The angiotensin-converting enzyme 2 (ACE2)-angiotensin 1-7 (A1-7)-A1-7 receptor (Mas) axis plays a protective role in the renin–angiotensin system (RAS). We recently found that ACE2 knockout (ACE2KO) mice exhibit earlier aging-associated muscle weakness, and that A1-7 alleviates muscle weakness in aging mice. In the present study, we investigated the role of the A1-7-Mas pathway in the effect of ACE2 on physiological aging. Male wild-type, ACE2KO, and Mas knockout (MasKO) mice were subjected to periodical grip strength measurement, followed by administration of A1-7 or vehicle for 4 weeks at 24 months of age. ACE2KO mice exhibited decreased grip strength after 6 months of age, while grip strength of MasKO mice was similar to that of wild-type mice. A1-7 improved grip strength in ACE2KO and wild-type mice, but not in MasKO mice. Muscle fibre size was smaller in ACE2KO mice than that in wild-type and MasKO mice, and increased with A1-7 in ACE2KO and WT mice, but not in MasKO mice. Centrally nucleated fibres (CNFs) and expression of the senescence-associated gene p16INK4a in skeletal muscles were enhanced only in ACE2KO mice and were not altered by A1-7. ACE2KO mice, but not MasKO mice, exhibited thinning of peripheral fat along with increased adipose expression of p16INK4a. A1-7 significantly increased bone volume in wild-type and ACE2KO mice, but not in MasKO mice. Our findings suggest that the impact of ACE2 on physiological aging does not depend on the endogenous production of A1-7 by ACE2, while overactivation of the A1-7-Mas pathway could alleviate sarcopenia and osteoporosis in aged mice.
Physiological aging is a long-term process that can result in diverse health conditions in older adults. This diversity creates a high-risk elderly population with health issues such as sarcopenia and frailty [1,2]. These health disorders can hamper independent living of the elderly and shorten their healthy lifespan [3,4]. Muscle weakness is a primary determinant of sarcopenia and frailty, and the prevention or alleviation of muscle weakness is widely recognised as a potential therapeutic target of these disorders. However, to date, an effective medical treatment is yet to be established for muscle weakness in older adults. Therefore, it is necessary to identify molecules associated with muscle weakness due to physiological aging .
We recently identified angiotensin-converting enzyme 2 (ACE2) as a candidate molecule associated with muscle weakness due to aging in mice. We found that ACE2 knockout (ACE2KO) mice exhibit earlier aging-associated muscle weakness with signatures of aging including the induction of p16INK4a, a senescence-associated gene, and increased numbers of centrally nucleated fibres (CNFs) in skeletal muscle . ACE2 plays a critical role in the renin–angiotensin system (RAS) as a carboxypeptidase to cleave angiotensin II to angiotensin 1-7 (A1-7) [7,8]. Recent findings suggest that the binding of A1-7 to its receptor, Mas, helps rescue the effects of RAS activation in various pathologies including muscle disorders . In a recent report, we also found that infusion of A1-7 for 4 weeks alleviates muscle weakness in both aged (24-month-old) ACE2KO and control wild-type mice . However, we did not find any effect of A1-7 infusion on expression of p16INK4a or number of central nuclei in skeletal muscle in both groups. It should also be noted that ACE2 has multiple functions outside the RAS that have never been tested for an association with aging [10–12]. Therefore, we could not conclude that the ablation of A1-7 production was the primary contributor to the early muscle aging phenotypes in ACE2KO mice.
In the present study, we investigated the role of the A1-7-Mas pathway in the effect of ACE2 on muscle aging in mice. Additionally, physiological aging was analysed in several tissues including bone, skin, and adipose tissue. We finally investigated whether the A1-7-Mas pathway could alleviate reduced bone volume along with muscle weakness in aged mice. We carried out the present study to clarify the molecular mechanism of physiological aging and aid drug discovery efforts for alleviating motor dysfunction and osteoporosis in the elderly.
In the present study, we tested the hypothesis that the accelerated aging phenotypes in ACE2KO mice are also observed in Mas knockout (MasKO) mice and that the effects of A1-7 on muscle function and bone volume are absent from aged MasKO mice. Male wild-type, MasKO, and ACE2KO mice (C57BL/6J background) were used for the experiment. ACE2KO mice were obtained by breeding between male ACE2KO mice (−/y) and heterozygous female ACE2KO mice (−/+). MasKO mice were obtained by breeding between heterozygous MasKO mice. Wild-type mice were randomly chosen from both colonies. These mice were maintained under specific pathogen-free conditions at 22°C under a 12-h light–dark cycle and received standard chow and water ad libitum. Forelimb grip strength and body weight were measured periodically at 3, 6, 12, 18, and 24 months of age. At 24 months, A1-7 (4332-v; PEPTIDE INSTITUTE, Japan) in 0.9% saline or vehicle (0.9% saline) was administered for 4 weeks at a dose of 0.1 µg/kg body weight/min via a subcutaneously implanted osmotic mini-pump (Alzet model 1004; Durect, U.S.A.) according to previously described methods . Thereafter, grip strength was measured, blood pressure (BP) was measured by a tail-cuff method, and tissues were extracted and frozen at −80°C for further analysis. Mice were killed by exsanguination under anaesthesia with medetomidine (0.75 mg/kg), midazolam (4 mg/kg), and butorphanol (5 mg/kg). All animal testing was conducted in the Institute of Experimental Animal Sciences in Osaka University. The Animal Care and Use Committee of Osaka University approved the study protocol. All experiments were performed strictly according to the U.S. National Institutes of Health Guide for the Care and Use of Laboratory Animals. All efforts were made to reduce animal suffering and to minimise the number of animals used.
Forelimb grip strength tests
Forelimb grip strength was measured on a digital force transducer (GPM-100; Melquest, Japan) as previously described [6,13]. We performed six consecutive measurements a day for 2 days. The average of 12 measurements was used to represent grip strength for each animal. The measurements were performed by investigators who were blinded with respect to the experimental group being tested. All test sessions were performed during the afternoon hours of the light cycle (11 a.m. to 5 p.m.) in the vivarium where the animals were housed.
RT-PCR array analysis
After total RNA was extracted from Extensor digitorum longus (EDL) muscle using TRIzol reagent (Thermo Fisher Scientific, U.S.A.), the RNA was amplified using the PreAMP cDNA Synthesis Primer Mix for Mouse Skeletal Muscle (QIAGEN, Germany). Then, a PCR array was performed in 384-well plates with an RT² Profiler™ PCR Array Mouse Skeletal Muscle: Myogenesis and Myopathy (QIAGEN, Germany) according to the manufacturer’s protocol. The numbers of the samples analysed were WT vehicle group n=3, ACE2KO vehicle group n=4, and ACE2KO A1-7 group n=3.
Reverse transcription and quantitative real-time PCR
Total RNA was extracted from mouse tissues using TRIzol reagent (Thermo Fisher Scientific, U.S.A.). cDNA was synthesised with a RevetraAce qPCR RT kit (FSQ-101, TOYOBO, Japan) according to the manufacturer’s instructions. Quantitative real-time PCR (RT-qPCR) was performed with the SYBR green qPCR system (TAKARA, Japan) and a 7900 HT Fast Real Time PCR instrument (Applied Biosystems, Japan). The data were analysed with the SDS 2.4 software (Applied Biosystems, Japan). Relative expression was calculated using the ΔΔCt method with normalisation to constitutively expressed genes. The PCR primer sequences of GAPDH, p16, p19, p21, and p53 were previously described . Commercially available PCR primer sets were used for BAP4 and Pax3 (PrimePCR SYBR Green Assay, Bio-Rad Laboratories, U.S.A.). The other sense and antisense primer sequences used in RT-qPCR analysis were as follows: Tnnt1, 5′-TGCACTAAAAGACCGCATTG-3′ and 5′-TCTTCTTGGCGTCATCCTCT-3′; Tnnc1, 5′-CAGCAAAGGGAAGTCTGAGG-3′ and 5′-TGCAGCATCATCTTCAGCTC-3′; Myf5, 5′-CAACCAGGAGGAGCGCGATCTCCG-3′ and 5′-AGGCGCTGTCCCAGTTGCATTC-3′; Mas, 5′-CCTCCCATTCTTCGAAGCTGTA-3′ and 5′- GCCTGGGTTGCATTTCATCTTT-3′; ACE2, 5′-TGAACACCATGAGCACCATT-3′ and 5′- TGCCCAGAGCCTAGAGTTGT-3′; AT1, 5′-GTTCCTGCTCACGTGTCTCA-3′ and 5′-CATCAGCCAGATGATGATGC-3′; AT2, 5′-GAATTACCCGTGACCAAGTCCT-3′ and 5′-GGAACTCTAAACACACTGCGGA-3′.
Quantification of muscle concentration of angiotensin II
Tissue lysate was extracted in saline with 0.1 N HCl and protease inhibitor cocktail (Thermo Fisher Scientific, U.S.A.) by mechanical disruption using TissueLyser LT (Qiagen, Germany) with zirconia balls. Angiotensin II concentration was measured by Angiotensin II EIA kit according to the manufacturer’s instructions (Enzo Life Sciences, U.S.A.). Protein concentration was measured by DC protein assay kit (Bio-Rad, U.S.A.), and muscle AII concentration was calculated by angiotensin II concentration divided by protein concentration in tissue lysate (pg/mg protein).
Histology and morphology
Skeletal muscles and skin were kept in 10% formalin for 24 h after excision. The samples were then dehydrated through an alcohol gradient and embedded in paraffin. Thereafter, the samples were sectioned and deparaffinised in xylene, rehydrated, and stained with Haematoxylin and Eosin (H&E). All skeletal muscles were cut to 10-μm thickness at the mid-abdomen of the muscle. The skin was cut perpendicular to the skin surface, 5–10 μm in the longitudinal direction. Basic muscle morphology was analysed with H&E staining. All stained, sectioned images were acquired under a microscope (Keyence BZ-X700, Keyence, Japan) at 20× magnification. All histological images were analysed using ImageJ software (ImageJ version 1.51, NIH, U.S.A.) as previously described . Single muscle fibre cross-sectional area (CSA) was measured using all EDL fibres except those with aspect ratio more than 2 in each microscopic image (number of total fibres were 4798, 5331, 4798, 5331, 4339, and 4331 in WT vehicle group, WT A1-7 group, ACE2KO vehicle group, ACE2KO A1-7 group, MASKO vehicle group, and MasKO A1-7 group, respectively). Single muscle fibre CSA was measured in all EDL fibres (WT vehicle group n=4798, WT A1-7 group n=5331, ACE2KO vehicle group n=4798, ACE2KO A1-7 group n=5331, MasKO vehicle group n=4339, MasKO A1-7 n=4331). The percentage of CNFs was counted at the mid-abdomen area of the EDL as previously reported . For skin analysis, the thicknesses of total skin layers, epidermis plus dermis layer, and the hypodermis layers were assessed using 30–50 random measurements from the skin surface in the vertical direction as previously reported .
Microcomputed tomography analysis
The femurs were incubated in 70% ethanol for 48 h after dissection. Three-dimensional microcomputed tomography (μCT) analyses were performed as described previously . Briefly, each femur was evaluated using a μCT system SMX-100CT-SV (Shimadzu, Japan). The images were reconstructed and bone volume per total volume (BV/TV) was evaluated with TRI3D-BON software (Ratoc System Engineering, Japan). BV/TV of metaphyseal regions in distal femur was analysed and presented as a representative image.
All data were analysed with GraphPad Prism Ver. 5.0 (GraphPad Software Inc., U.S.A.). Values are shown as mean ± standard error of mean (SEM) or as a percentage. Differences among the three phenotypes of mice were analysed with one-way ANOVA with a Bonferroni post-test. Differences in the therapeutic effect between A1-7 and vehicle were analysed with a Student’s t test, except for the comparison before and after the treatment in which a paired t test was used (Figure 1D). Regarding the distribution of single fibre area in the EDL, the frequency of the skeletal muscles was corrected as a percentage of the total number of the fibres and displayed as a histogram (Figure 2C–E). The differences of distribution were analysed with Kolmogorov–Smirnov test. P<0.05 was considered statistically significant.
Schematic overview, temporal changes in body weight and forelimb grip strength, and grip strength in aged mice treated with vehicle or A1-7
Muscle morphology in EDL muscle in aged mice treated with vehicle or A1-7
Mice lacking Mas did not mimic the early muscle weakness observed in mice lacking ACE2
We compared the temporal changes in muscle strength of ACE2KO and wild-type mice with mice lacking the primary A1-7 receptor, Mas (MasKO mice) (Figure 1A). There was no difference in body weight among three strains of mice during the experimental period (Figure 1B). Consistent with our recent report , ACE2KO mice exhibited lower forelimb grip strength than that of wild-type mice from 6 months until 24 months of age. In contrast, MasKO mice showed similar grip strength to that of wild-type mice except at 18 months of age (Figure 1C).
A1-7 infusion restored muscle weakness in old ACE2 KO and wild-type mice, but not in MasKO mice
Figure 1D demonstrates the grip strength of each mouse before and after administration of A1-7 or vehicle (saline) for 4 weeks. Grip strength of three strains of mice with vehicle treatment reduced significantly during the 4 weeks (Figure 1D). In contrast, grip strength did not reduce in A1-7-treated wild-type and ACE2KO mice, while grip strength reduced in Mas KO mice during A1-7 treatment. When the data were analysed as percent change of grip strength after drug treatment, the results indicated a significant improvement in muscle weakness upon A1-7 treatment in wild-type and ACE2KO mice, but not in MasKO mice (Figure 1E). We found any difference in BP neither among the genotypes nor between the treatments (Supplementary Table S1).
Muscle CSA was decreased by ACE2 deletion and increased by A1-7 treatment in a Mas-dependent manner
We previously observed no difference in size of the fast tibialis anterior (TA) muscle between ACE2KO and wild-type mice, and between mice with vehicle and A1-7 treatment at 25 months of age . In this study, we focussed on another fast-twitch muscle, the EDL muscle. Figure 2 shows the histological analysis of EDL muscles in 25-month-old mice. We found that the CSA of myofibres was significantly smaller in ACE2KO mice than wild-type and MasKO mice (Figure 2B). Myofibres were significantly larger in ACE2KO mice with A1-7 treatment than in ACE2KO mice with vehicle treatment (Figure 2B). A similar trend was observed in wild-type mice, but the difference did not reach statistical significance (P=0.08). There was no difference in fibre size between treatment groups in MasKO mice (P=0.24). By analysing the histogram of each single-fibre size, we found a rightward shift of the histogram in A1-7-treated mice compared with that in vehicle-treated mice, indicating an increased proportion of large muscle fibres upon A1-7 treatment in ACE2KO and wild-type mice (Figure 2C,E). In contrast, the single-fibre area histogram shifted left upon A1-7 treatment in MasKO mice (Figure 2D).
Aged MasKO mice exhibit distinct muscle aging signatures from those observed in ACE2KO mice
CNF is a hallmark of muscle regeneration but is also associated with muscle aging [15,18]. We previously found that CNFs increased in 25-month-old ACE2KO mice compared with wild-type mice in TA muscle . Consistently we found that the number of CNFs in EDL muscle of 25-month-old ACE2KO mice were larger than that of wild-type mice (Figure 2F). In contrast, there was no difference in CNFs between MasKO and wild-type mice (Figure 2F). We also found that expression of a senescence-associated gene, p16INK4a, increased prominently in EDL, TA, and the slow Quadriceps (QM) red muscle in ACE2KO mice, but not in MasKO mice, compared with expression in wild-type mice (Figure 3). There was no difference in the muscle expression of other senescence-associated genes p53, p21, and p19 among the three strains of mice (Supplementary Figure S1). Consistent with the previous report , A1-7 treatment did not alter CNFs or the muscular expression of senescence-associated genes in three strains of mice (Figure 3 and Supplementary Figure S1).
RT-PCR analysis of p16INK4a in skeletal muscles in old mice treated with vehicle or A1-7
Muscle expression of Pax3 increased upon A1-7 infusion in wild-type and ACE2KO mice, but not in MasKO mice
To screen genes altered by ACE2 deletion or A1-7 treatment in skeletal muscle, we first compared the expression of 84 genes associated with myogenesis and myopathy in EDL muscles among wild-type mice with vehicle treatment and ACE2KO mice with vehicle and A1-7 treatments using a commercially available PCR array kit. While no genes exhibited significant differences in expression upon multiple comparison among the three groups, additional comparisons between two groups revealed altered expression of five genes in EDL muscles of ACE2KO mice upon A1-7 treatment compared with expression in ACE2KO mice with vehicle treatment (Supplementary Table S2). The altered expression of these five genes (Pax3, Myf5, BMP4, Tnnt1, and Tnntc) was then validated by RT-qPCR using EDL muscles with a larger number of samples in three strains of mice with vehicle or A1-7 treatment. Pax3 expression was significantly greater in muscles upon A1-7 treatment than upon vehicle treatment in ACE2KO and wild-type mice, but not in MasKO mice (Figure 4). We did not find any change in the expression of the other four genes among all the groups (Supplementary Figure S2).
RT-PCR analysis of Pax3 in EDL muscle in aged mice treated with vehicle or A1-7
The aging signatures in adipose tissues were accelerated in aged ACE2KO mice but not in MasKO mice
Skin thinning is a signature of aging in mice [16,19]. In the histological analysis of skin layers, we found that the total thickness of the skin layers was significantly smaller in 25-month-old ACE2KO mice than that in wild-type and MasKO mice (Figure 5A, B). As shown in Figure 5C,D, this difference in the skin thickness appeared to result from a difference in the thickness of the hypodermis, which primarily consists of subcutaneous adipose tissue. A1-7 treatment did not alter skin thickness in the three strains of mice (Figure 5B). There was no difference in skin thickness between 4-month-old ACE2KO mice and wild-type mice (Supplementary Figure S3). We also found that the expression level of p16INK4a in epididymal adipose tissue increased in ACE2KO mice compared with wild-type and MasKO mice with vehicle treatment (Figure 5E). The expression level of p53 in epididymal adipose tissue was higher in wild-type mice than MasKO and ACE2KO mice, while that of p21 and p19 was equivalent between the three types of mice (Supplementary Figure S4) A1-7 treatment had no effect on the adipose expression of these genes (Figure 5E and Supplementary Figure S4).
Morphological analysis of skin and RT-PCR analysis of p16INK4a in adipose tissue of aged mice
Aged ACE2KO mice did not exhibit signatures of activation of the RAS
To investigate the influence of deletion in ACE2 or Mas on the local RAS, we evaluated the expression of ACE2, Mas, AT1a, and AT2 in QM muscle and epididymal fat. We found that there was no difference in the expression of these genes among the three types of mice except the influence of gene knockout in ACE2KO and MasKO mice (Figure 6A–D). A1-7 treatment also did not affect most of these genes except the unexpected increase in AT2 by A1-7 in epididymal fat in wild-type and ACE2KO mice but not in MasKO mice (Figure 6D). We also measured the concentration of AII in QM muscle. As a result, we did not find any difference in the concentration of these peptides among the genotypes or between the treatments (Figure 6E,F).
Tissue expression of RAS-associated genes and tissue concentration of angiotensin peptides
A1-7 treatment increased bone volume in old wild-type and ACE2KO mice, but not in MasKO mice
We carried out μCT analysis of femurs and found that the old male mice exhibited aging-associated loss of bone volume as indicated by decreased BV/TV compared with 3-month-old young male wild-type mice (Figure 7A,B). We found that BV/TV was higher in ACE2KO and wild-type mice with A1-7 treatment than in those with vehicle treatment (Figure 7B). There was no difference in BV/TV between MasKO mice treated with vehicle or A1-7. We also found no difference in BV/TV among the three strains of mice treated with vehicle (Figure 7B).
μCT analysis of bone volume in young mice, and aged mice treated with vehicle or A1-7
In the present study, we clarified the research question raised in our previous study : is the accelerated physiological muscle aging by ACE2 deletion primarily induced by the loss of its protective action through A1-7? We found that the early manifestation of muscle weakness with accelerated muscle aging signatures in ACE2KO mice was absent from MasKO mice. We also found that the fibre size of EDL muscles in ACE2KO mice, but not in MasKO mice, was smaller than that in wild-type mice, suggesting that aged ACE2KO mice exhibit clinical features of sarcopenia. MasKO mice unexpectedly exhibited reduced grip strength compared with wild-type mice at 18 months of age. Nevertheless, the absence of accelerated muscle aging phenotypes in MasKO mice at 25 months of age suggests that muscle aging was not accelerated in MasKO mice. Additionally, we found that prominent peripheral fat loss accompanied by increased adipose expression of the senescence-associated gene p16INK4a in old ACE2KO mice, but not in MasKO mice. These findings indicate that the impact of ACE2 on physiological systemic aging is independent of the A1-7-Mas pathway. In contrast, we found that A1-7 treatment restored muscle strength, muscle fibre size, and bone volume in aged mice via a Mas-dependent pathway, as shown by the absence of the effects of A1-7 in MasKO mice. While recent studies proposed several A1-7 receptors beside Mas including angiotensin type 1 receptor, MrgD, and angiotensin type 2 receptor, our findings suggest that A1-7 exerts its action through binding to Mas in skeletal muscles [20–22]. Collectively, our findings suggest that the endogenous production of A1-7 by ACE2 is not associated with physiological aging, but that overactivation of the A1-7-Mas pathway could alleviate sarcopenia and osteoporosis in aged mice (graphical summary in Supplementary Figure S5).
It should be noted that the potential increase in AII concentration by deletion of ACE2 can also be associated with accelerated aging. However, we did not find any alternation of muscle concentration of AII in ACE2KO mice. This is consistent with the previous findings that plasma concentration of AII and A1-7 are not altered in ACE2KO mice with normal aging [23,24]. These findings suggest that the potential influence of ACE2 on AII and A1-7 is compensated by alternative pathway in normal aging. In addition, the deletion of ACE2 did not affect the tissue expression of the RAS-associated genes including AT1 and AT2 (Figure 6). Taken together, the current findings suggest that the functions of ACE2 besides its role in RAS may be associated with the process of physiological aging. Besides its function as a carboxypeptidase to cleave angiotensin II into A1-7, ACE2 is known to have multiple functions, some of which may be involved in the aging process . As a carboxypeptidase, ACE2 also cleaves several molecules including dynorphin A [1–13], apelin, and des-Arg  bradykinin [8,26]. It was recently reported that mice deficient in either apelin or its receptor exhibit aging-associated muscle weakness, suggesting the protective role of apelin in muscle aging . Nevertheless, the antagonistic action of ACE2 in the apelin system is unlikely to explain the early muscle aging in ACE2KO mice. The C-terminus of ACE2 is highly homologous to collectrin, which functions as a chaperone protein of an amino acid transporter, B0AT1, while ACE2 and collectrin are predominantly expressed in the small intestinal epithelium and renal tubular cells, respectively . The interaction between B0AT1 and ACE2 in the small intestine is particularly important for tryptophan absorption, and tryptophan concentrations in blood and skeletal muscle are markedly decreased in ACE2-deficient mice [12,28]. Notably, tryptophan and its metabolic pathway are closely involved in regulating age-related diseases and lifespan . The impact of chronic tryptophan depletion on aging in mammals is still unclear. While early studies reported that animals fed a tryptophan-deficient diet showed increased longevity compared with control animals, it remains unknown whether tryptophan depletion itself or calorie restriction secondary to reduced food consumption caused by tryptophan depletion contributed to longevity in mammals . We found that food consumption in ACE2KO mice was similar to that in WT mice . We also found that the lifespan of ACE2KO mice is similar to that of WT mice, suggesting that early aging phenotypes in ACE2KO mice would not lead to shortened lifespan . Therefore, ACE2KO mice could be an ideal model to investigate the direct association between tryptophan depletion and physiological aging.
Aged ACE2KO mice exhibited increase in CNFs with reduced myofibre size in EDL muscle. We and others reported that CNFs increased by aging in skeletal muscle of mice, while the machinery of increased CNFs by aging can be different from that in regenerative process from acute muscle injury [6,15,18]. It is conceivable that increased muscle CNFs in ACE2KO mice is not a resultant of increased regenerative capacity, but a hallmark of accelerated muscle aging. p16INK4a, also known as cyclin-dependent kinase inhibitor 2A, promotes retinoblastoma (RB)-dependent cell-cycle arrest and contributes to systemic aging in mammals . Consistent with a previous report, we found that expression of p16INK4a increased prominently in both fast- and slow-twitch muscles in ACE2KO mice . In addition, we found that p16INK4a expression in epididymal adipose tissue increased along with prominent peripheral fat loss. We and others indicated that p16 INK4a in skeletal muscle was induced only at the oldest age during lifespan in wild-type mice [6,32]. p53, p19, and p21 were also reported to increase by aging, while the increase appears to be more modest and earlier than that in p16INK4a . We consistently observed that gene expression was prominently increased in p16INK4a but not in p53, p19, and p21 by ACE2 deletion. The specific increase in p16INK4a in ACE2KO mice is supported by no impact of ACE2 on p19 which shares the same INK4a locus. In aging tissues, p16INK4a is primarily expressed in progenitor cells, which are satellite cells and preadipocytes in skeletal muscle and adipose tissue, respectively . Aging progenitor cells expressing p16INK4a lose the ability to proliferate and differentiate, leading to impairment of tissue damage repair. It was reported that p16INK4a is not a marker of senescence, but has a causal role in the dysfunction of aged satellite cells . It was also reported that in BubR1 mutant mice, a mouse model of early senescence, p16INK4a depletion reversed subdermal thinning as well as muscle weakness and atrophy . Collectively, our findings will promote future investigations into whether p16INK4a is a marker or inducer of organ aging in ACE2KO mice.
Finally, we clearly found that A1-7 can rescue muscle weakness, muscle size, and bone volume in aged mice in a Mas-dependent manner. While A1-7 did not affect the size of fast-twitch TA muscle in a previous study , we found that A1-7 increased the fibre size of fast-twitch EDL muscle in old mice. We sought to find the gene signatures of muscle strengthening upon A1-7 treatment and found that out of 84 muscle-associated genes, only Pax3 expression was altered in A1-7-treated EDL muscle. Pax3 is a paralogue of Pax7, both of which are expressed in muscle progenitor cells but have distinct roles in myogenesis [37,38]. In various models of muscle disorders, A1-7 induced recovery of muscle function via mechanisms including AKT phosphorylation, prevention of increases in atrogin-1 and MuRF-1, reduction in p38 MAPK phosphorylation, and inhibition of TGFβ signalling [39–42]. Further investigation will be required to identify the signaling pathway of A1-7 that results in improved muscle weakness in relation to Pax3 overexpression in aged mice. We unexpectedly found that A1-7 increased the expression of AT2 in epididymal fat in wild-type and ACE2KO mice but not in MasKO mice. It was reported that AT2 in adipose tissue could positively modulate adipose metabolism by regulating stem cell differentiation to adipocytes . Further investigation will be required to elucidate how A1-7 increases the adipose AT2 and if the induction of AT2 by A1-7 would affect metabolism in old mice. The improvement of osteoporosis by A1-7 treatment was consistent with the previous report that A1-7 alleviates osteoporosis of ovariectomised rats by normalising bone expression of osteoprotegerin and receptor activator NF-κB ligand . The notion of locomotive dysfunction has been proposed in recent years as a concept encompassing the dysfunction of body parts involved in exercising, including muscles, bones, and joints, which causes the inability to walk, eventually increasing the risk of long-term care . In this regard, our results provide a basis for the future clinical application of A1-7 or Mas agonists in treating older adults [46,47].
We conducted the study to clarify the research question if the influence of ACE2 on physiological aging primarily induced by its protective action through activation of MAS by A1-7.
We found the phenotypes of enhanced physiological aging observed in ACE2KO mice were absent from MasKO mice. In contrast, A1-7 treatment alleviated muscle weakness, muscle size, and bone volume in aged mice via a Mas-dependent pathway.
Our findings suggest that the endogenous production of A1-7 by ACE2 is not associated with physiological aging, but that treatment for activating the A1-7-Mas pathway might alleviate sarcopenia and osteoporosis in older population.
We are most grateful to Hikari Kitamura and Yuka Nakao for their excellent technical assistance.
S.N. performed experiments and wrote the manuscript. K.Y. conceived the idea of the study, designed experiments, and wrote and edited the manuscript. M.M., M.H., and H.R. designed experiments. H.T., Y.N., Y.I., T.F., S.Y., M.N., M.T., K.H., H.A., Y. Takami, Y. Takeya, and K.S. performed experiments.
This work was supported by the Grant-in-Aid for Scientific Research [grant number J550703552].
The authors declare that there are no competing interests associated with the manuscript.
angiotensin-converting enzyme 2
bone volume per total volume
centrally nucleated fibre
extensor digitorum longus
Haematoxylin and Eosin
Mas-related G-protein coupled receptor type D
muscle RING-Finger Protein-1
quantitative real-time PCR