Fibroblast growth factor 21 regulates neuromuscular junction innervation through HDAC4 in denervation-induced skeletal muscle atrophy

Muscles remain in a state of partial contraction called muscle tone, which is important for generating reflexes, maintaining posture and balance and controlling the physiology of organs. When the nerve that stimulates the muscle is cut, it loses its tone and becomes flaccid. The tone of the striated skeletal muscles (SkM) depends on various factors, such as the balance between protein synthesis and degradation, cytokines and chemokines, and innervation by motoneurons.

Neuromuscular disorders, such as amyotrophic lateral sclerosis (ALS), are characterized by nerve degeneration and muscle atrophy, culminating – in the most severe cases – in death. Currently, few or no treatments effectively reverse these conditions, prompting searching for possible molecular candidates involved in neuromuscular pathophysiology that new drugs could target.

In physiological stress, striated skeletal muscles secrete the peptide hormone FGF21, negatively regulating muscle mass and function. Its involvement in muscle atrophy is not fully understood, however. In this preprint, the authors demonstrate the relationship between homeostasis in SkM, denervation, and the expression of Fgf21. They highlight the molecular targets, proteins and signalling pathways involved in this process, positioning FGF21 as a possible target for treating neuromuscular disorders.

Key findings of the preprint

• Fgf21 expression is consistently elevated in all public neurogenic muscle atrophy datasets (GSE18119, GSE48574, GSE52766, GSE87108, GSE49826) analyzed and upregulated in the tibialis anterior (TA) muscle denervation model in vivo. Although Fgf21 is related to the endocrine context, its expression in denervation/muscle atrophy has no contribution from endocrine organs. There is a positive correlation between the expression of Fgf21 and atrogenes, leading to loss of muscle weight, size, and grip strength, which may contribute to the progression of neurogenic muscle atrophy.
• Although partial muscle recovery occurs, denervation decreases Fgf21 expression and grip strength and suppresses protein degradation markers, suggesting that Fgf21 deficiency protects against muscle atrophy. Fgf21 overexpression followed by denervation increases circulating levels of FGF21 (an effect independent of denervation) and reduces grip strength, myofiber size and muscle weight. To confirm their findings, the authors overexpressed Fgf21 in the TA muscle of Fgf21-null animals, followed by denervation. The data obtained confirms that Fgf21 induces muscle atrophy, and this gene’s ablation can be considered a neuromuscular protection mechanism.
• Fgf21-dependent genetic alterations and denervation were associated with immunity and axon guidance, the latter of which is crucial for neuromuscular junction (NMJ) establishment and SkM function. The Fgf21 knockdown, followed by denervation, partially recovered NMJ structure, and in vitro, treatment with recombinant FGF21 reduced the expression of NMJ-associated genes and the intensity of acetylcholine receptors (AChR). These results suggest that the muscle atrophy-resisting effect of Fgf21 deficiency is associated with improved NMJ innervation.
  
• FGF21 affects HDAC4, a class IIa histone deacetylase (HDAC) essential for neural activity and the muscle’s response to denervation. NMJ denervation decreases FGF21 accumulation in the cytoplasm. The absence of FGF21, therefore, increases the cytoplasmic presence of HDAC4, promoting NMJ reinnervation and restoring muscle homeostasis.
  
• Denervation increases mRNA and protein TGFB1 levels, specifically in SkM, but not in plasma. TGFB1 in vitro treatment increased FGF21 expression at both the mRNA and protein levels through the non-canonical (JNK/c-Jun) rather than the canonical (Smad) pathway and induced muscle atrophy, similar to FGF21 treatment. The silencing of Fgf21 before TGFB1 treatment partially restored AChR intensity and blocked basal and TGFB1-induced Fgf21 upregulation in mature myotubes but not in myoblasts, suppressing TGFB1-induced atrophy. In physiological conditions, Fgf21 silencing had no effect without TGFB1 treatment.
  
• Fibro-adipogenic progenitors (FAPs) secrete Activin A (TGFB superfamily). In denervated muscle, the number of FAPs (Den-FAPs) and TGFB1 (both mRNA and protein levels) is significantly higher. In vitro, medium conditioned with Den-FAPs secretion reduced myotube diameter and AChR intensity and increased Atrogin1 protein level. This effect was mitigated with the silencing of Fgf21, indicating the interaction between Fgf21, TGFB1 and FAPs in muscle atrophy. In short, FAPs are triggered by denervation to release TGFB1, which promotes the expression of Fgf21 in myofibers, resulting in muscle atrophy.

Why I chose to highlight this preprint

The choice of this preprint can be understood in the context of the research I am developing as part of my PhD (= establishment of neuromuscular junctions in chicken embryos). Reflecting on new gene candidates involved in neuromuscular impairment that could play a role in muscle atrophy allows me to create more hypotheses to refine and argue my results more robustly, while at the same time allowing me to catch up on the scientific literature on this topic.

Future directions and questions for the authors.

I congratulate the authors on their approach and excellent work. While reading the preprint, I thought about how embryonic development could be one way to identify models and gene and protein pathways deregulated in muscle atrophy. I have a few questions:

1) Have you considered profiling the temporal expression of Fgf21 in myogenesis and axogenesis in vitro?

2) Have you investigated or are you interested in looking at the possible association between the TGFB, Wnt/PCP (non-canonical) and Agrin/Lrp4/MuSK pathways in muscle atrophy?

Given that the latter two participated in establishing NMJs, I would bet that they are also involved in muscle atrophy.

Tags: fgf21, hdac4, muscle atrophy, neuromuscular junction

doi: https://doi.org/10.1242/prelights.40273

Read preprint (No Ratings Yet)