Chronic metabolic conditions such as type 2 diabetes (T2D) currently place an unprecedented burden upon global health care systems( Reference Wang, McPherson and Marsh 1 , 2 ), and vast swathes of scientific literature are devoted to understanding their aetiology and the development of strategies for prevention and treatment. In this context, the postprandial period is of significant interest to researchers due to its acute effects on the metabolic and hormonal milieu, and the adverse health effects associated with chronic exposure to an impaired meal response. Given that western populations effectively spend most of the day in the postprandial state, interventions aiming to adjust the acute meal response are pertinent to preventing the deterioration in metabolic health that may precede the onset of overt symptoms.
In the years that precede the development of T2D, a progressive transition from normal glucose tolerance to impaired glucose tolerance occurs, characterised by a decline in both insulin action and early phase insulin secretion( Reference Monnier, Colette and Dunseath 3 ). Hyperglycaemia ensues, with postprandial glycaemic control rather than fasting glucose appearing to deteriorate first( Reference Monnier, Colette and Dunseath 3 , Reference Woerle, Pimenta and Meyer 4 ). Evidence suggests a considerable number of individuals with normal fasting glucose have an abnormal postprandial glucose level( 5 , Reference Gerstein 6 ) and postprandial hyperglycaemia has been identified as an independent and continuous risk factor for CVD in diabetic and non-diabetic populations( Reference Gerich 7 – Reference Bianchi, Miccoli and Penno 9 ). This remains the case when fasting glucose is in the normal range( Reference Ning, Zhang and Dekker 10 ).
Pharmacological agents such as metformin and thiazolidinediones are known to be effective in reducing blood glucose concentrations, with possible favourable cardiovascular effects. Their effectiveness in maintaining normoglycaemia over the long term is uncertain however, while significant side effects may limit use( Reference Tahrani, Bailey and Del Prato 11 , Reference Tahrani, Piya and Kennedy 12 ). The role of lifestyle in the prevention of onset and deterioration in metabolic health is therefore paramount, and evidence demonstrating that many chronic diseases are preventable through lifestyle is compelling( Reference Tuomilehto, Lindström and Eriksson 13 – Reference Salas-Salvadó, Martinez-González and Bulló 15 ). A key component of this is identifying dietary practices and food components, which reduce the risk of deterioration in metabolic health outcomes.
Dietary strategies including modification of complex carbohydrate intake or consumption of low glycaemic index (GI) diets may be beneficial in improving metabolic health outcomes( Reference Raben 16 ), however compliance may be difficult to maintain in the long term( Reference Brekke, Sunesson and Axelsen 17 ). A wealth of epidemiological evidence associates increased dairy consumption with reduced risk of metabolic disease and an inverse relationship has been demonstrated between dairy consumption and glycaemia( Reference Drehmer, Pereira and Schmidt 18 , Reference Da Silva, Julien and Couture 19 ) as well as incidence of T2D( Reference Aune, Norat and Romundstad 20 , Reference Díaz-López, Bulló and Martínez-González 21 ). Milk and other dairy products are complex and diverse foods containing a number of key nutrients and bioactive components may synergistically contribute to reduction of disease risk( Reference Lovegrove and Givens 22 ). It is therefore prudent to consider the effect of isolated components in randomised controlled trials. Dairy proteins appear to play a key role in the influence of dairy consumption on metabolic outcomes, with an accumulating body of evidence supporting the beneficial effects of whey protein consumption in particular( Reference Zhang, Tong and Wan 23 – Reference Mignone, Wu and Horowitz 25 ).
Bovine milk protein comprises of two distinct types of proteins, namely whey (approximately 20%) and casein (approximately 80%). Both whey and casein are complete proteins, however whey is a considerably richer source of the branched-chain amino acids (BCAA) leucine, isoleucine and valine( Reference Hall, Millward and Long 26 ), which may have important metabolic consequences. Casein is precipitated in the stomach by gastric acid resulting in coagulation and slowing of gastric emptying, whereas whey proteins are acid-soluble and rapidly emptied from the stomach and delivered to the small intestine intact( Reference Pal and Radavelli-Bagatini 24 ). This results in a greater increase in plasma amino acids, which may be important for augmenting the postprandial response. Studies in a variety of populations have identified the capacity of whey protein to augment the postprandial insulinaemic response, with potential concomitant effects on glycaemia. The present paper will review the current evidence and provide an overview of the potential mechanisms by which whey protein consumption may influence postprandial glycaemia.
Whey protein and postprandial glycaemia
It is likely that several mechanisms interact to affect postprandial glycaemia, with evidence suggesting that whey lowers glucose via insulin-dependent and insulin-independent mechanisms( Reference Akhavan, Luhovyy and Panahi 27 ). Although not fully understood, it appears that whey protein enhances postprandial insulinaemia via direct and indirect pathways. It is well established that protein, or more specifically amino acids, stimulate insulin secretion( Reference Floyd, Fajans and Conn 28 ). This effect is present in healthy individuals( Reference Nilsson, Stenberg and Frid 29 ) and in those with T2D (who have a blunted insulin response to increases in plasma glucose concentration), where coingestion of protein with carbohydrate amplifies the insulin response( Reference van Loon, Kruijshoop and Menheere 30 , Reference Manders, Hansen and Zorenc 31 ). Whey protein is a rich source of essential amino acids, known to be potent insulin secretagogues( Reference Floyd, Fajans and Pek 32 ). Leucine is of particular interest in this regard, exerting acute and chronic effects at multiple regulatory sites including pancreatic β-cells, liver, muscle and adipose tissue to influence glucose homeostasis( Reference Yang, Chi and Burkhardt 33 ).
Previous studies using β-cell lines or incubated primary islet cells have described two separate pathways where leucine directly stimulates insulin secretion via increases in the intracellular ATP:ADP ratio within β-cells. Firstly, intracellular catabolism of leucine acts to increase energy status via its deaminated metabolite α-ketoisocaproic acid, which is converted to acetyl-CoA before entering the tricarboxylic acid cycle( Reference Yang, Chi and Burkhardt 33 – Reference Gao, Ning and Wang 35 ). Secondly, leucine can allosterically activate glutamate dehydrogenase, which converts glutamate to α-ketoglutarate, an intermediate in the tricarboxylic acid cycle( Reference Yang, Chi and Burkhardt 33 , Reference Smith and Stanley 36 , Reference Liu, Jeppesen and Gregersen 37 ). The subsequent increase in ATP results in closure of ATP-sensitive potassium channels, which has the effect of depolarising the cell membrane, generating an influx of calcium ions. Oscillations in membrane potential are accompanied by oscillations in cytosolic calcium concentration and subsequent exocytosis of secretory granules, which discharge their contents (insulin, C-peptide and some proinsulin) into the extracellular space( Reference Seino, Shibasaki and Minami 38 ). It has also been proposed that some amino acids can directly depolarise the cell membrane through co-transport with Na+( Reference Newsholme, Brennan and Rubi 34 , Reference Salvucci, Neufeld and Newsholme 39 ), thus activating voltage-dependent calcium channels resulting in subsequent insulin secretion as described earlier. Additionally, BCAA can activate the mechanistic target of rapamycin pathway via stimulation of P70 S6 kinase( Reference Yang, Chi and Burkhardt 33 , Reference Kwon, Marshall and Pappan 40 ), with leucine described as the most effective in this regard( Reference Newsholme, Brennan and Rubi 34 ). This activation of the mechanistic target of rapamycin mitogenic signalling pathway is associated with mitochondrial signals generated by leucine metabolism, leading to stimulation of protein synthesis and enhanced β-cell function( Reference Xu, Kwon and Cruz 41 ).
When intact whey protein or a mixture of amino acids including BCAA were administered to healthy individuals (combined with 25 g glucose), both drinks raised insulin concentrations similarly, suggesting that the action of whey on postprandial glycaemia is largely associated with its BCAA content( Reference Nilsson, Holst and Bjorck 42 ). However, the glucose-dependent insulinotropic peptide response was 80% greater following intact whey compared with glucose control, whereas it was not different for the amino acid mixtures. This is indicative of bioactive peptides present in intact whey, or formed during digestion, influencing the incretin response. Indeed, whey also stimulates glucagon-like peptide-1 release in healthy males( Reference Akhavan, Luhovyy and Panahi 27 ). Both hormones have strong insulinotrophic effects( Reference Adams and Broughton 43 ), therefore whey stimulated incretin release may be responsible for an indirect influence on the insulin response, an effect observed in vitro ( Reference Salehi, Gunnerud and Muhammed 44 ).
The efficacy of the incretin hormone response is diminished by rapid degradation to inactive forms by the dipeptyl peptidase-IV (DPP-IV) enzyme( Reference Power, Nongonierma and Jakeman 45 ). Dairy proteins, including whey, have emerged as potential endogenous inhibitors of DPP-IV; however, evidence in human subjects is limited. Administration of whey has been associated with a significant reduction in DPP-IV activity in mice( Reference Gunnarsson, Winzell and Deacon 46 ), while a peptide identified from whey β-lactoglobulin hydrolysate is associated with moderate DPP-IV inhibition in vitro ( Reference Tulipano, Sibilia and Caroli 47 ). In human subjects with T2D, consuming 50 g whey protein compared with water prior to a high GI breakfast reduced postprandial glycaemia while stimulating insulin and glucagon-like peptide-1 responses; however no differences in DPP-IV activity were detected( Reference Jakubowicz, Froy and Ahren 48 ). One explanation may be that DPP-IV inhibition in the small intestine may not have been detectable in the plasma, with the possibility that this is another indirect mechanism where whey exerts an effect on insulin secretion.
The most likely mediator of insulin-independent actions of whey on glycaemia is a result of modulation of gastric emptying, since slowing of gastric emptying can diminish postprandial glucose excursions( Reference Horowitz, Edelbroek and Wishart 49 , Reference Marathe, Rayner and Jones 50 ). The addition of whey protein to a carbohydrate load reduces postprandial glucose while displaying delayed gastric emptying( Reference Hoefle, Bangert and Stamfort 51 ), and addition of whey to a preload has been reported to slow gastric emptying of a subsequent meal in healthy( Reference Akhavan, Luhovyy and Brown 52 ) and diabetic( Reference Ma, Stevens and Cukier 53 ) populations. The gastric emptying and incretin related properties of whey consumption in the context of attenuating postprandial glucose excursions appear to be interdependent, as glucagon-like peptide-1 also slows gastric emptying and suppresses glucagon secretion( Reference Mignone, Wu and Horowitz 25 ).
Acute intervention studies
The number of published acute studies investigating the effects of whey protein consumption on postprandial responses has increased considerably in recent years. A summary of these studies in normal-weight, overweight/obese and T2D populations is provided in Table 1. Studies in a variety of populations have identified the capacity of whey protein to augment the postprandial insulinaemic response, with potential concomitant effects on glycaemia. Studies with overweight or T2D populations are fewer in number than those with healthy sample groups, and it remains unclear whether the magnitude of the insulinotrophic effect following whey supplementation is sufficient to consistently reduce postprandial glycaemia in insulin resistant individuals, where hyperinsulinaemia may be commonplace.
Table 1. Summary of acute randomised trials investigating the effect of whey protein consumption on postprandial glycaemia
AA, amino acids; AUC, area under curve; BCAA, branched-chain amino acids; BM, body mass; CHO, carbohydrate; DPP-IV, dipeptidyl peptidase IV; GI, glycaemic index; GIP, glucose-dependent insulinotrophic polypeptide; GLP-1, glucagon-like peptide-1; SP, soy protein; T2D, type 2 diabetes; WP, whey protein; WPH, whey protein hydrolysate.
Zafar et al. ( Reference Zafar, Waslien and AlRaefaei 54 ) investigated the effects of 25 g whey protein or glucose, or a combination of both (50 g), on postprandial glycaemia for 120 min in fifteen normal-weight and fifteen overweight young females. Whey protein was effective in reducing postprandial glycaemia compared with glucose and glucose with whey, while the latter condition supressed blood glucose compared with control. Glucose area under curve (AUC) was 16% higher following the pure glucose drink in overweight compared with normal-weight participants, however this disparity in glucose tolerance between groups was corrected when whey protein was administered, with glucose attenuated to similar levels in both groups following whey protein beverages. The authors hypothesise that such an effect may be due to an enhancement of insulin sensitivity by whey protein in overweight participants; however insulin concentrations were not assessed in this study, making this a speculative assumption. Such an effect has, however, been observed following chronic whey protein consumption in both rats( Reference Belobrajdic, McIntosh and Owens 55 ) and human subjects( Reference Pal, Ellis and Dhaliwal 56 ).
A dose–response relationship between whey protein load and post-meal glucose reduction has been observed in several studies in normal weight( Reference Akhavan, Luhovyy and Brown 52 , Reference Gunnerud, Ostman and Bjorck 57 ) and obese populations( Reference Petersen, Ward and Bastian 58 ). Both studies in lean healthy individuals observed a linear effect of glucose reduction with whey protein loads ranging 5–40 g( Reference Akhavan, Luhovyy and Brown 52 ) and 4·5–18·0 g( Reference Gunnerud, Ostman and Bjorck 57 ); however, differential patterns in insulin response were reported between studies, which may be the result of a number of fundamental differences in study design. Akhavan et al. ( Reference Akhavan, Luhovyy and Brown 52 ) administered whey as a preload 30 min before consumption of a mixed-macronutrient meal, whereas Gunnerud et al. ( Reference Gunnerud, Ostman and Bjorck 57 ) did not use a preload design. The addition of 18 g whey to 25 g glucose has previously been found to significantly reduce glycaemia with a concomitant increase in insulin AUC( Reference Nilsson, Holst and Bjorck 42 ), and it has been proposed that amino acid availability may potentiate the increased insulin response. This is an attractive theory since plasma amino acids also increased in a dose-dependent manner in this study, while a previous investigation revealed that a combination of whey and free amino acids induced a rapid insulinogenic effect, which influenced early glycaemia( Reference Gunnerud, Heinzle and Holst 59 ).
A limited number of studies have assessed the glycaemic response following whey supplementation in non-diabetic overweight or obese populations. A significant reduction in glucose AUC was observed over 180 min following consumption of a 55 g liquid protein preload in comparison with lactose and glucose controls in overweight or obese men (BMI ranging from 26·8 to 40·4 kg/m2), however insulin concentrations were not affected by condition( Reference Bowen, Noakes and Trenerry 60 ). It is important to note that data for whey and casein protein preloads were pooled together and therefore the differential effects of whey and casein supplementation on a variety of markers( Reference Hall, Millward and Long 26 , Reference Boirie, Dangin and Gachon 61 ) cannot be distinguished. Nevertheless, the findings of Pal et al. ( Reference Pal, Ellis and Ho 62 ) support the observations described earlier. This group reported a 16% reduction in post-meal glycaemia over 360 min in a group of overweight or obese post-menopausal females following 45 g whey protein compared with an energy-matched glucose control. This intervention did not influence post-meal insulinaemia, indicating that milk-derived proteins may influence plasma glucose via insulin-independent mechanisms or via temporal changes in insulin sensitivity. Key differences exist in the design of these studies therefore this may influence any direct comparisons. Pal et al. ( Reference Pal, Ellis and Ho 62 ) administered protein alongside a mixed macronutrient breakfast meal, simulating the effect of consuming a realistic breakfast meal (bread roll with margarine and spread) with additional protein, which may be a more ecologically valid method than the provision of a test beverage only( Reference Bowen, Noakes and Trenerry 60 ).
In patients with T2D, substituting lean ham and lactose with the equivalent amount (18·2 g) of whey protein in high GI meals significantly increased the insulin response following both breakfast (18%) and lunch (49%), however glycaemia was similar following the breakfast meal, and only reduced after lunch (21%)( Reference Frid, Nilsson and Holst 63 ). It is possible that the post-breakfast increase in circulating insulin was not sufficient to overcome the greater insulin resistance observed in the post-absorptive state( Reference Plat, Byrne and Sturis 64 ). The post-lunch reduction in glycaemia is comparable with the 28% reduction in glucose AUC observed alongside a simultaneous 105% increase in insulin AUC by Jakubowicz et al. ( Reference Jakubowicz, Froy and Ahren 48 ). This study compared 50 g whey protein consumption with a control (water) drink before a high GI breakfast in participants with T2D. The higher dose of protein and the fact that it was administered as a preload 30 min prior to the breakfast meal in the latter study, in addition to the fact that it was a single meal test, means that these findings are not directly comparable. Despite this, the effect observed on postprandial glycaemia in both studies was greater than that detected over the same time period (180 min) following administration of nateglinide, a pharmacologic rapid-acting insulin secretagogue( Reference Gribble, Manley and Levy 65 ), emphasising the potential of dietary strategies for glycaemic management in individuals with impaired glucose metabolism. In a further study in participants with T2D, similar reductions in post-meal glycaemia were observed whether 55 g whey was given before (30 min) or alongside a high GI meal compared with no supplementation( Reference Ma, Stevens and Cukier 53 ). Rate of gastric emptying was also measured as part of this protocol and was inhibited by the presence of whey; however there was no evidence that the difference in emptying rates between the whey preload and whey in meal trials was reflected in postprandial glycaemia.
Longer-term intervention studies
Although many of the mechanistic pathways are yet to be fully elucidated, the potential benefits of whey ingestion have been described earlier, which encourages the possibility that metabolic health status may be improved if it is consumed chronically. Studies assessing the longer-term effects of whey supplementation are relatively few in number, with only one study having been carried out in normal-weight participants( Reference Astbury, Taylor and French 66 ) and a small number in overweight/obese( Reference Pal, Ellis and Dhaliwal 56 , Reference Pal, Radavelli-Bagatini and Hagger 67 – Reference Gouni-Berthold, Schulte and Krone 69 ) or diabetic( Reference Ma, Jesudason and Stevens 70 ) individuals.
The majority longer-term studies have been designed to address the impact of whey protein supplementation on appetite regulation and body weight management with limited data published on glycaemic control. It is important to consider however that reductions in body weight will positively impact insulin sensitivity and therefore have a direct effect on postprandial glycaemic control. To date, no longer-term whey protein intervention studies have measured postprandial glycaemic control but have included fasting glucose and insulin measurements. Two studies that involved supplementation of about 55 g/d whey protein for 12 and 23 weeks reported reductions in fasting insulin concentrations compared with carbohydrate supplemented groups, without reduction in fasting glucose, indicating an improvement in insulin sensitivity( Reference Pal, Ellis and Dhaliwal 56 , Reference Baer, Stote and Paul 68 ). The mechanisms explaining this improvement are unclear as this effect was independent of any change in body or fat mass in the latter study; however this effect of chronic whey supplementation on insulin sensitivity is supported by evidence from rat studies( Reference Belobrajdic, McIntosh and Owens 55 , Reference Tong, Li and Xu 71 ).
The limited evidence to date appears to show that chronic supplementation of the diet with whey protein is associated with possible metabolic health benefits including improved fasting lipid profile and insulin sensitivity, with possible effects on food intake and body mass. Caution must be taken when interpreting these findings however, as longer-term trials lack the control over extraneous variables, which exist in the laboratory environment. Variations in free-living physical activity or energy intake must also be considered in longer-term trials.
Implications for future research
Presently, many studies involving whey protein consumption have utilised high GI, carbohydrate-rich test meals such as powdered potato with( Reference Ma, Stevens and Cukier 53 ) or without( Reference Frid, Nilsson and Holst 63 ) additional glucose, or have mixed glucose into a whey protein beverage( Reference Nilsson, Holst and Bjorck 42 , Reference Gunnerud, Ostman and Bjorck 57 , Reference Petersen, Ward and Bastian 58 ) served as a test breakfast. This enables assessment of the efficacy of whey to attenuate greatly elevated postprandial glycaemia; however it may be valuable to investigate the response following consumption of whey with realistic breakfast-type foods. The consumption of test beverages in place of solid foods may also have implications when assessing appetite, since satiation is initially affected by sensory features( Reference Benelam 72 ).
There is a dearth of studies investigating the addition of whey to a meal as a supplemental strategy. A number of well controlled studies have matched comparison treatments for energy( Reference Bowen, Noakes and Clifton 73 , Reference Chungchunlam, Henare and Ganesh 74 ) or macronutrient content( Reference Mortensen, Holmer-Jensen and Hartvigsen 75 , Reference Gunnerud, Holst and Ostman 76 ); however adherence in a naturalistic setting may be easier when adding protein before or alongside a meal, rather than substituting it for other nutrients in that meal. Similar designs have supplemented whey by adding it to glucose( Reference Zafar, Waslien and AlRaefaei 54 , Reference Petersen, Ward and Bastian 58 ) or maltodextrin( Reference Hoefle, Bangert and Stamfort 51 ), but examples of addition to mixed-macronutrient meals are lacking.
The balance between providing an efficacious dose to reduce postprandial hyperglycaemia, ensuring palatability, and avoiding overconsumption of excess energy, should be carefully considered when designing studies. Studies in overweight/obese non-diabetic populations have all supplemented 45–55 g protein. A 45 g whey protein preload, as administered by Pal et al. ( Reference Pal, Ellis and Ho 62 ), contains about 750 kJ (about 180 kcal) which, if consumed regularly, may have a detrimental impact on energy balance.
To the author's knowledge, only one study( Reference Ma, Stevens and Cukier 53 ) has investigated the effect of timing of whey protein before or alongside a meal on subsequent metabolic responses. Several studies have administered whey at a set time before a standardised or ad libitum meal under all conditions, while fewer studies have administered whey protein alongside or immediately prior to a test meal. The large variation in preload timings (15–180 min) makes direct comparison of the most efficacious supplementation strategy problematic.
Further research is needed to develop optimal chronic supplementation strategies that can be incorporated into the diet in a free-living setting. Additionally, the implementation of longer-term supplementation protocols would allow assessment of potential negative consequences of whey protein supplementation on metabolic health. Increased postprandial insulinaemia in the absence of a reduction in glycaemia is suggestive of compromised insulin sensitivity, an effect that has previously been observed following acute whey protein ingestion( Reference Smith, Yoshino and Stromsdorfer 77 ). Thus there is potential for the chronic exposure to the insulinotrophic effects of whey protein to have a desensitising effect on sites of insulin action, which would have a detrimental effect on the prevention of metabolic disease.
Conclusions
Acute whey protein supplementation around meal times appears to improve postprandial glycaemia in normal weight, overweight and obese and T2D patients. In order for this to become a therapeutic strategy, further long-term intervention studies using realistic doses of whey protein provided in a format that can be incorporated easily into normal dietary practices are required.
Financial Support
None.
Conflicts of Interest
E. J. S. has received funding from Arla Food Ingredients for whey protein and glycaemic control research.
Authorship
E. J. S. and D. M. A. are solely responsible for the preparation of this manuscript.
