Postprandial hyperglycaemia is characteristic of type 2 diabetes mellitus (T2DM), however it is often present prior to clinical diagnosis and this is known as impaired glucose tolerance (IGT) or pre-diabetes [1
]. IGT is defined as a 2-h postprandial glucose concentration between 7.8 and 11 mmol/L [3
]. The use of bioactive phytochemicals from plants and plant-based foods to moderate hyperglycaemia is a growing area of nutrition research [4
]. Marine algae contain bioactive polyphenolic molecules with capacity to moderate postprandial hyperglycaemia, including phlorotannins, which are unique to marine algae [4
]. Potential mechanisms of action have previously been reviewed [4
] and include the inhibition of carbohydrate digestive enzymes, α-amylase and α-glucosidase, as demonstrated in chemical assay [4
] and the alteration of hepatic enzyme (inhibiting glucose-6-phosphatase and phosphoenolpyruvate carboxykinase) to promote glycogen production and the removal of glucose from the blood, as demonstrated in a diabetic mouse model [7
]. Marine algal polyphenols (MAPs) have also been shown to upregulate phosphorylation of AMPK (adenosine monophosphate-activated protein kinase), ACC (acetyl-CoA carboxylase) and Akt (a serine/threonine protein kinase) in diabetic mouse and rat models to increase the number of GLUT4 (glucose transporter 4) transporters at the cell membrane and increase glucose uptake at a cellular level [8
Human trials have shown that polyphenols (from plants and algae) have potential to lower postprandial blood glucose [10
], with one initial human trial showing reductions in postprandial blood glucose following treatment with MAPs specifically [15
]. However, the timing of polyphenol intake and the resulting effects on postprandial glycaemia are fundamentally unexplored in the literature. Typically, acute postprandial studies have been conducted in the morning only, whereas there is increasing evidence to indicate a time of day effect of carbohydrate consumption on postprandial glycaemia.
A small number of tightly controlled postprandial studies in human participants have shown significantly elevated glycaemia at night compared to during the day [16
]. This phenomenon is due to the influence of circadian rhythms on glucose metabolism. Glucose homeostasis is regulated in a diurnal rhythm based on the external light-dark cycle [18
] and internal feeding-fasting cycles [19
]. Typically, feeding occurs during the light phase of the cycle and fasting occurs during the dark phase of the cycle [19
]. When situations arise that challenge the status quo, such as eating during times intended for fasting, this results in a misalignment between the light-dark cycle and the feed-fast cycle resulting in postprandial hyperglycaemia. Consistent and prolonged evening hyperglycaemia may be a key contributor to the risk of T2DM and cardiovascular diseases (CVD), particularly as observed in shift working populations, who often have no choice but to eat late into the night [20
]. Strategies to modify evening postprandial hyperglycaemia may help lower the risk of T2DM and CVD [23
There is also evidence to suggest a role for the timing of treatment in influencing health outcomes. For example, taking blood pressure medications in the evening, rather than the morning, reduces overall risk of cardiovascular events in people with T2DM, largely through reducing blood pressure while asleep [24
]. Similarly, using polyphenols to moderate postprandial hyperglycaemia at night [18
], may help to improve blood glucose regulation and reduce the risk of T2DM and CVD.
This study investigated whether a polyphenol-rich extract from the marine macroalga Fucus vesiculosus moderated postprandial blood glucose and plasma insulin responses in healthy adults in the evening. It was hypothesised that healthy adults would exhibit postprandial hyperglycaemia in the evening, compared with the morning, and that the polyphenol-rich extract would reduce postprandial blood glucose, compared with placebo. A secondary outcome was the investigation of the influence of ethnicity and sex on postprandial responses.
2. Materials and Methods
2.1. Trial Design
A double-blind, placebo-controlled, randomised crossover trial was carried out in Melbourne, Australia from February 2017 to April 2018. This trial was registered with the ANZCTR, registration number ACTRN12616000126415p and is reported according to the CONSORT 2010 checklist (Supplementary Material
). Ethical approval was obtained from Monash University Human Research Ethics Committee, approval number CF16/53–2016000019. All procedures were carried out in accordance with the Declaration of Helsinki, with written informed consent given by all participants.
Participants were recruited from the public via online advertising and fliers. Volunteers were normotensive males and females, aged 18–65 years, with fasting blood glucose (FBG) < 5.5 mmol/L and body mass index (BMI) between 18.5 and 30 kg/m2. Participants were excluded if they had been diagnosed with any gastrointestinal, liver or thyroid conditions, were taking medication for blood sugar control or hypertension, were taking natural health products known to have similar actions to polyphenols for example, fish oil, had undergone recent major surgery, were pregnant, planning a pregnancy or breastfeeding, consumed > 9 standard drinks per week or > 4 standard drinks per day of alcohol, were a smoker or had a cardiac defibrillator. Testing sessions were avoided during the week prior to the beginning of menstruation for all female participants.
In the absence of similar studies with a polyphenol treatment in the evening, power analyses were based on data from a meal timing study [17
], which investigated postprandial blood glucose in the morning (8:00 a.m.), in the evening (8:00 p.m.) and at night (midnight), following an oral glucose tolerance test (OGTT). At 80% power, 15 participants were required to detect a difference in blood glucose incremental area under the curve (iAUC) of 150 mmol/L·2 h (G*Power 22.214.171.124 [26
]). The recruitment target was 19 participants, to allow for up to 20% dropout.
2.3. Test Products
The intervention product was a 2000 mg dose of a powdered extract from the brown seaweed F. vesiculosus, containing 560 mg polyphenols and 1340 mg fucoidan (a complex carbohydrate) (Marinova Pty Ltd., Tasmania, Australia). Two placebo products were used, one was a 2000 mg dose of a cellulose fibre—Medisca® Cellulose NF (Microcrystalline) (MEDISCA Australia Pty Ltd., New South Wales, Australia)—used to account for the fibre content of the seaweed extract. The other was a 2000 mg dose of commercially available rice flour (Ward McKenzie Pty Ltd., Altona, Victoria, Australia), which acted as the no treatment placebo. All test products were encapsulated in identical opaque size 0 capsules (The Melbourne Food Ingredient Depot, Brunswick, Australia).
2.4. Randomisation and Blinding
Computer generated randomisation was used to determine the order in which participants received the test products. Each supplement was coded with a corresponding letter to conceal its identity. The participants and the investigator carrying out participant enrolment, data collection and analysis (M.M.), were blinded as to which supplements participants received on each crossover. An investigator not involved in data collection (M.B.) carried out supplement order generation and allocation.
Initial screening was conducted via telephone interview. Eligible individuals were then screened in-person at the research facility to determine BMI, blood pressure and fasting blood glucose concentration. Those who remained eligible were invited to join the study and allocated a supplement order. Participants attended three testing sessions, during which they received the polyphenol-rich seaweed extract, cellulose and rice flour supplements in a randomised, crossover manner. Participants received a pre-prepared meal (pasta with tomato-based sauce; 3425 kJ, 122 g carbohydrate, 3.3 g fat, 20.4 g protein) to consume between 8–9 a.m. on the morning of each testing day. They were asked to then fast until testing was complete, with the exclusion of water. Participants were also given a list of foods (naturally high in polyphenols) to avoid and asked to avoid strenuous exercise for 24 h prior to each testing session.
Participants arrived at the testing facility at 7 p.m. after a fast of ≥10 h. Two initial finger prick tests were taken to assess fasting blood glucose (at −45 and −35 min) and plasma insulin (−45 min). At −30 min (7:15 p.m.) participants were given a 2000 mg dose of either the polyphenol-rich extract, cellulose or rice flour. At 0 min (7:45 p.m.) participants were served 50 g available carbohydrate in the form of white bread (108 g) and asked to consume the entire portion within 7 min. Finger prick blood samples were taken at regular intervals to measure blood glucose (15, 30, 45, 60, 90, 120, 150, 180 min) and plasma insulin (30, 60, 90, 120, 150, 180 min) (Figure 1
). This process was repeated three times in a crossover manner so that all participants received all three supplements. A one week wash out period was observed between each treatment.
A subset of eight participants also completed the testing protocol starting at 7 a.m. instead of 7 p.m. (with the standardised meal for dinner the night before) to confirm that elevated glycaemic responses were observed in the evening compared with the morning. In the morning, participants were only given the cellulose fibre placebo, as a comparator to the cellulose fibre placebo in the evening. Blood samples were collected for two hours instead of three because blood glucose returns to baseline levels within this time [17
2.6. Outcome Measures
2.6.1. Blood Glucose and Plasma Insulin
Capillary blood samples were obtained by pricking the fingertip with a Unistik® 3 Extra single-use lancing device (Owen Mumford Ltd., Oxfordshire, United Kingdom). Three droplets of blood were wiped away prior to collection in a HemoCue® Glucose 201 RT micro cuvette (Radiometer Pacific Pty Ltd., Mount Waverley, Victoria, Australia). Blood glucose concentrations were assessed immediately on collection of capillary blood samples using the HemoCue Glucose 201 RT System (Radiometer Pacific Pty Ltd., Mount Waverley, Victoria, Australia), according to standard procedures.
Plasma insulin was measured from capillary blood samples, which were collected using Safe-T-FillTM
Capillary Blood Collection GK Systems containing EDTA anti-coagulant (item no. 077001, Kabe Labortechnik GmbH, Cologne, Germany) (pictured in Figure 1
). At least 200 µL of whole blood was collected from the capillary at each time point. Whole blood was centrifuged at 4 °C and 1300 g
for 15 min (serial no. 5703BI110739, Eppendorf AG, Hamburg, Germany). Aliquots of plasma was stored at −80 °C until analysis. Insulin concentration was measured using the Millipore ELISA Kits for Human Insulin (Cat. # EZHI-14K and EZHI-14BK, Merck Millipore, Bayswater, Victoria, Australia), according to kit instructions. Each sample was assessed in duplicate and absorbance measured using the Rayto Microplate reader (450 nm wavelength, RT-2100C, Abacus ALS, Meadowbrook, Queensland, Australia). The lowest detectable insulin concentration for this assay was 1.0 µU/mL, therefore any values below this were rounded up to 1.0 µU/mL. The highest accurately detectable insulin concentration was 200 µU/mL. Samples that read above this value were diluted 2:1, using assay buffer as a diluent, and re-run. Units were converted from µU/mL to pmol/L prior to statistical analysis. Across all plates, the mean coefficient of variation was 8.5% (standard deviation (SD) 15.3).
2.6.2. Anthropometric Data
Height, weight and body composition were measured at the screening session. Participants removed shoes and socks for all anthropometric measures. Height was measured using the Harpenden Stadiometer (Holtain Ltd., Crymych, UK). Weight and body composition (% fat mass, % fat free mass, visceral fat (L)) were measured using the SECA mBCA 515 medical body composition analyser (SECA, Hamburg, Germany), with participants in light clothing. Waist circumference was measured over light clothing or bare skin at the narrowest point around the torso.
2.6.3. Intolerance Symptoms
An intolerance symptoms questionnaire [27
] was completed by participants 24 h after ingestion of each supplement to assess the occurrence and intensity of any side effects. Participants were asked to indicate whether they experienced intolerance symptoms as a result of taking the supplement (above any usual ailments) and whether they were of mild, moderate or severe intensity (scored as 1, 2 or 3, respectively). Side effects listed in the questionnaire were headache, anxiety, tiredness/exhaustion, lack of energy, tendency to become rapidly exhausted, reduction in appetite, increase in appetite, hiccups, nausea, vomiting, indigestion, stomach or abdominal pain, constipation, diarrhoea, gas, abdominal bloating, cardiac palpitations, balance disorders, reduced capacity to concentrate, feeling cold, muscle or joint pain, numbness, burning or itching sensations, dark or depressing thoughts.
2.6.4. Diaries and Questionnaires
Food intake data were collected using a 3-day food diary (over two weekdays and one weekend day) which was cross-checked with participants and assessed using FoodWorks 8 (Xyris Software (Australia) Pty Ltd., Spring Hill, Queensland, Australia) to establish participants’ usual dietary intake. A food frequency questionnaire (FFQ) was used to estimate participants’ dietary polyphenol intake. The FFQ was adapted for the Australian diet from a British FFQ used to assess dietary polyphenol intake [28
]. Daily dietary polyphenol intake was calculated using data from Phenol-Explorer 3.6 Database on polyphenol content in foods [29
] and the United States Department of Agriculture (USDA) Database for the Flavonoid Content of Selected Foods, Release 3.1 (December 2013) [30
]. The International Physical Activity Questionnaire (IPAQ) short version was used to assess physical activity habits. This questionnaire consists of four questions that assess the amount and intensity of physical activity and sitting time in participants’ daily lives [31
]. These data contributed to the description of characteristics of the sample population.
2.7. Quantification of Soluble Polyphenols
An adapted Folin-Ciocalteu methodology was used to quantify the total soluble polyphenols in the extract [32
] with phloroglucinol dihydrate used as standard (Sigma-Aldrich P38005, St. Louis, MO, USA). The extract was dissolved in 10 mL of distilled water and diluted to reach concentrations of 25, 50 and 100 µg/mL. The assay was performed by pipetting 2 mL of distilled water (blank), the phloroglucinol standards (5, 10, 15, 20, 30, 50 and 100 µg/mL) and the sample solutions into sequential vials. Folin-Ciocalteu reagent (500 µL) (Sigma-Aldrich F9252, St. Louis, MO, USA) was then added to each vial and allowed to stand for 5 min. Then 1500 µL of 7.5% w
sodium carbonate solution and 4000 µL of distilled water were added to each vial. The reaction was then incubated in the dark at room temperature for two hours. Analysis was conducted using a spectrophotometer at 765 nm, with the solutions in quartz cuvettes. All samples, standards and blanks were run in triplicate and absorbance values were recorded.
2.8. Statistical Analysis
Analyses were performed using Statistical Package for Social Sciences (SPSS) version 22.0 (SPSS Inc., Chicago, IL, USA), with the level of significance accepted as p
< 0.05. All results were assessed for normality using the Shapiro-Wilk test. Where results were not normally distributed the data were transformed, using the natural log and parametric tests applied. Where data could not be normalized, non-parametric tests were applied and data were reported as median (interquartile range (IQR)). Normally distributed data were reported as mean (SD). Postprandial blood glucose and insulin were assessed using iAUC and peak blood concentration. The iAUC was calculated using the trapezoidal method with baseline value removed and is expressed as mmol/L·3 h and pmol/L·3 h, respectively. Late phase insulin response was used as another measure of postprandial insulin and was calculated using the insulin iAUC from 30–120 min postprandial and is reported as pmol/L·90 min [17
A one-way repeated measures analysis of variance (ANOVA) or Friedman’s Two-Way Analysis of Variance by Ranks test (for data that did not meet relevant parametric assumptions) were used to determine differences between the treatments for fasting glucose and insulin, iAUC, late phase insulin and peak concentrations. Supplement sequence, age, sex and % fat mass were added to the one-way repeated measures ANOVA as covariates when assessing iAUC and peak concentration. Differences between the treatments for intolerance symptoms were assessed using the Friedman’s test because the data was not normally distributed. For comparison between the population sub groups (female/male and Asian/Caucasian) independent t tests or independent samples Mann-Whitney U tests (for data that did not meet relevant parametric assumptions) were used to assess differences in fasting, iAUC and peak postprandial glucose and insulin concentrations. For the comparison between morning and evening, glucose and insulin were assessed by iAUC and peak blood concentration. Paired samples t tests or Wilcoxon signed-rank tests, were used to determine differences between morning and evening for iAUC and peak concentration. The morning evening comparison (iAUC) was from 0–120 min.
In line with the literature, an elevated postprandial blood glucose response was observed in healthy individuals when carbohydrates were consumed in the evening compared with the morning. Though not clinically significant, a lowering effect of the polyphenol-rich extract on peak postprandial glucose concentration was observed among females in the evening, compared with cellulose and rice flour, with no discernible intolerance symptoms. Further research should investigate the glycaemic lowering effects of polyphenols in the evening. If polyphenol treatment can moderate postprandial hyperglycaemia in a state of IGT in the evening, it may also help people diagnosed with IGT to manage their blood glucose levels and prevent progression to T2DM. This study identified that Asian participants exhibited elevated postprandial insulin responses in the evening, compared with Caucasian participants, highlighting the need for further research investigating the effect of ethnicity and meal timing on postprandial glycaemic and insulinaemic responses.