50% in neonates to 25% in adults (23,24). In
preterm infants receiving enteral nutrition, as in enterally fed
piglets, a first-pass splanchnic extraction of threonine of 70%
was reported (25). Unfortunately, no measurement of first-pass
splanchnic extraction of threonine is available in adult humans,
but our data suggest that it could be much lower than previously
thought. The high threonine requirement of neonates could be
TABLE 3 Threonine whole body flux, and threonine fluxes across the PDV and the distal small
intestine in enterally fed minipigs 4 d before and 3 d after induction of ileitis1
d 3 ANOVA3
Item2 d 24 Control TNBS Pd Pt Pd3t
n 8 4 4
I.g. infusion, mmolkg21h21
Threonine 194.5 6 6.6 184.2 6 10.8 199.1 6 2.3 0.217 0.221 0.651
15N-Threonine 3.8 6 0.1 4.1 6 0.1 4.1 6 0.2 0.047 0.437 0.283
Whole-body flux, mmolkg21h21
Raiv 302.0 6 15.1 288.7 6 29.1 302.5 6 17.8 0.537 0.492 0.415 PDV flux, mmolkg21h21 FnetThr 144.2 6 9.3 149.8 6 6.7 102.1 6 12.5 0.043 0.422 0.003 Capt Thrart 41.8 6 9.4 24.9 6 14.3 171.0 6 34.6 0.005 0.040 0.001 Capt Thrlum 55.7 6 7.7 52.3 6 9.3 53.0 6 11.4 0.115 0.957 0.385 Ap Thrlum 142.6 6 7.7 136.0 6 10.1 150.2 6 9.4 0.907 0.275 0.303 Ap Threndo 43.4 6 9.3 38.6 6 10.5 123.0 6 18.6 0.012 0.057 0.008
First-pass extraction, % of i.g. infusion
Splanchnic 33.6 6 4.8 38.7 6 6.5 39.2 6 7.9 0.155 0.852 0.490
Portal 28.0 6 3.6 27.6 6 4.3 25.9 6 5.4 0.168 0.704 0.183
Hepatic 5.7 6 1.7 11.0 6 2.9 13.2 6 4.3 0.083 0.780 0.704
1Values are means 6 SEM, n
¼ 4 (d 3) or 8 (d 24).
2Ra
iv, threonine whole body flux calculated from intravenous tracer; FnetThr, threonine net flux; Capt Thrart, arterial threonine uptake; Capt
Thrlum, luminal threonine uptake; Ap Thrlum, luminal threonine release; Ap Threndo, non-luminal threonine release. 3Data were analyzed by a mixed-model ANOVA with subjects as random effect and day, group, day 3 group as factors. P
d, P for a day
effect; Pt, P for a group effect; PdxtP for an interaction between days and groups.
724 Re´mond et al.
related to enhanced mucin synthesis, because the neonate’s
acquired immune system is not fully functional in the intestine
and the neonate is dependent on the innate defenses of mucus
(26).
PDV, liver, and consequently splanchnic first-pass extraction
of dietary threonine were not affected by ileitis. The lack of
response of the liver in terms of threonine utilization agrees with
a focused inflammation in the digestive tract. Liver first-pass
extraction of threonine accounted for only 20% of the total
splanchnic extraction. This value is in agreement with that in
growing pigs (4), confirming the prominent role of the PDV in
the splanchnic first-pass extraction of dietary threonine. Because
the enteral solution used in the present study contained only free
amino acids, the largest part of the infused threonine was
probably rapidly absorbed in the proximal part of the small
intestine. Under these conditions, ileitis did not affect PDV
first-pass extraction of dietary threonine, suggesting that neither
absorption processes nor dietary threonine utilization by this gut
segment was affected.
Ileitis produced a 5-fold increase in PDV uptake of arterial
threonine. Whereas the majority of the utilized threonine derived
from the intestinal lumen in noninflamed minipigs, as previously
observed in piglets and growing pigs fed normal-protein diet
(3,4), ;80% of the threonine taken up by the PDV was derived
from arterial supply during the course of ileitis. Such a large
increase cannot be related to only the small inflamed area. Ileitis
through systemic or nervous signals could have induced an
upregulation of intestinal protein synthesis, and mucin synthesis
in particular, upstream and downstream from the inflamed area.
Different pathways could be involved in the metabolic fate of
the threonine taken up by the PDV, depending on the tissue.
Indeed, the intestine does not seem to oxidize threonine, whereas
the pancreas, which is drained by the portal vein, can degrade
threonine through the
L-threonine 3-dehydrogenase pathway
(3,27). In agreement, it was shown that PDV oxidize some of the
threonine that is taken up from the arterial delivery, but this
visceral threonine oxidation accounts for ,15% of the
whole-body threonine oxidation (4). The major metabolic fate of
threonine in the PDV is therefore protein synthesis and, more
particularly, mucin synthesis. Threonine utilization for mucin
synthesis could have been relatively low in our specific
condi-tions, because our minipigs received a low-fiber, nonprotein
enteral solution and it is has been reported that intestinal mucin
synthesis and secretion are stimulated by nutritional factors,
such as fiber or dietary peptides (28,29). Indeed, the mucin FSR
in the ileum was low (61%/d for the control group on d 24)
compared with the values reported for conventionally fed pigs
(20) or rats (30) (;140%/d). Nevertheless, a large part of the
increase in threonine utilization by the PDV in the course of
ileitis was probably related to the increased mucin synthesis.
This was supported by the specific increase in PDV retention of
threonine, serine, proline, glutamate, and leucine, which are
major constitutive amino acids of mucins.
The present study showed that in noninflamed minipigs,
;20% of the threonine that is released in the portal blood is not
of dietary origin. This endogenous release of threonine has 2
possible origins: reabsorption of threonine of secreted proteins
or proteolysis in gastrointestinal tissues. Because degradation by
the digestive enzymes of part of the mucins secreted in the upper
part of the gut is conceivable, a recycling of threonine is likely.
However, such recycling was not evident in the portal blood of
enterally fed piglets (5), suggesting that either mucins are
particularly resistant to digestive proteases or threonine released
from mucin degradation is immediately reincorporated into
intestinal proteins. Thus, although in the present work it was not
possible to distinguish its origin, the increased endogenous
release of threonine over the time course of ileitis suggested an
increased proteolysis in gastrointestinal tissues.
Thus, the main change in threonine trafficking within the
splanchnic area induced by ileitis was the increase in PDV uptake
of arterial threonine, which increased from 10 to 60% of the
whole-body flux of threonine. Because the latter was not
affected by ileitis, the increase in gastrointestinal utilization of
arterial threonine would be accompanied by a proportional
decrease in threonine utilization by peripheral tissues.
In conclusion, our data indicate that during acute ileitis,
the gastrointestinal tract requirement for threonine is largely
increased and that in the course of enteral nutrition containing
free amino acids, the arterial supply is the preferential source of
threonine used to meet the additional need. Because mucosal and
mucin protein synthesis in the proximal small intestine increases
with luminal threonine concentration (31), increased enteral
threonine supply would probably increase intestinal first-pass
extraction of threonine, limiting the availability of additional
dietary threonine for gut segments that do not benefit from
luminal threonine supply (stomach, ileum, large intestine). Thus,
during acute intestinal inflammation, the optimal solution could
be a combined approach using enteral nutrition (with adequate
threonine supply) to maintain intestinal mass and integrity (26)
and parenteral nutrition to supply additional threonine to meet
the increased PDV requirement. Finally, the increased threonine
requirement in our study during conditions of local gut
inflam-mation, only a 20-cm segment in the ileum, suggests this increase
would be even more dramatic in clinical conditions.
Acknowledgments
We thank D. Durand for surgeries in minipigs, C. Lafarge for
minipig care, P. Denis and J. Gimonet for their help in blood
sampling and biochemical analysis, C. Mettraux and J.
Vuichoud for purification and amino acid analyses of mucins,
and A. F. Mermoud for MS measurements.
Literature Cited
1. Reeds PJ, Burin DG, Stoll B, van Goudoever JB. Consequences and regulation of gut metabolism. In: Lobley GE, White A, MacRae JC, editors. Proceedings of the VIIIth International Symposium on Protein metabolism, and Nutrition; 1999 Sept 1–4; Aberdeen, UK. Wageningen (Netherlands): Wageningen Press; 1999. p. 127–53.
2. Roberton AM, Rabel B, Harding CA, Tasman-Jones C, Harris PJ, Lee SP. Use of the ileal conduit as a model for studying human small intestinal mucus glycoprotein secretion. Am J Physiol. 1991;261:G728– 34.
3. Le Floc’h N, Se`ve B. Catabolism through the threonine dehydrogenase pathway does not account for the high first-pass extraction rate of dietary threonine by the portal drained viscera in pigs. Br J Nutr. 2005;93:447–56.
4. Schaart MW, Scierbeek H, Van der Scoor SRD, Stoll B, Burrin DG, Reeds PJ, van Goudoever JB. Threonine utilization is high in the intestine of piglets. J Nutr. 2005;135:765–70.
5. Van der Schoor SRD, Reeds PJ, Stoll B, Henry JF, Rosenberger JR, Burrin DG, van Goudoever JB. The high metabolic cost of a functional gut. Gastroenterology. 2002;123:1931–40.
6. Gaudichon C, Bos C, Morens C, Petzke KJ, Mariotti F, Everwand J, Benamouzig R, Dare´ S, Tome´ D, et al. Ileal losses of nitrogen and amino acids in humans and their importance to the assessment of amino acid requirements. Gastroenterology. 2002;123:50–9.
7. Corfield AP, Myerscough N, Longman R, Silvester P, Arul S, Pignatelli M. Mucins and mucosal protection in the gastrointestinal tract: new prospects for mucins in the pathology of the gastrointestinal disease. Gut. 2000;47:589–94.
8. Faure M, Moe¨nnoz D, Montigon F, Fay LB, Breuille´ D, Finot PA, Balle`vre O, Boza J. Mucin production and composition is altered in dextran sulfate sodium-induced colitis in rats. Dig Dis Sci. 2003;48: 1366–73.
9. Faure M, Chone´ F, Mettraux C, Godin JP, Be´cherau F, Vuichoud J, Papet I, Breuille´ D, Obled C. Threonine utilization for synthesis of acute phase proteins, intestinal proteins, and mucins is increased during sepsis in rats. J Nutr. 2007;137:1802–7.
10. Faure M, Mettraux C, Moennoz D, Godin JP, Vuichoud J, Rochat F, Breuille´ D, Obled C, Corthe´sy-Theulaz I. Specific amino acids increase mucin synthesis and microbiota in dextran sulfate sodium-treated rats. J Nutr. 2006;136:1558–64.
11. Goh J, O’Morain CA. Review article: nutrition and adult inflammatory bowel disease. Aliment Pharmacol Ther. 2003;17:307–20.
12. Shibata Y, Taruishi M, Ashida T. Experimental ileitis in dogs and colitis in rats with trinitrobenzene sulfonic acid: colonoscopic and histopath-ologic studies. Gastroenterol Jpn. 1993;28:518–27.
13. Rodriguez-Cabezas ME, Galvez J, Lorente MD, Concha A, Camuesco D, Azzouz S, Osuna A, Redondo L, Zarzuelo A. Dietary fiber down-regulates colonic tumor necrosis factor a and nitric oxide production in trinitrobenzenesulfonic acid-induced colitic rats. J Nutr. 2002;132: 3263–71.
14. Millar AD, Rampton DS, Chander CL, Claxson AWD, Blades S, Coumbe A, Panetta J, Morris CJ, Blake DR. Evaluating the antioxidant potential of new treatments for inflammatory bowel disease using a rat model of colitis. Gut. 1996;39:407–15.
15. Krawisz JE, Sharon P, Stenson WF. Quantitative assay for acute intes-tinal inflammation based on myeloperoxidase activity. Gastroenterology. 1984;87:1344–50.
16. Van Eijk HMH, van der Heidjen AH, van Berlo CLH, Soeters PB. Fully automated liquid-chromatographic determination of amino acids. Clin Chem. 1988;34:2510–3.
17. Rosenblatt J, Chinkes D, Wolfe M, Wolfe RR. Stable isotope tracer anaylsis by GC-MS, including quantification of isotopomer effects. Am J Physiol. 1992;263:E584–96.
18. Godin JP, Faure M, Breuille´ D, Hopfgartner G, Fay LB. Determination of13C isotopic enrichment of valine and threonine by GC-C-IRMS after
formation of the N(O,S)-ethoxycarbonyl ethyl ester derivatives of the amino acids. Anal Bioanal Chem. 2007;388:909–18.
19. Montagne L, Toulec R, Lalle`s JP. Calf intestinal mucin: isolation, partial characterization, and measurement in ileal digesta with enzyme-linked immunosorbent assay. J Dairy Sci. 2000;83:507–17.
20. Libao-Mercado AJ, de Lange CFM. Refined methodology to purify mucins from pig colonic mucosa. Livest Sci. 2007;109:141–4. 21. Morris GP, Beck PL, Herridge MS, Depew WT, Scewczuk MR, Wallace
JL. Hapten-induced model of chronic inflammation and ulceration in the rat colon. Gastroenterology. 1989;96:795–803.
22. Merritt AM, Buergelt CD, Sanchez LC. Porcine ileitis model induced by TNBS-ethanol instillation. Dig Dis Sci. 2002;47:879–85.
23. Beaufre`re B, Fournier V, Salle B, Putet G. Leucine kinetics in fed low-birth-weight infants: importance of splanchnic tissues. Am J Physiol Endocrinol Metab. 1992;263:E214–20.
24. Boirie Y, Gachon P, Beaufre`re B. Splanchnic and whole-body leucine kinetics in young and elderly men. Am J Clin Nutr. 1997;65:489–95. 25. Van der Schoor SRD, Wattimena DL, Huijmans J, Vermes A, van
Goudever JB. The gut takes nearly all: threonine kinetics in infants. Am J Clin Nutr. 2007;86:1132–8.
26. 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: G1293–301.
27. Le Floc’h N, Thibault JN, Se`ve B. Tissue localization of threonine oxidation in pigs. Br J Nutr. 1997;77:593–603.
28. Satchithanandam S, Vargofcak-Apker M, Calvert RJ, Leeds AR, Cassidy MM. Alteration of gastrointestinal mucin by fiber feeding in rats. J Nutr. 1990;120:1179–84.
29. Claustre J, Toumi F, Trompette A, Trompette A, Jourdan J, Guinard H, Chayvialla JA, Plaisancie´ P. Effects of peptides derived from dietary proteins on mucus secretion in rat jejunum. Am J Physiol Gastrointest Liver Physiol. 2002;283:G521–8.
30. Faure M, Moe¨nnoz D, Montigon F, Fay LB, Breuille´ D, Finot PA, Balle`vre O, Boza J. Development of a rapid and convenient method to purify mucins and determine their in vivo synthesis rate in rats. Anal Biochem. 2002;307:244–51.
31. Nichols NL, Bertolo RF. Luminal threonine concentration acutely affects intestinal mucosal protein and mucin synthesis in piglets. J Nutr. 2008;138:1298–303.
726 Re´mond et al.
