Moderate Aerobic Exercise May Best Activate Brain Repair Pathways After Stroke

Aerobic exercise is already a cornerstone of stroke rehabilitation, but a new narrative review argues that the intensity of that exercise may be the single most important variable determining whether the brain actually repairs itself. Published in the Bulletin of Faculty of Physical Therapy, the review by Areej Fatima, Huiyi Jiang and Jifang Qiu examines how moderate versus high-intensity aerobic exercise differentially engages the brain-derived neurotrophic factor (BDNF), phosphoinositide 3-kinase (PI3K) and protein kinase B (Akt) signaling axis after ischemic stroke. Their synthesis points to a consistent conclusion: moderate-intensity training reliably switches on this pro-survival molecular circuit, while pushing intensity too high can abolish the benefit entirely.

Ischemic stroke, caused by a blockage of blood flow to the brain, remains a leading cause of disability worldwide. The authors cite 2019 global burden figures in which ischemic stroke affected 77.19 million people, caused 3.29 million deaths and accounted for 63.48 million disability-adjusted life years. Because prolonged drug therapy carries side effects and financial strain, non-pharmacological rehabilitation, particularly sustained activities such as treadmill training, swimming and cycling, has become central to recovery. Physical therapists use aerobic exercise to rebuild strength, coordination, balance and ambulation, and the review frames it as safer than high-strain anaerobic work for survivors with residual motor impairment, cardiovascular instability or hypertension.

The molecular story at the heart of the review begins with BDNF, a pleiotropic growth factor renowned for its roles in neuroplasticity and cell survival. In a healthy neuron, BDNF binds its receptor TrkB, which activates PI3K; PI3K generates lipid messengers that recruit and activate Akt at the plasma membrane. From this signaling hub, Akt phosphorylates a network of substrates that suppress apoptosis, regulate metabolism and drive gene transcription. During cerebral ischemia-reperfusion injury, this equilibrium collapses: BDNF signaling fails, Akt becomes inactivated, pro-apoptotic factors such as BAD and caspase-9 are unleashed, and the disinhibited kinase GSK-3β drives inflammatory and apoptotic programs, compounded by NF-κB activation and surges in tumor necrosis factor-alpha and interleukin-6.

Therapeutically, restoring this axis is a compelling target. Strategies using agents such as insulin, erythropoietin or recombinant BDNF seek to re-engage PI3K/Akt, allowing Akt to neutralize apoptotic threats, inhibit GSK-3β and activate downstream effectors including mTOR and CREB, the latter of which can upregulate BDNF expression in a self-reinforcing pro-survival loop. The review notes preclinical studies demonstrating reductions in infarct volume of roughly 30 to 40 percent and a blunted neuroinflammatory response with such pathway restoration, though these remain animal and laboratory findings rather than proven clinical benefits.

Where does exercise fit in? The review assembles evidence that aerobic activity upregulates BDNF and, through TrkB, triggers the PI3K/Akt cascade, promoting neuronal survival, synaptic plasticity and angiogenesis. In one Alzheimer’s disease model cited, treadmill exercise increased BDNF levels, activated downstream PI3K/Akt signaling and suppressed the apoptotic marker cleaved caspase-3. Another cited study reported that aerobic exercise upregulated BDNF expression approximately 1.8-fold, linked to increased CREB phosphorylation and potentiation of the PI3K/Akt pathway. In human models, cited work by Dadkhah and colleagues found aerobic exercise elevated BDNF, which in turn triggered Akt phosphorylation via the PI3K cascade. Critically, one cited study reported that PI3K inhibition completely prevented BDNF upregulation, underscoring the pathway’s role in translating physical activity into neural repair.

The intensity question is where the review delivers its most striking claims. Moderate-intensity aerobic exercise is defined as roughly 50 to 70 percent of an individual’s maximal capacity, translating clinically to about 55 to 69 percent of maximum heart rate. Cited work by Leger and colleagues suggested that skeletal muscle contraction during moderate exercise raises BDNF and irisin levels, with the BDNF increase highest after 30 to 60 minutes, activating PI3K/Akt essential for nerve survival and plasticity. In preclinical models, this corresponds to treadmill speeds of 8 to 18 meters per minute; a protocol of 15 to 18 meters per minute for 20 to 30 minutes daily, reported by Ploughman and colleagues, was described as facilitating post-stroke motor relearning. These regimens, typically 20 to 60 minutes per session over several weeks, were reported to upregulate BDNF and activate PI3K/Akt without the detrimental inflammation or oxidative stress associated with high-intensity exercise.

Human evidence cited in the review aligns with this window. Hill and colleagues employed training at 60 to 80 percent of heart rate reserve three times weekly for 12 weeks, resulting in motor improvements linked to increased serum BDNF and insulin-like growth factor 1. Hu and colleagues used a four-week moderate-intensity protocol of 30 minutes of daily treadmill running, five days per week, in an ischemic stroke rat model at approximately 50 to 70 percent of maximal oxygen consumption, reporting elevated BDNF and vascular endothelial growth factor supporting neuronal survival and angiogenesis. A 2024 Bayesian network meta-analysis encompassing 4,111 participants established that regimens of 8 to 12 weeks, with 20 to 40 minute sessions at 60 to 80 percent of heart rate reserve or 14 to 16 on the Borg exertion scale, three to five times per week, formed an efficacious framework for improving peak oxygen uptake and mobility.

The ceiling effect may be the review’s most provocative finding. In one cited study, moderate treadmill running robustly increased phosphorylated Akt levels after brain injury, while high-intensity running completely nullified that activation, establishing what the authors call a strict therapeutic ceiling. In stroke patients, a four-week moderate-intensity continuous training protocol of 40-minute sessions five days per week produced functional improvements despite a paradoxical drop in peripheral BDNF, a finding the authors interpret as supporting beneficial redistribution of BDNF to the brain, where it engages TrkB, activates PI3K/Akt and shifts the balance of Bcl-2 to Bax toward neuronal survival. The safety rationale is equally practical: survivors often face impaired balance, comorbidities and fatigue, and high-intensity exercise elevates risks of falls, cardiovascular strain and injury.

The review also connects this signaling axis to systemic metabolic health. Insulin resistance, associated with type 2 diabetes and metabolic syndrome, is identified as a pre-stroke risk factor and modifier of post-stroke outcomes. Glucagon-like peptide-1 receptor agonists such as liraglutide and semaglutide, and SGLT2 inhibitors, are described as having demonstrated direct neuroprotective effects in preclinical models by enhancing PI3K/Akt activity, reducing inflammation and elevating BDNF. Moderate-intensity training itself shows metabolic convergence: in diabetic models, an eight-week program enhanced muscular insulin sensitivity via PI3K/Akt activation, and in humans a single 45-minute session acutely upregulated PI3K/Akt gene expression in adipose tissue. Cross-condition work in a doxorubicin heart-injury model similarly found moderate treadmill running at 50 to 70 percent of maximal oxygen uptake activated a BDNF/AMPK/PI3K/Akt axis that blocked mitochondrial apoptosis, suggesting the mechanism is a conserved cellular defense rather than a stroke-specific quirk.

The authors are candid about limitations. Translation from rodents to humans is unreliable, since forced treadmill running in animals differs from human exercise. Patient heterogeneity across recovery stages, lesion types and comorbidities makes it difficult to isolate exercise effects. Optimal duration, frequency and post-injury timing remain unclear, the mechanisms behind the high-intensity ceiling effect are not well understood, and combination strategies with drugs, brain stimulation or diet lack systematic investigation. The review’s conclusions should therefore be read as a mechanistic synthesis and a rationale for personalized, intensity-driven prescriptions, not as settled clinical guidelines. Future work, the authors argue, should pursue dose-response studies, biomarker-guided exercise prescription and integration of neuroimaging with long-term functional outcomes to determine precisely how much, how hard and how soon after stroke the body’s own repair signaling can best be switched on.

Subject of Research: How aerobic exercise intensity modulates BDNF, PI3K and Akt signaling for neurorepair after ischemic stroke

Article Title: The effect of different intensities of aerobic exercise on neurorepair based on BDNF, PI3K, and Akt signaling pathways after ischemic stroke: a narrative review

Article References: Fatima, A., Jiang, H., & Qiu, J. (2026). The effect of different intensities of aerobic exercise on neurorepair based on BDNF, PI3K, and Akt signaling pathways after ischemic stroke: a narrative review. Bulletin of Faculty of Physical Therapy, 31(1), Article 34. https://doi.org/10.1186/s43161-026-00369-4

Image Credits: AI Generated

DOI: 10.1186/s43161-026-00369-4

Keywords: ischemic stroke, aerobic exercise, BDNF, PI3K, Akt, neuroplasticity, exercise intensity, neurorehabilitation, stroke recovery, signaling pathways, moderate-intensity training, narrative review