Physical activity prevents numerous disorders and improves many pathophysiological disease features. Exercise has been classically associated to muscular and metabolic benefits. However, the effects of exercise training on brain function, specialty known as “Neurobiology of exercise”, have recently received much attention.
The beneficial effects of exercise have been clearly established in several pathologies such as Alzheimer´s disease, Parkinson´s disease, amyotrophic lateral sclerosis, schizophrenia, bipolar disorder and depressive disorder, among others. These findings have lead to use exercise training as a therapeutic coadjutant strategy not only in research but also in the clinical practice. According to a large amount of evidence, physical exercise not only restores the altered physiology of several neuropsychiatric diseases but also improves the brain function, cognition and psychological condition in healthy people.
Exercise exerts its action on the brain through many molecular pathways and physiological mechanisms. Some of them include the prevention of oxidative damage, release of endorphins, restoration of dopamine signaling and those actions mediated by the neurotrophic factors. Neurotrophic factors are growth factors with different sources and pathways of action. Although there are several factors included in this family of proteins, we would like to highlight the insulin-like growth factor 1 (IGF-1), the vascular endothelial growth factor (VEGF), the brain-derived neurotrophic factor (BDNF), the nerve growth factor (NGF), and the neurotrophins 3 and 4/5 (NT-3 and NT-4/5 respectively). The main actions mediated by these factors include the hippocampal neurogenesis, neuron repair, axogenesis, dendrogenesis, synaptic transmission modulation, synaptogenesis, and thereby brain plasticity. The functional consequences of their modulation include long term potentiation, improvement of learning and memory, anxiolytic and antidepressant effects.
BDNF is one of the most important neurotrophic factors since it mediates pleiotropic trophic effects in the brain. In addition, it is also one of the main neurotrophic factors induced by both chronic and acute exercise. High levels of BDNF have been found in brains of exercised persons, through post mortem studies and jugular blood in vivo analysis, as well as in trained animals. The brain modulation is produced mainly in the hippocampus and limbic system and thereby it has been associated with several cognitive and psychological improvements.
There is yet no consensus about the adequate blood processing conditions to standardize peripheral BDNF assessment in exercise studies. Serum, plasma, whole blood, and platelets-rich plasma with several methodological processing conditions have been indistinctly used in the literature. This leads to inconsistencies in the studies.
This Doctoral Thesis aims to clarify the effects of acute exercise and training in human blood levels of neurotrophic factors as well as the appropriate methodological protocol for the BDNF analysis. We also aim to determine the molecular pathways involved in the beneficial effects of exercise training in two mice models of Alzheimer´s disease. Finally, we aim to test the possible synergistic beneficial effect of exercise training and a BDNF pharmacologic mimetic on brain function in rats.
In our first experimental model, healthy adolescents were divided into two groups according to their exercise habits. The trained group included members of an elite cyclist team, and thereby highly trained. The control group included sedentary matched controls. We compared the IGF-1 and BDNF blood levels of both groups in the pre-season and post-competition period. We evaluated the possible effect of the circulating BDNF on the cAMP response element-binding (CREB) activation in peripheral blood mononuclear cells. All participants were also evaluated through anthropometric, hematological and acelerometric analysis.
We found that the BDNF and IGF-1 blood levels were increased in the trained adolescents compared with the sedentary controls when we analyzed it during the pre-season, characterized by a moderate physical demand. This increment did not affect to the CREB. Moreover, the differences between both groups disappeared when we compared the neurotrophic factor levels in the post-competition period, characterized by a maximum physical performance requirement.
In our second experimental model we determined the effect of an acute bout of exercise on BDNF blood levels in healthy adults under different blood processing conditions in a time-course analysis (at baseline, immediately after exercise, at 30 and 60 minutes of recovery). The blood samples that we analyzed included serum coagulated 10 minutes and 24 hours; plasma with EDTA, with and without platelets; and whole blood.
We found an increment in BDNF levels after the acute exercise in the serum coagulated during 24 hours and in whole blood samples. These changes were not evident when analyzed in the serum coagulated during 10 minutes, total plasma and platelet-free plasma samples. The interference of the anticoagulants used for the plasma and the irregular platelet activation in the serum coagulated during 10 minutes led to a high variability in the BDNF levels. We have also found that the processing temperature of the samples and the hemoconcentration are relevant factors to take into account in these studies.
In our third experimental model we used non transgenic and double transgenic mice (2xTg) for Alzheimer´s disease. We divided the animals into two groups: sedentary and exercised. At 10 months of age, trained groups were subjected to 12 weeks of a training combination of forced and voluntary exercise. Different psychological and physical tests were performed to the animals. We also evaluated brain glucose uptake by positron emission tomography and biochemical markers related to amyloid-β (Aβ) (1-42) levels and its modulation through the low density lipoprotein receptor-related protein 1 (LRP1), BDNF pathway (tyrosine kinase type B (TrkB) levels and CREB levels and activation), oxidative damage (malondialdehyde, glutathione, protein carbonylation), antioxidant defense levels (superoxide dismutase dependent of copper/zinc and manganese (Cu/Zn-SOD and Mn-SOD), glutathione peroxidase (GPx) and catalase (CAT)), and mitochondrial content (cytochrome-C) and biogenesis (peroxisome proliferator-activated receptor-gamma coactivator 1α, PGC-1α).
The exercise training improved the behavior of the 2xTg mice as well as their physical performance. These improvements were accompanied by a hippocampal Aβ (1-42) reduction in the 2xTg mice. The cerebral and systemic oxidative damage, LRP1 and hippocampal BDNF levels were reduced in the 2xTg mice and the exercise did not affect it. Nevertheless, the brain glucose uptake was higher in the transgenic mice and the antioxidant defense, determined by the CAT, increased in the 2xTg mice after exercise.
Our forth experimental model included a triple transgenic mouse model (3xTg) of Alzheimer´s disease and non transgenic mice as control. We studied the protective effect of exercise in ovariectomized mice. The exercise protocol included 12 weeks of spontaneous wheel-running. The animals were divided into eight experimental groups which included an Alzheimer´s disease model and/or artificial climacteric and/or exercise treatment. We sacrificed animals and analyzed several brain biomarkers which included the Aβ and hiperphosphorylated tau levels, the amyloidogenic pathway (C99/APP), BDNF levels and its pathway through TrkB and CREB, PGC-1α, and the expression the antioxidant enzymes GPx, Mn-SOD and CAT.
The exercise training performed by the 3xTg mice and the ovariectomy did not affect to the brain Aβ and the tau hyperphosphorylated levels. Nevertheless, the exercise partially prevented activation of the amyloidogenic pathway in all cases. In addition, training incremented the hippocampal BDNF levels of the 3xTg, ovariectomized and sham. We obtained a CREB activation increment in the non transgenic and 3xTg non ovariectomized mice subjected to exercise. Moreover, the hippocampal expression of CAT increased in the 3xTg mice, trained and sedentary, whereas the ovariectomy reduced it. This was reverted through physical exercise.
Finally, in the fifth experimental model we aimed to evaluate the possible synergic effects of 6 weeks of exercise training and a pharmacological BDNF mimetic, 7,8-dihydroxyflavone (7,8-DHF), in healthy young rats. Thus, we used four experimental groups which included the trained group, the group with 7,8-DHF, the group subjected to a combination of exercise and 7,8-DHF and a control group. We evaluated the cognition of the animals thought the object recognition test and their behavioral condition by the open field test.
The treatment with 7,8-DHF and/or forced exercise training did not show a synergic effect on the psychological parameters analyzed. The learning and memory did not improve, whereas the exploratory behavior incremented with all treatments but especially with the exercise training.
Taking together our findings we can conclude that acute and chronic exercise increases BDNF peripheral levels in humans whereas exercise training improves the pathophysiological and functional features of Alzheimer´s disease in two different transgenic models. Exercise seems to contribute to brain benefits by a wide range of physiological mechanisms including the BDNF pathway.