Originally published: Exercise-linked FNDC5/irisin rescues synaptic plasticity and memory defects in Alzheimer's models (2019).
Despite decades of research, successful treatments for Alzheimer’s disease (AD) have remained elusive. Manifesting in progressive memory loss and cognitive decline, Alzheimer’s is characterized by synaptic degradation and the impairment of long-term potentiation (LTP). A key hallmark of Alzheimer's pathology is the accumulation of AβO, a soluble form of the amyloid-beta protein. AβO is found to cluster at synapses in the prefrontal cortex and hippocampus and is thought to contribute to the loss of dendritic spines.1 While pharmacological treatments addressing AβO buildup are still in development, researchers have recently described a novel behavioral mechanism to counter its associated degenerative effects. In Nature Medicine, Lourenco et al. present evidence to support the role of irisin, a peptide hormone released during muscle contraction, in decreasing AβO levels and rescuing memory loss.2 If replicated, their work has the potential to create a promising new direction for Alzheimer's research and promote exercise as an effective lifestyle intervention against future cognitive disease.
Irisin is a myokine, a muscle-derived hormone, created by the cleavage of the FNDC5 protein in response to exercise. Irisin has previously been linked to the upregulation of BDNF in the hippocampus, a neurotrophin that stimulates synaptic growth and development.3 With this, in their 2019 paper, Lourenco et al. asked whether exercise-evoked irisin could reduce AβO-related degeneration. Working primarily with mice, they first used Western blotting to confirm that irisin’s FNDC5 precursor was expressed in similar levels to the human hippocampus. Although AβO buildup is found in both the cortex and the hippocampus, the author’s decision to focus solely on the latter is likely due to key differences between the human and mouse cortex. Importantly, mice lack a dorsolateral prefrontal cortex, thus limiting the ability to translate findings to humans.4 Lourenco et al. (2019) set the stage for their primary results by comparing irisin levels in the cerebrospinal fluid (CSF) and blood plasma of clinical AD patients with controls. Here, they found decreased irisin concentration in the CSF of AD individuals, with no change to plasma levels relative to healthy participants. This indicated a connection between AD pathology and irisin deficiency.
The authors’ key findings came via in-vivo behavioral tasks, specifically a novel object recognition test (NOR) to study the impact of AβO and irisin on memory. They demonstrated that AβO-infused and irisin knock-down mice both performed poorly on the memory task compared to controls, spending equal amounts of time exploring the new and conditioned objects. As stated, AβO accumulation has been associated with reduced dendritic spine density, specifically in the thin spines that strengthen new synapses during learning. The performance of the AβO mice illustrates the memory decline seen in clinical AD as a result of impaired capacity for LTP. They then observed that AβO mice who had also been infused with irisin performed with equal competence as controls, demonstrating a boost in memory and suggesting a recovery in plasticity. These behavioral findings were contextualized functionally and histologically. Electrophysiology performed on subjects’ hippocampi revealed that samples with both AβO and irisin exhibited normal excitatory potentials, while those with only AβO were significantly reduced in magnitude. This indicated that irisin had prevented the synaptic impairment associated with amyloid-beta clusters. Similar effects were observed when analyzing spine density, with the irisin-infused AD mice resisting the pruning expected with AβO. These results suggest irisin has a potential protective effect against the neurodegeneration of Alzheimer’s.
Figure 1. Lourenco et al. (2019) investigated the effect of exercise on novel object memory tests (NOR) in mice and compared the performance of exercised and sedentary AβO subjects with controls. Success in the NOR is primarily increased exploration of new, rather than conditioned objects. 1.A reports these findings 24 hours after AβO infusion, at the conclusion of a 5-week exercise protocol, while 1.B details results 5 days post.
These findings are significant, but their translational relevance in treating human Alzheimer’s is less immediately apparent. The authors address this by using exercise, the impetus for irisin production, to replicate their results. Here, they subjected mice to 60 minutes of swimming for five sessions a week over five weeks. Critically, they observed that exercised AD mice performed with similar success to controls and far better than sedentary AβO recipients (1A, B). As anticipated, these mice were found to have higher levels of irisin than their non-exercised counterparts. This suggested that de novo irisin production, as a result of aerobic exercise, counteracted the effects of AβO accumulation. Notably, AβO infusions occurred after the exercise period concluded. This proposes that physical activity has a prophylactic effect against the development of Alzheimer’s.
The importance of Lourenco and colleagues’ findings cannot be understated. As it stands, there are no pharmaceutical interventions that target Alzheimer’s before the onset of symptoms. Available drugs like lecanemab and donanemab, target the buildup of AβO, but are exceedingly expensive and only delay disease progression.5 In addition to supporting exercise as a preventive behavior, Lourenco et al. (2019) also introduce irisin as a target for future pharmacological treatments. In this respect, their work leaves questions surrounding the mechanism of irisin’s effects. For example, it is unclear whether exercise-evoked irisin is produced peripherally or at FNDC5 proteins in the brain. The former would require irisin be capable of crossing the blood-brain barrier itself or communicate intracellularly with secondary messengers. As stated, Other work, such as that of Wrann *et al. (*2013) **suggests a connection between irisin and the expression of the neurotrophin BDNF. However, the underlying mechanism of this “muscle-brain axis,” including receptor targets and signaling pathways, remains to be described.3