FOXO1 (forkhead box O1), a forkhead-type transcription factor whose gene expression is up-regulated in the skeletal muscle during starvation, appears to be a key molecule of energy metabolism and skeletal muscle atrophy. Cathepsin L, a lysosomal proteinase whose expression is also up-regulated in the skeletal muscle during starvation, is induced in transgenic mice overexpressing FOXO1 relative to wild-type littermates. In the present study, we conducted in vivo and in vitro experiments focusing on FOXO1 regulation of Ctsl (cathepsin L gene; CTSL1 in humans) expression in the skeletal muscle. During fasting and refeeding of C57BL/6 mice, Ctsl was regulated in parallel with FOXO1 in the skeletal muscle. Fasting-induced Ctsl expression was attenuated in transgenic mice overexpressing a dominant-negative form of FOXO1 or in skeletal-muscle-specific Foxo1-knockout mice relative to respective wild-type controls. Using C2C12 mouse myoblasts overexpressing a constitutively active form of FOXO1, we showed that FOXO1 induces Ctsl expression. Moreover, we found FOXO1-binding sites in both the mouse Ctsl and human CTSL1 promoters. The luciferase reporter analysis revealed that the mouse Ctsl and human CTSL1 promoters are activated by FOXO1, which is abolished by mutations in the consensus FOXO1-binding sites. Gel mobility-shift and chromatin immunoprecipiation assays showed that FOXO1 is recruited and binds to the Ctsl promoter. The present study provides in vivo and in vitro evidence that Ctsl is a direct target of FOXO1 in the skeletal muscle, thereby suggesting a role for the FOXO1/cathepsin L pathway in fasting-induced skeletal muscle metabolic change and atrophy.
The skeletal muscle is the largest organ in the human body, with important roles in exercise, glucose uptake and energy expenditure. Skeletal muscle metabolism is changed by the supply of nutrients and circulating hormones [1,2]. Starvation and disease states (such as diabetes and cancer cachexia) lead to a rapid reduction in skeletal muscle mass (atrophy) . What is the physiological role of muscle atrophy? As the brain mainly uses glucose as an energy source, during starvation it needs to be supplied with glucose. Thus, for short periods of fasting, skeletal muscle increases utilization of lipids instead of glucose. On the other hand, for longer periods of fasting or starvation resulting in muscle atrophy, skeletal muscle protein is degraded and mobilized as a source of amino acids for gluconeogenesis that occurs mainly in the liver .
The FOXO (forkhead box O) members FOXO1, FOXO3a and FOXO4 belong to a subfamily of the forkhead transcription factors [4,5]. The FOXO family regulates a variety of biological processes such as metabolism, cell proliferation, apoptosis, stress response and longevity [6–9]. FOXO1 activates gluconeogenic enzyme genes in the liver, such as those for PEPCK (phosphoenolpyruvate carboxykinase) and G6Pase (glucose-6-phosphatase). A dominant-negative form of FOXO1 (DN-FOXO1), which contains the DNA-binding domain, but lacks the transcriptional activation domain, suppressed the fasting-induced increase of Pepck and G6Pase expression in liver cells . We showed previously that energy-deprived conditions in mice, such as fasting and diabetes, up-regulated expression of Foxo1 in skeletal muscle of mice . Several FOXO1 target genes have been identified in skeletal muscle. For instance, FOXO1 up-regulates PDK4 (pyruvate dehydrogenase kinase 4), a kinase that suppresses glycolysis , and LPL (lipoprotein lipase), an enzyme that increases lipid incorporation , and down-regulates SREBP1c (sterol-regulatory-element-binding protein 1c), a master regulator of lipogenesis . The FOXO1 target genes may be involved in the utilization of lipids instead of glucose in the skeletal muscle. On the other hand, forced expression of FOXO1 or FOXO3a up-regulates the expression of a variety of atrophy-related genes including the MuRF1 and atrogin/MAFbx ubiquitin ligases [14,15], as well as Bnip3 and LC3, important molecules for autophagy [16,17], thus inducing skeletal muscle atrophy in vitro and in vivo. We have created transgenic mice that overexpress FOXO1 in skeletal muscle (FOXO1 mice) and found that they exhibit skeletal muscle atrophy . Moreover, FOXO1 has been shown to induce Gadd45a (growth-arrest and DNA-damage-inducible protein 45α) , a suppressor of the cell cycle, thereby facilitating skeletal muscle atrophy. Thus identification and functional analysis of FOXO1 target genes will help facilitate a better understanding of skeletal muscle metabolism.
Cathepsin L is a lysosomal proteinase, whose expression is up-regulated during various forms of skeletal muscle atrophy including starvation [20–22]. In the skeletal muscle of the FOXO1 mice, Ctsl expression was markedly increased . Earlier findings showed that lysosomal proteolysis is activated upon skeletal muscle atrophy [23,24]. Although circumstantial evidence suggests that cathepsin L is involved in skeletal muscle atrophy, to our knowledge, there are no reports on the regulation of Ctsl by FOXO1. In the present study, we provide in vivo and in vitro evidence that Ctsl is a direct target of FOXO1 in skeletal muscle.
Genetically modified animals
The human skeletal muscle α-actin promoter  was kindly provided by Dr E.C. Hardeman and Dr K. Guven (Children's Medical Research Institute, Westmead, NSW, Australia). DN-FOXO1, a mutant version of FOXO1 containing amino acid residues 1–256, has been described previously . Transgenic plasmid containing the cDNA for DN-FOXO1 (see Figure 2A) was excised and purified for injection (at 2 ng·μl−1) . Fertilized eggs were recovered from C57BL/6 females crossed with C57BL/6 males and microinjected at Japan SLC Inc. (Hamamatsu, Japan). To obtain skeletal-muscle-specific Foxo1-knockout mice, we inactivated Foxo1 expression in the skeletal muscle by crossing mice homozygous for a floxed Foxo1 allele with myogenin-cre transgenics. Myogenin-cre and Foxo1flox mice were as described previously . The mice were maintained at a constant temperature of 24 °C with fixed artificial light (12 h light/12 h dark). All animal experiments were conducted in accordance with the guidelines of Tokyo Medical and Dental University Committee on Animal Research (No. 0090041) and National Institute of Health and Nutrition (No. 0706).
C2C12 cells and cell cultures
C2C12 mouse myoblasts (RIKEN Cell Bank, Tsukuba, Japan) were cultured in DMEM (Dulbecco's modified Eagle's medium) containing 10% (v/v) FBS (fetal bovine serum) until the cells reached confluence. The medium was then replaced with DMEM containing 2% (v/v) horse serum (differentiation medium) and incubated for 4 days to induce the formation of myotubes before each experiment. C2C12 myoblasts stably expressing FOXO1–ER (oestrogen receptor) fusion proteins were obtained as described previously . In brief, C2C12 cells were stably transfected with the empty pBABE retrovirus or pBABE vectors expressing fusion proteins containing a constitutively active form of human FOXO1 [FOXO1(3A)]  [where three Akt phosphorylation sites (Thr24, Ser256 and Ser319) are replaced by alanine residues] in-frame with a modified TAM (tamoxifen)-specific version of the murine ER-ligand-binding domain. FOXO1–ER plasmid was provided by Dr Terry G. Unterman (Department of Medicine, University of Illinois at Chicago, U.S.A.) Cells were selected with puromycin and colonies were pooled for studies, as reported previously . The fusion proteins are restricted to the cytoplasmic space until activation by treatment with TAM .
Quantitative real-time PCR
Quantitative real-time PCR was performed as described previously . Total RNA was prepared using Sepazol. cDNA was synthesized from 5 μg of total RNA using ReverTra Ace® (TOYOBO) with random primers. Gene expression levels were measured with an ABI PRISM 7700 Sequence Detection System using SYBR Green PCR Core Reagents (Applied Biosystems). Levels of mRNA were normalized to those of 36B4 mRNA. The primers used were as follows. Cathepsin L: forward, 5′-TCTCACGCTCAAGGCAATCA-3′, reverse, 5′-AAGCAAAATCCATCAGGCCTC-3′; GADD45α: forward, 5′-CGTAGACCCCGATAACGTGGTA-3′, reverse, 5′CGGATGAGGGTGAAATGGAT-3′; FOXO1: forward, 5′-ATTCGGAATGACCTCATGGA-3′, reverse, 5′-GTGTGGGAAGCTTTGGTTGG-3′; DN-FOXO1 (transgene specific): forward, 5′-GACTACAAGGACGACGATGA-3′, reverse, 5′-AGCGGCTCGAAGTCCGGGTC-3′, FOXO3a: forward, 5′-TCTGCGGGCTGGAAGAACT-3′, reverse, 5′-CTCTTGCCCGTGCCTTCAT-3′; FOXO4: forward, 5′-ATGGATGGTCCGCACGGTG-3′, reverse, 5′-CTTGCCAGTGGCCTCGTTG-3′; and 36B4: forward, 5′-GGCCCTGCACTCTCGCTTTC-3′, reverse 5′-TGCCAGGACGCGCTTGT-3′.
Cloning of the mouse
Ctsl and human CTSL1 promoters
The mouse Ctsl promoter has been described previously . The 4-kb mouse Ctsl promoter region was excised with BamHI from pMEPCAT3 and cloned into a pGL3-basic luciferase vector (Promega Corporation). The 4-kb mouse promoter was sequenced. The human CTSL1 promoter [30,31] was obtained by PCR from genomic DNA of HEK (human embryonic kidney)-293 cells. The PCR primers used were 5′-GTGGTGCGCGCCTGTAGTCC-3′ and 5′-GGCGCACTCCACGGATGCCG-3′. Mutations in the promoter sequences were introduced using a QuikChange® site-directed mutagenesis kit (Stratagene). Primers used were human DBE1: 5′-CTGGGACAGTCAGTGGGCAAGCCACGAACC-3′; human DBE2: 5′-GGGACAGTCAGTGGGCAAGCCACGAACC-3′; and mouse DBE: 5′-GTGATAGACTGAGTGGGCAAACATACAAAG-3′. DBE is DAF16 (decay-accelerating factor 16)-binding element, to which FOXO1 binds .
Transfection and luciferase assay
HEK-293 cells were plated at a density of 105 cells/12-well plate in DMEM containing 10% (v/v) FBS. Luciferase gene constructs containing a Ctsl promoter fragment with or without mutations of putative FOXO1-binding sites were prepared. The luciferase reporter plasmid (0.8 μg), the expression plasmid [pCAG-FOXO1(3A) or empty pCAG, 0.8 μg], and a phRL-TK vector (25 ng; Promega) as an internal control for transfection efficiency, were transfected into HEK-293 cells using Lipofectamine™ 2000 (Invitrogen). After an overnight transfection period, cells were lysed and assayed for luciferase activity using the dual-luciferase assay kit (Promega). The activity was calculated as the ratio of firefly luciferase activity to Renilla luciferase activity (internal control) and expressed as the average of triplicate experiments.
Gel mobility-shift assay
The gel mobility-shift assay was performed as described previously . In-vitro-translated human FOXO1 was generated from pCMX-FOXO1, using the TNT® T7 Quick Coupled Transcription/Translation System (Promega) according to the manufacturer's instruction. Double-stranded oligonucleotide probes used in gel mobility-shift assays were prepared by annealing both strands of each putative FOXO1-binding site in the human CTSL1 promoter (DBE1: 5′-ATCTCCAAAATAGTAAACAAATTCCTGCAG-3′, −145 to −152, numbering the first nucleotide of exon 1 as +1; DBE2: 5′-GGGACAGTCAGTAAACAAGCCACGAACC-3′, −1400 to −1407, numbering the first nucleotide of exon 1 as +1; DBE1 mutant: 5′-ATCTCCAAAATAGTGGGCAAATTCCTGCAG-3′; and DBE2 mutant: 5′-GGGACAGTCAGTGGGCAAGCCACGAACC-3′; underlining indicates sites of mutation) and labelling with [γ-32P]ATP (PerkinElmer Life Sciences) using T4 polynucleotide kinase (Roche Applied Science). The labelled probes (50000 d.p.m.) were incubated with extracts containing in-vitro-translated FOXO1 in a mixture (total volume of 25 μl) containing 10 mM Tris/HCl (pH 7.5), 50 mM NaCl, 1 mM DTT (dithiothreitol), 1 mM EDTA and 4.4% glycerol with 1 mg of poly(dI-dC)·(dI-dC) for 30 min on ice and then separated by electrophoresis on a 6% polyacrylamide gel in 45 mM Tris/HCl (pH 8.0), 45 mM borate and 1 mM EDTA. After electrophoresis, gels were dried and analysed with a BAS-2500 (Fuji Film).
ChIP (chromatin immunoprecipitation) assay
ChIP was carried out using a ChIP assay kit (Upstate Biotechnology) according to the manufacturer's guidelines [13,34]. Briefly, C2C12 myoblasts stably expressing FOXO1(3A)–ER were incubated for 24 h with or without 1 μM TAM. Proteins were cross-linked to DNA with the addition of formaldehyde (1% final concentration). Cells were washed and lysed in SDS lysis buffer, sonicated for 10 s and allowed to recover for 30 s over ice (this was repeated seven times). Lysates were cleared with Protein A–agarose for 30 min, pelleted and incubated overnight with an anti-FOXO1 antibody (sc-11350; Santa Cruz Biotechnology). Before the incubation, input samples were removed from the lysate and stored at 4 °C until extraction. Following incubation with the antibody, protein–DNA complexes were eluted (1% SDS and 0.1 M NaHCO3), and the cross-links were reversed. DNA was purified by phenol/chloroform extraction. PCR primers were designed to locate DBE of the Ctsl promoter: forward, 5′-AAAAGACAAGAGGATGCCTT-3′, and reverse, 5′-CTGGTGTTCTCAGGTTAGTC-3′. The amplified region was −3670 to −3339, numbering the first nucleotide of exon 1 as +1. PCR primers were also designed to locate non-DBE of the Ctsl promoter: forward, 5′-CCACGAAAAGAATTTCTACCA-3′ and reverse, 5′-AGTTGTAGATTAAAATGTGCAG-3′. The amplified region was −439 to −289, numbering the first nucleotide of exon 1 as +1.
All results are expressed as means±S.E.M. Statistical comparisons of data from experimental groups were made with a one-way ANOVA, and groups were compared using Fisher's PLSD (protected least-significant difference) test (Statview 5.0; Abacus Concepts, Berkeley, CA, U.S.A.). When differences were significant, groups were compared using Fisher's PLSD test. Statistical significance was defined as P<0.05.
Co-ordinate regulation of
Foxo1 and Ctsl expression in the mouse skeletal muscle during fasting and refeeding
To analyse the in vivo relationship between Foxo1 and Ctsl expression, we first examined their gene expression in the skeletal muscle of mice subjected to fasting and refeeding. Expression of mRNAs for Foxo1 and Gadd45a, a bona fide FOXO1-target gene , was increased in skeletal muscle after 24 and 48 h of fasting (Figures 1A and 1B). Ctsl mRNA expression was also markedly increased in the skeletal muscle during fasting (Figure 1C). The effect of fasting was reversed by refeeding. Foxo1 and Ctsl mRNA levels were increased in various regions of skeletal muscles, such as gastrocnemius (Figure 1), soleus, extensor digitorum longus and tibialis anterior (see Supplementary Figure S1 at http://www.BiochemJ.org/bj/427/bj4270171add.htm). In other tissues, such as the brain, kidney and adipose tissue, fasting did not markedly alter Foxo1 and Ctsl mRNA expression (Y. Kamei, unpublished work). These observations indicate that the expression of Foxo1 and Ctsl is co-ordinately regulated in the skeletal muscle during fasting and refeeding.
Foxo1, Gadd45a and Ctsl expression in skeletal muscle of fasted and refed mice
Transgenic mice overexpressing DN-FOXO1
Previously, we generated transgenic mice with skeletal-muscle-specific overexpression of human FOXO1 using the α-actin promoter (FOXO1 mice) . Skeletal muscle in these FOXO1 mice showed an increase in Ctsl mRNA levels . To examine the possible in vivo regulation of Ctsl by FOXO1, we also generated transgenic mice with skeletal-muscle-specific overexpression of DN-FOXO1 (Figure 2A), which suppresses FOXO1-mediated transcription. DN-FOXO1 contains the DNA-binding domain, but lacks the transcription activation domain, of FOXO1 [10,27,35]. DN-FOXO1 transgene expression was observed specifically in skeletal muscle (Figure 2B). Histologically, there was no appreciable difference in skeletal muscle between DN-FOXO1 and wild-type mice (results not shown).
Creation of DN-FOXO1 transgenic mice
expression is suppressed in the skeletal muscle of DN-FOXO1 mice Ctsl
We used 16 DN-FOXO1 mice and 16 gender- and age-matched wild-type mice. Eight mice each were allowed to eat freely (fed) or were fasted for 24 h. Foxo1, Gadd45a and Ctsl expression was increased in the skeletal muscle from wild-type mice (Figure 3). In DN-FOXO1 mice, fasting-increased endogenous Foxo1 expression was attenuated compared with wild-type mice (Figure 3A), suggesting that FOXO1 up-regulates its own gene expression. Moreover, induction of Ctsl as well as Gadd45a expression by fasting was markedly diminished in the DN-FOXO1 mice (Figures 3B and 3C). These observations indicate that FOXO1 significantly contributes to the up-regulation of Ctsl expression during fasting in vivo.
Gene expression in skeletal muscle of fed or fasted DN-FOXO1 mice
Ctsl expression is suppressed in the skeletal muscle of skeletal-muscle-specific Foxo1-knockout mice
To examine whether the induction of Ctsl expression in the muscles of fasting animals is dependent on FOXO1 or not, we used muscle-specific Foxo1-knockout mice (myogenin-cre/Foxo1flox) . Knockout and control mice were fed or were fasted for 24 h, and expression of FOXO family members (Foxo1, Foxo3a and Foxo4), Ctsl and Gadd45a was examined (Figure 4). In control mice, Foxo1, Foxo3a, Ctsl and Gadd45a expression was markedly up-regulated. The Foxo1 expression in the knockout mice that were fed was much lower than that in the control mice that were fed. We did not observe marked induction of Foxo1 expression in the fasted knockout mice relative to the fed knockout mice. In the knockout mice, the induction of Ctsl expression was suppressed (Figure 4C). Expression of FOXO4 did not differ among groups (results not shown). These observations indicate that FOXO1 is important for the up-regulation of Ctsl expression during fasting. However, since Ctsl expression during fasting was not completely suppressed, there may be additional factor(s). FOXO3a may be such a factor, as its expression was up-regulated during fasting both in control and knockout mice (Figure 4D).
Gene expression in the skeletal muscle of fed or fasted muscle-specific
Activation of FOXO1 in C2C12 myocytes promotes
To study the effects of FOXO1 on Ctsl expression in muscle cells, we first employed C2C12 cells stably expressing a constitutively active form of FOXO1 [FOXO1(3A)] in-frame with a modified form of the ER ligand-binding domain that responds selectively to TAM . Previous studies with these cells have shown that fusion proteins are restricted to the cytoplasmic space in the absence of ligand and then rapidly translocate to the nucleus upon treatment with TAM . Each mRNA signal in Figure 5(A) is the sum of endogenous Foxo1 mRNA and retrovirus-derived FOXO1(3A)–ER mRNA. The endogenous Foxo1 mRNA was very low in C2C12 cells. As expected, treatment with TAM did not change FOXO1(3A)–ER mRNA levels. Treatment with TAM resulted in a marked induction of Gadd45a expression, confirming that it successfully promoted the transcriptional activity of our FOXO1(3A)–ER-C2C12 myotubes. As shown in Figure 5, treatment with TAM also markedly increased the mRNA abundance of Ctsl as well as Gadd45a. No changes in Ctsl mRNA expression were observed in FOXO1(3A)–ER cells in the absence of TAM or in control C2C12 cells stably transfected with empty vector (Mock) (Figure 5C). These results suggest that the expression of Ctsl is up-regulated directly by the activation of FOXO1 in muscle cells.
Ctsl expression by FOXO1 in C2C12 muscle cells
Ctsl promoter is activated by FOXO1
The above data suggest that Ctsl is a direct transcriptional target of FOXO1 in muscle cells. FOXO1 is known to bind the sequence GTAAACAA or DBE . We therefore examined using a transient transfection-reporter assay whether the mouse Ctsl and human CTSL1 promoters are activated by FOXO1. The mouse Ctsl promoter has been cloned previously . We sequenced the 4-kb mouse Ctsl promoter and found a single consensus FOXO1-binding site (GTAAACAA) (−3528 to −3535, numbering the first nucleotide of exon 1 as +1). Plasmid constructs linking the mouse Ctsl promoter including the putative FOXO1-binding site to the luciferase reporter gene were analysed. FOXO1 increased the mouse Ctsl promoter (−4000 to +10)-driven reporter activity (Figure 6A). Furthermore, mutation in the consensus FOXO1-binding site abolished the FOXO1-induced luciferase activity (Figure 6A). Consistent with in vivo transgenic mice data (Figure 3), in the in vitro transfection reporter assay, DN-FOXO1 dose-dependently suppressed the FOXO1(3A)-induced transcriptional activity of the mouse Ctsl promoter (Figure 6B). These observations suggest that FOXO1 up-regulates the mouse Ctsl expression via the DBE sequence in its promoter.
Transient transfection-reporter assay of the effect of FOXO1 on
CTSL1 promoter is bound and activated by FOXO1
In the human CTSL1 promoter [30,31], we also found two potential FOXO-binding sites; there are two perfect DBEs (−145 to −152 and −1400 to −1407, numbering the first nucleotide of exon 1 as +1). FOXO1 increased the human CTSL1 promoter (−1600 to +10)-driven reporter activity in a transfection assay (Figure 6C). In addition, mutations in the consensus DBEs abolished the FOXO1-induced luciferase activity. Thus the results of the luciferase assay with the human CTSL1 promoter were similar to those with the mouse Ctsl promoters.
We also examined the binding of FOXO1 to the DBEs in the human CTSL1 promoter with a gel mobility-shift assay. FOXO1 that was synthesized in vitro clearly bound to oligonucleotides containing the putative FOXO1-binding sites of the human CTSL1 promoter, and did not bind to oligonucleotides with mutations in the consensus DBEs (Figure 7A). Moreover, we performed a ChIP analysis using C2C12 cells expressing FOXO1(3A)–ER (as used in Figure 5), and found that FOXO1 was recruited to the mouse Ctsl promoter containing the DBE (Figure 7B). These observations, taken together, suggest that FOXO1 up-regulates the mouse Ctsl and human CTSL1 expression via the DBE sequences of their promoters; CTSL1 is a direct target of FOXO1 in the skeletal muscle.
Recruitment of FOXO1 to the putative FOXO1-binding sites of the
FOXO1 signalling is important in linking nutritional and hormonal cascades to the regulation of skeletal muscle atrophy. As a transcriptional factor and/or cofactor, FOXO1 regulates many genes in a variety of biological processes. Identification and molecular analysis of FOXO1 target genes should help facilitate a better understanding of skeletal muscle metabolism. In the present study, we showed that FOXO1 directly activates Ctsl expression.
In the present study, we first conducted in vivo experiments focusing on FOXO1 regulation of Ctsl expression in skeletal muscle in the context of physiological nutritional change. During fasting and refeeding of C57BL6 mice, Ctsl was regulated in parallel with FOXO1 in skeletal muscle (Figure 1). Fasting-induced Ctsl expression was attenuated in DN-FOXO1 mice (Figure 3) and in skeletal-muscle-specific Foxo1-knockout mice (Figure 4) relative to respective wild-type controls. In this regard, we observed previously that Ctsl expression is markedly increased in skeletal muscle of FOXO1 mice . Taken together, our results suggest that FOXO1 activates Ctsl expression in vivo. The increase in Ctsl mRNA is delayed compared with that of Foxo1 (Figure 1). This could be explained as follows: (i) Ctsl mRNA may have a long half life, or (ii) Ctsl is activated by different transcription factors as well as FOXO1 during fasting. Indeed, it has been reported that addition of glucocorticoid, whose blood level is increased during fasting, has been reported to increase the level of Ctsl mRNA . Cathepsin L is considered to play a major role in the terminal degradation of proteins delivered to lysosomes by endocytosis or autophagy [24,36]. A previous study shows that pharmacological inhibition in rats of both cathepsin L and calpain, an intracellular Ca2+-dependent protease, prevents sepsis-induced bulk protein degradation  and suggests a role for cathepsin L in the degradation of various skeletal muscle proteins. It is therefore conceivable that FOXO1-induced transcriptional activation of Ctsl plays a role in fasting-induced autophagy and proteolysis. During fasting, a large number of genes show a change in their expression; some are changed directly as a physiological response, and others may be changed indirectly as secondary events. Nevertheless, Ctsl expression is likely to be regulated by FOXO1.
Using C2C12 myoblasts, we also showed that FOXO1 induces endogenous Ctsl expression in vitro (Figure 5). Moreover, we showed that FOXO1 can bind to and activate the Ctsl promoter (Figures 6 and 7). The promoter of Ctsl has been sequenced and analysed in humans, mice and rats. Transcription factors such as the specificity proteins Sp1/Sp3 have been reported to increase the basal activities of the promoter [31,38,39]. In the present study, we have provided the first evidence for transcriptional activation of the Ctsl promoter by an inducible transcription factor, FOXO1, thereby suggesting that Ctsl is a direct target gene of FOXO1. During fasting, among members of the cathepsin family, only Ctsl expression is markedly increased in skeletal muscle . Indeed, there are no consensus DBEs in the putative promoter of other cathepsins (−1.5 kb from the transcription start sites of cathepsins B, C, D, E, G, H, J, K S and Z in humans and mice; Y. Yamazaki and Y. Kamei, unpublished work), indicating that FOXO1 specifically activates Ctsl during fasting. Therefore cathepsin L probably plays a role in fasting-induced adaptive responses including skeletal muscle atrophy.
FOXO1 has been shown to activate the expression of Gadd45a, Pepck and G6Pase via direct binding to DBE in vitro [19,40]. On the other hand, adenoviral introduction of DN-FOXO1 can suppress the gene expression of Pepck and G6Pase in the liver in vivo . Previous studies suggest that FOXO1 can regulate gene expression in at least two different ways: (i) FOXO1 directly binds to and transactivates the promoter of its target genes , or (ii) FOXO1 interacts with other transcription factors via protein–protein interactions, without DNA binding, thereby regulating the expression of target genes [27,41]. Because FOXO1 interacts with nuclear receptors via its C-terminus , DN-FOXO1 can suppress the action of FOXO1 as a transcription factor without affecting its action as a transcriptional cofactor. Therefore FOXO1 appears to regulate Ctsl expression via a transcriptional mechanism.
We reported previously that FOXO1 mice have decreased skeletal muscle mass . In the present study, there was no appreciable histological difference in skeletal muscle between DN-FOXO1 and wild-type mice. We also observed no marked difference in body weight and skeletal muscle mass between DN-FOXO1 and wild-type mice (results not shown). This may be because DN-FOXO1 does not suppress all the FOXO1 actions in a dominant-negative fashion, as described in the above. A detailed phenotypic analysis of DN-FOXO1 mice is ongoing in our laboratory.
In conclusion, the present study provides in vivo and in vitro evidence that Ctsl is a direct target of FOXO1 in skeletal muscle. The results provide important clues towards understanding the molecular mechanism underlying FOXO1-mediated transcriptional regulation of gene expression in skeletal muscle. Further studies will better clarify the physiological and pathophysiological implication of FOXO1-induced Ctsl expression in skeletal muscle.
DAF16 (decay-accelerating factor 16)-binding element
Dulbecco's modified Eagle's medium
dominant-negative forkhead box O1
fetal bovine serum
forkhead box O
growth-arrest and DNA-damage-inducible protein 45α
human embryonic kidney
protected least-significant difference
Yasutomi Kamei, Tadahiro Kitamura, Takayoshi Suganami, Osamu Ezaki and Yoshihiro Ogawa led the design and overall implementation of the trial. Yasutomi Kamei wrote the initial draft of the paper in consultation with Yukio Hirata, Bruce Troen and Yoshihiro Ogawa. Yoshihiro Yamazaki, Satoshi Sugita, Fumiko Akaike, Sayaka Kanai, Shinji Miura and Ichizo Nishino were responsible for laboratory analyses. All authors contributed to interpretation of data and have seen and approved the final manuscript.
We thank Ms Chihiro Osawa for technical assistance of Foxo1-knockout mice analysis. We thank Dr Terry G. Unterman for providing the FOXO1(3A)–ER plasmid. S. Sugita is a Research Fellow of the Japan Society for the Promotion of Science.
This work was supported in part by a Grant-in-Aid for scientific research KAKENHI from the Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT, Tokyo, Japan), from the Japanese Ministry of Health, Labour and Welfare, and research grants from Astellas Foundation for Research on Metabolic Disorders, Ono Medical Research Foundation, Takeda Science Foundation, and Novo Nordisk Pharma Ltd Insulin Research 2009.