Exercise-induced histone acetylation – playing tag with the genome

Short-term and longitudinal studies have shown that endurance exercise training extends life expectancy and reduces risk for many chronic disorders, including obesity, insulin resistance and type 2 diabetes (Hawley, 2004). Endurance exercise training orchestrates numerous morphological and metabolic adaptations in skeletal muscle, including mitochondrial biogenesis and an enhanced capacity to oxidize glucose and fats (Hawley, 2004). We have previously utilized global transcriptome expression technologies such as oligonucleotide arrays, targeted gene expression analysis, etc., to demonstrate that these changes are the culmination of transcriptional adaptations induced with individual acute bouts of endurance exercise (Mahoney et al. 2005). What has received less attention is how alterations in exercise-mediated DNA transcription are a function of post-translational modifications of histone proteins, which regulate transcriptional repression and/or initiation via chromatin remodelling. Acetylation of histone 3, one of the four types of histones important for genomic DNA packaging, at lysine residues 9, 14 and 36 (H3K9/14/36) is associated with transcription initiation and elongation. Histone acetylation is modulated by histone acetyltransferase (HAT) and histone deacetylase (HDAC) activity, which in turn can be regulated by post-translational modifications such as phosphorylation, ubiquitination-mediated proteosomal degradation, sumolation, etc. Potthoff and colleagues (2007) have illustrated that over-expression of HDAC5 is negatively correlated with endurance training-mediated adaptations in mouse skeletal muscle. It remains unknown how histone acetylation and HDAC activity are modulated in response to an acute bout of exercise in humans. McGee et al. (2009), in a recent article in The Journal of Physiology, sought to address this issue by examining the effects of an acute bout of exercise on histone 3 modification, in conjunction with analysing the regulation of class IIa HDAC enzymes (McGee et al. 2009).

McGee and colleagues subjected young men to an acute bout of cycling exercise (60 min at ∼75%) (McGee et al. 2009). Muscle biopsies were obtained from the vastus lateralis muscle before and immediately after the exercise bout. Western blot analysis was carried out on nuclear and whole muscle extracts to access the global histone 3 acetylation and class IIa HDACs (isoforms 4, 5, 7 and 9) protein content in response to an acute bout of exercise. Real-time PCR was used to measure the mRNA content of the HDAC isoforms immediately post-exercise. The authors also measured global HDAC activity and immunoprecipitated ubiquitinated HDACs as a marker of HDAC targeted for proteasomal degradation. Lastly, phosphorylation-mediated activation of class IIa HDAC kinases, AMPK, PKD and CaMKII, were measured in whole muscle extracts.

Acute exercise increases histone 3 lysine 36 acetylation

McGee and colleagues demonstrated for the first time that an acute bout of exercise led to increases in H3K36 acetylation levels (McGee et al. 2009). The acetylation of this conserved residue has been shown to be associated with transcriptional elongation. Surprisingly, H3K9/14 acetylation, linked with transcriptional initiation, remained unchanged with exercise. The authors proceeded to evaluate if differential regulation of class IIa HDACs may explain the enhanced H3K36 acetylation post-exercise.

Acute exercise alters HDAC 4 and 5 subcellular localization without affecting their content and activity

Class IIa HDAC isoforms 4, 5, 7 and 9 mRNA were found to be stably expressed within muscle and no changes in mRNA content were observed with exercise. With the exception of HDAC7 protein content, which could not be measured consistently in this study, the content of all other isoforms remained unchanged with exercise. The authors measured global HDAC activity and found it to be unaffected by the acute bout of cycling. Despite no changes in total HDAC content and activity, the authors hypothesized that class IIa HDAC function can be homeostatically regulated via ubiquitin-mediated proteasomal degradation and/or subcellular localization. Ubiquitination of HDAC5 (negative regulator of skeletal muscle adaptations post-exercise) was significantly higher following acute exercise. No alterations were observed in the ubiquitination status of other HDAC isoforms. The authors postulate that the lack of change in steady-state protein content of HDACs indicates an absence of degradation immediately following exercise, despite increased ubiquitinated HDAC5. Lastly, the nuclear abundance of class IIa HDACs was assessed. Only HDAC4/5 were found to be lower in the nuclear fraction post-exercise. Since overall protein content was unchanged before and after exercise, this suggested nuclear-to-cytoplasmic shuttling of HDAC4/5. The authors then evaluated the activation of putative kinases in response to exercise that may target HDAC4/5 for nuclear export.

Acute exercise activates class IIa HDAC kinases

The authors observed a significant increase in skeletal muscle AMPK and CaMKII phosphorylation after an acute bout of endurance exercise. As these two kinases are known to phosphorylate class IIa HDACs, which attenuates HDAC activity via nuclear-to-cytoplasmic export, the authors postulated this to be the mechanism underlying the observed nuclear export of HDAC4/5.

Interpretation and implications

The report by McGee et al. (2009) is the first to document the exercise-induced alterations in global histone acetylation, which may regulate chromatin remodelling in human skeletal muscle immediately after exercise. The authors demonstrated a significant increase in H3K36 acetylation after an acute bout of exercise, despite no changes in overall class IIa HDAC content and activity. Since histone acetylation is a function of both a decrease in HDAC activity and upregulation of HAT activity, we believe that an increase in H3K36 acetylation post-exercise may also be due to enhanced recruitment of transcription factors with HAT activity at the promoter sites of exercise-responsive genes. Indeed, PGC-1α, central regulator of mitochondrial biogenesis, translocates into the nucleus in response to an acute bout of exercise (Wright et al. 2007). PGC-1α has been shown to recruit transcriptional cofactors such as steroid receptor coactivator-1 and cAMP response-binding element (CREB)-binding protein, p300, which possess HAT activity (Puigserver & Spiegelman, 2003). In addition, we suggest that it would be more informative to measure global HDAC activity in nuclear fractions, rather than whole muscle homogenate, as the effect of the nuclear export of HDAC4/5 would be detected more easily in nuclear extracts only. We hypothesize that a concomitant increase in HAT activity, via PGC-1α-mediated co-activation, together with an increased HDAC4/5 nuclear export, could explain the enhanced histone 3 acetylation reported. Furthermore, future endeavours should also include investigating other regulatory modifications that can alter chromatin remodelling in response to exercise, such as promoter CpG island methylation, and histone phosphorylation and methylation.

The authors noted an increase in ubiquitinated HDAC5, without any decrements in total HDAC5 content. The importance of recovery post-exercise to garner the full benefits of exercise is well known and has been previously established (Mahoney et al. 2005). Thus, we postulate that the proteosomal degradation of HDAC5 may take place during the recovery phase after exercise. In future, the authors may consider doing a time course after an acute bout of exercise to assess any potential decrease in HDAC5 content due to proteosomal degradation, assuming that ubiquitination targets HDAC5 for degradation (Potthoff et al. 2007). Alternatively, we hypothesize that the ubiquitination of HDAC5 could also be a mechanism to shuttle it out of the nucleus, in addition to phosphorylation-mediated export via AMPK and CaMKII. For example, the tumour suppressor protein p53 is targeted for nuclear export upon mono-ubiquitination (Li et al. 2003). It would be interesting to determine the number of ubiquitin moieties attached to HDAC5, which may dictate its fate – proteosomal degradation versus nuclear export.

In summary, a single bout of endurance exercise can induce histone modifications by modulating HDAC4/5 subcellular localization, thereby initiating exercise-specific transcriptional activity (McGee et al. 2009). The findings of this study further add to the complexity of coordinated transcriptional circuitry that controls skeletal muscle gene regulation following an acute bout of endurance activity. In future, a genome-wide chromatin immunoprecipitation assay coupled with promoter arrays may be utilized for an in-depth analysis of spatial and temporal changes in histone modifications immediately after exercise and during the recovery phase. Since endurance exercise is suggested to have therapeutic potential in obesity, type II diabetes, metabolic syndrome and associated co-morbidities, it is intriguing to elucidate the underlying mechanisms that are activated upon exercise. In addition to the canonical activation of signalling kinases and transcription factors, it is now evident that exercise may also play tag with our genome via chromatin remodelling, opening up new avenues of research and creating targets for exercise-based therapeutic interventions.