Paper

International Journal of Obesity (2003) 27, 663–668. doi:10.1038/sj.ijo.0802284

Effect of two breakfasts, different in carbohydrate composition, on hunger and satiety and mood in healthy men

W J Pasman1, V M Blokdijk1, F M Bertina2, W P M Hopman3 and H F J Hendriks1

  1. 1TNO Nutrition and Food Research, Department of Nutritional Physiology, Zeist, The Netherlands
  2. 2Lgen Knoopkazerne, TGTF, Utrecht, The Netherlands
  3. 3Department of Gastroenterology and Hepatology, University Medical Centre Nijmegen, The Netherlands

Correspondence: Dr WJ Pasman, TNO Nutrition and Food Research, Department of Nutritional Physiology, PO Box 360, 3700 AJ Zeist, The Netherlands. E-mail: pasman@voeding.tno.nl

Received 26 June 2002; Revised 13 January 2003; Accepted 14 January 2003.

Top

Abstract

OBJECTIVE: To study the effect of simple vs complex carbohydrates (SCHO and CCHO respectively) containing breakfasts on blood parameters, hunger and satiety and mood.

DESIGN: A 2-day, open, randomised, cross-over trial.

SUBJECTS: A total of 26 male subjects (34plusminus6 y; BMI 23.4plusminus2.2 kg m-2).

MEASUREMENTS: Blood glucose, insulin, triacylglycerols (TG), free fatty acids (FFA) and cholecystokinin (CCK) were determined repeatedly for 4 h on both test days after a breakfast containing SCHO or CCHO. Feelings of hunger and satiety were determined at similar time points as well. Mood state was examined 3 h after breakfast consumption.

RESULTS: Consumption of a SCHO breakfast resulted in higher glucose and insulin levels at 30 min after breakfast consumption. TG at 180 min, and FFA at 180 and 240 min were higher after SCHO breakfast than after CCHO breakfast. Satiety scores were higher after CCHO breakfast consumption for the first 90 min after intake. Furthermore, the item 'fatigue' was scored higher after SCHO breakfast than after CCHO breakfast intake.

CONCLUSION: Consumption of a CCHO breakfast is favourable in comparison to a SCHO breakfast, because of the lower perception of 'fatigue' and the higher degree of satiety after consumption.

Keywords:

carbohydrate, satiety, mood, obesity, breakfast

Top

Introduction

The 'global epidemic of obesity'1 is due to changes in society: modernisation results in overnutrition and a sedentary lifestyle.2 The increasing prevalence of obesity has been stated to be 'a normal adaptation of body weight to an abnormal environment'.3

Subjects suffering from obesity might have features related to the metabolic syndrome as well. The combination of metabolic disturbances (high plasma insulin, high blood pressure, high plasma triacylglycerols, low HDL-cholesterol, diabetes mellitus or impaired glucose tolerance) has been suggested to be affected by insulin resistance.4,5,6 Weight loss and food consumption of low-glycemic index (GI) food or of carbohydrates have been speculated to improve obesity as well as insulin resistance.6,7,8

In general, the fat consumption has decreased in the western societies, and therefore the rise in prevalence of obesity needs to be explained by other dietary and environmental factors, for example the increase of simple carbohydrates and the decrease in physical activity. The effect of consumption of a low-fat diet on body weight reduction and on HDL-cholesterol decrease has been debated before,9,10 illustrating that specific food consumption might have positive as well as negative effects.

A dietary factor that may influence body weight and insulin resistance is the carbohydrate composition of the diet. Owing to the present trend towards an increase in carbohydrate consumption, the effect of the type of carbohydrate on body weight gain and insulin sensitivity might be related to the increasing prevalence of obesity and of insulin resistance.

Variation of the amount of simple (SCHO) and complex carbohydrates (CCHO) will result in less pronounced variations in blood glucose and insulin levels. A low or high GI of food is known to have a minor or stronger effect on blood glucose after consumption of the food product in comparison with a standard glucose load.11,12

Several investigators have studied the effect of isolated macronutrients or a single food product on GI.11 In daily life, however, combinations of macronutrients are consumed varying in carbohydrate composition as well (complex or simple), both with meals and snacks. Therefore, in this study two normal breakfasts, based on a normal Dutch breakfast for men aged 22–50 y,13 were compared. These were similar in type of food products, but different in ratio of simple to complex carbohydrates.

The aim of this study was to investigate whether even small changes within the macronutrient composition of breakfast already have an effect on metabolic and subjective parameters of hunger and satiety. Therefore, the effect of two breakfasts, different in type of carbohydrate composition, on blood glucose, insulin, fat and cholecystokinin (CCK) in healthy men was studied. Since it has been demonstrated that especially carbohydrates affect mood and stress14 this was examined in the present study as a secondary parameter as well. By means of questionnaires the scores on hunger, satiety and mood were investigated.

Top

Methods

Subjects

Male volunteers, aged 25–45 y, were recruited from the pool of volunteers of TNO Nutrition and Food Research. Respondents received a verbal briefing together with the information in writing. All volunteers gave informed consent and filled in a questionnaire on lifestyle, medical history and dietary habits. Each of the respondents was physically examined and blood was collected after an overnight fast for routine blood chemistry.

All volunteers included had a body mass index (BMI) below 28 kg m-2, were healthy as indicated by the medical questionnaire and the physical examination, had a normal (Dutch) dietary pattern, and their body weight was stable during the month preceding the study (plusminus2 kg). Eligible subjects did not suffer from gastrointestinal complaints and/or a metabolic or endocrine disease. They were all nonsmoking, did not use any soft or hard drugs, nor did they consume more than 28 glasses of alcoholic beverages a week. In total, 26 men participated. Baseline characteristics are presented in Table 1.


Study protocol

The study was a 2-day, open, randomised, crossover trial. At day 01, subjects were randomly assigned to receive a breakfast with a high or low ratio of simple-to-complex carbohydrates. At day 08, after at least 1-week wash-out period, subjects received the alternate breakfast. The wash-out period between both test days ranged for subjects between 7 and 21 days. Randomisation on age, BMI and fasting glucose and insulin was successful.

The evening meal before both test days was controlled and supplied by the TNO metabolic ward (a 'ready-to-eat' dinner), in order to keep prior food intake similar. Also, the physical activity during the day preceding the test day was kept constant (daily life and sport activities). On both test days, the volunteers came to the TNO metabolic ward where breakfast was served. Before having breakfast a fasting blood sample was collected. After having breakfast, blood samples were obtained frequently till 4 h after starting breakfast. During this 4 h period, the volunteers were not allowed to eat or drink anything.

During each visit to the TNO metabolic ward a health questionnaire, to examine adverse events, was filled in by the volunteers and their body weight was registered. Compliance of evening meal consumption was checked by asking the volunteers to return the empty bag of the 'ready-to-eat' meal on the test day and intake of breakfast was controlled by a dietitian.

Adverse events were determined by the medical investigator based on the well-being questionnaire and on spontaneous reporting.

The study was performed according to the ICH Guideline for Good Clinical Practice and was approved by the independent Medical Ethics Committee of TNO.

Breakfast

The two breakfasts differed in subgroups of carbohydrates. The macronutrient composition was kept similar and was based on a normal Dutch breakfast for men aged 22–50 y.13 A differentiation was only made within the group of carbohydrates with respect to SCHO and CCHO, respectively (see Table 2).


The SCHO breakfast contained as little as possible poly- and as much as possible mono- and disaccharides and consisted of the following four food products: white bread (two slices, 60 g), coloured, fruit-flavoured sprinkles (40 g), low-fat margarine Becel (8 g) and milk (200 ml). The CCHO breakfast consisted of the following four food products: rye bread (two slices, 50 g), low-fat 20+ cheese spread (15 g), low-fat margarine Becel (5 g) and currant bread (two rolls, 100 g). Coffee and treat were allowed on both test days ad libitum up to 500 ml (no sugar added).

The ratio between simple (mono- and disaccharides) and complex carbohydrates (polysaccharides) in the CCHO breakfast (1 : 1.6) was reverse to the SCHO breakfast (1.7 : 1) (see Table 2).

Blood chemistry

Blood was collected after an overnight fast from the antecubital vein using Vacutainer tubes (10 ml). For serum collection, blood was collected in tubes containing clot activator. Blood was centrifuged within 15–30 min after collection (10 min, 2.000 times g, 4°C) and serum stored at -18°C. Prestudy clinical chemistry as well as safety parameters were determined. In study parameters measured were: glucose, triacylglycerols (TG) and free fatty acids (FFA), which were determined using commercial test kits (Boehringer, Mannheim, Germany) on a Hitachi 911 automatic analyser (Hitachi Instrument Division, Ibaraki-ken, Japan). Insulin was determined using AIA-600 Immunoassay Analysator. CCK was measured by a sensitive and specific radioimmunoassay as described elsewhere.15

Study parameters

On both test days, blood was collected at 0, 30, 60, 90, 120, 180 and 240 min after breakfast. Glucose, insulin, TG, FFA and CCK were determined.

On both test days body weight was registered. Before having breakfast and 30, 60, 90, 120, 180 and 240 min after starting breakfast the volunteers were asked to score their feelings of hunger and satiety by means of 100 mm visual analog rating scales (VARS). Also, at 180 min after the start of having breakfast the profile of mood states (POMS) questionnaire was filled in by the volunteers.16 POMS is a questionnaire consisting of 32 adjectives describing mood states, which could be scored between 0 and 4. The described mood states are depression, anger, fatigue, vigor and tension.

Statistics

Results were expressed as mean plusminus standard deviation (s.d.), unless stated otherwise.

Data were analysed using the SAS statistical software package (SAS/STAT Version 6, SAS Institute, Cary, NC, USA).

The obtained data (blood parameters and questionnaires) were analysed by means of t-tests to investigate the differences between the two sorts of breakfast. This was analysed for both the consecutive time points and curve characteristics, like area under the curve (AUC), time to reach the maximum value of a parameter (Tmax) and the maximum response (Cmax).

Statistical analysis showed that there was no carryover effect (P>0.05). Also, no interaction existed between treatment and the difference in amount of days between the two visits (P>0.05).

In all performed statistical analyses, the null hypothesis was rejected at the 0.05 level of probability.

Top

Results

General

All 26 participants completed the study. The baseline characteristics are presented in Table 1.

Compliance of the 'ready-to-eat dinner' consumed on the evening prior to the test-breakfast was 96%. One subject ate a self-chosen dinner. Since he consumed a similar meal the evening before the second test day and subjects were their own control because of the crossover design of the study, his data was used normally.

The compliance of breakfast-intake was 100%. On both test days, all volunteers were asked to finish their breakfast in 10 min (range 4–25 min).

Nine adverse events were reported during the study. All adverse events were not or unlikely to be related to the study.

Study parameters

Blood parameters
 

Generally, the blood parameters glucose, insulin, TG, FFA and CCK did not show different patterns after consumption of the two breakfasts. However, 30 min after having breakfast glucose tended to be higher after SCHO (7.5 mmol/l) than after CCHO (7.1 mmol/l) breakfast (P=0.05) (Figure 1). At this same time-point insulin was significantly higher after intake of SCHO (53.6 mU/l) than CCHO breakfast (38.0 mU/l) (P<0.05) (Figure 2). TG was only different after 240 min of having breakfast (P<0.05) (Figure 3). Significant differences on other time-points were not present. For FFA, significantly lower levels were found after intake of CCHO than SCHO breakfast at t=180 and 240 (P<0.05) (Figure 4). No differences in CCK patterns were found after consumption of either of the breakfasts (Figure 5).

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Serum glucose (mmol/l) is shown for 4 h after SCHO breakfast (——) and CCHO breakfast (- - - -) consumption. * Significant difference between both test conditions.

Full figure and legend (15K)

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Serum insulin (mU/l) is shown for 4 h after SCHO breakfast (——) and CCHO breakfast (- - - -) consumption. * Significant difference between both test conditions.

Full figure and legend (15K)

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Serum triacylglycerol (mmol/l) is shown for 4 h after SCHO breakfast (——) and CCHO breakfast (- - - -) consumption. * Significant difference between both test conditions.

Full figure and legend (16K)

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Serum free fatty acids (mmol/l) is shown for 4 h after SCHO breakfast (——) and CCHO breakfast (- - - -) consumption. * Significant difference between both test conditions.

Full figure and legend (16K)

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Serum cholecystokinin (pmol/l) is shown for 4 h after SCHO breakfast (——) and CCHO breakfast (- - - -) consumption.

Full figure and legend (14K)

Some curve characteristics were significantly different between the breakfasts. The maximum concentration of glucose was reached significantly earlier (Tmax) after SCHO (on average 31.2plusminus5.9 min) than CCHO breakfast (39.2plusminus14.1 min) (P<0.05). The AUC of blood insulin differed significantly (lower AUC for CCHO, see Figure 2). Besides the AUC, the Cmax of insulin differed significantly between the two breakfasts. The maximum concentration of insulin reached was 59.3plusminus25.1 mU/l after intake of SCHO breakfast vs 49.7plusminus20.3 mU/l for CCHO breakfast (P<0.05). For TG and FFA it is remarkable that after approximately 120 min the curve separates (see Figures 3 and 4). Tmax for TG, FFA and Cmax for FFA are significantly different (P<0.05).

Hunger and satiety scores
 

With respect to the scores on 'hunger' and 'satiety' the following differences were observed. After 60 and 180 min of having breakfast, higher scores on 'hunger' were reported with SCHO than CCHO breakfast (P<0.05) (Figure 6). For 'satiety' scores, significant differences were found after 0, 30, 60 and 90 min (P<0.05). The 'satiety' feelings were higher with CCHO breakfast than SCHO breakfast (Figure 7).

Figure 6.
Figure 6 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Hunger scores (mm) are shown for 4 h after SCHO breakfast (——) and CCHO breakfast (- - - -) consumption. * Significant difference between both test conditions.

Full figure and legend (15K)

Figure 7.
Figure 7 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Satiety scores (mm) are shown for 4 h after SCHO breakfast (——) and CCHO breakfast (- - - -) consumption. * Significant difference between both test conditions.

Full figure and legend (15K)

The curve characteristics AUC, Tmax, and Cmax did not differ between the two breakfasts.

POMS
 

As can be seen in Table 3, the POMS questionnaire was significantly different between the two breakfasts on the item 'fatigue'. After intake of SCHO breakfast the score on 'fatigue' was 2.7plusminus3. 3 and after intake of CCHO breakfast this score was 1.5plusminus1.7 (P<0.05). A trend was found for the item 'anger' (P=0.10).


Top

Discussion

The present study showed that differences within the macronutrient composition of carbohydrates in an isoenergetic breakfast can result in a favourable effect on perception of hunger and satiety, and mood state in healthy male subjects. The effect was shown with (subjective) questionnaires as well as with physiological (objective) parameters.

The effect on mood state, feeling less fatigue after CCHO breakfast consumption in the present study, is in line with other studies as well. In a study investigating the effect of fibre breakfast cereals, it was found that perception of fatigue was significantly lower after high-fibre breakfast cereal intake.17 As a result of the higher content of dietary fibre in the CCHO breakfast in the present study as well, it is difficult to indicate whether this is an effect of the CCHO in the diet or of the amount of dietary fibre.

The relation of carbohydrates and mood has been suggested to be affected by actions of blood–brain tryptophan and serotonin. It might be useful to know that the POMS questionnaire might be useful to examine this relation as well.

The positive effect found of the CCHO breakfast, reducing hunger and increasing satiety feelings, corresponds to related studies examining the effect of dietary fibres or low GI diets as well.18,19,20,21,22

The review by Roberts recently showed that foods with a high GI may increase hunger and promote overeating, as we would like to suggest in this study as well for the SCHO breakfast. As a result of the faster rise of hunger feelings, subjects will eat more frequently and therefore ingest in total more energy daily. This was also observed with regard to unavailable carbohydrates that reduced hunger feelings in a study carried out by Sparti and coworkers.19 The authors suggested that the lower hunger perception with a diet high in unavailable carbohydrates may be helpful for a better control of food intake.19 In a study with obese teenage boys, the concomitant rapid rise in glucose after consumption of a high GI meal was the trigger for excessive food intake.20.

A comparison between consumption of simple and complex carbohydrates has been drawn before. In a multicentre, randomised ad libitum feeding trial, Saris et al21 tested the effects of altering the ratio of fat to carbohydrate, as well as simple to complex carbohydrate per se, on body weight in overweight individuals. They found weight loss with both low-fat diets: a low-fat high simple carbohydrate and low-fat high complex carbohydrate diet. Weight loss was of 0.9 kg (P<0.05) and 1.8 kg (P<0.001) respectively, while the control diet group and seasonal control group gained weight (0.8 and 0.1 kg, NS). Fat mass changed by -1.3 kg (P<0.01), -1.8 kg (P<0.001) and +0.6 kg (NS) in the low-fat high simple carbohydrate, low-fat high complex carbohydrate and control diet groups, respectively. It was suggested that reduction of fat intake resulted in a modest but significant reduction in body weight and body fatness. The concomitant increase in either simple or complex carbohydrates did not indicate significant differences in weight change.21 Our present acute findings suggest that the reduction in body weight in the low-fat high complex carbohydrate group compared to the control group, as found by Saris et al.21 might be related to a longer duration of satiety because of a high intake of complex carbohydrates. However, the difference in dietary fibre content in the present study is a confounding factor, because as has been found before, dietary fibres itself affects satiety as well.22 Another 6 month intervention study examining SCHO vs CCHO showed a reduction of body weight with CCHO as well, which was not found for the group consuming an SCHO diet.23

The differences in complex and simple carbohydrates in a diet are related to the discussion whether our diet should contain mainly low or high GI food products. A number of studies stress that a diet high in GI will deteriorate insulin sensitivity and in the long-term end up in diabetes mellitus and obesity. It is however questionable whether this is the case in normal-weight healthy subjects in which glucose loads are well-tolerated, as is shown in the present study. The body is very well able to handle the amount of glucose, because of the normal flexibility of the system. Also, Pi-Sunyer24 concluded recently that the hypothesis of a high GI being detrimental for health is still controversial.

Based on the present findings and results presented so far, we suggest that a diet containing a high portion of complex carbohydrates might be useful for body weight control via reduction of food intake. This is in agreement with the health benefits of a low GI diet as was recently reviewed by Ludwig.25

Top

Conclusion

Consumption of a CCHO breakfast is favourable in comparison to a SCHO breakfast, because of the lower perception of 'fatigue' and the higher degree of satiety after CCHO consumption.

Top

References

  1. World Health Organisation. Obesity: preventing and managing the global epidemic. WHO: Geneva; 1998.
  2. Sørensen TIA. The changing lifestyle in the world. Body weight and what else. Diabetes Care 2000; 23 (Suppl 2): B1–B4.
  3. Seidell JC. Voeding en gezondheid in de volgende eeuw: problemen van grote omvang. Vrije Universiteit Press: Amsterdam; 1999.
  4. Godsland IF, Stevenson JC. Insulin resistance: syndrome or tendency? Lancet 1995; 346: 100–103.
  5. Meigs JB. Invited commentary: insulin resistance syndrome? Syndrome X? Multiple metabolic syndrome? A syndrome at all? Factor analysis reveals patterns in the fabric of correlated metabolic risk factors. Am J Epidemiol 2000; 152: 908–911. | Article | PubMed | ISI | ChemPort |
  6. Riccardi G, Rivellese AA. Dietary treatment of the metabolic syndrome—the optimal diet. Br J Nutr 2000; 83(Suppl 1): S143–S148. | PubMed | ISI | ChemPort |
  7. Ludwig DS. Dietary glycemic index and obesity. J Nutr 2000; 130: 280S–283S. | PubMed | ISI | ChemPort |
  8. Kirk T, Crombie N, Cursiter M. Promotion of dietary carbohydrate as an approach to weight maintenance after initial weight loss: a pilot study. J Hum Nutr Diet 2000; 13: 277–285.
  9. Connor WE, Connor SL, Katan MB, Grundy SM, Willett WC. Should a low-fat, high-carbohydrate diet be recommended for everyone? N Eng J Med 1997; 337: 562–567.
  10. Seidell JC. Optimizing fat intake: does a reduction in fat intake prevent obesity? Eur Heart J Suppl 1999; 1 (Suppl S): S118–S122.
  11. Jenkins DJA, Wolever TMS, Taylor RH, Barker H, Fielden H, Baldwin JM, Bowling AC, Newman HC, Jenkins AL, Goff DV. Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr 1981; 34: 362–366. | PubMed | ISI | ChemPort |
  12. Wolever TMS, Jenkins DJA, Jenkins AL, Josse RG. The glycemic index: methodology and clinical implications. Am J Clin Nutr 1991; 54: 846–854. | PubMed | ISI | ChemPort |
  13. Dutch food consumption survey. Zo eet Nederland 1998. Voedingscentrum: Den Haag; 1998.
  14. Markus CR, Klöpping-Ketelaars WI, Pasman WJ, Klarenbeek B, van den Berg H. Dose-dependent effect of alpha-lactalbumin in combination with two different doses of glucose on the plasma Trp/LNAA ratio. Nutr Neurosci 2000; 3: 345–355.
  15. Jansen JBMJ, Lamers CBHW. Radioimmunoassay of cholecystokinin in human tissue and plasma. Clin Chim Acta 1983; 131: 305–316. | Article | PubMed | ISI | ChemPort |
  16. Wald FDM, Mellenbergh GJ. Instrumenteel onderzoek. De verkorte versie van de Nederlandse vertaling van de Profile of Mood States (POMS). Ned Tijdschr voor Psychol 1990; 45: 86–90.
  17. Smith A, Bazzoni C, Beale J, Elliott-Smith J, Tiley M. High fibre breakfast cereals reduce fatigue. Appetite 2001; 37: 249–250.
  18. Roberts SB. High-glycemic index foods, hunger, and obesity: is there a connection? Nutr Rev 2000; 58: 163–169. | PubMed | ChemPort |
  19. Sparti A, Milon H, Di Vetta V, Schneiter P, Tappy L, Jéquier E, Schutz Y. Effects of diets high or low in unavailable and slowly digestible carbohydrates on the pattern of 24-h substrate oxidation and feelings of hunger in humans. Am J Clin Nutr 2000; 72: 1461–1468. | PubMed | ChemPort |
  20. Ludwig DS, Majzoub JA, Al-Zahrani A, Dallal GE, Blanco I, Roberts SB. High glycemic index foods, overeating, and obesity. Pediatrics 1999; 103: E26. | Article | PubMed | ChemPort |
  21. Saris WHM, Astrup A, Prentice AM, Zunft HJF, Formiguera X, Verboeket-van de Venne WPHG, Raben A, Poppitt SD, Seppelt B, Johnston S, Vasilaras TH, Keogh GF. Randomized controlled trial of changes in dietary carbohydrate/fat ratio and simple vs complex carbohydrates on body weight and blood lipids: the CARMEN study. Int J Obes Relat Metab Disord 2000; 24: 1310–1318. | Article | PubMed | ChemPort |
  22. Pasman WJ, Wauters MAJ, Westerterp-Plantenga MS, Saris WHM. Effect of one week of fibre supplementation on hunger and satiety or energy intake. Appetite 1997; 29: 77–87. | Article | PubMed | ISI | ChemPort |
  23. Poppitt SD, Keogh GF, Prentice AM, Williams DEM, Sonnemans HMW, Valk EEJ, Robinson E, Wareham NJ. Long-term effects of ad libitum low-fat, high-carbohydrate diets on body weight and serum lipids in overweight subjects with metabolic syndrome. Am J Clin Nutr 2002; 75: 11–20. | PubMed | ISI | ChemPort |
  24. Pi-Sunyer FX. Glycemic index. Obes Res 2000; 8(Suppl 1); 1S, I8.
  25. Ludwig DS. The glycemic index. Physiological mechanisms relating to obesity, diabetes, and cardiovascular disease. JAMA 2002; 287: 2414–2423. | Article | PubMed | ISI | ChemPort |

Extra navigation

.

natureevents

ADVERTISEMENT