**2. The potential role of exercise-induced neurotrophic factors on mental health**

It is well reported that the potential role of exercise in providing or maintaining beneficial effects on mental health and inhibiting the disorder's progression is apparent. However, studies show that differences in the therapeutic benefit of exercise appear to be due to variations between populations and exercise modalities. The present literature suggests exercise-induced mental health with three mechanistic hypotheses; a) physical/hedonic effects, (b) neurobiological mechanisms, or (c) cultivating behavioral mechanisms of change. It seems that exercise simultaneously influences mental health via neurobiological and behavioral mechanisms [1]. However, this chapter focuses only on neurobiological mechanisms. Because the underlying mechanisms responsible for these beneficial effects are poorly defined, although some myokines and metabolites released by the skeletal muscle are well known to affect brain health positively, the underlying mechanisms require further investigation. This section will examine how different exercise protocols affect these mechanisms and the relationship of this effect with mental health.

#### **2.1 Brain-derived neurotrophic factor (BDNF)**

BDNF stands out due to its active role in the neurogenesis and neuroplasticity of neurons, which has a potential role in the pathophysiology of many neuropsychiatric disorders in various brain regions [26]. The cumulative evidence is diverse, including major depression [27, 28] bipolar disorder [29–31], and schizophrenia [32–34] shows that BDNF levels are significantly reduced in neuropsychiatric disorders. In addition, it has been shown that a decrease in hippocampal BDNF levels is associated with stress-induced depressive behaviors, and antidepressant treatment increases BDNF expression [35].

In addition, it is thought that BDNF may have different roles in brain stress and reward systems [36, 37]. Previous research showed that stress could change BDNF expression in many critical brain regions [38]. Underpin of this idea comes that BDNF and stressors can affect both systems: the stress-related system, which mainly contains the hypothalamus-pituitary–adrenocortical (HPA) axis and the hippocampus, and the reward-associated system, which mainly contains the ventral tegmental area–nucleus accumbens pathway [39]. BDNF may exert notable influences on these two systems. BDNF is also closely associated with insomnia, which is known to cause stress-related mental disorders [40]. Furthermore, limited studies have shown that direct hippocampal infusions of BDNF can produce antidepressant effects in rodents [41, 42]. Few human studies show that depressed patients have much lower BDNF levels in postmortem brain tissue than healthy human subjects [43]. It is hypothesized that low BDNF levels in the brain may cause atrophy and increased cell loss in the hippocampus and prefrontal cortex, as observed in depressed subjects [35, 44]. Similarly, most studies show decreased BDNF levels in schizophrenia [45]. Exercise benefits panic disorder (type of anxiety disorder) by increasing serum BDNF concentration [46].

It is well accepted that BDNF is the most sensitive neurotrophin to the effects of exercise [47]. Kurdi and Flora [48] investigated the role of aerobic exercise (AE) in affecting BDNF levels in the elderly with depression. Thirty-five older women (age ≥ 50 y) with depression and 35 healthy coeval women underwent treadmill running/walking with a speed of 6 km/h for 15 minutes once a day for 28 days. Before the intervention, the depression group's BDNF levels were far lower than in the non-depression group. Although AE increased BDNF production in both groups, the increasing percentage of BDNF levels was higher in the depression group. But important to note that subjects with depression were first treated with an antidepressant of selective serotonin reuptake inhibitors for 3 months before the intervention. A single-blind, randomized clinical trial design study compared standard psychiatric treatment and a 12-week AE program (60 min/3d/w, 10-min warm-up, followed by 45-min AE, and ended with a 5-min cool-down, 60−75% of HRmax) utilizing exergame (whole-body exercise software) and traditional AE equipment (a stationary bike, and an elliptical machine) in 33 individuals with schizophrenia [49]. The results indicate that AE is effective in enhancing neurocognitive functioning by BDNF upregulation. Another found that a single exercise session (an incremental exercise test on a motorized treadmill) leads to a significant up-regulation and transient normalization of BDNF serum levels in elderly women (n = 35, age: 61.1 ± 7.2 y) with a remitted depressive episode of unipolar depression [50]. Kerling et al. [51] investigated the effect of 6 weeks AE (total 22 exercises, each exercise 45-min at moderate intensity, bicycle ergometer, treadmill, cross-trainer, and rowing) intervention on serum BDNF levels in guideline-based treated patients with MDD (n = 42, age: 44.2 ± 8.5 y). Authors indicated that AE given as an adjunct to standard guideline-based treatment has additional effects on serum BDNF levels in people with MDD. Moreover, Yeh et al. [52] observed that significant impact on the enhancement of BDNF levels and improvement of depression symptoms after 12 weeks AE (50-min/3d/w), with music in community women (n = 41, age: 53.2 ± 10.3 y).

A study by Hartmann et al. [53] revealed that 6 weeks of moderate-intensity AE (stationary cycling at 65% HRR for 35 min progressing to 70% for 40 min) reduces the severity of depression, anxiety, and psychological distress in people (age = 40.5 ± 9.84 y, n = 13) with mental health disorder compared to healthy counterparts. However, no difference was found between BDNF levels in the comparison between groups, and it was reported that there was no relationship between the effects of BDNF levels on disease symptoms. A randomized controlled study by Pereira et al. [54] investigated muscle strength exercises (hip flexion, abduction, adduction, and extension; knee flexion and extension; and mini-squat, 75% of 1RM) and AE (65%−80% of the HRmax, walking and calisthenics exercises) on the plasma levels of BDNF and depressive symptoms in 451 elderly women (community-dwelling older women, age: 65–89 y). Both protocols lasted 10 weeks, and 30 sessions (1-h sessions) in total were performed three times a week. Based on the findings, resistance exercise (RE) and AE improved depressive symptoms. Only SE significantly increased the plasma levels of BDNF. However, the authors indicated that BDNF did not mediate the positive effects of exercise on depressive symptoms in the elderly.

According to Birinci et al. [55], exergames, also known as active video games, are a new-generation exercise that can combine physical exercise and cognitive exercise at the same time. A randomized clinical trial by Anderson-Hanley et al. [56] investigated the effect of the cognitive benefit of exergame intervention and cognitive exercise in older adults (n = 14, age: 78.1 ± 9.9 y) with or at risk for mild cognitive impairment (MCI). Participants performed one of three conditions for 6 months:

#### *The Potential Role of Exercise-Induced Neurotrophic Factors for Mental Health DOI: http://dx.doi.org/10.5772/intechopen.106867*

exer-tour (low cognitive load, virtual scenic bike tour) or exer-score (high cognitive demand, videogame+biking or video game-only, no physical exercise). Exercises were conducted for at least 20 minutes at least twice a week and gradually increased exercise duration to 45 minutes and frequency to at least three to five times per week. Both exercise interventions had similarly positive effects on executive functions (Stroop and word memory test scores) after 6 months. In addition, increased BDNF level due to exercise was associated with increased gray matter volume in the prefrontal cortex and right anterior cingulate cortex. Byun and Kang [57] conducted aerobic and cognitive exercise separately in one session (50 min/4d/w, moderate-high intensity, cognitive exercise + AE, respectively) for 12 weeks in 24 women (age: 65–79 y). The cognitive exercise consisted of situation assessment, responding to language instruction, communication, counting with hand, drawing with hands, and following hand signals. The physical exercise included walking, stepping, knee up, leg up, hopping, jumping jack, squatting, lifting the heels, turning the leg, raising the arm, bending the wrist, and balancing exercises. Findings showed that the exercise program significantly increased the Mini-mental state examination scores and BDNF levels. Kim et al. [58] investigated the effect of a combined exercise [3d/w, RE: chest press, seated row, squat, shoulder press, biceps curl, triceps extension, calf raise, and reverse crunch with elastic band for 25 min, 12–13 RPE, and AE: moderate walking at 50−70% of heart rate reserve (HRR) for 25 min] program on circulating BDNF expression in 24 patients with schizophrenia for 12 weeks. Increased serum BDNF levels were observed following the combined exercise program. The authors pointed out that exercise-induced BDNF levels may provide an alternative treatment for the chronic schizophrenic population.

On the contrary, a randomized controlled trial by McGurk et al. [59] investigated the effect of adding a 30-h AE program over 10 weeks to an equally intensive cognitive remediation program on cognitive function. The study results showed a combination of physical exercise (stationary cycling, walking, or jogging 60–75% of HRmax, 3d/w/40 min) and cognitive exercise (computer-based games, 60 min/3d/w), and cognitive exercise alone improved cognitive functions. In addition, although there was no difference in BDNF levels between the groups, it was reported that moderateintensity AE did not affect BDNF levels in people with mental disorders.

Some studies have also evaluated the effectiveness of a combination of exercise and behavioral therapy. Gourgouvelis et al. [60] recruited 16 individuals (4 M/12 F; age: 39.31 ± 1.25 y) with MDD or anxiety and 22 healthy individuals (11 M/11 F; age: 20.95 ± 1.25 y) in their study. The authors investigated the effects of exercise on treatment outcomes in addition to antidepressant medication and behavioral therapy for 8 weeks. Exercise with behavioral therapy was conducted for 8 weeks [≥150 min/w AE at 60–80% HRmax and 2d/w RE (8–12 repetitions at 95% of 10 repeated maximum, 2–3 superset large muscle groups]. After the intervention, 75% of patients showed either a therapeutic response or a complete remission of symptoms, resulting in a greater reduction in depression symptoms compared to 25% of non-exercise. In a study by Szuhany and Otto [61], participants with MDD or dysthymia (n = 29, 22 F/7 M, age: 18–65 y) performed 16 weeks of behavioral therapy (experts performed individual interview, 60 min/3d/w) or behavioral therapy+exercise (150 min/w moderate-intensity AE or mild-intensity yoga) or behavioral therapy+stretching (150 min/w) intervention. BDNF is increased in the depressive sedentary population immediately after acute exercise. Although acute exercise has increased BDNF levels in depressed individuals, no correlation was found between BDNF changes and changes in depression symptoms. Ikai et al. [62] investigated the effects of Hatha

yoga therapy on resilience and BDNF levels activity in patients with schizophreniaspectrum disorders. While the yoga group performed weekly 1-hour Hatha yoga sessions [gentle yoga stretches and simple movements in coordination with breathing, movements of major muscle groups, asana (twisting poses and standing poses), deep relaxation, and breathing exercises] in addition to regular treatment, the control group recruited a daycare rehabilitation program (provided social skills and walking). No significant differences were found in changes in BDNF levels from baseline to week eight between the two groups. The authors also indicated that yoga therapy showed no beneficial effects on the resilience level of stress markers.

According to Szuhany and Otto [61], effect sizes indicate that studies with approximately 85 participants may be needed to obtain significant associations between BDNF response and depression outcome in cases where exercise is not the only treatment method. Furthermore, as previous studies have shown that BDNF responds differently to the gender variant, obtaining comparable samples of men and women may be essential to assess its effects [24, 63]. Studies generally evaluate 150 minutes of moderate-intensity exercise per week. However, the current literature reports that short-term high-intensity interval training (HIIT) would increase BDNF levels significantly [64, 65]. Therefore, future studies can add the HIIT method to exercise programs to get a clearer understanding of the exercise effect. In addition, it should consider that some drugs used by participants under treatment may create confusion when revealing the effect of exercise alone. Moreover, results from sample groups with wide age ranges may influence the inferences because it is well known that BDNF levels are affected by age [66].

#### **2.2 Irisin**

Inducing peroxisome proliferator-activated receptor-gamma coactivator-1 alpha (PGC-1α) by exercise leads to the production of the muscle protein called fibronectin type III domain containing 5 (FNDC5). Then FNDC5 cleaved and generates a polypeptide hormone named irisin. Irisin is a recently discovered exercise-induced myokine that can control numerous cellular signaling (white fat-browning, energy expenditure increase, anti-inflammatory effects, and mitochondrial function improvement) in many organs [67]. Irisin is expressed in skeletal muscles [68], adipose tissue, pancreas, cardiac muscle [69], and several regions of the brain [70, 71] and is also observed in Purkinje cells, neurons in the cerebellum [71], and cerebrospinal fluid [72]. Moreover, a large number of investigations reported that RE (eight exercises of 12 repetitions with 3–4 sets at 65% of 1RM) [73], prolonged moderate AE (a 90-min treadmill exercise protocol at 60% of VO2max) [74], and strenuous exercise (a 30-min cycling ergometry) [75] induce the peripheral irisin levels. Jedrychowski et al. demonstrated that irisin circulates at ∼3.6 ng/ml in sedentary individuals, while it can significantly increase in individuals undergoing AE up to ∼4.3 ng/ml [76].

Irisin may have an impact on the expression of many neuronal genes that modulate neuronal plasticity [77, 78] and that can potentially fight against neural deterioration [79] and are secreted strongly by exercise [68, 77]. Some studies also suggest that irisin is likely to have a beneficial effect hippocampus and dentate gyrus, which are memory-related brain regions [80, 81]. A review study by Farshbaf and Alvina, [67] indicated that irisin has multiple endocrine actions central nervous system and this molecular mechanism can show neuroprotection against different injuries and insults, including neurodegenerative disorders. Authors also pointed out that irisin release can modulate BDNF expression in the hippocampus leading to protection

#### *The Potential Role of Exercise-Induced Neurotrophic Factors for Mental Health DOI: http://dx.doi.org/10.5772/intechopen.106867*

against ischemia, acute stress, and neurodegenerative disorders. They suggested that pharmacologically FNDC5/Irisin could be an alternative as mimetic of exercise, in cases for example that exercise is not recommended or not possible. In a pioneering study, Lourenco et al. [79] found that the expression of FNDC5 in the hippocampus and cerebrospinal fluid is decreased in both human and mouse models with AD. Similarly, decreased PGC-1α expression was observed in human Parkinson's disease (PD) brain samples [82]. Yu et al. [83] showed that cerebral ischemia can reduce FNDC5 expression in skeletal muscle and circulating irisin levels in the peripheral blood.

Recently, irisin is thought to be a therapeutic hormone that can positively affect depressive behaviors through its effect on neuronal activities in the prefrontal cortex [84, 85]. Moreover, some studies suggest that exercise-induced irisin ameliorates depressive symptoms by activating the PGC-1α-FNDC5/irisin pathway in the hippocampus [21, 86]. Irisin upregulates BDNF expression in many brain areas (VTA, hippocampus, etc.), involved in learning, mood, attention, motivation, and reward system. Irisin levels lead to improve mitochondria function through PGC-1α signaling and elevate BDNF enhancing synaptic plasticity in the brain. Finally, this process may lead to the improvement of depressive neuropathology (**Figure 1**). Similarly, Siteneski et al. [87] indicated that central administration of irisin and BDNF elicited antidepressant-like behaviors in mice. Studies also observed decreased serum irisin levels in people with post-stroke depression [88] and anxiety disorder [89].

Papp et al. [90] revealed that irisin might link mood deterioration to the central effects of BDNF exerted in areas closely associated with reward-related processes involved in the evolution of depression in patients with chronic obstructive pulmonary disease (n = 74, age: 62.15 ± 9.70 y). Similarly, Zsuga et al. [91] indicated that exercise-induced irisin might have a role in motivation and reward-related processes. Wrann et al. [77] revealed that the proliferation of FNDC5 in the primary cortical neurons may have increased expression of BDNF. A study investigating the relationship between energy homeostasis regulation and coronary heart disease (CHD) patients comorbid with depression by Han et al. [92] indicated that interaction between irisin and BDNF could trigger the imbalance of energy homeostasis in the depression of CHD patients. They also demonstrated that irisin and BDNF serum levels were significantly lower in CHD patients with depression compared with CHD patients without depression. Similarly, a review study by Jo and Song [93] suggests that irisin is an essential molecule that may suppress several neuropathological mechanisms involved in depression. On the contrary, Hofmann et al. [94] investigated the effectiveness of irisin release with depressiveness, perceived stress, and anxiety symptoms in obese women (n = 98, age: 43.9 ± 12.5 y). According to their study results, irisin was not associated with depressiveness, anxiety, and perceived stress in female obese patients.

#### **2.3 Insulin-like growth factor-1 (IGF-1)**

IGF-1 can modulate many mechanisms that help to occur neuroplasticity in the human brain, such as synaptic processes (e.g., long-term potentiation) [95, 96], angiogenesis, axon outgrowth, dendritic maturation, and synaptogenesis [97, 98]. Moreover, there is a putative idea that IGF-1 plays a role in structural gray matter changes. Supporting neuronal survival via inducing proliferation, reducing apoptosis, and protecting neurons against toxicity may underpin this idea [99]. According to Calvo et al. [100] higher levels of serum IGF-1 are associated with better cognitive

#### **Figure 1.**

*The effect of exercise on the irisin pathway and related depressive neuropathology changes. Exercise-induced irisin is mainly derived from skeletal muscle and readily crosses the blood–brain barrier. Irisin upregulates BDNF expression in many brain areas involved in learning, mood, attention, motivation, and reward system. Irisin levels lead to improve mitochondria function through PGC-1α signaling and elevate BDNF enhancing synaptic plasticity in the brain. Finally, this process may lead to the improvement of depressive neuropathology. FNDC5, fibronectin domain-containing protein 5; PGC-1α, proliferator-activated receptor-gamma coactivator-1 alpha; UCP1, uncoupling protein 1; BDNF, brain-derived neurotrophic factor; TrkB, tropomyosin receptor kinase B.*

performance in persons (n = 31, age: 83.71 ± 3.59 y, 58% women) with MCI, particularly on tests of learning and memory. A cross-sectional study by Duron et al. [101] showed a significant association between low IGF-I serum levels and neurodegenerative process of AD in men, but not in women (218 men, 476 women; 78.6 ± 6.7 y). Parallel to this, evidence suggests that IGF-1 may have a role in beta-amyloid clearance and reducing hyperphosphorylation tau in AD [102].

A recent review revealed that IGF-1 serum levels might be affected by the type of exercise. However, there is no consensus on the effect of exercise conditions yet [103]. A study by Cassilhas et al. [104] showed that 24 weeks of intensive RE (6 exercises: chest press, leg press, vertical traction, abdominal crunch, leg curl, and lower back, 80% of the 1RM, 2 sets of 8 repetitions each with rests for 1-min 30-sec between them and 3-min between exercises) intervention improved mood, ameliorated anxiety, and increased IGF-1 serum concentration in older men (n = 20) ages 65 to 75 years. Yoon et al. [105] recruited 30 older women into the RE and interval training group (n = 10, age: 64.10 ± 3.35 y), the RE and AE group (n = 10, age: 65.20 ± 5.10 y), and the control group (n = 10, age: 63.20 ± 2.62 y). Authors aimed to determine the effects

#### *The Potential Role of Exercise-Induced Neurotrophic Factors for Mental Health DOI: http://dx.doi.org/10.5772/intechopen.106867*

of RE (chest press, lateral pull-down, arm curl, back extension, crunch, leg press, leg extension, leg curl, and heel raise, 8–15 repeat at the intensity of 1-RM, 60–80%, in 3 sets) + AE (stationary bicycle exercise for 30 mi at the target HR of 50–60%) and RE+ interval training (stationary bicycle exercise different durations at different HR, HR variability 40% between 90%) on the mental health in older women. Participants performed 12 weeks of exercise involving 30-minute RE followed by 30-min interval training or AE 3 times a week. Not depending on the type of exercise, both interventions effectively changed the IGF-1 levels, functional fitness, and sleep quality. However, no difference was found between the two types of exercise. A recent study investigated circulating IGF-1 levels in AD patients (n = 34) and older adults (n = 40) without dementia after acute exercise (Balke-Ware test: incremental test on a treadmill), [102]. After the intervention, higher levels of IGF-1 were observed in AD patients. Tsai et al. [106] investigated the effects of acute AE (a 30-minute bout of bicycle ergometer exercise at 65−75% of HRR) or RE (30-minute biceps curls, triceps extensions, bench presses, leg presses, leg extensions, and vertical butterflies exercises for two sets of 10 repetitions, at 75% 1RM, with a 90-second rest between sets, and a 2 minutes interval between each different exercise) interventions on neurocognitive performances and changes in circulating levels of neuroprotective growth factors (e.g., BDNF, IGF-1, and VEGF) in older adults (n = 55, age: 60–80 y) with amnestic MCI. Findings of the study showed that an acute bout of AE significantly increased serum levels of BDNF and IGF-1, but only serum IGF-1 levels increasing were observed after RE.

Ding et al. [107] indicated that IGF-1 also contributes to the BDNF pathway to mediate exercise-induced synaptic and cognitive plasticity. Thus, stimulation of the uptake of blood-borne IGF-I by nerve cells may lead to the synergetic neuroprotective effect by inducing BDNF levels [108]. Similarly, a review pointed out that there is a putative idea about the possible pharmacological and biochemical mechanisms of several types of antidepressants (selective serotonin reuptake inhibitors) and antipsychotics, in which activation of IGF-1 and BDNF pathways seem to drive the therapeutic effects [109]. Nevertheless, besides the IGF-1, mechanisms such as hormonal involvement, neurotransmitter levels, and the balance between sympathetic and parasympathetic central activity should also consider for the neurobiology of depression [104]. A recent review investigated the molecular mechanisms of exercise in the coronavirus disease 19 (COVID-19) pandemic on mental health [110]. It indicated that exercise enhances IGF-1 levels and the activity of the PGC-1α/FNDC5/Irisin pathway leading to neuronal survival and the maintenance of good mental health, which is poorly affected in the quarantine period during COVID-19.

#### **2.4 Vascular endothelial growth factor (VEGF)**

VEGF is a signal protein belonging to a subfamily of growth factors. It is produced by cells that stimulate the formation of blood vessels [111]. VEGF is classically associated with stimulation of angiogenesis and vasculogenesis [112]. However, recent evidence has shown that it also affects nerve cells and plays an important role in hippocampal neurogenesis and neuroprotection [113]. As BDNF or IGF-1, exercise-induced VEGF may have a role in ameliorating the pathological process of AD [114]. Moreover, VEGF, because of its active role in neurogenesis, may mediate antidepressant therapies [115]. These authors revealed that VEGF level changes were not associated with depressive symptom improvement due to repetitive transcranial magnetic stimulation but strongly correlated with reduction of anhedonia. Another

study showed higher serum VEGF levels in non-responders than in responder patients with MDD treated with antidepressants [116]. A recent study pointed out that VEGF is linked with increased BBB permeability [117]. It could be an important pathology linking stress and psychiatric disorders, including MDD. Results of the study indicated that VEGF might play an essential role in the pathogenesis of depression by increasing the BBB permeability. Parallel to these findings, Hadidi et al. [118] indicated various neurological and physical symptoms due to post-stroke depression (PSD), including a decrease in serum BDNF and VEGF levels A study revealed that VEGF-induced antidepressant-like effects involve modulation of norepinephrine and serotonin systems [119].

VEGF is a candidate anti-stress molecule because of its role in enhancing antidepressant-like effects in response to different external stimuli such as stress, learning and exercise [119]. An animal study showed that exercise can be leading the expression of VEGF, especially in the hippocampus [120]. In humans, an immediate increase was shown in serum VEGF levels after high-altitude exercise and its effect lasted up to 1 month after training [121]. On the contrary, another study indicate that exercise did not have a positive effect on VEGF [122].

Krogh et al. [123] investigated the effect of an AE (45 min/3d/w for 3 months, stationary bike at 80% of HRmax) intervention on hippocampal volume and serum BDNF, VEGF, and IGF-1 in patients with MDD (n = 79, age: 18–60 y). In conclusion, the authors observed no significant increase in resting levels of neurotrophins. But they pointed out a positive association between change in hippocampal volume and depressive symptoms. A randomized controlled trial revealed that in older adults (n = 55) with amnestic MCI, both exercise regimes (16 weeks) AE and RE are effective in increasing neurotrophins (BDNF, IGF-1, and VEGF, etc.,) reducing some inflammatory cytokines, and facilitating neurocognitive performance [124]. Another study [106] showed that VEGF levels tended to increase after AE but no significant changes were observed after RE. Du et al. [125] compared the acute effects of high (80% 1-RM) and low (40% 1-RM) RE (sitting pull-down, sitting shoulder press, sitting chest push, sitting leg kick) with or without blood flow restriction on perception, BDNF, and VEGF levels in patients (N = 24) with post-stroke depression (PSD). Based on the findings, there were no increases in BDNF and VEGF levels of each group compared to before exercise. After exercise, the BDNF levels of the low RE group increased significantly, but without change was observed in the VEGF level of the low RE group. In addition, blood flow restriction RE may increase the serum BNDF and VEGF levels of PSD patients by increasing the body's blood lactic acid concentration.
