Effects of photobiomodulation therapy on regulation of myogenic regulatory factor mRNA expression in vivo: A systematic review

Non‐invasive promotion of myogenic regulatory factors (MRFs), through photobiomodulation therapy (PBMT), may be a viable method of facilitating skeletal muscle regeneration post‐injury, given the importance of MRF in skeletal muscle regeneration. The aim of this systematic review was to collate current evidence, identifying key themes and changes in expression of MRF in in vivo models. Web of Science, PubMed, Scopus and Cochrane databases were systematically searched and identified 1459 studies, of which 10 met the inclusion criteria. Myogenic determination factor was most consistently regulated in response to PBMT treatment, and the expression of remaining MRFs was heterogenous. All studies exhibited a high risk of bias, primarily due to lack of blinding in PBMT application and MRF analysis. Our review suggests that the current evidence base for MRF expression from PBMT is highly variable. Future research should focus on developing a robust methodology for determining the effect of laser therapy on MRF expression, as well as long‐term assessment of skeletal muscle regeneration.

. To facilitate the healing and regeneration of these damaged tissues, electrotherapeutic modalities are often employed by rehabilitation therapists to assist in this endeavour [2]. Photobiomodulation therapy (PBMT) is one treatment solution available, which is non-invasive and has shown positive results in facilitating skeletal muscle recovery [3]. Following damage, skeletal muscle goes through a highly controlled and tightly regulated process to regenerate damaged muscle fibres. The inflammatory cascade activated in response to injury, results in the activation and propagation of myogenic stem cells, termed satellite cells (SCs) [4]. From these SCs, myogenic progenitor cells (MPCs) are formed [5]. Myogenic regulatory factors (MRFs) are essential in the control and regulation of MPCs commitment to the myogenic lineage and also the processes of determination and differentiation [6]. The result is the differentiation of myoblasts that will regenerate the damaged tissue, ultimately with the aim of restoring contractile function [7] as shown in Figure 1. The MRFs consist of myogenic determination factor 1 (MyoD), myogenic factor 5 (Myf5), myogenin and MRF4, which are all able to induce myoblastic traits within non-myogenic cell lineages and fibroblasts [9]. Each of the MRFs play a key temporal role: while SCs are activated and proliferating, they express MyoD and Myf5, and then later myogenin as the terminal differentiation factor [10]. The expression of MRF4 remains high at the end of regeneration and also during SC return to quiescence [5]. Once the regeneration of the tissue is complete, the progeny of the SCs return to quiescence to maintain the SC pool, and myostatin is expressed, inhibiting excessive myoblast proliferation and maintaining homeostasis within the tissue [11,12]. In some cases of significant injury, disease or mutations within these genes, MRFs may not be sufficiently expressed; therefore, muscle regeneration is severely limited or prevented altogether [13]. Furthermore, limited or no expression of MRFs can lead to potentially excessive scar formation and limited tissue recovery [14].
To promote skeletal muscle regeneration, accelerate the healing process and minimize excessive scar tissues formation, PBMT has been indicated as a viable noninvasive intervention [3]. Given the fundamental role MRFs play in the regeneration of skeletal muscle, it is necessary to characterize their expression in response to PBMT. Such evidence could be used to optimize treatments in terms of dose, timing and duration.
Cytochrome c oxidase, which resides within the mitochondria, is theorized to be the primary mechanism by which PBMT laser medium interacts with the targeted tissues [15]. This chromophore is sensitive to wavelengths within the red to infrared spectrum (λ660-1100 nm) [16]. When exposed to laser light at the appropriate wavelength, the chromophore's respiratory chain activity is promoted, enabling an increase in the cellular energy currency of adenosine triphosphate (ATP), facilitating an increased propagation and differentiation of myogenic cells [17,18]. Although there is evidence for the differentiation of myogenic cells, little attention has been paid to the expression of MRFs in response to different parameters of PBMT. Reports suggest that PBMT has mixed results on healing different tissue types, possibly as a result of incorrect dosing [16]. Insufficient tissue irradiation can lead to no benefit, while excessive irradiation can lead to an inhibitory effect on tissue healing [19].
F I G U R E 1 Timeline of the myogenic regulatory factor expression pathway, adapted from Zanou and Gailly [8]. Upon initial injury, the quiescent satellite cells express myogenic factor 5 (Myf5), proliferating into myoblasts, followed by myogenic determination factor (MyoD) expression. As the myotube begins formation, myogenin is expressed, leading to the fusion of myoblasts. Finally, MRF4 promotes myotube maturation and myofibre organization Likewise, unoptimized dosimetric parameters can lead to poor or insignificant outcomes, with heterogeneous treatment dosages leading to varied results [3,20].
Unoptimized treatment parameters, such energy-perpoint (Joules) or wavelength (nanometres), have been shown to lead to ineffective treatment outcomes, making it essential to establish an effective PBMT dosage [19]. Previous systematic reviews have sought to analyse the effect of PBMT on the entire inflammatory and regenerative process but did not review the effect of PBMT on MRF regulation in greater detail [3,21]. Given that no standardized protocol exists for promoting MRF upregulation, the collation of current evidence will be a key step towards establishing optimal dosimetric parameters to control these important factors necessary for muscle regeneration. Many of the studies in this area are largely animal-based and although these are critical in refining treatment parameters for human trials to demonstrate clinical applicability, the methodological quality in these studies often shows wide variation and applicability in human populations [22]. Through a controlled systematic review of the literature, it will be possible assess the findings of the existing studies in terms of quality and the potential for impact in humans.
To this end, this systematic review will analyse the current literature on MRF expression in response to PBMT within animal populations, intending to identify any optimal treatment parameters that may exist. The aim is that this study will facilitate further work in human-based research by demonstrating the potential efficacy of PBMT in animal models.

| METHODS
The review was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [23].

| Inclusion and exclusion criteria
The inclusion and exclusion criteria for this review are summarized in Table 1 and detailed below.

| Studies
All research articles included for the review were required to fit within the 'Populations, Intervention, Control and Outcome' framework. Articles that did not fit were excluded from this review. The review included experimental studies and randomized control studies. The eligible research had to include an injury control group (ICG), be accessible as full-text and be in English. Review studies and case reports were excluded.

| Populations
Trials that included any animal model were included. Models had to be healthy with no induced comorbidities, for example, diabetes. Any muscle damage needed to be induced by cryoinjury, incision, contusion, or exercise exertion, with studies being excluded if damage was induced by venom or toxins. Damage had to be applied only to skeletal muscle, and studies were omitted if lesions were in other tissue types, for example, skin, bone, cartilage or nerve. Fatigue recovery models, in vitro cell cultures and human studies were excluded, as were any studies that did not declare an animal welfare statement.

| Interventions
Trials that were selected had to include exposure to a form of PBMT within the red to infrared spectrum (λ660-1100 nm), either from a laser or an LED light source. The minimum dosimetric parameters required for inclusion included wavelength (nm), optical power output (mW), beam size, treatment

| Reference searching
Reference lists for all papers that met the study's inclusion criteria were screened to identify any additional studies which were not present during the initial literature search.

| Study selection
Throughout the literature search, all prospective studies were uploaded to EndNote X9.2 to be collated, with duplicates removed. Article titles and abstracts were reviewed for their eligibility. If the abstract was not sufficient to make a judgement or was unclear, the fulltext was analysed to determine the research's eligibility. Each piece of eligible research was screened through full-text analysis and then referred to as 'eligible' or 'excluded'. The search results and reasons for subsequent exclusions are presented in the PRISMA flow diagram ( Figure 2).

| Data extraction
Population size, trauma type, treatment variables, analysis time points, MRF expression assessment methods and results (including P values, where reported) were collated into a Microsoft Excel spreadsheet, which is summarized in Tables 2-4.

| Data synthesis
Given the considerable heterogeneity within extracted data, all results were reported narratively as opposed to statistical meta-analysis.

| Assessment of methodological quality
Studies underwent a risk of bias analysis, employing the 'SYstematic Review Centre for Laboratory animal Experimentation' tool (SYRCLE) to analyse the relevant methodological domains for quality and risk of bias for each study [34]. Every study was manually assessed and curated for each domain based on the information provided by the article, which was subsequently graded. Each of the SYRCLE domains are: 'sequence generation', 'baseline characteristics', 'allocation concealment', 'random housing', 'blinding (performance bias)', 'random outcome assessment', 'blinding (detection bias)', 'incomplete outcome data', 'outcome reporting', 'other forms of bias'. Each study was read and each of the above domains were scrutinized and graded 'unclear', 'low' and 'high' depending on the apparent level of bias and whether sufficient information was available to make a judgement (Table 5).

| RESULTS
The results of this systematic review show a broad array of outcomes from a mixture of treatment parameters. This concurs with previous findings in this field by Alves et al. and Teles et al., who stated that variety in dosage parameters leads to a difference in outcomes [3,21].

| Myogenic determination factor
MyoD was measured by all of the 10 included studies. The frequency of results is summarized in Figure 3A. MyoD was predominately upregulated by PBMT over the assessment periods measured by all 10 studies, demonstrating a consistent and robust effect.

| Myogenic factor 5
Only two studies investigated Myf5 expression in response to PBMT [25,32]. Morais et al. stated a significant increase of Myf5 mRNA expression was present following PBMT treatment at 4.9 J and 660 nm, but reported no inferential statistics to support this statement [25]. Only the treatment Rodrigues et al. [29] Male Wistar rats (n Vatansever et al. [33] Male Wistar rats (n = 56)

| Myogenin
Myogenin expression was measured by 8 of the 10 included studies [24][25][26][27][28][29][30]32]. Myogenin demonstrated heterogenous results from the included studies; overall, there was no definitive trend in the results, and based on the current scientific literature, it was not possible to conclude that PBMT either promoted increased or decreased myogenin expression over the course of muscle regeneration ( Figure 3B).

| Myogenic regulatory factor 4
Only one study examined MRF4, where a significant decrease (P < .05) in mRNA expression was observed when compared to the ICG, at a power output of 75Mw (0.5 J) and wavelength of 904 nm [32]. This is in contrast with the 25 mW (0.5 J) treatment group, which showed no significant change in MRF4 expression (P > .05). These data may suggest a power output-specificity in the response of MRF4.

| Risk of bias
The overall risk of bias was high throughout the studies, as summarized in Table 5. Most studies employed assessor blinding for histopathological assessment but did not blind the investigator for the measurement of MRF expression. This results in a high risk of bias which may potentially affect the overall results.

| Summary of results
Overall, the results of this systematic review exhibit a broad array of outcomes from a mixture of treatment parameters. This concurs with previous findings in this field by Alves et al. and Teles et al., who stated that variety in dosage parameters leads to a notable variation in outcomes [3,21]. MyoD is the primary factor responsible for promoting SC proliferation and myogenic

| Myogenic determination factor
Evidence conflicts regarding the time points of when MyoD is expressed post-injury, although this may be relative to how the skeletal muscle is injured. Contusion injury in rat models often results in change to MyoD expression within 4 to 8 hours post-injury, which declines by 8 days [5]. In contrast, prior research with cryoinjury models has shown that MyoD and Myf5 initially downregulate for 7 days, after which PAX7 downregulation occurs, enabling SC proliferation through expression of the early MRFs [35]. This may explain why MyoD expression changes were analysed more often from 7 days onwards, and why a significant increase was noted by most of the studies analysing expression at this time. Cryoinjury generally exhibits distinct and well-characterized layers of damage, which are known to result in different MRF expression and regulation depending on the extent of tissue damage and should be considered a factor when analysing MRF expression [35]. Given the early expression of MyoD in some cases of skeletal muscle injury, it is unexpected that only one study analysed the effect at day 1 [26].
While not directly comparable to in vivo studies, in vitro research has indicated some concurrence with the in vivo studies included within this review in terms of MyoD expression and PBMT dosages. These studies  [38]. While an in vitro cell line will prove to behave differently to an in vivo environment in response to PBMT, the above studies indicate that a differing dosimetric parameters still lead to an array of outcomes to varying degrees.

| Myogenic factor 5
Only two studies assessed the expression of Myf5, which is surprising given the factor's essential role in the early promotion of SC differentiation through the myogenic lineage [6]. While these two studies exhibited positive upregulation, Morais et al. only reported an increase in expression when strength training was used within the pre-treatment regime, with PBMT potentially enhancing the physiological benefits from exercise [25]. Conversely, Beasi et al. indicated that swimming exercise augments the effect of PBMT in promoting MRF expression, yet both studies show that PBMT mixed with exercise in some form may maximize PBMT in MRF expression enhancement [24]. As the two studies only observed a 14-day time point, early changes in Myf5 levels may have been missed when PBMT alone may have promoted a change without strength training. Trajano et al. only noticed a change in their 75 mW (0.5 J) treatment group at 5 days, suggesting that increased optical power may be more effective in promoting Myf5 expression [32]. Given no other skeletal muscle injury research has assessed Myf5, further research is needed to characterize its expression.

| Myogenin
Myogenin expression varied considerably from PBMT application, with some studies detecting a range of upregulation, downregulation, or no change, with no clear trend. Studies of myogenin expression, have shown that it follows a similar pattern as MyoD during contusion injuries, upregulating within 4 to 8 hours and eventually declining at 8 days [5]. Cryoinjuries can differ, exhibiting myogenin upregulation within the severely damaged areas of tissue but only a weak expression at 7 days [35]. Myogenin expression has only been observed in two PBMT in vitro studies by Zhang et al. and Mesquita-Ferrari et al., who observed myogenin expression at 24 hours and 24 and 48 hours, respectively [37,39]. Zhang et al. discovered a significant decrease at 24 hours in myogenin expression, while Mesquita-Ferrari et al. showed no significant changes at both time points [37,39]. This concurs with this current review's results, showing within the first few days post-injury, myogenin is not notably expressed. As myogenin is responsible for the terminal differentiation and fusion of myoblasts in mature muscle fibres, this may explain why it is not expressed during the early stages of injury [35]. Due to the limited numbers of studies MRF expression in the latter stages of injury at 7, 14 and 21 days, it is difficult to determine whether a trend of upregulation is present towards myotube maturation. Further quantitative studies are necessary to define whether this is the case.

| Myogenic regulatory factor 4
Trajano et al. were the only authors assessing MRF4 response to levels, noting a significant decrease in response to PBMT [32]. This decrease was only observed by the 75 mW (0.5 J) treatment group, suggesting that increased optical power output may be required to alter its expression, but this requires further investigation. As MRF4 has been shown to be expressed at the terminal phase of differentiation, it might be expected that these factor levels are naturally decreased at 5 days [5]. The results for MRF4 are distinctly limited by the restrictive time analysis, justifying the need for future research to analyse the effect of PBMT towards the effect of myotube maturation.

| Result variability
The collective results were highly variable, in part be due to the breadth of dosimetric parameters utilized in the included studies. Considering that the depth of tissue and chromophore activation may be relative to the treatment wavelength, it is perhaps unsurprising that the array of wavelengths used only demonstrates a trend in expression changes for MyoD [16]. Several studies that utilized different dosimetric parameters garnered varied results for MyoD expression; Trajano et al. demonstrated a significant increase in MyoD expression at 25 and 75 mW dosages; however, these both utilized an energy-per-point of 0.5 J at 904 nm, and were therefore comparable [32]. Conversely, the two studies by Rodrigues et al. resulted in varying outcomes for MyoD expression over the treatment, even though the authors employed the same treatment parameters across the two studies of 10 and 50 J/ cm 2 , being 0.4 and 2 J, respectively [29,30].
Freitas and Hamblin discuss how excessive dosage may result in an inhibitory or negative effect on cellular or tissue response, however the studies formerly discussed may suggest that energy output is not the determinant of MRF promotion in vivo, indicating other factors are necessary, especially wavelength [19]. Given that specific wavelengths interact with different molecular components, the wavelengths employed by the Rodrigues et al. studies (660 and 780 nm), alongside the aforementioned in vitro studies, may indicate that wavelength is more detriment in MRF activation than energy dosage alone [29,30]. Considering the distinct lack of homogenous treatment parameters and results, it is hard to discern what dosages are required to facilitate suitable regulation of MRFs throughout the regeneration period.
Animal numbers within each study may also have compounded result variability, especially as no study reported a power calculation method for animal numbers within their methods, thus risking underpowering, or using more animals than necessary. Furthermore, the studies included in this review analysed a wide variety of time points. Some studies examined MRF expression at critical time points of the healing process, enabling a timeline of MRF expression to be established in response to PBMT [26,29,30]. The remaining authors only assessed singular chronological points, specifically 4, 5, 7 and 14 days. This may result in studies missing critical time points for MRF regulation, as the factors are expressed in a highly controlled manner at different time points throughout the healing process.

| Limitations of included studies
Treatment blinding is difficult to accommodate in traumainduced PBMT research, due to the apparent injury presented to the researcher and the need to apply PBMT to an obvious injury site. This may lead to a different approach to handling and treatment of the animals, due to the imposed injury [34]. Another major limitation for all the studies was the level of risk of bias domains; this brings into consideration the methodological quality of the included studies, primarily due to the lack of blinding for MRF measurement and PBMT application. Prior animalbased research has shown to vary in methodological quality, resulting in a mixture of experimental quality, which ultimately poses ramifications on the translation of results to human trials [22]. While the intrinsic nature of PBMT application to a lesion area is problematic around the use of blinding, the inclusion of an injury group with sham treatment (ie, non-therapeutic red light), with the treatment parameters being blinded to the researcher applying the treatment may be a viable option.

| Review limitations
One limitation of this systematic review is the omission of research and results pertaining to inflammatory markers and infiltrate. Inflammatory cells are an important component in regulating muscular regeneration due to the infiltrate releasing growth factors and cytokines; however, this study was expressly focused at determining MRF expression within an in vivo environment due to their significance in regulating myogenesis.
Research implementing toxins as the lesion modality were also omitted from inclusion, as this research was aimed at the effect of laser in orthopaedic and musculoskeletal injuries through physical trauma. Additionally, some animals may show resistance to toxins, resulting in variable MRF expression [40]. Furthermore, given that these toxins can cause both local and systemic myotoxicity and significant disruption to the surrounding tissues, it may not be as reproducible as a contusion or cryoinjury, as featured throughout the included studies [41,42]. Studies were also excluded if the animals had induced comorbidities, such as diabetes, in order to represent healthy population and keep the results comparable.
Finally, studies were omitted from this review if they did not utilize qRT-PCR to analyse MRF expression, meaning three studies were excluded as they relied upon immunohistochemistry analysis. As immunohistochemistry analysis is only semi-quantitative and mRNA and protein expression are known to correlate poorly it would not allow a comparable cross-study assessment.
This study was also limited by the lack of any metaanalysis. Given the limited and often heterogenous results of the included studies, formal meta-analysis was not possible. This may also be considered a limitation of the included studies in this review since inappropriate or insufficient data were reported and any future studies should be encouraged to supply more information regarding results to facilitate meta-analyses and validate their inferential results.

| CONCLUSIONS AND RECOMMENDATIONS
While the results from this systematic review are varied, the collated research has shown some MRF regulatory changes in response to PBMT, especially MyoD, given the increased attention it received within the included studies. The studies showed a mixture of methodological quality, although the lack of blinding may cause concern on the presence of bias. The consensus amongst the included studies was that PBMT promotes lesion area reduction. However, this needs to be confirmed as a response to MRF upregulation and not just attenuation of the inflammatory infiltrate and natural healing occurring. Although these results may be limited clinically, the aim is that this study will encourage further, high-quality research in PBMT for skeletal muscle regeneration, focusing on the use of quantitative measures, being primarily qRT-PCR and protein expression analyses. It is critical that more studies attempt to assess MRF and related factor expression over the whole injury period, not a singular arbitrary time point in the healing process, as MRF expression seemingly changes throughout the period of the injury and important time points may be missed all together. This should be combined with histological tissue analysis alongside investigation into protein-level MRF expression to determine if MRFs are expressed for sufficient periods to result in functional myogenesis.
PBMT research is progressively improving and becoming more prevalent, yet more studies are required for the timely advancement of the field, especially research aiming to define the optimal parameters for stimulating MRF activation to promote muscle regeneration. It is apparent from the results of this study that several different sets of dosimetric parameters can produce comparable results suggesting a wide treatment window that should be relatively simple to define. Developing an understanding of the mechanisms regulating MRF expression would assist in defining the most efficacious dose for animal models, and eventually humans, to promote skeletal muscle regeneration. This may include using advanced in vitro cell and tissue models to explore the mechanisms through which Cytochrome c oxidase activation ultimately leads to a MRF expression.