Brain-derived neurotrophic factor (BDNF) is a member of the neurotrophin family, which includes nerve growth factor, neurotrophin-3, and neurotrophin-4/5 [
In addition to the role of BDNF in neurological disorders, recent studies reported that peripheral injection of BDNF exerts hypophagic and hypoglycemic effects on obese hyperglycemic animals but not on normal animals, pointing to antiobesity and antidiabetic effects [
This study was performed at the GOP Taksim Education and Research Hospital outpatient department of internal medicine. It included 88 patients (38 males and 50 females) and 33 control subjects (17 males and 16 females). The presence of diabetes was based on a previous diagnosis of T2DM or a random plasma glucose level of 200 mg/dL or higher, together with classical features of DM, such as polyuria, polydipsia, polyphagia, and weight loss, or a fasting blood glucose level of >126 mg/dL or higher or a HbA1C level of 6.5% or higher. Exclusion criteria were the presence of systemic diseases: neoplastic, inflammatory, and infectious diseases. The study protocol was approved by GOP Taksim Research and Education Hospital ethics committee, Istanbul. Informed written consent was obtained from all the participants (patients and controls) after receiving a full explanation about the study and its purpose.
Hypertension was defined as antihypertensive drug use or systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg. Body mass index (BMI) was obtained using the formula weight (kg)/height (m)2. Obesity was defined as a BMI >30 kg/m2. Waist circumference was measured in a standing position at the level of the umbilicus.
Blood samples were obtained after overnight fasting. Serum cholesterol, triglyceride, and high-density lipoprotein cholesterol (HDL-C) were measured by enzymatic colorimetric methods with commercially available kits (COBAS 311, Roche Diagnostics GmbH, Mannheim, Germany), and low-density lipoprotein cholesterol C (LDL-C) was calculated according to the Friedewald formula. Serum glucose measures were determined enzymatically using the hexokinase method (Roche Diagnostics GmbH, Mannheim, Germany). Blood HbA1c was determined with a COBAS 311 analyzer using the particle-enhanced immunoturbidimetric method (Roche Diagnostics, Mannheim, Germany). Final results were expressed as percent HbA1c of the total Hb according to the protocol of the Diabetes Control and Complications Trial/National Glycohemoglobin Standardization Program (DCCT/NGSP). The particle-enhanced immunoturbidimetric method with a Behring Nephelometer BN-100 (Behring Diagnostic, Frankfurt, Germany) was used to measure C-reactive protein (CRP). The sensitivity of the test was 0.1 mg/L. White blood cell (WBC) levels were measured with an automatic hematology analyzer (Beckman Coulter, Brea, CA, USA). The erythrocyte sedimentation rate (ESR) was determined with the Westergren method using an established normal range of 0–20 mm/1 hr. Ferritin was measured with an electrochemiluminescence immunoassay (Roche Hitachi Modular E 170). The insulin level was determined with an electrochemiluminescence immunoassay (Roche Diagnostics, Mannheim, Germany) on an automated Roche Cobas E 411 (Roche Diagnostics). The homeostasis model assessment of insulin resistance (HOMA-IR) index was calculated from the fasting blood glucose and fasting serum insulin concentrations by the formula: HOMA-IR = fasting serum insulin (
Number Cruncher Statistical System (NCSS) 2007 and Power Analysis and Sample Size (PASS) 2008 Statistical Software (Utah, USA) programs were used for the statistical analysis. Descriptive statistical methods (mean, standard deviation, frequency, ratio, minimum, and maximum) were used to evaluate the study data. Student’s
The clinical characteristics of the study subjects are shown in Table
Clinical characteristics of patients and controls.
Patient group ( |
Control group ( |
|||
---|---|---|---|---|
Mean ± s.d./ |
Mean ± s.d./ |
|||
Age (years) | 60.03 | ±12.22 | 58.82 | ±10.80 |
Gender | ||||
Male | 38 | 43.2% | 17 | 51.5% |
Female | 50 | 56.8% | 16 | 48.5% |
BMI (kg/m2) | 31.52 | ±5.80 | 26.87 | ±3.89*** |
Waist circumference (cm) | 108.75 | ±14.07 | 94.36 | ±9.75*** |
Systolic pressure (mmHg) | 130.51 | ±14.40 | 118.03 | ±8.00*** |
Diastolic pressure (mmHg) | 81.25 | ±8.17 | 74.24 | ±7.82*** |
BDNF (pg/mL) | 206.81 | ±107.32 | 130.84 | ±59.81*** |
Fasting blood glucose (mg/dL) | 170.36 | ±91.21 | 85.33 | ±11.49*** |
Fasting insulin ( |
9.26 | ±4.92 | 5.48 | ±2.13*** |
HOMA-IR | 3.73 | ±3.05 | 1.16 | ±0.48*** |
HbA1c (%) | 8.30 | ±2.37 | 5.42 | ±0.52*** |
Triglyceride (mg/dL) | 175.95 | ±108.67 | 156.30 | ±65.02 |
Total cholesterol (mg/dL) | 212.21 | ±35.97 | 199.20 | ±52.50 |
LDL (mg/dL) | 128.18 | ±32.17 | 115.52 | ±44.26 |
HDL (mg/dL) | 47.01 | ±13.54 | 48.27 | ±13.20 |
CRP (mg/L) | 7.84 | ±2.84 | 2.57 | ±1.07*** |
ESR (mm/hr) | 26.09 | ±10.54 | 19.79 | ±5.58*** |
White blood cell (×103/ |
7278.41 | ±2190.07 | 6624.24 | ±1683.38 |
Ferritin (ng/mL) | 84.88 | ±38.69 | 44.52 | ±26.58 |
Creatinine clearance (mL/min) | 91.74 | ±40.88 | 158.54 | ±83.23*** |
Oral antidiabetic drug | 56 | 64% | 0 | 0% |
Oral antidiabetic drug plus insulin | 16 | 18% | 0 | 0% |
Insulin | 16 | 18% | 0 | 0% |
Antihypertension | 54 | 61% | 0 | 0% |
Antilipid | 26 | 30% | 0 | 0% |
Smoking history | 31 | 35.2% | 0 | 0% |
Student’s
Statistical significance: ***
BMI: body mass index, BDNF: brain-derived neurotrophic factor, HOMA-IR: homeostasis model assessment of insulin resistance, LDL: low-density lipoprotein, HDL: high-density lipoprotein, CRP: C-reactive protein, and ESR: erythrocyte sedimentation rate.
35.2% of patients have cigarette smoking history (Table
The serum BDNF levels showed a positive correlation with HOMA-IR (
Relationship between HOMA-IR, triglyceride levels, WBC levels, and BDNF in T2DM patients.
Patient group | ||
---|---|---|
|
| |
BDNF and HOMA-IR | 0.281 |
|
BDNF and triglyceride | 0.265 |
|
BDNF and WBC | 0.355 |
|
Statistical significance: *
BDNF: brain-derived neurotrophic factor, HOMA-IR: homeostasis model assessment of insulin resistance, and WBC: white blood cell.
In the logistic regression analysis, age (
Logistic regression analysis of independent factors associated with T2DM.
|
Exp. ( |
95% CI for Exp. ( |
||
---|---|---|---|---|
Lower | Upper | |||
Age |
|
1.093 | 1.019 | 1.172 |
CRP |
|
1.806 | 1.129 | 2.888 |
BDNF |
|
9.987 | 2.417 | 41.267 |
BMI |
|
7.377 | 1.043 | 52.197 |
Statistical significance: *
CRP: C-reactive protein, BDNF: brain-derived neurotrophic factor, and BMI: body mass index.
The serum BDNF levels were significantly higher in the T2DM patients compared to the healthy controls (206.81 ± 107.32 pg/mL versus 130.84 ± 59.81 pg/mL,
Diagnostic screening tests results for BDNF.
BDNF | Sensitivity | Specificity | Positive predictive value | Negative predictive value | Validity |
---|---|---|---|---|---|
≥120 | 76.92 | 60.00 | 85.71 | 45.45 | 72.82 |
≥128 | 74.36 | 60.00 | 85.29 | 42.86 | 70.87 |
≥130 | 73.08 | 60.00 | 85.07 | 41.67 | 69.90 |
≥137 |
|
|
|
|
|
≥142 | 69.23 | 68.00 | 87.10 | 41.46 | 68.93 |
≥145 | 66.67 | 72.00 | 88.14 | 40.91 | 67.96 |
≥150 | 65.38 | 72.00 | 87.93 | 40.00 | 66.99 |
ROC curve.
In this study, we investigated changes in the plasma BDNF level and correlations between BDNF and clinical and biochemical parameters in T2DM patients. We found that the serum BDNF of T2DM patients was significantly higher than that of control subjects. A previous study showed that serum BDNF levels were elevated in newly diagnosed female T2DM patients compared to healthy subjects [
On the other hand, Fujinami et al. reported that serum BDNF levels were significantly lower in patients with advanced T2DM compared to control subjects [
In a previous study, BDNF heterozygous knockout (BDNF +/−) mice showed obesity and insulin resistance [
Previous studies described an association of plasma BDNF with inflammatory conditions [
In the logistic regression analysis, BDNF was independently associated with T2DM, irrespective of age, BMI, and CRP. Patients who had a serum BDNF level more than 137 pg/mL were a predictive value like HbA1c for the diabetes diagnosis with a sensitivity of 71.79% and a specificity of 68%. It had a positive predictive value of 87.5% and a negative predictive value of 43.59%. This novel research suggests that serum BDNF may be used as a prediction data for T2DM like HgA1c in future. More importantly, the standardization and cut-off values of serum BDNF had been controversial in previous studies. For the first time, current study suggests a cut-off point for serum BDNF of type 2 diabetes mellitus.
There are some limitations in this study. First, this was a cross-sectional study of a relatively small number of patients. A larger patient population should be recruited. Second, we measured levels of BDNF but not those of other neurotrophins. Third, we measured levels of WBCs and CRP to determine the association between BDNF values and inflammation but not those of other serum inflammatory markers, such as interleukin-6 or tumor necrosis factor.
In conclusion, serum BDNF level was higher in patients with T2DM, and the cut-off predictive value of BDNF was 137 pg/mL. The findings of this study suggest that BDNF may contribute to glucose and lipid metabolism and inflammation.
The authors declare that there is no conflict of interests regarding the publication of this paper.