Water‐soluble dietary fiber alleviates cancer‐induced muscle wasting through changes in gut microenvironment in mice

Abstract Cancer cachexia and the associated skeletal muscle wasting are considered poor prognostic factors, although effective treatment has not yet been established. Recent studies have indicated that the pathogenesis of skeletal muscle loss may involve dysbiosis of the gut microbiota and the accompanying chronic inflammation or altered metabolism. In this study, we evaluated the possible effects of modifying the gut microenvironment with partially hydrolyzed guar gum (PHGG), a soluble dietary fiber, on cancer‐related muscle wasting and its mechanism using a colon‐26 murine cachexia model. Compared with a fiber‐free (FF) diet, PHGG contained fiber‐rich (FR) diet–attenuated skeletal muscle loss in cachectic mice by suppressing the elevation of the major muscle‐specific ubiquitin ligases Atrogin‐1 and MuRF1, as well as the autophagy markers LC3 and Bnip3. Although tight‐junction markers were partially reduced in both FR and FF diet–fed cachectic mice, the abundance of Bifidobacterium, Akkermansia, and unclassified S24‐7 family increased by FR diet, contributing to the retention of the colonic mucus layer. The reinforcement of the gut barrier function resulted in the controlled entry of pathogens into the host system and reduced circulating levels of lipopolysaccharide‐binding protein (LBP) and IL‐6, which in turn led to the suppression of proteolysis by downregulating the ubiquitin‐proteasome system and autophagy pathway. These results suggest that dietary fiber may have the potential to alleviate skeletal muscle loss in cancer cachexia, providing new insights for developing effective strategies in the future.


| INTRODUC TI ON
Cancer cachexia is a complex metabolic disorder characterized by severe wasting of skeletal muscle and adipose tissue, loss of physical function, and anorexia. Sarcopenia, defined as reductions in the muscle mass, strength, and physical condition, is also an important disorder in cancer patients. Severe cachexia and sarcopenia are associated with high mortality, reduced quality of life, and poor outcomes of anticancer treatments, contributing to at least 20% of cancer deaths. 1 Cachexia is driven by inflammation; however, because of its multifactorial nature, the pathogenesis is still incompletely understood.
In catabolic conditions such as cancer, muscle wasting is induced by the continuous imbalance between protein synthesis and degradation. 2 Increased protein degradation is considered to be the main cause, supported by the fact that skeletal muscle mass cannot be restored by amino acid supplementation alone. 3 Previous reports demonstrated that atrogin-1/muscle atrophy F-box (MAFbx) and muscle RING finger 1 (MuRF1), the two ubiquitin ligases of the ubiquitin-proteasome pathway, play a major role in muscle proteolysis as part of cancer-related muscle wasting. 4,5 Recently, the autophagy/lysosomal pathway has also been regarded as important. 3,6 These pathways are mediated by procachectic cytokines such as IL-6, TNFα, and IFNγ and other regulators. 7 In particular, IL-6 is regarded as the key mediator for the progression of cachexia, with its levels associated with patients' longevity. 8 Recently, researchers have focused on the crosstalk between gut microbiota and the pathophysiology of age-related sarcopenia, the so-called "gut-muscle axis." 9,10 Moreover, gut barrier dysfunction has been demonstrated in several murine cachectic models, suggesting the relevance of gut microbiota to muscle metabolism during cancer. 11,12 Although a few studies on cachexia focusing on interventions to the gut environment have been reported, [13][14][15] limited evidence is available.
Dietary fiber, a nondigestible nutrient including polysaccharides, is drawing attention. The gut microbiome ferments dietary polysaccharides into short-chain fatty acids (SCFAs), absorbable substances that act as mediators of favorable metabolic activities in the host. 16 Recent studies reported that SCFAs partly mediate skeletal muscle metabolism and function in aging or obese mice; 17,18 however, their role in cancer-associated muscle metabolism remains unclear.
Partially hydrolyzed guar gum (PHGG), a water-soluble dietary fiber derived from guar seeds, is known for its high fermentability and stimulation of SCFA production. 19 Therefore, PHGG is regarded as one of the most beneficial dietary fibers and considered a functional food.
Based on these studies, we hypothesized that PHGG intake would exert a systemic anti-inflammatory effect through changes in the intestinal environment and barrier function, thereby suppressing muscle atrophy in cancer cachexia. We used a well-established colon-26 (C26) cachexia model and evaluated the characteristics and mechanisms.

| Cell culture and treatments
The colon-26 murine adenocarcinoma cell line was obtained from

| Animals and diet
Male BALB/c mice (6 weeks old, Shimizu Laboratory Supplies) were caged individually in a room with a temperature of 18-24°C, humidity of 40%-70%, and a 12-hour light/dark cycle. Two types of diet were prepared: a fiber-rich (FR) diet (cellulose-removed AIN-93G diet < Research Diet, Nihon Clea, Tokyo, Japan > supplemented with 5% PHGG), and fiber-free (FF) diet (cellulose-removed AIN-93G diet). The detailed components of diets are described in Table S1.

| Experimental design
The C26 cachexia model was designed by injecting 5 × 10 6 cells in 0.2 ml of saline subcutaneously into the right flank of mice anesthetized by isoflurane inhalation, as previously reported. 22 After one week of acclimatization, they were randomly assigned to four groups according to their body weight and fed the FR or FF diet for 2 weeks. Next, on day 0, either C26 cells or a saline solution was injected subcutaneously. Thus, the experiment comprised the following four groups: C26-FR, C26-FF groups (C26 cell-transplanted mice fed FR or FF diet) and CT-FR, CT-FF groups (sham-injected mice fed FR or FF diet). Two independent experiments in which mice were K E Y W O R D S cachexia, dietary fiber, gut microbiome, prebiotics, sarcopenia sacrificed on days 9 and 21 were performed to estimate the mechanisms underlying early and late phases of cachexia. The body weight, food intake, and tumor growth were recorded every other day. The tumor size was calculated as a × b 2 /2, with a the maximal and b the minimal diameter of the tumor.
For the first experiment, on day 21, mice (n = 7 or 8 per group) were adequately anesthetized with isoflurane gas and blood was collected by heart puncture, centrifuged 3 at 500 × g for 5 minutes, and the plasma was stored at −80°C. The gastrocnemius muscle, fecal-containing cecum, colon, terminal ileum, and tumor were harvested, weighed, and stored at −80°C. The colon was immediately preserved in Carnoy's solution for 3 hours and then transferred to fresh methanol until further use. For the second experiment, we performed the same protocols as above (n = 8 mice in the C26-FR, C26-FF groups and n = 5 mice in the CT-FR, CT-FF groups), sacrificing mice and collecting the sections as described above on day 9.

| Tissue and cell mRNA analyses
Total RNA was extracted and then reverse-transcribed into complementary DNA. Polymerase chain reaction (PCR) was conducted by the 7300 Real-Time PCR system (Applied Biosystems). Primer sequences are detailed in Table S2.

| RNA microarray analysis
Isolated RNA from the gastrocnemius muscle was subjected to microarray analysis using Affymetrix GeneChip Mouse Gene 1.0 ST Array (Thermo Fisher Scientific) according to the manufacturer's instructions (n = 3 per group sacrificed on day 21). Comparison of the mean differences was made between the two groups selected (C26-FF vs CT-FF group and C26-FR vs C26-FF group) using a twotailed parametric t test. We defined differentially expressed genes (DEGs) as those with an absolute fold change ≥1.5 with p < 0.05.
Next, for function annotation of DEGs, enrichment analysis based on the Gene Ontology (GO) category database was applied using the DAVID Bioinformatics database. 23 Categories of biological process (BP), cellular component (CC), and molecular function (MF) were analyzed, and we excluded the drawn categories if the annotated genes were less than five, showed no significant difference, or the fold change was <1.5. 23

| FISH and immunofluorescence staining with histological analysis
Fluorescence in situ hybridization and immunofluorescence staining for Muc2 and EUB was performed using colonic sections that were fixed and paraffinized. Sections were deparaffinized and dehydrated before being mounted with the eubacterial probe EUB-338-Cy5 (Sigma-Aldrich), which targets nearly all bacteria, for hybridization, and primary antibody staining was performed by covering the sections with 1:100 diluted rabbit anti-Muc2 antibodies (Santa Cruz Biotechnology) at 4°C overnight. For secondary antibody staining, the sections were covered by a 1:1,000 dilution of goat antirabbit IgG antibody conjugated with Alexa Fluor 594 (Abcam) and Phalloidin-iFluor 488 (Abcam) and incubated at room temperature for 2 hours. These were visualized using an Olympus FV10i confocal laser scanning microscope (Olympus). The colonic mucus layer thickness was measured using the method previously described. 24 C2C12 myotubes were cultured in differential medium with either LPS (1 µg/ml) or LPS and SCFA cocktail (10 mM) for 48 hours.
These cells were fixed with 4% paraformaldehyde, incubated with 1:250 diluted mouse anti-myosin heavy chain (MyHC) antibody (MAB4470, R&D Systems) and 1:500 diluted Alexa Fluor 488-conjugated secondary antibody(Abcam), and visualized by fluorescence microscopy. The diameter of myotubes in ten random fields was measured using Image J software version 1.53 (National Institutes of Health), as previously reported. 21

| Analysis of cecal bacteria
We extracted the genomic DNA of the bacteria from the collected cecal contents as reported previously. 25 A two-step PCR method was carried out using purified DNA samples to ac- The sequence data were analyzed using QIIME2 2020.8. The detailed quality statistics for each sample are summarized in Table S3.
Assignment of taxonomy was achieved through the Sklearn classifier algorithm against the Greengenes database 13_8 (99% OTUs full-length sequences). In this study, singletons and ASVs assigned to mitochondria and chloroplasts were removed. The phylogenetic tree was generated by SATé-enabled phylogenetic placement (SEPP). 26 Metrics regarding alpha (Chao1 and Shannon indices) and beta diversities (UniFrac distances) were calculated by QIIME2 by setting the sampling depth at 5000 reads. Significance of β-diversity was assessed by permutational multivariate analysis of variance (PERMANOVA) using QIIME2.

| Analysis of fecal SCFAs
Approximately 30 mg of cecal contents were suspended in 0.5 ml of 14% perchloric acid for protein elimination, centrifuged at 10,000 × g for 5 minutes at 4°C, and the supernatant solution was screened through a cellulose acetate membrane filter of 0.45 µm pore size.
The amount of organic acids was measured by ion-exclusion highperformance liquid chromatography, as described previously. 27

| Statistical analyses
Results are presented as the mean ± SEM, except for data from microbiome analysis where α-diversity and relative abundance of representative phyla and genera are presented by box-and-whisker plots with minimal and maximal values. The significance of differences was assessed using Student's t test or the Mann-Whitney U test when comparing two groups. When comparing multiple groups, one-way ANOVA followed by post hoc Tukey multiple comparison test or the Kruskal-Wallis test followed by the Steel-Dwass post hoc test was used. All statistical tests were two-sided, and p < 0.05 was set as the level of significance. Statistical analyses were performed using JMP ® pro 15 (SAS Institute Inc.) unless otherwise stated.

| Fiber-rich diet attenuates body weight and muscle volume in C26 murine model
The overall protocol of the experiments is shown in Figure 1A. First, we compared the body weight and muscle volume between mice fed FR and FF diets. The body weight started to decrease gradually from days 3 to 7 in the C26 groups, whereas it increased steadily in the CT groups, with a slight difference in food intake ( Figure 1B,C). Body weight was maintained in the latter period in the C26-FR group, while it continuously decreased in the C26-FF group, consequently leading to large but nonsignificant differences among groups in body and carcass weights when sacrificed on day 21 ( Figure 1D,E).
Importantly, C26 transplantation induced a significant loss of gastrocnemius muscle weight, but this was greater in the C26-FF group, resulting in a significant difference in the C26-FR and C26-FF groups ( Figure 1F). In addition, dietary treatment did not affect the tumor size in the C26 groups, or body and gastrocnemius weights in the CT groups ( Figure 1D,E,G).

| Fiber-rich diet reduces markers of skeletal muscle atrophy in cancer cachexia
To assess the mechanisms of the anticachectic effect of the FR diet, we initially measured markers related to proteolysis of gas-

| Fiber deficiency reduces colonic mucus layer and promotes bacterial invasion of host
We examined the thickness of the colonic mucus layer by immunofluorescence staining of Muc2 and also validated the location of bacteria using the FISH technique. On day 9, a significantly thicker mucus layer was found in both the C26-FR and CT-FR groups ( Figure 3I,K). Mucus layer measurements on day 21 revealed a thinner mucus layer in the C26 groups compared with the CT groups, although it was preserved in FR diet-fed mice and severely depleted in FF diet-fed mice, and this was accompanied by bacterial invasion into the colonic epithelium ( Figure 3J,L). Lipopolysaccharidebinding protein, an acute phase response protein mainly produced by the liver, is regarded as an indicator of the antigen load of bacteria. 29 Notably, circulating LBP levels showed a tendency to be  Figure   S3).

| Fiber-rich diet alters composition of gut microbiome in cachexia models
To elucidate the gut environment underlying this model, we ana-  Table S8.

| Analysis of fecal short-chain fatty acids and their effects on LPS-induced C2C12 myotube atrophy
The next step was to examine SCFAs in cecal contents among the four groups on day 9. Between the CT groups, the principal metabolites, acetate, propionate, and succinate, showed significantly higher levels in mice fed the FR diet. Although the C26 groups showed relatively lower SCFA levels, the same trend as described above was found, resulting in overarching preservation in the C26-FR groups ( Figure 5A-E). The results for other SCFAs are presented in Figure   S4. To explore the effects of SCFAs on skeletal muscle mass, we used a well-established in vitro myotube culture model. Exposure of C2C12 myotubes to LPS induced a significant decrease in the diameter, but this was reversed on administering a cocktail of SCFAs ( Figure 5F,G). The increased expression of Atrogin-1 by LPS was significantly reduced when LPS-stimulated myotubes were subjected to SCFA treatment ( Figure 5H). On the other hand, MuRF1 did not show any difference ( Figure 5I). We also demonstrated that the SCFA cocktail upregulated the expression of PGC1α and Tfam but not CPT1b in C2C12 myotubes with or without LPS exposure ( Figure 5J-L).

| DISCUSS ION
Herein, we demonstrated that a PHGG-containing FR diet ameliorated muscle wasting under cancer-cachectic conditions by reinforcing the gut barrier function through alterations in the gut microbiota and subsequent anti-inflammatory effects. Consistent with previous reports, 11,12 we showed that the gut barrier function of cachectic mice was disrupted. However, a significantly thicker colonic mucus layer was retained in mice receiving PHGG supplementation, compensating for this impairment. To our best knowledge, this is the first report to show that dietary fiber preserves the colonic mucus layer during cancer cachexia. The reinforcement of the mucus layer consequently reduced the translocation of pathogens into the host system and led to lower circulating levels of LBP and IL-6, which contributed to the suppression of proteolysis by downregulating the core factors, ubiquitin-proteasome system, and autophagy pathway.
Clinical approaches targeting the anabolic pathway have emerged sporadically, although there are very few challenges to control hypercatabolism in cancer cachexia. Therefore, it is noteworthy that suppression of the catabolic pathway was sufficient to achieve significant improvements in skeletal muscle mass and body weight. Microarray analysis offered us wider insights, showing that the genes upregulated by FR diet intake, which were depressed in response to C26, were related to the extracellular matrix and sarcolemma with enriched processes of muscle contraction and collagen fibril organization, and these all play essential roles in muscle structural integrity and function. 30 This suggests that a FR diet could improve not only the volume but also functional aspects of skeletal muscle.
The role of gut microbiota in cancer cachexia is receiving increasing attention, with evidence that altered intestinal homeostasis with increased gut permeability was present in conjunction with increased proinflammatory cytokines and cachectic conditions in mouse models. [11][12][13] These studies support the idea that a decrease in and Akkermansia by PHGG supplementation in C26 mice, which may explain the decreased diversity in microbial abundance and evenness in this group. S24-7, which was a predominant member of the gut microbiota in the C26-FR group, has been reported to possess fermentative pathways for producing succinate, acetate, and propionate. 37 We assume that this bacterium largely contributed to the preservation of fecal SCFAs. Of note, Bifidobacterium has been reported to restore the reduced intestinal mucus layer caused by consumption of a low-fiber Western diet. 32 Moreover, Akkermansia muciniphila has also been reported regarding its importance in improving the gut barrier function and reducing endotoxemia. 38 However, the detailed mechanisms underlying the effects of these bacteria on the host are still unclear, especially in the area of cancer cachexia; thus, further research is warranted.
The host's immunological system responds to the tumor by producing proinflammatory cytokines including IL-1β, IL-6, TNFα, and INFγ. 3,7 These cytokines are essential for driving inflammatory reactions needed to overcome the tumor burden, while also playing a critical role in muscle wasting by directly targeting muscle tissue.
IL-6 is widely known as a key mediator of cancer cachexia, as these inhibitors were reported to markedly prevent cancer-associated muscle wasting in previous studies. 39,40 In our study, IL-6, but not Previous studies indicated that SCFAs, major microbial metabolites, can strengthen the gut barrier by promoting mucus production and secretion. 41,42 Furthermore, these metabolites may be partially involved in the metabolism and function of skeletal muscle through modification of mitochondrial activity. 20,43 In light of these reports, while the role of SCFAs in the context of cancer cachexia has yet to be elucidated, beneficial effects are expected. Our data revealed that fecal SCFA levels were relatively low in cachectic mice, although the reduction was mitigated when PHGG supplementation was given. Indeed, this could be another explanation for the retention of the mucus layer. Markers related to mitochondria were not fully affected in our mouse model, although we did examine the possible protective effect of SCFAs on muscle atrophy by in vitro methods.
However, it should be noted as a limitation that the concentrations of LPS and SCFAs set in the in vitro experiment were higher than that in the circulating system. 11,44 Taken together, our findings suggest that SCFAs may be involved in preventing cancer-induced muscle wasting, although further investigation is warranted.
In conclusion, our observations suggest that dietary fiber alters the gut microbiota and restores the gut barrier function in cachectic mice and that these modifications induce a systemic antiinflammatory effect, thereby attenuating muscle wasting. Among the complex mechanisms of cancer-induced muscle atrophy, crosstalk between the gut environment and skeletal muscle may be at least partially involved. Although further research is required in this underexplored area, the novel insights obtained in this study may indicate a potential nutritional strategy against cancer cachexia.

ACK N OWLED G M ENTS
This work was partly supported by MAFF Commissioned project study on "Project for the realization of foods and dietary habits to extend healthy life expectancy" (Grant Number JPJ009842). This work was also supported by JSPS KAKENHI (Grant Number JP 20K07787).

D I SCLOS U R E
Yuji Naito received scholarship funds from Taiyo Kagaku Co., Ltd. and from EA Pharma Co. Ltd.; a collaboration research fund from Taiyo Kagaku Co., Ltd.; and lecture fees from Mylan EPD Co., Takeda Pharma.
Co. Ltd., and Miyarisan Pharma. Co. Ltd. The present research was partly supported by these funds. Neither the funding agency nor any outside organization has participated in the study design or has any compet-

DATA AVA I L A B I L I T Y S TAT E M E N T
The raw microarray data generated in this study are available in