Structural and immunoendocrine remodeling in gut, pancreas and thymus in weaning rats fed powdered milk diets rich in Maillard reactants.

Western diet is extending worldwide and suspected to be associated with various metabolic diseases. Many food products have skim milk powder added to it and, during processing, lactose reacts with milk proteins and Maillard reaction products (MRPs) are formed. Dietary MRPs are suggested risk factors for metabolic dysregulation, but the mechanisms behind are still enigmatic. Here we describe that weaning rats fed diets rich in MRPs are affected in both their immune and endocrine systems. Marked structural changes in pancreas, intestine and thymus are noted already after 1 week of exposure. The pancreatic islets become sparser, the intestinal mucosa is thinner, and thymus displays increased apoptosis and atrophy. Glucagon- like peptide-1 (GLP-1) seems to play a key role in that the number of GLP-1 expressing cells is up-regulated in endocrine pancreas but down-regulated in the intestinal mucosa. Further, intestinal GLP-1-immunoreactive cells are juxta positioned not only to nerve fibres and tuft cells, as previously described, but also to intraepithelial CD3 positive T cells, rendering them a strategic location in metabolic regulation. Our results suggest dietary MRPs to cause metabolic disorders, dysregulation of intestinal GLP-1- immunoreactive cells, arrest in pancreas development and thymus atrophy.

Skim milk powder with high content of Maillard reaction products affect weight gain, organ development and intestinal inflammation in early life in rats.

BACKGROUND: The intestinal tract is important for development of immune tolerance and disturbances are suggested to trigger autoimmune disorders. The aim of this study was to explore the effect of Maillard products in skim milk powder obtained after long storage, compared to fresh skim milk powder. METHODS: Young rats were weaned onto a diet based on skim milk powder with high concentration of Maillard products (HM-SM, n = 18) or low (C-SM, n = 18) for one week or four weeks. Weekly body weight and feed consumption were noted. At the end, organ weights, intestinal histology, permeability and inflammatory cytokines were evaluated. RESULTS: Rats fed with HM-SM had after one week, 15% less weight gain than controls, despite equal feed intake. After one week thymus and spleen were smaller, intestinal mucosa thickness was increased and acute inflammatory cytokines (IL-17, IL-1ß, MCP-1) were elevated. After four weeks, cytokines associated with chronic intestinal inflammation (fractalkine, IP-10, leptin, LIX, MIP-2, RANTES and VEGF) were increased in rats fed with HM-SM compared to C-SM. CONCLUSION: High content of Maillard products in stored milk powder caused an intestinal inflammation. Whether this is relevant for tolerance development and future autoimmune diseases remains to be explored.

Hydrolysis of milk fat globules by pancreatic lipase. Role of colipase, phospholipase A2, and bile salts.

Human milk fat globules require colipase to be hydrolyzed by pancreatic lipase in the presence of bile salts. This is contrary to a recent report in this Journal (J. Clin. Invest. 67: 1748-1752.) according to which inhibition of lipase by bile salt could be overcome by the addition of colipase or phospholipase A2. This latter finding is shown to be due to contamination of commercially available pancreatic phospholipase A2 by colipase.

High-fat diet impairs hippocampal neurogenesis in male rats.

High fat diets and obesity pose serious health problems, such as type II diabetes and cardiovascular disease. Impaired cognitive function is also associated with high fat intake. In this study, we show that just 4 weeks of feeding a diet rich in fat ad libitum decreased hippocampal neurogenesis in male, but not female, rats. There was no obesity, but male rats fed a diet rich in fat exhibited elevated serum corticosterone levels compared with those fed standard rat chow. These data indicate that high dietary fat intake can disrupt hippocampal neurogenesis, probably through an increase in serum corticosterone levels, and that males are more susceptible than females.

Sweet and fat taste preference in obesity have different associations with personality and eating behavior.

The aim of this study was to test associations between self-reported attitudes of sweet and fat taste preferences and psychological constructs of eating behavior and personality in obesity. Sixty obese patients were included. The Three Factor Eating Questionnaire was used for the assessment of psychological constructs of eating behavior, and the Swedish universities Scales of Personality was used for measuring personality traits. A strong sweet taste preference was associated with more neurotic personality traits (P=.003), in particular lack of assertiveness (P=.001) and embitterment (P=.002). Strong fat taste preference was rather related to lower levels of the eating characteristic cognitive restraint (P=.017), implying less attempts to restrict and control food intake. Whereas strong sweet taste preference was linked to a personality style in obesity, strong fat preference could be more an aspect of eating behavior. A psychobiological stress model is discussed in relation to the results on sweet preference and hampered personality functioning.

Measurement of the binding of colipase to a triacylglycerol substrate.

The binding between colipase and two triacylglycerol substrates, tributyrin and Intralipid, in the presence of bile salts have been determined quantitatively by a method based on equilibrium partition in an aqueous two-phase system. In the model proposed the triacylglycerol, in the form of spherical droplets covered with bile salt, is assumed to have a certain number of independent binding sites at the surface for colipase. The binding of colipase to tributyrin at pH 7.0 in the presence of 4 mM sodium taurodeoxycholate and 150 mM NaCl had a dissociation constant Kd = 3.3 . 10(-7) M; the concentration of binding sites was 1.2 . 10(-6) M in a 102 mM tributyrin emulsion. When tributyrin was dispersed in 1 mM and 12 mM sodium taurodeoxycholate the dissociation constant was somewhat higher, 6.3 . 10(-7) M and 6.0 . 10(-7) M, respectively. Thus the binding strength was optimal at 4 mM sodium taurodeoxycholate. At the same time the concentration of binding sites decreased from 4.1 . 10(-6) M for 1 mM sodium taurodeoxycholate to 1.4 . 10(-6) M for 12 mM sodium taurodeoxycholate. This indicated that at higher bile salt concentration the bile salt acted as non-competitive inhibitors on the binding of colipase to the substrate, thus binding to other sites than colipase to the substrate. The binding of colipase to Intralipid, an emulsion of a long-chain triacylglycerol stabilized with phosphatidylcholine and glycerol, was more complex with indications of several different binding sites with different affinity. The majority of these had a dissociation constant Kd = 1.2 . 10(-6) M in the presence of 4 mM sodium taurodeoxycholate and 150 mM. With each droplet having a diameter of 10(-4) cm, the number of binding sites on each droplet was determined to 1.96 . 10(5) and the average area available for each colipase molecule to 1600 A at saturation. Colipase on denaturation has a surface of 1320 A.

The effect of pancreatic procolipase and colipase on pancreatic lipase activation.

Intestinal fat digestion is carried out by the concerted action of pancreatic lipase and its protein cofactor colipase. Colipase is secreted from pancreas as a procolipase and is transformed into colipase by the trypsin cleavage of the Arg5-Gly6 bond during liberation of an N-terminal pentapeptide. The kinetic parameters for the lipase-colipase system compared to the lipase-procolipase system has been compared using trioctanoin and Intralipid as substrates. It was found that at pH 7.0 the Kmapp using Intralipid as substrate was the same for procolipase and colipase, 0.06 mM and 0.05 mM, respectively. At pH 8.0, however, the Kmapp were different-0.23 mM for procolipase and 0.08 mM for colipase. In a similar way the binding between colipase and lipase had a dissociation constant of 2.4 x 10(-6) M at pH 7.0, while for procolipase--lipase binding the dissociation constant was 4.1 x 10(-6) M with no significant difference. At pH 8.0 the binding between colipase and lipase was stronger, Kd being 2.0 x 10(-7) M, while weaker for procolipase and lipase, Kd being 1.0 x 10(-5) M. It is concluded that at the physiological pH value as is found in the intestine, the activation of procolipase to colipase has no influence on the hydrolysis of trioctanoin or Intralipid in the presence of bile salt.

Effect of phosphatidylcholine and free fatty acids on the activity of pancreatic lipase-colipase.

Mixed micelles of bile salt and phospholipids inhibit the lipase-colipase-catalysed hydrolysis of triacylglycerols. Free fatty acids can reverse this inhibition and reactivate lipase-colipase. This reactivation is either due to the formation of a high-affinity complex between lipase and colipase induced by free fatty acids and/or to a change of the quality of the interface. Lauric acid, oleic acid and linoleic acid are the most potent reactivators, while short-chain free fatty acids have no effect and long-chain, saturated free fatty acids inhibit the lipase-colipase activity further. The physiological relevance of these results is evident as the glyceride emulsion reaching the duodenum already contains free fatty acids due to the activity of lingual lipase in the stomach.

Importance of the N-terminal sequence in porcine pancreatic colipase.

Colipase exists in pancreatic juice in a pro-form which is activated by limited trypsin hydrolysis. During this activation, the N-terminal pentapeptide 1Val-Pro-Asp-Pro-5Arg is cleaved. The new N-terminal sequence formed, 6Gly-Ile-Ile-Ile-10Asn, contains three isoleucine residues. The importance of these for stimulating lipase activity has been investigated by successive Edman degradation of epsilon-acetimidolysine residues followed by limited trypsin hydrolysis. The epsilon-amidinated colipase obtained was fully active both with a phospholipid-covered triacylglycerol (Intralipid) and tributyrin as substrate. After removal of the three isoleucine residues, the activity of colipase was lost with Intralipid but not with tributyrin as substrate. The shortened colipases regained their Intralipid activity upon addition of long-chain fatty acids. The binding of colipase to lipase was not affected by removal of the isoleucine residues.

The identity and properties of two forms of activated colipase from porcine pancreas.

Colipase is excreted as a procolipase, colipase101. It is activated by low concentrations of trypsin, hydrolyzing the N-terminal pentapeptide. With higher concentrations of trypsin or in the presence of Ca2+ a smaller form of colipase containing 85 amino acids, appears. It has glycine as the N-terminal and arginine as the C-terminal amino acid residue and has lost 11 amino acids in the C-terminal chain. The ability of colipase85 to activate lipase with tributyrin as substrate is about the same as for colipase96 and procolipase. With fenfluramine, an anoretic agent, added to the tributyrin colipase assay system, the specific activities of colipase96 and colipase85 are similar and about five times higher than that of colipase101. With intralipid as substrate colipase85 enables lipase to reach the triacylglycerol substrate more rapidly than colipase96, having about six times shorter lagtime for a given concentration. Colipase84, obtained by splitting off the C-terminal arginine from colipase85, has a lagtime somewhere between colipase85 and colipase96, pointing to the importance of arginine85 for Intralipid activity. The binding between lipase and colipase has about the same strength for procolipase, colipase96 and for colipase85, Kd being about 10(-6) M either in buffer or in the presence of 2 mM taurodeoxycholate at pH 7. Addition of long chain fatty acids in the presence of bile salts increases the binding strength between colipase and lipase 100-fold, both for colipase96 and colipase85.

Measurement of the binding of human colipase to human lipase and lipase substrates.

Equilibrium partition in an aqueous two-phase system was the method used for quantitative determinations of the binding between human colipase and human lipase and three triacylglycerol substrates: Intralipid tributyrin and triolein. The measurements were performed in a dextran/polyethyleneglycol system at pH 7.0 in the presence of 2 mM sodium taurodeoxycholate and 150 mM NaCL. The binding of colipase to lipase had a dissociation constant Kd = 4.8 . 10(-8) M. The dissociation constants for the binding of colipase to Intralipid, tributyrin and triolein were found to be 2.10(-7) M, 4.8 . 10(-8) M and 6.2 . 10(-8) M, respectively.

The interaction between human pancreatic carboxylester hydrolase (bile-salt-stimulated lipase of human milk) and lactoferrin.

An interaction between lactoferrin and human pancreatic carboxylester hydrolase (carboxylic-ester hydrolase, EC 3.1.1.1) (of bile-salt-stimulated lipase from human milk) has been demonstrated using partition in an aqueous two-phase system. This binding was strongly increased by the presence of sodium taurocholate, giving an apparent dissociation constant of around 10(-7) M. With this constant, significant binding is expected to occur in the intestine of the newborn being breast-fed between lactoferrin and either the pancreatic carboxylester hydrolase or the milk bile-salt-stimulated lipase. For carboxylester hydrolase, the interaction with lactoferrin meant a 1.4-fold increase in hydrolytic activity against p-nitrophenylacetate and cholesterololeate. For the function of lactoferrin we have not studied the importance of this interaction.

Procolipase is produced in the rat stomach--a novel source of enterostatin.

Procolipase was identified in the stomach by in situ hybridisation. A strong autoradiographic labelling of chief cells was seen in the fundus region, declining more distally and being almost absent in antrum. There was no labelling seen in the intestine. Colipase activity was estimated in rat gastric juice following pentagastrin stimulation and was found to average 2 microM. Furthermore, enterostatin, the N-terminal pentapeptide of procolipase, has been identified in the rat gut and pancreas. Extracts from gastric mucosa, intestinal mucosa and pancreas were purified by gel filtration (Sephadex G25), ion-exchange chromatography (CM-Sepharose) and HPLC (C18 reverse phase). Using an ELISA assay with antibodies directed against enterostatin, two forms of the peptide were identified both in the gut and in the pancreas, with the amino-acid sequences APGPR and VPGPR, respectively. APGPR was found to be the predominant form of enterostatin, whereas only a small amount had the structure VPGPR. Enterostatin in the form of APGPR, when injected intracerebroventricularly in female Sprague-Dawley rats, significantly reduced high-fat food intake in a two-choice situation of low-fat (14% fat by energy) and high-fat (38% fat) food. It is concluded that procolipase is produced in the stomach and secreted into the gastric juice. This is also a novel source of enterostatin.

The interaction between pancreatic lipase and colipase: a protein-protein interaction regulated by a lipid.

Pancreatic lipase readily adsorbs to a triglyceride droplet. In the intestine the triglyceride droplets are covered with bile salt and phospholipids which will prevent the adsorption of lipase. In this situation the activity of lipase is restored by colipase, another pancreatic protein. Lipase and colipase in solution form a 1:1 molar complex. I emphasize the fact that the binding and conformation of the two proteins in the complex is dependent on the type of lipids present and suggest that this lipid-determined structure of the complex is responsible for the actual function of lipase/colipase. It determines whether colipase assists lipase in binding to the bile salt-covered triglyceride droplet as is the case with tributyrin as substrate, and whether colipase in addition activates lipase as is the case with a mixed trioctanoin/lecithin monolayer substrate. In other words, lipase activity is regulated by the combined action of colipase and the lipid substrate.

A possible physiological function of pancreatic pro-colipase activation peptide in appetite regulation.

Pancreatic pro-colipase activation peptide, a pentapeptide with the sequence VPDPR was found to significantly suppress food intake of 20 h fasted Sprague-Dawley rats in a dose-dependent way. A rat treated with pro-colipase-enriched pellets for 26 days showed decreased daily food intake and retarded growth, which were restored during a following period of regular feeding. Genetically obese Zucker rats (fa/fa) were found to contain a reduced content of pancreatic pro-colipase (60% reduction), whereas the pancreatic lipase content was normal. A physiological function of pancreatic pro-colipase activation peptide as an endogenous satiety signal is suggested.

Inhibition of insulin release by intraduodenally infused enterostatin-VPDPR in rats.

Enterostatin, an amino-terminal pentapeptide produced in the intestinal lumen after cleavage of pancreatic procolipase, has been shown to suppress fat intake in rats after intraduodenal infusion. In this study, female Sprague-Dawley rats fitted with a duodenal catheter were intestinally infused with enterostatin (Val-Pro-Asp-Pro-Arg, 11.3 and 22.6 nmol/kg/min) plus 20% Intralipid for 30 min. Plasma insulin levels were significantly reduced, whereas plasma glucose concentrations were not altered by enterostatin-VPDPR. The tripeptide Asp-Pro-Arg was also found to decrease the levels of plasma insulin. However, the pentapeptide with the sequence Val-Pro-Gly-Pro-Arg, des-Arg-enterostatin Val-Pro-Asp-Pro and the tripeptide Pro-Asp-Pro failed to cause the reduction of plasma insulin levels in rats following intestinal infusion of these peptides. Radiolabeled enterostatin ([3H]VPDPR) was identified in plasma by HPLC following intraduodenal infusion of the peptide, indicating that the appearance of an intact enterostatin-VPDPR in blood. It is concluded that intestinally administered enterostatin-VPDPR and its metabolites reduce plasma levels of insulin stimulated by Intralipid.

Binding of enterostatin to the human neuroepithelioma cell line SK-N-MC.

SK-N-MC cells were found to possess binding sites for enterostatin, a peptide with central effects on appetite and sympathetic activation of brown adipose tissue during high-fat feeding. Scatchard analyses of the binding indicated one high-affinity binding (Kd = 0.5-1.5 nM) and one low-affinity binding (Kd = 15-30 nM) for 3H-enterostatin (APGPR). 125I-YGGAPGPR showed similar binding parameters as for the low affinity binding of 3H-APGPR. Met-enkephalin and beta3-casomorphin1-5 were found to displace the binding of 3H-APGPR to the SK-N-MC cells. Affinity purification of solubilized cells revealed an APGPR-binding protein estimated to 53 kDa which may represent a distinct enterostatin receptor. Cross-linking of 125I-YGGAPGPR to intact cells labeled one major protein with the same molecular size. There was no binding of enterostatin to four other human neuroblastoma/neuroepithelioma cell lines, named IMR-92, LAN#5, NB-1 #14 and SH5-SY.

Metabolism of enterostatin in rat intestine, brain membranes, and serum: differential involvement of proline-specific peptidases.

The degradation of enterostatin (VPDPR), a potent inhibitor of food intake, by intestinal brush-border membranes, brain membranes, and rat serum has been investigated in the presence of specific inhibitors. Hydrolysis by intestinal membranes was found to be 10 and 100 times faster than in serum and brain membranes, respectively. Enterostatin hydrolysis by intestinal and brain membranes involves the removal of C-terminal arginine by carboxypeptidase P, leading to the production of des-Arg-enterostatin, and the splitting of the Pro2-Asp3 bond by dipeptidyl aminopeptidase IV (DPP IV). A small amount of the potent anorectic peptide Pro2-Asp3-Pro4 was released during hydrolysis of des-Arg-enterostatin by brain membranes and rat serum. In rat serum, enterostatin degradation was mainly due to DPP IV.

Enterostatin in gut endocrine cells--immunocytochemical evidence.

The presence of enterostatin, a pentapeptide acting as a potential satiety signal in rats, was investigated in rat intestine by immunocytochemical methods. Using antibodies directed against the C-terminal part of enterostatin, the peptide was identified in endocrine cells in the antral part of the stomach and in the small intestine of rat. The immunoreactive cells were more frequent in the antrum and duodenum and became gradually fewer towards the distal small intestine. In some of the labeled endocrine cells, a coexistence of enterostatin with serotonin was revealed by immunocytochemical double staining, implying that the cells were enterochromaffin cells. In the pancreas, no enterostatin-immunoreactive cells were detected, indicating enterostatin to be included in its parent molecule, procolipase. In addition, the existence of procolipase in the gastrointestinal tract, including the pancreas, was investigated. Procolipase immunoreactivity was also identified, except in the pancreas, in chief cells in the fundus region of the stomach. The number of labeled cells declined distally in the stomach, finally being absent in the intestine. Immunoreactive enterostatin was measured with a specific ELISA method. Intestinal content and serum were found to average 540 +/- 70 and 50 +/- 4 nM, respectively. Pancreatic duct ligation strongly reduced the levels of enterostatin in intestinal content to 5.4 +/- 1.5 nM (p < 0.001), and also reduced the serum enterostatin level to 35 +/- 5 nM (p < 0.05). It is concluded that the peptide enterostatin in the rat is produced both in the exocrine pancreas, as part of pancreatic procolipase, and in gut endocrine cells, both sources of peptide being important for the circulating enterostatin.

Enterostatin: a gut-brain peptide regulating fat intake in rat.

The effect of enterostatin on high-fat food intake has been investigated. After 18 h of food deprivation, rats were injected intravenously with enterostatin (VPGPR). A dose of 38 nmol of enterostatin gave a significant inhibition of high-fat food intake, while at a higher dose of 76 nmol the inhibiting effect was lost. During the first hour, after injection of enterostatin, there was even a slight increase in food intake. Binding studies of tritiated enterostatin to crude brain membranes indicated one binding site with high affinity (Kd = 0.5 nM) and one with low affinity (Kd = 170 nM). The two dissociation constants suggest different receptor subtypes and could explain why enterostatin both can inhibit and, at high doses, stimulate fat intake in rats.