612 Gut 1997; 41: 612–618 Specific adaptation of gastric emptying to diets

with diVering protein content in the rat: is Gut: first published as 10.1136/gut.41.5.612 on 1 November 1997. Downloaded from endogenous implicated?

G Shi, V Leray, C Scarpignato, N Bentouimou, S Bruley des Varannes, C Cherbut, J-P Galmiche

Abstract low protein diet (p<0.05) but not in rats on Background—Recent studies indicate that a high protein diet (p>0.05). As a conse- gastric emptying may be influenced by quence, postprandial plasma levels of patterns of previous nutrient intake. En- CCK in rats fed with a medium or low dogenous cholecystokinin (CCK), whose protein diet were significantly (p<0.05) synthesis and release can be aVected by higher than those in rats receiving a high dietary intake, has a major role in the protein diet. In rats on high and low regulation of gastric emptying. protein diets, dose response curves to Aims—To evaluate the influence of diets CCK-8 were virtually identical, suggesting with diVering protein content on gastric that dietary protein intake has no influ- emptying of diVering liquid test meals and ence on the eVect of exogenous CCK. plasma CCK levels in the rat and to check Conclusions—These results clearly show whether the inhibitory eVect of exogenous that gastric emptying of a protein contain- CCK on gastric emptying is modified after ing meal can be modified by previous long term intake of diets with diVering dietary protein intake. This eVect, which protein content. is time dependent and meal specific, may Methods—Rats were fed for three weeks be related to changes in endogenous CCK with high protein, medium protein (regu- release which will aVect emptying rate. lar), or low protein diet. On day 22 gastric While the exact mechanisms underlying emptying of a peptone meal was studied. this adaptive response need to be studied In addition, basal and postprandial CCK and clarified further, these results empha- levels after the diVerent dietary regimens sise the importance of dietary history in were measured by bioassay. The time

the evaluation and interpretation of gas- http://gut.bmj.com/ course of dietary adaptation was studied tric emptying data. and its specificity assessed through the use (Gut 1997; 41: 612–618) of diVerent (peptone, glucose, and methyl- cellulose) test meals. The eVect of exog- Keywords: diet; protein content; gastric emptying; enous CCK-8 on gastric emptying was cholecystokinin studied at the end of the adaptation period (three weeks). Results—Feeding the animals with a high Dietary factors, including the volume, osmo- on October 1, 2021 by guest. Protected copyright. protein diet for three weeks resulted in a larity, viscosity, pH, and content of the meal are significant (p<0.05) acceleration (by 21.2 important in the physiological control of 1 (8.2)%) of gastric emptying while feeding gastric emptying. Recent studies indicate that with a low protein diet was followed by a gastric emptying may actually be influenced by significant (p<0.05) delay (by 24.0 (6.2)%) patterns of previous nutrient intake. In hu- in the emptying rate. When the time mans, both gastric emptying and mouth to course of the eVect of dietary adaptation caecum transit are faster after a period on a on gastric emptying was studied, it ap- high fat diet than after the same period on a low Human Nutrition 2 Research Centre and peared that at least two weeks are required fat diet. Supplementation of the diet with glu- Department of for dietary protein to be eVective. The cose for several days accelerates the emptying Gastroenterology and regulatory eVect of dietary protein on of a hyperosmotic drink of glucose3 or the Hepatology, University gastric emptying proved to be dependent emptying of glucose and fructose4 from the of Nantes, Nantes, human stomach. Conversely, gastric emptying France on meal composition. Only the emptying G Shi rate of a protein containing meal (40% of glucose is delayed by a four day fast in V Leray peptone) was significantly modified by humans.5 Results obtained from animal experi- C Scarpignato previous dietary intake. No significant ments, performed in rats, have also shown that N Bentouimou (p>0.05) changes were observed with glu- dietary restriction results in a significant delay S Bruley des Varannes cose and methylcellulose meals whose in gastric emptying, which returns to normal C Cherbut 67 J-P Galmiche emptying rates were similar in rats receiv- after a period of free feeding. In both studies, ing a high protein or low protein diet. A however, restriction of energy intake was Correspondence to: peptone meal strongly and significantly excessive enough to result in malnutrition. The Professor Galmiche. (p<0.05) increased plasma CCK levels in eVect of changes in meal content (without any Accepted for publication rats fed a medium protein (regular) diet. restriction of energy intake) on gastric empty- 20 February 1997 Results were similar in rats receiving a ing has not yet been studied in the rat. Dietary adaptation of gastric emptying in rats 613

TABLE 1 Composition of the diets 12 hour light dark cycle, with lights on at 0800 and oV at 2000, and were fed with three diVer- Low protein diet Medium protein diet High protein diet Ingredient (g/100 g diet) (g/100 g diet) (g/ 100 g diet) ent diets (table 1) during diVerent periods. During the course of the experiments, food Gut: first published as 10.1136/gut.41.5.612 on 1 November 1997. Downloaded from Proteins 9.2 17.0 55.9 consumption and body weights were recorded Casein 3.3 50.0 Other proteins* 5.9 17.0 5.9 daily. Carbohydrates 75.4 58.7 28.7 Dextrin 55.8 9.1 Other carbohydrates† 19.6 58.7 19.6 MEASUREMENT OF GASTRIC EMPTYING Lipids 4.3 3.0 4.3 Gastric emptying was measured by the method Fibre 1.4 4.0 1.4 of Scarpignato 21 22 with some modifica- Water 4.0 12.0 4.0 et al Minerals 5.0 5.0 5.0 tions. Three diVerent test meals were used: two Vitamin mixture 0.7 0.3 0.7 nutrient (40% peptone and 25% glucose) and Energy content (MJ 100 g) 1.58 1.38 1.58 one non-nutrient (1.5% methylcellulose) *Other proteins include animal (fish) protein and vegetable (soya bean meal, yeast) protein at a meals. Both peptone and glucose were dis- ratio of 2:7. †Other carbohydrates are from cereals and cereal byproducts (including 13% starch). solved in distilled water to make their final High and low protein diets were prepared by adding casein and dextrin to the medium protein diet concentration; phenol red (1.0 mg/ml) was so that they contained 3.3% and 50% casein, respectively. Appropriate quantities of lipids, miner- als and the vitamin mixture were added. The medium protein diet was rat and mouse maintenance then added as a non-absorbable marker. These diet A-04 from UAR, France. two test meals were prepared a few days before experiments and stored at 4°C. The non- There is growing evidence in the literature nutrient test meal (the methylcellulose meal) suggesting the involvement of endogenous was prepared as previously described.21 The cholecystokinin (CCK) in the regulation of 8–15 pH values of 40% peptone, 25% glucose, and gastric motor function. On the other hand, 1.5% methylcellulose were 6.9, 7.0, and 7.2, CCK synthesis and release can be aVected by respectively. The osmolarity was 1840 dietary intake in both physiological and patho- mosmol/kg for 25% glucose and 8 mosmol/kg logical conditions. In elderly patients with pro- for 1.5% methylcellulose. The 40% peptone tein energy undernutrition, postprandial was actually a suspension of peptone in plasma CCK levels were found to be signifi- distilled water whose osmolarity cannot be cantly higher than those found in healthy (both measured. young and aged) subjects, suggesting that the All the prewarmed (37°C) meals (3 ml for hypersecretion is related to malnutri- 16 each rat) were given orally through a stainless tion instead of to aging. On the contrary, in steel tube after an 18 hour fast. Forty (for pep- obese patients, the plasma CCK response to tone test meal), 20 (for glucose test meal), or sham feeding was significantly lower when 15 (for methylcellulose meal) minutes later, compared with that found in normal weight 17 rats were killed by cervical dislocation. The (age and sex matched) subjects. When stomach was then ligated at the pylorus and compared with healthy controls, in patients cardia, opened along the greater curvature, and with anorexia nervosa, an earlier and greater gastric contents collected in graduated tubes. http://gut.bmj.com/ rise in plasma CCK levels after a liquid meal The stomach was carefully rinsed three times was observed, a pattern which was normalised 18 with 30 ml of distilled water and the rinsing with nutrition therapy. Animal experiments solutions were added to gastric contents. The have shown that—during adaptation to in- mixture was made up to 100 ml with water and creased dietary proteins—there is a notable, thoroughly mixed. Five millilitres of this although transient, increase in plasma CCK ° 19 mixture was centrifuged at 3100 rpm at 4 C for levels. On the contrary, after food deprivation, 15 minutes. Supernatant (1 ml) was then plasma CCK levels decrease rapidly and this on October 1, 2021 by guest. Protected copyright. added to 5 ml Na3PO4 (63.7 g/l) and vortexed. decrease is followed by a reduction in duodenal The phenol red concentration in each sample CCK mRNA and mucosal CCK peptide levels. (Csample) was determined spectrophotometri- Only one day of refeeding was suYcient to cally at 523 nm and gastric emptying calcu- restore to control levels all the above lated according to the following formula: parameters.20 No study has, however, investi- gated the eVects of changes in endogenous = Cstandard−Csample × CCK levels, induced by chronic intake of Gastric emptying % 100% Cstandard diVerent concentrations of protein, on gastric

emptying. where Cstandard represents the phenol red con- The aim of present study was therefore two- centration found in the stomach of rats fold: to evaluate the influence of diets with dif- sacrificed immediately after administration of fering protein content on gastric emptying of the test meal. diVerent liquid test meals, plasma CCK levels, and CCK mRNA in duodenal mucosa of rats; EXPERIMENTAL DESIGN and to check whether the inhibitory eVect of EVect of type of protein intake on gastric emptying exogenous CCK on gastric emptying is modi- In order to check whether diets with diVering fied after long term intake of diets with diVer- protein content could aVect emptying rate of a ing protein content. protein containing (peptone) meal, three sets of experiments were performed. Materials and Methods In the first set, three groups of rats (six to ANIMALS AND DIETS eight animals) were fed for 21 days with a high Male Wistar rats (aged 6–7 weeks, weighing protein, medium protein (regular), or low pro- 180–210 g at the start of the experiment) were tein diet (see table 1). On day 22, gastric emp- used in this study. They were maintained on a tying of the peptone test meal was measured. In 614 Shi, Leray, Scarpignato, Bentouimou, Bruley des Varannes, Cherbut, et al

order to check whether the modification of were centrifuged at 3500 rpm for 10 minutes at gastric emptying is reversible, two groups of 4°C. The plasma was applied onto Sep-Pak rats were fed for 21 days with a high protein or C18 Cartridges (Waters Associates, Millipore low protein diet, respectively, and then changed Corp., Milford, Molsheim, France), previously Gut: first published as 10.1136/gut.41.5.612 on 1 November 1997. Downloaded from to a medium protein (regular) diet for seven activated with 10 ml acetonitrile, 10 ml days. Thereafter, gastric emptying of the 40% ethanol, and 20 ml water. After washing with peptone meal was measured. 20 ml water, CCK was eluted with 1 ml In a second set of experiments, the time acetonitrile: water (1:1). This eluate was then course of dietary adaptation was studied. Eight lyophilised for 72 hours and the corresponding groups of animals were fed with a high or low extract stored at −20°C until assayed protein diet for 7, 14, 21, or 30 days and gastric Acini were isolated by collagenase digestion emptying of a peptone meal was quantified at of the pancreas. They were then suspended in a the end of the adaptation period. Tris buVer containing 40 mM Tris, 115 mM

Finally, in order to check whether the dietary NaCl, 0.5 mM CaCl2, 1.2 mM MgSO4, and

adaptation to protein aVects “specifically” gas- 1.2 mM KH2PO4 supplemented with 5 mM tric emptying of a protein test meal, the glucose, 2 mM glutamine, 2% of an essential influence of 21 days feeding with diets of amino acid solution, 1% of a non-essential diVering protein content on gastric emptying of amino acid solution, and 5 mg/ml bovine

nutrient or non-nutrient meals was investi- serum albumin, equilibrated with O2:CO2 gated. Six groups of rats were fed for three (95:5), and adjusted to pH 7.4 with HCl. weeks with a high or low protein diet and gas- Plasma CCK extracts were added to 1 ml tric emptying of a peptone, glucose, or methyl- aliquots of acinar suspension and incubated for cellulose meal was measured. 30 minutes at 37°C. Amylase release into the medium was measured using procion yellow EVect of diVerent types of protein intake on CCK starch as substrate. The total acinar amylase release content was measured after lysis of cells by To evaluate the eVect of dietary protein content Triton X-100. Amylase release, expressed as a on CCK release, six groups of rats (of six to percentage of total amylase content, was quan- nine animals each) were studied. They were fed tified using a standard curve constructed with with a low, medium (regular), or high protein CCK-8. diet for 21 days and on day 22 both basal plasma CCK levels (three groups) and plasma CCK mRNA ASSAY CCK levels eight minutes after intragastric Duodenal CCK mRNA levels in response to a administration of a protein meal (three groups) peptone test meal after a three week adaptation were measured by bioassay (see below). This to either a high protein diet or a low protein time interval has been chosen since published diet were measured six hours after the test results23 24 and our own preliminary data has meal. This time interval was chosen after some

shown that after meals, the time to peak plasma preliminary work where we observed the high- http://gut.bmj.com/ CCK levels ranges from five to 10 minutes. est mRNA levels six hours after the meal. This value was notably reduced after 24 hours and EVect of diVerent types of protein intake on CCK almost undetectable after 48 hours. mRNA In two other groups of rats fed for three weeks RNA isolation with a high or low protein diet, duodenal Immediately after sacrifice, the duodenum was mucosa was excised and CCK mRNA content removed and the luminal contents rinsed with measured by northern blot analysis (see iced phosphate buVered saline. The duodenum on October 1, 2021 by guest. Protected copyright. below). was then incised and the mucosa scraped oV and immediately frozen in liquid nitrogen until EVect of diVerent types of protein intake on RNA isolation. Total RNA was isolated from CCK-8 induced delay in gastric emptying duodenal mucosal scrapings by the acid To evaluate whether diets with diVering protein guanidinium thiocyanate-phenol-chloroform content could aVect the gastric inhibitory extraction method.25 Duodenal scrapings were activity of exogenous CCK-8, a dose response homogenised in 1 ml of 4 M guanidium curve (percentage inhibition versus dose) of the thiocyanate, 25 mM sodium citrate (per 100 peptide was constructed in animals fed for mg of tissue), 0.1 M 2-mercaptoethanol and three weeks with a high or low protein diet. 0.5% sarcosyl. Then, in order, 0.1 ml of 2 M Eight groups of rats were given the diets with sodium acetate pH 4, 1 ml of phenol saturated diVering protein content and, at the end of the in water, and 1 ml of chloroform:isoamylic adaptation period, gastric emptying of a alcohol (49:1) were added. The homogenate non-caloric liquid meal (methylcellulose) was was kept at 4°C for 15 minutes and centrifuged measured after saline or three diVerent doses of at 4°C for 20 minutes at 13 000 rpm. The exogenous CCK-8, injected intraperitoneally aqueous phase, containing RNA, was precipi- five minutes before the meal. tated with 1 ml of isopropanol for one hour at –20°C and centrifuged at 4°C for 20 minutes at CCK BIOASSAY 13 000 rpm. Pellets were dissolved with the Plasma levels of CCK were measured by the guanidinium thiocyanate solution, then pre- specific and sensitive bioassay described by cipitated with isopropanol for at least one hour Liddle et al.23 This method is based on the abil- at –20°C. After centrifugation for 10 minutes at ity of CCK to stimulate amylase release from 13 000 rpm, RNAs were washed with 75% isolated rat pancreatic acini. Blood samples ethanol, dried under vacuum, and dissolved in Dietary adaptation of gastric emptying in rats 615

water. The concentration of RNA was det- 30 ermined spectrophotometrically. RNAs were concentrated to obtain 20 µg per 7 µl. ** 25 * Gut: first published as 10.1136/gut.41.5.612 on 1 November 1997. Downloaded from

Northern blot analysis 20 ** * Total RNA (20 µg) was denatured for 15 min- ° High protein diet utes at 60 Cin23µlofbuVer (20 mM * * morphomolinopropane sulphonic acid, 5 mM 15 ** Medium protein diet Food intake (g/day) Low protein diet sodium acetate, and 1 mM EDTA) containing 0 50% formamide (vol/vol) and 6% formalde- 123 8 15 21 hyde (vol/vol). Time course (day) RNA samples were then electrophoresed in 1% agarose and 6% formaldehyde gel, and Figure 1: Food intake of rats fed with a high, medium, or low protein diet. Points represent the means. Vertical bars transferred to Hybond N nylon membranes are standard errors. *p<0.05 versus medium protein diet, (Amersham, Les Ulis, France). The nylon **p<0.01 versus medium protein diet. membranes were fixed in ultraviolet light for eight minutes, prehybridised for at least one Results hour at 42°C in hybridisation buVer containing BODY WEIGHT AND FOOD CONSUMPTION DURING 50% formamide (vol/vol), 0.08% bovine serum DIFFERENT DIETARY REGIMENS albumin (wt/vol), 0.08% Ficoll (wt/vol), 0.08% Body weights of rats fed with diets with diVer- polyvinyl pyrolidone (wt/vol), 50 mM phos- ing protein content were not dissimilar at any time interval during the adaptation periods phate buVer pH 6.5, 900 mM NaCl, 1% (data not shown). Animals fed with a low pro- sodium dodecyl sulphate (SDS) (wt/vol), 4% tein diet ate more and animals on a high dextran sulphate, 2 mM EDTA, and 0.2 mg/ml protein diet ate less than those receiving a of denatured herring sperm DNA. medium protein (regular) diet (fig 1). The membranes were then hybridised for 16 hours at 42°C in the same solution with a EFFECT OF DIFFERENT TYPES OF PROTEIN INTAKE cDNA probe of CCK (American Type Culture ON GASTRIC EMPTYING Collection), labelled by the random primer 32 Under our experimental conditions, the pep- method with P-dCTP. Blots were then tone test meal leaving the stomach (gastric washed twice for 10 minutes in 500 ml of 0.1% emptying) in control rats receiving a medium SDS, 0.5×SSC (20×SSC: 3 M NaCl, 0.3 M protein (regular) diet was 37.0 (1.6)%. Feeding sodium citrate) at room temperature, and twice the animals with a high protein diet for three for 10 minutes in the same solution at 55°C. weeks resulted in a significant acceleration (by Blots were subsequently stripped in a boiling 21.2 (8.2)%) of gastric emptying while feeding

solution of 0.1% SDS and then rehybridised with a low protein diet was followed by a http://gut.bmj.com/ with a 32P-labelled cDNA probe of phos- significant delay (by 24.0 (6.2)%) of emptying phoglycerate kinase (PGK), a housekeeping rate (fig 2). When rats on high or low protein gene. Signals were then quantified by a laser diets were put on a medium protein (regular) densitometer. Densitometric values obtained diet for just one week, gastric emptying with CCK were expressed after normalisation reverted back to normal (no significant diVer- with PGK values. As usual, results were ence was found between emptying rate of each expressed as the ratio of CCK mRNA to PGK of these two groups and that of control mRNA to correct for any diVerences in total animals) and no significant diVerence was on October 1, 2021 by guest. Protected copyright. RNA recovery between samples and any small found between the two groups (high versus low diVerences in the amount of RNA transferred to the Hybond N nylon membranes. 60 * *

CHEMICALS 50 Synthetic sulphated CCK-8 (Bachem, Buben- dorf, Switzerland) was dissolved in 0.9% NaCl 40 solution. Peptone (grade II), glucose, methyl- cellulose, phenol red, and routine chemicals 30 were all purchased from Sigma (St Louis, Mis- souri, USA). 20 Gastric emptying (%) STATISTICAL ANALYSIS As data are normally distributed, values are 10 expressed as mean (SEM). Comparison among means was performed through analysis of vari- 0 High Medium Low ance followed by post hoc tests (Dunnett mul- protein protein protein tiple comparisons test or Bonferroni test). A p diet diet diet value less than 0.05 was considered significant. Figure 2: Gastric emptying of a peptone meal (over 40 All the calculations were made by using the minutes) in rats fed for 21 days with a high, medium, or TM low protein diet. Each column represents the mean of the Instat software (GraphPad Software Inc., values obtained from six to eight animals. Vertical bars are Berkeley, California, USA). standard errors. *p<0.05. 616 Shi, Leray, Scarpignato, Bentouimou, Bruley des Varannes, Cherbut, et al

High protein diet 6.0 Low protein diet ** 60 Basal

* * ** 5.0 Post-prandial Gut: first published as 10.1136/gut.41.5.612 on 1 November 1997. Downloaded from

50 * 4.0 40 3.0 30

2.0

20 Plasma CCK (pmol/l)

Gastric emptying (%) 1.0 10

0 0 7 Days 14 Days 21 Days 30 Days High Medium Low protein protein protein Figure 3: Gastric emptying of a peptone meal (over 40 diet diet diet minutes) in rats fed with a high or low protein diet for 7, Figure 5: Basal and postprandial plasma CCK levels in 14, 21, or 30 days. Each column represents the mean of the rats fed for 21 days with a high, medium, or low protein values obtained from six to eight animals. Vertical bars are diet. Peptone induced CCK release was measured eight standard errors. *p<0.01 versus low protein diet, minutes after the test meal. Each column represents the **p<0.001 versus low protein diet. mean of the values obtained from six to nine animals. Vertical bars are standard errors. *p<0.05 versus basal protein diet: 44.9 (4.0)% versus 43.5 (3.3)%, value, **p<0.01 versus basal value. p>0.05). When the time course of the eVect of dietary EFFECT OF DIFFERENT TYPES OF PROTEIN INTAKE adaptation on gastric emptying was studied, it ON CCK RELEASE appeared evident that at least two weeks are Basal plasma CCK levels in control rats receiv- required for dietary protein to be eVective (fig ing a medium protein (regular) diet were 1.48 3). Indeed, only one week of feeding with diets (0.37) pmol/l and were not significantly diVer- of diVering protein content was not suYcient ent from those found in the plasma of animals to achieve a statistically significant diVerence fed with a high or low protein diet (fig 5). As between the two groups (fig 3). As expected, expected, a peptone meal strongly and signifi- the higher the body weight (and therefore the cantly increased plasma CCK levels in control age) of the animals, the slower the emptying rats. This also occurred in rats receiving a low rate.25 The inverse correlation between age and protein diet but not in rats on a high protein gastric emptying was significant in rats on both diet (fig 5). As a consequence, postprandial

high and low protein diets. plasma levels of CCK in rats fed with medium http://gut.bmj.com/ The regulatory eVect of dietary protein on or low protein diets were significantly higher gastric emptying proved to be dependent on than those found in rats receiving a high meal composition. Indeed, only the emptying protein diet. rate of a protein containing meal (40% peptone meal) was modified by previous dietary intake. EFFECT OF DIFFERENT TYPES OF PROTEIN INTAKE On the contrary, no significant changes were ON CCK mRNA observed with glucose and methylcellulose The postprandial mRNA levels (expressed as meals, whose emptying rates were similar in CCK mRNA to PGK mRNA ratio) in the on October 1, 2021 by guest. Protected copyright. rats receiving a high protein or low protein diet duodenal mucosa were 89.16 (2.95)% after (fig 4). adaptation to a high protein diet and 77.72 (5.36)% after adaptation to a low protein diet, 100 respectively (p=0.09). High protein diet 90 Low protein diet EFFECT OF DIFFERENT TYPES OF PROTEIN INTAKE 80 ON CCK INDUCED DELAY IN GASTRIC EMPTYING 26 70 As previously reported, CCK-8 significantly delayed gastric emptying of a non-caloric 60 liquid meal (methylcellulose meal) and this * 50 eVect proved to be dose dependent (fig 6). In rats on high and low protein diets, dose 40 response curves to CCK-8 were virtually iden- 30 tical, suggesting no influence of dietary protein

Gastric emptying (%) intake on the eVect of exogenous CCK. 20 10 Discussion 0 Results of the present investigation show that Peptone Glucose Methylcellulose gastric emptying of a liquid test meal contain- Figure 4: Gastric emptying of a peptone (over 40 ing protein is significantly faster after a 21 day minutes), glucose (over 20 minutes), or methylcellulose period on a high protein diet; the emptying rate (over 15 minutes) meal in rats fed for 21 days with a high is significantly slower after the same period on or low protein diet. Each column represents the mean of the values obtained from six to eight animals. Vertical bars are a low protein diet. This gastric motor adapta- standard errors. *p<0.01 versus low protein diet value. tion seems to be time dependent since at least Dietary adaptation of gastric emptying in rats 617

80 hormonal mediators. If this is the case, CCK might be the primary candidate since it has 70 been shown to be released from “I” cells by the presence of proteins in the small intestine24 28 Gut: first published as 10.1136/gut.41.5.612 on 1 November 1997. Downloaded from 60 and to be capable of delaying gastric empty- ing.15 26 Thus, if a high protein diet accelerates 50 gastric emptying by down regulating nutrient receptors in the small intestine, one would 40 expect the postprandial plasma CCK response to be reduced. The converse should be true for 30 the eVect of low protein intake. In fact, the High protein diet (y = 38.7 + 55.9 logx; r = 0.96) CCK response to a peptone meal was reduced 20 Low protein diet (y = 43.9 + 46.6 logx; r = 0.99) after feeding with a high protein diet and increased after low protein diet. The diet Inhibition of gastric emptying (%) 10 induced changes in CCK release are consistent with diet induced modifications in emptying 0 0 1.0 2.0 3.0 4.0 5.0 rate observed in our experimental conditions. CCK-8 (µg/kg) In fact, after a peptone meal, the small intesti- Figure 6: Dose response curves of the eVect of CCK-8 on gastric emptying of a nal mucosa of rats fed with low protein diet will methylcellulose meal in rats fed for 21 days with a high or low protein diet. Each point release (within a few minutes) more CCK, represents the mean of the values obtained from six to eight animals. Vertical bars are which will, in turn, inhibit gastric emptying of standard errors. The graph displays a semilogarithmic plot of percentage inhibition versus the remaining meal, so that the observed emp- dose. tying rate will be slower. On the contrary, when 14 days are required for diets to induce signifi- the duodenal mucosa has been desensitised by cant changes in gastric emptying. These feeding with a high protein diet, less CCK will changes could be returned to normal by one be released, so that one of the physiological week of refeeding with a medium protein brakes to emptying of gastric contents will be (regular) diet. Although these gastric motor lessened and, consequently, final gastric emp- changes are associated with an opposite trend tying will seem faster. Although one time point in food intake, any cause-eVect relationship was used for measurement of plasma CCK, seems unlikely since dietary restriction is these data suggest that endogenous CCK may usually followed by delayed, rather than accel- have a role in the adaptive process. Recent data erated, gastric emptying.67 The eVect of from our laboratory seem to confirm this dietary proteins proved to be specific to gastric hypothesis, since pretreatment of animals with emptying of a protein containing meal, because , a selective CCK-A receptor an- only the emptying rate of a peptone meal, but tagonist, was able to prevent the adaptation of gastric emptying to diets with diVering protein not that of a glucose or methylcellulose meal, 27 was modified after feeding with diets of diVer- content. http://gut.bmj.com/ ing protein content. Composition rather than The increase in postprandial CCK levels osmolarity of the test meal seems to confer this after feeding a low protein diet seems to reflect an increase in CCK release rather than CCK specificity. Indeed, as with gastric emptying of synthesis by “I” cells. Indeed, duodenal 40% peptone, the emptying rate of an 8% pep- mucosal levels of CCK mRNA were not statis- tone meal (which has a much lower osmolarity tically aVected by the diVerent protein intake. than a 25% glucose meal) was modified in a 27 These results are at variance from those similar way by previous dietary intake. on October 1, 2021 by guest. Protected copyright. reported by Liddle et al,29 who found CCK The presence of nutrients, such as glucose, release to be associated with changes in CCK protein, or fat in the small intestine exerts a mRNA levels. DiVerences in the experimental powerful regulatory eVect on the rate of gastric 1 design—that is, diVerent stimuli (protein meal emptying. Indeed, gastric emptying is modi- versus soybean trypsin inhibitor) and diVerent fied by previous dietary history. This important time intervals (six hours versus 12 or 24 hours) finding suggests that underexposure or overex- can be responsible for discrepancies in findings. posure of the small intestinal receptors to AdiVerent mechanism to explain the nutrients may alter the sensitivity of the adaptation of gastric emptying to diVerent mechanisms that regulate gastric emptying. types of protein intake could be a modification The changes in the gastric emptying rate, of gastric sensitivity to CCK. Obese rats, for observed in rats fed with a high or low protein instance, show a decreased sensitivity to CCK30 diet, support the hypothesis that small intesti- andithasbeenproposedthatthismaybeaconse- nal regulatory mechanisms had been desensi- quence of increased meal size.31 However, dose tised by a period of exposure to large amounts response curves to exogenous CCK-8 were the of proteins and sensitised by exposure to same in rats fed with high or low protein diets, relatively small amounts of proteins. These suggesting that dietary regimens are unable to changes, which are time dependent and meal aVect gastrointestinal sensitivity to exogenous specific, emphasise the importance of recent CCK. dietary intake in interpreting gastric emptying It is worth mentioning that CCK unrelated data. mechanisms also seem to be involved in the The mechanisms underlying the adaptive adaptive changes of gastric emptying or motil- responses could be mediated by changes in ity to diet since specific adaptation of gastric nutrient receptor density or, alternatively, emptying34or motility32 to glucose (which does could be related to release of, or sensitivity to, not release CCK) and rapid adaptive changes 618 Shi, Leray, Scarpignato, Bentouimou, Bruley des Varannes, Cherbut, et al

in the pyloric motor response to prolonged emptying of liquid nutrient loads in Rhesus monkeys. Am J 33 Physiol 1993; 265: G371–5. intraduodenal dextrose have been reported. 14 Moran TH, Kornbluh R, Moore K, Schwartz GJ. To summarise, gastric emptying of a protein Cholecystokinin inhibits gastric emtying and contracts the containing meal can be modified by previous pyloric sphincter in rats by interacting with low aYnity Gut: first published as 10.1136/gut.41.5.612 on 1 November 1997. Downloaded from receptor sites. Digestion 1994; 52: 165–72. dietary protein intake. This eVect, which is 15 Scarpignato C, Varga G, Corradi C. EVect of CCK and its time dependent and meal specific, seems to be antagonists on gastric emptying. J Physiol (Paris) 1993; 87: 291–300. related to changes in endogenous CCK release 16 Berthelemy P, Bouisson M, Vellas B, Moreau J, Vaysse N, which will aVect emptying rate. While the exact Albarede JL, Fbat A. Postprandial cholecystokinin secre- tion in elderly with protein-energy undernutrition. JAm mechanisms underlying this adaptive response Geriatr Soc 1992; 40: 365–9. need to be studied and clarified further, these 17 Wisen O, Bjorvell H, Cantor P, Johansson C, Theodorsson results emphasise the importance of dietary E. Plasma levels of regulatory in obesity following modified sham feeding (MSF) and a liquid test meal. Regul history in the evaluation and interpretation of Pept 1992; 39: 43–54. gastric emptying data. 18 Harty RF, Pearson PH, Solomon TE, McGuigan JE. Chole- cystokinin, vasoactive intestinal peptide and peptide histidine responses to feeding in anorexia ner- Preliminary data from the present investigation have been vosa. Regul Pept 1991; 36: 141–50. presented to the American Gastroenterological Association 19 Green GM, Levan VH, Liddle RA. Plasma cholecystokinin (New Orleans) and appeared in abstract form (Gastroenterology and pancreatic growth during adaptation to dietary 1994; 106: A566). This work was supported by a grant from the protein. Am J Physiol 1986; 251: G70–4. Institut National de la Santé et Recherche Médicale (INSERM, 20 Kanayama S, Liddle RA. Influence of food deprivation on CRI No. 9508). intestinal cholecystokinin and somatostatin. Gastroenterol- ogy 1991; 100: 909–15. 21 Scarpignato C, Capovilla T, Bertaccini G. Action of 1 Meyer JH. Motility of the stomach and gastrointestinal caerulein on gastric emptying in the conscious rat. junction. In: Johnson LR, ed. Arch Int Physiology of the gastrointestinal 1980; 246: 286–94. tract. 2nd edn. New York: Raven Press, 1987: 613–30. Pharmacodyn Ther 2 Cunningham KM, Daly J, Horowitz M, Read NW. 22 Scarpignato C. Mesure de la vidange gastrique des liquides Gastrointestinal adaptation to diets of diVering fat compo- chez le rat. J Pharmacol (Paris) 1983; 14: 261–8. sition in human volunteers. Gut 1991; 32: 483–6. 23 Liddle RA, Goldfine ID, Williams JA. Bioassay of plasma 3 Cunningham KM, Horowitz M, Read NW. The eVect of cholecystokinin in rats: eVects of food, trypsin inhibitor and short-term dietary supplementation with glucose on gastric alcohol. Gastroenterology 1984; 87: 542–9. emptying in humans. Br J Nutr 1991; 65: 15–9. 24 Liddle RA, Green GM, Conrad CK, Williams JA. Proteins 4 Horowitz M, Cunningham KM, Wishart JM, Jones KL, but not amino acids, carbohydrates, or fats stimulate chole- Read NW. The eVect of short-term dietary supplementa- cystokinin secretion in the rat. Am J Physiol 1986; 251: tion with glucose on gastric emptying of glucose and fruc- 243–8. tose and oral glucose tolerance in normal subjects. Diabeto- 25 Chomszynski P, Sacchi N. Single-step method of RNA iso- logia 1996; 39: 481–6. lation by acid guanidinium thiocyanate-phenol-chloroform 5 Corvilain B, Abramowicz M, Fery F, Schoutens A, extraction. Anal Biochem 1987; 162: 156–9. Verlinden M, Balasse E, Horowitz M. EVect of short-term 26 Scarpignato C, Kisfalvi I, D’Amato M, Varga G. EVect of starvation on gastric emptying in humans: relationship to dexloxiglumide and spiroglumide, two new CCK-receptor oral glucose tolerance. Am J Physiol 1995; 269: G512–7. antagonists, on gastric emptying and secretion in the rat: 6 Desai AJ, Moyer LL, Deveney CW, Sankaran H. Malnutri- evaluation of their receptor selectivity in vivo. Aliment Phar- tion significantly prolongs the delay in gastric emptying in macol Ther 1996; 10: 411–9. rats. Nutr Res 1993; 13: 715–22. 27 Shi G, Bruley des Varannes S, Cherbut C, Rozé C, 7 Robinson PH, Stephenson JS. Dietary restriction delays Galmiche JP. Diet-induced changes of gastric emptying in gastric emptying in rats. Appetite 1990; 14: 193–201. rats: involvement of endogenous CCK [abstract]. Gastroen- 8 Forster ER, Dockray GJ. The role of cholecystokinin in terology 1997; 112: A654. inhibition of gastric emptying by peptone in the rat. Exp 28 Liddle RA. Regulation of cholecystokinin synthesis and Physiol 1992; 77: 693–700. secretion in rat intestine. J Nutr 1994; 124: 1308–14. 9 Gregory PC, McFayden M, Rayner DV. Control of gastric 29 Liddle RA, Carter JD, McDonald AR. Dietary regulation of emptying in the pig: influence of cholecystokinin, somato- http://gut.bmj.com/ rat intestinal CCK gene expression. J Clin Invest 1988; 81: statin and prokinetic agents. Exp Physiol 1995; 80: 159–65. 2015–9. 10 Jansen JBMJ, Fried M, Hopman WPM, Lamers CBHW, Meyer JH. Relation between gastric emptying of albumin- 30 McLauglin CL, Baille CA. Decreased sensitivity of Zucker obese rat to the putative satiety agent cholecystokinin. dextrose meals and cholecystokinin release in man. Dig Dis Sci 1994; 39: 571–6. Physiol Behav 1980; 25: 543–8. 11 Konturek SJ, Kwiecien N, Obtulomicz W, Kopp B, Olesky J, 31 Baille CA, McLauglin CL, Della-Fera MA. Role of Rovati L. Cholecystokinin in the inhibition of gastric secre- cholecystokinin and opioid peptides in control of food tion and gastric emptying in humans. Digestion 1990; 45: intake. Physiol Rev 1986; 66: 172–234. 1–8. 32 Raybould HE, Zittel TT. Inhibition of gastric motility 12 Konturek JW, Thoer P, Maczka M, Stoll R, Domschke W, induced by intestinal glucose in awake rats: role of +

Konturek SJ. Role of cholecystokinin in the control of gas- Na -glucose co-transporter. Neurogastroenterol Motil 1995; on October 1, 2021 by guest. Protected copyright. tric emptying and secretory response to a fatty meal in nor- 7: 9–14. mal subjects and duodenal ulcer patients. Scand J 33 Edelbroek M, Horowitz M, Fraser R, Wishart J, Morris H, Gastroenterol 1994; 29: 583–90. Dent J, Akkermans L. Adaptive changes in the pyloric 13 Moran TH, Ameglio PJ, Schwartz GJ, Peyton HJ, McHugh motor response to intraduodenal dextrose in normal PR. Endogenous cholecystokinin in the control of gastric subjects. Gastroenterology 1992; 103: 1754–61.