Delayed onset muscle soreness (DOMS) has been defined as delayed onset soreness, muscle stiffness, swelling, loss of force-generating capacity, reduced joint range of motion, and decreased proprioceptive function [1]. Unaccustomed or strenuous exercises involving eccentric contractions are known to induce DOMS with neuromuscular changes for several days ^[2][3][4][2,3,4]. The pain of DOMS is not felt for approximately 8 h, peaks 1 or 2 days later [3], and subsides within 7 days after exercise [5]. Theodore Hough first described DOMS in 1902, attributing soreness to ruptures in the muscles [6]. However, the mechanism of DOMS is still far from entirely understood. Several theories, such as lactic acid, muscle spasm, inflammation, connective tissue damage, muscle damage, enzyme efflux and most recently neuronal damage and the Piezo2 microdamage theory attempt to explain the mechanism of DOMS ^[7][8][9][7,8,9], but no single theory or factor has answered the question entirely. Notably, DOMS is different from pain experienced during or immediately after exercise [10], as DOMS could be induced without muscle damage [11], similarly to vibration [12]. Furthermore, exercise-induced muscle damage could exist without DOMS, and even earlier findings seemed to conclude that DOMS inducement is independent of inflammation [13] or that it is not essential for DOMS [5].

2. Neural Microdamage of DOMS

It is important to note certain milestones that led to the neuronal microdamage theory of DOMS. Weerakkody et al. was the first to demonstrate the contribution of muscle spindle-derived proprioceptors in DOMS ^[12][14][12,14]. Later, Torres et al. showed that reduced proprioception in DOMS is muscle spindle-induced; however, they theorized that the eccentric exercise-derived damage is from intrafusal muscle fibers [2]. Proske and Gandevia reported that eccentric exercise indeed damages proprioception [15]. A significant milestone was the novel concept, put forward by Bennet et al., of neuronal terminal lesions, called terminal arbor degeneration (TAD), that could be learned from paclitaxel chemotherapy [16]. Kouzaki et al. demonstrated that eccentric exercise-induced muscle damage increased the M-wave latency and implicated reversible motoneuronal damage, but excluded muscle spindle origin [17]. However, it is known from Vincent et al. that platinum analogue chemotherapy causes complex Type Ia proprioceptive impairment, and this lesion could evolve in an acute and chronic fashion as well ^[18][19][18,19]. Furthermore, Alvarez et al. and Bullinger et al. showed permanent central synaptic disconnection of proprioceptors from motoneurons after nerve injury, and this phenomenon is associated with the loss of vesicular glutamate transporter (VGLUT) 1/Ia synapses on motoneurons ^[20][21][20,21]. Sonkodi et al. suggested that the muscle spindle origin of increased M-wave latency on motoneurons after eccentric exercise-induced muscle damage should not be excluded [22] and even proposed that the transiently impaired muscle spindle-derived proprioceptors and the resultant synaptic disconnection on motoneurons are responsible for the increased M-wave latency in DOMS, in line with the findings of Vincent, Alvarez and Bullinger et al. ^{[19][20][21][22]}[19,20,21,22]. The significant findings of Murase et al. that bradykinin and nerve growth factor (NGF) are pivotal in DOMS inducement were supplemented by Kubo et al. and Ota et al., who showed that C-fiber and transient receptor potential vanilloid (TRPV) 1/TRPV4 also play an essential role in the DOMS mechanism ^[23][24][25][23,24,25]. The earlier research work of Bewick et al. was fulfilled by Than et al. on glutamatergic autoexcitation in muscle spindles, substantially helping the understanding of glutamate-based signaling of static phase firing sensory encoding ^[26][27][26,27]. Sonkodi et al. hypothesized that DOMS is an acute stress response-induced acute compression proprioceptive axonopathy derived from muscle spindles [8]. Sonkodi et al. even suggested that the primary microinjury of the dichotomous noncontact injury mechanism of DOMS is a transient Piezo2 channelopathy at the peripheral terminal of Type Ia proprioceptors [9]. Notably, Piezo2 channels are shown to be the principal mechanotransduction channels for proprioception [28], involved in vibration sensing [29], and bone-derived osteocalcin was demonstrated to mediate an acute stress response [30]. Sonkodi also theorized that the result of acute Type Ia proprioceptive terminal microinjury is represented in the delayed latency of the medium latency response (MLR) of the affected stretch reflex due to a switch of monosynaptic Type Ia static phase firing sensory encoding to polysynaptic Type II static phase firing encoding ^[9][31][32][9,31,32]. Notably, MLR is commonly viewed by scientists as dominantly Type II afferent mediated ^{[33][34][35][36][37][38][39][40][41]}[33,34,35,36,37,38,39,40,41]. Indeed, Sonkodi et al. demonstrated that delayed latency of MLR is a consequence of unaccustomed fatiguing exercise that induced DOMS [32]. Most recently, Borghi et al. showed that intense acute DOMS-inducing swimming causes spinal cord neuroinflammation [42], as was hypothesized by Sonkodi et al. ^[8][31][8,31].

3. Noncontact Injury Mechanism of DOMS

Proprioception, described as our sixth sense by Sir Charles Bell in 1830 ^[43][82], is a mysterious, diagnostically challenging system referring to the sense of the positions and actions of the extremities and providing our postural control effortlessly ^[44][46]. It is often the no man’s land in clinical medicine because it rather pertains to the peripheral nervous system; however, it has profound preprogrammed pathways in the central nervous system. Morgan et al. ^[45][83] and Hody et al. ^[46][84] identified earlier the dichotomous injury mechanism of DOMS. The biphasic acute proprioceptive compression axonopathy or neuronal noncontact injury mechanism was first described through the new noncontact neuroinflammation theory of DOMS ^[8][9][8,9].

3.1. Primary Injury Phase or Piezo2 Channelopathy

The proprioceptive primary microinjury is hypothesized to occur when ASR-induced energy depletion at the primary afferent’s peripheral terminal prevails, and as a result, the mechano-energetically dysfunctional mitochondrial supply could impair the glutamate vesicular release system in addition to the autologous microinjury of the proprioceptive Piezo2 ion channels ^[8][9][31][8,9,31]. The realization that these types of lesions could behave in an analogous way that is experienced as a side effect of axonopathy-causing paclitaxel and platinum-analogue chemotherapy could enhance our understanding because it evolves in a dose-limiting, threshold-driven manner and is not associated with classical Wallerian axonal degeneration ^{[8][9][16][19][31]}[8,9,16,19,31]; however, it could disrupt the monosynaptic static phase firing sensory encoding on motoneurons ^[19][20][21][19,20,21].

3.2. Secondary Injury Phase or Axonopathy

As a result of the primary noncontact microdamage, a secondary injury phase could succeed in the form of harsher tissue damage ^[8][47][8,53]. This more pervasive tissue damage in DOMS is due to the primary impairment of proprioceptive protection, as itwe could be eexperienced in other noncontact injuries ^{[8][31][47][48][49]}[8,31,53,57,85]. The pivotal involvement of other sensory neurons, such as nociceptive neurons [24], and other ion channels, e.g., TRPV1 and TRPV4 [25], could occur in this stage of the noncontact pathophysiology of DOMS [8]. Notably, nociceptive C-fiber involvement in mechanical hyperalgesia of DOMS could be correspondingly secondary but pivotal because these neurons contribute to the slow temporal summation of pain ^[8][24][50][8,24,86] (see Table 1). Recent research is evolving in support of the noncontact neuroinflammation theory of DOMS ^[8][9][42][8,9,42]. Muscle spindle-derived proprioceptive large fiber involvement has been demonstrated in DOMS, as suggested by the acute proprioceptive compression axonopathy theory ^[8][12][8,12]. Moreover, eccentric exercise-derived damage, implicated within the muscle spindle, reduces proprioception immediately after DOMS-inducing exercise and not with a delayed onset [2]. The current researcheuthor interprets these findings to mean that muscle spindle-derived proprioceptive terminal Piezo2 microinjury is the critical gateway to pathophysiology in DOMS, but mechanical hyperalgesia cannot evolve without the secondary, harsher tissue injury and resultant C-fiber contribution [9] (see Table 1). A secondary preprogrammed compensatory pathway could come into play as a consequence of the primary noncontact Piezo2 microdamage to enhance postural control, enhance shock attenuation and support the body against gravity at the injured segmental level ^[9][31][47][9,31,53]. The basis for the switch to this secondary compensatory pathway is the aforementioned silent exchange of static phase firing sensory encoding and the heightened inducement of NaPICs or, even more likely, NMDA PICs ^[52][72] on motoneurons, resulting in compensatory exaggerated contractions and reduced range of motion [31]. Notably, the lost function of Piezo channels indeed induces exaggerated mechanoreflexes and contractions in compromised Aδ or Type III sensory endings ^[53][87]. However, the impaired Type III fiber-associated compensatory exaggerated contractions and reduced range of motion induced from extracellular matrix or muscles are suggested to evolve only in the secondary injury phase of DOMS, as hypothesized by the acute autologous proprioceptive compression axonopathy theory ^[8][31][8,31]. Accordingly, the C-fiber pain pathway is interlinked with Type III fibers during the secondary phase of the DOMS mechanism [8] due to muscle or extracellular matrix damage. Indeed, chemical and enzymatic destruction of the extracellular matrix impairs Piezo2 mechanogating putatively ^[44][54][46,88], and the deep fascia seems to be more sensitive to noxious chemical stimuli than skeletal muscles in DOMS ^[55][89]. However, it cannot be the primary damage because the impairment of proprioception could be experienced immediately after DOMS-inducing exercise, and certainly, the secondary damage is not muscle spindle derived [2] (see Table 2). It is important to note that muscle spindles cannot be viewed as entirely isolated anatomical structures, but rather as a continuum with extrafusal space. Most intrafusal muscle fibers extend beyond the ends of the muscle spindle capsule and are clearly attached to the surrounding connective tissue ^[56][90]. Furthermore, it is speculative but likely that numerous transverse connections exist from the intracellular space across the cell membrane to the extracellular matrix and between extrafusal fibers, distal muscle spindle capsules and terminating intrafusal fibers ^[57][91]. The current researcheuthor suggests that a good candidate for this extracellular matrix-based trans-spindle cross-talk between Type Ia Piezo2 and Type III Piezo2 channels could be the Piezo1 ion channels. Correspondingly, cellular traction forces produce spatially restricted Piezo1-mediated Ca²⁺ flickers on a noncontact or no external mechanical force basis ^[58][80]. Notably, the pores of Piezo channels are nonselective to cations; however, Piezo currents slightly favor Ca^2+ [44][59][46,58], and this preference could have special importance in Piezo2-based sensing of these Ca²⁺ flickers, especially in a “leaky” microinjured state. Indeed, recent findings put forward a force-from-filament or tether model based on Piezo channels tethered to the actin cytoskeleton, and the perturbation of this tethering could impair Piezo-transduced feedback ^[60][92]. The current researcheuthor interprets these findings to mean that enzymatically impaired Piezo2 mechanotransduction of Type III fibers exert an additional, but this time delayed onset presynaptic inhibitory, effect on the already mechano-energetically microinjured muscle spindle-derived proprioceptive central terminals. This mechanism could be analogous to that presented by Fernández-Trillo et al., where the eye blinks of Piezo2 knockout mice were lower when von Frey filaments were applied to corneal Aδ sensory fibers compared to those of wild-type mice ^[61][93]. Notably, the short latency blink reflex is induced by the stretching of extraocular muscles and elicited in the extraocular muscle spindles ^[62][94]. In summary, polymodal Aδ fibers that contain Piezo2 ion channels could be good candidates for cross-talk with nociceptive C-fibers in the secondary phase of the DOMS mechanism, as suggested earlier ^[8][31][8,31]. Notably, Borghi et al. found upregulated c-Fos at locations where proprioceptive primary afferents enter the spinal dorsal horn [42], further supporting the hypothesis and observation that the primary microdamage is rather at the Piezo2-containing Type Ia terminals ^[9][32][9,32], and the secondary damage is more related to Type III/C-fibers [8].

3.3. Tertiary or Longitudinal Injury Phase

Emerging evidence supports that DOMS has a tertiary or longitudinal injury phase, lasting up to a year, in the form of the repeated bout effect (RBE). RBE is a reduced DOMS effect of the initial one when the same exercise bout is repeated ^{[9][31][47][63]}[9,31,53,95]. Accordingly, the initial bout of severe DOMS-inducing unaccustomed exercise comprising eccentric contractions could evoke reduced DOMS symptoms for at least 6 months, but this adaptation is lost within 9 to 12 months ^[64][96]. This “adaptation” phenomenon is attributed to neural, connective tissue and cellular mechanisms [5]. The current researcheuthor proposes that the critical gateway to this “adaptation” process is the microinjury of the Piezo2 ion channels and is primarily orchestrated by the proposed intimate Piezo current-based cross-talk between Piezo2 and Piezo1 channels involving the sensory neurons, connective tissue, or more precisely, the extracellular matrix and peripheral cells. Notably, recent studies are emerging in support of this autologous injury mechanism theory since the aforementioned myosin-II-mediated cellular traction forces could induce spatially restricted Piezo1-transduced Ca²⁺ flickers even in the absence of external mechanical force ^[58][80]. Furthermore, the suggested proprioceptive Piezo2 channelopathy-induced NMDA receptor activation opens several memory pathways at the central terminal on the spinal dorsal horn, such as pain memory, inflammation, working and episodic memory [31]. This NMDA receptor activation process is in addition to the one that is suggested to be the result of ASR inducing osteocalcin in the CNS [31]. Notably, osteocalcin exerts a strong influence on spatial learning and memory ^[65][97].