Secondary Logo

Journal Logo

BDNF, Metabolic Risk Factors, and Resistance Training in Middle-Aged Individuals


Medicine & Science in Sports & Exercise: March 2008 - Volume 40 - Issue 3 - p 535-541
doi: 10.1249/MSS.0b013e31815dd057
BASIC SCIENCES: Original Investigations

Introduction and Purpose: Brain-derived neurotrophic factor (BDNF) and physical inactivity contribute to the development of the metabolic syndrome (MetS). There appears to be an association between BDNF and risk factors for MetS, and the effects of resistance training (RT) on BDNF and metabolic risk in middle-aged individuals with high and low numbers of metabolic risk factors (HiMF and LoMF, respectively) are unclear and are the focus of this research.

Methods: Forty-nine men (N = 25) and women (N = 24) aged 50.9 ± 6.2 yr were randomized to four groups, HiMF training (HiMFT), HiMF control (HiMFC), LoMF training (LoMFT), and LoMF control (LoMFC). Before and after 10 wk of RT, participants underwent tests for muscle strength and anthropometry, and a fasting blood sample was taken. Data were analyzed using Spearman correlations and repeated-measures ANOVA.

Results: BDNF was positively correlated with plasma triglycerides, glucose, HbA1C, and insulin resistance. BDNF was elevated in HiMF compared with LoMF (904.9 ± 270.6 vs 709.6 ± 239.8 respectively, P = 0.01). Training increased muscle strength and lean body mass but had no effect on BDNF levels or any examined risk factors.

Conclusion: BDNF levels correlated with risk factors for MetS and were elevated in individuals with HiMF. RT had no effect on BDNF levels or other risk factors for MetS. As RT has an effect on muscle strength and lean body mass, it should be added to other nonpharmacological interventions for middle-aged individuals with HiMF such as aerobic and/or diet.

1Centre for Ageing, Rehabilitation, Exercise and Sport, School of Human Movement, Recreation and Performance, Victoria University, Melbourne, AUSTRALIA; 2Cellular and Molecular Metabolism Laboratory, Baker Heart Research Institute, Melbourne, AUSTRALIA; 3Department of Cardiology and University of Melbourne, Austin Health, Melbourne, AUSTRALIA; 4Department of Endocrinology and University of Melbourne, Austin Health, Melbourne, AUSTRALIA; and 5School of Exercise and Nutrition Sciences, Deakin University, Melbourne, AUSTRALIA

Address for correspondence: Itamar Levinger, M.Sc., School of Human Movement, Recreation and Performance, Victoria University Footscray Park Campus, Ballarat Road, Footscray 3011, Melbourne, VIC, Australia; E-mail:

Submitted for publication August 2007.

Accepted for publication October 2007.

Brain-derived neurotrophic factor (BDNF) is part of the neurotrophic factor family and is produced from the ventromedial hypothalamus (15). BDNF contributes to the development of the brain by regulating synaptic plasticity, neurogenesis, and neural survival (27). It has been reported that low levels of BDNF are associated with the development of neuropsychiatric disorders such as depression and Alzheimer disease (37). In the past decade, animals studies have shown that BDNF also acts on the endocrine system, contributing to the control of energy balance, glucose levels, and insulin resistance (22,28), thereby improving glycemic control (15,22). Some studies, but not all (36), have reported that overweight individuals with type 2 diabetes mellitus (T2DM) have lower levels of BDNF compared with healthy individuals, and that there is an inverse correlation between BDNF levels and fasting plasma glucose, obesity, and triglyceride levels, and a positive correlation with high-density lipoprotein (HDL) (16,21).

There is substantial evidence to suggest that physical inactivity is a major contributor to the development of obesity, hypertension, dyslipidemia, and insulin resistance/T2DM (19,25), all components of the metabolic syndrome (MetS) (40). Aerobic (endurance) exercise training improves risk factors for MetS and T2DM (2), lowers risk for brain diseases (1), improves depression, and increases BDNF levels (8,34,39).

In recent years, resistance training (RT) has been of great interest for patients with elevated fasting blood glucose, insulin resistance, and T2DM. Recently, we have reported that RT improved functional capacity and quality of life in this population (23). In addition, benefits of RT include improvements in fasting blood glucose (3,24), HbA1c (5,11,24), insulin resistance (5,20), and blood pressure (4) have been reported. However, other studies have reported no change in the above metabolic risk factors after RT (7,9,12,31). It is difficult to assess the role of RT in individuals with a cluster of metabolic risk factors, as some of the above studies involved either hybrid exercise training regimens (i.e., combined aerobic and resistance exercise) or double interventions (e.g., including diet). Currently there is a lack of randomized controlled data on the effects of RT as a single intervention in these individuals (38).

The associations between BDNF and risk factors for MetS, and the effects of RT on both BDNF and metabolic risk in individuals with two or more metabolic risk factors (HiMF) and low numbers of metabolic risk factors (LoMF) are unclear, and are the focus of this study. No previous studies have examined the effect of RT on BDNF levels. We hypothesized that BDNF levels would correlate with risk factors for MetS and T2DM but might not be sensitive to a RT intervention. In addition, we hypothesized that RT would have a limited effects on metabolic risk factors in individuals with HiMF.

Back to Top | Article Outline


Power analysis to estimate the number of participant needed to test the hypotheses was performed prior to data collection using the data from Park et al. (29). Although a minimum sample size of eight participants per group (or 32 participants overall) was needed to achieve a power of 80% at an alpha of 0.05, a total of 17 extra participants were included because they were already participating in a larger clinical trial. This also allowed for withdrawals of some volunteers before completion of the study.

Forty-nine untrained men (N = 25) and women (N = 24) aged 50.9 ± 6.2 yr (range 40-69 yr) volunteered to participate in the study. Following initial assessment, each participant was classified as having zero, one, two, three, four, or five risk factors for MetS. The International Diabetes Federation (40) criteria for each metabolic risk factor were used to categorize individuals. Each participant was then classified as either HiMF (number of metabolic risk factors ≥ 2) or LoMF (number of metabolic risk factors ≤ 1). They were then randomly assigned to allocated to one of four groups: HiMF training (HiMFT), HiMF nonexercise control (HiMFC), LoMF training (LoMFT), and LoMF nonexercise control (LoMFC). The rationale for the HiMF group allocation is that individuals with two or more risk factors are at high risk of developing MetS and T2DM (30). Randomization was also stratified according to sex, to ensure similar number of males and females between the HiMFT and HiMFC and LoMFT and LoMFC (Table 1). Participants were taking a range of medications, including beta-blockers (N = 2), calcium channel blockers (N = 1), angiotensin converting enzyme inhibitors (N = 2), statins (N= 2), metformin (N = 1), and hormone therapy (N = 5). Participants were included if they had not been involved in regular aerobic exercise training exceeding 60 min·wk−1 during the previous 6 months, or RT in the previous 5 yr.



Participants were excluded if they had documented cardiac disease. Each participant received written and verbal explanations about the nature of the study, and those who choose to participate signed an informed consent document. The study protocol was approved by the human research ethics committees of both Victoria University and Austin Health.

Back to Top | Article Outline

Study Protocol

At baseline and after 10 wk of RT, participants completed a 3-d dietary log and underwent a series of anthropometric measurements and tests to examine their metabolic risk factors and muscle strength.

Back to Top | Article Outline

Standard overnight blood test.

A blood sample was collected after 12 h of fasting. The blood was centrifuged and analyzed (SYNCHRON LX System/Lxi725, Beckman) for triglycerides, HDL, and glucose levels.

HbA1c was analyzed by Primus CLC 330 analyzer (Kansas City, MO), and insulin was analyzed by automated Immulite 2000 immunoassay system (Immulite 2000, DPC). Insulin resistance was estimated by the homeostasis model assessment (HOMA). This model has been validated against a hyperinsulinemic-euglycemic clamp (26). HOMA was calculated from fasting glucose and insulin:

On a different day, after at least 3 h of fasting, 20 mL of blood was taken and centrifuged, and plasma was immediately stored at −20°C until assayed. BDNF was analyzed with a standard Human BDNF enzyme-linked immunosorbant assay (ELISA) method (Catalog number: DY248; Minneapolis, MN).

Back to Top | Article Outline

Blood pressure.

Blood pressure was measured using a mercury sphygmomanometer after the participant had rested in a seated position for 15 min. Systolic and diastolic blood pressures were recorded to the nearest 2 mm Hg.

Back to Top | Article Outline

Three-day dietary log.

Participants received a kitchen scale (Kitchen Scale Model: KCHC-009) and recorded their diet in detail for two weekdays and a weekend day. The logs were analyzed by FoodWorks Professional Edition (Version 3.02.581, Xyris Software). In addition, during the study participants were encouraged not to alter their diet.

Back to Top | Article Outline

Anthropometric measurements.

Height was measured barefoot using a stadiometer to the nearest 0.5 cm. Weight was measured without clothes, using a scale (August Sauter GmbH), to the nearest 0.05 kg. Waist circumference was measured with a steel tape and taken as the circumference between the iliac crest and the lower border of the ribs. Three measurements were taken, and the mean of the two closest measures was recorded.

Dual-energy x-ray absorptiometry (DXA) (GE Lunar Pxodigy, Software version 9.1, Madison, WI) was used to assess total body fat percentages, total body fat, and lean body mass (LBM). All DXA measurements were performed at the bone density unit, Austin Health.

Back to Top | Article Outline

Muscle strength.

The full description of the muscle strength test has been reported previously (23). In brief, total muscle strength was calculated as the sum of seven different resistance exercises. Muscle strength was evaluated using the one-repetition maximum (1RM) method. 1RM was defined as the heaviest weight a participant could lift just once with a correct lifting technique, without compensatory movements.

Back to Top | Article Outline

Resistance Training Protocol

The full description of the training protocol has been reported previously (23). In brief, RT was performed 3 d·wk−1 for 10 wk, with 48 h of recovery between sessions. Training consisted of the seven exercises that were used in the 1RM strength tests. In addition, participants performed one abdominal exercise (abdominal curl). Initial training intensity was two sets of 15-20 repetitions at 40-50% 1RM. From weeks 2-10, participants performed each exercise for three sets, 8-20 repetitions, at 50-85% 1RM. At each session, weights were adjusted according to the current capacity of the individual.

Back to Top | Article Outline


Multivariate analysis of variance (MANOVA) was used to examine the differences of BDNF and the metabolic risk factors between HiMF and LoMF groups before each group was randomized to training or nonexercise control. MANOVA was then applied to BDNF and the metabolic risk factors following randomization to the subgroups, but before the exercise training (i.e., baseline levels for HiMFT vs HiMFC and LoMFT vs LoMFC). The training data were analyzed by the repeated-measures ANOVA model, which was constructed to analyze the effect of primary interest by time (pre and post) for each group. Repeated-measure ANOVA was also used to examine the effect of training over time between each of the two metabolic risk groups (i.e., LoMFT vs LoMFC and between HiMFC and HiMFT, referred as P value "group × time") and between the two training groups (i.e., LoMFT vs HiMFT). ANCOVA was used to examine the training effects on the BDNF levels of the HiMF groups because the HiMFC group had a higher mean BDNF level at baseline compared with the HiMFT. Spearman rho correlations were used to assess correlations between BDNF and other individual risk factors, and stepwise multiple regression analysis was used to identify the variable with the strongest correlation when risk factors were analyzed as a group. All data are reported as means ± standard deviation, and all statistical analyses were conducted at the 95% level of significance.

Back to Top | Article Outline


Correlation between BDNF and Other Risk Factors

Spearman rho values show that BDNF levels were positively correlated with the actual number (zero, one, two, three, four, or five) of metabolic risk factors a person has (r = 0.36, P = 0.01), triglyceride (r = 0.39, P < 0.01), fasting plasma glucose levels (r = 0.37, P = 0.01), HbA1c (r = 0.29, P = 0.05), plasma insulin level (r = 0.36, P = 0.01), HOMA (r = 0.37, P = 0.01), and abdominal fat (r = 0.37, P = 0.01). BDNF was negatively correlated with HDL (r = −0.36, P = 0.02). No correlation was found between BDNF and blood pressure (P > 0.05). When factors were taken together within one stepwise multiple regression model, triglyceride was the strongest predictor for BDNF levels (r = 0.45, P < 0.01).

Back to Top | Article Outline

Metabolic Differences at Baseline between the HiMF and LoMF

BDNF was significantly elevated in HiMF compared with LoMF (904.9 ± 270.6 vs 709.6 ± 239.8 pg·mL−1, P = 0.01). The HiMF group had higher waist circumference (101.8 ± 10.8 vs 80.2 ± 9.1 cm, P < 0.01), systolic (133.0 ± 13.8 vs 117.3 ± 12.5 mm Hg, P < 0.01) and diastolic 87.5 ± 9.3 vs 77.8 ± 6.5 mm Hg, P < 0.01) blood pressure, triglyceride (1.5 ± 0.6 vs 0.7 ± 0.3 mM, P < 0.01) and lower HDL (1.5 ± 0.5 vs 1.8 ± 0.6 mM, P = 0.02) compared with the LoMF group. In addition, the HiMF group had higher fasting glucose (5.7 ± 0.5 vs 4.9 ± 0.3 mM, P < 0.01) and insulin level (54.8 ± 33.7 vs 24.1 ± 19.4 pM, P < 0.01). There were no significant differences between LoMFC and LoMFT, or between HiMFC and HiMFT, for sex, anthropometric measurements, blood pressure, lipid profile, or fasting glucose level (Table 1). Despite the randomization, at baseline, BDNF level was higher in the HiMFC versus the HiMFT (P = 0.01) (Table 1 and Fig. 1).



Back to Top | Article Outline

The Effect of Resistance Training

Adherence to training.

Forty-five of 49 participants completed the study (two from the HiMFT group and two from the LoMFC group withdrew from the study, and their data were excluded). Adherence to training was 88 ± 8.3% for the HiMFT group and 96 ± 6.5% for the LoMFT group. At baseline, two participants from the training and three from the control groups did not complete the 1RM test for the leg extension exercise, because each possessed strength exceeding the available loads on the machine. The strength test for leg extension was not conducted at end point for these individuals. One other participant from the HiMFT was not tested for leg extension because of an unrelated knee injury.

Back to Top | Article Outline

Diet comparison.

No changes in total energy intake were observed during the study within each group (P > 0.10), and there were no significant interactions between LoMFC and LoMFT (P = 0.65), nor between HiMFC and HiMFT (P = 0.20). No change was observed in the energy macronutrient composition of the diet within or between groups, with carbohydrate contributing 53.8 ± 9.8%, protein 24.2 ± 5.9%, and fat 22.0 ± 5.4%, of total energy intake.

Back to Top | Article Outline

The effect of training on muscle strength and body composition.

The effect of training on muscle strength and body composition has been described previously in detail (23). In brief, both HiMFT and LoMFT training significantly increased total muscle strength (25.0 ± 6.2%) and LBM (1.1 kg) and decreased total fat percentage (~1.2 kg).

Back to Top | Article Outline

The effect of training on BDNF and risk factors for MetS.

BDNF levels did not change significantly in response to training in any group (all P > 0.28) (Fig. 1). In addition, training had no significant effects on blood pressure (systolic and/or diastolic), triglyceride, HDL, or fasting plasma glucose (all P > 0.05), both within and between groups (Table 2). Waist circumference was significantly increased (1.6 cm, P < 0.05) in the LoMFC group compared with the pretraining value. Although the HiMFT group exhibited a 1.5-cm reduction in waist circumference, it did not reach statistical significance (P = 0.07).



Back to Top | Article Outline


The main findings of the current study are that BDNF was positively correlated with the number of metabolic risk factors, and with abdominal fat, triglyceride, fasting plasma glucose, HbA1c, plasma insulin, and insulin resistance, and BDNF was negatively correlated with HDL in middle-aged individuals. In addition, RT as a single intervention had no effect on BDNF level or other metabolic risk factors examined here for apparently healthy middle-aged individuals, and those with two or more metabolic risk factors.

Recently, it has been shown that BDNF may play a role in the development of risk factors associated with the MetS, also called the "metabolic syndrome-neurotrophic hypothesis" (18). It seems that BDNF levels and its associations with other risk factors may change with the duration of metabolic risk factor exposure and the prevalence of other chronic diseases, such as T2DM. For instance, normal or elevated BDNF levels have been described at early stages of T2DM, and low levels have been described as the disease progresses (16,18). In the current study, we found a positive correlation between BDNF and other metabolic risk factors in individuals who had not yet developed overt disease. Similarly, Suwa et al. (36) report a significant positive correlation between BDNF and BMI (r = 0.54), percentages of body fat (r = 0.56), triglyceride level (r = 0.47), fasting glucose level (r = 0.44), and insulin resistance (r = 0.51) in newly diagnosed females with T2DM. In addition, they report that BDNF was elevated in females with T2DM compared with those with normal glucose tolerance. The authors suggest that BDNF is upregulated as a compensatory mechanism in the early stages of the disease process. However, an increase in BDNF levels (oversecretion) at an early stage of the MetS leads to imbalances in the autonomic nervous system and in the interaction between the neuroimmunoendocrine systems that may lead to a reduction in BDNF levels, compared with healthy controls, in later stages of the disease process (18). This may help to explain why individuals with longstanding T2DM (16,21) and neuropsychiatric disorders (37) exhibited lower levels of BDNF compared with healthy controls. In addition, negative correlations of BDNF with BMI (r = −0.12), HbA1c (r = −0.11), and triglyceride (r = −0.21), and positive correlation with HDL (r = 0.34), have been reported in patients with chronic disease (16).

Reductions in physical activity levels and aging have been shown to be important contributors to the development of obesity, hypertension, dyslipidemia, and insulin resistance. Increases in physical activity should lead to improvements in metabolic risk profiles for apparently healthy middle-aged individuals, and for those with multiple numbers of metabolic risk factors. In the current study, we found that a single intervention of RT did not change BDNF levels or improve metabolic risk profiles of middle-aged individuals with low and high numbers of risk factors for MetS. Although the participants were randomized to HiMFT or HiMFC, at baseline the HiMFT had significantly lower BDNF levels compared with HiMFC. However, it appears that this did not affect the results, because no change was observed in the BDNF levels of the HiMFT (although they started from lower values). The lack of change in BDNF levels or other examined metabolic risk factors occurred despite an increase in muscle strength (on the order of 24%) and improvements in LBM and body fat. We have previously reported that increases in muscle strength and LBM contribute to an improvement in the capacity to perform activities of daily living and quality of life in this population (20). To our knowledge, this is the first study that has examined the effect of RT on BDNF. In light of the present data, and the time-dependent changes in BDNF outlined above, it appears RT does not change BDNF levels in middle-aged individuals or in those with high exposure to metabolic risk factors in the early stages of T2DM. However, it is possible that RT will cause favorable changes to BDNF levels in individuals with low levels of BDNF associated with chronic, overt disease.

Some studies have previously reported that RT can reduce fasting blood glucose and improve glycemic control (3,6,24) and insulin resistance (5,33) and reduce blood pressure (4) in people with T2DM and those with hypertension. However, the findings from the current study and other studies (7,9,12,31,32) suggest that RT has a limited effect on the metabolic risk profile of middle-aged individuals with HiMF. Some studies have reported that RT has no effect on plasma glucose and insulin levels at a "basal" state but can result in improvements to insulin action during insulin stimulation (hyperinsulinemic-euglycemic clamp) (13,32). This may indicate that RT has a limited effect on insulin resistance and insulin action at rest, as reported in the current study, but may improve tissue sensitivity and insulin action in the presence of elevated insulin (such as during postprandial periods). The contrasting findings between the studies that reported changes in risk factors after RT and those that did not find changes in risk factors after RT may also be related to different clinical populations, different methods to assess risk factors, age of participants and/or duration of the disease, differing exercise training protocols such as hybrid training protocol (aerobic and resistance exercises), or multiple interventions of exercise and diet. The data from this study and from others outlined above suggest that RT as a single intervention (i.e., no dietary modification) has little or no effect on metabolic risk factors for apparently healthy middle-aged individuals and for those at high risk for developing T2DM. RT may improve metabolic risk profiles in elderly individuals with overt T2DM or similar cohorts as studied here if RT is combined with aerobic training and diet.

A potential limitation of the study is the possibility for gender bias between the HiMF (male: female 19:10) and LoMF (6:14) groups. It is important to note that of the total sample of 49 volunteers, 25 were male and 24 were female. The allocation to HiMF or LoMF groups was based on objective IDF criteria. Another possible limitation is the measurement of resting, fasted blood samples before and after training, which may not adequately reflect the kinetics of BDNF release. It has been shown that acute exercise (at least aerobic exercise) can have an immediate or short-term effect on serum BDNF levels (increased BDNF) in humans (14) and may increase BDNF levels in the hippocampus (35) and skeletal muscle (10) of rats. The changes in BDNF appear to be related to exercise intensity, lactate production, and corticosterone (14,35).Thus, it may be possible that our sampling regimen missed these changes. It should be noted, however, that addressing the acute effects of exercise on BDNF was not the aim of this study.

Back to Top | Article Outline


BDNF was positively correlated with risk factors for MetS and T2DM in this study, and this may possibly be explained by the relatively early stage of risk factor exposure and absence of chronic overt disease. In addition, RT as a single intervention did not change BDNF levels or metabolic risk profiles of middle-aged individuals and those with multiple risk factors for MetS, even though the exercise was sufficient to produce significant improvements in muscle strength, LBM, and body fat. Longer training durations may be needed to elicit improvements in metabolic risk factors in individuals with HiMF. In middle-aged individuals and in those with multiple metabolic risk factors, RT should be added to aerobic training and/or diet.

Back to Top | Article Outline


1. Abbott RD, White LR, Ross GW, Masaki KH, Curb JD, Petrovitch H. Walking and dementia in physically capable elderly men. JAMA. 2004;292:1447-53.
2. American College of Sports Medicine. Exercise and type 2 diabetes. Med Sci Sports Exerc. 2000;32(7):1345-60.
3. Balducci S, Leonetti F, Di Mario U, Fallucca F. Is a long-term aerobic plus resistance training program feasible for and effective on metabolic profile in type 2 diabetic patients. Diabetes Care. 2004;27:841-2.
4. Blumenthal J, Siegel WC, Appelbaum M. Failure of exercise to reduce blood pressure in patients with mild hypertension. JAMA. 1991;266:2098-104.
5. Brooks N, Layne JE, Gordon PL, Roubenoff R, Nelson ME, Castaneda-Sceppa C. Strength training improves muscle quality and insulin sensitivity in Hispanic older adults with type 2 diabetes. Int J Med Sci. 2007;4:19-27.
6. Castaneda C, Layne JE, Munoz-Orians L, et al. A randomized controlled trial of resistance exercise training to improve glycemic control in older adults with type 2 diabetes. Diabetes Care. 2002;25:2335-41.
7. Cornelissen VA, Fagard RH. Effect of resistance training on resting blood pressure: a meta-analysis of randomized controlled trials. J Hypertens. 2005;23:251-9.
8. Cotman CW, Engesser-Cesar C. Exercise enhances and protects brain function. Exerc Sport Sci Rev. 2002;30(2):75-9.
9. Cuff DJ, Meneilly GS, Martin A, Ignaszewski A, Tildesley HD, Frohlich JJ. Effective exercise modality to reduce insulin resistance in women with type 2 diabetes. Diabetes Care. 2003;26:2977-82.
10. Cuppini R, Sartini S, Agostini D, et al. Bdnf expression in rat skeletal muscle after acute or repeated exercise. Arch Ital Biol. 2007;145:99-110.
11. Dunstan DW, Daly RM, Owen N, et al. High-intensity resistance training improves glycemic control in older patients with type 2 diabetes. Diabetes Care. 2002;25:1729-37.
12. Dunstan DW, Puddey IB, Beilin LJ, Burke V, Morton AR, Stanton KG. Effects of a short-term circuit weight training program on glycaemic control in NIDDM. Diabetes Res Clin Pract. 1998;40:53-61.
13. Ferrara CM, Goldberg AP, Ortmeyer HK, Ryan AS. Effects of aerobic and resistive exercise training on glucose disposal and skeletal muscle metabolism in older men. J Gerontol. 2006;61:480-7.
14. Ferris LT, Williams JS, Shen CL. The effect of acute exercise on serum brain-derived neurotrophic factor levels and cognitive function. Med Sci Sports Exerc. 2007;39(4):728-34.
15. Fox EA, Byerly MS. A mechanism underlying mature-onset obesity: evidence from the hyperphagic phenotype of brain-derived neurotrophic factor mutants. Am J Physiol. 2004;286:R994-1004.
16. Geroldi D, Minoretti P, Emanuele E. Brain-derived neurotrophic factor and the metabolic syndrome: more than just a hypothesis. Med Hypotheses. 2006;67:195-6.
17. Haffner SM, Kennedy E, Gonzalez C, Stern MP, Miettinen H. A prospective analysis of the HOMA model. The Mexico City Diabetes Study. Diabetes Care. 1996;19:1138-41.
18. Hristova M, Aloe L. Metabolic syndrome-neurotrophic hypothesis. Med Hypotheses. 2006;66:545-9.
19. Hu FB, Leitzmann MF, Stampfer MJ, Colditz GA, Willett WC, Rimm EB. Physical activity and television watching in relation to risk for type 2 diabetes mellitus in men. Arch Intern Med. 2001;161:1542-8.
20. Ishii T, Yamakita T, Sato T, Tanaka S, Fujii S. Resistance training improves insulin sensitivity in NIDDM subjects without altering maximal oxygen uptake. Diabetes Care. 1998;21:1353-5.
21. Krabbe KS, Nielsen AR, Krogh-Madsen R, et al. Brain-derived neurotrophic factor (BDNF) and type 2 diabetes. Diabetologia. 2007;50:431-8.
22. Lebrun B, Bariohay B, Moyse E, Jean A. Brain-derived neurotrophic factor (BDNF) and food intake regulation: a minireview. Auton Neurosci. 2006;126-127:30-8.
23. Levinger I, Goodman C, Hare DL, Jerums G, Selig S. The effect of resistance training on functional capacity and quality of life in individuals with high and low numbers of metabolic risk factors. Diabetes Care. 2007;30:2205-10.
24. Maiorana A, O‘Driscoll G, Goodman C, Taylor R, Green D. Combined aerobic and resistance exercise improves glycemic control and fitness in type 2 diabetes. Diabetes Res Clin Pract. 2002;56:115-23.
25. Martinez JA. Body-weight regulation: causes of obesity. Proc Nutr Soc. 2000;59:337-45.
26. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28:412-9.
27. Mattson MP, Maudsley S, Martin B. BDNF and 5-HT: A dynamic duo in age-related neuronal plasticity and neurodegenerative disorders. Trends Neurosci. 2004;27:589-94.
28. Nonomura T, Tsuchida A, Ono-Kishino M, Nakagawa T, Taiji M, Noguchi H. Brain-derived neurotrophic factor regulates energy expenditure through the central nervous system in obese diabetic mice. Int J Exp Diabetes Res. 2001;2:201-9.
29. Park SK, Park JH, Kwon YC, Kim HS, Yoon MS, Park HT. The effect of combined aerobic and resistance exercise training on abdominal fat in obese middle-aged women. J Physiol Anthropol Appl Human Sci. 2003;22:129-35.
30. Reaven GM. The metabolic syndrome: is this diagnosis necessary? Am J Clin Nutr. 2006;83:1237-47.
31. Rice B, Janssen I, Hudson R, Ross R. Effects of aerobic or resistance exercise and/or diet on glucose tolerance and plasma insulin levels in obese men. Diabetes Care. 1999;22:684-91.
32. Ryan AS, Hurlbut DE, Lott ME, et al. Insulin action after resistive training in insulin resistant older men and women. J Am Gerontol Soc. 2001;49:247-53.
33. Ryan AS, Pratley RE, Goldberg AP, Elahi D. Resistive training increases insulin action in postmenopausal women. J Gerontol Med Sci. 1996;51A:M199-205.
34. Sarbadhikari SN, Saha AK. Moderate exercise and chronic stress produce counteractive effects on different areas of the brain by acting through various neurotransmitter receptor subtypes: a hypothesis. Theor Biol Med Mod. 2006;3:33-50.
35. Soya H, Nakamura T, Deocaris CC, et al. BDNF induction with mild exercise in the rat hippocampus. Biochem Biophys Res Commun. 2007;358:961-7.
36. Suwa M, Kishimoto H, Nofuji Y, et al. Serum brain-derived neurotrophic factor level is increased and associated with obesity in newly diagnosed female patients with type 2 diabetes mellitus. Metabolism. 2006;55:852-7.
37. Tsai SJ. Brain-derived neurotrophic factor: a bridge between major depression and Alzheimer‘s disease? Med Hypotheses. 2003;61:110-3.
38. Willey KA, Fiatarone-Singh MA. Battling insulin resistance in elderly obese people with type 2 diabetes: bring on the heavyweights. Diabetes Care. 2003;26:1580-9.
39. Winter B, Breitenstein C, Mooren FC, et al. High impact running improves learning. Neurobiol Learn Mem. 2007;87:597-609.
40. Zimmet PZ, Alberti KG, Shaw JE. Mainstreaming the metabolic syndrome: a definitive definition. This new definition should assist both researchers and clinicians. Med J Aust. 2005;183:175-6.


©2008The American College of Sports Medicine