Metabolome analysis of masseter muscle in senescence-accelerated mouse-prone 8 (SAMP8)

Frailty is a vulnerable state that marks the transition to long-term care for the elders. Recently, the relationship between frailty and oral function has been attracting attention. By clarifying the specic metabolic changes in the masseter muscle, we aimed to contribute to maintenance of masticatory function. The purpose of this study is to clarify the changes in masseter muscle of senescence-accelerated mouse-prone 8 (SAMP8) mice metabolites and metabolic pathways due to aging. Capillary electrophoresis-mass spectrometry metabolome analysis was performed on the masseter muscle of 12-week-old, 40-week-old, and 55-week-old mice. Expression analysis was performed by reverse transcription polymerase chain reaction (RT-PCR) and immunouorescence for the metabolome pathways extracted by metabolome analysis that considered to be related to aging. Nineteen metabolites had a signicant difference in absolute quantitative values and were considered to affect the rst principal component by factor loading. The extracted metabolic pathways were glycolysis, polyamine metabolome pathway, and purine metabolome pathway. RT-PCR was performed on the extracted metabolome pathways. Expression of the spermidine synthase and hypoxanthine phosphoribosyl transferase genes with signicant differences by RT-PCR was conrmed by immunouorescence. The metabolic pathways considered to be related to aging in masseter muscle were glycolysis, polyamine metabolic pathway, and purine metabolic pathway. have Studies indicate that glycolytic metabolites of skeletal muscle decrease with aging [4] . In the gastrocnemius muscle, fructose 1,6-diphosphate, dihydroxyacetone phosphate, and glyceraldehyde 3-phosphate (upstream metabolites) decreased [4] . Our results indicated a decrease in glycolytic metabolites. These metabolites were 3-phosphoglyceric acid, 2-phosphoglyceric acid, and phosphoenolpyruvic acid (downstream metabolites). MyHC Type IIb bers decrease in the masseter muscle of SAMP8 mice, notably between 40-week-old and 55-week-old mice [8] , which decline could be associated with the aforementioned decrease. The dephosphorylation of 1,3-bisphosphoglyceric acid produces 3-phosphoglyceric acid. This reaction is accompanied by ATP production through the phosphorylation of ADP. Phosphoenolpyruvic acid is produced by the dehydration of 2-phosphoglyceric acid. This reaction is considered to produce pyruvic acid simultaneously to ATP by the transfer of a phosphate group to ADP. We suggest that the decrease in 3-phosphoglyceric acid, 2-phosphoglyceric acid, and phosphoenolpyruvic acid in the masseter muscle was caused by the consumption of a large amount of ATP required for masticatory movements.


Introduction
Frailty is a vulnerable state which marks the transition to long-term care for the elders and an increase in their risks of health problems, falls and death [1] . Sarcopenia is the main symptom of frailty. The early detection and prevention of sarcopenia is important to ensure an excellent quality of life for the aged.
Fundamental research on aging has been conducted with the use of aging animals. Leann et al. measured the cross-sectional area of the soleus muscle in Fischer 344 Brown Norway (FBNF1) rats and reported the atrophy of fast-twitch muscles and notable changes in old age [2] . Guo et al. observed morphological and immunohistochemical changes over time in the gastrocnemius muscle of senescence-accelerated mouse-prone 8 (SAMP8) mice and reported that sarcopenia appeared when they were 40 weeks old [3] . Furthermore, Uchitomi et al. performed metabolome analysis on the gastrocnemius muscle of aged C57BL/6J mice to search for metabolites that uctuate due to aging [4] . Consequently, they reported a decrease in glucose metabolites, a decrease in polyamine metabolites, and an increase in neurotransmitters.
Recently, the relationship between frailty and oral function has been attracting attention in the subject of numerous studies. For instance, in an epidemiological study, Murakami et al. conducted a survey of community-dwelling older people in Japan and reported that sarcopenia was associated with the masticatory function [5] . Meanwhile, Iwasaki et al. reported that maximum bite force was associated with frailty [6] . Moreover, the concept of oral frailty was proposed for the early detection of slight deterioration in the aged, and the characteristic of poor diadochokinesis, constitutes the rst step toward oral disfunction. Oral frailty is also associated with the prognosis of curtailed longevity [7] . We have noted that there are only a few de ning studies that relate the deterioration of oral health to aging. Hoshino et al. reported that the atrophy of the masseter muscle in SAMP8 mice had occurred by the time they were 40 weeks old and that there were subsequent changes in muscle contraction characteristics [8] . However, there are no reports of the molecular biological examination of the masseter muscle during aging. Given this context, we focused on the metabolites in this study. By clarifying the speci c metabolic changes in the masseter muscle, we aimed to contribute to maintenance of masticatory function. Hence, the purpose of this study is to clarify the changes in masseter muscle metabolites and metabolic pathways due to aging. We veri ed the results through a metabolome analysis of the masseter muscles of SAMP8 mice [8] , as the changes due to aging have already been con rmed, in this type of young mice (12-week-old), elderly mice (40-week-old), and later stage elderly mice (55-week-old).

Body weight
The mean body weight of the mice was 28.9 ±1.01 g at 12 weeks old, 39.22 ±3.43 g at 40 weeks old, and 39.96±3.79 g at 55 weeks old (Fig. 1a). While body weight increased between weeks 12 and 40, and between weeks 12 and 55, there was no difference in weight between 40-week-old and 55-week-old mice.

Amount of food intake
The mean weekly food intake was 36.2±3.88 g at 12 weeks old, 39.8±6.01 g at 40 weeks old, and 25.6±0.76 g at 55 weeks old (Fig. 1b). Food intake decreased between 12 weeks and 55 weeks, and between 40 weeks and 55 weeks, whereas there was no difference in food consumption between 12- week-old and 40-week-old mice.

Metabolome analysis (masseter muscle)
A principal component analysis (PCA) demonstrated that the metabolites contributing to the rst principal component could be clearly distinguished at each age (Fig. 2a). Hierarchical cluster analysis (HCA) also revealed changes in metabolite pro les (Fig. 2b). Moreover, clear metabolic uctuations were observed from 40 weeks to 55 weeks.
Metabolites that uctuated between 40 weeks and 55 weeks There were signi cant differences in the absolute quantitative values of 36 substances (Table. 1), 20 of which were elevated metabolites and 16 were lowered metabolites. Furthermore, the factor loadings that affected the rst principal component were 10 elevated substances and 9 lowered substances (Table 2). Moreover, signi cant changes were observed in the polyamine metabolic pathway, purine metabolic pathway, and glycolysis pathway.
Since the polyamine metabolic pathway is associated with cell proliferation, protein synthesis, and nucleic acid synthesis, a signi cant increase in S-adenosylmethionine and spermidine was observed (Fig.  3).
Since the purine metabolic pathway is related to purine nucleotide metabolism and ATP resynthesis, a signi cant increase in IMP, GMP, GDP, and hypoxanthine, and a signi cant decrease in GTP and ATP were observed (Fig. 4).
Additionally, we observed an increase in the neurotransmitter choline and a decrease in phosphocreatine, which is used for ATP synthesis and to produce creatinine, which is a reaction product.

Reverse transcription polymerase chain reaction (RT-PCR)
We con rmed the expression of enzymes that are involved in the polyamine and purine metabolic pathways and that uctuated between 40-week-old and 55-week-old mice. In the polyamine metabolic pathway, we observed a signi cant increase in the expression of the spermidine synthase gene (SRM) in 55-week-old mice compared with 40-week-old mice (p = 0.002) (Fig. 6 a). Furthermore, we observed a signi cant increase in the expression of the hypoxanthine phosphoribosyl transferase gene (HPRT), an enzyme used in the salvage circuit of ATP resynthesis (p = 0.048) in the purine metabolic pathway (Fig. 6 b).

Immunohistochemical observations
The SRM of the polyamine metabolic pathway was expressed surrounding the muscle bers, especially in the interstitium (Fig. 7). The HPRT of the purine metabolic pathway was also expressed surrounding the muscle bers. (Fig. 8 ).

Discussion
The body weight of SAMP8 mice is said to increase between the age of 8 weeks and 24 weeks [9] .
Similarly, in this study, the body weight tended to increase as the mice aged. Furthermore, there was no clear individual difference during the raising stage. Moreover, aging physiologically reduces food intake, and anorexia due to aging leads to energy malnutrition and weight loss [10] . We observed a decrease in food intake between 12-week-old and 55-week-old mice, and between 40-week-old and 55-week-old mice. However, since it is not accompanied by weight loss, it can be inferred that it was not a case of simple undernutrition. Consequently, to clarify the relationship between body weights and food intakes, we considered that other factors such as lean body mass and whole-body muscle mass should be measured. We also considered that nutritional indicators such as blood cholesterol level and blood hemoglobin should be examined.

Page 5/19
The masseter muscle consists of predominantly fast-twitch muscle, with many MyHC Type II bers distributed among the muscle bers. Fast-twitch muscle bers have high enzymatic activity and creatine kinase activity that compose glycolysis and have excellent anaerobic metabolic capacity [11] . Studies indicate that glycolytic metabolites of skeletal muscle decrease with aging [4] . In the gastrocnemius muscle, fructose 1,6-diphosphate, dihydroxyacetone phosphate, and glyceraldehyde 3-phosphate (upstream metabolites) decreased [4] . Our results indicated a decrease in glycolytic metabolites. These metabolites were 3-phosphoglyceric acid, 2-phosphoglyceric acid, and phosphoenolpyruvic acid (downstream metabolites). MyHC Type IIb bers decrease in the masseter muscle of SAMP8 mice, notably between 40-week-old and 55-week-old mice [8] , which decline could be associated with the aforementioned decrease. The dephosphorylation of 1,3-bisphosphoglyceric acid produces 3phosphoglyceric acid. This reaction is accompanied by ATP production through the phosphorylation of ADP. Phosphoenolpyruvic acid is produced by the dehydration of 2-phosphoglyceric acid. This reaction is considered to produce pyruvic acid simultaneously to ATP by the transfer of a phosphate group to ADP. We suggest that the decrease in 3-phosphoglyceric acid, 2-phosphoglyceric acid, and phosphoenolpyruvic acid in the masseter muscle was caused by the consumption of a large amount of ATP required for masticatory movements.
Spermidine is an autophagy-related autoinducer [12] . Its concentration has been reported to decrease with age [13] . Moreover, its concentration decreases signi cantly in the gastrocnemius muscle of C57/BL6J mice [4] . In this study, we observed a signi cant increase of spermidine in the masseter muscle. Furthermore, in the masseter muscle, the expression of SRM, a gene for spermidine synthase, was con rmed in the interstitium by uorescent immunohistochemical staining, and the expression was enhanced in the comparison between 40-week-old and 55-week-old mice. The same result was achieved by reverse transcription polymerase chain reaction (RT-PCR), and we believe that it participates in the synthesis of spermidine in the polyamine metabolic pathway. The masseter muscle's involvement in the masticatory function is essential to maintain a normal life. Additionally, there are reports that state autophagy is activated in comparison to the hind limbs, [14] and that masseter muscles in the young and elderly have similar performance to that of rats [15] . Based on these ndings, we think that the masseter muscle, which is used more frequently than the hind limbs, is more susceptible to fatigue and aging, and a compensatory increase in spermidine induces autophagy activation, which might be correlated with maintaining functioning. As spermine is necessary for spermidine resynthesis [16] , although no signi cant difference was observed in this research, it is considered that it decreases (p =0) at 40 weeks due to the increase of spermidine at 55 weeks.
Furthermore, S-adenosylmethionine is involved in DNA methylation and is reported to increase with aging [17][18][19] . DNA methylation is a sign of slight changes in tissue composition such as tissue in ammation and brosis [20] , and has become the subject of several studies as an epigenic biomarker of aging [21,22] .
In this study, we con rmed that SAMP8 increased from 40-week-old to 55-week-old mice. We consider that this may decrease Type II bers of the masseter muscle and the uctuation in the polyamine metabolic pathway.
The purine metabolic pathway is engaged in ATP resynthesis during movement. Phosphocreatine is required for ATP synthesis [23] , and hypoxanthine and IMP increases with movement [24] . According to the results, changes in glucose metabolism and the increase in hypoxanthine and IMP are maintained by the masticatory function in SAMP8's masseter muscle. We consider that the increased demand for energy sources ATP and GTP that leads to the activation of the salvage circuit for resynthesis may have caused the increase in HPRT. Furthermore, ATP was one of the factors responsible for the decrease in phosphocreatine. The expression of HPRT was con rmed by uorescent immunohistochemical staining and RT-PCR, thus suggesting the activation of the salvage circuit of the purine metabolic pathway.
A limitation of this study is that aging changes in SAMP8 mice may differ from those in normal human aging. Although SAMP8 mice have a shorter lifespan than aged animals [25] , and develop sarcopenia phenotype [3] , we expect that they will be used in future studies on aging. In our study, SAMP8 mice were used to understand the characteristics of aging in the masseter muscle. Moreover, since humans and mice have different masticatory forms and muscle ber composition, it may be unlikely that the results on mice will re ect the same changes that occur in humans.
However, these results will lead to the establishment of a masseter aging mechanism. Additionally, SAMR1 is often used as a control group for SAM. It has been previously con rmed that the masseter muscle in SAMP8 mice had histological changes due to aging when compared to SAMR1 mice [8] .
Moreover, to further clarify the phenotype of the masseter muscle with aging, there is a need to compare it with the younger mice as in previous studies [4] .
In conclusion, our results suggest that the age-related metabolic pathways of the masseter muscle on SAMP8 are likely to be the glycolysis and polyamine metabolome pathway and the purine metabolome pathway.

Methods
Animal experiment SAMP8 mice were used in this study. SAMP8 is a type of mouse whose age accelerates between 16 weeks and 20 weeks, and is often used in aging research. The mice were raised for each of the following groups: 12-week-old (young) (N=4) to a stage before the start of accelerated aging; 40-week-old (elderly) (N=4) in the mean lifespan of SAMP8; 55-week-old (late elderly) (N=5) to a stage that exceeded the mean lifespan. In this study, the same species and the same age as those used in the previous studies [8] were used. The mice were raised in a 125×213×125 mm aluminum cage and were freely given food (Lab MR-A1) and water. Upon administering general anesthesia with iso urane, the mice were euthanized by dislocating the cervical spines, and the masseter muscles were extracted. The experiments were conducted in accordance with the National Institutes of Health guidelines for care and use of animals and also approved by the Tokyo Dental College Institutional Animal Care and Use Committee (approval number: 202601). The study was carried out in compliance with the ARRIVE guidelines.
Body weight and the amount of food intake In this study, body weight was measured once a week with an electronic scale (AND EW-1500i) and the food was weighed concurrently with a balance from 12-week-old to 55-week-old. The following week, the feed was again measured concomitantly with the mice's bodyweight, and the difference in the mass of the food between the two weeks represented the amount of food intake.

Capillary electrophoresis-mass spectrometry (CE-MS) metabolome analysis
Approximately 50 mg of frozen tissue was plunged into 750 µL of 50% acetonitrile/Milli-Q water containing internal standards (H33041002, Human Metabolome Technologies, Inc., Tsuruoka, Japan) at 0 °C to inactivate enzymes. The tissue was homogenized three times at 1,500 rpm for 2 min using a tissue homogenizer (Micro Smash MS100R, Tomy Digital Biology Co., Ltd., Tokyo, Japan) and then the homogenate was centrifuged at 2,300 ×g and 4 °C for 5 min. Subsequently, 400 µL of the upper aqueous layer was centrifugally ltered through a Millipore 5-kDa cutoff lter at 9,100 ×g at 4 °C for 120 min to remove proteins. The ltrate was centrifugally concentrated and re-suspended in 50 µL of Milli-Q water for capillary electrophoresis-mass spectrometry (CE-MS) analysis.
The metabolome analysis was performed with capillary electrophoresis time-of-ight mass spectrometry (CE-TOFMS) for cation analysis and CE-tandem mass spectrometry (CE-MS/MS) for anion analysis based on the methods described previously. This analysis focused on 116 central metabolites. CE-TOFMS analysis was conducted using an Agilent CE capillary electrophoresis system equipped with an Agilent 6210 time-of-ight mass spectrometer (Agilent Technologies, Waldbronn, Germany).

Quantitative analysis of expression metabolites
Total RNA extraction of the masseter muscle was performed using the RNeasy Mini Kit (Qiagen, Hilden, Germany) and Proteinase K (Takara Bio, Shiga, Japan). The cDNA was prepared using the QuantiTect Reverse Transcription Kit (Thermo Fisher Scienti c, Waltham, MA, USA). The Double Delta Ct Value (ΔΔCt) method was used for quanti cation, and the relative expression level was compared to the expression of housekeeping genes in each sample.
For gene expression analysis, the TaqMan® gene expression assays (Thermo Fisher Scienti c, Massachusetts, USA) were used in the RT-PCR 7500 realtime polymerase chain reaction (PCR) system (Thermo Fisher Scienti c). This study focused on the polyamine and purine metabolic pathways, which were the most common metabolites that uctuated between 40-week-old and 55-week-old mice. In addition, the housekeeping gene β-actin (Mm00607939_s1) was used.

Immunohistochemical observations
The gathered masseter muscle was cut approximately 5 mm at the center of the belly of the muscle and xed to a cork plate perpendicular to the muscle bers. The specimens were rapidly frozen using isopentane cooled in liquid nitrogen. Frozen sections with a thickness of 4.5 μm and 20 μm were prepared using a cryostat M1950 (Leica, Nussloch, Germany), and then dried at room temperature for 60 min. In the immunohistochemical staining, the local expression of SRM in the polyamine metabolic pathway and HPRT in the purine metabolic pathway, which showed signi cant differences in gene expression, were observed in muscle bers. The sections used for the SRM antibody were xed with acetone for 10 min, and the sections used for the HPRT antibody were xed with methanol for 10 min.
After washing with PBS solution for 15 min, blocking was performed with 10% goat serum for 1 h and washed with PBS solution for 15 min. The primary antibody used was a 200fold diluted HPRT antibody (GeneTex, California, US) and SRM antibody (Proteintech, Chicago, US), and the specimen was incubated overnight at 4 °C after applying the primary antibody.
Furthermore, the secondary antibody, a goat anti-rabbit IgG antibody Alexa Fluor Plus 488 (Invitrogen, Massachusetts, US) was diluted 200-fold for the HPRT antibody and SRM antibody. To con rm the cytoskeleton, after applying the Acti-stain ™ 555 Fluorescent Phalloidin (Cytoskeleton Inc., Denver, US) secondary antibody, the sections were incubated at room temperature for 1 h under a shade. A universal microscope (Axioplan2 and Axiophot2, Carl Zeiss, Inc., Jena, Germany) and a confocal laser scanning microscope (LSM 880 with Airyscan, Carl Zeiss, Inc., Jena, Germany) were used for observation.

Statistical analyses
Statistical analyses were performed with IBM SPSS Statistics version 26 (IBM, Armonk, NY, USA) software. Moreover, for the statistics of body weight and food intake, Tukey's test was performed for the multiple comparison test. The metabolites, RT-PCR, and immunohistochemical stained areas of 40-weekold and 55-week-old mice were compared using Welch's t-test. The signi cance threshold was set at 5% (p < 0.05). The detection limit was described (p=0).

Declarations
"Ratio" is the comparative value of the relative areas (40-week-old vs. 55-week-old). The pvalue was calculated using Welch's t-test (***p < 0.001, **p < 0.01, *p < 0.05). Table. 2 List of significant changes in metabolites between 40-week-old and 55-week-old by absolute quantitative value and factor loading. 55-week-old / 40-week-old. a Body weight Changes in body weights from 12-week-old to 55-week-old in ve mice of 55-week-old. Body weights was measured weekly. ( ** p < 0.01). b Amount of feed intake