Male C57BL/6J mice (8-week old at the time of cisplatin administration) were purchased from Jackson Laboratory Japan, Inc. (Kanagawa, Japan). All animals were housed separately under environmentally controlled ambient temperature (23°C ± 3°C), humidity (50% ± 20%), and lighting (12 h light/dark cycle) conditions and had ad libitum access to water and food.

This study was approved by the Experimental Animal Ethics Committee of TSUMURA & CO. (Tokyo, Japan; permit no. 18-043, 19-013, 23-005), and all experimental procedures were performed in accordance with the ARRIVE guidelines and the guidelines of the National Institutes of Health for the Care and Use of Laboratory Animals.

Cisplatin was purchased from Fujifilm Wako (Osaka, Japan) and dissolved in saline at a concentration of 3 mg/10 mL. HET was manufactured by TSUMURA & CO. (Tokyo, Japan) as the intermediate product (extract powder, Lot No. 2240041020, 2200041020, 2210041010, 2230041010) of ‘TSUMURA Hochuekkito Extract Granules for Ethical Use’ as shown in their website (https://www.tsumura.co.jp/english/kampo/, and https://www.tsumura.co.jp/english/ir/library/integrated-report/). The extract quality was standardized based on good manufacturing practices, as defined by the Ministry of Health, Labour, and Welfare of Japan and are in accordance with the Japanese Pharmacopoeia (JP) (https://www.pmda.go.jp/english/rs-sb-std/standards-development/jp/0029.html, https://www.mhlw.go.jp/content/11120000/000904450.pdf). Briefly, it was manufactured by spray-drying the hot-water extract of the following 10 crude drugs: 4.0 parts JP Astragalus root [Astragalus membranaceus Bunge, or Astragalus mongholicus Bunge (Leguminosae), radix], 4.0 parts JP Atractylodes lancea rhizome [Atractylodes lancea De Candolle, or Atractylodes schinensis Koidzumi (Asteraceae), rhizome], 4.0 parts JP Ginseng [Panax ginseng C. A. Meyer (Panax schinseng Nees) (Araliaceae), radix], 3.0 parts JP Japanese angelica root [Angelica acutiloba Kitagawa, or Angelica acutiloba Kitagawa var. sugiyamae Hikino (Umbelliferae), radix], 2.0 parts JP Bupleurum root [Bupleurum falcatum Linné (Umbelliferae), radix], 2.0 parts JP Jujube [Ziziphus jujuba Miller var. inermis Rehder (Rhamnaceae), fructus], 2.0 parts JP Citrus unshiu peel [Citrus unshiu Marcowicz, or Citrus reticulata Blanco (Rutaceae), pericarpium], 1.5 parts JP Glycyrrhiza [Glycyrrhiza uralensis Fischer, or Glycyrrhiza glabra Linné (Leguminosae), radix], 1.0 parts JP Cimicifuga rhizome [Cimicifuga simplex Turczaninow, Cimicifuga dahurica Maximowicz, Cimicifuga foetida Linné, or Cimicifuga heracleifolia Komarov (Ranunculaceae), rhizome], and 0.5 parts JP ginger [Zingiber officinale Roscoe (Zingiberaceae), rhizome]. The plants were identified by their external morphology and marker metabolites as per the JP and company standards. A three-dimensional high-performance liquid chromatogram of HET is presented in Supplementary Figure S1. HET was administered to mice by mixing it into standard chow (MF; Oriental Yeast Co., Ltd., Tokyo, Japan) at a concentration of 1.5% or 3%, which are approximately 2 and 4 times the human equivalent dose, respectively. The doses were determined according to previous reports (Nahata et al., 2021; Nahata et al., 2022; Okugawa et al., 2024).

Total RNA was extracted from muscle samples using the RNeasy Mini Kit (74106; Qiagen, Valencia, CA, United States) and reverse transcribed to cDNA using the TaqMan High-Capacity cDNA Reverse Transcription Kit (4374967; Applied Biosystems, Waltham, MA, United States). DNA was extracted from muscle samples using the QIAamp DNA Mini Kit (51304; Qiagen, Valencia, CA, United States). mRNA expression and mitochondrial DNA (mtDNA) copy number was analyzed using quantitative real-time polymerase chain reaction (4444557; TaqMan Fast Advanced Master Mix, Applied Biosystems, Waltham, MA, United States) using TaqMan gene-specific primers (Thermo Fisher Scientific, Waltham, MA, United States) on a Quant Studio 7 Flex Real-Time PCR System (Applied Biosystems, Waltham, MA, United States). mRNA expression was normalized to that of glyceraldehyde 3-phosphate dehydrogenase (Gapdh), and the mtDNA copy number was normalized using the nuclear gene ribonuclease P RNA component H1 (Rpph1). The TaqMan gene-specific primers used for the experiments are provided in the Supplementary File.

The miRNeasy Mini Kit (217004; Qiagen, Valencia, CA, United States) was used for miRNA extraction from red muscle samples. miRNA sequence library preparation, sequencing, mapping, and read counting by RNA-seq by Expectation-Maximization (RSEM) differential expression and analysis by DESeq2 were conducted by DNAFORM (Yokohama, Kanagawa, Japan). Fold changes and statistical differences (padj; adjusted p-value) in miRNA expression levels between the two groups (vehicle versus cisplatin, cisplatin versus cisplatin + 3% HET) and baseMeans are shown in the Supplementary Material. Metascape (http://metascape.org) was used for enrichment analysis of Gene Ontology (GO) biological processes (Zhou et al., 2019). miEAA (https://ccb-compute2.cs.uni-saarland.de/mieaa/) (Aparicio-Puerta et al., 2023) was used to search for target genes.

The gastrocnemius and plantaris muscle samples were immunohistochemically stained for the slow muscle-specific myosin heavy chain 7 (MYH7) at Biopathology Institute Co., Ltd. (Oita, Japan). Briefly, muscle samples were fixed in neutral buffer solution with 10% formalin, paraffin-embedded and sectioned, and retrieved by heating for 10 min at 120°C. The sections were incubated with specific antibodies (Supplementary File) and visualized after color development using 3,3′-diaminobenzidine tetrahydrochloride solution. The cross-sectional area ratios and slow-twitch fiber counts were assessed using ImageJ. Hematoxylin and eosin staining of adjacent sections was performed at GenoStaff Co., Ltd. (Tokyo, Japan).

Muscle tissues were homogenized in RIPA buffer (Fujifilm Wako, Osaka, Japan) containing 1/100 volume of a protease inhibitor cocktail (P8340; Sigma-Aldrich, St. Louis, MO, United States) and PhosSTOP (04 906 845 001; Roche, Basel, Switzerland) and then centrifuged twice at 10,000 g at 4°C for 10 min to obtain the tissue proteins in the supernatants. Phospho- AMP-activated protein kinase (AMPK)/total-AMPK and sirtuin 1 (SIRT1) levels were assessed using Western blotting. Briefly, the samples (10 μg of total protein per lane) were subjected to SDS-PAGE, transferred onto PVDF membranes, blocked using SuperBlock (TBS) Blocking Buffer (37535; Thermo Fisher Scientific, Waltham, MA, United States) for 1 h at room temperature, and then incubated overnight at 4°C with specific primary antibodies and then secondary antibodies at room temperature for 1 h (Supplementary File). Blots were visualized using Pierce ECL Plus Western blotting Substrate (32132; GE Healthcare, Buckinghamshire, United Kingdom) and the Typhoon 9410 imager (GE Healthcare, Buckinghamshire, United Kingdom).

Statistical analyses were performed using GraphPad Prism version 8 (GraphPad Software, San Diego, CA, United States), R software version 4.3.1 (The R Foundation for Statistical Computing, Vienna, Austria; https://www.R-project.org/), and Stat Light 2000 (Yukms, Tokyo, Japan).

Line graphs with the mean ± standard error values were generated using GraphPad Prism, and two-way ANOVA followed by Tukey’s multiple comparisons post hoc test was used for statistical analysis. Box plots were generated in R using Tukey’s method. For the statistical analysis of differences between two groups, the Wilcoxon rank-sum test in R was used when outliers existed (any values that were more than 1.5 times the interquartile range above the third quartile or below the first quartile); otherwise, Student’s t-test or the Aspin–Welch t-test in Stat Light 2000 were used. For statistical analysis between multiple groups, Tukey’s multiple comparison test with a robust regression model was used in R when outliers existed (Yohai, 1987; Rabbi et al., 2022); otherwise, the Tukey–Kramer or Steel–Dwass tests in Stat Light 2000 were used. The miRNA data are displayed as MA-plots created in R. For significant differences, p values were calculated using the Wald method and adjusted using the Benjamini–Hochberg method. The R packages (ggbreak, ggplot2, ggrepel, MASS, multcomp, openxlsx, readxl, robust base, tidyr, and writexl) were obtained from the Comprehensive R Archive Network (CRAN). Further details are provided as Supplementary Material. Statistical significance was set at p < 0.05.

Following previous reports (Sakai et al., 2014), we established a mouse model of muscle atrophy by administering cisplatin for four consecutive days and observed the resulting induction of muscle atrophy and the subsequent recovery process. Maximum body weight decrease was observed on Day7 (p = 0.0003108) but recovered by Day14 to the same levels as those in the vehicle group (Figure 1B). Food intake was decreased on Day4 (p = 0.000666) but recovered to the same level or higher than that in the vehicle group after the cessation of cisplatin administration (Figure 1C). Hindlimb skeletal muscle (gastrocnemius and plantaris muscle) weights decreased persistently with cisplatin administration until Day14 (p = 0.01041) (Figure 1D), and skeletal muscle weight per gram body weight also tended to decrease on Day14 (p = 0.06496) (Figure 1E). Grip strength was significantly lower on Day7 (p = 0.006993) and Day14 (p = 0.04507) (Figure 1F). Thus, although the decrease in food intake was transient and the decrease in body weight had recovered on Day14, muscle atrophy and decreased muscle performance persisted.

Cisplatin-induced muscle atrophy involves both increased muscle proteolysis and decreased muscle regeneration and protein synthesis (Conte et al., 2020; Klassen et al., 2023). However, the recovery process from cisplatin-induced muscle atrophy has not been investigated in detail. Therefore, in addition to the period from Day4 and Day7, which is the acute phase of cisplatin-induced muscle atrophy, we also examined gene expression related to muscle proteolysis, synthesis, and regeneration on Day14, which can be considered as the recovery phase.

An analysis of mRNA expression of the ubiquitin ligase family members Atrogin-1 and MuRF-1, which are directly involved in muscle atrophy, revealed that Atrogin-1 mRNA expression was significantly upregulated by cisplatin administration on Day4 (p = 0.0040) and Day7 (p = 0.0001554), but was significantly reduced on Day14 (p = 0.01476) (Figure 2A). Likewise, MuRF-1 mRNA expression also tended to be increased by cisplatin administration on Day4 (p = 0.0896) and Day7 (p = 0.1304) but recovered to the same level as that in the vehicle group on Day14 (p = 0.1444) (Figure 2B). We also examined the mRNA expression of Myostatin, a myokine that causes muscle atrophy, and found that it tended to increase on Day7 (p = 0.0533) but had decreased to the same level as that in the vehicle group on Day14 (p = 0.7000) (Figure 2C).

Next, we investigated the pathways involved in muscle regeneration and differentiation. The mRNA expression of Insulin-like growth factor 1 (Igf-1), which is known to control both anabolic and catabolic pathways in skeletal muscle, was significantly decreased by cisplatin administration on Day4 (p = 0.00996) and Day7 (p = 0.0001554), but recovered to the same level as that in the vehicle group on Day14 (p = 1) (Figure 2D). Among the myogenic regulatory factors that regulate skeletal muscle regeneration and differentiation, Myogenic differentiation 1 (MyoD) mRNA expression was continuously decreased after cisplatin administration (p = 0.0006216 on Day14). However, we did not observe any significant change in Myogenin mRNA expression (p = 0.7168) on Day14 (Figures 2E,F).

As cisplatin-induced myotube atrophy is reported to be related to mitochondrial damage (Matsumoto et al., 2022; Klassen et al., 2023), we assessed the expression of mitochondria-related genes in this study. Expression of Peroxisome proliferator-activated receptor gamma coactivator 1α (Pgc-1α), the master regulator of mitochondrial biogenesis, was significantly decreased on Day7 (p = 0.002953) in the cisplatin group but had recovered to the levels comparable to that in the vehicle group on Day14 (p = 0.3823) (Figure 2G). The expression of Mitochondrial transcription factor A (Tfam) mRNA was unchanged by cisplatin administration on Day7 and Day14 (p = 0.8785, 0.6742) (Figure 2H). The relative mtDNA copy number was lower in the cisplatin group than in the vehicle group on Day4 and Day7, but not on Day14 (p = 0.0013, 0.08298, and 0.7030, respectively) (Figure 2I). Superoxide dismutase 2 (Sod2) mRNA localized in mitochondria decreased on Day7 (p = 0.0357), but its level on Day14 was comparable to that in the vehicle group (p = 0.6454) (Figure 2J). In the early stages of cisplatin-induced muscle atrophy (acute phase: Day4 and Day7), increased expression of proteolysis-related genes and decreased expression of protein synthesis-and mitochondrial biogenesis-related genes was expected based on previous reports. However, almost all changes had recovered by Day14, and only MyoD mRNA expression continued to decrease. This suggests that the persistence of muscle atrophy until Day14 may be related to the continued decline in MyoD mRNA expression.

Next, we investigated the effects of HET in the mouse model of cisplatin-induced muscle atrophy. There was no significant effect of HET on the decrease in body weight and food intake in either the acute (Day0–7) or recovery (Day7–14) phases (Figures 3B–G). We then investigated the effects of HET on grip strength and locomotor activity, which reflect skeletal muscle performance. The average grip strength in the cisplatin group was slightly lower than that in the vehicle group on Day13 (p = 0.755) and tended to be restored by 3% HET administration (p = 0.438), although the differences were not significant (Figure 3H). The cisplatin-induced decrease in locomotor activity peaked on Day4 and gradually recovered thereafter (Figure 3I). The cumulative locomotor activity for 13 days was significantly decreased in the cisplatin group compared to that in the vehicle group (p = 0.0293) and was significantly restored in the 3% HET group compared to that in the cisplatin group (p = 0.0225) (Figure 3J). The effects of HET on locomotor activity were also examined by dividing the study period into the acute (Day0–7) and recovery (Day8–13) phases. HET did not affect locomotor activity in the acute phase, and a significant increase in locomotor activity was observed only in the recovery phase (p = 0.00828) (Figures 3K,L). These results indicate that HET improved skeletal muscle performance without affecting body weight and food intake, suggesting that HET may not directly affect the process of induction of cisplatin-induced muscle atrophy, but rather enhances the recovery process from muscle atrophy.

As skeletal muscle consists of different types of muscle fibers—fast- and slow-twitch fibers—we evaluated the effects of HET on gastrocnemius muscle weight after dividing it into two regions: the fast-twitch fiber-dominant region (white muscle) and the slow-twitch fiber-containing region (red muscle). The total weight of the gastrocnemius muscle (without the plantaris muscle) was significantly lower in the cisplatin group than in the vehicle group (p < 0.001) and was significantly restored in the 3% HET group (p = 0.0165) on Day14 (Figure 4A). The cisplatin induced-decrease in muscle weight was observed only in white muscle (p = <1e-04), whereas there was a trend toward a decrease in the red muscle that was not statistically significant (p = 0.2768); in contrast, HET increased muscle weight only in the red muscle region (p = 0.0129 by 3% HET) but not in the white muscle region (p = 0.745) (Figures 4B,C). Whole gastrocnemius muscle, red muscle, and white muscle weight per gram body weight showed the same tendency as the absolute muscle weight (Figures 4D–F). However, no increase in muscle weight caused by HET was observed on Day7 (Supplementary Figure S2). Moreover, in cisplatin-untreated mice, HET did not increase the weight of the red muscle region or locomotor activity (Supplementary Figure S3). These findings suggest that HET may restore skeletal muscle performance by acting specifically on slow-twitch fibers during the recovery from cisplatin-induced muscle atrophy, rather than during the acute phase.

Since HET was suggested to affect specifically slow-twitch fibers, we then investigated its effects on mRNA expression of each myosin-heavy chain (MyHC) isoform: one slow-twitch fiber type (Myh7/MyHCI) and three fast-twitch fiber types (Myh2/MyHCIIa, Myh1/MyHCIIx, and Myh4/MyHCIIb) (Wang and Pessin, 2013) (Figures 5A–D). Cisplatin treatment did not result in any prominent changes in mRNA expression in the red muscle region; Myh7 and Myh4 mRNA expression tended to be slightly increased in the 3% HET group than in the cisplatin group, but the differences were not statistically significant (p = 0.6111 and 0.47154, respectively) (Figures 5A,D). In the white muscle region, neither cisplatin nor HET affected the expression levels of Myh7/2/1/4 mRNAs (Supplementary Figure S4).

We also assessed the protein levels of MyHCI (MYH7) using immunohistochemical staining (Figure 6A, lower panel). Typical images of hematoxylin and eosin staining are shown in the upper panel of Figure 6A. The MYH7-positive cross-sectional area was unchanged by cisplatin administration but significantly increased by 3% HET administration (p = 0.0075) (Figure 6B). In contrast, the MYH7-negative cross-sectional area was not changed by cisplatin or 3% HET administration (Figure 6C). Thus, although the increase in total expression of Myh7 mRNA in the red muscle region was not significant, HET appeared to increase the cross-sectional area of slow-twitch muscle fibers. These results suggest that HET enhances the recovery process from cisplatin-induced muscle damage mainly by increasing the mass of slow-twitch fibers in the mouse hindlimb.

Next, we investigated why HET specifically affected slow-twitch fibers in cisplatin-induced muscle atrophy. First, we examined the effects of HET on Atrogin-1, Myostatin, and Igf-1, which were altered by cisplatin in the acute phase (Day7), but no obvious effects of HET were observed on any of these in the red muscle regions (p = 0.6144, 0.1025, and 0.414) (Figures 7A–C). We also investigated the effect of HET on MyoD mRNA levels, since a sustained decrease in MyoD mRNA expression was assumed to be involved in the continued atrophy of the white muscle region in this model (Figures 2E, 4B,E). MyoD mRNA expression was also significantly decreased by cisplatin administration in white muscle (p = 0.0343), but there was no change in the red muscle region (p = 0.236) on Day14 (Supplementary Figure S5). Furthermore, HET did not affect the expression levels of MyoD in either region (Supplementary Figure S5). The absence of any observed change in expression in the red muscle regions after HET administration (p = 0.995) indicates that MyoD expression is not involved in the effect of HET increasing red muscle weight.

Myogenin, myocyte enhancer factor-2c (Mef2c), Nuclear factor of activated T cells 1 (Nfatc1), and Class Ⅱ histone deacetylase 4 (HDAC4) are known to be involved in the regulation of slow-twitch fiber differentiation and muscle hypertrophy (Hughes et al., 1993; Potthoff et al., 2007; Alexander et al., 2010; Ehlers et al., 2014; Anderson et al., 2015). Therefore, we examined the effects of HET on the expression of their mRNAs. Contrary to our expectations, we did not observe any corresponding significant changes in the red muscle regions after administration of 3% HET (p = 0.9998, 0.995, 1.000 and 0.878) (Supplementary Figure S6). Moreover, activation of the AMPK/SIRT1/PGC-1α signaling pathway has been reported as a mechanism by which polyphenols and arginine induce fiber type transformation from type II to type I (Chen et al., 2018; Wen et al., 2020; Xue et al., 2021); thus, we also examined the effects of HET on phospho-AMPK and SIRT1 protein levels, but did not observe any clear effect on them on Day14 (Supplementary Figure S7). As mitochondrial dysfunction has been reported to be involved in cisplatin-induced muscle atrophy (Sirago et al., 2017; Seo et al., 2021; Matsumoto et al., 2022), we then examined the effects of HET on the expression of mitochondria-related genes. The expressions of Pgc-1α, Tfam, Sirt1, Sirt3, Nuclear respiratory factor 1 (Nrf1), and Sod2 mRNA were also not significantly different between groups and remained unchanged by HET administration (p = 0.962, 0.900, 0.812, 0.9948, 0.975, and 0.9614 between the cisplatin and cisplatin + 3% HET groups) (Supplementary Figure S8). Administration of 3% HET significantly increased (p = 0.01609) the decrease in mtDNA level of cytochrome b (Cytb) (p = 0.00226) in the red muscle, but not of the copy numbers for NADH dehydrogenase subunit 1 (Nd1) and cytochrome c oxidase subunit 1 (Cox1) (p = 0.947093 and 0.999047) (Figures 8A–C).

As mentioned earlier, we did not find any definite HET-evoked changes in the expression levels of key regulators of skeletal muscle fiber-type, i.e., MyoD, Myogenin, Myostatin, MEF2, NFAT, HDAC4, SIRT1, and PGC-1α and attempts to identify possible mechanisms by which HET can specifically affect slow-twitch fibers based on the findings of previous reports did not yield any definitive answers. Therefore, we focused on miRNAs known to play a role in fine-tuning gene expression and translation in eukaryotes. In addition to myogenic regulatory factors, growth factors, hormones, and myokines, some miRNAs are not only reportedly involved in skeletal muscle development and muscle-related diseases, but also in defining muscle fiber types. Thus, we performed miRNA-seq analysis using the red muscle region of the gastrocnemius muscle, where the effect of HET was most prominent.

From the obtained DESeq2 data (Supplementary File), we extracted miRNAs of note according to the following threshold levels: padj <0.05, absolute Log2FoldChange >0.1, and baseMean >100. Based on a comparison of the vehicle versus cisplatin groups (Figure 9), we extracted nine upregulated miRNAs and nine downregulated miRNAs by cisplatin administration (a total of 18 miRNAs from 1003 miRNAs listed, Table 1). Gene Ontology (GO) analysis was performed using Metascape (https://metascape.org/) to identify the predicted functions of the miRNAs and the potential effects of cisplatin. As a result (Supplementary Figure S8), the miRNAs upregulated by cisplatin showed significant associations with long-term synaptic potentiation (GO:0060291), cellular response to leukemia inhibitory factor (GO:1990830), sensory perception of sound (GO:0007605), and cellular response to phenylalanine (GO:0071234). In contrast, the miRNAs downregulated by cisplatin were significantly associated with cellular response to leukemia inhibitory factor (GO:1990830), response to oxygen levels (GO:0070482), cellular response to lipid (GO:0071396), and cellular response to organic cyclic compound (GO:0071407).

We then compared our results to those of a previous study (Jang et al., 2021). We used miEAA to predict the target genes of the extracted miRNAs in our study and investigated whether the expression of those target genes was suppressed by cisplatin administration based on RNA-seq data in that study (Jang et al., 2021), in which row gene expression data (GSE147613) for Cisplatin and Normal were obtained from the Gene Expression Omnibus dataset of the National Center for Biotechnology Information (https://www.ncbi.nlm.nih.gov/geo) (Jang et al., 2021).

Of the 28 target genes predicted from the list of miRNAs with increased levels in our study (Supplementary File), most were also downregulated after cisplatin administration in a previous report (Jang et al., 2021). In contrast, miEAA predicted only Il-6 mRNA as the target gene of the miRNA that had reduced levels after cisplatin administration in our study, but its level was increased in the previous study and remained unchanged in our study (Supplementary Figure S10). This was probably because some of miRNAs with increased levels in our study were also predicted to target IL-6. Among the miRNAs increased by cisplatin, those that had large baseMean values were miR-29a/b and miR-34a. Thus, changes in their expression levels were estimated to have large effects.

Regarding the difference in miRNA expression between the cisplatin and cisplatin + 3% HET groups, three miRNAs (miR-1a-1, miR-1a-2, and miR-206, which facilitate satellite cell proliferation and differentiation) were extracted according to our threshold levels of the 1023 miRNAs (Figure 10; Table 2) (Chen et al., 2010; Salant et al., 2020). They were significantly upregulated by HET administration and showed the highest baseMean values, indicating large effects. As miR-206 has been reported to be involved in determining the slow muscle fiber type (Bjorkman et al., 2020), it may have played a role in the thickening of slow fiber muscle in the cisplatin + HET groups. By searching the miEAA database, we identified 70 experimentally validated target genes for miR-206 or miR-1 (Supplementary File). Among them, only HDAC4 has been reported to influence the induction of slow-twitch muscle fibers by suppressing the transcriptional activity of MEF2 (Potthoff et al., 2007). Therefore, HDAC4 was considered a plausible target gene of miR-206 or miR-1, which was upregulated by HET administration. However, in this study, no change in Hdac4 mRNA levels was observed following HET administration (Supplementary Figure S6D).

Multiple lines of evidence indicate that cisplatin causes skeletal muscle atrophy by both activating muscle protein degradation and downregulating protein synthesis and muscle regeneration systems (Conte et al., 2020; Klassen et al., 2023). However, the recovery process is yet to be studied precisely. Since recovery of body and muscle weight after cisplatin administration takes approximately 4–6 weeks (Garcia et al., 2013; Hojman et al., 2014), we conducted a detailed pathophysiological evaluation of the recovery phase in addition to the acute phase after cisplatin administration.

Our results indicate that cisplatin regulates specific genes to induce muscle atrophy by enhancing the protein degradation system, decrease protein synthesis, disrupt mitochondrial biogenesis and reduced mtDNA level in the acute phase (Day4 to Day7), and these findings are in good agreement with previous reports (Sirago et al., 2017; Seo et al., 2021; Matsumoto et al., 2022). Although these changes were found to be almost completely restored on Day14, the decrease in skeletal muscle weight and grip strength persisted. One possible explanation for this apparent discrepancy may be the persistent suppression of MyoD expression. MyoD is essential not only for myogenesis, but also for skeletal muscle regeneration, and several reports have shown that MyoD is downregulated during the acute phase after cisplatin administration (Chen et al., 2015; Wu et al., 2019; Jang et al., 2021). Furthermore, previous studies have indicated that MyoD is expressed at higher levels in fast-twitch fibers than in slow-twitch fibers (Hughes et al., 1997) and that its overexpression in muscle results in a slow-to-fast fiber type conversion (Ekmark et al., 2007), which is also in line with our results. Notably, our study showed, for the first time, that the decrease in MyoD expression persisted until Day14 after cisplatin administration, and this may be one of the reasons why the recovery of skeletal muscle weight and grip strength was delayed in cisplatin-induced muscle atrophy.

Our analysis showed that body weight and food intake were unaffected by HET administration in both the acute and recovery phases in our model of cisplatin-induced muscle atrophy. When we investigated the effects of HET on factors that reflect muscle performance, we found that the decreased average grip strength was partially recovered by HET administration, although the difference was not statistically significant. Moreover, the cisplatin-induced decrease in locomotor activity peaked on Day4 and gradually recovered thereafter. Although HET did not affect locomotor activities in the acute phase (Day0–7), it specifically increased them in the recovery phase (Day8–13), indicating that HET does not affect the induction of muscle atrophy, but rather enhances the recovery process. Since the induction of skeletal muscle hypertrophy has been reported to increase muscle performance (Dyle et al., 2014; Yoo et al., 2021), we assumed that restoration of gastrocnemius muscle weight by HET administration led to the restoration of decreased locomotor activity in this study. On the other hand, voluntary exercise prevented cisplatin-induced muscle wasting in mice and locomotive wheel running activity increased the weight of the gastrocnemius muscle in disuse muscle atrophy model mice (Hojman et al., 2014; Brooks et al., 2018) and endurance exercise strengthened slow-twitch muscles (Crupi et al., 2018). Thus, there is a possibility that HET might have increased locomotor activity and then restored skeletal muscle atrophy since it reportedly improves physical activity in elderly patients (Watanabe and Maeda, 2022) and aged mice (Nahata et al., 2021; Nahata et al., 2022). However, further investigations are required to clarify the precise relationships involved. Moreover, since the anxiolytic effect of cisplatin can reduce motor activity-related parameters (Vukovic et al., 2019) and HET improved avoidance behavior in a depression model (Tohda and Mingmalairak, 2013) and anxiety-like behavior in lipopolysaccharide-treated mice (Ushio et al., 2022), it is possible that the improvement of anxiety by HET led to an additional increase in locomotor activity.

The gastrocnemius muscle consists of both slow- and fast-twitch fibers and can be divided into two different regions: the “red muscle” region, characterized by the presence of “slow” isoforms of contractile proteins and a high oxidative enzyme content, and the “white muscle” region, which has a predominance of glycolytic enzymes and “fast” isoforms of contractile proteins (Burkholder et al., 1994; Glancy and Balaban, 2011; Dimauro et al., 2019; Lloyd et al., 2023). Thus, we evaluated the effect of HET by dividing the gastrocnemius muscle into red and white regions.

We found that, although the decrease in the weight of white muscle was unaffected, the weight of red muscle (containing slow-twitch fibers) was significantly increased by HET administration, indicating that HET specifically affected slow-twitch fibers in the red muscle region. To confirm this, we investigated the effects of HET on subtype-specific gene and protein expression patterns. Although there was no definite change in the mRNA expression levels of myosin heavy chain subtypes, including slow fiber-specific Myh7, immunohistochemical analyses showed that MYH7-positive slow-twitch fibers were significantly increased following HET treatment.

Grip strength is primarily considered to reflect the function of fast-twitch muscles (Song et al., 2021), whereas locomotor activity is more indicative of endurance, which reflects the function of slow-twitch muscles (Crupi et al., 2018). The insufficient recovery of grip strength even 2 weeks after cisplatin administration in this study is believed to be due to the continued atrophy of fast-twitch muscles, probably caused by the sustained decrease in MyoD levels. Cisplatin has been reported to cause disruption of neuromuscular junctions (Huot et al., 2022) which has been implicated in the pathogenesis of fast-twitch muscle-dominant atrophy observed in several neuromuscular disorders such as Duchenne muscular dystrophy, amyotrophic lateral sclerosis and sarcopenia (Lloyd et al., 2023). Thus, white muscle region may be more susceptible to cisplatin and resulted in the weight decrease. In contrast, locomotor activity naturally recovered after Day8, and HET further increased this activity which was consistent with the increase in slow-twitch muscle area. Thus, the changes in skeletal muscle performance reflected the changes in muscle fiber types.

We then investigated the mechanism by which HET specifically exerts its action on slow-twitch fibers but found no effect of HET on genes that were altered by cisplatin in the acute phase, such as muscle atrophy- and mitochondria-related genes. Additionally, HET did not restore the decreased MyoD mRNA expression, which is likely involved in persistent white muscle atrophy. Furthermore, HET administration did not lead to any significant changes in the gene expressions, that are involved in the differentiation and hypertrophy of slow-twitch fibers.

With regard to other models of skeletal muscle atrophy, HET has been reported to improve muscle atrophy by suppressing Atrogin-1 and activating AMPK in a disuse-induced muscle atrophy model (Yakabe et al., 2022), by suppressing IL-6 production in colon 26 adenocarcinoma (Yae et al., 2012), and inhibiting inflammation and oxidative stress in an amyotrophic lateral sclerosis mouse model (Cai and Yang, 2019). However, we did not detect any changes in the expression of these factors following HET administration in this study. Thus, different mechanisms were assumed to be involved in the effects of HET in this study.

All previous studies on the effects of drugs on cisplatin-induced muscle atrophy have focused on their preventive and therapeutic effects in the acute phase of muscle atrophy. In contrast, the newly discovered effects of HET were limited to the recovery phase in this study. Furthermore, HET had no effect on normal mice and its effect was observed only after cisplatin induced muscle injury. This suggests that HET specifically acts during the recovery process rather than nonspecifically affecting the muscles. Since activation, proliferation, and differentiation of skeletal muscle satellite cells, which are important for muscle regeneration, are reportedly affected by fibro/adipogenic progenitors, inflammatory/immune cells, etc., (Koike et al., 2022; Jiang et al., 2024), their involvement in the action mechanism of HET is also highly intriguing and requires future investigation.

We also attempted to determine the roles of the major regulators of skeletal muscle fiber-type with respect to the slow-twitch fiber-specific effects of HET. However, contrary to our expectations, we did not observe any changes in this regard, and focused on miRNAs. Recently, several miRNAs, including muscle-specific miRNAs (myomiRs), have been shown to significantly affect skeletal muscle differentiation, regeneration, atrophy, and functioning (Horak et al., 2016). Furthermore, some myomiRs (i.e., miR-1a-1/2, miR-133a/b, miR-206, miR-208b, miR-486, and miR-499) are known to be involved in skeletal muscle fiber-type regulation (van Rooij et al., 2009; Zhang et al., 2014; Bjorkman et al., 2020). Thus, we performed a comprehensive miRNA-sequence analysis to identify the miRNAs that may be related to the mechanism of action of HET.

Cisplatin treatment increased the levels of nine miRNAs and the predicted target genes were in good agreement with the changes in mRNA levels in previous study by Jang et al. (2021). This suggests that some of these miRNAs altered the expression levels of several mRNAs in skeletal muscle and may have thus contributed to the induction and/or regression of skeletal muscle atrophy.

Among the nine upregulated miRNAs that were significantly upregulated by cisplatin, miR-29a, miR-29b, and miR-34a appear to be particularly important, because their expression levels were much higher than those of the other miRNAs. miR-29a and miR-29b belong to the miR-29 family and have been reported to either suppress skeletal muscle atrophy by inhibiting MuRF-1, Ying Yang-1, and PTEN (Wang et al., 2011; Wang et al., 2020; Liu et al., 2021) or promote it by inhibiting IGF-1 and PI3K (p85α) (Li et al., 2017), respectively. Considering that the observed expression levels of miR-29a were much higher than that of miR-29b in our study, it is possible that the miR-29 family facilitated the overall recovery from muscle atrophy. Furthermore, both miR-29 and miR-34a significantly increase with age in the skeletal muscle and are senescence-associated miRNAs (Hu et al., 2014; Fulzele et al., 2019; Alfonzo et al., 2022), implying that their increase might have induced senescence-like conditions in the skeletal muscle after cisplatin administration.

HET administration significantly upregulated the expression of three miRNAs: miR-1a-1, miR-1a-2, and miR-206. Both miR-1 and miR-206 are increased after muscle damage and enhance the differentiation of skeletal muscle satellite cells by targeting Pax7 (Chen et al., 2010) and the induction of slow-twitch muscle by targeting HDAC4 (Potthoff et al., 2007), and were highly expressed in the red muscle region and increased by HET administration in this study; thus, we assumed that both effects were involved in increasing the weight of the red muscle region. MiR-1 is the most abundant miRNA in the heart tissue, accounting for 40% of the total miRNAs (Rao et al., 2009) and has been shown to play a critical role in heart development (Wei et al., 2014; Luo et al., 2018). In contrast, miR-206 is involved in skeletal muscle satellite cell proliferation and differentiation (Chen et al., 2010), regeneration of neuromuscular synapses by targeting HDAC4 (Williams et al., 2009) and repairment of neuromuscular junctions after nerve injury (Valdez et al., 2014). Moreover, miR-206 is highly expressed in slow-twitch soleus muscles (Siracusa et al., 2018), and its deficiency delays differentiation and formation of myotube (Salant et al., 2020). Moreover, as the cross-sectional area of type I (slow-twitch) myofibers in the soleus muscle of male miR-206-knockout mice (Bjorkman et al., 2020) was lower, miR-206 may have played a role in thickening the slow fiber muscles in the HET group in this study, indicating its involvement in the slow-twitch fiber-specific action mechanism of HET. Because miR-206 is important for mitochondrial and muscle function (Przanowska et al., 2020), it may also be involved in the increase in locomotor activity.

Based on the above findings, it is plausible that HET administration increases slow-twitch muscle fibers through the upregulation of miR-1 and/or miR-206. To further elucidate the underlying mechanism, we searched for target genes of miR-1 and/or miR-206 using the miEAA database. Although we identified 70 experimentally validated target genes for miR-1/miR-206, only HDAC4 has been reported to influence the regulation of slow-twitch muscle fibers by suppressing the transcriptional activity of MEF2 (Potthoff et al., 2007). Therefore, HDAC4 was considered a plausible target gene of miR-1/miR-206, which is upregulated by HET administration. However, in this study, no changes in Hdac4 mRNA levels were observed following HET administration. This suggests that the inhibition of Hdac4 translation rather than transcription may be involved, that target genes other than Hdac4 may play a role in the regulation of slow muscle fibers by HET, or that subtle changes in multiple target genes may collectively contribute to the induction of slow-twitch muscle differentiation/ regeneration.

This study has several limitations. First, although we found that HET treatment increased slow-twitch muscle fibers in the red region of gastrocnemius muscle and significantly increased miR-1/miR-206 in the same area, we could not establish a direct causal relationship between the two. To directly demonstrate this experimentally, it would be necessary to knock down miR-1/miR-206 using miRNA sponges or anti-microRNAs (Winkle et al., 2021) or to use muscle-specific miR-206-deficient mice (Williams et al., 2009; Bjorkman et al., 2020). Second, this study was unable to identify the target genes of miR-1 and miR-206 involved in slow muscle induction. To address this, a comprehensive analysis of both miRNAs and mRNAs will be necessary in the future to elucidate their mutual relationship. Third, although we demonstrated the positive effects of HET overall, we could not specifically determine which of the 10 crude drugs or their metabolites were involved in the restorative effects of HET on cisplatin induced-muscle atrophy in our model. Several reports have indicated the possibility of some of the metabolites of HET improving muscle atrophy and locomotor activities, specifically astragaloside Ⅳ from Astragalus root (Jiang et al., 2020), compound K (Kim et al., 2022) and Ginsenoside Rd (Wijaya et al., 2022) from ginseng, Saikosaponin A and D from Bupleurum root (Huang M. et al., 2023), ursolic acid from jujube (Tao et al., 2023), and nobiletin (Nohara et al., 2019; Wang et al., 2023) and naringin (Lv et al., 2023) from Citrus unshiu peel. In the future, it will be important to clarify which of these metabolites contribute to the effect of HET on muscle atrophy through in vitro studies using muscle cells, such as C2C12 and other approaches. Lastly, miRNA expression was investigated using only the red muscle region, although its weight was not decreased by cisplatin on Day14. Thus, further investigation of the changes in miRNA expression in white muscle will be needed in future studies to better understand the effects of cisplatin-induced muscle atrophy.