Article pubs.acs.org/jpr
Whey Protein Delays Gastric Emptying and Suppresses Plasma Fatty Acids and Their Metabolites Compared to Casein, Gluten, and Fish Protein Jan Stanstrup,*,† Simon S. Schou,† Jens Holmer-Jensen,§ Kjeld Hermansen,§ and Lars O. Dragsted*,† †
Department of Nutrition, Exercise and Sports, Faculty of Science, University of Copenhagen, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark § Department of Endocrinology and Internal Medicine, Aarhus University Hospital, Tage-Hansens Gade 2, DK-8000 Aarhus C, Denmark S Supporting Information *
ABSTRACT: Whey protein has been demonstrated to improve fasting lipid and insulin response in overweight and obese individuals. To establish new hypotheses for this effect and to investigate the impact of stomach emptying, we compared plasma profiles after intake of whey isolate (WI), casein, gluten (GLU), and cod (COD). Obese, nondiabetic subjects were included in the randomized, blinded, crossover meal study. Subjects ingested a high fat meal containing one of the four protein sources. Plasma samples were collected at five time points and metabolites analyzed using LC-Q-TOF-MS. In contrast to previous studies, the WI meal caused a decreased rate of gastric emptying compared to the other test meals. The WI meal also caused elevated levels of a number of amino acids, possibly stimulating insulin release leading to reduced plasma glucose. The WI meal also caused decreased levels of a number of fatty acids, while the GLU meal caused elevated levels of a number of unidentified hydroxy fatty acids and dicarboxylic fatty acids. Also reported are a number of markers of fish intake unique to the COD meal. KEYWORDS: metabolomics, whey protein, gastric emptying, fish markers
■
INTRODUCTION Obesity,1,2 together with metabolic syndrome and type 2 diabetes,3 has emerged as a threat to the continuation of the major health improvements that have occurred in most global regions of the world during the last few decades.4 One of the risk factors of obesity is dyslipidemia.5 While the presence of hypertriglyceridemia has traditionally been assessed using fasting blood samples, it has recently been demonstrated that cardiovascular disease, one of the main risk factors associated with obesity,6,7 is more strongly associated with postprandial lipemia.8−10 The magnitude of postprandial hyperlipidemia is primarily correlated to the diet in general and fat content of meals in particular.11 However, the extent of postprandial hyperlipidemia can also be modified by concomitant intake of protein and it has been demonstrated that specifically whey proteins have lipid-lowering and insulin secretion enhancing properties. It has been demonstrated that whey improves fasting lipids and insulin levels in overweight and obese individuals following a period of whey supplementation,12 and we recently demonstrated that intake of whey in combination with a high fat meal caused lower postprandial lipemia (plasma triglycerides and free fatty acids) in the obese nondiabetic subjects compared to supplementation with cod or gluten.13 Since postprandial lipemia is closely correlated to cardiovas© 2014 American Chemical Society
cular disease, long-term dietary supplementation with whey protein may prove beneficial in preventing cardiovascular disease in obese nondiabetic subjects. The mechanisms behind the differential effect are, however, not clarified. Better understanding of the mechanisms are important for translating the beneficial metabolic effects of whey to the clinic and possibly for identifying the active components as a possible future drug target for improving insulin response. We hypothesize that the acute differential effects between whey isolate and other proteins may be ascribed to differences in postprandial amino acids, lipids, and/or other metabolites in obese nondiabetic subjects and that such differences may be revealed by metabolomics using sensitive liquid chromatography coupled time-of-flight mass spectrometry (LC-TOF-MS). The objective of our study was to explore the acute metabolic changes after a lipid load together with whey protein isolate (WI) compared to intake of a similar load together with gluten (GLU) or cod (COD), exemplifying plant- and animal-derived proteins, or casein (CAS), the other major protein constituent in milk. Our specific aims were dual: (1) to explain the differential effects of whey compared to the other three major dietary proteins and, in addition, to assess the effect of stomach Received: December 9, 2013 Published: April 8, 2014 2396
dx.doi.org/10.1021/pr401214w | J. Proteome Res. 2014, 13, 2396−2408
Journal of Proteome Research
Article
The study volunteers were 11 obese nondiabetics, aged 40− 68 years with BMI 30.3−42.0. The isocaloric test meals consisted of an energy-free soup added 100 g of butter and 45 g of one of the protein sources. Twenty five grams of raw leek was added to make the soup more palatable. Forty five grams of carbohydrate was given in the form of white bread. The participants arrived fasted and delivered a urine spot sample. After a period of rest blood samples were drawn into EDTA tubes immediately before intake of the meal (T = 0) and at 1, 2, 4, and 8 h after T = 0. Plasma was thereafter separated by centrifugation at 2000 × g for 20 min at 4 °C. Plasma samples were stored at −80 °C until analysis. Urine was also collected during the study period (0−8 h after the meal) and samples stored at −20 °C until analysis. Clinical results, study design and subject characterization have been previously described in detail by Holmer-Jensen et al.13 The Central Denmark Region Committee (address: De Videnskabsetiske komiteer for Region Midtjylland, Skottenborg 26, DK-8800 Viborg, Denmark,
[email protected]) on Biomedical Research Ethics approved the study according to the Helsinki declaration and the protocol is registered on Clinicaltrials.gov (ID: NCT00863564) and the full trial protocol can be accessed from the Ethics committee.
emptying; (2) to identify any specific metabolites that may serve as markers of exposure or biological response to each of the protein treatments. The main outcomes of this investigation have been previously described by Holmer-Jensen et al.13
■
MATERIALS AND METHODS
Materials
Whey isolate (LACPRODAN DI-9224) and calcium caseinate (Miprodan 40) milk protein were kindly supplied by Arla Foods Ingredients Group P/S (Aarhus, Denmark) and the gluten protein (Gluvital 21000) was kindly provided by Cerestar Scandinavia A/S (Charlottenlund, Denmark). The cod fillet was purchased from the local market (Coop torskefilet (cod fillet); Royal Greenland A/S, Aalborg, Denmark). All aqueous solutions were prepared using purified distilled water from a Millipore Milli-Q system (Billerica, MA, USA). All organic solvents used and formic acid were LC-MS grade from Sigma-Aldrich (St. Louis, MO, USA). β-hydroxy-α-methylbutyric acid, 3-Hydroxy-2-methylbutyric acid, 1,2,3,4-tetrahydroβ-carboline-3-carboxylic acid and uridinediphosphoglucuronic acid (UDPGA) were purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA), 3-methyl-histidine and N6,N6,N6trimethyl-lysine from Bachem (Bubendorf, Switzerland), (±)-propionylcarnitine chloride from Tocris Bioscience (Bristol, UK), N-phenylacetyl-methionine from Enamine Ltd. (Kiev, Ukraine), α-hydroxydecanoic acid from Larodan Fine Chemicals AB (Malmö, Sweden) and glycolithocholic acid from Steroloids (Wilton, NH, USA). Human liver extract, all other reagents and all other authentic standards were obtained from Sigma-Aldrich. Liver extracts from 5 Balb/c mice were obtained from the Rodent Metabolic Phenotyping Center at the Panum Institute, University of Copenhagen (http://www. rmpc.ku.dk). The mice were euthanized without previous treatments and the livers quickly excised and cooled. The livers were homogenized in 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) pH 7.2 with 1 mM EDTA. γ-glutamyl-leucine, γ-glutamyl-valine, γ-glutamyl-methionine and β-asp-Leu were synthesized as previously described.14 Paracetamol sulfate and dopamine-3-O-sulfate was synthesized as previous described.15 In short, paracetamol or dopamine was buffered with TRIS (pH 7.5) and 3′phosphoadenosine-5′-phosphosulfate (PAPS) added. After stirring for 5 min at 37 °C a chilled human liver extract was added. Paracetamol glucuronide, sebacic acid monoglucuronide, dodecanedioic acid monoglucuronide and tetradecanedioic acid monoglucuronide was synthesized as previously described.16 In short paracetamol, sebacic acid, dodecanedioic acid or and tetradecanedioic acid was added to an aqueous solution of MgCl2 and UDPGA. After stirring for 5 min at 37 °C a chilled mouse liver extract was added. For both sulfation and glucuronidation the mixtures were incubated at 37 °C for 1 h. The reactions were stopped using cold methanol and after freezing for 10 min the mixture was centrifuged at 10,000 × g for 4 min. The supernatant was collected, evaporated to dryness and dissolved in water.
Sample Preparation
Before analysis, the plasma and urine samples were transferred to the metabolic profiling facility in Frederiksberg, Denmark. The samples were then thawed and distributed randomly in the 96-well sample tray such that all samples from the same individual were in the same analytical batch. This procedure ensures that the random noise in the more relevant intraperson variation is limited. The urine samples were diluted 1:1 with aqueous 5% 70:30 (v/v) acetonitrile (AcN):methanol (MeOH). For the plasma samples Sirrocco plates were used to precipitate and filter the proteins. 1:1 AcN:MeOH was used for precipitation and extraction. The plasma samples were redissolved in aqueous 5% 70:30 AcN:MeOH. The procedure in its entirety has been previously described by Gürdeniz et al.17 To all samples five labeled internal standards were added. LC-MS Analysis
The 220 plasma samples and 88 urine samples, in addition to blanks, metabolomics standards mix and pooled serum (collected at the department) samples, were divided into three 96 well plates and analyzed one at a time in both positive (ESI+) and negative mode (ESI-). The sample trays were kept at 4 °C during analysis. Ten μL of each sample was injected into a UPLC (Waters) equipped with a 1.7 μm 2.1 mm × 100 mm C18 BEH column (Waters) operated with a 6.0 min linear gradient from 0.1% formic acid in water to 0.1% formic acid in 20:80 acetone:acetonitrile. The eluate was analyzed by ESI-Q-TOF-MS (Waters Q-Tof Premier) controlled by MassLynx 4.0 (Waters) in both ESI+ and ESI-. A voltage of 2.8 or 3.2 kV was applied to the capillary in negative and positive modes, respectively. Data were collected in centroid mode using leucine-enkephalin as a lock-spray mass to calibrate mass accuracy every 10 s. A blank (0.1% formic acid) and a metabolomics standard mix containing 40 different physiological compounds in addition to a pooled serum sample were analyzed at the beginning, middle and end of the analysis of each 96-well tray. This metabolomics standard mix and pooled serum samples were used to check mass error (
