Carbohydrates, glycemic index and diabetes mellitus

Effects of whey proteins on glycaemia and insulinaemia to an oral glucose load in healthy adults; a dose–response study




Whey proteins have insulinogenic properties and the effect appears to be mediated from a postprandial plasma amino-acid (AA) response. The aim was to study the possible dose–response relationship between whey intake and glycaemic-, insulinaemic- and plasma AA responses.


Twelve healthy volunteers participated in the study. They were provided three whey protein drinks, containing 4.5, 9 or 18 g protein as breakfast meals in random order. All meals contained 25 g available carbohydrates (glucose). The same amount of glucose in water was used as reference.


Linear dose–response relations were found between whey protein intake and postprandial glycaemia, insulinaemia and plasma AAs. The two highest doses, 18 g and 9 g, significantly reduced postprandial glycaemia (incremental area under the curve (iAUC) 0–120 min; P0.05). The 18 g dose significantly increased the insulin response (iAUC 0–120 min; P0.05). All measured plasma AAs (15 in total), except glutamic acid, responded in a dose-dependent way, and the 9 and 18 g doses resulted in significantly higher plasma levels of AAs compared with the reference.


Whey protein affects glycaemia, insulinaemia and plasma AAs to a glucose load in a dose-dependent manner. Comparatively low doses of whey protein (9 g) reduced postprandial glycaemia significantly when added to a carbohydrate-rich meal.


The incidence of type 2 diabetes is increasing worldwide, and there is a global need for preventive actions. Observational studies indicate that milk and other dairy products might protect against type 2 diabetes and cardiovascular diseases.1, 2, 3 Several nutrients in the dairy products have been suggested to contribute to this protective effect, one of them being the protein fraction.4, 5 We have previously shown that co-ingestion of whey protein with carbohydrates (lactose), results in a higher insulin response than cheese (casein) or milk (whey and casein), respectively, indicating that the whey protein fraction may constitute an important insulinogenic component in milk.6 The insulin-stimulating effect has shown to correlate with a specific postprandial amino-acid (AA) pattern appearing in plasma after whey ingestion, with the highest response in the case of the five AAs isoleucine (ile), leucine (leu), lysine (lys), threonine (thr) and valine (val).7, 8 In addition, previous findings suggest that whey protein increases plasma insulin concentration without an additional effect on c-peptide or insulin clearance.9

When co-ingested with carbohydrates, whey protein reduces the postprandial glycaemic excursion in both healthy subjects and in type 2 diabetic subjects.6, 10 It has previously been demonstrated that intake of 18 g of whey protein concentrate significantly increased insulin response and lowers postprandial glycaemia, compared with white wheat bread or glucose controls.6, 7 Additionally, it was recently shown that a whey protein peptide fraction supplement (5–20 g) reduced glycaemia compared with a glucose reference drink in a dose-dependent way in obese subjects.11 However, insulin and AA analyses were not made in that study. It can be hypothesized that smaller doses of whey protein also have insulinogenic effects, acting through plasma AA responses. If so, this could facilitate the use of protein supplement at meals aiming at improved glycaemic regulation without increasing the overall protein intake to a large extent. The aim of the present study was therefore to investigate the effects of three different doses of whey protein co-ingested with a fixed glucose load with respect to glycaemic, insulinaemic and plasma AA responses in healthy subjects.

Subjects and methods

Test meals and study design

Three different whey doses (4.5, 9 and 18 g) were served with 25 g of available carbohydrates (glucose) as test drinks. Spray-dried whey protein isolate (LACPRODAN DI-9224, Arla Food Ingredients amba, Viby J, Denmark) was dissolved in 250 g water before serving. Twenty-five grams glucose in 250 g water was used as a reference drink.

The study was based on a glycaemic index measuring setting, as previously described.12 A single blind randomized within-subject trial was performed. The test drinks were provided as breakfasts on four different occasions in random order with approximately 1 week between each test. The participants were instructed to eat a standardized meal in the evening (between 2100 and 2200 hours) before the experimental day, consisting of white bread slices with an optional drink. After that, they were only allowed to drink small amounts of water until they reported to the laboratory in the morning. They were also instructed to avoid alcohol, excessive physical activity and food rich in dietary fibre on the day before each test. When the subjects arrived to the laboratory in the morning, a peripheral catheter was inserted into an antecubital vein. After drawing a fasting blood sample, the meals were served.

Test subjects

The recruitment of participants and the execution of the study took place in the autumn 2007. Twelve healthy, non-smoking volunteers (7W;5M), aged 20–30 years with normal body mass indices (22.9 kg/m2; mean±0.7 s.e.m.) and without drug treatment participated in the study. All test subjects had normal fasting capillary blood glucose concentrations (5.3±mM; mean±0.1 s.e.m.) and no history of malabsorption diseases. All test subjects gave their informed written consent and were aware of the possibility of withdrawing from the study at any time they desired. The approval of the study was obtained by the Central Ethical Review Board in Lund, Sweden (reference number 414/2007).

Blood analysis

Glucose and insulin were measured at 0, 15, 30, 45, 60, 90 and 120 min and AAs at 0, 15, 30, 45 and 60 min after meal consumption. Capillary blood samples were taken for immediate plasma glucose analysis (Glucose 201+, Hemocue AB, Ängelholm, Sweden), and venous blood was used for insulin and AA analysis. Serum and plasma (EDTA) tubes were left on ice to rest for approximately 30 min before centrifugation for 10 min (1800 × g, 4 °C). Thereafter, serum and plasma were immediately separated and the samples were frozen at −20 °C until analysis. Serum insulin analysis was performed on an integrated immunoassay analyser (CODA Open Microplate System; Bio-rad Laboratories, Hercules, CA, USA) using an enzyme immunoassay kit (Mercodia AB, Uppsala, Sweden). Free AAs in plasma were purified by mixing 100 μl of 10% sulfosalicylic acid with 400 μl plasma to precipitate high-molecular-weight proteins according to the method of Pharmacia Biochrom Ltd (Cambridge, UK). The AA solutions were filtered before analysing with an AA analyser (Biochrom 30; Pharmacia Biochrom Ltd) by using ion-exchange chromatography. The AAs were separated by using standard lithium citrate buffers of pH 2.80, 3.00, 3.15, 3.50 and 3.55. The postcolumn derivatization was performed with ninhydrin. The plasma concentration of aspartic acid and cysteine was close to the limit of detection and was therefore not evaluated.

Calculations and statistical methods

All data are expressed as mean±s.e.m. and values of P0.05 were considered statistically significant. Fasting values of glucose, insulin and AAs were analyzed with a general linear model (analysis of variance) using subject as a random variable. The incremental areas under the curves (iAUCs) for glucose, insulin and AAs were calculated for each subject and test meal, using the trapezoid model (GraphPad Prism, release 5.04, GraphPad software Inc., San Diego, CA, USA). The glycaemic index and insulinaemic index were calculated from the 120-min postprandial iAUC for plasma glucose and serum insulin, respectively, by using the glucose drink as a reference (glycaemic index and insulinaemic index =100). Incremental peaks (iPeaks) for the glucose and insulin of each individual were calculated as the maximum postprandial increase from baseline.

The data (iAUC) were analyzed using a mixed model analysis of covariance with subject as a random variable and the corresponding baseline (fasting values) as a covariate.

Differences between groups were identified using Tukey’s multiple comparisons tests (MINITAB, release 14, Minitab Inc., State College, PA, USA). In the cases of unevenly distributed residuals (tested with Anderson–Darling test), Box–Cox transformations were performed on the data before the analysis of variance and/or analysis of covariance, respectively.

Time × treatment interactions for glucose and insulin responses were analyzed using a mixed model (PROC MIXED in SAS release 8.01, SAS institute Inc., Cary, NC, USA) with repeated measures and an autoregressive covariance structure. Subjects were modelled as a random variable and corresponding baseline values (fasting values) were modelled as covariate. When significant interactions between treatment and time were found, Tukey’s multiple comparisons test were performed for each time point by using the MINITAB software.

Correlation analysis was conducted to evaluate the relation among dependent measures with the use of Spearman’s partial correlation coefficients controlling for subjects and corresponding baseline values (two-tailed test; SPSS software, version 18; SPSS Inc., Chicago, IL, USA).

Dose–response relations were tested with a mixed model analysis of covariance where the dose (0, 4.5, 9 and 18) and the fasting values were set as a covariates and subject as a random variable and the iAUC as the response. Linear regression was used to estimate the reduction/increase in iAUC per gram increase of whey protein.11

One subject was withdrawn from all statistical calculations due to deviant insulin responses throughout the data set (tested with Grubbs test for statistical outliers). One subject was withdrawn from all AA analysis due to analytical errors on the reference product.


Postprandial plasma blood glucose

The postprandial glucose responses are shown in Figure 1 and in Table 1. Significant linear dose-dependent reductions of glycaemia were found for iAUC 0–120 min (r=0.786; P<0.001) and iPeak (r=0.764; P<0.001) with increasing whey doses. All three whey doses (4.5, 9 and 18 g) reduced the glycaemic response (iAUC 0–120 min) compared with the reference meal (−25%, −37% and −46%, respectively), although only significant following the 9- and 18-g doses. Additionally, the linear regression analysis showed that for each additional gram of whey protein, the postprandial glycaemia (iAUC) was lowered with −3.8±1.4 mmol/l. Furthermore, the 9- and 18-g doses displayed significantly lower glucose iPeaks in comparison to the reference drink. Additionally, a main effect of treatment (P=0.0208) and a time × treatment interaction (P=0.0039) were found and the glucose response to the 9- and 18-g whey doses were significantly lower compared with the reference drink at 60 min.

Figure 1

Acute dose–response effects on blood glucose. Mean (±s.e.m.) incremental postprandial changes (Δ) in plasma glucose in response to equal amounts of carbohydrates from a reference drink and three whey drinks; n=11. Values with different letters are significantly different, P0.05 (analysis of covariance followed by Tukey’s multiple comparisons test). A significant treatment effect (P=0.0208) and a significant time × treatment interaction (P=0.0039) were found.

Table 1 Postprandial plasma glucose and serum insulin responsesa

Postprandial serum insulin

The insulin responses are shown in Figure 2 and Table 1. There was a linear dose-dependent increase of insulin expressed as iAUC 0–120 min (r=0.836; P<0.000) and iPeak (r=0.875; P<0.000). All whey doses resulted in higher iAUCs (0–120 min), compared with the reference, although only significantly different following the 18-g dose. Moreover, the iPeak was significantly higher following all the three whey doses in comparison to the reference. A treatment effect (P=0.0012) and a time × treatment interaction (P=0.0112) were found, and the insulin levels after 9- and 18-g whey doses were significantly higher compared with the reference drink at 30 min.

Figure 2

Acute dose–response effects on insulin. Mean (±s.e.m.) incremental postprandial changes (Δ) in serum insulin in response to equal amounts of carbohydrates from a reference drink and three test drinks containing different whey doses; n=11. Values with different letters are significantly different, P0.05 (analysis of covariance followed by Tukey’s multiple comparisons test). Both significant treatment effect (P=0.0012) and time × treatment interactions (P=0.0112) were found.

Postprandial plasma AAs

Incremental postprandial plasma AA responses (iAUC 0–60 min) are displayed in Figure 3 and Table 2. All AAs, except glutamic acid (glu), displayed a linear dose-dependent response (r=0.633–0.860; P0.005). The highest whey dose (18 g) yielded higher iAUC responses for all measured AAs except for glu, phenylalanine and proline, in comparison to the reference and the 4.5-g dose. The 9-g whey dose resulted in significantly higher iAUC of ile, leu, lys, thr, val, arginine, asparagine and serine than did the reference or the 4.5-g dose. Additionally, positive correlations were observed between the plasma AAs ile, leu, lys, thr and val (iAUC 0–60 min) and insulin iPeak (r>0.494; P<0.009).

Figure 3

Acute dose–response effects on specific plasma amino acids. Mean (±s.e.m.) incremental changes in plasma levels of the amino acids isoleucine (a), leucine (b), lysine (c), threonine (d), valine (e) and all the five combined (f) in response to different doses of whey protein (4.5, 9 and 18 g) when served with equal amounts of carbohydrates (25 g); n=11.

Table 2 Incremental postprandial areas under the curve (iAUC 0–60 min) for the different plasma amino acidsa


In the present study, we found that in healthy subjects, the postprandial glycaemic response to a glucose load gradually decreased when increasing the whey dose. In addition, the levels of insulin and AAs successively increased in a linear dose-dependent manner, which has not been reported before. Previously, we have shown that intake of 18 g whey protein increased the insulinaemic response and reduced the postprandial glycaemia to a 25-g carbohydrate load significantly.6 The present study demonstrates that 9 g whey also reduces glucose responses (iAUC 0–120 min and iPeak) but without significant impact on postprandial insulin (iAUC 0–120 min) to a glucose challenge (25 g). Furthermore, it was shown that also low doses of whey protein have insulinogenic properties, as the 4.5 g dose resulted in an increased insulin iPeak, compared with the glucose reference drink. As judged from correlation analysis, significant dose–response effects were observed for glycaemia, insulinaemia and plasma AAs following whey intake in healthy subjects. Altogether, it could be proposed that comparatively modest levels of whey protein can facilitate glycaemic regulation to a carbohydrate challenge through an insulinogenic effect in healthy subjects. In a recent study, we have shown that a combination of whey protein and free AAs taken just before a carbohydrate load resulted in a rapid insulinogenic effect with important impact on postprandial glycaemic regulation.13 The pre-meal whey protein and AAs intake preferentially increased the early insulin concentration (iAUC 0–15 min) with little influence on overall course of postprandial insulinaemia. In the present study, the insulinaemic response expressed as iAUC 0–30 min was significantly higher following the 9 and 18-g whey doses compared with the reference meal. Although not statistically significant, there was also a trend that all whey doses elicited higher insulin responses at 15 min after the test meal. This indicates that also low doses of whey protein have the potential to beneficially influence glycaemic regulation by inducing an early insulin response.

As a potential mechanism for the insulinogenic properties of whey, increased levels of plasma AAs, especially ile, leu, val, lys and thr, has been discussed.6, 8, 10 In the present study, all whey doses increased the plasma AA responses, which is in line with previous findings.6 Furthermore, we did observe a dose–response relation between the five AAs (ile, leu, val, lys and thr) and whey dose and the five AAs correlated positively with the insulin iPeak. Interestingly, it has previously been suggested that these five AAs could increase peripheral insulin levels by competitive binding to insulin receptors on hepatocytes and thus contribute to the increased insulin levels.9 In addition to these five AAs, also other AAs have been reported to affect insulin secretion both in vivo (alanine, arginine, glutamine, glutamic acid, methionine and phenylalanine)14, 15, 16 and in vitro (alanine and glutamine).17 Interestingly, it has recently been shown that AAs might both have gastric inhibitory polypeptide18, 19 as well as glucagon-like peptide-18 stimulatory effects. In addition, we recently showed that an early plasma AA responses correlated strongly with both gastric inhibitory polypeptide and glucagon-like peptide-1 (iAUC 0–15 min) as well as with the early insulin responses.13 Subsequently, both plasma AAs as well as the incretins could contribute to the insulin-stimulating mechanism. In the present study, we did not measure C-peptide and thus not the insulin secretion. However, insulin was elevated following whey intake and responded in a dose-dependent manner. It has previously been postulated that whey protein both increases C-peptide clearance and decreases hepatic insulin extraction,9 which could contribute to the elevated insulin concentrations.

This is the first study in healthy subjects where low doses of whey protein have been tested with respect to postprandial glycaemic, insulinaemic and plasma AA responses. Whey doses in the range 5–20 g have previously been shown to reduce postprandial glycaemia11 and affect short-term postprandial satiety in obese subjects.20 There are also dose–response studies published on whey protein and metabolic effects on healthy subject, but with a main focus on satiety and appetite regulating properties, and much larger doses (14–38 g) were tested.21, 22

A limitation of the present study is that the participant number was in the lower range (n=11). It cannot be excluded that a somewhat larger sample size would have resulted in a significant effect on insulin following the 9-g dose. A potential weakness of the study is that c-peptide and thus insulin secretion was not measured.

In conclusion, whey protein affect glycaemia, insulinaemia and plasma AA responses to an oral glucose load in a linear, dose-dependent way. We showed that low dose (9 g) of whey protein is effective in reducing glycaemia without significant increase in insulin. The results may have implications for the design of meals and meal supplements with the potential of facilitating post-meal glycaemic regulation.


  1. 1

    Elwood PC, Pickering JE, Fehily AM . Milk and dairy consumption, diabetes and the metabolic syndrome: the Caerphilly prospective study. J Epidemiol Community Health 2007; 61: 695–698.

    Article  Google Scholar 

  2. 2

    Malik VS, Sun Q, van Dam RM, Rimm EB, Willett WC, Rosner B et al. Adolescent dairy product consumption and risk of type 2 diabetes in middle-aged women. Am J Clin Nutr 2011; 94: 854–861.

    CAS  Article  Google Scholar 

  3. 3

    Jakubowicz D, Froy O . Biochemical and metabolic mechanisms by which dietary whey protein may combat obesity and type 2 diabetes. J Nutr Biochem 2012; 24: 1–5.

    Article  Google Scholar 

  4. 4

    Pfeuffer M, Schrezenmeir J . Milk and the metabolic syndrome. Obes Rev 2007; 8: 109–118.

    CAS  Article  Google Scholar 

  5. 5

    Luhovyy BL, Akhavan T, Anderson GH . Whey proteins in the regulation of food intake and satiety. J Am Coll Nutr 2007; 26: 704S–712S.

    CAS  Article  Google Scholar 

  6. 6

    Nilsson M, Stenberg M, Frid AH, Holst JJ, Bjorck IM . Glycemia and insulinemia in healthy subjects after lactose-equivalent meals of milk and other food proteins: the role of plasma amino acids and incretins. Am J Clin Nutr 2004; 80: 1246–1253.

    CAS  Article  Google Scholar 

  7. 7

    Nilsson M, Holst JJ, Bjorck IME . Metabolic effects of amino acid mixtures and whey protein in healthy subjects: studies using glucose-equivalent drinks. Am J Clin Nutr 2007; 85: 996–1004.

    CAS  Article  Google Scholar 

  8. 8

    Salehi A, Gunnerud U, Muhammed S, Ostman E, Holst J, Bjorck I et al. The insulinogenic effect of whey protein is partially mediated by a direct effect of amino acids and GIP on beta-cells. Nutr Metab 2012; 9: 48.

    CAS  Article  Google Scholar 

  9. 9

    Lan-Pidhainy X, Wolever TM . The hypoglycemic effect of fat and protein is not attenuated by insulin resistance. Am J Clin Nutr 2010; 91: 98–105.

    CAS  Article  Google Scholar 

  10. 10

    Frid AH, Nilsson M, Holst JJ, Bjorck IM . Effect of whey on blood glucose and insulin responses to composite breakfast and lunch meals in type 2 diabetic subjects. Am J Clin Nutr 2005; 82: 69–75.

    CAS  Article  Google Scholar 

  11. 11

    Petersen BL, Ward LS, Bastian ED, Jenkins AL, Campbell J, Vuksan V . A whey protein supplement decreases post-prandial glycemia. Nutr J 2009; 8: 47.

    Article  Google Scholar 

  12. 12

    Brouns F, Bjorck I, Frayn KN, Gibbs AL, Lang V, Slama G et al. Glycaemic index methodology. Nutr Res Rev 2005; 18: 145–171.

    CAS  Article  Google Scholar 

  13. 13

    Gunnerud UJ, Heinzle C, Holst JJ, Ostman EM, Bjorck IM . Effects of pre-meal drinks with protein and amino acids on glycemic and metabolic responses at a subsequent composite meal. PLoS One 2012; 7: e44731.

    CAS  Article  Google Scholar 

  14. 14

    Gannon MC, Nuttall FQ . Amino acid ingestion and glucose metabolism-a review. IUBMB Life 2010; 62: 660–668.

    CAS  Article  Google Scholar 

  15. 15

    Fajans SS, Floyd JC Jr., Knopf RF, Conn FW . Effect of amino acids and proteins on insulin secretion in man. Recent Prog Horm Res 1967; 23: 617–662.

    CAS  PubMed  Google Scholar 

  16. 16

    Floyd JC Jr., Fajans SS, Conn JW, Knopf RF, Rull J . Stimulation of insulin secretion by amino acids. J Clin Invest 1966; 45: 1487–1502.

    CAS  Article  Google Scholar 

  17. 17

    Newsholme P, Bender K, Kiely A, Brennan L . Amino acid metabolism, insulin secretion and diabetes. Biochem Soc Trans 2007; 35 (Pt 5), 1180–1186.

    CAS  Article  Google Scholar 

  18. 18

    Flatt PR, Kwasowski P, Howland RJ, Bailey CJ . Gastric inhibitory polypeptide and insulin responses to orally administered amino acids in genetically obese hyperglycemic (ob/ob) mice. J Nutr 1991; 121: 1123–1128.

    CAS  Article  Google Scholar 

  19. 19

    Thomas FB, Sinar D, Mazzaferri EL, Cataland S, Mekhjian HS, Caldwell JH et al. Selective release of gastric inhibitory polypeptide by intraduodenal amino acid perfusion in man. Gastroenterology 1978; 74: 1261–1265.

    CAS  PubMed  Google Scholar 

  20. 20

    Poppitt SD, Proctor J, McGill AT, Wiessing KR, Falk S, Xin L et al. Low-dose whey protein-enriched water beverages alter satiety in a study of overweight women. Appetite 2011; 56: 456–464.

    CAS  Article  Google Scholar 

  21. 21

    Veldhorst MA, Nieuwenhuizen AG, Hochstenbach-Waelen A, van Vught AJ, Westerterp KR, Engelen MP et al. Dose-dependent satiating effect of whey relative to casein or soy. Physiol Behav 2009; 96: 675–682.

    CAS  Article  Google Scholar 

  22. 22

    Veldhorst MAB, Nieuwenhuizen AG, Hochstenbach-Waelen A, Westerterp KR, Engelen MPKJ, Brummer RJM et al. A breakfast with alpha-lactalbumin, gelatin, or gelatin plus TRP lowers energy intake at lunch compared with a breakfast with casein, soy, whey, or whey-GMP. Clin Nutr 2009; 28: 147–155.

    CAS  Article  Google Scholar 

Download references


This study was funded by the Lund University Antidiabetic Food Center, a VINNOVA VINN Excellence Center.

Author information



Corresponding author

Correspondence to U J Gunnerud.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Contributors: UJG coordinated the study and was involved in the study design, the collection and analysis of the data, statistical analysis and the evaluation and the writing of the paper. EMÖ was involved in the study design, interpretation of data and in writing the paper. IMEB was involved in the study design and for securing the funding and was involved in the evaluation and in writing the paper. All authors read and approved the final manuscript.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Gunnerud, U., Östman, E. & Björck, I. Effects of whey proteins on glycaemia and insulinaemia to an oral glucose load in healthy adults; a dose–response study. Eur J Clin Nutr 67, 749–753 (2013).

Download citation


  • amino acids
  • glycaemic index
  • insulinaemic index
  • whey protein

Further reading