Journal Logo

Advanced Search
APPLIED SCIENCES

Sedentarism, Physical Activity, Steps, and Neurotrophic Factors in Obese Children

MORA-GONZALEZ, JOSE1; MIGUELES, JAIRO H.1; ESTEBAN-CORNEJO, IRENE1,2; CADENAS-SANCHEZ, CRISTINA1; PASTOR-VILLAESCUSA, BELÉN3,4; MOLINA-GARCÍA, PABLO1,5; RODRIGUEZ-AYLLON, MARÍA1; RICO, MARÍA C.4; GIL, ANGEL4,6,7; AGUILERA, CONCEPCIÓN M.4,6,7; ESCOLANO-MARGARIT, MARÍA VICTORIA8; GEJL, ANNE KAER9; ANDERSEN, LARS BO10; CATENA, ANDRÉS11; ORTEGA, FRANCISCO B.1

Author Information

1PROFITH “PROmoting FITness and Health through physical activity” Research Group, Sport and Health University Research Institute (iMUDS), Department of Physical and Sports Education, Faculty of Sports Science, University of Granada, Granada, SPAIN

2Department of Psychology, Northeastern University, Boston, MA

3LMU - Ludwig-Maximilians-Universität München, Division of Metabolic and Nutritional Medicine, Dr. von Hauner Children’s Hospital, University of Munich Medical Center, Munich, GERMANY

4Department of Biochemistry and Molecular Biology II, Institute of Nutrition and Food Technology “José Mataix,” Center of Biomedical Research, University of Granada, Granada, SPAIN

5Department of Rehabilitation Sciences, KU Leuven – University of Leuven, Leuven, BELGIUM

6Instituto de Investigación Biosanitaria ibs. GRANADA, Granada, SPAIN

7CIBER Fisiopatología de la Obesidad y la Nutrición (CIBEROBN), Madrid, SPAIN

8Hospital Universitario San Cecilio, Granada, SPAIN

9Department of Physiotherapy and Occupational Therapy, University College Copenhagen, Copenhagen, DENMARK

10Western Norway University of Applied Sciences, Campus Sogndal, Faculty of Education, Arts and Sport, NORWAY

11Department of Experimental Psychology, Mind, Brain and Behaviour Research Centre (CIMCYC), University of Granada, Granada, SPAIN

Address for correspondence: Jose Mora-Gonzalez, M.Sc., Department of Physical and Sports Education, Faculty of Sports Science, University of Granada; Carretera de Alfacar, 21. Granada 18071, Spain; E-mail: [email protected].

Submitted for publication December 2018.

Accepted for publication May 2019.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.acsm-msse.org).

Medicine & Science in Sports & Exercise: November 2019 - Volume 51 - Issue 11 - p 2325-2333
doi: 10.1249/MSS.0000000000002064

Abstract

Childhood obesity has shown to be negatively related to cognitive functions and detectable structural abnormalities in the brain (1,2). Likewise, obesity may also influence stored and circulating neurotrophic factors such as brain-derived neurotrophic factor (BDNF) in humans (3), although literature to this respect is inconsistent (4). It has been observed that BDNF missense mutations in its receptor, TrkB, have been associated with weight gain in both humans and mice (4). Further, evidence has shown a significant reduction of circulating BDNF levels in children with obesity compared with normal-weight peers (5,6). Importantly, BDNF plays a key role in synaptic plasticity, neuronal transmission, and cell growth and survival throughout the cortex (7). This factor is produced in the brain and in selected peripheral tissues such as platelets (8). Platelets are the major nonneural source of BDNF from which it reaches plasma and is able to pass the blood–brain barrier, only when it is not bound to platelets (9,10). Interestingly, the positive correlation between BDNF in the brain and circulating BDNF suggests that circulating BDNF levels may reflect the levels in the central nervous system (11). Furthermore, BDNF may be released from the brain to the periphery during the practice of physical activity (PA) (12). Apart from BDNF, other neurotrophic factors such as vascular endothelial growth factor (VEGF) or insulin growth factor-1 (IGF-1) are important for neural growth and neuron survival (13). Hence, it seems of relevance to examine how protective environmental factors, such as lifestyle behaviors (e.g., sedentarism or PA), may influence neurotrophic factors in a particularly vulnerable population such as children with overweight/obesity.

Emerging evidence suggest that PA has a beneficial effect on the brain and cognitive processes in children (14). Neurotrophic factors have been suggested as potential mechanisms underlying this relationship (15). From all these factors, BDNF may be the most important one that has been suggested to be upregulated by PA (13). Indeed, BDNF may play a crucial role in the PA’s influence on brain structure and as an underlying factor of the PA-induced cognitive improvement. However, in humans, there is inconsistent evidence on the role of PA on neurotrophic factors (16–19). Physical activity may increment serum BDNF concentrations in adolescents (16,19) and adults (18), although there are other studies showing a negative association between PA and BDNF (17). In children, to the best of our knowledge, there are only two observational studies, and they did not find significant associations (20,21). However, no previous cross-sectional studies have focused on children with obesity nor have analyzed the role of step-related behaviors on neurotrophic factors. In addition, the BDNF plays a key role in the energy homeostasis and the appetite regulation (22), which highlights even more the importance of examining the potential relationship of sedentary time and PA with brain in the context of obesity during childhood. Particularly, walking (hereinafter step-related behaviors) is the most popular PA behavior, as well as the ideal PA intervention to improve health across sedentary populations, such as the obese ones (23,24). Thus, the aim of the present study was to analyze the association of sedentary time, PA, and step-related behaviors with BDNF and other neurotrophic factors (i.e., VEGF and IGF-1) in children with overweight/obesity.

METHODS

Participants

The present cross-sectional study was developed under the framework of the ActiveBrains project (http://profith.ugr.es/activebrains) (25). A total of 110 children with overweight/obesity age 8 to 11 yr were recruited from Granada (Spain) after meeting the defined inclusion criteria, which have been described elsewhere (25). The study was conducted in three waves. The present cross-sectional analyses used baseline data from 97 children with overweight/obesity (10.0 ± 1.2 yr; 58% boys) with complete baseline data on sedentary time, time-based PA, steps-related behaviors, and neurotrophic factors. For VEGF analyses, a sample of 88 participants was used after excluding those children with lower VEGF levels than the kit could detect. The baseline data collection took part from November 2014 to February 2016.

A description of the purpose and characteristics of the study was given to the parents or legal guardian and written informed consent was provided by them allowing the child to participate. The ActiveBrains project was approved by the Ethics Committee on Human Research of the University of Granada and was registered in ClinicalTrials.gov (identifier: NCT02295072).

Sedentary time, PA, and steps metrics

Sedentary time, PA, and step-related behaviors were assessed by accelerometer (GT3X+; ActiGraph, Pensacola, FL), taking into account the latest advances in data processing (26). Children wore simultaneously two accelerometers located on the right hip and nondominant wrist for seven consecutive days (24 h·d−1). They were instructed to remove them only for water activities (i.e., bathing or swimming) and to record waking-up and sleep onset times during the 7 d on a diary. Raw data were collected at a sampling frequency of 100 Hz were loaded in ActiLife (ActiGraph) and processed then in R (v. 3.1.2, https://www.cran.r-project.org/) using the GGIR package (v. 1.6–0, https://cran.r-project.org/web/packages/GGIR/) (27). We calculated the Euclidean Norm Minus One G metric (ENMO, G ~ 9.8 m·s−2) after auto-calibrating the acceleration signal (27,28). The mean of ENMO with negative values rounded to zero was calculated over 5 s epochs. Simultaneously, we derived the number of steps per minute (step cadence) from the hip-worn accelerometer using the ActiLife software. Then, we imported steps information to R for further analyses in the GGIR package.

Accelerometric information processing in GGIR consisted in: (a) nonwear time detection by the Van Hees et al. approach (29). (b) Detection of abnormally and sustained high acceleration values (i.e., clipped time). (c) Replacement of the nonwear and clipped time by the mean acceleration recorded within the same time frame for the rest of the measurement (29). A replacement by 0 for all metrics was performed if no data were collected for a specific time frame for the rest of the days. (d) Identification of waking and sleeping hours based on an automatized algorithm guided by the diaries completed by the participants (30). The inclusion criterion for a valid day was wearing the accelerometer ≥16 h·d−1. A minimum of four valid days (three weekdays and one weekend day) per week was required to be included in the analyses. The compliance wearing the accelerometer was high, with 98% of the sample wearing the accelerometers for ≥6 d.

Sedentary time and PA were classified into different intensities following Hildebrand et al. hip- and wrist-based cutoff points for the ENMO metric (31,32). Because ActiGraph’s step detection algorithm is adapted to the hip location, the main analyses of the present study were performed using hip data, although analyses for sedentary time and PA were replicated using estimates from the nondominant wrist-worn accelerometer and presented as supplemental material (see Table, Supplemental Digital Content 1, Associations between sedentary time, PA, measured with Euclidian norm minus one metric in nondominant wrist, and neurotrophic factors, https://links.lww.com/MSS/B640).

The PA variables included in this study were total minutes per day at light, moderate, vigorous, and moderate-to-vigorous PA (MVPA) for hip and wrist. With regards to steps, the volume of steps per day and the peak 60-, 30-, and 1-min cadences were computed following previously published procedures (33). We also derived time spent in the following cadence bands intensities (i.e., steps per minute): 0 steps per minute (nonmovement), 1 to 19 steps per minute (incidental movement), 20 to 39 steps per minute (sporadic movement), 40 to 59 steps per minute (purposeful movement), 60 to 79 steps per minute (slow walking), 80 to 99 steps per minute (medium walking), 100 to 119 steps per minute (brisk walking), and 120+ steps per minute (faster locomotor movements, e.g., running) (33).

Neurotrophic factors

Blood samples were obtained for biochemical and hematological screening tests between 8.30 am and 10.30 am after a minimum of 8 h overnight fasting condition at the San Cecilio University Hospital and the Virgen de las Nieves Maternity Hospital (Granada, Spain). All participants had up to 11 mL of blood drawn from the antecubital vein. The blood for plasma samples was drawn into tubes containing ethylenediaminetetraacetic acid and kept on ice for around 60 min. After collection and transportation of the samples, they were centrifuged (10 min at 4°C, 1000g), aliquoted under cold conditions by ice, and immediately stored at −80°C in the Center of Biomedical Research (Granada, Spain) until analysis.

The analysis of mature BDNF, VEGF, and IGF-1 levels in plasma was performed using the Luminex IS 100/200 system (Luminex Corporation, Austin, TX), with the XMap technology and using human monoclonal antibodies (Milliplex Map Kit, Millipore, Billerica, MA). For mature BDNF, we used the Human Neurodegenerative Disease Magnetic Bead Panel 3 (catalog HNDG3MAG-36K; EMD Millipore Corporation, Billerica, MA); for VEGF, we used the Human Angiogenesis/Growth Factor Magnetic Bead Panel (catalog HAGP1MAG-12K; EMD Millipore), and for IGF-1, we used the Human IGF-1, II Magnetic Bead Panel (catalog HIGFMAG-52K; EMD Millipore). In the Luminex IS 100/200 system, assay sensitivities or minimum detectable concentrations for BDNF, VEGF-A, and IGF-1 assays were 0.23, 8.1, and 15 ng·mL−1, respectively. Those samples not reaching the minimum detectable were excluded from the analyses. The intra-assay percent coefficient of variation for BDNF, VEGF-A, and IGF-1 was estimated to be <5.4, 3.5, and 10, respectively, and interassay at <5.3, 10, 15, respectively.

Potential confounders

After testing with correlation analyses which of the variables could be a potential confounder, sex, peak height velocity (PHV), fat mass index, wave of participation, and the parental educational level were used as potential confounders in the analyses. Peak height velocity is an indicator of maturity offset during childhood and adolescence (34). We used age and anthropometric variables (height, girls, and seated height, boys) to calculate PHV following Moore’s equations (34). The difference in years between PHV and chronological age was defined as a value of maturity offset. Fat mass index (kg·m−2) was assessed by Dual-energy X-ray absorptiometry (discovery densitometer from hologic). Wave of participation was a categorical variable according to the first moment of participation of each child in the study (wave 1, 2, or 3). Parental educational level was assessed by a self-reported questionnaire completed by parents, and we combined responses of both of them as: neither had a university degree; one had a university degree; or both had a university degree (35).

Statistical analysis

The characteristics of the study sample are presented as means and SD or percentages. Nonnormally distributed outcomes are presented as median and interquartile range (IQR). Prior to all analyses, the extreme values were winsorized to limit their influence; this was done by replacing raw scores with less than the first percentile of the cohortwide distribution with the value of the first percentile and replacing scores greater than the 99th percentile with the 99th percentile value (36). Furthermore, all outcomes were checked for normal distribution and BDNF, VEGF, and IGF-1 were normalized because they showed skewed distributions. Interaction analyses were performed between sex and sedentary time, PA and steps-related behaviors on the neurotrophic factors. No significant interactions with sex were found (P ≥ 0.10); therefore, analyses were performed for all the participants together.

Linear regression analyses were performed to examine the association of estimations from hip-worn accelerometers of sedentary time, PA, and step-related behavior with neurotrophic factors (i.e., BDNF, VEGF, and IGF-1) adjusting by potential confounders. Sedentary time and PA analyses were replicated for the nondominant wrist-placement data. We also performed linear regression analyses to examine the association between time accumulated (min·d−1) in different cadence bands of 0, 1 to 19, 20 to 39, 40 to 59, 60 to 79, 80 to 99, 100 to 119, and 120+ steps per minute and the BDNF, adjusting by potential confounders. We performed collinearity diagnosis between PA intensities and between step cadences. No multi-collinearity was observed among any of the independent variables (variance inflation factor, VIF < 10). A significance level of P < 0.05 was used. All the statistical procedures were performed using the SPSS software for Mac (version 22.0, IBM Corporation).

RESULTS

Descriptive characteristics of the sample are shown in Table 1. Times accumulated at different cadence bands are shown in Table 2. A significant association was found between light PA, moderate PA, MVPA, and peak 60-min steps cadence with BDNF (β ranging from 0.195 to 0.242, all P < 0.037) (Table 3). An association was also found between light PA and VEGF (β = 0.207, P = 0.048). No significant associations were found for the relationship of sedentary time with any of the neurotrophic factors nor for the relationship between PA, step-related behaviors, and IGF-1 (P > 0.05). When performing analyses with nondominant wrist-placement data (see Table, Supplemental Digital Content 1, Associations between sedentary time, PA, measured with Euclidian norm minus one metric in nondominant wrist, and neurotrophic factors, https://links.lww.com/MSS/B640), the associations of light PA with BDNF and VEGF disappeared (all P > 0.05). However, moderate PA and MVPA remained significantly associated with BDNF (β = 0.220, P = 0.041 and β = 0.246, P = 0.027, respectively). An association was also observed between vigorous PA and BDNF (β = 0.244, P = 0.032).

T1
TABLE 1:
Descriptive characteristics of the study sample.
T2
TABLE 2:
Time (min·d−1) accumulated at different cadence band of steps.
T3
TABLE 3:
Associations of sedentary time, physical activity and steps (measured with Euclidian norm minus one metric in hip) with neurotrophic factors (N = 97).

Figure 1 shows the relationship between time accumulated at different steps cadence bands and BDNF, adjusting for potential confounders. Among all the cadence bands, a significant association was found of the time spent in the 40 to 59 steps per minute cadence band (i.e., “Purposeful movement”) and the time spent in 60 to 79 steps per minute cadence band (i.e., “Slow walking”) with BDNF (β = 0.198, P = 0.044, and β = 0.205, P = 0.040, respectively).

F1
FIGURE 1:
Relationship between time (min·d−1) accumulated at different cadence bands and levels of BDNF. These analyses were adjusted for the following covariates: sex, PHV, fat mass index, wave of participation and parental educational level. The asterisk (*) is used to highlight significance level at P < 0.05.

DISCUSSION

The main finding of the present study was that objectively-measured PA and step-related behaviors, but not sedentary time, were positively associated with BDNF in children. Particularly, light PA, moderate PA, MVPA, and peak 60-min steps cadency were related to BDNF, being the associations of moderate PA and MVPA consistent from either hip or wrist accelerometer data. No significant associations were found between PA and steps with VEGF and IGF-1, apart from the borderline association observed between light PA and VEGF. No association was found between sedentary time and the neurotrophic factors. In addition, the time spent in purposeful movements (i.e., 40–59 steps per minute) and slow walking (i.e., 60–79 steps per minute) was associated with BDNF. Our findings suggest that different intensities and types of PA, mainly moderate and MVPA and walking at slow-medium cadences may increase plasma BDNF levels in children with overweight/obesity. However, these findings must be interpreted with caution due to the methodological limitations when measuring neurotropic factors (37), as well as to the complexity of PA analyses and the emerging variety of methods to analyze it (26).

To the best of our knowledge, this is the first study that analyzes the association between objectively measured sedentary time, PA and step-related behaviors with neurotrophic factors (i.e., BDNF, VEGF and IGF-1) in a sample of children with overweight/obesity. Only two observational studies in healthy normal-weight children have previously analyzed this relationship. In line with our results, Gabel et al. (21) did not find any association between sedentary time and plasma BDNF levels in 7- to 10-yr-old children. In contrast to our cross-sectional results, a recent 2-yr longitudinal study did not find a relationship between objectively-measured PA and serum BDNF in children age 8 to 11 yr (20). When analyzing steps, our positive results between the peak 60-min cadence and BDNF are in contrast to the negative associations found by another study in adults (38). The inconsistency and contradictory findings regarding the relationship between PA, steps, and BDNF might be due to the differences between studies with respect to the sample’s characteristics (i.e., overweight/obese vs normal weight peers), the age group analyzed (i.e., children vs adolescents or adults), the study design (i.e., cross-sectional vs longitudinal), and the methodology followed for assessing and processing PA (i.e., objective vs subjective methods) or for analyzing the neurotrophic factors levels (i.e., differences regarding kits used, prestorage treatments of blood samples—clotting/icing time, centrifugation strategy—or the way BDNF is measured in peripheral blood—plasma BDNF vs serum BDNF). With respect to the differences in BDNF measurements, much higher concentrations of BDNF have been observed in serum in comparison to plasma (8,39). On one hand, the clotting time methodology chosen can be critical for serum BDNF levels (8,40). On the other hand, plasma is obtained from blood samples drawn into tubes containing anticoagulants, preventing coagulation and thereby activation of platelets and BDNF release. Due to the smaller amount of platelet-associated BDNF in plasma, BDNF measured in plasma may, to a higher extent than serum BDNF, reflect the concentration of free BDNF. However, there is still a need to better understand how much it reflects brain levels and how it relates to PA.

Despite findings from most observational studies suggesting an inverse relationship between PA and peripheral BDNF levels (18), the positive associations found in our study are supported by previous literature focusing on the effects of physical exercise on BDNF in humans (18,41). Particularly, two studies analyzed the changes in children’s BDNF level after a lifestyle intervention which included an exercise component. Corripio et al. (42) observed that BDNF in plasma was increased in prepubertal obese children after a 2-yr lifestyle intervention which included 30 to 45 min of moderate exercise three times per week. On the contrary, another study did not find any significant change in serum BDNF in children of different weight loss after 1-yr exercise therapy (i.e., physical games) once per week (6). In adults, a recent meta-analysis showed that both acute and regular programmed exercise had a significant impact on BDNF concentrations, reflecting a moderate and small effect size (Hedges’ g = 0.46, P < 0.001; and Hedges’ g = 0.28, P = 0.005, respectively, for acute and regular exercise intervention studies) (41). Another study found that the impact on adult’s BDNF levels might be exercise intensity-dependent (19). In fact, we observed a significant association between vigorous PA and BDNF when the wrist location data was used. No information is yet available regarding which accelerometer location is more valid and reliable in children (26), which highlight the need of reporting both hip and wrist data whenever this is feasible. In our study, moderate PA and MVPA intensities were consistently associated with BDNF when using either hip or wrist PA data. This fact suggest that a moderate intensity of PA could be a higher stimulus for children with overweight/obesity to increase BDNF levels. However, further investigations are needed to clarify the effects of different PA intensities accelerometer locations on neurotrophic factors.

Another interesting finding of this study was the consistently (with both hip and wrist data) no significant associations between sedentary time, PA and steps with VEGF and IGF-1 (only a borderline association was found between light PA and VEGF). Although BDNF, VEGF, and IGF-1 are all considered neurotrophic factors and have several characteristics in common, each of them has a different functionality. Whereas BDNF is an important nerve growth factor that facilitates the growth and survival of various neurons and regulates synaptic plasticity (7), both VEGF and IGF-1 contribute to the stimulation of angiogenesis and hippocampal neurogenesis (13). Thus, the influence of PA may be different depending on the factor, which could explain the significant associations found for BDNF and the nonassociations for VEGF and IGF-1.

When analyzing which of the steps cadence bands were associated to BDNF, we observed a significant association of the time accumulated in purposeful movement (i.e., 40–59 steps per minute) and in slow walking (60–79 steps per minute) with BDNF. In this regard, walking is the most popular PA behavior, as well as the ideal PA intervention to be recommended to improve health across sedentary populations, such as the one of the present study (23,24). Additionally, the fact that our sample only accumulate an average of 7.8 min·d−1 in bands over 100 steps per minute limits the possibility to detect any significant relation between these high cadences and BDNF. To the best of our knowledge, no previous studies have analyzed the relation between time in different cadence bands and neurotrophic factors. The cadence bands appearing significantly associated to BDNF in children with overweight/obesity could be considered as bands of light PA. This, together with the fact that we also found an association between light PA and BDNF, may suggest that light activities such as walking may be enough to increase levels of the BDNF in children with overweight/obesity. In fact, children with overweight/obesity have shown a higher metabolic cost when walking at same speeds in comparison with normal-weight peers (43). This fact may indicate that children with overweight/obesity could be more sensible to neurophysiological changes at lower absolute intensities, yet the relative intensity (e.g., percent of maximal heart rate) might be similar to higher cadences conducted by leaner children. Additionally, obese children do not achieve cadences that are as high as those reached by either overweight or normal-weight children, and therefore, it may be difficult to investigate whether high cadences are associated with neurotrophic factors (33,43). Taking into account the difficulties to perform physical activities of higher intensity for this type of population, walking may be of help to increase total PA levels and health (24), and therefore have neurotrophic benefits (44).

Several explanations have been suggested to physiologically explain our associations between PA and BDNF (18). First, BDNF can pass through the blood–brain barrier in both directions (10), and it may be speculated whether peripheral BDNF circulating in blood is more efficiently uptaken or released by the brain or platelets in physically active individuals (12). However, this must be interpreted with caution because platelets cannot pass the blood–brain barrier, and at least 80% of the BDNF in plasma comes from platelets (8). Second, exercise may have beneficial effects on platelet function, being platelets are the main storage for peripheral BDNF (45). Third, aerobic exercise increases hippocampal levels of BDNF in animals (13). Animal models have also shown that BDNF can pass the blood–brain barrier from the brain to the plasma (10), and it is likely that exercise cause a production of BDNF in human brain. All these neurobiological mechanisms may explain the association of PA and steps with BDNF in the present study. However, further studies are needed to elucidate the underlying mechanisms on the association between PA and BDNF.

Caution must be applied when interpreting our findings due to several limitations. First, the cross-sectional design does not allow inferences about causality to any of the associated outcomes. Second, plasma BDNF bound to platelets cannot cross the blood–brain barrier (46) and therefore the BDNF level in the brain may be rather reflected by the amount of free BDNF in plasma (not bounded to platelets) (39). Further, normal plasma still contains a large number of platelets after centrifugation, and because BDNF is released from platelets due to activation (e.g., when a blood vessel is punctured), this fact may highly affect the level of BDNF in plasma measured in vitro (8). Third, in our study we used a statistical approach to analyze PA that has been previously used in the literature focusing on neurotrophic factors and that allows us to make direct comparisons with previous studies. However, nowadays, it is complex to choose a way to analyze PA, which is reflected in the wide variety of statistical approaches to analyze PA in the literature. Many of these ways to analyze PA should be performed when a large sample size is available, as they require all predictors (i.e., sedentary time, light PA, moderate PA and vigorous PA) coexisting in the same model, therefore decreasing the degrees of freedom and, also, the statistical power. Our relatively small sample (N = 97) discourage any attempt of applying statistical models requiring larger sample sizes to answer these questions. Thus, to find a consensus and clarify which is the best method to analyze PA, future studies using larger sample sizes should address different type of PA analysis when analyzing its association with neurotrophic factors. On the other hand, the main strength of this study was its novelty, being the first study to investigate the relationship between sedentary time, PA, and steps with neurotrophic factors in a sample of children with overweight/obesity. Additional strengths include the objective measurements of sedentary time, PA, and steps using raw accelerations in two different locations (hip and nondominant wrist) and the use of the most advanced technology to analyze neurotrophic factors (i.e., Luminex 200).

CONCLUSIONS

The results of the present study suggest that light to moderate PA intensity and step-related behaviors, but not sedentary time, are positively associated to BDNF in children with overweight/obesity. Moderate PA and MVPA seem to be consistent in the association with BDNF regardless of the accelerometer location. Particularly, the time spent in walking at slow cadences may be stimulus enough to influence the levels of BDNF in children with overweight/obesity. No associations were found between PA, sedentary time, and VEGF and IGF. Importantly, we revealed for the first time that light PA, moderate PA, MVPA, and time spent in walking at slow cadences, but not sedentary time, were associated with BDNF in children with overweight/obesity. These findings shed light on that children in an overweight/obesity status may have more room for BDNF increments induced by PA. Further, walking at slow cadences may be stimulus enough for this population to influence levels of BDNF. Result from the present study must be interpreted with caution, taking into account the limitations and variety of methods used to measure neurotrophic factors and to analyze PA. Thus, further studies using other methods must confirm or contrast our results.

The ActiveBrains project was funded by the Spanish Ministry of Economy and Competitiveness/FEDER (DEP2013-47540, DEP2016-79512-R, RYC-2011-09011, and DEP2017-91544-EXP). J. M.-G. and J. H. M. are supported by the Spanish Ministry of Education, Culture and Sport (FPU14/06837 and FPU15/02645, respectively). I. E.-C. is supported by a grant from the Alicia Koplowitz Foundation. C. C.-S. is supported by a grant from the Spanish Ministry of Economy and Competitiveness (BES-2014-068829). P. M.-G. is supported by a grant from European Union’s Horizon 2020 research and innovation program (no 667302). Additional support was obtained from the University of Granada, Plan Propio de Investigación 2016, Excellence actions: Units of Excellence, Unit of Excellence on Exercise and Health (UCEES); by the Junta de Andalucía, Conserjería de Conocimiento, Investigación y Universidades and European Regional Development Fund (ERDF) (ref. SOMM17/6107/UGR); and by the Alicia Koplowitz Foundation). In addition, funding was provided by the SAMID III network, RETICS, funded by the PN I + D + I 2017–2021 (Spain), ISCIII—Sub-Directorate General for Research Assessment and Promotion, the European Regional Development Fund (ERDF) (ref. RD16/0022) and the EXERNET Research Network on Exercise and Health in Special Populations (DEP2005–00046/ACTI). The authors would like to thank all the families participating in the ActiveBrains project. The authors also acknowledge everyone who helped with the data collection and all of the members involved in the field-work for their effort, enthusiasm, and support. This work is part of a Ph.D. thesis conducted in the Biomedicine Doctoral Studies of the University of Granada, Spain.

The results of the present study do not constitute endorsement by ACSM. The authors declare that the results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.

REFERENCES

1. Ou X, Andres A, Pivik RT, Cleves MA, Badger TM. Brain gray and white matter differences in healthy normal weight and obese children. J Magn Reson Imaging. 2015;42(5):1205–13.
2. Kamijo K, Pontifex MB, Khan NA, et al. The negative association of childhood obesity to cognitive control of action monitoring. Cereb Cortex. 2014;24(3):654–62.
3. Lommatzsch M, Zingler D, Schuhbaeck K, et al. The impact of age, weight and gender on BDNF levels in human platelets and plasma. Neurobiol Aging. 2005;26(1):115–23.
4. Sandrini L, Di Minno A, Amadio P, Ieraci A, Tremoli E, Barbieri SS. Association between obesity and circulating brain-derived neurotrophic factor (BDNF) levels: Systematic review of literature and meta-analysis. Int J Mol Sci. 2018;19(8).
5. El-Gharbawy AH, Adler-Wailes DC, Mirch MC, et al. Serum brain-derived neurotrophic factor concentrations in lean and overweight children and adolescents. J Clin Endocrinol Metab. 2006;91(9):3548–52.
6. Roth CL, Elfers C, Gebhardt U, Müller HL, Reinehr T. Brain-derived neurotrophic factor and its relation to leptin in obese children before and after weight loss. Metabolism. 2013;62(2):226–34.
7. Binder DK, Scharfman HE. Brain-derived neurotropic factor. Growth Factors. 2004;22(3):123–31.
8. Fujimura H, Altar CA, Chen R, et al. Brain-derived neurotrophic factor is stored in human platelets and released by agonist stimulation. Thromb Haemost. 2002;87(4):728–34.
9. Yamamoto H, Gurney ME. Human platelets contain brain-derived neurotrophic factor. J Neurosci. 1990;10(11):3469–78.
10. Pan W, Banks WA, Fasold MB, Bluth J, Kastin AJ. Transport of brain-derived neurotrophic factor across the blood–brain barrier. Neuropharmacology. 1998;37(12):1553–61.
11. Klein AB, Williamson R, Santini MA, et al. Blood BDNF concentrations reflect brain-tissue BDNF levels across species. Int J Neuropsychopharmacol. 2011;14(3):347–53.
12. Rasmussen P, Brassard P, Adser H, et al. Evidence for a release of brain-derived neurotrophic factor from the brain during exercise. Exp Physiol. 2009;94(10):1062–9.
13. Cotman CW, Berchtold NC, Christie L-A. Exercise builds brain health: key roles of growth factor cascades and inflammation. Trends Neurosci. 2007;30(9):464–72.
14. Donnelly JE, Hillman CH, Castelli D, et al. Physical activity, fitness, cognitive function, and academic achievement in children: a systematic review. Med Sci Sports Exerc. 2016;48(6):1197–222.
15. Herting MM, Keenan MF, Nagel BJ. Aerobic fitness linked to cortical brain development in adolescent males: preliminary findings suggest a possible role of BDNF genotype. Front Hum Neurosci. 2016;10:327.
16. Jeon YK, Ha CH. The effect of exercise intensity on brain derived neurotrophic factor and memory in adolescents. Environ Health Prev Med. 2017;22(1):27.
17. Huang T, Gejl AK, Tarp J, Andersen LB, Peijs L, Bugge A. Cross-sectional associations of objectively-measured physical activity with brain-derived neurotrophic factor in adolescents. Physiol Behav. 2016;171:87–91.
18. Huang T, Larsen KT, Ried-Larsen M, Møller NC, Andersen LB. The effects of physical activity and exercise on brain-derived neurotrophic factor in healthy humans: a review. Scand J Med Sci Sports. 2014;24(1):1–10.
19. Ferris LT, Williams JS, Shen C-L. The effect of acute exercise on serum brain-derived neurotrophic factor levels and cognitive function. Med Sci Sports Exerc. 2007;39(4):728–34.
20. Arvidsson D, Johannesson E, Andersen LB, et al. A longitudinal analysis of the relationships of physical activity and body fat with nerve growth factor and brain-derived neural factor in children. J Phys Act Health. 2018;1–6.
21. Gabel L, Ridgers ND, Della Gatta PA, et al. Associations of sedentary time patterns and TV viewing time with inflammatory and endothelial function biomarkers in children. Pediatr Obes. 2016;11(3):194–201.
22. Spaeth AM, Kanoski SE, Hayes MR, Grill HJ. TrkB receptor signaling in the nucleus tractus solitarius mediates the food intake-suppressive effects of hindbrain BDNF and leptin. Am J Physiol Endocrinol Metab. 2012;302(10):E1252–60.
23. Ding D, Nguyen B, Learnihan V, et al. Moving to an active lifestyle? A systematic review of the effects of residential relocation on walking, physical activity and travel behaviour. Br J Sports Med. 2018;52(12):789–99.
24. Morris JN, Hardman AE. Walking to health. Sports Med. 1997;23(5):306–32.
25. Cadenas-Sánchez C, Mora-González J, Migueles JH, et al. An exercise-based randomized controlled trial on brain, cognition, physical health and mental health in overweight/obese children (ActiveBrains project): Rationale, design and methods. Contemp Clin Trials. 2016;47:315–24.
26. Migueles JH, Cadenas-Sanchez C, Ekelund U, et al. Accelerometer data collection and processing criteria to assess physical activity and other outcomes: a systematic review and practical considerations. Sports Med. 2017;47(9):1821–45.
27. van Hees VT, Gorzelniak L, Dean León EC, et al. Separating movement and gravity components in an acceleration signal and implications for the assessment of human daily physical activity. PLoS One. 2013;8(4):e61691.
28. van Hees VT, Fang Z, Langford J, et al. Autocalibration of accelerometer data for free-living physical activity assessment using local gravity and temperature: an evaluation on four continents. J Appl Physiol. 2014;117(7):738–44.
29. van Hees VT, Renström F, Wright A, et al. Estimation of daily energy expenditure in pregnant and non-pregnant women using a wrist-worn tri-axial accelerometer. PLoS One. 2011;6(7):e22922.
30. van Hees VT, Sabia S, Anderson KN, et al. A novel, open access method to assess sleep duration using a wrist-worn accelerometer. PLoS One. 2015;10(11):e0142533.
31. Hildebrand M, Hansen BH, van Hees VT, Ekelund U. Evaluation of raw acceleration sedentary thresholds in children and adults. Scand J Med Sci Sports. 2016;27(12):1814–23.
32. Hildebrand M, Van Hees VT, Hansen BH, Ekelund U. Age group comparability of raw accelerometer output from wrist-and hip-worn monitors. Med Sci Sports Exerc. 2014;46(9):1816–24.
33. Barreira TV, Katzmarzyk PT, Johnson WD, Tudor-Locke C. Cadence patterns and peak cadence in US children and adolescents: NHANES, 2005–2006. Med Sci Sports Exerc. 2012;44(9):1721–7.
34. Moore SA, Mckay HA, Macdonald H, et al. Enhancing a somatic maturity prediction model. Med Sci Sports Exerc. 2015;47(8):1755–64.
35. Huppertz C, Bartels M, de Geus EJC, et al. The effects of parental education on exercise behavior in childhood and youth: a study in Dutch and Finnish twins. Scand J Med Sci Sports. 2017;27(10):1143–56.
36. Sink KM, Espeland MA, Castro CM, et al. Effect of a 24-month physical activity intervention vs health education on cognitive outcomes in sedentary older adults. JAMA. 2015;314(8):781.
37. Pareja-Galeano H, Alis R, Sanchis-Gomar F, et al. Methodological considerations to determine the effect of exercise on brain-derived neurotrophic factor levels. Clin Biochem. 2015;48(3):162–6.
38. Nofuji Y, Suwa M, Moriyama Y, et al. Decreased serum brain-derived neurotrophic factor in trained men. Neurosci Lett. 2008;437(1):29–32.
39. Radka SF, Holst PA, Fritsche M, Altar CA. Presence of brain-derived neurotrophic factor in brain and human and rat but not mouse serum detected by a sensitive and specific immunoassay. Brain Res. 1996;709(1):122–301.
40. Tsuchimine S, Sugawara N, Ishioka M, Yasui-Furukori N. Preanalysis storage conditions influence the measurement of brain-derived neurotrophic factor levels in peripheral blood. Neuropsychobiology. 2014;69(2):83–8.
41. Szuhany KL, Bugatti M, Otto MW. A meta-analytic review of the effects of exercise on brain-derived neurotrophic factor. J Psychiatr Res. 2015;60:56–64.
42. Corripio R, Gónzalez-Clemente J-M, Jacobo P-S, et al. Plasma brain-derived neurotrophic factor in prepubertal obese children: results from a 2-year lifestyle intervention programme. Clin Endocrinol (Oxf). 2012;77(5):715–20.
43. Huang L, Chen P, Zhuang J, Walt S. Metabolic cost, mechanical work, and efficiency during normal walking in obese and normal-weight children. Res Q Exerc Sport. 2013;84(2 Suppl):S72–9.
44. Shultz SP, Anner J, Hills AP. Paediatric obesity, physical activity and the musculoskeletal system. Obes Rev. 2009;10(5):576–82.
45. Tang SW, Chu E, Hui T, Helmeste D, Law C. Influence of exercise on serum brain-derived neurotrophic factor concentrations in healthy human subjects. Neurosci Lett. 2008;431(1):62–5.
46. de Vries HE, Kuiper J, de Boer AG, Van Berkel TJ, Breimer DD. The blood–brain barrier in neuroinflammatory diseases. Pharmacol Rev. 1997;49(2):143–55.
Keywords:

ACCELEROMETRY; BDNF; COGNITION; OBESITY; WALKING; YOUTH

Supplemental Digital Content

Copyright © 2019 by the American College of Sports Medicine