Aerobic exercise promotes hippocampal neurogenesis through skeletal myofiber‐derived vascular endothelial growth factor

The beneficial effects of exercise, including improved cognitive performance, memory, and mood, are well documented and thought to be a result of increased hippocampal neurogenesis through the proliferation and differentiation of neural precursor cells (NPCs) within the dentate gyrus, although the mechanisms by which this occurs are unknown. Following aerobic exercise, several factors are released into circulation. Of these, vascular endothelial growth factor (VEGF), along with insulin growth factor‐1 (IGF‐1), plays a key role in promoting angiogenesis and hippocampal neurogenesis (Fabel et al. 2003; Cotman et al. 2007); however, the source of this VEGF and its mechanism of action are not well defined. A recent study by Rich et al. (2017) posited that VEGF present in skeletal myofibers may be released during exercise and influence downstream signalling pathways, playing an important role in the neurogenic process. The potential role of skeletal myofiber VEGF in neurogenesis supports the utilization of exercise regimes in the treatment of certain central nervous system (CNS) disorders.

In a series of elegantly designed experiments, Rich et al. (2017) demonstrated that skeletal myofiber VEGF is required to enhance hippocampal neurogenesis following exercise. To achieve this, conditional ablation of the skeletal myofiber‐specific VEGF gene was achieved in male NES‐GFP/VEGF LoxP mice, which were divided into two groups: (1) HSA‐CRE‐ER^T2 positive, VEGF gene‐ablated group (VEGF^HSA−/−) and (2) HSA‐CRE‐ER^T2 negative, non‐ablated group (VEGF^f/f). Each group was further subdivided into two subgroups: those that received exposure to a running wheel for 2 weeks or sedentary controls (Fig. 1). Cerebral blood flow (CBF), NPC and endothelial cell proliferation, and VEGF levels in skeletal muscle were compared between groups.

Figure 1. Experimental paradigm used by Rich et al. (2017).

A, mutant male NES‐GFP/VEGF LoxP mice on a C57BL/6 background were divided into four groups based on ablation of VEGF and exposure to exercise training. B, the experimental schedule lasted 5 weeks. VEGF ablation was achieved within 3 weeks by tamoxifen administration for 5 days (day 0–4). The exercise phase lasted 2 weeks (day 21–36). BrdU labelling was conducted twice during the training phase to label proliferating NPCs. Mice were tested for maximal exercise performance on day 36 and killed by perfusion on day 37 to measure VEGF levels and NPC proliferation.

Using an enzyme‐linked immunosorbent assay (ELISA) on tissue from the plantaris, a skeletal muscle used during running, Rich et al. (2017) show that the VEGF ablation paradigm resulted in a ∼60% reduction of VEGF expressed in the skeletal muscle of VEGF^HSA−/− mice compared to VEGF^f/f controls, confirming that the ablation paradigm effectively decreased the amount of skeletal myofiber VEGF. Interestingly, VEGF levels were not significantly increased in the plantaris of exercised mice compared to sedentary mice when measured following the exercise training period in either genotype. This could indicate that the skeletal myofiber VEGF was released into the circulation instead of being sequestered in the muscle, that transient expression of VEGF returned to baseline levels prior to measurement, or that exercise did not lead to increased VEGF expression.

The presence of skeletal myofiber VEGF was necessary for improved performance in maximal speed and endurance following training. Not only were VEGF^f/f mice that received exercise training the only group to show improvements in performance, but they also outperformed all the other groups. The authors suggest that differences in performance may be due to blood flow variations between the VEFG^f/f and VEFG^HSA−/− mice in response to exercise. Thus, the mechanism for improved physical performance in the exercise trained VEGF^f/f group may depend on increased peripheral VEGF‐dependent angiogenesis (Delavar et al. 2014) leading to improved tissue oxygenation.

To identify proliferating NPCs, 5‐bromo‐2′deoxyuridine (BrdU) was administered during the exercise training phase. Mice were killed 1 day following the completion of the training phase and tissue was collected for analysis. Hippocampal NPC and endothelial cell proliferation was assessed with immunostaining for Nestin⁺ and CD31⁺ expression, respectively.

Exercised VEGF^f/f mice had a higher number of proliferating NPCs in the hippocampus as measured by Nestin⁺ and BrdU⁺ immunostaining compared to both sedentary groups and the exercised VEGF^HSA−/− group, confirming that skeletal myofiber VEGF is necessary for promoting exercise‐mediated neurogenesis in the hippocampus. Notably, endothelial cell proliferation, as measured by CD31⁺/BrdU⁺ staining, was similar in all groups, indicating that proliferation was specific to NPCs and that angiogenesis was not the reason for increased neurogenesis. These findings suggest that a combination of skeletal myofiber VEGF and exercise is necessary for hippocampal‐specific neurogenesis. One possible explanation for this observation is that VEGF^HSA−/− mice had decreased hippocampal blood flow compared to VEGF^f/f mice, providing a potentially less favourable environment for neurogenesis, whereby fewer neurotrophic factors or signalling molecules, such as VEGF, IGF‐1, and fibroblast growth factor 2, reach the dentate gyrus (Olson et al. 2006). This finding also demonstrates the importance of VEGF in the maintenance of normal CBF rates.

One interesting consideration is that trained VEGF^f/f mice had both increased NPC proliferation and improved exercise performance in endurance and maximal speed. These findings allude to a link between neurogenesis in the dentate gyrus and physical performance, but the authors do not explicitly suggest the nature of that link. More work is required to determine whether these processes are directly related, indirectly related, or modulated by a common mechanism.

This research provides compelling evidence that skeletal myofiber VEGF is crucial for stimulating NPC proliferation with aerobic exercise, as only mice with normal levels of skeletal myofiber VEGF exhibited increased neurogenesis following the training phase. The mechanism by which skeletal myofiber VEGF induces neurogenesis has yet to be elucidated; nonetheless, sources from outside the CNS can play a key role. Indeed, peripheral inhibition of VEGF prevents exercise‐induced neurogenesis (Fabel et al. 2003), suggesting that circulating VEGF derived from skeletal muscle has the potential to induce direct effects on NPCs or influence downstream pathways that affect neurogenesis. Since VEGF can cross the blood–brain barrier (BBB), it is possible that circulating skeletal myofiber VEGF asserts its effects directly on NPCs by causing them to proliferate or increasing their survival (Lange et al. 2016). It is also possible that skeletal myofiber VEGF stimulates NPCs to further secrete VEGF, involving them in autocrine or paracrine signalling pathways which can promote neurogenesis and survival (Lange et al. 2016). Alternatively, skeletal myofiber VEGF might impact NPC proliferation and survival by influencing downstream signalling pathway cascades, such as the MEK–MAPK, PI3K–Akt and Src–eNOS pathways (Lange et al. 2016), or interacting with other factors, such as brain‐derived neurotrophic factor or IGF‐1 (Cotman et al. 2007). Further studies are necessary to determine other players in this signalling cascade and whether these findings apply to all tissues and sources of VEGF.

The ability of VEGF to disrupt the BBB provides another possible mechanism by which skeletal myofiber VEGF may affect NPC proliferation. Increasing BBB permeability provides circulating cells, factors, and signalling molecules with direct access to brain parenchyma. This can have implications in cases of acute cerebrovascular insults and neurodegenerative diseases whereby increased peripheral VEGF resulting from exercise or exogenous administration may modulate neurogenesis by promoting anti‐inflammatory cell and signalling molecule trafficking across the BBB, or by direct interaction with NPCs (Lange et al. 2016). Notably, increasing BBB permeability may also increase the risk of developing edema and hemorrhage or could exacerbate damage by recruiting pro‐inflammatory cells (Lange et al. 2016). Thus, the balance between dose, timing, and VEGF source must be carefully evaluated to determine the translational relevance of these findings.

The work by Rich et al. (2017) enhances our understanding of the intricacies between exercise and neurogenesis and provides convincing evidence that VEGF plays a role in this process. However, more work is necessary to clarify the mechanisms by which skeletal myofiber VEGF affects exercise‐induced neurogenesis and whether proliferating NPCs survive and give rise to mature neural cells. This could be accomplished by conducting long term experiments; exploring the link between VEGF, exercise‐induced neurogenesis, and resident and circulating inflammatory cells, astrocytes, and other neural cell types; investigating BBB permeability; and studying other cellular factors, both peripherally and within the CNS. Additionally, although VEGF has been shown to increase in specific regions of the brain following exercise (Tang et al. 2010), it would be interesting to explore the isoform and source of VEGF that increases and whether it co‐localizes with proliferating NPCs.

Rich et al. (2017) clearly show that skeletal myofiber VEGF is necessary for neurogenesis in the hippocampus resulting from aerobic exercise; it would be interesting to investigate whether their training regime has anatomical specificity limited to regions within the hippocampus (Vivar et al. 2016) or if a similar paradigm can promote neurogenesis in other brain regions, such as the subventricular zone, another neurogenic niche where NPCs reside. Likewise, it would be interesting to determine whether non‐cardiovascular based exercise regimes also have the ability to induce NPC proliferation and whether this process is also mediated by skeletal myofiber VEGF.

Further directions could also include examining the role that VEGF plays in exercise‐induced neurogenesis to promote functional recovery following injury or disease and determining whether the resulting neurogenesis is beneficial and has enough therapeutic potential to improve clinical outcomes. These studies could provide further evidence to support the prescription of exercise as a therapeutic intervention. Thus, similar experiments could be performed with neurodegenerative models of disease, since one of the key limitations to recovery in these conditions is the loss of neural cells. This would help the understanding of conditions and mechanisms that dictate positive or adverse neurogenic responses to exercise.

In summary, Rich et al. (2017) demonstrated that skeletal myofiber VEGF, in combination with exercise, promotes hippocampal neurogenesis. They report increased NPC proliferation in the dentate gyrus following aerobic exercise training and hypothesize that it is regulated by skeletal myofiber VEGF, which was shown to be an important link between exercise and neurogenesis. Their findings encourage further research that may uncover the mechanisms underlying this process and provide further rationale for the prescription of exercise regimes with the goal of enhancing regenerative strategies in certain CNS disorders.

Additional information

Competing interests

None declared.

Author contributions

Both authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.