Febuxostat Reduces Muscle Wasting in Tumor-bearing Mice via Inhibition of Reacrive Oxygen Species Generation

To analyze the effects of the administration of febuxostat, a selective xanthine oxidase (XO) inhibitor, on both in vitro and in vivo models of the waste of skeletal muscles, due to LM8 osteosarcoma cells. C2C12 myotubes were incubated in the conditioned medium of LM8 osteosarcoma cells. At the same time, febuxostat was added at a concentration of 3 µM and 30 µM, and reactive oxygen species(ROS) was measured using 2’,7’-dichlorouorescein diacetate (DCF-DA) at 2 h and 24 h after exposure. Furthermore, an in vivo study was performed on male C3H mice (5 weeks old) that were subcutaneously injected with LM8 osteosarcoma cells on their backs. Febuxostat was administrated in the drinking water at a concentration of 5 µg/ml (group F5), and 25 µg/ml (group F25). In addition, tumor-bearing mice without febuxostat (group TB) and control mice (group C) were established. At 4 weeks after the inoculation of tumor cells, body weight, wet weights of the gastrocnemius muscles, XO activity, 8-hydroxy-2’-deoxyguanosine(8-OHdG), and expression of TNF-α and IL-6 were evaluated and compared among the 4 groups.


Abstract Background
To analyze the effects of the administration of febuxostat, a selective xanthine oxidase (XO) inhibitor, on both in vitro and in vivo models of the waste of skeletal muscles, due to LM8 osteosarcoma cells.
Methods C2C12 myotubes were incubated in the conditioned medium of LM8 osteosarcoma cells. At the same time, febuxostat was added at a concentration of 3 µM and 30 µM, and reactive oxygen species(ROS) was measured using 2',7'-dichloro uorescein diacetate (DCF-DA) at 2 h and 24 h after exposure. Furthermore, an in vivo study was performed on male C3H mice (5 weeks old) that were subcutaneously injected with LM8 osteosarcoma cells on their backs. Febuxostat was administrated in the drinking water at a concentration of 5 µg/ml (group F5), and 25 µg/ml (group F25). In addition, tumor-bearing mice without febuxostat (group TB) and control mice (group C) were established. At 4 weeks after the inoculation of tumor cells, body weight, wet weights of the gastrocnemius muscles, XO activity, 8hydroxy-2'-deoxyguanosine , and expression of TNF-α and IL-6 were evaluated and compared among the 4 groups.

Results
ROS generation was clearly observed in C2C12 myotubes following incubation in the conditioned medium of LM8 osteosarcoma cells. ROS generation was signi cantly inhibited by febuxostat administration. Furthermore, LM8-bearing mice showed signi cant loss of body weight and wet weight of the gastrocnemius muscles, in which XO activity, 8-OHdG, and expression of IL-6 were signi cantly increased compared to those in mice without injections of the tumor cells. Further, febuxostat administration not only signi cantly improved the body weight and wet weight of the skeletal muscles, but also reduced markers of oxidative stress and expression of pro-in ammatory cytokines. Febuxostat did not show anti-tumor effects, including inhibition of lung metastasis and improved overall survival.

Conclusion
Febuxostat, which is clinically used for treatment of hyperuricemia, is effective against the waste of the skeletal muscles induced by LM8 osteosarcoma cells. Background Cachexia, due to malignant tumors, is a multifactorial syndrome characterized by an ongoing loss of skeletal muscle mass (with or without the loss of fat mass), that cannot be fully reversed by conventional nutritional support and leads to progressive functional impairment (1). This condition has been a challenging problem for orthopedists and oncologists, as cachexia has been shown to be associated with increased risk of postsurgical complications, functional impairment, and impaired responses to chemotherapy and anti-neoplastic treatments (2).
Although the pathophysiology of this disorder is unknown, we currently know that malignant tumors alter their surrounding environments via tumor-derived and host-derived paracrine factors (3). As one of the factors, reactive oxygen species (ROS) have been well known to contribute to various steps of tumor genesis, including transformation, growth promotion, and malignant progression. A large number of studies have been undertaken to study the role of ROS in the pathophysiology of malignant tumors (4)(5)(6)(7). The role of ROS in the important signaling events of skeletal muscle atrophy due to denervation, aging, and disuse (8,9), has also been implicated. It is reasonable that ROS could also be used as targets against skeletal muscle atrophy caused by malignant tumors as well as other disorders.
Febuxostat (2-(3-cyano-4-isobutoxy-phenyl)-4-methyl-1,3-thiazole-5carboxylic acid) is a selective inhibitor of xanthine oxidase (XO), which produces ROS via uric acid and also catalyzes the reduction of oxygen to superoxide. This drug is clinically used for treatment of patients with gout, and has been reported to be useful for treatment of various conditions related to ROS as well as hyperuricemia (10)(11)(12). However, no reports are available con rming its effectiveness against the deleterious effects of ROS in skeletal muscles under unfavorable circumstances such as malignant tumors.
Osteosarcoma is the most common primary malignant tumor of the bone. Due to the development of combination chemotherapy regimens, the curative rate has increased to 60-70% for 5-year survival, even though 5-20% patients survive using surgical resection alone (13). A therapeutic strategy is needed for the maintenance and improvement of daily and social activities in patients with osteosarcoma. In addition, previous research has reported 5-year survival rate of 35%, even in patients with lung metastasis, which occurs in approximately half of patients with osteosarcoma, of which 30-50% of the patients die due to lung metastasis (14)(15)(16). Chemotherapy is an essential component of osteosarcoma treatment. In various types of cancers, the waste of the skeletal muscles has been reported to be associated with the quality of the skeletal muscles (17)(18)(19). Further, inhibition of the muscle wasting can be also advantageous for osteosarcoma treatment.
Furthermore, increased in ammation is a characteristic feature of the cachexic condition in numerous diseases, including malignant tumors (20). Several cytokines, including TNF-α and IL-6, were shown to be implicated as mediators of cachexic progression in both human and animal models (21,22). The cytokines can regulate intracellular signaling, which results in muscle wasting and disrupted metabolic homeostasis in the skeletal muscles during the cachexic condition. Accordingly, the inhibition of in ammatory cytokines may be important in the control of cachexia.
The aim of this study is to analyze the effectiveness of febuxostat against cachexia, in vitro and in vivo.

Methods
Subjects C2C12 cell lines (ATCC, Manassas, VA, USA) were used for In vitro study. C2C12 murine myoblasts were cultured in a growth medium consisting of low-glucose Dulbecco's Modi ed Eagle's Medium (DMEM) containing 10% fetal bovine serum (FBS) in a 5% CO 2 incubator at a temperature of 37 °C. After culturing until 70-80% con uence, C2C12 cells were used for the following analysis.
An in vivo study was performed on 68 male C3H mice aged at 5 weeks (SLC, Hamamatsu, Japan). The animals were housed in a temperature-controlled environment and maintained on a 12-hour light-dark cycle with the standard chow diet for mice and rats (Rodent Diet CA-1, CLEA Japan Inc., Tokyo, Japan); water was available ad libitum. The experimental protocol was approved by the Animal Ethics Research Committee at Mie University.
Collection of conditioned medium LM8 osteosarcoma cells, which were derived from the murine osteosarcoma cell line Dunn osteosarcoma through a repetition of eight cycles of the procedure, were grown in 10% FBS at 37 °C in 5% CO 2 (23).
When the con uence reached 80-90%, the cells were washed with phosphate-buffered saline (PBS) and the medium was replaced with DMEM without phenol red or serum. After 48 h, the conditioned medium was collected by centrifugation at 12,000 rpm at 4 °C for 10 min and ltration with a 0.22 µm lter to remove debris (24). The conditioned medium was preserved at -80 °C for further experiments.
Antioxidant effect of myoblasts incubated in conditioned medium C2C12 cells were seeded at 1.0 × 10 4 cells/well on a 96 well-plate in DMEM with 10% FBS. After 24 h, the medium was replaced with DMEM containing 2% FBS for differentiation into the myotubes. At the same time, febuxostat was added at a concentration of either 3 µM or 30 µM for 24 h as premedication. Febuxostat was gratuitously provided by Teijin Pharma Limited (Tokyo, Japan). Following incubation in the conditioned medium for 2 and 24 h, levels of ROS were examined using 2',7'-dichloro uorescein diacetate (DCF-DA), according to the manufacturer's instructions (Abcam, Tokyo, Japan). Myotubes were washed with PBS and fresh DMEM without phenol red, and incubated with 10 µM DCF-DA for 30 min in darkness at room temperature. The cells were immediately analyzed, and ROS levels were measured by an increase in the DCF uorescence. DCF uorescence was measured at an excitation wavelength of 488 nm and an emission wavelength of 519 nm.
Animal models and febuxostat administration: Male C3H mice (5 weeks old; SLC, Hamamatsu, Japan) were inoculated with 1.0 × 10 7 LM8 osteosarcoma cells in 0.2 ml of PBS by subcutaneous injection into the backs of the mice. Control mice were injected with 0.2 ml of PBS. Febuxostat was administered in the drinking water, from the day of inoculation with the tumor cells. Four different groups were established: group F5, tumor-bearing mice with febuxostat 5 µg/ml; group F25, tumor-bearing mice with febuxostat tumor inoculation, mice were deeply anesthetized with an intraperitoneal injection of sodium pentobarbital (0.05 mg/g body weight), followed by body weight measurement. Then, muscles of gastrocnemius were fully excised from the attachment site of the bone, and wet weights were measured. Furthermore, lung metastasis was macroscopically observed.
Immunohistochemical analysis: The harvested gastrocnemius muscles were immediately xed in 4% paraformaldehyde at 4 °C overnight. Then, the specimens were embedded in para n, and were longitudinally cut at 4-mm thickness on a microtome. Specimens were dewaxed in xylene and rehydrated in graded ethanol (99-70% (v/v)) in distilled water. Endogenous peroxidase activity was quenched by 30min incubation in 0.3% (v/v) hydrogen peroxide in 99% methanol. Heat induced antigen retrieval was performed in 10 mM citrate buffer (pH 6.0) using a pressure cooker (Delicio 6L; T-FAL, Rumily, France).
The sections were then left to cool at room temperature in the citrate buffer. Nonspeci c staining was blocked by incubating the sections in a solution of 1% bovine serum albumin for 20 min at room temperature. The bovine serum albumin was then drained off, and specimens were incubated at room temperature overnight in each antibody (1: 50). Goat monoclonal 8OHdG antibodies (Japan Institute for Control of Aging, Tokyo, Japan) were used as primary antibodies. Bound primary antibodies were detected using secondary anti-goat Ig antibodies and anti-rabbit Ig antibodies conjugated to horseradish peroxidase (1:100; Dako Japan, Kyoto, Japan) for 1 h at room temperature. Bound antibodies were visualized by a reaction with 3, 30-diaminobenzidine. Between the incubation steps, sections were washed with PBS (3 × 5 min) to eliminate excess non-bound antibodies or reagents. Sections were counterstained with hematoxylin. We counted the number of 8OHdG labeled nuclei, and calculated the positive rate of 8OHdG expression by dividing this number by the total number of nuclei.
XO assay XO activity was measured using a commercially available kit (Sigma-Aldrich, Darmstadt, Germany). The harvested gastrocnemius muscles were homogenized with an assay buffer. The homogenate was centrifuged at 15,000 rpm for 10 min, and the supernatant was used for the assay. After the addition of reaction buffer, initial (T initial ) and nal (T nal ) measurements were performed at 570 nm on a microplate reader (FluoStar Galaxy®,BMG LABTECH GmbH, Ortenberg, Germany), and XO activity was calculated by the change in the measurements from T initial to T nal .

Measurements of in ammatory cytokine
The harvested gastrocnemius muscles were quickly frozen in liquid nitrogen, homogenized using Cryopress (Microtech, Chiba, Japan), and centrifuged, and the supernatant was preserved at -80 °C.

Statistical analyses
Page 6/19 SPSS Statistics Software version 23.0 (IBM, Armonk, NY, USA) was used. Results are shown as mean ± standard deviation (SD). The comparisons among groups were assessed using the Kruskal-Wallis test.
Further, overall survival was determined using the Kaplan-Meier survival analysis. Values of p < 0.05 were considered statistically signi cant.

Results
ROS generation and effects of febuxostat administration following exposure of the conditioned medium from LM8 osteosarcoma cells on C2C12 myotubes The DCF-DA assay (n = 5 in each group) showed that ROS were signi cantly upregulated to 1.9-fold in C2C12 myotubes incubated in the conditioned medium, compared to controls from the early phase (2 h). The upregulation of ROS was predominant at 24 h after incubation in the conditioned medium, where the ROS concentration increased by 3.0-fold compared to controls at 24 h. On the other hand, premedication of the febuxostat signi cantly inhibited upregulation of the ROS in the C2C12 myoblasts, with a dose of 30 µM at 2 h, and both doses of 3 µM and 30 µM at 24 h after exposure to the conditioned medium ( Fig. 1). In addition, no differences in the ROS concentration were observed in group F5 and F25 between 2 and 24 h after incubation in the conditioned medium, even though the ROS signi cantly increased in the C2C12 myoblasts that lacked any added febuxostat.

Evaluation Of The Mice Inoculated With Lm8 Osteosarcoma Cells
At 4 weeks after inoculation with LM8 osteosarcoma, the body weight was signi cantly decreased in group TB (24.9 ± 1.4 g) compared to group C (20.8 ± 1.7 g). The wet weight of the gastrocnemius muscles of group TB mice was signi cantly lower (88.3 ± 10.5 mg) than that of group C mice (118.3 ± 10.3 mg). In the skeletal muscles, signi cant differences were observed in the XO activity as well as the ratio of 8-OHdG positive nuclei in group TB and group C. In addition, expression of pro-in ammatory cytokines in the gastrocnemius muscles of group TB were 12.1 ± 1.3 pg/ml in TNF-α and 1.1 ± 0.3 pg/ml in IL-6, which are higher than the corresponding 10.0 ± 1.7 pg/ml in TNF-α and 0.8 ± 0.3 pg/ml in IL-6 of group C.
Inhibitory effects of febuxostat on the general condition and waste of skeletal muscles in tumor-bearing mice with LM8 osteosarcoma cells Administration of febuxostat signi cantly inhibited body weight loss in both F5 (23.2 ± 1.9 g) and F25 (23.5 ± 2.6 g) groups at 4 weeks, although approximately 15% loss of body weight was recorded in the group TB mice, compared to the group C mice (Fig. 2A). Administration of febuxostat in tumor-bearing mice also led to a signi cant improvement in skeletal muscle weight in both F5 and F25 groups (Fig. 2B). In the skeletal muscles, XO activity was signi cantly inhibited due to administration of febuxostat (group F5, 9.1 ± 1.6 µU/mg; and group F25, 8.1 ± 1.5 µU/mg) (Fig. 3). The ratio of 8-OHdG positive nuclei was also signi cantly reduced in the gastrocnemius muscles of group F5 and F25, compared to those of group TB (Fig. 4). Likewise, expression of TNF-α and IL-6 was also signi cantly reduced in the gastrocnemius muscles of group F5 compared to those of group TB, and IL-6 was also signi cantly inhibited in the gastrocnemius muscles of group F25 (Fig. 5). TNF-α and IL-6 correlated negatively with the wet weight of the gastrocnemius muscles of all mice of all the groups (Fig. 6).
Moreover, lung metastasis was macroscopically observed in mice of all the groups. The mean survival time was 49.8, 60.5, and 56.8 days for group TB, group F5, and group F25, respectively. There were insigni cant differences in overall survival among the groups (Fig. 7).

Discussion
Cachexia is a secondary condition to chronic diseases, including cancer, diabetic mellitus, chronic obstructive pulmonary disease, and chronic heart disease. Although the pathophysiology of cachexia depends on the types of diseases, a common feature of these conditions is alteration of several factors in the plasma with destructive effects on the local tissues (25). Especially in the cachexic condition, caused by malignant tumors, tumor-derived factors, including in ammatory cytokines and exosome, have been reported to be involved in waste of the skeletal muscles, as a result of imbalance of the degradation and/or synthesis on myo brillar proteins (26). Oxidative stress is one of the most common mechanisms in the signaling pathway of the imbalance under the cachexic condition in various diseases, and one of the important characteristics is increased ROS levels in the skeletal muscles of the cachexic patients (25,27). This study showed that secretion products from LM8 osteosarcoma cells had a harmful effect on the C2C12 myotubes due to upregulation of ROS.
The administration of the antioxidants is considered as one of the therapeutic strategies for the treatment of waste of the skeletal muscles under the cachexic condition. In fact, a number of studies have reported the clinical uses of antioxidants in the treatment of the waste of skeletal muscles in various conditions (28)(29)(30)(31)(32). In this in vitro study, it was demonstrated that administration of febuxostat signi cantly reduces the generation of ROS in C2C12 myoblasts of LM8 osteosarcoma cells incubated in conditioned medium. Febuxostat is a clinically used drug for the treatment of hyperuricemia and could concomitantly suppress the generation of ROS in inhibition of xanthine oxidase. Thus, febuxostat is likely to have an ability of protection for skeletal muscles against deteriorated effects due to oxidative stress by tumors. In fact, our in vivo study clearly showed that the administration of febuxostat signi cantly inhibited the loss of body weight and wet weight of the gastrocnemius muscle in the mice injected with LM8 osteosarcoma cells. The mechanism seemed to be a direct inhibition of oxidative stress due to the increase in ROS generation in the skeletal muscles, since administration of febuxostat signi cantly reduced XO activity and expression of oxidative stress markers  in the skeletal muscles of the LM8-induced mice.
The reduction of in ammatory cytokine expression was also observed in the skeletal muscles of this model by the administration of this drug, which is a selective XO inhibitor but not an anti-in ammatory agent. The in ammatory cytokine has been implicated as a critical mediator for progression of the cachexic condition of the skeletal muscles (33,34). TNF-α can promote protein degradation through the transcription of ubiquitin proteasome E3, MurF1, and atrogin-1, of which expression had been shown in the skeletal muscles of cachexic individuals and the animal models of cachexia, which results in the degradation of proteins in the skeletal muscles (21). IL-6 was also implicated as a critical regulator of in ammation-induced wasting of the skeletal muscles during cachexic progression (35). In fact, we found a signi cant correlation between reduction of wet weight of the gastrocnemius muscles and expression of TNF-α and IL-6. Besides, several studies showed anti-in ammatory effects by the administration of febuxostat in various organs, including liver, lung, and kidneys (36,37). In these studies, suppressive effects on in ammation was likely to be a result of the inhibition of ROS production, which is known to induce the expression of in ammation-related cytokines. In addition, injured cells could induce the in ammation, resulting in further damage to the tissues. Thus, we suggested that the administration of the febuxostat could also protect the skeletal muscles exposed to secreted products from the LM8 osteosarcoma cells via an indirect mechanism by suppression of the in ammatory cytokines, as a result of reduced ROS generation.
The use of LM8 osteosarcoma cells was a major limitation of this study for the induction of cachexic condition, since this cell line has the characteristic of a high ability of pulmonary metastasis. Therefore, it was unclear whether ROS generation in this study was induced by oxidative stress due to pulmonary metastasis or secreted products from tumor cells. However, regardless of pulmonary metastasis in all mice inoculated with LM8 osteosarcoma cells, the mice were inhibited from experiencing a reduced body weight and wet weight of the gastrocnemius muscle by administration of the febuxostat. In addition, pulmonary metastasis as well as cachexia were observed in patients of the terminal phases. We believe that this study shows the effectiveness of the febuxostat for the quality of life of patients who are at advanced stage of malignant tumors, even though the lung metastasis and overall survival rates were ineffective.
In conclusion, febuxostat was not only effective against ROS generation in C2C12 myotubes incubated in the conditioned medium of LM8 osteosarcoma cells, but also inhibited the loss of the body weight and wet weight of the gastrocnemius muscles in the mice following the subcutaneous injection of LM8 osteosarcoma cells. The skeletal muscles in the LM8 injected mice showed signi cant increase in the oxidative stress markers and pro-in ammatory cytokines' expression, while administration of febuxostat signi cantly inhibited the increase of the expressions in skeletal muscles. Febuxostat could be a useful treatment for waste of the skeletal muscles caused by LM8 osteosarcoma, even though the drug has been ineffective for tumorigenesis.

Not applicable
Availability of data and material The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Funding
Not applicable

Competing interests
The authors declare that they have no competing interests.

Authors' contributions
TT and MT designed the research and wrote the paper. TI and TA participated in the experimental design and techniques. All authors read and approved the nal manuscript.    Xanthine oxidase (XO) activity in the muscles. The XO activity was signi cantly increased in the skeletal muscles of group TB compared to group C. The activity was clearly reduced by the administration of febuxostat. Group TB, LM8-bearing mice without febuxostat; group F5, LM8-bearing mice with febuxostat 5 µg/ml; group F25, tumor-bearing mice with febuxostat 25 µg/ml; and group C, control mice. * p<0.05 (among each group).

Figure 4
Immunohistochemical analysis of 8-OHdG on gastrocnemius muscles. The pictures showed that the stained nuclei were markedly increased in group TB compared to group C. In fact, the number of stained nuclei was signi cantly increased due to inoculation of the LM8 osteosarcoma cells, whereas the number was decreased in the skeletal muscles of group F5 and group F25 compared to group TB. Group TB, LM8bearing mice without febuxostat; group F5, LM8-bearing mice with febuxostat 5 µg/ml; group F25, tumorbearing mice with febuxostat 25 µg/ml; and group C, control mice. * p<0.05 (among each group). TNF-α (A) and IL-6 (B) in the gastrocnemius muscles. TNF-α in the skeletal muscles were signi cantly reduced in group F5, along with the tendency in group F25. IL-6 was also upregulated due to the injection of LM8 osteosarcoma cells, and was signi cantly inhibited in both groups F5 and F25. Group TB, LM8bearing mice without febuxostat; group F5, LM8-bearing mice with febuxostat 5 µg/ml; group F25, tumorbearing mice with febuxostat 25 µg/ml; and group C, control mice. * p<0.05 (among each group).  Overall survival of the mice with inoculation of LM8 osteosarcoma cells. Mean duration values of the overall survival were 49.8, 60.5, and 56.75 days in group TB, group F5, and group F25, respectively. No differences among groups were found. Group TB, LM8-bearing mice without febuxostat; group F5, LM8bearing mice with febuxostat 5 µg/ml; and group F25, tumor-bearing mice with febuxostat 25 µg/ml