PTMs offer fast, reversible control of protein function and signaling during exercise, matching cell metabolism and structure to changing mechanical and energetic demands (1, 2). Distinct training modes leave unique PTM signatures: Endurance training mainly reshapes mitochondrial acetylation to promote oxidative metabolism, while resistance training primarily modifies contractile and cytoskeletal protein phosphorylation to follow strength changes (3, 4). Outside of skeletal muscle, long-term exercise induces cardioprotective remodeling, potentially through mechanisms involving titin mechanics and PTM regulation, and emerging evidence indicates metabolite-driven lactylation to be cardioprotective (5, 6).

At the signaling nexus, AMP-activated protein kinase (AMPK) demonstrates multi-post-translational modification (PTM) integration; its activity, localization, and stability are regulated by phosphorylation, acetylation, ubiquitination, and methylation, thereby connecting energetic stress to adaptive responses (7). Mechanotransduction also interacts with chemical regulation; the deformation of sarcomeric and focal adhesion proteins alters kinase/phosphatase accessibility and substrate conformation (8). Recent progress in mass-spectrometry proteomics facilitates the creation of system-level maps of exercise-regulated PTM landscapes and enables the shift from descriptive catalogs to pathway-level inference (9, 10). In this expansive domain, phosphorylation is the most thoroughly quantified and mechanistically accessible axis for linking acute exercise stimuli to synchronized metabolic and structural results. We propose that exercise adaptation is governed by kinase-phosphatase feedback circuits that establish a dynamic phosphorylation setpoint. This setpoint, which is established by the balance of kinase-driven signal amplification and phosphatase-mediated feedback, encodes the nature of the exercise stimulus and directs the specificity of long-term adaptive responses (2, 11–17).

Exercise perturbs cellular energy homeostasis, calcium handling, redox state, and cytoskeletal tension; all these inputs contribute to the activation of kinases with opposing phosphatase activity. These factors together influence the magnitude, duration, and localization of phosphorylation signals. This framework contains AMPK that incorporates energetic stress to increase glucose uptake, fatty-acid oxidation, and mitochondrial biogenesis, while mechanosensitive scaffold structures such as titin and focal-adhesion complex control kinase access and substrate conformations (5, 7, 8). Recent phosphoproteomic studies reveal a coordinated, mode- and intensity-dependent phosphorylation program in both humans and animal models, enabling inference of upstream kinases via kinase–substrate enrichment analysis (KSEA) and HIIT workload–based analyses allow inference of potential regulatory drivers and feedback loops (3, 9–11). These individual findings together point towards a more focused investigation of the kinase-phosphatase axis as an organizing principle for dynamic control of exercise adaptation.

Protein phosphorylation, the primary post-translational modification (PTM) that directs exercise-induced adaptations, is tightly regulated by the coordinated action of kinases and phosphatases (27, 55). This process also serves as a vital link for connecting energy stress signals, mechanical cue transmission, and interplays between various PTMs (9, 56). AMPK, mTOR, MAPK, and other key kinases are stimulated by exercise stimuli: AMPK mediates glucose uptake via phosphorylation of glucose transporter 4 (GLUT4) and fatty acid oxidation via phosphorylation of acetyl-coenzyme A carboxylase (ACC), sustaining energy homeostasis during exercise (7, 28); mTOR activates p70 ribosomal S6 kinase (p70S6K) and eukaryotic translation initiation factor 4E-binding protein 1 (4E-BP1) in skeletal muscle for protein synthesis needed in resistance training hypertrophy (25, 57); and the MAPK family, such as the ERK1/2 family, coordinates anabolic and catabolic transcriptional programs through regulating transcription factors like c-Fos (12, 58).

Meanwhile, phosphatases build up an essential negative feedback circuit: MAPK phosphatases MKPs inactivate MAPKs by dephosphorylation and inhibit excessive MAPK activation of inflammatory response (12); the PHLPP family inhibits Akt activity to prevent cell damage caused by sustained signal transmission (25); and P-type phosphatases PHPT1 and LHPP reverse the modification of histidine phosphorylation and keep the reversibility of non-canonical PTMs (16). These three groups of phosphatases ensure that the intensity and duration of exercise signals match physiological demands.

Different kinds of exercises have different ways to control these kinases and phosphatases; this is important for how they adapt to the exercise (59). Acute energetic stress from HIIT causes PHLPP2 expression. One important role is as a suppressor of Akt overactivity, which has been proposed to serve a protective role against an unhealthily high proliferation of muscle cells during metabolic stress (6); endurance training, on the other hand, keeps PHLPP activity low, which creates an environment that allows for the sustained Akt signaling needed for mitochondrial biogenesis. This effect is directly linked to the activation of oxidative metabolism genes by peroxisome proliferator-activated receptor γ coactivator 1α (PGC-1α) (7, 60). Notably, several research gaps remain: the expression differences of PHLPP1 and PHLPP2 between type I (oxidative) and type II (glycolytic) muscle fibers, as well as the impact of their subcellular localization on signal efficiency, have not been clarified (25); although histidine phosphorylation is known to regulate the activity of metabolic enzymes such as hexokinase (16, 61), the substrate profiles of its associated kinases (NME1/2) and phosphatases in exercise contexts still rely on indirect evidence from oncology and immunology fields (38). To directly address the knowledge gap regarding histidine phosphorylation, future work should incorporate acid-free sample preparation pipelines, rapid tissue freezing during in vivo contractions, and compartment-resolved isolation of mitochondria and myofibrils (53). These approaches are essential for preserving pHis stability during the transient metabolic transitions of acute exercise. Time-resolved phosphoproteomics specifically optimized for pHis detection would allow measurement of rapid histidine dephosphorylation events that may occur within seconds of contractile onset or metabolic recovery—events that are invisible to conventional workflows (16, 53). Furthermore, integrating pHis-sensitive proteomics with metabolomics and isotope-tracing (fluxomics) could reveal how histidine phosphorylation unites metabolic switches—such as glycolytic burst, PCr depletion, or mitochondrial engagement—to enzyme activity during exercise. Consequently, it is essential to perform exercise-oriented studies employing mass spectrometry for the examination of Ser/Thr/Tyr phosphorylation (9, 10), to address this knowledge deficiency (53).

Methodological limitations restrict the depth of mechanistic analysis. Static phosphoproteomics cannot capture transient dephosphorylation events of protein tyrosine phosphatase 1B (PTP1B) during HIIT (47, 62), necessitating the combination of time-resolved techniques and tissue-specific sample preparation (laser capture microdissection) to analyze phosphatase activity dynamics; chemoproteomics can identify novel substrates of phosphatases such as MKP-5 by capturing phosphatase-substrate complexes via activity-based probes (48, 63); and proximity labeling techniques (BioID) can clarify the localization of phosphatases in microdomains such as sarcomeres and mitochondria (63), explaining the spatial regulatory mechanisms underlying their substrate specificity (25). The combination of such technologies will move exercise phosphatase studies from “descriptive profiling” to “causal mechanistic” research. To address these limitations, we described the important methodological directions for combining time-resolved phosphoproteomics with spatially resolved and epigenomic technologies (Table 2).

Phosphorylation, regulated by kinases and phosphatases, is a key molecular axis implicated in exercise-induced adaptations. It controls important cellular processes like mitochondrial biogenesis, muscle hypertrophy, and metabolic flexibility. This balance is not merely a simple on-off switch; rather, it constitutes a complex regulatory hub that uses mechanical, metabolic, and hormonal signals to make sure that molecular responses match the needs of exercise. HIIT may induce an increase in PHLPP2 expression, which has been proposed to help restrain excessive Akt activation and prevent maladaptation, whereas during endurance training, PHLPP activity tends to remain low, supporting Akt-dependent mitochondrial biogenesis.

We currently lack essential knowledge about the functional interaction between phosphorylation dynamics and epigenetic regulation. Exercise is accompanied by histone acetylation, such as heightened acetylation at mitochondrial-gene promoters during endurance exercise, and can also trigger DNA methylation at loci linked to fiber type during resistance training. However, the mechanisms by which the kinase–phosphatase network facilitates these epigenetic modifications are not yet fully understood. One important question is whether AMPK, the main energy-sensing kinase, connects energetic stress to chromatin remodeling by adding phosphate groups to histone-modifying enzymes like acetyltransferases or deacetylases, and another key question is whether the regulation of Akt signaling by PHLPP alters the activity of methyltransferases like EZH2. These enzymes are responsible for creating the epigenetic memory of muscle fiber type. Furthermore, the function of non-canonical phosphorylation, particularly histidine phosphorylation, in exercise-induced epigenetic regulation is inadequately investigated and constitutes a significant area for future research.

Future work should integrate time-resolved phosphoproteomics with epigenomic profiling—such as ChIP-seq for histone marks and ATAC-seq for chromatin accessibility—to build causal paths from kinase–phosphatase activity to epigenetic modifications, gene expression, and ultimately physiological adaptation. The advancement of high-resolution quantitative phosphoproteomics and single-cell phosphoproteomics will be crucial for resolving dynamic kinase and phosphatase activities across fiber types and cellular subpopulations. Moreover, CRISPR-Cas9–mediated conditional knockouts and chemogenetic/optogenetic tools will help establish causal links between enzyme activity and adaptive phenotypes, moving beyond inferences from correlative networks.

Finally, fixing methodological inconsistencies is important for making the results clearer. For example, most studies on PTP1B and exercise focus on liver tissue, which doesn't really explain how it works in myocyte insulin signaling. To address these discrepancies, supplementary tissue- and cell-specific models are essential. Because phosphatase families have some overlap, like how MKP isoforms regulate MAPKs, we need to make several changes to stop mistakenly linking phenotypes to one enzyme (75, 76). Integrated mechanistic insights could inform translational strategies with potential applications in metabolic disease treatment, athletic performance, and sarcopenia mitigation—though substantial translational work will be required. These advances have the potential to shift aspects of exercise physiology towards more mechanistic, molecularly specific frameworks.